Butene isomerisation mechanisms

Fig. 10. Mechanism for the isomerisation of the n-butenes involving 7T-allylic intermediates.

The observed deuterobutene distributions together with the calculated AAprofiles, for those metals where unique A/-profiles could be obtained, the surface D/H ratio and the calculated deuterobutene distributions are shown in Table 29. One of the major features of these results is that, over all the metals studied, the trans- and cis-but-2-ene profiles show pronounced maxima at -N2 This clearly shows that the predominant route to the formation of but-2-ene was direct 1 4-addition of two hydrogen atoms to adsorbed buta-1 3-diene. 1 2-Addition of hydrogen to yield but-l-ene-A/j followed by isomerisation would have led to a zero value for but-2-ene-A/j and a maximum at but-2-ene-AA3 or higher depending upon the number of butene—butyl interconversions before desorption of the but-2-ene. The detailed interpretation of the A/-profiles has been discussed fully by Wells and co-workers [166,167] who have proposed the two mechanisms shown in Fig. 37. 

BF3 HjO and BF3 - 2 H2O, are formed at 25 C in equilibrium with free boron fluoride. In the same paper, the isomerisation of butene-2 by this catalytic mixture was examined kinetically. The rate of isomerisation was found to be proportional to the free BF3 as well as to its water-complex concentration, and to the olefin concentration. No complex between boron fluoride and mcmomer was identified, but in their conclusions the authors noted that a mechanism involving attack of the complex BF3 - H2O the complex butene-BF3 would satisfy the kinetics, but seems to have little else to recommend it . In fact we feel that precisely that mechanism was operative, through an Ad 3 intermediate, the Lewis acid-olefin complex being formed, but as in many other instances being too weak to be detected. In the light of present-day knowledge, this conclusion seems quite acceptable, but twenty years ago it probably seemed far-fetched. In any case, the authors clearly realised this possibility, but were too cautions about proposing it outri t. 

This rearrangement is well known in liquid phase and numerous mechanism studies of this [3,3] sigmatropic reaction on vanadates O = V(OR)3 appeared in the litterature [3,12]. But we found only four publications in heterogeneous vapor phase on that isomerisation, people from PUBLICKER [13] using phosphates catalysts obtained essentially methyl-2 butene-2 yne-3 3 (Scheme 2) like BERGMANN [9]. 

Comparison with other molecular sieves (Fig. 11) shows that the yields obtained with FER are very high indeed. As elaborated upon elsewhere [6-8], we have proposed that the isomerisation involves a bi-molecular mechanism in which e.g. di-methylhexene isomers crack selectively to isobutene and n-butene (Fig. 12). The mono-molecular mechanism requires the energetically unfavourable primary carbenium ions. Molecular modelling [7] has provided support for this mechanism in that the branched octenes can be formed in the intra-crystalline voids of FER but their diffusion out of the pores is hindered. 

Scheme 8.17 Unimolecxilar and bimolecular mechanisms for the isomerisation of butenes.

---