Hexagonal crystalline inclusion complexes of 4-iodophenoxy trimesoate

Article abstract of DOI:10.1039/b809592b

Bifurcated I...π and I...O=C halogen bonding interactions assist in formation of unique iodo-arene trimers leading to nanoscale channels in inclusion complexes of trimesic acid iodophenolate. The Royal Society of Chemistry.

Full text of DOI:10.1039/b809592b

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Hexagonal crystalline inclusion complexes of 4-iodophenoxy trimesoatew
F. Christopher Pigge,*a Venu R. Vangala,a Pradeep P. Kapadia,a Dale C. Swensona
and Nigam P. Rathb
Received (in Austin, TX, USA) 9th June 2008, Accepted 9th July 2008
First published as an Advance Article on the web 29th August 2008
DOI: 10.1039/b809592b
Bifurcated Iꢀ ꢀ ꢀp and Iꢀ ꢀ ꢀOQC halogen bonding interactions
assist in formation of unique iodo–arene trimers leading to
nanoscale channels in inclusion complexes of trimesic acid
iodophenolate.
bonded trimers are commonly observed motifs in these struc-
tures and, in combination with symmetry elements inherent to
the phenoxy triazine framework, give rise to hexagonal pack-
ing arrangements with large (B12 A diameter) channels.
We recently reported results of a study examining the
influence of halogen bonding interactions on the solid state
structures of 4-halo-substituted triaroylbenzenes (e.g., 1).9 In
continuing research, we sought to probe the interplay between
symmetry and electronic factors upon halogen bonding by
characterizing analogues of triaroylbenzenes in which haloaryl
moieties were substituted with electron donating or electron
withdrawing groups. As part of this work 1,3,5-tris(4-iodo-
phenoxycarbonyl)benzene (2) was prepared,10 and we wish to
disclose the results of preliminary crystallographic investiga-
tions. In particular, crystallization of 2 from CHCl3, pyridine,
and benzene produced hexagonal inclusion complexes posses-
sing nanometre-sized channels. Significantly, these porous
inclusion complexes appear to be mediated in part by unpre-
cedented Iꢀ ꢀ ꢀp trimer motifs supported in a bifurcated fashion
by Iꢀ ꢀ ꢀOQC interactions.
Crystalline inclusion complexes continue to provide inspira-
tion for the design of new organic and metal–organic solids.
Particularly intriguing from a functional materials standpoint
are inclusion complexes possessing sizable channels or pores
extending through the crystalline lattice. Such structures are
interesting as it is envisioned that crystal engineering of robust
tailor-made porous organic and metal–organic solids from
relatively simple molecular building blocks will afford valuable
new catalysts and molecular storage/separation devices.1 Not
surprisingly, the de novo synthesis of porous crystalline solids
(especially those comprised of purely organic components) has
proven to be non-trivial.2
Progress in this area, however, has been reported in recent
years. In particular, the tecton approach has been successfully
applied in the construction of organic inclusion complexes in
which large percentages (450%) of the crystal lattices are
solvent accessible.3 This strategy entails incorporation of
hydrogen bonding functionality along the periphery of rela-
tively rigid organic scaffolds. Strong hydrogen bonding inter-
actions may then become the dominant supramolecular
interaction in the crystal, sometimes at the expense of close
packing.
Non-covalent interactions between organic halogens and
nucleophiles (halogen bonding interactions) offer an alterna-
tive to hydrogen bonding as a structurally defining supramo-
lecular motif, and halogen bonding is emerging as an
important tool in crystal engineering.4,5 Halogen bonding
interactions between amines, nitrogen heterocycles, carbonyl
groups, and other organic halides (so-called type II halogen
bonding) is now well-documented and the strength of these
interactions can sometimes rival that of conventional hydro-
gen bonds.4,6 Porous organic crystalline networks mediated
through halogen bonding interactions, however, are rare,7
although notable exceptions are encountered in crystal struc-
tures of 4-chloro- and 4-bromophenoxy triazines.8 Halogen-
Crystals of 2 obtained from CHCl3 exhibited a hexagonal
morphology, and the structure was found to be an inclusion
complex of stoichiometry 2ꢀ3CHCl3 solved in the non-centro-
symmetric space group P65.z The triester 2 assembles in layers
stacked parallel to the c axis. Molecules of 2 adopt an
approximately symmetrical conformation, but slight differ-
ences in the angles of each iodophenoxy ring with respect to
the carbonyl linkages result in deviation from perfect C3
symmetry. The most striking feature of the structure is the
manner in which three iodophenoxy ester moieties from three
independent molecules come together in a triangular array
seemingly facilitated by the roughly trigonal symmetry of 2,
templating effects of the included solvent, and bifurcated Iꢀ ꢀ ꢀp
and Iꢀ ꢀ ꢀOQC interactions (Fig. 1). The metrics of the three
sets of halogen bonding interactions are not identical (reflect-
ing the absence of C3 symmetry in the molecular units) and all
Iꢀ ꢀ ꢀp and Iꢀ ꢀ ꢀO distances slightly exceed the sums of the
relevant van der Waals radii (see Fig. 1 caption).11 Despite
a Department of Chemistry, University of Iowa, Iowa City, Iowa
52242, USA. E-mail: chris-pigge@uiowa.edu;
Fax: +1 319-335-1270; Tel: +1 319-335-3805
b Department of Chemistry & Biochemistry, University of Missouri –
St. Louis, St. Louis, Missouri 63121, USA
w Electronic supplementary information (ESI) available: Preparation
of 2, DSC and PXRD data. CCDC reference numbers 688114–688117.
For ESI and crystallographic data in CIF or other electronic format,
see DOI: 10.1039/b809592b
ꢁc
This journal is The Royal Society of Chemistry 2008
4726 | Chem. Commun., 2008, 4726–4728

View Article Online
Fig. 1 2D layers in 2ꢀ3CHCl3. Triangular motif mediated by bifur-
cated Iꢀ ꢀ ꢀO and Iꢀ ꢀ ꢀp interactions (black lines): dIꢀ ꢀ ꢀO ¼ 3.59, 3.60, and
3.66 A; dIꢀ ꢀ ꢀp ¼ 3.59, 3.79, and 3.87 A. CHCl3 solvates not shown.
Fig. 3 View (down c) of 2ꢀ3C5H5N. Honeycomb network mediated
centroid
by Iꢀ ꢀ ꢀO (d ¼ 3.77 A) and Iꢀ ꢀ ꢀparene
(d ¼ 3.72 A) halogen
the apparent absence of strong non-covalent attractions (at
least according to distance criteria), the triangular motif
evident in Fig. 1 seems to constitute the dominant supramo-
lecular interaction that serves to position adjacent molecules
of 2 laterally to form two-dimensional layers. These interac-
tions differ significantly from halogen bonding exhibited by 1,
perhaps as a consequence of the electron-rich iodophenoxy
rings that attenuate the electrophilicity of the iodo groups but
increase the nucleophilicity of the arenes. The packing evident
in Fig. 1 outwardly resembles the layer structures observed in
Nangia and co-workers’ triphenoxy triazines, although these
latter networks are mediated by well-defined halogen trimer
synthons (type II halogen bonding).8
bonding (black lines). Hexagonal channels possess openings of B13 A
diameter. Disordered pyridine solvates shown with partial occupancy
for clarity.
except that the material crystallized in the centrosymmetric
space group P63/m. As shown in Fig. 3, 2 self-assembles to
form two-dimensional layers that stack in an ABAB pattern
resulting in generation of hexagonal channels parallel to the
c axis. As was the case with the CHCl3 solvate, layers of 2 in
2ꢀ3C5H5N exhibit bifurcated Iꢀ ꢀ ꢀp and Iꢀ ꢀ ꢀOQC interactions,
although in this instance the metrics of the respective interac-
tions are all identical as a consequence of crystalline symme-
try. Additionally, the closest Iꢀ ꢀ ꢀp contact is to the centroid of
a proximal arene rather than the midpoint of an arene p bond
(as is the case in the CHCl3 solvate). Pyridine molecules line
the hexagonal channels and are disordered over two sites.
While there is considerable precedent for pyridyl-Nꢀ ꢀ ꢀI halo-
gen bonds,4,6 there are no halogen bonding interactions
between the pyridine and aryl iodide groups in 2ꢀ3C5H5N.
The theoretical porosity of this structure was calculated to
be 39.7%.12
The 2D layers stack parallel to the c axis such that the
central 1,3,5-tricarbonylbenzene rings are positioned above
and below bifurcated halogen bonding trimer moieties and
every seventh layer is directly superimposed. This arrangement
produces large hexagonal channels that are filled with included
CHCl3 solvates (Fig. 2). The distance across these openings is
B1.3 nm and the channels extend uninterrupted along the
length of the c axis with the included CHCl3 molecules arrayed
in a helical fashion. The porosity of the network was deter-
mined to be 37.4%.12
Pyridine and chloroform, although disparate compounds,
appear to both play vital roles in templating formation of
hexagonal structures. Pyridine solvates are arrayed in a
C3-symmetrical arrangement within individual channels while
CHCl3 solvates are aligned around a six-fold helix. In light of
these observations, a series of co-crystallization experiments
was performed in which 2 and various three-fold symmetrical
co-crystallizing agents were combined in benzene.13 Examples
of co-crystallizing agents employed include 1,3,5-trihaloben-
zenes, triazine, mesitylene, trimethylamine-N-oxide, triphenyl-
phosphine, and trimethyl trimesoate. Almost all of these
crystallizations initially afforded hexagonal crystals isostruc-
tural to the 2ꢀ3C5H5N inclusion complex with disordered
benzene molecules taking the place of pyridine to the exclusion
of the co-crystallization additives.14z Continued contact with
the mother liquor, however, invariably resulted in slow dis-
appearance of the large hexagonal blocks concomitant with
the appearance of needle-shaped crystals (t ¼ days to weeks).
A similar sequence of events was also observed in crystals
obtained from pyridine. Determination of unit cell parameters
from randomly chosen samples from several crystallization
experiments revealed that these needles were isostructural with
crystals obtained directly from CH2Cl2. Material obtained in
this fashion was found to consist of a close-packed structure of
We attempted to crystallize 2 under other conditions in the
hope of obtaining analogous channel-type inclusion com-
plexes. Slow evaporation of a pyridine solution also deposited
large hexagonal crystals and X-ray diffractometry revealed a
channel-type inclusion complex of stoichiometry 2ꢀ3C5H5N.z
The structure was nearly identical to that of the CHCl3 solvate
Fig. 2 Extended packing in 2ꢀ3CHCl3 down c. Chloroform solvate
molecules reside in hexagonal channels of diameter B13 A.
ꢁc
This journal is The Royal Society of Chemistry 2008
Chem. Commun., 2008, 4726–4728 | 4727

View Article Online
We thank the Department of Chemistry, University of
Iowa. NPR thanks the US National Science Foundation
for funding the purchase of an APEXII diffractometer
(CHE-0420497).
Notes and references
z Crystallographic data. 2ꢀ3CHCl3: C30H18Cl9I3O6, M ¼ 1174.19,
hexagonal, P65, a ¼ 16.3917(3), b ¼ 16.3917(3), c ¼ 24.8007(12) A,
V ¼ 5770.9(3) A3, Z ¼ 6, Dc ¼ 2.027 g cmꢂ3, m ¼ 3.103 mmꢂ1, T (K) ¼
100(2), 178 452 reflections collected, 8874 unique, 7877 observed (I 4
2s(I)) reflections, 403 refined parameters, GOF ¼ 1.061, R1 ¼ 0.0541,
wR2 ¼ 0.1637, Flack x ¼ 0.48(4)—absolute structure could not be
determined, CCDC 688117. 2ꢀ3C5H5N: C42H30I3N3O6, M ¼ 1053.39,
hexagonal, P63/m, a ¼ 16.5590(18), b ¼ 16.5590(18), c ¼ 8.4752(9) A,
V ¼ 2012.6(4) A3, Z ¼ 2, Dc ¼ 1.738 g cmꢂ3, m ¼ 2.380 mmꢂ1, T (K) ¼
190(2), 29 038 reflections collected, 1640 unique, 1367 observed (I 4
2s(I)) reflections, 83 refined parameters, GOF ¼ 1.061, R1 ¼ 0.0470,
wR2 ¼ 0.1383, CCDC 688114. 2ꢀ1.5C6H6: C36H24I3O6, M ¼ 933.25,
hexagonal, P63/m, a ¼ 16.5473(17), b ¼ 16.5473(17), c ¼ 8.4380(9) A,
V ¼ 2000.9(4) A3, Z ¼ 2, Dc ¼ 1.549 g cmꢂ3, m ¼ 2.380 mmꢂ1, T (K) ¼
200(2), 53 859 reflections collected, 1700 unique, 1442 observed (I 4
2s(I)) reflections, 81 refined parameters, GOF ¼ 1.056, R1 ¼ 0.0385,
Fig. 4 View of the asymmetric unit in the close-packed structure of 2
and principal halogen bonding interactions: dIꢀ ꢀ ꢀp ¼ 3.48 A, C–Iꢀ ꢀ ꢀp ¼
174.91, dIꢀ ꢀ ꢀO ¼ 3.23 A, C–Iꢀ ꢀ ꢀO ¼ 142.31.
2.z A view of the two independent molecules in the asymmetric
unit along with their principal halogen bonding interactions is
shown in Fig. 4. Molecules in the asymmetric unit are part of
helical chains that intertwine to achieve close packing. Needle-
shaped crystals of 2 were more robust than their hexagonal
counterparts and DSC analysis revealed a single endothermic
transition corresponding to the melting point of 2 (see ESIw).15
Thus, it appears that the hexagonal crystals obtained from
pyridine and benzene are metastable products which are
eventually supplanted by the thermodynamically favored
close-packed form. Evidently, the combination of halogen
bonding, symmetry, and solvent inclusion in 2ꢀ3C5H5N and
2ꢀ1.5C6H6 is insufficient to maintain an open network, at least
under the crystallization conditions employed (i.e., room
temperature and pressure).
wR2
¼
0.1101, CCDC 688115. 2ꢀnon-solvated: C27H15I3O6,
M ¼ 816.09, monoclinic, P21/c, a ¼ 29.911(3), b ¼ 5.6474(6),
c ¼ 31.534(4) A, b ¼ 103.112(5)1, V ¼ 5187.8(10) A3, Z ¼ 8, Dc
¼
2.090 g cmꢂ3, m ¼ 3.655 mmꢂ1, T (K) ¼ 210(2), 39 978 reflections
collected, 11 892 unique, 8033 observed (I 4 2s(I)) reflections, 649
refined parameters, GOF ¼ 1.008, R1 ¼ 0.0369, wR2 ¼ 0.0761, CCDC
688116.
1 D. Maspoch, D. Ruiz-Molina and J. Veciana, Chem. Soc. Rev.,
2007, 36, 770; G. R. Desiraju, J. Mol. Struct., 2003, 656, 5; A.
Nangia, Curr. Opin. Solid State Mater. Sci., 2001, 5, 115.
2 D. V. Soldatov, J. Chem. Crystallogr., 2006, 36, 747.
3 K. E. Maly, E. Gagnon, T. Maris and J. D. Wuest, J. Am. Chem.
Soc., 2007, 129, 4306.
4 P. Metrangolo, G. Resnati, T. Pilati and S. Biella, Struct. Bonding,
2008, 126, 105.
5 K. E. Riley and K. M. Merz, J. Phys. Chem. A, 2007, 111, 1688; J.
P. M. Lommerse, A. J. Stone, R. Taylor and F. H. Allen, J. Am.
Chem. Soc., 1996, 118, 3108.
Wuest et al. have noted that maximization of both hydrogen
bonding and close-packing is mutually exclusive in certain organic
solids.3 This feature can lead to open crystalline architectures in
systems where hydrogen bonding interactions dominate. Halogen
bonding, though generally recognized as weaker than conven-
tional hydrogen bonding, may be capable of exerting a similar
influence over the crystallization process. In the system described
above, the combination of trigonal host 2 and appropriate solvent
guests produced inclusion complexes with nanometre-sized chan-
nels mediated by mutually reinforcing halogen bonding and
symmetry interactions. Spontaneous conversion of this solid state
network into a solvate-free close-packed structure provides evi-
dence that the open porous framework is thermodynamically less
stable. One can envision, however, that in the presence of stronger
halogen bonding interactions the thermodynamic preference for
close packing may be inverted in favor of porous architectures
(analogous to hydrogen bonded networks). Consequently, as the
phenomenon of halogen bonding becomes better understood, the
engineering of strong and directional halogen bonding interac-
tions should emerge as valuable design elements in the fabrication
of functional organic materials such as porous solids. Addition-
ally, the complementary nature of hydrogen and halogen bonding
has been noted,16 and the design of robust crystalline materials
that exhibit structurally defining and mutually reinforcing hydro-
gen/halogen bonding interactions is an attractive area for future
investigation.
6 Recent studies: S. Muniappan, S. Lipstman and I. Goldberg,
Chem. Commun., 2008, 1777; M. Vartanian, A. C. B. Lucassen,
L. J. W. Shimon and M. E. van der Boom, Cryst. Growth Des.,
2008, 8, 786; P. Metrangolo, F. Meyer, T. Pilati, D. M. Proserpio
and G. Resnati, Cryst. Growth Des., 2008, 8, 654; E. Cariati, A.
Forni, S. Biella, P. Metrangolo, F. Meyer, G. Resnati, S. Righetto,
E. Tordin and R. Ugo, Chem. Commun., 2007, 2590; L. Russo, S.
Biella, M. Lahtinen, R. Liantonio, P. Metrangolo, G. Resnati and
K. Rissanen, CrystEngComm, 2007, 9, 341; P. Metrangolo, F.
Meyer, T. Pilati, D. M. Proserpio and G. Resnati, Chem.–Eur. J.,
2007, 13, 5765.
7 Recent example: P. Metrangolo, F. Meyer, T. Pilati, G. Resnati
and G. Terraneo, Chem. Commun., 2008, 1635.
8 B. K. Saha, R. K. R. Jetti, L. S. Reddy, S. Aitipamula and A.
Nangia, Cryst. Growth Des., 2005, 5, 887.
9 F. C. Pigge, V. R. Vangala and D. C. Swenson, Chem. Commun.,
2006, 2123.
10 C.-H. Lee and T. Yamamoto, Bull. Chem. Soc. Jpn., 2002, 75, 615.
11 A. Bondi, J. Phys. Chem., 1964, 68, 441.
12 Calculated using PLATON: A. L. Spek, PLATON—A multipur-
pose crystallographic tool, University of Utrecht, The Netherlands,
1999.
13 For a similar strategy, see: V. S. S. Kumar, F. C. Pigge and N. P.
Rath, CrystEngComm, 2004, 6, 531.
14 Best fit of the diffraction data for these crystals was obtained using
stoichiometry 2ꢀ1.5C6H6.
15 Thermal analysis of 2ꢀ3C5H5N (DSC/TGA) has also been per-
formed (see ESIw).
¨
16 C. B. Aakeroy, M. Fasulo, N. Schultheiss, J. Desper and C.
Moore, J. Am. Chem. Soc., 2007, 129, 13772.
ꢁc
This journal is The Royal Society of Chemistry 2008
4728 | Chem. Commun., 2008, 4726–4728