-------
100.0-
RIC
VJ1
CO
412 S
I
1327100
100
5:00
—I
200
10:00
—I
300
15:00
—I
400
20:00
I
500 SCAN
25:00 TIME
Figure 22. C.C/MS chromatogram for purgeables in unspiked
Kansas City, Kansas, POTW primary sludge.
-------
349
o.
(X
m
3322
ta
422
269
RIC
184
298
JS
509
403
487
240
146
"I
100
5:00
—I
300
15:00
—I
400
20:86
I
200
10:00
I
500
25:00
SCAN
TIME
Figure 23. OC/MS ehromatogram for purgeables in unspiked
Industrial No. 2 primary sludge.
-------
Nonetheless, in most cases the recoveries observed were both good and
reproducible. Only 11% of all recovery determinations were less than 50%.
More than 63% fell within the range of 50 to 150% recovery. An additional
25% of the recovery determinations exceeded 150%. Spike recoveries in excess
of 100% may reflect slightly lower recoveries for the internal quantitation
standards relative to the anaytes. The relative standard deviations (RSD)
for triplicate recovery determinations were generally low. More than 90% of
the RSDs were 30% or less. Of those, 27% were less than 10% RSD. The method
reproducibility was surprisingly good considering the difficulty in removing
representative aliquots from a heterogeneous sample matrix. Considering the
variability of sludge characteristics, the heterogeneous nature of sludges,
and the detection limits achieved, the results of these precision and accuracy
determinations demonstrate that the proposed method can be reliably applied
to the analysis of purgeable compounds in municipal and industrial wastewater
treatment sludges.
TIME/COST ANALYSIS OF THE METHOD
In general, the per sample labor and costs required for analyses of sludge
samples by the purge-trap GC/MS protocol (Method Appendix) are similar to those
for wastewater analyses. However, the analysis of a higher percentage of spiked
and duplicate samples may be required to check the precision and accuracy of
the method for each sludge matrix. Estimates of the labor and cost were based
on several important conditions:
" Analyses would be conducted by professional analytical chemists who
are experienced in trace organic analysis of sludges or at least in trace
organic analysis of difficult sample matrices.
" No major equipment would be purchased. Only the costs of replacing
consumable supplies and glassware were estimated.
* Sample analyses would be conducted on lots of 20 to 100 samples.
Analysis of fewer samples would result in higher per sample costs.
" Analytical standard solutions would be prepared from commercially avail-
able mixed standard solutions. Preparation of mixed analytical standards from
the pure compounds requires a considerable labor investment that may not be
cost-effective for sample lots of 100 or less.
Under these conditions, the per sample time/cost estimate for analysis
of purgeables in sludge is 2.5 labor-hours of technical effort and direct
supervision, 1.0 h of GC/MS instrument time, and $10 for other direct costs.
Of the labor hours, 1.5 h are estimated for data retrieval, interpretation,
and reporting.
Many operations associated with routine analysis of purgeables in sludges
are dependent on the specific program objectives and are not included in these
estimates. However, they do contribute to the overall cost of the program.
60
-------
These operations include analysis of spiked and replicate samples for analyti
cal quality control and miscellaneous sample handling such as preparing com-
posite samples and spiking samples. In addition, reanalysis may be required
for samples with analyte responses outside the dynamic range of the GC/MS de-
tector. As many as 30 to 40% reruns are typically required for routine
analysis of POTW sludges from various sources.5
3 Hansen, E. M., Midwest Research Institute, personal communication (1980)
-------
SECTION 5
DEVELOPMENT AND EVALUATION OF METHODS FOR EXTRACTABLE COMPOUNDS
As was the case for purgeable compounds, the development of methods for
the analysis of extractable compounds in sludge focused initially on adapting
the EPA protocol for industrial wastewaters.1 The differences in character
between some untreated industrial and municipal wastewaters and many sludges
are minimal. In general, sludges contain higher levels of both dissolved and
suspended solids. The high solids content of sludges poses a significant bar-
rier to efficient extraction of the acidic, basic, and neutral compounds. In
addition, the high level of extractable materials in sludges necessitates the
use of selective extract preparation methods to mitigate coextracted interfer-
ences. Development of efficient and selective extraction methods for sludges
is a formidable challenge because of the wide range of physical and chemical
properties of the organic priority pollutants. The initial approach to achieve
these objectives was to develop an efficient but somewhat less selective ex-
traction procedure followed by an extract cleanup procedure to remove inter-
fering coextractants and produce an extract of sufficient quality for GC/MS
analysis.
This section describes the development of a preliminary method for ex-
tractable compounds in POTW sludge based on relatively expedient adaptations
of the wastewater screening protocol.. Following the development of a workable
preliminary method, several alternative extraction and extract cleanup proce-
dures were investigated and a capillary GC/MS method was developed. The re-
sults of these experiments led to the development of a revised protocol for
extractable compounds. The precision and accuracy of the revised method were
evaluated for five samples of POTW and industrial wastewater treatment sludge,
and the labor- and instrument-hours required for conducting analyses with this
protocol were estimated.
DEVELOPMENT OF A PRELIMINARY PROTOCOL FOR POTW SLUDGE
Under the industrial wastewater protocol, wastewater samples are sequen-
tially extracted with dichloromethane under basic and then acidic conditions.
The base/neutral (B/N) and acid extracts are individually concentrated and
analyzed by individual packed column GC/MS procedures. The B/N extract con-
tains the benzidenes and neutral compounds and the acid extract contains the
phenolic compounds. The analytical scheme for wastewaters is illustrated in
Figure 24.
62
-------
Wastewater
(1.0/)
Extract
Wastewater
Wastewater
Discard
Extract 3X
with CH2CI2
Adjust to pH^ 2
with 6M HCI
Extract 3X
with CH2CI2
Adjust to pH£ 11
with 6N NaOH
Determine Phenols
by GC/MS
on SP-1240-DA
Determine Base/Neutrals
& Pesticides by GC/MS
on SP-2250
Figure 24. Analysis scheme for extractable compounds
in industrial wastewater.
63
-------
The performance of these GC/MS procedures for analyses of B/N and acidic
compounds has been well demonstrated. Hence, the primary objective in adapt-
ing the wastewater methods was to develop procedures for sludge extraction
and extract cleanup to provide extracts of sufficient quality for GC/MS analy-
sis. Many of the extractable compounds were expected to strongly associate
with the sludge solids. Hence, it was anticipated that the wastewater extrac-
tion method, i.e., simple liquid-liquid partitioning with dichloromethane,
would not provide sufficient contact of the extracting solvent with the solids
to allow efficient extraction of those compounds from sludges and would be
hindered by formation of emulsions. A procedure using a high speed homoge-
nizer probe to provide vigorous mixing and blending of the sludge aliquot with
the extracting solvent was evaluated. The homogenized mixture was separated
by centrifugation and the extract withdrawn with a pipette.
Unfortunately, this vigorous homogenization/centrifugation procedure also
extracts large quantities of lipids, fatty acids, and other high molecular
weight compounds present in POTW sludges. These compounds can cause signifi-
cant interferences during GC/MS analysis and necessitate extract cleanup. A
preparative scale, low pressure, gel permeation chromatography (GPC) proce-
dure was selected for removing these high molecular weight materials from
sludge extracts. GPC techniques have been successfully applied to the removal
of lipid materials from fish extracts prepared for analysis of chlorinated
pesticides.6 Kloepfer and Bunn7 modified the procedure used for pesticides
in fish by developing a dichloromethane elution method for sludge extracts.
A scheme for the analysis of extractable compounds in sludge using basic
then acidic homogenization/centrifugation extraction and GPC cleanup is illus-
trated in Figure 25. In an attempt to further simplify a preliminary method
for extractable compounds, a single pH extraction scheme was proposed which
would be based on the recoveries for selected basic, neutral, and acidic com-
pounds from water at several pHs. Evaluations of the two proposed analytical
schemes are described below.
Evaluation of a Dual pH Extraction Scheme
The dual pff (basic then acidic) extraction scheme was evaluated by deter-
mining recoveries for the extractable analytes spiked into primary and second-
ary sludge from the Belmont POTW (Indianapolis, Indiana) and combined sludge
from the Muddy Creek POTW (Cincinnati, Ohio).
s Stalling, D. L., R. C. Tindle and J. L. Johnson. Cleanup of Pesticides
and Polychlorinated Biphenyl Residues in Fish Extracts by Gel Permeation
Chromatography. JAOAC 55:32-38 (1972).
7 Kloepfer, R. and V. Bunn, Region VII, Environmental Protection Agency
Kansas City, personal communication (1979).
64
-------
Sludge
{320 ml)
Sludge
Sjudge
Discard
Dry with NC12SO4
Dry with Na2S04
Adjust to pH< 2
with 6M HCi
Adjust to pH£ 11
with 6N NaOH
Determine Phenols
by GC/M5
on SP-1240-DA
Clean Up by GPC on
Bio Beads SX-3 Eluted
with CH2CI2
Determine Base/Neutrals
& Pesticides by GC/MS
on SP-2250
Clean Up by GPC on
Bio Beads SX-3 Eluted
with CH2CI2
Extract 3X with CH2CI2
by Homogenization/
Centrifugation
Extract 3X with CH2Ci2
by Homogenization/
Centrifugation
Figure 25. Dual pH extraction scheme for extractable compounds
in POTW sludges.
65
-------
Experimental—
Each 320-ml sludge sample was divided into four 80-ml aliquots in four
200-ml glass centrifuge screw-cap tubes. One sample for each sludge type was
spiked by adding the extractable spiking compounds (in acetone) to each sludge
aliquot so as to achieve a spike concentration of 310 |Jg/liter. The tubes
were sealed with TFE-lined screw caps and were held at 4°C overnight, with
tumbling, to allow equilibration of the spikes. Spiked and unspiked sludge
aliquots were extracted with three portions of dichloromethane as follows.
Each sludge aliquot was adjusted to pH 11 with 6 N sodium hydroxide. Sample
pH was determined using indicating test strips. An 80-ml portion of dichloro-
methane was added to each centrifuge tube. The contents of the tube were
homogenized with a Tekmar Tissumizer® blending probe for 1 rain and then cen-
trifuged at 3,000 rpm for 30 min. This produced three phases consisting of a
fairly clear aqueous layer on top, a dark solvent extract on the bottom, and
a firm mat of solids at the solvent-water interface. The lower solvent layer
was removed with a 50-ml pipette and the extraction repeated twice. All ex-
tracts from the four tubes constituting a single sample were combined. These
base neutral sludge extracts were dried by passage through a short column of
anhydrous sodium sulfate" prior to concentration to 5 to 25 ml in Kuderna-
Danish evaporators. The lowest practicable volumes were achieved while avoid-
ing excessive viscosity or solidification of the extracts. The sludge ali-
quots were then adjusted to pH 1 with 6 N hydrochloric acid, and the extrac-
tion, extract drying, and extract concentration steps were repeated to produce
the acidic sludge extracts.
The B/N and acidic extracts were cleaned by GPC on a 2.5-cm ID glass
column packed with 50 g of Bio-Beads SX-3 and eluted with dichloromethane.
The eluent flow was 5.0 ml/min with a column pressure of 350 to 700 millibars
(5 to 10 psi). The column eluent passed through a 254-nm ultraviolet detector
to provide some elution information. The column was calibrated by chromato-
graphing 1.0-g aliquots of corn oil and 100-jjg portions of bis(2-ethylhexyl)-
phthalate and pentachlorophenol in 5.0 ml of dichloromethane. The first 100 ml
eluted from the column was discarded, the second 150 ml was collected as the
cleaned extract, and the final 100 ml for each run was also discarded. The
"dump," "collect," and "wash" fractions were selected to provide i 85% removal
of the corn oil in the dump fraction, £ 85% recovery of the phthalate, and
100% recovery of the pentachlorophenol in the collect recovery fraction. The
wash fraction, 100 ml of elution prior to the next injection, was employed to
prevent carry-over. The sludge extracts were chromatographed in one or more
5.0 ml injections. The combined cleaned extract fractions were concentrated
to 1 to 5 ml for GC/MS analysis. Extracts that were still highly colored were
cleaned by a second pass through the GPC column.
Cleaned by extraction with dichloromethane, drying, and then heating to
650°C for 2 h. The cleaned material was stored at 116°C until just
prior to its use.
66
-------
The B/N and acidic extracts were analyzed by GC/MS according to the pro-
cedures described in the industrial wastewater protocol. The GC column used
for analyses of B/N extracts, a 1.8 m x 2 mm ID glass column packed with 1%
SP-2250 on 100/120 mesh Supelcoport, was programmed from 50 to 260°C at 10°C/
min following a 4-min initial hold. The column used for analysis of acidic
extracts, al.2mx2mmID glass column packed with 1% SP-1240-DA on 100/120
mesh Supelcoport, was programmed from 85 to 200°C at 10°C/min following a
4-min initial hold. The final column temperature was maintained for 10 min.
Both columns were eluted with helium at 30 ml/min. The base/neutral and
acidic compounds were quantitated by comparisons with standards prepared from
the spiking solutions. Analyte concentrations were calculated using internal
standard correction as follows:
B^g = area of internal standard peak in standard
Vj = volume of extract injected (pi)
V„ = total volume of extract (ml)
h
N = nanograms in standard
Vg = volume of sludge extracted (liter)
F = fraction of extract cleaned by GPC for analysis (e.g.,
pg/liter of sludge
where: A = area of peak in sample extract
B = area of peak in standard
Ajg = area of internal standard peak in sample extract
20 ml
= .80).
25 ml
67
-------
Results and Discussion—
The results of recovery determinations for sludges spiked with the ex-
tractable spiking compounds and analyzed with the dual pH extraction scheme
are shown in Table 20. Although the recoveries observed for most compounds
were fairly good, several compounds were not recovered from the primary sludge.
These compounds were hexachloroethane, benzidene, 3,3'-dichlorobenzidene, and
pentachlorophenol. Part of the loss of hexachloroethane can be attributed to
volatilization during extract concentration. Dilution of the primary sludge
extracts to reduce the concentration of interfering coextractants also reduced
the spike levels to at or near the detection limit. The zero recoveries ob-
served for bis(2-chloroethyl)ether and N-nitrosodimethylamine reflect, in part,
their characteristically poor chromatography.
Evaluation of a Single pH Extraction Scheme
Determination of the Recoveries of Selected Extractable Compounds
at Various pHs--
Although the dual pH extraction scheme evaluated in the preceding sec-
tion provided fairly good recoveries for the B/N and acidic compounds, the
extraction procedure was time-consuming and labor-intensive. In an attempt
to economize the effort and time required for sludge analyses, the feasibility
of developing a single pH extraction scheme was investigated. An optimum pH
for extraction of all of the B/N and acidic compounds was determined from the
recoveries observed for representative compounds spiked into clean water and
extracted at various pH levels. The optimum pH was then utilized in recovery
determinations for the same compounds from spiked sludge.
Experimental--Separate 1.0 liter aliquots of glass-distilled water,
spiked with 500 pg/liter each of phenol, 2,4-dinitrophenol, and pentachloro-
phenol were adjusted to pH 2, 3, 4, and 5 and extracted serially with three
portions of dichloromethane (200, 100, and 100 ml). Additional distilled
water aliquots were spiked with 500 |jg/liter each of anthracene, benzidine,
and 3,3'-dichlorobenzidine and similarly extracted at pH 2, 3, 4, and 11.
The extracts were dried by passage through a short column of anhydrous sodium
sulfate, concentrated to 1.0 ml in Kuderna-Danish evaporators, and analyzed
by GC/FID. The chromatographic columns and conditions used were the same as
described for GC/MS analysis of sludge extracts in the preceding section.
Results and discussion—The results of these recovery determinations are
shown in Table 21. Recoveries for benzidine were very low at pH 2 and 3 but
better at pH 4 and 11. Pentachlorophenol and 2,4-dinitrophenol recoveries
were good at pH 2, 3, and 4 but were lower at pH 5. Anthracene was inexplic-
ably recovered at only 54% at pH 4. Based on these results, the best compro-
mise pH for extraction of the entire range of acidic and basic compounds
appeared to be pH 4.
68
-------
TABLE 20. RESULTS OF EVALUATION OF A DUAL pH EXTRACTION SCHEME (BASIC THEN ACIDIC)
FOR THE ANALYSIS OF BASE/NEUTRALS AND ACIDS FROM POTW SLUDGES
Primary s
ludge3
Secondary sludge
Combined
sludge
Spike
Unfortified
Spike
Unfortified
Spike
Unfortified
Spike
Level
conc.
recovery
conc.
recovery
conc.
recovery
Compound
(pg/L)
(mr/l)
(%)
(pg/L)
(%)
(mr/l)
(%)
1,4-Dichlorobenzene
310
NDb
150
ND
43
ND
78
Hexachloroethane
310
ND
0
ND
70
ND
150
Bis(2-chloroisopropylether
310
ND
110
ND
56
ND
100
Bis(2-chloroethyl)ether
310
ND
0
ND
0
ND
0
Acenaphlhylene
310
ND
170
ND
75
ND
110
2,6-Dinitrotoluene
310
ND
94
ND
81
ND
110
Phenanthrene
310
3,600
-
ND
-
460
-
Fluoranthene
310
ND
320
ND
59
180
140
Benzidene
310
ND
0
ND
39
ND
34
3,3'-Dichlorobenzidene
310
ND
0
ND
67
ND
130
Bis(2-ethylhexyl)phthalate
310
3,500
68
420
66
630
330
Benz[aJpyrene
310
94
270
7.7
80
74
150
N-Nitrosodimethylamine
310
ND
0
ND
0
ND
0
Phenol
310
400
180
ND
130
200
62
2,4-Dichlorophenol
310
ND
130
ND
110
ND
27
Pentachlorophenol
310
ND
0
ND
80
180
100
a The primary sludge extracts were analyzed at a 1:5 dilution relative to other extracts because
of high levels of interfering materials.
b ND = not detected.
c Phenanthrene was not incLuded in the spikes.
-------
TABLE 21. RECOVERY OF SELECTED SEMIVOLATILE PRIORITY POLLUTANTS
EXTRACTED AT SELECTED pHs
Compound
Spike level
(|Jg/L)
pH 2
Extraction recovery (%)
pH 3 pH 4 pH 5
pH 11
Phenol
500
39
46
41
46
NAa
2,4-Dinitrophenol
500
95
95
112
44
NA
Pentachlorophenol
500
100
96
107
85
NA
Anthracene
500
73
89
54
NA
92
Benzidene
500
2
2
44
NA
73
3,3'-Dichlorobenzidene
500
70
87
71
NA
85
a NA = not analyzed.
70
-------
Determination of the Recoveries of Selected Extractable Compounds
from Sludge--
Experiniental—Aliquots (320 ml) of primary sludge from the Platte County
POTW and digested sludge from the Little Blue River POTW were prepared and
spiked in the same manner as aliquots prepared for dual pH extraction, de-
scribed in the preceding section. Triplicate spiked aliquots were prepared
for each sample. The aliquots were buffered at pH 4 with 5 ml of a 0.1-M
potassium hydrogen phthalate solution and extracted with dichlororaethane by
homogenization/centrifugation. The extracts were cleaned and concentrated by
the procedures previously described for the dual pH extraction scheme. B/N
and acidic compounds were determined in the extracts by GC/MS on both column
systems.
Results and discussion--The results of recovery determinations for spiked
sludge using the single pH extraction scheme are shown in Table 22. Recover-
ies were not calculated for bis(2-ethylhexyl)phthalate since it was present
in the unspiked sludges at concentrations much higher than the spike level.
The recoveries observed for most compounds are fairly good. However, the
benzidines and chloroalkylethers were not recovered from any of the spiked
sludge aliquots.
Overall, the dual pH extraction scheme gave somewhat better recoveries
than those obtained with the single pH scheme. In particular, the phenols,
benzidines, and chloroalkyl ethers were recovered better from spiked sludges
extracted at basic and then acidic pH. Although the single pH procedure was
considerably faster to perform, the dual pH scheme illustrated in Figure 25
was selected for the preliminary POTW sludge protocol4 based on the results
of the recovery determinations. Although the priority pollutant pesticides
were not included in the evaluation experiments, they are typically well-
behaved neutral compounds. As such, in the preliminary protocol, the pesti-
cides were considered as a subset of the B/Ns.
ADDITIONAL METHODS DEVELOPMENT FOR MUNICIPAL AND INDUSTRIAL
WASTEWATER TREATMENT SLUDGE
The combination of homogenization/centrifugation extraction and GPC ex-
tract cleanup described in the preliminary POTW sludge protocol provided satis-
factory recoveries for the B/N and acidic compounds spiked into POTW sludge.
However, the rigorous extraction procedure was time-consuming, labor-intensive,
and not very selective. The large quantities of lipids and other high molec-
ular weight biogenic compounds in many crude extracts exceeded the capacity
of the GPC cleanup procedure. Although the GPC method was designed to remove
^ 85% of up to 1 g of high molecular weight biogenic compounds, multiple GPC
runs were required for many extracts. The multiple runs were done as sequen-
tial 5.0-ml injections of crude extracts or by rechromatographing the cleaned
extracts. Some extracts that were cleaned twice and concentrated to 1.0 ml
still could not be analyzed by GC/MS because of severe interferences. The
time and cleanup problems were partially reduced for the EPA survey of POTW
71
-------
TABLE 22. RESULTS OF EVALUATION OF A SINGLE pH EXTRACTION SCHEME FOR THE ANALYSIS
OF BASE/NEUTRALS AND ACIDS FROM POTW SLUDGES
Primary s
ludge
Secondary
sludge
Combined
sludge
Spike
Unfortified
Spike
Unfortified
Spike
Unfortified
Spike
level
conc.
recovery
conc.
recovery
conc.
recovery
Compound
(Mft/L)
(|Jfc/L)
(%)
(Mg/L)
(%)
(Mg/L)
(%)
1,4-Dichlorobenzene
310
130
170
+
44a
12
110
+
16
29
140 ± 21
Hexa ch1o roe tha ne
310
ND
70
+
13
ND
57
+
22
ND
41 ± 12
Bis(2-chloroisopropyl)ether
310
ND
0
ND
0
ND
0
Bis(2-chloroethyl)ether
310
ND
0
ND
0
ND
0
Acenaphthylene
310
ND
78
+
3
ND
89
+
17
6.3
100 ± 30
2,6-Dinitrotoluene
310
ND
54
+
20
ND
68
+
11
15
37 ± 14
Phenanthrene
310
19
81
+
6
5.9
88
+
9
15
72 ± 5
Fluoranthene
310
19
75
+
14
7.9
67
+
20
4.9
58 ± 20
Benzidine
310
ND
0
ND
0
ND
0
3,3'-Di chlorobenzidene
310
ND
0
ND
0
ND
0
Bis(2-ethylhexyl)phthalate
310
3,100
c
1,500
c
2,500
c
Benzo[a]pyrene
310
7.5
80
+
25
3.7
90
±
53
7.5
100 ± 15
N-Nitrosodimethylamine
310
ND
0
ND
0
ND
0
Phenol
310
91
88
+
78
35
35
+
6
240
89 ± 32
2,4-Dichlorophenol
310
ND
67
+
40
2.0
37
+
4
ND
70 ± 19
Pentachlorophenol
310
30
110
+
76
21
45
+
5
22
89 ± 35
a Mean ± standard deviation for three determinations,
b ND = not detected.
c Recovery calculations are not representative since levels in the unfortified sludges
were much higher than the spiked levels.
-------
emissions by extracting a single 80-ml sludge aliquot for each sample.8 This
required only one centrifuge tube per sample and produced 25% of the inter-
fering coextractants in the crude extract. And, since the extracts from 80 ml
aliquots could be concentrated further following cleanup, the method sensi-
tivity was essentially the same. The experiments described below were de-
signed to evaluate alternate procedures to allow more efficient and selective
extraction, to evaluate alternate extract cleanup methods appropriate for the
best extraction procedure, and to develop capillary GC/MS methods to aid in
the determination of the extractable compounds in complex sludge extracts.
Extraction Methods
Three alternate extraction procedures were evaluated for sludges in an
attempt to develop a more efficient, selective, and less time-consuming ex-
traction method. The procedures evaluated include: continuous liquid-liquid
extraction, steam distillation, and microextraction.
Continuous Liquid-Liquid Extraction—
A continuous liquid-liquid extraction (CLLE) apparatus, previously de-
signed by MRI, has been successfully used for extracting industrial waste-
waters for analyses of base/neutral and acidic organic priority pollutants.
In addition to allowing unattended extraction, the continuous method allowed
extraction of wastewaters containing high levels of dissolved and suspended
solids without formation of emulsions. The apparatus, shown in Figure 26,
was designed for use with heavier-than-water solvents, e.g., dichloromethane.
This system was evaluated for the extraction of diluted sludge aliquots. The
sludge aliquots were diluted so that the suspended solids contents were sim-
ilar to those typical of high solids wastewaters. The effects of selected
operating parameters pertinent to adapting the CLLE procedure for sludges were
investigated. These parameters were the extraction time, the sludge aliquot
dilution ratio, and the addition of methanol. Methanol was added to encourage
desorption of compounds that associate strongly with the sludge solids. The
desorbed compounds would then be more easily extracted.
Extraction time--Aliquots of unspiked combined sludge were extracted for
10 and 20 h to investigate the time required for most efficient extraction.
Experimental--The basic CLLE procedure is described as follows.
The apparatus was assembled with the glass drip tip adjusted to a level near
the top of the sample compartment and the stopcock closed. To prevent sus-
pended solids from being carried into the boiling flask by the recycling sol-
vent, a plug of precleaned glass wool was placed in the bottom of the sample
chamber. Half of the dichloromethane, 200 ml, was loaded into the sample
compartment, and the other half was loaded into the boiling flask. The sam-
ple, consisting of 1.0 liter of water or a sludge aliquot diluted to 1.0 liter,
was added to the sample compartment and the stopcock was opened. The drip
tip was lowered until the solvent level just entered the tip and was secured
at that level with the two bored-through TFE unions. The extraction
5 Hansen, E. M. Midwest Research Institute, personal communication (1979).
73
-------
6 mm
Glass Rod
TFE Reducing Union,
10-6mm Bored Through at 6mm
5 24/40
I 45/50
Sample
Compartment
70 mm
Glass Drip Tip
500 mm
TFE Reducing Union, t0-6mm.
Bored Through at 6 mm
I 24/40
TFE Stopcock
TFE Reducing Union, 10-6mn
250 ml
TFE Tubing, 6mm O.D.
Figure 26. Continuous liquid-liquid extractor designed by MRI.
74
-------
was initiated by heating the boiling flask. The vaporized solvent was con-
densed in the water-cooled condenser, dripped through the sample, and col-
lected at the bottom of the sample chamber. The return tube maintained the
solvent in the sample chamber at the original level. At the end of the ex-
traction, the extracts were removed, dried by passage through a short column
of precleaned anhydrous sodium sulfate, concentrated to 1.0 ml in Kuderna-
Danish evaporators, and analyzed by the GC/MS procedures described in the
preliminary POTW sludge protocol.
For this particular experiment, two 100-ml aliquots of unspiked com-
bined sludge (2.1% solids) were adjusted to pH 11 with 6 N sodium hydroxide
and extracted for 10 and 20 h, respectively.
Results and discussion--The results of analyses of unspiked sludge
extracts prepared from 100-ml aliquots diluted to 1.0 liter and extracted at
pH 11 for 10 and 20 h are shown in Table 23. The concentrations observed
were essentially the same for all compounds identified except bis(2-ethylhexyl)-
phthalate. The difference in phthalate concentrations is likely due to spuri-
ous contamination. From these results, it was concluded that 10 h was a suf-
ficient extraction time.
Extract dilution—
Experimental—Two sludge aliquots, 100 and 300 ml, were spiked with
the B/N spiking compounds and extracted at pH 11 as described in the preceding
experiment. The extracts were analyzed by the GC/MS procedures described in
the preliminary POTW sludge protocol.
Results and discussion—The influence of sludge dilution on extrac-
tion efficiencies may be seen from the spike recoveries shown in Table 24.
The recoveries from the 100-ml aliquot were generally higher than from the
300-ml aliquot. During the course of extracting sludge aliquots, the suspended
solids settle to the water-solvent interface. The solids tend to settle faster
in the less diluted aliquot. Hence, it is likely that the sludge solids were
not extracted as efficiently in that aliquot.
Addition of methanol--Aliquots of spiked water and combined sludge were
extracted with and without the addition of methanol.
Experimental—Two 500-ml aliquots of glass-distilled water and two
100-ml aliquots of combined sludge were spiked with the extractable spiking
compounds and were allowed to equilibrate at 4°C for 4 h. One spiked water
aliquot and one spiked sludge aliquot were fortified with 200 ml of glass-
distilled methanol. All water and sludge aliquots were diluted to 1.0 liter,
adjusted to pH 11 with 6 N sodium sulfate, and extracted for 10 h. The B/N
extracts were removed and the systems were charged with 400 ml of fresh sol-
vent. The samples were then adjusted to pH 1 with 6 N hydrochloric acid and
extracted an additional 10 h. The resulting B/N and acidic extracts were
dried and analyzed by the preliminary POTW protocol GC/MS procedures.
75
-------
TABLE 23. CONCENTRATIONS MEASURED IN UNSPIKED COMBINED
SLUDGE EXTRACTS BY CONTINUOUS LIQUID-LIQUID
EXTRACTION: EFFECT OF EXTRACTION TIME
Compound
Concentration (|Jg/L
sludge)
20-h
extraction
10-h
extraction
1,4-Dichlorobenzene
12
11
Acenaphthylene
46
39
Fluoranthene
12
15
Pyrene
13
19
Bis(2-ethylhexyl)phthalate
650
2,000
Benzofluoranthenes
6
7
Benzo[a]pyrene
6
5
a Concentrations represent suras for these compounds, which
elute simultaneously and have the same major ions for
GC/MS.
76
-------
TABLE 24. RECOVERIES OF EXTRACTABLE COMPOUNDS FROM SPIKED
COMBINED SLUDGE BY CONTINUOUS LIQUID-LIQUID
EXTRACTION: EFFECT OF SLUDGE DILUTION
Compound
Spike level
(pg/L sludge)
Recovery (%)
100-ml aliquot3 300-ml aliquot3
1,4-Dichloroben.zene
315
33
57
Hexachloroethane
317
98
46
Acenaphthylene
318
88
35
2,6-Dinitrotoluene
319
121
81
Fluoranthene
274
33
25
Benzidene
310
37
69
3,3'-Dichlorobenzidine
364
68
43
n-Butylbenzylphthalate
358
20
25
Bis(2-ethylhexy1)phthalate
336
57
6
Benzo[ajpyrene
312
20
26
a Diluted to 1.0 L prior to extraction.
77
-------
Results and discussion--The results of recovery determinations for
spiked water and sludge aliquots extracted with and without the addition of
methanol are shown in Table 25. The recoveries observed from the spiked
sludge were generally lower than from the spiked water. The addition of
methanol had little influence on recoveries from water or sludge. Several
compounds, e.g., polynuclear aromatic compounds, were very poorly recovered
from spiked sludge. These compounds are likely to strongly associate with
suspended solids. Some form of mechanical agitation during extraction may
improve solids-solvent contact and increase extraction efficiencies for these
compounds.
Extractive Steam Distillation—
Extractive steam distillation has been shown to be an efficient and
selective procedure for concentrating chlorinated pesticides from water, soil,
sediment, and tissue samples. Veith and Kiwus9 developed an apparatus which
provided extraction of the steam distillate with a lighter-than-water solvent.
A distillation/extraction head is marketed by Kontes (Catalog No. K-523010)
which allows continuous extraction of the distillate with a heavier-than-water
solvent. This continuous extractive steam distillation system was evaluated
for the extraction of sludge distillate with dichloromethane. The apparatus
utilized, fashioned after the Kontes unit, is shown in Figure 27. The sample
is boiled and the condensed distillate is extracted in a manner similar to
the CLLE unit described previously. This system was initially evaluated with
spiked water with extraction times from 2 to 72 h. The optimum extraction
time as determined from these experiments was used in extraction studies with
spiked combined sludge.
Extraction time—The optimum time for efficient extractive steam distil-
lation was investigated by extracting aliquots of spiked water for 2 to 72 h.
Experimental—Aliquots of glass-distilled water (500 ml) were spiked
with selected neutral or acidic compounds. Aliquots spiked with neutral com-
pounds were adjusted to pH 7 and transferred to the sample flask. The solvent
flask was charged with 300 ml dichloromethane and the contents of both flasks
were heated to boiling. Individual neutral aliquots were extracted for 2, 6,
15, or 72 h. Aliquots spiked with acidic compounds, adjusted to pH 1, and
similarly extracted for 2, 6, 12, or 48 h. All extracts were dried with an-
hydrous sodium sulfate columns, concentrated to 1.0 ml, and analyzed by GC/FID
using the gas chromatographic procedures described in the preliminary POTW
sludge protocol.
s Veith, G. D. and L. M. Kiwus. An Exhaustive Steam Distillation and
Solvent Extraction Unit for Pesticides and Industrial Chemicals.
Bull. Environ. Contain. Toxicol. 17:631-36 (1977).
78
-------
TABLE 25. RECOVERIES OF EXTRACTABLE COMPOUNDS FROM SPIKED DISTILLED WATER AND SPIKED COMBINED
SLUDGE BY CONTINUOUS LIQUID-LIQUID EXTRACTION: EFFECT OF METHANOL (MeOH) ADDITION
Combined
sludge
Distilled
water
Without
MeOH
With
MeOH
Without
MeOH
With
MeOH
Spike
Spike
Spike
Spike
Spike
Spike
level
Spike
level
Spike
level
recovery
level
recovery
(pg/L
recovery
(Mg/L
recovery
Compound
O'g/L)
(%)
(Wg/L)
(%)
sludge)
a)
sludge)
(%)
1,4-Dichlorobenzene
315
113
315
119
312
32
315
38
Hexachloroethane
317
85
317
100
314
6
317
5
Bis(2-chloroisopropyl)-
318
110
318
113
364
69
364
87
ether
Bis(2-chloroethyl)ether
319
259
319
248
316
115
338
145
Acenaphthylene
318
69
318
73
315
3
318
11
2,6-Dinitrotoluene
319
120
319
118
316
51
319
60
Fluoranthene
274
73
274
77
271
8
274
8
Benzidine
310
60
310
60
316
10
316
52
3,3' -Diclilorobenzidine
364
101
364
116
371
35
371
44
n-Butylbenzylphthalate
358
34
358
25
354
14
358
14
Bis(2-ethylhexyl)-
336
44
336
52
333
a
336
a
phthalate
Benzo[a]pyrene
312
67
312
112
309
8
312
5
Phenol
335
99
341
69
344
102
341
94
2,4-Dimetliylphenol
351
4
341
13
355
19
359
27
2,4-Dichlorophenol
296
57
292
75
335
62
338
57
Pentachlorophenol
334
0
337
220
344
0
347
14
a Recovery not determined since the spike level was much lower than present in the unspiked sample.
-------
°ir
fRx)
jdn)
v
Sludge
Dichloromefhane
Figure 27. Continuous extractive steam distillation apparatus,
80
-------
Results and discussioa--The results of the extraction time study
are shown in Table 26. Maximum recoveries were achieved for most compounds
except phenol after only 2 h of extraction. However, since the phenol re-
covery was higher for 6-h extractions, subsequent steam distillation experi-
ments were conducted using 6 h of extraction. The low recoveries observed
for 1,4-dichlorobenzene (21 to 27%) were likely due to volatility losses.
Higher recoveries were observed in subsequent experiments when the condenser
cooling water was chilled.
Determination of recoveries from spiked water and sludge--
Experimental--A 500-ml aliquot of glass-distilled water, spiked with
the extractable spiking compounds, and a similar aliquot of unspiked water
were basified to pH 11 and extracted for 6 h with 300 ml of dichloromethane.
The extracts were removed and the water was acidified to pH 1 before extracting
with fresh solvent. Two 300-ml aliquots of combined sludge, one spiked and
one unspiked, were similarly extracted. All water and sludge extracts were
dried, concentrated, and analyzed by GC/MS according to the preliminary POTW
protocol.
Results and discussion--The results of recovery determinations for
spiked compounds in water and sludge extracted by steam distillation are shown
in Table 27. Recoveries observed for the benzidenes, butylbenzylphthalate,
and 2,4-dimethylphenol were low from both water and sludge. The poor recover-
ies for 2,4-dimethylphenol were surprising in light of the 54 to 67% recoveries
observed in the preceding experiment. This particular compound was frequently
and inexplicably observed to have poor recoveries throughout the course of
this program. The high recoveries observed for the chloroalkyl ethers was
likely due to the difficulties inherent in measuring the response of the broad
chromatographic peaks characteristic of the compounds. Nonetheless, the spike
recoveries from combined sludge were considerably lower than from water for
many compounds.
The probability that a particular compound can be efficiently steam
distilled from aqueous media is directly related to its vapor pressure at the
boiling point of the sample. Vapor pressure data are presented for selected
priority pollutants and other compounds in Table 28. These data were estimated
with the Antoine equation or taken from literature sources. The Antoine equa-
tion is:
log10 P = A - B/(C + T)
or
logio P =
-52.23B
(T + 273)
+ C
81
-------
TABLE 26. RECOVERIES OF EXTRACTABLE COMPOUNDS FROM SPIKED DISTILLED
WATER SAMPLES BY EXTRACTIVE STEAM DISTILLATION
Compound
Spike
level
(ur/l)
2
6
% Recovery
Time (h)
12 15
48
72
1,4-Dichlorobenzene
100
27
25
NAa
21
NA
22
Bis(2-chloroethyl)ether
100
36
35
NA
35
NA
32
2,6-Dinitrotoluene
100
100
99
NA
95
NA
74
Fluoranthene
100
81
88
NA
75
NA
72
Bis(2-ethylhexyl)phthalate
100
81
94
NA
80
NA
101
Phenol
100
42
71
73
NA
69
NA
2,4-Dimethylphenol
100
56
54
62
NA
67
NA
2,4-Dichlorophenol
100
76
61
131
NA
46
NA
Pentachlorophenol
100
99
106
125
NA
95
NA
a NA = Not analyzed.
82
-------
TABLE 27. RECOVERIES OF EXTRACTABLE COMPOUNDS FROM SPIKED DISTILLED WATER
AND SPIKED SLUDGE SAMPLES BY EXTRACTIVE STEAM DISTILLATION
Distilled water Combined sludge
Spike Spike
Compound
Spike level
(Mr/L water)
recovery
ra
Spike level
(yg/L sludge)
recove
(%)
1,4-Dichlorobenzene
315
102
630
103
Hexa chlo roe thane
317
106
634
0
Bis(2-chloroisopropyl)ether
386
143
772
154
Bis(2-chloroethyl)ether
362
176
492
258
Acenaphthylene
318
96
636
97
2,6-Dinitrotoluene
319
129
638
0
Fluoranthene
274
102
548
43
Benzidine
310
0
620
0
3,3'-Dichlorobenzidine
364
0
728
0
n-Butylbenzylphthalate
358
0
716
4
Bis(2-ethylhexyl)phthalate
336
71
672
17
Benzo[a]pyrene
316
75
624
3
Phenol
338
33
676
a
2,4-Dimethylphenol
348
3
696
0
2,4-Dichlorophenol
328
47
606
16
Penta chlo ropheno1
337
74
662
79
a Recovery not estimated since the concentration of phenol in the unspiked
sludge was larger than the concentration in the spiked sludge.
83
-------
TABLE 28. VAPOR PRESSURE DATA FOR EPA PRIORITY POLLUTANTS AND
SELECTED ALIPHATIC HYDROCARBONS
Compound
p
vap (on He)
Literature Ref.
Extractable Spiking Compounds
1,4-Dichlorobeazene
77.7
(100°C)
a
67
(100°C)
b
Hexachloroethane
40
(102.3°C)
c
Bis(2-chloroethyl)ether
60
(101.5°C)
c
Acenaphthylene
10'
*3-io"2
(20#C)
d
Fluoranthene
10'
*«-io"4
(20°C)
d
Bis(2-ethylhexyl)phthalate
0.07
(150°C)
e
Benzo(a]pyrene
5
x 10"9
(25°C)
d
Phenol
41.3
(100°C)
i
2,4-Dimethylphenol
10
(91.3°C)
c
20
(105°C)
c
2,4-Dichloropheno1
10
(93°C)
c
20
(108°C)
c
Pentachlorophenol
0.26
(100°C)
a
Other EPA Priority Pollutants
1,3-Dichlorobeozene
60
(92°C)
c
1,2-Dichlorobenzene
60
(99.5°C)
c
1,2,4-Trichlorobenzene
20
(97.2-C)
c
Naphthalene
20
(101.7°C)
c
Kitrobenzene
20
(99.3°C)
c
2-Chloronaphthalene
5
(104.8°C)
c
Isophorone
20
(96.S°C)
c
Fluorene
5
(129.3°C)
c
Hexachlorobenzeae
1
(114.4'C)
c
Phenanthcene
1
(118.2°C)
c
Anthracene
0.07
(100°C)
a
Dimethylphthalate
1
(100.3°C)
c
2-Nitrophenol
10
(90.4°C)
c
2-Chlorophenol
60
(92°C)
c
2,4,6-Trichlorophenol
5
(105.9°C)
c
Hydrocarbons
a-Keptane
796
(100°C)
a
n-Octane
351
(100°C)
a
n-Konane
158
(100°C)
a
n-Decane
72
UQ0°C)
a
n-Pentadecane
1.6
(100°C)
a
a Lange's Handbook of Chemistry, 11th edition. McGraw-Hill, 1973.
b Kirk-Othaer Encyclopedia of Chemical Technology, 2nd edition. John Wiley
and Sons, 1964.
c Chemical Engineer's Handbook, 5th edition. Peiry and Chilton, Eds.
McGraw-Hill, 1973.
d U.S. Environmental Protection Agency. Water-Related Environmental Fate
of 129 priority Pollutants, Vol. II. EPA-440/4-79-029»,b,
December 1979.
e Kirk-Othmer Encyclopedia of Chemical Technology, 1st edition. John Wiley
and Sons, 1953.
84
-------
where A , B , and C are constants characteristic of the compound and T
is the temperature in degrees Celsius. From these data, it is not surprising
that extractive steam distillation recoveries for benzo[a]pyrene are low.
Vapor pressure data were not available for the benzidenes; however, benzidene
boils at 400°C, so the vapor pressures for benzidene and 3,3'-dichlorobenzi-
dine are likely to be very low at 100°C and their potentials to steam distill
are thus not favorable.
Although the concentrated steam distillation extracts from the com-
bined sludge were very clear and were analyzed without cleanup, they contained
high levels of aliphatic hydrocarbons. This is not surprising considering
that the vapor pressures for the alkanes listed in Table 28 are high relative
to the extractable priority pollutants.
Microextraction--
Successful liquid-liquid extraction of aqueous samples with a small vol-
ume of solvent was reported by Rhoades and Millar.10 This method has been
adapted to allow simple and rapid extraction of organic priority pollutants
from a variety of industrial wastewaters.11 This procedure typically involves
shaking 100 ml of wastewater with 1.0 ml of hexane, diisopropyl ether, tolu-
ene, or other lighter-than-water solvent, as appropriate for the specific ana-
lytes, in a 100-ml volumetric flask. Following extraction the contents of
the flask are allowed to settle. The solvent accumulates in the constricted
neck and can be sampled with a syringe for a direct GC or GC/MS analysis.
Salt is added and/or the pH of the sample can be adjusted as required to opti-
mize recoveries. Although the procedure was designed for use with lighter-
than-water solvents, glassware can be envisioned that could allow extraction
with heavier-than-water solvents, such as dichloromethane. However, the sus-
pended solids in sludge and diluted sludge would likely hinder the recovery
of the extract from the bottom of the sample. In addition to its simplicity,
microextraction typically produces extracts that are relatively free of inter-
fering coextracted materials. The solvent to sample ratio (typically 1:100)
efficiently extracts only those compounds with favorable partitioning charac-
teristics. The interferences in many samples that are extracted by more rig-
orous procedures frequently have low partitioning coefficients but are present
in the samples in very high concentrations.
Microextraction procedures were evaluated for sludges with a few minor
modifications. Specifically, sludge samples were extracted in Babcock milk
butterfat testing tubes. Babcock tubes, shaped much like volumetric flasks
but designed to tolerate centrifugation, were used to assist extract recovery
from the extracted sludge aliquots. Following a preliminary experiment in
which spiked combined sludge was extracted with hexane, the influences of
salting and solvent selection were investigated with spiked water and com-
bined sludge.
1U Rhoades, J. W. and J. D. Millar.
tive Analysis of Fruit Flavors.
11 Rhodes, J. W. and C. P. Nulton.
communication (1979).
Gas Chromatographic Method for Compara
J. Agr. Food Chem. 13:5-9 (1965).
Southwest Rsearch Institute, personal
85
-------
Preliminary evaluations with spiked combined sludge—
Experimental—Two 15-ml aliquots of combined sludge (2.1% solids)
were transferred to Babcock tubes and spiked with selected B/N compounds.
All aliquots were diluted to 45 ml with glass-distilled water. After adjust-
ing to pH 11, a 0.5-ml aliquot of hexane was added to each tube, and the tubes
were hand-shaken for 2 min. In order to recover the extracts, the tubes were
centrifuged for 15 min at 1,500 rpm. The extracts were analyzed directly by
GC/FID using the SP-2250 packed column system described in the preliminary
POTW sludge protocol.
Results and discussion—The recoveries determined for spiked com-
pounds concentrated from combined sludge by microextraction are shown in
Table 29. Since the maximum enrichment achievable for the preliminary eval-
uations was 30-fold (0.5 ml extract from 15-ml sludge aliquot), high spike
levels were employed. Although the spike levels were rather high and some
differences were apparent between the recoveries observed for duplicates,
these results indicated that the microextraction procedure showed some promise.
Appropriate choices of extracting solvent may facilitate analysis of the en-
tire range of extractable compounds. The possible necessity of using several
extraction solvents on replicate sludge aliquots may still be preferable to a
laborious and difficult extraction and extract cleanup scheme. The extracts
appeared to be relatively free from coextracted interferences. This was evi-
denced by the successful determination of spiked compounds by GC/FID. The
GC/FID chromatograms for the microextracts of spiked and unspiked sludge shown
in Figure 28 illustrate that the spiked compounds were readily identified.
Recoveries from spiked water—Three solvents were evaluated and the in-
fluence of salting was investigated for microextraction of spiked water.
Experimental—Several aliquots (100 ml) of glass-distilled water
were extracted in 100-ml volumetric flasks. Separate aliquots were spiked
with the B/N and acidic spiking compounds and were adjusted to pH 11 or 1, as
appropriate. The water samples were extracted with 1.0-ml aliquots of hexane,
toluene, or diisopropylether. The extracts were removed with pipettes and
the process was repeated twice for a total of three extractions for the same
spiked water aliquot. An additional set of extractions was conducted in a
similar fashion for glass-distilled water saturated with sodium chloride.
All individual extracts were analyzed without concentration by GC/FID using
the chromatographic systems described in the preliminary POTW sludge protocol.
The contents of the extracts were quantitated against mixed standard solutions
prepared in the same solvent.
Results and discussion—The results of microextraction recovery
determinations with the three solvents and with or without salt are shown in
Tables 30 to 32. In general, toluene provided the best recoveries for most
compounds. Good recoveries were obtained with toluene for all compounds ex-
cept phenol. Although good recoveries were achieved with hexane for several
neutral compounds, recoveries were generally poor for the polar compounds and
for benzo[a]pyrene. Some of the polar compounds, notably bis(2-chloroethyl)-
ether, 3,3'-dichlorobenzidene, and the phenols, were extracted with diisopropyl
ether. Phenol has relatively high solubility in water, as noted in Section 7.
86
-------
TABLE 29. MICROEXTRACTION OF EXTRACTABLE COMPOUNDS
FROM SPIKED SLUDGE SAMPLES WITH HEXANE
Compound
Spike level
(iJg/L sludge)
Spike
recovery'
(7o)
1,4-Dichlorobenzene
5,250
53, 43
Hexachloroethane
5,280
52, 26
Acenaphthylene
5,300
101, 85
2,6-Dinitrotoluene
5,320
31, 35
Fluorantheae
4,570
77, 47
n-Butylbenzylphthalate
5,970
85, 54
Bis(2-ethylhexyl)phthalate
5,600
52, 74
Benzo[a]pyrene
2,000
8, 44
a Sludge samples were spiked after dilution 1:1 with
distilled water.
87
-------
iVv-v1^V
Figure 28. GC/FID chromatograms of hexane microextracts spiked (A)
and unspiked (B) combined sludge.
88
-------
TABLE 30. MICROEXTRACTION OF PRIORITY POLLUTANTS FROM
SPIKED WATER SAMPLES WITH HEXANE
Spike recovery (%)
Spike
With
salt
Without
salt
level
Extract
Extract
Compound
(Pg/L)
1
2
3
Sura
1
2
3
Sum
1,4-Dichlorobenzene
788
93
0
0
93
43
0
0
43
Hexachloroethane
792
81
0
0
81
42
0
0
42
Bis(2-chloroisopropyl)ether
324
0
0
0
0
0
0
0
0
Bis(2-chloroethyl)ether
211
45
16
4
65
83
4
0
87
Acenaphthylene
795
77
6
0
83
50
0
0
50
2,6-Dinitrotoluene
798
19
13
0
32
39
0
0
39
Fluoranthene
686
80
12
9
101
50
0
0
50
Benzidine
287
0
0
0
0
0
0
0
0
3,3 '-Dichlorobenzidine
272
28
14
17
59
75
6
3
84
n-Butylbenzylphthalate
896
53
5
0
58
60
0
0
60
Bis(2-ethylhexyl)phthalate
840
62
19
6
87
50
0
0
50
Benzo[a]pyrene
300
26
19
0
45
53
0
0
53
Phenol
338
0
0
0
0
0
0
0
0
2,4-Dimethy1phenol
348
2
2
0
4
34
15
13
62
2,4-Dichlorophenol
395
19
16
0
35
67
24
13
104
Pentachlorophenol
337
137
2
0
139
48
19
4
71
-------
TABLE 31. MICROEXTRACTION OF PRIORITY POLLUTANTS FROM
SPIKED WATER SAMPLES WITH TOLUENE
Spike recovery (%)
Spike
With
salt
Without salt
level
Extract
Extract
Compound
(Pg/L)
1
2
3
Sum
1
2
3
Sum
1,4-Dichlorobenzene
788
97
0
0
97
79
0
0
79
Hexachloroethane
792
73
0
0
73
73
0
0
73
Bis (2-cliloroisopropyl) ether
331
0
0
0
0
102
0
0
102
Bis (2-chloroetliyl)ether
209
114
13
11
138
114
0
0
114
Acenaphthylene
795
76
11
0
87
71
6
0
77
2,6-Dinitrotoluene
798
75
0
0
75
55
0
0
55
Fluoranthene
686
32
18
9
59
78
5
0
83
Benzidine
303
21
16
0
37
206
46
10
262
3,3'-Dichlorobenzidine
286
108
11
0
119
110
8
6
124
n-Butylbenzylphthalate
896
30
11
0
41
75
0
0
75
Bis(2-ethylhexyl)phthalate
840
35
17
7
59
80
7
0
87
Benzo[a]pyrene
300
9
0
0
9
65
0
0
65
Phenol
338
4
0
0
4
16
9
0
25
2,4-Dimethylphenol
348
35
18
11
64
87
27
11
125
2,4-Dichlorophenol
395
51
22
10
83
64
23
10
97
Pentachlorophenol
337
121
22
0
143
97
21
0
118
-------
TABLE 32. MICROEXTRACTION OF PRIORITY POLLUTANTS FROM
SPIKED WATER SAMPLES WITH DIISOPROPYL ETHER
Spike recovery (%)
Spike
With
salt
Without
salt
level
Extract
Extract
Compound
(Pg/L)
1
2
3
Sum
1
2
3
Sum
1,4-Dichlorobenzene
788
45
0
0
45
34
0
0
34
Hexachloroethane
792
54
0
0
54
30
0
0
30
Bis(2-chloroisopropyl)ether
334
0
0
0
0
0
0
0
0
Bis(2-chloroethyl)ether
211
54
12
19
85
90
0
0
90
Acenaphthy1ene
795
67
9
0
76
58
10
0
68
2,6-Dinitrotoluene
798
32
0
0
32
55
0
0
55
Fluoranthene
686
41
13
0
54
54
12
0
66
Benzidine
303
0
0
0
0
45
14
9
68
3,3'-Dichlorobenzidine
286
77
7
0
84
99
6
0
105
n-Butylbenzylphthalate
896
0
0
0
0
70
43
0
113
Bis ( 2-e thy lhexyl)plitha late
840
0
25
0
25
62
31
0
93
Benzo[a]pyrene
300
0
0
0
0
49
0
0
49
Phenol
338
24
19
16
59
34
37
17
88
2,4-Dimethylphenol
348
84
39
14
137
57
38
9
104
2,4-Dichlorophenol
395
140
43
9
192
58
45
11
114
Pentachlorophenol
337
68
35
20
123
61
42
31
134
-------
Since the success of microextraction depends on a very favorable partition
coefficient, high recoveries for phenol may be difficult to achieve with this
technique.
The influence of the addition of salt to the aliquots prior to
extraction varied with the compound spiked. In particular, recoveries for
1,4-dichlorobenzene were markedly higher when salt was added. Recoveries for
benzo[a]pyrene were higher from aliquots without salt. Salting decreased re-
coveries for 3,3'-dichlorobenzidine extracted with hexane but did not signif-
icantly affect its extraction with toluene or diisopropylether. Hence, a
successful microextraction method for sludges may necessitate extraction of
replicate aliquots at different pHs, with different solvents, and with or
without the addition of salt.
The largest fractions of analyte recovered by compound-solvent-salt com-
binations for which good total recoveries were achieved, were contained in
the first of the three successive extracts. Since the success of microectrac-
tion procedures depends on very favorable partitioning characteristics, this
is an important element in the potential for successful sludge microextraction.
Attempts to improve recoveries by compositing successive extracts increases
the likelihood of coextracting interfering materials with lower partition con-
stants. Similarly, compositing extracts from more than one solvent-salt com-
bination in order to analyze a wide range of compounds with a minimum of GC/MS
runs would increase the potential for interferences.
Recoveries from spiked sludge--The evaluations of three solvents and
salting for microextraction were repeated for spiked combined sludge (2.1%
solids). Two sludge spiking levels were employed.
Experimental—Spiked and unspiked 30-ml aliquots of combined sludge
were poured into Babcock tubes. Separate aliquots were spiked with the B/N
and acidic compounds nominally at 200 and 1,000 |Jg/liter. All sludge aliquots
spiked at 1,000 pg/liter were allowed to equilibrate at 4°C overnight with
tumbling. The aliquots were adjusted to pH 11 or 1, as appropriate, and were
diluted with 15 ml of glass-distilled water that had been saturated with the
extracting solvent. The diluted aliquots were extracted with 0.5-ml aliquots
hexane or toluene or 1.0 ml of diisopropylether by hand shaking for 2 min.
The extracts were recovered by centrifuging at 1,500 rpm for 15 min and remov-
ing the solvent with a 500-|Jl syringe. The volumes of extracts recovered were
read from the syringe. All extractions were conducted in duplicate. The ex-
tracts were analyzed by the GC/MS procedures described in the preliminary P0TW
sludge protocol.
Results and discussion—The results of microextraction recovery
determinations from spiked sludge are shown in Tables 33 to 35. Few analyte
recoveries were acceptable for an analytical method. The primary breakdown
of the method was extract recovery. Extract recoveries were both variable
and low. Essentially, no extract could be recovered from many of the acidi-
fied sludge aliquots.
92
-------
TABLE 33. HICROEXTRACTtON OF EXTRACT ABLE COMTOUHDS FRQH SPIKED SLUDGE SAMPLES WITH HEXANE
Spike Spike recovery (I) Splkeb Spike tecovery (X\
level With HaCl Without HaCl level With HaCl Without HaCl
Compound (ug/L sludge) Va " 330 pi V - 400 |il V - 230 pi V - 220 pi (pr/L sludge) V - B70 pi V - 1,630 pi V - 1,220 pi V - 1,200 pi
vO
1,4-Dlchlorobenzene
llcxachloroethane
Bi8(2-chloroiso-
propyl)ether
Bis(2-chloro-
etl»yl)ether
Acenaphthalene
2,6-Dlnltrotoluene
Fluorantliene
Benzidine
3,3*-Dichloro-
betizidlrte
n-Butylbeiizyl-
phtlialate
Bls(2-ethy lliexyl)
pht halate
Benzo(a)pyrene
Phenol
2,4-Dlnethylphenol
2,4-Dlchlorophenol
Pentachloropltenol
312
208
190
234
180
207
243
301
264
200
225
225
219
232
6
1
7
0
0
10
58
3
0
2.5
0
6
5
3
4
0
0
6
44
4
71
41
40
0
15
23
149
10
100 pi V * 50 pi V « 0 |il
92
45
45
0
15
20
167
11
V - 0 pi
a V - volume of extract recovered,
b Given aa the average value for the four sanplea.
c Not present In the spiking solution.
1,570
1,040
1,655
1,055
950
1,170
900
1,020
1,200
1,510
1,320
1,000
36
0
63
22
26
0
5
0
0
9
11
0.4
100 pi
0.4
0.3
1
0
34
0
70
26
22
0
0
0
4
0
0
0.2
V - 140 ill
0.3
0.3
1
0
51
0
48
10
51
0
31
0
10
84
23
9
10 pi
51
0
53
11
53
0
38
0
14
88
24
12
10 pi
0
0.01
0.06
0.07
-------
TABLE 34. M1C80EXTRACT10N OF EXTRACTABLE COMPOUNDS FROM SPIKEl) SLUDGE SAMPLES WITH TOLUENE
Spike Spike recovery (X) Spike®* Spike recovery (%)
Coapound
level
(pg/L sludge)
Ulth
NaCl
Without
NaCl
' level
(ng/L sludge)
Ulth NaCl
Without
NaCl
Vs - 100 ul
V - 690 ill
V - 380 pi
V - 320 ul
V - 1,380 ul
V - 615 ul
V - 830 |il
V ¦
- 45
1,4-Dlchlorobentene
312
0
0
10
Q
1,570
0
9
14
3
llexftchloroethane
208
0
0
0
0
1,040
0
0
0
0
BIs(2-chloroIso-
c
c
c
c
c
1,650
0
25
0
4
propyOether
Bls(2-ch1oro-
c
c
c
c
c
1,055
0
0
0
5
ethyl)ethcr
Acenaphthylene
190
19
20
11
10
950
0
7
35
4
2,6-Dlnltrotoluene
234
0
0
0
0
1,170
0
0
0
0
Fluoranthene
180
34
32
13
6
900
0
0.3
0
0.5
Benzidine
207
0
0
0
0
1,030
0
0
0
0
3,3'-Dichloro-
243
0
0
9
7
1,210
0
k
20
1.5
benzidlne
n-Butylbenzyl-
301
29
21
8
9
1,510
0
0
0
0.5
plithalate
BIs(2-ethylhexy1)
264
224
200
73
80
1,320
10
0
c
c
plithalate
Benzofa|pyrene
200
0
0
9
5
1,000
0
0
0
-12
V - 0 111 V
- 0 Ml V
- 0 ul v
- 0 pi
V - 10 pi
V - 70 |il
V - 0 ill
V
- 0
Phenol
225*
_
1,130
NA
U
_
_
2,4-l>lacthy)phenol
225*
-
-
-
-
1,160
NA
1
-
-
2,4-Dlchloroplienol
219<1
-
-
-
-
990
NA
1
-
-
Pentachlorophenol
232*
-
-
-
1,120
NA
0
-
-
a V « voluae of extract recovered,
b Given as the average value for the four saaplea.
c Hot present in the spiking aolution.
d No solvent could be recovered fro* the spiked sludge.
e Not estimated since concentration of phenol In the unsplked sludge w.in found (o be larger
than the concentration in the spiked sludge.
NA " not analyzed.
-------
TABI.E 35. H1CR0EXTRACTI0N OF EXTRACTABI.E COMPOUNDS FROfl SPIKED Sl.llDGE SAMPLES WITH DlISOPROrYL ETHER
Spike Spike recovery (Z) Spike" Spike recovery (*)
level Hi til NaCl Wltliout NaCl level With NaCl Without NaCl
Compound (ug/1. sludge) Va - 310 pi V - 430 pi V - 220 |il V - 195 pi (pg/L sludge) V - 750 pi V - 950 pi V - 40 pi V - 165 pi
1,4-Dichlorobenzene 312 0 0 0 0 1,570 19 19 3 9
Ilexacliloroethane 208 0 0 0 0 1,040 0 0 0 0
HI s < 2-ch I orol so- c c c c c 1,650 106 74 6 19
propyl )etlier
Bls(2-chlnro- c c c c t 1,055 45 47 1 7
etltyl)ether
Arenaphthylene 190 0 0 0 0 950 12 15 3 14
2,6-DIiiltrotoluenc 234 0 0 0 0 1,170 0 0 0 0
Fluoranthene 180 0.6 3 0.6 1.1 900 5 5 1 7
Benzidine 207 0 0 0 0 1,030 0 0 0 0
3,3'-Dlchloro- 243 0 0 0 0 1,210 13 54 8 49
benzidine
n-ButyIheniyl- 301 0.8 3.5 0.3 0 1,510 9 10 4 17
phthalate
Bls(2-ethyIhexyl) 264 12 27 4 4 1,320 35 40 12 39
phthalate
Bonzojnjpyreue 200 0 0 0 0 1,000 6 7 2 8
V - 180 p|J V - 880 pid V - 400 plJ V - 390 p1d V - 0 pi » - 0 pi V - 0 pi V - 0 pi
Phenol 225 1500 e - - - -
2,4-Dlnethylphenol 225 1.4 1.3 0 0 e - -
2,4-I>lchlorophenol 219 0.8 0 0 0 e - --
Pentnrlilorophenol 232 0000 e - -
a V « volume of extract recovered,
b (>lven as the average value for the four samples,
c Not present In llic spiking solution.
d An additional 0.5 nl of dllsopropyl ether was added after concentration.
e Experiment not performed since the solvent could not be recovered fr<« the unsplked sludge.
-------
Method Selection—
The alternate extraction procedures evaluated, continuous liquid-liquid
extraction, extractive steam distillation, and microextraction, were selected
because they are less time-consuming and less labor-intensive than horaogeniza-
tion/centrifugation and they may produce extracts that require little or no
cleanup prior to GC/MS analysis. However, none of the techniques tested pro-
vided efficient extraction of the entire range of base/neutral and acidic
compounds. Each of these procedures achieved promising recoveries for concen-
trating spiked compounds from clean water. Recoveries from spiked combined
POTW sludge, however, were generally lower. Continuous liquid-liquid extrac-
tion with dichloromethane proved ineffective in extracting polynuclear aro-
matic hydrocarbons and similar hydrophobic compounds from sludge solids. Ex-
tractive steam distillation with dichloromethane was ineffective in extracting
compounds with low vapor pressures. In addition, the steam distillate extracts
of sludge contained high levels of coextracted aliphatic hydrocarbons. Al-
though good recoveries were obtained for spiked water with microextraction and
the procedure is exceptionally simple, the physical recovery of the organic
solvent extracts from sludge aliquots was a major hindrance. Hence, the homo-
genization/centrifugation method was selected for the revised sludge protocol.
This selection necessitated additional evaluation of extract cleanup procedures
alternate or supplemental to the GPC method used in the preliminary POTW pro-
tocol.
Extract Cleanup Methods
Successful employment of a rigorous, nonselective extraction procedure
for sludges, such as homogenization/centrifugation, necessitates the use of
very selective extract cleanup to produce extracts of efficient quality for
reliable GC/MS determination. Two extract cleanup mechanisms were evaluated
to meet these requirements, molecular size discrimination and polarity selec-
tion. Various GPC procedures were evaluated to determine the optimum molec-
ular size fractionation procedure. The polarity-based cleanup methods eval-
uated were adsorption chromatography on silica gel, florisil, and cesium
silicate. The performance of adsorption procedures were evaluated for sludge
extracts both with and without GPC precleaning.
Gel Permeation Chromatography—
Extract cleanup by preparative GPC is a molecular size discrimination
technique using a low pressure column. The method employed in the prelimini-
nary POTW sludge protocol used a 2.5-cm ID column packed with Bio-Beads SX-3.
The column was operated at 350 to 700 millibars (5 to 10 psi) and was eluted
with 5.0 ml/min dichloromethane. Compounds that are larger than the molecular
exclusion limit of the bead pores are excluded from the beads and are eluted
with the first few column void volumes. Compounds that are smaller than the
exclusion limit are retained in the pores and are eluted later than the ex-
cluded material. Specifically, the large biogenic compounds in sludge ex-
tracts are excluded from the bead pores and are eluted earlier than the ana-
lytes retained in the pores. Since no residue remains on the column and the
column is not deactivated by the sample, the column can be reused many times
and is amenable to automatic operation. The GPC system utilized for this
method development was an automated system capable of processing 23 injections
unattended.
96
-------
Five GPC packings were evaluated with three solvent systems and one pack-
ing was evaluated with a fourth solvent system. The basic properties of the
gels are their molecular weight exclusion limits, i.e., pore size, and their
swelling characteristics. The nominal molecular weight exclusion limits for
the five gels are listed in Table 36. The effective exclusion limits are in-
fluenced somewhat by bead swelling. Swelling characteristics are largely de-
pendent on the bead composition but are also influenced by the solvent. The
solvent influence on swelling is sufficient that injection of an extract onto
a GPC column eluted with a different solvent can cause a dramatic change in
the pressure in the column. Hence, extracts should be prepared in the same
solvent as the eluting solvent. The solvent systems evaluated were designed
to investigate the effect of solvent polarity on the column elution character-
istics. The solvents tested were dichloromethane, 15% cyclohexane in dichlo-
romethane, and 50% cyclohexane in dichloromethane. In addition, a very polar
solvent 50% acetone in dichloromethane was tested with Bio-Beads SX-4.
Each gel-solvent combination was evaluated by chromatographing solutions
of corn oil, butylbenzylphthalate, and mixed phenols. Corn oil was used to
represent the large biogenic materials in sludge extracts. Butylbenzylphthal-
ate is one of the largest extractable priority pollutants and thus is one of
the first to elute. The mixed phenols solution contained phenol, 2,4-dichloro
phenol, and pentachlorophenol. These compounds form weak hydrogen bonds with
beads. As a result they are some of the later eluting analytes. The objec-
tive of the elution experiments was to achieve maximum separation between the
corn oil and the phthalate and to allow good recovery of the phenols. Recov-
eries of selected compounds were determined for the optimum GPC procedure.
Elution patterns—
Experimental—Columns were prepared from 60 to 70 g of each packing.
The gel was soaked overnight in the first solvent tested. The columns (60 cm
x 2.5 cm ID fitted with adjustable plungers on both ends) were prepared by
pouring the gel slurry into the column, initiating the eluent flow (up-flow
at 5.0 ml/min with a constant flow pump), and compressing the bed to achieve
350 to 700 millibars of solvent pressure. The column was allowed to equili-
brate at least 2 h with flow. Solutions of corn oil, butylbenzylphthalate,
and mixed phenols (phenol, 2,4-dichlorophenol, and pentachlorophenol) were
prepared in the elution solvent. The concentration of corn oil was 400 mg/ml.
The concentrations of phthalate and phenols in the solutions were 20 to 180
|Jg/ml. These solutions were introduced onto the column in 5.0-ml injections.
Column elution was monitored with a 254-nra absorption detector. The eluent
solvent was changed for the next experiment by lowering the pressure on the
bed by backing off the plungers and changing the contents of the solvent res-
ervoir. The beads were allowed to equilibrate with the new solvent by pump-
ing for 2 to 4 h before the bed was compressed again and equilibrated under
pressure for 2 h.
97
-------
TABLE 36. MOLECULAR WEIGHT EXCLUSION LIMITS FOR THE GPC PACKINGS EVALUATED
Manufacturer
Packing
Molecular weight exclusion limit
Bio-Rad Laboratories
Bio-Beads SX-2
2,700
Bio-Rad Laboratories
Bio-Beads SX-3
2,000
Bio-Rad Laboratories
Bio-Beads SX-4
1,400
Bio-Rad Laboratories
Bio-Beads SX-8
1,000
Pharmacia
Sephadex LH-20
5,000
a Obtained from manufacturer's catalog or technical representative.
-------
Results and discussion—The elution volumes for corn oil, butylben-
zylphthalate, and the phenols are shown in Table 37. Figure 29 shows the
superimposed chromatograms for the three injections on each gel eluted with
15% cyclohexane in dichloromethane. Figure 30 shows a similar comparison of
chromatograms for SX-3 eluted with the three solvents. The elution volumes
and peak profiles observed indicate that gels with lower molecular weight ex-
clusion limits provided better peak shapes. In particular, the phenols elute
in a narrow band on the SX-4 and -8 profiles shown in Figure 29. This may be
at least partially attributed to denser packing of the column. The gels with
higher exclusion limits swell more and result in columns that are more com-
pressible. In particular, columns packed with SX-2 and -3 were very elastic.
The beds could be compressed several centimeters with both plungers. Columns
packed with SX-8 were barely compressible.
Although the chromatograms shown in Figure 29 do not show discern-
ible trends for resolution between the corn oil and the phthalate, elution
data for all gel-solvent combinations indicate that the lower exclusion limit
gels provided generally poorer separations. In addition, as illustrated in
Figure 30 for SX-3, elution with higher percentages of cyclohexane in dichlo-
romethane resulted in poorer corn oil-phthalate separation. The best overall
separation was obtained with SX-3 eluted with dichloromethane, the method used
in the preliminary POTW sludge protocol. This system had the best combination
of separation and peak shape. A fraction cut selected at the valley between
the corn oil and phthalate achieved £ 85% elimination of the oil from the ana-
lytes with k 85% recovery of the phthalate (based on peak areas). The selec-
tion of dichloromethane as the elution solvent is directly compatible with
the homogenization/centrifugation sludge extracts.
Determination of recoveries—The recoveries of the extractable spiking
compounds through GPC on Bio-Bead SX-3 eluted with dichloromethane were deter-
mined by chromatographing spiked blank extracts.
Experimental—A 1.0-ml mixed standard solution containing the
extractable spiking compounds at 100 pg/ml in dichloromethane was chromato-
graphed on the Bio-Beads SX-3 column described above. The column was eluted
with dichloromethane at 5.0 ml/min. The initial 110 ml (22 min) of eluent
was discarded. The next 125 ml (25 min) was collected, evaporated to 1.0 ml,
and analyzed by GC/FID using the chromatographic procedures in the preliminary
POTW protocol. A second 1.0-ml aliquot of the spiked blank was diluted to
125 ml with dichloromethane and similarly evaporated and analyzed to check
for losses in evaporating the cleaned extract.
Results and discussion—The results of recovery determinations for
the extractable spiking compounds through GPC cleanup and extract concentra-
tion are shown in Table 38. Analyte recoveries from Kuderna-Danish concentra-
tion were excellent. Recoveries from GPC were good for all compounds except
bis(2-chloroethyl)ether. Low recoveries are frequently observed for this com-
pound, bis(2-chloroisopropyl)ether, and benzidine because their poor chromatog-
raphy makes determination difficult. The lower recovery for bis(2-ethylhexyl)-
phthalate reflects the selection of the dump-collect split to achieve effective
cleanup.
99
-------
TABLE 37. ELUTION VOLUMES FOR CORN OIL, PHTHALATES, AND PHENOLS
ON GPC USING VARIOUS GELS AND SOLVENTS
Elution volume (ml)
Eluent
Compound
LH-20
SX-2
SX-3
SX-
-4
SX-8
Dichloromethane
Corn oil
50-120
60-170
50-125
50-
-120
50-95
Butylbenzylphthalate
80-120
60-170
105-145
92-
-120
65-80
Phenols3
112-137
120-220
145-195
117-
-170
110-130
15% Cyclohexane in
Corn oil
65-145
50-160
45-135
40-
-110
65-135
dichloromethane
Butylbenzylphthalate
85-130
120-175
85-165
85-
-125
85-130
Phenols
130-155
160-220
b
135-
¦145
130-155
50% Cyclohexane in
Corn oil
50-115
50-175
45-145
45-
-120
70-140
dichloromethane
Butylbenzylphthalate
50-100
130-195
100-150
90-
-130
100-150
Phenols
90-110
175-245
145-225
135-
-200
b
50% Acetone in
Corn oil
b
b
b
100-
-175
b
dichloromethane
Butylbenzylphthalate
b
b
b
110-
-155
b
Phenols
b
b
b
110-
-150
b
a Phenol, 2,4-dichlorophenol, and pentachlorophenol.
b No determination.
-------
Butyl benzyl-
phthalate
Corn 011„
l Phenols
»iit
0 10 20 30 40 so
(A)
Butyl benzyl -
phthalate
Corn 011
Corn Oil
Butylbenzyl-
nhthalate
/
Phenols
10 20 30 40 so
(B)
Phenols
Butylbenzyl
phthalate
Corn 011
L
Butylbenzyl-
ohthai ate
Corn 011
(C)
Phenols
10 70 30
' ¦
10 20 30
(D)
(E)
Figure 29• GPC chromatograms on Sephadex LH-20 (A) and Bio-Beads
SX-2 (B), SX-3 (C), SX-4 (D), and SX-8 (E) eluted with
15% cvclohexane in dichloromethane.
101
-------
Butylbenzyl-
phthalate
Corn Oil
Phenols
Phenols
Butyl benzyl -
phthalate >
Corn Oil
J
10
L
0
40
50
20
30
Butylbenzyl-
^ phthalate
Corn Oil
Phenols
20
30
±
10
0
40
SO
Iu. T|m# <",n> <-4")
(A) (15) (C)
Figure 30. GPC c.hromatoj4rams on Bio-Beads SX-3 eluted wirh dicliloromethane (A), 15% cyclohexane
in di chlorometliane (B), and 50% cyclohexane in dichloromethane (C).
-------
TABLE 38. RECOVERIES OF EXTRACTABLE COMPOUNDS
Conrpouad
Spike
level
(MS)
% Recovery
GPC cleanup +
Kuderna-Danish Kuderna-Danish
concentration concentration
1,4-Dichlorobenzene
100
69
98
Hexachloroethane
100
70
100
Bis(2-chloroisopropyl) ether
100
63
85
Bis(2-chloroethyl) ether
100
37
103
Acenaphthylene
100
77
100
2,6-Dinitrotoluene
100
77
106
Fluoranthene
100
87
98
Benzidine
100
52
103
3,3'-Dichlorobenzidine
100
79
103
n-Butylbenzylphthalate
100
81
104
Bis(2-ethylhexyl)phthalate
100
57
110
Benzo[a]pyrene
100
79
108
Phenol
100
81
101
2,4-Dimethylphenol
100
90
94
2,4-Dichlorophenol
100
86
102
Pentachlorophenol
100
105
100
103
-------
Absorption Chromatography on Florisil and Silica Gel—
Absorption chromatography on florisil or silica gel is a widely used
extract cleanup procedure based on polarity discrimination. The activity of
the adsorbent can be adjusted to meet the requirements of a particular ana-
lytical system by adding a small amount of water to deactivate a dry, i.e.,
fully activated, adsorbent. Florisil and silica gel are two of the most
widely used adsorbents. A typical adsorption chromatography procedure uses
20 g of adsorbent in a 1- to 25-cm ID column. The sample extract is intro-
duced onto the column in a small volume and the column is eluted with a series
of solvents and solvent mixtures with increasing polarities. The first frac-
tion, typically a hydrocarbon solvent, elutes aliphatic hydrocarbons which
are discarded. The polarities of the subsequent eluents and the volumes used
are designed to elute the analytes while leaving more polar coextractants on
the column. Since small amounts of water in the extract partially deactivate
the adsorbent and very polar materials remain on the column after analyte
elution, the adsorbent is discarded after a single usage.
Adsorption chromatography could be useful for sludge extract cleanup
alone or as a supplement to GPC cleanup. Florisil or silica gel provides re-
moval of both aliphatic hydrocarbons and low molecular weight polar compounds
which are not removed by GPC. Although florisil and silica gel also remove
high molecular weight polar compounds, precleaning with GPC could make removal
of these compounds more efficient by reducing the capacity required of the
adsorbent.
Partially deactivated florisil and silica gel were evaluated for adsorp-
tion chromatographic cleanup of sludge extracts. The elution properties for
the extractable spiking compounds were determined and an appropriate elution
scheme was developed. Recoveries were determined for the analytes spiked into
homogenization/centrifugation extracts both with and without GPC precleanup.
Extracts from both 80- and 160-ral sludge aliquots were spiked and chromato-
graphed to test the capacity of the method.
Development of an elution scheme—
Experimental--Florisil (60/100 mesh, Floridin) was cleaned by a
16-h Soxhlet extraction with dichloromethane. The solvent was allowed to
evaporate from the adsorbent at ambient temperature prior to activating over-
night at 140°C. The clean, active adsorbent was stored at 140°C. Immediately
prior to use, the active florisil was deactivated with 1% water (by weight)
and was allowed to equilibrate for 4 h, with tumbling. Silica gel (Silica
Woelm, 70/150 mesh, ICN) was similarly cleaned, activated, and stored. The
silica gel was deactivated with 3% water (by weight) and allowed to equili-
brate 4 h, with tumbling prior to use. Columns were prepared by pouring
20 g of the adsorbent into a 28 cm x 2 cm ID glass column half filled with
hexane. After allowing the adsorbent to settle in the column, the hexane
level was lowered to just above the adsorbent.
The elution characteristics of the extractable spiking compounds on
the florisil column were investigated by chromatographing 1.0-ml aliquots of
hexane spiked with 27-50 (Jg of the B/N and acidic compounds. The elution
characteristics of B/N spiking compounds on the silica gel column were simi-
larly investigated. Since preliminary experiments indicated that significant
104
-------
quantities of 1,4-dichlorobenzene and hexachloroethane were eluted from silica
gel with 50 ml of hexane, the initial fraction (Fraction I) was eluted with
only 20 ml of hexane. Subsequent fractions were eluted with 50 ml each of
10% dichloromethane in hexane (fraction II) and 50% dichloromethane in hexane
(Fraction III), followed by 150 ml of 5% acetone in dichloromethane (Fraction
IV). The fractions eluted were concentrated to 1.0 ml in Kuderna-Danish
evaporators and the analytes determined by GC/FID using the preliminary POTW
protocol chromatographic methods.
Results and discussion--The recoveries for the spiked extractable
compounds in each fraction eluted from the florisil and silica gel columns
are shown in Tables 39 and 40, respectively. The distributions of the B/N
compounds within the fraction eluted were very similar for both the florisil
and silica gel column. That is, the performances of the florisil and silica
gel columns were roughly equivalent.
The recoveries observed for the phenols from the florisil column
and for the benzidenes from both columns were low. More polar eluents are
likely required to recover these compounds. However, more polar eluents are
likely to elute significant quantities of polar coextracting interferences
from sludge extracts. Subsequent development of florisil and silica gel
extract cleanup methods was focused on the B/N compounds only.
Recoveries from spiked sludge extracts--Recoveries were determined for
B/N compounds spiked into extracts prepared from 80- and 160-ml aliquots of
combined POTW sludge and cleaned up by florisil and silica gel chromatography.
This study was designed to test both the recoveries and cleanup capacities of
the columns. In addition, recoveries were determined for extracts spiked fol-
lowing GPC cleanup to investigate the possible enhancement of the performance
of adsorption chromatographic cleanup by removing high molecular weight bio-
genic compounds.
Experimental--Homogenization/centrifugation extracts were prepared
from 80- and 160-ml aliquots of combined POTW sludge. Half of all extracts
were cleaned by GPC, and all extracts were concentrated to 5 to 10 ml. Se-
lected extracts were spiked with the B/N spiking compounds. Extracts prepared
from 160-ml sludge aliquots were spiked at twice the levels for extracts from
80-ml aliquots. All combinations of spiked and unspiked extracts with and
without GPC cleanup were chromatographed on florisil and silica gel columns.
The extracts were loaded onto the columns by mixing with 2 g of the adsorbent,
removing the solvent under a gentle stream of dry nitrogen, and then transfer-
ring the adsorbed extract to the top of the column for elution. The columns
were eluted in four fractions as described above. Fraction I (20 ml hexane)
was discarded. Fractions II, III, and IV were composited or analyzed sepa-
rately for some extracts. All composited fractions (II through IV) and indi-
vidual fractions were concentrated and analyzed by GC/MS using the procedures
described in the preliminary POTW protocol.
105
-------
TABLE 39. RECOVERIES OF EXTRACTABLE COMPOUNDS FROM SPIKED BLANK EXTRACTS BY FLORISIL
CHROMATOGRAPHY ELUTED WITH HEXANE, DICHLOROMETHANE AND ACETONE
Si * Recovery
3 b C d
Compound (Mg) Fraction I Fraction II Fraction III Fraction IV Total
1,4-Dichlorobenzene
46.8
0
65
5.1
2.4
73
Hexachloroethane
31.2
0
70
2.6
0.6
73
Bis(2^chloroisopropyl)ether
49.7
0
0
20
41
61
Bis(2-chloroethyl)ether
31.7
0
0
3.8
29
33
Acenaphthylene
28.5
0
14
37
13
65
2,6-Dinitrotoluene
35.1
0
0
0
49
49
Fluoranthene
27.0
0.2
0.1
43
25
68
Benzidine
31.0
0
0
0
0
0
3,3'-Dichlorobenzidine
36.4
0
0
0
0
0
n-Butylbenzylphthalate
45.2
0.9
0
0
0.5
1.4
Bis(2-ethylhexyl)phthalate
39.6
0
0
6.1
0
6.1
Benzo[a]pyrene
30.0
0
0
8.1
22
30
Phenol
48.4
0
0
3.1
12
16
2,4-Dimethylphenol
35.4
0
7.1
11
11
29
2,4-Di chlorophenol
32.8
0
1.4
7.4
31
40
Pentachlorophenol
30.4
0
0
0
0
0
a Fraction I was eluted with 20 ml hexane.
b Fraction II was eluted with 50 ml 10% dichloromethane in hexane.
c Fraction III was eluted with 50 ml 50% dichloromethane in hexane.
d Fraction IV was eluted with 150 ml 5% acetone in dichloromethane.
-------
TABLE 40. RECOVERIES OF BASE/NEUTRAL COMPOUNDS FROM SPIKED BLANK EXTRACTS BY SILICA GEL
CHROMATOGRAPHY ELUTED WITH HEXANE, DICHLOROMETHANE AND ACETONE
Compound
Spike
level
(H8)
% Recovery
Fraction I3
Fraction
b c
II Fraction III
Fraction IV^
Total
1,4-Dichlorobenzene
46.8
0
62
8
0
70
Hexachloroethane
31.2
0
83
0
0
83
Bis(2-chloroisopropyl)ether
49.6
0
0
0
85
85
Bis(2-chloroethyl)ether
31.6
0
0
0
93
93
Acenaphthylene
28.5
0
0
79
0
79
2,6-Dinitrotoluene
35.1
0
0
0
96
96
Fluoranthene
27.0
0
0
90
0
90
Benzidine
31.0
0
0
0
0
0
3,3'-Dichlorobenzidine
36.4
0
0
0
50
50
n-Bu ty1benzylphtha1a te
45.2
0
0
0
97
97
Bis(2-ethylhexyl)phthalate
39.6
0
0
0
93
93
Benzo[a]pyrene
30.0
0
0
79
0
78
a Fraction I was eluted with 20 ml hexane.
b Fraction II was eluted with 50 ml 10% dichloromethane in hexane.
c Fraction III was eluted with 50 ml 50% dichloromethane in hexane.
d Fraction IV was eluted with 150 ml 5% acetone in dichloromethane.
-------
Results and discussion—Table 41 shows a sample of the recoveries
observed for spiked sludge extracts. This table shows the results for ex-
tracts from replicate 80-ml aliquots of combined sludge, spiked without GPC
cleanup, and chromatographed on florisil. Recoveries observed were generally
good and fairly reproducible for all compounds except benzidene.
The distribution of analytes among Fractions II, III, and IV is very
similar to that for the spiked blank shown in Table 41. The distribution of
compounds among the fractions could be helpful in cases where determination
of selected compounds is required. A significant portion of the background
material in the composite of Fractions II to IV was contributed by Fraction II.
The GC/MS chromatograms for the composite of Fractions II to IV for the exper-
iment represented in Table 41 are shown in Figure 31. Figures 32 to 34 show
the GC/MS chromatograms for the individual fractions. These chromatograms
indicate that determination of compounds eluting in Fractions III and IV would
be easier if those fractions were analyzed separately.
The recoveries obtained for spiked compounds in replicate extracts
from 80- and 160-ml sludge aliquots with and without GPC chromatographed on
florisil and silica gel are summarized in Table 42. The recoveries observed
for the florisil and silica gel columns were similar. Recoveries were also
similar for extracts with and without GPC cleanup. The GC/MS chromatogram
for the composite of Fractions II through IV for a spiked 80-ml sludge aliquot
cleaned by GPC is shown in Figure 35. The background in this chromatogram is
very similar to that for a replicate extract without GPC (Figure 31).
The recoveries observed for extracts prepared from 80- and 160-ml
sludge aliquots are comparable. However, the florisil-cleaned extracts of
160-ml sludge aliquots had to be analyzed at a larger final extract volume to
avoid overloading the GC/MS detection system. Hence, no real gain in method
sensitivity was realized by extracting larger sludge aliquots.
Adsorption on Cesium Silicate--
The use of cesium silicate has been reported for isolation of phenolics
from neutral compounds in dilute wastewater extracts.12'13 This procedure
was evaluated as a cleanup method for sludge extracts. Recoveries were deter-
mined for spiked blanks and then for spiked sludge extracts.
T2 Stalling, D. L., L. M. Smith and J. D. Petty. Approaches to Comprehensive
Analyses of Persistent Halogenated Environmental Contaminants. Measure-
ment of Organic Pollutants in Water and Wastewater. ASTM STD 686,
C. E. Van Hall, Ed., 1979, pp. 302-323.
13 DeWalle, F. and E. Chian. Presence of Priority Pollutants in Sewage and
Their Removal in Sewage Treatment Plants. First Annual Report to EPA,
Grant No. R806102, 1979.
108
-------
tmi a. mmm & mtmaruL mmm rmspim nmi trmm it mim attmnmm,
WITHOUT CLEANUP Blf CPC, ELUTED WITH liEXANE, DICHLOROHETIIANE, AND ACETONE
Spike
level
(MR)
Unspikc-d
sludge
level
(l«8)
Spiked
sludge 1
Spiked sludge 2
Average
recovery
(%)
Conpound
Fraction8
11-1V Uig)
Total
recovery (%)
Fraction 11^
(wO
Fraction 1I1C
(Mg)
Fraction IV*'
(Mg)
Total
recovery (%)
1,4-Dichlorobenzeue
46.8
33
60
58
70
0
0
29
68
llexachl orofllune
31.2
0
30
96
38
0
0
123
110
Bis(2-cfclorr,-
isopropyl JeLher
149.0
0
190
128
-------
479
581
495
529
455
121
514
393 *
•o
3 7?
294 319
RIC
No fiPC cleanup
spiked sludge
a
260
300
400
500
5:00 10:00 15:00 20:00 25:00 30:00 TIME
Figure 31. GC/MS chromatogram of the combined eluent (Fractions II-IV) obtained by
florisil cleanup of the spiked sludge (no (IPC cleanup).
-------
No GPC cleanup
spiked sludge
20:00
25,00
10:08
914227*.
100
5:00
600
30:08
SCAN
TIME
Figure 32. GC/MS chromatogram of Fraction TI obtained by florisil cleanup
of the spiked sludge (no GPC cleanup).
-------
i«e.«
8634360.
523
MC
i
m
10.00
r
20.00
500
25.00
£00
30.03
SCAN
TIIUE
Figure 33. GC/MS chromatogram of Fraction TTI obtained by florisil cleanup
of the spiked sludge (no GPC cleanup).
-------
4 2
€7
541
238
Q.
a
507
9 «*4
41 w
O CO
a
CM
CQ
600 SCAN
30;M TIME
3M
I5; W
Figure 34. GC/MC chromatogram of Fraction IV obtained by florisil cleanup
of the spiked sludge (no GPC cleanup).
-------
TABLE 42. SUMMARY OF RECOVERIES FOR TIIE EXIF ACT ADI.E COMPOUNDS FROH EXTRACTS OF 80- AND 160-nl
SI.UDCE AllQUOTS SPIKED AND THEN CLEANED WITH FL0R1SIL OR SILICA GEL
Average recovery (%)*
Spike level Florisil Silica gel
Compound
(Vg)
With
GPC
Without
GPC
With
80 ¦!
GPC
160 >1
Without
GPC
80 «1
160 al
80 ml
160 nl
80 al
160 tal
80 ml
160 ml
1,4-Dichlorobenzene
47
94
92
142
68
88
73
109
62
95
Hexachloroethane
31
62
79
97
110
70
95
54
116
47
Bis(2-chloroisopropyl)ether
149
298
100
102
99
111
102
112
129
62
Bis(2-chloroethyl)ether
95
190
103
49
74
53
85
66
116
93
Acenaphthylene
29
57
97
148
96
109
87
72
91
93
2,6-Dinitrotoluene
35
70
90
110
122
139
108
102
129
108
Fluoraothene
27
54
85
110
75
135
69
55
98
103
Benzidine
93
186
17
0
0
0
0
0
0
0
3,3-Dichlorobenzidine
109
218
106
168
90
167
63
155
70
1
n-Butylbenzylphthalate
45
90
170
b
I0B
b
148
b
103
b
Bis(2-ethylhexyl)phthal ate
40
80
152
b
48
b
108
b
270
b
Benzojajpyreue
30
60
52
155
52
130
47
114
81
99
a Average of two determinations.
b Recoveries were not determined because the levels in the unspiked extracts were much higher than the spike
level.
-------
loO.O-i
ftIC
213 2375
Spiked sludge after GFC
15: uu
7593S80
i00
5:00
100
16:60
c09 SCAN
0:O0 TltC
Figure 35. GC/MS chromatogram of combined eluent (Fraction II-IV) obtained by florisil
cleanup of the spiked sludge following GPC cleanup.
-------
Recoveries from spiked blank extracts—
Experimental—Cesium silicate was prepared by procedures reported
by DeWalle and Chian.13 Two hundred grams of silica gel (Davidson 923,
100/120 mesh) were added to 600 ml of a saturated solution of cesium hydrox-
ide in methanol. The mixture was stirred at room temperature for 1 hr, was
allowed to settle, and the supernatant was decanted. The residue was filtered
and the filtrate was washed with 250 ml of methanol followed by 250 ml of di-
chloromethane. The adsorbent was dried and stored at 140°C. Columns were
prepared by adding 20 g of cesium silicate into a 280 cm x 2 cm ID column half
filled with dichloromethane. The adsorbent was allowed to settle and the sol-
vent level was adjusted to just above the adsorbent. A dilute, spiked blank
was prepared by spiking a 150-ml aliquot of dichloromethane with the acidic
spiking compounds. The spiked blank was allowed to pass through the column
at 3 ml/min and the eluent was collected (Fraction I). The column was then
eluted with 20 ml of methanol (Fraction II) and 100 ml of methanol (Fraction
III). Since cesium silicate has some appreciable solubility in methanol, the
methanol eluents were concentrated to 4 ml in Kuderna-Danish evaporators and
then were partitioned between 6 ml of water (adjusted to pH 1) and 2 ml of
dichloromethane. The dichloromethane fraction was removed and the aqueous
fraction was extracted with two additional 2-ml aliquots of dichloroethane.
All three extracts were composited. The dichloromethane column eluent
(Fraction I) and the methanol extracts (Fractions II and III) were concen-
trated to 1.0 ml, and the analytes were determined by GC/FID using the chro-
matographic conditions described in the preliminary P0TW sludge protocol.
Results and discussion—The results of recovery determinations for
a spiked blank are shown in Table 43. No breakthrough of the analytes was
observed, i.e., none of the spiked phenols were found in Fraction I. It was
also determined that 20 ml of methanol was sufficient to elute the spiked
phenols from the column.
Recoveries from spiked sludge extracts—Recoveries were determined for
spiked compounds in an extract from an 80-ml aliquot of sludge. Recoveries
were also determined for spiked extracts from 160-ml aliquots of a different
sludge with and without GPC cleanup.
Experimental—Two extracts were prepared from 80-ml aliquots of
sludge and were spiked with the acidic spiking compounds. Both dilute ex-
tracts (240 ml) were passed through the column at 3 ml/min. The columns were
eluted with 50 ml of methanol and the methanol eluents were back extracted as
described in the preceding section. Spiked and unspiked extracts of 160-ml
sludge aliquots with and without GPC cleanup were processed on cesium silicate
columns in the same manner except that the original extract eluents were col-
lected. All fractions were concentrated and analyzed by GC/MS with the pro-
cedures described in the preliminary P0TW sludge protocol.
116
-------
TABLE 43. RECOVERIES OF ACIDIC COMPOUNDS FROM SPIKED BLANK
EXTRACTS BY CESIUM SILICATE CHROMATOGRAPHY
Spiked blank
Compound
Spike
level
(MJ?)
Fraction Ia
(Mg)
Fraction 11^
(MR)
Fraction IIIc
(Mg)
% Recovery
Phenol
48.4
0
39.5
0
82
2,4-Dime thylpheno1
35.4
0
23.7
0
67
2,4-Dichlorophenol
32.8
0
26
0
79
Pentachlorophenol
30.4
0
37
0
122
a Fraction I represents the dichloromethane eluate.
b Fraction II was eluted with 20 ml methanol,
c Fraction III was eluted with 100 ml methanol.
-------
Results and discussion—The results of recovery determinations for
spiked compounds from sludge extracts are shown in Table 44. The recoveries
observed for the extract of 80-ml aliquots of sludge were good. The very high
recoveries noted for phenol and pentachlorophenol in the extracts of 160-ml
sludge aliquots remain unexplained. However, significant portions of the
analytes spiked into these extracts eluted with the extract (Fraction I).
The breakthrough is probably due to the higher levels of coextracted materials
blocking the adsorption sites on the cesium silicate. Since the results were
no better for extracts that had been cleaned by GPC, the interferences are
likely low molecular weight compounds. Hence, the cleanup capacity of cesium
silicate may be somewhat limited.
Method Selection—
The GPC extract cleanup method used in the preliminary POTW sludge pro-
tocol, which consisted of Bio-Beads SX-3 eluted with dichloromethane, provided
the best combination of peak shape and separation between corn oil (used to
simulate high molecular weight biogenic interferences) and butylbenzylphthalate
(one of the largest extractable priority pollutants) of the gel-solvent combin-
ations tested. Good recoveries were observed for compounds in spiked blanks
processed by this GPC procedure. This molecular size based discrimination
was shown to be applicable for cleaning both acidic and B/N extracts.
Adsorption chromatography on florisil and silica gel provided good clean-
up and recoveries for B/N compounds. Recoveries for the acidic compounds was
poor. The florisil and silica gel procedures evaluated provided roughly equiv-
alent recoveries. No improvement in the performance of these procedures was
realized for extracts precleaned by GPC. Although recoveries observed for
compounds spiked into extracts from 80- and 160-ml sludge aliquots were very
similar, cleaned extracts from 160-ml sludge aliquots had to be analyzed at a
larger final volume. Hence, no real improvement in method sensitivity re-
sulted from extracting larger sludge aliquots.
Cesium silicate chromatography provided good recoveries for spiked blanks
and spiked extracts from 80-ral sludge aliquots. However, recoveries were poor
for compounds spiked into extracts from 160-ml sludge aliquots.
Selection of GPC or florisil or silica gel methods for cleaning B/N ex-
tracts should be based on the specific requirements of the sludge extract and
the determination method. The polarity based adsorption chromatography pro-
cedures provide more effective cleanup for extracts containing significant
quantities of low molecular weight polar coextractants. In addition, florisil
and silica gel remove aliphatic hydrocarbons. Hence, extracts cleaned by the
florisil or silica gel methods are more likely to be amenable to analysis us-
ing capillary column GC/MS. On the other hand, the GPC cleanup method can
easily be automated and produces extracts suitable for packed column GC/MS.
The poor recoveries observed for compounds spiked into extracts from 160-ml
sludge aliquots and cleaned with cesium silicate indicated that the column is
very susceptible to overloading. Hence, GPC is a more reliable procedure for
cleaning acidic extracts.
118
-------
TABLE 44. RECOVERIES OF ACIDIC COMPOUNDS FRCHI SPIKED SMIDGE EXTRACTS
BY CESIUM SILICATE CHROMATOGRAPHY
160^»|
BO-al " Without GPC " " Witli CM
Spike
lltispiked
Recovery
Fract ion
Spike
Unspike
Fraction
IB
level
Fraction
S[» i ke
Fract ion
recovery
Fraction
Unspi ke
Fraction
levc I
Fraction
Spike
Fraction
recovery
Fraction
level
level
lla
level
II
I
11
1
11
1
U
Coapound
(pk)
09
1
(I)
(Hg)
(ME)
(Hg)
(*>
(*)
(»'g)
(PR)
(I)
«)
Phenol
48.4
8.4
68
76.2
18
85
150
62
33
88
120
150
2,4-Diaethylphenol
35.4
NDC
75
49.2
HD
ND
68
43
ND
ND
63
33
2,4-Dichlorophenol
32.8
2.8
89
53.1
ND
NU
57
23
ND
ND
51
39
Pentachlorophenol
30.4
9.8
120
45.6
ND
1.7
32
160
ND
0.5
0
180
a Eluted with SO ¦! of Methanol,
b The crude extract etuate.
c NU = not detected.
-------
GC/MS Procedures
Capillary GC/MS procedures were evaluated as an alternative to the packed
column GC/MS methods described in the preliminary PQTW sludge protocol. The
performances of selected column systems were compared. The column system pro-
viding the best overall performance was evaluated with cleaned sludge extracts.
Relative retention times and response factors were determined for the B/N and
acidic priority pollutants.
Performance Evaluations for Selected Column Systems—
The performances of four wall-coated open tubular (WCOT) capillary col-
umns were evaluated by chromatographing commercially prepared capillary column
performance test solutions and solutions of the B/N and acidic spiking com-
pounds. Two columns were evaluated with both nitrogen and hydrogen carrier
gases to investigate the influence of the carrier gas on column performance.
Experimenta1--The WCOT capillary columns evaluated and the operating
conditions used in these evaluations are shown in Table 45. The columns were
tested in a Varian 3700 chromatograph with a flame ionization detector. This
chromatograph was fitted with a Varian Model 1070 split/splitless injection
system. Two commercially prepared column performance test solutions were chro-
matographed on each column. The contents of these test solutions are listed
in Table 46. The resulting chromatograras were used to calculate the height
equivalent to a theoretical plate (HETP), peak asymmetries (AS) and column
adsorption for selected compounds, and the acid-base character CpH) of the
columns. Solutions of the acidic and B/N spiking compounds were also chromato-
graphed on each column system. The amount injected for each compound ranged
50 to 100 ng for the B/Ns and 75 to 125 ng for the acids.
Results and discussion—Sample chromatograms obtained for performance
evaluation test mixtures on each column are shown in Figures 36 to 39. The
results of the column performance parameters determined are shown in Table 47.
The HETP for each column was calculated as a measure of the resolving power
according to the formula:
HETP =
where L is the column length and is the number of theoretical plates.
N Cjr is calculated as follows:
eff
Neff = 5'545 (w^)
where t 1 is the retention time of the selected compound less the retention
time for methane (t - t ) and VL . is the peak width at half height.
r m 0.5
120
-------
TAPI.E 45. CIIROrtAKHiRAl'HIC CONDITIONS USED IN THE EVALUATION OF CAPILLARY COMPOUNDS BY GC/FID
Parameter
Co Inan dimensions
Cartiti gas
Flow velocity
(c«/eer)
Makeup gas (nitrogen)
flow (ml/arin)
Injector temperature (VC)
Column temperature
program:
Pcrfonaaoce check
Base/neutrals
Acids
_SP;2100_
0.25 ¦ 10 ¦ 10 ¦
nitrogen
17
30
300
isothermal at 120°C
4 din at 50°C, then
to 2*0°C at 4<>C/min
2 ain at 70°C, then
to 270°C, at 8°C/mia
Fused silica SP-2I0O
(Carbowax 208
deactivated)
0.21 mm ID X 25 m
nitrogen
30
30
300
isothermal at 100°C
I ain at 50°C, then
to 280*C at 40C/min
I "in at 50®C, then
to 240°C at AT./*in
Colui»n type
M
a 60 cc/scc for base/neutral runs and 45 cm/sec for acid runs,
b Not tested with this solution.
SE-H
0.24 mm II) x 18 a
niIrogen
25
30
300
isothermal at 90°C
50-280'C at 4°C/siin
2 sin at 45°C, then
to 250"C at IO°C/min
SE-52
0.24 m ID i 18 a
hydrogen
25
30
300
Fused silica
J5E^54
0.24 mm W x 15
nitrogen
31
30
300
isothermal at 85°C
5 tain at 40°C, then 4 min at 50°C, then
to 280"C at l0oC/min to 280®C at 4°C/aiin
b 2 min at 70°C, then
to 280*C at 8sC/min
Fused silica
SE-54^
0.24 m 10 x 15 ¦
hydrogen
60, 45*
30
300
isothermal at S5°C
4 ain at 50°C, then
to 2B0°C at 4®C/»in
2 min at 70°C, then
to 280°C at 8°C/min
-------
TABLE 46. COMPOSITION OF CAPILLARY COLUMN PERFORMANCE
TEST MIXTURES
Test mixture Composition
Varian 2-octanone 0.2 (Jg/Ml
P/N 82-005049-01 1-octanol 0.2
(nonpolar) naphthalene 0.2
2,6-dimethylphenol 0.2
2,4-dimethylaniline 0.2
C12-alkane 0.2
C13-alkane 0.2
Alltech Ci3-alkane 0.1 (Jg/pl
TP-5 (polar) C14-alkane 0.1
C15-alkane 0.1
C16-alkane 0.1
1-octanol 0.5
5-nonanone 0.3
2,6-dimethylaniline 0.4
2,6-dimethylphenol 0.4
naphthalene 0.5
-------
0 5 10 15 20
Time, Minutes
Figure 36. HRGC/FID chromatogram of a performance test solution
on the SP-2100 capillary column.
123
-------
1 I I I I I !
0 5 10 15 20 25 30
Tim», Miriuttt
Figure 37. HRGC/FID chromatogram of a performance test solution on the
fused silica SP-2100 (Carbowax 20M deactivated) capillary column.
124
-------
Cl2
-o
i
!
1
10 15
Tim*,- Minute*
20
25
Figure 38. HRGC/FID chromatogram of a performance test solution
on the SE-52 capillary column eluted with nitrogen.
125
-------
"D
si
'14
-Z
a
o
c
V
£
i
>o
¦o
\
x
•o
m
ts
10.
Tir
20
I, Minutes
25
30
35
40
Figure 39. HRGC/FID chromatogram of a performance test solution on the
fused silica SE-54 capillary column eluted with nitrogen.
126
-------
TABLE 47. RESULTS OF THE WCOT CAI'ILLARY COLUMN EVALUATIONS
Coluan type
SP-2100
SP-2100 (Carbovax 20N)
SE-52
SE-54
Carrier
Nitrogen
Nitrogen
Nitrogen
Nitrogen
Hydrogen
Coopound
selected
for IIKTP
(deternined)
CIS
CIS
C1S
CIS
C|S
ItETP
("•")
1.45
0.48
0.6
0.36
0.25
1-Octanol
220
150
100
100
Asymetry (AS)
Naphthalene Octanone
AdsorptIon
1-Octanol/C,j* DHA/C,4b
120
100
210
110
130
a Ratio of peak heights, 1-octanol/Cl2.
b Ratio of peak heights, 2,4-di»ethylaniline/C,j.
c Ratio of peak heights, 2,6-di«ethyIphenol/C|j.
d Ratio of peak heights, 2,4-di«elhylani1ine/2,6-di«ethylphenol.
e Asywaetry not determined. Octanone was not present in the test nixture used.
f Ratio of peak heights, l-octaool/Ci0-
300
92
100
0.3
0.1
0.8
o.«of
0.99
1.12
0.75
0.84
0.71
DHP/C,jc plld
1.02
1.08
1.58
1.33
1.16
0.97
1.12
0.47
0.63
0.62
K>
-------
Peak asymmetry (AS), a measure of the adsorption activity, was calculated for
selected compounds from the formula:
W
AS = ^ x 100
F
where Wg and W^, are the back and front baseline widths of the peak mea-
sured from a line bisecting the peak maximum. AS values were calculated for
1-octanol, naphthalene, and octanone. A more reliable indication of a col-
umn's activity for adsorption is the ratio of peak heights for a compound sus-
ceptible to adsorption and an inert compound. These adsorption ratios were
calculated for 1-octanol, 2,4-dimethylaniline, and 2,6-dimethylphenol versus
dodecane. The aniline and phenol adsorption values, as well as the pH of the
column (the ratio of peak heights for the aniline and phenol), provide an in-
dication of the acidic or basic character of the columns.
Of the four columns tested, the SE-54 fused silica column exhibited the
best overall performance. The HETPs for the SE-54 column indicate much higher
resolving power. The asymmetry and adsorption parameters indicate that the
column, although slightly acidic, is fairly inert. The influence of the car-
rier gas on the performance of the SE-54 fused silica column was only appar-
ent in the lower HETPs observed with hydrogen. Helium, typically used for
GC/MS should provide similar performance with HETPs between those observed
for nitrogen and hydrogen.
The absolute and relative retention times and response factors for the
extractable spiking compounds are listed in Tables 48 and 49. All relative
retention times and response factors were calculated using the dio-anthracene
internal standard for reference. The shorter retention times observed for
compounds chromatographed on the SE-52 and SE-54 columns with hydrogen as the
carrier gas reflect the higher linear velocity needed to achieve maximum effi-
ciency. The use of hydrogen significantly shortened the total analysis time
over nitrogen elution with no sacrifice in resolution. The chromatography
observed was generally satisfactory for all compounds on all columns except
that two pairs of compounds, 3,3'dichlorobenzidine and bis(2-ethylhexyl)phthal
ate and pentachlorophenol and d^-anthracene, were not separated on the SE-52
column. The latter pair was also not separated on the SE-54 column. However,
the best chromatography for benzidine was observed on the SE-54 column.
Although good chromatography was observed for the acidic compounds in
standard solutions on all column systems, the phenols are likely to be very
susceptible to adsorption on a partially active column. Analysis of acidic
sludge extracts, cleaned by GPC only, would likely increase the adsorption
character of capillary columns. In addition, the extracts for many sludges
contain high levels of aliphatic hydrocarbons not removed by GPC which can
hinder GC/MS analysis. Hence, capillary GC/MS may not be practicable for
routine analyses of acidic sludge extracts.
128
-------
TABLE 48. RELATIVE RETENTION TIMES AND RELATIVE RESPONSE FACTORS DETERMINED
FOR THE BASE/NEUTRAL SPIKING COMPOUNDS BY CAPILLARY GC/FID
Fused silica SP-2100
SP-2100 (Cjrbowx 20M deactivated)
RT RF RT RF
Compound (am) RRT (nsa/ng) RRF (nun) RRT (nm/ng) RRF
1,4-Dichloroben2ene
14.57
0.
.35
0.63
0.
.60
8.56
0.
26
1.
,42
2.37
Hexachloroethane
17.52
0.
.42
0.27
0,
.26
10.63
0.
32
0.
.33
0.55
Bis(2-chloroisopropyI)ether
16.54
0.
,39
0.63
0.
.60
9.84
0.
30
0.
.64
1.07
Bis(2-chloroethyl)ether
13.19
0.
.31
0.26
0.
.25
7.78
0.
24
0.
.97
1.62
Acenapbtbylene
31.50
0.
.75
1.89
2.
.80
23.03
0.
70
1.
.60
2.67
2,6-Dinitrotoluene
30.91
0.
,74
0.41
0.
.39
23.52
0.
71
0.
.24
0.40
Fluoranthene
49.21
1.
.17
0.46
0.
.44
39.96
1.
21
0.
.26
0.43
Benzidine
51.38
1.
,23
0.007
0.
.007
ND
-
ND
-
3,3'-Dichlorobenzidine
59.45
1.
.42
0.003
0.
.003
46.46
1.
41
0.
.12
0.20
n-Butylbenzylphthalate
55.73
1.
,33
0.30
0,
.29
45.87
1.
39
0.
.23
0.38
Bis(2-ethylhexyl)pbthalate
60.24
1.
.44
0.56
0,
.53
51.08
1.
55
0
.06
0.10
Benzofa]pyrene
67.13
1.
.60
0.02
0,
.02
57.28
1.
74
0
.003
0.005
d10"Anthracene
41.93
1.
.00
1.05
1,
.00
32.97
1.
00
0.
.60
1.00
SE-52
(N5)
SE-52 (H?)
RT
RF
RT
RF
Compound
Imin)
RRT
(um/ng)
RRF
(min)
RRT
(mra/ag)
RRF
1,4-Dichlorobenzene
9.06
0.
.26
1.49
1,
.52
9.55
0.
45
2.
.53
3.24
Hexachloroethane
11.02
0
.32
0.29
0
.30
10.83
0.
,50
1
.43
1.83
Bis(2-chloroisopropyl)ether
10.43
0,
.30
0.59
0
.60
10.63
0.
.49
1
.05
1.35
Bis(2-chloroethyl)ether
8.46
0.
.25
1.02
1
.04
9.25
0.
,43
2
.34
3.00
Acenaphthylene
24.21
0.
.70
0.94
0
.96
17.13
0.
.80
3.
.81
4.88
2,6-Dinitrotoluene
25.39
0,
.74
0.04
0
.04
17.72
0.
.83
0
.32
0.41
Fluoranthene
41.54
1,
.21
0.03
0,
.03
24.41
1.
.14
0
.65
0.83
Benzidine
44.29
1,
.29
0.02
0
.02
ND
-
-
-
3,3'-Dichlorobeozidine
51.57
1,
.50
0.05
0
.05
28.54
1.
.33
b
-
n-Butylbenzylphthalate
48.43
1,
.41
0.07
0,
.07
26.97
1.
.26
0
.58
0."4
Bis(2-ethylhexyl)phthalate
52.36
1.
.52
0.08
0
.08
28.54
1.
,33
b
-
Benzol a]pyrene
59.06
1.
.71
0.03
0,
.03
31.79
1.
.48
0
.10
0.13
dio'Anthraceoe
34.45
1.
.00
0.98
1,
.00
21.46
1.
,00
0
.78
1.00
SE-54
(N,)
SE-
•54 (H»)
RT
RF
RT
RF
Compound
(nia)
RRT
CiBB'ng)
RRF
(min)
RRT
(mm/ng)
RRF
1,4-Dichlorobenzene
9.25
0.26
2.85
13.57
3.15
0.12
• 4.50
8.04
Hexachloroethane
11.81
0.33
0.60
2.86
4.72
0.18
0.63
1.13
Bis(2-chloroisopropyl)ether
11.22
0.31
0.66
3.14
4.53
0.17
0.92
1.64
Bis(2-chloroethyl)ether
8.66
0.24
1.77
8.43
2.76
0.10
3.33
5.95
Aceoaphthylece
25.39
0.71
0.92
4.38
17.32
0.65
1.51
2.70
2,6-Dinitrotoluene
26.57
0.74
0.05
0.24
18.50
0.70
0.34
1.62
Fluoranthene
42.72
1.19
0.20
0.95
33.27
1.25
0.27
0.48
Benzidine
44.49
1.24
0.02
0.10
35.43
1.33
0.03
0.14
3,3'-Dichlorobeozidine
52.36
1.46
0.02
0.10
42.32
1.59
0.06
0.29
n-Butylbenzylphthalate
49.61
1.38
0.15
0.71
40.35
1.52
0.33
0.59
Bis(2-ethylhexyl)phthalate
53.74
1.50
0.12
0.57
44.69
1.68
0.22
0.39
Benzo[a]pyrene
59.65
1.66
0.05
0.24
48.62
1.83
0.05
0.09
d10-Anthracene
35.83
1.00
0.21
1.00
26.57
1.00
0.56
1.00
a ND - not detected.
b Response factors were not determined for 3,3'-dichlorobenzidine and
bis(2-ethylhexyl)phthalate because tbey coeluted.
129
-------
TABLE 49. RELATIVE RETENTION TIMES AND RELATIVE RESPONSE FACTORS DETERMINED
TOR THE ACIDIC SPIKING COMPOUNDS BY CAPILLARY GC/FID
fused silica SP-2100
SP-2100 (Carbowax 20M deactivated)
Compound
RT
(mm)
RRT
RF
(ata/nx)
ROT
RT
(nin)
RRT
RF
(¦n/n*)
RRF
Phenol
9.06
0.37
0.44
0.85
11.22
0.34
0.81
0.79
2,4-Dinethylphenol
12.80
0.S3
0.12
0.23
15.35
0.47
0.76
0.75
2,4-Dichlorophenol
13.39
0.55
0.09
0.17
16.54
0.50
0.24
0.24
Pentachlorophenol
23.82
0.98
0.09
0.17
34.84
1.06
0.02
0.02
D1o~Anchracene
24.21
1.00
SE-52
0.52
(N,)
1.00
32.87
1.00
1.02
1.00
CoBoound
RT
(nin)
RRT
RF
(mn/ag)
RRF
Phenol
8.66
0.44
0.86
0.88
2,6-Diae tbylpheno1
11.02
0.56
1.06
1.08
2,4-Di chlorophenol
11.61
0.59
0.17
0.17
Pentachlorophenol
19.69
1.00
0.07
0.07
D]o"Anthracene
19.69
1.00
SE-54
0.98
(V,)
1.00
SE-
54 (H»)
Coocoucd
RT
(nin)
RRT
RF
(nm/n*)
RRF
RT
(ain)
RRT
RF
(mn/ng)
RRF
Phenol
4.53
0.24
3.31
5.34
2.17
0.14
5.16
15.18
2,4-DiBethylpbenol
7.68
0.41
2.11
3.40
4.92
0.32
0.90
2.65
2,4-Dichlorophenol
8.27
0.44
1.07
1.76
5.31
0.35
0.62
1.S2
Pentachlorophenol
18.5
0.99
ND4
-
15.34
1.00
0.12
0.35
D10-Anthracene
18.7
1.00
0.62
1.00
15.35
1.00
0.34
1.00
a ND * not detected.
130
-------
Evaluation of the SE-54 Fused Silica Column for Analyses of Sludge Extracts—
The suitability of the SE-54 fused silica column for analyses of sludge
extracts was evaluated by chromatographing the spiked sludge extracts prepared
for the evaluation of adsorption chromatographic cleanup procedures. In addi-
tion, retention time and response factor data were generated for the entire
list of B/N priority pollutants.
Experimental--Extracts that were prepared from 80-ml aliquots of sludge,
spiked with the B/N spiking compounds, and then cleaned with florisil or sil-
ica gel were analyzed by GC/MS using the 15-m SE-54 WCOT capillary column
eluted with helium.
Solutions of the B/N priority pollutant compounds at various concentra-
tions were analyzed with the same capillary GC/MS system, and retention times
and response factors were compiled.
Results and discussion--The capillary GC/MS, or high resolution GC/MS
(HRGC/MS), chromatograms for composited and individual fractions from silica
gel cleanup of spike sludge extracts are shown in Figures 40 to 43. Compari-
son of these chromatograms with the packed column GC/MS chromatograms for sim-
ilar fractions from florisil cleanup (Figures 31 to 34) illustrate the much
better chromatographic resolution provided by the capillary column. Interpre-
tation of the HRGC/MS data was made considerably easier by the separations
achieved.
The relative retention time data generated for the B/N compounds are
shown in Table 50. Table 51 shows the relative response factors determined
for many of the B/N priority pollutants. The relative response factors for
most of the compounds, except benzidine and 2,6-dinitrotoluene, were fairly
consistent over the range of 4 to 200 ng injected. Nonetheless, these data
confirm the principle that the response for a particular compound should be
checked over the entire range of concentrations quantitated.
SELECTION OF METHODS FOR EXTRACTABLE COMPOUNDS IN MUNICIPAL
AND INDUSTRIAL WASTEWATER TREATMENT SLUDGE
Although the methods described in the preliminary POTW sludge protocol
for analyzing base/neutral and acidic compounds provided acceptable precision
and accuracy for most compounds, some shortcomings were apparent. Homogeniza-
tion/centrifugation extraction of 320-ml aliquots of sludge was both labor-
intensive and time-consuming. The rigorous extraction also produced an extract
enriched in interfering coextractants which challenged the capacity of the
GPC cleanup. Unfortunately, none of the alternate extraction procedures was
found to provide recoveries for the broad range of the basic, neutral, and
acidic compounds that were comparable to those demonstrated for homogenization/
centrifugation. Hence, the homogenization/centrifugation extraction was recom-
mended for analyses of both municipal and industrial sludge with one modifica-
tion. Extraction of a single 80-ml aliquot of sludge was recommended to de-
crease the time required for extract preparation. Since extracts from 80-ml
aliquots were analyzed at a smaller final volume, there was no real decrease
in method sensitivity.
131
-------
Figure AO. I1RGC/MS chromatogram of the combined eluent (Fractions II-IV)
obtained by silica gel cleanup of the spiked sludge extract
(no GPC cleanup).
-------
100.
00
10:00
5341180.
M
iooo
29:88
1508
30:00
2000
25C0 SCAN
50:00 TIltE
Figure 41. I1RGC/MS chromatogram of Fraction II obtained by silica gel
cleanup of the spiked sludge extract (no CPC cleanup).
-------
1C9.0-
5996429.
RIC_
CO
->
mJ
ed
H
<0
JS
>\
X
01
JZ
lVAv.
j
500
10. Ci)
1000
20.80
1530
30; 00
20UO
48=60
25C8 SCAN
50; s» TIME
Figure 42. HRCC/MS chromatogram of Fraction III obtained by silica gel
cleanup of the spiked sludge extract (no GPC cleanup).
-------
56279M
100
BIC
c.
Figure 43. IIRGC/MS chromatogram of Fraction TV obtained by silica gel
cleanup of the spiked sludge extract (no GPC cleanup).
-------
TABLE 50. RELATIVE RETENTION TIMES FOR THE BASE/NEUTRA1
COMPOUNDS ANALYZED BY CAPILLARY COLUMN GC/MS
Compound name
RRTa
N-nitrosodimethylamine
0.04
Bis(2-chloroethyl)ether
0.21
1,3-Dichlorobenzene
0.22
1,4-Dichlorobenzene
0.22
1,2-Dichlorobenzene
0.25
Bis(2-chloroisopropyl)ether
0.29
Hexachloroethane
0.29
N-Nitrosodi-n-propylamine
0.31
Nitrobenzene
0.32
Isophorone
0.36
Bis(2-chloroethoxy)methane
0.41
1,2,4-Trichlorobenzene
0.42
Naphthalene
0.43
Hexachlorobutadiene
0.47
Hexachlorocyclopentadiene
0.60
2-Chloronaphthalene
0.63
Acenaphthylene
0.70
Dimethylphthalate
0.72
2,6-Dinitrotoluene
0.72
Acenaphthene
0.73
2,4-Dinitrotoluene
0.78
Fluorene
0.82
4-Chlorophenylphenylether
0.84
Diethylphthalate ^
0.84
N-Nitrosodiphenylaming
1,2-Diphenylhvdrazine
0.84
0.86
4-Bromophenylphenylether
0.92
Hexachlorobenzene
0.93
Phenanthrene
0.98
Anthracene
0.99
DiQ-Anthracene
1.00
Di-n-butylphthalate
1.14
Fluoranthene
1.19
Pyrene
1.23
Benzidine
1.24
Butylbenzylphthalate
1.40
Chrysene
1.45
3,3'-Dichlorobenzidine
1.47
Bis(2-ethylhexyl)phthalate
1.52
Benzo[k]fluoranthene
1.62
Di-n-octylphthalate
1.63
Benzo[a]pyrene
1.67
Dibenzo[a,h]anthracene
1.84
Benzo[g,h,i]perylene
1.87
a RRT = relative retention time based on Dioanthracene
(RRT = 1.00) for compounds chromatographed on a
SE-54 UCOT fused silica capillary (15 m x 0.24 ma ID)
eluted with helium at 10 psi and with a temperature
program of 50°C for 4 min, then 4°C/min to 320#C.
b Elutes a diphenylamine.
c Elutes as azobenzene.
136
-------
TABLE 51. GC/MS RELATIVE RESPONSE FACTORS FOR THE BASE/NEUTRAL PRIORITY POLLUTANTS
Ion used for
Relative response factor3
quantitation
Amount
injected
Compound
(m/e)
4 ng
20 ng
40 ng
80 ng
100 ng
200 ng
1,2-Dichlorobenzene
146
NAb
NA
1.26
1.31
NA
NA
1,3-DichIorobenzene
146
NA
NA
1.10
1.33
NA
NA
1,4-Dichlorobenzene
146
0.38
0.52
NA
NA
0.39
0.46
Hexachloroethane
117
0.25
0.30
NA
NA
0.22
0.25
Bis(2-chloroisopropyl)ether
45
NA
NA
NA
NA
NA
NA
Bis(2-chloroethyl)ether
93
0.42
0.59
NA
NA
0.57
0.54
Nitrobenzene
77
NA
NA
3.72
1.26
NA
NA
Hexachlorobutadiene
225
NA
NA
0.42
0.47
NA
NA
Naphthalene
128
NA
NA
3.32
2.59
NA
NA
2-Chloronaphthalene
162
NA
NA
1.54
1.38
NA
NA
Acenaphthylene
152
1.73
1.76
NA
NA
1.26
0.55
Acenaphthene
154
NA
NA
1.24
1.28
NA
NA
2,6-Dinitrotoluene
165
0.13
0.23
NA
0.40
0.21
0.17
Fluorene
166
NA
NA
1.11
1.31
NA
NA
4-Broinophenylphenylether
248
NA
NA
0.31
0.31
NA
NA
Fluoranthene
202
0.88
0.86
1.08
1.18
0.81
0.47
Pyrene
202
NA
NA
1.08
1.11
NA
NA
Benzidine
184
0.04
0.11
NA
NA
0.22
0.15
3,3'-Dichlorobenzidine
252
NA
NA
NA
NA
NA
NA
n-Butylbenzylphthalate
149
0.55
0.55
NA
NA
0.61
0.44
Bis(2-ethylhexyl)phthalate
149
0.94
0.98
NA
NA
0.93
0.47
Benzo[a]pyrene
252
0.24
0.25
NA
NA
0.28
0.21
djo'Anthracene
188
1.00
1.00
1.00
1.00
1.00
1.00
a Relative to djo_anthracene.
b NA = not analyzed.
-------
Both GPC and adsorption chromatography on florisil and silica gel pro-
vided acceptable extract cleanup and recoveries for the B/N compounds. The
florisil and silica gel adsorbent methods were essentially equivalent. The
florisil and silica gel methods removed aliphatic hydrocarbons and low and
high molecular weight polar interferences. Although the GPC method only re-
moves mostly high molecular weight biogenic materials, it is easier to auto-
mate and is more amenable to routine sludge analyses. GPC was the only clean-
up procedure that provided good recoveries for the acidic compounds. Hence,
GPC and adsorption chromatography on florisil or silica gel were recommended
as options for cleaning B/N extracts. GPC was recommended for cleanup of
acidic extracts.
Although the packed column GC/MS method described in the preliminary POTW
protocol provided reliable analyses of sludge extracts, the SE-54 WCOT fused
silica capillary column was recommended as the preferred option for the GC/MS
procedure for B/N extracts. The better resolution and inertness of the SE-54
capillary column provided GC/MS data that were easier to interpret. However,
the HRGC/MS method should only be used for extracts cleaned by chromatography
on florisil or silica gel. The packed column GC/MS method was recommended
for analyses of acidic extracts. The level of cleanup provided by the GPC
cleanup for acidic extracts may make it difficult to maintain the inactivity
of the capillary column.
The revised analytical scheme for extractable priority pollutants in
municipal and industrial wastewater treatment sludges is shown in Figure 44.
The detailed analytical protocol prepared from this scheme is appended to
this report.
EVALUATION OF THE PRECISION AND ACCURACY OF THE METHOD
The precision and accuracy of the extractables protocol (Appendix B)
were evaluated by determining recoveries of the spiked compounds from five
primary sludges. The sludges used in these evaluations were those described
in Table 18. Recoveries were determined for the extractable compounds forti-
fied into aliquots of these sludges at three spiking levels. The spiking
levels were chosen to be 2, 20, and 200 times the individual detection limit
typically observed for each compound in sludge. Two combinations of cleanup-
analysis options were evaluated for the B/N compounds. B/N extracts were ana-
lyzed by packed column GC/MS following GPC cleanup and by capillary GC/MS fol-
lowing florisil cleanup. All acidic sludge extracts were cleaned up by GPC
and analyzed by packed column GC/MS.
Eleven 80-ml aliquots of each of the five primary sludge samples were
placed in 200-ml centrifuge bottles fitted with TFE-lined septa. Three sets
of triplicate bottles were spiked with the extractable spiking compounds at
400, 4,000 and 40,000 |jg/liter sludge. Benzidine was spiked at 800, 8,000,
and 80,000 pg/liter sludge. The spikes were added in 2 to 3 ml of acetone
solution. Immediately following the addition of the spikes, samples were
homogenized for 30 to 45 s and stored at 4°C overnight prior to analysis.
The unspiked sludges were analyzed in duplicate and the spiked sludges were
analyzed in triplicate by the procedures described in Method Appendix.
138
-------
Sludge
(80 ml)
Extract
Sludge
OPTION B
OPTION A
Sludge
or
Extract Cleanup
or
Determine Base/ Neutrals and Pesticides
Discard
Adjust to pH£2
with 6M HCI
Adjust loph2;ll
with 6N NaOH
Adsorption
Chromatography
on Silica Gel
Capillary Column
GC/MS with
SE -54 WCOT
Packed Column
GC / MS on
SP - 2250
Determine Phenols
by GC/MSon
SP-1240-DA
Clean Up by GPC on
Bio Beods SX-3 Eluted
with Ch,CI9
Clean Up by GPC on
Bio Beads SX-3 Eluted
with CH2CI2
Extract 3X with CH^CI
by Homogenization/
Centrifugation
Extract 3X with CH2CI2
by Homogenization/
Centrifugation
Determine Base/Neutrals
and Pesticides by GC/MS
on SP - 2250
Figure 44. Scheme for analysis of extractable organics in sludge.
-------
Results and Discussion
The results of the precision and accuracy determinations for the B/N
compounds obtained with GPC cleanup and packed column GC/MS are shown in
Table 52. The GC/MS chroraatograms shown in Figures 45 to 52 are representa-
tive samples of the data generated. Figures 45 to 48 are chromatograms for
unspiked and spiked aliquots of Platte County POTW sludge. Chromatograms for
unspiked aliquots of sludges from the remaining four plants are shown in
Figures 49 to 52. Many of the recovery determinations were likely perturbated
by high concentrations of the analytes in the unspiked sludge samples. Recov-
eries were not calculated for cases where the unspiked concentration was greater
than 10 times the spike level. Although the recoveries varied somewhat from
sludge to sludge for many compounds, all compounds were recovered from all
five sludges spiked at 40,000 pg/liter. Recoveries were generally lower for
the polar compounds, including 2,6-dinitrotoluene, the benzidines, and bis-
(2-chloroethyl)ether. However, only 2,6-dinitrotoluene was not recovered from
all sludges spiked at 400 pg/liter. Of all recovery determinations, 49% fell
within the range of 50 to 150%. An additional 19% were lower than 50% recov-
ery, and only 8% of the recoveries were greater than 150%.
The accuracy of the method was somewhat dependent on the particular
sludge samples. However, the precisions of the recovery determinations were
very good. The relative standard deviations for triplicate determinations
were 30% or less for 95% of the measurements and 10% or less for 62% of the
measurements.
The recoveries for B/N compounds from spiked sludge obtained with florisil
cleanup and capillary GC/MS are shown in Table 53. Figures 53 to 60 show the
HRGC/MS chromatograms for unspiked and spiked sludges from the Platte County
POTW and for unspiked sludges from the other plants. Although these sludges
are replicate samples of those spiked and analyzed with the packed column GC/MS
method, the levels determined in the unspiked samples by the capillary column
GC/MS method were lower. This likely reflects losses during sample storage
as the samples were extracted for the capillary column GC/MS experiments 2 to
3 weeks later than for the packed column GC/MS study. In general, the recover-
ies were similar to those observed with the packed column GC/MS option. How-
ever, all compounds were recovered from all five sludges spiked at 400 |Jg/liter.
An investigation of the distribution of the recoveries shows that 48% were
within the range of 50 to 150%, 28% were less than 50%, and 13% were higher
than 150% recovery. The reproducibility of the determinations was slightly
poorer. Fifty-six percent of the relative standard deviation (RSD) values
were 30% or less. Of these, 30% are less than 10% RSD. The poorer precision
observed with the HRGC/MS method may reflect the influence of repeated injec-
tions of sludge extracts on the performance of the capillary column. Concen-
trated samples can overload the column and degrade the column performance.
The reproducibility of HRGC/MS determinations for extracts from the high level
spikes was better than for other extracts. Eighty-four percent of the extracts
from samples spiked at 40,000 jjg/liter had RSDs of less than 30%.
140
-------
TAB1.K 52. HKSlll lS OF 1'Rfc.CISlUN AND ACCURACY KVAI IIATIONS H»K THE EXTKACTABI.F.S METHOD
(I'ACKEI* W'UIMN CC/MS) AIMM.lhh TO 1 UK ANALYSIS OF BASF./NK4ITRAI. CONIHUINDS
IN MUNItU l»AI. AND 1N0US1KUI. WASTEWATER IRKAIMENT SLIUXIES
Platte County POWW
Spike Unspiked* Spike
conc. sludge Recovery cone.
Co»i»ound (yg/L) n< ent ut mmi determined in Lhr Uiib(i i kt-il sludge Wj.s
lii ^Itcr III.hi I Im- spike level.
Blue River POTW Kansas Cily, Ks Industrial No. 1 In«lust.ria_l No. 2
UnspikedJ Spike Unspikeda Spike Unspiked* Spike Unspiked*
conc. Recovery conc. conc. Recovery cone. conc. Recovery conc. rone. Recovery
- (hb'M... _ {%!-.. (tV1 L_ . Wl > ill. W il'S/LL (X). ..
47
37
+
21
400
21
U
4
3
400
NU
51
4
4
400
820
c
77
4
7
4,000
93
4
9
4,000
73
4
5
4,000
88
4
12
87
~
7
39,990
90
4
9
39,900
70
4
4
40,000
92
~
9
ND
0
400
Nl)
0
400
NO
17
4
5
400
ND
0
54
~
3
4,000
47
~
8
4,000
55
i
9
4,000
97
1
6
59
i
5
40,010
57
~
10
39,990
51
4
7
40,000
77
4
4
ND
)00
+
54
400
ND
0
400
ND
88
4
6
400
ND
0
380
~
37
4,000
510
t
123
4,000
120
4
14
4,000
200
i
4
280
~
8
39,990
410
4
53
40,000
95
4
11
40,000
130
*
7
1,4 30
43,
25C
400
ND
0
400
6
65
4
3
400
ND
200
4
34
42
~
7
4,000
82
~
12
4,000
89
4
5
4,000
130
t
52
41
~
2
40,000
98
i
10
39,000
88
4
3
40,000
93
4
2
NU
0
400
ND
0
400
Nl)
0
400
ND
0
20
~
I
4,000
0
4,000
59
4
8
4,000
150
4
9
73
±
1
40,040
78
~
20
39,990
99
4
14
40,000
97
4
5
8,330
e
400
124
16 i
2
400
14
56
4
8
400
670
c
59
~
18
4,000
52
~
14
4,000
82
t
7
4,000
120
4
1
20
~
2
40,000
100
i
6
39,920
100
4
14
40,000
95
4
4
ND
0
800
ND
0
800
ND
38
4
8
800
ND
0
0
8,000
0
8,000
82
t
10
8,000
120
4
7
96
1
12
80,000
160
f
13
80,1)00
130
4
6
70,000
50
4
2
NO
0
402
535
c
400
Nl)
0
NA
ND
.
100
~
14
4,020
no
~
13
4,000
0
NA
-
-
NA
-
NA
-
NA
3,090
c
400
5,950
e
400
ND
57
i
8
400
12
,500
e
150
t
17
4,000
210
~
I36C
4,000
54
t
5
4,000
c
31
4
2
40,000
300
4
30
401000
85
4
7
40,000
80
4
4
2,850
i
400
1,790
c
400
85
30
4
3
400
24
,200
C
120
I
5
4 ,000
190
4
80
4,000
42
4
4
4,000
230
4
4C
28
i
3
40,010
180
4
21
40,000
53
i
9
48,820
200
~
10
4,050
e
400
NU
0
400
Nl)
30
4
I
400
308
0
290
~
bf
4,000
80
4
8
4,000
33
t
4
4,000
140
4
1
68
*
10
40,010
140
4
19
40,080
48
4
4
40,000
97
~
6
e Com fill r^t ion determined in itic uuspikrd sludge was
IU t iBM-s higher Hi.iii the spike level .
1 NA = this tonpotiud was. not jvail.il>If when these
s.mplos wen' spiked.
d Nl) not del, , ted
-------
475
458
370 g
496
327
523
280
RIC
303
a
182
SCAN
TIME
400
20:30
600
38:00
300
15:00
180
5:00
Figure 45. CC/MS chromatogram for base/neutrals in unspiked Platte County primary POTW sludge.
-------
100*0"
RIC
U>
8538060
u
(U
JZ
u
« *»
c ^
« -H
N ~* V
c j= e
« «-» (B
X «l £
O O 4J
^ W. 01
0 o o
»-H r-4 M
43 £ O
U U »—4
-H I J2
O
-------
100.0-1
4>-
¦t-
I
400
20:00
r
500
25:00
I
200
10:00
30:00
15:00
4997110
SCAN
TIME
Figure 47. GC/MS chromatogram for base/neutrals in Platte County primary POTW sludge
spiked at 4,000 |jg/Lwitli selected compounds.
-------
6183838
JC.
RIC
O.
w
•C
464
X
553
247
154
537]
488
330 356
275
380
688 SCAN
188
288
488
5:88 18:08 15:86 28:88 25:88 38:88 TIME
Figure 48. GC/MS chromatogram for base/neutrals in Platte County primary POTW sludge
spiked at 40,000 yg/L with selected compounds.
-------
190.0-1
RIC
•p-
CTn
175 206
3346430
—I
600
30:08
—I
700
35:00
—I
100
5:00
—I
200
10:00
—I
300
15:00
l
400
20:00
I
500
25:00
SCAN
TIME
Figure 49.
GC/MS chromatogram for base/neutrals in unspiked Blue River primary POTW sludge.
-------
Figure 50. CC/MS chromatogram for base/neutrals in unspiked Kansas City, Kansas, primary POTW sludge.
-------
2461690
Figure 51. GC/MS chromatogram for base/neutrals in unspiked Industrial No. 1 primary sludge
-------
313
604
266
240
638
211
o.
178
139
09
RIC
S3
102
122
2'jO
5:00 10:08 15: ¦j3 2u:08 15:£>0 30:U0 35:00 TItE
Figure 52. GC/MS chromatogram for base/neutrals in unspiked Industrial No. 2 primary sludge.
-------
IABI.K 53. kfcSHU-S Of f'KtClSHW AND AlttURACY KVAU-ATIONS KOR THE tXTRACTABI.ES MEl'HOl)
iUAl' I I.I AKV CiHUMN CC/HS) AtTUhn TO THK ANALYSIS OK BASE/NEUTRAI. LOMMMINUS
IN MUNICIPAL AN1I INtHISTRtAI. WASTEUATKK I'kKATMKNT SI.IJDCES
Ui
o
Platte Coiinly POWW
_ Compound
Spike
coor.
(|Jg/L)
Ufjspiked
sludge
Recovery
U)
Spike
COBC .
w)
1,4-Di rlilorobenzene
*17
21
52
i
w>
400
4,160
100
t
1
4,000
39,990
62
±
28
39,900
Hen athloroel bane
441
NO
21
i
14
400
4,400
22
i
6
4,000
40,010
46
j
12
40,010
Bi&(2-cliloro«thyl )-
408
ND
)&
t
63
400
tlkrr
4,070
140
~
16
4,000
39,990
130
t
81
39,990
ActnapliUiy lene
411
ND
140
i
72
400
4,090
230
t
39
4,000
40,000
B8
i
26
40,000
2,fc-Oiuitrot oJueue
407
ND
43
t
27
400
4,050
5
~
1
4,000
40,040
78
~
5
40,040
Fluoranitieiie
428
27
230
t
122
400
4,270
190
t
26
4,000
40,000
89
±
23
40,000
Benzidine
789
ND
0
800
7,870
1
~
1
8,000
80,000
40 1
11
80,000
3,3' -Dithlorobenzidine
415
ND
5
£
8
NA*
4,130
140
~
32
NA
NA
-
HA
n-Uuly1 frenzylphlbalate
452
770
290
*
400
4,510
160
t
26
4,000
40,000
120
J
16
40,000
Di-n-otlylphlbalate
444
3,500
a
400
4,430
150
t
41
4,000
40,010
130
t
5?
40,010
ttenzo|a|i>yrene
406
84
350
i
202
400
4,050
100 1
21
4,000
40,010
140
j
25
40,010
a tied it ul l wo det cinmal ions •
b nvnti 1 stjinlaid deviation for Mircc Ucler*iiut ions.
r NO - net .li'lrUcd
Rliie Hivi
Itosptked*
i"onc.
l _
34
2,iS D
r wrrw
——
Sp i ke
Recovery
cone.
(I) _
(pg/U
9 ~
2
400
35 t
14
4,000
58 t
8
39,990
3 ~
I
400
10 i
4
4,D00
39 i
6
40 ,010
24 i
15
400
t>«* t
32
4,000
130 1
66
39,990
18 i
6
400
53 i
2
4,000
HO t
4
40,000
7 ~
6
400
4 1
4
4,000
41 t
43
40,040
1 i
2d
400
120 i
8
4,000
120 i
10
40,000
0
800
11 1
4
8,000
50 i
10
80,000
402
NA
-
m
80 1
26d
400
72 t
20
4,000
120 i
19
40,000
d
400
18 i
18
4,000
99 i
11
40,010
63 1
39d
400
46 t
IJ
4,000
no t
29
40,010
I, BOO
21
4
12
40D
41
~
20
4,000
SS
1
18
39,990
0
400
15
+
7
4,000
42
±
6
39.990
45
i
30
400
no
i
66
4,000
240
i
66
40,000
82
1
46
400
150
±
51
4,000
MO
±
9
J9.990
0
400
16
i
28
4,000
76
+
21
39,990
UO
t
69
400
120
i
31
4,000
no
~
4
39,920
6
1
10
800
1
t
1
8,000
81
j
9
60,010
6
1
1
NA
NA
-
NA
t
400
280
~
266d
4,000
no
1
11
40,000
d
400
200
1
91
4,00d
190
1
16
40,000
110
±
51
400
120
i
24
4,000
UO
1
13
40,080
Kansas Cit^i Ks
Uuspiked" Spike
cone. Recovery tone.
CjJjj/L) (X)
Industrial No. )
Dnspiked Spike
cone. Recovery cone.
Jitf/u (%)_ (Hfi/li.
ND
lodustrial No. 2
Un«pikeda
cone. Recovery
!*)_._
24
i
5
400
58
23
t
12
160
t
97
4,000
57
t
8
SS
t
8
40,000
70
±
19
7
t
10
400
ND
3
+
2
161
i
100
4,000
75
1
I
49
~
21
40,000
35
1
5
38
i
9
400
ND
0
130
i
56
4,000
63
~
2
100
i
g
40,000
92
~
19
55
t
5
400
730
d
no
t
35
4,000
93
1
3
100
t
2
40,000
114
1
27
n
t
13
400
1,090
6
160
t
80
4,000
62
i
8
90
i
3
40,000
121
t
32
67
1
3
400
125
0
160
t
64
4,000
121
*
6
88
i
19
40,000
120
1
25
0
780
ND
0
240
4
100
7,800
35
1
17
160
t
6
70,000
53
+
4
NA
_
-
NA
-
-
NA
-
67
t
8
400
9.9JO
t
230
+
36
4,000
d
72
t
ie
40,000
140
1
35
58
t
16
490
2,350
d
140
t
66
4,890
170
i
24
58
t
13
48,820
210
t
62
26
t
2
400
910
d
170
t
110
4,000
57
~
14
42
t
13
40,000
110
±
22
~r:
Z-
: --
e HA - tilts compound wjs not .iv.ii when tWse
staples were spiked.
/ Comviitrji iuh JfliTMtiHii in lh<* ufi?.piked sludge
wjs 10 t jut's higliei tlidii I lie h|>ikr level.
•1 IJijik rill I
-------
01
1682 524287*.
100. »n
1522
1068
1219
RIC.
1129
1366
1273
1135 ^
X'
982 =
445
549 649
74rt
—j 1 i 1 i 1 i 1 i 1 I 1 i 1 i 1
200 400 609 COD 100D 1200 1400 1600 SCAN
3:20 8; 10 10;00 13:20 16:10 20:00 23:20 26:10 TIME
Figure 53. HRGC/MS cliromatogram for base/neutrals in unspiked Platte County primary POTW sludge.
-------
too.e-i
1678
BIC.
Ul
NJ
V c*i
48M7M.
I
200
3; 28
I
400
6:49
—I
6C0
10.00
tiOO
13.20
11)00
16.10
1200
29:08
1400
H5;u
26:48
SCAN
hue
Figure 54. HRGC/MS chroniatogram for base/neutrals in Platte County primary POTW sludge
spiked at 400 ug/L with selected compounds.
-------
BQ
BIC_
o a
to
a.
a o
o
0 SCAN
2CC0
SCO
8:20 I6:-M 25;d!) 33:20 <;1:40 TIME
Figure 55. HRGC/MS cliromatogram for base/neutrals in Platte County primary POTW sludge
spiked at 4,000 pg/L with selected compounds.
-------
ICO.O-i
6Q
BIC.
a.
ta
2COO
33:20
1000
16:40
Figure 56. HRGC/MS chromatogram for base/neutrals in Platte County primary POTW sludge
spiked at AO,000 yg/L with selected compounds.
-------
to*.
Ln
Ln
•f= 750
5169150.
408
6:40
6ca
10; 90
803
13:20
toed
16:40
1200
20:00
1400
23:20
16yd
26:40
SCAN
TIME
Figure 57. HRGC/MS chromatogram for base/neutrals in unspiked Blue River primary POTW sludge.
-------
103.0-1
BIC
Ul
3067906.
600
10:00
600
13; 28
1000
16:49
1200
20:80
1400
23:20
<638
26:40
SCAN
TIME
Figure 58. HRCC/MS cliromatogram for base/neutrals in unsplked
Kansas City, Kansas, primary POTW sludge.
-------
189.
2891779.
lie
1498
Ui
^1
»
x
V
159
253
409
533
J I 1323
200
3;28
n
460
6:49
600
10; 09
—I
800
13; 20
1000
16; 40
1200
20; 00
1400
23; 20
1600
26:40
SCAN
TIME
Figure 59. 11RGC/MS chromatogram for base/neutrals in unspiked Industrial No. 1 primary sludge.
-------
lea.a-i
tic.
<_n
00
1581b
i960 1938
600
10:00
13:20
5(6915*.
1000
16:40
1200
28:00
1400
23:20
16C0
26:40
SCAN
TIKE
Figure 60. HRGC/MS chroniatogram for base/neutrals in unspiked Industrial No.
2 primary sludge.
-------
The recoveries determined for the acidic compounds are shown in Table 54.
Figures 61 to 68 show the GC/MS chromatograms for spiked and unspiked Kansas
City, Kansas, POTW sludge and for unspiked sludges from the remaining four
plants. The recoveries observed were generally good and were reproducible.
No zero recoveries were observed. Although many of the spike levels were
lower than the unspiked concentrations, 65% of all recoveries were in the
range of 50 to 150%, 21% were less than 50%, and 14% were greater than 150%
recovery. The RSD values were low for triplicate determinations, 70% were
30% or less RSD, and 40% were less than 10% RSD. Poor recoveries were deter-
mined for 2,4-dimethylphenol from the Platte County sludge samples. Recover-
ies observed for this compound during the method development experiments were
frequently anomalously low. This may indicate that 2,4-dimethylphenol is sus-
ceptible to biodegradation in some sludges.
In view of the complexity and diversity of municipal and industrial
wastewater treatment sludges, the precision and accuracy results presented
here demonstrate that the protocol developed can be reliably applied to the
analysis of the organic priority pollutants in sludge. The success of this
protocol for the variety of extractable compounds for which it was developed
and evaluated indicates that the methods included may also be useful for many
nonpriority pollutant analytes.
TIME/COST ANALYSIS OF THE METHOD
Unlike the similarity of the time/costs for conducting purgeables analy-
ses of sludge and wastewater, the time/costs for the extractable compounds
are considerably higher for sludge analyses than for wastewater analyses.
However, the same conditions and assumptions are applicable for the estimates
presented here. The analysts are experienced professionals, no major equip-
ment purchases are included, samples are analyzed in lots of 20 to 100, and
analytical standards are prepared from commercially available mixed solutions.
The estimated per sample labor and cost requirements for analysis of
extractable priority pollutants in sludges are shown in Table 55. The acidic
compounds are analyzed in the same manner in both options, i.e., extract clean-
up by GPC and analysis by packed column GC/MS. The two cost options shown in
Table 55 are for the two principal analytical options for the B/N compounds.
The labor and costs for analysis of spiked and replicate samples, sample compos
iting and spiking, and reanalysis of selected extracts must also be estimated
for each specific analysis program.
159
-------
ON
o
RESULTS OF PRECISION AND ACCURACY EVALUATIONS VOR THE EXTRACTAB1.ES METHOD
AI'I'l. I El) Hi Tilt ANALYSIS OK ACIDIC COMPOUNDS IN MUNICIPAL AND
INDUSTRIAL UASTEWAIEK 1KFATMKNT SI.UIK5ES
Platte County POWW^
Compound
Spike
conc.
. L
Uotpiked
sludge
S|>a ke
Recovery couc.
{%)_ . .Juk/l)
_Blue River POTW
Unspiked*
conc.
Recovery
(XI .
Spi ke
conc.
Jys/L)
Kansas Cityx Ks
llti&piked*
conc. Recovery
(M/L) (X)._.
Spike
conc.
(mb" '
Industrial Mo. 1
Unspiked*
conc.
(ys/i.)_
Recovery
Industrial No. 2__
Spike Unspiked*
conc. conc. Recovery
__(yg/U (yg/L)_ (XI
Phenol
464
323
86
1
I4b
400
160
67
1
22
400
2,800
c
400
112
48
t 9
400
800
101
t 48
4,630
52
~
8
4,000
89
t
A
4,000
72
t 71
4,000
49
t 6
4,000
47,
, 120
46,280
69
~
6
40,000
68
i
15
40,000
78
i 14
<•0,000
99,
, 88
40,000
57,
, 66
2,4-Diaetliy Iphenol
412
N0d
7
~
3
400
ND
94
£
10
400
NU
174
t 51
400
ND
88
t 7
400
1,370
c
4,110
4
i
1
4,000
45
t
2
4,000
21
t 9
4,000
90
1 II
4,000
94,
, 232
41,080
5
t
3
40,000
83
i
68
40,000
110,
, 93
190,
, 190
40,000
25,
, 28
2,4-Dichlorophcnol
399
NU
140
1
5
400
72
140
t
7
400
1,780
c
400
NO
69
t 9
400
114
380
t 95
1,970
66
~
13
4,000
93
t
7
4,000
92
1 64
4,000
S3
± 3
4,000
67,
, 140
39,750
89
i
8
40,000
78
i
7
40,000
69
t 17
40,000
102,
, 95
40,000
68,
, 79
Pentachloropbenol
401
323
130
1
29
400
660
70
~
10C
400
1,120
230,
, 230c
400
229
41
t 20
400
13,960
e
4,000
150
±
27
4,000
88
t
6
4,000
202
t 75
4,000
45
~ 3
4,000
c
40,010
79
~
101
39,970
71
~
12
39,970
85
4 47
40,000
79,
, 68
40,000
39,
. 53
a Mean for two deterainations.
b Mean t standard deviation for three deteraination».
c Concentration determined in the uuspiked sludge was higher than the spike level,
d NU - not detected.
e Concentiat ion deterained in the unspiked sludge was 10 tiaes higher than ihe spike level.
-------
285
243
RIC
J3
117
134
138
96
180
50
280
380 SCAN
150
2:30 5:88 7:38 10:08 12:38 15:08 TIME
Figure 61. GC/MS chromatogram for acids in unspiked Kansas City, Kansas, primary POTW sludge.
-------
ico.en
RIC
206
265
c 226
a
173
123
149
158
250
2:30 5:CJ 7:30 10:00 12:30 15:00 TIME
Figure 62. GC/MS chromatogram for acids in Kansas City, Kansas, primary POTW sludge
spiked at 400 yg/l, with selected compounds.
-------
260
8323879
jC
221
RIC
288
HE
129
17G
153
58
188
158
388 SCAN
2:38 5:08 7:38 18:08 12:38 15:88 TIME
Figure 63. GC/MS chromatogram for acids in Kansas City, Kansas, primary POTW sludge
spiked at 4,000 yg/L with selected compounds.
-------
338
7733240
259
RIC
219
0.
279
Ol,
122
148 172
400 SCAN
300
350
250
150
200
2:30 5:60 7:30 10:00 12:30 15:00 17:30 20:00 TIME
Figure 64. GC/MS chromatogram for acids in Kansas City, Kansas, primary POTW sludge
spiked at AO,000 iig/i> with selected compounds.
-------
264
8368838
JZ
279
RIC
228
282
178
a.
137
168
288
258
388 SCAN
188
2:38 5:08 7:38 18:88 12:38 15:88 TIME
Figure 65. GC/MS chromatogram for acids in unspiked Platte County primary POTW sludge.
-------
JC-
o.
BICl
250
12:38
158
7:38
208
18:88
5:88
Figure 66. GC/MS chromatogram for acids in unspiked Blue River primary POTW sludge.
-------
279
2551800
BIC
250
200
50
150
2:30 5:00 7:38 10:00 12:30 15:1* TU*
Figure 67.
GC/MS chromatogram for acids in unspiked Industrial No. 1 primary sludge.
-------
IW.»1
JtIC_
¦C
VJ
259
12.38
Figure 68. GC/MS chromatogram for acids in unspiked Industrial No. 2 primary sludge.
-------
TABLE 55. TIME/COST ESTIMATE (PER SAMPLE) FOR ANALYSIS
OF EXTRACTABLE COMPOUNDS IN SLUDGE
Option Aa
Option b'3
Operation
Labor
(h)
Other
GC/MS costs
(h) ($)
Labor
(h)
GC/MS
(h)
Other
costs
($)
Extraction
4.8
4.8
Extract cleanup
GPC
Florisil or silica gel
0.4
1.3
0.7
GC/MS
2.7
2.7
2.5
2.5
Data retrieval and
interpretation
2.8
2.8
Total
12.0
2.7 32
10.8
2.5
27
a B/N extracts cleaned up by florisil or silica gel chromatography
and analyzed by capillary column GC/MS.
b B/N and acid extracts cleaned up by GPC and analyzed by packed
column GC/MS.
169
-------
SECTION 6
EVALUATION OF THE APPLICABILITY OF THE PRELIMINARY POTW
SLUDGE PROTOCOL METHODS FOR SEDIMENT ANALYSIS
Following the development of the preliminary POTW sludge protocol,4 the
purging, extraction, and extract cleanup methods were briefly evaluated for
their applicability for analysis of organic priority pollutants in sediments.
Differences in the compositions of sediments and sludges necessitated some
minor modifications to the sludge methods in order to apply them to sediments.
Sediments typically contain higher solids contents than sludges, and the sus-
pended solids have a much broader range of densities, typically from near
1.0 g/ml for biogenic materials to 2.6 g/ml for siliceous minerals. In addi-
tion, sediments generally have a lower total organic carbon level compared to
sludges. The effects of these differences generally influenced only the purg-
ing and extraction procedures.
SAMPLE COLLECTION AND CHARACTERIZATION
Sediment samples were collected from three points along the Blue River
in Kansas City, Missouri. Samples were collected with a Ponar dredge from
bridges crossing the river at Grandview Road and at 15th Street. Samples were
collected with a shovel from the east bank of the river 15 to 20 ra north of
the bridge at Blue Parkway (Highway M-350). The Grandview Road site is 300
to 400 m downstream from a large electrical component manufacturing plant.
The Blue Parkway and 15th Street sampling sites are just downstream of asphalt-
ing and creosoting plants, respectively. All samples were dark-colored and
had strong odors. The samples were stored at 4°C immediately upon arrival at
MRI. The solids contents of each sample were determined gravimetrically by
drying 1- to 3-g aliquots at 110°C overnight. Total organic carbon was deter-
mined in aliquots of each sample by oxidation to carbon dioxide and infrared
detection. The solids and total organic carbon contents for each sediment
sample are listed in Table 56.
EVALUATION OF SEDIMENT SPIKING
Since the high solids content of sediments may hinder mixing of an or-
ganic spiking solution with sediment, the mixing times required to achieve
homogeneous spike distribution in aliquots of Blue River sediment were inves-
tigated. This study was simplified by using radiolabeled compounds represen-
tative of the three analytical classes. The compounds used were 14C-phenol
and 14C-2,5-dichlorophenol to represent acidic extractables, and 14C-fluorene
to represent base/neutral extractables.
170
-------
TABLE 56. BLUE RIVER SEDIMENT SAMPLES COLLECTED
FOR EVALUATING THE PRELIMINARY POTW PROTOCOL
Sampling location
Solids content
(%)
Total organic
carbon
(rag/g dry weight)
Grandview Road Bridge
65
4.4
Blue Parkway
48
11.0
15th Street Bridge
39
13.0
-------
Experimental
Aliquots of sediments from the 15th Street Bridge site (approximately
500 g) were placed in 500-ml (nominally 16-oz) glass screw-cap jars. The
contents of the jars were spiked with 1.0 ml of an acetone solution contain-
ing etier 14C-phenol at 10.0 |jCi/ml, 14C-2,5-dichlorophenol at 9.5 (JCi/ral, or
14C-fluorene at 5.5 pCi/ml. The jars were sealed with TFE-lined caps and were
tumbled. Aliquots were removed at several time periods from 0 to 104 h. At
each time period three 1-g aliquots were removed from different locations in
the jars. These aliquots were extracted with 6.0 ml of toluene with a vortex
mixer and centrifuged for 10 min at 3,400 rpm. A 1.0-ml portion of each ex-
tract was assayed by liquid scintillation counting. Determinations were made
with a Packard Model 2420 Tri-Carb Liquid Scintillation Spectrometer using a
scintillation cocktail containing 0.47% PP0 (2,5-diphenyl oxazole) as primary
scintillator and 0.006% POPOP (£-bis[2-(5-phenyloxazole)]-benzene) as second-
ary scintillator in toluene. All extracts were counted for 2 or 4 rain and
the counting efficiency was determined by an automatic external standard ratio
method. The dry weights of the sediment aliquots assayed were determined
after drying overnight at 110°C.
Similar experiments were conducted using 14C-benzene. The principal dif-
ferences were that larger sediment aliquots (approximately 630 g) were spiked
and the jars were topped off with a small volume of water each time they were
sealed to remaintain zero headspace during mixing. The samples were spiked
with 1.0 ml of a 10 |jCi/ml solution of 14C-benzene in methanol.
Results and Discussion
The results of long-term equilibration studies for fluorene, phenol, and
2,5-dichlorophenol was summarized in Figures 69 to 71. The ranges of 14C con-
tents of aliquots sampled at each time indicate that fairly uniform distribu-
tion for each compound was achieved within 8 h after spiking. The equilibrium
levels observed for phenol, 2,5-dichlorophenol, and fluorene at 8 h were 80,
85, and 100%, respectively, of the levels spiked. Fluorene levels in the
spiked sediment remained fairly constant and uniform throughout the 104-h
experiment. Although the mean phenol levels remained fairly constant and
uniform through 48 h, the range of phenol in aliquots taken at 104 h was
61% of the mean. Levels of 2,5-dichlorophenol remained uniform throughout
the experiment but decreased to about half of the initial equilibration level.
The loss may be attributable to degradation of the analyte. Hence, it may be
advisable to extract spiked sediment aliquots during the period of 10 to 48 h
following fortification.
The results of the spike equilibration study for benzene are summarized
in Figure 72. Uniform distribution of the spike was achieved after 6 h and
the equilibrium level remained fairly constant following equilibration. The
equilibrium level achieved was 35% of the theoretical spike level. Losses
are likely attributable to volatilization of the spiked benzene during spiking
or sample handling immediately following spiking.
172
-------
250 r
200 -
150 -
100 - -l
50 -
ol I ' I I I I I I I I 1 1 I I I
0 8 16 24 32 40 4 8 56 64 72 80 88 96 104 112 120
Equilibration Time (hours)
Figure 69. Results of equilibration of *''C-f luorene spiked in sediment.
-------
0
-L
_L
16 24 32 40 48 56 64 72
Equilibration Time (hours)
-J I I I I
80 88 96 104 112
Finurii 70. UesuLts of equilibration of 1''('-phenol spiked in sediment.
-------
X
_L
_L
_L
_L
J
0 8 16 24 32
40 48 56 64 72 80
Equilibration Time (hours)
88 96 104 112 120
Figure 71. Results of equilibration of 1'tC-2,5-dichloropijenol spiked in sediment.
-------
150
HO
130
120
110
100
90
80
70
60
50
40
30
20
10
0
T
I
r
i
J L
X
X
X
X
8 10 12 14 16 18 20 22 24
_J
126
Figure 72. Resul ts of equil ibration of ^''C-benzcne spiked in sediment.
-------
ANALYSIS OF PURGEABLE COMPOUNDS
Since the lower total organic carbon content of sediments relative to
sludges may influence purging efficiencies, a preliminary investigation of
purging from various concentrations of sediment in suspension was conducted.
Recoveries were then determined for compounds spiked into sediment and ana-
lyzed using the optimum suspension concentration.
Evaluation of the Effect of Solids Dilution on Purging
Experimental—
Suspensions containing 10, 30, 50, and 80% wet sediment (by weight) were
prepared with sediment from the 15th Street site. Ten-milliliter aliquots of
each suspension were spiked with 250 ng each of the purgeable spiking compounds
and analyzed by the purge-trap GC/MS method described in the preliminary POTW
sludge protocol. No foaming of the suspensions was observed.
Results and Discussion--
The quantities of the purgeable compounds determined in the sediment
suspensions are shown in Table 57. The quantities are also plotted versus
the sediment concentration in the suspensions in Figure 73. Levels for tol-
uene were not plotted, nor were the anomalously low levels observed in the
50% suspensions. From these results, it is apparent that chloroform, 1,1-di-
chloroethene, 1,2-dichloroethane, dichloromethane, and toluene were present
in the unspiked sediment at significant concentrations. The purging efficien-
cies for the various compounds appeared to be affected differently. The levels
of chlorobenzene, ethylbenzene, trichloroethene, and 1,1,1-trichloroethane
were decreased slightly at higher sediment concentrations. The purging effi-
ciency for 1,1,2,2-tetrachloroethane was much lower in the 80% suspensions.
Purging efficiencies appeared to be good for all compounds in the 30% suspen-
sion. Further dilution did not provide markedly higher recoveries but did
decrease the method detection limit. Hence, 30% sediment suspension was se-
lected for subsequent purging studies.
Determination of Recoveries from Spiked Sediment
Experimental--
Five aliquots for each Blue River sediment sample were placed in 40-mi
vials. Two stainless steel balls were added to each vial, and the vials were
sealed with no headspace. Three vials for each sample were spiked with 5.0
Hi of a solution containing the purgeable spiking compounds to achieve spike
concentrations of 55 to 60 ng/g wet sediment. The vials were tumbled overnight
at 4°C to allow equilibration of the spike. A single 3-g aliquot from each
vial was diluted to 10 ml and analyzed by the purge-trap GC/MS procedures de-
scribed in the preliminary sludge protocol. Duplicate unspiked aliquots and
triplicate spiked aliquots were analyzed for each sediment sample.
Results and Discussion--
The results of the recovery experiments are shown in Table 58. The recov-
eries observed were much higher for the Blue Parkway and 15th Street samples.
The compounds that were found at significant concentrations in unspiked 15th
Street samples in the preceding experiment were likely lost during storage.
177
-------
TABLE 57. RESULTS OF ANALYSES OF VARIOUS SEDIMENT-WATER SUSPENSIONS
(10-80% SEDIMENT) BY PURGE-TRAP GC/MS FOLLOWING FORTIFICATION WITH
250 ng EACH OF SELECTED COMPOUNDS
Weight of compounds observed (ng)
Compound 10% 30% 50% 80%
Benzene
201
262
141
221
1,1,2,2-Tetrachloroethane
196
144
34
1
Chloroform
215
319
253
452
1,1-Dichloroethene
264
281
403
607
1,2-Dichloroethane
233
343
371
614
Dichloromethane
273
406
429
472
Trichloroethene
243
178
107
149
1,1,1-Trichloroethane
211
192
99
118
Chlorobenzene
213
185
111
164
Toluene
1,390
1,670
1,250
1,900
Ethylbenzene
197
176
95
156
178
-------
700 r
\
| 400
5
6
200
too
C ^•mmm__L««J—
100 90 M 70 oO 50 40 30 20 10
S*4ia*ot CaM*ft« «4 S»ik«« S>wp«r*i—< (S by ¦»a'w— oi awl MdiaaM )
7®r
600 -
SOoU
I
400
t
if
•9
»
i
200
100
100
fetfintnf C«m*m ai SiiM Si—w«in ("J» by ve*we W wf i
Figure 73. Results of analyses of sediment-water suspensions by
purge-trap GC/MS following fortification with 250 ng
each of selected compounds.
179
-------
TABLE 58. RESULTS OF EVALUATION OF POTW SLUDGE PURGEABLES METHOD FOR SPIKED SEDIMENTS
Grandview Road Blue Parkway 15th Street
Compound
Spike
level
(ng/g)
Unspiked
level
(ng/g)
Spike
recovery
(%)
Unspiked
level
(ng/g)
Spike
recovery
(%)
Unspiked
level
(ng/g)
Spike
recovery
(%)
Benzene3
55
NDb
61
± 26C
ND
86
+
14
ND
187
± 42
Carbon tetrachloride
55
ND
0
ND
0
ND
131
± 136
Chlorobenzene
55
ND
26
± 26
ND
60
+
8
ND
155
± 42
Chloroform
55
ND
5
± 8
ND
152
+
42
ND
57
± 69
1,2-Dichloroethane
55
ND
42
± 14
ND
111
+
39
ND
162
± 94
Vinylidene chloride
55
ND
36
± 11
ND
109
+
8
ND
165
± 75
Ethylbenzene
55
ND
21
± 24
ND
123
+
20
ND
108
± 4
Tetrachloroethene
55
ND
22
± 30
ND
148
+
28
ND
122
± 13
1,1,1-Trichloroethane
55
ND
6
± 10
ND
81
+
2
ND
101
± 90
Trichloroethylene
55
ND
23
± 18
ND
101
+
14
ND
138
± 43
Vinyl chloride
55
ND
46
± 1
ND
132
+
29
ND
119
± 37
a Because of saturation of primary ions, secondary ions were used for quantitation,
b ND = not detected.
c Mean recovery ± standard deviation of Lhree replicates.
-------
Since the samples were collected for spike recovery experiments and not for
accurate characterization of ambient concentrations of purgeables, they were
stored without concern for maintaining zero headspace. The recoveries deter-
mined for triplicate spiked samples were quite variable for many compounds.
Part of the variability likely reflects the lack of homogeneity achieved in
the spiked aliquots.
The results of this evaluation indicated that the sludge purge-trap GC/MS
methods may be successfully applied to sediment analysis. As in the case of
sludges, the precision and accuracy of the method may be improved through the
use of mechanical mixing of the sediment suspension during purging, i.e., the
MRI-designed stirred purge system. However, the abrasive character of sandy
sediments may decrease the usable lifetime of the purging equipment.
ANALYSIS OF EXTRACTABLE COMPOUNDS
Determination of Recoveries from Spiked Sediment
Preliminary attempts to extract sediment aliquots with the homogeniza-
tion/centrifugation method used for sludges indicated that some modification
of the procedure was required. The sediment solids were not cleanly separated
from the extract by centrifugation because of the wide range of particle den-
sities. However, sediments that had been fractionated by centrifuging and
decanting could be homogenized with dichloromethane and centrifuged to cleanly
recover the extract. Therefore, the extraction procedure was modified to in-
clude this fractionation. The sediment supernatants were extracted separately.
Experimental—
Five hundred gram portions (in 16-oz jars) of Blue River sediment sam-
ples were spiked with 300 (Jg of each of the extractable spiking compounds.
Hence, the nominal spike concentrations were 600 ng/g wet sediment. After
tumbling overnight at 4°C to allow for equilibration of the spikes, the con-
tents of each jar were equally divided into three portions. The triplicate
portions were each subdivided into four 200-ml centrifuge tubes. The ali-
quots were basified to pH 11 and centrifuged. The supernatants were trans-
ferred to 125-ml separatory funnels. Dichloromethane (100 ml) was added to
each residue. The content of each tube was homogenized and centrifuged and
the extract removed. The aliquots were similarly extracted twice more and
the extracts combined. The aqueous supernatants were batch extracted three
times with an equal volume of dichloromethane and the extracts combined with
the solids extracts. The aqueous supernatant was then distributed equally
among the four solids aliquots, and the water-sediment mixtures were adjusted
to pH £ 1. The aliquots were again separated and extracted for acidic com-
pounds in a manner similar to extraction for base/neutral compounds. Acids
and base/neutral extracts were cleaned by GPC and analyzed by GC/MS by pre-
liminary POTW sludge protocol procedures. Concentrations of compounds iden-
tified were calculated on a dry solids weight basis.
Results and Discussion—
The results of the recovery determinations for compounds spiked into
sediments are shown in Table 59. The recoveries observed were generally good
for most compounds. The recoveries for the phenolic compounds were particu-
larly good and reproducible, except for 2,4-dimethylphenol. Many of the 3/N
181
-------
TABLE 59. RESULTS OF EVALUATION OF MODIFIED POTW SLUDGE
EXTRACTABLES METHOD FOR SPIKED SEDIMENTS
Grandview Road Blue Parkway 15th Street
Spike
Unspiked
Spike
Unspiked
Spike
Unspiked
Spike
level
level
recovery
level
recovery
level
recovery
Compound
(ng/g)
(ng/g)
«)
(ng/g)
(%)
(ng/g)
(%)
1,4-Dichlorobenzene
629
3h
47
+
9a
2
37 ±
2
17
50 ± 7
llexachloroethane
632
ND
0
ND
0
ND
0
Bis(2-chloroisopropyl)ethe
r 656
ND
c
ND
c
ND
51 ± 2
Bis(2-chloroethyl)ether
599
ND
51
+
16
ND
33 +
5
ND
46 ± 4
Acenaphthylene
634
4
58
+
14
5
57 ±
4
ND
86 ± 11
2,6-Dinitrotoluene
636
ND
0
ND
2 ±
4
ND
0
l'luoranthene
547
390
27
+
21
360
58 ±
28
1,400
85 ± 28
Benzidine
618
ND
29
+
14
ND
20 ±
1
ND
22 ± 3
3,3'-Dichlorobenzidine
726
ND
85
+
27
ND
64 ±
4
ND
70 ± 6
n-Butylbenzylphthalate
714
92
310
+
100
100
120 ±
110
550
82 ± 3
Bis(2-ethylhexyl)phthalate
670
830
31
+
29
490
26 ±
4
1,200
d
Benzo[a]pyrene
622
150
45
+
16
180
68 ±
13
380
76 ± 13
Phenol
674
50
67
+
6
10
65 ±
2
130
79 ± 10
2,4-Dimethylphenol
694
ND
28
i
5
ND
38 ±
4
160
52 ± 6
2,4-Dichlorophenol
655
ND
65
+
10
3
73 ±
4
ND
86 ± 7
Pentachlorophenol
673
ND
86
+
22
ND
73 ±
13
38
84 ± 4
a Mean ± standard deviation for three determinations,
b ND = not detected.
c This compound was not present in the spiking solution.
d Recoveries were not determined for this compound because of
high 1 evels in the unspiked sample.
-------
recoveries were somewhat variable. Although benzidine recoveries were low and
variable, 3,3'-dichlorobenzidine recoveries were good and fairly reproducible.
In general, the POTW sludge extractables method as modified appeared to
provide fairly good results for sediments. However, since separate extraction
of supernatants was required, the sediment extraction procedure was even more
labor-intensive than sludge extraction. Hence, alternate sediment extraction
methods were briefly investigated.
Evaluation of Alternate Extraction Methods
Two alternate procedures were evaluated for sediment extraction by com-
paring the concentrations determined for neutral extractable compounds in rep-
licate aliquots of unspiked Blue River sediment. The sediment was collected
from the 15th Street bridge but was not a replicate of the sample used in the
preceding experiments. Extractive steam distillation, described in Section 5,
and sequential Soxhlet extraction with a hydrophilic solvent and then a hydro-
phobic solvent, using methanol then dichloromethane, were compared with the
homogenization/centrifugation procedure described in the preceding experiment.
Experimental—
A 175-g aliquot of wet sediment was diluted with 300 ml of glass-distilled
water. The sediment-water suspension was extracted for 6 h with 300 ml of
dichloromethane using the extractive steam distillation system described in
Section 5. The extract was concentrated to 5.0 ml for analysis. A second
wet sediment aliquot (84.6 g) was centrifuged and the excess water was decanted.
The residue was transferred to a preextracted cellulose thimble measuring 30
mm x 77 mm and extracted with 200 ml of methanol for 8 h. The extract was
removed and the sediment extracted an additional 8 h with 200 ml of dichloro-
methane. The methanol extract, containing 10 to 15 ml of water, was concen-
trated to approximately 25 ml in a Kuderna-Danish evaporator and extracted
with three 25-ml portions of dichloromethane. This extract and the dichloro-
methane sediment extract were independently concentrated to 5.0 ml for analy-
sis. A third wet sediment aliquot of 160 g was extracted at ambient pH by
the homogenization/centrifugation procedure described in the preceding experi-
ment. The homogenization extract was cleaned by GPC and the cleaned extract
was concentrated to 5.0 ml for analysis. All extracts were screened by GC/FID
and then analyzed for all neutral organic priority pollutants by GC/MS using
the preliminary POTW sludge protocol procedures.
Results and Discussion—
The concentrations determined for neutral organic priority pollutants in
the sediment aliquots extracted by steam distillation, Soxhlet, and homogeni-
zation/centrifugation procedures are shown in Table 60. GC/FID chromatograms
for each extract are shown in Figures 74 to 77. The GC/FID chromatograms illus-
trate that the rigorous homogenization/centrifugation procedure, even with
GPC cleanup, produces extracts with higher levels of unresolved coextractants
than steam distillation or Soxhlet extraction. However, the highest concentra-
tions of the analytes were observed in the Soxhlet extracts. Further devel-
opment of sediment extraction procedures should be focused on techniques which
are more selective and efficient than homogenization/centrifugation.
183
-------
TABLE 60. ORGANIC PRIORITY POLLUTANTS IDENTIFIED IN BLUE RIVER SEDIMENT
BY SOXHLET EXTRACTION. STEAM DISTILLATION, AND HOMOGENIZATION
Concentration (ng/g drv sediment)
Soxhlet
Steam
Compound identified
CH,OH CH2Cl2 Total distillation Homogenization
Acenaphthene 48 42 90 87
Fluorene 60 40 100 84
Phenanthrene/ 380 430 810 450
Anthracene4
Dimethylphthalate 13 2 15 ND°
Diethylphthalate 33 11 50 ND
Fluoranthene 430 1,200 1,600 470
Pyrene 330 900 1,200 370
Di-n-butylphthalate 47 23 80 48
3utylbenzylphthalate ND 53 53 ND
Benzanthracene/ 170 780 950 76
Chrvsene3
Bis(2-ethylhexyl)- 180 90 270 260
phthalate
Benzofluoranthenes3 70 500 570 7
3enzo[a]pyrene 120 900 1,000 NT)
Indeno[1,2.3-cd]pyrene 5 50 56 ND
Dibenzf a, ji] anthracene ND 23 23 IID
3enzo[£,h,iJperylene 10 92 100 ND
21
24
120
ND
35
170
140
43
ai
130
44
38
ND
ND
a Concentrations represent sums o£ these compounds which alute simultane-
ously and have the same major ions for GC/MS.
b ND » not detected.
184
-------
0
Figure 74. GC/FID chromatogram of Blue River sediment extract by
steam distillation-continuous extraction - no cleanup.
185
-------
wvaAX^aA.
JL
15 20
Time (min )
25
30
Figure 75. GC/FID chromatogram of Blue River sediment extract by
Soxhlet with methanol - no cleanup.
186
-------
25
Time (min)
Figure 76. GC/FID chromatogram of Blue River sediment extract by
Soxhlet with dichloromethane (preextracted with
methanol) - rio cleanup.
137
-------
Time (min)
Figure 77. GC/FID chromatogram of Blue River sediment extract by
homogenization following cleanup by GPC.
188
-------
SECTION 7
DETERMINATION OF THE PHASE DISTRIBUTION OF SELECTED COMPOUNDS
IN 1% PRIMARY SLUDGE
The distributions of contaminants between solution and solids in waste-
waters and sludges are key factors in influencing the transport of those con-
taminants within the wastewater treatment system. In particular, compounds
with low aqueous solubilities may strongly associate with solids. Hence,
Solids wasting from a treatment system may be the primary emission mode;
i.e., solids removal from the wastewater may also provide efficient removal
of hydrophobic contaminants. The objective of the experiments described in
this section was to determine the distributions of selected purgeable and ex-
tractable compounds spiked in two samples of 1% primary POTW sludge at two or
more levels. The development of analytical methods for sludges, described in
the preceding sections, facilitated the experimental determination of phase
distributions.
The phase distributions determined for each compound were expressed as
follows:
c 1 - A
^ _ solids
distribution C _
supernatant
whe
re: K,. . ., is the phase distribution coefficient
distribution r
C ,. , is the concentration of the analyte in the suspended
solids i¦j r / \
solids (ng/g)
C ^ ^ is the concentration of the analvte in the sludge
supernatant , ,, .. .
supernatant (|Jg/liter)
The target spiking levels for the phase distribution experiments were
100 and 1,000 times the detection limit of the analytical method. However,
the aqueous solubility for each compound was a primary consideration in de-
termining the actual spiking levels. The spiking compounds may be found in
sludges at concentrations in excess of their aqueous solubilities because a
fraction is associated with the suspended solids. Hence, sludges can be
Spiked with many compounds at levels exceeding aqueous solubilities for the
purpose of determining analytical recoveries with some measure of confidence.
However, more accurate phase distribution information will result from experi-
ments where the suspended solids are exposed to the spiked compounds in solu-
tion. That is, spiking levels should be equal to or less than the aqueous
solubility for each compound.
189
-------
The solubilities of the purgeable spiking compounds in water are listed
in Table 61. The solubilities for all compounds except vinyl chloride are
greater than 150 mg/liter. Assuming a conservative analytical detection limit
of 5 pg/liter, all compounds except vinyl chloride could be spiked at 1,000
times the detection limit without exceeding their aqueous solubilities.
The aqueous solubilities for the extractable spiking compounds are
listed in Table 62. The extractable compounds were divided into four groups
based on solubility and analytical class to allow convenient spiking and analy-
sis. The first three groups are base/neutral compounds selected according to
solubilities and the fourth group contains the phenolic compounds. Only four
compounds from the more soluble Group I B/Ns and three phenols from Group IV
have solubilities exceeding 1,000 times a conservative detection limit of
;250 |jg/liter. The spike levels for the phase distribution experiments were
; selected to be lower than the solubility of the least soluble compound in each
Sgroup. The very low aqueous solubility of benzo[a]pyrene, the only Group III
compound, was too low to allow either spiking at soluble levels for analysis
of fractions by GC/MS methods. Hence, phase distributions were not determined
for benzo[a]pyrene.
The sludges used in the phase distribution experiments were the Platte
County, Kansas City, Kansas, and Blue River primary POTW sludges described in
Table 18. Experiments for purgeables were conducted with the Platte County
and Kansas City, Kansas, sludges. Experiments for extractables used the
Platte County and Blue River sludges. Sludge aliquots were diluted to 1%
suspended solids with purgeables-free water prior to spiking with the purge-
able compounds in methanol. Since the solubilities of many of the extract-
able compounds are much lower than the purgeable compounds, sludge aliquots
used in the phase distribution studies for extractable compounds were diluted
to 1% suspended solids with a solution of the spiking compounds in glass-
distilled water. Spiking was done in this manner to ensure that the less
soluble compounds would be dissolved before being exposed to the solids.
PURGEABLE COMPOUNDS
Although recoveries had been determined for selected purgeable compounds
spiked into primary sludge from the Platte County and Kansas City, Kansas,
POTWs (see Section 4), recoveries were also measured for the same compounds
spiked into sludge aliquots diluted to 1% suspended solids to provide an in-
dication of the precision and accuracy of the subsequent distribution experi-
®ents.
Determination of Recoveries from Spiked 1% Sludge
Experimental—
Aliquots of primary sludge were diluted to 1% suspended solids with
purgeables-free water, spiked, and analyzed as described in the Method
Appendix. Recoveries were determined in triplicate.
190
-------
TABLE 61. WATER SOLUBILITIES OF PURGEABLE SPIKING COMPOUNDS
Water solubility3
at 20°C
Compound (fjg/L)
Benzene 1,800,000 (25°C)
Carbon tetrachloride 785,000
Chloroform 8,200,000
1.1-Dichloroethene 400,000
Tetrachloroethene 150,000
Vinyl chloride 1,100 (25°C)
1.2-Dichloroethane 3,690,000
Trichloroethene 1,100,000
1,1,1-Trichloroethane 4,400,000
Chlorobenzene 500,000
Ethylbenzene 152,000
a U.S. Environmental Protection Agency. Water-Related
Environmental Fate of 129 Priority Pollutants,
Vol. II. EPA-440/4-79-029b, December 1979.
191
-------
TABLE 62. WATER SOLUBILITIES OF THE EXTRACTABLE SPIKING COMPOUNDS
Water solubility
at 20-25°C Literature
Compound (yg/L) reference
Group I
1,4-Dichlorobenzene 79,000 a
Hexachloroethane 50,000 b
Bis(2-chloroisopropyl)ether 1,700,000 b
Bis(2-chloroethyl)ether 10,200,000 c
2,6-Dinitrotoluene 300,000 d
Benzidine 400,000
3,3'-Dichlorobenzidine NA
e
Group II
Acenaphthylene 3,930 c
Fluoranthene 300 f
n-Butylbenzylphthalate 2,900 c
Bis(2-ethylhexyl)phthalate 400 S
Group III
Benzo[a]pyrene 4 f
c
Group IV
Phenol 93,000,000
2,4-Dimethylphenol 17,000,000 c
2,4-Dichlorophenol 4,500,000 a
Pentachlorophenol 14,000 c
a Chion, C. T. et al. Environ. Sci. Technol. _11:475 (1977).
b Kirk-Othmer Encyclopedia of Chemical Technology, 2nd edition,
John Wiley and Sons, 1979.
c U.S. Environmental Protection Agency, Water-Related Environmental
Fate of 129 Priority Pollutants, Vol. II. EPA-440/4-79-029B,
December 1979.
d Lange's Handbook of Chemistry, 11th edition, McGraw-Hill, 1973.
e Merck Index, 9th edition, M. Windholz (Ed.), Merck & Co., 1976.
f Mackay, D. and W. Y. Shiu, J. Chem. Eng. Data 22:399 (1977).
g Patty, F. A. Industrial Hygiene and Toxicology, Vol. 2, Science
Publishers, 1978, p. 1900.
NA » not available.
192
-------
Results and Discussion—
The results of the recovery determinations are shown in Table 63. Re-
coveries for all compounds except carbon tetrachloride were within the range
of 75 to 140% with good reproducibility. The concentrations of the spiking
compounds in the unspiked, diluted sludge were all at or near the detection
limit.
Determination of Phase Distributions
Experimental—
Diluted sludge (1% suspended solids) was poured into 200-ml screw-cap
centrifuge tubes containing a magnetic stirring bar. The samples were spiked
with the purgeable spiking compounds in 10 to 100 |Jl of methanol and sealed
with no headspace with TFE-lined lids. The spiked samples were stirred for
1 min and were allowed to equilibrate at 4°C overnight with tumbling. The
phases were separated by centrifuging the bottles at 2,500 rpm for 30 min.
The centrifuged aliquots were allowed to cool at 4°C before gently decanting
the supernatant. Aliquots of the supernatant were stored in 40-ml vials with
no headspace prior to analysis. Immediately following removal of the super-
natants, the centrifuge bottles were filled with purgeables-free water and
resealed with zero headspace. The reconstituted sludge solids were hand-
shaken 1 min and aliquots were removed for analysis. Triplicate aliquots of
supernatants and reconstituted solids were analyzed for each spiked sample by
the purge-trap GC/MS procedure described in the Method Appendix. The concen-
trations determined for the supernatants and the corresponding reconstituted
solids (dry basis) were used to calculate the phase distribution coefficients
for each sludge sample.
Results and Discussion—
The results of the phase distribution determinations are shown in
Table 64. The coefficients observed range from less than 10 for vinyl chlo-
ride and 1,2-dichloroethane to more than 200 for tetrachloroethene and ethyl-
benzene. The coefficients were higher for all compounds in the Kansas City,
Kansas, sludge. No clear relationships with concentration were apparent in
either sludge. The coefficient determinations for each spike level were very
reproducible.
EXTRACTABLE COMPOUNDS
In order to provide an indication of the precision and accuracy of the
subsequent phase distribution experiments, recoveries were determined for the
selected extractable compounds spiked into sludge supernatants and solids from
the diluted Platte County and Blue River P0TW sludges. In addition, the time
required for spikes to achieve equilibrium distribution between supernatant
aad solids was investigated.
193
-------
TABLE 63. RECOVERIES FOR PURGEABLE COMPOUNDS SPIKED
INTO PRIMARY POTW SLUDGE DILUTED TO 1% SOLIDS
Spike Platte County Kansas City, Kansas
level POTW recovery POTW recovery
Compound (mb/D (%) (%)
Benzene
56b
93
+
6a
100
± 3
560
100
+
15
120
± 7
Carbon tetrachloride
539b
62
+
40
62
± 30
5,390°
31
+
2
30
± 11
Chloroform
l01h
100
+
7
120
± 1
1,010
130
+
20
99
± 6
1,1-Dxchloroethene
271b
77
+
5
96
± 2
2,710
97
+
15
97
± 2
Tetrachloroethene
163
78
+
4
110
± 10
1,630
110
+
6
84
± 24
Vinyl chloride
170
78
+
1
90
(4)C
1,700
86
+
6
100
± 3
1,2-Dichloroethane
274b
98
+
3
92
± 4
2,740°
110
+
4
100
± 3
Trichloroethene
108,
76
+
12
92
± 5
1,080°
110
+
L
86
± 8
1,1,l-Trichloroethane
866,
120
+
4
140
± 5
8,600
130
+
3
140
± 19
Chlorobenzene
108
82
+
7
91
± 1
1,080
110
+
2
94
± 15
Ethylbenzene
268
79
+
1
100
± 3
2,680
110
+
2
93
± 3
a Mean ± standard deviation for three determinations.
b Analyzed from a 2.0-ml aliquot diluted to 10 ml prior to purging,
c Mean (average deviation) for two determinations.
194
-------
TABLE 64. DISTRIBUTION COEFFICIENTS DETERMINED FOR PURGEABLE COMPOUNDS
SPIKED INTO PRIMARY POTW SLUDGE
^distribution3
Spike level Platte County Kansas City, Kansas
Compound ((Jg/L) POTW POTW
Benzene
56
11
± lb
28
+
3
560
23
± 2
24
+
5
Carbon tetrachloride
539
34
(10)°
76.
.2
(6)
5,390
20
± 4
43
+
19
Chloroform
101
14
(0.2)
15
+
1
1,010
13
(0.8)
17
+
3
1, 1-Dichloroethene
271
6.9
± 2
23
+
3
2,710
48
± 31
20
+
3
Tetrachloroethene
163
160
± 27
380
a.
39
1,630
260
± 21
290
+
100
Vinyl chloride
170
d
6.6
+
0.4
1,700
4.4
± 0.8
3.4
+
0.3
1,2-Dichloroethane
274
1.7
+ 0.7
7.1
+
0.5
2,740
4.5
± 0.6
2.8
+
0.3
Trichloroethene
108
40
± 9
60
+
6
1,080
44
+ 3
69
+
20
1,1,1-Trichloroethane
866
41
± 4
80
+
4
8,600
45
± 2
61
+
10
Chlorobenzene
108
53
± 4
110
+
14
1,080
100
± 13
110
+
32
Ethylbenzene
268
110
± 2
230
+
10
2,680
190
± 23
240
+
84
3 ^distribution = concentration of analyte in solids (a^/g) .
concentration of analyte in supernatant (ng/mlj
b Mean ± standard deviation for three determinations.
c Mean (average deviation) for two determinations.
d Determination was not possible because of coeluting interferences.
195
-------
Determination of Recoveries from Spiked Fractions of 1% Sludge
Experimental—
Three 80-ml aliquots of each sludge diluted to 1% solids were fraction-
ated by centrifuging 30 min at 2,500 rpm and then decanting the supernatants.
The supernatants from two aliquots of each sludge were spiked with the Group I
and IV compounds at 1,000 )Jg/liter and the Group II compounds at 300 (jg/liter.
The sludge residues were reconstituted by diluting to 80 ml with glass-
distilled water and were spiked with the Group I, II, and IV at the same levels
as the supernatants. The diluted residues were homogenized for 1 min. All
supernatants and reconstituted solids were extracted and analyzed by the pro-
cedures described in the extractables protocol (Method Appendix) except that
no extract cleanup was performed. Analyses were conducted using packed column
GC/MS.
Results and Discussion--
The results of the recovery determinations are shown in Table 65. In
general, the recoveries observed were good. However, hexachloroethane and
2,6-dinitrotoluene were not recovered from the Blue River sludge supernatants.
Benzidine was not recovered from any of the spiked reconstituted solids sam-
ples. Hence, the phase distribution results for these compounds are not mean-
ingful .
Determination of Time Required for Equilibrium Distribution
of Spikes Between Phases
Experimental--
Sludge aliquots containing 0.80-g solids were diluted to 80 ml in 200-ml
centrifuge tubes with glass-distilled water spiked with the extractable spik-
ing compounds in order to achieve concentrations of 5,000 [Jg/liter for the
Group I and IV compounds and 300 [jg/liter for the Group II compounds. The
aliquots were stored at 4°C in the dark. At specific time intervals (0, 2,
4, 6, 8, 24, 48, and 72 h) aliquots were removed and fractionated. The
fractions were analyzed by the extractables protocol (Method Appendix) with
the exception that extract cleanup was omitted.
Results and Discussion--
The concentrations of Group I, II, 3nd IV compounds measured in the
supernatants and solids fractions of spiked Platte County primary sludge at
0 to 72 h following spiking are shown in Tables 66 to 68. Equilibration of
the spikes between the solids and supernatant fractions appeared to have been
achieved within the first 2 h following spiking. The recoveries of some com-
pounds were decreased at longer equilibration times. Hence, the phase distri-
bution experiments were conducted allowing only 2 h for equilibration. Sig-
nificant losses of the spiked compounds were not observed for most compounds,
with the exception of benzidine; the total recoveries were greater than 50%.
196
-------
TABLE 65. RECOVERIES OF THE EXTRACTABLE COMPOUNDS SPIKED
INTO PRIMARY POTW SLUDGE DILUTED TO 1% SOLIDS
Platte County
Supernatant
Reconsti tuted
1 solids
Spike
Unspiked
Spike
Unspiked
Spike
level
level
recovery
level
recovery
Compound
(Mg/L)
(|.lg/U
a)
(mk/l)
(%)
Group I
I,4-Dichlorobenzene
1,000
ND
61,
80
ND
80, 45
liexachlo roe thane
1,000
ND
210,
210
ND
80, 34
Bis(2-chloroethyl)ether
1,000
ND
46,
49
ND
190, 65
2,6-Dinitrotoluene
1,007
ND
110,
110
ND
220, 100
3,3'-Dichlorobenzidine
300
ND
150,
170
ND
47, 48
Benzidine
300
Nl)
74,
280
ND
0, 0
Group II
Acenaphthylene
300
ND
200,
170
ND
64, 61
Fluoranthene
300
6
200,
190
ND
63, 64
n-Buty1benzylphthalate
300
22
55,
100
65
63, 51
Bis(2-ethylhexyl)phthalate
300
252
18,
63
556
55, 40
Group IV
Phenol
1,180
ND
77,
84
ND
120, 110
2,4-Dimethylphenol
1,030
ND
77,
83
ND
74, 68
2,4-Dichloroplienol
1,160
ND
92,
95
ND
120, 120
Pentachlorophenol
1,160
ND
180,
96
ND
121, 134
(continued)
-------
Blue River
Supernatant
Reconstituted
solids
Spike
Unspiked
Spike
Unspiked
Spike
level
level
recovery
level
recover
Compound
(mk/D
(jJg/L)
(%)
(pg/L)
(%)
Group I
1,4-Dichlorobenzene
1,110
ND
52, 66
ND
70, 57
Hexachloroethane
1,110
ND
0, 0
ND
64, 70
Bis(2-chloroethyl)ether
1,010
ND
58, 88
ND
91, 94
2,6-Dinitrotoluene
1,010
ND
0, 0
ND
66, 64
3,3'-Dichlorobenzidine
1,040
ND
130, 120
ND
120, 93
Benzidine
985
ND
120, 180
ND
0, 0
Group II
Acenaphthylene
307
20
61, 75
84
110,
Fluoranthene
320
80
77, 89
478
100,
n-Butylbenzylphthalate
342
478
78, 67
2,270
180,
Bis(2-ethylhexyl)phthalate
332
382
35, 40
815
130,
Group IV
Phenol
1,080
63
110, 102
ND
130,
2,4-l)imethylphenol
1,030
9
120, 92
14
130,
2,4-Dichlorophenol
1,160
ND
140, 114
ND
160,
Pentachloropheriol
1,000
20
120, 90
73
200,
-------
TABLE 66. RESULTS OF EQUILIBRATION STUDY OF GROUP I COMPOUNDS
SPIKED INTO PLATTE COUNTY PRIMARY SLUDGE AT 5,000 pg/£
Concentration
Compound
Time
(h)
Supernatant
(ug/L)
Solids
(ng/g)
Average
spike
recovery (%)
1,4-Dich.lorobeazene
Hexa chlo roethane
Bis(2-chloroethyl)ether
2,6-Dinitrotoluene
Benzidine
3,3'-Dichlorobenzidine
0
390
400
1,380
1,580
40
2
660
750
2,680
1,680
58
4
860
910
2,790
2,580
71
6
375
1,040
2,910
2,960
73
8
960
940
2,560
2,630
71
24
960
840
2,400
3,140
73
48
900
1,160
2,410
2,080
66
72
800
860
2,510
2,560
67
0
160
250
2,025
2,150
46
2
260
300
3,560
2,460
66
4
600
510
2,560
3,000
67
6
310
625
3,510
3,520
80
8
340
370
1,890
2,360
50
24
375
390
1,240
1,090
31
48
260
300
1,890
1,750
42
72
210
240
525
375
14
0
2,100
3,110
ND
400
56
2
3,040
3,360
210
275
69
4
4,920
5,160
ND
ND
100
6
2,025
7,100
ND
NE
91
8
5,010
4,8S0
200
340
100
24
4,590
4,090
260
225
92
48
4,820
6,720
250
210
120
72
5,450
5,740
200
310
120
0
1,950
3,150
210
310
55
0
2,550
2,860
250
390
60
4
3,050
3.310
80
175
65
6
2,020
3,120
500
150
57
8
2,450
2,710
118
ND
52
24
1,950
1,880
68
58
39
48
1,280
1,440
210
150
30
72
1,650
1,380
ND
ND
30
0
ND
ND
ND
ND
C
2
ND
ND
ND
ND
0
4
78
82
ND
ND
V?
6
46
42
ND
ND
1 c
8
56
59
ND
ND
19
24
130
120
ND
ND
30
48
b
ND
ND
ND
0
72
ND
ND
ND
ND
0
0
54
50
240
200
90
2
44
41
260
300
110
4
44
54
200
200
83
6
46
42
ND
ND
i
8
46
67
410
760
220
24
300
225
700
750
330
48
b
ND
175
175
58
72
36
ND
200
190
71
a Spiking level 300 ppb.
b Extract lost during workup.
199
-------
TABLE 67. RESULTS OF EQUILIBRATION STUDY OF GROUP II COMPOUNDS SPIKED
INTO PLATTE COUNTY PRIMARY SLUDGE AT 300 Mg/L
Compound
Concentration
Average
Time
Supernatant
Solids
spike
(h)
(mk/L)
(ug/s)
recovery (%)
0
430
36
225
212
84
2
24
29
212
250
86
4
26
25
150
137
56
6
24
14
137
137
52
8
16
14
150
162
57
24
30
26
100
118
46
48
a
20
137
116
4S
72
26
27
119
137
52
0
25
26
300
275
100
2
20
21
287
375
110
4
19
21
225
187
73
6
18
12
212
187
71
3
13
15
300
375
110
24
35
34
312
325
120
48
a
18
237
175
73
72
30
23
225
212
30
0
16
14
287
250
73
2
10
11
287
312
79
4
18
19
225
225
62
6
15
21
187
137
46
3
ND
36
761
350
210
24
95
32
300
850
230
48
a
14
275
212
66
72
24
16
150
250
56
0
100
124
775
761
b
2
39
37
900
375
4
42
49
525
437
6
SO
107
625
562
8
61
225
3,060
3,700
24
425
362
2,400
2,825
48
a
64
575
437
72
77
76
512
512
Acenaphthylene
Fluoranthene
a-Butylbearylphthalate
Bis(2-ethylhexyl)phthalate
a Extract lost during workup.
b Recoveries were not determined since the unspiked level was higher
than the spike level.
200
-------
TABLE 68. EQUILIBRATION STUDY OF GROUP IV COMPOUNDS SPIKED
INTO PLATTE COUNTY PRIMARY SLUDGE AT 5,000 pg/L
Concentration
Average
Time
Supernatant
Solids
spike
Compound
Ch)
(pg/L)
fng/g)
recovery (%)
Phenol
0
6,160
5,890
790
725
140
2
5,050
5,700
300
540
120
4
4,760
5,710
550
660
120
6
6,560
6,250
600
410
140
8
7,600
6,725
425
425
150
24
4,850
4,880
425
410
110
48
4,740
4,910
440
410
110
72
a
5,440
a
350
120
2,4-Dimethylphenol
0
5,260
4,710
740
775
120
2
4,190
4,690
590
540
100
4
4,090
4,540
425
400
95
6
5,320
4,850
460
450
no
S
7,520
6,010
510
525
150
24
7,040
3,990
350
390
88
48
3,730
3,810
490
525
36
72
a
5,440
a
350
120
2,4-Dichloropheno1
0
3,350
2,950
2,050
2,180
110
2
2,780
3,040
2,050
2,100
99
4
2,560
2,950
1,890
2,150
96
6
5,025
2,390
2,:oo
1,520
95
8
4,200
3,580
1,950
2,060
120
24
2,300
2,440
1,680
1,720
31
48
2,110
2,340
1,520
1,790
79
a
2,340
a
1,600
89
Pentachlorophenol
0
1,450
1,430
3,550
3,580
b
2
S75
750
4,120
4,250
4
690
900
4,310
4,110
6
700
940
4,250
4,060
8
950
1,240
4,050
3,760
24
325
940
4,180
4,060
48
825
1,150
4,130
3,350
72
a
775
a
4,220
a Extract lost during workup.
b Recoveries were not determined. Analyses of pentachlorophenol in the solids fraction
were hindered by severe interferences. The amount in the solids was determined by
the difference between the spike level and the amount in the supernatant. Penta-
chlorophenol was not found in the unspiked sample and supernatant analyses had no
interferences.
201
-------
Determination of Phase Distributions
Experimental-
Sludge aliquots containing 0.80-g solids were diluted to 80 ml in 200-ml
centrifuge tubes with glass-distilled water spiked with the extractable spik-
ing compounds. The final spike concentrations were nominally 300, 1,000,
5,000, and 10,000 (Jg/liter for the Group I and IV compounds; 200, 300, and
3,000 (Jg/liter for the acenaphthylene and bis(2-ethylhexyl)phthalate; and 200
and 300 (Jg/liter for fluoranthene and butylbenzylphthalate. The spiked ali-
quots were stored at 4°C in the dark for 2 h before fractionating and analyz-
ing as described in the preceding experiment. All experiments were conducted
in duplicate.
Results and Discussion--
The concentrations of analytes determined in the supernatants and solids
of the spiked sludge aliquots are shown in Tables 69 to 71 along with the phase
distribution coefficients determined for each experiment. The phase distribu-
tion coefficients are summarized in Table 72. The phase distribution coeffi-
cients observed range from less than 10 for bis(2-chloroethyl)ether, 2,6-di-
nitrotoluene, benzidine, and phenol to greater than 4,000 for hexachloroethane,
fluoranthene, and butylbenzylphthalate. The large relative standard devia-
tions for the distribution coefficients shown in Table 72 for some compounds
may be indicative of a relationship between spike concentration and the dis-
tributions. However, no well-defined concentration-distribution relationships
were discernible.
202
-------
TABLE 69. PHASE DfSTMBUTIOH COEFFICIENTS DETERMINED BY GROUP I COMPOUNDS
SPIKED INTO PRIMARY POTV SI.UDGE
Compound
1 ,4-l)ichlorobcnzene
llexa cli 1 o roc thane
ro
O
LO
Bis(2-rtiloroettiyl )«>tlier
2,6~l)i»i t Kilo I none
Spike
level
Jl'sZL!
307
1,027
5,000
11,025
320
1,067
5,000
10,675
300
1,000
5,025
10,(100
360
1,200
5,040
12,025
Platte County rOIW sludge
Concentration
Blur River sludge
Supernal ants
tcg/l-J
39
32
325
237
950
1,075
8,260
6,050
Nl)b
ND
47
29
125
150
800
487
225
212
1,440
912
6,325
7,000
42,190
31 ,000
175
187
812
687
2,310
2,600
14,090
12,050
Sol ids
..i!fy _?si
5,900
5,500
32,500
38,800
431,000
270,000
470,000
617 ,000
12,500
11,900
50,000
50,000
625,000
431,000
622,000
940,000
ND
ND
6,800
6,300
16,200
21,200
11,200
28,700
2,200
2,200
8,400
10,600
21,200
23,700
84,000
109,000
. i. »
d i strlbution
151
172
Spike Concentration
level Siipernatants Solids „
(jig/l ) (pg/J.) (ng/g dry solids) distribution'
100
164
454
251
57
105
1,064
1,720
5,000
2,870
780
1,930
2 6
3
0.3
0.9
13
12
10
15
334
5,575
11,150
330
5.500
11,000
305
5,090
10,170
304
5,080
10,150
34
27
900
780
3,810
4,130
ND
ND
Nl)
ND
ND
110
170
133
4,100
3,380
19,400
19,600
ND
ND
ND
ND
ND
ND
14,160
16,420
231,800
41,400
1,816,900
1,343,700
12,740
12,650
163,600
49,800
1,509,300
888,100
Nl)
ND
22,580
20,030
77,660
32,120
11,250
ND
49,400
59,800
130,600
94,800
420
610
260
53
570
325
8,070
5.5
6
3 8
1.6
(conl ilined)
-------
TMB W (rnic)iiic«
Platte County I'OTW sluilgc
Spike Concentration
level Supernatants Solids „
Compound (l'g/I ) i(!B/I'i (ng/g dry solids) Jjstbut^on
Benzidine 295 Nl) NU c
NO ND c
990 1,660 16,000 10
3,050 15,000 5
9,850 30,500 253,750 8
39,270 258,750 7
3,3'-Diclilorobenzidine 310 38 93,800 2,470
48 93,800 1,950
1,035 360 365,000 1,010
275 325,000 1,180
10,350 3,060 2,010,000 660
2,620 2,000,000 760
a K = concentration of analyte in solids tng/g)
distribution ~ concentration of analyte in supernatant (|'g/l.)
to
O b ND = not detected.
¦P-
c *j- i .. .. could not be determined,
distribution
Blue River sludge
Spike Concent ration
level Suprrnatants Solids
(jJg/l.) (yg/l.) (ng/g dry solids) di stribut ioua
295 235 ND c
660 NO c
4,920 1,142 29,250 26
1,700 18,100 II
9,850 6,170 74,250 12
10,430 178,750 17
310 18 20,200 1,120
20 19,900 1,000
5,160 400 250,600 630
330 331,000 1,000
10,340 1,360 1,500,000 1,100
1,490 1,513,000 1,040
-------
TAIil£ 70. I'llASe DISTRIBUTION COEFFICIENTS DETERMINED FOR CROVI' II COMPOUNOS
SPIKED INTO I'RIHAKY POTW SLUDGE
InJ
O
Ui
Platte County POTW sludge
Spike
Concentration
lovel
Supernalants
Sol ids
K
Compound
(l'g/L)
... (i'g/q
(yg/B X-
distribution'
Acenaplilliylene
202
17
23,810
1,400
18
14,890
830
300
13
33,730
2,590
16
39,680
2,480
2,970
235
378,970
1,610
249
373,020
1,500
Fluoraiilhene
212
8
31,250
3,910
7
37,110
5,300
300
10
45,470
4,550
11
58,590
5,330
ii - Bui ylhrnzylplilha late
224
23
48,250
2,100
25
72,370
2,890
304
13
51,050
3,930
14
54,820
3,920
3,270
125
539,470
4,320
125
605,260
4,840
Bi s(2-ctliy 1 h<'xyl )|»lil ha I.ite
220
207
265,620
1,280
195
192,700
990
100
217
187,500
860
212
182,290
Rf>0
Spike
Irvel
ji»g/y
308
3,400
320
318
3,270
332
Blue River simlfce
Stipe rnatants
Goncentration
Sol ids
_l!!8/g dry solids) distribution
38
46
445
470
31
36
241
342
589
69 7
258
284
21,870
22,800
373,460
361,210
58,380
53,300
170,520
176,370
466,700
446,900
87,000
83,500
580
500
840
770
1,880
1,480
710
510
790
500
340
290
U _ concentration ol .in.ilyle in solid (ng/g)
distribution concentration ol aualytr in sii|icriiaiant (|ig/l.)
-------
TAH1.£ 71. PffASfc t)(SrK(tit(rtON COtttlClt.N fS OHMRHINEi) FOR GROUF {V COtlPOUNOS
SPIKKD INTO PRIMARY M)TW SLUDGE
ro
O
on
Compound
Phenol
2,4-Ui««*tlty 1 phenol
2, 4-Dichloropbenol
Pent acli I o ruplicnu 1
Platte
County P0TW sludge
----- ' — -
Spike
Conrentr.il ion
level
Supernatants
Sol ids
If
(pg/U
(yg/u
(ng/g dry sol ids)
distribnt
115
156
3,050
20
162
b
b
1,150
1,510
16,860
11
1,590
17,860
11
5,010
6,240
25,210
4
7,040
45,170
6
11,480
23,980
154,400
6
21,380
157,560
7
100
24
1,940
til
31
b
c
1,000
940
26,410
28
906
24,650
27
5,000
5,240
82,750
16
5,860
75,700
13
10,010
7,920
297,540
38
7,660
274,650
36
112
37
4,340
117
37
b
c
1,130
520
51,910
100
540
48,730
90
5,000
2,950
171,600
58
3,230
178,000
55
11,280
7,980
497,900
62
7,170
521,200
71
114
Nl>
8,960
b
ND
b
c
1 , 130
152
71,850
473
NO
89,570
b
5,000
630
324,800
515
540
334,650
620
11,310
2.4SO
618,100
248
2,5 10
614,200
243
Spi ke
level
tMK/l)
Blue River sludge
Concentration
323
Supernatants
(M*/i.)
5,390
10,780
5,140
347
5,790
11,580
300
5,000
10,000
280
433
6,260
6,320
9,420
10,100
550
808
15,250
13,300
8,190
8,480
99
142
2,620
2,780
5,500
5,610
49
52
1.260
1.250
2,040
2,020
Sol ids
(ng/g dry solids)
3,360
5,270
60,000
47,270
b
103,640
)6,360
23,640
510,900
498,200
b
258,200
14,580
22,200
425,000
435,400
l>
713,200
18,180
35,230
471,000
476,700
b
693,200
distribution
12
12
30
29
33
37
30
147
156
162
157
370
680
370
390
340
< oncctit r^t ion ot Jiiulylr in m»ImIs ("u/g)
J <11 si i i l*ut i oil com cut rat i cm «>l aiidlytc in siipc-i n.it .nit (|>g/i.)
b S.iutfilc lost .
i K could not be dclri'Mi iit-d.
ih ut rihution
-------
TABLE 72. SUMMARY TABLE OF PHASE DISTRIBUTION COEFFICIENTS
Compound
distribution
Platte County
POTW sludge
Blue River
sludge
1,4-Dichlorobenzene
182
+
124
437
± 151
Hexachloroethane
2,230
+
1,540
8,
, 070a
Bis(2-chloroethyl)ether
3.1
+
2.5
4.2
± 2
2,6-Dinitrotoluene
10.4
+
2.8
b
Benzidine
7.5
+
2
17
± 7
3,3'-Dichlorobenzidine
1,340
+
720
980
± 180
Acenaphthylene
1,740
+
680
670
± 159
Fluoranthene
4,770
+
680
1,680
± 282
n-Butylbenzylphthalate
3,670
+
1,000
627
± 145
Bis(2-ethylhexyl)phthalate
1,000
+
200
315
± 35
Phenol
9.3
+
5.4
10.4
± 1.7
2,4-Dimethylphenol
34
+
23
31.8
± 3.3
2,4-Dichlorophenol
79
+
24
150
± 14
Pentachlorophenol
420
+
168
430
± 140
a Data questionable.
b Distribution coefficient not determined.
207
-------
APPENDIX
METHOD 624-S
PROTOCOL FOR THE ANALYSIS OF PURGEABLE ORGANIC PRIORITY POLLUTANTS
IN INDUSTRIAL AND MUNICIPAL WASTEWATER TREATMENT SLUDGE
1. Scope and Application
1.1 This method is used for the determination of purgeable (volatile) or-
ganics. The complete list of compounds is provided in Table 1.
1.2 The method is for qualitative and quantitative analysis of these
compounds in municipal and industrial wastewater treatment sludges.
The procedure requires the use of a gas chromatography/mass spec-
trometer (GC/MS) as the final detector.
1.3 The method detection limit for each compound is very dependent on
the compound characteristics and the particular sludge analyzed.
However, typical detection limits are 2 to 5 pg/liter for publicly
owned treatment works (POTW) sludges with 1 to 5% total solids.
1.4 This method is restricted to use by or under the supervision of
analysts experienced in the use of purge and trap systems and GC/MS
and skilled in the interpretation of mass spectra. Each analyst
must demonstrate the ability to generate acceptable results with
this method using the procedure described in Section 8.
2. Summary of Method
2.1 An inert gas is bubbled through a 10-ml sludge aliquot contained in
a specially designed purging chamber at ambient temperature. The
purgeables are efficiently transferred from the aqueous phase to the
vapor phase. The vapor is swept through a sorbent column where the
purgeables are trapped. After purging is completed, the sorbent col-
umn is heated and backflushed with the inert gas to desorb the purge-
ables onto a gas chromatographic column. The gas chromatograph is
temperature programmed to separate the purgeables which are then de-
tected with a mass spectrometer.1'2
3. Interferences
3.1 Interferences coextracted from the samples will vary considerably
from source to source, depending upon the diversity of the indus-
trial complex or municipality being sampled. Impurities in the
208
-------
purge gas and organic compounds outgassing from the plumbing up-
stream of the trap account for the majority of contamination prob-
lems. The analytical system must be demonstrated to be free from
the interferences under the conditions of the analysis by running
method blanks. Method blanks are run by charging the purging device
with reagent water and analyzing it in the normal manner. The use
of non-TFE plastic tubing, non-TFE thread sealants, or flow control-
lers with rubber components in the purging device should be avoided.
3.2 Samples can be contaminated by diffusion of volatile organic mate-
rials (particularly dichloromethane) through the septum seal into
the sample during shipment and storage. A field blank prepared from
reagent water and carried through the sampling and handling protocol
can serve as a check on such contamination.
3.3 Cross-contamination can occur whenever high-level and low-level sam-
ples are analyzed sequentially. To reduce cross-contamination, it
is recommended that the purging device and sample syringe be rinsed
twice between samples with reagent water to check for cross-contami-
nation. For samples containing large amounts of water-soluble mate-
rials , suspended solids, high-boiling compounds or high organohalide
levels, it may be necessary to wash the purging device with a soap
solution, rinse with distilled water, and then dry in a 105°C oven
between analyses. The trap and other parts of the system are also
subject to contamination; therefore, frequent bakeout and purging
of the entire system may be required.
4. Safety
4.1 The toxicity or carcinogenicity of each reagent used in this method
has not been precisely defined; however, each chemical compound
should be treated as a potential health hazard. From this viewpoint,
exposure to these chemicals must be reduced to the lowest possible
level by whatever means available. The laboratory is responsible
for maintaining a current awareness file of OSHA regulations regard-
ing the safe handling of the chemicals specified in this method. A
reference file of material data handling sheets should also be made
available to all personnel involved in the chemical analysis. Addi-
tional references to laboratory safety are available and have been
identified3 5 for the information of the analyst.
4.2 The following parameters covered by this method have been tentatively
classified as known or suspected, human or mammalian carcinogens:
benzene, carbon tetrachloride, chloroform, 1,4-dichlorobenzene, and
vinyl chloride. Primary standards of these toxic compounds should
be prepared in a hood. A NIOSH/MESA approved toxic gas respirator
should be worn when the analyst handles high concentrations of these
toxic compounds.
209
-------
5. Apparatus and Materials
5.1 Sampling
5.1.1 Vial - 25-ml capacity or larger, equipped with a screw cap
with hole in center (Pierce No. 13075 or equivalent). Deter-
gent wash, rinse with tap and distilled water, and dry at
105°C before use.
5.1.2 Septum - TFE-faced (Pierce No. 12722 or equivalent). Deter-
gent wash, rinse with tap and distilled water, and dry at
105°C for 1 hr before use.
5.2 Sample Preparation
5.2.1 Purge and Trap System. Assemble the system as depicted in
Figures 1 and 2. A commercial version, such as the Tekmar
Liquid Sample Concentrator Model LSC-1, or its equivalent,
may be modified for use by replacing the standard purge tube
with the purge tube shown in Figure 3. The trap is packed
according to Figure 4. In order to function properly, the
trap must be packed in the following order: Place the glass
wool plug in the inlet end of the trap and follow with the
0V-1, Tenax®, silica gel, charcoal, and finally, a second
glass wool plug. Reversing the packing order, i.e., placing
the charcoal in the trap first, will cause the silica gel and
Tenax® layers to become contaminated with charcoal dust caus-
ing poor desorption efficiencies. Install the trap so that
the effluent from the purging device enters the Tenax® end of
the trap.
5.2.2 Glassware
5.2.2.1 Screw-cap vials, 40 ml with TFE-lined caps.
5.2.2.2 Graduated cylinder. 10 ml.
5.2.2.3 Volumetric flasks, 10 ml and 50 ml.
5.2.2.4 Round-bottom flask (optional), 250 ml for sample
compositing.
5.2.3 Analytical balance, for standards preparation.
5.2.4 Roller mill and 1/8-in. stainless steel ball bearings.
5.2.5 Catalytic gas purifier.
5.2.6 Purging gas, He or N2, water compressed, high-purity grade.
5.2.7 Syringes, 10 |Jl, 25 (Jl, 100 jjI, 1 ml, and 10 ml gas tight.
5.2.8 Magnetic stirring motor with 2- to 2.5 cm TFE-coated spin bar.
210
-------
5.3 For Quantitation and Identification
Gas chroraatograph/mass spectrometer/data system, Finnigan 4000 or
equivalent. The GC/MS interface should be a glass jet separator.
The computer system should allow acquisition and storage of repeti-
tive scan data throughout the GC/MS runs. Computer software should
be available to allow searching of GC/MS data for display of ex-
tracted ion current profiles (EICP) and integration of the peaks.
The GC/MS should be fitted with a stainless steel or glass column
packed with 0.2% Carbowax 1500 on Carbopack C, or equivalent. Typ-
ical column dimensions are 1/8-in. OD stainless steel or 2-mm ID
glass and 2.4 to 2.8 m in length.
6. Reagents
6.1 Trap Materials
6.1.1 Methylsilicone packing - 3% OV-1 on Chromosorb-W (60/80 mesh)
or equivalent.
6.1.2 2,6-Diphenylene oxide polymer - Tenax-GC® (60/80 mesh), chro-
matographic grade or equivalent.
6.1.3 Silica gel - Davison Grade 15 (35/60 mesh) or equivalent.
6.1.4 Coconut charcoal - Barnaby Chaney No. CA-580-26 (26 mesh),
Lot No. M-2649 or equivalent.
6.2 Reagent Water (purgeable organic-free). Generate organic-free water
by passing tap water through a carbon filter bed containing about 1
lb of activated carbon and purging overnight with prepurified nitro-
gen. A Millipore Super-Q Water System or its equivalent may be used
to generate organic-free deionized water. Organic-free water can
also be prepared by boiling distilled water for 15 min. While still
hot, transfer to a glass-stoppered bottle. Cool to room temperature.
Continuously purge the water during storage. Test organic-free water
daily by analyzing according to the method described in Section 10.
6.3 Methanol. "Distilled in Glass" or equivalent, stored in original
containers and used as received.
6.4 Analytical Standards
6.4.1 Primary Standards. Prepare standard stock solutions (at ap-
proximately 2 pg/fJl) by adding, from a 100—pi syringe, 1 to 2
drops of the 99+% pure reference standard to methanol (9.8 ml)
contained in a tared 10-ml volumetric flask (weighed to near-
est 0.1 mg). Add the component so that the two drops fall
into the alcohol and do not come in contact with the neck of
the flask. Prepare gaseous standards, e.g., vinyl chloride,
in a similar manner using a 10.0 ml valved gas-tight syringe
with a 2-in. needle. Fill the syringe (10.0 ml) with the gas-
eous compound. Weigh the 10-ml volumetric flask containing
9.8 ml of methyl alcohol to 0.1 mg. Lower the syringe needle
211
-------
to about 5 mm above the methyl alcohol meniscus and slowly
inject the standard into the flask. The gas rapidly dissolves
in the methyl alcohol. Reweigh the flask and use the weight
gain to calculate the concentration of the standard. Dilute
to volume, mix, and transfer to a 10-ml screw-cap vial with a
silicone rubber/TFE cap liner. Gas stock standards are gener-
ally stable for at least 1 week when maintained at less than
0°C. With the exception of 2-chloroethylvinyl ether, stock
standards of compounds that boil above room temperature are
generally stable for at least 4 weeks when stored at 4°C.
(Safety Caution: Because of the toxicity of most organohal-
ides, dilutions should be made in a glove box suitable for
handling carcinogens. It is advisable to use an approved
respirator when high concentrations of such materials must be
handled in a fume hood.)
6.4.2 Working Standards. From the primary dilutions, prepare 10 ml
of a multicomponent secondary dilution mixture in methyl alco-
hol at a concentration of 50 ng/(Jl containing each of the com-
pounds to be determined. Prepare 50 ml of a 50-ng/ml standard
from the 50 ng/pl standard by dosing 50.0 fjl into 50.0 ml of
organic-free water. Additional working standards should be
prepared as required to bracket concentrations of compounds
for referee analyses.
6.5 Internal Standard Spiking Solution. From stock standard solutions
prepared as above, add a volume to give 1,000 pg each of bromochloro-
methane, 2-bromo-l-chloropropane, and 1,4-dichlorobutane to 45 ml of
organic-free water (blank water) contained in a 50-ml volumetric
flask, mix, and dilute to volume. Dose 9.0 |Jl of this internal stan-
dard spiking solution into every sample and reference standard ana-
lyzed. Prepare a fresh method recovery spiking solution on a weekly
basis.
6.6 Surrogate Standard Spiking Solution. Select a minimum of three sur-
rogate compounds from Table 2. Prepare and store stock standard and
spiking solutions as described in Section 6.5
7. Calibration
7.1 Establish GC/MS operating parameters equivalent to those indicated
in Section 11. The purge-trap GC/MS system can be calibrated using
the external standard technique (Section 7.2) or the internal stan-
dard technique (Section 7.3).
7.2 External Standard Calibration Procedure
7.2.1 Prepare calibration standards at a minimum of three concen-
tration levels for each parameter of interest by adding vol-
umes of one or more stock standards to 10.0-ml aliquots of
reagent water. Analyze immediately. One of the external
standards should be at a concentration near, but above, the
212
-------
method detection limit, and the other concentrations should
correspond to the expected range of concentrations found in
real samples or should define the working range of the detec-
tor.
7.2.2 Analyze each calibration standard with the purge-trap GC/MS
system described in Section 11. Tabulate peak heights or
area responses against the concentration of the analyte in
the standard. The results can be used to prepare a calibra-
tion curve for each compound. Alternatively, if the ratio of
response to concentration analyzed (calibration factor) is a
constant over the working range (< 10% relative standard de-
viation, RSD), linearity through the origin can be assumed
and the average ratio or calibration factor can be used in
place of a calibration curve.
7.2.3 Verify the working calibration curve or calibration factor on
each working day by the measurement of one or more calibration
standards. If the response for any parameter varies from the
predicted response by more than ± 10%, the test must be re-
peated using a fresh calibration standard. Alternatively, a
new calibration curve or calibration factor must be prepared
for that compound.
7.3 Internal Standard Calibration Procedure. To use this approach,
select one or more internal standards that are similar in analytical
behavior to the compounds of interest. The analyst must demonstrate
that the measurement of the internal standard is not affected by
method or matrix interferences. The internal standards used for this
procedure should include bromochloromethane, 2-bromo-l-chloropropane,
and 1,4-dichlorobutane.
7.3.1 Prepare calibration standards at a minimum of three concen-
tration levels for each parameter of interest by adding vol-
umes of one or more stock standards to a 10.0-ml aliquot of
reagent water, add 9.0 |jl of the internal standard spiking
solution, and analyze immediately. One of the standards
should be at a concentration near, but above, the method de-
tection limit and the other concentrations should correspond
to the expected range of concentrations found in real samples
or should define the working range of the detector.
7.3.2 Analyze each calibration standard with the purge-trap GC/MS
system described in Section 11. Tabulate peak height or area
responses against concentration for each compound and
internal standard. Calculate relative response factors (RRF)
for each compound using Equation 1.
RRF = CAsB.s)/CAisBs) (Eq. 1)
213
-------
where: Ag = Response for the parameter to be measured.
A^g = Response for the internal standard.
B. = Mass of the internal standard (ng).
is v
Bs = Mass of the parameter to be measured (ng).
If the RRP value over the working range is a constant (< 10%
RSD), the RKF can be assumed to be nonvariant and the average
RRF can be used for calculations. Alternatively, the results
can be used to plot a calibration curve of response ratios,
A /A. , vs. RRF.
s' is'
7.3.3 Verify the working calibration curve or RRF on each working
day by the measurement of one or more calibration standards.
If the response for any parameter varies from the predicted
response by more than ± 10%, the test must be repeated using
a fresh calibration standard. Alternatively, a new calibra-
tion curve must be prepared for that compound.
7.4 Daily Calibration of the Gas Chromatography-Mass Spectrometry
(GC/MS) System
7.4.1 Evaluate the system performance each day that it is used for
the analysis of samples or blanks by injecting 20 ng of £-
bromofluorobenzene (BFB) into the GC inlet. Check to be sure
that the performance criteria listed in Table 3 are met. If
the criteria are not met, the instrument must be retuned to
satisfy those criteria before continuing.
To perform the calibration test, the following instrumental
parameters are required.
Electron energy, 70 eV (nominal)
Mass range, m/e 20-275
Scan time to provide at least 5 scans per peak, typi-
cally 5 s or less.
7.4.2 At the beginning of each working day, verify the calibration
of the system by analyzing a standard (Section 7.3.3).
8. Quality Control
Each laboratory that uses this method is required to operate a formal
quality control program. The minimum requirements of this program con-
sist of an initial demonstration of laboratory capability and the analy-
sis of spiked samples as a continuing check on performance. The labora-
tory is required to maintain performance records to define the quality of
data that is generated. Before performing any analyses, the analyst must
demonstrate the ability to generate acceptable accuracy and precision
214
-------
with this method. This ability is established as described in Section
8.1. In recognition of the rapid advances that are occurring in chro-
matography, the analyst is permitted certain options to improve the sepa-
rations or lower the cost of measurements. Each time such modifications
are made to the method, the analyst is required to repeat the procedure
in Section 8.1. The laboratory must spike all samples with surrogate
standards to monitor continuing laboratory performance. This procedure
is described in Section 8.4.
8.1 Demonstrate Acceptable Performance. Before performing any analyses,
the analyst must demonstrate the ability to generate acceptable accu-
racy and precision with this procedure.
8.1.1 For each compound to be measured, select a spike concentra-
tion representative of the expected levels in the samples.
Using stock standards, prepare a quality control check stan-
dard in methanol 1,000 times more concentrated than the
selected concentrations.
8.1.2 Syringe 10.0 |jl of the check standard to each of a minimum of
four 10.0-ml aliquots of reagent water. A representative
sludge can be used in place of the reagent water, but one or
more additional aliquots must be analyzed to determine back-
ground levels, and the spike level must exceed twice the back-
ground level for the test to be valid. Analyze the aliquots
according to the method beginning in Section 10.
8.1.3 Calculate average recovery (R) and standard deviation (s), in
percentage recovery, for the results. Sludge background cor-
rections must be made before R and s calculations are performed.
8.1.4 Using the appropriate data from Table 7, determine the recov-
ery and single operator precision expected for the method for
each parameter, and compare these results to the values cal-
culated in Section 8.1.3. If the data are not comparable,
the analyst must review potential problem areas and repeat
the test.
8.2 Precision and Accuracy Statement. The analyst must calculate method
performance criteria for each of the surrogate standards.
8.2.1 Calculate upper and lower control limits for method perfor-
mance for each surrogate standard, using the values for R and
s calculated in Section 8.1.3:
Upper Control Limit (UCL) = R + 3 s
Lower Control Limit (LCL) = R - 3 s
The UCL and LCL can be used to construct control charts6 that
are useful in observing trends in performance.
215
-------
8.2.2 For each surrogate standard, the laboratory must develop and
maintain separate accuracy statements of laboratory perfor-
mance for wastewater samples. An accuracy statement for the
method is defined as R ± s. The accuracy statement should be
developed by the analysis of four aliquots of sludge as de-
scribed in Section 8.1.2, followed by the calculation of R
and s. Alternately, the analyst may use four sludge data
points gathered through the requirement for continuing quality
control in Section 8.4. The accuracy statements should be
updated regularly.6
8.3 Surrogate Spikes. The laboratory is required to spike all of their
samples with the surrogate standard spiking solution to monitor spike
recoveries. If the recovery for any surrogate standard does not fall
within the control limits for method performance, the results reported
for that sample must be qualified as described in Section 13.3. The
laboratory should monitor the frequency of suspect data to ensure
that it remains at or below 5%.
8.4 System Blanks. Analyze daily an organic-free water blank to demon-
strate that all interferences from the analytical system are under
control. The intensities of EICPs for the internal standards gives
an overall check of the system sensitivity.
8.5 Fortified and Replicate Samples
8.5.1 Sample Selection. Select at least 30% of all samples or at
least one sample from each sampling location for fortified
and replicate aliquot analysis. Analyze the selected samples
in duplicate to determine unspiked analyte concentrations.
Using the procedures described in Section 8.5.2, spike and
analyze duplicate aliquots of the selected samples. Analyze
a third unspiked aliquot of the selected samples on the same
day that the corresponding spiked aliquots are analyzed.
8.5.2 Spiking Procedures. Determine the volume of an empty vial to
within 0.5 ml by filling the vial with volatile organics-free
water and then measuring the volume of water with a graduated
cylinder. Mark the volume on the vial and allow it to air-dry.
Add 3 to 5 precleaned ball bearings to the vial. Transfer the
sample to the vial and spike with the representative purgeable
compounds listed in Table 4 and other compounds identified in
the sample, as available, at two times the concentration found
in the unspiked sample or at 10 times the lower limit of de-
tection, whichever is greater. Seal the vial and place on a
roller mill in a 4°C cold room; tumble the sample for 16 h
before analysis.
8.6 Field Blanks. Analyze a field blank for each day of sampling at each
sampling location.
216
-------
8.7 Additional Quality Control. It is recommended that the laboratory
adopt additional quality assurance practices for use with this method.
The specific practices that are most productive depend upon the needs
of the laboratory and the nature of the samples. Whenever possible,
the laboratory should perform analysis of reference materials and
participate in relevant performance evaluation studies.
9. Sampling and Preservation
9.1 Sampling. Samples must be collected in 40-ml screw-cap vials with
zero headspace and sealed with TFE-lined septa. Before using, wash
all sample bottles and TFE seals in detergent and rinse with tap
water and finally distilled water. Allow the bottles and seals to
air-dry at room temperature, heat in a 105°C oven for 1 h, then al-
low to cool in an area known to be free of organics.
NOTE: Do not heat the TFE seals for extended periods of time (more
than 1 h) because the silicone layer slowly degrades at 105°C.
9.2 Preservation. Ice samples immediately after collection, refriger-
ate at 4°C, and purge within 10 days. Desorb and complete the analy-
ses immediately after purging.
9.3 Special Preservation for Aromatics. Experimental evidence indicates
that some aromatic compounds, notably benzene, toluene, and ethyl
benzene are susceptible to rapid biological degradation under certain
environmental conditions.2 Refrigeration alone may not be adequate
to preserve these compounds in sludges for more than 7 days. For
this reason, a separate sample should be collected, acidified, and
analyzed when these aromatics are to be determined. Collect about
500 ml of sample in a clean container. Adjust the pH of the sample
to about 2 by adding HC1 (1+1) while stirring. Check pH with narrow
range (1.4 to 2.8) pH paper. Fill a sample container as described
in Section 9.1. If chlorine residual is present, add sodium thio-
sulfate to another sample container and fill as in Section 9.1 and
mix thoroughly.
10. Sample Preparation and Purging
10.1 Sample Compositing. Sludge samples for the analysis of purgeable
compounds are typically collected as grab samples. Several samples
may be composited with minimal analyte losses. Chill the appropri-
ate sample vials and a clean 250-ml round-bottom flask by immersing
them in an ice bath. Gently pour the entire contents of each vial
into the flask and swirl gently. Vigorous mixing must be avoided
to prevent analyte losses. Prepare a number of composite aliquots
sufficient for subsequent analysis by gently pouring the composited
sludge into precleaned 40-ml vials. Seal the vials with zero head-
space using TFE-lined septa and screw-caps. Store the composited
samples at 4°C.
217
-------
10.2 Sample Purging. Condition the adsorbent trap at 200°C with nitro-
gen or helium flow. Allow the trap to cool to ambient room temper-
ature and turn off the gas flow through the trap and purge tube
before sample purging. Remove an aliqout of the sludge sample by
gently pouring the sludge into a 10-ml graduated cylinder. Fill
the cylinder just to the 10.0-ml mark. Dose the aliquot with 9 pi
of the internal standard spiking solution by slowly injecting the
solution (with a 10—|Jl syringe) 3 to 4 cm under the surface of the
sludge. Gently pour the spiked sludge aliquot into the purge tube.
If solids adhere to the inner walls of the graduated cylinder,
rinse the cylinder with a small amount of organics-free water and
add the rinsings to the purge tube. Seal the purge tube and turn
on the purge gas flow (40 ml/min). Purge the sample for 12 min
while maintaining the sample and trap at ambient room temperature.
11. Analysis of the Sample Purge
Analyze the sample purge by GC/MS using the Carbowax 1500/Carbopack C
column described in Section 5.2 operated with a helium carrier gas flow
of 30 ml/min. Heat the trap to 180 to 200°C. Backflush it for 3 min
into the gas chromatograph, with the oven at 40°C. At the end of the
3-min period, program the column at 10°C/min to 170°C. Hold at this
temperature until all compounds have eluted. An example of the separa-
tion achieved by this column is shown in Figure 5. Relative retention
times for the purgeable compounds are listed in Table 5. The purging
device must be removed from the instrument and thoroughly rinsed with
volatile organic-free water between each sample. The trap must be con-
ditioned at 200°C with flow for 10 min between each sample.
12. Qualitative Identification
12.1 Obtain EICPs for the primary ion (Table 5) and, if available, at
least two secondary ions for each parameter of interest. The fol-
lowing criteria must be met to make a qualitative identification.
12.1.1 The characteristic ion of each parameter of interest must
maximize in the same or within one scan of each other.
12.1.2 The retention time must fall within ± 30 s of the reten-
tion time of the authentic compound.
12.1.3 The relative peak heights of the three characteristic ions
in the EICPs must fall within ± 20% of the relative inten-
sities of these ions in a reference mass spectrum. The
reference mass spectrum can be obtained from a standard
analyzed in the GC/MS system or from a reference library.
12.2 Structural isomers that have very similar mass spectra and less
than 30 s difference in retention time, can be explicitly iden-
tified only if the resolution between authentic isomers in a stan-
dard mix is acceptable. Acceptable resolution is achieved if the
baseline to valley height between the isomers is less than 25% of
the sum of the two peak heights. Otherwise, structural isomers
are identified as isomeric pairs.
218
-------
13. Calculations
13.1 When a parameter has been identified, the quantitation of that
parameter should be based on the integrated abundance from the
EICP of the first listed characteristic ion given in Table 5. If
the sample produces an interference for the primary ion, use a sec-
ondary characteristic ion to quantitate. Quantitation may be per-
formed using the external or internal standard techniques.
13.1.1
13.1.2
A. = Area of the characteristic ion for the
i s
internal standard.
Bis = Mass of the internal standard (ng).
V = Volume of sample purged (ml).
13.2 Report results in micrograms per liter. The results for cis- and
trans-1,3-dichloropropene should be reported as total 1,3-dichloro-
propene (Storet No. 34561, CAS No. 542-75-6). When duplicate and
spiked samples are analyzed, report all data obtained with the
sample results.
13.3 If any of the surrogate standard recoveries fall outside the con-
trol limits which were established as directed in Section 8.3,
data for all parameters in that sample must be labeled as suspect.
14. Method Performance
Performance data for the application of this method to both POTW and
industrial sludges are shown in Table 7. These data were generated by
analyzing duplicate aliquots of unspiked sludge and triplicate aliquots
of sludge spiked at 2, 20, and 200 times the typical detection limits
for representative purgeable compounds. The method was evaluated using
If the external standard calibration procedure is used,
calculate the concentration of the parameter being mea-
sured from the area of the characteristic ion using the
calibration curve or calibration factor in Section 7.2.
If the internal standard calibration procedure was used,
calculate the concentration in the sample using the rela-
tive response factor (RRF) determined in Section 7.3 and
Equation 2.
(As)(Bis}
Concentration |Jg/L = —)(rrf)(v) (Eq* 2)
X s
where: A = Area of the characteristic ion for the
parameter or surrogate standard to be
219
-------
three POTW sludges and two industrial sludges. The recoveries and stan-
dard deviations represent the results from all determinations from each
sludge type. The mean RSD for triplicates illustrate the precision for
triplicate analyses. Although the recoveries observed were generally
good, many recovery determinations were influenced by relatively high
concentrations of the spiking compounds in the unspiked sludges. In most
cases, the results of recovery determinations for compounds present in
the unspiked samples at concentrations in excess of the spiking level
were not included in Table 7. The small mean RSD for triplicates observed
for most compounds indicate that the largest component of the method var-
iance can be attributed to differences in the characteristics of the test
sludges. These data represent the results from one laboratory.
15. References
1. Bellar, T. A., and J. J. Lichtenberg, Journal American Water Works
Association, 66, p. 739 (1964).
2. Bellar, T. A., and J. J. Lichtenberg, "Semi-Automated Headspace
Analysis of Drinking Waters and Industrial Waters for Purgeable
Volatile Organic Compounds," Measurement of Organic Pollutants in
Water and Wastewater, C. E. Van Hall, editor, American Society for
Testing and Materials, Philadelphia, PA. Special Technical Publi-
cation 686, 1978.
3. "Carcinogens - Working with Carcinogens," Department of Health,
Education, and Welfare, Public Health Service, Center for Disease
Control, National Institute of Occupational Safety and Health,
Publication No. 77-206, August 1977.
4. "OSHA Safety and Health Standards, General Industry," (29CFR1910),
Occupational Safety and Health Administration, OSHA 2206 (Revised
January 1976).
5. "Safety in Academic Chemistry Laboratories," American Chemical
Society Publication, Committee on Chemical Safety, 3rd Edition,
1979.
6. "Handbook of Analytical Quality Control in Water and Wastewater
Laboratories," EPA-600/4-79-019, U.S. Environmental Monitoring and
Support Laboratory, Cincinnati, Ohio 45268, March 1979.
16. Additional Sources
1. "Sampling and Analysis Procedures for Screening of Industrial
Effluents for Priority Pollutants." U.S. Environmental Protection
Agency, Environmental Monitoring and Support Laboratory, Cincinnati,
OH 45268, March 1977, Revised April 1977. Effluent Guidelines
Division, Washington, DC 20460.
220
-------
2. "Methods 330.4 (Titrimetric, DPD-FAS) and 330.5 (Spectrophotometric,
DPD) for Chlorine, Total Residual," Methods for Chemical Analysis of
Water and Wastes, EPA 600/4-79-020, U.S. Environmental Protection
Agency, Environmental Monitoring and Support Laboratory - Cincinnati,
OH 45268. In preparation.
3. "Preservation and Maximum Holding Time for the Priority Pollutants,"
U.S. Environmental Protection Agency, Environmental Monitoring and
Support Laboratory, Cincinnati, OH 45268. In preparation.
4. Budde, W. L., and J. W. Eichelberger, "Performance Tests for the
Evaluation of Computerized Gas Chromatography/Mass Spectrometry
Equipment and Laboratories," EPA-600/4-80-025, U.S. Environmental
Protection Agency, Environmental Monitoring and Support Laboratory,
Cincinnati, OH 45268, p. 16, April 1980.
5. Kleopfer, R. D., "Priority Pollutant Methodology Quality Assurance
Review," U.S. Environmental Protection Agency, Region VII, Kansas
City, KS. Seminar for Analytical Methods for Priority Pollutants,
Norfolk, VA, January 17-18, 1980, U.S. Environmental Protection
Agency, Office of Water Programs, Effluent Guidelines Division,
Washington, D.C. 20460.
221
-------
TABLE 1. PURGEABLE ORGANIC PRIORITY POLLUTANTS
Compound
CAS No.
Acrolein
107-02-8
Acrylonitrile
107-13-1
Chlororaethane
74-87-3
Dichlorodifluoromethane
75-71-8
Bromoraethane
74-83-9
Vinyl chloride
75-01-4
Chloroethane
75-00-3
Dichloromethane
75-09-2
Trichlorofluoromethane
75-69-4
1,1-Dichloroethene
75-35-4
1,1-Dichloroethane
75-34-3
trans-1,2-Dichloroethene
540-59-0
Chloroform
67-66-3
1,2-Dichloroethane
107-06-2
1,1,1-Trichloroethane
71-55-6
Carbon tetrachloride
56-23-5
Bromodichloromethane
75-27-4
1,2-Dichloropropane
78-87-5
Benzene
71-43-2
trans-1,3-Dichloropropene
542-75-6
Trichloroethene
79-01-6
cis-1,3-Dichloropropene
542-75-6
Dibromochloromethane
124-48-1
1,1,2-Trichloroethane
79-00-5
2-Chloroethylvinyl ether
110-75-8
Bromoform
75-25-2
Tetrachloroethene
127-18-4
Toluene
108-88-3
1,1,2,2-Tetrachloroethane
79-34-5
Chlorobenzene
108-90-7
Ethylbenzene
100-41-4
222
-------
TABLE 2
SUGGESTED SURROGATE AND INTERNAL STANDARDS
Retention
Time
Primary
Secondary
Compound
(minutes)
Ion
Ions
Surrogate Standards
Benzene d-6
17.0
84
4-Bromofluorobenzene
28.3
95
174, 176
1,2-Dichloroethane d-4
12.1
102
-
1,4-Difluorobenzene
19.6
114
63,88
Ethylbenzene d-5
26.4
111
-
Ethylbenzene d-10
26.4
98
-
Fluorobenzene
18.4
96
70
Pentafluorobenzene
23.5
168
-
Internal Standards
Bromochloromethane
9.3
128
49, 130, 51
2-Bromo-l-chloropropane
18.5
77
79, 156
1,4-Dichlorobutane
22.9
55
90, 92
a For chromatographic conditions, see Table 5.
223
-------
TABLE 3. £-BROMOFLUOROBENZENE IONS AND
ION ABUNDANCE CRITERIA3
m/e Ion abundance criteria
50 25-40% of m/e 95
75 30-60% of m/e 95
95 base peak, 100% relative abundance
96 5-9% of m/e 95
173 < 2% of m/e 95
174 > 50% of m/e 95
175 5-9% of m/e 174
176 > 95% but < 101% of m/e 95
177 5-9% of m/e 176
a Eichelberger, J. W. , L. E. Harris, and W. L. Budde,
"Reference Compound to Calibrate Ion Abundance
Measurement in Gas Chromatography-Mass Spectrom-
etry Systems," Analytical Chemistry, 47, 995-1000
(1979).
224
-------
TABLE 4. REPRESENTATIVE PURGEABLE COMPOUNDS
FOR RECOVERY STUDIES
Benzene
Carbon tetrachloride
Chlorobenzene
Chloroform
1,2-Dichloroethane
1,1-Dichloroethene
Ethyl benzene
Tetrachloroethene
1,1,1-Trichloroethane
Trichloroethene
Vinyl chloride
225
-------
TABLE 5. RELATIVE RETENTION TIMES OF PURGEABLE PRIORITY POLLUTANTSa
Compound RRT*5
Chloromethane 0.116
Bromomethane 0.146
Dichlorodifluoromethane 0.169
Vinyl chloride 0.179
Chloroethane 0.209
Dichloromethane 0.312
Trichlorofluoromethane 0.435
1,1-Dichloroethene 0.475
Bromochloromethane (IS) 0.502
1.1-Dichloroethane 0.545
trans-1,2-Dichloroethene 0.585
Chloroform 0.621
1.2-Dichloroethane 0.661
1.1.1-Trichloroethane 0.728
Carbon tetrachloride 0.751
Bromodichloromethane 0.784
c
Bis-chloromethyl ether Unknown
1,2-Dichloropropane 0.854
trans-1,3-Dichloropropene 0.870
Trichloroethene 0.900
Benzene 0.917
Dibromochloromethane 0.934
cis-1,3-Dichloropropene 0.937
1.1.2-Trichloroethane 0.937 ^
2-Chloroethylvinyl ether Unknown
2-Bromo-l-chloropropane (IS) 1.000
Bromoform 1.073
1,1,2,2-Tetrachloroethane 1.206
Tetrachloroethene 1.213
1,4-Dichlorobutane (IS) 1.236
Toluene 1.292
Chlorobenzene 1.412
Ethylbenzene 1.641
Acrolein Unknown
Q
Acrylonitrile Unknown
a These data were obtained under the following conditions: GC column -
glass, 8-ft long x 0.1 in. I.D. packed with Carbopack C (60/80 mesh),
coated with 0.2% Carbowax 1500; carrier flow - 30 ml/min; oven tempera-
ture - initial 40°C held for 3 min, programmed 10°C/min to 170°C and
held until all compounds eluted.
b Retention times relative to 2-bromo-l-chloropropane with an absolute
retention time of 903 s.
c Bis-Chloromethyl ether has a half-life of about 10 s in aqueous mixtures,
d 2-Chloroethylvinyl ether may be unstable in aqueous mixtures. Retention
time is unknown.
e Acrolein and acrylonitrile do not purge efficiently from aqueous mixtures.
Retention times under these conditions are not known.
226
-------
TABLE 6. CHARACTERISTIC IONS OF PURGEABLE ORGANICS
EI
ions
Ion used to
Compound
(relative
intensity)
quantify
oromethane
50(100);
52(33)
50
)mome thane
94(100);
96(94)
94
:hlorodif luoromethane
85(100);
87(33);
101(13); 103(9)
101
iyl chloride
62(100);
64(33)
62
.oroethane
64(100);
66(33)
64
•hloromethane
49(100);
51(33);
84(86); 86(55)
84
chlorofluororaethane
101(100)
; 103(66)
101
¦Dichloroethene
61(100);
96(80);
98(53)
96
•mo chl orome thane (IS)
49(100);
130(88);
128(70); 51(33)
128
-Dichloroethane
63(100);
65(33);
83(13); 85(8); 98(7);
63
100(4)
as-1,2-Dichloroethene
64(100);
96(90);
98(57)
96
oroform
83(100);
85(66)
83
>Dichloroethane
62(100);
64(33);
98(23); 100(15)
98
il-Trichloroethane
98(100);
99(66);
117(17); 119(16)
97
fbon tetrachloride
117(100)
; 119(96)
; 121(30)
117
MDodichlorome thane
83(100);
85(66);
127(13); 129(17)
127
'•chloromethyl ether
79(100);
81(33)
79
•"Dichloropropane
63(100);
65(33);
112(4); 114(3)
112
£s-1,3-Dichloropropene
chloroethene
75(100);
77(33)
75
95(100);
97(66);
130(90); 132(85)
130
'2ene
78(100);
52(15)
78
Tomochloromethane
129(100)
; 127(78)
; 208(13); 206(10)
127
:*1,3-Dichloropropene
75(100);
77(33)
75
>2-Trichloroethane
83(95);
85(60); 97(100); 99(63);
97
Ihloroethylvinyl ether
132(9)
; 134(8)
63(95);
65(32); 106(18)
106
lromo-1-chloropropane (IS)
77(100);
79(33);
156(5)
77
"Do form
171(50);
173(100)
; 175(50); 250(4);
173
252(11); 254(11); 256(4)
>2,2-Tetrachloroethane
83(100);
85(66);
131(7); 133(7);
168
166(5)
; 168(6)
.rachloroethene
129(64);
131(62);
164(78); 166(100)
164
•Dichlorobutane (IS)
55(100);
90(30);
92(10)
55
Uene
91(100);
92(78)
92
orobenzene
112(100)
; 114(33)
112
lylbenzene
91(100);
106(33)
106
'olein
26(49);
27(100);
55(64); 56(83)
56
7lonitrile
26(100);
51(32);
52(75); 53(99)
53
227
-------
TABI.E 7. ACCURACY AND PRECISION FOR PURGEABLE ORGANICS
lTOTW Sludges Two Industrial Sludges
Spike Recovery Spike Recovery
Mean Mean
Compound
Detemiin-
atious
Spike level
_(l'g/t) __ .
Hiu Hax
Mean
(%)
Standard
devia-
tion
NSI) for
trjpl .
(%)
Determi n-
alioiis
Spike level
_
Hin Hax
Hean
_cx)
Standard
devia-
tion
RSD for
tripl.
__(%)_
Belize hp
17
1
100
160
55
16
15
1
100
98
25
16
Chlot'ofom
23
2
200
100
58
20
17
2
200
76
22
20
1,l-Diiiiloroelhane
27
5
500
170
53
15
15
5
500
110
51
26
Tet radii oroethane
15
3
300
150
33
16
12
3
300
150
70
24
Vinyl chloride
18
50
500
130
38
14
16
5
500
110
47
13
1,2-l)i ill loroet li.ine
27
5
500
140
51
14
12
5
500
100
28
16
Trichloroelhene
18
20
200
160
69
24
12
2
200
140
44
21
1,1,1-Trichloroelliane
21
16
1,600
130
47
23
15
16
1,600
110
40
11
Clilorobenzene
21
2
200
120
36
16
15
2
200
160
62
14
Klliyl benzene
18
5
500
120
26
15
12
5
500
150
55
14
-------
Carrier Gas
Pressure Flow Control
Regulator
To Gas Chromatograph
Trap Inlet (Tenax End)
~"Resistance Wire
Purge Gas L,
Flow Control
Trap
Flow
Heater
Control
Off
6-Port
Valve
Vent
Purging
Device
Note:
All Lines Between Trap and GC
Should be Heated to 80 °C
Figure 1. Schematic of the purge-trap system in the purge mode.
229
-------
Carrier Gas
Pressure Flow Control
Regulator 1
— ' >'<
Purge Gas
Flow Control
6-Port
Valve
To Gas Chromatograph
/-Trap Inlet (Tenax End)
/ y-Resistance Wire
Trap
<(180°C
Heater
Control
Purging
Device
Note:
All Lines Between Trap and GC
Should be Heated to 80 °C
Figure 2. Schematic of the purge-trap system in the desorb mode.
230
-------
Medium
Frit
-—38 mm-O .D.
100 mm
Figure 3. Bottom frit purge tube.
231
-------
PACKING
CONSTRUCTION
Glass Wool 5mm
Activated
Charcoal
7.7 cm
Grade 15
Silica Gel
7.7cm
3% OV-1 1cm
Glass Woof 5 mm
I
I
Tanax 7.7 cm
ff
' A
i
7^1/Foot Resistance
Wire Wrapped Solid
(Double Layer)
15cm
70/Foot Resistance
Wire Wrapped Solid
(Single Layer)
Compression Fitting
Nut and Ferrules
Thermocouple/
Controller
Sensor
Electronic
Temperature
Control nnd
Pyrometer
Tubing, 25cm
0.105in. ID
0.122 in. OD
Stainless Steel
Trap Inlet
Figure 4. Trap packings and construction.
232
-------
100.0
RIC
rsa
Ca>
CO
~ w
0 4) 01 .C 91
-A O. «l B 4>
CM O
952320
5:80
10:00
15:00
28:00
25:80 TIME
Figure 5. CJC/MS chromalogram for the purgeable compounds.
-------
METHOD 625-S
PROTOCOL FOR THE ANALYSIS OF EXTRACTABLE ORGANIC PRIORITY POLLUTANTS
IN INDUSTRIAL AND MUNICIPAL WASTEWATER TREATMENT SLUDGE
1. Scope and Application
1.1 This method is used for the determination of the base, neutral, and
acid-extractable organic compounds listed in Table 1.
1.2 The method is for qualitative and quantitative analysis of these
compounds in municipal and industrial wastewater treatment sludges.
The procedure requires the use of a gas chromatograph/mass spectrom-
eter (GC/MS) as the final detector.
1.3 The method detection limit for each compound is dependent on
the compound characteristics and the particular sludge analyzed.
However, typical detection limits are 200 to 300 (jg/liter for pub-
licly owned treatment works (POTW) sludges with 1 to 5% total solids.
1.4 This method is restricted to use by or under the close supervision
of experienced analysts. Each analyst must demonstrate the ability
to generate acceptable results with this method using the procedure
described in Section 8.
2. Summary of Method
This method (Figure 1) uses repetitive solvent extraction aided by a high-
speed homogenizer. The extract is separated by centrifugation and removed
with a pipette or syringe. Sludges are extracted at pH I 11 to isolate
base/neutral compounds and at pH 5 2 to isolate acidic compounds. Extracts
containing base/neutral compounds are cleaned by silica gel or florisil
chromatography or by gel permeation chromatography (GPC). Extracts con-
taining the acidic compounds are cleaned by GPC. The organic priority
pollutants are determined in the cleaned extracts by capillary column or
packed column GC/MS. Option A, i.e., extract cleanup by silica gel or
florisil chromatography and analysis by capillary GC/MS (HRGC/MS) is
preferred since HRGC/MS allows easier data interpretation. Qualitative
identification of compounds is performed using the retention time and the
relative abundance of three characteristic ions. Quantitative analysis
is performed using external or internal standard techniques with a single
characteristic ion.
234
-------
3. Interferences
3.1 Method interferences may be caused by contaminants in solvents, re-
agents, glassware, and other sample processing hardware that lead to
discrete artifacts and/or elevated baselines in GC/MS chromatograms.
All of these materials must be routinely demonstrated to be free from
interferences under the conditions of the analysis by running labora-
tory reagent blanks as described in Section 8.4.
3.1.1 Glassware must be scrupulously cleaned. All glassware should
be cleaned as soon as possible after use by rinsing with the
last solvent used in it. This should be followed by detergent
washing with hot water, and rinses with tap water and reagent
water. It should then be drained dry, and heated in a muffle
furnace at 400°C for 15 to 30 min. Some thermally stable ma-
terials, such as PCBs, may not be eliminated by this treat-
ment. Solvent rinses with acetone and pesticide quality di-
chloromethane may be substituted for the muffle furnace heat-
ing. Volumetric glassware should not be heated in a muffle
furnace. After drying and cooling, glassware should be sealed
and stored in a clean environment to prevent any accumulation
of dust or other contaminants. It should be stored inverted
or capped with aluminum foil.
3.1.2 The use of high purity reagents and solvents helps to mini-
mize interference problems. Purification of solvents by
distillation in all-glass systems may be required.
3.2 Matrix interferences may be caused by contaminants that are coex-
tracted from the sample. The extent of matrix interferences will
vary considerably from source to source, depending upon the nature
and diversity of the wastewater treatment system being sampled. The
cleanup procedures in Section 11 can be used to overcome many of
these interferences, but unique samples may require additional
cleanup approaches to achieve acceptable detection limits.
4. Safety
4.1 The toxicity or carcinogenicity of each reagent used in this method
has not been precisely defined. However, each chemical compound
should be treated as a potential health hazard. From this viewpoint,
exposure to these chemicals must be reduced to the lowest possible
level by whatever means available. The laboratory is responsible
for maintaining a current awareness file of OSHA regulations regard-
ing the safe handling of the chemicals specified in this method. A
reference file of material data handling sheets should also be made
available to all personnel involved in the chemical analysis.
4.2 The following parameters covered by this method have been tentatively
classified as known or suspected, human or mamalian carcinogens;
benzo[a]anthracene, benzidine, 3,3'-dichlorobenzidine, benzo[a]pvrene,
a-BHC, p-BHC, 6-BHC, v-BHC, dibenzo[a,h]anthracene, N-nitrosodimethyl-
amine, 4,4'-DDT, and polychlorinated biphenyls.
235
-------
5. Apparatus and Materials
5.1 Sampling, Extraction, and Extract Cleanup
5.1.1 Emulsifier-Tekmar Tissumizer® or equivalent, high capacity.
5.1.2 Centrifuge, capable of handling 200 ml bottles.
5.1.3 Centrifuge bottles with TFE lined screw caps, 200 ml.
5.1.4 Kudema-Danish (K-D) glassware.
5.1.4.1 Snyder columns, 3 bulb, macro.
5.1.4.2 Snyder columns modified micro.
5.1.4.3 Evaporating flasks, 500 ml, 250 ml.
5.1.4.4 Receiver ampuls 5 and 10 ml, graduated, with spring
attachment.
5.1.4.5 Beakers, 100 ml.
5.1.5 Water or steam bath for Kuderna-Danish concentrations.
5.1.6 Chromatographic (drying) Column - Pyrex (400 mm x 20 mm ID)
without a fritted plate.
5.1.7 Separatory funnels, 500 ml with TFE stopcock.
5.1.8 Syringe, 100 ml, Pyrex, with long needle.
5.1.9 Graduated cylinder, 100 or 250 ml.
5.1.10 Vials, 2 ml, 4 ml, and 8 ml with TFE lined screw caps.
5.1.11 Sample bottles, 1,000 ml or 4,000 ml glass with TFE lined
screw caps.
5.1.12 Disposable pipettes, for transferring extracts.
5.1.13 Gel Permeation Chromatograph (GPC), Analytical Biochemical
Labs, Inc., GPC Autoprep 1002 or equivalent including:
5.1.13.1 Glass column 25 mm ID x 60-70 mm packed with 70 g of
Bio-Beads SX-3.
5.1.13.2 Chromatographic pump, operated at 5 ml/rain with
350-700 millibars (5-10 psi).
5.1.13.3 Injector with loop.
236
-------
5.1.13.4 Syringe filter holder, stainless steel and TFE,
Gelman 4310 or equivalent.
5.1.13.5 Spectrophotometric chromatographic detector, 254 nm,
with strip chart recorder (optional, for GPC cali-
bration) .
5.1.14 Chromatographic column, 200 mm x 20 mm ID, with solvent
reservoir (250 ml) and TFE stopcock.
5.1.15 Roller mill.
5.1.16 Bottles, 500 ml, brown glass.
5.2 For Identification and Quantitation
5.2.1 Gas chromatograph/mass spectrometer with data system, Finnigan
4000 or equivalent. The GC/MS interface should include a
glass jet separator and a direct capillary line. The computer
system should allow acquisition and storage of repetitive scan
data throughout the GC/MS runs. Computer software should be
available to allow searching of GC/MS data for display of ex-
tracted ion current profiles (EICPs) and integration of the
peaks. The GC injection port must be designed for on-column
injection when using packed columns and for splitless or on-
column injection when using capillary columns. GC columns re-
quired are:
5.2.1.1 0.9 to 1.8 m x 2 mm ID glass packed with 1% SP-1240-DA
on 100/120 mesh Supelcoport or equivalent.
5.2.1.2 15-30 m x 0.2 mm ID wall coated open tubular capil-
lary column coated with SE-54, or equivalent, pro-
viding at least 25,000 effective theoretical plates,
measured at C13, or a 1.8mx2mmID glass packed
with 3% SP-2250 on 100/120 mesh Supelcoport, or
equivalent.
5.2.2 Gas chromatograph/flame ionization detector with the same GC
columns as for the GC/MS system.
6. Reagents
6.1 For Extraction and Extract Cleanup
6.1.1 Dichloromethane, "Distilled in Glass" or
equivalent, stored in original containers and used as received.
6.1.2 Hexane, "Distilled in Glass" or equivalent, stored in original
containers and used as received.
6.1.3 Acetone, "Distilled in Glass" or equivalent, stored in original
containers and used as received.
237
-------
6.1.4 Hydrochloric acid solution (6N). Slowly add 100 ml of HC1
(12N) to 100 ml of reagent water.
6.1.5 Sodium hydroxide solution (6N). Dissolve 24 g NaOH in re-
agent water and dilute to 100 ml.
6.1.6 Sodium sulfate (Na2S04), Anhydrous, granular. Clean by over-
night Soxhlet extraction with dichloromethane, drying in an
oven at 110 to 160°C oven, and then heating to 650°C for 2
hr. Store in a glass jar tightly sealed with TFE-lined screw
cap.
6.1.7 Silica gel, 75/150 mesh. Clean silica gel for 16 h by
Soxhlet extraction with dichloromethane. Dry and activate
for 16 h at 160°C. Deactivate by adding 3% water (by weight)
and mixing for at least 4 h on a tumbler. Store at room tem-
perature in glass jars fitted with TFE lined screw caps.
6.1.8 Florisil 60/100 mesh, Floridan. Clean florisil for 16
hr by Soxhlet extraction with dichloromethane. Dry and acti-
vate for 16 h at 160°C. Deactivate by adding 1% water by
weight and mixing for at least 4 h on a tumbler. Store at
room temperature in glass jars fitted with TFE lined screw
caps.
6.1.9 Glass wool. Clean glass wool by thorough rinsing with hexane,
dried in a 100°C oven, and stored in a hexane rinsed glass
jar with TFE lined screw cap.
6.1.10 Boiling chips - silica or carborundum.
6.1.11 GPC Calibration Solutions
6.1.11.1 Corn oil, 200 mg/ml in dichloromethane.
6.1.11.2 Pentachlorophenol and di-n-octylphthalate,
4 mg/ml each in dichloromethane.
6.2 For Identification and Quantitation
6.2.1 Analytical Standards
6.2.1.1 Prepare stock solutions from the pure compounds by
dissolving 10 mg quantities into 10 ml of dichloro-
methane. If compound purity is certified at 96% or
greater, the weight can be used without correction
to calculate the concentration of the stock stan-
dard. Commercially prepared stock standards can be
used at any concentration if they are certified by
the manufacturer or by an independent source.
238
-------
6.2.1.2 Transfer the stock standard solutions into TFE
sealed screw cap bottles. Store at 4°C and protect
from light. Stock standard solutions should be
checked frequently for signs of degradation or evap-
oration, especially just prior to preparing calibra-
tion standards from them. Quality control check
standards are available from the U.S. Environmental
Protection Agency, Environmental Monitoring and Sup-
port Laboratory, Cincinnati, that can be used to de-
termine the accuracy of calibration standards.
6.2.1.3 Stock standard solutions must be replaced after
6 months, or sooner if comparison with check
standards indicate a problem.
6.2.1.4 Prepare mixed analytical standards by diluting ali-
quots of the stock solutions. The acids standard
should contain each of the phenolic compounds at
concentrations in the range of 50 to 200 ng/jjl.
The base, neutral, and pesticide (B/N/P) standard
should be prepared at concentrations in the range
of 20-100 ng/|Jl. All working standards must include
d10-anthracene at 20 ng/(Jl.
6.2.2 Analyte Spiking Solutions
Prepare mixed spiking solutions by serial dilution from the
individual stock solutions prepared as described in Section
6.2.1. The specifications for spiking solutions for indi-
vidual samples are described in Section 8.3.2.
6.2.3 Internal Standard Spiking Solution
Prepare djo-anthracene internal spiking solutions at 20 \Jg/
liter in dichloromethane as described in Section 6.2
6.2.4 Surrogate Standard Spiking Solution
Select a minimum of five B/N and two acid surrogate compounds
from Table 2. Prepare and store stock standard and spiking
solutions as described in Section 6.2.1.
6.2.5 DFTPP Standard
Prepare a 25 ng/pl solution of DFTPP in dichloromethane.
7. Method Calibration
7.1 Establish GC/MS operating parameters equivalent to those indicated
in Section 12. The GC/MS system can be calibrated using the external
standard technique or the internal standard technique.
239
-------
7.2
External Standard Calibration Procedure
7.2.1 Prepare calibration standards at a minimum of three concentra-
tion levels for each parameter of interest by adding volumes
of one or more stock standards to a volumetric flask and di-
luting to volume with dichloromethane. One of the external
standards should be at a concentration near, but above, the
method detection limit and the other concentrations should
correspond to the expected range of concentrations found in
real samples or should define the working range of the de-
tector.
7.2.2 Using injections of 1 to 2 pi of each calibration standard,
tabulate peak height or area responses against the mass in-
jected. The results can be used to prepare a calibration
curve for each compound. Alternatively, if the ratio of re-
sponse to amount injected in the calibration factor is a con-
stant over the working range (< 10% relative standard devia-
tion RSD), linearity through the origin can be assumed and the
average ratio or calibration factor can be used in place of a
calibration curve.
7.2.3 Verify the working calibration curve or calibration factor on
each working day by the measurement of one or more calibration
standards. If the response for any parameter varies from the
predicted response by more than ± 10%, the test must be re-
peated using a fresh calibration standard. Alternatively, a
new calibration curve or calibration factor must be prepared
for that compound.
7.3 Internal Standard Calibration Procedure. To use this approach, se-
lect one or more internal standards that are similar in analytical
behavior to the compounds of interest. The analyst must demonstrate
that the measurement of the internal standard is not affected by
method or matrix interferences. The internal standards used for
this procedure must include d10-anthracene.
7.3.1 Prepare calibration standards at a minimum of three concen-
tration levels for each parameter of interest by adding vol-
umes of one or more stock standards to a volumetric flask.
To each calibration standard, add a known constant amount of
one or more internal standards and dilute to volume with di-
chloromethane. One of the standards should be at a concen-
tration near, but above, the method detection limit, and the
other concentrations should correspond to the expected range
of concentrations found in real samples or should define the
working range of the detector.
240
-------
7.3.2 Using injections of 2 to 5 |Jl of each calibration standard,
tabulate peak height or area responses against concentration
for each compound and internal standard, and calculate rela-
tive response factors (RRF) for each compound using Equation 1.
EOT = (AsBis)/(A.sBs) (Eq. 1)
where Ag = Response for the parameter to be measured.
A^s = Response for the internal standard.
B. = Mass of the internal standard (ng).
IS
B = Mass of the parameter to be measured (ng).
s
If the RRF value over the working range is a constant (< 10%
RSD), the RRF can be assumed to be nonvariant and the average
RRF can be used for calculations. Alternatively, the results
can be used to plot a calibration curve of response ratios,
A /A. vs. RRF.
s is
7.3.3 Verify the working calibration curve or RRF on each working
day by the measurement of one or more calibration standards.
If the response for any parameter varies from the predicted
response by more than ± 10%, the test must be repeated using
a fresh calibration standard. Alternatively, a new calibra-
tion curve must be prepared for that compound.
Daily Calibration of the GC/MS System
7.4.1 At the beginning of each day, check the calibration of the
GC/MS system and adjust if necessary to meet decafluorotri-
phenylphosphine (DFTPP) specifications (Section 7.4.3). Each
day base/neutrals are measured, the column performance speci-
fication (Section 12) with benzidine must be met. Each day
the acids are measured, the column performance specification
(Section 12) with pentachlorophenol must be met. DFTPP can
be mixed in solution with either of these compounds to com-
plete two specifications with one injection, if desired.
7.4.2 To perform the calibration test of the GC/MS system, the fol-
lowing instrumental parameters are required.
Electron energy, 70 eV (nominal)
Mass range, m/e 40-475
Scan time to provide at least 5 scans per peak. Typical scan
times are 3 s or less for acids or base/neutrals with the
packed column, 1 s or less for base/neutrals with the capil-
lary column.
241
-------
7.4.3 Evaluate the system performance each day that it is to be
used for the analysis of samples or blanks by examining the
mass spectrum of DFTPP. Inject a solution containing 50 ng
DFTPP and check to ensure that performance criteria listed in
Table 3 are met. If the system performance criteria are not
met, the analyst must retune the spectrometer and repeat the
performance check. The performance criteria must be met be-
fore any samples or standards may be analyzed.
7.4.4 At the beginning of each working day, verify the calibration
of the GC/MS system by injecting a standard mixture (Section
7.3.3).
7.4.5 For instrument performance verification, to each acid and
base/neutral extract and each standard analyzed, add an
amount of djQ-anthracene immediately prior to injection such
that 40 ± 4 ng will be injected.
8. Quality Control
8.1 Demonstrate Acceptable Performance
Before performing any analyses, the analyst must demonstrate the
ability to generate acceptable accuracy and precision with this pro-
cedure.
8.1.1 For each compound to be measured, select a spike concentra-
tion representative of the expected levels in the samples.
Using stock standards, prepare a quality control check stan-
dard in acetone 1,000 times more concentrated than the
selected concentrations.
8.1.2 Add 80 yl of the check standard to each of a minimum of four
80 ml aliquots of reagent water with a syringe. A representa-
tive sludge can be used in place of the reagent water, but
one or more additional aliquots must be analyzed to determine
background levels, and the spike level must exceed twice the
background level for the test to be valid. Analyze the ali-
quots according to the method beginning in Section 10.
8.1.3 Calculate average recovery (R) and standard deviation (s), in
percentage recovery, for the results. Sludge background cor-
rections must be made before R and s calculations are performed.
8.1.4 Using the appropriate data from Tables 12-14, determine the
recovery and single operator precision expected for the method
for each parameter, and compare these results to the values
calculated in Section 8.1.3. If the data are not comparable,
the analyst must review potential problem areas and repeat
the test.
242
-------
8.2 Precision and Accuracy Statement. The analyst must calculate method
performance criteria for each of the surrogate standards.
8.2.1 Calculate upper and lower control limits for method performane
for each surrogate standard, using the values for R and s cal-
culated in Section 8.1.3:
Upper Control Limit (UCL) = R + 3 s
Lower Control Limit (LCL) = R - 3 s
The UCL and LCL can be used to construct control charts4
that are useful in observing trends in performance.
8.2.2 For each surrogate standard, the laboratory must develop and
maintain separate accuracy statements of laboratory perfor-
mance for wastewater samples. An accuracy statement for the
method is defined as R ± s. The accuracy statement should be
developed by the analysis of four aliquots of sludge as de-
scribed in Section 8.1.2, followed by the calculation of R
and s. Alternately, the analyst must use four sludge data
points gathered through the requirement for continuing qual-
ity control in Section 8.4. The accuracy statements should
be updated regularly.4
8.3 Surrogate Spikes. The laboratory is required to spike all of their
samples with the surrogate standard spiking solution to monitor spike
recoveries. Suggested surrogate compounds are listed in Table 5.
If the recovery for any surrogate standard does not fall within the
control limits for method performance, the results reported for that
sample must be qualified as described in Section 14.3. The labo-
ratory should monitor the frequency of suspect data to ensure that
it remains at or below 5%.
8.4 Method Blank. Before processing any samples, the analyst should
demonstrate through the analysis of an 80 ml aliquot of reagent water,
that all glassware and reagents are interference free. Each time a
set of samples is extracted or there is a change in reagents, a la-
boratory reagent blank should be processed as a safeguard against
laboratory contamination.
8.5 Fortified and Replicate Samples
8.5.1 Sample Selection. Select at least 30% of all samples or at
least one sample from each sampling location for fortified
and replicate aliquot analysis. Analyze the selected samples
in duplicate to determine unspiked analyte concentrations.
Using the procedures described in Section 8.5.2, spike and
analyze duplicate aliquots of the selected samples. Analyze
a third unspiked aliquot of the selected samples on the same
day that the corresponding spiked aliquots are analyzed.
24 3
-------
8.5.2 Spiking Procedures. Spike an 80 ml aliquot of sludge with
the representative semivolatile compounds listed in Table 4
and with other compounds identified in the sample, if avail-
able, at two times the observed concentration or at 10 times
the lower limit of detection, whichever is greater. Prepare
the spike as two acetone solutions with the acidic and neutral
compounds in the first solution and the basic compounds in the
second solution. The concentrations of the spiking solutions
should be such that 1 to 5 ml of each solution are added to
the sludge sample to attain the required spike concentration.
Homogenize the spiked sample for 45 to 60 s and store overnight
at 4°C with tumbling before extraction and analysis.
9. Sampling and Preservation
9.1 Sampling. Collect samples in glass containers (1,000 to 4,000 ml)
with a TFE lined screw cap. The container should be prewashed with
acetone and dried before use. Containers should be filled no more
than two-thirds full with sample to minimize breakage during freezing.
9.2 Preservation. Preferably, samples should be iced or refrigerated at
4°C for not more than 24 h before extraction. Where extraction can-
not be performed within 24 h, samples should be frozen. Samples
may be stored for up to 30 days at -20°C or indefinitely at -75°C.
In order to prevent breakage during storage, the containers should
not be slightly wanned and then recooled.
10. Sample Extraction
10.1 Preparation of Drying Column. Immediately prior to extracting a
sample, prepare a drying tube for the extract. Place a small
glass wool plug in the bottom of the column and add anhydrous so-
dium sulfate to a depth of 10 to 15 cm.
10.2 pH Adjustment. Thoroughly homogenize the sludge sample by mixing
with an emulsifier in the sample bottle for 1 min, then quickly
remove a 80 ml aliquot into a 250 ml graduated cylinder. Transfer
the aliquot into a 250 ml centrifuge bottle. Basify to pH £ 11
with a 6 N sodium hydroxide solution. Mix briefly with the
homogenizer to ensure uniform sample pH. (Note: If copious pre-
cipitation of carbonates is observed when sodium hydroxide is
added, make the sample slightly acidic with 6 N hydrochloric acid
and allow the carbon dioxide evolution to cease before basifying
the sample.)
10.3 Basic Extraction. Add 80 ml of dichloromethane to the centrifuge
bottle and homogenize for 45 to 60 s. Do not homogenize more
than 60 s to avoid heating the sample. Cap the centrifuge bot-
tles with TFE-lined screw caps and centrifuge at 2,500 to 3,000 rpm
for 30 min. Stir and repeat centrifugation if satisfactory phase
separation is not achieved. The mixture will separate into an
244
-------
aqueous layer over the dichloromethane extract with a solids cake
at the water-dichloromethane interface. Withdraw the extract from
each centrifuge bottle with a 100 ml syringe. Discharge the ex-
tracts into a 500 ml separatory funnel. Drain the dichloromethane
through the drying column into a Kuderna-Danish evaporator. Re-
tain any aqueous layer and return it to the centrifuge bottle.
Extract the sample two more times to achieve three-fold extraction.
Wash the drying column with an additional 100 ml of dichloromethane
and combine the eluent with the dried extracts.
10.4 Extract Concentration. Add a boiling chip to the extract in the
Kuderna-Danish evaporator and concentrate the extract to 8 ml using
an 85°C bath or a steam bath. If the extract is only slightly col-
ored and not notably viscous, fit a modified Snyder column onto
the Kuderna-Danish receiving tube and immerse the tube halfway in
a 35°C water bath. Direct a gentle stream of nitrogen directly on
the surface of the extract until the volume is ^ 4 ml. Transfer
the extract to a clean vial, seal it with TFE lined screw cap, and
store the extract at 4°C. If the extract is highly colored, vis-
cous or solidifies when concentrating to 4 ml, dilute the extract
to 12 ml with dichloromethane, transfer to a clean vial, and store
as described above.
10.5 Acidic Extraction. Acidify the sludge sample portions to pH ^ 2
with 6 N hydrochloric acid and extract the sample again by proce-
dures described in Sections 10.3 to 10.4. Discard the extracted
sludge aliquots.
Extract Cleanup
11.1 Base/Neutral Extracts. Clean the base/neutral sludge extracts by
adsorption chromatography on either florisil or silica gel. Al-
though there does not appear to be any significant difference in
the performance of these adsorbents, the adsorbent selected for
analyses of all quality assurance blanks, spiked blanks, and
spiked sludges associated with a specific sludge sample must be
the same as was selected for the original sample analysis. In
cases where the base/neutral extracts are to be analyzed by packed
column GC/MS, the GPC procedure (Section 11.2) may be used as an
alternative cleanup method.
11.1.1 Column Preparation. Prepare a 200 mm x 20 mm ID silica
gel or florisil column by pouring 20 g fresh 3% deactivated
silica gel or fresh 1% deactivated florisil, into a chro-
matographic column containing 60 to 70 ml hexane. Tap the
side of the clamped column with a stirring rod to aid pack-
ing and air bubble escape. If clumping should occur at
the bottom of the solvent reservoir, a stirring rod can be
used to break the clumps. Add hexane as needed. After
245
-------
most packing has settled, rinse down the florisil or sil-
ica gel adhering to the column walls with additional hex-
ane. Wash the column with an additional 20 ml hexane un-
til the solvent level is within 1 to 2 mm of the top of
the silica gel or florisil.
11.1.2 Column Calibration.
11.1.2.1 Check the elution pattern for each new batch of
adsorbent prepared by chromatographing a spiked
blank. Mix 1 ml of a dichloromethane solution
containing 100 |Jg of each of the specific analytes
of interest and the representative B/N/P compounds
included in quality assurance method spikes (de-
scribed in Section 8.3) with 2 g of adsorbant.
Evaporate the solvent with a gentle stream of dry
nitrogen and add the adsorbant to the column.
11.1.2.2 Elute the column with 20 ml of hexane (Fraction
I), 50 ml of 10% dichloromethane in hexane (Frac-
tion II), 50 ml of 50% dichloromethane in hexane
(Fraction III), and then 150 ml of 5% acetone in
dichloromethane (Fraction IV). Maintain the elu-
tion flow at 2 to 4 ml/min.
11.1.2.3 Collect each fraction in separate Kuderna-Danish
evaporators and concentrate the fractions accord-
ing to the procedures described in Section 7.4.
11.1.2.4 Analyze each fraction by GC/FID or GC/MS with the
SE-54 capillary or packed SP-2250 column. Deter-
mine the recovery of each compound in each frac-
tion by comparing the detector responses with those
from a duplicate aliquot of the original spiked
blank. Elution patterns and recoveries observed
for selected compounds chromatographed on silica
gel and florisil are shown in Tables 5 and 6.
11.1.2.5 Increasing the volume of Fraction I may be required
to provide acceptable cleanup for sludge extracts
containing high concentrations of aliphatic hydro-
carbons or other non-polar compounds. However,
additional hexane may elute significant portions
of the chlorinated benzenes, hexachloroethane,
hexachlorobutadiene, hexachlorocyclopentadiene,
2-chloronaphthalene, aldrin, and p,p'-DDE. The
elution patterns and recoveries observed for many
B/N/P compounds chromatographed on silica gel using
100 ml of hexane for Fraction I are shown in Table 7.
In cases where the volume of Fraction I must be
increased to achieve acceptable cleanup and the
analytes of interest include the compounds listed
246
-------
in this section, Fraction I must be analyzed sepa-
rately from Fraction II-IV.
11.1.3 Column Operation.
11.1.3.1 Place 2 g of fresh 1% deactivated florisil or 3%
deactivated silica gel into a small beaker and add
the concentrated base/neutral extract. Dry the
sample with a gentle nitrogen stream with stirring
to ensure fast and uniform drying. Load dried
florisil or silica gel containing the sludge com-
ponents into the column. Elute the column as de-
scribed in Section 11.1.2. Discard Fraction I.
11.1.3.2 Collect and composite in Kuderna-Danish evaporators
all fractions containing the compounds of interest
such as Fractions II through IV for all B/N/P com-
pounds. Fraction II frequently contains much higher
levels of interfering materials than to Fractions
III or IV. As a result, it should not be compos-
ited with Fractions III and IV if it contains no
compounds of interest. Fractions may be analyzed
separately if desired.
11.1.3.3 Concentrate all fractions collected to 1 ml ac-
cording to procedures described in Section 10.A.
11.2 Acidic Fraction. Gel permeation chromatography is used to remove
triglycerides and fatty acids.
11.2.1 GPC Column Preparation. Place 70 g of Bio-Beads SX-3 in a
400 ml beaker. Cover the beads with dichloromethane and
allow the beads to swell overnight before packing the col-
umns. Transfer the swelled beads to the column and start
pumping solvent, dichloromethane, through the column, with
upward flow at 5.0 ml/min. After ^ 1 h, adjust the pres-
sure on the column to 350 to 700 millibars (5 to 10 psi)
and pump an additional 4 h to remove air from the column.
Adjust the column pressure periodically as required to
maintain 350 to 700 millibars.
11.2.2 Column Calibration.
11.2.2.1 Calibrate the GPC column elution at least once a
week according to the following procedure. Load 5
ml of the corn oil solution into sample loop No. 1
and 5 ml of the phthalate-phenol solution into loop
No. 2. Inject the corn oil and collect 10 ml frac-
tions, for 36 min, changing the fraction at 2 min
intervals. Inject the phenols solution and col-
lect 10-ml fractions for 1 h.
247
-------
11.2.2.2 Determine the corn oil elution pattern by evapora-
tion of each fraction to dryness followed by a
gravimetric determination of the residue.
11.2.2.3 Analyze the phthalate-phenol fractions by GC/FID
on the SP-2250 and SP-1240-DA columns. Plot the
concentration of each component in each fraction
versus total eluent volume (or time) from the in-
jection points.
11.2.2.4 Choose a "dump time" which allows £ 85% removal of
the corn oil and S 85% recovery of the di-n-octyl-
phthalate. Choose the "collect time" to extend at
least 10 min after the elution of pentachlorophenol.
"Wash" the column 20 min between samples. Typical
parameters selected are: dump time, 20 min (100
ml); collect time, 30 min (150 ml); and wash time,
20 min (100 ml).
11.2.2.5 The column can also be calibrated using a 254-nm
UV detector to monitor the elution of the corn oil
and the phenols. Measure the peak areas at various
elution times to determine appropriate fractions.
11.2.3 Column Operation.
11.2.3.1 Prefilter or load all extracts via the filter
holder to avoid particulates that might cause flow
stoppage. Load the £ 4-ml extracts into the 5-ml
loops with a clean solvent on both sides of the
extract. Load the *v» 12-ml extracts in three con-
secutive loops. Use sufficient clean solvent after
the extract to transfer the entire aliquot into
the loop. Between extracts, purge the sample load-
ing tubing thoroughly with clean solvent. After
especially dirty extracts, run a GPC blank using
dichloromethane to check for carry-over. Process
the extracts using the dump, collect, and wash
parameters determined from the calibration and col-
lect the cleaned extracts in 250 ml brown bottles.
11.2.3.2 Concentrate the cleaned extracts, combining col-
lected fractions from multiple injections, to * 8
ml using Kuderna-Danish evaporators and then to S
1 ml using modified Snyder columns and nitrogen
blowdown.
11.2.3.3 Transfer the cleaned extracts to 8 ml graduated
tubes and dilute to 4 ml with dichloromethane.
Store at 4°C for GC/MS analysis. Intensely colored
extracts may require a second GPC cleanup.
248
-------
11.2.3.4 Note: Sample extracts must be in the same solvent
used for column elution. Injection of solutions
in other solvents can cause a significant change
in the bead swell. In extreme cases a traumatic
pressure increase may damage the column and cap,
or even break the glass column.
Sample Extract Analysis
12.1 Base/Neutral/Pesticide Extracts. Analyze the B/N/P extracts by
GC/MS using the capillary column or packed column systems described
in Section 5.2.1 under the conditions shown below. The capillary
column GC/MS procedure is preferred because this procedure provides
more unambiguous interpretation of the resulting data. However,
if satisfactory performance of the column cannot be demonstrated
or if the required Grob-type injection system with direct capillary
GC/MS interface is not available, the packed column GC/MS procedure
may be used.
12.1.1 Capillary GC/MS with a SE-54 WCOT Column
12.1.1.1 Column Temperature: 50°C for 4 min, 50 to 320°C
at 4°C/min, and 320°C until after the elution
time for benzo[g,h,i]perylene.
GC/MS interface, 300°C
Carrier gas helium at 10 psi
Splitless injection for 30 s
Injection size, 2 yl
Examples of the separations achieved by this
column are shown in Figure 2. Relative reten-
tion times for the base/neutral compounds on
this column are listed in Table 8.
12.1.1.2 Capillary Column GC/MS Performance Checks
12.1.1.2.1 Column Performance Screening. Fully evaluate
the performance of each new column by GC/FID
prior to installation in the GC/MS system.5
In addition, recheck columns that show a
marked decrease in performance on the GC/MS
(as described in Section 12.2.1.3) or that
have not been used the previous 2 weeks for
GC/MS analyses under this protocol. Install
the capillary column in a gas chromatograph
equipped with a Grob-type split/splitless in-
jector and flame ionization detector. Adjust
carrier gas at about 15 to 20 cm/s for N2
249
-------
or 30 to 40 cm/s when He or hydrogen are
used as the carrier gas. Set the splitter
flow at 30 to 50 ml/min, makeup gas at 30
ml/min, and column oven temperature at 85°C.
Using splitless injection, inject 2 Ml of a
column performance test mixture containing 20
ng each of 2,6-dimethylphenol, 2,6-dimethyl-
aniline, 1-octanol, 2-octanone, n-tridecane,
and n-tetradecane or an equivalent commercially
available test mixture. Table 9 shows the
contents of three appropriate commercial mix-
tures. Measure the pH of the column as the
ratio of the peak height of 2,6-dimethylan-
iline to that of 2,6-dimethylphenol. A value
of 0.5 to 1.5 is acceptable. Determine the
activity of the column toward polar compounds
by determining both the asymmetry for the
1-octanol peak and the ratio of the peak height
of 1-octanol to that of C13 alkane. The peak
asymmetry is evaluated by drawing a perpendicu-
lar from the apex of the peak to the baseline
and measuring the width from the front of the
peak to the perpendicular line (W^) and from
the back of the peak to the perpendicular line
(wB).
AS = x 100
WF
Peak asymmetries between 75 and 200 are ac-
ceptable. The ratio of peak height of the
1-octanol to that of C13 alkane should be
higher than 0.5. Determine the number of ef-
fective theoretical plates (N from the
tridecane peak using the equation
tr' ^
"eft*5-545
where tr' is the adjusted retention time of
tridecane and is the peak width at half
height. The number of effective plates of
acceptable columns must be at least 25,000.
12.1.1.2.2 System Performance. Evaluate the performance
of the capillary column in the GC/MS system
each time the column is installed and each
working day prior to its use. Using split-
less injection, inject 2 |Jl of a test mixture
containing 20 ng each of 2,6-dimethylaniline,
2,6-dimethylphenol, methylstearate, octadecene,
250
-------
and octadecane. Measure the pH of the column
by taking the ratio of peak heights of 2,6-
dimethylaniline to that of 2,6-dimethylphenol
A pH of 0.5 to 1.5 is acceptable. Determine
the resolution of the peaks corresponding to
octadecane and octadecene. The peaks should
be resolved with a 50% valley or better.
Determine the asymmetry of the methylstearate
peak according to the formula given in Sec-
tion 12.1.1.2.a. If the methylstearate peak
tails, the GC/MS transfer line is not ade-
quately heated.
12.1.2 Packed Column GC/MS with a SP-2250 Column
Analyze B/N extracts by GC/MS using the SP-2250 column
described in Section 5.2.1, operated under the following
conditions.
Column temperature, 60°C for 2 rain, 60 to 260°C at 80°C/
min, and 260°C until after the elution time for benzo-
[g,h,i]perylene.
Injector temperature, 225°C
GC/MS interface temperature, 275°C
Carrier gas, helium at 30 ml/min
Injection size, 2 (Jl
An example of the separation achieved by the column is
shown in Figure 3. Relative retention times for the base/
neutral and pesticide compounds on this column are listed
in Table 10.
12.2 Acid Extracts
Analyze acid extracts by GC/MS using the SP-1240-DA column de-
scribed in Section 5.2.1, operated under the following conditions.
Column temperature, 85°C for 2 min, 85 to 185°C at 10°C/
min and 185°C until after the elution time for 4-nitro-
phenol.
Injector temperature, 185°C.
GC/MS surface temperature, 275°C.
Carrier gas, helium at 30 ml/min.
Injection size, 2 pi.
An example of the separation achieved by the column is
shown in Figure 4. Relative retention times for the acid
compounds are listed in Table 11.
251
-------
13. Qualitative and Quantitative Determination
13.1 Using the characteristic mass spectral ions listed in Tables 8 or
10 for the base/neutral and pesticide compounds and in Table 11 for
the acid compounds, plot at least three extracted ion current plots
(EICPs) for each analyte.
13.2 Identify the presence of analytes by the coincidence of peaks in
the characteristic EICPs at the appropriate retention times and
with intensities in the characteristic ratios (Tables 8 or 10 and
11).
13.3 Record the area (intensity) of the peak in the EICP for the most
intense ion for each compound identified.
14. Calculations
14.1 Determine the concentration of individual compounds in the sample.
14.1.1 If the external standard calibration procedure is used,
calculate the amount of material injected from the peak
response using the calibration curve or calibration factor
in Section 7.2.2. The concentration in the sample can be
calculated from Equation 2:
(M)(V )
Concentration, |jg/liter = (Eq. 2)
i s
where: M = Mass of material injected (ng).
= Volume of extract injected (jjI).
Vg = Volume of total extract (ml).
Vg = Volume of wet sludge extracted (liter)
14.1.2 If the internal standard calibration procedure was used,
calculate the concentration in the sample using Equation 3.
(A)(Is)(Ve)
Concentration, |jg/liter = -tt—)(rrj)(v )(V ) ^Eq"
is i s
where: A = Area of peak in sample extract.
A. - Area of internal standard peak in sample
extract.
252
-------
I = Areas of internal standard injected (ng).
Vg = Total volume of extract (ml).
= Volume of extract injected (|Jl)-
V = Volume of sludge extracted (liter).
s
RRF = Relative response factor.
14.2 Report results in micrograms per liter without correction for re-
covery data. When duplicate and spiked samples are analyzed, re-
port all data obtained with the sample results.
14.3 If the surrogate standard recoverues fall outside the limits in
Section 8.2, data for all parameters in that sample must be
labeled as suspect.
15. Method Performance
Performance data for the application of this method to both POTW and
industrial sludges are shown in Tables 12 to 14. Table 12 shows data
for B/N compounds analyzed in POTW and industrial sludges via Option A,
i.e., extract cleanup by silica gel chromatography and analysis by
HRGC/MS. Table 13 shows data for B/N compounds analyzed via Option B,
i.e., extract cleanup by GPC and analysis by packed column GC/MS, and
acidic compounds in aliquots of the same sludges. Data for B/N compound
were analyzed via Option B and acidic compounds in sludges from 40 POTWs.
Most of these data represent results from one laboratory. Approximately
25% of the data in Table 14 was contributed by a second laboratory.
16. References
1. "Carcinogens - Working With Carcinogens," Department of Health,
Education, and Welfare, Public Health Service, Center for Disease
Control, National Institute for Occupational Safety and Health,
Publication No. 77-206, August 1977.
2. "OSHA Safety and Health Standards, General Industry" (29CFR1910),
Occupational Safety and Health Administration, OSHA 2206 (Revised
January 1979).
3. "Safety in Academic Chemistry Laboratories," American Chemical Society
Committee on Chemical Safety, 3rd Edition Publication (1979).
4. "Handbook of Analytical Quality Control in Water and Wastewater
Laboratories," EPA-600/4-79-019, U.S. Environmental Monitoring
and Support Laboratory, Cincinnati, Ohio 45268, March 1979.
5. Grob, K., Jr., G. Grob, and K. Grob, "Comprehensive Standardized
Quality Test for Glass Capillary Columns," J. Chrom., 156, 1-20
(1978).
253
-------
Additional Sources
1. "Sampling and Analysis Procedures for Screening of Industrial
Effluents for Priority Pollutants." U.S. Environmental Protection
Agency, Environmental Monitoring and Support Laboratory, Cincinnati,
OH 45268, March 1977, Revised April 1977. Effluent Guidelines
Division, Washington, DC 20460.
2. "Methods 330.4 (Titrimetric, DPD-FAS) and 330.5 (Spectrophotometric,
DPD) for Chlorine, Total Residual," Methods for Chemical Analysis
of Water and Wastes, EPA 600/4-79-020, U.S. Environmental Protection
Agency, Environmental Monitoring and Support Laboratory - Cincinnati,
OH 45268. In preparation.
3. "Preservation and Maximum Holding Time for the Priority Pollutants,"
U.S. Environmental Protection Agency, Environmental Monitoring and
Support Laboratory, Cincinnati, OH 45268. In preparation.
4. Budde, W. L., and J. W. Eichelberger, "Performance Tests for the
Evaluation of Computerized Gas Chromatography/Mass Spectrometry
Equipment and Laboratories," EPA-600/4-80-025, U.S. Environmental
Protection Agency, Environmental Monitoring and Support Laboratory,
Cincinnati, OH 45268, p. 16, April 1980.
5. Kleopfer, R. D., "Priority Pollutant Methodology Quality Assurance
Review," U.S. Environmental Protection Agency, Region VII,
Kansas City, KS. Seminar for Analytical Methods for Priority
Pollutants, Norfolk, VA, January 17-18, 1980, U.S. Environmental
Protection Agency, Office of Water Programs, Effluent Guidelines
Division, Washington, DC 20460.
254
-------
TABLE 1. EXTRACTABLE ORGANIC PRIORITY POLLUTANTS
Compound CAS No.
Acids
2-Nitrophenol
88-75-5
4-Nitrophenol
100-02-7
Pentachlorophenol
87-86-5
Phenol
108-95-2
2,4,5-Trichlorophenol
88-06-2
4-Chloro-3-methylphenol
59-50-7
2-Chlorophenol
95-57-8
2,4-Dichlorophenol
120-83-2
2,4-Dimethylphenol
105-67-9
4,6-Dinitro-2-methylphenol
534-32-1
Bases
Benzidine
93-87-5
3,3'-Dichlorobenzidine
92-94-1
Polycyclic Aromatic Hydrocarbons
Acenaphthene
83-32-9
Acenaphthylene
208-96-8
Anthracene
120-12-7
Benz[a]anthra cene
56-55-3
Benzolbjfluoranthene
205-99-2
Benzo[k]fluoranthene
207-08-9
Benzo[g,h,i]perylene
191-24-2
Benzo[ajpyrene
50-32-8
Chrysene
218-01-9
Dibenz[a,h]anthra cene
53-70-3
Fluoranthene
206-44-0
Fluorene
86-73-7
Indeno[1,2,3-cd]pyrene
193-39-5
Naphthalene
91-20-3
Phenanthrene
85-01-8
Pyrene
129-00-0
Phthalates
Bis(2-ethylhexyl)phthalate
117-81-7
Butylbenzylphthalate
85-68-7
Diethylphthalate
84-66-2
Dime thylphthalate
131-11-3
Di-n-butylphthalate
84-74-2
Di-n-octylphthalate
117-84-0
(continued)
255
-------
TABLE 1 (continued)
Compound CAS No.
Chlorinated Hydrocarbons
2-Chloronaphthalene 91-58-7
1.2-Dichlorobenzene 95-50-1
1.3-Dichlorobenzene 541-73-1
1.4-Dichlorobenzene 106-46-7
Hexachlorobenzene 118-74-1
Hexachloro-l,3-butadiene 87-68-3
Hexachloroethane 67-72-1
Hexachlorocyclopentadiene 77-47-4
1,2,4-Trichlorobenzene 129-82-1
Chloroalkyl Ethers
Bis(2-chloroethyl)ether 111-44-4
Bis(2-chloroethoxy)methane 111-91-1
Bis(2-chloroisopropyl)ether 39638-32-9
Miscellaneous Neutrals
4-Bromophenyl phenyl ether 101-55-3
4-Chlorophenyl phenyl ether 7005-72-3
2,4-Dinitrotoluene 121-14-2
N-Nitrosodi-n-propylamine 621-64-7
N-Nitrosodimethylamine 62-75-9
2,6-Dinitrotoluene 606-20-2
Isophorone 78-59-1
Nitrobenzene 98-95-3
N-Nitrosodiphenylamine 86-30-6
1,2-Diphenylhydrazine 122-66-7
Pesticides
p-Endosulfan 33213-65-9
a-BHC 319-84-6
Y-BHC 58-89-9
|3-BHC 319-85-7
Aldrin 309-00-2
Heptachlor 76-44-8
Heptachlor epoxide 1024-57-3
a-Endosulfan 959-98-8
Dieldrin 60-57-1
4,4'-DDE 72-55-9
4,4'-DDD 72-54-8
4,4'-DDT 50-29-3
(continued)
256
-------
TABLE 1 (concluded)
Compound CAS No.
Endrin 72-20-8
Endosulfan sulfate 1031-07-8
6-BHC 319-86-8
Chlordane 57-74-9
Toxaphene 8001-35-2
PCB-1242 53469-21-9
PCB-1221 11104-28-2
PCB-1254 11097-69-1
PCB-1232 11141-16-5
PCB-1248 12672-29-6
PCB-1260 11096-82-5
PCB-1016 12674-11-2
257
-------
TABLE 2. SUGGESTED SURROGATE COMPOUNDS
Base/neutral fraction Acid fraction
Aniline-ds
Anthracene-dxo
Benzo[a]anthracene-di2
4,4'-Dibromobiphenyl
4,4'-Dibromooctafluorobiphenyl
Decafluorobiphenyl
2,2'-Difluorobiphenyl
2-Fluoraniline
1-Fluoronaphthylene
2-Fluoronaphthylene
Naphthalene-dg
Nitrobenzene-d5
1,2,3,4,5-Pen.taf luorobiphenyl
Pyridine-d5
2-Fluorophenol
Pentafluo ropheno1
Phenol-d5
2-Perfluoromethyl phenol
258
-------
TABLE 3. DFTPP KEY IONS AND ION ABUNDANCE CRITERIA
m/e Ion abundance criteria
51 30-60% of m/e 198
68 < 2% of m/e 69
70 < 2% of m/e 69
127 40-60% of 198
197 < 1% of mass 198
198 Base peak, 100% relative abundance
199 5-9% of ra/e 198
275 10-30% of m/e 198
365 1% of m/e 198
441 Present and < m/e 443
442 40% of m/e 198
44 3 17-23% of m/e 442
a Eichelberger, J. W., L. E. Harris, and W. L.
Budde, "Reference Compound to Calibrate Ion
Abundance Measurement in Gas Chromatography-
Mass Spectrometry," Analytical Chemistry, 47,
995 (1975).
TABLE 4. REPRESENTATIVE EXTRACTABLE COMPOUNDS
FOR RECOVERY STUDIES
Acenaphthylene
Benzidine
Benzo[a]pyrene
Bis(2-chloroethyl) ether
Bis(2-chloroisopropyl) ether
B i s ( 2 - e thylhexy1)phthala te
Butylbenzyl phthalate
3,3'-Dichlorobenzidine
2,4-Dichlorophenol
2,4-Dimethylphenol
2,6-Dinitrotoluene
Fluoranthene
Hexachloroethane
N-Nitrosodimethylamine
1,4-Dichlorobenzene
Pentachlorophenol
Phenol
Dieldrin
a-BHC
£,£*-DDE
Heptachlor
259
-------
TABI.E 5. ELUT10N 1'ATTF.RHS AND RECOVERIES OF SELECTED BASE/HF.UTRAI.S CHROHATOCWAPI1EP OH SILICA CF.la
Fraction II Fraction III Fraction IV
Compound
Spike
level
litlialate
39.6
+
93
B»nzo| a Ipyrrne
30.0
*
78
a 3% deactivated silica gel.
h Analyses done liy packed colimn GC/MS.
c + indicates ¦ore than IX recovery in that fraction.
-------
TABI.E 6. ELUT10H PATTERNS ANN REUOVERl fcS OF SKI .FXTEI) BASE/NEUTRALS CUKOHATOCRAPIIEP ON FLORISILa
Fraction II Fraction 111 Fraction IV
Compound
Spike
level
(h«L
Fraction I
hexane
<2«Lull
10% dicliloroBH'thane/
liexane
(50jrt)
50% diclilo rone thane/
hexane
(50 ml)
5% acetone/
dichloronethane
(150 ml)
Total ^
Recovery
(X)
1,4-l)ichlorol>enzene
93.6
,c
~
+
48
liexacli 1 oroet liane
62.A
~
-t
~
67
Bis(2-chloroi sopropyl )etlier
42.2
+
•f
150
Bis(2-cliloroetliyl)ether
66.2
+
+
68
Aceiiapht tiy 1 ene
57.0
+
4-
92
2,6-Oinitrotoluene
70.2
¦f
130
Flnoranthrne
54.0
*
~
~
120
3,3'-Dichlorohenzidine
72.8
\
65
n-Riit yllionzy Iplit lialate
90.4
4
110
Ris(2-etliy Ihr-xyl )phthalate
79.2
¦f
92
Bcnzo|aIpyrene
60.0
¦f
87
a 1% doactivated florisil.
b Analyses done liy packed column GC/MS.
c » indicates More than IX recovery in th.it (raclion.
-------
TABI.F. 7. F.UITION PATTERNS AND RECOVERIES OF SELECTED BASE/NEUTRM./PF.STICIDES C1IROMATOGRAPHED ON SILICA GEI.
WITH AN Al.TERNATE EI.UTION*' SCIIEHE
i\>
PO
CoHpoiind _ _
Bis(2-chloroethyl )etlier
1,3-Dichlorobeiizene
1,2-Dichlorohenzene
llexachloroet hanc
N-ni t rosoil j -n-propylaaine
Nit robenzene
Ris(2-chloroethoxy)aiethane
1,2,4-Triclilorobeiizeiie
Naphthalene
llexacli] orolmt adiene
llexacli 1 ororycl opentad i ene
2-I'll I oronaplit lia I ene
Acenaplilliylene
2,6-Oinil rotoluetic
Ac cnaphl lia I cue
2,4 -I) j ii i t ro t o 1 iiene
Fliiorene
4 -Chi oropli<-ny I pliciiy I el her
Diethylplitlialate
4- Brnnoplieny I ptieny I ellier
llcxarlilorohciizciie
Phriiaiif Incite
Ant hi acetic
Fraction I
Itcxane
(100 ml)
Fraction II
I OX tlichloronethane/
hexane
(50 ml)
Fraction III
50% dichloronethnne/
hexane
(50 nl)
Fraction IV
5% acetone/
dichloromethane
(150 nl)
Total
Recovery
(%)
95 + 22
64 i 6
67 + 9
73 + 14
94 J 12
67 i 3
66 ± 5
70+17
58 + 7
140 + 51
300 + 55
94 + 5
77 + 21
93 ± 19
88 + 12
95 ± 14
87 i 22
110 ± 12
90 ± 41
120 ± 15
85 i 30
85 ± 30
83 ± 28
(cont imied)
-------
ro
CT>
CO
(¦«nt|K)uil(t
I)i-n-liut y Iplil ha late
Fliioranthcne
f'y rone
Eudosulfan sulfate
n-Hiityllieiizylphthalate
Chrysene
3,3' -Dirhlorol>enz idine
Bi s(2-etliylhexyl )plithalate
Ren?.o|a (pyrene
Dilienz|a ,li|anlhraceiie
Y-I1IIC
Aldrin
lleptaclilorepox ide
p,|i' -DDE
F.ndrin
p,p'-1)1)1)
p,p'-l)l)T
ff-Kiidosul (an
Fraction 1
liexane
(100 ml)
__ TABIE 7 (continued)^
Fraction II
10% didil oroiuetliane/
liexane
(50 ml)
Fraction III
50H dichlorometli.-ine/
liexane
(50 ml)
Fraction IV
acetone/
dichloromethane
(150 nl)
Total
Recovery
__tt)
80 ± 51
51 ~ 20
63 ± 12
67 ± 17
87 ± 35
72 + 5
4 0+2
95 t 61
70 i 34
84 1 17
66 ± 27
67 i 14
88 ± 3
80 1 6
77 + 18
78 ± 7
91 i 17
86 i 17
a Data from l.opez-Avi la, Viorica, Raymond V. Northcult, Joii Onslot, ami Margie Wickliam. "Analysis of the NBS Sediment
liy the MHI !e Protocol," Final Report prepared for the F.uviroiimeutal Protection Agency under KPA Contract
No. 68-03-2711 (1981).
I> 3% react ivatcd silica gel.
c Recovery's determined in duplicate at 8-|ig and RO-pg levels lor each compound.
Analyses done by fused silica capill.ny liC/HS.
d I indicates more than 1% recovery in that fraction.
-------
TABLE 8. CHROMATOGRAPHIC CONDITIONS AND CHARACTERISTIC EI IONS FOR
THE BASE/NEUTRAL COMPOUND ANALYZED BY CAPILLARY COLUMN GC/MS
a k Characteristic EI ions
Compound RRT ' (relative intensity)
N-Nitrosodimethylamine
0
.04
42 (
00),
74(88),
44(21)
Bis(2-chloroethyl)ether
0
.21
93 (
00),
63(99),
95(31)
1,3-Dichlorobenzene
0
.22
146
100),
148(64)
, 113(12)
1,4-Dichlorobenzene
0
.22
146
100),
148(64)
, 113(11)
1,2-Dichlorobenzene
0
.25
146
100),
148(64)
, 113 (11)
Bis(2-chloroisopropyl)ether
0
.29
45 (
00),
77(19),
79(12)
Hexachloroethane
0
.29
117
100,
199(61),
201(99)
N-Nitrosodi-n-propylamine
0
.31
130
22),
42(64),
101(12)
Nitrobenzene
0
.32
77 (
00),
123(50),
65(15)
Isophorone
0
.36
82(
00),
95(14),
138(18)
Bis(2-chloroethoxy)methane
0
41
93(
00),
95(32),
123(21)
1,2,4-Trichlorobenzene
0
42
74(
00),
109(80),
145(52)
Naphthalene
0
43
128
100)
223(63)
, 227(65)
Hexachlorobutadiene
0
47
225
100)
223(63)
, 227(65)
Hexachlorocyclopentadiene
0
60
237
100)
235(63)
, 272(12)
2-Chloronaphthalene
0
63
162
100)
164(32)
, 127(31)
Acenaphthylene
0
70
152
100)
153(16)
, 151(17)
Dimethylphthalate
0
72
163
100)
164(10)
, 194(11)
2,6-Dinitrotoluene
0
72
165
100)
63(72),
121(23)
Acenaphthene
0
73
154
100)
155(95)
, 152(13)
2,4-Dinitrotoluene
0
78
165
100)
63(72),
121(23)
Fluorene
0
82
166
100)
165(80)
, 167(14)
4-Chlorophenyl phenyl ether
0
84
204
100)
206(34)
, 141(29)
Diethylphthalate
0
84
149
100)
178(25)
, 150(10)
N-Nitrosodiphenylamingc
1,2-Diphenylhydrazine
0
84
169
100)
168(71)
, 167(50)
0
86
77(
00),
93(58),
105(28)
4-Bromophenyl phenyl ether
0
92
248
100)
250(99)
, 141(45)
Hexachlorobenzene
0
93
284
100)
142(30)
, 249(24)
Phenanthrene
0
98
178
100)
179(16)
, 176(15)
Anthracene
0
99
178
100)
179(16)
, 176(15)
djo"Anthracene
1
00
188
100)
94(19),
80(18)
Di-n-butylphthalate
1
14
149
100)
150(27)
, 104(10)
Fluoranthene
1
19
202
100)
101(23)
, 100(14)
Pyrene
1
23
202
100)
101(26)
, 100(17)
Benzidene
1
24
184
100)
92(24),
185(13)
Butylbenzylphthalate
1
40
149
100)
91(50)
Chrysene
1.
45
228
100)
229(19)
, 226(23)
3,3'-Dichlorobenzidene
1
47
252
100)
254(66)
, 126(16)
Bis(2-ethylhexyl)phthalate
1.
52
149
100)
167(31)
, 279(26)
Benzo[k]fluoranthene
1.
62
252
100)
253(23)
, 125(16)
(continued)
264
-------
TABLE 8 (continued)
Compound
Characteristic EI ions
(relative intensity)
Di-n-octylphthalate
Benzo[a]pyrene
Dibenz [a,h]anthracene
Benzo[g,h,i]perylene
1.63
1.67
1.84
1.87
149(100), 167(29), 279(22)
252(100), 253(23), 125(21)
278(100), 139(31), 279(12)
276(100), 138(37), 277(25)
a Relative to d10-anthracene. Retention times data were not determined
for the priority pollutant pesticides and a few base/neutral compounds.
However, their retention properties should be roughly equivalent to
those on SP-2250 (see Table 6).
b SE-54 WCOT glass capillary (15 m x 0.24 mm ID), He at 10 psi, program:
50°C for 4 min, then 4°C/min to 320°C.
c Elutes as diphenylamine.
d Elutes as azobenzene.
265
-------
TABLE 9. COMPOSITION OF CAPILLARY COLUMN PERFORMANCE
TEST MIXTURES
Test mixture
Composition
Varian
P/N 82-005049-01
(nonpolar)
Alltech
TP-5 (polar)
Analabs
Test probe LPK-013F
(general purpose)
2-octanone
1-octanol
naphthalene
2,6-dimethylphenol
2,4-dimethylaniline
C12-alkane
C13-alkane
C13-alkane
C14-alkane
C15-alkane
C16-alkane
1-octanol
5-nonanone
2,6-dimethylaniline
2,6-dimethylphenol
naphthalene
C12-acid methyl ester
Cn-acid methyl ester
CiQ-acid methyl ester
C10-alkane
Cxj-alkane
1-octanol
Nonanal
2,3-butanediol
2,6-dimethylphenol
2,6-dimethylaniline
dicyclohexylamine
2-ethylhexanoic acid
0.2 |Jg/Ml
0.2
0.2
0.2
0.2
0.2
0.2
0.1 Mg/Pl
0.1
0.1
0.1
0.5
0.3
0.4
0.4
0.5
41 ng/jJl
41
42
28
29
36
40
53
32
32
31
38
266
-------
TABLE 10. CHROMATOGRAPHIC CONDITIONS AND CHARACTERISTIC IONS
FOR THE BASE/NEUTRAL AND PESTICIDE COMPOUNDS ANALYZED
BY PACKED COLUMN GC/MS
„ u
Characteristic EI ions
Compound
RRT '
(relative intensity)
N-Nitrosodimethylamine
0.15
42(100),
74(88), 44(21)
1,3-Dichlorobenzene
0.31
146(100),
148(64), 113(12)
1,4-Dichlorobenzene
0.33
146(100),
148(64), 113(11)
Hexachloroethane
0.35
117(100),
199(61), 201(99)
1,2-Dichlorobenzene
0.35
146(100),
140(64), 113(11)
Bis(2-cbloroisopropyl)ether
0.37
45(100),
77(19), 79(12)
N-Nitrosodi-n-propylamine
0.42
130(22), 42(64), 101(12)
Nitrobenzene
0.45
77(100),
123(50), 65(15)
Isophorone
0.47
82(100),
95(14), 138(18)
Hexachlorobutadiene
0.48
225(100),
223(63), 227(65)
1,2,4-Trichlorobenzene
0.49
74(100),
109(80), 145(52)
Bis(2-chloroethoxy)methane
0.50
93(100),
95(32), 123(21)
Naphthalene
0.51
128(100),
127(10), 129(11)
Bis(2-chloroethyl)ether
0.55
93(100),
63(99), 95(31)
Hexachlorocyclopentadiene
0.60
237(100)
235(63), 272(12)
2-Chloronaphthalene
0.68
162(100)
164(32), 127(31)
Acenaphthylene
0.75
152(100)
153(16), 151(17)
Acenaphthene
0.77
154(100)
153(95), 152(53)
Dimethyl phthalate
0.78
163(100)
164(10), 194(11)
2,6-Dinitrotoluene
0.81
165(100)
63(72), 121(23)
Fluorene
0.85
166(100)
165(80), 167(14)
2,4-Dinitrotoluene
4-Chlorophenyl phenyl ether
0.85
165(100)
63(72), 121(23)
0.85
204(100)
206(34), 141(29)
Diethyl phthalate ^
0.87
149(100)
178(25), 150(10)
1,2-Diphenylhydrazine
N-Nitrosodiphenylamine
0.88
77(100),
93(58), 105(28)
0.89
169(100)
168(71), 167(50)
Hexachlorobenzene
0.92
284(100)
142(30), 249(24)
4-Bromophenyl phenyl ether
0.92
248(100)
250(99), 141(45)
Phenanthrene
0.99
178(100)
179(16), 176(15)
Anthracene
0.99
178(100)
179(16), 176(15)
Deuterated anthracene (D-10)
1.00
188(100)
94(19), 80(18)
D-n-butylphthalate
1.09
149(100)
150(27), 104(10)
Fluoranthene
1.18
202(100)
101(23), 100(14)
Pyrene
1.22
202(100)
101(26), 100(17)
Benzidine
1.27
184(100)
92(24), 185(13)
2,3,7,8-Tetrachlorodibenzo-
1.33
322(100)
320(90), 59(95)
£-dioxin
Butylb enzylphthalate
1.34
149(100)
91(50)
Bis(2-ethylhexyl)phthalate
1.37
149(100)
167(31), 279(26)
Chrysene
1.40
228(100)
229(19), 226(23)
(continued)
267
-------
TABLE 10 (continued)
Compound
RRT
a ,b
Characteristic EI ions
(relative intensity)
Benz[a]anthracene
Benzo[b]fluoranthene
Benzo[k]fluoranthene
3,3'-Dichlorobenzidine
Di-n-octylphthalate
Benzo[a]pyrene
Indeno[1,2,3-cd ]pyrene
Benzo [£,h,i]perylene
Bis(chloromethyl) ether
p-Endosulfan
ot-BHC
y-BHC
g-BHC
6-BHC
Aldrin
Heptachlor
Heptachlor epoxide
a-Endosulfan
Dieldrin
4,4*-DDE
4,4'-DDD
4,4'-DDT
Endrin
Endosulfan sulfate
Chlordane
Toxaphene
PCB-1242
PCB-1254
1.40
1.43
1.43
1.45
1.50
1.50
1.86
1.98
d
0.47
0.94
1.00
1.03
1.04
1.05
1.06
1.13
1.14
1.18
1.20
1.22
1.27
1.30
1.30
1.05-1.26
1.12-1.35
0.86-1.14
1.09-1.30
228(100)
252(100)
252(100)
252(100)
149(100)
252(100)
276(100)
276(100)
45(100),
201(100)
183(100)
183(100)
181(100)
183(100)
181(100)
183(100)
66(100),
100(100)
201(100)
79(100),
246(100)
235(100)
81(100),
272(100)
373(19),
(231, 233
(224, 260
(294, 330
229(19)
253(23)
253(23)
254(66)
167, 279
253(23)
138(28)
138(37)
49(14),
283(48)
109(86)
109(86)
183(93)
109(86)
183(93)
109(86)
220(11),
353(79)
283(48)
263(28),
248(64)
237(76)
82(61),
226(19)
125(15)
125(16)
126(16)
125(21)
277(27)
277(25)
51(5)
278(30)
181(91)
181(91)
109(62)
181(90)
109(62)
181(90)
263(73)
351(60)
278(30)
279(22)
176(65)
165(93)
263(70)
387(75), 422(25)
375(171, 377(10)
235 n
294)*
362)
a 3% SP-2250 on 100/120 mesh Supelcoport in a 1.8 ra x 2 mm ID glass column;
He at 30 ml/min. Program: 60°C for 2 min, then 8°C/min to 260°C and
hold for 15 min.
b Elutes as azobenzene.
c Elutes as diphenylamine.
d No data.
e These three ions are characteristic for the a and "y forms of chlordane.
No stock should be set in these three for other isomers.
f These ions are listed without relative intensities since the mixtures they
represent defy characterization by three masses.
268
-------
TABLE 11. CHROMATOGRAPHIC CONDITIONS AND CHARACTERISTIC IONS
FOR THE ACID COMPOUNDS ANALYZED BY PACKED COLUMN GC/MS
Compound
RRTa,b
Characteristic EI ions
(relative intensity)
Chlorophenol
0.51
128(100), 64(54), 130(31)
2-Nitrophenol
0.55
139(100), 65(35), 109(8)
Phenol
0.61
94(100), 65(17), 66(19)
2,4-Dimethylphenol
0.67
122(100), 107(90), 121(55)
2,4-Di chlo ropheno1
0.69
162(100), 164(50), 98(61)
2,4,6-Trichlorophenol
0.79
196(100), 198(92), 200(26)
4-Chloro-m-cresol
0.86
142(100), 107(80), 144(32)
2,4-Dinitrophenol
1.04
184(100), 63(59), 154(53)
4,6-Dinitro-o-cresol
1.04
198(100), 182(35), 77(28)
Pentachlorophenol
1.13
266(100), 264(62), 268(63)
4-Nitrophenol
1.63
65(100), 139(45), 109(72)
a Relative to d ^"anthracene.
b 1% SP-1240-DA on 100/120 Supelcoport in a 1.2 o x 2 mil
ID glass column; He at 30 ml/min. Program: 85°C
for 2 min, then 10°C/min to 185°C and hold for 5 rain.
269
-------
TABLE 12. ACCURACY AND PRECISION FOR BASE/NEUTRAL EXTRACTABI.E ORCANICS ANALYZED VIA OPTION A
(SILICA GEL CLEANUP, 1IRGC/HS DETERMINATION) IN THREE POTV AND TWO INDUSTRIAL SLUDGES
Three POTW Kludges Two industrial sludges
Spike recovery " Spike recovery
Coaipound
Deteraina-
tions
Spike level
(mr/1.)
Miniaiua Max i«im
Mean
(*>
Standard
deviation
Hean
RSD Cor
Tripl.
(X)
Determina-
tions
Spike level
(|l*/L)
Hinimui Maxima
Mean
(I)
Standard
deviation
Hean
RSD ror
Tripl.
tt)
1,4-Dichlorobcnzene
27
400
40
,000
44
25
37
18
400
40,
,000
65
59
32
llexachl oroet hane
25
A 00
40,
,000
24
17
33
18
400
40,
,000
55
65
55
Bis(2-rhlor«cthyl)ether
21
400
40
,000
no
77
54
16
400
40,
,000
80
44
20
Acenaphlliylene
21
400
40,
,000
110
66
26
15
400
40,
,000
95
27
14
2,6-DiniI rotolncne
25
400
40,
,000
33
35
12
15
400
40,
,000
89
63
42
Fluoranthene
24
400
40,
,000
140
64
25
15
400
40,
,000
110
42
18
Benzidine
18
7,900
80,
,000
31
31
49
12
780
80,
,000
120
96
25
3,3' -Diclilorobenzidine
9
402
4,
,130
50
78
66
a
-
-
-
-
-
Butylhenzylphthalate
18
4,000
40,
,000
140
59
18
12
400
40,
,000
110
92
19
Di-n-octylplithalate
18
400
40,
,000
130
75
40
15
400
48,
,800
130
71
30
Benzol aIpyrene
21
400
40,
,000
no
43
24
15
400
40,
,100
82
70
30
a This ronpound was not spiked' into the industrial sludges.
-------
TABLE 13. ACCURACY AND PRECISION FOR BASE/NEUTRAL EXTRACT ABLE ORGANICS ANALYZED VIA OPTION B
(GPC CLEANUP, GC/HS DETERMINATION AND ACIDIC EXTRACTABLE ORGANICS
IN THREE POTW AND TWO INDUSTRIAL SLUDGES
Three POTW sludges _ Two Industrial sludges
Spike recovery Spike recovery
Meau Hean
Spike level RSD for Spike level RSD for
Coapound
Deteraina-
tions
CMB/l)
Hiniaua Maxima
Mean
(X)
Standard
deviation
Tripl.
(X)
Deteraina-
tions
(MB/t)
Hiniaua Maxima
Hean
(X)
Standard
deviation
Trlpl.
(X)
1,4-Dirhlorobenzene
27
40
41,600
69
28
16
15
400
40,000
75
16
9
Hexachloroethane
23
400
44,100
48
17
10
16
400
40,000
56
31
14
Bls(2-chloroctliyl (ether
25
400
40,700
220
190
17
16
400
40,000
120
51
7
Acenaphthy1ene
21
400
40,000
65
26
9
18
400
40,000
no
47
12
2,6-Oinitrotoluene
14
4,000
40,000
50
31
10
12
4,000
40,000
100
35
10
Fluoranlhene
24
400
40,000
51
27
14
15
400
40,000
91
22
8
Benzidine
1*
4,870
80,000
79
59
8
16
800
70,000
77
41
10
3,3*-Dichlorobenzidine
13
415
4,130
69
40
11
3
4,000
4,000
110
13
13
Buty Ibcnzy I plilhal at e
IB
4,500
40,000
130
120
7
12
400
40,000
69
15
9
Di-n-octylphthalate
28
4,000
40,000
98
77
13
12
400
48,800
81
73
10
Benzo|i)pyrcne
16
400
40,600
120
94
13
16
400
40,000
66
47
7
Dirnol
24
400
46,300
73
26
27
14
400
40,000
72
34
26
2,4-DiwthyIpheiioI
26
400
41,100
5B
62
37
12
400
40,000
no
64
23
2,4-Dichlorophenol
21
399
40,000
89
32
18
15
400
40,000
140
96
15
Pcntachlorophenol
26
400
40,000
120
67
37
10
400
40,000
50
16
28
-------
TABLE 14. ACCURACY AND PRECISION FOR BASE/NEUTRAL EXTRACTABLE ORGANICS
ANALYZED VIA OPTION B (GPC CLEANUP, GC/MS DETERMINATIONS) AND
ACIDIC EXTRACTABLE ORGANICS IN PRIMARY SLUDGES FROM 40 POTWs3
Spike recovery*5
Determina-
Mean
Standard
Mean Avg. Rel.
Compound
tions
(%)
Deviation
Dev.
for Dupl.
1,3-Dichlorobenzene
27
86
39
11
1,2-Dichlorobenzene
24
68
25
11
Nitrobenzene
13
49
28
17
Hexachlorobutadiene
27
71
22
7
Naphthalene
27
94
37
8
2-Chloronaphthalene
27
89
33
8
Acenaphthaiene
24
79
27
8
2,6-Dinitrotoluene
23
59
38
15
Fluorene
27
94
34
10
N-Nitrosodiphenylamine
24
140
84
11
4-Bromophenylphenylether
27
80
27
5
Di-n-butylphthalate
26
79
43
9
Fluoranthene
26
85
40
9
Pyrene
27
90
47
8
Butylbenzylphthalate
26
95
64
11
Bis(2-ethylhexyl)phthalate
22
65
34
8
Benzo[a]pyrene
27
85
49
11
1,4-Dichlorobenzene
27
82
34
11
Hexachloroethane
21
49
25
21
Isophorone
27
51
27
15
1,2,4-Trichlorobenzene
27
83
32
9
Bis(2-chloroethoxy)methane
24
54
27
12
Hexachlorocyclopentadiene
20
0
0
0
Acenaphthylene
19
76
24
9
Dimethylphthalate
27
50
33
20
2,4-Dinitrotolueae
26
65
50
14
Diethylphthalate
27
65
35
12
Hexa chlo robenzene
27
63
21
10
Phenanthrene/anthracene
26
71
39
11
Chrysene/Benz[a]anthracene
27
71
27
11
Benzo[g,h,i]perylene
19
31
15
10
N-Nitrosodi-N-propylamine
24
77
54
13
1,2-Diphenylhydrazine
23
64
30
9
Benzidine
21
8
11
15
3,3'-Dichlorobenzidine
27
56
58
14
2-Chlorophenol
27
59
16
14
2-Nitrophenol
27
46
39
17
4-Nitrophenol
25
22
23
15
Phenol
26
66
28
11
(continued)
272
-------
TABLE 14 (continued)
„ b
Spxke recovery
Compound
Determina-
tions
Mean
(%)
Standard
Deviation
Mean Avg
Dev. for
2,4-Dimethylphenol
27
34
39
20
2,4-Dichlorophenol
27
68
25
11
2,4,6-Trichlorophenol
27
71
25
13
1,3-Dichlorobenzene
27
86
39
11
j>-Chloro-m-cresol
22
58
32
9
2,4-Dinitrophenol
24
14
27
11
4,6-Dinitro-o-cresol
26
13
24
15
Pentachlorophenol
27
84
43
15
a-BHC
27
39
32
10
Dieldrin
27
72
30
8
4,4'-DDE
27
57
19
11
Heptachlor
26
76
35
10
a Swanson, Stephen E., T. Murry Williams, Lloyd M. Petrie, and Earl M. Hansen,
"Survey of Analysis of POTW Sludges for Priority Pollutants: Part II
Quality Assurance Data," Final Report prepared under EPA Contract No.
68-01-5915, Task 35 (1981).
b Minimum spike level 125 (Jg/L; maximum spike level 1,250 |Jg/L.
273
-------
Sludge
(80 ml)
Extract
Sludge
OPTION A
OPTION B
Sludge
or
or
Determine Base/Neutrals and Pesticides
Discard
Adjust to pH£2
with 6M IICI
Adjust to pti2 11
with 6N NaOH
Adsorption
Chromatography
on Florisil
Adsorption
Chromatography
on Silica Gel
Packed Column
GC/MS on
SP-2250
Capillary Column
GC/MS with
SE-54 WCOT
Determine Plienols
by GC/MS on
SP- 1240-DA
Clean Up by GPC on
Bio Beads SX-3 Eluted
with CtUCIj
Clean Up by GPC on
Bio Beads SX-3 Eluted
with CH2CI2
Extract 3X with CHoCI'
by Homogenization/
Centrifugalion
Extract 3X with CH2CI
by Homogenization/
Centrifugalion
Determine Base/Neutrals
and Pesticides by GC/MS
on SP-2250
Figure 1. Scheme for analysis of extractable organics in sludge.
-------
100.0.,
V
X
1
?
8
N
C
a
c
X
|
%
O
o
s
CM
.c
u
X
u
1
o
I
Q
15
cn
CN
234
-
ro
en
923 x
I TOO
Ll
147 171
350
11:40
900
30:00
26:40
20:00
23:20
100.0-
-?»403
3
8
1353?
191? 1983
S1294 =
1200
40:00
36:40
43:20
1000 SCAN
33:20 IIME
1000
33:20
1400
46:40
1500
50:00
53:20
1700
56:40
1800
60:00
ITOO
63:20
2000 SCAN
66:40 TIME
Figure 2. CC/MS chromatogram of base/neutral compounds on a fused silica capillary column.
-------
tW.0-1
BIC
fN5
o>
fU Q>
8622C3.
T
8:29
16:49
25:69
33:29
41:49
TIME
Figure 3.
GC/MS chromatogram for the base/neutral compounds on a packed column.
-------
a.
CN
2W =
BIC.
o. a.
_c
a a
si
u
226
211
>87
2:30
5:00
7:30
10:00
12:30
Figure 4. OC/MS chromatogram of the acidic compounds.
-------