5575
J . Am. Chem. SOC.1981,103,5575-5577
Observation of a Very Large Orbital Interaction through Four Bonds. An Alternative Model of Orbital Interactions through Bonds Michael N. Paddon-Row,*+*Harish K. Patney,* R. S.Brown,g and K. N. Houk* Departments of Chemistry New South Wales Institute of Technology PO Box 123,Broadway, New South Wales, Australia 2007 Louisiana State University, Baron Rouge, Louisiana 70803 University of Pittsburgh, Pittsburgh, Pennsylvania I5260 University of Alberta, Edmonton, Alberta, Canada T6G 2G2 Received November 25, 1980 In 1968, Hoffmann, Imamura, and Hehre reported calculations on diradical model systems which led to the now classic conceptual dissection of orbital interactions into through-space and through-bond varieties. These calculations indicated that the magnitude of orbital interactions through n bonds (OIT-n-B) should diminish regularly with increasing n.’ Experimental measures of this damping effect are not conclusive, despite the wealth of data that has been collected on OIT-3-B?q3 because only a few examples of OIT-4-B and OIT-5-B have appeared,4d and these are often in systems where the magnitude of interaction can only be inferred indirectly. We now present direct experimental evidence that OIT-4-B can be as large as OIT-3-B. We also describe a simple general model of OITB that readily explains and predicts magnitudes and stereochemical aspects of OITB. In the course of our studies of the chemical consequences of orbital interactions,’ we have measured the He I photoelectron spectra (PES) of diene 1, monoene 2,and the saturated analogue.*
6
Figure 1. Photoelectron spectra of saturated, diene, and monoene dimethanonaphthalenes. n ionizations are the relatively sharp features to the right in the second and third spectra.
&&& 5
4a
I
3
2
The PES of these molecules are given in Figure 1, and the ionization potentials (IPS) of these and related compounds are listed in Table I. The two A IPSof 1 differ by 0.87 eV, in spite of the ‘Visiting Scholar at the University of Pittsburgh and the LSU,on leave from the New South Wales Institute of Technology. Send correspondence to the Australian address. *N.S.W. Institute of Technology. f University of Alberta. *University of Pittsburgh. (1) (a) Hoffmann, R.; Imamura, A.; Hehre, W. J. J. Am. Chem. SOC. 1968, 90, 1499. (b) Hoffmann, R. Acc. Chem. Res. 1971, 4 , 1. (2) Review: Gleiter, R. Angew. Chem., Int. Ed. Engl. 1974, 13, 696. (3) (a) Gleiter, R.; Heilbronner, E.; Hekman, M.; Martin, H. D. Chem. Ber. 1973, 106, 28. (b) Bischof, P.; Hashmall, J. A.; Heilbronner, E.; Hornung, V. Tetrahedron Lett. 1969,4025. (c) Brogli, F.; Eberbach, W.; Haselbach, E.; Heilbronner, E.; Hornung, V.; Lemal, D. M. Helu. Chim. Acta 1973,56, 1933. (d) Paddon-Row, M. N.; Hartcher, R. J . Am. Chem. SOC. 1980, 102, 662. (4) Paddon-Row, M. N.; Hartcher, R. J. Am. Chem. Soc. 1980,102,671. ( 5 ) (a) Martin, H. D.; Schwesinger, R. Chem. Ber. 1974,107, 3143. (b) Bartetzko, R.; Gleiter. R.; Muthard, J. L.; Paquette, L. A. J. Am. Chem. Soc. 1978, 100, 5589. (c) Pasman, P.; Verhoeven J. W.; de Boer, Th. J. Tetrahedron Lett. 1977, 207. (6) A particularly clear cut example of the dependence of OITB-n-B on n concerns the series I-(CbC)-I. It was found that OITB between the u lone pairs and the u framework were attenuated with increasing m,being negligible for m = 3 (corresponding to OIT-7-B): Haink, H. J.; Heilbronner, E.; Hornung, V.; Kloster-Jensen, E. Helu. Chim. Acta 1970, 53, 1073. Heilbronner, E.; Hornung, V.; Maier, J. P.; Kloster-Jensen, E. J. Am. Chem. Soc. 1974, 96,4252. Bieri, G.; Heilbronner, E.; Jones, T. B.;Kloster-Jensen, E.; Maier, J. P. Phys. Scr. 1977, 16, 202. (7) Paddon-Row, M. N.; Lap, B.V.; Patney, H. K.;Warrener, R. N. Aust. J. Chem. 1980, 33, 1493. (8) The synthesis of these molecules will be reported in the full paper.
0002-7863/81/1503-5575$01.25/0
64
-H-
oJb%
ds
4-
6-a mixing in 1 and 2. The a and top two u orbitals before interaction are given in the middle. On the left: n (basis)-HO-a-MO mixing. On the right: mixing of a f n with HOa-MO and SHO-U-MO.
Figure 2. Schematic representation of
Table I. Vertical n Ionization Potentials (k0.03 eV) and Orbital Energies ( E ~ ~of Dimethanonaphthalenes ~ ~ - ~ ~ ) compd
IP (n),eV
EnSTO-sG, eV
1
8.48 9.35 8.85 8.465a 8.9OSa 8.18 8.65
-1.46 -8.35 -1.85 -1.54 - 8.06 -1.19 -1.69
2 5 6 7
assignment n-n
n+n n n- n n+n n n
impossibility of direct overlap, and the separation of A orbitals by four u bonds. This 0.87-eV splitting energy is barely less than that typically found in those molecules exhibiting purely ?r OIT-3-B such as 3 (0.97 eV)3aand 4 (0.85 eV).3c39 Indeed, the A 0 1981 American Chemical Society
5516
Communications to the Editor
J. Am. Chem. Soc., Vol. 103, No. 18, 1981
IP split in 1 and analogous molecules results from differential mixing of the A - a and a a orbitals with framework occupied u orbitals of appropriate symmetry. This provides not only a more satisfactory explanation of experiment but also a simpler predictive method, as described below. The essential features of our revised OITB model are as follows: (1) a (or lone pair) orbitals of closed-shell systems interact through bonds principally by mixing only with occupied u orbitals; (2) the orbital energies of the a f a combinations of dienes, and of the corresponding monoene a basic orbital, are all raised through mixing with the u orbitals but by different amounts.” This model is best understood, for example, as diagrammed in Figure 2. An important feature of our model is the raising of the a basis leuel, by an amount x , through its mixing with HOa-MO (there are no symmetry restrictions to this mixing). It is easily shown that the level of the symmetry adapted pair, 1/[42(7r - a)], is raised by twice the amount, 2x, through its mixing with the antisymmetric HO-U-MO.‘~However, the level of the symmetry adapted pair, (1/42)(a 3. a), is only marginally raised (
