Induced-fit in the gas phase: Conformational effects on the enantioselectivity of chiral tetra-amide macrocycles

Article abstract of DOI:10.1021/ja073287+

The structure, stability, and reactivity of proton-bound diastereomeric [M?H?A]⁺ complexes between some amino acid derivatives (A) and several chiral tetra-amide macrocycles (M) have been investigated in the gas phase by ESI-FT-ICR and ESI-ITMS-CID mass spectrometry. The displacement of the A guest from the diastereomeric [M?H?A]⁺ complexes by reaction with the 2-aminobutane enantiomers (B) exhibits a distinct enantioselectivity with regards to the leaving amino acid A and, to a minor extent, to the amine reactant B. The emerging selectivity picture, discussed in the light of molecular mechanics calculations, provides compelling evidence that the most stable conformers of the selected chiral tetraamide macrocycles M may acquire in the gas phase a different conformation by induced fit on complexation with some representative amino acid derivatives A. This leads to the coexistence in the gas phase of stable diastereomeric [M?H?A] ⁺ eq-eq and ax-ax structures, in proportions depending on the configuration of A and M and characterized by different stability and reactivity toward the 2-aminobutane enantiomers. The enantioselectivity of the gas-phase A-to-B displacement in the diastereomeric [M?H?A]⁺ complexes essentially reflects the free energy gap between the homo- and heterochiral [M?H?A]⁺ complexes, except when the tetra-amidic host presents an additional macrocycle generated by a decamethylene chain. In this case, the measured enantioselectivity mostly reflects the stability difference between the relevant diastereomeric transition structures.

Full text of DOI:10.1021/ja073287+

Published on Web 12/21/2007
Induced-Fit in the Gas Phase: Conformational Effects on the
Enantioselectivity of Chiral Tetra-Amide Macrocycles
Francesco Gasparrini, Marco Pierini, Claudio Villani,* Antonello Filippi, and
Maurizio Speranza*
Dipartimento degli Studi di Chimica e Tecnologia delle Sostanze Biologicamente AttiVe,
UniVersita` La Sapienza, 00185 Roma, Italy
Received May 9, 2007; E-mail: maurizio.speranza@uniroma1.it
Abstract: The structure, stability, and reactivity of proton-bound diastereomeric [M•H•A]+ complexes
between some amino acid derivatives (A) and several chiral tetra-amide macrocycles (M) have been
investigated in the gas phase by ESI-FT-ICR and ESI-ITMS-CID mass spectrometry. The displacement of
the A guest from the diastereomeric [M•H•A]+ complexes by reaction with the 2-aminobutane enantiomers
(B) exhibits a distinct enantioselectivity with regards to the leaving amino acid A and, to a minor extent, to
the amine reactant B. The emerging selectivity picture, discussed in the light of molecular mechanics
calculations, provides compelling evidence that the most stable conformers of the selected chiral tetra-
amide macrocycles M may acquire in the gas phase a different conformation by induced fit on complexation
with some representative amino acid derivatives A. This leads to the coexistence in the gas phase of
stable diastereomeric [M•H•A]+ eq-eq and ax-ax structures, in proportions depending on the configuration
of A and M and characterized by different stability and reactivity toward the 2-aminobutane enantiomers.
The enantioselectivity of the gas-phase A-to-B displacement in the diastereomeric [M•H•A]+ complexes
essentially reflects the free energy gap between the homo- and heterochiral [M•H•A]+ complexes, except
when the tetra-amidic host presents an additional macrocycle generated by a decamethylene chain. In
this case, the measured enantioselectivity mostly reflects the stability difference between the relevant
diastereomeric transition structures.
dextrins,6 and calixarenes.7 Macrocyclic amides form another
group of synthetic receptors that have been given considerable
Introduction
Shape-specific noncovalent attractive and repulsive interac-
tions constitute the basis for information transfer between
molecules in living systems as well as in synthetic supramo-
lecular structures. Despite ever more accurate descriptions of
biological systems and a dramatically increasing diversity of
synthetic host/guest complexes, a systematic and general
understanding of the underlying intermolecular forces is still
in its infancy. Organic chemistry makes possible the targeted
construction of infinitely variable macrocyclic compounds with
multiple centers interacting in complementary ways.1 Their
remarkable selectivity toward a variety of neutral and ionic
guests illustrates the principle of molecular recognition and is
the basis of many applications.2
attention. Interest in this category of macrocyclic hosts comes
from the amphiphilic properties of their amido groups in dipolar
or H-bonding interactions, the carbonyl acting as a dipole donor
and a H-bond acceptor and the N-H as a dipole acceptor and
a H-bond donor. Indeed, a number of recent publications have
presented convincing evidence that, depending on the experi-
mental conditions, macrocyclic amides can bind cations or
anions.8 In general, it is thought that the complexation process
will be energetically more favorable if the preferred conforma-
(4) Lehn, J.-M. Acc. Chem. Res. 1978, 11, 49.
(5) (a) Cui, C.; Cho, S. J.; Kim, S. J. Phys. Chem. A 1998, 102, 1119. (b) Lee,
W. Y.; Park, C. H.; Kim, S. J. Am. Chem. Soc. 1993, 115, 1184.
(6) Szejtli, J.; Osa, T. ComprehensiVe Supramolecular Chemistry; Cyclodex-
trines; Atwood, J. L., Davies, J. E. D., MacNicol, D. D., Vo¨gtle, F., Eds.;
Elsevier: Oxford, U.K., 1996; Vol. 3.
Perhaps the most widely known and extensively investigated
macrocycles are crown ethers,3 cryptands,4 starands,5 cyclo-
(7) (a) Mandolini, L.; Ungaro, R. Calixarenes in Action; Imperial College
Press: London, U.K., 2000. (b) Perrin, M.; Oehler. In Calixarenes, a
Versatile Class of Macrocyclic Compounds; Vicens, J., Bo¨hmer, V., Eds.;
Kluwer Academic Publishers: Dordrecht, The Netherlands, 1991.
(8) (a) Kim, H. Y.; Calabrese, J.; McEwen, C. J. Am. Chem. Soc. 1996, 118,
1545. (b) Kim, K. S.; Cui, C.; Cho, S. J. Phys. Chem. B 1998, 102, 461.
(c) Malinowska, E.; Wro´blewski, W.; Ostaszewski, R.; Jurczak, J. Pol. J.
Chem. 2000, 74, 701. (d) Chmielewski, M.; Jurczak, J. Tetrahedron Lett.
2004, 45, 6007. (e) Szczepan´ska, A.; Sałan´ski, P.; Jurczak, J. Tetrahedron
2003, 4775. (f) Hossain, Md. A.; Llinares, J. M.; Powell, D.; Bowman-
James, K. Inorg. Chem. 2001, 40, 2936. (g) Hossain, Md. A.; Kang, S. O.;
Powell, D.; Bowman-James, K. Inorg. Chem. 2003, 42, 1397. (h) Suh, S.
B.; Cui, C.; Son, H. S.; U., J. S.; Won, Y.; Kim, K. S. J. Phys. Chem. B
2002, 106, 2061. (i) Chmielewski, M.; Szumna, A.; Jurczak, J. Tetrahedron
Lett. 2004, 45, 8699. (j) Chmielewski, M.; Jurczak, J. Chem.-Eur. J. 2005,
11, 6080.
(1) (a) Lehn, J.-M. Supramolecular Chemistry; VCH: Weinheim, Germany,
1995. (b) Schneider, H.-J.; Yatsimirsky, A. Principles and Methods in
Supramolecular Chemistry; Wiley: Chichester, U.K., 2000. (c) Steed, J.
W.; Atwood, H. L. Supramolecular Chemistry; Wiley: Chichester, U.K.,
2000. (d) Dietrich, B.; Viout, P.; Lehn, J.-M. Macrocyclic Chemistry;
VCH: Weinheim, Germany, 1993.
(2) (a) Schneider, H. J. Angew. Chem., Int. Ed. Engl. 1991, 30, 1417. (b) Seel,
C.; Vo¨gtle, F. Angew. Chem., Int. Ed. Engl. 1992, 31, 528. (c) Izatt, R.
M.; Pawlak, K.; Bradshow, J. S.; Bruening, R. L. Chem. ReV. 1991, 91,
1721.
(3) (a) Pedersen, C. J. J. Am. Chem. Soc. 1967, 89, 2495. (b) Pedersen, C. J.
Angew. Chem., Int. Ed. Engl. 1988, 27, 1021. (c) Glendening, E. D.; Feller,
D.; Thompson, M. A. J. Am. Chem. Soc. 1994, 116, 10657. (d) Glendening,
E. D.; Feller, D. J. Am. Chem. Soc. 1996, 118, 6052.
9
522
J. AM. CHEM. SOC. 2008, 130, 522-534
10.1021/ja073287+ CCC: $40.75 © 2008 American Chemical Society

Induced-Fit in the Gas Phase
A R T I C L E S
Chart 1.a
a *Configurational descriptors for chiral macrocycles 1-3 are abbreviated in the text as either r or s, meaning the configuration at the four stereogenic
carbons are either R,R,R,R or S,S,S,S.
tion of the macrocycle in the free state resembles closely the
conformation it will adopt upon interaction with the ionic guest.
Thus, an important research endeavor is the study of the
conformational equilibria of macrocyclic amide receptors and
their sensitivity to the noncovalent interactions with potential
guests.9
We were recently engaged in a preliminary mass spectro-
metric (MS) study of the chiral recognition of some representa-
tive amino acid derivatives (A), i.e., phenylalanine amide
(pheNH2), phenylalanine ethyl ester (pheOEt), and naphthyla-
lanine ethyl ester (naphOEt) (Chart 1), by several specifically
designed macrocyclic tetra-amides (M), i.e., r-1, s-1, and r-2,
in the gas phase where interference from the solvent and the
counterion is safely excluded.10 The enantioselectivity of the
selected M hosts was checked by introducing into a Fourier
transform ion cyclotron resonance mass spectrometer (FT-ICR-
MS), equipped with an electrospray ionization source (ESI), the
proton-bonded two-body complexes [M•H•A]+ and by measur-
ing the rate of the displacement reaction 1 where B is either
(R)-(-)- (BR) and (S)-(+)-2-butylamine (BS).
surprisingly is the observation of a single structure for the
heterochiral [r-1•H•s-pheOEt]+ complex, whereas its homochiral
[r-1•H• r-pheOEt +
]
congener displays a reaction kinetics con-
sistent with the coexistence of two stable [1•H•r-pheOEt +
]
fast
and [1•H•r-pheOEt +
] slow structures. No clear-cut explanation for
such a spectacular configurational effect is presently available.
This intriguing kinetic picture prompted us to widen the
investigation to the macrocyclic tetra-amides s-1D, r-3, and s-3
and to a larger set of amino acidic guests (Chart 1). The purposes
of this research is to: (i) verify the effects of the functional
groups of the guest A on the enantioselectivity of the gas-phase
reaction 1; (ii) elucidate the structural features of the diaster-
eomeric [M•H•A]+ complexes; (iii) explain the origin of the
spectacular configurational effect on the reactivity of isomeric
forms of [M•H•A]+ (A ) pheOEt and naphOEt); and (iv) shed
some light on the dynamics and the mechanism of reaction 1.
Results and Discussion
All of the experiments concerning the gas-phase thermody-
namics and kinetics of the selected [M•H•A]+ complexes have
been carried out by using two complementary mass spectro-
metric techniques. Indications about the nature of proton bonding
in the ESI-formed diastereomeric [M•H•A]+ complexes have
been derived from their collision-induced decomposition (CID)
performed in a ion-trap mass spectrometer.11-16 With the same
instrument, information on the relative stability of diastereomeric
[M•H•A]+ + B f [M•H•B]+ + A
(1)
According to the relevant reaction kinetics, a single structure
is attributed to the diastereomeric [M•H•A]+ (M/A ) r-1/
pheNH2 and r-2/pheOEt) complexes. In contrast, both the hetero-
[r-1•H•s-naphOEt + and the homochiral [s-1•H•s-naphOEt +
]
]
adducts behave as a combination of two stable, non-intercon-
verting isomeric forms, one (henceforth marked with the fast
subscript) much more reactive toward B than the other
(henceforth marked with the slow subscript). Even more
(11) Cooks, R. G.; Kruger, T. L. J. Am. Chem. Soc. 1977, 99, 1279.
(12) McLuckey, S. A.; Cameron, D.; Cooks, R. G. J. Am. Chem. Soc. 1981,
103, 1313.
(13) Cooks, R. G.; Patrick, J. S.; Kotiaho, T.; McLuckey, S. A. Mass Spectrom.
ReV. 1994, 13, 287.
(14) Cooks, R. G.; Koskinen, J. T.; Thomas, P. D. J. Mass Spectrom. 1999, 34,
85.
(9) Parra, R. D.; Yoo, B.; Welmhoff, M. J. Phys. Chem. A 2006, 110, 4487.
(10) Filippi, A.; Gasparrini, F.; Pierini, M.; Speranza, M.; Villani, C. J. Am.
Chem. Soc. 2005, 127, 11912.
(15) Armentrout, P. B. J. Am. Soc. Mass Spectrom. 2000, 11, 371.
(16) Dang, T. T.; Pedersen, S. F.; Leary, J. A. J. Am. Soc. Mass Spectrom.
1994, 5, 452.
9
J. AM. CHEM. SOC. VOL. 130, NO. 2, 2008 523

A R T I C L E S
Gasparrini et al.
Table 1. Enantioselectivity Factors R from CID of Diastereoisomeric [(M)2•H•A]+ Adducts
Enantioselectivity Factor, R
Guest A
thrOEt
host
(M)
alaOEt
valOEt
leuOEt
n-valOEt
serOEt
phgOEt
pheOEt
fpheOEt
mpheOEt
naphOEt
r-1 1.93 ( 0.04 2.87 ( 0.03 3.13 ( 0.06 5.00 ( 0.01 3.00 ( 0.06 2.34 ( 0.05 2.35 ( 0.05 4.52 ( 0.09 3.07 ( 0.06 3.29 ( 0.07 10.00 ( 0.09
r-2 0.80 ( 0.02 1.18 ( 0.02 1.14 ( 0.02 1.10 ( 0.02 1.03 ( 0.02 1.19 ( 0.02 1.00 ( 0.02 1.81 ( 0.04 1.60 ( 0.03 1.48 ( 0.03 2.43 ( 0.05
r-3 1.37 ( 0.03 1.98 ( 0.04 1.61 ( 0.03 3.40 ( 0.07 1.19 ( 0.02 1.83 ( 0.04 1.56 ( 0.03 2.07 ( 0.04 2.22 ( 0.04 2.14 ( 0.04 1.51 ( 0.03
[M•H•A]+ was obtained through CID of the corresponding
[M2•H•A]+ aggregates.17-19 The kinetic enantioselectivity of
the exchange reaction 1 was determined by using an ESI-FT-
ICR mass spectrometer. The enantioselectivity factors were
obtained by comparing the rate constants of eq 1 when using
the [M•H•A]+ diastereoisomers with a given 2-butylamine (e.g.,
BS) or, alternatively, by using a given [M•H•A]+ diastereoisomer
with the 2-butylamine enantiomers. In the next sections, these
results will be summarized and integrated with those coming
from dedicated molecular modeling and docking studies.
Gas-Phase Stability of Diastereomeric [M•H•A]+ Com-
plexes. Information about the relative stability of the homochiral
and heterochiral [M•H•A]+ complexes has been derived from
CID of the corresponding three-body [(M)2•H•A]+ adducts.
Their CID patterns are characterized by the exclusive formation
of the relevant [M•H•A]+ and [(M)2•H]+ fragments in propor-
tions which depend on the relative basicity of the M/A and M/M
adducts, the configuration of both the guest A and the host M,
and the collision energies. At equal collision energies (see
Experimental Section), the relative stability of the homochiral
and heterochiral [M•H•A]+ complexes from CID of the relevant
[(M)2•H•A]+ adducts provides an expression of the enantiose-
lectivity factor R of [M•H]+ toward A (the subscripts hetero/
homo refer to the relative configuration of the guest and the
host in the complexes):17-19
not produce any effect on the selectivity. With aromatic A
guests, the enantioselectivity factor R, measured with M ) r-1,
appears to strongly increase with the size of the π-systems (cfr.
A ) pheOEt and A ) naphOEt) and with the chain flexibility
(cfr. A ) phgOEt and A ) pheOEt). These factors play only a
minor role with M ) r-2 and r-3. The presence of a para-
substituent on the aromatic ring of A does not modify ap-
preciably the enantioselectivity (cfr. A ) pheOEt, fpheOEt, and
mpheOEt). The relative stability of diastereomeric [M•H•A]+
complexes can be evaluated from the experimental R factors of
Table 1 in terms of the following equation: ∆∆GCID ) (∆Ghetero
- ∆Ghomo CID ) RTeff ln R, at the “effective temperature” Teff,
)
which is an empirical parameter identified as the temperature
of a canonical ensemble of clusters for which fragmentation
would yield the same branching ratios as observed experimen-
tally.20
Gas-Phase Reactivity of Diastereomeric [M•H•A]+ Com-
plexes. As pointed out above, the R factors of Table 1 reflect
the relative stability of the heterochiral vs the homochiral
[M•H•A]+ fragments, and therefore, they express the thermo-
dynamic enantioselectivity of the chiral [M•H]+ hosts toward
the chiral A guests, measured at the undefined “effective
temperature” Teff. Now, we are aimed at verifying whether the
thermodynamic stability of [M•H•A]+ is the only factor
controlling its reactivity toward the amine B (eq 1). In other
words, to what extent the relative stability of the diastereomeric
[M•H•A]+ complexes controls their enantioselectivity toward
B.
[M•H•A]+
+
(
)
[(M)2•H]
homo
R )
(2)
[M•H•A]+
Table 2 reports the ESI-FT-ICR kinetic parameters of the
exchange reaction 1, measured in present and previous10 studies.
The kinetics were carried out by monitoring the appearance of
the exchanged product [M•H•B]+ and the decay of the reactant
[M•H•A]+ as a function of time t. If I is the intensity of complex
[M•H•A]+ at the delay time t and I0 is the sum of the intensities
of [M•H•A]+ and [M•H•B]+, monoexponential ln(I/I0) vs t plots
were obtained for all the systems with M ) r-1 and A ) phe
and pheNH2, and with M ) r-2 and A ) phe, pheNH2, and
pheOEt. The good linearity of their decay curves (corr. coeff. r2
g 0.986, see Supporting Information) confirms the view that
the [M•H•A]+ complexes are thermally equilibrated when
reacting with B. The pseudo-first-order rate constants k′ of
reaction 1 were obtained from the slopes of the relevant ln(I/
I0) vs t linear plots. The corresponding second-order rate
constants k are calculated from the ratio between the slope of
the first-order plots and the B concentration (k ) k′/[B]). Their
values, compared with the relevant collision rate constants (kC),
estimated according to Su’s trajectory calculation method,21
provides directly the efficiency of the reaction (eff ) k/kC).
+
(
)
[(M)2•H]
hetero
Table 1 reports the enantioselectivity factors R for the
[M•H•A]+ complexes formed by combining all the r-M hosts
and most of the A guests shown in Chart 1. Stereochemical
preference is always in favor of the homochiral association (R
> 1), with the only exception of alaOEt with host 2. The R values
dramatically depend on the functional groups of both M and
A. In general, they increase in the following M order: r-2 <
r-3 < r-1, thus suggesting that the substituents bound to the
stereogenic carbons of M play a major role. Another significant
effect is attached to the decamethylene chain in 3, which
somewhat lowers its enantioselectivity relative to that of 1.
Comparison of the R factors of the complexes with A ) alaOEt
,
valOEt, leuOEt, and n-valOEt, as guests, reveals an increase in
enantioselectivity with the size of the aliphatic group of A. The
presence of the OH functionality in A ) serOEt and thrOEt does
(17) For a recent review, see: Speranza, M. Ad. Phys. Org. Chem. 2004, 39,
147.
(18) Yao, Z.; Wan, T. S. M.; Kwong, K.; Che, C. Chem. Commun. 1999, 20,
2119.
(19) Tao, W. A.; Zhang, D.; Wang, F.; Thomas, P.; Cooks, R. G. Anal. Chem.
1999, 71, 4427.
(20) Laskin, J.; Futrell, J. H. J. Phys. Chem. 2000, 104, 8829.
(21) Su, T. J. Chem. Phys. 1988, 88, 4102, 5355.
9
524 J. AM. CHEM. SOC. VOL. 130, NO. 2, 2008

Induced-Fit in the Gas Phase
A R T I C L E S
Table 2. Exchange Rate Constants (×10-11 cm3 Molecule-1 s-1
)
BR
)
(R)-(
−)- C4H9NH2
BS ) (S)-(+)- C4H9NH2
host (M)
guest (A)
k a
Fb
k a
Fb
êb
r-1
r-1
r-1
r-1
r-1
r-1
r-1
r-1
s-1D
s-1D
s-1D
s-1D
s-1
r-1
s-1
r-1
r-2
r-2
r-2
r-2
r-2
r-2
s-3
r-3
s-3
r-3
r-phe
77.5 ( 1.0 (0.68)
89.7 ( 1.5 (0.79)
33.6 ( 0.7 (0.30)
40.4 ( 0.7 (0.36)
26.3 ( 3.0 (0.24)
20.6 ( 0.3 (0.18)
4.0 ( 0.2 (0.04)
20.6 ( 0.3 (0.18)
26.2 ( 1.4 (0.23)
26.3 ( 0.7 (0.23)
4.44 ( 0.27 (0.039)
26.3 ( 0.7 (0.23)
1.22 ( 0.04 (0.010)
26.3 ( 1.3 (0.22)
59.3 ( 0.6 (0.52)
61.9 ( 0.8 (0.55)
42.1 ( 0.5 (0.37)
48.1 ( 1.1 (0.42)
25.2 ( 0.6 (0.22)
26.3 ( 0.3 (0.23)
3.6 ( 0.2 (0.03)
26.3 ( 0.3 (0.23)
23.0 ( 1.9 (0.20)
15.6 ( 1.9(0.14)
2.77 ( 0.15 (0.024)
15.6 ( 1.9(0.14)
1.25 ( 0.07 (0.011)
25.0 ( 2.5 (0.21)
0.058 ( 0.007 (0.0005)
1.20 ( 0.06(0.010)
62.1 ( 2.4(0.54)
62.9 ( 1.1 (0.55)
39.6 ( 0.5 (0.35)
42.5 ( 1.3 (0.37)
19.3 ( 0.6 (0.17)
31.0 ( 1.2 (0.27)
24.9 ( 0.6(0.21)
1.31 ( 0.03
1.45 ( 0.04
0.80 ( 0.02c
0.84 ( 0.03c
1.04 ( 0.17
0.78 ( 0.02
1.11 ( 0.12
0.78 ( 0.02
0.88 ( 0.12d
0.59 ( 0.09d
0.62 ( 0.06d
0.59 ( 0.09d
0.98 ( 0.09c
1.05 ( 0.18c
1.10 ( 0.18c
1.02 ( 0.04c
1.44 ( 0.08
1.03 ( 0.03
1.11 ( 0.07
0.99 ( 0.06
0.99 ( 0.04c
0.94 ( 0.05c
0.96 ( 0.05
0.60 ( 0.07
0.83 ( 0.14
1.15 ( 0.19
0.86 ( 0.03
0.83 ( 0.03
1.28 ( 0.18
0.19 ( 0.02
1.00 ( 0.08
0.17 ( 0.01
0.96 ( 0.02
0.87 ( 0.03
0.96 ( 0.02
0.14 ( 0.01
1.47 ( 0.35
0.18 ( 0.03
s-phe
r-pheNH
s-pheNH
2
2
r-pheOEt
s-pheOEt
r-pheOEt
s-pheOEt
s-pheOEt
r-pheOEt
s-pheOEt
r-pheOEt
s-naphOEt
s-naphOEt
s-naphOEt
s-naphOEt
r-phe
fast
slow
fast
slow
fast
fast
slow
slow
0.046 ( 0.004
0.052 ( 0.07
1.39 ( 0.07
1.05 ( 0.06
0.65 ( 0.02
1.15 ( 0.28
0.032 ( 0.005
0.050 ( 0.009
0.048 ( 0.09
0.99 ( 0.05
0.93 ( 0.06
0.62 ( 0.05
0.72 ( 0.06
0.044 ( 0.005
0.064 ( 0.003 (0.0005)
1.23 ( 0.09 (0.010)
89.6 ( 1.9(0.78)
s-phe
64.6 ( 1.8(0.56)
r-pheNH
s-pheNH
2
2
43.8 ( 1.4 (0.38)
41.9 ( 1.3 (0.37)
19.1 ( 0.2 (0.17)
29.2 ( 0.3 (0.26)
23.9 ( 4.3 (0.19)
20.8 ( 1.1 (0.18)
0.042 ( 0.003 (0.0004)
1.32 ( 0.12 (0.012)
r-pheOEt
s-pheOEt
s-naphOEt
s-naphOEt
s-naphOEt
s-naphOEt
fast
fast
slow
slow
34.7 ( 2.0 (0.29)
0.051 ( 0.003 (0.0004)
1.15 ( 0.09(0.010)
a k × 10-11 cm3 molecule-1 s-1; the values in parentheses represent the reaction efficiency expressed as the ratio between the measured rate constants and
b
the corresponding collision constant kC, calculated using the trajectory calculation method (Su, T.; Chesnavitch, W. J. J. Chem. Phys. 1982, 76, 5183).
F)khomo/khetero (referred to the configuration of M and A); ê ) kR/kS (referred to the configuration of B). c Reference 10. d ê ) kS/kR (referred to the configuration
of B).
Like the above-mentioned systems, the heterochiral [r-1•H•s-
suggest the presence of two persistent isomeric structures with
largely different reactivity toward B (see Supporting Informa-
tion).
pheOEt +
] complex follows a monoexponential kinetics when
reacting with the B enantiomers (corr. coeff. r2 g 0.994, see
Supporting Information). Contrariwise, its homochiral [r-1•H•r-
An estimate of the relative abundance of the isomeric forms,
responsible for the above biexponential kinetics, can be derived
from the deconvolution of the corresponding [MHA]+ decay
curve. Accordingly, the time dependence of the more reactive
isomer [M•H•A]+fast can be inferred from the overall [M•H•A]+
decay after subtracting the first-order decay of the less reactive
isomer [M•H•A]+slow. The Y-intercepts of the first-order decay
pheOEt +
] analogue follows a biexponential kinetics under the
same experimental conditions. As pointed out in related stud-
ies,10,22 this kinetic behavior conforms to the coexistence of two
stable isomeric structures for [r-1•H•r-pheOEt]+, one less
reactive ([r-1•H•r-pheOEt +
1•H•r-pheOEt +
exhibited by the heterochiral [r-1•H•s-pheOEt +
]
slow) and the other more reactive ([r-
fast). In contrast, the monoexponential kinetics,
complex, is
]
]
of [MHA]+
and [MHA]+
provide an estimate of the
slow
fast
attributed to the occurrence of a single structure or, alternatively,
of several stable isomers, but with comparable reactivity toward
B. The same rationale applies to the monoexponential kinetics
observed for the guest exchange reactions in the [r-1•H•s/r-
pheNH2]+ and [r-2•H•A]+ complexes.
relative abundance of the [MHA]+ isomers (Table 3). As
expected, the same [M•H•A]+fast/[M•H•A]+slow ≈ 0.6 distribu-
tion is observed for the homochiral [r-1•H•r-pheOEt +
and
]
[s-1D•H•s-pheOEt]+complexes. With the diastereomeric [1•H•s-
naphOEt]+ and [3•H•s-naphOEt]+ complexes, the [M•H•A]+
/
fast
[M•H•A]+
distribution is more unbalanced in favor of the
We sought a definitive check of the spectacular configura-
slow
tional effect observed with [r-1•H•pheOEt +
]
by replacing the
r-1 host with its hexadeuterated enantiomer s-1D. The reaction
are extremely
[M•H•A]+slow component([M•H•A]+fast/[M•H•A]+slow ≈0.1÷0.5).
Kinetic Enantioselectivity. Table 2 reports the enantiose-
lectivity of reaction 1 involving some representative [M•H•A]+
complexes. The reaction enantioselectivity is expressed by the
F ) khomo/khetero ratio, when referred to the configuration of the
A guest and the M host, or by the ê ) kR/kS one, when referred
to the configuration of the amine B. A F > 1 value indicates
that the amine B displaces the guest from the homochiral
complex faster that the guest from the heterochiral complex
diastereomeric [M•H•AD]+ and [M•H•AL]+ complexes. The
opposite is true when F < 1. A F ) 1 value corresponds to
equal displacement rates. Analogously, a ê > 1 value indicates
that the displacement of the A guest from a given [M•H•A]+
diastereomer is faster with the r-amine (BR) than with the s-one
kinetics of the corresponding [s-1D•H•pheOEt +
]
reassuring because, as expected, the heterochiral [s-1D•H•r-
pheOEt]+ complex displays a monoexponential kinetics, whereas
the homochiral [s-1D•H•s-pheOEt]+ one exhibits a biexponential
decay plot. A different picture comes from the decay of the
diastereomeric [1•H•s-naphOEt + and [3•H•s-naphOEt +
] ] com-
plexes by reaction with the enantiomers of B. Both diastereo-
meric adducts show in fact biexponential decay curves, which
(22) (a) Lebrilla, C. B. Acc. Chem. Res. 2001, 34, 653. (b) Wu, L.; Cooks, R.
G. Anal. Chem. 2003, 75, 678. (c) Gal, J. F.; Stone, M.; Lebrilla, C. B. Int.
J. Mass Spectrom. 2003, 222, 259. (d) Ahn, S.; Ramirez, J.; Grigorean,
G.; Lebrilla, C. B. J. Am. Soc. Mass Spectrom. 2001, 12, 278. (e) Grigorean,
G.; Lebrilla, C. B. Anal. Chem. 2001, 73, 1684.
9
J. AM. CHEM. SOC. VOL. 130, NO. 2, 2008 525

A R T I C L E S
Gasparrini et al.
Table 3. Percent Distribution of eq-eq and ax-ax Conformations for pheNH2, pheOEt, and naphOEt Guests in [M•H•A]+ Complexes
[M
•
H
•
A]+
[M
•
H
•
A]+
[M
•
H
•
A]+
[M
•
H
•
A]+
slow
∆∆
H
°
(fast-slow)a
fast
slow
fast
1
host (M)
guest (A)
[M
•H•
A]+
experimental
experimental
calculated (config.)
calculated (config.)
(kcal mol-
)
1
pheOEt
homochiral
heterochiral
homochiral
heterochiral
homochiral
heterochiral
homochiral
heterochiral
homochiral
heterochiral
homochiral
heterochiral
37 ( 2
63 ( 2
17 (ax-ax)
3 (eq-eq)
17 (ax-ax)
3 (eq-eq)
27 (ax-ax)
32 (eq-eq)
13 (eq-eq)
21 (eq-eq)
83 (eq-eq)
97 (ax-ax)
83 (eq-eq)
97 (ax-ax)
73 (eq-eq)
68 (ax-ax)
87 (ax-ax)
79 (ax-ax)
0.9
2.0
0.9
2.0
0.6
0.4
1.1
0.8
100
1D
1
pheOEt
35 ( 8
65 ( 8
100
100
100
pheNH
2
1
naphOEt
pheOEt
12 ( 3
29 ( 4
88 ( 3
71 ( 4
2
100
100
100 (eq-eq)
100 (eq-eq)
3
naphOEt
13 ( 2
34 ( 2
87 ( 2
66 ( 2
48 (eq-eq)
8 (eq-eq)
52 (ax-ax)
92 (ax-ax)
0.0
1.5
a Derived from the MM calculated [M•H•A]+fast and [M•H•A]+slow relative abundances (columns 6 and 7); ∆∆H°(fast-slow) ) ∆H°fast - ∆H°slow ) -RT
ln([M•H•A]+fast/[M•H•A]+slow), T ) 298 K.
Table 4. Experimental and Calculated Enantioselectivities for Selected Host-Guest Combinations
a
b
c
c
∆∆G#
∆∆GCID
∆∆
G
°
∆∆H°
th
FT
-
ICR
th
1
1
1
1
[M
•H•
A]+
(kcal mol-
)
(kcal mol-
)
(kcal mol-
)
(kcal mol-
)
+
[r-1•H•pheOEt
]
1.05
0.10
1.77
0.27
2.04
0.89
0.45
1.36
0.35
0.24
0.91
0.21
0.93
0.54
0.20
0.92
0.24
0.83
-0.25
1.29
0.09
0.03
+
[r-1•H•pheNH2
[r-1•H•naphOEt
]
+
]
+
[r-2•H•pheOEt
]
+
[r-3•H•naphOEt
]
d
rmsd FT-ICR/ th
rmsd CID/ th
0.95
0.35
d
a ∆∆G#
) ∆G#
- ∆G#
) -RT ln F (Table 2), T ) 298 K. b ∆∆GCID ) (∆Ghetero - ∆Ghomo
)
) RTeff ln R (Teff ) 298 K). c ∆∆X°th
)
CID
FT-ICR
(∆X°hetero - ∆X°homo
homo
hetero
)
th
(X ) G, H), calculated at T ) 298 K. d rmsd ) root-mean-square deviation.
(BS). Again, the opposite is true when ê < 1. A ê ) 1 value
corresponds to equal displacement rates.
between heterochiral [s-1D•H•r-pheOEt]+ complex and a given
B enantiomer (eff ) 0.23 (BR); 0.14 (BS)) must be compared
From the measured kinetic enantioselectivity factors F of
with that between [r-1•H•s-pheOEt +
]
and the opposite enanti-
omer of B (eff ) 0.23 (BS); 0.18 (BR)). Similarly, the reaction
efficiencies of the homochiral [s-1D•H•s-pheOEt +
fast (eff ) 0.23
(BR); 0.20 (BS)) and [s-1D•H•s-pheOEt +
slow structure (eff ) 0.04
(BR); 0.02 (BS)) closely correspond to those of their undeuterated
[r-1•H•r-pheOEt + and [r-1•H•r-pheOEt +
analogues (eff
Table 2, it is possible to derive the ∆∆G#
difference
FT-ICR
between the activation barriers for the displacement reaction 1
]
involving the homochiral and the heterochiral [M•H•A]+ clusters
]
(∆∆G#
) ∆G#
- ∆G#
) -RT ln F). These
FT-ICR
homo
hetero
quantities have been reported in Table 4 together with the
]
]
slow
fast
∆∆GCID ) (∆Ghetero -∆Ghomo CID ) RTeff ln R values, calculated
from the CID results of Table 1, by taking 298 K as the
“effective temperature” Teff.
)
) 0.22 (BS); 0.24 (BR), and eff ) 0.03 (BS); 0.04 (BR),
respectively). The coincidence of these efficiency values, while
reassuring about the accuracy and reproducibility of the
experimental ESI-FT-ICR results, indicates no significant OCX3
(X ) H,D) kinetic isotope effects involved in their reaction 1.
Both the more and less reactive isomers of the diastereomeric
The kinetic results of Table 2 confirm the view that the
efficiency of the gas-phase reaction 1 may depend on the
structure and the configuration of M, A, and B. The [r-1•H•A]+
(A ) phe and pheNH2) complexes exhibit a small enantiose-
lectivity as regard to both the A (0.8 < F < 1.0) and the B
configuration (0.8 < ê < 1.4). Replacing r-1 with r-2 as host
does not modify the picture appreciably (0.9 < F < 1.4 and 1.0
< ê < 1.4). In contrast, a more pronounced enantioselectivity,
referred to the A configuration, is observed in the reaction with
[1•H•s-naphOEt +
complexes display the same exceptional
]
enantioselectivity, referred to the A configuration (F ) 0.050),
and a negligible selectivity toward the B enantiomers (1.0 < ê
< 1.1). A somewhat different picture is observed by replacing
1 with 3 as hosts. Here, the efficiency of reaction 1 involving
the diastereomeric [3•H•s-naphOEt +
]
structures are almost
fast
[r-2•H•pheOEt +
]
(F ≈ 0.6), whereas no effect of the B
identical to that measured for the heterochiral [r-1•H•s-
configuration is detected (0.9 < ê < 1.0). As pointed out before,
the decay curve of the heterochiral [r-1•H•s-pheOEt]+ complex
reflects the occurrence of a single stable structure with a reaction
efficiency of 0.18 (BR) and 0.23 (BS)). It is interesting to note
that the reaction efficiency of the more reactive isomeric
structure of the homochiral [r-1•H•r-pheOEt]+complex (eff )
0.24 (BR); 0.22 (BS)) is comparable to that of the single isomeric
structure evidenced for the heterochiral [r-1•H•s-pheOEt]+ one.
The heterochiral [s-1D•H•r-pheOEt]+ complex shows the same
reaction efficiency (eff ) 0.23 (BR); 0.14 (BS)) of its undeu-
terated [r-1•H•s-pheOEt]+ analogue (eff ) 0.18 (BR); 0.23 (BS)).
It should be considered that the efficiency of the reaction
naphOEt + and [s-1D•H•r-pheOEt +
]
]
fast
ones (eff ≈ 0.2-0.3).
fast
The reaction enantioselectivity of the [3•H•s-naphOEt +
]
fast
structures, relative to the A configuration (0.72 < F < 1.15), is
significantly lower than that measured with the [1•H•s-
naphOEt +
]
homologues (F ) 0.050). A reversed order is
fast
observed for the less reactive [M•H•s-naphOEt +
]
(M ) 1,
slow
3) structures. In fact, 0.032 < F < 0.044 for [3•H•s-
naphOEt +slow, whereas F ) 0.050 for [1•H•s-naphOEt +
Similarly to [1•H•s-naphOEt +
[3•H•s-naphOEt +fast and [3•H•s-naphOEt +
selectivity toward the B enantiomers (0.6 < ê < 1.1). The
reduced enantioselectivity of the [3•H•s-naphOEt +
fast structures
]
]
.
slow
]
fast and [1•H•s-naphOEt +
slow exhibit a limited
] slow, both
]
]
]
9
526 J. AM. CHEM. SOC. VOL. 130, NO. 2, 2008

Induced-Fit in the Gas Phase
A R T I C L E S
Figure 1. Relevant minimum energy structures of the r-1, r-2, and r-3 hosts found by conformational search.
(F ) 1.15 ( 0.28 (BR); 0.72 ( 0.06 (BS)) parallels the
moderating effects of the decamethylene chain in 3 on the
enantioselectivity factor R measured by ESI-MS-CID experi-
ments. Furthermore, the presence of the hydrocarbon chain has
no dramatic effects on the reaction efficiency, except for the
(eqs 3-5) and are collected in Table 4.
∆H°R/S
)
H °R/S × pi
(3)
∑
i
i)AD
∆S°R/S ) -
p × ln(p )
(4)
(5)
[s-1•H•s-naphOEt +
]
(eff ) 0.01) and [s-3•H•s-naphOEt +
]
∑
i
i
fast
fast
i)AD
(eff ) 0.20) pair.
∆G°R/S ) ∆H°R/S - T∆S°R/S
Molecular Modeling. Experimental results have clearly
highlighted the ability of the tetra-amide macrocyclic hosts 1-3
to act as very effective selectors for the enantiomers of amino
acid derivatives. Structural variations on both hosts and guests
have sizable effects on enantioselectivity. To gain a better
understanding of the factors determining the enantioselectivity
of chiral hosts 1-3 toward the enantiomeric guests, we have
performed an in-depth theoretical investigation based on mo-
lecular modeling calculations. Owing to the quite large size,
flexibility, and complexity of the isolated hosts and their adducts,
a reasonably complete and homogeneous sampling of the
potential energy hypersurfaces related to the conformational
variability of the single species and to the intermolecular host-
guest interactions can only be obtained using computationally
non-demanding methods. Our approach uses molecular me-
chanic models, that are computationally light and provide an
excellent account of conformational energy differences for
organic compounds. First, preliminary exhaustive conforma-
tional searches and relevant analyses have been carried out on
structures of macrocycles M ) r-1, r-2, and r-3 and guests A
) pheOEt, pheNH2, and naphOEt by means of molecular
mechanics using the MM2*force field (see Experimental
Section). The more stable conformers of each species within a
suitable energy window have been selected as the representative
ensemble of their structure and employed to perform molecular
docking simulations of [M•H•A]+adducts, with guests in both
R and S configuration (r-1 with r/s-pheOEt, r/s-pheNH2 and r/s-
naphOEt; r-2 with r/s-pheOEt; r-3 with r/s-naphOEt). In turn,
each type of proton-bound two-body complex is represented
by an ensemble of adducts, again selected within an ap-
propriately wide energy window, that describes geometry and
stability of the complex as average properties. Standard
structural enthalpy, entropy, and free energy changes (∆X°hetero
and ∆X°homo, with X ) H, S, or G) for each complexation
process, and the corresponding differential terms between
diastereomeric complexes (∆∆X°th ) (∆X°hetero - ∆X°homo)th)
have been derived by the statistical treatment of energetic data
(AD ) number of adducts constituting the complex ensemble;
pi) Boltzmann population of the io adduct inside the considered
ensemble).
The geometry of the most stable proton-bound two-body
complexes has been re-optimized at the B3LYP/6-31G* level
of theory (see Supporting Information). Their inspection gave
useful information as regards to the protonation site and the
relative disposition/orientation of the components in each
supramolecular adduct. In addition, the relative basicity of the
host and the guest molecule within their supramolecular
complexes were studied by quanto-mechanic calculations.
Calculated Structure and Energetics of the Tetramidic
Hosts M. All the structures, obtained by the MM2* conforma-
tional searches of r-1, r-2, and r-3 within an energy window of
4.5 kcal mol-1, have been considered in the analysis. The
calculated geometries of r-1, r-2, and r-3 can be conveniently
clustered in ensembles of three types, according to the disposi-
tions assumed in each conformer by the alkyl or phenyl
fragments on two contiguous stereogenic carbons of the host,
denoted as C* in Chart 1. These geometries are classified as:
equatorial-equatorial (eq-eq), axial-axial (ax-ax), and axial-
equatorial (ax-eq) (Figure 1). Note that the cyclohexyl ring in
r-2 locks the position of the alkyl fragments on C* and only
the eq-eq disposition is possible. In both r-1 and r-3, the eq-
eq dispositions are energetically favored, with Boltzmann
populations (BP) amounting to 98.40% and 99.95%, respec-
tively. In all eq-eq geometries, the macrocyclic host presents
a C2-symmetric folded structure with a concave side and a
convex one (Figure 2). Alternate and diverging CO and NH
groups are located on the outer margins of the concave surface,
F1, whereas the same, but now converging functionalities are
placed on the central folding of the convex side, F2. Such
molecular framework, reminding the structure of a saddle roof,
is stabilized by two strong H-bonds between two facing amide
moieties. For the three hosts, an additional distinction regards
9
J. AM. CHEM. SOC. VOL. 130, NO. 2, 2008 527

A R T I C L E S
Gasparrini et al.
eq-eq, two ax-ax and one ax-eq), two for r-2 (open eq-eq
and closed eq-eq) and three for r-3 (open eq-eq and closed
eq-eq and one ax-ax).
Calculated Structure and Energetics of the Guest Mol-
ecules A. Molecular structures of the guests pheOEt, pheNH2,
and naphOEt have been modeled by starting the conformational
search from their protonated forms at the amino groups. During
the docking simulation, the protonated amino group of each
guest acts as a probe for the basic sites present in the
macrocyclic host. Conformers of each guest species for the
docking experiments were selected within a 3 kcal mol-1 energy
window among the large ensembles obtained by the conforma-
tional searches. To adequately simulate the high flexibility of
the guests, such minimal ensembles were built according to the
geometric criterion to warrant an exhaustive sampling of the
wider dihedral angles diversity of all the rotatable bonds. The
number of geometries considered in each cluster were 9 for
pheOEt (corresponding to 100% of BP), 5 for pheNH2 (corre-
sponding to 100% of BP) and 6 for naphOEt (corresponding to
71% of BP).
Figure 2. C2-symmetric saddle-roof structure of the selected r-1, r-2, and
r-3 hosts. Substituents on the stereogenic carbons are omitted for clarity.
the relevant eq-eq conformer ensembles. On the basis of the
extent of their surface folding, it is possible to distinguish open
and closed geometries (Figure 3) in which the angle formed by
the two isophthalic rings is either larger or smaller than 100°,
respectively. In the open structures, H-bonds are seen between
CdO and N-H groups located on the F2 side, whereas in the
closed structures these interactions are located on the F1 side.
In both r-1 and r-2, the open eq-eq conformation results
strongly favored over the closed one, while the reverse is true
in 3. This is certainly due to the decamethylene chain that
connects the isophthalic rings and forces them in a nearly
parallel disposition (Figure 1). Instead, within the less populated
ax-eq and ax-ax geometries of r-1 (1.6% of BP) and r-3
(∼0.1% of BP) and also in the quasi-planar geometries of 2 (of
negligible BP; Figure 1), the two isophthalic rings assume a
wide-open disposition with the angle between the rings close
to 150° in ax-ax conformers, and practically 180° in the other
cases (Figure 3). Also in these structures, however, the four
amidic fragments are complementarily arranged and establish
a couple of intramolecular H-bonds. Inspection of both r-1 and
r-2 molecular models clearly suggests that, in their eq-eq and
ax-ax geometries, the F1 and F2 surfaces may easily inter-
change by inversion of the macrocycle through rotation around
the aryl-CO bonds.
Relative Basicity of the Host and the Guest in the
[M•H•A]+ Complexes. In all the [M•H•A]+ systems investi-
gated, [M•H]+ is invariably the predominant, if not the exclusive
CID fragment observed. These findings indicate that all the
selected macrocyclic tetramides 1-3 of Chart 1 are significantly
more basic than all the listed A guests. However, this basicity
order is valid for M and A in their isolated gaseous state,
whereas no indication about the proton location in the [M•H•A]+
complexes may be derived by the experimental data. Protonation
sites in the [M•H•A]+ complexes have been therefore investi-
gated by considering the adduct structures coming from the
docking simulations both as such and after their further
optimization by ab initio DFT-calculations, at the B3LYP/6-
31G* level of theory (some of them have been re-optimized at
the B3LYP/6-311+G** level of theory without any appreciable
structural differences). In particular, optimizations were carried
out: (i) directly on the more stable simulated complexes
[r-1•H•r-pheOEt]+ and [r-1•H•s-pheOEt]+ having the host with
different conformations (eq-eq, ax-ax, and ax-eq) and in
which H+ is bound to the amino group of the guest; (ii) on the
same complexes previously modified by transferring the H+
from the protonated amino group of the guest to the opposite
Docking experiments were then performed on highly repre-
sentative conformers of the hosts molecules: six for r-1 (three
Figure 3. Models of the r-1, r-2, and r-3 macrocycles showing their different folding patterns. Substituents on the stereogenic carbons are omitted for
clarity.
9
528 J. AM. CHEM. SOC. VOL. 130, NO. 2, 2008

Induced-Fit in the Gas Phase
A R T I C L E S
carbonyl oxygen of the host. All the B3LYP/6-31G*-optimized
structures are reported as Supporting Information (Figure S1).
Inspection of these structures reveals that, irrespective of the
input protonation site of the complex (whether on the most basic
CO groups23 of M or on the amino group of A), after DFT
optimization the proton is invariably found on the amino group
of A. This suggests that the pheOEt molecule, which is less basic
than M in the isolated state, becomes more basic than M when
“solVated” by the polar enVironment of the host. The energetic
cost for the proton transfer from M to the amino group of A in
[M•H•A]+ is largely counterbalanced by the extra-stabilization
of the protonated A due to the establishment of multiple H+-
with the host in the minority closed conformation shows this
latter in the open form.
For the homochiral [r-1•H•r-pheOEt]+ and [s-1D•H•s-pheOEt]+-
complexes, MM calculations point to their eq-eq structures as
0.9 kcal mol-1 more stable than the corresponding ax-ax ones
(∆∆H°(fast-slow) in Table 3). The order is reversed in the
heterochiral [r-1•H•s-pheOEt]+ and [s-1D•H•r-pheOEt]+complexes,
whose eq-eq forms are 2.0 kcal mol-1 less stable than the
relevant ax-ax structures. For both [r-1•H•s/r-naphOEt +
] dia-
stereomers, the ax-ax form is more stable than the eq-eq
congener by 1.1 kcal mol-1, in the case of the homochiral
complex, and by 0.8 kcal mol-1, in the case of the heterochiral
one. The same stability order is calculated for the heterochiral
[r-3•H•s-naphOEt]+ complex (∆∆H°(fast-slow) ) 1.5 kcal mol-1),
while the eq-eq and ax-ax structures of its homochiral
congener are almost degenerate (Table 3).
+
bondings between its NH3 group and the amido carbonyls of
M. Obviously, when these H+-interactions are weakened in the
CID excited [M•H•A]+, such an extra-stabilization is largely
removed and the proton moves to M because of the reduced
basicity of the A guest.
The relative abundances, estimated for the more- and less-
stable couple of isomers of each complex, are compared in Table
3 with those derived from the experimental kinetic curves. At
first glance, the agreement between the two sets of data may
appear only partial and qualitative. However, a deeper inspection
of Table 3 may remove most of the apparent discrepancies
among the reported figures. The occurrence of two stable
Structure and Energetics of the [M•H•A]+ Complexes.
Supramolecular generation of some of the [M•H•A]+ adducts
studied in the present work was simulated by multiconforma-
tional molecular docking procedures, characterized by high-grid
sampling levels of both the host approaching surface and the
guest geometrical orientation.24 We modeled the following
diastereomeric pairs of complexes: [r-1•H•s/r-pheOEt]+, [r-1•H•s/
r-pheNH2]+, [r-1•H•s/r-naphOEt]+, [r-2•H•s/r-pheOEt]+, and
[r-3•H•s/r-naphOEt]+, with a number of adducts of each species
generated in the first step of the docking procedure equal to
763 776, 424 320, 509 184, 254 592, and 254 592, respectively.
Inspection of the more stable complexes shows that, irrespective
of the type and configuration of host and guest species, a pattern
of strong H-bonds is established between the two molecular
units at the center of the macrocycle convex side. In particular,
the protonated amino group of the guest is always H+-bonded
to the converging CdO groups present on the F2 surface. By
contrast, all the adducts formed with the guest approaching the
host from the concave surface are computed about 17 kcal mol-1
higher in energy. Interestingly enough, only the eq-eq and ax-
ax conformations of the macrocycles are found in the ensembles
of the more stable adducts and in proportions different from
those of the uncomplexed hosts. In particular, host r-1, that in
the free form is calculated to assume almost exclusively the
eq-eq conformation ([eq-eq BP] > 98%), acquires a predomi-
nant ax-ax geometry by induced fit on complexation with
s-pheOEt, s-pheNH2, and r/s-naphOEt guests (68%< [ax-ax BP]
<97%; Table 3). Among the minority eq-eq conformations,
the more stable ones show open geometries. Similarly, host r-3,
that is almost exclusively eq-eq in the free state (eq-eq BP ≈
100%), is calculated to assume a predominant ax-ax geometry
by induced fit on complexation with s-naphOEt ([eq-eq BP]
≈ 8%) and, to a lesser extent, on complexation with the
r-naphOEt ([eq-eq BP] ≈ 48%). Even in this case, the more
stable among the minority eq-eq conformations adopt only open
geometries. In contrast, host r-2 undergoes only a population
change of its own eq-eq geometry by induced fit on complex-
ation with the r/s-pheOEt guest enantiomers: after addition of
the guest, a significant fraction of the adducts generated starting
isomeric forms for the homochiral [r-1•H•r-pheOEt +
(and
]
[s-1D•H•s-pheOEt]+) complex accounts for its biexponential
decay with B. On the grounds of their relative stability, the most
stable eq-eq structure is associated with the less reactive
[r-1•H•r-pheOEt +
]
slow (and [s-1D•H•s-pheOEt +
]
slow) complex and
the less stable ax-ax structure to the more reactive [r-1•H•r-
pheOEt + (and [s-1D•H•s-pheOEt +
fast) one. In contrast, the
pronounced stability gap (2.0 kcal mol-1) between the ax-ax
(and
]
]
fast
and eq-eq forms of the heterochiral [r-1•H•s-pheOEt +
]
[s-1D•H•r-pheOEt]+) complex justifies the large predominance
of a single stable ax-ax isomer, which is responsible of the
observed monoexponential decay with B. An analogous rationale
can be advanced for the biexponential decay registered for the
diastereomeric [r/s-1•H•s-naphOEt +
diastereomeric complexes show a stable ax-ax structure,
associated with the less reactive [r/s-1•H•s-naphOEt +
slow com-
plexes, accompanied by a less stable eq-eq form, associated
with the more reactive [r/s-1•H•s-naphOEt +
isomers. The
] complexes. Indeed, both
]
]
fast
same stability order is observed with the heterochiral [r-3•H•s-
naphOEt +
] complex, whereas the ax-ax and eq-eq forms of
the homochiral [s-3•H•s-naphOEt]+ one are almost degenerate.
At variance with the heterochiral [r-1•H•s-pheOEt]+ and [s-1D•H•r-
pheOEt +
] complexes, the apparent monoexponential decay of
the diastereomeric [r-1•H•r/s-pheNH2]+ pair when reacting with
B is not ascribed to the large predominance of a single isomer,
but rather to the limited stability difference between their ax-
ax and eq-eq forms (0.4÷0.6 kcal mol-1; Table 3). If the
reactivity of the diastereomeric [r-1•H•r/s-pheNH2]+ pair toward
B is mainly determined by their thermodynamic stability (Vide
infra), their ax-ax and eq-eq isomeric forms should display
similar reaction 1 rate constants and, therefore, exhibit an
apparent monoexponential decay. The same conclusion would
be reached for the almost degenerate ax-ax and eq-eq forms
of the homochiral [s-3•H•s-naphOEt]+ complex, if their reactiv-
ity toward B were mainly determined by their thermodynamic
stability. However, the observation of a biexponential kinetics
for the reaction of the homochiral [s-3•H•s-naphOEt]+ complex
(23) Bagno, A. J. Phys. Org. Chem. 2000, 13, 574.
(24) (a) Alcaro, S.; Gasparrini, F.; Incani, O.; Mecucci, S.; Misiti, D.; Pierini,
M. and Villani, C. J. Comput. Chem. 2000, 21, 515. (b) Alcaro, S.;
Gasparrini, F.; Incani, O.; Caglioti, L.; Pierini, M.; Villani, C. J. Comput.
Chem. 2007, 28, 1119.;
9
J. AM. CHEM. SOC. VOL. 130, NO. 2, 2008 529

A R T I C L E S
Gasparrini et al.
[r/s-3•H•s-naphOEt]+ complexes (Table 2), express the kinetic
enantioselectivity of these systems in the guest exchange reaction
with B since essentially reflecting the stability difference of the
relevant diastereomeric transition structures.
Approach of the Amine B to the [M•H•A]+ Complexes.
As already mentioned, analysis of the F and ê factors arising
from the kinetic experiments involving the proton-bound
complexes between the structurally related hosts M ) 1 and 3
and the guest s-naphOEt indicates that the decamethylene chain
in 3 does not influence appreciably the efficiency of the A-to-B
displacement, except with the homochiral [s-M•H•s-naphOEt +
]
fast
isomer. This latter exception provides a further evidence of the
thermodynamic vs kinetic factors controlling the reactivity of
the homochiral [s-1•H•s-naphOEt + and [s-3•H•s-naphOEt +
]
]
complexes, respectively, whereas the reactivity of the hetero-
Figure 4. Full red line describes the relationship between the experimental
activation barriers of the slow reaction of several two-body [M•H•A]+
complexes with amine B (obtained from the relevant F values from Table
1; ∆∆G# ) ∆G#homo - ∆G#hetero ) -RT ln F) and the MM computed relative
chiral [r-1•H•s-naphOEt + and [r-3•H•s-naphOEt +
]
]
complexes
are invariably governed by thermodynamic factors. According
to conformational analysis, both hosts 1 and 3 in their eq-eq
and ax-ax conformations, are characterized by a saddle roof
shape that differentiates the two sides of the macrocycle.
Docking simulations show that all the guests add to the host on
the convex surface. Note that, for 3, this is the only possible
approach because the decamethylene chain hinders the approach
from the concave side.
enthalpies of the same [M•H•A]+ complexes (∆∆H°th ) (∆H°hetero
-
∆H°homo)th). The full blue line describes the relationship between the
experimental ∆∆G# ) ∆G# - ∆G#
) -RT ln F values and the
homo
hetero
relative stability of the same [M•H•A]+ complexes, calculated at Teff
298 K from the R values of Table 1 (∆∆GCID ) (∆Ghetero - ∆Ghomo
)
)
)
CID
RTeff ln R). If the effective temperature Teff is taken equal to 457 K (broken
blue line), both the slopes of the ∆∆G# vs ∆∆H°th and the ∆∆G# vs ∆∆GCID
straight lines coincide and are close to unit (slope ) 0.921). This means,
by one side, that MM computations correctly estimate the relative stability
of the selected [M•H•A]+ complexes and, by the other side, that the stability
gap of the diastereomeric [M•H•A]+ reactants (except [3•H•s-naphOEt]+)
is mainly responsible of the enantioselectivity F factors of Table 1. All the
In principle, the B amine may approach either the concave
or the convex sides of the [1•H•A]+ complexes to give the most
stable [1•H•B]+ product with B located on the convex side of
the host. In the first case, the host must undergo a macrocycle
inversion to put B on the new convex side and release A (a
backside displacement). In the second case, B approaches the
complex from the same side where A is located (a frontside
displacement), and therefore, the release of the guest A does
occur without any significant conformational changes of the
host. Clearly, the presence of the decamethylene chain in 3
prevents any backside displacement in favor of the frontside
one. That the same frontside approach takes place also with
the other [M•H•A]+ structures is suggested by performing a
dedicated docking experiment. The global minimum structure
points refer to reactions with BR, except that of [s-1D•H•pheOEt + with BS.
]
with B strongly suggests that this hypothesis might be not
acceptable in this case (Vide infra).
The experimental and calculated enantioselectivity factors for
hosts 1-3, listed in Table 4, are graphically summarized in
Figure 4. A reasonable linear correlation (red line in Figure 4;
r2 ) 0.961; slope ) 0.921; y-intercept ) -0.205) does exist
between the ∆∆H°th ) (∆H°hetero - ∆H°homo)th
and ∆∆G# )
∆G#homo - ∆G#hetero for all the [M•H•A]+ adducts investigated,
except [r/s-3•H•s-naphOEt]+. The linear relationship suggests
that the kinetic enantioselectivity, measured in the FT-ICR
experiments at 298 K, is essentially an expression of the stability
gap between the corresponding diastereomeric [M•H•A]+
reactants. This conclusion is further supported by the linear
of the [r-1•H•r-pheOEt +
complex has been employed in a
]
docking experiment with amine BR, to form the protonated
trimer [r-1•H•r-pheOEt•BR]+. The results confirm that the
approach of BR is energetically more favorable on the convex
side by about 3 kcal mol-1. Here, the BR places its amine group
just above the couple of converging CdO fragments on the F2
surface of the host, in a favorable disposition to realize the guest
exchange after receiving a proton from the facing NH3+ group
of the protonated aminoacid (Figure 5).
correlation between ∆∆GCID ) (∆Ghetero - ∆Ghomo CID
)
) RTeff
ln R (taken a Teff ) 298 K) and ∆∆G# ) ∆G# - ∆G#
homo
hetero
(blue solid line in Figure 4; r2 ) 0.971; slope ) 0.601;
y-intercept ) +0.290). Indeed, if Teff is taken as equal to 457
K, the ∆∆G# vs ∆∆GCID linear correlation becomes parallel to
that of the ∆∆G# vs ∆∆Hth one, though shifted upside by ca.
0.65 kcal mol-1 (blue broken line in Figure 4). The pronounced
deviation of the [r/s-3•H•s-naphOEt]+ systems from the relation-
ships of Figure 4 implies that the enantioselectivity, measured
in the FT-ICR experiments at 298K, reflects not only their
stability difference, but also the effects of the decamethylene
chain of the r-3 host on the relative stability of the diastereo-
meric transition structures of eq 1. This conclusion is consistent
with the observation of a biexponential kinetics for the reaction
of B with the homochiral [s-3•H•s-naphOEt]+ complex, despite
the degeneracy of the ax-ax and eq-eq forms of the homochiral
[s-3•H•s-naphOEt]+ complex. It is concluded that the F factors,
arising from the FT-ICR experiments with the diastereomeric
In this frame, the largely different efficiency observed in the
exchange reaction 1 on the homochiral [M•s-naphOEt
]
(M
fast
) s-1 (eff ) 0.01) and s-3 (eff ) 0.20)) isomers can be attributed
to the effects of the decamethylene chain in the open eq-eq
conformation of s-3 which may favor the disruption of intramo-
lecular H-bond interactions between the alternate converging
CO and NH functionalities placed on the convex F2 face of the
host in favor of intermolecular interactions with the B amine
approaching the same face.
The Chiral Recognition Model. The formation of the
selected [M•H•A]+ complexes is mainly driven by the establish-
ment of a pattern of three strong H-bonds between the
protonated amino group of the guest A and the couple of
9
530 J. AM. CHEM. SOC. VOL. 130, NO. 2, 2008

Induced-Fit in the Gas Phase
A R T I C L E S
+
Figure 5. Simulation by docking experiments of the r-pheOEt displacement from the [r-1•H•r-pheOEt
]
complex by BR. The most stable structure of the
protonated trimer [r-1•H•r-pheOEt•BR]+, obtained by docking procedure, is depicted in the center (LP is the lone-pair of BR).
+
Figure 7. Structures of global minima of the ax-ax [r-1•H•s-pheOEt
]
+
and [r-1•H•r-naphOEt
] complexes showing additional attractive edge-to-
face and face-to-face interactions between the aromatic moiety of the guest
and one isophthalic ring of the host.
thalic rings of the host. The magnitude of these interactions
depends on the nature and the configuration of the guest and is
somewhat tempered by repulsive interactions, as demonstrated
by the significant increase of the distance between the centroids
of the opposite phenyl rings of the host when the pheOEt guests
(8.80 Å with r-pheOEt and 8.97 Å with s-pheOEt) are replaced
by the larger naphOEt ones (9.50 Å with r-naphOEt and 9.45 Å
with s-naphOEt). This may account in part for the significant
MM-calculated stability difference between the [r-1•H•r-na-
phOEt]+ and [r-1•H•s-naphOEt]+ ax-ax structures (ER - ES )
-1.02 kcal mol-1, Figure 8a), which significantly reduces and
Figure 6. Structures of global minima of the [M•H•r-phe
]
OEt + and [M•H•r-
naphOEt
]
+ complexes having the host in eq-eq conformation showing the
additional H-bond between the carbonyl oxygen of the guest and one amide
N-H of the host.
converging CdO functionalities placed on the F2 surface of
the host M. The guest is placed above the center of the host
convex side and establishes further weaker interactions with the
peripheral zones of the host. In the eq-eq structures, an
additional H-bond is observed between the carbonyl oxygen of
the guest and one amide N-H of the host (Figure 6). In the
ax-ax structures with aromatic guests, additional attractive
edge-to-face and face-to-face interactions operate between the
aromatic moiety of the guest and one isophthalic ring of the
host (Figure 7). The extent to which the side-chain groups of
the guest either favor or hinder complexation with a given host,
depends on the stereochemistry of the guest, and on the size
and structure of the side chain itself. This is evident if we
consider the very different discrimination ability of host 1 toward
the enantiomers of pheOEt and naphOEt. We inspected eight
geometries of the minimum-energy complexes [R-1•H•R/S-
pheOEt]+ and [r-1•H•r/s-naphOEt]+ obtained by docking experi-
ments. Four of these are the more stable ones having the host
in ax-ax conformation (Figure 8a) and the other four are the
more stable with the host in eq-eq conformation (Figure 8b).
Figure 8a clearly shows that all the guests interact with host 1
in ax-ax conformation in a similar way, placing their aromatic
moiety in a face-to-face arrangement above one of the isoph-
even inverts between the [r-1•H•r-pheOEt +
and [r-1•H•s-
]
pheOEt]+ ones (ER - ES ) 0.34 kcal mol-1, Figure 8a). Figure
8b reveals that, also with the host in eq-eq conformation, the
enantiomers of naphOEt place their aromatic ring above one
isophthalic ring of the host with stabilizing face-to-face (r-
enantiomer) or edge-to-face (s-enantiomer) orientations (ER -
ES ) -1.32 kcal mol-1, Figure 8b). In contrast, the most stable
eq-eq [r-1•H•r-pheOEt +
] structure shows the phenyl ring of
the r-pheOEt guest placed just above two adjacent equatorial
phenyl rings of the host, thus in a position suitable for intense
edge-to-face attractive interactions. Similar interactions are much
weaker in the eq-eq [r-1•H•s-pheOEt +
] structure because of
the removed position of the guest aromatic ring (ER - ES )
-2.00 kcal mol-1, Figure 8b). Edge-to-face attractive interac-
tions are weak in the most stable ax-ax [r-1•H•s-pheOEt +
]
structure as well. This accounts for the stability difference
9
J. AM. CHEM. SOC. VOL. 130, NO. 2, 2008 531

A R T I C L E S
Gasparrini et al.
+
+
Figure 8. Structure and relative stability of the calculated most stable [r-1•H•pheOEt
]
and [r-1•H•naphOEt
]
diastereomeric complexes having the host
either in the ax-ax (a) or in the eq-eq (b) conformation employed to analyze the different selectivity showed by host 1 toward the enantiomers of pheOEt
and naphOEt
·
between the most stable eq-eq and ax-ax [r-1•H•pheOEt +
]
nation abilities compared to eq-eq-locked host 2. The reduced
stability gap between the diastereomeric complexes with 3 does
not seem to play any appreciable role on the large kinetic
enantioselectivity, measured for the same complexes in reaction
eq
ax
structures (ER
-
eq - ES
-
ax ) -0.56 kcal mol-1, Figure 8a,b).
The aptitude of the 1 and 3 hosts to assume the ax-ax
conformation by complexation enhances their enantiodiscrimi-
9
532 J. AM. CHEM. SOC. VOL. 130, NO. 2, 2008

Induced-Fit in the Gas Phase
A R T I C L E S
(Bruker Spectrospin) and a resonance cell (“infinity cell”) situated
between the poles of a superconducting magnet (4.7 T). Stock solutions
of the macrocyclic tetra-amides M ) r-1, s-1D, r-2, r-3, and s-3 (1 ×
10-5 M) in CH3OH, containing an equimolar amount of the appropriate
amino acid derivative A (A ) r/s-phe, r/s-pheOEt, and s-naphOEt), were
electrosprayed through a heated capillary (130 °C) into the external
source of the FT-ICR mass spectrometer. The resulting ions were
transferred into the resonance cell by a system of potentials and lenses
and quenched by collisions with methane pulsed into the cell through
a magnetic valve. Abundant signals, corresponding to the natural
isotopomers of the proton-bound complex [M•H•A]+, were monitored
and isolated by broad-band ejection of the accompanying ions. The
[M•H•A]+ family was then allowed to react with the chiral amine B
present in the cell at a fixed pressure whose value ranges from 6.0 ×
10-8 to 2.9 × 10-7 mbar depending upon its reactivity. Accurate
measurement of the B pressure in the resonance cell necessitates the
use of an ion gauge whose sensitivity is dependent on the nature of
the chemical species. The correction of the ionization gauge reading is
achieved by first determining the rate constant of the reaction between
1, because the presence of the decamethylene chain probably
influence dramatically the relative stability of the corresponding
transition structures.
Conclusions
The present MS and computational study provides compelling
evidence that the most stable conformers of the selected chiral
tetra-amide macrocycles M ) 1-3 may acquire in the gas phase
a different conformation by induced fit on complexation with
some representative amino acid derivatives A. This leads to the
coexistence in the gas phase of stable diastereomeric [M•H•A]+
eq-eq and ax-ax structures, in proportions depending on the
configuration of A and M and characterized by different stability
and reactivity toward the 2-aminobutane enantiomers.
The gas-phase reaction of the diastereomeric [M•H•A]+
complexes with the enantiomers of 2-aminobutane B obeys
either a monoexponential or a biexponential kinetics, depending
upon the number and the reactivity of coexisting eq-eq and
ax-ax structures. The monoexponential kinetics indicate the
occurrence of a single [M•H•A]+ structure (as with the
+
the CH4 radical cation and CH4 in the FT-ICR instrument at a given
nominal methane pressure and then by comparing the obtained result
with the average value of reported rate constants for this process (1.13
× 10-9 cm3 molecule-1 s-1).25 Subsequently, the correction factor
needed for amine B may be estimated with the method based on an
indicated linear dependence of the response of the ionization gauge
with the polarizability of the base in question.26
Computational Details. Molecular mechanics calculations and
docking simulations in Vacuo of the bimolecular adducts between the
chiral macrocycles r-1, r-2 and r-3 and the protonated aminoacidic
derivatives pheOEt, pheNH2, and naphOEt, as well as between the
[r-1•H•r-pheOEt]+ complex and the r-2-aminobutane BR, were achieved
in three steps:
(i) conformational search of host and guest molecules was carried
out by Batchmin and Macromodel version 4.5 (Columbia University,
NY) using the following options: MM2* Force Field, Montecarlo
stochastic algorithm with 3000 generated structures, minimization by
PR conjugate gradient. All the rotatable bonds were explored. The
obtained geometries were analyzed by the home made computer
program C.A.T.24 to exclude twin molecules and to make clusters based
on energetic and geometric criteria;
heterochiral [r-1•H•s-pheOEt + and [s-1D•H•r-pheOEt +
] ] com-
plexes) or, alternatively, of several stable [M•H•A]+ regioiso-
mers with comparable reactivity (as with the diastereomeric
[r-1•H•r-pheNH2]+ pair). In contrast, the biexponential kinetics,
observed with the other complexes investigated, are ascribed
to the coexistence of two stable [M•H•A]+ isomeric forms with
largely different reactivity toward B.
The gas-phase reaction 1 between the selected diastereomeric
[M•H•A]+ complexes and the B enantiomers exhibits an
enantioselectivity which strongly depends on the structure and
the configuration of A and M, but not on that of B. It is verified
that the measured kinetic enantioselectivity essentially reflects
the free energy gap between the homo- and heterochiral
[M•H•A]+ complexes, except when the tetra-amidic host
presents the intramolecular decamethylene chain. In this case,
the measured enantioselectivity mostly reflects the stability
difference between the eq 1 diastereomeric transition structures.
(ii) rigid docking was performed on each couple of host and guest
molecules using the MolInE program. The host-guest approach options
set were: 52 directions of translation and 272 relative orientations of
guest to host for each couple of host-guest conformations;
(iii) the ensemble of adducts obtained in the second step was
submitted to selection by using either energetic and geometric criteria.
The geometry of the so achieved complexes was optimized by full
relaxing their structure using the Batchmin program with the following
options set: MM2* Force Field, PR conjugate gradient minimization.
All the conformer ensembles were analyzed by C.A.T. program to
exclude twin molecules, make energetic clusters and perform calcula-
tions of both Boltzmann populations and thermodynamic quantities
Experimental Section
Mass Spectrometric Experiments. Mass spectra were obtained on
a LCQ-Deca XP Plus ion-trap mass spectrometer fitted with an
electrospray ionization (ESI) source and a syringe pump. Operating
conditions of the ESI source are as follows: spray voltage ) + 5.0
kV; sheath gas ) 15 AU (Arbitrary Units); capillary voltage ) + 27
V; capillary temperature ) 210 °C; tube lens offset ) 50 V. Methanolic
solutions are infused via a syringe pump at a flow rate of 3 µL/min.
ESI of solutions of the macrocyclic tetra-amides M ) r-1, r-2, and r-3
(1 × 10-5 M), containing an equimolar amount of the appropriate amino
acid derivative A (Chart 1), leads to the formation of appreciable
amounts of the corresponding [M•H•A]+ and [(M)2•H•A]+ complexes.
After individual isolation by broad-band ejection of the accompanying
ions, the isolated ions are then subjected to a supplementary ac signal
to resonantly excite them and cause fragmentation by collisions with
He gas (CID). Ion excitation time for CID is 30 ms with the amplitude
of the excitation AC kept the same for the measurements of two
enantiomeric guests with the same host. Optimized instrumental values
of relative collision energy of 12-15% give exclusively the fragment
ions of interest (relative collision energy values range 0-100%
corresponding to 0-5 V resonant excitation potential). Spectra acquired
in the centroid mode are the average of about 100 scans, each consisting
of three averaged microscans.
related to the simulated complexes.
+
Ab initio DFT calculations on the [r-1•H•r-pheOEt
]
and [r-1•H•s-
pheOEt
]
complexes were carried out using the Gaussian 03 suite of
+
programs27 installed on dual processor Opteron workstations. The
calculations were carried out at the B3LYP/6-31G*28 level of theory.
At the same level of theory, frequency calculations were performed
for all the optimized structures to ascertain their minimum or transition
(25) Ikezoe, Y., Matsuoka, S., Takebe, M., Viggiano, A. A. Gas-Phase Ion -
Molecule Reaction Rate Constant through 1986; Maruzen Company, Ltd.:
Tokyo, 1987.
(26) Bartmess, J. E.; Georgiadis, R. M. Vacuum 1983, 33, 149.
(27) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 1372, 5648. (b) Lee, C.; Yang,
W.; Parr, R. G. Phys. ReV. B 1988, 37, 785.
(28) Frisch, M. J.; et al. Gaussian 03, revision C.02; Gaussian,Inc.: Wallingford,
CT, 2004.
Kinetic experiments were performed at room temperature in an
APEX 47e FT-ICR mass spectrometer equipped with an ESI source
9
J. AM. CHEM. SOC. VOL. 130, NO. 2, 2008 533

A R T I C L E S
Gasparrini et al.
+
state nature. The ax-ax and eq-eq conformers of [r-1•H•r-pheOEt
]
the Consiglio Nazionale delle Ricerche (CNR). MS and AF
express their gratitude to F. Angelelli for technical assistance.
have been reoptimized at the B3LYP/6-311+G** level of theory. No
appreciable structural differences have been observed relative to those
calculated at the B3LYP/6-31G* level of theory.
Supporting Information Available: Synthesis of macrocycle
3. Most stable B3LYP/6-31G*-optimized structures of the eq-
eq, ax-ax, and ax-eq conformers of the diastereomeric
[r-1•H•pheOEt +
complexes. Kinetic plots. This material is
]
Acknowledgment. WorksupportedbytheMinisterodell’Istruzione
dell’Universita` e della Ricerca (MIUR, PRIN contract no.
2005037725) and FIRB contract no. RBPR05NWWC_003) and
available free of charge via the Internet at http://pubs.acs.org.
JA073287+
9
534 J. AM. CHEM. SOC. VOL. 130, NO. 2, 2008