Systematic study of the formation of the lanthanoid cubane cluster motif mediated by steric modification of diketonate ligands

Article abstract of DOI:10.1039/c1dt10580a

The treatment of ortho ring-functionalised 1-phenylbutane-1,3-dione ligands bearing nitro (Hnpd, Hnmc), methoxy (Hmmc) or fluoro (Hfpp) groups with hydrated lanthanoid salts has provided [Er₄(μ₃-OH) ₄(H₂O)₂

Full text of DOI:10.1039/c1dt10580a

View Article Online / Journal Homepage / Table of Contents for this issue
This article is published as part of the Dalton Transactions themed issue entitled:
Self-Assembly in Inorganic Chemistry
Guest Editors Paul Kruger and Thorri Gunnlaugsson
Published in issue 45, 2011 of Dalton Transactions
Image reproduced with permission of Mark Ogden
Articles in the issue include:
PERSPECTIVE:
Metal ion directed self-assembly of sensors for ions, molecules and biomolecules
Jim A. Thomas
Dalton Trans., 2011, DOI: 10.1039/C1DT10876J
ARTICLES:
Self-assembly between dicarboxylate ions and a binuclear europium complex: formation of stable
adducts and heterometallic lanthanide complexes
James A. Tilney, Thomas Just Sørensen, Benjamin P. Burton-Pye and Stephen Faulkner
Dalton Trans., 2011, DOI: 10.1039/C1DT11103E
Structural and metallo selectivity in the assembly of [2 × 2] grid-type metallosupramolecular
species: Mechanisms and kinetic control
Artur R. Stefankiewicz, Jack Harrowfield, Augustin Madalan, Kari Rissanen, Alexandre N. Sobolev
and Jean-Marie Lehn
Dalton Trans., 2011, DOI: 10.1039/C1DT11226K
Visit the Dalton Transactions website for more cutting-edge inorganic and organometallic research
www.rsc.org/dalton

Dynamic Article Links
Dalton
Transactions
Cite this: Dalton Trans., 2011, 40, 12169
www.rsc.org/dalton
PAPER
Systematic study of the formation of the lanthanoid cubane cluster motif
mediated by steric modification of diketonate ligands†
Philip C. Andrews, William J. Gee, Peter C. Junk* and Jonathan G. MacLellan
Received 5th April 2011, Accepted 5th July 2011
DOI: 10.1039/c1dt10580a
The treatment of ortho ring-functionalised 1-phenylbutane-1,3-dione ligands bearing nitro (Hnpd,
Hnmc), methoxy (Hmmc) or fluoro (Hfpp) groups with hydrated lanthanoid salts has provided
[Er4(m3-OH)4(H2O)2(npd)8] (3), [Ln4(m3-OH)4(nmc)8] (Ln = Gd (4), Tb (5), Dy (6) and Er (7)),
[Er4(m3-OH)4(mmc)8] (8) and [Er4(m3-OH)4(H2O)2(fpp)8] (9), respectively. The products were all obtained
as cubane clusters in the solid state, as distinct from previous diketonato clusters, with control over
motif formation attributed to the steric influence of the ortho-positioned functional groups at the
cluster periphery. This work highlights a means of targeting a specific lanthanoid cluster motif by the
rational modification of ligands at key locations.
a controlled synthesis of the lanthanoid cubane cluster motif
Introduction
by the manipulation of steric bulk at precise locations of a 1-
The ability to control the identity of lanthanoid cluster motifs
phenylbutane-1,3-dione ligand. Much research has been under-
would represent an important breakthrough relevant to many
taken on diketonate stabilised lanthanoid cage clusters, both
by us,18–21 and others,10–17,22,23 prompting the exploration of this
postulate using such ligands. Of particular relevance was the
isolation of a cubane motif by the inclusionofa ferrocene moietyto
a diketonate scaffold.22 It has been demonstrated that the cubane
motif is common to lanthanoid cluster chemistry,24–28 yet only
three examples are reported stabilised by diketonate ligands.17,22,29
As previously mentioned, the size of the lanthanoid ion has great
influence over the cluster motif, hence it should be noted that the
primary focus of this report has been limited to the intermediate
and heavy lanthanoid ions.
We believe the lack of diketonate cubane examples in the
literature is ligand mediated, resulting from the conjugated planar
nature of the diketone ligand. The planarity of 1-phenyl-1,3-
diketonate ligands allows for closely packed parallel arrangements
around the periphery of clusters, typically observed within many of
the motifs stabilised by such ligands. In addition to the amenability
of closely-packed, planar diketonate arrays, interactions such
as CH–p and p–p stacking have been observed in the solid
state.30 Both the parallel arrangements and p–p interactions
observed at the cluster periphery are susceptible to disruption
by manipulating ligand sterics, promoting variation in ligand
number, and ultimately, motif. We postulate that increased steric
bulk imparted by the ferrocene diketonate ligands reported by
Roesky et al.22 disrupted the bridging modes typical for 1-phenyl-
1,3-diketonate ligands, resulting in the cubane motif. A review
of the Cambridge Crystallographic Database shows that in the
solid state, ortho substituted 1-phenyl-1,3-diketone species exhibit
disrupted planarity between the phenyl and diketone group,
increasing the steric profile of the ligand.31–37 Cumulatively, this
minor change in the dihedral angle of the substituted phenyl
fields of inorganic chemistry, allowing for the design of specific
clusters suited to target applications. As an example, such control
could facilitate the design of finely tuned, potent luminescent1–5
or magnetic6,7 devices. This stems from the strong influence
interatomic distances and relative orientations of lanthanoid
centres have on the electronic configuration responsible for the
above applications. Traditionally, the discovery and isolation of
lanthanoid oxo/hydroxo cage cluster motifs has been considered
a serendipitous undertaking. More recently however, targeted
approaches to new lanthanoid motifs have begun to appear,
such as the use of anionic templates.8,9 Furthermore, an inverse
relationship has been established regarding the size of the ligand
relative to the resultant cluster, exemplified by the diverse range
of clusters stabilised by diketonate ligands of varying size, such
as the nonanuclear clusters stabilised by pentane-2,4-dione9,10 and
ethyl 3-oxobutanoate,12 and penta- and tetranuclear clusters by the
sterically more demanding dibenzoylmethane.13–16 Indeed a recent
report by Zheng et al. elegantly and thoroughly demonstrated
this principle utilising the diketonate ligand acetylacetone.17 Each
of these reports has brought the controlled design of lanthanoid
cluster motifs closer to a reality.
By adopting the same systematic approach when studying
cluster motif as Zheng,17 we herein are able to demonstrate
School of Chemistry, Monash University, Clayton, Victoria, 3800, Australia.
E-mail: peter.junk@monash.edu
† Electronic supplementary information (ESI) available: Molecular struc-
ture of 1 and 2, tables of bond lengths and angles for 3, 7–9, and examples
of ‘capping’ vs. ‘peripheral’ for cubane spaces 3, 7 and 8. CCDC reference
numbers 801771–801778 and 819703. For ESI and crystallographic data
in CIF or other electronic format see DOI: 10.1039/c1dt10580a
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 12169–12179 | 12169
©

ring effects a major change in the resultant cluster motif when
compared to clusters stabilised by the parent ligand.
Herein we demonstrate the effect of a single ortho ring sub-
stitution using small functional groups (nitro, methoxy, fluoro)
on two diketone ligand controls: dibenzoylmethane and (Z)-2-
(hydroxy(phenyl)methylene)cyclopentanone (Hpmc). Dibenzoyl-
methane is commercially available and, used as a ligand, has previ-
ously been found to stabilise the planar tetranuclear and pentanu-
clear lanthanoid oxo/hydroxo cage cluster motif.13,14 The synthesis
of Hpmc is reported, with subsequent reactions found to generate
both the tetranuclear and nonanuclear lanthanoid oxo/hydroxo
cage cluster motif dependent on the metal used. Modification
of the control ligands by the addition of an ortho functional
group (o-nitro, o-methoxy, o-fluoro) was found to convert the
parent cluster motifs to cubane cluster motifs. The syntheses of
two o-nitrophenyl, a o-methoxyphenyl and a pentafluorophenyl
ligand are reported, as well as the structural characterisation
of their corresponding lanthanoid cubane clusters. These results
give a straightforward, targeted synthesis of a predetermined
cluster motif and represent an important step away from the
serendipitous isolation of lanthanoid cluster motifs in favour of
the deliberate design of target species. Such a design strategy will
prove essential in improving the efficiency and scope of lanthanoid
cluster materials that have been heralded as prerequisite to a range
of important technological advances.
Fig. 1 Complete structures of Hpmc, Hnpd, Hnmc, Hmmc and Hfpp.
1H-NMR studies performed on each of the ligands showed
the enolic tautomer was dominant in solution. A ratio of
approximately 1 : 2 (keto : enol) was observed for Hpmc, Hnmc and
Hmmc, with the remainder exclusively in the enolic conformation.
In each instance, tautomers were assigned by the presence of either
an enolic proton in the range of 13.8 to 15.9 ppm, or by the
methylene signal located between the carbonyl functionalities at
4.27, 3.67 and 4.33 ppm respectively for the keto tautomer.
Reaction of two equivalents of the parent ligand Hpmc with
one equivalent of [LnCl2·H2O)6]Cl (Ln = Nd, Er) in methanol
using excess triethylamine as base (Scheme 2) generated two
cluster motifs. Subsequent removal of the methanol under reduced
pressure and dissolving the residue in toluene precipitated the
by-product triethylamine hydrochloride, enabling isolation of sol-
vated pure compounds after filtration. Single crystals were grown
by slow diffusion of n-hexane into separate toluene solutions
of 1 and 2. Dependent on the size of the Ln3+ ionic radii, the
tetranuclear planar diamond motif with neodymium, [Nd4(m3-
OH)2(pmc)10] (1), and a nonanuclear cluster exhibiting an anionic
hourglass motif for erbium, [HNEt3][Er9(m4-O)2(m3-OH)8(pmc)16]
(2) were isolated. Both of these motifs have been described
previously in the literature.14,39 Products were characterised by
IR spectroscopy, single crystal X-ray diffraction and elemental
analysis. The crystallographic data for 1 and 2 can be found
in the ESI.† It has been surmised that both dibenzoylmethane
and the parent diketonate ligand Hpmc are able to generate this
range of motifs owing to the ability of diketonates to easily adopt
Results and discussion
To best observe the effects of ortho ring-substitution of 1-phenyl-
1,3-diketonate ligands on lanthanoid cluster motifs, the chemistry
of the parent ligands was first evaluated. The first of the two control
ligands, dibenzoylmethane, is commercially available and known
to stabilise two common motifs: a planar diamond tetranuclear
motif exhibited by Ln3+ ions with larger ionic radii (Pr, Nd, Sm,
Yb, Lu), and the square-based pyramidal pentanuclear motif for
Ln3+ metals with smaller ionic radii (Eu, Er, Tm).14,15 The second
control ligand, (Z)-2-(hydroxy(phenyl)methylene)cyclopentanone
(Hpmc), was chosen to promote formation of higher nuclearity
clusters. The ligand Hpmc is not commercially available, and was
synthesised according to a modified literature procedure, whereby
an acid chloride precursor was treated with the lithium enolate
of cyclopentanone.38 Four ortho ring functionalised b-diketone
ligands were synthesised analogously to Hpmc by the use of
ring substituted acid chloride precursors and, where appropriate,
exchanging cyclopentanone for acetophenone (Scheme 1). Each of
the ligands have been depicted in full in Fig. 1. Moderate to good
yields were obtained for each ligand after column chromatography,
ranging from 60% for Hmmc to 91% for Hpmc respectively.
2
bridging modes between metals, such as the (m-O)2-h and (m-O)-
2
h coordination modes.14 In such cases, any intermediate dimeric
species formed would likely be bridged through the ligand as
opposed to hydroxy bridging, which is thought necessary for
cubane formation.26,27
Scheme 2 Cluster species stabilised by parent ligand Hpmc with varying
ionic radii of Ln3+
.
Each of the ortho substituted ligands (Hnpd, Hnmc, Hmmc,
Hfpp) yielded cubane clusters when treated with hydrated lan-
thanoid chloride salts using identical conditions to that which
gave clusters 1 and 2 (Scheme 3). With the exception of Hnmc,
erbium has been the sole target of investigation with each ligand
to limit the influence of lanthanoid ionic radius on motif. The
ortho nitro ligand Hnpd yielded a cubane cluster of composition
[Er4(m3-OH)4(H2O)2(npd)8] (3). The related ligand Hnmc gave the
Scheme 1 Synthetic pathway to b-diketone ligands.
12170 | Dalton Trans., 2011, 40, 12169–12179
This journal is
The Royal Society of Chemistry 2011
©

Scheme 3 Various compositions of cubane cluster motif observed
dependent on the ligand employed.
cubane series [Ln4(m3-OH)4(nmc)8] (Ln = Gd (4), Tb (5), Dy (6)
and Er (7)). Substitution of the nitro group for a methoxy group
resulted in a continuance of the cubane trend, as demonstrated by
Hmmc which yielded the cubane cluster [Er4(m3-OH)4(mmc)8] (8).
The pentafluoro ligand Hfpp yielded a cubane cluster analogous
to 3 with a composition of [Er4(m3-OH)4(H2O)2(fpp)8] (9). Single
crystals were obtained by slow evaporation of concentrated
solutions of crude product in either neat or 5% acetone solutions
of toluene. In each case, characterisation was achieved using
infrared spectroscopy, elemental analysis and single crystal X-ray
diffraction. Yields for each cubane species varied from moderate
to good. To allow meaningful comparisons of bond lengths and
angles in the solid state, erbium cluster 7 has been chosen to
represent the cubane series 4–7. A summary of the crystal data
for clusters 1–9 is given in Table 1.
The triethylamine-mediated reaction of Hnpd with hydrated
erbium chloride resulted in the successful isolation of a cubane
product 3 (Fig. 2). The asymmetric unit of 3 contained a neutral
2
2
[Er4(m3-OH)4(H2O)2(h -npd)6(m-h -npd)2] cluster complete with
two molecules of toluene in the lattice which are disordered over
four positions. Lists of selected bond lengths and angles have
been generated for each of the cubane clusters discussed and
can be found in the ESI.† Each of the four erbium centres are
eight coordinate, coordinating solely to oxygen atoms originating
from either the hydroxide anions in the cubane core, chelating
ligands or coordinated aqua ligands. Six of the eight ligands are
2
solely h chelating to an erbium centre. The remaining two ligands
Fig.
2 Molecular structure of the cubane cluster [Er4(m3-OH)4-
2 2
(H2O)2(h -npd)6(m-h -npd)2] (3). Lattice solvent and hydrogen atoms have
been omitted for clarity.
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 12169–12179 | 12171
©

2
exhibit a m-h bridging mode between two erbium centres and are
crystallographically to form an isomorphous series. Each of
the four compounds possesses a crystallographically imposed
4 symmetry. The structure of the erbium moiety 7 is given in
Fig. 4. While 4–7 are isomorphous, only the erbium analogue (7)
will be discussed in detail.
located on opposing faces of the cubane. The sterically crowded
cluster periphery is further imbued with two coordinated aqua
ligands located on Er(3) and Er(4). The retention of aqua ligands
is atypical, as diketonate ligands will usually displace such co-
¯
2
ligands by adopting the m-h bridging mode. The grounds for the
retention of aqua ligands will be discussed in detail below. All six
faces of the cubane are highly distorted but four of the faces have
similar Er–O–Er angles which range from 107.5(3)◦–111.9(3)◦. O–
Er–O angles range from 74.0(3)◦–76.5(3)◦ for two of these faces
with the other two in the range 68.9(3)◦–71.0(3)◦. The remaining
faces each have Er–O–Er angles in the range 97.4(3)◦–98.6(3)◦ and
O–Er–O angles ranging from 68.2(3)◦–69.8(3)◦. Average Er–O(H)
˚
˚
bond lengths are 2.36 A and lie in the range 2.31(1)–2.43(1) A. The
2
˚
h -diketonates have average Er–O bond lengths of 2.30 A while the
two bridging ligands have one short (non-bridging) Er–O distance
˚
˚
averaging 2.29 A and two longer contacts averaging 2.50 A. The
˚
˚
two Er–O(H2) distances are 2.44(1) A and 2.41(1) A. Two of the
m3-OH groups act as hydrogen bond donors to one of the nitro O
atoms of a neighbouring ligand. The second oxygen atom in turn
acts as a hydrogen bond acceptor from one of the coordinated
water molecules. These interactions are designated in both cases
as R22(8) by Etter’s notation (Fig. 3).40
Fig. 4 Molecular structure of the cubane cluster [Er4(m3-OH)4(nmc)8] (7).
Lattice solvent and hydrogen atoms have been omitted for clarity.
Unlike the structure of 3, which contained one entire molecule
in the asymmetric unit, the structure of 7 contains only a single
erbium chelated by two nmc ligands and coordinated by a single
hydroxy group. As a result of this symmetry, the entire cluster is
comprised of the tetranuclear cubane core with four m3-bridged
hydroxy anions and eight ligands positioned around the exterior.
Each erbium atom is coordinated by two ligands binding in a
2
h (O,O¢) fashion and three m3-bridged hydroxy anions, making
each metal seven coordinate. This gives the cluster an overall
composition of [Er4(m3-OH)4(nmc)8]. The cubane core is somewhat
distorted with two opposing faces having both Er–O–Er angles
of 109.40(11)◦ and O–Er–O angles of 69.86(11)◦. The two Er–
˚
˚
O distances within this face are 2.31(1) A and 2.38(1) A. The
four remaining faces have Er–O–Er bond angles of 101.38◦ and
103.55◦ and O–Er–O angles of 73.35(13)◦ and 74.67(13)◦, with
Fig. 3 Enlargement of cluster 3 showing intra-cluster hydrogen-bonding
interactions as dashed lines. Ligands not exhibiting hydrogen-bonding
interactions and the lattice solvents have been omitted for clarity. Hydrogen
atoms attached to the coordinated water could not sustain refinement,
hence only the donor–acceptor interactions are shown.
˚
two Er–O bond lengths of 2.33(1) A and the other two of
˚
˚
2.31(1) A and 2.38(1) A. Despite the additional symmetry in
7 these values correspond well with those of 3. There are two
crystallographically distinct diketonates in the structure of 7 and
both act only in a h (O,O¢) fashion; this results in shorter Er–O
˚ ˚
distances of 2.27(1) A and 2.26(1) A in ligand 1 and the slightly
The inter-oxygen donor–acceptor distances between the pair of
2
˚
core m3-OH groups and the nitro termini are 2.99(1) A for O(1)–
˚
O(16) and 2.90(1) A for O(2)–O(7), while the coordinated aqua
˚
˚
˚
ligands to the opposite nitro termini distances are 2.80(1) A for
more asymmetric interaction, 2.24(1) A and 2.31(1) A, in ligand
2. These shorter distances also reflect the lower coordination
number of 7 as compared to 8 in 3. The planes of the phenyl
rings are oriented at angles of 74.4(2)◦ and 55.84(2)◦ relative to
the respective conjugated keto–enol planes of each nmc. The nitro
groups of the phenyl rings aligned at 74.4(2)◦ are positioned such
that they can accept hydrogen bond donation from the hydroxide
core. This geometric positioning is apparent in each of the solid
˚
O(37)–(O15) and 3.05(1) A for O(38)–O(8). To date no examples
of lanthanoid intracluster hydrogen-bonding have been reported,
however close inspection of a recent cluster reported by Roesky
et al. identified a nitro to m4-OH hydroxy interaction with an inter-
41
˚
oxygen distance of 2.970 A.
The ligand Hnmc reacted with hydrated lanthanoid chlorides
to yield cubane clusters containing lanthanoid centres with a
reduced coordination number of seven, owing to the reduced steric
bulk of the ligand. Each of the structures 4–7 were determined
˚
state structures of 4–7. In 7 the H ◊ ◊ ◊ O distance is 2.66 A, which
is within the limit of the van der Waals interaction for these two
12172 | Dalton Trans., 2011, 40, 12169–12179
This journal is
The Royal Society of Chemistry 2011
©

˚
atoms (2.72 A). One ligand of the asymmetric unit is involved
sation of the cluster was hindered by poorly diffracting crystals,
coupled with a high concentration of electron density located at the
cluster core, thermal motion and disorder, all of which culminated
in poorly defined ligands and solvent molecules. Therefore, ligands
were modelled to give the best possible representation of the
structure present. Due to these problems the bond lengths and
angle data for peripheral ligands may not be of high precision, but
there is little doubt about the connectivity.
in a p–p stacking interaction with an aligned ligand from a
neighbouring cluster (by translation -x, 1 - y, x), to give a 2-
D sheet structure (Fig. 5). The centroid–centroid distance was
found to be 3.872 A for 7, slightly offset due to the proximity
of the cyclopentanone ring. Reciprocal N–O dipole interactions
˚
strengthen the p–p stacking, with parallel N O, O N distances
of 3.976 A observed.
˚
The asymmetric unit contains the entire neutral cubane species
and two molecules of toluene. The overall structure of the cluster
in 8 varies from that of 3 only due to the absence of the two
coordinated water molecules. Three of the eight diketonate ligands
2
participate in a m-h bridging mode which ensures Er(1), Er(2)
and Er(4) are eight coordinate. The fourth diketone ligand cannot
achieve the proximity required for Er(3) to bridge, and as a result,
the erbium centre remains seven coordinate. The five non-bridging
2
ligands are h chelating. Intracluster hydrogen-bonding is again
observed, whereby the protons from two of the m3-OH groups
that comprise the core interact with the methoxy oxygen atoms
from two adjacent ligands, assigned as R11(8) by Etter’s notation
(Fig. 7).40 Donor–acceptor distances of 3.04(3) A between O(1)–
˚
˚
H and O(19) and 3.095(19) A between O(4)–H and O(16) were
observed, respectively. The interatomic distances are greater than
those observed for cluster 3, suggesting the intracluster hydrogen-
bonding interactions are weaker, however the inward alignment of
the methoxy groups towards the core is strongly indicative of real
interaction.
Fig. 5 2-D sheet interactions of clusters 4–7 formed by p–p and reciprocal
dipole interactions.
The methoxy bearing ligand Hmmc reacts with hydrated
erbium chloride to yield a cubane cluster of composition [Er4(m3-
OH)4(mmc)8] (8) (Fig. 6). Unlike the previous clusters of analogous
composition (4–7), this erbium cluster was found to contain seven
and eight coordinate metal centres. Crystallographic characteri-
Fig. 7 Enlargement of cluster 8 showing weak intracluster hydro-
gen-bonding interactions (Donor–acceptor distances: O(1) to O(19):
˚
˚
3.036 A; O(4) to O(16): 3.095 A). Ligands not exhibiting hydrogen-bonding
interactions and the lattice solvents have been omitted for clarity.
The pentafluoro ligand yielded cubane species 9 that was
also found to retain coordinated aqua ligands (Fig. 8). The
retention of aqua ligands is likely due to similarities between the
steric profiles of fpp and npd diketonate chelates. The asymmet-
2
ric unit of 9 contains a neutral [Er4(m3-OH)4(H2O)2(h -npd)6(m-
2
h -npd)2] cluster with single molecules of acetone and toluene.
The coordinated aqua ligands participate in weak intracluster
hydrogen-bonding with the fluorinated rings, with oxygen–fluorine
˚
˚
distances of 3.386 A for O(21)–F(21) and 3.149 A for O(22)–
F(36). Additionally, hydrogen-bonding between a coordinated
aqua ligand and a molecule of acetone present in the lattice was
Fig. 6 Molecular structure of the cubane cluster [Er4(m3-OH)4(mmc)8]
(8). Lattice solvent and hydrogen atoms have been omitted for clarity.
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 12169–12179 | 12173
©

Fig. 10 View from above of a capping face (left) and a peripheral face
(right) of cubane 7 as determined by observed symmetry.
It is likely that ortho-substitution of the ligands promotes
formation of dimeric intermediates, which have previously been
implicated as precursors to cubane species. The association of two
dimers (capping faces) to yield the cubane core coincides with
initial nucleophilic displacements of aqua ligands from the m2-OH
bridged hydroxy groups (Fig. 11, Step 1). A postulated secondary
wave of aqua ligand displacements appears to be greatly influenced
by the steric bulk of the ligand. Crystallographic evidence suggests
that the secondary displacements (Fig. 11, Step 2) are effected
by bridging of the diketonate ligands over the peripheral faces.
Furthermore, as there are known examples of cubane clusters
that are free of coordinated aqua ligands and generated from
ligands lacking nucleophilic ortho functionalities, these groups
are not thought to actively displace the aqua ligands.22 Two
clusters 3 and 9, [Er4(m3-OH)4(H2O)2(X)8] (X = npd; fpp), were
each found to retain two residual aqua ligands which we postulate
are owing to the increased size of the dibenzoylmethane scaffold.
Presumably the crowded ligand environment hinders bridging of
the diketonate ligands at the peripheral faces, such that only two
of the four possible nucleophilic displacements occur. Formation
of the cubane core (Step 1) must occur prior to generation of
the hydrogen bonding interactions observed for clusters 3 and 8,
hence such hydrogen bonding interactions are not thought to be a
factor in determining formation of the cubane motif. Evidence of
the dimeric intermediates postulated in Step 1 were obtained from
the methanolic reaction mixtures several hours after the addition
of base using mass spectrometry. Electrospray mass spectrometry
(ESI-MS) has been extensively used for solution state detection of
inorganic species, as well as analysis of solution equilibria, despite
drawbacks such as gas-phase reaction prior to the analyser.42
Intermediate species were identified for three of the four motifs,
specifically for 3, 8 and 9. Table 2 gives a list of intermediate ionic
species observed by ESI-MS. In each case, only the most prominent
isotope is listed. Typically, a dicationic sodium or potassium
salt with the composition [Ln2(OH)2(L)4] (L = npd, mmc, fpp)
was obtained, coupled with varying degrees of solvation from
methanol and residual water. While the intermediate associated
with synthesis of clusters 4–7 was not identified, we believe it
unlikely that an alternate mechanism should occur for formation
of this motif.
Fig.
8 Molecular structure of the cubane cluster [Er4(m3-OH)4-
2 2
(H2O)2(h -fpp)6(m-h -fpp)2] (9). Lattice solvent and hydrogen atoms have
been omitted for clarity.
Fig. 9 Enlargement of cluster 9 showing hydrogen-bonding interactions
as dashed lines. Ligands not exhibiting hydrogen-bonding interactions
have been omitted for clarity. Hydrogen atoms attached to the coordinated
water could not sustain refinement, hence only the donor–acceptor
interactions are shown.
˚
observed, with interoxygen distances of 2.769 A for O(21)–O(23)
(Fig. 9).
While prior reports have postulated a likely mechanism for
cubane cluster formation, the wealth of structural information
obtained in this study warrants a further discussion of the
associative pathway yielding cubane clusters from m2-OH bridged
dimers.26,27 Our descriptions of the stepwise cluster formation
utilise the terms ‘capping’ to represent the two faces derived from
the precursor dimeric species, and ‘peripheral’ for the four faces
generated by their subsequent fusion. The capping faces remain
identifiable, owing to higher perceived symmetry when viewed
from above, as exemplified in Fig. 10 (see ESI†).
The change in coordination number observed for the erbium
containing species 3 and 9, 7 and 8, has been attributed to
both ligand sterics and variations in intracluster interactions.
The relationship between coordination number and sterics is
exemplified by the tetranuclear eight coordinate cubane motif
12174 | Dalton Trans., 2011, 40, 12169–12179
This journal is
The Royal Society of Chemistry 2011
©

Fig. 11 Postulated mechanistic pathway for cubane formation whereby two dimeric species (capping faces) associate by the nucleophilic displacement
of aqua ligands (dashed arrows). The degree of secondary aqua displacements is dependent on the steric bulk of the ligands and mediated by bridging of
the diketonate. For clarity, only half of the cubane species has been shown for Step 2, however identical interactions on the opposing face simultaneously
occur. Similarly, the ligands have been simplified to a basic representative scaffold.
Table 2 Assignment of charged intermediate species to clusters 3, 8 and
9 by ESI-MS
Ionic Species
m/z (calculated) m/z (observed)
Na2[Er2(OH)2(npd)4]2+·2MeOH
Na2[Er2(OH)2(npd)4]2+·4MeOH
[Er2(OH)4(npd)3]-·MeOH
779.1
811.1
1238.1
1265.3
780.3
812.1
1239.2
1264.2
1309.3
866.8
877.9
939.9
951.0
961.9
[Er2(MeO)(mmc)4]+·MeOH
Na[Er2(OH)2(mmc)4]+·H2O, MeOH 1309.2
Na2[Er2(OH)2(fpp)4]2+·2H2O, MeOH 867.0
Na2[Er2(OH)2(fpp)4]2+·5H2O
K2[Er2(OH)2(fpp)4]2+·10H2O
878.0
939.0
Fig. 12 Observed CH–p interactions, highlighted in gold, thought to
K2[Er2(OH)2(fpp)4]2+·6H2O, 3MeOH 951.0
K2[Er2(OH)2(fpp)4]2+·9H2O, 2MeOH 962.0
˚
mediate alignment of peripheral ligands. CH–p distances of 2.719 A were
˚
observed for cluster 7 (left), and 3.181 A for cluster 8 (right).
which were found to encourage intracluster interactions, which, in
the case of clusters 4–7, gave weak interactions, and in the case of 3
and 8 resulted in lanthanoid intracluster hydrogen-bonding. While
we have demonstrated that synthesis of the cubane motif can be
rationally controlled by ligand design, foreseeing the more subtle
interactions governing the final stages of cluster growth remains a
considerable challenge.
3, which is stabilised by the sterically demanding ortho nitro
ligand Hnpd. Here the observed coordination number of eight was
achieved by retention of two coordinated aqua ligands and two
2
bridging diketonate ligands coordinating in a m-h -npd fashion
across a peripheral face, as discussed above. The absence of
residual aqua ligands in clusters 4–8 suggests that ligand bridging
across the peripheral faces does occur, but is transient for each of
2
the seven coordinate metal centres. The peripheral m-h bridging
Conclusion
motif appears reliant on the orientation of the peripheral ligands,
which in turn, are influenced by the angle the substituted phenyl
rings exhibit on the capping faces. The size and nature of the ortho
functional group on the capping faces determine this angle by
maximising interaction of the ortho functional group with the m3-
OH hydroxo groups of the core (Fig. 12). The peripheral ligands
align parallel to the ortho substituted phenyl ring to maximise
CH–p interactions. Thus, the orientation of the substituted phenyl
rings at the capping positions influence bridging of the peripheral
ligands to a degree dependent on functional group identity, which,
in turn, determines the coordination number of the metal centres.
Herein the amenability of diketonate ligands to stabilise a
lanthanoid cubane motif by the addition of ortho phenyl substi-
tuted functionalities has been demonstrated and the scope of the
reaction has been investigated using three ligands of varying bulk
and substitution. The functionalities employed were nucleophilic,
We have demonstrated that manipulation of ligand sterics is a valid
approach to predetermining lanthanoid cluster motifs. Control
over sterics can be imparted by judicious ligand design. This
represents an important step towards controlling the construction
of lanthanoid cluster species and achieving strategies that allow
predetermination of a target motif. The controlled synthesis of
tetranuclear cubane clusters was achieved by the addition of
a substituent (nitro, methoxy, fluoro) at the ortho position of
two diketonate ligands that previously were known to stabilise
unrelated motifs common to lanthanoid cluster chemistry. Seven
cubane clusters have been synthesised and characterised, having
the composition [Ln4(OH)4(H2O)xL8] (L = npd, nmc, mmc, fpp;
x = 0, 2). The motifs varied in terms of intracluster interactions
dependent on the ligand employed, with an eight coordinate,
diaqua erbium cluster species 3 observed for Hnpd; clusters
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 12169–12179 | 12175
©

generated by Hnmc were found to contain solely seven coordinate
lanthanoid metals (Ln = Gd, Tb, Dy, Er) 4–7; while a mixed seven
and eight coordinate erbium cluster 8 was observed for Hmmc
and finally a diaqua erbium cluster 9 was isolated for Hfpp with
analogous composition to 3. A mechanistic discussion based on
crystallographic evidence obtained for the products 3–9 allowed
us to postulate that ligand sterics drives formation of the cubane
motif, while fluctuations in metal coordination number were a
result of varying degrees of intracluster hydrogen bonding.
128.37, 131.01, 133.47, 168.54, 210.67; IR n: 3060 m, 2964 m, 1739
m, 1672 m, 1595 s, 1569 s, 1492 m, 1447 s, 1363 s, 1280 m, 1248 s,
1219 w, 1178 w, 1151 m, 1070 m, 1025 m, 995 m, 883 m, 824 m;
HRMS(ESI): m/z = 211.0729 [M + Na]+.
(Z)-3-Hydroxy-3-(2-nitrophenyl)-1-phenylprop-2-en-1-one (Hnpd)
The title compound was obtained as orange powder after ◦column
chromatography. Yield: 870 mg, (85%); m.p. 111.0–111.7 C; 1H-
NMR (500 MHz, CDCl3, 298 K): d = 6.48 (s, 1 H, COCHCO);
3
3
7.47 (t, JH,H = 7.5 Hz, 2 H, PhH); 7.56 (t, JH,H = 7.5 Hz, 1 H,
Experimental
3
3
PhH); 7.62 (t, JH,H = 7.5 Hz, 1 H, o-NO2PhH); 7.65 (d, JH,H
=
7.5 Hz, 1 H, o-NO2PhH); 7.70 (t, 3JH,H = 7.5 Hz, 1 H, o-NO2PhH);
General considerations
3
3
7.92 (d, JH,H = 7.5 Hz, 2 H, PhH); 7.96 (d, JH,H = 7.5 Hz, 1 H,
o-NO2PhH); 15.86 (s, 1 H, OH); 13C-NMR (125 MHz, CDCl3,
298 K): d = 96.45, 124.57, 127.32, 128.88, 129.27, 131.31, 133.02,
133.15, 133.64, 133.99, 145.54, 182.73, 188.86; IR n: 3067 m, 2869
m, 1594 s, 1514 s, 1352 s, 1284 m, 1224 s, 1112 m, 1065 m, 1021 m,
963 w, 924 m, 853 m, 815 w; HRMS(ESI): m/z = 270.0760 [M +
H]+, 287.1023 [M + NH4]+, 292.0578 [M + Na]+.
All chemicals were purchased from Sigma-Aldrich and were used
as received. Solvents used were reagent grade and purified by
standard literature methods.43 1H and 13C NMR spectra were
recorded on a Varian Unity Nova 500 spectrometer. Chemical
shifts were recorded on the d scale and referenced against the
solvent. Solid-state IR spectra were recorded using a Perkin Elmer
1600 series FTIR or a Bruker Equinox 55 Infrared Spectrometer
fitted with a Specac Diamond ATR source. Infrared band
frequencies are reported in wavenumbers (cm-1) and intensities are
reported as strong (s), medium (m) or weak (w). All positive and
negative electrospray ionisation mass spectroscopy (ESI( )-MS)
was performed on a Micromass Platform II QMS spectrometer
with cone voltages varying from 35 to 50 volts. Only the isotope
of highest intensity is reported. High-resolution mass spectra
(HRMS) were recorded on a Bruker BioApex 47e FTMS fitted
with an Analytical electrospray source using NaI for accurate
mass calibration. Melting points were measured on a Stuart
Scientific Melting Point Apparatus SMP3 in an open capillary
and were uncalibrated. Elemental Analyses were performed by
Campbell Microanalytical Laboratory, Department of Chemistry,
University of Otago, Dunedin, New Zealand.
(Z)-2-(Hydroxy(2-nitrophenyl)methylene)cyclopentanone (Hnmc)
The title compound was obtained as an orange oil after column
chromatography. Yield: 606 mg, (65%); 1H-NMR (500 MHz,
3
CDCl3, 298 K): d = 1.92 (quint, JH,H = 7.5 Hz, 2 H, H2); 2.43
(t, 3JH,H = 7.5 Hz, 2 H, H3); 2.53 (t, 3JH,H = 8.0 Hz, 2 H, H1); 7.50
(d, 3JH,H = 7.5 Hz, 1 H, o-NO2PhH); 7.58 (t, 3JH,H = 7.5 Hz, 1 H,
3
o-NO2PhH); 7.68 (t, JH,H = 7.5 Hz, 1 H, o-NO2PhH); 8.02 (d,
3JH,H = 8.5 Hz, 1 H, o-NO2PhH); 15.65 (s, 1 H, OH); 13C-NMR
(125 MHz, CDCl3, 298 K): d = 20.81, 26.55, 37.05, 110.99, 124.65,
128.01, 129.65, 130.76, 133.41, 147.48, 171.46, 204.88; IR n: 2948
m, 1593 m, 1566 s, 1529 s, 1437 m, 1357 s, 1308 w, 1283 w, 1243 s,
1165 m, 1026 m, 884 w, 832 m; HRMS(ESI): m/z = 256.0587 [M +
Na]+.
Synthesis of (Z)-2-(Hydroxy(2-methoxyphenyl)-
General synthesis of Hnpd, Hnmc, Hmmc and Hpmc
methylene)cyclopentanone (Hmmc)
To an anhydrous THF (10 mL) solution of ketone (4.0 mmol, ace-
tophenone for Hnpd, cyclopentanone for Hnmc and Hpmc) cooled
to -78 ◦C was added LiHMDS (1.6 M in hexanes, 4.2 mmol). One
equivalent of acid chloride (4.0 mmol, 2-nitrobenzoyl chloride
for Hnpd and Hnmc, benzoyl chloride for Hpmc) was next added
with rapid stirring and the solution allowed to warm to room
temperature. The reaction was quenched with water (~30 mL) and
acidified to pH 5 with 1 M HCl. The solution was extracted with
ethyl acetate (3 ¥ 50 mL) and the organic extracts dried (MgSO4)
and concentrated to dryness under reduced pressure. The crude
product was then purified by column chromatography yielding the
b-diketone.
The title compound was obtained as a yellow oil after column
chromatography. Yield: 520 mg, (60%); 1H-NMR (500 MHz,
3
CDCl3, 298 K): d = 1.85 (quint, JH,H = 7.0 Hz, 2 H, H2); 2.49
(t, 3JH,H = 8.0 Hz, 4 H, H3 H1); 3.84 (s, 3 H, CH3); 6.93 (d, 3JH,H
=
3
8.0 Hz, 1 H, PhH); 6.99 (t, JH,H = 7.5 Hz, 1 H, PhH); 7.39 (t,
3JH,H = 7.5 Hz, 1 H, PhH); 7.42 (d, JH,H = 7.5 Hz, 1 H, PhH);
3
13.94 (s, 1 H, OH); 13C-NMR (125 MHz, CDCl3, 298 K): d =
21.13, 27.12, 38.09, 55.55, 111.35, 112.12, 120.48, 123.88, 129.74,
131.76, 158.52, 168.55, 214.18; IR n: 3366 m, 2969 m, 1743 s, 1605
brs, 1488 s, 1461 s, 1436 m, 1367 m, 1278 m, 1238 s, 1159 m, 1116
w, 1065 m, 1020 s, 952 m, 909 w, 879 m, 825 s; HRMS(ESI): m/z =
219.1016 [M + H]+, 241.0837 [M + Na]+.
(Z)-2-(Hydroxy(phenyl)methylene)cyclopentanone (Hpmc)
Synthesis of (Z)-3-hydroxy-3-(perfluorophenyl)-1-
phenylprop-2-en-1-one (Hfpp)
The title compound was obtained as a brown oil after column
chromatography. Yield: 685 mg, (91%); 1H-NMR (500 MHz,
3
CDCl3, 298 K): d = 1.96 (quint, JH,H = 7.5 Hz, 2 H, H2); 2.49
The title compound was obtained as an off-white solid. Yield:
3
3
(t, JH,H = 8.0 Hz, 2 H, H3); 2.87 (t, JH,H = 7.0 Hz, 2 H, H1);
941 mg, (75%); 1H-NMR (500 MHz, CDCl3, 298 K): d = 6.50 (s, 1
3
7.41–7.46 (m, 2 H, PhH); 7.77 (d, JH,H = 7.0 Hz, 2 H, PhH);
H, COCHCO); 7.49 (t, 3JH,H = 7.5 Hz, 2 H, PhH); 7.59 (t, 3JH,H
=
8.01 (d, 3JH,H = 8.0 Hz, 1 H, PhH); 14.49 (s, 1 H, OH); 13C-NMR
(125 MHz, CDCl3, 298 K): d = 21.35, 28.40, 37.62, 109.30, 128.11,
7.5 Hz, 1 H, PhH); 7.95 (d, 3JH,H = 7.5 Hz, 1 H, PhH); 15.90 (s, 1
H, OH); 13C-NMR (125 MHz, CDCl3, 298 K): d = 94.68, 99.80,
12176 | Dalton Trans., 2011, 40, 12169–12179
This journal is
The Royal Society of Chemistry 2011
©

127.45, 127.58, 128.92, 129.13, 133.38, 133.79, 143.76, 145.76,
178.86, 184.33; IR n: 3366 m, 2923 m, 1650 m, 1595 m, 1555
m, 1477 s, 1328 m, 1258 m, 1177 m, 1098 w, 1064 w, 982 s, 942 s,
864 m, 844 w, 821 m, 779 s, 711 m, 686 m; HRMS(ESI): m/z =
313.0289 [M - H]-.
1390 s, 1345 m, 1253 s, 1031 w, 886 m, 848 m, 825 m, 684 m;
C96H84N8O36Tb4 (2561.4): calcd. C 45.0, H 3.3, N 4.4; found C
44.6, H 3.5, N 4.0.
Synthesis of [Dy4(l3-OH)4(nmc)8] (6)
Synthesis of [Nd4(l3-OH)2(pmc)10] (1)
Synthesised analogously to 1. Recrystallisation of the residue by
slow evaporation of a concentrated toluene solution yielded yellow
block crystals of 6. Yield: 247 mg, (64%); m.p. 175–180 ◦C (dec);
IR n: 3605 w, 2949 m, 2852 m, 1600 s, 1524 m, 1484 s, 1445 s,
1394 s, 1279 m, 1043 w, 892 m, 851 m, 787 m, 685 m, 631 w;
C96H84Dy4N8O36 (2576.2): calcd. C 44.8, H 3.3, N 4.3; found C
44.9, H 3.4, N 4.3.
To a round bottomed flask equipped with a stirrer bead was added
Hpmc (225 mg, 1.2 mmol) and hydrated neodymium chloride
(213 mg, 0.6 mmol). Once dissolved in methanol a slow, dropwise
addition of an excess of triethylamine (1 mL) commenced. The
reaction proceeded for 12 h. After this time the solvent was
removed and the residue stirred in toluene (~2 mL) for 1 h.‡The
solution was then filtered to remove any salts and left in an open
vial to slowly evaporate yielding pale yellow crystals that were
harvested prior to complete loss of solvent. Yield: 21 mg, (7%);
m.p. 200–215 ◦C (dec); IR n: 3330 brw, 3057 m, 1595 s, 1565 s,
1495 w, 1445 s, 1369 s, 1312 w, 1267 s, 1191 w, 1075 w, 1026 w, 923
w, 880 w, 826 m; C120H122Nd4O22·Tol (2483.1): calcd. C 58.0, H 4.6;
found C 57.5, H 4.9.
Synthesis of [Er4(l3-OH)4(nmc)8] (7)
Synthesised analogously to 1. Recrystallisation of the residue by
slow evaporation of a concentrated toluene solution yielded yellow
block crystals of 7. Yield: 243 mg, (63%); m.p. 175–180 ◦C (dec);
IR n: 3610 w, 2952 m, 2857 m, 1604 s, 1527 m, 1487 s, 1449 s,
1412 s, 1347 m, 1280 s, 1260 s, 1031 w, 891 m, 850 m, 788 m, 684
m, 632 w; C96H84Er4N8O36 (2594.8): calcd. C 44.4, H 3.3, N 4.3;
found C 44.0, H 3.1, N 4.2.
Synthesis of [HNEt3][Er9(l4-O)2(l3-OH)8(pmc)16] (2)
Synthesised analogously to 1. Orange crystals were obtained by
slow diffusion of hexa◦ne into a toluene solution of 2. Yield: 64 mg,
(18%); m.p. 180–190 C (dec); IR n: 3610 w, 3333 brw, 3056 m,
1596 s, 1568 s, 1495 w, 1446 s, 1377 s, 1313 w, 1276 s, 1193 w, 1075
m, 1027 m, 923 w, 880 w, 827 m; C198H200Er9NO42 (4772.3): calcd.
C 49.8, H 4.3, N 0.4; found C 49.4, H 4.0, N 0.1.
Synthesis of [Er4(l3-OH)4(mmc)8] (8)
Synthesised analogously to 1. Recrystallisation of the residue by
slow evaporation of a concentrated toluene solution yielded pink
◦
needles of 8. Yield: 93 mg, (25%); m.p. 212–221 C (dec); IR n:
Synthesis of [Er4(l3-OH)4(H2O)2(npd)8] (3)
3645 w, 3330 m, 2943 m, 2835 m, 1593 s, 1463 s, 1400 s, 1314 w,
1267 s, 1240 s, 1179 w, 1119 m, 1021 m, 885 m, 829 m, 748 s, 629 s;
C104H108Er4O28·1.5Tol (2475.0): calcd. C 52.6, H 4.6; found C 52.6,
H 4.6.
Synthesised analogously to 1. Recrystallisation of the residue
by slow evaporation of a concentrated toluene solution yielded
yellow needles of 3. Yield: 100 mg, (23%); m.p. 135–140 ◦C
(dec); IR n: 1598 s, 1512 s, 1452 s, 1395 s, 1281 m, 1216 m,
1024 m, 855 m, 731 s, 692 s; MS(ESI): m/z = 702.2 [Er2(npd)4]2+,
781.1 [Er2(OH)(npd)4]K+·2MeOH, 2H2O. C120H84Er4N8O36·1.5Tol
(3011.2): calcd. C 50.7, H 3.2, N 3.7; found C, 51.2; H, 3.3; N, 3.4.
Synthesis of [Er4(l3-OH)4(H2O)2(fpp)8] (9)
Synthesised analogously to 1. Recrystallisation of the residue by
slow evaporation of a concentrated toluene solution spiked with
5% acetone yielded orange needles of 8. Yield: 166 mg, (34%); m.p.
Synthesis of [Gd4(l3-OH)4(nmc)8] (4)
◦
185–193 C (dec); IR n: 3604 w, 2924 w, 1707 w, 1649 w, 1598 s,
Synthesised analogously to 1. Recrystallisation of the residue by
slow evaporation of a concentrated toluene solution yielded yellow
block crystals of 4. Yield: 187 mg, (49%); m.p. 170–175 ◦C (dec);
IR n: 3608 w, 2949 m, 2854 m, 1600 s, 1524 m, 1484 s, 1444 s,
1393 s, 1344 m, 1278 s, 1030 w, 892 m, 851 m, 787 m, 685 m, 631
w; C96H84Gd4N8O36 (2554.7): calcd. C 45.1, H 3.3, N 4.4; found C
45.1, H 3.4, N 4.0.
1567 s, 1481 s, 1450 m, 1402 s, 1319 m, 1261 m, 1194 m, 1177
m, 1114 m, 1064 m, 1026 m, 986 s, 963 s, 872 m, 712 m, 689 m;
C120H56Er4F40O22·Acetone (3278.7): calcd. C 44.3, H 1.9; found C
44.2, H 1.7.
Crystal data were collected using either an Enraf-Nonius
Kappa CCD or a Bruker X8 APEX CCD instrument with
˚
monochromated Mo-Ka radiation, l = 0.71073 A. All data were
collected at 123 K, maintained using an open flow of nitrogen
from an Oxford Cryostreams cryostat. Crystalline samples were
mounted on glass fibres in viscous hydrocarbon oil. Structural
solution and refinement was carried out using SHELXL-9744
utilising the graphical interface X-Seed.45 Data were corrected
for absorption using the SADABS46 or SORTAV47 packages.
Synchrotron data was collected at 123(2) K on the MX2 beam-
line at the Australian Synchrotron, Victoria, Australia. The data
collection and integration were performed within Blu-Ice48 and
XDS49 software programs.
Synthesis of [Tb4(l3-OH)4(nmc)8] (5)
Synthesised analogously to 1. Recrystallisation of the residue by
slow evaporation of a concentrated toluene solution yielded yellow
block crystals of 5. Yield: 227 mg, (59%); m.p. 175–180 ◦C (dec);
IR n: 3600 w, 2949 m, 2855 m, 1599 s, 1524 m, 1484 s, 1444 s,
‡ In some instances product was found to precipitate from solution
allowing collection by simple filtration followed by recrystallisation to
yield pure products.
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 12169–12179 | 12177
©

Variata
4 J. Kido and Y. Okamoto, Chem. Rev., 2002, 102, 2357.
5 J.-C. G. Bunzli and C. Piguet, Chem. Soc. Rev., 2005, 34,
1048.
6 C. Benelli and D. Gutteschi, Chem. Rev., 2002, 102, 2369.
7 R. Sessoli and A. K. Powell, Coord. Chem. Rev., 2009, 253,
2328.
8 X.-J. Kong, Y. Wu, L.-S. Long, L.-S. Zheng and Z. Zheng, J. Am. Chem.
Soc., 2009, 131, 6918.
9 R. Y. Wang, H. D. Selby, H. Liu, M. D. Carducci, T. Z. Jin,
Z. P. Zheng, J. W. Anthis and R. J. Staples, Inorg. Chem., 2002, 41,
278.
10 M. Addemo, G. Bombieri, E. Foresti, M. D. Grillione and M. Volpe,
Inorg. Chem., 2004, 43, 1603.
(1): The phenyl ring of ligand 5 is disordered over two positions
with refined occupancy of 76% for the major component. There
is also disorder in the C-3 position of the five membered ring in
ligand 4. Due to this disorder the phenyl rings were constrained
as perfect hexagons and the minor component of disorder was
refined as one.
(2): The counter ion and the solvent molecules within the void
could not be adequately modelled therefore the residual electron
density was removed using platon. The total solvent accessible
void was calculated to be 4155.8 A3 and the electron count per cell
was 3089. All phenyl rings are constrained as perfect hexagons,
the five membered ring on ligand 1 had its bond lengths fixed to
ideal values and all other rings were constrained to be the same as
ligand 1. The structure was refined as a racemic twin.
(3): Ligand 7 was constrained to be the same as ligand 2. Ligand
6 is disordered with the nitro group residing on one phenyl ring
with 76.5% occupancy and on the other with 24.5% occupancy.
There are three isolated toluene solvent molecules; two have 50%
occupancies each and the third is disordered over two positions
70 : 30% occupancies. All three molecules were refined isotropically
as perfect hexagons. Additionally the methyl group in the first
molecules was fixed at an idealised distance to the ring.
11 M. Volpe, G. Bombieri and N. Marchini, J. Alloys Compd., 2006, 408,
1046.
12 L. G. Hubert-Pfalzgraf, N. Miele-Pajot, R. Papiernik and J. Vaisser-
mann, J. Chem. Soc., Dalton Trans., 1999, 4127.
13 R. Xiong, J. Zuo, Z. Yu, X. You and W. Chen, Inorg. Chem. Commun.,
1999, 2, 490.
14 V. Baskar and P. W. Roesky, Z. Anorg. Allg. Chem., 2005, 631,
2782.
15 P. W. Roesky, G. Canseco-Melchor and A. Zulys, Chem. Commun.,
2004, 738.
16 M. R. Bu¨rgstein, M. T. Gamer and P. W. Roesky, J. Am. Chem. Soc.,
2004, 126, 5213.
17 Y. Wu, S. Morton, X. Kong, G. S. Nichol and Z. Zheng, Dalton Trans.,
2011, 40, 1041.
18 P. C. Andrews, G. B. Deacon, R. Frank, B. H. Fraser, P. C. Junk, J. G.
MacLellan, M. Massi, B. Moubaraki, K. S. Murray and M. Silberstein,
Eur. J. Inorg. Chem., 2009, 6, 744.
19 P. C. Andrews, T. Beck, B. H. Fraser, P. C. Junk, M. Massi, B.
Moubaraki, K. S. Murray and M. Silberstein, Polyhedron, 2009, 28,
2123.
20 P. C. Andrews, T. Beck, C. M. Forsyth, B. H. Fraser, P. C. Junk, M.
Massi and P. W. Roesky, Dalton Trans., 2007, 5651.
21 P. C. Andrews, W. J. Gee, P. C. Junk and J. G. MacLellan, Inorg. Chem.,
2010, 49, 5016.
22 V. Baskar and P. W. Roesky, Dalton Trans., 2006, 676.
23 A. K. Jami, P. V. V. N. Kishore and V. Baskar, Polyhedron, 2009, 28,
2284.
24 X.-J. Kong, L.-S. Long, L.-S. Zheng, R. Y. Wang and Z. P. Zheng,
Inorg. Chem., 2009, 48, 3268.
25 R. Y. Wang, H. D. Selby, H. Liu, M. D. Carducci, T. Z.
Jin, J. W. Anthis and R. J. Staples, Inorg. Chem., 2002, 41,
278.
26 Z. Zheng, Chem. Commun., 2001, 2521.
27 R. Wang, H. Liu, M. D. Carducci, T. Jin, C. Zheng and Z. P. Zheng,
Inorg. Chem., 2001, 40, 2743.
(8): The cyclopentyl ring in ligand 2 has been modelled as
the same as in ligand 1. The phenyl groups in ligands 2 and
8 are disordered over two positions with refined occupancies of
50% and have been constrained as perfect hexagons, as have the
toluene solvent molecules. The thermal parameters for each of
these fragments have been restrained as one. The methyl group in
the disordered toluene molecules could not be located and the rings
were refined without hydrogen atoms. The two components of this
disorder were refined as 25% and 12.5% occupancies. There are
still large solvent accessible voids however there was insufficient
electron density to locate any further solvent molecules. Only the
(MO)4 core could sustain isotropic refinement.
(9): The thermal parameters for O2, C123 and C96 were
constrained using the ISOR command. The thermal parameters
for the carbon atoms C40 to C45 in the phenyl ring of ligand 3
were refined as one parameter. The toluene solvent molecule was
constrained as a perfect hexagon and with refined occupancies of
50%.
28 X.-M. Chen, Y.-L. Wu, Y.-X. Tong, Z. Sun and D. N. Hendrickson,
Polyhedron, 1997, 24, 4265.
29 J. C. Plakatouras, I. Baxter, M. B. Hursthouse, K. M. A. Malik, K. J.
McAleese and S. R. Drake, J. Chem. Soc., Chem. Commun., 1994,
2455.
30 K. Manseki and S. Yanagida, Chem. Commun., 2007, 1242.
31 S. Abram, U. Abram and T. Zimmermann, Acta Crystallogr., Sect. C:
Cryst. Struct. Commun., 1994, 50, 765.
32 U. Casellato, R. Graziani, G. Maccarone, R. R. Purello and M. Vidali,
J. Crystallogr. Spectrosc. Res., 1987, 17, 323.
33 G. A. Echeverria, J. L. Jios, J. C. Autino and G. Punte, Struct. Chem.,
2000, 11, 367.
34 A. Hori and T. Arii, CrystEngComm, 2007, 9, 215.
35 V. Bertolasi, P. Gilli, V. Ferretti and G. Gilli, J. Am. Chem. Soc., 1991,
113, 4917.
Acknowledgements
We gratefully acknowledge the Australian Research Council and
Monash University for supporting this work including, but not
limited to, the granting of a Postgraduate Publication Award.
This research was, in part, undertaken on the macromolecular
crystallography (MX1) beamline at the Australian Synchrotron,
Victoria, Australia.
36 V. Bertolasi, P. Gilli, V. Ferretti, G. Gilli and L. Pretto, J. Am. Chem.
Soc., 2004, 126, 3845.
37 J. Langer, M. Gartner, H. Goris and D. Walther, Synthesis, 2006,
2697.
38 S. T. Heller and S. R. Natarajan, Org. Lett., 2006, 8(13),
2675.
References
39 G. Xu, Z. Wang, Z. He, Z. Lu, C. Liao and C. Yan, Inorg. Chem., 2002,
41, 6802.
1 J.-C. G. Bu¨nzli, Acc. Chem. Res., 2006, 39, 53.
2 N. Sabbatini and M. Guardigli, Coord. Chem. Rev., 1993, 123, 201.
3 G. F. de Su, O. L. Malta, C. D. Donega, A. M. Simas, R. L. Longo,
P. A. Santa-Cruz and E. F. da Silva, Coord. Chem. Rev., 2000, 196,
165.
40 J. Bernstein, R. E. Davis, L. Shimoni and L.-N. Chang, Angew. Chem.,
Int. Ed. Engl., 1995, 34, 1555.
41 S. Datta, V. Baskar, H. Li and P. W. Roesky, Eur. J. Inorg. Chem., 2007,
4216.
12178 | Dalton Trans., 2011, 40, 12169–12179
This journal is
The Royal Society of Chemistry 2011
©

42 V. B. Di Marco and G. G. Bombi, Mass Spectrom. Rev., 2006, 25,
347.
43 W. L. F. Armarego and C. L. L. Chai, Purification of Laboratory
Chemicals, 5th Edition, Butterworth-Heinemann, 2003.
44 G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Crystallogr., 2008,
64, 112.
46 G. M. Sheldrick, SADABS, version 2.03, University of Go¨ttingen,
Go¨ttingen, Germany, 1997.
47 R. H. Blessing, J. Appl. Crystallogr., 1997, 30, 421.
48 T. M. McPhillips, S. E. McPhillips, H. J. Chiu, A. E. Cohen, A. M.
Deacon, P. J. Ellis, E. Garman, A. Gonzalez, N. K. Sauter, R. P.
Phizackerley, S. M. Soltis and P. Kuhn, J. Synchrotron Radiat., 2002, 9,
401.
45 L. J. Barbour, XSEED: A graphical interface for use with the SHELX97
program suite, J. Supramol. Chem., 2001, 1, 189.
49 W. Kabsch, J. Appl. Crystallogr., 1993, 26, 795.
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 12169–12179 | 12179
©