CO2 insertion

Reaction between butadiene and CO2 has been extensively studied (171) since the reaction was first demonstrated (167—170). This reaction has been shown to be catalyzed by Pd (172,173), Ni (174), Ru (175), Pt (178), and Rh (172,173) catalysts. Products include gamma (5) and delta lactones (6), acids (7,8), and esters (9). Mechanistic studies have shown that butadiene initially forms a dimer (Pd, Ru, Ni) or trimer (Rh) intermediate followed by CO2 insertion (171). The fate of these intermediates depends on the metal, the ligands, and the reaction conditions. 

If the starting material contains M-H or M-C bonds a further complication can arise due to the possibility of a CO2 insertion reaction. Thus, both [Ru(H)2(N2)(PPh3)3] and [Ru(H)2(PPh3)4] react to give the formate [Ru(H)(OOCH)(PPh3)3], and similar CO2 insertions into M-H are known for M = Co, Fe, Os, Ir, Pt. These normal insertion reactions are consistent with the expected bond polarities M +-H and 0 =C +=0, but occasionally abnormal insertion occurs to give metal carboxylic acids... 

CO2 insertion into M C bonds has, of course, been known since the first papers of V. Grignard in 1901 (p. 134). Organo-Li (and other M and M") also react extremely vigorously to give salts of carboxylic acids, RCO Li, (RC02)2Be, etc. Zinc dialkyls are much less reaetive towards CO2, e.g. 

Insertion reactions of CS2 arc known for all the elements which undergo CO2 insertion... 

Reactions of the cyclopentadienyl-amidinate-supported imidotitanium complexes with CO2 proceed via initial cycloaddition reactions, but depending on the imido Af-substituent go on to yield products of either isocyanate extrusion or unprecedented double CO2 insertion (Scheme 89). ... 

Figure 5. Optimized structures participating in the CO2 insertion into the Cu-H bond of CuH(PH3)2. 2 A is the reactant, 3A is the transition state, and 4A-6A are products [34b],...

The CO2 insertion into the Cu(I)-H bond leading to Cu(V-COOH)(PH3)2 was compared with the insertion leading to Cu(,n1-OCOH)(PH3)2. Because the transition state (TS) of the C02 insertion leading to the Cu-(V-COOH) species could not be optimized with the Hartree-Fock method used in the geometry optimizations, the assumed TS-like structure was calculated with the O-H distance and the Cu-H-0 angle arbitrarily taken to be 2.0 A and 100... 

We recently investigated [40] the reason why C02 is inserted into the Rh(I)-H bond with a significantly lower barrier than into the Rh(III)-H bond, as shown in Table 2. As discussed above, charge-transfer from the metal-hydride moiety to the K orbital of CO2 is very important in the CO2 insertion reaction, and, at the same time, the metal-formate moiety is very much stabilized by the donation of electrons from the metal fragment. Since the Rh(I) center is more electron-rich than Rh(III), the charge-transfer from the Rh(I)-H moiety to the k orbital of C02 is favored, and the formate moiety is provided with sufficient electrons. Consequently, CO2 is more easily inserted into the Rh(I)-H bond than into the Rh(III)-H bond. 

Aluminum porphyrins first came to attention with the discovery that the simple alkyl complex Al(TPP)Et was capable of activating CO2 under atmospheric pressure. Both irradiation with visible light and addition of 1-methylimidazole were required for the reaction, which was proposed to proceed by initial coordination of the base to aluminum. The aluminum porphyrin containing direct product of CO2 insertion was not isolated, but was proposed on the basis of IR data to be (TPP)A10C(0)Et, which was then treated with HCl gas, presumably liberating propanoic acid, subsequently isolated as the butyl or methyl ester after reaction with 1-butanol or diazomethane, respectively [Eq. (5)]. Insertion of CO2 into the Al—C bond of an ethylaluminum phthalocyanine complex has also been reported. ... 

Aluminum porphyrins with alkoxide, carboxylate, or enolate can also activate CO2, some catalytically. For example, Al(TPP)OMe (prepared from Al(TPP)Et with methanol) can bring about the catalytic formation of cyclic carbonate or polycarbonate from CO2 and epoxide [Eq. (6)], ° - and Al(TPP)OAc catalyzes the formation of carbamic esters from CO2, dialkylamines, and epoxide. Neither of the reactions requires activation by visible light, in contrast to the reactions involving the alkylaluminum precursors. Another key difference is that the ethyl group in Al(TPP)Et remains in the propionate product after CO2 insertion, whereas the methoxide or acetate precursors in the other reactions do not, indicating that quite different mechanisms are possibly operating in these processes. Most of this chemistry has been followed via spectroscopic (IR and H NMR) observation of the aluminum porphyrin species, and by organic product analysis, and relatively little is known about the details of the CO2 activation steps. 

Consecutive CO2 insertion has not been observed (and is assumed to be thermodynamically unfavored). However, consecutive epoxide ring opening is common, particularly for Lewis acidic catalysts like zinc derivatives (Scheme 4). 

Electrochemical reductions of CO2 at a number of metal electrodes have been reported [12, 65, 66]. CO has been identified as the principal product for Ag and Au electrodes in aqueous bicarbonate solutions at current densities of 5.5 mA cm [67]. Different mechanisms for the formation of CO on metal electrodes have been proposed. It has been demonstrated for Au electrodes that the rate of CO production is proportional to the partial pressure of CO2. This is similar to the results observed for the formation of CO2 adducts of homogeneous catalysts discussed earlier. There are also a number of spectroscopic studies of CO2 bound to metal surfaces [68-70], and the formation of strongly bound CO from CO2 on Pt electrodes [71]. These results are consistent with the mechanism proposed for the reduction of CO2 to CO by homogeneous complexes described earlier and shown in Sch. 2. Alternative mechanistic pathways for the formation of CO on metal electrodes have proposed the formation of M—COOH species by (1) insertion of CO2 into M—H bonds on the surface or (2) by outer-sphere electron transfer to CO2 followed by protonation to form a COOH radical and then adsorption of the neutral radical [12]. Certainly, protonation of adsorbed CO2 by a proton on the surface or in solution would be reasonable. However, insertion of CO2 into a surface hydride would seem unlikely based on precedents in homogeneous catalysis. CO2 insertion into transition metal hydrides complexes invariably leads to formation of formate complexes in which C—H bonds rather than O—H bonds have been formed, as discussed in the next section. 

Formate production has also been reported for electropolymerized films of [Co(4-vinylterpyridine)2] " on glassy carbon electrodes in dimethylformamide solutions [63]. Interestingly, the product of this same catalytic system in aqueous solutions is formaldehyde [81]. Other heterogeneous systems that produce formate include Cd, Sn, Pb, In, and Zn electrodes in aqueous media [12] (see also Vol VII 5.2.3). It is likely that the pathway to formate formation on metal electrodes follows the sequence of M—H bond formation followed by CO2 insertion to form a M—0C(0)H species followed by desorption from the electrode surface. 

CO2 insertion into TT-allylpalladium complexes exhibits the opposite regioselectivity, with, for example, exclusive reaction at the more substituted terminus of a butenylpalladium complex (equation... 

Alkylation of RhCl(PPh3)3 yields unstable alkyls that undergo CO2 insertion Rh(OCOPh)(PPh3)3 has monodentate benzoate (X-ray). 

Tn iridium chemistry, some indirect CO2 insertions have been reported by English and Heiskovitz [54]. Reaction of Ir(depe)C) and COj in acetonitrile yields an iridium carboxylate hydride. 

The catalytic mechanism of the enzyme carbonic anhydrasc, already mentioned in Section 2.1. is thought to be an attack of OH on a 7 n(ll) complex in enzymatically-bound carbon dioxide, Tltis biochemical reaction can also be interpreted as a CO2 insertion into an. M-0 bund [83,84). 

Chisholm extensively studied the CO2 insertion into Group VI metal atkoxides [87-89]. The complexes of formula M2(OR), readily and reversibly react with... 

Berke and coworkers have shown how hydrides can be activated by a trans nitrosyl so that CO2 inserts into the MoH bond of wer-Mo(CO)H(NO)(PMe3)3 to give a formato-complex. ... 

These complexes are easily reoxidized by dioxygen to TppFe R.i Alkyliron(III) porphyrins will undergo CO and CO2 insertion into the Fe-C bond, forming acyliron(III) (PFeC(O)R) and alkyl carboxylate (PFeOC(O)R) complexes, respectively. ... 

The CO insertion reaction into the metal hydride bond is in fact a member of the class of ligand insertion reactions to which much theoretical work has been devoted (28,29-35). Some years ago we analyzed the ethylene insertion into the rhodium hydride bond of a Rh(III) hexacoordinated complex (. We later focused our attention on the CO insertion reaction into the Mn-H bond ofHMn(CO)5 (37-39) and very recently we have undertaken the study of the CO2 insertion reaction into the Cr-H bond of HCr(CO)5 (C. Bo and A. Dedieu, Inorg. Chem., in press). We will concentrate here on the CO insertion reaction and compare it to the two other insertion reactions. The study of the reaction (1) was carried out at both the SCF and... 

---