Characterization of chiral host-guest complexation in fast atom bombardment mass spectrometry

Article abstract of DOI:10.1021/ac950912x

A new method has been developed for the characterization of complexion between host and guest molecules. Adduct formation between chiral crown ethers 1 and 2 and enantiomeric ammonium ions 4 and 5 was examined. The reference compound 3 (achiral host) was chosen to be similar in structure to the chiral crown ethers for quantitative measurements. Our approach is based on a formalism assuming an equilibrium: [chiral host + H]⁺ + [achiral host + chiral guest]⁺1 [chiral host + chiral guest]⁺ + [achiral host + H]⁺. The equlibrium constant for this process was calculated using the relative peak intensities of the corresponding species in the FAB mass spectra. It was found that these provide significantly better reproducibility and more reliable results than the relative peak intensity method described before (Sawada, M.; et al. J. Am. Chem. Soc. 1992, 114, 4405; 1993, 115, 7381; Org. Mass Spectrom. 1993, 28, 1525).^1-3 In the examples studied, the equilibrium constants corresponding to the formation of heterochiral adducts (S,S-R or R,R-S) were higher than those for the formation of homochiral aggregates (S,S-S or R,R-R).

Full text of DOI:10.1021/ac950912x

Anal. Chem. 1996, 68, 792-795
Characterization of Chiral Host-Guest
Complexation in Fast Atom Bombardment Mass
Spectrometry
Gabriella Po´ csfalvi,† Miklo´ s Lipta´k,† Pe´ ter Huszthy,‡ Jerald S. Bradshaw,§ Reed M. Izatt,§ and
Ka´ roly Ve´ key*,†
Mass Spectrometry Group, Central Research Institute for Chemistry of the Hungarian Academy of Sciences,
H-1025 Budapest, Pusztaszeri u 59-67, Hungary, Department of Organic Chemistry, Technical University of Budapest,
H-1111 Budapest, Gelle´rt te´r 4, Hungary, and Brigham Young University, Provo, Utah 84602
A new method has been developed for the characterization
of complexion between host and guest molecules. Adduct
formation between chiral crown ethers 1 and 2 and
enantiomeric ammonium ions 4 and 5 was examined. The
reference compound 3 (achiral host) was chosen to be
similar in structure to the chiral crown ethers for quan-
titative measurements. Our approach is based on a
formalism assuming an equilibrium: [chiral host + H]+
+ [achiral host + chiral guest]+ h [chiral host + chiral
guest]+ + [achiral host + H]+. The equlibrium constant
for this process was calculated using the relative peak
intensities of the corresponding species in the FAB mass
spectra. It was found that these provide significantly
better reproducibility and more reliable results than the
relative peak intensity method described before (Sawada,
M.; et al. J . Am . Ch em . Soc. 1 9 9 2 , 114, 4 4 0 5 ; 1 9 9 3 ,
115, 7 3 8 1 ; Or g. Ma ss Spectr om . 1 9 9 3 , 28, 1 5 2 5 ).1 -3
In the examples studied, the equilibrium constants cor-
responding to the formation of heterochiral adducts
(S,S-R or R,R-S) were higher than those for the formation
of homochiral aggregates (S,S-S or R,R-R).
Figure 1. Chiral macrocycles and primary ammonium cations
studied.
and gave higher RPI values than their homochiral (RR or SS)
counterparts. A more detailed investigation2 was also carried out
to demonstrate the usefulness of the RPI method. The time
dependence of RPI values was examined, and they were found to
be nearly constant in a 1 min interval. The degree of chiral
recognition (RPIR/ RPIS) reached 1.6 in some cases. In another
study,3 RR and SS homochiral adducts were found to be more
stable than RS or SR adducts. With the help of Fourier transform
ion cyclotron resonance measurements4 the equilibrium constants
for the complexation between a chiral crown ether and the
enantiomers of 1-(R-naphthyl)ethylammonium cation were deter-
mined. The heterochiral cluster was found to be more stable than
the homochiral one, the stability difference being ∆∆G ) 4.2 (
Determination of the discrimination between guest enanti-
omers by a chiral host has always been a challenge for mass
spectrometrists. The use of chiral crown ethers as host molecules
and the investigation of their adduct formation with chiral guest
compounds by mass spectrometry have gained a lot of interest
during the last few years.1-4 Sawada and co-workers1 reported
the use of modified carbohydrates and modified chiral crown
ethers as host molecules (Chir) for the discrimination of enan-
tiomeric alkylammonium ions (A). An internal standard com-
pound (Ref) structurally related to the host was employed and
0.4 kJ mol-1
.
In the present paper we describe a new approach which
provides a more reliable and reproducible method for the
determination of the degree of chiral discrimination than the
above-mentioned (RPI) technique. The usefulness of our meth-
odology is demonstrated by two chiral crown ethers (1 and 2 )
as host molecules and two enantiomeric pairs of chiral ammonium
salts (4 R, 4 S and 5 R, 5 S] (see Figure 1) as guests.
+
+
the relative peak intensities (RPI) [I(Chir+A )/ I(Ref+A )] were calcu-
lated for the relevant diastereomeric adduct ions. It was found
that the studied heterochiral adducts (RS or SR) were more stable
† Central Research Institute for Chemistry.
‡ Technical University of Budapest.
§ Brigham Young University.
(1) Sawada, M.; Shizuma, M.; Takai, Y.; Yamada, H.; Kaneda, T.; Hanafusa, T.
J. Am. Chem. Soc. 1 9 9 2 , 114, 4405.
(2) Sawada, M.; Okumura, Y.; Shizuma, M.; Takai, Y.; Hidaka, Y.; Yamada, H.;
Tanaka, T.; Kaneda, T.; Hirose, K.; Misumi, S.; Takahashi, S. J. Am. Chem.
Soc. 1 9 9 3 , 115, 7381.
(3) Sawada, M.; Okumura, Y.; Yamada, H.; Takai, Y.; Takahashi, S.; Kaneds,
T.; Hirose K.; Misumi, S. Org. Mass Spectrom. 1 9 9 3 , 28, 1525.
(4) Chu, I.-H.; Dearden, D. V.; Bradshaw, J. S.; Huszthy, P.; Izatt, R. M. J. Am.
Chem. Soc. 1 9 9 3 , 115, 4318.
EXPERIMENTAL SECTION
1 was prepared by a method different from that described in
the literature.5 It was prepared from dimethyl-2,6-pyridinedicar-
(5) Jones, B. A.; Bradshaw, J. S.; Izatt, R. M. J. Heterocycl. Chem. 1 9 8 2 , 19,
551.
792 Analytical Chemistry, Vol. 68, No. 5, March 1, 1996
0003-2700/96/0368-0792$12.00/0 © 1996 American Chemical Society

Table 1. Relative Stability Constants of Crown
Ether-Ammonium Cation Complexesa
FAB
solution
1
2
1
2
4
5
1.00 ( 0.02
0.92 ( 0.02
0.17 ( 0.01
0.12 ( 0.01
1.0 ( 0.09
1.2 ( 0.04
6.1 ( 0.09
6.1 ( 0.09
a
Mean values of homo- and heterochiral adducts are given relative
to the complex 1 -4 . These data are determined from the relative
stability constants (K) measured in the present work, characterizing
FAB conditions, and K values determined in methanol solution.8
boxylate and (2S,12S)-(+)-4,7,10-trioxatridecane-2,12-diol in the
presence of 4 Å molecular sieves and a catalytic amount of sodium
methoxide by base-catalyzed transesterification reaction as de-
Figure 2. FAB mass spectrum obtained by mixing 1, 3, and 4S.
scribed for the preparation of the di-tert-butyl analogue.6
1
obtained by this base-catalyzed transesterification method gave
the same optical rotation, melting point, and IR and 1H NMR
spectra as 1 prepared by the literature procedure.5 2 6 and 3 7
were prepared as described in the literature. The hydrogen
perchlorate salts of 4 and 5 were prepared according to the
method described.8
calculated and used for the characterization of host-guest
interaction phenomena. Our approach involves the assumption
of an equilibrium between protonation and organic ammonium
cation exchange, which can be described by the following
equations:
Fast atom bombardment mass spectra were recorded on a VG-
ZAB-2SEQ (VG Analytical) reverse geometry mass spectrometer
operating at an accelerating voltage of 8 kV. The sample was
bombarded with a beam of Cs+ ions having an energy of 30 keV.
Scans were collected for 20 min in the mass range of 100-1000
Da using a 5 s/ scan rate. The spectra were averaged for every
minute, and the relative peak intensities listed by the data system
were used to calculate the equilibrium constants.
ChirH+ + Ref h Chir + RefH+
RefA+ + Chir h Ref + ChirA+
(1)
(2)
where Chir is the chiral crown ether, Ref is the achiral (reference)
crown ether, and A+ is the organic ammonium cation. The sum
of these two processes is
The host (1 and 2 ), reference (3 ), and guest (perchlorate salts
of 4 R or 4 S and 5 R or 5 S) compounds were dissolved in ethyl
acetate (0.1 M). The solutions of host and reference compounds
were mixed (1:1), and these mixtures were used for the studies.
Glycerol (1 µL), the solution of the host + reference (1 µL), and
solution of the guest compound (1 µL) were mixed on the FAB
probe and introduced into the ion source. In the experiments, at
least three parallel measurements were performed and the aver-
age K values were calculated. The results are summarized in
Table 1.
ChirH+ + RefA+ h ChirA+ + RefH+
(3)
An equilibrium constant (K) can be defined corresponding to
eq 3
[ChirA+][RefH+]
[ChirH+][RefA+]
K )
(4)
In eq 4, the brackets indicate concentrations. As only ionic
species are present whose concentration is proportional to their
relative peak abundances, K can be determined from the mass
spectra.
RESULTS AND DISCUSSION
The adducts due to host-guest complexes gave reasonably
abundant peaks in the FAB mass spectra. Figure 2 shows the
mass spectrum obtained by mixing 1 , 3 , and 4 S. The protonated
molecules of the reference (Ref, an achiral host) and the chiral
crown ether (Chir) give rise to peaks at m/ z 326 and 354,
respectively. m/ z 364 and 392 are the potassium adducts due to
K+ traces in the sample, whereas the peaks at m/ z 447 and 475
are due to the reference + amine guest (Ref + A+) and to the
chiral host + amine guest (Chir + A+) adducts, respectively.
Methods for Data Evaluation. In earlier studies,1-4 the
Figure 3 shows the results of a typical experiment reproduced
three times under identical conditions where the spectra were
averaged for every minute between 1 and 20 min measurement
time and the RPI and K values were calculated. It can clearly be
seen that the K value remains constant (or shows a very slight
linear time dependence) whereas RPI values show a relatively
large scattering. We believe that this dramatic improvement
shows the advantage of using an equilibrium constant instead of
a somewhat arbitrary peak ratio. We do not claim that there is a
true equilibrium in the FAB process (it may only approximate an
equilibrium or a steady state), but using this formalism we get
much more reproducible results. From another point of view,
the scattering using RPI values is related to a change in the
[RefH+]/ [ChirH+] ratio. It is a typical effect in FAB that the
spectra of mixtures show a time dependence; in the present case,
+
+
relative peak intensity RPI ) I(Chir+A / I(Ref+A values were
)
)
(6) Huszthy, P.; Bradshaw, J. S.; Zhu, C.-Y.; Izatt R. M.; Lifson, S. J. Org. Chem.
1 9 9 1 , 56, 3330.
(7) Izatt, R. M.; Lamb, J. D.; Assay, R. E.; Maas, G. E.; Bradshaw, J. S.;
Christensen, J. J. J. Am. Chem. Soc. 1 9 7 7 , 99, 6134.
(8) Izatt, R. M.; Wang, T.-M.; Hathaway, J. K.; Zhang, X.-X.; Curtis, J. C.;
Bradshaw, J. S.; Zhu C.-Y.; Huszthy, P. J. Inclusion Phenom. Mol. Recognit.
Chem. 1 9 9 4 , 17, 157.
Analytical Chemistry, Vol. 68, No. 5, March 1, 1996 793

Figure 4. Comparison of enantiomers 5R and 5S using chiral crown
ether 1: KSS-R/KSS-S ) 1.17.
Table 2. Chiral Discrimination (Ratios of Stability
Constants, KRR-S/KRR-R and KSS-R/KSS-S) for Hetero- vs
Homochiral Complexesa
FAB
solution
1
2
1
2
4
5
1.10 ( 0.02
1.17 ( 0.02
1.29 ( 0.10
1.70 ( 0.15
1.7 ( 0.31
2.7 ( 0.19
1.4 ( 0.28
1.5 ( 0.28
a
FAB data refer to the present measurement of equilibrium
constants (K); solution data were determined in methanol; the stability
Figure 3. RPI and K values vs time. (RPI - [ChirA+]/[RefA+], K )
[RH+] [ChirA+] / [RefA+] [ChirH+]); Chir ) 1; Ref ) 3; A ) 5S.
constants are taken from the literature.8
The stability of crown ether-ammonium cation complexes
studied here is determined mainly by two attractive and one
repulsive types of interactions:8,9 three hydrogen bonds between
the ammonium protons of the ammonium cation and the pyridine
nitrogen and two alternating macroring oxygens of the crown
ether; π-π interaction between the phenyl (4 ) or naphthyl (5 )
group of the ammonium cation and the pyridine ring of the crown
ether; and steric hindrance between the ammonium cation and
the methyl (1 ) or phenyl (2 ) substituents of the crown ether. In
the complexes studied, this third interaction, steric hindrance, is
the main difference. This is likely the primary cause of enantio-
meric discrimination as well.
The stability of crown ether-ammonium cation complexes can
be characterized by mass spectrometry using the relative equi-
librium constant (K) described above. The precise mechanism
and place [solution, selvedge (the relatively high-pressure gas
close to the matrix surface), or gas phase] of the formation of
these complexes under FAB conditions is, however, unclear. Ion
abundances probably reflect steady state rather than true equi-
librium conditionssso the equilibrium constants described in this
paper should be regarded rather as an analytically useful char-
acterization of the complexes and of chiral recognition than as
true equilibrium values characterizing either the gas or the liquid
phase. Even so, the stability of and chiral effects in the complexes
studied under FAB conditions show a remarkable, and chemically
very reasonable, dependence on the substituents. This behavior
contrasts somewhat with that determined in solution (see Tables
1 and 2). This suggests that stabilities determined by FAB
characterize complexes significantly isolated from their molecular
environment, as in the gas phase, contrasting with that in solution
the [RefH+]/ [ChirH+] ratio changes. This ratio could change for
several reasons, e.g., slightly different concentration of Ref and
Chir in the matrix, surface effects (one of these compounds may
be concentrated at the surface, and this may change during the
experiment), etc. In experiments where very small differences
have to be detected between the stabilities of different adduct
species, it is very important to use a precise and reproducible
method for data evaluation and sample characterization. As these
results indicate, the equilibrium method developed here is much
more suitable and accurate one for the study of host-guest
complexation than the RPI method described before.1
When the concentrations of host + reference and those the
guest were changed independently by ∼1 order of magnitude, it
was found that the calculated K values corresponding to a
particular system changed only slightly (∼3%). Thus, if only a
very small amount of sample is available it is not necessary, that
the reference compound and the chiral analyte be applied in the
same concentration.
Comparison of R and S Enantiomers. As mentioned in the
introduction, in some cases3 homochiral and in others1 hetero-
chiral aggregates were found to give larger RPI values. In a study
by Sawada et al.,2 regardless of the configuration of the two applied
+
+
crown ethers, I(Chir+A ) / I(Chir+A
) ratios were found to be 1.2 for
R
S
1-(R-naphthyl)ethylammonium ion complexation. Our data show
that heterochiral complexation is more favorable than the homo-
chiral one; that is, K values corresponding to S,S-R and R,R-S were
higher than the values of S,S-S and R,R-R, respectively. In Figure
4, the results of the 1 -5 R and 1 -5 S complexation experiments
are shown. The K values exhibit a slight time dependence, but
the two enantiomers of ammonium ions can unambiguously be
distinguished, the two lines are well separated throughout the
measurement.
(9) Hathaway, J. K.; Izatt, R. M.; Zhu, C.-Y.; Huszthy P.; Bradshaw, J. S.
Supramol. Chem. 1 9 9 5 , 5, 9.
794 Analytical Chemistry, Vol. 68, No. 5, March 1, 1996

where interactions with solvent molecules have a crucial role in
determining stability constants.
of phenyl, compared to that of the smaller (and spherically
approximately symmetrical) methyl substituent. In one diaste-
reomer (the heterochiral complex) the steric hindrance between
substituents is smaller, so this complex has higher stability. The
larger the substituents, the more important this repulsive effect
is. This effect decreases stability but increases chiral discrimina-
tion at the same time.
In solution, the conformation of the crown ether, especially
the plane of the phenyl substituent, is likely to be more flexible
as the mobile solvent molecules can stabilize many possible
conformations. This distorts both the “intrinsic” stability and the
“intrinsic” chiral recognition between the two enantiomers.
Therefore the correlation between substituent size and stability,
between substituent size and chiral discrimination; and between
stability and chiral discrimination becomes more difficult to
rationalize in solution than in FAB conditions.
The study of chiral recognition by FAB has two main
advantages compared to other studies: (1) the high sensitivity of
mass spectrometry allows characterization using very small sample
amounts, and (2) under FAB conditions, as the results in the
present paper show, “intrinsic” chiral effects can be studied and
good correlations between substituent size and complex stability
and also between substituent size and enantioselectivity can be
found.
As shown in Table 1, the stability of the complexes under FAB
conditions depends very significantly on the size of the substit-
uents both of the ammonium cation and of the crown ether
(methyl vs phenyl). Steric hindrance between larger substituents
results in lower stability in going from the phenyl (4 ) to the
naphthyl (5 ) group of the ammonium cation and especially from
methyl (1 ) to phenyl (2 ) substituents of the crown ether. This
behavior contrasts with that in solution, where the difference in
the stability of complexes with phenyl- and naphthylethylammo-
nium cations is negligible.8 In solution, the diphenyl-substituted
crown ether (2 ) gives more stable complexes than the dimethyl-
substituted one (1 )sopposite to that observed in FAB. This
undoubtedly reflects the effect of solvent molecules on complex
stability and likely stereochemical behavior (enantioselectivity).
The heterochiral complexes (S,S-R or R,R-S) studied in the
present paper are more stable than homochiral ones (S,S-S or
R,R-R) both in FAB and in solution. (In the case of the complexes
of 1 -5 R and 1 -5 S this order of stability was also observed in
the crystalline state10 .) This stability order is explained by steric
repulsion between the substituents, destabilizing mainly the
homochiral complex. Enantioselectivity is inversely proportional
to the stability of the ammonium cation-crown ether complex in
FAB: 2 -5 is 8 times less stable with 7 times larger enantiose-
lectivity than 1 -4 . Substituent size has a smaller effect on the
enantioselectivity in solution-phase equilibrium measurements,8
and under such conditions, there is also little correlation between
enantioselectivity and complex stability.
ACKNOWLEDGMENT
Financial support of the Hungarian National Science Fund
(Grant OTKA-T 4140) and the Office of Naval Research (USA) is
gratefully acknowledged. The authors to thank David V. Dearden
for his useful comments.
A possible explanation of this behavior is that, under FAB
conditions (which is likely to be close to a solvent-free environ-
ment), optimum conformation of the crown ether, especially the
plane of the phenyl substituent, could be fairly rigid. This changes
during complex formation, decreasing complex stability in the case
Received for review September 13, 1995. Accepted
December 1, 1995.
AC950912X
(10) Davidson, R. B.; Bradshaw, J. S.; Jones, B. A.; Dalley, N. K.; Christensen, J.
J.; Izatt, R. M.; Morin F. G.; Grant, D. M. J. Org. Chem. 1 9 8 4 , 49, 353.
X Abstract published in Advance ACS Abstracts, January 15, 1996.
Analytical Chemistry, Vol. 68, No. 5, March 1, 1996 795