146

Ref.

[17]

[18,19]

[18,19]

[20-22]

[24]

a- hex. = hexagonal, rhomb. = rhombohedral; b, g-cm°.

a- hex. = hexagonal, rhomb. = rhombohedral; b, g-cm°.

The different forms of carbynes were assumed to be polytypes with different numbers of carbon atoms in the chains lying parallel to the hexagonal axis and different packing arrangements of the chains within the crystallite. Heimann et al [23] proposed that the sizes of the unit cells were determined by the spacing between kinks in extended carbon chains, Fig. 3A. They were able to correlate the cQ value for the different carbyne forms with assumed numbers of carbon atoms, n (in the range n = 6 to 12), in the linear parts of the chains.

Fig. 3. A, A kinked polyyne chain model for linear carbynes (after [23]); B, cyclo C-18 carbyne [25].

Recently, Tanuxna and Palnichenko [24] have reported a new form of carbon which they call 'Carbolite' formed by quenching high temperature carbon vapour onto a metal substrate. Hexagonal Carbolite I was formed from an Ar-rich gas; a rhombohedral form, Carbolite II, was formed from an Ar-H2 gas mixture.

X-ray diffraction peaks were rather broad with coherence lengths as low as 20 nm and this was attributed to rapid quenching. It was proposed that the carbon atoms are arranged in polyyne chains (n = 4) along the c-axis. The density of Carbolite (1.46 g-cm"3) is lower than values for other carbynes and for diamond and graphite - hence the name - and this was attributed to a rapid quenching process.

Molecular orbital calculations indicate that cyclo C-18 carbyne should be relatively stable and experimental evidence for cyclocarbynes has been found [25], Fig. 3B. Diederich et al [25] synthesised a precursor of cyclo C-18 and showed by laser flash heating and time-of flight mass spectrometry that a series of retro Diels-Alder reactions occurred leading to cyclo C-18 as the predominant fragmentation pattern. Diederich has also presented a fascinating review of possible cyclic all-carbon molecules and other carbon-rich nanometre-sized carbon networks that may be susceptible to synthesis using organic chemical techniques [26].

Despite many publications on carbynes, their existence has not been universally accepted and the literature has been characterised by conflicting claims and counter claims [e.g., 27-29]. This is particularly true of meteoritic carbynes. An interesting account of the nature of elemental carbon in interstellar dust (including diamond, graphite and carbynes) was given by Pillinger [30]. Reitmeijer [31] has re-interpreted carbyne diffraction data and has concluded that carbynes could be stratified or mixed layer carbons with variable heteroelement content (H,0,N) rather than a pure carbon allotrope.

In addition to questions over interpretation of diffraction data, there are reservations about the stability of carbynes. Lagow et al [32] note that the condensation of the compound Li-C=C-Br to form carbon chains is potentially explosive. There is also the possibility of cross-linking between carbyne chains and the nature of the termination of the carbyne chains is unclear. Eastmond et al [33] showed that polyyne compounds of the type:

are stabilised by the bulky silyl end-groups. Lagow et al [32] also synthesised and determined the crystal structure of a polyyne with tertiary butyl end groups:

that was stable to ~130°C. They also found mass spectrometric evidence both where R = phenyl and n = 16-28, and of carbyne chains with lengths up to C300 after laser ablation of graphite in the presence of C2N2 and C2F6. The presumption was that these carbynes were stabilised by nitrile and trifluoromethyl end caps. For composites of carbynes and alkali metal fluorides produced by reduction of fluoropolymers with alkali metal amalgams, it is argued that the alkali metal matrix suppresses cross-linking of the carbyne

Despite the scepticism in some quarters, a large number of chemical and physical methods have been developed for producing carbynoid materials. These include: dehydropolymerisation of acetylene, dehydrohalogenation of polyvinylidene halides and reductive dehalogenation of poly(tetrafluoroethylene) and related compounds, condensation of carbon vapour produced by various means, e.g., laser ablation and arc discharge, shock compression of graphite and other solid forms of carbon [35], At present, no allcarbon carbynoid material has been isolated in large single crystal form and, consequently, full X-ray structural analyses and bulk property measurements have not been performed. (Note. An extensive review of carbynes by Russian workers [36] was published after this Section of the Chapter was completed.)

Fullerenes are described in detail in Chapter 2 and therefore only a brief outline of their structure is presented here to provide a comparison with the other forms of carbon. The C60 molecule, Buckminsterfullerene, was discovered in the mass spectrum of laser-ablated graphite in 1985 [37] and crystals of C60 were first isolated from soot formed from graphite arc electrodes in 1990 [38], Although these events are relatively recent, the C60 molecule has become one of the most widely-recognised molecular structures in science and in 1996 the co-discoverers Curl, Kroto and Smalley were awarded the Nobel prize for chemistry. Part of the appeal of this molecule lies in its beautiful icosahedral symmetry - a truncated icosahedron, or a molecular soccer ball, Fig. 4A.

The C60 molecule contains 12 pentagons and 20 hexagons. This type of hexagonal-pentagonal structure closely resembles the geodesic domes developed by the architect and engineer R. Buckminster Fuller, after whom the molecule is named. In the C60 molecule each carbon atom is bonded to three molecule is named. In the C60 molccule each carbon atom is bonded to three others by two longer bonds (length -145 pm) and one shorter bond (bond length -140 pm). These are conventionally referred to in the Fullerene literature as two C-C single bonds and one C=C double bond, although their bond orders are intermediate between a pure Csp3-Csp3 bond and a purely aromatic Car-Car bond, being close to the value for the C-C bond in the graphite basal plane, cf. Tables 1 and 2. The double bonds lie between two hexagons and are therefore known as 6:6 bonds whereas the single bonds link a hexagon to a pentagon and are known as 6:5 bonds. It follows that there is bonding anisotropy in the CfJ1 molecule since bonds around a pentagon are all single bonds and bonds around a hexagon are alternately single bonds and double bonds. It appears therefore that the bonding in Cso is mainly sp2 with delocalised n electrons, but with some sp3 character resulting from curvature of the C-C bonds.

Crystals of C60 formed by vacuum sublimation have a face-centred cubic (fee) crystal structure at room temperature, a0 = 1417 pm [38,39], Those grown from solution have a variety of crystal structures depending upon the solvent used, e.g., fee, hexagonal close packed, hep, or orthorhombic structures [40, 41]; some of these structures may be stabilised by solvent molecules. Solid state 13C nuclear magnetic resonance, nmr, and other spectroscopic studies show that, despite the bonding anisotropy, all carbon atoms in the C60 molecule are equivalent [42-44], This is because at room temperature the C60 molecules are re-orienting rapidly on their lattice sites. As the temperature is reduced, there is a phase transition at —260K to a primitive cubic structure [45] as a result of orientational ordering of some of the C60 molecules. At 86K there is a glass transition in which the orientational ordering is frozen [46, 47].

Fullerenes are a range of stable closed-shell carbon molecules and their derivatives, of which C60 is the archetype. The next highest stable member of the series is C70 which is found in small quantities with C^ in arc electrode soot. C70 may be regarded as a C60 molecule with an extra belt of hexagons inserted at the the stability of Fullerenes and the occurrence of 'magic numbers' in the Fullerene series. The rule states that closed carbon cages in which the pentagons are isolated from each other are likely to be more stable than those in which pentagons are in contact. Cso is the smallest closed shell carbon cluster that avoids abutting pentagons and O,0 is the next smallest. Other Fullerenes such as C7S, C78, C82, Cg4, have been isolated [50,51], Mass fragments in the range C^o-C9S0 detected in mass spectra [52,53] have been attributed to giant Fullerenes but none has been isolated.

As expected from its oblate spheroidal shape (a molecular rugby ball), the C-,a molecule has lower symmetry than the CM molecule. There are also five different types of C atom sites and eight different types of C-C bond in the C70 molecule. The structural chemistry of C70 crystals is also much more complex than for C60 crystals. At high temperatures an fee structure is found (a0 = 1501 pm). As the temperature is progressively lowered there is a complex series of transitions to rhombohedral, hexagonal, and monoclinic phases [54], Limited crystal structure studies on the higher Fullerenes, C72-C84, using scanning tunnelling microscopy [55,56] and micro-diffraction [57] show fee structures in each case. An excellent monograph on Fullerenes and carbon nanotubes has been published recently [58 ].

2.5 Some properties of the crystalline forms of carbon

It is outside the scope of this Chapter to undertake a comprehensive review of structure-property relationships for the different forms of carbon. However, a limited comparison of properties is useful for illustrating the influence of chemical bonding upon the properties of diamond, graphite and Buckminsterfullerene, C60, Table 4. Carbynes are omitted from the comparison since insufficient is known of their properties.

Table 4. Some properties of crystalline forms of carbon

Property

diamond *

graphite b

C c 60

0 0

Post a comment