Coordination chemistry of N-aminopropyl pendant arm derivatives of mixed N/S-, and N/S/O-donor macrocycles, and construction of selective fluorimetric chemosensors for heavy metal ions

Article abstract of DOI:10.1039/b506785e

The coordination chemistry of the N-aminopropyl pendant arm derivatives (L^1c-4c) of the mixed donor macrocyclic ligands [12]aneNS ₂O, [12]aneNS₃, [12]aneN₂SO, and [15]aneNS ₂O₂ (L^1a-4a) towards Cu^II, Zn ^II, Cd^II, Hg^II, and Pb^II in aqueous solution has been investigated. The protonation and stability constants with the aforementioned metal ions were determined potentiometrically and compared, where possible, with those of the unfunctionalised macrocycles. The measured values show that Hg^II and Cu^II in water have the highest affinity for all ligands considered, with the N-aminopropyl pendant arm weakly coordinating the metal centres. Crystals suitable for X-ray diffraction analysis were grown for the perchlorate salt (H₂L ^1c)(ClO₄)₂·dmf, and for the 1:1 complexes [Cd(L^3a)(NO₃)₂] (1), [Cu(L ^4a)(dmf)](ClO₄)₂ (2), [Zn(L ^1c)(ClO₄)]ClO₄ (3), [Cd(L^1c)(NO ₃)]NO₃ (4), and [Hg(L^2c)](ClO₄) ₂ (5). Their structures show the macrocyclic ligands adopting a folded conformation, which for the 12-membered systems can be either [2424] or [3333] depending on the nature of the metal ion. L^1c-4c were also functionalised at the primary amino pendant group with different fluorogenic subunits. In particular the N-dansylamidopropyl (L^nd, n = 1-4), and the N-(9-anthracenylmethyl)aminopropyl (L^ne, n = 1, 2, 4, Scheme 1) pendant arm derivatives of L^1a-4a were synthesised and their optical responses to the above mentioned metal ions were investigated in MeCN/H ₂O (4: 1 v/v) solutions. The Royal Society of Chemistry 2005.

Full text of DOI:10.1039/b506785e

View Article Online / Journal Homepage / Table of Contents for this issue
F U L L P A P E R
Coordination chemistry of N-aminopropyl pendant arm derivatives
of mixed N/S-, and N/S/O-donor macrocycles, and construction
of selective fluorimetric chemosensors for heavy metal ions†
M. Carla Aragoni,a Massimiliano Arca,a Andrea Bencini,b Alexander J. Blake,c
Claudia Caltagirone,a Antonello Decortes,a Francesco Demartin,d Francesco A. Devillanova,a
Enrico Faggi,b Luisa Stella Dolci,e Alessandra Garau,a Francesco Isaia,a Vito Lippolis,*a
Luca Prodi,e Claire Wilson,c Barbara Valtancolib and Nelsi Zaccheronie
a Dipartimento di Chimica Inorganica ed Analitica, Universita` degli Studi di Cagliari,
Complesso Universitario di Monserrato, S.S. 554 Bivio per Sestu, 09042, Monserrato, CA,
Italy. E-mail: lippolis@unica.it
b Dipartimento di Chimica, Universita` di Firenze, Polo Scientifico, Via della Lastruccia 3,
50019, Sesto Fiorentino, Florence, Italy
c School of Chemistry, The University of Nottingham, University Park, Nottingham,
UK NG7 2RD
d Dipartimento di Chimica Strutturale e Stereochimica Inorganica, Universita` degli Studi di
Milano, Via G. Venezian 21, 20133, Milano, Italy
e Dipartimento di Chimica “G. Ciamician”, Universita` di Bologna, Via Selmi 2, 40126, Bologna,
Italy
Received 13th May 2005, Accepted 30th June 2005
First published as an Advance Article on the web 26th July 2005
The coordination chemistry of the N-aminopropyl pendant arm derivatives (L1c–4c) of the mixed donor macrocyclic
ligands [12]aneNS2O, [12]aneNS3, [12]aneN2SO, and [15]aneNS2O2 (L1a–4a) towards CuII, ZnII, CdII, HgII, and PbII in
aqueous solution has been investigated. The protonation and stability constants with the aforementioned metal ions
were determined potentiometrically and compared, where possible, with those of the unfunctionalised macrocycles.
The measured values show that HgII and CuII in water have the highest affinity for all ligands considered, with the
N-aminopropyl pendant arm weakly coordinating the metal centres. Crystals suitable for X-ray diffraction analysis
were grown for the perchlorate salt (H2L1c)(ClO4)2·dmf, and for the 1 : 1 complexes [Cd(L3a)(NO3)2] (1),
[Cu(L4a)(dmf)](ClO4)2 (2), [Zn(L1c)(ClO4)]ClO4 (3), [Cd(L1c)(NO3)]NO3 (4), and [Hg(L2c)](ClO4)2 (5). Their structures
show the macrocyclic ligands adopting a folded conformation, which for the 12-membered systems can be either
[2424] or [3333] depending on the nature of the metal ion. L1c–4c were also functionalised at the primary amino
pendant group with different fluorogenic subunits. In particular the N-dansylamidopropyl (Lnd, n = 1–4), and the
N-(9-anthracenylmethyl)aminopropyl (Lne, n = 1, 2, 4, Scheme 1) pendant arm derivatives of L1a–4a were synthesised
and their optical responses to the above mentioned metal ions were investigated in MeCN/H2O (4 : 1 v/v) solutions.
optical signal expressed as an enhancement or quenching of the
Introduction
fluorophore emission. The sensor performance can be improved
The design of artificial molecular sensors for the selective
by either increasing the selectivity of the host–guest interaction
detection of different target species is a growing area of
chemistry. The intense research interest in this field is driven by
via an accurate design of the receptor unit, or by improving the
transduction mechanism leading to a change of the fluorescence
the great demand for extremely sensitive and selective analytical
emission of the sensor. In this respect, a useful approach implies
tools able to recognize, sense and evaluate at subnanomolar
a direct interaction of the fluorophore with the target species.
level the concentration of charged and neutral substrates in
By adopting a “receptor–spacer–fluorophore” model, many
biological and environmental samples and in industrial waste
selective and sensitive sensors and switches for metal ions feature
effluent samples.
In this respect, fluorescent chemosensors have been widely
chromophores as anthracene, 8-hydroxyquinoline, dansylamide,
or phenanthroline as signalling sites,1–9 either connected to
studied in the past decade because of the low detection limits
a macrocyclic ligand as coordinating sidearms or inserted
that can be achieved and relatively cheap equipment required.1–16
as a nonpendant integral part of a macrocyclic structure.
These compounds are usually made of a fluorogenic fragment
In fact, macrocyclic receptors represent the first choice as
(signalling unit) covalently linked to a receptor unit. With
receptor moieties for metal cations because of the extensive
such a supramolecular approach, the selective recognition of
possibilities they can offer for modulation of host coordination
the target species by the receptor unit is converted into an
environments, thereby allowing an easy route to achieving
selective interactions with the substrate of interest. Many of the
reported fluoroionophores feature polyoxa-,17,18 polyaza-,19–22
† Electronic supplementary information (ESI) available: Synthetic de-
tails including analytical and spectroscopic data for L1c–e, L2b–e, L3b–d
,
and aza/oxa-macrocycles23–26 as the guest binding site, whereas
the potential of mixed N/S-, and N/S/O-donating macrocyclic
sites as selective receptors in molecular sensors remains more or
less untapped.27–30
and the isolated metal complexes. Protonation constants (log K)
of L3a, L1c, L2c, L3c, and L4c (Table S1). Distribution diagrams for the
systems ZnII/L1c, and CdII/L1c (Fig. S1), ZnII/L2c (Fig. S2), CuII/L2c,
and HgII/L2c (Fig. S3), ZnII/L3c, and CdII/L3c (Fig. S4). Changes in the
UV-Vis spectrum of L4d (Fig. S5), and L2e (Fig. S6) upon addition of
increasing amounts of HgII. See http://dx.doi.org/10.1039/b506785e
In this context, following our interest in both coordina-
tion chemistry of mixed thia-aza macrocycles,31–34 and their
2 9 9 4
D a l t o n T r a n s . , 2 0 0 5 , 2 9 9 4 – 3 0 0 4
T h i s j o u r n a l i s
T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 5
©

View Article Online
Scheme 1
analytical applications as selective ionophores towards heavy
and transition metal ions, we describe here the coordination
The complexes [Cd(L3a)(NO3)2] (1), [Cu(L4a)(dmf)](ClO4)2
(2), [Zn(L1c)(ClO4)]ClO4 (3), [Cd(L1c)(NO3)]NO3 (4), and
[Hg(L2c)](ClO4)2 (5) were synthesized by following a standard
procedure which includes mixing of the appropriate metal salt
and the ligand in a 1 : 1 molar ratio in MeCN and the isolation
of the product as crystals from the reaction mixture stirred for
a few hours at room temperature by diffusion of Et2O vapor.
In the case of 2 crystals were obtained by crystallization from
dmf/Et2O of the product isolated from the reaction mixture by
diffusion of Et2O vapor. Synthetic details including analytical
data have been deposited as ESI.†
properties of the N-aminopropyl pendant arm derivatives (L1c–4c
)
of the NS2O-, NS3-, N2SO-, and NS2O2-donating macrocy-
cles L1a–4a (Scheme 1) towards CuII, ZnII, CdII, HgII, and
PbII. The N-dansylamidopropyl (Lnd, n = 1–4), and the N-
(9-anthracenylmethyl)aminopropyl (Lne, n = 1, 2, 4) pendant
arm derivatives of L1a–4a were also synthesised and their optical
response to the above mentioned metal ions investigated in
MeCN/H2O (4 : 1 v/v) solutions. Although in these systems the
fluorescent moiety is remote from the cavity of the macrocyclic
receptor, the linker heteroatom of the fluorophore (i.e. the amine
or amide nitrogen atom of the pendant arm/s) can participate
in complexation to the metal ion, thereby possibly improving
the signal transduction mechanism and leading to a reliable
substrate-specific response by the sensor.
Caution! Most of the metal complexes synthesised were
−
isolated in the solid state as ClO4 salts. We worked with these
complexes on a small scale without any explosive incident.
Despite these observations, the unpredictable behaviour of
−
ClO4 salts necessitate extreme care in handling.
Experimental
Synthesis of L4b. A mixture of L4a (4.4 g, 17.5 mmol) and
acrylonitrile (31.5 mL) was refluxed for 24 h under N2. After
cooling, the excess acrylonitrile was removed under reduced
pressure. The crude product was purified by flash silica gel
chromatography using CH2Cl2 : MeOH (90 : 10 v/v) as eluant.
A white solid was obtained (2.7 g, 8.87 mmol, 51% yield). Mp:
40–42 ◦C. Elem. Anal.: found (calc. for C13H24N2S2O2): C, 51.10
(51.28); H, 8.04 (7.95); N, 9.55 (9.20); S, 21.21 (21.06). 1H-NMR
(400 MHz, 298 K, CDCl3): dH 2.44 (2H, t, J = 7.2 Hz), 2.70–
2.83 (14H, m), 3.60 (4H, s), 3.74 (4H, t, J = 5.2 Hz). 13C-NMR
(400 MHz, 298 K, CDCl3): dC 16.90, 30.19, 31.18, 50.89, 54.36,
70.62, 73.57, 118.72.
All melting points are uncorrected. Microanalytical data were
obtained using a Fison EA CHNS-O instrument operating at
1000 ◦C. ESI mass spectra were recorded at the School of
Chemistry of The University of Nottingham (UK). 1H and 13
C
NMR spectra were recorded on a Varian VXR300 or VXR400
spectrometer. Absorption spectra were recorded with a Varian
Model Cary 5 UV-Vis-NIR spectrophotometer. Uncorrected
emission, and corrected excitation spectra were obtained with a
Varian Cary Eclipse Fluorescence spectrophotometer. In order
to allow comparison among emission intensities, we performed
corrections for instrumental response, inner filter effect, and
phototube sensitivity.35 A correction for differences in the
refraction index was introduced when necessary. Luminescence
quantum yields (uncertainty 15%) were determined using
quinine sulfate in 1 M H2SO4 aqueous solution (U = 0.546) as a
reference. MeCN (Uvasol, Merck) and Millipore grade water
were used as solvents for spectrophotometric measurements.
Solutions of the ligands were 2.5 × 10−5 M. Other solvents
and starting materials were purchased from commercial sources
where available.
Synthesis of L4c. A mixture of L4b (2.54 g, 8.34 mmol) and
BH3·THF 1 M (250 mL) was refluxed for 48 h under N2.
After cooling, water (150 mL) was added and the solvent was
removed under reduced pressure. The residue was taken up in
2 M HCl (150 mL) and the resulting mixture was refluxed for 2 h.
The solvent was then removed under reduced pressure and the
resulting white solid was dissolved in water and passed through
a Dowex 1 × 8–200 ion-exchange resin. A colourless oil was
obtained (2.23 g, 7.25 mmol, 87% yield). 1H-NMR (400 MHz,
298 K, CDCl3): dH 1.55–1.58 (2H, m), 2.44 (2H, t, J = 6.8 Hz),
2.60–2.73 (10H, m), 3.32 (2H, br s), 3.53–3.55 (10H, m), 3.67
(2H, t, J = 5.4 Hz). 13C-NMR (300 MHz, 298 K, CDCl3):
dC 29.80, 31.08, 40.27, 53.07, 54.13, 62.10, 70.63, 73.37.
Synthesis of ligands and complexes
L1a–L4a, and L1b were synthesised according to procedures
reported in the literature.32,36,37 Here we report the procedure
adopted for the synthesis of L4b–e (see below). The same
procedure has been adopted for the synthesis of L1c–e, L2b–e
,
Synthesis of L4d. A solution of dansyl chloride (0.19 g,
0.713 mmol) in anhydrous MeCN (25 ml) was added drop-wise
to a mixture of L4c (0.2 g, 0.648 mmol) and K2CO3 (0.29 g,
and L3b–d, for which spectroscopic and microanalytical data are
supplied within the ESI.†
D a l t o n T r a n s . , 2 0 0 5 , 2 9 9 4 – 3 0 0 4
2 9 9 5

View Article Online
2.139 mmol) in anhydrous MeCN (25 ml). This mixture was
stirred at room temperature for 48 h under N2. After filtration,
the solvent was removed under reduced pressure. The residue
(a yellow oil) was dissolved in CHCl3 and washed with water.
The organic extracts were dried over Na2SO4, filtered and the
solvent was removed under reduced pressure to yield a yellow oil.
The product was purified by flash chromatography on silica gel
using CH2Cl2 : MeOH (90 : 10 v/v) as eluant to afford a yellow
solid (0.27 g, 0.5 mmol, 77% yield). Mp: 88–89 ◦C. Elem. Anal.:
found (calc. for C25H39N3S3O4): C, 55.54 (55.42); H, 7.88 (7.26);
N, 7.60 (7.76); S, 17.99 (17.75).1H NMR (400 MHz, CDCl3):
dH 1.50–1.52 (2H, m), 2.38 (2H, t, J = 5.6 Hz), 2.58–2.73 (12H,
m), 2.84 (6H, s), 2.93–2.96 (2H, m), 3.59 (4H, s), 3.70–3.72 (4H,
m), 6.59 (1H, br s), 7.14 (1H, d, J = 7.6 Hz), 7.44–7.55 (2H, m),
8.19 (1H, d, J = 7.6 Hz), 8.31 (1H, d, J = 8.4 Hz), 8.47 (1H,
d, J = 8.4 Hz). 13C NMR (400 MHz, CDCl3): dC 25.53, 29.26,
31.06, 43.04, 45.24, 53.66, 54.07, 70.61, 73.43, 114.93, 118.88,
122.97, 128.02, 129.28, 129.50, 129.65, 129.92, 134.80, 151.67.
Mass Spectrum EI+: m/z 542 ([C25H39N3S3O4]+).
Synthesis of L4e. 9-anthraldehyde (0.21 g, 1.0 mmol) was
added in portions to a solution of L4c (0.31 g, 1.0 mmol) in dry
EtOH : MeCN (1 : 1 v/v, 24 mL). The mixture was stirred at
room temperature for 24 h under N2. The solvent was removed
under reduced pressure. The residue was dissolved in dry EtOH
(50 mL) and NaBH4 (1.44 g, 38 mmol) and NaBH3CN (2.39 g,
38 mmol) were added to the resulting solution. The mixture
obtained was stirred at room temperature for 48 h under N2. The
solvent was removed under reduced pressure and the residue
was taken up in water and basified to pH = 14 with 5 M
NaOH. The product was extracted with CH2Cl2, the organic
phases were dried over Na2SO4, filtered and the solvent removed
under reduced pressure. The crude product was purified by flash
silica gel chromatography using CH2Cl2 : MeOH (90 : 10 v/v)
as eluant. A yellow solid was obtained (0.199 g, 0.40 mmol, 40%
yield). Elem. Anal.: found (calc. for C28H38N2S2O2): C, 66.99
(67.43); H, 7.91 (7.68); N, 5.40 (5.62); S, 12.83 (12.86).1H NMR
(400 MHz, CDCl3): dH 2.36–2.88 (20H, m), 3.37–3.50 (4H, m),
3.59–3.62 (2H, m), 4.12 (1H, br s), 4.72 (2H, s), 7.38–7.52 (4H,
m), 7.92 (2H, d, J = 8.4 Hz), 8.26 (2H, d, J = 8.4 Hz), 8.33
(1H, s). 13C NMR (300 MHz, CDCl3): dC 28.30, 29.32, 31.04,
44.93, 48.64, 53.68, 53.73, 70.64, 73.31, 123.85, 124.98, 126.46,
127.43, 129.60, 130.12, 131.33. Mass Spectrum EI+: m/z 499
([C28H38N2S2O2]+).
Potentiometric measurements
All pH-metric measurements were carried out in degassed 0.1 M
NMe4NO3 water solutions, at 298.1 K, using the equipment
and procedure described elsewhere.38 The combined Ingold 405
S7/120 electrode was calibrated as a hydrogen concentration
probe by titrating amounts of HCl with CO2-free NMe4OH so-
lutions and determining the equivalent point by Gran’s method39
which allows the determination of the standard potential E◦,
and the ionic product of water (pKw = 13.83(1) at 298.1 K
in 0.1 M NMe4NO3). Ligand and metal ion concentrations of
1.0 × 10−3 M and 0.5 × 10−3 to 1.0 × 10−3 M, respectively, were
employed in the potentiometric measurements, which involved
performing three titration experiments, each of ca. 100 data
points, in the pH range 2–10.5. The relevant e.m.f. data were
analysed by means of the computer program HYPERQUAD.40
Crystallography
Crystal data and refinement details of all structure deter-
minations appear in Table 1. Only special features of the
analyses are mentioned here. Single crystal data collection for
(H2L1c)(ClO4)2·dmf, [Cd(L3a)(NO3)2] (1), [Cu(L4a)(dmf)](ClO4)2
(2), [Zn(L1c)(ClO4)]ClO4 (3), and [Hg(L2c)](ClO4)2 (5) was per-
formed on a Bruker SMART CCD diffractometer using x
scans. For [Cd(L1c)(NO3)]NO3 (4) data were acquired on a
2 9 9 6
D a l t o n T r a n s . , 2 0 0 5 , 2 9 9 4 – 3 0 0 4

View Article Online
Enraf-Nonius CAD-4 diffractometer using x scans. All data
sets were corrected for Lorentz, polarization and absorption
effects as specified in Table 1. All structures were solved
by direct methods using SHELXS9742 followed by difference
Fourier syntheses and refined with full-matrix least squares using
SHELXL97.43 All non-H atoms were refined anisotropically and
H atoms were introduced at calculated positions and thereafter
incorporated into a riding model with Uiso(H) = 1.2Ueq(C). In 2,
3, and 5 some of the oxygen atoms in perchlorato anions display
slightly high displacement parameters which might be indicative
of partial disorder. However, any attempt to split these atoms
into two components offered no advantage.
CCDC reference numbers 271763–271768.
See http://dx.doi.org/10.1039/b506785e for crystallographic
data in CIF or other electronic format.
Results and discussion
Synthesis of Lnc (n = 1–4) and their coordination properties
towards CuII, ZnII, CdII, HgII, and PbII
Mixed O/S-, N/S-, and N/O/S-donor macrocycles feature hard
and soft binding sites in close proximity and, therefore, should
ensure strong complexation to a great variety of metal ions.
Compared to homoleptic thioether and polyaza macrocycles,
the chemistry of these ligands, especially those containing
three different donors, has been relatively neglected due to
synthetic difficulties encountered in their preparation.36,37,44–48
Recently, Reedijk et al. have developed a facile route for the
synthesis of a range of mixed N/O/S-donor macrocycles which
offers the advantage of an easier deprotonation step of the
nitrogen atom after the cyclization reaction and a consequently
higher overall yield.36 This has prompted us to study mixed
N/S-, and N/O/S-donor macrocycles as receptors in redox
and fluorescence molecular sensors for heavy and transition
metal (HTM) ions.32,33 We surmised that a diversified donor
set, featuring both hard and soft donor atoms, would confer
selectivity to the recognition of HTM ions by the receptor, thus
leading to a specific substrate-response of the signalling unit
in a molecular sensor. According to the modular approach
outlined in the introduction for the design and construction of
fluorescence sensors, we linked the macrocycles L1a–4a to dansyl
or 9-anthracenylmethyl fluorescent moieties via aminopropyl
spacers.
Fig. 1 Partial view of the packing diagram for (H2L1a)(ClO4)2·dmf
with the numbering scheme adopted. i = x, 3/2 − y, z; ii = 1 − x,
1
2
−
+ y, −z.
hydrogen bonds [C(2iii)H(2biii) · · · O(12) 2.47, C(2iii) · · · O(12)
◦
iii
ii
˚
3.383(6) A, C(2 )–H(2b ) · · · O(12) 157 , iii = −1/2 + x, 3/2 −
y, 1/2 − z] to afford an extended 3D-network.
Solution studies. Before proceeding with the synthesis of
the target fluorescent sensors (Scheme 1) we investigated the
protonation of L1c–4c and complex formation with CuII, ZnII,
CdII, HgII, and PbII by means of potentiometric measurements
in aqueous solutions (298.1 K, 0.1 M NMe4NO3) in the pH
range 2.0–10.5.
The protonation constants of the ligands are in the range
generally observed for polyamine compounds and are supplied
within ESI (Table S1).† The complexes formed and the corre-
sponding stability constants are reported in Table 2, together
with those previously reported for the complexes with L1a,32
and those determined for the complexes with L3a. Precipitation
at alkaline pHs of hydroxo complexes precluded the study of
some systems in the alkaline pH region (Table 2), but since the
process of complex formation generally takes place at acidic pH
values, the relevant stability constants for the [ML]2+ species
were determined in all cases, with the only exceptions of the
complexes ofPbII complexes withL1c andL2c andof CdII with L4c.
In fact, in these cases the low solubility of the complexes from
acidic to alkaline pHs did not allow a speciation study by using
potentiometric techniques.
Functionalisation of L1a–4a to their N-aminopropyl pendant
arm derivatives L1c–4c was achieved via reaction of L1a–4a with
acrylonitrile and subsequent reduction of the isolated pendant
arm nitrile derivatives, L1b–4b, with BH3·THF.
Single crystals of (H2L1c)(ClO4)2·dmf were obtained by dif-
fusion of Et2O vapour into a dmf solution of the crude
perchlorate salt of L1c obtained by addition of concentrated
HClO4 into a solution of L1c in EtOH. An X-ray diffraction
analysis (Fig. 1) reveals that the macrocyclic framework adopts a
“butterfly” [2424] conformation with the plane of the ring folded
along the S · · · S vector. This conformation is presumably de-
termined by the intramolecular N(1)H · · · O(7) hydrogen bond
The stability of the complexes with L3a, which contains an
N2SO-donor set within its cyclic framework, increases in the
order ZnII < PbII < CdII < CuII < HgII. HgII and CuII form
the most stable complexes, as generally observed in the case of
polyamine ligands. The lower stability of the ZnII complex with
respect to the CdII and PbII ones is rather unusual for polyamine
ligands, which generally show a similar binding ability for
these three metal cations. This result could be explained by the
presence in the donor set of a soft sulfur atom, which generally
displays a higher affinity for the softer CdII and PbII metal ions.
A similar trend has also been observed in the case of other
mixed N/O/S-donor tetradentate 12-membered macrocycles
(see the stability constants of the metal complexes with L1a
and L2a in Table 2).32,33
˚
[N(1)H · · · O(7) 2.05, N(1) · · · O(7) 2.909(5) A, N(1)–H · · · O(7)
141◦]. The protonated N-aminopropyl pendant arm points out
+
of the ring cavity, and the terminal-NH3 ammonium group
forms three hydrogen bonds, one of which is with a molecule
˚
of dmf [N(16)H(16a) · · · O(2) 1.77, N(16) · · · O(2) 2.703(6) A,
N(16)–H(16a) · · · O(2) 179◦].
Symmetry related (H2L1c)2+ cations are bridged by two
perchlorate ions via CH · · · O [C(3ii)H(3bii) · · · O(13) 2.51,
◦
ii
ii
ii
˚
C(3 ) · · · O(13) 3.270(6) A, C(3 )–H(3b ) · · · O(13) 135 , ii =
1 − x, −1/2 + y, −z] and NH · · · O hydrogen bonds at
the terminal ammonium groups [N(16)H(16b/c) · · · O(13) 2.23,
◦
˚
N(16) · · · O(13) 2.995(6) A, N(16)–H(16b/c) · · · O(13) 140 ] thus
forming undulating ribbons which propagate along the [010]
direction. Ribbons of this type are connected via CH · · · O
L1c and L2c contain an aminopropyl side arm attached
respectively to a [12]aneNS2O or [12]aneNS3 cyclic unit. Their
D a l t o n T r a n s . , 2 0 0 5 , 2 9 9 4 – 3 0 0 4
2 9 9 7

View Article Online
Table 2 Formation constants (log K) of the metal complexes with L1a–3a, L1c–4c (I = 0.1 M, 298.1 K)a
Reaction
CuII
ZnII
CdII
PbII
HgII
8.65(3)b
M2+ + L1a ꢀ [ML1a]2+
8.13b
—
—
7.90b
4.20b
b
M2+ + L2a ꢀ [ML2a]2+
7.93(2)c
9.81(6)
6.7(1)
9.67(1)
5.25(3)
6.11(5)
2.91(6)
9.23(3)
6.15(3)
4.28(5)
12.9(1)
7.6(1)
5.3(1)
4.4(1)
6.9(1)f
—
5.8(1)c
8.10(5)
4.27(4)c
—
c,d
c,e
M2+ + L3a ꢀ [ML3a]2+
5.10(5)
6.93(5)g
10.1(1)g
3.7(1)
9.62(3)
7.09(4)
2.93(5)
—
[ML3a]2+ + OH− ꢀ [ML3a(OH)]+
—
—
—
—
f
f
f
M2+ + L1c ꢀ [ML1c]2+
8.95(2)
6.68(2)
6.74(6)
—
8.14(4)
6.68(2)
6.63(2)
7.5(1)
8.3(1)
6.5(1)
4.3(1)
5.4(1)f
—
8.08(3)
5.02(8)
2.98(8)
—
h
[ML1c]2+ + H+ꢀ [M(HL1c)]3+
[ML1c]2+ + OH− ꢀ [ML1c(OH)]+
[ML1c(OH)]+ + OH− ꢀ [ML1c(OH)2]
M2+ + L2c ꢀ [ML2c]2+
—
—
—
8.20(3)f
7.22(4)
—
—
10.88(4)
6.05(5)
2.88(4)
11.8(1)
8.1(1)
6.7(1)
3.7(1)
7.8(1)f
7.3(1)
h
[ML2c]2+ + H+ꢀ [M(HL2c)]3+
[ML2c]2+ + OH− ꢀ [ML2c(OH)]+
M2+ + L3c ꢀ [ML3c]2+
—
—
9.0(1)
8.7(1)
6.3(1)
4.3(1)
9.2(1)
8.7(1)
5.3(1)
[ML3c]2+ + H+ꢀ [M(HL3c)]3+
[M(HL3c)]3+ + H+ꢀ [M(H2L3c)]4+
[ML3c]2+ + OH− ꢀ [ML3c(OH)]+
M2+ + L4c ꢀ [ML4c]2+
f
—
i
—
6.0(1)f
—
[ML4c]2+ + H+ꢀ [M(HL4c)]3+
—
a For L4a it was not possible to study the complexation with the metal ions under consideration due to the impossibility to prepare a weighable salt of
this macrocycle. b Values from reference 32. c Data for the complexation of L2a were provided by M. Schro¨der et al. as a private communication. d The
stability of the complex is too low to be confidently determined. e Precipitation of the HgII complex does not allow the determination of the stability
of this complex. f Precipitation at slightly alkaline pH values of hydroxo complexes does not allow the study of the system in the alkaline pH region.
g The Log K value for the PbII and HgII complexes are in good agreement with those previously found by Afshar et al. (Log K = 6.6 for [PbL3a]2+ and
Log K = 10.5 for [HgL3a]2+, see reference 37). h The low solubility of the PbII complexes with L1c and L2c does not allow the potentiometric study of
these systems. i The low solubility of the CdII complexes with L4c does not allow the potentiometric study of this system.
complexation to CuII, ZnII, CdII and HgII occurs at acidic pH
values, with the formation of [M(HL)]3+ species (M = CuII, ZnII,
CdII or HgII, L = L1c or L2c), followed by deprotonation to
afford [ML]2+ complexes. Deprotonation of a coordinated water
molecule takes place at neutral (ZnII) or alkaline (CuII and HgII)
pH values to give hydroxo complexes, as shown in Fig. 2 for the
CuII and HgII complexes with L1c (the distribution diagrams for
the other complexes with L1c and L2c are supplied within ESI,
Figs. S1, S2 and S3).† CdII hydroxo-complexes with L2c were not
detected in solution, due to their low solubility.
than L1a and L2a, indicating that in these species the amine group
of the pendant arm is indeed involved in metal coordination.
Furthermore, in contrast to the complexes with L1a and L2a, the
complexes [ML1c]2+ and [ML2c]2+ display a marked tendency to
bind a proton to afford monoprotonated [M(HL)]3+species, as
shown by the rather high values (Table 2) of the protonation
constants of the [ML]2+ complexes (L = L1c or L2c).
This suggests that protonation of these complexes takes place
on the amine group of the pendant arm, with a consequent
detachment of this nitrogen donor from the metal.
We were interested in comparing the coordination properties
of L1c and L2c with those of the corresponding macrocycles not
bearing an N-aminopropyl side arm, L1a and L2a, respectively. As
shown in Table 2, L1c and L2c form more stable [ML]2+ complexes
L3c contains two aminopropyl pendant arms attached to
a [12]aneN2SO moiety. Among the ligands investigated, L3c
forms the most stable complexes (Table 2), in agreement with
the higher number of N-donors potentially available for metal
coordination. Analogously to L1c and L2c, the presence of
aminopropyl side arms gives rise to an increased stability of the
[ML3c]2+ complexes with respect to the corresponding complexes
with the underivatised macrocycle L3a. This suggests that at least
one aminoalkyl pendant arm is involved in metal coordination.
At the same time, all metal complexes show a marked tendency
to form mono- and di-protonated complexes, as shown in Fig. 3
for CuII and HgII (distribution diagrams for the systems ZnII/L3c
and CdII/L3c are supplied within ESI, Fig. S4).†
The fact that complexes with L3a do not show any tendency to
bind protons indicates that protonation of [ML3c]2+ takes place
on the primary amine group of the side arms. Furthermore,
the constant for the addition of the first proton to the [ML3c]2+
complexes ([ML3c]2+ + H+ ꢀ [M(HL3c)]3+) ranges between 7.6
and 8.7 log units; these high values are generally attributed to
protonation of amine groups not bound to metal cations.49 The
second protonation constant is lower than the first one (5.3–
6.7 log units), as expected for protonation of an aminopropyl
group weakly involved in metal binding.49 Therefore, in the
[ML3c]2+ complexes only one aminopropyl side arm would be
bound to the metal, while the other one would not be involved
in any interaction with the metal. Consequently, binding of
the first proton to the [ML3c]2+ complexes, observed at slightly
alkaline pH values, takes place on the aminopropyl group
not coordinated to the metal, while the second protonation
step ([M(HL3c)]3+ + H+ ꢀ [M(H2L3c)]4+), at acidic pH values,
occurs on the metal-bound aminopropyl side arm and causes its
displacement from the metal.
Fig. 2 Distribution diagrams for the systems CuII/L1c (a), and HgII/L1c
(b) ([MII] = [L1c] = 1 × 10−3 M, T = 298.1 K, I = 0.1 M).
2 9 9 8
D a l t o n T r a n s . , 2 0 0 5 , 2 9 9 4 – 3 0 0 4

View Article Online
diffraction were obtained by diffusion of Et2O vapour. A
single-crystal structure analysis confirmed the complex to be
the species [Cd(L3a)(NO3)2] (1), the asymmetric unit consisting
of two complexed macrocyclic molecules (1-I and 1-II). Due
to the similarity of the two independent molecules in unit
cell, only one (1-I) will be discussed (Fig. 4, Table 3). The
coordination sphere around each metal centre consists of
the N2SO-donor set of the macrocyclic framework, and two
asymmetrically bidentate nitrato ligands (in 1-I, for one of
−
the NO3 ions, one Cd–Onitrato distance is only slightly longer
than those found for a clearly bidentate nitrato ligand (Fig. 4,
Table 3),44 whereas in 1-II this happens for both coordinated
−
NO3 ions with short Cd–Onitrato distances of 2.308(4) and
˚
˚
2.325(4) A, and longer ones of 2.704(5) and 2.678(4) A). The
CdII ion is displaced out of the plane defined by the four donors
of L3a (by 1.0 A for both independent molecules 1-I and 1-II)
˚
and the macrocylic ligand adopts a [3333] conformation with
N(1)–Cd–N(7) and S(4)–Cd–O(10) angles for 1-I of 114.8(2)
˚
˚
[113.2(2) for 1-II] and 123.98(10) A [122.63(10) A for 1-II],
respectively. A similar coordination sphere around the metal
centres, and conformation for the macrocyclic ligands has been
observed in the complexes [Hg(L3a)(NO3)2],37 [Pb(L3a)NO3)2],37
[Hg([12]aneNSO2)(NO3)2] ([12]aneNSO2 = 1,7-dioxa-4-thia-
10-azacyclododecane),45 [Cd(cis-[12]aneN2S2)(NO3)2] (cis-
[12]aneN2S2
=
1,4-dithia-7,10-diazacyclododecane),44 and
[Cd(L2a)(NO3)2],44 thus confirming the inability of 12-membered
mixed-donor tetradentate macrocycles to bind a metal in the
plane of the donors, and their preference for a folded [2424]
conformation with small metal ions such as CuII and ZnII, or a
more planar [3333] conformation with larger metals such as CdII,
HgII, and PbII that lack a particular stereochemical preference.
Fig. 3 Distribution diagrams for the systems CuII/L3c (a) and HgII/L3c
(b) ([MII] = [L3c] = 1 × 10−3 M, T = 298.1 K, I = 0.1 M).
Among the N-aminopropyl pendant arm functionalised
macrocycles under investigation, L4c displays the lowest binding
ability for metal cations (Table 2). A tentative explanation of
the lower stability of the complexes with L4c may involve the
different molecular architecture of this ligand, which contains
one nitrogen donor and four weakly coordinating sulfur or
oxygen atoms within its 15-membered cyclic framework. Such
a disposition could leave a sulfur or oxygen donor excluded
from coordination to the metal in the [ML4c]2+ complexes,
thus generating a destabilising 8-membered chelate ring. At the
same time, in contrast with the other macrocycles bearing an
aminopropyl pendant arm reported here, the [ML4c]2+ complexes
(M = CuII, ZnII or PbII) do not display a particular tendency
to form protonated species. Presumably, because of their low
stability, protonation of the [ML4c]2+ species on the metal-
bound aminopropyl pendant arm (and consequent detachment
from the metal) would lead to decomplexation rather than
to the formation of a protonated [M(HL4c)]3+ species. As a
matter of fact, only the most stable complex, [HgL4c]2+, gives
a monoprotonated species, [Hg(HL4c)]3+.
Solid state studies. A feature of 12-membered tetradentate
macrocycles is their inability to host divalent transition metal
ions within their cavity: it is not possible to achieve coplanarity
between the four donor atoms and the guest metal. This is well
documented for ligands with four identical donors, whereas
structural data for mixed tetradentate N/S- and N/S/O-
donor 12-membered macrocycles are sparse.37,44,45,50 Recently, we
reported the X-ray crystal structures for the 1 : 1 metal complexes
of L1a with CuII, CdII, and HgII.32 L1a acts a tetradentate ligand
and adopts folded [2424] or [3333] conformations resembling an
open book with the spine along the S(4)–M–S(10) direction and
N(1)–M–O(7) angles ranging from 81.50(12)◦ in [Cd(L1a)(NO3)2]
to 111.2(4)◦ in [Hg(L1a)MeCN]+. In each case anionic ligands
or solvent molecules complete the coordination sphere around
the metal ion. Similar folded conformations are adopted by
[12]aneN3S (1-thia-4,7,10-triazacyclododecane) in its complexes
with CuII, ZnII, HgII, PbII, and NiII, with the more planar
[3333] conformer being dominant for larger metal ions (HgII
and PbII).50
Fig.
4
ORTEP view of one (1-I) of the two independent
[Cd(L3a)(NO3)2] units in 1 with the adopted numbering scheme. Dis-
placement ellipsoids are drawn at 30% probability. Hydrogen atoms
have been omitted for clarity except those on the macrocyclic N-donors.
Even fewer structural data are available for metal complexes
with 15-membered mixed N/S/O-donor pentadentate macro-
cycles such as L4a than for the smaller 12-membered macrocyclic
ligands.44,47,51 The greater flexibility and the larger cavity of these
systems allows very different conformations to be adopted
from a “butterfly” or folded geometry in [Cd(L4a)(NO3)2],44
[Ni([15]aneN2S2O)(H2O)](NO3)2,47 [Ag([15]aneN2S2O)(SCN)],51
([15]aneN2S2O = 1-oxa-7,10-dithia-4,13-diazacyclopentadecane),
[Ni([15]ꢀaneN2S2O)(NO3)]NO3,47 and [Cu([15]ꢀaneN2S2O)]-
(ClO4)2 ([15]ꢀaneN2S2O = 1-oxa-4,13-dithia-7,10-diazacyclo-
pentadecane),47 to a flattened geometry in [Cd([15]aneN2SO2)-
(NO3)]NO3 ([15]aneN2SO2
=
1,4-dioxa-10-thia-7,13-diaza-
Reaction of L3a with Cd(NO3)2 in MeCN afforded a
colourless solution from which crystals suitable for X-ray
cyclopentadecane)44 with the metal centres encapsulated within
the ring cavity.
D a l t o n T r a n s . , 2 0 0 5 , 2 9 9 4 – 3 0 0 4
2 9 9 9

View Article Online
◦
3a
4a
1c
1c
˚
Table 3 Selected bond distances (A) and angles ( ) for [Cd(L )(NO3)2] (1), [Cu(L )(dmf)](ClO4)2 (2), [Zn(L )(ClO4)]ClO4 (3), [Cd(L )(NO3)]NO3
(4), and [Hg(L2c)](ClO4)2 (5)a
1-Ib
2
3
4d
5f
M–N(1)
M–S(4)
M–X(7)
M–S(10)
2.303(4)
2.710(2)
2.311(5)
1.979(3)
2.4038(10)
2.347(2)
2.150(5)
2.486(4)
2.195(4)
2.466(3)
2.441(3)
2.6681(10)
2.666(2)
2.500(9) [2.505(9)]
2.772(3) [2.737(3)]
2.544(3) [2.563(3)]
2.697(3) [2.657(3)]
2.5970(9)
M–O(10)
M–S(13)
M–Y(1)
M–Y(2)
2.511(4)
2.433(3)
2.4069(10)
1.925(2)
2.409(4) [2.522(4)]
2.297(4) [2.778(5)]
78.88(13)
2.030(5)
2.362(7)
90.3(4)
97.5(2)
84.3(4)
2.239(3)
2.334(2) [2.838(3)]
76.94(7)
94.37(8)
80.22(6)
2.190(9) [2.169(8)]
N(1)–M–S(4)
N(1)–M–X(7)
N(1)–M–S(10)
N(1)–M–O(10)
N(1)–M–S(13)
N(1)–M–Y(1)
N(1)–M–Y(2)
S(4)–M–X(7)
S(4)–M–S(10)
S(4)–M–O(10)
S(4)–M–S(13)
S(4)–M–Y(1)
S(4)–M–Y(2)
X(7)–M–S(10)
X(7)–M–O(10)
X(7)–M–S(13)
X(7)–M–Y(1)
X(7)–M–Y(2)
O(10)–M–S(13)
O(10)–M–Y(1)
O(10)–M–Y(2)
Y(1)–M–S(10)
Y(1)–M–S(13)
Y(2)–M–S(10)
Y(1)–M–Y(2)
86.79(9)
77.5(2) [76.0(2)]
126.7(2) [129.2(2)]
77.6(2) [79.0(2)]
114.8(2)
95.17(10)
72.4(2)
93.75(10)
88.67(8)
176.97(11)
90.73(15) [137.26(15)]
122.9(2) [73.9(2)]
77.93(13)
102.5(2)
174.6(5)
79.1(3)
90.20(10)
159.37(9) [152.02(8)]
70.23(5)
94.0(3) [95.8(3)]
78.35(6)
80.33(10) [82.22(10)]
132.63(9) [132.16(10)]
156.61(7)
133.23(3)
123.98(10)
146.55(6)
134.58(3)
93.52(8)
153.83(11) [143.83(11)]
86.76(12) [75.59(10)]
101.2(4)
91.7(8)
79.1(3)
103.88(7)
82.47(7) [127.06(6)]
71.50(5)
105.0(3) [106.9(2)]
83.29(9) [82.97(9)]
72.4(2)
68.28(8)
147.06(7)
81.95(9)
128.0(2) [82.80(15)]
115.6(2) [149.8(2)]
160.1(2)
78.0(2)
171.43(8)
79.66(9) [83.01(8)]
138.6(3) [134.7(2)]
78.84(6)
84.33(9)
73.95(14) [77.30(15)]
149.03(15) [135.64(14)]
102.2(4)
116.51(7)
116.4(3) [115.7(2)]
93.24(7)
91.9(8)
82.1(2)
115.73(7)
51.73(13) [48.94(13)]c
93.55(10) [96.31(10)]e
a X(7) = N (1), O (2, 3, and 4), S (5); Y(1) = N(2)O3 (1), O(1S) (2), N(16) (3, 4, and 5); Y(2) = N(3)O3 (1), Cl(2)O4 (3), N(2)O3 (4). b Selected bond
o
˚
lengths (A) and angles ( ) for one of the two independent molecules of 1 are reported; where two values are reported, the first refers to the bond
distances and angles involving O(21) or O(31), and the second to bond distances and angles involving O(22) or O(32) (see Fig. 4). c Y(1)–M–Y(2) =
O(21)–Cd–O(22) [O(31)–Cd–O(32)] (see Fig. 4). d Where two values are reported, the first refers to the bond distances and angles involving O(20),
and the second to bond distances and angles involving O(22) (see Fig. 7). e Y(1)–M–Y(2) = N(16)–Cd–O(20) [N(16)–Cd–O(22)] (see Fig. 7). f Selected
o
˚
bond lengths (A) and angles ( ) for both independent molecules of 5 ([5-I] and [5-II]) are reported.
The X-ray crystal structure of the compound
[Cu(L4a)(dmf)](ClO4)2 (2) obtained by reacting L4a with
Cu(ClO4)2 in MeCN followed by crystallization in dmf/Et2O, is
not an exception. In this structure the stereochemical demands
of the metal centre for an octahedral coordination environment
results in a “butterfly” configuration of the macrocycle, with
the plane of the ring folded at the S · · · S vector. The site in
the pseudo-octahedral coordination sphere left free by L4a is
occupied by a molecule of dmf (Fig. 5, Table 3). The CuII ion
lies in the equatorial mean plane defined by the atoms S(4),
S(13), O(7), and O(10), and the apical N(1) atom is hydrogen
bonded to a perchlorato ion [NH · · · O 2.31, N · · · O 3.146(4)
◦
˚
A, N–H · · · O 147.5 ].
Despite numerous attempts to isolate the solid state crystalline
1 : 1 complexes of Lnc (n = 1–4) with the metal ions under inves-
tigation, only for [Zn(L1c)(ClO4)]ClO4 (3), [Cd(L1c)(NO3)]NO3
(4), and [Hg(L2c)](ClO4)2 (5) were crystals of diffraction quality
grown. In 3, the coordination environment around the metal
center is pseudo-octahedral with five positions occupied by
the N2S2O-donor set of L1c, and the remaining site taken up
by a monodentate perchlorato ligand (Fig. 6, Table 3). The
ZnII ion is displaced 0.23 A out of the equatorial mean plane
˚
Fig. 5 ORTEP view of the [Cu(L4a)(dmf)]+ complex cation interacting
with a perchlorato ion in 2 with the adopted numbering scheme.
Displacement ellipsoids are drawn at 30% probability. Hydrogen atoms
except that on the macrocyclic N-donor, and one perchlorato ion have
been omitted for clarity.
defined by the atoms N(1), S(4), S(10), and O(20) towards the
apical N(16) atom which is hydrogen bonded to a perchlorato
◦
˚
ligand [NH · · · O 2.24, N · · · O 3.121(15) A, N–H · · · O 168 ], and
provides the shortest Zn–donor interaction [Zn–N(16) 2.030(5)
˚
A]. Despite the presence of a coordinating 3-aminopropyl
pendant arm, the macrocyclic framework of L1c in 3 is folded
along the S · · · S vector with a N(1)–Zn–O(7) folding angle of
97.5(2)◦, and adopts a [2424] conformation having the same
distribution of the torsion angles observed in the 1 : 1 CuII
3 0 0 0
D a l t o n T r a n s . , 2 0 0 5 , 2 9 9 4 – 3 0 0 4

View Article Online
˚
shortest Cd-donor bound [Cd–N(16) 2.239(3) A] in this com-
plex, and is also hydrogen bonded to a nitrato ligand [NH · · · O
◦
˚
2.28, N · · · O 3.124(4) A, N–H · · · O 160 ]. As a consequence,
˚
the Cd–O(7) bond length [2.666(2) A], although well within
II
˚
the range 2.36–2.84 A reported for similar bonds in other Cd
complexes,45 is significantly longer than the corresponding bond
1a
32
˚
distance [2.561(3) A] observed in [Cd(L )(NO3)2] where the
macrocylic ligand is not functionalised and a bidentate nitrato
ligand formally occupies the position site of the 3-aminopropyl
pendant group of L1c in 4. Furthermore, the conformation
adopted upon coordination by the [12]aneNS2O moiety of L1c
in 4 is neither [2424] nor [3333] as confirmed by the torsion
angles at the C–N(1) bonds which are one anti [175.2(3)◦] and
one gauche [67.5(4)◦] (absolute values) instead of both anti (see
above for 3), and by the placements at the C–S(4) bonds which
assume absolute values of 95.5(3)◦ and 51.3(3)◦.
A genuine [3333] conformation characterised by pairs of
consecutive gauche torsion angles separated by an anti angle
is instead adopted upon coordination by the [12]aneNS3 moiety
of L2c in both independent units of [Hg(L2c)](ClO4)2 in 5. In
this complex the coordination sphere around the metal center in
Fig. 6 ORTEP view of [Zn(L1c)(ClO4)]ClO4 (3) with the adopted num-
bering scheme. Displacement ellipsoids are drawn at 30% probability.
Hydrogen atoms except those on the pendant amino group have been
omitted for clarity.
both independent molecules is a distorted square-based pyramid
II
˚
(Fig. 8, Table 3) with the Hg ions displaced about 0.9 A from
and CdII complexes of the unfunctionalised ligand L1a.32 In
particular, the torsion angles at the C–N(1) and C–O(7) bonds
assume anti arrangements [absolute values of the angles ranging
from 158.7(14) to174.4(14)◦], whereas oneof each pair of torsion
angles about the C–S bonds is gauche [52.9(12), 77.8(11)◦], the
other two torsion angles assumes the values of 126.4(12), and
122.9(12)◦. A gauche disposition is preferred at the C–C bonds
[48.0(14)–70.0(12)◦].
the mean basal coordination planes defined by the atoms N(1),
S(4), S(7), and S(10), towards the apical N(16) atoms which
provide the shortest Hg–donor interactions. The 3-aminopropyl
pendant group of both independent [Hg(L2c)](ClO4)2 molecules
is hydrogen bonded to two perchlorato ligands as shown in Fig. 8
˚
for 5-I [NH(16a) · · · O(33) 2.37, N(16) · · · O(33) 3.042(16) A, N–
H(16a) · · · O(33) 132◦; NH(16b) · · · O(22) 2.39, N(16) · · · O(22)
◦
˚
3.264(15) A, N–H(16b) · · · O(22) 164 ]. Hg–O distances ranging
˚
from 3.541(14) to 3.209(10) A between the metal centers in both
With respect to 3, the coordination sphere around the metal
centre in [Cd(L1c)(NO3)]NO3 (4) is much more distorted (Fig. 7,
Table 3). The CdII ion is effectively six-coordinate with the
macrocyclic moiety of L1c bound to one side of the metal ion
via its NS2O-donor set, and with the N-aminopropyl pendant
group and an essentially monodentate nitrato ligand completing
the coordination environment [the Cd–O(22) bond distance of
independent [Hg(L2c)](ClO4)2 molecules and the O-atoms of the
perchlorato ions are too long to be indicative for significant
interactions.
˚
2.838(3) A is at the extreme upper end of the commonly accepted
range where the nitrato ligand can be considered bidentate].
The CdII ion is displaced well out of the ring cavity of the
˚
[12]aneNS2O framework [1.09 A out of the mean plane defined
by the atoms N(1), S(4), S(10), and O(7)] by the interaction with
the 3-aminopropyl pendant group of L1c which is involved in the
Fig.
8
ORTEP view of one (5-I) of the two independent
[Hg(L2c)](ClO4)2 units in 5 with the adopted numbering scheme.
Displacement ellipsoids are drawn at 30% probability. Hydrogen atoms
have been omitted for clarity except those on the pendant amino group.
Synthesis of Lnd (n = 1–4) and Lne (n = 1, 2, 4) and optical
response of these new fluorescent sensors in the presence of CuII,
ZnII, CdII, HgII, and PbII.
Lnd (n = 1–4) were synthesized by reacting Lnc (n = 1–4)
with a stoichiometric amount of dansyl chloride in MeCN. Lne
(n = 1, 2, 4) were instead synthesized by direct Schiff-base
condensation of Lnc (n = 1, 2, 4) with 9-anthraldehyde in dry
EtOH : MeCN (1 : 1 v/v) followed by the reduction of the
Fig. 7 ORTEP view of [Cd(L1c)(NO3)]NO3 (4) with the adopted num-
bering scheme. Displacement ellipsoids are drawn at 30% probability.
Hydrogen atoms except those on the pendant amino group have been
omitted for clarity.
D a l t o n T r a n s . , 2 0 0 5 , 2 9 9 4 – 3 0 0 4
3 0 0 1

View Article Online
Table 4 Photophysical properties of Lnd (n = 1–4) and Lne (n = 1, 2, 4) in MeCN/H2O (4 : 1 v/v) at pH 7.0, and fluorescence intensity changes as a
function of the metal (Irel(%))a
Absorption
Fluorescence
Ligand
L1d
k/nm
e/dm3mol−1cm−1
k/nm
U
CuII
—
CdII
—
HgII
11.0
1.9
251
338
252
338
253
338
253
339
350
367
387
350
367
387
350
368
388
19960
4260
15530
4800
29770
8400
15150
4260
4260
6240
5550
4660
6760
5990
4620
6650
5860
534
532
530
528
416
0.24
0.24
0.17
0.16
0.065
L2d
39.0
22.0
—
—
L3d
—
27.0
5.6
L4d
—
L1e
21.0
160.4b
8.0
L2e
L4e
417
416
0.057
0.086
12.5
—
—
2.7
64.9c
10.0
a Irel (%) represents the percentage of the residual fluorescence intensity for a L/MII molar ratio of 2; ZnII does not affect the fluorescence intensity of
any of the ligands, while an insoluble precipitate forms upon addition of PbII to solutions of all ligands. b A smooth and monotonic increase of the
luminescence intensity is observed upon addition of CdII to L1e. c A smooth and monotonic decrease of the luminescence intensity is observed upon
addition of CuII to L4e.
crude products obtained with NaBH4/NaBH3CN (1 : 1) in dry
EtOH. Unfortunately, despite many attempts we were unable to
a prepare a pure sample of L3e (Scheme 1) using this synthetic
procedure. The photophysical properties of these compounds
(kex, eex, kem, fluorescent quantum yield, U) are reported in
Table 4.
The free ligands Lne (n = 1, 2, 4) showed in MeCN/H2O (4 :
1 v/v) at pH 7.0 the typical absorption and emission due
to the presence of the anthracenyl group with the maximum
in the absorption band at about 367 nm and the maximum
in the fluorescence band at 416 nm (Table 4). The values
measured for the fluorescence quantum yield ranged from 0.057
to 0.086 (Table 4). These values are remarkably lower than that
observed for anthracene in MeCN (0.27) due to a photoinduced
electron transfer process from the secondary nitrogen of the
propylamino pendant arm to the p system of the fluorescent
unit.53
The optical response of Lnd (n = 1–4) and Lne (n =
1, 2, 4) to CuII, ZnII, CdII, HgII and PbII was studied in
MeCN/H2O (4 : 1 v/v) solutions buffered with HEPES [4-(2-
hydroxyethyl)piperazine-1-ethanesulfonic acid] at pH 7.0.
The absorption spectra of all free dansylamidopropyl deriva-
tives, L1d–4d, in MeCN/H2O (4 : 1 v/v) at pH 7.0 present a
large, unstructured band around 338 nm and a more intense one
around 252 nm, see Table 4. These species are also luminescent,
exhibiting an emission band maximum in the range 528 (L4d)–
534 nm (L1d). The highest fluorescent quantum yield is that
measured for L1d and L2d (U = 0.24), while for L3d and L4d the
values of U measured are 0.17 and 0.16, respectively.
Changes in the absorption and emission spectra of the four
ligands were only observed upon addition of CuII or HgII to
MeCN/H2O (4 : 1 v/v) solutions of Lnd (n = 1–4) buffered
at pH 7.0; an insoluble precipitate was formed in the case of
PbII. In particular, the bands at 338 nm for L1d–L3d and 339 nm
for L4d shift to lower wavelengths (less than 6 nm) and increase
slightly in intensity (see ESI Fig. S5 for the L4d/HgII system†).
However, the intensity increase was more evident upon addition
of CuII.
Similar effects, and in general not dramatic, were observed on
the absorption properties of Lne (n = 1, 2, 4) as a consequence
of metal ion complexation: a small decrease in extinction
coefficients was observed only upon addition of CuII or HgII
(a small increase of e was noticed instead in the case of L4e with
both metal ions) accompanied in some cases by very small red
shifts (less than 4 nm for L1e and L2e with both metal ions); no
changes were recorded upon addition of ZnII and CdII, while
an insoluble precipitate formed in the case of PbII with all three
anthracenyl derivatives. For the titrations of L1e with both CuII
and HgII, and of L2e with HgII, well defined isosbestic points
were observed (see ESI Fig. S6 for the L2e/HgII system†).
In contrast to L1d (see above), a quenching of the luminescence
intensity was observed for L1e upon addition of both CuII
and HgII (Irel 21.0 and 8.0%, respectively). The plateau values
were reached after the addition of about one, and two molar
equivalents of HgII and CuII, respectively. A smooth monotonic
increase of the luminescence intensity was also monitored upon
addition of CdII (Irel 160.4% for a L1e/CdII molar ratio of 2). The
behavior of L2e and L4e is instead very similar to that observed for
their structural analogous, L2d and L4d, bearing the dansylamido
moiety. In fact, a quenching of the fluorescence was caused by
the addition of 1 molar equivalent of CuII and HgII (Irel 12.5
and 2.7%, respectively) in the case of L2e (Fig. 9c), whereas only
the addition of HgII determined a significant quenching (Irel
10%) in the case of L4e (Fig. 9d). However, for L4e, a smooth
and monotonic decrease of the luminescence intensity was also
observed upon addition of CuII (Irel 64.9% for a L1e/CuII molar
ratio of 2), see Fig. 9d and Fig. 9b for comparison with L4d.
The decrease of fluorescence intensity upon addition of CuII or
HgII ions to Lne (n = 1, 2, 4), as described above, provides clear
evidence that in these systems the quenching process due to
Interestingly, while the fluorescence of L2d and L3d was
quenched by both CuII (Irel 39.0 and 22.0% for L2d and L3d,
respectively) and HgII (Irel 1.9 and 27.0% for L2d and L3d,
respectively, see Fig. 9a for the L3d), for L1d and L4d a quenching
of the luminescence intensity was observed only upon addition
of HgII up to a HgII/Lnd (n = 1, 4) molar ratio of 1 (Irel 11.0 and
5.6%, respectively, Fig. 9b for the L4d). In all cases no shift of
the emission bands was observed. The strong quenching effect
observed for L2d and L3d with CuII can be ascribed to either a
dansyl-to-metal energy-transfer (ET) or to a metal-to-dansyl
electron-transfer (eT) mechanism. Data obtained for similar
complexes52 have suggested that at least a contribution from the
ET mechanism is operating. In the case of HgII, a contribution
to fluorescence quenching from the heavy-atom effect cannot be
excluded, while the ET mechanism cannot be effective for this
d10 ion.52
3 0 0 2
D a l t o n T r a n s . , 2 0 0 5 , 2 9 9 4 – 3 0 0 4

View Article Online
Fig. 9 Fluorescent intensity/molar ratio plot for L3d (a), L4d (b), L2e (c) and L4e (d) (2.5 × 10−5 M, MeCN/H2O (4 : 1 v/v), pH = 7.0) in the presence
of increasing amounts of CuII (᭹), HgII (ꢁ), ZnII (+), and CdII (ꢂ). An insoluble precipitate formed upon addition of PbII to solutions of all four
ligands.
eT or ET mechanisms involving the metal ion prevails over the
eT process among the secondary N-donor and the anthracene
that would be interrupted upon complexation of the function-
alised pendant arm to the metal ion with consequent switching
on of the fluorescence. The fluorescence intensity changes
are tabulated as a function of metal and sensor number in
Table 4.
in aqueous solutions (see ref. 54 for analytical details such as
membrane preparation, response mechanism and measuring
principle, measuring range and dynamic response time, and
interference of co-existing cationic species). These results should
aid further development and application of selective chemical
sensors for “soft” metal cations which employ S-containing
macrocycles as receptor units. In fact, there are only a few reports
of fluorescent chemosensors for heavy metal ions which feature
S-donor macrocycles as receptors units. Our results once more
point out that within a supramolecular modular approach to
the design of “soft” metal ions-selective fluorescent sensors,
the choice of the “read-out” unit is critical to both sensor-
performance and sensor-selectivity.33
Conclusions
Following the “receptor–spacer-signalling unit” model, we have
synthesised the fluorescent chemosensors Lnd (n = 1–4) and Lne
(n = 1, 2, 4, Scheme 1) in which mixed N/S- and N/S/O-donor
macrocycles are connected through an aminopropyl linker to
a dansyl or a 9-anthracenylmethyl fluorescent moiety. Their
optical response towards the “borderline” and the “soft” metal
ions CuII, ZnII,CdII, HgII and PbII which are of great importance
from an industrial, biological, and environmental points of
view, has been investigated at pH 7.0 in MeCN/H2O (4 :
1 v/v) solutions. As expected, and in line with our studies
performed both in solution and in the solid state on the
coordination chemistry of the N-aminopropyl pendant arm
precursors L1c–4c (Scheme 1), the presence of S-donor atoms in
the macrocyclic frameworks can confer specificity and selectivity
to the interaction of the present molecular sensors with the
metal ions considered. In particular, L1d and L4d both containing
a dansylamido fluorogenic fragment as signalling unit, and
respectively, a NS2O- and a NS2O2-donor set in the cyclic
receptor unit, respond selectively to HgII by a quenching in
the fluorescence emission. In particular, L1d has already found
analytical application in the construction of a fluorimetric bulk
optode membrane for selective subnanomolar detection of HgII
Acknowledgements
We thank the Ministero Italiano dell’Universita` e della Ricerca
Scientifica e Tecnologica (MIUR) for financial support. (FIRB
project RBAU01HFNS and FISR project SAIA). We thank
EPRSC (UK) for funding of the diffractometer.
References
1 A. P. de Silva, H. Q. G. Gunaratne, T. Gunnlaugsson, A. J. M. Huxley,
C. P. McCoy, J. T. Rademacher and T. E. Rice, Chem. Rev., 1997, 97,
1515.
2 Chemosensors of Ions and Molecular Recognition, ed. A. W. Czarnik
and J.-P. Desvergene, NATO ASI Ser., Ser. C, vol. 492, Kluwer
Academic Publishers, Dordrecht, 1997.
3 E. Kimura and T. Koike, Chem. Soc. Rev., 1998, 27, 179.
4 (a) B. Valeur and I. Leray, Coord. Chem. Rev., 2000, 205, 3; R.
Me´tivier, I. Leray and B. Valeur, Chem. Commun., 2003, 996.
5 V. Balzani, P. Ceroni, S. Gestermann, C. Kauffmann, M. Gorka and
F. Vo¨gtle, Chem. Commun., 2000, 853.
D a l t o n T r a n s . , 2 0 0 5 , 2 9 9 4 – 3 0 0 4
3 0 0 3

View Article Online
6 (a) L. Prodi, F. Bolletta, M. Montalti and N. Zaccheroni, Coord.
Chem. Rev., 2000, 205, 59; (b) L. Prodi, New. J. Chem., 2005, 29, 20.
7 K. Rurack, Spectrochim. Acta, Part A, 2001, 57, 2161.
8 K. Parack and U. Resch-Genger, Chem. Soc. Rev., 2002, 31, 116.
9 L. Fabbrizzi, M. Licchelli and A. Taglietti, Dalton Trans., 2003, 3471.
10 J. Zao, M. G. Davidson, M. F. Mahon, G. Kociok-Ko¨hn and T. D.
James, J. Am. Chem. Soc., 2004, 126, 16179.
11 Y. D. Fernandez, A. Pe´rez Gramatges, V. Amendola, F. Foti, C.
Mangano, P. Pallavicini and S. Patroni, Chem. Commun., 2004, 1650.
12 R. Pohl, D. Aldakov, P. Kuba´t, K. Jurs´ıkova´, M. Marquez and P.
Anzenbacher, Jr., Chem. Commun., 2004, 1282.
13 Y. Zheng, J. Orbelescu, X. Ji, F. M. Andreopoulos, S. M. Pham and
R. M. Leblanc, J. Am. Chem. Soc., 2003, 125, 2680.
14 W. Yang, J. Yan, H. Fang and B. Wang, Chem. Commun., 2003, 792.
15 B. K. Kaletas, R. M. Williams, B. Ko¨nig and Luisa De Cola, Chem.
Commun., 2002, 776.
Trans., 2003, 901; (b) L. Tei, A. J. Blake, P. A. Cooke, C. Caltagirone,
F. Demartin, V. Lippolis, F. Morale, C. Wilson and M. Schro¨der,
J. Chem. Soc., Dalton Trans., 2002, 1662.
33 A. J. Blake, A. Bencini, C. Caltagirone, G. De Filippo, L. S. Dolci,
A. Garau, F. Isaia, V. Lippolis, P. Mariani, L. Prodi, M. Montalti,
N. Zaccheroni and C. Wilson, Dalton Trans., 2004, 2771.
34 (a) M. Shamsipur, M. Javanbakht, M. F. Mousavi, M. R. Ganjali,
V. Lippolis, A. Garau and L. Tei, Talanta, 2001, 55, 1047; (b) M.
Shamsipur, M. Javanbakht, V. Lippolis, A. Garau, G. De Filippo,
M. R. Ganjali and A. Yari, Anal. Chim. Acta, 2002, 462, 225;
(c) M. Shamsipur, M. Javanbakht, M. R. Ganjali, M. F. Mousavi,
V. Lippolis and A. Garau, Electroanal., 2002, 14, 1691; (d) M.
Shamsipur, M. Hosseini, K. Alizadeh, M. F. Mousavi, A. Garau,
V. Lippolis and A. Yari, Anal. Chem., 2005, 77, 276.
35 A. Credi and L. Prodi, Spectrochim. Acta, Sect. A, 1998, 54,
159.
16 E. M. Nolan, S. C. Burdette, J. H. Harvey, S. A. Hildebrand and S. J.
Lippard, Inorg. Chem., 2004, 43, 2624.
17 X. Xu, H. Xu and H.-F. Ji, Chem. Commun., 2001, 2092.
18 A. P. de Silva, G. D. McClean and S. Pagliari, Chem. Commun., 2003,
2010.
19 A. Bencini, A. Bianchi, C. Lodeiro, A. Masotti, A. J. Parola, F. Pina,
J. Seixas, de Melo and B. Valtancoli, Chem. Commun., 2000, 1639.
20 S. Aoki, S. Kaido, H. Fujioka and E. Kimura, Inorg. Chem., 2003,
42, 1023.
36 L. G. A. van de Water, W. Buijs, W. L. Driessen and J. Reedijk,
New. J. Chem., 2001, 25, 243.
37 S. Afshar, S. T. Marcus, L. R. Gahan and T. W. Hambley,
Aust. J. Chem., 1999, 52, 1.
38 C. Bazzicalupi, A. Bencini, V. Fusi, C. Giorgi, P. Paoletti and B.
Valtancoli, Inorg. Chem., 1998, 37, 941.
39 G. Gran, Analyst, 1952, 77, 661.
40 P. Gans, A. Sabatini and A. Vacca, Talanta, 1996, 43, 807.
41 SADABS Area-Detector Absorption Correction Program, Bruker
AXS, Inc., Madison, WI, USA, 2000.
21 F. Sancenon, A. Benito, F. J. Herna´ndez, J. M. Lloris, R. Mart´ınez-
Ma´n˜ez, T. Pardo and J. Soto, Eur. J. Inorg. Chem., 2002, 866.
22 P. V. Bernhardt, E. G. Moore and M. J. Riley, Inorg. Chem., 2002,
41, 3025.
42 G. M. Sheldrick, SHELXS 97, Acta Crystallogr., Sect A, 1990, 46,
467.
43 G. M. Sheldrick, SHELXL 97, Universita¨t Go¨ttingen, Germany,
1997.
44 M. W. Glenny, L. G. A. van de Water, W. L. Driessen, J. Reedijk,
A. J. Blake, C. Wilson and M. Schro¨der, Dalton Trans., 2004, 1953
and references therein.
45 K. A. Byriel, L. R. Gahan, C. H. L. Kennard and C. J. Sunderland,
J. Chem. Soc., Dalton Trans., 1993, 625.
46 J. Powell and C. J. May, J. Am. Chem. Soc., 1982, 104, 2636.
47 F. Arnaud-Neu, M.-J. Schwing-Well, R. Louis and R. Weiss, Inorg.
Chem., 1979, 18, 2956.
48 M.-T. Youinou, J. A. Osborn, J.-P. Collin and P. Lagrange, J. Am.
Chem. Soc., 1986, 25, 453.
49 A. Bencini, A. Bianchi, E. Garcia-Espan˜a, M. Micheloni and J. A.
Ramirez, Coord. Chem. Rev., 1999, 188, 97.
50 S. T. Marcus, P. V. Bernhardt, L. Grøndahl and L. R. Gahan,
Polyhedron, 1999, 18, 3451.
51 R. Louis, F. Arnaud-Nu, R. Weiss and M. J. Schwing-weill, Inorg.
Nucl. Chem. Lett., 1977, 13, 31.
23 J. P. Geue, N. J. Head, A. D. Ward and S. F. Lincoln, Dalton Trans.,
2003, 521.
24 Y. Al Shihadeh, A. Benito, J. M. Lloris, R. Mart´ınez-Ma´n˜ez, T. Pardo,
J. Soto and M. D. Marcos, J. Chem. Soc., Dalton Trans., 2000, 1199.
25 (a) L. Prodi, F. Bolletta, M. Montalti, N. Zaccheroni, J. S. Bradshaw,
P. B. Savage and R. M. Izatt, Tetrahedron Lett., 1998, 39, 5451;
(b) L. Prodi, C. Bargossi, M. Montalti, N. Zaccheroni, N. Su, J. S.
Bradshaw, R. M. Izatt and P. B. Savage, J. Am. Chem. Soc., 2000,
122, 6769; (c) L. Prodi, M. Montalti, N. Zaccheroni, J. S. Bradshaw,
R. M. Izatt and P. B. Savage, Tetrahedron Lett., 2001, 42, 2941; (d) L.
Prodi, M. Montalti, J. S. Bradshaw, R. M. Izatt and P. B. Savage,
J. Inclusion Phenom. Macrocyclic Chem., 2001, 41, 123.
26 H. He, M. A. Mortellaro, M. J. P. Leiner, R. J. Fraatz and J. K. Tusa,
J. Am. Chem. Soc., 2003, 125, 1468.
27 R. T. Bronson, J. S. Bradshaw, P. B. Savage, S. Fuangswasdi, S. C.
Lee, K. E. Krakowiak and R. M. Izatt, J. Org. Chem., 2001, 66, 4752.
28 S. C. Lee, R. M. Izatt, X. X. Zhang, E. G. Nelson, J. D. Lamb, P. B.
Savage and J. B. Bradshaw, Inorg. Chim. Acta, 2001, 317, 174.
29 G. De Santis, L. Fabbrizzi, M. Licchelli, C. Mangano, D. Sacchi and
N. Sardone, Inorg. Chim. Acta, 1997, 257, 69.
52 L. Prodi, F. Bolletta, M. Montalti and N. Zaccheroni, Eur. J. Inorg.
Chem., 1999, 901.
53 (a) See for example,P. R. Ashton, R. Ballardini, V. Balzani, M.
Gomez-Lopez, S. E. Lawrence, M. V. Martinez-Diaz, M. Montalti,
A. Piersanti, L. Prodi, J. F. Stoddart and D. J. Williams, J. Am.
Chem. Soc., 1997, 119, 10641; (b) A. W. Czarnik, in Fluorescent
Chemosensor for Ion and Molecule Recognition, ed. A. W. Czarnik,
ACS Symposium series, Washington, DC, 1992, p. 104.
54 M. Shamsipur, M. Hosseini, K. Alizadeh, N. Alizadeh, A. Yari, C.
Caltagirone and V. Lippolis, Anal. Chim. Acta, 2005, 533, 17.
30 J. V. Ros-Lis, R. Mart´ınez-Ma´n˜ez, K. Rurack, F. Sanceno´n, J. Soto
and M. Spieles, Inorg. Chem., 2004, 43, 5183.
31 M. C. Aragoni, M. Arca, F. Demartin, F. A. Devillanova, F. Isaia,
A. Garau, V. Lippolis, U. Papke, M. Shamsipur, L. Tei, A. Yari and
G. Verani, Inorg. Chem., 2002, 41, 6623.
32 (a) C. Caltagirone, A. Bencini, F. Demartin, F. A. Devillanova, A.
Garau, F. Isaia, V. Lippolis, U. Papke, L. Tei and G. Verani, Dalton
3 0 0 4
D a l t o n T r a n s . , 2 0 0 5 , 2 9 9 4 – 3 0 0 4