Specific recognition of fluoride anion using a metallamacrocycle incorporating a uranyl-salen unit

Article abstract of DOI:10.1039/b806149a

The design and synthesis of a novel fluoride receptor that uses a salen-complexed Lewis acidic uranyl center as the sole binding site is reported here. This receptor binds fluoride anions in DMSO with a high affinity constant (K > 10⁶ M^-1) and exhibits a negligible affinity (K < 10 M^-1) towards otherwise effective competitors, such as acetate, phosphate and cyanide anions. The Royal Society of Chemistry and the Centre National de la Recherche Scientifique.

Full text of DOI:10.1039/b806149a

View Article Online / Journal Homepage / Table of Contents for this issue
LETTER
www.rsc.org/njc | New Journal of Chemistry
Specific recognition of fluoride anion using a metallamacrocycle
incorporating a uranyl-salen unitwz
Massimo Cametti,ab Antonella Dalla Cort,a Luigi Mandolini,*a Maija Nissinenb
and Kari Rissanen*b
Received (in Montpellier, France) 11th April 2008, Accepted 25th April 2008
First published as an Advance Article on the web 19th May 2008
DOI: 10.1039/b806149a
The design and synthesis of a novel fluoride receptor that uses a
salen-complexed Lewis acidic uranyl center as the sole binding
site is reported here. This receptor binds fluoride anions in
to hard anions, as shown by studies in solution and in the solid
state.5 The affinity of the parent uranyl-salophen compound,
1, for fluoride is revealed by the orange to bright yellow color
change induced by the addition of tetrabutylammonium fluor-
ide (TBAF) to a 1 mM solution of 1 in DMSO. The reversible
nature of the interaction was proved by the finding that the
addition of water lead to the recovery of the original absorp-
tion spectrum of 1 (Fig. S1w). This is easily understood in
terms of hydrogen bonding interactions between water and
fluoride anions, and possibly by the coordination of an oxygen
of a water molecule to the uranyl center. The spectroscopic
changes observed when a very dilute solution of 1 was treated
with increasing quantities of TBAF is shown in Fig. 1. There is
a definite tendency for the sigmoid behaviour of the titration
curves to become less pronounced upon increasing the analyte
concentration (Fig. S2w). When the concentration approaches
1 mM, the titration profiles are sharp curves, reaching a
plateau at 1 equiv. of added fluoride (Fig. 2). The same
DMSO with a high affinity constant (K 4 106
M
ꢀ1) and
exhibits a negligible affinity (K o 10 Mꢀ1) towards otherwise
effective competitors, such as acetate, phosphate and cyanide
anions.
The molecular recognition of fluoride anion is currently
attracting great interest due to its chemical and biological
importance, as witnessed by a host of papers on the subject
published in the last few years.1 The construction of specific
chemosensors calls for the design and synthesis of receptors
capable of binding fluoride anions with both high efficiency
and selectivity. In the vast majority of reported systems,
fluoride anion is coordinated using hydrogen bonding inter-
actions.2 Less numerous, yet still well documented, are recep-
tors containing a Lewis acidic center, such as boron or silicon.3
More rarely reported are those receptors based on a metal
center, either alone or in combination with additional binding
sites.4 Here, we report a highly effective and selective neutral
receptor that makes use of an immobilized UO22+ dication as
the sole binding site for fluoride.
1
behavior is experienced in the H NMR titration reported in
Fig. S3.w
This finding, combined with the existence of sharp isosbestic
points (Fig. 1), clearly indicates that during the titration,
receptor 1 is a two state system, and that only one strong
complex of 1 : 1 composition is formed with fluoride anions.6
No spectral changes were observed upon adding 200 mol
equiv. of the tetrabutylammonium (TBA) salts of chloride,
2ꢀ
When complexed with the tetradentate N2O2
unit of
either salen or salophen ligands, the uranyl dication uses its
fifth coordination site in the equatorial plane to bind strongly
a
`
IMC-CNR and Dipartimento di Chimica, Universita ‘‘La Sapienza’’,
Box 34, Roma 62, 00185 Roma, Italy.
E-mail: luigi.mandolini@uniroma1.it; Fax: +39 06 490421;
Tel: +39 06 49913624
b Nanoscience Center, Department of Chemistry, University of
Jyvaskyla, PO Box 35, FIN- 40014 JYU, Finland.
¨
¨
E-mail: Kari.Rissanen@jyu.fi; Fax: +358 142602651;
Tel: +358 142602672
w Electronic supplementary information (ESI) available: Typical UV-
vis and NMR titrations; X-ray data; 1H and 13C NMR spectra of
receptors 2 and 3. See DOI: 10.1039/b806149a
Fig. 1 UV-vis spectral changes of 3.9 ꢁ 10ꢀ5 M 1 (DMSO, 25 1C)
upon addition of TBAF. The inset shows the spectral changes
at 411 nm.
z CCDC 670306. For crystallographic data in CIF or other electronic
format see DOI: 10.1039/b806149a
ꢂc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2008 New J. Chem., 2008, 32, 1113–1116 | 1113

View Article Online
Table 1 Log K values (K/Mꢀ1) for the association of receptors 1–3
with various anions in DMSO at 25 1C
Aniona
AcOꢀ
H3PO4
CNꢀ
Fꢀ
1
2
3
3.41 ꢄ 0.03
4.00 ꢄ 0.05
2.36 ꢄ 0.05
6.4 ꢄ 0.1b
3.62 ꢄ 0.04
2.32 ꢄ 0.08
2.32 ꢄ 0.08
6.5 ꢄ 0.2b,c
o1
o1
o1
46d
ꢀ
a
b
TBA salts (except for acetate, which is its TMA salt). From
competitive titrations in the presence of excess TMAAcO (see
c
ESI). Titration of dilute solutions of 2 displayed a sigmoid behavior
d
analogous to that seen in Fig. 1. See text.
Fig. 2 UV-vis titration of 8.7 ꢁ 10ꢀ4 M 1 with TBAF in DMSO. The
tions. As a first effort to develop a more selective fluoride
receptor based on the uranyl-salophen unit, we considered the
tert-butylated derivative, 2, which easily prepared by the
condensation of 3,5-di-tert-butyl-salicylaldehyde with 1,2-di-
aminobenzene in the presence of UO2(AcO)2ꢃ2H2O in metha-
nol. The introduction of bulky tert-butyl groups was meant to
promote steric control over an anion’s approach to the
equatorial binding site. Table 1 shows that the tert-butyl
groups flanking the uranyl center did indeed decrease the
inset shows 1 : 1 complex formation.
bromide or iodide anions. This was also the case with weakly
basic oxyanions, such as ClO4ꢀ and SO42ꢀ. Although, because
of the sigmoid-shaped titration curve (Fig. 1), the affinity of
complex 1 towards fluoride anions could not be measured on
the basis of the absorption titration, the above observations
clearly highlight the high selectivity of 1 for fluoride anions
over other halides and weakly basic anions. However, the
ꢀ
affinity towards H2PO4 by 50-fold, but were not bulky
situation with more basic anions, such as AcOꢀ and H2PO4
,
enough to hinder the complexation of the flat AcOꢀ anion.7
Consequently, on the basis of molecular models, we envisaged
cyclophane 3, whose small cavity seemed to host selectively the
tiny fluoride ion, with the exclusion of the bulkier anions. At
first, we actually considered the analogous macrocycle struc-
ture based on the salophen moiety. However, any attempt at
reacting precursor 5 (Scheme 1) with 1,2-diaminobenzene in
the presence of UO2(AcO)2 failed to yield the desired macro-
cycle. From this evidence, we reasoned that the more flexible
salen moiety could be more easily incorporated into a strained
macrocyclic structure than the rigid salophen one. The syn-
thesis of target receptor 3 was accomplished by a two-step
procedure, starting from readily available materials
(Scheme 1). The X-ray crystal structure of its methanol solvate
is shown in Fig. 4.z The cavity where the UO22+-bound
methanol is hosted has a 5.3 ꢁ 4.5 A rectangular shape, which
can obviously accommodate a fluoride anion, but it does not
appear to be spatially suitable for linear coordination in the
equatorial plane of even an anion as small as CNꢀ.
ꢀ
was different. Fig. 3 shows the spectroscopic changes experi-
enced by 1 upon titration with tetramethylammonium acetate
(TMAAcO). Unlike in Fig. 1, the titration plot in Fig. 3 has
the expected shape for a 1 : 1 binding isotherm. Similar
behaviors were observed for titrations with TBA H2PO4.
The results of the titration experiments are listed in the first
column of Table 1, including an estimate of the binding affinity
towards fluoride ion based on competition experiments, in
which 1 was titrated with TBAF in the presence of a large
excess of TMAAcO (Fig. S4w).
While the parent uranyl-salophen complex appears to bind
fluoride ion significantly more strongly than the other basic
anions, with a selectivity that compares well with, or even
exceeds, those experienced by more structured receptors,1 even
higher selectivities are considered desirable for many applica-
The receptor properties of 3 were in line with our expecta-
tions. No spectral variations were induced by the addition of
ꢀ
up to several hundred equivalents of the TBA salts of H2PO4
,
CNꢀ or TMAAcO to solutions of 3 in DMSO (K o 10 Mꢀ1).
Titration with TBAF instead caused spectral variations that
were very similar to those experienced by 1 and 2 (compare
Fig. S5w with Fig. 1), showing that the affinity of 3 towards
fluoride anions is comparable to those of the salophen
Fig. 3 UV-vis spectral changes of 3.9 ꢁ 10ꢀ5 M 1 (DMSO, 25 1C)
upon addition of TMAAcO. The inset shows the fit of the experi-
mental data to a 1 : 1 binding isotherm (415 nm).
Scheme 1
ꢂc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2008
1114 | New J. Chem., 2008, 32, 1113–1116

View Article Online
Compound 5
To a suspension of n-pentane-washed NaH (1.1 g, 60% in oil)
in DMSO (10 mL) was added a solution of 2,3-dihydroxyben-
zaldehyde (1.5 g, 11.3 mmol) in DMSO (10 mL) at 20–25 1C.
After stirring for 1 h, a solution of 4 (2 g, 5.63 mmol) in
DMSO (10 mL) was added at 20–25 1C. Stirring was con-
tinued for 24 h, whereupon the mixture was poured into water
(50 mL) and extracted with CHCl3 (2 ꢁ 100 mL). The aqueous
layer was acidified with 6 M HCl to adjust the pH to 3, and it
was again extracted with CHCl3 (4 ꢁ 80 mL). The combined
CHCl3 layers were washed with 1 M HCl (2 ꢁ 50 mL) and
then dried over MgSO4. The solvent was evaporated and the
residue then purified by column chromatography (silica gel,
CHCl3) to give pure dialdehyde 5 as a yellow solid (yield 60%).
Elemental analysis: calc. for C29H24O6: C, 74.35; H, 5.16;
found: C, 74.20; H, 4.98%. Mass spectrum (ESI): m/z 469.35
Fig. 4 X-Ray crystal structure of 3ꢃCH3OH. Left: ORTEP plot (50%
probability level). Right: CPK model. The UO22+-bound methanol is
omitted for clarity, and only one of the uranyl-salen–methanol com-
plexes in the asymmetric unit is shown.
receptors.8 It appears therefore that the selectivity of 3 to-
wards fluoride anions relative to a large variety of anion
competitors is no less than 105 higher, which is, to the best
of our knowledge, among the highest ever recorded.
To sum up, we have shown that a remarkable specificity for
fluoride anions in DMSO solution can be achieved by the
rational design of a uranyl-salen receptor, in which the space
surrounding the metal centre is delimited by a short bridge
featuring a diphenylmethane unit.
1
[M + H]+ (calc. for C29H24O6 469.16). H NMR (CDCl3): d
11.04 (s, 2H), 9.91 (s, 2H), 7.36 (d, J = 8 Hz, 4H), 7.15 (m,
8H), 6.89 (t, J = 8 Hz, 2H), 5.15 (s, 4H) and 3.98 (s, 2H). 13
C
NMR (CDCl3): d 196.6, 152.6, 147.6, 138.1, 133.7, 129.5,
127.7, 125.4, 121.2, 120.5 119.3, 71.6 and 41.2.
Receptor 3
To a refluxing solution of 5 (0.4 g, 0.85 mmol) in methanol (40
mL) was added ethylenediamine (0.51 g, 0.60 mL, 0.85 mmol)
dropwise. After 1.5 h, UO2(AcO)2ꢃ2H2O (0.36 g, 0.85 mmol)
was added and reflux continued for a further 30 min, where-
upon the mixture was allowed to cool to room temperature
overnight. The red solid formed was filtered and crystallized
from methanol/DMSO (99 : 1) to afford 3 (yield 19%).
Elemental analysis: calc. for C31H26N2O6ꢃCH3OH: C, 48.49;
H, 3.81; N, 3.53; found: C, 49.10; H, 3.78; N, 3.34%. Mass
spectrum (ESI): m/z 783.46 (calc. for C31H26N2O6UNa
[M + Na]+ 783.57) and 815.25 (calc. for C32H30N2O7UNa
[M + CH3OH + Na]+ 815.61). 1H NMR (DMSO-d6): d 9.40
(s, 2H), 7.39 (m, 6H), 7.20 (d, J = 7.9 Hz, 2H), 7.01 (d, J =
6.6 Hz, 4H), 6.60 (t, J = 7.9 Hz, 2H), 5.19 (s, 4H), 4.44 (s, 4H)
and 3.87 (s, 2H). 13C NMR (DMSO-d6): d 168.5, 160.7,
151.9, 140.9, 136.2, 133.8, 128.8, 128.4, 126.4, 124.2, 121.4,
119.7, 115.2, 72.8 and 63.1.
The support and sponsorship provided by COST Action
D31/0016 is acknowledged. The Academy of Finland is grate-
fully acknowledged by M. N. and K. R. (proj. no. 122350) for
financial support.
Experimental
NMR spectra were recorded on either an AC 200 or AC 300
Bruker spectrometer. UV-vis spectra were recorded in the
thermostated cell compartment of a Cary 300 spectrophot-
ometer. Association constants were determined according to
standard UV-vis titration procedures.9 Receptor
1 was
available from previous work.5d Bis(4-bromomethylphenyl)-
methane (4) was prepared according to a literature proce-
dure.10
Receptor 2
X-Ray crystal structure determination
ortho-Phenylenediamine (0.115 g, 1.06 mmol) was added to a
refluxing solution of 3,5-di-tert-butyl-2-hydroxybenzaldehyde
(0.5 g, 2.13 mmol) in methanol (50 mL) with stirring.
UO2(AcO)2ꢃ2H2O (0.45 g, 1.06 mmol) was added after 1 h,
whereupon the solution was cooled to room temperature and
stirred overnight. The red precipitate of 2 that formed was
filtered-off and crystallized from hexane to give red needles
(0.53 g, yield 62%). Elemental analysis: calc. for
C36H46N2O4U: C, 53.46; H, 5.73; N, 3.46; found: C, 53.66;
H, 5.53; N, 3.53%. Mass spectrum (ESI): m/z 831.45
[M + Na]+ (calc. for C36H46N2O4UNa 831.78). 1H NMR
(DMSO-d6): d 9.63 (s, 2H), 7.71 (m, 2H), 7.66 (s, 4H), 7.48 (m,
2H), 1.88 (s, 9H) and 1.47 (s, 9H). 13C NMR (DMSO-d6): d
146.8, 138.9, 137.4, 129.9, 128.1, 123.9, 119.9, 35.0, 33.4, 31.3
and 30.0.
X-Ray data for 3 crystallized from methanol were collected
from an orange needle-like crystal of size 0.05 ꢁ 0.3 ꢁ 0.5 mm
on a Nonius Kappa CCD diffractometer using graphite-
monochromatized Mo-Ka radiation and a temperature of
173.0 K.
Crystal data for 3
4C31H26N2O6Uꢃ5.5CH3OH, M = 3218.50 g molꢀ1, triclinic,
P1 (no. 1), a = 11.3081 (2), b = 13.0762 (3), c = 21.1935 (4)
A, a = 78.420 (1), b = 84.284 (1), g = 88.925(1)1, V =
3054.8(1) A3, Z = 1, m = 5.365 mmꢀ1, 20 665 reflections
collected of which 15 597 were unique (Rint = 0.057), final
R1 = 0.044 and wR2 = 0.112 for I 4 2s(I). Flack parameter
0.011(7).z
ꢂc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2008 New J. Chem., 2008, 32, 1113–1116 | 1115

View Article Online
K. Perdikaki and G. Reeske, Eur. J. Inorg. Chem., 2004, 2283–2288;
(e) A. C. Ion, I. Ion, M. M. G. Antonisse, B. H. M. Snelink-Ruel and
D. N. Reinhoudt, Russ. J. Gen. Chem., 2001, 71, 159–161.
References
1. For recent papers, see: (a) I. V. Korendovych, M. Cho,
P. L. Butler, R. J. Staples and E. V. Rybak-Akimova, Org. Lett.,
2006, 8, 3171–3174; (b) R. Nishiyabu and P. Anzenbacher, Jr, Org.
Lett., 2006, 8, 359–362; (c) Z. Lin, S. Ou, C. Duan, B. Zhang and
Z. Bai, Chem. Commun., 2006, 624–626; (d) Z. Lin, Y. Zhao,
C. Duan, B. Zhang and Z. Bai, Dalton Trans., 2006, 3678–3684;
(e) E. R. Libra and M. J. Scott, Chem. Commun., 2006, 1485–1487;
5. (a) M. M. G. Antonisse and D. N. Reinhoudt, Chem. Commun.,
1998, 443–448; (b) M. Cametti, M. Nissinen, A. Dalla Cort,
L. Mandolini and K. Rissanen, J. Am. Chem. Soc., 2005, 127,
3831–3837; (c) M. Cametti, M. Nissinen, A. Dalla Cort, K. Rissanen
and L. Mandolini, Inorg. Chem., 2006, 45, 6099–6101;
(d) M. Cametti, M. Nissinen, A. Dalla Cort, L. Mandolini and
K. Rissanen, J. Am. Chem. Soc., 2007, 129, 3641–3648.
(f) T. W. Hudnall, M. Melaımi and F. P. Gabbaı, Org. Lett., 2006,
¨
¨
8, 2747–2749; (g) G. W. Bates, P. A. Gale and M. E. Light, Chem.
Commun., 2007, 2121–2123; (h) F. Han, Y. Bao, Z. Yang,
T. M. Fyles, J. Zhao, X. Peng, J. Fan, Y. Wu and S. Sun,
Chem.–Eur. J., 2007, 13, 2880–2892; (i) C.-I. Lin, S. Selvi,
J.-M. Fang, P.-T. Chou, C.-H. Lai and Y.-M. Cheng, J. Org.
Chem., 2007, 72, 3537–3542.
6. We speculate that the ‘‘disturbance’’ occurring in the low con-
centration domain is caused by an impurity in the DMSO solvent,
which sequesters a significant fraction of the added fluoride. The
observed phenomena are qualitatively explained if the above
species is spectroscopically silent, is present at a concentration in
the order of 10ꢀ5 M and has an affinity toward fluoride anions
comparable to that of 1. Such a ‘‘phantom’’ sequestering agent
might well be an acidic impurity causing the formation of (HF2)ꢀ,
but not adventitious water (see ref. 2a) because the influence of
water on the stability of 1ꢃFꢀ becomes significant only at very high
water concentrations (Fig. S1w).
2. For review articles, see: (a) J. L. Sessler, S. Camiolo and P. A. Gale,
Coord. Chem. Rev., 2003, 240, 17–55; (b) C. R. Bondy and
S. J. Loeb, Coord. Chem. Rev., 2003, 240, 77–99; (c) P. A. Gale,
Acc. Chem. Res., 2006, 39, 465–475; (d) V. Amendola,
M. Bonizzoni, D. Esteban-Gomez, L. Fabbrizzi, M. Licchelli,
F. Sancenon and A. Taglietti, Coord. Chem. Rev., 2006, 250,
1451–1470; (e) V. Amendola, D. Esteban-Gomez, L. Fabbrizzi
and M. Licchelli, Acc. Chem. Res., 2006, 39, 343–353 and refer-
ences therein. See also ref. 1 for selected recent papers.
7. We believe that the slight stability increase of the AcOꢀ complex
on going from 1 to 2 is a real phenomenon. It might be a
consequence of the operation of weak attractive van der Waals
interactions between the methyl group of the bound acetate and
the flanking tert-butyl groups.
3. (a) M. Melaımi, S. Sole, C.-W. Chiu, H. Wang and F. P. Gabbaı,
¨
Inorg. Chem., 2006, 45, 8136–8143; (b) C. W. Chiu and F. Gabbaı,
¨
8. Titration plots of dilute solutions (4–5 ꢁ 10ꢀ5 M) of receptors 1, 2
and 3 show that saturation is reached when the concentration of
added fluoride anion is ca. 1 ꢁ 10ꢀ4 M in all cases. As for
receptors 1 and 2, the conclusion that similar titration profiles
imply similar affinities toward fluoride anions is fully corroborated
by their virtually identical log K values in Table 1. Thus, if the
affinity of 2 is very similar to that of 1, then, by inference, so is that
of 3.
¨
J. Am. Chem. Soc., 2006, 128, 14248–14249; (c) K. Parab,
K. Venkatasubbaiah and F. Jakle, J. Am. Chem. Soc., 2006,
¨
128, 12879–12885; (d) M. H. Lee, T. Agou, J. Kobayashi,
T. Kawashima and F. P. Gabbaı, Chem. Commun., 2007,
¨
¨
1133–1135; (e) T. W. Hudnall and F. P. Gabbaı, J. Am. Chem.
Soc., 2007, 129, 11978–11986; (f) G. Kwak, M. Fujiki and
T. Masuda, Macromolecules, 2004, 37, 2422–2426.
4. (a) M.-L. Lehaire, R. Scopelliti, H. Piotrowski and K. Severin, Angew.
9. V. van Axel Castelli, A. Dalla Cort, L. Mandolini, V. Pinto,
D. N. Reinhoudt, F. Ribaudo, C. Sanna, L. Schiaffino and B. H.
Chem., Int. Ed., 2002, 41, 1419–1422; (b) M. Melaımi and
¨
¨
F. P. Gabbaı, J. Am. Chem. Soc., 2005, 127, 9680–9681; (c) I. H.
M. Snellinkl-Ruel, Supramol. Chem., 2002, 14, 211–219.
¨
3205–3211.
¨
10. H.-F. Grutzmacher and W. Husemann, Tetrahedron, 1987, 43,
A. Badr and M. E. Meyerhoff, J. Am. Chem. Soc., 2005, 127,
5318–5319; (d) N. Chaniotakis, K. Jurkschat, D. Mueller,
ꢂc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2008
1116 | New J. Chem., 2008, 32, 1113–1116