Aza-crown macrocycles as chiral solvating agents for mandelic acid derivatives

Article abstract of DOI:10.1021/jo2018203

A series of new chiral macrocycles containing the trans-1,2- diaminocyclohexane (DACH) subunit and arene-and oligoethylene glycol-derived spacers has been prepared in enantiomerically pure form. Four of the macrocycles have been characterized by X-ray crystallography, which reveals a consistent mode of intramolecular N-H···N hydrogen bonding and conformational variations about the N-benzylic bonds. Most of the macrocycles were found to differentiate the enantiomers of mandelic acid (MA) by ¹H NMR spectroscopy in CDCl₃; within the series of macrocycles tested, enantiodiscrimination was promoted by (i) a meta-linkage geometry about the arene spacer, (ii) the presence of naphthalene-rather than phenylene-derived arene spacers, and (iii) increasing length of the oligoethylene glycol bridge. ¹H NMR titrations were performed with optically pure MA samples, and the data were fitted to a simultaneous 1:1 and 2:1 binding model, yielding estimates of 2:1 binding constants between some of the macrocycles and MA enantiomers. In several cases, NOESY spectra of the MA: macrocycle complexes show differential intramolecular correlations between protons adjacent to the amine and carboxylic acid groups of the macrocycles and MA enantiomers, respectively, thus demonstrating geometric differences between the diastereomeric intermolecular complexes. The three most effective macrocycles were employed as chiral solvating agents (CSAs) to determine the enantiomeric excess (ee) of 18 MA samples over a wide ee range and with very high accuracy (1% absolute error) (Figure presented).

Full text of DOI:10.1021/jo2018203

Article
pubs.acs.org/joc
Aza-Crown Macrocycles as Chiral Solvating Agents for Mandelic Acid
Derivatives
Thomas P. Quinn,† Philip D. Atwood,† Joseph M. Tanski,‡ Tyler F. Moore,†
and J. Frantz Folmer-Andersen*,†
†Department of Chemistry, The State University of New York at New Paltz, 1 Hawk Drive, New Paltz, New York 12561, United
States
‡Department of Chemistry, Vassar College, 124 Raymond Avenue, Box 601, Poughkeepsie, New York 12604, United States
S
* Supporting Information
ABSTRACT: A series of new chiral macrocycles containing
the trans-1,2-diaminocyclohexane (DACH) subunit and arene-
and oligoethylene glycol-derived spacers has been prepared in
enantiomerically pure form. Four of the macrocycles have been
characterized by X-ray crystallography, which reveals a
consistent mode of intramolecular N−H···N hydrogen
bonding and conformational variations about the N-benzylic
bonds. Most of the macrocycles were found to differentiate the
enantiomers of mandelic acid (MA) by 1H NMR spectroscopy
in CDCl3; within the series of macrocycles tested,
enantiodiscrimination was promoted by (i) a meta-linkage
geometry about the arene spacer, (ii) the presence of
1
naphthalene- rather than phenylene-derived arene spacers, and (iii) increasing length of the oligoethylene glycol bridge. H
NMR titrations were performed with optically pure MA samples, and the data were fitted to a simultaneous 1:1 and 2:1 binding
model, yielding estimates of 2:1 binding constants between some of the macrocycles and MA enantiomers. In several cases,
NOESY spectra of the MA:macrocycle complexes show differential intramolecular correlations between protons adjacent to the
amine and carboxylic acid groups of the macrocycles and MA enantiomers, respectively, thus demonstrating geometric
differences between the diastereomeric intermolecular complexes. The three most effective macrocycles were employed as chiral
solvating agents (CSAs) to determine the enantiomeric excess (ee) of 18 MA samples over a wide ee range and with very high
accuracy (1% absolute error).
supramolecular chemistry as a means of selectively addressing
enantiomers.
INTRODUCTION
■
The development of synthetic receptors that are capable of
differentially binding enantiomeric substrates is an important
goal,1 relating both to the advancement of fundamental
knowledge of molecular recognition phenomena2 and to
more practical applications focused on discriminating enan-
tiomers for preparative3 and analytical4 purposes. Thermody-
namic enantioselection is predicated on the ability of a chiral
host molecule to form energetically distinct diastereomeric
complexes with enantiomeric guests. Despite sustained interest
in the problem since soon after the inception of modern
supramolecular chemistry,5 it remains a considerable challenge
to realize binding enantioselecivities in synthetic systems that
are comparable to those of protein−substrate complexes. For
example, the blood coagulation protein thrombin binds small-
A key analytical application that has emerged from
enantioselective host−guest chemistry is the use of chiral
solvating agents (CSAs) for NMR spectroscopy.4e,8 CSAs are
essentially chiral receptors that differentiate the NMR signals of
enantiomeric guests upon complexation, usually allowing for
the enantiomeric excess (ee) of the guest sample to be
determined but also providing absolute configurations of guests
in some cases. Among the most familiar CSAs are chiral
lanthanide complexes8a,b which induce large, differential H
1
NMR shifts in a wide range of enantiomeric donor groups upon
metal coordination. Despite the versatility of lanthanide-based
CSAs, complications associated with paramagnetic line broad-
ening at higher field strengths and difficulties in elucidating the
mechanisms of chiral discrimination have encouraged the
development of alternative CSAs. Unlike chiral derivatizing
agents,8a,b,9 which rely on covalent bond formation to
desymmetrize enantiomers and have also been used extensively
molecule inhibitors with enantioselectivities (K(R)/K(S)
)
approaching 800:1 (ΔΔGenantio ∼ 4 kcal mol−1 at 298 K),6
whereas selectivity factors of 10 (ΔΔGenantio = 1.4 kcal mol−1 at
298 K) are generally considered impressive for synthetic host
molecules.2c,5e,7 Such disparities are likely related to differing
degrees of substrate encapsulation and structural complemen-
tarity, and their magnitudes highlight the latent potential of
Received: September 2, 2011
Published: October 31, 2011
© 2011 American Chemical Society
10020
dx.doi.org/10.1021/jo2018203 | J. Org. Chem. 2011, 76, 10020−10030

The Journal of Organic Chemistry
Article
Figure 1. Structures of the receptors examined in this study.
for ee measurement by NMR, CSAs do not require chemical
modification of the sample, thus obviating derivatization
chemistry and facilitating sample recovery. CSAs also offer a
significant advantage over chromatographic methods of ee
analysis, in that they avoid the inherently laborious process of
physically separating enantiomers. In light of these features, the
development of synthetic receptors as CSAs represents a
promising approach toward new methods of rapid and facile ee
analysis.10
Binding studies indicate that the macrocycles generally form
simultaneous 1:1 and 2:1 complexes with the enantiomers of
MA, which are driven by interactions between the amine and
carboxylic acid groups of the macrocycles and substrates,
respectively. In most cases, the 2:1 MA:macrocycle complexes
are found to be responsible for the observed enantioselectivity.
Several of the macrocycles are shown to function as effective
CSAs for MA derivatives, allowing for the accurate determi-
1
nation of ee by H NMR spectroscopy.
Understanding how CSAs differentiate enantiomers by NMR
is essential both to their optimal use and to the rational design
of improved variants. Because the association/dissociation
kinetics of many binding processes are fast on NMR time
scales, experiments employing CSAs typically yield NMR
spectra of weighted averages of free and bound species.8a,b,11
Consequently, enantiodiscrimination may arise from energetic
and/or spectral (magnetic) differences between diastereomeric
substrate:CSA complexes. If one substrate enantiomer is bound
more strongly by the CSA (energetic discrimination), that
enantiomer will be complexed to a greater degree in a
competitive regime in which the substrate is present in excess
over the CSA. In such cases, discrimination may be observed
even if the limiting bound chemical shifts of the diastereomeric
complexes are very similar. On the other hand, if the inherent
NMR spectra of the diastereomeric complexes differ signifi-
cantly (spectral discrimination), discrimination will result even
when both are present in similar amounts, for example, once
the substrate has been saturated with the CSA. An ideal CSA
would provide useful levels of differentiation over wide
stoichiometric and concentrations ranges, thus allowing for
substoichiometric use but not requiring the achievement of any
specific molar ratios.
DESIGN CRITERIA
■
As a byproduct of industrial nylon production, DACH is an
inexpensive and ubiquitous chiral subunit12 that has been
widely incorporated into schemes for enantioselective molec-
ular recognition,2d,5e,11a,c,13 synthesis,14 sensing,4i,15,16 and
separation.17 Structurally, DACH is preorganized by a preferred
diequatorial conformation, which persists in water regardless of
protonation state,18 and presumably enhances its efficacy as a
chelating Lewis base (ca. 3-fold higher affinity than ethylenedi-
amine for Zn(II), Cu(II), and Ni(II)19). The elaboration of the
N atoms is a general strategy for the creation of chiral
molecular clefts centered on the vicinal diamine group. In
particular, di(N-benzylated) DACH derivatives and related
compounds have been highly successful as ligands for
asymmetric catalysts14a,14c and as enantioselective recep-
tors.2d,4i,11b,16a Our interest in the effects of constraining the
N-benzylated DACH subunit within macrocyclic systems led us
to prepare the compounds shown in Figure 1. Although
numerous highly symmetrical chiral macrocycles comprised of
multiple N-benzylated DACH subunits2d,8f,16,20 have been
reported, macrocycles containing a single DACH unit,21,22
which may be more amenable to investigations of binding
phenomena, have been less well studied.
Herein, we report the synthesis of a series of chiral aza-crown
macrocycles, of which each member contains a di(N-
benzylated) trans-1,2-diaminocyclohexane (DACH) subunit.
X-ray crystal structures of several macrocycles are presented,
along with evidence of enantiomeric discrimination of mandelic
acid (MA) derivatives by 1H NMR spectroscopy in chloroform.
Figure 1 shows the structures of the receptors described in
this study. With the exception of (R,R)-8, which serves as a
nonmacrocyclic control compound, each receptor contains a
DACH moiety that is joined through N-benzylic linkages to a
flexible oligoethylene glycol bridge. The DACH subunits
10021
dx.doi.org/10.1021/jo2018203 | J. Org. Chem. 2011, 76, 10020−10030

The Journal of Organic Chemistry
Article
a
Scheme 1. (A) Representative Synthesis of (R,R)-5; (B) Intermediate Compounds Used in the Syntheses of the Macrocycles
Shown in Figure 1
a
Reagents: (i) p-hydroxybenzaldehyde, K2CO3, NaI, MeCN; (ii) (a) (R,R)-1,2-diaminocyclohexane, MeOH, (b) NaBH4.
impart chirality and C2 symmetry onto the structures, while
directing the vicinal amino groups toward the interior of
relatively rigid clefts enforced by the arene spacers. With the
exception of compounds (R,R)-121 and the non-macrocyclic
control compound (R,R)-8,15 the receptors described here are,
to the best of our knowledge, previously unreported. The ortho-
linked naphthyl groups of (R,R)-6 and (R,R)-7 are derived from
readily available 2-hydroxy-1-naphthaldehyde and are expected
to increase the steric bulk about the DACH center relative to
the ortho-phenylene group of (R,R)-1, while potentially
enabling fluorescence monitoring of binding events in future
work. The arene linkage geometry and length of the
oligoethylene glycol bridge will influence the overall shape
and flexibility of the macrocycles and, in turn, their binding
properties. However, Smithrud et al. have shown that for a
related series of di(N-benzylated) DACH-containing cyclo-
phanes the arene linkage geometry significantly affects the
basicity of the amino groups by allowing for differing degrees of
solvation and cation−π stabilization of the conjugate acid
ammonium ions,22a which implies that both steric and
electronic factors may contribute to changes in binding
properties that accompany structural variations among the
macrocycles.
Structural Analysis. The structures of four of the
macrocycles shown in Figure 1 were determined by X-ray
crystallography using a Cu-based X-ray source that allows for
independent confirmation of absolute configuration. Figure 2
shows ORTEP diagrams for the series of meta-linked
macrocycles (R,R)-2, (R,R)-3, and (R,R)-4. All of the structures
display intramolecular hydrogen bonding between the vicinal
amino groups and gauche relationships between both benzylic
C atoms and the nearest cyclohexyl methine H atoms. The
smallest macrocycle (R,R)-2 (Figure 2A) adopts a conforma-
tion in which both phenylene rings reside anti to the respective
adjoining cyclohexyl ring C atoms (dihedral angles: C6−N2−
C24−C22 = 179°; C1−N1−C7−C8 = 173°), giving the
molecule a relatively planar overall shape, whereas the larger
macrocycles (R,R)-3 and (R,R)-4 (Figure 2B,C) exhibit near-
perpendicular arrangements of the two phenylene rings (73°
and 80° angles for (R,R)-3 and (R,R)-4, respectively), which in
both cases is caused by gauche relationships between one of the
phenylene rings and the nearest cyclohexane C atom attached
to the intramolecular hydrogen bond donor N−H group
(dihedral angle for (R,R)-3: C21−N21−C27−C28 = 58°; for
(R,R)-4: C1−N1−C7−C8 = 57°). In addition to the
intramolecular hydrogen bonding seen in all of the structures,
(R,R)-2 features intermolecular N−H···N hydrogen bonding
that runs in two different directions within the crystal, thus
producing “methylene-like” disorder of the N−H bonds
(omitted from Figure 2A). These interactions are presumably
facilitated by the approximate planarity of (R,R)-2, as
intermolecular hydrogen bonding is absent in the structures
of (R,R)-3 and (R,R)-4.
RESULTS AND DISCUSSION
■
Synthesis. Each of the macrocycles was prepared as a
single enantiomer according to routes analogous to that shown
for (R,R)-5 in Scheme 1A. The dialdehyde 9 was prepared from
the corresponding ditosylate 10 and cyclized with optically pure
(R,R)-DACH through a reductive amination protocol. Under
suitably dilute conditions, this approach reliably yielded the
desired [1 + 1] macrocycles as the major product, although [2
+ 2] products were occasionally detected in crude reaction
mixtures by thin layer chromatography and mass spectrometry.
In all cases, the NMR spectra demonstrate the expected C2
symmetry, and no surprising correlations were observed in the
NOESY spectra. The structures of intermediate compounds
used in the preparation of the remaining macrocycles are shown
in Scheme 1B.
The crystal structures shown in Figure 2 are thought to
reflect the increasing flexibility of the meta-linked macrocycles
with ring size. Molecular modeling23 suggests that (R,R)-2 is
relatively rigid and that anti conformations are required about
both benzylic C−N bonds (anti-anti conformation) to avoid
excessive strain within the macrocycle. As a consequence, the
two chemically equivalent aryl H atoms that reside ortho to
both ring substituents (referred to as diortho) are firmly
directed toward the interior of the macrocyclic cavity of (R,R)-2
and constrained within the anisotropic deshielding region of the
10022
dx.doi.org/10.1021/jo2018203 | J. Org. Chem. 2011, 76, 10020−10030

The Journal of Organic Chemistry
Article
transannular arene ring. However, the more flexible macro-
cycles (R,R)-3 and (R,R)-4 are capable of accessing
conformations in which one of the benzylic C−N bonds is
gauche (anti-gauche, as observed in the X-ray structures and
predicted by molecular modeling), which allows the two
phenylene rings to move apart, thus diminishing transannular
deshielding anticipated in the anti-anti conformation. As
1
expected on this basis, the H NMR chemical shifts of the
diortho H resonances vary considerably among the series, with a
significant upfield shift observed with increasing ring size (see
Figure S-15). This is taken as evidence that within the meta-
linked series, the anti-gauche conformation becomes more
accessible as the length of the flexible oligoethylene glycol
linker increases and that conformational flexibility about the
DACH-centered cleft increases appreciably in the order (R,R)-2
< (R,R)-3 < (R,R)-4.
Figure 3 shows the ORTEP diagram of (R,R)-7, determined
from crystals grown from a concentrated CDCl3 solution. A
single solvent of crystallization is present in the structure and
has been omitted for clarity. The two naphthyl rings of (R,R)-7
are nearly parallel (16.7° angle) and are oriented away from the
cyclohexyl group (dihedral angles: C1−N2−C34−C29 = 166°;
C6−N1−C7−C8 = 179°), defining a cleft in which the vicinal
diamines reside. This cleft is further arranged in a pseudosym-
metrical fashion by the gauche relationship between both N-
benzyl bonds and their respective cyclohexyl methine H atoms,
as observed in all structures in Figure 2. The vicinal amino
groups participate in an internal hydrogen bond (N1−N2
distance = 2.792 Å), which is bifurcated through an interaction
with one of the aryl ether oxygen atoms (N1−O1 distance
2.952 Å, Figure 3B). The ortho-linkage geometry causes the
triethylene glycol chain to span the naphthyl rings on one side
of the average plane of the cyclohexyl moiety, imparting a
pronounced kink onto the 20-membered ring.
Chirality Recognition. The receptors shown in Figure 1
were evaluated as CSAs for MA, which has been widely used as
a model substrate for enantioselective recognition stud-
ies.4d,e,h,5d,n,7b,8d−g,j,16b,20c,24 The addition of a receptor to a
solution of racemic MA in CDCl3 caused the benzylic C−H
resonance in the 1H NMR spectrum of MA to shift upfield and,
in most cases, to split into two equal-intensity singlets that drift
downfield slightly and reconverge as [receptor]/[MA]
approaches unity. Representative spectra for the most effective
compound (R,R)-4 are presented in Figure 4A. The two
singlets are assigned to the MA enantiomers, which become
desymmetrized through their interaction with the macrocycle.
Figure 2. Views of the ORTEP diagrams of (A) (R,R)-2, because of
disorder caused by intramolecular H-bonding, four N−H bonds were
observed, of which two are omitted, (B) (R,R)-3 and (C) (R,R)-4;
thermal ellipsoids are scaled to 50% probability.
Figure 3. Side (A) and top (B) views of the ORTEP diagram of (R,R)-7. Thermal ellipsoids are scaled to 50% probability. Most of the hydrogen
atoms and a CDCl3 molecule of crystallization have been omitted for clarity.
10023
dx.doi.org/10.1021/jo2018203 | J. Org. Chem. 2011, 76, 10020−10030

The Journal of Organic Chemistry
Article
Figure 4. (A) Variations in part of the 300 MHz 1H NMR spectrum corresponding to the benzylic C−H resonance of (rac)-mandelic acid (MA, 10.0
mM in CDCl3) upon increasing [(R,R)-4]; additional studies of optically pure MA samples reveal that (R)-MA undergoes the greater upfield shift.
(B) Chemical shift nonequivalencies (Δδ) of the enantiomeric α-H signals of (rac)-MA (10.0 mM in CDCl3) as a function of molar ratio for the
effective receptors; curves intended only to guide the eye; spectra collected at 23 °C.
a
Scheme 2. A Proposed Simultaneous 1:1 and 2:1 Binding Model for the Interaction of Mandelic Acid and (R,R)-3
a
Binding geometries and protonation states are speculative.
As expected, analogous experiments performed with enantio-
merically pure MA samples yielded similar chemical shift
variations, but no splitting of the signals, and indicated that (R)-
MA undergoes the larger upfield shift. The overall upfield shift
of the benzylic C−H resonance, and the fact that the most
Figure 4B plots the chemical shift nonequivalencies (Δδ) of
MA enantiomers observed for each receptor as a function of
molar ratio. In most cases, the enantiomeric separation
increases sharply as the molar ratio rises from 0 to about
0.25 and then steadily declines to zero as [receptor]/[(rac)-
MA] is increased further. The naphthyl-containing macrocycle
(R,R)-7 represents an exception to this trend, as saturation of
Δδ is observed near [(R,R)-7]/[(rac)-MA] = 0.25. No
observable enantiodiscrimination was provided by the smaller
diethylene glycol-linked macrocycles (R,R)-2 and (R,R)-6
under the conditions studied. Apparently, the smaller cavities
of these receptors do not allow for inclusion of MA to the
extent that molecular shape can be efficiently discriminated. It
is evident from Figure 4B that, among the series of isomeric
triethylene glycol-linked macrocycles, the meta-linkage of
(R,R)-3 provides the greatest enantiodiscrimination, whereas
the ortho- and para-linked compounds (R,R)-1 and (R,R)-5 are
not significantly more effective than the non-macrocyclic
receptor (R,R)-8. Within the series of meta-linked ((R,R)-2,
(R,R)-3, and (R,R)-4) and naphthyl-linked ((R,R)-6 and (R,R)-
1
significant corresponding changes in the H NMR spectrum of
(R,R)-4 are downfield shifts of N−C−H signals (see Figure S-
17), suggest that the effect is driven primarily by Brønsted−
Lowery acid/base interactions between the carboxylic acid and
amine groups of the substrate and receptor, respectively. The
observed chemical shift dependence of the benzylic C−H
resonance on molar ratio (first upfield and then slightly
downfield) is uncommon among CSAs8b and likely signifies a
binding stoichiometry beyond a simple 1:1 complexation. The
maximal chemical shift difference between the signals (Δδ)
achieved in Figure 4A is roughly 0.1 ppm, and baseline
resolution is sustained over a fairly wide [(R,R)-4] range (ca.
0.5−6.5 mM), thus demonstrating the potential of this
macrocycle as a CSA for MA.
10024
dx.doi.org/10.1021/jo2018203 | J. Org. Chem. 2011, 76, 10020−10030

The Journal of Organic Chemistry
Article
1
Figure 5. (A) Variations in the region of the 300 MHz H NMR spectrum corresponding to the benzylic C−H of (S)-mandelic acid ((S)-MA,
varying concentration), in the presence of (R,R)-3 (440 μM). (B) Binding isotherms for (S)-MA and (R)-MA under conditions from (A); spectra
collected in CDCl3 at 23 °C.
a
Table 1. Limiting Bound Chemical Shifts and Macroscopic 2:1 Binding Constants
limiting bound chemical shifts (ppm)
2:1 binding constants (M−1
)
receptor
δ1:1(R)
δ1:1(S)
δ2:1(R)
δ2:1(S)
K2:1(R)
K2:1(S)
(R,R)-1
(R,R)-2
(R,R)-3
(R,R)-6
(R,R)-7
4.92
4.93
4.94
4.97
4.84
4.90
4.93
4.94
4.96
4.89
3.84
4.96
2.99
4.52
4.51
3.68
4.96
4.42
4.64
4.11
(3.3 ± 0.5) × 102
(4.8 ± 0.7) × 103
(2.2 ± 0.5) × 102
(7.5 ± 1.2) × 102
(9.6 ± 1.5) × 102
(1.9 ± 0.3) × 102
(5.6 ± 0.8) × 103
(7.4 ± 1.1) × 102
(9.3 ± 1.4) × 102
(1.8 ± 0.3) × 102
a
Determinations carried out in CDCl3 at 23 °C by fitting binding isotherms such as those shown in Figure 5B with the computer program NMRTit
27
1
HGG; the determined K1:1 values are too large for accurate analysis by H NMR.
7) macrocycles, enantiodiscrimination is shown to increase with
ring size, and additionally, the greater effectiveness of (R,R)-7
relative to (R,R)-1 demonstrates the beneficial affect of the
naphthyl groups in the ortho-linked system.
spectroscopically similar, the enantiodiscrimination would
diminish. The persistence of nonequivalence over this range
for the macrocycle (R,R)-7 implies that for this macrocycle the
diastereomeric 1:1 complexes are thermodynamically and/or
magnetically dissimilar, which is likely the result of increased
steric demand and/or magnetic anisotropy about the binding
site provided by the naphthyl rings. While the overall analysis
presented above neglects the involvement of heterochiral
ternary complexes (e.g., (R)-MA:(R,R)-3:(S)-MA), the ob-
served Δδ maxima near [receptor]/[(rac)-MA] = 0.25
nevertheless qualitatively support 2:1 MA:macrocycle com-
plexes as the major source of chirality discrimination.
The variations in Δδ with molar ratio shown in Figure 4B
can be tentatively explained in terms of a binding model in
which (rac)-MA simultaneously forms complexes with the
receptors of both 2:1 and 1:1 stoichiometries, and with the
exception of (R,R)-7, only the ternary complexes exhibit
significant enantioselectivity. The occurrence of both 2:1 and
1:1 complexes, as depicted in Scheme 2 for (R,R)-3, is
consistent with the assumption that proton transfer and/or
hydrogen bonding drive the association and has been
previously reported for the enantioselective recognition of
MA by a calix[4]arene bearing two chiral amine subunits.24f In
that study also, the 2:1 MA:receptor complexes were
determined to be the major source of enantioselectivity. In
addition, (R,R)-1,2-diphenylethane-1,2-diamine is known to be
most effective as a CSA for chiral carboxylic acids at a
carboxylic acid:diamine ratio of 2:1.8j Early in the titrations
represented in Figure 4B, as the molar ratio proceeds from 0 to
0.25, 2:1 complexes are statistically favored by stoichiometry
imbalance. Throughout this range, there is sufficient MA to
allow the receptor to selectively bind the enantiomer that forms
the more stable 2:1 complex, while preferentially leaving its
antipode free in solution. As the molar ratio further increases
from 0.25 to 0.5, 2:1 complexes remain statistically favored, and
so the majority of receptor added over this regime may interact
with the leftover, weaker-binding MA enantiomer in a 2:1
fashion, causing a convergence of the chemical shifts. As the
molar ratio proceeds from 0.5 through unity, 1:1 complexes
become increasingly significant, and assuming these to be
1
Additional H NMR experiments were performed in which
the concentrations of single enantiomers of MA were gradually
increased in the presence of constant concentrations of the
macrocycles. As before, the chemical shift of the benzylic C-H
resonance of MA was monitored as a function of molar ratio,
defined here as the inverse of that from Figure 4. Figure 5
shows typical results, in the case where [MA]/[(R,R)-3] is
raised incrementally from ∼0.2 to ∼6.0. The chemical shift
variations shown in Figure 5B exhibit changes in slope near
[MA]/[(R,R)-3] = 1 and [MA]/[(R,R)-3] = 2, which provides
evidence for the formation of both 1:1 and 2:1 complexes.25 It
is generally observed that when the substrate:macrocycle ratio
is <1, δ is quite insensitive to the molar ratio, as virtually all of
the MA present under these conditions is expected to bind the
macrocycle in a 1:1 fashion (provided 1/K1:1 < [(R,R)-3]t =
440 μM). The δ value observed when the receptor is in great
excess over MA is therefore considered to be close to the
limiting bound chemical shift of the 1:1 complex (R,R)-3:MA.
The coincidence of chemical shifts for MA enantiomers over
this range confirms the nonselectivity of 1:1 complexes, and as
10025
dx.doi.org/10.1021/jo2018203 | J. Org. Chem. 2011, 76, 10020−10030

The Journal of Organic Chemistry
Article
expected, is observed for all of the macrocycles except (R,R)-7
(see Figure S-24). As [MA]/[(R,R)-3] increases beyond unity,
the signals initially shift downfield sharply and diverge, while
gradually approaching saturation at higher molar ratios. This
behavior is interpreted in terms of the accumulation of
unbound MA along with the gradual onset of spectroscopically
and/or energetically distinct 2:1 complexes when MA is present
in significant excess.
The chemical shift variations shown in Figure 5B were fitted
to the model depicted in Scheme 2 using the computer
program NMRTit HGG,26 which provides estimates of K1:1,
K2:1, and the associated limiting bound chemical shifts δ1:1 and
δ2:1. The excellent fits obtained for most macrocycles (see
Supporting Information) confirm the proposed binding model
in those cases, but the determined K1:1 values are on the order
of 105−106 M−1, which are too large to be directly quantified by
NMR.27 However, the K2:1 values were reproducibly
determined and are presented along with the corresponding
limiting bound chemical shifts in Table 1. Of the receptors for
which binding constants were measured, (R,R)-7 show the
greatest binding enantioselectivity, with differences in 2:1
binding free energies of ∼1 kcal mol−1 between MA
enantiomers. Binding parameters were not obtained for
receptors (R,R)-4 and (R,R)-5, as the latter showed significant
variations in δ with molar ratio for both MA enantiomers when
[MA]/[(R,R)-5] < 1 (Figure S-22) and the former displayed an
anomalous discontinuity in the binding isotherm of (R)-MA at
[MA]/[(R,R)-4] = 1.5 (Figure S-21), suggesting that the
binding equilibria operative in these systems may be more
complicated than that depicted in Scheme 2. Despite the
unavailability K1:1 values, in most cases the strong correlations
between experimental and theoretical binding isotherms, and
the similarity of δ1:1 values between enantiomers further
implicates the 2:1 complexes as the major source of chirality
discrimination.
Although the involvement of ternary complexes obscures a
structural analysis of binding enantioselectivity in these systems,
rudimentary structural information was obtained through
nuclear Overhauser effect spectroscopy (NOESY). 2D
NOESY spectra of the complexes of (rac)-MA with (R,R)-2,
(R,R)-3, (R,R)-4, and (R,R)-7 were collected under conditions
of maximal chemical shift nonequivalence. The presence of the
meta-linked macrocycles ((R,R)-2, (R,R)-3, and (R,R)-4) was
found to increase the solubility of MA in CDCl3 by about an
order of magnitude, whereas (R,R)-7 did not influence MA
solubility appreciably. Under conditions of increased concen-
trations, much stronger intermolecular NOEs were observed,
and for (R,R)-3 and (R,R)-4, substantially larger benzylic C−H
resonance nonequivalencies were achieved (Δδ = 0.098 and
0.125 ppm for (R,R)-3 and (R,R)-4, respectively, at [MA] = ca.
100 mM). The complexes of all four macrocycles showed
intermolecular NOEs between the enantiomeric benzylic C−H
resonances of MA and the diastereomeric benzylic C−H
resonances of the macrocycles, with more intense correlations
to the pro-S H atoms of the macrocycles (Ha in Figure 6). A
portion of the NOESY spectrum of the (rac)-MA:(R,R)-7
complexes at a near 1:1 molar ratio is shown in Figure 6. The
benzylic C−H resonances of both MA enantiomers are
correlated with the Ha signal (circled in blue), but only that
of the stronger-binding (R)-MA correlates to Hb (green). This
result confirms that the MA enantiomers are bound by (R,R)-7
in structurally distinct environments and suggests that (R)-MA
forms the tighter 1:1 complex. Similar analyses of MA
Figure 6. A portion of the 300 MHz NOESY spectrum of a solution of
(rac)-mandelic acid (8.70 mM) and (R,R)-7 (9.04 mM) in CDCl3 at
23 °C; intermolecular correlations to Ha and Hb are circled in blue and
green, respectively; assignments of Ha and Hb made on the basis of an
intramolecular NOE correlation of Hb with the cyclohexyl methine
proton.
complexes with (R,R)-3 and (R,R)-4 also reveal nonequivalent
intermolecular NOEs between MA enantiomers; however, the
differences between MA enantiomers are less prominent (see
Figures S-26 and S-27).
Several of the macrocycles were tested as CSAs for a range of
MA derivatives, for which maximal Δδ values are presented in
Table 2. For each macrocycle, the variations in Δδ with molar
Table 2. Maximal Chemical Shift Nonequivalencies for
a
Mandelic Acid Derivatives
receptor
guest
(R,R)-3
(R,R)-4
(R,R)-5
(R,R)-7
MA
0.077
0.077
0.060
0.114
0.078
0.090
0
0.102
0.079
0.074
0.130
0.128
0.105
0.025
0.043
0.047
0.046
0.034
0.042
0.022
0.030
0.054
0.033
0.052
0.048
0.038
0.032
0.031
ortho-Cl-MA
meta-Cl-MA
para-Cl-MA
para-Br-MA
para-OMe-MA
α-MPA
a
1H NMR chemical shift differences (Δδ, ppm) between the benzylic
C−H resonances of enantiomers of (rac)-mandelic acid (MA)
derivatives (ca. 10 mM) observed upon titration of macrocycles in
CDCl3 at 23 °C. Entries in bold denote the achievement of clear
baseline resolution of signals on a 300 MHz instrument. For each
macrocycle and MA derivative, variations in Δδ with molar ratio were
qualitatively similar to those shown for MA in Figure 4B. α-MPA = α-
methoxyphenylacetic acid.
ratio were qualitatively similar to those presented for MA in
Figure 4B, with maximal Δδ near [receptor]/[MA] = 0.25 for
(R,R)-3, (R,R)-4, and (R,R)-5 and near saturation of Δδ
occurring at [(R,R)-7]/[MA] > 0.25 (see also Figure S-28).
10026
dx.doi.org/10.1021/jo2018203 | J. Org. Chem. 2011, 76, 10020−10030

The Journal of Organic Chemistry
Article
Figure 7. (A) Portion of the 300 MHz 1H NMR spectra of nonracemic mandelic acid samples (8.7 mM, various ee values) and (R,R)-7 (8.5 mM) in
CDCl3. (B) Correlation between ee values determined gravimetrically and by integration of the signals such as those shown in (A); ee defined in
terms of (S)-MA.
1
The larger meta-linked macrocycles (R,R)-3 and (R,R)-4 are the
most widely successful of those studied and are capable of
resolving enantiomeric signals for a variety of arene-substituted
MA derivatives, with especially large nonequivalencies observed
for para-substituted compounds. The poor differentiation
generally observed toward the enantiomers of α-methoxyphe-
nylacetic acid (α-MPA) implies that hydrogen bond donation
by the α-hydroxyl group of MA is crucial for enantiodiscrimi-
nation. None of the macrocycles tested elicited changes in the
1H NMR spectrum of (rac)-methyl mandelate, thus providing
further support for a binding model involving proton transfer or
hydrogen bonding.
In order to demonstrate the practical utility of the
macrocycles as CSAs, ee values of multiple nonracemic MA
samples were determined by integration of the enantiomeric
benzylic C−H resonances in the presence of near-optimal
quantities of receptors (R,R)-3, (R,R)-4, and (R,R)-7. Figure 7
shows that (R,R)-7, which provides the smallest enantiodiscri-
mination of the three macrocycles employed, maintains
analytical resolution over a wide ee range and that each of
the three macrocycles yields an excellent correlation between
the ee values determined by NMR and those determined
gravimetrically. The average absolute error in the 18 ee
measurements plotted in Figure 7B is ∼1%. Further, a
consistent relationship between the chemical shifts and the
configuration of MA was observed among the three macro-
cycles tested, with (R)-MA undergoing the greater upfield shift
in each case.
oligo(ethylene glycol) bridge. H NMR studies demonstrate
that in most cases MA binds to the macrocycles to form
complexes of both 1:1 and 2:1 stoichiometry and that the 2:1
complexes generally provide the source of enantiodiscrimina-
1
tion. H NMR titration data yielded estimates of 2:1 binding
constants to several of the macrocycles; however, the 1:1
binding constants are too large to be quantified by this method.
Differential intramolecular NOEs were observed between the
enantiomers of MA and the most effective macrocycles. The
meta-linked macrocycles (R,R)-3 and (R,R)-4 as well as the
naphthyl-containing ortho-linked macrocycle (R,R)-7 are shown
to be effective CSAs for MA derivatives and derivatives and
were used to accurately determine the ee of a number of MA
samples.
EXPERIMENTAL SECTION
■
General. 1H NMR spectra were recorded at 300 MHz and are
referenced to the solvent. Analytical thin layer chromatography was
performed using 0.2 mm silica gel plates. Column chromatography was
performed with silica gel 60 (230−400 mesh). Reagents were used as
received from commercial suppliers. All reactions were carried out
under an atmosphere of dry N2. Anhydrous CH2Cl2 was obtained by
passage through columns of activated molecular sieves. Acetonitrile
was distilled over P2O5. Anhydrous methanol was prepared by stirring
for 12 h over 3 Å molecular sieves. Ammonia-saturated methanol was
obtained by bubbling NH3(g) through anhydrous methanol (500 mL)
for 45 min at 0 °C. Anhydrous K2CO3 was stored at 110 °C for at least
72 h prior to use and allowed to cool in a desiccator. (R,R)-DACH was
resolved according to literature precedent,28 free-based from the L-
tartrate salt, and used immediately. Compounds (R,R)-815 and 10−
1229 were prepared according to literature precedent. Yields refer to
homogeneous, analytically pure (1H NMR) material and have not
been optimized.
SUMMARY
■
A series of new, enantiomerically pure DACH-based macro-
cycles containing arene and oligo(ethylene glycol) spacers have
been prepared, and four members have been characterized by
X-ray crystallography. Most of the macrocycles were found to
1H NMR Analyses. CDCl3 was passed through a column of
activated alumina prior to use, although virtually identical chemical
shift nonequivalencies were achieved when untreated CDCl3 was used.
Enantiodiscrimination studies of rac-MA derivatives (Figure 4 and
Table 2) were carried out by adding small aliquots of the macrocycle
stock solution (typically near 200 mM in CDCl3) to an NMR tube
containing of the MA derivative (550 μL and typically near 10 mM),
so that changes in the analyte solution volume were small (<5%).
Binding titrations (Figure 5 and Table 1) were performed by mixing
stock solutions of enantiomerically pure MA (typically near 8 mM in
CDCl3) and a given macrocycle (typically near 200 mM in CDCl3) to
1
enantioselectively associate with MA by H NMR on the basis
of interactions between the carboxylic acid and amine groups of
the substrates and macrocycles, respectively. The abilities of the
macrocycles to differentiate the 1H NMR spectra of the
enantiomers of MA by binding in CDCl3 was found to be
markedly dependent on (1) the nature of the arene group, (2)
the arene linkage geometry, and (3) the length of the
10027
dx.doi.org/10.1021/jo2018203 | J. Org. Chem. 2011, 76, 10020−10030

The Journal of Organic Chemistry
Article
obtain an analyte solution (typically 1 mL and near 500 μM
macrocycle in CDCl3) and a titrant solution (typically 5 mM in MA
and always of identical macrocycle concentration as present in the
analyte solution; also in CDCl3), thus ensuring that the macrocycle
concentration remained constant over the course of the experiment.
Titrations of a given macrocycle with enantiomeric MA samples were
performed in parallel. The observed chemical shift changes were fitted
with the computer program NMRTit HGG.26 The macroscopic 2:1
binding constants reported in Table 1 have been statistically
corrected30 from those reported by NMRTit HGG to account for
the fact that the program returns microscopic binding constants.
Crystallographic Data Collection and Refinement. Crystals of
(R,R)-2 and (R,R)-3 were grown by evaporation from concentrated
Et2O solutions, whereas crystals of (R,R)-4 and (R,R)-7 were obtained
by slow evaporation from solutions in CDCl3. X-ray diffraction data
were collected on a CCD platform diffractometer (Cu Kα (λ = 1.541
78 Å)) at 125 K. Crystals were mounted in a nylon loop with
cryoprotectant oil. The structures were solved using direct methods
and standard difference map techniques and were refined by full-
matrix least-squares procedures on F2 with SHELXTL (Version
6.14).31 All non-hydrogen atoms were refined anisotropically.
Hydrogen atoms on carbon were included in calculated positions
and were refined using a riding model. Hydrogen atoms on nitrogen
were located in the difference map and refined semifreely with the help
of a distance restraint. PLATON32 was used to verify all stereo-
chemical configuration designations. ORTEP-333 was used to generate
Figures 2 and 3. Crystallographic data for the structures reported in
this paper are given in the Supporting Information.
Typical Procedure for Synthesis of Dialdehydes. To a flame-
dried 250 mL round-bottom flask was added solution of ditosylated
ethylene glycol (10 mmol) in freshly distilled acetonitrile (60 mL),
followed by the phenol derivative (22 mmol), anhydrous K2CO3 (4.14
g, 30 mmol), and NaI (ca. 50 mg). The reaction was stirred at reflux
for 12 h, allowed to cool to room temperature, and concentrated under
reduced pressure. The resulting residue was partitioned between
CH2Cl2 (150 mL) and 1 M NaOH (50 mL). The organic phase was
washed with an additional portion of 1 M NaOH (50 mL) and
saturated brine (50 mL), dried over Na2SO4, and concentrated to yield
the product, which was of sufficient purity to be used in the
subsequent macrocyclization step but could be further purified by
column chromatography or recrystallization. Yields refer to material
obtained directly after extraction.
Diethylene Glycol Di(1-formyl-2-naphthyl) Ether (17). Yellow
solid (3.296 g, 82%); Rf (1.5:1 EtOAc:hexanes): 0.53. H NMR (300
1
MHz, CDCl3): δ 10.90 (s, 1H), 9.21 (d, J = 8.7 Hz, 1H), 7.98 (d, J =
9.0 Hz, 1H), 7.72 (bd, J = 7.9 Hz, 1H), 7.60−7.55 (m, 1H), 7.41−7.36
(m, 1H), 7.26−7.23 (m, 1H), 4.39 (t, J = 4.6 Hz, 2H), 3.99 (t, J = 4.6
Hz, 2H). 13C NMR (75 MHz, CDCl3): 192.3, 163.4, 137.7, 131.6,
131.6, 130.0, 129.0, 128.4, 125.2, 117.5, 114.2, 70.3, 69.6.
Triethylene Glycol Di(1-formyl-2-naphthyl) Ether (18). Yellow
1
solid (4.305 g, 94%); Rf (2:1 EtOAc:hexanes): 0.47. H NMR (300
MHz, CDCl3): δ 10.91 (s, 1H), 9.24 (d, J = 8.8 Hz, 1H), 7.98 (d, J =
9.2 Hz, 1H), 7.72 (bd, J = 8.1 Hz, 1H), 7.61−7.56 (m, 1H), 7.42−7.37
(m, 1H), 7.24 (d, J = 9.2 Hz, 1H), 4.35 (t, J = 4.6 Hz, 2H), 3.90 (t, J =
4.6 Hz, 2H), 3.74 (s, 2H). 13C NMR (75 MHz, CDCl3): δ 192.5,
163.7, 137.7, 131.7, 130.0, 128.9, 128.4, 125.2, 125.1, 117.4, 114.3,
71.2, 69.9, 69.5.
Typical Procedure for Synthesis of Macrocycles (R,R)-1
through (R,R)-5. To a flame-dried 1 L round-bottom flask was added
a solution of (R,R)-DACH (10.8 mmol, 1.23 g) in anhydrous MeOH
(200 mL) and a solution of the dialdehyde (10.8 mmol) in anhydrous
MeOH (200 mL). The reaction was stirred at room temperature for
16 h, after which NaBH4 (50 mmol, 1.89 g) was carefully added in
small portions over a period of 1 h. The reaction was then brought to
reflux for 4 h, concentrated under reduced pressure, and dissolved in a
bilayer of CH2Cl2 (300 mL) and 1 M NaOH (100 mL). The organic
phase was then washed sequentially with an additional portion of 1 M
NaOH (100 mL) and brine (100 mL), dried over Na2SO4, and
concentrated under reduced pressure to give the crude products
(typically foamy solids), which were purified by column chromatog-
raphy on silica gel (∼50 cm3 g−1), using 1−10% NH3-saturated
MeOH:EtOAc as the eluent.
Tri(ethyleneoxy)-o-phenylene-Linked Macrocycle ((R,R)-1). Col-
1
orless oil (2.106 g, 44%); Rf (10% NH3(sat)MeOH/EtOAc): 0.56. H
NMR (300 MHz, CDCl3): δ 7.14−7.04 (m, 2H), 6.84−6.73 (m, 2H),
4.09−4.02 (m, 1H), 3.98−3.92 (m, 2H), 3.77−3.55 (m, 6H), 2.12
(bdd, J = 5.2 Hz, J = 3.2 Hz 1H), 3.46 (d, J = 12.8 Hz, 1H), 2.18−2.10
(m, 2H), 1.89−1.85 (bm, 1H), 1.58−1.54 (bm, 1H), 1.13−1.07 (bm,
1H), 0.98−0.90 (bm, 1H). 13C NMR (75 MHz, CDCl3): δ 157.4,
130.6, 129.9, 128.2, 120.8, 111.7, 71.6, 70.0, 68.4, 62.5, 48.5, 31.7, 25.2.
HRMS (ESI, m/z) calcd for C26H37N2O4[M + H]+: 441.2748; found:
441.2751.
Di(ethyleneoxy)-m-phenylene-Linked Macrocycle ((R,R)-2).
White foamy solid (2.377 g, 55%); Rf (3% NH3(sat)MeOH/EtOAc):
1
0.39. H NMR (300 MHz, CDCl3): δ 7.16 (t, J = 7.9 Hz, 1H), 7.09
Triethylene Glycol Di(p-formylphenyl) Ether (9). Yellow solid
1
(bt, 1H), 6.79−6.73 (m, 2H), 4.14−4.11 (m, 2H), 3.95−3.88 (m, 3H),
3.56 (d, J = 12.8 Hz, 1H), 2.28−2.19 (bm, 2H), 1.84 (bd, J = 8.6 Hz,
1H), 1.28−1.21 (bm, 1H), 1.08−1.01 (bm, 1H). 13C NMR (75 MHz,
CDCl3): δ 159.3, 142.7, 129.0, 120.9, 114.6, 112.5, 71.4, 67.5, 61.3,
50.8, 31.6, 25.1. HRMS (ESI, m/z) calcd for C24H33N2O3[M + H]+:
397.2486; found: 397.2497.
(3.150 g, 88%); Rf (2:1 EtOAc:hexanes): 0.43. H NMR (300 MHz,
CDCl3): δ 9.86 (s, 1H), 7.79 (d, J = 8.9 Hz, 2H), 6.98 (d, J = 8.9 Hz,
2H), 4.18 (m, 2H), 3.87 (t, J = 4.7 Hz, 2H), 3.74 (s, 2H). 13C NMR
(75 MHz, CDCl3): 191.1, 164.0, 132.2, 130.3, 115.1, 71.1, 69.8, 67.9.
Triethylene Glycol Di(o-formylphenyl) Ether (13). Pale yellow oil
1
(3.364 g, 94%); Rf (1.5:1 EtOAc:hexanes): 0.47. H NMR (300 MHz,
Tri(ethyleneoxy)-m-phenylene-Linked Macrocycle ((R,R)-3).
CDCl3): δ 10.48 (s, 1H), 7.78 (dd, J = 7.7 Hz, J = 1.8 Hz, 1H), 7.52−
7.46 (m, 1H), 7.02−6.94 (m, 2H), 4.21 (t, J = 4.7 Hz, 2H), 3.88 (t, J =
4.6 Hz, 2H), 3.72 (s, 2H). 13C NMR (75 MHz, CDCl3): 190.1, 161.4,
136.1, 128.5, 125.3, 121.2, 113.1, 71.2, 69.8, 68.4.
White foamy solid (2.880 g, 61%); Rf (3% NH3(sat)MeOH/EtOAc):
1
0.29. H NMR (300 MHz, CDCl3): δ 7.15 (t, J = 7.9 Hz, 1H), 6.97
(bt, 1H), 6.80−6.76 (m, 2H), 4.04−4.01 (m, 2H), 3.90−3.82 (m, 3H),
3.72 (s, 2H), 3.58 (d, J = 12.6 Hz, 1H), 2.24−2.16 (bm, 2H), 1.72 (bd,
J = 9.8 Hz, 1H), 1.28−1.14 (bm, 1H), 1.06−0.99 (bm, 1H). 13C NMR
(75 MHz, CDCl3): δ 159.2, 142.6, 129.1, 120.9, 113.6, 113.0, 70.9,
69.9, 67.3, 60.9, 50.8, 31.5, 25.1. HRMS (ESI, m/z) calcd for
C26H37N2O4[M + H]+: 441.2748; found: 441.2761.
Diethylene glycol Di(m-formylphenyl) Ether (14). White solid
1
(2.875 g, 93%); Rf (1:1 EtOAc:hexanes): 0.53. H NMR (300 MHz,
CDCl3): δ 9.94 (s, 1H), 7.44−7.38 (m, 3H), 7.20−7.16 (m, 1H), 4.21
(t, J = 4.7 Hz, 2H), 3.95 (t, J = 4.7 Hz, 2H). 13C NMR (75 MHz,
CDCl3): 192.2, 159.5, 138.0, 130.3, 123.9, 122.3, 113.0, 70.0, 67.9.
Triethylene Glycol Di(m-formylphenyl) Ether (15). Pale yellow
Tetra(ethyleneoxy)-m-phenylene-Linked Macrocycle ((R,R)-4).
1
White foamy solid (2.682 g, 57%); Rf (3% NH3(sat)MeOH/EtOAc):
solid (3.217 g, 89%); Rf (1.5:1 EtOAc:hexanes): 0.50. H NMR (300
1
MHz, CDCl3): δ 9.91 (s, 1H), 7.42−7.35 (m, 3H), 7.17−7.13 (m,
0.28. H NMR (300 MHz, CDCl3): δ 7.13 (t, J = 7.9 Hz, 1H), 6.88
1H), 4.15 (t, J = 4.7 Hz, 2H), 3.85 (t, J = 4.7 Hz, 2H) 3.72 (s, 2H). 13
C
(bt, 1H), 6.79−6.76 (m, 2H), 4.04−3.77 (m, 5H), 3.66 (s, 4H), 3.55
(d, J = 13.0 Hz, 1H), 2.24−2.163 (bm, 3H), 1.70 (bd, J = 8.4 Hz, 1H),
1.23−1.17 (bm, 1H), 1.07−1.00 (bm, 1H). 13C NMR (75 MHz,
CDCl3): δ 159.3, 142.6, 129.2, 120.9, 113.9, 113.3, 71.0, 70.9, 69.8,
67.5 60.9, 50.8, 31.5, 25.2. HRMS (ESI, m/z) calcd for C26H37N2O4[M
+ H]+: 485.3010; found: 485.3018.
NMR (75 MHz, CDCl3): 192.2, 159.5, 137.9, 130.2, 123.8, 122.1,
113.0, 71.1, 69.8, 67.8.
Tetraethylene Glycol Di(m-formylphenyl) Ether (16). Colorless oil
1
(3.296 g, 82%); Rf (2:1 EtOAc:hexanes): 0.48. H NMR (300 MHz,
CDCl3): δ 9.93 (s, 1H), 7.44−7.36 (m, 3H), 7.19−7.15 (m, 1H), 4.16
(t, J = 4.8 Hz, 2H), 3.85 (t, J = 4.7 Hz, 2H) 3.73−3.65 (m, 4H). 13C
NMR (75 MHz, CDCl3): 192.3, 159.6, 138.0, 130.3, 123.8, 122.3,
113.1, 71.1, 70.1, 69.8, 67.9.
Tri(ethyleneoxy)-p-phenylene-Linked Macrocycle ((R,R)-5). Col-
1
orless oil (1.959 g, 41%); Rf (10% NH3(sat)MeOH/EtOAc): 0.48. H
NMR (300 MHz, CDCl3): δ 7.16 (d, J = 8.5 Hz, 2H), 6.81 (d, J = 8.5
10028
dx.doi.org/10.1021/jo2018203 | J. Org. Chem. 2011, 76, 10020−10030

The Journal of Organic Chemistry
Article
Hz, 2H), 4.09 (bt, 2H), 3.84−3.72 (m, 5H), 3.49 (d, J = 12.9 Hz, 1H),
2.26−2.13 (bm, 2H), 1.71 (bd, J = 8.1 Hz, 1H), 1.23−1.18(bm, 1H),
1.07−1.01 (bm, 1H). 13C NMR (75 MHz, CDCl3): δ 157.9, 133.6,
129.3, 114.7, 71.1, 70.1, 67.7, 61.0, 50.4, 31.8, 25.3. HRMS (ESI, m/z)
calcd for C26H37N2O4[M + H]+: 441.2748; found: 441.2745.
Typical Procedure for Synthesis of Macrocycles (R,R)-6 and
(R,R)-7. To a flame-dried 1 L round-bottom flask was added a
solution of (R,R)-DACH (4.9 mmol, 0.56 g) in dry CH2Cl2 (200 mL)
and a solution of the dialdehyde (4.9 mmol) in dry CH2Cl2 (200 mL).
The reaction was stirred at room temperature for 16 h. After this time,
MeOH (100 mL) was added and NaBH4 (28 mmol, 1.06 g) was
carefully introduced in small portions over a period of 1 h. The
reaction was then refluxed for 4 h, concentrated under reduced
pressure, and dissolved in a bilayer of CH2Cl2 (300 mL) and 1 M
NaOH (100 mL). The organic phase was then washed sequentially
with an additional portion of 1 M NaOH (100 mL) and brine (100
mL), dried over Na2SO4, and concentrated under reduced pressure to
give the crude products (typically foamy solids), which were purified
by column chromatography on silica gel (∼50 cm3 g−1), using 1−5%
NH3-saturated MeOH:EtOAc as the eluent.
Di(ethyleneoxy)-1,2-naphthyl-Linked Macrocycle ((R,R)-6). Yel-
low oil (0.875 g, 36%); Rf (3% NH3(sat)MeOH/EtOAc): 0.23. 1H
NMR (300 MHz, CDCl3): δ 8.01 (d, J = 8.8 Hz, 1H), 7.75−7.68 (m,
2H), 7.46−7.40 (m, 1H) 7.32−7.27 (m, 1H) 7.19 (d, J = 8.9 Hz, 1H)
4.33−4.18 (m, 4H) 4.01−3.87 (m, 2H), 2.43−2.41 (bm, 1H), 2.21−
2.17 (bm, 1H), 1.68−1.66 (bm, 1H) 1.24−1.10 (bm, 2H). 13C NMR
(75 MHz, CDCl3): δ 154.6, 133.5, 129.5, 128.8, 128.6, 126.7, 123.6,
123.4, 123.4, 114.8, 70.2, 69.1, 61.6, 41.3, 32.1, 25.1. HRMS (ESI, m/
z) calcd for C32H37N2O3[M + H]+: 497.2799; found: 497.2810.
Tri(ethyleneoxy)-1,2-naphthyl-Linked Macrocycle ((R,R)-7). Yel-
low foamy solid (1.342 g, 51%); Rf (10% NH3(sat)MeOH/EtOAc):
0.58. 1H NMR (300 MHz, CDCl3): δ 7.97 (d, J = 8.5 Hz, 1H), 7.72−
7.66 (m, 2H), 7.28−7.23 (m, 1H) 7.18−7.13 (m, 1H) 7.05 (d, J = 8.9
Hz, 1H) 4.45 (d, J = 11.1 Hz, 1H) 4.05 (bt, J = 9.3 Hz, 1H), 3.76 (d, J
= 11.1 Hz, 1H), 3.65−3.38 (m, 4H), 3.27−3.22 (m, 1H) 2.37−2.34
(bm, 2H), 1.78 (bd J = 8.6 Hz, 1H), 1.36−1.30 (bm, 1H) 1.21−1.13
(bm, 1H). 13C NMR (75 MHz, CDCl3): δ 154.7, 133.5, 129.4, 128.5,
128.1, 126.5, 124.0, 123.5, 123.2, 115.3, 71.5, 70.5, 69.4, 63.1, 41.9,
32.1, 25.3. HRMS (ESI, m/z) calcd for C34H41N2O3[M + H]+:
541.3061; found: 541.3065.
REFERENCES
■
(1) (a) Diederich, F. In Cyclophanes; Stoddart, J. F., Ed.; Monographs
in Supramolecular Chemistry 2; The Royal Society of Chemistry:
Cambridge, UK, 1991; pp 222−245. (b) Webb, T. H.; Wilcox, C. S.
Chem. Soc. Rev. 1993, 22, 383−395. (c) Zhang, X. X.; Bradshaw, J. S.;
Izatt, R. M. Chem. Rev. 1997, 97, 3313−3361. (d) Stibor, I.;
Zlatuskova, P. Top. Curr. Chem. 2005, 255, 31−63.
(2) (a) Jadhav, V. D.; Schmidtchen, F. P. J. Org. Chem. 2008, 73,
1077−1087. (b) Schneider, H.-G. Angew. Chem., Int. Ed. 2009, 48,
3924−3977. (c) Carrillo, R.; López-Rodríguez, M.; Martín, V. S.;
Martín, T. Angew. Chem., Int. Ed. 2009, 48, 7803−7808. (d) Gonzal
́
ez-
dez,
́
Alvarez, A.; Alfonso, I.; Díaz, P.; García-Espana, E.; Gotor-Fernan
́
̃
V.; Gotor, V. J. Org. Chem. 2008, 73, 374−382. (e) Rekharsky, M. V.;
Yamamura, H.; Ionue, C.; Kawai, M.; Osaka, I.; Arakawa, R.; Shiba, K.;
Sata, A.; Ko, Y. H.; Selvapalam, N.; Kim, K.; Inoue, Y. J. Am. Chem. Soc.
2006, 128, 14871−14880. (f) Schmuck, C.; Wich, P. Angew. Chem., Int.
Ed. 2006, 45, 4277−4281. (g) Gong, J.; Gibb, B. C. Chem. Commun.
2005, 3319−3321. (h) Rossi, A.; Kyne, G. M.; Turner, D. L.; Wells, N.
J.; Kilburn, J. D. Angew. Chem., Int. Ed. 2002, 41, 4233−4236.
(i) Sirikulkajorn, A.; Tuntulani, T.; Ruangpornvistuti, V.;
Tomapataget, B.; Davis, A. P. Tetrahedron 2010, 66, 7423−7428.
(3) (a) Harmata, M. Acc. Chem. Res. 2004, 37, 862−873. (b) Ema, T.;
Tanida, D.; Sugita, K.; Saki, T.; Miyazawa, K.; Ohnishi, A. Org. Lett.
2008, 10, 2365−2368. (c) Snyder, S. E.; Carey, J. R.; Pirkle, W. H.
Tetrahedron 2005, 61, 7562−7567. (d) Breccia, P.; Van Gool, M.;
́ ́
Perez-Fernandez, R.; Martín-Santamaría, S.; Gago, F.; Prados, P.; de
Mendoza, J. J. Am. Chem. Soc. 2003, 125, 8270−8284. (e) Schuur, B.;
Verkuijl, B. J. V.; Minnaard, A. J.; de Vries, J. G.; Heeres, H. J.; Feringa,
B. L. Org. Biomol. Chem. 2011, 9, 36−51. (f) Lewandowski, K.; Murer,
P.; Svec, F.; Frechet, J. M. J. Chem. Commun. 1998, 2237−2238.
(g) Bunzen, J.; Kiehne, I.; Benkhauser-Schunk, C.; Lutzen, A. Org. Lett.
2009, 11, 4786−4789.
̈
̈
(4) (a) Pu, L. Chem. Rev. 2004, 104, 1687−1716. (b) Hembury, G.
A.; Borokov, V. V.; Inoue, Y. 2008, 108, 1−73. (c) Accetta, A.;
Corradini, R.; Marchelli, R. Top. Curr. Chem. 2011, 300, 175−216.
(d) Zhu, L.; Zhong, Z.; Anslyn, E. V. J. Am. Chem. Soc. 2005, 127,
4260−4269. (e) Ema, T.; Tanida, D.; Sakai, T. J. Am. Chem. Soc. 2007,
129, 10591−10596. (f) Liu, S.; Pestano, J. P. C.; Wolf, C. J. Org. Chem.
2008, 73, 4267−4270. (g) James, T. D.; Sandanayake, K. R. A. S.;
Shinkai, S. Nature 1995, 374, 345−347. (h) Han, C.; Hou, X.; Zhang,
H.; Guo, W.; Li, H.; Jiang, L. J. Am. Chem. Soc. 2011, 133, 7644−7647.
(i) Yu, S.; DeBerarinis, A. M.; Turlington, M.; Pu, L. J. Org. Chem.
2011, 76, 2814−2819. (j) Reetz, M. T.; Sostmann, S. Tetrahedron
2001, 57, 2515−2520. (k) Kubo, Y.; Maeda, S.; Tokita, S.; Kubo, M.
Nature 1996, 382, 522−524. (l) Tsubaki, K.; Tanima, D.;
Nuruzzaman, M.; Kusumoto, T.; Fuji, K.; Kawabata, T. J. Org.
Chem. 2005, 70, 4609−4616.
ASSOCIATED CONTENT
■
S
* Supporting Information
Characterization data for macrocyclic compounds, results of
molecular modeling, 1H NMR and NOESY studies. This
material is available free of charge via the Internet at http://
pubs.acs.org.
(5) (a) Kyba, E. B.; Koga, K.; Sousa, L. R.; Siegel, M. G.; Cram, D. J.
J. Am. Chem. Soc. 1973, 95, 2692−2693. (b) Lehn, J.-M.; Sirlin, C. J.
Chem. Soc., Chem. Commun. 1978, 949−951. (c) Petti, M. A.; Shepodd,
T. J.; Barrans, R. E. Jr.; Dougherty, D. A. J. Am. Chem. Soc. 1988, 110,
AUTHOR INFORMATION
■
Corresponding Author
́
6825−6840. (d) Echavarren, A.; Gala, A.; Lehn, J.-M.; de Mendoza, J.
*E-mail: andersef@newpaltz.edu. Fax: (845) 257 3791.
J. Am. Chem. Soc. 1989, 111, 4994−4995. (e) Yoon, S. S.; Still, W. C. J.
Am. Chem. Soc. 1993, 115, 823−824. (f) Reetz, M. T.; Huff, J.;
Rudolph, J.; Tollner, K.; Deege, A.; Goddard, R. J. Am. Chem. Soc.
̈
ACKNOWLEDGMENTS
■
1994, 116, 11588−11589. (g) Pieters, R. J.; Diederich, F. Chem.
Commun. 1996, 2255−2256. (h) Asakawa, M.; Brown, C. L.; Pasini,
D.; Stoddart, J. F.; Wyatt, P. G. J. Org. Chem. 1996, 61, 7234−7235.
(i) Ogoshi, H.; Mizutani, T. Acc. Chem. Res. 1998, 31, 81−89. (j) Chin,
J.; Lee, S. S.; Lee, K. J.; Park, S.; Kim, D. H. Nature 1999, 401, 254−
257. (k) Rivera, J. M.; Martín, T.; Rebek, J. Jr. J. Am. Chem. Soc. 2001,
123, 5213−5220. (l) Fiedler, D.; Leung, D. H.; Bergman, R. G.;
Raymond, K. N. J. Am. Chem. Soc. 2004, 126, 3674−3675. (m) Ikeda,
T.; Hirata, O.; Takeuchi, M.; Shinkai., S. J. Am. Chem. Soc. 2006, 128,
16008−16009. (n) Heo, J.; Mirkin, C. A. Angew. Chem., Int. Ed. 2006,
45, 941−944. (o) Shoji, Y.; Tashiro, K.; Aida, T. J. Am. Chem. Soc.
2010, 132, 5928−5929.
We thank Professor Christopher Hunter of the University of
Sheffield providing NMRTit HGG. X-ray diffraction facilities
were provided by the US National Science Foundation (grants
0521237 and 0911324 to J.M.T.) and Vassar College. Mass
spectrometry was performed using instrumentation supported
by funding from a National Science Foundation Major
Research Instrumentation Grant (#1039659, Teresa A. Garrett,
P. I.) Acknowledgement is made to the donors of the American
Chemical Society Petroleum Research Fund for support of this
research through a New Undergraduate Investigator Award
(PRF# 49510-UNI3) to J.F.F.-A. and to the State University of
New York at New Paltz for financial support.
(6) (a) Fokkens, J.; Klebe, G. Angew. Chem., Int. Ed. 2006, 45, 985−
989. (b) Obst, U.; Betschmann, P.; Lerner, C.; Seiler, P.; Diederich, F.;
10029
dx.doi.org/10.1021/jo2018203 | J. Org. Chem. 2011, 76, 10020−10030

The Journal of Organic Chemistry
Article
Gramlich, V.; Weber, L.; Banner, D. W.; Schonholzer, P. Helv. Chim.
(21) Correa, W. H.; Scott, J. L. Molecules 2004, 9, 513−519.
̈
Acta 2000, 83, 855−909.
(7) (a) Miyaji, H.; Hong, S.-H.; Jeong, S.-D.; Yoon, D.-W.; Na, H.-K.;
Hong, J.; Ham, S.; Sessler, J. L.; Lee, C.-H. Angew. Chem., Int. Ed.
(22) (a) Dvornikovs, V.; Smithrud, D. B. J. Org. Chem. 2002, 67,
2160−2167. (b) Murahashi, S.-I.; Noji, S.; Komiya, N. Adv. Synth.
Catal. 2004, 346, 195−198.
(23) See Supporting Information for details of molecular modeling
studies.
2007, 46, 2508−2511. (b) Olivia, A. I.; Simon
́
, L.; Muniz, F. M.; Sanz,
F.; Moran, J. R. Org. Lett. 2004, 7, 1155−1157. (c) Gavin, J. A.; Garcia,
̃
́
́
́ ́
(24) (a) Busto, E.; Gonzalez-Alvarez, A.; Gotor-Fernandez, V.;
M. E.; Benesi, A. J.; Mallouk, T. E. J. Org. Chem. 1998, 63, 7663−7669.
(8) (a) Parker, D. Chem. Rev. 1991, 91, 1441−1457. (b) Wenzel, T. J.
Discrimination of Chiral Compounds Using NMR Spectroscopy; Wiley:
Hoboken, NJ, 2007. (c) Pham, N. H.; Wenzel, T. J. J. Org. Chem.
2011, 76, 986−989. (d) Yang, D.; Li, X.; Fan, Y.-F.; Zhang, D.-W. J.
Am. Chem. Soc. 2005, 127, 7996−7997. (e) Ma, F.; Al, L.; Shen, X.;
Zhang, C. Org. Lett. 2007, 9, 125−127. (f) Tanaka, K.; Fukuda, N.
Tetrahedron: Asymmetry 2009, 20, 111−114. (g) Moon, L. S.; Pal, M.;
Kasetti, Y.; Bharatam, P. V.; Jolly, R. S. J. Org. Chem. 2010, 75, 5487−
5498. (h) Iwaniuk, D. P.; Wolf, C. J. Org. Chem. 2010, 75, 6724−6727.
(i) Chin, J.; Kim, D. C.; Kim, H.-J.; Panosyan, F. B.; Kim, K. M. Org.
Lett. 2004, 6, 2591−2593. (j) Fulwood, R.; Parker, D. J. Chem. Soc.,
Perkin Trans. 2 1994, 57−64.
Alfonso, I.; Gotor, V. Tetrahedron 2010, 66, 6070−6077. (b) Demirtas,
H. N.; Bozkurt, S. B.; Durmaz, M.; Yilmaz, M.; Sirit, A. Tetrahedron:
Asymmetry 2008, 19, 2020−2025. (c) Altava, B.; Burgete, M. I.; Carbo,
́
N.; Escorihuela, J.; Luis, S. V. Tetrahedron: Asymmetry 2010, 21, 982−
989. (d) Wu, Y.; Guo, H.; James, T. D.; Jhao, J. J. Org. Chem. 2011, 76,
5685−5695. (e) He, X.; Zhang, Q.; Wang, W.; Lin, L.; Liu, X.; Feng,
X. Org. Lett. 2011, 13, 804−807. (f) Zheng, Y.-S.; Zhang, C. Org. Lett.
2004, 13, 1189−1192.
(25) Connors, K. A. Binding Constants, The Measurement of Molecular
Complex Stability; Wiley-Interscience: New York, 1987; pp 147−149.
(26) Bisson, A. P.; Hunter, C. A.; Morales, J. C.; Young, K. Chem.
Eur. J. 1998, 4, 845−851.
(27) Fielding, L. Tetrahedron 2000, 56, 6151−6170.
(28) Schanz, H.-J.; Linseis, M. A.; Gilheany, D. G. Tetrahedron:
Asymmetry 2003, 14, 2763−2769.
(29) Bonger, K. M.; van den Berg, R. J. B. H. N.; Heitman, L. H.;
IJzerman, A. P.; Oosterom, J.; Timmers, C. M.; Overkleeft, H. S.; van
der Marel, G. A. Bioorg. Med. Chem. Lett. 2007, 15, 4841−4856.
(30) Connors, K. A. Binding Constants, The Measurement of Molecular
Complex Stability; Wiley-Interscience: New York, 1987; pp 21−24.
(31) Sheldrick, G. M. Acta Crystallogr. 2008, A64, 112−122.
(32) Spek, A. L. Acta Crystallogr. 2009, D65, 148−155.
(33) Farrugia, L. J. J. Appl. Crystallogr. 1997, 30, 565.
́
(9) (a) Seco, J. M.; Quinoa, E.; Riguera, R. Chem. Rev. 2004, 104,
̃
17−177. (b) Fukui, H.; Fukushi, Y. Org. Lett. 2010, 12, 2856−2859.
(10) (a) Tsukamamoto, M.; Kagan, H. B. Adv. Synth. Catal. 2002,
344, 453−463. (b) Reetz, M. T. Angew. Chem., Int. Ed. 2001, 40, 284−
310. (c) Nieto, S.; Dragna, J. M.; Anslyn, E. V. Chem.Eur. J. 2010,
16, 227. (d) Matsushita, M.; Yoshida, K.; Yamamoto, N.; Wirsching,
P.; Lerner, R. A.; Janda, K. D. Angew. Chem., Int. Ed. 2003, 42, 5984−
5987. (e) Abato, P.; Seto, C. T. J. Am. Chem. Soc. 2001, 123, 9206−
9207. (f) van Delden, R. A.; Feringa, B. L. Angew. Chem., Int. Ed. 2001,
40, 3198−3120.
(11) CSAs that rely on metal coordination are often exceptions. See:
(a) Staubach, B.; Buddrus, J. Angew. Chem., Int. Ed. Engl. 1996, 35,
1344−1346. (b) Kim, H.-J.; Kim, W.; Lough, Kim, B.; Moon, Chin, J.
J. Am. Chem. Soc. 2005, 127, 16776−16777. (c) Kim, W.; So, S. M.;
Chagal, L.; Lough, A. J.; Lim, B. M.; Chin, J. J. Org. Chem. 2006, 71,
8966−8968.
(12) Bennani, Y. L.; Hanessian, S. Chem. Rev. 1997, 97, 3161−3195.
(13) (a) Nativi, C.; Francesconi, O.; Gabrielli, G.; Vacca, A.; Roelens,
S. Chem.Eur. J. 2011, 17, 4814−4820. (b) Ryan, K.; Gershell, L. J.;
Still, W. C. Tetrahedron 2000, 3309−3318. (c) Amendola, V.;
Boiocchi, M.; Estiban-Gomez, D.; Fabbrizzi, L.; Monzani, E. Org.
́
Biomol. Chem. 2005, 3, 2632−2639.
(14) (a) Taylor, M. S.; Jacobsen, E. N. Angew. Chem., Int. Ed. 2006,
45, 1520−1543. (b) Trost, B. M.; Miller, J. R.; Hoffman, C. M. J. Am.
Chem. Soc. 2011, 133, 8165−8167. (c) Evans, D. A.; Mito, S.; Seidel,
D. J. Am. Chem. Soc. 2007, 129, 11583−11592. (d) Boga, C.; Fiorelli,
C.; Savoia, D. Synthesis 2006, 2, 285−292.
(15) Folmer-Andersen, J. F.; Lynch, V. M.; Anslyn, E. V. J. Am. Chem.
Soc. 2005, 127, 7986−7987.
(16) (a) Yang, X.; Lui, X.; Shen, K.; Zhu, C.; Cheng, Y. Org. Lett.
2011, 13, 3510−3513. (b) Li, Z.-B.; Lin, J.; Pu, L. . Angew. Chem., Int.
Ed. 2005, 44, 1690−1693.
(17) (a) Gasparrini, F.; Misiti, D.; Still, W. C.; Villani, C.; Weenmers,
H. J. Org. Chem. 1997, 62, 8221−8224. (b) Reeve, T. B.; Cros, J.-P.;
Gennari, C.; Piarulli, U.; de Vries, J. G. Angew. Chem., Int. Ed. 2006, 45,
2449−2453. (c) Scrimin, P.; Tecilla, P.; Tonellato, U. Tetrahedron
1995, 51, 217−230. (d) Gawronski, J.; Grycz, P.; Kwit, M.;
Rychlewska, U. Chem.Eur. J. 2002, 8, 4210−4215.
(18) Sumeier, J. L.; Reilley, C. N. Anal. Chem. 1964, 36, 1707−1712.
(19) Bertsch, C. R.; Fernelius, W. C.; Block, B. P. J. Phys. Chem. 1958,
62, 444−450.
(20) (a) Gualandi, A.; Cerisoli, L.; Stoekli-Evans, H.; Savoia, D. J.
Org. Chem. 2011, 76, 3399−3408. (b) Gao, J.; Zingaro, R. A.;
Reibenspies, J. H.; Martell, A. E. Org. Lett. 2004, 6, 2453−2455. (c) Li.,
Z.-B.; Lin, J.; Sabat, M.; Hyacinth, M.; Pu, L. J. Org. Chem. 2007, 72,
4905−4916. (d) Hodacova,
́
J.; Chadim, M.; Zav
García-Espana, E.; Luis, S. V.; Miravet, J. F. J. Org. Chem. 2005, 70,
́
ada, J.; Aguilar, J.;
̃
2042−2047. (e) Gawronski, J.; Brzostowska, M.; Kwit, M.; Plutecka,
A.; Pychlewska, U. J. Org. Chem. 2005, 70, 10147−10150. (f) Padmaja,
M.; Periasamy, M. Tetrahedron: Asymmetry 2004, 15, 2437−2441.
10030
dx.doi.org/10.1021/jo2018203 | J. Org. Chem. 2011, 76, 10020−10030