A highly selective fluorescent molecular sensor for potassium based on a calix[4]bisazacrown bearing boron-dipyrromethene fluorophores

Article abstract of DOI:10.1039/b504408a

The synthesis of a ditopic fluorescent sensor for cations associating a 1,3 alternate calix[4]bisazacrown-5 as an ionophore and substituted boron-dipyrromethene dyes as fluorophores is reported. Photophysical studies revealed that in medium and high polarity solvents an efficient charge transfer (CT) reaction occurs in the excited state leading to a dual emission and a strong quenching of fluorescence. This CT process is either totally suppressed by amino group protonation or hampered by cation complexation; in both cases, a strong fluorescence enhancement is observed. This "switching-on" of the emission is more pronounced when two cations are coordinated. The stability constants of complexes with sodium, potassium, caesium, calcium and barium cations were measured in acetonitrile and ethanol. The system shows a high sensitivity and selectivity for potassium over other metal ions. A test under physiological conditions was successfully achieved. The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2005.

Full text of DOI:10.1039/b504408a

View Online / Journal Homepage / Table of Contents for this issue
P A P E R
A highly selective fluorescent molecular sensor for potassium based
on a calix[4]bisazacrown bearing boron-dipyrromethene
fluorophores
Jean-Pierre Malval,ab Isabelle Leray*ab and Bernard Valeurab
a
´
CNRS UMR 8531 Laboratoire de Photophysique et Photochimie Supramoleculaires et
Macromoleculaires, Departement de Chimie, ENS-Cachan, 61 Avenue du President Wilson,
´
´
´
F-94235 Cachan Cedex, France
b
´ ´
´
Laboratoire de Chimie Generale, Conservatoire National des Arts et Metiers, 292 rue Saint-
Martin, F-75141 Paris Cedex, France. E-mail: icmleray@ppsm.ens-cachan.fr
Received (in Montpellier, France) 30th March 2005, Accepted 31st May 2005
First published as an Advance Article on the web 8th July 2005
The synthesis of a ditopic fluorescent sensor for cations associating a 1,3 alternate calix[4]bisazacrown-5 as
an ionophore and substituted boron-dipyrromethene dyes as fluorophores is reported. Photophysical studies
revealed that in medium and high polarity solvents an efficient charge transfer (CT) reaction occurs in the
excited state leading to a dual emission and a strong quenching of fluorescence. This CT process is either
totally suppressed by amino group protonation or hampered by cation complexation; in both cases, a strong
fluorescence enhancement is observed. This ‘‘switching-on’’ of the emission is more pronounced when two
cations are coordinated. The stability constants of complexes with sodium, potassium, caesium, calcium and
barium cations were measured in acetonitrile and ethanol. The system shows a high sensitivity and selectivity
for potassium over other metal ions. A test under physiological conditions was successfully achieved.
sensor with a 1,3 alternate calix[4]arene bearing a crown-5
Introduction
ether and an azacrown-5 ether at the rims, and also a pyrene
fluorophore was connected to the nitrogen atom leading to a
PET sensing process: K1/Na1 selectivity was perfectly main-
tained. This kind of specific receptor combined with a fluoro-
phore that can be excited at higher wavelengths can offer a
better alternative sensor for biological applications. Boron-
dipyrromethene (BDP) is a good candidate as a fluorophore
since it fulfils the photophysical requirements: high molar
absorption coefficient, high fluorescence quantum yield, good
photostability, long wavelength of excitation (B500 nm).11
Kollmannsberger et a1.12 reported an azacrown-5 substituted
boron-dipyrromethene; despite its lack of selectivity this fluoro-
ionophore showed an extremely high fluorescence enhancement
in the presence of alkali and alkaline-earth metal ions.
The present paper describes the synthesis of a new fluores-
cent molecular sensor 1 Calix-Bodipy. Spectroscopic proper-
ties, protonation effects and binding ability are reported.
Suitable potassium recognition in the presence of interfering
cations is demonstrated. A test under physiological conditions
is described.
The increasing demand of sensors for application in environ-
mental science, food technology, medical diagnosis has cur-
rently stimulated considerable research devoted to the design
and synthesis of abiotic molecular devices capable of signaling
a guest molecule or ion. Among the numerous analytical tools,
fluorescent molecular sensors offer several distinct advantages
in terms of sensitivity, selectivity, time response and spatial
resolution.1 Typical modular construction of such a system is
based on a link between a fluorophore and a recognition
subunit (ionophore) leading to a fluoroionophore. Photophy-
sical sensing processes are varied and can be for example
Photoinduced Electron Transfer (PET), Photoinduced Charge
Transfer (PCT), Energy Transfer, or excimer formation.2
Target substrates of particular interest are alkali and alka-
line-earth metal ions like K1, Na1, and Ca21 since their
concentration fluctuations play a major regulatory role in
many biological processes.3 A current challenging issue is the
specific detection of K1: the main difficulty lies in interference
and/or competing reactions with Na1 present in much larger
concentrations (the average physiological concentrations of
Na1 and K1 are 165 mM and 5 mM, respectively).3 K1 vs.
Na1 selectivity can be achieved by synthesis of concave
receptors having a suitable size and capable of forming specific
and stable interactions with its guest cation such as sandwich
crown-ether complexes,4 cryptands5 or calixarenes.6 In parti-
cular, 1,3 alternate calix[4]crown-5 or azacrown-5 ligands
appear to be good candidates since they showed a very high
affinity for K1.7 A K1 selective fluoroionophore has been
designed by using a naphthalene connected to a calix[4]-
crown-5.8 Kim et al.9 reported a 1,3 alternate calix[4]aza-
crown-5 with an appended nitrophenol chromophore; this
system used as a cation-extractant showed a high selectivity
towards K1 versus Na1. The same authors10 designed another
Experimental
General procedures
1H NMR spectra were recorded at room temperature on a
Bruker AC300 spectrometer using tetramethylsilane as refer-
ence; chemical shifts are reported in ppm. Elemental analyses
were performed at the Institut de Chimie des Substances
Naturelles (France). All chemicals were of reagent grade and
were used without further purification unless otherwise noted.
Toluene, tetrahydrofuran (THF), dimethylformamide (DMF),
dichloromethane and acetonitrile, received from Aldrich or
SDS, were distilled prior to use following standard methods.
T h i s j o u r n a l i s & T h e R o y a l S o c i e t y o f C h e m i s t r y a n d t h e
C e n t r e N a t i o n a l d e l a R e c h e r c h e S c i e n t i f i q u e 2 0 0 5
N e w J . C h e m . , 2 0 0 5 , 2 9 , 1 0 8 9 – 1 0 9 4
1089

View Online
were added to the mixture and the solution was stirred for 30
Synthesis
4-{Bis-[2-(2-hydroxyethoxy)ethyl]amino}benzaldehyde 3. 2-(2-
min at room temperature. Then, 1.5 ml of N-ethyl-N,N-diiso-
propylamine and 1.5 ml of BF3 ꢀ Et2O were added to the
mixture. After 30 min of stirring at room temperature, the
reaction mixture was washed with water, dried over Na2SO4
and chromatographed (eluent CH2Cl2–MeOH 90 : 10) fol-
lowed with precipitation in MeOH give 44 mg of pure com-
pound 1 (yield 23%). dH (300 MHz, CDCl3) 1.56 (s, 12 H), 2.56
(s, 12 H), 3.41 (t, J ¼ 6 Hz, 8 H), 3.58 (t, J ¼ 6 Hz, 8 H), 3.72 (s,
8 H), 3.8 (m, 16 H), 5.97 (s, 4 H), 6.60 (t, J ¼ 6 Hz, 4 H), 6.84
(d, J ¼ 9 Hz, 4 H), 7.14 (d, J ¼ 9 Hz, 12 H). dC (CDCl3, 100
MHz) 14.5 (CH3), 14.7 (CH3), 37.5 (CH2), 51.4 (CH2), 69.1
(CH2), 71.2 (CH2), 72.5 (CH2), 77.1 (CHAr), 111.4 (CHAr),
120.7 (CHAr), 121.5 (CHAr), 121.9 (CqAr), 129.2 (CHAr),
130.8 (CHAr), 132.1 (CqAr), 133.5 (CqAr), 142.9 (CqAr),
143.0 (CqAr), 147.7 (CqAr), 154.7 (CqAr), 157.1 (CqAr).
ESI: m/z calculated for C82H88B2F4N6O8: 1383.2, found
1383. Anal. Calc. for C82H88B2F4N6O8 ꢀ H2O: C 70.29; H
6.47; N 6.00. Found: C, 69.72; H, 6.54; N, 5.73%.
{[2-(2-Hydroxyethoxy)ethyl]phenylamino}ethoxy)ethanol (14
g, 52 mmol) and hexamethylenetetramine (10.4 g, 74.2 mmol,
1.4 eq) in 12 ml of absolute ethanol was refluxed for 1 h under
argon. 45 ml of mixture glacial acetic acid–formic acid (50 : 50
v/v) was slowly added to the mixture and the solution was
refluxed for 3 h. 100 ml of HCl 0.5 M were slowly added to the
mixture and the solution was stirred for 30 min. The solution
was neutralized with K2CO3 and extracted with dichloro-
methane. The combined organic phases were dried (MgSO4)
and concentrated under vacuum. The residue was purified by
silica gel chromatography (dichloromethane–acetone 80 : 20)
to give 6.1 g (39%) of an oil. dH (300 MHz, CDCl3) 2.48 (br s, 2
H), 3.56 (t, J ¼ 3 Hz, 4 H), 3.70 (m, 12 H), 6.74 (d, J ¼ 8.8 Hz, 2
H), 7.70 (d, J ¼ 8.8 Hz, 2 H), 9.70 (s, 1 H). 13C NMR (CDCl3,
100 MHz): 51.2 (CH2), 61.6 (CH2), 68.5 (CH2), 72.6 (CH2),
111.2 (CHAr), 125.5 (CqAr), 132.1 (CHAr), 152.7 (CqAr),
190.1 (CHO).
Spectroscopic measurements
Ditosylate of 4-{bis-[2-(2-hydroxyethoxy)ethyl]amino}benzal-
dehyde 4. A mixture of 3 (2.30 g, 7.75 mmol), p-toluenesulfonyl
chloride (2.97 g, 15.5 mmol) in dichloromethane (60 ml) was
cooled at 0 1C. Then, triethylamine (2.17 ml, 16 mmol) was
added dropwise After 24 h, the reaction mixture was neutra-
lized to pH B1 by addition of 1 M HCl. The organic layer was
dried over MgSO4, filtered and evaporated under reduced
pressure. The crude product was chromatographed on a silica
column with a 90 : 10 mixture of dichloromethane–acetone as
eluent to give pure ditosylate 4 (2.9 g; 61%) as a pale yellow oil.
UV/Vis absorption spectra were recorded on a Varian Cary5E
spectrophotometer. Corrected emission spectra were obtained
on a Jobin-Yvon Spex Fluoromax spectrofluorimeter. Cou-
marin 540 in ethanol (ff 0.78) and DCM in methanol (ff
0.43)12 were used as fluorescence standards. The complexation
constants were determined by global analysis of the evolution
of all absorption and/or emission spectra by using the Specfit
Global Analysis System V3.0 for 32-bit Windows system. This
software uses singular value decomposition and non-linear
regression modelling by the Levenberg–Marquardt method.13
Fluorescence intensity decays were obtained by the single-
photon timing method with picosecond laser excitation using a
Spectra-Physics set-up composed of a Titanium-Sapphire Tsu-
nami laser pumped by an argon ion laser, a pulse detector, and
doubling (LBO) and tripling (BBO) crystals. Light pulses were
selected by optoacoustic crystals at a repetition rate of 4 MHz.
Fluorescence photons were detected through an interferential
filter by means of a Hamamatsu MCP R3809U photomultiplier
connected to a Becker & Hickl electronic board (SPC 630). Data
were analysed by a nonlinear least-squares method using Glo-
bals software (Globals Unlimited, University of Illinois at
Urbana-Champaign, Laboratory of Fluorescence Dynamics).
d
H (300 MHz, CDCl3) 2.42 (s, 6 H), 3.62 (m, 12 H), 4.12 (t, J ¼
3.3 Hz, 4 H), 6.67 (d, J ¼ 9 Hz, 2 H), 7.31 (d, J ¼ 8.1 Hz, 4 H),
7.68 (d, J ¼ 9 Hz, 2 H), 7.75 (d, J ¼ 8. Hz, 4H), 9.72 (s, 1 H). dC
(CDCl3, 100 MHz) 21.5 (CH3Ph), 51.0 (CH2), 68.5 (CH2), 68.6
(CH2), 69.1 (CH2), 111.0 (CH2), 125.4 (2 CHAr), 127.8
(CHAr), 129.8 (CHAr), 132.0 (CHAr), 132.8 (CqAr), 144.9
(CqAr), 152.4 (CqAr), 190.1 (CHO). Anal. Calc. for
C29H35NO9S2. H2O: C 55.84; H 5.98; N 2.25; found: C,
55.62; H, 5.44; N, 2.21%.
Compound 5. Calix[4]arene (1.24 g; 2.9 mmol) and K2CO3
(4.04 g; 29.3 mmol) were stirred in acetonitrile (170 ml) at room
temperature for 1 h. Then, ditosylate 4 (1.77 g; 2.93 mmol)
dissolved in 10 ml of acetonitrile) was added and the reaction
mixture was refluxed for 7 days. Another portion (1.77 g; 2.93
mmol) of 2 dissolved in 10 ml of CH3CN and (4.04 g; 29.3
mmol) of K2CO3 was added to the solution mixture. After
refluxing the solution for 7 days, the solution was cooled to
room temperature and the solvents were removed under re-
duced pressure. The organic layer was washed with water,
dried over Na2SO4, filtered and evaporated. Chromatography
on silica column with a 95 : 5 mixture of dichloromethane–
acetone as eluent followed by precipitation with methanol gave
product 5 (0.58 g; 21%) as a beige solid. dH (300 MHz, CDCl3)
3.54 (m, 24 H), 3.72 (m, 16 H), 6.67 (t, J ¼ 6 Hz, 4 H), 6.77 (d,
J ¼ 8 Hz, 2 H), 7.10 (d, J ¼ 6 Hz, 8 H), 7.79 (d, J ¼ 8 Hz, 4 H),
9.77 (s, 2H). dC (CDCl3, 100 MHz) 37.8 (CH2), 51.1 (CH2),
68.4 (CH2), 70.5 (CH2), 71.2 (CH2), 111.8 (CHAr), 121.8
(CHAr), 125.4 (CqAr), 130.3 (CHAr), 132.2 (CHAr), 133.6
(CqAr), 152.4 (CqAr), 157.0 (CqAr), 190.0 (CHO). Anal. Calc.
for C58H62N2O10 þ CH2Cl2: C 68.66; H 6.25; N 2.71. Found:
C, 68.78; H, 6.41; N, 2.58%.
Result and discussion
Synthesis
The synthesis of the fluoroionophore 1 was performed accord-
ing to Scheme 1. The aldehyde 3 was obtained according to the
procedure described by Duff14 by reacting the dialcohol 2 with
hexamethylenetetramine in EtOH with a 39% yield. Ditosylate
4 is synthesized by reaction of 3 with toluene-p-sulfonyl
chloride in the presence of triethylamine in CH2Cl2 in 76%
yield. The biscalixcrown 5 is obtained in 21% yield by con-
densing calix[4]arene with 2 eq of ditosylate 4 and 10 eq of
K2CO3 in CH3CN thanks to our the previously reported
procedure for similar compounds.15 The fluoroionophore 1
was synthesized according to the procedure described by
Kollmannsberger12 in 23% yield from 2,4-dimethylpyrrole
and biscalixcrown 5, followed by oxidation in situ with 2,3-
dichloro-5,6-dicyano-1,4-benzoquinone (DDQ) and further re-
action with boron trifluoride etherate in the presence of
ethyldiisopropylamine.
Compound 1. 130 mg of compound 5 (0.137 mmol) and 52
mg of dimethylpyrrole were dissolved in 50 ml of dichloro-
methane under argon atmosphere. After addition of one drop
of trifluoroacetic acid, the reaction mixture was left for 2 h at
room temperature. 57 mg of dichlorodicyanobenzoquinone
Photophysical properties
Fig. 1 depicts the absorption and fluorescence spectra of 1 in
several solvents of different polarity and Table 1 summarizes
the spectroscopic data.
1090
N e w J . C h e m . , 2 0 0 5 , 2 9 , 1 0 8 9 – 1 0 9 4

View Online
Scheme 1 Reagents and conditions: i hexamethylenetetramine, HCl, ii, TsCl, NEt3, iii, 4, K2CO3, iv dimethylpyrrole TFA, DDQ, NEtPri2–BF3 ꢀ
Et2O.
1 shows a narrow absorption band at 500 nm with a shoulder
Stokes-shifted. In solvents more polar than hexane, the fluor-
escence is strongly quenched: the fluorescence quantum yield
drops from 0.27 in hexane to 0.003 in acetonitrile. Moreover, a
second fluorescent band appears at 630 nm. This broad and
unstructured band is well developed in solvents of medium
polarity like tetrahydrofuran but completely disappears in
polar solvents. The red emission is attributed to a species
having a strong charge transfer (CT) character and formed
by a fast reaction in the excited state.12 The driving force for
the formation of this CT-state can be estimated from the Rehm
–Weller equation18 (DGeT ¼ Eox ꢁ Ered ꢁ DE00) neglecting the
Coulomb part of the stabilization energy. With the assumption
at the short wavelength side (Fig. 1) characteristic of the BDP
chromophore.11 This band is insensitive to the polarity of the
solvent since it is slightly blue shifted when going from hexane
to acetonitrile. The molar absorption coefficient at the max-
imum wavelength (140 000 L molꢁ1 cmꢁ1 in acetonitrile) is
twice as large as the value of a typical BDP connected to a
phenyl group,16 suggesting that the two chromophores do not
interact in the ground state. No charge transfer band is
observed at long wavelengths. In the UV-region, a sharp and
intense band is located around 260 nm (Fig. 2); this band could
correspond to a transition centered on the N-alkyl aniline
moieties.17 In contrast to the absorption spectrum, the fluor-
escence spectrum is strongly solvent dependent (Fig. 1). In non-
polar solvents, the highly radiative locally excited state (LE)
exhibits a fluorescence band of mirror image shape, weakly
of electron transfer from the N,N dialkyl anilino group (Eox
¼
0.77 V vs. SCE)19 to the BDP moiety (Ered ¼ ꢁ1.15 V vs.
SCE)20 and with a zero–zero energy transition equal to 2.47 eV
in acetonitrile, the estimated driving force is strongly exer-
gonic: DGeT ¼ ꢁ0.55 eV.
Time-resolved fluorescence measurements were performed in
non-polar and polar solvents. In hexane the fluorescence decay
profile is mono-exponential (2.20 ns) yielding a radiative rate
constant of 0.13 ꢂ 109 sꢁ1 in good accordance with analogous
fluorophores.12,21 In polar solvents like acetonitrile, the fluor-
escence decay profile is more intricate since a sum of three
exponentials was required for a satisfactory fit. The fast
component (13 ps) with a relative amplitude of 0.83 corre-
sponds to the lifetime of the CT species. The slow component
(2.80 ns) is attributed to the lifetime of the LE state. This latter
value is markedly shorter than the lifetime of a phenyl sub-
stituted BDP (3.2 ns);12 this is consistent with a precursor–
successor mechanism where a first excited emitting species
leads to a second one during a reaction in the excited state.
Observation of a rise time in the spectral region where the CT
species fluoresces is expected; however, this very short rise time
could not be accurately measured because of the limitations of
our experimental setup. Besides, a third component whose
relative amplitude is smaller than 0.05 has a corresponding
lifetime of 165 ps. This intermediate decay time was also
Fig. 1 Absorption and fluorescence spectra of 1 in hexane (A), 1,4
dioxane (B), tetrahydrofuran (C), dichloromethane (D), acetonitrile
(E) and ethanol (F).
N e w J . C h e m . , 2 0 0 5 , 2 9 , 1 0 8 9 – 1 0 9 4
1091

View Online
Table 1 Absorption and emission properties of 1 in different solvents
Solvent
labs/nm
lem (LE)/nm
lem (CT)/nm
fCT/fLE
t1/psc
t2/psc
t3/nsc
a
b
fF
2
wr
Hexane
Dioxane
CHCl3
THF
498
499
502
499
500
496
498
497
509
512
514
512
513
509
512
509
—
596
609
658
672
—
0
0.28
—
—
2.20 ꢄ 0.03
1.15
1.36
3.98
2.28
5.34
3.03
0
0.052
0.043
0.019
0.014
0.003
0.006
0.008
o10
272 ꢄ 20
2.9 ꢄ 0.1
CH2Cl2
MeCN
EtOH
MeOH
a
B10
B10
165 ꢄ 40
212 ꢄ 30
2.8 ꢄ 0.1
3.1 ꢄ 0.2
1.15
1.12
—
—
0
0
b
c
The CT band is deconvoluted from the global fluorescence spectrum by using a Gaussian function. Error: 5–10%,
l
ex: 483 nm, lem ¼ 515 nm.
observed by Kollmannsberger et al.12 The authors suggested
the existence of a double minimum in the potential surface of
the excited CT state as a function of the twist angle between the
donor and acceptor group.
induced effects on fluorescence are similar to those observed
when amino groups are protonated and correspond to an
enhancement of the LE emission band (Fig. 3). Fluorimetric
titration was performed by analysing the evolution of the
whole emission spectrum by means of the SPECFIT program
(using global analysis with 180 wavelengths). The constants
K11 and K21 of the successive equilibria are defined as:
Protonation effects
The fast reaction from LE state to the CT state can be
suppressed by protonation of the amino group. Addition of
trifluoroacetic acid to a solution of 1 in ethanol shifts the BDP
absorption band by 3 nm whereas the band located at 260 nm
disappears completely (Fig. 2). A concomitant drastic enhance-
ment of fluorescence intensity is observed (inset of Fig. 2); the
fluorescence quantum yield rises from 0.003 to 0.67. The
fluorescence decay profile of the protonated species is mono-
exponential (t1 ¼ 4.04 ns) and leads to a radiative rate constant
of 0.17 ꢂ 109 sꢁ1, i.e. similar to the value calculated in hexane.
½MLꢃ
M þ L Ð ML K11
¼
ð1Þ
½Mꢃ½Lꢃ
½M2Lꢃ
M þ ML Ð M2L K21
¼
ð2Þ
½MLꢃ½Lꢃ
The global equilibrium for the formation of the complex
M2L is:
½M2Lꢃ
2M þ L Ð M2L b21
¼
ð3Þ
½MLꢃ½Lꢃ
Complexation studies
The values of K11 and K21 for K1 are reported in Table 2. In
contrast to the high value of K11 for K1 (log K11 ¼ 4.32), the
values for Cs1, Ca21, Ba21 were found to be very small (log
K11 o 0.5). K11 is slightly higher for Na1 (log K11 ¼ 1.32). A
typical fit of the fluorescence intensity evolution in the presence
of K1 is presented in the inset of Fig. 3.
Metal binding abilities of 1 were investigated in acetonitrile
and ethanol by addition of thiocyanate salts of Na1, K1,
perchlorate salts of Ca21, Ba21 and caesium acetate. The
absorption spectrum of 1 does not change significantly in the
presence of Na1, Cs1, Ca21, Ba21 (from 0 to 0.1 mol Lꢁ1
)
whereas increasing the concentration of K1 leads to a pro-
gressive decrease of the band at 260 nm. An isosbestic point is
then maintained at 251 nm, which confirms the existence of two
interconvertible species in the ground state. The complexation-
It should be noted that the value of K11 for K1 is of the same
order as those previously reported for calix[4]crown-5
ethers.7,9,10 The selectivity towards potassium ion with respect
to the other cations, expressed as the ratio of the stability
Fig. 2 Evolution of the absorption spectrum of 1 upon addition of
trifluoroacetic acid (solvent: ethanol). Inset: evolution of the fluores-
cence spectrum of 1 by addition of trifluoroacetic acid (lexc ¼ 475 nm;
solvent: ethanol).
Fig. 3 Fluorescence spectra of 1 (6.9 ꢂ 10ꢁ7 mol Lꢁ1) with increasing
concentration of K1 in ethanol (lexc ¼ 475 nm). Inset: titration curve
at 510 nm.
1092
N e w J . C h e m . , 2 0 0 5 , 2 9 , 1 0 8 9 – 1 0 9 4

View Online
Table 2 Stability constants of the complexes of 1 with K1 in
acetonitrile, in ethanol and in a mixture EtOH–H2O (75 : 25 v/v)
Solvent
Acetonitrile
Ethanol
EtOH–H2O (75 : 25 v/v))
Log K11
Log K21
Log b21
4.32 ꢄ 0.04
1.73 ꢄ 0.04
6.05 ꢄ 0.04
4.47 ꢄ 0.14
0.78 ꢄ 0.14
5.26 ꢄ 0.14
2.4 ꢄ 0.10
1.25 ꢄ 0.15
3.5 ꢄ 0.1
constants, is higher than 1000 and this very high value is
attributed to the size fit of the calix[4] azacrown-5 iono-
phore7,9,10 and to a preferential cation p-interaction of K1
with the phenyl moieties of the 1,3 alternate calix[4]arene.6
Furthermore a second step of complexation is observed for
K1 at higher cation concentrations (millimolar domain), and is
attributed to the [2 : 1] complex formation (metal : ligand).
The fluorescence intensity increases more gradually, and the
derived association constant (K21) has a value of 1.73 and 0.78
(logarithmic scale) in acetonitrile and ethanol, respectively.
These constants which are more than two orders of magnitude
smaller than K11 show that the complexation of the second
cation is more difficult. The ratio K21/ K11 (B0.002), which is
far below the statistical value of 1/4 observed if the two binding
sites were independent and equivalent,22 is evidence for an
anticooperative process. The molecular structures of the com-
plexes [1 : 1] and [2 : 1] optimized by the AM1 semi-empirical
method23 indicate that the average distance between K1 and
the nitrogen atom decreased by 1.28 A, from 4.28 A to 3.00 A,
when two K1 are coordinated. Electrostatic repulsion between
the two cations can be here invoked.24 The extrapolated
fluorescence spectra of the [1 : 1] and [2 : 1] complexes in
acetonitrile are displayed in Fig. 4. The fluorescence quantum
yield of the [2 : 1] complex (f21) is 10 times higher than that of
the [1 : 1] complex (f11). This result suggests a more efficient
inhibition of the excited CT process in the [2 : 1] complex, and
this is consistent with the calculated structural data since the
proximity of potassium ions reduces the electron-donating
character of the amino group. The intensity of the LE band
is therefore enhanced.
Fig. 5 Concentration profiles of 1, and its [1 : 1] and [2 : 1]
complexes with K1 in ethanol ([1] ¼ 6.8 10ꢁ7 mol Lꢁ1).
yield of the [2 : 1] complex is 75 and 154 times more emissive
than that of the [1 : 1] complex and 1, respectively. The species
distribution profiles as a function of the concentration of K1
were calculated for 1 and for its relevant complexes (Fig. 5).
With an initial free ligand concentration of 6.8 ꢂ 107 mol Lꢁ1
,
the [1 : 1] complex outnumbers the other species in the
millimolar domain (98.5% of the species). However, the [2 : 1]
complex also starts being formed and because of its high
relative fluorescence quantum yield, a slight increase of its
concentration leads to a high enhancement of the fluorescence
signal. Thus, above 1 mmol Lꢁ1 of K1, the total fluorescence
intensity is mainly governed by the fluorescence intensity
contribution of the [2 : 1] complex, as displayed in Fig. 6.
It is now of interest to examine the practical use of this new
fluoroionophore 1 for the determination of K1 in aqueous
solutions. It turned out that 1 cannot be used in pure water
The selectivity of K1 is even better in ethanol (polar protic
solvent) since no significant fluorescence change was observed
upon addition of a large excess of Na1, Cs1, Ca21, Ba21
(cM > 0.1 mol Lꢁ1). Moreover, the extrapolated fluorescence
spectra of each complex indicate that the fluorescence quantum
Fig. 4 Extrapolated fluorescence spectra of 1 and its [1 : 1] and [2 : 1]
complexes with K1 in ethanol.
Fig. 6 Relative fluorescence intensity of each species as a function of
potassium concentration in ethanol ([1] ¼ 6.8 ꢂ 10ꢁ7 mol Lꢁ1).
N e w J . C h e m . , 2 0 0 5 , 2 9 , 1 0 8 9 – 1 0 9 4
1093

View Online
the millimolar range of K1 concentration. Thus, the high K1/
Na1 selectivity under physiological conditions combined with its
good sensitivity make this system a suitable K1-sensor for
analytical applications.
References
1
(a) Chemosensors of Ion and Molecular Recognition, ed. J.-P.
Desvergne and A. W. Czarnik, NATO ASI Series, Kluwer Aca-
demic, Dordrecht, 1997, vol. C492; (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) B.
Valeur and I. Leray, Coord. Chem. Rev., 2000, 205, 3–40.
B. Valeur, Molecular Fluorescence. Principles and Applications,
Wiley-VCH, Weiheim, 2002.
2
3
4
A. Minta and R. Y. Tsien, J. Biol. Chem., 1989, 264, 19449.
W. S. Xia, R. H. Schmehl and C. J. Li, J. Am. Chem. Soc., 1999,
121, 5599.
5
(a) D. B. Lehn, J. P. Sauvage and J. Blanzat, Tetrahedron, 1973,
29, 1629; (b) A. P. de Silva, H. Q. N. Gunaratne and K. R. A. S.
Sandanayake, Tetrahedron Lett., 1990, 31, 5193; (c) R. Crossley,
Z. Goolamali and P. G. Sammes, J. Chem. Soc., Perkin Trans. 2,
1994, 1615; (d) H. He, M. A. Mortellaro, M. J. P. Leiner, R. J.
Fraatz and J. K. Tusa, J. Am. Chem. Soc., 2003, 125, 1468.
(a) C. D. Gutsche, Calixarenes Revisited, The Royal Society of
Chemistry, Cambridge, 1998; (b) Calixarenes 2001, ed. Z. Asfari,
Fig. 7 Fluorescence intensity response of 1 (6 ꢂ 10ꢁ7 mol Lꢁ1) in
water–ethanol (1 : 3, v : v), lexc ¼ 475 nm, lem ¼ 515 nm. Potassium,
sodium concentration are (Na1/K1 in 10ꢁ3 mol Lꢁ1): (a) 0/0; (b) 44/0;
(c) 44/0.52; (d) 44/1.05; (e) 44/1.57; (f) 44/2.1; (g) 44/2.62; (h) 44/3.15;
(i) 44/3.67; (j) 44/4.2; (k) 44/4.72.
6
7
V. Bohmer, J. Harrowfield and J. Vicens, Kluwer Academic,
¨
Dordrecht, 2001.
(a) A. Casnati, A. Pochini, R. Ungaro, C. Bocchi, F. Ugozzoli,
R. J. M. Egberink, H. Strujik, R. Lugtenberg, F. de Jong and
D. N. Reinhoudt, Chem. Eur. J., 1996, 2, 436; (b) F. Arnaud-Neu,
Z. Asfari, B. Souley and J. Vicens, New J. Chem., 1996, 20, 453;
(c) J. S. Kim, W. K. Lee, D. Y. Ra, Y. I. Lee, W. K. Choi, K. W.
Lee and W. Z. Oh, Microchem. J., 1998, 59, 464; (d) K. N. Koh,
K. Araki, S. Shinkai, Z. Asfari and J. Vicens, Tetrahedron Lett.,
1995, 36, 6095; (e) S. K. Kim, S. H. Lee, J. Y. Lee, J. Y. Lee, R. A.
Bartsch and J. S. Kim, J. Am. Chem. Soc., 2004, 126, 16499.
J. H. Bu, Q. Y. Zheng, C. F. Chen and Z. T. Huang, Org. Lett.,
2004, 6, 3301.
because of self-aggregation, as previously reported,25 that
alters the photophysical properties. However, a mixture EtOH
–H2O (3 : 1 v/v)) was found to be still suitable, in spite of the
decrease in affinity for K1. It should be noticed that in contrast
to the previously system described by Kim et al.9,10 with a
tertiary amine linked to a calix[4]azacrown-5 our system
composed of a dialkylaniline with a boron-dipyrromethene as
an acceptor group in the para position exhibits a much lower
pKa,21c which enables application under neutral conditions.
The titration curve remains consistent with the formation of
two successive [1 : 1] and [2 : 1] complexes (log K11 ¼ 2.48 and
log K21 ¼ 1.25).
8
9
J. S. Kim, O. J. Shon, J. W. Ko, M. H. Cho, I. Y. Yu and
J. Vicens, J. Org. Chem., 2000, 65, 2386.
10 J. S. Kim, O. J. Shon, J. A. Rim, S. K. Kim and J. Yoon, J. Org.
Chem., 2002, 67, 2348.
11 (a) E. Vos de Wal, J. Pardoen, J. A. Van Koeveringe and J.
Lugtenburg, Recl. Trav. Chim. Pays-Bas, 1977, 96, 306; (b) J.
Karolin, L. B.-A. Johansson, L. Strandberg and T. Ny, J. Am.
Chem. Soc., 1994, 116, 7801; (c) T. Werner, C. Huber, S. Heinl, M.
Kollmannsberg, J. Daub and O. S. Wolfbeis, Fresenius’ J. Anal.
Chem., 1997, 359, 150.
12 M. Kollmannsberger, K. Rurack, U. Resch-Genger and J. Daub,
J. Phys. Chem. A, 1998, 102, 10211.
13 H. Gampp, M. Maeder, C. J. Meyer and A. D. Zuberbuhler,
¨
Talanta, 1985, 32, 95–101.
14 J. C. Duff, J. Chem. Soc., 1945, 276.
15 I. Leray, Z. Asfari, J. Vicens and B. Valeur, J. Chem. Soc., Perkin.
Trans. 2, 2002, 1429.
16 P. Toele, H. Zhang, C. Trieflinger, J. Daub and M. Glasbeek,
Chem. Phys. Lett., 2003, 368, 66.
17 (a) M. Ballester and J. Riera, Spectrochim. Acta A, 1967, 23,
51533; (b) S. Kapelle, W. Rettig and R. Lapouyade, Chem. Phys.
Lett., 2001, 348, 416; (c) J. P. Malval, J. P. Morand, R. La-
Then, a test of the analytical performance of 1 for the
determination of K1 in extracellular conditions (serum or
whole blood) was achieved. In such conditions, the concentra-
tion of K1 is about 5 mM, and that of Na1 about 150 mM.
Thus, considering a biological sample that is diluted with an
ethanol solution (containing 1 at the appropriate concentra-
tion) by a factor of 4, the question arises as to whether one can
detect about 1 mM of K1 in the presence of about 40 mM of
Na1. Fig. 7 shows that this is indeed possible: increasing the
concentration of Na1 from 0 to 44 mM has almost no effect on
the fluorescence intensity, whereas addition of K1 from 0 to
5 mM in the presence of 44 mM Na1 leads to a quasi-linear
increase in fluorescence intensity. This demonstrates the sensi-
tivity and selectivity of 1 towards K1 at physiological concen-
trations.
pouyade, W. Rettig, G. Jonusauskas, J. Oberle, C. Trieflinger and
´
J. Daub, Photochem. Photbiol. Sci., 2004, 3, 939.
18 D. Rehm and A. Weller, Isr. J. Chem., 1970, 8, 259.
19 D. R. Worall, S. L. Williams, F. Wilkinson, J. E. Crossley, H.
Bouas-Laurent and J. P. Desvergne, J. Phys. Chem., 1999, 109,
9255.
20 R. Y. Lai and A. J. Bard, J. Phys. Chem. B, 2003, 107, 5036.
21 (a) M. Kollmannsberger, K. Rurack, U. Resch-Genger, W. Rettig
and J. Daub, Chem. Phys. Lett., 2000, 329, 363; (b) K. Rurack, M.
Kollmannsberger, U. Resch-Genger and J. Daub, J. Am. Chem.
Soc., 2000, 122, 968; (c) K. Rurack, M. Kollmannsbergerb and
J. Daub, New J. Chem., 2001, 25, 289.
22 K. A. Connors, Binding constants. The measurement of Molecular
Complex Stability, John Wiley & Sons, New York, 1987.
23 HYPERCHEM, v. 7.03, Hypercube, Inc., Gainesville, FL, 2002.
24 I. Leray, Z. Asfari, J. Vicens and B. Valeur, J. Fluoresc., 2004, 14,
451.
Conclusion
The synthesis of a new fluoroionophore associating a 1,3 alter-
nate calix[4] azacrown-5 and a boron–dipyrromethene as a
fluorophore is reported. The ionophore, which possesses two
binding sites whose size perfectly fits the potassium cation,
exhibits an excellent K1/Na1 selectivity in acetonitrile, ethanol,
and ethanol–water mixtures. Cation coordination to the amino
group blocks the efficient CT mechanism, which leads to a strong
fluorescence enhancement of the LE band. The enhancement is
drastically larger with the [2 : 1] species (Fluorescence Enhance-
ment Factor ¼ 154 in ethanol); this result is consistent with a
closer proximity between the cation and the nitrogen atom of the
aniline moiety. Fluorescence signal amplification is all the more
improved as the [2 : 1] complex is produced which is the case in
25 F. Bergstrom, I. Mikhalyov, P. Haggof, R. Wortmann, T. Ny and
¨
L. B. A. Johansson, J. Am. Chem. Soc., 2001, 124, 196.
¨
¨
1094
N e w J . C h e m . , 2 0 0 5 , 2 9 , 1 0 8 9 – 1 0 9 4