This perspective article provides an overview of the advancements in the field of non-precious metal complexes featuring anionic PCP pincer ligands with the inclusion of aliphatic systems. It covers research from the beginning in 1976 until late 2015 and provides a summary of key developments in this area, which is, to date, limited to the metals nickel, cobalt, iron, and molybdenum. While the research in nickel PCP complexes is already quite extensive, the chemistry of cobalt, iron, and molybdenum PCP complexes is comparatively sparse. With other non-precious metals such as copper, manganese, chromium or vanadium no PCP complexes are known as yet. In the case of nickel PCP complexes already many catalytic applications such as Suzuki-Miyaura coupling, C-S cross coupling, Kharasch and Michael additions, hydrosilylation of aldehydes and ketones, cyanomethylation of aldehydes, and hydroamination of nitriles were reported. While iron PCP complexes were found to be active catalysts for the hydrosilylation of aldehydes and ketones as well as the dehydrogenation of ammonia-borane, cobalt PCP complexes were not applied to any catalytic reactions. Surprisingly, only one molybdenum PCP complex is reported, which was capable of cleaving dinitrogen to give a nitride complex. This perspective underlines that the combination of cheap and abundant metals such as nickel, cobalt, and iron with PCP pincer ligands may result in the development of novel, versatile, and efficient catalysts for atom-efficient catalytic reactions.

Anhydrous CoCl2 or [NiCl2(DME)] reacts with the ligand PCPMe-iPr (1) in the presence of nBuLi to afford the 15e and 16e square planar complexes [Co(PCPMe-iPr)Cl] (2) and [Ni(PCPMe-iPr)Cl] (3), respectively. Complex 2 is a paramagnetic d7 low-spin complex, which is a useful precursor for a series of Co(I), Co(II), and Co(III) PCP complexes. Complex 2 reacts readily with CO and pyridine to afford the five-coordinate square-pyramidal 17e complexes [Co(PCPMe-iPr)(CO)Cl] (4) and [Co(PCPMe-iPr)(py)Cl] (5), respectively, while in the presence of Ag+ and CO the cationic complex [Co(PCPMe-iPr)(CO)2]+ (6) is afforded. The effective magnetic moments μeff of all Co(II) complexes were derived from the temperature dependence of the inverse molar magnetic susceptibility by SQUID measurements and are in the range 1.9 to 2.4 μB. This is consistent with a d7 low-spin configuration with some degree of spin–orbit coupling. Oxidation of 2 with CuCl2 affords the paramagnetic Co(III) PCP complex [Co(PCPMe-iPr)Cl2] (7), while the synthesis of the diamagnetic Co(I) complex [Co(PCPMe-iPr)(CO)2] (8) was achieved by stirring 2 in toluene with KC8 in the presence of CO. Finally, the cationic 16e Ni(II) PCP complex [Ni(PCPMe-iPr)(CO)]+ (10) was obtained by reacting complex 3 with 1 equiv of AgSbF6 in the presence of CO. The reactivity of CO addition to Co(I), Co(II), and Ni(II) PCP square planar complexes of the type [M(PCPMe-iPr)(CO)]n (n = +1, 0) was investigated by DFT calculations, showing that formation of the Co species, 6 and 8, is thermodynamically favorable, while Ni(II) maintains the 16e configuration since CO addition is unfavorable in this case. X-ray structures of most complexes are provided and discussed. A structural feature of interest is that the apical CO ligand in 4 deviates significantly from linearity, with a Co–C–O angle of 170.0(1)°. The DFT-calculated value is 172°, clearly showing that this is not a packing but an electronic effect.

The 15e square-planar complexes [Co(PCPMe-iPr)Cl] (2a) and [Co(PCP-tBu)Cl] (2b), respectively, react readily with NaBH4 to afford complexes [Co(PCPMe-iPr)(η2-BH4)] (4a) and [Co(PCP-tBu)(η2-BH4)] (4b) in high yields, as confirmed by IR spectroscopy, X-ray crystallography, and elemental analysis. The borohydride ligand is symmetrically bound to the cobalt center in η2-fashion. These compounds are paramagnetic with effective magnetic moments of 2.0(1) and 2.1(1) μB consistent with a d7 low-spin system corresponding to one unpaired electron. None of these complexes reacted with CO2 to give formate complexes. For structural and reactivity comparisons, we prepared the analogous Ni(II) borohydride complex [Ni(PCPMe-iPr)(η2-BH4)] (5) via two different synthetic routes. One utilizes [Ni(PCPMe-iPr)Cl] (3) and NaBH4, the second one makes use of the hydride complex [Ni(PCPMe-iPr)H] (6) and BH3·THF. In both cases, 5 is obtained in high yields. In contrast to 4a and 4b, the borohydride ligand is asymmetrically bound to the nickel center but still in an η2-mode. [Ni(PCPMe-iPr)(η2-BH4)] (5) loses readily BH3 at elevated temperatures in the presence of NEt3 to form 6. Complexes 5 and 6 are both diamagnetic and were characterized by a combination of 1H, 13C{1H}, and 31P{1H} NMR, IR spectroscopy, and elemental analysis. Additionally, the structure of these compounds was established by X-ray crystallography. Complexes 5 and 6 react with CO2 to give the formate complex [Ni(PCPMe-iPr)(OC(C=O)H] (7). The extrusion of BH3 from [Co(PCPMe-iPr)(η2-BH4)] (4a) and [Ni(PCPMe-iPr)(η2-BH4)] (5) with the aid of NH3 to yield the respective hydride complexes [Co(PCPMe-iPr)H] and [Ni(PCPMe-iPr)H] (6) and BH3NH3 was investigated by DFT calculations showing that formation of the Ni hydride is thermodynamically favorable, whereas the formation of the Co(II) hydride, in agreement with the experiment, is unfavorable. The electronic structures and the bonding of the borohydride ligand in [Co(PCPMe-iPr)(η2-BH4)] (4a) and [Ni(PCPMe-iPr)(η2-BH4)] (5) were established by DFT computations.

The synthesis and reactivity of a Co(I) pincer complex [Co(ϰ(3) P,CH,P-P(CH)P(NMe) -iPr)(CO)2](+) featuring an η(2)-C(aryl)-H agostic bond is described. This complex was obtained by protonation of the Co(I) complex [Co(PCP(NMe) -iPr)(CO)2]. The Co(III) hydride complex [Co(PCP(NMe) -iPr)(CNtBu)2(H)](+) was obtained upon protonation of [Co(PCP(NMe) -iPr)(CNtBu)2]. Three ways to cleave the agostic C-H bond are presented. First, owing to the acidity of the agostic proton, treatment with pyridine results in facile deprotonation (C-H bond cleavage) and reformation of [Co(PCP(NMe) -iPr)(CO)2]. Second, C-H bond cleavage is achieved upon exposure of [Co(ϰ(3)P,CH,P-P(CH)P(NMe) -iPr)(CO)2](+) to oxygen or TEMPO to yield the paramagnetic Co(II) PCP complex [Co(PCP(NMe) -iPr)(CO)2](+). Finally, replacement of one CO ligand in [Co(ϰ(3) P,CH,P-P(CH)P(NMe) -iPr)(CO)2](+) by CNtBu promotes the rapid oxidative addition of the agostic η(2) -C(aryl)-H bond to give two isomeric hydride complexes of the type [Co(PCP(NMe) -iPr)(CNtBu)(CO)(H)](+).

Iridium complexes bearing PCP-type pincer ligands are the most effective catalysts reported to date for the low-temperature (≤ca. 200 °C) dehydrogenation of alkanes. To investigate the activity of formally isoelectronic ruthenium complexes, we have synthesized the neutral 2,7-di-tert-butyl-4,5-bis­(diisopropyl­phosphino)-9,9-dimethyl­thioxanthene (iPrxanPSP) pincer ligand and several Ru complexes thereof. The (iPrxanPSP)Ru complexes catalyze alkane transfer dehydrogenation of the benchmark cyclooctane/t-butylethylene (COA/TBE) couple with turnover frequencies up to ca. 1 s–1 at 150 °C and 0.2 s–1 at 120 °C, the highest rates for alkane dehydrogenation ever reported at such temperatures. Dehydrogenation of n-octane, however, is much less effective. A combination of experiment and DFT calculations allow us to explain why (iPrxanPSP)Ru is more effective than (iPrPCP)Ir for dehydrogenation of COA, while the reverse is true for dehydrogenation of n-alkanes. Considering only in-cycle species and simple olefin complexes, the (iPrxanPSP)Ru fragment is calculated to be much more active than (iPrPCP)Ir for dehydrogenation of both COA and n-alkanes. However, the resting state in the (iPrxanPSP)­Ru-catalyzed transfer dehydrogenation of n-alkane is a very stable linear-allyl hydride complex, whereas the corresponding cyclooctenyl hydride is much less stable.