Printer Friendly

Turning molecules into metals: 20th century alchemy.

The incorporation of p-dopants may be a useful way to generate more conducive materials

The design of molecular materials which conduct electricity is a concept that has captivated the imagination and challenged the skills of many synthetic chemists.

With the exception of the recent and exciting discovery of the conducting and super-conducting properties of alkali metal-doped |C.sub.60~ |1~, most synthetic molecular conductors are based on the use of charged |pi~-radicals, either in the form of charge transfer salts such as TTF TCNQ or radical ion salts such as those based on the TMTSF and BEDT-TTF donors |2~.

An alternative approach, one which obviates the need for a counterion, involves the use of neutral rather than charged radicals. Although this concept was first proposed nearly twenty years ago |3~, the development of these materials has, for practical reasons, been relatively slow. Practical issues aside, however, these systems contain a design flaw. As in any one-dimensional structure with a half-filled electronic band, these stacked radicals are prone to a charge density wave or Peierls instability, i.e., a tendency to dimerize.

This solid state distortion of evenly spaced radical plates opens up an energy gap at the Fermi level, and the materials are, at best, semiconductors. In order to stabilize highly conducting states the dimensionality of the structure must be increased; extended networks of lateral as well as longitudinal interdimer contacts must be developed. There are both physical and chemical approaches to achieving this end.

Our interest in this topic developed from on-going studies of the chemistry of unsaturated inorganic heterocycles containing thiazyl (SN) and selenazyl (SeN) linkages. While many closed shell ring systems are known |4~, the discovery of a variety of stable neutral 7|pi~-electron radicals prompted us to investigate the possibility of generating molecular materials with conductive properties. Our early work in this area focused on the use of pseudo odd-alterant ring systems, but we have since moved towards derivatives of the 1,2,3,5-dithiadiazolyl and diselena-diazolyl rings 1 (E = S, Se) |5~.

Radicals of this type can be prepared by reduction of the corresponding cations, which can be made by a variety of routes, the most versatile of which involves the condensation of a per-silylated amidine with S|Cl.sub.2~ or Se|Cl.sub.2~. In the solid state the radicals associate as dimers. For simple derivatives (e.g., R = H, Me, Ph) these materials are relatively volatile solids. For larger R-groups both volatility and solubility in organic media are low; the most effective method for purification and crystal growth is vacuum sublimation.

As indicated above, the central issue in molecular conductor design is the understanding and control of the chemical features which influence molecular packing. Early structural work on dithiadiazolyl derivatives, however, provided no indication that ordered stacks of dimers, were possible. The chemical challenge therefore consisted in the judicious (perhaps serendipitous) selection of R-group such that, in the solid state, the radical dimer sub-units were aligned in a columnar fashion, with tight contacts both within and between the stacked columns.

One design strategy that we have pursued involves the incorporation of a cyano-functionality within the R-group attached to the radical, the intent being to link radicals together by CN---S/Se contacts and favour ribbon-like rather than herring-bore packing patterns. A variety of cyanophenyl and cyanofuryl systems have now been characterized and, as is illustrated in Fig. 2 for 1 (R = 2-cyanophenyl, E = Se) the cyano groups do indeed serve as "molecular ties", producing the desired solid state ordering |6~. The mean intradimer Se - Se distance is 3.31 |Angstroms~, while the mean interdimer Se - Se contact is 4.08 |Angstroms~. The vertical stacking mode, however, carries a cost; the steric bulk of the organic substituent tends to force the stacked columns apart, reducing the inter-columnar contacts and thereby the widening the valence to conduction band gap.

A second, more ambitious approach involves the design and synthesis of bi- and even trifunctional radicals. Figure 3 illustrates the girder-like structure of the 1,3,5-benzene bridged triradical |C.sub.6~|H.sub.3~|(C|N.sub.2~|S.sub.2~).sub.3~ 2 |7~. Of particular interest is the fact that while each dimer head associates with a near neighbor, the molecular plates do not associate; instead a polymeric array of plates bound in a zig-zag fashion by weak S---S bonds is observed. This zig-zag arrangement of stacked plates is also found for the 1,3-phenylene bridged diradicals |C.sub.6~|H.sub.4~|(C|N.sub.2~|E.sub.2~).sub.2~ 3 (E = S, Se) |8~.

Consistently with the higher density of radicals interdimer contacts are both more numerous and tighter in 2 and 3 than in 1. We have also discovered that more than one phase is possible for some compounds. In the case of 3 (E = Se), for example, a second phase, consisting of chain-like arrays of discrete dimers, has been characterized |9~. Clearly, the energetic differences between different packing motifs are very subtle.

Extended Huckel band structure calculations on a range of mono- and polyfunctional systems reveal similar features for E = S and Se, with tighter band gaps for the latter. Trends in valence and conduction band dispersion, and hence band gap, correlate well with the number and magnitude of interdimer contacts.

We have probed the magnetic and conductivity characteristics of several sulphur and selenium based derivatives. At ambient temperatures the compounds are essentially diamagnetic, with low residual magnetism arising from defects. At elevated temperatures both compounds 2 and 3 (E = S) show a marked hysteretic increase in paramagnetism, a phenomenon that is not accompanied by a significant increase in conductivity. We conclude that this spin break-out does not serve as a source of carriers, as would be expected in most semiconductors.

A similar but very much smaller increase in magnetic susceptibility is observed in 3 (E = Se) at elevated temperatures. Variable temperature single crystalconductivity measurements on the stacked (|alpha~) and chain (|beta~) phases of 3 (E = Se) reveal Arrhenius-like behaviour at lower temperatures, suggestive of a conduction mechanism involving thermal activation; the derived band gaps for the two phases are 0.55 and 0.77 eV respectively.

In summary, the structures studied to date reveal that while the sulphur derivatives exhibit rather poor conductivity, the selenium compounds are semiconductors with conductivities ranging as high as |10.sup.-2~ S |cm.sup.-1~. Modification of the R-group of 1 constitutes an effective way of favouring the desired stacking mode, and also allows for subtle improvements in interdimer contacts and hence conductivity. Preliminary high pressure experiments indicate that physical pressure can significantly improve conductivity. Incorporation of p-dopants (e.g., iodine) also appears to a useful way to generate more conductive materials. Research in both these areas is currently in progress.

This work is a collaborative venture involving workers at the University of Guelph, the University of Arkansas and AT&T Bell Laboratories. Efforts at Guelph are funded by NSERC.

References

1) R.C. Haddon, Acc. Chem. Res. 25, 127 (1992).

2) J.R. Ferraro and J.M. Williams, Introduction to Synthetic Molecular Conductors, Academic Press, 1987.

3) R.C. Haddon, Aust. J. Chem. 28, 2343 (1975).

4) R.T. Oakley, Prog. Inorg. Chem. 36, 299 (1988).

5) A.W. Cordes, R.C. Haddon and R.T. Oakley, in "The Chemistry of Inorganic Ring Systems", Elsevier, Amsterdam, Ed. R. Steudel, p. 295 (1992).

6) A.W. Cordes, R.C. Haddon, R.G. Hicks, R.T. Oakley and T.T.M. Palstra, Inorg. Chem. 31, 1802 (1992).

7) A.W. Cordes, R.C. Haddon, R.G. Hicks, R.T. Oakley, T.T.M. Palstra, L.F. Schneemeyer and J.V. Waszczak, J. Am. Chem. Soc. 114, 5000 (1992).

8) M.P. Andrews, A.W. Cordes, D.C. Douglass, R.M. Fleming, S.H. Glarum, R.C. Haddon, R.T. Oakley, T.T.M. Palstra, L.F. Schneemeyer, G.W. Trucks, R. Tycko, J.V. Waszczak, K.M. Young and N.M. Zimmerman, J. Am. Chem. Soc. 113, 3559 (1991).

9) A.W. Cordes, R.C. Haddon, R.G. Hicks, R.T. Oakley, T.T.M. Palstra, L.F. Schneemeyer and J.V. Waszczak, J. Am. Chem. Soc. 114, 1729 (1992).
COPYRIGHT 1992 Chemical Institute of Canada
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1992 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Oakley, Richard T.
Publication:Canadian Chemical News
Date:Nov 1, 1992
Words:1370
Previous Article:Silicon-germanium for electronics and optoelectronics.
Next Article:Molecular materials for nonlinear optics.
Topics:

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters