Publications

A procedure for the general synthesis of nickel(II)–aryl bromide oxidative addition complexes without aryl substitution at the 2,6-positions was achieved by inverse addition of the requisite bipyridyl nickel(0)–1,5-cyclooctadiene (COD) complex to the aryl bromide, yielding 2,2’-bipyridyl (bpy) and 4,4’-di-tert-butyl-2,2’-bipyridyl (dtbbpy) supported nickel(II)–aryl bromide compounds in 38–93% yield. During reductive C(sp²)–C(sp³) cross-electrophile cross-coupling in DMA at 60 °C, UV-vis absorbance spectroscopy determined (dtbbpy)nickel(II)–aryl bromide complexes to be the predominant catalyst resting states. The observed catalyst resting states were implicated in forming the organic products of C(sp²)–C(sp³) cross-coupling and of C(sp²)–C(sp²) homocoupling, homocoupling of which was more predominant with electron-donating aryl substituents. Dissolution of (dtbbpy)nickel(II)–aryl bromide compounds in 0.5 mM DMA in the absence of excess aryl bromide established decay of the compounds within minutes to form aryl homodimer and dimeric (dtbbpy)nickel(I) bromide as the exclusive products, in which compounds with the fastest rates of decay were those bearing electron-donating substituents (OMe, t-Bu), correlated with more nucleophilic character of the aryl and weaker Ni–Aryl bond strength. Determination of concentration-dependence for homodimerization of (dtbbpy)NiBr(4-F-C₆H₄) and (bpy)NiBr(4-F-C₆H₄) compounds measured by UV-visible absorbance spectroscopy exhibited 1.5-order and 1.3-order dependence, respectively, rationalized as exhibiting concurrent bimolecular and unimolecular mechanisms, namely bimolecular transmetalation/reductive elimination, and unimolecular autoreduction, the latter of which was informed by rapid comproportionation of nickel(II)–aryl with nickel(0) yielding nickel(I)–aryl intermediates detected by X-band EPR spectroscopy at 77 K. These studies informed kinetically relevant aryl homodimerization as the predominant mechanism for low-valent nickel formation, initiating the nickel(I/II/III) alkyl radical chain mechanism by homogeneous turnover of the (dtbbpy)nickel(II)–aryl bromide catalyst.

Synthesized from benzothiazolones, 2-alkoxy benzothiazoles served as competent C(sp2)–O electrophiles in (bisphos-phine)cobalt(II)-catalyzed Kumada arylation, forming 2-arylbenzothiazole products. Employing catalytic (5 mol%) bis(diphenylphosphino)propane (dppp)–cobalt(II) dibromide precatalyst, the Kumada arylation of heteroaryl ethers was gen-eral towards 2-alkoxybenzoxazole, 2-alkoxyquinoline, and 2-alkoxythiazole substrates (17–65% yield). Investigation of cata-lytically relevant precatalysts indicated kinetic incompetence of cobalt(I) and cobalt(0) complexes, supporting proposal of a redox-neutral cobalt(II)-catalyzed reaction involving rate-determining heteroaryl ether substitution.

Cross-coupling of alkyl electrophiles by first row transition metal catalysts represents a productive approach of incorporating C(sp³) fragments into pharmaceutically relevant organic molecules. This Account highlights the development of cobalt-catalyzed Negishi C(sp²)–C(sp³) cross-coupling reactions of redox active electrophiles, particularly thiol-derived unactivated alkyl(pyridyl)sulfones. Informed by catalyst characterization and mechanistic investigation, this account proposes key principles of substrate and catalyst design, which may enable development of future methodologies and robust cobalt catalysts for medicinal and process chemistry applications.

A well-defined, single-component precatalyst N,N,N′,N′-tetramethylethylenediamine (TMEDA)cobalt(II) dibromide catalyzed Negishi arylation reactions of alkyl bromide, N-hydroxyphthalimide ester, and (hetero)aryl halide electrophiles, exhibiting both C(sp²)─C(sp²) and C(sp²)─C(sp³) bond-formation. Organometallic reactions of relevant weak field ligand cobalt(II) dihalide complexes with arylzinc reagents demonstrated monoarylation, yielding a series of isolable high spin (S = 3/2) cobalt(II)–monoaryl bromide compounds supported by TMEDA, 3,5-lutidine, or bis(oxazoline) (BOX) ligands. Relevant to C(sp²)–C(sp²) cross-coupling, cobalt(II)–monoaryl bromide complexes reacted with arylzinc nucleophiles to yield TMEDA-supported and bpy-supported low-spin (S = 1/2) cobalt(II)–diaryl complexes, which underwent reductive elimination in acetonitrile to provide direct evidence for cobalt(0/II) cycles. In C(sp²)─C(sp³) Negishi cross-coupling reactions, the isolated TMEDA-supported and BOX-supported cobalt(II)–monoaryl complexes were determined to be catalyst resting states by ¹H and ¹⁹F NMR spectroscopies in acetonitrile-d₃. Stoichiometric reactions demonstrated high spin (TMEDA)cobalt(II)–monoaryl to be competent for alkyl bromine atom abstraction and for alkyl radical capture, yielding C(sp²)─C(sp³) product by reductive elimination at (TMEDA)cobalt(III). These data demonstrated the versatile reactivity of high-spin (S = 3/2) cobalt(II)–monoaryl and low-spin cobalt(II)–diaryl (S = 1/2) complexes, engaging productive closed-shell cobalt(0/II) or open-shell cobalt(I/III) cross-coupling mechanisms depending on the aryl or alkyl electrophile substrate.

The formation of manganese–oxo catalysts involved in C(sp³)–H bond oxidation was explored in the targeted synthesis of (salen/salophen)manganese complexes that varied axial ligand identity and varied oxidation state of the manganese center. Isolated compounds included dinuclear (salen/salophen)manganese(III)–μ-hydroxo and trinuclear (salen)manganese(IV/III/IV)–μ-oxo, the latter of which formed by oxidation with catalytically relevant oxidant iodosylbenzene. The X-ray structure of trinuclear complex (salen)manganese(IV/III/IV)–μ-oxo indicated a Mn(IV)–O–Mn(III)–O–Mn(IV) motif, with nearly linear Mn–O–Mn angles of 166.19(12)° and 172.47(15)°, Mn(IV)–O bond lengths of 1.948(2) and 1.998(2) Å, and Mn(III)–O bond lengths of 2.102(2) and 2.118(2) Å. All well-defined (salen/salophen)manganese hydroxo and oxo compounds served as precatalysts for oxidation of C(sp³)–H substrates 9,10-dihydroanthracene (>99% conversion), fluorene (52–70% conversion), and phenylcyclohexane (with lower 18–23% conversion), albeit with lower rate of activity for the isolated trinuclear μ-oxo compound, allowing its assignment as an off-cycle catalyst aggregate. These data supported the proposal of a manganese(III/V) cycle for C(sp³)–H oxidation, which involved monomerization of the dinuclear (salen)manganese(III)-μ-hydroxo catalyst prior to rate-determining C(sp³)–H activation.

A cobalt(II) catalyst supported by the ligand 2-(diphenylphosphino)phenol (P,O) was developed for the C(sp²)–C(sp³) Negishi arylation of alkyl(pyridyl)sulfones, which are bench-stable, nonorganohalide C(sp³)–S electrophiles. Employing the catalyst generated in situ from 5 mol % P,O ligand and 5 mol % cobalt(II) bromide, a variety of (hetero)aryl C(sp²)–C(sp³) products were synthesized derived from primary and secondary alkyl sulfones, including difluoromethylation using 2-((difluoromethyl)sulfonyl)pyridine, and cross-coupling of sulfone derived from the thiol-containing ACE inhibitor captopril. Freeze-quench X-band EPR spectroscopy of a catalytic reaction established the catalyst resting state as low-spin (S = 1/2), square-pyramidal (P,O)cobalt(II)–aryl, a rare example of a cobalt(II)–aryl complex detected during a cross-coupling reaction. These data informed the cobalt(II/III/I/0) catalytic cycle involving alkyl radical capture at the (P,O)cobalt(II)–aryl catalyst resting state, enabling selective formation of the C(sp²)–C(sp³) product.

To explore arylboron transmetalation at manganese(I), reactions of 4-fluorophenylborates with pentacarbonylmanganese(I) hexafluorophosphate cation (CO)₅Mn(MeCN)(PF₆) were evaluated for the formation of 4-fluorophenylmanganese(I) pentacarbonyl (CO)₅Mn(4-F–C₆H₄). The optimal reagent was neopentylglycol 4-fluorophenylboronic ester activated with n-butyllithium, which reacted with (CO)₅Mn(MeCN)(PF₆) to give (CO)₅Mn(4-F–C₆H₄) in 58% yield. These conditions were extrapolated to reactions involving other neopentylglycol esters to yield a scope of seven (CO)₅Mn(I)–aryls with varied substitutions on the aryl ring. The bench-stable, diamagnetic (CO)₅Mn(I)–aryl compounds were purified by flash column chromatography on silica and were characterized by IR spectroscopy and by ¹H, ¹³C, ¹⁹F, and ⁵⁵Mn NMR spectroscopies, with solid-state molecular structures verified by X-ray crystallography. Finally, (CO)₅Mn(4-F–C₆H₄) was explored as a synthetic arylating reagent in reactions with various electrophiles and nucleophiles, with organic products including those from aryl C(sp²)–X and aroyl C(sp²)CO–X bond formation.

Iron-catalyzed Kumada cross-coupling was explored for C(sp³)–O arylation of activated cyclohexanol derivatives, revealing cyclohexyl tosylate as a competent substrate. Investigation of the effect of bromide additives indicated that cyclohexyl tosylate underwent bromide substitution in a reaction with MgBr₂─the salt byproduct generated during cross-coupling. The single-turnover reaction of 1 equivalent of cyclohexyl tosylate with 1 equivalent of 4-fluorophenylmagnesium bromide in the presence of bis(diphenylphosphino)benzene (dppbz)iron(II) dichloride showed no conversion to arylated product, indicating that cyclohexyl tosylate was not activated by catalytically relevant iron intermediates and that tosylate-to-bromide substitution was necessary for productive cross-coupling. A two-step method was developed, which involved in situ bromide substitution of alkyl tosylate substrates using MgBr₂·OEt₂, followed by (dppbz)iron(II)-catalyzed Kumada arylation, which was used to convert 16 C(sp³)–OTs substrates to the corresponding C(sp²)–C(sp³) arylated products in 31–84% yield.