Design of co-crystals/salts of aminopyrimidines and carboxylic acids through recurrently occurring synthons

Article abstract of DOI:10.1021/cg3005954

Fourteen crystals involving various substituted organic acid molecules and 2-amino-4,6-dimethylpyrimidine (MP)/2-amino-4,6-dimethoxypyrimidine (MOP) were prepared and characterized by using single crystal X-ray diffraction. Among the 14 crystals, proton transfer from the carboxylic acid to pyrimidine base has occurred only in 5 crystals while the remaining were co-crystals. It is obvious that the R₂²(8) ring motif occurs in co-crystals/salts when an acid interacts with a pyrimidine base. The present work has been focused on this aspect, confining the study to MP and MOP, where the results surely indicate that there are three main synthons, which regularly occur, linear heterotetramer (LHT), cyclic heterotetramer (CHT), and heterotrimer (HT). Among the whole lot, in both the present study and the literature, LHT is dominant as it occurs in 30 out of the existing 54 structures, while CHT and HT correspond to 9 each with the remaining structures forming newer synthons. Our results are also consistent with a recent paper in Crystal Growth and Design (2008, 8, 4031-4044), where LHT is predicted to be more stable among other synthons. Furthermore in the present study, attempts have been made to correlate the formation of salts/co-crystals using ΔpK_a for both MP and MOP molecular complexes. The failure to explain MOP using ΔpK_a also suggests that the use of Pk_a values in predicting co-crystals/salts could vary from system to system.

Full text of DOI:10.1021/cg3005954

Article
pubs.acs.org/crystal
Design of Co-crystals/Salts of Aminopyrimidines and Carboxylic
Acids through Recurrently Occurring Synthons
Samuel Ebenezer and P. Thomas Muthiah*
School of Chemistry, Bharathidasan University, Tiruchirappalli-620 024, India
S
* Supporting Information
ABSTRACT: Fourteen crystals involving various substituted
organic acid molecules and 2-amino-4,6-dimethylpyrimidine
(MP)/2-amino-4,6-dimethoxypyrimidine (MOP) were pre-
pared and characterized by using single crystal X-ray
diffraction. Among the 14 crystals, proton transfer from the
carboxylic acid to pyrimidine base has occurred only in 5
crystals while the remaining were co-crystals. It is obvious that
2
the R2 (8) ring motif occurs in co-crystals/salts when an acid
interacts with a pyrimidine base. The present work has been
focused on this aspect, confining the study to MP and MOP,
where the results surely indicate that there are three main
synthons, which regularly occur, linear heterotetramer (LHT),
cyclic heterotetramer (CHT), and heterotrimer (HT). Among
the whole lot, in both the present study and the literature,
LHT is dominant as it occurs in 30 out of the existing 54
structures, while CHT and HT correspond to 9 each with the
remaining structures forming newer synthons. Our results are
also consistent with a recent paper in Crystal Growth and
Design (2008, 8, 4031−4044), where LHT is predicted to be more stable among other synthons. Furthermore in the present
study, attempts have been made to correlate the formation of salts/co-crystals using ΔpKa for both MP and MOP molecular
complexes. The failure to explain MOP using ΔpKa also suggests that the use of Pka values in predicting co-crystals/salts could
vary from system to system.
among the whole lot, hydrogen bonding has been more
prevalent because of its strength, directionality, and
predictability.9−13 Homosynthons and heterosynthons are
the two distinctively categorized types of synthons.14,15 The
preference in the formation of supramolecular heterosynthons
over homosynthons is evident from the large number of
reports in the literature.16−21 Moreover in the context of the
present study it is very clear from the systematic analysis of the
bimolecular ring motifs in organic crystal structures that the
occurrence of supramolecular heterosynthons (2-substituted
nitrogen heterocyclic system with carboylic acid group) are
statistically high when compared to the individual homo-
synthons.22,23 More specifically in the present study we focus
herein upon the ability of carboxylic acids to form other
reliable and stable supramolecular heterosynthons with the
pyrimidine derivatives.
INTRODUCTION
■
Numerous organic molecular motifs form the essential structural
components of biological systems and drugs. The three-
dimensional structures and interaction of bioorganic motifs
are of immense importance in the existence of biological systems
and in molecular recognition. Molecular recognition refers to
the selective binding of two or more molecules through
noncovalent interactions such as hydrogen bonding, hydro-
phobic forces, van der Waals forces, π−π interactions and
electrostatic forces.1 They play a very important role in biological
systems as demonstrated in antigen−antibody, DNA−protein,
sugar−lectin, and many other interactions.2−6 They are also vital
parts of many biochemical processes such as enzyme action,
molecular transport, genetic information transfer and protein
assembly.7,8
From the perspective of crystal engineering, a detailed
understanding of the supramolecular chemistry of functional
groups in a compound is prerequisite for rationally designing
salts and co-crystals with desired structural features. The key to
this has been the identification of the recurrently occurring
supramolecular synthons. Use of synthetic strategies such as
hydrogen bonding, halogen bonding, and stacking interac-
tions have been very common in building these synthons and
Pyrimidines and aminopyrimidine derivatives are bio-
logically important compounds and they manifest themselves
in nature as components of nucleic acids. The functions of
nucleic acids are explicitly determined by hydrogen bonding
Received: May 2, 2012
Published: May 20, 2012
© 2012 American Chemical Society
3766
dx.doi.org/10.1021/cg3005954 | Cryst. Growth Des. 2012, 12, 3766−3785

Crystal Growth & Design
Article
Scheme 1. Molecular Components of Different Co-crystals and Salts
of 2-chlorobenzoic acid/3-methylbenzoic acid/cinnamic acid/
4-nitrobenzoic acid/4-methoxybenzoic acid/isophthalic acid/chloro-
acetic acid/trichloroacetic acid/5-chloro salicylic acid/3-hydroxy
picolinic acid in 1:1 molar ratio and were allowed to warm over a
water bath for half an hour. The mixtures were cooled slowly and kept
at room temperature. After a few days, colorless prismatic crystals
(for compounds 1, 3, 6, 9, and 10), colorless plate-like crystals (for
compounds 2 and 5) and colorless needle-like crystals (for compounds
4, 7, and 8) separated out of the mother liquor.
Compounds 11, 12, and 14 were prepared by mixing hot methanolic
solution of 2-amino-4,6-dimethylpyrimidine with hot methanolic
solution of 5-chloro salicylic acid/malonic acid/adipic acid in 1:1
molar ratio and warmed in a water bath for few minutes. The mixtures
were then allowed to cool slowly at room temperature. After a few days
colorless needle-like crystals (for compound 11, 12, and 14) separated
out of the mother liquor. Similarly compound 13 was prepared by
mixing ethanolic solution of 2-amino-4,6-dimethylpyrimidine with hot
ethanolic solution of glutaric acid in 1:1 molar ratio. It was allowed to
warm over a water bath for half an hour and then left for slow
evaporation at room temperature. After a few days, colorless plate-like
crystals separated out of the mother liquor.
patterns including base pairing which is responsible for
genetic information transfer.8,24 Aminopyrimidine derivatives
also present themselves as the key components in many
drugs.25 Their interactions with carboxylic acids are of utmost
importance since they are involved in protein−nucleic acid
recognition and drug−protein recognition processes (where
the pyrimidine moiety of a drug forms hydrogen bonding with
the carboxyl group of the protein).26−28 It is for this reason
that the present work has been mainly focused on salts and co-
crystals of aminopyrimidine derivatives (2-amino-4,6-dimeth-
yl pyrimidine (MP) and 2-amino-4,6-dimethoxy pyrimidine
(MOP)) . Moreover in the recent years there has been massive
interest in the co-crystals/salts involving Active Pharmaceut-
ical Ingredients because of their applications in drug
formulation.29−32 Therefore to gain deep knowledge and
understanding of a detailed supramolecular organization in
the salts/co-crystals, carboxylic acids with different mod-
ifications have been tried and attempts have been made to
identify their effect on the hydrogen-bonded supramolecular
motifs. Scheme 1 shows the molecular structure of various
carboxylic acids along with the pyrimidine derivatives which
give rise to different salts and co-crystals. A large number of
MP/MOP-carboxylate/carboxylic acid complexes has earlier
been reported from our laboratory.33−46
2-amino-4,6-dimethoxypyrimidine and 2-amino-4,6-dimethylpyrimi-
dine were purchased from Sigma-Aldrich, Inc. Chloroacetic acid,
cinnamic acid and glutaric acid were purchased from S.d. Fine chemicals
Pvt. Ltd., India, whereas the rest of the acids were purchased from Loba
chemie Pvt. Ltd., India.
X-RAY CRYSTALLOGRAPHY
■
EXPERIMENTAL SECTION
Single crystal diffraction data for all crystal structures were
collected on Bruker SMART APEX-II diffractometer equipped
by a CCD-detector with graphite monochromated Mo Kα
■
Compounds 1−10 were prepared by mixing hot methanolic solution
of 2-amino-4,6-dimethoxypyrimidine with hot methanolic solution
3767
dx.doi.org/10.1021/cg3005954 | Cryst. Growth Des. 2012, 12, 3766−3785

Crystal Growth & Design
Article
3768
dx.doi.org/10.1021/cg3005954 | Cryst. Growth Des. 2012, 12, 3766−3785

Crystal Growth & Design
Article
radiation (λ-=0.71073 Å). Data reduction and cell refinement
were carried out using Bruker SAINT and the necessary
absorption corrections were performed by multiscan method
using SADABS.47 In all cases, the structures were solved in the
WinGX suite of programs by direct methods using SHELXL-86/
SHELXL-97 and refined using full-matrix least-squares techni-
ques on F2 using SHELXS-97.48 All non-hydrogen atoms were
refined anisotropically and thereafter, all hydrogen atoms were
placed in their geometrically idealized positions and constrained
to ride on their parent atoms. Diagrams and publication material
were generated using PLATON49 and MERCURY.50
RESULTS
■
Crystallographic data for all the crystals are summarized in Table 1
and the hydrogen bonding parameters listed in Table 2. A detailed
structural description of all the co-crystals and salts are given below
in succession.
2-Amino-4,6-dimethoxypyrimidine-2-chlorobenzoic
acid (1). The crystal structure of 1 crystallizes in triclinic space
group, P1, with a molecule of 2CBA and a molecule of MOP in
̅
2
the asymmetric unit. The main motif (R2 (8)) is assembled via a
complementary hydrogen-bond interaction between the carbox-
ylic acid and the amino-pyrimidine moiety to form a dimeric unit.
Adjacent dimeric units are further connected through self-
complementary secondary N−H···N hydrogen bonds to form a
four-component supermolecule. Thus the primary and secon-
dary hydrogen bonds, O−H···N, N−H···O, and N−H···N
combine to form a linear heterotetramer motif. The supra-
molecular tetramer in 1 is planar and the adjacent tetramer
supermolecules, produced from the weak hydrogen bonds, are
connected into infinite 1-D tape via two symmetry related
hydrogen bonds involving H8B of the methoxy group and O8 of
another methoxy oxygen with both neighboring pyrimidine rings
2
forming a ring motif through graph set notation R2 (6) as shown
in Figure 1 (symmetry codes −x, 1 − y, 1 − z). These tapes are
interconnected to neighboring tapes through weak hydrogen
bonds involving C12 of an acid ring and O7 of another methoxy
group attached to the pyrimidine ring to form a supramolecular
sheet (Figure 2) along (122) plane (symmetry codes x, 1 + y,
−1 + z). Each layer of supramoleuclar sheets are stacked one with
another through Π−Π stacking interactions. Stacking between
the pyrimidine−pyrimidine and acid−acid rings are involved
with centroid-to-centroid (Cg−Cg) distance of 3.5119(9) and
3.9200(12) Å and a slip angle (the angle between the centroid
vector and the normal to the plane) of 11.79° and 23.68°,
respectively.
2-Amino-4,6-dimethoxypyrimidine-3-methylbenzoic
acid (2). 2 crystallizes in triclinic P1 space group with the
̅
asymmetric unit possessing a molecule of MOP and 3MBA. The
2
primary R2 (8) motif connects the acid and amino-pyrimidine
moiety to form a dimeric unit. A similar four component
supermolecule is formed as in the previous case, through primary
and secondary hydrogen bonds (O−H···N, N−H···O, and
N−H···N) involving the amino and hydroxyl protons of
carboxylic acid as donors and the two pyrimidine nitrogens
including the oxygen of the carboxyl group as acceptors. Adjacent
tetramers are connected via C−H···O hydrogen bond, involving
the C−H and the oxygen of methoxy group of pyrimidine ring
on either side (C8−H8C···O8). As a result adjacent tetramers
are assembled into infinite 1-D tapes (Figure 3). However this
time, in contrast to the previous structure, each tape is not
interconnected with the neighboring ones through C−H···O
bonds since they lie far away from each other. They also form a
3769
dx.doi.org/10.1021/cg3005954 | Cryst. Growth Des. 2012, 12, 3766−3785

Crystal Growth & Design
Article
Table 2. Hydrogen Bond Metrics for Compounds 1−14
́
́
D- - -H···A
H···A (Å)
D···A (Å)
∠DH···A
symmetry operation
x, y, z
MOP·2CBA (1:1)
2.699(2)
O(1)- - -H(1)···N(1)
N(2)- - -H(2A)···N(3)
N(2)- - -H(2B)···O(2)
C(8)- - -H(8B)···O(8)
C(12)- - -H(12)···O(7)
1.89
2.31
1.97
2.59
2.52
171
166
173
152
161
3.156(2)
2 − x, −y, 1 − z
x, y, z
2.829(2)
3.473(2)
−x, 1 − y, 1 − z
x, 1 + y, −1 + z
3.415(2)
MOP·3MBA (1:1)
2.6817(19)
3.161(2)
O(1)- - -H(1)···N(1)
N(2)- - -H(2A)···N(3)
N(2)- - -H(2B)···O(2)
C(8)- - -H(8C)···O(8)
1.87
2.31
2.06
2.58
168
169
173
151
x, y, z
2 − x, −y, 1 − z
x, y, z
2.912(2)
3.458(2)
−x, 1 − y, 1 − z
MOP·CA (1:1)
2.712(2)
O(1)- - -H(1)···N(1)
N(2)- - -H(2A)···O(2)
N(2)- - -H(2B)···N(3)
C(14)- - -H(14)···O(1)
1.90
2.02
2.28
2.68
169
174
169
159
x, y, z
2.878(2)
x, y, z
3.130(2)
−x, −y, −z
2 − x, 1 − y, 1 − z
3.565(2)
MOP·4NBA (1:1)
2.643(2)
O(1)- - -H(1)···N(1)
N(2)- - -H(2A)···O(2)
N(2)- - -H(2B)···N(3)
1.83
2.01
2.29
170
173
157
x, y, z
2.869(2)
x, y, z
3.098(2)
2 − x,1 − y, 1 − z
MOP·4MOBA (1:1)
2.7270(15)
2.9142(17)
2.8479(18)
MOP·IPA (2:1)
2.7807(16)
2.8267(19)
2.6606(18)
2.851(2)
O(1)- - -H(1)···N(1)
N(2)- - -H(2A)···O(2)
N(2)- - -H(2B)···O(2)
1.91
2.11
2.00
172
155
170
x, y, z
1 − x, 1 − y, 1 − z
x, y, z
O(1)- - -H(1)···N(1A)
1.97
1.99
1.85
2.00
2.21
2.55
172
166
170
168
149
153
x, y, z
N(2A)- - -H(2A1)···O(2)
O(3)- - -H(3A)···N(1B)
N(2B)- - -H(2B1)···O(4)
N(2B)- - -H(2B2)···O(7A)
C(7B)- - -H(7B3)···O(2)
x, y, z
x, y, z
x, y, z
2.976(2)
2 − x,1 − y, −z
x, 1 + y, 1 + z
3.433(2)
MOP·CAA (1:1)
2.6351(18)
2.8495(18)
3.187(2)
MOP+·TCAA− (1:1)
2.734(2)
O(1)- - -H(1)···N(1)
N(2)- - -H(2A)···O(2)
N(2)- - -H(2B)···N(3)
1.82
1.99
2.34
170
173
167
x, y, z
x, y, z
1 − x, 2 − y, 1 − z
N(1)- - -H(1)···O(1)
N(2)- - -H(2A)···O(2)
N(2)- - -H(2B)···O(2)
1.89
2.16
1.93
165
137
175
x, y, z
2.845(2)
1 − x, 1 − y, 1 − z
x, y, z
2.786(2)
MO+P·5CSA− (1:1)
2.6582(15)
3.1627(18)
2.7742(18)
2.5273(18)
3.3311(18)
MOP+.3HPA‑ (1:1)
2.9571(17)
2.7146(14)
2.7870(16)
2.5280(17)
2.6934(14)
2.5214(17)
3.0491(19)
2.7885(16)
3.5781(15)
MP+·5CSA− (1:1)
2.799(6)
N(1)- - -H(1)···O(1)
N(2)- - -H(2A)···N(3)
N(2)- - -H(2B)···O(2)
O(3)- - -H(3)···O(2)
C(7)- - -H(7B)···O(1)
1.80
2.32
1.94
1.81
2.42
178
166
165
145
160
x, y, z
x, 3/2 − y, 1/2 + z
x, y, z
x, y, z
1 − x, 1 − y, 1 − z
N(2A)- - -H(2A1)···N(4A)
N(1A)- - -H(1A)···O(1A)
N(2A)- - -H(2A2)···O(2A)
O(3A)- - -H(3A)···O(1A)
N(1B)- - -H(1B)···O(1B)
O(3B)- - -H(3B)···O(1B)
N(2B)- - -H(2B1)···N(4B)
N(2B)- - -H(2B2)···O(2B)
C(8B)- - -H(8B3)···O(8B)
2.10
1.86
1.94
1.80
1.83
1.79
2.20
1.94
2.66
172
176
168
147
178
148
172
169
161
1 − x, 1 − y, −z
x, y, z
x, y, z
x, y, z
x, y, z
x, y, z
2 − x, 1 − y, 1 − z
x, y, z
1 − x, −y, 1 − z
N(2A)- - -H(2A1)···O(2B)
N(2B)- - -H(2B1)···O(2A)
N(2A)- - -H(2A2)···O(2A)
N(2B)- - -H(2B2)···O(2B)
N(1A)- - -H(1A)···O(1A)
1.99
2.02
1.95
1.93
1.82
158
156
167
176
175
x, y, z
x, y, z
x, y, z
x, y, z
x, y, z
2.833(5)
2.790(6)
2.792(5)
2.678(6)
3770
dx.doi.org/10.1021/cg3005954 | Cryst. Growth Des. 2012, 12, 3766−3785

Crystal Growth & Design
Table 2. continued
D- - -H···A
Article
́
́
H···A (Å)
D···A (Å)
∠DH···A
symmetry operation
x, y, z
MP+·5CSA− (1:1)
N(1B)- - -H(1B···O(1B)
O(3A)- - -H(3A)···O(1A)
O(3B)- - -H(3B)···O(1B)
C(5A)- - -H(5A)···O(3B)
C(5B)- - -H(5B)···O(3A)
1.87
1.80
1.79
2.53
2.56
2.728(5)
175
147
147
151
150
2.524(5)
x, y, z
x, y, z
2.523(5)
3.375(6)
1 − x, 1/2 + y, 3/2 − z
−x, −1/2 + y, 1/2 − z
3.401(6)
MP+·MA− (1:1)
2.652(2)
N(1)- - -H(1)···O(1)
N(2)- - -H(2A)···N(3)
N(2)- - -H(2B)···O(2)
O(3)- - -H(3)···O(2)
C(5)- - -H(5)···O(3)
1.79
2.14
1.92
1.72
2.61
175
179
170
155
139
x, y, z
2.998(2)
1 − x, 1 − y, −z
x, y, z
2.775(2)
2.482(2)
x, y, z
3.401(3)
2 + x, −1 + y, z
MP·GA (2:1)
2.612(7)
O(1)- - -H(1)···N(1A)
1.82
2.19
2.10
1.82
2.11
2.21
161
172
169
160
168
168
x, y, z
N(2A)- - -H(2A1)···N(3B)
N(2A)- - -H(2A2)···O(2)
O(3)- - -H(3)···N(1B)
N(2B)- - -H(2B1)···O(4)
N(2B)- - -H(2B2)···N(3A)
3.047(7)
1/2 + x, −y, z
x, y, z
2.948(8)
2.606(7)
x, y, z
2.960(8)
x, y, z
3.053(8)
−1/2 + x, −y, z
MP·AA (2:1)
2.649(3)
O(1)- - -H(1)···N(1A)
1.85
2.14
2.19
1.86
2.18
2.13
165
170
171
165
177
173
x, y, z
N(2A)- - -H(2A1)···O(2)
N(2A)- - -H(2A2)···N(3B)
O(3)- - -H(3)···N(1B)
N(2B)- - -H(2B1)···N(3A)
N(2B)- - -H(2B2)···O(4)
2.993(3)
x, y, z
3.045(3)
1 + x, y, −1 + z
x, y, z
2.659(3)
3.043(3)
−1 + x, y, 1 + z
x, y, z
2.980(3)
2
Figure 1. View of infinite tape connected by R2 (6) synthon in 1.
planar sheet (Figure 4) along the (122) plane which stacks one
over the other through Π−Π interactions. The stacking
interactions are similar to the previous structure which occur
between two pyrimidine and two acid moieties with Cg−Cg
distance of 3.4113(9) and 3.9030(12) Å and a slip angle of 7.98°
and 15.36°, respectively.
Stacking interactions between these parallel sheets of molecules
are far from the expected range.
2-Amino-4,6-dimethoxypyrimidine-4-nitrobenzoic
Acid (4). MOP.4NBA crystallizes in monoclinic P21/c space
group with a molecule of MOP and 4NBA in the asymmetric
unit. The primary supermolecule in the crystal structure are
connected through the acid and amino-pyrimidine moieties
engaging in O−H···N and N−H···O hydrogen bonds. The
secondary synthons involving the N−H···N hydrogen bonds are
responsible for the formation of the linear heterotetramer. All the
heterotetramers do not lie in the same plane unlike the previous
structures rather they are in two different planes. Inspite of the
presence of the two strong acceptor atoms (oxygen atoms of
the nitro group in the aromatic acid), no extra hydrogen bonds
are found within the limits. Moreover no stacking interactions are
observed in this structure.
2-Amino-4,6-dimethoxypyrimidine-cinnamic Acid (3).
3 also crystallizes in triclinic P1 space group. The asymmetric unit
̅
consists of a molecule of CA and MOP. A similar four
component supermolecule (linear heterotetramer) is formed
which is essentially the same like the previous structures. The
adjacent tetrameric units are held together through inversely
related C14−H14···O1 hydrogen bonds, involving C−H of the
acid ring and hydroxyl oxygen of carboxylic group in the acid
molecule, to form an infinite 1D tape (Figure 5). These parallel
tapes lie alongside each other in the (113) plane (Figure 6).
̅
3771
dx.doi.org/10.1021/cg3005954 | Cryst. Growth Des. 2012, 12, 3766−3785

Crystal Growth & Design
Article
Figure 2. Supramolecular sheet formed by interconnection of tapes along (122) plane in 1.
2
Figure 3. Extension of infinite tape connected by R2 (6) synthon in 2.
2-Amino-4,6-dimethoxypyrimidine-4-methoxyben-
zoic Acid (5). MOP·4MOBA crystallizes in the P21/c
monoclinic space group where the asymmetric unit consists of
single molecule of 4MOBA acid and MOP. A couple of N−H···O
and O−H···N hydrogen bonds is responsible for the formation of
a dimer which acts as the primary hydrogen bonds. The two acid-
aminopyrimidine dimers are again interlinked together with
N−H···O bonds (secondary hydrogen bonds) between the free
amino H-atom of one of the dimer and the carbonyl oxygen of
the acid in the other dimer producing a DADA (D-hydrogen
bonded donor and A- hydrogen bonded acceptor) array of
quadruple hydrogen bonds which is represented by graph-set
2-Amino-4,6-dimethoxypyrimidine-isophthalic Acid
(6). 6 crystallizes in triclinic P1 space group with two
̅
crystallographically independent molecules of MOP and one
molecule of IPA in the asymmetric unit. The trimer in the crystal
is the result of hydrogen-bond interactions between the two
carboxylic acid groups in IPA and the two MOP molecules. The
primary synthons in this structure are N−H···O and O−H···N
hydrogen bonds. In the 2:1 ligand-acid trimer, the free amino
proton in the B molecule of the pyrimidine moiety interacts with
an oxygen of the methoxy group in the neighboring trimer
(symmetry code 2 − x, 1 − y, −z) to form a hexameric
supermolecule via N2B−H2B2···O7A hydrogen bonds(Figure 9).
One of the amino group hydrogens (H2A of the A pyrimidine
molecule) and pyrimidine nitrogens (N3 of both pyrimidine
molecules) in pyrimidine moiety have not participated in any of
the hydrogen bonding interactions. The reason could be the
presence of two methoxy groups lying adjacent to each other in
the hexameric supermolecule which sterically hinders any donor
hydrogens from approaching the acceptors. The oval-shaped
hexameric supermolecules are further fused to neighboring ones
through weak C−H···O bonds (symmetry code x, 1 + y, 1 + z)
and extends itself along a direction to form a tape. Many parallel
2
2
2
notation R2 (8), R4 (8), and R2 (8) to form a cyclic
heterotetrameric synthon (Figure 7). These tetramers are also
arranged in a two-dimensional pattern without any neighboring
interaction along the (211) plane. There are two types of Π−Π
̅
stacking noted between the aminopyrimidine and the acid
moieties. The pyrimidine ring present in a plane, exhibits
stacking interactions with both acid molecules lying in parallel
planes above and below it with Cg−Cg distance of 3.6309(8) and
3.8526(8) Å and a slip angle of 15.38° and 25.35°, respectively
(Figure 8).
tapes lie alongside on the (122) plane as shown in Figure 10.
̅
3772
dx.doi.org/10.1021/cg3005954 | Cryst. Growth Des. 2012, 12, 3766−3785

Crystal Growth & Design
Article
Figure 4. Supramolecular sheet along (122) plane with no interaction by neighboring tapes in 2.
2
Figure 5. Extension of infinite tape connected by R2 (16) synthon in 3.
Two types of Π−Π stacking are noted; one between the
pyrimidine−acid and the other between two pyrimidine−
pyrimidine moieties. The tapes that extend along a one-
dimensional direction stack with adjacent tapes, one lying on
top (through pyrimidine A−acid stacking) and the other at the
bottom (pyrmidine A−pyrimidine B stacking) with Cg−Cg
distance of 3.6971(9) and 3.7526(8) Å and a slip angle of 21.23°
and 23.47°, respectively.
2-Amino-4,6-dimethoxypyrimidine-chloroacetic Acid
(7) and 2-Amino-4,6-dimethoxypyrimidinium Trichloro-
acetate (8). Both 7 and 8 crystallizes in the same monoclinic
P21/c space group. The asymmetric unit of 7 contains a neutral
molecule of MOP and CAA like the previous cases whereas 8
consists of a pyrimidinium cation (MOP+) and a triacetate anion
(TCAA¯). The pyrimidinium cation is a consequence of the
proton transfer from the acid to the pyrimidine ring. This is
reflected by an increase in C−N−C bond angle compared to the
neutral molecule.
the N3 pyrimidine nitrogen atom extends the architecture into a
four-component supermolecule which is a linear heterotetramer.
These supermolecules align themselves along a line without any
strong interaction to form a chain (although a formation of
2
R2 (6) motif is viable, the hydrogen bonds are not within range).
Such type of chains interpenetrate between each other in almost
perpendicular direction and look like a three-dimensional grid
(Figure 11). No stacking interactions are observed in this
structure.
Similarly secondary N−H···O¯ hydrogen bonds, formed from
the free amino proton and the carbonyl oxygen, extend 8 into a
four-component cyclic heterotetramer with DDAA array of
quadruple hydrogen bonds. The cyclic tetramer extends itself by
interacting through Cl···O bonds (involving Cl2 of the
trichloroacetate anion and O1 of the heterocycle), with
Cl2···O1 distance of 3.2192(16) Å and C−Cl···O angle of
162.2° (18) along the x-axis to form a supramolecular tape
(Figure 12). The supramolecular tapes are interconnected to the
neighboring ones through chlorine-chlorine interaction with
Cl···Cl distance of 3.4675(9) Å. No stacking interactions are
present in this structure (Figure 13).
The primary motif in 7 is composed of MOP and CAA which is
connected through a couple strong hydrogen bonds (O−H···N
and N−H···O hydrogen bonds), whereas in 8, it is made up of
charge-assisted hydrogen bond interactions between the
carboxylate and the amino-pyrimidinium moieties through
N−H···O¯ and N−H+···O¯ hydrogen bonds. In 7, the secondary
N−H···N hydrogen bonds formed from the amino proton and
2-Amino-4,6-dimethoxypyrimidinium 5-chlorosalicy-
late (9). The crystal structure of 9 crystallizes in orthorhombic
Pbca space group. The asymmetric unit contains one MOP+
cation and one 5CSA¯ anion. After the formation of a dimer
3773
dx.doi.org/10.1021/cg3005954 | Cryst. Growth Des. 2012, 12, 3766−3785

Crystal Growth & Design
Article
Figure 6. Supramolecular sheet along (113) plane without interaction by adjacent tapes in 3.
̅
Figure 7. Supramolecular sheet formed by cyclic tetramers in 5 along (211) plane with no interaction by neighboring ones.
̅
through the primary synthons (charge-assisted N−H···O¯ and
N−H+···O¯ hydrogen bonds between the carboxylate and the
amino-pyrimidinium moieties) in the crystal structure of 9, the
secondary N−H···N hydrogen bonds take over to form the four-
component linear heterotetramer. In the anion, a typical
intramolecular hydrogen bond is formed between the hydroxyl
group and the carboxylate anion with graph-set notation S(6).
Weak C−H···O bonds, involving the C−H of the methoxy group
and the carbonyl oxygen of the anion (C7−H7B···O1 hydrogen
bond), link the dimers to form a zigzag supramolecular chain
along the c-axis (Figure 14). Adjacently arranged tetramer units
in two different planes are stacked through Π−Π interactions
3774
dx.doi.org/10.1021/cg3005954 | Cryst. Growth Des. 2012, 12, 3766−3785

Crystal Growth & Design
Article
Figure 8. Stacking between two different planes in 5.
Figure 9. View of hexameric supermolecule in 6.
between an anion and a pyrimidinium base with Cg−Cg distance
of 3.6916(9) Å and a slip angle of 19.94° (Figure 15).
Figure 11. View of a segment of interpenetrated chains that form grid
like structure in 7.
2-Amino-4,6-dimethoxypyrimidinium 3-hydroxypico-
linate (10). MOP+·3HPA¯ crystallizes in triclinic space group P1
with the asymmetric unit containing two MOP+ cations and two
3HPA¯ anions. The formation of an organic salt is confirmed
from the protonation at the pyrimidine ring (evident from the
increase in C−N−C bond angle compared to the neutral
molecule). The salt displays exactly the same primary motif
distinct from the other two cases. Typical intramolecular
hydrogen bond exists between the hydroxyl −OH group and
the carboxylate group of the 3-hydroxypicolinate anion to form a
six-membered hydrogen-bonded ring [S(6)]. The two molecules
in the asymmetric unit initially form the same type of cyclic
heterotetramer and differ only in the further aggregation. The
cyclic heterotetramers formed by both molecules (A and B)
arrange themselves along a plane in different ways to form a long
strip of tape propagating along different directions (Figure 16).
Molecule A does not involve in any hydrogen bonds to form the
tape whereas molecule B links the neighboring ones through
̅
exhibited by MOP+·TCAA¯ and MOP+·5CSA¯, composed of
intermolecular interactions: N−H...O¯ and N−H+...O¯. The
primary motif is interlinked by the free amino hydrogen atom
and the acceptor nitrogen in the picolinate anion through
secondary hydrogen bonds to form a cyclic heterotetramer
Figure 10. Supramolecular sheet in 6 formed by hexamers along (122) plane with no interaction by neighboring tapes.
̅
3775
dx.doi.org/10.1021/cg3005954 | Cryst. Growth Des. 2012, 12, 3766−3785

Crystal Growth & Design
Article
Figure 12. View of supramolecular tape formed through chlorine−oxygen interaction in 8.
MP+·5CSA¯ is also a salt which is confirmed by the protonation
at the N1 position, reflected by an increase in C−N−C bond
angle compared to the neutral molecule. The two independent
molecules look almost identical with one another having similar
molecular interactions. They differ from each other with slight
variations in the bond distances, bond angles and hydrogen bond
lengths. The two crystallographically independent molecules are
related by pseudosymmetry. All alternate space groups are
eliminated by an analysis using PLATON. The primary synthons
are formed as result of two sets of inequivalent charge-assisted
N−H···O¯ and N−H+···O¯ hydrogen bonds in structure 11. The
ion pairs are linked through charge-assisted N−H···O¯ formed
from the free amino proton and carboxylate moiety to form a
cyclic heterotetramer. The tetramer is composed of small motifs
2
2
(represented by the graph-set notation R2 (8), R4 (8), and
2
R2 (8)) leading to DDAA array of quadruple hydrogen bonds as
in the case of 8. These larger synthons in turn aggregate in
(−101) plane through C5A−H5A···O3B and C5B−H5B···O3A
hydrogen bonds involving a larger ring motif to form a
supramolecular sheet as shown in Figure 18.
2-Amino-4,6-dimethylpyrimidinium hydrogen malo-
nate (12), 2-Amino-4,6-dimethylpyrimidine-glutaric acid
(13), and 2-Amino-4,6-dimethylpyrimidine-adipic Acid
(14). 12 crystallizes in triclinic P1, 13 in orthorhombic Pca2 and
̅
1
14 in P21/c space groups. MP·GA and MP·AA are sustained by
three-molecule aggregates consisting of two molecules of MP
and one molecule of acid in the asymmetric unit whereas
MP+·MA¯ consists of a single MP+ cation and a single MA¯ anion
in the asymmetric unit. The primary synthons in 12 are charge-
assisted N−H+···O−and N−H···O− hydrogen bonds forming a
dimer. The dimer is linked to one end by a cation through base
pair (couple of N−H···N bonds) and the other end by an anion
through a pair of weak C−H···O bonds involving the C−H of
methylene carbon and the carboxylate oxygen extending itself
Figure 13. Interlinking of adjacent tapes through Cl···Cl interation in 8.
2
weak C8B−H8B3···O8B bonds involving the R2 (6) motif
(symmetry code 1 − x, −y, 1 − z). Both tapes lie crisscross in the
overall pattern to form a grid like network as shown in Figure 17.
2-Amino-4,6-dimethylpyrimidinium 5-chlorosalicylate
(11). The crystal structure 11 crystallizes in monoclinic P21/c
space group. The asymmetric unit contains two crystallo-
graphically independent MP+ cations and 5CSA¯ anions.
2
into a zigzag tape with three different R2 (8) motifs. These zizzag
Figure 14. Formation of supramolecular chain through weak C−H···O interaction in 9.
3776
dx.doi.org/10.1021/cg3005954 | Cryst. Growth Des. 2012, 12, 3766−3785

Crystal Growth & Design
Article
Figure 15. Stacking interactions observed between two aminopyrimidine rings in 9 is shown.
2
Figure 16. Tetramers of molecule A lying alongside each other and of molecule B connected via R2 (6) synthon to form a supramolecular tape in 10.
Figure 17. View of interpenetrated strips of molecule A (represented by ball and stick) and molecule B (represented by space filled mode) in 10.
3777
dx.doi.org/10.1021/cg3005954 | Cryst. Growth Des. 2012, 12, 3766−3785

Crystal Growth & Design
Article
Figure 18. Interlinking cyclic heterotetramers through weak C−H···O bonds leads to the formation of supramolecular sheet in 11.
Figure 19. View of supramolecular corrugated sheet along (120) plane in 12.
tapes are interlinked through C5−H5···O5 bonds with neigh-
boring ones to form a planar corrugated sheet along the (120)
plane (Figure 19). In 13 and 14, the primary intermolecular
O−H···N and N−H···O hydrogen bonds between dicarboxylic
acids and MP are essentially the same on both sides of the diacid.
The trimers are again arranged into extended 1-D zigzag infinite
chains connected by the terminal MP at both ends through
base pairing (Figure 20 and 21). In both the compounds, alternative
chains have a similar pattern of arrangements having their base
pair aligned in a same direction whereas the neighboring sym-
metry related chains have the base pair aligned in the opposite
direction. The difference lies only in the dimensionality main-
tained by both the structures. The zigzag chains arranged in 14 lie
along the same plane unlike 13 showing 3D nature. The planes of
the two carboxylic acid moieties (involving the carbonyl carbon
and the two oxygen atoms) in 14 are slightly twisted with respect
to each other, having an angle of 4.82°, which suggests that the
trimeric supermolecule is almost in a single plane which in turn
leads to a planar arrangement of the entire zigzag chains along the
(101) plane. In 13, the angle of twist between both carboxylic
acid groups with respect one another is 79.38°, which eventually
leads to the nonplanar trimer. The extended infinite chain
obviously lies out of plane unlike 14. No stacking interactions are
found in both 12 and 13. The planar sheets of 14 are stabilized
through Π−Π stacking interactions between two of the
pyrimidine B molecules with interplanar distance, Cg−Cg
distance and slip angle of 3.4698(9) Å, 3.6048(13) Å and
15.73° respectively (Figure 22).
DISCUSSION
■
By using Cambridge Structural Database (CSD), several re-
currently occurring hydrogen-bonded synthons have been re-
cognized and identified by Desiraju and Allen et al.22,23,51 One
such frequently occurring bimolecular cyclic hydrogen bonded
2
synthon is a ring motif with graph set notation, R2 (8), which is
readily formed between a carboxylic acid and an aminopyri-
midine group. However we are much interested on what happens
after this step and how they go on to self-assemble into larger
supramolecular assemblies.
In the present study, several small and large sized ring motifs
2
2
2
4
2
are observed, such as R2 (6), R2 (8), R4 (8), R4 (14), R2 (16),
4
6
6
R6 (22), R8 (24), and R6 (32). But not all of them occur
3778
dx.doi.org/10.1021/cg3005954 | Cryst. Growth Des. 2012, 12, 3766−3785

Crystal Growth & Design
Article
Figure 20. Infinite zigzag chains formed in 13.
Figure 21. Infinite zigzag chains of 14 lying parallel to each other along the (101) plane.
Figure 22. Π−Π stacking between pyrimidines moieties of different planar chains in 14.
3779
dx.doi.org/10.1021/cg3005954 | Cryst. Growth Des. 2012, 12, 3766−3785

Crystal Growth & Design
Article
2
changes are observed. Subsequent levels of aggregation form
either a linear heterotetramer (LHT) or cyclic heterotetramer
(CHT) or heterotrimer (HT) or any other preferred stable
synthon. Scheme 3 shows the common synthons that are formed
while Table 3 and Table 4 lists out the type of synthons involved in
each of the molecular complexes.
Scheme 2. Three Types of R2 (8) Motifs
Among the 14 structures, 9 of them readily form linear hetero-
tetrameric synthon (1, 2, 3, 4, 7, 9, 12, 13, and 14), 3 of them
form a cyclic tetramer (5, 8, and 11), 1 of them forms a trimer
(6), and the other (10) forms a relatively different type of cyclic
tetramer.
The 9 linear heterotetramers are formed by the combination of
2
three R2 (8) motifs. Structures 1, 2 and 3 have the linear
tetramers arranged in a plane and incidentally all of them have
crystallized in the same space group, P-1. Each linear tetramer is
surrounded by six such tetramers in proximity to form a planar
sheet as shown in Figure 23. The similar type of arrangement of
the sheets has lead the closeness in the lattice parameters in
structure 1 and 2 but not 3. This is explained by the fact that the
linear heterotetramers in 1 and 2 are connected to its neighbor
repeatedly, as most of them arise only in specific cases.
2
2
2
Nevertheless R2 (6), R2 (8), and R4 (8) motifs have been the
most recurrently occurring ones in the present study. Among the
three the later two are involved through the formation of strong
hydrogen bonds whereas the former one is formed through weak
2
through R2 (6) motif whereas 3 has not got the luxury because of
2
bonds. The high-probability R2 (8) motif is special because they
its structural chemistry. Tetrameric structures 4, 7, and 9 do not
lie in a plane and therefore have distinct patterns. 12 also
forms a linear heterotetramer like the rest of the structures
and further organizes through other hydrogen bonds to form
larger supramolecular architectures. Although 13 and 14 give
the impression that they have formed a heterotrimer initially, the
extended trimeric infinite chain has revealed that they are built
emerge through three different ways (by involvement of different
hydrogen bonds) as shown in Scheme 2.
In all the 14 structures, the primary motif is the formation of
pyrimidine-carboxylic acid dimer through the predominant
2
R2 (8) synthon. The primary intermolecular bonds involved
are the O−H···N/O−···H−N+ and N−H···O/N−H···O− bonds
and it is only in next step that the higher levels of structural
Scheme 3. Representation of Three Types of Synthons Formed by Combination of Pyrimidine and Carboxylic Acid
3780
dx.doi.org/10.1021/cg3005954 | Cryst. Growth Des. 2012, 12, 3766−3785

Crystal Growth & Design
Article
Table 3. Co-crystals/Salts of 2-Amino-4,6-dimethyl Pyrimidine Are Listed in the Increasing Order of ΔpKa, Calculated Using
a
Advanced Chemistry Development Lab Software and SPARC Software, Respectively (Δ pKa = pKa(base) − pKa(acid))
Advanced Chemistry Development Lab Software
SPARC software
salt/co-
salt/co-
crystal
base (pKa)
MP
acids (pKa)
ΔpKa
crystal
ref
base (pKa)
acids (pKa)
ΔpKa
synthons
4.68 2-amino benzoic acid 4.94
4-amino benzoic acid 4.86
−0.26
−0.18
0.11
co-crystal
co-crystal
co-crystal
45
33
34
MP 5.24 4-amino benzoic acid
4.75
4.65
4.64
0.49
0.59
0.6
co-crystal
co-crystal
co-crystal
LHT
LHT
LHT
adipic acid
4-hydroxy benzoic
acid
4.57
4.53
4.47
4.44
4.39
4.34
4.33
2-amino benzoic acid
3-benzoyl propionic
acid
0.15
0.21
0.24
0.29
0.34
0.35
0.41
0.44
co-crystal
co-crystal
co-crystal
co-crystal
co-crystal
co-crystal
co-crystal
co-crystal
33
52
52
glutaric acid
4.51
4.42
4.39
4.28
4.27
4.18
4.17
4.16
0.73
0.82
0.85
0.96
0.97
1.06
1.07
1.08
co-crystal
co-crystal
co-crystal
co-crystal
co-crystal
co-crystal
co-crystal
co-crystal
LHT
LHT
HT
4-methoxy benzoic
acid
4-methoxy benzoic
acid
3,4-dimethyl benzoic
acid
3 benzoyl propionic
acid
adipic acid
present
52
3,4-dimethyl benzoic
acid
HT
3,5-dimethyl benzoic
acid
succinic acid
HT
glutaric acid
present
53
3,5-dimethyl benzoic
acid
HT
3-methyl benzoic acid 4.27
4-hydroxy benzoic
acid
others
LHT
succinic acid
benzoic acid
4.24
35
2,4-dimethyl benzoic
acid
4.20
4.18
0.48
0.50
co-crystal
co-crystal
52
52
cinnamic acic
4.11
4.1
1.13
1.14
co-crystal
co-crystal
HT
2,4-dimethyl benzoic
acid
2,4,6-trimethyl
benzoic acid
LHT
2,5-dimethyl benzoic
acid
3.99
0.69
0.71
co-crystal
co-crystal
52
53
3-methyl benzoic acid 4.1
1.14
1.18
co-crystal
co-crystal
LHT
LHT
4-chloro benzoic acid 3.97
2,5-dimethyl benzoic
acid
4.06
4.03
2-methyl benzoic acid 3.95
0.73
0.80
0.83
co-crystal
co-crystal
co-crystal
53
53
52
benzoic acid
1.21
1.26
1.49
co-crystal
co-crystal
co-crystal
LHT
LHT
LHT
cinnamic acid
3.88
3.85
2-methyl benzoic acid 3.98
4-chloro benzoic acid 3.75
2,4,6-trimethyl
benzoic acid
terephthalic acid
3-nitro benzoic acid
acetyl salicylic acid
4-nitro benzoic acid
fumaric acid
3.49
3.48
3.48
3.42
3.15
3.01
1.19
1.20
1.20
1.26
1.53
1.67
1.71
1.73
1.75
co-crystal
co-crystal
co-crystal
co-crystal
co-crystal
salt
36
53
52
53
37
38
53
33
52
terephthalic acid
acetyl salicylic acid
fumaric acid
3.73
3.63
3.47
3.35
3.21
3.19
1.51
1.61
1.77
1.89
2.03
2.05
2.12
2.16
2.27
co-crystal
co-crystal
co-crystal
co-crystal
co-crystal
co-crystal
co-crystal
salt
LHT
LHT
HT
phthalic acid
HT
4-nitro benzoic acid
3-nitro benzoic acid
LHT
LHT
LHT
LHT
LHT
salicylic acid
2-chloro benzoic acid 2.97
co-crystal
co-crystal
co-crystal
2-chloro benzoic acid 3.12
phthalic acid
2.95
2.93
salicylic acid
malonic acid
3.08
2.97
2,3-difluoro benzoic
acid
salt
malonic acid
2.92
2.77
1.76
1.91
2.04
salt
salt
salt
present
39
2,3-difluoro benzoic
2.87
2.37
2.59
2.88
co-crystal
salt
LHT
CHT
LHT
acid
3,5-dinitro benzoic
acid
5-chloro salicylic acid 2.65
5-chloro salicylic acid 2.64
present
3,5-dinitro benzoic
acid
2.36
salt
a
In addition to this all the type of synthons formed are also denoted for the corresponding salts and co-crystals.
through discrete heterotetrameric units (combination of three
R2 (8) motifs).
guidelines. 3-hydroxy picolinic acid possesses an additional
strong donor and acceptor, respectively. The first level of organi-
zation in this case is the pyrimidine-carboxylic acid interaction as
discussed earlier and differs only in the higher level of supra-
molecular organization (whether to form a linear heterotetramer,
cyclic heterotetramer or a heterotrimer). However in structure
10 after the formation of dimer, the strong acceptor N4 of
picolinic acid readily tends to accept a donor N2−H2, amino
group of the pyrimidine to form a strong bond in preference to
O2 (carboxyl oxygen of the acid molecle) thereby disrupting the
formation of a regular cyclic heterotramer.
2
The next three structures which include 5, 8, and 11, form
2
2
cyclic heterotetramer through the combination of R2 (8), R4 (8),
2
and R2 (8) motifs. Two of them are co-crystals and one forms a
salt. Among these, only 5 and 11 form planar supramolecular
sheets.
2
Structure 6 forms the lone trimer built upon by three R2 (8)
motifs. This trimer couples with inversely related trimer to form a
hexamer, where such hexamer units are interlinked through weak
bonds to form infinite ribbons.
One can reason out why structure 10 has not formed one of
the expected supermolecule based on the best donor/acceptor
The above results in conjunction with the literature survey
of co-crystals/salts of MP and MOP shows that 30 out of
3781
dx.doi.org/10.1021/cg3005954 | Cryst. Growth Des. 2012, 12, 3766−3785

Crystal Growth & Design
Article
Table 4. Co-crystals/Salts of 2-Amino-4,6-dimethoxy Are Listed in the Increasing Order of ΔpKa, Calculated Using Advanced
a
Chemistry Development Lab Software and SPARC Software, respectively (Δ pKa = pKa(base) − pKa(acid))
advanced chemistry development lab software
SPARC software
salt/co-
salt/co-
crystal
base (pKa)
MOP
acids (pKa)
ΔpKa
−2.01
crystal
ref
40
base (pKa)
MOP
acids (pKa) ΔpKa
synthons
CHT
2.93 2-amino benzoic
4.94
4.86
4.57
4.49
4.47
4.33
4.27
4.24
3.88
co-crystal
2.85 4-amino benzoic
4.75
4.66
4.64
4.42
4.27
4.17
4.11
4.1
−1.9
co-crystal
co-crystal
co-crystal
co-crystal
co-crystal
salt
acid
acid
4-amino benzoic
acid
−1.93
−1.64
−1.56
−1.54
−1.4
co-crystal
salt
41
indole-3-acetic acid
−1.81
−1.79
−1.57
−1.42
−1.32
−1.26
−1.25
others
LHT
CHT
CHT
LHT
LHT
LHT
4-hydroxy benzoic
acid
42
2-amino benzoic
acid
indole-3-acetic acid
co-crystal
co-crystal
salt
46
4-methoxy benzoic
acid
4-methoxy benzoic
acid
present
37
succinic acid
gallic acid
4-hydroxy benzoic
acid
3-methyl benzoic
acid
−1.34
−1.31
co-crystal
co-crystal
present
35
cinnamic acid
co-crystal
co-crystal
succinic acid
3-methyl benzoic
acid
cinnamic acid
−0.95
−0.49
co-crystal
co-crystal
present
present
gallic acid
3.98
3.87
−1.13
−1.02
salt
salt
LHT
4-nitro benzoic acid 3.42
3-hydroxy picolinic
acid
others
isophthalic acid
3.53
−0.6
co-crystal
present
pyridine-2,6-
dicarboxylic acid
3.9
−1.05
salt
CHT
fumaric acid
maleic acid
tartaric acid
salicylic acid
3.15
3.15
3.07
3.01
2.97
−0.22
−0.22
−0.14
−0.08
−0.04
salt
37
isophthalic acid
citric acid
3.72
3.71
3.47
3.47
3.35
−0.87
−0.86
−0.62
−0.62
−0.5
co-crystal
salt
HT
salt
37
others
CHT
others
HT
salt
42
fumaric acid
maleic acid
salt
salt
43
salt
2-chloro benzoic
acid
co-crystal
present
phthalic acid
co-crystal
pyridine-2,6-
dicarboxylic acid
2.97
−0.04
salt
42
4-nitro benzoic acid 3.21
−0.36
co-crystal
LHT
phthalic acid
citric acid
2.95
2.93
−0.02
0
co-crystal
salt
44
37
tartaric acid
3.16
3.12
−0.31
−0.27
salt
others
LHT
2-chloror benzoic
acid
co-crystal
2,4-dichloro
benzoic acid
2.68
0.25
co-crystal
37
salicylic acid
3.08
−0.23
salt
CHT
chloro acetic acid
2.65
2.64
0.28
0.29
co-crystal
salt
present
present
chloroacetic acid
2.97
2.83
−0.12
0.02
co-crystal
co-crystal
LHT
CHT
5-chloro salicylic
acid
2,4-dichlro benzoic
acid
3-hydroxy picolinic
acid
1.14
1.79
2.84
salt
salt
present
present
5-chloro salicylic
acid
2.65
0.66
0.2
salt
salt
LHT
CHT
trichloro acetic acid. 0.09
trichloroacetic acid
2.19
a
In addition to this all the type of synthons formed are also denoted for the corresponding salts and co-crystals. LHT = linear heterotetramer,
CHT = cyclic heterotetramer, and HT= heterotrimer.
54 (around 56%) leads to the formation of linear heerotetramer
synthon, 9 of them form cyclic heterotetramers (around 17%),
9 others form heterotrimers (around 17%) and the rest form
other new synthons (around 11%). This clearly states that the
linear heterotetramer synthon is the most predominantly existing
synthon. It is also predicted to be the most stable one formed
when an aminopyrimidine and carboxylic acid interact with each
other which is exemplified by a crystal structure prediction study
carried out by Thakur and Desiraju.52 In fact one of our recently
published work also clearly demonstrates that this stable linear
heterotetrameric synthon is responsible for the formation several
isostructural co-crystals.53
The present work has produced a list of multicomponent
crystals which include salts and co-crystals. It is well-known that
when an acid crystallizes with a base it either forms a salt or co-
crystal (Scheme 4). The rationalization to the formation of these
salts/co-crystals involving MOP and MP with various acids has
also been attempted to explain in terms of pKa values. Among the
14 structures, 9 of them are co-crystals while the remaining are
salts. From the literature it is clear that the acid dissociation
constant, pKa, have been used quite successfully in predicting
and rationalizing the formation of salts/co-crystals. We have
tried to reason the observed structural behavior through a
comparative analysis of the calculated pKa values for all the
structures. The two pyrimidine derivatives are basic com-
pounds with calculated pKa values of 4.68 (5.24) and 2.93
(2.85) respectively.54,56 Molecular complexes of these
pyrimidines with various carboxylic acids (present work and
the reported structures) in the increasing order of pKa are
presented in Table 3 and Table 4.
The C−N−C bond angle in MOP and MP, which involves
acid−base interaction range from 115 to 118° in most cases
(indicative of co-crystal fomation) and increases to a higher range
of 120−122° indicating the formation of salts. This is supported
by the C−O bond distances determined through X-ray crystallo-
graphic study. A plot for all the structures including those in the
literature, with X-axis corresponding to the ΔpKa [where ΔpKa =
pKa(base) − pKa(acid)] and Y-axis corresponding to ΔDC−O
3782
dx.doi.org/10.1021/cg3005954 | Cryst. Growth Des. 2012, 12, 3766−3785

Crystal Growth & Design
Article
Figure 23. View of a supramolecular sheet formed in 1, 2, and 3 where each tetramer is surrounded by six adjacent tetramers. The ellipses denote the
2
formation of R2 (6) synthon between adjacent methoxy groups in 1 and 2.
Scheme 4. Scheme Combination of Acid and Base to Give Either a Co-crystal or Salt
Figure 24. Plot of ΔpKa vs ΔDc-o for all the MP (left) and MOP (right) co-crystals/salts with various carboxylic acids.
(difference between the lengths of the two C−O bonds in a carboxyl
group) bond lengths, is shown in Figure 24. The scattergram
indicates that the densely populated top left corner corresponds
to co-crystals having larger ΔDC−O and the scarcely populated
bottom right corner to formation of salts. It is observable that for
MP there is some low negative corelation whereas for MOP the
regularity has been unseen and no correlation. A similar plot has
been reported in the literature.57
Even if the usefulness of pKa values is dubious in the ability to
predict the formation of co-crystals/salts, the approach has made
possible to identify the occurrence of proton transfer atleast
before the actual experiment has been carried out. Several papers
have reported the use of pKa to predict salt/co-crystals.57−55 It
has been identified that the reliability and the applicability have
been uncertain when the ΔpKa ranges between 0 and 3. It is
expected that salt formation is predominant when ΔpKa is
3783
dx.doi.org/10.1021/cg3005954 | Cryst. Growth Des. 2012, 12, 3766−3785

Crystal Growth & Design
Article
greater than 3, and likewise they correctly form co-crystals when
the ΔpKa is negative. There is poor predictive power between the
range of 0 to 3,57 and this narrow continuum differs for different
system. Infact we have tried to identify the narrow continuum
that exists for our system containing two aminopyrimidine bases.
The ΔpKa calculated using Advanced Chemistry Develop-
ment Software54 shows good agreement with the results ob-
tained with only few exceptions, which is likely when a large set of
data is analyzed (Table 3). The narrow transition where the salts
and co-crystals overlap with each other is between ΔpKa 1.50
and 1.75 in the case of MP. In order to cross-check whether the
theoretical calculations performed were reliable, the same sets of
calculations were performed with another software named
SPARC.56 The results substantiated the earlier results
qualitatively but failed quantitatively. The change in pKa from
4.68 (MP) to 2.93 (MOP) is expected to form more number of
co-crystals, in contrast the percentage of co-crystals formed are
88% and 52% for MP and MOP respectively. It is also evident
that the pKa of MOP does not clearly predict the formation of co-
crystals despite the fact that ΔpKa values for majority of the
complexes lie below zero leading to some ambiguity (Table 4).
Not all systems that are considered strictly follow the pKa
guidelines. The results do not show the expected outcome
which could be due to several factors such as the environment of
the crystal structure, methods of reaction conditions used in the
preparation of the complexes, nature of solvents and even the
unreliable theoretical values of pKa used in the calculation.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: tommtrichy@yahoo.co.in Phone: ++91-431-2407053
Fax: ++91-431-2407045, 2407030.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The authors thank the DST-India (FIST programme) for the use
of Bruker SMART APEX II CCD diffractometer at the School of
Chemistry, Bharathidasan University.
REFERENCES
■
(1) Legon, A. C.; Millen, D. J. In Principles of Molecular Recognition;
Buckingham, A. D., Legon, A. C., Roberts, S. M., Eds.; Blackie Academic
and Professional: London, 1993.
(2) von Hippel, P. H.; Berg, O. G. Proc. Natl. Acad. Sci. U. S. A. 1986, 83,
1608−1612.
(3) Hippel, P. H. V. ; Berg, O. G. In DNA-Ligand Interactions; Plenum
Publishing Corp.: New York, 1987.
(4) Sundberg, E. J.; Mariuzza, R. A. Adv. Protein Chem. 2002, 61, 119−
160.
(5) Weis, W. I. Annu. Rev. Biochem. 1996, 65, 441−473.
(6) Ofek, I.; Sharon, N. Infect. Immun. 1988, 56, 539−547.
(7) Lehn, J. M. Supramolecular Chemistry: Concepts and Perspectives;
VCH: Weinheim, Germany, 1995.
(8) Atwood, J. L.; Davies, J. E. D.; MacNicol, D. D.; Vogtle, F. In
Comprehensive Supramolecular Chemistry; Lehn, J.-M., Ed.; Pergamon:
Oxford, U.K., 1996.
(9) Aakeroy, C. B.; Beatty, A. M. Aust. J. Chem. 2001, 54, 409−421.
(10) Steiner, T. Angew. Chem., Int. Ed. 2002, 41, 48−76.
(11) Bernstein, J.; Davis, R. E.; Shimoni, L.; Chang, N. L. Angew. Chem.,
Int. Ed. Engl. 1995, 34, 1555−1573.
̈
CONCLUSION
■
This systematic study reveals that the complementary pair of O−
H···N/O¯···H−N+ and N−H···O¯/N−H···O hydrogen bonds in
the family of MOP/MP- carboxylic acids readily assemble to
2
form R2 (8) motif which further self-assembles to aid in building
a larger supermolecule - LHT, CHT, HT and others. Since 56%
of the structures form the LHT, the high supramolecular yield
implicates the reliability of this synthon.
(12) Kollman, P. A. J. Am. Chem. Soc. 1972, 94, 1837−1842.
(13) Aakeroy, C. B.; Seddon, K. R. Chem. Soc. Rev. 1993, 22, 397−407.
(14) Walsh, R. D. B.; Bradner, M. W.; Fleischman, S.; Morales, L. A.;
Moulton, B.; Rodriguez-Hornedo, N.; Zaworotko, M. J. Chem. Commun.
2003, 186−187.
In the absence of other strong functional groups in the
aromatic ring, it is the weak C−H···O, Cl···Cl, and stacking inter-
actions that determine the overall arrangement of the tetramers.
Therefore the heterotetrameric synthons can be used extensively
for the construction of larger supramolecular architectures by
substituting suitable functional groups in the aromatic acid with
the knowledge on the hierarchy of hydrogen bonds.
(15) Bis, J. A.; Zaworotko, M. J. Cryst. Growth Des. 2005, 5, 1169−
1179.
(16) Vishweshwar, P.; Nangia, A.; Lynch, V. M. Cryst. Growth Des.
2003, 3, 783−790.
(17) Aakeroy, C. B.; Beatty, A. M.; Helfrich, B. A. J. Am. Chem. Soc.
2002, 124, 14425−14432.
2
Apart from all these synthons, a new R2 (6) motif, formed pri-
(18) Bhogala, B. R.; Vishweshwar, P.; Nangia, A. Cryst. Growth Des.
2002, 2, 325−328.
marily between a couple of inversely related methoxy group
seems to be frequently occurring in our study. Although in most
cases this is far from the range of the regular C−H···O bond
lengths, a thorough investigation could provide some insight into
this new synthon. Structures 1, 2, and 10 show this type of
synthons.
The above results and the synthetic strategies can also be
employed practically to other biological relavant systems.
On the basis of molecular co-crystals/salts presented here as
well as those reported from literature, the ΔpKa clearly demon-
strates that MP is a good co-crystal former when combined with
carboxylic acids in contrast to MOP, which is of unpredictive
nature.
(19) Fleischman, S. G.; Kuduva, S. S.; Mcmahon, J. A.; Moulton, B.;
Walsh, R. D. B.; Rodriguez-Hornedo, N.; Zaworotko, M. J. Cryst. Growth
Des. 2003, 3, 909−919.
(20) Reddy, L. S.; Nangia, A.; Lynch, V. M. Cryst. Growth Des. 2004, 4,
89−94.
(21) Vangala, V. R.; Mondal, R.; Broder, C. K.; Howard, J. A. K.;
Desiraju, G. R. Cryst. Growth Des. 2005, 5, 99−104.
(22) Allen, F. H.; Raithby, P. R.; Shields, G. P.; Taylor, R. Chem.
Commun. 1998, 1043−1044.
(23) Allen, F. H.; Motherwell, W. D. S.; Raithby, P. R.; Shields, G. P.;
Taylor, R. New J. Chem. 1999, 25−34.
(24) Jeffrey, G. A.; Saenger, W. Hydrogen Bonding in Biological
Structures; Springer-Verlag: Berlin, 1991.
(25) Schmidt, L. H.; Harrison, J.; Rossan, R. N.; Vaughan, D.; Crosby,
R. Am. J. Trop. Med. Hyg. 1977, 26, 837−849.
ASSOCIATED CONTENT
* Supporting Information
Crystallographic information files (CIF) of structures 1−14.
This material is available free of charge via the Internet at http://
pubs.acs.org.
■
(26) Vallee, B. L.; Auld, D. S. Acc. Chem. Res. 1993, 26, 543−551.
(27) Hunt, W. E.; Schwalbe, C. H.; Bird, K.; Mallinson, P. D. J. Biochem.
1980, 187, 533−536.
S
(28) Baker, B. R.; Santi, D. V. J. Pharm. Sci. 1965, 54, 1252−1257.
(29) Serajuddin, A. T. M. Adv. Drug Delivery Rev. 2007, 59, 603−616.
3784
dx.doi.org/10.1021/cg3005954 | Cryst. Growth Des. 2012, 12, 3766−3785

Crystal Growth & Design
Article
(30) Blagden, N.; de Matas, M.; Gavan, P. T.; York, P. Adv. Drug
Delivery Rev. 2007, 59, 617−630.
(31) Reddy, L. S.; Bethune, S. J.; Kampf, J. W.; Hornedo, N. R. Cryst.
Growth Des. 2009, 9, 378−385.
(32) Stanton, M. K.; Tufekcic, S.; Morgan, C.; Bak, A. Crystal Cryst.
Growth Des. 2009, 9, 1344−1352.
(33) Balasubramani, K. Ph.D. Thesis, School of Chemistry,
Bharathidasan University, 2007.
(34) Balasubramani, K.; Muthiah, P. T.; Lynch, D. E. Acta Crystallogr.
2006, E62, o2907−o2909.
(35) Devi, P. Ph.D. Thesis, School of Chemistry, Bharathidasan
University, 2009.
(36) Devi, P.; Muthiah, P. T. Acta Crystallogr. 2007, E63, o4822−
o4823.
(37) Thanigaimani, K. Ph.D. Thesis, Bharathidasan University, 2009.
(38) Muthiah, P. T.; Balasubramani, K.; Rychlewska, U.; Plutecka, A.
Acta Crystallogr. 2006, C62, o605−o607.
(39) Subashini, A.; Muthiah, P. T.; Lynch, D. E. Acta Crystallogr. 2008,
E64, o426.
(40) Thanigaimani, K.; Muthiah, P. T.; Lynch, D. E. Acta Crystallogr.
2008, E64, o107−o108.
(41) Thanigaimani, K.; Muthiah, P. T.; Lynch, D. E. Acta Crystallogr.
2006, E62, o2976−o2978.
(42) Thanigaimani, K.; Muthiah, P. T.; Lynch, D. E. Acta Crystallogr.
2007, C63, o295−o300.
(43) Thanigaimani, K.; Muthiah, P. T.; Lynch, D. E. Acta Crystallogr.
2007, E63, o4555−o4555.
(44) Thanigaimani, K.; Muthiah, P. T.; Lynch, D. E. Acta Crystallogr.
2007, E63, o4212.
(45) Ebenezer, S.; Muthiah, P. T. Acta Crystallogr. 2010, E66, o516.
(46) Ebenezer, S.; Muthiah, P. T. Acta Crystallogr. 2010, E66, o2634−
o2635.
(47) Bruker. SMART APEX2, SAINT, and SADABS; Bruker AXS Inc.:
Madison, WI, U.S.A., 2008.
(48) Sheldrick, G. M. Acta Crystallogr. A. 2008, 64, 112−122.
(49) Spek, A. L. J. Appl. Crystallogr. 2003, 36, 7−13.
(50) Macrae, C. F.; Bruno, I. J.; Chisholm, J. A.; Edgington, P. R.;
McCabe, P.; Pidcock, E.; Rodriguez-Monge, L.; Taylor, R.; Streek, J. V.
D.; Wood, P. A. J. Appl. Crystallogr. 2008, 41, 466−470.
(51) Desiraju, G. R. Angew. Chem., Int. Ed. Engl. 2003, 34, 2311−2327.
(52) Thakur, T. S.; Desiraju, G. R. Cryst. Growth Des. 2008, 8, 4031.
(53) Ebenezer, S.; Muthiah, P. T.; Butcher, R. J. Cryst. Growth Des.
2011, 11, 3579−3592.
(54) Calculated using Advanced Chemistry Development (ACD/
Labs) Software V 8.00 for Microsoft windows (1994−2009 ACD/Labs)
(55) Hilal, S. H.; Karickhoff, S. W.; Carreira, L. A. Quant. Struc. Act. Rel
1995, 14, 348−355.
(56) Childs, S. L.; Stahly, G. P.; Park, A. Mol. Pharmaceutics 2007, 4,
323−338.
(57) Aakeroy, C. B.; Desper, J.; Urbina, J. F. Chem. Commum. 2005,
̈
2820−2822.
(58) Aakeroy, C. B.; Desper, J.; Scott, B. M. T. Chem. Commun. 2006,
̈
1445−1447.
(59) Aakeroy, C. B.; Beatty, A. M.; Helfrich, B. A. Angew. Chem., Int. Ed.
2001, 40, 3240−3242.
(60) Lynch, D. E.; McClenaghan, I. Acta Crystallogr. 2001, C57, 830−
̈
832.
3785
dx.doi.org/10.1021/cg3005954 | Cryst. Growth Des. 2012, 12, 3766−3785