CONSPECTUS: Decarboxylation chemistry has a rich history, and in more recent times, it has been recruited in the quest to develop cheaper, cleaner, and more efficient bond-coupling reactions. Thus, over the past two decades, there has been intense investigation into new metal-catalyzed reactions of carboxylic substrates. Understanding the elementary steps of metal-mediated transformations is at the heart of inventing new reactions and improving the performance of existing ones. Fortunately, during the same time period, there has been a convergence in mass spectrometry (MS) techniques, which allows these catalytic processes to be examined efficiently in the gas phase. Thus, electrospray ionization (ESI) sources have been combined with ion-trap mass spectrometers, which in turn have been modified to either accept radiation from tunable OPO lasers for spectroscopy based structural assignment of ions or to allow the study of ion-molecule reactions (IMR). The resultant "complete" gas-phase chemical laboratories provide a platform to study the elementary steps of metal-catalyzed decarboxylation reactions in exquisite detail. In this Account, we illustrate how the powerful combination of ion trap mass spectrometry experiments and DFT calculations can be systematically used to examine the formation of organometallic ions and their chemical transformations. Specifically, ESI-MS allows the transfer of inorganic carboxylate complexes, [RCO2M(L)n](x), (x = charge) from the condensed to the gas phase. These mass selected ions serve as precursors to organometallic ions [RM(L)n](x) via neutral extrusion of CO2, accessible by slow heating in the ion trap using collision induced dissociation (CID). This approach provides access to an array of organometallic ions with well-defined stoichiometry. In terms of understanding the decarboxylation process, we highlight the role of the metal center (M), the organic group (R), and the auxiliary ligand (L), along with cluster nuclearity, in promoting the formation of the organometallic ion. Where isomeric organometallic ions are generated and normal MS approaches cannot distinguish them, we describe approaches to elucidate the decarboxylation mechanism via determination of their structure. These "unmasked" organometallic ions, [RM(L)n](x), can also be structurally interrogated spectroscopically or via CID. We have thus compared the gas-phase structures and decomposition of several highly reactive and synthetically important organometallic ions for the first time. Perhaps the most significant aspect of this work is the study of bimolecular reactions, which provides experimental information on mechanistically obscure bond-formation and cross-coupling steps and the intrinsic reactivity of ions. We have sought to understand transformations of substrates including acid-base and hydrolysis reactions, along with reactions resulting in C-C bond formation. Our studies also allow a direct comparison of the performance of different metal catalysts in the individual elementary steps associated with pro...

Copper-mediated allylic substitution reactions are widely used in organic synthesis, whereas the analogous reactions for silver and gold are essentially unknown. To unravel why this is the case, the gas-phase reactions of allyl iodide with the coinage metal dimethylmetallates, [CH(3)MCH(3)](-) (M = Cu, Ag and Au), were examined under the near thermal conditions of an ion trap mass spectrometer and via electronic structure calculations. [CH(3)CuCH(3)](-) reacted with allyl iodide with a reaction efficiency of 6.6% of the collision rate to yield: I(-) (75%); the cross-coupling product, [CH(3)CuI](-) (24%); and the homo-coupling product, [C(3)H(5)CuI](-) (1%). [CH(3)AgCH(3)](-) and [CH(3)AuCH(3)](-) reacted substantially slower (reaction efficiencies of 0.028% and 0.072%). [CH(3)AgCH(3)](-) forms I(-) (19%) and [CH(3)AgI](-) (81%), while only I(-) is formed from [CH(3)AuCH(3)](-). Because the experiments do not detect the neutral product(s) formed, which might otherwise help identify the mechanisms of reaction, and to rationalize the observed ionic products and reactivity order, calculations at the B3LYP/def2-QZVP//B3LYP/SDD6-31+G(d) level were conducted on four different mechanisms: (i) S(N)2; (ii) S(N)2'; (iii) oxidative-addition/reductive elimination (OA/RE) via an M(III) η(3)-allyl intermediate; and (iv) OA/RE via an M(III) η(1)-allyl intermediate. For copper, mechanisms (iii) and (iv) are predicted to be competitive. Only the Cu(III) η(3)-allyl intermediate undergoes reductive elimination via two different transition states to yield either the cross-coupling or the homo-coupling products. Their relative barriers are consistent with homo-coupling being a minor pathway. For silver, the kinetically most probable pathway is the S(N)2 reaction, consistent with no homo-coupling product, [C(3)H(5)AgI](-), being observed. For gold, no C-C coupling reaction is kinetically viable. Instead, I(-) is predicted to be formed along with a stable Au(III) η(3)-allyl complex. These results clearly highlight the superiority of organocuprates in allylic substitution reactions.

No abstract

A combination of gas-phase 3D quadrupole ion trap mass spectrometry experiments and density functional theory (DFT) calculations have been used to examine the mechanism of thermal decomposition of fluorinated coinage metal carboxylates. The precursor anions, [CF(3)CO(2)MO(2)CCF(3)](-) (M = Cu, Ag and Au), were introduced into the gas-phase via electrospray ionization. Multistage mass spectrometry (MS(n)) experiments were conducted utilizing collision-induced dissociation, yielding a series of trifluoromethylated organometallic species and fluorides via the loss of CO(2), CF(2) or "CF(2)CO(2)". Carboxylate ligand loss was insignificant or absent in all cases. DFT calculations were carried out on a range of potentially competing fragmentation pathways for [CF(3)CO(2)MO(2)CCF(3)](-), [CF(3)CO(2)MCF(3)](-) and [CF(3)CO(2)MF](-). These shed light on possible products and mechanisms for loss of "CF(2)CO(2)", namely, concerted or step-wise loss of CO(2) and CF(2) and a CF(2)CO(2) lactone pathway. The lactone pathway was found to be higher in energy in all cases. In addition, the possibility of forming [CF(3)MCF(3)](-) and [CF(3)MF](-), via decarboxylation is discussed. For the first time the novel fluoride complexes [FMF](-), M = Cu, Ag and Au have been experimentally observed. Finally, the decomposition reactions of [CF(3)CO(2)ML](-) (where L = CF(3) and CF(3)CO(2)) and [CH(3)CO(2)ML](-) (where L = CH(3) and CH(3)CO(2)) are compared.

A combination of multistage mass spectrometry experiments and DFT calculations were used to examine the synthesis and reactivity of dimethylaurate. Collision induced dissociation (CID) of [(CH(3)CO(2))(4)Au](-) proceeded via reductive elimination of acetylperoxide to yield the diacetate [CH(3)CO(2)AuO(2)CCH(3)](-), which in turn underwent sequential CID decarboxylation reactions to yield the organoaurates [CH(3)CO(2)AuCH(3)](-) and [CH(3)AuCH(3)](-). The unimolecular chemistry of the dimethylaurate proceeds via a combination of bond homolysis to yield the methyl aurate radical anion [CH(3)Au] (-) as well as formation of the gold dihydride [HAuH](-). DFT calculations reveal that the latter anion is formed via a 1,2-dyotropic rearrangement to yield the isomer [CH(3)CH(2)AuH](-), followed by a beta-hydride elimination reaction. Ion-molecule reactions of [CH(3)AuCH(3)](-) with methyl iodide did not yield any products even at relatively high concentrations of the neutral substrate and longer reaction times, indicating a reaction efficiency of less than 1 in 20 000 collisions. DFT calculations were carried out on two different potential energy surfaces (PES) for the reaction of [CH(3)AuCH(3)](-) with CH(3)I: (i) an S(N)2 mechanism proceeding via a side-on transition state; and (ii) a stepwise mechanism proceeding via oxidative addition followed by reductive elimination. Both pathways have significant endothermic barriers, consistent with the lack of C-C bond coupling products being formed in the experiments. Finally, the reactivity of [CH(3)AuCH(3)](-) is compared to the previously studied [CH(3)AgCH(3)](-) and [CH(3)CuCH(3)](-), as well as condensed phase studies on dimethylaurate salts.

The gas-phase fragmentation reactions of the group 11 organometallates [CH 3 were studied by density functional theory (DFT) calculations and, for those accessible in the gas phase, via collision-induced dissociation (CID) and selected deuterium labeling experiments. The mixed metallates [CH 3 MR]were found to fragment via a diverse set of pathways, including bond homolysis, bond heterolysis, and β-hydride elimination. A 1,2-dyotropic rearrangement was observed for R = Ph. DFT calculations suggest that the M-C bond energy, the availability of metal orbitals for π-bonding, and the nature of the ligand R group substituents are the main factors that control the observed reactivity. Comparisons with the known solution-phase reactivity of organocopper and organosilver species are made.

The homocuprates [MeCuMe]- and [EtCuEt]- were generated in the gas phase by double decarboxylation of the copper carboxylate centers [MeCO2CuO2CMe]- and [EtCO2CuO2CEt]-, respectively. The same strategy was explored for generating the heterocuprates [MeCuR]- from [MeCO2CuO2CR]- (R = Et, Pr, iPr, tBu, allyl, benzyl, Ph). The formation of these organocuprates was examined by multistage mass spectrometry experiments, including collision-induced dissociation and ion-molecule reactions, and theoretically by density functional theory. A number of side reactions were observed to be in competition with the second stage of decarboxylation, including loss of the anionic carboxylate ligand and loss of neutral alkene via beta-hydride transfer elimination. Interpretation of decarboxylation of the heterocarboxylates [MeCO2CuO2CR]- was more complex because of the possibility of decarboxylation occurring at either of the two different carboxylate ligands and giving rise to the possible isomers [MeCuO2CR]- or [MeCO2CuR]-. Ion-molecule reactions of the products of initial decarboxylation with allyl iodide resulted in C-C coupling to produce the ionic products [ICuO2CR]- or [MeCO2CuI]-, which provided insights into the relative population of the isomers, and indicated that the site of decarboxylation was dependent on R. For example, [MeCO2CuO2CtBu]- underwent decarboxylation at MeCO2- to give [MeCuO2CtBu]-, while [MeCO2CuO2CCH2Ph]- underwent decarboxylation at PhCH2CO2- to give [MeCO2CuCH2Ph]-. Each of the heterocuprates [MeCuR]- (R = Et, Pr, iPr, allyl, benzyl, Ph) could be generated by the double decarboxylation strategy. However, when R = tBu, intermediate [MeCuO2CtBu]- only underwent loss of tBuCO2-, a consequence of the steric bulk of tBu disfavoring decarboxylation and stabilizing the competing channel of carboxylate anion loss. Detailed DFT calculations were carried out on the potential energy surfaces for the first and second decarboxylation reactions of all homo- and heterocuprates, as well as possible competing reactions. These reveal that in all cases the first decarboxylation reaction is favored over loss of the carboxylate ligand. In contrast, other reactions such as carboxylate ligand loss and beta-hydride transfer become more competitive with the second decarboxylation reaction.

CO2 activation mediated by [LTiH](+) (L=Cp2 , O) is observed in the gas phase at room temperature using electrospray-ionization mass spectrometry, and reaction details are derived from traveling wave ion-mobility mass spectrometry. Wheresas oxygen-atom transfer prevails in the reaction of the oxide complex [OTiH](+) with CO2 , generating [OTi(OH)](+) under the elimination of CO, insertion of CO2 into the metal-hydrogen bond of the cyclopentadienyl complex, [Cp2 TiH](+) , gives rise to the formate complex [Cp2 Ti(O2 CH)](+) . DFT-based methods were employed to understand how the ligand controls the observed variation in reactivity toward CO2 . Insertion of CO2 into the Ti-H bond constitutes the initial step for the reaction of both [Cp2 TiH](+) and [OTiH](+) , thus generating formate complexes as intermediates. In contrast to [Cp2 Ti(O2 CH)](+) which is kinetically stable, facile decarbonylation of [OTi(O2 CH)](+) results in the hydroxo complex [OTi(OH)](+) . The longer lifetime of [Cp2 Ti(O2 CH)](+) allows for secondary reactions with background water, as a result of which, [Cp2 Ti(OH)](+) is formed. Further, computational studies reveal a good linear correlation between the hydride affinity of [LTi](2+) and the barrier for CO2 insertion into various [LTiH](+) complexes. Understanding the intrinsic ligand effects may provide insight into the selective activation of CO2 .