the Science of the
                                                                                  Total Environment
ELSEVIER                The Science of the Total Environment 256 (2000) 227-232                           -
                                                                            www.elsevier.com/locate/scitotenv

                                      Rapid communication

      Perchlorate  uptake  by salt cedar  (Tamarbc ramosissima}

                in the Las Vegas Wash riparian ecosystem


         Edward T. Urbansky*, Matthew L. Magnuson, Catherine A. Kelty,
                                      Stephanie K. Brown
      United States Environmental Protection Agency, National Risk Management Research Laboratory, Water Supply and Water
                    Resources Division, 26 West Martin Luther King Drive, Cincinnati, OH 45268, USA

                               Received 24 March 2000; accepted 25 March 2000
Abstract

  Perchlorate ion (C1OJ) has been identified in samples of dormant salt cedar (Tamarix ramosissima) growing in the
Las Vegas Wash. Perchlorate is an oxidant, but its reduction is kinetically hindered. Concern over thyroid effects
caused the Environmental Protection Agency  (EPA) to add  perchlorate  to the  drinking  water  Contaminant
Candidate List (CCL). Beginning in 2001, utilities will look for perchlorate under the Unregulated Contaminants
Monitoring Rule (UCMR). In wood samples acquired from the same plant growing in a contaminated stream,
perchlorate concentrations were found as follows: 5-6 u-g g"1  in dry twigs extending above the water and 300 |xg g~'
in stalks immersed in the stream. Perchlorate was leached from samples of wood, and the resulting solutions were
analyzed by ion chromatography after clean-up. The identification was confirmed  by electrospray ionization mass
spectrometry after complexation of perchlorate with decyltrimethylammonium cation. Because salt cedar is regarded
as an invasive species, there are large scale programs aimed at eliminating it. However, this work suggests that salt
cedar might play a role in the ecological distribution of perchlorate as an environmental contaminant. Consequently,
a thorough investigation of the fate and transport of perchlorate in tamarisks is required to assess  the effects that
eradication might have on perchlorate-tainted riparian ecosystems, such as the Las Vegas Wash. This is especially
important since water from the wash enters Lake Mead and the Colorado River and has  the potential to affect the
potable water source of tens of millions of people as well  as irrigation water used on a variety of crops, including
much of the lettuce produced in the USA. © 2000 Elsevier Science B.V. All rights reserved.

Keywords:  Perchlorate; Salt cedar; Tamarix; Riparian ecosystem; Drinking water; 1C; ESI-MS; Eradication
  * Corresponding author. Tel.: +1-513-569-7655; fax: + 1-513-569-7658.
  E-mail address: urbansky.edwardlS'epa.gov. (E.T. Urbansky).

0048-9697/00/S - see front matter  2000 Elsevier Science B.V. All rights reserved.
PII: 30048-9697(00)00489-7

-------
 228
                    E.T. Urbansky et al. / The Science of the Total Environment 256 (2000) 227-232
 1. Introduction

   Salt cedar (Tamarix ramosissima) is  a woody
 plant, indigenous to Asia now found as an inva-
 sive  species in western riparian  (streambank)
 ecosystems  (Westbrooks,  1998).   Southwestern
 states experience the greatest effects due to the
 favorable climate (Muzika and Swearingen, 1999)
 with  more than a million acres of land  occupied
 (Deuser et al., 1998). T. ramosissima accumulates
 salt, which it then secretes through its leaves. This
 brackifies the nearby soil to the extent that native
 genera (e.g. Salix)  cannot  grow  (Muzika and
 Swearingen, 1999). Because of the serious impact
 on native flora, projects to eradicate salt  cedar
 have  been undertaken by the Natural Park Ser-
 vice (Deuser et al.,  1998). Tamarisks have colo-
 nized much of the land surrounding the Las Ve-
 gas Wash and can be found growing in shallow
 surface waterways, such as creeks or springs. The
 plant consumes and transpires incredible amounts
 of water so that it can bring about fluctuations in
 the local water table when  it  becomes active,
 roughly from March  to December.
  Perchlorate has  been identified as a contami-
 nant  in the soil, surface water, and  ground water
 of the Las Vegas Wash (Urbansky, 1998; Renner,
 1999). The contamination in this region has been
 attributed to the production  of  ammonium per-
 chlorate  by defense  and aero-space contractors
 dating back several  decades  (Damian and Pon-
 tius,  1999).  Perchlorate's  nature  makes  both
 potable water  treatment and site   remediation
 difficult (Urbansky and Schock,  1999; Espenson
 2000). Perchlorate affects the thyroid (Wolff, 1998;
 Clark, 2000), and was placed on the  Contaminant
 Candidate List (CCL) (Environmental Protection
Agency, 1998; Perciasepe, 1998) and the  Unregu-
lated  Contaminants  Monitoring Rule  (UCMR)
(Browner, 1999). Water from the Las Vegas Wash
flows  into Lake  Mead and becomes part of the
Colorado River.  Lake Mead and  the downstream
portion of Colorado River serve as the major
source of potable water for southern California
(including Los Angeles), southern Nevada (greater
Las Vegas),  and parts of Arizona. Consequently,
the drinking water of tens of millions of people is
potentially  at  risk.   Moreover,  Colorado  River
 water and other contaminated waters are used to
 irrigate farms that supply significant portions of
 the produce consumed in the USA. Therefore,
 this contamination has the potential to affect the
 nation's food supply if perchlorate is absorbed by
 food plants.
   Because salt cedar influences the  hydrology of
 the Las Vegas  Wash and perchlorate  salts  are
 very water-soluble, it seems reasonable to con-
 sider whether salt cedar can take up perchlorate
 along with other salts.
 2. Experimental section


   Sections of tamarisk twigs, branches, and stalks
 ranging from 2 to 12 mm were collected from
 plants growing in the Las  Vegas Wash. Samples
 were taken from sections of plant submerged in
 the water and from sections extending above the
 surface of the water. A mass of 80 g (wet mass) of
 the  submerged  stalks  (3-12 mm diameter) was
 soaked in running deionized (DI) water for 2 min
 and then rinsed with a spray of DI water. A mass
 of 20 g (diy mass) of the exposed twigs (2-7 mm
 diameter) was similarly cleaned. Branch sections
 were shredded in a blender; and leached at 5°C
 for ~ 60 h; the plant matter to water ratio was 20
 g  dl~'. The mixture was  then suction  filtered
 through Corning (Corning, NY,  USA) 0.45-(jum
 cellulose acetate with a glass fiber pre-filter.
  Ion chromatography (1C) standards were  pre-
 pared in DI water using sodium perchlorate (GFS
 Chemicals, Columbus,  OH, USA). A stock solu-
 tion containing 1000  jig  ml"'  was  diluted as
 needed for spiking,  constructing  a calibration
 curve, and  determining retention time. The  col-
 lected shredded wood  filtrates prepared at 20 g
 df1  were diluted at a ratio of 1:6 v/v with DI
water. The diluted solution from the submerged
branches was cleaned up  by passing a  5.0-ml
aliquot through either Supelco (Bellefonte,  PA,
USA) Envi-Carb (carbon),  Envi-Chrom P (poly-
styrene), or Waters (Milford, MA, USA) Sep-Pak
CliS cartridge.
  A Dionex  DX300 (Sunnyvale.  CA, USA)  1C
(500-(jil loop) equipped  with  AG11 and ASH

-------
                     E.T. Urbansky et al. / The Science of the Total Environment 256 (2000) 227-232
                                             229
columns (4 mm diameter) was used with isocratic
0.10 M NaOH (aq.) at 1.00 ml min~'  with sup-
pressed  conductivity  detection  (Jackson  et al.,
1999; Wirt et al., 1998).
  To prepare a sample for electrospray ionization
mass spectrometry (ESI-MS), a 5.0-ml aliquot of
dry twig 20 g dl~' filtrate was run through six C1S
cartridges. The 1C  was operated  in preparative
mode (2.00-ml  loop). The eluate was collected
from 7.1  to  9.0 min. This collected fraction of
eluate was  reinjected in its  entirety using the
same sample loop, and the peak at 8 min (C1O^~)
was collected.
  Complexation ESI-MS has  been applied to
water samples (Magnuson et al., 2000a,b; Urban-
sky and Magnuson, 2000; Urbansky et al., 2000).
Perchlorate forms a stable ion pair with a quater-
nary ammonium cation. This complex is extracted
into CH2C12  and injected without separation. The
following were added to the collected fraction: 50
(jul  of 0.20  M  C10H21N(CH3)3Br(aq.)  solution
(Fluka, Buchs, Switzerland) and 300 fjul of CH,C1,
(J.T. Baker Ultra Resi-analyzed, Phillipsburg, NJ,
USA). After  extraction,  the  CH2C12 phase  was
drawn off with a syringe.  Aliquots of 25 JJL!  were
injected  into the ESI-MS system  as  described
(Magnuson et al., 2000a,b)
3. Results and discussion

   Chromatograms for preparative and analytical
1C are shown in  Fig. 1. Negative  ion ESI-MS
signals  at  m/z 380 and 400  jjum correspond
to  C10H21NMe3(Br)(ClO4r  and C]0H21NMe3-
(C1O4)2, respectively. Peaks for the  two ions are
shown in Fig. 2. The average signal-to-blank ratio
(S/B)  using the  sum, (AX(}u + A4mj,  was  4.2.
Even separately, (5/6)380u = 4.9 and (S/B\Wu =
2.1, either of which confirms the identification.
   The 1:6 v/v dilution of the 20-g dl~' filtrate of
the submerged stalks was found to contain 10 jxg
ml ~'  by 1C, while that of the dry twigs was found
to contain  200 ng ml"1.  These  concentrations
should be viewed as minima. There is no standard
material by which to gauge the  effectiveness of
the leaching  procedure;  therefore,  this prelimi-
nary finding is limited by the state of the science.
a
                       retention time, min
                                      perchiorata
                       456
                     retention time, m!n
Fig. 1. (a) Preparative ion chromatograph of the 20 g dl~
filtrate of dormant, dry (above water level) Taniarix mmosis-
sinia twigs. Peak 5 is perchlorate. Injection was made using a
2-ml sample loop. The fraction from 7 to 9 min was collected
and reinjected for analysis, (b) Analytical ion chromatograph
of the 7-9-min collected fraction of eluate from Fig. 1. Peak 4
is the  perchlorate peak.  Injection was made using a 2-ml
sample loop. The fraction from 7.9 to 9.2 min was collected.
The identification of perchlorate was confirmed by complexa-
tion electrospray ionization mass spectrometry (cESI-MS). The
perchlorate was extracted into dichloromethane using decyl-
trimethylammonium  cation (see Fig. 2 and text for more
details).
No difference was observed between the clean-up
cartridges, suggesting that perchlorate is  not re-
tained.
  Submerged samples contained  300 pug  g"1,
while exposed samples contained 5-6  (Jig g"  .It
is difficult to draw a conclusion  about rate and
selectivity of  uptake  in T. ramosissima, but it is
clear that immersed stalks  can  absorb/adsorb
significant amounts.  Perhaps this  could  be  ex-
ploited   for   remediation.  Perchlorate-reducing
microorganisms,  especially  facultative  anaerobic
bacteria  are  readily  cultured in the  laboratory
(Coates  et al., 2000; Giblin  et al.,  2000;  Logan,

-------
 230
                     E. T. Urbansky et al. / Tlie Science of the Total Environment 256 (2000) 227-232
        m/z: 380
so -
40 ~
20 -
0 -J
9.59
HH 608
HA 16658
11.51
MH 584
MA 20324
13.87
MH 631
MA 17321

10.0
m/z: 400


»!,'

100
80 -
60 -
40 •
20 -
0-

9.54
MH 230
MA 5654


11.46
HK 225
HA 6991


13.97
HH 214
HA 7946

                 10.0
                                        15.0
         23.77
 ,, „-   HH 8674
MH  6194   «* S8908
HA  78964
                   25.93
                  MH 875B
                  MA 99S15
            2B.20
           HH 7578
           MA 89003
 22.09    23.77
MH 1270   MH 1073
MA 16367  HA 15292
 25.93
HH 998
HA 14860
                                                                               28.20
                                                                             KB 919
                                                                             HA 14167
                                                                                        30.11
                                                                                       KH 6634
                                                                                       HA 80230
                                                                                       30.0
                                       30.36
                                      MX 740
                                      HA 130B1
                                                                                      30.0
Fig. 2. Confirmation of perchlorate by negative ion complexation ESI mass spectrum obtained by extracting peak 4, which was
collected from eluate in  Fig. 2, with a cationic surfactant (decyltrimethylammonium bromide) into dichloromethane. Both the
bromoperchiorato and bis(perchlorato) complex anions are observed in the mass spectrum. Ci0H2iNMe3(BrXClO4)~ has m/z - 380
u; C10H21NMe3(ClO4)f has m/z = 400 u. The first three injections are the blank (extract of 1C eluate of just eluent after going
through the suppressor). The last five injections are the dichloromethane extract of peak 4. Injection volume was 25 jxl.
1998). Phytoremediation of perchlorate by willows
(genus Salix) has been reported by Nzengung and
Wang (1998). Because significant incorporation of
perchlorate  would  be  expected with  long-term
rather than short-term or one-time exposure, diel
fluctuations  in  concentrations may be averaged
out, providing a more  uniform measure of expo-
sure as a biological indicator.
4. Conclusions
  For better or worse, salt  cedar must  now be
regarded as a major part of the ecosystem in the
American  Southwest  and certainly  in the  Las
Vegas  Wash.  Accordingly,  plans  to  eradicate
Tamarix must be evaluated in terms of total eco-
logical impact, including effects on aqueous phase
pollutants, such as perchlorate.  It is not possible
at this time to state what effect Tamarix has on
perchlorate distribution in the Las Vegas Wash.
Only two things are known for certain: Tamarix
consumes  an enormous  amount of water  and
Tamarix can take up perchlorate.
       Several lines of research are suggested by these
     facts. It may be possible to use Tamarix to con-
     centrate  the highly soluble perchlorate salts  as
     part of a remediative strategy at some sites or to
     accumulate  it in the wood. On the other hand,
     high rates of transpiration in salt cedar may make
     the perchlorate concentrations in ground water
     higher  if perchlorate ion is selectively excluded
     when water  is taken into the root system. Conse-
     quently, it will be  important to answer the  fol-
     lowing  questions: (1) Do tamarisks absorb all  of
     the perchlorate along with the water they take in?
     (2) Do tamarisks secrete perchlorate salts on their
     leaves or retain them within the plant? (3) Can
     tamarisks harbor microbes in the rhizosphere that
     are  capable  of  reducing  perchlorate?  (4)  Do
     tamarisks  metabolize  (reduce) perchlorate   to
     other oxyanions of chlorine or to  chloride?  (5)
     Can perchlorate levels found in  tamarisk wood be
     used to assess long-term contamination of ripar-
     ian ecosystems?
      The nature, rate, and  selectivity of perchlorate
     uptake  and  processing  by salt cedar are all  un-
     known, therefore, we cannot say whether this
     plant is improving or reducing water quality. Fur-
     thermore, without accurate information on  the

-------
                       E. T. Urbansky et al. / The Science of the Total Environment 256 (2000) 227-232
                                                 231
transport and fate of perchlorate in Tamarix, the
impact of eradication  programs on perchlorate
cannot be evaluated. Many programs rely on herb-
icides;  however, any perchlorate in the  plant is
likely to be returned to the soil and ground water
as the  plant deteriorates and decomposes.  This
could potentially compound a problem if the salt
cedar acts as a reservoir and  the perchlorate is
released  upon the death  of the plant. In order to
ensure the protection  of native species  and the
waterways and sound environmental management
practices, it is necessary to completely understand
the  relationship among source water, salt  cedar,
and perchlorate in  ecosystems such as  the Las
Vegas Wash.
Acknowledgements
   We acknowledge Professor Jacimaria R. Batista
(University of Nevada-Las  Vegas) for assistance
in procuring  samples of salt cedar and Brenda
Pohlmann  (Nevada  Division  of  Environmental
Protection) for  helpful discussions. Mention  of
specific brand names or  models  should  not be
construed to suggest endorsement or recommen-
dation by  the US government. This  paper was
produced by  US  government employees in the
course of  their  official  duties and is,  therefore,
exempt from copyright.
References
Browner  C. Part II. Environmental Protection Agency. 40
   CFR parts 9, 141 and 142. Revisions to unregulated con-
   taminant monitoring regulation for  public water systems;
   final rule. Fed Regist 1999;64<180):50555-50620.
Clark JJJ. Toxicology of perchlorate [Ch. 3 and references
   cited therein]. In: Urbansky ET, editor. Perchlorate in the
   environment. New York: Kluwer/Plenum, 2000.
Coates JD, Michaelidou U, O'Connor SM. Bruce RA, Achen-
   bach LA. The diverse microbiology  of perchlorate reduc-
   tion [Ch. 24 and references  cited therein].  In: Urbansky
   ET, editor.  Perchlorate in the environment. New York:
   KJuwer/Plenum, 2000.
Darnian P, Pontius  FW. From rockets to remediation: the
  perchlorate problem. Environ Prot 1999:10:24-31.
Deuser C, Haley J, Torrence I. Lake Mead 'SWAT'  team
  attacks tamarisk. In Natural Resource Year in Review —
  1997. National Park Service. Washington  DC. Document
  No. D-1247, 1998.
Environmental Protection Agency. Drinking Water Contami-
  nant Candidate List. Document No. 815-F-98-002, 1998.
Espenson JH. The Problem and perversity of perchlorate [Ch.
  1]. In: Urbansky ET, editor. Perchlorate  in the environ-
  ment. New York: Kluwer/Plenum, 2000.
Giblin TL, Herman DC, Frankenberger WT. An autotrophic
  system for the bioremediation of perchlorate from ground-
  water [Ch. 19, and references cited therein]. In: Urbansky
  ET,  editor. Perchlorate  in  the environment. New York:
  Kluwer/Plenum, 2000.
Jackson PE, Laikhtman M, Rohrer J. Determination of trace
  level perchlorate  in drinking and ground water by ion
  ' chromatography. J Chromatogr A 1999;850:131-135.
Logan BE. A review of chlorate- and perchlorate-respiring
  microorganisms. Biorem  J 1998;2:81-95.
Magnuson ML, Urbansky ET, Kelty CA. Determination of
  perchlorate at  trace levels in drinking water by ion-pair
  extraction with electrospray ionization mass spectrometry.
  Anal Chem 2000a;72:25-29.
Magnuson ML, Urbansky ET, Kelty CA. Microscale extrac-
  tion  of perchlorate in drinking water with  low level detec-
  tion  by electrospray-mass  spectrometry. Talanta; 2000b;
  52:285-291.
Muzika RM, Swearingen JM. National  Park Service website,
  1999.  URL:   http://www.nps.gov/plants/alien/fact/
  tamal.htm.
Nzengung VA, Wang C. Influences on phytoremediation of
  perchlorate-contaminated water [Ch. 21  and references
  cited therein]. In:  Urbansky ET, editor. Perchlorate in the
  environment. New York: Kluwer/Plenum, 2000.
Perciasepe R. Part  III. Environmental Protection Agency.
  Announcement of the drinking water contaminant candi-
  date list: notice. Fed Regist 1998;63(40):10273-102S7.
Renner R. Perchlorate-tainted wells spur government action.
  Environ Sci Technol  I999;33:110A-111A.
Urbansky ET.  Perchlorate chemistry: implications for analysis
  and remediation. Biorem J  1998:2:81-95.
Urbansky ET,  Schock MR.  Issues in managing the risks  asso-
  ciated with  perchlorate in drinking water. J Environ Man-
  age 1999:56:79-95.
Urbansky ET, Magnuson ML. Sensitivity and selectivity- en-
  hancement  in perchlorate  anion quantitation using  com-
  plexation-electrospray ionization-mass spectrometry [Ch. vS].
  In: Urbansky ET. editor. Perchlorate in the environment.
  New York:  Kluwer/Plenum, 2000.
Urbansky ET.  Gu B, Magnuson ML. Brown GM. Kelty CA. A
  survey of bottled  waters tor perchlorate using ion chro-
  matography and electrospray ionization mass spectrometry.
  Submitted for publication. 2000.

-------
232                 '    E.T. Urbansky et al. / The Science of the Total Environment 256 (2000) 227-232

Westbrooks  R. Invasive  Plants: Changing the  Landscape      Wirt K,  Laikhtman  M, Rohrer J,  Jackson  PE. Low level
   of America — Factbook. Federal Interagency Committee         perchlorate analysis in drinking water and ground water by
   for  the  Management  of  Noxious  and Exotic  Weeds.         ion chromatography. Am Environ  Lab 1998;10(1):5.
   Washington  DC,  1998.  Available  on  the  web;  URL:      Wolff J. Perchlorate and  the thyroid  gland.  Pharm  Rev
   http://www.denix.osd.mil/denix/public/es-programs/con-         1998;50:89-105.
   servation/invasive/deserts.html.

-------