About the use of an amide group as a linker in fluoroionophores: Competition between linker and ionophore acting as chelating groups

Article abstract of DOI:10.1039/b501613d

The photophysical and complexing properties of a series of aza-crown fluoroionophores based on coumarin 343 and on 3- and 6-methoxynaphthoic amides in acetonitrile are reported. The goal of the work was to probe the participation of the amide bridge linking the fluorophore and the ionophore in the metal chelation. The use of 3- and 6-methoxy substituents in the naphthoic amide fluorophores allowed us to maintain the charge transfer character of the system and to probe the participation of the methoxy group as ancillary ligand. The aza-crown unit is no longer complexing when the amide linker is included in a β-dicarbonyl sub-structure. The amide function itself is still able to form complexes, even if weaker, with the cations. The Royal Society of Chemistry 2005.

Full text of DOI:10.1039/b501613d

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/materials | Journal of Materials Chemistry
About the use of an amide group as a linker in fluoroionophores:
competition between linker and ionophore acting as chelating groups
Laetitia Maton, Doroth e´ e Taziaux, Jean-Philippe Soumillion and Jean-Louis Habib Jiwan*
Received 1st February 2005, Accepted 20th April 2005
First published as an Advance Article on the web 4th May 2005
DOI: 10.1039/b501613d
The photophysical and complexing properties of a series of aza-crown fluoroionophores based on
coumarin 343 and on 3- and 6-methoxynaphthoic amides in acetonitrile are reported. The goal of
the work was to probe the participation of the amide bridge linking the fluorophore and the
ionophore in the metal chelation. The use of 3- and 6-methoxy substituents in the naphthoic
amide fluorophores allowed us to maintain the charge transfer character of the system and to
probe the participation of the methoxy group as ancillary ligand. The aza-crown unit is no longer
complexing when the amide linker is included in a b-dicarbonyl sub-structure. The amide function
itself is still able to form complexes, even if weaker, with the cations.
selectivity of the detection of alkaline-earth cations versus
Introduction
lithium cation. Whereas the complex of C343-crown with
Fluorescent molecular sensors for the selective detection of
lithium in acetonitrile was nearly as stable as those with
metal ions have attracted considerable attention due to their
magnesium or calcium, the stability constant of C343-
interest in analytical chemistry, biology, medicine (clinical
benzocrown with lithium is two orders of magnitude lower
1
,2
diagnosis), environment, etc. The design principles of such
than those with alkaline-earth cations. In the case of C343-
dibenzocrown, the interaction between the ligand and the
lithium cation is so weak that the stability constant could not
be determined.
3
,4
sensors, so-called fluoroionophores, have been reviewed. In
this frame, fluoroionophores based on coumarin 343 linked to
monoaza-crown ether via a methylene group, were previously
5
–7
studied.
These probes contain an electron-withdrawing
Surprisingly, the stability constants for magnesium with all
these C343-fluoroionophores were found to be very close, in
group, the carbonyl group of the lactone, conjugated to an
electron-donating group, the nitrogen atom of the julolidyl
ring. This gives to the absorption of light a charge transfer
character. Since the carbonyl may interact with the cation
chelated in the aza-crown site, marked photophysical changes
S
acetonitrile (log K 5 4.27, 4.59 and 4.24 for C343-crown,
C343-benzocrown and C343-dibenzocrown respectively). Since
the structural modification of the crown ether does not seem to
influence the stability of the magnesium complex, this cation
was supposed to be chelated by the two carbonyl groups and
not by the crown ether heteroatoms. This kind of supra-
7
are expected upon complexation, and were observed.
Bathochromic shifts and hyperchromicity were recorded in
acetonitrile and ethanol principally in the cases of magnesium
and calcium chelation.
1
1
molecular structure was already suggested by Alonso et al.
It has also been shown that naphthalenic compounds
bearing amide groups may be used as fluorescent chemo-
Photosensing coumarins linked to the aza-crown by an
8
amide group were reported in a subsequent work (C343-
1
2,13
sensors for alkali and alkaline-earth metal ions.
crown shown hereunder), and in a further step the monoaza-
crown receptor was replaced by a monoazabenzo-crown or a
monoazadibenzo-crown in order to modify the selectivity of
Considering these results, we firstly decided to synthesize
and study C343-diethylamide (C343-dea, see structure below),
a fluoroionophore with a similar b-dicarbonyl sub-structure
but without crown ether. In a second step, naphthalenic
fluorophores linked to an aza-crown ether via an amide
group or bearing a diethylamide function and substituted by a
methoxy group were studied. These naphthalenic probes
also contain an electron-donating group, the methoxy group,
conjugated to an electron-withdrawing group, the carbonyl
9
,10
the fluorescent probe.
The consequences of these structural
modifications were examined in terms of photophysical and
complexing properties and indeed it was observed that the
rigidification of the complexing cavity greatly improves the
*habib@chim.ucl.ac.be
2
928 | J. Mater. Chem., 2005, 15, 2928–2937
This journal is ß The Royal Society of Chemistry 2005

View Article Online
amide group, and therefore are expected to behave as charge
transfer sensors. The methoxy group was placed at a position
where it is supposed to participate in the cation coordination
as an ancillary complexing group while in other molecules the
position of the methoxy substituent was changed in order to
suppress the ancillary effect. Greater participation of the
crown ether in the cation chelation should be expected in these
cases, when compared to the coumarinic examples.
salts to 2.5 ml of a ligand solution of the same concentration
2
5
(concentration range: 1–4 6 10
M) in order to avoid
dilution effects. For C343-dea, the absorption spectrum was
recorded after each addition. For the naphthalenic probes,
titrations were done by fluorescence and in that case the
emission spectra were recorded after each cation addition and
were corrected for small variations of absorption if necessary.
The stability constants of complexes were determined from
absorption or fluorescence intensity evolution upon metal
addition using equations reported in ref. 10 and processed by
The aim of this work was to study the photophysical and
complexing properties of the ligands and those of their
+
+
+
E
complexes with alkaline (Li , Na , K ) and alkaline-earth
2+
Origin software.
2
Mg , Ca ) cations in acetonitrile. From these results, it
+
(
The apparatus and method for fluorescence decay times
10
measurements were described in a previous paper.
should be possible to outline the function of the crown
ether and/or the amide bridge in the cation chelation and
we could determine if the methoxy group acts as an auxiliary
complexing group when it is placed in position 3 in the
naphthalenic compounds.
NMR spectra were recorded on a BRUKER Avance 500
spectrometer. NMR spectra of complexes were recorded using
a 10 mm probe. The concentrations of solutions in deuterated
23
1
22
13
acetonitrile were 10 M for H spectra and 10 M for
spectra.
C
Experimental
Infrared spectra were obtained on a Biorad FTS 135
spectrophotometer, using KBr pellets or CDCl solutions.
3
Solvents and reagents
Mass spectra were obtained on a LCQ apparatus with an
Coumarin 343, 3- and 6-methoxy-2-naphthoic acids were
purchased from Aldrich and 1-aza-18-crown-6 was from
Fluka. PyBOP and diisopropylethylamine were ACROS
products. All reagents were used as received. Absorption and
fluorescence spectra were recorded in acetonitrile from Fisher
Scientific (HPLC grade). Lithium, sodium, alkaline-earth
perchlorates and potassium thiocyanate were purchased from
Alpha (highest quality available) and were vacuum-dried over
APCI source (Thermo FINNIGAN, San Jose, CA, USA).
Melting points were determined with a B U¨ CHI apparatus.
HPLC analyses were performed on a SpectraPhysics SP 8800,
equipped with an Econosil normal phase or a C18 Biorad
reversed phase column.
HRMS and ESI-MS analyses of the complexes were
performed in the laboratory of bioorganic mass spectrometry
in the University Louis Pasteur of Strasbourg (Prof. A. Van
Dorsselaer). Spectra of the complexes were obtained on a triple
quadrupole apparatus Quattro II (Micromass, Altringham,
UK). The ESI source was warmed to 70 uC and a potential of
2 5
P O prior to use. Deuterated chloroform and acetonitrile for
NMR measurements were Aldrich products.
Apparatus and methods
1
0 V was applied on the entry cone in order to transfer the
supramolecular species to the gas phase without fragmenta-
tion. Every scan was recorded from 200 to 2000 m/z and
several scans were added to obtain the final spectrum. Every
spectrum was reproduced three times. Ligand concentrations
UV-Vis absorption spectra were obtained on a Varian Cary
5
0 spectrophotometer. Emission spectra were obtained on a
SPEX Fluorolog 1681 spectrofluorometer and on a Varian
Cary Eclipse spectrofluorometer. Fluorescent quantum yields
were determined using naphthalene in cyclohexane as reference
2
4
23
were 10
M and 10
M for 3MN-crown and 6MN-dea
1
4
(
WF 5 0.23 ¡ 0.02, lexc 5 267 nm ) or coumarin 314 in
respectively. These concentrations are higher than those used
for UV-Vis spectroscopy experiments. Significant formation of
complexes is then allowed without using a high cation
concentration that is unfavourable for the ESI process.
1
ethanol (WF 5 0.68, lexc 5 408 nm ).
5
The titrations were realized by progressive additions of small
aliquots of a ligand solution highly concentrated in cations
This journal is ß The Royal Society of Chemistry 2005
J. Mater. Chem., 2005, 15, 2928–2937 | 2929

View Article Online
Synthesis
NMR (CDCl
3
, 500 MHz), d (ppm): 171.56, 158.43, 134.84,
1
32.43, 129.89, 128.25, 127.02, 125.85, 124.66, 119.57, 105.75,
+
C343-dea, 3- and 6MN-dea were obtained from coumarin 343,
5
5.42, 43.23, 39.45, 14.39, 13.07. MS (APCI): 258 (MH ), 185.
2
3
- or 6-methoxy-2-naphthoic acids and diethylamine using the
1
3
FTIR (CDCl ): nCLO: 1613 cm . HPLC (normal phase,
1
6
coupling agent PyBOP. The same method was tested to
synthesize 3- and 6MN-crown but purification difficulties were
encountered due to by-products formed during the reaction.
Similar problems were encountered with DCC as coupling
agent. The amides 3- and 6MN-crown were finally synthesized
by the reaction of the corresponding naphthoic acid with
CH
2 2
Cl –iPrOH 95 : 5), purity: 99.5%. HRMS calculated for
+
[M + H] : 258.1494; found: 258.1494.
Synthesis of 3MN-crown. Thionyl chloride (120 ml, 1.05 eq.)
dissolved in 20 ml of dry CH Cl , was added dropwise to a
2
2
solution of 315 mg (1 eq.) of 3-methoxy-2-naphthoic acid
dissolved in 20 ml of CH Cl containing one drop of
1
-aza-18-crown-6 via the formation of the acyl chloride.
2
2
dimethylformamide. The reaction mixture is refluxed for
5 min. The solvent (40 ml) was eliminated by normal pressure
distillation during which time 20 ml of fresh CH Cl were
added. After cooling, 0.8 ml (3.88 eq.) of Et N were added to
the mixture, followed by dropwise introduction of 488 mg
1.19 eq.) of 1-aza-18-crown-6 dissolved in 10 ml of CH Cl .
Synthesis of C343-dea. A mixture of 498 mg (1 eq.) of
4
coumarin 343, 184 mg (1.4 eq.) of diethylamine, 966 mg
1.8 eq.) of PyBOP and 890 mg (4 eq.) of diisopropylethyl-
amine was stirred overnight at room temperature in 70 ml
of a 1 : 1 mixture of CH Cl and CH CN. After solvent
evaporation, the crude product was purified by column
chromatography on silica (CHCl –iPrOH 9 : 1) resulting in
yellow thick oil that slowly crystallized. Yield: 89%.
2
2
(
3
2
2
3
(
2
2
The mixture was refluxed for 30 min and left stirring overnight
at room temperature. After being washed twice with 20 ml of
aqueous ammonia (pH 9) and three times with 20 ml of
deionised water, the organic phase was evaporated under
reduced pressure. Finally, the crude product is chromato-
3
1
Mp: 148 uC. H NMR (CDCl
H), 6.87 (s, 1H), 3.53 (m, 2H), 3.30 (m, 6H), 2.86 (t, 2H),
3
, 500 MHz), d (ppm): 7.73 (s,
1
2
5
1
3
1
3
.75 (t, 2H), 1.96 (m, 4H), 1.20 (m, 6H). C NMR (CDCl3,
00 MHz), d (ppm): 166.21, 159.64, 151.98, 146.74, 143.16,
25.61, 118.90, 117.02, 107.58, 106.23, 50.10, 49.70, 43.37,
2 2
graphed on silica gel (CH Cl –iPrOH 9 : 1) affording a
colourless oil. Yield: 85%.
1
3
H NMR (CDCl , 500 MHz), d (ppm): 7.75 (m, 2H), 7.69 (s,
9.55, 27.53, 21.36, 20.44, 20.25, 14.29, 12.97. MS (APCI): 341
+
1
4
1
1
H), 7.46 (m, 1H), 7.36 (m, 1H), 7.16 (s, 1H), 3.94 (s, 3H), 3.4–
(
MH ), 300, 268. HPLC (normal phase, CHCl
3
–EtOH 95 : 5),
+
1
.0 (broad, 24H). C NMR (CDCl , 500 MHz), d (ppm):
3
3
purity: 99.1%. HRMS calculated for [M + Na] : 363.1685;
found: 363.1692.
69.56, 153.63, 134.64, 128.49, 128.28, 128.02, 127.48, 127.11,
26.74, 124.37, 105.93, 70.92–69.75, 55.72, 49.55, 46.16. MS
+
21
(
APCI): 448 (MH ), 185. FTIR (CDCl
3
): nCLO: 1627 cm
–EtOH 97.5 : 2.5), purity: 99.9%.
.
Synthesis of 3MN-dea. A mixture of 296 mg (1 eq.) of
HPLC (normal phase, CHCl
3
3
8
-methoxy-2-naphthoic acid, 220 ml (1.46 eq.) of diethylamine,
+
HRMS calculated for [M + H] : 448.2335; found: 448.2339.
30 mg (1.10 eq.) of PyBOP and 2 ml (7.84 eq.) of
diisopropylethylamine was stirred overnight at room tempera-
ture. After solvent evaporation, the crude product was purified
Synthesis of 6MN-crown. The fluoroionophore 6MN-crown
was synthesised following the method used for 3MN-crown
from 125 ml (1.14 eq.) of thionyl chloride, 315 mg (1 eq.) of
by column chromatography on silica (CHCl –iPrOH 9 : 1)
3
and was recrystallized in ethyl acetate to give white crystals.
6
4
-methoxy-2-naphthoic acid, 0.8 ml (3.88 eq.) of Et N and
3
Yield: 96%.
88 mg (1.19 eq.) of 1-aza-18-crown-6. A colourless oil was
1
Mp: 111 uC. H NMR (CDCl , 500 MHz), d (ppm): 7.75 (m,
3
obtained. Yield: 66%.
1
2
H), 7.67 (s, 1H), 7.46 (m, 1H), 7.36 (m, 1H), 7.16 (s, 1H), 3.93
H NMR (CDCl , 500 MHz), d (ppm): 7.85 (s, 1H), 7.75 (d,
3
(
1
s, 3H), 3.62 (m, 2H), 3.16 (m, 2H), 1.29 (t, 3H), 1.03 (t, 3H).
3
1
3
5
1
5
H), 7.75 (d, 1H), 7.48 (m, 1H), 7.17 (m, 1H), 7.13 (m, 1H),
1
3
C NMR (CDCl , 500 MHz), d (ppm): 168.40, 153.73, 134.45,
3
3
.94 (s, 3H), 3.5–3.9 (broad, 24H). C NMR (CDCl ,
1
5
28.75, 128.42, 127.87, 126.88, 126.88, 126.59, 124.19, 105.72,
+
00 MHz), d (ppm): 173.54, 158.54, 134.96, 132.02, 130.02,
5.57, 42.85, 38.91, 13.92, 12.92. MS (APCI): 258 (MH ), 185.
2
28.27, 127.08, 126.53, 125.08, 119.61, 105.80, 70.83–69.82,
+
1
FTIR (KBr pellet): nCLO: 1624 cm . HPLC (normal phase,
CHCl –EtOH 97.5 : 2.5), purity: 99.4%. HRMS calculated for
5.49, 50.32, 46.34. MS (APCI): 448 (MH ), 185. FTIR
21
3
(
CDCl ): n
: 1623 cm . HPLC (normal phase, CHCl3–
CLO
+
3
[
M + Na] : 280.1313; found: 280.1309.
EtOH 97.5 : 2.5), purity: 99.8%. HRMS calculated for [M +
Na] : 470.2155; found: 470.2167.
+
Synthesis of 6MN-dea. A mixture of 305 mg (1 eq.) of
6
8
-methoxy-2-naphthoic acid, 220 ml (1.41 eq.) of diethylamine,
Results and discussion
67 mg (1.14 eq.) of PyBOP and 1 ml (3.92 eq.) of
diisopropylethylamine was stirred for 30 h at room tempera-
ture. After solvent evaporation, the crude product was
A) C343-dea
chromatographed twice on a silica column (CHCl
: 1 and Et O–CHCl 1 : 1), yielding the product as a colour-
less oil. Yield: 82%.
3
–iPrOH
Absorption. The absorption spectra of C343-dea as free
ligand and as complexes with various alkaline and alkaline-
earth cations are reported in Fig. 1 and detailed in Table 1.
Complexation by a cation induces a bathochromic shift as
expected and a hyperchromicity as already observed for C343-
9
2
3
1
3
H NMR (CDCl , 500 MHz), d (ppm): 7.80 (s, 1H), 7.75 (d,
1
3
H), 7.75 (d, 1H), 7.44 (dd, 1H), 7.18 (dd, 1H), 7.14 (m, 1H),
1
3
8
+
+
+
2+
.93 (s, 3H), 3.2–3.7 (broad, 4H), 1.1–1.3 (broad, 6H).
C
crown. Among the tested cations (Li , Na , K , Mg and
2
930 | J. Mater. Chem., 2005, 15, 2928–2937
This journal is ß The Royal Society of Chemistry 2005

View Article Online
Table 2 Emission properties of free and complexed C343-dea in
acetonitrile
Charge
Radius/A density/q A
2
1
˚
˚
lem/nm
W
F
F i
t /ns (f )
Ligand
Li
Mg
472
491
501
0.56 3.70
0.50
+
0.90
0.86
1.11
2.32
2+
0.52 4.18 (0.92)
.34 (0.08)
0.51 4.18 (0.96)
.61 (0.04)
fluorescence quantum yield; t:
: fractional contribution to the steady-state
1
2+
Ca
1.14
1.75
497
0
a
em F
l : emission maximum; W :
fluorescence lifetime; f
intensity (satisfactory values—below 1.2—of the reduced x were
obtained in all cases).
i
2
Fig. 1 Absorption spectra of C343-dea: (a) free ligand; (b) its
+
2+
2+
complex with Li ; (c) its complex with Ca ; (d) its complex with Mg
.
2+
+
+
Ca ) only Na and K do not induce spectral modifications.
The observed properties are very close to that registered for
8
C343-crown.
,10
Emission. The emission properties of C343-dea and its com-
plexes in acetonitrile are reported in Table 2 and illustrated in
Fig. 2.
Emission maxima of C343-dea and its complexes are close
to the maxima of C343-crown but the fluorescence quantum
yields are somewhat lower (0.56 for C343-dea and 0.65 for
1
0
C343-crown ). This might be due to a greater flexibility of the
diethylamine group compared to the aza-crown ether.
The fluorescence decay time of free C343-dea is mono-
exponential, but in the cases of complexes a biexponential
decay is observed consisting of a longer (4.18 ns) and a shorter
decay time (1.3 ns for the magnesium complex and 0.6 ns for
the calcium complex). The longer decay time that represents
more than 90% of the decay may be tentatively assigned to a
simple 1 : 1 complex in which the metal chelation induces a
rigidification of the structure. The short component might
reveal the presence of a second kind of complex with another
stoichiometry for which the close proximity of the two
coumarinic rings leads to some fluorescence quenching.
Fig. 2 Emission spectra of C343-dea: (a) free ligand; (b) its complex
+
2+
2+
with Li ; (c) its complex with Ca ; (d) its complex with Mg
.
B) Naphthalenic fluoroionophores
Absorption. The absorption spectra of the four naphthalenic
ligands show that the replacement of the diethylamine by the
aza-crown ether does not significantly modify the ligand
absorption spectra. The modifications of absorption spectra
upon addition of the tested cations are very small and will not
be detailed here.
Table 1 Absorption properties of free and complexed C343-dea in
acetonitrile
Emission. The emission characteristics of the four ligands
and their complexes are presented in Table 3. Quantum yields
and fluorescence lifetimes were only determined when full
complexation could be reached. This is not always possible due
to the weak stability of some complexes. In the other cases,
the observed trends in the photophysical changes induced by
the cations are indicated.
2
3
Radius/
˚
Charge
density/q A
l
max
nm
/
l
iso
nm
/
e 6 10 /
M
2
1
21 21
˚
A
cm
Ligand
Li
407.5
418.5
457.0
443.0
26.0
26.5
40.7
33.7
+
0.90
0.86
1.14
1.11
2.32
1.75
410.5
424.0
418.5
2
+
Mg
Ca
2
+
As for the absorption spectra, the replacement of the
diethylamine by the crown ether does not influence the
emission properties.
a
lmax: maximum absorption wavelength; liso: isosbestic point, e:
extinction coefficient.
This journal is ß The Royal Society of Chemistry 2005
J. Mater. Chem., 2005, 15, 2928–2937 | 2931

View Article Online
Table 3 Emission properties of naphthalenic ligands and their complexes in acetonitrile
Cations 3MN-dea
6MN-dea
em/nm
2
1
˚
Radius/A
˚
Charge density/q A
lem/nm
W
F
t
F
/ns (f
i
)
l
W
F
F i
t /ns (f )
Ligand
367
0.003
0.11 (0.30)
.28 (0.70)
352
0.022
0.07 (0.01)
1.40 (0.99)
0
+
Li
Na
0.90
1.16
1.52
0.86
1.14
1.11
0.86
0.66
2.32
1.75
367
367
367
393
389
q
—
—
—
—
—
353
353
353
354
352
Q
—
—
—
—
—
+
0.003
0.003
qqq
qq
0.022
0.022
QQQ
QQ
+
K
Mg
2
+
2
Ca
+
3
MN-crown
6MN-crown
em/nm
lem/nm
W
F
t
F
/ns (f
i
)
l
W
F
F i
t /ns (f )
Ligand
367
0.007
0.15 (0.18)
.60 (0.82)
—
0.28 (0.16)
1.66 (0.84)
—
353
0.019
0.23 (0.02)
1.42 (0.99)
—
0.99 (0.19)
1.90 (0.81)
—
0.10 (0.21)
0.28 (0.34)
0.84 (0.45)
0.17 (0.02)
1.23 (0.37)
3.34 (0.61)
0
+
Li
Na
0.90
1.16
1.11
0.86
368
370
qq
0.016
353
353
q
0.020
+
+
K
Mg
1.52
0.86
0.66
2.32
369
416
q
0.055
353
400
0.019
0.024
2+
0.63 (0.04)
4
9
.03 (0.34)
.40 (0.62)
2+
Ca
1.14
1.75
376
0.028
0.20 (0.02)
354
0.025
0
2
.78 (0.16)
.19 (0.82)
a
F
lem: emission maximum; W : fluorescence quantum yield; q, Q: increase, decrease of the fluorescence intensity (when full complexation
cannot be reached); t: fluorescence lifetime; f
i
: fractional contribution to the steady-state intensity (satisfactory values—below 1.2—of the
2
reduced x were obtained in all cases).
The molecules bearing a diethylamide function are able to
interact only with cations of high charge density. A variation
of the fluorescence intensity is induced by these cations with
the effect being in opposite directions for 3MN-dea (increase
of intensity) and 6MN-dea (decreasing intensity). This
opposite effect will be discussed later. A red shift of 26 and
22 nm is observed for 3MN-dea when the solution is saturated
with magnesium and calcium cations respectively.
The emission spectra of probes bearing an aza-crown ether
are modified by almost all the tested cations and full com-
+
2+
2+
plexation can be reached for both probes with Na , Mg , Ca
(Fig. 3 and Fig. 4). In all these cases, the metal chelation
2
5
25
Fig. 4 Emission spectra of (a) 6MN-crown (c 5 5.8 6 10 M) and
its complexes (at full complexation) with (b) Na ; (c) Ca ; (d) Mg in
Fig. 3 Emission spectra of (a) 3MN-crown (c 5 5.9 6 10 M) and
its complexes (at full complexation) with (b) Na ; (c) Ca ; (d) Mg in
+
2+
2+
+
2+
2+
acetonitrile; lexc: 315 nm.
acetonitrile; lexc: 267 nm
2
932 | J. Mater. Chem., 2005, 15, 2928–2937
This journal is ß The Royal Society of Chemistry 2005

View Article Online
induces an increase of the fluorescence intensity. Large red
shifts of 49 and 47 nm are observed for magnesium complexes
with 3- and 6MN-crown respectively.
The observed bathochromic shifts increase significantly with
the charge density of the cation. This is coherent with the
stronger interaction between the ligand carbonyl group and
the cation and the greater resulting stabilisation of the excited
charge transfer state. The variation of fluorescence intensity
is also related to the cation charge density. For most of the
complexes, cation binding leads to an increase of the
fluorescence quantum yield. It is accepted that naphthalenic
or aromatic compounds having amide groups are strongly
quenched by intersystem crossing to a triplet sate and/or by the
rotational relaxation linked to excited state rotations around
the three naphthalene–CO, CO–NH and N–alkyl bonds. This
may explain why, when complexed by cations, these rotors are
1
3
less available for relaxation. An increase of the rigidity of the
system by cation complexation may therefore be an explana-
tion for the fluorescence enhancement. The case of 6MN-dea
will be discussed later.
Two exponentials were needed in all cases to fit the
experimental decay curves. In the case of the free ligands,
both contributions are significant for 3MN derivatives while
the shorter component may be considered negligible for the
Fig. 5 Evolution of the emission spectrum of (a) 3MN-crown (c 5
25
25
25
1
1
4
4
.8 6 10 M) upon NaClO (c 5 0; 1.2 6 10 ; 2.3 6 10 ; 3.5 6
25 25 25 25 24 24
0
; 4.6 6 10 ; 6.9 6 10 ; 9.8 6 10 ; 1.4 6 10 ; 2.2 6 10 ;
24 24 23
.0 6 10 ; 9.5 6 10 ; 2.8 6 10 M) addition, lexc 5 330 nm.
6
MN derivatives. This means that the methoxy group, when
close to the amide group, is probably responsible for an
excited state conformational equilibrium leading to two
different emitting conformers. This seems coherent with the
observed since the stability constant of the complex drops
from log K 5 3.26 for C343-crown to 1.7 for C343-dea.
s
Calcium and magnesium cations are complexed by 3- and
observed increase of W when these rotations are slowed down
F
6
MN-dea with stability constants of the same order of
in the presence of the cation. In the case of the complexes, the
situation is still more complicated which might be the sign of
the presence of different complexes with various structures or
stoichiometries.
magnitude. Since the position of the methoxy group does not
modify the stability of these complexes, it seems that no
ancillary chelation occurs with this substituent. Although
satisfactory fits are obtained using 1 : 1 (metal : ligand)
stoichiometry for 6MN-dea complexes, these complexes
C) Stoichiometry and stability constants of the complexes in
acetonitrile
The stability constants were obtained from the evolution of
the absorption or emission spectra upon cation addition (cf.
Apparatus and methods section). An example of the evolution
4
of 3MN-crown emission spectrum upon NaClO addition and
the corresponding titration curve are presented in Fig. 5 and 6.
The ML stoichiometry was firstly tested in every case and
found to be adapted to most of the complexes. The stability
constants calculated for that stoichiometry are given in Table 4
and defined according to eqn. (1):
½
MLꢀ
MzL'ML Ks~
(1)
½
Mꢀ|½Lꢀ
The stability constant of the C343-dea magnesium complex
is very close to the values obtained for the C343-crown
derivatives. This clearly demonstrated that the b-dicarbonyl
sub-structure of our coumarinic compounds is responsible for
the chelation of magnesium cation. For the calcium complex, it
is not possible to simply compare this new result with the
previous ones obtained for the crowned fluoroionophores
since complexes with a stoichiometry differing from 1 : 1 were
observed for C343-crown and C343-benzocrown. Lithium
cation weakly binds C343-dea. For lithium, a drastic effect is
25
Fig. 6 Titration curve of 3MN-crown (c 5 1.8 6 10
NaClO at 371 nm and fit to 1 : 1 stoichiometry.
M) by
4
This journal is ß The Royal Society of Chemistry 2005
J. Mater. Chem., 2005, 15, 2928–2937 | 2933

View Article Online
Table 4 Stability constants of complexes according to the hypothesis of 1 : 1 stoichiometry
Log K
s
C343-dea
3MN-dea
6MN-dea
3MN-crown
6MN-crown
+
a
a
a
a
a
a
a
Li
Na
1.70 ¡ 0.01
2.61 ¡ 0.02
4.13 ¡ 0.01
2.62 ¡ 0.02
(3.78 ¡ 0.02)
5.26 ¡ 0.01
+
a
3.95 ¡ 0.02
+
a
a
K
Mg
2
+
b
b
4.46 ¡ 0.01
3.81 ¡ 0.01
2.55 ¡ 0.02
2.63 ¡ 0.02
2.98 ¡ 0.02
2.92 ¡ 0.02
b
(4.34 ¡ 0.01)
5.71 ¡ 0.04
2
+
Ca
a
Cation-induced spectroscopic changes too small to determine a constant. Stoichiometry 1 : 1 used for calculation but found inadequate by
visual inspection of the fitted curve.
showed a surprising decrease of fluorescence intensity we have
not explained yet. A proposed hypothesis is the formation of
complexes of stoichiometry 2 : 2 and not 1 : 1. Indeed, in this
case, a cation would be complexed by the methoxy group of
2 1 2
K 5 9.14 ¡ 0.02, K 5 3.63 ¡ 0.02 and K 5 7.90 ¡ 0.08 for
the same complexes, respectively. Although these stoichiome-
tries (proposed structure of M L showed below) fit satisfac-
3
2
torily with experimental data, they cannot be considered as
established on the basis of these fittings.
6
MN-dea and at the same time by the carbonyl group of
another ligand molecule. This would bring the aromatic
rings close together and could lead to the formation of less
fluorescent excimeric complexes.
Stability constants are of the same order of magnitude for
3
3
- and 6MN-crown with calcium cations. As mentioned for
- and 6MN-dea, the methoxy group does not seem to
participate in the complexation. Normally the 18-crown-6
ether cavity is adapted for cations of the size of potassium
cations. But it appears that smaller cations, like sodium cation,
are more strongly bound by this aza-crown cavity than
potassium cation. We suppose that the oxygen atoms of the
crown ether are coordinating the cation. The nitrogen atom,
whose electron-donating character is diminished by its con-
jugation in the amide link, is probably not involved in the
cation complexation, leaving a smaller chelating cavity. Such a
sodium complex of 18-membered ring crown amide was
studied in ref. 17 by X-ray crystallography. In this solid state
structure, the sodium cation is coordinated in the macroring
and the non-coordinating amide nitrogen atom is puckered
outward. Furthermore this X-ray study reveals that this
sodium complex is in a dimeric form with the carbonyl group
of another ligand molecule used to fulfil the coordination
sphere of the sodium cation.
D) NMR experiments on the complexation of 3MN-crown and
6
MN-dea
NMR spectroscopy can be used to determine if a particular
8
1
group of a ligand participates in cation complexation or not.
Two complexes were studied by NMR: 3MN-crown with Na
+
The 1 : 1 stoichiometry is not adapted for magnesium
complexes of 3- and 6MN-crown. Better results were obtained
considering M L and M L stoichiometries. Stability constants
2
+
and 6MN-dea with Mg . The first one was chosen because of
its high stability constant and its unexpected complexation
selectivity. The second one was studied because of its parti-
cular photophysical behaviour since a decrease of fluorescence
intensity is observed upon metal chelation while an increase
was observed in all other studied cases. A possible explanation
for this different behaviour would be the formation of a 2 : 2
complex. NMR experiments were done to probe this proposal.
2
3 2
for these stoichiometries are defined as follows:
MLꢀ
Mꢀ|½Lꢀ
½
MzL'ML K1~
(2)
(3)
½
½
M3L2ꢀ
2MLzM'M3L2 K2~
2
½
MLꢀ |½Mꢀ
1
H NMR spectra of 3MN-crown and its complex with
or
sodium were recorded. The protons of the crown ether part of
the ligand appeared as a set of multiplets from 3.30 ppm to
½
M2Lꢀ
MLzM'M2L K2~
(4)
3
4
.78 ppm and are shifted, upon sodium chelation, from 3.50 to
.08 ppm (Fig. 7). A singlet corresponding to the methoxy
½
MLꢀ|½Mꢀ
With the hypothesis of M
2
L stoichiometry, the magnesium
group appeared at 3.92 ppm and was not affected by the
presence of sodium cation. No shift was observed for aromatic
signals. This confirms our view of this complex: the sodium
cation is well fitted in the crown ether cavity and the methoxy
group is not involved in the cation chelation.
complex stability constants obtained were K 5 3.77 ¡ 0.03
1
and K 5 3.96 ¡ 0.02 for 3MN-crown, K 5 4.35 ¡ 0.04
1
2
and K
2
5 4.45 ¡ 0.01 for 6MN-crown. In the case of
M
3
L
2
stoichiometry, the results were K
1
5 3.69 ¡ 0.02 and
2
934 | J. Mater. Chem., 2005, 15, 2928–2937
This journal is ß The Royal Society of Chemistry 2005

View Article Online
1
Fig. 7 Part of H NMR spectra of A: 3MN-crown (1.41 10 M) and B: its complex with Na .
23
+
1
3
Table 5 Chemical shift differences between C peaks of 3MN-crown
(
electron-donating part of the sensor. Moreover, Van den
20
Bergh et al. reported that, for calcium complexation by a
2
3
+
c 5 9.74 6 10 M) and its complex with Na
fluoroionophore, excited and ground state stability constants
may be strongly different.
1 13
H and C NMR spectra of 6MN-dea and its complex with
magnesium are shown in Fig. 8 and Table 6.
The formation of a 2 : 2 complex (as previously supposed) is
expected to lead to peak shifts for the carbonyl and methoxy
groups. The peaks of the carbonyl group and atoms nearby
are shifted, showing that the cation is complexed by this
13
C
1
13
Dd (ppm)
group. However, the positions of the H and C peaks of the
methoxy group are not influenced by the presence of the cation
and the hypothesis of a 2 : 2 complex is therefore to be rejected.
Furthermore, the rotation of the C–N amide link is shown by
the NMR spectra to be difficult due to the double bond
character of this link and to be completely hindered when the
cation is coordinated to the carbonyl group: two well defined
quadruplets of two non-equivalent ethyl groups are apparent
in the complex while they are clearly less discriminated in the
free ligand. This is an observation that, on one hand, confirms
the rigidification of such a system by complexation but, on the
other hand, does not explain the fluorescence decrease
observed in the case of the 6MN-dea magnesium complex.
1
3
5
1
8
4
1
1
1
2
20.13
20.12
0.12
20.34 to +0.05
0.12
0.15
20.16 to +2.61
0.02
20.24
, 2, 6, 7, 9, 10
4 to 23
1
3, 24
21.44
1
3
C NMR spectra of the same complex showed significant
1
3
shifts for C of the crown ether but not for the methoxy group
nor for the carbonyl group (Table 5). Therefore, it is concluded
that the cation does not interact with the carbonyl group in the
ground state. But, since red shifts of the emission spectrum and
fluorescence enhancement are observed upon metal chelation
by 3MN-crown, we supposed that the cations interact with the
carbonyl group in the excited state. So, after light excitation,
the ligand undergoes intramolecular charge transfer and the
electronic density on the carbonyl group is increased. This may
yield the formation of a tighter complex with the cation
located closer to the carbonyl group in the excited state. This
effect would be in the opposite direction compared to the
E) Mass spectrometry analysis of 3MN-crown and 6MN-dea in
2+
the presence of Mg
Mass spectrometry measurements were performed on com-
plexes of 3MN-crown and 6MN-dea with magnesium, trying
to get further information about stoichiometries: fluorimetric
2+
titration of 3MN-crown by Mg could not be fitted as a 1 : 1
complex and a 2 : 2 stoichiometry was considered (and
disproved by NMR measurements) in the case of the
magnesium complex of 6MN-dea. Mass spectrometry is used
in supramolecular chemistry thanks to electrospray ionisation
(ESI). This technique is a soft ionisation technique and allows
the supramolecular species to be transferred from the solution
1
9
cation photoejection reported by Martin et al. In this case,
the fluorescent sensor was built in a reversed way compared to
our probes: their fluorophore presents also a light induced
charge transfer character but the cation interacts with the
2
1–23
to the gas phase.
This journal is ß The Royal Society of Chemistry 2005
J. Mater. Chem., 2005, 15, 2928–2937 | 2935

View Article Online
1
23
Fig. 8 Part of H NMR spectra of A: 6MN-dea (c 5 1.28 6 10 M) and B: its complex with magnesium.
1
3
Table 6 Chemical shift differences between C peaks of 6MN-dea
(
(stoichiometry 1 : 1) in the case of 6MN-dea with magnesium
22
c 5 1.01 6 10 M) and its complex with magnesium
contains four molecules of solvents. More extra ligands are
present in this case as can be expected since 6MN-dea offers
fewer coordination sites than the 3MN-crown ligand.
The role of these co-ligands is very often neglected in this
type of studies and this may well be of importance for
explaining why stoichiometries are not always easy to
determine: small spectral differences may be expected between
two 1 : 1 complexes according to the nature of the co-ligands
13
C
13
C
Dd (ppm)
Dd (ppm)
(
acetonitrile, residual water molecule or anion) and the type of
: 1 complex that is formed may change during the titration.
1
7
5
2
9
1
4
2
1.35
0.64
0.50
23.54
0.18
3
8
6
20.44
0.59
0.21
0.14
1.21
1
Association with solvent molecules is expected to dominate at
the beginning but anions may be more important at the end of
the titration if a large excess of salt is necessary to reach full
complexation, which is often the case especially in absorption
or emission titrations. Of course in ESI-MS experiments, the
evaporation step may influence the nature of the co-ligands
according to their relative volatility.
11
15 + 13
0
20.30
0.20
1.74
20.35
16 + 14
In the case of the magnesium complex of 3MN-crown, this
preliminary study tends to show that the stoichiometry 1 : 1 is
The ESI-MS experiments reveal the presence of complexes
of different stoichiometries and the possible change of co-
ligands. Both may explain the deviations observed in fittings
in the majority as indicated by signals at m/z 256.3
+
2+
(
3 4 3 2 2
[LMgCH CN] ) and 570.0 ([LMgClO ] ). M L and M L
were not found and this leads us to reject our hypothesis. In
the same way, and confirming the NMR results, the mass
spectra of magnesium complex of 6MN-dea never showed a 2 : 2
complex but the 1 : 1 complex is once again the major species
(
see for instance note b of Table 4).
Conclusion
2+
as recorded by the signals at m/z 222.6 ([LMg(CH CN) ] )
3
The close similarity between the complexation constants
found for C343-dea and in preceding works for C343-crown,
C343-benzocrown and C343-dibenzocrown clearly demon-
strates that the b-dicarbonyl substructure common to these
molecules is responsible for the magnesium complexation and
that the aza-crown cavity does not seem to play a role in the
alkaline-earth cation chelation. Things may be less clear-cut
for calcium complexes where competitive complexation
seems possible. With the naphthalenic molecules it is clearly
demonstrated that amide functions, when alone, are still able
to complex calcium and magnesium but with weaker
constants, while alkali metal ions do not affect the photo-
physics of these molecules.
4
+
and 462.2 ([LMg(CH
or ML (for instance, [(6MN-dea)
10.2 or [(6MN-dea)
in both cases and in a larger proportion for 6MN-dea than for
MN-crown complex because this first ligand does not have a
3
CN)
2
ClO
4
] ). Minority species like ML
2
2+
3
2
Mg(CH
3
CN)
2
]
at m/z
2+
3
3 3
MgCH CN] at m/z 418.4) are present
3
well defined ionophore.
It is also important to note that the observed complexes of
+
2
3
MN-crown with magnesium are as follows: [LMgCH
+
3
CN]
4
] . It seems that the coordination sphere of
and [LMgClO
magnesium cannot be completed by the chelating atoms of
MN-crown and the cation needs an extra ligand that can be
the solvent or the anion. The major observed complex
3
2
936 | J. Mater. Chem., 2005, 15, 2928–2937
This journal is ß The Royal Society of Chemistry 2005

View Article Online
3
4
B. Valeur, in Topics in Fluorescence spectroscopy Vol 41: Probe
design and chemical sensing, ed. J. R. Lakowicz, Plenum, New York,
1994, pp. 21–48.
B. Valeur and I. Leray, Coord. Chem. Rev., 2000, 205, 3–40.
Even if this group was shown to help complexation in other
2
4
cases, no ancillary chelating effect of the methoxy substituent
was demonstrated in the 3MN-fluoroionophores.
In almost all cases a 1 : 1 stoichiometry has been
demonstrated. NMR and mass spectrometry measurements,
for the cases studied, give arguments favouring a major 1 : 1
complex even when the fluorimetric titration does not fit for
this stoichiometry. The presence of minor complexes of
different stoichiometries and the possibility of changes in the
co-ligands are possible reasons for the deviations observed in
certain cases.
5 J. Bourson, M. N. Borrel and B. Valeur, Anal. Chim. Acta, 1992,
57, 189–193.
J. Bourson, F. Badaoui and B. Valeur, J. Fluoresc., 1994, 4,
75–277.
J. Bourson, J. Pouget and B. Valeur, J. Phys. Chem., 1993, 97,
4552–4557.
2
6
7
8
9
2
J.-L. Habib Jiwan, C. Branger, J.-Ph. Soumillion and B. Valeur,
J. Photochem. Photobiol. A, 1998, 116, 127–133.
I. Leray, J.-L. Habib Jiwan, C. Branger, J.-Ph. Soumillion and
B. Valeur, J. Photochem. Photobiol. A: Chem., 2000, 135, 163–169.
0 D. Taziaux, J.-Ph. Soumillion and J.-L. Habib Jiwan, J. Photochem.
1
1
1
1
Photobiol. A: Chem., 2004, 162, 599–607.
Acknowledgements
1 M.-T. Alonso, E. Brunet, O. Juanes and J.-C. Rodrigues-Ubis,
J. Photochem. Photobiol. A: Chem., 2002, 147, 113–125.
2 J. Kawakami, A. Fukushi and S. Ito, Chem. Lett., 1999,
955–956.
The authors gratefully acknowledge the ‘‘Fonds National de la
Recherche Scientifique’’ for its financial support. L. M. is a
Research Fellow from the FNRS. Professor Alain Van
Dorsselaer, Dr Emmanuelle Leize and Haiko Herschbach
are thanked for fruitful collaboration in mass spectrometry.
The authors thank also Dr David Chapon for NMR
measurements.
3 J. Kawakami, R. Miyamoto, A. Fukushi, K. Shimozaki and S. Ito,
J. Photochem. Photobiol. A: Chem., 2002, 146, 163–168.
4 D. F. Eaton, Pure Appl. Chem., 1988, 60, 7, 1107–1114.
5 G. A. Reynolds and K. H. Drexhage, Opt. Commun., 1975, 13, 222.
1
1
16 J. Coste, D. Le-Nguyen and B. Castro, Tetrahedron Lett., 1990, 31,
, 205–208.
2
1
1
1
2
7 E. S. Meadows, L. J. Barbour, R. Ferdani and G. W. Gokel,
J. Supramol. Chem., 2001, 1, 111–115.
8 T. Gunnlaugson, M. Nieuwenhuyzen, L. Richard and V. Thoss,
J. Chem. Soc., Perkin Trans. 2, 2002, 141–150.
9 M. M. Martin, P. Plaza, N. Dai Hung, Y. H. Meyer, J. Bourson
and B. Valeur, Chem. Phys. Lett., 1993, 202, 5, 452–430.
0 V. Van den Bergh, N. Boens, F. C. De Schryver, M. Ameloot,
P. Steels, J. Gallay, M. Vincent and A. Kowalczyck, Biophys. J.,
Laetitia Maton, Doroth e´ e Taziaux, Jean-Philippe Soumillion and
Jean-Louis Habib Jiwan*
Universit e´ Catholique de Louvain, unit e´ CMAT, 1 Place Louis Pasteur,
B-1348 Louvain-la-Neuve, Belgium. E-mail: habib@chim.ucl.ac.be
References
1
995, 68, 3, 1110–9.
1
Chemosensors of ion and molecule recognition, NATO ASI series,
ed. J.-P. Desvergnes, A. W. Czarnik, Kluwer Academic Publishers,
Dordrecht, 1997.
21 C. A. Schalley, Mass Spectrom. Rev., 2001, 20, 253–309.
22 W. M. David and J. S. Brodbelt, J. Am. Soc. Mass Spectrom., 2003,
14, 383–392.
2
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, 1515–1566.
23 S. M. Blair, J. S. Brodbelt, A. P. Marchand, H.-S. Chong and
S. Alihodzic, J. Am. Soc. Mass Spectrom., 2000, 11, 884–891.
24 A. Minta and R. Y. Tsien, J. Biol. Chem., 1998, 264, 32, 19449–57.
This journal is ß The Royal Society of Chemistry 2005
J. Mater. Chem., 2005, 15, 2928–2937 | 2937