Metal-controlled assembly and selectivity of a urea-based anion receptor

Article abstract of DOI:10.1021/ic060160x

1-(3-[1,10]phenanthrolin-2-yl-phenyl)-3-(4-trifluoromethyl-phenyl)-urea (1) forms a solution-stable 1:2 complex with copper(I), [Cu^I(1) ₂]⁺, which behaves as an anion receptor in aprotic media. The [Cu^I(1)₂]⁺ receptor may adopt a geometrical arrangement in which the two facing urea subunits give tetrafurcate H-bond interaction with a spherical anion (halides and H₂PO₄ ^-). In the presence of the Y-shaped acetate, the [Cu ^I(1)₂]⁺ receptor rearranges to interact with two CH₃COO^- ions, according to two stepwise equilibria. Thus, in the presence of substoichiometric amounts of the anions, [Cu ^I(1)₂]⁺ discriminates H₂PO ₄^- over CH₃COO^-, inverting the basicity trend.

Full text of DOI:10.1021/ic060160x

Inorg. Chem. 2006, 45, 6138−6147
Metal-Controlled Assembly and Selectivity of a Urea-Based Anion
Receptor
Valeria Amendola,† Massimo Boiocchi,‡ Benoˆıt Colasson,§ and Luigi Fabbrizzi*,†
Contribution from Dipartimento di Chimica Generale and Sezione INSTM, Viale Taramelli 12,
UniVersita` di PaVia, 27100 PaVia, and Centro Grandi Strumenti, Via Bassi 10, 27100 PaVia, Italy
Received January 28, 2006
1-(3-[1,10]phenanthrolin-2-yl-phenyl)-3-(4-trifluoromethyl-phenyl)-urea (1) forms a solution-stable 1:2 complex with
copper(I), [CuI(1)2]+, which behaves as an anion receptor in aprotic media. The [CuI(1)2]+ receptor may adopt a
geometrical arrangement in which the two facing urea subunits give tetrafurcate H-bond interaction with a spherical
anion (halides and H2PO4-). In the presence of the Y-shaped acetate, the [CuI(1)2]+ receptor rearranges to interact
with two CH3COO- ions, according to two stepwise equilibria. Thus, in the presence of substoichiometric amounts
-
of the anions, [CuI(1)2]+ discriminates H2PO4 over CH3COO-, inverting the basicity trend.
Introduction
stable H-bonded complex [LH‚‚‚F]-, which, on addition of
a second equivalent of F-, may release a HF molecule to
give the deprotonated form of the receptor, L-, and the self
complex [HF2]-.5 H-bond donor tendencies of receptors on
the basis of a single urea subunit can be enhanced through
the covalent linking to electron-withdrawing substituents
(e.g., nitrophenyl, cyanophenyl, trifluoromethylphenyl), but
the increased acidity does not alter the affinity pattern, which
is ultimately determined by the basicity of the anion. Thus,
selectivity can be achieved only by positioning two or more
urea subunits on a rigid platform that has been designed by
taking into account the geometrical binding preferences of
the anion (e.g., a calix[6]arene,6 a 1,3-xylyl fragment,7 a 1,2-
cyclohexane subunit).8
According to a different approach, metals can be used to
properly assemble H-bond donor subunits (e.g., urea deriva-
tives) and to generate binding sites suitable for selective anion
binding.9 In particular, the H-bond donor subunit must be
covalently linked to a ligand for metal ions. Then, when two
or more molecules of the functionalized ligand react with a
chosen metal, a coordination complex forms, which may
properly orientate the H-bond donor subunits and offer a
convenient binding site for anions. The shape and size of
Since the pioneering works of Wilcox1 and Hamilton,2 urea
has become a widely used fragment for the design of neutral
receptors and sensors for anions.3 The single urea subunit
typically behaves as a donor of two H-bonds and establishes
complementary interactions with two oxygen atoms of a
given oxoanion, e.g., the Y-shaped acetate.4 However, the
-
normally observed selectivity pattern (CH3COO- > H2PO4
-
-
-
-
> NO2 > HSO4 > NO3 > ClO4 ) does not rely on
structural features (e.g., the more or less favorable matching
of the urea NH fragments and anion oxygen atoms) but
simply reflects the decreasing basicity of the anion, ex-
pressed, for instance, by the value of the partial negative
charge on its oxygen atoms.4 On the other hand, urea
establishes with spherical halide ions bifurcate hydrogen
bonding interactions whose intensity decreases with the
charge density of the anion (Cl- > Br- > I-). Fluoride
exhibits a unique behavior; it establishes a monofurcate
interaction with one NH fragment of urea, giving the rather
* To whom correspondence should be addressed. E-mail:
luigi.fabbrizzi@unipv.it.
† Dipartimento di Chimica Generale, Universita` di Pavia.
‡ Centro Grandi Strumenti, Universita` di Pavia.
§ Sezione INSTM, Universita` di Pavia.
(1) Smith, P. J.; Reddington, M. V.; Wilcox, C. S. Tetrahedron Lett. 1992,
41, 6085.
(2) Fan, E.; van Arman, S. A.; Kincaid, S.; Hamilton, A. D. J. Am. Chem.
Soc. 1993, 115, 369.
(5) Boiocchi, M.; Del Boca, L.; Esteban-Go´mez, D.; Fabbrizzi, L.;
Licchelli, M.; Monzani, E. Chem.sEur. J. 2005, 11, 3097.
(6) Scheerder, J.; Engbersen, J. F. J.; Casnati, A.; Ungaro, R.; Reinhoudt,
D. N. J. Org. Chem. 1995, 60, 6448.
(3) Gale, P. A. Amide- and Urea-Based Anion Receptors. In Encyclopedia
of Supramolecular Chemistry; Marcel Dekker: New York, 2004, pp
31-41.
(4) Boiocchi, M.; Del Boca, L.; Esteban-Go´mez, D.; Fabbrizzi, L.;
Licchelli, M.; Monzani, E. J. Am. Chem. Soc. 2004, 126, 16507.
(7) Nishizawa, S.; Bu¨hlmann, P.; Iwao, M.; Umezawa, Y. Tetrahedron
Lett. 1995, 36, 6483.
(8) Amendola, V.; Boiocchi, M.; Esteban-Go´mez, D.; Fabbrizzi, L.;
Monzani, E. Org. Biomol. Chem. 2005, 3, 2632.
(9) Beer, P. D.; Hayes, E. J. Coord. Chem. ReV. 2003, 240, 167.
6138 Inorganic Chemistry, Vol. 45, No. 16, 2006
10.1021/ic060160x CCC: $33.50
© 2006 American Chemical Society
Published on Web 04/19/2006

Metal-Controlled Urea-Based Anion Receptor
the binding site are determined by the coordinative prefer-
ences of the metal (toward a tetrahedral, square, or octahedral
geometry) and by the nature and length of the spacer
connecting the ligand to the H-bond donor subunit. For
instance, a PtII center coordinated four urea functionalized
isoquinolines at the corners of a square, giving a substitu-
tionally inert complex; the four urea subunits formed a
appended to the other phenyl ring to enhance the H-bond
donor tendencies of the urea moiety and induce an optical
charge-transfer transition, which is useful for reporting
purposes. Binding studies have been carried out in a DMSO
solution and in a 4:1 v/v THF:MeCN mixture, and pertinent
association constants have been determined through the
nonlinear least-squares analysis of data from UV-vis spec-
trophotometric titration experiments. The crystal structure
of the CuI(1)2Cl salt is reported.
-
H-bond donor cavity suitable for interaction with H2PO4
-
and HSO4 in a DMSO solution.10 Because of the extreme
kinetic stability of the d8 (low-spin) metal center, the [PtII-
(isoquinoline)4]2+ moiety provides a rigid platform that has
been compared to the purely organic scaffold of calix[4]-
arenes. Other inert metal centers (e.g. RuII, d8 low-spin, which
imposes an octahedral coordination geometry)11 have been
utilized to build molecular cavities that are capable of
accommodating anions through interaction with H-bond
donors appended to the ligand(s). On the other hand,
substitutionally labile d10 metal centers allow for reversible
assembling of functionalized ligands, a feature which may
impart a higher versatility in anion recognition. A first report
refers to the 1:2 tetrahedral CuI complex of a 1,10-
phenanthroline ligand linked to an acylaminopyridine moiety
(H-bond donor through the amide NH fragment and H-bond
acceptor through the pyridine nitrogen atom) that has been
shown to form H-bond complexes with dicarboxylic acids
in chloroform.12 More recently, it has been shown that a
[AgIL2]+ complex, in which L is a pyridine covalently linked
to a urea subunit, forms a stable adduct with the nitrate ion,
which is H-bonded to both urea fragments. The [AgIL2‚‚‚
NO3] ternary complex has been characterized both in the
solid state13 and in acetone solution.14 In particular, the
coordination to the metal induced a 30-fold enhancement of
Experimental Section
Materials and Methods. All reagents for syntheses were
purchased from Aldrich/Fluka and used without further purification.
2-Chloro-1,10-phenanthroline (2) was prepared in three steps from
commercial 1,10-phenanthroline according to the literature proce-
dure.15 UV-vis spectra were recorded on a Varian CARY 100
spectrophotometer with a quartz cuvette (path length 0.1-1.0 cm).
In the titrations with anions, UV-vis spectra of the samples were
recorded after the addition of aliquots of an alkylammonium salt
solution of the envisaged anion ([Bu4N]+ for CH3COO-, H2PO4
NO2-, HSO4-, F-, Br-, and I-; [BzMe3N]+ for Cl-). All spectro-
photometric titration curves were processed with the HyperQuad
program.16 1H NMR spectra were obtained on a Bruker AVANCE
400 spectrometer (400 MHz) operating at 9.37 T.
-
,
Syntheses. 2-(3-Nitro-phenyl)-1,10-phenanthroline (3). 2-chloro-
1,10-phenanthroline (2) (362 mg, 1.69 mmol) was dissolved in
toluene (10 mL). While the solution was degassed with bubbling
nitrogen, 3-nitrophenylboronic acid (295 mg, 1.77 mmol) was
added, followed by Pd(PPh3)4 (97 mg, 0.084 mmol). Finally, 10
mL of a degassed aqueous solution of Na2CO3 (2 M) was added.
The solution was then heated at reflux under nitrogen for 15 h. A
spot-to-spot conversion is observed on TLC (Al2O3, CH2Cl2/MeOH
(1%)). The solution was cooled, and 30 mL of CH2Cl2 and 30 mL
of water were added. After the first separation of the two phases,
the aqueous phase was extracted with 2 × 50 mL of CH2Cl2. The
organic phase was filtered through a plug of Celite, dried with Na2-
SO4. After filtration, the solvent was removed to yield 530 mg of
crude product. This crude material was purified by chromatography
over alumina, eluting with CH2Cl2/MeOH (2%) to yield pure 2 (500
-
the affinity of the urea-based receptor toward NO3 .
1
mg, 97% yield). H NMR (400 MHz, CDCl3): 9.27 (d, 1H), 9.05
(s, 1H), 8.81 (d, 1H), 8.47 (d, 1H), 8.30 (d, 1H), 8.27 (d, 1H), 8.20
(d, 1H), 7.83 (s, 2H), 7.73 (t, 1H), 7.68 (t, 1H). ES-MS m/z: 302.1
[M + H]+, calcd 302.1. Anal. Calcd for C18H11N3O2: C, 71.75; H,
3.68; N, 13.95. Found: C, 71.89; H, 3.67; N, 14.25.
We report here on the anion binding tendencies of the 1:2
CuI complex with the urea-functionalized 1,10-phenanthro-
line 1, 1-(3-[1,10]phenanthrolin-2-yl-phenyl)-3-(4-trifluoro-
methyl-phenyl)-urea. The linking positions in the phenylurea
fragment and in the phenanthroline moiety have been chosen
in order to favor the formation of a not-too-large anion
binding site following metal complex assembly, as suggested
by modeling studies. A trifluoromethylphenyl group has been
3-[1,10]Phenanthrolin-2-yl-phenylamine (4). 2-(3-Nitro-phen-
yl)-1,10-phenanthroline (3) (322 mg, 1.07 mmol) was dissolved in
30 mL of ethanol. SnCl2‚2H2O (1.6 g, 7.1 mmol) was added, and
the resulting mixture was heated at reflux for 3 h under nitrogen.
The reaction was monitored by TLC (Al2O3, CH2Cl2/MeOH (2%)),
and a clean spot-to-spot conversion was observed. Ethanol was
removed under vacuum. The solid was taken up in CH2Cl2/H2O
(pH was adjusted to 12 with NaOH). The aqueous phase was
extracted twice with CH2Cl2 (30 mL). The combined organic phase
was washed twice with basic water (pH 12) and once with distilled
water. After being dried over Na2SO4 and filtered, the solvent was
removed to yield 3-[1,10]phenanthrolin-2-yl-phenylamine (7). This
isolated yellow oil turned rapidly into a black oil, indicating that
decomposition was taking place. Nevertheless, a good mass of a
(10) Bondy, C. R.; Gale, P. A.; Loeb, S. J. J. Am. Chem. Soc. 2004, 126,
5030.
(11) Wallace, K. J.; Daari, R.; Belcher, W. J.; Abouderbala, L. O.; Boutelle,
M. G.; Steed, J. W. J. Organomet. Chem. 2003, 666, 63.
(12) Goodman, M. S.; Jubian, V.; Linton, B.; Hamilton, A. D. J. Am. Chem.
Soc. 1995, 117, 11610.
(13) Turner, D. E.; Spencer, E. C.; Howard, J. A. K.; Tocher, D. A.; Steed,
J. W. Chem. Commun. 2004, 1352.
(14) Turner, D. R.; Smith, B.; Spencer, E. C.; Goeta, A. E.; Radosavljevic-
Evans, I.; Derek, A.; Tocher, D. A.; Howard, J. A. K.; Steed, J. W.
New J. Chem. 2005, 29, 90.
(15) Lewis, J.; O’Donoghue, T. D. J. Chem. Soc., Dalton Trans. 1980,
736.
(16) Gans, P.; Sabatini, A.; Vacca, A. Talanta 1996, 43, 1739.
Inorganic Chemistry, Vol. 45, No. 16, 2006 6139

Amendola et al.
Scheme 1
Å), 2θmax ) 46°, 46 463 measured reflections, 13 t778 independent
reflections (Rint ) 0.049), 11 693 strong reflections [Io > 2σ(Io)],
1262 refined parameters, R1 ) 0.0928 (strong data) and 0.1078
(all data), wR2 ) 0.2240 (strong data) and 0.2341 (all data), GOF
) 1.150, 0.87 and -0.38 maximum and minimum residual electron
density.
Single crystals suitable for X-ray diffraction study were always
characterized by diffracted intensities corresponding to the P21 space
group. In particular, the systematic extinctions due to the c glide
plane were not obeyed. Several attempts to solve the structure by
direct methods (SIR-97)18 in the P21 space group produced a first
partial atomic model, which exhibited the symmetry of the P21/c
space group. The presence of the forbidden reflections in the
diffraction pattern was thus ascribed to merohedral twinning.
Refinement by full-matrix least-squares procedures on F2 using all
reflections (SHELXL-97)19 was performed assuming a second
component in the crystal, rotated by 180° about [100]. The
agreement index strongly improved, and the atomic model was
completed by analyzing the ∆F map. The refined twin ratio result
was 0.52(1):0.48(1). All non-hydrogen atoms were refined with
anisotropic temperature factors; hydrogen atoms were placed using
the appropriate AFIX instructions. CCDC 289232 contains the
supplementary crystallographic data for this paper.
fraction that was kept in solution could be obtained (ES-MS m/z:
272.4 [M + H]+, calcd 272.3). TLC of this sample indicated that
no decomposition was observed when 4 was kept in solution. For
the following step (i.e., the formation of the urea-phenanthroline
conjugate (1)), after filtration, the solution of CH2Cl2 containing 4
was concentrated under vacuum and used immediately.
1-(3-[1,10]Phenanthrolin-2-yl-phenyl)-3-(4-(trifluoromethyl-
phenyl)-urea (1). To a CH2Cl2 solution of 3-[1,10]phenanthrolin-
2-yl-phenylamine (1.07 mmol on the basis of a 100% conversion
of 3) was added 1.1 equiv of R,R,R-trifluoro-p-tolylisocyanate (165
µL). The solution was stirred at room temperature for 2 h. The
white precipitate that appeared in the course of the reaction was
filtered and washed with CH2Cl2. The urea-phenanthroline con-
jugate 1 was obtained as a white solid (305 mg, 62% yield). For a
visual depiction of the total synthesis, see Scheme 1
Results and Discussion
1. Design of the Receptor. The CuI ion gives with 1,10-
phenanthroline a tetrahedral complex [CuI(phen)2]+. How-
ever, the coordination geometry is not exactly tetrahedral
(N-Cu-N bond angles differ from the canonic value of
109.5°) because of steric constraints imposed by the rigid
phenanthroline framework.20 Steric effects exerted by small
substituents on the 2,9 positions (e.g., methyl21 and triflu-
oromethyl)22 force the phenanthrolines to lie perpendicular
each other, minimizing the interligand repulsions. However,
aromatic substituents on the heterocyclic rings may induce
variable distortions; for instance, in the symmetric 2,9-
diphenyl-1,10-phenanthroline complex of CuI, the angle
between phen planes is ca. 59°,23 whereas in the monosub-
stituted 2-dimethylaminophenyl-phenanthroline, the angle is
ca. 68°.24
Notice that in ligand 1, the linking positions in the
phenylurea fragment and in the heterocyclic ring have been
judiciously selected, generating a rather small and strictly
defined binding site that is suitable for interaction with a
single anion. In this regard, modeling studies where carried
out using a semiempirical method (PM3) for the H-bond
complex [CuI(1)2‚‚‚Cl]+. The pertinent calculated structure
is shown in Figure 1.
1
. H NMR (400 MHz, d6-DMSO): 9.17 (d, 3H), 8.60 (d, 1H),
8.53 (dd, 1H), 8.45 (s, 1H), 8.35 (d, 1H), 8.05 (m, 3H), 7.80 (m,
2H), 7.52 (d, 1H), 7.48 (d, 1H), 7.55 (t, 1H). ES-MS m/z: 459.1
[M + H]+, calcd 459.1. Anal. Calcd for C26H17N4F3O: C, 68.12;
H, 3.74; N, 12.22. Found: C, 68.66; H, 3.77; N, 13.06.
[Cu(1)2](ClO4). 1 (24.1 mg, 2 equiv) was dissolved in a CH2-
Cl2/MeOH mixture (10:3 mL). The solution was degassed, and [Cu-
(CH3CN)4] (ClO4) (8.6 mg, 1 equiv) in a degassed solution of
CH3CN was added. The solution immediately turned red. It was
allowed to stir at room temperature for 1 h. The solvent was then
removed under vacuum to quantitatively yield [Cu(1)2](ClO4) (29
mg) as a red solid. ES-MS m/z: 979.2 [M - ClO4]+, calcd 979.3.
Once this complex was isolated, it was very difficult to dissolve it
in any solvent except DMSO. For the titration in the THF/CH3CN
mixture, the complex was prepared in situ. Caution: perchlorate
salts are potentially explosiVe and must be handled with care. In
particular, they should neVer be heated as solids.
X-ray Crystallographic Studies. Diffraction data of the [CuI-
(1)2]Cl complex salt were collected at ambient temperature on a
Bruker-AXS Smart-Apex CCD-based diffractometer. Omega rota-
tion frames (scan width 0.3°, scan time 50 s, sample-to-detector
distance 7 cm) were processed with the SAINT software (Bruker-
AXS Inc.), and intensities were corrected for Lorentz and polariza-
tion effects. Absorption effects were analytically evaluated by the
SADABS software,17 and correction was applied to the data (0.72
and 0.98 minimum and maximum transmission factors). Crystal
data for [Cu(I)L]Cl complex: C52H34ClCuF6N8O2, Mr ) 1015.87,
T ) 293 K, crystal dimensions 0.13 × 0.08 × 0.04 mm3,
monoclinic, P21/c (No. 14), a ) 15.412(5), b ) 25.957(5), c )
25.514(3) Å, â ) 107.58(1), V ) 9730(4) Å3, Z ) 8, Fcalcd ) 1.387,
F(000) ) 4144, µ ) 0.576 mm-1, Mo KR X-radiation (λ ) 0.7107
(18) Altomare, A.; Burla, M. C.; Camalli, M.; Cascarano, G. L.; Giaco-
vazzo, C.; Guagliardi, A.; Moliterni, A. G. G.; Polidori, G.; Spagna,
R. J. Appl. Crystallogr. 1999, 32, 115.
(19) Sheldrick, G. M. SHELX97: Programs for Crystal Structure Analysis;
University of Go¨ttingen: Go¨ttingen, Germany, 1997.
(20) Clarke, R.; Latham, K.; Rix, C.; White, J. Acta Crystallogr., Sect. C
2003, 59, m7-m9.
(21) Kovalevsky, A. Y.; Gembicky, M.; Novozhilova, I. V.; Coppens, P.
Inorg. Chem. 2003, 42, 8794.
(22) Kovalevsky, A. Y.; Gembicky, M.; Coppens, P. Inorg. Chem. 2004,
43, 8282.
(23) (a) Geoffroy, M.; Wermeille, M.; Buchecker, C. O.; Sauvage, J.-P.;
Bernardinelli, G. Inorg. Chim. Acta 1990, 167, 157. (b) Klemens, F.
K.; Palmer, C. E. A.; Rolland, S. M.; Fanwick, P. E.; McMillin, D.
R.; Sauvage, J.-P. New J. Chem. 1990, 14, 129.
(17) Sheldrick, G. M. SADABS: Siemens Area Detector Absorption
Correction Program; University of Go¨ttingen: Go¨ttingen, Germany,
1996.
6140 Inorganic Chemistry, Vol. 45, No. 16, 2006

Metal-Controlled Urea-Based Anion Receptor
Figure 1. Hypothesized structure of the [CuI(1)2‚‚‚Cl]+ complex, which
has been modeled through a semiempirical method (PM3).
In the model, the chloride ion receives four H-bond from
the two facing urea subunits, according to a rather distorted
tetrahedral geometry (the angle between the two NH-Cu-
NH planes is 60.8°). On the other hand, the CuI center
experiences a regular tetrahedral coordination; in particular,
the angle between the phenanthroline planes is 89.2°. In any
case, the model suggests that the [CuI(1)2]+ complex may
act as a receptor for a single spherical anion.
Figure 2. Ortep view of the [CuI(1)2]Cl complex salt (names are only
show for Cu, N. and Cl atoms; thermal ellipsoids are drawn at the 30%
probability level; only hydrogen atoms bonded to the Nurea atoms are
reported). The crystal contains two nonsymmetrically equivalent [CuI(1)2]+
molecules and their counterions; each Cl- ion interacts with four NHCONH
(urea) groups as a proton acceptor of weak hydrogen bonds (drawn with
dashed lines).
2. Crystal and Molecular Structure of the CuI(1)2Cl
Salt. On diffusion of n-hexane on a 4:1 v/v THF:MeCN
solution containing [CuI(CH3CN)4]ClO4, 1 (2 equiv), and Et3-
BzNCl (1 equiv), red-purple crystals of a salt of formula
CuI(1)2Cl that were suitable for X-ray diffraction studies
formed. The corresponding ORTEP plot of the complex salt
is shown in Figure 2.
The crystal structure consists of two nonsymmetrically
equivalent [CuI(1)2]Cl molecules with similar features, in
which each Cu atom is coordinated by four nitrogen atoms
of two phenanthroline subunits. However, in contrast to what
is suggested by modeling studies, the diarylurea pendant arms
of each metal complex do not converge to provide a
tetrahedral tetrafurcate binding site for chloride, but point
toward opposite directions, each one establishing bifurcate
hydrogen bonds with different Cl- ions. Moreover, it appears
that the two nonsymmetrically equivalent [CuI(1)2]+ mol-
ecules are held together by π-π interactions between two
phenanthroline subunits, which are arranged in an offset
manner, exhibiting a dihedral angle between the phenan-
throline rings of 2.0(1)° and a centroid-centroid distance
of 3.67(1) Å.
Noticeably, similar interactions occur in the whole crystal
between the phenanthroline fragments of symmetrically
related molecules (as shown in the stick model in Figure 3).
In particular, π-π stacking interactions are established
between [Cu(1)L2]+-[Cu(1)L2]+ and [Cu(2)L2]+-[Cu(2)-
L2]+ couples (L ) 1), the separation between the centroids
of overlapping phenanthrolines being 3.62(1) and 4.03(1) Å,
respectively. Further face-to-face π-stacking interactions are
also in place between adjacent diarylurea groups still
arranged according to an offset manner: the centroid-
centroid distances between superimposed diarylureas are in
the range 3.77(1)-3.92(1) Å. All these π-π interactions are
responsible for the array of infinite molecular layers placed
normal to the c axis. Quite interestingly, chloride ions play
an important role in stabilizing such an ordered crystalline
arrangement. In particular, each Cl- ion is H-bonded to two
urea fragments belonging to the copper(I) complexes of two
different cells. In particular, each chloride ion receives four
H-bonds, according to a rather distorted tetrahedron, with
the angles between the two NH-Cl-NH planes being 52.9-
(3) and 53.5(3)°. The hydrogen-bond motif results in infinite
chains formed by Cl(1)--[Cu(1)L]+-Cl(2)--[Cu(2)]L+-
Cl(1)-. Relevant structural parameters of the N-H‚‚‚Cl
interactions are reported in Table 1.
Extended π-π interactions and hydrogen bonds induce
severe distortions on the coordination geometry of both CuI
centers; in particular, each CuI center is placed out of the
best plane of both the chelate phenanthroline fragments:
displacements are 0.35(1) and 0.46(1) Å for Cu(1) and 0.27-
(1) and 0.35(1) Å for Cu(2). Therefore, the angles between
the two CuN2 planes (65.3(2)° for Cu(1) and 67.5(2)° for
Cu(2)) are different than those between the two phenathroline
rings bound to the same metal (52.2(1)° for Cu(1) and 56.8-
(1)° for Cu(2)). Both features are to be compared to the value
of 90° of an ideal tetrahedral [CuI(phen)2]+ complex (with
Cu in the plane of both ligands and the phenanthrolines
normal to each other) and to the value of 89.2° from the
structural modeling study.
For both metal centers, the bond distances with respect to
the two N atoms of the ortho-substituted aromatic rings
(2.187(7) Å for Cu(1) and 2.136(6) Å for Cu(2)) are longer
than those observed for the nitrogen atoms of the unsubsti-
(24) Bardwell, D. A.; Cargill Thompson, A. M. W.; Jeffery, J. C.; Tilley,
E. E. M.; Ward, M. D. J. Chem. Soc., Dalton Trans. 1995, 835.
Inorganic Chemistry, Vol. 45, No. 16, 2006 6141

Amendola et al.
Figure 3. Simplified sketch of the molecular layer normal to the c axis; the extensive face-to-face π-stacking interactions between the aromatic groups of
the [CuI(1)2]+ molecules are observed. Only H atoms bonded to the urea groups are shown, and the hydrogen-bond motifs involving NHCONH groups and
Cl- ions are drawn with dashed lines. Each of the chloride ions, Cl(1) and Cl(2), receives four hydrogen bonds.
Table 1. Structural Parameters of the N-H···Cl Interactionsa
donor group
D‚‚‚A (Å) H‚‚‚A (Å) D-H‚‚‚A (deg) acceptor atom
N(3a)-H(1Na) 3.256(7)
N(4a)-H(4Na) 3.212(8)
N(3b)-H(3Nb) 3.258(7)
N(4b)-H(4Nb) 3.268(7)
N(3c)-H(3Nc) 3.330(8)
N(4c)-H(4Nc) 3.184(8)
N(3d)-H(4Nd) 3.368(8)
N(4d)-H(4Nd) 3.243(7)
2.460(7)
2.447(8)
2.432(7)
2.451(7)
2.535(8)
2.387(8)
2.551(8)
2.420(7)
154.3(4)
148.6(5)
161.4(5)
158.9(5)
154.1(5)
154.5(5)
159.1(5)
160.5(5)
Cl(1)
Cl(1)
Cl(2)
Cl(2)
Cl(1′)
Cl(1′)
Cl(2′′)
Cl(2′′)
a Symmetry codes: (′) ) 1 - x, 2 - y, 1 - z; (′′) ) 2 - x, 1 - y,
1 - x.
Table 2. Bond Distances (Å) and Angles (deg) around the Two Metal
Centers
Figure 4. (a) Family of spectra obtained over the course of the titration
of a solution 3 × 10-4 M in 1 with a standard solution of [CuI(CH3CN)4]]-
ClO4 in a 4:1 v/v THF:MeCN medium; (b) titration profile based on the
absorbance of the band growing at 434 nm.
Cu(1)-N(1a)
1.964(6) Cu(2)-N(1c)
2.187(7) Cu(2)-N(2c)
2.029(7) Cu(2)-N(1d)
2.111(7) Cu(2)-N(2d)
81.30(24) N(1c)-Cu(2)-N(2c)
2.009(6)
2.136(6)
2.001(8)
2.127(8)
80.74(25)
Cu(1)-N(2a)
Cu(1)-N(1b)
Cu(1)-N(2b)
N(1a)-Cu(1)-N(2a)
N(1a)-Cu(1)-N(1b) 134.72(25) N(1c)-Cu(2)-N(1d) 134.40(27)
N(1a)-Cu(1)-N(2b) 132.30(25) N(1c)-Cu(2)-N(2d) 133.44(27)
N(2a)-Cu(1)-N(1b) 131.73(26) N(2c)-Cu(2)-N(1d) 130.82(28)
The formation and stability of the [CuI(1)2]+ complex in
the THF/MeCN mixture were investigated by titrating a
solution of [CuI(CH3CN)4]ClO4 with a standard solution of
1. On addition of the phenanthroline derivative, the solution
took a red-purple color and an absorption band, centered at
434 nm, appeared. The family of spectra recorded over the
course of the titration is shown in Figure 4a. The band, which
derives from a metal-to-ligand charge-transfer (MLCT)
transition within the [CuI(phenanthroline)2]+ complex, reaches
its maximum on addition of 0.5 equiv of CuI, indicating the
quantitative formation of the [CuI(phenanthroline)2]+ com-
plex (see the titration profile in Figure 4b). The absence of
any curvature suggests that the constant â2 of the Cu+ + 2L
a [CuIL2]+ complex formation equilibrium is too large to
be determined through a spectrophotometric titration experi-
ment. In any case, it can be assumed that in a THF/MeCN
solution, 1 × 10-6 to 1 × 10-4 M in [CuI(CH3CN)4]ClO4
and 2 × 10-6 to 2 × 10-4 M in 1, the conditions used in
this study, the [CuIL2]+ complex forms quantitatively.
Titration experiments were then carried out by adding a
N(2a)-Cu(1)-N(2b)
N(1b)-Cu(1)-N(2b)
93.16(27) N(2c)-Cu(2)-N(2d)
82.56(27) N(1d)-Cu(2)-N(2d)
96.21(27)
81.54(29)
tuted rings (1.964(6) for Cu(1) and 2.009(6)). Structural
parameters pertinent to metal-ligand interactions are shown
in Table 2.
3. Anion Binding in Solution: Poorly Polar Medium.
Solid-state results seem to indicate unfavorable arrangement
for selective anion binding, which requires face-to-face
juxtaposition of the two urea moieties, within the same metal
complex. However, geometrical features in the solid state
are determined by intermolecular interactions (π-π stacking
and Cl- bridging through H-bonds), which, in a diluted liquid
solution, can be disrupted in favor of solute-solvent interac-
tions. In any case, anion-binding studies were carried out in
a 4:1 v/v THF:MeCN solution. Such a medium was chosen
in order to maximize the solubility of 1 and of its copper(I)
complex.
6142 Inorganic Chemistry, Vol. 45, No. 16, 2006

Metal-Controlled Urea-Based Anion Receptor
Figure 5. (a) Family of spectra obtained over the course of the titration
of a solution 7.5 × 10-6 M in 1 with a standard solution of [Bu3BzN]Cl in
a 4:1 v/v THF:MeCN medium; (b) titration profile based on the absorbance
at 282 nm.
Figure 6. (a) Family of spectra obtained over the course of the titration
of a solution 7.5 × 10-6 M in 1 with a standard solution of [Bu4N]H2PO4
in a 4:1 v/v THF:MeCN medium; (b) titration profile based on the
absorbance at 282 nm.
The occurrence of the two stepwise equilibria can be visually
perceived in the titration profile in Figure 6b. In fact, a flex
point is observed that corresponds to the addition of 0.5 equiv
standard solution of the tetrabutylammonium salt of the
envisaged anion, [Bu4N]X, to a solution of the receptor
(either 1 or the [CuI(1)2]+ complex).
-
of H2PO4 . In particular, during the addition of the first 0.5
Figure 5a shows the family of spectra obtained over the
course of the titration of a solution 7.5 × 10-6 M in 1 with
a standard solution of [Bu3BzN]Cl. System 1 shows a band
centered at 270 nm, to be associated to a charge-transfer
transition along the NH-to-CF3 dipole, across the benzene
ring. On chloride addition, the band at 270 nm undergoes a
red shift to the limiting value of 282 nm, while its absorbance
increases, a behavior that is ascribed to the formation of the
[L‚‚‚Cl]- H-bond complex. In particular, on complex forma-
tion, negative charge is transferred from the anion to the
nitrogen atom of the urea fragment, which increases the
intensity of the dipole and accounts for the red shift of the
band. Multiwavelength least-squares analysis of spectral data,
using the HyperQuad program,17 indicated the occurrence
of the equilibrium L + Cl- a [L‚‚‚Cl]-, with log K ) 4.46
( 0.01. Formation of a 1:1 H-bond complex was also
observed for Br- (log K ) 3.75 ( 0.01). In the case of I-,
even at a 1 × 10-4 M concentration level, very moderate
spectral modifications were observed, and it can be assessed
only that log K e 2. On the other hand, on titration with the
strongly basic acetate anion, the titration profile clearly
indicated the formation of a 1:1 H-bond complex, but the
absence of any curvature even at a receptor concentration
of 1 × 10-6 M pointed toward an association constant log
K > 7.
equiv, the [L‚‚‚X‚‚‚L]- complex forms; then, on further anion
addition, the [L‚‚‚X‚‚‚L]- complex decomposes to give two
1:1 [L‚‚‚X]- complexes. The formation of a rather stable
-
2:1 complex is ascribed to the ability of the H2PO4 ion to
receive four H-bonds from two urea subunits, as tentatively
shown in structural sketch 5.
Note that such an opportunity is not allowed by the more-
basic Y-shaped CH3COO- anion, which can only establish
a bifurcate interaction with a single urea receptor. A similar
two-step interaction has been observed for the spherical
fluoride ion (see pertinent log K values in Table 3). At this
point, one could ask why spherical chloride and iodide ions
do not form the [L‚‚‚X‚‚‚L]- complex in the early stages of
the titration. In this instance, it can be suggested that both
Cl- and Br- are too poorly basic to give a stable H-bond
interaction with a second molecule of the urea receptor.
Similar titration experiments were then carried out on
solutions of the [CuI(1)2]+ complex. Figure 7a shows the
family of spectra obtained over the course of the titration of
a solution 1.3 × 10-4 M in the [CuI(1)2]+ receptor with
[Me3BzN]Cl. Also in this case, on chloride addition, a red
shift of the charge-transfer band was observed, whereas the
titration profile (Figure 7b) clearly indicated the formation
of a 1:1 receptor:anion adduct. It is therefore suggested that
the following association equilibrium occurs: [CuI(1)2]+ +
Cl- / [CuI(1)2‚‚‚Cl]. However, the titration profile does not
-
H2PO4 and F- anions displayed a different behavior. In
particular, Figure 6a shows the family of spectra obtained
over the course of the titration of a solution 7.5 × 10-6
M
in 1 with a solution of [Bu4N]H2PO4. Nonlinear least-squares
fitting of the titration data over the 250-350 nm wavelength
interval indicated the occurrence of two consecutive equi-
-
libria (X- ) H2PO4 )
L + X- / [L‚‚‚X]-; log K11 ) 5.35 ( 0.04
L + [L‚‚‚X]- / [L‚‚‚X‚‚‚L]-; log K21 ) 5.11 ( 0.12 (2)
(1)
Inorganic Chemistry, Vol. 45, No. 16, 2006 6143

Amendola et al.
Table 3. Association Constants (log K values) for the Interaction of Receptors L and [CuI(L)2]+ (L ) 1), in a 4:1 v/v THF:MeCN Mixture at 25 °C,
with Anions
anion
L + X- / [L‚‚‚X]-
L + [L‚‚‚X]- / [L‚‚‚X‚‚‚L]-
4.36 ( 0.12
[CuI(L)2]+ + X- / [CuI(L)2‚‚‚X]
F-
5.20 ( 0.03
4.46 ( 0.01
3.75 ( 0.01
<2
>7
5.35 ( 0.04
>7
>7
Cl-
Br-
6.15 ( 0.06
4.52 ( 0.01
>7
I-
CH3COO-
-
H2PO4
5.15 ( 0.12
>7
Formation of a [CuI(1)2‚‚‚X] complex was observed for
all the other investigated anions. In particular, remarkable
enhancement of the association constant was observed for
bromide (log K ) 6.15 ( 0.06) and iodide (log K ) 4.52 (
0.01). In the case of H2PO4- and F- ions, very steep titration
profiles were observed, which set log K values beyond the
detectable limit under the experimental conditions (log K >
7).
Thus, it appears that the preorientation effect exerted by
the CuI center allows for tetrafurcate H-bond formation with
the poorly basic chloride and bromide anions. On the other
-
hand, H2PO4 and F- ions are able to establish tetrafurcate
interaction with two urea subunits even in absence of
the templating metal center (vide supra). However, the
[L‚‚‚X‚‚‚L]- complex is not especially stable and decom-
poses on addition of an excess of X-, giving two 1:1
complexes. This is not the case of the corresponding [CuI-
(L)2‚‚‚X] complexes, which maintain their integrity even after
the addition of a large excess of X-. Only in the case of Cl-
was decomposition observed, but it was of a different nature.
In particular, excess addition of chloride caused the decrease
and disappearance of the MLCT band centered at 434 nm.
This indicates the occurrence of decomposition of the [CuI-
(phenanthroline2]+ complex, probably due to metal extrusion
and formation of the especially stable [CuICl2]- species. On
the other hand, excess addition of other investigated anions
did not modify the MLCT band.
Figure 7. (a) Family of spectra obtained over the course of the titration
of a solution 1.3 × 10-4 M in [CuI(1)2]+ with a standard solution of [Bu4-
BzN]Cl in a 4:1 v/v THF:MeCN medium; (b) titration profile based on the
absorbance at 282 nm.
show any curvature, which prevents the determination of a
reliable association constant. No curvature was observed on
titration of a 50-fold diluted solution of the [CuI(1)2]+
complex (2.6 × 10-6 M) either, which indicates that log K
> 7. Thus, the binding constant of chloride to the [CuI(1)2]+
receptor is at least 350-fold higher than that of the simple
receptor 1. Such a difference can hardly be ascribed to the
favorable electrostatic effect exerted by the Cu+ ion toward
the incoming anion, in view of relatively high distance and
ligand shielding effects. Rather, it is suggested that the [CuI-
(1)2]+ receptor, whatever its geometrical arrangement before
complexation, reorganizes to chelate the chloride ion with
the two urea pendant arms.
In the case of acetate, the effect of the metal on anion
binding could not be investigated, because even metal-free
receptor 1 formed a 1:1 H-bond complex [L‚‚‚CH3COO]-
with a log K > 7. Thus, we decided to study the interaction
of CH3COO- and H2PO4- with 1 and [CuI(1)2]+ in the more-
polar and competing solvent DMSO. In fact, because of the
higher solvating tendencies of DMSO, the anion desolvation
term in this medium is especially endergonic, which reduces
the values of association constants. This circumstance would
permit us to discriminate the behavior of more-basic anions
and to evaluate their recognition selectivity.
A tentative sketch of the 1:1 H-bond complex is shown
below, as structural formula 6, which reflects the calculated
structure in Figure 1.
4. Anion Binding in Solution: Highly Polar Medium.
DMSO played an effective discriminating role in anion
recognition by receptors 1 and [CuI(1)2]+. In particular,
addition of even a large excess of tetraalkylammonium
halides, including F-, did not cause any modification of the
absorption spectra of both 1 and [CuI(1)2]+. Also, on titration
of a solution 2.6 × 10-4 M in 1 with a standard solution of
[Bu4N]H2PO4, no detectable shift of the charge-transfer band
centered at 270 nm was observed, indicating a very low
solution stability of the H-bond complex (log K < 2).
However, on titrating a solution containing the [CuI(1)2]+
6144 Inorganic Chemistry, Vol. 45, No. 16, 2006

Metal-Controlled Urea-Based Anion Receptor
Figure 8. (a) Family of spectraobtained over the course of the titration of a solution 1.3 × 10-4 M in [CuI(1)2]+ with a standard solution of [Bu4N]H2PO4
in a DMSO medium; (b) titration profile based on the absorbance at 282 nm; (c) family of spectra obtained over the course of the titration of a solution 1.3
× 10-4 M in [CuI(1)2]+ with a standard solution of [Bu4N]CH3COO; (d) titration profile based on the absorbance at 282 nm.
Figure 9. Family of spectra obtained over the course of the titration of a solution 6.0 × 10-3 M in [CuI(1)2]+ with [Bu4N]H2PO4] in a 4:1 v/v DMSO-
d6:CD3CN medium.
receptor with [Bu4N]H2PO4, a pronounced red shift of the
charge-transfer band and substantial modifications of the
spectral pattern were observed, as shown in Figure 8a. The
titration profile, shown in Figure 8b, indicates the formation
of a H-bond complex of 1:1 stoichiometry, for which an
association constant log K ) 4.15 ( 0.01 was calculated
through nonlinear least-squares treatment of spectral data.
The metal-induced enhancement of the association constant
suggests that in the [CuI(1)2‚‚‚H2PO4] complex, the dihy-
drogenphosphate ion is chelated by two facing urea subunits,
as sketched in structural formula 6.
the broadening of the signals, indicating a fluxional process
on the NMR time scale, together with important structural
changes characterized by large shifts (∆δ) of Ha, Hb, Hc,
and Hd. Protons of the urea groups are shifted downfield,
indicating the formation of a H-bonded adduct with the anion
-
H2PO4 . Meanwhile, the protons located on the m-phenylene
linker undergo upfield shifts (∆δ(Ha) ) -0.81, ∆δ(Hc) )
-1.01, ∆δ(Hd) ) -0.52) and a downfield shift (∆δ(Hb) )
0.57). This behavior is reminiscent of the shifts observed
during the formation of helices in other m-phenylene-bridged
phenanthroline systems25 and can be attributed to the
chelation of H2PO4- by the two urea fragments (in agreement
with the UV-vis titration), leading to a helical structure in
the ion pair-receptor adduct in solution.
1
Such a hypothesis has been corroborated by a H NMR
titration experiment. Figure 9 shows the spectra obtained over
the course of the titration of a 4:1 v/v DMSO-d6:CD3CN
solution of [Cu(1)2](ClO4) (6 × 10-3 M) with [Bu4N](H2-
PO4). Spectra indicate that anion binding is responsible for
A quite different behavior was observed when the interac-
tion of receptors 1 and [CuI(1)2]+ with acetate was studied.
Inorganic Chemistry, Vol. 45, No. 16, 2006 6145

Amendola et al.
Following the spectrophotometric titration of plain receptor
1, an association constant log K ) 3.30 ( 0.01 was
calculated for H-bond complex [L‚‚‚CH3COO]-. On the
other hand, Figure 8c shows the family of spectra obtained
during the titration with [Bu4N]CH3COO of a solution of
the [CuI(1)2]+ receptor. The spectral pattern is qualitatively
similar to that obtained on titration with dihydrogenphosphate
(in Figure 8a), but the titration profile displayed in Figure
8d is distinctly less steep. In particular, a best fitting of the
spectral data was obtained on assuming the occurrence of
two consecutive equilibria (3) and (4), characterized by
stepwise association constants of a rather similar magnitude.
[CuI(1)2]+ + CH3COO- /
[CuI(1)2‚‚‚CH3COO] (3); log K1 ) 3.50 ( 0.01
Figure 10. Medium: DMSO, 1 × 10-4 M in [CuI(1)2]+. Solid line:
percent concentration of the [CuI(1)2‚‚‚H2PO4] complex over the course of
the titration with H2PO4-. Dashed lines: percent concentration of the [CuI-
(1)2‚‚‚CH3COO] complex (long dashes) and the [CH3COO‚‚‚CuI(1)2‚‚‚
CH3COO]- complex (short dashes) over the course of the titration with
CH3COO-. Percent concentration is given by the ratio of the current
concentration of the receptor-anion complex [CuI(1)2]+ over the total
concentration of the [CuI(1)2]+ receptor.
[CuI(1)2‚‚‚CH3COO] + CH3COO- /
[CuI(1)2‚‚‚(CH3COO)2]- (4); log K2 ) 3.24 ( 0.04
It should be noted that the CH3COO- ion can establish
bifurcate hydrogen bonding interaction with only one urea
subunit and cannot profit from the tetrafurcate binding site
made by two facing urea subunits offered by the [CuI(1)2]+
receptor, as H2PO4- and spherical halide anions do. Thus, it
is suggested that each acetate ion goes to interact with a urea
moiety oriented toward the outside. The arrangement of the
1:2 H-bonded complex, [CuI(1)2‚‚‚(CH3COO)2]-, is tenta-
tively sketched as formula 7. Notice that the log K1 and log
K2 values are rather close to the log K value determined for
acetate interaction with metal-free receptor 1 (log K ) 3.30)
and that the [log K1 - log K2] difference is close to the value
expected on pure statistical bases (∆log K ) log 4 ) 0.6).
This suggests independent binding of each CH3COO- ion
at remote and noninterfering urea subunits, as tentatively
illustrated by formula 7.
tions of the phosphate and acetate complexes vary over the
course of the titration of solutions 1 × 10-4 M in the [CuI-
(1)2]+ receptor.
It is observed that in the titration with phosphate, on
-
addition of 1 equiv of anion, 49.9% of H2PO4 has been
taken by the [CuI(1)2]+ receptor. On the other hand, in the
titration with acetate, after 1 equiv addition, only 23.7% of
the total CH3COO- has been removed from the solu-
tion (18.8% as [CuI(1)2‚‚‚CH3COO] and 2 × 2.5% as
[CH3COO‚‚‚CuI(1)2‚‚‚CH3COO]-). Recognition selectivity
-
of H2PO4 over CH3COO- increases as the concentration
of the [CuI(1)2]+ receptor further decreases.
Conclusion
As recently pointed out,26 the young discipline of anion
coordination chemistry27 has learned some useful tricks from
its older and more-experienced sister, metal coordination
chemistry,28 which includes the chelate effect.29 The neutral
bifurcate H-bond donor urea, L, binds a spherical anion X-,
just as the neutral bidentate ligand ethylenediamine (en)
forms two coordinative bonds with a transition-metal ion
Mn+. However, whereas a second en molecule can further
coordinate the metal without any substantial reduction of
affinity, giving a [M(en)2]n+ complex with a variety of
transition metals in solvents of different polarity, including
water, a second molecule of urea is able to establish
hydrogen-bonding interactions with the anion, giving the
[L‚‚‚X‚‚‚L]- H-bond adduct, only with especially basic ions
(25) (a) Dietrich-Buchecker, C. O.; Colasson, B.; Jouvenot, D.; Sauvage,
J.-P. Chem.sEur. J. 2005, 11, 4374. (b) Dietrich-Buchecker, C. O.;
Sauvage, J.-P. Chem. Commun. 1999, 615. (c) Dietrich-Buchecker,
C. O.; Sauvage, J.-P.; De Cian, A.; Fischer, J. J. Chem. Soc., Chem.
Commun. 1994, 2231.
(26) Bowman-James, K. Acc. Chem. Res. 2005, 38, 671.
(27) Graf, E.; Lehn, J.-M. J. Am. Chem. Soc. 1976, 98, 6403.
(28) Werner, A. Z. Anorg. Allg. 1893, 3, 267.
Thus, it appears that the [CuI(1)2]+ receptor, at least in
the presence of a stoichiometric amount of anion, inverts
the selectivity trend on the basis of basicity and chooses to
-
bind H2PO4 in the presence of the more-basic CH3COO-.
As an example, Figure 10 shows how the percent concentra-
(29) Schwarzenbach, G. HelV. Chim. Acta 1952, 35, 2344.
6146 Inorganic Chemistry, Vol. 45, No. 16, 2006

Metal-Controlled Urea-Based Anion Receptor
-
(e.g., F- and H2PO4 ) and in poorly polar media. Such a
K+) and an anion (e.g., Cl-) in close contact; in the presence
of K+, the receptor’s affinity toward Cl-, in a DMSO
solution, was 13-fold enhanced because of a cooperative
electrostatic effect.31 On the other hand, Beer has shown that,
in a ditopic receptor containing oxa-crown and amide binding
sites, the presence of K+ inverts the selectivity toward Cl-
and H2PO4- because of the different inference of cooperative
electrostatic and conformational allosteric effects.32 The
conjugate urea-phenanthroline system described here is an
example of a self-assembling heteroditopic receptor for d10
metal salts, e.g., CuIX. The metal-induced enhancement of
the affinity toward anions, which excludes any electrostatic
contribution and is simply related to conformational effects,
varies over 2-3 orders of magnitude and is much higher
than that observed in previously reported examples.
failing behavior is to be ascribed to the intrinsic weakness
of hydrogen-bonding interactions compared to metal-ligand
interactions. However, if two urea subunits are linked
together by a spacer of suitable length, anion chelation may
occur, essentially because of a favorable entropy contribution.
In previously quoted examples,6-8 the urea subunits have
been linked covalently through a spacer or a platform whose
nature defines the geometrical features of the receptor’s donor
set. Alternatively, as shown here, the urea subunits are
allowed to assemble together, through coordination to a metal
center, which acts as a template. The coordinative geo-
metrical requirements of the metal as well as the design of
the urea-ligand conjugate determine size and orientation of
the H-bond binding site and help to define the selectivity of
anion recognition. Moreover, the use of a kinetically labile
metal ion (e.g., CuI) imparts versatility to the receptor, which
can switch from a locked arrangement, suitable to the
recognition of spherical anions, including dihydrogenphos-
phate, to an unlocked arrangement, which favors the interac-
tion with nonspherical anions, such as the Y-shaped acetate.
Finally, it should be observed that there exists a current
and increasing interest on ion-pair recognition,30 i.e., the
simultaneous selective complexation of a metal ion and an
anion by a heteroditopic receptor. For instance, Smith has
designed rigid polycyclic receptors containing amide N-H
fragments that are capable of accommodating a cation (e.g.,
Acknowledgment. The financial support of the European
Union (RTN Contract HPRN-CT-2000-00029) and the Italian
Ministry of University and Research (PRINsDispositivi
Supramolecolari; FIRBsProject RBNE019H9K) is gratefully
acknowledged.
Supporting Information Available: X-ray crystallographic files
in CIF format for the [CuI(1)2]Cl complex salt. This material is
available free of charge via the Internet at http://pubs.acs.org.
IC060160X
(31) Mahoney, J. M.; Beatty, A. M.; Smith, B. D. J. Am. Chem. Soc. 2001,
123, 5847.
(30) Beer, P. D.; Gale, P. A. Angew. Chem., Int. Ed. 2001, 40, 486.
(32) Beer, P. D.; Dent, S. W. Chem. Commun. 1998, 825.
Inorganic Chemistry, Vol. 45, No. 16, 2006 6147