COMMUNICATION [Link].

org/chemcomm | ChemComm

First snapshot of a nonpolymeric hydrogelator interacting with its

gelling solvents{
D. Krishna Kumar, D. Amilan Jose, Amitava Das* and Parthasarathi Dastidar*
Received (in Cambridge, UK) 17th May 2005, Accepted 17th June 2005
First published as an Advance Article on the web 13th July 2005
DOI: 10.1039/b506941f

A pyridyl urea based low molecular weight supramolecular Scheme 1) have been readily synthesized following a reported
hydrogelator has been synthesized; crystallized from its gelling procedure.6 While 1 shows remarkable gelation ability with pure
solvents, the single crystal structure of the gelator molecule water (Minimum Gelator Concentration, MGC 5 0.8 wt %), the
interacting with its gelling solvents reported herein is the first corresponding positional isomers 2 and 3 are nongelators. To the
example in the literature. best of our knowledge, compound 1 reported herein represents
the lowest molecular weight nonpolymeric urea based hydrogelator.
The majority of low molecular mass organic gelators (LMOGs) In a typical experiment, 1 is dissolved in pure water (1 wt %,
discovered to date display gelation ability with organic solvents.1 w/v) by heating. An opaque thermo-reversible gel is formed after a
LMOGs capable of gelling pure water or aqueous solvents are few minutes upon cooling at rt under ambient conditions.
known as hydrogelators.2 Hydrogels are an important class of However, due to its relatively poor solubility in pure water,
materials that display many interesting applications.3 The gelation experiments on 1 cannot be performed beyond 1.4 wt %
occurrence of LMOG derived hydrogelators in limited numbers (w/v) concentration of the gelator. Therefore, gelation study of 1 is
compared to their organogelator counterparts is presumably due performed in water containing a co-solvent, namely ethylene glycol
to the fact that a careful balance between hydrophobic interactions (EG). To estimate the thermal stability of the gel, the gel to sol
and hydrogen bonding in an aqueous solvent is essential to achieve melting temperature, Tgel{, is plotted against the gelator concen-
the required three-dimensional elastic networks involving small trations (wt % w/v) in 10% EG in water (v/v) and also against
gelator molecules within which the solvent molecules are varying EG concentration in water (% v/v) at a fixed gelator
immobilized. Therefore, designing a nonpolymeric LMOG based concentration of 1 wt % as depicted in Fig. 1.
hydrogelator is a challenging and daunting task. To design a The increase of Tgel with the increase of gelator concentration
gelator molecule, it is important to understand the supramolecular and also the remarkably low minimum gel concentration for 1
architecture (crystal structure) of the meta-stable gel fiber in its indicate that self-assembly in the gel state is driven by strong
native (gel) form. However, it is virtually impossible to determine intermolecular interactions. Tgel decreases with the increase of EG
the crystal structure of a gel fiber; only an indirect method using concentration, presumably due to the higher solubility of the
X-ray powder diffraction (XRPD) data may be applied.4 gelator in excess EG. FT-IR experiments have been carried out to
However, recording good quality XRPD data of the gel fibers in gain insights into the hydrogen bonding environment of the
their native form generally suffers from the scattering contribution carbonyl group (CLO) of the urea moiety in 1 under various
of the solvent molecules and the less crystalline nature of the gel conditions. In the bulk solid, the CLO band appears at 1740 cm21
fibers and therefore, in most cases, attempts to record XRPD of whereas 1735 cm21 is the band position in the gel (D2O) state.
gel fibers turn out to be a major disappointment. On the other
hand, correlating the single crystal structure of a molecule in its
thermodynamically more stable crystalline state with its gelling/
nongelling behaviour seems to be more practical and our efforts in
this regard have been quite useful in designing new LMOGs.5
Thus, it is clear that single crystal structure information on a
gelator molecule is as important. To the best of our knowledge, no Scheme 1
such data are available for a supramolecular hydrogelator.
Our previous experience with a pyridyl amide based hydro-
gelator5e prompted us to investigate pyridyl urea based compounds
as potential hydrogelators. Thus N,N9-bis(4-pyridyl) urea 1 and the
corresponding 3- and 2-pyridyl derivatives (2 and 3 respectively,

Analytical Science Discipline, Central Salt & Marine Chemicals

Research Institute, G. B. Marg, Bhavnagar, 364 002, Gujarat, India.
E-mail: parthod123@[Link]; dastidar@[Link];
amitava@[Link]; Fax: +91-278-2567562
{ Electronic supplementary information (ESI) available: analytical data
for 1 and 2, Tgel measurement, crystallographic details including data
collection, structure solution and refinement, hydrogen bonding para- Fig. 1 Plot of Tgel at various conditions. (¤) Tgel vs. varying EG
meters and ORTEP diagrams for 1?EG?H2O, 1?HCl?xH2O, 2 and concentration in water (v/v) at a fixed concentration of 1 wt % gelator. (&)
2?xH2O. See [Link] Tgel vs. gelator concentration; gel prepared in 10% EG/water (v/v).

This journal is ß The Royal Society of Chemistry 2005 Chem. Commun., 2005, 4059–4061 | 4059
Thus a 5 cm21 shift in the CLO band may be indicative of further
hydrogen bonding of the carbonyl groups in the gel state.
The gelation behavior of 1 is quite pH sensitive. The pH of the
aqueous solution of 1 is found to be 8.3 at MGC. When the pH is
changed to 5.3 using 1% AcOH (v/v), it fails to form gel. On the
other hand, when gelation experiments are performed at pH 9–10
using NaHCO3, gels are formed. These results clearly indicate that
the free pyridine nitrogen atoms might be contributing towards the
required self-assembly of the gelator molecules through possible
hydrogen bonding. To provide further support for this hypothesis,
a monohydrochloride salt of the gelator 1 has been synthesized
and it also turns out to be a nongelator. This indicates that both the Fig. 3 Various hydrogen bonding interactions of the gelator with the
ring nitrogen atoms of the gelator 1 must be free from protonation gelling solvents as observed in the single crystal structure of 1?EG?H2O.
in order to form gel and also the relative arrangement of the ring
nitrogen atoms in the molecule must be linear since the positional also participating in hydrogen bonding interactions with water and
isomers, namely 2 and 3 fail to show any gelation property. the EG molecule (N…OHO 2
5 2.774 s, /N…H–OHO 2
5 177.1u;
SEM pictures of xerogel of 1 in pure water (Fig. 2a) and in 10% … …
N OEG 5 2.718 s, /N H–OEG 5 176.7u) (Fig. 3).
EG/water (v/v) (Fig. 2b) display a typical fibrous network. The The supramolecular assembly of the gelator molecule and the
morphology of the several mm long slender fibers in pure water interacting solvents in the crystal lattice displays microporous
appears to be rod shaped with varying thickness, whereas highly architecture. A continuous channel is formed along the crystal-
intertwined tape type fibers are observed in the EG/water system. lographic a-axis and the solvent molecules (both water and EG)
Understandably, the solvent molecules are immobilized in such a are firmly held in the channel through hydrogen bonding
network of fibers resulting in gel formation. interactions with the gelator molecules (Fig. 4).
Analysis of the single crystal structure of a gelator molecule To the best of our knowledge, the present crystal structure is the
provides useful information about the detail of the intermolecular first example wherein the LMOG based hydrogelator molecule is
interactions which might be responsible for gel fiber formation.5,7 shown to interact with its gelling solvents.
However, crystallizing a gelator molecule is a daunting task. Attempts to record XRPD patterns of the gel fiber in its native
Moreover, crystallizing a gelator molecule from its gelling solvents state have proved unsuccessful due to a heavy scattering
is even more difficult and to the best of our knowledge, no such contribution arising from the solvent molecules. XRPD patterns
example in supramolecular hydrogelators is known. Efforts to of the xerogel (dried gel) prepared under various conditions are
grow a single crystal of 1 from pure water have proved inconsistent presumably due to inhomogeneous solvent loss from
unsuccessful. However, 1 can be crystallized from its gelling the crystal lattice during xerogel formation. The XRPD pattern of
solvent systems such as EG/water (1 : 9 and 1 : 1, v/v) when the gel the xerogel recorded after drying the sample at 120 uC for 6 h
is kept at rt under ambient conditions in an open test tube for shows a high degree of noncrystallinity. Based on these results, it is
about a month. The crystal structure of 1 thus obtained from its reasonable to assume that the gelling solvent molecules might be
gelling solvents is found to contain both EG and water molecules! included in the crystal lattice of the gel fibers (in the native state)
The crystal 1?EG?H2O belongs to a noncentrosymmetric space and escape of the solvent molecules from the crystal structure of
group (orthorhombic, P212121) and the asymmetric unit contains the gel fibers during xerogel formation results in the partial
one gelator molecule, one EG and one water.{ The hydrogen collapse of the crystal lattice and thus a high degree of
bonding sites of the gelator molecule are fully satisfied through noncrystallinity of the XRPD pattern of the xerogel (Figure S1).{
various hydrogen bonding interactions with the solvent molecules. Since the gelation property of 1 is pH sensitive, it was considered
The urea moiety does not display the usual propagating worthwhile to examine the single crystal structure of the
hydrogen bonding network involving the CLO group and NH monoprotonated gelator molecule 1 in the form of its hydro-
hydrogen atoms;8 instead the carbonyl group is involved in chloride salt. Thus, a monohydrochloride salt of 1 was crystallized
hydrogen bonding with a water molecule (O…O 5 2.945 s, from a solution of 1 in 0.01 M HCl and its single crystal structure
/O–H…O 5 148.2u) and the urea nitrogen atoms are interacting was investigated. It was observed that the crystal structure of
with oxygen atoms of the EG molecule (N…O 5 2.805–2.864 s; 1?HCl?xH2O (x 5 1.66) belongs to a centric triclinic space group
/N–H…O 5 160.0–173.8u). Both the pyridine nitrogen atoms are

Fig. 4 Overall supramolecular assembly of the gelator molecule display-

Fig. 2 SEM micrographs of xerogels of a) 1 in pure water at MGC ing microporous channel architecture. The space filling model on the right
(bar 5 20 m); b) 1 in 10% EG/water (v/v) at 1.2 wt % (bar 5 100 m). shows one such channel. Hydrogen atoms are omitted for clarity.

4060 | Chem. Commun., 2005, 4059–4061 This journal is ß The Royal Society of Chemistry 2005
(P-1).{ In the asymmetric unit, two monoprotonated urea, three The Department of Science & Technology, New Delhi and the
water molecules and two Cl2 ions are located in general positions. Ministry of Environment and Forests, New Delhi are gratefully
The excess electron densities (two peaks) amounting to y1.2 e/s3 acknowledged for financial support. D.K.K. and D.A.J. acknowl-
have been assigned as disordered oxygen atoms presumably arising edge CSIR for an SRF fellowship.
from solvent water molecules.{ The monoprotonated gelator
molecules are held together by a N–H…N hydrogen bond Notes and references
(N…N 5 2.737 s; /N–H…N 5 176.8u) resulting in 1-D
hydrogen bonded chains which are further arranged in a parallel { Crystal data for 1?EG?H2O: C13H18N4O4, FW 5 294.31, Orthorhombic,
P212121, a 5 8.1836(7), b 5 10.6011(10), c 5 16.4345(16) s,
fashion held together by various hydrogen bonding interactions V 5 1425.8(2) s3, Z 5 4, Dc 5 1.371 g cm23, F(000) 5 624, T 5 100 K.
involving solvent water molecules and Cl2 ions (Figure S2).{ Thus Final residuals (for 258 parameters) were R1 5 0.0360 for 1807 reflections
the single crystal structure of 1?HCl?xH2O clearly demonstrates with I . 2s(I), and R1 5 0.0402, wR2 5 0.0894, GOF 5 1.051 for all 1956
that protonation of one of the ring nitrogens favours intermole- reflections. Crystal data for 1?HCl?xH2O (x 5 1.66): C22H28Cl2N8O5.32,
FW 5 560.54, Triclinic, P-1, a 5 9.8236(12), b 5 10.2628(12),
cular hydrogen bonding through N–H…N interactions thereby c 5 13.9993(17) s, a 5 101.696(2), b 5 95.602(2), c 5 101.734(2)u,
preventing the molecule from forming hydrogen bonds with the V 5 1338.9(3) s3, Z 5 2, Dc 5 1.390 g cm23, F(000) 5 585, T 5 100 K.
solvent molecules involving ring nitrogen atoms. Final residuals (for 382 parameters) were R1 5 0.0493 for 4359 reflections
It is remarkable that by going from 4-pyridyl to 3-pyridyl and with I . 2s(I), and R1 5 0.0718, wR2 5 0.1639, GOF 5 1.113 for all 5734
reflections. Crystal data for 2: C11H10N4O, FW 5 214.23, Orthorhombic,
2-pyridyl derivatives (2 and 3 respectively), the gelation ability Aba2, a 5 13.713(6), b 5 6.977(3), c 5 10.013(4) s, V 5 957.9(7) s3, Z 5 4,
disappears and therefore, it is important to study the molecular Dc 5 1.485 g cm23, F(000) 5 448, T 5 100 K. Final residuals (for 94
packing of the nongelators, if possible. Two different crystal parameters) were R1 5 0.0348 for 583 reflections with I . 2s(I), and
R1 5 0.0355, wR2 5 0.0868, GOF 5 1.099 for all 592 reflections. Crystal
structures of 2, one devoid of any solvent molecule (crystallized
data for 2?xH2O (x 5 1.33): C16.50H19N6O3.50, FW 5 357.38, Monoclinic,
from EtOH/water 1 : 1 v/v) and one with one water molecule C2/c, a 5 13.073(4), b 5 11.793(7), c 5 22.678(8) s, b 5 100.03(3)u,
included in the crystal lattice, 2?xH2O (x 5 1.33) (crystallized from V 5 3443(3) s3, Z 5 8, Dc 5 1.379 g cm23, F(000) 5 1504, T 5 298 K.
MeOH), have been obtained. The urea moiety in 2 is located on a Final residuals (for 276 parameters) were R1 5 0.0420 for 1494 reflections
with I . 2s(I), and R1 5 0.0780, wR2 5 0.1039, GOF 5 0.976 for all 2239
2-fold axis (orthorhombic, Aba2).{ The molecules in 2 are held
reflections. Details of data collection, structure solution and refinement for
together by N–H…N hydrogen bonding involving pyridine all the crystals are given in ESI. CCDC 267563–267566. See http://
ring nitrogen and urea nitrogen atoms (N…N 5 3.056 s; [Link]/10.1039/b506941f for crystallographic data in CIF or other
/N–H…N 5 166.9u) resulting in 2-D nonplanar sheets which are electronic format.
further packed in a herringbone fashion in the crystal lattice
1 For excellent reviews on LMOGs see: (a) P. Terech and R. G. Weiss,
(Figure S3).{ On the other hand, the hydrated crystal of 2 displays
Chem. Rev., 1997, 97, 3133; (b) D. J. Abdallah and R. G. Weiss, Adv.
a crystal structure belonging to a monoclinic centrosymmetric Mater., 2000, 12, 1237; (c) O. Grownwald and S. Shinkai, Curr. Opin.
space group C2/c, the asymmetric unit of which contains one Colloid Interface Sci., 2002, 7, 148; (d) J. H. van Esch and B. L. Feringa,
molecule of 2 and two molecules of water in general positions and Angew. Chem., 2000, 112, 2351; (e) J. H. van Esch and B. L. Feringa,
Angew. Chem., Int. Ed., 2000, 39, 2263.
half a molecule of 2 in a crystallographic 2-fold axis.{ The overall
2 L. A. Estroff and A. D. Hamilton, Chem. Rev., 2004, 104, 1201 and
assembly can best be described as a 3-D hydrogen bonded network references cited therein.
involving the urea and water molecules (Figure S4).{ The crystal 3 (a) F. Ilmain, T. Tanaka and E. Kokufuta, Nature, 1991, 349, 400; (b)
structure of 3 (reported recently in a different context9) shows J. H. Holtz and S. A. Asher, Nature, 1997, 389, 829; (c) L. Simonsen,
L. Hovgaard, P. B. Mortensen and H. Brondsted, Eur. J. Pharm. Sci.,
hydrogen bonded dimers which are packed in the crystal lattice
1995, 3, 329; (d) Y. Osada and J. P. Gong, Adv. Mater., 1998, 10, 827; (e)
further by van der Waals forces (Figure S5).{ It is noteworthy that J. M. Weissman, H. B. Sunkara, A. S. Tse and S. A. Asher, Science, 1996,
all the nongelator single crystal structures, namely 1?HCl?xH2O, 2, 274, 959; (f) A. K. Andrianov, S. Cohen, K. B. Visscher, L. G. Payne,
2?xH2O and 3 do not display any microporous architecture as H. R. Allcock and R. J. Langer, J. Controlled Release, 1993, 27, 69; (g)
Y. Osada and A. R. Khokhlov, Polymer Gels and Networks; Marcel
observed in the solvated gelator structure 1?EG?H2O. It therefore
Dekker, New York, 2002; (h) Y. H. Chu, J. K. Chen and G. M.
appears that effective hydrogen bonding with the gelling solvent Whitesides, Anal. Chem., 1993, 65, 1314; (i) K. Lee and S. A. Asher,
molecules is important in the present example for gelation. J. Am. Chem. Soc., 2000, 122, 9534; (j) J. C. Tiller, Angew. Chem., Int.
In summary, we have reported the first crystal structure Ed., 2003, 42, 3072.
4 E. Ostuni, P. Kamaras and R. G. Weiss, Angew. Chem., Int. Ed. Engl.,
description of a supramolecular hydrogelator interacting with its 1996, 35, 1324; E. Ostuni, P. Kamaras and R. G. Weiss, Angew. Chem.,
gelling solvents. It is also the first report of a nonpolymeric 1996, 108, 1423.
hydrogelator crystallized from its gelling solvents. It may be 5 (a) D. Krishna Kumar, D. A. Jose, P. Dastidar and A. Das, Langmuir,
reasonable to assume, based on the XRPD data of both gel and 2004, 20, 10413; (b) D. R. Trivedi, A. Ballabh and P. Dastidar, Chem.
Mater., 2003, 15, 3971; (c) A. Ballabh, D. R. Trivedi and P. Dastidar,
xerogel (see text), that the gel fibers in the native state might have a Chem. Mater., 2003, 15, 2136; (d) D. R. Trivedi, A. Ballabh, P. Dastidar
similar (may not be identical) microporous supramolecular and B. Ganguly, Chem.–Eur. J., 2004, 10, 5311; (e) D. Krishna Kumar,
environment wherein the gelling solvent molecules are embedded. D. A. Jose, P. Dastidar and A. Das, Chem. Mater., 2004, 16, 2332; (f)
While we recognize the importance of the identification of the P. Dastidar, S. Okabe, K. Nakano, K. Iida, M. Miyata, N. Tohnai and
M. Shibayama, Chem. Mater., 2005, 17, 741.
crystal morph and crystal structure of the gel fiber in its native (gel) 6 D. B. Grotjahn and C. Joubran, Tetrahedron: Asymmetry, 1995, 6, 745.
state, the single crystal structure information on the thermo- 7 (a) R. Luboradzki, O. Gronwald, M. Ikeda, S. Shinkai and
dynamically stable crystalline state of the gelator is as important if D. N. Reinhoudt, Tetrahedron, 2000, 56, 9595; (b) S.-I. Tamaru,
an alternative and more practical approach to correlating the R. Luboradzki and S. Shinkai, Chem. Lett., 2001, 336.
8 M. D. Hollingsworth and K. D. M. Harris, in Comprehensive
crystal structure of a molecule with its gelling/nongelling property Supramolecular Chemistry, Vol. 6, ed. D. D. MacNicol, F. Toda and
is to be probed. The results reported herein provide such attempts R. Bishop, Pergamon, Oxford, 1996, pp. 177.
for the first time in a nonpolymeric hydrogelator. 9 P. S. Corbin and S. C. Zimmerman, J. Am. Chem. Soc., 2000, 122, 3779.

This journal is ß The Royal Society of Chemistry 2005 Chem. Commun., 2005, 4059–4061 | 4061