Cyclochiral resorcin[4]arenes as effective enantioselectors in the gas phase

Article abstract of DOI:10.1002/jms.2028

The effect of cyclochirality of rccc-2,8,14,20-tetra-n-decyl-4,10,16,22- tetra-O-methylresorcin[4]arene (C) on the enantiodiscrimination of a number of chiral bidentate and tridentate aromatic and aliphatic biomolecules (G) has been investigated by nano-electrospray ionization (nano-ESI)-Fourier transform ion cyclotron resonance mass spectrometry. The experimental approach is based on the formation of diastereomeric proton-bound [C·H·G]⁺ complexes by nano-ESI of solutions containing an equimolar amount of quasi-enantiomers (C) together with the chiral guest (G) and the subsequent measurement of the rate of the G substitution by the attack of several achiral and chiral amines. In general, the heterochiral complexes react faster than the homochiral ones, except when G is an aminoalcoholic neurotransmitter whose complexes, beyond that, exhibit the highest enantioselectivity. The kinetic results were further supported by both collision-induced dissociation experiments on some of the relevant [C₂·H·G] ⁺ three-body species and Density functional theory (DFT) calculations performed on the most selective systems. Copyright

Full text of DOI:10.1002/jms.2028

Research Article
Received: 8 June 2011
Revised: 9 September 2011
Accepted: 14 November 2011
Published online in Wiley Online Library
(wileyonlinelibrary.com) DOI 10.1002/jms.2028
Cyclochiral resorcin[4]arenes as effective
enantioselectors in the gas phase
Caterina Fraschetti,a,b* Matthias C. Letzel,a Marlene Paletta,a
Jochen Mattay,a Maurizio Speranza,b Antonello Filippi,b
Massimiliano Aschic and Alexander B. Rozhenkoa,d
The effect of cyclochirality of rccc-2,8,14,20-tetra-n-decyl-4,10,16,22-tetra-O-methylresorcin[4]arene (C) on the enantiodiscri-
mination of a number of chiral bidentate and tridentate aromatic and aliphatic biomolecules (G) has been investigated by
nano-electrospray ionization (nano-ESI)-Fourier transform ion cyclotron resonance mass spectrometry. The experimental
approach is based on the formation of diastereomeric proton-bound [CꢀHꢀG]+ complexes by nano-ESI of solutions containing
an equimolar amount of quasi-enantiomers (C) together with the chiral guest (G) and the subsequent measurement of the rate
of the G substitution by the attack of several achiral and chiral amines. In general, the heterochiral complexes react faster than
the homochiral ones, except when G is an aminoalcoholic neurotransmitter whose complexes, beyond that, exhibit the high-
est enantioselectivity. The kinetic results were further supported by both collision-induced dissociation experiments on some
of the relevant [C2ꢀHꢀG]+ three-body species and Density functional theory (DFT) calculations performed on the most selective
systems. Copyright © 2012 John Wiley & Sons, Ltd.
Supporting information may be found in the online version of this article.
Keywords: cyclochirality; resorc[4]arene; enantiodiscrimination; gas phase; FT-ICR; inherent chirality; ligand exchange; supramolecular
complex
methodology.[23] To shed more light on the intrinsic factors
governing enantiodiscrimination by cyclochiral receptors, we gen-
INTRODUCTION
One of the most important goals of the supramolecular chemistry is
the construction of artificial receptors, which can mimic enzyme
activity and selectivity, and are applicable to asymmetric syntheses.
The noncovalent interactions established between chiral host
molecules, for example, proteins, with a variety of guests, as well
as the origin of diastereoselection in the nucleophilic capture by
chiral cation, have been intensively investigated in the gaseous
phase, where solvation effect and ion pairing do not interfere.[1–6]
In general, the gas-phase studies of diastereomeric complexes
provide an efficient probe to evaluate the intrinsic factors govern-
ing enantiorecognition processes.[7] Chiral resorcin[4]arenes are
commonly recognized as instances of potential receptors success-
fully employed in enantiorecognition and catalysis.[8–10] Regarding
resorcin[4]arenes, chirality can be due to (1) the presence of stereo-
genic centers in their side chains, and (2) the spatial arrangement of
achiral subunits forming a structurally chiral macrocyclic scaffold.
The concepts of ‘cyclochirality’ and ‘inherent chirality’ apply to
the latter case. ‘Cyclochirality’ is defined as molecular asymmetry
due to the sense of direction of the rings in a XXXX system
composed by the same building blocks, chemically locked as in
the case of rotaxanes[11] or noncovalently locked by hydrogen
bonds[12], whereas the ‘inherent chirality’ is referred to as a cova-
lently blocked curvature of a XXYZ or WXYZ system.[13] To date,
the recognition capability of gaseous resorcin[4]arenes has been
kinetically explored using predominantly chiral macrocycles of
type 1).[14–22] A first study on the complexation behavior of a cyclo-
chiral resorcinarene towards several chiral quaternary ammonium
ions has been investigated by us using the ESI-mass spectrometry
erated proton-bound adducts between resorcinarenes C (Chart 1)
and several chiral bidentate or tridentate biomolecules (G, Table 1),
and their reactivity towards some primary amines (Chart 1; B:
CH3CHWNH2 with W = H, CH3, C2H5) has been measured in the
gaseous phase. The cyclochirality of the macrocycles of Chart 1 is
due to the presence of four methoxy and four hydroxy groups on
the aromatic scaffold, which are arranged either clockwise or
counterclockwise. Following the stereochemical nomenclature for
compounds of axial symmetry, we denote the two cyclochiral enan-
tiomers of Chart 1 as (M,R)-C and (P,S)-C (CM,R and CP,S in Chart 1). M
and P notations are referred to the axial chirality of the compound,
whereas R and S describe the inter-ring configuration centers.[24]
*
Correspondence to: Caterina Fraschetti, Department of Chemistry Bielefeld
University, P.O. Box 100131, 33501, Bielefeld, Germany. E-mail: caterina.fraschetti@
uniroma1.it
a Department of Chemistry Bielefeld University, P.O. Box 100131, 33501, Bielefeld,
Germany
b Department of Chemistry and Technologies of Drug, Sapienza-University of
Rome, Piazzale Aldo Moro, 5, 00185, Rome, Italy
c
Department of Chemistry, L’Aquila University, Via Vetoio s.n.c. Coppito,
L’Aquila, Italy
d Institute of Organic Chemistry of the National Academy of Sciences of Ukraine,
Murmans’ka str. 5, 02094 Kyiv, Ukraine
J. Mass. Spectrom. 2012, 47, 72–79
Copyright © 2012 John Wiley & Sons, Ltd.

Cyclochirality in the gas phase
The absolute configuration of the cyclochiral resorcinarenes is
known from earlier studies by Heaney et al.[24] The CD and NMR
spectroscopic details are described in Supporting Information.
All the other employed substances are commercially available.
ESI-FT-ICR experiments
The experiments were performed at room temperature in an APEX
III 7.0 Tesla Fourier transform ion cyclotron resonance (FT-ICR) mass
spectrometer (Bruker Daltonic GmbH, Bremen) fitted with a nano-
ESI source (Apollo) and a resonance cell (infinity cell). The starting
CH3OH/CHCl3 (10:1) solutions contained the cyclochiral hosts
CP,S (1ꢂ 10ꢁ5 M), CM,R (1ꢂ 10ꢁ5 M) and the selected guest G
(6ꢂ 10ꢁ5 M). The resulting ions were transferred into the reso-
nance cell by using an accumulation hexapole and a system of
potentials and lenses. Finally, the ions were cooled by collisions
with Argon, pulsed into the cell through a magnetic valve. The
mass spectra of the ESI solutions are characterized by the almost
exclusive presence of the intense signals corresponding to the
proton-bound complexes [CꢀHꢀG]+. These noncovalently charged
aggregates were isolated by broad-band ejection of the accompa-
nying ions and were allowed to react with a neutral gas B
(B = C2H5NH2, iso-C3H7NH2, (R) or (S)-sec-C4H9NH2), present in the
cell at controlled pressure (from 2.0 ꢂ 10ꢁ9 to 4.5 ꢂ 10ꢁ8 mbar,
depending upon their reactivity). Comparison of the kinetic results
H
D
H3C
W
B:
W=H, CH3,C2H5
CH
NH2
Chart 1. Cyclochiral Resorcinarenes (C) and neutral amine (B) used in the
gas-phase investigaton.
It is well known that naked cyclochiral macrocycles may have
either a C4 (crown) or a C2 (boat) symmetry and that resorcinarene
C is a C4-symmetrical macrocycle[25] with a chiral concave surface
suitable for noncovalent interactions with selected polar guests. It
should be noted that the resorcinarene conformation may be
somewhat influenced in solution by the polarity of the medium.
The crown⇄boat conformational equilibrium is shifted to the
right in going from aprotic to protic solvent because the intramo-
lecular hydrogen bond network stabilizing the C4 symmetry is
partially disrupted by intermolecular hydrogen bonding with the
protic molecules of the solvent.[26] The orientation of guests on
the upper aromatic cavity of C should be further favored by the
presence of a second but apolar cavity on the opposite side
formed by the very long aliphatic chains. Their terminal-CX3
methyl groups (X = H, D) allow the use of quasi-racemic mixtures
of the cyclochiral hosts, thus enabling the simultaneous measure-
ments of the reactivity of the corresponding quasi-diastereomeric
obtained with the quasi-diasteromeric [CP,S ꢀHꢀG]+/[CM,R ꢀHꢀG]+
H
D
mixture with the reversed combination [CP,S ꢀHꢀG]+/[CM,R ꢀHꢀG]+
safely excludes any isotope effect on the kinetics of Eqn (1). Each
kinetics has been repeated not less than three times to evaluate
the experimental error.
D
H
ESI-CID experiments
Collision-induced dissociation (CID) experiments were performed
on a Micromass Quattro LCZ (Waters-Micromass, Manchester, UK)
equipped with a nano-electrospray ionization (nano-ESI) source.
Operating conditions of the nano-ESI source are as follows: capil-
lary voltage = +1.29 kV, cone = 41 V, collision energy = 14.2 eV,
and source block temperature = 60 ꢃC. Methanolic solutions were
introduced via a nanospray needle. CID experiments were
performed in the following way: after isolation in the first quadru-
pole, the precursor ion was transferred to the collision chamber,
where it was fragmented by collisions with Ar. The resulting ions
were analyzed in the second quadrupole. The nominal pressure
of argon in the collision chamber was 8.0 ꢂ 10ꢁ4 mbar. The
relative abundance of fragments results from the area of peaks
of the spectra acquired in profile mode. Multiple scans were
accumulated until a satisfying s/n was achieved.
H
D
complexes, (e.g., [CP,S ꢀHꢀG]+ and [CM,R ꢀHꢀG]+) (Eqn (1)).[27]
Â
Ã
Â
Ã
CP;S ꢀHꢀG þ þ B ! CP;S ꢀHꢀB þ þ G
(1a)
H
H
Â
Ã
Â
Ã
CM;R ꢀHꢀG þ þ B ! CM;R ꢀHꢀB þ þ G
ð1bÞ
D
D
EXPERIMENTAL SECTION
Materials
D
The CP,SD/CM,R racemate was synthesized as reported in
Scheme 1 and purified by recrystallization.[28–32] The deuteration
level of the racemate was measured by means of ESI-mass spec-
trometry technique as exceeding 99%. The CP,SD/CM,RD racemate,
Computational details
For more rapid calculations, the decamethylene lateral chains
were replaced by methyl groups and the obtained adducts were
optimized using the TURBOMOLE suit program[33] (version 6.02).
For every diastereoisomer, 8–10 different structures were fully
optimized, and their total energies were compared. Noteworthy,
the adduct structures with lowest total energies are indeed
stabilized by a H–O, H–N, and H–p hydrogen bonds network.
Such approach does not ensure finding true global minima but
provides several indicative conformations, which possess lowest
total energy values and hence should strongly contribute into
the distribution of conformational isomers. The BP86 DFT
as well as the CP,SH/CM,R one, were resolved by reaction with
H
freshly prepared (S)-(+)-10-camphorsulfonyl chloride, K2CO3, and
dimethylamphetamine in a THF/acetonitril mixture (2:3 v:v) to
yield the functionalized diastereomers, subsequently separated
by silica gel high-performance liquid chromatography. The dia-
stereomeric excess was determined by means of NMR spectros-
copy as >99%. The monocamphorsulfonates diastereomers were
then hydrolyzed under alkaline conditions into the enantiomeri-
cally pure resorcin[4]arenes (ꢁ)-CM,RH and (+)-CP,SH and their deu-
D
D
terated quasi-enantiomeric counterparts (ꢁ)-CM,R and (+)-CP,S
.
J. Mass. Spectrom. 2012, 47, 72–79
Copyright © 2012 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/jms

C. Fraschetti et al.
Table 1. Investigated Guests, G
Free AA
Aminoalcohol
R
R’
L-phenylalanine
H
H
A1
L-tyrosinol
N1
3,4-dihydroxy-L-phenylalanine
A2
(1R,2R)-2-amino-1-phenyl-1,3-propandiol
N2
L-tryptophan
H
H
H
A3
A4
A5
L-epinephrine
N3
5-hydroxy-L-tryptophan
L-norepinephrine
N4
L-tyrosine
AA ester
L-phenylalanine ethyl ester
C2H5
CH3
E1
L-tyrosine methyl ester
E2
O
(b)
(a)
CD3
CD3
OTos
H
9
9
9
+
HO
(c)
OMe
CMD,R
CPD,S
+
D
Scheme 1. Synthesis of the labeled quasi-racemate of CM,R and CP,SD. (a) 1.9 eq. d3-MeMgI (1.0 M in Et2O), 3 mol% CuCl2 and 12 mol% 1-phenyl
propyne in THF, 12 h (95% isolated yield), (b) 3.6 mol% RuCl3 (H2O)3, 2.5 eq. NaIO4, MeCN/H2O (92:8 v:v), 16 h (62% isolated yield), (c) 2 eq. BF3ꢀOEt2
in dichloromethane, ꢁ10 ꢃC, 1 h (56% isolated yield).
wileyonlinelibrary.com/journal/jms
Copyright © 2012 John Wiley & Sons, Ltd.
J. Mass. Spectrom. 2012, 47, 72–79

Cyclochirality in the gas phase
functional[34,35] was used in combination with the standard SV(P)
basis sets. The Resolution of the Identity algorithm[36] implemen-
ted in the TURBOMOLE packet[37] was employed for the high
calculation efficiency. Every optimized structure was modified
to increase the number of hydrogen bonds in the adduct and
was then re-optimized. The procedure was repeated until no
more lowering total energy was observed. The most favored
structures were re-optimized using the same functional and the
TZVP extended basis sets (TZV basis sets[38] of triple-zeta quality
plus one p-function set for hydrogen, TZV plus one d-function
set for all the other atoms). Finally, to take into account the
electronic dispersion effects, we re-optimized the most favored
structures using Grimme’s B97-D functional[39,40] and the above-
mentioned TZVP basis sets. The vibrational frequencies were
calculated at the RI-BP86/TZVP and RI-B97-D/TZVP levels of
theory, deriving gradients numerically.[41] The relative energy (ΔE)
magnitudes were computed using zero-point energy (ZPE)-
corrected total energy values, whereas the relative free Gibbs
energy magnitudes (ΔG) were obtained from the same total energy
values corrected by chemical potential (CP) correction values (Table
S5 in Supporting Information). The default scaling factor (0.9914)
was used for the ZPE and CP correction values. All the thermo-
dynamic parameters were computed using freeh routine for
T = 298.15 K and P = 0.1MPa. The calculated structures were
presented using the visual molecular dynamics program set.[42]
Eqn (1). The relevant second-order constants (k) are calculated as
k = k’/[B]. Comparison of k with the corresponding thermal
capture collision rate kcap provides an estimate of the reaction
efficiency (eff= 100k/kcap; Supporting Information).[43] The effect
of the resorcinarene’s cyclochirality is described by the ratio
r = khomo/khetero, where khomo refers to the rate constant
measured when the host and the guest have the same chirality
and khetero to that measured when the host and the guest have
opposite nominal configuration. The obtained enantioselectivity
ratios r = khomo/khetero are summarized in Table 2 (the individual
rate constants are reported in Tables S1–S3). It should be noted that
the enantioselectivity of the competing reactions 1 increases as the
r values diverges from unity. Control experiments have been
performed by employing the glycine ethyl ester as an achiral
guest, where no appreciable deviation of r from unity has been
obtained ([CP,S ꢀHꢀG]+/[CM,R ꢀHꢀG]+ r = 0.98 ꢄ 0.06 and [CP,S ꢀHꢀG]+/
H
D
D
H
[CM,R ꢀHꢀG]+ r = 1.00 ꢄ 0.06). Ligand exchange reactions of
CH3CH2NH2 on [CꢀHꢀG]+ complexes of N2, N3, and N4 do not occur
under the used experimental conditions. As shown in Table 2,
most of the enantioselectivity values fall at r ≤ 1.00, thus
indicating that the heterochiral complexes react faster than the
homochiral homologues. The opposite behavior (r > 1.00) is
observed for the [CꢀHꢀG]+ complexes with G = A4, N2, N4. In
Fig. 1, an instance of representative mass spectra describing the
reaction of [CꢀHꢀN2]+ towards iso-propylamine has been reported:
as the delay time increases, the diastereoisomer precursor ions
disappear and the exchanged product is formed with different
efficiency, depending on the configuration of C. Besides, taking into
account that the experimental uncertainty associated with the
measured kinetics does not exceed ꢄ10%, the comparison of the
Kinetic results
1
According to H-NMR evidence, competitive solvation prevents
the formation of neutral [CꢀG] aggregates (in CDCl3 or CDCl3/
CD3OD solutions; Supporting Information). Thus, the proton-
bound [CꢀHꢀG]+ complexes are essentially generated in the evap-
orating nano-ESI nanodroplets. The kinetic profile of reactions 1
was analyzed by plotting the log([CꢀHꢀG]t+/[CꢀHꢀG]0+) versus the
reaction time, where [CꢀHꢀG]t+ is the precursor abundance at time
t and [CꢀHꢀG]+0 is the sum of the [CꢀHꢀG]t+ and the [CꢀHꢀB]+
complexes. All the kinetic plots exhibit a mono-exponential
decay with an excellent correlation (0.998< r2 < 0.999), which
could suggest the predominance of a single [CꢀHꢀG]+ structure or
the coexistence of structures that react with undistinguishable k.
The slopes of the mono-exponential kinetic plots provide a mea-
sure of the corresponding pseudo-first order rate constants (k’) of
[CP,S ꢀHꢀG]+/[CM,R ꢀHꢀG]+ and [CP,S ꢀHꢀG]+/[CM,R ꢀHꢀG]+ pairs
confirms that their reactions 1 are not affected by significant
isotope effects, except for G = N1 with B = iso-C3H7NH2 (Table S4;
KIE = 0.74).[44]
H
D
D
H
Theoretical results
To explain the origin of the measured enantioselectivity, we
performed a theoretical analysis to elucidate the structural features
of the [CꢀHꢀG]+ noncovalent adducts. In this view, some prelimi-
nary aspects should be discussed: (1) the actual location of the
proton in the [CꢀHꢀG]+ complexes, whether preferentially bound
Table 2. khomo/khetero measured for reactions of [CHꢀHꢀG]+/[CDꢀHꢀG]+ complexes with B
W = H
W = CH3
W = C2H5 (R)-
W = C2H5; (S)-
P.A.
G
210
212.5
214.1
214.1
khomo/khetero (r)
A1
A2
A3
A4
A5
E1
0.99 ꢄ 0.07
0.39 ꢄ 0.01
0.56 ꢄ 0.02
1.10 ꢄ 0.04
0.93 ꢄ 0.06
0.95 ꢄ 0.06
1.31 ꢄ 0.08
0.95 ꢄ 0.05
0.77 ꢄ 0.03
0.82 ꢄ 0.04
0.54 ꢄ 0.03
3.94 ꢄ 0.23
1.02 ꢄ 0.04
1.57 ꢄ 0.09
1.10 ꢄ 0.08
0.94 ꢄ 0.04
1.08 ꢄ 0.07
1.44 ꢄ 0.09
0.99 ꢄ 0.06
0.82 ꢄ 0.08
0.84 ꢄ 0.05
0.76 ꢄ 0.03
2.65 ꢄ 0.38
0.78 ꢄ 0.03
1.32 ꢄ 0.08
0.70 ꢄ 0.05
2.03 ꢄ 0.11
0.53 ꢄ 0.03
0.59 ꢄ 0.03
0.74 ꢄ 0.03
0.80 ꢄ 0.04
0.47 ꢄ 0.02
3.89 ꢄ 0.29
0.89 ꢄ 0.04
1.51 ꢄ 0.09
E2
N1
N2
N3
N4
P.A. values are reported in kcal molꢁ1 and have been by taken from a web source.[46]
J. Mass. Spectrom. 2012, 47, 72–79
Copyright © 2012 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/jms

C. Fraschetti et al.
protonation site is consistent with the ESI-CID experiments on
some representative [CꢀHꢀG]+ complexes, which gave [HꢀG]+ as
unique fragmentation product, thus pointing to the amino group
of the guest as the most basic site within the complexes. The
aliphatic OH is engaged in O-Hguest –O-CH3host hydrogen bond-
ing, whereas the aromatic hydroxyl group is preferably located
outside the cavity. The corresponding couple of diastereoiso-
meric structures of [CꢀHꢀN2]+ aggregates is shown in Fig. 3. For
the homochiral adduct, the ammonium group is again inside
the host cavity, whereas in the heterochiral [CꢀHꢀN2]+, the ammo-
nium group is shifted towards the corner of the host possessing
the crown conformation. In both cases, there is a hydrogen
bond network involving two O-Hguest –O-CH3host interactions.
DISCUSSION
Figure 1. Time dependent mass spectra of the reaction of [CꢀHꢀN2]+
complexes towards iso-propylamine.
The effect of the amine configuration ((R)-(ꢁ)-2-C4H9NH2 = k(R) vs
(S)-(+)-2-C4H9NH2 = k(S)) on reactions 1 is inferred from the rele-
vant x = k(R)/k(S) ratios reported in Tables S1–S3. The observation
that the measured x values, in most cases, significantly diverges
from unity strongly indicates that the rate-determining step of
reaction involves a direct interaction between 2-C4H9NH2 and the
diastereoisomeric scaffold of the complex. Because the B-to-G
displacement requires a prototropic transfer between the two
species, it is concluded that the chiral amine B must approach
the proton-bound complex frontside by developing an effective
interaction of the latter with the chiral upper rim of the host before
releasing the neutral guest.
to the C or to the G moieties, and (2) the actual location of guest
G in the proton-bound [CꢀHꢀG]+ complexes. The optimized
structures of adducts of simplified rccc-2,8,14,20-tetra-methyl-
4,10,16,22-tetra-O-methylresorcin[4]arene (C’) with the more
efficiently enantiorecognized guests (N1, N2) are shown in
Figs. 2–3. In the most favored structures of the N1 adducts,
the guest is located in the upper chiral rim of the host; mean-
while, its -NH+3 charged head is located exactly in the center of
the basket, engaged in NH–p multiple interactions. This predicted
Figure 2. Optimized structures of the diastereomeric heterochiral and homochiral [C’ꢀHꢀN1]+ complexes.
Figure 3. Optimized structures of the diastereomeric heterochiral and homochiral [C’ꢀHꢀN2]+ complexes.
wileyonlinelibrary.com/journal/jms
Copyright © 2012 John Wiley & Sons, Ltd.
J. Mass. Spectrom. 2012, 47, 72–79

Cyclochirality in the gas phase
In principle, the enantioselectivity r values of Table 2 may
essentially reflect the different stability of diastereomeric [CꢀHꢀG]+
complexes, coupled with quasi-isoenergetic transition structures
(thermodynamic enantioselectivity). Alternatively, the measured
r values are due to different transition structures stability
connecting quasi-degenerate diastereomeric [CꢀHꢀG]+ complexes
to products (kinetic enantioselectivity). Of course, the experimen-
tally observed kinetic scenario may reflect a combination of these
possibilities.
their relative stability as well as that of the transition structures
involved across the reaction coordinate.
CONCLUSION
The interactions between a cyclochiral resorcin[4]arene and repre-
sentative bidentate and tridentate guests have been explored in
the gas phase. The results represent the first example of kinetic
enantiodiscrimination exclusively due to the structural asymmetry
of a macrocycle scaffold. Mostly, the heterochiral adducts react
faster than their homochiral counterparts, with few exceptions
(G = A4; N2; N4). In the most enantioselective system, the guest is
an aromatic aminodiol, thus suggesting that basically, the guest
does need two hydroxyl groups to selectively interact with the
upper region of the chiral scaffold and one basic amino group to
keep the charge and behave as the effective hook by sticking on
the aromatic complex counterparts by cation–p interactions. This
structural arrangement is confirmed by DFT calculations. The
pronounced effect of the neutral gas proton affinity unambigu-
ously indicates that the major enantioselectivity reason is the
intimate interaction developed in the diastereomeric transition
structures. This view is supported by CID of the three-body
Analysis of Tables S1–S3 reveals that in all cases, the reaction
efficiency (eff) for a given [CꢀHꢀG]+ complex increases by increas-
ing the proton affinity of B and by decreasing that of the coordi-
nated guest G (in the order N ! E ! AA).[46] Furthermore, in all
virtually examined cases, the increase of the B proton affinity,
which should make the relevant transition structures more similar
to the reactant [CꢀHꢀG]+ structures than to the product [CꢀHꢀB]+, is
accompanied by the tendency of enantioselectivity r terms to
the unity. On these grounds, it is unlikely that the measured
enantioselectivities simply reflect the different stability of diaste-
reomeric [CꢀHꢀG]+ complexes, but rather they are thought to arise
from the different stability of the transition structures on the
reaction pathway to products. This conclusion finds an indepen-
dent support from the relative stability of the diastereomeric
[CꢀHꢀG]+ complexes obtained by collision-induced dissociation
[CM,R ꢀCP,S ꢀHꢀN1]+ complexes that points to the same stability for
X
Y
X
Y
diastereomeric [CM,R ꢀHꢀN1]+ and [CP,S ꢀHꢀN1]+ complexes. Never-
(CID) of the [CM,R ꢀCP,S ꢀHꢀG]+ (G = N1; N2; X,Y = H,D or D,H) ions.
X
Y
theless, the CID result obtained for the [CM,R ꢀCP,S ꢀHꢀN2]+ adduct
indicates the combination of thermodynamic and kinetic factors
as the origin of the FT-ICR selectivity, the less stable homochiral
species being even the more reactive one.
X
Y
CID of [CM,R ꢀCP,S ꢀHꢀG]+ (X,Y = H,D or D,H) is expected to yield
X
Y
the corresponding [CM,R ꢀHꢀG]+ or [CP,S ꢀHꢀG]+ (X,Y = H,D or D,H)
fragments after loss of the other host molecule. Although it is
well recognized that Cook’s method is not particularly suited for
large and polydentate systems, wherein collisional dissociation
may involve not negligible activation barriers and entropy
factors, nevertheless, in this specific case, it can be assumed that
the differential ΔΔS{ = ΔSh{omo-ΔS{hetero contributions are negligi-
ble, as related to the loss of two enantiomeric forms of the same
X
Y
Acknowledgements
Dr. Heinrich Luftmann (Org. Chem. Inst. der Universität Münster)
is gratefully acknowledged for the ESI-CID experiments. The
financial support was provided by the Deutsche Forschungsge-
meinschaft (SFB 613).
host. It follows that the relative abundance of the [CM,R ꢀHꢀG]+
H
and [CP,S ꢀHꢀG]+ (or [CM,R ꢀHꢀG]+ and [CP,S ꢀHꢀG]+) fragments
D
D
H
from CID of their [CM,R ꢀCP,S ꢀHꢀG]+ precursors can represent a
X
Y
Supporting information
reliable estimate of their relative stability. In this frame, the
X
Y
diastereomeric [CM,R ꢀHꢀN1]+ and [CP,S ꢀHꢀN1]+ complexes can
Supporting information may be found in the online version of
this article.
be considered as nearly degenerate because of their almost
equal abundance from CID of [CM,R ꢀCP,S ꢀHꢀN1]+ (X,Y = H,D or D,H)
X
Y
([CM,R ꢀHꢀN1]+/[CP,S ꢀHꢀN1]+ = 1.15ꢄ 0.12; Fig. S1a). In contrast, the
X
Y
REFERENCES
X
Y
homochiral [CM,R ꢀHꢀN2]+ and the heterochiral [CP,S ꢀHꢀN2]+
[1] F. Hollfelder, A. J. Kirby, D. S. Tawfik. Off-the-shelf proteins that rival
tailor-made antibodies as catalysts. Nature 1996, 383, 60.
[2] J. A. Loo. Studying noncovalent protein complexes by electrospray
ionization mass spectrometry. Mass Spectrom. Rev. 1997, 16, 1.
[3] J. A. Loo. Electrospray ionization mass spectrometry: a technology
for studying noncovalent macromolecular complexes. Int. J. Mass
Spectrom. 2000, 200, 175.
[4] A. J. Heck, R. H. H. Van den Heuvel. Investigation of intact protein
complexes by mass spectrometry. Mass Spectrom. Rev. 2004, 23, 368.
[5] A. Filippi, N. Trout, A. P. Brunelle, W. Adcock, T. S. Sorensen, M. Speranza.
Origin of diasterofacial selectivity in tertiary 2-adamantyl cations. J. Am.
Chem. Soc. 2001, 123, 6396.
[6] A. Filippi, N. Trout, A. P. Brunelle, W. Adcock, T. S. Sorensen, M. Speranza.
Importance of entropy in the diastereoslectivity of 5-substituted
2-methyladamant-2-yl cations. J. Org. Chem. 2004, 69, 5537.
[7] a) Y. Liang, J. S. Bradshaw, D. V. Dearden. The thermodynamic basis
for enantiodiscrimination: gas-phase measurement of the enthalpy
and entropy of chiral amine recognition by dimethyldiketopyri-
dino-18-crown-6. J. Phys. Chem. A 2002, 106, 9665;
fragments arising from CID of [CM,R ꢀCP,S ꢀHꢀN2]+ (X,Y= H,D or D,H)
X
Y
exhibit a relative abundance significantly below unity ([CM,
ꢀHꢀN2]+/[CP,S ꢀHꢀN2]+ = 0.72 ꢄ 0.10)[47] (Fig. S1b). This result is in
X
R
Y
full agreement with the more pronounced reactivity of the
homochiral [CM,R ꢀHꢀN2]+ complexes, relative to that of
X
Y
the heterochiral [CP,S ꢀHꢀN2]+ one, towards the same amine B in
the FT-ICR experiments. The calculated relative energies of the
[C’ꢀHꢀN1]+ diastereomeric adducts (Fig. 2; ΔE = Ehomo-Ehetero = 0.12
kcal molꢁ1) are very close, whereas the corresponding values for
[C’ꢀHꢀN2]+ differ more significantly (ΔE = 1.22kcal/mol).[48] This
result apparently correlate well with the CID of [CM,R ꢀCP,S ꢀHꢀN1]+
X
Y
X
Y
and [CM,R ꢀCP,S ꢀHꢀN2]+. Therefore, the same stability of the
diastereomeric [CꢀHꢀN1]+ adducts indicates that the enantioselec-
tivity values measured for this system mostly reflect the relative free
Gibbs energy values of the corresponding transition structures. On
the other hand, the combination of CID and kinetic results
strongly supports the view that the enantioselectivity mea-
sured for the diastereomeric [CꢀHꢀN2]+ adducts reflect either
b) W. A. Tao, R. G. Cooks. Chiral analysis by MS. Anal. Chem.. 2003, 75,
25A and references therein.
[8] a) S. Starnes, D. M. Rudkevich, J. Rebek. Cavitand-porphyrins. J. Am.
Chem. Soc. 2001, 123, 4659; b) U. Lücking, F. C. Tucci, D. M. Rudkevich,
J. Mass. Spectrom. 2012, 47, 72–79
Copyright © 2012 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/jms

C. Fraschetti et al.
J. Rebek. Self-folding cavitands od nanoscale dimensions. J. Am. Chem.
Soc. 2000, 122, 8880; c) D. M. Rudkevich, J. Rebek. Depening cavitands.
J. Eur. Org. Chem. 1999, 1991; d) D. M. Rudkevich. Nanoscale molecular
contaniners. Bull. Chem. Soc. Jpn. 2002, 75, 393.
[27] M. Gangoda, R. K. Gilpin, J. Figueirinhas. Deiterium nuclear magnetic
resonance studies of alkyl-modified silica. J. Phys. Chem. 1989, 93, 4815.
[28] M. Sawada. Chiral recognition in host-guest complexation deter-
mined by FAB mass spectrometry. J. Mass Spectrom. Soc. Jpn. 1997,
45(3), 439.
[29] X. Quin, T. Tzvetkov, X. Liu, D. C. Lee, L. Yu, D. C. Jacobs. Site-selective
abstraction in the reaction of 5–20 eV O+ with a self-assembled
monolayer. J. Am. Chem. Soc. 2004, 126, 13232.
[30] J. Terao, H. Todo, S. A. Begum, H. Kuniyasu, N. Kambe. Copper-
Catalyzed cross-coupling reaction of grignard reagents with primary-
alkyl halides: remarkable effect of 1-phenylpropyne. Angew. Chem.
Int. Ed. 2007, 46, 2086.
[9] D. J. Cram, J. M. Cram. Container Molecules and Their Guests. The
Royal Society of Chemistry: Cambridge; 1994.
[10] a) T. Heinz, D. M. Rudkevich, J. Rebek. Pairwise selection of guests in
a cylindrical molecular capsule of nanometre dimension. J. Nature
1998, 394, 764; b) D. M. Rudkevich. Calixarenes 2001; c) Z. Asfari,
V. Böhmer, J. Harrowfield, J. Vicens (Eds). Kluwer: Dordrecht
The Netherlands. 2001, 155; (d) F. Hof, S. L. Craig, C. Nuckolls,
J. Rebek. Molecular encapsulation. Angew. Chem. Int. Ed. 2002, 41,
1488.
[31] M. Gaddis, R. Ellis, G. T. Currie. Girard T Reagent for Carbonyls. Nature
1961, 191, 1391.
[11] C. Reuter, R. Schmieder, F. Vőgtle. From rotaxanes to knots. Templat-
ing, hydrogen bond patterns, and cyclochirality. Pure Appl. Chem.
2000, 12, 2233.
[12] A. Szumna. Cyclochiral conformers of resorcin[4]arenes stabilized by
hydrogen bonds. Org. Biomol. Chem. 2007, 5, 1358.
[32] M. Klaes, C. Agena, M. Köhler, M. Inoue, T. Wada, Y. Inoue, J. Mattay.
First synthesis, isolation and characterization of enantiomerically
pure and inherently chiral resorc[4]arenes by Lewis acid cyclization of
a resorcinol monoalkyl ether. Eur. J. Org. Chem. 2003, 8, 1404.
[33] R. Buckley, P. C. Bulman Page, Y. Chan, H. Heaney, M. Klaes,
M. J. McIldowie, V. McKee, J. Mattay, M. Mocerino, E. Moreno,
B. W. Skelton, A. H. White. The preparation and absolute configurations
of enantiomerically pure C4-symmetric tetraalkoxyresorcin[4]arenes
obtained from camphorsulfonate derivatives. Eur. J. Org. Chem. 2006,
22, 5135.
[13] a) A. Dalla Cort, C. Mandolini, L. Pasquini, L. Schiaffino. “Inherent
chirality” and curvature. New J. Chem. 2004, 28, 1198; b) Y. S. Zheng,
J. Luo. Inherently chiral calixarenes: a decade’s review. J. Incl. Phenom.
Macrocycl. Chem. 2011, DOI 10.1007/s10847-011-9935-4.
[14] M. Vincenti, A. Irico. Gas-phase interactions of calixarene and
resorcinarene cavitands with molecular guests studied by mass
spectrometry. Int. J. Mass Spectrom. 2002, 214, 23 and references
therein.
[15] M. W. Makinen, P. Vainiotalo, M. Nissinen, K. Rissanen. Ammonium
ion mediated resorcarene capsules: ESI-FT-ICRMS study on gas-
phase structure and ammonium ion affinity of tetraethyl resorcarene
and its per-methylated derivative. J. Am. Soc. Mass Spectrom. 2003,
14, 143.
[16] M. W. Makinen, P. Vainiotalo, K. Rissanen. Alkali metal mediated
resorcarene capsules: an ESI-FTICRMS study on gas-phase structure
and cation binding of tetraethyl resorcarene and its per-methylated
derivative. J. Am. Soc. Mass Spectrom. 2002, 13, 851.
[17] M. Vincenti, A. Irico, E. Dalcanale. Adv. Mass Spectrom. 1998, 14,
Chaps. 7/129 7/150, and references therein.
[34] R. Ahlrichs, M. Baer, M. Haeser, H. Horn, C. Koelmel. Electronic struc-
ture calculations on workstation computers: the program system
turbomole. Chem. Phys. Lett. 1989, 162, 165.
[35] A. D. Becke. Density-functional exchange-energy approximation
with correct asymptotic behavior. Phys. Rev. A 1988, 38, 3098.
[36] J. P. Perdew. Density-functional approximation for the correlation energy
of the inhomogeneous electron gas. Phys. Rev. B. 1986, 33, 8822.
[37] a) B. I. Dunlap, J. W. Conolly, J. R. Sabin. On some approximations in
applications of Xa theory. J. Chem. Phys. 1979, 71, 3396; b) O. Vahtras,
J. Almlöf, M. W. Feyereisen. Integral approximations for LCAO-SCF
calculations. Chem. Phys. Lett. 1993, 213, 514; K. Eichkorn, O. Treutler,
H. Öhm, M. Häser, R. Ahlichs. Auxiliary basis sets to approximate
Coulomb potentials. Chem. Phys. Lett. 1995, 240, 283.
[38] K. Eichkorn, O. Treutler, H. Ohm, M. Haser, R. Ahlrichs. Auxiliary basis
sets to approximate Coulomb potentials. Chem. Phys. Lett. 1995, 242,
652; R. Ahlrichs, M. Arnim. Methods and Techniques in Computa-
tional Chemistry, MET ECC-95 (ed.) E. Clementi, G. Corongiu, STEF:
Cagliari; 1995, 509.
[39] A. Schaefer, C. Huber, R. Ahlrichs. Fully optimized contracted Gauss-
ian basis sets of triple zeta valence quality for atoms Li to Kr. J. Chem.
Phys. 1994, 100, 5829.
[18] M. Vincenti, E. Dalcanale. Host–guest complexation in the gas phase.
Investigation of the mechanism of interaction between cavitands
and neutral guest molecules. J. Chem. Soc. Perkin Trans. 1995, 2, 1069.
[19] B. Botta, I. D’Acquarica, L. Nevola, F. Sacco, Z. V. Lopez, G. Zappia,
C. Fraschetti, M. Speranza, A. Tafi, F. Caporuscio, M. C. Letzel, J. Mattay.
Bis(diamido)-bridged basket resorcin[4]arenes as enantioselective
receptors for amino acids and amines. Eur.J. Org. Chem. 2007, 5995,
and the references therein.
[40] S. Grimme. Semiempirical GGA-Type Density functional constructed with
a long-range dispersion correction. J. Comput. Chem. 2006, 27, 1787.
[41] S. Grimme, J. Antony, T. Schwabe, C Mück-Lichtenfeld. Density
functional theory with dispersion corrections for supramolecular
structures, aggregates, and complexes of (bio)organic molecules.
Org. Biomol. Chem. 2007, 5, 741.
[42] P. Deglmann, K. May, F. Furche, R. Ahlrichs. Nuclear second analytical
derivative calculations using auxiliary basis set expansions. Chem.
Phys. Lett. 2004, 384, 103; P. Deglmann, K. May, F. Furche, R. Ahlrichs.
An efficient implementation of second analytical derivatives for den-
sity functional methods. Chem. Phys. Lett. 2002, 362, 511; P. Deglmann,
K. May, F. Furche. Efficient characterization of stationary points on
potential energy surfaces. J. Chem. Phys. 2002, 117, 9535.
[43] VMD for WIN-32, Version 1.8.3 (Februar, 15, 2005); see also W. Humpfrey,
A. Dalke, K. Schulten. VMD: visual molecular dynamics. J. Mol. Graphics.
1996, 14, 33.
[20] B. Botta, A. Tafi, F. Caporuscio, M. Botta, L. Nevola, I. D’Acquarica,
C. Fraschetti, M. Speranza. Modelling amphetamine/receptor inter-
actions: a gas-phase study of complexes formed between amphet-
amine and some chiral amido[4]resorcinarenes. Chem. Eur. J. 2008,
14, 3585.
[21] B. Botta, C. Fraschetti, F. R. Novara, A. Tafi, F. Sacco, L. Mannina,
A. P. Sobolev, J. Mattay, M. C. Letzel, M. Speranza. Interactions of
vinca alkaloids with chiral amido[4]resorcinarenes:
a dynamic,
kinetic, and spectroscopic study. Org. Biomol. Chem. 2009, 7, 1798,
and the references therein.
[22] B. Botta, C. Fraschetti, I. D’Acquarica, M. Speranza, F. R. Novara,
J. Mattay, M. C. Letzel. Gas-phase enantioselectivity of chiarl N-linked
peptidoresorc[4]arene isomers toward dipeptides. J. Phys. Chem. A
2009, 113, 14625, and the references therein.
[23] A. Mehdizadeh, M. C. Letzel, M. Klaes, C. Agena, J. Mattay. Chiral
discrimination on the host-guest-complexation of resorc[4]arenes
with quarternary amines. Eur. J. Mass Spectrom. 2004, 10, 649.
[24] R. Buckley, J. Y. Boxhall, P. C. Bulman Page, Y. Chan, M. R. J. Elsegood,
H. Heaney, K. E. Holmes, M. J. McIldowie, V. McKie, M. J. McGrath,
M. Mocerino, A. M. Poulton, E. P. Sampler, B. W. Skelton, A. H. White.
Mannich and O-alkylation reactions of tetraalkoxyresorcin[4]arenes –
the use of some products in ligand-assisted reactions. Eur. J. Org. Chem.
2006, 22, 5117.
[25] M. J. McIldowie, M. Mocerino, B. W. Skelton, A. H. White. Facile Lewis
acid catalyzed synthesis of C4 symmetric resorcinarenes. Org. Lett.
2000, 24, 3869.
[26] C. Schiel, G. A. Hembury,V. V. Borovkov, M. Klaes, C. Agena, T. Wada,
S. Grimme, Y. Inoue, J. Mattay. New insight into the geometry of
resorc[4]arenes: solvent-mediated supramolecular conformational
and chiroptical control. J. Org. Chem. 2006, 71, 976.
[44] a)T. Su, W. J. Chesnavitch. Parametrization of the ion–polar molecule
collision rate constant by trajectory calculations. J. Chem. Phys. 1982,
76(10), 5183; b) T. Su. Erratum: trajectory calculations of ion-polar
molecule capture rate constants at low temperatures [J. Chem.
Phys. 1988, 88, 4102]. J. Chem. Phys. 1988, 89(8), 5355.
[45] Kinetic isotope effect expressed as kH/kD.
[46] The used gas phase basicities have been taken from the http://
webbook.nist.gov/chemistry/ site.
[47] For instance ΔPA = PAethanolamine-PAglycine = 10.4 kcal molꢁ1 (http://
webbook.nist.gov/chemistry/) and ΔPA = PAmethyl ester-PAglycine = 3.8
kcal molꢁ1 (M. J. Locke, R. T. McIver Jr. Effect of solvation on the
acid/base properties of glycine. J. Am. Chem. Soc. 1983, 105, 4226).
[48] The diastereomeric excess found for [CP,S ꢀHꢀN2]+ adduct corresponds to
Y
the relative free energy difference of only approximately 0.2 kcal molꢁ1
.
wileyonlinelibrary.com/journal/jms
Copyright © 2012 John Wiley & Sons, Ltd.
J. Mass. Spectrom. 2012, 47, 72–79