&EFK
United States
Environmental Protection
Agency
Industrial Environmental Research
Laboratory
Cincinnati OH 45268
Research and Development
EPA-600/S2-80-206 Feb. 1981
Project Summary
Development of a Chemical
Toxicity Assay for Pulp Mill
Effluents
J. M. Leach and L. T. K. Chung
A chemical analysis procedure was
developed to rapidly measure com-
pounds responsible for the toxicity of
pulp mill effluents to fish. These results
were used to estimate effluent toxicity
as measured by standard laboratory bio-
assays, and to determine loadings of
various toxic compounds from all the
pulp mills on two river systems.
The analytical procedure involved
adsorption of toxic compounds from
effluent onto Rohm and Haas Amber-
lite XAD-2 polymeric resin, elution of
adsorbed constituents from the resin
with ether and methanol, methylation
of the extract, and quantitative analysis
by glass capillary column gas chroma-
tography. Analysis time was less than
one and one-half hours.
Results for 113 samples of raw and
biologically-treated bleached and un-
bleached kraft, sulphite and ground-
wood effluents were converted via
toxic units into estimates of acute lethal
toxicity. Agreement between 96-h me-
dian lethal concentrations calculated
by this method and values from bio-
assays of the effluents using rainbow
trout was within ± 30% for 73% of the
samples examined. Calculated 96-h
LCBO's and those measured by bioas-
jsay differed by more than 30% for the
other 27% of samples examined. At
present, therefore, toxicity determined
from chemical analysis can be a valu-
able supplement to, but not a substitute
for, direct measurement by bioassy.
Samples were collected and analysed
for five consecutive days from all pulp
mills on the Willamette River, Oregon,
and the Androscoggin River flowing
through New Hampshire and Maine,
to determine loadings of known toxic
compounds into and out of the bioba-
sins. Biotreatment at most of the mills
was very effective in removing toxicants
Estimates based on theoretical dilution
capacity of the receiving waters showed
that secondary treatment was highly
beneficial in protecting the rivers against
the toxic effects of pulp mill effluents.
Minimum dilution of pulp mill effluents
discharged to the Willamette River
during the study period (October,
1979) was 93-fold, assuming rapid and
complete mixing. In the Androscoggin
River, the minimum effluent dilution
during the study period (February,
1980) was 17-fold.
This Project Summary was developed
by EPA's Industrial Environmental
Research Laboratory, Cincinnati, to
announce key findings of the research
project that is fully documented in a
separated report of the same title (see
Project Report ordering information at
back).
This Project Summary was devel-
oped by EPA's Industrial Environ-
mental Research Laboratory, Cincin-
nati. 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 order-
ing information at back).
-------
Introduction
In recent years, more than 30 organic
compounds have been identified as con-
tributing to the toxicity of pulp mill pro-
cess effluents to fish. Chemical analysis
procedures using gas chromatography
(GC) have been developed for toxicants
in the various waste streams, and toxicity
of the effluents to fish has been esti-
mated by summing toxic unit equivalents
of the measured toxicant concentrations.
For many of the streams studied, good
correlation has been obtained between
the toxicity calculated from results of
GC analyses and values obtained from
acute lethaf bioassays using rainbow
trout.
As an extension of the previous work,
the primary objectives of this study were
to:
A. Develop a chemical analysis pro-
cedure for rapid measurement of
the concentrations of toxic materi-
als in bleached and unbleached
kraft whole mill, groundwood, and
sulfite mill effluents.
B. Relate the chemical assay results
to effluent toxicity measured in bio-
assays using rainbow trout.
C. Measure toxicant loadings from
kraft whole mill effluents.
D. Measure toxicant loadings from
groundwood and sulfite mills.
E. Make a preliminary assessment of
the environmental significance of
mill biotreatment systems for re-
ducing toxic loadings in effluents.
F. Measure toxicant loadings from all
mills on selected river systems.
The work was carried out in two phases;
(I) development and evaluation of an an-
alytical procedure for predicting effluent
toxicity (objectives A, B); (II) use of the
procedure for estimating toxicant loadings
(objectives C-F).
Conclusions
Phase I
The sample (20-50 ml of untreated
effluent; 100 ml of biotreated effluent)
adjusted to pH 9 with 10% NaOH solution,
and to a specific conductance greater
than 2 mmho/cm using saturated NaCI
solution, was passed from a 250-ml sep-
aratory funnel at 15 ml/min through a
glass column (1.6 x 22 cm) containing
Amberlite XAD-2 resin (12ml), previously
rinsed with deionized water (3x15 ml).
The funnel was rinsed once with a small
amount of NaCI solution at pH 9, and
residual effluent was expressed from the
column using a stream of nitrogen. Ad-
sorbed constituents were extracted by
allowing 9:1 ether/methanol (15 ml) to
stand in contact with the resin for 10
min before eluting into a 100-ml glass
centrifuge tube at a flow rate of 15 ml/
min. Additional ether/methanol (2x10
ml) was then passed through the resin
to complete the extraction of adsorbed
toxicants. Effluent trapped by the resin
and eluted during passage of solvent was
separated from the solvent layer by im-
mersing the centrifuge tube in an ace-
tone/dry-ice bath for 2 min. The organic
layer was decanted from the frozen aque-
ous layer into a 100-ml round-bottom
flask. The ice was thawed, washed with
diethyl ether (2x5 ml), frozen again, and
the organic extracts were combined.
Heptadecanoic acid (1 ml of a solution
containing 0.05 mg/ml in methanol) was
added to the solvent extract as an inter-
nal standard. The solvent was evaporated
almost to dryness using a rotary evapo-
rator, transferred using ether (6 ml) to a
15-ml graduated centrifuge tube, and
concentrated to 2 ml by gentle warming
on a water bath. Methanol (0.2 ml) was
added and the solution was methylated
with diazomethane for 10 min. The solu-
tion was evaporated to 0.1 ml on a warm-
water bath, and aliquots (2-5 /J\) were
analysed by capillary column GC using a
Hewlett-Packard Model 5830 FID instru-
ment. Column specifications were: 16
m glass (0.25 mm i.d.) WCOT OV-101;
initial temperature, 190°C, programmed
at 1 °C/min (20 min), 2°C/min (10 min),
then 3°C/min to 250°C. Toxic compounds
of interest were identified routinely by
matching retention times with those of
standards using a Hewlett-Packard Model
18850A integrator terminal with settings
of attenuation 16, slope sensitivity 0.1,
area reject 100, and retention time win-
dow 2%. Identities of the constituents
were authenticated in some samples by
combination gas chromatography-mass
spectrometry and in others by confirming
that retention times coincided in samples
"fortified" with added amounts of the
pure compounds. Chlorinated guaiacol
concentrations were checked by electron-
capture GC (Hewlett-Packard Model
5710 instrument containing a glass col-
umn, 0.32 cm x 1.8 m, packed with 8%
OV-1 on Chrom W (HP), 80-100 mesh;
temperature, 175°C; carrier gas 95%
argon, 5% methane; flow rate 20 ml/min)
using aldrin as an internal standard.
Samples from each mill were analysed
before methylation to detect neutral com-
pounds. Analysis time, excluding this last
step, was 1 h 25 min.
Recovery efficiency was measured
using portions of the effluents to which
known amounts of the toxic constituents
had been added. Aliquots (4 or 8 ml) of
stock solutions containing each toxicant
in methanol (0.25 mg/ml) were mixed
and diluted to 20 ml with methanol to
give a solution containing 0.05 or 0.10
mg/ml of each compound. An aliquot (1
ml) of this solution was added to the
effluent sample 20-100 ml), which was
then processed through XAD-2 resin.
Recovery efficiencies were in the range
of 80-103% for toxicants added to effluent
samples, reflecting a level of accuracy
compatible with other complex environ-
mental analyses of organic constituents.
Analytical conditions were varied dur-
ing development of the procedure to opti-
mize accuracy and minimize analysis
time for aqueous standard solutions of
toxic constituents and for effluent sam-
ples. Effects of the following variables
on analytical accuracy and precision
were studied:
1. Ionic strength and pH of toxicant
solution or effluent.
2. Use of ether or ether/methanol
for elution of adsorbed materials
from XAD-2 resin.
3. Solvent evaporation technique.
4. Drying technique.
5. Internal standard for GC analysis.
6. GC conditions. (
7. Variation of GC retention time and
response factor with time and con-
centration.
Analytical procedures were evaluated
for mixtures of the following compounds,
which have been identified as important
toxic constituents of various pulp mill
process streams.
abietic acid
dehydroabietic acid
isopimaric acid
pimaric acid
sandaracopimaric acid
oleic acid
linoleic acid
linolenic acid
juvabione
juvabiol
dihydrojuvabione
pimarol
isopimarol
trichloroguaiacol
tetrachloroguaiacol
monochlorodehydroabietic acid
dechlorodehydroabietic acid
epoxystearic acid
dichlorostearic acid
Primary- and secondary-treated
bleached and unbleached softwood kraft
and sulphite whole mill effluents from
26 mills (Table 1) were then analysed^
-------
Table 1. Mills from which Samples were Obtained for Bioassay and Analysis
Code
Mill Location
Type
Effluent Sample
A British Columbia Coastal
B Oregon
C British Columbia Interior
D British Columbia Interior
E British Columbia Interior
F British Columbia Interior
G Mid-West USA
H East Coast
I Ontario
J Ontario
K Mar/times Canada
L N. Carolina
M British Columbia Interior
N Washington
0 Mid-West USA
P Washington
0. Oregon
R Washington
S Western Canada
T Washington
U Oregon
V Oregon
W Oregon
X Washington
Y Oregon
Z Washington
Unbleached softwood kraft
Bleached softwood kraft
Bleached softwood kraft
Bleached softwood kraft
Bleached softwood kraft
Bleached and unbleached
softwood kraft
Bleached hardwood kraft
Unbleached low-yield
sodium-base sulphite
(softwood)
Unbleached low-yield
sodium-base sulphite
(softwood/
Bleached calcium/magne-
sium-base sulphite (softwood)
Bleached ammonium-base
sulphite (softwood)
Bleached softwood and hard-
wood kraft
Refiner groundwood
Bleached and unbleached
softwood kraft; NSSC
Bleached hardwood kraft;
hardwood NSSC
Bleached calcium-base
sulphite (softwood)
Bleached ammonium-base
sulphite (softwood)
Bleached ammonium-base
sulphite (softwood)
Unbleached softwood kraft
Refiner groundwood
(softwood)
Unbleached softwood kraft;
NSSC
Unbleached magnesium-base
sulphite; refiner groundwood
(softwood)
Unbleached softwood kraft
Refiner groundwood
Bleached softwood and
hardwood kraft
Bleached softwood kraft and
magnesium-base sulphite
Primary
Primary
Primary
Primary, secondary
Primary, secondary
Primary, secondary
Secondary
Primary
Primary
Primary
Primary,
Primary,
Primary,
Primary,
Primary,
Primary,
Primary.
Primary
secondary
secondary
secondary
secondary
secondary
secondary
secondary
Primary
Primary, secondary
Primary, secondary
Primary, secondary
Primary, secondary
Primary, secondary
Primary, secondary
^^aci
Toxicity of the effluents to rainbow trout-
were measured by bioassay, and the re-
sults were similar to values calculated
from concentrations of the individual
toxicants, using the toxic unit concept.
Agreement between 96-h LCBO's esti-
mated from analysis results and the
values from bioassays was within 30%
for 73% of 113 effluent samples exam-
ined (Figure 1). For approximately 10%
of the samples, analysis results predicted
a high toxicity value whereas bioassay
indicated low toxicity, or vice versa. Resin
iids were the predominant toxic constit-
uents in primary-treated effluent samples
from mills operating on softwood species.
Levels of these compounds were gener-
ally, though not invariably, a good indi-
cator of the effluent toxicity. None of the
secondary-treated effluent samples ex-
amined from mills in the United States
were toxic to rainbow trout.
Phase II
The analysis procedure was then used
to measure the concentrations of poten-
tially toxic compounds in the influents to
and effluents from biotreatment ponds
at all mills on the Willamette River, Ore-
gon, and the Androscoggin River, New
Hampshire and Maine (Table 2). The re-
sults were used to assess, in a prelimi-
nary way, the environmental significance
of mill biotreatment systems on the two
rivers. Retention times in the various
biobasins were from 6 hours to 14 days.
All except one mill on the Willamette
River and one on the Androscoggin River
provided excellent removal of the toxic
compounds analysed. Calculated total
daily loadings discharged from all mills
to the Willamette River were 2 kg chlori-
nated guaiacols, 52 kg Ci8 unsaturated
fatty acids, and 71 kg resin acids. Approx-
imately 20% of the fatty acids and 67%
of the resin acids originated from one
mill. Secondary treatment systems at the
remaining mills removed 60-90% of un-
saturated fatty acids and 85-99+% of re-
sin acids, based on loadings in influents
and effluents.
Calculated total daily loadings dis-
charged from all mills to the Androscog-
gin River were 55 kg Cie unsaturated
fatty acids, 115 mg resin acids, 21 kg tri-
and tetra-chlorinated guaiacols, and 2.7
kg chlorinated dehydroabietic acids. One
mill accounted for 56% of the resin acid
loading to the river. Secondary treatment
at mills on the Androscoggin River re-
moved 91 -96% of the unsaturated fatty
acids, 92-97% of resin acids (except at
one mill where resin acid removal was
80%), and amounts of the chlorinated
guaiacols and chlorinated resin acids
ranging from 50-90%.
Table 2. Mills Sampled for Toxicant
Loadings Studies
Willamette River, Oregon
Unbleached kraft
Unbleached magnesium-base sulfite,
refiner groundwood, and
de-inking (2 mills)
Stone groundwood
Unbleached kraft/NSSC
Bleached kraft
Defibrated wood pulp
Bleached ammonium-base sulfite
—all softwood mills
Androscoggin River. New Hampshire.
Maine
Bleached softwood and hardwood
kraft
Bleached softwood and hardwood
kraft, groundwood
Refiner groundwood
Bleached softwood and hardwood
kraft, NSSC
-------
/ I/IS
80
60
40
_^^
Q)
| 20
o
•Q
1 10
C f T°0
FofW <"
- |i™
I *
-p
o
e
*
i
"" i \
Oy Q9999
O o O
„
il T ,
T ' X
° *«
O 1
f
T
,{°
!
0 to
1
X
, ,
x «•
O
r
o
o
it
6°
V
0 ' 1
fii 1 °o
0 _
o° 0°
X °
0 0
o
]
°o0 (l
X
$
KEY
96-h LC50 and
l confidence limits
' 6/ bioassay
0 96-hLC50
estimated from
analysis
.
Softwoods Hard- \ Primary- Secondary-Primary- Secondary-
woods i treated treated treated treated
Primary-treated unbleached \ Sulphite effluents \ Groundwoodt Biotreated bleached kraft effluents
and bleached kraft effluents I effluents '
Figure 1. Comparison of calculated and measured effluent toxicities.
Recommendations
Reasons for the lack of agreement in
some cases between calculated and mea-
sured toxicity should be investigated. As
a first step, the toxic components should
be identified in effluents from mills that
use hemlock, cedar, or hardwoods.
Since most research in this area to
date, has been carried out with cold water
fish species, a similar study comparing
calculated and measured toxicity using
warm water test fish species would be
of relevance and value to mills in the
south.
Consideration should be given to using
"standard" 96-h LC50 values of the indi-
vidual toxicants to calculate the toxicity
of effluents to fish species. This would
avoid problems of variations in bioassay
results caused by the effects of seasonal
and genetic differences on sensitivity be-
tween different stocks of the same spe-
cies.
This study was concerned with acute
lethal toxicity. Pulp mill effluents cause
detectable sublethal effects in fish in the
laboratory at concentrations of 0.1 -0.2
of the 96-h LC50. Studies should be un-
dertaken to determine the degree of efflu-
ent treatment necessary to ensure that
concentrations of toxicants in receiving
waters are below these levels.
-------
J. M. Leach andL. T. K. Chung are with B.C. Research, Vancouver, B.C., Canada
V6S 2L2.
D. L. Wilson and M. R. Strutz are the EPA Project Officers (see below).
The complete report, entitled "Development of a Chemical Toxicity Assay for
Pulp Mill Effluents," (Order No. PB 81-126 369; Cost: $11.00, 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 Officers can be contacted at:
Industrial Environmental Research Laboratory
U.S. Environmental Protection Agency
Cincinnati, OH 45268
« US. GOVERNMENT PRINTING OFFICE: 1M1 -757-064/0262
-------
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Agency
EPA 335
Official Business
Penalty for Private Use $300
T ££RL0169064
1
------- |