Download Metalloid Al- and Ga-clusters: a novel dimension in organometallic

View Article Online / Journal Homepage / Table of Contents for this issue
FRONTIER
Dalton
H. Schnöckel
University of Karlsruhe, Institute for Inorganic Chemistry, Engesserstrastr. 15, Karlsruhe,
D-76128, Germany. E-mail: [email protected]
www.rsc.org/dalton
Metalloid Al- and Ga-clusters: a novel dimension in organometallic
chemistry linking the molecular and the solid-state areas?†
Received 17th May 2005, Accepted 9th August 2005
First published as an Advance Article on the web 31st August 2005
Downloaded by University of Washington on 27/03/2013 22:34:44.
Published on 31 August 2005 on http://pubs.rsc.org | doi:10.1039/B507002N
Formation and fragmentation of metal–metal bonds on the way between stable metal compounds in which the metal
atoms are oxidised (e.g. isolated species in solution or metal salts in bulk) and the bulk metal are the fundamental
steps to understand this process in which formation and chemical behaviour of metalloid Al and Ga clusters as
intermediates are essential. Many examples of metalloid Al and Ga clusters show that their formation reflects a high
degree of complexity like that of the simple seeming formation of the bulk metal itself: starting from metastable Al(I)
and Ga(I) solutions containing small molecular entities, metalloid clusters grow during many self-organization steps
including aggregation as well as irreversible redox cascades. This novel class of clusters seems to open a new
dimension in chemistry between the molecular and the solid-state area, because, for the first time, it is shown that
under well selected conditions definite molecular species, i.e. metalloid clusters, grow via the formation of additional
metal–metal bonds and that the solid metal represents the final step.
Introduction
Most elements of the periodic system are metals. Their chemistry
and especially their formation and dissolution belong to the
oldest technical chemical processes. In general, however, only
the bulk metals themselves, on the one hand, and their stable
compounds (e.g. salts, oxides, sulfides, in solution or in bulk), on
the other, are well known. It seems strange that intermediates in
† Electronic supplementary information (ESI) available: Fig. 1 : The
Aln core of the metalloid clusters [Al7 R6 ]− , [Al12 R8 ]− , [Al14 R6 I6 ]2− (R =
N(SiMe3 )2 ). Fig. 2 : The Aln core of the metalloid clusters [Al69 R18 ]3− and
[Al77 R20 ]2− (R = N(SiMe2 )2 ) and the different arrangements of the central
Al13 units. Fig. 3 : The Al22 core of the halides Al22 X20 ·12D (X = Br, Cl; D =
THF). Fig. 4 : Section of the normal-pressure (a, b, c, d) and the highpressure modifications (GaII, GaIII) of solid Ga. The fcc high-pressure
modification GaIV is not shown. Fig. 5 : Gan core of the clusters Ga18 R8
and Ga22 R8 (R = SitBu3 ). See http://dx.doi.org/10.1039/b507002n
DOI: 10.1039/b507002n
Hansgeorg Schnöckel studied chemistry at the University
of Münster, where he gained his PhD under H. J. Becher in
1970. Subsequently he started spectroscopic matrix investigations
of reactive high-temperature molecules which were the basis for
later synthetic work. In 1987 he became professor and in 1989 he
changed to the University of Munich. Since 1993 he has held the
chair for analytical chemistry at the University of Karlsruhe.
H. Schnöckel
This journal is
©
this process of the formation and breaking of metal–metal bonds
(MM) are mostly unknown. This fundamental process and
molecular intermediates exhibiting MM bonding are therefore
central to this contribution. These molecular intermediates are
mostly addressed as metal atom clusters.1 However, since this
definition of F. A. Cotton is not restricted to species containing
mainly MM bonding, we have introduced the term metalloid
clusters.2 Such clusters contain more MM contacts than ML
(metal–ligand) bonds and mostly show similarities with respect
to the topology to the arrangements of atoms in the elements
themselves. Though most investigations on metalloid cluster
species have been, and are still, performed in the field of precious
metals (e.g. Au, Pd)3 because of their relatively straightforward
synthesis, their stability and their inertness even in air and
sometimes in water, it seems to be extremely difficult to obtain
crystalline materials in order to characterize them via crystal
structure analysis.4 The largest cluster of this type is a Pd145 cluster for which—although only two crystals have been obtained so
far—the structure could be solved by L. F. Dahl et al.:5 There is a
centre of 55 naked Pd atoms surrounded by 90 ligand-bearing Pd
atoms. The many synthetic results for these clusters of precious
metals and the few detailed structure determinations on the one
hand, and the failure of analysis for clusters of base metals until
1997 on the other, may give an idea of the great surprise when
8 years ago a metalloid cluster was presented containing 77 Al
atoms, of which 57 were naked and only 20 were ligand-bearing
(Fig. 1).6 However, this result was assumed to be a singularity
or a curiosity.7 We will show here that an exciting story has been
developed from this starting point and one which will also have
impact on the development of chemistry in general.
Though the organometallic chemistry of aluminium became
technically important in the second half of the last century—e.g.
through the work of Ziegler8 —the chemistry of the group 13
elements was assumed in general to be completely discovered
by the end of the last century. The exception seemed to be the
chemistry of boron. In retrospect, the “curiosities” in boron
chemistry have been a stroke of luck for the whole of chemistry,
for these results were the basis for novel ideas of bonding which
have influenced many areas of chemistry.9
In order to develop a new chemistry, possibly similar to that
of boron, for the heavier elements Al and Ga, spectroscopic
and synthetic work was started in the 1980s. The spectroscopic
work concentrated on Al and Ga monohalides (AlX, GaX)
The Royal Society of Chemistry 2005
Dalton Trans., 2005, 3131–3136
3131
Downloaded by University of Washington on 27/03/2013 22:34:44.
Published on 31 August 2005 on http://pubs.rsc.org | doi:10.1039/B507002N
View Article Online
Fig. 1 The arrangement of 77 Al atoms in the metalloid cluster anion
[Al77 {N(SiMe3 )2 }20 ]2− . The 20 ligand-bearing Al atoms are blue and the
44 Al atoms of the second “shell” are metallic-grey. The centre of the
cluster exhibits a single Al atom surrounded by 12 Al atoms in a distorted
cuboctahedral/icosahedral arrangement.
generated at high temperatures and trapped in solid Ar and
then allowed to aggregate to form e.g. dimers or to react with
other molecules such as O2 or HCl.10 In the synthetic area the
beginning of a new chemistry came with the detection of the first
molecular compound containing a direct Al–Al bond via the
careful reduction of R2 AlX species (R = organic ligand such as
[HC(SiMe3 )2 ].11 From the matrix work on AlX and GaX species,
a synthetic upscaling led to metastable AlX and GaX solutions
and through substitution reactions to the first Al(I) organic
species AlCp*.12 Simultaneously, synthetic routes via reduction
of RAlX2 or R2 AlX compounds have been developed.13 Both
strategies have been successfully applied in the last decade,
with the technique of trapping AlX or GaX species and the
reduction route resulting in Aln Rm and Gan Rm species (n ≈ m),
as described in several reviews. Most contributions involving
the latter technique came from a few groups worldwide, e.g.
those of W. Uhl,13 G. H. Robinson,14 G. Linti,15 P. P. Power,16
P. Jutzi,17 H. W. Roesky,18 R. A. Fischer,19 and N. Wiberg.20 This
chemistry of Al(I) and Al(II) species and their Ga analogues
should be called “modern classical chemistry”; reference has
been made elsewhere to as a renaissance of Al chemistry,18 since
spectacular analogies to the molecular chemistry of boron, e.g.
in the case of the icosahedral Al12 R12 2− , cannot be overlooked.21
From this platform of Al(I) species, e.g. [AlCp*]4 , as the first
organo-Al(I) compound containing an Al4 tetrahedral core,22
a new dimension of chemistry is now visible, as convincingly
illustrated by the recently published Al50 Cp*12 compound23
(Fig. 2). This largest structurally characterised homoleptic
organometallic compound containing 50 Al, 120 C and 180 H
atoms may be a milestone in the metal-rich chemistry with
respect to fundamental questions and possibly also applications.
Therefore, this example may be an ideal starting point to
introduce and explain the fundamental aspects of metalloid Al
and Ga clusters through the medium of the following sections:
1. Synthetic concepts
2. Variety of structural motifs within the crystalline compounds
3. Properties of the crystalline compounds
4. Bonding of metalloid clusters via experimental and theoretical investigations
1. Synthetic concepts
The synthetic concept for metalloid clusters is based on a technically sophisticated preparation of metastable solutions formed
via trapping of the high-temperature subhalide molecules AlX
and GaX and a solvent mixture (mostly toluene and a donor
component such as Et2 O), for which experimental details have
been described before.12,20b,24,25 If these solutions are heated
from −80 ◦ C to about +80 ◦ C, disproportionation to the bulk
3132
Dalton Trans., 2005, 3131–3136
Fig. 2 The arrangement of 50 Al atoms and 12 C5 (CH3 )5 moieties of
the Al50 Cp*12 cluster. The 12 ligand-bearing Al atoms are blue; 30 of the
remaining 38 Al atoms are metallic-grey, 8 Al atoms in the centre are
pink. The topology of the 60 CH3 -groups is similar to that of the 60 C
atoms of fullerene.
metal (Al/Ga) and the trihalide is observed: e.g. 3 AlCl →
2 Almetal + AlCl3 . During this process, many steps of aggregation
and elimination (e.g. redox chemistry) are necessary, so that
the overall change exhibits a very high degree of complexity.
Nevertheless, the process is fast and therefore intermediates
cannot usually be trapped. Such trapping becomes possible
generally only if bulky substituents R− (R− substitutes e.g. Cl− )
are added during the disproportionation. Fortunately, some
of these substituted species can be isolated and structurally
characterised as snapshots on the way to the bulk metal.
By varying the reaction conditions (e.g. temperature, solvent,
amount of the donor compound in the solvent and the amount
of the substitution reagent, e.g. LiR), the size of the metalloid
clusters, i.e. the degree of proceeding to the formation of the
bulk metal, can be influenced.
To sum up, the reproducible detection of definite intermediate
clusters demonstrates that this process leading to the formation
of metals follows definite rules which are completely unknown
so far, and for which even the few snapshots illustrate the high
complexity.
2. Variety of structural motifs within the crystalline compounds
The variety of metal atom topology in the crystalline compound
formed by a metalloid cluster often reflects the topology of the
metal itself. This principle is shown for [N(SiMe3 )2 ]− ligandstabilized Al7 , Al12 , Al14 , Al69 and Al77 cluster compounds20b,24,25
in which distorted arrangements of the Al atoms similar to those
in the close-packed structure of the metal are visible (ESI,†
Fig. 1 ). Nevertheless, each of these clusters exhibits topologies of
the metal atoms that are singular in cluster chemistry.26 There are
small but significant differences even where the clusters are quite
similar, e.g. in the largest clusters [Al69 R18 ]3− and [Al77 R20 ]2− ;27
the central Al atoms have a decahedral shell (Al69 ) and a distorted
cuboctahedral coordination (Al77 ) of 12 Al atoms surrounding
the unique central Al atom (ESI,† Fig. 2 ). This means that small
changes in the cluster shell are reflected also in the cluster centre.
Consequently it must be assumed that changes on any metal
surface e.g. by oxidation will have topological and/or electrical
influence on the metal atoms in the bulk below the surface. This
influence may be of special importance for the properties of any
nanoscaled metallic entities and devices.
Besides clusters exhibiting the atomic arrangement of the
known structure of the bulk metal, it is possible in principle
also to form metastable, hitherto unknown modifications of
Downloaded by University of Washington on 27/03/2013 22:34:44.
Published on 31 August 2005 on http://pubs.rsc.org | doi:10.1039/B507002N
View Article Online
Fig. 3 The [Ga51 (PtBu2 )14 Br6 ]3− cluster anion with its highly symmetrical arrangement of the surrounding ligands (6 Br atoms (green) and
6 + 2 terminally bonded and 6 bridging PtBu2 entities (purple)). The
centred Ga13 unit with a volume similar to that of a Ga13 section of the
high pressure fcc-modification of the metal is focussed.
the metals under these mild reaction conditions, as revealed
by the existence of cluster analogues. For Al a non-metallic
modification analogous to that of a- or b-boron is conceivable
because we were able to isolate a potential precursor species
Al22 X20 in which naked Al12 icosahedra are directly bonded to
10 AlX2 moieties [Al12 (AlX2 )10 ]28,29 (ESI,† Fig. 3 ). On the basis
of quantum chemical calculations, it seems possible in principle
to connect the Al12 icosahedra to form a hypothetical metastable
b-Al. This is a challenge for future work to prepare metastable
modifications of elements, and not only of Al, as new and
unusual materials. Especially for gallium with its already known
seven modifications (ESI,† Fig. 4 ), there are realistic chances
of producing new modifications like that of a-boron deduced
from quantum chemical calculations and recent experimental
observations.30 Obviously the metal atoms of Ga are very
“flexible” with respect to different topologies, and consequently
this flexibility allowing different connectivities is also reflected
in the variety of metalloid Ga clusters prepared as nanoscaled
element modifications: thus, [Ga10 (GaR)8 ] and [Ga14 (GaR)8 ]
(R = supersilyl = SitBu3 ) are convincing examples that within
a similar “box-volume” of 8 Ga ligand atoms arranged at the
vertices of a cube there are 10 Ga atoms arranged in the topology
of the normal pressure modification of b-Ga or 14 Ga atoms
arranged as in the high-pressure modification Ga(III)31 (ESI,†
Fig. 5 ). In order to investigate the influence of ligands on the
arrangement of metal atoms in the cluster core, the [SitBu3 ]−
ligand was changed for the bridging ligand [P(tBu)2 ]− .
This substitution leads to a compression of the central core
of the naked Gan atoms and therefore, as in the Ga51 cluster, the
central cuboctahedral Ga13 unit is very close to the unit of 13
Ga atoms in the high-pressure modification Ga(IV)32 (Fig. 3).
Furthermore, the easy variation of connectivities of the Ga
atoms could be demonstrated recently by the preparation of
four metalloid Ga clusters with four different arrangements of
22 Ga atoms (Fig. 4).33
The largest structurally characterized metalloid cluster containing 84 Ga atoms is protected by the same kind and number
of ligands [N(SiMe3 )2 ] as the Al77 cluster compound34 (Fig. 5).
This Ga84 cluster exhibits many singularities, e.g. its high pseudosymmetry (D5h ) which may have been the basis for addressing
Fig. 4 Four different arrangements of 22 Ga atoms in the cluster species
(a)–(d): (a) Ga22 R8 (R = Si(SiMe3 )3 , Ge(SiMe3 )3 , SitBu3 ), the central Ga
atom has a coordination number of 13 (3, 6, 4). The directly bonded
ligand atoms are grey. (b) [Ga22 R10 ]2− (R = N(SiMe3 )2 , the N atoms are
red), the central Ga atom has a coordination number of 11. In contrast to
(a) and (b) there is no central Ga atom in the distorted icosahedral species
presented in (c) and (d): (c) [Ga22 {N(SiMe3 )2 }10 Br10 ]2− (projection along
the top and bottom Ga atoms), Br atoms (green), N atoms (red); (d) via
the bridging ligands PtBu2 (P atoms purple) this cluster Ga22 R12 has a
low average oxidation number for the Ga atoms, i.e. this cluster, more
so than (c) presents a snapshot on the way to one of the modifications
of bulk Ga.
it as a fragment of a quasi-crystal.35 Here we want to draw
attention only to the Ga2 unit in the centre of the cluster with
a short Ga–Ga distance not much longer than the so-called
Ga–Ga triple bond.36,37 This topology of a Ga2 unit (as in aGa) surrounded by a cage of about 32 other naked Ga atoms
is unique and noteworthy, especially with respect to possible
rotations of this moiety. The unusual topology of the Ga84 cluster
itself and especially the arrangement of the clusters in the crystal
(Fig. 5) are responsible for the remarkable properties of this
compound shortly to be discussed in the next section.
3. Properties of the crystalline compounds
In order to measure reproducible physical properties of a
metalloid cluster compound as an intermediate on the way to the
bulk metal, well-defined crystalline material is required. Since,
even for large clusters of precious metals, no suitable crystals
were available so far, no reproducible data exist. For Al and Ga
clusters these measurements are especially difficult since many of
them spontaneously ignite in air. Nevertheless, after three years
of hard work (involving measurements on crystalline materials,
e.g. solid-state Ga-NMR, EPR, SQUID and conductivity), we
have succeeded in showing that crystals of the Ga84 compound
with its special arrangement of the Ga84 entities (Fig. 5) exhibit
electrical conductivity and below 7 K even superconductivity.38,39
These unexpected and yet to be conclusively interpreted results
have been investigated intensively in the last years.40 One point
may be of fundamental importance: e.g. besides the Ga84 cluster
compound containing the [Ga84 R20 ]4− anions, we were able to
crystallize a compound containing the analogous [Ga84 R20 ]3−
species with a slightly different arrangement of the cluster
entities.41 This observation may be significant for the entire
cluster chemistry because cluster doping (e.g. a small amount
of [Ga84 R20 ]3− cluster may replace the [Ga84 R20 ]4− anions in
the lattice of the latter cluster compound or vice-versa), which
may occur during the formation of crystals depending on
the redox potential of the mother-liquor, may be responsible
for the resulting properties and especially the conductivity
Dalton Trans., 2005, 3131–3136
3133
Downloaded by University of Washington on 27/03/2013 22:34:44.
Published on 31 August 2005 on http://pubs.rsc.org | doi:10.1039/B507002N
View Article Online
3134
Fig. 5 The single [Ga84 R20 ]4− cluster (R = N(SiMe3 )2 , N atoms red) (left) with its 20 ligand-bearing Ga atoms (blue) and the arrangement of these
clusters in the crystal (right) via the “naked” top and bottom Ga atoms (dark-grey, left). The space between these Ga atoms (1.4 nm) is filled with
two toluene molecules (right).
mentioned above. In the crystalline state in particular, metalloid
clusters seem to exhibit unpredictable properties which may be
important ultimately for understanding fundamental processes,
e.g. conductivity or superconductivity,42 and consequently they
may also be suitable for special applications.
4. Bonding of metalloid clusters via experimental and theoretical
investigations
The description of bonding within the large variety of different topologies of metalloid Al and especially Ga clusters
seems to be a real problem. This is not unexpected since
the group 13 elements boron, aluminium and gallium, though
containing three valence electrons each, show in their elemental
forms that, depending on small changes of conditions (p, T),
different connectivities of the metal atoms are achieved in
the different modifications. Consequently, the metalloid cluster
species reflecting this variety cannot be understood via a single
electron-counting rule.43 In order to understand the bonding,
therefore, the electronic situation of every metalloid Al and Ga
cluster species has to be discussed thoroughly and individually
with the help of quantum chemical calculations. However, a
qualitative description of the topology can often be obtained
via comparison with modifications of the elements themselves
(cf . above). Gallium with its seven known modifications is a
nearly ideal case in order to make plausible this concept that
metalloid clusters are intermediates during the formation of the
element, i.e. that metalloid clusters are snapshots taken during
these highly complex processes of self-organisation. Such a
hypothetical description is visualized in Fig. 6 demonstrating via
a cartoon that starting from GaX species via disproportionation
different routes to the 7 modifications of gallium can be expected
to proceed via certain metalloid Ga clusters.
Besides topological arguments for the structure of the clusters,
recent mass spectrometric results obtained via FT-ICR-MS
provide an experimental basis for the bonding description of
these clusters. After the structurally characterised cluster anion
[Ga19 R6 ]− [R = C(SiMe2 )3 ]2 has been transferred to the gas phase
without decomposition by electrospray ionisation (ESI) (this
being the first time for a metalloid cluster),44 the weakest bonds
could be dissociated via collision experiments (Fig. 7). Thus, the
weakest bonds are those between the cuboctahedral Ga13 − core
and the six neutral GaR units which have been shown previously
to be stable in the gas phase. Since the stable Ga13 − anion finally
results (the analogous Al13 − has been called a superhalogenide
species recently because of its stable electronic configuration
Dalton Trans., 2005, 3131–3136
Fig. 6 This cartoon with Mt. Bromo, East Java, Indonesia, may
illustrate in a simple manner the hypothetical routes to the formation of
different modifications of bulk gallium via different metalloid clusters as
snapshots of these highly complex processes of self-organisation starting
with the high-temperature gallium monohalide molecules GaX.
(jellium model) and high electron affinity of the neutral species45 )
we have combined the high-vacuum world of physicists dealing
with naked metal atom clusters as intermediates between the
bulk metal and metal atoms and the world of chemists, which is
usually the solution from which the crystalline compounds are
grown (Fig. 8).
This description of metalloid Al and Ga clusters as naked
metal atom cluster cores which are oxidised on the outside to M+
or M2+ , and for which the bonds between the core and the MR or
MR2 entities are weak, is also evident in the structure of the cluster species presented in Fig. 9. On the basis of the structures of so
many metalloid clusters, a hypothetical mechanism for the dissolution of a bulk metal can be deduced. This is visualized in Fig. 9.
Ongoing experiments with naked Aln clusters by FT-ICR mass
spectrometry will confirm this mechanism in the gas phase for a
fundamental process in chemistry, i.e. the dissolution of metals
and, via retro-synthetic considerations, also the formation of a
metal from a salt-like solution.46
Downloaded by University of Washington on 27/03/2013 22:34:44.
Published on 31 August 2005 on http://pubs.rsc.org | doi:10.1039/B507002N
View Article Online
Fig. 9 (a) A simplified picture of the hypothetical stepwise oxidation
of a surface of bulk aluminium. The mechanism is based on some
structurally characterized metalloid clusters as intermediates, i.e. as
snapshots during this highly complex process. (b) The metalloid clusters
containing a shell of M+ atoms (blue) are [Al7 R6 ]− [R = N(SiMe3 )2 ],23,24
[Ga13 R6 ]− [R = C(SiMe3 )3 ]2 and Al50 Cp*12 .22 (c) The metalloid clusters
containing the oxidized M2+ species (orange) are: [Al5 Br8 ]− ,23 Al22 Cl20 27,28
and Ga24 Br22 .29
Conclusions and outlook
Fig. 7 The relation between the metalloid cluster anions [Ga13 (GaR)6 ]−
(R = C(SiMe3 )3 ) and the naked Ga13 − cluster. SORI-CAD FT-MS
experiments after electrospray (ESI) show that the six weakest bonds
between the Ga13 core and the GaR units can be fragmented, in a stepwise
fashion or removed in a single step via LDI (laser desorption ionisation).
Fig. 8 Visualization of the generation of two types of cluster containing mainly metal–metal bonds; left: metalloid clusters starting from
diatomic monohalide molecules (the oxidation states of the metal atoms
are given in parentheses); right: naked metal atom clusters starting from
the bulk metal. It is a challenge for future work to connect these two
“worlds” of physicists and chemists.
Formation and fragmentation of metal–metal bonds on the way
between stable metal compounds in which the metal atoms
are oxidised (e.g. isolated species in solution or metal salts in
bulk) and the bulk metal are the essential steps to understand
the formation and chemical behaviour of metalloid Al and Ga
clusters as intermediates during this process. These clusters seem
to open a new dimension in chemistry between the molecular
and the solid-state area, because, for the first time, definite
molecular species grow only via the formation of additional
metal–metal bonds. In solid-state chemistry, this process of
forming larger metal-containing entities (e.g. Zintl anions) has
been fruitfully expanded e.g. by Corbett, Fässler, Nesper, von
Schnering, and Simon.47 However, the molecular chemistry of
species containing mainly metal–metal bonds may now open
new territory since unusual properties may be expected for
these novel substances, as has been shown for the Ga84 cluster
compound. This metal–metal chemistry would be a natural
development of classical chemistry of former periods when
(a) many non-metal elements (e.g. C, N, P, Si) are connected
to molecular entities also by sophisticated methods, such as
those of modern organic chemistry, or (b) metal atoms and
non-metal atoms are combined, as in traditional inorganic
chemistry. However, since even the synthesis of e.g. C60 fullerene
species via a stepwise route does not seem to be possible so
far, stepwise synthetic procedures are even more unlikely to
succeed in the case of metalloid clusters. Therefore, the present
method of trapping these clusters starting from metastable Al(I)
or Ga(I) solutions on the way to the bulk metal seems to be
the only way known to date for forming molecular species
containing mainly metal–metal bonds. In our opinion, however,
this process of forming metalloid clusters, which might seem
simple at first glance, exhibits a high degree of complexity; i.e.
the self-organization steps within this process–aggregation steps
as in supermolecular chemistry as well as irreversible redox
cascades–may not be much simpler than in many reactions in
biology,48 but experimentally more difficult to detect because
all the intermediates in the case of Al and Ga clusters are
sensitive to attack by air and moisture. Furthermore, since most
of the elements of the Periodic Table are metals, and since
every metallic element may be formed via alternative routes, this
novel chemistry of metalloid clusters exhibits a greater diversity
than many biological options.48 In the next decades, therefore,
much work in theory, spectroscopy, physics, and especially via
synthetic chemistry is needed to characterise more intermediates
Dalton Trans., 2005, 3131–3136
3135
View Article Online
as snapshots during this highly facile process in order to gain a
deeper understanding of the fundamental mechanism of metal
formation and dissolution.
Acknowledgements
We thank the Deutsche Forschungsgemeinschaft and the Centre
for Functional Nanostructures (CFN) and the Fonds der
Chemischen Industrie for financial support. Special thanks to
P. Hauser, M. Kayas, and S. Schneider for essential help in
preparing this manuscript and the figures.
27
28
29
30
Downloaded by University of Washington on 27/03/2013 22:34:44.
Published on 31 August 2005 on http://pubs.rsc.org | doi:10.1039/B507002N
References
3136
1 F. A. Cotton, Q. Rev. Chem. Soc., 1966, 20, 397.
2 A. Schnepf, G. Stößer and H. Schnöckel, J. Am. Chem. Soc., 2000,
122, 9178.
3 (a) Cluster and Colloids, ed. G. Schmid, Wiley-VCH, Weinheim, 1994;
(b) Metal Clusters in Chemistry, ed. P. Braunstein, L. A. Oro and
P. R. Raithby, Wiley-VCH, Weinheim, 1999; (c) Nanoparticles, ed.
G. Schmid, Wiley-VCH Weinheim, 2004.
4 Although suitable crystals of the Au55 cluster compounds for an X-ray
structure determination are not available so far, a face-centred cubic
arrangement of the Au atoms with shorter average distances than
in the bulk metal could be inferred recently via EXAFS, XANES
and WAXS: R. E. Benfield, D. Grandjean, M. Kroell, R. Pugin,
T. Sawitowski and G. Schmid, J. Phys. Chem. B, 2001, 105, 1961.
5 N. T. Tran, D. R. Powell and L. F. Dahl, Angew. Chem., 2000, 112,
4287; N. T. Tran, D. R. Powell and L. F. Dahl, Angew. Chem., Int.
Ed., 2000, 39, 4121.
6 A. Ecker, E. Weckert and H. Schnöckel, Nature, 1997, 387, 379.
7 F. A. Cotton, in ref. 3(b), pp. 3–7.
8 K. Ziegler, Angew. Chem., 1964, 76, 545; see also http://
nobelprize.org/chemistry/laureates/1963/ziegler-lecture.pdf, 2005.
9 W. N. Lipscomb, Angew. Chem., 1977, 89, 685; see also http://
nobelprize.org/chemistry/laureates/1976/lipscomb-lecture.pdf, 2005.
10 J. Bahlo, H.-J. Himmel and H. Schnöckel, Angew. Chem., 2001, 113,
4820; J. Bahlo, H.-J. Himmel and H. Schnöckel, Angew. Chem., Int.
Ed., 2001, 40, 4696; H.-J. Himmel, J. Bahlo, M. Haussmann, F. Kurth,
G. Stößer and H. Schnöckel, Inorg. Chem., 2002, 41, 4952.
11 W. Uhl, Z. Naturforsch., Teil B, 1988, 43, 1113.
12 C. Dohmeier, D. Loos and H. Schnöckel, Angew. Chem., 1996, 108,
141; C. Dohmeier, D. Loos and H. Schnöckel, Angew. Chem., Int.
Ed. Engl., 1996, 35, 129.
13 W. Uhl, Angew. Chem., 1993, 105, 1449; W. Uhl, Angew. Chem., Int.
Ed. Engl., 1993, 32, 1386.
14 G. H. Robinson, Adv. Organomet. Chem., 2001, 47, 283.
15 G. Linti and H. Schnöckel, Coord. Chem. Rev., 2000, 285, 206.
16 P. P. Power, J. Organomet. Chem., 2004, 689, 3904.
17 P. Jutzi, Pure Appl. Chem., 2003, 75, 483.
18 H. W. Roesky, Inorganic Chemistry in Focus II, ed. G. Meyer,
D. Naumann and L. Wesemann, Wiley-VCH, Weinheim, 2005,
pp. 89–103; H. W. Roesky, Inorg. Chem., 2004, 43, 7284.
19 C. Gemel, T. Steinke, M. Cokoja, A. Kempter and R. A. Fischer,
Eur. J. Inorg. Chem., 2004, 4161.
20 (a) N. Wiberg and P. P. Power, Molecular Clusters of the Main
Group Elements, ed. M. Driess and H. Nöth, Wiley-VCH, Weinheim,
2004, pp. 188–208; (b) G. Linti, H. Schnöckel, W. Uhl, N. Wiberg,
Molecular Clusters of the Main Group Elements, ed. M. Driess and
H. Nöth, Wiley-VCH, Weinheim, 2004, pp. 126–168.
21 W. Hiller, K.-W. Klinkhammer, W. Uhl and J. Wagner, Angew. Chem.,
1991, 193, 182; W. Hiller, K.-W. Klinkhammer, W. Uhl and J. Wagner,
Angew. Chem., Int. Ed. Engl., 1991, 30, 179.
22 C. Dohmeier, C. Robl, M. Tacke and H. Schnöckel, Angew. Chem.,
1991, 103, 594; C. Dohmeier, C. Robl, M. Tacke and H. Schnöckel,
Angew. Chem., Int. Ed. Engl., 1991, 30, 564.
23 J. Vollet, J. R. Hartig and H. Schnöckel, Angew. Chem., 2004, 116,
3248; J. Vollet, J. R. Hartig and H. Schnöckel, Angew. Chem., Int.
Ed., 2004, 43, 3186.
24 A. Schnepf and H. Schnöckel, Angew. Chem., 2002, 114, 3682; A.
Schnepf and H. Schnöckel, Angew. Chem., Int. Ed., 2002, 41, 3532.
25 A. Schnepf and H. Schnöckel, Adv. Organomet. Chem., 2001, 235.
26 However, only for a neutral In12 R8 cluster has a similar arrangement
of the metal atoms as in the Al12 R8 − anion been observed: N. Wiberg,
Dalton Trans., 2005, 3131–3136
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
T. Blank, H. Nöth and W. Ponikwar, Angew. Chem., 1999, 111, 887;
N. Wiberg, T. Blank, H. Nöth and W. Ponikwar, Angew. Chem., Int.
Ed., 1999, 38, 839.
H. Köhnlein, A. Purath, C. Klemp, E. Baum, I. Krossing, G. Stößer
and H. Schnöckel, Inorg. Chem., 2001, 40, 4830.
Ch. Klemp, R. Köppe, E. Weckert and H. Schnöckel, Angew. Chem.,
1999, 111, 1851; Ch. Klemp, R. Köppe, E. Weckert and H. Schnöckel,
Angew. Chem., Int. Ed., 1999, 38, 1739.
C. Klemp, M. Bruns, J. Gauss, U. Häussermann, G. Stößer, L. van
Wüllen, M. Jansen and H. Schnöckel, J. Am. Chem. Soc., 2001, 123,
9099.
T. Duan, E. Baum, R. Burgert and H. Schnöckel, Angew. Chem.,
2004, 116, 3252; T. Duan, E. Baum, R. Burgert and H. Schnöckel,
Angew. Chem., Int. Ed., 2004, 43, 3190; H. von Schnering and
R. Nesper, Acta Chem. Scand., 1991, 45, 870; U. Häussermann,
S. Simiak, I. Abrikosov and S. Lidin, Chem. Eur. J., 1997, 3, 904;
X. G. Gong, G. Chiarotti, M. Parinello and E. Tasatti, Phys. Rev.
B, 1991, 43, 14277; U. Häussermann, S. I. Simiak, R. Ahuja and
B. Johansson, Phys. Rev. Lett., 2003, 90, 65701.
A. Donchev, A. Schnepf, G. Stößer, E. Baum, H. Schnöckel, T. Blank
and N. Wiberg, Chem. Eur. J., 2001, 7, 3348.
J. Steiner, G. Stößer and H. Schnöckel, Angew. Chem., 2004, 116,
305; J. Steiner, G. Stößer and H. Schnöckel, Angew. Chem., Int. Ed.,
2004, 43, 302.
References to all Ga22 species in Fig. 4 are mentioned by: J. Steiner, G.
Stößer and H. Schnöckel, Angew. Chem., 2004, 116, 6712; J. Steiner,
G. Stößer and H. Schnöckel, Angew. Chem., Int. Ed., 2004, 43, 6549.
A. Schnepf and H. Schnöckel, Angew. Chem., 2001, 113, 734; A.
Schnepf and H. Schnöckel, Angew. Chem., Int. Ed., 2001, 40, 712.
R. B. King and P. V. R. Schleger, Molecular Clusters of the Main
Group Elements, ed. M. Driess and H. Nöth, Wiley-VCH, Weinheim,
2004, pp. 1–33.
J. Su, X.-W. Li, R. C. Crittendon and G. H. Robinson, J. Am. Chem.
Soc., 1997, 119, 5471; Y. Xie, R. S. Grev, J. Ga, H. F. Schäfer III, P.
von R. Schleyer, J. Su, X.-W. Li and G. H. Robinson, J. Am. Chem.
Soc., 1998, 120, 3773; N. Takagi, M. W. Schmidt and S. Nagase,
Organometallics, 2001, 20, 1646; A. J. Bridgeman and L. R. Ireland,
Polyhedron, 2001, 2841.
R. Köppe and H. Schnöckel, Z. Anorg. Allg. Chem., 2000, 626, 1095;
H.-J. Himmel and H. Schnöckel, Chem. Eur. J., 2002, 8, 2397; H.-J.
Himmel and H. Schnöckel, Chem.–Eur. J., 2003, 9, 748.
J. Hagel, M. T. Kelemen, G. Fischer, B. Pilawa, J. Wosnitza, E.
Dormann, H. v. Löhneysen, A. Schnepf, H. Schnöckel, U. Neisel
and J. Beck, J. Low Temp. Phys., 2002, 129, 133.
O. N. Bakharev, N. Zelders, H. B. Brom, A. Schnepf, H. Schnöckel
and L. Jos de Jongh, Eur. Phys. J. D, 2003, 24, 101.
J. Frenzel, S. Gemming and G. Seifert, Phys. Rev. B, 2004, 70, 235404.
A. Schnepf, E. Herrling, B. Pilawa, E. Weckert, A. Meents, B. Jee
and H. Schnöckel, Inorg. Chem., 2003, 629, 2168.
The results of the first NMR measurements on metal atoms within a
crystalline metalloid cluster compound during the transition to the
superconducting state are about to be published (O. Bakharev, H. B.
Brom, L. J. de Jongh, A. Schnepf and H. Schnöckel, Nature, to be
submitted).
K. Wade, Adv. Inorg. Chem. Radiochem., 1976, 18, 1; R. W. Rudolph,
Acc. Chem. Res., 1976, 9, 446; R. E. Williams, Adv. Inorg. Chem.
Radiochem., 1976, 18, 67; D. M. Mingos, Nature, 1972, 336, 99.
K. Weiß and H. Schnöckel, Z. Anorg. Allg. Chem., 2003, 629, 1175.
D. E. Bergeron, P. J. Roach, A. W. Castleman, Jr., N. O. Jones and
S. N. Khanna, Science, 2005, 307, 231.
First results with Cl2 as an oxidising agent reveal different chemical
behaviours for different Aln clusters (R. Burgert, M. Olzmann,
K. Bowen and H. Schnöckel, Angew. Chem., submitted). The planned
oxidation with O2 promises to be especially complex because even Al
atoms under matrix conditions yield unexpected species (e.g. AlO6 :
G. Stößer and H. Schnöckel, Angew. Chem., 2005, 117, 4334; G.
Stößer and H. Schnöckel, Angew. Chem., Int. Ed., 2005, 44, 4261.
Of the large number of publications in this field only two are selected
which contain references to many other contributions: J. D. Corbett,
Angew. Chem., 2000, 112, 682; J. D. Corbett, Angew. Chem., Int. Ed.,
2000, 39, 692; T. F. Fässler and S. D. Hoffmann, Angew. Chem., 2004,
116, 6400.
J.-M. Lehn, Supramolecular Chemistry, Concepts and Perspectives,
Wiley-VCH, Weinheim, 1995, pp. 204–206. In this chapter (Chemistry and Biology, Creativity and Art), the author compares chemistry
and biology with respect to complexity and breadth or diversity.