Propellanes isomerization

It may be useful either at the outset or post factum to use molecular mechanics to calculate which of these products may be the more stable. A priori there is no way to tell whether either of the two products is the generally preferred one. In the present case such calculations do not appear to have been carried out at the outset not surprisingly. This sort of thing was not done at the time the work was conducted. I know of only one very recent paper in which such calculations appear, comparing intermediates between a dispiran with its isomeric propellane (see below 40)). 

Although we are dealing with work described in the literature by means of a posteriori molecular calculation it is useful to see the relative calculated heats of formation of the isomeric propellanes and dispirans and note particularly the right-most columns so as to be in a position to gauge these against the experimental results or vice versa. 

We tried in 1968, as it turned out, unsuccessfully, to prepare a propellane having the [4.2.2] nucleus (actually it was a [4.2.2]propellene). But we were most successful in preparing the isomeric dispiran 7. None of the desired propellane was formed (see reaction scheme) ... 

In the abovementioned cases preparation of the propellanes was direct. A very nice instance exists, however, of rearrangement of a dispiran, 26, to Dewar benzenes 27 which happen to be [n.2.2]propelladienes. Silver ion (silver perchlorate at —20 °C) promotes the isomerization, as shown 10). 

There are many more syntheses of heterocyclic propellanes from 1,1,2,2-substituted carbocyclic starting materials. The tetrol discussed above, when treated with KHS04 at 170-190 °C affords the dioxa[3.3.2]propellane shown no isomeric spiran is mentioned. Although the yield is only 50% perhaps some dispiran is hiding in the brauner Ruckstand from which the propellane diether is either crystallized at low... 

In the Baeyer-Villiger oxidation of [4.3.3]propellane-8,ll-dione 74 the propellane-bis-lactones formed, 75 and 76, are accompanied by a dispirolactone 77 22. Different product mixtures result when different (acidic or more basic) reaction conditions are employed but it has been shown experimentally for the head-to-tail propellane bis-lactone 75, vis-a-vis the isomeric dispiran 77, the latter appears to be the thermodynamically more stable product, resulting from the former under acidic conditions (p-TsOH/ C6H6, 7 days, r.t.). The structures were established by means of X-ray diffraction and H- and 13C-NMR spectroscopy. 

Since our work on the preparation of a dispiran rather than the isomeric propellane 7>, I have lived with the feeling, alas, the feeling, by no means the certainty, that propellanes should be available from dispirans when the relative stability permits it. (Cf. Ref. 3 b). 

Further, a very recent paper has reported a cascade rearrangement, under acidic conditions of the simpler educt, a dispiro[3.0.3.3]undecane derivative 93 to the dehydrated isomeric propellane 94 40). It is somewhat reminiscent of the analogous case of 26 where silver ion is the catalyst10). Treatment of the dispiro-alcohol 78 when heated for 2 hrs at 70 °C with p-toluenesulfonic acid in benzene gives in quantitative yield the [3.3.3]propellene 94. The following cascade is proposed to explain the rearrangement. 

It appears that if one wishes to anticipate whether a 1,2-dispiran or an isomeric propellane would be formed, it would be wise, before beginning trial and error, to carry out simple molecular mechanics calculations as to which product experiment is likely to give. 

The oxidation of 3,6-dehydrohomoadamantane (52) with NO+BF4, photo-excited tetracyanobenzene, and under anodic conditions has been found to involve a common radical cation intermediate. The study has shown that the activation of propellane cTc-c bonds with strong oxidizing electrophiles occurs by a sequence of single-electron transfer steps. These findings are supported by ab initio computations showing that the isomeric radical cations can equilibrate with low barriers and lead to a common product. ... 

It is interesting to note that head-to-tail dimeric structures are proposed for these products. On the other hand, the head-to-head cyclobutane dimers 15 and 16 are obtained on isomerization of the propellanes 13 and 14. respectively.8 The bridgehead alkenes have been proposed as intermediates as is evident from trapping of the adducts with furan. An X-ray crystallographic study of 15 confirmed the configurations of both cyclobutanes 15 and 16 indicating syn head-to-head dimerization, as 15 could cleanly be converted to 16 by catalytic hydrogenation. 

Sulfuration of acenaphtho[l,2- ]acenaphthylene 80 with elemental sulfur (1 molar amount as Sg) in DMF at 130°C gave the pentathiepane 7, with a [5.3.3]propellane structure, as the sole product, in 88% yield (Scheme 15) <1998TL2605, 1999EJ0597>. In the case of 3-phenylacenaphtho[l,2- ]acenaphthylenes 83, pentathiepanes were separated as a mixture of two conformers 8 and 9 in the equilibrium ratio of 55 45. Both isomers 8 and 9 slowly isomerized into each other in solution at room temperature. 

There are numerous other examples of stereochemical correspondence in propeller molecules. The propellane 19> 1e has only one energetically reasonable isomerization mechanism available a twisting motion about the C3 axis which corresponds to the three-ring flip as well as to the trigonal twist rearrangement (Fig. 5). Similarly, only one isomerization mechanism (stereochemically correspondent in the permutational sense to the zero-ring flip) is energetically reasonable for tri-o-thymotide (Id). 8,20)... 

In the first approach shown in Scheme 9, ketoester 77 was alkylated successively with 4-bromobutene and 1,3-dibromopropene. After decarboxylation, 78 was converted into iV-aziridinylimine 79 in good yield. The pivotal radical cyclization reaction proceeded smoothly to produce a mixture of isomeric propellane compounds 80, which was purified after the epoxidation step. For the synthesis of modhephene, the mixture of epoxides was rearranged into the corresponding allylic alcohols 81 and then the allylic alcohols were oxidized, giving a separable mixture of unsaturated ketones 82a and 82b. The major product 82a possessed the correct stereochemistry of the methyl group of modhephene. Since 82a had already been converted into modhephene, a formal total synthesis of dZ-modhephene has thus been completed. The isomeric ratio of 80 reflects the stereoselectivity during the radical cyclization reaction. The selectivity was very close to the ratio reported by Sha in his radical cyclization reaction. ... 

A special case of a transannular cycloisomerization occurs with the ethyl acrylate derivative 22. lO a.b Nickel(O), as well as palladium(O), catalysts effect this reaction in generally high yields, leading to the formation of the [3.3.3]propellane derivative 23. Even in the nickel(0)-catalyzed reaction, no isomeric compound arising from proximal cleavage of the methylenecyelopropane is formed, presumably due to activation of the distal bond arising from the tetrasubstitution of the methylene group. 

The reaction of propellane 13 with nitrosobenzene was found to afford a 2 1 mixture of isomeric anti/syn Diels-Alder adducts [Eq. (8)].17... 

The cyclopentenones (70) do not undergo cycloaddition reactions with cyclohexene. The only reaction encountered in the irradiation of these molecules is the facile isomerization by a 1,3-alkyl shift to (71). In contrast with this, irradiation of cyclopentenone (72a) in the presence of ethylene yields (73a). The cyclopentenone (72b) yields (73b) with cyclohexene. It is clear from these results that there is some structural phenomenon within the molecules which make some, the (4,3,2)-propellanes prone to rearrange, while others, the (3,3,3)-propellanes, undergo cycloaddition. 

Ring expansion reactions of the type shown in equations (17) and (18), in which a more equitable distribution of carbon atoms between two rings of a polycyclic hydrocarbon is produced, provide the link between the decyclization reactions above and the annulation reactions considered later. It would seem that the greater the strain in the reactant, the larger the number of potential catalysts. The complexes of seven different metals catalyze the reaction shown in equation (17). Of the possible catalysts, only the silver (I) salts bring about any byproduct formation.The isomerization of [l.l.l]propellane (equation 18) is catalyzed by [RhCl(PPh3)3]. ... 

Treatment of the bridgehead dibromide 157 with sodium in refluxing triglyme (216 °C), with direct distillation of the product into a cold trap, furnishes 75% of the parent [3.1.1]propellane (158). This compound is stable in toluene solution at room temperature but polymerizes in the absence of solvent. It reacts at — 78 °C with methanol to give largely a mixture of the three isomeric ethers 159, 160 and 161. 

Silver nitrate-catalysed rearrangements in methanol of dibromocyclopropyl propellanes have been examined. For example, (80 R = Br) gave a mixture of the corresponding methyl ketal (80 R = OMe) and bicyclo[5,4,0]undec-l(7)-en-2-one, together with minor products. One step in a reported synthesis of 1,7-methano-[12]annulene involved addition of dibromocarbene to tricyclo[4,4,l,0 ]undeca-3,8-diene (81) followed by silver acetate-catalysed ring-opening of the resultant bis-dibromocarbene adduct to give a mixture of isomeric acetates. ... 

The diene (1119) has been made in low yield from (1118) by base treatment of the double a-chloro-sulphone and is stable at 0°C but polymerizes above 100 °C, apparently to (1121). There is no evidence of isomerization to the very strained (1120). The photocycloaddition of alkenes to (1122) gives moderate yields of [5,3,2]propella-nones. Reaction of (1123) with diazomethane and CuBr gives 87% of (1124), which opens on heating to (1125) and reacts with HO Ac slightly faster than [3,2,1]-propellane. Dimethyl butyndioate reacts with (1124X giving (1126) and (1127). ... 

The pentalenone 6 is an intermediate in the synthesis of 3.3.3. propellane sesquiterpenes like modhephene. But, again, poor yields were observed by Paquette in the multi-step synthesis of by acylation of cyclopentene or 1-trimethylsilylcyclopentene and cyclization of the corresponding divinyl-ketone. 41,42 Moreover, two isomeric pentalenones were obtained and it was necessary to isomerize with rhodium trichloride. 

Several papers have appeared dealing with the synthesis of strained bridgehead olefins (anti-Bredt olefins). Conditions are described whereby a 10 1 mixture of the olefins (25) and (26) is formed by the vacuum pyrolysis of the bridgehead chloro-compound (24). ° The olefin (25) can form a reversible stabilized complex with [Pt°(PPh3)2], and this same catalyst can also effect irreversible isomerization to (27). The lead tetra-acetate-induced oxidative decarboxylation of the propellane carboxylic acid (28) produces the stable olefin (29) in good yield. ... 

A study of the consequences of through-bond versus through-space coupling in strained propellanes indicates the probable importance of a new kind of isomerism. The isomers are related by a simple bond-stretching process. In tricyclo[2,2,2,0 ]octane, extended Hiickel calculations show that isomers... 

In the course of these mechanistic studies a number of propellanes have been prepared. Paquette has used the route via a-chlorosulphones described last year to prepare a series of substituted [4,4,2]propellanes. Pyrolysis of triene (267) affords the cyclo-octatetraene (268) and similarly (269) gives (270). Labelling studies show that rearrangement proceeds via an intramolecular +, 2J cycloaddition followed by valence isomerization. Further Diels-Alder additions to (271) and related propellanes are described, and the e.s.r. spectrum of the radical obtained by electrolytic reduction of (272) is recorded. Other important synthetic studies concern [4,2,l]propellanes, [4,3,1]-propellanes, and [4,2,2]propellanes. ... 

Cargill etal. have described an interesting synthesis of the propellane (110) by isomerization of the tricycle (109) resulting from irradiation of the cyclohexenone... 

---