Copper(I) templated synthesis of a 2,2-bipyridine derived 2-catenane: Synthetic, modelling, and X-ray studies

Article abstract of DOI:10.1071/CH09253

Following molecular modelling employing molecular mechanics and semi-empirical (PM3) calculations as an aid to ligand design, a new copper(i)-containing [2]-catenane has been prepared by the reaction of 6,6′-bis[(4″-formylphenoxy)methyl]-2,2′-bipyridyl wi

Full text of DOI:10.1071/CH09253

RESEARCH FRONT
Full Paper
CSIRO PUBLISHING
Aust. J. Chem. 2009, 62, 1014–1019
www.publish.csiro.au/journals/ajc
Copper(I) Templated Synthesis of a 2,2ꢀ-Bipyridine Derived
2-Catenane: Synthetic, Modelling, and X-ray Studies∗
Jason R. Price,A Jack K. Clegg,A Ronald R. Fenton,A Leonard F. Lindoy,A,E,§
John C. McMurtrie,B George V. Meehan,C Andrew Parkin,D
David Perkins,A and Peter TurnerA
ASchool of Chemistry, The University of Sydney, NSW 2006, Australia.
BSchool of Physical and Chemical Sciences, Queensland University of Technology,
GPO Box 2434, Brisbane, Qld 4001, Australia.
CSchool of Pharmacy and Molecular Sciences, James Cook University, Townsville,
Qld 4811, Australia.
DDepartment of Chemistry, University of Glasgow, G12 8QQ, UK.
ECorresponding author. Email: lindoy@chem.usyd.edu.au
§Visiting Professor, Department of Chemistry, National University of Singapore, 2008.
Following molecular modelling employing molecular mechanics and semi-empirical (PM3) calculations as an aid to ligand
design, a new copper(i)-containing [2]-catenane has been prepared by the reaction of 6,6ꢀ-bis[(4ꢀꢀ-formylphenoxy)methyl]-
2,2ꢀ-bipyridyl with 1,6-hexanediamine in the presence of a copper(i) template, followed by in situ sodium borohydride
reduction of the tetra-imine species generated. The corresponding [2]-catenane was obtained as a deep orange crystalline
solid. The X-ray structure of this product along with that of the corresponding bis-dialdehyde precursor copper(i) complex
have been determined. In the latter case, the structure is pre-organized for catenane formation. The new catenane appears
to be the first example of a tetrahedral metal template product incorporating a bis-bipyridyl derivative core. Isolation of the
metal-free [2]-catenane has also been achieved following treatment of the copper(i) catenane complex in dimethylsulfoxide
with tetrabutylammonium cyanide.
Manuscript received: 28 April 2009.
Manuscript accepted: 26 May 2009.
Introduction
been synthesized using zinc(ii) as a five coordinate core and a
mixed-ligand combination of tridentate (2,2ꢀ:6ꢀ,2ꢀꢀ-terpyridine)
and bidentate (2,9-disubstituted 1,10-phenanthroline) deriva-
tives occupying the coordination sphere.[7] Catenanes with
dimetallic cores have also been achieved using two copper(i)
metal centres.[8]
The use of ligand systems designed to allow secondary bond-
ing interactions to assist in the pre-organization of the mutual
embrace for catenane synthesis gives rise to the prospect of uti-
lizing more flexible precursor ligands and hence to widen the
scope of such syntheses in terms of both the form and function
of the resulting catenanes that might be generated.
In the present study, we have investigated the use of the
relatively flexible 2,2ꢀ-bipyridyl dialdehyde 2 (Scheme 1) as
a precursor ligand for catenane formation, with 2 designed to
promote favourable π–π stacking interactions with a second lig-
and when coordinated in an orthogonal embrace to a tetrahedral
metal ion (copper(i) in the present study).
The design and synthesis of mechanically linked molecular
systems, of which [2]-catenanes and [2]-rotaxanes are the sim-
plest examples,[1] continues to be a rapidly expanding area of
supramolecular chemistry. To a large degree, the current intense
interest in such systems is due to their potential for incorpo-
ration as functional elements into a wide variety of nano-scale
molecular machines and devices.[2]
The use of a metal template strategy has been a continuing
theme in the formation of many new catenane derivatives.[3] In
one commonly used procedure, the metal ion directs mutually
orthogonal coordination of two appropriately functionalized lig-
ands, with this being followed by double ring closure to produce
the corresponding metal-containing [2]-catenane. Several such
metal templated syntheses, the first example being reported by
Sauvage et al. for formation of a copper(i) catenane,[4] have
been based on use of rigid 2,9-disubstituted, 1,10-phenantholine
derivatives of type 1. The use of mutually orthogonal embraces
have also been demonstrated for other ligand types as well as for
different metal ion templates; for example, an octahedral metal
centre can be used in conjunction with two essentially rigid tri-
dentate ligands, such as 2,2ꢀ:6ꢀ,2ꢀꢀ-terpyridine derivatives[5] or
pyridine diimine ligands.[6] Mixed ligand catenanes have also
Results and Discussion
Modelling Studies
Initially, a series of preliminary molecular modelling studies
was undertaken as an aid to catenane ligand design. Inspection
∗Dedicated to the memory of our co-author, the late Dr Andrew Parkin, a fine young chemist whose untimely passing is a great loss to us all.
© CSIRO 2009 10.1071/CH09253 0004-9425/09/091014

RESEARCH FRONT
Copper(i) Templated Synthesis of a 2,2ꢀ-Bipyridine Derived 2-Catenane
1015
60
50
N
N
O
40
30
O
N
N
20
10
R
R
0
10
20
30
O
O
3
4
5
6
7
2
1
I
I
Catenane favoured
Scheme 1.
40
Length of alkyl chain (n)
(a)
Fig. 2. Comparison of the relative computed energy values for corre-
sponding isomeric diimine catenated and spiral macrocyclic complexes on
variation of the alkyl chain length between the imine groups.
macrocycle in preference to the required catenane. The result-
ing 2:2 macrocyclic complex was expected to adopt an extended
spiral geometry. The calculated relative energies for this and
the ‘pre-organized for catenane formation’conformer discussed
above indicate that the latter structure (Fig. 1a) is 36 kJ mol−1
lower in energy than the isomer given by Fig. 1b. While it is
realized that in solution there will likely exist a dynamic equilib-
rium between these (and possibly other) conformers, the relative
energies of the present two conformers clearly suggest that
the dominant species will be that predicted to favour catenane
formation. It is noted that both the above minimum energy con-
formations have aromatic rings suitably aligned for significant
face-to-faceπ-stackingtooccur(seeFig. 1)andthisundoubtedly
contributes to the overall stabilities of both structures.
(b)
As mentioned above, Schiff base ring closure of the dialde-
hyde components 2 in [Cu(2)2]+ was planned to involve reaction
with a flexible diamine of suitable length. Further modelling
studies were undertaken in order to probe the ideal dimensions
for such a diamine. The energy minimized conformers shown
in Figs 1a and 1b were used as the basis for constructing the
starting structures for the corresponding Schiff base products
formed on condensation with the series of linear diamines of
type H2N(CH2)nNH2 (where n = 3–7) in both their catenane
and spiral forms. The computations were again carried out as
described above.A comparison of the results from the respective
semi-empirical (PM3) calculations is illustrated schematically
in Fig. 2. When n = 3, the spiral 2:2 macrocyclic Schiff base
complex is of lower energy than the corresponding (isomeric)
catenane, while for n = 4–7 the calculations indicate that for-
mation of the catenane is favoured over the spiral form. The
resultsalsopredictthattheformationofthecatenanewillbemost
favoured when the bridge between the imine groups corresponds
to around six methylene groups.
Fig. 1. (a) Energy minimized conformer of [Cu(2)2]+ pre-organized for
catenane formation. (b) Energy minimized conformer of [Cu(2)2]+ with a
spiral form. Both figures illustrate how the aromatic rings in each case can
give rise to offset face-to-face π-stacking interactions.
of molecular models indicated that the 4-hydroxybenzaldehyde
derivative 2, rather than its isomers derived from 2- or 3-
hydroxybenzaldehyde, appeared most promising for catenane
formation and this isomer was thus chosen for further investiga-
tion. Models of different conformers of the tetrahedral [Cu(2)2]+
complexwerethenconstructedusingSpartan,[9] andtheirenergy
minimizations were carried out using molecular mechanics
(MMFF force field).The minimized structures were then used as
the basis for single point, semi-empirical calculations (PM3) and
the relative energies of the different conformers obtained. These
results, strongly suggested that the conformer shown in Fig. 1a
displays favourable pre-organization for catenane formation,
with its aldehyde groups so positioned to facilitate ring closure
(via Schiff base condensation with an appropriate diamine).
In contrast to the above structure, a second minimum energy
conformer (Fig. 1b) has pairs of aldehyde groups originating
from different ligands orientated in close proximity so that ring
closure would more likely favour the formation of a large 2:2
Synthetic and X-ray Studies
The synthesis of dialdehyde 2 involved condensation of 6,6ꢀ-
bis(chloromethyl)-2,2ꢀ-bipyridine[10] with 4-hydroxybenzalde-
hyde in a 1:2 ratio in dimethylformamide in the presence of
NaHCO3. Addition of two molar equivalents of 2 to an aceto-
nitrile solution of tetrakis(acetonitrile)copper(i) hexafluoro-
phosphate led to the formation of dark red crystals on allowing
the reaction solution to stand. The microanalysis and 1H NMR
spectrum of this product were in agreement with its formulation
as a bis-ligand complex of type [Cu(2)2]PF6.

RESEARCH FRONT
1016
J. R. Price et al.
the diamines employed have incorporated only two or three
methylene groups linking the amine groups.
High dilution conditions were employed for the reaction
of 1,6-hexanediamine with [Cu(2)2](PF6) in methanol. The
course of the reaction was monitored by mass spectrometry-
electrospray ionization (MS-ESI) (aliquots of the reaction solu-
tion were periodically removed and subjected to mass spectral
analysis); the results showed the formation of a peak at m/z
1071.47 corresponding to that expected for the required tetra-
imine catenane species, [Cu(3)](PF6). The reaction behaviour
was probed further in a separate (parallel) Schiff base conden-
sation experiment conducted on an NMR scale. To 10 mL of
CD3OD containing 10 mg of [Cu(2)2](PF6) was added a stoi-
chiometric amount of 1,6-hexanediamine in CD3OD; the NMR
spectra of successive aliquots from this solution showed the rise
of an imine proton peak at 9.02 ppm (relative toTMS).The Schiff
basereactionappearedtoproceedtocompletion–asindicatedby
the disappearance of the aldehyde proton signal at 9.62 ppm after
1 h. An ESI-MS of this solution showed a peak corresponding to
the required tetra-imine catenane species, [Cu(3)]+.
Despite the above results, attempts to isolate a pure prod-
uct from the main reaction mixture were unsuccessful, perhaps
reflecting the well established susceptibility of imine groups to
undergo hydrolytic or solvolytic cleavage[12] (perhaps occurring
in the present case during the attempted isolation process). In
view of this, the generation of a more robust catenane deriva-
tive was attempted through performing a sodium borohydride
reduction step directly on the above reaction solution in order
to reduce the imine groups in situ to their corresponding sec-
ondary amines. Following completion of the borohydride step,
the reaction mixture was subjected to MS-ESI analysis which
indicated that complete reduction of all four imine groups had
occurred. The procedure resulted in a crude orange microcrys-
talline product which was then dissolved in aqueous isopropanol.
On standing the solution deposited a small quantity of deep
orange crystals which, however, were isolated in a disappoint-
ingly low yield of 7%. The ESI-MS spectrum of this product
showed a strong parent ion peak at m/z 1080.1 (see Accessory
Publication Fig. AP2 (a)) in agreement with the presence of the
target tetra-amine containing catenane species, [Cu(4)]+. Micro-
analysis of the dried product gave a composition corresponding
to [Cu(4)]PF6·5H2O. Treatment of a solution of this product
in aqueous propanol with potassium cyanide led, after workup,
to a product whose mass spectrum showed the major peak at
m/z 1056.1, corresponding to the potassium salt of catenane 4
(see Accessory Publication Fig. AP2 (b)); gratifyingly, no peak
was present with an m/z value corresponding to the presence of
a single free macrocycle. Substitution of tetrabutylammonium
cyanide[4] for potassium cyanide in the above procedure gener-
ated a product yielding a mass spectral peak (HRMS-ESI: LH+,
m/z 1017.571) corresponding to the metal-free catenane (4).
Of course, it needs to be noted that while collectively the
above evidence is in accord with formation of the targeted
catenane, it is non-specific in that the results also agree with
formation of the isomeric ‘double ring’ spiral complex ((b) in
Scheme 2). However, a further exploratory experiment mitigated
against formation of the latter species. In this, a small sam-
ple of the [Cu(4)]PF6·5H2O product was dissolved in aqueous
methanol and the pH was adjusted to <1 using hydrochloric acid.
This solution suffered no loss of colour compared with a control
sample at neutral pH, even after a period of weeks, in keep-
ing with the expected kinetic inertness of the target catenane.[4]
Such a degree of inertnessisnot expected for a non-mechanically
Fig. 3. X-ray structure of the complex cation in the copper(i) dialdehyde
complex, [Cu(2)2]PF6·MeCN, showing how the aromatic ring systems are
arranged for offset face-to-face π-stacking.
The X-ray structure of this complex was determined (Fig. 3)
and the result confirmed the above stoichiometry. As expected,
the copper is coordinated in a distorted tetrahedral fashion by
four heterocyclic nitrogens from the two bipyridine derivatives,
with theplanesofthebipyridine units lying at ∼73◦ toeach other.
In accord with the expectations from the modelling studies, the
structurecorrespondstotheconformershowninFig. 1a. Namely,
the structure shows the aldehyde groups most favourably posi-
tioned for catenane formation, with the shortest interatomic
distance between the two aldehyde oxygens of the same lig-
and being 8.61 Å, while the shortest distance between aldehyde
groups in different ligands is 13.1 Å. Also as predicted, each
‘benzaldehyde’ring is appropriately positioned for π-interaction
in an offset face-to-face manner with a coordinated bipyridine
fragment. It is noted that the π–π separations for the respective
benzaldehyde groups belonging to each ligand differ; the dis-
tances between the respective aryl ring centroids and the closest
atom in an adjacent bipyridine moiety are 3.46 Å and 3.75 Å.
The structure closely parallels that obtained from the mod-
elling studies (compare Fig. 1a) although some differences
between the X-ray structure and the energy minimized (gas
phase) structure are apparent. For example, in the modelled
structure the mutual orientation of the bipyridine rings between
bound ligands is orthogonal, not skewed from 90◦ as observed
in the crystal structure. Also, the di-aldehyde ligands within the
modelled structure are related by S4 symmetry, whereas for the
crystal structure the symmetry is C2.
Confirmation that the solid state configuration of the above
copper-containing precursor was that predicted to favour cate-
nane formation, gave us some confidence that this structure
might also persist as a significant (and perhaps dominant) con-
former in solution. Based on the results of our modelling studies,
1,6-hexanediamine was chosen as the double Schiff base ring-
closing reagent for reaction with the tetra-aldehyde [Cu(2)2]+.
In this context it is noted that while related double Schiff base
condensations represent a common procedure for achieving ring
closure in macrocyclic chemistry,[11] in the majority of cases

RESEARCH FRONT
Copper(i) Templated Synthesis of a 2,2ꢀ-Bipyridine Derived 2-Catenane
1017
ꢁ
O
O
O
O
R
N
N
Cu
N
ꢁ
NH2
NH2
N
O
(where R ꢂ –(CH2)n–)
O
O
O
(a)
(b)
ꢁ
N
ꢁ
N
N
N
O
O
N
N
N
O
O
O
O
N
N
Cu
R
R
R
R
Cu
N
N
N
O
O
N
N
N
N
Scheme 2. Two of the products that may arise from ring closure via Schiff base condensation using a linear
aliphatic diamine of appropriate length: (a) the catenane and (b) the corresponding spiral 2:2 macrocyclic complex.
A third possibility also exists in which two non-interlinked macrocycles would be generated but initial modelling
studies suggested that this possibility will not be favoured.
Conclusion
In this manuscript we report the template synthesis, properties,
and X-ray structure of a new copper(i) catenane incorporat-
ing, to the best of our knowledge, the first example based on
a bis(bipyridyl)copper(i) core. Our results demonstrate the pref-
erential generation, albeit in low yield, of this species over
the other two possible isomeric structures incorporating non-
catenane species. Isolation of the metal-free [2]-catenane has
also been achieved, opening the way to the future synthesis of
other metal ion derivatives of this interesting interlocked system.
Experimental
2-Bromo-6-picoline,[13] 6,6ꢀ-dimethyl-2,2ꢀ-bipyridine,[14] and
tetrakis(acetonitrile)copper(i) hexafluorophosphate[15] were
synthesized and purified according to the literature methods.
6,6ꢀ-Bis(chloromethyl)-2,2ꢀ-bipyridine was also synthesized by
the literature method[10] but was purified by flash chromato-
graphy on silica (CHCl3 as eluent).
Fig. 4. X-ray structure of the [Cu(4)2]+ catenane cation.
interlocked (and more flexible) ‘spiral’complex of the 2:2 ‘large
ring’ macrocycle.
NMR spectra were determined at 298 K on a Bruker
AM300 spectrometer in CD3SOCD2H or CD3OD; δH values
for CD3SOCD2H are relative to this solvent at 2.50 ppm and
TMS for CD3OD; δC values are relative to CD3SOCD2H at
39.52 ppm. High-resolution ESI mass spectra were recorded on
a Bruker BioApex 7T FTICR mass spectrometer (with an Ana-
lytica source), while low-resolution ESI spectra were obtained
on a Finnigan LCQ-8 spectrometer.
An X-ray structure determination on the above product was
undertaken using a deep orange crystal taken directly (without
drying) from the aqueous isopropanol recrystallization solu-
tion. The structure (Fig. 4) shows that the product is indeed
the required [2]-catenane with an overall stoichiometry of
([Cu(4)]PF6)·(iPrOH)1.5·(H2O)2.75. The water and isopropanol
molecules are involved in a hydrogen bonding network with the
secondary amines from the macrocyclic rings (see Accessory
Publication Fig. AP4). Each aryl ring in the catenane cation is
arranged such that an offset face-to-face π-interaction occurs
between it and an adjacent bipyridine moiety, with the separation
between rings being ∼3.6 Å.
6,6ꢀ-Bis[(4ꢀꢀ-formylphenoxy)methyl]-2,2ꢀ-bipyridyl 2
Sodium hydrogen carbonate (0.70 g, 4.2 mmol) and potas-
sium iodide (0.03 g, 0.2 mmol) were added to a solution of

RESEARCH FRONT
1018
J. R. Price et al.
a
g
b
f
ꢁ
e
HN
c
NH
N
N
O
O
O
d
d
O
e
h
c
b
f
i
N
a
j
N
N
Cu
g
N
O
O
O
O
Scheme 3.
NH
HN
6,6ꢀ-bis(chloromethyl)-2,2ꢀ-bipyridine (0.48 g, 1.9 mmol) and 4-
hydroxybenzaldehyde (0.51 g, 4.2 mmol) in dimethylformamide
(170 mL). This mixture was heated at 140◦C for 16 h, filtered
hot and then reduced in volume to 10 mL by evaporation under
reduced pressure. On allowing the solution to stand overnight
at 4◦C, a microcrystalline product precipitated and was iso-
lated by filtration and recrystallized from dimethylformamide to
yield an off-white powder, which was dried under vacuum over
phosphorus pentoxide (0.43 g, 53%). mp 252◦C (Found: LH+,
m/z 425.149 (HRMS-ESI). C26H21N2O4 requires 425.150). δH
((CD3)2SO, 300 MHz) 5.43 (4H, s, H-d), 7.29 (4H, d, J 9, H-e),
7.62 (2H, d, J 8, H-c), 7.90 (4H, d, J 9, H-f), 8.02 (2H, t, J 8,
H-b), 8.33 (2H, d, J 8, H-a), 9.84 (2H, s, H-g) (Scheme 3).
Scheme 4.
directly for the X-ray structure determination. The remaining
product was separated by filtration and dried under vacuum over
phosphorus pentoxide (43 mg, 7%) (Found: (CuL)+, m/z 1079.5
(MS-ESI). C64H72N8O4Cu requires 1079.49) (Found: C, 58.4;
H, 6.0; N, 8.6. Calc. for C64H72N8O4CuPF6·5H2O: C, 58.42; H,
6.28; N, 8.52%). δH ((CD3)2SO, 300 MHz) 1.48 (4H, m, H-j),
1.55 (4H, m, H-i), 2.47 (4H, m, H-h), 3.36 (4H, s, NH), 3.45
(4H, s, H-g), 4.70 (4H, s, H-d), 5.95 (8H, d, J 8, H-f), 6.68 (8H,
d, J 8, H-e), 7.42 (4H, d, J 8, H-c), 7.81 (4H, d, J 7, H-a), 7.93
(4H, t, J 8, H-b) Scheme 4.
Free catenane (4) was generated by a similar method to that
described by Sauvage et al.[4] Treatment of the above orange
NMR solution of [Cu(4)]PF6 with tetrabutylammonium cyanide
resulted in a colourless solution from which the demetalated
product, 4, was obtained on workup. (Found: LH+, m/z 1017.571
(HRMS-ESI). C64H72N8O4 requires 1017.575). δH ((CD3)2SO,
300 MHz) 0.08 (4H, m, H-j), 0.43 (4H, m, H-i), 1.48 (4H, m,
H-h), 2.98 (4H, s, H-g), 3.36 (4H, s, NH), 5.27 (4H, s, H-d), 6.75
(8H, d, J 9, H-f), 6.80 (8H, d, J 9, H-e), 7.52 (4H, d, J 7, H-c),
7.85 (4H, t, J 8, H-b), 8.33 (4H, d, J 7, H-a). Proton designations
use the same convention as for the above copper(i) complex.
[Cu(6,6ꢀ-bis[(4ꢀꢀ-formylphenoxy)methyl]-2,2ꢀ-
bipyridyl)2]PF6 [Cu(2)2]PF6
Dialdehyde 2 (0.40 g, 0.94 mmol) was suspended in hot aceto-
nitrile (20 mL) and tetrakis(acetonitrile)copper(i) hexafluoro-
phosphate (0.18 g, 0.48 mmol) was added, the mixture was
stirred at reflux for 15 min and then the hot solution was fil-
tered. The filtrate was allowed to cool and then let stand at
4◦C. Dark red crystals formed overnight; a crystal from this
crop was used directly for the X-ray structure determination.
The remaining product was dried under vacuum over phos-
phorus pentoxide (0.29 g, 58%) (Found: [Cu(3)2]+, m/z 911.1
(MS-ESI). C52H40N4O8Cu requires 911.21) (Found: C, 59.3;
H, 3.6; N, 5.3. Calc. for C52H40N4O8CuPF6: C, 59.07; H, 3.81;
N, 5.30%). δH ((CD3)2SO, 300 MHz) 4.91 (4H, s, H-d), 6.46
(4H, d, J 9, H-e), 7.52 (4H, d, J 9, H-f), 7.84 (2H, d, J 8, H-c),
8.06 (2H, t, J 8, H-b), 8.23 (2H, d, J 8, H-a), 9.78 (2H, s, H-
g). δC (CD3SOCD2H, 75 MHz) 70.61, 113.78, 122.08, 126.62,
129.74, 131.33, 139.25, 151.07, 153.15, 161.95, 191.10.
Structure Determinations
Data were collected with ω scans to ∼56◦ 2θ using a
Bruker SMART 1000 diffractometer employing graphite-
monochromated MoKα radiation generated from a sealed tube
(0.71073 Å) at 150(2) K. Data integration and reduction were
undertaken with SAINT and XPREP.[16] Subsequent com-
putations were carried out using the WinGX-32 graphical
user interface.[17] Structures were solved by direct methods
using either SHELXS-97[18] or SIR97.[19] Multi-scan empiri-
cal absorption corrections, when applied, were applied to the
dataset using the program SADABS.[20,21] Data were refined and
extended with SHELXL-97.[22] In general, non-hydrogen atoms
were refined anisotropically. Carbon-bound hydrogen atoms
were included in idealized positions and refined using a rid-
ing model. Oxygen bound hydrogen atoms were first located
in the difference Fourier map before refinement. Where these
hydrogen atoms could not be located, they were not modelled.
Catenane [Cu(4)]PF6·5H2O
[Cu(6,6ꢀ-bis[(4ꢀꢀ-formylphenoxy)methyl]-2,2ꢀ-bipyridyl)2]PF6
(483 mg, 0.46 mmol) was dissolved in warm methanol (1 L),
activated 4 Å molecular sieves were added to the warm stirred
solution followed by dropwise addition of 1,6-hexanediamine
(106 mg, 0.91 mmol) in methanol (500 mL) over 1 h. The reac-
tion mixture was then heated at reflux for 30 min and sodium
borohydride (173 mg, 4.6 mmol) was carefully added to the hot
solution. The yellow reaction solution was refluxed for a fur-
ther 30 min, filtered hot, and the solvent removed under reduced
pressure. The crude residue that remained was then added to
a stirred dichloromethane/ethyl acetate (∼20/100 mL) solvent
mixture and the insoluble off-white inorganic salts that remained
were removed by filtration and discarded. The filtrate was taken
todrynessunderreducedpressuretoyieldthecrudeproductasan
orange microcrystalline solid. This was purified by slow recrys-
tallization from aqueous isopropanol to yield a small quantity
of deep orange crystals; a crystal from this solution was used
[Cu(2)2]PF6·CH3CN
Formula C56H46CuF6N6O8P, M 1139.50, monoclinic, space
group C2/c(#15), a 26.7343(17), b 12.2824(7), c 16.2932(10) Å,
β 100.863(3), V 5254.2(6) Å3, Dc 1.441 g cm−3, Z 4, crys-
tal size 0.28 by 0.20 by 0.07 mm, colour red, habit plate,
T
150(2) K, λ(MoKα) 0.71073 Å, µ(MoKα) 0.529 mm−1
,
T(SADABS)min,max 0.925, 1.000, 2θmax 57.48, hkl range −35

RESEARCH FRONT
Copper(i) Templated Synthesis of a 2,2ꢀ-Bipyridine Derived 2-Catenane
1019
26, −16 15, −21 21, N 25282, Nind 6352(Rmerge 0.0569), Nobs
4673(I > 2σ(I)), Nvar 384, residuals∗ R1(F) 0.0754, wR2(F2)
(b) E. R. Kay, D. A. Leigh, Pure Appl. Chem. 2008, 80, 17.
doi:10.1351/PAC200880010017
(c) B. Champin, P. Mobian, J.-P. Sauvage, Chem. Soc. Rev. 2007, 36,
358. doi:10.1039/B604484K
(d) V. Balzani, A. Credi, S. Silvi, M. Venturi, Chem. Soc. Rev. 2006,
35, 1135. doi:10.1039/B517102B
0.1907, GoF(all) 1.113, ꢀρmin,max −1.328, 1.304 e− Å−3
.
*R1 = ꢁ||Fo| − |Fc||/ꢁ|Fo|
for
Fo > 2σ(Fo);
wR2 =
[ꢁw(Fo2 − Fc2)2/ꢁ(wFc2)2]1/2 all reflections w = 1/[σ2(Fo2) +
(0.0504P)2 + 38.0461P] where P = (Fo2 + 2F2c)/3.
[3] (a) See, for example: D. A. Leigh, P. J. Lusby, R. T. McBurney,
A. Morelli, A. M. Z. Slawin, A. R. Thomson, D. B. Walker, J. Am.
Chem. Soc. 2009, 131, 3762. doi:10.1021/JA809627J
(b) S. M. Goldup, D. A. Leigh, P. J. Lusby, R. T. McBurney,
A. M. Z. Slawin, Angew. Chem. Int. Ed. 2008, 47, 6999. doi:10.1002/
ANIE.200801904
Specific details: One of the ligand arms was refined as dis-
ordered over two positions with total occupancy of 1 (0.55 and
0.45 respectively), and the PF6 anion was refined as a rigid body.
([Cu(4)]PF6)·(iPrOH)1.5·(H2O)2.75
(c) D. A. Leigh, J. Am. Chem. Soc. 2009, 131, 3762.
doi:10.1021/JA809627J
(d) S. M. Goldup, D. A. Leigh, P. J. Lusby, R. T. McBurney,
A. M. Z. Slawin, Angew. Chem. Int. Ed. 2008, 47, 6999. doi:10.1002/
ANIE.200801904
(e) P. Mobian, J.-M. Kern, J.-P. Sauvage, Inorg. Chem. 2003, 42, 8633.
doi:10.1021/IC030181G
(f) P. Mobian, J.-M. Kern, J.-P. Sauvage, J. Am. Chem. Soc. 2003, 125,
2016. doi:10.1021/JA0209215
Formula C68.50H84CuF6N8O8.25P, M 1359.94, monoclinic,
space group P21/n(#14), a 13.144(3), b 40.160(10), c
13.131(3) Å, β 89.990(5), V 6931(3) Å3, Dc 1.298 g cm−3, Z 4,
crystal size 0.322 by 0.298 by 0.144 mm, colour orange, habit
prism, T 150(2) K, λ(MoKα) 0.71073 Å, µ(MoKα) 0.413 mm−1
,
2θmax 56.68, hkl range −17 16, −53 37, −17 14, N 45239,
Nind 16247(Rmerge 0.0406), Nobs 11232(I > 2σ(I)), Nvar 812,
residuals∗ R1(F) 0.0621, wR2(F2) 0.1782, GoF(all) 1.060,
ꢀρmin,max −0.896, 1.040 e− Å−3. *R1 = ꢁ||Fo| − |Fc||/ꢁ|Fo| for
[4] C. O. Dietrich-Buchecker, J. P. Sauvage, J. M. Kern, J.Am. Chem. Soc.
1984, 106, 3043. doi:10.1021/JA00322A055
[5] J. P. Sauvage, M. Ward, Inorg. Chem. 1991, 30, 3869.
doi:10.1021/IC00020A019
Fo > 2σ(Fo); wR2 = [ꢁw(F2o − Fc2)2/ꢁ(wFc) ]
2 2 1/2 all reflections
w = 1/[σ2(Fo2) + (0.0983P)2 + 1.0277P]whereP = (Fo2 + F2c)/3.
Specific details: The metric symmetry was found to be tetra-
gonal, however no solution could be found with this symmetry.
The crystal symmetry subsequently proved to be monoclinic
with pseudo-merohedral twinning arising from a 90 degree rota-
tion about the monoclinic b axis. Application of the appropriate
twin law with a refined twin fraction of 0.5 reduced the con-
ventional residual from 0.22 to 0.07. The same twinning was
observed in a second crystal for which data were also collected.
An absorption correction was not applied to the data. The asym-
metric unit contains one metal complex cation, one PF6 anion,
two partial occupancy isopropanol molecules (occupancy 0.75
each) and three water molecules (O(1)-O(3)), one of which was
modelled with partial occupancy (0.75). Partial occupancy non-
hydrogen atoms were refined with isotropic thermal parameters.
No O-bound hydrogen atoms were included in the model.
Crystallographic data (excluding structure factors) for the
structures reported in this paper have been deposited with the
Cambridge Crystallographic Data Centre as supplementary pub-
lication nos. CCDC 729163–729164. Copies of the data can be
obtained free of charge on application to the CFCDC, 12 Union
Road, Cambridge, CB2 1EZ, UK (fax: (+44) 1223 336 033;
email: deposit@ccdc.ccam.ac.uk).
[6] D. A. Leigh, P. J. Lusby, S. J. Teat, A. J. Wilson, J. K. Y. Wong,
Angew. Chem. Int. Ed. 2001, 40, 1538. doi:10.1002/1521-3773
(20010417)40:8<1538::AID-ANIE1538>3.0.CO;2-F
[7] C. Hamann, J. M. Kern, J.-P. Sauvage, Inorg. Chem. 2003, 42, 1877.
doi:10.1021/IC020381C
[8] M. Hutin, C. A. Schalley, G. Bernardinelli, J. R. Nitschke, Chem. Eur.
J. 2006, 12, 4069. doi:10.1002/CHEM.200501591
[9] Spartan ’02 Windows 2002 (Wavefunction: Irvine, CA).
[10] G. R. Newkome, W. E. Puckett, G. E. Keifer, V. K. Gupta,
Y. Xia, M. Coreil, M. A. Hackney, J. Org. Chem. 1982, 47, 4116.
doi:10.1021/JO00142A022
[11] (a) P. A. Vigato, S. Tamburini, Coord. Chem. Rev. 2004, 248, 1717.
doi:10.1016/J.CCT.2003.09.003
(b) L. F. Lindoy, Chem. Soc. Rev. 1975, 4, 421. doi:10.1039/
CS9750400421
[12] L. F. Lindoy, Quart. Rev. 1971, 379.
[13] G. R. Newkome, D. C. Pantaleo, W. E. Puckett, P. L. Zeifle, W. A.
Deutsch, J. Inorg. Nucl. Chem. 1981, 43, 1529. doi:10.1016/0022-
1902(81)80331-4
[14] T. M. Cassol, F.W. J. Demnitz, M. Navarro, E.A. d. Neves,Tetrahedron
Lett. 2000, 41, 8203. doi:10.1016/S0040-4039(00)01310-1
[15] G. J. Kubas, Inorg. Synth. 1979, 19, 90. doi:10.1002/9780470132500.
CH18
[16] SMART, SAINT and XPREP 1995 (Bruker Analytical X-ray Instru-
ments Inc.: Madison, WI).
[17] G.Win, X-32: System of programs for solving, refining, and analyzing
singlecrystalX-raydiffractiondataforsmallmolecules, L. J. Farrugia,
J. Appl. Cryst. 1999, 32, 837.
[18] G. M. Sheldrick, SHELXL-97: Programs for Crystal Structure Anal-
ysis 1997 (University of Göttingen: Germany).
[19] A. Altomare, M. C. Burla, M. Camalli, G. L. Cascarano,
C. Giocavazzo, A. Guagliardi, A. G. C. Moliterni, G. Polidori,
R. Spagna, J. Appl. Cryst. 1999, 32, 115. doi:10.1107/
S0021889898007717
Accessory Publication
Further details of the MS and 1H NMR spectra for the catenane
and selected precursors as well as X-ray details of the hydro-
gen bonded network present in the solid state for the copper(i)
catenane complex are available on the Journal’s website.
Acknowledgement
We thank the Australian Research Council for support.
[20] R. H. Blessing, Acta Crystallogr. 1995, A32, 837.
[21] G. M. Sheldrick, SADABS: Empirical Absorption and Correction
Software 1996 (University of Göttingen: Germany).
[22] G. M. Sheldrick, SHELXL-97: Programs for Crystal Structure Anal-
ysis 1997 (University of Göttingen: Germany).
References
[1] L. F. Lindoy, I. M. Atkinson, Self-assembly in Supramolecular
Chemistry 2000 (Royal Society for Chemistry: Cambridge, UK).
[2] (a) S. Bonnet, J.-P. Collin, Chem. Soc. Rev. 2008, 37, 1207.
doi:10.1039/B713678C