Troger's base molecular scaffolds in dicarboxylic acid recognition

Article abstract of DOI:10.1021/jo9909204

Artificial receptors (1-5) have been designed and synthesized from simple precursors. The chain length selectivity studies of dicarboxylic acids within the cavities of new fluorescent Troger's base molecular frameworks (1- 3) have been carried out with a critical examination of their role of rigidity as well as flexibility in selective binding in comparison to receptor 5. The chiral resolution of the racemic Troger's base receptors (1 and 2) by chiral recognition with (+)- camphoric acid using hydrogen-bonding interactions has been studied.

Full text of DOI:10.1021/jo9909204

J . Org. Chem. 2000, 65, 1907-1914
1907
Tr oger ’s Ba se Molecu la r Sca ffold s in Dica r boxylic Acid
Recogn ition
Shyamaprosad Goswami,* Kumaresh Ghosh, and Swagata Dasgupta
Department of Chemistry, Bengal Engineering College (Deemed University),
Botanic Garden, Howrah 711 103, India
Received J une 7, 1999
Artificial receptors (1-5) have been designed and synthesized from simple precursors. The chain
length selectivity studies of dicarboxylic acids within the cavities of new fluorescent Troger’s base
molecular frameworks (1-3) have been carried out with a critical examination of their role of rigidity
as well as flexibility in selective binding in comparison to receptor 5. The chiral resolution of the
racemic Troger’s base receptors (1 and 2) by chiral recognition with (+)- camphoric acid using
hydrogen-bonding interactions has been studied.
In tr od u ction
Receptor 4 was synthesized for binding studies with the
metal ions. 2-Aminopyridine, a hydrogen-bonding motif
for the carboxylic acid group3,4 has been used by Hamilton
et al. to design dicarboxylic acid receptors by a common
strategy that involved the linking of the 2-aminopyridine
groups through rigid and semirigid spacers. Use of
transition metals as templates for the assembly of
hydrogen-bonding sites to create a functioning receptor
for the selective recognition of dicarboxylic acids5 is also
of particular interest. In this context, selective binding
of glutaric acid has been reported by a resorcinol-
aldehyde (dodecyl aldehyde and pentyl aldehyde) cy-
clotetramer6 and of dibutylmalonic acid by a receptor
having a benzophenone spacer between the two urea or
phosphonamide type binding moieties in the cavity for
the carboxylic acid.7
In search for a conformationally and geometrically
well-defined linker to hold the hydrogen-bonding groups,
we report here Troger’s base as a unique structural
feature for the construction of novel molecular receptors
for dicarboxylic acid recognition studies. The chiral
complexation of (+)-camphoric acid with the (()-Troger’s
base receptors is also reported.
The understanding of noncovalent intermolecular in-
teractions has become the central focus in the diverse
field of chemistry. The design and synthesis of model
receptors to recognize substrates of biochemical signifi-
cance utilizing these weak forces to mimic biological
events are of keen interest in molecular recognition
research.1,2 During the last decade, a lot of synthetic
receptors for dicarboxylic acids have been designed and
synthesized on the basis of the use of a relatively small
number of interacting H-bonding groups. The most
striking feature of the recognition of dicarboxylic acids
is their selective binding by synthetic receptors.3-7 The
precise selectivity between two purely aliphatic dicar-
boxylic acids (close homologues) is more difficult. This
happens owing to the fact that the bond rotation in an
open chain can occur freely and the two carboxyl ends in
the extended chain compromise to come closer for hy-
drogen-bonding contacts with the receptor binding groups
if they demand so. However, the binding constant may
be less if the receptor having the same binding groups is
at a more separated distance. Having a desire to create
new spacers, we designed novel Troger’s base receptors
(1-3), which have conformationally well-defined geom-
etry for the selective recognition of dicarboxylic acids.
Resu lts a n d Discu ssion
During the course of our attempts to construct molec-
ular recognition sites on rigid as well as semirigid
templates of different topography, we felt that the
Troger’s base unit could potentially serve as a rigid
molecular building block for the construction of artificial
receptors for dicarboxylic acids.8 Accordingly, we designed
and synthesized the receptors 1, 2, and 3, where ami-
nopyridine heterocyclic units are the hydrogen-bonding
sites for the formation of strong multihydrogen-bonded
complexes with dicarboxylic acids. The other semirigid
receptor 5, where the binding arms are assembled by a
flexible methylene bridge, was also synthesized and
studied to investigate its role in the selective binding of
dicarboxylic acids.
(1) Dugas, H. Bioorganic Chemistry, Springer-Verlag, Inc.: New
York, 1996.
(2) (a) Hosseini, M. W.; Lehn, J . M. J . Am. Chem. Soc. 1982, 104,
3525-3527. (b) Rebek, J ., J r.; Nemeth, D.; Ballester, P.; Lin, F. T. J .
Am. Chem. Soc. 1987, 109, 3474-3475. (c) Breslow, R.; Rajagopalan,
R.; Schwartz, J . J . Am. Chem. Soc. 1981, 103, 2905-2907.
(3) (a) Etter, M. C.; Adsmond, D. A. J . Chem. Soc., Chem. Commun.
1990, 589-591. (b) Tellado, F. G.; Geib, S. J .; Goswami, S.; Hamilton,
A. D. J . Am. Chem. Soc. 1991, 113, 9265-9269. (c) Vicent, C.; Fan, E.;
Hamilton, A. D. Tetrahedron Lett. 1992, 33, 4269-4272. (d) Geib, S.
J .; Vincent, C.; Fan, E.; Hamilton, A. D. Angew. Chem., Int. Ed. Engl.
1993, 32, 119-121. (e) Tellado, F. G.; Goswami, S.; Chang, S. K.; Geib,
S. J .; Hamilton, A. D. J . Am. Chem. Soc. 1990, 112, 7393-7394. (f)
Scott, E. F.; Arman, A. V.; Kinacid, S.; Hamilton, A. D. J . Am. Chem.
Soc. 1993, 115, 369-370.
(4) Karle, I. L.; Ranganathan, D.; Haridas, V. J . Am. Chem. Soc.
1997, 119, 2777-2783.
(5) (a) Goodman, M. S.; Hamilton, A. D.; Weiss, J . J . Am. Chem.
Soc. 1995, 117, 8447-8455. (b) Goodman. M. S.; J ubian, V.; Hamilton,
A. D. Tetrahedron Lett. 1995, 36, 2551-2554. Halter, I. P.; Smith, T.
J .; Weiss, J . J . Org. Chem. 1997, 62, 2186-2192.
In all the designed receptors 1, 2, and 3, the two
pyridine units are angularly disposed by the Troger’s base
(6) Tanaka, Y.; Kato, Y.; Aoyama, Y. J . Am. Chem. Soc. 1990, 112,
2807-2808.
(7) Raposo, C.; Luengo, A.; Almaraz, M.; Martin, M.; Mussons, M.;
Caballero, M. C.; Moran, J . R. Tetrahedron 1996, 52, 12323-12332.
(8) (a) Goswami, S.; Ghosh, K. Tetrahedron Lett. 1997, 38, 4503-
4506. (b) Goswami, S.; Ghosh, K.; Halder, M. Tetrahedron Lett. 1999,
40, 1735-1738.
10.1021/jo9909204 CCC: $19.00 © 2000 American Chemical Society
Published on Web 03/04/2000

1908 J . Org. Chem., Vol. 65, No. 7, 2000
Goswami et al.
Sch em e 1
Sch em e 2
spacer. The hydrogen-bonding groups are well arranged
in a concave face to bind dicarboxylic acids, and all of
them can have three possible conformations of compa-
rable energy values ( in-in, in-out, out-out ) in the
solution phase. MMX calculations9 on the “in-in” con-
formation of 1 (Emin ) 52.85 kcal/mol) indicates a distance
of separation of 12.99 Å between the two pyridine ring
nitrogen atoms. In receptor 3 (Emin ) 44.44 kcal/mol), the
N-N (pyridine ring nitrogens) and NH-NH distances
are ca. 11.60 and 10.56 Å, respectively. Thus the cavity
of the receptors 1 and 3 can accommodate dicarboxylic
acids of various chain lengths, and the selective binding
occurs only when the cavity dimensions and the chain
lengths of dicarboxylic acids are matched together.
Similarly, 2 and 5 having different cavity dimensions can
also bind dicarboxylic acids of different chain lengths.
Troger’s base analogues 1 and 2 have been synthesized
from the reaction of the corresponding amines with
hexamethylenetetramine in the presence of trifluoroace-
tic acid (Scheme 2). It was initially planned to synthesize
both receptors 1 and 2 according to Scheme 1. But this
approach was frustrating owing to the very low yield of
the intermediate 6, which on subsequent hydrolysis by
lithium hydroxide followed by coupling with 2-amino-6-
methylpyridine gave also the desired receptor 1 but in
very poor yield and which could not be isolated in pure
form. Thus this route was abandoned.
(PTLC) from a mixture of other undesirable side prod-
ucts. Compound 2 was obtained in the same way in 62%
yield. Formation of both 1 and 6 in poor yields is ascribed
to the presence of electron-withdrawing groups in the
para position of aniline derivatives.11
The other Troger’s base analogues 3 and 4 were
synthesized according to Scheme 3. 2-Acetylamino-6-
methylpyridine was coupled with 4-nitrobenzaldehyde in
the presence of Ac2O at 180 °C to give 10 (cis and trans
mixtures; major trans).12a Both nitro and ethylenic double
bonds in 10 were reduced by hydrogenation to give the
corresponding amine 11, which was then reacted with
hexamethylenetetramine in the presence of trifluoroace-
tic acid to generate 3. In a similar way, the other
analogue 4 was synthesized in 70% yield. The synthetic
Schemes 2 and 3 show an elegant approach to attach the
heterocyclic units to the Troger’s base molecular frame-
works, which could potentially serve as hydrogen-bonding
Compound 1 was synthesized by Scheme 2 where 2-(4-
aminobenzoyl amino)-6-methylpyridine 8 was subjected
to react with hexamethylenetetramine in the presence
of trifluoroacetic acid (nonaqueous medium).10 Compound
1 was isolated in 15-20% yield by careful purification
(11) (a) Sucholiki, I. V. L. P.; Wilcox, C. S. J . Org. Chem. 1988, 53,
98-104. (b) Webb, T. H.; Wilcox, C. S. J . Org. Chem. 1990, 55, 363-
365.
(12) (a) Williams, J . L. R.; Adel, R. E.; Carbson, J . M.; Reynolds, G.
A.; Borden, D. G.; Ford, J . A., J r. J . Org. Chem. 1963, 387-390. (b)
Goswami, S.; Mahapatra, A. K. Tetrahedron Lett. 1998, 39, 1981-1984.
(9) Energy minimization was carried out by MMX (PC Model Serena
Software 1993). Molecular modeling was done using standard con-
stants, and the dielectric constant was maintained at 1.5.
(10) Adrian, J . C., J r.; Wilcox, C. S. J . Am. Chem. Soc. 1992, 114,
1398-1403.

Troger’s Base Molecular Scaffolds
J . Org. Chem., Vol. 65, No. 7, 2000 1909
Sch em e 3
F igu r e 1. Mode of complexation of 1 with dicarboxylic acids.
Ta ble 1. Bin d in g Con sta n ts of Recep tor 1 w ith
Dica r boxylic Acid s
diacids
glutaric
adipic
suberic
sebacic
benzene-1,4-diacetic
Ka (at 25 °C), M-1 ∆G (at 25 °C), kcal/mol
1.0 × 103
1.69 × 103
1.5 × 104
3.10 × 103
2.88 × 102
-4.09
-4.40
-5.69
-4.76
-3.35
Ta ble 2. Bin d in g Con sta n t Va lu es Ka (M-1) of Recep tor s
2 a n d 3 w ith Dica r boxylic Acid s
glutaric
adipic
suberic
sebacic
Sch em e 4
receptor 2
receptor 3
2.4 × 102
3.5 × 102
4.8 × 102
5.8 × 102
1.10 × 102
5.3 × 102
1.01 × 103
6.54 × 103
1). In the NMR spectrum of a 1:1 complex of 1 (in CDCl3)
with suberic acid (dissolved in CDCl3 with 2% d6-DMSO),
the considerable downfield shift of the pyridine amide
protons (∆δ 1.68 ppm) and of the phenyl ring as well as
the methylene bridge protons of the Troger’s base frame
(∆δ 1.03 ppm) indicates the formation of a stronger
complex (Figure 1). The dilution of the 1:1 complex of
suberic acid with the receptor 1 in CDCl3 (∆δ 2.0 ppm of
pyridine amide protons) also shows negligible change in
receptors for various guests. Compound 5 was synthe-
sized according to Scheme 4. The C-C bond formation
in 15 was achieved by the cobalt coupling method.12b
Then alkaline hydrolysis of 15 afforded the acid 16, which
on coupling with 2-amino-6-methylpyridine amine gave
the receptor 5 in good yield.
Com p lexa tion Stu d ies. Complexation studies of di-
carboxylic acids (both chiral and achiral) were carried out
using NMR and fluorescence experiments.
1
the H NMR spectrum. This supports a strong complex-
ation. The other receptors 2 and 3 were explored in
complexation study with the same dicarboxylic acids,
used for 1 (Table 1) to see the chain length selectivity by
their cavities. The comparative binding constant values
are presented in Table 2. Figure 2A (Emin ) 41.98 kcal/
mol) and Figure 2B (Emin ) 39.88 kcal/mol) are the
energy-minimized forms of the complexes of suberic acid
with receptors 1 and 3, respectively. In receptor 1, suberic
acid is well-arranged (Figure 2A) with the hydrogen-
bonding sites into the cavity compared to receptor 3 in
Figure 2B.
(a ) NMR Titr a tion . In NMR titrations, hosts were
dissolved in dry CDCl3 and the stock solutions of guests
were prepared by dissolving aliphatic dicarboxylic acids
in dry CDCl3. But owing to the partial solubility of the
diacids into CDCl3, a few drops of dry d6-DMSO (2% in
each case) were added to the stock solution to prepare a
homogeneous solution of the guest. On addition of the
guest to the host solution under controlled conditions, a
downfield chemical shift of the amide protons of host was
observed. The chemical shift values were noted as a
function of the concentration of the variable component.
Foster-Fyfe analysis13a of the titration data at 25 °C gave
binding constant values. The binding constant values for
1 with dicarboxylic acids of various chain lengths are
given in Table 1. The binding study shows that the cavity
of Troger’s base analogue 1 is selective for suberic acid.
Integration of the proton signals of the 1:1 complex in
Both 2 and 3 bind adipic acid with a 1:1 stoichiometry.
Receptor 2 also makes a 1:1 complex with glutaric acid.
The receptor 2 shows a marginal increase in Ka with
suberic acid giving a 1:2 (guest:host) stoichiometry
whereas 3 binds with long chain sebacic acid with a 1:2
(guest:host) stoichiometry. All the stoichiometries of the
complexes are determined from the break in the titration
curves (∆δ vs Cguest/Chost). The reduced values of Ka with
2 and 3 are explained in terms of their flexibility. The
attachment of the pyridine amide binding moieties of
receptors 2 and 3 with the tetrahedral methylene carbons
instead of being directly bonded to the benzene ring (as
in receptor 1) into the respective cavity may also have a
steric influence on the stoichiometry of the complex.
Owing to the presence of the -CH2- group between
pyridine and Troger’s base benzene ring, the energy
barrier between the conformations is reduced and the
pyridine ring can adopt different orientations in solution
the 1H NMR spectrum as well as the break in the
13b
titration curve (∆δ vs Cguest/Chost
)
at 1.0 establishes a
1:1 stoichiometry for the dicarboxylic acids with 1 (Table
(13) (a) Foster, R.; Fyfe, C. A. Prog. Nucl. Magn. Reson. Spectrosc.
1969, 4, 1-89. (b) Blanda, M. T.; Horner, J . H.; Newcomb, M. J . J .
Org. Chem. 1989, 54, 4626-4636.

1910 J . Org. Chem., Vol. 65, No. 7, 2000
Goswami et al.
F igu r e 2. Energy-minimized structures of (A) 1 with suberic acid and (B) 3 with suberic acid.
F igu r e 3. NMR spectrum of 1:1:1 mixture of receptor 1, benzene-1,4-diacetic acid, and suberic acid.
phase very easily. In the case of 1, the higher association
constant for suberic acid compared to glutaric and adipic
acids, respectively, is due to the complementary cavity
size with suberic acid but not with shorter dicarboxylic
acids. The selectivity of suberic acid was proved by its
exclusive extraction (∼90%) in pure CDCl3 from its 1:1
mixture with benzene 1,4-diacetic acid as predicted by
comparison of their NMR integrations in Figure 3.
The other receptor 5 shows weaker binding with
suberic and adipic acids. In the energy-minimized struc-
ture (Figure 4A) of 5 (Emin ) 33.51 kcal/mol), the N-N
(ring nitrogens) distance is ca. 11.22 Å and NH-NH
distance is ca. 9.50 Å. Although these distances are
comparable with the values of the receptor 1, it does not
bind suberic acid strongly (Ka ) 3.0 × 102 M-1). More
interestingly, during NMR titration of 5 with suberic
acid, the amide peak was broad and split into a doublet.
This situation indicates the different chemical environ-
ments of amide protons, which probably arise from the
loss of symmetry present in 5 owing to the C-C bond
rotation in the bridge during complexation. This is also
evident from the energy-minimized structure of suberic
acid with receptor 5 (Figure 4B; Emin ) 23.37 kcal/mol)
where one phenyl ring at the bridge is more tilted from
the plane of the other phenyl ring compared to receptor
5 itself in Figure 4A. This simple experiment thus
indicates that the flexibility introduced in the design of
the receptor does not have a significant role in the strong
as well as selective binding of dicarboxylic acids until
rigid or semirigid spacers are considered as the building
blocks for the construction of such molecular receptors.
In all the cases, binding constants are sacrificed to
some extent using DMSO as it is a competitive guest.
This was evidently proved by taking glutaric acid as a
representative with the receptor 1. The value of binding
constant measured in CDCl3 for glutaric acid (Ka ) 6.01
× 103 M-1) was found to be higher than the value in
CDCl3 containing 2% d6-DMSO (Ka ) 1.0 × 103 M-1
)
given in Table 1. We are now using the Troger’s base

Troger’s Base Molecular Scaffolds
J . Org. Chem., Vol. 65, No. 7, 2000 1911
F igu r e 4. Energy-minimized structure of (A) receptor 5 and (B) receptor 5 with suberic acid.
analogue 4 in metal binding.14 Its strong affinity with
CuII-metal ion was noted from the broadening of the
peaks of the complex in the NMR spectrum.
F lu or escen ce Exp er im en ts. Fluorescence experi-
ments were carried out on the Troger’s base receptors
with suberic acid, which showed strong complexation
characteristics in the NMR titration data of suberic acid
with receptor 1. Receptors were dissolved in CHCl3, and
guests were taken in CHCl3 containing 2% DMSO.
Interestingly, here also a great affinity for suberic acid
was noted with a considerable change in fluorescence
intensity. The excitation wavelength was at 330 nm for
this set of studies with the emission being monitored at
404 nm. Figure 5A shows the enhancement of fluores-
cence intensity with subsequent additions of the guest
suberic acid. A plot of emission intensity against the
concentration of suberic acid showed a nonlinear increase
in the fluorescence emission as a function of suberic acid
concentration. This gives a binding constant (Ka)15a of 2.54
× 104 M-1, which is close to the predicted value in 1H-
NMR titration experiments. The experiment was re-
peated with receptor 2. In the case of receptor 2, a
significant enhancement is not observed (Figure 5B) as
in the case of receptor 1. Excitation in the second case
was also at 330 nm with the emission wavelength being
set at 384 nm. The hydrogen bonding perturbs the lowest
excited singlet state of nπ* type of the receptor 1 in a
destabilizing manner.15b This increases the energy of the
nπ* excited state of receptor 1 to such an extent that the
fluorescence quantum yield increases to a considerable
extent. Such perturbation is less in the case of receptor
2, thereby indicating its poor complexation through
hydrogen bonding with suberic acid. The lower affinity
for the guest suberic acid in this case can be attributed
F igu r e 5. Fluorescence titration spectra of (A) 1 (2.04 × 10-4
to the flexibility of the arms of the Troger’s base receptor
that cannot take part in effective complexation as is
possible in the case of receptor 1. Studies with sebacic
acid for the receptor 1 did not reveal any significant
enhancement of fluorescence intensity.
M) and (B) 2 (2.57 × 10-4 M) at excitation at 330 nm with
suberic acid (4.14 × 10-3 M) at 25 °C.
tempted us to study the chiral recognition process by
using chiral acid guests with racemic Troger’s base
receptors. We studied the binding process of (()-Troger’s
base receptors 1 and 2 with (+)-camphoric acid. (+)-
Camphoric acid was used as the chiral dicarboxylic acid
guest because of its approximate size matching with the
cavity dimension of 1 and 2. During the complexation
study, the amide protons of 1 (Figure 6A) and 2 (Figure
6B) are split into two sets as shown by NMR studies,
establishing an equilibrium between the two diastereo-
meric complexes in solution. The separation between the
Such hydrogen-bonding studies of dicarboxylic acids of
various chain lengths with the Troger’s base analogues
(14) Yashima, E.; Akashi, M.; Miyauchi, M. Chem. Lett. 1991, 1017-
1020. Metal binding studies with receptor 4 and others are in progress
in our laboratory. This will be published in due course.
(15) (a) Tecilla, P.; Dixon, R. P.; Slobodkin, G.; Alavi, D. S.; Waldeck,
D. H.; Hamilton, A. D. J . Am. Chem. Soc. 1990, 112, 9408-9410. (b)
Prasanna de Silva, A.; Gunaratne, H. Q. N.; Gunnlaugsson, T.; Huxley,
A. J . M.; McCoy, C. P.; Rademacher, J . T.; Rice, T. E. Chem. Rev. 1997,
97, 1515-1566.

1912 J . Org. Chem., Vol. 65, No. 7, 2000
Goswami et al.
F igu r e 6. Variation of chemical shift of (A) 1 and (B) 2 on gradual addition of (+)-camphoric acid solution.
two sets slowly decreased on gradual addition of guest
solution during the course of titrations at 25 °C. At 1:1
stoichiometry, the two peaks are merged and broadened.
The two different sets of amide protons are obtained from
the generation of intrinsically nonidentical chemical
shifts in some of the sensor nuclei of the diastereomeric
complexes that are found between (+)-camphoric acid and
(()-Troger’s bases 1 and 2.
using chiral dicarboxylic acid [(+)-camphoric acid] as a
chiral stationary phase may be feasible. But practically
it was very difficult to resolve the racemic Troger’s bases
(1 and 2) using (+)-camphoric acid both by crystallization
as well as by chromatographic technique. This may be
due to the fast equilibration between the diastereomeric
complexes formed in the solution phase.
The chemical shift variation can be explained by
assuming the different stability of diastereomeric associ-
ates of the enantiomers in solution. The variation of
chemical shift has been plotted (Figure 6A, B) as a
function of the overall host and guest concentrations at
25 °C. In the case of 1, two diastereomeric complexes with
(+)-camphoric acid [(+)-Troger’s base receptor 1‚(+)-
camphoric acid and (-)-Troger’s base receptor 1‚(+)-
camphoric acid] differ by a binding constant (∆Ka) value
of 2.5 × 102 - 3.0 × 102 M-1 whereas for 2 [(+)-Troger’s
base receptor 2‚(+)-camphoric acid and (-)-Troger’s base
receptor 2‚(+)-camphoric acid] the value of ∆Ka is reduced
to 1 × 102 M-1. Thus the NMR resolution of (()-Troger’s
base analogues (1 and 2) proceeds from a difference
between the stability constants for the diastereomeric
cavities, rather than from a purely spectroscopic effect
on the induced chemical shift (δ0 - δR).16 We thus
expected that a chromatographic resolution17 of 1 and 2
Con clu sion
We have been able to readily construct a family of
Troger’s base molecular frameworks with heterocyclic
binding motifs for the carboxyl group that selectively
binds dicarboxylic acids of various chain lengths with
moderate association constants. These Troger’s base
receptors can be used as fluorescent probes in monitoring
the carboxylic acid recognition process. In the design, the
flexible receptors show weak binding with dicarboxylic
acid guests in spite of having comparable chain lengths
with the cavity dimensions. This demonstrates that the
rigidity of the host molecule has a special role in the
selective binding of dicarboxylic acids.
(16) Canceill, J .; Lacombe, L.; Collet, A. J . Am. Chem. Soc. 1985,
107, 6993-6996.
(17) Prelog, V.; Wieland, P. Helv. Chim. Acta. 1944, 27, 1127-1134.

Troger’s Base Molecular Scaffolds
J . Org. Chem., Vol. 65, No. 7, 2000 1913
which was first purified by column chromatography. Further
purification was done by PTLC using EtOH in C6H6 (5%) to
give the pure product 1 (0.09 g, 15%), mp 110-2 °C: 1H NMR
(200 MHz, CDCl3) δ 8.46 (s, 2H, NH), 8.12 (d, 2H, J ) 8 Hz),
7.72 (d, 2H, J ) 8 Hz), 7.60 (t, 2H, J ) 8 Hz), 7.56 (s, 2H),
7.23 (d, 2H, J ) 8 Hz), 6.90 (d, 2H, J ) 8 Hz), 4.78 (d, 2H, J
) 16 Hz), 4.35 (s, 2H), 4.29 (d, 2H, J ) 16 Hz), 2.45 (s, 6H);
13C NMR (50 MHz, CDCl3) δ 164.9, 156.8, 156.6, 151.8, 138.7,
129.9, 127.9, 126.7, 126.4, 125.3, 119.3, 110.9, 66.7, 58.7, 29.7;
IR (CHCl3) νmax 3361, 2924, 2857, 2327, 1670, 1603, 1528, 1390,
1297, 1201, 1090 cm-1; UV λmax (log ꢀ) 293 (4.40), 254 (4.14) [c
1.43 × 10-4 M, CHCl3]; mass (EI, m/e) 490.2 (M+, 69%), 383.3
(100%); Anal. calcd for C29H26N6O2: C, 71.00; H, 5.34; N, 17.13.
Found: C, 70.65; H, 5.25; N, 16.93.
6-Meth yl-2-(4-n itr op h en yla cetyla m in o)p yr id in e 9. To
a solution of 4-nitrophenylacetylchloride (0.505 g, obtained by
heating 4-nitrophenylacetic acid with an excess of thionyl
chloride at 60 °C for 12 h) in dry CH2Cl2 (10 mL), 2-amino-6-
methylpyridine (0.302 g, 2.8 mmol) and triethylamine (1 mL)
were added dropwise at room temperature under nitrogen
atmosphere. The mixture was stirred overnight. The organic
layer was washed with water followed by saturated sodium
bicarbonate solution and dried over Na2SO4. After evaporation
of the solvent, the product was purified by silica gel chroma-
tography ( 0.58 g, 81%) mp 118 °C: 1H NMR (200 MHz, CDCl3)
δ 8.23 (d, 2H, J ) 8 Hz), 7.95 (d, 1H, J ) 8 Hz), 7.85 (s, 1H,
-NHCOCH3), 7.58 (t, 1H, J ) 8 Hz), 7.52 (d, 2H, J ) 8 Hz),
6.89 (d, 1H, J ) 8 Hz), 3.80 (s, 2H), 2.42 (s, 3H).
We have also studied the chiral resolution process of
racemic Troger’s base receptors by using a chiral acid
through hydrogen bonding interaction, which is thus a
manifestation of chiral recognition. Though the chiral
resolution was successful in NMR studies, chromato-
graphically it did not work. This is likely to be of practical
interest when interactions with various chiral guest
molecules are to be investigated.
Exp er im en ta l Section
Gen er a l. All reactions were conducted under dry nitrogen
and stirred magnetically. The following solvents were freshly
distilled prior to use: tetrahydrofuran (THF) from sodium and
benzophenone, ethanol from CaO, magnesium turnings and,
I2, and methylene chloride and triethylamine from calcium
hydride. All other solvents and reagents were of reagent-grade
quality and used without further purification. All chemical
shift values shown are in δ scales. Melting points were
recorded in open capillaries and are uncorrected. Elemental
analyses were done on a CHN analyzer. Mass spectral data
are given as m/z (% abundance). Infrared spectra were taken
in CHCl3 and as a thin film on KBr plates. The NMR titrations
were carried out in CDCl3 at 25 °C. Error limits of binding
constant values are within (5%.
2,8-Bis(eth oxycar bon yl)-6H,12H-5,11-m eth an odiben zo-
[b,f][1,5]-d ia zocin e 6. A mixture of p-aminoethylbenzoate (0.1
g, 0.60 mmol) and hexamethylenetetramine (0.08 g, 0.60 mmol)
in trifluoroacetic acid (TFAA) (2 mL) was stirred at 60-70 °C.
After 60 h, TFAA was removed by distillation. The residue
was taken in 5 mL of water, poured into a separatory funnel,
and basified by the addition of concentrated NH4OH. The
aqueous layer was then extracted with CH2Cl2 (3 × 50 mL).
The combined organic phases were dried (Na2SO4), filtered,
and concentrated in vacuo to give the crude product as a yellow
glass foam. The crude product was purified by column chro-
matography using EtOAc in CHCl3 (10%) as the eluent to give
the product 6 (0.05 g, 23%), mp 126-128 °C: 1H NMR (200
MHz, CDCl3) δ 7.81 (d, 2H, J ) 8 Hz), 7.62 (s, 2H), 7.16 (d,
2H, J ) 8 Hz), 4.73 (d, 2H, J ) 16 Hz), 4.38 (m, 4H), 4.28 (s,
2H), 4.24 (d, 2H, J ) 16 Hz), 1.34 (t, 6H, J ) 6 Hz); 13C NMR
(50 MHz, CDCl3) δ 165.7, 152.3, 128.7, 127.2, 126.1, 124.7,
66.6, 60.5, 58.6, 14.3; IR (CHCl3) νmax 2891, 2913, 1712, 1609,
1281, 1180, 1110, 1022 cm-1; UV λmax (log ꢀ) 284 (4.33), 250
(4.39) [c 1.09 × 10-4 M, CHCl3].
2,8-Bis((2-ca r boxa m id o-6-m eth ylp yr id in e)m eth yl)-6H,-
12H-5,11-m eth a n od iben zo[b,f][1,5]-d ia zocin e 2. The nitro
compound 9 was reduced to the corresponding amine by
hydrogen in presence of Pd/C as catalyst. The catalyst was
filtered off, and the solvent was evaporated in vacuo to give
the amine. The crude amine (0.14 g, 0.58 mmol) was next
dissolved in trifluoroacetic acid (TFAA) (4 mL). Hexamethyl-
enetetramine (0.08 g, 0.58 mmol) was added, and the mixture
was stirred at 60 °C for 60 h under N2 atmosphere. TFAA was
removed by distillation, and the residue was diluted with
water. The mixture was basified by the addition of NH4OH,
and the aqueous layer was extracted with three 50 mL portions
of CH2Cl2. The combined organic layer was dried over Na2-
SO4, filtered, and concentrated in vacuo to give a brown glass
foam. After initial purification by column chromatography,
final pure product 2 (0.20 g, 62%) was obtained by PTLC using
EtOH in C6H6 (5%) as the eluent, mp 70 °C: 1H NMR (200
MHz, CDCl3) δ 7.97 (d, 2H, J ) 8 Hz), 7.81 (s, 2H, NH), 7.54
(t, 2H, J ) 8 Hz), 7.12 (s, 4H), 6.87 (s, 2H), 6.85 (d, 2H, J ) 8
Hz), 4.68 (d, 2H, J ) 16 Hz), 4.30 (s, 2H), 4.16 ((d, 2H, J ) 16
Hz), 3.58 (s, 4H), 2.38 (s, 6H); 13C NMR (75 MHz, CDCl3) δ
169.4, 156.6, 150.4, 147.4, 138.6, 129.6, 128.8, 128.4, 127.8,
125.7, 119.3, 110.8, 66.7, 58.5, 44.4, 29.6; IR (KBr): νmax 3530,
2-(4-Am in oben zoyla m in o)-6-m eth ylp yr id in e 8. To a
stirred solution of 4-nitrobenzoyl chloride (1.71 g, 9.24 mmol)
in dry dichloromethane (30 mL) was added dropwise 2-amino-
6-methylpyridine (1 g, 9.24 mmol) and triethylamine (1.3 mL)
at room temperature under nitrogen. The mixture was left
overnight under stirring, and water (20 mL) was added. The
organic layer was washed with water followed by saturated
sodium bicarbonate solution and dried (Na2SO4). After evapo-
ration of the solvent, the product 7 was further purified by
silica gel chromatography (2.14 g, 90%), mp 128 °C. A mixture
of compound 7 (0.52 g, 2.02 mmol), Pd/C (10%) (0.1 g), and
ethanol (15 mL) was next stirred at room temperature under
hydrogen atmosphere overnight. The catalyst was filtered off,
and the solvent was evaporated in vacuo to give product 8 (0.41
g, 89%), which was almost pure for the next run. 1H NMR (200
MHz, CDCl3) δ 8.40 (s, 1H, NH), 8.16 (d, 1H, J ) 8 Hz), 7.77
(d, 2H, J ) 8 Hz), 7.60 (t, 1H, J ) 8 Hz), 6.86 (d, lH, J ) 8
Hz), 6.69 (d, 2H, J ) 8 Hz), 4.02 (bs, 2H, -NH2), 2.46 (s, 3H).
2,8-Bis(2-ca r boxa m id o-6-m eth ylp yr id in e)-6H,12H-5,11-
m eth a n od iben zo[b,f][1,5]-d ia zocin e 1. A solution of com-
pound 8 (0.4 g, 1.76 mmol) and hexamethylenetetramine (0.25
g, 1.76 mmol) in 5 mL of trifluoroacetic acid (TFAA) was stirred
at 60-70 °C. After 70 h the TFAA was removed by distillation.
The residue was taken up in 20 mL of H2O, poured into a
separatory funnel, and basified by the addition of NH4OH. The
aqueous layer was then extracted with CH2Cl2 (3 × 50 mL).
The combined organic phases were dried over Na2SO4, filtered,
and concentrated in vacuo to give a yellow brown glass foam,
2920, 1669, 1600, 1574, 1538, 1451, 1391, 1296, 1150 cm-1
;
UV λmax (log ꢀ) 280 (4.29), 241 (4.45) [c 7.72 × 10-5 M, CHCl3];
mass (EI, m/e) 519.4 (MH+, 26%), 518.2 (M+, 7%), 384.2 (4%),
275.2 (12%), 135.2 (36%), 109.1 (100%); Anal. calcd for
C
31H30N6O2: C, 71.79; H, 5.83; N, 16.20. Found: C, 71.50; H,
5.70; N, 16.32.
2-Acetyla m in o-6(4-n itr ostyr yl)p yr id in e 10. A solution
consisting of 2-acetylamino-6-methylpyridine (0.8 g, 5.33
mmol), 4-nitrobenzaldehyde (0.805 g, 5.33 mmol), and 5 mL
of acetic anhydride was heated 175 °C for 50 h. The cooled
solution was poured onto ice, and the mixture was then made
alkaline with 40% aqueous NaOH. The product was then
extracted from water mixture with ethyl acetate. Evaporation
of the solvent in vacuo gave the impure desired product.
Purification by column chromatography [using benzene:ethyl
acetate (3:1)] gave the product 10 (0.72 g, 48%), mp 175-178
°C: 1H NMR (200 MHz, CDCl3) δ 8.24 (d, 2H, J ) 8 Hz), 8.13
(t, 1H, J ) 8 Hz), 7.98 (bs, 1H, NH), 7.75-7.64 (m, 4H), 7.60
(d, 1H, J ) 16 Hz), 7.11 (d, 1H, J ) 8 Hz), 2.25 (s, 3H).
2-Acetyla m in o-6-[(4-a m in op h en yl)eth yl)]p yr id in e 11.
Compound 10 (0.3 g, 1.06 mmol) and 10% Pd/C (0.1 g) were
taken in ethanol and stirred overnight at room temperature
under hydrogen atmosphere. The catalyst was filtered off, and
the solvent was evaporated in vacuo to give compound 11

1914 J . Org. Chem., Vol. 65, No. 7, 2000
Goswami et al.
(0.19g, 70 %): 1H NMR (200 MHz, d6-DMSO) δ 9.58 (s, 1H,
NH), 8.14 (d, 1H, J ) 8 Hz), 7.63 (t, 1H, J ) 8 Hz), 6.92 (d,
2H, J ) 8 Hz), 6.82-6.71 (m, 3H), 4.53 (bs, 2H, -NH2), 2.94
(bs, 4H), 2.20 (s, 3H).
the pure product 16 (0.86 g, 78%), mp 218 °C: 1H NMR (200
MHz, d6-DMSO ) δ 7.98-7.90 (m, 4H), 7.58 (s, 2H), 7.42 (d,
2H, J ) 4 Hz), 3.07 (s, 4H).
1,2-Bis[3-(2-ca r b oxa m id o-6-m et h ylp yr id in e)p h en yl]-
eth a n e 5. To a stirred solution of diacid 16 (0.3 g, 1.1 mmol)
in dry CH2Cl2, oxalyl chloride (0.19 mL, 2.2 mmol) and a drop
of dry DMF (catalytic) were added. The mixture was stirred
at room temperature under N2 atmosphere for 1.5 h. The
solvent was evaporated under N2 atmosphere. The solid mass
was redissolved in dry CH2Cl2, and a solution of 2-amino-6-
methylpyridine (0.19 g, 1.74 mmol) containing Et3N (0.5 mL)
was added dropwise. The mixture was stirred overnight. The
usual workup and purification by silica gel column chroma-
tography using petroleum ether in CHCl3 (30%) gave the title
compound as semisolid (0.53 g, 68%). 1H NMR (200 MHz,
CDCl3) δ 8.60 (bs, 2H, -NH), 8.20 (d, 2H, J ) 8 Hz), 7.91 (m,
2H), 7.87-7.54 (m, 6H), 7.32 (d, 2H, J ) 6 Hz), 6.92 (m, 2H),
2.48 (s, 4H), 2.43 (s, 6H); IR (KBr) νmax 3295, 3050, 2920, 1673,
2,8-Bis((2-a cet yla m in o-6-m et h ylp yr id in e)et h yl)-6H ,-
12H-5,11-m eth a n od iben zo[b,f][1,5]-d ia zocin e 3. A mixture
of the amine 11 (0.09 g, 0.35 mmol), hexamethylenetetramine
(0.05 g, 0.35 mmol), and TFAA (3 mL) was initially stirred at
room temperature (25 °C) for 3 h and then heated at 60 °C
with stirring for 50 h under N2 atmosphere. The mixture was
cooled to rt, diluted with water, and neutralized by NH4OH.
The mixture was extracted with CH2Cl2 (4 × 5 mL). The
organic layer was next washed with saturated NaHCO3
solution and dried over anhydrous Na2SO4. The volatiles were
removed, and the crude product was initially purified by
column chromatography on silica gel using EtOH in C6H6 (6%)
as the eluent. The final purification by PTLC using the same
solvent as the developing agent afforded the product 3 (0.12
g, 63%), mp 80 °C: 1H NMR (200 MHz, CDCl3) δ 7.98 (d, 2H,
J ) 8 Hz), 7.86 (bs, 2H, NH), 7.54 (t, 2H, J ) 8 Hz), 7.06-
6.99 (m, 4H), 6.78 (d, 2H, J ) 8 Hz), 6.71 (s, 2H), 4.64 (d, 2H,
J ) 18 Hz), 4.29 (s, 2H), 4.09 (d, 21, J ) 18 Hz), 2.86 (s, 8H),
2.19 (s, 6H); 13C NMR (50 MHz, CDCl3) 166.6, 160.4, 150.6,
146.0, 139.5, 136.9, 127.6, 127.3, 126.6, 124.9, 118.6, 111.0,
66.8, 58.5, 39.5, 35.1, 25.1; IR (KBr) νmax 3400, 3040, 2915,
1682. 1595, 1572, 1537, 1449 cm-1; UV λmax (log ꢀ) 281 (4.22),
240 (4.35) [c 9.16 × 10-5 M, CHCl3]; mass (EI, m/e) 547.5 (MH+,
27%), 339.1 (14%), 55.0 (100%); Anal. calcd for C33H34N6O2:
C, 72.25; H, 6.26; N, 15.37. Found: C, 71.98; H, 6.19; N, 15.23.
4′-Nitr o-4-stilba zole 12. A mixture of 4-nitrobenzaldehyde
(0.80 g, 5.29 mmol), 4-picoline (0.49 g, 5.30 mmol), and 10 mL
of acetic anhydride was refluxed at 180 °C, poured onto ice,
and neutralized with 50% sodium hydroxide solution. The
mixture was extracted with EtOAc and dried (Na2SO4). The
solvent was evaporated in vacuo, and the crude product was
purified by column chromatography using EtOAc in petroleum
ether (20%) as the eluent to give the pure compound 12 (0.80
g, 57%), mp 140-142 °C: 1H NMR (200 MHz, CDCl3) δ 8.63
(m, 2H), 8.25 (d, 2H, J ) 8 Hz), 7.68 (d, 2H, J ) 8 Hz), 7.41
(m, 2H), 7.28 (d, 1H, J ) 10 Hz), 7.17 (d, 1H, J ) 16 Hz).
2,8-Bis((4-pyr idyl)eth yl)-6H,12H-5,11-m eth a n od iben zo-
[b,f][1,5]-d ia zocin e 4. The crude amine 13 (0.10 g, 0.40
mmol), obtained from catalytic hydrogenation (H2, Pd/C) of 12,
was dissolved in TFAA (3 mL). Hexamethylenetetramine was
added, and the mixture was heated at 60 °C with stirring
under N2 atmosphere for 40 h. The mixture was cooled to rt,
diluted with water, and neutralized by NH4OH. The mixture
was extracted with CH2Cl2 (4 × 5 mL). The organic layer was
thoroughly washed with saturated NaHCO3 solution and dried
over anhydrous Na2SO4. Solvent was removed in vacuo, and
the crude product was purified by column chromatography on
silica gel using EtOH in C6H6 (6%) as the eluent to afford the
1598, 1573 cm-1
.
Exp er im en ta l P r oced u r e for NMR Titr a tion . As a
specific example, the control titration experiment of Troger’s
base receptor 1 with suberic acid will be described here. A 3.46
× 10-3 M solution of 1 in dry CDCl3 was taken in an NMR
tube. A 1.33 × 10-2 M solution of suberic acid in dry CDCl3
containing 2% d6-DMSO was prepared. An initial NMR
spectrum of 1 was taken, and the initial chemical shift of the
amide proton was determined to be 8.38 ppm. The solution of
guest suberic acid was then progressively added, in 10, 20,
30, 40 µL proportions, etc., to the solution of 1, and the
chemical shift of amide proton was recorded after each
addition. Such addition was continued until no further change
in the chemical shift of the amide proton was observed. The
chemical shift of the amide proton at the saturation point was
1.70 ppm. The temperature of the NMR probe was 25 °C. The
change in chemical shift (∆δ) values were calculated by
subtracting the chemical shift at each titration point from the
chemical shift value of pure host. Thus, a titration curve of
∆δ vs concentration ratio of guest/host and a binding constant
curve of ∆δ/Cguest vs ∆δ were plotted. Foster-Fyfe analysis of
the titration data at 25 °C gave the value of Ka (1.5 × 104 M-1),
and ∆G was calculated from the relation ∆G ) -RT ln Ka.
F lu or escen ce Exp er im en t. Stock solutions of receptors
were prepared in CHCl3, and guests were dissolved in CHCl3
containing 2% DMSO. The titrations were carried out by
successive addition of guest solution to the solution of receptor
placed in a fluorescence curette. Fluorescence spectra were
measured using a PERKIN ELMER LS 50B spectrofluorim-
eter, using a conventional 1 × 1 cm quartz cell at 25 °C with
the excitation and emission slits of 6 nm width. The solution
of receptor 1 at a concentration of 2.04 × 10-4 M was excited
at 330 nm with the emission being monitored at 404 nm, and
the fluorescence intensity at the emission maximum was used
to determine the complex stability constant. A similar experi-
ment was done with receptor 2 (2.57 × 10-4 M) at excitation
and emission wavelengths 330 and 384 nm, respectively, with
suberic acid (4.14 × 10-3 M) at 25 °C.
1
low-melting yellowish solid (0.12 g, 70%). H NMR (200 MHz
CDCl3): δ 8.44 (d, 4H, J ) 8 Hz), 7.07-6.94 (m, 8H), 6.68 (s,
2H), 4.64 (d, 2H, J ) 18 Hz), 4.29 (s, 2H), 4.09 (d, 2H, J ) 18
Hz), 2.80 (bs, 8 H); 13C NMR (50 MHz, CDCl3) 150.3, 149.6,
146.3, 136.3, 127.7, 127.3, 126.6, 125.0, 123.8, 66.9, 59.8, 36.9,
35.9; IR (CHCl3) νmax 3420, 3015, 2890, 1669, 1595, 1552, 1487,
1410 cm-1; UV λmax (log ꢀ) 294 (3.79), 251 (4.16) [c 1.39 × 10-4
M, CHCl3]; mass (EI, m/e) 432.3 (M+, 30%), 340.3 (22%), 281.3
(50%), 188.2 (100%).
3,3′-Eth ylen ed iben zoic Acid 16. 1-Carboethoxy-3-bro-
momethylbenzene 14 (1 g, 4.1 mmol) was treated with Co-
(PPh3)3Cl (4.34g, 4.9 mmol) in degassed dry benzene (70 mL)
at room temperature for 20 min under Ar atmosphere. The
reaction mixture was filtered, and the filtrate was washed with
water and concentrated to dryness. The crude 15 was hydro-
lyzed by 60 mL of 3N NaOH for 3-4 h at 80 °C. The solution
was then neutralized by dilute HCl and extracted by EtOAc.
The solvent was evaporated, and the residue was chromato-
graphed on a silica gel column using EtOAc:MeOH (3:1) to give
Ack n ow led gm en t. We thank DST and CSIR, Gov-
ernment of India for financial support. We appreciate
the help of IACS, Calcutta, for elemental analysis and
IIT, Kharagpur, for NMR spectra and fluorescence
studies. We acknowledge Dr. R. S. Pilato of University
of Maryland, College Park, MD, for the mass spectra.
We thank Reshmi Mukherjee for her assistance.
Su p p or tin g In for m a tion Ava ila ble: Spectral data for
1-6 and binding constant curve, titration curve, and fluores-
cence binding curve for suberic acid with 1. This material is
available free of charge via the Internet at http://pubs.acs.org.
J O9909204