close

Вход

Забыли?

вход по аккаунту

?

Fullercages without CarbonЧFulleranes Fullerenes Space-filler-enes.

код для вставкиСкачать
HIGHLIGHTS
features-here both the tertiary structure and the active site-in
the catalysis of completely different reactions. The future will
certainly bring further structural analyses : for example, if
samples crystallized in another buffer or containing the active
site in a reduced state were to be analyzed, more detailed
hypotheses of the reaction mechanism could be proposed. Possibly a larger complex of proteins A, B, and C could be crystallized, above all to study the immense influence of the regulatory
protein on the structure of the active site.
German version: Angew. Chem. 1994, 106, 889
[l] A. C. Rosenzweig, C. A. Frederick, S. J. Lippard, P. Nordlund, Narure 1993,
366, 537 - 543.
[2] M. A. Holmes, I. L. Trong, S. Turley, L. C. Sieker, R. E. Stenkamp, J. Mol. Biol.
1991,218. 583-593.
[3] P. Nordlund, H. Eklund, J. Mol. Biol. 1993, 232, 123-164.
[4] L. C. Sieker, S. Turley, B. C. Prickril, J. LeGall, Proteins: Struct. Funrf. Genet.
1988,3,184-186.
[5] N. Strdter, R. Frohlich, A. Schiemann, B. Krebs, M. Korner, H . Suerbaum, H.
Witzel, J. Mol. Biol. 1992, 224, 511-513.
161 P. H. Pritchard, C . E Costa, Environ. Sci. Techno/. 1991, 25, 372-379.
[7] J. Green, H. Dalton, J. Biol. Chem. 1989, 264, 17698-17703.
[8] J. G. DeWitt, J. G. Bentsen, A. C. Rosenzweig, B. Hedman, J. Green, S. Pilkington, G. C. Papaefthymiou, H. Dalton, K. 0. Hodgson, S. J. Lippard, J. Am.
Chem. SOC.1991, 113, 9219-9235.
[9] P. Nordlund, H. Dalton, H. Eklund, FEES Lerr. 1992, 307, 257-262.
Fullercages without Carbon-Fulleranes,
Fullerenes, Space-filler-enes?
Reinhard Nesper *
Is the fullerene structural principle, revealed in recent years
with the new modifications of carbon, of far more wide-ranging
significance than has previously been assumed? Can the special
geometry of fullercages form the basis for energetically favorable, or at the very least kinetically stable structures, with elements quite different from carbon? Such questions relate not
only to derivatives of carbon fullerenes with metallic or
organometallic species but also to what has previously been
known as incorporation, or the replacement of carbon in fullerenes with heteroatoms. A new study by Corbett and Sevov entitled “Carbon-Free Fullerenes” has now provided further food
for thought in a quite different direction.[’]
The two authors describe the ternary indides Na,,,In,,,Ni,
and Na,,In,,Z, (Z = Ni, Pd, Pt), in which Na,,In,,, In,,, and
In,, cages, each with fullerene-type structures, form characteristic structural elements. The fullercages are present, however, not
as isolated clusters, but as two-dimensional layers in a quasiclosest-packed arrangement, connected through In-In bonds.
The spaces between the layers are filled by the Na and Z atoms.
The fullercages in these indides are not empty. They contain
additional interesting polyhedra, whose corners are located beneath the surface of the larger external polyhedra, in accord
with the duality principle. These enclosed polyhedra can be
designated endohedral units, thereby leading to the formulations Na,,@[Na,,In,,],
Na,,@In,,, and Na,,@In,,. The inner polyhedra are still fairly large and are each filled with one
additional smaller polyhedron. In this way one of the most
important structural principles of the intermetallic phases is also
[*I Prof. Dr. R. Nesper
Laboratorium fur Anorganische Chemie
Eidgenossische Technische Hochschule, ETH Zentrum
CH-8092 Zurich (Switzerland)
Telefax: Int. code + (1)262-0718
Angen. Chem. Int. Ed. Engl. 1994, 33, No. 8
0 VCH
adhered to : a quasi-closest-packed arrangement in every sense.
For the larger cages the overall structure of the multilayered
polyhedra can, with respect to the central atom Z, be represented as In,,@Na,~@[Na,,~n,,l, Ni@In,,@Na,,@~n,,, and
Z@In,,@Na,,@In,,
(Fig. lc). The discrepancies with the
overall formulas are due to additional interstitial atoms.
Substitution of spheres for the large cages in Na,,In,,Z, gives
rise to an arrangement that corresponds very well to the nickel
arsenite structure. The expression “kierarchical structures” was
recently used by von Schnering et al. to describe arrangements
in which large units containing many atoms, in this case fullerspheres, take the place of smaller, topologically similar species,
here, simple atoms.[’’ This comparison points to the fact that
compounds that differ both in terms of stoichiometry and chemistry can apparently exist in similar arrangements, if these arrangements prove to be particularly favorable in a number of
respects. Thus, a kind of fractal behavior or scale-symmetry is
indicated in structural c h e m i ~ t r y . ~ ~ ]
We now come to the question as to whether fullercages are
among the more favored arrangements. I believe so. Upon close
scrutiny of the literature, it becomes clear that such polyhedra
have indeed been frequently detected in chemical compounds
since the 1950s; the first unambiguous example was P-rhombohedral boron.I4] However, intermetallic compounds may also
contain fullercages (Table 1) formed between the more electronegative components in each compound, in which homopolar bonds are preferred.
In P-rhombohedra1 boron a B,, fullercage is present as the
doubly endohedrally filled unit B,,@B,,@B,,
with connections in three dimensions (Fig. la).14]As all the atoms are of the
same type, all the bonds are covalent, even the shorter, stronger
bonds between the B,, polyhedra. It is therefore difficult to
consider the B,, polyhedra as distinct units in a chemical bond-
Verlagsgesellschaft mbH, 0-69451 Weinheim. 1994
0570-0833/94j0808-0843 $10.00+ .2S/O
843
HIGHLIGHTS
Table 1. Intermetallic compounds with endohedrally filled fullercages N,,, N,, ,
N,,, and N76.The maximum average number of electrons per atom of the electronegative component is calculated assuming a complete transfer of charge from the
electropositivecomponent (d electrons are not taken into account for Cu, Zn, and
Cd) .
Compound
Fullercage
Symmetry
VEC... [a] Ref.
Li,CuAl,
(ALWa9Mg,, [bl
Li,,Cu,Ga,,
Li,,Zn3,Ga6,
Na,,Cd,,Gas,
Na13K4Ga49.6
Na,,,InI9,Ni,
Na96bZz
(Cu+%o
(Zn.AO,,
@a),,
(Zn,Ga),,
(Cd,Ga),o
(Cia),,
(In),,
(Nab)60
(In),,
3.0
3.0
3.0
3.0
3.1
3.3
3.9
[lo1
[111
[121
1131
4.0
[11
LiMgAl,
(A1176
T, - a3m
T, - a3m
T, - 43m
D,, - 6m2
D,, - Jm
D,, - Jm
C,, - 2m
D,, - 3m
D,, - 6m2
C,,- 3m
4.5
t141
~ 5 1
ti1
~3~71
[a] VEC = valence electron number = average number ofelectrons per atom.[b] see
Fig. lb.
ing sense. Quantum mechanical investigations have not yet been
carried out for the fullerpolyhedra in this boron modificationan additional challenge for courageous theoreticians.
What is certain is that the average number of valence electrons-three per boron atom-does not suffice for the formation of a fullerene or hence a system with VEC = 4. At first
glance, the inwardly and outwardly orientated terminal bonds
present a confusing picture (see Fig. la). Upon closer examination, however, multicenter and two-electron/two-center bonds
can be assigned, in accordance with the Wade rules for electrondeficient polyhedra. All the five-membered rings, and hence all
the bonds in the five-membered rings, comprise constituent
parts of icosahedra, and thus Wade polyhedra. All other bonds,
in other words those joining the five-membered rings, bridge the
icosahedra and are thereby among the stronger two-electron/
two-center bonds. It may be coincidental that the shorter and
longer bonds in C,, and B,, are topologically equivalent.
On the basis of these considerations and by analogy with
boranes the B,, polyhedron may be considered an electron-deficient fullerane with free terminal valences rather than a
fullerene. Fulleranes constructed with single bonds and terminally linked on all sides have the same theoretical electron requirement as isolated fullerenes-VEC = 4. Thus, the boron
structure has too few electrons for either structural form. However, boron may compensate for this through the formation of
electron-deficient polyhedra, in this case icosahedra, which as a
consequence must be integrated into the fullerane surface.
Should /3-rhombohedra1boron therefore be perceived as a polymeric, electron-deficient variation of C,,? As is the case for the
boron fullercage, N,, polyhedra are present in many of the
compounds shown in Table 1. These polyhedra should correspond to the stronger covalent interactions, as in each case they
are composed of the more electronegative components A1 and
Ga and display relatively short edge-dimensions.
Although all the phases listed in Table 1 display intermetallic
character, the partial transfer of charge from the more strongly
electropositive alkali and alkaline earth elements should be taken into account. But the charge transfer should be less pronounced than that in Zintl phases. Thus, it is not clear how
many electrons are available for a given fullercage, in other
words, how many electronic states are occupied. For the novel
indides"] and LiMgAl,['] only simplified theoretical investigaa44
0 VCH Verlagsgesellschafl mbH, 0-69451Weinheim. 1994
tions have been conducted to date with the extended-Hiickel
method. The maximum average number of electrons possible
for the electronegativecomponents, VEC,,,, is attained when all
the valence electrons are assigned to these atoms (see Table 1).
Thus for N,, fulleranes (N = Al, Ga) VEC,,, = 3.0-3.3; for
the In,,, In,,, and (NaJn),, fulleranes in Na,,In,,Z,
and
Na,,,In,,,Ni,
VEC,,, x 4; and finally for Al,, in LiMgAl,,
the largest metal fullerane known to date, VEC,,, = 4.5.
The indium atoms in these phases can thus formally obtain up
to about four valence electrons, thereby permitting fullerane- or
fullerene-type structures. No inward homopolar interactions
are present, because no indium atoms are contained within the
middle inner polyhedron; however, there are many terminal
In-In bonds to the neighboring cages. Multiple bonding should
be absent here, as a consequence of the low tendency towards
such interactions among the heavy main-group elements, and
thus, in analogy to the boron polyhedron, a fullerane formulation is proposed.
The largest cages to date are the doubly endohedrally filled
Al,, fulleranes in LiMgAl,, Al,M,@M,o@A1,,, where M =
Li, Mg (Fig. Id).[','] Its size is compared to that of C,, in
Figure 1f. The Al,, cages form a quasi-cubic-closest-packed
arrangement, again in line with the hierarchical structure relationship. Since all the Al,, cages are linked through Al-A1
bonds, there should be no double bonding contribution, and the
large cage can be termed a fullerane. The results of simplified
band structure calculations indicate that the fullerane A1 atoms
have an electron requirement of V E C x 4 to achieve optimal
bonding.r51The electronic prerequisites for this are derived from
the stoichiometry (see Table 1). In this way, only a single formal
transfer of charge, (Li,Mg)+ Al-, is required per formula unit.
With this relatively simple classification of such phases by
means of a more or less estimated valence electron count, it is
clear that only a limited understanding of the chemical bonding
can be achieved; however, this may,still be useful for the synthesis of new compounds of similar type. In support of this preliminary qualitative hypothesis, we should emphasize that the
structures of many aluminides, gallides, and indides can be rationalized by simple, formal charge-transfers and the Wade
rules. Thus, the phases discussed here are not exceptional.[6- *I
The formation of metal fulleranes appears to begin with an
icosahedral cluster which forms the innermost shell of a threelayer cluster that is typically completed by a fullerane sphere.
For phases that obey this construction principle the term "icosogene" has been coined.[9]The shell-like construction, in which
electropositive and electronegative atoms are segregated into
different layers, is strongly reminiscent of tenside emulsions and
micellar systems. For the two metal fulleranes the formation of
the inner surfaces seems to be of considerable significance.
Viewed superficially, the phases NaIn and Na,,In,,Z, scarcely
differ from one another in stoichiometry and valence electron
count; however, the latter phase contains Z atoms as nucleation
centers.
This leads to the obvious question: when does growth from
nucleation centers of icosahedral or pentagonal symmetry result
in classical crystals and when in quasi-crystals? Interestingly,
quasi-crystalline LiMgAl, has also recently been reported[l6]
(see also ref.91). One wonders whether fullerenes can also serve
as nuclei for the growth of quasi-crystals.
0570-0833jsl94~0808-0844
8 10.00 i
.2S/O
Angew. Chem. Int. Ed. Engl. 1994,33, No. 8
HIGHLIGHTS
Fig. 1. Structures of endohedrally tilled fullercages: a) b-rhombohedra1 boron. In the fullercage B,, all the five-membered rings form part of B,, icosahedra (red and blue
rods; yellow rods for the corresponding fullerene five-memberedrings), of which only half are represented completely. The central icosahedron (white rods) is linked through
terminal bonds (thick magenta rods; only half are depicted for clarity) to all the icosahedra in the B,, surface. b) (AI,Zn),,Mg,, with (Al,Zn),, fullerane cage. The pentagon
dodecahedron shown in red corresponds to the layer of atoms between the fullercage and the inner icosahedron. The fullercage shows distinct deformations which displace
the atoms from the mean plane of the surface of the sphere. c) Multiply endohedrally filled Na,,In,,Z,: Z@In,,@Na,,@In,,.
Only the middle inner polyhedron Na,, is
visible as depicted. The outwardly bridging In-In bonds are indicated. d) LiMgAl, with a doubly endohedrally filled Al,, fullerane: A13M7@M40@A17,(M = Li, Mg). The
outwardly bridging AI-A1 bonds (transparent golden rods) are shown on the fullerene surface (yellow rods). e) A1a,O,,o polyhedra in Sr3,Bi22A14aO14,with AlO, tetrahedra
(violet). This polyhedron is also filled by two further polyhedron shells, but these are not shown here. Since the polyhedron surfaces of the fullercage are depicted as
transparent, the tilting of the A10, tetrihedra can be discerned clearly. f ) Comparison between C,, (left) and, Al,, (in (AI,Zn),,Mg,, ,sylid, blue-green) in Al,, (right, yellow
rods). The average sphere radii are 3.4, 6.8, and 7.4 A.
Fullercages have already been identified in the structures of
insulators, again in an arrangement in which distinct interactions lead to onion-skin-like constructions on the molecular
scale. The compound Sr36BizzA14EO141
is a complex oxide composed of a doubly endohedrally filled A14E0210
polyhedron with
12 five-membered rings and 32 six-membered rings["] (Fig. le).
Here, each 0 atom is situated not quite midway between two A1
atoms (A1,,0, 16). The three-way coordination of the A1 atoms
on the surface of the fullercage is increased to tetrahedral
(AlE40,,,) through a further, terminal oxygen atom, whereby
the linked A10, tetrahedra give rise to a complex fullercage. It
is therefore reasonable to anticipate more such structures in the
chemistry of heteropolyanions.
Multiply endohedrally filled units are likewise known for the
larger C, fullerenes, for example, the "onion-skin fullerenes"
constructed entirely from carbon atoms." 91 It is, however, also
possible to include here other molecular species that form van
der Waals complexes with C, fullerenes, in other words, hostguest compounds. Access to clathrates of this type is not very
likely with the Kratschmer process,"E1 as few potential guest
Angew. Chem. Inr. Ed. Engl. 1994, 33, No. 8
0 VCH
molecules have sufficiently high thermal stability. Rather, fullerenes must first be synthesized at significantly lower temperatures before such inclusion compounds would appear completely conceivable.
Finally, there remains the question as to whether other heavy
elements are able to form fullerenes or fulleranes. Since silicon
and germanium tolerate a greater range of bond angles, they
would be expected to form cages that would be geometrically
stabilized by endohedral filling. Whether a multiply bonded
structure would have any energetic chance over a singly bonded
fullerane is another open question. For the tricoordinate elements phosphorus, arsenic, antimony, and bismuth, endohedrally stabilized fulleranes seem entirely imaginable.
But why should we stop here? A great number of multicomponent compounds can be discussed in this manner and may
even have been prepared in a few laboratories. Binary compounds are suitable only with certain preconditions because of
the unavoidable frustration in the five-memberd rings. Perhaps
other large polyhedra with rings having an even number of
members and whose geometry and solid-state structure are also
Verlagsgesellschaft mbH. 0-69451 Weinheim,1994
0570-0833/94/0808-0845 St0.00-k ,2510
845
HIGHLIGHTS
known are more favorable-they
fullercages!
do not always have to form
German version: Angew. Chem. 1994,106,891
[l]S.C. Sevov, J. D. Corbett, Science 1993,262, 880.
[2]W. Carillo-Cabrera, N. Caroca-Canales, K. Peters, H. G. von Schnering, Z.
Anorg. Allg. Chem. 1993,619,1556.
[3]R. Nesper, Angew. Chem. 1991,103,805;Angew. Chem. I n t . Ed. Engl. 1991,30,
789.
[4]B. Kolakowski, Acta Phys. Pol. 1962,22, 439;J. L. Hoard, D. B. Sullenger,
C. H. L. Kennard, R. E. Hughes, J. Solid Sfate Chem. 1970,1,268;D. Geist,
R. Kloss, H. Follner, Acta Crystallogr. Sect. B 1970,26, 1800;J. Donohue,
Structures of the Elements, Krieger, Malaba, FL, USA, 1974,S.61.
[5] R. Nesper. unpublished results.
[6]C. Belin, M.Tillard-Charbonnel,Prog. Solid State Chem. 1993,22,59.
[IR.Nesper, Juhrb. Akud. Wiss. Gdrringen, Vandenhoek & Ruprecht. Gottingen,
1989, S. 15: R. Nesper, Habilitationschrift, Universitat Stuttgart, 1988;R.
Nesper, Prog. Solid Stutc. Chem. 1990.20, 1.
846
0 VCH VerlagsgesellschafrmbH. 0-69451 Weinheim. 1994
[8]U. Frank-Cordier, G. Cordier, H. Schafer, Z . Narurforscfi. B 1982,37, 119,
127;W Blase, G.Cordier, ibid. 1989,44,1011-1014.
[9]R. B. King, Inorg. Chem. 1989,28, 2796.
[lo]H. K. Hardy, J. M. Sicock, J. Inst. Met. 1956,24,423.
[ll]G. Bergman, J. L. T. Waugh, L. Pauling, Acra Crysfallogr. 1957,10, 254.
[I21 M. Tillard-Charbonnel, C. Belin, J. Solid State Chem. 1991,90,270.
[13]M.Tillard-Charbonnel, N. Chouaibi, C. Belin, J. Lappaset, Eur. .
I
SolidSrate
Inorg. Chem. 1992,29,347.
[14]M. Tillard-Charbonnel, N. Chouaibi, C. Belin, Mater. Res. Bull. 1992, 27,
1277.
[IS]C. Belin, M.Charbonnel, J. Solid. Stare Chem. 1993,64,57.
[I61 A. Niikura, A. P. Tsai, A. Inoue, T. Masumoto, A. Yamamoto, Jpn. 1 Appl.
Phys. Part 2 1993.32, 1160.
[17]H. G. von Schnering,T.Popp, unpublished; T. Popp, Dissertation,Universitat
Stuttgart, 1990.
[18]W. Kratschmer, L. D. Lamb, K. Fostiropoulos, D. R. Huffman, N a f w e 1990,
347,354.
[19]D. Ugarte, Nature 1992,359,707.
05704833~94~08084846
$ iO.OO+ ,2510
Angew. Chem. Int. Ed. Engl. 1994,33,No. 8
Документ
Категория
Без категории
Просмотров
1
Размер файла
538 Кб
Теги
without, space, enes, filled, fullerenes, fullercages, carbonчfulleranes
1/--страниц
Пожаловаться на содержимое документа