Versatility and dynamics of the copper(I) coordination sphere in sterically hindering tris(pyrazolyl)methane-incorporating macrobicycles

Article abstract of DOI:10.1039/b816275a

Two arene-capped macrobicycles (1 and 2) incorporating the tris(pyrazolyl)methane (Tpm) chelate have been prepared from a benzylthiol-functionalized Tpm precursor (3). Reaction of either macrobicycle with Cu(CH₃CN)₄⁺ leads

Full text of DOI:10.1039/b816275a

View Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/njc | New Journal of Chemistry
Versatility and dynamics of the copper(I) coordination sphere in sterically
hindering tris(pyrazolyl)methane-incorporating macrobicycleswz
Leyong Wang,ab Jean-Claude Chambron*a and Enrique Espinosaac
Received (in Montpellier, France) 22nd September 2008, Accepted 28th November 2008
First published as an Advance Article on the web 8th January 2009
DOI: 10.1039/b816275a
Two arene-capped macrobicycles (1 and 2) incorporating the tris(pyrazolyl)methane (Tpm)
chelate have been prepared from a benzylthiol-functionalized Tpm precursor (3). Reaction of
+
either macrobicycle with Cu(CH3CN)4 leads to tetrahedral or trigonal-planar, fluxional
complexes incorporating the Cu(CH3CN)+ subunit ([Cu(1)(CH3CN)]+ and [Cu(2)(CH3CN)]+).
The acetonitrile ancillary ligand does not fit inside the macrobicycle cavity and can be removed
by heating under vacuum, which produces the [Cu(1)]+ and [Cu(2)]+ species probably involving
intramolecular thioether coordination. The [Cu(1)(CH3CN)]+ complex was shown to convert
slowly in wet acetone into a helical coordination polymer, which is formulated as
[Cu(1)(H2O)]nn+. In the crystal, the metal cation in tetrahedral geometry is bound to two
pyrazole nitrogens and bridges two macrobicyclic subunits by intermolecular thioether sulfur
coordination. Its coordination sphere is remarkably completed by a water molecule that is located
inside the macrobicycle and is bound to the uncoordinated pyrazole fragment and to the
mesitylene cap by OHꢀ ꢀ ꢀN and OHꢀ ꢀ ꢀp hydrogen bonding interactions.
these approaches being inspired by the concept of enzyme
Introduction
pockets, and producing confined environments such as those
offered by metal-modified zeolites,5 or self-assembled cages.6
The stereoelectronic control of the coordination sphere of
transition metal cations by multidentate ligands is an important
Multiple examples of sterically encumbered ligands can be
issue in the design of mimics of the active sites of metallo-
found in the scorpionate family, especially the anionic tris-
(pyrazolyl)hydroborates (TpR0,R00).7 Substitution of the pyrazole
enzymes and the development of homogeneous catalysts.
Ligands that favor coordinatively unsaturated transition metal
3-position (R0) by sterically hindering groups enforces tetra-
complexes and enforce unusual coordination numbers are
hedral coordination geometry at the metal and allows control
particularly relevant in this context, as these systems are often
of the fourth, apical coordination site. For example, four-
particularly reactive.1 As a matter of fact, the activation of
coordinate monomeric Cu(II) thiolato complexes, that bear
otherwise inert molecules, such as dinitrogen, has amply relied
spectroscopic similarities to blue copper proteins, have been
on these principles.2
obtained with the TpiPr,iPr ligand.8 The use of bulky, hydro-
phobic substituents such as t-butyl in TptBu,Me has led to the
Sterically encumbering ligands can be designed by grafting
bulky substituents in close proximity to the coordinating sites,3
discovery of a Zn(II) complex that is a structural and functional
or by incorporating the latter in macro(poly)cyclic structures,4
mimic of carbonic anhydrase.9 The Cu(I) complex of trans-
Tppm, a tris(pyrazolyl)hydroborate ligand featuring a deep,
a
´
´
Institut de Chimie Moleculaire de l’Universite de Bourgogne
(ICMUB, UMR CNRS n1 5260), 9 avenue Alain Savary,
BP 47870, 21078 Dijon cedex, France
C3-symmetric chiral cavity, gave 85% ee for the cis product
of the catalytic enantioselective cyclopropanation of styrene by
ethyl diazoacetate.10
E-mail: jean-claude.chambron@u-bourgogne.fr;
Fax: +33 (0)3 8039 6117; Tel: +33 (0)3 8039 6116
b Key Laboratory of Mesoscopic Chemistry of Ministry of Education
and School of Chemistry and Chemical Engineering, Nanjing
A few metal complexes with sterically hindered tris(pyrazolyl)-
methane (Tpm) analogues of the Tp ligands have been reported.
Silver(I) forms complexes of 1 : 1 stoichiometry only in the
presence of the bulky TpmtBu ligand,11 but cationic tetrahedral
copper(I) carbonyl and acetonitrile complexes with TpmPh or
University, 210093 Nanjing, P.R. China. E-mail: lywang@nju.edu.cn
´
Mineraux et Biologiques (LCM3B, UMR CNRS n1 7036),
c
´
Laboratoire de Cristallographie et Modelisation des Materiaux
´
´
Nancy-Universite, 1 boulevard des Aiguillettes,
54606 Vandæuvre-les-Nancy, France.
TpmtBu do not differ from those with TpmMe,Me 12
.
`
We recently communicated on the synthesis of macrobicycle
that incorporates the tris(pyrazolyl)methane chelate
E-mail: enrique.espinosa@lcm3b.uhp-nancy.fr;
Fax: +33 (0)3 8340 6492; Tel: +33 (0)3 8368 4953
w Dedicated to Prof. Jean-Pierre Sauvage on the occasion of his 65th
birthday.
1
(Scheme 1).13 As suggested by handling of Corey–Pauling–
Koltun molecular models, the cavity left over copper(I)
coordinated to the three pyrazole nitrogens is not large enough
to accommodate a small, rigid rod axial ligand such as
CH3CN. We were therefore intrigued to know how this metal,
which shows preference for tetrahedral coordination geometry
in normal conditions, but turns to trigonal-planar (or even
z Electronic supplementary information (ESI) available: Synthesis
of compound
3
and its precursors; NOESY 2D 1H/1H NMR
spectra of 1 and [Cu(1)(CH3CN)]PF6, VT 1H NMR spectra of 1,
[Cu(1)(CH3CN)]PF6, [Cu(2)(CH3CN)]PF6, [Cu(1)]PF6, and [Cu(2)]PF6;
X-ray crystallographic data for 1 and [Cu(1)(H2O)n](PF6)n. CCSC
710458–710459. For ESI and crystallographic data in CIF or other
electronic format see DOI: 10.1039/b816275a
ꢁc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
New J. Chem., 2009, 33, 327–336 | 327

View Online
This suggests that one C(methine)–N(pyrazole) bond is
cleaved under laser irradiation. This fragmentation process
presumably also happens in the case of the macrobicycles.
However, since the released pyrazole-containing arm is still
attached to the rest of the molecule, the observed signal
corresponds to the molecular peak.
Macrobicycles 1 and 2 have similar 1H NMR spectra
(CDCl3, CD2Cl2) and display C3v symmetry in solution
(Table 1). The spectrum of 1 was assigned using 1H/1H 2D
NOESY, which shows the following through space correla-
tions: 50-H/CH, 40-H/4-H, ArH/2-H, ArCH2S/ArH, SCH2/2-H,
and SCH2/6-H (Fig. S1, ESIz). The correlation ArH/2-H
shows that the molecule has a dwarfed conformation obtained
by squeezing the benzene cap towards the Tpm chelate. The
methylene protons are enantiotopic, even at 190 K, where two
close, although slightly broadened singlets are still observed
(Fig. S2, ESIz). Comparison of 1 and 2 with precursor 3,
which has an open, acyclic structure indicates that methine
protons are mostly affected by the conformational change
(Dd E ꢂ0.35 ppm), whereas pyrazole protons 40-H and 50-H
are not. Other protons that show significant shifts are 2-H and
4-H of the phenyl rings (7Dd7 Z 0.2 ppm for 2). Their ortho
position to the pyrazole subunit makes the former obviously
sensitive to conformational changes about the pyrazole–
phenyl bond. The methylene protons are also affected
(Dd E ꢂ0.22 ppm).
Scheme 1 Structural formulae of the macrobicycles.
linear) under steric pressure,14 would respond to the constraints
imposed by the macrobicyclic ligand.
This report adds a new member to this family of sterically
hindered scorpionate molecules, 2, which derives from 1 by
methyl substitution of the benzene cap, and examines their
coordination chemistry with copper(I).
Results and discussion
Synthesis of the macrobicycles
Macrobicycles 113 and 2 were prepared by tripod–tripod
coupling reaction between benzylthiol-functionalized tris-
(pyrazolyl)methane 313 and either 1,3,5-tribromomethylbenzene
4 or 1,3,5-tribromomethyl-2,4,6-trimethylbenzene 5, in the pre-
sence of K2CO3 as base, in N,N-dimethylformamide (DMF) at
60 1C under high dilution (1.7 mM) (Scheme 2). In these
conditions, they were obtained in 28%13 and 33% yields for 1
and 2, respectively, after purification by column chromatography.
Macrobicycle 1 formed suitable X-ray quality crystals by
slow evaporation of a dichloromethane–heptane solution, and
its crystal structure was solved. Two non-equivalent molecules
share the unit cell (Fig. 1): one encapsulates a water molecule,
the other is empty. The two macrobicycles are perpendicular
to each other, the m-phenylene spacer of the empty one
showing p–p stacking interactions with the benzene cap of
the other, with a 3.706 A distance between centroids and a 2.21
angle between the mean planes of the aromatic fragments.
The m-phenylene spacers of the empty molecule are nearly
coplanar with the pyrazole subunits, the angles between their
mean planes ranging from 14.21 to 19.71. Two spacers of the
other molecule are tilted to some extent, with angles of 32.11
and 33.71. This is presumably the result of the encapsulation of
a water molecule, which is equally disordered on two posi-
tions, one outside and the other inside the macrobicycle. The
Structures of the macrobicycles
Comparison of the MALDI-TOF spectra of 1 and 2 with the
spectrum of acyclic precursor 3 attests to the macrobicyclic
nature of the former. Whereas for the closed structures 1 and 2
the highest m/z corresponds to the molecular peak (respec-
tively, 694 and 738), in the case of the open system 3 it
corresponds to a bis(pyrazolyl)methane fragment (m/z 391),
the expected molecular peak at m/z 581 being not observed.
Scheme 2 Preparation of macrobicycles 1 and 2.
ꢁc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
328 | New J. Chem., 2009, 33, 327–336

View Online
Table 1 1H NMR data of tripod precursor 3, macrobicycles 1 and 2, and their copper(I) complexesa
CDCl3
d6-acetone (CD2Cl2)
d6-acetone
Solvent
Compoundb
1
2
3
1
[Cu(1)ACN]+
CIS (1)
[Cu(2)ACN]+
[Cu(1)]+ (Dd)f
[Cu(2)]+ (Dd)g
CH
50
2
4
6
5
ArH
40
ArCH2S
CH2S
ArCH3
CH3CN
8.13
7.66
7.66
7.42
7.38
7.31
7.03
6.63
3.62
3.60
8.17
7.73
7.58
7.43
7.32
7.29
8.50
7.68
7.79
7.68
7.31
7.36
9.02 (8.16)
8.04 (7.73)
7.96 (7.68)
7.51 (7.45)
7.43 (7.42)
7.32 (7.34)
7.23 (7.08)
6.70 (6.67)
3.81 (3.67)
3.73 (3.63)
9.41d,e (8.82)
8.44 (8.24)
8.14 (8.05)
7.67 (7.61)
7.62 (7.51)
7.46 (7.43)
7.32 (7.19)
6.99 (6.69)
3.92 (3.83)
3.85 (3.79)
+0.39 (+0.66)
+0.40 (+0.51)
+0.18 (+0.37)
+0.16 (+0.16)
+0.19 (+0.09)
+0.14 (+0.09)
+0.09 (+0.11)
+0.29 (+0.02)
+0.11 (+0.16)
+0.12 (+0.16)
9.40
8.45
8.17
7.63
7.50
7.39
9.55 (+0.14)
8.46 (+0.02)
8.03 (ꢂ0.11)
7.62 (ꢂ0.05)
7.57 (ꢂ0.05)
7.43 (ꢂ0.03)
7.29 (ꢂ0.03)
6.97 (ꢂ0.02)
3.90 (ꢂ0.02)
3.75 (ꢂ0.10)
9.39 (ꢂ0.01)
8.45 (0.00)
8.06 (ꢂ0.11)
7.60 (ꢂ0.03)
7.45 (ꢂ0.05)
7.35 (ꢂ0.04)
6.64
3.73
3.52
2.04
6.69
3.78
7.01
4.09
3.91
2.45
2.03
6.98 (ꢂ0.03)
4.07 (ꢂ0.02)
3.82 (ꢂ0.09)
2.42 (ꢂ0.03)
2.05 (1.97)c
1.90 (1.66)
–0.15 (–0.31)
a
b
Unless otherwise stated, the spectra were recorded at 500 MHz and 300 K. Chemical shifts are given in ppm downfield from SiMe4. Numbering
c
d
e
f
scheme as shown in Fig. 2. Refers to free CH3CN in d6-acetone (respectively, CD2Cl2). 298 K. 600 MHz. Dd (ppm) refers to the chemical
shift difference between homologous protons of [Cu(1)]+ and [Cu(1)ACN]+
.
Same as (f) for the complexes of 2.
g
could be estimated by comparing relevant geometric
parameters for the empty cage and related copper(I) tris-
(pyrazolyl)methane complexes. For example, in the case of
[Cu(TpmtBu)(CH3CN)]PF6,12 Nꢀ ꢀ ꢀN distances between
coordinating atoms range from 2.925 to 2.987 A, while they
are slightly longer for 1 (2.915–3.514 A). Other representative
parameters, the angles between the vectors joining the methine
carbon and the coordinating nitrogen atoms, are within
the same limits for both compounds (Table S1, ESIz).
The complexation of Cu(I) by the macrobicycles
The coordination chemistry of macrobicycles 1 and 2 was
investigated with copper(I), a prototypical metal cation that
shows preference for trigonal-planar and tetrahedral coordina-
tion environments, depending on the number of available
donor atoms.
Reaction of equimolar amounts of 1 and [Cu(CH3CN)4]PF6
in CH2Cl2–CH3CN 2 : 1 for 2 h, followed by evaporation of
the solvents and drying under vacuum at 35 1C for 12 h left a
1
material that was examined by H and 13C NMR. The color-
less product of the reaction did not dissolve completely in
CD2Cl2, but was apparently more soluble in d6-acetone.y The
mixtures were therefore clarified by filtration. The spectra of
the resulting solutions differ from that of the free macrobicycle
in the corresponding solvents, which indicates that complexa-
tion has indeed taken place. Complexation induced shifts
(CIS) of macrobicycle 1 are all positive (Table 1). In CD2Cl2,
the highest CIS are observed for the methine CH (+0.66 ppm)
and 50-H (+0.51 ppm), while 40-H shows the lowest
(+0.02 ppm). 2-H is also affected significantly (+0.37 ppm),
which suggests some reorientation of the m-phenylene spacers
upon complexation, but methylene (+0.16 ppm), benzene
(+0.11 ppm) and other aryl protons (+0.09 ppm) are, by
comparison, only slightly shifted. In addition to the signals of
the complexed macrobicycle a singlet at d 1.66 ppm integrating
for E3 H is due to Cu(I)-bound CH3CN. It is shielded by
ꢂ0.31 ppm with respect to free CH3CN (d 1.97 ppm in CD2Cl2).
All the peaks broaden upon decreasing the temperature,
Fig. 1 ORTEP view of a pair of non-equivalent molecules in the
crystal structure of 1. The depicted H2O solvate (shown in (b)) is
disordered over two positions, one inside, the other outside the cavity.
Ellipsoids are drawn at 50% probability level.
inside water molecule is well centered in the cavity of the cage,
with very similar Oꢀ ꢀ ꢀN bond distances that range from
3.297(6) A to 3.447(6) A, and a distance of 4.521 A between
the oxygen atom and the centroid of the mesitylene cap. The
corresponding angles range from 141.11 to 156.11. These data
indicate that the inside water molecule forms hydrogen bonds
with two pyrazole N atoms.
Macrobicycles 1 and 2 have the unique property of
incorporating a tris(pyrazolyl)methane scorpionate ligand in
its conformation for making metal complexes with trigonal-
pyramidal or tetrahedral geometry. The degree of preorganization
y This made the determination of yields difficult.
ꢁc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
New J. Chem., 2009, 33, 327–336 | 329

View Online
especially those of the aryl and methylene groups, with a
maximum broadening of the macrobicycle protons at 210 K,
which is characteristic of coalescence (Fig. S3, ESIz). Below,
smaller peaks separate from the main set of signals, while the
methine CH, 40-H and 50-H of the latter sharpen again. This
indicates that, at low temperature (182 K), the Cu(I) complex of
1 has a major conformation that is accompanied by minor
one(s). The singlet of the CH3CN protons broadens as the
temperature is lowered, but its chemical shift does not change
throughout. At least two very broad, overlapping peaks have
separated in the 182 K spectrum. The spectral evolution is
different in d6-acetone, as simple monotonous broadening is
observed upon cooling (Fig. S4, ESIz). At 182 K at least three
broad signals are observed in the region of the methylene
protons, which probably indicates that the latter form dia-
stereotopic pairs below this temperature, their coalescence
occurring at E190 K.
By contrast, the Cu(I) complex of 2 prepared similarly
dissolves easily in d6-acetone. However, its 1H NMR spectrum
in this solvent shows broad features at 300 K, except for the
CH3CN protons. As increasing the temperature sharpens
the signals, line broadening is diagnostic of dynamic effects
in the complex (Fig. 3). Indeed, macrobicycle 2 itself has sharp
resonances at 300 K. As shown in Table 1, analogous protons
of 1 and 2 in their complexes have similar d, with the exception
of protons 6-H and ArCH2S, which are close to the mesityl
cap, and show |Dd| 4 0.10 ppm. This is not surprising, as the
macrobicycles differ by the very nature of the cap. Similar
features are found in the 1H/1H NOESY 2D spectrum of
[Cu(2)(CH3CN)]+, except that the correlation ArCH2S/ArH
of [Cu(2)(CH3CN)]+ is replaced by correlations ArCH2S/CH3
and CH2S/CH3. However, the correlation CH3/CH3CN is
notably absent.
The dynamic NMR features of the Cu(I) complex of 2 in
d6-acetone differ significantly from those of [Cu(1)(CH3CN)]+.
Contrary to the latter case, decreasing the temperature gradually
sharpens the signals at first, with a maximum at 240 K (Fig. 3).
Further cooling considerably broadens the signals of the
2-H and methylene protons, while the others just broaden a
little. The signal of CH3CN broadens monotonously between
300 and 182 K, and shifts downfield from 2.02 to 2.21 ppm
(Fig. S6, ESIz).
Except for 40-H (+0.29 vs +0.02 ppm in CD2Cl2), the CIS
values are all decreased in acetone, including the bound
acetonitrile molecule (ꢂ0.15 vs ꢂ0.31 ppm in CD2Cl2). The
nitrile carbon (d 106.2) also resonates upfield by comparison
with uncomplexed CH3CN in d6-acetone (d 117.6 ppm). The
1H/1H NOESY 2D spectrum (Fig. S5, ESIz) shows the
following correlations: 50-H/CH, 40-H/4-H, ArCH2S/ArH,
SCH2/2-H, and SCH2/6-H, which are indicated in Fig. 2. In
addition to these features, there is a remarkable correlation
spot between the protons of the bound CH3CN and those of
the benzene cap (ArH). All these observations allow formula-
At first glance macrobicycle 1 forms a C3v symmetric
copper(I) complex at the 1H NMR timescale, with a bound
acetonitrile molecule. However, this ancillary ligand is not
located inside the macrobicycle, otherwise the resonance of the
methyl protons would appear much more upfield. The data
also rule out the direct interaction of copper(I) with
the mesitylene cap, which would exclude acetonitrile. There are
indeed examples of competition between aryl and acetonitrile
coordination in copper(I) complexes.15 However, in the present
case, the aryl moiety is too far for binding to the metal. The
differences in the room temperature NMR properties of
[Cu(1)(CH3CN)]+ and [Cu(2)(CH3CN)]+ suggest that these
complexes are fluxional, the process being faster for the former.
Fluxionality can be observed if the CH3CN ligand is bound
tion of the complex as [Cu(1)(CH3CN)]+. The latter can be
+
formed directly in d6-acetone from 1 and Cu(CH3CN)4
.
Only one singlet for the CH3CN molecules (integrating
for 12 protons) is observed at 2.02 ppm (vs 1.90 ppm in the
case of pure [Cu(1)(CH3CN)]+ dissolved in d6-acetone), which
indicates that the bound and free acetonitrile molecules are
1
in fast exchange on the H NMR timescale. Removal of the
solvent under vacuum at 40 1C overnight left a colorless
solid, which again did not completely dissolve in d6-acetone,
but analyzed as [Cu(1)(CH3CN)]+, as shown by 1H NMR
spectroscopy.
Fig. 2 1H NMR spectrum of [Cu(1)(CH3CN)]PF6 in d6-acetone at 300 K. The arrows on the structure show NOE correlations.
ꢁc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
330 | New J. Chem., 2009, 33, 327–336

View Online
1
Fig. 3 Variable temperature H NMR spectra of [Cu(2)(CH3CN)]PF6 in d6-acetone.
outside the cavity of the macrobicycles, as represented in the inset
of Fig. 2, a hypothesis that is supported by the NOESY
experiment. Accommodation of the ancillary ligand within one
of the three portals of the macrobicycles is made possible by
deformations involving the aryl caps, these being more hindered,
and therefore less flexible in 2 by comparison with 1. The
coordination sphere of copper(I) is probably close to a deformed
tetrahedron. The resulting complexes having C2v ‘‘instantaneous’’
symmetry, the fluxional processes change it to the observed
average C3v symmetry, which involves simultaneous hopping
of copper(I) and the bound CH3CN between the three
openings of the macrobicycles (left side of Scheme 3).
This model has however to be refined in order to account for
the peculiar low temperature behavior of [Cu(2)(CH3CN)]+.
In particular, the relatively well-resolved spectrum obtained at
B240 K could be due to a species in which the coordination
sphere of copper(I) is trigonal planar. This is, for example,
possible if the Cu+ cation is bound in k2 coordination mode to
two pyrazole nitrogen atoms of the macrobicycle, leaving the
third one uncoordinated, and to the acetonitrile molecule
(right side of Scheme 3). The transition between the tetra-
hedral and trigonal geometries relieves the steric strain
associated with the former at the expense of a decrease of the
coordination number of the Cu+ cation. In both coordination
modes, decoordination of the ancillary CH3CN ligand leaves
trigonal pyramidal Cu(I) as transient intermediate.
Scheme 3 Dynamic processes occurring in the acetonitrile Cu(I)
complexes of macrobicycles 1 and 2.
The bound acetonitrile molecule can be removed to form
[Cu(1)]+ and [Cu(2)]+. This is accomplished by treatment of
[Cu(1)(CH3CN)]+ and [Cu(2)(CH3CN)]+, respectively, with
acetone, evaporation of the solvent, and heating the solid residue
in vacuum. Noticeably, whereas removal of bound CH3CN
occurs readily at 35–45 1C in the case of [Cu(2)(CH3CN)]+,
the process has to be repeated several times with heating at
100 1C in the case of [Cu(1)(CH3CN)]+. The 1H NMR spectra of
[Cu(1)]+ and [Cu(2)]+ show signals that are broadened and
shifted by comparison with those of [Cu(1)(CH3CN)]+ and
[Cu(2)(CH3CN)]+ (Table 1). Strongest shifts are experienced
by protons 2-H (ꢂ0.11 and ꢂ0.11 ppm) and CH2S (ꢂ0.10 and
ꢂ0.09 ppm) of [Cu(1)]+ and [Cu(2)]+, respectively, which are in
the vicinity of the bridging sulfur atoms. In addition, the methine
proton of 1 experiences a strong upfield shift (+0.14 ppm),
which is not observed in the case of [Cu(2)]+ (ꢂ0.01 ppm). The
1H/1H NOESY 2D NMR spectrum of [Cu(1)]+ shows the same
correlations (CH3CN/ArH excepted) as those found in the case
of [Cu(1)(CH3CN)]+. Cooling a solution of [Cu(1)]+ down to
190 K only broadens the signals, as for [Cu(1)(CH3CN)]+
1
(Fig. S7, ESIz). By contrast, the 190 K H NMR spectrum of
[Cu(2)]+ is dramatically different from that of [Cu(2)(CH3CN)]+
(Fig. S8, ESIz), as it shows a multitude of relatively sharp signals,
ꢁc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
New J. Chem., 2009, 33, 327–336 | 331

View Online
except in the region of the methylene protons (3.2–5.0 ppm).
This suggests that [Cu(2)]+ exists under different, highly un-
symmetrical conformations at this temperature. Upon warming
up, these signals gradually disappear at the expense of those
1
observed at 298 K. As noted for the VT H NMR spectra of
[Cu(2)(CH3CN)]+ transient sharpening of the signals of [Cu(2)]+
is also observed at intermediate temperature (240 K). All of
these observations suggest that CH3CN coordination in
[Cu(1)(CH3CN)]+ and [Cu(2)(CH3CN)]+ is replaced by intra-
molecular thioether coordination in [Cu(1)]+ and [Cu(2)]+,
respectively. The resulting species are highly unsymmetrical and
strained, especially in the case of [Cu(2)]+, which accounts for
the spectroscopic differences observed. In addition, the dynamic
processes that have been identified for the CH3CN adducts
(Scheme 3) certainly operate also in the case of [Cu(1)]+ and
[Cu(2)]+.
Although all manipulations were done in an oxygen-free
atmosphere, the complexes are not air-sensitive in solution, at
least for short periods, and samples for MALDI-TOF experi-
ments could be prepared as usual in the open atmosphere.
They showed a peak at m/z corresponding to the monocationic
copper(I) adduct of the ligand, a property that was reported in
earlier studies for [tris(pyrazolyl)methane]copper(I) acetonitrile
complexes.12
Fig. 4 Ball-and-stick representation of part of the helical coordina-
tion polymer in the crystal structure of [Cu(1)(H2O)]n(PF6)n viewed
along the threefold axis of the helix.
As mentioned, once dried, [Cu(1)(CH3CN)]PF6 did not show
complete solubility in acetone, even less in dichloromethane.
Moreover, white precipitates developed from standing, clarified
solutions, indicating that a chemical transformation was
taking place. Fortunately, good X-ray quality colorless crys-
tals were obtained from a filtered acetone solution of
[Cu(1)(CH3CN)]PF6 that had been left for a few days at room
temperature under argon. They belong to the centrosymmetric
ꢀ
trigonal (hexagonal axes) space group R3. Unexpectedly, the
crystals are not made from individual complex molecules, but
consist of unidimensional coordination polymers assembled
from copper(I) and macrobicycle 1 (Fig. 4). The complex
subunits of the coordination polymer are arranged helically
with a ternary periodicity, along the (001) direction of the unit
cell whose parameter (/c/ = 15.523 A) corresponds to the helix
pitch, and their mean axis makes a 21.51 angle with the helix
axis. Instead of being equally coordinated by all three pyrazole
fragments of the cage, a feature which would have placed it
inside the cavity, copper(I) is bound to only two of them
(average distance: 2.098(3) A), and is located outside, in one of
the portals of the macrobicycle (Fig. 5). This allows the metal
cation to reach the sulfur atom (distance Cuꢀ ꢀ ꢀS(12a):
2.2268(9) A) of another metallated macrobicycle, and the
formation of the coordination polymer. The remaining of
the tetrahedral coordination sphere of copper(I) is completed
by a water molecule (distance Cuꢀ ꢀ ꢀO: 2.143(2) A) that is
located roughly in the center of the cage and oriented by
OHꢀ ꢀ ꢀN and OHꢀ ꢀ ꢀp hydrogen bonding interactions to the
uncomplexed pyrazole nitrogen atom (distance Oꢀ ꢀ ꢀN(2b):
2.942(3) A) and to the mesitylene cap (distance Oꢀ ꢀ ꢀC(17):
4.806(4) A). The distance between the oxygen atom and the
centroid of the mesitylene cap is 4.747 A, slightly longer than
that observed for macrobicycle 1 (4.521 A). To our knowledge,
coordination of water to soft copper(I) is quite unusual and
represents a unique feature of this system. As a matter of fact,
Fig. 5 ORTEP view of two consecutive molecular subunits in the
X-ray crystal structure of [Cu(1)(H2O)]n(PF6)n, showing the coordina-
tion sphere of Cu(I) and the encapsulated H2O molecule. Solid lines
represent hydrogen bonds. Ellipsoids are drawn at 50% probability
level.
rare copper(I)-oxygen bonds have been observed, and it was
with carboxylate ligands or in THF adducts.16 Therefore, the
coordination polymer described above can be formulated as
[Cu(1)(H2O)]n(PF6)n.
The k2 coordination mode of macrobicycle 1 in the
coordination polymer is far from being unprecedented. It is
well known that tris(pyrazolyl)methane ligands and their
borate analogues can act as simple bidentate chelates, without
involving the third arm. Excluding square-planar complexes,
this is especially the case for sterically hindered systems, such
as [Cu{HB(3-(4-tBuPh)-5-Mepz)3}P(tBu)3], in which Cu(I) is
tricoordinate,17 or for metals whose coordination spheres are
fulfilled with strong field ligands such as CO or PMe3.18
The following scenario can be proposed for the complexation
of copper(I) by macrobicycle 1 incorporating the tris(pyrazolyl)-
methane chelate. Reaction with [Cu(CH3CN)4]PF6 produces the
ꢁc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
332 | New J. Chem., 2009, 33, 327–336

View Online
soluble complex species [Cu(1)(CH3CN)]PF6 at first, which
then slowly evolves to the insoluble coordination polymer
[Cu(1)(H2O)]n(PF6)n. In the latter system two consecutive
macrobicyclic ligands bind the bridging metal cation through
k2-N(pyrazole) and k1-S(thioether) coordination, respectively.
This places copper(I) slightly outside the cage and allows the
metal to satisfy its favored tetrahedral coordination geometry
without deformation of the macrobicyclic structure, because there
is full room for the fourth ligand (a water molecule) inside the
cage. The proposed structure of [Cu(1)(CH3CN)]PF6 (Fig. 2)
differs from that of the monomer complex described above by the
following features: (i) acetonitrile ligation, which is replaced by
thioether coordination in the polymer, and (ii) probable direct,
weak interaction of copper(I) with the pyrazole nitrogen atom
that is involved in hydrogen bonding with a water molecule in the
polymer. Ketones being known to catalyze substitution reactions
in coordination chemistry,19 and copper(I) being thiophilic, crys-
tallization of the coordination polymer [Cu(1)(H2O)]n(PF6)n from
[Cu(1)(CH3CN)]PF6 in acetone is not unrealistic. The inclusion of
a water molecule, which bridges the metal cation and a pyrazole
nitrogen atom via hydrogen bonding, could endow the copper(I)
complex fragment in the former with increased stability as
synthesized from acyclic tris(pyrazolyl)methane precursors
bearing appropriate functions. The X-ray crystal structure of
the prototypical member of this family of macrobicycles (1)
showed that the tris(pyrazolyl)methane ligand fragment is
highly preorganized for coordination to metal cations favoring
tetrahedral geometries, while the surrounding cavity provides
only limited space for an ancillary ligand. Acetone solutions
obtained after combination of 1 and [Cu(CH3CN)4]PF6
evolved to a crystalline material that could be analyzed by
X-ray crystallography. The crystals contain helical arrange-
ments of copper-bridged macrobicycles, which form coordina-
tion polymers. Unexpectedly, instead of sitting inside the cages
the copper(I) cations link two adjacent molecules by coordina-
tion to two pyrazole nitrogens and one thioether sulfur. Their
coordination sphere is completed by a water molecule
that resides inside the cage and is hydrogen bonded to the
remaining, free pyrazole nitrogen. Therefore the coordination
polymers can be formulated as [Cu(1)(H2O)]n(PF6)n. Com-
1
parison of the H NMR properties of the soluble adducts of
macrobicycles 1 and 2 with copper(I) suggests that these are
fluxional complexes, which are direct precursors of the
coordination polymers by ligand substitution: in acetone, the
compared to the latter. As shown in eqn (1), conversion of
[Cu(1)(CH3CN)]+ into [Cu(1)(H2O)]n
acetonitrile complex [Cu(1)(CH3CN)]+ is slowly converted
n+
n+
involves release of
into the insoluble coordination polymer [Cu(1)(H2O)]n
.
CH3CN with concomitant capture of water from the solvent.
Comparison of the 1H NMR spectra of a freshly prepared
solution of [Cu(1)(CH3CN)]+ with that of the solution left after
deposition of the solid materialz shows that the broad singlet of
residual H2O at 3.17 ppm in the former has vanished in the latter.
On the contrary, the sharp singlet of CH3CN at 1.90 ppm is left
unchanged, which confirms that bound and free acetonitrile
molecules are in exchange at the NMR timescale.
Water complexation and activation in transition metal
complexes featuring hydrogen bonding sites that are not
involved in metal coordination have been described.20
However the prototypical tris(pyrazolyl)methane-incorporating
macrobicycle 1 has the rare and interesting ability to bind
simultaneously with the same type of donor atoms (pyrazole
nitrogen) a metal cation (coordination bond N–Cu+) and
a water molecule (hydrogen bond Nꢀ ꢀ ꢀH(O)). This result
is an illustration of a possible consequence of squeezing the
coordination environment of a metal.
n[Cu(1)(CH3CN)]+ + nH2O - [Cu(1)(H2O)]nn+ + CH3CN
(1)
As the stability in solution of the copper(I) complexes
of macrobicycle 2 is increased as compared to those of 1,
further work will study the reactivity of [Cu(2)(CH3CN)]+
and [Cu(2)]+ with small anionic ligands and investigate
the possibility of these complexes to mimic active sites of
metalloproteins, such as nitrite reductase.21
It is also interesting to note that, in spite of the fact that H2O
could fit as axial ligand in the surrounding macrotricycle
cavity, whereas CH3CN would not, [Cu(1)(CH3CN)]+ is
formed rather than the C3v-symmetric [Cu(1)(H2O)]+. This
happens at the expense of deformations of the macrobicycle
and distortion of the coordination sphere of the metal cation.
In the scenario detailed above the formation of the coordina-
tion polymer is driven by its low solubility. Alternatively, the
complexation of copper(I) by macrobicycle 1 could be analyzed in
terms of kinetic and thermodynamic control, the kinetic product
being the soluble [Cu(1)(CH3CN)]PF6 complex, and the thermo-
dynamic product being the insoluble coordination polymer
[Cu(1)(H2O)]n(PF6)n. As noted previously, [Cu(2)(CH3CN)]PF6
gave longstanding clear solutions in d6-acetone, a feature which
would indicate increased stability of the latter with respect to
[Cu(1)(CH3CN)]PF6.
Experimental
General
1,3,5-Tribromomethylbenzene (4),22 1,3,5-tribromomethyl-
2,4,6-trimethylbenzene (5),23 and [Cu(CH3CN)4](PF6)24 were
prepared according to the literature. Unless otherwise stated
all reactions were performed under argon using standard
Schlenk techniques. Dimethylformamide was filtered on alu-
mina and stored over 3 A molecular sieves. Silica for column
chromatography was from Merck (Kieselgel 60). NMR spec-
tra were obtained using Bruker Avance 300 and DRX 500
spectrometers. Mass spectra were obtained with a Bruker
Daltonix Proflex III spectrometer (MALDI-TOF, dithranol
Summary and conclusion
Two macrobicycles (1 and 2) incorporating the tris(pyrazolyl)-
matrix). Melting points were measured with a Buchi Melting
¨
methane scorpionate ligand and closed by aryl caps have been
Point B-545 apparatus and are uncorrected. Elemental
analyses were run with an EA1108 CHNS Fisons Instrument
analyser.
z Wet acetone was employed, as checked by 1H NMR prior to
complexation studies.
ꢁc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
New J. Chem., 2009, 33, 327–336 | 333

View Online
Synthesis
Macrobicycle 1. A solution of 3 (0.210 g, 0.360 mmol) and
136.8, 133.7, 131.6, 129.3, 128.7, 127.3, 124.9, 105.5, 31.7, 16.7
(3 missing signals). m/z (MALDI-TOF) 738.02 (M + H+).
1,3,5-tribromomethylbenzene 4 (0.129 g, 0.360 mmol) in DMF
(150 cm3) was added dropwise to a stirred suspension of
K2CO3 (0.300 g, 2.16 mmol) in DMF (300 cm3) at 55 1C over
4 days. After 24 h stirring the solvent was removed in vacuo
and the residue partitioned between CH2Cl2 and water. The
organic layer was washed with brine and dried over MgSO4.
Purification of the crude product by chromatography (silica
70–230 mesh, CH2Cl2) followed by crystallization from
CH2Cl2–heptane afforded 0.070 g of 1 as colorless crystals.
Yield: 28%. mp 4 300 1C (dec). Found: C, 68.21; H, 5.12; N,
11.71; S, 12.92. Calc. for C40H34N6S3ꢀ0.5 H2O: C, 68.26; H,
5.02; N, 11.95; S, 13.64%. dH(500 MHz, CDCl3; Me4Si) 8.13
(1 H, br s, CH), 7.66 (3 H, s, 2-H), 7.66 (3 H, d, 2J 2.5, 50-H),
7.42 (3 H, d, 2J 7.6, 4-H), 7.38 (3 H, d, 2J 7.6, 6-H), 7.31 (3 H, t,
2J 7.6, 5-H), 7.03 (3 H, s, ArH), 6.63 (3 H, d, 2J 2.5, 40-H), 3.62
(6 H, s, CH2), 3.60 (6 H, s, CH2). dH(500 MHz, CD2Cl2;
Me4Si) 8.16 (1 H, s, CH), 7.73 (3 H, d, 3J 2.5, 50-H), 7.68 (3 H,
[Cu(1)(CH3CN)]PF6.
A solution of [Cu(CH3CN)4]PF6
(0.012 g, 0.0295 mmol) in CH3CN (2 cm3) was added to a
solution of 1 (0.0210 g, 0.0302 mmol) in CH2Cl2 (3 cm3). After
3 h stirring, the solvents were removed in vacuo leaving
a colorless residue which was dried overnight at 35 1C.
dH(500 MHz, CD3CN; Me4Si) 8.59 (1 H, s, CH), 8.01 (3 H,
3
3
d, J 2.6, 50-H), 7.91 (3 H, br s, 2-H), 7.51 (3 H, ddd, J 7.7,
4J 1.5, 4-H), 7.47 (3 H, ddd, J 7.7, J 1.5, 6-H), 7.36 (3 H, t,
3J 7.7, 5-H), 7.13 (3 H, s, ArH), 6.73 (3 H, d, 3J 2.6, 40-H), 3.74
(12 H, s, CH2). dH(600 MHz, d6-acetone; Me4Si) 9.41 (1 H, s,
CH), 8.44 (3 H, d, 3J 3.0, 50-H), 8.14 (3 H, br s, 2-H), 7.67 (3 H,
dt, 3J 7.8, 4-H), 7.62 (3 H, dt, 3J 7.8, 6-H), 7.46 (3 H, t, 3J 7.8,
3
4
3
5-H), 7.32 (3 H, s, ArH), 6.99 (3 H, d, J 3.0 Hz, 40-H), 3.92
(6 H, s, ArCH2S), 3.85 (6 H, s, CH2S), 1.90 (3 H, s, CH3CN).
dH(500 MHz, CD2Cl2; Me4Si) 8.82 (1 H, s, CH), 8.24 (3 H, d,
3J 2.7, 50-H), 8.05 (3 H, br s, 2-H), 7.61 (3 H, d, J 7.6, 4-H),
3
3
7.51 (3 H, d, J 7.6 Hz, 6-H), 7.43 (3 H, t, J 7.6, 5-H), 7.19
3
4
3
4
3
(3 H, s, ArH), 6.69 (3 H, d, J 2.7, 40-H), 3.83 (6 H, s,
t, J 1.5, 2-H), 7.45 (3 H, ddd, J 7.5 Hz, J 1.5, 4-H), 7.42
3
(3 H, ddd, J 7.5, J 1.5, 6-H), 7.34 (3 H, t, J 7.5, 5-H), 7.08
4
3
ArCH2S), 3.79 (6 H, s, CH2S), 1.66 (3 H, s, CH3CN).
dC(125 MHz, d6-acetone; Me4Si) 154.6, 139.7, 138.7, 135.3,
131.5, 130.8, 130.2, 128.9, 127.3, 127.2, 106.2, 78.7, 55.0, 37.0,
36.2, 3.7. m/z (MALDI-TOF) 756.82 (Cu(1)–CH3CN–PF6)+.
3
(3 H, s, ArH), 6.67 (3 H, d, J 2.5, 40-H), 3.67 (6 H, s,
ArCH2S), 3.63 (6 H, s, CH2S). dH(500 MHz, d6-acetone;
3
Me4Si) 9.02 (1 H, s, CH), 8.04 (3 H, d, J 2.5, 50-H), 7.96
3
(3 H, br s, 2-H), 7.51 (3 H, br d, J 7.7, 4-H), 7.43 (3 H, br d,
3J 7.7, 6-H), 7.32 (3 H, t, 3J 7.7, 5-H), 7.23 (3 H, s, ArH), 6.70
(3 H, d, 3J 2.5, 4’-H), 3.81 (6 H, s, ArCH2S), 3.73 (6 H, s,
CH2S). dC(75 MHz, CDCl3; Me4Si) 152.7, 138.9, 138.1, 133.5,
130.9 (50-C), 129.0 (5-C), 128.6 (6-C), 128.4 (Ar-C), 127.4
(2-C), 124.8 (4-C), 105.1 (40-C), 81.9 (CH), 36.4 (CS) 36.3
(CS); dC(125 MHz, CD2Cl2; Me4Si) 152.8 (30-C), 139.3
(ArC-C), 138.4 (1-C), 135.6 (3-C), 131.1 (50-C), 129.0 (5-C),
128.9 (6-C), 128.3 (ArH-C), 127.4 (2-C), 124.9 (4-C), 105.1
(40-C), 81.8 (CH), 36.7 (ArCH2S), 36.4 (CH2S). dC(125 MHz,
d6-acetone; Me4Si) 152.7, 140.4, 138.6, 134.8, 132.4, 129.5,
129.5, 128.7, 127.6, 125.4, 104.5, 81.1, 37.6, 37.1. m/z
(MALDI-TOF) 694.87 (M+).
[Cu(1)(H2O)]ꢀPF6ꢀ1/3C3H6Oꢀ1/9H2O. [Cu(1)(CH3CN)]PF6
was dissolved in d6-acetone, leaving a colorless residue that
was discarded. The filtrate was placed in an NMR tube and
stored at room temperature. Needle-like crystals suitable for
X-ray diffraction appeared after 5 days.
[Cu(2)(CH3CN)]PF6.
A solution of [Cu(CH3CN)4]PF6
(0.0110 g, 0.0295 mmol) in CH3CN (2 cm3) was added to a
suspension of 2 (0.0218 g, 0.0296 mmol) in CH2Cl2 (4 cm3).
After 3 h stirring, the solvents were removed in vacuo leaving a
colorless residue, which was dried overnight at 35 1C.
dH(500 MHz, d6-acetone, 298 K; Me4Si) 9.40 (1 H, s, CH),
8.45 (3 H, br s, 50-H), 8.17 (3 H, br s, 2-H), 7.63 (3 H, br s,
4-H), 7.50 (3 H, br s, 6-H), 7.39 (3 H, br s, 5-H), 7.01 (3 H, s,
40-H), 4.09 (6 H, br s, ArCH2S), 3.91 (6 H, br s, CH2), 2.45
(9 H, br s, CH3), 2.03 (3 H, s, CH3CN). m/z (MALDI-TOF)
798.08 (Cu(2)–CH3CN–PF6)+.
Macrobicycle 2. A solution of 3 (0.200 g, 0.344 mmol)
and 1,3,5-tribromomethyl-2,4,6-trimethylbenzene 5 (0.142 g,
0.356 mmol) in DMF (50 cm3) was added dropwise to a stirred
suspension of K2CO3 (0.296 g, 2.142 mmol) in DMF (150 cm3)
at 60 1C over 8 h. After overnight stirring the solvent was
removed in vacuo. Purification of the crude product by flash
column chromatography (silica 230–400 mesh, 2% AcOEt in
CH2Cl2) afforded 2 (0.090 g) as a colorless solid. Yield: 32%.
mp 272 1C (dec). Found: C, 68.57; H, 5.76; N, 10.94; S 11.96.
Calcd for C43H40N6S3ꢀH2O: C, 68.40; H, 5.60; N, 11.13; S,
12.74%. dH(500 MHz, CDCl3; Me4Si) 8.17 (1 H, s, CH), 7.73
[Cu(1)]PF6. Mixing [Cu(CH3CN)4]PF6 (0.00270 g, 7.22 ꢃ
10ꢂ6 mol) and 1 (0.00502 g, 7.22 ꢃ 10ꢂ6 mol) in d6-acetone
produced a solution, which was evaporated in vacuo. Further
drying at 45 1C for 22 h afforded [Cu(1)(CH3CN)]PF6
(1H NMR in d6-acetone). The same procedure was repeated
three times at 100 1C to leave [Cu(1)]PF6 along with insoluble
matter. dH(500 MHz, d6-acetone, 298 K; Me4Si) 9.55 (1 H, s,
CH), 8.46 (3 H, s, 50-H), 8.03 (3 H, s, 2-H), 7.62 (3 H, d, 3J 7.5,
4-H), 7.57 (3 H, d, 3J 7.5, 6-H), 7.43 (3 H, t, 3J 7.5, 5-H), 7.29
3
(3 H, d, J 2.5, 50-H), 7.58 (3 H, br s, 2-H), 7.43 (3 H, ddd,
4
3
4
3J 7.5, J 1.5, 4-H), 7.32 (3 H, ddd, J 7.5, J 1.5, 6-H), 7.29
3
(3 H, s, ArH), 6.97 (3 H, d, J 2.5, 40-H), 3.90 (6 H, s,
3
3
(3 H, t, J 7.5, 5-H), 6.64 (3 H, d, J 2.5, 40-H), 3.73 (6 H, s,
ArCH2S), 3.75 (6 H, s, CH2).
CH2), 3.52 (6 H, s, CH2), 2.04 (3 H, s, CH3). dH(500 MHz,
3
CD2Cl2; Me4Si) 8.23 (1 H, s, CH), 7.78 (3 H, d, J 2.5, 50-H),
[Cu(2)]PF6. Mixing [Cu(CH3CN)4]PF6 (0.00255 g, 6.84 ꢃ
10ꢂ6 mol) and 1 (0.00495 g, 6.72 ꢃ 10ꢂ6 mol) in d6-acetone
produced a solution, which was evaporated in vacuo. Further
drying at 35 1C for 18 h afforded [Cu(2)(CH3CN)]PF6
(1H NMR in d6-acetone). The same procedure was repeated
3
7.52 (3 H, br s, 2-H), 7.60 (3 H, ddd, J 7.5, 4J 1.6, 4-H), 7.36
(3 H, ddd, 3J 7.5, 6-H), 7.31 (3 H, t, 3J 7.5, 5-H), 6.71 (3 H, d,
3J 2.5, 40-H), 3.73 (6 H, s, ArCH2S), 3.50 (6 H, s, CH2S), 2.01
(3 H, s, CH3). dC(125 MHz, CD2Cl2; Me4Si) 152.6, 139.8,
ꢁc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
334 | New J. Chem., 2009, 33, 327–336

View Online
once at 35 1C overnight to leave [Cu(2)]PF6 quantitatively.
dH(500 MHz, d6-acetone, 298 K; Me4Si) 9.39 (1 H, br s, CH),
8.45 (3 H, br s, 50-H), 8.06 (3 H, s, 2-H), 7.60 (3 H, br s, 4-H),
7.45 (3 H, br s, 6-H), 7.35 (3 H, br s, 5-H), 6.98 (3 H, br s,
40-H), 4.07 (6 H, br s, ArCH2S), 3.82 (6 H, br s, CH2), 2.42
(9 H, br s, CH3).
factors of 0.5. The two highest residuals (1.54 and 1.06 e Aꢂ3
)
appear as a partial unachieved modelling of the disorder close
to the position of two sulphur atoms. In order to drive the
refinement towards a meaningful model, 383 restraints were
used. They were applied to all disordered parts of the
molecular structure and concern bonding distances (DFIX
instruction), flatness of phenyl rings (FLAT instruction), rigid
bond and isotropic restraints on the anisotropic displacement
parameters (DELU and ISOR instructions).
X-Ray crystallography
Colorless single-crystal specimens of prismatic shapes of
21ꢀ1.5H2Oꢀ0.5CH2Cl2ꢀ0.5C2H6O and [Cu(1)(H2O)]ꢀPF6ꢀ1/
3C3H6Oꢀ1/9H2O were selected for the X-ray diffraction
experiments at T = 110(2) and 115(2) K, respectively. They
were mounted with silicon grease on the tips of glass capillaries.
Diffraction data were collected on a Nonius KappaCCD
diffractometer25 equipped with a nitrogen jet stream low-
temperature system (Oxford Cryosystems). The X-ray
source was graphite-monochromated Mo-Ka radiation
(m = 0.71073 A) from a sealed tube. In both cases, lattice
parameters were obtained by a least-squares fit to the
optimized setting angles of the collected reflections. Intensity
data were recorded as j- and o-scans with k offsets. No
significant intensity decay or temperature drift was observed
during the data collections. Data were reduced by using
DENZO software26 without applying absorption corrections,
the missing absorption corrections being partially com-
pensated by the data scaling procedure in the data reduction.
The structures were solved by direct methods using the SIR97
program.27 Refinements were carried out by full-matrix least-
squares on F2 using the SHELXL97 program28 and the
complete set of reflections. Anisotropic thermal parameters
were used for non-hydrogen atoms. In both crystal structure
determinations, most of the hydrogen atoms were located in
the Fourier synthesis. They were placed at calculated positions
using a riding model, except one of the water molecule
encapsulated by 1 and those of the water molecule in the
coordination sphere of the metal center, which were, respec-
tively, constrained and restrained to the mean Ow–H bond
distance (B0.98 A) observed from neutron experiments.29 The
isotropic thermal factors of all hydrogen atoms were fixed at
1.3 times the equivalent isotropic displacement parameter of
the corresponding bonded atom. The hydrogen atoms of a
water molecule with a site occupation factor of 1/9 in the
copper complex crystal structure, and another one of the
disordered water molecule encapsulated by 1, were not found.
Compound 21ꢀ1.5H2Oꢀ0.5CH2Cl2ꢀ0.5C2H6O crystallizes in
Compound [Cu(1)(H2O)]ꢀPF6ꢀ1/3C3H6Oꢀ1/9H2O crystallizes
ꢀ
in the trigonal system (space group R3, hexagonal axes). The
asymmetric unitꢂconsists of one copper complex [Cu(1)]+, one
counter-ion PF6 and co-crystallized water and acetone mole-
cules. One water molecule is found in the coordination sphere of
the metal center, which is tetrahedral. Two remaining C3H6O
and H2O solvent molecules were refined with site occupation
factors of 1/3 and 1/9, the acetone being disordered around a
threefold inversion axis at (0, 0, 0) and the water molecule
occupying a special position on this axis at (0, 0, z). The
refinement of the water molecule with this low site occupation
factor (1/3 lower than the site symmetry) was based on
the residual initially observed at the oxygen atom position
(1.83 e Aꢂ3) and on the refinement of its anisotropic displace-
ment parameters. In addition, even if higher residuals (not
necessarily corresponding to any atom) are expected on special
positions, the presence of a solvent molecule is also presumed
based on large accessible regions in the unit cell. Indeed, even
after determination of the crystal structure, voids of 217 A3 are
observed close to the regions occupied by this water molecule. A
careful examination of the residual density seems to indicate no
further molecules. However, the consideration of another dis-
ordered solvent molecule instead of water cannot be completely
excluded due to the presence of several diffuse residuals in the
same region (the two highest residuals, 0.90 and 0.69 e Aꢂ3
,
were found out of the threefold inversion axis at 1.11 and
2.30 A from the oxygen atom of this water molecule). The
m-phenylene spacers of two arms of the macrobicycle were
found partially disordered (each one over two close positions)
and they were refined with sites occupation factors of 0.6/0.4
and 0.7/0.3. In order to drive the refinement towards a
meaningful model, 201 restraints were used. They were applied
to all disordered parts of the molecular structure and concern
bonding distances (DFIX instruction), flatness of phenyl rings
(FLAT instruction), rigid bond and isotropic restraints on
the anisotropic displacement parameters (DELU and ISOR
instructions).
ꢀ
the triclinic system (space group P1). The asymmetric unit
Relevant experimental crystal data and refinement details
are summarized in Table S2 (ESIz). Crystallographic views
were generated using ORTEP III for windows.30
consists of two non-equivalent macrobicycles 1, one non-
disordered and one disordered, as well as co-crystallized water,
dichloromethane and ethanol solvent molecules. The disorder
of the macrobicycle is observed over two close positions
(site occupation factors 0.5/0.5) and involves one of the three
m-phenylene spacers and the benzene cap. This macrobicycle
encapsulates a disordered water molecule over two positions,
one inside and one outside 1, exhibiting site occupation factors
of 0.5/0.5. The dichloromethane molecule was found close to
the disordered macrobicycle, sharing its occupied region with
a couple of ethanol and water molecules. These CH2Cl2,
H2O and C2H6O molecules were refined with site occupation
References
1 K. Maruoka, A. B. Conception, N. Hirayama and H. Yamamoto,
J. Am. Chem. Soc., 1990, 112, 7422–7423; M. Ray, G. P. A. Yap,
A. L. Rheingold and A. S. Borovik, J. Chem. Soc., Chem.
Commun., 1995, 1777–1778; M. H. Chisholm and N. W. Eilerts,
Chem. Commun., 1996, 853–854; M. Ray, B. S. Hammes, G. P.
A. Yap, A. L. Rheingold and A. S. Borovik, Inorg. Chem., 1998,
37, 1527–1532; J. R. Hagadorn, L. Que, Jr. and W. B. Tolman,
J. Am. Chem. Soc., 1998, 120, 13531–13532; X. Hu, I. Castro-
Rodriguez and K. Meyer, Chem. Commun., 2004, 2164–2165;
ꢁc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
New J. Chem., 2009, 33, 327–336 | 335

View Online
I. Nieto, F. Cervantes-Lee and J. M. Smith, Chem. Commun., 2005,
3811–3813; C. Zimmermann, F. W. Heinemann and
6331–6332; S. Yokota, Y. Tachi, N. Nishiwaki, M. Ariga and
S. Itoh, Inorg. Chem., 2001, 40, 5316–5317; F.-B. Xu, Q.-S. Li,
L.-Z. Wu, X.-B. Leng, Z.-C. Li, X.-S. Zeng, Y. L. Chow and
Z.-Z. Zhang, Organometallics, 2003, 22, 633–640; B. A.
Gandhi, O. Green and J. N. Burstyn, Inorg. Chem., 2007, 46,
3816–3825.
A. Grohmann, Eur. J. Inorg. Chem., 2005, 3506–3512; U. Lunig
¨
and F. Fahrenkrug, Eur. J. Org. Chem., 2006, 916–923.
2 R. R. Schrock, Acc. Chem. Res., 2005, 38, 955–962; P. L. Holland,
Can. J. Chem., 2005, 83, 296–301.
3 Early examples: J. P. Collman, R. R. Gagne, C. A. Reed,
T. R. Halbert, G. Lang and W. T. Robinson, J. Am. Chem. Soc.,
1974, 97, 1427–1439; A. R. Battersby, D. G. Buckley, S. G. Hartley
and M. D. Turnbull, J. Chem. Soc., Chem. Commun., 1976,
879–881; J. E. Baldwin, T. Klose and M. Peters, J. Chem. Soc.,
Chem. Commun., 1976, 881–883; M. Momenteau, B. Loock,
J. Mispelter and E. Bisagni, Nouv. J. Chim., 1979, 3, 77–79.
4 D. H. Camacho, E. V. Salo, J. W. Ziller and Z. Guan, Angew.
Chem., Int. Ed., 2004, 43, 1821–1825.
15 T. Osako, Y. Ueno, Y. Tachi and S. Itoh, Inorg. Chem., 2003, 42,
8087–8097.
16 J. R. Hagadorn, T. I. Zahn, L. Que, Jr and W. B. Tolman, J. Chem.
Soc., Dalton Trans., 2003, 1790–1794.
17 R. R. Conry, G. Ji and A. A. Tipton, Inorg. Chem., 1999, 38,
906–913.
18 A. Macchioni, G. Bellachioma, G. Cardaci, V. Gramlich,
H. Ruegger, S. Terenzi and L. M. Venanzi, Organometallics,
¨
1997, 16, 2139–2145.
5 K. Maruszewski, D. P. Strommen and J. R. Kincaid, J. Am. Chem.
Soc., 1993, 115, 8345–8350.
6 M. Yoshizawa and M. Fujita, Pure Appl. Chem., 2005, 77,
1107–1112; J. Rebek Jr, Angew. Chem., Int. Ed., 2005, 44,
2068–2078; D. Fiedler, D. H. Leung, R. G. Bergman and
K. N. Raymond, Acc. Chem. Res., 2005, 38, 349–358.
19 E. Engeldinger, D. Armspach, D. Matt and P. G. Jones,
Chem.–Eur. J., 2003, 9, 3091–3105.
´
20 J. C. Mareque-Rivas, R. Prabaharan and R. Torres Martın de
Rosales, Chem. Commun., 2004, 76–77.
21 M. Kukimoto, M. Nishiyama, M. E. P. Murphy, S. Turley,
E. T. Adman, S. Horinouchi and T. Beppu, Biochemistry, 1994,
33, 5246–5252; H. J. Wijma, L. J. C. Jeuken, M. P. Verbeet,
F. A. Armstrong and G. W. Canters, J. Biol. Chem., 2006, 281,
16340–16346 and references therein.
7 S. Trofimenko, Chem. Rev., 1993, 93, 943–980.
8 N. Kitajima, K. Fujisawa and Y. Moro-oka, J. Am. Chem. Soc.,
1990, 112, 3210–3212; N. Kitajima, K. Fujisawa, M. Tanaka and
Y. Moro-oka, J. Am. Chem. Soc., 1992, 114, 9232–9233; D. Qiu,
L. Kilpatrick, N. Kitajima and T. G. Spiro, J. Am. Chem. Soc.,
1994, 116, 2585–2590.
22 E. Dı
J. Rodrı
Chem., 2001, 66, 5664–5670.
´
ez-Barra, J. C. Garcı
´
a-Martı
´
nez, S. Merino, R. del Rey,
and J. Tejeda, J. Org.
´
guez-Lopez, P. Sanchez-Verdu
´
´
´
9 R. Alsfasser, S. Trofimenko, A. Looney, G. Parkin and
H. Vahrenkamp, Inorg. Chem., 1991, 30, 4098–4100; A. Looney,
G. Parkin, R. Alsfasser, M. Ruf and H. Vahrenkamp, Angew.
Chem., Int. Ed. Engl., 1992, 31, 92–93; R. Alsfasser, M. Ruf,
S. Trofimenko and H. Vahrenkamp, Chem. Ber., 1993, 126,
703–710; A. Looney, R. Han, K. McNeill and G. Parkin, J. Am.
Chem. Soc., 1993, 115, 4690–4697.
23 S. K. Armstrong, S. Clunas and K. W. Muir, Synthesis, 1999,
993–998.
24 G. J. Kubas, Inorg. Syntheses, 1990, 28, 90–92.
25 B. Nonius, in COLLECT, Data Collection Software, BV Nonius,
Delft, The Netherlands, 1998.
26 Z. Otwinowski and W. Minor, Methods Enzymol., 1997, 276,
306–326.
10 M. C. Keyes, B. M. Chamberlain, S. A. Caltagirone, J. A. Halfen
and W. B. Tolman, Organometallics, 1998, 17, 1984–1992.
11 D. L. Reger, J. E. Collins, A. L. Rheingold, L. M. Liable-Sands
and G. P. A. Yap, Organometallics, 1997, 16, 349–353.
12 D. L. Reger, J. E. Collins, A. L. Rheingold and L. M. Liable-Sands,
Organometallics, 1996, 15, 2029–2032.
27 A. Altomare, M. C. Burla, M. Camalli, G. L. Cascarano,
C. Giacovazzo, A. Guagliardi, A. G. G. Moliterni, G. Polidori
and R. Spagna, J. Appl. Crystallogr., 1999, 32, 115–119.
28 G. M. Sheldrick, in SHELXS-97, Program for the Solution of
Crystal Structures, University of Gottingen, Gottingen, Germany,
1997.
¨
¨
13 L. Wang and J.-C. Chambron, Org. Lett., 2004, 6, 747–750.
14 A. H. Lewin, R. J. Michl, P. Ganis and U. Lepore, J. Chem. Soc.,
Chem. Commun., 1972, 661–662; P. Roussel and P. Scott, J. Am.
Chem. Soc., 1998, 120, 1070–1071; B. A. Jazdzewski, V. G. Young,
Jr. and W. B. Tolman, Chem. Commun., 1998, 2521–2522;
P. L. Holland and W. B. Tolman, J. Am. Chem. Soc., 2000, 122,
29 E. Espinosa, C. Lecomte, E. Molins, S. Veintemillas, A. Cousson
and W. Paulus, Acta Crystallogr., Sect. B, 1996, 52, 519–534;
I. Mata, E. Espinosa, E. Molins, S. Veintemillas,
W. Maniukiewicz, C. Lecomte, A. Cousson and W. Paulus, Acta
Crystallogr., Sect. A, 2006, 62, 365–378.
30 L. J. Farrugia, J. Appl. Crystallogr., 1999, 32, 837–838.
ꢁc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
336 | New J. Chem., 2009, 33, 327–336