Non-coplanar aromatic carboxylic acids: Unusual conformation-dependent self-assembly and pseudopolymorphism of di(3-carboxymesityl)methane

Article abstract of DOI:10.1016/j.molstruc.2007.10.019

The dicarboxylic acids 3 and 4, i.e., di(3-carboxy-2,4,6-trimethylphenyl)methane and di(3-carboxyphenyl)methane, are created by a methylene tethering of mesitoic acid and benzoic acid, respectively. These diacids may explore two low energy conformations, viz., syn and anti. Whereas the syn-diacid 3 is found to undergo self-assembly in the solid state via a very rare tetrameric motif, the anti-diacid 3 is found to exhibit pseudopolymorphism with guest molecules such as DMSO and PhOH-H₂O. The unusual patterns of assembly and the occurrence of pseudopolymorphism for the syn and anti conformers, respectively, appear to emanate from an unique structural feature that emerges as a consequence of tethering. It appears that the molecules that contain strongly interacting functional groups in non-coplanar aryl rings will suffer from packing problems, which manifests itself in a new mode of packing and the phenomenon of pseudopolymorphism.

Full text of DOI:10.1016/j.molstruc.2007.10.019

Available online at www.sciencedirect.com
Journal of Molecular Structure 885 (2008) 139–148
www.elsevier.com/locate/molstruc
Non-coplanar aromatic carboxylic acids: Unusual
conformation-dependent self-assembly and pseudopolymorphism
of di(3-carboxymesityl)methane
Jarugu Narasimha Moorthy *, Palani Natarajan
Department of Chemistry, Indian Institute of Technology, Kanpur 208 016, India
Received 3 September 2007; received in revised form 13 October 2007; accepted 18 October 2007
Available online 26 October 2007
Abstract
The dicarboxylic acids 3 and 4, i.e., di(3-carboxy-2,4,6-trimethylphenyl)methane and di(3-carboxyphenyl)methane, are created by a
methylene tethering of mesitoic acid and benzoic acid, respectively. These diacids may explore two low energy conformations, viz., syn
and anti. Whereas the syn-diacid 3 is found to undergo self-assembly in the solid state via a very rare tetrameric motif, the anti-diacid 3 is
found to exhibit pseudopolymorphism with guest molecules such as DMSO and PhOH–H2O. The unusual patterns of assembly and the
occurrence of pseudopolymorphism for the syn and anti conformers, respectively, appear to emanate from an unique structural feature
that emerges as a consequence of tethering. It appears that the molecules that contain strongly interacting functional groups in non-
coplanar aryl rings will suffer from packing problems, which manifests itself in a new mode of packing and the phenomenon of
pseudopolymorphism.
Ó 2007 Elsevier B.V. All rights reserved.
Keywords: Pseudopolymorphism; Solid state; X-ray crystallography; Carboxylic acids; Self-assembly
1. Introduction
organization based on supposedly robust supramolecular
synthons.
The ultimate goal of crystal engineering [1,2], which is
concerned with the understanding of intermolecular inter-
actions in the context of crystal packing and exploiting
such an understanding to design new solids with desired
physical and chemical properties, is to be able to predict
crystal packing and synthesize crystals with predetermined
molecular ordering. Although the supramolecular synthon
approach involving identification of crystals as retrosyn-
thetic targets [2] has led to a tremendous advancement in
predictively engineering crystal structures, subtle variation
of structural attributes thwart the expected crystal packing
based on supposedly robust supramolecular synthons [3].
Thus, there is a need to understand how, when and why
the crystal packing may take a departure from the expected
One of the functional groups that have been extensively
explored in the realm of supramolecular chemistry is the
carboxylic acid. The COOH group may, in principle, exist
in either of the two geometries, viz., syn and anti. Leisero-
witz and co-workers [4] comprehensively analyzed various
patterns in which the COOH groups assemble in the crys-
tals. Accordingly, two primary patterns in which the acids
have been found to self-assemble are the centrosymmetric
dimer motif and the polymeric catemer motif (Fig. 1).
The Cambridge Structural Database (CSD) has been ana-
lyzed from time-to-time to gauge the frequency of occur-
rence of patterns in which the carboxylic acids are found
to associate [5]. In general, the dimer motif has been found
to occur in >90% of the acid structures in which other func-
tional groups that potentially compete for hydrogen bond-
ing are absent. The catemer and other patterns that include
dimer motifs expanded by solvents, tetramers, a hexamer
motif and a helical motif are observed in the remaining
*
Corresponding author. Tel.: +91 512 2597438; fax: +91 512 2597436.
E-mail address: moorthy@iitk.ac.in (J.N. Moorthy).
0022-2860/$ - see front matter Ó 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.molstruc.2007.10.019

140
J.N. Moorthy, P. Natarajan / Journal of Molecular Structure 885 (2008) 139–148
R
O
R
H
O
H
O
O
O
O
R
H
O
O
R
O
O
O
O
H
H
H
H
O
H
anti
R
R
O
syn
O
R
H
H
O
O
O
O
O
R
R
R
O
O
H
O
H
H
O
O
R
R
O
R
H
O
O
R
H
H
O
O
O
Dimeric motif: syn-syn
O
R
R
Catemeric motifs: syn-syn
H
H
O
O
O
H
R
R
O
H
O
R
O
R
R
O
O
R
O
O
O
H
H
O
H
H
H
O
O
O
O
R
R
O
R
O
Catemeric motif: syn-anti
Fig. 1. Various motifs that are possible for the self-assembly of carboxylic acids.
Catemeric motifs: anti-anti
10% of the structures [5d]. Given that the carboxylic acid
serves as one hydrogen bond donor and one hydrogen
bond acceptor in both dimeric as well as catemeric motifs,
what indeed causes the acids to crystallize with such a bias
in favor of the centrosymmetric dimer motif predominantly
is intriguing. Understanding of factors that promote adop-
tion of either dimeric or catemeric motif is important in the
utilization of carboxyl group as a supramolecular synthon
reliably in crystal engineering. Recently, Desiraju et al.
showed that ancillary C–Hꢀ ꢀ ꢀO hydrogen bonds support
prevalence of the rare syn–anti catemeric motif, cf. Fig. 1,
in a family of phenylpropiolic acids [5c] and cubanecarb-
oxylic acids [6].
From a recent CSD analyses of solvates, Jacco van de
Streek [7] pointed out an observation that the molecules
that contain strongly interacting functional groups in
non-coplanar aromatic rings, e.g., 1,10-binaphthyl-2,20-
dicarboxylic acid 1 (Chart 1), exhibit more than average
tendency to form solvates. We were delighted to recognize
that our own discovery of inclusion of a hexameric planar
Me
COOH
Me
HOOC
Me
Me
Me
Me
Me
HOOC
COOH
COOH
Me
Me
Me
Me
Me
COOH
syn
2
1
Me
Me
HOOC
Me
Me
Me
Me
Me
Me Me
Me
COOH
HOOC
COOH
COOH
COOH
Me
Me
3
4
anti
Chart 1. Structures of the diacids 3 and 4, and their conformations.

J.N. Moorthy, P. Natarajan / Journal of Molecular Structure 885 (2008) 139–148
141
water network in the crystal lattice of bimesityl-3,30-dicar-
boxylic acid 2, which we reported some time ago [8], falls
in line with this observation. This prompted us to examine
the crystal structures of sterically hindered as well as unhin-
dered benzoic acids, viz., bis(3-carboxy-2,4,6-trimethyl-
phenyl)methane 3 and di(3-carboxyphenyl)methane 4, as
the two carboxyaryl rings would be related in these cases
by a non-coplanar tetrahedral angle. In addition to explor-
ing the occurrence of solvatomorphism/pseudopolymor-
phism [9] – crystallization of a given compound with
solvents – in these cases, we were also motivated to exam-
ine:
solvent was removed in vacuo. The pure product was iso-
lated as a colorless solid by silica-gel column chromatogra-
phy using petroleum ether as an eluent, yield 3.3 g (67%);
1
mp 242–244 °C; H NMR (CDCl3, 400 MHz) d 2.14 (s,
6H), 2.28 (s, 6H), 2.49 (s, 6H), 4.09 (s, 2H), 6.86 (s, 2H);
13C NMR (CDCl3, 100 MHz) d 16.28, 20.14, 21.62,
30.83, 132.19, 136.95, 138.21, 139.16, 142.10.
2.3. Preparation of bis(3-cyano-2,4,6-trimethylphenyl)
methane
To a solution of 2.6 g (28.0 mmol) of CuCN in 20 mL of
DMF was added 3.0 g (7.3 mmol) of bis(3-bromo-2,4,6-tri-
methylphenyl)methane and the resultant mixture was heated
at reflux (150–160 °C) for 6 h. Subsequently, the reaction
mixture was cooled to room temperature and a solution of
FeCl3 (4.2 g in 5.0 mL of 2.4 N HCl) was added. The mixture
was heated at 70–80 °C for 2 h and cooled. The organic mat-
ter was extracted with chloroform, washed with water, dried
over anhyd Na2SO4 and the solvent removed in vacuo. The
pure product was isolated asa colorless solid by silica-gel col-
umn chromatography using CHCl3 and petroleum ether
mixture (20:80) as an eluent, yield 1.7 g (70%); IR (KBr)
ꢁ How the tethering with a methylene spacer influences
the self-assembly via dimer motif, which is predomi-
nantly observed in simple benzoic acids as well as ste-
rically-hindered acids such 2,6-dimethylbenzoic acid
(Refcode: DMBNZA10), mesitoic acid (TMBZAC),
pentamethylbenzoic acid (TUSOIH), etc.
ꢁ The possibility of conformation-dependent crystal
packing due to two low-energy syn and anti
conformations that the molecules may exist in
(Chart 1).
1
cmꢂ1 2241, 3083; H NMR (CDCl3, 400 MHz) d 2.10 (s,
Herein, we report a rare tetrameric assembly of COOH
groups in the crystal lattice of solvent-free diacid 3 and its
conformation-dependent guest inclusion behavior, i.e.,
pseudopolymorphism, with DMSO and PhOH–H2O.
6H), 2.28 (s, 6H), 2.45 (s, 6H), 4.03 (s, 2H), 6.91 (s, 2H);
13C NMR (CDCl3, 100 MHz) d 19.07, 20.45, 21.35, 31.33,
112.55, 117.86, 130.49, 135.38, 139.94, 140.23, 141.67.
2.4. Synthesis of bis(3-carbamoyl-2,4,6-trimethylphenyl)
methane
2. Experimental
2.1. General aspects
A solution of 1.5 g (5.21 mmol) of bis(3-cyano-2,4,6-tri-
methylphenyl)methane in ca. 5.0 mL of AcOH–H2SO4
(2:1, v/v) was heated at reflux. The progress of the reaction
was monitored by TLC analysis. After completion of the
reaction (16 h), the reaction mixture was poured into ice-cold
water. The product that precipitated out was filtered to
obtain bis(3-carbamoyl-2,4,6-trimethylphenyl)methane as
a colorless powder, yield 1.4 g (83%), mp >280 °C; IR
(KBr) cmꢂ1 1647, 3156, 3343; 1H NMR (DMSO-d6,
400 MHz) d 2.00 (s, 6H), 2.03 (s, 6H), 2.16 (s, 6H) 3.93 (s,
2H), 6.82 (s, 2H), 7.38 (s, 2H), 7.57 (s, 2H); 13C NMR
(DMSO-d6,100 MHz) d 16.91, 18.74, 20.58, 30.99, 129.57,
130.0, 131.81, 134.84, 135.74, 137.86, 171.98.
Anhydrous tetrahydrofuran (THF) was freshly distilled
over sodium prior to use. All other solvents were distilled
prior to use. Column chromatography was conducted with
silica gel (Acme, Mumbai, 60–120 lm mesh). H and 13C
1
NMR spectra were recorded on
a JEOL-Lambda
(400 MHz) spectrometer using CDCl3 as a solvent. IR
spectra were recorded on a Bruker Vector 22 FT-IR spec-
trophotometer. The melting points were determined with a
Perfit Melting point apparatus (India) and are uncorrected.
The commercial chemicals were used as received.
The diacid 4, i.e., bis(3-carboxyphenyl)methane, was
prepared by following the reported procedure [10].
2.5. Synthesis of bis(3-carboxy-2,4,6-trimethylphenyl)
methane
2.2. Synthesis of bis(3-bromo-2,4,6-trimethylphenyl)methane
[11]
To a suspension of 1.1 g (3.25 mmol) of bis(3-carbam-
oyl-2,4,6-trimethylphenyl)methane in 10 mL of 60%
H2SO4 was added 5 mL of TFA with vigorous stirring
and heated at 80 °C for ca. 3 h. Subsequently, the solution
was concentrated and the organic material was extracted
with ethyl acetate. The combined extract was evaporated.
The pure product was isolated as a colorless solid by sil-
ica-gel column chromatography using ethyl acetate and
petroleum ether mixture (50:50) as an eluent; yield 0.9 g
To a solution of 8.5 g (48.3 mmol) of NBS and 3.0 g
(12 mmol) dimesitylmethane [12] in 50 mL of CH3CN
was added a catalytic amount of TFA with vigorous stir-
ring. The reaction mixture was allowed to stir for 8 h at
room temperature. Subsequently, the solution was concen-
trated and the organic material was extracted with CHCl3.
The combined extract was washed with 10% NaOH solu-
tion followed by water, dried over anhyd Na2SO4 and the

142
J.N. Moorthy, P. Natarajan / Journal of Molecular Structure 885 (2008) 139–148
1
(81%); IR (KBr) cmꢂ11691, 3413; H NMR (DMSO-d6,
400 MHz) d 2.10 (s, 6H), 2.12 (s, 6H), 2.26 (s, 6H) 4.08
(s, 2H), 6.88 (s, 2H); 13C NMR (DMSO-d6,100 MHz) d
17.73, 19.34, 21.05, 31.74, 130.83, 131.83, 133.22, 135.22,
136.37, 138.16, 171.64.
sealed Siemens ceramic diffraction tube (k = 0.71073 A)
˚
and a highly oriented graphite monochromator operating
at 50 kV, 30 mA. The lattice parameters and standard devi-
ations were obtained by a least squares fit using 25 frames
with 20 s/frame exposures with the Bruker SMART soft-
ware. The data were processed and reduced using Bruker
SAINTPLUS and empirical absorption correction was
made using Bruker SADABS. The structure was solved
by Direct Methods using SHELXTL package, and refined
by the full matrix least-squares based on F2 using
SHELX97 program (Table 1). The hydrogen atoms were
located from the difference Fourier for all cases, except
for phenol hydrogen and hydrogen atoms of the guest
water molecule in 3–PhOH. In this case, the missing hydro-
gens were not fixed and the refinement was done without
hydrogen atoms.
2.6. X-ray crystallographic structure determinations
The crystals were mounted in a glass capillary, cooled to
100 K and the intensity data were collected on a Bruker
Nonius SMART APEX CCD detector system with Mo
Table 1
The crystal data for diacid 3-free and its phenol and DMSO
pseudopolymorphs
Property
3-Free
3–DMSO
3–PhOH
CCDC 658513–658515 contain the supplementary crys-
tallographic data. These data can be obtained free of
charge via www.ccdc.cam.ac.uk/conts/retrieving.html (or
from the Cambridge Crystallographic Data Centre, 12,
Union Road, Cambridge CB2 1EZ, UK; fax: +44 1223
336033).
Chemical formula
Formula weight
Crystal system
Space group
C
340.4
21H24O4
C25H36O6S2 C27H30O6
496.66
450.51
Orthorhombic
Pbca
Monoclinic Monoclinic
C2/c
C2/c
˚
a (A)
21.320(3)
7.5687(11)
22.508(3)
90
100.103(5)
90
3575.8(9)
8
100(2)
0.086
33.844(4)
8.8859(10)
8.6693(10)
90
16.574(3)
11.793(3)
23.350(5)
90
˚
b (A)
˚
c (A)
a (°)
b (°)
c (°)
102.720(2)
90
90
90
3. Results and discussion
3
˚
Volume (A )
2543.2(5)
4
4563.9(17)
8
The diacid 3 as well as its sterically-unhindered analog 4,
were conveniently synthesized by following the protocol
shown in Scheme 1; for 3, dimesitylmethane was subjected
to bromination, cyanation and hydrolysis sequence to
afford initially the diamide [13] followed by further
hydrolysis to give the diacid 3 (Scheme 1). The diacid 4
was prepared by direct condensation of benzoic acid with
para-formaldehyde in H2SO4 [10].
Z
T (K)
100(2)
0.247
100(2)
0.092
l (mmꢂ1
F(000)
)
1456
9673
3487
1064
1920
Reflections collected
Unique reflections
6915
2491
24251
4448
Reflections with I > 2r(I) 2365
2067
3214
qcalc (g cmꢂ3
)
1.265
1.022
0.060
0.1336
1.297
1.311
Goodness-of-fit
Final R1 [I > 2r(I)]
Final wR2
1.060
1.001
0.048
0.056
The crystallization of diacid 3 was tried in a variety of
solvents. While it precipitated out as a powdery material
0.1166
0.1402
Me
Me
Me
Me
Me
(HCHO)n
i) NBS, CH3CN, rt, 8 h
HCOOH, 80 oC, 2 h
(85%)
ii) CuCN, DMF, reflux, 4 h
(71%)
Me
Me Me
Me
Me
Me
Me Me
Me
Me
Me
Me
CN
CN
Me
Me
H2SO4-H2O
H2SO4
TFA, 80 oC, 3 h
(64%)
Me
Me Me
Me
COOH
Me
Me Me
Me
CONH2
CH3COOH, reflux, 16 h
(83%)
COOH
CONH2
3
O
(HCHO)n
H2SO4, O0
OH
rt, 3h
COOH
COOH
4
Scheme 1. Preparation of diacids 3 and 4.

J.N. Moorthy, P. Natarajan / Journal of Molecular Structure 885 (2008) 139–148
143
in several of the solvents upon slow evaporation of their
solutions, crystals that appeared like shiny chunks grew
from the solution of EtOAc–CHCl3 in the presence of
added phenol. In a similar manner, the solution of the acid
in DMSO yielded thick blocks upon allowing the solution
of 3 in DMSO for a period of 3 weeks. A slow diffusion of
HCl vapor into the solution of acid 3 in NaHCO3 led to
square-shaped crystals. While the crystals obtained via dif-
fusion of HCl into the solution of the disodium salt of the
diacid 3 were found to be guest free (3-Free), those of the
crystals grown from DMSO were found to contain two sol-
vent molecules in the crystal lattice (3–DMSO). In a similar
manner, the crystals that were grown from phenol-contain-
ing solutions of the diacid 3 were found to include one phe-
nol and one adventitious water molecule (3–PhOH). The
crystal data for the three cases are given in Table 1. Intrigu-
ingly, the diacid 4 did not crystallize in a number of crystal-
lization attempts under a variety of conditions.
The diacid 3-free crystallizes in the monoclinic crystal
system (C2/c space group) with Z0 = 1. In Fig. 2 are shown
the molecular structure as well as the crystal packing dia-
gram. As expected, the two carboxyl groups lie almost per-
pendicular to the planes of the mesitylene rings to which
they are connected (h = 82.68° and 80.28°). The angle
between the planes of the mesitylene rings that are joined
by the methylene spacer is 116.3°, and the molecule as a
Fig. 2. The molecular structure of the diacid 3-free (a), dimeric association via two O–Hꢀ ꢀ ꢀO hydrogen bonds (b), further assembly into a tetramer (c) and
the overall crystal packing (d).

144
J.N. Moorthy, P. Natarajan / Journal of Molecular Structure 885 (2008) 139–148
whole assumes the shape of ‘V’. Most importantly, the con-
formation with respect to the carboxyl groups present in
two rings is found to be syn. The crystal packing in
Fig. 2 reveals a rather unusual pattern for the assembly
of the carboxyl groups. Whereas the dimer and the cate-
meric forms are not observed, the centrosymmetrically-
related acids are found to form a dimer via two O–Hꢀ ꢀ ꢀO
hydrogen bonds involving both of the two carboxyl groups
so far (Ref. codes: BODPEP, CLACET01, EJEQOZ and
ETYTAC). It is noteworthy that the tetrameric assembly
observed for the diacid 3 constitutes the first instance for
any aromatic carboxylic acid in particular.
The crystals of the diacid 3 grown from DMSO were
found to include the latter in the crystal lattice. The crystals
belonged to monoclinic system (space group: C2/c). The
asymmetric unit cell was found to consist of only half of
the diacid 3 with associated DMSO solvent molecule,
which suggests the stoichiometry between the diacid 3
and DMSO to be 1:2; this relative ratio was also estab-
lished by 1H NMR analysis of the crystals dissolved in ace-
tone-d6. Due to the 2-fold symmetry, the carboxyl groups
of the diacid 3 assume anti orientation. The crystal packing
of the 3–DMSO lattice inclusion complex is shown in
Fig. 3. One may draw a close analogy between the molec-
ular packing in this crystal with that observed in the
˚
˚
(D = 2.58 A, d = 1.80 A and hO–Hꢀ ꢀ ꢀO = 153.6°), thereby
leaving open one O–H donor and one C@O acceptor for
further self-assembly. Two such dimers assemble further
˚
˚
via 2 O–Hꢀ ꢀ ꢀO bonds (D = 2.68 A, d = 1.84 A and hO–
Hꢀ ꢀ ꢀO = 171.0 °), thereby completing a tetrameric motif
R44(16) motif as shown in Fig. 2. The tetrameric assembly
is a rare occurrence for carboxylic acids in general. The rig-
orous CSD analyses carried out recently [5d] show only
four cases for which such an assembly has been observed
Fig. 3. The molecular structure of the diacid 3 with DMSO (a), the solvent-expanded dimer (b) and crystal packing (c). Notice that the diacids form a tape
via the intermediacy of DMSO solvent molecules.

J.N. Moorthy, P. Natarajan / Journal of Molecular Structure 885 (2008) 139–148
145
guest-freelatticediscussedabove. Eachofthecarboxylgroups
isfound to be hydrogen bondedtothe oxygenatom of DMSO
˚
analyses shows that the two carboxyl hydrogens of the
two acids that are translation-symmetry related are bonded
to phenol–water couple via strong O–Hꢀ ꢀ ꢀO hydrogen
via
a
strong O–Hꢀ ꢀ ꢀO hydrogen bond (D = 2.58 A,
˚
˚
˚
d = 1.71 A and hO–Hꢀ ꢀ ꢀO = 164.4°). Two such centrosymmet-
rically-related pairs assemble via 2 C–Hꢀ ꢀ ꢀO hydrogen bonds
bonds (D = 2.54 A, d = 1.73 A, hO–Hꢀ ꢀ ꢀO = 173.8° and
˚
˚
D = 2.64 A, d = 1.85 A, hO–Hꢀ ꢀ ꢀO = 161.2°) as shown in
Fig. 4. Interestingly, the acceptor carbonyl oxygen atoms
of the two carboxyl groups involve in strong C-Hꢀ ꢀ ꢀO hydro-
˚
˚
(D = 3.21 A, d = 2.31 A and hC–Hꢀ ꢀꢀO = 156.0 °) to form sol-
vent-expanded dimers that propagate as tapes, Fig. 3.
We have analyzed the crystal structures of carboxylic
acids containing DMSO deposited in CSD (ver. 1.9,
2007). There is a variety of ways by which the DMSO is
included in the crystal lattices of the carboxylic acids.
The modes of DMSO inclusion I–V are schematically
shown in Chart 2. The association mode observed in the
case of 3–DMSO corresponds to the pattern II. The solvent
molecules are firmly bound in the crystal lattice, and the
TGA analyses show that solvent molecules begin to escape
at ca. 198 °C.
˚
˚
gen bonds (D = 2.82 A, d = 1.91 A, hC–Hꢀ ꢀ ꢀO = 130.5° and
˚
˚
D = 3.22 A, d = 2.54 A,
hC–Hꢀ ꢀ ꢀO = 130.2°) with the
b-glide-related another molecule as shown in Fig. 4. The
inclusion of phenol by acids is indeed very rare. The CSD
analyses reveal only 4 examples of phenol inclusion com-
plexes of carboxylic acids (refcodes: FEYYUE, GOSPOL,
JIDWEY01, XIBQUU). We believe that the role of water
is also space-filling in addition to bridging the phenol and
the carboxyl group of the diacid via hydrogen bonding.
Clearly, the molecular topology of the diacid 3 appears to
be the sole cause for the observed guest inclusion.
As mentioned at the outset, the motivation to analyze
the self-assembly of diacids 3 and 4 was to explore if the
non-coplanar aromatic dicarboxylic acids exhibit increased
propensity for guest inclusion. Although molecules tend to
pack as closely as possible in the crystals, strong and spe-
cific directional interactions may lead to open networks
depending on the fashion in which the interactions are pre-
destined in the molecular building blocks. Due to two
strong and directional hydrogen bonds, the dimer motif
of carboxyl groups has been exploited reliably in molecular
tectonics to create open-framework structures in which the
guest molecules occupy the void spaces. For example, the
The crystals of 3–PhOH were found to correspond to
orthorhombic crystal system (space group: Pbca). The crys-
tal structure shows that the diacid molecules exist in anti
conformation in this case as well. Further, one molecule
of phenol and adventitious water are found to be included
in the crystal lattice for each molecule of the diacid. Due to
the sterics of the methyl groups at ortho positions, the car-
boxyl groups are twisted by 82.39° and 70.28° with respect
to the mesityl rings to which they are bonded. In Fig. 4 is
shown the crystal packing down a-axis, which reveals the
formation of channels down a-axis between the layers of
the ‘V’-shaped diacids. The guest phenols that are hydro-
gen bonded to water reside in these channels. A careful
O
O
H
CH2
S
H
R
O
O
O
O
O
H
O
H2C
S
H
O
R
R
CH2
O
H
O
H
R
H
H
H
R
H
H
H2C
CH3
O
O
S
O
O
H
S
O
H3C
H3C
CH3
H2C
O
O
H
S
O
R
R
CH2
I
II
III
O
R
H
O
CH2
S
O
H
O
O
H
R
H3C
O
S
H
H2C
H
H2C
H
O
O
H
CH3
R
O
O
H
S
S
O
R
H2C
O
O
H
H2C
H2C
O
H
R
iV
V
H
Chart 2. Various patterns in which the dimer motif of carboxyl groups is expanded by DMSO solvent molecules.

146
J.N. Moorthy, P. Natarajan / Journal of Molecular Structure 885 (2008) 139–148
Fig. 4. The mode of assembly of the carboxyl groups of 3 with lattice included phenol and water (top). The crystal packing of 3–PhOH down b-axis
(bottom). The included phenol and water oxygen are shown in a space-filling model. It is noteworthy that the association of acids with water leads to
channels down a-axis, within which the guest phenols are located.
2-dimensional trimesic acid [14] and related derivatives [5a]
and hexakis(p-carboxyphenyl)benzene [15] and 3-dimen-
sional 1,10-binaphthyl-2,20-dicarboxylic acid [16] and spiro-
bifluorene-2,20,7,70-tetracarboxylic acid [17] crystallize with
inclusion of guest molecules in the networks of the acids
assembled by the dimer motif. Indeed, Nangia and co-
workers [18] exploited the dimer motif to develop host-
guest inclusion compounds. However, the trade-off
between close-packing and exploitation of strong-hydrogen
bonding interactions appears to be one of the reasons for
the observation of structures in which neither of the
dimeric and catemeric motifs of carboxylic acid is
observed. For example, tetrakis(p-carboxyphenyl)methane
[19] crystallizes with inclusion of toluene and DMSO,
and the carboxyl groups are found to be bonded to DMSO
solvent. Similarly, 3,30,5,50-tetrakis(p-carboxyphenyl)bime-
sityl, reported by us [20], also crystallizes with DMSO
bonded to the carboxyl group.
For exploitation of carboxylic acids in the molecular
self-assembly, it is important to understand the factors that
influence adoption of a particular motif, i.e. dimeric or
catemeric. As mentioned earlier, Desiraju et al. recently
showed that the rare catemeric motif predominates in a
unique class of arylpropiolic acids, in which auxiliary C–
Hꢀ ꢀ ꢀO hydrogen bonds augment the catemeric motif
[5c,5d]. Clearly, the structural attributes play an important
and decisive role in the determination of ultimate crystal
packing and the observation of a particular packing/bind-
ing motif. We sought to investigate this aspect also in the
conformationally-flexible diarylmethane dicarboxylic acids
3 and 4, which may in principle exhibit two extreme low-
energy conformations, i.e., syn and anti, cf. Chart 1. Inso-
far as the diacid 3 is concerned, the rigidity of the carboxyl
groups due to steric hindrance from the ortho methyl
groups imposes barrier on the directionality for the self-
assembly via carboxyl groups. Hence, the diacid 3 may

J.N. Moorthy, P. Natarajan / Journal of Molecular Structure 885 (2008) 139–148
147
be considered as a rigid ‘V’-shaped module with freedom
for rotation of each of the carboxymesityl rings as a whole
with respect to the central methylene carbon. The rotation
of one of the carboxymesityl ring with respect to the other
leads to two low-energy conformations, i.e., syn and anti.
Thus, the diacid 3 offers the opportunity to analyze confor-
mation-dependent crystal packing behavior in addition to
exploring the influence of structural attributes involving
non-coplanar aryl rings on the solid-state self assembly of
dicarboxylic acids. The diacid 4, the sterically-unhindered
analog of 3, may also explore the two extreme conforma-
tions, but the carboxyl groups in this case are not posed
with any barrier to lie in the plane of the phenyl rings.
Let us first consider the behavior of the diacid 3. What is
noteworthy is the fact that the molecules adopt syn confor-
mation in the solvent-free modification of the diacid of 3,
while they explore anti conformation in DMSO and in
the presence of phenol. Neither of the dimeric and catemer-
ic motifs is observed for the self-assembly of 3 in 3-free, but
a very rare tetrameric association is found, cf. Fig. 2. As
mentioned earlier, this pattern is known for only four
instances so far. Is it the steric hindrance that prohibits
adoption of dimeric/catemeric motifs? There are several
instances of sterically-hindered acids that include 2,6-dim-
ethylbenzoic acid, mesitoic acid, pentamethylbenzoic acid,
etc. for which the dimer motif is observed. It is known that
steric factors may influence the adoption of catemeric motif
in favor of the dimer motif [4,5c]. Thus, the reason why the
latter is not observed in 3-free should be traceable to the
tethering of two such acids by a methylene carbon, which
evidently changes the shape of the molecule from a planar
constituent mesitoic acid to a more complex ‘V’-shaped
rigid structure. We believe that sterically-hindered dicar-
boxylic acids are very sensitive to adopting the dimeric
motif such that a simple perturbation by way of electronic
factors or structural variation of the kind such as tethering
in 3 may result in altogether new and rare pattern of
assembly; we have previously shown that dibromo/
dichloro-substitution of mesitylene carboxylic acid leads
to helical self-assembly [3]. Presumably, the close-packing
of the diacid 3 in its syn and anti conformations via
dimeric/catemeric motif is difficult because of the tethering
such that the rare tetrameric assembly as shown in Fig. 1
emerges ultimately as the final outcome. We were not suc-
cessful in identifying any other guest-free modification in
which the relative disposition of carboxyl groups is anti.
However, it is this conformation that is found in the pseud-
opolymorphs of the diacid 3 with DMSO and PhOH–H2O.
This shows that the anti-diacid presumably cannot undergo
close packing except with appropriate guest molecules. It is
noteworthy that the guest species in both DMSO and
PhOH are directly hydrogen bonded to the carboxyl
groups of the diacid 3.
appears that the earlier attempts to obtain crystals were also
in vain, which is reflected from the structure determination
using powder XRD data (Ref. code: BZPHCC). This further
attests to the difficulty in the crystal growth of simple meth-
ylene-tethered carboxylic acids, which assume a complex
enough shape to pose severe problems with crystallization.
4. Conclusions
We have examined the solid-state self-assembly of a steri-
cally-hindred methylene-tethered dicarboxylic acid 3, which
may explore two extreme and low-energy conformations,
viz., syn and anti. The molecular topology in conjunction
with steric hindrance appear to divert the syn-diacid 3 from
adopting the otherwise prevalent dimer motif in favor of a
very rare tetrameric association. The same attributes appear
to impart guest-binding propensity to the diacid molecules in
their anti conformations resulting in the occurrence of
pseudopolymorphism with DMSO and PhOH–H2O. Given
that the carboxyl group is one of the important functional
groups that is exploited in crystal engineering, the conforma-
tion-dependent crystal packing and observation of pesudo-
polymorphism for the diacid 3 in its syn and anti
conformation suggest that the sterically-hindered acids are
highly susceptible to exhibiting variations in association
modes. The fact that simple unhindered dicarboxylic acid 4
does not crystallize in various attempts bears out the influ-
ence of structural changes through methylene tethering on
the self-assembly of aromatic carboxylic acids.
Acknowledgements
J.N.M. thanks the Department of Science and Technol-
ogy (DST), India, for generous financial support through
Ramanna fellowship. P.N. is grateful to University Grants
Commission (UGC), India, for a research fellowship.
References
[1] G.R. Desiraju, Crystal Engineering. The design of Organic Solids,
Elsevier, Amsterdam, The Netherlands, 1989.
[2] G.R. Desiraju, Angew. Chem., Int. Ed. (1995) 2311.
[3] J.N. Moorthy, R. Natarajan, P. Mal, P. Venugopalan, J. Am. Chem.
Soc. 124 (2002) 6530, and references therein.
[4] L. Leiserowitz, Acta Crystallogr. B 32 (1976) 775.
[5] (a) S.V. Kolotuchin, E.E. Felon, S.R. Wilson, C.J. Loweth, S.C.
Zimmerman, Anew. Chem., Int. Ed. 34 (1995) 2654;
(b) T. Steiner, Acta Crystallogr. B 57 (2001) 103;
(c) D. Das, G.R. Desiraju, Chem. Asian J. 1 (2006) 231;
(d) D. Das, Ph.D. Thesis, University of Hyderabad, 2006.
[6] S.S. Kuduva, D.C. Craig, A. Nangia, G.R. Desiraju, J. Am. Chem.
Soc. 121 (1999) 1936.
[7] J. van de Streek, Cryst. Eng. Commun. 9 (2007) 350.
[8] J.N. Moorthy, R. Natarajan, P. Venugopalan, Angew. Chem., Int.
Ed. 41 (2002) 3417.
[9] (a) J. Bernstein, Polymorphism in Molecular Crystals, Oxford
University Press, Oxford, 2002;
Strangely, we could not obtain the crystals of the diacid 4
from a number of crystallization attempts in a variety of sol-
vents that include EtOAc, MeOH, dioxane, CHCl3,
NO2C6H5, CH3CN, etc. and mixed solvent systems. It
(b) There is a controversy associated with the usage of the term
‘pseudopolymorphism’, see: A. Nangia, Cryst. Growth Des. 6 (2006)
2 and references therein.

148
J.N. Moorthy, P. Natarajan / Journal of Molecular Structure 885 (2008) 139–148
[10] R.W. Beattie, R.H.F. Manske, Can. J. Chem. 42 (1964) 223.
[11] H.B. Sun, R. Hua, Y. Yin, Tetrahedron Lett. 47 (2006) 2291.
[12] J.H. Cornell Jr., M.H. Gollis, Org. Syn. 42 (1962) 57.
[13] J.N. Moorthy, N. Singhal, J. Org. Chem. 70 (2005) 1926.
[14] (a) F.H. Herbstein, Crystalline Molecular Complexes and Com-
pounds, vol. 1, Oxford University Press, London, 2005;
(b) F.H. Herbstein, Top. Curr. Chem. 40 (1987) 107.
[15] K. Kobayashi, T. Shirasaka, E. Horn, N. Furukawa, Tetrahedron
Lett. 41 (2000) 89.
[17] P. Holy, M. Havlik, M. Tichy, J. Zavada, I. Cisarova, Collect. Czech.
Chem. Commun. 13 (2006) 71.
[18] (a) V.S.S. Kumar, A. Nangia, M.T. Kirchner, R. Boese, New. J.
Chem. 27 (2003) 224;
(b) V.S.S. Kumar, A. Nangia, A.K. Katz, H.L. Carrell, Cryst.
Growth Des. 2 (2002) 313;
(c) G. Smith, M.G. Coyne, J.M. White, Aust. J. Chem. 53 (2000) 203.
[19] N. Malek, T. Maris, M. Simard, J.D. Wuest, Acta Crystallogr. E 61
(2005) o518.
[16] L. Izotova, B. Ibragimov, J. Ashurov, S. Talipov, E. Weber, Cryst.
Growth Des. 6 (2006) 2523.
[20] J.N. Moorthy, R. Natarajan, P. Venugopalan, J. Org. Chem. 70
(2005) 8568.