H+, Na+ and K+ modulated lanthanide luminescent switching of Tb(III) based cyclen aromatic diaza-crown ether conjugates in water

Article abstract of DOI:10.1039/b307055g

The cyclen based aromatic diaza-15-crown-5 and 18-crown-6 ether conjugates 1Tb-4Tb were designed as luminescent switches for sodium and potassium where the delayed Tb(III) emission, occurring as line-like emission bands between 490-622 nm, was 'switched o

Full text of DOI:10.1039/b307055g

H+, Na+ and K+ modulated lanthanide luminescent switching of Tb(III
based cyclen aromatic diaza-crown ether conjugates in water
)
Thorfinnur Gunnlaugsson* and Joseph P. Leonard
Department of Chemistry, Trinity College Dublin, Dublin 2, Ireland. E-mail: gunnlaut@tcd.ie; Fax: 00 353
1 671 2826; Tel: 00 353 1 608 3459
Received (in Cambridge, UK) 19th June 2003, Accepted 5th August 2003
First published as an Advance Article on the web 22nd August 2003
The cyclen based aromatic diaza-15-crown-5 and 18-crown-
upon addition to diethyl ether (100 mL). Whereas the cationic
complexes 1Tb and 2Tb were purified on neutral alumina, the
neutral complexes 3Tb and 4Tb did not require further
purifications. These complexes were characterised by conven-
tional methods; the ESMS of all showed the correct isotopic
patterns, and the 1H NMR (CD3CN) gave the expected shifted
resonance for the equatorial and the axial protons of the cyclen
ring, and the a-CH2, appearing for 1Tb at 20.4, 18.6, 0.9, 0.3,
23.4 and 28.2 ppm indicating a monocapped square anti-
prismatic geometry as the major isomer in solution.5–7
The photophysical properties of 1Tb–4Tb were investigated
in water and in buffered solution at pH 7.4. Firstly, the hydration
state (q) of all the complexes was determined by measuring their
Tb(III) excited state lifetimes by direct excitation of the metal
ion at 366 nm in (un-buffered) H2O (tH2O) and D2O (tD2O). For
1Tb these were found to be 0.181 and 0.189 ms, giving q =
0.95, indicative of single metal coordinated water molecule.
Similarly, for 2Tb–4Tb, q-values of 0.9–1.35 were determined.
The pH dependence of the ground and the excited state were
evaluated by exciting the aromatic antenna. In alkaline solution
( ~ pH 12) the complexes gave rise to two absorption bands. For
1Tb these appeared at ca. 211 and 251 nm with a shoulder at
279 nm, whilst in acidic solution ( ~ pH 2) these bands were blue
shifted with concomitant hypochromic effect (ESI). Monitoring
the changes at 279 nm as a function of pH gave two sigmoidal
curves that were consistent with a simple ion-equilibrium and 1
: 1 binding (ESI).2 From these, pKas of 11.0 (±0.2) and 5.5
(±0.2) were determined, assigned to the deprotonation of the
receptor amide and the protonation of one of the aniline units of
the receptor. Small changes were also seen at lower pH, which
we assigned to the protonation of the aniline unit conjugated to
6 ether conjugates 1Tb–4Tb were designed as luminescent
switches for sodium and potassium where the delayed Tb(III
)
emission, occurring as line-like emission bands between
490–622 nm, was ‘switched on’ upon recognition of these
ions in pH 7.4 buffered water solution.
The development of luminescent sensors, switches, and other
luminescent devices is an active area of research in both
supramolecular and material chemistry.1,2 Often, these systems
employ fluorescence, emitting in the ns-range, whereas more
recently, transition metal ions (such as Ru(II), Os(II), Ir(III) etc.)
complexes and conjugates, that emit in the sub-ms range have
been developed.3 Of these, the latter have a particular advantage
over the former for on-line and real-time chemosensing of
biological species as their long excited state lifetimes overcome
auto-fluorescence and short lived light scattering in biological
samples.2 We are interested in the development of fluorescent
and colorimetric chemosensors.4 Recently, we have developed
cyclen based lanthanide luminescent devices.5–7 Here the
lanthanide emission of Tb(III) and Eu(III) was modulated by
either a single ionic,5 or more (ionic and molecular) ‘inputs’
such as H+ or O2 (giving rise to luminescent logic-gate
mimics).6 Furthermore, we have developed Tb(III) complexes as
sensors for organic aromatic carboxylates.7 These compounds
have several advantages over many other luminescent devices
in addition to emitting at long wavelength, they have line-like
emission bands ( ~ 10–15 nm), long lived exited state (delayed
emission) occurring in the ms time range. With the aim of
further advancing this research, using different ionic ‘inputs’,
we have developed the ligands 1–4, tetra-substituted cyclen
conjugates using aromatic diaza-15-crown-58,9 and diaza-
18-crown-69 ethers for the recognition of Na+ and K+ in water
respectively. In these, the aromatic crown ethers would
function, not only as receptors for these ions, but also as antenna
for the population (sensitisation) of the excited state of the
lanthanide ion.2c,5–7 As a result, we propose that this sensitisa-
tion process and consequently, the emission, would be modu-
lated upon Na+ or K+ recognition, e.g. by ‘switching’ the
lanthanide emission either ‘off’ or ‘on’. Herein we describe the
syntheses of 1–4 and their Tb(III) complexes 1Tb–4Tb, and
their luminescent properties in water and at pH 7.4. To the best
of our knowledge these are the first examples of such stable and
highly coordinatively saturated Tb(III) conjugates as lumines-
cence switches for Na+ and K+ in water.10
The synthesis of 1Tb–4Tb is shown in Scheme 1. The
synthesis of 5 and 6 have been previously reported by us.9 The
corresponding a-chloroamides 7 and 8, were made in 62% yield
by reacting 5 or 6 with a-chloroacetic anhydride in the presence
of Na2CO3 in 1,4-dioxane. The formation of the cyclen
conjugates 1–4 was achieved by reacting first 7 or 8 with either
97 or 1011 in dry CH3CN and Cs2CO3 under inert atmosphere at
72 °C for 72 hours, giving 1, 2, 11 and 12 respectively. The
acids 3 and 4 were then made by ester hydrolysis of 11 and 12
using TFA–CH2Cl2, followed by precipitation from MeOH
upon addition of diethyl ether. The Tb(III) complexes 1Tb–4Tb
were formed by refluxing an equimolar amount (0.1 mM) of 1–4
with Tb(CF3SO3)3 in dry CH3CN, followed by the precipitation
Scheme 1 The synthesis of the ligands 1, 2, 3 and 4, and the corresponding
Tb(III) complexes 1Tb, 2Tb, 3Tb and 4Tb.
2424
CHEM. COMMUN., 2003, 2424–2425
This journal is © The Royal Society of Chemistry 2003

the cyclen unit. Similar results were observed for 2Tb–4Tb.
The associated changes in the fluorescence emission, a broad
emission band lmax ~ 378 nm, gave rise to efficient quenching
upon acidification with two pKas at ca. 5.6 and 10.5.
The most striking and interesting changes were seen for the
Tb(III) luminescence, which occurred as narrow bands at 490,
5
7
547, 586 and 622 nm for the D4 ? FJ (J = 6, 5 ,4 and 3)
transitions, where the Tb(III) emission was reversibly ‘switched
on’ between pH 9.8–12, and pH 1.8–4.0 for 1Tb and 2Tb, with
pKas of 2.8(±0.2) and 3.0 (±0.2), and 11.2 (±0.2) and 11.1 (±0.2)
for the two profiles respectively. The tH O were also pH
dependent, being 1.29, 0.18 and 0.43 ms for 22Tb at pH ~ 1, 7,
and ~ 12 respectively. However, between pH 4–9 the lantha-
nide emission was not pH dependent, and remained ‘switched
off’ as is evident from plotting the changes of the emission at
547 nm vs. pH (Fig. 1). Such ‘on-off-on’ switching is quite
unique, but we have recently seen the reverse effect i.e. the ‘off-
on-off’ switching for a Eu(III) cyclen conjugate having a single
phen ligand as the antenna.12 The lower pKas determined for
1Tb and 2Tb are attributed to the protonation of the aniline
moiety adjacent to the cyclen complex, which is not clearly
observed in the absorption or the fluorescence emission.4d
However, the above trend was not observed for 3Tb and 4Tb.
Even though the Tb(III) emission was switched on in acidic
solution, mirroring the above effects, the changes in the alkaline
solution did not occur. To investigate this further, we evaluated
the q-values of 2Tb and 4Tb as a function of pH. For the
cationic complexes, q was found to be ~ 1 between pH 1–12,
whereas neutral complexes gave q = 1 between pH 1–8,
however, in alkaline solution a meaningful q value could not be
determined. This suggests that the coordination environment, or
the charge of the complex, might play a part here, masking the
deprotonation process of the amide. We are currently investigat-
ing this in greater detail. However, the importance of the above
behaviour lies in the fact that the Tb(III) emission was, for all the
complexes, ‘switched off’ and pH independent in the physio-
logical pH range. Hence, the changes in the Tb(III) emission
upon addition of either Na+ or K+ would signify the recognition
of these ions. To investigate this we carried out a titration of
1Tb–4Tb using a range of Group I and II cations (as AcO2 and
Cl2 salts) at pH 7.4 (TRIS buffer) and 0.1 M tetramethyl
ammonium chloride to maintain constant ionic strength. For all,
the Tb(III) emission (Fig. 2, insert) was clearly ‘switched on’
upon titrations with either Na+ or K+. In contrast the
fluorescence emission of 1Tb–4Tb was quenched, and the
Fig. 2 The changes in the emission at 547 nm for 2Tb upon titration of
several group I and II cations. Inserted are the corresponding changes in the
Tb(III) emission, which is ‘switched on’ when titrated using K+. Excitation
at 300 nm, [2Tb] = 4 mM. pM = 2log [M]; M = cation.
either Na+ or K+ by the aromatic diaza-crown ethers gives rise
to major structural changes in the receptor, which cause the ring
to be almost deconjugated from the aromatic moieties as shown
by X-ray crystallography of the two precursors of 5 and 6 and
their corresponding Na+ and K+ complexes.8,9 This increases
the oxidation potential of the receptor.8 We propose that such
interactions enhance the efficiency of the populations of the
Tb(III) 5D4 state (which occur via sensitisation from the S1 ? T1
of the antenna), causing the large Tb(III) luminescent enhance-
ments.2d,4,6,7 We are currently investigating this phenomenon
and improving the design of these complexes for the biological
chemistry of Na+ and K+.
In summary, we have developed novel luminescent switches
using cyclen diaza-aromatic crown ether conjugates, which give
rise to stable nine coordination complexes with q ~ 1. Whereas
at pH 7.4 the Tb(III) emission was ‘switched off’ the recognition
of Na+ or K+ gave rise to large Tb(III) luminescent enhance-
ments. We thank Kinerton Ltd, Enterprise Ireland, TCD and
BMRI for financial support, Dr Hazel M. Moncrieff for helpful
discussion and Dr John E. O’Brien for assisting with NMR.
Notes and references
1 V. Balzani, A. Credi and M. Venturi, Pure Appl. Chem., 2003, 75, 541;
R. Ballardini, V. Balzani, A. Credi, M. T. Gandolf and M. Venturi, Acc.
Chem. Res., 2001, 36, 445.
2 (a) A. P. de Silva, D. B. Fox, A. J. Huxley and T. S. Moody, Coord.
Chem. Rev., 2000, 205, 41; (b) A. P. de Silva, H. Q. N. Gunaratne, T.
Gunnlaugsson, A. J. M. Huxley, C. P. McCoy, J. T. Rademacher and T.
E. Rice, Chem. Rev., 1997, 97, 1515; (c) D. Parker and J. A. G.
Williams, J. Chem. Soc., Dalton Trans., 1996, 3613; (d) C. Li and W. T.
Wong, Chem. Commun., 2002, 2034.
3 M. Licini and J. A. G. A. Williams, Chem. Commun., 1999, 1943;
Harriman and R. Ziessel, Chem. Commun., 1996, 1707.
4 (a) T. Gunnlaugsson, A. P. Davis, J. E. O’Brien and M. Glynn, Organic
Lett., 2002, 4, 2449; (b) T. Gunnlaugsson, B. Bichell and C. Nolan,
Tetrahedron Lett., 2002, 43, 4948; (c) T. Gunnlaugsson, M. Nieu-
wenhuyzen, L. Richard and V. Thoss, J. Chem. Soc., Perkin Trans. 2,
2002, 141; (d) T. Gunnlaugsson and J. P. Leonard, J. Chem. Soc., Perkin
Trans. 2, 2002, 1980; (e) T. Gunnlaugsson, A. P. Davis and M. Glynn,
Chem. Commun., 2001, 2556.
absorption spectra was blue shifted. The changes in the Tb(III
)
emission of 2Tb as a function of pM (2log[M+]) is shown in
Fig. 2 for several ions. In general, even though the emission was
only fully switched on at high ion concentrations (possibly due
to inductive effect of the Tb(III) ion)13 the Tb(III) luminescent
enhancement was up to 40 fold for all the complexes.
Furthermore, the selectivity towards various ions was as
expected, e.g. the emission of 1Tb and 3Tb (15-crown-5), was
most efficiently switched on by Na+, whereas for 2Tb and 4Tb
(18-crown-6), it was switched on by K+. The effect of other ions
demonstrated poor selectivity towards Li+, Mg2+ and Ca2+ as
demonstrated for 2Tb in Fig. 2. The reason for the efficient
luminescent switching by these complexes upon recognition of
5 T. Gunnlaugsson, Tetrahedron Lett., 2001, 42, 8901.
6 T. Gunnlaugsson, D. A. Mac Dónaill and D. Parker, J. Am. Chem. Soc.,
2001, 123, 12866; T. Gunnlaugsson, D. A. Mac Dónaill and D. Parker,
Chem. Commun., 2000, 93.
7 T. Gunnlaugsson, A. Harte, J. P. Leonard and M. Nieuwenhuyzen,
Chem. Commun., 2002, 2134.
8 A. P. de Silva, H. Q. N. Gunaratne, T. Gunnlaugsson and M.
Nieuwenhuyzen, Chem. Commun., 1996, 1967.
9 T. Gunnlaugsson, H. Q. N. Gunaratne, M. Nieuwenhuyzen and J. P.
Leonard, J. Chem. Soc., Perkin Trans. 1, 2002, 1954.
10 A. P. de Silva, H. Q. N. Gunaratne, T. E. Rise and S. Stewart, Chem.
Commun., 1997, 1891.
11 A. Dadabhoy, S. Faulkner and P. G. T. Sammes, J. Chem. Soc., Perkin
Trans. 2, 2002, 348.
12 Gunnlaugsson, J. P. Leonard, K. Senechal and A. J. Harte, J. Am. Chem.
Soc., 2003, 125, in press.
Fig. 1 The changes in the emission at 547 nm for 1Tb and 2Tb as a function
of pH. [1Tb] and [2Tb] = 10mM . Excitation 300 nm.
13 Reany, T. Gunnlaugsson and D. Parker, J. Chem. Soc., Perkin Trans. 2,
2000, 1819.
CHEM. COMMUN., 2003, 2424–2425
2425