Skip to main content

Full text of "DTIC AD1016093: Rain-Induced Wash-Off of Chemical Warfare Agent (VX) from Foliar Surfaces of Living Plants Maintained in a Surety Hood"

See other formats


EDGEWOOD CHEMICAL BIOLOGICAL CENTER 



U.S. ARMY RESEARCH, DEVELOPMENT AND ENGINEERING COMMAND 

Aberdeen Proving Ground, MD 21010-5424 


ECBC-TR-1405 


RAIN-INDUCED WASH-OFF OF CHEMICAL WARFARE 
AGENT (VX) FROM FOLIAR SURFACES OF 
LIVING PLANTS MAINTAINED IN A SURETY HOOD 


Mark V. Haley 
Ronald T. Checkai 
Michael Simini 
Richard J. Lawrence 

RESEARCH AND TECHNOLOGY DIRECTORATE 


Michael W. Busch 

EXCET, INC. 
Springfield, VA 22150-2519 

September 2016 


Approved for public release; distribution is unlimited. 


0 us ARMY 

RDECOM 


Disclaimer 


The findings in this report are not to be construed as an official Department of the Army 
position unless so designated by other authorizing documents. 



REPORT DOCUMENTATION PAGE oMBNaom^es 

Public reporting burden for this collection of information is estimated to average 1 h per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data 
needed, and completing and reviewing this collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this 
burden to Department of Defense, Washington Headquarters Services, Directorate for Information Operations and Reports (0704-0188), 1215 Jefferson Davis Highway, Suite 1204, Arlington, VA 22202-4302. 
Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to any penalty for failing to comply with a collection of information if it does not display a currently valid 

OMB control number. PLEASE DO NOT RETURN YOUR FORM TO THE ABOVE ADDRESS. 

1. REPORT DATE (DD-MM-YYYY) 2. REPORT TYPE 

XX-09-2016 Final 

3. DATES COVERED (From - To) 

May 2014-Sep 2015 

4. TITLE AND SUBTITLE 

Rain-Induced Wash-Off of Chemical Warfare Agent (VX) from Foliar 

Surfaces of Living Plants Maintained in a Surety Hood 

5a. CONTRACT NUMBER 

5b. GRANT NUMBER 

5c. PROGRAM ELEMENT NUMBER 

6. AUTHOR(S) " 

Haley, Mark V.; Checkai, Ronald T.; Simini, Michael; Lawrence, Richard J. 
(ECBC); and Busch, Michael W. (Excet) 

5d. PROJECT NUMBER 

WBS R.0013813.81.4 

5e. TASK NUMBER 

5f. WORK UNIT NUMBER 

7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 

Director, ECBC, ATTN: RDCB-DRT-M, APG, MD 21010-5424 

Excet, Inc.; 6225 Brandon Ave., Suite 360; Springfield, VA 22150-2519 

8. PERFORMING ORGANIZATION REPORT 

NUMBER 

ECBC-TR-1405 

9. SPONSORING / MONITORING AGENCY NAME(S) AND ADDRESS(ES) 

Defense Threat Reduction Agency; 8725 John J. Kingman Road, MSC 6201; 
Fort Belvoir, VA 22060-6201 

10. SPONSOR/MONITOR’S ACRONYM(S) 

DTRA 

11. SPONSOR/MONITOR’S REPORT NUMBER(S) 


12. DISTRIBUTION / AVAILABILITY STATEMENT 

Approved for public release; distribution is unlimited. 


13. SUPPLEMENTARY NOTES 


14. ABSTRACT: 

In separate, replicated experiments, we experimentally established wash-off coefficients (k w ) for the chemical agent VX on 
grass, utilizing 1 and 3 pL VX droplets at 0.017 (1 min), 1, and 4 h after dissemination. A 10 mm (0.39 in.) rain event at 
0.017 h after dissemination washed off 95 and 83% of the 1 and 3 pL VX droplets, respectively. At 1 h after dissemination, a 
10 mm rain event washed off 0.03 and 0.5% of the 1 and 3 pL VX droplets, respectively. At 0.017 h after dissemination, the k w 
values for 1 and 3 pL droplets were 0.095 and 0.083 mm" 1 , respectively. At 1 and 4 h after dissemination, the k w values for VX 
were approximately 3 orders of magnitude less than those at 0.017 h. Grass contaminated with 3 pL VX droplets was exposed 
to multiple (lOx) 100 pL light rain events, followed by multiple (lOx) surface wipes at 0.017, 0.5, 1, 4, and 24 h after 
dissemination. The cumulative proportions of 3 pL VX droplets washed off by the rain events at 0.017 and 1 h after 
dissemination were 75.3 and 0.99%, respectively, and the cumulative proportions of VX wiped off after those rain events were 
0.8 and 0.3%, respectively. Results from these investigations of agent-plant interactions provide input for predictive models 
and functional information for “Go/No-Go” decisions that can affect soldiers on agent-contaminated battlefields. 


15. SUBJECT TERMS 

Wash-off coefficient Echinochloa crus-galli Foliage 

Chemical warfare agent (CWA) O-cthy I-.S’-(2-di isopropyl am inocthyl) methyl phosphonothiolate (VX) Barnyard grass 


16. SECURITY CLASSIFICATION OF: 

17. LIMITATION OF 
ABSTRACT 

18. NUMBER OF 
PAGES 

19a. NAME OF RESPONSIBLE PERSON 

Renu B. Rastogi 

a. REPORT 

b. ABSTRACT 

c. THIS PAGE 



19b. TELEPHONE NUMBER (include area code) 

U 

u 

u 

uu 

28 

(410) 436-7545 


Standard Form 298 (Rev. 8-98) 
Prescribed by ANSI Std. Z39.18 


































Blank 


n 



PREFACE 


The work described in this report was authorized under project no. 

WBS R.0013813.81.4 The work was started in May 2014 and completed in September 2015. 

The use of either trade or manufacturers’ names in this report does not constitute 
an official endorsement of any commercial products. This report may not be cited for purposes of 
advertisement. 


This report has been approved for public release. 


Acknowledgments 


The authors acknowledge Dr. David J. McGarvey (U.S. Army Edgewood 
Chemical Biological Center [ECBC]) for conducting VX purity determinations using nuclear 
magnetic resonance spectroscopy. The authors would also l ik e to thank Dr. Bruce King (ECBC) 
and Dr. Morgan Minyard (Defense Threat Reduction Agency [DTRA]) for consultation 
throughout the duration of the project. Funding for this project was provided by DTRA. 



Blank 


IV 



CONTENTS 


1. INTRODUCTION.1 

2. METHODS.2 

2.1 Plant Selection.2 

2.2 Chemicals.2 

2.3 Dissemination of VX Droplets onto Leaves.3 

2.4 Simulated Rainfall and Leaf Surface Wipe Events.5 

2.5 Extractability of VX from Wipes.6 

2.6 VX Stability Studies.7 

2.7 Analytical Determination of VX.7 

3. RESULTS.9 

3.1 Extractability of VX from Wipes and Glass.9 

3.2 VX Stability Studies.9 

3.3 Wash-Off Coefficients.9 

3.4 Multiple Rain and Leaf Wipe Events.10 

4. DISCUSSION.13 

5. CONCLUSION.14 

LITERATURE CITED.15 

ACRONYMS AND ABBREVIATIONS.17 


v 





















FIGURES 


1. E. crus-galli grass plants in surety hood with plant stands.3 

2. E. crus-galli leaf secured in horizontal position using tape folded in half 

lengthwise (to prevent leaf contact with the adhesive).4 

3. Rain induced wash-off from VX-contaminated grass leaves.12 

4. Residual VX recovery from surface wipes.12 


TABLES 


1. PEARL Model Wash-Off Coefficient Classes.5 

2. Volume of Rain Used to Produce a 10 mm Rain Event.6 

3. HPLC Gradient Table for VX Quantitation.8 

4. MRM Mass Transitions.8 

5. Single 10 mm Rain Event Applied to a Grass Leaf Contaminated with 1 pL 

of VX (n = 3).10 

6. Single 10 mm Rain Event Applied to a Grass Leaf Contaminated with 3 pL 

of VX (n = 3).10 

7. Total VX Recovery from Multiple Rain Events (10 x 100 pL) and Surface 

Wipes from Contaminated Grass Leaves.11 


vi 














RAIN-INDUCED WASH-OFF OF CHEMICAL WARFARE AGENT (VX) FROM 
FOLIAR SURFACES OF LIVING PLANTS MAINTAINED IN A SURETY HOOD 


1. INTRODUCTION 

Defensive capabilities are needed to identify threats in the event of the release of 
chemical warfare agent (CWA) into the natural environment. Commanders who have soldiers 
under battlefield conditions must be armed with functional information for “Go/No-Go” 
decisions related to the exposure of soldiers to CWA on agent-contaminated battlefields. Little 
scientific information exists that describes the hazards to the solider associated with agent-plant 
interactions. Without a more-complete understanding of these interactions, it is difficult to 
predict the presence and persistence of the potential exposure hazard posed by CWA- 
contaminated foliage. 

The dissemination of CWA in the field can be subject to many environmental 
pathways that affect its fate and may ultimately pose an exposure hazard. A particularly 
important exposure pathway is the agent-plant interface, where many environmental 
functionalities occur including CWA absorption into plant tissue, evaporation from the leaf 
surface, transformation, fixation, photodegradation, and rain-induced wash-off from foliar 
surfaces (Hulbert et al., 2011; Van Emon et al., 1998). The research presented in this report 
focuses on rain-induced wash-off of (9-ethyl -S-( 2-di isopropy 1 aminocthy 1) methyl 
phosphonothiolate (VX) from contaminated grass (Echinochloa crus-galli [L. 1 ] P. Beauv 2 ) 
leaves (also referred to by the common names barnyard grass and Japanese millet). 

The fate of chemical compounds in the environment is influenced by their 
physical and chemical properties as well as ambient meteorological conditions and the 
environmental material with which the compounds interact (Talmage et al., 2007). Persistence on 
plant foliage is also a function of the biochemical, physiological, and micromorphological 
properties of the plant leaves, including leaf epicuticular waxes and cuticle (Sanyal et al., 2006). 
The extent of persistence, penetration (absorption), or evaporation of agent droplets on foliage 
will depend on the vapor pressure, the hydrophilic nature of the agent, and the extent of 
hydrophobicity of the foliar surfaces (Rothamsted, 2013). Therefore, the extent of absorption and 
persistence of CWAs on foliar surfaces and within plant tissues among different plant species 
will vary depending on the compound. 

Our objective in investigating agent-plant interactions in this study was to 
determine whether rainfall reduces the exposure hazard to the Warfighter by decreasing the 
quantity of CWA that remains available on natural living vegetation. This, in turn, would reduce 
the potential for contact transfer of agent onto Army uniforms and provide data for models used 
to predict the fate of CWA and related chemicals in the environment. The pesticide industry uses 
rainfastness characteristics (a measure of the post-dissemination time that pesticides will remain 
effective on foliage following a rain event) to help determine the interval for pesticide 


'L. indicates that Carl Linnaeus is the authority for the species name. 

2 P. Beauv. indicates that Palisot de Beauvois was the author of this botanical name. 


1 



re-application (Wise, 2015). Typically, pesticide studies used to determine rainfastness involve 
the uniform application of a dilute mixture of pesticide to outdoor test plots or in controlled 
chambers (Hulbert et al., 2011; Van Emon et al., 1998). In our studies, we could not use outdoor 
plots or spray into chambers. We used high-purity VX (neat) to simulate battlefield conditions. 
We applied single droplets of neat VX to the live foliar surfaces of the grass, E. crus-galli, which 
was contained in a chemical surety hood. At predetermined time intervals after agent 
dissemination, we applied a simulated rainfall directly onto the agent droplet site using ASTM 
Type I water that was allowed to equilibrate to ambient conditions. The water runoff was 
collected and analyzed to determine the amount of agent that was found in the wash-off solution. 
Immediately after the rain event, we wiped the leaf surface to determine the amount of agent 
remaining that may be available for transfer to Army uniforms by contact. 

The data generated from this study were directly applicable for input into the 
Pesticide Emission Assessment at Regional and Local scales (PEARL) model. The PEARL 
model is used to predict the fate and transformation of a pesticide in soil-plant systems (Leistra, 
2001; Van den Berg and Leistra, 2004). Because VX is an organophosphate (OP) compound, 
with physical properties similar to OP pesticides, the PEARL model is considered to be one of 
the models reasonably useful for predicting the fate of VX in the environment. 


2. METHODS 

2.1 Plant Selection 

We selected the grass species E. crus-galli for research investigations described in 
this report. E. crus-galli is one of the most-prevalent natural grass species worldwide, is tolerant 
of both dry and wet natural habitats, and is used as forage for grazing animals as well as for 
wildlife food and habitat (USDA, 2015). We used novel methods that were developed to enable 
and sustain the culture of living, physiologically healthy plants within a chemical agent surety 
hood, as described by Simini et al. (2016). 

2.2 Chemicals 

The CWA used in this study was VX at 93% purity, Chemical Agent Standard 
Analytical Reference Material (CASARM) grade, Chemical Abstracts Service (CAS) no. 50782- 
69-9), which was stabilized with 5% by weight diisopropylcarbodiimide (CAS no. 693-13-0; 
Sigma-Aldrich Company; St. Louis, MO). Reagent-grade isopropyl alcohol (IPA; CAS no. 67- 
63-0; Sigma-Aldrich) was used as an extractant. The simulated rainwater was ASTM Type I 
water (18 MQ cm) (ASTM, 2004) that was allowed to equilibrate to ambient conditions. 
Miracle-Gro Water Soluble All Purpose Plant Pood (Scotts Miracle-Gro Company; Marysville, 
OH) fertilizer (24% total nitrogen [calculated as N], 8% available phosphate [calculated as 
P 2 O 5 ], 16% soluble potash [calculated as K 2 O], 0.02% boron, 0.07% copper [water soluble], 
0.15% iron [chelated], 0.05% manganese [chelated], 0.0005% molybdenum, 0.06% zinc [water 
soluble], and 1.14% ethylenediaminetetraacetic acid chelating agent) was used to prepare the 
dilute phytonutrient solution (530 mg/L) with ASTM Type I water. 


2 



2.3 


Dissemination of VX Droplets onto Leaves 


Plant stands were constructed to hold the pots in a fixed position. A hole was cut 
in each Petri dish cover, and each pot was placed through the hole and onto the Petri dish 
(Figure 1). Each pot was secured to a ring stand with an adjustable ring clamp. Plant leaves were 
laid horizontally across a ring near the top of the plant canopy and secured to the ring by lengths 
of clear plastic (cellulose acetate) tape that was folded in half lengthwise (thus preventing sticky 
contact of acrylate adhesive to leaf surface) and placed across the leaf surface. The ends of the 
folded tape were then secured to the ring with additional tape, while maintaining slight pressure 
on the leaf surface (Figure 2). This method of securing individual leaves in a horizontal position 
prevented any possible leaf surface damage caused by tape removal and ensured that 
disseminated agent droplets contacted the leaf surface at the point intended and that those 
locations were easily identified for further investigation. Individual leaves on the living plants 
remained secured in this horizontal position during and after dissemination of VX to prevent 
uncontrolled agent deposition throughout testing. Physiologically healthy plants were maintained 
within the chemical agent surety hood environment (Simini et al., 2016). The temperature within 
the surety hood was maintained at 22 ± 2 °C, and the relative humidity was maintained at 
50 ± 10%. The average airflow through the hood was measured at 1.5 ± 0.09 mph (measured at 
the face of the hood using an Airdata Multimeter; ADM-870C; Shortridge Instruments, Inc. 
[Scottsdale, AZ]). 



Figure 1. E. crus-galli grass plants in surety hood with plant stands. 


3 




Figure 2. E. crus-galli leaf secured in horizontal position using tape folded in half lengthwise 

(to prevent leaf contact with the adhesive). 


In separate experiments, single 1 or 3 pL droplets of CASARM-grade VX 
(0.8278 ± 0.0386 mg or 3.010 ± 0.137 mg, respectively) that represent the range of droplet sizes 
expected from CWA dissemination under field conditions (TOP, 2011) were individually 
dispensed onto plant leaves using a calibrated 10 pL Hamilton (Reno, NV) gastight syringe. A 
single droplet was apphed onto a single live leaf that was still attached to the plant; this 
prevented the droplets from merging on the foliage (for the purpose of subsequent 
measurements). The disseminated VX droplets were allowed to equilibrate on the foliage for 
0.017, 1, and 4 h before the treated leaf was removed from the plant. The treated leaf was then 
subjected to a single 10 mm rain event, as required when developing data for PEARL model 
input. Each of these experiments was conducted using quadruplicate replication {n = 4). 

The PEARL model is used to predict the fate and transformation of a pesticide in 
a soil plant system (Leistra et al., 2001; Van den Berg and Leistra, 2004). Because VX is an OP 
compound, with properties similar to OP pesticides, the PEARL model can be used to reasonably 
predict the fate of VX in the natural environment. PEARL model input requires a wash-off 
coefficient (k w ), which is based on the percentage of pesticide washed off the foliage from a 
10 mm rain event. The PEARL model user’s guide (Van den Berg and Leistra, 2004) identifies 
five wash-off classifications on the basis of the percentage of wash-off from a 10 mm rainfall 
(Table 1) and suggests interpolation to more accurately determine k w values from generated 
experimental data. Analysis of the PEARL model wash-off coefficient classes (Table 1), by 
performing a linear regression of these percent compound wash-off versus k w values, yielded a 
coefficient of determination of unity (r 2 = 1.00). We used linear interpolation with our wash-off 
data to determine the values of k w for VX on grass at time points after dissemination. 


4 



Table 1. PEARL Model Wash-Off Coefficient Classes 


Compound 

Wash-Off 

Wash-Off 

Coefficient 

(%) 

(mm 2 ) 

90 

0.09 

70 

0.07 

50 

0.05 

30 

0.03 

10 

0.01 


2.4 Simulated Rainfall and Leaf Surface Wipe Events 

In separate simulated rainfall experiments, the k w values required for input into 
the PEARL model were determined for 1 and 3 pL droplets of VX on foliage. The rain 
application used in these separate studies was a single, 10 mm rain event that was required by the 
PEARL model to represent a moderate rainfall (Llasat, 2001; Hunsche et al., 2007) onto VX- 
contaminated foliage. 

After the disseminated VX had equilibrated with the leaf surface for a 
predetermined amount of time, the contaminated leaf was removed from the plant by holding the 
leaf with a forceps at a position approximately 3 in. (7.5 cm) from the end closest to the stem of 
the leaf containing the VX droplet spread. The leaf was then cut next to the forceps (on the side 
of the forceps nearer the stem). While still holding the cut leaf with the forceps, the end of the 
leaf containing the absorbed VX droplet was inserted into a 50 mL collection vial. The collection 
vial and the leaf were tilted to approximately 45 degrees, then the simulated raindrops 
(cumulatively, 10 mm ± 1.09%) were applied onto the leaf above the site of the agent. This 
caused the applied raindrops to run down the leaf surface and across the target area of agent 
droplet spread, which simulated wash-off that can occur during a natural rain event. The 
simulated rain was applied using a calibrated 5 mL Gilson Pipetman (Middleton, WI) pipette 
(raindrop size of 45.6 ± 2 pL). 

The vials used to collect the wash-off from the 10 mm rain events were prefilled 
with a volume of IPA equal to the volume of raindrops applied. This was done to slow the 
degradation of VX in water (see the VX stability study in Section 2.6), which resulted in a 1:1 
dilution of rainwater with IPA. The volume of raindrops applied was calculated based on the area 
of agent droplet spread at predetermined time points after dissemination (Simini et al., 2016). 
During a 4 h period, the VX droplet spread increased in area from 132 to 163 mm 2 for the 
1 pL droplets and from 166 to 303 mm 2 for the 3 pL droplets. The volumes of rain applied for 
the 1 and 3 pL agent droplets to yield 10 mm (0.39 in.) of rainfall were calculated based on the 
target area of the agent droplet spread at the specific sampling times after dissemination 
(Table 2). 


After the 10 mm rain event was completed, the leaf was removed from the vial 
and placed onto a clean plastic Petri dish. Using forceps, a 2.5 x 2.5 cm swatch of VectraR 
QuanTex TX1080 wipe (Texwipe; Kernersville, NC), folded in half, was used to apply a one- 
pass wipe to the surface of the leaf. The wiping motion started at the stem end of the leaf 


5 




segment above the VX droplet application site, wiped through the VX application site, and ended 
at the tip of the leaf. A force of 16 g (± 2.9 g) was applied to the wipe during this process. The 
wipe was then immediately placed into 1000 pL of IPA for a minimum of 1 h to extract VX from 
the wipe. 


Table 2. Volume of Rain Used To Produce a 10 mm Rain Event 2 


Time after 
Dissemination 

(h) 

Area of 1 pL b 
Droplet Spread 
(mm 2 ) 

Area of 3 pL b 
Droplet Spread 
(mm 2 ) 

Rain Applied 
to 1 pL VX 
Droplet Target 
Area 
(mL) 

Rain Applied 
to 3 pL VX 
Droplet Target 
Area 
(mL) 

0.017 

132 

166 

1.32 

1.66 

1 

163 

301 

1.63 

3.01 

4 

135 

303 

1.35 

3.03 


“Data from Simini et at., 2016. 

b Based on the area of the VX droplet spread after dissemination. 


Additional rainfall studies were conducted separately using 3 pL VX droplets that 
were equilibrated on the leaf for 0.017, 0.5, 1, 4, and 24 h. The individual contaminated leaves 
were removed from living plants, and each of these leaves was then subjected to multiple (lOx) 
consecutive rain events of 100 pL (101.7 ± 0.3 pL) each, which represented individual light 
rainfalls that cumulatively produced a moderate rainfall (Hunsche et al., 2007; Llasat, 2001). The 
cumulative rainfall from each series of multiple 100 pL rain events was less than a single 10 mm 
(0.39 in.) rainfall, and their equivalency ranged from 30 to 60% of 10 mm. A calibrated 200 pL 
Gilson Pipetman pipette was used to apply the 100 pL rain events (raindrop size, 17.2 ±1.8 pL). 
After the consecutive 100 pL rain events were completed, the leaf was removed from the 
collection vial, placed into a clean plastic Petri dish, and wiped in the manner described 
previously. However, after the conclusion of the multiple rainfalls onto these contaminated 
leaves, each leaf received 10 separate single-pass wipes, using a fresh, single layer for each wipe. 
Each wipe was immediately placed into a vial containing 1000 pL of IPA and extracted for a 
minimum of 1 h. These additional simulated rainfall experiments were conducted with a 
minimum of three and maximum of six replications. 

2.5 Extractability of VX from Wipes 

To determine the extractability of VX from wipes, a 2.5 x 2.5 cm 2 piece of wipe 
was placed into a 50 mL collection vial, and then a 3 pL droplet of neat VX was placed on the 
wipe. Within 15 s, 1000 pL of IPA was added to the vial, and it was capped. The wipe was 
extracted for a minimum of 1 h before subsamples of the IPA were removed for analysis. To 
determine the VX-absorbing efficiency of the wipe from a glass surface, we conducted 
comparative studies. A 3 pL droplet of VX was placed onto a clean glass disk 
(3.68 cm diameter x 0.071 cm thick) and allowed to equilibrate for 1 min. Using forceps, a wipe 
was drawn across the VX droplet (one pass) in the manner described in Section 2.4. The wipe 
was then placed into 1000 pL of IPA to extract the VX. This was followed with four additional 
separate, single-pass wipes, and then each of these wipes was also extracted in IPA. The glass 
disk was then placed in IPA and extracted for a minimum of 1 h before the IPA was analyzed. 


6 




Positive-control samples were obtained by placing a 3 pL VX droplet directly onto the glass 
disk, immediately placing this disk into IPA, and extracting the contaminated glass disk for a 
minimum of 1 h before analyzing the IPA. All extractability experiments were replicated in 
quadruplicate (n = 4). 

2.6 VX Stability Studies 

In this study, a simulated rain consisting of ASTM Type I water was allowed to 
equilibrate with the atmosphere, similar to natural rainwater, and was then used to displace VX 
from plant leaves. Results of previous reports have shown that the persistence of VX in aqueous 
solution varies depending on the properties of the solution, temperature, and concentration 
(Safety Data Sheet, 2015; Epstein et al., 1974). 

In our studies, the wash-off from a leaf exposed to a single rain event (10 mm of 
rain) was collected into a vial containing enough IPA to produce a 1:1 mixture of IPA and 
rainwater. We hypothesized that if the wash-off samples were diluted 1:1 with IPA and analyzed 
within 24 h, the amount of degradation of VX by hydrolysis subsequent to sampling would be 
insignificant. To test our hypothesis, we conducted VX stability studies by transferring neat VX 
into a 1:1 mixture of IPA and rainwater leaf rinse (rainwater after rinsing a non-contaminated 
leaf) to yield a final VX stock concentration of 929.6 ng/pL. The stock was then immediately 
diluted using the 1:1 mixture of IPA and rainwater leaf rinse to yield additional VX 
concentrations of 93.9 and 9.3 ng/pL. Samples were removed from each VX concentration, 
placed into separate sample vials, and capped. This resulted in three 1 mL samples for each 
concentration to allow the analysis of replicates at each of four time intervals. The concentrations 
of VX in solution were analytically determined at 4, 24, 48, and 168 h after preparation. 

The multiple rain events (100 pL per event) were collected into vials containing 
1000 pL of IPA (final sample contained approximately 90% IPA). Stability studies were also 
conducted in triplicate using a mixture of 10% leaf water rinse and 90% IPA that contained a VX 
concentration of 46.5 ng/pL. The VX concentrations were subsequently determined at 0, 24, and 
96 h. 

A one-way analysis of variance (ANOVA) was used to determine whether there 
were significant differences (probability [p\ < 0.05) among VX concentrations across the time 
intervals used in these stability studies of VX in IPA-water mixtures. The Fisher’s multiple 
comparison test was used to determine significant differences (p < 0.05) in VX concentrations 
between the individual time intervals. 

2.7 Analytical Determination of VX 

Quantitative analysis of VX was conducted using Agilent 6890 gas 
chromatography (GC; Agilent Technologies; Santa Clara, CA) equipped with a flame 
photometric detector. Quantification was achieved using an Agilent DB-5 fused silica column 
(30 m x 0.32 mm, 0.5 mm film thickness). The sample volumes of 1 pL were injected into the 
GC using an Agilent (7683B series) autosampler. Sample-inlet temperature was maintained at 
225 °C, in splitless mode. The initial oven temperature was 80 °C with a temperature ramp rate 
of 45 to 300 °C. A nine-point calibration curve (0.014, 0.072, 0.14, 0.73, 1.45, 3.91, 5.81, 11.62, 


7 



and 23.82 ng/pL) was used to determine the VX concentration in the wash-off solution. The r 2 
for the linear regression of the standard curve throughout these studies was 0.9995 ± 0.0003. The 
instrument limit of detection was 0.005 ng/pL, based on peak-to-peak background noise for this 
method. 


Quantitative analytical determinations for low levels of VX and confirmation of 
GC results were conducted using high-performance liquid chromatography (HPLC) linked with 
tandem mass spectrometry (MS/MS) (Agilent 1260 liquid chromatograph triple-quadrupole mass 
spectrometer with MassHunter data acquisition and analysis software). The HPLC system was 
fitted with an Agilent Eclipse XDB-Cix column (5 pm, 4.6 x 150 mm). Sample injections were 1 
pL. A 13 min separation method was used; the composition of mobile phase A was 0.1% formic 
acid (v/v) in H 2 O, and mobile phase B was 0.1% formic acid (v/v) in methanol (MeOH). The 
gradient conditions used for HPLC separation are shown in Table 3. 


Table 3. H 

[PLC Gradient Table for VX Quantitation 

Time 

(min) 

Mobile Phase A 

(%) 

Mobile Phase B 

(%) 

0 

99.9 

0.1 

2 

99.9 

0.1 

7 

5.0 

95.0 

8 

5.0 

95.0 

11 

99.9 

0.1 

13 

99.9 

0.1 


The HPLC column eluent was delivered to an electrospray ionization source that 
was maintained in positive ion mode. MS/MS discrimination was performed via the multiple 
reaction monitoring (MRM) technique that incorporated an isotope dilution (VX-c/ 5 ) and used the 
following three mass transitions: VX quantitation, VX confirmation, and VX-J 5 internal standard 
(Table 4). 


Table 4. MRM Mass Transitions 


Analyte 

Precursor Mass (Da) 

Product Mass (Da) 

VX-J 5 internal standard 

273 

128 

VX quantitation 

268 

128 

VX qualifier 

268 

86 


Calibration was conducted by plotting the relative responses of VX and \X-d5 as 
a function of concentration. An 11-point calibration curve (5.0-5000 pg/pL of VX, each with 
50 pg/pL of VX-J 5 ) was used to construct a linear calibration curve (1/v weighting). All 
analyzed samples were prepared to contain 50 pg/pL of VX-J 5 as the internal standard, and 
reported VX concentrations were calculated by applying the equation of fit and dilution factors, 
as applicable. The instrument limit of detection was 0.5 pg/pL, as based on the peak-to-peak 
background noise for this method. 


8 




3. 


RESULTS 


3.1 Extractability of VX from Wipes and Glass 

In the wipe extractability experiments, we applied VX to the wipe and then 
allowed the VX to equilibrate for 15 s before extracting the contaminated wipe in IPA. After 
extracting the wipe for 1 h in IPA, we recovered 99.5% (±4.7%) of the VX that had been added 
to the wipe. 


Compared with IPA positive controls, we recovered 96.1% (±3.1%) of the VX 
applied to glass from the first wipe. We recovered a cumulative average of an additional 1.4% 
(±0.4%) of the VX from the additional four wipes that were used in subsequent consecutive 
wipes of glass. Following the wiping procedures, an additional 0.8% (±0.1%) residual amount of 
the initial VX was recovered from the wiped glass disks by direct extraction with IPA. In total, 
98.3% of the initial VX that was placed on the glass disks was recovered. 

3.2 VX Stability Studies 

When significant differences (p < 0.05) were found in the VX concentrations 
among the respective time intervals used in the stability studies of VX in IPA-water mixtures 
using ANOVA, then the Fisher’s multiple comparison test was used to determine the significant 
differences (p < 0.05) in VX concentrations between time intervals. 

When VX was placed in a 1:1 mixture of IPA and rainwater leaf rinse, there was 
no significant decline in VX concentration up to 48 h after preparation (p > 0.05). However, after 
168 h, there was an average decline of 27.1 ± 11% in VX concentration in the 1:1 IPA-to- 
rainwater leaf rinse, which was a significantly different concentration when compared with the 
VX concentration after 4 h (p < 0.05). When VX was in a mixture consisting of 10% rainwater 
leaf rinse and 90% IPA, there was no significant decline in VX concentration up to 96 h 
(p > 0.05). These results confirmed that analytical concentrations of VX determined within 48 h 
of producing rainwater wash-off samples (in corresponding mixtures containing IPA) are 
accurate well within statistical boundaries. All rainwater wash-off samples containing IPA and 
VX were analyzed within 24 h of being produced for all definitive experiments described here. 

3.3 Wash-Off Coefficients 

The k w values from 10 mm rainfall events were determined in separate 
experiments using 1 and 3 pL VX droplets at 0.0172, 1, and 4 h after dissemination. A single 
10 mm rain event at 0.017 h after dissemination washed off 95% of the 1 pL VX droplet and 
83% of the 3 pL VX droplet from the contaminated grass leaf. After 1 h, only 0.03 to 0.5% of 
the respective 1 and 3 pL VX droplets were washed off the leaf by a 10 mm rainfall (Tables 5 
and 6). The k w values for the 1 and 3 pL droplets at 0.017 h after dissemination were 0.095 and 
0.083 mm -1 , respectively. The k w values for the 1 and 3 pL VX droplets at 1 and 4 h after 
dissemination were approximately 3 orders of magnitude less than these respective values at 
0.017 h (Table 5). A single, one-pass wipe was conducted after the 10 mm rain event to 
determine the amount of dislodgeable residual agent remaining on grass leaves after rainfall. The 


9 



amounts of residual VX recoverable by one single-pass wipe of the leaf after a 10 mm rainfall at 
0.017 h after dissemination were 2.7 and 1.7% for the 1 and 3 pL droplets, respectively. After a 
10 mm rainfall at 1 and 4 h after dissemination, the amounts of VX recovered from a single wipe 
were orders of magnitude less than those at 0.017 h after dissemination for both the 1 and 3 pL 
VX droplets (Tables 5 and 6). 


Table 5. Single 10 mm Rain Event Applied to 
a Grass Leaf Contaminated with 1 pL of VX (n = 3) 


Contact 
Time on 
Leaf(h) 

Agent 

Spread Area 
(mm 2 ) 1 

Volume of Rain 
Applied (mL) b 

Recovery 

k w 

(mm 1 ) 

% by Rain 
(±SD) 

% by Wipe c 
(±SD) 

0.017 

132 

1.32 

95.2 (8.7) 

2.7 (2.6) 

0.0952 

1 

163 

1.63 

0.03 (0.03) 

0.02 (0.01) 

3 x 10 5 

4 

135 

1.35 

0.02 (0.002) 

0.006 (0.003) 

2 x 10 5 


“Data from Simini et al. (2016). 

b Volume of 10 mm rain applied was based on surface area of the drop spread at reference time after dissemination. 
“Wipe recovery was conducted after the 10 mm rain event. 

SD: standard deviation (shown in parentheses). 


Table 6. Single 10 mm Rain Event Applied to 
a Grass Leaf Contaminated with 3 pL of VX (n = 3) 


Contact 
Time on 
Leaf(h) 

Agent 

Spread Area 
(mm 2 ) 1 

Volume of Rain 
Applied (mL) b 

Recovery 

kw 

(mm 1 ) 

% by Rain 
(±SD) 

% by Wipe c 
(±SD) 

0.017 

166 

1.66 

83.0 (5.1) 

1.7 (0.9) 

0.083 

1 

301 

3.01 

0.5 (0.7) 

0.08 (0.07) 

5 x 10- 4 

4 

303 

3.03 

0.4 (0.2) 

0.3 (0.2) 

4 x lO^ 4 


“Data from Simini et al., 2016. 

b Volume of 10 mm rain applied was based on surface area of the drop spread at reference time after dissemination. 
“Wipe recovery was conducted after the 10 mm rain event. 

SD: standard deviation (shown in parentheses). 


3.4 Multiple Rain and Leaf Wipe Events 

Grass leaves, each contaminated with a single 3 pL droplet of VX, were subjected 
to multiple (lOx) 100 pL rain events at 0.017, 0.5, 1, 4, and 24 h after dissemination. The total 
cumulative amount of VX washed from the leaf surface at 0.017 h after dissemination was 
2.266 mg (Table 7), which represents approximately 75.3% of the VX originally added to the 
surface of the leaf. At 0.5 h after dissemination, the amount of VX removed from the leaf surface 
by the consecutive rainfall events was approximately 1 order of magnitude less than that 
removed at 0.017 h after dissemination (Table 7). The cumulative amount of VX recovered from 
rain events at 0.017 h after dissemination was significantly greater ip < 0.05) than the 
corresponding recoveries at 0.5, 1, 4, and 24 h after dissemination. There was no significant 
difference (p < 0.05) in the cumulative amount of VX recovered from rain events across the 0.5, 
1, 4, and 24 h post-dissemination time points. The total cumulative amount of VX recovered 
from surface wipes after 10 consecutive 100 pL rain events at 0.5 h after dissemination was 


10 





significantly greater (p < 0.05) than the recovery from surface wipes at 0.017, 1, 4, and 24 h after 
dissemination. There were no significant differences (p > 0.05) in the cumulative amounts of VX 
recovered from surface wipes across the 0.017, 1, 4, and 24 h post-dissemination time points. In 
Figure 3, the cumulative mass of VX recovered from rain events was plotted to show incremental 
recovery in the wash-off of VX from multiple rain events. The estimated human percutaneous 
lethal dose for 50% of the population (LD50) level of 3 mg/70 kg soldier (Safety Data Sheet, 
2015) is included as a reference point. The cumulative amount of VX recovered from rainwater 
after consecutive 100 pL rain events on a grass leaf contaminated with a single 3 pL VX droplet 
was within the same order of magnitude as the estimated human percutaneous LD50. 


Table 7. Total VX Recovery from Multiple Rain Events (10 x 100 pL) 
and Surface Wipes from Contaminated Grass Leaves 


Time after 
Dissemination 

(h) 

VX Recovery from 
Multiple Rain 
Events 
(mg) (±SD) 

VX Recovery from 
Multiple Surface 
Wipes 
(mg) (±SD) 

Total 

VX Recovery 
(%) 

0.017 

2.266 (0.493) 

0.024 (0.013) 

76.1 

0.5 

0.385 (0.268) 

0.094 (0.050) 

15.9 

1 

0.030 (0.039) 

0.009 (0.006) 

1.29 

4 

0.012 (0.008) 

0.002 (0.0003) 

0.46 

24 

0.065 (0.039) 

0.019 (0.005) 

2.79 


SD: standard deviation (shown in parentheses). 


The cumulative mass of VX recovered from wipes after multiple rain events is 
shown in Figure 4. The cumulative recovery of VX from wipes at 0.5 h after dissemination was 
significantly greater (p < 0.05) than that at the 0.017, 1, 4, and 24 h post-dissemination time 
points. The cumulative amount of VX recovered from surface wipes applied to a leaf that was 
contaminated with a single 3 pL droplet was orders of magnitude below the estimated human 
percutaneous LD 50 of 3 mg. 


11 





0 2 4 6 8 10 


Rain Events 

*The amount of agent washed off the leaf at 0.017 h was significantly 
greater (p < 0.05) than that at any of the 0.5, 1, 4, and 24 h time points 
after dissemination. 

Figure 3. Rain induced wash-off from VX-contaminated grass leaves. Mass of VX recovered 
after multiple (lOx) rain events (100 pL each) from leaves that were contaminated with a single 
3 pL droplet of VX. The amount of VX in this 3 pL droplet, disseminated onto each leaf, 
corresponded to the estimated human percutaneous LD50 (Safety Data Sheet, 2015). 



Figure 4. Residual VX recovery from surface wipes. Cumulative mass of VX recovered from 
multiple wipes (10 separate, consecutive wipe events) after VX-contaminated leaves were 
subjected to 10 consecutive 100 pL rain events. 


12 









4. 


DISCUSSION 


We determined the wash-off coefficient (k w ) in separate experiments for 1 and 
3 pL droplets of VX on the grass species E. crus-galli using one 10 mm moderate rain event 
(Hunsche et al., 2007; Llasat, 2001) at several time points after dissemination. The k w results are 
critical data points needed for the PEARL model to predict the fate and transformation of 
compounds in a plant and soil system. Incorporation of the data collected in this study into the 
PEARL model will be discussed in subsequent reports. However, in producing k w values for the 
model, we observed several key characteristics that may be helpful for commanders of 
Warfighters in battlefield conditions when Go/No-Go decisions are critical under stressful 
situations. 


At 1 h after dissemination onto E. crus-galli leaves, VX becomes highly rainfast. 
(The designations are as follows: H is highly rainfast with <30% washoff, M is moderately 
rainfast with <50% washoff, and L is low rainfast with <70% washoff [Wise, 2015].) In other 
words, 83% (3 pL droplets) to 95% (1 pL droplets) of the VX was washed off grass leaves when 
a single 10 mm (0.39 in.) moderate rain event was applied at 0.017 h (1 min) after VX 
dissemination. At 1 h after dissemination, <1% of VX (1 or 3 pL droplets) was removed from 
grass leaves by rain displacement that was followed by a single surface wipe. 

We also conducted experiments on grass leaves that were each contaminated with 
a single 3 pL droplet of VX using multiple 100 pL rain events. Individually, these rain events 
represented light rainfalls, and cumulatively, they produced a moderate rainfall that ranged from 
30 to 60% of 10 mm (Llasat, 2001; Hunsche et al., 2007). The cumulative amount of VX washed 
off a grass leaf from 10 rain events at 0.017 h after dissemination was approximately 75.3% 
(2.266 mg of VX). The cumulative amount of CWA recovered from 10 consecutive surface 
wipes after the series of rain events was approximately 0.8% (0.024 mg of VX), which resulted 
in 76.1% VX recovery of the 3 pL droplets. The total amount of VX recovered from both the 
rain and wipe events at 1 h after dissemination was 1.3% (0.039 mg of VX). In our research, the 
average proportion of 3 pL VX droplets on grass removed by moderate rainfall (single and 
multiple rain events) at 0.017 h (1 min) after dissemination of VX was approximately 79.2%, 
which corresponds with field studies conducted decades ago, wherein grass sod was 
contaminated with VX then rinsed with water within minutes after dissemination to remove 
approximately 66% of the VX applied. 

These results indicate that moderate rainfall, occurring within 1 min after 
dissemination, should reduce the potential amount of VX immediately available for direct 
contact transfer of VX from grass leaf surfaces to the soldier. However, the wash-off from that 
foliage would contain high concentrations of VX, which poses additional types of hazards. 
Depending upon the rainwater pH and buffering capacity of the soil, the potential contact hazard 
could remain with contaminated rainwater and reside within soil. 


13 



5. 


CONCLUSION 


The time after dissemination of VX on grass-like species (e.g., E. crus-galli) can 
be a decisive factor for reducing the threat of hazard to the Warfighter. Within minutes after 
dissemination of VX onto grass, the majority of the VX (approximately 75-95%) can be 
removed by wash-off from a moderate rainfall (>0.6 mm) or with excess water. Based on these 
results, wash-off can reduce the potential amount of VX immediately available for direct 
exposure of the soldier to VX on the surface of contaminated foliage; however, the hazard is then 
transferred to water and soil. 

When the time interval after dissemination of VX is known to be at least 1 h, the 
proportion of VX remaining immediately accessible on the external surface of grass (i.e., as 
estimated by measured VX wash-off) is greatly reduced to approximately 1 % of that 
disseminated (based on results for 1 or 3 pL droplets of VX with recoveries of 
0.8278 ± 0.0386 mg or 3.010 ± 0.137 mg of VX, respectively). 


14 



LITERATURE CITED 


ASTM International. Standard Specification for Reagent Water, ASTM D1193-99; ASTM 
International: West Conshohocken, PA, 2004. 

http://www.astm.org/DATABASE.CART/HISTORICAL/Dl 193-99.htm (accessed December 
2015). 

Epstein, J.; Callahan, J.J.; Bauer, V.E. The Kinetics and Mechanisms of Hydrolysis of 
Phosphonothiolates in Dilute Aqueous Solution. Phosphorus 1974, 4, 157-163. 

Hulbert, D.; Isaacs, R.; Vandervoort, C.; Wise, J. Rainfastness and Residual Activity of 
Insecticides to Control Japanese Beetle (Coleoptera: Scarabaeidae) in Grapes. J. Econ. Entomol. 
2011 ,104 (5), 1656-1664. 

Hunsche, M.; Scherhag, H.; Schmitz-Eiberger, M.; Noga, G. Influence of Rain Intensity and 
Rapeseed Oil Ethoxylate Adjuvants on Biological Efficacy of Glyphosate J. Plant Dis. Prot. 
2007, 114(4), 176-182. 

Leistra, M.; Van der Linden, A.; Boesten, J.; Tiktak, A.; Van den Berg, F. PEARL Model for 
Pesticide Behavior and Emissions in Soil-Plant Systems; Descriptions of the Processes in 
FOCUS PEARL v l.l.L; Alterra-rapport 013; Alterra, Green World Research: Wageningen, The 
Netherlands, 2001. 

Lethal Nerve Agent (VX); Safety Data Sheet; U.S. Army Edgewood Chemical Biological Center: 
Aberdeen Proving Ground, MD, 2015. 

Llasat, M.C. An Objective Classification of Rainfall Events on the Basis of Their Convective 
Features: Application to Rainfall Intensity in the Northeast of Spain. Int. J. Climatol. 2001, 21, 
1385-1400. 

Rothamsted Research. Pesticide Chemistry. http://www.rothamsted.ac.uk/bch/PCGroup/ 
translocation.html (accessed April 2013). 

Sanyal, D.; Bhowmil, P.C.; Reddy, K.N. Influence of Leaf Surface Micromorphology, Wax 
Content, and Surfactant on Primisulfuron Droplet Spread on Barnyardgrass ( Echinochloa crus- 
galli) and Green Foxtail ( Setaria viridis ). Weed Science 2006, 54 (4), 627-633. 

Simini, M.; Checkai, R.T.; Haley, M.V. Visaed Characterization ofVX Droplets on Plant 
Foliage', ECBC-TR-1393; U.S. Army Edgewood Chemical Biological Center: Aberdeen Proving 
Ground, MD, 2016; UNCLASSIFIED Report. 

Talmage, S.S.; Munro, N.B.; Watson, A.P.; King, J.F.; Hauschild, V. The Fate of Chemical 
Warfare Agents in the Environment. In Chemical Warfare Agents: Toxicology and Treatment, 

2 nd ed.; Marrs, T.C., Maynard, R.L., Sidell, F.R., Eds.; John Wiley and Sons, Ltd.: Hoboken, NJ, 
2007. 


15 



Test Operations Procedure (TOP). Chemical and Biological Contamination Survivability 
(CBCS), Large Item Exteriors', TOP 08-2-510A; U.S. Army Dugway Proving Ground: Dugway, 
UT, 2011; UNCLASSIFIED Procedure. 

U.S. Department of Agriculture (USDA), Natural Resource Conservation Service (NRCS); The 
PLANTS Database; National Plant Data Team: Greensboro, NC, 2015. http://plants.usda.gov 
(accessed March 2015). 

Van den Berg, F.; Leistra, M. Improvements of the Model Concept for Volatilisation of 
Pesticides from Soils and Plant Surfaces in PEARL, Description and User’s Guide for PEARL 
2.1.1-C1, Version 1.1\ Alterra-rapport; Alterra, Green World Research: Wageningen, The 
Netherlands, 2004. 

Van Emon, J.M.; Gerlach, C.L.; Reed, A.W.; Hardwick, B.C. Foliar Dislodgeable Residue 
Analysis: A New Scientific Approach to a Regulatory Concern. Food Tech. Biotechnol. 1998, 36 
(2), 119-124. 

Wise, J. Rainfast Characteristics of Insecticides on Fruit, 2016, Michigan State University 
Extension, Department of Entomology, http://msue.anr.msu.edu/news/rainfast_characteristics_ 
of_insecticides_on_fruit (accessed June 2015). 


16 



ACRONYMS AND ABBREVIATIONS 


ANOVA 

analysis of variance 

CAS 

Chemical Abstract Service 

CASARM 

Chemical Agent Standard Analytical Reference Material 

CWA 

chemical warfare agent 

GC 

gas chromatography 

HPLC 

high-performance liquid chromatography 

IPA 

isopropyl alcohol 

kw 

wash-off coefficient 

LD 50 

lethal dose for 50% of the population 

MRM 

multiple reaction monitoring 

MS/MS 

tandem mass spectrometry 

OP 

organophosphate 

P 

probability 

PEARL 

Pesticide Emission Assessment at Regional and Local scales 

r 2 

coefficient of determination 

SD 

standard deviation 

VX 

0-cthyl-S-(2-diisopropylaminocthyl) methyl phosphonothiolate 

VX-J 5 

VX isotope internal standard 


17 




DISTRIBUTION LIST 


The following individuals and organizations were provided with one Adobe 
portable document format (pdf) electronic version of this report: 


U.S. Army Edgewood Chemical 
Biological Center (ECBC) 
Molecular Toxicology Branch 
RDCB-DRT-M 
ATTN: Haley, M. 

Kristovich, R. 

Checkai, R. 

Simini, M. 


ECBC Analytical Toxicology Branch 

RDCB-DRT-T 

ATTN: Lawrence, R. 

Busch, M. 


Defense Threat Reduction Agency 
J9-CBS 

ATTN: Graziano, A. 


Department of Homeland Security 

DHS-ORD-CSAC 

ATTN: Famini, G. 


G-3 History Office 
U.S. Army RDECOM 
ATTN: Smart, J. 


ECBC Technical Library 
RDCB-DRB-BL 
ATTN: Foppiano, S. 
Stein, J. 


Office of the Chief Counsel 

AMSRD-CC 

ATTN: Upchurch, V. 


Defense Technical Information Center 
ATTN: DTIC OA 


ECBC Rock Island 
RDCB-DES 
ATTN: Lee, K.