United States
 Environmental Protection
 Agency
 Environmental Monitoring and Support^
 Laboratory
 Cincinnati OH 45268
 Research and Development
 EPA-600/S4-81-071  Dec  1981
 Project Summary
 Determination  of  Volatile
 Organics  in  Industrial  and
 Municipal  Wastewaters
 Jerry L. Wilson
  This program was undertaken in an
effort to develop analytical methodol-
ogies for the determination of volatile
organics in industrial and municipal
wastewaters. The efforts were directed
exclusively toward analytical tech-
niques employing gas chromatography
with flame ionization and  halogen-
specific detectors.
  A number of gas chromatographic
column packings were evaluated in an
effort to find  a set of optimum
operating conditions. Direct aqueous
injection, solvent extraction, and
purge and trap techniques were
evaluated as means of sample intro-
duction. The most satisfactory results
were obtained using the  Bellar-Lich-
tenberg purge and trap method with
both FID and Coulson detectors.
  Stability of the test compounds as
solutions in methanol and n-butanol
was evaluated after 30-, 60-, and 90-
day periods. Preservation  of aqueous
samples containing residual chlorine
for a one week period was studied;
aliquots of reducing agents,  added to
the sample before storage, were found
to aid in maintaining cfiemical stability.
  The methods developed were tested
on spiked industrial wastewater  sam-
ples in order to evaluate their applica-
bility to real world situations. The
methods were found to be satisfactory
for relatively "clean" samples. For
grossly contaminated samples, how-
ever, the methods suffer  from inter-
ferences caused by lack of  detector
specificity.
  This report was submitted in fulfill-
 ment of Contract No. 68-03-2635 by
 California Analytical Laboratory work-
 ing as a subcontractor to The Carbor-
 undum Company under the sponsor-
 ship of the U.S. Environmental Pro-
 tection Agency. This report covers the
 period May, 1978 to June, 1979.
  This Project Summary was developed
 by  EPA's Environmental Monitoring
 and Support Laboratory, Cincinnati,
 OH, to announce key findings of the
 research project that is fully docu-
 mented in a separate  report of the
 same title (see Project Report ordering
 information at back),

Introduction
  Under provisions of the Clean Water
Act, the United States Environmental
Protection  Agency (EPA) is required to
promulgate guidelines establishing test
procedures for the analysis of pollutants.
The Clean Water Act Amendments of
1977 emphasize the control of toxic
pollutants and declare the 65 "priority"
pollutants and classes of pollutants to
be toxic under Section 307(a) of the Act.
This report  is one of a series that
investigates the analytical behavior of
selected priority pollutants and suggests
a suitable test procedure for their
measurement.
  The purpose of the current study was
to develop and evaluate various tech-
niques for  analysis of volatile organics
by gas chromatography. Thirty-six
compounds were studied. While this
study developed  analytical methods

-------
specifically for these compounds, they
could apply to analysis of other com-
pounds as well. A list of the compounds
studied is found in Table 1.
  As  part of this study, several  gas
chromatographic column  packing  ma-
terials were  evaluated with respect to
their ability to separate as many of the
compounds as possible. Detection was
accomplished by means of flame ioniza-
tion,  electron  capture,  and Coulson
electrolytic  conductivity detectors.
Direct aqueous  injection, liquid/liquid
extraction, and  purge and trap tech-
niques were studied as methods of
  sample introduction. The effects  of
  storage conditions upon the  chemical
  stability of aqueous samples containing
  trace levels  of test compounds and
  residual chlorine were  studied.  An
  effort was  made to determine a  set of
  optimum conditions for water sample
  storage.
    The methods developed were applied
  to the analysis of industrial wastewaters.
  Samples from five industrial categories
  were analyzed by direct injection using
  FID  and Coulson detectors,  by  liquid
  extraction  using the electron capture
  detector, and by purge and trap using
Table 1.    Retention Times of Test Compounds
                                            Retention times Imin.)
          Compound
Column 1    Column 2    Column 3
1 dichlorodifluoromethane
2 methyl chloride
3 vinyl chloride
4 methylbromide
5 chloroethane
6 trichlorofluoromethane
7 1,1 -dichloroethylene
8 methylene chloride
9 1 ,2-trans-dichloroethylene
10 acrolein
11 1,1 -dichloroethane
12 chloroform
13 1,1,1 -trichloroethane
14 carbon tetrachloride
15 acrylonitrile
16 benzene
17 1 ,2-dichloroethane
18 trichloroethylene
19 dichlorobromomethane
20 2,3-dichloropropene
21 1 ,2-dichloropropane
22 dibromomethane
23 1 ,3-trans-dichloropropylene
24 2-chloroethylvinyl ether
25 toluene
26 1,3-cis-dichloropropene
27 1,1,2-trichloroethane
28 chlorodibromomethane
29 1 -chlorocyclohexene
30 tetrachloroethylene
31 1 ,2-dibromoethane
32 chlorobenzene
33 ethylbenzene
34 1, 1 ,2,2-tetrachloroethane
35 bromoform
36 p-dichlorobenzene
1.7
0.7
—
1.4
2.5
6.3
6.9
4.1
8.8
4.7
8.2
9.3
10.9
11.3
5.4
14.0
10.0
13.6
11.8
12.9
12.9
10.1
13.1
15.0
19.1
14.2
14.2
14.0
18.7
18.1
14.9
20.0
21.8
18.0
16.2
27.3
0.95
0.54
1.0
0.8
1.3
2.3
3.6
4.5
5.2
4.2
4.5
6.7
8.8
9.3
4.8
13.8
7.5
12.7
10.5
11.5
11.6
10.0
12.1
14.0
23.3
13.5
13.6
13.6
—
18.3
14.4
20.4
26.7
18.6
16.3
—
1.3
1.8
1.6
2.2
2.5
5.0
3.4
3.6
4.7
2.5
5.6
5.9
9.2
8.2
3.2
8.8
8.2
9.9
10.2
10.0
10.3
10.4
10.8
10.7
11.7
11.0
11.3
12.6
14.3
13.4
12.7
13.0
13.7
14.5
14.1
17.2
Column 1: 1.8 m x 2 mm glass with  1% SP1000 on Carbopack B. 60-80 mesh;
Program-Inject at 50°C; hold 5 min, then up 10°c/mih to 225°C for 5 min
Column 2: 1.8 m x 2 mm glass with 0.2% Carbowax  15OO on Carbopack C. 60-80
mesh; Program Inject at 40°C; hold 4 min, up  10°C/min to 175°C, hold 10 min.
Column 3: 4 m x 2 mm i.d. glass with20%SP2100/0.1%Carbowax 1500on 100-125
mesh Supelcoport; Program-Inject at 50°C; hold 4 min', up 10°C/min to 170°C, hold
4 min.
FID and Coulson detectors. The results"
of the various analyses were compared
for each of the industrial samples.

Development of Analytical
Methods

Chroma to graph y
  Three approaches for analyzing
wastewater for thirty-six volatile organic
compounds were evaluated:  the purge
and trap technique, the direct aqueous
injection approach, and the solvent
extraction  approach.  The  first two
approaches employed  both  the flame
ionization detector (FID) and the Coulson
electrolytic  conductivity detector. The
third  approach utilized the electron
capture  detector (ECD). Except as
subsequently specified, all the chroma-
tographic column packing  materials
tested were compatible with  any of the
analytical approaches.
  No gas chromatographic column was
found which could separate all the test
compounds in a single programmed
run. Tenax GC,  3%  SP-2510,  and
Chromosorb 102 were tried and found
to be unsatisfactory for general  use.
These columns were rejected primarily
because of inadequate separation of a
number of compounds, or,  in some
cases, because of poor chromatography.
For a specific  subset of the purgeables,
however, these column materials may
be suitable in a given wastewater.
  Three  columns  which  did  show
adequate separation of most compounds,
in addition to good chromatography
(sharp peaks, no tailing), were 1% SP-
1000  on Carbopack B  (60-80 mesh),
0.2% Carbowax 1500 on Carbopack C
(60-80 mesh), and 20% SP-2100/0.1%
Carbowax 1500 on Supelcoport (100-
120 mesh). The retention times of each
compound on these columns along with
the column dimensions and temperature
programs are  given in Table  1. A close
examination  will show that  those
compounds which do not separate on
the 1% SP-1000 column, for example,
will separate on the 20% SP-2100, and
vice versa. Thus, by comparing retention
times  on two columns, sufficient
separation  of  all  compounds  can be
achieved.

Purge  and  Trap
  The purge  and trap technique was
evaluated with respect to its ability to
detect 1 ppb of all the test compounds. A
key factor is the purgeability of the test
compounds from aqueous solution. The
recovery of test compounds from water

-------
was determined by comparing  the
detector response from a purge and trap
analysis to that from a direct injection of
the same mass  of a given compound
into the same chromatographic system.
Except where noted, the following purge
and trap conditions were employed:
 Purge time:
 Purge
  Temperature:
 Desorb Time:
 Desorb
  Flow Rate:
 Desorb
  Temperature:
 Sample Size:
10-15 minutes

ambient (22-25°)
5 minutes

25 mL/minute

200° C
5 mL
  The purging efficiencies of all the test
compounds were measured using the
approach  described above. Measure-
ments were made at 10 ppb. Results for
two traps are shown in Table 2. The data
shows that, in some cases, recoveries
determined using the Tenax/Carbosieve
B (HT) trap (Trap  2) are greater  than
those determined using the Tenax/silica
trap (Trap  1).
  Particularly noteworthy are the results
for acrolein and acrylonitrile. The Trap 2
recoveries equal those obtained  with
Trap 1 even after  increasing  purging
temperature  and  ionic strength.  The
results suggest that polymerization  or
irreversible absorption  are the cause of
low  recovery,  and not poor purging
efficiency.
  Compounds  having  a lower boiling
point than trichlorofluoromethane were
not trapped effectively by the Tenax/
silica trap. Therefore,  various Tenax/
charcoal traps were prepared and tested
for their ability to trap these gaseous
compounds.  Data  were obtained for
Tenax/Carbosieve B (HP) traps and for
Table 2.    Purge and Trap Efficiencies of Test Compounds
 Compound
                                           Purge and Trap Efficiency*
                          Trap /c
                      Trap 2C
trichlorofluoromethane
1, 1 -dichloroethylene
methylene chloride
1 ,2-trans-dichloroethylene
1, 1 -dichloroethane
chloroform
1, 1, 1 -trichloroethane
carbon tetrachloride
trichloroethylene
1 ,2-dichloroethane
dichlorobromomethane
1 ,2-dichloropropane
1 ,3-trans-dichloropropylene
1 ,3-cis-dichloropropylene
1, 1 ,2-trichloroethane
chlorodibromomethane
tetrachloroethylene
chlorobenzene
1. 1 ,2,2-tetrachloroethane
bromoform
dibromomethane
2-chloroethylvinyl ether
1 , 2 -dibromoethane
2,3-dichloropropene
1 -chlorocyclohexene
p-dichlorobenzene
benzene
toluene
ethylbenzene
acrolein
acrylonitrile
78
82
82
104
100
76
86
60
77
74
85
83
87
58
89
72
82
110
28
49
93
21
56
87
68
100
41
38
28
9
15
85
120
140
92
95
83
108
87
71
100
134
88
72
60
85
91
48
81
82
77
81
45
60
93
105
120
82
123
48
44
41
"Determined by comparing the  response by purge  and trap to that obtained
 by direct injection of the test compound.
^Tenax/silica trap obtained from Tekmar, Inc. Purge time 15 min.
°Tenax/Carbosieve  BfHT) (Supelco. Inc.. Bellefonte, Pa) trap prepared in our
 laboratory, 14 cm Tenax + 7 cm Carbosieve B (HT). Purge time 15 min.
Tenax/SKC  charcoal traps.  For  each
trap, purge times of 10 and 15 minutes
were tested.
  Based upon the results, a number of
conclusions can be drawn: 1) Charcoal/
Tenax traps are effective adsorbants for
the compounds tested; 2) five to seven
cm of charcoal  are  required for  a
retentive trap; 3) recoveries of some
compounds are lower using the longer
15  minute purge  times. This indicates
that gaseous compounds are easily
desorbed  from these  traps at room
temperature.


Solvent Extraction/Electron
Capture Detection
  The  purposes of this aspect of the
study were to determine the optimum
conditions of  extraction using a simple
extraction procedure, the accuracy and
precision of  the extraction process at
the one ppb level; and the suitability of
the extraction solvents n-pentane and
isooctane.
  The first extraction techniques tested
were those in which no head space was
allowed in the extraction vessel. Various
methods of agitation were tested and
the pentane  layer analyzed by electron
capture  gas chromatography. The
extraction efficiency for each compound
was determined  by comparison  with
standards  of  known  concentration.  It
quickly became  apparent that  very
vigorous agitation  methods are needed
to get efficient extractions when there
are no headspace bubbles to help mix
the two phases.
  If, on the other hand, one allows  a
small amount of headspace, very  good
extraction  efficiencies can be obtained
by simple hand shaking for only one
minute. Trichlorofluoromethane and
compounds  boiling above  trichloro-
fluoromethane could be extracted with
high efficiency.
  The  variation  in response  of the
electron capture detector to the test
compounds is quite pronounced. For
those compounds containing three or
more halogens, the response is suffi-
ciently high that one can readily detect 1
ppb  or less.  The monohalogenated
compounds  and  even the dichloro-
ethanes and  dichloropropanes cause
such a poor  response that  100 ppb
would be difficult to detect at the same
detector setting used for the trihalogen-
ated compounds. Typical estimated
detection limits and extraction efficien-
cies for the more sensitive compounds
are presented  in Table 3. For the

-------
Table 3.    Method Detection Limit and Extraction Efficiencies for Polyhalogenated
           Compounds
Compound
dichlorodifluoromethane
trichlorofluoromethane
chloroform
1, 1, 1 -trichloroethane
carbon tetrachloride
trichloroeth ylene
chlorodibromomethane
dibromomethane
1, 1,2-trichloroethane
chlorodibromomethane
tetrachloroethylene
1, 1 ,2,2-tetrachloroethane
bromoform
MDL*
(ppb)
1
0.8
0.3
0.02
0.02
0.04
0.1
0.1
0.3
0.1
0.02
0.04
0.04
Extraction
Efficiency
NA
NA
88
100
100
97
100
65
91
100
100
100
100
*MDL=minimum detection limit; calculated as follows:
                MDL=   (Area)            (2 mL)
                      (Response area units) (35 mL) (5 uL)
                                   gm
             in gm
               uL
remaining compounds, the sensitivity of
the detector is relatively poor.
  The MDLs for Table 3 were calculated
using a minimum area of 200 units. The
response was determined from the
regression analysis of each curve.
Further assumptions  were  35 mL
sample extracted with 2  mL pentane
and an injection volume of 5 microliters.
  The solvents tested were pentane and
isooctane.  Both were  shown to be
satisfactory. However, even Nanograde
Quality batches  from three  different
suppliers contained low boiling, elec-
tron capture responsive contaminants.
These could  not  be eliminated by
distillation.  Furthermore, elution of the
solvent  normally caused  a negative
peak, which  would interfere with
quantitation of any compound eluting at
the same time.

Direct Aqueous Inject/on
  Direct injection of aqueous solution of
the test compounds was studied using
the  FID  and  Coulson detectors  to
evaluate the  utility of the method  in
detecting the compounds at ppm levels.
Using the same  chromatographic col-
umns as used for the other methods, it
has been demonstrated that one ppm of
almost every compound can be detected.
The  exceptions were methyl chloride
(10 ppm) and  p-dichlorobenzene  (2
ppm). Columns may require an injection
of blank water prior to analysis until a
clean blank run  is obtained, because
water seems to  clean the column  of
strongly  absorbing compounds. The
20% SP-2100 column  is particularly
good for aqueous injection.  The 0.2%
carbowax 1500 column, however, does
not work well for aqueous injections.
The water appears to uncover active
sites  which  bind some  compounds
irreversibly and thus show a progressive
loss of sensitivity during repeated runs.

Preservation Study
  The aim of the preservation study was
to determine  the effects of certain
storage parameters on the integrity or
chemical stability of clean water samples
containing trace levels  of  the test
compounds at the ppb level. Parameters
investigated were temperature (4°Cand
25°C), pH  (2, 7 and  10),  and the
presence of reducing agents (ascorbate,
thiosulfate, and sulfite) added to destroy
any residual  chlorine. Samples were
prepared and stored for  seven days
under various sets of conditions and
analyzed by purge and trap to determine
recovery of each of the compounds.
  1,1,2,2-tetrachloroethane is not pre-
served at pH .10 under any conditions.
Loss of 2-chloroethylvinyl  ether occurs
at pH 2  and  7. Formation of trace
quantities  of  unidentified compounds
was observed in many of the unpreserved
chlorinated samples. These results
alone indicate the  value  of  adding a
reducing agent when collecting chlorine
treated wastewater effluents for volatile
organic analyses. The value of adding a
reducing reagent may be even greater
for actual wastewater samples which
may contain many non-volatile organics.
since it is well established that chlorine
reacts with compounds such as humic
acids to produce trihalomethanes.

Application Phase
  The methods described in preceding
sections  were tested on industrial and
municipal wastewater effluents. Anal-
ysis  of  actual wastewater samples
allows one to evaluate the effect of the
presence of non-volatile organics and
other dissolved solids which are typically
present  in wastewater. One can  also
evaluate the flexibility of each method in
handling wide variations in concentra-
tion and  number of compounds.


Selection and Preparation of
Industrial Samples
  The industrial categories used in this
study were selected to give a range of
difficulty  in analysis. The five industries
selected were:
  1. Organic Chemicals Industry  (In-
     dustrial Category #12)
  2. Publicly Owned Treatment Works
     (POTW)
  3. Auto and Other Laundries (Indus-
     trial Category #19)
  4. Coal Mining Industry  (Industrial
     Category #11)                  (
  5. Pulp and Paper Industry (Industrial
     Category #16)
  Previous work  with samples from
these  industries  indicated that their
effluents contained a wide  range of
volatiles  concentration. In each of the
above industries, a number of samples
were available in  California Analytical
Laboratory.  Several samples  in each
industry were composited to yield two
liters of wastewater from each industry.
These samples were representative of
the  industry and  are  not typical of a
single plant or treatment facility. Thus,
these composited samples may repre-
sent a "worst case" for each industry.
  Each composited  industrial  sample
was  analyzed five  ways: 1) direct
aqueous injection using the  FID; 2)
direct aqueous injection using the
Coulson detector; 3) liquid/liquid
extraction  with the electron capture
detector; 4) purge and  trap with the
Coulson detector; and 5) purge and trap
with the  FID. All the GLC data collected
in the application  phase were obtained
using a  4mx2mm(id) glass column
packed with 20% SP-2100/0.1 % carbo-
wax 1500 on 100-120 mesh supelcoport.
  After the initial analyses, the samples
were spiked with the test compounds at
20 ppb and divided into two portions.

-------
 One portion was analyzed immediately
 by purge and trap. The second portion
 was placed in a septum sealed 40 mL
 vial and placed in the refrigerator to be
 analyzed one week later.
 Analysis of Composited
 Samples

 Direct Aqueous Injections
   The results of direct aqueous injection,
 for the POTW sample, using both the
 FID and the Coulson detectors are given
 in Table 4. Peaks in the samples which
 matched  a  standard retention time
 within 0.1  minute were  listed  as
            positive identifications. The only other
            criterion used was the selectivity of the
            Coulson detector.  For the halogenated
            compounds, if a peak appearing in the
            FID run  was not confirmed by  the
            Coulson run, then the peak was ignored.
            Furthermore, for those  halogenated
            compounds having a high FID response,
            both detectors had to see a peak before a
            quantitation  was  reported. Since  the
            Coulson detector is highly selective for
            halogenated compounds, all peaks
            found by that detector which correspond
            to the retention time of one of the  test
            compounds are indicated in the tables,
            but only those peaks which are confirmed
            by the FID analysis are quantitated.
Table 4.    Results of Analysis of Publicly Owned Treatment Works Composited
           Sample
        Compound
Reten-
 tion   Direct  Inject   Solvent  Purge and Trap
 Time                 Extraction
 (min)    FID   Colson    EC      FID   Coulson
 dichlorodifluoromethane         1.3
 vin yl chloride                  1.6
 methyl chloride                 1.8
 methyl bromide                 2.2
 chloroethane                  2.5
 acrolein                       2.5
 acrylonitrile                    3.2
 1,1 -dichloroethylene            3.4
 methylene chloride             3.6
 1,2-trans-dichloroethylene       4.7
 trichlorofJuoromethane          5.0
 1.1 -dichloroethane             5.6
 chloroform                     5.9
 carbon tetrachloride            8.2
 1,2 -dichloroethane             8.2
 benzene                       8.8
 1,1.1 -trichloroethane            9.2
 trichloroethane                 9.9
 2.3-dichloropropene          10.0
 dichlorobromomethane        10.2
 1.2-dichloropropane          10.3
 dibromomethane             10.4
 2-chloroethylvinyl ether       10.7
 1,3-dichloroproylene          10.8
 1,3-cis-dichloropropylene      11.0
 1,1,2-trichloroethane          11.3
 toluene                     11.7
 chlorodibromomethane        12.6
 1,2-dibromoethane           12.7
 chlorobenzene               13.0
 tetrachloroethylene           13.4
 ethylbenzene                13.7
 bromoform                   14.1
 1 -chlorocyclohexene          14.3
 1,1,2,2-tetrachloroethane      14.5
 p-dichlorobenzene            17.2
          60
                         5
                         0.4
         300
                                 0.5     2
                        11       11
 All results are given in parts per billion (ppb).
*Peak detected; not quantitated.
  It should be noted that two industrial
category samples, organic chemicals
and auto and other laundries, yielded a
large number of peaks by FID analysis,
but only a small  number by  Coulson
analysis. This made  identification
difficult in some cases.  In any event, it
seems certain that there are a number
of volatile compounds present in these
effluents, at the ppm level, which are
not compounds under study here.

Analysis by Solvent Extraction
and Electron Capture Detection
  The results of this analysis are found
in column five of Table 4. As expected,
this  method is capable  of detecting
levels of some compounds not seen by
the purge and trap  method. However,
there  are a few instances  where
compounds  are  reported at a level
which  should have  been seen by the
other methods but  they  were not. In
general,  the agreement  between  the
extraction method and purge and trap is
good.

Analysis by Purge and Trap
  Columns six and seven of Table 4con-
tain the results of the purge and trap
analyses. The same criteria for quantita-
tion was used as in the direct injection
analysis. The quantitative agreement
between the two detectors is poor, but
using the selectivity of the Coulson as a
guide  and  an examination  of  the
chromatograms indicate  the  FID is
responding to other compounds that are
present in the sample. In almost every
case,  the FID is  measuring  higher
values  than  the  Coulson, which is
consistent with the above hypothesis.

Analysis of Spiked Samples
  As described previously, each of the
industrial composites were spiked with
20 ppb of each of the test compounds.
The spiked solutions were transferred to
six 40 mL glass screw capped vials and
sealed  with Teflon lined silicon rubber
septums. Three of  these vials were
immediately analyzed by purge and trap,
while three were stored m a refrigerator
for one week, then analyzed. The results
for the POTW sample are seen in Table
5. The  concentration shown for  the
spiked  compounds are  the average of
three determinations ± the  relative
standard deviation.
  The gases dichlorodifluoromethane,
methyl chloride,  methyl bromide and
chloroethane were not  seen, probably
due to a very rapid loss from the working
solution. In  order to see these com-

-------
Table 5.    Analysis of Publicly Owned Treatment Works Spiked Sample
Compound
dichlorodifluoromethane
methyl chloride
methyl bromide
chloroethane
ac role in*
acrylonitrile*
1, 1 -dichloroethylene
methylene chloride
1 ,2-trans-dichloroethylene
trichlorofluoromethane
1 , 1 -dichloroethane
chloroform
carbon tetrachloride
1 ,2-dichloroethane
benzene*
1,1,1 -trichloroethane
trichloroethylene
2, 3 -dichloropropene
dichlorobromomethane
1 ,2-dichloropropane
dibromomethane
2-chloroethylvinyl ether*
1 ,3-trans-dichloroproplyene
1 ,3cis-dichloropropylene
1, 1 ,2-trichloroethane
toluene*
chlorodibromomethane
1 , 2 -dibromoethane
chlorobenzene*
tetrachloroethylene
ethyl benzene
bromoform
1 -chlorocyclohexene*
1, 1 ,2,2-tetrachloroethane
p-dichlorobenzene*
Spike
level
(ppb)
30.7
26.9
33.1
28.0
33.6
17.0
23.8
21.2
21.4
14.9
20.8
24.0
17.9
20.2
22.0
22.6
23.0
20.0
21.4
22.4
20.0
20.0
11.9
11.9
25.0
22.8
26.6
23.6
21.4
20.8
19.9
19.9
20.0
20.8
20.6
Concentration Determined (ppb)
Back- Spiked Spiked
ground Day 0 Day 7




20±58%
6±16%
40±16%
70±7%
31±13%
Masked
21±11%
9 35+2%
21 ±4%
33±7%
36±38%
25+19%
21±1%
20±17%
2 22+3%
20+17%
29±1%
13±24%
12+4%
10+8%
28±3%
0.5 19±101%
30±2%
19+7%
11 ±22%
9 36+4%
1 + 16%
20+12%
12±18%
24±0%
17±16%




18+67%
109±59%
16±48%
83±41%
11±111%
Masked
38±3%
30±5%
11 ±3%
15+29%
20+39%
10±34%
16+14%
12±9%
45±3%
12+9%
14±14%
13±7%
19±29%
16+44%
27+1%
11+7%
13±9%
49±2%
5±6%
15+1%
0.2+0%
71 ±20%
11+6%
20+27%
16±2%
*These compounds were analyzed using the FID. All others were analyzed using
 Coulson Detector.
pounds, an alternative spiking procedure
must be used. However, their rapid loss
strongly suggests that they are not likely
to be found in an  industrial  effluent
unless sampling is done very soon after
the gases enter the effluent.
  In general, the precision of  analysis
was poor. Onesignificantfactorwasthe
frequent  presence of interfering com-
pounds and the occasional high levels of
some test  compounds  in the sample
prior to spiking. These situationscaused
sufficient skewing or distortion of the
chromatograms such that correct inte-
gration of peaks was not accomplished.
It was usually not possible to check peak
heights against integrator values, but in
the few instances where it was possible,
the precision was usually within 10%
relative  standard deviation. A second
factor was the carryover between
replicate runs. High  levels of some
compounds made carryover a significant
problem.  Due  to  time constraints,
blanks were not run between replicates,
but only between different samples.

Conclusions and
Recommendations
  Methodologies for the analysis of
volatile organic compounds in industrial
and municipal wastewater have been
developed  and evaluated,  using  gas
chromatography with the flame ioniza-
tion,  electron capture,  and Coulson
detectors.  Several  improvements in
existing analytical methods  have been
realized. Some of the current techniques
require further development.
  Although no single GC column was
found that gives  complete chromato-
graphic separation of all the test com-
pounds, several GC packing materials
were identified  that  provide gooc
chromatographic separation of all but a
few of the test compounds.
  All the test compounds were detectec
by  one  or  more of the  analytics
approaches studied at levels of 1 ppb oi
less. The nonhalogenated compounds
are not detected by Coulson or electror
capture.  Many of the compounds are
detectable at levels of 0.1 ppb or lower
  The stability of the test compounds as
standards in  organic and as "clean'
samples in aqueous media was studiec
in detail. The addition of preservatives tc
reduce residual  chlorine  in aqueou:
samples was found to help maintain the
integrity (chemical stability) of aqueous
samples  containing residual  chlorine
and trace quantities of  the test com
pounds.
  The major limitation of all the analyti
cal approaches  investigated  in this
study was the lack of specificity o
detection. The FID  was the only detectoi
of those studied that was able to detec
the nonhalogenated compounds. How
ever,  it  was the  least  specific  of the
detectors studied.  For relatively "clean'
wastewaters,  a combination of  the
analytical methods  developed on this
project have been found to usually be
both specific  and sensitive. However
for grossly polluted wastewaters
interferences became such  a severe
problem  that qualitative  identificatior
was not feasible.
  Further study is  recommended tc
develop  either sample cleanup  proce
dures for use prior to analysis, or to fine
an analytical scheme more specific tc
the test compounds of  interest  before
these or  similar methods can become
generally applicable for  analysis o
complex industrial or municipal waste
waters. Other areas recommended foi
further study include: further evaluatior
of the precision and accuracy  of th«
methods  developed  herein, evaluatior
of alternative trap materials to be usec
in the purge  and trap  approach, ant
development of alternative methods foi
the analysis of acrolein and acrylonitrile

-------
Jerry L. Wilson is with the California Analytical Laboratory, Sacramento, CA
  95814.
James E.  Longbottom is the EPA Project Officer (see below).
The complete report, entitled "Determination of Volatile Organics in Industrial
  and Municipal Wastewaters," (Order No. PB  82-119 090; Cost: $10.50,
  subject to change) will be available only from:
        National Technical Information Service
        5285 Port Royal Road
        Springfield. VA 22161
        Telephone:  703-487-4650
The EPA Project Officer can be contacted at:
        Environmental Monitoring and Support Laboratory
        U.S. Environmental Protection Agency
        Cincinnati,  OH 45268
                                                                            . S. GOVERNMENT PRINTING OFFICE: I98I/559-092/3346

-------
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Postage and
Fees Paid
Environmental
Protection
Agency
EPA 335
Official Business
Penalty for Private Use $300
       US  ENVTR  PROTECTION  AGENCY
       REGION  5  LIBRARY
       230  S  DEARBORN STREET
       CHICAGO  IL  60604

-------