Exploring the structural landscape of 2-aminopyrazines via co-crystallizations

Article abstract of DOI:10.1039/c2ce25516b

A correlation between the electrostatic charge on the hydrogen-bond acceptor sites of 2-aminopyrazine derivatives and the ability of the compound to form intermolecular hydrogen bonds with carboxylic acids in the solid state has been established. The charge on the hydrogen-bond acceptor can be modulated which leads to a predictable lowering of the supramolecular yield of the reaction. The outcome of all reactions was screened using IR spectroscopy, and twelve new crystal structures are reported to verify the spectroscopic assignments, and to examine the exact nature of the primary intermolecular interactions. The binding preference of carboxylic acids towards the two possible binding sites of 2-aminopyrazines has also been examined, and the main driving force for the assembly of the heteromer between bases and carboxylic acids is the two-point O-H...N/O...H-N synthon. However, seven out of twelve times carboxylic acids also bind via a single-point O-H...N synthons. This 'synthon crossover' is unavoidable due to highly competitive binding sites present in the N-heterocyclic bases chosen.

Full text of DOI:10.1039/c2ce25516b

View Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
CrystEngComm
Cite this: CrystEngComm, 2012, 14, 5845–5853
www.rsc.org/crystengcomm
PAPER
Exploring the structural landscape of 2-aminopyrazines via co-
crystallizations{
Christer B. Aakero¨y,* Prashant D. Chopade, Claudia Ganser, Arbin Rajbanshi and John Desper
Received 6th April 2012, Accepted 28th May 2012
DOI: 10.1039/c2ce25516b
A correlation between the electrostatic charge on the hydrogen-bond acceptor sites of
2-aminopyrazine derivatives and the ability of the compound to form intermolecular hydrogen bonds
with carboxylic acids in the solid state has been established. The charge on the hydrogen-bond
acceptor can be modulated which leads to a predictable lowering of the supramolecular yield of the
reaction. The outcome of all reactions was screened using IR spectroscopy, and twelve new crystal
structures are reported to verify the spectroscopic assignments, and to examine the exact nature of the
primary intermolecular interactions. The binding preference of carboxylic acids towards the two
possible binding sites of 2-aminopyrazines has also been examined, and the main driving force for the
…
…
assembly of the heteromer between bases and carboxylic acids is the two-point O–H N/O H–N
…
synthon. However, seven out of twelve times carboxylic acids also bind via a single-point O–H
synthons. This ‘synthon crossover’ is unavoidable due to highly competitive binding sites present in
the N-heterocyclic bases chosen.
N
other high-value chemicals7 without altering molecular struc-
ture.8
Although many non-covalent interactions are used in co-
Introduction
Supramolecular chemistry is founded upon reversible non-
covalent interactions that are capable of error correction through
thermodynamic equilibration. Supramolecular synthesis can,
under ideal circumstances, be performed with a minimum
amount of effort by using modular subunits encoded with
specific functional groups that produce pre-determined archi-
tectures through robust and directional molecular recognition
events.1 Supramolecular synthons2 are kinetically defined
structural units that represent the essence of crystals in terms
of molecular recognition. A fully rational design of specific
molecular solid-state architectures requires a detailed under-
standing of synthons, but even so, crystal engineering3 is always
likely to offer formidable challenges arising from the use of
reversible interactions and the inherent limitations in one-pot
reactions.
crystal synthesis ranging from weak p–p interactions9 to recently
emerging halogen bonds,10 hydrogen bonding11 remains the
primary synthetic tool in this area. Our working strategy for the
synthesis of binary co-crystals is based on controlling the affinity
between different molecules bearing complementary hydrogen-
bonding moieties, thereby manipulating the balance between
homomeric and heteromeric interactions (Scheme 1).12 As part
of this strategy, we first need to map out the structural landscape
that surrounds interactions between a wide range of molecules
bearing a multitude of functional groups.13
We have recently shown how hydrogen bonds and halogen
bonds can be used side-by-side without synthon crossover as
long as the primary molecular recognition events are designed
In the late 1980’s, Etter and co-workers systematically
employed co-crystallization reactions for probing binding pre-
ferences and patterns of functional-group recognition,4 and this
approach still offers arguably the best experimental method for
ranking the relative importance of intermolecular interactions.5
In addition, co-crystals have recently attracted considerable
interest as they provide opportunities for changing physical
properties of active pharmaceutical ingredients (APIs),6 and
Department of Chemistry, Kansas State University, Manhattan, KS,
66506, USA. E-mail: aakeroy@ksu.edu
{ Electronic supplementary information (ESI) available: Details of the
crystallographic work and cif files. CCDC reference numbers 876776–
876787. For ESI and crystallographic data in CIF or other electronic
format see DOI: 10.1039/c2ce25516b
Scheme 1 Recrystallization (top) yields homomeric solids. The co-
crystallization process (bottom) leads to a heteromeric product.
This journal is ß The Royal Society of Chemistry 2012
CrystEngComm, 2012, 14, 5845–5853 | 5845

View Online
around a careful combination of geometric and electrostatic
complementarity.14
In order to develop versatile supramolecular synthetic
strategies based on HB (hydrogen bond) driven self assembly,
it is necessary to identify building blocks that display reliable
binding preferences in the presence of a range of chemical
functionalities. In this study, we have designed ditopic molecules
(with two different HB acceptor sites) to test the binding
preferences in order to identify a ranking of synthons within an
intermolecularly competitive framework. Our choice of building
block is guided by a survey of the CSD,15 which plays a vital part
in providing structural information for pattern identification. A
search of carboxylic acids with 2-aminopyridine fragments
yielded 60 hits and a combination of carboxylic acids with
pyridine produced 644 hits (Scheme 2).16 This shows that
carboxylic acids can bind effectively to either moiety through a
Scheme 3 Ditopic supramolecular reagents for probing selectivity.
…
…
…
…
combination of O–H N/O H–N or O–H N/O H–C syn-
thons, respectively. To establish the relative strength (and, in
effect, supramolecular selectivity) of these two moieties we
decided to carry out systematic co-crystallization studies on a
probe molecule containing both types of acceptor sites with a
series of carboxylic acids. A potential complication with a py/2-
NH2-py competition would result from the fact that most CSD
hits corresponding to 2-aminopyridine and carboxylic acids
resulted in salts (49 of 60) which would make the comparison
flawed. Therefore we needed a backbone containing two sites (in
terms of their relative strengths) that are comparable to the
pyridine-based system.
Our test system of choice is 2-aminopyrazine as it contains the
2-NH2-pyrazine moiety and a second ‘‘open’’ heterocyclic
nitrogen atom on the same backbone. Importantly, this is a
significantly weaker base which should reduce the proclivity for
salt-formation compared to that displayed by 2-NH2-pyridine.
In addition, we also wanted to explore the role that electrostatic
charge has on the efficiency of the co-crystal synthesis; as most
hydrogen bonds are primarily electrostatic in nature, a dimin-
ished charge on a hydrogen-bond acceptor is expected to reduce
the success rate of co-crystal formation. We opted for three test
molecules in this study, Scheme 3, all of which have two main
binding sites, and the charge is reduced on the nitrogen atoms
through subsequent additions of one and two electron with-
drawing substituents, respectively.
Scheme 4 Potential homo- and hetero synthons in salts and co-crystals
of 2-aminopyrazine and a carboxylic acid.
Methodology
Three ditopic supramolecular reagents (Scheme 3, B1–B3) were
combined with thirty carboxylic acids to establish (i) binding
preferences and (ii) supramolecular yield. The bromine sub-
stituents alter the charge on the endo-cyclic nitrogen atoms, N(1)
and N(4), and the electrostatic surface potential for each
compound was obtained using semi-empirical PM3 calculations.
Co-crystallizations were carried out using ‘solvent assisted
grinding’17 (methanol as a solvent18) on a mixture of bases and
carboxylic acids and the solids from all 90 (3 x 30) reactions were
characterized by IR spectroscopy to establish if the result was a
reaction (co-crystallization or salt formation) or a no reaction
(re-crystallization). The outcome is shown in Table 1. If the
result of the reaction was a co-crystallization, based on IR, we
attempted to grow crystals suitable for single-crystal X-ray
diffraction through slow evaporation. A total of 12 crystal
structure determinations were carried out.
A collection of postulated homo- and hetero synthons
resulting from reactions between B1–B3 and a carboxylic acid
is shown in Scheme 4.
Experimental section
All carboxylic acids and 2-aminopyrazine were purchased from
Aldrich and utilized without further purification methods. The
aminopyrazine derivatives B2–B3 were synthesized according to
previously reported methods.14 To determine melting points we
used a Fisher–Johns melting point apparatus. Infrared spectro-
Scheme 2 CSD structural analysis results for synthons under investiga-
tion.
scopy was carried out on a Nicolet 380 FT-IR. 1H NMR and 13
C
5846 | CrystEngComm, 2012, 14, 5845–5853
This journal is ß The Royal Society of Chemistry 2012

View Online
Table 1 Select IR bands in all solids resulting from combinations of A1–A30 and B1–B3
Acids
B1
B2
B3
A1
A2
A3
A4
A5
A6
A7
A8
4-Aminobenzoic acid
4-Hydroxybenzoic acid
3,5-Dihydroxybenzoic acid
2,4-Dimethoxybenzoic acid
3,4-Dihydroxybenzoic acid
2,3-Dihydroxybenzoic acid
2,5-Dihydroxybenzoic acid
2,3-Dimethylbenzoic acid
2,5-Dimethylbenzoic acid
Benzoic acid
4-Nitrobenzoic acid
4-Fluorobenozoic acid
Pentafluorobenzoic acid
3,5-Dinitrobenzoic acid
2,4-Dinitrobenzoic acid
3-Nitrobenzoic acid
2-Chloro-6-fluorobenzoic acid
2,6-Difluorobenzoic acid
3-Fluorobenzoic acid
4-Cyanobenzoic acid
Oxalic acid
Malonic acid
Succinic acid
Glutaric acid
Adipic acid
Pimelic acid
Suberic acid
Azelaic acid
Sebacic acid
Dodecanedioic acid
—
—
1898, 2521
1878, 2607
1865, 2475
1891, 2448
—
1865, 2410
—
—
—
—
—
—
—
1898, 2521
1878, 2607
1865, 2475
1891, 2448
—
—
1885, 2400
—
1818, 2481
—
A9
—
—
—
—
—
—
—
—
A10
A11
A12
A13
A14
A15
A16
A17
A18
A19
A20
A21
A22
A23
A24
A25
A26
A27
A28
A29
A30
1911, 2441
1881, 2409
1891, 2521
2050, 2400
1860, 2687
2018, 2381
1898, 2428
—
2004, 2614
1891, 2475
1879, 2438
2006, 2522
1854, 2455
1920, 2414
1854, 2400
1920, 2520
1877, 2495
1881, 2495
1861, 2483
1887, 2475
1895, 2507
—
1879, 2407
1851, 2355
—
1851, 2481
1865, 2468
—
1871, 2495
—
—
—
—
1851, 2521
1859, 2448
1911, 2349
1918, 2528
—
—
—
1838, 2427
—
—
—
—
—
—
—
—
—
—
1859, 2545
1869, 2413
—
2450, 1850
1834, 2518
—
NMR spectra were recorded using Varian Unity plus 400 MHz
spectrometer in CDCl3.
A4, 3,4-dihydroxybenzoic acid A5, 2,3-dihydroxybenzoic acid
A6, 2,5-dihydroxybenzoic acid A7, 2,3-dimethylbenzoic acid A8,
2,5-dimethylbenzoic acid A9, benzoic acid A10, 4-nitrobenzoic
acid A11, 4-fluorobenozoic acid A12, pentafluorobenzoic acid
A13, 3,5-dinitrobenzoic acid A14, 2,4-dinitrobenzoic acid A15,
3-nitrobenzoic acid A16, 2-chloro-6-fluorobenzoic acid A17, 2,6-
difluorobenzoic acid A18, 3-fluorobenzoic acid A19, 4-cyano-
benzoic acid A20 and ten aliphatic dicarboxylic acids: oxalic acid
A21, malonic acid A22, succinic acid A23, glutaric acid A24,
adipic acid A25, pimelic acid A26, suberic acid A27, azelaic acid
A28, sebacic acid A29 and dodecanedioic acid A30. These acids
were chosen in such a way that we could cover a range of
aliphatic and aromatic carboxylic acids, both weak and strong,
to ensure that the nature of the carboxylic acids does not affect
the binding preference. For co-crystallization, stoichiometric
amounts of bases B1–B3 and acids, either 1 : 1 (monoacids, A1–
A20) or 2 : 1 (diacids, A21–A30), were mixed together.
Synthesis of B2 and B3
N-bromosuccinimide (2.4 g, 13.4 mmol) in 100 ml methylene
chloride solution was added dropwise to 2-aminopyrazine (1.0 g,
10.6 mmol) dissolved in methylene chloride (200 mL) cooled to
0–5 uC. The reaction mixture was stirred at 0 uC for 2 h. Upon
completion, the reaction mixture was quenched with 10% sodium
bicarbonate and 10% sodium sulfite solution followed by
filtration. The precipitate was washed with water. The filtrate
was extracted using methylene chloride and dried over anhy-
drous magnesium sulfate. Excess solvent was removed through
rotary evaporation. Isolation of a residue was carried out
through column chromatography on silica with hexane : ethyl
acetate (10 : 0A4 : 6) mixture as the eluant. 2-amino-5-bromo-
pyrazine was isolated as a yellowish–white powder and 2-amino-
3,5-dibromopyrazine was a white powder and recrystallized from
ethyl acetate : hexane. B2 (2-amino-5-bromopyrazine), (2.70 g,
62%). Mp. 105–107 uC (Lit. Mp. 105–110 uC);19 1H NMR (dH;
200 MHz, CDCl3): 8.09 (s, 1H), 7.78 (s, 1H), 4.64 (br, 2H). B3 (2-
amino-3,5-dibromopyrazine), (0.76 g, 12%). Mp. 111–113 uC
(Lit. Mp. 109–110 uC);20 1H NMR (dH; 200 MHz, CDCl3): 8.05
(s, 1H), 5.05 (br, 2H).
2-Aminopyrazine 4-nitrobenzoic acid (1 : 2), B1?A11. 2-
Aminopyrazine (0.020 g, 0.21 mmol) was dissolved in 1 mL of
methanol. To this solution was added 4-nitrobenzoic acid
(0.035 g, 0.21 mmol) in 3 mL of methanol. The resulting
solution was warmed and allowed to stand for slow evaporation
at room temperature. Prism-shaped crystals were obtained after
5 days. M.p. 200–202 uC.
Synthesis of co-crystals and salts
2-Aminopyrazinium
3,5-dinitrobenzoate,
B1?A14.
2-
B1–B3 were subjected to co-crystallization reactions with thirty
different carboxylic acids. We used twenty aromatic monocar-
boxylic acids: 4-aminobenzoic acid A1, 4-hydroxybenzoic acid
A2, 3,5-dihydroxybenzoic acid A3, 2,4-dimethoxybenzoic acid
Aminopyrazine (0.020 g, 0.21 mmol) was dissolved in 1 mL of
methanol. To this solution was added 3,5-dinitrobenzoic acid
(0.044 g, 0.21 mmol) in 3 mL of methanol. The resulting solution
was warmed and allowed to stand for slow evaporation at room
This journal is ß The Royal Society of Chemistry 2012
CrystEngComm, 2012, 14, 5845–5853 | 5847

View Online
temperature. Block-shaped crystals were obtained after 7 days.
2-Amino-3,5-dibromopyrazine 2,5-dihydroxybenzoic acid (1 : 1),
B3?A7. 2-Amino-3,5-dibromopyrazine (0.020 g, 0.08 mmol) was
dissolved in 1 mL of methanol. To this solution was added 2,
5-dihydroxybenzoic acid (0.012 g, 0.08 mmol) in 3 mL of
methanol. The resulting solution was warmed and allowed to
stand for slow evaporation at room temperature. Prism-shaped
crystals were obtained after 7 days. M.p. 180–182 uC.
M.p. 158–160 uC.
2-Aminopyrazine 3-nitrobenzoic acid (1 : 1), B1?A16. 2-Amino-
pyrazine (0.020 g, 0.21 mmol) was dissolved in 1 mL of
methanol. To this solution was added 3-nitrobenzoic acid
(0.035 g, 0.21 mmol) in 3 mL of methanol. The resulting
solution was warmed and allowed to stand for slow evaporation
at room temperature. Block-shaped crystals were obtained after
5 days. M.p. 134–136 uC.
2-Amino-3,5-dibromopyrazine pentafluorobenzoic acid (1 : 1),
B3?A13. 2-Amino-3,5-dibromopyrazine (0.020 g, 0.08 mmol) was
dissolved in 1 mL of ethanol. To this solution was added
pentafluorobenzoic acid (0.016 g, 0.08 mmol) in 3 mL of ethanol.
The resulting solution was warmed and allowed to stand for slow
evaporation at room temperature. Block-shaped crystals were
obtained after 4 days. M.p. 98–100 uC.
2-Aminopyrazine succinic acid (1 : 1), B1?A23. 2-Aminopy-
razine (0.020 g, 0.21 mmol) was dissolved in 1 mL of methanol. To
this solution was added succinic acid (0.006 g, 0.11 mmol) in 3 mL
of methanol. The resulting solution was warmed and allowed to
stand for slow evaporation at room temperature. Plate-shaped
crystals were obtained after 7 days. M.p. 140–142 uC.
2-Amino-3,5-dibromopyrazine 3,5-dinitrobenzoic acid (1 : 1),
B3?A14. 2-Amino-3,5-dibromopyrazine (0.020 g, 0.08 mmol)
was dissolved in 1 mL of ethanol. To this solution was added 3,5-
dinitrobenzoic acid (0.016 g, 0.08 mmol) in 3 mL of ethanol. The
resulting solution was warmed and allowed to stand for slow
evaporation at room temperature. Plate-shaped crystals were
obtained after 7 days. M.p. 128–130 uC.
2-Aminopyrazine glutaric acid (1 : 1), B1?A24. 2-Aminopy-
razine (0.020 g, 0.21 mmol) was dissolved in 1 mL of methanol.
To this solution was added glutaric acid (0.013 g, 0.11 mmol) in
3 mL of methanol. The resulting solution was warmed and
allowed to stand for slow evaporation at room temperature. Plate-
shaped crystals were obtained after 4 days. M.p. 126–128 uC.
2-Amino-3,5-dibromopyrazine 2,6-difluorobenzoic acid (1 : 1),
B3?A18. 2-Amino-3,5-dibromopyrazine (0.020 g, 0.08 mmol) was
dissolved in 1 mL of ethanol. To this solution was added 2,6-
difluorobenzoic acid (0.012 g, 0.08 mmol) in 3 mL of ethanol.
The resulting solution was warmed and allowed to stand for slow
evaporation at room temperature. Block-shaped crystals were
obtained after 3 days. M.p. 144–146 uC.
2-Aminopyrazine pimelic acid (1 : 1), B1?A26. 2-Aminopy-
razine (0.020 g, 0.21 mmol) was dissolved in 1 mL of methanol.
To this solution was added pimelic acid (0.016 g, 0.11 mmol) in
3 mL of methanol. The resulting solution was warmed and
allowed to stand for slow evaporation at room temperature. Plate-
shaped crystals were obtained after 3 days. M.p. 108–110 uC.
2-Aminopyrazine suberic acid (1 : 1), B1?A27. 2-Aminopy-
razine (0.020 g, 0.21 mmol) was dissolved in 1 mL of methanol.
To this solution was added suberic acid (0.018 g, 0.11 mmol) in
3 mL of methanol. The resulting solution was warmed and
allowed to stand for slow evaporation at room temperature. Plate-
shaped crystals were obtained after 5 days. M.p. 210–212 uC.
Electrostatic charges
To determine the electrostatic charges on the hydrogen-bond
donors and acceptors of the molecules (B1–B3) we first used
semiempirical PM3 calculations to optimize molecular geome-
tries and then probed the MEPS (0.002 e au21 isosurface, from
PM3 calculations) with a point charge. Calculated charges are
represented in kJ mol21 (Table 2).
2-Aminopyrazine sebacic acid (1 : 1), B1?A29. 2-Aminopy-
razine (0.020 g, 0.21 mmol) was dissolved in 1 mL of methanol.
To this solution was added sebacic acid (0.021 g, 0.11 mmol) in
3 mL of methanol. The resulting solution was warmed and
allowed to stand for slow evaporation at room temperature. Plate-
shaped crystals were obtained after 4 days. M.p. 124–126 uC.
X-Ray crystallography
Data sets were collected on Bruker Kappa APEX II system or a
SMART APEX II system, at 120 K using APEX2 software. An
Oxford Cryostream 700 low-temperature device was used to
Table 2 Charge calculations (PM3) and outcome summary of co-crystallization experiments between pyrazines and carboxylic acids
B1, 2-aminopyrazine B2, 2-amino-5-bromopyrazine B3, 2-amino-3,5-dibromopyrazine
Charges kJ mol21
Reaction
N(1) 2251
N(4) 2255
N(1) 2238
N(4) 2245
N(1) 2222
N(4) 2240
25
5
12
18
40
6
24
20
No reaction
% yield
83
5848 | CrystEngComm, 2012, 14, 5845–5853
This journal is ß The Royal Society of Chemistry 2012

View Online
This journal is ß The Royal Society of Chemistry 2012
CrystEngComm, 2012, 14, 5845–5853 | 5849

View Online
control temperature. Mo-Ka radiation was used. Initial cell
constants were found by small widely separated ‘‘matrix’’ runs.
Data collection strategies were determined using COSMO. Scan
speeds and scan widths were chosen based on scattering power
and peak rocking curves. Crystallographic data for all twelve
compounds are summarized in Table 3.
When B1 was combined with 3,5-dinitrobenzoic acid A14, a
1 : 1 salt was formed (Fig. 3). The primary intermolecular
2
interaction is the charge-assisted two-point N–H O /N –
+
…
2
…
H
O
synthon (Scheme 4, Synthon IV). In addition, the
nearest neighbor of the anti-proton of the amino group, is an
oxygen atom of a nitro group.
The crystal structure determination of B1?A16 shows 1 : 2
stoichiometry and the expected hydrogen-bond interactions
primarily responsible for the supramolecular assembly
Results
…
…
The distinction between reaction (salt/co-crystal) and no reaction
was made based on the presence/absence of broad stretches near
(Fig. 4a), O–H N and CLO H–N (Synthon III and V).
Similarly, the B1?A11 structure showed co-crystal formation
(Fig. 4b) via previously observed synthon III and V (Scheme 4).
Due to the disorder in B1?A11 and B1?A16 it was not possible to
establish the secondary interaction of the anti-proton of the
amino group.
1850 and 2500 cm21 which are indicative of intermolecular O–
…
H
N(heterocycle) hydrogen bonds and which will only appear
in salts and co-crystals, and not in physical mixtures of the two
reactants (Fig. 1 and Fig. 2). Based on IR spectroscopy we
determined that B1, the strongest base of the three, showed an
83% supramolecular yield followed by 40% and 20% yields for
base B2 and B3, respectively.
In the next set, we changed the aromatic acids to aliphatic
dicarboxylic acids to test whether changing the nature of the
carboxylic acids affects the synthon selectivity. In the structure of
B1?A23, synthon I (Scheme 4) prevailed leaving the carboxylic
acid only one option for heteromeric interaction i.e. the N(4)
binding site of B1. The heteromeric synthon observed in co-
The results from IR spectroscopy were further supported and
complemented by subsequent single crystal X-ray diffraction
analysis. In this particular case, where our key goal is to map out
the structural landscape of 2-aminopyrazine derivatives, struc-
tural analysis provided useful insight. We obtained twelve crystal
structures. The relevant X-ray data are summarized in Tables
S3–S4 (in ESI).{
…
crystal B1?A23 was O–H N(4) (Synthon V). To be noted, the
…
potential self-complementary C–H O hydrogen bond from the
C–H of B1 to the carbonyl of A23 does not form even though the
…
˚
two molecules are coplanar. The C–H O distance is 2.875 A,
which is well beyond the van der Waals distance for a hydrogen
˚
and oxygen contact, 2.72 A. The combination of synthon I and
heterosynthon V give rise to infinite 1D chains, moreover, these
…
chains are extended into 2D sheets via amino anti-NH OLC
secondary interactions (Fig. 5).
When B1 and glutaric acid A24 were combined, the outcome
was 1 : 1 co-crystal, with the acid binding to both the sites of B1.
Fig. 1 IR spectrum of attempted co-crystallization of 3,5-dibromo-2-
aminopyrazine B3 and 2,4-dimethoxybenzoic acid (no reaction).
Fig. 3 The primary synthon in the crystal structure of B1?A14.
Fig. 2 IR spectrum of successful co-crystallization of 3,5-dibromo-2-
aminopyrazine B3 and pentafluorobenzoic acid.
Fig. 4 Thermal ellipsoid plots (50% probability level) of (a) B1?A11 and
(b) B1?A16.
5850 | CrystEngComm, 2012, 14, 5845–5853
This journal is ß The Royal Society of Chemistry 2012

View Online
Fig. 7 Extended 2D network formed in the co-crystal of B1?A26.
Fig. 5 2D network formed in co-crystal of B1?A23.
Synthons III and V (Fig. 6) persisted even when different
aliphatic dicarboxylic acids were used (B1?A26, B1?A27,
B1?A29).21 In addition, in all four structures, the anti NH
proton interacts with a carbonyl oxygen atom, giving rise to a 2D
extended network (Fig. 7).
Fig. 8 1 : 1 co-crystal of B3?A13.
When B3 was combined with pentafluorobenzoic acid A13, it
yielded a 1 : 1 co-crystal with the desired two-point hydrogen
bonding synthon I, Scheme 4 (Fig. 8); the same primary synthon
was found in the structures of B1?A7 and B1?A18. However, B3
with 3,5-dinitrobenzoic acid (A14) shows an unexpected O–
be due to a prevailing acid–acid dimer even in the presence of
N-heterocylic bases.22
Synthon classification. In B1?A11/A16 the carboxylic acid
binds to both the ends of B1, and in B1?A14, where proton
transfer from acid to B1 occurs, it opts for the predicted 2-amino
end of B1.
… …
N/O H–C synthon VII, Scheme 5 (Fig. 9). In co-crystals
H
B3?A13 and B3?A18, the anti-proton of the amino group
hydrogen bonds to an oxygen atom of a carbonyl group. In
the case of B3?A14, the anti-proton hydrogen bonds to an
oxygen atom of a nitro group of A14, whereas in B3?A7 the anti
NH proton is structurally inactive.
The next set of crystal structures we examined were from a
combination of B1 and aliphatic dicarboxylic acids. The
observed synthon I (B1 dimer) in the case of B1?A23 does not
persist upon use of longer chain aliphatic dicarboxylic acids
(B1?A26, B1?A26, B1?A27, B1?A29). This could be due to
prevailing heteromeric interactions over observed homomeric
interactions. Also, except B1?A23, all remaining four co-crystals
from the combination of aliphatic dicarboxylic acids and B1,
shows remarkable synthon consistency (Table 4). However,
selective binding of carboxylic acids at the expected N1 site of B1
could not be achieved presumably due to the presence of the
competitive binding site N4. Although the stronger two-point
interaction should prevail over a single-point interaction, but
more often for synthon selectivity to occur, geometric and
electrostatic factors need to be chosen meticulously.23
Discussion
Controlling supramolecular yield
Based on IR spectroscopy the strongest base of the three, B1
showed an 83% supramolecular yield followed by 40% and 20%
for B2 and B3, respectively. These results can be explained in
terms of reduced electrostatic charges on the plausible binding
sites of 2-aminopyrazine derivatives due to addition of electron
withdrawing substituents, Table 2.
Our systematic analysis also clarifies that heteromeric product
formation in this system is primarily determined by the charge on
the hydrogen-bond acceptor sites of B1–B3 and not by the
strength of the carboxylic acid (Table S2, ESI{). We analyzed
the results in the context of charges of the carboxylic acids (the
charge on the acidic proton i.e. COOH); however, a better
correlation was found between the charges on the heterocyclic
nitrogen atom and the supramolecular yield. For example, a
stronger acid, 2,5-dihydroxybenzoic acid A7, gave ‘no reaction’
with B1 and B2, but a ‘reaction’ with the weakest base B3. This
inconsistency of results based on carboxylic acid strengths could
B3 the weakest of the three bases, yielded four co-crystals
B3?A7, B3?A13, B3?A14, B3?A18. Three out of four times
synthon III was observed and in one exception (B3?A14) it
showed a never observed synthon VII.24
Binding preferences. We investigated the binding preference of
carboxylic acids with 2-aminopyrazine derivatives, based on the
crystal structures obtained. The structural analysis reveals that
ten out of twelve times the carboxylic acids bind, as expected, to
the 2-amino end of B1–B3. However, in seven out of twelve cases
the carboxylic acids bind to N(4) of B1–B3.
We performed the CSD analysis for carboxylic acid binding
preference when both 2-aminopyridine and pyridine sites are
available. The results show that 13 out of 14 times the acid binds
to 2-aminopyridine end with one exception where 2-aminopyr-
idine forms a homodimer with the acid binding to a single-point
pyridyl nitrogen (Table S4, ESI{). The deviation of the observed
results in the 2-aminopyrazine case could be rationalized by
examining comparable charges on the two possible binding sites
Fig. 6 Infinite 1D chains formed in the co-crystal of B1?A24.
This journal is ß The Royal Society of Chemistry 2012
CrystEngComm, 2012, 14, 5845–5853 | 5851

View Online
Scheme 5 The structural landscape inhabited by co-crystals of 2-aminopyrazine and carboxylic acids.
(N(1) and N(4) nitrogen atoms) thereby giving more alternatives
to carboxylic acids for binding.
Conclusions
Our experimental results successfully establish a relationship
between the electrostatic charge on the N-heterocyclic base and
the ability of the compound to form intermolecular hydrogen
bonds in the solid state. By adding appropriate substituents via
conventional covalent synthesis, we have been able to modulate
the charge on the hydrogen-bond acceptor and this leads to a
predictable lowering of the supramolecular yield of the reaction.
The main driving force for assembly of a heteromer between
Fig. 9 The primary hydrogen bonds in the crystal structure of B3?A14.
Table 4 Summary of observed synthons in crystal structures
Synthons
…
…
bases and carboxylic acids was found to be the O–H N/O H–
N synthon. However, synthon crossover was observed in the
structural landscape of 2-aminopyrazine derivatives combined
with carboxylic acids and was unavoidable due to competitive
binding sites present in the N-heterocyclic bases chosen. We are
currently investigating the role of stoichiometry in controlling
the binding ability of carboxylic acids with 2-aminopyrazine
derivatives.
Crystal Structure
I
II
III
3
IV
V
VI
VII
VIII
B1?A11
B1?A14
B1?A16
B1?A23
B1?A24
B1?A26
B1?A27
B1?A29
B3?A7
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
Notes and references
B3?A13
B3?A14
B3?A18
3
1 J.-M. Lehn, Supramolecular Chemistry: Concepts and Perspectives,
VCH, Weinheim, 1995.
3
5852 | CrystEngComm, 2012, 14, 5845–5853
This journal is ß The Royal Society of Chemistry 2012

View Online
2 G. R. Desiraju, Angew. Chem., Int. Ed. Engl., 1995, 34, 2311.
3 G. R. Desiraju, Crystal engineering: The design of organic solids,
Elsevier, Amsterdam, 1989.
4 (a) M. C. Etter, J. Phys. Chem., 1991, 95, 4601; (b) M. C. Etter and S.
M. Reutzel, J. Am. Chem. Soc., 1991, 113, 2586; (c) M. C. Etter, Z.
Urbanczyk-Lipkowska, M. Zia-Ebrahimi and T. W. Panunto, J. Am.
Chem. Soc., 1990, 112, 8415; (d) M. C. Etter, Acc. Chem. Res., 1990,
23, 120.
5 (a) C. B. Aakero¨y, A. Beatty, M. Nieuwenhuyzen and J. Desper,
Tetrahedron, 2000, 56, 6693; (b) C. B. Aakero¨y, A. M. Beatty and
B. A. Helfrich, J. Am. Chem. Soc., 2002, 124, 14425; (c) C. B.
Aakero¨y, J. Desper and M. E. Fasulo, CrystEngComm, 2006, 8, 586;
(d) C. B. Aakero¨y, J. Desper, D. J. Salmon and M. M. Smith, Cryst.
Growth Des., 2006, 6, 1033–1042; (e) C. B. Aakero¨y, I. Hussain and J.
Desper, Cryst. Growth Des., 2006, 6, 474–480.
6 (a) C. B. Aakero¨y, S. Forbes and J. Desper, J. Am. Chem. Soc., 2009,
131, 17048; (b) D. P. McNamara, S. L. Childs, J. Giordano, A. Iarriccio,
J. Cassidy, M. S. Shet, R. Mannion, E. O’Donnell and A. Park, Pharm.
Res., 2006, 23, 1888; (c) P. Vishweshwar, J. A. McMahon, J. A. Bis and
M. J. Zaworotko, J. Pharm. Sci., 2006, 95, 499.
35, 565; (e) B. Moulton and M. J. Zaworotko, Chem. Rev., 2001, 101,
1629; (f) G. R. Desiraju, Angew. Chem., Int. Ed. Engl., 1995, 34, 2311;
(g) M. Wenger and J. Bernstein, Angew. Chem., Int. Ed., 2006, 45,
7966; (h) S. L. Childs and K. I. Hardcastle, CrystEngComm, 2007, 9,
364; (i) E. Bosch, CrystEngComm, 2007, 9, 191; (j) V. R. Pedireddi, S.
Chatterjee, A. Ranganathan and C. N. R. Rao, J. Am. Chem. Soc.,
1997, 119, 10867; (k) C. B. Aakero¨y, J. Desper and J. F. Urbina,
CrystEngComm, 2005, 7, 193; (l) C. B. Aakero¨y, N. Schultheiss and J.
Desper, J. Mol. Struct., 2010, 972, 35; (m) C. B. Aakero¨y, J. Desper
and B. M. T. Scott, Chem. Commun., 2006, 1445.
12 (a) J. A. Bis and M. J. Zaworotko, Cryst. Growth Des., 2005, 5, 1169;
(b) K. Merz and V. Vasylyeva, CrystEngComm, 2010, 12, 3989; (c)
G. R. Desiraju, J. Mol. Struct., 2003, 656, 5; (d) N. R. Goud, N. J.
Babu and A. Nangia, Cryst. Growth Des., 2011, 11, 1930.
13 (a) B. Sarma, N. K. Nath, B. R. Bhogala and A. Nangia, Cryst.
Growth Des., 2009, 9, 1546; (b) J. A. Bis, P. Vishweshwar, D. Weyna
and M. J. Zaworotko, Mol. Pharmaceutics, 2007, 4, 401; (c) C. B.
Aakero¨y, A. M. Beatty and B. A. Helfrich, Angew. Chem., Int. Ed.,
2001, 40, 3240; (d) E. Corradi, S. V. Meille, M. T. Messina, P.
Metrangolo and G. Resnati, Angew. Chem., Int. Ed., 2000, 39, 1782.
14 (a) C. B. Aakero¨y, P. D. Chopade, C. Ganser and J. Desper, Chem.
Commun., 2011, 47, 4688; (b) C. B. Aakero¨y, P. D. Chopade and J.
Desper, Cryst. Growth Des., 2011, 11, 5333.
7 N. Schultheiss, M. Roe and S. X. M. Boerrigter, CrystEngComm,
2011, 13, 611.
8 J. D. Dunitz, in Perspectives in Supramolecular Chemistry: The
Crystal as a Supramolecular Entity, ed. G.R. Desiraju, Wiley,
Amsterdam, 1995.
15 F. A. Allen, Acta Crystallogr., Sect. B: Struct. Sci., 2002, 58, 380.
16 CSD search carried out on ConQuest Version 1.12. (Updated till
November 2011).
9 (a) N. Barooah, R. J. Sarma and J. B. Baruah, CrystEngComm, 2006,
8, 608; (b) N. Motohiro, CrystEngComm, 2004, 6, 130.
ˇ ´
17 S. L. James, C. J. Adams, C. Bolm, D. Braga, P. Collier, T. Frisˇcic, F.
10 (a) P. Metrangolo, H. Neukirch, T. Pilati and G. Resnati, Acc. Chem.
Res., 2005, 38, 386; (b) P. Metrangolo, T. Pilati, G. Resnati and A.
Stevenazzi, Chem. Commun., 2004, 1492; (c) A. De Santis, A. Forni,
R. Liantonio, P. Metrangolo, T. Pilati and G. Resnati, Chem.–Eur.
J., 2003, 9, 3974; (d) R. B. Walsh, C. W. Padgett, P. Metrangolo, G.
Resnati, T. W. Hanks and W. T. Pennington, Cryst. Growth Des.,
Grepioni, K. D. M. Harris, G. Hyett, W. Jones, A. Krebs, J. Mack,
L. Maini, A. G. Orpen, I. P. Parkin, W. C. Shearhouse, J. W. Steed
and D. C. Waddell, Chem. Soc. Rev., 2012, 41, 413.
18 We attempted ‘solvent assisted grinding’ using different solvents and
melt experiments for co-crystal screening. However, while using
different solvents (e.g. dichloromethane, chloroform) for grinding,
solubility problem of acids/bases encountered and when melted, the
compounds decomposed.
ˇ ´
2001, 1, 165; (e) D. Cincic, T. Frisˇcic and W. Jones, J. Am. Chem.
ˇ ´
Soc., 2008, 130, 7524; (f) D. Cincic, T. Frisˇcic and W. Jones, Chem.–
Eur. J., 2008, 14, 747; (g) T. Shirman, D. Freeman, Y. D. Posner, I.
Feldman, A. Facchetti and M. E. van der Boom, J. Am. Chem. Soc.,
2008, 130, 8162; (h) H. L. Nguyen, P. N. Horton, M. B. Hursthouse,
A. C. Legon and D. W. Bruce, J. Am. Chem. Soc., 2004, 126, 1617; (i)
S. Triguero, R. Llusar, V. Polo and M. Fourmigue´, Cryst. Growth
Des., 2008, 8, 2241; (j) E. A. Meyer, R. K. Castellano and F.
Diederich, Angew. Chem., Int. Ed., 2003, 42, 1210; (k) G. M.
Espallargas, L. Brammer and P. Sherwood, Angew. Chem., Int. Ed.,
2006, 45, 435.
19 R. C. Ellingson and R. L. Henry, J. Am. Chem. Soc., 1949, 71, 2798.
20 W. W. Paudler and M. V. Jovanovic, J. Org. Chem., 1983, 48, 1064.
21 Crystal structure B1?A29 has amino group disordered in 50 : 50 ratio.
22 (a) P. L. Wash, E. Maverick, J. Chiefari and D. A. Lightner, J. Am.
Chem. Soc., 1997, 119, 3802; (b) C. B. Aakero¨y, A. M. Beatty and
K. R. Lorimer, J. Chem. Soc., Dalton Trans., 2000, 3869.
23 (a) C. B. Aakero¨y, J. Desper, B. A. Helfrich, P. Metrangolo, T. Pilati,
G. Resnati and A. Stevenazzi, Chem. Commun., 2007, 4236; (b) C. B.
Aakero¨y, N. Schultheiss, A. Rajbanshi, J. Desper and C. Moore,
Cryst. Growth Des., 2009, 9, 432.
11 (a) C. B. Aakero¨y and D. J. Salmon, CrystEngComm, 2005, 7, 439;
(b) L. MacGillivray, CrystEngComm, 2004, 6, 77; (c) J.-M. Lehn,
Science, 2002, 295, 2400; (d) G. R. Desiraju, Acc. Chem. Res., 2002,
24 CSD search shows no hits on this type of synthon in presence of
2-aminopyridine based heterocycles.
This journal is ß The Royal Society of Chemistry 2012
CrystEngComm, 2012, 14, 5845–5853 | 5853