EPA/600/R-03/003
                                                                   September 2004
                                   Final Report



A STUDY ON THE ACCUMULATION OF PERCHLORATE IN YOUNG HEAD LETTUCE

                                      By:
                             Stacy Lewis Hutchinson*
                           Ecosystems Research Division
                                Athens, GA 30605
                      * Current Address: Kansas State University
                 Department of Biological and Agricultural Engineering
                                  147SeatonHall
                               Manhattan, KS 66506
                       National Exposure Research Laboratory
                         Office of Research and Development
                        U.S. Environmental Protection Agency
                         Research Triangle Park, NC 27711

-------
                                   DISCLAIMER

       The information in this document has been funded by the United States Air Force under
IAG #57938313-01-0 to the United States Environmental Protection Agency.  It has been
subjected to the Agency's peer and administrative review and has been approved for publication
as an EPA document. Mention of trade names or commercial products does not constitute
endorsement or recommendation for use.

-------
                                       PREFACE
       The Colorado River is contaminated with perchlorate at low levels (5-9 parts-per-billion,
ppb).  Much of the lettuce consumed in the winter months in the U.S. is irrigated by Colorado
River water.  Results from 5,650 drinking water sources in California show perchlorate
detections in only 319 sources above the reporting level of 4 [ig/L (parts-per-billion, ppb).
However, perchlorate levels of up to 260 ppb were detected in wells near weapons
manufacturing facilities in Sacramento and Los Angeles counties. Perchlorate has also been
detected at levels of 17 ppb in Lake Mead as a result of releases from two ammonium
perchlorate manufacturing facilities in Nevada. The primary sources of perchlorate
contamination appear to be from industrial  and military operations that use perchlorate as an
oxidizing agent. Perchlorate contamination in water is of concern because of uncertainties about
toxicity and health effects from low levels in drinking water sources, the impact on ecosystems,
and possible indirect exposure pathways to humans from agricultural and other activities. Anion
exchange resins, microbial-mediated reduction, and phytoremediation are under investigation as
ways to remove perchlorate from contaminated waters.

       Phytoremediation is the use of plants to remove both ppb and parts-per-million (ppm)
levels of organic and inorganic pollutants from contaminated soil and water. Since 1999
research into the ability of terrestrial and aquatic plants to degrade or accumulate perchlorate has
been reported by several groups including EPA/NERL-Athens. The Athens researchers observed
accumulation of perchlorate in aquatic species such as blue-hyssop and parrot-feather that
suggested that leafy vegetables such as cabbage and lettuce might accumulate perchlorate.  For
these reasons, potential accumulation of perchlorate in lettuce leaves was identified as one of six
high-priority research needs at the U.S. Air Force's (USAF) Little Rock Eco Summit in April
1999.  This study was part of the work plan of an interagency agreement between the USAF and
EPA to investigate the fate of, and potential exposures to, perchlorate.  This greenhouse study
was designed as a narrow screening-level test to determine the degree of perchlorate uptake
(from fortified nutrient solution) and subsequent accumulation in lettuce leaves.

       The report was submitted for review, consistent with a level 2 EPA product.  Specifically,
this report underwent five external technical reviews, one internal editorial review, and one
internal QA review. Overall, most reviewer comments were calls for more information and
details. Incorporating the requested information produced this modified report that more clearly
communicates the important finding that lettuce accumulates perchlorate  from fortified nutrient
solution.  However,  follow-up studies will be required before this potential perchlorate exposure
route can be fully characterized. This successful demonstration of the uptake and accumulation
of perchlorate by lettuce in the greenhouse  study is information that other researchers can use in
further research on uptake of perchlorate by lettuce grown under field conditions. EPA/NERL-
Athens concluded research on perchlorate in June 2002 and does not plan any further research on
perchlorate.

                                               Rosemarie C. Russo, Ph.D.
                                               Director
                                               Ecosystems Research Division
                                               Athens, Georgia
                                           in

-------
                               TABLE OF CONTENTS
                                                                                Page
Disclaimer	ii
Preface	iii
Table of Contents	v
List of Figures	vi
List of Tables	vii
Acknowledgements	viii
Introduction	1
Experimental Methods	1
Results and Discussion	3
Summary and Conclusions	5
References	6
                                          IV

-------
                                  LIST OF FIGURES
                                                                             Page

Figure 1. Lettuce dry weights over time. Top graph shows dry mass over time over time,
bottom graph shows dry weight linear regression lines (n=3). The plants
were repotted on day-51	9

Figure 2. Concentration of perchlorate in dry leaf tissue over
time.  Error bars represent standard deviation among triplicate samples (n=3)	10

Figure 3. Mass of perchl orate extracted from above-ground lettuce tissue as
compared to the mass of perchlorate added	11

-------
                                   LIST OF TABLES
                                                                                 Page

Table 1.  Cumulative volume of perchlorate-fortified water, total mass of
perchlorate added, and total mass of perchlorate recovered at each sample
point (n = 3). Data are also shown in Figure 2	12

Table 2.  The mass of perchl orate recovered from each treatment based on best-fit linear
regression lines through the entire data set (days 14-95, Figure 3). Coefficient of
determination is for comparison to the significant values of r2 at p=0.01 (r2 s;g = 0.708)
for n=10 error degrees of freedom	13

Table 3.  The mass of perchlorate recovered from each treatment at the final
takedown (day-95) based on the average data for the takedown period (n = 3)	13

Table 4.  Wet plant concentrations, mass of perchlorate recovered
from plant tissue, and the resulting concentration factor for the outer leaves
and inner head of lettuce at the final takedown (day-95, n = 3)	14
                                           VI

-------
                              ACKNOWLEDGEMENTS

       The U.S. Air Force (IAGRW57938313-01-0) provided funding for this research.  The
author would like to acknowledge Dr. John Evans' help with the development of the analytical
procedures, and Dr. Evans and Dr. Jackson Ellington for assistance with data analysis. In
addition, the author wishes to thank Dr. Boyd Turner, Ms. Linda Smith, Mr. Alan Sealock and
Ms. Georgia Wood for help in maintaining the plants and preparing samples for analysis.
                                          Vll

-------
1.0    INTRODUCTION

       Perchlorate (C1O4") releases have been confirmed in 20 states throughout the United
States. The majority of the releases are in California, Nevada, Arizona, and Texas1'2. In
California, detections have been primarily in groundwater sources in the counties of Los
Angeles, San Bernardino, and Riverside, as well as sources containing water from the Colorado
River3. Results from 5,650 drinking water sources in California show perchlorate detections in
only 319 sources above the reporting level  of 4 (ig/L (parts-per-billion, ppb). Perchlorate levels
of up to 260 ppb were detected in wells near weapons manufacturing facilities in Sacramento and
Los Angeles counties4. Perchlorate has also been detected at low levels (5-9 ppb) in the
Colorado River5 and up to 17 ppb in Lake Mead as a result of releases from two ammonium
perchlorate manufacturing facilities in Nevada2.

       The primary sources of perchlorate contamination appear to be from industrial and
military operations that use perchlorate as an oxidizing agent6. Perchlorate is water soluble,
exceedingly mobile in aqueous systems, and can persist for many decades under typical ground
and surface water conditions6.  Perchlorate contamination in water is of concern because of
uncertainties about toxicity and health effects from low levels in drinking water sources, the
impact on ecosystems,  and possible indirect exposure pathways to humans from agricultural and
other activities1. Anion exchange resins, microbial-mediated reduction, and phytoremediation
are under investigation as ways to remove perchlorate from contaminated waters1'5'7.

       Phytoremediation is the use of plants to remove both ppb and parts-per-million (ppm)
levels of organic and inorganic pollutants from contaminated soil and water. The concentrations
ppb and ppm differ by a factor of 1,000 i.e., 1 ppm = 1,000 ppb. Since 1999 research into the
ability of terrestrial and aquatic plants to degrade or accumulate perchlorate has been reported by
several groups7"11.  A large part of the national supply of winter fruits and vegetables, including
lettuce, are grown in southern California and Arizona and  are irrigated with Colorado River
water12.  Currently, there are very limited data about the possible uptake of perchlorate into
agricultural products caused by irrigation with low ppb levels of perchlorate-contaminated
water.  Accumulation of perchlorate in aquatic species  such as blue-hyssop and parrot-feather7
suggested that leafy vegetables such as cabbage and lettuce might accumulate perchlorate.

       The overall objective of this study therefore was to demonstrate in a greenhouse study
the potential for incorporation of perchlorate from  aqueous solutions of 10, 50,  100, 500, 1,000,
5,000, and 10,000 ppb into an agricultural food crop (lettuce; Lactuca sativa\ which is typically
grown under irrigated conditions. A sand matrix amended with water containing known amounts
of perchlorate was used as the growth medium to accentuate uptake.  The successful
demonstration of the uptake and accumulation of perchlorate by lettuce in the greenhouse study
was seen as information that other researchers could use in further research on uptake of
perchlorate by lettuce grown under field conditions.

-------
2.0    EXPERIMENTAL METHODS
       The growing conditions described in the standard method for conducting seedling growth
tests (ASTM E1598-94)13 were used as a guideline for plant growth throughout the study.
Lettuce plants (Lactuca sativa) were grown from seed (Burpee's Iceberg Crisphead, Packaged
for 2000 Lot 1, W. Atlee Burpee & Co., Warminter, PA, 90-120 days to maturity) in conical
plastic containers, 14 cm deep and 3.8 cm in diameter at the top. The conical containers held
approximately 135 g washed sand and were used for the first 51 days of the experiment. On day-
51 the plants were repotted, without disrupting the root ball, by transferring the plant and the
sand from the smaller containers into 10 cm x 10 cm x 8 cm plastic containers with
approximately 550 g of additional washed sand for expanded root growth. In both containers the
sand was within 2 cm of the top of the container. The bottoms of both types of containers were
lined with glass wool to prevent loss of sand from the containers.

       The plants were grown in a greenhouse in Athens, GA from the last week in February
until the first week in June 2000.  Germination (appearance of the first leaf above the sand
surface) occurred at day-7. The day-95  samples were collected 95 days after germination for a
total study duration of 102 days from seeding. The greenhouse was not temperature controlled,
but was equipped with an electric fan that was installed opposite a screen door to provide cross-
ventilation airflow. The fan was controlled by a thermostat set to activate the fan at 29.5 °C (85
°F).  Additionally, the greenhouse was equipped with fluorescent grow lights that were operated
14 hrs per day throughout the study to enhance light intensity and increase the photoperiod.

       Children's play box sand was purchased locally and washed with tap water until the wash
water was clear.  A sample of the last wash water was analyzed for perchlorate by ion
chromatography.

       Plant nutrient solution was prepared from Peter's Professional plant food (20-20-20,
1.9% nitrate nitrogen) purchased locally in Athens, GA. The nutrient solution was prepared per
directions stated on the bag by adding 3.5 g grab sample of the solid to 1 L of  18 MQ water and
mixing thoroughly. The nutrient solution was prepared as needed. Before the  start of the study,
the first preparation of nutrient solution was analyzed for perchl orate before application to the
plants. The plants were fertilized once per week throughout the study with 5 mL of the nutrient
solution by slowly releasing the nutrient solution from a pipette into the sand at the base of the
plant.

       Seeds were germinated with application of water as needed to keep the  sand moist and
once per week 5 mL of the nutrient solution was added to all the plants; beginning when the
seedlings were 14 days old, plants were also watered with 10, 50, 100, 500, 1,000, 5,000 and
10,000 ppb solutions  of perchl orate.  The treatment solutions were made one time  by weighing
the calculated amount of solid sodium perchlorate (Fisher Scientific) into a 500-mL volumetric
flask and bringing to volume with 18  MQ water. Nutrient and perchlorate treatment solutions
were added at a rate to prevent dripping from the bottom of the containers. Perchlorate treatment

-------
solution was applied to the containers three times per week at a maximum of 10 mL/day.  To
maintain plant health, supplemental watering was done with tap water on days during the latter
stages of growth when no perchlorate solution was added. The final perchlorate treatment was
day-93.

       Based on the analysis of three lettuce plants at seven perchl orate treatment levels plus a
control for 12 sampling dates, the total number of plants sampled during the study was 288. In
the greenhouse, the plants for each treatment level and control were grouped together in a
shallow container and weekly the groups were rotated to different locations in the greenhouse.
During the study, the plants were spaced such that leaves of adjacent plants did not touch to
allow air circulation. On designated days during the study, the perchl orate content of the leaves
and roots at each level of treatment was measured in three separate lettuce plants.  The first
sampling event occurred on day-21 from seeding (14 days after germination) with the final day-
95 sampling occurring 102 days from seeding.  Three control plants were also analyzed at each
sampling event. Each lettuce plant was separated into above (leaf)- and below (root)-ground
biomass before analysis. The day-86 and day-95 samples had small heads that were separated
from the outer leaves and the inner (head) and outer leaves were analyzed separately.

       A published method14 was modified and used for the extraction of plant tissue and for
instrumental analysis of an aliquot of water from the last sand wash, the nutrient solutions, and
the aqueous extracts of the lettuce leaves and roots. One modification entailed oven drying and
pulverizing the dried plant tissue rather than freeze-drying and grinding.  Additionally, the
analytical  column was 4 mm  internal diameter (ID) rather than 2 mm and the injected volume
was 100 |jL rather than 1,000 |jL.  The fresh plant tissue was weighed, washed with 18 MQ
water, and dried at 104°C for 24 hours in uncapped glass vials. The tissue dry weight and
percent moisture were recorded for each sample.  Perchlorate was extracted  from the dry  and
pulverized plant material with 18 MQ water at an approximate mass to volume ratio of 0.6 g to
30 mL  depending  on the whole plant dry mass. Water (18 MQ) was added and the vials were
capped and placed in a boiling water bath for 30 minutes.  The vials were cooled and stored at
4°C for 24 hrs.  The aqueous  extract was filtered through 1 layer of Kimwipes and the filtrate
was centrifuged at 20,000 x g for 30 min to remove any residual plant tissue. Organic acids and
interfering ions were removed from the supernatant by adding 0.5 g of DD-6 alumina per mL of
extract. The extract was allowed to remain on the DD-6 for 24 hr at 4°C.  An aliquot was
removed, filtered through a 0.4 micron Acrodisc® filter, and analyzed for perchlorate using an
isocratic ion chromatographic (1C) procedure. The sand and glass wool in the plant containers
were discarded and not analyzed for perchlorate.

       A Dionex Ion Chromatograph (1C) equipped with a GP40 gradient pump, AD20
absorbance detector, CD20 conductivity detector, AS3500 autosampler, and LC20
chromatography enclosure was used for  analysis of perchlorate in the tissue  and water samples.
Ion analysis was performed with an lonpac AS 16-HC (4-mm X 250 mm) analytical column. A
guard column preceded the analytical column to prevent sample contaminants from eluting onto
the analytical column.  The column flow rate of eluent (sodium hydroxide 50 mM) was 1.0 mL
min"1. The injection loop volume was 100 jiL, and the run time for perchlorate analysis was 20
min. An anion self-regenerating suppressor (ASRS) was used for suppressed-conductivity
detection.  Distilled and deionized water was used for regeneration of the ASRS. Standard

-------
curves were calculated from injection of 500 ppb to 25,000 ppb calibration standards. Based on
injection of 100 |j,L injections, the instrument detection limit was 300 ppb. The estimated
method detection limit was 500 ppb (25 ppb on a dry weight basis) based on the quantification of
the extracts of perchlorate-fortified lettuce.

      The first nutrient solution prepared from each bag of Peter's plant food and a sample of
the water collected from the final washing of the sand were analyzed on this 1C after filtration
through a 0. 45 (Dm filter. After detection of perchlorate in the control samples midway through
the study, the most recently prepared nutrient solution and the sand washing sample were
analyzed on a second 1C with a lower limit of detection (see Results and Discussion).

      At each sampling event, three plants of equal size were removed from the green house
and analyzed for perchlorate. The remaining plants were rearranged to maintain equal spacing
and airflow. The plant dry mass and perchlorate concentration means and standard deviations
from the analysis of the three plants at each sampling event were used to plot plant accumulation
of dry mass over time as well as total perchlorate accumulation in the roots and leaves of the
entire plant at each sampling event.

3.0   RESULTS AND DISCUSSION

      The photoperiod, temperature, water, and nutrient conditions to which the plants were
exposed in  the greenhouse were sufficient to maintain steady biomass  accumulation (Figure 1).
The greenhouse was not heated. The lowest recorded temperature during the study was 10°C and
the highest 35°C.  When the ambient temperature in the greenhouse reached 29.5 °C  (85 °F) a
fan was  automatically activated to circulate air and to moderate the temperature. Based on
information supplied with the seed, the time to maturity for the Crisphead lettuce is 90-120 days.
The day-95 plants had well defined heads and the typical green color of lettuce. Dry mass
accumulation of the lettuce with respect to days from seeding is shown in Figure 1 starting from
day-21.  It is evident from Figure 1 that the biomass accumulation of all the plants was similar.
Exposure to perchlorate, even at 10,000 ppb, did not affect plant growth and there was no visible
or textural difference in thelO,000 ppb plants and the control plants.

      For consistency in lettuce sample size, at each sampling event three plants  of similar size
were chosen from each treatment level and control  plants. The drop in biomass for the day-72
and day-86 samples could be attributed to the fact that at this time in the study the overall
number  of plants at each treatment level was reduced to a level such that only smaller plants of
similar size remained. The last perchlorate treatment was on day-93.  The almost doubling in
plant mass  between day-86 and day-95 samples was attributed to the supplemental water that
was added to the plants daily for the latter days of the study.  This additional water, plus
continued root expansion, possibly allowed the roots to grow and transpire nutrients that had
been deposited out of the root zone during the previous wetting/drying events.  Since perchlorate
would migrate with the nutrients, deposition of perchlorate outside the root zone is one
explanation for the less than 100 % recovery of perchlorate reported in Table 1.

      Even though the water from the last sand washing and each weekly nutrient solution were
tested for perchlorate, perchlorate was observed in the control plants in the day-35 samples.  The

-------
nutrient solution that had been prepared for the post day-35 plants and the sand-washing sample
were reanalyzed on a second 1C with a lower detection limit (1-4 ppb). The tap water sample
from the sand washing did not contain perchlorate; this fact also indicated the tap water used to
water the plants in the latter stages of the study did not contain perchlorate. However, the Peter's
nutrient solution was found to contain 100 ppb perchlorate.  The addition of perchlorate from the
nutrient solution was considered to be constant throughout the study and included in the
calculation of the total mass of perchlorate added to the plant at each perchlorate treatment level
(Table 1). As can be seen in Table 1, the perchlorate added via Peter's nutrient solution
contributed a small fraction of the total amount of perchlorate added to treatments 1,000, 5,000,
and 10,000 ppb perchlorate.  However, in the lower concentrations (100 to 500 ppb), the amount
of perchlorate added in the Peter's solution ranged from approximately 30% in the 100 ppb
treatment to around 6% of the 500 ppb treatment.  The 10 and 50 ppb treatment level data were
not reported due to the high percentage of perchlorate added by the nutrient solution.

       Post day-35 preparations of nutrient solution were not analyzed for perchlorate, but the
solutions were prepared from the same bag as was used to prepare the nutrient solution that
contained 100 ppb perchlorate. Since each preparation of nutrient solution was essentially a grab
sample from a heterogeneous mixture of solid ingredients the perchlorate concentrations of the
nutrient solutions may not have been constant at 100 ppb. Thus more or less than the calculated
amount of perchlorate may have been added than was accounted for based on the single
measurement.  This is evident in Table 1 where the mass of perchlorate recovered from the
control plants on days 72, 86, and 95 was much greater than the corresponding calculated
amount.

       In the 100 ppb and higher treatment samples, measurable perchlorate was observed in the
above-ground biomass (leaves) on day-21 (Figure 2 and Table 1). The first lettuce samples were
collected and analyzed for perchlorate seven days after the initial perchlorate treatments. In
Figure 2 the perchlorate concentration in the leaf dry mass appeared to be concentration
dependent and increased steadily with time over the first 6 to 7 weeks of growth. Also, in Figure
2, the decline in the perchlorate concentration after day 51 may be related to the repotting, which
occurred at this time. As seen in Figure 1, the plant biomass increased at a steeper rate for the
day-58 and day-65 samples; the increased rate of biomass is reflected in the decline in the Figure
2 day-58 and day-65 perchlorate concentration data.

       A second sudden increase in plant biomass is evident in Figure 1 between the day-86 and
day-95 samplings.  Because of high daily temperatures experienced in the latter stages of the
study, the volume of additional non-contaminated water added each week increased but was
added at a rate that did not cause dripping from the bottom of the container. The amount of
perchlorate-amended water was kept at 10 mL per application, 3 days per week to ensure no loss
of perchlorate due to leaching.  The additional water enhanced plant growth and yielded an
increased level of dry biomass in the day-95 samples. Possibly the growth spurt was caused by
the extra water solubilizing accumulated nutrient solution and making it available to the roots.
The increase in the day-95 dry biomass diluted the accumulated perchlorate, compared to the
day-86 concentrations, as shown in the Figure 2 dry biomass concentration of perchlorate in the
day-95 samples.  In Figure 2, the concentrations ranged from a maximum of 3,600,000 ppb dry
plant material (480,000 ppb wet plant material) with the addition of 10,000 ppb perchlorate

-------
solution to a low accumulation of 90,000 ppb dry plant material (5,000 ppb wet plant material) in
the 100 ppb treatment.

       In Figure 3 total perchlorate extracted from the plant tissue was plotted against total
perchlorate added in solution (data also in Table 1). From this graph, it is obvious that
perchlorate uptake was continuous throughout the plant growth cycle. The calculated percent
recovery of perchlorate from each treatment based on best-fit linear regression lines through the
entire data set (days 14-95, Figure 3) is reported in Table 2. The coefficient of determination
was included as an indicator of best-fit line.  At treatment levels of 500 ppb to 10,000 ppb, the
amount of uptake into the above-ground biomass in Table 2 accounts for 73 to 82% of the
applied perchlorate.  This would indicate that perchlorate was carried with the transpiration
stream and was potentially 100% translocated at these concentrations.

       Percent recovery from the final takedown (day 95) is listed in Table 3.  Perchlorate
recovered from treatments 500-10,000 ppb was similar to the calculated values reported in Table
2 and ranged from 66 to  90% of the applied perchlorate.  The perchlorate content determined in
the roots of the 10,000 ppb treatment on day 95 was 1.3% (30 |j,g) of the applied amount.  The 30
|j,g of perchlorate recovered in the 10,000 ppb day-95 samples was the largest amount recovered
in roots in any samples during the study.  At all the treatment levels, the perchlorate not
accounted for in the above ground biomass and the roots was assumed to be in the sand/glass
wool that remained in the container when the plant was removed. The method14 used for  the
extraction of perchlorate has been shown to quantitatively recover perchlorate from lettuce
tissue. The boiling water bath extraction does not degrade perchlorate but disrupts cell walls and
liberates bound perchlorate.

       By the final takedown (day 95 from planting), the plant was large enough to separate the
older, outer leaves from the inner leaves, which had formed a small lettuce head. These small
lettuce heads (inner leaves) were  analyzed separately from the older outer leaves. The inner head
and outer leaf perchlorate concentrations and concentration factors (CFs) are shown in Table 4.
The CFs were calculated by dividing the wet plant concentration by the treatment concentration.
For the four treatment levels the wet plant concentration (ppb) of perchlorate in the outer  leaves
was from 2 to 6 times higher than the inner leaves. These data indicate that more perchlorate is
accumulated in the older, outer leaves of the lettuce plant with lower concentrations in the newly
formed inner leaves (head).  Because the majority of water transpired by lettuce during head
formation and growth is  through the exposed outer leaf material, these results strongly suggest
that perchlorate moves with the transpiration stream and the mature inner head will contain
substantially less perchlorate.  Specifically, these data show that the inner lettuce head that was
not exposed to light, and also transpired substantially less water, resulted in concentration factors
that were 68% to 89 % less than the outer exposed leaf material, depending on treatment level.
4.0    SUMMARY AND CONCLUSIONS

       The irrigation study indicates that perchlorate was transported from the root zone and
incorporated into lettuce leaf tissue. The recovery values for the 500 ppb and higher treatments
were approximately 79% (average of Table 3 recovery values); this suggests the majority of the

-------
perchl orate was incorporated into plant tissue and remained there until the end of the study.  The
unaccounted for 21% possibly remained in the sand, which was not analyzed after removal of the
intact plant; the lettuce tissue extracts were not analyzed for products of degradation.  Another
possibility is that because of the weekly additions of nutrient solution nutrient anions
accumulated and competed with perchlorate for uptake.  The study also suggested that the uptake
into lettuce was related to mass transport into the plant as driven by transpiration. Passive mass
transport uptake of perchlorate was reported in a recent study that showed perchlorate was
quantitatively transported via the transpiration stream to tobacco leaf from hydroponics nutrient
solution15.

       The data from this bench-scale greenhouse study indicate that perchlorate is accumulated
in the older, outer leaves of the lettuce plant, with significantly lower concentrations in the newly
formed inner leaves (head). The CFs derived by calculating the wet plant concentration by the
treatment concentration are 17-28 for the outer leaves, and 3-9 in the emerging head. The
greenhouse conditions in this study did not mimic field conditions and the perchlorate-fortified
treatment solution was added in a way designed to maximize uptake. Thus, the accumulation of
perchlorate was possibly enhanced over what would be accumulated in lettuce grown in the field
and irrigated with perchlorate-contaminated water.

       Recent reports have shown that perchlorate does not appreciably sorb to soils and that its
mobility and fate in surface and groundwater are largely influenced by the flow of the water and
the presence or absence of organisms that degrade perchlorate1'5'16. Another study showed that
perchlorate that was present as a natural constituent in the applied fertilizer accumulated in leaf
of tobacco grown under field conditions17. These recent reports support the potential for lettuce
uptake of perchlorate grown under field conditions, but follow-up studies are required before one
can fully characterize this exposure route in the above-ground vegetation.

-------
REFERENCES

  1.    EPA Groundwater and Drinking Water: Perchlorate
        August 4, 2003.
  2.    Urbansky, E.T. (2000) Perchlorate in the Environment, Kluwer Academic/Plenum
       Publishers: New York, pp. 177,189.
  3.    Perchlorate in California Drinking Water: Monitoring Update
        August 4,
       2003.
  4.    Office of Environmental Health Hazard Assessment: Proposed Public Health Goal for
       Perchlorate  August 4,
       2003.
  5.    Herman, D.C. and W.T. Frankenberger, Jr.  (1998) Microbial-mediated reduction of
       perchlorate in groundwater, Journal of Environmental Quality, 27, 750-754.
  6.    Urbansky, E.T. (2000) Quantitation of perchlorate ion: practices and advances applied to
       the analysis of common matrices. Critical Review sin Analytical Chemistry, 30, 311-343.
  7.    Susarla, S., S.T. Bacchus, S.C. McCutcheon, and N.L. Wolfe (1999) Potential species for
       Phytoremediation of perchlorate, EPA/600/R-99/069.
  8.    Nzengung, V.A. and C.H. Wang (1999) Influences on Phytoremediation of perchlorate-
       contaminated water, Abstracts of Papers of the American Chemical Society, 218:37-
       ENVR.
  9.    Susarla, S., S.T. Bacchus, G. Harvey, and S.C. McCutcheon (2000) Phytotransformation
       of perchlorate contaminated waters, Environmental Technology, 21, 1055-1065.
 10.    Van Aken, V. and J.L. Schnoor (2002) Evidence of perchlorate (C1O4" ) reduction in plant
       tissues (poplar tree) using radio-labeled (C1(V )-C-36, Environmental Science and
       Technology, 36, 2783-2788.
 11.    Nzengung, V.A., C.H. Wang, and G. Harvey (1999) Plant-mediated transformation of
       perchlorate into chloride, Environmental Science and Technology, 33, 1470-1478.
 12.    California River Water Users Association: Agriculture Uses
        August 4, 2003.
 13.    ASTM International (1994) Standard practice for conducting early seedling growth tests,
       ASTM E 1598-94, Philadelphia, PA.
 14.    Ellington, JJ. and JJ. Evans (2000) Determination of perchlorate at parts-per-billion
       levels in plants by ion chromatography, Journal of Chromatography A 898, 193-199.
 15.    Sundberg, S.E., JJ. Ellington, J. J. Evans, D. A. Keys, and J. W. Fisher (2003)
       Accumulation of perchlorate in tobacco plants: development of a plant kinetic model,
       Journal of Environmental Monitoring 5, 505-512.
 16.    Urbansky, E.T. and S. K. Brown (2003) Perchlorate retention and mobility in soils,
       Journal of Environmental Monitoring 5, 455-462.
 17.    Ellington, J.J., N.L. Wolfe,  A.W. Garrison, JJ. Evans, J.K. Avants, and Q. Teng (2001)
       Determination of perchlorate in tobacco plants and tobacco products, Environmental
       Science and Technology, 35, 3213-3218.

-------
   2500
   2000 -
 S 1500 -

Q
 *> 1000 -
    500 -
       • Control
       • 100 ppb
       • 500 ppb
       • 1,000 ppb
       • 5,000 ppb
       • 10,000 ppb
           21     28     35     38     42     45     51     58
                                         Days from Planting
                                                       65
                                         72
                                         86
                      95
    1800
    1600-
    1400-
 'SJ
  g 1200-
   Control y = 89.552x - 199.42
  100 ppb y = 73.804x-71.061
   500 ppb y = 71.381x- 139.39
 1,000 ppb y = 61.997x-52.727
 5,000 ppb y = 90.44 Ix - 189.03
10,000 ppb y = 81.846x-215
                  28
              35
38
42     45     51     58
   Days from Planting
65
72
86
95
Figure 1.  Lettuce dry weights over time. Top graph shows dry mass over time, bottom graph
shows dry weight linear regression lines (n=3).  The plants were repotted on day-51.

-------
   5000

   4500  -
S*
|  4000  -

3  3500
                 10,000 ppb
                 5,000 ppb
                 1,000 ppb
                 500 ppb
        10       20       30       40       50      60
                                       Days from Planting
                                                            70
80
90
100
Figure 2. Concentration of perchlorate in dry leaf tissue over time. Error bars represent standard
deviation among triplicate samples (n=3).
                                            10

-------
    200




    180  -




    160  -




    140  -
                O 500 ppb


                X 1,000 ppb
                         50
                                         100              150




                                   Total Perchlorate Added (jig)
 200
 250
   2000  -




   1800  -




   1600  -




!* 1400  -




1  1200  -
•4rf
c_


^  1000  -
I   800  -]

u
i.


^   600  -\
«
    400  -




    200  -
       0
               A 5,000 ppb



               D 10,000 ppb
                        500
                                        1000            1500



                                    Total Perchlorate Added (jig)
2000
2500
Figure 3.  Mass of perchlorate extracted from above-ground lettuce tissue as compared to the

mass of perchlorate added.
                                              11

-------
Table 1.  Cumulative volume of perchlorate-fortified water, total mass of perchlorate added, and
total mass of perchl orate recovered at each sample point (n = 3). Data is also shown in Figure 2.


Volume of
contaminated
water
15
30
50
65
75
90
100
120
140
170
200
230

Days
from
planting
21
28
35
38
42
45
51
58
65
72
86
95
Control
ng CICV
Added
in
nutrient
solution
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
7.0
Recovered
ND
ND
3.8
5.5
ND
2.0
3.7
6.5
5.9
13.0
18.0
28.0
Std
Dev
—
—
0.0
1.6
—
2.8
0.4
0.2
3.4
0.8
6.5
7.6
100 ppb Treatment
ng CICV
Added
as
treatment
1.5
3.0
5.0
6.5
7.5
9.0
10.0
12.0
14.0
17.0
20.0
23.0
Total
added
w/control
3.0
5.0
7.5
9.5
11.0
13.0
14.5
17.0
19.5
23.0
26.5
30.0
Recovered
TR
2
2
10
10
13
14
16
24
30
16
37
Std
Dev
0.5
0.4
0.0
1.1
1.6
2.8
1.0
1.0
1.2
1.3
1.5
3.5
500 ppb Treatment
ng cicv
Added
as
treatment
7.5
15.0
25.0
32.5
37.5
45.0
50.0
60.0
70.0
85.0
100.0
115.0
Total
added
w/control
9.0
17.0
27.5
35.5
41.0
49.0
54.5
65.0
75.5
91.0
106.5
122.0
Recovered
2
11
19
26
34
38
45
49
72
72
82
110
Std
Dev
1.7
0.5
0.9
6.2
3.3
6.3
3.7
8.9
14
15
17
1.8
ND = not detected
TR = Trace


Volume of
contaminated
water added
15
30
50
65
75
90
100
120
140
170
200
230

Days
from
planting
21
28
35
38
42
45
51
58
65
72
86
95
1,000 ppb Treatment
Ud CIO/
Added
as
treatment
15.0
30.0
50.0
65.0
75.0
90.0
100.0
120.0
140.0
170.0
200.0
230.0
Total
added
w/control
16.5
32.0
52.5
68.0
78.5
94.0
104.5
125.0
145.5
176.0
206.5
237.0
Recovered
0.9
23.
37
51
66
76
91
100
120
160
110
170
Std
Dev
1.5
1.8
0.1
16
6.1
3.2
1.3
0.4
25
3.4
8.3
45
5,000 ppb Treatment
ud CICV
Added
as
treatment
75.0
150.0
250.0
325.0
375.0
450.0
500.0
600.0
700.0
850.0
1000.0
1150.0
Total
added
w/control
76.5
152.0
252.5
328.0
378.5
454.0
504.5
605.0
705.5
856.0
1006.5
1157.0
Recovered
18
110
210
280
300
340
360
420
550
580
930
790
Std
Dev
22
11
25
1.0
27
6.4
83
17
20
7.2
130
140
10,000 ppb Treatment
Ud CIO/
Added
as
treatment
150.0
300.0
500.0
650.0
750.0
900.0
1000.0
1200.0
1400.0
1700.0
2000.0
2300.0
Total
added
w/control
151.5
302.0
502.5
653.0
753.5
904.0
1 004.5
1205.0
1 405.5
1706.0
2006.5
2307.0
Recovered
24
160
190
340
490
660
470
660
800
1300
1700
2000
Std
Dev
30
66
41
46
95
21
100
60
130
3.6
120
55
                                           12

-------
Table 2.  The mass of perchlorate recovered from each treatment based on best-fit linear
regression lines through the entire data set (days 14-95, Figure 3). Coefficient of
determination is included for comparison to the significant values of r2 at p=0.01 (r2 sig =
0.708) for n=10 error degrees of freedom.
Treatment (ppb)
500
1,000
5,000
10,000
Percent Recovered
82
74
76
73
r2
0.977
0.892
0.944
0.914
Table 3.  The mass of perchl orate recovered from each treatment at the final takedown
(day 95) based on the average data for the takedown period (n = 3).
Treatment
500
1,000
5,000
10,000
Perchlorate Added
(us)
122
237
1157
2307
Perchlorate
Recovered (ug)
110± 1.8
170 ±45
790 ± 140
2000 ±55
Percent Recovered
90 ±2
72 ± 19
68 ± 12
87 ±2
                                        13

-------
Table 4.  Wet plant concentrations, mass of perchlorate recovered from plant tissue, and
the resulting concentration factor for the outer leaves and inner head of lettuce at the final
takedown (day-95, n = 3).




Outer
Leaves

Inner
Leaves
(head)
Treatment Level
500 ppb

Wet
Plant
ppb

14,000

3,000
Std
Dev


4000

1000
CF.*


28

6
1,000 ppb

Wet
Plant
ppb

21,000

9,000
Std
Dev


2000

6000
CF.*


21

9
5,000 ppb

Wet
Plant
ppb

84,000

16,000
Std Dev


23,000

4,000
CF.*


17

O
10,000 ppb

Wet Plant
ppb

210,000

32,000
Std Dev


29,000

4,000
CF.*


21

3
* The concentration factor (CF) is the ratio of the wet plant concentration to the treatment
concentration.
                                         14

-------