PA/600/R-01/047
I States
mmental Protection
nyency . »
Office of Research and
Development
Cincinnati-, OH 45268
EPA/600/R-01/047
May 2001
www.epa.gov
Survey of Fertilizers and
Related Materials for
Perchlorate (CIO4~)
Final Report
I
-------
EPA/600/R-01/047
May 2001
Survey of Fertilizers and Related
Materials for Perchlorate (CIO4~)
Final Report
By
Edward Todd Urbansky
U.S. Environmental Protection Agency
Cincinnati, OH 45268
Timothy W.Collette
U.S. Environmental Protection Agency
Athens, GA 30605
Wayne P. Robarge
North Carolina State University
Raleigh, NC 27695-7619
William L. Hall
IMC-Global
Mulberry, FL 33860
James M. Skillen
The Fertilizer Institute
Washington, DC 20002
Peter F. Kane
Office of Indiana State Chemist and Seed Commisioner
West Lafayette, IN 47907-1154
This study was conducted in cooperation with North Carolina State University,
IMC-Global, The Fertilizer Institute, the Office of Indiana State Chemist and
Seed Commissioner, and the International Fertilizer Development Center.
National Risk Management Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, OH 45268
-------
Notice
The U.S. Environmental Protection Agency, through the National Risk Management Research
Laboratory of its Office of Research and Development, managed and collaborated in the research
described here. 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. In an effort to save paper and reduce printing
costs, this report is being issued by the EPA only as an Adobe Acrobat portable document format (pdf)
file. Adobe Acrobat Reader is available free of charge via the Adobe website.
-------
Foreword
The U.S. Environmental Protection Agency is charged by Congress with protecting the nation's
land, air, and water resources. Under a mandate of national environmental laws, the agency strives
to formulate and implement actions leading to a compatible balance between human activities and the
ability of natural systems to support and nurture life. To meet this mandate, EPA's research program
is providing data and technical support for solving environmental problems today and building a science
knowledge base necessary to manage our ecological resources wisely, understand how pollutants
affect our health, and prevent or reduce environmental risks in the future.
The National Risk Management Research Laboratory is the agency's center for investigation
of technological and management approaches for preventing and reducing risks from pollution that
threatens human health and the environment. The focus of the laboratory's research program is on
methods and their cost-effectiveness for prevention and control of pollution to air, land, water, and
subsurface resources; protection of water quality in public water systems; remediation of contaminated
sites, sediments and ground water; prevention and control of indoor air pollution; and restoration of
ecosystems. NRMRL collaborates with both public and private sector partners to foster technologies
that reduce the cost of compliance and to anticipate emerging problems. NRMRL's research provides
solutions to environmental problems by developing and promoting technologies that protect and improve
the environment; advancing scientific and engineering information to support regulatory and policy
decisions; and providing the technical support and information transfer to ensure implementation of
environmental regulations and strategies at the national, state, and community levels.
This publication has been produced as part of the laboratory's strategic long-term research
plan. It is published and made available by EPA's Office of Research and Development to assist the
user community and to link researchers with their clients.
E. Timothy Oppelt, Director
National Risk Management Research Laboratory
-------
Contents
Notice ii
Foreword iii
Abstract iv
Contents v
Tables and figures vi
1 Environmental perchlorate- what role for fertilizer? 1
1.1 Introduction 1
1.2 Fertilizer manufacture and use 1
1.2.1 Nutrient availability 1
1.2.2 Nitrogen sources 2
1.2.3 Phosphate sources 5
1.2.4 Potassium sources 5
1.2.5 Information sources 6
1.3 Previous fertilizer analysis studies 7
2 Survey of fertilizers and related materials 9
2.1 Objectives 9
2.2 Phase 1—Evaluation of laboratories 9
2.3 Phase 2—Analysis of samples 10
2.3.1 Sampling and analysis strategy 10
2.3.2 Results 11
2.3.3 Discussion 11
3 Implications for vascular plants 18
3.1 Introduction 18
3.2 Complicating factors 18
3.2.1 Chemical influences on ion transport 19
3.2.2 Concentration 19
3.2.3 Tissue-specific accumulation 20
3.2.4 Soil sorption 20
3.2.5 Summary 20
3.3 Difficulties in analysis 21
3.4 Implications of perchlorate absorption and accumulation 21
References 23
-------
Tables and Figures
Tables
1.1 Some common agricultural fertilizers used in production farming 2
1.2 Consumption (in tons) of nitrate salts in regions of the continental United States
for the year ending June 30,1998 4
1.3 Annual consumption/application (in tons) of some nitrogen fertilizers for several states 4
1.4 Annual production (tonnage) of nitrate salts by some manufacturers who
supplied the U.S. marketplace 5
2.1 Fertilizers and related materials surveyed for perchlorate 13
2.2 Summary results for perchlorate concentration detected by replicate
analyses of samples listed in Table 2.1 15
2.3 Tested fortifications (spikes) and recoveries in solutions of samples listed in
Tables 2.1-2.2 17
Figures
1.1 Schematic of urea production 2
1.2 Synthesis of nitric acid by the redox reaction of ammonia and atmospheric oxygen 4
1.3 Schematic of ammonium nitrate production 4
1.4 Operational schematic of phosphate rock processing 5
1.5 Bucketwheel used in mining phosphate rock 5
1.6 Agricultural DAP 5
1.7 The mineral sylvite (KCI) 6
1.8 Canadian potash mines can be kilometers below the surface 6
1.9 Sylvite and sylvinite can be mined by dissolving the minerals in the water and
pumping the brine to the surface 6
-------
Acknowledgments
We acknowledge the following individuals and institutions for their assistance in the completion of this project. This includes the services of
participating laboratories, state regulatory agencies, and industry employees.
Jay Johnson & Kent W. Richman
American Pacific Corporation
Cedar City, Utah
Stanley Kobata, Harry Welte & Paul Roos
California Department of Food and Agriculture
Sacramento, California
Thomas E. Carville, Carol F. Schexnaydor & Jesus I. Peralta
CF Industries
Donaldsonville, Louisiana, and Washington, D.C.
D.H. Thomas, P.E. Jackson & J.S. Rohrer
Dionex Corporation
Sunnyvale, California
Catherine A. Kelty, Stephanie K. Brown, Matthew L. Magnuson, J.
Jackson Ellington & John J. Evans
U.S. Environmental Protection Agency
Cincinnati, Ohio, and Athens, Georgia
David F. Graves & John B. Sargeant
Florida Department of Agriculture and Consumer Services
Tallahassee, Florida, and Lakeland, Florida
Steve Gamble
IMC-Kalium
Carlsbad, New Mexico
Linda Weber, David Gadsby, Bonnie Lingard,
Daphne Williams, Charles Kinsey, Julie Thompson,
Pam Burgess & Dennis Sebastian
IMC-Phosphates Environmental Laboratory
Bradley, Florida
George A. Kennedy, Celia G. Calvo, Bobby W. Biggars,
& Steven J. Van Kauwenbergh
International Fertilizer Development Center
Muscle Shoals, Alabama
Ali Haghani and Andrew Eaton
Montgomery Watson Laboratories
Pasadena, California
Tina Sack
New Mexico Department of Agriculture
Roswell, New Mexico
Michael Hancock
Office of Indiana State Chemist and Seed Commissioner
West Lafayette, Indiana
George W. Latimer
Office of Texas State Chemist
College Station, Texas
Guillermo Ramirez
North Carolina State University
Raleigh, North Carolina
Baohua Gu
Oak Ridge National Laboratory
Oak Ridge, Tennessee
Richard A. Beidelschies
Ohio Department of Agriculture
Reynoldsburg, Ohio
B.H. James, A.A. Coe & Ray Poole
PCS Phosphate
Aurora, North Carolina
Gene H. Williams & Sanford Simon
Pursell Industries
Orrville, Ohio, and Birmingham, Alabama
-------
Chapter 1
Environmental perchlorate: what role for fertilizer?
1.1. Introduction
Perchlorate was discovered in U.S. waterways in the late 1990s.
Most perchlorate salts are used as solid oxidants or energy boosters in
rockets or ordnance; therefore, much of the perchlorate-tainted
waterways in the U.S. can be traced to military operations, defense
contracting, or manufacturing facilities. Perchlorate ion is linked to
thyroid dysfunction, due to its similarity in ionic radius to iodide
(Clark, 2000). Becauseperchlorate-tainted waters are used for irrigation,
there are questions about absorption, elimination, and retention in food
plants. Furthermore, recent reports have suggested that fertilizers could
represent another source of perchlorate in the environment. Sporadic
findings of perchlorate in fertilizers were initially alarming because of
the widespread use of fertilizers in production farming. Because of the
dependence ofU.S. agriculture on chemical commodity fertilizers, it was
clear that assessment of any possible role of fertilizers would require
investigation.
Attention has been drawn to the possible roles of fertilizers in
environmental perchlorate contamination for two reasons. First,
perchlorate-tainted agricultural runoff could lead to pollution of natural
waterways used as drinking water sources. Second, there is a potential
for food plants to take up soluble compounds—including perchlorate
salts—and thus provide an alternate route of exposure. It has long been
known that Chile possesses caliche ores rich in sodium nitrate (NaNO3)
that coincidentally are also a natural source of perchlorate (C1O4~)
(Schilt, 1979; Ericksen, 1983). The origin of the sodium perchlorate
(NaClO4) remains an area of debate, but it is nonetheless present and
can be incorporated into any products made from the caliche.
1.2. Fertilizer manufacture and use
American farmers apply about 54 million tons of fertilizer yearly.
As with many commodity chemicals, large scale purchases are dictated
by cost of raw materials, which are in turn normally influenced by
transportation costs. Thus, proximate (rather than distant) sources of
agricultural chemicals are likely to play the greatest roles in local
ecosystems involved in production farming. In addition, processingaids
(e.g., clays) are likely to be derived from the closest possible source.
Production farming relies on cheap and available sources of primary
plant nutrients (macronutrients): nitrogen, phosphorus, and potassium.
Fertilizer grades (or guaranteed analyses) are expressed using the
convention of Samual Johnson, a student of Justus Liebig, who has been
called the "father of the fertilizer industry." In the Johnson convention,
nitrogen is expressed as a mass fraction (weight percent) as it was
originally determined by the Kjeldahl method. Phosphorus and
potassium are denoted in terms of the minerals formed by ashing at high
temperature. Thus, phosphorus is expressed as the mass fraction of the
empirical formula of its oxide, diphosphorus pentoxide (P2O5), and
potassium is expressed as the mass fraction of its oxide, potassium
oxide (K2O). This nomenclature does not reflect the actual composition
of the fertilizer, however, nor current practice in assaying it.
1.2.1. Nutrient availability
To minimize the need for multiple applications and to prevent
overdosing, timed-, delayed-, or controlled-release fertilizers are used in
both agricultural and horticultural applications. There are two
mechanisms to delay nutrient release. The first is to use essentially
insoluble minerals that are not readily converted to absorbable aqueous
phase nutrients, for example phosphate rock or other calcium
phosphates. The second is to coat the soluble fertilizer with an
insoluble material, such as a urea-based polymer or sulfur. This is often
done with consumer products, e.g., lawn fertilizers. Most urea-based
polymers are methylene ureas or urea-methanal blends. As urea
polymers are hydrolyzed, they too serve as a source of nutrients.
Many commodity chemicals used as agricultural fertilizers contain
fairly high concentrations of one, or sometimes two, of the primary
plant nutrients (Table 1.1). Trace metals (e.g., boron, copper,
magnesium) can be applied separately or along with these primary
nutrients on a farm site. Fertilizer application in production farming is
highly dependent on the crop and the native soil. Crops are influenced
by climate, weather, topography, soil type, and other factors that are
generally similar within a geographical region; therefore, crops and
fertilizer use are also similar within such a region. This is of course
unsurprising and consistent with agricultural production of dairy foods,
com, tobacco, wheat, etc. Because all plants require the same primary
nutrients, there is some usage to provide theseregardless of crop. Local
soil conditions also dictate what nutrients should be augmented, and
there can be large regional variations.
1
-------
Table 1.1. Some common agricultural fertilizers used in produc-
tion farmingf
Chemical name
Chemical
formula
Grade
(N-P-K
ratio)
Ammonia
(anhydrous)
Ammonium nitrate
Urea
Ammonium
monohydrogen
phosphate
(diammonium
phosphate)
Potassium chloride
(pure)
Potassium
magnesium sulfate
Triple super-
phosphate
NH,
82-0-0
NH4NO3
(NH2)2CO
(NH4)2HP04
34-0-0
46-0-0
18-46-0
KCI
K2Mg2(S04)3
0-0-62
0-0-22
Ca(H2PO4)2'H2O* 0-46-0
tMostsolidfertilizers contain processing aids (e.g., clays) that
keep the product from becoming a solid mass. Granular
products are often dusted with these. 'Triple superphosphate
is a somewhat ill-defined substance, being a slightly hydrous
mixtureofcalcium hydrogen phosphates (variable protonation).
It is represented here as calcium dihydrogen phosphate mono-
hydrate, but that compound is the main component and not the
only component.
On the other hand, consumer products can be distributed over
larger geographical regions because of the nature of the market. For
example, major manufacturers have a limited number of sites dedicated
to blending multiple-nutrient formulations. These products are often
sold as bagged fertilizers through home-improvement centers, nurseries,
florists, horticulturists, and department (or other retail) stores. Unlike
agricultural fertilizers, consumer products are usually multi-nutrient
formulations. In addition, trace metals are sometimes incorporated
directly into them. Because the average user will apply only a very
small amount of trace metals (or even primary nutrients) relative to a
production farm, it is more economical, more practical, and more
convenient touse multiple-nutrient formulations. Moreover, the average
consumer does not have the wherewithal to disperse careful doses of
several single-component fertilizers at the appropriate times of the
growing season.
Multiple-component fertilizers can be timed (controlled) releaseor
soluble blends. Many multiple-component products are intended for
soil amendment to lawns or gardens, e.g., 10-10-10, and other common
multiple-macronutrient formulations. Water-soluble blends are used to
supply nutrients rapidly to growing plants and are generally applied
repeatedly during a growing season (as with each watering), whereas
timed-release fertilizers allow water t o leach nutrients slowly for release
to the soil and plants. They are applied perhaps once or twice a year,
e.g., a lawn winterizer. Obviously, soluble and insoluble fertilizers
cannot be entirely identical chemically. However, the distinction is
essentially irrelevant for agricultural fertilizers, which are applied to
fortify particular nutrients. Of course, allowances must be made forthe
bioavailability of these nutrients. As a general rule, agricultural
fertilizers are soluble chemicals. Because fertilizer application on
production farms is geographically delimited, there is considerable
interest in knowing which commodity chemicals might contain
perchlorate and how much. Such information might suggest regions for
further investigation. Moreover, it will be important to know what
crops might potentially be affected—if any.
1.2.2. Nitrogen sources
urea •«H1ttntriftigt"
Figure 1.1. Schematic of urea production.
The simplest nitrogen source is anhydrous liquid ammonia. Liquid
ammonia is stored in bulk tanks and injected directly into the soil. No
fertilizer has a higher nitrogen content. Ammonia is made using the
Haber process, which entails heating desiccated nitrogen (separated
from liquified air) and hydrogen (usually from methane) in the presence
of a catalyst at 500-700 °C.
Urea is also a common source of nitrogen. Highly soluble in water,
urea hydrolyzes to carbonic acid and ammonia, given time. The
industrial process used to synthesize urea is shown in Figure 1.1. In
addition to its use as a fertilizer, a special feed grade of urea is used to
supplement cattle feed.
Nitrate salts are also used as fertilizers. Ammonium nitrate is the
primary nitrate salt used in production farming. Most—if not
all—ammonium nitrate today is made from atmospheric gases. None of
the major nitrogen fertilizer producers [Potash Corporation of
Saskatchewan (PCS), Agrium, Coastal, Mississippi Chemical, Kemira
Dansmark, and IMC] use natural saltpeters in manufacturing. IMC and
PCS do not sell nitrate-based fertilizers, focusing instead on urea,
ammonium phosphates, and similar nitrogenous compounds.
Consequently, perchlorate contamination is not possible from the raw
materials. Ammonium nitrate is prepared from nitric acid and ammonia.
Nitric acid is manufactured by the process shown in Figure 1.2.
Ammonium nitrate is produced as shown in Figure 1.3.
-------
stone
dehydrating
tower
90%HN03gas
-75% HzSO4 liquid
Figure 1.2. Synthesis of nitric acid by the redox reaction of
ammonia and atmospheric oxygen. A1:9 ratio of ammonia
and air pass through Rh/Pt gauze. The cooled nitric oxide
and oxygen pass into an absorber where more air is added
and the NO is converted to nitrogen dioxide. The nitrogen
dioxide undergoes hydrolytic disproportionationtogive nitric
acid with NO as a byproduct. After dehydration, the nitric
acid vapor is recovered from the gas phase.
95%
NH«NOj
IMeter •-
product
Figure 1.3. Schematic of ammonium nitrate production.
Aqueous nitric acid and anhydrous ammonia are
combined to produce the salt, which is then prilled and
dried.
The alkali metal saltpeters (sodium andpotassiumnitrates) are also
used as nitrogen sources. Their mineral forms are known as soda niter
(nitratine) and potash niter (nitrine), respectively. Chile saltpeter
(NaNO3) is mined from caliche ores in the North. The mined rock
contains veins rich in sodium nitrate. The ore is crushed and mixed with
water to dissolve the soluble salts. The sodium nitrate is then recovered
from the leachate. Chile's Sociedad Quimica y Minera S.A. (SQM)
reports annual production of about 992,000 tons of nitrate products.
SQM North America sold some 75,000 tons to U.S. farmers in 1998.
The company touts its products primarily for cotton, tobacco, and
citrus fruits. It is the caliche ores that contain naturally occurring
perchlorate. No other company sells a product derived from caliche as
of this writing; however, Potash Corporation of Saskatchewan does
own caliche mines in Chile (Searls, 1999; EIU, 1999)
Because nitrate salts (saltpeters) find use as fertilizers, these
natural resources have been mined and refined to produce commercial
fertilizers for domestic use or for export. Chilean nitrate fertilizers
(NaNO3 and KNO3) are manufactured by SQM. SQM markets its
products in the U.S. under the name Bulldog Soda. Chilean nitrate salts
are sold to agricultural operations, chemical suppliers, and consumer-
oriented companies such as Voluntary PurchasingGroups, Inc., or A.H.
Hoffman, Inc., who repackage and resell it as Hi-Yield* or Hoffman®
nitrate of soda, respectively. Also, secondary users may incorporate
Chilean nitrate salts into water-soluble plant foods, lawn fertilizers, and
other retail (specialty) products.
Due to their cost and availability, Chilean nitrates are niche
fertilizers. SQM markets its products to growers of tobacco, citrus
fruits, cotton, and some vegetable crops, particularly emphasizing that
the products are low in chloride content (CNC, 1999). As noted above,
typical American fertilizer consumption is 54 million tons per year;
consequently,most U.S. fertilizers arederived from other raw materials.
For example, ammonium nitrate (NH4NO3), which is often used for
purposes similar to NaNO3, is manufactured from methane, nitrogen,
and oxygen. There is no evidence that any ammonium nitrate is derived
from Chilean caliche. On account of its low usage, perchlorate from
Chilean nitrates cannot represent a significant anthropogenic source of
perchlorate nationwide, regardless of the perchlorate content. Recent
examination of two manufacturing lots found perchlorate concentrations
below 2 mg g"1, i.e., < 0.2% w/w, with some lot-to-lot variability
(Urbansky, 2001). However, in a recent letter to EPA, SQM North
America's President Guillermo Farias indicated that SQM had modified
its refining process to produce fertilizer containing less than 0.01%
perchlorate (<0.1 mg g"1); this corresponds to a reduction of 90-95%.
SQM monitors its production stream every 2 hours to verify the
perchlorate concentration. Accordingly, previous data on perchlorate
content are only applicable in a historical sense rather than being
reflective of ongoing fertilizer use.
As Table 1.2 indicates, there is limited application of natural
saltpeters as fertilizers in the U.S. based simply on total consumption.
There just is not enough production of the natural materials. Some
states keep detailed records on fertilizer use, especially of chemical
commodities used in production farming, but others do not. For
example, the Office of Indiana State Chemist is required to keep track
of only the top ten fertilizers; even ammonium nitrate is not among the
top ten in Indiana. As a result, it is not easy to discern the potential
distribution of minor fertilizers known to contain traces of perchlorate
salts. Table 1.2 gives the tonnage for a few nitrogen fertilizers for
several states. Many states do not keep records as detailed as those of
the Office of the Texas State Chemist, which tracks each fertilizer by
grade (N-P-K ratio) and tonnage. The Corn Belt relies heavily on urea
-------
and anhydrous ammonia as nitrogen sources, as shown by Indiana and
Ohio consumption of these two chemicals in Table 1.3, while
ammonium nitrate finds greater use in tobacco-farming states.
Table 1.2. Consumption (in tons) of nitrate salts in regions of
the continental United States for the year ending June 30,1998
Region
New
England
Mid-Atlantic
South
Atlantic
Midwest
Great Plains
East South
Central
West South
Central
Rocky
Mountain
Pacific
U.S. Total*
NH4*
2,469
33,556
162,035
81,585
436,371
489,603
338,618
254,168
148,340
1,946,868
Na*
194
260
14,87
0
189
409
7,786
4,192
831
10,28
1
39,01
3
K*
142
12,06
4
17,30
8
1,496
296
2,122
652
9,022
146
46,10
0
Na*/K*
0
0
21,76
2
5
55
914
2,851
0
0
22,76
2
Key: New England (ME, NH, VT, MA, Rl, CT); Mid-Atlantic (NY,
NJ, PA, DE, MD, WV); South Atlantic (VA, NC, SC, GA, FL);
Midwest (OH, IN, IL, Ml, Wl); Great Plains (MN, IA, MO, ND, SD,
NE, KS); East South Central (KY, TN, AL, MS); West South
Central (AR, LA, OK, TX); Rocky Mountain (MT, ID, WY, CO, NM,
AZ, UT, NV); Pacific (CA, OR, WA); *Total U.S. includes HA, AK,
PR. Source: Association of American Plant Food Control
Officials/The Fertilizer Institute, Commercial Fertilizers 1998.
D.L. Tern/and B.J. Kirby, Eds. Universityof Kentucky: Lexington,
KY, 1998.
Sodium and potassium nitrates make up a small fraction of the
nitrate application in the United States; however, prior to the establish-
ment of nitric acid and ammonia factories, natural saltpeters played
significant roles in American agriculture. In addition, ammonium nitrate
was manufactured from Chile saltpeter before the industrialized
oxidation of ammonia to nitric acid became commonplace in the 1940s.
Decades ago, ammonium nitrate was prepared from Chilean sodium
nitrate by ion exchange rather than from gaseous reactants. Historical
use of ammonium nitrate previous to or in the first half of the 20th
century might be linked to contaminated groundwater, and has been
attributed toonemanufacturingfacility in Arizona (EPA, 1999). On the
other hand, a recent survey of water supplies forperchloratewas unable
to detect perchlorate in nearly all of them, and concluded that
perchlorate contamination is generally localized and related to point
sources (Gullick, 2001). Reliable data on the use of natural saltpeters
appear to be unavailable. While most nitrate salt manufacturers rely on
ammonia oxidation (Table 1.4), which is a chlorine-free process, the
possibility of contamination from products derived from Chile saltpeter
cannot be ignored.
The lack of information on natural attenuation as well as limited
knowledge of hydrogeology makes it difficult to determine where and
how such problem sites might be found. For this reason, monitoring for
perchlorateunderthe EPA's Unregulated Contaminant MonitoringRule
for drinking water can be expected to provide some of the most useful
information. Meanwhile, it is instructive to consider the processes by
which major nutrients are produced so as to evaluate the possibilities
for contamination.
Table 1.3. Annual consumption/application (in tons) of some
nitrogen fertilizers for several states
NH3'
82-0-0
Urea
46-0-0
NH4NO3 KNO3
34-0-0 14-0-46
State
Arkansas* 1,207 465,737 62,003 N.R.t
Georgia0 3,859 15,084 56,215 N.R.
Indiana* 193,347 48,478 N.R. N.R.
Maryland§ 1,155 11,614 9,518 N.R.
New 13,747 34,348 2,725 N.R.
Mexico*
Ohio* 86,499 115,180 7,516 N.R.
Texas" 142,383 74,235 80,120 1,115
West 390 4,476 1,283 N.R.
Virginia*
•Anhydrous. *1998. §July 1998-June 1999. *1997. **March-
August 1998. tN.R. = not reported. DJuly 1998-June 1999.
Sources: Arkansas State Plant Board; Association of American
Plant Food Control Officials; PlantFood, Feed, & Grain Division,
Georgia Department of Agriculture; Office of Indiana State
Chemist and Seed Commissioner; Maryland Department of
Agriculture, Office of Plant Industries and Pest Management,
State ChemistSection; Feed, Seed, andFertilizerBureau.New
Mexico Department of Agriculture; Office of the Texas State
Chemist; West Virginia Agricultural Statistics Service.
-------
Table 1.4. Annual production (tonnage) of nitrate salts by some manufacturers who supplied the U.S. marketplace*
Company
Agrium§
Calgary, Alberta, Canada
Coastal Chem, Inc.
Cheyenne, WY§§
Kemira Danmark§
Fredericia, Denmark
Vicksburg Chemical Co.§
Vicksburg, MS
ammoxt
KNO3
npfl
np
3239
151,000
ammox
NaNOa
np
np
np
np
ammox
NH,NO3
630,000
306,180
np
np
ammox
HN03
np
245,164
np
np
Chilean
NaNOj
546J
np
np
np
Chilean
KNO3
652t
np
np
np
*Anumberof companies with smaller production outputs make NH4NO3 and urea, such as Apache Nitrogen Products Inc., which
manufactured 50,000 tons of fertilizer in 2000. This table excludes sales from SQM.fTheterm "ammox" refers to use of ammonia
oxidation to produce nitric acid. §Actual or estimated production for 2000. IfThe designation "np" means that the manufacturer
reported to EPA that this material was not a product. JThis material was sold by Western Farm
Services, an Agrium subsidiary, and was reported to have "little to no detectable perchlorate" *«>(:
according to information provided by Agrium to EPA. §§Actual production for 1999.
1.2.3. Phosphate sources
Phosphate rock is mined in a number of states, including Florida,
Idaho, and Montana. Florida's phosphate rock deposits are near the
surface (~8 m down) and formed 5-15 million years ago. They are
obtained by removingthe overburden (coveringrock). Figure 1.4 shows
Figure 1.5. Bucketwheel used
in mining phosphate rock.
Photo © PCS Phosphate. Used
permission.
phosphate rock processing In
North Carolina, overburden is
taken up via bucketwheel (Figure
1.5). Natural phosphate rocks
usually contain a mixture of the
apatite minerals. Fluoroapatite
has the empirical formula
Ca5(PO4)3F.
Ammonium monohydrogen
phosphate, (NH4)2HPO4, is commonly referred to as DAP in the
agricultural chemical industry, short fordiammonium phosphate. DAP
is a source of both nitrogen and phosphorus and has a grade of 18-46-0.
The ACS reagent is a white crystalline material; MIST sells J.T. Baker
Ultrex ammonium monohydrogen phosphate as SRM 694 for use by
fertilizer manufacturers in assaying this material. However, agricultural
DAP is generally a mixture of gray, brown, and/or black pellets (Figure
1.6). Some of this color is due to
residual calcium minerals (e.g.,
apatites, gypsum) and some is
due to natural organic matter.
Most agricultural DAP contains
6-15 mol% NH4H2PO4 or
monoammonium phosphate
(MAP).
Figure 1.6. Agricultural DAP,
In the Corn Belt, granular product provided courtesy of CF
triple superphosphate or GTSP industries.
(0-46-0) and DAP, dominate the
market. About 82% of the ort/io-phosphoric acid produced in the U.S.
goes into fertilizer manufacture, with 49% into DAP and 10% into
MAP.
Figure 1.4. Operational schematic of phosphate
rock processing. MGA refers to merchant grade
acid. © PCS Phosphate. Used with permission.
-------
Figure 1.7.The mineral sylvite (KCI). The red or brown color
comes from occluded hematite (Fe203) or limonite
(Fe(O)OHv?H2O). Materials provided courtesy of Kova, Inc.,
(left) and CF Industries, Inc. (right).
1.2.4. Potassium sources
Essentially no potash (K2O/K2CO3) is used as a fertilizer today,
but many products are referred to as potash for historical reasons.
Potassium nitrate and potassium chloride (muriate of potash, MOP)
dominate this market. Sylvite (KCI) is mined in Canada, and is the most
popular potassium source in the Corn Belt. Both Saskatchewan and
New Brunswick have sylvite and/or sylvinite mines. The red material
on the left in Figure 1.7 has a guaranteed analysis of 0-0-60, while the
purer material on the right is guaranteed at 0-0-62, which represents the
maximum, within the limits of experimental error.
Sylvinite (43% KCI, 57% NaCl) deposits occur in New Mexico
and can be refined to remove much of the halite (NaCl). New Mexico
also has reserves of sylvite and langbeinite [potassium magnesium
sulfate, K2M§2(864)3 or 2K2SO4'MgSO4]. Langbeinite is popular in
dairy country as a source of magnesium.
Like Chile saltpeter, these minerals are marine evaporites resulting
from the drying up of terminal inland seas. In most cases, deposits of
H
/\
[1-OreSMBB
D
Figure 1.8. Canadian potash mines can be kilometers
below the surface. © PCS Potash. Used with
permission.
Ewaottn
Floating \
Dredge V
Injection
Figure 1.9. Sylvite and sylvinite can be mined by
dissolving the minerals in water and pumping the brine
to the surface. © PCS Potash. Used with permission.
sylvite are hundreds or thousands of meters below the surface, having
been covered over by sedimentary rock formations over some 300
million years.
The market is dominated by two producers, IMC-Kalium and
PCS. PCS has deep mines that rely on traditional techniques (Figure
1.8) and solution mines that make use of the solubility of the minerals
(Figure 1.9).
1.2.5. Information sources
Information on fertilizer production and application comes from
a variety of sources, including trade organizations, manufacturers, and
government (state/federal) agencies. Both the U.S. Census Bureau and
the U.S. Geological Survey track fertilizer commodities. The Census
Bureau's Economics and Statistics Administration publishes an annual
report (MA325B, formerly MA28B) as well as quarterly reports
(MQ28B) on inorganic fertilizer materials and related products. The
Geological Survey publishes reports on fertilizer minerals that cover
manufacture, use, regulation, litigation, and other matters (Cf. Lemons
1996;Searls, 1999). Natural Resources Canada also publishes aminerals
yearbook (Cf. Prud'homme, 1998). Most publications are available
online.
Because data are obtained through many sources, it is common for
there to be inconsistencies as well as apparent inconsistencies.
Apparent inconsistencies sometimes stem from how materials are
tracked. For example, an apparent inconsistency in the data between
Tables 1.2 and 1.4 relates to ammonium nitrate production (-940,000
tons) versus consumption (1.8 million tons). However, Table 1.4
neither includes all manufacturers, nor product blends. In addition, the
tables do not account for normal fluctuations in inventory. In other
-------
words, a material produced in one year may be sold or used the
following year. Taken together, these factors increase the difficulty in
monitoring the application of or tracking transactions involving
perchlorate-containing materials.
1.3. Previous fertilizer analysis studies
Aside from the analyses of Chilean caliche, there were no studies
to suggest that any other processed fertilizer or raw material might
contain perchlorate prior to 1998. That year the Ecosystems Research
Division of the EPA's National Exposure Research Laboratory (EPA-
ORD-NERL-ERD) found perchlorate in several samples that were not
derived from Chile saltpeter (Susarla, 1999D). This finding was later
duplicated by other investigators from the North Carolina State
University Department of Soil Science. However, the presence of
perchlorate could only be confirmed in consumer products, and not in
agricultural fertilizers. Moreover, subsequent analyses of different bags
(likely different lots) of many of the same brands and grades did not
show perchlorate (Susarla, 2000; Williams, 2001). The choice of
fertilizers is problematic because the same raw materials may be used
in a variety of products at a point in time. Additionally, a few major
companies are responsible for makinga large number of products under
several brand names. Furthermore, some companies rely on toll
manufacturing so that the products are actually made by another
company to meet a specific formulation. Accordingly, an error or
contamination associated with one raw material could affect a variety of
products without regard to company or application. Since those early
days, each subsequent study on fertilizer has attempted to address more
issues, and study designs have been continually refined based on what
was learned in previous investigations.
The EPA-ORD-NERL-ERD study brought to light a number of
important issues for trace analysis of fertilizers. First, most of the
research on determining perchlorate to that time had been focused on
either finished potable water or raw source water (Urbansky, 2000F).
Second, fertilizers are considerably more complicated matrices than
dilute water solutions. Third, solid fertilizers are not homogeneous. In
fact, some are macroscopically heterogeneous, for example, multi-
component formulations used as lawn and garden fertilizers. It is
possible to sort out the particles by hand. Thus, representative
samplingbecomes akey issue. Fourth, the effectiveness of the leaching
step must be evaluated. Fifth, the materials must be carefully selected
to properly reflect the market of interest, e.g., production farming, lawn
treatment, vegetable gardens, houseplant foods.
About the same time as the EPA-ORD-NERL-ERD study, the
U.S. Air Force Research Laboratories (AFRL) performed a study to
assess interlaboratory corroboration, that is, the ability of different labs
to analyze the same sample and get the same result (Eldridge, 2000).
Samples of a variety of lawn and garden fertilizers were selected from
around the country. The AFRL study used a number of products.
These included four brand name products from New York, three from
Missouri, and one from California. Overall, these represented multiple
lots of four brand name fertilizer products, with some different lots of
the same products coming from different cities. The AFRL study did
not account for the sources of the commodity chemicals blended into
these products, and it did not link manufacturing lots with lots of raw
materials. Even bagged fertilizers from different manufacturing lots may
be comprised from some of the same raw materials. Therefore,
limitations in choice of products prevent extrapolating the results to
large scale fertilization (as in production agriculture).
An AFRL contractor was tasked with leaching or dissolving the
materials and parceling out the liquid samples to 7 laboratories running
1C and 3 using other techniques (capillary electrophoresis, tetra-
phenylstibonium titrimetry, and Raman spectrometry). Each laboratory
was sent a sample of the same test solution for analysis. Laboratories
were permitted to use any means of analysis. Ion chromatography,
capillary electrophoresis, titrimetry, and Raman spectrometry were all
employed. Interlaboratory agreement was generally good, indicatingthat
laboratories were able to determine perchlorate in the fertilizer
solutions, despite matrix complexity. Quality control checks were
limited to performance on deionized water blanks and perchlorate
solutions prepared from deionized water; laboratories were not required
to show performance on fortified samples (spike recovery). Some of the
samples failed to show Raman scattering lines consistent with their
major components; this matter was not addressed.
While interlaboratory agreement was good on the liquid solutions,
values for leachates derived from different samples of the same lot of
material varied substantially in some cases. Heterogeneity of bagged
fertilizer products (especially multiple component products) was
therefore demonstrated to be a significant matter by the AFRL study.1
Specifically, duplicate samples of solid from bagged fertilizers gave
aqueous leachates withconsiderable differences in measured perchlorate
concentration: 5.8 ± 0.5 mgg"1 versus 2.8 ± 0.3 mg g~' for one product
and 0.98 ± 0.14 mg g~' versus 2.7 ± 0.3 mgg~' for another [values here
are the average and estimated standard deviation for 10 results (each of
which is an average for one of 10 methods]. Such variation may be
attributable to sampling error and does not necessarily reflect an actual
difference. EPA-ORD-NERL-ERD has observed even wider variation
in grab samples, with some measured concentrations as much as 5-7
times larger than others. When some of the same products were
subjected to sampling scheme intended to yield a more representative
sample, considerably lower intersample variability was observed
(Williams, 2001). Less striking variation in perchlorate distribution
within and among bags of sodium nitrate fertilizer has been seen for
small grab samples of solids, but can be eliminated by more rigorous
sampling or by choosing larger sample sizes (Urbansky, 2001).
For one product in the AFRL study, perchlorate concentrations in
the first solid sample taken were below the detection limit in one
'Here, we call these products bagged fertilizers even though some are sold
in boxes or jars; this contrasts with bulk fertilizers sold by railway car or
track and then used for agricultural purposes (or eventually incorporated
into a bagged fertilizer). Such products are usually referred to as specialty
fertilizers within the industry. While some products are specifically
manufactured so as to assure uniform distribution of macronutrients and
micronutnents, it is unclear whether perchlorate contamination would also
be homogeneous.
-------
leachate, according to all labs. A second solid sample contained 2.7 ±
0.3 mg g"1, based on its leachate [value is the average and estimated
standard deviation for 10 results (each of which is an average for one of
10 methods)]. It is unclear whether sampling error alone can account for
the disparate values. The analytical laboratories were sent the same
leachate (one for each solid) and were not working with different
portions of the solid. While the AFRL used the data only to assess
interlaboratory corroboration of fertilizer analysis by ion chroma-
tography, the results also confirm that there was perchlorate in some of
the materials purchased during a specific time period. However, no
conclusions should be drawn regarding perchlorate concentrations in
specific lots or brands of fertilizer because the inconsistencies in the
data were not explored further and the sampling approach was
restrictive in nature.
Subsequently, the Water Supply and Water Resources Division of
the EPA's National Risk Management Research Laboratory (EPA-
ORD-NRMRL-WSWRD) conducted its own survey of fertilizers in a
collaboration with the Department of Energy's Oak Ridge National
Laboratory (Urbansky, 2000A; Urbansky, 2000D). In addition to a
variety of products purchased from home improvement, garden supply
or department stores, products were purchased from farming supply
stores (e.g., 50-lb bags of urea or ammonium nitrate) in Indiana, Ohio,
Kentucky, Pennsylvania, and Tennessee. In addition, commodity
chemical samples were collected from local distributors in Ohio and
Indiana. These included urea, potassium chloride, ammonium
monohydrogen phosphate, and granular triplesuperphosphate, among
others. Samples were leached or dissolved and subjected to
complexationelectrospray ionization mass spectrometry (cESI-MS) or
ion chromatography (1C). All 1C analyses were performed by the Oak
Ridge National Laboratory, which received separate portions of the
solid materials. The only products that were found to contain any
perchlorate were those based on Chile saltpeter. While this study was
the first to include the same products used on production farms, it did
not address the issue of sampling.
Sampling strategies for commodity chemicals are alwaystricky.lt
is essentially impossible to sample 23,000 tons of urea piled in a
warehouse representatively. These products are moved with heavy
machinery and transferred to railcars and trucks from barges using
conveyor systems and other heavy equipment. It is worth noting that
the equipment used for transfer and transportation of these
commodities is not normally cleaned thoroughly between uses for
different commodities. A common practice is to run a few tons of gravel
through the system. Accordingly, traces from residual products are
possible.
Fertilizers are normally sampled by taking repeated cores through
the piles using a Missouri D tube sampler. The core samples are
combined, riffled, divided, and analyzed. These practices are standard
within the fertilizer industry and regulatory bodies. Across the nation,
state chemists or agriculture departments are obligated to examine
fertilizers to verify the manufacturers' reported grades. Sampling
practices have evolved to fill those needs. Previous studies generally did
not take these practices into account, concentrating instead on the
analysis of the solid once a grab sample had been collected. Sampling is
of course important to obtain representative results. The distribution of
perchlorate is not uniform in Chilean sodium nitrate. For example, in
two lots with average concentrations of 1.5 and 1.8 mg g"1, individual
10.0-g grab samples ranged from 0.74 to 1.96mgg~' (Urbansky, 2001).
Information from previous studies and from standard industry practices
was used to guide the study reported on in Chapter 2.
-------
Chapter 2
Survey of fertilizers and related materials
2.1. Objectives
In an effort to take into account the difficulties of the matrix, the
problems with sampling, and the applicability of the results, this study
was undertaken. No single study can say once and for all whether there
is perchlorate in fertilizers. However, it is possible to provide a
snapshot of current fertilizers commodities. The EPA entered into a
collaboration with The Fertilizer Institute, the International Fertilizer
Development Center, the Fertilizer Section of the Office of Indiana
State Chemist and Seed Commisioner, the North Carolina State
University Department of Soil Science (which serves as part of the
North Carolina Agricultural Research Service), and IMC-Global for this
purpose.
This study was composed of two distinct phases. Phase 1 was
designed to evaluate laboratory performance and the ruggedness of the
method. Laboratory participation was on a voluntary basis.
Laboratories were required to use ion chromatography, but were
permitted to choose columns and operating conditions on their own,
within certain limits. Phase 1 test samples included a wide variety of
fertilizer matrices. In Phase 2, samples of materials were collected from
around the nation and sent to the participating laboratories. All data
were provided to EPA for evaluation and analysis.
2.2. Phase 1—Evaluation of laboratories
These spanned commodity chemicals, water-soluble plant foods,
and granulated/pelletized lawn fertilizers. Phase 1 samples were
prepared by EPA-ORD-NRMRL-WSWRD and sent directly to
participating laboratories under custody seal. A combination of solid
and liquid (aqueous) samples was sent to each laboratory. All of the
liquid samples were made by leaching or dissolving the solid in
deionized water at a ratio of 10 g dL~' (-10% w/w). All of the raw
materials were either American Chemical Society (ACS) reagent grade
or demonstrated to be perchlorate-free (within the limits of experimental
error) by EPA analyses (Urbansky,2000A). Cations of the perchlorate
salts used to fortify these materials are identified in parentheses.
Laboratories were required to demonstrate recovery of fortifications,
reproducibility, and ruggedness in real matrices. They were also required
to supply experimental details and calibration data. Lastly, quality
control specifications were established with regard to number of
replicates, blanks, spike recovery, accuracy, and precision. Laboratories
were required to estimate limits of detection within each matrix using a
standardized procedure. The method has been released as EPA/600/R-
01/026, and details covered in the method will not be repeated here.
A set of performance evaluation samples was prepared by EPA-
ORD-NRMRL-WSWRD. The identities of these were as follows:
1. agricultural GTSP
2. 1:1 bentonite:kaolinite + 990 u.g g"1 C1O4" (Na+)
3. agricultural DAP
4. agricultural grade NH4NO3 + 620 ng g'1 CKV (Na+)
5. agricultural grade NH4NO3 + 310 jig g"1 C1O4" (Na+)
6. duplicate of no. 5
7. ACS reagent urea + 530 ng g'1 C1O4~ (Na+)
8. ACS reagent urea
9. agricultural KC1 (0-0-60)
10. duplicate of no. 9
11. 10 g dL"1 aq. ACS reagent urea, K2SO4, NaCl, (NH4)2HPO4, and
KNO3; xylene cyanole FF
12. duplicate of no. 11
13. no. 11 + 170 ng mL"1 CKV (Na+ salt)
14. no. 11 + 170 ng mL"1 C1O4~ (NH4+ salt)
15. Vigoro lawn winterizer, 22-3-14
16. Scotts lawn winterizer, 22-3-14
17. duplicate of no. 16
18. no. 16 + 1.4 mg g"1 C1O4"
19. 10 g dL"1 (total solids/liquid) aq. agricultural (NH4)2HPO4; ACS
reagent urea, KC1, NaNO3, MgSO4, bentonite, CoCl2«6H2O, and
NiCl2-6H2O
20. no. 19 + 82 ng mL-1 C1CV (Na+ salt)
21. no. 19 + 82 ng mL"1 CKV (K+ salt)
22. no. 19 + 82 ng mL"1 C1O4" (NH4+ salt)
23. no. 19 + 82 ng mL-' C1O4" (NBu4+ salt)
24. no. 19 + 82 ng mL"1 C1O4" (NOct/ salt)
25. aq. 6.2 ng mL"1 C1O4" + FD&C Blue No. 1
26. aq. 34 ng mL"1 C1O4" in Cincinnati tap water
27. aq. 1.34 ng mL"1 C1O4" in Cincinnati tap water
28. 10 g dL"1 aq. Peter's water soluble plant food, 20-20-20
29. 10 g dL~' aq. Miracle Gro tomato food, 18-18-21
30. 10 gdL'1 aq. ACSreagent (NH4)2HPO4 + 12 ng mL"1 C1O4" (Na+)
31. 10 g dL~' aq. ACS reagent KC1 + 74 ng mL-1 CKV (Na+)
Laboratories were required to demonstrate their ability to detect
perchlorate in fertilizers. Some laboratories failed to adequately test
-------
aqueous samples with concentrations in the parts-per-billion, and were
cautioned not to dilute samples initially. In Phase 1, the matrix was not
identified to the laboratories. This made the process difficult, time-
consuming, and resulted in higher errors because it was not possible to
initiate corrective measures to handle the needs of a particular matrix.
For this reason, major chemical constituents were identified for the
laboratories in Phase 2.
The following laboratories successfully completed Phase 1:
California Department of Food and Agriculture, Dionex Corporation,
American Pacific Corporation (AMPAC), NCSU Department of Soil
Science, Montgomery Watson Laboratories (MWL), and IMC-
Phosphates Environmental Laboratory.
2.3. Phase 2—Analysis of samples
2.3.1. Sampling and analysis strategy
Phase 2 consisted of the testing of fertilizer samples. The materials
sampled represented current major suppliers of production farm
fertilizers. Because these were raw materials, they also represented
materials used in consumer-oriented fertilizers. Clay additives and other
fertilizer products were included, too (e.g., ammonium sulfate).
Sampling sites were chosen to reflect national supplies and to provide
geographic coverage of raw materials. Samples were procured under the
direction of state regulatory agents duly authorized by their agencies
and recognized by the Association of American Plant Food Control
Officials (AAPFCO) for the collection and examination of fertilizers
using accepted procedures.
A number of objectives were established for the choice of samples
and the means of collection. Appropriate measures were then taken to
achieve these objectives.
1. To verify the method for analyzing fertilizer materials for
perchlorate ion. To this end, a large number of materials was
chosen to represent common matrices and ions that might
confound the analytical process. These materials included anions
that might interfere with perchlorate determination, e.g., sulfate,
chloride, phosphate, and nitrate.
2. To identify individual sources of contamination if
perchlorate ion is found by testing a broad cross section of
individual raw materials and nutrient sources throughout
the U.S. and include materials from each category
(nutrients and fillers) that represent ingredients common
to all products. To this end, we (1) enlisted the help of a large
number of material manufacturers and producers; (2) selected
products, and consequently materials, from diverse geographic
areas; (3) gathered at least three distinct samples of each major
nutrient source material, for example, three DAP samples each
from different suppliers in diverse geographic locations; (4)
identified certain materials as likely candidates for contamination
to be sampled without fail, specifically, Chilean nitrate salts,
langbeinite ore, and potassium chloride; (5) included at least two
samples of leading minor contributor sources such as iron oxide,
limestone filler, and clay conditioners that are typically part of
fertilizer blends. While many materials have been tested, the list is
not exhaustive and was not intended to be. Rather, it focuses on
major sources of many chemical commodities used in fertilizer
manufacture. Sources ofmacronutrients and several micronutrients
were the principal focus. While some fillers and minor ingredients
were tested, the coverage was not comprehensive.
Raw materials rather than finished products were focused on
for two reasons: (1) If perchlorate were present in blended
fertilizers, it would have had to come from a specific raw material
in the blend (assuming it was not intentionally added); finished
product sampling would not identify a specific contaminant. (2)
Raw materials should contain higher concentrations than
formulated products (i.e., blending dilutes the analyte) and their
analyses are likely to suffer fewer matrix effects. Products are
listed in Table 2.1.
3. To ensure correct sampling and third party validation, and
to maintain a chain of custody throughout the sampling,
transport, preparation, and distribution process. To this
end, samples were taken under the supervision of state officials
authorized by state statutes to collect samples of fertilizers, such
as state chemists or agriculture department officers. In addition to
manuals from The Fertilizer Institute (TFI), Association of
Official Analytical Chemists (AOAC, 2000a, 2000b), and the
AAPFCO (1999), written instructions were issued to each
inspector outlining the specific protocol to be used. These
instructions were in addition to and complementary to the
standard sampling practices. In a few cases, state inspectors were
unavailable, and knowledgeable practioners from industry were
permitted to collect the sample; this has been noted in Table 2.1.
Sampling was conducted at the actual production site whenever
possible to eliminate potential for contamination by other
materials. The collected samples were sent under custody seal to
the International Fertilizer Development Center (IFDC) for
riffling, grinding, and packaging. The IFDC sent the custody-sealed
packages to the participating laboratories.
4. To obtain consensus from governmental, industrial, and
academic interests on how to accomplish these goals. All
documents and proposals were reviewed by TFI, IMC-Agrico,
EPA-ORD-NERL-ERD, EPA-ORD-NRMRL-WSWRD, NCSU
Department of Soil Science, Office of Indiana State Chemist
(OISC) Fertilizer Section, U.S. Air Force Research Laboratories,
and the AOAC Fertilizer Referee.
A series of quality control samples were prepared for Phase 2 by
EPA-ORD-NRMRL-WSWRD and sent to the IFDC for distribution to
the participating laboratories. Each laboratory submitted its results to
EPA-ORD-NRMRL-WSWRD for evaluation and tabulation.
Laboratories were required to submit chromatograms for the sample
solutions and fortifications they made to demonstrate acceptable
recovery in the matrix. Laboratories were also required to submit proof
of the goodness of their calibrations, instrument performance, detection
limit, and sensitivity.
10
-------
2.3.2. Results
Table 2.2 shows the results obtained by the laboratories that
elected to participate in Phase 2. All of the samples were tested by
NCSU, CDF A, IMC, and AMP AC; not all of the samples were tested
by Dionex. Positive hits above the assured reporting level (ARL) were
rechecked. The calculation of the assured reporting level is described in
detail in the method itself, EPA/600/R-01/026 (Collette, 2000).
Other than nonzero results reported by one lab, perchlorate was
detectable only in materials derived from Chilean caliche. It must be
emphasized that thesenonzero results were still below the ARL except
for item 48. In the case of item 48, a granular triplesuperphosphate, the
same lab obtained discrepant results of 26 ngg~' (with an ARL of 20 jig
g~') and undetectable (with an ARL of 100 p.g g~'). Given that no other
laboratory obtained similarresults, this suggests that the 26 jigg~' value
is erroneous, and it has been rejected as an outlier. Initially, this lab
found 44.5 ug g~' in item 23, an ammonium dihydrogen phosphate
(MAP). However, it reported that perchlorate was undetectable when
requested to reanalyze the material. In addition, a portion of an archived
sample of item 23 was sent to this lab by EPA-ORD-NRMRL-
WSWRD for analysis; this portion also failed to show detectable
perchlorate. Consequently, the original value was discarded as faulty.
Samples of materials 5, 23, 40, and 42 were analyzed indepen-
dently by MWL and by EPA-ORD-NERL-ERD. MWL found 2675 ^g
g~' in item 42, while NERL found 2845 ug g~l. These values are
consistent with results reported by other laboratories. Perchlorate was
undetectable in 5,23, and 40, according to both MWL and EPA-ORD-
NERL-ERD.
Recoveries of fortifications ranged from 81% to 111%, regardless
of the specific increase in concentration resulting from spiking. The
increases in concentration, A[C1O4"], and the recoveries are reported in
Table 2.3. All laboratories demonstrated satisfactory performance in
this regard. Although not explicitly reported here, the ARLs can be
computed from the information in Table 2.3.
2.3.3. Discussion
It is worth pointing out at the U.S. Geological Survey and Air
Force Research Laboratories have found perchlorate in isolated samples
of sylvite taken from New Mexico (Harvey, 1999). USGS is engagedin
additional sampling of North American mining sites in order to assess
whether there are natural mineral deposits of potassium perchlorate in
sylvite or sylvinite. Because little is known about the mechanisms of
perchlorate formation in the natural environment (which are assumed to
be meteorological in nature), it is not clear whether these findings
represent a low-level background to be expected in evaporite mineral
deposits or not. There is, of course, the possibility for anthropogenic
perchlorate formation to occur during blasting operations to mine
potassium chloride, but it has not been observed. Nonetheless,
perchlorate has not been detected in any samples of agricultural grade
potassium chloride (0-0-62 or 0-0-60) taken under the auspices of the
EPA (items 4,8, and 28). Nor was it found inalangbeinitesample (item
13), acquired via blasting. Current data are inadequate to demonstrate
that KC1 or K2Mg2(SO4)3 suffer from inclusions of perchlorate salts to
an environmentally relevant extent.
Previous EPA-ORD-NERL-ERD and AFRL findings cannot be
applied to fertilizers as a whole because of the composition and nature
of the products investigated. Those findings may have reflected a
temporal contamination of one or more raw materials or an error in
manufacture. The only fertilizers unequivocally and consistently
demonstrated to contain perchlorate were bagged products deriving
some or all of their nitrogen from Chilean nitrate salts, which are known
to vary in perchloratecontent. Even though perchlorate was previously
identified in some fertilizer products and was presumably introduced
through a contaminated raw material, this incident appears to have been
isolated and the source of the perchlorate remains an enigma, for some
of these products had no known link to Chile saltpeter. Furthermore,
awareness within the fertilizer industry and the environmental
community is now substantially heightened.
Based on the studies thus far (Collette, 2000; Gu, 2000; Urbansky,
2000A;Urbansky, 2000D; Urbansky, 2000E; Robarge, 2000), there is
a consensus among researchers that there is insufficient evidence for
fertilizers to be viewed as contributors to environmental perchlorate
contamination, except for imported Chile saltpeter or products derived
from it. The potential future influence of such products is further
reduced by SQM's modifiedrefiningprocess to lower perchlorate con-
centration in its products. Within the limits of experimental error and
current ion chromatographic technology,agricultural products tested by
EPA-ORD-NERL-ERD,ORNL,andEPA-ORD-NRMRL-WSWRDall
were devoid of perchlorate as were all of the products in this survey
except for those known to contain or to be derived from mined Chile
saltpeter. While the results presented herein reflect another such
snapshot, it is the most comprehensive examination of fertilizers and
related materials to date. As with any environmental contaminant, an
individual study is limited in applicability to the future. Nonetheless,
the weight of evidence obtained to date largely argues against fertilizers
as sources of environmental perchlorate. That notwithstanding, it would
not be imprudent for manufacturers, trade associations, or the industry
to establish periodic monitoringpractices to screen either raw materials
or finished products for perchlorate in a manner similar to how finished
drinking water is tested, perhaps with some allowance for periodic
phenomena.
11
-------
Table 2.1. Fertilizers and related materials surveyed for perchlorate
Item Composition
Manufacturer/supplier
Authority of
1 Lawn fertilizer, 22-3-14
2 Ammonium monohydrogen phosphate (DAP)
3 Urea
4 Potassium chloride (MOP)
5 Iron oxide
6 Limestone
7 Potassium magnesium sulfate (Sul-Po-Mag)
8 Potassium chloride (MOP)
9 Osmocote 18-6-12
10 Miracle Gro lawn fertilizer, 36-6-6
11 Miracle Gro plant food, 20-20-20
12 Langbeinite, mechanical mining
13 Langbeinite, drill and blast
14 Potassium magnesium sulfate (Sul-Po-Mag)
15 Limestone
16 Ammonium sulfate
17 Urea
18 Ammonium sulfate
19 Ammonium monohydrogen phosphate (DAP)
20 Potassium magnesium sulfate (Sul-Po-Mag)
21 Potassium sulfate
22 Ammonium sulfate
23 Ammonium dihydrogen phosphate (MAP)
24 Iron oxide
25 Limestone
26 Urea
27 Clay
28 Potassium chloride
29 Urea
30 Ammonium nitrate
Scotts Company, Marysville, OH ODA
IMC-Agrico ODA
Potash Corporation of Saskatchewan ODA
Potash Corporation of Saskatchewan ODA
Sims Agriculture ODA
Millersville Lime ODA
IMC-Kalium, Carlsbad, NM CDFA§
Mississippi Potash, Carlsbad, NM CDFA§
Scotts, Marysville, OH ODA
Scotts, Port Washington, NY ODA
Scotts, Port Washington, NY ODA
IMC-Kalium, Carlsbad, NM NMDA
IMC-Kalium, Carlsbad, NM NMDA
IMC-Kalium, Carlsbad, NM NMDA
Chemical Lime, Salinas, CA CDFA§
Simplot, Pocatello, ID CDFA§
Unocalt CDFA§
Dakota Gasification, Bismarck, ND CDFA§
IMC-Agrico FDACS$
IMC-Kalium FDACSJ
IMC-Kalium FDACSJ
Dutch State Mines (DSM) FDACSJ
IMC-Agrico FDACSt
Fritt Industries, Ozark, AL FDACSJ
E.R. Jahna, Lake Wales, FL FDACSJ
Unocalt FDACSrJ:
Ag Sorb FDACSJ
PCS FDACSt
CF Industries, Donaldsonville, LA supplier
LaRoche, Atlanta, GA IFDC°
12
-------
Table 2.1 continued
Item Composition
Manufacturer/Supplier
Authority of
31 Ammonium monohydrogen phosphate (DAP)
32 Ammonium dihydrogen phosphate (MAP)
33 Potassium sodium nitrate
34 Ammonium nitrate
35 Potassium nitrate
36 Sodium nitrate
37 Ammonium nitrate
38 Granular triplesuperphosphate
39 Ammonium dihydrogen phosphate (MAP)
40 Ammonium monohydrogen phosphate (DAP)
41 Limestone
42 Potassium nitrate
43 Plant food, 10-10-10
44 Clay
45 Potassium magnesium sulfate
46 Potassium nitrate
47 Ammonium monohydrogen phosphate (DAP)
48 Granular triplesuperphosphate
Agrium, Soda Springs, ID
Agrium, Soda Springs, ID
SQM FDACS®
Mississippi Chemical, Yazoo City, MS FDACS®
SQM FDACS®
SQM FDACS®
Mississippi Chemical, Yazoo City, MS FDACS®
Cargill, Riverview, FL
Simplot, Pocatello, ID
Simplot, Pocatello, ID
Georgia Marble
SQM
SSC, Statesville, NC supplier
Oil Dry Co., Ripley, MS Pursell IndJ
IMC-Kalium, Carlsbad, NM Pursell Ind.H
Vicksburg Chemical Co., Vicksburg, MS manufacturer
PCS, Aurora, NC manufacturer
IMC-Agrico, Mulberry, FL manufacturer
Quality control samples prepared by EPA-ORD-NRMRL-WSWRD (all solids)
49 Potassium chloride + 6.8 mg CIO4~ g~'
50
51
52
Peter's water soluble plant food, 20-20-20
+ 6.2mgCICvg'
J.T. Baker (Phillipsburg, NJ) ACS reagent EPA
KCI and KCIO4
Scotts (Marysville, OH) + GFS (Columbus, EPA
OH) ACS reagent NaCIO4
Granular triple superphosphate + 2.7 mg CIGy g ' A.H. Hoffman (Lancaster, PA) + GFS ACS EPA
reagent NaCIO4
Urea + 1.8 mg CIO4~g '
53 Potassium chloride (no analyte added)
D.W. Dickey & Son (Lisbon, OH) + Aldrich EPA
(Milwaukee, Wl) ACS reagent NH4CIO4
54
55
Ammonium nitrate (no analyte added)
Chilean sodium nitrate (Chile saltpeter)0
CF Industries (Cincinnati, OH)
+ A. H.Hoffman
Cargill (Shelbyville, KY)
A.H. Hoffman (mfd. by SQM)
EPA
EPA
EPA
Key: Fertilizers: Sul-Po-Mag = sulfate of potash/magnesia, MOP = muriate of potash, DAP = diammonium phosphate, MAP =
monoammonium phosphate. Manufacturers/suppliers: PCS = Potash Corporation of Saskatchewan; SQM = Sociedad Quimica
y Minera. Authorities: CDFA = California Department of Food and Agriculture; FDACS = Florida Department of Agriculture and
Consumer Services; NMDA= New Mexico Departmentof Agriculture; ODA= Ohio Department of Agriculture. fUnocal is now owned
by Agrium. ^Sampled at Pursell Industries in Winterhaven, Florida. ®Sampled at Gro-Mor Co., Inc., Plant City, FL. §Sampled at
Pursell Industries, Los Angeles, California. IJPursell Ind. = Pursell Industries, a supplier for the manufacturers listed (located in
Orrville, Ohio). DMaterial was purchased by IFDC for this study. °This material had previously been found to contain 1.7 mg CIQ,~
g*' by EPA-ORD-NRMRL-WSWRD; no perchlorate salts were added to increase this value.
13
-------
Table 2.2. Summary results for perchlorate concentration detected by replicate analyses of samples listed in Table 2.1TI
Item
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
3D
Calif. Dept. of Food
& Agric.
[00.1 dilnt
pg g~1 1 to
u§
u
u
u
u
u
u
u
u
u
u
u
u
u
u
u
u
u
u
u
u
u
u
u
u
u
u
u
u
II
100
50
25
50
10
5
500
10
100
50
500
50
50
50
5
25
25
25
100
500
50
10
200
10
5
25
2
25
25
fi
NCSU Dept. of Soil
Science
[CIO/] diln
pg g-1 1 to
u
u
u
u
u
u
u
u
u
u
u
u
u
u
u
u
u
u
u
u
u
u
u
u
u
u
u
u
u
II
100
500
100
100
100
100
100
100
500
500
500
100
100
100
100
100
100
100
500
100
100
100
500
100
100
100
100
100
100
mn
American Pacific
Corporation
[00,1 diln
pg g-1 1 to
0.3*
u
2*
0.3*
0.6*
u
u
u
u
u
u
u
u
u
u
u
u
u
u
u
u
1.5*
u
1
u
0.4*
1.5*
0.3*
0.4*
II
10
100
50
10
10
1
10
10
100
1000
100
10
50
50
5
50
2
50
10000
10
50
50
50
10
1
10
50
10
10
?n
IMC Environ.
Laboratory
[CK)41 diln
tig g~1 1 to
u
u
u
u
u
u
u
u
u
u
u
u
u
u
u
u
u
u
LI
u
u
u
u
u
u
u
u
u
u
II
10
100
50
100
10
100
100
10
100
10
100
100
10
10
10
50
10
100
50
100
100
10
100
100
10
100
10
100
50
•m
Dionex Corporation
[CIO,1 diln
|ig g-1 1 to
u 100
u 100
u 100
u 100
u 100
u 100
u 100
u 100
u 100
u 100
u 100
14
-------
Table 2.2 continued
Item
31
31dJ
32
33
33d
34
34d
35
36
37
38
39
40
41
42
43
44
45
46
47
48
48d
49
50
51
52
53
54
55
Calif. Dept. of Food
& Agric.
[CICvl
M9JT1
u
u
u
3700
3925
u
u
2624
1860
u
u
u
u
u
2424
u
u
u
u
u
u
u
5720
5730
2540
1540
u
u
1550
dilnt
1to
100
500
250
4000
5000
5
5
4000
2000
5
1000
500
200
5
4000
100
2
50
25
250
1000
500
10000
5000
4000
2000
10
5
2000
NCSU Dept. of Soil
Science
[CICvl
Mgr1
u
u
u
4020
4270
u
u
2380
1950
u
u
u
u
u
2542
u
u
u
u
u
u
u
5960
5190
2700
1790
u
u
1580
diln
1to
500
500
500
5000
5000
50
10
2000
2000
100
500
500
500
100
2000
100
100
100
100
500
500
500
5000
5000
2000
2000
100
100
2000
American Pacific
Corporation
[CI041
M9£T1
u
u
u
4200
3700
0.9*
u
2100
1800
u
u
u
2.5*
u
2050
u
0.3*
u
u
u
26**
u**
6200
5100
2080
1430
0.5
u
1440
diln
1to
500
50
1000
10000
10000
10
50
5000
5000
10
1000
100
50
1
5000
100
10
50
5
100
100
1000
10000
10000
2000
1000
10
10
1000
IMC Environ.
Laboratory
[CI041
M9ST1
u
u
u
4066
4136
u
u
2288
2054
u
u
u
u
u
2455
u
u
u
u
u
u
u
6377
5737
2376
1511
u
u
1680
diln
1to
100
100
100
5000
5000
100
10
2000
2000
10
50
1000
100
10
2000
10
10
50
50
10
100
100
5000
5000
2000
1000
10
10
1000
Dionex Corporation
[CI04]
pgcr1
u
3860
3850
u
2240
1920
u
u
u
2340
u
u
u
6140
5430
2300
1810
1700
diln
1to
100
5000
5000
100
2000
2000
1000
1000
100
2000
100
1000
1000
10000
10000
10000
10000
2000
USample loops were 1000 uL exceptfor Dionex, which used 500 \iL. tDilution factor refers to subsequentdilution of a stock solution
prepared by leaching or dissolving the material at 10 g dL~1 (-10% w/w)4Samples 31, 33, 34, and 48 were supplied to the
laboratories as blind duplicates. §The designation "u" means undetected, i.e., no peak could be distinguished from the baseline.
*Values falls below the assured reporting level and therefore cannot be viewed as meaningful; see the method (EPA/600/R-01/026)
for details on how the ARL is determined. "See text for explanation.
15
-------
Table 2.3. Tested fortifications (spikes) and recoveries in solutions of samples listed in Tables 2.1-2.2
Item
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
Calif. Dept. of Food
& Agric.
A[CIO4'] recov
ugg-1 %
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
98
102
107
109.5
98
104
93
108
104
99
101
100
92
96
105
106
90
106
93
101
95
107
98
107
97
96
92
102
105
100
NCSU Dept. of Soil
Science
A[CICV] recov
M9JT1 %
10
50
50
10
10
10
50
30
50
50
50
10
50
50
10
50
10
50
50
50
50
50
50
10
10
10
10
30
10
10
101
98
93
93
87
98
85
96.5
101
101
95
88
89
88
96
86.5
96
94
97
88
91
81.5
99
84
96
97
92
94
95
95
American Pacific
Corporation
A[CIO4~] recov
M9ST1 %
10
10
10
10
10
10
10
10
10
10
10
10
10
20
10
10
10
10
10
10
10
10
50
10
10
10
10
10
10
10
93
90
103
97
97
93
106
93
103
93
109
91
92
90
91
95
108
97
107
106
96
95
92
106
99
106
93
94
91
101
IMC Environ.
Laboratory
A[CIO^ recov
M9ST1 %
200
200
200
200
200
200
200
200
200
200
200
200
200
200
200
200
200
200
200
200
200
200
200
200
200
200
200
200
200
57
85
100
102
98
89
100
103
87.5
96.5
97
99
104
80.5
88
117
99
93
97.5
97
102
103
85
95
103
93
102.5
107
98.5
105
96.5
Dionex Corporation
A[CICV] recov
ugg-1 %
10
10
10
10
10
10
10
10
10
10
10
107
103
96
101
98
96.5
108
106
110
106
110
16
-------
-------
Table 2.3 continued
Item
31
31dt
32
33
33d
34
34d
35
36
37
38
39
40
41
42
43
44
45
46
47
48
48d
49
50
51
52
53
54
55
Calif. Dept of Food
& Agric.
A[CI041
10
10
10
50
50
10
10
10
10
10
10
10
10
10
50
10
10
10
8
10
10
10
50
40
100
50
10
10
50
recov
100
95
97
90
102
100
102
109
107
94
95
101
94
93
108
96
94
91
104
103
110
92.5
104
95
93
105
102
101
103
NCSU Dept of Soil
Science
A[CI041
50
50
50
100
50
50
10
100
100
10
50
50
50
10
100
50
10
10
10
50
50
50
100
100
100
100
10
10
100
recov
95
97
97
100
107.5
81
97
98
98
99
108
95
98
98
99
93
91
91
94.5
99
99
105
97
109
111
97.5
92.5
100
100
American Pacific
Corporation
A[CI041
10
10
10
20
20
20
10
30
20
10
10
10
10
10
30
10
10
10
10
10
20
10
50
50
60
50
10
10
50
recov
91
94
93
95
103
106
102
104
105
102
100
92
93
103
98
109
93
102
105
92
102
103
95
93
105
91
91
109
94
IMC Environ.
Laboratory
A[CI041
200
200
400
57
57
57
200
57
57
57
200
200
200
200
57
200
200
200
200
200
200
200
20
57
20
20
200
200
57
recov
98
98
98
98
97
93.5
85
98.5
98
91
88
103
101
97
97
88.5
88.5
109
101
81
91
94
101
95
98
100
94
110
96
Dionex Corporation
A[CI041
10
100
100
10
100
100
10
10
10
100
10
10
10
50
40
20
20
100
recov
103
102
101
102
106
103
98
110
99
105
100
96
105
95
99
105
102
102
17
-------
Chapter 3
Implications for vascular plants
3.1. Introduction
In the laboratory setting, some plant species will absorb
perchlorate when exposed to perchlorate via irrigation water. This has
been explored for possible phytoremediation (Nzengung, 1999; 2000).
Some investigators have speculated that bacteria are responsible for
perchlorate reduction in plants. Perchlorate-reducing monera have been
identified by several laboratories, and cultured from a variety of sources
(including Las Vegas Wash sediments, food processing sludge, and
sewage sludge) (Logan, 1998; Coates, 1999; Coates, 2000). This
suggests that perchlorate-reducingbacteriaare active in the environment.
On the other hand, the bacteria isolated thus far prefer oxygen over
nitrate over perchlorate. In order for perchlorate reduction t o occur, the
water must be anoxic, and all of the nitrate must have been consumed.
Moreover, thesebacterial cultures require a suitably moist environment.
Arid soils or regions with low rainfall may not sustain their growth.
Natural attenuation probably varies around the nation, depending on
local factors; therefore, it is not possible to draw any conclusions about
the ecological impact of using fertilizers that contain perchlorate, for
they may be applied in areas where bacterial degradation occurs.
Due to the reported occurrence of perchlorate in certain water
resources and in certain fertilizer products, several groups have begun
to address the extent and significance of perchlorate uptake by plants.
For example, if produce is grown using perchlorate-tainted irrigation
water or fertilizers and the perchlorate is retained in the edible portions,
this might constitute a route of human exposure. The possibility of
exposure would be further increased if perchlorate were shown to
survive various types of processing. Unfortunately, experimental
results that definitively gauge the extent of risk from this route of
exposure have not yet been published. However, some significant
progress toward this goal has been made. Work is ongoing in this area,
and several projects are proceedingat the Instituteof Environmental and
Human Health of Texas Tech University.
3.2. Complicating factors
One problem with uptake studies is the possibility of convolved
influences on uptake. There are perchlorateabsorption dataavailable for
only a few species ofvascularplants. The absorption and accumulation
of anionic solutes can be affected by many physical and chemical
properties, such as concentration, size, charge density, and aquation.
3.2.1. Chemical influences on ion transport
Ion transport through plasma membranes occurs via transport
proteins (Raven, 1999). At present, there are no published reports on
the transmembrane transport of perchlorate in plants at the molecular
or cellular level. In the absence of studies specifically examining
influences on transmembrane perchlorate transport, it is necessary to
consider whether previously published or ongoinguptake studies have
accounted for likely influences.
Pertechnetate (TcO4~) is similar in size (ionic radius) and Gibbs
free energy of aquation (Moyer, 1997). In aerobic aquatic environments,
pertechnetate is highly mobile, accumulated by plants, and not
appreciably retained by many soils (Sheppard, 1991); therefore, it is
like perchloratein many respects. One respect in which it differs is its
ease of reduction to an immobile species, TcO2 (Tagami, 1996). Unlike
perchlorate, many papers have been published on the fate and transport
of pertechnetate in food plants due to its release after use in medical
imaging or from nuclear installations. Given the similarities between
C1O4 and TcO4 , those factors that influence pertechnetate uptake
should be either accounted for or ruled out in perchlorate uptake
studies. Depending on the plant, pertechnetate absorption and
accumulation can be affected by nitrate (Kriger, 2000; Echevarria, 1996;
1998), sulfate (Cataldo, 1978; 1983), and phosphate (Cataldo, 1978;
1983; Echevarria, 1996; 1998). No studies on perchlorate uptake have
attempted to account for the impacts of other ionic species.
3.2.2 Concentration
Active transport of perchlorate has not been reported. However,
both active and passive transport of an ion are affected by its
concentration (Raven, 1999). All ions experience an electrochemical
gradient which is determined in part by the ion's concentration. Past
and ongoing studies have exposed plants to much higher concentrations
than are present in the environment.
For example, Hutchinson and coworkers are presently studying
greenhouse-grown lettuce irrigated with perchlorate-tainted water.
Lettuce is of particular importance for assessingthe risk of perchlorate
to the food supply since much of the lettuce produced in the U.S. is
irrigated with perchlorate-tainted water. Also, lettuce has a high water
content and virtually the entire aboveground plant is consumed without
cooking or processing. Lettuce plants are watered with one of five
different concentrations of perchlorate (0.1,0.5, 1.0, 5.0, and 10.0 \ig
mL~') for a period of 90 days following planting. At various intervals
18
-------
of time, whole plants were harvested and divided into green tissue and
root samples; each sample was analyzed for perchlorate. The analytical
method was adapted from Ellington and Evans (2000). Levels of
perchlorate rise steadily over the first 50-60 days, and then generally
level off. The amount of perchlorate detected in the leaves correlates
with dose. As an example, at about 50 days into the experiment, the
lettuce irrigated with 10.0 ng mLT1 perchlorate exhibits a perchlorate
content of about 3 mgg~' on a lettuce dry matter basis. Since lettuce is
about 90% water, this would amount to about 0.30 mg g~' on a wet
mass basis. Experiments are underway todetermine whether lettuce has
the capability to degrade perchlorate if the supply of the contaminant
is stopped; however, initial data suggest that increases in biomass are
responsible for the apparent reduction in perchlorate content.
Therefore, a decline in concentration (e.g., expressed as mgg~') does not
adequately reflect the situation. While the preliminary results from
these studies (Hutchinson, 2000) represent progress in understanding
perchlorate in consumable produce, it must remembered that the
perchlorate concentrations are about 500 times the concentrations in
perchlorate-tainted irrigation water and that results obtained under
laboratory growingconditions cannot be directly extrapolated to edible
agricultural produce.
3.2.3. Tissue-specific accumulation
All studies of perchlorate occurrence in plants have focused on the
leaves, stalks, and wood. While the leaves of some plants (e.g., lettuce)
are consumed, the leaves of many others are not. Other than lettuce, no
studies have been conducted on absorption and accumulation in edible
fruits and vegetables. Therefore, information on uptake by Tamarix
ramosissima (Urbansky, 2000C), Salix (Nzengung, 1999; 2000)
tobacco (Wolfe, 1999) and Myriophyllum aquaticum (Susarla,
1999B; 1999C) cannot be directly applied or extrapolated to food
crops. While many ions are transported via the xylem, the phloem
supplies many of the nutrients to fruit. Although dissolved salts are
carried into the xylem alongwiththe water, the phloem relies on active
transport. Because the edible portions of many plants (e.g., tubers,
roots, some fruits) experience little to no transpiration, the
phloem—rather than the xylem—carries both inorganic and organic
materials for nutrition (Salisbury, 1992). For example, nitrate is almost
never present in phloem (Salisbury, 1992), and is rarely found in fruit.
It is worth pointing out that studies on pertechnetate uptake
showed that this anion was selectively accumulated in specific parts of
plants and generally not in the consumable tissues. Cataldo (1986),
Echevarria (1997), and Gast (1978) demonstrated that a variety of food
plants could absorb and accumulate pertechnetate, withmost of the ion
in the roots or leaves rather than edible portions. While TcO4~ was
shown to enter tomato leaf tissue via the xylem (Krijger, 1999), the
fruit—which is actually consumed—was never tested. Accordingly,
these observations demonstrate the importance of samplingconsumable
plant tissues and not just foliage. Furthermore, they warn against
drawing inferences about fruit (or other edible portions) based on
occurrence in foliage.
3.2.4. Soil sorption
It is generally accepted that perchlorate adsorbs to soil particles
through outer-sphere complexes where the ions engage in simple
electrovalent bonds and serve to balance electric charge on the surface
(Sparks, 1995; Sposito, 1989). Such adsorption is often influenced by
pH due to protonation of mineral oxo moieties, as is the case with
goethite (Gurol, 2000; Sasaki, 1983) or y-Al2O3 (Sasaki, 1983). Similar
outer-sphere behavior has been observed by others as well, especially
when perchlorate salts have been used to vary ionic strength and to
provide a competitor to probe adsorption of another anion (Bourg,
1978; Gisler, 1980; Hundal, 1994; Ji, 1992; Kummert, 1980; Rhue,
1990; Sadusky, 1991; Sigg, 1981; Zachara, 1988; Zhang, 1996). The
number of available binding/exchange sites is finite, but very large,
permitting soil to act as a reservoir for anions. Outer-sphere complex
formation is reversible and generally labile, but the sheer number of
active sites can lead to a buffering effect on aqueous ion concentration,
especially if the soil is far from saturated with the ion. This
phenomenon is further affected by ionic strength. Under high ionic
strength conditions (i.e., large concentrations of soluble and dissociable
salts), less adsorption is possible for any one particular ion. In the case
of diffuse ion swarms, where aquated counterions (anions in this case)
hover near the surface in an ion-pair like situation, association with the
positively charged surface is fleeting and constantly changing, with no
real effect on availability of the ion to the aqueous phase.
Even though soil scientists routinely use perchlorate salts as
indifferent electrolytes, it is possible that some soil types could resist
desorption. In addition, sorption phenomena in soil are often more
complicated than those in simpler (e.g., aqueous) chemical systems.
Despite the principle of microscopic reversibility, which requires that
the same pathway for adsorption be available for desorption, the
thermodynamics of the sorption equilibrium affect the activation energy
of the desorption process. Consequently, it is not unusual for desorp-
tion to occur via a different mechanism than the simple reverse of the
one for adsorption (Sparks, 1995). Sorption phenomena have not been
accounted for in any of the adsorption-accumulation studies conducted
thus far. Transport of ions through soil is also a complexprocess (Jury,
1992; Goldberg, 1992; Selim, 1992). It can be difficult to account for in
the laboratory setting due to the additional impacts of hydrologic,
meteorologic, and geomorphologic factors, which are location-
dependent. Moreover, ion exchange is often transport-controlled
(Sposito, 1994).
3.2.5. Summary
To be applicable, studies of perchlorate absorption and accumu-
lation must control for a variety of complicating factors. The presence
of other anions either from co-administration in fertilizers or background
salts present in the water supply may suppress uptake. Interspecies
variation in absorption mechanisms may lead to differing levels of
absorption and differing locations of accumulation. It is important to
know if accumulation occurs in fruits versus in leaves or as a result of
foliar application versus root application (as in irrigation).The rate of
harvesting may lead to different rates of uptake by disrupting normal
physiological processes in the plants. Lastly, the effect of soil
(primarily sorptive in nature) must be considered. Careful agronomic
studies are required to account for such influences, which are likely to
complicate studies on the impact of contaminated irrigation water, too.
19
-------
3.3. Difficulties in analysis
Another problem that has delayed accurate and definitive studies
of perchlorate uptake by edible plants is the difficulty of analyzing for
perchlorate in plant materials. Ion chromatography is currently the
recommended method for routine analysis of inorganic ions such as
perchlorate. It is a sensitive, reliable, and easily-implemented technique
when perchlorate occurs in a matrix that has a relatively low level of
total dissolved solids (IDS). Unfortunately, in a matrix with high IDS
(which largely equates to high ionic strength), other ions can swamp the
conductivity detector and effectively mask the signal from perchlorate.
This has proven to be a very difficult problem with extracts of plant
materials, many of which exhibit high TDS. In addition to potentially
high levels of other inorganic ions, extracts of plant materials typically
contain amino acids, other carboxylates or carboxylic acids (e.g., citrate,
ascorbate, fatty acids) sugars, and nucleotides; all of which contribute
to the ionic environment of the sample (Ellington, 2000). Ion chromato-
graphy is not alone in this regard. Other techniques and methods
suitable for reasonably dilute drinking water matrices (Urbansky,
2000C; Magnuson, 2000A; Magnuson, 2000B) cannot be readily
applied to fertilizers or botanical and physiological fluids. The problems
of trace ionic analysis have led to development of other methods that
rely on instrumentation that is common in high-end research
laboratories, such as asymmetric waveform ion mobility mass
spectrometry (Handy, 2000; Ells, 2000) or tandem MS-MS systems
(Koester, 2000). Another option used for eliminating dissolved matter
is selective preconcentration of the perchlorate on an anion-exchange
resin (personal communication from Baohua Gu).
Recently, Ellington and Evans (2000) have reported an IC-based
method for the analysis of perchlorate in plant materials, which greatly
reduces interferences from high TDS. Their method involves first
freeze-drying the plant material and then grindingit through a 30-mesh
screen. The ground material is then mixed with water and heated for 0.5
hours in a boiling water bath in order to saturate the dry material and to
precipitate proteins. The saturated samples are then shaken and stored
at 3°C overnight.Next, the samples arecentrifuged and the supernatant
is filtered. With most plant materials, this produces a very highly
colored solution that contains numerous inorganic and organic ions.
Prior to analysis, Ellington and Evans add these highly colored extracts
to alumina sorbent (DD-6) and allow them to stand for 20 hours at 3
°C. Finally, the extracts are filtered through an activated cartridge that
contains divinylbenzene. The highly colored extracts are rendered
colorless, and the ionic level is reduced dramatically. The minimum
reporting level (MRL) of perchlorate in lettuce and tomato was found
to be approximately 250 ng g~' on a wet mass basis.
Lettuce and tomato were chosen as representative plants because
they are considered high priority candidates for screening studies of
perchlorate in foodstuffs (Ellington, 2000) However, it should be noted
that in this work, native perchlorate was not detected in any produce,
nor was the method applied to any edible plants that were grown with
intentional exposure to perchlorate. Instead, perchlorate was spikedinto
the extraction water for one half of the duplicate freeze-dried samples,
while one half were extracted with pure water; consequently, sorptive
loss to plant tissue is not entirely precluded. Ellington and Evans (2000)
were able to evaluate the efficacy of the clean-up procedure for lettuce
and tomato in light of the impact on the known perchlorate
concentration in the spiked extracts. They observed no loss of
perchlorate in spite of the efficient removal of other inorganic (and also
organic) ions. This is tentatively attributed to preferential and
competitive adsorption on the alumina. This assertion is supported by
the observation that 40% of perchlorate is lost from solution when
exposed to alumina in the absence of other ions. However, loss of
perchlorate is not expected when the method is applied to plant material
becausemost extracts have high levels of TDS. Despite some remaining
uncertainties, the method of Ellington and Evans (2000) advances the
assessment of potential risk posed by eating produce grown in the
presence of perchlorate-tainted irrigation water or fertilizer.
3.4. Implications of perchlorate
absorption and accumulation
An obvious concern raised by finding measurable perchlorate
concentrations in plant tissues is whether this ion can affect food crops.
Most domestic crops are fertilized usingcommodity chemicals with no
known link to perchlorate contamination. Some crops (e.g., corn, wheat,
and rice) are fertilized with nitrogen fertilizers that should be
perchlorate-free because of the manufacturing processes. There is no
reason to suspect any perchlorate associated with growing grains.
The only crops with documented use of Chilean nitrate products
are tobacco and citrus fruits. Data on application of perchlorate-
containing fertilizers is sparse or nonexistent, and it is not possible to
estimate the ecological impact in any meaningful way. Modest
information is available on uptake by tobacco, but this is not a food
crop, and the use of Chilean nitrate salts appears to be locale-
dependent.
Depending on the season, sources of fresh and processed fruits and
vegetables vary considerably between domestic and imported. EPA has
no data regarding the reliance on perchlorate-containing fertilizers for
food crops grown outside the U.S. It is not known whether fruits or
vegetables absorb and retain perchlorate ion under typical growing
conditions (Cf. §3.2); accordingly, it is not possible to say whether
produce can serve as an exposure route at this time. There are no
published data on perchlorate in imported produce, no published data
on perchlorate in domestic produce, and minimal data from controlled
laboratory or field experiments on absorption and accumulation in food
crops; therefore, it is impossible to assess whether foodstuffs
contribute to perchlorate consumption in humans or whether drinking
water provides the entire body burden.
Even if many food plants can be shown to absorb and retain
perchlorate under some conditions, the primary source of this
contaminant is irrigation water. However, merely exposing the growing
plant to perchlorate does not imply absorption or accumulation in the
edible portions. In the studies on uptake by lettuce seedlings, the plants
have been subjected to perchlorate concentrations many times higher
than thoseencounteredin irrigation water. In addition, the confounding
20
-------
factors described previously and the harvest conditions have not been
accounted for.
Because aerospace industries, perchlorate manufacturers, and
military bases that use perchlorate salts are fairly localized
geographically, most of the country's agricultural products should be
free from exposure via tainted irrigation water. On the other hand, some
produce is largely supplied by regions that irrigate with Colorado River
water, which is known to contain perchlorate. Therefore, such produce
represents a potential exposure route for consumers. There are
currently no investigations underway toexamine food crops with docu-
mented exposure to perchlorate via irrigation or fertilization. In the
meantime, the gatheringof data by fertilizer manufacturers who choose
to screen their finished products as part of routine quality control
measures to prevent any recurrence of perchlorate contamination is
welcome. Some state or local authorities (e.g., agriculture departments
or state chemists' offices) may choose to conduct periodic screenings
in the course of their normal operations of assaying fertilizers to
validate the grade. Unfortunately, it is impossible to be more specific
since the source of the contamination previously observed has never
been identified, but appears to have reflected an episodic, if not isolated,
incident. There is inadequate evidence to suggest widespread or long-
term perchlorate contamination in fertilizers used for the bulk of
production farming operations nationwide. In fact,mostevidencepoints
to the contrary, but a modest level of continued vigilance would not be
inappropriate. While the likelihood of exposure via agricultural sources
is small due to the low consumption of Chilean nitrate salts and the low
perchlorate concentrations therein, the significance of whatever
exposure does occur is unknown in terms of food plant uptake or
ecological impact
21
-------
References
AAPFCO. Inspection Manual, 6th ed. Association of American Plant
Food Control Officials: Lexington, KY, 1999.
AOAC. Association of Official Analytical Chemists International.
Official Method 929.01. Sampling of Solid Fertilizers, 2000a.
AOAC. Association of Official Analytical Chemists International.
Official Method 929.02. Preparation of Fertilizer Test Sample,
2000b.
Bourg, A.C.M.; Schindler, P.W. Ternary surface complexes, 1. Complex
formation in the system silica-Cu(II)-ethylenediamine. Chimia,
1978, 32, 166^168.
Cataldo,D.A.; Garland, T.R.; Wildung, R.E. Plant root absorption and
metabolic fate of technetium in plants. In Technetium in the
Environment, G. Desmet and C. Myttenaere, Eds. Elsevier:
London, England, 1986; Chapter 22. ISBN 0-85334-421-3.
Cataldo, D.A.; Wildung, R.E.;Garland,T.R. Technetium accumulations,
fate, and behavior in plants. In Environmental Chemistry and
Cycling Processes, D.C. Adriano and I. Lehr Brisbin Jr., Eds.
Technical Information Center, U.S. Department of Energy:
Springfield, VA, 1978; pp. 538-549. Available from National
Technical Information Service as CONF-760429. ISBN 0-87079-
302-0.
Cataldo, D.A.; Wildung, R.E.; Garland, T.R. Root absorption and
transport behavior of technetium in soybean. Plant Physiology,
1983, 73, 849-852.
Chilean Nitrate Corporation (CNC) website. 1999. URL:
http://www.cncusa.com.
Clark, J.J.J. Perchlorate toxicology. In Perchlorate in the Environ-
ment, E.T. Urbansky, Ed. Kluwer/Plenum: New York, NY, 2000;
Chapter 3, and references cited therein.
Coates, J.D.; Michaelidou, U.; Bruce, R.A.; O'Connor, S.M.; Crespi,
J.N., Achenbach, L. A. The ubiquity and diversity of dissimilatory
(per)chlorate-reducing bacteria. Applied and Environmental
Microbiology, 1999, 65, 5234-5241, and references cited
therein.
Coates, J.D.; Michaelidou, U.; O'Connor, S.M.; Bruce, R.A.;
Achenbach, L.A. The diverse microbiology of (per)chlorate
reduction. In Perchlorate in the Environment, E.T. Urbansky,
Ed. Kluwer/Plenum: New York, NY, 2000; Chapter 24, and
references cited therein.
Collette, T.W.; Williams, T.L. Perchlorate in fertilizers? Analysis by
Raman spectrometry. Paper presented before the Division of
Fertilizer and Soil Chemistry at the American Chemical Society
220th National Meeting, Washington, D.C., August 20-24, 2000.
Collette, T.W.; Robarge, W.P.; Urbansky, E.T. Ion chromatographic
determination of perchlorate: analysis of fertilizers and
related materials. U.S. Environmental Protection Agency:
Cincinnati, OH, 2001. EPA/600/R-01/026.
Echevarria, G. Contribution a la prevision des transferts sol-
plante des radionucleides. Ph.D. thesis. 1'Institut National
Polytechnique de Lorraine: Nancy, France, 1996.
Echevarria, G.; Vong, P.C.; Leclerc-Cessac, E.; Morel, J.L.
Bioavailability of technetium-99 as affected by plant species and
growth, application form, and soil incubation. Journal of
Environmental Quality, 1997, 26, 947-956.
Echevarria, G.; Vong, P.C.; Morel, J.L. Effect of NO3~ on the fate of
99TcO4~ in the soil-plant system. Journal of Environmental
Radioactivity, 1998, 38, 163-171.
Eldridge, J.E.; Tsui, D.T.; Mattie, D.R.; Crown, J.; Scott, R.
Perchlorate in Fertilizers. U.S. Air Force Research Labora-
tories: Wright Patterson AFB, OH, 2000. Report in press.
Ellington, J.J.; Evans, J.J. Determination of perchlorate at parts-per-
billion levels in plants by ion chromatography. Journal of
Chromatography A, 2000, 898, 193.
Ellington J.J.; Wolfe, N.L.; Garrison, A.W.; Evans, J.J.; Avants, J.K.;
Teng, Q. Analysis of perchlorate in tobacco plants and tobacco
products, Environmental Science & Technology, 2001.
Submitted for publication.
Ells, B.; Barnett, D.A.; Purves, R.W., Guevremont, R. Trace level
23
-------
determination of perchlorate in water matrices and human urine
using ESI-FAIMS-MS. Journal of Environmental Monitoring
2000, 2, 393-397.
EPA, Apache Powder Superfund Site: Status of Apache Cleanup
Activities. St. David, Arizona. U.S. Environmental Protection
Agency (Region 9): San Francisco, CA: May 1999.
Ericksen, G.E. The Chilean nitrate deposits. American Scientist,
1983, 71, 366-374, and references cited therein.
Gast,R.G.;Thorvig, L.H.; Landa, E.R.; Gallagher, K.J. Technetium-99
toxicity to plants and sorption by soils. In Environmental
Chemistry and Cycling Processes, D.C. Adriano and I. Lehr
Brisbin Jr., Eds. Technical Information Center, U.S. Department
of Energy: Springfield, VA, 1978; pp. 550-567. Available from
National Technical Information Service as CONF-760429. ISBN
0-87079-302-0.
Gisler, A. Die Adsorption von Aminosauren an Granzftachen
Oxid-Wasser. Ph.D. Thesis. University of Bern: Bern,
Switzerland, 1980.
Goldberg, S. Use of surface complexation models in soil chemical
systems. Advances in Agronomy, 1992, 47, 234—330.
Gu,B.; Brown, G.M.; Urbansky,E.T.; Magnuson, M.L. Determination
of perchlorate in fertilizers by ion chromatography. Paper
presented before the Division of Fertilizer and Soil Chemistry at
the American Chemical Society 220th National Meeting,
Washington, D.C., August 20-24, 2000.
Gullick, R.W.; LeChevallier, M.W.; Barhorst, T.S. Occurrence of
perchlorate in drinking water sources. Journal of the American
Water Works Association, January 2001, 93 (1), 66-77.
Gurol, M.; Kim, K. Investigation of perchlorate removal in drinking
water sources by chemical methods. In Perchlorate in the
Environment, E.T. Urbansky, Ed. Kluwer/Plenum: New York,
NY, 2000; Chapter 10.
Handy, R.; Barnett, D.A.; Purves, R.W.; Horlick, G., Guevremont, R.
Determination of nanomolar perchlorate in water by ESI-F AIMS-
MS. Journal of Analytical Atomic Spectrometry, 2000, 15,
907-911.
Harvey, G.J.; Tsui, D.T.; Eldridge, J.E.; Orris, G.J. 20th Annual
Meeting Abstract Book; Society of Environmental and
Toxicological Chemists, 1999. Abstract PHAOI5, p. 277.
Hundal, L.S.; Pasricha, N.S. Thermodynamic parameters of potassium
exchange as characterized by equilibrium and kinetic approaches in
chloride and perchlorate background anions. Soil Science, 1994,
157, 356-364.
Hutchinson, S.L.; Susarla, S.; Wolfe, N.L.; McCutcheon, S.C.
Perchlorate accumulation from contaminated irrigation water and
fertilizer in leafy vegetation, Second International Conference on
Remediation of Chlorinated and Recalcitrant Compounds, May
22-25,2000, Monterey, CA.
Jackson, P.E.; Laikhtman, M.; Rohrer, J. Determination of trace level
perchlorate in drinking water and ground water by ion
chromatography. Journal of Chromatography A, 1999, 850,
131-135.
Jackson, P.E.; Gokhale, S.; Rohrer, J.S. Recent developments in the
analysis of perchlorate using ion chromatography. In Perchlorate
in the Environment, E.T. Urbansky, Ed. Kluwer/Plenum: New
York, NY, 2000; Chapter 5.
Ji, G.; Kong, X. Adsorption of chloride, nitrate, and perchlorate by
variable charge soils. Pedosphere, 1992,2,317-326.
Jury, W.A.; Fluhler, H. Transport of chemicals through soil:
mechanisms, models, and field applications. Advances in
Agronomy, 1992, 47, 142-202.
Koester, C.J.; Seller, H.R.; Halden, R.U. Analysis of perchlorate in
groundwater by electrospray ionization mass spectrometry/mass
spectrometry. Environmental Science & Technology, 2000,
54,1862-1864.
Krijger, G.C.; Kolloffel, C.; Wolterbeek, H.T. Effect of nitrate on
uptake of pertechnetate by tomato plants. Journal of
Environmental Quality, 2000, 29, 866-870.
Krijger, G.C.; Van Vliet, P.M.; Wolterbeek, H.T. Metal speciation in
xylem exudate of Lycopersicon esculentum Mill — technetium.
Plant and Soil, 1999, 212, 165-173.
Kummert,R.; Stumm, W. The surface complexation of organic acids on
hydrous y-A[2Oy Journal of Colloid Interface Science,
1980, 75, 373 385.
Lemon, J.F. Jr. Nitrogen. U.S. Geological Survey Minerals Infor-
mation. 1996. URL: http://minerals.usgs.gov/minerals/
pubs/cornmodity/nitrogen/480496.pdf.
Logan, B.E. A review of chlorate- and perchlorate-respiring
microorganisms. Bioremediation Journal, 1998, 2, 69-79, and
references therein.
Magnuson, M.L.; Urbansky, E.T.; Kelty, C.A. Determination of
perchlorate at trace levels in drinking water by ion-pair extraction
with electrospray ionization mass spectrometry. Analytical
Chemistry, 2000A, 72, 25-29.
Magnuson, M.L.; Urbansky, E.T.; Kelty, C.A. Microscale extraction
of perchlorate in drinking water by ion-pair extraction with low-
24
-------
level detection by electrospray-mass spectrometry. Talanta
20008,52,285-291.
Moyer, B.A.; Bonnesen, P.V. Physical factors in anion separations. In
Supramolecular Chemistry of Anions, A. Bianchi, K.
Bowman-James, and E. Garcia-Espana, Eds. Wiley-VCH: New
York, NY, 1997; Chapter 1. ISBN 0-471-18622-8.
Nzengung, V.A.; Wang, C.; Harvey, G. Plant-mediated transformation
of perchlorate into chloride. Environmental Science &
Technology, 1999,33, 1470-1478.
Nzengung, V.A.; Wang, C. Influences on phytoremediation of
perchlorate-contaminated water. In Perchlorate in the Environ-
ment, E.T. Urbansky, Ed. Kluwer/Plenum: New York, NY, 2000;
Chapter 21. ISBN 0-306-46389-X.
Prud'homme, M. Potash. Natural Resources Canada. 1998. URL:
http://www.nrcan.gc.ca/mms.cmy/content/1998/46.pdf.
Raven, P.H.; Evert, R.F.; Eichorn, S.E. Biology of Plants, 6th ed.
Freeman: New York, NY, 1999; Chapter 4.
Renner, R. Study finding perchlorate in fertilizer rattles industry.
Environmental Science & Technology, 1999, 33, 394A-
395 A.
Rhue, R.D.; Reve, W.H. Exchange capacity and adsorbed-cation charge
as affected by chloride and perchlorate. Soil Science Society of
America Journal, 1990, 54, 705-708.
Robarge, W.P.; Duffera, M.; Ramirez, G. Evaluation of the composition
of fertilizer source materials: perchlorate. Paper presented before
the Division of Fertilizer and Soil Chemistry at the American
Chemical Society 220th National Meeting, Washington, D.C.,
August 20-24, 2000.
Sadusky, M.C.; Sparks, D.L. Anionic effects on potassium reactions in
variable-charge Atlantic coastal soils. Soil Science Society of
America Journal, 1991, 55, 371-375.
Salisbury, F.B.; Ross, C.W. Plant Physiology, 4th ed. Wadsworth:
Belmont, CA, 1992; Chapter 1.
Sasaki, M.; Moriya, M.; Yasunaa, T.; Astumian, R.D. A kinetic study
of ion-pair formation on the surface of ot-FeOOH in aqueous
suspensions using the electric field pulse technique. Journal of
Physical Chemistry, 1983, 87, 1449-1453. Also unpublished
data on Y-M2O3 cited in Hayes, K.F. Equilibrium,
spectroscopic, and kinetic studies of ion adsorption at the
oxide/aqueous interface. Ph.D. Thesis. Stanford University:
Stanford, CA, 1987 (available through University Microfilms
International, #8723014).
Schilt, A.A. Perchloric Acid and Perchlorates. GFS Chemicals,
Inc.: Columbus, OH, 1979, p. 35, and references therein.
Searls, J.P. Potash—1999. U.S. Geological Survey Minerals
Yearbook. 1999. URL: http://minerals.usgs.gov/minerals/
pubs/commodity/potash560499.pdf
Selim, H. M. Modeling the transport and retention of inorganics in soils.
Advances in Agronomy, 1992, 47, 331-384.
Sheppard, S.C.; Evenden, W.G. Can aquatic macrophytes mobilize
technetium by oxidizing their rhizosphere? Journal of
Environmental Quality, 1991, 20, 738-744, and references
cited therein.
Sigg, L.; Stumm, W. The interaction of anions and weak acids with the
hydrous goethite (ct-FeOOH) surface. Colloids and Surfaces,
1981,2,101-107.
Sparks, D.L. Environmental Soil Chemistry. Academic Press: San
Diego, CA, 1995; Chapter 5.
Sposito, G. The Chemistry of Soils. Oxford University Press: New
York, NY, 1989, Chapters 7-8.
Sposito, G. Chemical Equilibria and Kinetics in Soil. Oxford
University Press: New York, NY, 1994, Chapters 4-5.
Susarla, S.; Bacchus, S.T.; McCutcheon S.C.; Wolfe, N.L. Potential
Species for Phytoremediation of Perchlorate. U.S.
Environmental Protection Agency: Athens, GA, August 1999A.
EPA/600/R-99/069.
Susarla, S.; Bacchus, S.T.; Wolfe, N.L.; McCutcheon, S.C.
Phytotransformation of perchlroateand identification of metabolic
products in myriophyllum aquaticum. International Journal of
Phytoremediation, 1999B, /, 97.
Susarla, S.; Bacchus, S.T.; Wolfe, N.L.; McCutcheon S.C.
Phytotransformation of perchlorate using parrot-feather. Soil and
Ground-water Cleanup, February/March 1999C, 20.
Susarla, S.;Collette,T.W.; Garrison, A.W.; Wolfe, N.L.; McCutcheon,
S.C. Perchlorate identification in fertilizers. Environmental
Science & Technology, 1999D, 33, 3469- 3472.
Susarla, S.; Collette, T.W.; Garrison, A.W.; Wolfe, N.L.; McCutcheon,
S.C. Correction to perchlorate identification in fertilizers.
Environmental Science & Technology, 2000,34,224.
Tagami, K.; Uchida, S. Microbial role in immobilization of technetium
in soil under waterlogged conditions. Chemosphere, 1996, 33,
217-225.
Urbansky, E.T.; Magnuson, M.L.; Kelty, C.A.; Gu, B.; Brown, G.M.
Comment on "Perchlorate identification in fertilizers" and the
25
-------
subsequent addition/correction. Environmental Science &
Technology, 2000A, 34,4452-4453.
Urbansky, E.T.; Magnuson, M.L.; Kelty, C.A.; Brown, S.K.
Perchlorate uptake by salt cedar (Tamarix ramosissima) in the
Las Vegas Wash riparian ecosystem. The Science of the Total
Environment, 2000B, 256, 227-232.
Urbansky, E.T.; Gu, B.; Magnuson, M.L.; Brown, G.M.; Kelty, C.A.
Survey of bottled waters for perchlorate by electrospray
ionization mass spectrometry (ESI-MS) and ion chromatography
(1C). Journal of the Science of Food and Agriculture,
2000C, 80, 1798-1804.
Urbansky, E.T.; Magnuson, M.L.; Gu, B.; Brown, G.M.; Kelty, C.A.
Development of an electrospray mass spectrometric method for
determining perchlorate in fertilizers. Paper presented before the
Division of Fertilizer and Soil Chemistry at the American Chemical
Society 220th National Meeting, Washington, D.C., August
20-24, 2000D.
Urbansky, E.T.; Kelty, C.A.; Brown, S.K.; Magnuson, M.L.
Distribution of perchlorate in samples of sodium nitrate (Chile
saltpeter) fertilizer derived from natural caliche. Paper presented
before the Division of Fertilizer and Soil Chemistry at the
American Chemical Society 220thNational Meeting, Washington,
D.C., August 20-24, 2000E.
Urbansky,E.T.Quantitation of perchlorate ion: Practices and advances
applied to the analysis of common matrices. Critical Reviews in
Analytical Chemistry, 2000F, 30, 311-343.
Urbansky, E.T.; Brown, S.K.; Magnuson, M.L.; Kelty, C.A.
Perchlorate levels in samples of sodium nitrate fertilizer derived
from Chilean caliche. Environmental Pollution, 2001, 112,
299-302.
Williams, T.L.; Martin, R.B.; Collette, T.W. Raman spectroscopic
analysis of fertilizers and plant tissue for perchlorate. Applied
Spectroscopy, 2001, in press.
Wolfe, N.L.; Ellington J.J.; Garrison, A.W.; Evans, J.J.; Avants, J.K.;
Teng, Q. Accumulation of perchlorate in tobacco plants and
tobacco products. Paper presented before the Division ofFertilizer
and Soil Chemistry at the American Chemical Society 220th
National Meeting, Washington, D.C., August 20-24,2000.
Zachara, J.M.; Cowan, C.E.; Schmidt,R.L.; Ainsworth, C.C. Chromate
adsorption by kaolinite. Clays and Clay Minerals, 1988, 36,
317-326.
Zhang, G.Y.; Brummer, G.M.; Zhang, X.N. Effect of perchlorate,
nitrate, chloride and pH on sulfate adsorption by variable charge
soils. Geoderma, 1996, 73,217-229.
26
------- |