Fluorescent hydrogel formation from carboxyphenyl-terpyridine

Article abstract of DOI:10.1039/c2cc37842f

We report the analysis of a novel terpyridine based supramolecular hydrogel which shows fluorescent properties in the gel but not in the sol form; the gel forms in a narrow pH range in aqueous solutions specifically in the presence of sodium ions, and con

Full text of DOI:10.1039/c2cc37842f

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Fluorescent hydrogel formation from
carboxyphenyl-terpyridine†
Cite this: Chem. Commun., 2013,
49, 731
Ashleigh Griffith, Thomas J. Bandy, Mark Light and Eugen Stulz*
Received 29th October 2012,
Accepted 28th November 2012
DOI: 10.1039/c2cc37842f
www.rsc.org/chemcomm
We report the analysis of a novel terpyridine based supramolecular
hydrogel which shows fluorescent properties in the gel but not in
the sol form; the gel forms in a narrow pH range in aqueous
solutions specifically in the presence of sodium ions, and contains
between 98% and 99.2% water.
Scheme 1 Synthesis of carboxy-terpyridine 1.
Terpyridines (terpys) are well known ligands for transition metal
complexes, which find wide applications in supramolecular chem-
istry, dye sensitisation of photovoltaic cells, and in catalysis. The
large diversity of physical properties such as luminescence and
electrochemistry can be tuned by both changing the metal and the
substitution pattern on the terpy ligand, making terpyridines
very suitable building blocks in supramolecular chemistry.1 In
connection with other photoactive electron donor and acceptor
moieties such as porphyrins,2 terpy complexes have been shown to
provide suitable systems for charge separation in artificial photo-
synthesis. In addition, a few specifically substituted terpyridine
complexes have previously been shown to be good gelators in
aqueous solutions, where metal complexation aids formation
of the supramolecular gel network.3 Hydrogels4 have attracted
general interest, for example due to their applications in
medicine5 for drug delivery,6 as biological matrices for enzyme
immobilisation7 or in electrochemical capacitors.8 However,
the conditions for gel formation seem to depend strongly on
several parameters such as substituents, concentration and
salts used for all compounds investigated. We have found that
40-para-phenylcarboxyl-2,20:60,200-terpyridine 1 is able to form a
reversible supramolecular hydrogel, which was observed during
crystallisation attempts and is rather a serendipitous finding.
The gel formed is opaque white and fluorescent under UV light,
is thermally reversible and only formed in a narrow range of pH
and terpyridine concentrations.
The synthesis of 1 was performed according to literature
procedures (Scheme 1)9 from p-carboxy benzaldehyde 2 and
2-acetyl pyridine 3 in the presence of ammonia. After hydrolysis
of the methyl ester using aqueous KOH and acidic workup, 1
readily forms a gel under specific conditions. Investigating the
gelation further, a full set of conditions systematically varying
the pH and the concentration of terpyridine 1 yielded a phase
diagram which showed a very well defined set of conditions
where the gel phase forms.† Both the concentration of sodium
hydroxide and the concentration of the terpyridine are critical
to gel formation. The gelation phase is defined between 0.5 and
0.75 M NaOH, and 8 to 20 mg mLꢀ1 of the terpyridine in water,
as shown in Fig. 1. Outside these boundaries, 1 either remains
in solution or forms an insoluble precipitate with no visible
higher order structure. This corresponds to a minimum water
School of Chemistry, University of Southampton, Highfield, Southampton SO17 1BJ,
UK. E-mail: est@soton.ac.uk; Tel: +44 (0)2380 599 369
† Electronic supplementary information (ESI) available: Experimental details of
synthesis, gelation, metallation and crystallisation of the terpyridine, SEM
pictures and analytical data of the gels. CCDC 909638. For ESI and
crystallographic data in CIF or other electronic format see DOI: 10.1039/
c2cc37842f
Fig. 1 Phase diagram of 1, with conditions for forming clear solutions (’), gels
(K) or insoluble aggregates (m).
c
This journal is The Royal Society of Chemistry 2013
Chem. Commun., 2013, 49, 731--733 731

View Article Online
Communication
ChemComm
content of 98%, and an upper limit of 99.2% of water in the gel. transitions at Tg = 42 1C (1 and 2 1C minꢀ1) or at 46 1C
It should be noted that gel formation was not observed when the (0.5 1C minꢀ1). Gel formation, on the other hand, is strongly
protonated form of 1 was used, or if sodium was exchanged to dependent on the cooling rate and a large hysteresis can be seen:
ammonia or lithium; with potassium a gel was observed only upon gelation was observed at slower cooling rates, and at 2 1C minꢀ1
storing at 4 1C overnight, thus gelation is very specific to Na+. This no proper gel was obtained as seen by the weaker fluorescence of
might be explained by a preferential interaction of sodium with the system at 10 1C. Interestingly, heating of this intermediate
carboxylates compared to other cations (particularly lithium and state from 10 to 20 1C reinitiated the formation of a hydrogel.
potassium).10 This is further corroborated by the observation that This behaviour is consistent for all systems investigated, thus is
addition of an equal amount of 18-crown-6 per sodium induced largely independent of the concentration of both NaOH and 1.
complete dissolution of the gels within two days. Solvents other The hysteresis, and the gel formation dependence on the cooling
than water such as methanol, DMF or DMSO also did not induce rate but not on the heating rate, indicates that the supra-
gelation. The specific substitution pattern also seems to be crucial molecular assembly is not at thermodynamic equilibrium during
because replacing the carboxylate with bromide11 inhibited gel the annealing process.
formation, as did the attachment of sterically bulky groups such as
One of the samples containing 10 mg mLꢀ1 of 1 and 0.5 M
tBu at the terpyridine unit.
NaOH was used to test additional water uptake.† The gel (0.5 mL)
Spectral analysis showed that 1 has an absorbance at l = was layered with 250 mL of water and left to stand at room
275 nm which does not significantly change on going from the sol to temperature. After considerable time (three weeks) the water phase
gel phase.† The luminescence spectra (lex = 275 nm), however, still appeared to be separated from the hydrogel, thus the water was
revealed that in addition to an emission maximum at lem = 376 nm, not taken up by the gel phase. However, upon tilting of the sample
a second peak at lem = 528 nm appears in the gel form, with a vial the water phase did not move.† Therefore a minimum amount
relative intensity of 18% compared to the first peak. We have of the terpy had diffused into the water phase turning it into a gel,
used this characteristic gel emission peak to monitor the which remained translucent and formed a clear separate gel phase
thermal behaviour as it provides us with an ‘‘all-or-nothing’’ compared to the opaque gel.
handle; in particular reversibility and influence of the cooling
The structure of the gel was also examined using scanning
rate on gelation (Fig. 2) were studied. At higher temperatures electron microscopy (SEM).† The samples were dried under
(>50 1C), the gel-phase melts and a clear solution of the vacuum for the analyses. Depending on the initial gel forming
terpyridine is present; the actual time to dissolve the gel is in conditions, a range of different morphologies were observed,
the order of minutes, but is significantly longer at higher but all mainly showing long non-crystalline fibrous structures
concentrations of 1. Monitoring the emission at 528 nm at with interconnecting thinner rod-like fibres. Fig. 3a and b show
variable heating rates (0.5, 1 and 2 1C minꢀ1) showed consistent
Fig. 3 Representative SEM images of the fresh gels: (a) [1] = 9 mg mLꢀ1, [NaOH] =
Fig. 2 (a) Melting (solid lines) and annealing (dashed lines) profiles at variable
rates, measured by fluorescence emission (lex = 275 nm; lem = 528 nm). The inset
shows the fluorescent gel under a hand-held UV lamp. (b) Change in the
fluorescence spectrum upon cooling (front to back).
0.5 M, scale bar left: 200 mm, right: 5 mm. (b) [1] = 5 mg mLꢀ1, [NaOH] =
0.5 M, scale bar left: 50 mm, right: 10 mm. SEM images of aged samples:
(c) [1] = 4 mg mLꢀ1, [NaOH] = 0.75 M, scale bar: 20 mm. (d) [1] = 8 mg mLꢀ1
[NaOH] = 0.75 M, scale bar: 10 mm.
,
c
732 Chem. Commun., 2013, 49, 731--733
This journal is The Royal Society of Chemistry 2013

View Article Online
ChemComm
Communication
concentration, and the melting transition can be varied to a limited
extent. The gel does not swell upon addition of further water but
rather some of the terpy diffuses into a supernatant water layer,
transforming the water layer into a gel of different consistency
instead. Metal binding to give both M(1) and M(1)2 complexes
(M = CoIII, CuII) inhibits gel formation, most likely due to the
coordination geometry on the metal which acts as a sterical barrier;
bulky groups at the periphery of the terpy also inhibit gel formation
Fig. 4 XRD structure of Co(1)2(Cl)2 (atoms are drawn at the 50% probability level as does removal of the carboxy group. The XRD structure of the
after anisotropic refinement).
cobalt complex could be solved, which demonstrates that the
octahedral environment around the cobalt centre is not suitable
for efficient p–p stacking of the terpy units. The gels are now under
some representative examples of the SEM pictures obtained.
investigation for further applications in electrochemistry and ion
These seem to be the preferred morphologies for the freshly
transport, because the terpy unit could be used as an electro-
formed terpy gels. Upon ageing for several weeks in air, the
chemically active sensor moiety in gel-loaded electrodes.
morphology of the gels alters significantly, and either columnar
This work was supported by the EPSRC (EP/F009186/1).
structures or formation of micro-crystalline structures were
observed (Fig. 3c and d). The gels obviously rearrange their
structure over time when exposed to moist air.
Notes and references
Since the terpyridine is a versatile ligand for transition metals,
and some further substituted complexes have shown gelation,3b,12
we also investigated whether metal complexation has an influence
on the behaviour of 1 towards hydrogel formation. We studied both
cobalt and copper due to their ease of formation, and the com-
plexes were made in situ. A 0.5 : 1, 1 : 1 and 2 : 1 mixture of metal
chloride to 1 was used for the formation of the complexes, using
either water or DMF as solvent. The formation of the complexes
could be detected using UV-vis spectroscopy, but they remained
largely insoluble in water or DMF, thus gelation was not occurring
with these systems. Similarly, preformed ruthenium or platinum
complexes were not soluble enough under the gelation conditions.
‡ Data were collected on a Bruker Nonius KappaCCD using a Mo
rotating anode generator; standard procedures were followed. The
crystals lost solvent very rapidly and data collection, structure solution
and refinement were very problematical, however the basic structure is
clear. Crystal data for 2011sot0106 (AG/6043/33/F2) C62H72Cl2CoN12O10
,
%
Mr = 1275.15, T = 120(2) K, triclinic, space group P1, a = 8.7683(15), b =
18.903(3), c = 19.765(4) Å, a = 76.568(10), b = 80.430(10), g = 85.719(11)1,
V = 3139.8(10) Å3, rcalc = 1.349 g cmꢀ3, m = 0.426 mmꢀ1, Z = 2, reflections
collected: 16592, independent reflections: 8528 (Rint = 0.0778), GOF =
1.961, final R indices [I > 2s(I)]: R1 = 0.1998, wR2 = 0.4262, R indices
(all data): R1 = 0.2688, wR2 = 0.4543, largest difference peak and
hole = 0.975 and ꢀ0.865 e Åꢀ3
.
1 U. S. Schubert and C. Eschbaumer, Angew. Chem., Int. Ed., 2002, 41,
2893–2926.
2 (a) A. C. Benniston, A. Harriman, C. Pariani and C. A. Sams, J. Phys.
Chem. A, 2007, 111, 8918–8924; (b) P. P. Kumar, G. Premaladha and
B. G. Maiya, Chem. Commun., 2005, 3823–3825.
2+
However, we were able to grow single crystals of the Co(1)2
complex from the DMF solution, a complex which was previously
not analysed by single crystal XRD (Fig. 4).‡ The complex forms
a distorted octahedral meridional complex and crystallises as the
di-chloride with six solvent DMF molecules. Both chlorides are
hydrogen bonded to the adjacent carboxylate hydrogens and a 3D
network is supported by C–HꢁꢁꢁCl contacts. The DMF molecules sit
in pockets within the structure. The crystals lost solvent very
rapidly, therefore the rather low quality of the diffraction and the
structure refinement did not allow for further detailed analysis of
the structure. However, the fact that the metallated terpyridines
crystallise rather than forming a gel supports the fact that a planar
structure is essential for gelation and p-stacking plays a central role
in the process.
In summary, the metal ligand 40-para-phenylcarboxy-terpyridine
1 has been shown to form a thermally reversible fluorescent gel
with a water content of up to 99.2%. The gelation is likely due to
efficient p–p stacking of the hydrophobic terpy units, combined
with formation of the sodium induced salt bridge between terpy-
carboxylates and hydrogen bonded networks with the enclosed
water clusters and between the terpy units, as was observed
3 (a) E. J. Cho, I. Y. Jeong, S. J. Lee, W. S. Han, J. K. Kang and
J. H. Jung, Tetrahedron Lett., 2008, 49, 1076–1079; (b) A. Gasnier,
G. Royal and P. Terech, Langmuir, 2009, 25, 8751–8762;
(c) M. Kimura, Y. Nakagawa, N. Adachi, Y. Tatewaki, T. Fukawa
and H. Shirai, Chem. Lett., 2009, 382–383; (d) K. Hanabusa,
T. Hirata, D. Inoue, I. Kimura and H. Shirai, Colloids Surf., A,
2000, 169, 307–315.
4 (a) L. A. Estroff and A. D. Hamilton, Chem. Rev., 2004, 104, 1201–1217;
(b) M. George and R. G. Weiss, Acc. Chem. Res., 2006, 39, 489–497.
5 D. L. Elbert, Acta Biomater., 2011, 7, 31–56.
6 (a) D. Macaya and M. Spector, Biomed. Mater., 2012, 7, 012001;
(b) Y. L. Dai, P. A. Ma, Z. Y. Cheng, X. J. Kang, X. Zhang, Z. Y.
Hou, C. X. Li, D. M. Yang, X. F. Zhai and J. Lin, ACS Nano, 2012, 6,
3327–3338; (c) A. C. Lima, P. Sher and J. F. Mano, Expert Opin. Drug
Delivery, 2012, 9, 231–248.
7 (a) K. M. Gray, B. D. Liba, Y. F. Wang, Y. Cheng, G. W. Rubloff,
W. E. Bentley, A. Montembault, I. Royaud, L. David and G. F. Payne,
Biomacromolecules, 2012, 13, 1181–1189; (b) D. Patel, S. E. Vandromme,
M. E. Reid and L. J. Taite, Biomacromolecules, 2012, 13, 1420–1428.
8 N. A. Choudhury, S. Sampath and A. K. Shukla, Energy Environ. Sci.,
2009, 2, 55–67.
9 E. C. Constable, E. L. Dunphy, C. E. Housecroft, M. Neuburger,
S. Schaffner, F. Schaper and S. R. Batten, Dalton Trans., 2007, 4323–4332.
10 J. S. Uejio, C. P. Schwartz, A. M. Duffin, W. S. Drisdell, R. C. Cohen
and R. J. Saykally, Proc. Natl. Acad. Sci. U. S. A., 2008, 105, 6809–6812.
11 T. Ru¨hl and E. Stulz, Supramol. Chem., 2010, 22, 103–108.
previously with pyridine systems.13 In their case, intermolecular 12 F. Camerel, R. Ziessel, B. Donnio, C. Bourgogne, D. Guillon,
hydrogen bonding with weak p–p stacking was accounted for
M. Schmutz, C. Iacovita and J. P. Bucher, Angew. Chem., Int. Ed.,
2007, 46, 2659–2662.
13 T. H. Kim, J. Seo, S. J. Lee, S. S. Lee, J. Kim and J. H. Jung, Chem.
fluorescence in the gel, which could equally be the case in our
system. The gel forms in a narrow range of pH and terpy
Mater., 2007, 19, 5815–5817.
c
This journal is The Royal Society of Chemistry 2013
Chem. Commun., 2013, 49, 731--733 733