Di-and Trimerization of Butadiene

Both di- and trimerization of butadiene with soluble nickel catalysts are well-established homogeneous catalytic reactions. The precatalyst having nickel in the zero oxidation state may be generated in many ways. Reduction of a Ni2+ salt or a coordination complex such as Ni(acac)2 (acac = acetylacetonate) with alkyl aluminum reagent in the presence of butadiene and a suitable tertiary phosphine is the preferred method. The nature of the phosphine ligand plays an important role in determining both the activity and selectivity of the catalytic 

The catalytic cycle proposed for the dimerization of butadiene is shown in Fig. 7.8. As shown by 7.24, two molecules of butadiene coordinate to NiL. A formal oxidative addition, as shown by Eq. 7.8, produces two nickel-carbon bonds and the carbon-carbon bond required for ring formation. The structure of 7.25 with two nickel-carbon bonds (see Fig. 7.8), is a hypothetical one that helps us to understand the carbon-carbon bond formation process. The actual catalytic intermediates that have been observed by spectroscopy have an rf-allyl type of bonding. As shown by reaction 7.9, species 7.25 can reductively eliminate 1,5-cyclooctadiene and the zerovalent nickel complex Ni-L. 

Similar oxidative additions involving the inner carbon atoms of the butadiene molecules can generate complexes having the formal structures 7.26 and 7.27. These may also be formed from 7.25 through tautomerization. Regeneration of 7.24 from these species involves elimination of vinyl cyclohexene and divinyl cyclobutane, respectively. 

The evidence for the proposed mechanism as shown in Fig. 7.8 comes mainly from in situ NMR studies and X-ray structures of isolated model complexes. Rapid equilibrium between species 7.25 to 7.27 involving an 7j3-allyl type of interaction results in a species of the type 7.28. This species has been observed by NMR. Similar model complexes such as 7.29 and 7.30 have been characterized by single crystal X-ray studies. 

In the absence of added phosphine the main product is the cyclic trimer of butadiene—cyclododecatriene. The presence of three double bonds in this molecule means other geometric isomers apart from the one shown in Fig. 7.6 exist. Identification of the species 7.31 by NMR is evidence for the involvement of rf-allyl intermediates. The complex 7.31 reductively eliminates cyclododecatriene. 

In 1959 butadiene polymerization was attempted by Wilke with a Zeigler type of catalyst made from NiCacac) plus Et AKOEt). Surprisingly, in this reaction di- and trimerization of butadiene rather 

The intermediate 7.28 with two rj -allyl ligands can be observed by in situ NMR spectroscopy. As shown by reaction 7.2.4.1, conversion of 7.28 to 7.29 can be treated as a formal reductive elimination reaction where the product, 1,5-COD remains coordinated to nickel before being eliminated. Similarly, as shown by reaction 7.2.4.2, the formation of vinyl cyclohexene can also be explained satisfactorily by arrow pushing formabsm. 

---

Figure 11 Representation of di- and trimerization of butadiene on nickel(0) to give bis-allylic intermediates and cyclic olefins.

Finally, as shown by reaction 2.3.3.4, two butadiene molecules can form two fj -allyl bonds to a nickel atom to give a nine-membered metallacycle. This type of reaction is of importance in the catalytic di-and trimerization of butadiene (see Section 7.2.5). 

In the previous chapters we discussed alkene-based homogeneous catalytic reactions such as hydrocarboxylation, hydroformylation, polymerization, and oligomerization. In this chapter we discuss a number of other homogeneous cataljTic reactions where an alkene is the only or one of the principal reactants. Some of the industrially important reactions that fall under the former category are selective di-, tri-, and tetramerization of ethylene, dimerization of propylene, and di-and trimerization of butadiene. 

An important question in light of the ease of chelation in the synthesis of the carbonyl complexes is whether it is possible to decoordinate the phos-phane arm, possibly to create a vacant coordination site for further chemistry. The question was addressed by treatment of 327 with 1,5-cyclooctadiene under photochemical reaction conditions, using the diene as the solvent, and resulted in a 41% yield of nonchelated cyclooctadiene complex 336 (Scheme 61). Treatment of this complex with diphenylethyne under reaction conditions normally allowing alkyne di- or trimerization reactions gave tetraphenylcyclo-butadiene complex 337 in 64% yield, showing that chemistry at the cobalt atom is possible without inhibition by a chelating phosphane arm. ... 

Alkene and Alkyne Dimerization and Trimerization. The low-valent mono-Cp zirconium compound CpZrMe(DMPE)2 (DMPE = l,2-bis(di-methylphosphino)ethane) catalyzes the dimerization of ethylene to 1-butene at low frequency (3 t.o./d), a process wherein the 1,3-butadiene complex 64 is presumed to be the catalyst (256). The clean dimerization of olefins using the Cp2ZrCl2/MAO (Al/Zr = 1) catalyst has recently been reported (257). Recently,... 

Recent studies on the allylation of alkynes with bis (7r-allyl) nickel have revealed that the Ni(0) generated in this process causes the trimeri-zation and, more importantly, the reductive dimerization of a portion of the alkyne (8). A deuterolytic work-up led to the terminally di-deuter-ated diene (5), supporting the presence of a nickelole precursor (4) (Scheme 1). The further interaction of 4 with 1, either in a Diels-Alder fashion (6) or by alkyne insertion in a C-Ni bond (7), could lead to the cyclic trimer 8 after extrusion of Ni(0), thereby accounting for the trimerizing action of Ni(0) on alkynes. This detection of dimer 5 then provided impetus for the synthesis of the unknown nickelole system to learn if its properties would accord with this proposed reaction scheme. Therefore, E,E-l,4-dilithio-l,2,3,4-tetraphenyl-l,3-butadiene (9) was treated with bis (triphenylphosphine) nickel (II) chloride or l,2-bis(di-phenylphosphino ethane)nickel(II) chloride to form the nickelole 10 (9) (Scheme 2). The nickelole reacted with dimethyl acetylenedicarboxylate to yield 11 and with CO to produce 12. Finally, in keeping with the hypothesis offered in Scheme 1, 10a did act as a trimerizing catalyst toward diphenylacetylene (13) to yield 14. 

---