Critical Reviews in Analytical Chemistry, 30(4):311—343 (2000)
Quantitation of Perchlorate Ion: Practices and
Advances Applied to the Analysis of Common
Matrices'
Edward Todd Urbansky
United States Environmental Protection Agency, Office of Research and Development,
National Risk Management Research Laboratory, Water Supply and Water Resources Division,
Cincinnati, Ohio 45268. E-mail: Urbansky,Edward©EPA.gov; Fax: 513-569-7658; Phone:
513-569-7655.
ABSTRACT: In 1997, low-level perchlorate contamination (<50 ng mL-1 or parts per billion)
was discovered in the western U.S. Since that time, it has been found in sites scattered around
the nation. Although the Environmental Protection Agency has not established a regulation for
perchlorate in drinking water, it has placed perchlorate on the contaminant candidate list (CCL)
and the unregulated contaminants monitoring rule (UCMR). A "provisional and unenforceable
concentration of 18 ng mL-1 will stand until at least late 2000 when EPA hopes to issue a revised
toxicological assessment. However, the need for techniques and methods for determining per-
chlorate is not constrained to environmental chemistry. Perchlorate salts are used pharmaceuti-
cally in Europe to treat Graves* disease and amiodarone-induced thyrotoxicosis. Ammonium
perchlorate is used as a solid oxidant in space shuttles and intercontinental ballistic missiles.
Thus, methods and techniques are necessary for quality control and quality assurance. Moreover,
analysis of explosives and post-explosion residues have made quantitation of perchlorate impor-
tant in forensic chemistry. A variety of techniques is available: gravimetry, spectrophotometry,
electrochemistry, ion chromatography, capillary electrophoresis, mass spectrometry—each has
its strengths and weaknesses. Within each technique, assorted methods are available with
corresponding limits of detection. As the breadth of matrices undergoing analysis expands from
potable water to agricultural runoff, fertilizers, fruit juices, or physiological and botanical fluids,
the risk for interference becomes greater. As toxicologists demand lower and lower limits of
detection, it falls to analytical chemists to ensure selectivity and sensitivity go hand-in-hand. In
the near future, we can expect refinements in sample pre treatment and clean-up as well as
analytical methods geared toward analyzing more complex matrices. Ion chromatography, cap-
illary electrophoresis, Raman spectrometry, and eiectrospray ionization mass spectrometry will
all play roles in environmental analysis; however, IC should be expected Jo dominate dnnkng
water analysis. This review describes the state of the science and how it might be applied, and
looks forward to where it is going and how it might get there.
I. INTRODUCTION
Investigations of techniques and meth-
ods for quantitating perchlorate ion have been
undertaken for several reasons. First, per-
chlorate salts, especially ammonium perchlo-
rate, are used in missile, rocket, and space
shuttle propulsion systems. Second, perchlo-
rate salts, usually KC104, have been used
pharmaceutical^. Third, electrochemical
sensors are often evaluated in terms of
Hofmeister behavior, for which perchlorate
represents an extreme case. Fourth, and most
recently, perchlorate has turned up in water
Note: This paper is an original U.S. government work and is not subject to copyright.
! 040-8347/00/'$.50
© 2000 by CRC Press LLC
311
-------
supplies around the U.S. at < 50 ng mL_1
(ppb) concentrations. Fifth, perchlorate and
similar oxyanions are seen in residues from
explosions and there is a need to qualita-
tively and quantitatively analyze soil and
dust for these analytes.
In the first two cases, there is a need for
analysis to ensure quality control in produc-
tion. Ammonium perchlorate is regarded as
vital to national security. The importance of
this chemical commodity can be inferred
from what occurred after the Pacific Engi-
neering and Production Company of Nevada
(PEPCON) ammonium perchlorate plant
exploded on May 4, 1988. PEPCON was
one of two American production facilities at
the time. Ammonium perchlorate is used in
so many aerospace and military programs
that it was questioned whether the govern-
ment should have its own plants rather than
relying on private suppliers. At the time of
the explosion, U.S. government programs
consumed 90% of the NH4C104 produced as
more than 40 different weapon systems or
aerospace programs relied on ammonium per-
chlorate.1 Ammonium perchlorate continues
to be important today in the propulsion sys-
tems of intercontinental ballistic missiles
(ICBMs), such as the Peacekeeper missile
(Figure 1), and the space shuttles.
With either chemotherapeutic use or
drinking water contamination, there is a need
to determine trace concentrations where per-
chlorate is a minor constituent relative to
assorted organic and inorganic anions. At
present, EPA's interest in analyzing surface,
ground, and drinking water for perchlorate
relates to possible health effects and the
potential need for regulation. Many of the
recent developments in perchlorate
quantitation are a direct result of concerns
over potable water. Beginning in 1997, per-
chlorate ion was discovered in natural wa-
terways and aquifers around the western
U.S.2"6 Affected regions include southern
California (e.g., metropolitan Los Angeles),
Nevada (especially greater Las Vegas), north-
western Arizona, and parts of Utah. This
perchlorate problem is believed to be the
legacy of years of legal dumping of waste-
waters and most likely dates back several
decades. The original salt was probably
ammonium perchlorate, which continues to
be used as a solid oxidant and energetics
FIGURE 1. The launch of an LGM-118 Peacekeeper missile, one of many U.S. defense systems that relies
upon ammonium perchlorate. (Courtesy of U.S. Air Force.)
312
-------
booster in missiles and rockets. Over time,
ammonium perehlorate decomposes and must
be replaced. The first step in this process is
referred to as hog out—the removal of the
old material. Some hog out sites as well as
ordnance storage areas are known to suffer
from perehlorate contamination.
Perehlorate is legendary in inorganic
chemistry for its inertness to reduction de-
spite the high oxidation state of the chlorine,
+7; however, this behavior is due to kinetic
reasons and not thermodynamic ones,2-3,7,8
Perehlorate's effects on human health and
mechanism of action in the thyroid gland
have been described elsewhere.2,4'9-11 The ion
interferes with iodide uptake due to similar-
ity in size and hydration and thus also inter-
feres with thyroid hormone production. On
account of the possible risks to human health
through drinking water, EPA's Office of
Water added perehlorate ion to the Contami-
nant Candidate List (CCL) in 19981243 and
more recently to the Unregulated Contami-
nants Monitoring Rule (UCMR) in 1999.14
Although the analytical chemistry of
perehlorate ion was briefly summarized in
1997,2 the last 2 years have seen a dramatic
increase in research, especially in that dedi-
cated to trace quantitation in raw and fin-
ished drinking water supplies. While the area
of greatest focus has been ion chromatogra-
phy, developments have been made in other
techniques as well In particular, perehlorate
quantitation in explosion residues has been
an area of research in forensic chemistry.
The primary focus of this review is on meth-
ods applicable to raw and finished potable
water supplies; nevertheless, significant de-
velopments in other matrices have been made
and are included as they may be directly
applicable to water supplies or materials that
have the potential to influence water sup-
plies. It is worth pointing out that many
samples can be converted to a dilute aqueous
solution. Thus, techniques and methods de-
veloped for the analysis of aqueous solu-
tions (e.g., drinking water) may be applied
to a variety of matrices that are themselves
soluble or from which the ionic components
may be leached.
II. GRAVIMETRY AND
SPECTROPHOTOMETRY
For historical reasons, we first consider
gravimetric and spectrophotometric deter-
minations of perehlorate. The first attempt
to review and evaluate the methods in the
literature was undertaken by Lamb and
Marden in 1912.15 The next major contribu-
tion was Nabar and Ramac'nandran's 1959
paper on the colorimetric determination of
perehlorate with methylene blue,16 This set
the foundation on which other authors would
build future spectrophotometric methods.
Several alternative classic methods preceded
Nabar and Ramachandran, and these were
summarized in their work.16 A notable ex-
ception was the gravimetric determination
based on nitron, which is still in use today,
but suffers from a variety of interferences as
nitron precipitates many large anions, in-
cluding perehlorate, iodide, nitrate, tungstate,
bromide, and perrhenate.17-18 All of the ana-
lytical methods up to this point suffered from
a wide variety of interferences as well as
limitations in sensitivity. The lower limits of
detection (LLODs) were generally inad-
equate.
After Wyngaarden et al.19 demonstrated
perehlorate could be taken up by the thyroid
gland in place of iodide in 1952, the interest
in methods that could measure trace concen-
trations in physiological fluids was height-
ened. In 1968, Collinson and Boltz devel-
oped the first indirect atomic absorption
method based on a perchloratocuprous com-
plex.20 This was the first reasonably selec-
tive method in the presence of common ions
and had an LLOD of about 0.7 p.g mL"1. In
1972, Weiss and Stanbury modified
Collinson and Boltz's method by employing
ion exchange resins and applied the new
313
-------
method to analyzing biological fluids (i.e.,
serum and urine).21 Subsequently, other de-
velopments were described in a previous
review.2 The last few years have seen addi-
tional work and that is covered here.
Burns et al. developed a method based
on the extraction of an ion pair with
protriptylinium cation.22 They include a use-
ful review of spectrophotometric determina-
tions. Their method suffers from interfer-
ences by permanganate, bromide, molybdate,
thiocyanate, chlorate; however, evaporation
with HC1 alleviated these. The LLOD was
20 |0.g C104". Thus, it would be necessary to
evaporate off 4 L of water containing 5 ng
mL-1. In real water supplies, there are
involatile organic anions that could substi-
tute for perchlorate. Typical drinking water
samples have [C]org = 300 ng mL~'; thus,
there would 300 fig of organic carbon for the
20 jig of perchlorate. If only 10% of it is
ionic, the interference would still be severe.
Another recent method is analogous to
the methylene blue method and is based on
complexation with Astraphloxine FF and
other dyes.23 Typical of dye extraction-based
methods, it was not very selective for per-
chlorate over other anions and is dye de-
pendent. In the best case, e= 1.22 x 105 M~l
cm-1 (units not specified, but assumed); this
suggests an LLOD of 0.24 \iM = 24 g jiL-1
= 24 ng mL~!. In a slightly different twist, a
chemoreceptor-dye interaction is disrupted
by the perchlorate anion; this causes the
absorbance change.24 Unfortunately, nitrate
and bromide are selected preferentially over
perchlorate.
Ensafi and Rezaei developed a flow in-
jection analysis method with automated ex-
traction.25 However, the method does not
work in presence of chlorate, iodide, or ni-
trate. The authors report that heating to dry-
ness with HC1 eliminated these interferents,
but they did not address the problem of
involatile organic species. The LLOD was a
promising 3 ng mL '. At 50 ng mL-1 per-
chlorate, the authors claim 500 ppm (10,000
times based on mass) chloride has no effect.
Unfortunately, the sample preparation steps
mean the method cannot be applied conve -
niently to drinking water because of chlorate
and nitrate. Sample sizes of less than 1 mL
are used; however, a fair amount of care is
involved in the sample preparation. As a
rule, interference and dye impurities would
appear to pose high barriers to spectrophoto-
metric or colorimetric methods at low (ppb)
concentrations, but they might work in a
QC/QA laboratory where the major anion in
a material is perchlorate.
Of special interest is the possibility of
spectrochemistry by design.26 Hisamoto et
al. synthesized selective dyes for use in
optode (optical sensor) construction. These
multiinformation dyes have not been applied
to real matrices yet. Because selectivity is so
important, molecular design and construc-
tion appear to be key areas of research if
perchlorate is to be directly quantitated with-
out separation whenever other common an-
ions are present in high concentrations. Se-
lectivity is still limited as illustrated in Figure
2, but this field is still developing. Although
the authors did not test it, there is the poten-
tial for application to electrochemical sen-
sors. A highly selective dye might be able to
eliminate the selectivity problems associated
with electrochemical detection without prior
use of a separation technique (vide infra).
More promising than colorimetry or spec-
trophotometry is Raman scattering. The
Raman phenomenon takes advantage of the
unique changesin scattering frequencies as-
sociated with changes in the polarizability
tensor (a function of the point group and the
bond strengths).27 Infrared spectrometry,
which depends on changes in the dipole
moment of the species during vibration, is
hindered by the presence of water except in
special applications (i.e., attenuated total
reflectance or mirrored internal reflectance,
which suffer from poor throughput). Unlike
IR spectroscopy, Raman scattering is much
less affected by water. In addition, molecules
314
-------
KD-U7
UTMCIQ4
i tru cio.
400 500 600 700
Wavelength(nm)
O 0.8
w 0,6
KD-4M7
lOQC^nion
FIGURE 2. Typical characteristics of multi-information dyes used as optodes
for perchlorate. Sal" = salicylate. (Reprinted from Anal. Chim. Acta 1998,
373, 271-289, with permission from Elsevier Science © 1998.)
and ions without net dipole moments can be
Raman active even though they are IR inac-
tive. Although perchlorate has no net dipole
when tetrahedral, it can have a net dipole
during bending, stretching, or wagging; there-
fore, it is in fact IR active. While Raman
bands of solids tend to be sharper and better
defined, this technique is readily applied to
aqueous solutions. Generally, argon-ion or
similar lasers are focused on a microscopic
volume of solution. The light is scattered as
it interacts with a perchlorate ion (or any
chemical species) in solution. Scattered light
with the same frequency as the incident light
is called Rayleigh scattering (v0) and is not
analytically significant. However, when the
scattered light is shifted in frequency, this
phenomenon is analytically significant. If
the ion gains energy from the light-particle
interaction, the scattered photon has less
energy than the incident photon (v0~ 6), and
a Stokes band is seen. On the other hand,
when the ion imparts energy to the photon
(v0 + 8 ), an anti-Smokes band is seen. To
minimize detection of Rayleigh and incident
radiation, the detector is normally positioned
perpendicular to the laser beam. Raman scat-
tering can be used to definitively rule out the
presence of perchlorate. Chlorine oxyanions
have been studied by Raman spectroscopy
in some detail and are well understood.28
The Raman bands can be shifted by sodium
or other cations, but this effect is systematic
and is readily accounted for.29 Although
normally limited to minimum analyte con-
centrations of 1 to 10 mM, Raman spectros-
315
-------
copy has been used successfully to measure
as little as 1 jig mL~' after preconcentration
(Figure 3).30 Most recently, it has been ap-
plied to aqueous leachates or solutions of
fertilizers.31'32
III. ELECTROCHEMISTRY
In this category, we consider recent de-
velopments in ion-selective electrodes
(ISEs) as well as membrane field effect tran-
sistors (MEMFETs) and other sensors based
on electricity. As noted above, perchlorate is
used routinely in electrochemical studies as
it occupies a key location in the Hofmeister
series.33"35 Anions positions in the Hofmeister
series are generally considered to be a mea-
sure of Gibbs free energies of aquation and
solubility of the cation-anion pair. The
Hofmeister series for selectivity of an ion-
selective electrode is as follows:34 C104~ >
SCN- > I > N03- > Br > N02- > CI".
Many perchlorate ISEs are based on
quaternary ammonium cations; these are
heavily influenced by the solubilities of the
perchlorate salts. More recent developments
have concentrated on macrocyclic chemo-
sensors or chemoreceptors.36-40 Some of the
new electrodes demonstrate deviations from
Hofmeister behavior (usually referred to as
anti-Hofmeister even though the order is not
necessarily reversed, but often jumbled).37-39
Macrocyclic chemoreceptors used in some
Nitrate
Perchlorata
Raman Shift (cm-1)
FIGURE 3. Raman spectra of nitrate (100 s, 1047 cm-1) and perchlorate (107 s, 934 cm"1) after electrophoretic
separation. A Nd-Yag laser (532 nm) was used as the source; Rayleigh scattering was rejected with a
holographic filter. Capillary: fused silica, coated with 3% T-linear polyacrylamide, 20 cm long x 75 nm i.d.
Conditions: 0.50 mM in each analyte; elect rokinetic injection for 15 s at -2 kV. Spectra were flat-fielded and
ratioed to the background electrolyte spectrum. A 5-point Savitsky-Golay quadratic smooth was employed.
Inset is the raw spectrum at 100 s. (Reprinted from Appl. Spectrosc. 1995, 49, 1183-1188, with permission
from the Society for Applied Spectroscopy © 1995.)
316
-------
electrodes lead to deviations from Hofmeister
behavior, and some electrodes response pref-
erentially to iodide36 or thiocyanate40 over
perchlorate.
A liquid membrane electrode based on
Brilliant Green has an LLOD of 20 jxM (2 fig
rnL-1) for perchlorate.41 The most serious
interferences come from periodate and per-
manganate. While neither of those is likely
to be encountered in water, chloride (usually
present at-3 mM = 100 ppm) has selectivity
coefficient about 10"6 that of perchlorate,
but is present at a molar concentration about
106 times that of perchlorate.41 Another ar-
ticle by the same authors attempted to link
potentiometric selectivity with partition co-
efficients.42 In it, they assumed 100% of pre-
cipitate formed by combining the cationic
dye with the perchlorate was in fact the dye-
perchlorate complex even though high salt
concentrations were used to promote crys-
tallization and the solid was not character-
ized.42 A selective fluorinated polyether (non-
ionic) chemosensor for perchlorate reached
a LLOD = 1 j±M = 100 ng rnL-1, even in high
chloride (0.10 A/).43 Unfortunately, it still
suffers from typical effects of nearby
Hofmeister series anions.43 Errachid et al.
1000
wrote a solid paper on ionophore develop-
ment and testing.44 They produced
MEMFETs and ISEs based on a phospha-
dithiamacrocycle. They saw well-behaved
response (Figure 4) for perchlorate down to
1 jxM and they obtained an LLOD of
100 nM44
Two papers were published on carbon
paste electrodes.43-46 In the first case, an
LLOD = 1 jiM = 100 ng mlr1 was obtained,
but interference problems would make it
difficult to apply to drinking water matrices.
It was nevertheless used successfully for
cetylpyridium titrations to determine perchlo-
rate in solution at higher concentrations.45 In
the second case, use of electrochemically
generated thallium(O) to reduce perchlorate
was applied to real tap water samples spiked
with perchlorate. An LLOD of 50 ng mL~l
was obtained, but interferences from labile
redox-active species (e.g., nitrate, nitrite,
chlorate, bromate, arsenate, and arsenite)
could confound the analysis.46
One contribution was particularly novel
and fascinating, but it is not ready for imple-
mentation. An electrochemiluminescent de-
termination of perchlorate was reported in a
recent communication by Xu and Dong.47
>
e
W
FIGURE 4. Perchlorate response for a membrane field-effect transistor (MEMFET)
and ion-selective electrode (ISE) based on a phosphadithiamacrocyclic chemosensor
designed for use in PVC membranes. (Reprinted from Sens. Actuators 81997, 43,
206-210, with permission from Elsevier Science © 1997.)
317
-------
An LLOD of 50 nM (5 ng mL_1) was ob-
tained, but the method suffers from interfer-
ence from iodide. Although iodide is not
generally found in drinking water supplies,
it would be present in the bloodstream and
intracellular thyroid fluids. Incomplete in-
formation on interference from other anions
precludes a thorough discussion of the appli-
cability of this method for drinking water
analysis. However, this appears to be the
first attempt at assaying perchlorate by
electrochemiluminescence, even though the
Ru(bpy)32+ complex has been used for other
analytes. Such combination techniques are
probably necessary to obtain adequate sensi-
tivity and selectivity.
At present, direct application of electro-
chemical methods is restricted to substances
that are comprised primarily of perchlorate
salts. Selectivity limitations make them likely
to be applicable only to those solutions where
it is already known that perchlorate is a pre-
dominant anion, as in a quality control labo-
ratory or online, real-time process-monitor-
ing system. In that capacity, they have rapid
response, sufficient selectivity, and high sen-
sitivity. However, interferences from other
large, poorly aquated anions are severe
enough to render these methods almost use-
less for typical drinking or raw waters. While
selectivity for perchlorate relative to com-
mon ions such as chloride, bromide, sulfate,
and nitrate is several orders of magnitude
higher, the concentration of perchlorate is
several orders of magnitude lower than the
concentrations of the other ions. Curiously,
the selectivity for perchlorate over other com-
mon anions is almost equally balanced by
the concentrations at which these species
normally occur. It is unclear whether stan-
dard additions could be used to vitiate ma-
trix effects (primarily interferent anions). If
a true blank could be measured by sequester-
ing perchlorate with other reagents to ac-
count for the background signal of the
interferents, then standard additions would
permit broader application of electrochemi-
cal methods. This would require a highly
selective sequestrant and would probably be
difficult given the behavior of precipitants
such as nitron or tetraphenylarsonium, which
react with a wide variety of large anions.
The candidates that immediately suggest
themselves are Rb+ and Cs+; the perchlorate
salts are sparingly soluble as perchlorate
solubilities go (of course, there will be a
counterion to balance the charge). Alter-
nately, a carefully designed macro- or poly-
cyclic molecule of the right size and shape
might do the job as well. These areas of
research remain unexplored. Despite these
weaknesses of selectivity, highly sensitive
electrochemical sensors may readily be ap-
plied if ions are separated before detection
using ion chromatography or capillary elec-
trophoresis, as we shall see in the sections
that follow.
IV. CAPILLARY ELECTROPHORESIS
Capillary electrophoresis (CE) has been
a rapidly advancing subdivision of separa-
tion science; more than 8000 papers using
CE were published in 1999. This rapid pro-
liferation of papers makes it almost impos-
sible to keep up to date. At its most funda-
mental level, capillary electrophoresis is
based on the differentiable migration of ions
under the influence of an electric field. Dur-
ing its developmental phase, it was known
by several synonyms, including capillary ion
analysis, capillary ionophoresis, and capil-
lary zone electrophoresis; however, the cur-
rent usage favors capillary electrophoresis.
The key term is capillary. Use of the capil-
lary allows rapid analysis (<10 min) with a
minimum of Joule heating (the effect of a
current flowing through a salty solution).
Readers unfamiliar with CE are referred to
several books on the subject.48 51
The discovery of trace perchlorate in
ground and surface waters around California
in 1997 is generally attributed to advance-
318
-------
ments in ion chromatography that made low-
level detection possible. However, Nann and
Pretsch52 actually made the first report of
determining perchlorate at 10 nM (1 ng mlr
') in tap water by CE with an ion-selective
microelectrode (ISME) in 1994 (Figure 5),
In 1997, Ehmann et al. showed that CE could
be used to determine 50 nM (5 ng mlr')
perchlorate in the presence of 22 other
equimolar anions (Figure 6),53 In order to
accomplish this, they used electrokinetic in-
jection with isotaehophoretic precon-
centration. Unfortunately, both discoveries
went completely unnoticed by the drinking
water industry that has exclusively favored
ion chromatography over CE (partly due to
EPA Methods), Because many of the devel-
opments that take place in research laborato-
ries are do-it-yourself and require time to
construct, they are often overlooked or ig-
nored by government and industry laborato-
ries until they become commercially avail-
able.
Papers on CE can usually be divided into
two groups: those that test new mobile phases
(also known as running buffer or background
electrolyte solutions), often with a new chro-
mophore or fluorophore, or those that test
new detectors. CE itself is a separation tech-
nique and thus requires a detector. If the
separation is sufficient, a fairly nonselective
mode of detection may be used. Much of CE
analysis is done using indirect detection, in
which a loss of signal (e.g., fluorescence or
absorbance) indicates the analyte is eluting
in place of the background electrolyte. Di-
rect detection by unselective electrochemi-
cal devices is also possible. Any measurable
property of the analyte may be used for di-
rect detection if the sensitivity of the detec-
tor is high enough.
Forensic CE has been applied to the
analysis of residues from explosives by the
Federal Bureau of Investigation (FBI)54 and
other law enforcement agencies.55-57
Hargadon and McCord report that recovered
fragments of incendiary devices generally
have sufficient residues remaining to meet
the the 5 fig mlr1 detection limit they re-
port.54 The FBI uses comparative analysis
FIGURE 5. Electropherogram of 10 nM (1 ng mi.-') perchlorate in tap water (solid line). Detection by
ion specific micro electrode. Conditions: running electrolyte: 20 mM sodium hydrogen suifate/sodium
sulfate, pH 2.5; electrokinetic injection potential: 10 kV for 10 s; separation potential: 30 kV; uncoated
capillary. Dotted line shows effect of adding 20 mMsulfate. Peaks: 1 = bromide, 2 = chloride, 3 = nitrate,
4 = perchlorate, 5 = unknown. (Reprinted from J. Chromatogr. A1994, 676,437-442, with permission
from Elsevier Science © 1994.)
319
-------
50 mSf
32 M
wwV ¦¦¦
ill mTOnniii HjFliiH
2,0
_r_
2J
-T—
3.0
3J
¦™r™
4,S
i [mill]
—I-
5.0
I
13
—r~
8.0
FIGURE 6. Electropherogram of 50 nM each on a Dionex CES 1 using electrokinetic injection and
isotacnopnoretic preconcentration: Peaks: 1 = thiosulfate, 2 = bromide, 3 = chloride, 4 = sulfate, 5 = nitrite,
6 = tetrafluoroborate, 7 = nitrate, 8 = sulfosuccinate, 9 = oxalate, 10 = mclybdate, 11 = perchlorate (5 ppb),
12 = thiocyanate, 13 = tungstate, 14 = chlorate, 15 = citrate, 16 = malonate, 17 = malate, 18 = tartrate, 19
= fluoride, 20 = bromaie, 21 = formate, 22 = phosphate, 23 = arsenate, (Reprinted from Chromatographia
1397, 45, 301-311, with permission from Elsevier Science © 1997.)
with CE and IC to confirm identifications.
Chromate is very popular for indirect detec-
tion in CE, and Figure 7 shows a typical
result. Kishi el al. demonstrated that CE could
be applied to other physical evidence, such
as a cotton glove (Figure 8) used to handle
fireworks.55 Comparative analyses were also
done by X-ray fluorescence spectroscopy and
X-ray diffraction.
Using chromate as a chromopfaore can
limit sensitivity with indirect detection. For
that reason, a wide variety of chromophores,
running electrolytes, and/or electroosmotic
flow modifiers has been used (Figure 9),
such as sulfosalicylate/imidazole (LLOD =
800 ng mL-1),58-59 phthalate (LLOD = 600 ng
ml,"1),60-61 indigotetrasulfonate (LLOD = 100
ng mL-1),62 pyromellitic acid,63 naph-
thalenesulfonates,64 or cationic polymers.65
Cyclodextrins can be used to change electro-
phoretic mobility and improve separation as
shown in Figure 10.66
Alternative detectors can improve sensi-
tivity over that observed for indirect uv-ab-
sorbance. The use of conductivity for IC is
common, and was applied to CE in 1993,
giving a perchlorate LLOD of about 10 ng
mL-1,67 Fluorescence detection has been
used68 and is now commercially available.
In 1999, Maeka et al. reported on the use of
a copper electrode for detection in CE.69
This technology has not been applied to real
samples yet and is not optimized, but initial
work puts the LLOD at to 3 jig mL-1.69
Kappes et al. used coated-wire electrodes
for detection.7" Based on their electrophero-
grams (Figure 11), perchlorate should have
an LLOD of ~ 100 to 200 ng mL"1. One
advantage of the coated-wire electrodes over
micropipet electrodes is their relative dura-
bility. Althoughsubstantial tailing is observed
after 5 days, the coated-wire electrodes last
2 to 3 days.
Better techniques of sample volume re-
duction have allowed mass spectrometry
(MS) to be used for detection, and CE-
ionspray MS has been applied to solutions
containing 1 mM perchlorate.71 Figure 12
demonstrates that the m/z = 99 u peak for
perhlorate can be discriminated on the basis
of m/z from the overlapping (coeluting) peaks
of selenite, phosphite, and iodate.
320
-------
o.ooao
S.2
5,#
4.4
&.0
4.8
1.4
time, min
FIGURE 7, Electropherogram from the FBI Laboratory on a Dionex CES 1. Capillary: 65 cm x 75 m Id.;
Conditions: 3.8 mM chromate, 1 mM diethyienetriamine. borate buffer, pH 7.8; hydrostatic (gravity) injection:
-50 nL; indirect uv-detection: A = 205 nm; B = 280 nm. Peaks: 1 = chloride, 2 = nitrite, 3 = nitrate, 4 = sulfate,
5 = perchlorate, 6 = thiocyanate, 7 = chlorate. Note that nitrite, nitrate, and thiocyanate all absorb at 205 nm,
thereby producing positive peaks. This additional information can be used to confirm or rule out identies
obtained by elution time using the signal at 280 nm. Concentrations not reported. (Reprinted from J.
Chromatogr. 1992, 602, 241-247, under U.S. government authority [not subject to copyright].)
aimUf
mlgratloa time
FIGURE 8. Electropherogram of an aqueous extract of a cotton glove used to handle fireworks. (Reprinted
from Electrophoresis 1998, 19, 3-5, with permission from John Wiley & Sons, Inc. © 1998.)
321
-------
rrr wit
4 g
0.0
0.5 mAU
T i
9a (left)
14.0 0.0
1.0
9a (right)
UJ
18 2.1
Z» 3.1
Tim® (mln)
9b
a.a
2 3
TIME (minutes)
9c
FIGURE 9. Electropherograms of mixtures of anions, (a) Capillary: 34.5 cm x 50 |im i.d. Conditions: 3.06 mM
imidazole; applied potential: 15 kV; injection 15 s; current: 0.5 to 3.0 |iA; left: pH 6.52, right: pH 7.04; indirect uv
detection at 214 nm. Peaks: 2 = sodium, 3 = lithium, 4 = water dip, 5 = phosphate, 6 = fluoride, 7 = formate, 8
= chlorate, 9 = perchlorate (8.7 g mL_1). (Reprinted from Electrophoresis, 1998, 19, 2243-2251, with
permission of John Wiley & Sons, Inc. © 1998.) (b) Inset conditions: 5 m/Wchromium (VI), 20 mWdiethanolamine,
0.50 mM tetradecyl (myristyl) trimethylammonium bromide. Main figure conditions: 200 \iM potassium
indigotetrasulfonate, 10 mM glutamic acid, 0.1% Carbowax, pH 3.22. applied potential: 30 kV; pressure injection:
0.6 s at 12.5 torr; indirect uv detection at 314 nm; 20 |iM each anion. Peaks: 1 = sulfate, 2 = nitrate, 3 =
perchlorate (2 fig mLr1), 4 = chlorate, 5 = bromate. (Reprinted from J. Chromatogr. >41998, 804,327-336, with
permission from Elsevier Science © 1998.) (c) Capillary: 51.2 cm x 14 (im i.d. Conditions: 0.25 mM salicylic acid/
sodium salicylate; pH 4.0; electrokinetic injection for 0.7 s at 30 kV; current: 5.7 nA; applied potential: 30 kV;
indirect fluorimetric detection using an argon ion laser. Peaks: 1 = chloride, 2 = nitrate, 3 = perchlorate (25 jaM
= 2.5 |ig mLr1), 4 = permanganate, 5 = dichromate, 6 = iodate, 7 = phosphate, 8 = salicylate. (Reprinted from
J. Chromatogr. 1989, 480, 169-178, with permission from Elsevier Science © 1989.)
322
-------
1.475E-2
s
s
1J75E-2-
3.0
3.5
4.0
migration time (min)
•3
>
g
U
I
a
I
tt-cjfclodexlria (mmol/L)
FIGURE 10. (A) Electropherogram of a mixture of anions (5 mM each) using a-cyclodextrin to affect ionic
mobility. Conditions: 5 mAf sodium chromate, 20 mM a-cyclodextrin, pH 8; hydrostatic injection for 15 s from
a height of 10 cm (-100 jiM); applied electric field: 333 V cnr1. Peaks: 1 = nitrate, 2 = iodide, 3 = bromate,
4 = thiocyanate, 5 = perchlorate (500 jig mL"1). (B) Influence of a-cyclodextrin concentration on electro-
phoretic mobility. Conditions: 5 mM potassium hydrogen phthalate + 0.002% poly(1,1 -dimethyl-3,5-
dimethylenepiperidinium chloride) (molar mass = 200 kDa), as electroosmotic flow modifier. Line identifica-
tions 1-5 see (A); 6 = iodate, 7 = ethanesulfonate, 8 = butanesulfonate, 9 = pentanesulfonate, 10 =
octanesulfonate. (Reprinted from Can. J. Chem. 1998, 76, 194-198, with permission from the National
Research Council of Canada © 1998.)
323
-------
cr
T
*00 10:00 11:00 11:00 13:00 14.00 13:00
I/nia
30 mV
9:00
10:00
11:00
12:00
(/nun
13:00
14:00
—1
15:00
FIGURE 11. (A) Electropherogram of a mixture of anions. Capillary: 100 cm x 25 jam i.d., uncoated fused silica.
Conditions: 10 mM potassium sulfate; electrokinetic injection for 7 s at -5.0 kV; applied potential: -30 kV.
Concentrations: 10 \iM chloride, bromide, nitrite, nitrate; 2 \lM perchlorate, thiocyanate; 1 \iM iodide. (A) Using
a micropipet ISE based on a Mn(III) porphyrin (MnTPP). (B) Using a coated-wire ISE based on MnTPP.
(Reprinted from Anal. Chim. Acta 1997, 350, 141-147, with permission from Elsevier © 1997.)
A variation of CE is capillary electro-
chromatography (CEC), which is included
here because it operates under an applied
electric field and can therefore take advan-
tage of ionic (electrophoretic) mobilities.
However, unlike CE, CEC relies on the types
of secondary interactions between the analyte
and the stationary phase that occur in any
chromatographic process. In this respect, it
is a hybrid of IC and CE because it uses the
mechanisms of both. As shown in Figure 13,
it has been applied to the determination of
perchlorate at 50 mM (5 mg mL"1)-60 Simi-
larly, Hsu et al.72 modified the capillary wall
by bonding a macrocyclic polyamine. The
polyamine structure is similar to an am-
pholyte and allows a continuous variation of
electrophoretic mobility as a function of pH
as shown in Figure 14. Hauser et al. modi-
fied a capillary with a polyacrylamide that
incorporated a quaternary amine and added
an ion-selective microelectrode, reaching
10 mM (1 mg mL~') as a detection limit for
perchlorate.73 One of the problems in CEC is
reduced electroosmotic flow, as noted by
Hilder et al., who used a Hypersil C18 pack-
ing.74 At first glance, there appears to be
little benefit to using CEC over CE for quan-
titating perchlorate; however, CEC can bet-
ter separate lipophilic species. Consequently,
324
-------
HK>,-
I**
15.775
mi m
I\
OOj
2.17J
flVS$?
T
<323
!\
Mill
KL
L
f 10
Turn (mi»)
FIGURE 12. CE-ion spray-MS for inorganic anions (1 mW) using 2.5 mJW pyromellitic acid in 20% MeOH, pH
7.8. 'Singly protonated. "Doubly protonated. (Reprinted from Anal. Chem. 1996, 68, 2155-2163, with
permission from the American Chemical Society © 1996.)
tlktline cations
Anions
FIGURE 13. Electropherogram of a mixture of ions by capillary electrochromatography. Capillary: 30.4 cm long
x 50 jim i.d.; packing: TSK IC-Anion-SW. Conditions: 90% v/v 5 mM phthalic acid and 5 mM
hexamethylenediamine containing 0.15% HEPES, pH 6.8; 10% v/v MeOH; applied potential: 4 kV; indirect uv-
detection at 236 nm. 1 = lithium, 2 = sodium, 3 = potassium, 4 = chloride, 5 = nitrite, 6 = nitrate, 7 = iodide,
8 = sulfate, 9 = perchlorate; concentrations not specified. (Reprinted from J. Micro. Sep. 1997, 9, 347-356,
with permission from John Wiley & Sons, Inc. © 1997.)
325
-------
PH
FIGURE 14. Eiectrophoretic mobility as a function of pH. Conditions: fused silica capillary
covalently modified with [24]ane-Ne x5Q cm (100 pm Ld.}; 5 m/Wsodium chromate; applied
potential: 15 kV; analyte concentration: 10 (1 ug ml""1 for perchlorate); electrokinetic
injection for 5 s at -15 kV; indirect uv detection at 250 nrn. (Reprinted from the Analyst
1997, 122, 1393-1398, with permission from the Royal Society of Chemistry © 1997.)
it may be applicable to some samples of
ground water, soil leachates, physiological
fluids, and other matrices that potentially
contain large amounts of soluble, polar, or-
ganic molecules (e.g., phenols, polyalcohols).
Surface waters contain natural organic mat-
ter (humic and fulvic acids) at several ppm
(as carbon), which commonly interfere in
standard methods. CEC can simultaneously
determine mixtures of hydrophilic and hy-
drophobic species, depending on the pack-
ing material used.
Meissner et al. used a related technique,
isotachophoresis, to measure a series of an-
ions.75 In isotachophoresis, the terminal and
leading background electrolytes are of low
and high mobility, respectively (relative to
each other and the analyte anions). The elec-
troosmotic flow is often stopped by coating
the capillary. As the anions migrate under
the influence of the electric field, they are
separated within the plug of injected solu-
tion.
The most widely available equipment and
reagents for CE are presently restricted by
LLODs of 100 ng mL-1 (ppb) to 10 g mi r1
(ppm). This makes them readily amenable to
the analysis of explosives and other solid
materials where perchlorate anion is present
at > 1 mg kg"1. Without using more sophis-
ticated preconcentration steps or detectors,
however, CE cannot reach the LLODs re-
quired for the routine analysis of potable
water or raw waters, but could be used for
some highly contaminated sites, where per-
chlorate concentrations in the parts-per-mil-
lion range are common. Forensic analytical
chemists have unquestionably demonstrated
that CE is applicable to the determination of
perchlorate in explosives and post-explosion
residues. Therefore, we can expect to see CE
play a role in that discipline. In those cases
where perchlorate is much lower than other
anions, peak overlap can be a serious prob-
lem in undiluted samples; however, dilution
raises the net detection limit. Thus, there is
an interplay between sensitivity of detectors
and matrix effects that must be considered.
It is worth pointing out that the limits of
CE with respect to analyzing water samples
are related to practical sample size problems
more than the sensitivity of available detec-
tors. Indirect fiuorimetric detection can reach
down to the attomole (10"iS mol) region,
which is fewer than 106 ions. Detecting 50
amol of perchlorate in a drinking water
326
-------
sample containing 5 ng mL"1 would require
an injection of 1 nL. In a capillary of 20 um
i.d, that would be a 2.5-mrn length of capil-
lary filled with solution. Such a plug of so-
lution would stop ionophoresis as it would
limit the current. However, detecting 1 amol
would require only 20 picoliters (pL), and
thus only 50 jim of the capillary to be used.
Depending on the fluorophore, a few parts-
per-billion is the practical lower limit of
detection for CE. As pointed out by Nann and
Pretsch,52 detecting an analyte present at 10
aM requires a detector capable of measuring
about 20 zmol s_I (1 zeptomole = 10~21 mol).
Other than fluorescence, ISMEs and conduc-
tivity are adequately sensitive for analyzing
potable water samples. Nevertheless, until
such time as these devices are common on
commercial instruments and make their way
into methods approved by the EPA and other
regulatory agencies, CE will continue to be
displaced by IC for the determination of many
ions found in drinking water and other liquid
matrices.
V. ION CHROMATOGRAPHY
The fundamentals of ion chromatogra-
phy (IC) are covered elsewhere.76 In this
review, we focus primarily on specific ad-
vances which make IC the technique of
choice for drinking water at the present.
Along with thiocyanate and iodide, perchlo-
rate has a low charge density and is poorly
aquated. Consequently, it is strongly retained
on most columns and suffers from poor peak
shape if it can be discerned eluting at all.
Several actions can be taken to promote elu-
tion: (1) add an organic solvent (usually
methanol) to the mobile phase,77 (2) use a
more hydrophobic anion in the eluent to dis-
place the perchlorate,78"81 (3) increase the
concentration of the eluent anion, or (4) make
the stationary phase more hydrophobic.82"85
Biesaga et al. illustrated the separation
of perchlorate from other chlorine oxyanions
in tap water.61 The IC retention time was 30
to 40 min, but the peak was still well shaped.
Unfortunately, the LLOD was 1.5 fig mL"1.
They obtained a CE LLOD of 0.6 jig mL-1;
at that time, CE outperformed IC. Phthalate
was used as the eluent anion for IC. Relative
to the concentration ranges that needed to be
measured, this was inadequate.
One of the first breakthroughs was the
development of a method based on
p-cyanophenol/p-cyanophenoxide by
Okamoto et al. at the California Department
of Health Services in 1997.78-79 Miura et al.
used 1,3,5-benzenetricarboxylate in the same
way.80 Also in 1997, Maurino and Minero
were developing a method based on cyanu-
ric acid/hydrogen cyanurate (cyanuric acid
= 2,4,6-trihydroxy-l,3,5-triazine).81 Both
Okamoto's and Maurino's method are based
on hydrophilic columns (Dionex IonPac AS5
and AS4, respectively) using hydrophobic
anions to displace the perchlorate. Okamoto
et al. showed that satisfactory peak shape
could be maintained withp-cyanophenoxide
eluent even in the presence of high chloride,
sulfate, and hydrogen carbonate. (Figure 15).
The peak shape with p-cyanophenoxide is
better than that obtained with cyanurate (Fig-
ure 16). In addition, Maurino and Minero
reached an LLOD of 1 [ig mL"1, but Okamoto
et al. were able to detect down to -1 ng mL"1.
Because perchlorate is quite hydropho-
bic as anions go, Dionex has pursued use
and development of columns intended to
permit breakthrough of well-aquated ions
(e.g., chloride, fluoride, bromide).82"83 Jack-
son et al. showed that the more hydrophobic
IonPac AS11 column can be used with a
large sample loop.82-85 The AS11 is limited
in high sulfate84-86 as demonstrated in Figure
17; recoveries are unsatisfactorily low.
Alltech has a methacrylate-based column for
hydrophobic anions, but has not established
methods that reach LLODs in the low ppb.87
While the AS 11 column can be used under
appropriate conditions, the Dionex IonPac
AS 16 was developed specifically to take into
327
-------
200
HCO,~ added—mg/L
600
1,000
Retention Time—minutes
15a
0.0
-0.5
200
S04 ~ added—mg/L
600
1,000
Reagent water
0.0
-0.5
1
5
Retention Time—minutes
i
10
15b
FIGURE 15. Ion ehromatogram of perchiorate on a Dionex AGS (guard) and ASS (analytical) columns using
p-cyanophenol/p-cyanophenoxide eluent. (A) Effect of hydrogen carbonate. (B) Effect of sulfate. (Reprinted
from J. Am. Water Works Assoc. 1999, 91, 73-04, with permission from the American Water Works
Association © 1999.)
328
-------
2.0
CfO,
so,
cr
E
u
1JS
M
a.
>
o
O
°:i3
25
0
5
10
IS
20
2.0
cr
E
u
CrO<
15 -
1.0
0.5
o
O
25
15
20
S
10
0
Retention tima, min
FIGURE 16. Ion chromatogram of perchlorate on a Dionex AGS (guard) and ASS (analytical) columns using
p-cyanophenol/p-cyanophenoxide eluent. (A) Effect of hydrogen carbonate. (B) Effect of sulfate. (Reprinted
from J. Am. Water Works Assoc. 1999, 91, 73-84, with permission from the American Water Works
Association © 1999.)
' ' ' t *"¦ (in. p.,,,,,,,,,,,, j
0 ZO 4.0 8.0 8.0 10.0 12.0
Mlnytes
FIGURE 17. Ion chromatogram of 20 ng mL-1 perchlorate on the Dionex lonPac AG11 (guard) and AS11
(analytical) columns. Conditions: 0.10 jxMNaOH(aq) eluent; flow rate; 1.0 mL min-1; suppressed conductivity;
1000 uL injection. Stacked chromatograms show the effect of adding sulfate (bottom to top) 0, 50, 200, 600,
and 1000 ng mL-1. Peak 1 = perchlorate, (Reprinted from J. Chromatogr. A 1999, 850, 131-135, with
permission from Elsevier Science © 1999.)
329
-------
account the need for a column resistant to
high concentrations of hydrophilic salts (e.g.,
NaCl, NajSOJ.84,85,88 Even in a high ionic
strength synthetic pound water matrix, a
well-shaped peak is obtained using the AS 16
column (Figure 18), The U.S. Air Force
Research Laboratory has actively pursued
validation of the AS 16 column under a wide
variety of conditions.89
Methods for the AS5 and AS 11 are likely
to be abandoned in favor of the AS 16, which
is the most resistant to common matrix ef-
fects. The Alltech methacrylate column is
unlikely to play a major role unless further
development is done to meet the needs of the
drinking water industry. The Dionex lonPac
AS 16 column can be expected to dominate
the analysis of drinking water, especially
with the issuance of EPA Method 314,90
which is expected to be promulgated for use
in the UCMR. This resistance will be even
more important as ion chromatography is
applied to analyzing aqueous solutions of
fertilizers, botanical fluids, and physiologi-
cal fluids, all of which contain high concen-
trations of ions as well as assorted organic
molecules that change the nature of the
mobile phase or sorb to the column them-
selves.
The most common detector for ion chro-
matography is a conductivity cell, and it is
the standard detector used on Dionex instru-
0.8
pS
0.0
ments. However, conductivity is not selec-
tive; consequently, there will be a need for
confirmation. Buch'oerger and Haider com-
bined IC with particle beam mass spectrom-
etry. Perchlorate was fragmented by elec-
tron impact to give 7 ions; no fragmentation
was observed using chemical ionization.91
Corr and Anacleto also used MS for detec-
tion with IC the same as they did for CE,71
While the previous papers are very prag-
matic and practical in nature—geared to-
ward a very specific application, fundamen-
tal research in ion chromatography has not
been neglected. A variety of properties and
models has been been used to explain the
retention behavior observed for perchlorate.
Daignault et al. attempted to correlate polar-
izability with the capacity factor (Figure
19).92 However, the formula used by
Daignault et al. was taken out of context and
misapplied. This does not entirely negate the
relationship they saw, but still merits correc-
tion. The energy associated with London
dispersion forces between two species A and
B is related to the polarizability of the spe-
cies by Eq. 1:93
E-~-
4 h
\!a+ib )
(1)
where / represents ionization energy, a rep-
resents polarizability, and R represents the
2.0 4.0 6.0 8.0
Minutes
10.0 12.0
FIGURE 18. Ion chromatogram of 5 ng mL-1 perchlorate in synthetic ground water on the Dionex lonPac AG16
(guard) and AS16 (analytical) columns. Synthetic ground water contains 200 ng mL-1 chlohde, 50 ug ml"'
nitrate. 200 |ig mL"1 carbonate, 1000 p.g mL"' sulfate. Peak 1 = perchlorate. (Reprinted from J. Chromatogr.
A 1999, 850, 131-135, with permission from Elsevier Science © 1999.)
330
-------
¦**
e>o
o
0.40
0.30
0.20
0.10
0.00
-•
mf
-¦
¦ e /
..
d
a y/b
' c
10 20 30 40
„ 2
50
FIGURE 19. Log of capacity factor versus the square of the ionic polarizability. See text for additional detail.
(Adapted from J. High Resol. Chromatogr. 1990, 13, 293-294, with permission from John Wiley & Sons ©
1990.)
radial distance between the two species. How-
ever, when A and B are the same molecule
or ion, Eq. 1 is reduced to Eq. 2.
E = -U4* (2)
4 R6
Daignault took Eq. 2 from Huheey et al.,94
and is reminiscent of the attraction portion
of the Lennard-Jones (6-12) potential, which
applies to small nonpolar molecules. In an
ion chromatography column, the retentive
behavior is more likely to be related to the
London forces between elements of the col-
umn (especially the quaternary ammonium
moieties) and the anions rather than Lon-
don attractive forces between the anions
themselves. Accordingly, one might expect
a relationship between a(C104), a(-NR3+),
and k'. It is worth noting that the electro-
static force for two anions is repulsive, but
this does not affect the London dispersion
forces.
Martin studied the effects of organic
solvents on capacity factors.95 Adding aceto-
nitrile, in particular, can reduce retention
times to 1/3 their original value. Such be-
havior is not observed with methanol. Mar-
tin also considered the effect of an anion's
hydration energy. The Dionex IonPac AS11
column was used for these experiments; thus,
observations about the effects of MeCN,
MeOH, or DMSO might be used to improve
separations of refractory samples.
Several other studies examined funda-
mental behavior; however, a thorough analy-
sis and interpretation is beyond the scope of
this article. Okada modeled capacity factors
in terms of the electric double layer theory to
explain perchlorate's retention behavior.96
Watanabe and Kubota had previously de-
scribed this behavior in terms of streaming
potential.97 Pirogov et al. demonstrated that
temperature dependence in retention behav-
ior was similar for both methylmethacrylate
and styrene-divinylbenzene polymer matri-
ces.98
A variety of studies have looked at alter-
nate stationary phases. Takeuchi et al. showed
that alumina could be used to separate per-
chlorate from chlorate, chlorite, and chlo-
ride.99 Elkafir et al. used graphitic carbon to
separate perchlorate from phosphate, sulfate,
and nitrate with carboxylate-based eluents.100
Muenter et al. used an 8-hydroxyquinoline-
based stationary phase to separate perchlor-
ate from chloride, bromide, and nitrate using
a water/acetonitrile mobile phase.101 Based
on electrochemical studies with cobalt(HI)
331
-------
phthalocyanines, Kocsis et al. incorporated
these moieties into an octadecylsilane (Clg)
column for use in anion chromatography;
they separated perchlorate, nitrate, chloride,
iodide, and thiocyanate using acetate as the
eluent anion.102 Umemura et al. used
octadecylsilane columns coated sulfobetaine
surfactants to separate a mixture of anions,
including chloride, bromide, perchlorate,
chlorate, nitrate, iodide, thiocyanate, nitrite,
and sulfate.103 Pirogov et al. have tested a
series of ionenes as modifiers for the mobile
phase.104 Ionenes are polymers of the form
[NR2+-(CH2)a-NR2+-(CH2)6]n and are named
as ionene a-b.Tht quaternary ammonium
moieties of the ionene mimic the functional
groups of the stationary phase of a standard
IC column. As Figure 20 indicates, excellent
separation is possible with a standard
octadecylsilane/silica stationary phase pack-
ing.104 For now, these alternative stationary
phases are developmental and unlikely to
figure into routine monitoring anytime in the
near future.
Ion chromatography is destined to be the
main technique used for potable water in the
U.S. for the forseeable future. It couples low
detection limits (< 5 ng mL~l) with ease of
use, selectivity, and general availability. Low
detection limits permit many samples to be
diluted sufficiently to eliminate untoward
matrix effects. IC tends to be nigged in a
variety of matrices, and it has an established
history in the drinking water industry in the
U.S.
VI. MASS SPECTROMETRY
Because perchlorate is a small inorganic
anion, certain ionization techniques suggest
themselves: thermospray, ionspray, and
electrospray (ESI-MS). Descriptions can be
found elsewhere.105-106 While mass spectrom-
etry can be used as a detector subsequent to
separation by IC or CE, it is possible to
perform mass spectrometric analysis of a
sample without separation. Barnett and
Horlick used ESI-MS to obtain an LLOD of
5 ng mL~' for perchlorate.107 They examined
solutions of quaternary ammonium com-
pounds (mouthwashes) for perchlorate. At
the U.S. Air Force Research Laboratories,
Clewell et al. lyophilized water samples and
redissolved the the residue in acetonitrile/
acetic acid.108 109 Figure 21 shows a mass
spectrum of the perchlorate anion (m/z = 99
u). One advantage to this approach is that
carbonate is driven off. They obtained a
detection limit of 340 pg mL-1. Typical nega-
tive ion ESI-MS response is shown in Figure
22 for several concentrations of perchlorate
(single ion monitoring).
At the EPA, Urbansky et al. showed that
perchlorate could be complexed with
nonnucleophilic bulky organic bases
(diazabicyclo compounds) to give molecular
ions of the form HB(C104)2~; the best sensi-
tivity was obtained with chlorhexidine, which
is a minimally nucleophilic base.110 This
phenomenon was further exploited by
Magnuson et al., who demonstrated that
perchlorate could be extracted into methyl-
ene chloride with quaternary ammonium
cations (e.g., C10H21NMe3+); see Figure 23.111
The concentration of perchlorate in real water
samples was determined by standard addi-
tions using ESI-MS with an LLOD of about
30 to 300 pg mL"1. The matrix can affect the
sensitivity, but the variation was less than a
factor of 10 among tested water samples
(Figure 24). This work was followed by the
development of a microextraction proce-
dure"2 and an alternate procedure using
methyl isobutyl ketone (MIBK).113 The
LLOD in MIBK is about 5 ng mL"1 or 10 to
100 times that in CH2C12. The MIBK method
was applied to the analysis of bottled waters,
many of which contain high concentrations
of dissolved minerals. Although sparkling
mineral waters do not normally pose serious
problems for IC, they can be refractory to
ESI-MS methods, presumably due to com-
petition for the cationic surfactant and/or
332
-------
*8?
A.
¦ m » » m m Run
3-4 lonene
L
2-8 lonene
» i w
3-8 lonene
* « I
3-6 lonene
« mm
4-6 lonene
6-8 lonene
6-10 lonene
FIGURE 20. Typical ion chromatograms obtained using ionenes. Conditions; 3 x 50 mm column packed with
Silasorb S; 0.3 mM potassium hydrogen pnthalate eluent; pH 6.8; flow rate 1.0 mL min-1; uv-detection at 254
nm. (Reprinted from J. Chromatogr. A 1999, 850, 53-63, with permission from Elsevier Science © 1999.)
333
-------
• 9.11
S.IZ^
1 , If i ii ~ I ^fi in! nfijlliiii^nti ^Bifi
¦**r
FIGURE 21. Electrospray ionization mass spectrum of perchtorate in acetonitrile/acetic acid. Molecular ions:
CI04- (m/z = 99 u), CH3CN»CI(V" {m/z = 140 u), CH3C02H»CI04- (m/z = 159 u). (Courtesy of Air Force
Research Laboratories, Wright Patterson AFB [not subject to copyright].)
SO ppb
5 . 14
AH '2066405X2
AA 2j€2 63.37«0
25 ppb
4.46
AH 139230496
AA 1^54037632
i ppb
t> o.sppb a>ai
I 1.(4 AK #S3B80?
231X6 AH 4839>H 7«35«S1J
UOOC AA 34SS1SSO I
10 ppb
3. 76
AH 72824160
_ . AA 560769664
S ppb f
3 .22
AH 42425326
AA 34907932
T
2 • o
FIGURE 22. Typical ESI-MS response for perchtorate (single ion monitoring, m/z = 99 u) obtained In
acetonitrile/acetic acid. Courtesy of Air Force Research Laboratories, Wright Patterson AFB [not subject to
copyright].)
334
-------
too
ClO-BrCJO.
3SO
SO
£
Eft
a &
~
« 40
ct
¦3
is
340 Jtt 3M 4M 42#
m/z
FIGURE 23. ESI mass spectrum of complex anions found when extracting
perchiorate with decyltri methylammonium ion, C10H21N(CH3)3+, supplied as the
bromide salt. Observable complexes contain two anions. Three different anionic
complexes are seen: Br/Br, Br/CI04, and CIO4/CIO4. (Reprinted from Anal. Chem.
2000, 72, 25-29, under U.S. government authority [not subject to copyright].)
CIC-arBr
see
C10004004
m
«
"5
* De ionized
Cincinnati Tap !
4 Raw Ohio River
Simulated Tap
perchiorate concentration ([ig/L)
FIGURE 24. Single ion monitoring (m/z = 380 u) by ESI-MS for the CH2CI2-extracted
complex anion C10H21NMe3(Br)(CIO4)- in several different water matrices. (Reprinted
from Anal. Chem. 2000, 72, 25-29, under U.S. government authority [not subject to
copyright].)
-------
electrospray suppression due to ionic
strength. This method has also been applied
to determining perchlorate in tamarisk, a plant
growing in affected regions.114 A tandem
ESI-MS-MS system has been used for
groundwater,115 although the equipment is
not widely available.
One of the most recent developments is
high field asymmetric waveform ion mobil-
ity mass spectrometry (also known as
FAIMS). Bamett et al. have recently reported
using FAIMS to measure the concentrations
of perchlorate—even with high sulfate con-
centrations, but environmental samples were
not specifically tested.116 Chlorate and bro-
mate were readily identified by FAIMS. We
can expect continuing progress from FAIMS,
which appears to be capable of detection
limits in the parts-per-trillion.
Although mass spectrometry is a useful
research tool, it probably will not find wide-
spread use in the drinking water industry for
primary monitoring. This is primarily due to
operating costs (including expertise) and
instrument cost. Nonetheless, mass spectro-
metric methods may be expected to play
important roles in confirmatory identifica-
tion of perchlorate found using an IC or CE
retention time.
VII. ANALYZING COMPLEX
MATRICES
Although ground, surface, and treated
potable water are not the simplest matrices,
they are relatively easy to analyze compared
with fruit juices, sludges, milk, blood, sap
and other matrices likely to be of interest in
assessing possible alternate exposure routes
for perchlorate—not just in humans, but in
animals and plants that inhabit the affected
ecosystems. Our own laboratory has been
investigating IC, CE, and ESI-MS for the
quantitation of perchlorate in fertilizers,
where it is present in trace amounts if at all,
while other anions are the main components
(e.g., phosphate, nitrate, chloride).117-119 The
EPA's National Exposure Research Labora-
tory has just reported a new method for the
analysis of plant matter; the method focuses
on substantial clean-up prior to injection.120
The EPA has recently begun validation pro-
cedures for Method 317.0, which uses IC to
separate anions, but indirect detection based
on a redox reaction between bromate (or
other oxyanions) and a chromopho-
rogenic species, o-dianisidine (3,3'-
dimethoxy-4,4'-diaminobiphenyl or 3,3'-
dimethoxybenzidine).121 The reaction occurs
after the separation (post-column) and is
monitored photometrically. Although per-
chlorate reduction is kinetically hindered, it
is not known if it is possible to use post-
column chromogenic reaction for indirect
detection of perchlorate. Because some metal
ions are known to catalyze perchlorate re-
duction,2 it may be possible to use a redox
reaction by appropriate adjustment of condi
tions. This certainly would be an area wor-
thy of exploration.
Analyses of more complicated (especially
physiological or biological) matrices are hin-
dered by organic anions (e.g., sulfonates,
carboxylates, phenoxides), polysaccharides,
fatty molecules (e.g., polyols, alkaloids, li-
poproteins, phospholipids, sterols), proteins
(e.g., albumin, casein). Removing these spe-
cies is not necessarily straightforward. Fur-
thermore, verifying that sample pre-treatment
does not also remove the analyte will be
important. At present, separation techniques
such as CE and-IC have been applied with-
out extensive sample pretreatment, but fruit
juices and milk are refractory to determining
perchlorate when meaningful limits of de-
tection (<1 fig mL"1) are required. Online
dialysis has been applied successfully to both
IC and CE.122 It is undoubtedly a technique
worthy of investigation.
In order to estimate the potential for eco-
logical impact, it is necessary to survey rep-
resentative organisms for exposure. This is
where techniques amenable to single cell
336
-------
sampling and analysis are useful.123"125 They
make it possible to assess insects, annelids,
helminths, nematodes, cnidarians, or small
crustaceans (e.g., amphipods, isopods, cla-
docerans) with total body volumes of 1 to
100 mm3; these creatures can be useful indi-
cators of water quality. In addition to mak-
ing it possible to obtain a sample from or-
ganisms otherwise too small, minimizing the
volume of sample needed reduces the stress
on individual reptiles, fish, amphibians, birds,
or small mammals, for whom even 1 mL
represents a significant volume of blood.
Minimizing the need for sacrificing organ-
isms, combined with lower costs of sample
collection, storage, and shipping make mi-
croanalytieal techniques and methods ex-
tremely attractive for use in ecological or
biological studies. Human erythrocytes have
been subjected to this technique (total vol-
ume 90 fL; 1 femtoliter = 10~12 L); however,
enzymes have been a focus of these investi-
gations. Intrathyroid perchlorate has not been
determined this way, but perhaps it could be,
that is, by harvesting individual follicular
cells. EPA, USDA, and FDA have an inter-
est in determining whether perchlorate can
be transported and/or accumulated through
the food chain, but thorough ecological as-
sessments are time consuming and expen-
sive. Therefore, measuring perchlorate at the
lowest levels of the food chain is attractive
in terms of saving time and money.
VIII. CONCLUSIONS
Certainly in the near future, we can ex-
pect ion chromatography to dominate envi-
ronmental analytical chemistry both because
of the limit of detection and the availability
of the instrumentation in many laboratories.
Capillary electrophoresis appears to have
carved out a niche in forensic analysis and
could be used in many environmental appli-
cations if sufficiently sensitive detectors were
available. It is likely that Raman spectrom-
etry and mass spectrometry will be used for
secondary confirmation because they rely
on properties unrelated to hydrophilicity or
ionic mobility, both of which influence re-
tention/elution time. Electrochemical sensors
are probably best suited to online process
monitoring for quality control, but can be
expected to make no inroads into environ-
mental analysis due to their limited selectiv-
ity unless combined with separation tech-
niques such as IC or CE. Lastly, the need to
analyze complex matrices (e.g., foods, bev-
erages, or body/plant fluids) will exact new
demands on all of these techniques and will
require the implementation of more sophis-
ticated sample clean-up and pre treatment
steps prior to analysis. Hyphenated tech-
niques are apt to become important in the
analysis of more complex matrices because
they can improve selectivity, for example,
LC-MS-MS, CE-MS, or CE-Raman spec-
trometry29 as has been done already. In the
end, the choice of techniques and methods
for the quantitation of perchlorate will come
down to central issues'of analytical chemis-
try: selectivity and sensitivity. Within limits
of cost and availability, whatever instrumen-
tation can meet those needs will be pursued.
ACKNOWLEDGMENTS
The assistance of Jennifer Heffron,
Raymond A. Hauck, Betty L. Merriman, and
Jennie Thomas in gathering materials for
this manuscript is recognized.
REFERENCES
1. Linke, S.R. Managing Crises in Defense In-
dustry: The PEP CON and Aviex Cases;
McNair Papers, No. 9. Institute for National
Strategic Studies: Washington, D.C., 1990.
2. Urbansky, E.T. Biorem. Perchlorate chem-
istry: implications for analysis and
remediation. 1998, 2, 81-95, and references
therein.
337
-------
3. Urbansky, E.T.; Schock, MR. J. Environ.
Manage., Issues in managing the risks associ-
ated with perchlorate in drinking water, 1999,
56, 79-95, and references therein.
4. D ami an. P.; Pontius, F.W, Environ. Prot.,
From rockets to remediation: the perchlorate
problem. 1999, 10, 24-31.
5. Renner, R. Environ. Set Technol., EPA draft
almost doubles safe dose of perchlorate in
water. 1999, 33, 110A-111A.
6. Renner, R. Environ. Sci. Technol Perchlor-
ate-tainted wells spur government action.
1998, 32, 21OA.
7. Espenson, J.H. The Problem and Perversity
of Perchlorate. In Perchlorate in the Environ-
ment; Urbansky, E.T., Ed. Plenum: New York:
2000, Ch. 1, and references therein.
8. Schilt, A.A. Perchloric Acid and Perchlor-
ates. GFS Chemicals: Columbus, OH, 1979,
and references therein. GFS advises that a
2nd edition is in preparation.
9. Clark, J.I.J. Toxicology of Perchlorate. In Per-
chlorate in the Environment; Urbansky, E.T.,
Ed. Plenum: New York: 2000; Ch. 3, and
references therein.
10. Wolff, J. Pharm. Revs., Perchlorate and the
thyroid gland. 1998, 50, 89-105, and refer-
ences therein.
11. Foye, W.O., Ed. Principles of Medicinal
Chemistry, 3rd ed. Lea & Febiger: Philadel-
phia, PA, 1989; pp. 612-613.
12. Perciasepe, R. Fed. Re gist., Part III. Envi-
ronmental Protection Agency. Announce-
ment of the drinking water contaminant can-
didate list; notice. 1998, 63 (40),
10273-10287.
13. Environmental Protection Agency. Drinking
Water Contaminant Candidate List, February
1998, EPA Doc. No. 815-F-98-002.
14. Browner, C. Fed. Re gist.. Part II. Environ-
mental Protection Agency. 40 CFR parts 9,
141 and 142. Revisions to unregulated con-
taminant monitoring regulation for public
water systems; final rule. 1999, 64 (180),
50555-50620.
15. Lamb, A.B.; Marden, I.W. J. Am. Chem. Soc.,
The quantitative determination of perchlor-
ates. 1912, 34, 812-817.
16. Nabar, G.M.; Ramachandran, C.R. Anal.
Chem., Quantitative determination of perchlo-
rate ion in solution. 1959, 31, 263-265.
17. Loebich, O. Z Anal. Chem. A new method
for the determination of perchloric acid (in
German). 1926, 68, 34-36.
18. Welcher, F.J. Organic Analytical Reagents.
Van Nostrand: New York, NY, 1947; Vol. 3,
pp. 138-147.
19. Wyngaarden, I.B.; Wright, B.M.; Ways, P.
Endocrinology, Effect of certain anions on
accumlation and retention of iodide by the
thyroid gland. 1952, 50, 537-549.
20. Collinson, W.J.; Boltz, D.F. Anal. Chem,, In-
direct spectrophotometric and atomic absorp-
tion methods for determination of perchlor-
ate. 1968, 40, 1896-1898, and references
therein.
21. Weiss, J.A.;Stanbury, J.B. Anal. Chem., Spec-
trophotometric determination of micro
amounts of perchlorate in biological fluids.
1972,44,619-620. See also references therein
for other methods at the time.
22. Bums, D.T.; Dunford, M.D.; Sutthivaiyakit,
P. Anal. Chim. Acta, Spectrophotometric de-
termination of perchlorate after extraction as
protriptylinium perchlorate. 1997, 356, 141-
143.
23. Kormosh, 2.A.; Bazel, Y.R. J. Anal. Chem.,
Extraction of oxyanions with basic
polymethine dyes from aqueous and aque-
ous-organic solutions: extraction-photomet-
ric determination of rhenium(VH) and
tungsten(VI). 1999, 54, 607-611.
24. Niikura, K.; Bisson, A.P.; Anslyn, E.V. J.
Chem. Soc. Perkin Trans. 2, Optical sensing
of inorganic""anions employing a synthetic
receptor and ionic colorimetric dyes. 1999,
1111-1114.
25. Ensafi, A.A.; Rezaei, B. Anal. Lett., Auto-
matic liquid-liquid extraction flow injection
analysis determination of trace amounts of
perchlorate with spectrophotometric detection.
1998, 31, 167-177.
26. Hisamoto, H.; Tohma, H,; Yamada, T.
Yamauchi, K.; Siswanta, D,; Yoshioka, N.;
Suzuki, K. Anal. Chim. Acta, Molecular de-
sign, characterization, and application of
338
-------
multi-information dyes for multi-dimensional
optical chemical sensing. Molecular design
concepts of the dyes and their fundamental
spectral characteristics. 1998, 373, 271-289.
27. Drago, R.S. Physical Methods for Chemists,
2nd ed. Saunders: Ft. Worth, TX, 1992; Ch. 6.
28. Eysel, H. Spectrochim. Acta, Raman intensi-
ties of liquids: absolute scattering activities
and CI- 0 bond EOPs of CIO, C102", C103~,
and C104" ions in aqueous solution. 1988,
44A, 991-997.
29. Miller, A.G.; Macklin, J.A. Anal. Chem.,
Matrix effects on the Raman analytical lines
of oxyanions. 1980, 52, 807 812.
30. Kowalchyk, W.K.; Walker, P.A. Ill; Morris,
M.D. Appl. Spectrose., Rapid normal Raman
spectroscopy of sub-ppm oxy-anion solutions:
the role of electrophoretic preconcentration.
1995, 49, 1183-1188.
31. Susarla, S.; Collette, T.W.; Garrison, A.W.;
Wolfe, N.L.; McCutcheon, S.C. Environ. Sci.
Technol, Perchlorate identification in fertil-
izers. 1999, 33, 3469-3472. See also correc-
tion thereto. 2000, 34, 224.
32. Williams, T.L., Martin, R.B.; Collette, T.W.
Appl. Spectrosc., Raman spectral analysis of
perchlorate contamination in commonly-used
fertilizers. 2000, submitted for publication.
33. Hofmeister, F.Arch. Exp. Pathol. Pharmakoi,
Zur Lehr von der Wirkungder Salze. liber
Regelmassigkeiten in der eiweissfallenden
Wirkung der Salze und ihre Beziehung zum
physiologischen Verhalten derselben. 1888.
24, 247-260.
34. Koizumi, S.; Imato, T.; Ishibashi. N. J. Membr.
Sci., Bi-ionic membrane potential across a
liquid anion-exchange membrane containing
triphenyltin chloride. 1997, 132, 149-158.
35. Koizumi, S.; Imato, T.; Ishibashi, N. Fresenius
J. Anal. Chem., Effect of concentrations of
triphenyltin chloride and anion exchanger on
the anion selectivity of a plasticized PVC
membrane electrode. 1997, 357, 37-43.
36. Ying, M.; Yuan, R.; Zhang, X.-M.; Song, Y.-
Q.; Li Z.-Q.; Shen, G.-L.; Yu, R.-Q. Analyst,
Highly selective iodide polyvinyl chloride)
membrane electrode based on a nickel(II)
tetraazaannulene macrocyclic complex. 1997,
122, 1143-1146.
37. Volf, R.; Shishkanova, T.V.; Matjka, P.;
Hamplova, M.; Krai, V. Anal. Chim. Acta,
Potentiometric anion response of poly(5,15-
bis-(2-aminophenyl)porphyrin) electro-poly-
merized electrodes. 1999, 381, 197-205.
38. Zhou, Z.; Wang, Y.; Tao, J.; Fan, Y.; Wu, Y. J.
Inclusion Phenom. Mol. Recognit. Chem., A
substituted tetraaza porphyrinogen as an
electroactive component for a polymeric mem-
brane anion-selective electrode. 1998, 32, 69-
80.
39. Lin, X.M.; Umezawa, K.; Tohda, K.; Furuta,
H.; Sessler, I.L.; Umezasa, Y. Anal. Sci., Po-
tentiometric responses of expanded porphy-
rin incorporated liquid membrane electrodes
toward a series of inorganic and organic an-
ions. 1998, 14, 99-108. •
40. Li Z.-Q.; Wu, Z.-Y.; Yuan, R.; Ying, M.;
Shen, G.-L.; Yu, R.-Q. Electrochim. Acta,
Thiocyanate-selective PVC membrane elec-
trodes based on Mn(II) complex of N,N.-bis-
(4- phenylazosalicylidene) o-phenylenedi-
amine as a neutral carrier. 1999, 44,
2543-2548.
41. Negash, N.; Moges, G.; Chandravanshi, B.S.
Chem. Anal. (Warsaw), Liquid membrane
electrode based on Brilliant Green-hydrogen
phthalate ion pair. 1997, 42, 579-588.
42. Negash, N.; Chandravanshi, B.S.; Moges, G.
Bull. Chem. Soc. Ethiop., Ion pair extraction
constants of some anions and selectivity of
liquid membrane electrodes based on Crystal
Violet and Brilliant Green. 1997, 11, 93-99.
43. Siswanta, D.; Takenaka, J.; Suzuki, T,;
Sasakura, H.; Hisamoto, H.; Suzuki, K. Chem.
Lett., Novel neutral anion ionophores based
on fluorinated (poly)ether compounds as a
sensory molecule for an ion-selective elec-
trode. 1997, 195-196.
44. Errachid, A.; Perez-Jimenez, C.; Casab6, J.;
Escriche, L.; Mufioz, J.A.; Bratov, A.;
Bausells, J. Sens. Actuators B, Perchlorate-
selective MEMFETs and ISEs based on a
new phosphadithia macrocycle. 1997, 43,
206-210.
45. Jezkova, J.; Musilovd, J.; Vytras, K. Elec-
troanalysis, Potentiometry with perchlorate
and fluoroborate ion-selective electrodes.
1997, 9, 1433-1436.
339
-------
46. Neuhold, C.C.; Kalcher, K.; Cai, X.; Raber,
G. Anal. Lett., Catalytic determination of per-
chlorate using a modified carbon paste elec-
trode. 1996, 29, 1685-1704.
47. Xu, G.; Dong, S. Electrochem, Commun.,
Electrochemiluminescent determination of
perchlorate by solvent extraction in the pres-
ence of Ru(bpy)32+, 1999, 1, 463-466.
48. Grossman, P.D.; Colbum, J C, Eds. Capil-
lary Electrophoresis: Theory and Practice.
Academic Press: San Diego, CA, 1992.
49. Jandik, P.; Bonn, G. Capillary Electrophore-
sis of Small Molecules and Ions. VCH: New
York, NY, 1993. Perchlorate is specifically
mentioned on pages 160 and 261.
50. Baker, Dale R. Capillary Electrophoresis.
Wiley-Interscience; New York, NY, 1995.
Perchlorate is specifically mentioned on page
43.
51. Li, S.F.Y. Capillary Electrophoresis: Prin-
ciples, Practice and Applications. Elsevier:
Amsterdam, the Netherlands, 1992.
52. Nann, A.; Pretsch, E. J. Chromatogr. A, Po-
tentiometric detection of anions separated by
capillary electrophoresis using an ion-selec-
tive electrode. 1994, 676, 437-442.
53. Ehmann, T.; Bachman, K.; Fabry, L.; Riifer,
H.; Pahlke, S.; Kotz, L. Chromatographic,
Optimization of the eiectrokinetic sample in-
troduction in capillary electrophoresis for the
ultra trace determination of anions on silicon
wafer surfaces. 1997, 45, 301-311.
54. Hargadon, K.A.; McCord, B.R. J.
ChromatogrExplosive residue analysis by
capillary electrophoresis and ion chromatog-
raphy. 1992, 602, 241-247.
55. Kishi, T.; Nakamura, I.; Arai, H. Electro-
phoresis, Application of capillary electro-
phoresis for the determination of inorganic
ions in trace explosives and explosive resi-
dues. 1998, 19, 3-5.
56. Warren, D.; Hiley, R.W.; Phillips, S.A.;
Ritchie, K. Set Justice» Novel technique for
the combined recovery, extraction and clean-
up of forensic organic and inorganic trace
explosives samples. 1999, 39, 11-18.
57. Park, S.-W.; Jin, K.-H.; You, J.-H.; Kim, T.-
J,; Paeng, K.-J.; Kong, K.-H. Anal. Set., Analy-
sis of inorganic anions by capillary zone elec-
trophoresis. 1997, 13, 243-246.
58. Xiong, X.; Li, S.F.Y. J. Chromatogr. A, De-
sign of background electrolytes for indirect
photometric detection based on a model of
sample zone absorption in capillary electro-
phoresis. 1999, 835, 169-185.
59. Xiong, X.; Li, S.F.Y. Electrophoresis» Dual
UV-absorbing background electrolytes for si-
multaneous separation and detection of small
cations and anions by capillary zone electro-
phoresis. 1998, 19, 2243-2251.
60. Kitagawa, S.; Tsuji, A.; Watanabe, H.;
Nakashima, M.; Tsuda, T. J. Micro. Sep.,
Pressurized flow-driven capillary
electrochromatography using ion exchange
resins. 1997, 9, 347-356.
61. Biesaga, M.; Kwiatkowska, M.; Trojanowicz,
M. J. Chromatogr. A, Separation of chlorine-
containing anions by ion chromatography and
capillary electrophoresis. 1997,777,375-381.
62. Doble, P.; Macka, M.; Haddad, P.R. J.
Chromatogr. A, Use of dyes as indirect detec-
tion probes for the high-sensitivity determi-
nation of anions by capillary electrophoresis.
1998, 804, 327-336.
63. Harrold, M.P.; Wojtusik, M.J.; Riviello, I.;
Henson, P. J. Chromatogr., Parameters influ-
encing separation and detection of anions by
capillary electrophoresis. 1993,640,463-471.
64. Shamsi, S.A.; Danielson, N.D. Anal. Chem.,
Naphthalenesulfonates as electrolytes for cap-
illary electrophoresis of inorganic anions,
organic acids, and surfactants with indirect
photometric detection. 1994, 66, 3757-3764.
65. Krokbin, O.V.; Hoshino, H.; Shpigun, O.A.;
Yotsuyanaffi, T. J. Chromatogr. A, Use of
cation ic polymers for the simultaneous deter-
mination of inorganic anions and metal-4-(2-
pyri dylazo)resoreinolato chelates in kinetic
differentiation-mode capillary electrophore-
sis. 1997, 776, 329-336.
66. Stathakis, C.; Cassidy, R.M. Can. J. Chem.,
Control of relative migration of small inor-
ganic and organic anions with cyclodextrios
in capillary electrophoresis (CE). 1998, 76,
194-198.
67. Avdalovic, N.; Pohl, C.A.; Rocklin, R.D.;
Stillian, J.R. Anal, Chem., Determination of
340
-------
cations and anions by capillary electrophore-
sis combined with suppressed conductivity
detection. 1993, 65, 1470-1475.
68. Gross, L.; Yeung, E.S. J. Chromatogr., Indi-
rect fluorimetric detection and quantification
in capillary zone electrophoresis of inorganic
anions and nucleotides. 1989,480, 169-178.
69. Macka, M,; Gerhardt, G.; Andersson, P.;
Bogan, D.; Cassidy, R.M.; Haddad, PR. Elec-
trophoresis, Capillary electrophoresis with
end-capillary potentiometric detection using
a copper electrode. 1999, 20, 2539-2546.
70. Kappes, T.; Schnierle, P.; Hauser, P.C. Anal.
Chim. Acta, Potentiometric detection of inor-
ganic anions and cations in capillary electro-
phoresis with coated-wire ion-selective elec-
trodes. 1997, 350, 141-147.
71. Corr, J.J.; Anacleto, I. F. Anal. Chem., Analy-
sis of inorganic species by capillary electro-
phoresis-mass spectrometry and ion exchange
chromatography-mass spectrometry using an
ion spray source. 1996, 68, 2155-2163.
72. Hsu, J.-C.; Chen, W.-H.; Liu, C.-Y. Analyst,
Preparation of a macrocyclic polyamine-
bonded column for the electrophoretic sepa-
ration of inorganic and organic anions. 1999,
122, 1393-1398.
73. Hauser, P.C.; Hong, A.P.C.; Renner, N. J.
Cap. Elec., Surface charge reversal for inor-
ganic anion determination in capillary elec-
trophoresis with an ion-selective microelec-
trode as detector. 1995, 2, 209-212.
74. Hilder, E.F.; Macka, M.; Haddad, P R. Anal.
Commutt., Mixed-mode capillary electro chro-
matographic separation of anionic analytes.
1999, 36, 299-303.
75. Meissner, T.; Eisenbeiss, F.; Jastorff, B. J.
Chromatogr. A, New leading electrolyte for
the direct determination of chloride and other
anions in analytical isotachophorcsis. 1999,
838, 81-88.
76. Small, H. Ion Chromatography. Plenum Press:
New York, NY, 1990.
77. Okada, T. J. Chromatogr. A, Nonaqueous on-
ion-exchange chromatography. I. Role of solva-
tion in anion-exchange resin. 1997, 758.19-28.
78. California Department of Health Services,
Sanitation and Radiation Laboratories Branch.
Determination of Perchlorate by Ion Chro-
matography, Rev. 0. Jun. 3, 1997.
79. Okamoto, H.S.; Rishi, D.K.; Steeber, W.R.;
Baumann, F.J.; Perera, S.K. J. Am. Water
Works Assoc„ Using ion chromatography to
detect perchlorate. 1999, 91, 73-84.
80. Miura, Y.; Kimura, M.; Ocfai, M.; Koh, T,
Proc. - School ScL, Tokai Univ., Effect of
eluents on suppressed ion chromatography
using anion exchange column of Dionex
HPIC-AS5. 1996, 31, 83-90.
81. Maurino, V,; Minero, C. Anal. Chem., Cya-
nuric acid-based eluent for suppressed anion
chromatography. 1997, 69, 3333-3338.
82. Wirt, K.; Laikhtman, M.; Rohrer, J.; Jackson,
P.E. Am. Environ. Lab., Low level perchlor-
ate analysis in drinking water and ground
water by ion chromatography, 1998,10, 1,5.
83. Jackson, P.E.; Laikhtman, M.; Rohrer, J. Proc.
- Water Qual. Technol. Conf, An improved
chromatographic method for the determina-
tion of trace level perchlorate in drinking water
and ground water. AWWA: San Diego, CA,
Nov. 1-4, 1998, 2058 2065.
84. Jackson, P.E.; Laikhtman, M.; Rohrer, J.S. J.
Chromatogr. A, Determination of trace level
perchlorate in drinking and ground water by
ion chromatography. 1999, 850, 131-135.
85. Jackson, P.E.; Gokhale, S.; Rohrer, J.S. Re-
cent Developments in the Analysis of Per-
chlorate Using Ion Chromatography. In Per-
chlorate in the Environment; Urbansky, E.T.,
Ed. Kluwer,'Plenum: 2000; Ch. 5.
86. Harrington, P.; Shen, Y. Proc. - Water Qual.
Technol, Conf., Improved analysis of perchlo-
rate in water. AWWA: San Diego, CA, Nov.
1-4, 1998, 1405-Ml 2.
87. Nair, L.M.; Saari-Nordhaus, R.; Montgomery, R.M.
J. Chromatogr. A, Applications of a new metfaaay-
late-based anion stationary phase for the separation
of inorganic anions. 1997, 789,127-134.
88. Eaton, A.; Haghani, A.; Cox, N.; Wong. E.
Proc, - Water Qual. Technol. Conf, A com-
parison of the raggedness of different analyti-
cal methods for perchlorate. AWWA: San
Diego, CA, Nov. 1-4, 1998, 1413-1423.
89. Tsui, D.T.; Clewell, R.A.; Eldridge, J.E.;
Mattie, D.R. Perchlorate Analysis with the
341
-------
AS 16 Column. In Perchlorate in the Envi-
ronment; Urbansky, E.T., Ed. Kluwer/Plenum:
2000; Ch. 7.
90. Hautman, D.P.; Munch, D.J.; Eaton, A.D.;
Haghani, A.W. Method 314.0. Determination
of Perchlorate in Drinking Water Using Ion
Chromatography, Revision 1.0. Environmen-
tal Protection Agency: Cincinnati, OH, 1999.
EPA Doc. No. 815-B-99-0Q3.
91. Buchberger, W.; Haider, K. J. Chromatogr.
A, Studies on the combination of ion chromato
graphy-particle-beam mass spectrometry with
capillary columns. 1997, 770, 59-68.
92. Daignault, L.G.; Rillema, D.P.; Jackman, D.C.
J. High Resol. Chromatogr., Retention be-
havior of ions on high performance ion chro-
matography columns. Part II: Retention be-
havior and polarizability. 1990,13, 293-294.
93. Atkins, P.W. Molecular Quantum Mechan-
ics, 2nd ed. Oxford University Press: Oxford,
UK, 1983; p. 364.
94. Huheey, J.E.; Keiter, E.A; Keiter, R.L. Inor-
ganic Chemistry: Principles of Structure and
Reactivity, 4th ed. Harper-Collins: New York,
NY, 1993; p. 299.
95. Martin, M.W. Microchem. J., Selectivity of
some ion chromatography stationary phases
for small anions in solvent-water mixtures
with hydroxide. 1999, 62, 203-222.
96. Okada, T. J. Chromatogr. A, Interpretation of
chromatographic behavior of ions based on
the electric double-layer theory. 1999, 850,
3-8.
97. Watanabe, T.; Kubota, M. Anal. Chim. Acta,
Electrostatic effect of streaming potential on
ion- exchange phenomena in resins for ion
chromatography. 1990, 22, 61-67.
98. Pirogov, A.V.; Obrezkov, O.N.; Shpigun, OA.
J. Anal. Chem., Temperature as a parameter
of optimization of the ion chromatographic
determination of anions. 1997, 52, 152-156.
99. Takeuchi, T.; Suzuki, E.; Ishii, D.
Chromatographia, Ion chromatography with
micropacked alumina columns. 1988,25,480-
482.
100. Elfakir, C.; Chainbault, P.; Dreux, M. J.
Chromatogr. A, Determination of inorganic
anions on porous graphitic carbon using
evaporative light scattering detection. Use of
carboxylic acids as electronic competitors.
1998, 829, 193 199.
101. Muenter, M.M.; Stokes, K.C.; Obie, R.T.;
Jezorak, J.R. J. Chromatogr. A, Simultaneous
separation of inorganic ions an neutral organ-
ics on ion-exchange stationary phases. 1999,
844, 39 51.
102. Kocsis, E.M.; Koizummi, S.; Sakane, T.;
Takahashi, K.; Imato, T. Anal. Sci., Separa-
tion behavior of anions on ODS modified
with phthalocyanine cobalt(III) complex.
1997, 13, 195-198.
103. Umemura, T.; Kamiya, S.; Itoh, A.; Chiba,
K.; Haraguchi, H. Anal. Chim. Acta, Evalua-
tion of sulfobetaine-type zwitterionic station-
ary phases for ion chromatographic separa-
tion using water asa mobile phase. 1997,349,
231-238.
104. Pirogov, A.V.; Platonov, M.M.; Shpigun, O.A.
J. Chromatogr. A, Polyelectrolyte sorbents
based on aliphatic ionenes for ion chromatog-
raphy. 1999, 850, 53-63.
105. Snyder, A.P. ACS Symposium Series 619:
Biochemical and Biotechnological Applica-
tions of Electrospray Ionization Mass Spec-
trometry. American Chemical Society: Wash-
ington, DC, 1995.
106. Watson, J.T. Introduction to Mass Spectrom-
etry, 3rd ed. Lippincott-Raven: Philadelphia,
1997.
107. Barnett, D.A.; Horlick, G. J. Anal. At.
Spectrom., Quantitative electrospray mass
spectrometry of halides and halogenic anions.
1997, 12, 497-501.
108. Clewell, R.E.; Tsui, D. Prepr. Ext. Abstr.
ACS Natl. -Meet., Am. Chem. Soc. Div.
Environ. Chem. Presented at the 218th Ameri-
can Chemical Society national meeting. New
Orleans, LA, USA, Aug. 22-26, 1999, 39,
71-74.
109. Clewell, R.; Chaudhuri, S.; Dickson, S.;
Cassady, R.S.; Wallner, W.M.; Eldridge,
J.E.; Tsui, D. Analysis of Trace Level Per-
chlorate in Drinking Water and Ground
Water by Electrospray Mass Spectrometry.
In Perchlorate in the Environment;
Urbansky, E.T., Ed. Kluwer/Plenum: 2000;
Ch. 6.
342
-------
110. Urbansky, E.T.; Magr.uson, M.L.; Freeman,
D.; Jelks, C. J. Anal. At, Spectrom.
Quantitation of perchlorate ion by electrospray
ionization mass spectrometry (ESI-MS) us-
ing stable association complexes with organic
cations and bases to enhance selectivity, 1999,
14, 1861-1866,
111. Magnuson,M.L.;Urbansky,E.T.; Kelty,C.A.
Anal. Chem., Determination of perchlorate at
trace levels in drinking water by ion-pair ex-
traction with electrospray ionization mass
spectrometry. 2000, 72, 25-29,
112. Magnuson, M.L.; Urbansky, E.T.; Kelty, C.A.
Talanta, Microscale extraction of perchlorate
in drinking water with low level detection by
electrospray-mass spectrometry. 2000, 52,
285-291.
113. Urbansky, E.T.; Gu, B ; Magnuson, M.L.;
Brown, G.M., Kelty, C.A. J. Sci. FoodAgric.,
Survey of bottled waters for perchlorate by
electrospray ionization mass spectrometry
(ESI-MS) and ion chromatography (IC). 2000,
80. 1798-1804..
114. Urbansky, E.T.; Magnuson, M.L.; Kelty, C.A.;
Brown, S.K. Sci. Tot. Environ., Perchlorate
uptake by salt cedar (Tamarix ramosissima)
in the Las Vegas Wash riparian ecosystem.
2000, 256, 227-232.
115. Koester, C.J.; Beller, H.R.; Halden, R.U.
Environ. Sci. Technol., Analysis of perchlor-
ate in groundwater by electrospray ionization
mass spectrometry/mass spectrometry. 2000,
34, 1862-1864.
116. Handy, R.; Barnett, D.A.; Purves, R.W.;
Horlick, G.; Guevremont, R. J. Anal. At.
Spectrom., Determination of nanmolar levels
of perchlorate in water by ESI-FAJMS-MS.
2000, 15, 907-911..
117. Urbansky, E.T.; Brown, S.K.; Magnuson,
M.L.; Kelty, C.A. Environ. Pollut., Distribu-
tion of perchlorate in samples of sodium ni -
trate fertilizer derived from Chilean caliche.
2000, 110, in press.
118. Urbansky, E.T.; Gu, B,; Furman, C.S.;
Magnuson, M.L.; Brown, G.M.; Kelty, C.A.
Quantitation of perchlorate ion in fertilizers:
a comparison of ion chromatography,
electrospray ionization mass spectrometry, and
capillary electrophoresis. In preparation.
119. Urbansky, E.T.; Gu, B.; Magnuson, M.L.;
Brown, G.M.; Kelty, C.A. Environ. Sci.
Technol., Comment on 'perchlorate identifi-
cation in fertilizers.' 2000, 34, in press.
120. Ellington, J. J.; Evans, I J. J. Chromatogr. A,
Determination of perchlorate at parts-per-bil-
lion levels in plants by ion chromatography.
2000, in press,
121. Wagner, H.P.; Pepich, B.V.; Hautman, D.P.;
Munch, D.J. Determination of Inorganic
Oxyhalide Disinfection Byproducts in Drink-
ing Water Using Ion Chromatography with
the Addition of a Postcolumn Reagent for
Trace Bromate Analysis. Environmental Pro-
tection Agency: Cincinnati, OH, 1999; Draft
Method 317.0.
122. Kuban, P.; Kariberg, B .Anal. Chem., On-line
dialysis coupled to a capillary electrophoresis
system for determination of small anions.
1997,68, 1169-1173, and references therein.
123. Hogan, B.L.; Yeung, E.S. Trends Anal. Chem.,
Single-cell analysis at the .level of a single
human erythrocyte. 1993, 12, 4-9.
124. Sims, C.E.; Meredith, G.D.; Krasieva, T.B.;
Berns, M.W.; Tromberg, B.J.; Allbritton,
N.L. Anal. Chem., Laser-micropipet combi-
nation for single-cell analysis. 1998, 70,
4570-4577.
125. Swanek, F.D.; Ferris, S.S.: Ewing, A.G. Cap-
illary Electrophoresis for the Analysis of
Single Cells: Electrochemical, Mass Spectro-
metric, and Radiochemical Detection. In
Handbook of Capillary Electrophoresis, 2nd
ed.; Landers, J.P., Ed. CRC Press: Boca Raton,
FL, 1997; 495-521.
126. Lillard, S.J.; Yeung, E.S. Capillary Electro-
phoresis for the Analysis of Single Cells:
Laser-Induced Fluorescence Detection, In
Handbook of Capillary Electrophoresis, 2nd
ed.; Landers, J.P., Ed. CRC Press: Boca
Raton,,FL, 1997; 523-544.
343
------- |