Halide ion inclusion into a dicopper(II) bistren cryptate containing 'active' 2,5-dimethylfuran spacers: The origin of the bright yellow colour

Article abstract of DOI:10.1016/j.ica.2008.03.049

The inclusion of halide ions into a dicopper(II) bistren cryptate complex containing 2,5-dimethylfuran spacers has been investigated through spectrophotometric titration experiments in MeCN solution. X-ray diffraction studies on the 1:1 chloride inclusion complex have shown that the encapsulated halide ion and the furan oxygen atoms lie at an interacting distance. Such an interaction perturbs the energy of the halide-to-copper(II) charge transfer transition, which is shifted to the visible region. As a consequence, an intense yellow colour develops on halide inclusion. Such a colour change is not observed on chloride or bromide inclusion into the dicopper(II) bistren cryptate containing spacers which are not capable to interact with the encapsulated halide and do not perturb the charge transfer transition, e.g. 1,3-xylyl fragments.

Full text of DOI:10.1016/j.ica.2008.03.049

Inorganica Chimica Acta 361 (2008) 4038–4046
Contents lists available at ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
Halide ion inclusion into a dicopper(II) bistren cryptate containing ‘active’
2,5-dimethylfuran spacers: The origin of the bright yellow colour
a
Valeria Amendola a, Greta Bergamaschi b, Massimo Boiocchi c, Luigi Fabbrizzi a, , Antonio Poggi ,
*
Michele Zema d
a Dipartimento di Chimica Generale, Università di Pavia, via Taramelli 12, 27100 Pavia, Italy
b Sezione INSTM, Università di Pavia, Italy
c Centro Grandi Strumenti, Università di Pavia, via Bassi 21, 27100 Pavia, Italy
d Dipartimento di Scienze della Terra, via Ferrata 1, Università di Pavia, 27100 Pavia, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
The inclusion of halide ions into a dicopper(II) bistren cryptate complex containing 2,5-dimethylfuran
spacers has been investigated through spectrophotometric titration experiments in MeCN solution.
X-ray diffraction studies on the 1:1 chloride inclusion complex have shown that the encapsulated halide
ion and the furan oxygen atoms lie at an interacting distance. Such an interaction perturbs the energy of
the halide-to-copper(II) charge transfer transition, which is shifted to the visible region. As a conse-
quence, an intense yellow colour develops on halide inclusion. Such a colour change is not observed
on chloride or bromide inclusion into the dicopper(II) bistren cryptate containing spacers which are
not capable to interact with the encapsulated halide and do not perturb the charge transfer transition,
e.g. 1,3-xylyl fragments.
Received 17 February 2008
Received in revised form 9 March 2008
Accepted 11 March 2008
Available online 16 March 2008
Dedicated to Dante Gatteschi.
Keywords:
Cryptate complexes
Anion recognition
Ó 2008 Elsevier B.V. All rights reserved.
Spectrophotometric titrations
Charge transfer transitions
Colorimetric anion sensors
1. Introduction
Anion recognition is a topic of current interest in supramolecu-
lar chemistry [1]. Recognition refers to the selective interaction of
the host system – the receptor – with the anion – the guest. Selec-
tivity mainly depends upon (i) the energy of the interaction and (ii)
the match of the geometrical features of the receptor with those
of the anion. The major part of the anion receptors described
during the last two decades operated through electrostatic interac-
tions (which include hydrogen bonding). Due to the intrinsic weak-
ness of these interactions, formation of stable complexes requires a
high level of preorganization of the receptor, which has to be de-
signed taking into account the shape and the size of the anion.
The highest level of preorganization is achieved with tridimen-
sional systems, e.g. cages [2].
Indeed, one of the first deliberately designed receptors for an-
ions was the hexa-ammonium ion derived from the bistren crypt-
and 1, which is capable to include a variety of inorganic anions in
* Corresponding author. Tel.: +39 0382 987 328; fax: +39 0382 528 544.
E-mail address: luigi.fabbrizzi@unipv.it (L. Fabbrizzi).
0020-1693/$ - see front matter Ó 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.ica.2008.03.049

V. Amendola et al. / Inorganica Chimica Acta 361 (2008) 4038–4046
4039
an acidic aqueous solution (pH 6 5), to give stable 1:1 complexes
[3]. Acidity is required in order to keep protonated the six second-
ary amine nitrogen atoms of the parent bistren cage and to provide
the receptor a multi-positive charge. However, a positive charge
can be more conveniently constituted by placing inside the crypt-
and two transition metal centres, e.g. CuII [4]. In particular, each
metal goes to occupy a tren subunit, giving a complex species of
trigonal bipyramidal geometry. When bound to the tren moiety,
each CuII ion is coordinatively unsaturated and maintains one of
the axial positions available to the coordination of the donor atom
of a further ligand. As an example, the bistren cryptand 2, in which
the two tetramine subunits are linked by 1,3-xylyl spacers, is able
to include, according to a cascade process, first two CuII ion, then
an ambidentate ion Xꢀ, which bridges the two metal centres, to
tances between the halide ion and furan oxygen atoms is instru-
mental for interpreting the unique spectral features of the halide
inclusion complexes.
2. Experimental
2.1. General procedures and materials
Cryptands 2 and 4 were prepared following described proce-
dures [5,11]. In titration experiments, UV–Vis spectra were regis-
tered on a scanning Varian Cary 100 spectrophotometer. The cell
holder was thermostatted at 25.0 °C, through circulating water.
Tipically, aliquotes of a fresh tetrabutylammonium salt standard
solution of the envisaged anion were added and the UV–Vis spectra
of the samples were recorded. For chloride, the [Et3Bn]Cl salt was
used, All spectrophotometric titration curves were fitted with the
HYPERQUAD program [14].
3þ
give the ternary dinuclear complex ½CuI2Ið2ÞXꢁ [5]. In particular,
complexes with Xꢀ = N3 [6], NCOꢀ [6], HCO3 [7] have been iso-
lated and structurally characterized through X-ray diffraction stud-
ies. Moreover, titration experiments in aqueous solution disclosed
a nice and unprecedented geometrical selectivity in anion recogni-
tion. In particular, a peak diagram was observed when plotting
logK of the inclusion equilibrium versus anion bite, i.e. the distance
between two consecutive donor atoms of the anion [8]. The highest
ꢀ
ꢀ
2.2. X-ray crystallographic studies
Synthetic single crystals of ½CuIIð4ÞClꢁ (ClO4)3 complex were of
2
small dimension and of poor X-ray diffraction quality. Diffraction
data were collected in the h range 2–20° at ambient temperature
by means of an Enraf-Nonius CAD4 four circle diffractometer,
working with graphite-monochromatized Mo Ka X-radiation
(k = 0.7107 Å). Intensities of reflections having h greater than 20°
were unobservable. Data reductions (including intensity integra-
tion, background, Lorentz and polarization corrections) were per-
formed with the WinGX package [15]. Absorption effects were
evaluated with the psi-scan method [16] and absorption correction
was applied to the data (min./max. transmission factors were
0.562/0.915).
ꢀ
stability was observed with the N3 anion, which is capable of
placing its terminal donor atoms in the axial positions of the two
CuII(tren)2+ moieties of the dinuclear cryptate, without inducing
any endoergonic rearrangement of the cage framework. Anions
providing either a shorter (e.g. NO3ꢀ) or a longer bite length (e.g.
NCSꢀ) cause an unfavourable conformational reorganization of
the cryptate skeleton, which is reflected in a drastic decrease of
the inclusion constant (up to two orders of magnitude). Changing
of the spacers linking the two tren subunits may change the recog-
nition tendencies of the dimetallic cryptate. This behaviour is quite
obvious when the length of the spacer is drastically modified. As an
example, the dicopper(II) complex of cryptand 3 includes both aro-
matic and aliphatic dicarboxylates [9], and is tailor made for the
Crystal structure was solved by direct methods (SIR 97 [17]) and
refined by full-matrix least-square procedures on F2 using all
reflections (SHELXL 97) [18]. However, the poor X-ray diffraction
quality of the crystal prevented an anisotropic structure refine-
ment for all non-hydrogen atoms. Therefore, only the two CuII cen-
tres and the Cl atoms were refined with unconstrained anisotropic
encapsulation of the L-glutamate neurotransmitter (as well as of
the glutarate ion, in which the two –COOꢀ groups are separated
by the same number of carbon atoms: two [10]). However, drastic
changes of anion recognition properties have been observed also
when the nature of the spacer is changed, even if its geometrical
feature are not seriously modified. This is the case of the dicop-
per(II) complex of cryptand 4, which contains 2,5-dimetylfuran
spacers. The length of 2,5-dimethylfuran fragment is rather close
to that of the 1,3-xylyl spacer; however, anion recognition proper-
Table 1
Crystal data for investigated crystals
½CuI2Ið4ÞClꢁ (ClO4)3
4þ
ties of the ½CuI2Ið4Þꢁ complex are quite different from those of the
Formula
M
Colour
C30H48Cl4Cu2N8O15
1029.66
orange
½CuI2Ið2Þꢁ4þ analogue. In particular, the ½CuI2Ið4Þꢁ4þ cryptate forms sta-
ble 1:1 inclusion complexes also with monoatomic anions like ha-
lides, which show an intense bright yellow colour [11]. Equilibrium
studies in aqueous solution indicated the formation of the perti-
Dimension (mm)
Crystal system
Space group
a (Å)
b (Å)
c (Å)
0.35 ꢃ 0.14 ꢃ 0.07
monoclinic
P21/c (no. 14)
16.441(6)
9.969(5)
27.463(7)
100.77(3)
4422.0(26)
4
1.547
2120
1.275
x scans
3þ
nent the ½CuI2Ið4ÞXꢁ inclusion complex at pH ꢂ 5. On increasing
pH, the halide was displaced from the cage by the OHꢀ ions, which
formed a very stable emerald green complex, in which the hydrox-
ide ion bridged the two metal centres, placed at an unusually short
b (°)
V (Å3)
4þ
distance [12]. The strikingly different behaviour of the ½CuI2Ið4Þꢁ
Z
qcalc (g cmꢀ3
F(000)
)
4þ
receptor with respect to ½CuI2Ið2Þꢁ was basically ascribed to a
greater flexibility of the 2,5-dimethylfuran compared to 1,3-xylyl
[13].
l Mo Ka (mmꢀ1
Scan type
)
In this article, we intend to explore in detail the formation of the
h Range (°)
2–20
4210
4090
0.1513
1160
327
0.1190, 0.2733
0.3746, 0.3514
0.885
0.984/ꢀ0.843
4þ
halide inclusion complexes of the ½CuI2Ið4Þꢁ receptor, with a spe-
Measured reflections
Unique reflections
Rint
Strong data [I0 > 2r(I0)]
Refined parameters
R1, wR2 (strong data)
R1, wR2 (all data)
cial reference to the development of the intense yellow colour (of
interest for the design of a colorimetric sensor for halides). In order
to avoid the interference of the hydroxide ion, we have carried out
equilibrium studies in anhydrous MeCN. The crystal and molecular
3þ
structure of the chloride inclusion complex ½CuI2Ið4ÞClꢁ has been
Goodness-of-fit
determined and compared with that previously reported for the
Maximum/minimum residuals (e Åꢀ3
)
3þ
bromide analogue ½CuI2Ið4ÞBrꢁ [11]. The surprisingly short dis-

4040
V. Amendola et al. / Inorganica Chimica Acta 361 (2008) 4038–4046
Table 2
atom displacement factors. The O atoms of the perchlorate counte-
rions were treated as anisotropic species with Uij components re-
strained to simulate an isotropic behaviour (ISOR instruction of
SHELXL-97) and the C, N and O atoms forming the cryptand moiety
were refined with isotropic atom displacement factors. All hydro-
gen atoms were placed at calculated positions with the appropriate
AFIX instructions and refined using a riding model.
The slightly elevated R1 and wR2 agreement indexes measured
for all reflections are due to the poor X-ray diffraction quality of
the crystal, that shows only 28% of intensities with an I/r(I) ratio
greater than 2 and an average I/r(I) = 2.52. However, the final re-
fined model does not exhibit any unacceptable chemical features
and the obtained crystal structure is more than suitable for the
aims of this study. Crystal data for studied crystal are shown in
Table 1.
Selected bond distances (Å) and bond angles (°) involving the two CuII centres
Cu(1)–Cl(1)
Cu(1)–N(1)
Cu(1)–N(2)
Cu(1)–N(5)
Cu(1)–N(7)
2.337(9)
Cu(2)–Cl(1)
Cu(2)–N(4)
Cu(2)–N(3)
Cu(2)–N(6)
Cu(2)–N(8)
2.338(9)
2.007(24)
2.029(22)
2.073(18)
2.089(23)
2.058(26)
2.052(29)
2.116(19)
2.073(25)
Cl(1)–Cu(1)–N(2)
Cl(1)–Cu(1)–N(5)
Cl(1)–Cu(1)–N(7)
N(1)–Cu(1)–N(2)
N(1)–Cu(1)–N(5)
N(1)–Cu(1)–N(7)
N(2)–Cu(1)–N(5)
N(2)–Cu(1)–N(7)
N(5)–Cu(1)–N(7)
94.1(7)
93.3(6)
96.8(7)
87.9(10)
84.2(9)
83.5(10)
119.3(9)
122.6(8)
116.1(9)
Cl(1)–Cu(2)–N(3)
Cl(1)–Cu(2)–N(6)
Cl(1)–Cu(2)–N(8)
N(4)–Cu(2)–N(3)
N(4)–Cu(2)–N(6)
N(4)–Cu(2)–N(8)
N(3)–Cu(2)–N(6)
N(3)–Cu(2)–N(8)
N(6)–Cu(2)–N(8)
93.9(8)
97.6(7)
93.6(8)
84.6(11)
83.8(10)
86.8(11)
118.5(10)
126.2(10)
113.1(10)
3. Results and discussion
3þ
3.28(1)–3.32(1) Å for the ½CuI2Ið4ÞBrꢁ analogue. Quite significantly,
these distances are markedly higher than the those between the
halide anion and the proximate furan oxygen atoms. Actually,
Oꢄ ꢄ ꢄX distances range from 3.14(2) to 3.18(2) Å for the chloride
inclusion complex and from 3.17(1) to 3.21(1) Å for the bromide
analogue. These values are appreciably lower than the sum of the
average van der Waals radii (1.75 + 1.52 = 3.27 Å for Clꢄ ꢄ ꢄO;
1.85 + 1.52 = 3.37 Å for Brꢄ ꢄ ꢄO) [19]. This suggests the occurrence
in both dimetallic complexes of a significant interaction between
the furan oxygen atoms of the cage and the encapsulated halide
anion. Such an evidence will be important for interpreting the un-
3.1. The crystal and molecular structure of the ½CuIIð4ÞClꢁ (ClO4)3
2
complex salt
The X-ray crystallographic study showed that the crystal of the
½CuIIð4ÞClꢁ (ClO4)3 cryptate salt and that of the previously studied
2
½CuIIð4ÞBrꢁ (ClO4)3 cryptate salt [11] are isostructural. An ORTEP
2
view of the ½CuI2Ið4ÞClꢁ3þ complex cation is displayed in Fig. 1, show-
ing that the chloride anion is placed at the centre of the cage and
bridges the two metal centres according to a collinear arrangement
(Cu(1)–Cl(1)–Cu(2) angle = 179.1(5)°).
3þ
3þ
ique spectral features of the ½CuI2Ið4ÞClꢁ and ½Cu2IIð4ÞBrꢁ com-
Selected geometrical features are reported in Table 2. Each CuII
centre shows a trigonal-bipyramidal coordination geometry. The
complex exhibits a C3h molecular symmetry, with each metal cen-
tre slightly displaced from the best plane of the three secondary
amine atoms (N3) towards the chloride ion. Displacement from
the N3 plane is 0.17(1) Å for Cu(1) and 0.18(1) Å for Cu(2). The
plexes (vide infra).
4þ
3.2. The behaviour of the ½CuI2Ið4Þꢁ complex in MeCN solution:
chloride and bromide ion inclusion
4þ
Formation and stability of the ½CuI2Ið4Þꢁ complex in MeCN solu-
CuIIꢄ ꢄ ꢄCuII distance (4.86(1) Å) is expectedly lower than that ob-
3þ
served with the previously described ½CuI2Ið4ÞBrꢁ
complex
tion were investigated by carrying out spectrophotometric titra-
tion experiments. Fig. 2a shows the family of spectra obtained
over the course of the titration of a solution 5.4 ꢃ 10ꢀ4 M of 4 with
a standard solution of CuII(CF3SO3)2.
(4.67(1) Å), due to the smaller size of the bridging halide ion. How-
ever, the difference of the CuIIꢄ ꢄ ꢄCuII distances (0.19 Å) is distinctly
smaller than the difference between the diameters of the two
halide ions: 0.30 Å. Quite interestingly, the distances between the
halide anion and the centroid of the furan rings are similar in
the two dicopper(II) cryptates. In particular, pertinent values
On metal addition, well definite bands develop both in the UV
and visible regions. High energy bands (300–400 nm) have an
LMCT nature and refer to amine nitrogen atom to CuII charge trans-
fer transitions. The broad band in the 600–900 nm interval (Fig. 2b)
is the envelope of the d–d transitions pertinent to a d9 cation in a
trigonal bipyramidal coordinative environment (each remaining
axial position being probably occupied by a MeCN molecule). The
titration profiles shown in Fig. 2c show a sharp discontinuity after
the addition of 2 equiv. of CuII, which clearly indicates the forma-
tion of a stable complex of 1:2 ligand/metal ratio, whose formation
3þ
are in the range 3.25(2)–3.33(2) Å for ½CuI2Ið4ÞClꢁ complex and
constant is higher than 106. It derives that solutions 10ꢀ3–10ꢀ4
M
in 4 and 2 ꢃ 10ꢀ3–2 ꢃ 10ꢀ4 M in CuII(CF3SO3)2 contain more than
4þ
99% of the ½CuI2Ið4Þꢁ complex. Thus, in anion recognition studies,
4þ
solutions of the ½CuI2Ið4Þꢁ receptor have been safely prepared by
dissolving in MeCN the bistren cryptand
CuII(CF3SO3)2.
4 and 2 equiv. of
Fig. 3a shows the complete family of spectra obtained over the
course of the titration of an MeCN solution 5.3 ꢃ 10ꢀ4 M of
4þ
½CuI2Ið4Þꢁ with a standard MeCN solution of [BnEt3N]Cl. On chlo-
4þ
ride addition the pale blue solution of the ½CuI2Ið4Þꢁ receptor be-
gins to take a bright yellow colour, while drastic spectral changes
are observed. In particular, an intense band centred at 410 nm
develops up to more than 6000 Mꢀ1 cmꢀ1, which is responsible
for the intense yellow colour. Definite modifications are observed
also in the d–d region of the spectrum (500–900 nm) (see
Fig. 3b). The band centred at 410 nm results from a chloride-to-
copper(II) charge transfer transition and is characterized by an
3þ
Fig. 1. ORTEP view of the ½CuI2Ið4ÞClꢁ cationic cryptate (perchlorate counterions
and hydrogen atoms have been omitted for clarity; ellipsoids are drawn at the 30%
probability level). Only Cu and Cl atoms were refined with anisotropic atom disp-
lacement parameters, whereas isotropic displacement parameters were used for all
remaining atoms.

V. Amendola et al. / Inorganica Chimica Acta 361 (2008) 4038–4046
4041
Fig. 2. (a) Spectra recorded over the course of the titration of a MeCN solution 5.4 ꢃ 10ꢀ4 M in 4, with a standard solution of CuII(CF3SO3)2. (b) d–d absorption bands.
(c) Titration profiles at 370 and 750 nm.
Fig. 3. (a) Spectra recorded over the course of the titration of a MeCN solution 5.3 ꢃ 10ꢀ4 M in ½Cu2IIð4Þꢁ4þ, with a standard solution of [BnEt3N]Cl. (b) d–d absorption bands.
(c) Titration profiles at 440 and 650 nm.
ꢀ
2þ
½CuI2Ið4ÞꢁClꢁ3þ þ Cl ¡ ½CuIIð4ÞꢁCl2ꢁ
ð3Þ
energy much lower than observed in mononuclear complexes of
type [CuII(tren)Cl]+, where the Clꢀ-to-CuII transition occurs at
k 6 350 nm and overlaps with the amine-to-CuII transition (and
the colour of the ternary complex is pale blue-green). Interesting
pieces of information on the stoichiometry of the species which
form in solution over the course of the titration experiment can
be taken from the profiles displayed in Fig. 3b. It is observed that
the Clꢀ-to-CuII CT band (filled triangles in Fig. 3c) increases until
the addition of 1 equiv. of Clꢀ, then it decreases in intensity to dis-
appear after the addition of 2 equiv. Discontinuity after the addi-
tion of 1 equiv. of chloride is observed also in the d–d region (see
open triangles in Fig. 3c), suggesting the occurrence of drastic geo-
metrical changes in the complex species present in solution. More-
over, a careful inspection of both titration profiles reported in Fig. 3
a indicates the existence of an inflection at 0.5 equiv.
2
Corresponding equilibrium constants are reported in Table 3 as
logK values.
Fig. 4 shows the concentration profiles of the species which
form over the course of the titration.
It can be noted that, on the very first additions of chloride, the
7þ
complex species f½CuI2Ið4Þꢁ ꢄ ꢄ ꢄ Cl ꢄ ꢄ ꢄ ½CuI2Ið4Þꢁg forms, according to
Eq. (1). In this species, which reaches its maximum concentration,
ca. 30%, on addition of 0.5 equiv. of Clꢀ, the chloride ion is not in-
cluded into the cage, but bridges two CuII ions of two distinct cryp-
tate complexes. On further addition of chloride, the 1:1 complex
3þ
½CuI2Ið4ÞꢁClꢁ appears as a major species. It is suggested that in this
species the chloride ion is included into the cryptate and bridges
the two CuII centres, as observed in the complex salt structurally
investigated by X-ray diffraction studies. To the 1:1 inclusion com-
plex the intense yellow colour corresponds, as indicated by the sat-
isfactory superimposition of the absorbance profile of the Clꢀ-to-
CuII charge transfer band (taken at 440 nm, open triangles in
The entire set of spectrophotometric titration data was ana-
lyzed with a non-linear least-squares procedure, using the HYPER-
QUAD program [14]. Best fitting of titration data was obtained by
assuming the occurrence of the following equilibria:
3þ
Fig. 4) with the concentration profile of the ½CuI2Ið4ÞꢁClꢁ species.
ꢀ
7þ
2½CuI2Ið4Þꢁ4þ þ Cl ¡ f½Cu2IIð4Þꢁ ꢄ ꢄ ꢄ Cl ꢄ ꢄ ꢄ ½CuI2Ið4Þꢁg
ð1Þ
ð2Þ
The inclusion complex reaches its maximum concentration (more
than 90%) on addition of 1 equiv. of Clꢀ. On further chloride
ꢀ
f½CuI2Ið4Þꢁ ꢄ ꢄ ꢄ Cl ꢄ ꢄ ꢄ ½Cu2IIð4Þꢁg7þ þ Cl ¡ 2½CuI2Ið4ÞꢁClꢁ
3þ

4042
V. Amendola et al. / Inorganica Chimica Acta 361 (2008) 4038–4046
Table 3
LogK values for the equilibria occurring in a MeCN solution, at 25 °C with the systems ½CuI2Ið4Þꢁ4þ=Xꢀ (X = Cl, Br)
Equilibrium
Clꢀ
Brꢀ
Iꢀ
4þ
7þ
2½CuI2Ið4Þꢁ þ Xꢀ ¡ f½Cu2IIð4Þꢁ ꢄ ꢄ ꢄ X ꢄ ꢄ ꢄ ½CuI2Ið4Þꢁg
13.2 0.1
4.2 0.1
4.5 0.1
>6
10.6 0.1
2.8 0.1
4.0 0.1
>6
4.46 0.05
5.4 0.1
7þ
3þ
f½CuI2Ið4Þꢁ ꢄ ꢄ ꢄ X ꢄ ꢄ ꢄ ½Cu2IIð4Þꢁg þ Xꢀ ¡ 2½CuI2Ið4ÞꢁXꢁ
3þ
2þ
½CuI2Ið4ÞꢁXꢁ þ Xꢀ ¡ ½Cu2IIð4ÞꢁX2ꢁ
½CuI2Ið4Þꢁ4þ þ Xꢀ ¡ ½Cu2IIð4ÞꢁXꢁ
4.94 0.03
3þ
pp level of chloride, which acquires an essentially antibonding
character and from which the electron is excited to the dz2 level
of the metal. The energy of the transition, following Clꢄ ꢄ ꢄO interac-
tion, is therefore reduced, which makes the pertinent absorption
band shift to the visible region. The Clꢄ ꢄ ꢄO interaction cannot be
3þ
established in the f½CuI2Ið4Þꢁ ꢄ ꢄ ꢄ X ꢄ ꢄ ꢄ ½CuI2Ið4Þꢁg species, where the
bridging chloride ion is outside of the cage and far away from
2þ
the furan oxygen atoms, and in the ½CuIIð4ÞꢁCl2ꢁ complex, in
2
which the two coordinated anions, in order to minimize electro-
static repulsions, probably protrude outside of the cage. Thus, in
3þ
the complexes f½Cu2IIð4Þꢁ ꢄ ꢄ ꢄ X ꢄ ꢄ ꢄ ½Cu2IIð4Þꢁg and ½CuIIð4ÞꢁCl2ꢁ2þ, the
2
normal Clꢀ-to-CuII charge transfer transition takes place.
A completely analogous behaviour was observed with bromide.
Fig. 6a shows the family of spectra taken over the course of the
4þ
titration of an MeCN solution 5.3 ꢃ 10ꢀ4 M of ½Cu2IIð4Þꢁ with a
standard MeCN solution of [Bu4N]Br. Development and decrease
of the absorption band associated to the Brꢀ-to-CuII charge transfer
transition are observed (see the absorbance profile in Fig. 6b). The
Fig. 4. Concentration profiles of the complex species forming over the course of the
spectrophotometric titration illustrated in Fig. 3. Profiles of the absorbances at 440
and 650 nm have been superimposed on the diagram.
same three bromide containing complex species (f½CuI2Ið4Þꢁ ꢄ ꢄ ꢄ Br ꢄ ꢄ ꢄ
½CuI2Ið4Þꢁg7þ, ½CuI2Ið4ÞꢁBrꢁ and ½CuIIð4ÞꢁBr2ꢁ2þ) formed in solution
3þ
2
3þ
addition, the concentration of the ½CuI2Ið4ÞꢁClꢁ complex decreases
over the course of the titration. LogK values associated to the step-
wise equilibria of type (1)–(3) are reported in Table 3. In all cases,
logK values are lower than observed in the case of corresponding
equilibria involving chloride. This reflects the lower coordinating
tendencies of bromide compared to chloride, generally observed
in transition metal coordination chemistry. In Table 3, the logK val-
ues associated to the inclusion Eq. (4) are also compared.
and the yellow colour progressively vanishes, while the new spe-
2þ
cies ½CuIIð4ÞꢁCl2ꢁ forms. It is suggested that in this complex two
2
chloride ions are not included into the cage (a circumstance
strongly disfavoured by electrostatic repulsions), but each anion
should be coordinated to a CuII centre in a rather distorted mode.
This may induce a rearrangement of the coordination sphere of
each metal ion, which is signalled by a modification of the d–d
absorption, as also indicated by the discontinuity of the titration
½CuI2Ið4Þꢁ4þ þ Xꢀ ¡ ½Cu2IIð4ÞꢁXꢁ
ð4Þ
3þ
profile on the d–d band (filled triangles in Fig. 4). On excess addi-
It is observed that inclusion of Clꢀ is favoured by two orders of mag-
nitude with respect to Brꢀ, in agreement with what observed in an
aqueous solution adjusted to pH = 5 [11]. In the present case, the
halide-to-metal charge transfer band is shifted to 535 nm, which
accounts for the intense egg-yellow colour. A nice superimposition
of the absorbance profile (filled triangles in Fig. 6b) with the % con-
2þ
tion of chloride, the concentration of the ½CuIIð4ÞꢁCl2ꢁ complex
2
tends to 100%, while the solution takes a pale blue-green colour.
The development and the geometrical modifications of the three
anion containing species over the course of the titration are picto-
rially illustrated in Fig. 5.
3þ
Thus, it appears that the intense yellow colour pertains specifi-
cally to the 1:1 inclusion complex ½CuI2Ið4ÞꢁClꢁ3þ. The origin of this
intense absorption at an unusually low energy has to be accounted
for. It has been mentioned before that in typical CuII(tetramine)Cl+
complexes the Clꢀ-to-CuII charge transfer transition takes place in
the UV region. However, it has been observed in the crystal struc-
centration curve of the ½CuI2Ið4ÞꢁBrꢁ inclusion complex is observed
also in the present case. The establishing of a significant interaction
between the encapsulated halide ion and the furan oxygen atoms,
documented by the short low Brꢄ ꢄ ꢄO distance, crystallographically
determined [11], has again to be invoked for justifying the low en-
ergy of the charge transfer band.
The question now is: does the intense yellow colour develop
also on titration with chloride and bromide of dicopper(II) cryp-
tates containing different spacers? Looking at the cryptate
½CuI2Ið2Þꢁ4þ, which possesses 1,3-xylyl spacers, the answer is not.
3þ
ture of the ½CuI2Ið4ÞꢁClꢁ complex that the distance between the
encapsulated chlorine atom and the furan oxygen atom is well be-
low the sum of average van der Waals radii, indicating interaction.
It is now suggested that such an interaction raises the energy of the
Fig. 5. Pictorial illustration of stepwise equilibria involving the three species: f½CuI2Ið4Þꢁ ꢄ ꢄ ꢄ Cl ꢄ ꢄ ꢄ ½CuI2Ið4Þꢁg7þ, ½CuI2Ið4ÞꢁXꢁ3þ, ½CuIIð4ÞꢁX2ꢁ2þ, which form over the course of the
2
4þ
titration of a MeCN solution of ½CuI2Ið4Þꢁ with Clꢀ. The largest sphere represents the halide ion.

V. Amendola et al. / Inorganica Chimica Acta 361 (2008) 4038–4046
4043
Fig. 6. (a) Spectra recorded over the course of the titration of a MeCN solution 5.3 ꢃ 10ꢀ4 M in ½Cu2IIð4Þꢁ4þ, with a standard solution of [Bu4N]Br. (b) Concentration profiles of
the complex species forming over the course of the spectrophotometric titration experiment. The profiles of the absorbance at 460 nm has been superimposed on the
diagram. On titration, the solution takes an egg-yellow colour.
In fact, on addition of either chloride or bromide, the pale blue
3.3. Iodide ion interaction with dicopper(II) bistren cryptates and with
a model [CuII(tren)]2+ complex
4þ
solution of ½CuI2Ið2Þꢁ turns pale blue-green, a behaviour which
can be accounted for by the spectrophotometric titration experi-
ment. As an example, Fig. 7a displays the family of spectra taken
over the course of the titration of an MeCN solution 5.02 ꢃ
4þ
Spectrophotometric titration of the ½CuI2Ið4Þꢁ receptor with io-
dide gave different results. Fig. 8a shows the family of spectra ob-
tained over the course of the titration of a MeCN solution
4þ
10ꢀ4 M of ½Cu2IIð2Þꢁ with a standard MeCN solution of [Bu4N]Br.
4þ
The Brꢀ-to-CuII charge transfer transition takes place at a much
higher energy than observed for the furan containing cryptate and
appears at 350–370 nm as a shoulder of the amine nitrogen-to-me-
6.15 ꢃ 10ꢀ4 M in ½Cu2IIð4Þꢁ with a standard solution of [Bu4N]I.
On iodide addition, two rather intense bands develop around
400 nm. These bands have a halide ion-to-copper(II) charge trans-
fer nature, but appear quite different from what observed on titra-
tal transition. The same three complex species observed for the
4þ
4þ
½CuI2Ið4Þꢁ
receptor form over the course of the titration:
tion of ½CuI2Ið4Þꢁ receptor with chloride and bromide. First, they
f½Cu2IIð2Þꢁ ꢄ ꢄ ꢄ Br ꢄ ꢄ ꢄ ½Cu2IIð2Þꢁg7þ, ½CuI2Ið2ÞꢁBrꢁ and ½CuIIð2ÞꢁBr2ꢁ2þ. The
show two well distinct transitions, at 370 and 440 nm; second,
they occur at an energy higher than expected; in fact, the Xꢀ-to-
CuII transition has been observed at 415 nm for chloride and at
435 nm for bromide. Thus, in view of the higher reducing tenden-
cies, such a band would be expected for iodide at 455 nm or more.
Moreover, best fitting of spectrophotometric data was obtained by
assuming the presence at the equilibrium of only two complex spe-
3þ
2
absorbance profile at 360 nm (filled triangles in Fig. 7c) does not
show any inflection or discontinuity, but keeps increasing as far
as the cryptate uptakes bromide ions. This indicates that the ha-
lide-to-copper(II) charge transfer transition is independent upon
the environment, is not affected by the cage framework and, in par-
ticular, is not perturbed by the spacers. In fact, the 1,3-xylyl frag-
ment cannot play any ‘active’ role.
7þ
cies: f½CuI2Ið4Þꢁ ꢄ ꢄ ꢄ I ꢄ ꢄ ꢄ ½CuI2Ið4Þꢁg and ½CuI2Ið4ÞIꢁ3þ, whose concentra-
Fig. 7. (a) Spectra recorded over the course of the titration of a MeCN solution 5.02 ꢃ 10ꢀ4 M in ½Cu2IIð4Þꢁ4þ, with a standard solution of [Bu4N]Br. (b) d–d spectum.
(c) Concentration profiles of the complex species forming over the course of the spectrophotometric titration experiment. The profiles of the absorbance at 360 nm has been
superimposed on the diagram. On titration, the solution takes a pale blue-green colour.

4044
V. Amendola et al. / Inorganica Chimica Acta 361 (2008) 4038–4046
Fig. 8. (a) Spectra recorded over the course of the titration of a MeCN solution 6.15 ꢃ 10ꢀ4 M in ½Cu2IIð4Þꢁ4þ, with a standard solution of [Bu4N]I. (b) Portion of d–d spectra.
(c) Lines: concentration profiles of the complex species forming over the course of the spectrophotometric titration experiment; symbols: absorbances at 440 and 700 nm
(right vertical axes). On titration, the solution takes a canary-yellow colour.
tion profiles are shown in Fig. 8c. The 1:1 iodide inclusion complex
tion of 4 equiv. of Iꢀ reaches ca. 80%; (iv) the complex of type
3þ
2þ
½CuI2Ið4ÞIꢁ is poorly stable and its concentration reaches only ca.
½CuIIð2ÞꢁX2ꢁ does not form.
2
80% even after the addition of a large excess of anion (up to
4 equiv.). The low stability of the 1:1 inclusion complex is quanti-
tatively expressed by the constant associated to Eq. (4), which is
65-fold lower than observed for bromide and 6000-fold lower than
for chloride. Notice that the complex of type ½CuI2Ið4ÞꢁX2]2+ does not
form, a behaviour which may be ascribed to the intrinsic weakness
of the CuII–Iꢀ coordinative bond and to the difficulty experienced
by the dimetallic cryptate to accommodate two rather large anions
in a sterically constrained arrangement.
The similarity of the results obtained for the two dinuclear
cryptate complexes suggests that the spectral behaviour is not re-
lated to the cage structure, but rather depends on each separated
CuII(tren)2+ subunit. This prompted us to investigate the binding
tendencies towards iodide of the copper(II) complex of tren deriv-
ative 5.
At this stage, it seemed useful to consider, for comparative pur-
4þ
poses, the recognition tendencies of the ½CuI2Ið2Þꢁ cryptate toward
NH
N
the iodide ion. Fig. 9a displays the spectra taken over the course of
4þ
the titration of an MeCN solution 5.00 ꢃ 10ꢀ4 M in ½Cu2IIð2Þꢁ with
[Bu4N]I. Results are very similar to those obtained with the
4þ
½CuI2Ið4Þꢁ receptor. In particular, (i) two rather intense bands cen-
NH
tred at ca. 370 and 440 nm are observed; (ii) the same complex
NH
7þ
species are formed on iodide titration: f½CuI2Ið2Þꢁ ꢄ ꢄ ꢄ I ꢄ ꢄ ꢄ ½Cu2IIð2Þꢁg
3þ
and ½CuI2Ið2ÞIꢁ3þ; (iii) the 1:1 anion inclusion complex ½CuI2Ið2ÞIꢁ
5
is not particularly stable (the logK value for Eq. (4), 4.82 0.04, is
very close to that observed for the ½CuI2Ið4Þꢁ4þ receptor) and on addi-
Fig. 9. (a) Spectra recorded over the course of the titration of a MeCN solution 5.00 ꢃ 10ꢀ4 M in ½Cu2 ð2Þꢁ4þ, with a standard solution of [Bu4N]I. (b) Portion of d–d spectra.
(c) Lines: concentration profiles of the complex species forming over the course of the spectrophotometric titration experiment; symbols: absorbances at 430 and 700 nm
(right vertical axes). On titration, the solution takes a canary-yellow colour.
II

V. Amendola et al. / Inorganica Chimica Acta 361 (2008) 4038–4046
4045
Fig. 10. (a) Spectra recorded over the course of the titration of a MeCN solution 3.15 ꢃ 10ꢀ4 M in [CuII(5)]2+, with a standard solution of [Bu4N]I. (b) Portion of d–d spectra.
(c) Lines: concentration profiles of the complex species forming over the course of the spectrophotometric titration experiment; symbols: absorbances at 450 and 800 nm
(right vertical axes). On titration, the solution takes a canary-yellow colour.
Fig. 10a shows the family of spectra obtained over the course of
the titration of a MeCN solution 3.15 ꢃ 10ꢀ4 M of [CuII(5)]2+ with a
standard solution of [Bu4N]I.
particular, we have shown that the dicopper(II) cryptate
4þ
½CuI2Ið4Þꢁ by means of its ‘active’ furan spacers allows specific rec-
ognition and colorimetric sensing of chloride and bromide ions.
The concept of the ‘active’ spacers can be profitably exploited in
the design of optical sensors for anions. Research work in this per-
spective is currently under way in our Laboratory.
Noticeably, on iodide addition, two bands develop at 370
4þ
and 440 nm, as observed for the ½CuI2Ið4Þꢁ receptor, and, less
4þ
evidently, with the ½CuI2Ið2Þꢁ
cryptate. Two iodide containing
species form over the course of the titration: the dimeric com-
plex {[CuII(5)]ꢄ ꢄ ꢄIꢄ ꢄ ꢄ[CuII(5)]}3+ and the five-coordinate complex
[CuII(5)I]+, whose concentration profiles are shown in Fig. 10c.
Thus, in the case of iodide, any cryptate effect vanishes for the
5. Supplementay material
CCDC 678004 contains the supplementary crystallographic data
for this paper. These data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/
data_request/cif.
4þ
½CuI2Ið4Þꢁ complex and the spectral response simply derives from
each individual CuII(tren)2+ fragment.
Moreover, an important feature to be pointed out is the high
stability of the 1:1 complex [CuII(5)I]+ in comparison with the
3þ
1:1 inclusion complexes ½CuI2Ið2ÞIꢁ and ½Cu2IIð4ÞIꢁ4þ. In particular,
Acknowledgement
the constant associated to Eq. (4) (logK = 5.2 0.1) is slightly, still
perceptibly higher than those associated to iodide inclusion into
the two investigated cryptates (equilibria of type (4)). Thus, the fol-
lowing paradox is observed: anion binding to a single CuII cation, in
the tren complex, is favoured with respect to the binding of two
CuII centres prepositioned within the bistren cage. However, this
apparent inconsistency can be accounted for by considering that
the inclusion of the large Iꢀ anion may induce a severe conforma-
tional rearrangement of the bistren cage, whose endergonicity
more than compensates the advantage of making two distinct
coordinative bonds. Moreover, it comes out that with iodide all
the special features related to the presence of the 2,5-dimethylfu-
ran spacers disappear. In particular, probably due to anion size and
the associated conformational rearrangement of the cage, which
may involve a change in the orientation of the furan spacer, any ha-
lide ion-furan oxygen atom interaction fades away, thus cancelling
the unique spectral properties.
The financial support of the Italian Ministry of University (FIRB
– Project RBNE019H9K; PRIN – Project: Supramolecular devices) is
gratefully acknowledged.
References
ˇ
[1] (a) A. Bianchi, K. Bowman-James, E. García-Espana (Eds.), Supramolecular
Chemistry of Anions, Wiley-VCH, New York, 1997;
(b) F.P. Schmidtchen, M. Berger, Chem. Rev. 97 (1997) 1609;
(c) T.S. Snowden, E.V. Anslyn, Curr. Opin. Chem. Biol. 3 (1999) 740;
(d) P. Beer, P.A. Gale, Angew. Chem., Int. Ed. 40 (2001) 486;
(e) V. Amendola, D. Esteban-Gomez, L. Fabbrizzi, M. Licchelli, Acc. Chem. Res.
39 (2006) 343;
(f) P.A. Gale, Acc. Chem. Res. 39 (2006) 465;
(g) J.L. Sessler, P.A. Gale, W.-S. Cho, Anion Receptor Chemistry, in: J.F. Stoddart
(Ed.), Monographs in Supramolecular Chemistry, Royal Society of Chemistry,
Cambridge, 2006;
(h) V. Amendola, M. Bonizzoni, D. Esteban-Gómez, L. Fabbrizzi, M. Licchelli, F.
Sancenón, A. Taglietti, Coord. Chem. Rev. 250 (2006) 1451;
(i) J.W. Steed, Chem. Commun. (2006) 2637.
[2] K. Bowman-James, Acc. Chem. Res. 38 (2005) 671.
[3] J.-M. Lehn, E. Sonveaux, A.K. Willard, J. Am. Chem. Soc. 100 (1978) 4914.
[4] J.-M. Lehn, S.H. Pine, E. Watanabe, A.K. Willard, J. Am. Chem. Soc. 99 (1977)
6766.
[5] L. Fabbrizzi, P. Pallavicini, A. Perotti, L. Parodi, A. Taglietti, Inorg. Chim. Acta
238 (1995) 5.
[6] C.J. Harding, F.E. Mabbs, E.J.L. MacInnes, V. McKee, J. Nelson, J. Chem. Soc.,
Dalton Trans. (1996) 3227.
[7] R. Menif, J. Reibenspies, A.E. Martell, Inorg. Chem. 30 (1991) 3446.
[8] L. Fabbrizzi, A. Leone, A. Taglietti, Angew. Chem., Int. Ed. 40 (2001) 3066.
[9] M. Boiocchi, M. Bonizzoni, L. Fabbrizzi, G. Piovani, A. Taglietti, Angew. Chem.,
Int. Ed. 116 (2004) 3935.
4. Conclusions
There exists a current interest in the development of colorimet-
ric molecular sensors for anions [20]. Most of them are constituted
by a specific receptor covalently linked to or integrated with an or-
ganic chromogenic subunit. This work has demonstrated that tran-
sition metal containing receptors can provide a powerful signal for
communicating anion recognition, i.e. the colour change associated
to the development of an anion-to-metal charge transfer band. In

4046
V. Amendola et al. / Inorganica Chimica Acta 361 (2008) 4038–4046
[10] M. Bonizzoni, L. Fabbrizzi, G. Piovani, A. Taglietti, Tetrahedron 60 (2004)
11159.
[16] A.C.T. North, D.C. Phillips, F.S. Mathews, Acta. Crystallogr., Sect. A 24 (1968)
351.
[11] V. Amendola, E. Bastianello, L. Fabbrizzi, C. Mangano, P. Pallavicini, A. Perotti,
A. Manotti Lanfredi, F. Ugozzoli, Angew. Chem., Int. Ed. Engl. 112 (2000) 3039.
[12] C.J. Harding, V. McKee, J. Nelson, Q. Lu, J. Chem. Soc., Chem. Commun. (1993)
1768.
[13] V. Amendola, L. Fabbrizzi, C. Mangano, P. Pallavicini, A. Poggi, A. Taglietti,
Coord. Chem. Rev. 219–221 (2001) 821.
[17] A. Altomare, M.C. Burla, M. Camalli, G.L. Cascarano, C. Giacovazzo, A.
Guagliardi, A.G.G. Moliterni, G. Polidori, R. Spagna, J. Appl. Crystallogr. 32
(1999) 115.
[18] G.M. Sheldrick, SHELX 97 Programs for Crystal Structure Analysis, University of
Göttingen, Germany, 1997.
[19] A. Bondi, J. Phys. Chem. 68 (1964) 441.
ˇ
[14] P. Gans, A. Sabatini, A. Vacca, Talanta 43 (1996) 1739. <http://www.hyperquad.
co.uk/index.ht>.
[20] (a) R. Martínez-Mánez, F. Sancenón, Chem. Rev. 103 (2003) 4419;
(b) C. Suksai, T. Tuntulani, Chem. Soc. Rev. 32 (2003) 192.
[15] L.J. Farrugia, J. Appl. Crystallogr. 32 (1999) 837.