Establishing supramolecular control over solid-state architectures: A simple mix and match strategy

Article abstract of DOI:10.1021/cg300180b

With the help of robust principles of crystal engineering, it is possible to construct co-crystals where two or more different molecular entities coexist in the same crystalline lattice; the supramolecular assembly is driven by noncovalent interactions, most commonly by hydrogen bonds. We have synthesized two ditopic amide based ligands (N-(4-pyridin-2-yl)isonicotinamide) and (N-(3-pyridin-2-yl)nicotinamide) and systematically established their binding preferences when faced with aliphatic dicarboxylic acids with an odd and even number of carbon atoms. Each ligand was co-crystallized with four odd and four even-chain dicarboxylic acids, and 13/16 reactions produced crystals suitable for single-crystal structure determination. On the basis of these results, it is clear that carefully selected systems can be manipulated to produce assemblies in the solid state with very precise control over topology and dimensionality. These ligands can be made to produce either 0-D or 1-D architectures simply by fine-tuning the choice of co-crystallizing agent in the supramolecular synthesis. This mix-and-match strategy allows us to mimic the reliability and versatility of covalent synthesis, in terms of successfully preparing a target with predetermined connectivity and metrics.

Full text of DOI:10.1021/cg300180b

Article
pubs.acs.org/crystal
Establishing Supramolecular Control over Solid-State Architectures:
A Simple Mix and Match Strategy
Christer B. Aakeroy,* Sheelu V. Panikkattu, Baillie DeHaven, and John Desper
̈
Department of Chemistry, Kansas State University, 213 CBC Building, Manhattan, Kansas 66506, United States
S
* Supporting Information
ABSTRACT: With the help of robust principles of crystal
engineering, it is possible to construct co-crystals where two or
more different molecular entities coexist in the same crystalline
lattice; the supramolecular assembly is driven by noncovalent
interactions, most commonly by hydrogen bonds. We have
synthesized two ditopic amide based ligands (N-(4-pyridin-2-yl)-
isonicotinamide) and (N-(3-pyridin-2-yl)nicotinamide) and
systematically established their binding preferences when faced
with aliphatic dicarboxylic acids with an odd and even number of
carbon atoms. Each ligand was co-crystallized with four odd and
four even-chain dicarboxylic acids, and 13/16 reactions produced crystals suitable for single-crystal structure determination. On
the basis of these results, it is clear that carefully selected systems can be manipulated to produce assemblies in the solid state
with very precise control over topology and dimensionality. These ligands can be made to produce either 0-D or 1-D architectures
simply by fine-tuning the choice of co-crystallizing agent in the supramolecular synthesis. This mix-and-match strategy allows us to
mimic the reliability and versatility of covalent synthesis, in terms of successfully preparing a target with predetermined connectivity
and metrics.
rystal engineering1,2 involves a combination of synthesis
and structural chemistry and requires a detailed under-
self-assembly to take place during supramolecular synthesis.6 An
example where molecular geometry has a profound effect on
properties is readily observed in the melting points of even-
number aliphatic dicarboxylic acids, which decrease monotoni-
cally with increasing number of methylene groups in the chain.
The melting points of odd-number diacids, however, are
significantly lower than their even-number analogues due to
considerable differences in crystal structure/packing.7,8 Similarly,
melting points of co-crystals9 of such diacids with bipyridyl
molecules that differ in the total number of carbon atoms
connecting the two pyridyl rings10 display a pattern that can
be correlated with the sum of carbon atoms in the repeating
unit.
In the present study, we examine how two fundamentally
different motifs, rings, and chains can be produced through the
appropriate co-crystallizations of odd and even chain aliphatic
dicarboxylic acids with two pyridine-based molecules differing
in the relative orientation of the two hydrogen-bond acceptor
sites. The two synthons employed in the assembly strategy are
(a) the carboxylic acid···pyridine synthon11,12 and (b) the
carboxylic acid····2-aminopyridine synthon,13 and since both
have ample precedents in the literature, we can focus primarily
on manipulating the relative positions of the binding sites in
order to develop direct and rational links between molecular
geometry and supramolecular structure. Even-chain diacids
C
standing of intermolecular interactions. Since any supra-
molecular synthesis by definition relies on reversible and
relatively weak interactions (when compared to most covalent
bonds), several problems must be overcome before we can
attain a reliability and complexity that begin to approach what
can be achieved in conventional covalent synthesis. Despite
substantial progress in the last decades, it is still very challenging
to utilize specific intermolecular interactions in the construction
of solid-state assemblies with desired dimensionality and
composition.3 However, if we were able to develop a larger
library of robust and predictable supramolecular synthons,4 we
would undoubtedly get closer to the goal of being able to attach
specific molecular-recognition sites onto discrete molecules
in such a way that we can preprogram subsequent assembly
processes that carry over from solution into the solid state.
Hydrogen bonds5 largely depend on the chemical nature and
electronic environment around the binding site, but at the
supramolecular level, the relative orientation and geometry of the
groups can determine shape, dimension, and metrics of the
resulting noncovalent assembly. Therefore, synthesis of a
desirable 3-D architecture in the solid state requires a careful
choice of target building blocks with complementary geometries
and binding patterns. Furthermore, it would be very useful if
information about molecular geometry of individual covalent
building blocks could be translated into a prediction (or at least
an expectation) regarding the structure and properties of the
resulting assembly. In short, we need a method for equipping
each building block with a code that allows for a specific
Received: February 7, 2012
Revised: March 8, 2012
Published: March 28, 2012
© 2012 American Chemical Society
2579
dx.doi.org/10.1021/cg300180b | Cryst. Growth Des. 2012, 12, 2579−2587

Crystal Growth & Design
Article
(referring to the number of carbon atoms in the chain) typically
have their molecular recognition sites arranged in a colinear
fashion, whereas in odd-chain diacids, they are positioned at an
angle of approximately 120°, with respect to each other
(Scheme 1). To complement these two types of orientation of
of 13 new co-crystals will be analyzed in the context of these
postulated architectures.
EXPERIMENTAL SECTION
■
All precursors for the synthesis of 3N and 4N were purchased from
Aldrich and used without further purification. Melting points were
determined using Fischer-Johns Mel-Temp melting point apparatus
and uncorrected. 1H NMR spectra were recorded on Varian 400 MHz
and 200 MHz spectrometer. IRs of compounds and co-crystals were
taken on Nicolet 380 FT-IR.
Scheme 1
Synthesis of 4N and 3N. (N-(4-pyridin-2-yl)isonicotinamide),
4N. The synthesis of 4N was carried out using a modified
procedure.14 Isonicotinic acid (1.21 g, 10 mmol) and triethylamine
(1.011 g, 10 mmol) were dissolved in 30 mL of chloroform.
Thionyl chloride (1.189 g, 10 mmol) was added dropwise under a
dinitrogen atmosphere maintaining the temperature between 5 and
10 °C. After the addition was complete, the mixture was heated
under reflux for 3 h at the end of which a brown solution was
obtained. The mixture was cooled to r.t., then a solution of amino
pyridine (1.08 g, 10 mmol), triethylamine (1.01 g, 10 mmol), and
acetonitrile (20 mL) was added in situ and stirred at r.t. overnight.
The organic layer was washed with water, and the solvent was
removed. Column chromatography gave a white powder as the
product. Yield 41%, mp 138−140 °C (lit. 136−138 °C).15 1H
NMR (δH, 400 MHz, MeOD): 8.748 (s, 2H), 8.363 (s, 1H), 8.245
(d, J = 8.2 Hz, 1H), 7.902−7.801 (m, 3H), 7.208−7.153 (m, 1H).
(N-(3-pyridin-2-yl)nicotinamide), 3N. Nicotinic acid (1.21 g, 10
mmol) and triethylamine (1.011 g, 10 mmol) were dissolved in 30 mL
of chloroform. Thionyl chloride (1.189 g, 10 mmol) was added
dropwise under a dinitrogen atmosphere maintaining the temperature
between 5 and 10 °C. After the addition was complete, the mixture
was heated under reflux for 3 h at the end of which a brown solution
was obtained. The mixture was cooled to r.t., then a solution of amino
pyridine (1.08 g, 10 mmol), triethylamine (1.01 g, 10 mmol), and
acetonitrile (20 mL) was added in situ and stirred at r.t. overnight. The
organic layer was washed with water, and the solvent was removed,
followed by column chromatography to yield an off-white powder as
the product. Yield 35%, mp 142−145 °C (lit. 141−143 °C).16 1H
NMR (δH, 400 MHz, DMSO): 11.094 (s, N−H 1H), 9.135 (s, 1H),
8.761 (d, 1H), 8.409 (d, 1H), 8.340 (d, 1H), 8.120 (d, 1H), 7.845 (t,
1H), 7.541 (t, 1H), 7.189 (t, 1H).
hydrogen-bond donor sites, we will examine (N-(3-pyridin-2-yl)-
nicotinamide), 3N, where the hydrogen-bond acceptor sites
are almost parallel and (N-(4-pyridin-2-yl)isonicotinamide), 4N,
where the binding sites are positioned at an angle of
approximately 60° with respect to each other (Scheme 2).
Scheme 2
If these complementary molecules form co-crystals, it is
possible to envision different assemblies, depending upon how
the donor/acceptor sites are arranged on the two participating
tectons. In Scheme 3, we have outlined how the building-blocks
could produce either 0-D rings or 1-D chains held together by
alternating −COOH···N(py) and −COOH···N(py)-NH2
synthons. The results from the crystal structure determinations
Synthesis of Co-crystals of 4N and 3N. 4N and 3N were co-
crystallized with four odd-chain diacids (malonic, glutaric, pimelic, and
azelaic acid) and four even-chain diacids (succinic, adipic, suberic. and
sebacic acid). The reactions included both a melt and a solvent-assisted
Scheme 3. Possible Architectures from Building Blocks 3N and 4N
2580
dx.doi.org/10.1021/cg300180b | Cryst. Growth Des. 2012, 12, 2579−2587

Crystal Growth & Design
Article
Figure 1. (a) IR spectrum of 4N; (b) IR of the product resulting from a combination of 4N and pimelic acid (4N-PA) showing two broad stretches
near the 1850 and 2500 cm−1 regions; (c) IR spectrum of 3N; (d) IR of the product resulting from the combination of 3N and adipic acid (3N-AA)
showing two broad stretches near the 1850 and 2450 cm−1 regions.
grinding experiment after which the resulting powder/product was
characterized using IR spectroscopy to determine if a co-crystal had
formed. A total of sixteen experiments were performed, and all IR spectra
displayed broad stretches near 1850 and 2450 cm−1 (Figure 1) indicative
of intermolecular O−H···N hydrogen bonds confirming co-crystal
formation in every case. Stoichiometric amounts of base and acid
(1:1) were subsequently dissolved separately in suitable solvents and
then mixed together. The resulting solution was warmed and then set
aside for slow evaporation at room temperature. Crystals suitable for
single-crystal X-ray diffraction were obtained for 13 of the 16 products.
(N-(4-pyridin-2yl)isonicotinamide) Glutaric Acid (1:1), 4N-GA. A
solution of 4N (10 mg, 0.05 mmol) in 2 mL of ethyl acetate was mixed
with glutaric acid (6.6 mg, 0.05 mmol) in 2 mL of ethyl acetate and
warmed. Transparent rod-like crystals were obtained within 20 days.
Mp 127−129 °C.
(N-(4-pyridin-2yl)isonicotinamide) Pimelic Acid (1:1), 4N-PA. A
solution of 4N (10 mg, 0.05 mmol) in 2 mL of ethyl acetate was mixed
with pimelic acid (8.0 mg, 0.05 mmol) in 2 mL of ethyl acetate and
warmed. Transparent needle-like crystals were obtained. Mp 123−
125 °C.
(N-(4-pyridin-2yl)isonicotinamide) Azelaic Acid (1:1), 4N-AzA. A
solution of 4N (10 mg, 0.05 mmol) in 2 mL of ethyl acetate was mixed
with azelaic acid (9.41 mg, 0.05 mmol) in 2 mL of ethyl acetate and
warmed. Transparent needle-like crystals were formed after 45 days.
Mp 92−95 °C.
(N-(4-pyridin-2yl)isonicotinamide) Succinic Acid (1:1), 4N-SA. A
solution of 4N (10 mg, 0.05 mmol) in 2 mL of ethyl acetate was mixed
with succinic acid (5.9 mg, 0.05 mmol) in 2 mL of ethyl acetate and
warmed. Transparent needles were obtained. Mp 122−125 °C.
(N-(4-pyridin-2yl)isonicotinamide) Adipic Acid (1:1), 4N-AA. A
solution of 4N (10 mg, 0.05 mmol) in 2 mL of ethyl acetate was mixed
with adipic acid (7.31 mg, 0.05 mmol) in 2 mL of ethyl acetate and
warmed. Transparent block-shaped crystals were formed. Mp 142−
145 °C.
(N-(4-pyridin-2yl)isonicotinamide) Suberic Acid (1:1), 4N-SuA. A
solution of 4N (10 mg, 0.05 mmol) in 2 mL of ethyl acetate was mixed
with suberic acid (8.71 mg, 0.05 mmol) in 2 mL of ethyl acetate and
warmed. Transparent rod-shaped crystals were obtained within few
weeks. Mp 133−135 °C.
(N-(4-pyridin-2yl)isonicotinamide) Sebacic Acid (1:1), 4N-SeA. A
solution of 4N (10 mg, 0.05 mmol) in 2 mL of ethyl acetate was mixed
with sebacic acid (10.11 mg, 0.05 mmol) in 2 mL of ethyl acetate and
warmed. Transparent rod-like crystals were formed. Mp 104−107 °C.
(N-(3-pyridin-2yl)nicotinamide) Malonic Acid (1:1), 3N-MA. A
solution of 3N (10 mg, 0.05 mmol) in 2 mL of ethyl acetate was mixed
with malonic acid (5.23 mg, 0.05 mmol) in 2 mL of ethyl acetate and
warmed. Transparent rod-like crystals were obtained within few weeks.
Mp 125−127 °C.
(N-(3-pyridin-2yl)nicotinamide) Pimelic Acid (1:1), 3N-PA. A
solution of 3N (10 mg, 0.05 mmol) in 2 mL of ethyl acetate was
2581
dx.doi.org/10.1021/cg300180b | Cryst. Growth Des. 2012, 12, 2579−2587

Crystal Growth & Design
Article
a
Table 1. Hydrogen Bond Geometries of the Co-Crystals of 4N
structure
4N-GA
D−H···A
O31 H31 N11
D-H (Å)
H···A (Å)
D···A (Å)
D−H−N (deg)
0.962(17)
0.855(14)
0.898(15)
0.90(3)
1.712(17)
2.164(14)
1.786(15)
2.08(3)
2.6713(11)
3.0103(11)
2.6570(11)
2.970(3)
174.6(14)
170.7(13)
162.9(15)
173(2)
N17 H17 O36_#1
O35 H35 N21_#1
N17 H17 O32
4N-PA
4N-AzA
4N-SA
O31 H31 N21
0.95(3)
1.74(3)
2.679(3)
169(3)
O37 H37 N11
1.10(3)
1.57(3)
2.665(3)
172(2)
O31 H31 N21
0.942(15)
0.862(14)
0.957(15)
0.89(2)
1.738(15)
2.063(14)
1.785(15)
2.12(2)
2.6756(11)
2.9200(12)
2.7413(12)
3.005(2)
173.6(13)
172.7(12)
176.3(13)
174(2)
N17 H17 O32
O39 H39 N11_#1
N17_1 H17_1 O52_1
N17_2 H17_2 O52_2
N17_3 H17_3 O52_3
O51_1 H51_1 N21_1
O51_2 H51_2 N21_2
O51_3 H51_3 N21_3
O54_1 H54_1 N41_1
O54_2 H54_2 N41_2
O54_3 H54_3 N41_3
O61_1 H61_1 N11_1
O61_2 H61_2 N11_2
O61_3 H61_3 N11_3
O64_1 H64_1 N31_1_#1
O64_2 H64_2 N31_2_#1
O64_3 H64_3 N31_3_#1
N17 H17 O42
0.92(2)
2.13(2)
3.050(2)
174.8(19)
176.5(18)
170(2)
0.91(2)
2.09(2)
3.001(2)
0.97(2)
1.70(2)
2.661(2)
0.96(2)
1.75(2)
2.691(2
165(2)
1.00(2)
1.69(2)
2.681(2)
170(2)
0.98(2)
1.72(2)
2.685(2)
171(2)
0.97(2)
1.74(2)
2.679(2)
162.4(19)
165(2)
0.94(2)
1.79(2)
2.702(2)
0.96(2)
1.72(2)
2.667(2)
173(2)
0.93(2)
1.75(2)
2.668(2)
170(2)
0.97(2)
1.72(2)
2.683(2)
173(2)
0.90(2)
1.80(3)
2.688(2)
168(2)
0.93(2)
1.74(2)
2.670(2)
175(2)
0.94(3)
1.72(3)
2.659(2)
170(2)
4N-AA
4N-SuA
4N-SeA
0.856(18)
0.98(2)
2.058(18)
1.73(3)
2.9131(14)
2.6945(15)
2.6772(14)
2.7133(15)
2.9228(13)
2.6583(13)
2.6693(12)
2.8556(12)
2.6327(12)
177.4(17)
169(2)
O31 H31 N11
O41 H41 N21
0.94(2)
1.74(2)
172.3(19)
179.3(18)
173.6(14)
175.3(14)
174.2(13)
176.2(13)
172.0(13)
O31 H31 N11
1.03(2)
1.69(2)
N17 H17 O39_#1
O38 H38 N21_#2
O31 H31 N11
0.850(16)
1.120(17)
0.999(15)
0.855(14)
1.013(15)
2.077(16)
1.541(17)
1.674(15)
2.002(14)
1.626(15)
N17 H17 O41_#1
O40 H40 N21_#2
a
Symmetry Codes: 4N−GA: #1 1 − x,−y, 1 − z. 4N−PA: #1 2.5 − x, −0.5 + y, 1.5 − z. 4N−AzA: #1 −1 − x, −y, 1 − z. 4N−SA: #1 −1 + x, −1 + y,
z. 4N−SuA: #1 0.5 − x, −0.5 + y, 1.5 − z. #2 0.5 − x, 0.5 + y, 1.5 − z. 4N−SeA: #1 0.5 − x, 0.5 + y, 1.5 − z. #2 0.5 − x, −0.5 + y, 1.5 − z.
mixed with pimelic acid (8.0 mg, 0.05 mmol) in 2 mL of ethyl acetate
and warmed. Transparent rod-like crystals were obtained. Mp 119−
120 °C.
small widely separated matrix runs. Scan speed and scan width were
chosen based on scattering power and peak rocking curves. Unit cell
constants and orientation matrix were improved by least-squares
refinement of the reflections threshold from the entire data set.
Integration was performed with SAINT,17b using this improved unit
cell as a starting point. Precise unit cell constants were calculated in
SAINT from the final merged data set. Lorenz and polarization
(N-(3-pyridin-2yl)nicotinamide) Azelaic Acid (1:1), 3N-AzA. A
solution of 3N (10 mg, 0.05 mmol) in 2 mL of ethyl acetate was mixed
with azelaic acid (9.41 mg, 0.05 mmol) in 2 mL of ethyl acetate and
warmed. Needle-like crystals were formed. Mp 90−93 °C.
(N-(3-pyridin-2yl)nicotinamide) Succinic Acid (1:1), 3N-SA. A
solution of 3N (10 mg, 0.05 mmol) in 2 mL of ethyl acetate was mixed
with succinic acid (5.9 mg, 0.05 mmol) in 2 mL of ethyl acetate and
warmed. Transparent needle-shaped crystals were obtained. Mp 145−
147 °C.
(N-(3-pyridin-2yl)nicotinamide) Adipic Acid (1:1), 3N-AA. A
solution of 3N (10 mg, 0.05 mmol) in 2 mL of ethyl acetate was
mixed with adipic acid (7.31 mg, 0.05 mmol) in 2 mL of ethyl acetate
and warmed. Transparent block-shaped crystals were formed. Mp
140−142 °C.
́
corrections were applied. Laue symmetry, space group, and unit cell
contents were found with XPREP. Data were reduced with
SHELXTL.17c The structures were solved in all cases by direct
methods without incident. Coordinates for the amide and carboxylic
acid protons were allowed to refine; all other hydrogen atoms were
assigned to idealized positions and were allowed to ride. Heavy atoms
were refined with anisotropic thermal parameters. Selected hydrogen
bond geometries are shown in Tables 1 (4N) and 2 (3N), and
crystallographic parameters of 4N and 3N co-crystals are listed in
Tables 3 and 4, respectively.
(N-(3-pyridin-2yl)nicotinamide) Sebacic Acid (1:1), 3N-SeA. A
solution of 3N (10 mg, 0.05 mmol) in 2 mL of ethyl acetate was mixed
with sebacic acid (10.11 mg, 0.05 mmol) in 2 mL of ethyl acetate and
warmed. Needle-shaped crystals were formed. Mp 104−106 °C.
X-ray Crystallography. X-ray data were collected on a Bruker
SMART APEX II (for 4A-GA, 4N-SuA, 3N-AA, 4N-AzA, 3N-SeA,
and 3N-MA or Bruker Kappa APEX II CCD (for 4N-PA, 4N-SeA,
4N-AA, 3N-SA, 4N-SA, 3N-PA, and 3N-AzA) diffractometer at 120 K
using, in all cases, a fine-focus molybdenum Kα tube. Data were
collected using APEX217a software. Initial cell constants were found by
RESULTS
■
Three crystal structures (out of four possible) were obtained
from combinations of 4N with an odd-chain diacid; 4N-GA,
4N-PA, and 4N-AzA. No proton transfer was observed, which
was consistent with the information obtained from the IR data,
Table 5.
In the structure of 4N-GA, the two primary intermolecular
interactions (−COOH···N(py) and −COOH···N(py)-NH2),
2582
dx.doi.org/10.1021/cg300180b | Cryst. Growth Des. 2012, 12, 2579−2587

Crystal Growth & Design
Article
a
Table 2. Hydrogen Bond Geometries of the Co-Crystals of 3N
structure
3N-MA
D−H···A
N17 H17 O32
D−H (Å)
H···A (Å)
D···A (Å)
D−H−N (deg)
0.921(16)
0.969(16)
0.971(17)
0.90(2)
2.225(16)
1.647(16)
1.666(17)
2.08(2)
2.06(2)
1.63(2)
1.78(2)
1.77(2)
1.64(2)
1.71(2)
2.074(19)
1.71(2
3.1144(13)
2.5945(13)
2.6344(13)
2.975(2)
162.1(13)
164.8(14)
174.9(15)
172.3(19)
172(2)
O31 H31 N21
O33 H33 N11_#1
N17_1 H17_1 O32_1
N17_2 H17_2 O32_2
O31_1 H31_1 N21_1
O31_2 H31_2 N21_2
O37_1 H37_1 N11_2
O37_2 H37_2 N11_1_#1
O31 H31 N11
3N-PA
0.89(2)
2.943(2)
1.04(2)
2.6672(19)
2.6611(19)
2.672(2)
172.5(19)
176(2)
0.88(2)
0.91(2)
172.1(19)
172.7(18)
171.3(18)
176.7(16)
169.6(19
172(2)
1.01(2)
2.647(2)
3N-AzA
3N-SA
1.00(2)
2.7055(17)
2.9785(17)
2.6491(17)
2.6581(19)
2.9674(19)
2.6702(18)
2.6832(12)
2.9271(11)
2.6785(11)
2.6898(14)
2.6579(14)
2.9131(13)
N17 H17 O40_#1
O39 H39 N21_#2
O31 H31 N11
0.906(19)
0.95(2)
0.96(2)
1.70(2)
2.12(2)
1.79(2)
1.780(17)
2.009(14)
1.721(17)
1.795(18)
1.79(2)
2.05
N17 H17 O35_#1
O34 H34 N21_#1
O31 H31 N11
0.85(2)
170(2)
0.90(2)
169(2)
3N-AA
3N-SeA
0.909(17)
0.924(14)
0.970(17)
0.898(18)
0.89(2)
172.0(14)
172.2(12)
168.6(14)
174.4(16)
166.3(18)
168.0
N17 H17 O36_#1
O37 H37 N21_#1
O31 H31 N11
O41 H41 N21_#1
N17 H17 O42_#1
0.88
a
Symmetry Codes: 3N−MA: #1 x,1 − y,−0.5 + z. 3N−PA: #1 −2 + x, 1 + y, 1 + z. 3N−AzA: #1 −0.5 −x, 0.5 + y, 0.5 − z. #2 −0.5 − x, −0.5 + y,
0.5 − z. 3N−SA: #1 −x, 1 − y, 1 − z. 3N−AA: #1 −x, 1 − y, −z. 3N−SeA: #1 2 − x, 1 − y, 1 − z.
Table 3. Crystallographic Parameters of 4N Co-Crystals
4N-PA
4N-GA
4N-SeA
4N-AA
4N-SA
4N-AzA
4N-SuA
formula moiety (C11H9N3O)
(C7H12O4)
(C11H9N3O)
(C5H8O4)
(C11H9N3O)
(C10H18O4)
(C11H9N3O)
(C6H10O4)
(C11H9N3O)
(C4H6O4)
(C11H9N3O)
(C9H16O4)
(C11H9N3O)
(C8H14O4)
empirical
formula
C18H21N3O5
C16H17N3O5
C21H27N3O5
C17H19N3O5
C15H15N3O5
C20H25N3O5
C19H23N3O5
molecular
weight
359.38
331.33
401.46
345.35
317.30
387.43
373.40
color, habit
colorless plate
monoclinic
colorless prism
triclinic
colorless prism
monoclinic
P2(1)/n, 4
11.0319(11)
9.9500(9)
18.7354(18)
90.00
colorless prism
monoclinic
P2(1)/c, 4
11.9508(11)
19.8752(19)
7.3996(8)
90.00
colorless prism
triclinic
colorless prism
triclinic
colorless prism
monoclinic
P2(1)/n, 4
11.4438(7)
10.1874(5)
16.5671(9)
90.00
crystal system
space group, Z P2(1)/n, 4
P1, 2
̅
P1, 12
̅
P1, 2
̅
a (Å)
7.5855(12)
21.675(3)
11.1031(16)
90.00
5.1663(6)
9.9211(10)
15.8829(17)
75.339(3)
82.551(3)
78.820(3)
769.84(14)
0.71073
11.2578(7)
19.6623(11)
19.8130(12)
93.236(3)
91.236(3)
90.298(3)
4377.6(5)
0.71073
5.6707(3)
10.8221(6)
17.0848(10)
104.161(2)
94.599(2)
104.840(2)
971.22(9)
0.71073
b (Å)
c (Å)
α (deg)
β (deg)
γ (deg)
volume (Å3)
96.050(9)
90.00
90.323(4)
90.00
105.384(4)
90.00
95.023(3)
90.00
1815.4(5)
0.71073
2056.5(3)
0.71073
1694.6(3)
0.71073
1924.02(18)
0.71073
X-ray
wavelength
μ (mm−1
)
0.097
0.108
none
0.093
none
0.101
none
0.111
0.096
none
0.094
none
absorption corr none
reflections
multiscan
collected
16 010
4889
15 920
5171
23 866
6752
29 113
6437
88 500
27578
14 241
>2σ(I)
22 510
6431
23 456
6179
independent
observed
2078
3934
4902
3940
4786
4312
threshold
expression
>2σ(I)
>2σ(I)
>2σ(I)
>2σ(I)
>2σ(I)
>2σ(I)
R1 (observed)
wR2 (all)
0.0558
0.0470
0.0452
0.0511
0.0646
0.0486
0.0488
0.1850
0.1416
0.1240
0.1626
0.1850
0.1509
0.1507
Δρ max/min
0.238/−0.185
0.341/−0.340
0.292/−0.271
0.325/−0.253
0.433/−0.386
0.400/−0.272
0.236/−0.214
Table 1, combine to give rise to a tetrameric 0-D ring, (Figure 2a).
The very same synthons and overall supramolecular motif were
also found in the crystal structure of 4N-AzA (Figure 2b). The
one outlier in this series of three is the structure of 4N-PA,
where the building blocks were assembled into an infinite 1-D
chain (Figure 3).
All four reactions between 4N and an even-chain diacid
produced crystals suitable for single-crystal structure analysis;
2583
dx.doi.org/10.1021/cg300180b | Cryst. Growth Des. 2012, 12, 2579−2587

Crystal Growth & Design
Article
Table 4. Crystallographic Parameters of 3N Co-Crystals
3N-AA
3N-SeA
3N-MA
3N-SA
3N-PA
3N-AzA
formula moiety
(C11H9N3O)
(C6H10O4)
(C11H9N3O)
(C10H18O4)
(C11H9N3O)
(C3H4O4)
(C11H9N3O)
(C4H6O4)
(C11H9N3O)
(C7H12O4)
(C11H9N3O)
(C9H16O4)
empirical formula C17H19N3O5
molecular weight 345.35
C21H27N3O5
401.46
C14H13N3O5
303.27
C15H15N3O5
317.30
C18H21N3O5
359.38
C20H25N3O5
387.43
color, habit
crystal system
space group, Z
a (Å)
yellow plate
triclinic
yellow prism
triclinic
gold plate
monoclinic
C2/c, 8
gold plate
triclinic
colorless prism
triclinic
colorless prism
monoclinic
P2(1)/n, 4
13.5588(13)
9.4660(10)
15.8384(15)
90.00
P1, 2
̅
P1, 2
̅
P1, 2
̅
P1, 4
̅
8.9361(8)
9.8031(9)
9.8270(9)
83.130(2)
74.507(3)
79.462(2)
813.31(13)
7.5703(8)
9.6882(10)
13.9977(14)
85.455(4)
81.210(4)
79.358(4)
995.74(18)
0.71073
31.2168(18)
4.7940(3)
22.7743(13)
90.00
4.8289(5)
11.1223(11)
14.3787(14)
67.856(4)
89.550(4)
87.878(4)
714.79(12)
0.71073
7.5711(5)
11.9820(8)
20.1758(13)
82.560(4)
80.081(4)
85.195(4)
1784.2(2)
0.71073
b (Å)
c (Å)
α (deg)
β (deg)
128.375(2)
90.00
105.946(4)
90.00
γ (deg)
volume (Å3)
2671.9(3)
0.71073
1954.6(3)
0.71073
X-ray wavelength 0.71073
μ (mm−1
)
0.105
none
0.096
0.117
0.113
0.099
0.096
absorption corr
reflections
collected
none
none
none
multiscan
multiscan
21 410
5266
13 375
5863
16 865
4729
10 209
4585
42 683
11 575
6192
15 672
6247
independent
observed
4361
4414
3553
3013
3981
threshold
expression
>2σ(I)
>2σ(I)
>2σ(I)
>2σ(I)
>2σ(I)
>2σ(I)
R1 (observed)
wR2 (all)
0.0465
0.0519
0.0433
0.0576
0.0622
0.0604
0.1432
0.1508
0.1241
0.1907
0.1777
0.1813
Δρ max/min
0.551/−0.240
0.436/−0.297
0.463/−0.246
0.360/−0.381
0.322/−0.316
0.316/−0.321
Table 5. IR Carbonyl CO Stretch and O−H···N Bands for
Three of the four reactions between an odd-chain diacid and
3N produced crystals suitable for single-crystal structure
analysis; 3N-MA, 3N-PA, and 3N-AzA. All three of them
contained infinite 1-D chains constructed by a combination of
alternating −COOH···N(py) and −COOH···N(py)-NH2
hydrogen-bond interactions, (Figure 5a−c).
Finally, we obtained structural data for three of the four
products obtained when 3N was allowed to react with even-
chain acids; 3N-SA, 3N-AA, and 3N-SeA (Figure 6). Again, the
−COOH···N(py) and −COOH···N(py)-NH2 synthons were
the primary intermolecular interactions, Table 2. All three
structures contained tetrameric 0-D rings, (Figure 6a−c).
Co-Crystals of 3N and 4N
compd
O−H···N stretches (cm−1
)
CO stretch (cm−1
)
4N-GA
4N-PA
4N-AzA
4N-SA
2492; 1862
2357; 1883
2488; 1875
2481; 1890
2492; 1862
2471; 1907
2483; 1877
2439; 1871
2487; 1877
2495; 1887
2406; 1887
2430; 1895
2509; 1886
1687
1687
1683
1685
1687
1683
1684
1701
1686
1684
1676
1695
1683
4N-AA
4N-SuA
4N-SeA
3N-MA
3N-PA
3N-AzA
3N-SA
DISCUSSION
■
3N-AA
3N-SeA
Combining 3N and 4N in stoichiometric amounts with a series
of aliphatic diacids can in principle offer three different
outcomes, i.e., a co-crystal, a salt, or no reaction at all. Through
a systematic IR screen, it is possible to anticipate the outcome
before any single crystal data becomes available. For instance,
an IR spectrum for a co-crystal shows two very prominent
broad O−H···N stretches around the 1850−2500 cm−1 region
(Figures 1b and 1d; Table 5), and the observed carbonyl CO
stretch for the acid is always above 1675 cm−1 (Table 5).
(4N-SA), (4N-AA), (4N-SuA) and (4N-SeA). As was the case
with 4N and odd-chain acids, the main intermolecular
interactions were −COOH···N(py) and −COOH···N(py)-
NH2 synthons, Table 1. The combination of these syn-
thetic vectors resulted, in each case, in infinite 1-D chains,
Figure 4a−d.
Figure 2. (a) Tetrameric ring in the crystal structure of 4N-GA. (b) Tetrameric ring in the crystal structure of 4N-AzA.
2584
dx.doi.org/10.1021/cg300180b | Cryst. Growth Des. 2012, 12, 2579−2587

Crystal Growth & Design
Article
as the primary building block. Two distinctly different types of
assemblies i.e., chains (0-D) and rings (1-D) were obtained
depending upon the choice of acid−base pair in the co-
crystallization experiments.
The following combinations, 4N-odd acid (pimelic, glutaric,
and azelaic acid) and 3N-even acid (succinic, adipic, and
sebacic acid), produce 0-D ring architectures in five out of six
structures. The only exception in this series was 4N with
pimelic acid, which gave a chain instead of the expected/
intended ring. This preference for a cyclic assembly in five of six
structures represents a clear pattern of behavior, especially since
we are dealing with molecules with considerable molecular
flexibility coupled with the fact that the synthesis is driven by
relatively weak and reversible interactions. There is no obvious
reason why 4N-PA contains infinite chains, but it is possible
that a ring-containing polymorph could appear if this
compound were subjected to a rigorous polymorph screen.
In the combinations of 4N-even acid (succinic, adipic,
suberic, and sebacic acid) and 3N-odd acid (malonic, pimelic,
and azelaic acid), we found 1-D chain architectures in all seven
of the crystal structures that we were able to obtain. All the
chains are assembled through a combination of −COOH···N-
(py) and −COOH···N(py)-NH2 hydrogen bonds, and the two
components alternate throughout each chain. It is interesting to
note that in 12 out of 13 crystal structures, the carbonyl group
and pyridine N moiety are arranged in an anti-configuration
with respect to each other with an exception in the structure of
Figure 3. One-dimensional hydrogen-bonded chain in the crystal
structure of 4N-PA.
On the basis of the IR data alone, we could establish that all 16
experiments yielded co-crystals. This assessment was also supported
by the result that we obtained from crystallographic data for 13 of
these 16 compounds; all were found to be co-crystals. Thus, the IR
data is very consistent with the single crystal analysis and can be
considered as a reliable tool to detect co-crystal formation.
All 13 crystal structures are assembled through acid−
COOH···N(py) and −COOH···N(py)-NH2 heterosynthons
Figure 4. Section of the infinite 1-D chain in the crystal structure of (a) 4N-SA; (b) 4N-AA; (c) 4N-SuA; and (d) 4N-SeA.
Figure 5. View of the infinite 1-D chain in the crystal structure of (a) 3N-MA; (b) 3N-PA; and (c) 3N-AzA.
2585
dx.doi.org/10.1021/cg300180b | Cryst. Growth Des. 2012, 12, 2579−2587

Crystal Growth & Design
Article
Figure 6. Tetrameric hydrogen bonded rings in the crystal structures of (a) 3N-SA; (b) 3N-AA; and (c) 3N-SeA.
Table 6. Result of Co-Crystallization Experiments of 4N and
3N with Several Even and Odd Chain Diacids
the molecule. In the case of 4,4-bipyridine, the two pyridyl
binding sites are linear with respect to each other, whereas in
the case of 4N and 3N, the pyridyl nitrogen and amino pyridine
binding sites are at an angle (Scheme 4). This difference in the
geometry of the binding sites controls the overall outcome of
the assembly process, which is evident from our results. The
connections between structural packing and solid-state derived
properties are well-known, but it is interesting to note that such
fundamental differences extend from homomeric solids19 to
heteromeric co-crystals. Furthermore, the relative orientations
of the synthetically important binding sites are clearly dictated
by inherent molecular geometries, but these differences can be
taken advantage of in a constructive sense when bringing
discrete molecular building blocks together into extended solid-
state architectures.
ligand
diacids
chains
rings
4-N
3-N
4-N
3-N
odd-number
even-number
even-number
odd-number
1/3
0
2/3
3/3
0
4/4
3/3
0
3N-malonic acid co-crystal (Figure 5a) where they are syn with
respect to each other; however, this discrepancy does not affect
the overall assembly, and chains do appear in this structure as
expected. A summary of the results is shown in Table 6.
On the basis of the success of our results, we can conclude
that the number of carbon atoms within the acid, combined
with the precise orientation of the binding sites in the co-
crystallizing agent, does indeed control the resulting supra-
molecular assembly in a very reliable manner. Despite the fact
that any number of motifs and assemblies could in principle be
formed when 3N and 4N are brought together with aliphatic
diacids, there is a remarkable consistency to the way in which
these flexible molecules recognize, bind, and organize
themselves in the solid state. Only two different primary
motifs can be identified, rings and chains, and to a very large
extent (twelve of thirteen cases), the appearance of one or the
other architecture can be rationalized a priori. Such correlations
are not always going to be found; for example, attempts to find
links between the appearance of tapes or rings and the number
of atoms in the participating carbon chains have been made
using 4,4′-bipyridyl and carboxylic acids as a test-bed. However,
in that case, it was not possible to deduce any obvious
connection between the molecular and the supramolecular.18 It
was finally concluded that tape-like structures were preferably
formed irrespective of the odd−even nature of the acid. The
notable difference between the bipy-system and our 3N/4N
compounds is in the relative orientation of the binding sites on
ASSOCIATED CONTENT
* Supporting Information
CIF files of the crystal structures. This material is available free
of charge via the Internet at http://pubs.acs.org.
■
S
AUTHOR INFORMATION
Corresponding Author
*E-mail: aakeroy@ksu.edu.
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
We are grateful for financial support from NSF (CHE-
0957607).
■
REFERENCES
■
586.
(1) Aakeroy, C. B. Acta Crystallogr., Sect. B: Struct. Sci. 1997, 53, 569−
̈
(2) (a) Braga, D. Chem. Commun. 2003, 2751−2754. (b) Aakeroy,
̈
C. B.; Champness, N.; Janiak, C. CrystEngComm 2010, 12, 22−43.
(3) (a) Aakeroy, C. B.; Desper, J.; Fasulo, M.; Hussain, I.; Brock, L.;
̈
Scheme 4. From Discrete Building Blocks to Rings and Chains.
Schultheiss, N. CrystEngComm 2008, 10, 1816−1821. (b) Aakeroy,
̈
C. B.; Chopade, P.; Ganser, C.; Desper, J. Chem. Commun. 2011, 47,
4688−4690.
(4) (a) Desiraju, G. R. Crystal Engineering: The Design of Organic
Solids; Elsevier: Amsterdam, The Netherlands, 1989. (b) Lehn, J. M.
Supramolecular Chemistry; VCH: Weinheim, Germany, 1995.
(c) Desiraju, G. R Perspectives in Supramolecular Chemistry; Wiley:
Chichester, U.K., 1995; Vol. 2.
2586
dx.doi.org/10.1021/cg300180b | Cryst. Growth Des. 2012, 12, 2579−2587

Crystal Growth & Design
Article
(5) (a) Aakeroy, C. B.; Salmon, D. J. CrystEngComm 2005, 7, 439.
̈
(b) MacGillivray, L. R. CrystEngComm 2004, 6, 77. (c) Lehn, J.-M.
Science 2002, 295, 2400. (d) Desiraju, G. R. Acc. Chem. Res. 2002, 35,
565. (e) Moulton, B.; Zaworotko, M. J. Chem. Rev. 2001, 101, 1629.
(f) Desiraju, G. R. Angew. Chem., Int. Ed. 1995, 34, 2311. (g) Wenger,
M.; Bernstein, J. Angew. Chem., Int. Ed. 2006, 45, 7966. (h) Childs,
S. L.; Hardcastle, K. I. CrystEngComm 2007, 9, 364. (i) Bosch,
E. CrystEngComm 2007, 9, 191. (j) Aakeroy, C. B.; Desper, J.; Salmon,
̈
D. J.; Smith, M. M. Cryst. Growth Des. 2006, 6, 1033−1042. (k) Aakeroy,
̈
C. B.; Desper, J.; Urbina, J. F. CrystEngComm 2005, 7, 193.
(6) (a) Braga, D.; Desiraju, G. R.; Miller, J. S.; Orpen, G. A.
CrystEngComm. 2002, 4, 500−509.
(7) (a) Thalladi, V. R.; Nusse, M.; Boese, R. J. Am. Chem. Soc. 2000,
̈
122, 9227−9236. (b) Vishweshwar, P.; Nangia, A.; Lynch, V. M. Cryst.
Growth Des. 2003, 3, 783−790. (c) Morrison, J. D.; Robertson, J. M.
J. Chem. Soc. 1949, 2, 980. (d) Verweel, H. J.; MacGillavry, C. H.
Nature 1938, 142, 161.
(8) (a) Baeyer, A. Ber. Chem. Ges. 1877, 10, 1286. (b) Thalladi, V. R.;
Nusse, M. J. Am. Chem. Soc. 2000, 122, 9227−9236.
(9) Desiraju, G. R. CrystEngComm 2003, 5, 466−467.
(10) (a) Braga, D.; Dichiarante, E.; Palladino, G. CrystEngComm
2010, 12, 3534−3536. (b) Bond, A. D. CrystEngComm 2006, 8, 333.
(c) Callear, S. K.; Hursthouse, M. B.; Threlfall, T. L. CrystEngComm
2010, 12, 898−908.
(11) (a) Mei, X.; Wolf, C. Eur. J. Org. Chem. 2004, 4340−4347.
(b) Dale, S. H.; Elsegood, M. R. J.; Hemmings, M.; Wilkinson, A. L.
CrystEngComm 2004, 6, 207. (c) Bhogala, B. R.; Nangia, A. Cryst.
Growth Des. 2003, 3, 547. (d) Bensemann, I.; Gdaniec, M.;
Lakomecka, K.; Milewska, M. J.; Polonski, T. Org. Biomol. Chem.
2003, 1, 1425. (e) Hguyen, T. L.; Fowler, F. W.; Lauher, J. W. J. Am.
Chem. Soc. 2001, 123, 11057.
(12) (a) Aakeroy, C. B.; Beatty, A. M.; Helfrich, B. A. J. Am. Chem.
̈
Soc. 2002, 124, 14425. (b) Vishweshwar, P.; Nangia, A.; Lynch, V. M.
Cryst. Growth Des. 2003, 3, 783.
(13) (a) Vishweshwar, P.; Nangia, A.; Lynch, V. M. J. Org. Chem.
2002, 67, 556. (b) Leiserowitz, L.; Nader, F. Acta Crystallogr., Sect. B:
Struct. Sci. 1977, 33, 2719. (c) Huang, C. M.; Leiserowitz, L.; Schmidt,
G. M. J. J. Chem. Soc., Perkin Trans. 2 1973, 5, 503.
(14) Kim, T. H.; Seo., J. Chem. Mater. 2007, 19, 5815−5817.
(15) Krishna, K.; Jose, D. A.; Dastidar, P.; Das, A. Langmuir 2004, 20,
10413−10418.
(16) Kushner, S.; Dalalian, H.; Robert, T. C. J. Org. Chem. 1948, 13,
834−836.
(17) (a) APEX2, v2.2.0; Bruker AXS: Madison, WI, 2007. (b)
SAINT, v7.46a; Bruker AXS: Madison, WI, 2007. (c) SHELXTL,
v6.10; Bruker AXS: Madison, WI, 2001.
(18) Pedireddi, V. R.; Chatterjee, S.; Ranganathan, A.; Rao, C. N.
Tetrahedron 1998, 54, 9457−9474.
(19) Sarma, J. A. R. P; Nangia, A.; Desiraju, G. R.; Zass, E.; Dunitz, J.
Nature 1996, 384, 320−320.
2587
dx.doi.org/10.1021/cg300180b | Cryst. Growth Des. 2012, 12, 2579−2587