Through the Looking Glass: Using the Lens of [SNS]-Pincer Ligands to Examine First-Row Metal Bifunctional Catalysts.

ConspectusHomogeneous catalysis is at the forefront of global efforts to innovate the synthesis of fine chemicals and achieve carbon-neutrality in energy applications. For decades, the push toward sustainable catalysis has focused on the development of first-row transition metal catalysts to supplant widespread use of precious metals. Metal-ligand cooperativity is an effective strategy to yield high-performing first-row metal molecular catalysts. Despite remarkable progress, state of the art catalysts often employ phosphorus-based ligands which are air-sensitive, potentially toxic, and on occasion offset the cost-savings of the metal. Thus, the development of simple and economical ligands composed of biomimetic donors should be a key focus that cannot be overlooked in the pursuit of sustainable catalyst candidates. This is an Account of our group's efforts to develop first-row transition metal complexes which use [SNS]-pincer ligands for bifunctional catalysis. We have synthesized two potentially tridentate ligands, one bearing an amido and two thioether donors [(SMeNSMe), L1] and one which includes thiolate, imine, and thioether donors [(SNSMe), L2], and used them as platforms upon which to explore the reaction pathways of first-row metals. The [SNS] ligand, L1, leads to formation of high-spin paramagnetic metal complexes of the type M(L1)2 in which the 6-membered ring thioether donor is hemilabile (M = Mn, Fe, Co). This allows Mn(L1)2 to function as a carbonyl hydroboration catalyst that operates by a novel hydride-free, inner-sphere reaction pathway. Exploring the reactivity of L2 with Fe and Ni revealed unique coordination chemistry and a variety of mono-, di-, tri-, and tetranuclear complexes enabled by bridging thiolates. Further studies showed L2 undergoes selective Caryl-S bond cleavage upon coordination to a metal with electron-rich phosphine donors, yielding a new (CNS)2- pincer ligand. The analogous reaction with L1 afforded a new (CNSMe)- pincer ligand via both Caryl-S and benzylic C-H bond cleavage. In an attempt to prepare Fe(L2)2, we obtained instead an Fe(N2S3) complex in which imine C-C bond formation affords a potentially hexadentate redox-active ligand. The Fe(N2S3) complex is a selective catalyst for hydroboration of aldehydes and appears to operate through a complicated mechanism. In contrast, a mechanistic study of Mn(L2)(CO)3-photocatalyzed dihydroboration of nitriles indicated that both the flexibility of the κ3-SNSMe ligand (fac- vs mer-coordination) and ability of Mn to undergo a spin-state change are required to access low energy barriers for this transformation. To effectively compare the reactivity of the thiolate vs amido donor, we prepared two Cu complexes, Cu(L1)(IPr) and Cu(L2)(IPr) [IPr = 1,3-bis(2,6-diisopropylphenyl)imidazol-2-ylidene], showing that, while both served as carbonyl hydroboration catalysts, only the amido complex was an effective catalyst for carbonyl hydrosilylation. In addition, complexes of the type Zn(L1)2, Zn(L2)2, and Zn(L1)(L2), were also effective for catalytic carbonyl hydroboration. While Zn(L1)(L2) was most active, catalyst speciation studies showed that each undergoes bifunctional catalyst activation to form a Zn bis(alkoxide) catalyst. Overall, our observations using [SNS] ligands with first-row transition metals show how the absence of traditional phosphine donors leads to different fundamental reactivity. Furthermore, this Account demonstrates the gap of knowledge which exists in understanding the reactivity of sulfur-based ligands to promote more widespread adoption of sustainable ligands.

Spin-state crossover in photo-catalyzed nitrile dihydroboration via Mn-thiolate cooperation.

The role of S-donors in ligand-assisted catalysis using first-row metals has not been broadly investigated. Herein is described a combined experimental and computational mechanistic study of the dihydroboration of nitriles with pinacolborane (HBpin) catalyzed by the Mn(i) complex, Mn(κ3-SMeNS)(CO)3, that features thioether, imine, and thiolate donors. Mechanistic studies revealed that catalysis requires the presence of UV light to enter and remain in the catalytic cycle and evidence is presented for loss of two CO ligands. Stoichiometric reactions showed that HBpin reduces the imine N[double bond, length as m-dash]C of the ligand backbone in the absence of nitrile, forming an inactive off-cycle by-product. DFT calculations showed that the bifunctional thiolate donor, coordinative flexibility of the SMeNS ligand, and access to an open-shell intermediate are all crucuial to accessing low-energy intermediates during catalysis.

Correlating Thermodynamic and Kinetic Hydricities of Rhenium Hydrides.

The kinetics of hydride transfer from Re(Rbpy)(CO)3H (bpy = 4,4'-R-2,2'-bipyridine; R = OMe, tBu, Me, H, Br, COOMe, CF3) to CO2 and seven different cationic N-heterocycles were determined. Additionally, the thermodynamic hydricities of complexes of the type Re(Rbpy)(CO)3H were established primarily using computational methods. Linear free-energy relationships (LFERs) derived by correlating thermodynamic and kinetic hydricities indicate that, in general, the rate of hydride transfer increases as the thermodynamic driving force for the reaction increases. Kinetic isotope effects range from inverse for hydride transfer reactions with a small driving force to normal for reactions with a large driving force. Hammett analysis indicates that hydride transfer reactions with greater thermodynamic driving force are less sensitive to changes in the electronic properties of the metal hydride, presumably because there is less buildup of charge in the increasingly early transition state. Bronsted α values were obtained for a range of hydride transfer reactions and along with DFT calculations suggest the reactions are concerted, which enables the use of Marcus theory to analyze hydride transfer reactions involving transition metal hydrides. It is notable, however, that even slight perturbations in the steric properties of the Re hydride or the hydride acceptor result in large deviations in the predicted rate of hydride transfer based on thermodynamic driving forces. This indicates that thermodynamic considerations alone cannot be used to predict the rate of hydride transfer, which has implications for catalyst design.

Same ligand, three first-row metals: comparing M-amido bifunctional reactivity (Mn, Fe, Co).

The bifunctional reactivity of three metal SNS (bis)amido complexes was computationally assessed by comparing the nucleophilicity of the M-Namido donor (Mn, Fe, Co). Hirshfeld charges identified the Mn-Namido donor as most nucleophilic and Fe as most electrophilic metal. Reaction energy profiles of a model bifunctional H2 activation showed Mn with the lowest reaction barrier (17 kcal mol-1), followed by Fe and Co (21 and 29 kcal mol-1).

Strategies and mechanisms of metal-ligand cooperativity in first-row transition metal complex catalysts.

The use of metal-ligand cooperation (MLC) by transition metal bifunctional catalysts has emerged at the forefront of homogeneous catalysis science. Specially designed ligands can serve a Lewis base or Lewis acid function, as an aromatization/dearomatization shuttle, or as an electron reservoir with reversible redox activity. This review encapsulates advances that have been made in this field over the last ten years, focusing exclusively on first-row transition metals, and highlighting significant contributions to mechanistic understanding.

Cu(i)-SNS complexes for outer-sphere hydroboration and hydrosilylation of carbonyls.

Two new NHC-Cu(i)-[κ2-SNS] complexes were synthesized to directly compare the bifunctional catalytic activity of a hard amido vs. a soft thiolate donor. The Cu thiolate complex catalyzed ketone hydroboration but not hydrosilylation, while the Cu amido complex is a high-performing outer-sphere carbonyl reduction catalyst using boranes and silanes.

Nickel-Catalyzed C-H Silylation of Arenes with Vinylsilanes: Rapid and Reversible ß-Si Elimination.

The reaction of C6F5H and H2C═CHSiMe3 with catalytic [iPr2Im]Ni(η2-H2C═CHSiMe3)2 (1b) exclusively forms the C-H silylation product C6F5SiMe3 with ethylene as a byproduct ([iPr2Im] = 1,3-di(isopropyl)imidazole-2-ylidene). Catalytic C-H bond silylation is facile with partially fluorinated aromatic substrates containing two ortho fluorine substituents adjacent to the C-H bond and 1,2,3,4-tetrafluorobenzene. Less fluorinated substrates react slower. Under the same reaction conditions, catalytic [IPr]Ni(η2-H2C═CHSiMe3)2 (1a) ([IPr] = 1,3-bis[2,6-diisopropylphenyl]-1,3-dihydro-2H-imidazol-2-ylidene) provided only the alkene hydroarylation product C6F5CH2CH2SiMe3. Mechanistic studies reveal that the C-H activation and ß-Si elimination steps are reversible under catalytic conditions with both catalysts 1a and 1b. With catalytic 1a, reversible ethylene loss after ß-Si elimination was also observed despite its inability to catalyze C-H silylation; the reductive elimination step to form the silylation product is much slower than reductive elimination to form the alkene hydroarylation product. Reversible ethylene loss was not observed with 1b, which suggests that the rate-limiting step in the reaction is neither C-H activation nor ß-Si elimination but either ethylene loss or reductive elimination of cis-disposed aryl and SiMe3 moieties.