ESI-MS Detection of Very Weak π-Stacking Interactions in the Mixed-Ligand Sandwich Complexes Formed by Substituted Benzo-Crown Ethers and Metal Cations

Article abstract of DOI:10.1016/j.jasms.2009.12.018

The stability of the mixed-ligand complex formed by amino-benzo-15-crown-5 and nitro-benzo-15-crown-5, with a metal cation, [NH₂B15C5 + NO₂B15C5 + K]⁺, was found to be enhanced by π-stacking interactions. This conclusion was deduced by comparison of the abundance of the mixed-ligand complex with the abundances of homo-ligand complexes ([(NH₂B15C5)₂ + K]⁺, [(NO₂B15C5)₂ + K]⁺), as well as with those of 1:1 complexes ([NH₂B15C5 + K]⁺, [NO₂B15C5 + K]⁺). Some solvents and some metal cations with large radii were also found to prevent the existence of π-stacking interactions.

Full text of DOI:10.1016/j.jasms.2009.12.018

ESI-MS Detection of Very Weak ␲-Stacking
Interactions in the Mixed-Ligand Sandwich
Complexes Formed by Substituted
Benzo-Crown Ethers and Metal Cations
Rafał Fran´ski and Błaz˙ej Gierczyk
Faculty of Chemistry, Adam Mickiewicz University, Poznan´, Poland
The stability of the mixed-ligand complex formed by amino-benzo-15-crown-5 and nitro-benzo-
15-crown-5, with a metal cation, [NH2B15C5 ϩ NO2B15C5 ϩ K]ϩ, was found to be enhanced
by ␲-stacking interactions. This conclusion was deduced by comparison of the abundance of
the mixed-ligand complex with the abundances of homo-ligand complexes ([(NH2B15C5)2 ϩ
K]ϩ, [(NO2B15C5)2 ϩ K]ϩ), as well as with those of 1:1 complexes ([NH2B15C5 ϩ K]ϩ,
[NO2B15C5 ϩ K]ϩ). Some solvents and some metal cations with large radii were also found to
prevent the existence of ␲-stacking interactions. (J Am Soc Mass Spectrom 2010, 21,
545–549) © 2010 American Society for Mass Spectrometry
rown ethers are popular hosts, and their inclu-
sion complexes have found a vast number of
Experimental
The ESI mass spectra were obtained on a Waters/
Micromass (Manchester, UK) ZQ2000 mass spectrome-
ter (single quadrupole type instrument, Z-spray, soft-
ware MassLynx ver. 3.5). The sample solutions were
prepared in methanol (or other solvent), at crown ether
concentrations of 0.7 ϫ 10Ϫ5 mol/dm3 (or similar as
indicated). At this low concentration, the 2:1 complexes
were characterized by low abundances in comparison
to those of the 1:1 complexes; therefore the m/z range of
the 2:1 complexes is magnified 10 times in Figure 1.
Sherman et al. [7] used a concentration of 5 ϫ 10Ϫ5
mol/dm3; however, we employed a lower concentra-
tion to minimize contamination of the instrument.
Signals of [M ϩ Na]ϩ, [M ϩ K]ϩ, and [M2 ϩ Na]ϩ
were clearly seen (M ϭ crown ether molecule) without
adding inorganic salts (Figure 1). Thus, the concentra-
tions of Naϩ and Kϩ were not known. ESI mass spectra
were also obtained with KClO4, KCl, or NaCl added. To
observe complexes with rubidium or cesium, RbClO4 or
CsClO4 were added to the solutions analyzed (concen-
tration about 0.4 ϫ 10Ϫ5 mol/dm3). The experiments
were repeated at least three times. In the repeated
measurements, the mass spectrum was recorded for
both the previously used solution and for the newly
prepared solution.
C
practical applications [1]. Most studies devoted
to crown ether complexes with metal cations have
focused on complexes with 1:1 stoichiometry. However,
complexes of 2:1 stoichiometry (“sandwich” type com-
plexes) are also known. Complexes of this type are
formed when the radius of the metal cation is greater
than the size of the crown ether cavity [2]. A number of
sandwich type complexes have been successfully stud-
ied in detail in the gas phase [3–5]; some of them
contained ammonium cation instead of a metal cation
[6]. An interesting paper devoted to the sandwich type
complexes formed by crown ethers is that published by
Sherman et al. [7], who demonstrated that the stability
of mixed-ligand sandwich type complexes can be en-
hanced by ␲-stacking interactions between two aro-
matic rings, one of which is enriched in electrons, while
the other is poor in electrons; ␲-stacking interactions
have been also observed for some complexes of crown
ethers with substituted ammonium cation [8, 9]. En-
couraged by this paper [7], and taking into account the
importance of ␲-stacking interactions [10–12], we have
examined whether such interactions exist in the sand-
wich complexes formed by amino-benzo-15-crown-5
(NH2B15C5) and nitro-benzo-15-crown-5 (NO2B15C5).
Due to resonance effects, the aromatic ring of the first
compound is rich in electrons while the aromatic ring in
the latter compound is poor in electrons. For compari-
son benzo-15-crown-5 (B15C5) has been also included
in the study (Scheme 1).
The sample solutions were infused into the ESI
source using a Harvard pump at a flow rate of 80
␮L/min. The ESI source potentials were capillary 3 kV,
lens 0.5 kV, extractor 4 V, and cone voltage 5 V. This last
parameter had the most profound effect on the mass
spectra obtained. Increase in this voltage leads to the
so-called “in-source” fragmentation/dissociation, but a
too low cone voltage may cause a drop in sensitivity. A
Address reprint requests to Dr. R. Fran´ski, Faculty of Chemistry, Adam
Mickiewicz University, Grunwaldzka 6, 60-780 Poznan´, Poland. E-mail:
franski@amu.edu.pl
Published online January 7, 2010
© 2010 American Society for Mass Spectrometry. Published by Elsevier Inc.
1044-0305/10/$32.00
doi:10.1016/j.jasms.2009.12.018
Received August 11, 2009
Revised December 22, 2009
Accepted December 22, 2009

´
546
FRANSKI AND GIERCZYK
J Am Soc Mass Spectrom 2010, 21, 545–549
Nitrogen was used as the nebulizing and desolvating
gas at flow-rates of 100 and 300 l h–1, respectively.
B15C5 and NH2B15C5 were obtained from Sigma-
Aldrich and used without further purification.
NO2B15C5 was synthesized by nitration of B15C5 with
a nitric acid (60%) and sulfuric acid (98%) mixture (1:3,
vol/vol) at 10 °C. After dilution with water, the prod-
ucts were collected by filtration and purified by crys-
tallization from ethanol; the yield was about 70%.
Scheme 1. Crown ethers studied in this work.
Results and Discussion
low cone voltage of 5 V minimizes the fragmentation/
dissociation process. The source temperature was
120 °C and the desolvation temperature was 300 °C.
Sherman et al. [7] took into consideration the following
reaction:
Figure 1. ESI mass spectra of methanol solutions containing (a) B15C5 and NH2B15C5, (b) B15C5
and NO2B15C5, (c) NH2B15C5 and NO2B15C5. [B15C5 ϩ Na]ϩ m/z 291, [B15C5 ϩ K]ϩ m/z 307—this
peak overlaps with isotope peak of [NH2B15C5 ϩ Na]ϩ m/z 306, [NH2B15C5 ϩ K]ϩ m/z 322,
[NO2B15C5 ϩ Na]ϩ m/z 336, [NO2B15C5 ϩ K]ϩ m/z 352, [(B15C5)2 ϩ K]ϩ m/z 575, [(NH2B15C5)2 ϩ K]ϩ
m/z 605, [(NO2B15C5)2 ϩ K]ϩ m/z 665, [NH2B15C5 ϩ B15C5 ϩ K]ϩ m/z 590, [NH2B15C5 ϩ NO2B15C5 ϩ
K]ϩ m/z 635, [B15C5 ϩ NO2B15C5 ϩ K]ϩ m/z 620.

J Am Soc Mass Spectrom 2010, 21, 545–549
ESI-MS STUDY OF ␲-STACKING INTERACTIONS
547
[A2M]ϩ ϩ [B2M]ϩ ⇔ 2[ABM]ϩ
equal affinity toward M). Because the crown ethers
NH2B15C5 and NO2B15C5 definitely do not have the
same affinity toward alkali metal cations, it is reason-
able to conclude that the existence of ␲-stacking inter-
actions in [NH2B15C5 ϩ NO2B15C5 ϩ K]ϩ compensate
for the lower complexing ability of NO2B15C5. It must
be stressed that the ESI mass spectrum of a methanol
solution containing NH2B15C5 and NO2B15C5 is the
only one in which the mixed-ligand complex is more
abundant than the homo-ligand complexes (Figure 1).
Different peak intensities in the ESI mass spectra can be
a result of different ionization efficiencies of the ions
analyzed. For example, hydrophobic effects can play a
role. However, it was demonstrated that for alkali cation
complexes, this effect is important for complexes of low
stabilities [16]. Complexes of crown ethers with alkali
metal ions can be regarded as rather stable ones, at least
those of 1:1 stoichiometry. It is also expected that because
the interactions of the NH2 group with methanol involve
hydrogen bond formation in solution, the ionization effi-
ciency (desolvation) of the NH2B15C5 complexes will be
lowered. This was observed for copper complexes [17],
which are less stable than alkali metal complexes. As
shown in Figure 1, NH2-B15C5 complexes are abundant.
The results of ESI-MS analysis of the host–guest
complexes (including those involving macrocycles) cor-
relate well with the measurements of the equilibriums
in solution [7, 18]. It has been clearly demonstrated by
Brodbelt and coworkers that the host–guest complexes,
which have similar conformations and structural fea-
tures, such as a simple crown ether binding to two
different alkali metal ions, have similar electrospray
ionization efficiencies [19].
To verify whether the relative peak intensities in the
ESI mass spectra (for the complexes of the crown ethers
used in this work) reflect the relative concentrations in
solution, the ESI mass spectra were taken at high excess
of alkali metal cations, Naϩ and Kϩ (Figure 2). Under
such conditions, almost all of the crown ether molecules
(NH2B15C5 and NO2B15C5) are in the form of com-
plexes with the alkali metal cation. If the concentrations
of NH2B15C5 and NO2B15C5 are the same, the resulting
concentrations of the complexes (e.g., [NH2B15C5 ϩ
Na]ϩ and [NO2B15C5 ϩ Na]ϩ) will also be the same or
very similar. If the ESI responses (spray efficiencies) of
the complexes are similar, the respective peak intensi-
ties, obviously, will be also similar.
where A and B are two different crown ethers and M is
a metal cation. The equilibrium constant of this
reaction is:
([ABM]ϩ)2
K ϭ
[A2M]ϩ [B2M]ϩ
As clearly justified by the authors [7], the concentra-
tions of the homo-ligand complexes ([A2M]ϩ and
[B2M]ϩ) as well as the concentration of the mixed-
ligand complex ([ABM]ϩ) can be replaced by respective
peak intensities in the ESI mass spectra obtained. Fur-
ther, in the same paper the Gibbs free-energy was
calculated as ⌬G ϭ ϪRT lnK, and it was concluded that
“large positive ⌬G values signify that the homo-ligand
complexes are favored, near-zero values mean that the
homo-ligand and mixed-ligand complexes are equally
favored, and large negative values reflect a strong prefer-
ence for formation of mixed-ligand sandwich complexes.”
In our opinion, in the situation where the homo-
ligand and mixed-ligand complexes are equally favored,
the concentration/peak intensity ratio of [A2M]ϩ:[ABM]ϩ:
[B2M]ϩ must be 1:2:1, assuming that A and B have equal
affinity toward M. When the affinity of A toward M is
two times higher than that of B toward M, the ratio
should be 4:4:1, and when the affinity of A toward M is
three times higher than that of B toward M, the ratio
should be 9:6:1 (see Supplemental Information, which
can be found in the electronic version of this article, for the
calculation). In all cases, K ϭ 4 and the calculated Gibbs
free-energy is ⌬G ϭ Ϫ8.31 ϫ 298 ϫ ln4 ϭ Ϫ3433 J/mol.
Figure 1 shows the ESI mass spectra of methanol
solutions, each of which contains two of the three crown
ethers studied (similar spectra were obtained when Kϩ ion
was added to the analyzed solution; see Supplemental
Information). The diameter of the 15-crown-5 cavity is
170–220 pm [13] and the diameters of the molecules of the
benzo derivatives studied in this work can be considered
as nearly the same (or slightly smaller [14]. The diameters
of Naϩ and Kϩ cations are 196 pm and 266 pm, respec-
tively [15]. Therefore, the latter cation is able to form
sandwich type complexes (complexes of stoichiometry
2:1) with the crown ethers studied.
As follows from the results in Figure 1, B15C5 is
much more prone to form complexes with alkali metal
cations than NO2B15C5, and only slightly more prone
than NH2B15C5. This can be concluded on the grounds
of peak intensities of both 1:1 and 2:1 complexes (Figure
1a and b). Therefore, it can be expected that NH2B15C5
is also much more prone to form complexes with alkali
metal cations than NO2B15C5. As shown in Figure 1c, this
is true only for 1:1 complexes. Signal intensities of 2:1
complexes decrease in the order [NH2B15C5 ϩ NO2B15C5 ϩ
K]ϩ Ͼ [(NH2B15C5)2 ϩ K]ϩ Ͼ [(NO2B15C5)2 ϩ K]ϩ. The
peak intensity ratio is close to 1:2:1 ([A2M]ϩ:[ABM]ϩ:[B2M]ϩ,
the situation when the homo-ligand and mixed-ligand
complexes are equally favored and A and B have
As demonstrated in Figure 2, the respective peak
intensities are very similar, especially for the ions
[NH2B15C5 ϩ Na]ϩ and [NO2B15C5 ϩ Na]ϩ; the latter
has 95% relative intensity of the former, indicating that
complexes of the crown ethers have similar ESI re-
sponses. Ions [NH2B15C5 ϩ Na]ϩ and [NO2B15C5 ϩ
Na]ϩ are the only crown ether complexes obtained in
the solution containing excess Naϩ cations. The peaks
of the ions [NH2B15C5 ϩ K]ϩ and [NO2B15C5 ϩ K]ϩ
are also similar, and the latter has 88% relative intensity
of the former; however, this result is slightly worse than
that obtained for sodium complexes. The ESI mass

´
548
FRANSKI AND GIERCZYK
J Am Soc Mass Spectrom 2010, 21, 545–549
Figure 2. ESI mass spectra of methanol solutions containing (a) NH2B15C5, NO2B15C5, and an excess of
(a) Naϩ cations (NaCl was added), (b) Kϩ cations (KCl was added). [NH2B15C5 ϩ Na]ϩ m/z 306,
[NO2B15C5 ϩ Na]ϩ m/z 336, [NH2B15C5 ϩ K]ϩ m/z 322, [NO2B15C5 ϩ K]ϩ m/z 352, [(NH2B15C5)2 ϩ K]ϩ
m/z 605, [NH2B15C5 ϩ NO2B15C5 ϩ K]ϩ m/z 635, [(NO2B15C5)2 ϩ K]ϩ m/z 665. Clusters [(NaCl)n ϩ Na]ϩ
(a) and clusters [(KCl)n ϩ K]ϩ (b) are clearly seen—cluster [(KCl)4 ϩ K]ϩ overlaps with ion [NO2B15C5 ϩ Na]ϩ.
spectrum obtained for the solution containing crown
ethers and an excess of Kϩ cations indicates that not all
crown ether molecules form 1:1 complexes with Kϩ
cations. There are complexes with Naϩ cations and
there are complexes with Kϩ cations of 2:1 stoichiome-
try. The explanation is that the Kϩ cation is too large to
fit well into the crown ether cavities (in contrast to the
Naϩ cation) and, therefore, other complexes are also
formed. It can be assumed that if 1:1 complexes have
similar ESI responses, the 2:1 complexes (formed by the
same crown ethers) will have similar responses as well.
peared. We have searched for a solvent in which the
␲-stacking interactions could be enhanced, in compari-
son to methanol, for example methanol/chloroform.
The Effect of Solvent and Metal Cation Radius
If the discussed ␲-stacking interactions exist in solution,
a change in solvent should lead to visible changes in
relative peak intensities. Acetonitrile is expected to
decrease the ␲-stacking interactions in the complexes
studied since this solvent can be a source of ␲-stacking
interactions itself. Water, a solvent much more polar
than methanol, is also expected to affect the ␲-stacking
interactions. Figure 3 shows the ESI mass spectra, in the
m/z range of 2:1 complexes, obtained by using metha-
nol/acetonitrile and methanol/water as solvents, both
in 1:1 stoichiometry.
Figure 3. ESI mass spectra, in m/z range of 2:1 complexes,
obtained by using methanol/acetonitrile (1/1) and methanol/
water (1/1) as solvents. In methanol/acetonitrile the complexes
have low abundance because coordination of the alkali metal
cation by acetonitrile disturbs formation of the complexes [20].
As clearly seen, in both cases the formation of the
mixed-ligand complex is not favored in comparison to
homo-ligand complexes, indicating that the ␲-stacking
interactions observed in pure methanol have disap-

J Am Soc Mass Spectrom 2010, 21, 545–549
ESI-MS STUDY OF ␲-STACKING INTERACTIONS
549
References
1. Gokel, G. W.; Leevy, W. M.; Weber, M. E. Crown Ethers: Sensors for
Ions and Molecular Scaffolds for Materials and Biological Models. Chem.
Rev. 2004, 104, 2723–2750.
2. Leize, E.; Anne Jaffrezic, A.; Van Dorsselaer, A. Correlation Between
Solvation Energies and Electrospray Mass Spectrometric Response
Factors. Study by Electrospray Mass Spectrometry of Supramolecular
Complexes in Thermodynamic Equilibrium in Solution. J. Mass Spec-
trom. 1996, 31, 537–544.
3. Shen, N.; Pope, R. M.; Dearden, D. V. Fundamental Factors Controlling
the Exchange of Multidentate Ligands: Displacement of 12-Crown-4
and Triglyme from Complexes with Divalent Alkaline Earth Cations.
Int. J. Mass Spectrom. 2000, 195/196, 639–652.
4. Nicoll, J. B.; Dearden, D. V. Reactions of Multidentate Ligands with
Ligated Alkali Cation complexes: Self-Exchange and “Sandwich” Com-
plex Formation Kinetics of Gas Phase Crown Ether–Alkali Cation
Complexes. Int. J. Mass Spectrom. 2001, 204, 171–183.
5. Chu, I.-H.; Zhang, H.; Dearden, D. V. Macrocyclic Chemistry in the Gas
Phase: Intrinsic Cation Affinities and Complexation Rates for Alkali
Metal Cation Complexes of Crown Ethers and Glymes. J. Am. Chem. Soc.
1993, 115, 5736–5744.
6. Oshima, T.; Matsuda, F.; Fukushima, K.; Tamura, H.; Matsubayashi, G.;
Arakawa, R. Selective Cation Binding of Crown Ether Acetals in Electros-
pray Ionization Mass Spectrometry. J. Chem. Soc. Perkin Trans. 1998, 2,
145–148.
7. Sherman, C. L.; Brodbelt, J. S.; Marchand, A. P.; Poola, B. Electrospray
Ionization Mass Spectrometric Detection of Self-Assembly of a Crown
Ether Complex Directed by ␲-Stacking Interactions. J. Am. Soc. Mass
Spectrom. 2005, 16, 1162–1171.
8. Yamaguchi, N.; Gibson, H. W. Noncovalent Chemical Modification of
Crown Ether Side-chain Polymethacrylates with a Secondary Ammo-
nium Salt: A Family of New Polypseudorotaxanes. Macromol. Chem.
Phys. 2000, 201, 815–824.
9. Bryant, W. S.; Guzei, I. A.; Rheingold, A. L.; Merola, J. S.; Gibson H. W.
A Study of the Complexation of Bis(m-Phenylene) Crown Ethers and
Secondary Ammonium Ions. J. Org. Chem. 1998, 63, 7634–7639.
10. Meyer, E. A.; Castellano, R. K.; Diederich, F. Interactions with Aromatic
Rings in Chemical and Biological Recognition. Angew. Chem. Int. Ed.
2003, 42, 1210–1250.
11. Ma, J. C.; Dougherty, D. A. The Cation-␲ Interaction. Chem. Rev. 1997,
97, 1303–1324.
12. Gallivan, J. P.; Dougherty, D. A. Cation-␲ interactions in structural
biology. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 9459–9464.
13. Maleknia, S.; Brodbelt, J. Cavity-Size-Dependent Dissociation of Crown
Ether/Ammonium Ion Complexes in the Gas Phase. J. Am. Chem. Soc.
1993, 115, 2837–2843.
14. Colton, R.; Mitchell, S.; Traeger, J. C. Interactions of Some Crown Ethers,
Cyclam, and Its Tetrathia Analogue with Alkali, Alkali Earth, and Other
Metal Ions: An Electrospray Mass Spectrometric Study. Inorg. Chim.
Acta 1995, 231, 87–93.
15. Emsley, J. The Elements. Oxford Chemistry Guides 2nd ed.; Clarendon
Press: Oxford, 1991; p. 156, 186.
16. Fran´ski, M.; Przybysz, I.; Borkowska, M.; Kozik, T. Cationization Versus
Surface Activity—The Effect on Electrospray Ionization. Eur. J. Mass
Spectrom. 2006, 12, 15–18.
17. Fran´ski, R.; Gierczyk B.; Schroeder, G. Anion-␲ Interactions—Interactions
Between Benzo-Crown Ether Metal Cation Complexes and Counter Ion.
J. Am. Soc. Mass. Spectrom. 2009, 20, 257–262.
18. Leize, E.; Jaffrezic A. A.; Van Dorsselaer, A. Correlation Between
Solvation Energies and Electrospray Mass Spectrometric Response
Factors. Study by Electrospray Mass Spectrometry of Supramolecular
Complexes in Thermodynamic Equilibrium in Solution. J. Mass Spec-
trom. 1996, 31, 537–544.
19. Reyzer, M. L.; Brodbelt, J. S.; Marchand, A. P.; Chen, Z.; Huang Z.;
Namboothiri, I. N. N. Determination of Alkali Metal Binding Selectiv-
ities of Caged Crown Ligands by Electrospray Ionization Quadrupole
Ion Trap Mass Spectrometry. Int. J. Mass Spectrom. 2001, 204, 133–142.
20. Fran´ski, R.; Schroeder, G.; Gierczyk, B.; Niedziałkowski, P.; Ossowski,
T. Formation of Stoichiometric Complexes between Dibenzo-30-
Crown-10 and Guanidinium Moiety Containing Compounds. Int. J.
Mass Spectrom. 2007, 266, 180–184.
Figure 4. ESI mass spectra, in m/z range of 2:1 complexes,
obtained after adding RbClO4 (left) and CsClO4 (right) to the
analyzed methanol solution; [(NH2B15C5)2 ϩ Rb]ϩ m/z 651,
[(NO2B15C5)2 ϩ Rb]ϩ m/z 711, [NH2B15C5 ϩ NO2B15C5 ϩ Rb]ϩ
m/z 681, [(NH2B15C5)2 ϩ Cs]ϩ m/z 699, [(NO2B15C5)2 ϩ Cs]ϩ m/z
759, [NH2B15C5 ϩ NO2B15C5 ϩ Cs]ϩ m/z 729.
However, a significant increase in the abundance of the
mixed-ligand complex relative to the homo-ligand com-
plexes was never observed.
As described in detail by Sherman et al. [7], the size of
the metal cation is of crucial importance for the contribu-
tion of ␲-stacking interactions in the sandwich complexes.
As shown in Figure 4, the mixed-ligand complexes are
more abundant than the homo-ligand complexes for ru-
bidium but not for cesium.
Because of the large radius of the Csϩ cation, the
interaction between aromatic rings is not possible.
This finding is additional confirmation that in the
[NH2B15C5 ϩ NO2B15C5 ϩ K]ϩ ions (and also in
[NH2B15C5 ϩ NO2B15C5 ϩ Rb]ϩ ions), ␲-stacking
interactions are important.
Appendix A
Supplementary Material
Supplementary material associated with this article
may be found in the online version at doi:10.1016/
j.jasms.2009.12.018.