United States
Environmental Protection
Agency
Water Engineering
Research Laboratory
Cincinnati OH 45268
Research and Development
EPA/600/S2-85/050 Aug. 1985
Project Summary
Determination of Toxic
Chemicals in Effluent from
Household Septic Tanks
Foppe B. DeWalle, David A. Kalman, Donald Norman, and Gary Plews
The presence of volatile organics was
evaluated in raw domestic sewage that
was generated in a subdivision and
treated by a large 5-year-old commuar-
nity septic tank from which the solids
had been pumped just before this
study. Analyses showed the presence
of priority pollutants in the raw
sewage. Essentially no removal of
these compounds occurred during the
2-day detention in the septic tank. The
priority pollutants generally showed
higher levels during the weekend, prob-
ably reflecting leisure activities and use
of related chemicals (paint thinners,
grease removers, toilet bowl cleaners,
etc.). Most of the other volatile com-
pounds were hydrocarbons, and their
removal by the septic tank generally de-
creased with increasing molecular
weight. Several organosulfur com-
pounds showed substantial increase as
a result of anaerobic degradation pro-
cesses in the septic tank.
This Project Summary was devel-
oped by EPA's Water Engineering Re-
search Laboratory, Cincinnati, OH, to
announce key findings of the research
project that is fully documented in a
separate report of the same title (see
Project Report ordering information at
back).
Introduction
Onsite disposal of sewage is widely
practiced in rural areas and urban
fringes, but it has received relatively lit-
tle attention from regulatory agencies
or research institutes. For example, little
comprehensive information is currently
available to document successes and
failures of onsite disposal systems and
the resulting health implications. In the
State of Washington, the census noted
403,910 septic tanks or cesspools and
14,464 other individual systems such as
aerobic units or ponds. Such systems
represented 34.7% of all housing units.
A current survey in the State of Wash-
ington also indicates the presence of
more than 500 large onsite systems
serving hospitals, schools, restaurants,
and subdivisions. Because the septic
tank effluent is directly returned to the
soil through infiltration into the subsur-
face drainfield, considerable public
health concern exists over groundwater
contamination and pollution of drinking
water wells. A recent study showed
such widespread contamination in
major parts of central Pierce County
south of Tacoma.
In previously published studies, only
the efficiency of the septic tank with re-
spect to BOD, SS, and grease removal
was evaluated; the trace organics pres-
ent in domestic sewage and in septic
tank effluent were not measured. The
present study was conducted to collect
such data. The aim was to measure the
presence of volatile priority pollutants
in domestic sewage as it entered a large
community septic tank system. The
study also evaluated the removal of
these compounds in the anaerobic sep-
tic tank by analyzing effluent samples
collected from the distribution box.
Materials and Methods
This study used a community septic
tank serving 97 homes in the Oakbrook
6 subdivision located south of Tacoma,
-------
Pierce County, Washington. The homes
are located on four streets served by a
200-mm (8-in.) gravity sewer that dis-
charges into a wetwell. A 10.1-L/s (160
gpm) submersible centrifugal pump
transports the waste to the septic tank.
The unit contains 84,935 L (22,440 gal)
liquid volume in the first compartment
and 42,468 L (11,220 gal) in the second
compartment. The raw sewage sample
was collected through the manhole
from the inlet-T before the sewage
mixed with the contents of the first com-
partment. The effluent sample was col-
lected from the distribution box located
4.57 m (15 ft) downstream from the
effluent-T.
The sewage was collected with an all
glass and teflon, custom-made sampler.
The sewage was drawn by suction
through a 0.5-in. telfon intake line with a
100-ml glass syringe. When the syringe
was in the drawn position, a teflon
solenoid valve closed the intake line and
opened the discharge line. When the sy-
ringe content was subsequently dis-
placed, it flowed through the discharge
line to a collection device with a floating
plunger to prevent losses of volatile or-
ganics. The septic tank effluent was col-
lected using a similar sampler that drew
from below the liquid surface in the dis-
tribution box. The samples were col-
lected as 24-hr composites during a
7-day continuous period.
The volatile organics were removed
from the aqueous sample with a purge
and trap method. A modified Hewlett/
Packard 7675A* purge and trap sampler
was used to purge 10 ml of liquid with
nitrogen gas at a rate of 20 ml/min. The
volatile organics that were stripped
from the liquid were subsequently ab-
sorbed when the nitrogen passed
through a Tenax GC trap. The trap was
subsequently heated, and the volatile
organics were backflushed and trapped
in the initial 0.5-m portion of a 30-m
fused silica WCOT column with an
SE-54 stationary phase chilled by liquid
nitrogen. After removal of the cryotrap
and flash heating of the column, the
volatiles were separated in the gas chro-
matograph and detected by mass spec-
trometry. A computer library system
containing spectra of 30,000 com-
pounds was used for a spectral compar-
ison with each detected compound.
"Mention of trade names or commercial products
does not constitute endorsement or recommenda-
tion for use.
Results and Discussion
Flow Measurements
Flow data were obtained both from
the water usage records of the Lake-
wood Water District Company and from
measurements in the wetwell. The
water usage data show a baseline of
58.7 L/min (15.5 gpm), which usage
triples to 219.5 L/min (58 gpm) during
the summer months because of exten-
sive usage. The frequency distribution
of the flow rates shows several maxima
corresponding to usage in a household
with 1, 2, 3,4, 5, 6, or 7 persons, respec-
tively. The median household with 3.2
persons uses 897 L/day (237 gal/day).
The usage ranges from 329.3 L/ person
per day (87 gal/ person per day) for a
one-person household to 242.2 L/ per-
son per day (64 gal/ person per day) for
a seven-person household.
The flow measurement at the wetwell
consisted of determining the interval
between sequenced pump switch-on
times. The flow rate during the day was
calculated with the holdup volume in
the wetwell. The average calculated
flow was 44.7 L/min (11.8 gpm), which is
24% less than that calculated with the
usage data. This discrepancy indicates
that about a quarter of the used water
does not reach the septic tank and is lost
through evaporation (clothes dryer,
plant evapotranspiration, sewer leaks,
etc.). The highest discharge rates and
the highest standard deviation were
noted around 9 p.m.
Presence of Trace Organics
The efficiency of the volatile organic
analysis (VOA) using the purge and trap
unit was evaluated using surrogate
compounds spiked at 20 ppb in the
liquid before purging. The results
show a median recovery of 91% for
bromochloromethane, 90% for 1,4-
dichlorobutane, and 82% for De-
benzene (Figure 1). However, the stand-
ard deviations were substantial at this
tow concentration level.
The organics were measured in the
samples during an intensive, week-long
monitoring followed by six additional
samplings, the last of which occurred
on January 23, 1982. The results are
summarized in Table 1 and indicate that
dichloromethane was found in all sam-
ples. Toluene followed in frequency of
detection. These two compounds were
also found in the water collected from
the 125-ft-deep monitoring well adja-
cent to the drainfield. The analysis dur-
ing the week-long monitoring was initi-
200
750
01
50
20 iig/L Spiking Level
o Bromochloromethane
A 1,4-Dichlorobutane
a D6-Benzene
JO
30 50 70
90
98
Percentage of Samples with Less Than
Corresponding Recovery
Figure 1. Normal frequency distribution of
surrogate recovery.
ated on Monday, September 22, and it
was terminated on Sunday, September
28, 1980. The volatile organic fraction
typically contained 40 to 50 compounds
at a concentration >1 ppb, but only five
were identified as priority pollutants.
Toluene was the most prevalent
among the priority pollutants, with an
average concentration of 34.6 (xg/L in
the raw sewage and 38.8 u-g/L in the ef-
fluent. The toluene, which originates
from cleaning solvents and paint thin-
ners, reached its maximum concentra-
tion of 47.8 p-g/L in the influent on Fri-
day, and 56.9 |xg/L in the effluent on
Sunday. The shift in the maximum
toluene concentration by 2 days may be
a result of the detention time, which
was estimated at approximately 2 days.
Dichloromethane, which likely origi-
nated from chlorinated tapwater, was
present in the next highest concentra-
tion (9 M-9/U and also showed its highest
level on Sunday. No lag was detected
between effluent and influent, however.
Chloroform showed its maximum con-
centration (5 u.g/L) in the influent on Sat-
urday, and the effluent showed a sec-
ond maximum on Sunday (0.8 (Jtg/L),
representing a 1-day shift. Tetrachlo
-------
Table 1. Occurrence of Volatiles in Septic Tank Samples
Percent of Samples in which Compound was Present
Compound
Dichloromethane
Chloroform
Trichlorofluoromethane
Bromomethane
1, 1,2-Trichloroethane
Trichloroethane
1, 1, 1 -Trichloroethane
1,3- Trichloropropene
Benzene
Chlorobenzene
Toluene
Ethylbenzene
Tap
n=2
WO
0
0
0
0
0
0
0
0
0
0
0
Influent
n=13
WO
62
0
0
0
8
0
8
0
0
85
15
Effluent
n = 13
100
62
0
15
8
8
8
0
IS
0
85
23
Scum
n=2
WO
0
50
100
8
50
0
0
100
0
50
100
Sludge
n=2
WO
0
0
50
0
0
0
0
50
50
0
WO
Well
n = 1
100
0
0
0
0
0
0
0
0
0
100
0
presence of a thick scum layer and dur-
ing subsurface drainage and migration.
Some biodegradation of several organ-
osulfurs and hydrocarbons is likely to
occur during soil migration, especially
since aerobic conditions generally pre-
vail. However, the chlorinated organics
and most of the branched or cyclic hy-
drocarbons are less likely to be de-
graded by bacterial action and are ex-
pected to migrate considerable dis-
tances in the soil. A related study noted
substantial migration of low-molecular-
weight chlorinated solvents away from
a landfill in Delaware.
The full report was submitted in fulfill-
ment of Grant No. R806102 by the Uni-
versity of Washington under the spon-
sorship of the U.S. Environmental
Protection Agency.
roethene was generally low during the
week, but it reached a maximum on
Monday (8 p,g/L in the influent and 1
|j,g/L in the effluent). Benzene was de-
tected only on Wednesday.
A log-normal frequency distribution
graph of the aromatic compounds
showed the presence of benzene only in
the scum and sludge layer. Although
benzene was not detected in the influ-
ent, it was detected on two occasions in
the effluent, possibly as a result of dis-
charged solids from the scum or sludge
layer. Toluene showed no removal in
the septic tank and very little accumula-
tion in the scum and sludge layer. Low
removals were also noted for the chlori-
nated compounds.
The generally low removals of the
volatile organics was further reflected
by the similarity of the reconstructed
total ion current of the volatile organics
in the influent and effluent of the septic
tank. A tabulation of the 48 compounds
(Table 2) shows that the majority are
hydrocarbons, including both aliphatic
and cyclic structures. Several com-
pounds reflect the presence of anaero-
bic degradation processes occurring in
the sewer line or septic tank. High con-
centrations were noted for 2-propanone,
2-ethyl-4-methyl-1-pentanol, and 4-
methyl-2-propyl-1-pentanolall likely
originating from anaerobic decomposi-
tion processes. Large increases were
ilso noted for several biogenic organ-
osulfur compounds such as carbon di-
sulfide, methanethiol (methylmercap-
tan), dimethyl disulfide, and dimethyl
trisulfide. The largest increase was
noted for methanethiol. Small increases
were noted for compounds with larger
molecular weights, probably reflecting
the greater difficulty with which bacte-
ria generate these larger compounds.
The low-molecular-weight alkylated
benzenes show a significant removal,
probably because of biodegradation;
but the results show no removal for
higher-molecular-weight compounds.
The hydrocarbons showed the .highest
removals at intermediate molecular
weight and lower removals at larger
molecular weights, probably because of
reduced volatilization. The increase
noted for several of the low-molecular-
weight hydrocarbons may be a result of
their formation as intermediates in the
breakdown of larger-molecular-weight
hydrocarbons.
Conclusions
The present study has important im-
plications for assessing the environ-
mental impact of septic tanks. Since lit-
tle removal of the volatile priority
pollutants occurs, these compounds
will be discharged through the subsur-
face drainfield and may enter the
groundwater. Volatilization is a removal
mechanism in the septic tank, but this
mechanism will not be effective in the
-------
Table 2. Volatile Compounds Found in Septic Tank (Sunday)
ng/L in
Volatile Compounds Scan Influent
Methane, dichlorodifluoro
Methanethiol
2-Propanone
Methane, thiobis
Unknown hydrocarbon
Carbon disulfide + dichloromethane
Cj hydrocarbon
Hexane, 3-methyl
Heptane
Disulfide, dimethyl
Benzene, methyl
Hexane, 2,5-dimethyl
Heptane, 3-methyl
Cyclohexane, 1,3-dimethyl, cis
Cyclohexane, 1,3-dimethyl, trans
Cyclohexane, 1-ethyl-2-methyl
Heptane, 2,4-dimethyl
C;0 cyclic hydrocarbon
Pentane, 2,2,3,4-tetramethyl
Trisulfide, dimethyl
Heptane, 6,6-dimethyl-2-methylene
Hexane, 2,2,5,5-tetramethyl
Branched C10 hydrocarbon
Nonene, 4,6,8-tri methyl
Hexane, 3,3,4-trimethyl
Hexane, 2,4-dimethyl
Benzene, 1,4-dichloro
Pentane, 2,2,3-trimethyl
Cw cyclic hydrocarbon
Benzene, 1-methyl-4 (1-methylethyl)
Heptane, 2,2,4,6,6-pentamethyl
Cyclohexane, 1-methyl-4-{ 1-methylethyl)
Cjo hydrocarbon
Hexane, 2,2,5-trimethyl
Hexane, 3,3-dimethyl
Hexane, 2,2,3-trimethyl
1,4-cyclohexadiene,
l-methyl-411-methylethyl)
Butane 2,2,3-trimethyl
1-pentanol, 2-ethyl-4-methyl
Pentane, 2,3,4-trimethyl
Pentane, 2,2,4,4-tetramethyl
Cyclohexane, 1-methyl-4-(1-methylethylidene
1 pentanol, 4-methyl-2-propyl
Hexane 2,2,3-trimethyl
Cyclohexane (1-methylethyl)
Bicycloheptane 3,7,7-trimethyl
Heptane, 5-ethyl-2-methyl
Benzene, 1,2,3-trichloro
Total '
1380
7407
7475
1419
1422
1536
1558
1612
1754
1842
1843
1875
1883
1890
1971
2009
2947
2960
2997
3000
3044
3074
3091
3124
3149
3157
3179
3211
3221
3229
3238
3240
3252
3326
3346
3349
3367
3408
3415
3453
3500
3583
3650
3717
0.64
<0.7
78.2
23.0
2.0
4.2
73.0
8.1
6.2
11.6
74.4
14.9
5.3
5.3
1.2
1.0
15.3
<0.7
9.7
11.4
6.2
8.0
3.7
3.4
3.2
8.3
16.7
2.5
7.2
15.4
17.1
126
3.6
7.2
38.7
45.8
3.9
3.9
21.2
11.9
4.0
9.2
13.5
5.9
<0.7
5.3
5.7
<0.7
623
Scan
7395
1410
1418
1421
1424
1527
7729
7825
2770
2947
3734
3768
3205
3272
3227
3237
3242
3378
3362
3402
3470
3449
3497
3579
3606
3652
3714
3745
\j.g/L in
Effluent
<0.7
128
70.3
84.4
4.2
10.0
13.6
<0.7
<0.7
29.7
16.7
<0.7
<0.7
<0.7
<0.7
<0.7
<0.7
3.3
<0.7
72.7
<0.7
<0.7
<0.7
<0.7
<0.7
<0.7
77.5
<0.7
25.7
77.6
7.2
707
<0.7
75.5
27.7
24.7
<0.7
<0.7
73.6
7.6
7.8
75.7
70.0
3.4
7.9
47.6
37.7
2.7
793
-------
Foppe B. DeWalle, David A. Kalman. and Donald Norman are with University of
Washington, Seattle. WA 98195; and Gary Plews is with Department of Social
and Health Services, Olympia, WA 98504.
Ronald F. Lewis is the EPA Project Officer (see below).
The complete report, entitled "Determination of Toxic Chemicals in Effluent from
Household Septic Tanks," (Order No. PB 85-196 798/AS; Cost: $8.50. 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:
Water Engineering Research Laboratory
U.S. Environmental Protection Agency
Cincinnati, OH 45268
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use $300
EPA/600/S2-85/050
US 6
L I = s?
------- |