Dinuclear Nickel(I) and Palladium(I) Complexes for Highly Active Transformations of Organic Compounds

molecules

Review Dinuclear Nickel(I) and Palladium(I) Complexes for Highly Active Transformations of Organic Compounds

Takahiro Inatomi ID , Yuji Koga and Kouki Matsubara * ID

Fukuoka University, 8-19-1 Nanakuma, Fukuoka 814-0180, Japan; [email protected] (T.I.); [email protected] (Y.K.) * Correspondence: [email protected]; Tel.: +81-92-871-6631

Received: 7 December 2017; Accepted: 9 January 2018; Published: 11 January 2018

Abstract: In typical catalytic organic transformations, transition metals in catalytically active complexes are present in their most stable valence states, such as palladium(0) and (II). However, some dimeric monovalent metal complexes can be stabilized by auxiliary ligands to form diamagnetic compounds with metal–metal bonding interactions. These diamagnetic compounds can act as catalysts while retaining their dimeric forms, split homolytically or heterolytically into monomeric forms, which usually have high activity, or in contrast, become completely deactivated as catalysts. Recently, many studies using group 10 metal complexes containing nickel and palladium have demonstrated that under speciﬁc conditions, the active forms of these catalyst precursors are not mononuclear zerovalent complexes, but instead dinuclear monovalent metal complexes. In this mini-review, we have surveyed the preparation, reactivity, and the catalytic processes of dinuclear nickel(I) and palladium(I) complexes, focusing on mechanistic insights into the precatalyst activation systems and the structure and behavior of nickel and palladium intermediates.

Keywords: monovalent nickel; monovalent palladium; dinuclear complexes; catalytic process; DFT calculations

1. Introduction Interest in the development of catalytic organic transformations using well-defined palladium and nickel complexes as active catalysts continues to increase [1–10]. Catalytic transformations using these catalysts, such as addition, polymerization, allylic substitution, cross coupling, and C–H substitution reactions, significantly contribute to the production of useful organic chemicals in industry. Interestingly, several dinuclear Pd(I) and Ni(I) complexes have been reported to demonstrate notably higher activity as precatalysts compared to typical monomeric zerovalent or divalent metal complexes [11–16]. Detailed mechanistic studies revealed that dinuclear or mononuclear active catalysts are generated from the corresponding dinuclear precatalysts to achieve highly efficient and/or selective catalytic reactions. Although cooperative activation of organic compounds by bimetallic complexes has been reported in organometallic reactions of cluster compounds [17–20], the mechanism by which dinickel(I) or dipalladium(I) complexes act in catalytic transformations is poorly understood. For example, the in situ formation of dinuclear complexes has been reported to lead to the deactivation of catalytic pathways of highly active mononuclear catalysts. Several other reports proposed that heterolytic or homolytic cleavage of the metal–metal bond of dinuclear Pd(I) or Ni(I) complexes could produce active metal species in situ, such as highly unsaturated “mono-ligated metal complexes”. However, the chemistries of dinuclear Ni(I) and Pd(I) complexes in catalysis have not been classified and compared with each other, although the chemistry of such nickel and palladium complexes has been summarized individually [11–16]. Therefore, the survey and comparison of the reactivity and

Molecules 2018, 23, 140; doi:10.3390/molecules23010140 www.mdpi.com/journal/molecules Molecules 2018, 23, 140 2 of 21

Molecules 2018, 23, 140 2 of 21 catalytic performance of dinuclear nickel(I) and palladium(I) complexes, as well as their generation processesas well as their and structures,generation processes should be and useful structures, and informative. should be useful In this and review, informative. the structural In this review, features the of structuraldinuclear features Pd(I) and of Ni(I)dinuclear complexes Pd(I) and are Ni(I) discussed, complexes and are their discussed, catalytic and applications their catalytic and mechanisticapplications anddetails mechanistic are surveyed details in are order surveyed to uncover in order the to key uncover points the of their key points chemistry. of their chemistry.

2. Preparation, Structure, and Properties of Dinuclear Nickel(I) and Palladium(I) Complexes Numerous examples of dinuclear nickel(I) and palladium(I)palladium(I) complexes have been reported in the past decades [11–16]. [11–16]. Most of these dinuclear complexes have b beeneen efficiently efﬁciently synthesized using several established methods. One One of of the most acce accessiblessible routes to these compounds is the treatment of nickel(II) or palladium(II)palladium(II) complexes with the corresponding zerovalent metal complex in the presence of bridging bidentate or tridentate ligandsligands (Method A in Chart1 1).). OtherOther methodsmethods toto obtainobtain such dinuclear dinuclear complexes complexes include include 1e 1e reduction reduction of ofmononuclear mononuclear divalent divalent complexes complexes (Method (Method B), the B), oxidationthe oxidation of zerovalentof zerovalent complexes complexes (Method (Method C), C), and and the the oxidative oxidative addition addition to to zerovalent zerovalent complexes complexes to form unsaturated divalent complexes, which are stabilized by forming a metal–metalmetal–metal bonding interaction with with a a second second zerovalent zerovalent complex complex (Metho (Methodd D). D). Rare Rare examples examples of ofthe the photoirradiation photoirradiation of alkylmetalof alkylmetal complexes, complexes, in inwhich which the the carbon–met carbon–metalal bond bond was was homolytically homolytically broken broken and and smoothly dimerized from from the the unstable unstable mono monomericmeric metal metal radical radical to form to form a me atal–metal metal–metal bond bond (Method (Method E), have E), alsohave been also shown.been shown. With multiple With multiple routes routesto access to su accessch complexes, such complexes, the use theof dinuclear use of dinuclear Pd(I) or Pd(I)Ni(I) complexesor Ni(I) complexes as catalyst as precursors catalyst precursors is increasingly is increasingly reported, because reported, these because compounds these compounds show remarkable show activitiesremarkable with activities respect with to catalytic respect transformations. to catalytic transformations.

ChartChart 1. 1. VariousVarious methods methods (Methods (Methods A–E) A–E) to to access access dinuclear dinuclear monovalent monovalent complexes complexes of of Ni Ni and and Pd. Pd.

In Figure 1, representative examples of dinuclear palladium(I) complexes are shown. In 1996, In Figure1, representative examples of dinuclear palladium(I) complexes are shown. In 1996, Mingos et al. reported the first example of catalytically active monovalent dipalladium complexes, Mingos et al. reported the ﬁrst example of catalytically active monovalent dipalladium complexes, 1a. The structure of these complexes was determined by X-ray crystallography, which showed the 1a. The structure of these complexes was determined by X-ray crystallography, which showed the existence of a Pd(I)–Pd(I) bonding interaction with bridging halogen ligands [21]. The stability of the existence of a Pd(I)–Pd(I) bonding interaction with bridging halogen ligands [21]. The stability dinuclear Pd(I) framework in the presence of only halide bridging ligands was discussed recently by of the dinuclear Pd(I) framework in the presence of only halide bridging ligands was discussed Schoenebeck et al. [22]. When tricyclohexylphosphine was used as the ancillary ligand, the halide-bridged recently by Schoenebeck et al. [22]. When tricyclohexylphosphine was used as the ancillary dimer did not form. Instead, the use of bis(t-butyl)isopropylphosphine, whose steric bulk is ligand, the halide-bridged dimer did not form. Instead, the use of bis(t-butyl)isopropylphosphine, intermediate between that of P(t-Bu)3 and PCy3, yielded the stable dimeric complex 1c, which was whose steric bulk is intermediate between that of P(t-Bu) and PCy , yielded the stable dimeric similar to Mingos’s dimer. 3 3 complex 1c, which was similar to Mingos’s dimer. In the presence of bridging ligands such as monomethylene-bridged bisphosphine, the dinuclear In the presence of bridging ligands such as monomethylene-bridged bisphosphine, the dinuclear Pd(I) complexes 6 were synthesized and characterized decades earlier than the other complexes, Pd(I) complexes 6 were synthesized and characterized decades earlier than the other complexes, although only reactions with small molecules were investigated [19,23]. Kurosawa and co-workers although only reactions with small molecules were investigated [19,23]. Kurosawa and co-workers developed the first example of a series of complexes 3 in which μ:η3-allenyl or propargyl ligands developed the ﬁrst example of a series of complexes 3 in which µ:η3-allenyl or propargyl ligands bridged the Pd(I)–Pd(I) bond, showing that dipalladium(I) structure was also stabilized by bridging bridged the Pd(I)–Pd(I) bond, showing that dipalladium(I) structure was also stabilized by bridging π-ligands. This bond was stable even when smaller phosphine ligands were used instead of P(tBu)3 [24]. π-ligands. This bond was stable even when smaller phosphine ligands were used instead of P(tBu) [24]. The complexes were formed by the reaction of Pd(II) allenyl or propargyl complexes with a Pd(0)3 The complexes were formed by the reaction of Pd(II) allenyl or propargyl complexes with a Pd(0) source. Murahashi and Kurosawa et al. reported a good starting complex for the preparation of source. Murahashi and Kurosawa et al. reported a good starting complex for the preparation of various cationic dinuclear Pd(I) complexes via ligand exchange reaction, [Pd2(NCCH3)6]2+ (2) [25]. For example, this complex was used as a precursor for the bis-dmpm bridged dipalladium(I) dication, [Pd2(dmpm)2(NCCH3)2]2+, where dmpm is bis(dimethylphosphino)methane.

Molecules 2018, 23, 140 3 of 21

2+ various cationic dinuclear Pd(I) complexes via ligand exchange reaction, [Pd2(NCCH3)6] (2)[25]. For example, this complex was used as a precursor for the bis-dmpm bridged dipalladium(I) dication, 2+ [PdMolecules2(dmpm) 2018, 232(NCCH, 140 3)2] , where dmpm is bis(dimethylphosphino)methane. 3 of 21

FigureFigure 1.1. RepresentativeRepresentative examplesexamples ofof dinucleardinuclear PdPd complexes.complexes. In parenthesis,parenthesis, thethe syntheticsynthetic methodsmethods A–EA–E areare shown.shown.

Vilar et al. successfully developed dipalladium(I) complex 4 using a bulky biphenylphosphine Vilar et al. successfully developed dipalladium(I) complex 4 using a bulky biphenylphosphine ligand, bis(tert-butyl)biphenylphosphine [26]. The complex was obtained by coupling of the Pd(0) ligand, bis(tert-butyl)biphenylphosphine [26]. The complex was obtained by coupling of the Pd(0) phosphine complex with a Pd(II) halide. Complex 4 mediated the Buchwald–Hartwig amination of phosphine complex with a Pd(II) halide. Complex 4 mediated the Buchwald–Hartwig amination of aryl halides efficiently at room temperature. Barder reported the similar dipalladium(I) complex 5. aryl halides efﬁciently at room temperature. Barder reported the similar dipalladium(I) complex 5. This complex was obtained by the reduction of palladium(II) dicyclohexyl(biaryl)phosphine complex This complex was obtained by the reduction of palladium(II) dicyclohexyl(biaryl)phosphine complex with silver tetrafluoroborate salt, leading to the coordination of the μ:η2:η2-aryl moiety of the with silver tetraﬂuoroborate salt, leading to the coordination of the µ:η2:η2-aryl moiety of the biarylphosphine ligand as a bridge between the two unsaturated metals [27]. The existence of 2’,6’- biarylphosphine ligand as a bridge between the two unsaturated metals [27]. The existence of dimethoxy groups on the biaryl substituent was essential to stabilize the dimeric structure. Therefore, 20,60-dimethoxy groups on the biaryl substituent was essential to stabilize the dimeric structure. the Pd-π interaction strongly stabilized the dipalladium framework. Therefore, the Pd-π interaction strongly stabilized the dipalladium framework. In 2004, Milstein et al. found that treatment of (2-methylallyl)PdCl2 with dmobp (dmobp = di- In 2004, Milstein et al. found that treatment of (2-methylallyl)PdCl with dmobp (dmobp = tert-butyl[(2,6-dimethoxyphenyl)methyl]phosphine) in the presence of NaOH2 yielded the π-allyl- di-tert-butyl[(2,6-dimethoxyphenyl)methyl]phosphine) in the presence of NaOH yielded the bridged dipalladium(I) complex 7a [28]. Denmark and Baird later reported the analogous complex π-allyl-bridged dipalladium(I) complex 7a [28]. Denmark and Baird later reported the analogous 7b, which used tri-tert-butylphosphine instead of the benzylphosphine. This complex has high catalytic complex 7b, which used tri-tert-butylphosphine instead of the benzylphosphine. This complex has activity toward Hiyama cross-coupling reactions of aryl halides with heterocyclic silanolates [29]. Bulky high catalytic activity toward Hiyama cross-coupling reactions of aryl halides with heterocyclic N-heterocyclic carbenes (NHCs) can also be used to stabilize the dinuclear Pd(I) system. Barcells and silanolates [29]. Bulky N-heterocyclic carbenes (NHCs) can also be used to stabilize the dinuclear Pd(I) Hazari et al. synthesized the IPr-supported μ-π-allyl Pd(I) dimers 7c, which are rare examples with system. Barcells and Hazari et al. synthesized the IPr-supported µ-π-allyl Pd(I) dimers 7c, which are 1-substituted μ-allyl ligands, by the reaction of Nolan’s IPrPd(II)Cl(π-allyl) complex with K2CO3 in rare examples with 1-substituted µ-allyl ligands, by the reaction of Nolan’s IPrPd(II)Cl(π-allyl) complex the presence of ethanol [30]. Bis-μ-allyl dipalladium(I) complexes 8 were also prepared by reacting with K CO in the presence of ethanol [30]. Bis-µ-allyl dipalladium(I) complexes 8 were also prepared mononuclear2 3 bis-π-allyl palladium(II) with free ligand, or by the reaction of μ-allyl palladium(II) by reacting mononuclear bis-π-allyl palladium(II) with free ligand, or by the reaction of µ-allyl chloride with free ligand and the sequential addition of an allyl Grignard reagent. Interestingly, CO2 palladium(II) chloride with free ligand and the sequential addition of an allyl Grignard reagent. was inserted into the palladium–carbon bond of 8 as an electrophile to form bridging carboxylate Interestingly, CO was inserted into the palladium–carbon bond of 8 as an electrophile to form complexes, which2 led to further catalytic processes as described below [12,31]. bridging carboxylate complexes, which led to further catalytic processes as described below [12,31]. Pfaltz et al. achieved isolation of an intermediary dipalladium(I) complex 9 using a chelating NP ligand for catalytic nucleophilic substitution of allylbenzoate with malonate [32]. Ozerov et al. successfully synthesized the bridging-ligand-free dipalladium(I) complexes 10 using monoanionic tridentate PNP pincer-type ligands. This was the representative example of a complex without bridging ligands that clearly showed the existence of a Pd(I)-Pd(I) bonding interaction [33] other than Murahashi and Kurosawa’s complex, [Pd2(NCCH3)6]2+ (2).

Molecules 2018, 23, 140 4 of 21

Pfaltz et al. achieved isolation of an intermediary dipalladium(I) complex 9 using a chelating NP ligand for catalytic nucleophilic substitution of allylbenzoate with malonate [32]. Ozerov et al. successfully synthesized the bridging-ligand-free dipalladium(I) complexes 10 using monoanionic tridentate PNP pincer-type ligands. This was the representative example of a complex without bridging Molecules 2018, 23, 140 4 of 21 ligands that clearly showed the existence of a Pd(I)-Pd(I) bonding interaction [33] other than Murahashi 2+ andCompared Kurosawa’s to complex, the analogous [Pd2(NCCH palladium3)6] (2 ).chemistry, dinuclear nickel(I) complexes and their catalyticCompared applications to the have analogous been po palladiumorly explored. chemistry, This may dinuclear be due nickel(I) to the greater complexes stability and theirof ‘monomeric’ catalytic applications have been poorly explored. This may be due to the greater stability of ‘monomeric’ monovalent nickel complexes compared to the corresponding palladium(I) complexes. Many monovalent nickel complexes compared to the corresponding palladium(I) complexes. Many examples examples of monomeric nickel(I) complexes have been reported [34], such as those relevant to the of monomeric nickel(I) complexes have been reported [34], such as those relevant to the oxygen oxygen activation system in bioinorganic chemistry [35]. Additionally, a great difference between the activation system in bioinorganic chemistry [35]. Additionally, a great difference between the number number of studies of the catalytic applications of palladium and nickel can be observed. of studies of the catalytic applications of palladium and nickel can be observed. Initial progress in the chemistry of dinuclear nickel(I) complexes has been promoted by the great Initial progress in the chemistry of dinuclear nickel(I) complexes has been promoted by the great π interestinterest in inunusual unusual interactions interactions of of conjugated conjugated π-ligands withwith thethe dinickel dinickel framework, framework, as as discussed discussed belowbelow [34]. [34 Preparative]. Preparative methods methods to toaccess access the the dinick dinickel(I)el(I) complexes complexes are are quite quite similar similar to to those those of of the palladiumthe palladium complexes, complexes, as might as might be expected. be expected. Dinickel Dinickel(I)(I) complexes complexes can be can prepared be prepared by simply by simply mixing unsaturatedmixing unsaturated monovalent monovalent and divale and divalentnt nickel nickel complexes complexes (Method (Method A), A), by by 1e-oxidation 1e-oxidation ofof nickel(0) nickel(0) complexescomplexes (Method (Method B),B), andand by by oxidative oxidative addition addition on Ni(0) on inNi(0) which in thewhich unsaturated the unsaturated adduct is stabilized adduct is stabilizedby forming by aforming metal–metal a metal–metal bonding interaction bonding withinteraction a second with Ni(0) a second complex Ni(0) (Method complex D). Additionally, (Method D). Additionally,in these processes, in these bridging processes, ligands bridging and/or ligands a coordinatively and/or a coordinatively unsaturated unsa site inturated one of site the in metal one of thecenters metal leadcenters to the lead facile to formationthe facile offormation the dimeric of structures,the dimeric similar structures, to procedures similar usingto procedures bridgingaryl, using bridgingallyl, and aryl, cyclopentadienyl allyl, and cyclopentadienyl groups in palladium groups in chemistry. palladium In Figurechemistry.2, representative In Figure 2, examplesrepresentative of examplesdinuclear of Ni(I)dinuclear complexes Ni(I) arecomplexes shown. are shown.

Figure 2. Representative examples of dinuclear Ni(I) complexes. In parenthesis, the synthetic methods Figure 2. Representative examples of dinuclear Ni(I) complexes. In parenthesis, the synthetic methods A–EA–E are are shown. shown.

In Inthe the earliest earliest stage stage of of the the development development of such complexes,complexes, Ni(I) Ni(I) hydride hydride dimers dimers stabilized stabilized by by bulkybulky chelating chelating bisphosphines bisphosphines (11 (11) )were were reported reported byby WilkeWilke andand Jonas Jonas [ 36[36].]. Those Those complexes complexes were were prepared by the reaction of NiCl2 (bisphosphine) with NaBMe3H, and then obtained via an alternative prepared by the reaction of NiCl2 (bisphosphine) with NaBMe3H, and then obtained via an alternative synthesis from the complex Ni(0) (bisphosphine) (benzene) and H2 gas at low temperature [37]. These synthesis from the complex Ni(0) (bisphosphine) (benzene) and H2 gas at low temperature [37]. complexesThese complexes were thermallywere thermally quite quitestable stable in the in so thelid solid state state and and did did not not decompose decompose toto the the monomeric monomeric form. However, [Ni(dippe)]2(μ-H)2, (dippe = 1,2-bis(diisopropylphosphino)ethane) reacts readily with cyclohexene sulfide to form sulfur-bridged dimers accompanied by the liberation of H2 gas [38], strongly indicating the potential of such Ni(I) hydride dimers as catalyst precursors. Bidentate nitrogen ligands with bulky substituents were also effective as auxiliary ligands in dinuclear Ni(I) complexes. Chirik et al. prepared diamine-supported dinickel(I) hydride complex 12, which can efficiently mediate the catalytic hydrosilylation of terminal alkenes [39]. In 2005, the preparation of chloride-bridged dinickel(I) complex 13a bearing IPr, where IPr is one of the bulky NHCs, 1,3-bis(2,6-diisopropylphenyl)imidazol-2-ylidene, was reported by Sigman et al. [40]. This complex has a 30e unsaturated nature and is kinetically stabilized by the bulky IPr

Molecules 2018, 23, 140 5 of 21

form. However, [Ni(dippe)]2(µ-H)2, (dippe = 1,2-bis(diisopropylphosphino)ethane) reacts readily with cyclohexene sulfide to form sulfur-bridged dimers accompanied by the liberation of H2 gas [38], strongly indicating the potential of such Ni(I) hydride dimers as catalyst precursors. Bidentate nitrogen ligands with bulky substituents were also effective as auxiliary ligands in dinuclear Ni(I) complexes. Chirik et al. prepared diamine-supported dinickel(I) hydride complex 12, which can efficiently mediate the catalytic hydrosilylation of terminal alkenes [39]. In 2005, the preparation of chloride-bridged dinickel(I) complex 13a bearing IPr, where IPr is one of the bulky NHCs, 1,3-bis(2,6-diisopropylphenyl)imidazol-2-ylidene, was reported by Sigman et al. [40]. This complex has a 30e unsaturated nature and is kinetically stabilized by the bulky IPr ligand. Subsequently, analogous dinuclear nickel(I) complexes with other bridging ligands, such as µ-arylimido and alkylene ligands, were produced by Hillhouse et al. through the reaction of Sigman’s dimer with arylazide and diazocarbene reagents [41,42]. These researchers also synthesized the dibromide analogue 13b [43]. Hazari et al. synthesized a series of dinuclear nickel(I) complexes 14 bearing bridging cyclopentadienyl and indenyl ligands, and compared their reactivity and stability to their palladium analogues [44]. According to this report, decomposition via heterolytic cleavage of the monovalent metal–metal bond to form a pair of zerovalent and divalent metal complexes is unfavorable for dinuclear nickel(I), but likely occurs for dinuclear palladium(I). Moreover, the metal–metal bonds in nickel(I) dimers are more resistant to splitting into monomeric complexes than those of the palladium analogues. Recently, Matsubara et al. serendipitously found that the reaction of aryl chloride with equal amounts of IPr and Ni (0)(cod)2 efficiently formed the dinuclear nickel(I) µ-σ-aryl complex 15 bearing the IPr ligand, rather than the monomeric oxidative adduct [45]. This dinickel(I) complex can also be prepared by the reaction of Sigman’s dimer with 1 equiv. of arylmagnesium chloride via transmetallation. Subsequent reaction with aryl chloride gave the biaryl and regenerated Sigman’s dimer, probably as a result of oxidative addition of aryl chloride and reductive elimination of the biaryl. Therefore, these complexes could potentially be active as catalysts in the Kumada–Tamao–Corriu coupling of aryl halides. Ghadwai et al. attempted to prepare the bromide analogue. However, similar treatment using bromoarenes with Ni(cod)2 and IPr provided the 2-arylated imidazolium salt rather than the dinuclear compound, probably due to the lower stability of the bromide analogue [46]. Thiolate is an effective bridging ligand that binds more strongly to metal centers than halide ligands. Tatsumi et al. reported that substitution of the bridging halide ligands with sulfide anions resulted in efficient formation of the corresponding µ-sulfide Ni(I) dimer 16a [47]. Olechnowicz et al. also synthesized the NHC analogue 16b, in which the bridging thiolate was µ–SH, from Sigman’s dimer [48]. Interestingly, this complex was also obtained from the heterolytic cleavage of H2 in the presence of the sulfido-bridged Ni(II) dimer [Ni(IPr)(µ-S)]2 [49]. In addition to the H–H bond, the H–B bond of pinacolborane was also heterolytically cleaved on the dinuclear nickel centers. Zargarian et al. reported that the decomposition of a divalent nickel complex after anion exchange with a non-coordinating anion such as tetraphenylborate provided the dimeric nickel(I) complex 17a in which two nickel centers were bridged by a µ-phosphide ligand [50]. Interestingly, one of the phenyl groups bound to borane also bridged the two nickel centers. Similar µ-phenyl bridging coordination was achieved more simply by Johnson et al. [51]. One of the halide ligands on a monovalent nickel complex was exchanged for tetraphenylborate to form an unsaturated nickel(I) center, which combined with another nickel(I) complex to yield dinuclear nickel(I) complex 17b, which features µ-phenyl bridging coordination by tetraphenylborate. Agapie et al. reported an interesting finding for the well-defined dinuclear nickel(I) complex 18, which is stabilized by a terphenylene-bridged diphosphine ligand that reacts with biphenyldimagnesium halide to form a transmetalated µ-biphenylene complex. This complex coupled a pair of biphenylene molecules on the two metal centers to form a bridging C–C bond to give tetraphenylene [52]. The complex also formed a fluorene molecule from biphenylene and dichloromethane. Johnson et al. Molecules 2018, 23, 140 6 of 21 reported that treatment of biphenylene with a monomeric Ni (0) complex resulted in the oxidative addition of biphenylene to form dinuclear bis(biphenyl) complex 19 containing a Ni(III)–Ni(I) bond [53]. Subsequent reductive coupling of the C–C bond provided a dinuclear tetraphenylene nickel(I) complex, which produced tetraphenylene catalytically in the presence of biphenylene upon heating, accompanied by regeneration of the dinuclear bis(biphenyl) complex. These results indicated that the unsaturated dinickel(I) system can also be effective for catalytic reactions involving concerted activation of substrates by a pair of active metal centers. π-Coordination bridging by a benzene ring can also stabilize the dinickel(I) framework, as seen in palladium complex 5. Jones et al. reported that a chelating three-atom-centered guanidinate ligand bearing a bulky DIP substituent (DIP = 2,6-diisopropylphenyl) coordinated to the Ni(I) center in an η1 fashion through the nitrogen atom, while one of the aromatic DIP substituents bridged the two Ni(I) centers to stabilize the Ni(I) dimer 20 [54]. In 2014, Uyeda et al. developed redox-active naphthyridine diimine-supported dinickel complexes [55]. Dinuclear Ni(I) complex 21 was formed upon comproportionation of Ni(cod)2 and NiBr2 in the presence of the naphthyridine diamine ligand. Notably, because of its relatively weak ligand field, the dinuclear Ni(I) complex was paramagnetic, S = 1, in contrast to the diamagnetic phosphine and NHC complexes. Moreover, oxidation and reduction were both found to be possible during electrochemical analysis, and reaction with oxidizing and reducing agents allowed the formation of a series of complexes in different 5e states. The authors also reported several catalytic reactions using the dinickel complex bearing naphthyridine diimine, which retained its dinuclear framework during catalysis [56].

3.1. Catalysis Using Palladium Complexes Several possible mechanisms for catalytic transformation by dinuclear palladium(I) complexes have been proposed: (1) Activation of substrates by the intact bimetallic complex; (2) disproportionation of the complex into Pd(0) and Pd(II), which are involved in the catalytic cycle; and (3) homolytic splitting of the complex into two Pd(I) molecules, which are smoothly reduced to Pd(0) or oxidized to Pd(II) (Scheme2). In the case of (2), an equilibrium between the regenerated dinuclear complex, which exists as a deactivated intermediate, and the active mononuclear complex may sometimes occur [62]. The most probable mechanism is the formation of highly active mono-ligated Pd(0) complexes from the corresponding dimer via mechanism (2) or (3). When monomeric complexes bearing more than two auxiliary ligands are used as catalyst precursors, Molecules 2018, 23, 140 6 of 21

In 2014, Uyeda et al. developed redox-active naphthyridine diimine-supported dinickel complexes [55]. Dinuclear Ni(I) complex 21 was formed upon comproportionation of Ni(cod)2 and NiBr2 in the presence of the naphthyridine diamine ligand. Notably, because of its relatively weak ligand field, the dinuclear Ni(I) complex was paramagnetic, S = 1, in contrast to the diamagnetic phosphine and NHC complexes. Moreover, oxidation and reduction were both found to be possible during electrochemical analysis, and reaction with oxidizing and reducing agents allowed the formation of a series of complexes in different 5e states. The authors also reported several catalytic reactions using the dinickel complex bearing naphthyridine diimine, which retained its dinuclear framework during catalysis [56].

3. Catalytic Applications Many reports of catalytic processes mediated by dinuclear palladium(I) complexes used as catalyst precursors have been published. Notably, almost all of these studies reported remarkably high efficiencies for the catalytic reactions, which included Suzuki–Miyaura coupling [28], Sonogashira coupling [57], Hiyama coupling [29], Buchwald–Hartwig amination [26], Kumada–Corriu and Negishi coupling [58] (Scheme 1), and others [59–61], suggesting that dinuclear Pd(I) complexes can generate species with unusually high activity as key intermediates. However, in all but a few examples,Molecules 2018it was, 23, 140unclear how these reactions proceeded from the dinuclear Pd(I) complexes.7 of 21 Additionally, there are several examples of catalytic transformations of organic compounds mediated by dinuclear nickel(I) complexes among the huge number of reported catalytic reactions using nickel liberation of these ligands requires harsh conditions. On the other hand, the Pd(I)–Pd(I) bond in complexes. The mechanism of these catalytic reactions has been studied in detail. This section focuses dinuclear complexes can be broken quite easily to form the active mono-ligated complex without on the roles of Pd and Ni bimetallic complexes in catalysis. Several factors, such as the choice of breaking the stronger metal–ligand bonds. This is probably the mechanism by which dinuclear Pd(I) bridgingcomplexes and/or are auxiliary active in ligands, these catalytic the type systems. of reaction, The activation and the metal of substrates center, bycritically unsaturated affected Pd(I) these roles.dimers Some that of the do nottransformations undergo splitting were into activate mononucleard via the complexes formation is of rare mononuclear in palladium mono-ligated, chemistry. unsaturatedHowever, concertedmetal complexes bimetallic activationafter disproportiona of the substratestion. by In two such adjacent complexes, metal centers re-formation could possibly of the dinuclearinduce complex highly active by comproportionation and/or selective transformations deactivated in the catalysis. complexes, Therefore, or served new reactionsto stabilize can active be form.developed In other bytransformations, designing bimetallic the intact activation dinucl processes.ear complex activated the substrates efficiently.

(a) Suzuki-Miyaura (2004/Milstein, D.) Cl R Pd cat. (1 mol%) R (tBu) P Pd Pd P(tBu) Cl + B(OH)2 2 2 o-xylene, CsF O O 40 oC O 7a O (b) Sonogashira (2009/Moore, J. S.) 32-43% Molecules 2018, 23, 140 Pd cat. (1 mol%) Cl 7 of 21 R ZnCl2 (10 mol%) R t t ( Bu)3P Pd Pd P( Bu)3 Br + R' H R' disproportionation of the complex intoTHF, Pd(0)HN(iPr)2 and Pd(II), which are involved in the catalytic cycle; rt 7b and (3) homolytic splitting of the complex into two Pd(I) 76-96molecules, % which are smoothly reduced to (c) Hiyama (2006/Denmark, S. E.) Cl Pd(0) or oxidizedR to Pd(II) (Scheme 2). In the case of R(2), an equilibriumX between the regenerated 1. NaH / toluene (tBu) P Pd Pd P(tBu) dinuclear complex, whichBr + exists as a deactivated intermediate, and the active3 mononuclear3 complex X SiMe2OH 2. Pd cat. (2.5 mol%) o may sometimes occur [62].X =The O, S most toluene,probable 50 C mechanism is the formation 7bof highly active 60-86 % (d) Buchwald-Hartwig amination (2004/Vilar, R.) t mono-ligated Pd(0) complexes from the corresponding dimer via mechanism (2)Bu or (3).X When monomeric tBu complexes bearing more than two auxPdiliary cat. (1 mol%)ligands are used as catalyst precursors,P Pd PdliberationX of these t t Bu Br + HNPh2 Bu NPh2 ligands requires harsh conditions. On THF,the rtother hand, the Pd(I)–Pd(I) bond in dinuclear4 complexes 86 % X = Cl, Br can be broken(e) Kumada-Corriu, quite easily Negishi to (2017/Schoe form thenebeck, active F.) mono-ligated complex without breaking the stronger metal–ligand bonds. This is probably the mechanism by which dinuclear Pd(I) complexesI are active R R' Pd cat. (2.5 mol%) R R' (tBu) P Pd Pd P(tBu) in these catalytic systems.Br + The M-Bractivation of substrates by unsaturated 3Pd(I) dimers that3 do not Tol/THF, rt I 1a air undergo splitting into mononuclearM = Mg, Zn complexes is rare in palladium chemistry. However, concerted bimetallic activation of the substrates by two adjacent metal71-98% centers could possibly induce highly active and/orScheme Schemeselective 1. Representative 1. Representativetransformations examples examples in catalysi of of catalytic catalytics. Therefore, processes using using new Pd(I)–Pd(I) Pd(I)–Pd(I)reactions complexes. cancomplexes. be developed by designing bimetallic activation processes. 3.1. Catalysis Using Palladium Complexes Several possible mechanisms for catalytic transformation by dinuclear palladium(I) complexes have been proposed: (1) Activation of substrates by the intact bimetallic complex; (2)

SchemeScheme 2. Three 2. Three possible possible activation activation pathways pathways for for dinuclear Pd(I) Pd(I) complexes complexes in in catalysis: catalysis: (1) (1) activation activation of substratesof substrates by by an an intact dimer;dimer; (2) (2) disproportionation disproportionation of adimer of a intodimer Pd(0) into and Pd(0) Pd(II); and and Pd(II); (3) splitting and (3) splittingof the of dimer the dimer into two into molecules two molecules of Pd(I). of Pd(I).

3.1.1.3.1.1. Monomeric Monomeric Palladium Palladium Catalysts Catalysts from from Dinuclear Dinuclear ComplexesComplexes In 2002,In 2002, Hartwig Hartwig et al. et reported al. reported the achievement the achievement of efficient of efﬁcient cross-coupling cross-coupling reactions, reactions, the Suzuki– the MiyauraSuzuki–Miyaura coupling and coupling the Buchwald–Hartwig and the Buchwald–Hartwig amination, amination, of aryl of chlorides aryl chlorides at room at room temperature temperature using using the dinuclear palladium(I) complex [Pd(PtBu )] (µ-X) (X = Cl, Br) (1b)[63]. The oxidative the dinuclear palladium(I) complex [Pd(PtBu3)]2(μ-X)23 (X2 = Cl,2 Br) (1b) [63]. The oxidative addition of addition of aryl chloride was revealed to be the rate-determining step, and the addition of PtBu3 to a aryl chloride was revealed to be the rate-determining step, and the addition of PtBu3 to a palladium(0) palladium(0) precursor showed similar activity toward the cross-coupling reactions. Therefore, it was precursor showed similar activity toward the cross-coupling reactions. Therefore, it was proposed that proposed that disproportionation of the dinuclear Pd(I) complex gave the active mononuclear Pd(0) disproportionation of the dinuclear Pd(I) complex gave the active mononuclear Pd(0) intermediate intermediate [Pd(0)(PtBu3)], which also could be formed by the reaction of the palladium(0) precursor [Pd(0)(PtBu3)], which also could be formed by the reaction of the palladium(0) precursor and PtBu3. and PtBu3. Barder showed that Suzuki–Miyaura coupling of aryl chloride could be accomplished using cationic palladium(I) dimer 5 bearing a biarylphosphine ligand, in which the π-electrons of the dimethoxy-substituted aryl moieties stabilize the unsaturated palladium centers [27]. Because the dimer complex did not activate aryl chloride at low temperature, it was proposed that formation of active monomeric complex at a temperature above 60 °C may be necessary (Scheme 3). Disproportionation of the bimetallic Pd(I) complex gives not only Pd(0) but also Pd(II) complexes, which could be easily reduced to Pd(0) by transmetallation with arylboronic acid to form biphenyl as a by-product.

Scheme 3. Proposed disproportionation pathway for the initiation of Suzuki–Miyaura coupling.

The Pd(I) dimer [Pd(IPr)]2(μ-Cl)(μ-π-allyl) (7c) was detected and determined to be a reaction intermediate in the Suzuki–Miyaura coupling of aryl chloride in the presence of [Pd(II)(IPr)(π-allyl)Cl] and a base by Balcells and Hazari et al. [30]. The dimer complex 7c can easily disproportionate into the

Molecules 2018, 23, 140 7 of 21 disproportionation of the complex into Pd(0) and Pd(II), which are involved in the catalytic cycle; and (3) homolytic splitting of the complex into two Pd(I) molecules, which are smoothly reduced to Pd(0) or oxidized to Pd(II) (Scheme 2). In the case of (2), an equilibrium between the regenerated dinuclear complex, which exists as a deactivated intermediate, and the active mononuclear complex may sometimes occur [62]. The most probable mechanism is the formation of highly active mono-ligated Pd(0) complexes from the corresponding dimer via mechanism (2) or (3). When monomeric complexes bearing more than two auxiliary ligands are used as catalyst precursors, liberation of these ligands requires harsh conditions. On the other hand, the Pd(I)–Pd(I) bond in dinuclear complexes can be broken quite easily to form the active mono-ligated complex without breaking the stronger metal–ligand bonds. This is probably the mechanism by which dinuclear Pd(I) complexes are active in these catalytic systems. The activation of substrates by unsaturated Pd(I) dimers that do not undergo splitting into mononuclear complexes is rare in palladium chemistry. However, concerted bimetallic activation of the substrates by two adjacent metal centers could possibly induce highly active and/or selective transformations in catalysis. Therefore, new reactions can be developed by designing bimetallic activation processes.

Scheme 2. Three possible activation pathways for dinuclear Pd(I) complexes in catalysis: (1) activation of substrates by an intact dimer; (2) disproportionation of a dimer into Pd(0) and Pd(II); and (3) splitting of the dimer into two molecules of Pd(I).

3.1.1. Monomeric Palladium Catalysts from Dinuclear Complexes In 2002, Hartwig et al. reported the achievement of efficient cross-coupling reactions, the Suzuki– Miyaura coupling and the Buchwald–Hartwig amination, of aryl chlorides at room temperature using the dinuclear palladium(I) complex [Pd(PtBu3)]2(μ-X)2 (X = Cl, Br) (1b) [63]. The oxidative addition of aryl chloride was revealed to be the rate-determining step, and the addition of PtBu3 to a palladium(0) precursorMolecules showed2018, 23, 140 similar activity toward the cross-coupling reactions. Therefore, it was proposed8 of 21that disproportionation of the dinuclear Pd(I) complex gave the active mononuclear Pd(0) intermediate [Pd(0)(PtBuBarder3)], showedwhich also that could Suzuki–Miyaura be formed by coupling the reac oftion aryl of chloride the palladium(0) could be accomplishedprecursor and using PtBu3. cationicBarder palladium(I) showed that dimer Suzuki–Miyaura5 bearing a biarylphosphinecoupling of aryl ligand, chloride in whichcould thebe accomplishedπ-electrons of theusing cationicdimethoxy-substituted palladium(I) dimer aryl 5 moieties bearing stabilize a biarylphosphine the unsaturated ligand, palladium in which centers the [π27-electrons]. Because of the the dimethoxy-substituteddimer complex did not aryl activate moieties aryl stabilize chloride th ate low unsaturated temperature, palladium it was proposed centers [27]. that formationBecause the dimerof activecomplex monomeric did not activate complex aryl atchloride a temperature at low temperature, above 60 it◦ wasC may proposed be necessary that formation (Scheme of 3active). monomericDisproportionation complex at ofa temperature the bimetallic above Pd(I) 60 complex °C may gives be necessary not only (Scheme Pd(0) but 3). also Disproportionation Pd(II) complexes, of thewhich bimetallic could Pd(I) be easily complex reduced gives to Pd(0) not only by transmetallation Pd(0) but also with Pd(II) arylboronic complexes, acid which to form could biphenyl be easily as reduceda by-product. to Pd(0) by transmetallation with arylboronic acid to form biphenyl as a by-product.

SchemeScheme 3. Proposed 3. Proposed disproportionation disproportionation pathway pathway for the initiationinitiation of of Suzuki–Miyaura Suzuki–Miyaura coupling. coupling.

The Pd(I) dimer [Pd(IPr)]2(μ-Cl)(μ-π-allyl) (7c) was detected and determined to be a reaction The Pd(I) dimer [Pd(IPr)]2(µ-Cl)(µ-π-allyl) (7c) was detected and determined to be a reaction intermediateintermediate in inthe the Suzuki–Miyaura Suzuki–Miyaura coupling coupling of of aryl aryl chloride inin thethe presence presence of of [Pd(II)(IPr)( [Pd(II)(IPr)(π-allyl)Cl]π-allyl)Cl] Molecules 2018, 23, 140 8 of 21 andand a base a base by Balcells by Balcells and andHazari Hazari et al. et [30]. al. [The30]. dimer The dimer complex complex 7c can7c easilycan disproportionate easily disproportionate into the π monomericinto the monomeric complexes IP complexesr-Pd(0) and IPr-Pd(0) Pd(II) π-allyl and to Pd(II) initiate -allylthe catalytic to initiate process; the moreover, catalytic addition process; π ofmoreover, an equivalent addition of an[Pd(II)(IPr)( equivalentπ-allyl)Cl] of [Pd(II)(IPr)( was found-allyl)Cl] to wasdeactivate found tothe deactivate process. theTherefore, process. π aTherefore, comproportionation a comproportionation process producin processg producingan equilibrium an equilibrium between the between Pd(I) theπ–allyl Pd(I) dimer–allyl and dimer the mononuclearand the mononuclear complexes complexes was also was identified, also identiﬁed, demonstrating demonstrating that the that Pd(I) the Pd(I)dimer dimer 7c forms7c forms as an as off-cyclean off-cycle product. product. Several Several studies studies identifying identifying a dinu a dinuclearclear Pd(I) Pd(I) complex complex as the as the resting resting state state outside outside of theof the catalytic catalytic cycle cycle have have also also been been reported reported [64–67]. [64–67 ]. Pfaltz et al. reported the palladium-catalyzed nucleophilic substitution of allylbenzoate with malonate, an example of the formation of a dinuclea dinuclearr structure proposed to be an off-cycle product that stabilized stabilized the the active active mononuclear mononuclear intermediate intermediatess [32]. [32 ].They They found found a reversible a reversible pathway pathway from from the π mononuclearthe mononuclear Pd(0) Pd(0) and Pd(II) and Pd(II) π-allyl-allyl complexes complexes to the to dinuclear the dinuclear Pd(I) complex Pd(I) complex 9, in which9, in a which π-allyl a π moiety-allyl moietybridged bridged the two the metal two centers. metal centers. However, However, further further reaction reaction of the ofdinuclear the dinuclear complex complex with malonatewith malonate did not did yield not the yield substituti the substitutionon product, product, indicating indicating that the thatdimer the is dimer not active is not in activethe catalytic in the transformation.catalytic transformation. This dinuclear This dinuclear Pd(I) complex Pd(I) complexreversibly reversibly produced produced the mononuclear the mononuclear Pd(II) π Pd(II)-allyl π complex-allyl complex in the presence in the presence of allylbenzoate of allylbenzoate (Scheme (Scheme 4). 4).

Scheme 4. Proposed mechanism of Pd-catalyzed allylic substitution. Scheme 4. Proposed mechanism of Pd-catalyzed allylic substitution. Schoenebeck et al. reported that the initiation process of the Heck reaction did not involve simple Schoenebeck et al. reported that the initiation process of the Heck reaction did not involve heterolytic dissociation of the dinuclear palladium(I) complex [Pd(PtBu3)]2(μ-Cl)2 to form the correspondingsimple heterolytic mono-ligated dissociation Pd(0) of the active dinuclear catalyst palladium(I) [68]. Instead, complex formation [Pd(PtBu of 3the)]2( µmonomeric-Cl)2 to form Pd(I) the corresponding mono-ligated Pd(0) active catalyst [68]. Instead, formation of the monomeric Pd(I) complex [Pd(PtBu3)Cl] by homolytic cleavage of the Pd(I) dimer is more favorable, as it proceeds with a much lower free energy (ΔGdiss = 24.8 kcal mol-1) than the heterolytic formation of [Pd(0)(PtBu3)] and [Pd(II)(PtBu3)Cl2] (ΔGdiss = 38.1 kcal mol-1), as estimated by DFT calculations. Additionally, the dimer complex itself, as well as its decomposition product Pd(0)(PtBu3)2, which was gradually generated from the dimer in the presence of aryl boronic acid and KF, were revealed to be less active toward catalysis. Therefore, these authors proposed that the true active catalyst was most likely to be the mono-ligated or anionic complex, [Pd(0)(PtBu3)] or [Pd(0)(PtBu3)Cl]-, formed by reduction of the monomeric Pd(I) intermediate [Pd(PtBu3)Cl] (Scheme 5).

Scheme 5. Possible pathways for the activation of the Pd(I) dimer.

3.1.2. Proposed Mechanisms Involving Dinuclear Palladium Catalysts The possibility of catalytic processes involving an active dipalladium(I) complex that retains its dinuclear framework has also been proposed. In 2010, Bera et al. reported that a dinuclear Pd(I) complex bearing tridentate 1,8-naphthyridine derivative ligands mediated Suzuki–Miyaura cross coupling of aryl halides [69]. Although 3-bromopyridine was effective for cross coupling, 2- bromopyridine was not suitable for the coupling reaction, and yielded only the starting material. They therefore proposed the presence of a metal coordination site adjacent to the active metal center

Molecules 2018, 23, 140 8 of 21 monomeric complexes IPr-Pd(0) and Pd(II) π-allyl to initiate the catalytic process; moreover, addition of an equivalent of [Pd(II)(IPr)(π-allyl)Cl] was found to deactivate the process. Therefore, a comproportionation process producing an equilibrium between the Pd(I) π–allyl dimer and the mononuclear complexes was also identified, demonstrating that the Pd(I) dimer 7c forms as an off-cycle product. Several studies identifying a dinuclear Pd(I) complex as the resting state outside of the catalytic cycle have also been reported [64–67]. Pfaltz et al. reported the palladium-catalyzed nucleophilic substitution of allylbenzoate with malonate, an example of the formation of a dinuclear structure proposed to be an off-cycle product that stabilized the active mononuclear intermediates [32]. They found a reversible pathway from the mononuclear Pd(0) and Pd(II) π-allyl complexes to the dinuclear Pd(I) complex 9, in which a π-allyl moiety bridged the two metal centers. However, further reaction of the dinuclear complex with malonate did not yield the substitution product, indicating that the dimer is not active in the catalytic transformation. This dinuclear Pd(I) complex reversibly produced the mononuclear Pd(II) π-allyl complex in the presence of allylbenzoate (Scheme 4).

Scheme 4. Proposed mechanism of Pd-catalyzed allylic substitution.

Schoenebeck et al. reported that the initiation process of the Heck reaction did not involve simple heterolyticMolecules 2018 dissociation, 23, 140 of the dinuclear palladium(I) complex [Pd(PtBu3)]2(μ-Cl)2 to form9 of 21the corresponding mono-ligated Pd(0) active catalyst [68]. Instead, formation of the monomeric Pd(I) complex [Pd(PtBu3)Cl] by homolytic cleavage of the Pd(I) dimer is more favorable, as it proceeds with complex [Pd(PtBu3)Cl] by homolytic cleavage of the Pd(I) dimer is more favorable, as it proceeds with a a much lower free energy (ΔGdiss = 24.8 kcal mol−1 -1) than the heterolytic formation of [Pd(0)(PtBu3)] much lower free energy (∆Gdiss = 24.8 kcal mol ) than the heterolytic formation of [Pd(0)(PtBu3)] and and [Pd(II)(PtBu3)Cl2] (ΔGdiss = 38.1 kcal −mol1 -1), as estimated by DFT calculations. Additionally, the [Pd(II)(PtBu3)Cl2](∆Gdiss = 38.1 kcal mol ), as estimated by DFT calculations. Additionally, the dimer 3 2 dimercomplex complex itself, itself, as well as as well its decomposition as its decomposition product Pd(0)(PtBu product Pd(0)(PtBu3)2, which was) , graduallywhich was generated gradually generatedfrom the from dimer the in dimer the presence in the presence of aryl boronic of aryl acid boronic and KF,acid were and revealedKF, were to revealed be less activeto be less toward active towardcatalysis. catalysis. Therefore, Therefore, these these authors authors proposed proposed that the that true the active true ac catalysttive catalyst was most was likely most tolikely be the to be - - themono-ligated mono-ligated or or anionic anionic complex, complex, [Pd(0)(PtBu [Pd(0)(PtBu3)]3)] or or [Pd(0)(PtBu [Pd(0)(PtBu3)Cl]3)Cl], formed, formed by by reduction reduction of theof the monomericmonomeric Pd(I) Pd(I) intermediate intermediate [Pd(PtBu [Pd(PtBu33)Cl])Cl] (Scheme (Scheme5 5).).

SchemeScheme 5.5. PossiblePossible pathways pathways for for the activationof of the the Pd(I) Pd(I) dimer. dimer.

3.1.2.3.1.2. Proposed Proposed Mechanisms Mechanisms Involving Involving Dinuclear Dinuclear Palladium CatalystsCatalysts TheThe possibility possibility of ofcatalytic catalytic processes processes involving involving an an active active dipalladium(I) dipalladium(I) complex complex that that retains retains its dinuclearits dinuclear framework framework has also has alsobeen been proposed. proposed. In 20 In10, 2010, Bera Bera et al. et reported al. reported that thata dinuclear a dinuclear Pd(I) complexPd(I) complex bearing bearing tridentate tridentate 1,8-naphthyridine 1,8-naphthyridine derivative derivative ligands ligands mediated mediated Suzuki–MiyauraSuzuki–Miyaura cross cross couplingcoupling of of aryl aryl halideshalides [69 [69].]. Although Although 3-bromopyridine 3-bromopyridine was effective was for effective cross coupling, for cross 2-bromopyridine coupling, 2- bromopyridinewas not suitable was for thenot couplingsuitable reaction, for the andcoupling yielded re onlyaction, the startingand yielded material. only They the therefore starting proposed material. Theythe therefore presence of proposed a metal coordination the presence site of adjacent a metal to coor the activedination metal site center adjacent that could to the be active easily bridgedmetal center by 2-bromopyridine in the oxidative addition process. However, deﬁnitive identiﬁcation of the bimetallic system in catalysis would require more detailed study. Subsequently Hazari et al. proposed that an intact dinuclear palladium(I) complex acts as a bimetallic catalyst in the allylic carboxylation of allylstannane and allylboronate with CO2 [31]. They found that stoichiometric reaction of the bis-µ-π-allyl Pd(I) complex with CO2 afforded a bridged mono-carboxylate dipalladium(I) complex via facile insertion of CO2 into the palladium-π-allyl bond. Subsequent addition of allylstannane and allylboronate gave the starting π-allyl complex accompanied by liberation of allylcarboxylate (Scheme6). Detailed theoretical calculations also indicated that the dinuclear framework of the dipalladium(I) catalyst remains intact during the catalytic allylic carboxylation [70]. However, the addition of weakly coordinating ligands, such as acetonitrile and 1-octene, accelerated the catalytic reaction. Moreover, the further addition of these ligands or the addition of other ligands yielded mononuclear Pd(0) complexes and palladium black [71]. Therefore, another catalytic pathway involving insertion of CO2 into a mononuclear Pd(II) complex was proposed. In this pathway, coordination of the ligand L’ stabilizes the disproportionated mononuclear Pd(0) complex in an equilibrium with the dinuclear complex, and the accompanying formation of the σ-allyl Pd(II) complex enables the insertion of CO2 into the Pd-carbon bond. The most remarkable process involving dinuclear Pd(I) complexes is a series of halogen exchange reactions of aryl halides studied by Schoenebeck et al. [72,73]. Interestingly, the reaction was not catalyzed by mononuclear Pd(II) halide. However, when a dinuclear Pd(I) catalyst was used, the two bridging halogen ligands of the complex underwent stepwise exchange, as observed using 31P NMR spectroscopy. A detailed theoretical study revealed that the oxidative addition of aryl iodide occurs at a single metal center that retains its dinuclear framework. After the site exchange process of the halogen ligand on the Pd(II) dimer, the semi-stable dipalladium(II) adduct intermediate eliminates aryl bromide to generate the Pd(I) dimer (Scheme7). A similar reaction mechanism was proposed for - the halogen exchange reaction of aryl iodide with SCF3 or SeCF3 [74,75]. Molecules 2018, 23, 140 9 of 21 that could be easily bridged by 2-bromopyridine in the oxidative addition process. However, definitive identification of the bimetallic system in catalysis would require more detailed study. Subsequently Hazari et al. proposed that an intact dinuclear palladium(I) complex acts as a bimetallic catalyst in the allylic carboxylation of allylstannane and allylboronate with CO2 [31]. They found that stoichiometric reaction of the bis-μ-π-allyl Pd(I) complex with CO2 afforded a bridged mono-carboxylate dipalladium(I) complex via facile insertion of CO2 into the palladium-π-allyl bond. Subsequent addition of allylstannane and allylboronate gave the starting π-allyl complex accompanied by liberation of allylcarboxylate (Scheme 6). Detailed theoretical calculations also indicated that the dinuclear framework of the dipalladium(I) catalyst remains intact during the catalytic allylic carboxylation [70]. However, the addition of weakly coordinating ligands, such as Molecules 2018, 23, 140 9 of 21 acetonitrile and 1-octene, accelerated the catalytic reaction. Moreover, the further addition of these ligands or thethat addition could beof easilyother bridgedligands byyielded 2-bromopyridine mononuclear in thePd(0) oxidative complexes addition and process.palladium However, black [71]. Therefore,definitive another identification catalytic of pathway the bimetallic involving system insertion in catalysis of would CO2 into require a mononuclea more detailedr Pd(II) study. complex was proposed.Subsequently In this Hazari pathway, et al. proposedcoordination that anof intactthe ligand dinuclear L’ palladium(I)stabilizes the complex disproportionated acts as a mononuclearbimetallic Pd(0)catalyst complex in the allylic in an carboxylation equilibrium of with allylstannane the dinuclear and allylboronate complex, and with the CO accompanying2 [31]. They formationfound ofthat the stoichiometric σ-allyl Pd(II) reaction complex of enablesthe bis-μ the-π-allyl insertion Pd(I) complexof CO2 into with the CO Pd-carbon2 afforded abond. bridged mono-carboxylate dipalladium(I) complex via facile insertion of CO2 into the palladium-π-allyl bond. Subsequent addition of allylstannane and allylboronate gave the starting π-allyl complex accompanied by liberation of allylcarboxylate (Scheme 6). Detailed theoretical calculations also indicated that the dinuclear framework of the dipalladium(I) catalyst remains intact during the catalytic allylic carboxylation [70]. However, the addition of weakly coordinating ligands, such as acetonitrile and 1-octene, accelerated the catalytic reaction. Moreover, the further addition of these ligands or the addition of other ligands yielded mononuclear Pd(0) complexes and palladium black [71]. Therefore, another catalytic pathway involving insertion of CO2 into a mononuclear Pd(II) complex was proposed. In this pathway, coordination of the ligand L’ stabilizes the disproportionated Moleculesmononuclear2018, 23 Pd(0), 140 complex in an equilibrium with the dinuclear complex, and the accompanying10 of 21 formation of the σ-allyl Pd(II) complex enables the insertion of CO2 into the Pd-carbon bond.

Scheme 6. Proposed catalytic cycle of allylic carboxylation using a bis(μ-π-allyl)dipalladium(I) complex, which is supported by experimental results.

The most remarkable process involving dinuclear Pd(I) complexes is a series of halogen exchange reactions of aryl halides studied by Schoenebeck et al. [72,73]. Interestingly, the reaction was not catalyzed by mononuclear Pd(II) halide. However, when a dinuclear Pd(I) catalyst was used, the two bridging halogen ligands of the complex underwent stepwise exchange, as observed using 31P NMR spectroscopy. A detailed theoretical study revealed that the oxidative addition of aryl iodide occurs at a single metal center that retains its dinuclear framework. After the site exchange process Scheme 6. Proposed catalytic cycle of allylic carboxylation using a bis(μ-π-allyl)dipalladium(I) of the halogenScheme 6.ligandProposed on catalytic the Pd(II) cycle of dimer, allylic carboxylation the semi-stable using a bis(dipalladium(II)µ-π-allyl)dipalladium(I) adduct complex, intermediate complex, which is supported by experimental results. eliminateswhich aryl is bromide supported to by generate experimental the results. Pd(I) dimer (Scheme 7). A similar reaction mechanism was - proposedThe for themost halogen remarkable exchange process reaction involving of aryl dinuclear iodide Pd(I)with complexesSCF3 or SeCF is a3 [74,75].series of halogen exchange reactions of aryl halides studied by Schoenebeck et al. [72,73]. Interestingly, the reaction was not catalyzed by mononuclear Pd(II) halide. However, when a dinuclear Pd(I) catalyst was used, the two bridging halogen ligands of the complex underwent stepwise exchange, as observed using 31P NMR spectroscopy. A detailed theoretical study revealed that the oxidative addition of aryl iodide occurs at a single metal center that retains its dinuclear framework. After the site exchange process of the halogen ligand on the Pd(II) dimer, the semi-stable dipalladium(II) adduct intermediate eliminates aryl bromide to generate the Pd(I) dimer (Scheme 7). A similar reaction mechanism was proposed for the halogen exchange reaction of aryl iodide with -SCF3 or SeCF3 [74,75].

SchemeScheme 7. Mechanism 7. Mechanism of halogen of halogen exchange exchange of of iodobenzen iodobenzene,e, proposed proposed based based on on theoretical theoretical studies. studies.

The chemistry of Pd(I) dimer in catalysis has been summarized as follows. The above studies on catalytic systems mediated by dinuclear Pd(I) complexes revealed that the formation of monomeric

Pd(0) and/or Pd(II) complexes as the true active catalytic species is the key process to initiate most reactions. Coordinatively unsaturated, mono-ligated palladium complexes of the form Pd(0)L are

proposed to be very active catalysts that enable facile activation of substrates, whereas typical Pd(0)L2 complexesScheme are 7. less Mechanism active for of thesehalogen reactions. exchange Several of iodobenzen pathwayse, proposed to generate based the on theoretical monomeric studies. complexes can be proposed, namely, the simple disproportionation of the dimer into zerovalent and divalent complexes, and the reduction or oxidation of Pd(I) complexes produced by homolytic cleavage of the Pd(I)–Pd(I) bond. Additionally, in situ generated, inactive dinuclear Pd(I) complexes sometimes

stabilize the unstable monomeric active species as resting states, and it is quite rare for intact dinuclear complexes to act as catalysts. It has also been reported that only one of the metal centers directly activate the C–X bond of aryl halide in the oxidative addition process. In many cases, it is very difﬁcult to clarify whether the active compound is a mononuclear or dinuclear complex. However, theoretical calculations combined with experimental results can suggest possible pathways. These results reveal the nature of dinuclear Pd(I) complexes and new possible transformations using these complexes, thus enabling progress in the development of highly efﬁcient reactions using bimetallic palladium systems in the near future.

The chemistry of Pd(I) dimer in catalysis has been summarized as follows. The above studies on catalytic systems mediated by dinuclear Pd(I) complexes revealed that the formation of monomeric Pd(0) and/or Pd(II) complexes as the true active catalytic species is the key process to initiate most reactions. Coordinatively unsaturated, mono-ligated palladium complexes of the form Pd(0)L are proposed to be very active catalysts that enable facile activation of substrates, whereas typical Pd(0)L2 complexes are less active for these reactions. Several pathways to generate the monomeric complexes can be proposed, namely, the simple disproportionation of the dimer into zerovalent and divalent complexes, and the reduction or oxidation of Pd(I) complexes produced by homolytic cleavage of the Pd(I)–Pd(I) bond. Additionally, in situ generated, inactive dinuclear Pd(I) complexes sometimes stabilize the unstable monomeric active species as resting states, and it is quite rare for intact dinuclear complexes to act as catalysts. It has also been reported that only one of the metal centers directly activate the C–X bond of aryl halide in the oxidative addition process. In many cases, it is very difficult to clarify whether the active compound is a mononuclear or dinuclear complex. However, theoretical calculations combined with experimental results can suggest possible pathways. These results reveal the nature of dinuclear Pd(I) complexes and new possible transformations using these complexes, thus enabling progress in the development of highly efficient reactions using bimetallic palladium systems in the near future.

3.2. Catalysis Using Nickel Complexes As noted above, studies on catalytic transformations using dinuclear nickel(I) complexes have been less reported compared with the corresponding palladium chemistry (Schemes 8 and 9). The studiesMolecules that2018 have, 23, 140been performed using nickel(I) complexes indicated the catalytic reactions11 of 21 were initiated by similar processes: Homolysis or heterolysis of Ni(I)–Ni(I) bonds to form mononuclear active catalysts (Scheme 8), and catalysis by intact dinuclear complexes (Scheme 9). When the active catalysts (Scheme8), and catalysis by intact dinuclear complexes (Scheme9). When the dinuclear dinuclearstructures structures remain remain intact during intact catalysis, during catalyticcatalysis, transformations catalytic transf usingormations dinuclear using nickel(I) dinuclear complexes nickel(I) complexescan proceed can proceed similarly similarly to those to using those palladium. using palla Ondium. the other On the hand, other the hand, greatest the difference greatest indifference the in thechemistry chemistry of nickelof nickel is that is that its paramagnetic, its paramagnet monovalentic, monovalent mononuclear mononuclear complexes complexes are much moreare much stable more stablethan than the the corresponding correspond palladiuming palladium complexes. complexes. Therefore, Therefore, homolytic homolytic cleavage cleavage of the Ni(I)–Ni(I) of the Ni(I)–Ni(I) bond bondoccurs occurs readily readily to formto form the coordinativelythe coordinatively unsaturated unsaturated Ni(I) complexes, Ni(I) complexes, which are anticipatedwhich are toanticipated have to havehigh high catalytic catalytic activity. activity.

SchemeScheme 8. Examples 8. Examples ofof catalytic catalytic processes processes usingdinuclear dinuclear Ni(I) Ni(I) catalyst catalyst precursors. precursors.

Molecules 2018, 23, 140 12 of 21 Molecules 2018, 23, 140 11 of 21

Scheme 9. Examples of catalytic processes using dinuclear Ni complexes, in which the complexes Scheme 9. Examples of catalytic processes using dinuclear Ni complexes, in which the complexes were were proposedproposed to to retain retain their their dinuclear dinuclear frameworks. frameworks.

3.2.1. 3.2.1.Mononuclear Mononuclear Nickel Nickel Catalysts Catalysts from from Dinuclear Dinuclear Complexes Complexes In contrastIn contrast to the to thechemistry chemistry of palladium, of palladium, there there have have been been few few reports reports of of mononuclear mononuclear Ni complexesNi complexes generated generated from the from corresponding the corresponding Ni(I) Ni(I)dimers dimers playing playing a key a keyrole rolein catalysis. in catalysis. Some catalyticSome reactions catalytic involvingreactions involving dinuclear dinuclear Ni(I) hydride Ni(I) hydride complexes complexes have have been been studied, studied, along along with with their mechanisms,their mechanisms, and mononuclear and mononuclear active activenickel nickelcatalysts catalysts were were proposed. proposed. Abu-Omar Abu-Omar reported reported the catalyticthe catalyticdehydrocoupling dehydrocoupling of orga ofnohydrosilanes, organohydrosilanes, and and also also isolated isolated thethe mononuclearmononuclear Ni(II) Ni(II) silyl silyl complex formed by the reaction of [Ni(dippe)(µ-H)]2 (11) (dippe = 1,2-diisopropylphosphinoethane) complex formed by the reaction of [Ni(dippe)(μ-H)]2 (11) (dippe = 1,2-diisopropylphosphinoethane) with HSiCl3 [76]. Interestingly, the dinickel hydride complex reacts with chlorobenzene to form a Ni(II) with HSiCl3 [76]. Interestingly, the dinickel hydride complex reacts with chlorobenzene to form a oxidative addition product, via generation of 2 equiv. of Ni(0) accompanied by evolution of hydrogen Ni(II)gas. oxidative Vicic et addition al. also performed product, avia stoichiometric generation reactionof 2 equiv. using of Ni(0) Ni(0) complexes accompanied generated by evolution from the of hydrogensame gas. Ni(I) Vicic hydride et al. complex also performed (11), involving a stoichiometric the oxidative reaction addition ofusing aryl Ni(0) halides, complexes transmetallation generated from withthe alkylsilane,same Ni(I) and hydride reductive complex elimination (11 of), alkylareneinvolving (Schemethe oxidative 10)[77]. Therefore,addition of Ni(I) aryl hydride halides, transmetallationdimers are a goodwith precursor alkylsilane, to generate and highlyreductiv activee elimination Ni(0) catalysts of in alkylarene situ. (Scheme 10) [77]. Therefore, Ni(I) hydride dimers are a good precursor to generate highly active Ni(0) catalysts in situ.

Scheme 10. Stoichiometric reactions of dinuclear Ni(I) hydride complexes.

Molecules 2018, 23, 140 11 of 21

Scheme 9. Examples of catalytic processes using dinuclear Ni complexes, in which the complexes were proposed to retain their dinuclear frameworks.

3.2.1. Mononuclear Nickel Catalysts from Dinuclear Complexes In contrast to the chemistry of palladium, there have been few reports of mononuclear Ni complexes generated from the corresponding Ni(I) dimers playing a key role in catalysis. Some catalytic reactions involving dinuclear Ni(I) hydride complexes have been studied, along with their mechanisms, and mononuclear active nickel catalysts were proposed. Abu-Omar reported the catalytic dehydrocoupling of organohydrosilanes, and also isolated the mononuclear Ni(II) silyl complex formed by the reaction of [Ni(dippe)(μ-H)]2 (11) (dippe = 1,2-diisopropylphosphinoethane) with HSiCl3 [76]. Interestingly, the dinickel hydride complex reacts with chlorobenzene to form a Ni(II) oxidative addition product, via generation of 2 equiv. of Ni(0) accompanied by evolution of hydrogen gas. Vicic et al. also performed a stoichiometric reaction using Ni(0) complexes generated from the same Ni(I) hydride complex (11), involving the oxidative addition of aryl halides, transmetallation with alkylsilane, and reductive elimination of alkylarene (Scheme 10) [77]. Molecules 2018, 23, 140 13 of 21 Therefore, Ni(I) hydride dimers are a good precursor to generate highly active Ni(0) catalysts in situ.

Molecules 2018, 23, 140 12 of 21 SchemeScheme 10. Stoichiometric 10. Stoichiometric reactions reactions of dinucleardinuclear Ni(I) Ni(I) hydride hydride complexes. complexes. Chirik et al. reported the use of the α-diimine-chelated nickel(I) hydride dimer 12 as a catalyst α precursor forChirik the efficient et al. reported catalytic the usehydrosilylation of the -diimine-chelated of terminal nickel(I) aliphati hydridec alkenes dimer in an12 anti-Markovnikovas a catalyst precursor for the efﬁcient catalytic hydrosilylation of terminal aliphatic alkenes in an anti-Markovnikov manner [39]. Interestingly, in contrast to the electronic structure of the diphosphine analogue, the manner [39]. Interestingly, in contrast to the electronic structure of the diphosphine analogue, the dimer dimer was representedrepresented as as a a complex complex with with two two Ni(II) Ni(II) centers centers with with two two 1e reduced 1e reducedα-diimine α-diimine ligands. ligands. Therefore,Therefore, the dimer the dimer waswas proposed proposed to to form mononuclearmononuclear Ni(II) Ni(II) monohydride. monohydride. DFT DFT calculations calculations indicatedindicated that smooth that smooth insertion insertion of ofthe the alkene alkene into into thethe Ni–HNi–H bond bond can can form form monomeric monomeric Ni(II) Ni(II) alkyl alkyl complexcomplex (Scheme (Scheme 11). 11).

Scheme 11. Proposed mechanism for the hydrosilylation of 1-octene with HSi(OEt)3. The α-diimine Scheme 11. Proposed mechanism for the hydrosilylation of 1-octene with HSi(OEt)3. The α-diimine ligandsligands are redox-active, are redox-active, and and 1e 1ereduction reduction of of the the α-diimine forms forms Ni(II) Ni(II) centers. centers.

HazariHazari et al. etreported al. reported the thesynthesis synthesis of of dinuclear dinuclear Ni(I)Ni(I) complexes complexes bearing bearingbridging bridging cyclopentadienyl cyclopentadienyl and indenyland indenyl ligands, ligands, [Ni(NHC)] [Ni(NHC)]2(μ2-Cl)((µ-Cl)(μµ-Cp)-Cp) ( (14a)) andand [Ni(NHC)][Ni(NHC)]2(µ-Cl)(2(μ-Cl)(µ-Ind)μ-Ind) (14b )(14b (Cp) =(Cp = cyclopentadienylcyclopentadienyl and Ind and = Ind indenyl) = indenyl) [44]. [ 44The]. dinuclear The dinuclear Cp Cpcomplex complex 14a14a undergoesundergoes homolytic homolytic Ni–Ni 5 bond cleavageNi–Ni bond and cleavage exists in and an existsequilibrium in an equilibrium between betweenmononuclear mononuclear Ni(NHC)( Ni(NHC)(η5-Cp)η and-Cp) the and dichloride the 13a Sigman’sdichloride dimer Sigman’s13a at room dimer temperature;at room temperature; however, however, a similar a similarequilibrium equilibrium was was not not observed observed for the for the bridging indenyl analogue 14b. This suggests that the Cp ligand stabilizes the mononuclear bridging indenyl analogue 14b. This suggests that the Cp ligand stabilizes the mononuclear Ni(I), but Ni(I), but the indenyl ligand does not (Scheme 12). Interestingly, the Cp complexes mediated the the indenylSuzuki–Miyaura ligand does cross not coupling (Scheme of 4-chlorotoluene, 12). Interestingly, but the dinuclear the Cp indenylcomplexes complex mediated did not, strongly the Suzuki– Miyaurasuggesting cross coupling that only theof mononuclear4-chlorotoluene, Ni(I) complex but the has dinuclear activity toward indenyl this catalyticcomplex transformation. did not, strongly suggestingMatsubara that only the et al. mononuclear found that Sigman’s Ni(I) complex dimer 13a hasis in activity equilibrium toward with this the catalytic stable mononuclear transformation. MatsubaraNi(I) complex et al. in found the presence that Sigman’s of an appropriate dimer ligand,13a is in such equilibrium as phosphine, with phosphite, the stableor pyridinemononuclear Ni(I) complex(Scheme 13in)[ the78 ,79presence]. They of also an reported appropriate catalytic ligand, reactions such using as phosphine, these mononuclear phosphite, complexes or pyridine in the presence of an excess amount of the ligands, such as Kumada–Tamao–Corriu cross (Scheme 13) [78,79]. They also reported catalytic reactions using these mononuclear complexes in the coupling, Suzuki–Miyaura cross coupling, and Buchwald–Hartwig amination of aryl halide. presenceNotably, of an theexcess mononuclear amount ofNi(I) the complexes ligands, such bearing as aKumada–Tamao–Corriu bulky NHC were thermally cross stable coupling, even in the Suzuki– Miyauraabsence cross ofcoupling, a redox-active and ligand.Buchwald–Hartwig amination of aryl halide. Notably, the mononuclear Ni(I) complexes bearing a bulky NHC were thermally stable even in the absence of a redox-active ligand.

Scheme 12. Thermal equilibrium between the dinuclear Ni(I) Cp complex and the catalytically active mononuclear Ni(I) Cp complex. This equilibrium does not occur for the indenyl analogue.

Scheme 13. Equilibrium between dinuclear and mononuclear Ni(I) complexes in the presence of a ligand (L). Both complexes are active for cross-coupling reactions.

Molecules 2018, 23, 140 12 of 21 Molecules 2018, 23, 140 12 of 21 Chirik et al. reported the use of the α-diimine-chelated nickel(I) hydride dimer 12 as a catalyst precursorChirik for et the al. efficientreported catalytic the use hydrosilylation of the α-diimine-chelated of terminal aliphatinickel(I)c alkeneshydride in dimer an anti-Markovnikov 12 as a catalyst precursormanner [39]. for theInterestingly, efficient catalytic in cont hydrosilylationrast to the electronic of terminal structure aliphati ofc alkenesthe diphosphine in an anti-Markovnikov analogue, the mannerdimer was [39]. represented Interestingly, as a incomplex contrast with to thetwo electronic Ni(II) centers structure with twoof the 1e reduceddiphosphine α-diimine analogue, ligands. the dimerTherefore, was representedthe dimer was as a proposedcomplex withto form two Ni(II)mono nuclearcenters withNi(II) two monohydr 1e reducedide. α DFT-diimine calculations ligands. Therefore,indicated that the smoothdimer wasinsertion proposed of the to alkene form intomono thenuclear Ni–H Ni(II)bond canmonohydr form monomericide. DFT calculations Ni(II) alkyl indicatedcomplex (Scheme that smooth 11). insertion of the alkene into the Ni–H bond can form monomeric Ni(II) alkyl complex (Scheme 11).

Scheme 11. Proposed mechanism for the hydrosilylation of 1-octene with HSi(OEt)3. The α-diimine Schemeligands are11. redox-active,Proposed mechanism and 1e reduction for the hydrosily of the α-diiminelation of forms 1-octene Ni(II) with centers. HSi(OEt) 3. The α-diimine ligands are redox-active, and 1e reduction of the α-diimine forms Ni(II) centers. Hazari et al. reported the synthesis of dinuclear Ni(I) complexes bearing bridging cyclopentadienyl and Hazariindenyl et al.ligands, reported [Ni(NHC)] the synthesis2(μ-Cl)( of dinuclearμ-Cp) ( 14aNi(I)) andcomplexes [Ni(NHC)] bearing2(μ br-Cl)(idgingμ-Ind) cyclopentadienyl (14b) (Cp = andcyclopentadienyl indenyl ligands, and Ind [Ni(NHC)] = indenyl)2(μ [44].-Cl)( Theμ-Cp) dinuclear (14a) andCp complex [Ni(NHC)] 14a 2undergoes(μ-Cl)(μ-Ind) homolytic (14b) (CpNi–Ni = cyclopentadienylbond cleavage and and exists Ind in = anindenyl) equilibrium [44]. The between dinuclear mononuclear Cp complex Ni(NHC)( 14a undergoesη5-Cp) and homolytic the dichloride Ni–Ni bondSigman’s cleavage dimer and 13a exists at room in an temperature; equilibrium betweenhowever, mononuclear a similar equilibrium Ni(NHC)( wasη5-Cp) not and observed the dichloride for the Sigman’sbridging indenyldimer 13a analogue at room 14b temperature;. This suggests however, that the a Cp sim ligandilar equilibrium stabilizes the was mononuclear not observed Ni(I), for butthe bridgingthe indenyl indenyl ligand analogue does not 14b (Scheme. This suggests 12). Intere that stingly,the Cp ligand the Cp stabilizes complexes the mononuclearmediated the Ni(I), Suzuki– but theMiyaura indenyl cross ligand coupling does ofnot 4-chlorotoluene, (Scheme 12). Intere but stingly,the dinuclear the Cp indenyl complexes complex mediated did not,the stronglySuzuki– Miyaurasuggesting cross that couplingonly the mononuclear of 4-chlorotoluene, Ni(I) complex but the has dinuclear activity towardindenyl this complex catalytic did transformation. not, strongly suggestingMatsubara that onlyet al. the found mononuclear that Sigman’s Ni(I) dimercomplex 13a has is activityin equilibrium toward withthis catalytic the stable transformation. mononuclear Ni(I)Matsubara complex in et the al. presencefound that of Sigman’san appropriate dimer li13agand, is in such equilibrium as phosphine, with the phosphite, stable mononuclear or pyridine Ni(I)(Scheme complex 13) [78,79]. in the They presence also reportedof an appropriate catalytic reactions ligand, such using as these phosphine, mononuclear phosphite, complexes or pyridine in the (Schemepresence 13)of an [78,79]. excess They amount also of reported the ligands, catalytic such reactions as Kumada–Tamao–Corriu using these mononuclear cross coupling, complexes Suzuki– in the presenceMiyaura crossof an excesscoupling, amount and Buchwald–Hartwigof the ligands, such aminas Kumada–Tamao–Corriuation of aryl halide. Notably, cross coupling, the mononuclear Suzuki– Molecules 2018, 23, 140 14 of 21 MiyauraNi(I) complexes cross coupling, bearing a and bulky Buchwald–Hartwig NHC were thermally amin stableation even of aryl in the halide. absence Notably, of a redox-active the mononuclear ligand. Ni(I) complexes bearing a bulky NHC were thermally stable even in the absence of a redox-active ligand.

SchemeScheme 12. Thermal 12. Thermal equilibrium equilibrium between between the the dinuclear dinuclear Ni(I) Ni(I) Cp Cp complex complex and and the catalyticallythe catalytically active active Schememononuclearmononuclear 12. Thermal Ni(I) Ni(I)Cp equilibrium complex. Cp complex. This between This equilibrium equilibrium the dinuclear does notnot Ni occur (I)occur Cp for forcomplex the the indenyl indenyl and analogue. the analogue. catalytically active mononuclear Ni(I) Cp complex. This equilibrium does not occur for the indenyl analogue.

Scheme 13. Equilibrium between dinuclear and mononuclear Ni(I) complexes in the presence of a Schemeligand (L). 13. Both Equilibrium complexes between are active dinuclear for cross-coupling and mononuclear reactions. Ni(I) complexes in the presence of a Molecules 2018Scheme, 23, 140 13. Equilibrium between dinuclear and mononuclear Ni(I) complexes in the presence of a 13 of 21 ligand ligand(L). Both (L). Bothcomplexes complexes are are active active for for cross-coupling cross-coupling reactions. reactions. 3.2.2. Proposed Mechanisms Involving Dinuclear Nickel Catalysts 3.2.2. Proposed Mechanisms Involving Dinuclear Nickel Catalysts In 2009,In 2009, Hillhouse Hillhouse et etal. al. reported reported a a dinuclear dinuclear Ni(I) complexcomplex bearing bearing a bridginga bridging imide imide ligand ligand as a as a productproduct of the of the reaction reaction of ofSigman’s Sigman’s dimer, dimer, [Ni(IPr)( [Ni(IPr)(µμ-Cl)]-Cl)]22,, withwith mesityl mesityl azide azide (N (N3Mes).3Mes). Subsequently, Subsequently, the thecoupling coupling reaction reaction of ofthe the imide imide ligand ligand with with isonitrile isonitrile gavegave isocyanate isocyanate derivatives derivatives accompanied accompanied by by the regenerationthe regeneration of ofthe the starting starting Sigman’s Sigman’s dimer dimer (Sch (Schemeeme 14 14))[ 41[41].]. This This reaction reaction sequence sequence can can be appliedbe applied to catalyticto catalytic nitrene nitrene transfer transfer reactions reactions of of N N33MesMes and isonitrile.isonitrile. It It was was unclear unclear whether whether mononuclear mononuclear speciesspecies were were generated generated or or not not in in this this reaction reaction sequence.

SchemeScheme 14. 14.ProposedProposed mechanism mechanism for for the the catalytic catalytic nitrenenitrene transfertransfer reaction reaction of of azide, azide, based based on on stoichiometricstoichiometric reactions. reactions.

Recently,Recently, Matsubara Matsubara et et al. al. reported reported that that Sigman’s Sigman’s dimerdimer13a 13aacts acts as as a highlya highly efﬁcient efficient catalyst catalyst in in the theKumada–Tamao–Corriu Kumada–Tamao–Corriu cross cross coupling coupling of of aryl halideshalides [ 45[45].]. Generally, Generally, a Ni(II)a Ni(II) oxidative oxidative addition addition productproduct is obtained is obtained from from the the reaction reaction of aa Ni(0)Ni(0) precursor precursor with with aryl aryl chloride. chloride. However, However, the reaction the reaction of of arylaryl chloride withwith Ni(0)(cod) Ni(0)(cod)2 in2 thein the presence presence of 1 equiv.of 1 equiv. of the bulkyof the IPr bulky ligand IPr unexpectedly ligand unexpectedly yielded yieldedthe dinuclear the dinuclear Ni(I) adduct Ni(I) 15adductrather 15 than rather mononuclear than mononuclear Ni(II) (Scheme Ni(II) 16). This (Scheme dinuclear 16). Ni(I) Thisµ -dinuclearσ-aryl Ni(I) μ-σ-aryl complex 15 was also active in the Kumada coupling, and can alternatively be prepared by the reaction of Sigman’s dimer 13a with phenylmagnesium chloride via transmetallation. Further stoichiometric reaction of the μ-σ-aryl complex 15 with aryl chloride afforded the biaryl as the cross- coupling product, accompanied by regeneration of Sigman’s dimer 13a. Therefore, the authors suspected that both dinuclear Ni(I) complexes were involved in the catalytic cycle. However, because the distribution of the biaryl products from the dinickel(I) μ-σ-aryl complex in the stoichiometric reaction was not consistent with that in the catalysis, an alternative pathway involving the oxidative addition of aryl halide to Sigman’s dimer 13a was proposed (Scheme 15), in which the μ-σ-aryl complex 15 could act as a precursor in catalysis. DFT calculations indicated that the addition of chlorobenzene to the dimer occurs without loss of the dinuclear framework, resulting in a semi-stable dinuclear Ni(II) complex similar to that reported for the palladium analogue [72]. Subsequent transmetallation with phenylmagnesium chloride forms the biaryl and Sigman’s dimer. Notably, the Ni(II)–Ni(0) complex with a dative bond from Ni(0) to Ni(II) is temporally formed just before the oxidative addition to one of the nickel centers to give the Ni(II)–Ni(II) complex. This catalytic reaction afforded no homo-coupling product, probably because the equilibrium of homolysis into mononuclear intermediates was not possible, which prevented the intermolecular aryl exchange process. This fact also strongly supports the existence of the dinuclear system during catalysis. Schoenebeck et al. also reported a theoretical study on the oxidative addition of iodobenzene to the Ni(I) iodide dimer [Ni(SIPr)]2(μ-I)2, which is analogous to Sigman’s dimer 13a. The results indicated that this reaction occurred via a pathway similar to Matsubara’s process [80], involving the efficient catalytic substitution of aryl iodide with Me4NSeCF3 to yield trifluoromethylselenoarene. Similarly, the biaryl homo-coupling product did not form in this transformation, and no signals were assigned as mononuclear Ni(I) species in the EPR spectra, suggesting that the existence of side reactions involving mononuclear Ni(I) species can be ruled out. The authors isolated the diselenide Ni(I) dimer, [Ni(SIPr)(μ-SeCF3)]2, as a possible intermediate complex. In contrast to the selectivity of

Molecules 2018, 23, 140 15 of 21

complex 15 was also active in the Kumada coupling, and can alternatively be prepared by the reaction of Sigman’s dimer 13a with phenylmagnesium chloride via transmetallation. Further stoichiometric reaction of the µ-σ-aryl complex 15 with aryl chloride afforded the biaryl as the cross-coupling product, accompanied by regeneration of Sigman’s dimer 13a. Therefore, the authors suspected that both dinuclear Ni(I) complexes were involved in the catalytic cycle. However, because the distribution of the biaryl products from the dinickel(I) µ-σ-aryl complex in the stoichiometric reaction was not consistent with that in the catalysis, an alternative pathway involving the oxidative addition of aryl halide to Sigman’s dimer 13a was proposed (Scheme 15), in which the µ-σ-aryl complex 15 could act as a precursor in catalysis. DFT calculations indicated that the addition of chlorobenzene to the dimer occurs without loss of the dinuclear framework, resulting in a semi-stable dinuclear Ni(II) complex similar to that reported for the palladium analogue [72]. Subsequent transmetallation with phenylmagnesium chloride forms the biaryl and Sigman’s dimer. Notably, the Ni(II)–Ni(0) complex with a dative bond from Ni(0) to Ni(II) is temporally formed just before the oxidative addition to one of the nickel centers to give the Ni(II)–Ni(II) complex. This catalytic reaction afforded no homo-coupling product, probably because the equilibrium of homolysis into mononuclear intermediates was not possible, which prevented the intermolecular aryl exchange process. This fact also strongly supports the existence of the dinuclear system during catalysis. Schoenebeck et al. also reported a theoretical study on the oxidative addition of iodobenzene to the Ni(I) iodide dimer [Ni(SIPr)]2(µ-I)2, which is analogous to Sigman’s dimer 13a. The results indicated that this reaction occurred via a pathway similar to Matsubara’s process [80], involving the efﬁcient catalytic substitution of aryl iodide with Me4NSeCF3 to yield triﬂuoromethylselenoarene. Similarly, the biaryl homo-coupling product did not form in this transformation, and no signals were assigned as mononuclear Ni(I) species in the EPR spectra, suggesting that the existence of side reactions Moleculesinvolving 2018, 23 mononuclear, 140 Ni(I) species can be ruled out. The authors isolated the diselenide Ni(I) dimer,14 of 21 [Ni(SIPr)(µ-SeCF3)]2, as a possible intermediate complex. In contrast to the selectivity of mononuclear mononuclearNi(0) catalysts, Ni(0) DFT catalysts, calculations DFT supportedcalculations the chemoselectivitysupported the chemoselectivity of the oxidative addition of the ofoxidative aryl additioniodide of rather aryl iodide than aryl rather selenide than to aryl dinuclear selenide Ni(I) to (Scheme dinuclear 16 ).Ni(I) (Scheme 16).

Scheme 15. Proposed mechanism of the Kumada–Tamao–Corriu coupling of aryl halide on a dinuclear Scheme 15. Proposed mechanism of the Kumada–Tamao–Corriu coupling of aryl halide on a dinuclear Ni(I)Ni(I) system, system, supported supported by by DFT DFT calculations. calculations.

Scheme 16. Selectivity divergence between Ni(0) and Ni(I)–Ni(I) in the trifluoromethylselenolation of aryl iodide. The calculated transition states for the oxidative addition of the Ph–X bond to Ni(0) and Ni(I)–Ni(I) showed different favored substituents X, supporting the chemoselectivity of the catalysis.

In sharp contrast to other dinuclear Ni(I) systems, the dinickel(I) complex reported by Uyeda et al. (iPrNDI)Ni2Br2 (21) (iPrNDI = 2,6-diisopropylphenyl-substituted naphthyridine diimine) has two unique features: (1) The positions of the two nickel centers are fixed by the tetradentate π-conjugated auxiliary ligand; and (2) the NDI ligand is redox-active and has two nickel centers, enabling reversible oxidation and reduction with intersystem spin crossing [55]. Moreover, using analogous Ni(0) benzene complex, several catalytic reactions, such as the hydrosilylation and cyclotrimerization of alkynes, were performed on the dinickel centers without breakage of the dinuclear framework. Detailed theoretical studies were conducted by Uyeda and Ess et al. to reveal the mechanism of catalytic alkyne cyclotrimerization (Scheme 17) [81]. Interestingly, oxidative C–C bond coupling of terminal alkynes occurs on the dinickel centers, and migratory insertion of the third alkyne provides chemoselective formation of benzene derivatives, while the reaction occurs via another pathway in the mononuclear catalyst system to provide different product selectivity. Catalytic reactions using dinuclear nickel(I) complexes are summarized. Although there are very few examples, both mechanisms involving highly active mononuclear complex and intact dinuclear complex as key intermediates are proposed in the efficient catalysis, as in the palladium chemistry. Dinuclear μ-hydride complexes can generate coordinatively unsaturated mononuclear Ni(0) complexes irreversibly as a result of elimination of hydrogen gas, whereas simple homolytic cleavage of the Ni–Ni bond occurs in dinuclear complexes bearing other bridging ligands to form monomeric Ni(I) complexes. Although such mononuclear Ni(I) complexes are proposed as key intermediates in catalysis, it should be noted that direct observation of the Ni(I) complexes as key intermediates has not been reported until now. On the other hand, theoretical calculation is a powerful method to

Molecules 2018, 23, 140 14 of 21 mononuclear Ni(0) catalysts, DFT calculations supported the chemoselectivity of the oxidative addition of aryl iodide rather than aryl selenide to dinuclear Ni(I) (Scheme 16).

Scheme 15. Proposed mechanism of the Kumada–Tamao–Corriu coupling of aryl halide on a dinuclear Molecules 2018, 23, 140 16 of 21 Ni(I) system, supported by DFT calculations.

Scheme 16.Scheme Selectivity 16. Selectivity divergence divergence between between Ni(0) Ni(0) and and Ni(I)–Ni(I)Ni(I)–Ni(I) in the intriﬂuoromethylselenolation the trifluoromethylselenolation of of aryl iodide.aryl The iodide. calculated The calculated transition transition states states for for th thee oxidative addition addition of the of Ph–X the bond Ph–X to Ni(0)bond and to Ni(0) and Ni(I)–Ni(I) showed different favored substituents X, supporting the chemoselectivity of the catalysis. Ni(I)–Ni(I) showed different favored substituents X, supporting the chemoselectivity of the catalysis.

In sharp contrast to other dinuclear Ni(I) systems, the dinickel(I) complex reported by Uyeda et al. iPr iPr In sharp( NDI)Nicontrast2Br to2 ( 21other)( NDIdinuclear = 2,6-diisopropylphenyl-substituted Ni(I) systems, the dinickel(I) naphthyridine complex diimine) reported has by two Uyeda et al. (iPrNDI)Ni2Brunique2 (21 features:) (iPrNDI (1) The = positions2,6-diisopropylphenyl-substituted of the two nickel centers are fixed by naphthyridine the tetradentate π-conjugated diimine) has two unique features:auxiliary (1) ligand; The andpositions (2) the NDI of the ligand two is redox-active nickel centers and has are two fixed nickel by centers, the tetradentate enabling reversible π-conjugated oxidation and reduction with intersystem spin crossing [55]. Moreover, using analogous Ni(0) auxiliary ligand;benzene and complex, (2) the several NDI catalyticligand is reactions, redox-active such as and the hydrosilylationhas two nickel and centers, cyclotrimerization enabling reversible oxidation andof alkynes, reduction were with performed intersystem on the dinickel spin cro centersssing without [55]. breakageMoreover, of the using dinuclear analogous framework. Ni(0) benzene complex, severalDetailed catalytic theoretical reactions,studies were such conducted as the hy bydrosilylation Uyeda and Ess and et al. cyclotrimerization to reveal the mechanism of alkynes, of were performed catalyticon the dinickel alkyne cyclotrimerization centers without (Scheme breakage 17)[81]. of Interestingly, the dinuclear oxidative framework. C–C bond couplingDetailed of theoretical terminal alkynes occurs on the dinickel centers, and migratory insertion of the third alkyne provides studies werechemoselective conducted formation by Uyeda of benzene and derivatives, Ess et whileal. to the reactionreveal occursthe mechanism via another pathway of catalytic in the alkyne cyclotrimerizationmononuclear (Scheme catalyst 17) system [81] to. provideInterestingly, different product oxidative selectivity. C–C bond coupling of terminal alkynes occurs on the dinickelCatalytic reactions centers, using and dinuclear migratory nickel(I) inse complexesrtion of are the summarized. third alkyne Although provides there are chemoselective very few examples, both mechanisms involving highly active mononuclear complex and intact dinuclear formation of benzene derivatives, while the reaction occurs via another pathway in the mononuclear complex as key intermediates are proposed in the efficient catalysis, as in the palladium chemistry. catalyst systemDinuclear to provideµ-hydride different complexes canproduct generate selectivity. coordinatively unsaturated mononuclear Ni(0) complexes Catalyticirreversibly reactions as a using result ofdinuclear elimination nickel(I) of hydrogen comp gas,lexes whereas are simplesummarized. homolytic Although cleavage of there the are very few examples,Ni–Ni both bond mechanisms occurs in dinuclear involving complexes highly bearing active other bridging mononuclear ligands to complex form monomeric and intact Ni(I) dinuclear complexes. Although such mononuclear Ni(I) complexes are proposed as key intermediates in catalysis, complex asit key should intermediates be noted that direct are observationproposed ofin the the Ni(I) efficient complexes catalysis, as key intermediates as in the palladium has not been chemistry. Dinuclear μreported-hydride until complexes now. On the can other generate hand, theoretical coordinatively calculation unsaturated is a powerful mononuclear method to propose Ni(0) the complexes irreversiblyintact as a dinuclear result activeof elimination complexes involved of hydrogen in the catalytic gas, whereas cycles. Finally, simple appropriate homolytic design ofcleavage the of the Ni–Ni bondbridging occurs ligands, in dinuclear naphthyridine complexes diamine, can bearing form dinuclear other Nibridging complex, whichligands has ato robust form bimetallic monomeric Ni(I) framework. A series of catalytic transformations with specific chemoselectivity clearly demonstrated complexes.the Although importance such of the chemistrymononuclear of dinuclear Ni(I) Ni(I) complexes complexes as are catalysts. proposed as key intermediates in catalysis, it should be noted that direct observation of the Ni(I) complexes as key intermediates has not been reported until now. On the other hand, theoretical calculation is a powerful method to

Molecules 2018, 23, 140 15 of 21 propose the intact dinuclear active complexes involved in the catalytic cycles. Finally, appropriate design of the bridging ligands, naphthyridine diamine, can form dinuclear Ni complex, which has a robustMolecules bimetallic2018, 23, 140framework. A series of catalytic transformations with specific chemoselectivity17 of 21 clearly demonstrated the importance of the chemistry of dinuclear Ni(I) complexes as catalysts.

SchemeScheme 17. 17.ProposedProposed system system for for the the cyclotrimerization cyclotrimerization ofof terminalterminal alkynes alkynes via via oxidative oxidative C–C C–C bond bond couplingcoupling and and migratory migratory insertion insertion of of the the third third alkynealkyne onon thethe dinickel(I) dinickel(I) centers, centers, which which provides provides chemoselectivechemoselective formation formation of of the the benzene benzene derivatives. derivatives.

4. Conclusion4. Conclusion and and Perspectives Perspectives In thisIn this mini-review, mini-review, thethe recentlyrecently reported reported structures, structures, catalytic catalytic applications, applications, and mechanistic and mechanistic details detailsof dinuclear of dinuclear Pd(I) Pd(I) and Ni(I) and complexesNi(I) complexes are surveyed. are surveyed. As noted As in thenoted Introduction, in the Introduction, highly efficient highly efficientcatalytic catalytic transformations transformations have been have achieved been achi undereved mild under conditions mild usingconditions these complexesusing these as catalystcomplexes as catalystprecursors. precursors. However, However, these involve these several involve different several activation different processes, activation depending processes, on depending the metal on the atoms,metal ligandatoms, structures, ligand structures, and kinds ofand reactions kinds employed.of reactions One employed. of the most One interesting of the pointsmost ofinteresting these pointsactivation of these processes activation is thatprocesses pairs of is active that pairs metal of sites active were metal able tosites activate were substrates able to activate in a concerted substrates fashion while retaining their dinuclear frameworks. For instance, the oxidative addition of aryl halide in a concerted fashion while retaining their dinuclear frameworks. For instance, the oxidative to M(I)–M(I) (M = Pd and Ni) generates not the mononuclear M(III) species, but instead the dinuclear addition of aryl halide to M(I)–M(I) (M = Pd and Ni) generates not the mononuclear M(III) species, adduct, a M(II)-M(II) complex stabilized by bridging ligands. On the other hand, the in situ “reversible” but formationinstead the of dinuclear dinuclear complexesadduct, a M(II)-M(II) may contribute comple to thex stabilized stabilization by of bridging highly active ligands. mononuclear On the other hand,catalysts, the in situ which “reversible” are essential formation for efficient of catalyticdinuclea transformations.r complexes mayThe contribute achievement to the of a stabilization dissociation of highlyequilibrium active mononuclear between semi-stable catalysts, complexes which with are weak essential metal–metal for effi bondscient that catalytic act as “bimetallic transformations. dormant The achievementspecies” facilitates of a dissociation the formation equilibrium of highly activebetween mononuclear semi-stable complexes complexes without with using weak an metal–metal excess of bondsancillary that act ligands, as “bimetallic and moreover, dormant avoids species” the decomposition facilitates the of formation such active of complexes. highly active mononuclear complexesAlthough without the using above an chemistry excess of of ancillary the dinuclear ligands, metal and complexes moreover, in catalysis avoids has the been decomposition revealed to of suchhave active great complexes. promise for the development of various useful organic transformations, it is still only at the startingAlthough line. the The above concerted chemistry activation of the of substratesdinuclear andmetal the complexes development in ofcatalysis highly efficienthas been and/or revealed to havechemoselective great promise catalytic for the processes development show great of various potential useful to expand organic widely transformati in the fieldsons, ofit is organic still only at thesynthesis starting and line. applications. The concerted activation of substrates and the development of highly efficient and/orAcknowledgments: chemoselectiveThis catalytic work was processes supported byshow the Japan grea Societyt potential for the to Promotion expand ofwidely Science (Grant-in-Aidin the fields of organicfor Scientific synthesis Research and applications. 25410123). We thank Editage (www.editage.jp) for editing a draft of this manuscript. Author Contributions: The authors contributed equally: T.I., Y.K. and K.M. researched reports and wrote the review. Acknowledgments:Conflicts of Interest: ThisThe work authors was declaresupported no conflict by the of Japan interest. Society for the Promotion of Science (Grant-in-Aid for Scientific Research 25410123). We thank Editage (www.editage.jp) for editing a draft of this manuscript.

AuthorReferences Contributions: The authors contributed equally: T.I., Y.K. and K.M. researched reports and wrote the review.1. Tamaru, Y. Modern Organonickel Chemistry; Wiley-VCH: Weinheim, Germany, 2005. Conflicts2. Montgomery, of Interest: The J. Organonickel authors declare chemistry. no conflict In Organometallics of interest. in Synthesis: Fourth Manual; Lipshutz, B.H., Ed.; John Wiley & Sons, Inc.: Hoboken, NJ, USA, 2013; pp. 319–428.

3. Garduno, J.A.; Arevalo, A.; Garcia, J.J. Bond and small-molecule activation with low-valent nickel complexes. Dalton Trans. 2015, 44, 13419–13438. [CrossRef][PubMed]

Molecules 2018, 23, 140 18 of 21

4. Hoshimoto, Y.; Ohashi, M.; Ogoshi, S. Catalytic transformation of aldehydes with nickel complexes through η2 coordination and oxidative cyclization. Acc. Chem. Res. 2015, 48, 1746–1755. [CrossRef][PubMed] 5. Eberhardt, N.A.; Guan, H. Nickel hydride complexes. Chem. Rev. 2016, 116, 8373–8426. [CrossRef][PubMed] 6. Mu, H.; Pan, L.; Song, D.; Li, Y. Neutral nickel catalysts for oleﬁn homo- and copolymerization: Relationships between catalyst structures and catalytic properties. Chem. Rev. 2015, 115, 12091–12137. [CrossRef][PubMed] 7. Thakur, A.; Louie, J. Advances in nickel-catalyzed cycloaddition reactions to construct carbocycles and heterocycles. Acc. Chem. Res. 2015, 48, 2354–2365. [CrossRef][PubMed] 8. Kantchev, E.A.B.; O’Brien, C.J.; Organ, M.G. Palladium complexes of N-heterocyclic carbenes as catalysts for cross-coupling reactions–A synthetic chemist’s perspective. Angew. Chem. Int. Ed. 2007, 46, 2768–2813. [CrossRef][PubMed] 9. Musaev, D.G.; Figg, T.M.; Kaledin, A.L. Versatile reactivity of Pd-catalysts: Mechanistic features of the mono-N-protected amino acid ligand and cesium-halide base in Pd-catalyzed C-H bond functionalization. Chem. Soc. Rev. 2014, 43, 5009–5031. [CrossRef][PubMed] 10. He, J.; Wasa, M.; Chan, K.S.L.; Shao, Q.; Yu, J.Q. Palladium-catalyzed transformations of alkyl C-H bonds. Chem. Rev. 2017, 117, 8754–8786. [CrossRef][PubMed] 11. Lin, S.; Agapie, T. Cross-coupling chemistry at mononuclear and dinuclear nickel complexes. Synlett 2011, 1–5. [CrossRef] 12. Hazari, N.; Hruszkewycz, D.P. Dinuclear PdI complexes with bridging allyl and related ligands. Chem. Soc. Rev. 2016, 45, 2871–2899. [CrossRef][PubMed] 13. Paton, R.S.; Brown, J.M. Dinuclear Palladium complexes-precursors or catalysts? Angew. Chem. Int. Ed. 2012, 42, 10448–10450. [CrossRef][PubMed] 14. Zimmermann, P.; Limberg, C. Activation of small molecules at Ni(I) moieties. J. Am. Chem. Soc. 2017, 139, 4233–4242. [CrossRef][PubMed] 15. Bonney, K.J.; Schoenebeck, F. Experiment and computation: A combined approach to study the reactivity of palladium complexes in oxidation states 0 to IV. Chem. Soc. Rev. 2014, 43, 6609–6638. [CrossRef][PubMed] 16. Liu, Q.; Dong, X.; Li, J.; Xiao, J.; Dong, Y.; Liu, H. Recent advances on palladium radical involved reactions. ACS Catal. 2015, 5, 6111–6137. [CrossRef] 17. Dyson, P.J. Catalysis by low oxidation state transition metal (carbonyl) clusters. Coord. Chem. Rev. 2004, 248, 2443–2458. [CrossRef] 18. Takao, T.; Suzuki, H. Skeletal rearrangement of hydrocarbyl ligands on a triruthenium core induced by chemical oxidation. Coord. Chem. Rev. 2012, 256, 695–708. [CrossRef] 19. Murahashi, T.; Kurosawa, H. Organopalladium complexes containing palladium-palladium bonds. Coord. Chem. Rev. 2002, 231, 207–228. [CrossRef] 20. Weiss, M.; Peters, R. Bimetalllic catalysis: Cooperation of carbophilic metal centers. In Cooperative Catalysis; Peters, R., Ed.; Wiley-VCH: Weinheim, Germany, 2015; pp. 227–262. t 21. Vilar, R.; Mingos, D.M.P.; Cardin, C.J. Synthesis and structural characterisation of [Pd2(µ-Br)2(PBu 3)2], an example of a palladium(I)–palladium(I) dimer. J. Chem. Soc. Dalton Trans. 1996, 23, 4313–4314. [CrossRef] 22. Proutiere, F.; Lyngvi, E.; Auﬁero, M.; Sanhueza, I.A.; Schoenebeck, F. Combining the reactivity properties of

PCy3 and PtBu3 into a single ligand, P(iPr)(tBu)2. Reaction via mono- or bisphosphine palladium(0) centers and palladium(I) dimer formation. Organometallics 2014, 33, 6879–6884. [CrossRef] 23. Kullberg, M.L.; Kubiak, C.P. Organometallic chemistry in aqueous solution: Water-soluble bis(dimethylphosphino)methane-bridged dipalladium complexes. Organometallics 1984, 3, 632–634. [CrossRef] 24. Ogoshi, S.; Nishida, T.; Tsutsumi, K.; Ooi, M.; Shinagawa, T.; Akasaka, T.; Yamane, M.; Kurosawa, H. Carbon-carbon bond formation by electrophilic addition at the central carbon of the µ-η3-allenyl/propargyl ligand on the Pd-Pd bond. J. Am. Chem. Soc. 2001, 123, 3223–3228. [CrossRef][PubMed] 25. Murahashi, T.; Nagai, T.; Okuno, T.; Matsutani, T.; Kurosawa, H. Synthesis and ligand substitution reactions of a homoleptic acetonitrile dipalladium(I) complex. Chem. Commun. 2000, 17, 1689–1690. [CrossRef] 26. Christmann, U.; Vilar, R.; White, A.J.P.; Williams, D.J. Synthesis of two novel dinuclear palladium(I) complexes and studies of their catalytic activity in amination reactions. Chem. Commun. 2004, 11, 1294–1295. [CrossRef][PubMed] Molecules 2018, 23, 140 19 of 21

27. Barder, T.E. Synthesis, structural, and electron topographical analyses of a dialkylbiaryl phosphine/arene-ligated palladium(I) dimer: Enhanced reactivity in Suzuki-Miyaura coupling reactions. J. Am. Chem. Soc. 2006, 128, 898–904. [CrossRef][PubMed] 28. Weissman, H.; Shimon, L.J.W.; Milstein, D. Unsaturated Pd(0), Pd(I), and Pd(II) complexes of a new methoxy-substituted benzyl phosphine. Aryl-X (X = Cl, I) oxidative addition, C-O cleavage, and Suzuki-Miyaura coupling of aryl chlorides. Organometallics 2004, 23, 3931–3940. [CrossRef] 29. Denmark, S.E.; Baird, J.D. Palladium-catalyzed cross-coupling reactions of heterocyclic silanolates with substituted aryl iodides and bromides. Org. Lett. 2006, 8, 793–795. [CrossRef][PubMed] 30. Hruszkewycz, D.P.; Balcells, D.; Guard, L.M.; Hazari, N.; Tilset, M. Insight into the efﬁciency of cinnamyl-supported precatalysts for the Suzuki−Miyaura reaction: Observation of Pd(I) dimers with bridging allyl ligands during catalysis. J. Am. Chem. Soc. 2014, 136, 7300–7316. [CrossRef][PubMed] 31. Hruszkewycz, D.P.; Wu, J.; Hazari, N.; Incarvito, C.D. Palladium(I)-bridging allyl dimers for the catalytic

functionalization of CO2. J. Am. Chem. Soc. 2011, 133, 3280–3283. [CrossRef][PubMed] 32. Markert, C.; Neuburger, M.; Kulicke, K.; Meuwly, M.; Pfaltz, A. Palladium-catalyzed allylic substitution: Reversible formation of allyl-bridged dinuclear palladium(I) complexes. Angew. Chem. Int. Ed. 2007, 46, 5892–5895. [CrossRef][PubMed] 33. Fafard, C.M.; Adhikari, D.; Foxman, B.M.; Mindiola, D.J.; Ozerov, O.V. Addition of ammonia, water, and dihydrogen across a single Pd-Pd bond. J. Am. Chem. Soc. 2007, 129, 10318–10319. [CrossRef][PubMed] 34. Lin, C.-Y.; Power, P.P. Complexes of Ni(I): A “rare” oxidation state of growing importance. Chem. Soc. Rev. 2017, 46, 5347–5399. [CrossRef][PubMed] 35. Kieber-Emmons, M.T.; Riordan, C.G. Dioxygen activation at monovalent nickel. Acc. Chem. Res. 2007, 40, 618–625. [CrossRef][PubMed] 36. Jonas, K.; Wilke, G. Hydrogen bonds in a Ni-Ni system. Angew. Chem. Int. Ed. 1970, 9, 312–313. [CrossRef] 37. Bach, I.; Goddard, R.; Kopiske, C.; Seevogel, K.; Pörschke, K.R. Syntheisis, structure, and reactivity of t t t t ( Bu2PC2H4P Bu)Ni(CH3)2 and {( Bu2PC2H4P Bu)Ni}2(µ-H)2. Organometallics 1999, 18, 10–20. [CrossRef] 38. Vicic, D.A.; Jones, W.A. Evidence for the existence of a late-metal terminal sulﬁdo complex. J. Am. Chem. Soc. 1999, 121, 4070–4071. [CrossRef] 39. Pappas, I.; Treacy, S.; Chirik, P.J. Alkene Hydrosilylation using tertiary silanes with α-diimine nickel catalysts. Redox-active ligands promote a distinct mechanistic pathway from platinum catalysis. ACS Catal. 2016, 6, 4105–4109. [CrossRef] 40. Dible, B.R.; Sigman, M.S.; Arif, A.M. Oxygen-induced ligand dehydrogenation of a planar bis-í-chloronickel(I) dimer featuring an NHC ligand. Inorg. Chem. 2005, 44, 3774–3776. [CrossRef][PubMed] 41. Laskowski, C.A.; Hillhouse, G.L. Group-transfer reactions of Ni(II)-Ni(II) bridging imido complexes. catalytic formation of carbodiimides and isocyanates via nitrene transfer from organoazides. Organometallics 2009, 28, 6114–6120. [CrossRef] 42. Laskowski, C.A.; Hillhouse, G.L. Synthesis and carbene-transfer reactivity of dimeric nickel carbene cations supported by N-heterocyclic carbene ligands. Chem. Sci. 2011, 2, 321–325. [CrossRef] 43. Laskowski, C.A.; Bungum, D.J.; Baldwin, S.M.; Del Ciello, S.A.; Iluc, V.M.; Hillhouse, G.L. Synthesis and reactivity of two-coordinate Ni(I) alkyl and aryl complexes. J. Am. Chem. Soc. 2013, 135, 18272–18275. [CrossRef][PubMed] 44. Wu, J.; Nova, A.; Balcells, D.; Brudvig, G.W.; Dai, W.; Guard, L.M.; Hazari, N.; Lin, P.H.; Pokhrel, R.; Takase, M.K. Nickel(I) monomers and dimers with cyclopentadienyl and indenyl ligands. Chem. Eur. J. 2014, 20, 5327–5337. [CrossRef][PubMed] 45. Matsubara, K.; Yamamoto, H.; Miyazaki, S.; Inatomi, T.; Nonaka, K.; Koga, Y.; Yamada, Y.; Veiros, L.F.; Kirchner, K. Dinuclear systems in the efﬁcient nickel-catalyzed Kumada−Tamao−Corriu cross-coupling of aryl halides. Organometallics 2017, 36, 255–265. [CrossRef] 46. Ho, N.K.T.; Neumann, B.; Stammler, H.G.; Menedes da Silva, V.H.; Watanabe, D.G.; Braga, A.A.C.; Ghadwal, R.S. Nickel-catalysed direct C2-arylation of N-heterocyclic carbenes. Dalton Trans. 2017, 46, 12027–12031. [CrossRef][PubMed] 47. Ito, M.; Matsumoto, T.; Tatsumi, K. Synthesis and reactions of mono- and dinuclear Ni(I) thiolate complexes. Inorg. Chem. 2009, 48, 2215–2223. [CrossRef][PubMed] Molecules 2018, 23, 140 20 of 21

48. Olechnowicz, F.; Hillhouse, G.L.; Jordan, R.F. Synthesis and reactivity of NHC-supported 2 2 2 Ni2(µ -η ,η -S2)-bridging disulﬁde and Ni2(µ-S)2-bridging sulﬁde complexes. Inorg. Chem. 2015, 54, 2705–2712. [CrossRef][PubMed] 49. Olechnowicz, F.; Hillhouse, G.L.; Cundari, T.R.; Jordan, R.F. Heterolytic H-H and H-B bond cleavage reactions

of {(IPr)Ni(µ-S)}2. Inorg. Chem. 2017, 56, 9922–9930. [CrossRef][PubMed] 50. Chen, Y.; Sui-Seng, C.; Zargarian, D. Tetraphenylborate as a novel bridging ligand in a zwitterionic nickel(I) dimer. Angew. Chem. Int. Ed. 2005, 44, 7721–7725. [CrossRef][PubMed] 51. Beck, R.; Johnson, S.A. Dinuclear Ni(I)-Ni(I) complexes with syn-facial bridging ligands from Ni(I) precursors or Ni(II)/Ni(0) comproportionation. Organometallics 2013, 32, 2944–2951. [CrossRef] 52. Velian, A.; Lin, S.; Miller, A.J.M.; Day, M.W.; Agapie, T. Synthesis and C-C coupling reactivity of a dinuclear NiI-NiI complex supported by a terphenyl diphosphine. J. Am. Chem. Soc. 2010, 132, 6296–6297. [CrossRef] [PubMed] 53. Beck, R.; Johnson, S.A. Mechanistic implications of an asymmetric intermediate in catalytic C–C coupling by a dinuclear nickel complex. Chem. Commun. 2011, 47, 9233–9235. [CrossRef][PubMed] 54. Jones, C.; Schulten, C.; Fohlmeister, L.; Stasch, A.; Murray, K.S.; Moubaraki, B.; Kohl, S.; Ertem, M.Z.; Gagliardi, L.; Cramer, C.J. Bulky guanidinato nickel(I) complexes: Synthesis, characterization, isomerization, and reactivity studies. Chem. Eur. J. 2011, 17, 1294–1303. [CrossRef][PubMed] 55. Zhou, Y.Y.; Hartline, D.R.; Steiman, T.J.; Fanwick, P.E.; Uyeda, C. Dinuclear nickel complexes in five states of oxidation using a redox-active ligand. Inorg. Chem. 2014, 53, 11770–11777. [CrossRef][PubMed] 56. Uyeda, C.; Steiman, T.J.; Pal, S. Catalytically active nickel-nickel bonds using redox active ligands. Synlett 2016, 27, 814–820. [CrossRef] 57. Finke, A.D.; Elleby, E.C.; Boyd, M.J.; Weissman, H.; Moore, J.S. Zinc chloride-promoted aryl bromide-alkyne cross-coupling reactions at room temperature. J. Org. Chem. 2009, 74, 8897–8900. [CrossRef][PubMed] 2 2 58. Kalvet, I.; Magnin, G.; Schoenebeck, F. Rapid room-temperature, chemoselective Csp -Csp Coupling of Poly(pseudo)halogenated Arenes Enabled by Palladium(I) Catalysis in air. Angew. Chem. Int. Ed. 2017, 56, 1581–1585. [CrossRef][PubMed] 59. Krishna, H.; Krishnamurthy, S.S.; Nethaji, M.; Murugavel, R.; Prabusankar, G. Unusual reactivity of a i 3 sterically hindered diphosphazane ligand, EtN{P(OR)2}2, (R = C6H3(Pr )2-2,6) towards (η -allyl)palladium precursors. Dalton Trans. 2007, 27, 2908–2914. [CrossRef][PubMed] 60. Ogura, T.; Yoshida, K.; Yanagisawa, A.; Imamoto, T. Optically active dinuclear palladium complexes containing a Pd-Pd Bond: Preparation and enantioinduction ability in asymmetric ring-opening reactions. Org. Lett. 2009, 11, 2245–2248. [CrossRef][PubMed] 61. Murray, S.A.; Liang, M.Z.; Meek, S.J. Stereoselective tandem bis-electrophile couplings of dibolylmethane. J. Am. Chem. Soc. 2017, 39, 14061–14064. [CrossRef][PubMed] 62. Proutiere, F.; Aufiero, M.; Schoenebeck, F. Reactivity and stability of dinuclear Pd(I) complexes: Studies on the active catalytic species, insights into precatalyst activation and deactivation, and application in highly selective cross-coupling. J. Am. Chem. Soc. 2012, 134, 606–612. [CrossRef][PubMed] 63. Stambuli, J.P.; Kuwano, R.; Hartwig, J.F. Unparalleled rates for the activation of aryl chlorides and bromides: Coupling with amines and boronic acids in minutes at room temperature. Angew. Chem. Int. Ed. 2002, 41, 4746–4748. [CrossRef][PubMed] 64. Oldenhof, S.; Lutz, M.; de Bruin, B.; van der Vlugt, J.I.; Reek, J.N.H. Well-defined bisMETAMORPhos PdI−PdI complex: Synthesis, structural characterization, and reactivity. Organometallics 2014, 33, 7293–7298. [CrossRef] 65. Srinivas, V.; Balaraman, E.; Sajna, K.V.; Kumara, S.K.C. Catalyst-free and catalysed addition of P(O)–H bonds to allenyl/alkynyl-phosphonates and -phosphane oxides: Use of a robust, recoverable dinuclear palladium(I) catalyst. Eur. J. Org. Chem. 2011, 22, 4222–4230. [CrossRef] 66. Hill, L.L.; Crowell, J.L.; Tutwiler, S.L.; Massie, N.L.; Hines, C.C.; Griffin, S.T.; Rogers, R.D.; Shaughnessy, K.H.; Grasa, G.A.; Seechurn, C.C.C.J.; et al. Synthesis and X-ray structure determination of highly active Pd(II), Pd(I), and Pd(0) complexes of di(tert-butyl)neopentylphosphine (DTBNpP) in the arylation of amines and ketones. J. Org. Chem. 2010, 75, 6477–6488. [CrossRef][PubMed] 67. Jaworski, J.N.; McCann, S.D.; Guzei, I.A.; Stahl, S.S. Detection of palladium(I) in aerobic oxidation catalysis. Angew. Chem. Int. Ed. 2017, 56, 1–7. [CrossRef][PubMed] Molecules 2018, 23, 140 21 of 21

68. Sperger, T.; Stirner, C.K.; Schoenebeck, F. Bench-stable and recoverable palladium(I) dimer as an efﬁcient catalyst for Heck cross-coupling. Synthesis 2017, 49, 115–120. [CrossRef] 69. Das, R.K.; Saha, B.; Rahaman, S.M.W.; Bera, J.K. Bimetallic catalysis involving dipalladium(I) and diruthenium(I) complexes. Chem. Eur. J. 2010, 16, 14459–14468. [CrossRef][PubMed]

70. Wang, M.; Fan, T.; Lin, Z. DFT studies on the reaction of CO2 with allyl-bridged dinuclear palladium(I) complexes. Polyhedron 2012, 32, 35–40. [CrossRef]

71. Hruszkewycz, D.P.; Wu, J.; Green, J.C.; Hazari, N.; Schmeier, T.J. Mechanistic studies of the insertion of CO2 into Palladium(I) bridging allyl dimers. Organometallics 2012, 31, 470–485. [CrossRef] 72. Bonney, K.J.; Proutiere, F.; Schoenebeck, F. Dinuclear Pd(I) complexes—solely precatalysts? Demonstration of direct reactivity of a Pd(I) dimer with an aryl iodide. Chem. Sci. 2013, 4, 4434–4439. [CrossRef] 73. Kalvet, I.; Bonney, K.J.; Schoenebeck, F. Kinetic and computational studies on Pd(I) dimer-mediated halogen exchange of aryl iodides. J. Org. Chem. 2014, 79, 12041–12046. [CrossRef][PubMed] 74. Yin, G.; Kalvet, I.; Schoenebeck, F. Triﬂuoromethylthiolation of aryl iodides and bromides enabled by a bench-stable and easy-to-recover dinuclear palladium(I) catalyst. Angew. Chem. Int. Ed. 2015, 54, 6809–6813. [CrossRef][PubMed]

75. Aufiero, M.; Sperger, T.; Tsang, A.S.K.; Schoenebeck, F. Highly efficient C-SeCF3 coupling of aryl iodides enabled by an air-stable dinuclear PdI catalyst. Angew. Chem. Int. Ed. 2015, 54, 1–6. [CrossRef][PubMed] 76. Smith, E.E.; Du, G.; Fanwick, P.E.; Abu-Omar, M.M. Dehydrocoupling of organosilanes with a dinuclear nickel hydride catalyst and isolation of a nickel silyl complex. Organometallics 2010, 29, 6527–6533. [CrossRef] 77. Dubinina, G.G.; Brennessel, W.W.; Miller, J.L.; Vicic, D.A. Exploring trifluoromethylation reactions at nickel: A structural and reactivity study. Organometallics 2008, 27, 3933–3938. [CrossRef] 78. Matsubara, K.; Fukahori, Y.; Inatomi, T.; Tazaki, S.; Yamada, Y.; Koga, Y.; Kanegawa, S.; Nakamura, T. Monomeric three-coordinate N-heterocyclic carbene nickel(I) complexes: Synthesis, structures, and catalytic applications in cross-coupling reactions. Organometallics 2016, 35, 3281–3287. [CrossRef] 79. Nagao, S.; Matsumoto, T.; Koga, Y.; Matsubara, K. Monovalent nickel complex bearing a bulky N-heterocyclic carbene catalyzes BuchwaldHartwig amination of aryl halides under mild conditions. Chem. Lett. 2011, 40, 1036–1038. [CrossRef] 80. Dürr, A.B.; Fisher, H.C.; Kalvet, I.; Truong, K.N.; Schoenebeck, F. Divergent reactivity of a dinuclear (NHC)Nickel(I) catalyst versus nickel(0) enables chemoselective trifluoromethylselenolation. Angew. Chem. Int. Ed. 2017, 56, 13431–13435. [CrossRef][PubMed] 81. Kwon, D.H.; Proctor, M.; Mendoza, S.; Uyeda, C.; Ess, D.H. Catalytic dinuclear nickel spin crossover mechanism and selectivity for alkyne cyclotrimerization. ACS Catal. 2017, 7, 4796–4804. [CrossRef]

© 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).