Miniature flowing atmospheric-pressure afterglow ion source for facile interfacing of CE with MS.

Abstract

Here, we present a miniaturized version of the flowing atmospheric-pressure afterglow (miniFAPA) ion source and use it for sheathless coupling of CE with MS. The simple design of the CE-miniFAPA-MS interface makes it easy to separate the electric potentials used for CE and for ionization. A pneumatically assisted nebulization of the CE effluent transfers the analytes from the liquid phase into the gas phase before they are ionized by interacting with reactive species produced by the FAPA. An important advantage of this interface is its high stability during operation: optimization of five different parameters indicated that the interface is not sensitive to minor deviations from the optimum values. Other advantages include ease of construction and maintenance, as well as relatively low cost. Samples with complex matrices, such as yeast extract, soil extract and urine, spiked with the test compounds, were successfully analyzed using the CE-miniFAPA-MS setup.

Full text

Page 1

Research Article
Miniature flowing atmospheric-pressure
afterglow ion source for facile interfacing
of CE with MS
Here, we present a miniaturized version of the flowing atmospheric-pressure afterglow
(miniFAPA) ion source and use it for sheathless coupling of CE with MS. The simple
design of the CE-miniFAPA-MS interface makes it easy to separate the electric potentials
used for CE and for ionization. A pneumatically assisted nebulization of the CE effluent
transfers the analytes from the liquid phase into the gas phase before they are ionized by
interacting with reactive species produced by the FAPA. An important advantage of this
interface is its high stability during operation: optimization of five different parameters
indicated that the interface is not sensitive to minor deviations from the optimum values.
Other advantages include ease of construction and maintenance, as well as relatively low
cost. Samples with complex matrices, such as yeast extract, soil extract and urine, spiked
with the test compounds, were successfully analyzed using the CE-miniFAPA-MS setup.
Keywords:
Ambient ionization techniques / CE / CE-MS / Flowing atmospheric-pressure
afterglow / MSDOI 10.1002/elps.201000350
1Introduction
Analysis of small molecules in complex sample matrices can
be challenging. MS is a potent analytical technique which
permits identification of analytes based on their mass-to-
charge ratios as well as fragmentation patterns. However,
MS often needs to be hyphenated on-line with a separation
technique in order to cope with analysis of complex samples
and matrices. CE is a versatile separation platform, which
generally offers high separation efficiency and can handle
nanolitre volumes of samples.
Various on-line interfaces for ion formation in CE-MS,
which fulfill the requirements of diverse analytical work
regimes, have been introduced to date. These include CE-
ESI [1], CE-ICP ionization [2, 3], CE-atmospheric pressure
chemical ionization (APCI) [4], and CE-atmospheric pres-
sure photoionization [5]. The most common CE-MS inter-
faces are based on the ESI process (for recent reviews see
[6–22]). These are capable of providing excellent sensitivity
in analyses even of complex molecules. Moreover, CE-ESI-
MS systems are available commercially. There are three
main types of CE-ESI-MS interfaces: (i) the sheathless
interface, (ii) the sheath-flow interface, and (iii) the liquid
junction interface [14]. The sheath-flow design has become
especially popular, probably due to its high stability, ease of
operation, and flexibility [14, 20, 21]. However, its disad-
vantage is the dilution of the CE effluent resulting from the
relatively high flow rates of the sheath liquid (up to 100 ?
the flow rate of the CE effluent). For this reason, the
sensitivity obtained with the sheath-flow interfaces is typi-
cally lower than that observed for sheathless interfaces.
Irrespective of the presence or absence of sheath liquid, the
ESI process requires application of an electric potential
between the capillary outlet and the MS inlet cone. Since CE
itself needs an electric field for the separation of molecules,
implementation of CE-ESI-MS interfaces often requires
coupling two power supplies, one for the CE separation and
one for the ESI [13]. The present study demonstrates a
proof-of-concept of an alternative sheathless CE-MS inter-
face, incorporating a miniature flowing atmospheric-pres-
sure afterglow (miniFAPA) ion source, which enables
decoupling of the separation and ionization voltages and is
characterized by relatively stable operation.
Matthias C. Jecklin?
Stefan Schmid?
Pawel L. Urban?
Andrea Amantonico
Renato Zenobi
Department of Chemistry and
Applied Biosciences, ETH Zurich,
Zurich, Switzerland
Received July 1, 2010
Revised August 7, 2010
Accepted August 9, 2010
Abbreviations:
ionization; FAPA, flowing atmospheric-pressure afterglow;
miniFAPA,
miniatureflowing
afterglow
APCI,
atmosphericpressure chemical
atmospheric-pressure
?These authors have contributed equally to this work.
Correspondence: Professor Renato Zenobi, Department of
Chemistry and Applied Biosciences, ETH Zurich, CH-8093 Zurich,
Switzerland
E-mail: zenobi@org.chem.ethz.ch
Fax: 141446321292
& 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.electrophoresis-journal.com
Electrophoresis 2010, 31, 3597–3605
3597

Page 2

The FAPA ion source, also referred to as flowing after-
glow atmospheric pressure glow discharge [23, 24], is an
example of a direct current plasma source [25]. It operates by
the formation of a glow discharge in helium between a pin
cathode and plate anode, with voltages between 400 and
500 V and electric currents of roughly 20–25 mA [26]. This
ion source was recently introduced by Andrade et al. for the
analysis of gaseous samples [23] as well as for direct
desorption/ionization from solid surfaces [24]. Direct
desorption/ionization FAPA-MS has already been used for
the analysis of pesticides in fruit [27], as well as for char-
acterizing polymer samples and plasticizers [28], but the
potential of FAPA in interfacing separation techniques with
MS, and the possibility to decouple the electric potentials
used for CE and ionization, has not yet been explored.
CE, when used together with MS, can be considered
alternative and orthogonal to the classical chromatographic
separation techniques routinely used with MS [19, 20].
Nonetheless, hyphenation of CE and MS is still seen to be
difficult [21, 22]. The study uses a downscaled version of the
FAPA ion source for hyphenation of CE with MS. We
believe that it could be easily adapted to miniature sample
preparation and separation systems, including microchips.
By way of example, it will be shown that with the simple CE-
miniFAPA-MS interface, it is possible to detect several test
compounds in unpurified samples containing complex
matrices.
2 Materials and methods
2.1 Materials
Acetic acid, pyridine, 3-methylpyridine, 2,4,6-trimethylpyr-
idine, tripropylamine, p-chloraniline, and N,N-dimethylben-
zamide were all purchased from Sigma-Aldrich (Buchs,
Switzerland). Sodium hydroxide was from Siegfried Handel
(Zofingen, Switzerland); ethanol from Scharlau Chemie
(Sentmenat, Spain);methanol
(Loughborough, UK). Nanopure water (r418 MOcm) was
obtained from a NANOpure Diamond system (Skan, Basel,
Switzerland).
Urine was collected from a healthy volunteer on the day
of the experiment. Soil extract was purchased from Carolina
Biological Supply (Burlington, NC, USA). A yeast extract
was prepared as follows: ?1 g of dry yeast (Saccharomyces
cerevisiae, type I; YSC1; Sigma-Aldrich) was heated with
20 mL of water in a microwave oven at 700 W for the total
time of 3 min. All the real samples were filtered with a
polyvinylidene fluoride syringe filter (pore size 0.45 mm;
Whatman, Clifton, NJ, USA).
from FisherScientific
2.2 CE
For the CE separation, a polyimide-coated fused silica
capillary manufactured by Polymicro (id 75 mm, od 363 mm,
length 58 cm; BGB Analytik, Boeckten, Switzerland) was
used. A home-built CE system, used in a previous study [29],
was adapted for the purpose of the study. When used for the
first time, the capillary was flushed with 1 M NaOH using a
1 bar overpressure for 1 h, followed by briefly flushing with
water and BGE. Then, on regular basis (approximately every
2 h), it was flushed for several minutes with 1 M NaOH,
water, and the BGE. When the capillary was conditioned,
the nebulizer gas was switched off and tissue paper was
placed at the outlet of the separation capillary to collect the
effluent.
2.3 CE-MS interface
A schematic representation of the CE-MS interface is shown
in Fig. 1A. Upstream of the interface, a UV imaging
detector (ActiPix D100; Paraytec, York, UK) was mounted to
provide a UV absorption signal; a 200 nm UV filter was
used. The outlet of the CE capillary (?10 cm) was
coated with silver paint (‘‘Leitsilber 200’’, Evonik Degussa,
Du ¨sseldorf, Germany) (Fig. 1B). The silver paint was
applied using a hair-free brush (ITW Texwipe, Mahwah,
NJ, USA). The coated outlet section of the CE capillary was
inserted into a wider-bore stainless steel capillary mounted
inside the T-junction (Fig. 1A). The added stiffness provided
by the silver paint ensures on-axis alignment of the two
capillaries. A nebulizer gas (N2) was used to vaporize the
effluent of the CE capillary. A miniature CMOS camera
(size: 15?22 mm, lens: f53.6 mm, 250 K (NTSC)) was
positioned close to the CE-miniFAPA-MS interface; this
enabled real-time monitoring of the interface and straight-
forward alignment of the main interface components
(a picture taken by the CMOS camera is reproduced in
Fig. 1C).
2.4 Miniature FAPA ion source
The miniFAPA ion source is a redesigned version of the one
used in previous study [23, 27] (Fig. 1D). It consists of a
10-cm-long glass tube with an id of 6 mm. A copper cap
(8 mm diameter and 5 mm length) with a central 0.4-mm
diameter hole, serving as the anode, was fixed to the front
end of the glass body. A stainless steel pin (;51.1 mm), the
cathode, is centered inside the glass body using a PTFE
spacer and fixed with a ferrule in the rear end of the glass
body. The PTFE-spacer contains eight small holes to permit
gas to flow through. The source was operated at 15.2 mA
and 470–490 V using a power supply (HCN 140–3500,
F.u.G. Elektronik, Rosenheim, Germany). A gas connection
was glued to the glass body enabling application of helium
(99.999%; PanGas, Dagmersellen, Switzerland) at a pres-
sure of ?0.4 bar. This resulted in a helium gas flow rate of
?1 L/min. The miniFAPA ion source was positioned such
that the outer electrode (the copper cap) faced the spray
path, as shown in Figs. 1A and C.
Electrophoresis 2010, 31, 3597–3605
3598
M. C. Jecklin et al.
& 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.electrophoresis-journal.com

Page 3

2.5 MS
A hybrid QTOF MS (Q-TOF Ultima; Waters/Micromass,
Manchester, UK) was used for MS analysis. All the
measurements were performed in positive ion mode. The
cone voltage and RF1 lens voltage were set to 50 and 60 V,
respectively, and the source temperature was kept at 501C.
Calibration of the instrument was performed using cesium
iodide clusters generated by electrospraying a solution of
cesium iodide in water/2-propanol (1:1, v/v) at a concentra-
tion of 2 mg/mL.
2.6 Data acquisition and treatment
MS data acquisition was controlled by the MassLynx (v 4.0)
software (Waters/Micromass). During the optimization
stage, mass spectra were background-subtracted using the
Mass Lynx (polynomial order: 4; subtraction below curve:
25.0%; tolerance: 0.01). In the case of urine sample spiked
with the test analytes, the mass spectra were further
analyzed using MATLAB software (v 7.6.0.324 (R2008a);
MathWorks, Natick, MA, USA). All the ASCII files
corresponding to a single CE run were automatically loaded
and displayed as a two-dimensional shaded surface using
MATLAB (defined by ‘‘migration time’’ and ‘‘m/z’’ axes), in
which each point is related to MS signal. Finally, a
background subtraction procedure was performed on the
data: a ‘‘background spectrum’’ corresponding to a certain
migration time, where no spiked analytes were detected (e.g.
from the beginning of the CE-MS run), was subtracted
arithmetically from each spectrum constituting the 2D
graph. UV absorption electropherograms were recorded
using the ActiPix D100 software (2008, v 1.0.1120; Paraytec);
an exponential filter with a time constant of 0.5 s was
applied.
Figure 1. Experimental setup
used for coupling CE with MS
using a minFAPA ion source.
(A) Schematic representation
showing key components of
the setup including
right to left): CE inlet vial, CE
separation capillary,
detector, nebulizer T-element
assembly with
section of CE capillary, mini-
FAPA ion source, and MS inlet
cone. The main geometrical
parameters adjusted during
interfaceoptimization
marked as a–d and a. Figure
not drawn to scale. (B) SEM of
the CE capillary tip coated
with‘‘Leitsilber
Picture of the CE-miniFAPA-
MS interface recorded by the
CMOS camera (cf. Supporting
Information). (D) Schematic of
the miniFAPA ion source.
(from
UV
thecoated
are
200’’.(C)
Electrophoresis 2010, 31, 3597–3605
Microfluidics and Miniaturization
3599
& 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.electrophoresis-journal.com

Page 4

3Results and discussion
3.1 Optimization of the interface
During optimization, a mixture of three compounds
(3-methylpyridine, p-chloraniline and tripropylamine) was
continuously infused via the CE capillary by application of
an overpressure of 1 bar to the inlet vial. The following
parameters were optimized, one at a time, within the
technically allowed ranges (Fig. 1A): length of the spray path
(a: 5–30 mm), distance from the orifice of the miniFAPA ion
source to the orifice of the MS (b: 0–15 mm), distance from
the orifice of the miniFAPA ion source to the spray path
(c: 2.5–20.0 mm), angle measured between the spray path
and central axis of the miniFAPA ion source (a: 30–1501),
and the nebulizer gas flow (n: 150–350 L/h). Based on the
results (Fig. 2), we chose the following values to be used in
further experiments: a515 mm, b57.5 mm, c52.5 mm,
a5901, and n5250 L/h. The distance d, the protrusion of
the CE capillary from the nebulizer capillary, was set to
?0.5 mm, based on preliminary tests. In most cases, the
trends observed for the three test compounds were the
same. However, some of the parameters are not completely
independent from one another. For example, as shown in
Fig. 2, the optimum values for parameters a and c are 15
and 2.5 mm, respectively. Nonetheless, with these values,
the stability of the signal diminished (cf. error bars in the
bottom panel of Fig. 2). This may be due to an effect of the
helium gas coming out of the orifice of the miniFAPA ion
source on the nebulized CE effluent; probably this effect is
more significant when the miniFAPA ion source is very
close to the spray path (c52.5 mm). Therefore, in order to
minimize possible deflection of the analyte species near the
miniFAPA orifice, we decided to use a value a510 mm,
instead of 15 mm. Also in the case of the angle a, we
decided to use 901 instead of 1201, since a lower ion
background was observed for 901. Ionization of the
analytes could be observed even when the miniFAPA ion
source was placed relatively far from the spray path
(c520 mm) (Fig. 2, bottom panel). While the FAPA (the
visible plume) extends over o10 mm (cf. Fig. 1C), this
observation suggests that the zone beyond the afterglow
region also contains a certain amount of reactive species
which can ionize the nebulized molecules of the analytes
(see also Section 3.5).
Based on the information presented in Fig. 2, we
conclude that the CE-miniFAPA-MS interface is relatively
insensitive to minor changes to the arrangement of the
interface components (cf. Fig. 1A). For all the parameters,
the ionization efficiency varies within only one order of
magnitude (Fig. 2). One possible explanation is that since
there is almost no electric field in the area adjacent to the
capillary tip, the interface is less affected by deviation of the
relative positioning of its key parts (capillary tip, miniFAPA
device, MS orifice). Since this interface does not require
frequent maintenance and is characterized by stable opera-
tion (Supporting Information), it is suitable for applications
where frequent maintenance and adjustment of key para-
meters is not possible. Such applications include on-site
quality control in production plants and environmental
monitoring. Use in conjunction with microfluidic devices,
in which the introduction of the ionization potential might
affect execution of the sample treatment steps performed in
the microfluidic channel (upstream of the MS interface)
should also be of interest.
3.2 Establishing the electric contact for CE
A technical challenge in CE-MS interfaces in general is to
establish a stable electric contact at the end of the capillary
while preserving optimum conditions for separation as well
as ionization [15]. In the CE-miniFAPA-MS interface, no
sheath liquid needs to be used; the electric contact with the
silver-coated capillary tip was found to be sufficient for the CE
separation. Neither is there a need to apply an electric
potential for ionization, which occurs mainly in the gas
phase. In general, narrow capillary emitter tips, used in
sheathless ionization, are susceptible to clogging [30] and to
deterioration of the conductive coating used as an electric
contact [13]. In the interface presented here, the effluent of
the CE capillary is nebulized with pure nitrogen gas, and
there is no requirement to tune the electric field strength by
sharpening the capillary end. The silver paint, used to
establish the electric contact for the CE separation in the
CE-miniFAPA-MS, adhered very well to the external poly-
imide layer (Fig. 1B), therefore we did not need to remove the
polymer cladding prior to spreading silver paint over the
untapered capillary end. Although there is a physical gap
(?140 mm) between the capillary lumen and the silver coating
outside capillary (Fig. 1B), this did not seem to cause
problems with the stability of the electric current during CE
runs (tested with 20kV). In fact, the CE current was stable,
with fluctuations normally not exceeding ?5%.
3.3 Coupling CE with miniFAPA-MS
In the next step, we applied CE separation together with the
miniFAPA-MS detection. Since dilute acetic acid is frequently
used as BGE in CE-MS [12, 16], we decided to use 50 mM
acetic acid (pH53.1) in this first demonstration of the
CE-miniFAPA-MS setup. Figure 3 presents UV electropher-
ograms as well as MS extracted ion currents obtained during
analysis of a standard analyte mixture (Table 1). The time
shift between the UV and MS traces is due to positioning of
the UV detector ?8 cm upstream of the CE-miniFAPA-MS
interface (cf. Fig. 1A). The migration order of p-chloraniline,
N,N-dimethylbenzamide, and the other four compounds
taken together, is as expected, with the migration times being
related to the solution-phase charge and hydrodynamic
radius. Baseline separation was achieved for almost all pairs
of electrophoretic peaks in MS traces (Fig. 3). The
CE resolution, as monitored by UV absorption detection,
Electrophoresis 2010, 31, 3597–3605
3600
M. C. Jecklin et al.
& 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.electrophoresis-journal.com

Page 5

does not seem to deteriorate following ionization of analytes
by miniFAPA and detection with Q-TOF-MS (Fig. 3). The
baseline in Fig. 3 is flat, even though no smoothing was
applied before the data display. LODs obtained here range
from 74737 to 6107143fmol for 2,4,6-trimethylpyridine
and tripropylamine, respectively (Table 1). In terms of
sensitivity, these figures locate the current CE-miniFAPA-
MS version at the level of some of the existing sheath-flow
CE-ESI-MS interfaces (see, for example, the review article
[12]). However, the CE-miniFAPA-MS is simpler to operate:
there is no need for applying sheath liquid or an ionization
potential to the CE capillary outlet. Some embodiments of
sheathless CE-ESI-MS interfaces have been reported to
provide sub-femtomole LODs (see, for example, [1, 31, 32]).
The sensitivity of our design could probably be improved, e.g.
when operating it with more sensitive mass analyzers than
the one we had available for the present demonstration. On-
line pre-concentration steps can also be incorporated into CE-
MS methods [17, 33], offering benefits with respect to the
concentration sensitivity.
3.4 Analysis of samples with complex matrices
Despite the high resolution of TOF analyzers, interferences
from the sample matrix and other analytes are often
problematic; therefore, separation prior to MS is indispen-
sable [17]. In the next experiment, samples within complex
matrices, such as a yeast extract, a soil extract and urine,
spiked with the test compounds, were successfully analyzed
using the proposed setup. Figure 4A illustrates three MS
electropherograms of p-chloraniline (extracted ion current
traces for m/z 128) in yeast extract, soil extract, and urine.
Figure 4B shows CE-UV and CE-MS electropherograms
obtained during analysis of a urine sample spiked with the
same standard compounds as in Table 1. Clearly, the
miniFAPA-MSsystemenables
compounds even whenthey
complex matrix (urine). In addition, by performing CE-
miniFAPA-MS/MS analysis of selected precursor ions, it
was straightforward to obtain fragmentation spectra (data
not shown).
detection
are
ofall
with
the
injecteda
Figure 2. Results of the optimi-
zation study of five key para-
meters: a (b5a/2, c55mm,
a5901,
n5250L/h),
15mm,c55mm,
n5250L/h),c
b57.5mm, a5901, n5250L/h),
a
(a515mm,
c55mm, n5250L/h), n (a5
15mm, b57.5mm, c55mm,
a5901). Refer to Fig. 1A for
description of a, b, c, a and n.
Each point corresponds to the
mean value of three measure-
ments, each one being a sum
of intensities recorded over a
30-s period. The dotted lines
markthevalues
for further experiments. Test
mixture:3-methylpyridine,
1.02mM; p-chloraniline, 0.951
mM;tripropylamine,
mM. Sample flow rate: ?8mL/
min.
b
a5901,
(a5
(a515mm,
b57.5mm,
selected
0.526
Electrophoresis 2010, 31, 3597–3605
Microfluidics and Miniaturization
3601
& 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.electrophoresis-journal.com

Page 6

Some of the species – e.g. m/z 142, 144, and 145 –
appear at the same time (Fig. 4B); therefore, one can
assume a common origin of these three species. The peak at
m/z 144 corresponds to protonated tripropylamine, whereas
the peak at m/z 145 (?) is the corresponding13C-isotope
peak. The peak at m/z 142 (??) represents the species
produced by loss of H2 from the tripropylamine ion.
Isotope peaks cannot be seen for most analytes (except for
tripropylamine and p-chloraniline) due to the signal
intensity threshold chosen for displaying the 2D plot
(Fig. 4B).
3.5 Comparison with the APCI interface and final
considerations
In general, most APCI interfaces are designed for LC-MS
and function properly only under high flow-rate conditions;
consequently, relatively low S/N ratios were observed when
high sheath flow rates were used in CE-APCI-MS [34]. In
APCI, the initial ionization in corona discharge takes place
in a relatively small volume near the needle tip [35]. An
interesting APCI interface for CE-MS has been presented by
Tanaka et al. [4]. This interface is very dissimilar to our
setup: (i) it has a needle to generate a corona discharge, (ii) it
uses a sheath liquid, and (iii) it incorporates a vaporizer
that heats up to 4001C. Similarly to CE-APCI-MS, the
CE-atmospheric pressure photoionization-MS also incorpo-
rates a vaporizer [36]. However, in this study using
CE-miniFAPA-MS, the transfer of analytes from the liquid
phase into the gas phase was conducted with the optimized
nebulizer gas flow rate. Importantly, this did not cause any
obvious oscillations of the capillary end.
The miniaturization of FAPA is believed to limit
dispersion of the gas-phase species in front of the MS
orifice. The reaction time in the present version of the
CE-miniFAPA-MS interface is roughly estimated to be in
the order of ?0.1 ms. One could expect that, in general,
downscaling the afterglow region poses a limit on the
interaction time between the analyte and reactive species.
However, the resulting decrease of sensitivity is outweighed
Figure 3. Representative UV and MS electropherograms showing
electrophoretic separation of the standard mixture: pyridine (pyr),
0.124mM;3-methylpyridine(3mp), 0.102mM;2,4,6-trimethylpyridine
(tmp), 0.0751mM; tripropylamine (tpa), 0.0526mM; p-chloraniline
(cha),0.0951mM; N,N-dimethylbenzamide
The sample was prepared in a 5mM solution of acetic acid.
CE-MSparameters:fusedsilica
58cm?75mm?363mm; BGE, 50mM acetic acid (pH53.1); injection,
?7mbar 5s; separation voltage, 120kV; hydrodynamic pressure
applied to the inlet vial during separation, ?7mbar; UV detection
(50cm downstream from the CE capillary inlet), 200nm; ionization,
miniFAPA (a510mm, b55mm, c52.5mm, a5901, n5250L/h);
MS detection, Q-TOF-MS. The injection volume, calculated using a
rearranged Poiseuille equation, was ?4.5nL. CE current: ?4.1mA.
Color bars mark the corresponding time windows of the UV and MS
traces. ? indicates an unidentified feature.
(dmba), 0.168mM.
capillary,length?id?od:
Table 1. Molecular weights (Mw), dissociation constants (pKa), migration times (tM), observed m/z values (protonated form), and LODs
(3?S/N criterion) calculated according to the MS traces obtained by CE-miniFAPA-MS
CompoundMw(g/mol)pKa
a)
tM(min)m/z [M1H]1 b)
LODc)(fmol)
Pyridine
3-Methylpyridine
2,4,6-Trimethylpyridine
Tripropylamine
p-Chloraniline
N,N-Dimethylbenzamide
79.10
93.13
121.18
143.27
127.57
149.19
5.2
5.6
6.6
10.6
4.0
n.f.
3.0
3.2
3.7
3.9
5.5
6.5
80
94
277722
149720
74737
6107143
251747
42577
122
144
128
150
a) Different on-line sources.
b) m/z for analyte peaks were rounded to the nearest integer value.
c) 70.5?spread.
n.f. Not found.
Electrophoresis 2010, 31, 3597–3605
3602
M. C. Jecklin et al.
& 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.electrophoresis-journal.com

Page 7

Figure 4. (A) Three MS electropherograms (m/z 128) of p-chloraniline (0.951 mM) in yeast extract, soil extract, and urine sample. The
samples were prepared by mixing 0.5 mL of filtered yeast extract, soil extract, or urine with appropriate amounts of the standards, 100 mL
of 50 mM acetic acid, and adjusted to the volume of 1 mL with water. Migration times have been aligned for clarity of graphical
presentation. (B) CE-UV and CE-MS electropherograms showing the result of analysis of a urine sample spiked with the test compounds
(Table 1) using CE-miniFAPA-MS. Spiked standard compounds: pyridine (pyr), 1.24 mM; 3-methylpyridine (3 mp), 1.02 mM; 2,4,6-
trimethylpyridine (tmp), 0.751 mM; tripropylamine (tpa), 0.526 mM; p-chloraniline (cha), 0.951 mM; N,N-dimethylbenzamide (dmba),
1.68 mM. One asterisk (?) indicates an isotope variant (m/z 145) of tripropylamine (m/z 144); two asterisks (??) indicate [M1H-H2]1(m/z
142), related to hydrogen loss of tripropylamine ion; most isotope variants of the spiked analytes as well as matrix components are not
seen due to the preset ion intensity threshold. Dashed lines approximately mark common points on the separation time line, accounting
for the shift due to positioning of the UV detector, ?8 cm upstream from the CE capillary outlet; fluctuations of migration velocity might
cause additional misalignment of the corresponding UV and MS features. Mass increment has been reduced to 0.1 m/z values for clarity
of graphical presentation. CE-MS parameters in (A) and (B) are the same as in Fig. 3.
Electrophoresis 2010, 31, 3597–3605
Microfluidics and Miniaturization
3603
& 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.electrophoresis-journal.com

Page 8

by the compatibility with the low flow rates (down to sub-
microliter per minute) of effluents from CE capillaries or
microchips. An investigation of the ionization mechanism is
not the topic of this study; like for other compounds
analyzed with FAPA-MS [23, 24, 27, 28], we also observed
formation of protonated molecules [M1H]1. As described
by Andrade et al. [23], the FAPA ionization process resem-
bles the one observed in most common APCI sources,
operated with corona discharge; therefore, yielding mostly
charged water clusters ([H2O]nH1) as active species.
4Concluding remarks
Implementation of the miniaturized FAPA ion source in
CE-MS brings some practical advantages:
(i)The outer plate electrode of miniFAPA is grounded,
which eliminates technical obstacles associated with
application of electric potentials for CE and ionization.
The interface offers good stability of operation; the ion
intensity was not strongly affected by slight deviation
of the interface alignment from the optimum.
Downscaling the ion source body to the size of a pen
(8 mm in diameter) enables straightforward position-
ing and rearrangement of the miniFAPA ion source
and a microscale capillary tip.
Construction of the ion source for the interface
described here is relatively simple and achieved using
inexpensive materials (o$10).
While the miniFAPA ion source is generally main-
tenance-free, its disassembly is very easy and can be
accomplished within 1 min allowing for fast but
thorough cleaning of the copper electrode.
(ii)
(iii)
(iv)
(v)
The LOD achieved in this study were in the femtomole
range. Further work should thus focus on improving the
sensitivity, for instance, by fine-tuning the ion collection
efficiency at the MS orifice, choice of another mass analyzer,
and optimization of the acquisition mode (e.g. multiple
reaction monitoring). Due to the minuscule technical
requirements for operating miniFAPA and its anticipated
ruggedness, following its further development, the new
interface may be useful in conjunction with portable ion
trap mass spectrometers designed for performing analysis
of target species of interest outside the laboratory [37], when
coupled with microscale capillaries and microchips, or in
industrial quality control systems; complementing the list of
the established CE-MS interfaces in current use.
We would like to thank Mr. Konstantin Barylyuk, Mr.
Christoph Ba ¨rschti, Mr. Heinz Benz and Dr. Frank Krumeich
for their support. We are also grateful to Dr. Mebs Surve and
Paraytec (York, UK) for providing the ActiPix D100 UV
imaging detector. M.C.J. was funded by the Swiss National
Science Foundation (SNSF, Grant No. 200020_124663). S.S.
was funded by Askair (Askokoro, Switzerland). P.L.U. was
funded by a Marie Curie Intra European Fellowship received
within the 7th European Community Framework Programme
(Contract No. PIEF-GA-2008-219222 - MESEL).
The authors have declared no conflict of interest.
5 References
[1] Olivares, J. A., Nguyen, N. T., Yonker, C. R., Smith, R. D.,
Anal. Chem. 1987, 60, 1230–1232.
[2] Lu, G. H., Bird, S. M., Barnes, R. M., Anal. Chem. 1995,
67, 2949–2956.
[3] Olesik, J. W., Kinzer, J. A., Olesik, S. V., Anal. Chem.
1995, 67, 1–12.
[4] Tanaka, Y., Otsuka, K., Terabe, S., J. Pharm. Biomed.
Anal. 2003, 30, 1889–1895.
[5] Mol, R., de Jong, G. J., Somsen, G. W., Electrophoresis
2005, 26, 146–154.
[6] Ohnesorge, J., Neusu ¨X, C., Wa ¨tzig, H., Electrophoresis
2005, 26, 3973–3987.
[7] Smyth, W. F., Electrophoresis 2005, 26, 1334–1357.
[8] Klampfl, C. W., Electrophoresis 2006, 27, 3–34.
[9] Monton, M. R. N., Soga, T., J. Chromatogr. A 2007, 1168,
237–246.
[10] Scriba, G. K. E., J. Chromatogr. A 2007, 1159, 28–41.
[11] Zamfir, A. D., J. Chromatogr. A 2007, 1159, 2–13.
[12] Herrero, M., Iban ˜ez, E., Cifuentes, A., Electrophoresis
2008, 29, 2148–2160.
[13] Maxwell, E. J., Chen, D. D. Y., Anal. Chim. Acta 2008,
627, 25–33.
[14] Mischak, H., Coon, J. J., Novak, J., Weissinger, E. M.,
Schanstra, J. P., Dominiczak, A. F., Mass Spectrom. Rev.
2009, 28, 703–724.
[15] Nesbitt, C. A., Zhang, H., Yeung, K. K. C., Anal. Chim.
Acta 2008, 627, 3–24.
[16] Pantu ˚c ˇkova ´, P., Gebauer, P., Boc ˇek, P., Kr ˇiva ´nkova ´, L.,
Electrophoresis 2009, 30, 203–214.
[17] Ramautar, R., Somsen, G. W., de Jong, G. J., Electro-
phoresis 2009, 30, 276–291.
[18] Staub, A., Schappler, J., Rudaz, S., Veuthey, J. L.,
Electrophoresis 2009, 30, 1610–1623.
[19] Simpson, D. C., Smith, R. D., Electrophoresis 2005, 26,
1291–1305.
[20] Gaspar, A., Englmann, M., Fekete, A., Harir, M., Schmitt-
Kopplin, P., Electrophoresis 2008, 29, 66–79.
[21] Klampfl, C. W., Electrophoresis 2009, 30, S83–S91.
[22] Kolch, W., Neusu ¨X, C., Pelzing, M., Mischak, H., Mass.
Spectrom. Rev. 2005, 24, 959–977.
[23] Andrade, F., Shelley, J., Wetzel, W., Webb, M., Gamez,
G., Ray, S., Hieftje, G., Anal. Chem. 2008, 80, 2646–2653.
[24] Andrade, F., Shelley, J., Wetzel, W., Webb, M., Gamez,
G., Ray, S., Hieftje, G., Anal. Chem. 2008, 80, 2654–2663.
[25] Roth, J. R., Industrial Plasma Engineering: Volume 1,
Principles. IOP Publishing, Bristol, UK 1995.
Electrophoresis 2010, 31, 3597–3605
3604
M. C. Jecklin et al.
& 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.electrophoresis-journal.com

Page 9

[26] Andrade, F. J., Wetzel, W. C., Chan, G. C. Y., Webb,
M. R., Gamez, G., Ray, S. J., Hieftje, G. M., J. Anal. At.
Spectrom. 2006, 21, 1175–1184.
[27] Jecklin, M. C., Gamez, G., Touboul, D., Zenobi, R., Rapid
Commun. Mass Spectrom. 2008, 22, 2791–2798.
[28] Jecklin, M. C., Gamez, G., Zenobi, R., Analyst 2009, 134,
1629–1636.
[29] Amantonico, A., Urban, P. L., Zenobi, R., Analyst 2009,
134, 1536–1540.
[30] Gibson, G. T. T., Mugo, S. M., Oleschuk, R. D., Mass
Spectrom. Rev. 2009, 28, 918–936.
[31] Wu, Y. T., Chen, Y. C., Anal. Chem. 2005, 77, 2071–2077.
[32] Edwards, J. L., Chisolm, C. N., Shackman, J. G.,
Kennedy, R. T., J. Chromatogr. A 2006, 1106, 80–88.
[33] Janini, G. M., Zhou, M., Yu, L. R., Blonder, J., Gignac,
M., Conrads, T. P., Issaq, H. J., Veenstra, T. D., Anal.
Chem. 2003, 75, 5984–5993.
[34] Shamsi, S. A., Miller, B. E., Electrophoresis 2004, 25,
3927–3961.
[35] Hayen, H., Michels, A., Franzke, J., Anal. Chem. 2009, 81,
10239–10245.
[36] Nilsson, S. L., Andersson, C., Sjoberg, P. J. R., Bylund,
D., Petersson, P., Jornten-Karlsson, M., Markides, K. E.,
Rapid Commun. Mass Spectrom. 2003, 17, 2267–2272.
[37] Ouyang, Z., Cooks, R. G., Annu. Rev. Anal. Chem. 2009,
2, 187–214.
Electrophoresis 2010, 31, 3597–3605
Microfluidics and Miniaturization
3605
& 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.electrophoresis-journal.com