Unsaturated catalyst precursors

Electron Unsaturated Catalyst Precursors. As analyzed in previous Sections, the trinuclear cluster H20s3(CO)io is an electronically as well as coordinately unsaturated species. The hydrogenation of olefins in the presence of H20s3(CO)io thus constitutes a typical example of catalysis induced by intrinsic unsaturation of the catalyst precursor. In these cases, the activation energy for the formation of the complex catalyst-substrate is rather low. The mechanism proposed for this kind of catalysis is illustrated in the scheme in Fig. 2.71. 

Recently, the possibility that complexes of unsaturated "C ligands other than alkylidenes might also serve as catalyst precursors in olefin metathesis has... 

The discussion of the activation of bonds containing a group 15 element is continued in chapter five. D.K. Wicht and D.S. Glueck discuss the addition of phosphines, R2P-H, phosphites, (R0)2P(=0)H, and phosphine oxides R2P(=0)H to unsaturated substrates. Although the addition of P-H bonds can be sometimes achieved directly, the transition metal-catalyzed reaction is usually faster and may proceed with a different stereochemistry. As in hydrosilylations, palladium and platinum complexes are frequently employed as catalyst precursors for P-H additions to unsaturated hydrocarbons, but (chiral) lanthanide complexes were used with great success for the (enantioselective) addition to heteropolar double bond systems, such as aldehydes and imines whereby pharmaceutically valuable a-hydroxy or a-amino phosphonates were obtained efficiently. 

Involvement of trans-PtH2L2 in the catalytic cycle was confirmed by the wgs reaction, employing trans-PtH2L2 as the catalyst precursor, from which was also isolated trans-[PtH(C0)L]0H as its BPh salt. The following two processes (eq. 15,16) would complete the catalytic cycle. The formation of Pt3(C0)3L4 (v(C0) 1840,1770 enf ) in the reaction of PtH2L2 with CO is considered indirect evidence for the intermediacy of the coordinatively unsaturated Pt-(C0)L2. A simplified scheme of the cycle may then be depicted... 

This finding is the consequence of the distribution of various ruthenium(II) hydrides in aqueous solutions as a function of pH [RuHCl(mtppms)3] is stable in acidic solutions, while under basic conditions the dominant species is [RuH2(mtppms)4] [10, 11]. A similar distribution of the Ru(II) hydrido-species as a function of the pH was observed with complexes of the related p-monosulfo-nated triphenylphosphine, ptpprns, too [116]. Nevertheless, the picture is even more complicated, since the unsaturated alcohol saturated aldehyde ratio depends also on the hydrogen pressure, and selective formation of the allylic alcohol product can be observed in acidic solutions (e.g., at pH 3) at elevated pressures of H2 (10-40 bar [117, 120]). (The effects of pH on the reaction rate of C = 0 hydrogenation were also studied in detail with the [IrCp (H20)3]2+ and [RuCpH(pta)2] catalyst precursors [118, 128].)... 

For unsaturated lactones containing an endocyclic double bond also the two previously described mechanisms are presumably involved and the regio-selectivity of the cyclocarbonylation is governed by the presence of bulky substituents on the substrate. Inoue and his group have observed that the catalyst precursor needs to be the cationic complex [Pd(PhCN)2(dppb)]+ and not a neutral Pd(0) or Pd(II) complex [ 148,149]. It is suggested that the mechanism involves a cationic palladium-hydride that coordinates to the triple bond then a hydride transfer occurs through a czs-addition. Alper et al. have shown that addition of dihydrogen to the palladium(O) precursor Pd2(dba)3/dppb affords an active system, in our opinion a palladium-hydride species, that coordinates the alkyne [150]. 

The indirect cyclisation of bromoacetals via cobaloxime(I) complexes was first reported in 1985 [67], At that time the reactions were conducted in a divided cell in the presence of a base (40yo aqeous NaOH) and about 50% of chloropyridine cobaloximeflll) as catalyst precursor. It was recently found that the amount of catalyst can be reduced to 5% (turnover of ca. 50) and that the base is no longer necessary when the reactions are conducted in an undivided cell in the presence of a zinc anode [68, 69]. The method has now been applied with cobaloxime or Co[C2(DOXDOH)p ] to a variety of ethylenic and acetylenic compounds to prepare fused bicyclic derivatives (Table 7, entry 1). The cyclic product can be either saturated or unsaturated depending on the amount of catalyst used, the cathode potential, and the presence of a hydrogen donor, e.g., RSH (Table 7, entry 2). The electrochemical method was found with some model reactions to be more selective and more efficient than the chemical route using Zn as reductant [70]. 

A wide range of carbon, nitrogen, and oxygen nucleophiles react with allylic esters in the presence of iridium catalysts to form branched allylic substitution products. The bulk of the recent literature on iridium-catalyzed allylic substitution has focused on catalysts derived from [Ir(COD)Cl]2 and phosphoramidite ligands. These complexes catalyze the formation of enantiomerically enriched allylic amines, allylic ethers, and (3-branched y-8 unsaturated carbonyl compounds. The latest generation and most commonly used of these catalysts (Scheme 1) consists of a cyclometalated iridium-phosphoramidite core chelated by 1,5-cyclooctadiene. A fifth coordination site is occupied in catalyst precursors by an additional -phosphoramidite or ethylene. The phosphoramidite that is used to generate the metalacyclic core typically contains one BlNOLate and one bis-arylethylamino group on phosphorus. 

A number of ex situ spectroscopic techniques, multinuclear NMR, IR, EXAFS, UV-vis, have contributed to rationalise the overall mechanism of the copolymerisation as well as specific aspects related to the nature of the unsaturated monomer (ethene, 1-alkenes, vinyl aromatics, cyclic alkenes, allenes). Valuable information on the initiation, propagation and termination steps has been provided by end-group analysis of the polyketone products, by labelling experiments of the catalyst precursors and solvents either with deuterated compounds or with easily identifiable functional groups, by X-ray diffraction analysis of precursors, model compounds and products, and by kinetic and thermodynamic studies of model reactions. The structure of some catalysis resting states and several catalyst deactivation paths have been traced. There is little doubt, however, that the most spectacular mechanistic breakthroughs have been obtained from in situ spectroscopic studies. 

Although the catalytic reactions described above involve mononuclear Rh and Rh complexes, dinuclear Rh compoimds have also been studied as catalyst precursors in oxygenation reactions. The system [Rh2(p.-OAc)4]/ f-BuOOH is effective in the oxidation of cyclic alkenes such as cyclopentene, cyclohexene and cycloheptene, mainly to o, /i-unsaturated ketones and allylic acetates, but with poor yields (Eq. 4) [30,31]. 

As has already been mentioned (Section 5.2.1), the catalyst precursor RhH(CO)L3 has to undergo ligand dissociation to generate an unsaturated, cat-alytically active intermediate. In the absence of phosphines CO is the main ligand of the various catalytically active intermediates. It is obvious, therefore, that in the presence of both CO and L there are several possible equilibria. Under mild conditions, the ones shown in Fig. 5.3 have been observed by multinuclear ( ll, 13C, 31P) temperature-variable NMR. 

In the presence of appropriate ruthenium catalyst precursors, diallyl and allyl homoallyl ethers do not lead to the expected metathesis or cycloisomerization products, but undergo first isomerization to form allyl vinyl ethers, and then a Claisen rearrangement which gives unsaturated aide-... 

Although kis expected to be very large, occasionally the coordinatively unsaturated species, L M, will be intercepted by another substrate (e.g., a Si—C bond) and give rise to some net reaction, such as redistribution of groups on silicon. Stated another way, one function of the Si—H bond is to generate an active catalyst from some catalyst precursor [e.g., Eq. (23)] (19). A fuller discussion of mechanistic implications is reserved for Section III. 

Strictly speaking, a catalyst is some species directly involved in the catalytic cycle and, in the reactions discussed here, these species are usually low-valent, coordinatively unsaturated transition metal complexes. Metal halides, e-.g., chloroplatinic acid, PdCl, etc., although often claimed as catalysts are more properly catalyst precursors, since in the presence of silyl hydrides the metal halides are reduced. If no stabilizing ligands, e.g., olefins, phosphines, etc. are present, the reduction normally proceeds to a finely divided form of the metal or to insoluble metal silyl/hydride clusters which may act as heterogeneous catalysts. 

Unsaturated residue formed during catalytic reactions that produced paraffins and olefins is the source of alkyl aromatics and nonvolatile residue. When HZSM-5 catalyst is employed, aromatic alkyl chain sizes are restricted to C4 or smaller. The pores of HZSM-5 are large enough to allow formation of small alkyl aromatics by cyclization and dehydrogenation of surface species, but formation of fused unsaturated coke precursors are inhibited. Unlike HZSM-5, larger HY pores facilitate the formation of larger nonvolatile unsaturated coke precursors. 

---