Topological equivalences between coordination polymer and co-crystal: A tecton approach in crystal engineering

Article abstract of DOI:10.1021/cg401858s

The ligand benzene-1,3,5-triyltriisonicotinate (1) was shown to form topological equivalent organic and metal-organic networks upon treatment with trimesic acid (H₃TMA) and Ag(I) salt, respectively. Both exhibit interpenetrated (10,3)-b networks, signifying the tecton equivalency between Ag(I) and H₃TMA. The metal-organic network is doubly interpenetrated, while the organic network is 8-fold interpenetrated which highlights the size differences between Ag(I) and H₃TMA. Despite interpenetration, both the networks were found to incorporate solvents MeOH and CHCl₃ in their channels. Further, the repeat of reaction of Ag(I) with 1 in the presence of a pyrene guest was found to produce crystals of open honeycomb (6,3) network with inclusion of pyrene molecules in the channels.

Full text of DOI:10.1021/cg401858s

Communication
pubs.acs.org/crystal
Topological Equivalences between Coordination Polymer and Co-
crystal: A Tecton Approach in Crystal Engineering
Gargi Mukherjee and Kumar Biradha*
Department of Chemistry, Indian Institute of Technology, Kharagpur 721302, India
S
* Supporting Information
ABSTRACT: The ligand benzene-1,3,5-triyltriisonicotinate (1) was
shown to form topological equivalent organic and metal−organic
networks upon treatment with trimesic acid (H3TMA) and Ag(I) salt,
respectively. Both exhibit interpenetrated (10,3)-b networks,
signifying the tecton equivalency between Ag(I) and H3TMA. The
metal−organic network is doubly interpenetrated, while the organic
network is 8-fold interpenetrated which highlights the size differences
between Ag(I) and H3TMA. Despite interpenetration, both the
networks were found to incorporate solvents MeOH and CHCl3 in
their channels. Further, the repeat of reaction of Ag(I) with 1 in the
presence of a pyrene guest was found to produce crystals of open
honeycomb (6,3) network with inclusion of pyrene molecules in the
channels.
rediction of the crystal structure for a given molecular
Scheme 1. (a) Synthon I and (b) synthon II
P
structure is accepted to be one of the difficult tasks in the
crystal engineering, as several factors govern the crystallization
reaction outcome.1 The failure of such predictions was
provocatively referred to as “one of the continuing scandals”
by Maddox in 1988.2 The tecton approach that is purely based
on molecular symmetry and the placement of functional groups
on the molecule and the supramolecular synthon approach
which deals with the robustness of the functional group
interactions serve the purpose of structural prediction to some
extent in cases of highly symmetrical molecules/supramolecular
architectures.3−5 In short, in the tecton approach, molecular
building blocks are highlighted, while in the synthon approach,
interactions are given importance; however, the same under-
lying principle of molecular programming is there in both of
them. Even in the case of symmetrical molecules, several other
possible structures exist which differ significantly from
predictable ones. For example, a 3-fold molecule can exhibit a
honeycomb network as well as (10,3)-a or -b networks;
similarly, a S4-symmetrical molecule can exhibit a diamondoid
or 2D-layer structures.6,7 The generation of crystal structures of
single organic components by using two or multi organic
components, cocrystals or salts, is also one of the current
challenges in crystal engineering. In this regard, the honeycomb
network that was formed by trimesic acid (H3TMA) via
synthon-I was shown to be formed by H3TMA (node) and
4,4′-bipy or other spacers using synthon-II (Scheme 1).8,9
Similarly, the formation of extended diamondoid networks via
synthon-II using two components was also shown very
recently.7,10 Further, the 3-fold symmetric molecules when
linked (two-connected) through metal nodes were known to
form (10,3)-a and (10,3)-b networks for a long while.11
However, only recently the organic counterpart of (10,3)-a
network was reported, but to date no (10,3)-b network was
reported to the best of our knowledge.12
On the other hand, the generation of topologically equivalent
purely organic and metal−organic networks is also one of the
challenging tasks of crystal engineering.13 Herein, we report
one such example using the molecule, benzene-1,3,5-
triyltriisonicotinate (1), which was linked to form topologically
equivalent networks via 3-connected organic (H3TMA) or
inorganic nodes/tectons [Ag(I)]. In the case of H3TMA and 1,
the networks are propagated through synthon-II, whereas in
the case of Ag(I) and 1, the networks are propagated by
coordination bonds. Recently, it was shown by us that the
molecule 1 forms two-dimensional iso-structural CPs that
exhibit remarkable ability of exchanging cations/anions in
single-crystal-to-single-crystal manner and also exhibit breath-
ing behavior in N2 sorption.14,15 In continuation of our studies
on CPs of 1, the molecule 1 was reacted with Ag(I) salts using
the layering technique.
Layering of methanolic solution of AgPF6 over the
chloroform solution of 1 afforded needle-shaped crystals of
Received: December 13, 2013
Revised: January 9, 2014
Published: January 13, 2014
© 2014 American Chemical Society
419
dx.doi.org/10.1021/cg401858s | Cryst. Growth Des. 2014, 14, 419−422

Crystal Growth & Design
Communication
cocrystal 2, {[Ag(1)(PF6)]·(CH3OH)·(H2O)·(CHCl3)}n (Fig-
ure 1). The single crystal X-ray analysis reveals that complex 2
CHCl3−MeOH (Figure 2). The crystal structure analysis of 3
reveals that it is crystallized in P21/c space group, and it is a
Figure 1. Illustrations for the crystal structure of 2: (a) T-shape
coordination geometry of Ag(I), (b) the shortest circuit formed by five
each of Ag(I) atoms and molecules of 1, (c) 3D-network with (10,3)-b
topology, (d) 2-fold interpenetrated network, Ag(I) and 1 are
represented as nodes, and (e) interpenetration of networks through
π···π and dipolar interactions between the central phenyl rings and the
keto groups of ester, respectively.
Figure 2. Illustrations for the crystal structure of 3: (a) the shortest
circuit formed by five molecules of 1 and five H3TMA molecules, (b)
organic network with (10,3)-b topology, (c) 8-fold interpenetrated
network, and (d) interpenetration of networks through π···π
interactions between the C6 rings, between the pyridyl arms, and
H3TMA and pyridyl arms of 1.
is crystallized in the C2/c space group. The asymmetric unit is
constituted by one unit each of 1, Ag(I), and PF6 anion. The
Ag(I) ion adopts a distorted trigonal pyramid geometry: the
three pyridyl groups occupy the trigonal plane (N−Ag: 2.375,
2.210, and 2.209 Å; N−Ag−N: 99, 114, and 146°) and the
disordered PF6 anion is coordinated at the apical position. The
Ag(I) ion is elevated from the N3 plane by 0.128 Å, and the
pyridyl rings make angles of 47°, 28°, and 36° with the N3
plane. In the crystal structure, both Ag(I) and 1 act as 3-
connected nodes and propagate the (10,3)-b framework. The
shortest circuit is composed of five units each of Ag(I) and 1. It
is interesting to note here that within the (10,3)-b network, the
coordinated PF6 ions point in one-direction such that the
chirality of the network is maintained. The remaining space of
the cavities of network was filled by self-interpenetration of the
network which has the opposite sense and solvent molecules
which occupy 15.6% of the crystal volume. The inter-
penetration of the networks is assisted by the π−π interactions
between the central phenyl rings (3.979 Å) and dipole−dipole
interactions between the CO groups of ester (3.079; 3.207
Å) (Figure 1e). It is interesting to note that the CO that
exhibits shorter dipole−dipole interaction has longer C−O
distance (1.174 vs 1.257 Å), although the IR spectra shows only
one peak at 1751 cm−1.
cocrystal. The C−O and CO bond lengths of −COOH of
H3TMA in 3 are 1.303 and 1.207; 1.304 and 1.216; and 1.304
and 1.214 Å, which clearly indicates that the −COOH groups
are not deprotonated. The molecules assemble through
synthon-II (3.273, 2.6771; 2.694, 3.295; 3.226, 2.632 Å) to
form a (10,3)-b network containing two organic components.
The ligand geometry in the cocrystal deviates from the 3-fold
symmetry similar to the one observed in complex 1. The
interplanar angles between the central phenyl plane and the
pyridyl planes are 31.38°, 35.38°, and 39.45°. In 3, the shortest
circuit is composed of five each of H3TMA and 1. We note here
that the both nodes in 3 show trigonal planar environment with
angles close to 120°, whereas in 2, Ag(I) severely deviates from
trigonal planar geometry (150.15°, 113.25°, and 93.42°),
although 1 has nearly 120° angles. In 3, the (10,3)-b networks
are 8-fold interpenetrated in contrast to 2-fold interpenetration
of 1. The increase in interpenetration could be the consequence
of longer arm length of the H3TMA tecton compared to the
Ag(I) ion. The interpenetration occurs through the plethora of
π···π interactions as follows: The units of 1 stack on each other
in a staggered fashion such that central C6 units interact with
each other (3.837 Å). In between the pyridyl arms of the stacks,
the adjacent pyridyl arms (3.768 Å) and H3TMA interdigitate
(3.754 Å) in a 4:2 ratio via π···π interactions (Figure 2e).
With the production of the topological equivalent (10,3)-b
networks of 1 with 3-connected tectons, the possibility of
eschewing the interpenetration of these networks in favor of
As the Ag(I) serves as the 3-connected node in the structure,
it was thought that H3TMA as a probable tecton to produce
organic cocrystal of 1 and H3TMA containing (10,3)-b network
via synthon-II. Crystals of cocrystal 3, {[1·H3TMA]·(CH3OH)·
2(CHCl3)}n, suitable for single crystal X-ray diffraction were
obtained by dissolving 1:1 equivalents of 1 and H3TMA in
420
dx.doi.org/10.1021/cg401858s | Cryst. Growth Des. 2014, 14, 419−422

Crystal Growth & Design
Communication
guest inclusion was explored by carrying these reactions in the
presence of aromatic guest molecules.16 The layering of
methanolic solution of AgPF6 over the chloroform solution of
1 and pyrene afforded colorless, block-shaped crystals of
complex 4, {[Ag2(1)2]·2(PF6)·3(pyrene)·2(CHCl3)}n. The
single crystal X-ray analysis reveals that the complex 4
H3TMA and 1 (Figure 4). The bands of 2 and 3 varied
considerably from those of 1 and H3TMA. 1 and H3TMA
crystallized in P1, and the asymmetric unit is constituted by
̅
one each of 1, Ag(I) and the PF6 anion, and 1.5 units of pyrene.
In 4, Ag(I) exhibits planar coordination geometry, unlike the
trigonal pyramid geometry in 2. However, similar to 2, both
Ag(I) ion and 1 act as a three-connected node but forms an
open honeycomb network with an arm length of 9.8 Å (Figure
3). As anticipated, the interpenetration was eschewed in favor
Figure 4. Diffuse reflectance spectra of 1 (red line), H3TMA (purple
line), 2 (blue line), and 3 (green line).
showed peaks at 268 and 298 nm, respectively. However, both
2 and 3 were found to exhibit three absorption peaks; for 2 the
absorption peaks were observed at 211, 256, and 330 nm, while
those of 3 are located at 213, 260, and 294 nm.
In conclusion, it was demonstrated that H3TMA is an
equivalent tecton to Ag(I) to produce topologically equivalent
(10,3)-b networks upon self-assembling with 1. The size of the
tecton altered the degree of interpenetration, smaller Ag(I)
produced doubly interpenetrated and the bigger tecton
H3TMA produced an 8-fold interpenetrated network. The
ligand−ligand interactions and Ag(I)−ligand interactions
played a significant role in the interpenetration of networks.
The cocrystal 3 is the first of its kind as there is no report on
cocrystal of 1 to date, and also it represents the first example of
the organic two component (10,3)-b network. The presence of
the guest such as pyrene was found to influence the network
geometry and interpenetration in the case of Ag(I) and 1 but
not in case of H3TMA and 1. Smaller aromatic guests than
pyrene failed to produce single crystals, even in case of the
reaction of Ag(I) and 1.
Figure 3. Illustrations for the crystal structure of 4: (a) honeycomb
network formed by Ag(I) and 1 with pyrene dimers inside the
hexagonal cavities; packing of layers (b) top view and (e) side view,
channeled pyrenes were shown in parrot green and cylinder mode, the
pyrenes between the bilayers were shown in dark green and a space
filling mode; (d) column of pyrene dimers; and (e) intercalation of
pyrene between the bilayers via π···π and CO···π interactions.
of the inclusion of the pyrene guest molecule. The honeycomb
layers form double layers via Ag···π (4.507 Å) and OC···π
(3.460 Å) interactions. These double layers pack in slipped
manner such that there exits continuous channels which are
occupied by a column of pyrene dimers and PF6 ions. Between
the double layers, one of the pyrene molecules is sandwiched by
π−π interactions between pyrene and Py-CO- moieties of 1
(4.659 Å). We note here that recently reported structures of 1
with Ag(I) salts have honeycomb networks which have offset
packing unlike the ones reported here. In those examples,
solvent molecules were found to act as guest molecules and
form unstable crystals.17
Efforts to obtain cocrystal of H3TMA with 1 in the presence
of pyrene or other aromatic guest molecules were not
successful. We note here that complex 4 bears striking
structural resemblance with that of [TMA·tpt]·2pyrene, as
both are honeycomb networks, and the channels are occupied
by pyrene molecules.18 However, the differences between the
present structure and that of [TMA·tpt]·2pyrene is that in
[TMA·tpt]·2pyrene the channels are filled by four molecules of
pyrene. In contrast in the present structure, those are filled by
two pyrene and two PF6 anions. Longer length of H3TMA and
the absence of anions may be reason of inclusion of two extra
pyrene molecules in that structure.
ASSOCIATED CONTENT
* Supporting Information
Experimental details, IR spectra, XRPD patterns, and TGA
data. This material is available free of charge via the Internet at
http://pubs.acs.org.
■
S
AUTHOR INFORMATION
Corresponding Author
*E-mail: kbiradha@chem.iitkgp.ernet.in. Fax: +91-3222-
282252. Tel: +91-3222-283346.
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We gratefully acknowledge the DST for the financial support
and DST-FIST for the single-crystal X-ray facility. G.M. thanks
CSIR for a research fellowship.
REFERENCES
■
Solid state diffuse reflectance spectra (DRS) of 2 and 3 were
recorded at room temperature and compared with those of
(1) (a) Jones, P. L.; Byrom, K. J.; Jeffery, J. C.; McCleverty, J. A.;
Ward, M. D. Chem. Commun. 1997, 1361−1362. (b) Beer, P. D.;
421
dx.doi.org/10.1021/cg401858s | Cryst. Growth Des. 2014, 14, 419−422

Crystal Growth & Design
Communication
́
Hayes, E. J. Coord. Chem. Rev. 2003, 240, 167−189. (c) Dıaz, P.;
Benet-Buchholz, J.; Vilar, R.; White, A. J. P. Inorg. Chem. 2006, 45,
1617−1626. (d) Lopez, S.; Keller, S. W. Inorg. Chem. 1999, 38, 1883−
1888. (e) Dalgarno, S. J.; Hardie, M. J.; Raston, C. L. Cryst. Growth
Des. 2004, 4, 227−234. (f) Mahata, P.; Sundaresan, A.; Natarajan, S.
Chem. Commun. 2007, 4471−4473.
(2) Maddox, J. Nature 1988, 335, 201.
(3) Simard, M.; Su, D.; Wuest, J. D. J. Am. Chem. Soc. 1991, 113,
4696−4698.
(4) Wuest, J. D. Chem. Commun. 2005, 5830−5837.
(5) Desiraju, G. R. Angew. Chem., Int. Ed. 1995, 34, 2311−2327.
(6) Rao, K. P.; Higuchi, M.; Duan, J.; Kitagawa, S. Cryst. Growth Des.
2013, 13, 981−985.
(7) (a) Roy, S.; Biradha, K. Cryst. Growth Des. 2013, 13, 3232−3241.
(b) Reddy, D. S.; Dewa, T.; Endo, K.; Aoyama, Y. Angew. Chem., Int.
Ed. 2000, 39, 4266−4268.
(8) (a) Duchamp, D. J.; Marsh, R. E. Acta Crystallogr., Sect. B 1969,
25, 5−19. (b) Herbstein, F. H. Top. Curr. Chem. 1987, 140, 107−139.
(9) (a) Weyna, D.; Shattock, R. T.; Vishweshwar, P.; Zaworotko, M.
J. Cryst. Growth Des. 2009, 9, 1106−1123. (b) Sharma, C. V. K.;
Zaworotko, M. J. Chem. Commun. 1996, 2655−2656. (c) Santra, R.;
Ghosh, N.; Biradha, K. New J. Chem. 2008, 32, 1673−1676.
(10) (a) Roy, S.; Mahata, G.; Biradha, K. Cryst. Growth Des. 2009, 9,
5006−5008. (b) Men, Y.-B.; Sun, J.; Huanga, Z.-T.; Zheng, Q.-Y.
CrystEngComm 2009, 11, 978−979.
(11) (a) Carlucci, L.; Ciani, G.; Proserpio, D. M.; Sironi, A. J. Am.
Chem. Soc. 1995, 117, 12861−12862. (b) Carlucci, L.; Ciani, G.;
Proserpio, D. M.; Sironi, A. J. Am. Chem. Soc. 1995, 117, 4562−4569.
(c) Abrahams, B. F.; Batten, S. R.; Hamit, H.; Hoskins, B. F.; Robson,
R. Chem. Commun. 1996, 1313−1314. (d) Yaghi, O. M.; Li, H. J. Am.
Chem. Soc. 1996, 118, 295−296. (e) Yaghi, O. M.; Davis, C. E.; Li, G.;
Li, H. J. Am. Chem. Soc. 1996, 119, 2861−2868. (f) Abrahams, B. F.;
Jackson, P. A.; Robson, R. Angew. Chem., Int. Ed. 1998, 37, 2656−
2659. (g) Biradha, K.; Fujita, M. Angew. Chem., Int. Ed. 2002, 41,
3392−3395. (h) Abrahams, B. F.; Haywood, M. G.; Hudson, T. A.;
Robson, R. Angew. Chem., Int. Ed. 2004, 43, 6157−6160. (i) Eubank, J.
F.; Walsh, R. D.; Eddaoudi, M. Chem. Commun. 2005, 2095−2097.
(j) Ke, Y.-X.; Collins, D.-J.; Sun, D.-F.; Zhou, H.-C. Inorg. Chem. 2006,
45, 1897−1899. (k) Halper, S. R.; Do, L.; Stork, J. R.; Cohen, S. M. J.
Am. Chem. Soc. 2006, 128, 15255−15268.
(12) Shattock, T. R.; Vishweshwar, P.; Wang, Z.; Zaworotko, M. J.
Cryst. Growth Des. 2005, 5, 2046−2049.
(13) (a) Reddy, D. S.; Craig, D. C.; Rae, A. D.; Desiraju, G. R. J.
Chem. Soc., Chem. Commun. 1993, 1737−1739. (b) Reddy, D. S.;
Craig, D. C.; Desiraju, G. R. J. Chem. Soc., Chem. Commun. 1995, 339−
340.
(14) Mukherjee, G.; Biradha, K. Chem. Commun. 2012, 48, 4293−
4295.
(15) Mukherjee, G.; Biradha, K. Chem. Commun. 2014, 50, 670−672.
(16) Santra, R.; Biradha, K. Cryst. Growth Des. 2009, 9, 4969−4978.
(17) Xiong, Y.; Yang, R.; Chen, S.; Jiang, J.; Su, C. -Y. CrystEngComm
2013, 15, 9751−9756.
(18) Ma, B.-Q.; Coppens, P. Chem. Commun. 2003, 2290−2291.
422
dx.doi.org/10.1021/cg401858s | Cryst. Growth Des. 2014, 14, 419−422