A'"
  f i
 -'.' '    I
                     ENVIRONMENTAL ASSESSMENT OF POLYMERS
                    UNDER THE U.S. TOXIC SUBSTANCES CONTROL
                                          ACT
                       Robert S. Boethling1* and J. Vincent Nabholz2
       1 Chief, Environmental Fate Section, Exposure Assessment Branch, Economics, Exposure and
        Technology Division-7406; and 'Senior Biologist, Environmental Effects Branch, Health and
           Environmental Review Division-7403, Office of Pollution Prevention and Toxics, U.S.
            Environmental Protection Agency, 401 M St., S.W., Washington, D.C.  20460-0001
     "To whom correspondence should be addressed

-------

-------
BACKGROUND

In the U.S.  the safety of specific chemical substances is evaluated
primarily under three statutes.  Substances used as food additives,
drugs,  and  cosmetics  are  registered  by  the  Food  and  Drug
Administration (FDA)  under the Federal Food, Drug,  and Cosmetic Act
(FFDCA).  Chemical substances proposed  for use as pesticides are
registered by the Environmental Protection Agency  (EPA) under the
Federal Insecticide,  Fungicide,  and Rodenticide Act  (FIFRA), which
imposes a host of data  requirements  for any submitter seeking to
register  the  substance as  an  active ingredient.    For  the vast
majority  of polymers,  however,  the Toxic  Substances Control Act
(TSCA) is the  applicable  statute.  TSCA  (Public  Law 94-469) was
enacted by  Congress  in  1976,  in  response  to a perceived need to
limit  exposure  to industrial  "environmental  chemicals"  such as
polychlorinated  biphenyls  (PCBs).   As  stated in  the Act,  its
primary purpose was

     "...to assure that...innovation and commerce in...
     chemical substances and mixtures do not present an
     unreasonable risk of injury to health or the envi-
     ronment."   (TSCA,  section 2(b))

As  defined  in TSCA,  "chemical  substances"  specifically  exclude
substances already regulated  under FFDCA  and  FIFRA (unless they
have non-FFDCA or non-FIFRA uses)  as well as alcohol, tobacco, and
certain other materials;   but all other substances are included.
Thus, most polymers  used  in water treatment,  coatings,  household
laundry  products,  and  manufactured  goods  are  subject  to  the
requirements of TSCA.

     These requirements are different for existing substances and
substances not yet in production,  i.e.,  new chemicals.  One of the
first tasks of the newly  created  Office of Toxic Substances (now
the Office of Pollution Prevention and Toxics or OPPT) was in fact
to assemble and  publish a  list  of chemical substances already in
commerce. This was accomplished by July, 1979 as the TSCA Chemical
Substance Inventory,  commonly  referred to simply as the Inventory,
which listed approximately 50,000  substances then  in production or
being imported into  the U.S.   Since that  time the Inventory has
grown to  include over 70,000 substances  by the  addition  of new
chemicals.

     Anyone who wishes to manufacture or import into the U.S. for   \
commercial purposes a substance not listed on the Inventory and not
otherwise excluded by TSCA (pesticides,  drugs, etc.) must submit
formal notification of  their intent to do so.  Such a  submission is
called a Premanufacture Notice (PMN), and for most  new chemicals it
must be submitted to EPA at least 90 days prior to manufacture or
import.  New chemicals not regulated by EPA are then added to the
Inventory  and  become   existing   chemicals  when  EPA receives  a
required  Notice of  Commencement  (NOC).   The NOC  declares the

-------
submitter's intent to commence manufacture or import.  Since 1979
EPA  has  received  more  than  30,000  valid  PMNs  (Fig.
submissions currently average well over 2,000 per year.

THE NEW CHEMICAL REVIEW PROCESS
                                                         1) ,  and
     Receipt of a PMN sets in motion a process of review that has
evolved since July 1979 to meet the unique requirements established
by TSCA.  The  fundamental  purpose of  PMN  review is stated in the
law and is to determine whether

     "...the manufacture, processing,  distribution in com-
     merce, use, or disposal [of a new chemical substance]
     or any combination of such activities presents or may
     present an unreasonable risk of injury to health or
     the environment..."   (TSCA section 5(b))

However,  TSCA  imposed  serious challenges  to  EPA's ability  to
accomplish  this.    The  most  significant  are,  first,  that  PMN
submitters are only required to furnish health and safety studies
already in  their possession  (if  any) , and  are not  required  to
conduct any testing  as a  precondition for  notification;   and
second, that the  review  must be done  within  90 days.  The '90-day
review  period  may  be  extended  to  180   days  under  certain
circumstances.  If no action to regulate the substance is taken by
EPA, the submitter is free  to commence manufacture or import after
expiration of the PMN review period.   The burden of proof that a
substance presents or may present an unreasonable risk rests on the
Agency's shoulders. EPA must make sound decisions based on few or
no submitted test data within 90 days.   This situation is obviously
unlike that for pesticides  in the U.S., but is  also unlike that for
new commercial chemicals in the European Union, where a series of
tests that supply the "minimum premarket data set" are prescribed
by law.

     Risk is defined as the probability of  occurrence of an adverse
health  or environmental effect associated  with exposure  to the
substance.  OPPT's new chemical review program  (NCP, USEPA 1995d.)
is, therefore, at its core a risk assessment process.  According to
the National  Research  Council (NRC)  risk assessment consists of
four components:  hazard (or effect) identification, dose-response
assessment,  exposure assessment,  and  risk  characterization (NRC
1994) .  The  NCP in  OPPT includes  these steps, but generally does
not  adhere to  the exact  standards set  by  the  NRC due  to the
aforementioned lack of data.  In addition, OPPT's determination of
which risks are "unreasonable"  also includes assessment of relative
risk, i.e., comparison of relative hazards of the PMN substance and
similar   existing  substances, and  certain  non-risk  factors.
Foremost  among these are  economic  factors  such as  the costs or
benefits  of  the new chemical,  the cost of any additional  testing
that  may  be required,  and the  economic impact  of testing  or
regulation  on  the  submitter;   and  the  pollution prevention

-------
potential (if any) associated with manufacture and use of the new
chemical.

     Fig. 2 outlines the NCP  in  OPPT.   More detailed information
about the NCP can be obtained  from Nabholz' et al. 1993a, Wagner et
al. 1995, and Moss et al.  1996.  As  a  result  of the review a new
chemical substance may be:

     •    Dropped from review because there is low concern with
          respect  to   its  toxicity  towards   humans   and  the
          environment;

          Dropped from review because there is low risk with
          respect to its manufacture, distribution, use, or
          disposal;

     •    Dropped from review, but with specified concerns
          related to its manufacture, distribution, use, or
          disposal which are communicated to the submitter via
          letter and/or a significant new use rule (SNUR,
          USEPA 1995a);

     •    Regulated due to either potential risk or potential
          significant release to the environment after the
          results of an initial review;

          Subjected to either (!) detailed review  ("standard
          review") or (2) an immediate request to the submitter
          for testing and/or additional information to determine
          if the substance should be regulated.  Immediate
          testing and/or information can only be requested if the
          chemical belongs to a Sec.  5(e) chemical category
          (Moss et al. 1996);   or

          Regulated or dropped from further review as a result of
          the standard review or the additional information
          and/or testing.

     Section  5(f)  of TSCA  grants  EPA  the   authority to  take
immediate action if there is a reasonable basis to  conclude that a
new  chemical's  manufacture,  processing, use,"  or disposal  will
present an unreasonable risk.   However,  in practice, 5(f) findings
are seldom  if ever made.   Regulation  under a  sec.  5(e) Consent
Order,  issued when it is  found  that a  substance  may present an
unreasonable  risk, is much more common.  Through a Consent Order,
the manufacturer  or importer  of  a  new  chemical  consents  to the
order's requirements in exchange for being permitted to manufacture
or  import  the   substance.    Section  5(e)  findings  are  always
predicated on insufficient information to adequately  assess risk,
and typically include requirements such  as:

     •    Use of protective equipment; e.g., gloves,  respirators,

-------
          and goggles for workers;

     •    Use restrictions;

          Restrictions to certain manufacturing, processing, use,
          and disposal sites;

     *    Warning labels and MSDS statements;  and

          Testing to resolve uncertainties regarding toxicity or
          exposure.

Testing  is a  frequent  requirement under  sec.  5(e),  but more
specific information about  releases  and uses are also requested.
If exposures can be controlled and  potential  risks sufficiently
reduced, then the testing is  generally  triggered such that it is
only  required   if   and  when  a predetermined  production  volume
{identified in  the  economic analysis)   is reached.   If exposures
and, thus,  potential risk  cannot  be controlled,  then "upfront"
testing (i.e.,  not triggered) is generally required.

     Section 5(e) findings  may  be  either risk-based or exposure-
based.  Prior to 1988 nearly all regulations under 5(e) were mainly
risk-based, which in practice means that more emphasis is placed on
toxicity than on exposure in  the  "may present" finding.   In 1988
OPPT implemented an alternative strategy,  exposure-based review
(XBR), that emphasized substantial or significant human exposure or
substantial release to the environment.   Criteria that defined the
terms substantial  and  significant were  developed  for production
volume, worker  exposure,  consumer exposure, ambient and general
population exposure, and environmental release  both total releases
to the environment and releases to surface waters (Table 1).  EPA
estimates these  parameters  based on  data submitted in the PMN as
well as its own databases and models.  If appropriate criteria are
met, the outcome is usually an  exposure-based  5(e)  Consent Order
for health  effects,  environmental  toxicity,  and/or environmental
fate testing.

       Full or partial exemption  from review may also be granted
for new chemicals meeting certain  requirements.   EPA has limited
reporting requirements for:

     •    Substances manufactured in small quantities solely for
          research  and development, as  long as special procedural
          and recordkeeping requirements are met;

     •    Substances submitted as Test Market  Exemption (THE) re-
          quests.   TMEs undergo expedited (45-day) review,  since
          the exposure assessment generally considers only  the
          volume, number of customers,  and period of time speci-
          fied  in the notice;

-------
                                                                 6

     •    Substances  to  be manufactured in  quantities of  10,000
          kg or less per year, which are submitted as  Low Volume
          Exemption (LVE) requests (USEPA 1995b, see 40 CFR part
          723).   Low  Volume  PMNs undergo an abbreviated (30-day)
          review and,  like TME requests,  the  LVE is either granted
          or denied.  LVEs are not added to the Inventory but are
          maintained in a separate list;

     •    Substances  expected   to  have  low  release   to  the
          environment and  low exposures  (LoREx  exemption,  USEPA
          1995b,  see  40  CFR part  723) , which  also  undergo  an
          abbreviated (30-day)  review  and are either  granted or
          denied; and

          Certain classes of polymers that are not chemically
          active or bioavailable (USEPA 1995c and 40 CFR part
          723);  these are discussed in more detail in the
          following section.

Chemistry review

     The PMN review process can be separated conceptually into four
phases:     chemistry  review,   toxicity  evaluation,   exposure
assessment, and  risk  assessment/risk management.   The chemistry
review phase is the first  step  and begins as soon as the PMN has
been  checked  for completeness  of  required information,  which
includes  all  available  data  on  chemical   identity,  production
volume,  byproducts,  proposed   use(s),   environmental  release,
disposal practices, and  human exposure.  The first step  in the
chemistry review is to check the adequacy of the submitted chemical
name.  EPA  now requires  the chemical name to be  consistent with
Chemical Abstracts Service (CAS) nomenclature policies as well as
names  used  for  similar  substances  already on  the  Inventory.
Consistency between  chemical name,  chemical structure, and the
chemical's manufacturing process is assured.  A full report  on the
chemistry of the new substance, containing submitted information on
the chemical identity, route of synthesis, impurities or byproducts
of the synthesis, and some physical/chemical properties,  is then
prepared.

     For most  PMNs,  however, the  Agency is unable  to conduct a
meaningful  review based  on submitted data alone.   In  large part
this is  due to TSCA's stipulation that  only data "known to...or
reasonably  ascertainable by"  (i.e., already  in the possession of)
the submitter must be  provided to EPA.  The problem of missing data
is manifested  in  the  earliest phases of review,  as even the most
basic information on properties  such as melting point and boiling
point  temperatures  and  vapor pressure  is  often absent.   In one
study of submitted PMN data,  for example, only 300 chemicals with
submitted  data for any  of  several  physical/chemical  properties
important in environmental assessment were identified  from  15,000
PMNs for the period 1979-1989  (Lynch et  al.   1991).  Although many

-------
of the  15,000  were polymers, at least half  of  these submissions
were for  class 1 substances, i.e., single compounds composed of
particular atoms arranged in a definite,  known structure.

     Polymers  are class  2  substances,  meaning  that they  have
variable compositions and/or are composed of complex combinations
of reactants, such that only a representative molecular structure
can be drawn.  Properties such as melting point and boiling point
temperature  are  relevant  for discrete (class 1)  substances but,
except for water  solubility  or  water  dispersibility, have little
meaning for  polymers.   For polymers the  chemistry review focuses
on:

     •     The monomers of which the polymer is composed and their
          mole percentages;

     •     The molecular weight (MW) distribution, including the
          number-average molecular weight (MWn) ,  how it was
          determined, and the oligomer content of the polymer,
          i.e., percentage of oligomers with a molecular mass
          less than 1,000 and 500 daltons;

          The equivalent weight of any reactive functional
          groups and/or cationic charge density;  and

     •     Properties, such as, physical form, particle size
          distribution, swellability,  water solubility, and water
          dispersibility.
             \
Water solubility, water dispersibility,  HW,  charge, and cationic
charge  density are  the most important  physical properties  for
aquatic toxicity assessment,  but basic information is often missing
or  is   not   reported  in  a  way  that  is  most   useful  for  risk
assessment.   For example, PMN submitters are not  required  to report
typical MWn  values,  but this  is  often  quite useful,  especially if
the typical  and lowest values  are far  apart.    In addition, MW
values given as greater than or  less  than some  MW are  not very
helpful unless the number  indicates the  actual  MW  (e.g., >10,000
daltons  is   not helpful  if   the   actual  MW  is  more  accurately
described as >100,000  daltons).   Finally,  monomer composition is
sometimes incomplete or the  structural diagram  fails to show the
most likely  types of linkages between monomers.   For example, a
random  reaction between monomers  rather than a blocked reaction
between  monomers  is  a  very  important distinction  that  most
submitters fail to make explicit.   Monomer  linkage is particularly
important information  for polymers claimed to be  biodegradable.

       Based on its experience from reviewing over 10,000 PMNs for
polymeric substances, EPA has identified a group  of polymers  that
it believes  poses low  to  no unreasonable risk  of  harm to human
health or the  environment due to,  generally, low  toxicity.  As of
30 May 1995  (USEPA 1995b and  40 CFR part 723)  these polymers became

-------
fully exempt  from reporting.   To qualify  for this exemption the
polymer  (USEPA 1996) must:

         •Belong to one of 12 acceptable polymer classes: poly-
          esters, polyamides and imides, polyacrylates, poly-
          urethanes and ureas, polyolefins, aromatic polysul-
          fones, polyethers, polysiloxanes, polyketones, aro-
          matic polythioethers, polymeric hydrocarbons, and
          phenol-formaldehyde copolymers;

          Oligom'er content must be less than 25 percent by weight
          below 1,000 daltons and less than 10 percent by weight
          below 500 daltons;

          Have no more than the permissible level of cationic
          character which is a functional group equivalent
          weight for cationic groups j>5,000 daltons;

     •    Have no reactive functional groups, or only reactive
          functional groups specifically allowed based on OPPT's
          risk assessment experience, e.g., blocked isocyanates,
          or a reactive functional group equivalent weight no
          less than a defined threshold, e.g., for pendant meth-
          acrylates, the equivalent weight threshold is 5,000
          daltons; and

          For polymers with MWn >10,000  daltons,  must  not be
          capable of absorbing their weight of water.

Since the majority of polymers with aquatic uses, e.g., polymeric
flocculants used  in  water treatment, are  cationic  in character,
they are not  eligible for exemption under  the polymer exemption
criteria.
                          t
Environmental assessment; exposure

     For more than  99 •  percent  of  new chemicals  the  focus  of
environmental assessment under TSCA  is on the aquatic environment.
Contamination of surface water is possible for landfilled chemicals
having certain  properties,  e.g., high  water  solubility  and low
biodegradability, if  the landfill is in  hydrologic contact  with
ground water.   Since  ground water constitutes  a  pathway for
potential  human  exposure  via  drinking water,  it is routinely
evaluated along with.other potential exposure pathways.  However,
for polymers it is generally assumed that releases to landfills and
deep  well  injection  do not  result in significant  aquatic-  or
terrestrial-ecological exposures.

     Environmental risk assessment under TSCA is accomplished using
the quotient method (Nabholz 1991, Nabholz et al.  1993, Rodier and
Mauriello  1993) .     The  quotient   method  compares  a  concern
concentration  (CC;   see  Environmental  assessment;    hazard)  or

-------
                                                                9

effective   concentration   (EC)   to  an   actual  or   predicted
environmental concentration  (PEC).   PECs  are determined from the
amount of  chemical released, the  extent  of removal  (if  any)  in
wastewater treatment, the extent of dilution  of the wastestream by
the receiving  environment,  and  the fate  of the  substance post-
release.    Depending  on  the specific  substance,  release  to the
environment  may  result  from   the manufacture,  processing,  or
industrial use  of  the substance;   from commercial use;   or from
consumer use  of household  products disposed of down  the drain.
Except for release from consumer uses, available release data are
summarized  in  an  Engineering  Report  prepared  prior  to  the
environmental exposure assessment.  Releases are expressed  in terms
of kilograms (kg) of the chemical substance released per day over
a specified  number of days  per year,  for each  specific  site of
release if the  sites are known  and for generic  sites otherwise.
Release data transmitted via the Engineering  Report do not include
consideration  of  wastewater  treatment  either  on-site  or  in
publicly-owned-treatment works  (POTWs or sewage treatment plants),
which is factored into the assessment in a subsequent step.

       The proper  starting  point for any  realistic assessment of
environmental exposure is a  sound understanding of the substance's
physical and  chemical properties.   To  fill the gaps  created by
missing data OPPT uses the Estimation Programs Interface  (EPI) to
access computerized estimation methods for melting point,  boiling
point and vapor pressure (MPBPVP);  water solubility (WSKOW; Meylan
et al. 1995) ; octanol/water partition coefficient (LOGKOW; Meylan
and Howard 1995);  Henry's Law constant (HENRY; Meylan and Howard
1991) ; and soil/sediment  sorption coefficient (PCKOC;  Meylan et al.
1992).    Also   used  to predict  the  octanol/water  partition
coefficient  is  the  computer program, CLOGP (Leo and Weininger
1985).    These  methods are based on 'fragment  contribution and
fragment-based correction factor approaches,  and provide state-of-
the-art estimates for most class 1  substances.

     For  both class  1 and  class  2  substances  it  is generally
assumed that water releases are  subjected to primary treatment by
gravitational settling and activated sludge secondary  treatment, at
a minimum.   If  additional or specialized  treatment processes are
employed, this information is also considered in the  assessment if
sufficient  data are  provided.    A  frequent shortcoming  of PMK
submissions  is  that they  contain detailed information  on the
treatment system, but only general information on efficiency.  For
example,  removal of  biochemical oxygen  demand  (BOD)  in on-site
treatment  may demonstrably  exceed  98% in  order  to satisfy water
permit  requirements, but usually there  is  no way  to determine
actual removal  of  the PMN substance itself from this information.
Whatever the  characteristics of the wastewater treatment system,
efficiency  of  removal  nearly  always must be estimated  for new
chemicals.   For  class  l substances  a  fugacity-based multimedia
model  (STP;  Clark et al. 1995)  is often  used to provide removal
estimates that integrate the key removal processes of  sorption, air

-------
                                                                 10
stripping,  and  biodegradation.    However,  the  model is  neither
useful  nor  is  it  needed for  most polymeric  substances,  since
sorption and/or precipitation are likely to be the only significant
processes.

     The  final  step in  the exposure  assessment  process is  to
calculate  a  PEC  based  on  post-treatment  release  of  the  PMN
substance.  In initial review,  if discharge is to a water body with
unidirectional  turbulent flow,   i.e., a river or stream,  surface
water concentrations are usually calculated using a simple dilution
model and fate  is considered qualitatively.  More complex aquatic
ecosystem models such as  EXAMS II  (USEPA 1985),  in which  site-
specific  ecosystem  characteristics  as  well  as   transport  and
transformation parameters  can be quantitatively  incorporated,  are
employed only  in detailed review  ("standard  review"; Fig. 2  and
USEPA 1994a).   If  the location of the release site is known,  the
stream  flow  for the  specific  river reach to  which the  facility
discharges is  used.   PECs are  calculated for mean and low  flow
conditions  using U.S.  Geological  Survey (USGS)  data.    If  the
location is  not  known or if multiple release sites are possible,
mean and low flows  from the stream flow distribution for industrial
facilities in the most relevant  Standard Industrial  Classification
(SIC) codes  are  used  (50th and  10th  percentile  flows  for  any
distribution) .   If this analysis results in  PECs that exceed  the
CC, the number of  days  of exceedance per year is estimated using
the Probabilistic Dilution Model {PDM; USEPA 1988).  In addition to
the CC the model requires input  data on kg released per day, number
of  days of  release  per  year,   and  either location  of  specific
site(s)  or  relevant  SIC code.    PDM performs a simple  dilution
calculation—i.e., does not account for transport or  transformation
processes after  release—but does  account for natural variability
in  stream  and  effluent  flow  by  incorporating   a  probability
distribution,   i.e.,   flows  are  assumed   to  be   log-normally
distributed, into the calculations.

Environmental assessment:  toxicity

     The purpose of  environmental toxicity assessment is (1)  to  j
identify as many of the potential  effects of a chemical substance
towards organisms in  the environment as possible and (2) to predict
the  potency  of  each effect.    Environmental  toxicity  data  are
usually  expressed   as   effective  concentrations  (EC)   of   the
substance, which indicate both  the  type and  seriousness of  the
effect.  The compilation of available ECs constitutes the hazard
profile or  the  toxicity profile  of  the substance,  and for  new
chemicals it contains,  at a minimum, six effects with effective
concentrations based on  100% active  ingredients  (ai):

     •    Fish acute value  (96-h LC50) ;
          Aquatic invertebrate  (usually  Daphnia  macma) acute
          value  (48-h LC5Q) ;
     •    Green  algal toxicity  value  (96-h EC50) ;
                                                 US EPA Headquarters Library
                                                      Mail code 3201
                                                 1200 Pennsylvania Avenue N\ V
                                                   Washington DC 20460

-------
                                                                11

      •    Fish chronic value  (ChV) from an early life stage
          toxicity test;
          Aquatic invertebrate ChV from the daphnid partial life-
          cycle toxicity test;  and
          Green algal ChV value from the 96-h toxicity test.

Since  OPPT  considers  that  the  96-h  algal  toxicity test  also
constitutes a test for chronic toxicity, the geometric mean of the
96-h lowest-observed-effect concentration (LOEC) artd the 96-h no-
observed-effect concentration (NOEC)  is considered to be the algal
ChV  if the  LOEC  is less  than  the  96-h  EC50.   The  ECs in  a
substance's toxicity profile  can  be  either  measured or predicted
using structure activity relationships  (SAR).  Measured values are
preferred, but  since less  than  4.8%  of  PMNs  contain  submitted
environmental toxicity  data  (Nabholz  et al.  1993a) ,  ECs  must be
predicted  for  the great  majority of  substances (Wagner  et  al.
1995).  Over 95% of the toxicity profiles for new chemicals under
TSCA consist entirely of predicted ECs.  When ecotoxicity data are
submitted with PMNs,  they are mostly for acute toxicity to fish.

     The need to predict the toxicity to chemicals based on their
structure and properties has led to the development of a variety of
methods   collectively   referred   to   as   structure   activity
relationships (SAR).   SARs used in OPPT are always quantitative in
that they are used to predict the toxic potency of  a chemical or
the EC.  Predictions are  obtained through the use  of two general
SAR methods:   nearest analog analysis or mathematical  equation.
SAR analysis via nearest analog analysis compares a chemical with
one  or  more  analogs.    A   prediction  is  obtained  by  either
interpolation   between   analogs,   i.e.,    geometric   mean,   or
extrapolation from one  or more  of the  analogs.   SAR analysis via
mathematical equation generally involved correlation  of chemical
structure or physical  properties  as surrogates  for structure to
measured toxicity values via statistical regression analysis.  The
chemicals in.a SAR are  homologous chemicals,  generally,  from the
same chemical class.   SARs for environmental  toxicity  now exist
(Clements and  Nabholz 1994,   Clements  1988,  and Clements  et  al.
1993)  for many  classes  of  substances though by  no means  all
substances subject to  TSCA  review,  and they  have  been immensely
useful in assessing new chemicals. Most SARs  for ecotoxicity have
been developed from data on class 1 substances;  SARs for class 2
substances were more difficult to  develop.  SARs are available for
all major  classes of polymers.   Validation of these SARs is an
ongoing process within OPPT and has been successful  (Nabholz et al.
1993b, and USEPA 1994b).

     Determination  of  a  CC  is  the  last  step  in the  toxicity
assessment   process   for  aquatic  toxicity.    A   CC   is  that
concentration  of  the   substance which,   if  exceeded  in ' the
environment, may cause a significant risk of harm. If the exposure
assessment does  not  yield PECs that  exceed  the CC,  it is assumed
that  the  probability  of  significant environmental  risk  from

-------
                                                                12

exposure to the substance is too low to warrant direct regulatory
action, for example to limit release.  The CC for aquatic life is
homologous  to  the Organization  for  Economic Co-operation and
Development's (OECD) and European Union's  (EU) predicted-no-effect
concentration  (PNEC)  and analogous to a  reference  dose  (RfD) in
risk assessment for human health (Barnes and Dourson 1988).  The CC
is expressed  in  the same units as  for PECs;  i.e.,  milligrams or
micrograms  per liter.   It  is calculated  from the  ECs  in the
toxicity profile by the  application of assessment  factors (AsF),
which are homologous to uncertainty factors  (Health Council of the
Netherlands 1989)  but  less similar to margins of safety or exposure
(USEPA 1990).

       In general,  AsFs are used to adjust ECs downward in order to
account for the amount and quality  of data in the toxicity profile
and their relevance for assessment of chronic toxicity under actual
environmental conditions.   For example,  an AsF of  1,000  is used
when the toxicity profile contains only one acute toxicity value.
This  EC  is  divided by  1,000  to  derive the  CC.    If data are
available from several acute but no chronic toxicity tests a factor
of 100 is used, and a factor of 10  is'used  when the ChV is known
from laboratory test data for the presumed most sensitive species,
but field data are  not available to confirm the ChV.  The CC is set
equal to the ChV,  i.e.,  the  AsF is  \, only when the ChV is derived
from  field  data  or, in some cases, microcosms  and/or  mesocosms.
Currently, if the exposure assessment identifies endangered species
that exist in the area where release is expected (OPPT has linked
its database for river flows for specific  streams and reaches with
an endangered species database),  then the CC is adjusted downward
by another factor of 10 for comparison to the PEC.

TESTING

Environmental toxicity

     Reliable test data are the foundation of risk assessment and
should always be considered preferable to  predictions based on SARs
or other estimation methods. In environmental toxicity assessment,
the  issue  of which tests  should constitute  a minimum data set
("base set") or first  tier of testing surfaced at an early stage in
the implementation of  TSCA  (Zeeman  and Gilford 1993).  For aquatic
exposures the base set of tests corresponds to the acute toxicity
attributes in the toxicity profile  for new chemicals, and consists
of:
          Fish acute toxicity test;
          Aquatic invertebrate acute toxicity test;
          Green algal toxicity test.
and
For terrestrial exposures the base set consists of a rodent acute
oral toxicity test, plant early seedling growth test, an earthworm
acute toxicity test, a soil  microbial community toxicity test, and

-------
                                                                13

an  avian acute  oral toxicity  test.    The aquatic  base set  is
frequently invoked under TSCA as tier  one testing for both new and
existing chemicals,  and  some  or all of  the tests  may be mandated
depending upon  the specific data requirements  for  the substance
under review.  Higher tier tests consist of subchronic and chronic
toxicity  tests,  toxicity  tests using contaminated  sediments,
microcosms,  and field studies (Smrchek et al.  1993).

     OPPT's aquatic  toxicity  test guidelines  as well  as those of
other  groups such  as the  OECD,  EU,  Standard  Methods, and  the
American Society for Testing and Materials  (ASTM) were designed to
generally apply to class  l substances.   These guidelines typically
employ  "clean  dilution  water"  test systems  with  low  levels  of
dissolved organic carbon  (DOC) and total suspended solids (TSS) in
fish  and invertebrate  tests and  growth  media with  low  water
hardness in algal  toxicity  tests.   However, studies (Gary  et al.
1987 and Nabholz  et al.   1993a)  have  shown that the  toxicity of
certain types of polymers may be affected by water hardness or DOC
concentrations, which also  vary with  environmental  conditions.
Toxicity testing   for  polymers   has  therefore  been modified  to
address  the  effects  of  DOC  and  hardness  and  the  realistic
conditions  encountered  in the  natural  environment.    This  is
discussed more thoroughly in ENVIRONMENTAL CONCERNS FOR POLYMERS.

Environmental fate

     The environmental behavior of  chemical substances is  highly
variable and ultimately  reflects  the  controlling influence  of
molecular structure.   Although  it  is  true that  among  the many
physical properties  of chemicals that might be  measured some are
more  important  than others,  the most  Important parameter(s)  to
measure is not necessarily the same  for  all substances.  Thus from
an  environmental  perspective  vapor  pressure  may be the  most
critical property for one substance, whereas for another it may be
the substance's water solubility.  Transformation processes, e.g.,
biodegradation and hydrolysis, are even  more varied and in general
are also substantially affected by environmental conditions.  For
these reasons it has not  been possible to develop a fixed yet still
affordable  set of  tests  for  physical/chemical  properties  and
environmental fate under TSCA.

     Testing to obtain data on  key  properties and transformation
processes is sometimes required  in risk-based  5(e) Consent Orders.
In such cases the nature of the  testing  generally depends upon the
exposed  populations and  routes  of  exposure,  and the outcome of
required testing  is  often used  as a trigger for toxicity testing.
For example, if a  high concern for aquatic toxicity  suggests  a need
for base set testing to resolve  uncertainty regarding ecotoxicity,
the latter nevertheless may be triggered only  if the PMN substance
is not readily biodegradable  as determined by an appropriate OECD
biodegradation  test.  As another example,  if a compound's water
solubility or vapor pressure strongly  influences the likelihood of

-------
                                                                14

exposure via some specific route, health or environmental effects
testing  may be  made contingent  upon determination  of relevant
properties.

     A fate testing strategy designed to account for the complexity
of  environmental  transport  and  transformation  phenomena  was
developed  for application  in  exposure-based  PMN review   (XBR).
Since  most polymers used  in  cleaning  products,  as  wastewater
flocculants, and in  industrial water treatment are expected to be
released to water in significant volumes, it is likely that the XBR
fate testing  strategy  will  apply to them.   The strategy borrows
some aspects of the "base set" approach in ecotoxicological testing
but is also intended to be flexible.  Aspects relevant to aquatic
exposures are  outlined  in Table 2.   An important feature is that
EPA generally limits required fate testing  to one or two tests for
substances  with  estimated production  volumes  >100,000  kg/year,
whereas if the volume is >l,000,000 Ib/year three or more tests may
be required if the data are deemed essential.

     Existing OPPT test guidelines in the principal areas that the
XBR fate  strategy addresses  are  primarily intended to  apply  to
class 1 substances,  not polymers.   Nevertheless,  the strategy is
interpreted broadly  by OPPT and acceptable test protocols  do not
need to be  based  strictly on  existing  published guidelines.  For
example, data  from  an  appropriate jar test  (ASTM  1994)  or other
laboratory  test designed to measure  performance of sedimentation
aids in wastewater clarification  (Halverson and Panzer 1980) may be
more relevant  for determining  fate  of a  polymeric  substance  in
wastewater treatment.  It  is prudent for the chemical manufacturer
or importer to submit the protocol to OPPT  for review and approval
before testing begins.

FATE ASSESSMENT FOR POLYMERS

     Electronic charge,  MW, and  solubility/dispersibility in water
are the most important  properties of polymers for fate assessment.
Charge density, i.e., number of charges per unit of  MW, may also be
important but  probably  plays a greater  role in toxicity assessment
(see  ENVIRONMENTAL  CONCERNS  FOR  POLYMERS).     Polymers  are
conveniently and pragmatically divided into four classes based on
charge:   anionic (negative  charge), cationic  (positive charge),
amphoteric  (positive and negative  charges  present  on  the same
molecule) and nonionic (electronically neutral).

     The  vast majority of  synthetic  polymers are  essentially
nonbiodegradable, a  fact that has been known for many years (for
example, see Alexander 1973).  Even modified natural polymers such
as carboxymethylcellulose and cellulose acetate butyrate having an
appreciable   degree  of  substitution  are   not   significantly
biodegradable.    Some  modified  natural  polymers  and  synthetic
polymers with  ester  linkages  or other labile groups incorporated
into the main  chain of  the polymer are  biodegraded  under favorable

-------
                                                                15
conditions.    Examples  include poly(/3-hydroxybutyric  acid),  a
naturally  occurring  bacterial  energy  storage product,  poly(6-
caprolactone),   polyfglyoxylic   acid)   and  poly(tetramethylene
succinate)  (Bailey and Gapud 1986,  Park et al. 1989, Pranamuda et
al.  1995).    However,  rates  of  biodegradation  that  may  be
significant  relative  to typical  residence times  in terrestrial
environments  (especially landfills) may  not be so for wastewater
treatment systems, where retention times are much shorter.  Thus
even  so-called   biodegradable   polymers   are   not  necessarily
significantly biodegraded in sewage treatment.

     All  PMN substances,  including  polymers, are  evaluated for
their biodegradability based on molecular structure if submitted
data are insufficient. In practice, the few biodegradable polymers
among the many recalcitrant  ones are easily identified because the
polymer's  biodegradability  is  usually  a major  reason  for the
submitter's desire to  market the substance.  In these cases the PMN
generally contains experimental data on biodegradability and this
characteristic figures prominently in a Pollution Prevention (PP)
Claim (see THE NEW CHEMICAL PROCESS).

     Since air stripping is obviously  not a major loss mechanism
for polymers  the  assessment of  removal of polymers in wastewater
treatment essentially reduces  to an  assessment of  their likely
removal  by  sorption  and/or precipitation.    In  review  of new
chemicals  under  TSCA, treatability  of  a polymer  discharged  to
wastewater treatment is inferred from the polymer's charge, MW, and
solubility/dispersibility unless  actual  data accompany  the PMN
(rarely the  case).  It is generally  assumed that the sequence of
treatment processes is limited to primary clarification followed by
activated sludge secondary treatment, which is typical for POTWs.
Nonionic, cationic, and amphoteric polymers with MWn >1,000 daltons
are assumed  to partition mainly to  the solids phase and be 90%
removed relative to total influent concentration.  The 90% figure
was  selected because  it represents  a  typical  level  of solids
removal in  POTWs  (USEPA  1982) .   The  remaining  fraction (10%)  is
thus assumed to  be discharged  to  receiving  waters,  although of
course it is  likely that this material is in the form of polymer
sorbed to sludge  solids  (organic matter).   For polymers of 500 <
MWn < 1,000,  a lower  removal rate  (typically 50  to 90%)  may be
assumed depending on  the polymer's structure  and properties.

     Anionic  polymers  with  negligible  water  solubility  and
dispersibility  are   assumed  to  behave   similarly  to  nonionic
polymers, but lower removal  rates are assumed for anionic polymers
having  appreciable solubility  or  dispersibility.   Removal rate
varies with MW but values are typically assigned as follows:
          MWn < 5,000
          5,000 < MWn < 20,000
          20,000 < MWn < 50,000
          MWn > 50,000
0 to 50%
  50%
  75%
  90%

-------
                                                                16

This   scheme  is   consistent   with  data   for  elimination  of
poly(carboxylic  acids)  in  the Semi-Continuous  Activated Sludge
(SCAS) and OECD Confirmatory tests (Table 3;  unpublished Procter &
Gamble data  summarized  in Opgenorth  1992),  and it is  consistent
with the expectation that  removal of anionic polymers will be lower
than  for cationic  or  neutral  polymers  due  to  the  net  negative
charge of  microbial cell surfaces.   Polycarboxylates are widely
used  in  detergents  and  other cleaning products and are generally
either homopolymers  of  acrylic acid or copolymers of acrylic and
maleic acids.   Their environmental properties  and toxicity have
been extensively reviewed by Opgenorth (1992).

     As  a  general rule,  for purposes of estimating environmental
concentrations polymers discharged from  wastewater treatment are
assumed not to undergo further removal by  in-stream fate processes.^
Precipitation of anionic polymers by alkali metal cations,  e.g, Ca
and Mg ;   polymer binding to dissolved organic carbon (DOC);  and
other  sorption  phenomena  control the  bioavailability  of  the
discharged material  (Gary et al.  1987 and Nabholz et al. 1993a).
These factors affecting bioavailability are considered when setting
the CC for a  polymeric substance.  These areas are addressed in the
following  section.

ENVIRONMENTAL CONCERNS FOR POLYMERS

     In toxicity assessment, distinction is made between polymers
with minimal  low  molecular  weight (LMW)  material, i.e.,  MWn>1000
with <25%  <1000  and <10% <500 daltons,  and  polymers with either
MWn<1000  or  polymers with  significant  amounts of  LMW  material,
i.e., >25% <1000 and >10% <500 daltons (USEPA  1996).  Water-soluble
or  dispersible polymers  with  either MWn  <1000  or  significant
amounts  of   LMW  material   are   of  concern  because  of  their
toxicologically similarity to polymers with MWn >1000, but  they may
also be  absorbed  through  biological membranes  and cause systemic
effects.   Environmental toxicity assessments for these polymers are
generally  based  on  both  the  type of  polymer  and  the  type  of
functional group(s)   in the LMW components (Nabholz et al. 1993aJ.
Polymers with  MWn >1000 are not absorbed through the respiratory
membranes  of aquatic organisms,  and  thus  toxicity is manifested
either through direct surface-active effects  on outer membranes of
aquatic  organisms   or  indirectly  via  chelation  of  essential
nutrients,  or both  (Nabholz  et  al.  I993a).   All  polymers  are
assesses as polymers, but, in addition, polymers with MWn <1000 and
polymers  with  significant  amounts  of  LMW oligomers   are  also
assessed as monomers.

     Insoluble polymers are  not expected  to be toxic unless in the
form  of  finely  divided particles.   The  toxicity  of  insoluble
particles  does not  depend  upon  the chemical  structure  of  the
polymer  and  results from  occlusion of respiratory organs such as
gills.   In this case toxicity occurs only at concentrations which
are considered of low concern (Wagner et al. 1995): acute  toxicity

-------
                                                                17

values are generally >100 mg/L and chronic values
>10 mg/L.

Anionic polymers

     Polyanionic polymers with MWn >1000 that are water soluble or
dispersible  in water  are  of concern  for aquatic  toxicity from
either direct or indirect toxicity.   Anionic polymers are divided
into  two  subclasses  for  assessment  purposes:    poly(aromatic
sulfonates/carboxylates) and poly(aliphatic acids).

     Poly(aromatic sulfonate/carboxvlatel polymers with MWn >1000
are of moderate concern for toxicity to  aquatic organisms, i.e.,
base set  LC5g/EC50 values  between 1  and  100  mg/L (Wagner  et al.
1995),  depending on the monomers  they contain (Table 4, polymers 1
thru 15) .  Dominant monomers associated with polymer toxicity thus
far  identified are:   carboxylated  diphenolsulfemes,  sulfonated
diphenolsulfones,    sulfonated   phenols,    sulfonated   cresols,
sulfonated   diphenylsulfones,   and  sulfonated  diphenylethers.
Monomers of low concern are sulfonated naphthalene and sulfonated
benzene.      The   strongest  evidence  for  direct   toxicity  of
poly(aromatic sulfonates/carboxylates) polymers is  based  on data
for polymers of carboxylated biphenolsulfone (1).  This is the only
poly(aromatic  sulfonate/carboxylate) polymer   which  had  an MWn
clearly >1000 and nil percentages of LMW oligomers.   In addition,
the  green algal  toxicity  test  was repeated  three times  with
growth/test medium with moderate hardness, i.e., 46, 152,  and 160
mg/L as CaCO,, to determine if the observed toxicity was due to the
indirect effect of  over-chelation of nutrient  divalent elements.
There was no reduction  in toxicity with increased hardness, i.e.,
algal 96-h EC50 values of 24, 20, and 47 mg/L, respectively, and the
polymer was observed to be algicidal. The mode-of-toxic action is
unknown but it appears to be a direct acting toxicant.

     The base set toxicity values for the other toxic polymers (2
to 13) ranged  to  a low value  of 2.0 mg/L (8).   The toxicity of
these polymeric  substances is undoubtedly a  combination  of the
direct toxicity from the polymer  itself and systemic toxicity from
the  LMW  oligomers  which  are capable  of being absorbed  through
respiratory membranes.

     It appears that  any  of the  monomers identified with polymer
toxicity can either be  substituted with carboxylic acid or sulfonic
acid.  Although  there are  no data  for  polymers with substituted
phosphoric  acid  monomers,  toxicity of  poly(aromatic phosphate)
polymers  should be assumed  to  be  equivalent  to  poly(aromatic
sulfonates/carboxylates) until test data can be obtained.  It also
appears that the most  common chemical moiety for this subclass is
phenol.   Thus, other  polymers  similar  to the  polymers  in this
subclass but differing only in the type of phenolic moiety should
be assumed to have similar toxicity.  The nearest analog SAR method
is  used  by  OPPT  to predict  the toxicity of  new polymers which

-------
                                                                18
belong to this subclass.
     Poly(aliphatic  acids);    Toxicity.     Poly(aliphatic  acid)
polymers are a subclass of polyanionic polymers and have been shown
to  contain  three acids:   carboxylic acid,  phosphinic acid,  and
sulfonic acid.  Polymers can be either homopolymers relative to one
acid or can contain mixtures of acids.  The polymers at
pH  7  are of  concern  only for  their indirect  toxicity to green
algae.  Toxic.ity to algae as indicated by the 96-h EC50  for growth
inhibition  is considered  moderate.    Toxicity towards  fish  and
aquatic  invertebrates  (Table 5)  is  consistently low,  i.e.,  LC50
values >100 mg/L (Wagner et al.  1995).  The average toxicity value
for polyanionic polymers  in Table  5 towards  fish  and aquatic
invertebrates is >225.0 mg/L.  Measured acute toxicity values for
an additional 43 polyanionic polymers  submitted to the NCP under
sec. 5 of TSCA had an average fish acute value >780.0 mg/L (n = 43)
and a average daphnid acute value >560.0 mg/L  (n = 22) at pH near
7 and 100% active ingredients (ai) or 100% polymer solids.

     Test data  suggest  that the  mechanism of  toxicity is  over-
chelation of nutrient elements needed by algae  for growth  (Nabholz
et al.  1993a), especially calcium and magnesium,  and probably iron.
Test data further suggest  that  the potency of a polymer to cause
indirect toxicity  is directly  related  to the  distances between
acids.   The distance between acids  control the  strength of  the
chelate  formed  between  two   acids  and   polyvalent  metals.
Homopolymers  of  acrylic  acid  (polyacrylates)   which  contain
carboxylic  acid substituents on  every  other carbon  in the  main
carbon  backbone  of  the  polymer   have   been  the  most  toxic
poly(aliphatic acids)  towards green algae  (Table 5, polymers 17 to
22} with a geometric mean 96-h EC50 value of 8.6 mg/L.  In contrast,
if you move the acids  closer  together or farther apart relative to
their position along  the polymer backbone, then  toxicity decreases
based on test data for PMN  polymers whose  chemical  identity  is
confidential under TSCA.  Test data for homopolymers  of maleic acid
(16),  which have COOH on every carbon of the polymer backbone,  (a)
indicate low toxicity to  algae  (96-h  EC50 = 560 mg/L) ,  (b)  65 times
less toxicity relative to  poly(acrylic acid), and  (c) suggesting a
weak ability  to chelate  nutrient elements.  Other  test data  for
polymers which  have acids  farther apart than 1 carbon, i.e.,  1.5
carbon separation  (23)  and 2.0  carbon separation  (24),  showed 8
times  and   17  times  less  toxicity,  respectively,  relative  to
poly(acrylic acid) (17 to 22).

          In addition, the toxicity of a polyanionic polymer can be
reduced  relative to the  toxicity of  poly(acrylic acid) by  (a)
adjusting the distance between acids by moving  some acids farther
from the polymer backbone or  (b) randomly  diluting a monomer which
chelates, e.g., acrylic acid, with a  monomer  which does not, e.g.,
acrylamide  and  maleic acid.   Polymer 25  is a polymer in  which
poly(acrylic  acid)  acid has been randomly  reacted with another
monomer in which still has a  carboxylic acid  on  every other carbon

-------
                                                                19

but the carboxylic acid is pendant to the backbone by a distance of
3 carbons.   The  green  algal  96-h  EC50 of polymer 25 is 57.0 mg/L?
7 times less toxic than poly(acrylic acid).  Test  data for polymer
26 (Table 5)  shows the  effect on toxicity of diluting a homopolymer
of acrylic  acid by just one  third with a non-chelating  or weak
chelating monomer.  The algal 96-h EC50 of this polymer was >500.0
mg/L which  is  a toxicity reduction of over 60  times  relative to
poly(acrylic acid).

     Therefore,  based  these  data,  poly(aliphatic acid)  polymers
have been grouped by the average distance between  the acids on the
polymer  (Table  5)  and  the  toxicity  of these  polymers ^ can  be
predicted by using the nearest analog SAR method and knowing
(a)  the monomer  composition  of a polymer;  and (b) the reaction
sequence, i.e.,  random reaction of monomers or blocked reactions.
In addition, the toxicity of  poly(aliphatic  acids) towards green
algae  is  (a) an  indirect  effect  via  over-chelation  of nutrient
elements, (b)  is sensitive  to changes in the chemical structure
which result in changing the distance between acids, and (c) will
only be observed in soft natural waters which have  hardness of less
than 30.0 mg/L as CaC03.  The  hardness of test/growth  media  as
specified in EPA, OECD, EU, and ASTM test methods is 15 to 24 mg/L.
The toxicity measured in the standard green algal toxicity tests is
an overestimate of toxicity for most natural surface waters.

     Poly(aliphatic acids):   Mitigation of Toxicitv.   If the Ca2*
salt of a poly(aliphatic acid)  polymer is tested in standard algal
media or if the polymer is tested  in media with a  hardness of near
150.0 mg/L CaC03, then  the  toxicity is mitigated  (Table 6).   OPPT
began  to  recommend toxicity  mitigation testing  for  polyanionic
polymers in  1989  after  toxicity mitigation testing  for polycationic
polymers was initiated.  PMN test data for polyacrylates (19, 20,
and 22) demonstrated mitigation  factors from 14 to 600 times as the
amount of Ca  in the testing  environment  increased. Test data for
two polymers intrinsically less toxic than  polyacrylate polymers
(23 and 25)  indicated less mitigation with mitigation factors of 12
and  8.9,   respectively.     However,   the   mitigated  effective
concentrations were similar.   Green algal 96-h EC5o values v^ere all
in the range of  500.0  to 950.0  mg/L in the presence of Ca .  All
of  this mitigation  testing  was  done  near  pH  7.5.    Toxicity
mitigation of poly(aliphatic acids) fails when the acid is'tested
un-neutralized.  The toxic effects of the H*,  i.e.,  low pH, cannot
be overcome by adding Ca  either to  the polymer or to the media and
dominate all testing results.

     The manifestation of toxicity  via over-chelation of nutrient
elements may be largely confined to laboratory conditions, +such as,
release  of   the  polymer as  the  soluble salt,  e.g.,  Na  or  K ,
directly to naturally soft surface waters without sewage treatment.
Standard  algal toxicity tests  generally seek  to maximize algal
growth and measure intrinsic or baseline toxicity and, thus, employ
media with a hardness  of about  15 to  24  mg/L, which is soft.  Yet

-------
                                                                20

many  poly(aliphatic  acids)  are  used  as   scale  inhibitors  in
industrial  and  commercial  applications  and are released  to the
environment already  chelated  with Ca* and  Mg".   Moreover, other
major uses such as in detergents and  cleaning products should lead
to  release  of the polymers to wastewater  treatment,  where both
chelation and removal by sorption occurs.  For any polymer released
in sewage treatment effluent, mitigation of toxicity may also occur
from  divalent cations  in  receiving waters,  since the  average
hardness  of  freshwater  in  the U.S.  is  about  120  mg/L  (CaC03
equivalent; Nalco 1988).

     Recommended algal toxicity testing  for poly(aliphatic acid)
polymers under sec.  5  of  TSCA generally  consists  to three tests:
(1) neat polymer at pH 7.5 tested  in  standard algal growth medium;
(2) polymer with  stoichiometrically  equivalent Ca +  added  to the
polymer  stock solution at  pH  7.5  and tested  in  standard  growth
medium;  and  (3) neat  polymer at  pH  7.5  tested in modified algal
growth medium.  Modified  growth medium contains  calcium alone or
calcium and magnesium added to attain'a measured hardness of about
150 mg/L.  If  Ca * and Mg  are added together they should be present
in a 2:1 (Ca:Mg)  ratio. The purpose  of this scheme is to simulate
various conditions of release.  For example,  polymers  used as scale
inhibitors are expected to be  released to the environment chelated
with Ca   and Mg   and  should  therefore be  tested  with  all three
tests.  Test  (1)  measures intrinsic toxicity.  If the 96-h EC50 is
>100.0 mg/L and the Chv is >10.0 mg/L, then no further testing need
be done  because of low concern for intrinsic toxicity.   Test (2)
simulates the release  from  use as  a  scale inhibitor and test (3)
simulates the release  of  the  neat polymer from manufacturing and
processing.    For  polymers  not  expected to  be  used  as  scale
inhibitors or released as the Ca+ and/or the Mg+, test (2)  can be
eliminated.   Of  course,  if the site of release is known and the
hardness of receiving water is  known at the site, then the hardness
of  the  medium  can   be   specified  to  match  the  site-specific
conditions.  In some  cases,  PMN submitters have combined tests (2)
and  (3)  into  one  test,   i.e.,  testing the  calcium salt  of the
polymer  in moderately hard medium.  This  is  acceptable but keep in
mind that the.green algal 96-h EC50 due to hardness alone is equal
to 1140.0 mg/L and the ChV =80.0 mg/L.

Nonionic polymers

     Nonionic polymers with MWn >1000, <25% <1000,  and <10% <500
are generally of  low  concern for ecotoxicity because  they have
negligible  water  solubility.    If a  nonionic polymer  is water
soluble  or dispersible and  has monomers reacted via  random order,
then aquatic  toxicity is  still low  with  base  set  LC50/EC50 values
expected to be >100.0 mg/L.  However,  if monomers are blocked in
order  to use  the  polymer  as a surfactant or dispersant, then the
polymer  could be  toxic to aquatic  organisms  through  a surface-
active detergent-type mode-of-toxic  action and should be tested.
Nonionic polymers  with significant  amounts of oligomer content,

-------
                                                                21

i.e.,  >25%  <1000 and >10%  <500,  may be of concern based on the
bioavailability  and toxicity  of the  LMW material  and  the LMW
oligomers are assessed as monomers.  And finally, nonionic polymers
with MWn <1000 are assessed as monomers.

Cationic polymers;  Toxicitv

     Cationic  polymers  of  concern for  environmental  toxicity
includes polymers  that  contain a net  positively  charged atom or
that contain  groups that can reasonably be anticipated to become
cationic in water  (USEPA 1996).  Atoms with a net positive charge
includes, but are not limited to, quaternary ammonium, phosphonium,
and sulfonium.   Groups anticipated to become cationic  in water
include, but are not limited to,  aliphatic primary, secondary, and
tertiary amines.  Forms of nitrogen not included are (a) aromatic
nitrogens, unless they are  quaternarized;   (b) nitrogens which are
directly substituted  to benzene, such as, aniline;   (c) amides;
(d)  nitriles;   and  (e)  nitro  groups.     As with  polyanionic
polymers, polycationic polymers of concern have to be either water
soluble  or  dispersible in water  as  either micro-  or  macro-
dispersions,  cationic polymers that are solids and are only to be
used  in the  solid  phase   are  of  low  concern.    specifically,
dispersed beads of cationic polymers are of low concern.

     Polycationic polymers  are assessed according to their type of
polymer backbone:  carbon-based,  silicone-based, e.g.,  [Si-O], or
natural, e.g.,  chitin,  starch, and  tannin,  because  the toxicity
data for these polymers have suggested that  the type of backbone
can  influence toxicity and  some  physical/chemical  properties.
Therefore,  the  toxicity  data  for   cationic  polymers  have  been
grouped according to polymer backbone  ib Table 7.

     The intrinsic aquatic toxicity of  polycationic  polymers in
clean water, i.e., water with a total organic  carbon (TOG) content
of <2 mg/L,  increases  exponentially  with increasing charge density
until toxicity becomes asymptotic (Table 7) . Polycationic polymers
have been shown to elicit acute toxic effects  in aquatic organisms
by physically disrupting respiratory,  e.g.,  gill  membranes, thus
interfering with O2 exchange  (Biesinger  and  Stokes  1986).  It is
presumed  that  polycationic   polymers strongly  adsorb  to  all
biological membranes which are negatively charged or anionic.  It
is  also presumed that chronic  toxicity occurs  thru  the  same
mechanism for polymers with minimal  amounts of LMW  oligomers,  i.e.,
<25% <1000 and <10% <500 daltons.   For cationic polymers with KWn
<1000 and polymers with  significant  amounts of LMW  oligomers,  i.e.,
>25%  <1000 and  >10% <500,  systemic  toxicity is  also  possible
(Nabholz et al.  1993a).

     The aquatic toxicity of polycationic polymers is most strongly
influenced by cationic  charge density and type of  polymer backbone
(Table 7)  rather than  (a)  if the  charge  is  permanent, such as,
quaternary  ammoniums, or dependant  on  pH,  e.g., aliphatic 1°, 2',

-------
                                                                22

or  3°  amines;   (b)  the position  of the cation  relative to  the
polymer  backbone,  such as,  in  the backbone  or  pendant from  the
backbone;   or (c)  MW.  For  example,  for polymers with  silicone-
based  backbones  (polymers  62  to  70,   Table  7),  the  largest
difference  between daphnid 48-h LC50 values due to charge  density
is about  2300  times,  i.e.,  62  compared to 69, but the difference
between  toxicity  due  to  converting  tertiary   amines  . (65)  to
quaternary  ammoniums  (66)   is  only  5  times.    Another  example
compares  the  apparent  effect of MW between two polymers with  the
same charge density.   Polymers 28  and  29  have  the  same charge
density  of  0.7% amine-N  but differ greatly  in  MNn,  i.e.,  1800
versus  8 million.   The  average  difference  between  fish acute
toxicity  values  (n  =  3)  is  <2  times and  the average difference
between daphnid acute  values (n = 2)  is 9 times.  The differences
between green algal 96-h EC50 and chronic values are >160  times  and
150 times,  respectively.

     Cationic  charge density is based  on percent amine-nitrogen
(%a-N)   because more than  99.9% of  all  polymers that  have been
submitted under sec.  5 of TSCA have had their  cationic group based
on nitrogen.  However, %a-N, cation equivalent weight (EQWT),  and
number of cations per 1000 MW (#C/K) are all equivalent expressions
of cationic charge density.  To convert from  %a-N to EQWT  or #C/K
the following equations can be used:

     1400 * %a-N = N-EQWT or cation EQWT;  and

     %a-N x 0.714286 = #C/K.

To convert cation EQWT to #C/K:

     1000 + EQWT = #C/K.

     The  structure activity  relationships (SAR)  for polycationic
polymers are grouped according  to type of toxic effect and  type of
polymer backbone. All  SARs are based on toxicity data for polymers
listed in Table 7;  100% active ingredients (ai) of polymer  or 100%
polymer solids;  test  dilution water with less than  2.0 mg/L of
total organic carbon (TOC) and  hardness  <180.0 mg/L as CaCO3;  pH
near 7;  and nominal concentrations.

A.  Fish acute toxicity:  96-h LC50

     (1) Carbon-based backbone

          (a) %a-N <3.5
log fish 96-h LC50 (mg/L) = 1.209 -  0.462  %a-N;
where n = 19 and R  = 0.66;

          (b) %a-N >3.5%
fish 96-h LC50  =  0.280  mg/L;  where  n = 34;

-------
                                                                23

      (2) Silicone-based backbone

           (a) %a-N <3.5
log fish 96-h LC5Cfe (mg/L)  =  2.203  -  0.963  %a-N;
where n = 4 and R  =0.73;

           (b) %a-N >3.5%
fish 96-h LC50 = 1.17 mg/L;  where n =  1;  and

      (3) Natural-based backbone

           (a) %a-N <3.5

These polymers will have  either similar toxicity to carbon-based
backboned polymers or less toxicity.  Polymers based on  tannin  (77,
78) have similar toxicity; polymers based on chitin or glucosamine
(79) are 2000 times less  toxic;   and polymers based on starch  (71
thru 76) are  either  10 times more  toxic  or 80  times less toxic.
SAR  analysis for  cationic polymers  with  natural-based polymer
backbones will require the  nearest  analog method to predict fish
acute toxicity.

           (b) %a-N >3.5%

The only datum available (79) indicate less toxicity than predicted
using SARs for carbon-based polymer backbones.
B.  Daphnid acute toxicity:  48-h LC,

     (1) Carbon-based backbone
50
          (a) %a-N <3.5
log daphnid 48-h LC50  (mg/L) =  2.839 -  1.194  %a-N;
where n = 7 and R  =0.90;

          (b) %a-N >3.5%
daphnid 48-h LC50 = 0.100  mg/L;  where  n =  13;

     (2) Silicone-based backbone

Data for  silicone-based  polymers  (Table  7)  indicate that  acute
toxicity towards daphnids will be either similar to  carbon-based
backboned polymers or less toxic.   SAR analysis will require  the
nearest analog method to  predict daphnid acute  toxicity:   use  the
most toxic nearest analog.

     (3) Natural-based backbone

          (a) %a-N <4.3
log daphnid  48-h LC50  (mg/L) =  2.77  -  0.412 %a-N;
where n = 6  and R  =0.82;

-------
                                                                24

           (b) %a-N >4.3%
daphnid 48-h LC50 < 11.0 mg/L;   where  n = 1;

Cationic polymers with natural-based  backbones are generally  less
toxic  than predicted relative  to  carbon-based  polymers with the
same charge density.

C.  Green  algal  toxicity:  96-h EC50

     (1) Carbon-based backbone

           (a) %a-N <3.5
log green  algal  96-h EC50  (mg/L) =  1.569  -  0.97 %a-N;
where n =  5 and  R  =0.54;

           (b) %a-N >3.5%
green algal 96-h EC50 = 0.040 mg/L;  where  n = 12;

     (2) Silicone-based backbone

Data for silicone-based polymers (Table 7) indicate that toxicity
towards  green  algae  will  be  either  similar  to  carbon-based
backboned  polymers or  less toxic.   SAR analysis will require the
nearest analog  method  to  predict algal  96-h EC50 toxicity value:
use the most toxic nearest analog.

     (3) Natural-based backbone
Cationic polymers with natural-based backbones  are less toxic than
predicted relative to  carbon-based  polymers with the same charge
density.  SAR analysis  will require the nearest analog method:  use
the most toxic nearest analog.

D.  Fish Chronic Toxicity:  Only one polymer (82) has been tested
for fish chronic  toxicity using a fish early-life-stage toxicity
test.   The chronic value (ChV)  was 0.018 mg/L which resulted in an
acute-to-chronic ratio  (ACR)  of 18.0.

E.   Daphnid Chronic Toxicity:   Only  one polymer  (82)  has been
tested  for  daphnid  chronic  toxicity using  the  daphnid  21-d
reproductive  inhibition toxicity test.   The ChV was  0.022 mg/L
which resulted in an acute-to-chronic ratio  (ACR) of 14.0.

F.  Green algal chronic toxicity:  96-h ChV

     (1) Carbon-based backbone

          (a) %a-N <3.5
log green algal ChV  (mg/L) = 1.057 - 1.0 %a-N;
where n = 5 and R  =0.53;

          (b) %a-N >3.5%
                                                                    i

-------
                                                                25

green algal 96-h EC50 =  0.020 mg/L;  where n =  11;

     (2) Silicone-based backbone

The single datum for silicone-based polymers  (Table 7, polymer 62)
indicated that the  algal  ChV was  similar to  the ChVs for carbon-^
based backboned polymers.   SARs  for carbon-based polymers can be
used to predict the  ChV for silicone-based polymers until more test
data are obtained.

     (3) Natural-based backbone

Cationic polymers with natural-based backbones  are less toxic than
predicted relative  to carbon-based polymers  with the same charge
density.  SAR analysis will require the nearest  analog method:  use
the most toxic nearest analog.

Amphoteric polymers:  Toxicity

     Polyamphoteric  polymers  contain  both  cationic and  anionic
moieties,  e.g.,  carboxylic  acids,  in  the  same polymer.    The
toxicity of polyamphoteric polymers is largely determined by the
cationic charge density and the cation-to-anion ratio  (CAR).   As
the  cationic  charge density  increases  the  toxicity to  aquatic
organisms increases, and,  when charge density  is constant, toxicity
also increases with increases in the CAR.  And vice-versa, as the
CAR decreases, toxicity decreases.  The toxicity of polyamphoteric
polymers can be predicted in four  steps:   (1)   The cationic charge
density, i.e., percent  amine-nitrogen  (%a-N),  and the cation-to-
anion  ratio   (CAR)   is  calculated  from  the  polymer  chemical
structure.  (2)  The toxicity of the polymer is predicted assuming
that it is a polycationic polymer  using the charge density and the
SARs for cationic polymers  with carbon-based  backbones.   (3)   A
toxicity reduction  factor (TRF) is calculated  for the polymer for
each effect, e.g.,  fish acute toxicity  and daphnid acute toxicity,
using SARs between the TRF and the CAR.   (4) Finally,  the predicted
toxicity values  (based  on the  assumption that the polyamphoteric
polymer is a polycationic polymer) are multiplied by the TRF.

     The SARs between TRF and CAR  were  developed by (a) predicting
the toxicity of each polymer in Table 8 using the  SARs for carbon-
based  backboned  cationic polymers;     (b)   dividing the  actual
measured toxicity value for the polyamphoteric polymer (Table 8) by
the  predicted toxicity value  and calculating a  TRF for  each
polymer;   and (c)  regressing the  TRFs against the  CARs for each
effect.  The SARs between the TRF  and  the CAR  are:

A.  Fish acute toxicity:  96-h LC50

log TRF - 1.411 - 0.257 CAR;  where n  =  3 and  R2 = 0.86;

B.  Daphnid acute toxicity:  48-h  LC50

-------
                                                                26
log TRF = 2.705 - 0.445 CAR;  where n = 2 and R2 = 1.0;
C.  Green algal toxicity:  96-h EC50

                                  re

D.  Green algal chronic toxicity:  96-h ChV
log TRF = 1.544 - 0.049 CAR;  where n = 2 and R2  =  1.0;   and
log TRF = 1.444 - 0.049 CAR;  where n = 2 and R2 = 1.0.

Cationic and amphoteric polymers:  Mitigation of Toxicity

     From 1970 thru  1984 many polycationic polymers submitted under
sec.  5 of  TSCA as  new chemicals  were  assessed as having  the
potential  of  presenting  an  unreasonable  risk  to  the  aquatic
environment.    As  a  result,  many  polycationic polymers  were
regulated by restricting uses and/or release sites, requiring on-
site  treatment in  addition to  sewage  treatment, and  requiring
toxicity and fate testing.  The impact on the polymer manufacturing
industry became so great that,  in late 1984, the Synthetic Organic
Chemicals Manufacturers Association (SOCMA) began discussions with
the  Environmental   Effects  Branch   (EEB)  of   OPPT  about  the
environmental  risk and  fate of cationic polymers.  SOCMA claimed
(and it was  generally known and accepted by OPPT) that polycationic
polymers were designed to react with dissolved organic carbon (DOC)
in water and  formed neutral insoluble  complexes  (or  floe)  which
were  assumed  to settle  on sediment  surfaces  and accumulate in
sediment. There was  general information in the literature  (1960 to
1970s)  that the toxicity of.  cationic polymers towards  fish  was
reduced in the presence of suspended solids in water.  In addition,
it was proposed that  cationic  polymers  should  not  be  toxic to
aquatic organisms as long as there  was  sufficient DOC in natural
waters to satisfy the exchange capacity of the polymer.  It quickly
became apparent to  SOCMA  and  OPPT that fate studies  would  be of
little use in supplying adequate test data to refine the PEC in the
environment because  (1) CCs were in the low M9/L  (Ppb) range,  (2)
analytical methods  for cationic  polymers  had  detection limits in
the 1  mg/L  (ppm)  range,  and (3)  even the best  analytical method
could  not distinguish between dissolved ancl DOC-reacted polymer.
OPPT wanted  adequate rigorous testing data using standardized test
guidelines to prove that cationic polymers were of low toxicity in
the presence of DOC.  In addition, OPPT wanted to known if cationic
polymers in sediment were bioavailable  to organisms  which fed on
sediment.

     In 1986,  EEB recommended that OPPT consider developmental of
a TSCA sec.  4 test  rule for  existing polycationic polymers because
discussions with SOCMA had resulted in no significant progress on
testing and it was known that there were many unpublished toxicity
studies for cationic polymers  on  the Inventory.  A test rule would
also automatically generate sec.  8(a)  and 8(d)  rules which would
obtain industry data on the manufacturing  of cationic polymers and

-------
                                                                27

those unpublished studies, respectively.   In the same year, OPPT
begin to  design a   standardized  test guideline, which  would be
acceptable to everyone, to measure  the  mitigation of toxicity of
cationic chemicals in the presence of DOC.

     The two most important factors to be decided were (1) what DOC
to use  in the test  and  (2)  what  was the  mean  concentration and
distribution of DOC in U.S.  surface fresh waters.  Humic acid was
selected as the  surrogate DOC based on research done by Gary et al.
1987.  Cary et  al. measured  the mitigation of 4 suspended solids
and five  types  of purified dissolved-organic-carbon  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  (MF) for humic acid were
closest to the  mean  MF for  all of the DOCs  tested.   The mean MF
factor  was  calculated by  OPPT for  each  of  the polymer/species
combinations.  The MF of each DOC was compared to the mean KF 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 was easily
available from chemical supply companies.

     OPPT decided that 10 mg/L of total organic carbon (TOG) was to
be the amount of DOC  at which it would assess the risk of cationic
polymers  to  water-column  aquatic organisms.   This  decision was
based on three facts:  (a) concentrations of humic acid in natural
waters were rarely measured, (b) the average measured  amount of TOC
in natural freshwater  of the U.S. was about 6.79 mg  TOC/L (Lynch
1987) , and (c) 10 mg  TOC/L was a round number close to 7 which errs
on the side of safety.  Lynch (1987) analyzed the USEPA Office of
Water's STORET  Data  Base  for measured  amounts  of  TOC  in U.S.
waters.   Lynch  found 67,994  measurements of  TOC taken  from 1977
through 1987 from all over  the U.S., i.e.,  19  of 23 major river
basins.   These  TOC measurements were lognormally distributed and
skewed toward larger concentrations  of TOC.  , The  geometric mean of
these  data  was  6.79 mg  TOC/L.    Since OPPT  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 DOC.

     By November 1988,  negotiations between SOCMA and  OPPT resulted
in the OPPT test guideline, "Fish Acute Toxicity  Mitigated by Humic
Acid (OPPTS 850.1085)." The  guideline was  distributed in December
1988  and  by October  1989  seven  acute  toxicity   and  toxicity
mitigation  tests using  fish  had  been  submitted  to  OPPT  and
validated by EEB.  As a result, EEB recommended to OPPT that the
proposed  sec.  4 test  rule,  sec.  8 (a),  and sec. 8 (d)  rules for
cationic  polymers be withdrawn.   There  is no current activity on
these polymers within the OPPT Existing Chemical Program.

-------
                                                                28

     The  toxicity  of polycationic  polymers is mitigated  in the
presence  of DOC  {Table 9) .   Fish  acute  toxicity data  for 16
polycationic polymers  have been evaluated  and  validated by EEB.
All of these polymers were  submitted under sec.  5 as new chemicals
and  all  were dropped  from  further  review  after  the mitigation
factor (MF) was  integrated into the risk assessment.   The MF is
defined as  the  amount  of toxicity mitigation due  to  10 mg TOC/L
relative to fish acute toxicity as a  96-h LC50.  Mitigation factors
for cationic polymers which are  (a)  random  reactions  of monomers
and  (b) have minimal oligomer  content,  i.e.,  <25%  <1000 and <10%
<500, are directly  correlated  with  charge density, i.e., percent
amine-N (%a-N).   The structure activity relationships (SAR) are:

A.  For charge densities >3.5 %a-N:
                            MF  =  110.0
where n = 7;
B.  For charge densities between 3.5 to 0.7 %a-N:

                   log MF = 0.858 + 0.265 %a-N
where n = 4 and R  =0.61;   and

C.  For charge densities <0.7 %a-N, MFs have not yet been measured,
but are expected to be <7.0.

     Cationic polymers  which have significant  oligomer content,
i.e., >25% <1000 and  >10%  <500,  can  have MFs significantly lower
than predicted  for polymers with minimal  oligomer  content.   For
example,  polymers 47 and  48  have predicted MFs of 110, however,
their measured MFs were  21  and 26,  respectively.  In addition, MFs
for blocked polymers cannot, at this time, be accurately predicted
using  these  SARs.   Polymer  27   is  a blocked  polymer with  the
cationic group attached  to  the polymer tips and is atypical of the
other polymers in Table 9 which are randomly reacted  polymers.  The
measured MF was  130 while the predicted MF was 11; a difference of
12 times.

          The toxicity of polycationic polymers  is also mitigated
when mixed with sediment.  Toxicity testing with natural sediment
contaminated with cationic polymers and with species which ingest
sediment has shown that  cationic  polymers with charge densities of
£4.2% a-N (or >3 cations/1000 MW or a N-equivalent weight <333.0)
are  not bioavailable to cause  toxicity,   i.e.,  48-h  no-effect-
concentrations  (NEC) >100.0 mg/kg dry weight sediment  (Rogers and
Witt  1989) ,   and  thus  are of  low concern once  transported  to
sediments.  A word of caution is  needed here.  No cationic polymer
with significant oligomer content, i.e.,  >25% <1000  and >10% <500,
has  been  tested.    The   oligomers  are   expected  to  be  more
biologically available for  uptake due to the lower MFs observed for
polymers  with significant  oligomer  content (Table 9) .   Another
possible  exception occurs  when a  cationic polymer is  formulated

-------
                                                                29

with acid in excess of the amount needed to neutralize the polymer,
such  that  the  product's  pH is  approximately 2.   Under  these
conditions the polymer fails to  flocculate DOC  in the water column
and  DOC  apparently does  not  mitigate  acute  toxicity to  fish
(Nabholz et al. 1993a).

     The testing scheme for cationic polymers under TSCA generally
requires  5  tests:    (1)  fish acute toxicity with  clean dilution
water,  i.e., TOC  <2 mg/L;   (2)  daphnid acute toxicity  with clean
dilution  water;    (3)  algal  toxicity  test  with  standard  growth
medium;   (4) fish acute toxicity with 10 mg of humic acid/L added
to clean dilution water;   and (5)  fish acute  toxicity  with 20 mg
humic acid/L added to dilution water.   All tests are done with the
static method;  nominal concentrations of polymer based  on 100% ai;
and TOC concentrations  are measured in the controls of all five
tests just prior to test initiation.  When the testing  results are
validated by EEB,  the three  fish  96-h LC50 values  are regressed
against TOC concentration.   The fish 96-h LCc0 due to the addition
of 10 mg TOC/L is predicted using the regression equation, and this
value is divided by the fish LC50 measured  in clean water to obtain
the toxicity mitigation factor (MF)  due to 10 mg TOC/L.  The MF is
then used to decrease the toxicity or adjust upward all effective
concentrations  (EC)  in the  polymer's  toxicity profile  that are
based on the intrinsic  toxicity of  the cationic polymer in clean
water.  A new CC is then determined from the ECs for the polymer in
the presence of 10 mg TOC/L and  a new risk assessment is done.  If
there is still a potential risk to aquatic organisms in the water
column, chronic toxicity testing with fish and  invertebrates in
clean dilution water is recommended using flow-through or static-
renewal   methods   (with  renewals  every  24   h)   and  nominal
concentrations.

SUMMARY AND CONCLUDING REMARKS

     In the U.S.  the environmental  safety of  chemical  substances
with non-food,  non-drug and non-pesticidal uses is  evaluated by EPA
under the  authority granted by the Toxic Substances Control Act
(TSCA).   For polymers the  scope of TSCA's applicability is very
broad  and includes  such use categories as water  and  wastewater
treatment,  coatings,  household,  and industrial cleaning products
and  manufactured  goods.    TSCA clearly  distinguishes  between
existing  chemical  substances and new  substances  not on the TSCA
Inventory.   To date most  of the environmental  assessments for
polymers  have  focused on  new  polymers,  for  which   prospective
manufacturers or importers must file a  Premanufacture Notice  (PMN)
under  sec.  5  of  TSCA.    The  PMN  process essentially involves.
weighing potential  risks and benefits  for each new substance.

     Polymers  that  meet  certain  requirements  are exempt from
reporting under sec. 5, but  this does not  include most polymers
with aquatic uses such as in water treatment.   Non-exempt polymers
are  subject to  the  normal  PMN  review process,  which includes

-------
                                                                30

chemistry  review,  toxicity evaluation,  exposure  assessment, and
risk assessment/risk management phases.  A unique feature of this
process  is  its high  level  of  reliance  on  structure  activity
relationships  (SAR) to predict missing data.  SAR development has
become a welcome indirect benefit as a result of the absence of any
specific or implied requirement in TSCA that PMN submitters perform
testing as  a precondition for approval.  Spurred by  the need to
make informed  judgments  often with few  or  no  experimental  data,
estimation  methods have  been developed for  most  physical  and
chemical   properties,   some   transformation   processes,   e.g.,
biodegradation,  and  many  toxic  effects  relevant  to  aquatic
exposures.  Unfortunately, these methods, although on balance very
successful for discrete  (class 1)  substances, are of little benefit
for  predicting  the  fate  of polymers.    The  result  is  that
experimental data continue to  play a critical role in environmental
fate assessment of polymers.  SAR methods for environmental aquatic
toxicity are available for all of the broad  classes of polymers.

     Most polymers  in commerce are essentially nonbiodegradable.
Environmental fate assessment  therefore is reduced to an evaluation
of their  potential for  sorption  or precipitation under various
conditions.  Rules of thumb are available to  inform judgments about
these processes, but in general they are based on a  heavy dose of
faith and relatively few  data.  Clearly it  is  in  industry's best
interest to  furnish EPA with relevant fate data.

     Environmental toxicity assessment are based on polymer charge:
anionic, nonionic,  cationic, and amphoteric.  Anionic;  Polyanionic
polymers  are   divided  up  into   two   classes:    poly(aromatic
sulfonates/carboxylates)    and  poly(aliphatic   acids).      Many
poly(aromatic  sulfonates/carboxylates)  are  moderately  toxic  to
aquatic  organisms thru  an unknown  direct acting  mode-of-toxic
action.   The nearest analog-SAR  method is  used  to  predict the
toxicity of these polymers by identifying the dominant monomer(s)
of the polymer.  Poly(aliphatic acids)  show toxicity only towards
green algae.   These polymers indirectly effect algal growth  by
over-chelating  the nutrient  elements   needed  for growth.    The
potency of a polymer to cause toxicity  is directly related to the
ability of the  polymer to chelate  divalent  metals.   The  distance
between acids on the polymer controls the strength of the chelate.
The toxicity of these polymers in  soft  water  environments can be
predicted by determining the average distance between acids.  The
toxicity of  poly(aliphatic  acids)  is mitigated by changing  their
salt to a  divalent salt,  e.g.,  Ca *,  or  by  releasing  them  to
moderately hard water.  Environmental risk from these polymers will
most likely be observed in soft natural water which have hardness
of <30.0 mg/L as CaC03.  Nonionic:   Nonionic  polymers are generally
of  low  concern  for  ecotoxicity,   however,  if hydrophobic  and
hydrophilic monomers are blocked in order to use the polymer as a
detergent  or  dispersant,  then  it  could  be  toxic  to aquatic
organisms.    Likewise,   polymers  with  significant  amounts  of
oligomers whose MW <1000 may be of concern due to the toxicity of

-------
                                                                31

the  oligomers  and the oligomers should be  assesses  as monomers.
Cationic;    Cationic  polymers  can be  highly  toxic to  aquatic
organism and should be used with caution.  Cationic polymers can be
grouped by the type of polymer backbone:  carbon-based, silicone-
based, and backbones using a natural polymer such as starch.  The
aquatic toxicity of cationic polymers is primarily related to their
cationic charge density, and, if soluble or dispersible as micro-
or macro-emulsions, can be reasonably  predicted by  SARs based on
regression equations.  The toxicity of Cationic polymers based on
silicone or  a natural  polymer  can be  less toxic relative  to a
polymer  based  on   carbon,   but  not  always.     Amphoteric:
Polyamphoteric polymers contain  both cationic and anionic moieties
in the same polymer.  The toxicity  of these polymers is determined
by the  cationic  charge density and  the cation-to-anion  ratio.
Toxicity increases  as charge density increases  and,  when charge
density is held constant, toxicity  increases with increases in the
cation-to-anion ratio.  Polyamphoteric polymers  can also be highly
toxic to aquatic organisms.   SARs for polyamphoteric polymers are
based on calculating the cationic charge density  and the cation-to-
anion ratio from polymer chemical structure.

     The toxicity  of  cationic   and araphoteric  polymers can  be
mitigated by dissolved organic carbon (DOC) which occurs naturally
in surface waters.  Toxicity  mitigation in  the  presence of 10 mg
total organic  carbon/L  is related  to cationic  charge density and
can be predicted by SARs.  If  cationic and amphoteric polymers are
used according to current standard  practices, then they are of low
environmental risk.

     Environmentally safe manufacture and use of polymers is best
assured if EPA and  industry  cooperate  to  the extent  and from the
earliest phase of  product  development practicable  for the PMN
submitter.    Much more use of the New  Chemical's Program's Pre-
Notice Communication process  could be made  than is  the case now,
and the effect can  only be beneficial  for all  stakeholders.  The
most  important information  needed by  EPA  assessors  is  polymer
chemical structure.   Specific  data about  polymer  structure  is
critical to the whole assessment process.  Data needs include, but
are not limited to, monomer identity;  polymer composition, e.g.,
mole ratios of monomers;  reaction sequence of monomers:  random or
blocked;   average-number  molecular  weight;    molecular  weight
distribution;    and oligomer content,  i.e.,  percent  <1000 and
percent <500 daltons.

     If  testing  is  considered  necessary,  either   by  voluntary
agreement  or  Consent  Order,  to  resolve  uncertainties,  test
protocols should be reviewed  by EPA before any testing commences.
Further, what is being tested should be understood and agreed to by
stakeholders; obviously this is  particularly important for class 2
substances.   Through  a combination of (1) testing  tailored to
specific substances subject to PMN review and
(2)  more  fundamental  research  on the  mechanisms of  polymer

-------
                                                                32

sorption,  precipitation,  and  toxicity  and  its  mitigation,  the
database needed for truly  informed ris.k assessment can be realized.
DISCLAIMER

This document  has been  reviewed by  OPPT,  EPA and  approved for
publication.    Approval  does  not  signify  that  the  contents
necessarily reflect the views and policies of the Agency nor does
mention of tradenames or commercial products constitute endorsement
or recommendation for use.

-------

-------
                                                                 33

LITERATURE  CITED

Alexander,  M.    1973.    "Nonbiodegradable and other  recalcitrant
     molecules,1* Biotechnology and Bioengineering.   15:611-647.
American  Society  for   Testing  and  Materials   (ASTM).     1994.
     "Standard  practice for coagulation-flocculation jar  test of
     water,"  D  2035-80,  in 1994 Annual  book of ASTM  standards.
     Section  11.    Water  and  environmental  technology,   volume
     11.02, pp.  751-753.   Philadelphia,  PA:  ASTM.
Bailey,   W.J.   and  B.   Gapud.     1986.     "Synthesis   of   new
     biodegradable  polymers,"  in  Biodeterioration 6.  Proceedings
     of  the sixth  international  biodeterioration  symposium,  S.
     Barry  and  D.R. Houghton,  eds., pp.  605-610.  Slough,  United
     Kingdom: CAB International.
Barnes, D.G. and M.L. Dourson.   1988.   "Reference dose  (RfD):
     description and use  in health risk assessment," Regulatory
     Toxicology  and Pharmacology.   8:471-486.
Biesinger,  K.E.  and G.N.  Stokes.    1986.   "Effects of synthetic
     polyelectrolytes on selected aquatic organisms," Journal of
     the Water Pollution  Control Federation.  58:207-213.
Gary, G.A., J.A.  McMahon  and W.J.  Kuc.   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.
Clark, B.,  J.G.  Henry and  D. Mackay.   1995.  "Fugacity analysis
     and  model  of  organic chemical  fate in a  sewage treatment
     plant," Environmental  Science and Technology.  29:1488-1494.
Clements, R.G.  (ed) .   1988.    .Estimating toxicity  of industrial
     chemicals  to  aquatic  organisms  using structure  activity
     relationships,  EPA report no.  560/6-88-001  (NTIS  no. PB89-
     117592).  Washington, DC:   Office of Pollution Prevention  and
     Toxics.
Clements, R.G.  and  J.  V. Nabholz.   1994.   ECOSAR:   A computer
     program for estimating the  ecotoxicity of industrial chemicals
     based on structure  activity relationships, EPA report no. 748-
     R-93-002  (NTIS no.   PB94-500485  [program]   and  PB94-104668
     [user's  guide]).     Washington,   DC:    Office  of  Pollution
     Prevention  and  Toxics.
Clements,  R.G.,  J.V.   Nabholz, D.E.  Johnson  and  M.G.   Zeeman.
     1993.      "The  use  of   quantitative  structure-activity
     relationships  (QSARs) as  screening tools  in  environmental  j
     assessment,"    in    .Environmental   toxicology   and   risJc
     assessment:  2nd volume, ASTM STP  1216, J.W.  Gorsuch, F.J.
     Dwyer, C.G.  Ingersoll and T.W.  LaPoint, eds.  Philadelphia,
     PA: American Society for Testing  and Materials.
Halverson,  F. and H.P.  Panzer.   1980.   "Flocculating agents,"  in
     Kirk-Othmer Encyclopedia  of Chemical Technology, volume  10,
     3rd edition, pp. 489-523.  New  York: Wiley.                    j
Health Council  of the  Netherlands (HCN).   1989.   Assessing  the
     risk of toxic  chemicals for ecosystems.  The Hague, The
     Netherlands: HCN.                           U.S. EPA Headquarters Ubrarv
                                                     _Ma" code 3201
                                                     Pennsylvania Avenue NW
                                                   Washington DC 20460

-------
                                                                34

Leo, A. and D. Weininger.  1985.  "CLOGP Version 3.3:  Estimation
     of the n-octanol/water partition coefficient for organics in
     the  TSCA Industrial  Inventory."  Computer Program.   Pomona
     College, Claremont, CA.
Lynch, D.G.   1987.   "Summary  of STORET data on dissolved organic
     carbon  (DOC)  levels  in surface  waters,"  Memorandum dated 15
     June.  Washington,  DC:  Engineering,  Economics, and Technology
     Division, Office of Pollution Prevention and Toxics, USEPA.
Lynch,  D.G.,  N.F.  Tirado, R.S.  Boethling,  G.R.  Huse  and G.C.
     Thorn.    1991.    "Performance of  on-line  chemical  property
     estimation   methods   with   TSCA   Premanufacture   Notice
     chemicals," Ecotoxicology and Environmental Safety.  22:240-
     249.
Meylan, W.M.  and P.H.  Howard.  1991.   "Bond  contribution method
     for   estimating   Henry's   Law   constants,"   Environmental
     Toxicology and Chemistry.  10:1283-1293.
Meylan, W.M. and P.H. Howard.   1995.   "Atom/fragment contribution
     method for estimating octanol-water partition coefficients,"
     Journal of Pharmacological Science.  84:83-92.
Meylan, W.M., P.H. Howard  and R.S.  Boethling.  1992.  "Molecular
     topology/fragment  contribution  method  for predicting soil
     sorption    coefficients,"   Environmental   Science    and
     Technology.  26:1560-1567.
Meylan, W.M., P.H.  Howard and R.S.   Boethling.   1995.   "Improved
     method  for estimating water solubility  from octanol/water
     partition   coefficient,"   .Environmental   Toxicology   and
     Chemistry,   in press.
Moss,  K.,  D.  Locke, and  C.  Auer.   1996.   "EPA's  New Chemicals
     Program,"  Chemical health & Safety. Jan/Feb:29-33.
Nabholz, J.v.   1991.   "Environmental  hazard  and risk assessment
     under the United States Toxic Substances  Control Act (TSCA),"
     The Science of the Total Environment. 109/110:649-665.
Nabholz, J.V.,  P.  Miller and M.  Zeeman.   1993a.   "Environmental
     risk assessment of new chemicals  under the Toxic Substances
     Control  Act (TSCA) section  5," in Environmental toxicology
     and risk assessment, ASTM STP 1179, W.G.  Landis, J.S. Hughes
     and M.A. Lewis, eds., pp. 40-55.  Philadelphia, PA: American
     Society  for Testing and Materials.
Nabholz, J.V., R. G. Clements, M. G. Zeeman,  K. C. Osborn, and R.
     Wedge.      1993b.      "Validation  of   structure  activity
     relationships  used  by  the  USEPA's  Office  of  Pollution
     Prevention and Toxics for the environmental hazard assessment
     of industrial chemicals," in Environmental toxicology and risk
     assessment:  2nd volume, ASTM STP 1216,  J. w. Gorsuch, F. J.
     Dwyer, C. G. Ingersoll, and T.  W.  LaPoint,  eds., pp. 571-590.
     Philadelphia, PA: American Society for Testing and Materials.
Nalco  Chemical  Company.   1988.   The NALCO water  handbook.  New
     York, NY:   McGraw-Hill Book  Company.
National Research  Council  (NRC).   1994.  Science and judgment in
     risk  assessment.     Washington,   DC:  National  Academy  of
     Sciences.
Opgenorth, H.J.  1992.  "Polymeric materials polycarboxylates," in

-------
                                                                35

     The Handbook  of Environmental Chemistry, 0. Hutzinger,  ed.,
     Volume 3 Part F,  Anthropogenic compounds.  Detergents, N.T.  de
     Oude, ed., pp.  337-350.  Berlin: Springer-Verlag.
Park,  J.K.,   D.   Jenkins,  T.M.  Holsen,  T.W.  Warnock  and  W.E.
     Gledhill.   1989.   "Fate  of  the  detergent  builder,  sodium
     polyglyoxylate,  in wastewater  treatment,"  Journal  of the
     Water Pollution  Control Federation.  61:491-499.
Pranamuda,  H.,  T.  Tokiwa  and  H. Tanaka.    1995.   "Microbial
     degradation  of an aliphatic  polyester with a high melting
     point,   poly(tetramethylene    succinate),"    Applied   and
     Environmental Microbiology.   61:1828-1832.
Rodier, D.J.  and  D.A.  Mauriello.   1993.  "The quotient method  of
     ecological  risk  assessment  and  modeling  under  TSCA:  a
     review,"  in  Environmental  toxicology  and  risk assessment,
     ASTM  STP 1179,  W.G.  Landis,  J.S.  Hughes  and  M.A.   Lewis,
     eds., pp.  80-91.   Philadelphia,  PA:  American  Society for
     Testing  and Materials.
Rogers, J. R. Jr.  and W.  T.  Witt.   1989.   "Effects of sediments
     flocculated with cationic  polyelectrolytes  when fed upon  by
     Daphnia  magna.    Denton,   TX:     Department   of  Biological
     Sciences, University of North Texas.   Available from  SOCMA,
     1330 Connecticut Avenue, N.W., Washington, DC  20036.
Smrchek, J.,  R.   Clements,  R.   Morcock,  and W.  Rabert.    1993.
     "Assessing  ecological  hazard  under  TSCA:    methods  and
     evaluation of  data,"  in  Environmental toxicology  and risk
     assessment,  ASTM STP 1179, W.G.  Landis, J.S. Hughes and
     M.A.  Lewis,  eds., pp.  22-39.   Philadelphia,   PA:  American
     Society  for Testing and Materials.
United States Environmental Protection Agency (USEPA).  1982.
     Fate  of priority pollutants  in  publicly  owned  treatment
     works, EPA report  no. 440/1-82-303.   Washington,  DC:  Office
     of Water.
United  States Environmental  Protection Agency  (USEPA).   1985.
     .Exposure analysis  modeling system  (EXAMS): user  manual and
     system documentation,  EPA  report  no.  600/3-82-023 (NTIS no.
     PB82-258096).    Washington,  DC:  Office  of  Research  and
     Development.
United States Environmental Protection Agency {USEPA).  1988.
     User's guide  to PDM3.   Washington,  DC:  Office of Pollution
     Prevention and Toxics, USEPA.
United States Environmental Protection Agency (USEPA).  1990.
     Statistical methods for estimating exposure above the
     reference dose,  EPA report no.  600/8-90-065 (NTIS no. PB90-
     261-504AS).       Washington,   DC:    Office   of   Health  and
     Environmental Assessment, USEPA.
United  States Environmental  Protection Agency  (USEPA).   1994a.
     A  review of  ecological assessment cas studies from a risk
     assessment perspective.   Volume II, Part II, Section 1:  New
     chemical case study, EPA report no.  630-R-94-003. Washington,
     DC:  Office of Research and Development, USEPA.
United  States Environmental  Protection Agency (USEPA).   1994b.
     U.S. EPA/EC joint project  on the evaluation of  (quantitative)

-------
                                                                36

     structure activity relationships, EPA report no. 743-R-94-001.
     Washington, DC:  Office of Prevention, Pesticides, and Toxic
     Substances, USEPA.
United States Environmental Protection Agency  (USEPA).  1995a.
     Amendment for expedited process to issue significant new use
     rules  for selected  new  chemical  substances;    final  rule,
     Federal .Register.  60:16311-16316.
United States Environmental Protection Agency  (USEPA),  1995b.
     Premanufacture  notification  for  exemption;    revision  of
     exemption  for  chemical   substances   manufactured  in  small
     quantities;  low release  and exposure  exemption;  final rule,
     Federal .Register.  60:16336-16351.
United States Environmental Protection Agency  (USEPA).  1995c.
     Premanufacture   notification  exemptions;     revisions  of
     exemptions  for  polymers;    final rule,  Federal  Register.
     60:16316-16336.
United States  Environmental  Protection Agency  (USEPA).   1995d.
     New  chemicals   program.     EPA  report  no.   743-F-95-001.
     Washington, DC:  Office  of  Pollution  Prevention and Toxics,
     USEPA.
United states Environmental Protection Agency  (USEPA).  1996.
     Polymer exemption guidance manual.  Washington, DC: Office of
     Pollution Prevention and Toxics, USEPA.  Available through the
     TSCA Assistance Information Service at 202-554-1404  (voice) or
     202-554-5603 (facsimile).
Wagner, P.M.,  J.  V.  Nabholz,  and R. J. Kent.   1995.   "The new
     chemicals  process at  the  Environmental  Protection  Agency
     (EPA):      structure-activity   relationships   for   hazard
     identification  and  risk assessment,"  Toxicology  Letters.
     79:67-73.
Zeeman, M. and J. Gilford.   1993.  "Ecological hazard evaluation
     and risk assessment under EPA's Toxic Substances Control Act
     (TSCA):  an  introduction,"  in .Environmental toxicology and
     risk assessment, ASTM STP 1179, W.G.  Landis, J.S. Hughes and
     M.A.   Lewis,  eds.,  pp.   7-21.    Philadelphia,   PA:  American
     Society for Testing and Materials.

-------
                         FIGURE LEGENDS

FIGURE 1.  Total number of valid Premanufacture Notices received
           by EPA for the period 1979-1994.

FIGURE 2.  Premanufacture Notice review process.

-------

-------
Number of Submissions
                        i
                              W

                              30
                              (D
                              o
                              2.
                              a

                              
-------

-------
    Duusaj.
                     IU9SUOO (a)s
                         I
bHNS
UJ80UOO
                   JO dOJQ
(*9IAajAep-&) SUOUdUiaxliaveHl&l*
  ajnpejnuew oj aajj BJB
Aam jBgj pa^iioN Auedutoo
       uo punoj
                        «u
                                         snooj
            wjodapofaftpsseg
               sasBajay PUB
            ajnsodjg p samoj
-------

-------
TABLE 1.  Criteria  for  substantial  or significant exposure in
Exposure-Based Review  (XBR)  of  Premanufacture Notices
Type of exposure
Exposure criterion
Worker exposure:
substantial or significant

     High number of workers

     Acute worker  inhalation
     Chronic worker exposure
          Inhalation
          Dermal
Non-worker exposure

     Consumer exposure via
     direct contact with
     consumer products
     General population:
     significant exposure
          Surface drinking
          water

          Air

          Ground drinking
          water

     Environment:
     substantial exposure
          Aggregate ambient
          exposure through
          surface water, air
          and ground water

          Substantial  release
          to water
>. 1,000 workers exposed

>_ 100 workers exposed by
inhalation to >. 10 mg/day
> 100 workers exposed to  1-10
mg/day for > 100 days/yr

.> 250 workers exposed by
routine dermal contact
> 100 days/yr
Presence of the chemical in any
consumer product where the
physical state or manner of use
make exposure likely
> 70 mg/yr


>. 70 mg/yr via ambient air

> 70 mg/yr
Total release to environmental
media > 10,000 kg/yr
Release to surface water
> 1,000 kg/yr after treatment
                                              U.S. EPA Headquarters Library
                                                   Mail code 3201
                                              1200 Pennsylvania Avenue NW
                                                Washington DC 20460

-------

-------
TABLE 2.  Fate testing strategy for aquatic exposures in
Exposure-Based Review (XBR)
Exposure criterion
Recommended tests
Production volume > 100,000
kg/yr and at least one other
XBR criterion8
Production volume > 454,000
      (1,000,000 lb/yr) and at
least one other XBR criterion
Wastewater treatment removal:
Semi-Continuous Activated
Sludge (SCAS),  Porous Pot,  or
Activated Sludge Sorption
Isotherm

-and-

One of the following: water
solubility,  vapor pressure,
soil/sediment sorption iso-
therm, ready biodegrada-
bility, or hydrolysis

Wastewater treatment removal:
as above

-and-

At least two of the following:
water solubility, vapor
pressure, soil/sediment sorption
isotherm, ready biodegrada-
bility, anaerobic biodegrada-
bility, or hydrolysis
  General  population:  exposure via  surface  drinking  water of
  > 70 mg/yr; and environment: release to surface water of
  > 1,000 kg/yr.

-------

-------
TABLE 3.  Biodegradation and elimination of polycarboxylates in
tests with activated sludge8
Polycarboxylate5 Ultimate SCAS test
biodegradation (DOC
(TC02), %c loss), %d
P(AA) 1,000
P(AA) 2,000
P(AA) 4,500
P{AA) 10,000
P(AA) 60,000
P(AA-MA) 12,000
P(AA-MA) 70,000
45
20
9
16
ND
31
20
45
21
40
58
93
83
95
Confirmatory
test
(removal) , %
NDe
ND
27
ND
ND
70
82
8 Source:  Opgenorth (1992).

b p = poly;  AA = acrylic acid; MA = maleic acid; numbers given
  are approximate MWn.

c Percent  of theoretical C02  formation after  contact  with
  activated sludge for 30-90 d.

d SCAS = Semi-Continuous Activated Sludge; DOC = dissolved
  organic carbon.

e ND = not done.

-------

-------
                                                                          1
Table  4.    Aquatic  toxicity  data  for  polyanionic  polymers:
poly{aromatic   sulfonates)    and   poly(aromatic   carboxylates).
Polymers are grouped by dominant monomer.  Data elements are/ from
left  to  right:     ID  No.   =  Polymer  identification  number;
MWn/1/40/15 5.0 145.0 20.0
1/7/15 5.1 188.8 44.0
>0. 4/100/10 9.7 136.0 9.3
1/7/15 23.0 950.0 12.0
Biphenol sulfone-SO,H and cresol-SO,H
GAChV


<12 .5
7.0
18.0
<12. 5

<7. 5
6.2
  7
  7
  8
  9
 10
 11
1.1/0/0          3.2

    Biphenol sulfone-S03H and biphenvl ether-S03H

                           120.0
0.6/95/40
0.6/95/40
12/7/0
1
80.0
32.0
                 2.0
               510.0
                 3.0
                <2.3
                           Phenol-S03H
                           Cresol-SO3H

-------
Table 4. Continued.
ID   MWr/500.0      340.0

      Benzene-S03H

600.0       900.0      800.0
                                                    44.2

-------
Table  5.    Aquatic  toxicity  data  for  polyanionic  polymers:
poly(aliphatic acids).   Polymers  are grouped by distance between
acids which is measured by counting the  average number of polymer-
backbone-carbons between acids*  Acids (A) include carboxylic acid,
phosphinic acid, and sulfonic acid.  Data elements are, from left
to right:  ID No. = Polymer  identification number;   KWn/339.0      560.0
16
16
0.5
0.5
1140.0
2500.0
Distance between acids; 1 carbon:  [C(A)C],
 17  3
 18  3.5
 19  2.5/5/3
 20  0.9/49/19
 21  10/35/5
 22  1.4/41/8
                   330.0
                  >500.0
                  >225.0
                          >440.0
                          >500.0

                         >1000.0
                          7.44
                          3.13
                         37.4
                         11.0
                          7.6
                          5.5
Distance between acids; 1.5 carbons:  [C(A)C(A)C]X

 23  0.6/88/69                  500.0       66.0

Distance between acids; 2 carbons;  [C(A)CCC(A]X

 24  3/0/0                                 150.0
280.0
 24.0
  4.7
  0.500
  0.540
                                                       93.0
Distance between acids; l carbon;  distance from backbone;
3 carbons;  [C(A)C]X + [C(CCCA)C]X

 25 1/15/5        >1000.0      1800.0       57.0        36.0

Distance between  acids;  1 carbon;chelating monomer diluted bv
non-chelatina monomer;  [C(COOH)C]3 + [C(CONH2)C]1

 26 >1/10/1         >40.0       >40.0     >500.0      >100.0

-------

-------
Table  6.    Mitigation  of  toxicity  towards  green  algae  from
polyanionic polymers by  calcium  ions  (Ca *).   Calcium is added by
either  increasing the  hardness of  the algal  growth  medium  or
testing the Ca * salt of the polymer.  Data have been grouped by the
average  distance  between  acids  in  terms  of  polymer-backbone-
carbons.  Data elements are,  from left to right:  ID No. = polymer
identification  number;    MWr/
-------

-------
Table 7.  Aquatic toxicity data for polycationic polymers grouped
by  type of polymer  backbone.  • Data  elements are,  from  left to
right:   ID  No.  = Polymer identification number;   A-N = cationic
charge density in percent amine nitrogen;  MW(/5.0
>5.0
>5.0
1/38/12
1/32/23
Pll



P4
P4
P4
B
B
P2
P2
P5
P5
P4
P2
PI
P3
P4
P4
P4
P4.
P5
P4
B
B
B
B
B
3



4
4
4
3
2
3,4
3,4
3
3
3
3
4
2,3
4
4
4
5 4
4
4
4
4
4
4
2
4
9



.
.
8
3
3
10
10
30
17







1


0






0
1
.
0
2
0
1
0
0
0
.
0
0
•
0
0
0
0
0
0
•
*
6
2
.5
.9
.3
.61
.0
.0
0
.970
.3
.640
.2
.840
.900
.320
5
.600
.300
730
.310
.900
.760
.150
.160
.290
720
0

300

.0
310.0




28.2
40.0

16







0


0






0
2

.0
1.7






.090


.180






.073
.9
                                                  1.3       0.160
                                                  2.2       0.880
                                                 >360.0     130.0
                                                  0.520     0.270
                                                  0.035     0.006
                                                  0.025     0.013
                                                  0.014     0.006
                                                  0.015     0.007

-------
Table 7. continued.
ID
NO.
A-N MW,/ Cat. C
(%) 
-------
Table 7. Continued.
ID   A-N    MW,/   Cat. Cat.   F96
No.  (%)  850
>1000
>1000
>1000
1
0

0
1.
570
*
*
m
•
•
*

•
5
*
0
0
0
0
18
370

860

0

117.
>1000.
>1000.


130.

26.0
11.
>1000.0
0
0
0


0

0.620
0 >480.0
>1000







.0







0.390
>480
.0

-------

-------
Table 8.  Aquatic toxicity data for polyamphoteric polymers.  Data
elements are, from left to right:  ID No. = Polymer identification
number;  A-N = cationic charge density in percent amine nitrogen;
MWr/
-------

-------
Table 9.   Mitigation of  aquatic toxicity from  polycationic and
polyamphoteric polymers by  humic acid.  Data  elements are, from
left to  right:   ID No. = Polymer identification number;   A-N =
cationic charge density in percent amine nitrogen;  MWr/
-------
Table 9. Continued.
ID
No.
44
44
44
44
47
47
47
82
82
82
82
82
82
82
82
82
48
48
48
83
83
83
54
54
54
54
54
56
56
56
56
56
58
58
58
58
58
58
A-N
(%)
4
4
4
4
6
6
6
6
6
6
6
6
6
6
6
6
7
7
7
8
8
8
11
11
11
11
11
14
14
14
14
14
14
14
14
14
14
14
•
*
*
*
m
*
*
•
•
•
•
•
•
*
•
•
•
»
•
*
•
•
•
•
•
»
*
*
•
•
•
•
*
*
*
•
*
•
6
6
6
6
4
4
4
6
6
6
6
6
6
6
6
6
8
8
8
4
4
4
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
MW,/
2
>2
>20
>20
0
0
3
, 5
10
0
0
0
2
5
9
•
*
•
•
•
*
•
*
•
*
•
•
•
•
*
•
•
•
•
•
•
•
•
•
*
•
•
•
•
•
•
•
*
•
•
•
•
•
900
0
0
0
720
2
9
280
380
600
6
6
1
2
0
0
0
0
0
0
0
0
200
0
0
0
0
100
160
0
36
2
072
084
160
2
4
6
Mitigation
Factor



76.0


21.0








290.0


26.0


27.0




>180.0




140.0





170.0

-------