Heterobimetallic catechol-phosphine complexes with palladium and a group-13 element: Structural flexibility and dynamics

Article abstract of DOI:10.1039/c4dt00785a

Group-13 metal acetylacetonates [M(acac)₃] (M = Al, Ga, In) or Al(OiPr)₃ react with a complex [Pd(catphosH)₂] that may act as chelating ligand towards a second metal, or with a mixture of catechol phosphine (catphosH₂) and [PdCl₂(cod)], to give heterometallic complexes featuring either dinuclear M(catphos)₂Pd or trinuclear M{(catphos)₂Pd}₂ motifs. Characterisation of the products by crystallographic and solution NMR studies gives insight into the structural diversity and flexibility of the coordination environments of the group-13 elements and their impact on the stability of the multinuclear complexes. The results indicate that gallium and indium are the most suitable elements for the stabilisation of di- and trinuclear assemblies, respectively. Dynamic NMR spectroscopy allowed to follow the dynamic averaging of the coordination environments of the four distinguishable catechol phosphines in the indium complex [M{(catphos)₂Pd}₂]H. The results revealed that the isomerisation follows a complicated pathway involving several distinguishable proton transfer steps, and allowed to propose a mechanistic explanation for the observed isomerisation.

Full text of DOI:10.1039/c4dt00785a

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Heterobimetallic catechol-phosphine complexes
with palladium and a group-13 element: structural
flexibility and dynamics†
Cite this: Dalton Trans., 2014, 43,
8911
G. Bauer,a M. Niegerb and D. Gudat*a
Group-13 metal acetylacetonates [M(acac)3] (M = Al, Ga, In) or Al(OiPr)3 react with a complex [Pd(cat-
phosH)2] that may act as chelating ligand towards a second metal, or with a mixture of catechol phos-
phine (catphosH2) and [PdCl2(cod)], to give heterometallic complexes featuring either dinuclear
M(catphos)2Pd or trinuclear M{(catphos)2Pd}2 motifs. Characterisation of the products by crystallographic
and solution NMR studies gives insight into the structural diversity and flexibility of the coordination
environments of the group-13 elements and their impact on the stability of the multinuclear complexes.
The results indicate that gallium and indium are the most suitable elements for the stabilisation of di- and
trinuclear assemblies, respectively. Dynamic NMR spectroscopy allowed to follow the dynamic averaging
of the coordination environments of the four distinguishable catechol phosphines in the indium complex
[M{(catphos)2Pd}2]H. The results revealed that the isomerisation follows a complicated pathway involving
several distinguishable proton transfer steps, and allowed to propose a mechanistic explanation for the
observed isomerisation.
Received 15th March 2014,
Accepted 30th April 2014
DOI: 10.1039/c4dt00785a
www.rsc.org/dalton
phosphines like 1, which was used as platform for larger
supramolecular architectures,3 or 2, which forms a scaffold for
Introduction
Functional phosphines with additional, electronically different the construction of defined heterometallic complexes of type A
coordinating groups receive persistent attention as hybrid (e.g. with M′ = Pd, Pt and M(L)n = MoO2, WO2, VO,4 or M(L)n =
ligands for uses in coordination chemistry and catalysis.1 For GaCl, SnCl2, BiCl5) and B (M(L)n = Zr, x = 05 or M(L)n
=
instance, hydroxyphosphines2 combine a phosphorus atom, Y(DMF), x = 1;6 Scheme 1). The terminal oxygen atoms of cate-
which is generally considered a “soft” donor centre, with one cholate ligands in A and B still exhibit some basicity which
or several alcohol or phenol moieties that can serve as neutral
or (after deprotonation) anionic “hard” metal binding sites.
If the molecular backbone permits to connect electronically
different functions with the same metal atom, such ligands
can form chelate complexes which may have potential appli-
cations in catalysis.1,2 However, the dissimilar binding prefer-
ences of the coordinating groups can as well be used to direct
metal atoms specifically to the hard (O) and soft (P) donor
site. This feature enables to control the ambident behaviour of
hydroxyphosphines in mononuclear complexes, but permits to
employ such ligands also as smart building blocks for the
selective assembly of heterometallic complexes. Particularly
illustrative examples have been reported for ditopic catechol
aInstitut für Anorganische Chemie, University of Stuttgart, Pfaffenwaldring 55, 70550
Stuttgart, Germany. E-mail: gudat@iac.uni-stuttgart.de; Fax: +49 711 685 64186
bLaboratory of Inorganic and Analytical Chemistry, University of Helsinki, P.O. Box
55 (A.I. Virtasen aukio 1), FIN-00014 Helsinki, Finland
†Electronic supplementary information (ESI) available: Cif-files, selected NMR
spectra of 7 and 8. CCDC 990577–990580. For ESI and crystallographic data in
CIF or other electronic format see DOI: 10.1039/c4dt00785a
Scheme 1 Molecular structures of catechol phosphine ligands 1, 2
(= H2catphos) and complexes [M(L)n(μ-catphos)2M’] (A) and [M(L)n-
{(μ-catphos)2Pd}2]x− (B).
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 8911–8920 | 8911

View Article Online
Paper
Dalton Transactions
gallium (M(L)n = GaCl; M′ = Pd(II)).5 The total cation charge in
these species is, like in all other previously prepared type A
complexes, four or larger. Since at the same time attempts to
prepare anionic complexes with lower cation charges failed,8 it
seems that formation of a stable bimetallic framework requires
a high electrostatic contribution to the metal–ligand bonding,
and we consider the use of trivalent group-13 cations therefore
an essential prerequisite for successful synthesis of the target
compounds. As thallium exhibits a strong preference for the
+I-oxidation state and Tl(III) compounds are often potent oxi-
dants, we focused on complexes combining palladium with
one of the lighter group-13 elements, boron through indium.
In the course of these studies, it further turned out that all
attempts to prepare boron-containing complexes remained
unsuccessful. We attribute this failure to the fact that a boron
atom is unable to adopt a coordination number of 5 and
form a neutral type A complex [B(X)(catphos)2Pd], whereas a
hypothetical cation [B(catphos)2Pd]+ is presumably too
strained and vulnerable to attack of nucleophilic solvents or
counter ions to be stable.
Scheme 2 Synthetic approaches to heterometallic complexes A.
allows them to act as hydrogen bond acceptors4 or undergo
protonation.6
Interestingly, it was found that the bite angle of the multi-
dentate array varies with the size and preferred coordination
geometry of the template, and the choice of suitable templates
from different groups of the periodic table allowed thus to con-
struct e.g. trans-chelating bisphosphine moieties with bite
angles close to 180° 7 and cis-chelating bisphosphine units
with bite angles between 102° to 97° 5 from the same catechol
phosphine building block. The P2O2 donor set around the Pd
atom in type A complexes shows hemilabile behaviour, and
some complexes were found to exhibit catalytic activity in CC
cross-coupling reactions.5
The synthesis of the oligonuclear complexes A, B can be
either accomplished in a single step by spontaneous conden-
sation of ligand 2 with two simple metal salts,5,6 or stepwise
by introducing the metal atoms one after the other
(Scheme 2).4,7 In the sequential reactions, the first metal atom
interacts with one coordination site of two catechol phos-
phines to produce complexes of types C or D, respectively,
which then act as bi- or multidentate donors towards a second
metal atom. Since the coordinative bonds between the catechol
phosphine and the metal template are kinetically labile and
their formation is in principle reversible, complexes C, D were
also denoted supramolecular ligands.4,5
Preparation of bimetallic Pd,M(III) complexes with M = Al,
Ga, In, was pursued via both the “self-assembly” route (path
(i), Scheme 2),5 and a stepwise process involving reaction of
catechol phosphine 2 with [PdCl2(cod)] (cod = 1,5-cycloocta-
diene) to give chelate complex D and further condensation
with metal trichlorides or tris-acetylacetonates (reactions (iv),
(v) in Scheme 2).4 The one-step reaction, which had previously
provided
a high yielding access to Pd,Ga-complex 3a
(Scheme 3, (i)),5 proved unfeasible for the synthesis of alu-
minium or indium complexes; spectroscopic studies (31P
NMR, ESI-MS) gave no evidence for the formation of defined
products. In contrast, reaction of 2 with [PdCl2(cod)] and [Ga-
(acac)3] allowed to access a Pd,Ga-complex under base-free
conditions (Scheme 3, (ii)). The product was isolated by direct
crystallization from the reaction mixture, and characterisation
by NMR data and single-crystal X-ray diffraction (see below)
revealed the presence of a solvate complex 3b which differs
from 3a in the coordination of an additional solvent molecule
(DMF) at the gallium centre.
As in the case of type A complexes with palladium and early
transition metals,4 otherwise inaccessible Pd,Al- and Pd,In-
complexes were readily available from the pre-formed Pd-
chelate [Pd(catphosH)2] (4, type D with M′ = Pd in Scheme 2)
In this work, we report on the outcome of a systematic
survey of heterometallic complexes containing palladium and
a group-13 element. The results provide additional insight into
the structural flexibility and dynamics of the template-centred
multidentate donor units, and cast some light on the factors
that govern the aggregation of individual ligating units.
Results and discussion
Syntheses
Scheme 3 Synthesis of heterometallic Pd,Ga-complexes from 2 and
Examples of complexes A with group-13 elements are as yet
GaCl3 or [Ga(acac)3] (dmf
= dimethyl formamide; cod = 1,5-
known with boron (M(L)n = B; M′ = Cu(I), Ag(I), Au(I))7 and cyclooctadiene).
8912 | Dalton Trans., 2014, 43, 8911–8920
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Scheme 5 Synthesis of heterometallic Pd,In-complexes from 4.
in the (+)-ESI-MS is attributable to a cation [Pd(catphos)2In]+
formed by loss of chloride from the basic binuclear unit of the
complex, but the spectrum also contains a low intensity peak
at higher m/z ratio that is attributable to a trinuclear Pd2In
species. The deliberate synthesis of a complex with this struc-
Scheme 4 Synthesis of heterometallic Pd,Al-complexes from 4.
and metal tris-acetylacetonates [M(acac)3] (M = Al, In) or InCl3, tural motif was feasible by reacting equimolar amounts of 4
respectively. Reactions of 4 with [Al(acac)3] in polar aprotic sol- and [In(acac)3] at 50 °C in methanol. The product was isolated
vents were sluggish, but clean transformations were observed in excellent yield (86% with respect to 4) after partial evapor-
in methanol at 50 °C (in order to overcome the low solubility ation of the solvent and crystallization. Even if the material
of 4 at ambient temperature); presumably, the protic nature of was unsuitable for single-crystal X-ray diffraction, its identity
this solvent and its ability to act as hydrogen bridge donor as neutral complex 8 was established by peaks of pseudo-mole-
facilitate the proton transfer steps associated with the displace- cular ions [In(catphos)4Pd2]H2 and [In(catphos)4Pd2]NaH+ in
+
ment of the acac-ligands on Al under concomitant generation the (+)-ESI-MS, and confirmed by solution NMR studies
of the weak acid acacH. Reaction of equimolar amounts of 4 (see below).
and [Al(acac)3] under these conditions afforded thus after
work-up a moderate yield of Pd,Al-complex 5 (Scheme 4) which
Crystallographic studies
was identified by a single-crystal XRD study (see below).
Spectroscopic characterisation of the product was rather
difficult due to its high tendency to hydrolyse in solution.
ESI-MS and NMR data confirmed the presence of the intact
complex (see Experimental), but indicated also that extensive
protolysis to give 4 along with acacH and unidentified Al-con-
taining species took place. ESI mass spectra allowed further to
The molecular structures of 3b, 5–7 and selected structural
parameters are displayed in Fig. 1–4. All four complexes share
a distorted square planar coordination at palladium. The
coordination number at the group-13 element is always six,
+
detect a cation [Pd2(μ-catphos)4Al]H2 , the presence of which
might point to the formation of a trinuclear complex, possibly
of type B (Scheme 1), in the decomposition process.
Since 4 also undergoes condensation with metal alkoxides,4
we explored its behaviour towards Al(OiPr)3. Reaction of equi-
molar amounts of reactants in isopropanol at 70 °C gave a crys-
talline precipitate that was identified as trinuclear type B
complex 6 (Scheme 4) by solid-state NMR and single-crystal
XRD studies. Even if the ESI-MS displayed the ion peak of the
protonated complex, NMR data gave no conclusive results. The
spectra of fresh solutions displayed broadened signals that
prevented a concise structural assignment. Aged solutions
showed evidence for hydrolysis.
Reaction of 4 and InCl3 requires no protic solvent but pro-
ceeds, by analogy to the formation of gallium complex 3a,
smoothly in a CH2Cl2–Et3N mixture. Work-up afforded a
product which was readily identified as tetranuclear complex 7
(Scheme 5) by a single-crystal XRD study. The dominant peak
Fig. 1 ORTEP-style representation of the molecular structure of 3b in
the crystal. Thermal ellipsoids are drawn at 50% probability level; hydro-
gen atoms are omitted for clarity. Selected distances (in Å) and angles
(in °): Pd1–P1 2.247(2), Pd1–P2 2.248(2), Pd1–O1 2.060(3), Pd1–O20
2.056(3), Ga1–O1 2.025(4), Ga1–O2 1.872(4), Ga1–O20 1.998(4), Ga1–
O21 1.891(4), Ga1–Cl1 2.195(2), Ga1–O1D 2.823(4), O20–Pd1–O1
74.8(1), P1–Pd1–P2 102.2(1).
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 8911–8920 | 8913

View Article Online
Paper
Dalton Transactions
Fig. 2 ORTEP-style representation of the molecular structure of 5 in
the crystal. Thermal ellipsoids are drawn at 50% probability level; hydro-
gen atoms in CH bonds and the second solvent molecule in the struc-
ture are omitted for clarity. Selected distances (in Å) and angles (in °):
Pd1–P1 2.224(2), Pd1–P2 2.230(2), Pd1–O1 2.017(5), Pd1–O20 2.087(5),
Al1–O1 1.920(5), Al1–O2 1.834(5), Al1–O20 1.964(5), Al1–O21 1.862(5),
Al1–O3 1.920(5), Al1–O4 1.851(6), O21⋯O1M 2.652(9), O1–Pd1–O20
75.1(2), P1–Pd1–P2 106.1(1).
Fig. 4 ORTEP-style representation of the molecular structure of 7 in
the crystal (top) and reduced representation showing the metal coordi-
nation environment in one half of the molecule (bottom). Thermal ellip-
soids are drawn at 50% probability level; hydrogen atoms are omitted for
clarity. Selected distances (in Å) and angles (in °): In1–O1 2.271(5), In1–
O2 2.097(5), Pd1–P1 2.250(2), Pd1–P2 2.254(2), Pd1–O1 2.080(4), Pd1–
O20 2.082(5), In1–O20 2.231(4), In1–O21 2.191(5), In1–O59 2.162(4),
In1–Cl1 2.413(2), Pd2–P3 2.257(2), Pd2–P4 2.263(2), Pd2–O39 2.078(4),
Pd2–O58 2.082(5), In2–O21 2.153(4), In2–O39 2.267(5), In2–O40
2.098(5), In2–O58 2.218(4), In2–O59 2.201(5), In2–Cl2 2.414(2), O1–
Pd1–O20 80.6(2), P1–Pd1–P2 100.0(1), O39–Pd2–O58 81.2(2), P3–
Pd2–P4 101.8(1).
but individual complexes show a remarkable variety in coordi-
nation geometries and bonding modes of the bridging cate-
chol phosphine units, as well as a large conformational
flexibility of the four-membered PdMO2-rings that form the
core of the heterometallic assembly.
The square pyramidal layout of the chlorine and catecho-
lato-oxygen atoms around the gallium centre and the bowl-like
arrangement of the catecholate ring planes in 3b (Fig. 1)
resemble closely the characteristic structural features of 3a.5
The oxygen atom of the extra DMF ligand occupies a trans-
position with respect to the chloride, but the Ga1–O1D separ-
ation (2.823(4) Å) is noticeably longer than the other Ga–O dis-
tances (1.872(4)–2.025(4) Å) and suggests to describe the
coordination number at gallium as 5 + 1 rather than 6. In
total, the geometrical distortions induced by the additional
ligand are – apart from the different orientation of phenyl
rings – relatively small, which is also illustrated in the simi-
larity of O–Pd–O and P–Pd–P bite angles in 3a5 (O–P–O
74.3(1)°, P–Pd–P 101.8(1)°) and 3b (O–P–O 74.8(1)°, P–Pd–P
102.2(1)°).
Fig. 3 ORTEP-style representation of the molecular structure of 6 in
the crystal (top) and reduced representation without phenyl rings
(bottom). Thermal ellipsoids are drawn at 50% probability level; hydro-
gen atoms in CH bonds are omitted for clarity. Only one of two split
positions for the hydrogen atom on O2 is displayed. Selected distances
(in Å) and angles (in °): Pd1–P1 2.214(3), Pd1–P2 2.220(3), Pd1–O1
2.052(5), Pd1–O20 2.005(6), Al1–O1 2.145(8), Al1–O20 2.094(7), Al1–
O21 2.021(7), O1–Pd1–O20 76.3(3), P1–Pd1–P2 97.2(1).
The molecular structure of 5 (Fig. 2) differs from that of 3b
mainly in the fact that the oxygen atoms of the three bidentate
ligands in the Al-tris-chelate unit occupy mutual cis-positions,
and that the two catecholate units have therefore markedly
8914 | Dalton Trans., 2014, 43, 8911–8920
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
different orientations: one μ-bridging oxygen atms (O1) retains of 9 are linked via chloride bridges, whereas those of 7 are
a nearly trigonal planar coordination sphere (sum of bond held together by μ-bridging oxygen donors and the chlorides
angles 347(1)°), and the adjacent benzene ring remains nearly act as terminal ligands; furthermore, the cisoid orientation of
parallel to the central AlO2Pd ring plane; the coordination at the catecholate ligands in 9 imposes effective local mirror sym-
the other μ-bridging oxygen atom (O20) is pyramidal (sum of metry around the BiO2Pd rings, whereas 7 exhibits a transoid
bond angles 311(1)°), and the flanking benzene ring exhibits a alignment with effective local C2-symmetry for each
strong tilt with respect to the AlO2Pd ring. The Al–O distances In(catphos)2Pd unit.
in 5 (1.834(5)–1.964(5) Å) are generally shorter than the Ga–O
Comparing the local environment of the palladium atom in
distances in 3b (1.872(4)–2.025(4) Å), in line with the smaller 3b, 5–7 confirms previous evidence5 that variation of the M(L)n
effective ion radius for Al3+ (0.535 Å, vs. 0.620 Å for Ga3+ ref. 9). template in a complex [M(L)n(catphos)2Pd] allows to tune
As had been previously noted,5 decreasing the size of the tem- P–Pd–P bite angles over a remarkably wide range (from 97° in
plate increases the bite angle of the bis-phosphine unit at pal- 5 to 106° in 6). The structural consequences of a formal
ladium (P–Pd–P 106.1(1) vs. 102.2(1) in 3b). Attachment of a exchange of templates are, however, not foreseeable, and the
hydrogen bridge donor to one of the terminal catecholato- finding that both the largest and smallest bite angles occur in
oxygen atoms has precedence in an analogous Ti,Pd-complex4 aluminium-containing complexes indicate that the variations
and induces a slight lengthening of the Al–O bond (1.862(5) Å cannot be explained by simple rules. The conjecture that
for O21–Al1 vs. 1.834(5) for O2–Al1).
larger templates decrease the separation between the phos-
Complex 6 (Fig. 3) has crystallographic C2-symmetry. The phorus atoms and enforce thus smaller bite angles5 holds thus
aluminium atom binds to six of the eight oxygen atoms of two at best for a very limited subset of compounds and cannot be
chelating {(catphos)2Pd} “pseudo-ligands” in a highly distorted generalised.
coordination environment that is intermediate between
regular octahedral and trigonal prismatic. Analysis of the dis-
Solution NMR studies
tortion with the “twist angle” model reveals a twist of some Apart from helping in the constitutional assignment of com-
22° between the basal faces of the prism, which corresponds plexes 3b, 5–8, solution NMR studies yielded information on
to a 37% distortion towards octahedral.10 The three oxygen conformational dynamics and ligand exchange processes
donor atoms in each unit are found at larger distances (Al–O which provide deeper insight into the flexibility and stability of
2.021(7)–2.145(8) Å) than in 5. The distance to the remaining the multimetallic frameworks.
1
oxygen atom (Al–O2 3.626(9) Å) is close to the sum of van-der-
The H NMR spectra of Al(III) complexes 5, 6 contain broad
Waals radii (3.75 Å11). Electroneutrality requires the presence signals that cannot be assigned to individual species but
of one further proton which is split between two sites close to suggest that the complexes have experienced a structural col-
the non-bound oxygen atoms. The O2⋯O21A distance (2.673(10) lapse or underwent partial hydrolysis in solution. This
Å) and the large O2–H⋯O21A angle (153°) suggest that interpretation was confirmed by detection of the signal of 4 as
this proton is involved in a hydrogen bond which connects the major constituent in 31P NMR spectra of aged solutions of
chelating pseudo-ligands. The coordination at the μ-bridging both complexes. Likewise, the 27Al chemical shifts of 6 in solu-
O1 atom of the monodentate catechol(ate) is planar (sum of tion (77 ppm) and in the solid state (5 ppm) display a large dis-
bond angles 360°) whereas the O20 atom in the bidentate cate- crepancy which is incompatible with the presence of the same
cholate is slightly pyramidal (sum of bond angles 340°). The molecular structure in both phases; the solution shift is rather
O20–Pd1–O1 angle (76.3(3)°) is slightly larger and the P1–Pd1– indicative of an acetylacetonato- or catecholato-complex with
P2 bite angle (97.2(1)°) distinctly smaller than in 5. Both five-coordinate aluminium,12 or a dynamically exchanging
effects together render the deviation from regular square mixture of species with different metal coordination numbers.
planar coordination at palladium less pronounced; a compari-
In combination with ESI-MS data, these findings led us to
son of Al–O distances in 5 and 6 suggests, however, that this conclude that the binuclear complexes 5, 6 exhibit limited
geometric relaxation is offset by destabilisation of the alu- stability in solution and tend to disassemble into different
minium coordination sphere.
species. The formation of substantial amounts of palladium
Tetranuclear complex 7 (Fig. 4) is best described as the complex 4 suggests cleavage of aluminium–catecholate-bonds
product of mutual donor–acceptor interaction of two binuclear as major fragmentation pathway. In view of the abundance of
complexes [In(Cl)(μ-catphos)2Pd]. The pairing creates a new known Al(III) catecholates,13 the instability of 5, 6 is presum-
rhombic In2O2 ring in which bridging In1–O59 and In2–O21 ably not caused by the incompatibility of metal and bidentate
distances (2.153(4)–2.162(4) Å) are shorter than internal In1– catecholate ligands, but by a mismatch between the preorgani-
O21 and In2–O59 distances (2.191(5)–2.201(5) Å) between sation of the two catechol units on the Pd(II) template of 4 and
atoms in the same fragment. The dimerization is presumably the steric requirements of the metal centre – the molecular
induced by the coordinative unsaturation of the large group-13 structure of 4 is apparently too rigid to arrange the catecholate
template (ionic radius of In3+ 0.80 Å9) in an isolated [In(Cl)- units for optimum binding of the small Al(III) ion.
(catphos)2Pd] fragment. Even if this aggregation has pre-
The 1H and 31P NMR spectra of the gallium complexes 3a,b
cedence in a bismuth complex [Bi(Cl)(catphos)2Pd] (9),5 both display defined signals which do not change with time. The
species show some notable structural differences: the subunits benzylic protons in 3b give rise to two separate doublets rather
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 8911–8920 | 8915

View Article Online
Paper
Dalton Transactions
than a single broad signal as in 3a,5 which implies a slower
inversion rate of the square-pyramidal ClO5-array. This behav-
iour is readily accounted for if one assumes that the DMF
ligand in 3b is not lost in solution, and dynamic isomerization
at the six-coordinate gallium occurs less easily than at the five-
coordinate metal centre in 3a. The persistence of 3a,b in solu-
tion suggests that these complexes are thermodynamically
stable, which is also in accord with the observation that their
formation via spontaneous condensation of ligand 2 with
appropriate metal salts is highly specific and does not yield
detectable amounts of side-products.
The 1H and 31P NMR spectra of indium complex 7 reveal
severely broadened resonances which sharpen, however, at low
temperature (−70 °C) to produce defined signals with a similar
pattern as 3a,b. The room temperature lineshape changes
upon prolonged storage of the solutions eventually into a
different set of broadened signals which appear at comparable
chemical shifts as the resonances of 8 (see below). Similar
changes were also observable when residual water was present
in the samples. Whereas the initial (reversible) temperature
dependent signal broadening is presumably associated with
conformational dynamics (e.g. inversion of the non-planar
chelate rings at the palladium atom) and possible dissociation
equilibria involving separation of the molecular subunits of 7,
the long term changes suggest that structural reorganisation
and concomitant formation of a new species take place.
Complex 8 is peculiar as its 1H and 31P NMR spectra exhibit
resonances of four distinguishable catechol phosphine
ligands. The pairwise coupling of the 31P NMR signals to yield
two AB-patterns indicates that the ligands form two
Pd(catphos)2 units with cis-coordinated phosphine donors. An
additional 1H NMR signal at low field reveals a single acidic
proton which is presumably involved in an intramolecular
hydrogen bond. In connection with the ESI-MS data, these
findings support the presence of a complex containing a
central In(III) ion ligated by a mono-anionic {(catphosH)-
(catphos)2Pd}− and
a
dianionic {(catphos)2Pd}2− unit
(Scheme 5). Two-dimensional 1H and 31P EXSY spectra dis-
close, however, that the molecular structure of 8 is not static,
but that reversible exchange processes induce dynamic aver-
aging of the unlike ligand environments. Analysis of the
changes in cross-peak patterns with temperature and mixing
time reveal that full equalisation is accomplished in several
stages which exhibit different activation energies. A consistent
mechanistic picture can be developed if 8 is assigned a similar
structure as aluminium complex 6. The transformation with
the lowest energy barrier involves pairwise exchange between
catechol phosphines in different {(catphos)2Pd} units but does
not mix signals of ligands at the same palladium centre
(Fig. 5a), and can be associated with a transfer of the hydro-
xylic proton along the H-bridge to a metal-bound oxygen atom
in the other {(catphos)2Pd} unit, and subsequent rearrange-
ment of the catecholates (Scheme 6, reaction (i)).
Fig. 5 Expansion of the benzylic region of 1H EXSY spectra of 8 (reson-
ances marked with an asterisk are due to MeOH). Top: spectrum
recorded at 253 K with 125 ms mixing time. Red circles indicate cross-
peaks arising from intramolecular NOE and J-coupling breakthrough;
red arrows denote peaks due to exchange between different catphos
ligands. Bottom: spectrum recorded at 303 K with 50 ms mixing time.
Red circles denote missing cross-peaks which would indicate exchange
between anisochronous benzylic protons in the same ligand; these
peaks become visible when longer mixing times (>200 ms) are employed
(see ESI†).
remote free catecholate site (Scheme 6, reaction (ii)). Finally,
exchange between inequivalent benzylic protons in each
ligand indicates the onset of conformational inversion of the
non-planar, six-membered Pd-centred chelate rings.
The conservation of the multiplet structure in the 31P NMR
spectra suggests that all exchange processes are intramolecu-
lar. It must be admitted that the available spectroscopic data
At the next stage, averaging between the ligands in the
same {(catphos)2Pd} unit is observed (Fig. 5b), which we
explain as a result of a transfer of the hydroxylic proton to the
8916 | Dalton Trans., 2014, 43, 8911–8920
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
(L = –) or accommodates a secondary ligand (L = DMF) seem to
provide a perfect match, which is reflected in both high for-
mation tendency and increased stability in solution. The still
larger In(III) cation supports formation of a complex [InCl-
(catphos)2Pd], which dimerises in the solid state under for-
mation of intermolecular In–O bonds but rearranges in solu-
tion to give rise to a trinuclear assembly of composition
[In{(catphos)2Pd}2]H. The superior stability of this species is
underlined by its formation as main product in the reaction of
equimolar amounts of In(acac)3 and [Pd(catphosH)2], and con-
firms the hypothesis5,6 that the tendency of large metal
cations to adopt higher coordination numbers provides a
strong driving force for formation of the M{(catphos)2Pd}2
binding motif. NMR studies allowed to follow the dynamic
averaging of the four different catechol phosphine environ-
ments in the trinuclear framework, and to propose a mechan-
ism for this process which involves a sequence of discernible
proton transfer and ligand rearrangement steps, and may serve
as a model for the explanation of analogous dynamics in
related complexes.
Scheme 6 Schematic representation of the proposed mechanism
for dynamic averaging between distinguishable catechol phosphine
ligands in 8.
do not allow secure assignment of the coordination geometry
around the central indium atom. However, the mechanistic
explanation remains valid even if the metal centre extends its
coordination by binding of one or both of the “free” catecho-
late oxygen donors, and may thus serve as a general model
which can also explain the rapid dynamic averaging of ligand
environments in other protonated catechol phosphine
complexes.6
The finding that even extended storage of solutions of 8
does not result in perceptible disintegration attests this
complex a higher stability compared to binuclear 7 (which is
also fully in accord with the observed formation of 8 during
degradation of 7). This trend is in contrast to the case of the
gallium complexes 3a,b, which give no evidence for the for-
mation of an analogous trinuclear assembly, and highlights
once more the importance of template size as a critical factor
for the stabilisation of multimetallic aggregates with bridging
catechol phosphines.
Experimental
Materials and methods
All manipulations were carried out under dry argon using stan-
dard Schlenk techniques. Solvents were purified by standard
procedures. NMR spectra: Bruker Avance 250 (1H: 250.1 MHz,
31P: 101.2 MHz) or Avance 400 (1H: 400.1 MHz, 27Al:
104.0 MHz, 31P: 162.9 MHz) at 30 °C. Chemical shifts were
referenced to ext. TMS (1H), 1.1 M Al(NO3)3 (27Al; Ξ =
26.056859 MHz), or 85% H3PO4 (31P; Ξ = 40.480742 MHz).
Coupling constants are given as absolute values. Elemental
analyses: Perkin-Elmer 2400 CHSN/O Analyser. (+)-ESI mass
spectra: Bruker Daltonics MicroTOF Q instrument. MeOH was
used as solvent unless specified otherwise. Melting Points
were determined with a Büchi B-545 melting point apparatus
in sealed capillaries.
Complex 3b. [Ga(acac)3] (104 mg, 0.32 mmol), [PdCl2(cod)]
(102 mg 0.32 mmol) and Et3N (0.2 ml, 1.44 mmol) were added
to a solution of 2 (200 mg, 0.65 mmol) in 10 ml of DMF. The
Conclusions
The ability of catechol phosphine 2 to support di- and trinuc- mixture was stirred for 3 h and was then evaporated to
lear heterometallic complexes containing palladium and the dryness. Recrystallization of the red brown residue from DMF–
group-13 elements aluminium, gallium, and indium was Et2O (1 : 1) yielded 162 mg (0.19 mmol, 60%) of 3b of
1
demonstrated. The target complexes were either accessed via m. p. 215–216 °C. – H NMR (CDCl3): δ = 7.62–7.25 (m, 8 H,
spontaneous reactions between the ligand and two simple o-Ph), 7.09–6.87 (m, 12 H, m-/p-Ph), 6.61 (d, 3JHH = 7.8 Hz, 2 H,
metal salts, or by condensation of pre-formed palladium C6H3), 6.44 (t, 3JHH = 7.8 Hz, 2 H, C6H3), 6.02 (d, 3JHH = 7.8 Hz,
2
chelate [Pd(catphosH)2] with group-13 element acetylaceto- 2 H, C6H3), 3.95 (dd, JHH = 17.0 Hz, 3JPH = 12.2 Hz, 2 H, CH2),
2
2
nates or alkoxides. Crystallographic and solution NMR data 3.35 (dd, JHH = 17.0 Hz, JPH = 11.9 Hz, 2 H, CH2). – 31P{1H}
indicate that the match between the size of the group-13 ion NMR (CDCl3): δ = 64.1 (s). – (+)-ESI-MS: m/z = 786.99 [M − Cl]+.
and the preorganisation of the donor centres in the rigid
– (–)-ESI-MS: m/z = 840.97 [M + – Anal. for
OH]−.
{Pd(catphos)2} unit has a critical effect on complex stabilities. C38H30ClGaO4P2Pd·DMF (897.28 g mol−1): calcd C 54.88,
Heterometallic assemblies with Al(III) are unstable in solution H 4.16, N 1.56%; found C 53.88, H 4.56, N 1.12%.
since this cation is obviously too small for the Pd-centred
Complex 5. [Al(acac)3] (54 mg, 0.17 mmol) was added to a
pseudo-chelate ligand. Complexes [GaCl(L)(catphos)2Pd] built suspension of 4 (120 mg, 0.17 mmol) in 15 ml of MeOH. The
around a larger Ga(III) centre that is either pentacoordinate mixture was heated to 50 °C and stirred for 3 h until all solids
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 8911–8920 | 8917

View Article Online
Paper
Dalton Transactions
had dissolved. The red solution formed was then concentrated dark red, microcrystalline solid which was collected by fil-
to approx. one third of its original volume, allowed to cool to tration and dried under reduced pressure to yield 168 mg
r.t., and stored for 48 h at 4 °C. The red crystalline precipitate (0.11 mmol, 86% with respect to 4) of 8, m. p. 136–137 °C. –
formed was collected by filtration and dried under reduced 1H NMR (400 MHz, 303 K, CD2Cl2): δ = 11.32 (s, 1 H, OH), 8.08
1
pressure. Yield: 43 mg (51 μmol, 31%), m. p. 108–110 °C. – H (m, 2 H, Ph), 7.50 (m, 1 H, Ph), 7.45 (m, 2 H, Ph), 7.41–7.17
NMR (CD2Cl2): δ = 7.63–7.52 (br m, 2 H, Ph), 7.50–6.78 (m, 18 (m, 18 H, Ph), 7.17–6.94 (m, 16 H, Ph + 1 H, C6H3), 6.94–6.84
3
H, Ph), 6.75–6.54 (m, 4 H, C6H3), 6.20 (m, 2 H, C6H3), 5.51 (s, (m, 1 H, Ph + 2 H, C6H3), 6.74 (d, JHH = 7.4 Hz, 1 H, C6H3),
3
3
1 H, CH), 3.39 (br, 4 H, CH2), 1.93 (br s, 6 H, CH3). – 31P{1H} 6.55 (d, JHH = 7.4 Hz, 2 H, C6H3), 6.40 (t, JHH = 7.6 Hz, 2 H,
NMR (CD2Cl2): δ = 79.4 (s). – (+)-ESI-MS: m/z = 745.05 [M − C6H3), 6.21 (t, 3JHH = 7.5 Hz, 1 H, C6H3), 5.90 (d, 3JHH = 7.7 Hz,
acac]+, 845.10 [M + H]+. – Anal. for C43H37AlO6P2Pd·CH3OH 1H, C6H3), 5.85 (d, JHH = 7.9 Hz, 1 H, C6H3), 5.70 (d, JHH
=
3
3
(877.15 g mol−1): calcd C 60.25, H 4.71%; found C 60.58, H 7.2 Hz, 1 H, C6H3), 5.61 (d, 3JHH = 7.2 Hz, 1 H, C6H3), 4.30 (dd,
2
2
4.66%.
2JPH = 6.2 Hz, JHH = 12.4 Hz, 1 H, CH2), 3.94 (dd, JPH
=
2
2
Complex 6. Al(OiPr)3 (34 mg 0.17 mmol) was added to a sus- 4.3 Hz, JHH = 12.0 Hz, 1 H, CH2), 3.08 (dd, JPH = 16.8 Hz,
pension of 4 (120 mg, 0.17 mmol) in 10 ml iPrOH. The 2JHH = 12.4 Hz, 1 H, CH2), 3.03 (dd, 2JPH = 18 Hz, 2JHH = 14 Hz,
2
mixture was heated to 70 °C and stirred for 3 h until all solids 1 H, CH2), 2.55 (t, JPH
=
2JHH = 14 Hz, 1 H, CH2), 2.44 (d,
had dissolved. The red solution formed was then concentrated 2JHH = 13.0 Hz, 2 H, CH2), 1.98 (d, JPH = 5.6 Hz, JHH
=
2
2
to approx. one third of its original volume, and allowed to cool 12.3 Hz, 1 H, CH2). – 31P{1H} NMR (161.9 MHz, 303 K,
2
2
to r.t. A bright red solid precipitated within 24 h. The super- CD2Cl2): δ = 71.6 (d, JPP = 49 Hz), 68.2 (d, JPP = 49 Hz), 54.9
natant solution, which had become almost colourless, was dis- (dd, JPP = 33 Hz), 48.9 (dd, JPP = 33 Hz). H NMR (400 MHz,
carded, and the crystalline residue dried under reduced 253 K, CD2Cl2): δ = 11.35 (s, 1 H, OH), 8.03 (m, 2 H, Ph), 7.53
2
2
1
pressure. Single-crystals suitable for
a XRD study were (m, 1 H, Ph), 7.47 (m, 2 H, Ph), 7.41–7.17 (m, 17 H, Ph),
obtained by recrystallization from CH2Cl2 at −20 °C. Yield 7.14–6.93 (m, 14 H, Ph + 1 H, C6H3), 6.90–6.77 (m, 2 H, Ph + 3
3
3
85 mg (58 μmol, 35%), m. p. 156–157 °C. – 1H NMR (CD2Cl2): δ H, C6H3), 6.55 (d, JHH = 7.8 Hz, 1 H, C6H3), 6.37 (t, JHH
=
3
= 7.38 (s br, 1 H, OH), 7.37–7.31 (m, 8 H, Ph), 7.28–7.21 (m, 16 7.6 Hz, 1 H, C6H3), 6.36 (t, JHH = 7.6 Hz, 1 H, C6H3), 6.25 (t,
H, Ph), 7.20–7.13 (m, 16 H, Ph), 6.73 (d, JHH = 7.6 Hz, 4 H, 3JHH = 7.6 Hz, 1 H, C6H3), 6.22 (t, JHH = 7.6 Hz, 1 H, C6H3),
3
3
C6H3), 6.29 (t, 3JHH = 7.6 Hz, 4 H, C6H3), 6.18 (d, 3JHH = 7.3 Hz, 5.97 (d, JHH = 7.4 Hz, 1 H, C6H3), 5.82 (d, JHH = 7.2 Hz, 1 H,
3
3
4 H, C6H3), 3.44–3.39 (m, 8 H, CH2). – 31P{1H} NMR (CD2Cl2): δ C6H3), 5.70 (d, 3JHH = 7.2 Hz, 1 H, C6H3), 5.64 (d, 3JHH = 7.4 Hz,
= 61.3 (s). – 27Al NMR (CD2Cl2): δ = 77.0 (s). – 31P{1H} CP-MAS 1 H, C6H3), 4.30 (dd, JHH = 13.0 Hz, JPH = 6.8 Hz, 1 H, CH2),
2
2
NMR: δ = 53.0, 49.0; JPP = 31 Hz (from 2D J-resolved). – 27Al 3.95 (dd, 2JPH = 5.2 Hz, 2JHH = 12.0 Hz, 1 H, CH2), 3.07 (t, 2JPH
=
2
{1H} CP-MAS NMR: δ = 5.0. – (+)-ESI-MS: m/z = 1465.13 [M + 2JHH = 12.3 Hz, 1 H, CH2), 3.03 (t, JPH = JHH = 13.3 Hz, 1 H,
2
2
H]+. The presence of varying amounts of residual solvent in CH2), 2.56 (dd, JPH = 16.4 Hz, JHH = 13.2 Hz, 1 H, CH2), 2.48
2
2
2
2
2
bulk samples prevented to obtain a satisfactory elemental (dd, JPH = 18.0 Hz, JPH = 14.3 Hz, 1 H, CH2), 2.50 (t, JPH
=
analysis.
18.0 Hz, 1 H, CH2), 2.25 (dd, 2JPH = 7.0 Hz, 2JHH = 14.0 Hz, 1 H,
Complex 7. InCl3 (24 mg, 0.11 mmol) and Et3N (0.05 ml, CH2), 1.83 (dd, JHH = 6.0 Hz, JPH = 12.6 Hz, 1 H, CH2). – 31P
3
2
0.36 mmol) were added to a solution of 4 (80 mg, 0.11 mmol) {1H} NMR (161.9 MHz, 253 K, CD2Cl2): δ = 71.3 (d, JPP
=
2
2
2
in 12 ml of CH2Cl2. The mixture was stirred overnight and 47 Hz), 68.5 (d, JPP = 47 Hz), 54.2 (d, JPP = 32 Hz), 48.2 (d,
then evaporated to dryness. Recrystallization of the red brown 2JPP = 32 Hz). Anal. for C76H61InO8P2Pd2·4 MeOH (1682.04):
residue from acetone produced 45 mg (51 μmol, 46%) of 7, calcd C 57.12, H 4.61%, found C 56.91, H 4.47%. – (+)-ESI-MS
m. p. >250 °C (dec). – H NMR (acetone-d6, 303 K): 7.6–7.0 (br, (MeOH): m/z = 832.98 [In(L)2Pd]+, 1555.06 [M + H]+, 1577.04
1
40 H, Ph), 7.0–5.6 (br, 12 H, C6H3, 3.3–3.1 (br, 8 H, CH2). – 31P [M + Na]+.
{1H} NMR (acetone-d6, 303 K): δ = 71.2 (br d, 2JPP = 42 Hz), 67.9
2
X-ray diffraction studies
(br d, JPP = 42 Hz), 55.2 (br), 50.8 (br). – 1H NMR (CD2Cl2,
203 K): 7.32 (br m, 12 H, Ph), 7.22–7.12 (br m, 12 H, Ph), 6.88 Diffraction data were collected at 123(2) K using a Bruker-
3
3
(br m, 16 H, Ph), 6.58 (d, JHH = 8 Hz, 4 H, C6H3), 6.29 (t, JHH Nonius Kappa-CCD diffractometer with Mo-Kα radiation (λ =
= 8 Hz, 4 H, C6H3), 5.62 (d, 3JHH = 8 Hz, 4 H, C6H3), 3.79 (br d, 0.71073 Å). A combination of ω and Φ scans was carried out to
2JHH = 12.4 Hz, 4 H, CH2), 2.98 (br m, 4 H, CH2). – 31P{1H} obtain at least
a
unique data set. Direct methods
NMR (CD2Cl2, 203 K): δ = 74.8 (s). – (+)-ESI-MS: m/z = 832.98 (SHELXS-9714) were used for structure solution (dual space
[In(L)2Pd]+. Anal. for C76H60O8P4Pd2In2Cl2·4 acetone methods for 5, using SHELXD14). Refinement was carried out
(1970.91): calcd C 53.63, H 4.30%; found C 54.06, H 4.02%.
using SHELXL-9714 (full-matrix least-squares on F2). Non-
–
Complex 8. [In(acac)3] (103 mg, 0.25 mmol) were added to a hydrogen atoms were refined anisotropically and hydrogen
suspension of 4 (180 mg, 0.25 mmol) in 20 ml of MeOH. The atoms with a riding model (H(O) free). Semi-empirical absorp-
mixture was heated to 50 °C and stirred for 3 h until all solids tion corrections were applied.
had dissolved. The red solution formed was then concentrated
One chloroform molecule in 3a is disordered. Three
to approx. one third of its original volume, and allowed to cool acetone molecules in 7 are disordered, and some phenyl
to r.t. Et2O was added until the clear solution became turbid. groups in 6 and 7 show high displacement parameters, indi-
The resulting mixture was stored for 48 h at 4 °C to produce a cating a possible disorder which could not be resolved due to
8918 | Dalton Trans., 2014, 43, 8911–8920
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Table 1 Summary of crystallographic data for 3a, 3b, 5–7
3b
5
6a
7
Formula
CCDC
Mol. Wt.
Crystal system
C41H37ClGaNO5P2Pd C43H37AlO6P2Pd – 2 MeOH C76H61AlO8P4Pd2 – 2 CH2Cl2 C76H60Cl2In2O8P4Pd2 – 4 acetone
990577
990578
909.13
990579
990580
1970.77
Triclinic
897.23
Monoclinic
P21/n
1635.76
Monoclinic
C2
Triclinic
P1
ˉ
ˉ
P 1
Space group
a/Å
b/Å
c/Å
α/°
12.644(1)
16.798(2)
17.833(2)
12.393(2)
13.325(2)
13.465(2)
96.11(1)
112.09(1)
97.42(1)
2013.8(5)
2
22.608(2)
13.568(1)
13.497(1)
11.116(1)
19.005(2)
20.826(2)
100.09(1)
103.19(1)
99.94(1)
4112.3(7)
2
1.592
β/°
104.01(1)
116.15(1)
γ/°
V/Å3
3675.0(7)
4
1.622
1.430
1816
3716.4(5)
2
1.462
0.780
1660
Z
Calcd density (g cm−3
)
1.499
0.617
936
Abs. coeff. (mm−1
F(000)
)
1.188
1984
Crystal size (mm)
θ Range (°)
R(int)
Reflections measured
Unique reflections
0.20 × 0.12 × 0.08
2.94 to 25.02
0.087
26 201
6417
0.16 × 0.08 × 0.04
2.98 to 25.00
0.124
28 365
7058
0.12 × 0.06 × 0.02
3.00 to 25.02
—
6509
6509
0.18 × 0.09 × 0.03
2.93 to 25.03
0.072
43 026
14 483
Unique reflections [I > 2σ(I)] 4380
4608
99.6%
4126
99.8%
8938
99.6%
Completeness to θ
99.2%
Absorption correction
equivalents
Max. and min. transmission 0.8620 and 0.7504
Semi-empirical from Semi-empirical from
Semi-empirical from
equivalents
0.9801 and 0.7985
6509/1/412
0.95
0.070
0.148
Semi-empirical from
equivalents
0.9703 and 0.8188
14 483/224/981
1.02
0.057
0.124
equivalents
0.9801 and 0.6572
7058/37/520
1.04
Data/restraints/parameters
Goodness of fit on F 2
R1 (I > 2σ(I))
6417/0/471
1.07
0.053
0.076
wR(F 2) (all data)
0.088
0.586 and −0.550
0.180
1.125 and −0.724
Largest diff. (e Å−3
)
0.759 and −0.661
−0.10(5)
1.637 and −1.016
Flack parameter
a Use of SQUEEZE (see ESI and cif-files).
limited data quality. Four highly disordered solvent molecules
(CH2Cl2) in the crystal structure of 6 were removed using the
SQUEEZE routine in the program Platon.15 Details of the
crystal structure determinations are listed in Table 1 and in
the cif-files (see ESI†). Crystallographic data (excluding struc-
ture factors) have been deposited with the Cambridge Crystal-
lographic Data Centre as supplementary publication no.
CCDC-990577 (3b), CCDC-990578 (5), CCDC-990579 (6), and
CCDC-990580 (7).
2011, 9, 1704; (e) S. Lühr, J. Holz and A. Börner, Chem-
CatChem, 2011, 3, 1708.
2 Selected reviews: (a) W. Malisch, B. Klupfel, D. Schumacher
and M. Nieger, J. Organomet. Chem., 2002, 661, 95;
(b) T. V. RajanBabu, Y.-Y. Yan and S. Shin, Curr. Org. Chem.,
2003, 7, 1759; (c) A. Börner, in Aqueous-Phase Organometal-
lic Catalysis, ed. B. Cornils and W. A. Herrmann, Wiley-
VCH, Weinheim, 2nd edn, 2004, p. 187; (d) J. Heinicke,
N. Peulecke, M. Köhler, M. He and W. Keim, J. Organomet.
Chem., 2005, 690, 2449; (e) M.-N. Gensow and A. Börner,
Multiphase Homogeneous Catalysis, 2005, vol. 1, p. 89;
(f) V. Andrushko and A. Börner, Phosphorus Ligands in
Asymmetric Catalysis, 2008, vol. 2, p. 633.
Acknowledgements
3 (a) X. Sun, D. W. Johnson, D. L. Caulder, R. Powers,
K. N. Raymond and E. H. Wong, Angew. Chem., Int. Ed.,
1999, 38, 1303; (b) X. Sun, D. W. Johnson, D. L. Caulder,
R. E. Powers, K. N. Raymond and E. H. Wong, J. Am. Chem.
Soc., 2001, 123, 2752.
We thank the Deutsche Forschungsgemeinschaft (DFG, grant
Gu 415/12-1) for financial support and the Academy of Finland
for a Research Fellowship. We further thank K. Wohlbold and
J. Trinkner for measurement of mass spectra.
4 G. Bauer, D. Förster, M. Nieger and D. Gudat, Z. Anorg. Allg.
Chem., 2014, 640, 325.
5 S. Chikkali, D. Gudat, F. Lissner, M. Niemeyer, Th. Schleid
and M. Nieger, Chem. – Eur. J., 2009, 15, 482.
6 S. H. Chikkali, M. Nieger and D. Gudat, New J. Chem., 2010,
34, 1348.
7 (a) S. H. Chikkali, D. Gudat, F. Lissner, M. Nieger and
T. Schleid, Dalton Trans., 2007, 35, 3906; (b) G. Bauer,
Notes and references
1 Selected reviews: (a) L.-C. Liang, Coord. Chem. Rev., 2006,
250, 1152; (b) F. Boeda, C. Beneyton and C. Crevisy, Mini-
Rev. Org. Chem., 2008, 5, 96; (c) J. I. van der Vlugt and
J. N. H. Reek, Angew. Chem., Int. Ed., 2009, 48, 8832;
(d) J. Wassenaar and J. N. H. Reek, Org. Biomol. Chem.,
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 8911–8920 | 8919

View Article Online
Paper
Dalton Transactions
Z. Benkö, M. Nieger, J. Nuss and D. Gudat, Chem. – Eur. J.,
2010, 16, 12091.
Polyhedron, 2004, 23, 2393; (d) M. H. Chisholm, J. Gallucci,
D. Navarro-Llobet and H. Zhen, Polyhedron, 2003, 22, 557;
(e) W. Ziemkowska, Main Group Met. Chem., 2000, 23, 337;
(f) S. N. Mhatre, S. B. Karweer, P. Pradhan, R. K. Iyer and
P. N. Moorthy, Dalton Trans., 1994, 3711; (g) T. Kiss,
K. Atkari, M. Jezowska-Bojczuk and P. Decock, J. Coord.
Chem., 1993, 29, 81; (h) T. B. Karpishin, T. D. P. Stack and
K. N. Raymond, J. Am. Chem. Soc., 1993, 115, 182;
(i) E. W. Meijer, Polyhedron, 1987, 6, 525.
8 G. Bauer, PhD Thesis, Stuttgart, 2013.
9 R. D. Shannon, Acta Crystallogr., Sect. A: Cryst. Phys., Diffr.,
Theor. Gen. Cryst., 1976, 32, 751.
10 E. L. Muetterties and J. L. Guggenberger, J. Am. Chem. Soc.,
1974, 96, 1748.
11 S. Alvarez, Dalton Trans., 2013, 42, 8617.
12 E. J. Bierschenk, N. R. Wilk and T. P. Hanusa, Inorg. Chem.,
2011, 50, 12126.
14 G. M. Sheldrick, Acta Crystallogr., Sect. A: Fundam. Crystal-
logr., 2008, 64, 112.
13 Selected references: (a) A. V. Piskunov, A. V. Maleeva,
G. K. Fukin, E. V. Baranov and O. V. Kuznetsova, 15 (a) A. L. Spek, Platon, A Multipurpose Crystallographic Tool,
Russ. J. Coord. Chem., 2010, 36, 161; (b) V. Kriegisch and
C. Lambert, Eur. J. Inorg. Chem., 2005, 4509; (c) S. Giroux,
S. Aury, P. Rubini, S. Parant, J.-R. Desmurs and M. Dury,
Utrecht University, Utrecht, The Netherlands, 2004;
(b) A. L. Spek, Acta Crystallogr., Sect. D: Biol. Crystallogr.,
2009, 65, 148.
8920 | Dalton Trans., 2014, 43, 8911–8920
This journal is © The Royal Society of Chemistry 2014