Transition from low molecular weight non-gelating oligo(amide-triazole)s to a restorable, halide-responsive poly(amide-triazole) supramolecular gel

Article abstract of DOI:10.1039/c3cc49323g

A self-assembled poly(amide-triazole) physical gel (1) was found to show responsive behaviour towards halide anions, while the corresponding monomeric (2) and dimeric homologues (3) were non-gelating. In the presence of halide anions, polymer gel 1 collapsed to become a solution, but could be restored back to the gel state after addition of a AgNO₃ salt. The Royal Society of Chemistry.

Full text of DOI:10.1039/c3cc49323g

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Transition from low molecular weight non-
gelating oligo(amide-triazole)s to a restorable,
halide-responsive poly(amide-triazole)
supramolecular gel†
Cite this: Chem. Commun., 2014,
50, 3064
Received 9th December 2013,
Accepted 7th January 2014
Siu-Lung Yim,ab Hak-Fun Chow*ab and Man-Chor Chana
DOI: 10.1039/c3cc49323g
www.rsc.org/chemcomm
A self-assembled poly(amide-triazole) physical gel (1) was found or amide NH did not possess gelating properties. Since triazole-rich
to show responsive behaviour towards halide anions, while the compounds are known to interact with anions via H bonding,6
corresponding monomeric (2) and dimeric homologues (3) were therefore we were interested to investigate the anion binding proper-
non-gelating. In the presence of halide anions, polymer gel 1 ties of polymer 1. Herein we report a rare example that polymer 1
collapsed to become a solution, but could be restored back to based on the amide-triazole repeating unit was found to be a halide-
the gel state after addition of a AgNO3 salt.
responsive physical gel, whereas its monomeric 2 and dimeric 3
homologues possessing the same repeating unit were found to be
Supramolecular self-assembled gels that show responsive behaviour devoid of gelation properties.7 Furthermore, physical gel 1 was found
towards external stimuli are highly interesting compounds because to collapse and become a solution after treatment with halide anions,
they can function as molecular sensors, diagnostic agents and but could be restored to its gel form upon addition of silver ions.
materials for drug delivery and tissue engineering.1 These gelating
materials are generally prepared from low molecular weight (LMW)
compounds which form thermo-reversible physical gels due to non-
covalent self-assembly.2 On the other hand, polymer chemical gels,
in which the gel network is constructed via covalent crosslinking,
generally lack responsive properties towards external stimuli as
the network structure is very stable as the gelation process is
irreversible.3 Hence, polymer gels that exhibit responsive properties
are rare. In the literature, there are only a few examples of stimuli-
responsive polymer physical gels, in which the gelation is based on
non-covalent interactions.4 We earlier reported the synthesis
and thermo-reversible organogelating property of polymer 1.5 In
our previous study attention was focused mainly on the conforma-
tional rigidity of the intramolecularly H bonded pyridine-2,6-
dicarboxamide unit on the polymerization efficiency of click
polymerization. It was found that the presence of such rigid units
in the polymer backbone not only produced polymers with higher
In order to have a better understanding of the anion binding
polymerization efficiency, but also conferred them with gelation capability of polymer 1, its LMW homologues 2–4 were prepared.
properties. Hence, structurally similar polymers lacking this kind of Compounds 2 and 3 actually represent the corresponding monomer
conformational rigidity due to the removal of the pyridine nitrogen and dimer of polymer 1, while 4 is the monomeric analogue without
the triazole units in the binding motif. The synthesis and structural
a Department of Chemistry, The Chinese University of Hong Kong, Shatin, HKSAR.
characterization of LMW homologues 2–4 are described in the
ESI.† In contrast to literature precedents,8 none of these LMW
compounds were organogelators, but they were all shown to interact
with halide anions. The host–guest chemistry of 2 and ClÀ was
investigated by 1H NMR titration experiments involving the addition
of tetrabutylammonium chloride (TBACl) to a [D8]-THF solution
of 2 (2.6 mM), and significant downfield shifts of the amide
E-mail: hfchow@cuhk.edu.hk
b The Center of Novel Functional Molecules and Institute of Molecular Functional
Materials, UGC-AoE Scheme, The Chinese University of Hong Kong, Shatin,
HKSAR
† Electronic supplementary information (ESI) available: Synthesis and character-
ization data for all new compounds; determination of binding constants. See
DOI: 10.1039/c3cc49323g
3064 | Chem. Commun., 2014, 50, 3064--3066
This journal is ©The Royal Society of Chemistry 2014

View Article Online
Communication
ChemComm
of halide anions. As compound 2 showed the strongest association
towards ClÀ, our subsequent investigations were then focused only
on this anion.
Our next attention was then turned to the dimeric receptor 3 with
two binding sites, which could be considered as the simplest proto-
type that possessed multiple binding sites. In particular, it was
interesting to know if the two binding sites could bind two halide
anions independently or whether allosteric binding can occur. As
expected, the amide NH and triazole CH resonance signals showed a
similar downfield shift upon addition of TBACl.9 A Job plot analysis
showed that 3 bound to ClÀ in a 1 : 2 stoichiometry.9 The chemical
shift values of amide NH were then fitted to a 1 : 2 binding model,9,10
and the K1 and K2 were found to be 100 Æ 10 and 740 Æ 120 MÀ1
,
respectively. The K1 value should not be directly compared to the Ka
value of the monomer 2–halide complex, as K1 is affected by the
second binding process. In principle, the interaction parameter a,
defined as 4K2/K1, should be o1, = 1, 41, for positive allosteric, non-
cooperative, and negative allosteric bindings, respectively.10 In our
case, the a value was 30, indicating that binding of the first ClÀ
enhanced the binding of the second.12 A Hill plot was constructed
and the Hill coefficient was 2.0,9 reinforcing the fact that positive
allosteric binding was indeed happening. At this moment, we were
uncertain of the mechanism of such positive allosteric interaction
since the two binding sites appeared to be isolated from each other.
One possible explanation could be an induced structural reorganiza-
tion of the second binding site after the first ClÀ binding, providing
better shape complementarity to the second ClÀ binding.
Fig. 1 Stacked partial 1H NMR spectra ([D8]-THF, 700 MHz) of compound
2 upon addition of TBACl.
Due to the inherent structural heterogeneity of polymer 1, it
was difficult to obtain its binding constant towards halide
anions. Nonetheless, based on the study on the dimeric species
3, it was believed that positive allosteric binding should also exist
in the polymer system. Similar to the findings for oligomers 2
and 3, 1H NMR titration experiments revealed significant down-
field shifts of the amide NH and triazole CH signals upon
addition of TBACl to a [D8]-THF solution of 1 (Fig. 3). Further-
more, it was found that the polymer–chloride complex gave
sharper 1H resonance signals than that of pure polymer 1 itself,
suggesting the breakdown of interpolymer chain H bonding
interaction upon ClÀ binding.
Fig. 2 Space filling models of 2ÁClÀ TBA+ (left) and 3Á2ClÀ 2TBA+ (right)
(grey: C; white: H, red: O; blue: N; green: Cl). The alkoxy side chains were
omitted for clarity.
NH (Dd ~1.5 ppm) and triazole CH (Dd B 0.4 ppm) signals were
Interestingly, treatment of a 2% w/v polymer toluene gel 1 with
observed (Fig. 1), indicating that both of them were interacting with TBA salts of halide (Cl, Br and I) triggered a gel-to-sol transition. It was
the ClÀ. The corresponding Job plot confirmed that the binding found that 0.2 equiv. (with respect to the total number of binding
stoichiometry was 1 : 1.9 Hence, it was proposed that 2 folded into a sites) of TBACl, 0.3 equiv. of TBABr, or 0.5 equiv. of TBAI, were
hairpin-like structure to create a ‘‘binding site’’ for ClÀ (Fig. 2). The
chemical shift values of the amide NH signal were then fitted to a
1:1 binding model,10 and the binding constant (Ka) of 2 for ClÀ was
found to be 1930 Æ 150 MÀ1.9 Similarly, the Ka values of 2 for BrÀ
and IÀ were determined to be 310 Æ 10 MÀ1 and 31 Æ 2 MÀ1
,
respectively, upon titration with TBABr and TBAI.9 These values were
comparable to those of other open-chain analogues reported
before.11 The gradual decrease of the binding constants towards
the larger halide anions suggested that the binding pocket had a
relatively small cavity, and was best fitted to the smallest ClÀ.
Incidentally, compound 4 without the two triazole units also binds
to ClÀ, albeit with a lower binding constant (1530 Æ 50 MÀ1).9 This
implied that the presence of the triazole CHs enhanced the binding
Fig. 3 Stacked partial 1H NMR ([D8]-THF, 400 MHz) spectra of 1 upon
addition of TBACl.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 3064--3066 | 3065

View Article Online
ChemComm
Communication
polymer regained its abilities to re-form the gelation network
via interpolymer chain H bonding. Unfortunately, it was found
that such reversible halide-responsive gelation behavior could
not be repeated for the second time as the insoluble AgCl and
TBAX salts perturbed subsequent gel formation.
In summary, we reported a rare example of a restorable, halide
anion-responsive polymer physical gel 1 based on reversible H
bonding interactions. This work also highlights a few advantages
of using polymer-based materials for chemo-sensing applications.
First, polymers may exhibit properties (e.g. gelation as in this case)
whereas their LMW oligomeric homologues do not. Second, the
presence of multiple binding sites in a polymer can promote positive
allosteric binding affinity towards guest molecules and can improve
its sensitivity. Both of these are useful attributes associated with
polymer-based sensors and carrier systems.
Fig. 4 (top) Gel-to-sol responsive behaviour of polymer toluene gel 1
upon addition of TAB salts of halide. (bottom) Sol-to-gel responsive
behaviour of polymer 1–ClÀ complex upon addition of a AgNO3 salt.
required for the complete breakdown of the gel (Fig. 4, top), in which
the whole process was completed within hours.
FTIR studies were then performed to reveal the H bonding
environment of the oligomer–halide and the polymer–halide
complexes.9 Upon addition of one equiv. of TBACl to a THF solution
of monomer 2, the NH stretching frequency was significantly red
shifted (from 3315 to 3170 cmÀ1). A similar red shift of the NH
stretching frequency (from 3313 to 3182 cmÀ1) was also noted when
TBACl was added to polymer 1 in THF. These observations indicated
the weakening of the NH stretching due to the formation of the
H bond between the amide NH and ClÀ. Moreover, there was almost
no change in the CQO stretching frequencies upon addition of
TBACl, which provided evidence that ClÀ binding was via the NH
but not the CQO functionality. Unfortunately, we were unable to
monitor the triazole CH stretching due to background absorption of
the THF solvent (2700–3150 cmÀ1).
The halide-responsive property of polymer gel 1 could be ration-
alized by replacement of the inter-polymer chain H bonding network
with the NHÁÁÁhalide and triazole–CHÁÁÁhalide H bonding complex,
which resulted in the segregation of individual polymer chains.
Incidentally, SEM images of the air-dried 1–chloride complex
showed segregated polymer fibers clearly, which provided
additional evidence of the proposed breakdown of the gelating
network upon ClÀ complexation (Fig. 5).
The work was supported by grants from the RGC (CUHK
400810) and partially from the UGC (AoE/P-03/08) of HKSAR.
Notes and references
1 (a) M. Barboiu, S. Cerneaux, A. van der Lee and G. Vaughan, J. Am. Chem.
Soc., 2004, 126, 3545; (b) H. Maeda, Chem.–Eur. J., 2008, 14, 11274;
(c) M.-O. M. Piepenbrock, G. O. Lloyd, N. Clarke and J. W. Steed, Chem.
Rev., 2009, 110, 1960; (d) G. O. Lloyd and J. W. Steed, Nat. Chem., 2009,
1, 437; (e) M.-O. M. Piepenbrock, G. O. Lloyd, N. Clarke and J. W. Steed,
Chem. Rev., 2010, 110, 1960; ( f ) J. W. Steed, Chem. Soc. Rev., 2010,
´
39, 3686; (g) D. D. Dıaz, D. Ku¨hbeck and R. J. Koopmans, Chem. Soc. Rev.,
2011, 40, 427; (h) G. O. Lloyd, M.-O. M. Piepenbrock, J. A. Foster, N. Clarke
and J. W. Steed, Soft Matter, 2012, 8, 204; (i) L. Marin, B. Simionescu and
M. Barboiu, Chem. Commun., 2012, 48, 8778.
2 For reviews and monographs on LMW gelators, see (a) N. M. Sangeetha
and U. Maitra, Chem. Soc. Rev., 2005, 34, 821; (b) Molecular Gels.
Materials with Self-Assembled Fibrillar Networks, ed. R. G. Weiss and
P. Terech, Springer, Dordrecht, The Netherlands, 2005; (c) M. George
and R. G. Weiss, Acc. Chem. Res., 2006, 39, 489; (d) P. Dastidar, Chem.
Soc. Rev., 2008, 37, 2699.
3 Polymer chemical gels that show volume change had been reported.
However, such gels seldom turn into solutions upon stimulation, see
Y. Takashima, S. Hatanaka, M. Otsubo, M. Nakahat, T. Kakuta,
A. Hashidzume, H. Yamaguchi and A. Harada, Nat. Commun., 2012, 3, 1270.
4 For examples, see (a) F. Peng, G. Li, X. Liu, S. Wu and Z. Tong, J. Am. Chem.
Soc., 2008, 130, 16166; (b) S. Tamesue, Y. Takashima, H. Yamaguchi,
S. Shinkai and A. Harada, Angew. Chem., Int. Ed., 2010, 49, 7461;
(c) M. Nakahata, Y. Takashima, H. Yamaguchi and A. Harada, Nat.
Commun., 2011, 2, 511; (d) H. Yamaguchi, Y. Kobayashi, R. Kobayashi,
Y. Takashima, A. Hashidzume and A. Harada, Nat. Commun., 2011, 3, 603;
(e) E. A. Appel, J. del Barrio, X. J. Loh and O. A. Scherman, Chem. Soc. Rev.,
2012, 41, 6195; ( f) M. Zhang, D. Xu, X. Yan, J. Chen, S. Dong, B. Zheng and
F. Huang, Angew. Chem., Int. Ed., 2012, 51, 7011.
The gelation power of the polymer–halide complex could be
restored by competitive removal of the ClÀ. Hence, upon
addition of a slight excess (1.5 equiv. with respect to ClÀ) of
AgNO3 powder to a 2% w/v polymer–halide complex solution in
toluene, a translucent gel was formed after a warming and
cooling process (Fig. 4, bottom). It was suggested that the Ag+
ion removed the ClÀ from the polymer–ClÀ complex, thus the
5 S.-L. Yim, H.-F. Chow, M.-C. Chan, C.-M. Che and K.-H. Low,
Chem.–Eur. J., 2013, 19, 2478.
6 Y. Hua and A. H. Flood, Chem. Soc. Rev., 2010, 39, 1262.
7 The organogelating property of the polymer 1, in contrast to the
non-gelating properties of the lower oligomers 2 and 3, was due to a
synergistic H bonding effect after polymerization, see (a) K.-N. Lau,
H.-F. Chow, M.-C. Chan and K.-W. Wong, Angew. Chem., Int. Ed.,
2008, 47, 6912; (b) J. Zhang, H.-F. Chow, M.-C. Chan, G. K.-W. Chow
and D. Kuck, Chem.–Eur. J., 2013, 19, 15019.
8 H.-F. Chow and G.-X. Wang, Tetrahedron, 2007, 63, 7407.
9 See ESI† for details.
10 P. Thordarson, Chem. Soc. Rev., 2011, 40, 1305.
11 (a) H. Juwarker, J. M. Lenhardt, J. C. Castillo, E. Zhao,
S. Krishnamurthy, R. M. Jamiolkowski, K.-H. Kim and S. L. Craig, J. Org.
Chem., 2009, 74, 8924; (b) V. Haridas, S. Sahu and P. Venugopalan,
Tetrahedron, 2011, 67, 727.
12 The data were also fitted to a non-cooperative 1 : 2 binding model,
with K1 fixed as 4 Â K2. However, the fitting result was far from
satisfactory. See ESI† for details.
Fig. 5 SEM image of a freeze-dried sample of 1 at 12 K magnification (left) and
an air-dried polymer 1–ClÀ complex from toluene at 200 K magnification (right).
3066 | Chem. Commun., 2014, 50, 3064--3066
This journal is ©The Royal Society of Chemistry 2014