CHRO~T~
ELSEVIER
A
Journal of Chromatography A, 703 (1995) 667-692
Review
Capillary electrophoresis-mass spectrometry
Jianyi Cai, Jack Henion*
Analytical Toxicology, Cornell University, 927 Warren Drive, Ithaca, NY 14850, USA
Abstract
As an on-line separation method, capillary electrophoresis (CE)-mass spectrometry (MS) distinguishes analytes
by both their differences in electrophoretic mobilities and structural information. CE-MS combines the advantages
of CE and MS so that information on both high separation efficiency and molecular masses and/or fragmentation
can be obtained in one analysis. During the past few years CE-MS has undergone significant development both in
instrumentation and application. Several ionization methods have been used for CE-MS. These include
electrospray ionization, ion spray or pneumatically assisted electrospray ionization, and continuous-flow fast atom
bombardment. The direct coupling of CE to desorption MS has not yet been reported, although publications have
appeared on the off-line coupling of CE with matrix-assisted laser resorption ionization and Z52Cf plasma
desorption MS using fraction collection. Numerous new applications of CE-MS have been published in the areas of
biological sciences, pharmaceutical and drug metabolism, and environmental analysis. The majority of applications
of CE-MS have been in the field of biological and biochemical studies.
Several limitations associated with CE-MS have precluded the technique being widely accepted for routine
analysis. The major limitation is its relatively poor concentration sensitivity. The concentration detection limits of
currently available CE-MS instrumentation are too high for most real-world applications. Other drawbacks with
CE-MS include the fluctuation in analyte migration time and limitations in electrolyte selections. Approaches to
improve the concentration sensitivity of CE-MS include on-line precoacentration either by capillary isotachophoresis or chromatographic methods. Another solution to increasing the -ensitivity of CE-MS is the development
of alternative types of mass spectrometers which offer the potential for greater sensitivity, such as ion trap, Fourier
transform ion cyclotron resonance, and time-of-flight (TOF) mass spectrometers. Coated capillaries are useful in
improving separation efficiencies of biomolecules by minimizing their adsorption onto the CE capillary walls.
At present, CE-MS is still not generally considered for routine analysis mainly due to its limited concentration
3ensitivity. As a complementary separation method to LC-MS with extremely high efficiency, CE-MS has the
potential for wide acceptance in the future. The popularity of CE-MS will continue to grow, as more sensitive MS
instrumentation and CE-MS interface are developed.
Contents
1. Introduction and historical review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2. CE-MS interface overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.1. Electrospray . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.1.1. Sheath-flow interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
* Corresponding author.
0021-9673/95t509.50
©
1995 Elsevier Science B.V. All rights reserved
SSD! 0021-9673(94)01178-8
668
669
670
670
668
J. Cai, J. Henion I J. Chromatogr. A 703 (1995) 667-692
2.1.2. Sheathlessinterface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.1.3. Comparison o f sheath-flowand sheathlessconfigurations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.2. Ion spray . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.2.1. Liquid junction interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.2.2. Sheath-flow interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.2.3. Comparison of liquid junction and sheath-flowconfigurations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.3. Continuous-flow fast atom bombardment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.3.1. Liquid junction interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.3.2. Sheath-flow interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.3.3. Comparison of liquid junction and sheath-flowconfigurations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.4. Comparison o f electrospray and ion spray with continuous-flow fast atom bombardment . . . . . . . . . . . . . . . . . . . . .
2.5. Other approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.5.1. Other interfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.5.2. Off-line C E - M S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3. Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.1. Biologically important components . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.I.1. Peptides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.1.2. Proteins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
: .......................................
3.1.3. Enzymaticdigests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.1.4. Other biopolymers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.1.5. Non-covalent binding study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.2. Pharmaceuticaland drug metabolism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.2.1. Drugs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.2.2. Drug metabolites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.3. Environmental . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.4. Other applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4. Limitationsand solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.1. Limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.2. Analyticalmethodology improvement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.2.1. On-line preconcentrationby capillary isotachophoresis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.2.2. Chromatographicon-line preconcentration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.2.3. Other analytical approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.3. Instrumentaldevelopments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.3.1. Coated capillaries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.3.2. Trapped ion mass spectrometers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.3.3. Time o f flight . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.3.4. Use of an arraydetector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1. Introduction and historical review
Microscale separation techniques, including
high-performance capillary liquid chromatography and capillary electrophoresis have a number of practical advantages ever conventionalscale analytical separation methods. Among
them are high separation efficiency, high speed,
and economy of sample size. High-performance
capillary electrophoresis (HPCE) is an important
microseparation technique in life sciences, as
well as biotechnology and environmental research areas. Unlike high-performance liquid
671
673
673
673
674
674
674
675
675
676
676
676
676
677
678
678
678
678
679
679
679
679
679
681
682
683
683
683
684
684
685
686
686
686
687
688
688
688
688
chromatography (HPLC), in which separation is
due to the partition of solutes between the
mobile phase and stationary phase, separation by
CE is based on the difference in charge-to-mass
ratio of the analytes. Therefore, a totally different selectivity is expected fox the analytes providing a complementary separation method to
HPLC.
Since the first demonstration of high separation efficiency with CE [1], the technique has
significantly advanced. CE exhibits unparalleled
resolvir.g power for condensed-phase separations. It is advantageous in solving problems
J. Cai, J. Henion / Z Chromatogr. A 703 (1995) 667-692
Time(m~n)
669
I
Fig. 1. Generic illustration of a CE-MS instrumental arrangement. HV= High voltage.
where gh separation efficiencies are required in
the case of analyzing very complex biological
mixtures and where the sample amount is limited. An extreme case is the analysis of the
contents of a single cell [2]. However, the limited
Ioadability of CE also poses high demands on the
detector sensitivity. This has led to a great deal
of effort towards the developmeL~ and improvement of CE instrumentation. A number of
commercial CE instruments have become availeble. As a result, CE continues to attract interest. The subject has been reviewed extensively
by a number of authors. The most recent comprehensive review summarizes the latest development and applications of the technique [3].
As new applications of CE continue to appear,
the advantages and importance of CE in conjunction with mass spectrometry (MS) also become better appreciated. Analytical chemists are
faced with the challenge of increasing sample
complexity and decreasing sample quantities.
Because of the complexity observed with most
biological mixtures, there continues to be a need
for the development of a highly efficient separation technique in conjunction with a sensitive
and specific detector. The low quantities of
analytes often available require nanoseparation
techniques. The mass spectrometer is a selective
and broadly applicable detector for analytical
separations. It can provide information regarding
the structure of unknown components present in
a sample mixture with high specificity and sensitivity. The coupling of CE with MS combines
the extremely high resolving power and structural information in one system. Like any other
coupled separation technique such as GC-MS
and LC-MS, the principal advantage of CE-MS
is that analytes are identified both by their
differential sep~ration and their molecular masses and/or fragnentation patterns. An analytical
separation that precedes MS analysis is often
necessary to assure correct interpretation of the
mass spectral data.
Since the first demonstration of the coupled
technique [4], CE-MS has undergone significant
development both in instrumentation and applications. A schematic diagram of CE-MS instrumental arrangement is shown in Fig. I.
Several reviews on the coupling of CE and MS
have appeared [5-7]. A large number of new
publications on the subject have emerged since
the last review. While this review is intended to
be comprehensive, the papers cited here will be
on the basis of publication date with special
emphasis on those appeared during the last two
years.
2. CE-MS interface overview
From an MS perspective, the combination of
CE and MS has relied on interfaces to allow
670
J. Cai, J. Henion / J. Chromatogr. A 703 (1995) 667-692
efficient transfer of analytes on-line from the
electrophoretic capillary to the mass spectrometer without sacrificing separation efficiency. CE
by its nature is particularly well suited to the
separation of polar compounds readily ionizable
in solution. These types of molecules have posed
a challenge to conventional ionization techniques
such as electron and chemical ionization due to
thermal decomposition of non-volatile or thermally labile compounds. Over the last decade,
several new ionization methods have been developed including fast atom bombardment
(FAB), thermospray (TSP), atmospheric pressure ionization (API) methods, plasma resorption (PD) and matrix-assisted laser desorption
ionization (MALDI). All of them except TSP
are capable of producing ions from the condensed phase without high temperatures. There
are several interfaces based on thi: API design,
including electrospray (ESI), pneumatically assisted electrospray or ion spray (ISP), and the
heated pneumatic nebulizer (HPN).
CE-MS requires the direct coupling of the
ionization method to the liquid-phase separation
techniques to allow MS detection. During the
past few years, improvements have been made in
interface reliability and reproducibility. The ionization techniques which have been successfully
used for CE-MS include continuous-flow FAB
(CF-FAB), ESI and pneumatically assisted
electrospray or ISE Desorption methods such as
MALDI and PD have also been investigated
although no direct on-line CE-MS results with
these ionization methods have been reported
to-date.
2.1. Electrospray
In ESI, ions may exist in solution as protonated molecules or adducts such as sodium and
ammonium adducts in the positive-ion mode or
deprotonated molecules in the negative-ion
mode. According to Kebade and Tang [8] there
are four major processes involved in ESI-MS.
Charged droplets are formed at the ESI capillary
tip by an ion separation mechanism. These
charged droplets are rapidly reduced in size by
solvent evaporation and repeated droplet dis-
integration resulting in extremely small and
highly charged droplets capable of forming gasphase ions. Two different mechanisms have been
proposed to account for the gas-phase ion production. Iribarne and Thomson [9] describe this
process as ion evaporation where gas-phase ions
are "evaporated" from the highly charged droplets. Schmeizeisen-Redeker et al. [10] suggested
a desolvation mechanism in which single ions are
released from very small droplets by solvent
evaporation. Finally, these gas-phase ions undergo secondary processes by which they are modified in the atmospheric and sampling regions of
the mass spectrometer.
ESI was first demonstrated by Zeleny in 1917
[11]. However, the recent breakthrough reported
by Fenn and co-workers [12,13] reported the
observation of multiply charged species with an
ESI interface. This finding has revolutionized the
applicability of conventional mass analyzers of
limited ma.,'--to-charge range to molecular mass
determination and amino acid 3equencing in
protein biochemical investigations [14]. ESI has
been successfully used for the coupling of separation techniques such as LC and CE. So far,
two types of ESI interfaces known as sheath-flow
and sheathless interfaces have gained general
acceptance for CE-ESI-MS.
2.1.1. Sheath-flow interface
The first successful coupling of CE with MS
was reported by the research group of Olivares
et al. [4]. In this initial work, the cathode end of
the CE capillary terminated within a stainlesssteel capillary sheath where electrical contact
was made, thus completing the CE circuit and
initiating the electrospray. While no sheath liquid
was used in this first report, there was a relatively large dead volume at the capillary terminus. An impr3ved version of electrospray CEMS interface was developed by the same research group where the metal contact at the CE
terminus was replaced with a thin sheath of
liquid flow [15]. A schematic diagram of ESI
sheath-flow interface is shown in Fig. 2A. Since
the CE capillary extended to the tip of the ESI
interface and no additional mixing volumes and
metal surface were involved, the sheath-flow
J. Cai, I. Henion / I. Chromatogr. A 703 (1995) 667-692
(A)
cl ~
(e)
(c)
w~,~tmm~dUp
•
mvonqe
Fig. 2. Schematic diagram of CE-ESI-MS interfaces. (A)
Coaxial sheath-flow configuration, (B) sheathless interface
with tapered and conductive coated CE capillary, and (C)
sheathless interface using a gold wire electrode.
interface provided both good sensitivity and
separation efficiency. This system has been
studied for better understanding and utilization
[16-18]. A significant number of applications of
CE-MS using electrospray ionization with a
sheath-flow interface have appeared over the
past three years, including applications in biological samples [19-21], drug metabolism [2224] and impurity determination [25,26]. So far,
the sheath-flow configuration has been the most
widely accepted interface for CE-ESI-MS applications.
Perkins et al. [27] reported a comparison study
between CE and nanoscale capillary liquid chromatography (ncLC) in conjunction with MS via a
sheath-flow ESI interface. CE-ESI-MS was
found to be more rapid and to provide better
671
absolute sensitivity, while ncLC-ESI-MS provided promise for the analysis of dilute samples.
In a similar study [28], CE-ESI-MS showed
improved peak shapes while ncLC-ESI-MS gave
better chromatographic resolution and was less
susceptible to sample overloading. Both studies
concluded that the two techniques are complementary.
One of the disadvantages associated with the
use of sheath liquid is the unavoidable addition
of ionic and neutral species in the sheath-flow
which compete for available charge in the ESI
process, thus lowering the maximum sensitivity
obtainable [29]. !n addition, difficulties have
been experienced with these types of sources for
obtaining a stable electrospray for extended
periods of time, especially with a water sheath.
In addition, studies by Karger and co-workers
[30,31] have shown that the migration times of
analytes in a CE-ESI-MS separation can be
affected by sheath liquid composition. Due to
the potential gradient across the CE capillary,
anions from both the sheath-flow and background electrolyte migrate towards the anode.
When the anions differ in the two solutions, a
moving ionic boundary is formed inside the CE
capillary. Since the pH may be different within
this moving boundary, the electroosmotic flowrates, the effective charge on the analytes and
their migration rates will change once they enter
the boundary. The sheath-flow composition was
found to have changed the migration order of
the analytes, as is shown in Fig. 3 [31].
2.1.2. Sheathless interface
A sbeathless configuration for a particular ESI
interface was developed for the characterization
of proteins in aqueous solutions [32]. This early
version of sheathless source was constructed
from a stainless-steel syringe needle electropolished to a tapered tip to provide a stable
electrospray for the characterization of proteins
in aqueous solution without pneumatic nebulization. Since the onset potential required to obtain
a stable electrospray decreases with a reduction
in the O.D. of the sprayer needle, a sharp
capillary tip allows the spray of aqueous solutions before the onset of a corona discharge. The
J. Cai, J. Henion I J. Chromatogr. A 703 (1995) 667-692
672
(A)
2
6
8
i
b
to
Time (miQ)
100.
(e)
/7
5
\
1
SO.
$
4
6
- 8ido
16:40
25:00
)):20
Time ( r a i n )
41"40 . . . . . . .
2
(c)
6
L
.__£--.--_
Time (rain)
Fig. 3. Effect of liquid sheath on the separation of proteins
by CE-ESI-MS. (A) CE-UV (214 am) with no sheath flow;
(B) CE-ESI-M$ with 1% acetic acid in methanol-water
(50"50) as sheath flow; (C) CE-UV (214 nm) with 1% acetic
acid in methanol-water (50:50) as sheath flow. Peaks: 1 =
lysozyme; 2 -- cytochrome c; 3 -- aprotinin; 4 = ribonuclease
A; 5 = myoglobin, 6 = a-chymotrypsinogen A; 7 --/3-1actoglobulin B; 8--O-lactoglobulin A; 9=carbonic anhydrase.
From Ref. [31].
concept was adopted by Gale and Smith [29]
who used a small-I.D, fused-silica capillary with
an etched tip instead of the stainless steel needle
for a pressure-infusion ESI source. A coaxial
sheath gas of sulfur hexafluoride (SFG) was
applied to suppress the corona discharge at the
sprayer tip.
A similar sheathless interface was adapted for
CE-ESI-MS applications [33]. The sheathless
design used a gold conductive coating at the CE
capillary terminus to establish electrical contact
with the CE effluent. Such a sheathless interface
is illustrated in Fig. 2B. A publication by the
same group [34] demonstrated the separation of
a tryptic digest using the sheathless interface and
a 10/zm I.D. CE capillary. The electrical contact
was made by applying a silver conductive coating
to the capillary terminus. The interface was
found to produce and maintain a stable electrospray signal during a CE-ESI-MS experiment.
The silver coating at the CE capillary terminus
caused the formation of additional peaks which
were believed to be the doubly charged adduct
ion, (M + H + Ag) z+, a phenomenon that may
make molecular mass determination more problematic. Another sheathless interface, the
"pinhole design" was also briefly investigated
[33]. A microhole was created in the CE capillary wall by short circuit using high voltage. The
microhole was covered with gold coatings. Preliminary results showed good performance with
this interface design.
A sheathless ESI interface was also developed
and coupled to a time-of-flight mass spectrometer by Fang et al. [35]. The electrical contact was
made by inserting a gold electrode into the outlet
of the CE capillary. The electrode was connected
to a needle assembly where voltage was applied
for ESI which completed the CE circuit (see Fig.
2C). By removing the polyimide coating and
sharpening the silica capillary tip, they were able
to increase the sensitivity by a factor of 4 to 5. A
stable electrospray current was observed using
90% water in the CE electrolyte, which makes it
possible for the direct transfer of optimum CE
separation condition to CE-ESI-MS. Recently,
Kriger et al. [36] incorporated a sheathless
interface into a CE-ESI-MS system on an ion
J. Cai, J. Henion / J. Chromatogr. A 703 0995) 667-692,
trap mass spectrometer. A 50 /~m I.D. fusedsilica capillary was tapered to 45° angle at the
exit tip and then gold-plated. The gold coating
was prepared with a proprietary process and
demonstrated excellent physical and chemical
stability. A good electrospray was obtained with
a wide range of CE electrolytes, including 100%
aqueous solutions and solutions of high ionic
strength. Using the sheathless interface they
were able to demonstrate a detection limit for
leucine-enkephalin in the total ion electropherograms on the order of 20 fmol and 10 fmol in the
extracted ion current profile, indicating the improvement in sensitivity without the use of
sheath-flow.
2.1.3. Comparison of sheath-flow and sheathless
configurations
Gale and Smith [29] compared the performance of the two ESI interfaces using pressure
infusion. Their study showed several advantages
of the sheathless version, including great sensitivity, low required flow-rates and long-term
stability. In another study [33l, the performance
of a gold-coated sheathless interface was compared with that of a sheath-flow interface regarding their dependence on buffer system and
concentration, as well as capillary I.D. The
sheathless interface turned out to offer better
analyte detectability. It was found that the
sheathless configuration also eliminated the possible interferences from the sheath solvents such
as charge state distribution shift [36].
2.2. Ion spray
ISP is closely related to ESI, the difference
being the application of a nebulizing gas which
permits stable electrospray operation at flowrates up to 1 ml/min [37] whereas pure electrospray has been restricted to flow-rate below 10
/~I/min [13], Ikonomou et al. [38] have published
a comprehensive study on the comparison of the
mechanisms and performance of ESI and ISP.
Since the nebulization of the sol~ent for ESI is
not dependent on the electrohydrodynamic instability at the capillary tip, iSP is able to
673
produce charged droplets and mass spectra even
when the capillary voltage is below that required
for the onset of electrospray [38]. In addition,
ISP provides superior sensitivity and signal
stability at higher flow-rates as compared with
pure ESI. ISP provides a mild ionization via an
ion evaporation process which results in primarily molecular mass information in the form of
singly and multiply charged ions. In fact, the
mass spectra of basic compounds are essentially
the same with either ESI or ISP [38]. To-date,
two types of CE-MS coupling have been described with ISP interfaces: the liquid junction
and coaxial sheath-flow configurations.
2.2.1. Liquid ]traction interface
A liquid junction design was developed by
Henion and co-workers [39,40] for the coupling
of CE to MS using the pneumatically assisted
electrospray approach. The liquid junction was
constructed from a stainless-steel tee. The
cathode end of the CE capillary and the end of
the ISP needle were positioned in the center of
the tee opposite each other with a gap of 10 to 25
/Lm which could be adjusted by tightening or
loosening the capillary fittings. The top opening
in the tee was fitted with a make-up liquid
reservoir while the bottom opening was used for
alignment and adjustment of the gap between
the CE capillary and the ISP needle. A
schematic diagram of the liquid junction interface is shown in Fig. 4A. The interface was
found to compensate for the different flow-rates
required by the ISP and the CE.
The utility of the system was demonstrated by
applications including the determination of a
synthetic mixtures of pesticide [39], peptide
standards and tryptic digests [40,41], sulfonylurea herbicides from soil [42], sulfonated
azo dyes in waste water extracts [43], synthetic
drug mixtures and drug residues from human
urine [44]. MS detection was achieved in either
the positive- or negative-ion modes. Separation
efficiencies as high as 300 000 theoretical plates
were obtained indicating negligible peak
broadening due the interface [39]. In addition,
the liquid junction interface was successfully
used for the coupling of gel-filled CE to MS [45].
J. Cai, I. Henion I J. Chromatogr. A 703 (1995) 667-692
674
(N
nebull~nggas
h~et
them are the separation and characterization of
peptides and proteins [47,51-53], drug-protein
conjugates [54], drug metabolites [55], natural
products [49], inorganic materials [56,57], textile
dyes [58] and degradation products of chemical
warfare agents [59]. CE-ISP-MS has also been
applied to the areas of environmental importance such as the analysis of paralytic shellfish
poisons in sea food [46,60], and the determination of agrochemicals and industrial diisocyanates [61].
2.2.3, Comparison of liquid junction and
sheath-flow configurations
(e)
6heidJs-How ne~dLdngtlN
leAel
Inlet
9~ilnhlms,aeLql~ ~ l n ~ m ~
~, PEEKteo
Ito
Fig, 4. Schematic diagram of CE-ISP-MS interfaces, (A)
Liquid junction and (B) coaxial sheath-flow configurations,
2.2.2. Sheath-flow interface
The ISP sheath-flow interface is similar to that
described for the ESI system with the difference
being the addition of a coaxial nebulizing gas as
is shown in Fig. 4B. It has been described by
Thibault and co-workers [46,47]. Later, Pleasence et ai. [48] modified a commercial ISP
interface so that both the liquid junction and
coaxial approaches could be used. Henion et al.
[49] have modified a coaxial sheath-flow ISP
interface for the coupling of CE to a benchtop
ion trap mass spectrometer. The interface
showed excellent stability and sensitivity. Using
finger-tight fittings, the interface allows easy
exchange of the CE capillary when necessary.
More recently, Tetler et al. [50] have studied the
influence of the dimensions of the capillaries in a
coaxial CE-MS interface on the stability and
sensitivity of the systems.
ISP using the coaxial sheath-flow configuration
has been used in numerous applications. Among
Pleasance et al. [48] reported a comparison
study between the liquid junction and coaxial
interfaces with regard to ruggedness, ease of use,
sensitivity, and electrophoi'etic performance. A
modified commercial ISP interface was constructed which allowed the use of both liquid
junction and sheath-flow configurations on the
same source. They found that both interface
designs were capable of providing efficient coupling of CE to MS with the coaxial sheath-flow
configuration being more robust and reproducible. An additional advantage with the sheathflow interface is that it provides the potential for
flow injection analysis since the make-up flow is
delivered independently of the CE effluent.
While the liquid junction CE-MS coupling approach provided improved sensitivity and ion
current stability when properly assembled, the
optimum setup of this system was more challenging than that of the sheath-flow system.
2.3. Continuous.flow fast atom bombardment
FAB was first introduced by Barber et al. [62].
This technique dramatically extended the capability of MS in the early 1980s for the determination of fragile and polar compounds. FAB was
considered a static analytical technique until Ito
et al. [63] reported continuous-flow (CF) FAB.
The combination of CE with CF-FAB-MS was
first demonstrated by Minard et al. [64]. Since
then there has been a great deal of effort to
improve its performance. The low flow-rates
required to maintain the high efficiencies of CE
675
J. Cai, J. Henion I J. Chromatogr. A 703 0995) 667-692
are incompatible with the typical CF-FAB flov:rates, which require a typical flow-rate of 5/zl/
rain. To circumvent this problem, interlaces
using sheath-flow or liquid juncticn configurations have been introduced to couple CE to
CF-FAB.
(A)
,/FIw T~ N~tx,,
!'a8 ~
2.3.!. Liquid junction interface
The first CE-CF-FAB-MS work was performed using a liquid junction inter~ace [64]. Its
construction involved alignment of the CE capillary and the FAB flow capillary. A 20-/zm wide
junction between the two capillaries was carefully adjusted and immersed in the FAB matrix
containing methanol or methanol-acetonitrile
with 4 to 20% glycerol. This interface provided
relatively low separation efficiency. Only 6800
theoretical plates were obtained for the separation of a synthetic peptide mixture, qeinhoud
et al. [65] have studied the performance of a
similar liquid junction interface in terms of peak
broadening and found that although the liquid
junction resulted in some loss of resolution in
separation, plate numbers over 10000 were
achievable. They demonstrated the concentration detection limit of 1 ng/ml of dextromethorphan. The liquid junction interface has been
redesigned to allow easy mounting and alignment of the CE and flow FAB capillaries, as well
as the inter-capillary gap adjustment [66]. The
interface was further modified by Caprioli et al.
[67], who reported CE-CF-FAB-MS and liquid
junction interface for the analysis of synthetic
mixtures of peptides and protein digests. A
liquid junction-like interface was constructed and
was used for the identification of deoxynucleoside-polyaromatic hydrocarbon adducts [68].
While similar to the early version in certain
respects, it contained several unique features
including the assembly which consisted of two
Nalgene tees joined by a glass capillary flamedrawn to a suitable I.D. to provide a snug
alignment for the CE and FAB capillaries. The
liquid junction interface for CE-CF-FAB-MS is
very similar to those for CE-ISP-MS with the
exception of the absence of nebulizing gas as is
shown in Fig. 5A.
(e)
FAll ~
_
_ ~
~;=. ~ ¢ ~ r , g .
h ....
,,.
FtmaS.u~",,
t a e m e ~ a w y ",,,
,.
~
~/"
,,"
Fig. 5. Schematic diagram of CE-CF-FAB-MS interfaces.
(A) Liquid junction and (B) coaxial sheath-flow configurations.
2.3.2. Sheath-flow interface
The coaxial sheath-flow interface is another
approach to couple CE and CF-FAB-MS, which
was first designed for the coupling of open
tubular liquid chromatography to CF-FAB [69].
The interf?ce was adapted for CE-CF-FAB-MS
analysis [70,71]. The coaxial configuration allows
independent optimization of the composition and
flow-rates of CE effluents and the FAB matrix,
minimal band broadening and precludes any
deleterious effects of the polar, viscous FAB
matrix on the CE separation process. The
sheath-flow interface for CE-CF-FAB-MS is
almost identical to those for CE-ESI-MS with
the exception of sheath flow being a FAB matrix
solution. A schematic representation of sheath
flow CE-CF-FAB-MS interface is shown in Fig.
5B. These initial demonstrations of on-line CECF-FAB-MS using coaxial sheath-flow interface
676
J, Cai, J. Henion / J. Chromatogr. A 703 (1995) 667-692
showed the separation of peptide mixtures with
separation efficiencies of several hundred
thousand theoretical ~!ates with limits of detection of less than 10 fmol [71]. In order to
minimize the vacuum induced flow, a small-I.D.
(13 ~m) CE capillary was used with the end
being withdrawn several mm back into the
sheath capillary (see Fig. 5B). The sheath-flow
interface was used for the characterization of
peptide and protein mixtures, as well as protein
digests [72,73].
Comparisons between CE and ncLC in conjunction with CF-FAB-MS have been reported
[74,75]. These findings suggest that the analysis
time is significantly shorter for CE than for cLC.
The detection limit by CE in terms of absolute
amount injected is several orders of magnitude
higher than for cLC. The two separations also
gave different elution order of the same analytes.
The main disadvantage of CE is that overloading
occurs at much lower levels than for cLC.
2.3.3. Comparison of liquid ]unction and
sheath-flow configurations
The two interface configurations for CE-CFFAB-MS have been compared in terms of performance and ruggedness [76]. The advantages
of a coaxial interface include no dead volume
and higher theoretical plate numbers. However,
the coaxial interface is difficult to handle and set
up due to the greater manipulation of relatively
brittle capillaries and the problems associated
with high voltage arcing through the thin capillary wall. In addition, ion source vacuum-induced parabolic flow present in the CE capillary
resulted in shorter electrophoresis times and loss
of CE performance. Only small-I.D, capillaries
have been used to minimize the effect associated
with the mass spectrometer vacuum. The vacuum-induced flow-rates in a coaxial interface
were found to be 4/~l/min for a 1 m x 75/~m
I.D. CE capillary and 3 ni/min for an equal
length of capillary with 13 ~m I.D. [70]. The
liquid junction interface was found to offer equal
or greater overall separation efficiencies. It was
relatively easy to set up and operate, and allows
larger-I.D, capillaries to be used to improve the
sample load. The disadvantages of liquid june-
tion interface are the large dead volume and
lower overall sensitivity. Both interfaces were
found to have degraded a significant degree of
CE performance.
2.4. Comaarison of electrospray and ion spray
with continuous-flowfast atom bombardment
Several research groups have compared the
two ionization methods with regard to the performance of CE-MS [74,77]. There are several
attractive features of CE-MS using the ESI
interface. Since larger-I.D, capillaries can be
used for the ESI interface, the loading capacity
should be greater than those using the coaxial
sheath-flow CF-FAB interface where only smallI.D. capillaries (ca. 10 ~m) can be used due to
the pressure drop at the source region which lead
to hydrodynamic flow within the CE capillary
distorting the plug flow of electroosmotic flow
and causing band broadening. The ESI interface
also gives reduced background noise and higher
sensitivity [78]. In the study conducted by Nichols et ai. [77], a micro liquid junction configuration was used for both CF-FAB and ISP interfaces for the investigation of the same analytes
which included aromatic sulfonates, quaternary
amines and a synthetic peptide mixture. The two
analytical approaches were found to provide
comparable sensitivity and analytical information. However, CF-FAB provided greater peak
broadening (see Fig. 6). Both systems were
considered as less than routine with respect to
analytical ruggedness.
2.5. Other approaches
2.5.1. Other interfaces
A microflow ultrasonic nebulizer interface was
developed by Tarr et al. [79]. The nebulizer was
able to provide stable nebulization and essentially 100% transport efficiency at flow-rates of 5 to
20/zl/min suggesting good potential for application in interfacing CE and microcolumn LC to
MS. However, no actual CE-MS experiments
have been conducted with thio interface.
J. Cai, J. Henion # J. Chromatogr. A 703 (1995) 667-692
677
(A)
IO0
3
I
~_ 50,
Time
:5
,
"
-
,
,,m,,,,
1~
o
8
ls
Time ( , . i n )
(B)
I0.02 AU
;"
6
i,X
'O
i,
Time (rni.)
2
"''---i
5
-
-------i:-':--~::--ir-~---~--"~---~
I0
15
-'---
20
Fig. 6. Total selected ion current and CE-UV (inset) electropherograms of a sulfonated aromatic mixture obtained by (A)
CE-ISP-SIM-MS and (B) CE-CF-FAB-SIM-MS, using liquid junction interfaces. From Ref. [77].
2.5.2. Off-line CE-MS
While on-line methods are preferred, they are
often not available to interested practitioners.
Ionization by desorption methods such as PD
and MALDI are very useful for the characterization of large molecules. Molecular masses up to
several hundred kilodaltons can be determined
by MALDI-MS by irradiation of small sample
678
J. Cai, J. Henion I J. Chromatogr. A 703 (1995) 667-692
spots [80]. However, on-line CE coupling of
desorption MS is not yet been reported.
Takigiku et al. [81] described the use of a
"porous glass joint" near the cathodic end of the
CE column to collect fractions for desorption
MS. Off-line coupling of CE with laser desorption MS was used by Keough et al. [82] for the
determination of proteins. Weinmann and coworkers [83,84] have developed a method using
nitrocellulose.coated targets for the isolation and
transfer of polypeptides and proteins from CE
aration for subsequent characterization by
f PD-MS. Several modes of off-line coupling
of CE-MALDI-MS have been investigated by
Van Veelen et al. [85]. The best performance was
obtained using a moving belt-like system in a
stepwise manner and a sheath-flow of matrix
solution followed by scanning of laser desorption
target. Tomer et al. [86] have recently described
methodologies for combining CE and affinity
chromatography off-line with MALDI-TOF-MS.
Using a coaxial sheath-flow to avoid memory
effects, CE fractions were collected and then
analyzed by MALDI-TOF-MS. Another approach was described by Blakley et al. [87],
where a computer-controlled x-y table and an
effluent transfer tube were employed for fraction
collection and MALDI-TOF-MS analysis. A
method for fraction collection from an automated micellar electrokinetic capillary chromatographic system has been described [88]. Aliquots
of 5 ~1 of fractions were collected with recoveries of more than 75%. How.mer, the ion
suppression effect of sodium dodecyl sulfate
present in the buffer prohibited the direct analysis by FAB-MS.
3. Applications
3.1. Biologically important components
One of the advantages of incorporating ESIMS and ISP-MS with the determination of biological compounds with high molecular masses
lies in the multiple charging of the analytes that
can occur under ESI conditions. The majority of
applications of CE-MS have been in the field of
biological and biochemical studies.
3.1.1. Peptides
A number of reports have appeared on the
characterization of synthetic mixtures of peptides
and proteins by CE-MS [17,41,89]. A CE-ESIMS system was optimized for the analysis of
peptide mixtures [20], Full-scan mass spectra
were acquired using 160 fmol of peptide standards loaded on-column. A synthetic mixture of
six dipeptides was examined by CE-ISP-MS
[51]. Standard peptide mixtures have also been
characterized by CE-ESI-MS on an ion trap [90]
and TOF [35] mass spectrometers. Rosnack et
al. [91] discussed the capability of CE-ESI-MS
in the analysis of synthetic peptide impurities, a
practical problem encountered in a peptide synthesis laboratory. Recently, a synthetic 37-residue peptide fragment of a monoelonal antibody
against Herpes virus was characterized by CEISP-MS and by off-line 252Cf PD-TOF-MS [53].
Five impurities formed in the synthesis were
separated and identified by CE-ISP-MS. Weinmann et al. [83] carded out structural characterization of polypeptides and proteins using off-line
coupling of CE and z52CfPD-MS analysis. Tomlinson and co-workers [92,93] described on-line
sample enrichment and/or desalting of peptides
for subsequent determination by CE-ESI-MS
using a precolumn packed with HPLC packing.
3.1.2. Proteins
The determination of recombinant bovine and
porcine somatotropins (rbSt and rpSt) was reported by Tsuji et al. [19]. A detection limit of
100 fmol was obtained. The separation and
determination of a synthetic protein mixture was
presented by Cole et al. [94] using a hydrophilic
derivatized CE capillary. Thompson and coworkers [30,31] studied the liquid sheath effects
on the CE-ESI-MS separations of standard
proteins in a coated capillary. The same authors
have also investigated on-column transient isotachophoretic sample preconcentration to improve protein detection limits by CE-ESI-MS
[95]. Wahl et al. [17] used the small-diametercapillary approach to increase peptide and protein detection sensitivity in CE-ESI-MS. Volk et
al. [52] demonstrated the rapid and efficient
separation and characterization of closely related
glycoforms of glycoproteins and structurally simi-
J. Cai, J. Henion / J. Chromatogr.A 703 0995) 667-692
lar proteolytic fragments using CE-ISP-MS. A
synthetic drug-protein conjugate mixture was
analyzed by CE-ISP-MS [54]. Lysozyme and its
conjugates with one to three naproxen molecules
were separated and identified. CE-ESI-MS has
been used for the analysis of a complex mixture
of peptides and small proteins, snake venoms of
Black Mamba [21]. Molecular mass and some
sequence information for those toxins were obtained using CE-selected ion monitoring (SIM)
MS and full-scan CE-MS analysis. Fig. 7 shows
the CE-ESI-SIM-MS electropherograms of
some peptides observed in the whole venom
from Black Mamba [21].
3.1.3. Enzymatic digests
Tryptic digests have been characterized by a
numbers of workers as model systems [41,74,96].
Separation of tryptic digests of bovine Candida
krusei and equine cytochrome c has been reported using CE-ESI-MS and an aminopropylamine-treated capillary [34]. By using a 10
tzm I.D. capillary, sample amounts in the 30
fmoi region were loaded into the capillary.
Tryptic digests of bovine, equine and tuna cytochrome c were characterized by CE-ESI-MS
using a 20/~m I.D. capillary [33]. The total ion
electropherograms obtained with these enzymatic diges:s are shown in Fig. 8. The
chymotryptic digest of ubiquitin was studied by
CE-ESI-MS using Fourier transform ion cyclotron resonance (FTICR) MS [97]. Takigiku et al.
[98] presented characterization of protein digests
by CE-ISP-MS. The hydrolysis of protein was
carried out in a fused-silica capillary with immobilized trypsin (capillary microreactor). This
technique may be advantageous for sequencing
minute quantities of protein with minimal sample
handling.
679
ESI-MS. Characterization of deoxynucleosidepolyaromatic hydrocarbon adducts by CE-CTFAB-MS was demonstrated by Wolf et al. [68].
The adducts were formed by covalent attachment of four polycyclic aromatic hydrocarbon
(PAH) and amino-PAH compounds to deoxyguanosine. Analysis of DNA-carcinogen a~<lucts
by CE-ESI-MS was investigated [102]. A reactive metabolite of benzo[a]pyrene, benzo[a]pyrene-7,8-dihydrodiol 9,10-epoxide metabolite,
was reacted with DNA in vitro to form adducts
which were subsequently cleaved into individual
nucleotides and analyzed by CE-ESI-MS.
3.1.5. Non-covalent binding study
CE-MS has been found to be a useful technique for the study of biomolecular non-covalent
interactions because it combines separation, and
characterization as well as quantitation in one
operation. Recently, we have investigated the
non-covalent binding of FKS06 and rapamycin to
FKBP by CE-ISP-MS [103]. The observed relative binding affinities of rapamycin and FK506 to
FKBP is 9/1 which is consistent with the published K+ values of the two ligands. Goodlett et
at. [104] studied the thermal stability of ribonuclease S, an enzymaticaUy active non-covalent
complex, by CE-ESI-MS. The intensity of
RNase S molecular ion peaks were observed to
decrease with the increasing temperatures at
heated capillary-skimmer interface and the in
the sample solution. Uzabiaga et al. [105] have
recently presented a study on the non-covalent
complex of human antithrombin III with the
pentasaccharide SR 90107A by CE-ESI-MS.
CE-ESI-MS was found to be superior to direct
infusion.
3.2. Pharmaceutical and drug metabolism
3.1.4. Other biopolymers
Separation and characterization of lipopolysaccharides from B. catarrhalis and H. influenzae
type B has been demonstrated using CE-ISP-MS
[99]. CE-ESI-MS was used for the analysis of
the oligosaccharides isolated from glycoproteins
[100]. Zhao et al, [101] presented the determination of a nucleotide and its radiation-damaged
products by capillary isotachophoresis (clTP)-
3.2.1. Drugs
CE-MS is finding increasing applications in
the analysis of sample for pharmaceuticals, therapeutic and xenobiotic metabolites. CE-ISP-MS
was used for the determination of small drug
molecules such as sulfonamides and benzodiazepines using the liquid junction interface [44]. The
analysis of an extract of human urine of a person
680
J. Cai, Y. Henion I ]. Chrommogr. A 703 (1995) 667-692
m/z=1055.5
'7
.
l
i
i
.
-.
--
""
-
.
:
,p
m/z=I s13
U
lO0
m/z,,I
"__
•
5
100t . . . . .
I
m/z-9?2.5
'- I
'
10(:)
.
.
.
.
.
.
.
.
.
nan.
-
.
I
-
--
.--
36
_
. . . . . .
- -
_
40
,.I
"
.
I
-
I
: - ~
ram,
----
6
_
iiii
1001l
8
-:: - ....
_
J
_ j
4.9
41
r~z,.llsl
UU0
.
11
,.~
.
.
.
.
.
.
.
!
~ - = . . :
-:
---
=-.-
.
. ~
nW.,,1006
13
iOo
--
.
.
_
• .
.
.
.
.
100
loo]
.
.
.
.
.
' ' -
.
I L ~
:
~
. . . .
:
ram,
,
.
~
.
~
. . . . . . . .
~
....
.
.
I
- - ] :
"llrilj
25
lOO
28
.
i
..i
. . . .
i
.
.
.
.
47
.
.4
iii
48
.
.
.
.
50
6~)
271,29
rnlz.i 145
100t
31
~132
m/z-lO01.5
12~X)
TIME(MEN)
g1Vb,1034
~t
.
]
6:00
.
.It
~oo
1001
-
~'
_
22
IO0
,
45
1 " 14
1819
.
: -..
42
12:00
TIME(MIN)
18:00
18:00
Fig. 7. CE-ESI-SIM-MSdeterminationof I~ptides in the whole venom from Dendroaspis polylep~s polylepis. From Ref. [21].
.I. Col, Z Henion / I. Chromatogr. A 703 (1995) 667-692
gated in our research group [112]. Several synthetic chiral drug mixtures were studied vsing
cyclodextfin as a chiral selector in the electrolyte. Both the protonated molecules of the
enantiomers of terbutaline and their respective
cyclodextrin-drug complex were shown in the
CE-ISP-SIM-MS electropherograms (Fig. 9).
10o Bovine
~
681
o
.~ '°°I Equ'ne
3.2.2. Drug metab~,,; "~
(9
rr
100
50
0.0
!
2.0
4.0
6,0
ti
8,0
10,0
Time (mir,)
CE-MS is gaining popularity in drug metabolism studies. Tomlinson and co-warkers [2224,113,114] recently published their applications
of CE-ESI-MS in drug metabolism with a particula~" emphasis on neuroleptic drugs. The use of
a non-aqueous solution as electrolyte was emphasized due to the low solubility of the drugs in
pure aqueous solutions. Methanol (100%) containing acetic acid and/or ammonium acetate
was used as CE electrolyte [22,23,114]. Shown in
Fig. 10 are the CE-ESI-SIM-MS electrophero-
Fig, 8, Total ion electropherograms t'tom ttyptic digests of
bovine, equine and ~una cytochrom~ c by CE-ESI-MS using
a sheathless interface. From Ref. [33].
(A)
nP,,z=226
100,
who had received an oral dose of flurazepam
dihydrochloride was demonstrated. The separation of sulfonamidc drugs was also reported
using CE-ESI-MS [27,106]. Parker et al. [28]
reported the analysis of macrolide antibiotics by
nanoscale packed capillary LC and CE-ESI-MS.
An application of CE-ISP-MS in the study of
Chinese herbal medicine remedy in tree barks
and Chinese herbal medicine tablets was published using a modified benchtop ion trap mass
spectrometer [49,107]. Quantitative analysis was
carried out using an internal standard.
Impurity determination is also very impnrtant
in pharmaceutical preparations. A CE-ISP-MS
system was evaluated for the detection and
characterization of trace impurities in sample
mixtures using model peptides and small drug
molecules [108]. The characterization of chiral
drugs has drawn increased attention in the pharmaceutical industry community [109]. Chiral
separation by CE has been reviewed recently
[110,111]. The separation and characterization of
chiral drugs by CE-ISP-MS have been investi-
45 720
IO
s:
0
•
I
•
•
•
•
i
•
•
•
•
!
•
•
•
•
I
•
•
I
"
"
(e)
n~:lSS7
100
m
,no
50
_i
0
•
!
.
20.0
.
.
.
3O.0
4O.0
Tim (rain)
Fig. 9. CE-ISP-MS etectropherograms showing the separation of (A) terbutaline enantiomers and (B) their respective cyclodextfin-drug complexes. From Ref. [i 12].
682
J. Cai, J. Henion I J. Chromatogr. A 703 (1995) 667-692
2.7 m/z:160
,.-.,.
k'9
0
,,....
m
I I
7.6~m/z=188
0
~
:
:
:
:
1°°
150m/1.310
-
~
666
o
<r~
m 84"3]m/z:229 /~
I
Z
o_
2.8 ]iz=245
. . . .
I.~
5.'00
,
. . . .
15:00
25100
TIME(MINUTES)
,
. . . .
,~
. . . .
,
. . . .
,
. . . .
,
Fig. 1O. CE-ESI-MS electropherograms for the analysis of a
metabolite mixture derived form a rat hepatic microsomal
incubation of Mifendine (MIF) using 5 mM ammonium
acetate and 100 mM acetic acid in 100% MeOH as electrolyte. From Ref. [114].
gram of a metabolite mixture derived from a rat
hepatic microsomal incubation of Mifentidine.
We have demenstrated the identification of metabolites of lysergic acid diethylamide (LSD) in
human hepatic microsomal incubation by both
LC-MS-MS and CE-MS-MS using a sheathflow ISP inter~ace [55]. CE-ISP-MS as a complementary separation tool greatly facilitated the
identification of unknown metabolites. A representative electropherogram is shown in Fig. 11.
S 3. Environmental
CE-MS has been applied to the determination
of compounds of environmental concern, such as
agrochemicals, pesticides, inorganic compounds
and dyes. A fast CE-ISP-MS determination of
sulfonylurea crop-protection agents was demon-
0.0 5.0 10.0 15.020.025.030.0
Time (rain)
Fig. 11. CE-ISP-MS electropherograms for the identification
of metabolites derived from human hepatic microsomal
incubation of LSD. Peaks: 1 = LSD; 2 = iso-LSD; 3 = norLSD; 4 = iso-nor-LSD; 5 = lysergic acid ethylamide (LAE);
6 = iso-LAE; 7 = 2-oxo-LgD; 8-11 = unknown metabolites.
From Ref. [55].
strated by Garcia and Henion [42]. Full-scan
CE-ISP-MS collision-induced dissociation mass
spectra were obtained from 35-pmol levels of
standard injected. The analysis of ionic textile
dyes using CE-ISP-MS and MS-MS was presented by Lee et al. [43]. In this early study, the
presence of three sulfonated azo dyes in a waste
water extract was confirmed by CE-ISP-MS-MS
experiments. CE-ISP-MS has been used for the
determination of quaternary ~mmonium herbicides [61]. The limit of detection at 5 ppm and
linear calibration graphs were obtained with
correlation coefficients above 0.995. The applica-
J, Cai, J. Henion I Y. Chromatogr. A 703 (1995)667-692
tion of CE-ISP-MS to the analysis of mixtures
containing a variety of antibiotic classes used in
the fish aquaculture industry has been reported
by Pleasance et al. [48]. The presence of several
antibiotics in shellfish extracts was confirmed at
low ppm levels. Lamoree et al. [115] have
demonstrated the application of CE-ESI-MS for
the determination of/3-agonists which are being
used illegally in the cattle industry as repartitioning agents to increase meat production. By using
on-capillary cITP for loadability enhancement,
concentration detection limits in the ng/ml range
were obtained. The determination of some /3agonists in spiked calf urine was demonstrated.
Hines et al. [116] have evaluated a CE-ESI-MS
system for the detection and quantification of
Fumonsin B~, the most abundant fumonisin in
corn samples.
286460
I Rb
ntt~85
! 288 160
I
ts+
m/z 133
I K+
m/z39
+
110360
_
._=.
.
m/z SS
.
l
.
.
.
.
19940
.
.
.
-
.
.
~
-
1 srs+
I Ca +
_~z.~
. . . . . . .. .. ............. . .+. .+. .. , .~.. ..... .. . . . . .+. .
4. Limitations and solutions
.+
.
16720
4.1. Limitations
I1 060
Although numerous publications have appeared on CE-MS, this technique is still not
widely accepted for routine use. The major
limitation of CE is the limited sample volumes
that can be analyzed without compromising
separation efficiency. Consequently, the concentration detection limit for CE is usually several
orders of magnitude higher than that of chromatographic methods. Using the currently available instrumentation, CE-MS detection fimits
for most applications are too high so that it
seems unlikely for it to be used in routine
analysis. For example, a study was conducted
comparing the performance of CE-MS with that
of microbore LC-MS in the determination of
endogenous amounts of leucine-enkephalin and
........
28 020
I Mn +
.mJz_.5~..,e..... ,-J...... Jl.t+..dL.l . . . . . . l . . . . . . . . . .
J .....
I Cd2+
12940
t Co z+
18 420
,, INib
11640
m/z 114
m/z 59
~zsg
2
4
6
Pleasance et al. [60] reported the application
of CE-ISP-MS to the analysis of paralytic shellfish toxins isolated from marine matrices. The
application of CE-ISP-MS to the determination
of textile dyes has been reported [58,117]. A
CE-ESI-MS system optimization for cationic
and anionic laser dye analysis has been described
using opposite polarities at the injector and
interface [25]. The technique was employed to
evaluate the purity of the laser dyes. Kostiainen
et al. [59] have demonstrated the identification
of degradation products of some chemical warfare agents such as substituted organophosphorus acids by CE-ISP-MS using the negative
ionization mode. Quantitative analysis was performed with good linearity. Another application
of CE-ISP-MS is the detection of inorganic ions
[56,57]. Example electropherograms obtained by
CE-ISP-SIM-MS are shown in Fig. 12 where ten
inorganic cations are shown separated. The analysis of an aqueous acidic extract of used aircraft
engine oil revealed high levels of lead as well as
lower levels of other metal ions such as
chromium and nickel [56].
,.- .........
l Ba +
m/z 138
- ___=__
3.4. Other applications
8
10
12
14
16
18
Time(~)
Fig. 12. CE-ISP-MS electropherograms of ten inorganic
cations. From Ref. [57].
684
I. Cai, J. Henion I I. Chromatogr. A 703 (1995) 667-692
methionine-enkephalin in equine cerebrospinal
fluid using identical sample clean-up and enrichment procedures [118]. CE-MS was found to be
limited in its concentration detection capacity
owing to its much smaller injection volume.
Another drawback with CE-MS is that migration times tend to fluctuate with a change of
temperature in the environment. Although some
manufacturers have incorporated temperaturecontrolling system into their CE instruments,
these devices cannot be effect.ively utilized in
CE-MS applications because a large portion of
the CE capillary is extended between the CE
instrument and the mass spectrometer. For applications such as regulatory work or those
involving unknown components in a mixture, the
use of a suitable internal standard would be
necessary [48,49]. The chemical condition of the
CE capillary inner walls also plays an important
role in CE separation. The reproducibility and
ruggedness of CE-MS are not currently as good
as those of LC-MS.
Like LC-MS, the use of non-volatile buffers
in CE-MS is generally avoided. Compromises
are often made in choosing appropriate operational conditions for CE-MS. Sensitivity limitations as well as ion source plugging problems
created by the use of non-volatile buffers have
impeded the direct transfer of CE separation
conditions to on-line CE-MS. Non-volatile additives such as cyclodextrins are widely used in the
separation of closely related analytes including
optical isomers [110-112]. However, the concentrations of these additives are often limited
by practical restrictions [26,119]. The MS sensitivity may deteriorate as the bulk flow of
surfactant enters the mass spectrometer source
region.
4.2. Analytical methodology improvement
,1.2.1. On-line preconcentration by capi!lary
isotachophoresis
A number of techniques have been introduced
to improve the concentration sensitivity of CE.
On-column sample concentration approaches,
particularly on-line analyte focusing by clTP,
have drawn the most attention [95,100,115,119-
125]. In contrast to CE which lacks good concentration sensitivity necessary for most realworld applications while providing very high
separation efficiency, clTP offers excellent concentration sensitivity but suffers from the resolution required to characterize complex mixtures. In cITP, the isotachophoretically stacked
bands may be several orders of magnitude more
concentrated than was originally loaded into the
capillary. Extremely narrow zones, frequently
less than 1 s wide, overlap one another, a
characteristic found in cITP separations with
trace quantities of compounds which requires the
mass spectrometer to scan at extremely high
speed in order to generate useful data. Although
spacers may be used to improve the separation,
it is generally impractical when a complex mixture of unknown components is involved. The
combination of CE and clTP has the potential to
overcome the disadvantages of both techniques
while retaining their advantages.
Some applications using cITP preconcentration involve two separate capillaries
[120,121,124]. In these designs, preconcentration
by clTP takes place in one capillary. The isotachophoreticaUy focused sample plugs are then
transferred into a second capillary where CE
separation takes place. The coupling of cITP and
CE is accomplished by inserting the CE capillary
into the clTP capillary through a septum. A
schematic illustration of the system is show in
Fig. 13. A 200-fold improvemeat in concentration sensitivity for the determination of anthracyclines was achieved by Van der Greef and
co-workers [120,124]. The authors have derived
an equation for the calculation of the splitting
ratio of clTP zones into the CE-ESI-MS system
[121].
The coupled use of cITP and CE within the
same capillaryapproach seems to be more effective in terms of ease of application and interfacing to the mass spectrometer [95,100,115,
119,122,123,125]. The technique can be applied
to any C E - M S system without modification of
instrumentation. Using on-line transient cITP
preconcentration techniques, the detection limit
of some protein samples was improved by two
orders of magnitude [95,125]. An example full-
J. Cai, J. [-lenion I J. Chromatogr. A 703 (1995) 667-692
1 on4~ UV
•
..7
685
1
Spectrot~mter
Fig. 13. Schematicillustrationof on-lineclTP preconcentrationfor CE-MS usingtwo separate capillaries.
example full-scan CE-ESI-MS electropherogram
of a synthetic protein mixture obtained using
on-line transient clTP preconcentration is shown
in Fig. 14. As much as 750 nl of sample was
injected. Full-scan CE-ESI-MS spectra were
obtained with analytes at concentrations as low
as ca. 10 -7 M, which was 100-fold lower than
CE-ESI-MS without preconcentration. A similar
approach was described by Lamoree and coworkers [115,123]. A sample volume of 870 nl
test mixture containing several /3-agonists was
analyzed without sacrificing separation resolu• I.O~
I.¢1|
J
m
I ,t
0,
!
2
Z_
Fig. 14. TransientclTP-MShdl-seanreconstructedion dectropherogram of syntheticprotein mixture with 750 nl injection. Peak: 1 = cytochrome c; 2 = myoglobin; * = from the
rear boundary of loading buffer. From Ref. [95].
tion. The capability of the clTP-CE-ESI-MS
system was demonstrated for the trace analysis
of a complex matrix such as calf urine extracts.
Recently, several research groups have evaluated
the clTP-CE-ESI-MS approach [100,119,122].
On-line clTP preconcentration has lowered the
concentration detection limits of CE-MS by two
orders of magnitude for the analysis of oligosaccharides [100]. Similar results were obtained
by Mosely [122] who observed a 200-fold improvement in concentration limits by using the
preconcentration approach, with up to 45% of
the CE capillary being initially filled with sample. It was found that the greater the difference
in the electrophoretic mobilities between the
terminating electrolyte and the analyte, the better the cITP focusing becomes. Analytes with
low electrophoretic mobilities were not preconcentrated. In all cases, the on-line cITP preconcentration has reduced the separation efficiency. But they were still r~luch better than
those of conventional LC separations.
4.2.2. Chromatographic on-line preconcentration
Another approach for on-line analyte concentration for CE is the chromatographic method.
This technique involves a precolumn, either
having walls derivatized with Cis and other
functionalities [126,127] or packed with HPLC
packing materials [128,129]. Recently, this concept has been incorporated into CE-ESI-MS
686
J. Cai, 1. Henion I J. Chromatogr. A 703 (1995) 667-692
applications [92,93]. The precolumn was made of
a small bed of HPLC packing in a FIFE cartridge attached to the inlet of the CE capillary. It
serves as an on-line device for sample cleanup
such as desalting and preconcentration for subsequent analysis by CE-ESI-MS.
4.2.3. Other analytical approaches
The use of small-I.D, capillaries was found to
have improved the absolute sensitivity of CEESI-MS [16,17]. This improvement in sensitivity
is due to the high ionization efficiency associated
with the very low bulk flow-rate from the smallI.D. capillary into the mass spectrometer. Since
the ESI ion current is nearly independent of
flow-rate, ESI-MS will operate as a mass-sensitive detection system when the ESI current is
limited by the flow-rate of the charged species in
the solution to the ESI source [34]. This is
illustrated in Fig. 15 where a 25- to 50-fold
increase in absolute sensitivity is obtained by
reducing the I.D. of CE capillary from 100 to 10
b~m.
Frequently, the high resolution observed with
CE separation results in extremely narrow peaks
Re~stive
t.bt~]100I~m
O"
'"1°'1 5Opm
~r.r')
05
"0
~
0""
0,25
• "
C
20pm
0.04
°
o"
~
_
.I
J
A
II
._
*a~a'l 10pm
0.01
Time (min)
Fig. 15. CE-ESI-SIM-MStotal ion current electropherogramsof a syntheticpeptidemixtureusing100,50, 20, 10/~m
I.D. capillaries. From Ref. [17].
and are sometimes not suitable for performing
on-line CE-MS, especially full-scan on-line MS
analysis. By reducing the applied electric field
strengt'l just prior to the elution of the analyte,
the peaks were broadened sufficiently to allow
enough scans be obtained and averaged across
each peak [18,100].
Recently Kirby et al. [130] demonstrated that
by carefully choosing the experimental conditions, eurfactants can be used successfully in
CE-ESI-MS with only minimal loss in analyte
sensitivity. They also found that the use of mixed
surfactant resulted in much less analyte signal
suppression than what is expected by using
sodium dodecyl sulfate alone.
4.3. Instrumental developments
4.3. I. Coated capillaries
Coated capillaries are often used in the analysis of biological samples such as basic peptides
and proteins to minimize band broadening and
peak tailing due to adsorption of analytes onto
the capillary walls [131]. The factors that attribute to protein adsorption include ionic, hydrophobic, and hydrogen-bonding interactions [3].
The recent advances in CE capillary coatings
have been reviewed [132]. Coated capillaries are
currently available from several commercial
sources. Aminopropyl silylated fused-silica capillaries have frequently been used in CE-MS
applications [17,18,20,21,33,34,36,73,74,89,104,
133]. Capillaries coated with polyacrylamide
have been used to minimize protein adsorption
and electroosmotic flow within the CE capillary
for the analysis of proteins by CE-MS
[30,31,95,130]. A different hydrophilic derivatized capillary was prepared for protein analysis
by CE-MS [94]. Kostiainen et al. [54] used a
commercially available deactivated fused-silica
capillary for the determination of a synthetic
drug-protein conjugate mixture by CE-MS.
Thibault et al. [47] described an CE-MS application using a non-covalent coated CE capillary
with an overall positive charge. The use of
coated capillaries to eliminate the electroosmotic
flow m~y also be useful in minimizing the flow of
non-volatile additives into the mass spectrometer
J. Cai, J. Henion I J. ehromatogr. A 703 (1995) 667-692
[130]. This may be of particular importance ir~
applications such as chiral separations by CEMS.
687
lllO~O,
i,i
x
liOtltB ~
4.3.2. Trapped ion mass spectrometers
One solution to improving the sensitivity of
CE-MS is the development of alternative types
of mass spectrometers which offer the potential
for greater sensitivity. Currently the efficiency of
ion transport from the atmospheric pressure ESI
source to the MS detector is low with only
approximately 10 -5 of the ions formed in ESI
being detected [134]. Trapped ion mass spectrometers, include the quadrupole ion trap (IT)
and the FTICR mass spectrometer may be useful
for improving CE-MS sensitivity. IT mass spectrometers provide significantly higher transmission efficiencies than beam-type mass spectrometers [135]. The coupling of CE with IT-MS has
been demonstrated in several research groups
[36,49,90,136,137]. Ramsey et al. [136] reported
chemical background and noise reduction in CEMS on a quadrupole IT mass spectrometer. By
using the combination of broad-band collisional
activation and resonance ejection, they were
able to significantly reduce background noise in
the CE-ESI-MS total ion electropherograms.
The electropherograms obtained before and
after the background reduction are shown in Fig.
16. Recently we have demonstrated the coupling
of CE to a highly modified benchtop IT mass
spectrometer for quantitative determination
some Chinese herbal remedies [49]. A signal-tonoise ratio better than 10 was obtained in the
extracted ion current with injected quantities
between 370 to 510 attomcle (see Fig. 17).
FTICR-MS has the advantages of high sensitivity, ultra-high resolution and extensive capabilities for tandem MS (MS") [138,139]. The
combination of CE and FTICR-MS has been
reported [97,140,141]. An off-line coupling of
CE and MALDI-MS was demonstrated on a
FTICR mass spectrometer [142]. Hofstadler et
al. [140] have modified a FTICR instrument
which allows rapid cycling between the different
pressure regimes m the trapped ion cell for both
optimum trapping efficiency and high-resolution
detection. The CE-ESI-MS experiments on the
l"''i
r'~
....
I ....
'''';i
....
'''71i
....
'7;;;i;"
ii
oil
IIE
x
....
I ....
I ....
I'''~'''1'"
'1'~''1
' ....
t''
Fig 16. Tolal ion electropherograms(TIC) of a synthetic
peptide mixture obtained (A) without active background
reduction measures and (B) with combined broad-band
coHisional activation and resonance ejection. Peaks: B-bradykinin; X~-xenopsin; N=neurotensin. Time iu min.
From Ref. [I36].
II ~II~"uUHIHI'
41MI"--' A A l~l'l~ill~'
44eat--I
.'.i11 •
z.
• ........
.
! .........
I.,
.....
,-!
.........
.
.
[
.
I .....
, • ,'~[
i
4m
LSiM
16:ill
lLi:ll
ltl:M
2i:M
'illili(NN)'
Fig. 17. Full-scanCE-ISP-MSextracted ion currentelectropherograms of a syntheticmixture co,'.tainingisoquinoline
alkaloidnaturalproductsat low levelsobtainedon a modified
benchtop ion trap mass spectrometer. Low "up-front"collision induced dissociation(CID)= 52.4 V; 25/tm I.D. capillary; mass range 150-400; 100%= 27 408. From Ref. [49].
688
Z Cai, J. Henion / J. C,$romatogr. A 703 (1995) 667-692
FTICR instrumentation allows fast spectrum
acquisition speed. The potential of the CE-ESIFHCR-MS system was further evaluated for a
chymotryptic digest of ubiquitin [97]. The on-line
combination of CE with ESI-FTICR-MS has also
been presented by Johnson et al. [141]. A pulsed
collision gas of nitrogen was used to collisionally
trap ions. Trapping efficiency of ca. 1% was
reported. However, due to the limited pumping
speed, high-resolution mass spectra could not be
obtained immediately after ion accumulation.
The ions were stored in the FTICR cell for 10
rain before a reasonably low base pressure could
be reached for high-resolution detection.
4.3.3. Time o f flight
TOF mass spectrometers offer very rapid scan
rates with the ability to record a full mass
spectrum from an injection pulse [14]. This
makes it potentially good for on-line applications
such as CE-MS and LC-MS. in addition, TOF
analyzers can provide improved resolution by
using a reflectron configuration. Also, they are
relatively simple and inexpensive, and are well
suited for pulsed ionization techniques such as
MALDI. The use of TOF MS on-line with CE
has been published by Zare and co-workers
[35,143]. The applicability of the system was
demonstrated using a sheathless ESI interface
for the characterization of synthetic peptide
mixtures. Recently, the same research group has
reported a modified ion optics system which
greatly improved ion transmission efficiencies
[144]. Detection limits at tens of femtomole
injected on-column were obtained for synthetic
peptides. The off-line combination of CE and
MALDI-TOF-MS has also been reported using
fraction collection [86].
4.3.4. Use o f an array detector
The use of a position and time-resolved ion
counting (PATRIC) detector was found to have
dramatically improved the sensitivity of mass
spectral detection in CE-MS [145]. Such PATRIC array detectors are available on Finnigan
MAT Model 900 mass spectrometers. While the
static detection mode was found to be useful in
target compound analysis, scanning array detec-
tion was more universal. By using scanning array
detection, a significant gain in sensitivity (10- to
100-fold) can be obtained over a conventional
secondary electron multiplier detector in combination with scanning. A same instrument
equipped with PATRIC has been used in CEESI-MS applications of drug metabolism studies
[22,24,113,114].
Acknowledgements
We acknowledge Beckman Instruments for the
loan of the P/ACE 2050 CE instrument, and
Eastman Kodak and PE Sciex for partial support
of our CE-MS research.
References
[1] J. Jorgenson and K. Lukacs, Anal. Chem., 53 (1981)
1298-1302.
[2] R. Wallingfordand A. Ewing,Anal. Chem., 60 (1988)
1972-1975.
[3] C.A. Manning and R.D. Kennedy, Anal. Chem., 66
(1994) 280R-314R.
[4] J.A. Olivares, N.T. Nguyen, C.R. Yanker and R.D.
Smith, Anal. Chem., 59 (1987) 1230-1232.
[5] R.D. Smith, J.H. Wahl and D.R. Goodlett, Anal.
Chem., 65 (1993) 574A-584A.
[6] W.M.A. Niessen, U.R. Tjaden and J. van der Greef,
J. Chromatogr., 636 (1993) 3-19.
[7] J. van der Greef and W.M.A. Niessen, Int. J. Mass
Spectrom. Ion Processes, 118/119 (1992) 857-873.
[8] E Kebarle and L. Tang, Anal. Chem., 65 (1993)
972A-986A.
[9] J.V. Iribarne and B.A. Thom~n, J. Chem. Phys., 64
(1976) 2287-2282.
[10] G. Schmelzeisen-Redeker, L. Buttering and F.W.
Rollgen, Int. J. Mass Spectrom. Ion Processes, 90
(1989) 139-150.
[11] J. Zeleny, Phys. Rev., 10 (1917) 1-6.
[12] M. Yamashita and J.B. Fenn, J. Phys. Chem., 88
(1984) 4451-4459.
[13] C.M. Whitehouse, R.N. Dreyer, M. Yamashita and
J.B. Fenn, Anal. Chem., 57 (1985) 675-679.
[14] A.L. Burlingame,R.K. Boydand S.J. Gaskeil, Anal.
Chem., 66 (1994) 634R-683R.
[15] R.D. Smith, C.J. Barinaga and H.R. Udseth, Anal.
Chem., 60 (1988) 1948-1952.
[16] J.H. Wahl, D.R. Goodlett, H.R. Udseth and R.D.
Smith, Anal. Chem., 64 (1992) 3194-3196.
J. Cai, J. Henion I J. Chromatogr. A 703 (1995) 667-692
[17] J.H. Wahl, D.R. Goodlett and H.R. Udseth, Electrophoresis, 14 (1993) 448-457.
[18] D.R. Goodlett, J.H. Wahl, H.R. Udseth and R.D.
Smith, J. MicrocoL Sep., 5 (1993) 57-62.
[19] K. Tsuji, L. Baczynskj and G.E. Bronson, Anal
Chem., 64 (1992) 1864-1870.
[20] M.A. Moseley, J.W. Jorgenson, J. Shabanowitz, D.F.
Hunt and K.B. Tomer, J. Am. Soc. Mass Spectrom., 3
(1992) 289-300.
[21] J.R. Perkins, C.E. Parker and K.B. Tomer, Electrophoresis, 14 (1993) 458-468.
[22] A.J. Tomlinson, L.M. Benson and S. Naylor, Am.
Lab., June (1994) 29-36.
[23] A.J. Tomlinson, L.M. Benson and S. Naylor, LC. GC,
12 (1994) 122-130.
[24] A.J. Tomlinson, L.M. Benson, K.L. Johnson and S,
Naylor, J. Chromatogr., 621 (1993) 239-248.
[25] J. Varghese and R.B. Cole, J. Chromatogr., 639 (1993)
303-316.
[26] J.V. Varghese and R.B. Cole, J. Chromatogr. A, 652
(1993) 369-372.
[27] J.R. Perkins, C.E. Parker and K.B. Tomer, J. Am.
Soc. Mass Spectrom., 3 (1992) 139-149.
[28] C.E. Parker, J.R. Perkins, K.B. Tomer, Y. Shida, K.
O'Hara and M. Kono, J. Am. Soc. Mass Spectrom., 3
(1992) 563-574.
[29] D.C. Gale and R.D. Smith, Rapid Commun. Mass
Spectrom., 7 (1993) 1017-t021.
[30] T.J. Thompson, F. Foret, D.E Kirby, P. Vouros and
B.L, Karger, presented at the 41st ASMS Conference
on Mass Spectrometry and Allied Topics, San Francisco, CA, 3I May-4 June 1993.
[31] T.J. Thompson, F. Foret, B.L. Karger and P. Vouros,
presented at the 42nd ASMS Conference on Mass
Spectrometry and Allied Topics, Chicago, IL, 29 May3 June 1994.
[32] S.K. Chowdhury and B.T. Chait, Anal. Chem., 63
(1991) 1660-1664.
[33] J.H. Wahl and R.D. Smith, J. Cap. Elec., I (1994)
62-71.
[34] J.H. Wahl, D.C. Gale and R.D. Smith, J. Chromatogr.
A, 659 (1994) 217-222.
[35] L. Fang, R. Zhang and R.N. Zare, presented at the
41st ASMS Conference on Mass Spectrometry and
Allied Topics, San Francisco, CA, 31 May-4 June
1993.
[36] M.S. Kriger, R.S. Ramsey and K.D. Cook, presented
at the 42nd ASMS Conference on Mass Spectrometry
and Allied Topics, Chicago, IL, 29 May-3 June 1994.
[37] G. Hopfgartner, T. Wachs, K. Bean and J. Henion,
Anal. Chem., 65 (1993) 439-446.
[38] M.G. Ikonomou, A.T. Blades and E Kebarle, Anal
Chem., 63 (1991) 1989-1998.
[39] E.D. Lee, W. Muck, J.D. Henion and T.R. Covey,
Biomed. Environ. Mass Spectrom., 18 (1989) 844-850.
[40] E.D. Lee, W. Muck, J.D. Henion and T.R. Covey, J.
Chromatogr., 458 (1988) 313-321.
689
[41] J.M. Johan:son. E.C. Huang, J.D. Henion and J.
Zweigenbaum, J. Chrontatogr.. 554 (199I) 311-327.
[42] F. Garcia and J. Henion, J. Chromatogr., 606 (1992)
237-47.
[43] E.D. Lee, W. Muck, J.D. Henion and T.R. Covey,
Biomed. Environ. Mass Spectrom., 18 (1989) 253-257.
[44] J.M. Johansson, R. Pavelka and j.D. Henion, J.
Chromatogr., 559 (1991) 515-528.
[45] F. Garcia and J.D. Henion, Anal. Chem., 64 (1992)
985-990.
[46] P. "Faibault, S. Pleasance and M.V. Laycock, J. Chromalogr., 542 (1991) 483-501.
[47] E Thibault, C. Paris and S. Pleasance, Rapid Commun. Mass Spectrom., 5 (1991) 484-490.
[48] S. Pleasance, P. Thibault and J. Kelly, J. Chromatogr.,
591 (1992) 325-339.
[49] J. Henion, A. Mordehai and J. Cai, Anal. Chem., 66
(1994) 2103-2109.
[50] L.W. Tetler, P A. Cooper and B. Powell, pre~nted at
the 42rid ASMS Conference on Mass Spectrometry and
Allied Topics, Chicago, IL, 29 May-3 June 1994.
[51] D. Wycherley, M.E. Rose, K. Giles, S.A. Jarvis and
M.A. McDowall, presented at the 41st ASMS Conference on Mass Spectrometry and Allied Topics, San
Francisco, CA, 31 May-4 June 1993.
[52] K.J. Volk, J. Liu, M.S. Lee, S.E. Klohr, E.H. Kerns
and I.E. Rosenberg, presented at the 41st ASMS
Conference on Mass Spectrometry and Allied Topics,
San Francisco, CA, 31 May-4 June 1993.
[53] R. Kostiainen, E. Lasonder and W. Bloemhoff, B/oL
Mass Spectrom., 23 (I994) 346-352.
[54] R. Kostiainen, E.J.F. Franssen and A.P. Bruins, J.
Chromatogr., 647 (1993) 361-365.
[55] J. Cai, H.K. Lira and J. Henion, presented at the 42nd
ASMS Conference on Mass Spectrometry and Allied
Topics, Chicago, 1L, 29 May-3 June 1994.
[56] T.G. Huggins and J.D. Henion, Electrophoresis, 14
(1993) 531-539.
[57] J.J. COLT,T.R. Covey and J.F. Anacleto, presented at
the 42rid ASMS Conference on Mass Spectrometry and
Allied Topics, Chicago, IL, 29 May-3 June 1994.
[58] L.W. Tetler, RA. Cooper and C.M. Can', Rapid
Commun. Mass Spectrom., 8 (1994) 179-182.
[59] R. Kostiainen, A.P. Bruins and V.M.A. Hakkinen, J.
Chromatogr., 634 (1993) 113-118.
[60] S. Pleasance, S.W. Ayer, M.V. Laycock and E
Thibault, Rapid Commun. Mass Spectrom., 6 (1992)
14-24.
[61] D. Wycherley, M.E. Rose, D. Rimmer, K. Giles, S.A.
Jarvis and M.A. McDowall, presented at the 41st
ASMS Conference on Mass Spectrometry and Allied
Topics, San Francisco, CA, 3I May-4 June 1993.
[62] M. Barber, R.S. Bordoli, R.D. Sedgwickand A. Tyler
N., J. Chem. Soc., Chem. Commun., (1981) 325-327.
[63] Y. Ito, T. Takeuchi, D. Ishi and M. Goto, J. Chromatogr., 346 (1985) I61-166.
[64] R.D. Minard, D. Chin-Fatt, P.J. Carry and A.G.
690
J. Cai, I. Henion I Z Chromatogr. A 703 0995) 667-692
Ewing, presented at the 36th ASMS Conference on
Mass Spectrometry and Allied Topics, San Francisco,
CA, 5-10 June 1988.
[65] N.J. Reinhoud, W.M.A. Niessen, U.R. Tjaden, L.G.
Gramberg, E.R. Verheij and J. van der Greef, Rapid
Commun. Mass Spectrom., 3 (I989) 348-351.
[66] R.D. Minard, D. Luckenbill, P.J. Curry and A.G.
Ewing, Adv. Mass Spectrom., 1I (1989) 436-437.
[67] R.M. Caprioli, W.T. Moore, M. Martin, B.B. DaGue,
K. Wilson and S. Modng, J. Chromatogr., 480 (1989)
247-257.
[68] S.M. Wolf, E Vouros, C. Norwood and E. Jakim, J.
Am. Soc. Mass Spectrom., 3 (1992) 757-761.
[69] J.S.M. de Wit, L.J. Deterding, M.A. Moseley, K.B.
Tomer and J.W. Jorgenson, Rapid Commun. Mass
Spectrom., 2 (1988) 100-104.
[70] M.A. Moseley, L.J. Deterding, K.B. Tomer and J.W.
Jorgenson, Rapid Commun. Mass Spectrom., 3 (1989)
87-93.
[71] M.A. Moseley, L.J. Deterding, K.B. Tomer and J.W.
Jorgenson, J. Chromatogr., 480 (1989) 197-209.
[72] M.A. Moseley, L.J. Deterding, K.B. Tomer and J.W.
Jorgenson, J. Chromatogr., 516 (1990) 167-173.
[73] M.A. Moseley, L.J. Deterding, K.B. Tomer and J.W.
Jorgenson, Anal. Chem., 63 (1991) 109-II4.
[74] L.J. Deterding, C.E. Parker, J.R. Perkins, M.A.
Moseley, J.W. Jorgenson and K.B. Tomer, J. Chromatogr., 554 (1991) 329-338.
[75] K.B. Tomer, L.J. Deterding, S.R. McGown and J.R.
Perkins, presented at the 40th ASM$ Conference on
Mass Spectrometry and Allied Topics, Washington, DC,
31 May-5 June 1992.
[76] M.J.-F. Suter and R,M. Caprioli, J. Am. Soc. Mass
Spectrom., 3 (1992) 198-206.
[77] W. Nichols, J. Zweigenbaum, F. Garcia, M. Johansson
and J. Henion, LC. GC, 10 (1992) 676-686.
[78] C.G. Edmonds, J.A. Loo and R.R. Ogorzalek Loo,
Biochem. Soc. Trans., 19 (1991) 943-947.
[79] M.A. Tart, G. Zhu and R.F. Browner, Anal. Chem.,
65 (1993) 1689-1695.
[80] M. Karas, U. Bahr, A. Ingendoh, E. Nordhoff, B.
Stahl, K. Stupat and F. Hillenkamp, Anal. Chim.
Acta, 241 (1990) 175-185.
[81] R. Takigiku, T. Keough, M.P. Lacey and R.E.
Schneider, Rapid Commun. Mass Spectrom., 4 (1990)
24-29.
[82] T. Keough, R. Takigiku, M.P. Lacey and M. Purdon,
Anal. Chem., 64 (1992) 1594-1600.
[83] W. Weinmann, K. Baumeister, I. Kaufmann and M.
Przybylski, J. Chromatogr., 628 (1993) 111-121.
[84] W. Weinmann, C.E. Parker, K. Baumeister, C. Maier,
K.B. Tomer and M. Przybylski, Electrophoresis, 15
(1994) 228-233.
[85] P.A. van Veelen, U.R. Tjaden and J. van der Greef, J.
Chromatogr., 647 (1993) 367-374.
[86] K.B. Tomer, W. Weinmann, D. Papac, L. Palm, C.E.
Parker, L. Deterding, M. Przybylski and J. Hoyes,
presented at the 42nd ASMS Conference on Mass
Spectrometry and Allied Topics, Chicago, IL, 29 May3 June 1994.
[87] C.R. Blakley, M.L. Vestal and A. Verentchikov, presented at the 42nd AJM$ Conference on Mass Jpec.
trometry and Allied Topics, Chicago, IL, 29 May-3
June 1994.
[88] A.F. Lecoq, S. Di Blase and L. Montanarella, J.
Chromatogr., 638 (1993) 363-373.
[89] D.F. Hunt, J.E. Alexander, A.L. McCormack, P.A.
Martino, H. Michel, J. Shabanowitz and N. Sherman,
in J.J. ViUafranca (Editor), Techniques in Protein
Chemistry II, Academic Press, New York, 1991, pp,
441-454.
[90] R.S. Ramsey, K.G. Asano, KJ. Hart, D.E. Goetinger, G.J. van Berkel and S.A. Mcluckey, presented at
the 41st ASM$ Conference on Mass Spectrometry and
Allied Topics, San Francisco, CA, 31 May-,; June
i993.
[91] K.J. Rosnuck, J.G. Stroh, G.C. Adrews and D.H.
Singleton, presented at the 41st AJMS Conference on
Mass Spectrometry and Allied Topics, San Francisco,
CA, 31 May-4 June 1993.
[92] A.J. Tomlinson, W.K. Nevala, W.D. Braddock, M.A.
Strausbauch, P.J. Wet~steinand S. Naylor, presented at
the 42nd ASM$ Conference on Mass Spectrometry and
Allied Topics, Chicago, IL, 29 May-3 June 1994.
[93] A.J. Tomlinson, L.M, Benson, M.A. Strausbauch,
G.F. Scanlan, P.J. Wettstein, F. Rusnak and S. Naylor,
presented at the 42nd ASMS Conference on Mass
Spectrometry and Allied Topics, Chicago, IL, 29 May3 June 1994.
[94] R.B. Cole, J. Varghese and R.M. McCormick, presented at the 42nd ASMS Conference on Mass Spectrometry and Allied Topics, Chicago, IL, 29 May-3
June 1994.
[95] T.J. Thompson, F. Foret, E Vouros and B.L. Karger,
Anal. Chem., 65 (1993) 900-906.
[96] R.D. Smith, H.R. Udseth, C.J. Barinaga and C.G.
Edmonds, J. Chromatogr., 559 (1991) 197-208.
[97] J.H. Wahl, S.A. Hofstadler, Z. Zhao, D.C. Gale and
R.D. Smith, presented at the 42nd ASMS Conference
on Mass Spectrometry and Allied Topics, Chicago, IL,
29 May-3 June 1994.
[98] R. Takigiku, T. Keough, M.P. Lacey, M.P. Purdon, L.
Licklider and W.G. Kuhr, presented at the 42nd ASM$
Conference on Mass Spectrometry and Allied Topics,
Chicago, IL, 29 May-3 June 1994.
[99] J.F. Kelly, E Thibault, H. Masoud, M.B. Perry and
J.C. Richards, presented at the 41st ASMS Conference
on Mass Spectrometry and Allied Topics, San Francisco, CA, 31 May-4 June 1993.
[100] J.F, Kelly, P. Thibault, S. Locke and L.R. Ramaley,
J. Cai, J. Henion I J. Chromatogr. A 703 (1995) 667-692
presented at the 42nd ASMS Conference on Mass
Spectrometry and Allied Topics, Chicago, IL, 29 May3 June 1994.
[I01] Z. Zhao, J.H, Wahl, A.F. Fuciarelli, D.C. Gale, H.R.
Udseth and R.D. Smith, presented at the 42nd ASMS
Conference on Mass Spectrometry aud Allied Topics,
Chicago, IL, 29 May-3 June 1994.
[lO21 J.P. Barry, P. Vouros and C.B. Norwood, presented at
the 42nd ASMS Conference on Mass Spectrometryand
Allied Topics, Chicago, IL, 29 May-3 June 1994.
[IO31F.Y.-L. Hsieh, J. Cai, Y.-T. Li, B. Ganem and J.
Henion, J. Am. Sac. Mass Spectrom., in press.
[104] D.R. Goodlett, R.R.O. Loo, J.A. Loo, J.H. Wahl,
H.R. Udseth and R.D. Smith, J. Am. Soc. Mass
Spectrom., 5 (1994) 614-622.
[IO51F. Uzabiaga, A. Tuong, J.C. Lonneau, M. Petitou and
C. Pieard, presented at the 42nd ASMS Conferenceon
Mass Spectrometry and Allied Topics, Chicago, IL, 29
May-3 June 1994.
[io6] I. Mylchreest, presented at the 40th ASMS Conference
on Mass Spectrometry and Allied Topics, Washington,
DC, 31 May-5 June I992.
[1o71J. Cai, A. Mordehai and J. Henion, presented at the
lOth Symposium on Liquid Chromatography~Mass
Spectrometry, Ithaca, NY, 21-23 July 1993.
[1OSl F.Y.-L. Hsieh, J. Cai and J. Henion, presented at the
42nd ASMS Conference on Mass Spectrometry and
Allied Topics, Chicago, IL, 29 May-3 June 1994.
[I09] S.C. Stinson, Chem. Eng. News, (1994) 38-72.
[1101 T,J. Ward, Anal. Chem., 66 (1994) 633A-640A.
[111] M. Novotny, H. Soini and M. Stefansson, Anal.
Chem., 66 (1994) 646A-655A.
[1121 R. Sheppard, X. Tong, J. Cai and J. Henion, presented at the 42nd ASMS Conference on Mass Spectrometry and Allied Topics, Chicago, IL, 29 May-3
June 1994.
[113]A,J. Tomlinson, L.M. Benson and S. Naylor, Electrophoresis, 15 (1994) 62-71.
[114] A.J. Tomlinson, L.M. Benson and S. Naylor, J. High
Resolut. Chromatogr., 17 (1994) 175-177.
[1151M.H. Lamoree, N.J. Reinhoud and U.R. Tjaden,
Biol. Mass Spectrom., 23 (1994) 339-345.
[1161H.B. Hines, M. Holcomb and E. Brueggemann, presented at the 42nd ASMS Conference on Mass Spectrometry and Allied Topics, Chicago, IL, 29 May-3
June I994.
[1171 M.D. LeBlanc and L. Ramaley, presented at the 41st
ASMS Conference on Mass Spectrometry and Allied
Topics, San Francisco, CA, 31 May-4 June 1993.
[118] W.M. Muck and J.D. Henion, J. Chromatogr., 495
(1989) 41-59,
[119] J.C. Severs and D.E. Games, presented at the 42nd
ASMS Conference on Mass Spectrometry and Allied
Topics, Chicago, IL, 29 May-3 June 1994.
[120] A.P. Tinke, N.J. Reinhoud, W.M.A. Niessen, U.R.
691
Tjaden and J. van der G~ef, Rapid Comm~m. Mass
Spectrom., 6 (1992) 560-563.
[121] N.J. Reinhoud, A.P. Tinke, U.R. Tjaden, W.M.A.
Niessen and J. van der Oreef, J. C h r ~ r . ,
627
(1992) 263-271.
[122] M.A. Mo~ley, presented at the 42nd ASMS Confer.
ence on Mass Spectrometryand Allied Topics, Chicago,
IL, 29 May-3 June 1994.
[123] M.H. Lamoree, N.J. Reinhoud, U.R. Tjaden,W.M.A.
Niessen and J. Van der Greef, prcsent~l at the 41st
ASM$ Conference on Mass Spectrometry and Allied
Topics, San Francisco, CA, 31 May-4 June 1993.
[124] J. Van der Greef, A.P. Tinke, N.J. Reillhoud, W.M.A.
Niessen and U.R. Tjaden, presented at the 40th ASMS
Conference on Mass Spectrometry and Allied Topics,
Washington, DC, 31 May-5 June 1992.
[125] T.J. Thompson, E Vouros, F. Foret and B. Karger,
presented at the 40th ASMS Conference on Mass
Spectrometry and Allied Topics, Washington, DC, 31
May-5 June 1992.
[126l J. Cai and Z. El Rassi, J. Liq. Chromatogr., 15 (1992)
2007-2024.
[127] J. Cai and Z. El Rassi, J. Liq. Chromatogr., 15 (1992)
1179-1192.
[128] I. Morita and J.I. Sawada, J. Chromatogr., 641 (1993)
375-381.
[129] M.E. Swartz and M. Merion, J. Chromatogr., 632
(1993) 209-213.
[130] D. Kirby, K.F. Greve, F. Foret, P. Vouros, B.L.
Karger and W. Nashabeh, presented at the 42nd ASMS
Conference on Mass Spectrometry and Allied Topics,
Chicago, IL, 29 May-3 June 1994.
[131] J. Kohr and H. Engelhardt, J. Chromatogr., 652
(1993) 309-316.
[132] T. Wehr, LC. GC. 11 (1993) 14-20.
[133] W. Weinmann, M. Przybylski, C.E. Parker and K.B.
Tomer, presented at the 41st ASMS Conference on
Mass Spectrometry and Allied Topics, San Francisco,
CA, 31 May-4 June 1993.
[134] R.D. Smith, J.A. Loo, C.G. Edmonds, C.J. Barinaga
and H.R. Udseth, J. Chromatogr., 516 (1990) 157165.
[135] S.A. McLuckey, G.J. van Berkel, D.E. Goeringer and
G.L. Glish, Anal. Chem., 66 (1994) 689A-696A.
[136] R.S. Ramsey, D.E. Goeringer and S.A. Mclnckey,
Anal. Chem., 65 (1993) 3521-3524.
[137] J.C. Schwartz and I. Jardine, presented at the 40th
ASMS Conference on Mass Spectrometry and Allied
Topics, Washington, DC, 31 May-5 June 1992.
[138] K.D. Henry, J.P. Qainn and F.W. McLafferty, J. Am.
Chem. Soc., 113 (1991) 5447-5450.
[139] M.V. Gorshkov, E.N. Nikolaev, S. Guan, K.Vermillion
and A.G. Marshall, presented at the 4151 ASMS
Conference on Mass Spectrometry and Allied Topics,
San Francisco. CA, 31 May-4 June 1993.
692
J. Cai, J. Henion I J. Chromatogr. A 703 (1995) 667-692
[140] S.A. Hofstadler, J.H. Wahl, J.E. Bruce and R.D.
Smith, J. Am. Chem. Soc., 115 (1993) 6983-6984.
[141] P.J. Johnson, ED. Mills, D.S. Gross, ED. Schnier and
E.R. Williams, presented at the 42nd ASMS Conference on Mass Spectrometry and Allied Topics, Chicago,
IL, 29 May-3 June 1994.
[142] J.A. Castoro, R.W. Chiu, C.A. Monning and C.L.
WiIkins, J. Am. Chem. Soc., 114 (1992) 7571-7572.
[143] L. Fang, R. Zhang and R.N. Zar¢, ~resented at the
42nd ASMS Conference on Mass Spectrometry and
Allied Topics, Chicago, IL, 29 May-3 June I994.
[144] R. Zhang, L. Fang and R.N. Zare, presented at the
#2nd ASMS Conference on Mass Spectrometry and
Allied Topics, Chicago, IL, 29 May-3 June 1994.
[145] N.J. Reinhoud, E. Schroder, U.R. Tjaden, W.M.A.
Niessen, M.C. ten Noever de Brauw and J. van de
Greef, 1. Chromatogr., 516 (1990) 147-155.