Electrospray-assisted laser desorption/ionization mass spectrometry for continuously monitoring the states of ongoing chemical reactions in organic or aqueous solution under ambient conditions

Article abstract of DOI:10.1021/ac800952e

Electrospray-assisted laser desorption/ionization (ELDI) combined with mass spectrometry allows chemical and biochemical compounds to be characterized directly from hydrophilic and hydrophobic organic solutions mixed with carbon powders under ambient conditions. Organic and inorganic compounds dissolved in polar or nonpolar solvent such as methanol, tetrahydrofuran, ethyl acetate, toluene, dichloromethane, or hexane can be detected using this ambient ionization technique without prior pretreatment. We have used this technique to monitor the progress in several ongoing reactions: the epoxidation of chalcone in ethanol, the chelation of ethylenediaminetetraacetic acid with copper and nickel ions in aqueous solution, the chelation of 1,10-phenanthroline with iron(II) in methanol, and the tryptic digestion of cytochrome c in aqueous solution. Liquid-ELDI analyses simply require irradiation of the surface of the sample solution with a pulsed ultraviolet laser; the laser energy is adsorbed by the carbon powder presuspended in the sample solution; the absorbed laser energy is then transferred to the surrounding solvent and to the analyte molecules in the solution, leading to their desorption; the desorbed gaseous analyte molecules are then postionized within an electrospray (ESI) plume to generate ESI-like analyte ions.

Full text of DOI:10.1021/ac800952e

Anal. Chem. 2008, 80, 7699–7705
Electrospray-Assisted Laser Desorption/Ionization
Mass Spectrometry for Continuously Monitoring
the States of Ongoing Chemical Reactions in
Organic or Aqueous Solution under Ambient
Conditions
Chi-Yuan Cheng,† Cheng-Hui Yuan,† Sy-Chyi Cheng,† Min-Zong Huang,† Hui-Chiu Chang,‡,§
Tien-Lu Cheng,‡,| Chen-Sheng Yeh, and Jentaie Shiea*,†,‡
Department of Chemistry, National Sun Yat-Sen University, Kaohsiung, Taiwan, National Sun Yat-Sen
UniversitysKaohsiung Medical University Joint Research Center, Kaohsiung, Taiwan, Graduate Institute of Medicine,
and Department of Biomedical Science and Environmental Biology, Kaohsiung Medical University, Kaohsiung,
Taiwan, and Department of Chemistry, National Cheng-Kung University, Tainan, Taiwan
Electrospray-assisted laser desorption/ionization (ELDI)
combined with mass spectrometry allows chemical and
biochemical compounds to be characterized directly from
hydrophilic and hydrophobic organic solutions mixed with
carbon powders under ambient conditions. Organic and
inorganic compounds dissolved in polar or nonpolar
solvent such as methanol, tetrahydrofuran, ethyl acetate,
toluene, dichloromethane, or hexane can be detected
using this ambient ionization technique without prior
pretreatment. We have used this technique to monitor the
progress in several ongoing reactions: the epoxidation of
chalcone in ethanol, the chelation of ethylenediaminetet-
raacetic acid with copper and nickel ions in aqueous
solution, the chelation of 1,10-phenanthroline with iro-
n(II) in methanol, and the tryptic digestion of cytochrome
c in aqueous solution. Liquid-ELDI analyses simply
require irradiation of the surface of the sample solution
with a pulsed ultraviolet laser; the laser energy is ad-
sorbed by the carbon powder presuspended in the sample
solution; the absorbed laser energy is then transferred to
the surrounding solvent and to the analyte molecules in
the solution, leading to their desorption; the desorbed
gaseous analyte molecules are then postionized within an
electrospray (ESI) plume to generate ESI-like analyte ions.
organic solvents for most abiotic reactions), it is necessary for
any developed technique to respond rapidly to changes in the
states of the chemicals (including reactants, intermediates, and
products) present in solution. Other preferable requirements for
such an analytical technique include (1) operation under ambient
conditions, (2) simultaneous response to multiple components,
(3) no sample pretreatment, and (4) no response to solid materials,
such as catalysts, in the sample solution. Most previously existing
techniques are based on measuring changes in the spectroscopic
or electrochemical characteristics of the chemical compounds
involved in the reaction.1-4 One of the disadvantages of using
these techniques is that structural characterization of the indi-
vidual compounds involved in the reaction is nearly impossible,
especially when the solution comprises mixtures of organic or
biological compounds.
Although mass spectrometry is an extremely useful technique
for characterizing small organic and large biochemical compounds,
several sample pretreatment steps are usually required to ef-
fectively ionize them: heating or drying a sample solution for the
analyte to be ionized in the gaseous state under vacuum (e.g.,
electron impact ionization and chemical ionization (;5-8 dissolving
the analyte in a suitable organic solvent or acidic solution to
generate ions under a high electric field or through heating [e.g.,
electrospray ionization (ESI) and atmospheric pressure chemical
ionization];9-12 or mixing the analyte with a matrix to increase
The development of analytical techniques for continuous
monitoring of the states of ongoing chemical and biochemical
reactions occurring in various solvents has been a formidable
challenge for analytical chemists. Because most reactions occur
in liquid phases (aqueous solutions for most biochemical reactions,
(1) Wu, Q.; Yang, S.; Luo, J.; Huang, W. Spectrosc. Spectral Anal. 2000, 20,
606–609
(2) Anderson, J. A.; Fernandez-Garcia, M. Chem. Eng. Res. Des. 2000, 78, 935–
942
(3) Vidu, R.; Quinlan, F. T.; Stroeve, P. Ind. Eng. Chem. Res. 2002, 41, 6546–
6554
(4) Blanco, M.; Castillo, M.; Beneyto, R.; Porcel, M. Spectrosc. Lett. 2005, 38,
825–837
(5) Fountain, S. T.; Lubman, D. M. Anal. Chem. 1993, 65, 1257–1266
(6) Drewnich, F.; Wieser, P. H. Rev. Sci. Instrum. 2002, 73, 3003–3006
(7) Cotter, R. J. Anal. Chem. 1980, 52, 1767–1767
(8) Perchalski, R. J.; Yost, R. A.; Wilder, B. J. Anal. Chem. 1983, 55, 2002–
2005
(9) Yamashita, M.; Fenn, J. B. J. Phys. Chem. 1984, 88, 4451–4459
(10) Fenn, J. B.; Mann, M.; Meng, C. K.; Wong, S. F.; Whitehouse, C. M. Science
1989, 246, 64–71
.
.
.
* To whom correspondence should be addressed. E-mail: jetea@
mail.nsysu.edu.tw.
.
.
† National Sun Yat-Sen University.
.
‡ National Sun Yat-Sen UniversitysKaohsiung Medical University Joint
Research Center.
.
§ Graduate Institute of Medicine, Kaohsiung Medical University.
| Department of Biomedical Science and Environmental Biology, Kaohsiung
Medical University.
.
.
National Cheng-Kung University.
.
10.1021/ac800952e CCC: $40.75 2008 American Chemical Society
Published on Web 09/20/2008
Analytical Chemistry, Vol. 80, No. 20, October 15, 2008 7699

the efficiency of desorption and ionization (e.g., fast atom
bombardment and matrix-assisted laser desorption/ionization).13-16
These pretreatment processes make it extremely difficult to use
mass spectrometry to continuously monitor the ongoing state of
a chemical reaction under ambient conditions.
Recently, the development of ambient ionization methods [e.g.,
electrospray laser desorption/ionization (ELDI) and desorption
electrospray ionization (DESI)] has made it unnecessary to heat
the sample solution or mix it with a matrix; nevertheless, only
solid samples are suitable for such analysis, meaning that drying
of the sample remains inevitable prior to ionization.17-22 An ideal
approach toward continuously monitoring the stages of ongoing
reaction would be to desorb and ionize the analyte molecules
directly from the solutions without drying the sample. Herein, we
report a modified ELDI techniquesliquid ELDIsthat allows
analyte ions to be generated directly from organic solvents and
aqueous solutions.21-26
In essence, the liquid ELDI technique presented herein is a
combination of two ionization methods: ELDI and surface-assisted
laser desorption/ionization (SALDI) operated under ambient
conditions.20-22,27-30 The possible mechanisms of ELDI and
SALDI have been discussed previously.20,21,27 Recently, the
principle of ELDI has been adapted in several techniques,
including matrix-assisted laser desorption/ionization electrospray
ionization and laser ablation electrospray ionization.31-33 In SALDI,
a sample solution is mixed with an “inorganic liquid matrix”
comprising a suspension of micrometer-sized carbon powder in
a mixture of glycerol, sucrose, and methanol.27,28 Because SALDI
is performed under vacuum, the sample solution must be mixed
with a viscous liquid matrix, such as these glycerol/sucrose
solutions, to prevent it from drying in the source. It has been
proposed that, in SALDI, the carbon powder is the energy-transfer
medium coupling the laser UV energy into the molecules in the
viscous liquid solution under vacuum.27,28 For liquid ELDI, the
laser energy is first adsorbed by the fine carbon particles
suspended in the sample solution; the energy is subsequently
transferred to the solvent and the analyte molecules in the sample
solution; the desorbed analyte molecules are then postionized
through ESI processes. In this study, we applied liquid ELDI
combined with mass spectrometry to monitor the progress of
several ongoing chemical and biochemical reactions in different
solution.
EXPERIMENTAL SECTION
Hemin chloride, cytochrome c, trypsin, 1,10-phenanthroline
(PA), nickel chloride, copper sulfate, iron(II) diammonium dis-
ulfate, and ethylenediaminetetraacetic acid (EDTA), and the
organic solvents (HPLC grade) were purchased from Sigma or
Aldrich (Milkaukee, WI) and used without further purification.
Carbon graphite powder was purchased from Merck (Darmstadt,
Germany). The synthesis of magnetic nanoparticles coated with
trypsin has been described previously.34,35
The sample preparation process for liquid ELDI-MS analysis
is extremely simple: merely suspend a small amount of the fine
carbon powder (4 mg/mL) in the sample solution and then subject
it to analysis without any further sample pretreatment. Commonly,
a small amount of the sample solution (5-10 µL) deposited on
an acrylic sample plate [5 (L) × 2 (W) cm] using a micropipet
was sufficient for complete liquid ELDI analysis. The sample plate
was positioned on an acrylic plate placed on a XYZ-stage, which
was set in front of the sampling capillary (inlet) of an ion trap
mass spectrometer. The surface of the sample solution was then
exposed to a pulsed nitrogen laser operated at a wavelength of
337 nm (Physical Science), a frequency of 10 Hz [controlled by a
sweep function generator (model FG-101A, Kowa, Taiwan)], a
pulsed energy of ∼150 µJ, and length of 4 ns. The laser beam
(spot size: ∼100 µm × 150 µm) was focused through an objective
lens. The strongest ion signal (using cytochrome c as the standard
for evaluation) was obtained at an incident laser angle of ∼45°
and a focal length of ∼25 cm.
(11) Fenn, J. B.; Mann, M.; Meng, C. K.; Wong, S. F.; Whitehouse, C. M. Mass
Spectrom. Rev. 1990, 9, 37–70
(12) Coon, J. J.; Harrison, W. W. Anal. Chem. 2002, 74, 5600–5605.
(13) Karas, M.; Hillenkamp, F. Anal. Chem. 1988, 60, 2299–2301
(14) Tanaka, K. Rapid Commun. Mass Spectrom. 1988, 2, 151–153
(15) Hillenkamp, F.; Karas, M.; Beavis, R. C.; Chait, B. T. Anal. Chem. 1991,
63, 1193A–1203A
(16) Barber, M.; Bordoli, R. S.; Sedgwick, R. D.; Tyler, A. N. Nature 1981,
293, 270–275
(17) Takats, Z.; Wiseman, J. M.; Gologan, B.; Cooks, R. G. Science 2004, 306,
471–473
(18) Wiseman, J. M.; Puolitaival, S. M.; Takats, Z.; Cooks, R. G.; Caprioli, R. M.
Angew. Chem., Int. Ed. 2005, 44, 7094–7097
(19) Cooks, R. G.; Ouyang, Z.; Takats, Z.; Wiseman, J. M. Science 2006, 311,
1566–1570
(20) Shiea, J.; Huang, M.-Z.; Hsu, H.-J.; Lee, C.-Y.; Yuan, C.-Y.; Beech, I.; Sunner,
J. Rapid Commun. Mass Spectrom. 2005, 19, 3701–3704
(21) Huang, M. Z.; Hsu, H. J.; Lee, J.; Jeng, J.; Shiea, J. J. Proteome Res. 2006,
5, 1107–1116
(22) Huang, M. Z.; Hsu, H. J.; Wu, C. I.; Lin, S. Y.; Ma, Y. L.; Cheng, T. L.;
Shiea, J. Rapid Commun. Mass Spectrom. 2007, 21, 1767–1775
.
.
.
.
.
.
.
.
.
.
The configuration of the ESI source for liquid ELDI was similar
to those used in solid ELDI, pyrolysis/ESI, and fused-droplet ESI
sources.20-22,36-39 The capillary electrosprayer was aligned paral-
lel to, and ∼3.5 mm above, the acrylic sample plate. The ESI plume
was directed toward the ion sampling orifice of the mass analyzer
(i.e., parallel to the sample plate). The electrospray needle and
the sample plate potentials were both held at 0 V (grounded);
.
(23) Shiea, J.; Yuan, C. H.; Huang, M. Z. 5th HUPO Annual World Conference,
Long Beach, CA, Oct. 28-Nov. 1, 2006; Poster W199.
(24) Yuan, C. H.; Hung, W. C.; Chang, H. C.; Lin, H. Y.; Liu, J. J.; Shiea, J. 5th
HUPO Annual World Conference, Long Beach, CA, Oct. 28-Nov. 1, 2006;
Poster W 203.
(25) Jeng, J.; Yuan, C. H.; Cheng C. Y.; J. Shiea. 55th ASMS Conference on
Mass Spectrometry, Indianapolis, IN, June 3-7, 2007; Poster TPA 10.
(26) Yuan, C. H.; Ma, Y. L.; Shiea, J. 55th ASMS Conference on Mass
Spectrometry, Indianapolis, IN, June 3-7, 2007; Poster TPA 18.
(34) Jeng, J.; Lin, M.; Cheng, F.; Yeh, C.; Shiea, J. Rapid Commun. Mass Spectrom.
(27) Sunner, J.; Dratz, E.; Chen, Y. Anal. Chem. 1995, 67, 4335–4342
(28) Kraft, P.; Alimpiev, S.; Dratz, E.; Sunner, J. J. Am. Soc. Mass Spectrom. 1998,
9, 912–924
(29) Chen, Y.; Shiea, J.; Sunner, J. J. Chromatogr., A 1998, 826, 77–86
(30) Han, M.; Sunner, J. J. Am. Soc. Mass Spectrom. 2000, 11, 644–649
(31) Sampson, J. S.; Hawkridge, A. M.; Muddiman, D. C. J. Am. Soc. Mass
Spectrom. 2006, 17, 1712–1716
(32) Sampson, J. S.; Hawkridge, A. M.; Muddiman, D. C. Rapid Commun. Mass
Spectrom. 2007, 21, 1150–1154
(33) Nemes, P.; Vertes, A. Anal. Chem. 2007, 79, 8098–8106
.
2007, 21, 3060–3068
(35) Chang, F.; Yang, Y.; Huang, W.; Yeh, C.; Tsai, C.; Wu, C.; Shieh, D.
Biomaterials 2005, 26, 729–738
(36) Hsu, H. J.; Kuo, T. L.; Wu, S. H.; Oung, J. N.; Shiea, J. Anal. Chem. 2005,
77, 7744–7749
(37) Shiea, J.; Wang, C. H. J. Mass Spectrom. 1997, 32, 247–250
(38) Shiea, J.; Chang, D. Y.; Lin, C. H.; Jiang, S. J. Anal. Chem. 2001, 73, 4983–
4987
(39) Shiea, J.; Lee, C. C.; Chang, D. Y.; Jeng, J. J. Mass Spectrom. 2002, 37,
115–117
.
.
.
.
.
.
.
.
.
.
.
.
7700 Analytical Chemistry, Vol. 80, No. 20, October 15, 2008

the sampling cone voltage in the ion trap mass spectrometer was
maintained at -4.5 kV. The electrospray solutions comprised
methanol/water mixtures with or without acetic acid; a typical
composition of the ESI solution used in the ELDI analysis was
80% MeOH/H2O. The ESI solution was delivered through a
capillary by a syringe pump at a flow rate of 2 µL/min. A
nebulizing gas, commonly used in conventional ESI sources, was
not used during ELDI. The ions generated in this manner were
further analyzed using a quadrupole/time-of-flight (Bruker Dal-
tonic, Bio-TOF-q, Billerica, MA) or an ion trap (Bruker Daltonics,
Esquire 3000 plus) mass analyzer. The mass spectra were
recorded at a scan rate of ∼2 s/scan.
In this study, four chemical reactions were performed in a
range of organic solvents in glass beakers, with ELDI-MS used
to continuously monitor the extent of each chemical’s ongoing
transformation. To continuously monitor the ongoing state of the
reactions 1-3 (see the Experimental Section below), the surface
of the solution containing reactants and carbon powder in a beaker
was continuously irradiated with an UV laser beam. The laser-
ablated molecules subsequently entered into an electrospray
plume and were ionized there. For reaction 4, every minute, a
drop of the solution was withdrawn (through a dropper), deposited
on the acrylic sample plate, and rapidly analyzed using liquid
ELDI-MS.
Reaction 1. Epoxidation of 1,3-Diphenyl-1-propen-3-one
(Chalcone) in Ethanol.40,41 An ethanol solution (3 mL) contain-
ing chalcone (10-3 M) and carbon powder (4 mg/mL) was stirred
for 1.7 min and then the catalyst, H2O2 (30% aqueous solution;
200 µL) was added. Because this catalyst functions only under
basic conditions, 1 M NaOH (200 µL) was added to catalyze the
peroxidation reaction. The solution was continuously stirred with
a magnetic bar while adding the reagent and catalyst solution.
Reaction 2. Chelation of EDTA with Copper and Nickel
Ions in Aqueous Solution.42 An aqueous solution (3 mL) of
EDTA (1.5 × 10-3 M) and carbon powder (4 mg/mL) in a beaker
was stirred continuously with a magnetic bar while CuSO4 solution
(1.5 × 10-2 M, 100 µL) was added. After 20 s, a NiCl2 solution
(1.5 × 10-2 M, 100 µL) was added to the mixture.
Figure 1. Liquid ELDI system. (a) Schematic representation of the
desorption and ionization of molecules dissolved in liquid during liquid
ELDI. a, ESI capillary; b, Taylor cone; c, charged ESI droplet; d, laser
beam; e, analyte ions; f, desorbed analyte; g, glass tube; h, sample
plate; i, carbon powder; j, stirring bar; k, magnetic stirrer. (b, c)
Photographic images of the liquid ELDI system displaying (b) the
electrospray plume and (c) the mixing of the analyte droplets
(generated through laser beam irradiation of a methanol solution
containing carbon powder) with the electrospray plume.
ELDI technique is performed under ambient conditions; no
viscous liquid matrix is necessary: the aqueous sample solution
itself is used as the medium for transferring the adsorbed laser
energy from the carbon particles to the analyte molecules. This
feature greatly simplifies the sample pretreatment process.
Panels b and c in Figure 1 display photographic images of the
electrospray plume only (laser off) and the mixing of the analyte
droplets (generated by irradiating the sample solution containing
carbon powder with an UV laser beam) with the electrospray
plume, respectively, in the liquid ELDI system. We found that, as
the surface of the sample solution was continuously irradiated with
the UV laser beam, a stable fine mist was generated from the
surface of the sample solution. The mist moved upward initially
and then turned 90° toward the inlet of the mass analyzer as it
joined the ESI plume.
Direct analysis of a hemin solution (in methanol) without
carbon powder revealed that the solution was transparent to the
UV laser light: no hemin signal (m/z 616) was detected (Figure
2a). Only after we had added 4 mg/mL or more of the carbon
powder into the sample solution did we observe the molecular
ion of hemin (Figure 2b). The intensity of the analyte ion signals
increased upon increasing the amount of carbon powder added
to the sample solution, reaching a plateau at concentrations of
carbon powder of 8 mg/mL and higher (Figure 2c).
Reaction 3. Chelation of 1,10-Phenanthroline with Iro-
n(II) in Methanol.43 A methanol solution (3 mL) containing PA
(1.5 × 10-3 M) and carbon powder (4 mg/mL) was stirred
continuously with a magnetic bar while aqueous (NH4)2Fe(SO4)2
solution (2.5 × 10-2 M, 100 µL) was added.
Reaction 4. Tryptic Digestion of Cytochrome c in Aqueous
Solution. A solution of cytochrome c (10-4 M) and fine carbon
powder (4 mg/mL) was mixed with a solution of magnetic
nanoparticles coated with trypsin (2.5 µg/µL), and the solution
was continuously stirred with a magnetic bar.
RESULTS AND DISCUSSION
Figure 1a illustrates the possible desorption/ionization pro-
cesses occurring during liquid ELDI. Unlike SALDI, the liquid
The detection of the hemin ion signal directly from the
methanol solution when carbon powder was present indicates that
the laser energy was adsorbed by the carbon powder and
transferred to the surrounding methanol and hemin molecules,
leading to their desorption. It seems that the liquid that dissolves
the analyte plays no role in desorption or ionization of the
analytesapart from being used as the energy-transfer medium.
Therefore, chemical compounds dissolved in the liquid medium,
(40) Bougauchi, M.; Watanabe, S.; Arai, T.; Sasai, H.; Shibasaki, M. J. Am. Chem.
Soc. 1997, 119, 2329–2330
(41) Berkessel, A.; Gasch, N.; Glaubitz, K.; Koch, C. Org. Lett. 2001, 3, 3839–
3842
(42) Reilley, C. N.; Schmid, R. W. Anal. Chem. 1958, 30, 947–953
.
.
.
(43) Harris, D. C. Quantitative Chemical Analysis, 6th ed.; W.H. Freeman and
Co.: New York, 2003.
Analytical Chemistry, Vol. 80, No. 20, October 15, 2008 7701

Figure 2. Positive liquid ELDI mass spectra of hemin (m/z 616, M+) dissolved in methanol (MeOH) (10-4 M, 10 µL) containing carbon powder
at concentrations of (a) 0, (b) 4, and (c) 8 mg/mL.
Figure 3. Liquid ELDI spectra of small molecules. Positive-ion liquid ELDI mass spectra of (a) hemin (m/z 616, M+) dissolved in methanol
(MeOH), (b) 18crown-6 ether (m/z 287 [MNa+] and m/z 303 [MK+]) dissolved in THF, (c) 1-hexadecylamine (m/z 242 [MH+]) dissolved in EA,
(d) methyl(triphenylphosphoranylidene) acetate (m/z 335 [MH+]) dissolved in dichloromethane, (e) cinnamic acid benzyl ester (m/z 261 [MNa+])
dissolved in toluene, (f) cetylpyridium chloride (m/z 304, [M - Cl]+) dissolved in hexane, (g) C40H28IrN4 ·PF6 (m/z 757 [M - PF6]+) dissolved in
dichloromethane, and (h) C33H26F4IrN6 ·PF6 (m/z 775 [M - PF6]+) dissolved in dichloromethane. The insets to (g) and (h) display the isotope
patterns of the molecular ions containing Ir.
which may include hydrophilic or hydrophobic and volatile or
nonvolatile organic solvents, can be characterized using ELDI.
To test this hypothesis, we performed liquid ELDI analyses of
six organic compounds dissolved in a range of polar and nonpolar
organic solvents, namely, MeOH, tetrahydrofuran (THF), ethyl
acetate (EA), dichloromethane, toluene, and hexane. Pure metha-
nol was used for the electrospray to ionize the molecules desorbed
by the UV laser. As expected, we detected the molecular ions of
hemin (M+, m/z 616),18crown-6 (MNa+, m/z 287; MK+, m/z 303),
1-hexadecylamine (MH+, m/z 242), methyl(triphenylphosphora-
7702 Analytical Chemistry, Vol. 80, No. 20, October 15, 2008

Figure 4. Using liquid ELDI to continuously monitor the state of the epoxidation of chalcone in ethanol. Positive-ion liquid ELDI mass spectra
were recorded at three different stages: (a) immediately after H2O2 had been added into the sample solution and (b) immediately and (c) 1 min
after the sample solution had been changed to basic through the addition of 1 M NaOH. (d) Extract ion chromatograms of the ions at m/z 209
([M1H]+ for the reactant chalcone) and 247 ([M2Na]+ for the product).
nylidene) acetate (MH+, m/z 335), cinnamic acid benzyl ester
(MNa+, m/z 261), and cetylpyridium chloride ([M - Cl]+, m/z
304)(Figure 3a-f). To test whether compounds having higher
molecular weights could also be detected, we synthesized two
inorganic compounds containing iridiumsC40H28IrN4 · PF6 ([M -
PF6]+, m/z 757) and C33H26F4IrN6 · PF6 ([M - PF6]+, m/z
775)sand subjected them to liquid ELDI analysis. Again, we
detected the molecular ions of both compounds from their
CH2Cl2/carbon powder solutions. The insets to Figure 3g and h
clearly display the isotope patterns expected for these two
molecular ions.
We expected, therefore, that liquid ELDI could be used to
continuously monitor many of the chemical reactions that occur
in organic solvents under ambient conditions. Figure 4 displays
the results obtained when using liquid ELDI-MS to monitor the
epoxidation of chalcone (1,3-diphenyl-1-propen-3-one).34,35 The
reaction was performed in a small glass beaker using an ethanol
solution containing carbon powder. We found that, even through
the chalcone solution was stirred with a magnetic bar, a stable
total ion current from the reactant ions (chalcone; m/z 209 for
M1H+; m/z 231 for M1Na+, Figure 4d) was obtained when the
surface was irradiated continuously with the UV laser pulse. The
catalyst, H2O2, was added into the solution 1.7 min after the onset
of stirring, resulting in the generation of gas bubbles and an abrupt
increase in the intensity of the reactant ion (m/z 209; Figure 4a
and d). Because the catalyst functions only under basic conditions,
we did not detect any ion signal from products1,3-diphenyl-1,2-
epoxy- propan-3-one (M2Na+: m/z 249; see Figure 4a and d; M2,
epoxided chalcone molecule) at this stage. Only after making the
solution basic, by adding NaOH solution (1 M, 200 µL) 2.4 min
after the onset of stirring, did we detect the ion of the reactant
(Figure 4c and d). The decrease in intensity of the ion signal of
the reactant coincided with the increased intensity of the product.
Figure 4d indicates clearly that the state of an ongoing reaction
can be evaluated continuously by monitoring the changes in the
intensities of the ion signals of the reactant (M1) and product (M2)
over time. However, due to certain mixing and diffusion time is
required to bring the reactant, reagent, and catalyst together, the
ion signal of reactant did not fall abruptly as the product signal
rises.
Figure 5 displays the results obtained when using ELDI-
MS to monitor the chelation of EDTA with copper and nickel
ions in aqueous solution. A glass beaker was charged with a
solution (3 mL) of EDTA (1.5 × 10-3 M) and carbon powder
(4 mg/mL). For ELDI analysis, the solution was stirred while
the surface was irradiated continuously with the UV laser pulse;
a stable extract ion current was obtained from the negatively
charged reactant ions (EDTA; m/z 291 for [M - H]-; Figure
Analytical Chemistry, Vol. 80, No. 20, October 15, 2008 7703

Figure 5. Continuously monitoring the state of the chelation reactions of EDTA with copper and nickel ions in aqueous solution. (b) Extract ion
chromatograms of the reactant at m/z 291 ([EDTA - H]-) and the product at m/z 352 ([EDTA + Cu - H]-) and 347 ([EDTA + Ni - H]-).
Negative-ion liquid ELDI mass spectra were recorded at three different stages: (c) EDTA solution only, (d) 10 s after adding CuSO4 (1.5 × 10-2
M, 100 µL) solution, and (e) 20 s after adding NiCl2 (1.5 × 10-2 M, 100 µL) solution.
Figure 6. Continuously monitoring the state of the chelation reaction of PA with iron(II) in methanol. (a) Extract ion chromatograms of the
reactant at m/z 181 ([PA)H]+) and product at m/z 298 ([(PA)3 + Fe]2+). Positive-ion liquid ELDI mass spectra were recorded at three different
stages: (d) PA solution only, (e) 6 s after adding (NH4)2Fe(SO4)2 (2.4 × 10-2 M, 100 µL) solution, and (f) 30 s after adding (NH4)2Fe(SO4)2
solution. The reaction was performed using a buffer solution at pH 10. The inset to (f) displays the isotope pattern of the molecular ion of [(PA)3
+ Fe]2+
.
5b). The ELDI mass spectrum recorded from the aqueous
solution featured the [M - H]- ion of EDTA predominantly
(Figure 5c). Immediately after CuSO4 solution was added into
the EDTA solution, a stable signal for the EDTA · Cu complex
ion (m/z 352) was detected (Figure 5b). The ELDI mass
spectrum featured the [EDTA]- (m/z 291) and (EDTA · Cu)-
(m/z 352) ions predominantly (Figure 5d). A NiCl2 solution
was subsequently added into the solution now containing
EDTA, EDTA · Cu, and Cu2+. We observed a stable signal for
the [EDTA · Ni]- ion (m/z 347). The ELDI mass spectra
featured the [EDTA · Cu]- and [EDTA · Ni]- ions predominantly
(Figure 5e). The results in Figure 5 indicate clearly that the
states of ongoing EDTA-copper and EDTA-nickel chelation
reactions can be evaluated continuously by monitoring the
changes in the intensities of the signals of the [EDTA · Cu]-
and [EDTA · Ni]- ions over the time.
We also used liquid ELDI/MS to monitor the reaction of PA
with iron(II) in methanol. We recorded positive-ion ELDI mass
spectra as the chelating reaction proceeded in a methanol/carbon
powder solution. Panels b and c in Figure 6 present extract ion
chromatograms of the reactant at m/z 181 [(PA)H]+ and the
product at m/z 298 [(PA)3 + Fe]2+. The mass spectra in Figure
7704 Analytical Chemistry, Vol. 80, No. 20, October 15, 2008

Figure 7. Using liquid ELDI to monitor the state of a tryptic digestion. Positive-ion liquid ELDI mass spectra recorded from sample droplets
collected (a) 0, (b) 15, and (c) 30 min after mixing a solution containing trypsin-coated magnetic nanoparticles with a cytochrome c/carbon
powder solution. b, Multiply charged cytochrome c ions; O, peptide ions from cytochrome c.
6d-f reveal three different stages of the reaction: (d) the PA
solution only and (e, f) the mixtures obtained (e) 6 s and (f) 20 s
after adding (NH4)2Fe(SO4)2 into the PA solution. The inset to
Figure 6f displays the isotope pattern of the molecular ion of [(PA)3
CONCLUSION
We have demonstrated that ELDI can be used to characterize
organic, inorganic, and biological compounds directly from liquids
(including aqueous solution and polar and nonpolar volatile
organic solvents) under ambient conditions. The presence of fine
carbon particles in the sample solution assists the desorption of
the analyte molecules during irradiation of the solution surface
with a pulsed laser beam. The states of ongoing biochemical and
organic reactions occurring in the solution can, therefore, be
monitored continuously. Because this liquid ELDI technique
utilizes an extremely simple sample pretreatment procedure and
provides rapid analyses under ambient conditions, we hope that
it will allow physical organic chemists and biochemists to monitor
a wide variety of reactions through continuous monitoring of the
changing ion intensities of their reactants, intermediates, and
products. Although it is not demonstrated in this paper, obtaining
useful kinetic information about a typical chemical reaction is
possible if suitable internal standards are used for quantification
of the reactants and products and the lifetime of the chemicals
(especially for intermediates) involved in the reaction is long
enough to be detected.
+ Fe]2+
.
The progress of ongoing biochemical reactionssin this case,
the tryptic digestion of cytochrome c in aqueous solutionscan
also be monitored using liquid ELDI. In this example, we mixed
an aqueous solution containing a cytochrome c standard (10-4
M) with the fine carbon powder. Magnetic nanoparticles coated
with trypsin were then added into the solution to digest the protein
in the solution.36,37 Every minute, a drop of the solution was
withdrawn (through a dropper), deposited on the acrylic sample
plate, and rapidly analyzed using ELDI. Figure 7 displays
representative ELDI mass spectra obtained from sample droplets
collected after 0, 15, and 30 min, respectively. We observe a
gradual increase in the peptide ion signals upon increasing the
digestion time. Interestingly, as the reaction proceeded, the charge
state of cytochrome c ion was changed to lower number (from
+9 to +7). It seems that the presence of the peptides originated
from cytochrome c in the sample solution will deprotonate the
cytochrome c ion during ionization. The ELDI mass spectra
display no interference from any ions derived from trypsin; this
phenomenon is probably due to the magnetic nanoparticle-bound
trypsin not being desorbed by the laser pulse or ionized in the
ESI plume during ELDI.
Received for review May 8, 2008. Accepted August 13,
2008.
AC800952E
Analytical Chemistry, Vol. 80, No. 20, October 15, 2008 7705