	UNITED STATES ENVIRONMENTAL PROTECTION AGENCY

	WASHINGTON, D.C.  20460

OFFICE OF PREVENTION, PESTICIDES AND TOXIC SUBSTANCES

March 25, 2002

MEMORANDUM

SUBJECT:	Diquat Dibromide  HED Risk Assessment for Tolerance
Reassessment Eligibility Document (TRED.)  PC Code No:  032201;  DP
Barcode: D281890; Submission Barcode: S611057

FROM:	Becky Daiss, Environmental Health Scientist

Reregistration Branch 4

Health Effects Division (7509C)

THROUGH:	Susan V. Hummel, Branch Senior Scientist

Reregistration Branch 4

Health Effects Division (7509C)

TO:		Tyler Lane, Chemical Review Manager

Betty Shackleford, Branch Chief

Registration Branch

Special Review and Reregistration Division (7508C)

Attached is the Health Effects Division (HED’s) revised risk
assessment conducted to  support a Tolerance Reassessment Eligibility
Decision (TRED) for diquat dibromide.  This document updates the March
6, 2001 version by incorporating revisions to the drinking water
exposure assessment provided by the Environmental Fate and Effects
Division (EFED).  The disciplinary science chapters and other supporting
documents are included as appendices as follows:

Use Closure Memo. Tyler Lane (10/31/01)

Report of the Hazard Identification Assessment Review Committee. Linda
Taylor ( 12/14/01, HED DOC NO 014670)

Report of the FQPA Safety Factor Committee.  C. Christiansen (11/27/10,
TXR NO. 0050293 )

Report of the Metabolism Assessment Review Committee. T. Morton
(9/13/01, D277764)

Product & Residue Chemistry Chapter. T. Morton  (12/12/01, D277710)

Residential  Exposure Assessment. T. Brennan (12/14/01, D279507)

Dietary Exposure and Risk Estimates for Tolerance Reassessment.  B.
Daiss (12/11/01, D277766)

Incident Report.   M. Spann and J. Blondell, Ph.D (10/15/01, D278482)

Tier I Drinking Water and Aquatic Ecological Exposure Assessments for
Diquat Dibromide. J Breithaupt  (3/5/02, D281199)

Toxicology Chapter, Linda Taylor (12/14/01, D279696)

	Table of Contents					

          										 	                     pg.

1.0 	EXECUTIVE SUMMARY	3

2.0 	PHYSICAL/CHEMICAL PROPERTIES CHARACTERIZATION		8

3.0 	HAZARD CHARACTERIZATION	9

3.1 	Hazard Profile		9

3.1.1	Acute Toxicity	9

3.1.2	Toxicity Profile	10

3.1.3	Hazard Characterization	14

3.2 	FQPA Considerations	14

3.3 	Dose Response Assessment	15

3.3.1 	Dietary Endpoints	15

3.3.2 	Residential Exposure Endpoints	16

3.4 	Endocrine Disruptor Effects	18

4.0 	EXPOSURE ASSESSMENT	19

4.1 	Summary of Registered Uses	19

4.2 	Dietary Exposure/Risk Pathway	20

4.2.1 	Residue Profile	20

4.2.2  	Dietary Exposure/Risk Assessment	24

4.3	Drinking Water Exposure/Risk Pathway	26

4.3.1 	Environmental Fate Assessment	26

4.3.2	Estimated Environmental Concentrations	26

4.4	Residential Exposure/Risk Assessment	29

4.4.1	Home Uses	29

4.4.2	Recreational	36

4.5	Incident Report Summary	38

4.5.1	Residential Handler Incidents	39

4.5.2	Post-Application and Recreational Incidents	39

5.0	AGGREGATE RISK ASSESSMENT AND RISK CHARACTERIZATION	39

5.1	Acute Aggregate Risk Assessment	39

5.2	Short-Term Aggregate Risk Assessment	40

5.3	Chronic Aggregate Risk Assessment	42

6.0	CUMULATIVE EXPOSURE AND RISK	43

7.0	DATA NEEDS	43

1.0 	EXECUTIVE SUMMARY

EPA issued a Reregistration Eligibility Decision (RED) for diquat
dibromide in July, 1995.  The RED recommended revisions to some of the
existing tolerances for diquat dibromide.  The purpose of this document
is to reassess the findings and conclusions presented in the RED to
determine whether infants and children exhibit enhanced sensitivity from
dietary and/or residential exposure to diquat dibromide.  This action is
required under the Food Quality Protection Act (FQPA) of 1996. 
Occupational exposure/risk is not considered for this TRED.  However, a
separate but concurrent reassessment of occupational exposure/risk is
also being conducted to address the registrant’s request to modify
diquat dibromide label requirements (T. Brennan, D279612).

Use Profile

Diquat dibromide is a non-selective contact herbicide, algicide,
dessicant, and defoliant.   As a herbicide/algicide, it is used to
control aquatic and terrestrial weeds.  It is used as a preharvest
dessicant/defoliant to facilitate the harvest of potatoes and various
crops grown for seed.  Its largest use is as a dessicant on potato
crops.  Other minor food use applications include use as a dessicant on
crops grown for seed that is used for feed, i.e., alfalfa, sorghum,
soybean, and clover.  Additionally, use of irrigation water containing
residues of diquat may result in diquat residues in plants and
livestock.  Non-food use applications include use as a preharvest
dessicant on carrot, radish and turnip grown for seed; as a post-harvest
dessicant on cantaloupe, cucumber, pepper, squash, tomato, and
watermelon; and as a dessicant/defoliant for  commercial greenhouses and
nurseries, ornamental seed crops, and commercial and residential
landscaping and grounds maintenance.  Diquat dibromide is rapidly
absorbed by green plant tissue and interacts with the photosynthesis
process.  It works as an herbicide/dessicant by reacting with molecular
oxygen to produce a superoxide anion in treated plants.  The oxidative
activity, which occurs subsequent to formation of the oxygen radicals,
rapidly destroys plant cell membranes.  Herbicidal activity is usually
quite rapid with effects visible in a few days.

Diquat dibromide is formulated as a soluble concentrate and ready-to-use
liquid.  As an  herbicide/algicide used to control algae and aquatic
weeds, it is applied by direct pouring, hand-held or mechanical sprayer,
and injection below the water surface.  As a terrestrial herbicide and
crop dessicant/defoliant, it is applied by hand-held sprayer, aircraft,
or ground equipment. 

Currently, 43 products containing the active ingredient diquat dibromide
are registered and marketed under the trade name diquat dibromide.   The
largest markets for diquat, in terms of total pounds of active
ingredient, include: aquatic uses (40%), potatoes (35%), home & garden
(10%), and alfalfa for seed (5%).  Most of the usage is in FL, ME, ND,
NY, WA, and WI (T. Lane, Use Closure Memo, 10/31/01).

Regulatory History

In July, 1995, EPA issued a RED for diquat dibromide.  In the RED, the
Agency recommended changes to published tolerances for several
commodities, including fish (increase from 0.1 to 2.0 ppm) and fruiting
vegetables (increase from 0.02 to 0.05 ppm).  The recommended changes to
the tolerances have not yet been implemented.  Based on the risk
assessment conducted for the RED, EPA concluded that acute and chronic
dietary risks from exposure to diquat are minimal, but found risks of
concern for some worker and residential exposure scenarios.  To address
these risks EPA required the following: closed mixing/loading of diquat
dibromide for aerial applications; a longer interim restricted entry
level (REI) and more stringent personal protective equipment (PPE) for
uses within the scope of the Worker Protection Standard for Agricultural
Pesticides; a four-day reentry interval to reduce post-application
exposure for workers not covered by the worker protection standard
(e.g., golf course workers); and a ban on broadcast treatments to reduce
the potential for post-application residential exposure.  Diquat labels
have not yet been revised in accordance with the requirements outlined
in the RED.  The RED also addressed the occurrence of ethylene dibromide
(EDB) which is used as a starting material in the manufacture of diquat
dibromide and may be present as a process impurity in final
formulations.  EDB is considered a carcinogen and all pesticide uses of
EDB have been canceled.  Previous EPA assessments indicate that the
presence of EDB in diquat formulations does not pose a significant
dietary risk.  In addition, the registrant certified an upper limit of
10 ppm EDB in diquat dibromide, and demonstrated that EDB does not
persist as an impurity in diquat dibromide. 

In January, 1998, HED conducted a human health risk assessment to
evaluate the effect of a change in application and use patterns on
potatoes and a new use on dried shelled peas and beans (W Cutchin,
1/23/98, D220714).  Results from the 1998 assessment were consistent
with the RED, as were the recommendations for additional restricted
entry and PPE requirements. 

Only manufacturing use products (MPs) are subject to a reregistration
eligibility decision. There are currently only 2 diquat dibromide MPs;
41.1% and 37.3% a.i. formulation intermediates.  Both of these MPs are
currently registered to Syngenta Crop Protection, Inc.  The products
were transferred from Zeneca to Syngenta and the product chemistry
information for the Zeneca MPs was a repack of a registered product
which is now canceled.  From the confidential statements of formula 
provided by Syngenta, it appears that the manufacturing process for the
current MPs has changed.  Therefore, the product chemistry data
submitted for the canceled products do not satisfy the requirements for
the Syngenta products and a complete updated product chemistry data
package on the currently registered technicals must be submitted to
support continued registration.  

Hazard Identification and Dose Response Assessment

The toxicology data base is adequate to characterize the toxicity of
diquat dibromide, Diquat dibromide exhibits low acute toxicity via the
oral (Toxicity Category II for technical, III for formulation) and
inhalation (Toxicity Category III) routes of exposure but is
moderately-to-severely toxic via the dermal route of exposure (Toxicity
Categories I for technical and II for formulation).  Diquat dibromide is
not an acute skin irritant (Toxicity Category IV) nor a dermal
sensitizer, but it is considered a moderate-to-severe eye irritant
(Toxicity Category II). 

Subchonic and chronic studies in several species indicate multiple
target sites for diquat dibromide toxicity.  In subchronic dermal
exposure studies in rats, diquat dibromide showed evidence of severe
systemic toxicity, i.e. high mortality and clinical signs.  In
subchronic inhalation study in rats, the lung was determined to be the
primary target site for inhalation toxicity.  Chronic feeding studies in
dogs, rats, mice, and rabbits indicate that target sites include the
eyes and kidneys in both males and females and the adrenals and
epididymides in males.  Developmental toxicity was observed in rat,
rabbit, and mouse studies, and reproductive toxicity was observed in the
rat in both generations.  Rat and rabbit studies provided evidence of
maternal toxicity.  Acute and subchronic studies in mice and rats
provided no evidence of neurotoxicity.  Available data provide no
evidence of endocrine disruption following exposure to diquat dibromide.
 Carcinogenicity studies in rats and mice provided no evidence of an
increased tumor incidence and diquat dibromide was classified as a
Category E (evidence of non-carcinogenicity to humans) by the HED Cancer
Peer Review Committee based on the 1999 EPA Draft Proposed Guidelines
for Carcinogen Risk Assessment.  The weight of the evidence was
predominantly negative for mutagenicity. The data provided no indication
of increased sensitivity of rats, mice, or rabbits to in utero and/or
early postnatal exposure to diquat dibromide.   HED’s FQPA committee
determined that the FQPA safety factor could be removed (1x) in
assessing the risk posed by this chemical because the toxicological
database is complete for FQPA assessment, there is no indication of
quantitative or qualitative increased susceptibility of rats or rabbits
to in utero and/or postnatal exposure, a developmental neurotoxicity
study is not required, and the dietary (food and drinking water) and
residential exposure assessments will not underestimate potential
exposures for infants and children.  There are no toxicological study
data gaps at this time.  

Risk assessments were conducted for the exposure scenarios listed below.
 Route-specific endpoints were selected for all but the dermal scenario.
 The dermal exposure endpoint is based on oral toxicity studies and
therefore requires application of a dermal absorption factor (4.1%).

– acute dietary(general population):  	NOAEL= 75 mg/kg/day  	RfD=0.75
mg/kg/day

– chronic dietary:  			NOAEL= 0.5   		RfD = 0.005 mg/kg/day

– short-term oral:  			NOAEL= 1 mg/kg/day	MOE=100 

– short-term dermal:  			NOAEL= 1 mg/kg/day  	MOE=100 

Exposure Assessment 

Analysis of dietary, drinking water, and residential exposure pathways
were included in  the risk assessment for the diquat TRED.  Sources of
dietary exposure include potatoes and various crops grown for seed that
is used for feed (e.g. alfalfa, sorghum) to which diquat has been
applied as a dessicant.  Residues in plants and livestock feeds may also
occur through irrigation of crops with water containing diquat residues,
resulting in dietary exposure through consumption of crops and secondary
residues in livestock tissues.  Drinking water exposure may occur due to
run-off from terrestrial use of diquat and use of diquat in lakes,
ponds, streams, etc., to control aquatic weeds.  Residential application
and post-application exposure may occur through the use of diquat on
turf grass and weeds.  Potential residential exposure routes include
inhalation and dermal exposure to adult applicators/handlers,
postapplication dermal exposure to adults and children, and incidental
ingestion of residue by toddlers via hand-to-mouth activity and
ingestion of grass or soil from a treated area.

Risk Assessment and Risk Characterization

Risk assessments were conducted for dietary, drinking water, and
residential exposure pathways.  An aggregate assessment of risk from the
combined food, drinking water and residential pathways was also
conducted.  A cumulative risk assessment considering risks from other
pesticides or chemical compounds having a common mechanism of toxicity
has not been conducted for this TRED because HED has not yet determined
if there are any other chemical substances that have a mechanism of
toxicity common with that of diquat dibromide.  

Food Pathway Exposure and Risk 

HED conducted acute and chronic dietary exposure analysis using the
Dietary Exposure Evaluation Model (DEEM().   The acute and chronic
dietary exposure/risk analyses were conducted using a conservative
deterministic (Tier I) methodology.  The Tier I analysis assumes that;
1) residues are present at published tolerances for registered uses and
at recommended tolerances for proposed new uses, and 2) 100% crop
treated (CT) for all commodities with existing and/or recommended
tolerances.  Tier I acute and chronic dietary analyses were conducted
for the general U.S. population and all population subgroups.  

The acute and chronic dietary risks are expressed as a percentage of the
acute or chronic Population Adjusted Dose (aPAD or cPAD).  A dietary
risk of 100% of the PAD is the target level of exposure that should not
be exceeded, (i.e., the estimated risk that is less than 100% of PAD is
not of concern). The Population Adjusted Dose (PAD) is the Acute
reference dose (RfD) or the Chronic RfD modified by the FQPA Safety
Factor.  In the acute dietary assessment for diquat, acute exposure
(mg/kg/day) was compared to the aPAD which is based on the acute RfD and
a FQPA safety factor of 1x..  For the chronic dietary assessment,
chronic exposure was compared to the cPAD which is based on the chronic
RfD and an FQPA factor of 1x.  Based on these analyses, acute and
chronic dietary risk associated with exposure to diquat from existing
and proposed uses are below the Agency’s level of concern for the
general US population and population subgroups.  The 95th percentile
acute exposure estimates were < 100% of the aPAD.  The highest acute
exposures ( 0.0054 mg/kg/day) were in Children 1-6 years old (<1% aPAD).
  The chronic exposure estimates were <100% of the cPAD, with the
highest chronic exposure (0.0031 mg/kg/day) occurring in children 1-6
years old (62% cPAD).

Residential Pathway Exposure and Risk 

Potential residential scenarios include exposures to handlers (mixers,
loaders, applicators, etc.) that occur during handling and application
of diquat and/or to persons entering treated sites after its
application.  This TRED estimates exposure and risk for four residential
handler scenarios four post-application scenarios, and two recreational
scenarios.  Residential handler scenarios include: 1) mixing, loading,
and applying with a low pressure handwand (for lawns and backyard
ponds), 2) mixing, loading, applying with a backpack sprayer(for lawns
and backyard ponds), 3) applying with an aerosol sprayer (for lawns),
and 4) applying with a trigger pump sprayer.  Postapplication scenarios
include: 1) dermal exposure to treated turf grass (adults and children);
2) toddler ingestion of treated turf grass via object-to-mouth
activities; 3) toddler ingestion of residue via hand-to-mouth activity
while on treated turf grass; and 4) toddler ingestion of soil from
treated areas.  Recreational exposures include: 1) golfer exposure from
playing on treated turf grass; and 2) exposure from swimming in treated
lakes and ponds.  A target Margin of Exposure (MOE) of 100 is considered
adequate for all residential exposure routes.  With one exception, the
MOEs estimated for all of the residential exposure scenarios described
above showed no risks of concern (i.e. all MOEs were 100 or greater). 
An assessment of aggregate residential risks to children 1-6 years from
post-application exposure to broadcast treated lawns resulted in an
estimated MOE of 70, which exceeds the Agency’s level of concern.  The
residential aggregate risk combines screening level risk estimates from
individual exposure pathways and should be viewed as a highly
conservative estimate which is certain to over-estimate risk.  A refined
analysis would result in lower exposure estimates and higher MOEs.

Drinking Water Pathway Exposure and Risk 

The Environmental Fate and Effects Division (EFED) performed a Tier I
drinking water assessment for diquat dibromide for both terrestrial and
aquatic uses.  EFED used both computer models and monitoring data to
estimate environmental concentrations of diquat in surface and ground
water.  For terrestrial uses, EFED assessed preharvest application to
potatoes, alfalfa, and clover with aerial and ground equipment and
application to clover trees, vines, small fruits, and vegetables using
directed spray from ground equipment.  The FQPA Index Reservoir
Screening Tool (FIRST) model was used to estimate environmental
concentrations in drinking water from surface water contaminated by
terrestrial use of diquat.  The surface water estimated environmental
concentrations (EECs) generated by FIRST ranged from 6.3- 13.2 ug/L
(ppb) for peak exposure and 0.2-0.4 ug/L for average annual exposure.   

EFED relied on monitoring data to estimate concentrations in
groundwater contaminated by terrestrial uses.  Based on municipal
monitoring data, EFED recommends use of EPA’s Office of Waters (OW)
maximum contaminant level (MCL) of 20 ppb for both peak and average
annual  EECs for groundwater concentrations due to terrestrial use of
diquat.  EFED also used monitoring data to estimate surface water and
ground water concentrations from aquatic uses of diquat.  For surface
water, EFED recommends use of the MCL of 20 ppb for both acute (peak)
and chronic (annual average) EECs.  For the aquatic use EFED relied on
use information from the U.S. Corp of Engineers and the states of
Florida, Minnesota, and Michigan.  The assessed aquatic uses include
application to lakes and flowing streams.  EFED also recommends use of
the MCL of 20 ppb for both acute and chronic EECs for groundwater.  
EFED recommends use of the same EECs for both surface and groundwater
contaminated by aquatic uses of diquat because diquat is used near wells
located next to lakes and ponds resulting in interaction between surface
and ground water. 

Aggregate Exposures and Risks 

The aggregate risk assessment combines the exposure assessments
conducted for dietary, drinking water, and residential exposure.  Since
there is potential for concurrent exposure via the food, water, and
residential pathways, the combined or aggregate exposures are estimated
and expressed in terms of an aggregate MOE.  The aggregate MOE is
compared with the target MOE to determine whether there is an aggregate
exposure of concern.  All routes of diquat dibromide exposure were
considered in the aggregate assessment for this TRED.  Aggregate
exposure pathways for adults include dietary, drinking water, and dermal
exposures from application and post-application activities.  Aggregate
exposure pathways for children include dietary, drinking water, oral and
dermal exposures from post-application exposure. A target MOE of 100 is
considered adequate for aggregate exposure/risk.  Acute and chronic
aggregate MOEs were not a risk concern, nor was the short term aggregate
MOE for adults.  The short-term screening level aggregate MOE for the
toddler (child 1-6) exposure scenario was 55.  The short-term aggregate
risk combines screening level risk estimates from individual exposure
pathways and should be viewed as a highly conservative estimate which is
certain to over-estimate risk.  A refined analysis would result in lower
exposure estimates and higher MOEs. 

Data Gaps

All product chemistry data are required for the Syngenta 41.1% and 37.3%
Formulation Intermediates (EPA Reg. Nos. 10-1062 and 100-1063).  
Magnitude of the residue in plants studies are required for sorghum
aspirated grain fractions and  soybean aspirated grain fractions. 

2.0	PHYSICAL/CHEMICAL PROPERTIES CHARACTERIZATION

Common Name:	Diquat Dibromide 

Chemical Name:	6,7-dihydrodipyrido(1,2-a;2',1'-c)pyrazinediium dibromide

Trade Names:		Reglone, Weedkiller D, Aquacide, Dextrone, FB 2, Reglox,
Reward 

Empirical Formula:	C12H12Br2N2

CAS No.:		85-00-7

PC Code:		032201

Structure:              

Molecular Weight:	344 [Cation - 184]

Physical State:		Crystals		

Color:			Colorless to Yellow

Odor:			Odorless				

Solubility in Water:	700 g/L at 20(C

Vapor Pressure:	< 0.03 mPa

Melting Point:		300(C

Density:		1.22-1.27 at  20(C

 Pure diquat dibromide is an odorless, colorless to yellow crystal.  It
is very soluble in water, slightly soluble in alcohol and hydroxylic
solvents, and insoluble in non-polar organic solvents.  It is
susceptible to ultraviolet decomposition.  Ethylene dibromide (EDB) may
be present as a process impurity in final formulations of diquat
dibromide.  However, the registrant has certified an upper limit of 10
ppm EDB in diquat dibromide and has demonstrated that EDB does not
persist as an impurity in diquat dibromide and will slowly dissipate
over time. 

3.0	HAZARD CHARACTERIZATION

3.1	Hazard Profile

The toxicity data base for diquat dibromide is adequate for the
selection of doses and endpoints for use in risk assessment.  HED’s
Hazard Identification Assessment Review Committee (HIARC) evaluated the
acceptable studies available in the database and established acute and
chronic reference doses (RfD), as well as doses and endpoints for short
and intermediate-term incidental oral exposure,  and short-term,
intermediate-term, and long-term dermal and inhalation exposure
scenarios.  The Acute Reference Dose (Acute RfD) is an estimate (with
uncertainty spanning an order of magnitude or greater) of a single day
oral exposure level for the human population, including the sensitive
subpopulations, that is likely to be without an appreciable risk of
deleterious effects.   The Chronic Reference Dose (Chronic RfD) is an
estimate (with uncertainty spanning an order of magnitude or greater) of
a daily oral exposure level for the human population, including
sensitive subpopulations, that is likely to be without an appreciable
risk of deleterious effects during a lifetime.   The acute RfDs are
calculated by dividing the NOAEL by the Uncertainty Factors (UF). 
Uncertainty Factors are used to account for differences between  humans
(intraspecies variability) and for differences between the test animals
and humans (interspecies extrapolation).  For residential exposures,
uncertainty factors are used to determine adequate margins of exposure
(MOEs).  The MOE is the ratio of the route appropriate NOAEL to
estimated exposure.  The HIARC also evaluated available studies to
determine if there is a special sensitivity for infants and children. 
The toxicological data are summarized below. 

3.1.1 	Acute Toxicity

Diquat dibromide exhibits low acute toxicity via the oral (Toxicity
Category II for technical, III for formulation) and inhalation (Toxicity
Category III) routes of exposure but is moderately-to-severely toxic via
the dermal route of exposure (Toxicity Categories I for technical and II
for formulation).  Diquat dibromide is not an acute skin irritant
(Toxicity Category IV) nor a dermal sensitizer, but it is considered a
moderate-to-severe eye irritant (Toxicity Category II).  Acute toxicity
categories for diquat are shown in Table 1.

Table 1. Acute Toxicity Data on DIQUAT DIBROMIDE



Guideline  No.	

Study Type	

MRID #(S)	

Results	

Toxicity Category



870.1100	

acute oral - rat  

Diquat Water Weed Killer	

00081506

	

rat LD50 = 810 mg/kg (

rat LD50 = 600 mg/kg (	

III





870.1200	

acute dermal - rat

 Diquat Water Weed Killer	

00100614

	

rabbits LD50 = 262 mg/kg (

rabbits LD50 = 315 mg/kg (

rabbits LD50 = 288.5 mg/kg (+(	

II





870.1300	

acute inhalation - rat

Diquat Water Weed Killer	

26385(	

rat LC50 = 0.80 mg/L (

rat LC50 = 1.09 mg/L (

rat LC50 = 0.97 mg/L (+(	

III



870.2400	

primary eye irritation -

Diquat Water Weed Killer	

00081507	

rabbit slight to severe eye irritant	

II



870.2500	

primary skin irritation - 

Diquat Water Weed Killer	

00107903	

slight irritation	

IV



870.2600 	

Dermal Sensitization	

00107903	

not a dermal sensitizer	

N/A

     ( Accession No.; With the exception of the dermal sensitization
study, which was conducted with the technical diquat [Diquat Herbicide
Concentrate], the above studies were conducted with the end-use products
[Diquat Water Weed Killer and Diquat 2 Spray (eye irritation study
only)] and not in terms of the Diquat ion. Because the only difference
between the technical diquat and the end-use products is 2.15% of water,
studies with the end-use products have been accepted to satisfy the
generic data requirements for acute studies. The above LD and LC values
are expressed in terms of the test material and not, as is commonly done
with diquat, in terms of the diquat cation.

3.1.2	Toxicity Profile

Table 2 identifies and summarizes guideline studies conducted for diquat
dibromide.

.

Table 2. Toxicity Profile of  DIQUAT DIBROMIDE



GUIDELINE	

STUDY	

MRID	

RESULTS



OPPTS 870.3100	

90-day oral toxicity - rat	

	

no study available



OPPTS 870.3150	

subchronic nonrodent oral toxicity - 90-day	

	

no study available



OPPTS 870.3200	

repeated dose dermal toxicity -21/28 days - rat

20.64% diquat cation

[doses: 0, 5, 20, 40, and 80 mg/kg/day]

Classification: Acceptable/Guideline	

40308101

	

Dermal Toxicity NOAEL < 5 mg/kg/day

Dermal Toxicity LOAEL = 5 mg/kg/day, based on dermal irritation at
application site [erythema, edema, atonia, and desquamation] and tissue
destruction [necrosis and eschar formation]..

Systemic Toxicity NOAEL = 5 mg/kg/day

Systemic Toxicity LOAEL = 20 mg/kg/day, based on a dose-related
mortality and clinical signs [hypothermia, hypoactivity, dyspnea,
cyanosis, pale extremities, general poor condition, and emaciated
appearance]. Based on the severity of the effects observed [death and
clinical signs (hypothermia, hypoactivity, dyspnea, cyanosis, pale
extremities, general poor condition, and emaciated appearance)] and the
extent of  skin toxicity [dermal irritation (erythema, edema, and
desquamation) and tissue destruction (necrosis and eschar formation);
sores, severe erythema, fissures, acute necrotizing purulent dermatitis,
and degeneration of the hair follicles and sebaceous glands] observed at
the site of application in one or both sexes, it was concluded that the
study was not appropriate for use in risk assessment since the NOAEL is
artificially low due to the fact that the skin is compromised



OPPTS 870.3200	

repeated dose dermal toxicity -20 days - rabbit

 [doses: 20, 40, 80, 160 mg diquat cation/kg/day]

Classification: Acceptable/Non-Guideline	

00140576	

Dermal Toxicity NOAEL = <20 mg/kg/day

Dermal Toxicity LOAEL = 20 mg/kg/day, based on erythema and scabbing.

Systemic Toxicity NOAEL = 20 mg/kg/day

Systemic LOAEL = 40 mg/kg/day, based on weight loss, unsteadiness,
muscular weakness and inability to stand, and pathological changes in
the distal convoluted renal tubules with cell necrosis [thought to be
associated with an electrolyte imbalance].



OPPTS 870.3250	

subchronic dermal toxicity - 90 days	

	

no study available



OPPTS 870.3465	

subchronic inhalation toxicity [21-day]

21.6% and 23.5% diquat cation

[doses: 0, 0.49, 1.1, 3.8 µg/L; 0, 0.1 µg/L]

Classification: Acceptable/Guideline, when both studies are considered
together. NOTE: Duration of study is 21 days, not 90 days.	

40301701 40640801	

NOAEL: 0.1 µg/L/day [males 0.024 mg/kg/day; females 0.026 mg/kg/day],
based on increased lung weights and microscopic lesions [mottling and
reddening of the lungs in females, multifocal, chronic, interstitial
pneumonia and alveolar macrophages in both sexes] in the lungs at the
LOAEL of 0.49 µg/L [males 0.117 mg/kg/day; females 0.128 mg/kg/day].



OPPTS 870.3500	

preliminary developmental toxicity screen	

	

no study 



OPPTS 870.3600	

inhalation developmental toxicity study	

	

no study



OPPTS 870.3700	

prenatal developmental toxicity study - rat

26.2% diquat cation

[doses: 0, 4, 12, 40 mg/kg/day]

Classification: Acceptable/Guideline	

41198902	

Maternal NOAEL: 4 mg/kg/day

Maternal LOAEL: 12 mg/kg/day, based on decreased body-weight gains and
food consumption during dosing. At HDT, there was one death [GD 15], and
clinical signs [piloerection and subdued activity] were observed [GD
13-22].

Developmental NOAEL: 12 mg/kg/day 

Developmental LOAEL: 40 mg/kg/day, based on decreased fetal, litter, and
gravid uterine weights, an increased incidence of fetuses with
hemorrhagic kidney, and delayed skeletal ossification [increased
incidence of minor skeletal variations, including unossified ventral
tubercle, unossified cervical vertebral centra, and delayed ossification
of the 2nd and 5th sternebrae].



OPPTS 870.3700	

prenatal developmental toxicity study - rabbit

26.2% diquat cation

[doses: 0, 1, 3, 10 mg/kg/day]

Classification: Acceptable/Guideline	

41198901	

Maternal NOAEL: 1 mg/kg/day

Maternal LOAEL: 3  mg/kg/day, based on body-weight loss [GD 7-10] and
decreased food consumption [GD 7-10]. At HDT, there were deaths and
clinical signs [diarrhea, subdued activity, thin appearance, mucus,
blood, little or no feces in tray]

Developmental NOAEL: 3 mg/kg/day 

Developmental LOAEL: 10 mg/kg/day, based on decreased fetal body weight,
an increased incidence of friable/mottled livers, and an increased
incidence of minor skeletal alterations [partially ossified ventral
tubercle of cervical vertebrae, partially ossified 6th sternebrae, and
unossified 6th sternebrae and presacral vertebrae].



OPPTS 870.3700	

prenatal developmental toxicity study - mouse

analytical standard

[doses: 0, 1, 2, 4 mg/kg/day]

Classification: Acceptable/Non-Guideline	

00061637	

Maternal NOAEL: 1 mg/kg/day [as diquat cation]

Maternal LOAEL: 2 mg/kg/day, based on mortality, clinical signs
[piloerection, respiratory sounds], and decreased body-weight gain
[during dosing period]. At HDT, additional clinical signs [abnormal
posture (hunched or tail raised), lethargy, tremors, unsteadiness on
feet, emaciation, ptosis] and a slight decrease in body weight [91% of
control] at termination were observed also. NOTE: Poor dosing technique;
incidental exposure into lungs.

Developmental NOAEL: 2 mg/kg/day 

Developmental LOAEL: 4 mg/kg/day, based on decreased fetal body weight
and an increased incidence of overall skeletal alterations.



OPPTS 870.3800	

reproduction and fertility effects

20.09% diquat cation

[doses: 0, 16, 80, 400/240 ppm (0, 0.8, 4, 20/12 mg/kg/day)]

Classification: Acceptable/Guideline	

 41531301	

Parental NOAEL: 16 ppm [0.8 mg/kg/day]

Parental LOAEL: 80 ppm [4 mg/kg/day], based on clinical signs,
ulceration of the tongue, and partial/total cataract. At HDT, increased
incidence of clinical signs including ophthalmoscopic signs and lack of
grooming, and gross and microscopic findings [ulceration of the tongue
and partial/total cataract].

Reproductive/Developmental NOAEL: 80 ppm [4 mg/kg/day]

Reproductive/Developmental LOAEL: 400/240 ppm [12 mg/kg/day], based on
decreased number of live pups per litter on days 1-22, decreased pup
body-weight gain during lactation, and increased incidence of kidney
lesions.



OPPTS 870.4100	

chronic toxicity [feeding] study - beagle dog

26.7% diquat cation

[doses: 0, 0.5, 2.5, 12.5 mg/kg/day for 52 weeks]

Classification: Acceptable/Guideline	

41730301	

NOAEL: 0.5 mg/kg/day 

LOAEL: 2.5 mg/kg/day, based on unilateral cataracts in females, and
decreased weights of epididymides and adrenals in males. At the HDT,
bilateral lenticular opacity [cataracts], macroscopically and
microscopically [in all dogs, both sexes]; inflammatory lesions in the
large intestine [in all dogs, both sexes]; increased kidney weight [both
sexes].



OPPTS 870.4200	

carcinogenicity -CD-1 mouse

21.09% diquat cation

[doses: 0, 30, 100, 300 ppm (males 0, 3.56, 11.96, 37.83 mg/kg/day;
females 0, 4.78, 16.03, 48.27 mg/kg/day)] for 104 weeks

Classification: Acceptable/Guideline	

42219801 42880701 42905901 42919501	

NOAEL = 30 ppm [356/4.78 mg/kg/day]

LOAEL: 100 [11.96/16.03 mg/kg/day], based on clinical signs [eye
discharge (males), subdued behavior (females)], and decreased body
weight/body-weight gain [males]. Diquat dibromide was not carcinogenic
in male or female CD-1 mice.



OPPTS 870.4300	

combined chronic toxicity/carcinogenicity - rat

26.5% diquat cation

[doses: 0, 5, 15, 75, and 375 ppm (0, 0.19, 0.58, 2.91, and 14.88
mg/kg/day for males and 0, 0.24, 0.72, 3.64, and 19.44 mg/kg/day for
females)] for 104 weeks.  

Classification: Acceptable/Guideline	

00145855 00155474 41085601	

NOAEL: 15 ppm [0.58/0.72 mg/kg/day]

LOAEL: 75 ppm [2.91/3.64 mg/kg/day], based on eye lesions [total
cataracts]. NOTE: The incidence of eye lesions increased with time on
study, and effect observed at lower dose level with time. Opacity first
occurred in week 10 at 75 ppm and week 11 at 375 ppm. There was no
treatment-related increase in tumor incidence in either sex. 



OPPTS 870.5100-5915	

genetic toxicity tests 

Classification: Acceptable/Guideline	

	

negative for mutagenicity in a bacterial gene mutation [Ames] assay.;
however, positive for gene mutations in a mammalian cell line [mouse
lymphoma] but at cytotoxic doses. Also, clastogenic in cultured human
lymphocytes, but the response was generally weak and observed at
cytotoxic levels. In contrast, there was no evidence of clastogenicity
in somatic cells [mouse bone marrow] or germinal cells [mouse
spermatogonia] or unscheduled DNA synthesis [UDS] in rat hepatocytes in
a series of in vivo studies. Overall, the data suggest that there is no
concern for mutagenicity.



OPPTS 870.6100	

delayed neurotoxicity of OPs following acute/28-day exposures	

-	

no study



OPPTS 870.6200	

neurotoxicity screening battery - acute [rat]

20.1% diquat cation

[doses: 0, 25, 75, and 150 mg/kg]

Classification: Acceptable/Guideline	

42666801	

NOAEL = 75 mg/kg

LOAEL = 150 mg/kg/day, based on clinical signs [piloerection, diarrhea,
staining around nose, urinary incontinence, upward curvature of the
spine, tip toe gait, hunched posture, subdued behavior, and pinched
sides] and decreased body-weight gains.

NOTE: Agency concluded that signs may not be due to direct
neurotoxicity.



OPPTS 870.6200	

subchronic neurotoxicity - Alpk:APfSD rats

20.8% diquat cation

doses: 0, 20, 100, and 400 ppm [males 0, 1.6, 8.0, 32.4
mg/kg/day/females 0, 1.9, 9.5, and 38.5 mg/kg/day]

Classification: Acceptable/Guideline	

42616101	

Systemic Toxicity NOAEL = 100 ppm [8.0/9.5 mg/kg/day

Systemic Toxicity LOAEL = 400 ppm [32.4/38.5 mg/kg/day, based on
evidence of cataracts and decreased body-weight gain and food
utilization in both sexes. There was no evidence of neurotoxicity at any
dose level.



OPPTS 870.6300	

developmental neurotoxicity study	

-	

no study



OPPTS 870.7485	

metabolism and pharmacokinetics [disposition and metabolism]

Classification: Acceptable/Guideline when considered with other two
studies	

00065592	

Diquat dibromide did not accumulate in tissues and was slowly absorbed
following  either oral or iv exposure to single doses of  [14C] Diquat
dibromide to rats or mice of both sexes. Following a single dose of 60
mg/kg, only 5.5% of the radiolabel was excreted in urine within 7 days.
Following an oral [feeding] dose of unlabeled Diquat dibromide [250 ppm]
to rats of both sexes for 2, 4, or 8 weeks, no retention of Diquat was
observed in the brain, liver, lung, stomach, small and large intestine,
muscle, and blood, and little retention in the kidneys [0.18, 0.25, and
1.17 ppm during weeks 2, 4, and 8, respectively]. Ten minutes after iv
injection, there were indications that Diquat concentrated in
cartilaginous tissues, liver, and urinary bladder, as well as the brain
and spinal cord. After 24 hours, radiolabel was detected only in the
urinary bladder and intestines.  



	

Metabolism - excretion

Classification: Acceptable/Guideline when considered with other two
studies	

00065593	

Rats excreted 6.3% and 89.3% of [14C] in the urine and feces,
respectively, within 4 days following oral exposure, with most being
excreted within the first 48 hours In urine, most [5.3%] of the [14C]
was unchanged Diquat, whereas the remaining 1% was associated with:
diquat monopyridone [0.2%], diquat dipyridone [0.1%], and unidentified
metabolites [0.3%]. In feces, 65.5% of excreted [14C] was detected in 
sulfuric acid-extractable fraction and 15.7% in the ammonium
sulfate-unextractable fraction. In the sulfuric acid fraction, [14C] was
distributed as follows: unchanged diquat [57.1%], diquat monopyridone
[4.3%], and unidentified material [4.1%]. 



	

Metabolism

Classification: Acceptable/Guideline when considered with other two
studies	

00055107	

Following oral dosing, (90% of  [14C] was eliminated in the feces,
indicating that Diquat was poorly absorbed for the gastrointestinal
tract. Following a subcutaneous injection of  [14C]-Diquat to circumvent
the intestine, nearly all  [14C] was recovered in the urine within 2
days.



OPPTS 870.7600	

dermal penetration

Classification: Acceptable/Guideline	

 41238701	

dermal absorption of Diquat dibromide through intact rat skin is
considered very low.



OPPTS 870.7800	

immunotoxicity	

-	

no study



	

3.1.3 	Hazard Characterization

The toxicity data base for diquat is complete and clearly defines the
toxicity of the compound.  Subchonic and chronic studies in several
species indicate multiple target sites for diquat dibromide toxicity. 
In subchronic dermal exposure studies in rats, diquat dibromide showed
evidence of severe systemic toxicity, i.e. high mortality and clinical
signs.  While a dermal study was available, the dermal endpoint is 
based on a rabbit developmental gavage study because the dermal study
was determined to be inappropriate for endpoint selection. Route
specific endpoints are available for all other exposure pathways.  In a
subchronic inhalation study in rats, the lung was determined to be the
primary target site for inhalation toxicity.  Chronic feeding studies in
dogs, rats, and mice indicate that target sites include the eyes and
kidneys in both males and females and the adrenals and epididymides in
males.  Developmental toxicity was observed in rat, rabbit, and mouse
studies, and reproductive toxicity was observed in the rat in both
generations.  Rat and rabbit studies provided evidence of maternal
toxicity.  The acute and subchonic neurotoxicity studies in rats
provided no evidence of neurotoxicity.  Available data provide no
evidence of endocrine disruption following exposure to diquat dibromide.
 Carcinogenicity studies in rats and mice provided no evidence of
increase tumor incidence and diquat dibromide was classified as a
Category E (evidence of non-carcinogenicity to humans) by the HED
Reference Dose (RfD)/Peer Review Committee based on the 1999 EPA Draft
Proposed Guidelines for Carcinogen Risk Assessment.  The weight of the
evidence was predominantly negative for mutagenicity. The data provided
no indication of increased sensitivity of rats, mice, or rabbits to in
utero and/or early postnatal exposure to diquat dibromide.  The terminal
residue of concern is the parent compound, diquat cation.  There is no
indication that metabolites are present in significant quantities. 

3.2	FQPA Considerations					

HED’s FQPA committee determined that the FQPA safety factor could be
removed (1x) in assessing the risk posed by this chemical because the
toxicological database is complete for FQPA assessment, there is no
indication of quantitative or qualitative increased susceptibility of
rats or rabbits to in utero and/or postnatal exposure, a developmental
neurotoxicity study is not required, and the dietary (food and drinking
water) and residential exposure assessments will not underestimate the
potential exposures for infants and children. The HIARC and FQPA
Committee determined that a developmental neurotoxicity study was not
required for diquat dibromide based on the fact that (1) there is no
indication of abnormalities in the development of the fetal nervous
system in prenatal developmental toxicity studies in rats, mice, and
rabbits at oral dose levels that were maternally toxic, (2) there was no
evidence of neuropathology in either the acute or subchronic
neurotoxicity studies, (3) the clinical and functional observational
battery observations in the acute neurotoxicity study, which could not
be unequivocally correlated to an effect on the nervous system, were not
observed in the subchronic neurotoxicity study, and (4) no neurotoxic
effects were observed in the brain weights or histopathology of the
nervous system in the chronic toxicity studies with diquat in several
species.

3.3	Dose Response Assessment

Doses and toxicological endpoints selected for various exposure
scenarios are summarized in Table 3. 

 

Table 3. Summary of Toxicological Dose and Endpoints for Diquat for Use
in Human Risk Assessment



Exposure Scenario	

Dose	

Endpoint	

Study



Acute Dietary

general population incl infants and children	

NOAEL= 75 mg/kg  

UF = 100	

LOAEL of 150 mg/kg based on clinical signs and decreased body-weight
gain. 	

Acute neurotoxicity

rat

	

FQPA = 1x     Acute RfD = 0.75 mg/kg     Acute PAD = 0.75 mg/kg



Chronic Dietary	

NOAEL = 0.5 mg/kg/day

UF = 100	

LOAEL of 2.5 mg/kg/day based on cataracts in females and decreased
adrenal and epididymides weights in males. 	

chronic toxicity dog

	

FQPA = 1x     Chronic RfD = 0.005 mg/kg/day    Chronic PAD = 0.005
mg/kg/day



Short-Term Oral

(1day - 1 month)	

NOAEL = 1 mg/kg/day 

MOE = 100	

LOAEL of 3 mg/kg/day based on body-weight loss and decreased food
consumption. 	

Developmental toxicity -

rabbit



Short-Term Dermal a    

(1 day - 1 month)	

NOAEL = 1 mg/kg/day 

MOE = 100	

LOAEL of 3 mg/kg/ day based on body-weight loss and decreased food
consumption.	

Developmental toxicity -

rabbit



Short-Term

Inhalation

(1 day - 1 month)	

NOAEL = 0.1 ug/l (0.024 mg/kd/d male, 0.026 mg/kd/d female)

MOE = 100	

LOAEL of 0.49 ug/L (0.117 mg/kg/day male, 0.128female) due to increased
mean lung weight in males, mottling and reddening of lungs in females,
and lung lesions. 	

21-day inhalation toxicity

rat

 a  Since an oral value was selected, route-to-route extrapolation
should be followed.  A dermal absorption factor is required for  this
risk assessments.								

3.3.1  Dietary Exposure Endpoints

3.3.1.1	Acute Reference Dose

 The HIARC selected an acute RfD of 0.75 mg/kg based on an acute
neurotoxicity gavage study in the rat which showed clinical signs of
systemic toxicity (e.g., piloerection, diarrhea, urinary incontinence,
upward curvature of the spine, subdued behavior) and decreased
body-weight gains at the systemic Lowest Observed Adverse Effect Level
(LOAEL).  The study is considered appropriate for selection of an acute
endpoint because effects were seen after a single dose.  The HIARC
recommended a (UF) of 100x (10x for interspecies and 10x for
intraspecies is extrapolation) for calculation of the RfD. 

The acute endpoint selected by the HIARC for the current risk assessment
differs from that used in assessments conducted for the 1995 RED and
1998 assessment of risk based new and revised uses.  The acute dietary
endpoint used in previous assessments was based on a rat developmental
study, which showed transitory decrease in mean body-weight gain of the
maternal rat following four days of exposure to diquat dibromide.  The
HIARC concluded that this endpoint was not a single dose effect and so
was not appropriate for use in an acute risk assessment.  The rabbit and
mouse developmental toxicity studies were also re-evaluated by the HIARC
for the current analysis.  The rabbit study was deemed inappropriate
because the maternal body-weight loss observed during gestation was not
a single dose effect.  The mouse study was rejected for endpoint
selection because of poor dosing technique which caused inappropriate
introduction of test material into the lungs.   

Acute RfD  = 	75 mg/kg (NOAEL = 0.75 mg/kg

        100 (UF)				

3.3.1.2	Chronic Reference Dose

The HIARC selected a chronic RfD of 0.005 mg/kg/day from a chronic oral
dog toxicity study.  The endpoint is based on unilateral cataracts in
males and females and decreased adrenal and epididymides weights in
males at the LOAEL of 2.5 mg/kg/day.  The study is considered
appropriate for selection of an endpoint for a chronic exposure scenario
because cataracts and decreased adrenal and epididymides weights are
found consistently in repeat exposure studies in both the dog and the
rat.  The rat chronic toxicity/carcinogenicity study which also showed
evidence of cataracts, was considered co-critical for this endpoint. 
The HIARC recommended a UF of 100 (10 interspecies; 10 intraspecies). 
The chronic endpoint selected for the current analysis was used in both
of the previous diquat risk assessments. 

Chronic RfD = 0.5 mg/kg/day (NOAEL) = 0.005 mg/kg/day

            100 (UF)

3.3.2	Residential Exposure Endpoints

					

3.3.2.1 Short-Term Oral Exposure (1 day - 1 month)

A short-term oral NOAEL of 1 mg/kg/day was selected from a rabbit
developmental gavage study which showed maternal body-weight loss during
gestation days (GD) 7-10 and decreased food consumption during dosing
(GD 7-10) at the LOAEL of 3 mg/kg/day.  The study is considered
appropriate as a basis for endpoint selection for infants and children,
the population of concern for incidental oral exposure.  A target MOE of
100 was determined to be adequate for this exposure pathway by the
HIARC.  A short-term oral endpoint was not included in previous HIARC or
risk assessment documents for this chemical.  

3.3.2.2 Short-Term Dermal Exposure (1 day - 1 month)

The HIARC selected a short-term dermal NOAEL of 1 mg/kg/day.  The
endpoint is  based on a rabbit developmental gavage study which resulted
in maternal body-weight loss during gestation days (GD) 7-10 and
decreased food consumption during dosing (GD 7-10) at the LOAEL of 3
mg/kg/day.  The rat chronic toxicity/carcinogenicity study was
considered co-critical for this endpoint.  The recommended target MOE
was 100.  For the diquat risk assessment conducted for the 1995 RED, and
endpoint from a 20 day dermal toxicity study in rabbits was selected for
use in assessing risk from dermal exposure.  For the 1998 revised
assessment, the HIARC determined that the 21-day rat dermal toxicity
study was more appropriate for use in selecting a dermal endpoint
because the rabbit study is a non-guideline study.  Only 3 rabbits/group
were used and hematology, clinical chemistry, and urinalysis parameters
were not monitored.  The guideline states that ten animals/sex/dose are
needed for a NOAEL for risk assessment.  The HIARC then re-evaluated the
rat dermal toxicity study for this TRED.  Based on the most recent
review, the HIARC concluded that the rat study was inappropriate for use
in endpoint selection because the animal’s skin was compromised
resulting in increased severity of effects and an artificially low
NOAEL.  

3.3.2.3	Dermal Absorption

Since the dermal exposure endpoints were selected from oral toxicity
studies, a dermal absorption factor is required to convert the oral dose
to an equivalent dermal dose for the risk assessment.  In an
acceptable/guideline in vivo per cutaneous absorption study, three
aqueous doses of [14C] diquat dibromide were applied dermally to 12
Sprague-Dawley rats (four rats per dose).  The rats were dermally dosed
at levels of 0.05, 0.5, and 5 mg diquat cation/rat for exposure periods
of 2, 10 and 24 hours.  The respective 2, 10, and 24 hour absorption
rates were 4.1, 5.5, and 7.4% at the low dose (0.05 mg/rat), absorption
was 2.8, 5.3, and 4.7% at the middle dose (0.5 mg/rat), and 2.6, 2.5,
and 3.3% at the high dose (5 mg/rat).  (Total percent absorbed = percent
absorbed  + percent remaining in/on the skin.) 

The 4.1% (dermal absorption factor corresponding to the 2 hour, low dose
(0.05 mg/rat) exposure was used for the non-occupational handler and
postapplication exposure scenarios where children and adults are assumed
to have a 2 hour exposure duration.  This is a conservative assumption
in which total percent absorbed = percent absorbed (1.2%) + percent
remaining in/on the skin (3.9%).  

3.3.2.4	Short Term Inhalation Exposure (1 day - 1 month)

The selected NOAEL for inhalation exposure is 0.1 µg/L (0.023 mg/kd/day
male, 0.027 mg/kg/day female) based on a repeated dose 21-day inhalation
toxicity study in rats which resulted in on increased lung weights and
microscopic lesions in the lungs at the LOAEL of 0.49 µg/L (0.113
mg/kg/day male, 0.134 female).  The route and duration of exposure are
appropriate for selecting a short term inhalation endpoint. 

3.3.2.5	Intermediate and Long-Term Endpoints - The HIARC also selected
endpoints for intermediate-term oral, and intermediate, and long-term
dermal and inhalation exposure scenarios.  However, since only
short-term exposures are anticipated for residential uses, endpoints for
longer term exposures were not used in the exposure and risk assessments
for this TRED.

3.3.2.6	Common Toxicological Endpoints for Exposure Routes

A common toxicological endpoint (decrease in maternal body weight and
food consumption) was selected for assessment of short-term incidental
oral and dermal (oral equivalent) exposures.  Therefore, these exposure
routes can be aggregated.  The toxicological endpoint (increased lung
weight and microscopic lung lesions) observed via the inhalation route
is unique.  Therefore, this exposure route cannot be aggregated with the
oral and dermal exposures.

3.4  Endocrine Disruptor Effects

Available toxicity data suggest that there is no evidence of endocrine
disruption following exposure to diquat dibromide. EPA is required under
the FFDCA, as amended by FQPA, to develop a screening program to
determine whether certain substances (including all pesticide active and
other ingredients) "may have an effect in humans that is similar to an
effect produced by a naturally occurring estrogen, or other such
endocrine effects as the Administrator may designate."  Following the
recommendations of its Endocrine Disruptor Screening and Testing
Advisory Committee (EDSTAC), EPA determined that there was scientific
bases for including, as part of the program, the androgen and thyroid
hormone systems, in addition to the estrogen hormone system.  EPA also
adopted EDSTAC’s recommendation that the Program include evaluations
of potential effects in wildlife.  For pesticide chemicals, EPA will use
FIFRA and, to the extent that effects in wildlife may help determine
whether a substance may have an effect in humans, FFDCA authority to
require the wildlife evaluations.  As the science develops and resources
allow, screening of additional hormone systems may be added to the
Endocrine Disruptor Screening Program (EDSP).

When the appropriate screening and/or testing protocols being considered
under the Agency’s EDSP have been developed, diquat dibromide may be
subjected to additional screening and/or testing to better characterize
effects related to endocrine disruption.

4.0	EXPOSURE ASSESSMENT

4.1	Summary of Registered Uses

Currently registered uses are summarized in Table 4.  Table 5 summarizes
the volume of diquat dibromide usage by crop.

Table 4.  Diquat Dibromide Use Groups and Sites



Use Group	

Use Sites



Aquatic Food Crop	

agricultural drainage & irrigation systems; lakes/ponds/reservoirs with
human or wildlife use 



Aquatic Non-food Industrial	

drainage systems, lakes/ponds/reservoirs without human or wildlife use



Aquatic Non-food Outdoor	

aquatic and intermittently flooded areas, streams/rivers/channeled water




Aquatic Non-food Residential	

ornamental ponds/aquaria



Greenhouse Food Crop	

in-use greenhouse



Indoor Food	

storage areas - empty and full



Indoor Non-food	

empty greenhouse



Outdoor Residential	

residential lawns, household/domestic dwellings outdoor premises



Terrestrial Feed Crop	

alfalfa, Bermuda grass, clover



Terrestrial Non-Food Crop	

carrot, cucumber, melons, pepper, radish, squash, turnip, tomato



Terrestrial Food & Feed Crop	

potato, sorghum, soybeans



Terrestrial Non-food Crop	

agricultural fallow/idleland, structures/equipment, rights-of way/fence
rows/hedge rows, and uncultivated areas; airports/landing fields;
commercial/industrial lawns, premises and equipment (outdoor); golf
course turf; nonagricultural outdoor structures/rights-of-way/fence
rows/hedge rows



Table 5. Volume of Diquat Dibromide Usage by Crop



Crop	

% Treated	

Pounds A.I. (1000)	

Application Rate (lbs A.I)	

Major States



	

	

	

	





Alfalfa/Seed	

50	

25	

0.4	

CA, OR, WA



 ADVANCE \d6 Potatoes	

 ADVANCE \d6 30	

 ADVANCE \d6 170   	

 ADVANCE \d6 0.3	

 ADVANCE \d6  ME, ND, NY



 ADVANCE \d6 Soybeans	

 ADVANCE \d6 <1	

 ADVANCE \d6 1   	

 ADVANCE \d6 0.4	

 ADVANCE \d6 -



 ADVANCE \d6 Sorghum	

 ADVANCE \d6 <1	

 ADVANCE \d6 2   	

 ADVANCE \d6 0.4	

 ADVANCE \d6  -



 ADVANCE \d6 Dry Beans	

 ADVANCE \d6 <1	

 ADVANCE \d6 1   	

 ADVANCE \d6 0.3	

 ADVANCE \d6  - 



 ADVANCE \d6 Grapes	

 ADVANCE \d6 <1	

 ADVANCE \d6 1   	

 ADVANCE \d6 0.4	

 ADVANCE \d6 -



 ADVANCE \d6 Home & Garden	

 ADVANCE \d6 <1	

 ADVANCE \d6 100   	

 ADVANCE \d6 0.3	

 ADVANCE \d6 -



 ADVANCE \d6 Aquatic Uses	

 ADVANCE \d6 <1	

 ADVANCE \d6 200   	

 ADVANCE \d6 1.5	

 ADVANCE \d6  CA, FL



 ADVANCE \d6 Total	

	

 ADVANCE \d6 500   	

	





4.2	Dietary Exposure/Risk Pathway

4.2.1	 Residue Profile

4.2.1.1  Nature of the Residue - Plants and Livestock

Plants

The reregistration requirements for plant metabolism are fulfilled (T.
Morton, 11/20/01, D277710).  The qualitative nature of the residue in
plants is adequately understood based on an acceptable potato metabolism
study and a rat bioavailability study.  The HED Metabolism Assessment
Review Committee (MARC) has concluded that the terminal residue of
concern in plants is the diquat cation (T. Morton, 11/8/01, D277765). 
The established tolerance expression for residues of diquat dibromide
in/on plant commodities is appropriate and no changes are required.

The potato metabolism study indicated that no metabolism of diquat
occurred in potato tubers following preharvest application of
[14C]diquat as a desiccant to potato stalks and stems.  Previously
submitted soybean and wheat metabolism studies were deemed marginal
because of inadequate characterization and identification of
14C-residues in the commodities of concern.  In lieu of additional crop
metabolism studies, the Agency recommended several alternatives for this
 requirement.  The registrant opted to conduct a bioavailability study. 
The results of the bioavailability study showed that diquat plant
residues are largely not bioavailable; <5% of the 14C is absorbed as a
result of feeding diquat field residues in/on wheat chaff to rats.  The
retention of diquat residues in tissues was negligible ((0.004 ppm
diquat equivalents) following dosing at  (25x the maximum human dietary
intake.

Livestock

The reregistration requirements for animal metabolism are fulfilled. 
The qualitative nature of the residue in livestock is adequately
understood based on acceptable poultry, ruminant, and fish metabolism
studies. The HED Metabolism Assessment Review Committee (MARC) has
concluded the residue of concern for livestock is the diquat cation (T.
Morton, 11/8/01, D277765).  The established tolerance expression for
residues of diquat dibromide in animal commodities is appropriate and no
changes are required.

In the poultry metabolism study, laying hens were dosed with
ring-labeled [14C]diquat at 32 ppm in the diet for 4 days.  The total
radioactive residues (TRR; expressed as diquat equivalents) ranged from
<0.001-0.004 ppm in egg yolks, egg whites, fat, and muscle, 0.042-0.058
ppm in kidney, and 0.030-0.045 ppm in liver.  The predominant
metabolites identified were diquat cation (48% of TRR in liver), and
diquat monopyridone (15.1% of TRR in kidney) 

The metabolism of diquat dibromide in ruminants has been extensively
investigated. Ruminant data confirm that the residue of concern in
ruminant milk and tissues is diquat cation. Ethylene bridge-labeled
[14C]diquat dibromide was administered at 5 ppm to three cows.  In
addition, one cow was dosed with ethylene bridge-labeled [14C]diquat at
20 ppm and with bypyridyl-labeled [14C]diquat at 5 ppm.  The highest TRR
value in milk was 0.077 ppm in the 72-hour milk sample from the cow
dosed at 20 ppm.  In milk samples from the high-dose cow, residues of
diquat per se were quantitated at <0.002 ppm and did not concentrate in
the fat, casein or whey.  No residues (<0.01 ppm) were found in the leg
muscle samples from one of the cows dosed at 5 ppm.  A bull calf was
administered ethylene bridge-labeled [14C]diquat dibromide at 8 ppm and
sacrificed 24 hours after dosing.  The TRR were 1.071 ppm in kidney,
0.033 ppm in liver and <0.04 ppm in other tissues.  Residues of diquat
cation were 0.03 ppm and <0.01 ppm in the kidney and liver samples,
respectively.

In the fish metabolism study, trout and carp were exposed to an initial
concentration of 1 ppm of bridge-labeled [14C]diquat in the water for 7
days.  The TRR (expressed as diquat equivalents) in carp head and tail,
viscera, and body with skin were 0.025-0.077 ppm, 0.135-0.946 ppm, and
0.013-0.024 ppm, respectively.  The TRR in skin and flesh without skin
were 0.015-0.023 ppm and 0.006-0.016 ppm, respectively.  The TRR in
trout head, tail, and flesh were 0.025-0.051 ppm, 0.059-0.239 ppm, and
0.008-0.01 ppm, respectively.  Approximately 65% of the radioactivity in
carp flesh and trout viscera was identified as diquat. 

4.2.1.2	Residue Analytical Method - Plants and Livestock

Enforcement methods

The Pesticide Analytical Manual (PAM) Vol. II. lists a
spectrophotometric method, designated as Method A as available for the
enforcement of tolerances for residues of diquat in/on plant and in
livestock commodities.  The limit of detection is 0.01 ppm.  The
registrant has proposed new enforcement methods, RM-5B-1 and RM-5C, for
plant and livestock commodities, respectively.  The stated limit of
detection is 0.005 ppm for RM-5B-1; the limit of detection for RM-5C is
not clearly specified.  Both methods have been adequately validated by
the registrant; however, an independent laboratory validation must be
conducted followed by validation by the Agency's Analytical Chemistry
Branch before they can be considered fully adequate for enforcement
purposes.  Once a successful EPA method validation has been performed,
these methods will be sent to FDA for inclusion in PAM Vol. II.

Data collection

Residue data submitted for tolerance reassessment were collected using
the current (PAM Vol. 2 Method A) or proposed (RM-5B and RM-5C)
enforcement methods.  The registrant provided adequate method validation
data to verify the suitability of these methods for data collection.

4.2.1.3	Multiresidue Methods

The FDA's PESTDATA dated 11/6/90 (PAM Vol. I, Appendix) indicates that
recovery of diquat dibromide using Multiresidue Protocols is unlikely. 
The updated PESTDATA dated 08/93 does not have an entry for diquat
dibromide.

4.2.1.4	Storage Stability Data

The requirements for storage stability data are fulfilled for purposes
of tolerance reassessment.  Adequate storage stability data on diquat
dibromide are available to support the storage conditions and intervals
of samples from magnitude of the residue studies in plants and
livestock.  Residues of diquat are stable under frozen (-20(C) storage
conditions for: up to six months in/on bell pepper, carrot roots, clover
(hay and seed), lettuce, potato, rice (grain and straw), sorghum grain,
soybean, tomato and tomato processed fractions, and wheat (grain and
straw); up to 8 months in processed fractions of sorghum grain and
soybean; and up to 2 months in water and seafood samples.

4.2.1.5	Magnitude of the Residue in Crop Plants

All data for magnitude of the residue in plants have been evaluated and
deemed adequate except for the following deficiencies: residue data
required for sorghum aspirated grain fractions, soybean aspirated grain
fractions.  All field residue data have been re-evaluated and plant
commodity tolerances reassessed (where residue data are available) for
TRED purposes.  There are no registered uses of diquat dibromide on
sugarcane and vetch.  Therefore, field residue data for these crops are
no longer required and the established tolerance for residues in/on
sugarcane should be revoked.

4.2.1.6	Magnitude of the Residue in Processed Food/Feed

The data for magnitude of the residue in processed food/feed have been
evaluated and deemed adequate to determine the extent to which residues
of diquat concentrate in food/feed items upon processing of raw
agricultural commodities.  Acceptable potato, soybean, and sorghum
processing studies have been submitted and evaluated.  Acceptable
processing data are also available for tomato wet pomace, tomato juice,
and tomato paste.  However, additional residue data are required for
sorghum and soybean aspirated grain fractions and tomato puree.

4.2.1.7	Magnitude of the Residue in Meat, Milk, Poultry, and Eggs

The reregistration data requirements for magnitude of the residue in
livestock are fulfilled. The HED MARC has concluded that the residue of
concern for livestock is the diquat cation (T. Morton 11/08/01) and
acceptable animal feeding studies have been submitted and evaluated.  

A new maximum dietary burden (MDB) estimate has been calculated based on
reassessed established/proposed tolerances of feed commodities and
revised Table II of OPPTS 860 Guidelines.  (T. Morton, 11/20/01,
D277710).  The established tolerances of 0.02 ppm for diquat residues in
the fat, meat, and meat byproducts of cattle, goats, hogs, horses, and
sheep may be raised to 0.05 ppm to achieve compatibility with the Codex
maximum residue limit (MRL).  The established 0.02-ppm tolerance level
for diquat residues in poultry fat, meat, meat byproducts and eggs may
be raised to 0.05 ppm to achieve compatibility with the Codex MRL.  

4.2.1.8 Magnitude of the Residue in Fish and Shellfish

All data requirements for magnitude of the residue in fish and shellfish
have been evaluated and deemed adequate to reassess the tolerances for
diquat; no additional data are required regarding this topic.  The
available data indicate that residues of diquat in fish and shellfish
will exceed the established tolerances following tests reflecting the
current maximum registered use patterns.  The registrant must submit a
petition requesting tolerance increases from 0.1 ppm to 2.0 ppm for fish
and 20 ppm for shellfish to cover all residues of diquat which may occur
as a result of the currently registered uses.

4.2.1.9  Magnitude of the Residue in Irrigated Crops

The available data concerning diquat residues following irrigation of
carrot, corn (sweet), cowpea, peach, and rice are adequate to support
the established 0.02 ppm tolerances for diquat residues in/on all
members of the crop groups containing these commodities.  However, the
data also indicate that residues in/on cowpea, blackberry, strawberry,
orange, mustard greens, pasture grass, and tomato may exceed the
tolerances for the respective crop groups.  The registrant must propose
a higher tolerance level of 0.05 ppm for citrus fruit, small fruits,
fruiting vegetables, legume vegetables, and Brassica leafy vegetables. 
The registrant is required to propose a higher tolerance level of 0.20
ppm for forage grasses.

No data are available for the commodities avocado, cottonseed, hops, and
sugarcane for which tolerances currently exist.  However, data for other
crops can be translated to these commodities.  Based on the highest
residues found in other crops irrigated with water containing diquat
residues, HED recommends that the registrant propose tolerances of 0.20
ppm for these crops.  If lower tolerances are desired, additional data
will be required.  

4.2.1.10 Confined and Field Accumulation in Rotational Crops

The data requirements for confined rotational crops have been reviewed
and deemed adequate by the Environmental Fate and Effects Division
(EFED).  The requirements for field rotational crop studies have been
waived at this time.

4.2.1.11  Codex/International Harmonization

Several maximum residue limits (MRLs) for diquat have been established
by Codex in various commodities.  Tolerance levels for some commodities
with an MRL have been revised in the TRED to achieve compatibility with
Codex.  The U.S. tolerance for eggs, poultry, meat, and offal
(mammalian) may be raised to 0.05 ppm to achieve harmonization.  Further
harmonization of U.S. tolerances and Codex MRLs on other commodities are
not feasible at this time because of differences in agricultural
practices. 

4.2.2 	Dietary Exposure/Risk Assessment

4.2.2.1	Consumption Data and Dietary Risk Analysis

Diquat acute and chronic dietary exposure assessments were conducted
using the Dietary Exposure Evaluation Model (DEEM() software Version
7.73, which incorporates consumption data from USDA’s Continuing
Surveys of Food Intake for Individuals (CSFII), 1989-1992.  The 1989-92
data are based on the reported consumption of more than 10,000
individuals over three consecutive days, and in total represent more
than 30,000 unique “person days” of data.  Foods “as consumed”
(e.g., apple pie) are linked to raw agricultural commodities and their
food forms (e.g., apples-cooked/canned or wheat-flour) by recipe
translation files internal to the DEEM software.  Consumption data are
averaged for the entire US population and within population subgroups
(e.g., children one to six years old) for chronic exposure assessment,
but are retained as individual consumption items for acute exposure
assessment.

For chronic exposure and risk assessment, estimates of average residues
for foods (e.g., orange) or food-forms (e.g., orange-juice) of interest
are multiplied by the averaged consumption estimate of each
food/food-form of each population subgroup.  Exposure estimates are
expressed in mg/kg body weight/day and as a percent of the cPAD. 

For acute exposure assessments, individual one-day consumption data are
used on an individual-by-individual basis. The reported consumption
amounts of each food item can be multiplied by a residue point estimate
and summed to obtain a total daily pesticide exposure for a
deterministic (Tier 1 or Tier 2) exposure assessment, or “matched”
in multiple random pairings with  residue values and then summed in a
probabilistic (Tier 3/4) assessment.  The resulting distribution of
exposures is expressed as a percentage of the aPAD on both a user (i.e.,
those who reported eating relevant commodities/food forms) and a
per-capita basis.

4.2.2.2 Acute Dietary Exposure/Risk Assessment

An acute dietary exposure analysis was conducted for diquat dibromide
using the DEEM( software (B. Daiss, D277766, 12/11/01).  The acute
dietary exposure/risk analysis was conducted using a conservative
deterministic (Tier I) methodology.  The Tier I analysis assumes that;
1) residues are present at published tolerances for registered uses and
at recommended tolerances for proposed new uses, and 2) 100% crop
treated (CT) for all commodities with existing and/or recommended
tolerances.  Tier I acute dietary analyses were conducted for the
general U.S. population and all population subgroups.  Based on this
analyses, acute dietary risk associated with exposure to diquat from
existing and proposed uses are below the Agency’s level of concern for
the general US population and population subgroups.  The 95th percentile
acute exposure estimates were < 100% of the acute Population Adjusted
Dose (aPAD).   The highest acute exposure ( 0.0054 mg/kg/day) was in
children 1-6 years old (<1% aPAD).  The results are presented in Table
6.

Table 6.  Acute Dietary (Food) Exposure Estimate and Percent of Acute
RfD at the 95th Percentile - Tier 1 Exposure Analysis (Assumes 100% Crop
Treated and Tolerance Level Residues)



	Subgroups	

	95th Percentile Exposure (mg/kd/day)	

	% aPAD at 95th Percentile Exposure



US Population (total)	

0.0039	

< 1



All Infants (<1 year old)	

    0.0035    	

< 1



Children 1-6 years old	

 0.0054	

< 1



Children 7-12 years old	

 0.0033	

< 1



Females 13-50 years old	

 0.0032	

< 1



Males 13-19 years old	

  0.0030	

< 1



Males 20+ years old	

  0.0035	

< 1



Seniors 55+ years old	

  0.0023	

< 1



4.2.2.3	Chronic Dietary Exposure/Risk Assessment

A chronic dietary exposure analysis was conducted for diquat dibromide
using the DEEM( software.  The chronic dietary analysis was conducted
using a conservative deterministic (Tier I) methodology (i.e., residues
present at tolerance levels and 100 %CT).   Tier I chronic dietary
analyses were conducted for the general U.S. population and all
population subgroups.  Based on this analyses, chronic dietary risk
associated with exposure to diquat from existing and proposed uses are
below the Agency’s level of concern for the general US population and
population subgroups.  The chronic exposure estimates were < 100% of the
chronic PAD (cPAD) with the highest chronic exposure (0.0031 mg/kg/day)
occurring in children 1-6 years old (62% cPAD).  Results are presented
in Table 7.



Table 7. Chronic Dietary (Food) Exposure Estimate and Percent of Chronic
RfD - Tier 1 Exposure Analysis  (Assumes 100% Crop Treated and Tolerance
Level Residues)



	Subgroups	

	Mean Exposure (mg/kd/day)	

	% cPAD at Mean Exposure



US Population (total)	

0.0019	

38



All Infants (<1 year old)	

0.0017	

33



Children 1-6 years old	

0.0031	

62



Children 7-12 years old	

0.0021	

42



Females 13-50 years old	

0.0017	

34



Males 13-19 years old	

0.0018	

26



Males 20+ years old	

0.0019	

38



Seniors 55+ years old	

0.0015	

31



4.3. 	Drinking Water Exposure/Risk Pathway

4.3.1. 	Environmental Fate Assessment

Diquat is persistent but essentially immobile in the environment,
indicating that it will most likely be associated with the soil and
sediment instead of water.  The primary route of environmental
dissipation of diquat used in terrestrial settings is strong adsorption
to soil.  When used as an aquatic herbicide, diquat is removed from the
water column by adsorption to sediment, aquatic vegetation, and organic
matter.  Based on acceptable guideline studies, diquat does not
hydrolyze or photodegrade and is resistant to microbial degradation
under aerobic and anaerobic conditions.  There were no major degradates
isolated from any of the environmental fate studies.  Environmental fate
data indicate that diquat is miscible in water (7 x 105 ppm), is stable
to hydrolysis and photolysis, and metabolism, but is essentially
immobile in soil and sediment.  The extent of adsorption appears to be
related to soil pH, with is consistent with cation exchange in soil. 

4.3.2	Estimated Environmental Concentrations/Monitoring Results

The Environmental Fate and Effects Division (EFED) performed a Tier I
drinking water assessment for diquat dibromide for both terrestrial uses
and aquatic uses (D281199, J. Breithaupt, 3/5/01).  EFED used both
computer models and monitoring data to estimate environmental
concentrations of diquat in surface and ground water.  For terrestrial
uses, EFED assessed preharvest application to potatoes, alfalfa, and
clover with aerial and ground equipment and application to clover trees,
vines, small fruits, and vegetables using directed spray from ground
equipment. The assessed aquatic uses include application to lakes and
flowing streams. 

4.3.2.1	Terrestrial Uses

Surface Water	

The FQPA Index Reservoir Screening Tool (FIRST) model was used to
estimate environmental concentrations in drinking water from surface
water contaminated by terrestrial use of diquat.  FIRST is based upon
the linked Pesticide Root Zone Model (PRZM) which simulates pesticides
in field run-off and Exposure Analysis Modeling System (EXAMs) which
simulates pesticide fate and transport in an aquatic environment. 
However, where previous PRZM-EXAMs models used a standard field pond
scenario, FIRST uses an Index Reservoir which is based on Shipman City
Lake in Illinois ( 13 acres in area, 9 feet deep, and  a watershed area
of 427 acres).  In addition, FIRST uses a Percent Cropped Area (PCA)
factor which translates to reduction of area within the reservoir that
is planted to modeled crop. Due to the change from the standard pond to
Index Reservoir, the physical scenario as well as the treatment of spray
drift is different in FIRST.  FIRST is designed to produce more
realistic estimates of pesticides in surface water that is used as a
source of drinking water.  

FIRST was used to model the following diquat scenarios:
trees/vines/small fruits/vegetables (maximum use rate of 1 lb a.i./A,
single application, 87 PCA); potatoes (0.5 lb a.i./A, 2 applications, 14
day application interval, 87 PCA); and alfalfa/clover/non-crop (0.5 lb
a.i./A, 1 application, single application, 87 PCA) .  The surface water
estimated environmental concentrations (EECs) generated by FIRST ranged
from 6.3- 13.2 ppb (ug/L) for peak exposure and 0.2-0.4 ppb for annual
average exposure.  

Groundwater

EFED used the Screening Concentration in Ground Water (SCI-GROW) model
to estimate diquat concentrations in groundwater contaminated by
terrestrial uses.  SCI-GROW is a regression-based model that uses few
input parameters: pesticide’s organic carbon partition coefficient
(Koc), aerobic soil degradation half-life, and product label application
rate and frequency (Barrett, 1997).  It provides a groundwater screening
concentration for use in determining potential risk to human health from
drinking water contaminated with a pesticide.  The groundwater
concentration is estimated based on the maximum application rates in
areas where groundwater is exceptionally vulnerable to contamination. 
These vulnerable areas are characterized by high rainfall, rapidly
permeable soil, and shallow aquifer. The SCI-GROW model estimated
terrestrial use groundwater EECs of 0.006 ppb.  However, EFED recommends
that monitoring data be used in lieu of the SCI-GROW results because
higher concentrations were observed from monitoring data.  

The sources of available monitoring data include the South Florida
Water Management District (SFWMD), and potable water modeling studies
from the registrant, and monitoring data from the EPA Office of Water
(OW).  The SFWMD data contained a total of 42 samples that were taken
from April 1992 to November 2000 on approximately a 1-3 month interval. 
For diquat, the only detection observed in surface water was 0.0045 ppm
in 1994.   Further monitoring beyond 1994 has not shown any detections
in surface water.  Also in the SFWMD, diquat was detected in 9 sediment
samples from canals with a maximum concentration of 3.1 ppm (LOD of 2.5
ppm, Miles and Pfeuffer, Pesticides in Canals of South Florida, Arch.
Environ. Contam. Toxicol. 32:337-345, 1997).    In the potable water
modeling studies, highly variable estimates of water concentrations have
been observed.  The modeling concentrations at reservoir pumps ranged
from non-detections (LOD of 0.003 ppm) to 0.26 ppm.  These modeling
estimates were not used in this assessment because of available
monitoring data obtained by OW.  OW has monitored for diquat at intake
pumps at drinking water utilities that use surface water and ground
water.  Data from eight states in the years 1993-1997 were included in
the report.  In these eight states, 0.06 percent of combined surface
water and ground water systems reported exceedences of the 0.02 ppm
(mg/L) Maximum Contaminate level (MCL) set by OW, resulting in a
population exceedence of 0.27 % (Occurrence of Regulated Contaminants in
Drinking Water: First Stage Occurrence and Exposure Report for Six-Year
Regulatory Review, Working Draft 5/12/2000).  Based on municipal
monitoring data, EFED recommends use of the MCL of 0.02 ppm (20 ppb) for
both peak and average terrestrial use EECs for groundwater.  

4.3.2.2	Aquatic Uses

EFED also relied on monitoring data to estimate surface water and ground
water concentrations from aquatic uses of diquat.  For the aquatic use,
EFED also included use information from the U.S. Corp of Engineers and
the states of Florida, Minnesota, and Michigan.   For aquatic use, EFED
recommends use of the MCL of 0.02 ppm (20 ppb) for both peak and annual
average EECs.  EFED recommends use of the same annual average EECs for
both surface and groundwater contaminated by aquatic uses of diquat
because diquat is used near wells located next to lakes and ponds
resulting in interaction between surface water and ground water.   EFED
notes that, while concentrations in excess of 20 ppb in private wells
cannot be ruled out, they are unlikely because of the tendency of diquat
to sorb nearly irreversably to soil and sediment.  

EECs for surface and groundwater terrestrial and aquatic uses are
provided in the Tables 8 and 9 respectively.

Table 8.  FIRST EEC’s for Diquat for Drinking Water Assessment from
Surface Water



Type of Use	

	Use Site	

Peak (ppb)	

Average Annual (ppb)



Terrestrial	

Trees, vines, small fruits, vegetables	

13.2	

0.4

	

Alfalfa, Clover	

6.3	

0.2

	

Non-Crop (fallow land)	

6.6	

0.2

	

Potatoes	

12.7	

0.4



Aquatic	

Ditches, reservoirs, and rivers	

201	

20

1 MCL from Office of Water



	Table 9. Monitoring EEC’s for Diquat for Drinking Water Assessment
from Ground Water



Type of Use	

Use Site	

 Peak (ppb)	

Average Annual (ppb)



Terrestrial	

Trees, vines, small fruits, vegetables, alfalfa, clover, non-crop
(fallow land), and potatoes	

20	

20



Aquatic	

Ditches, reservoirs, and rivers	

20	

20



4.4	Residential Exposure/Risk Assessment

A regulatory review of residential exposure to diquat dibromide was
conducted for this TRED because there is potential exposure to
non-occupational (residential) handlers (mixers, loaders, applicators,
etc.) during handling and application of diquat and/or to persons
entering treated sites after its application (T. Brennan, 12/14/01,
D279507).  Only short-term exposures are expected/assessed for
residential exposure scenarios.

4.4.1	Home Uses

4.4.1.1	Handler

Handler Exposure Scenarios

Diquat dibromide can be applied to turf and backyard ponds for general
weed control and can be used in and around home & garden sites for weed
control and landscape uses by residential handlers.  Diquat dibromide is
applied to residential turf grass at application rates ranging from 0.25
to 0.5 lb ai/acre.  Therefore, for this assessment both low end and high
end MOEs were assessed based on the range of application rates.  Diquat
dibromide is formulated for residential uses both as a Liquid Ready to
Use, and as a Soluble Concentrate/Liquid that can be mixed with water
and then applied with a low pressure handwand or backpack sprayer.  This
TRED estimates exposure and risk for four residential handler scenarios:
  	

1) Mixing, loading, and applying with a low pressure handwand - This
residential handler scenario estimates exposure and risk to a mixer,
loader, applicator applying diquat dibromide with a low pressure
handwand to control weeds in and around lawns, gardens, around
buildings, driveways, fence lines and other such edge areas.  The dermal
and inhalation unit exposures for this scenario were obtained from the
Outdoor Residential Exposure Task Force (ORETF) chemical handler
exposure studies.

2) Mixing, loading, applying with a backpack sprayer - This scenario
estimates exposure and risk to a mixer, loader, applicator applying
diquat dibromide with a backpack sprayer to control residential weeds in
edge areas.

3) Applying with an aerosol sprayer - This scenario estimates exposure
and risk to an applicator applying diquat with an aerosol spray can to
control residential weeds in edge areas. 

4) Applying with a trigger pump sprayer - This scenario estimates
exposure and risk to an applicator applying diquat dibromide with an
trigger pump sprayer to control residential weeds.  The dermal and
inhalation unit exposures for this scenario were obtained from a
carbaryl Occupational and Residential Exposure Task Force (ORETF) study
on ready-to-use insect sprayer application to home garden vegetables.

The handler assessment was developed using standard residential
application techniques and PHED unit exposure data.  It is the policy of
the HED to use data from the Pesticide Handlers Exposure Database (PHED)
Version 1.1 to assess handler exposures for regulatory actions when
chemical-specific monitoring data are not available.  PHED was designed
by a task force of representatives from the US. EPA, Health Canada, the
California Department of Pesticide Regulation, and members of the
American Crop Protection Association.   PHED is a software system
consisting of two parts; 1) a database of measured exposure values for
workers involved in the handling of pesticides under actual field
conditions, and 2) a set of computer algorithms used to subset and
statistically summarize the selected data.  Currently, the database
contains values for over 1,700 monitored individuals (i.e., replicates).
 For the handler scenarios female adult applicators were assessed for
dermal exposures because the dermal toxicity endpoint was based on
developmental toxicity effects.  A complete summary of the handler
dermal exposure and risk calculations, critical assumptions, and results
is provided in Table 10.  Handler exposure assumptions are summarized
below.

Handler Exposure Assumptions

The following assumptions were made in the exposure calculations:

Average body weight of an adult handler is 70 kg/day.

·	Application rates range from a low-end rate 0.009 to a high-end 0.018
lb ai/gallon of sprayer.

·	Backpack sprayers and low pressure handwands were applied at a rate
of 5 gallons per day.  Aerosol cans and triggers sprayers were applied
at a rate of 0.125 gallons per day.

·	No protective clothing was factored into the assessment for these
residential handler exposure/risk scenarios.  Clothing assumptions
include short pants, short-sleeved shirt, and no gloves.

Exposure frequency - The residential handlers are expected to have a
short-term exposure duration (less than 30 days).

Handler Exposure and Risk Estimates

A target MOE of 100 for the dermal and inhalation routes are considered
adequate for the handler risk assessment. Results of handler exposure
assessment are presented in Table 10 and are summarized below. 

1) Low pressure handwand scenario - The adult applicator dermal MOEs
were 670 for the low end scenario and 330 for the high end scenario. 
The inhalation MOEs were 8,800 for the low end scenario and 4,400 for
the high end scenario.  

2) Backpack sprayer scenario - The adult applicator dermal MOEs were
7,500 for the low end scenario and 3,700 for the high end scenario.  The
inhalation MOEs were 1,400 for the low end scenario and 720 for the high
end scenario.

3) Aerosol can scenario - The adult applicator dermal MOEs were 6,900
for the low end scenario, and 3,500 for the high end scenario.  The
inhalation MOEs were 720 for the low end scenario, and 360 for the high
end scenario.

4) Trigger pump spray scenario -  The adult applicator dermal MOEs were
24,390 for the low end scenario, and 14,000 for the high end scenario. 
The inhalation MOEs were > 24,000,000 for the low end scenario, and >
12,000,000 for the high end scenario.

Table 10: Diquat Dibromide Residential Handler Short Term Exposure and
Risk Assessment



Exposure Scenario 

(Scenario #)	

Dermal Unit Exposure  (mg/lb ai)	

Inhalation Unit Exposure    (ug/lb ai)	

Crop

(rate)	

Application Rate

 (lb ai per gal)	

Amount Treated (gal per day)       	

Dermal Absorbed Dose c (mg/kg/d)	

Dermal MOE d	

Inhalation Dose e (mg/kg/d)	

Inhalation MOE f



Mixer/Loader/App



Mixing/Loading/Applying Liquids for Low Pressure Handwand application
(1)a	

100	

30	

high end 	

0.018	

5	

0.003	

330	

0.0000055	

4400



Mixing/Loading/Applying Liquids for Low Pressure Handwand application
(2)a	

100	

30	

low  end 	

0.009	

5	

0.0015	

670	

0.0000028	

8800



Mixing/Loading/Applying Liquids for Backpack sprayer application (3)b	

5.1	

30	

high end 	

0.018  	

5 	

0.00027	

3700	

0.000038	

720



Mixing/Loading/Applying Liquids for Backpack sprayer application (4)b	

5.1	

30	

low  end 	

0.009 	

5 	

0.00013	

7500	

0.000019	

1400



Applicator



Aerosol can application (5)b	

220	

2400	

high end	

0.018 	

0.125 	

0.00029	

3500	

0.000077	

360



Aerosol can application (6)b	

220	

2400	

low  end	

0.009 	

0.125 	

0.00014	

6900	

0.000039	

720



Trigger Sprayer (7)a	

53	

0.067	

high end	

0.018 	

0.125 	

0.00007	

14000	

2E-09	

12000000



Trigger Sprayer (8)a	

53	

0.067	

low  end	

0.009 	

0.125 	

0.000041	

24390	

1E-09	

24000000

Footnotes:

a	Baseline dermal and inhalation exposure derived from the ORETF
chemical handler exposure studies.

b	Baseline dermal unit exposure represents short pants, short-sleeved
shirt, no gloves, open mixing/loading, open cab/tractor.  Values from
PHED V1.1.  Baseline inhalation unit exposure represents no respirator. 
Values from PHED V1.1.

c	Dermal daily dose (mg/kg/day) = daily unit exposure (mg/lb ai) x
application rate (lb ai/acre) x amount handled per day (acres) / bw (60
kg).

d	Dermal MOE = NOAEL (1 mg/kg) / [daily dose (mg/kg/day) x dermal
absorption factor (4.1%)].

e	inhalation daily dose (mg/kg/day) = inhalation unit exposure (µg/lb
ai) x application rate (lb ai/acre) x amount handled per day (acres) x
conversion factor (1 mg/1,000 µg) / body weight (70 kg). 

f	Inhalation MOE = NOAEL (0.1 ug/L) / daily dose (mg/kg/day).  

4.4.1.2  Postapplication

Diquat dibromide is used on dormant turf grass for weed control.  Diquat
dibromide will kill all types of vegetation and can be used to kill turf
grass. Based on this and a review the labels, HED conducted a lawn
post-application analysis.  Diquat dibromide is specifically labeled to
be applied on dormant Bermuda and zoysia grass.  Therefore re-entry must
be addressed.  Only short-term exposures (1 to 30 day period of
exposure) are assessed because residents can have post-application
exposure to treated lawns in the initial hours and days following
treatment.  Diquat dibromide is applied to residential turf grass at
application rates ranging from 0.25 to 0.5 lb ai/acre.  Therefore, for
this assessment both low end and high end MOEs were assessed based on
the range of application rates.  	

Postapplication Exposure Scenarios

Four post-application scenarios were assessed:

1) Dermal exposure to treated turf grass - adults and children - This
post-application scenario estimates the dermal exposures and risk to
adults and toddlers from dermal contact with turf treated with diquat
dibromide.  This scenario assumes that diquat dibromide residues are
transferred to the skin of adults/toddlers who enter treated yards for
recreation, yard work, or other homeowner activities. 

2) Toddler ingestion of treated turf grass via object-to-mouth
activities - This post-application scenario estimates doses among
toddlers from incidental ingestion of residential turf grass that has
been previously treated with pesticides.  This scenario assumes that
turf is ingested (or just mouthed) by toddlers who play in the treated
areas. 

 

3) Toddler ingestion of residue via hand-to-mouth activity while on
treated turf grass - This post-application scenario estimates potential
ingestion of pesticide residues from previously treated turf.  This
scenario assumes that pesticide residues are transferred to the skin of
toddlers playing on treated yards and are subsequently ingested as a
result of hand-to-mouth transfer.

4) Toddler ingestion of soil from treated area - This post-application
scenario estimates doses among toddlers from incidental ingestion of
soil containing pesticide residues.  This scenario assumes that
pesticide residues in soil are ingested by toddlers who play in treated
areas.  

The turf, post-application assessments were developed using the
Residential SOP guidance.  The exposure and risk calculations, critical
assumptions, and results are provided in Table 11 for postapplication
dermal scenarios and in Table 12 for postapplication incidental oral
exposure.  Post-application exposure assumptions are summarized below.

Postapplication Exposure Assumptions

The following assumptions were made in the dermal postapplication
exposure calculations:

Application rates range from a low-end rate of 0.25 to a high-end rate
of 0.50 lbs ai/acre.

Turf transferable residue is equal to 5 % of the application rate.

Turf transfer coefficient is 14,500 cm2/hr for adult and 5,200 cm2/hr
for children.

Exposure duration is 2 hours for exposure to residential lawns, and 4
hours for exposure to golf course turf. 

Body weight is 70 kg for adults, and 15 kg for toddlers.

The following assumptions were made in the oral postapplication exposure
calculations:	

Application rates range from a low-end rate of 0.25 to a high-end rate
of 0.50 lbs ai/acre.

Hand and object transfer efficiency is equal to 5 % of the application
rate available for transfer from treated turf to wet hands and objects.

Surface portion of  hand put in mouth is 20 cm2.

Hand-to-mouth exposure frequency is 20/times per hour.

Body weight is 15 kg for toddlers.

Exposure time 2 hours.

Saliva extract factor (for hand-to-mouth only) is 50 percent.

Postapplication Exposure and Risk Estimates

A target MOE of 100 for both the dermal and incidental oral routes is
considered adequate for the postapplication risk assessment.  Results of
post application assessment are presented in Tables 11 and 12 and are
summarized below. 

1) Dermal exposure to treated turf grass - adults and children - The
adult MOEs for re-entering treated lawns were not a risk concern with a
low end MOE of 480, and a high end MOE of 220.  The toddler MOEs for
this scenario were also not a risk concern with a low end MOE of 270 and
a high end MOE of 130. 	

2) Toddler ingestion of treated turf grass via object-to-mouth
activities - The toddler MOEs for ingesting (or mouthing) of treated
turf grass were not a risk concern with the low end MOE of 1,110, and a
high end MOE of 560.

3) Toddler ingestion of residue via hand-to-mouth activity while on
treated turf grass - The toddler MOEs from incidental exposure arising
from hand-to-mouth transfer of diquat dibromide were not a risk concern
with the low end scenario MOE of 290 and the high end scenario MOE of
145.

4) Toddler ingestion of soil from treated area - The toddler MOEs for
ingesting (or mouthing) of treated soil were not a risk of concern with
the low end MOE of 83,300, and a high end MOE of 41,670.

Table 11.  Dermal Post-application Risks to Toddlers and Adults When
Reentering Treated Lawns on Day 0  after Sprays have Dried



Scenario	

Range

Finder1	

Application

Rate

(lb ai/acre)	

Fraction

of Residue

Retained	

Transfer 

Coefficient

(cm2/hr)	

Exposure 

Duration

(hours)	

Body 

Wt

(kg)	

Daily 

Dermal 

Dose2 (mg/kg/d)	

Dermal MOE3





Toddler	

Low End	

0.25	

0.05	

5,200	

2	

15	

0.09	

270

	

High End	

0.5	

0.05	

5,200	

2	

15	

0.19	

130



Adult	

Low End	

0.25	

0.05	

14,500	

2	

70	

0.051	

480

	

High End	

0.5	

0.05	

14,500	

2	

70	

0.11	

220

1	Low end ranges are derived from the lowest labeled application rates,
while the high end ranges are derived from the highest labeled rates EPA
Reg. No. 10182-404.  These application rates represent broadcast
application to dormant, established turf grass. 

2	Dermal potential dose rates are calculated as follows:

PDRt = DFRt * CF1 * Tc * ET

where:													

PDRt	=	potential dose rate on day “t” (mg/day).

DFRt	= 	dislodgeable foliar residue on day “t” (ug/cm2).

CF1	=  	weight unit conversion factor to convert ug units in the DFR to
mg for the  daily dose (0.001 mg/ug)

Tc		=  	transfer coefficient (cm2/hr).

ET		= 	exposure time (hr/day).

and

DFRt + AR * F * (1-D)t * CF2 * CF3 

where:

AR	= 	application rate (lbs ai/ft2 or lb ai/acre).

F		= 	fraction of ai retained on foliage (unitless).

D		= 	fraction of residue that dissipates daily (unitless).

t		=	post-application day on which exposure is being assessed.

CF2	= 	weight unit conversion factor to convert the lbs ai in the
application rate to ug for the DFR value (4.54E8 ug/lb)

CF3	=	area unit conversion factor to convert the surface area units
(ft2) in the    application rate to cm2 for the DFR value (1.08E-3
ft2/cm2 or 24.7E-9 acre/cm2 if the application rate is per acre)	

3 	Post-application Dermal MOE = Oral NOAEL (1 mg/kg/day)/[Daily Dermal
Dose (mg/kg/day) x Dermal Absorption Value (4.1%)].  MOEs are reported
to two significant figures.



Table 12.    Oral Post-application Risks to Toddlers from Oral Exposures
When Reentering Treated Lawns



Type of

Exposure	

Range

Finder1	

Application

Rate

(lb ai/acre)	

Fraction

of Residue

Available on Foliage	

Ingestion Rate or Other Assumptions	

Saliva Extraction Factor 	

Exposure 

Duration

(hours)	

Body 

Wt

(kg)	

Daily 

Oral 

Dose2 

(mg/kg/d)	

MOE 3





Exposure to  Treated Turf Grass via object-to-mouth activities5	

Low End	

0.25	

0.20	

25 cm2/day ingestion	

NA	

2	

15	

0.0009	

1,110

	

High End	

0.5







0.0018	

560



Hand-to-Mouth Activity While on Treated Turf Grass	

Low End	

0.25	

0.05	

20 hand-to-mouth events per hour;

20 cm2/day exposed surface area per event	

50%

	

0.0035	

290

	

High End	

0.5







0.007	

145



Soil6	

Low End	

0.25	

NA	

100 mg/day ingestion

	

NA

	

0.000012	

83,330

	

High End	

0.5







0.000024	

41,670

1	Low end ranges are derived from the lowest labeled application rates,
while the high end ranges are derived from the highest labeled rates EPA
Reg. No. 10182-404.  These application rates represent broadcast
application to dormant, established turf grass. 

2	Object-to-mouth exposure to treated turf grass potential dose rates
from ingestion are calculated as follows:

PDRt = GRt * IgR * CF1

where:

PDRt	=	potential dose rate on day “t” (mg/day)

GRt	=	grass (and plant matter) residue on day “t” (ug/cm2)

IgR	=	ingestion rate of grass (cm2/day)

CF1	=	weight unit conversion factor to convert ug units in the DFR to mg
for the  daily dose (0.001 mg/ug)

Hand-to-mouth potential dose rates from ingestion are calculated as
follows:

PDRt = DFRt * SA * FQ * SEF * ET * CF1

where:

PDRt	=	potential dose rate on day “t” (mg/day).

DFRt	=	dislodgeable foliar residue on day “t” (ug/cm2 turf).

SA	=	surface area of the hands (cm2/event).

SEF	=	saliva extraction factor (%)

FQ	=	frequency of hand-to-mouth activity (events/hr).

ET	=	exposure time (hr/day).

CF1	=	weight unit conversion factor to convert ug units in the DFR value
to mg for the daily exposure (0.001 mg/ug)

Ingestion of soil from treated area potential dose rates from ingestion
are calculated as follows:

PDRt = SRt * Ing * CF1

where:

PDRt	=	potential dose rate on day ‘t” (mg/day)

SRt 	=	soil residues on day “t” (ug/g)

IgR	=	ingestion rate of soil (mg/day)

CF1	=	weight unit conversion factor to convert ug units in the residues
on the soil to g for the  daily dose (1.0E-6 g/ug) 

3Postapplication oral MOE = Oral NOAEL(1 mg/kg/day)/Daily Oral
Dose(mg/kg/day).  Oral NOAEL determined from a rabbit developmental
study.  MOEs are reported to two significant figures; an acceptable MOE
is at least 100.   

Aggregate Post-Application Exposure and Risk Estimate

The short term aggregate post-application risk is the estimated risk
associated with combined risks from the short term dermal and oral
post-application exposures.  Given the observed effects at the
recommended NOAELs for oral and dermal pathways, HED believes that risk
from these exposure routes can be reasonably added.  The inhalation
risks are not aggregated because the NOAEL for this exposure pathway is
based on a distinct target organ effect.  Aggregate post-application
risk is estimated for the child exposure scenario only since the child
may be exposed via both oral and dermal pathways while the adult
exposure is from the dermal route only.  The short term aggregate MOE
for the child  is calculated by adding exposure estimates from the oral
and dermal pathways using the formula presented below.  The child
aggregate risk combines the highest exposures from the post-application
scenarios (i.e., high-end oral and dermal post-application reentry
exposure).   The calculated short-term aggregate MOE is presented in
Table 13.   The short-term aggregate post-application MOE for the
high-end exposure scenario for 1-6 year old children is 70.  The
residential aggregate risk combines screening level risk estimates from
individual exposure pathways and should be viewed as a highly
conservative estimate which is certain to over-estimate risk.  The
estimated risk from the individual pathways is based on high-end
assumptions, i.e., highest application rates, child reentry/play on the
day of treatment, and a high-end turf transferable residue factor of 5%.
 In addition, based on a human dermal absorption study cited by the
registrant which shows dermal absorption of 0.3%, use of a 4.1% dermal
absorption factor is likely to result in a further overestimation of
risk.  A refined analysis would result in lower exposure estimates and
higher MOEs.

MOE Post-Application CHILD 	=                        1                  
  

               1          +         1         	 		          		       			
	           MOEDERMAL           MOEORAL	

where:		

MOEdermal = Short Term Dermal NOAEL (mg/kg/day) ( (Short Term Dermal
Exposure (mg/kg/day)  x Dermal Absorption Factor)

MOEoral = Short Term Oral NOAEL (mg/kg/day) ( Short Term Oral Exposure
(mg/kg/day)	

	Table 13. Short Term Aggregate Post-Application Risk to Child



Population

Subgroup	

Short Term Oral NOAEL

mg/kg/d	

Short Term Dermal NOAEL

mg/kg/d	

Short Term Dermal Re-entry Exposure 

mg/kg/d1	

Short Term Oral    - Hand to Mouth - Exposure

mg/kg/d	

Oral MOE	

Dermal MOE	

Post-Application Aggregate MOE



Child 1-6	

1	

1	

0.0078	

0.007	

145	

130	

70

1 Includes Dermal Absorption Factor of  4.1% (0.19 mg/kg/d x 0.041 =
0.0078)	

4.4.2	Recreational								

 

Two recreational post-application exposure scenarios were assessed:

1)  Recreational golfer exposure from playing on treated turf grass
(adults) - This post-application scenario estimates dermal exposures and
risk to adult golfers from dermal contact with turf grass on golf
courses that has been previously treated with diquat dibromide.  The
scenario assumes that a golfer re-enters the course after diquat
dibromide sprays have dried and then play a four hour round of golf.

2)  Swimming exposure to treated ponds and lakes - This
post-application scenario estimates dermal exposures and risk to adult
and 7-10 year old swimmers who re-enter treated ponds and lakes. Several
of the diquat dibromide labels intended for aquatic weed control uses
have swimming re-entry intervals of 0 days (example: EPA Reg. No.
10182-404).

The exposure and risk calculations, critical assumptions, and results
for the recreational golfer exposure scenarios are provided in Table 14.
 Exposure assumptions for the golfer scenario are summarized below.

Recreational Golfer Exposure Assumptions

The following assumptions were made in the dermal exposure calculations
for recreational golfer exposure:	

Application rates range from a low-end rate of 0.25 to a high-end rate
of 0.50 lbs ai/acre.

Turf transferable residue is equal to 5 % of the application rate.

Turf transfer coefficient is 14,500 cm2/hr for adult

Exposure duration 4 hours for exposure to golf course turf. 

Body weight is 70 kg for adults

Table 14.  Dermal Post-application Risks Adults When Reentering Golf
Courses on Day 0 after Sprays have Dried



Scenario	

Range

Finder1	

Application

Rate

(lb ai/acre)	

Fraction

of Residue

Retained	

Transfer 

Coefficient

(cm2/hr)	

Exposure 

Duration

(hours)	

Body 

Wt

(kg)	

Daily 

Dermal 

Dose2 (mg/kg/d)	

Dermal MOE3





Adult Golfer	

Low End	

0.25	

0.05	

500	

4	

70	

0.0034	

7,100

	

High End	

0.5	

0.05	

500	

4	

70	

0.0077	

3,200

1 Golfer durations are assumed to be 4 hours for an 18-hole round of
golf.

2 Dermal potential dose rates are calculated as follows:  PDRt = DFRt *
CF1 * Tc * ET

3 Post-application Dermal MOE = Oral NOAEL (1 mg/kg/day)/[Daily Dermal
Dose (mg/kg/day) x Dermal Absorption Value (4.1%)].  MOEs are reported
to two significant figures.

Swimmer Exposure Assumptions

In order to assess potential exposures to swimmers who re-enter treated
ponds and lakes, HED used the Swimmer Exposure Assessment Model
(SWIMODEL).   The SWIMODEL was developed for estimating the human
exposure doses to the pesticides and toxic pollutants in swimming pools.
 This model is a modification of a study used by J. A. Beech (1980) for
estimating exposure to Trihalomethanes (THM) in swimming pools.  Clearly
swimming in ponds and lakes is different than pools in many ways;
however, the basic exposure to chemicals in the water column are similar
enough to warrant using this model for this TRED. 

The model is based on exposure routes and age-specific contact factors,
exposure duration and frequency, chemical/physical properties of the
pollutant, and pollutant concentration, total exposure doses can be
approximated by the model.  For this TRED child (age 7-10) and adult 
swimmers were modeled.  One diquat dibromide concentration in the lake
was modeled for this analysis: 20 ppb – the maximum contaminant goal
which was reported by EFED as a high end, monitoring data endpoint.  
The exposure and risk calculations, critical assumptions, and results
for the swimmer scenario are provided in Table 15.

Table 15. Child and Adult Exposure and Risk to Swimming in Treated Lakes
on Day 0



Exposed Population	

Concentration in Lake (ppm)1	

Total Exposure (mg/event)2	

Body Weight (kg)3	

Total Dose (mg/kg/day)4	

MOE5



Child (age 7-10) 	

20	

0.062	

37.8	

0.0016	

630



Adult	

20	

0.0087	

84.4	

0.0001	

10,000

1  The concentration in the lake was run at 20 ppb (the maximum
contaminant goal reported by EFED as a high end, monitoring data
endpoint).

2 Total exposure is a combination of exposures via the following routes:
oral, dermal, buccal/sublingual, orbital/nasal, aural, and inhalation.

3 Body weights represent the 90% for the population being modeled.

4 Total dose (mg/kg/day) = Total exposure (mg/event) / body weight (kg)

5 MOE = Oral NOAEL 1 mg/kg/day / Total dose (mg/kg/day)

Recreational Exposure and Risk Estimates

A target MOE of 100 for both the dermal and incidental oral routes is
considered adequate for the recreational risk assessment.  Results of
the recreational exposure assessment are presented in Tables 14 and 15
and are summarized below.

1)  The MOEs for an adult playing a round of golf on a treated golf
course were not a risk concern.  MOEs ranged from 7,100 for the low-end
exposure scenario to 3,200 for the high-end scenario.

2) Estimated MOEs were not a risk of concern for child or adult
swimmers.  MOEs for children age 7-10 ranged from 630 to 180.  MOEs for
adults ranged from 10,000 to 770.

4.5	Incident Reports 	

The vast majority of incident reports for diquat fall into one of two
categories: (1) people exposed while handling the product for its
intended purpose; and, (2) people who drank the pesticide either by
accident or on purpose (i.e., in order to harm themselves) (M. Spann and
J. Blondell, 10/15/01,  D278482).  According to a several of pesticide
incident sources, diquat dibromide ranked relatively high on the list of
pesticides with reported incidents.  Detailed descriptions of 112 cases
submitted to the California Pesticide Illness Surveillance Program
(1982-1999) were reviewed.  In 76 of these cases, diquat was used alone
or  was judged to be responsible for the health effects.  Only cases
with a definite, probable or possible relationship were reviewed. 
Diquat ranked 50th as a cause of systemic poisoning in California based
on data for 1982 through 1999.   Similarly, on the list of the top 200
chemicals for which National Pesticide Telecommunications Network
received calls from 1984-1991 inclusively, diquat was ranked 54th with
74 incidents in humans reported. From the review of California data, it
appears that a majority of cases involved systemic, eye, and skin
illnesses such as eye irritation/pain, skin rash, chemical burns,
chemical conjunctivitis, nausea, and vomiting.  Poison Control Center
data tend to support the California data, eye irritation/pain, erythema,
rash, skin itching, corneal abrasion, lacrimation, dyspnea, coughing,
and choking were the most common effects reported due to exposure.  Oral
exposure to even modest amounts can lead to severe poisoning and even
death.

4.5.1	Residential Handler Incidents

 Of all the handler incident reports, “applicator” was associated
with more exposures than any other category.  These illnesses included
symptoms of burning and itching eyes, dizziness, nausea, dyspnea,
rashes, chemical conjunctivitis, vomiting, dermatitis, and chemical
burns. 

4.5.2 	Postapplication and Recreational Incidents

Not many cases of post-application exposures have been reported, however
there was one incident involving swimmers.  In this case, two boys swam
in a lake that had been treated with diquat three days prior to the
activity.  Both boys developed a rash, one much more severe than the
other.  The nine year old boy with the more serious rash was treated by
a physician while the other boy was treated at home.  No information was
provided on levels of diquat in the treated lake.

5.0 	AGGREGATE RISK ASSESSMENT AND RISK CHARACTERIZATION

The aggregate risk assessment integrates the assessments conducted for
dietary and residential exposure.  Since there is potential for
concurrent exposure via the food, water and residential pathways, the
combined exposures are estimated using the methodology described below
and are compared with monitoring-based estimates of drinking water
contamination determined by EFED.  All routes of diquat dibromide
exposure have been considered.  Aggregate exposure pathways for adults
include dietary, drinking water, and dermal and inhalation exposures
from application and post-application activities.  Aggregate exposure
pathways for children include dietary, drinking water, and dermal and
oral exposures from post-application exposure. 

5.1  	Acute Aggregate Risk Assessment

The acute aggregate risk is the estimated risk associated with combined
acute food and drinking water exposure.  The acute aggregate MOE is
calculated by adding acute dietary exposure estimates with drinking
water levels obtained from monitoring data using the formula presented
below.  The target MOE for acute aggregate risk is 100.  The calculated
acute aggregate

MOEs are presented in Table 16.  Acute aggregate risk MOEs are below
EPA’s level of concern (i.e., (to 100) for all population subgroups. 
Since the estimated risk from the individual pathways are considered to
be screening level, the aggregates should be considered as a highly
conservative estimate of risk. 

MOE Aggregateacute	=   	                  1                    

        1         +           1       	        	

            		MOEFOOD          MOEWATER

where:

 MOEfood = Acute Dietary NOAEL (mg/kg/day) ( Acute Dietary Exposure
(mg/kg/day)

 MOEwater= Acute Dietary NOAEL (mg/kg/day) ( Acute Water Exposure
(mg/kg/day)

and: 

Acute Water Exposure = Acute Water EEC (mg/L) x Drinking Water
Consumption (L/day) 

(mg/kg/d)			Body Weight (kg)	

	Table 16.  Acute Aggregate Risk 



Population Subgroup	

Acute Dietary NOAEL (mg/kg/day)	

Acute Dietary Exposure

(mg/kg/day)	

Acute Water

EEC

(mg/L)	

DW Consumption 

(L/day)	

Body Weight

(kg)	

Acute Water Exposure

(mg/kg/day)	

Acute

Aggregate MOE



US Population (total)	

75	

0.0039	

0.02	

2	

70	

0.00057	

16773



All Infants (<1 year old)	

75	

    0.0035    	

0.02	

1	

10	

0.002	

13636



Children 1-6 years old	

75	

 0.0054	

0.02	

1	

10	

0.002	

10135



Children 7-12 years old	

75	

 0.0033	

0.02	

1	

10	

0.002	

14151



Females 13-50 years old	

75	

 0.0032	

0.02	

2	

60	

0.00067	

19397



Males 13-19 years old	

75	

  0.0030	

0.02	

2	

70	

0.00057	

21000



Males 20+ years old	

75	

  0.0035	

0.02	

2	

70	

0.00057	

18421



Seniors 55+ years old	

75	

  0.0023	

0.02	

2	

70	

0.00057	

26119

	

5.2 	Short-Term Aggregate Risk Assessment

The short term aggregate risk is the estimated risk associated with
combined risks from the following pathways: chronic dietary intake,
average annual drinking water exposures, and short term dermal and oral
(if applicable) exposures.  Given the observed effects at the
recommended NOAELs for each of these pathways, HED believes that risk
from these exposure routes can be reasonably added.  The inhalation
risks are not aggregated because the NOAEL for this exposure pathway is
based on a distinct target organ effect.  The short term aggregate MOE
is calculated by adding exposure estimates from each of these pathways
using the formula presented below.  The adult aggregate risk combines
the highest exposures from the various application and post-application
scenarios (i.e., high-end, low pressure handwand applicator exposure,
and high-end adult reentry exposure).  Likewise, the highest dermal and
oral exposures from the assessed scenarios are added for the estimated
child risk aggregate.  The calculated short-term aggregate MOEs are
presented in Table 14.  The short term aggregate MOE for the high-end
exposure scenario for the adult applicator is 100.  The short-term
aggregate MOE for the high-end toddler exposure scenario is 55.  

The short-term aggregate risk combines screening level risk estimates
from individual exposure pathways and should be viewed as a highly
conservative estimate, certain to over-estimate risk.  A refined
analysis would result in lower exposure estimates and higher MOEs. 
Possible refinements include: 1) revising the dietary assessment to
account for actual percent crop treated (i.e., no use involves 100% crop
treated; most involve <1% CT) and to reflect residues based on field
trial data, which are generally much lower than tolerance levels; 2)
revising estimated environmental concentrations in drinking water
sources to reflect lower concentrations in a sizeable majority of
monitoring samples (i.e., monitoring data from 1993-1997 showed an MCL
exceedence rate of < 1%); and 3) refining the residential exposure
scenarios to include more plausible assumptions regarding a) the use of
protective clothing for applicators (i.e., no protective clothing was
factored into assessments of residential exposure), b) application rates
(i.e., highest rates are assumed), c) day of reentry (i.e., children and
adults are assumed to reenter and work/play on the lawn on the day of
treatment), d) turf transferable residue (i.e., TTR is assumed to be 5%
of the application rate - a high-end assumption), and e) dermal
absorption factor (i.e., dermal absorption is assumed to be 4.1% based
on a rat study; a human dermal absorption study cited by the registrant 
estimates dermal absorption to be about 0.3% (Feldman RJ and Maibach HI,
“Percutaneous penetration of some pesticides and herbicides in man”
Tox. Appl. Pharm. 28 126-132, 1974)). 

MOE Aggregate SHORT TERM	=                                              
    1                                               

                 1         +           1          +           1         
 +        1                   	 		          	      MOEFOOD              
MOEWATER    MOEDERMAL        MOEORAL*	* If applicable

where:	

MOEfood  = Short Term Oral NOAEL (mg/kg/day) ( Chronic Dietary Exposure
(mg/kg/day)

MOEwater = Short Term Oral NOAEL (mg/kg/day) (  Chronic Water Exposure
(mg/kg/day)

MOEdermal = Short Term Dermal NOAEL (mg/kg/day) ( (Short Term Dermal
Exposure (mg/kg/day)  x Dermal Absorption Factor)

MOEoral = Short Term Oral NOAEL (mg/kg/day) ( Short Term Oral Exposure
(mg/kg/day)

and: 	

Water Exposure = Chronic  Water EEC (mg/L) x Drinking Water Consumption
(L/day) 

      (mg/kg/d)				Body Weight (kg)			

	Table 17. Short Term Aggregate Risk



Population

Subgroup	

Short Term Oral NOAEL

mg/kg/d	

Chronic Dietary Exposure

mg/kg/d	

Annual Avg Water Exposure

mg/kg/d	

Short Term Dermal NOAEL

mg/kg/d	

Short Term Dermal Exposure - Handwand Applicator

mg/kg/d1	

Short Term Dermal Exposure Re-entry

mg/kg/d1	

Short Term Oral    - Hand to Mouth - Exposure

mg/kg/d	

Short Term Aggregate MOE



Male 20+	

1	

0.0018	

0.00057	

1	

0.003	

0.0045	

NA	

100



Child 1-6	

1	

0.0031	

0.002	

1	

NA	

0.0078	

0.007	

55

1 Includes Dermal Absorption Factor of  4.1% (0.11 mg/kg/d x 0.041 =
0.0045; 0.19 mg/kg/d x 0.041 = 0.0078)

5.3 	Chronic Aggregate Risk Assessment

The chronic aggregate risk is the estimated risk associated with
combined chronic food and drinking water exposure.  The chronic
aggregate MOE is calculated by adding chronic  dietary exposures and
estimated average annual drinking water concentrations using the formula
presented below.  The target MOE for chronic aggregate risk is 100.  The
calculated chronic aggregate MOEs are presented in Table 18.   Chronic
aggregate risk MOEs are below EPA’s level of concern (i.e., (to 100)
for all population subgroups.  Again, since the aggregate combines
screening level risks from individual exposure pathways, the chronic
aggregate risk estimate is highly conservative.

MOE AggregateCHRONIC	=   	                  1                    

        1         +           1       	        	

            		MOEFOOD          MOEWATER

where:

 MOEfood = Chronic Dietary NOAEL (mg/kg/day) ( Chronic Dietary Exposure
(mg/kg/day)

 MOEwater= Chronic Dietary NOAEL (mg/kg/day) ( Chronic Water Exposure
(mg/kg/day)

and: 

Chronic Water Exposure = Chronic Water EEC (mg/L) x Drinking Water
Consumption (L/day) 

(mg/kg/d)				   Body Weight (kg)

	Table 18.  Chronic Aggregate Risk



Population Subgroup	

Chronic Dietary NOAEL (mg/kg/day)	

Chronic Dietary Exposure

(mg/kg/day)	

Chronic Water

EEC

(mg/L)	

DW Consumption 

(L/day)	

Body Weight

(kg)	

Annual Avg Water Exposure

(mg/kg/day)	

Chronic 

Aggregate MOE



US Population (total)	

0.5	

0.0019	

0.02	

2	

70	

0.00057	

202



All Infants (<1 year old)	

0.5	

0.0017	

0.02	

1	

10	

0.002	

135



Children 1-6 years old	

0.5	

0.0031	

0.02	

1	

10	

0.002	

98



Children 7-12 years old	

0.5	

0.0021	

0.02	

1	

10	

0.002	

122



Females 13-50 years old	

0.5	

0.0017	

0.02	

2	

60	

0.00067	

211



Males 13-19 years old	

0.5	

0.0018	

0.02	

2	

70	

0.00057	

211



Males 20+ years old	

0.5	

0.0019	

0.02	

2	

70	

0.00057	

202



Seniors 55+ years old	

0.5	

0.0015	

0.02	

2	

70	

0.00057	

241



6.0 	CUMULATIVE RISK

The Food Quality Protection Act (1996) stipulates that when determining
the safety of a pesticide chemical, EPA shall base its assessment of the
risk posed by the chemical on, among other things, available information
concerning the cumulative effects to human health that may result from
dietary, residential, or other non-occupational exposure to other
substances that have a common mechanism of toxicity.  The reason for
consideration of other substances is due to the possibility that
low-level exposures to multiple chemical substances that cause a common
toxic effect by a common mechanism could lead to the same adverse health
effect as would a higher level of exposure to any of the other
substances individually.  A person exposed to a pesticide at a level
that is considered safe may in fact experience harm if that person is
also exposed to other substances that cause a common toxic effect by a
mechanism common with that of the subject pesticide, even if the
individual exposure levels to the other substances are also considered
safe.

HED did not perform a cumulative risk assessment as part of this TRED
for diquat dibromide because HED has not yet initiated a review to
determine if there are any other chemical substances that have a
mechanism of toxicity common with that of diquat dibromide.   For
purposes of this TRED, EPA has assumed that diquat dibromide does not
have a common mechanism of toxicity with other substances.

On this basis, the registrant must submit, upon EPA’s request and
according to a schedule determined by the Agency, such information as
the Agency directs to be submitted in order to evaluate issues related
to whether diquat dibromide shares a common mechanism of toxicity with
any other substance and, if so, whether any tolerances for diquat
dibromide need to be modified or revoked.  If HED identifies other
substances that share a common mechanism of toxicity with diquat
dibromide, HED will perform aggregate exposure assessments on each
chemical, and will begin to conduct a cumulative risk assessment once
the final guidance HED will use for conducting cumulative risk
assessments is available.    

HED has recently developed a framework that it proposes to use for
conducting cumulative risk assessments on substances that have a common
mechanism of toxicity.  This guidance was issued for public comment on
June 30, 2000 (65 FR 40644-40650) and is available from the OPP Website
at: http://www.epa.gov/fedrgstr/EPA-PEST/2000/June/Day-30/6049.pdf   In
the draft guidance, it is stated that a cumulative risk assessment of
substances that cause a common toxic effect by a common mechanism will
not be conducted until an aggregate exposure assessment of each
substance has been completed.  The proposed guidance on cumulative risk
assessment of pesticide chemicals that have a common mechanism of
toxicity is expected to be finalized by December 2001.

Before undertaking a cumulative risk assessment, HED will follow
procedures for identifying chemicals that have a common mechanism of
toxicity as set forth in the “Guidance for Identifying Pesticide
Chemicals and Other Substances that Have a Common Mechanism of
Toxicity” (64 FR 5795-5796, February 5, 1999).

7.0 	DATA NEEDS/LABEL REQUIREMENTS

All product chemistry data are required for the Syngenta 41.1% and 37.3%
Formulation Intermediates (EPA Reg. Nos. 10-1062 and 100-1063). 
Magnitude of the residue in plants studies are required for sorghum
aspirated grain fractions and  soybean aspirated grain fractions. 

((((( PAGE  2 ((((( NUMPAGES  45 

((((( PAGE  27 ((((( NUMPAGES  45 

