UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, O.C. 20460
2-1-91
OFFICE OF
PESTICDES AND TOXIC
SUBSTANCES
MEMORANDUM
SUBJECT: Environment
FROM:
lymeri
TO:
J. Vincent Nabnolz, Ph'.D./3£hd Maurice G.
Senior Scientist and Chi&T, respectively
Environmental Effects Branch
Health and Environmental
Review Division (TS-796)
Mary E. Cushmac
New Chemicals Branch
Chemical Control Division (TS-794)
eeman, Ph.D.
As per your request, this is the current status of the
environmental concerns of polymers (MW > 1000). Vince Nabholz,
EEB, prepared this response.
ENVIRONMENTAL CONCERNS OF POLYMERS
All polymers are divided into four classes depending on the
type of electronic charge of the polymer: nonionic (neutral);
anionic (negative charge); cationic (positive charge); and
amphoteric (mixture of positive and negative charges on same
molecule) polymers.
I. POLYMERS WITH MW < 1000.
Polymers with MW < 1000 and some water solubility may be of
concern because of their potential to act like polymers whose MW
> 1000 and of their potential to be absorbed through biological
membranes and cause systemic effects.
A. Polynonionic (neutral) polymers.
Small neutral polymers are generally assessed based on the
type of functional group, e.g, aniline, phenol, alcohol, epoxide,
etc. Polynonionic polymers are evaluated on the basis of their
octanol/water partition coefficients (K^ or P), melting point
(mp), water solubility (SH20), and predicted toxicity to aquatic
Page 1 of 9
Printed on Recycled Paper
-------
organisms using structure activity relationships (SAR, e.g.,
quantitative structure activity relationships [QSAR]) just as you
would assess a monomer with the same functional group. For
example, polyphenols are assessed as you would assess a phenolic
monomer using the phenol SARs (Clements 1988).
B. Polyanionic (negatively charged1 polymers.
Small polyanionic polymers are assessed by using the nearest
analog method. The chemical structure of the anionic group(s),
e.g., carboxylic acid, phosphoric acid, sulfonic acid, is
compared to analogous larger polymers with significant amounts of
low molecular weight (LMW) components or polyanionic monomers
(e.g., EDTA) for which there is environmental test data. In
general, the concerns for these small polymers are the same as
for the large polymers whose MW > 1000 with the additional
concern for the potential absorption and subsequent systemic
toxicity.
C. Polycationic (positive charged) polymers.
Polycationic polymers are assessed by using either the SARs
for polycationic polymers (Clements 1988), the aliphatic amine
SARs, the SARs for quaternary ammonium surfactants (Clements
1988), or the generic review of small quaternary ammonium (non-
surfactant) compounds.
D. Polyamphoteric (polymers containing cationic and anionic
charges withinthe same molecule) polymers.
Small polyamphoteric polymers with equal numbers of cationic
charges and anionic charges, or with greater cations than anions
are treated as a polycationic compound. The toxicity of the
polycationic portion of the compound is reduced based on the
number of anionic charges. When there are greater numbers of
anions than cations, the compound is treated as a polyanionic
polymer.
II. POLYMERS WITH MW > 1000.
Hazardous polymers with MW > 1000 are expected to be water
soluble (or self-dispersing), are not expected to be absorbed
through biological membranes, and are expected to assert their
toxicity by affecting the outer membranes of aquatic organisms or
the near environment of the organism (e.g., over-chelation of
nutrient elements). Insoluble polymers are not expected to be
toxic unless they are ground up into fine particles. The
toxicity of finely ground particles is due to indirect (physical)
toxicity (e.g., the clogging of respiratory organs such as gills)
and only occurs at high concentrations, i.e., acute toxicity
Page 2 of 9
-------
values of greater than 1000.0 mg/L and chronic toxicity values of
greater than 50.0 mg/L. The toxicity of finely ground insoluble
polymers does not depend upon the chemical structure of the
polymer.
A. Konioi\ic (neutral 1 polymers.
Polynonionic polymers which have MW > 1000 are of low
concern.
B. Anionic (negatively charged) polymers.
Polyanionic polymers which have MW > 1000 and which are
water soluble (miscible or self-dispersing) are of concern for
aquatic toxicity. Polyanionic polymers are divided into three
subclasses: poly(carboxylie acids), poly(aromatic sulfonates),
and poly(aliphatic sulfonates).
1. Poly(carboxylie acidsl are of concern only for their
toxicity to green algae. Toxicity to algae as defined by the 96-
h EC50 for growth inhibition, is moderate with toxicity values
ranging from 1 to 100 mg/L (ppm). It appears that the mode of
toxic action of these poly(carboxylie acids) is over-chelation of
nutrient elements needed by algae for growth. When enough
calcium (as divalent cation) is added to a polymer to satisfy its
anionic charges, toxicity to algae is mitigated. It is unknown
if calcium (as calcium carbonate in water with hardness of 100 to
150 mg/L as CaC03) added to algal growth medium will also
mitigate toxicity to an equal degree.
a. Structural requirements. Poly(acrylic acid) is
moderately toxic to green algae and appears to be the most potent
form of poly(carboxylie acid) in its ability to chelate nutrient
elements. It's chemical structure is ~[CC(COOH)]— where you
have a carboxylic acid on every other (or alternating) carbon(s)
in the polymer backbone. The carboxylic acids are paired and
equal distance from the polymer backbone. Test data for
poly(maleic acid) indicated low toxicity to algae, i.e., 96-h
EC50 = 560.0 mg/L, and, thus, a weak ability to chelate nutrient
elements. In this polymer, there is a carboxylic acid on every
carbon of the polymer backbone, i.e.,
—[C(COOH)C(COOH)]—. Additional PMN test data have suggested
that (1) when the carboxylic acids are further separated, e.g., a
carboxylic acid on every fifth carbon of the polymer backbone, or
(2) when the carboxylic acids are different distances from the
polymer backbone, the polymer's ability to chelate nutrient
elements is reduced. The test data to support these last two
conclusions are weak. In summary, the most potent structure for
poly(carboxylic acid) polymers is paired acids which are equal
distant from the polymer backbone and which have one acid on
alternating carbons.
Page 3 of 9
-------
b. Mitigation of toxicitv. The toxicity of poly(acrylic
acid) polymers has been shown to be mitigated 13 times by the
addition of one equivalent of Ca to the polymer before testing.
PMN test data have indicated that the 96-h EC50 increased from
37.4 mg/L to 500.0 mg/L if you chelate the polymer with divalent
ions before exposure to algae. This indirect toxicity to algae
via over-chelation of nutrient elements may be an laboratory
artifact because (1) many poly(carboxylic acid) polymers are used
a scale inhibitors and are released to the natural environmental
chelated with Ca and Mg, (2) these polymers are initially tested
as the Na or K salt, and (3) the OTS Environmental Test Guideline
recommends a growth medium which has a hardness of only about
15.0 mg/L as CaCO3. This represents very soft water and the
average hardness of freshwater in the United States is about
120.0 mg/L. In those cases where the polymer is not used as a
scale inhibitor and is released to the environment as the Na or K
salt, the hardness of the receiving waters (i.e., 120.0 to 150.0
mg/L) may cause a substantial mitigation of the toxicity relative
to the toxicity observed in the standard algal toxicity test.
However, there are not sufficient test data to demonstrate the
amount of mitigation due to moderately hard growth medium.
c. Testing scheme. (1) Polymers used as scale inhibitors
and released to the environment chelated with calcium and
magnesium ions should be tested three times with freshwater green
algae: (a) testjchemical as is, (b) test chemical with an
equivalent of Ca** added to the stock solution, and (c) test
chemical as is but tested with modified algal test/growth medium.
Calcium alone or Ca and Mg is added to attain a measured hardness
of about 150.0 mg/L as CaCO3. If Ca and Mg are added together,
then add Ca and Mg in the ratio of 2 Ca to every Mg. Test
results from the test with an equivalent of Ca added to the
polymers will be used to assess releases from use. Test results
with the polymer as the Na or K salt will be used to assess
releases from manufacturing and processing. All stock solutions
should be adjusted to pH 7 before testing because PMN test data
have shown that, if the polymer is tested as the acid or with
excess acid, the toxicity from the acid was greater than the
toxicity of the polymer via over-chelation. (2) Polymers not
used as scale inhibitors and are released to the environment as
produced (generally as the Na and K salts) should be tested twice
with green algae: (a) the polymer as is, and (b) the polymer
with modified algal test/growth medium.
2. Polv(aromatic sulfonatesl. a. Polymers showing
toxicity. Poly(aromatic sulfonate) polymers with MW greater than
1000 may be of moderate concern for acute toxicity towards fish
and green algae. Polymers in this class have the following
characteristic monomers: sulfonated phenols, sulfonated cresols,
sulfonated diphenolsulfones, sulfonated diphenyloxides, and
sulfonated diphenylsulfones. This concern is based on two facts.
Page 4 of 9
-------
The first is test data for 8 PMN polymers which indicated acute
toxicity values of about 5.0 mg/L for fish (as a 96-h LC50) and
20.0 mg/L for algae (as a 96-h EC50). This evidence is weak
because while these polymers have average number MWs equal to or
greater than 1000, 1;hey has substantial amounts of low molecular
weight (LMW) components: > 10% < 500 and > 25% < 1000. The
observed toxicity could have been due to the LMW components.
However, the second fact is that the Agency has recently received
data for a PMN polyner which had a MW = 12,200, <0.1% <500, and
<0.1% <1000, contained a carboxylic acid substituted
diphenolsulfone, and was moderately toxic to aquatic organisms,
i.e., fish 96-h LC50 = 72.0 mg/L, daphnid 48-h LC50 - 86.0 mg/L,
and green algal 96-h EC50 » 40.0 mg/L (hardness [H] of medium »
18.0 mg/L as CaC03) . This polymer was tested three additional
times with harder growth medium. There was no significant
mitigation of the toxicity, i.e., H = 46 mg/L, EC50 » 24.0 mg/L,
H = 152 mg/L, EC50 « 20 mg/L, and H = 160 mg/L, EC50 =47.0 mg/L.
The only common monomer between these two sets of polymers was
the acid substituted diphenolsulfone.
b. Polymers showing low toxicitv. Poly(aromatic sulfonate)
polymers which have been shown to have low toxicity (i.e., acute
toxicity values gresiter than 100.0 mg/L) or are highly suspected
of having low toxicity are composed of the following monomers:
benzene sulfonates and sulfonated naphthalene.
3. Polvfalipheitic sulfonates) . There are not enough test
data for these polymers to draw any firm conclusions about their
toxicity. However, it is suspected that if these polymers show
toxicity to aquatic organisms it will be to algae as was observed
for the poly(carboxylic acid) polymers.
C. Polvcationic (positively charged) polymers.
Polycationic polymers include polyamines (primary amines,
secondary amines, and tertiary amines); quaternary amines';
polysulfoniums; and polyphosphoniums. Polymers of concern have
MWs >1000 and are water soluble (miscible or self-dispersing).
Polymers based on polyglucosamines (i.e., chitosan) are much less
toxic than predicted, and are no longer of concern.
1. Toxicity. Aquatic toxicity in clean water (i.e., total
organic carbon [TOG] < 2 mg/L) increases exponentially with
increasing cationic charge density, i.e., protonated and/or
quaternarized-N, S or P. An SAR for polycationic polymers has
been published by Clements (1988). Charge density is measured as
per cent amine-N for nitrogen-based polymers; equivalent weight
of N, S, or P; or % cations/1000 MW. Toxicity to aquatic
organisms increases exponentially until about 2.5 cations/1000 MW
(or 3.5% amine-nitrcgen or an equivalent weight = 400),
thereafter, toxicity becomes asymptotic. Acute toxicity values
Page 5 of 9
-------
to fish and daphnids (i.e., aquatic invertebrates) are > 100 mg/L
[i.e., low concern] at < 0.2 cations/1000 MW (or equivalent
weight >5000 or <0.3% amine-nitrogen); 100 to 1 mg/L [i.e.,
moderate concern] at charge densities of >0.2 to 1.6 cations/1000
MW (or equivalent weights between 5000 and 625, or percent amine-
nitrogen between 0.3 and 2.2); and < 1 mg/L [i.e., high concern]
at > 1.6 cations/1000 MW (or equivalent weights < 625 or percent
amine-nitrogen > 2.3). Green algae are about 6 times more
sensitive than fish (i.e., algal 96-h EC50 versus fish 96-h
LC50). The mode of toxic action for these polymers is surface
active (i.e., they react with biological membranes), however,
when MW falls below 1000, some systemic toxicity may also occur.
2. Mitigation of toxicity. The aquatic toxicity of these
polymers with MW > 1000, < 10% <500, and < 25% <1000 is highly
mitigated by the presence of dissolved organic carbon (DOC) in
water. For polymers with charge densities »> 2.4 cations/1000 MW
(or equivalent weight < 400 or > 3.3% amine-nitrogen), the acute
toxicity to fish is reduced about 94 times when the measured TOC
in water is 10 mg/L as the result of adding humic acid to
dilution water (i.e., TOC = 10 mg/L is equivalent to 27.6 mg/L
humic acid, sodium salt; CASRN [1415-93-6]; Aldrich HI,675-2;
Merck Index 10,4649). This mitigation factor of 94 is based on
the results of testing 8 polymers.
The toxicity of polymers with MW > 1000 and significant
amounts of LMW material, i.e., > 10% < 500 and > 25% < 1000 is
mitigated less by dissolved organic matter in the water column.
Two polymers with significant amounts of LMW material (i.e., one
with MW = 1000, 23.3% <500, and 32.3% <1000; and another with MW
= 1030, 12% <500, and 38% <1000) had mitigation factors of only
26 and 21, respectively.
The aquatic toxicity of polymers with charge densities =>
2.4 cations/1000 MW (or equivalent weights <400 or >3.3% amine-
nitrogen) are now of low concern for aquatic organisms living in
the water column of the aquatic environment because of the
predicted low risk.
There is one known exception to the mitigation/low
bioavailability scenario for these polymers. When these polymers
are formulated with excess acid, e.g., 20% excess acid more than
you need to protonate all of the amines in a polymer; and the
product pH is about 2, these polymers will fail to flocculate DOC
in the water column and DOC does not appear to mitigate the acute
toxicity to fish.
3. Partitioning to sediments. Many polycationic polymers
are designed to react with DOC in the water column to form an
insoluble flocculent. This flocculent eventually settles on
sediments and accumulates in sediment. Sediment toxicity testing
Page 6 of 9
-------
with species which ingest sediment has shown that polycationic
polymers with charge densities of => 3.0 cations/1000 MW (or =>
3.0 cations/1000 MW or an equivalent weight =< 333.0) are not
bioavailable to cause any toxicity and are thus of low concern in
sediments (Rogers and Witt (1989).
4. Selection of humic acid for mitigation testing. Humic
acid was selected as the representative dissolved organic carbon
(DOC) based on research done by Gary et al (1987). Gary et al
measured the mitigation of 4 suspended solids and five dissolved
organic carbon (DOC) compounds on the acute toxicity of four
cationic polyelectrolytes to freshwater fish and aquatic
invertebrates. Humic acid was about average in its ability to
mitigate toxicity. Analysis of Table 4 in Gary et al indicated
that the mitigation factors (HF) for humic acid were closest to
the mean MF for all of the DOCs tested. The mean MF factor was
calculated for each of the polymer/species combinations. The MF
of each DOC was compared to the mean MF and the absolute value of
the difference was averaged for that DOC. Humic acid had the
lowest average difference or, in other words, the MFs for humic
acid were closest to the mean MF for each polymer/species
combination, i.e., lignin > tannic acid > fulvic acid > lignosite
> humic acid. In addition, humic acid is easily available from
chemical supply companies.
5. Selection of 10 ma TOC/L to set the mitigation factor.
Ten mg TOC/L has been used in Agency hazard and risk
assessments for three reasons: (a) concentrations of humic acid
in natural waters are rarely measured, (b) the average measured
amount of TOC in natural freshwater of the US is about 6.79 mg
TOC/L (Lynch 1987), and (c) 10 mg TOC/L is a round number close
to 7 which errs on the side of safety. Lynch (1987) analyzed
the EPA Office of Water's STORET Data Base for measured amounts
of TOC in US waters. Lynch found 67,994 measurements of TOC
taken from 1977 through 1987 from all over the US (i.e., 19 of 23
major river basins in the US). These TOC measurements were
lognormally distributed and skewed toward the larger amounts of
TOC. The geometric mean of these data was 6.79 mg TOC/L. Since
the Agency does generic risk assessments for most chemicals, at
least the first time they are assessed, it was decided to use the
average amount of TOC in natural waters as the benchmark amount
of dissolved organic carbon.
6. Testing Scheme. a. The base set of environmental
toxicity tests are done in clean dilution water. The base set of
environmental toxicity tests include: (1) the fish acute
toxicity test, the daphnid acute toxicity test, and the green
algal toxicity test. All of the above tests will be done with
the static method and will be based on nominal concentrations
corrected to 100% active ingredients. Clean dilution water is
defined as water with < 2 mg TOC/L.
Page 7 of 9
-------
b. The fish acute toxicity test will be done at least two
more times with different concentrations of humic acid (HA)
dissolved in the dilution water (CFR §795.115): the first test
will be done with 20 mg HA/L dissolved in the dilution water. If
the huroic acid and the cationic chemical forms a flocculent,
precipitate, or a viscous mixture which significantly interferes
with the behavior of the fish or causes physical toxicity (e.g.,
clogging of gills), then the concentration of humic acid will be
reduced (e.g., to 15 mg HA/L) until physical toxicity is not
significant; the second test will be done with a humic acid
concentration which is lower than the first, e.g., if the first
test is done with 20 mg/L humic acid then the second test will be
done with 10 mg HA/L.
c. Total Organic Carbon (TOC) determinations need to be
done for the clean dilution water and for each concentration of
humic acid; three TOC determinations for the clean dilution
water and three determinations for each humic acid control.
. d. If humic acid reduces the toxicity significantly, i.e,
toxicity is substantially reduced in the presence of 10 mg TOC/L
and the risk to water column organisms has been eliminated, then
the only further testing will be sediment toxicity testing.
e. Sediment toxicity testing may include (1) tadpoles
gavaged with contaminated sediment for 30 days (CFR §795.145),
(2) tadpoles exposed to contaminated sediments in same tank for
30 days (CFR §795.145), or (3) adult daphnids exposed to
contaminated sediments in same tank for 30 days (CFR §795.135).
f. If humic acid does not reduce toxicity, and there is
still a significant risk to water column organisms in the
presence of 10 mg TOC/L, then chronic toxicity testing for fish
and aquatic invertebrates will be required. I
g. The chronic toxicity test for fish (CFR §797.1600) is
the fish early life stage toxicity test using clean dilution
water, the flow-through method and nominal concentrations based
on 100% ai. The chronic toxicity test for aquatic invertebrates
(CFR §797.1350) is the daphnid partial life cycle toxicity test
using clean dilution water, flow-through method, and nominal
concentrations based on 100% ai.
D. Polyamphoteric (polymers containing cationic and anionic
charges within the same molecule) polymers.
Polyamphoteric polymers with equal numbers of cationic
charges and anionic charges or with greater cations than anions
are treated as a polycationic polymer. The toxicity of the
polycationic polymer portion of the polymer is reduced based on
the number of anionic charges. When there are greater numbers of
anions than cations, the polymer is treated as a polyanionic
polymer.
Page 8 of 9
-------
I I
III. REFERENCES.
Cary, G A, McMahon, J A, and Kuc, W J. 1987. The effect of
suspended solids and naturally occurring dissolved organics in
reducing the acute toxicities of cationic polyelectrolytes to
aquatic organisms. Environmental Toxicology and Chemistry 6:469-
474.
Clements, R G (editor). 1988. Estimating toxicity of industrial
chemicals to aquatic organisms using structure activity
relationships. Washington, DC: Environmental Effects Branch,
Health and Environmental Review Division (TS-796), Office to
Toxic Substances, United States Environmental Protection Agency.
EPA-560-6-88-001. Available from NTIS, Springfield, Virginia
22161, PB89-117592.
Lynch D G. 1987. Summary of STORET data on dissolved organic
carbon (DOC) levels in surface waters. Washington, DC: Exposure
Assessment Branch, Exposure Evaluation Division (TS-798), Office
of Toxic Substances, United States Environmental Protection
Agency. Memorandum.
Rogers, J H, Jr. and Witt, W T. 1989. Effects of sediments
flocculated with cationic polyelectrolytes when fed upon by
Dapjinia maana. Denton, TX: Department of Biological Sciences,
University of North Texas. Available from SOCHA, 1330
Connecticut Avenue, NW, Washington, DC 20036. Unpublished
manuscript.
Page 9 of 9
-------
------- |