Heavier tetrylene- and tetrylyne-transition metal chemistry: it's no carbon copy

	Fig. 6 Examples of addition cyclic silylene derivatives.

The 5-membered cyclic bis(alkyl)silylene 3.172 has been utilised in generating mono(silylene) Ni⁰ (3.174, 3.175, and 3.176), Pd⁰ (3.177 and 3.178), Pt⁰ (3.179, and 3.180) complexes, through direct reactions with M⁰ precursors (M = Ni–Pt), followed by ligand exchange processes (Scheme 39).¹³⁵ Due to the electron-deficient nature of the Si^II center in this silylene ligand compared to N-heterocyclic derivatives, significant backbonding from the M⁰ centers to Si^II is observed, resulting in contracted M–Si double bonds.

	Scheme 39 Synthesis of nickel(0), palladium(0), and platinum(0) complexes bearing a cyclic bis(alkyl)silylene ligand, with additional arene and phosphine ligands.

The 14-electron bis(silylene)palladium complex 3.181 was also obtained by combining two equivalents of 3.172 with [(Cy₃P)₂Pd], which demonstrated the ability to readily cleave H₂ at room temperature, yielding the Pd^II disilane complex 3.182 (Scheme 40).¹³⁶ Isoelectronic cationic Cu^I and Ag^I complexes were also synthesised by directly combining silylene 3.172 with cationic Cu and Ag fragments, resulting in 3.183 and 3.184.¹³⁷

	Scheme 40 Synthesis of palladium(0) and coinage metal complexes bearing a cyclic bis(alkyl)silylene ligand, and dihydrogen activation by the former.

Further examples of Ni⁰ and Pt⁰ complexes were reported (viz. 3.185, 3.186, and 3.187, Scheme 41), accessed by combining silylene 3.172 with [M₂(dvtms)₃], followed by ligand exchange with CO for the Ni system.^138,139 In regards to the cyclic (alkyl)(amido)silylene 3.173, one group 10 complex has been reported, namely the Pt⁰ bis(alkene) species 3.188, synthesised by combination of the silylene ligand with [Pt₂(dvtms)₃] (Scheme 42).¹⁴⁰ Additionally, coinage metal complexes have also been reported, whereby addition of two equiv. of the ligand to [MCl] (M = Cu, Ag, Au) led to insertion into the M–Cl bond by one silylene ligand, forming (silyl)(silylene) complexes 3.189, 3.190, and 3.191. This was also shown to occur with the related cyclic bis(alkyl)silylene, forming complexes 3.192, 3.193, and 3.194.¹⁴¹

	Scheme 41 Synthesis of nickel(0) and platinum(0) complexes bearing a cyclic bis(alkyl)silylene ligand, with additional alkene or carbonyl ligands.

	Scheme 42 Formation of a platinum(0) complexes bearing a cyclic (alkyl)(amido)silylene, and related reactivity towards coinage metal halides.

The unique phospha- and bora-ylidic heterocyclic silylenes 3.195 and 3.196 have demonstrated their ability to act as particularly strong donor ligands in complexes 3.197, 3.198, and 3.199 (for 3.195), and 3.200, and 3.201 (for 3.196) (Scheme 43).^142,143 Notably, the CO stretching frequencies in [Ni(CO)₃] complexes were compared to those of known phosphine, carbene, and silylene complexes, suggesting that both 3.195 and 3.196 exhibit stronger donor characteristics than the majority of these ligands. However, this observation has not been implemented in catalytic or further reactivity studies.

	Scheme 43 Transition metal complexation of unique heterocyclic silylenes featuring bora- and phospha-ylidic moieties.

Only one structurally characterised acyclic silylene group 4 complex is known, being a hafnocene complex (Scheme 44).⁶³ This was obtained by reacting dilithiosilane 3.202 with hafnocene dichloride, initially forming the unstable 16-electron complex 3.203. Under these conditions, C–H activation occurred, resulting in the formation of 3.204. However, the addition of PMe₃ led to the isolation of a stable 18-electron complex, 3.205, which exhibits significant Si–Hf double bond character.

	Scheme 44 Synthesis of a hafnium complex bearing an acyclic silylene ligand, via salt-metathesis.

Given the ability of group 6 metals to form alkylidyne complexes, they have been extensively investigated for the formation of their heavier analogues, metallo–silylidyne complexes, which will be discussed later in this review. Several synthetic routes to access these species involve acyclic–silylene complexes, which themselves are relatively rare. The first example of a base-free silylene complex of a group 6 metal was the tungsten species 3.206, featuring the Mes₂Si ligand (Scheme 45). This complex possesses a very short Si–W double bond (2.3850 Å), particularly when compared to the Si–W single bond in the same molecule (2.6456 Å).¹⁴⁷ The formation of this complex involved the photolysis of the methyl tungsten complex, [Cp*WMe(CO)₃], in the presence of the disilane HMe₂Si–SiMeMes₂, leading to initial dealkylation followed by a series of migratory reactions resulting in the formation of the silylene complex.

	Scheme 45 Synthesis of a tungsten complex bearing an acyclic silylene ligand, via σ-metathesis and group migration.

The (mesityl)(chloro)silylene complex 3.207 was obtained using a similar alkyl elimination strategy, involving the reaction of MesSiClH₂ with [Cp*Mo(dmpe)Bz] (Scheme 46).¹⁴⁸ Initially, the elimination of toluene takes place, followed by H-migration to Mo, resulting in the formation of the desired silylene complex. This species was utilised in accessing the first example of a metal complex with some degree of M–Si triple bond character (e.g. 3.208). However, due to the bridging nature of the hydride ligand, the triple bond is somewhat perturbed. The same research group employed the benzyl elimination approach to access a range of bis-alkyl (3.209, 3.210), bis-aryl (3.211), and (alky)(aryl)-silylene (3.212) complexes of molybdenum hydrides, as well as (aryl)(hydrido)silylene complexes 3.213, 3.214, 3.215, exhibiting varying degrees of Mo–H⋯Si interactions.¹⁴⁹ Similar chemistry was achieved using a C–H activated Cp* tungsten complex (Scheme 47). Addition of silane leads to protonation and regeneration of the Cp* ligand, resulting in the formation of base-free silylene complexes 3.216 and 3.217.¹⁵⁰

	Scheme 46 Synthesis of molybdenum complexes bearing acyclic silylene ligands, via σ-metathesis and hydride migration.

	Scheme 47 Synthesis of a tungsten complex bearing an acyclic silylene ligand, via σ-metathesis, chloride abstraction, and hydride migration.

An alternative route to sterically encumbered (alkyl)(hydrido)silylene tungsten complexes was subsequently reported. Here, the direct reaction of Cp-bound tungsten methyl complexes (i.e. [CpW(CO)₂Me L]; L = CO, MeCN) with monoalkylsilanes led to methyl elimination and H-migration to form 3.218 (Scheme 48).¹⁵¹ Addition of acetone to 3.218 triggers various reactive processes, including the insertion of the CO bond into the Si–H bond, forming 3.219, and CO bond metathesis, forming 3.220. The reversible addition of one C–H bond of acetone is also observed, forming 3.221. Furthermore, it was later discovered that the addition of ^iPrNHC (^iPrNHC = [{(Me)CN(ⁱPr)}₂C:]) to 3.218 results in hydride abstraction from this complex, forming anionic 3.222 with a contracted Si–W bond length, indicative of the “formal” silylene character (i.e. without bridging hydride interaction).^152,153

	Scheme 48 Synthesis of a tungsten complex bearing a hydride-bridged acyclic hydrido-silylene ligand, and its reactivity.

En route to a stable molybdenum silylidyne complex, the NHC-stabilised (aryl)(chloro)silylene 3.223 was reacted with the anionic molybdenum salt Li[CpMo(CO)₃], resulting in metathesis and the formation of 3.224; the most likely resonance form features a positively charged carbene (i.e. imidzolium) ligand, and an anionic molybdenum center. As such. 3.224 represents a formal silylene complex (Scheme 49).¹⁵⁴ Subsequent NHC abstraction led to the first example of a molybdenum silylidyne complex, 3.225. Addition of nucleophiles to this species generally resulted in addition at Si, yielding a range of novel molybdenum silylene complexes, such as 3.226, 3.227, and 3.228.¹⁵⁵ This indicates the electrophilic nature of the silicon center in triply-bonded M–Si species. The [2+2] cycloaddition of carbodimides and ketones to metal silylidyne complexes also leads to metal silylene complexes, forming 4-membered metallacycles with metal–silicon double bonds. Complex 3.229 exhibits such reactivity, giving rise to 3.230, 3.231, and 3.232, all of which contain a silylene fragment (Scheme 50).^156,157

	Scheme 49 Synthesis of molybdenum complex bearing acyclic silylene ligands, via salt metatheses.

	Scheme 50 Synthesis of a tungsten complexes bearing acyclic silylene ligands, via cycloaddition of a tungsten-silylidyne complex.

Two acyclic silylene complexes of group 7 metals have been identified. They are obtained through the addition of diethyl or diphenyl silane to the alkene-coordinated Mn^I hydride complex [(dmpe)₂MnH(C₂H₄)], resulting in the formation of 3.233 and 3.234, accompanied by the elimination of ethane (Scheme 51).¹⁵⁸ While the diethyl derivative 2.233 exclusively exists as the trans-isomer without Mn–H⋯Si interactions, the cis-isomer is observed in the case of the diphenyl derivative 3.234. Both complexes react with H₂ to yield hydride-bridged silyl complexes 3.235 and 3.236.

	Scheme 51 Synthesis of manganese complexes bearing an acyclic silylene ligands, via dihydrogen elimination.

The first base-free silylene complexes were reported in 1990 in the cationic ruthenium species 3.237 and 3.238, synthesised through triflate abstraction from their silyl complexes.¹⁵⁹ Four years later, silylene complexes without π-donating substituents at Si^II were reported (3.239 and 3.240, Scheme 52).¹⁶⁰ Since then, several group 8 silylene complexes have been documented.

	Scheme 52 Synthesis of a rutheneium complexes bearing acyclic silylene ligands.

Iron complexes featuring an acyclic silylene ligand are rather rare. The first, 3.241, was synthesised using a similar approach as tungsten complex 3.206. Irradiation of a mixture of [Cp*FeMe(CO)₂] and disilane HMe₂Si–SiMeMes₂ resulted in the initial loss of methane, followed by Me₃Si migration.¹⁶¹ Upon addition of tert-butyl isocyanide, evidence of both 1,2- and 1,3-group migrations in 3.241 was observed, leading to the formation of silyl complex 3.242 (Scheme 53). These findings provide insights into the fluxional binding behavior in such complexes. More recently, a handful of additional Fe complexes were reported, accessed from ‘masked’ (phosphinimide)(silyl)silylenes (Scheme 54). These precursors, akin to earlier reported 3-membered silacycles,^162–165 can eliminate their cyclohexane unit in a retro-[2+1] reaction, releasing cyclohexene. The resulting silylenes are trapped by additional substrates, e.g. [Fe(CO)₄].¹⁶⁶ In this manner, complexes 3.243, 3.244, and 3.245 were isolated.

	Scheme 53 Synthesis of an iron complex bearing an acyclic silylene ligand, via σ-metathesis and group migration.

	Scheme 54 Synthesis of an iron complex bearing an acyclic silylene ligand, via cyclohexene elimination.

Osmium complexes with related silylene ligands were accessed through initial dealkylation reactions of Os^II alkyls. Specifically, the formation of bis(alkyl)silylene complexes 3.246 and 3.247 involved this initial dealkylation reaction, followed by the conversion of (chloro)silyl complex [Cp*Os(PMe₃)₂(SiClR₂)] to the corresponding triflate, and subsequent triflate abstraction (Scheme 55).¹⁶⁷ The direct dealkylation of Os^II benzyl species [Cp*Os(PⁱPr₃)(Bz)] with aryl silanes, followed by spontaneous hydride migration, led to the formation of (hydrido)(aryl)silylene complexes 3.248 and 3.249.¹⁶⁸

	Scheme 55 Synthesis of osmium complexes bearing acyclic silylene ligands, via salt elimination, and σ-metathesis and hydride migration.

Further examples of ruthenium–silylene complexes have been synthesised through alkyl abstraction and hydride/group migration methods.^169–171 Specifically, examples of (alkyl)(hydrido)- (3.250) and (aryl)(hydrido)-silylene (3.251, 3.252, 3.253, and 3.254) complexes having been accessed utilising a [Cp*Ru] moiety, as well as the unique scorpionate complex 3.255,¹⁷² where silane addition leads to H-transfer to one activated Ph group of the ligand (Scheme 56). Additionally, the (chloro)(silyl)silylene complex 3.256 was prepared by initially combining the cationic NHC-stabilised silylene [(IMe)₂(^tBu₃Si)Si]Cl with arene-coordinated [{RuCl₂(p-cym)}₂] (p-cym = 1-Me-4-ⁱPrC₆H₄), followed by reduction with KC₈ (Scheme 56).¹⁷³

	Scheme 56 Synthesis of numerous ruthenium complexes bearing acyclic silylene ligands.

Further investigations into the chemistry of several ruthenium–silylene complexes also reveal intriguing points. The (hydrido)silylene complex 3.250 demonstrated reactivity with nitriles, whereby insertion into the Ru–Si double bond and subsequent bond migration processes result in the formation of silyl–isocyanide complexes 3.257 and 3.258 (Scheme 57).¹⁶⁹ In a more practical context, the closely related hydrido complex 3.254 exhibited effective catalytic activity in the hydrosilylation of alkenes, for which the postulated mechanism involves alkene insertion into the Si–H bond of the silylene ligand (Scheme 58).¹⁷¹ This previously unobserved mechanism highlights the potential significance of silylene–metal complexes as intermediates in industrially applicable hydrosilylation catalyses.

	Scheme 57 Reactivity of an acyclic–silylene ruthenium complex torwards organonitriles.

	Scheme 58 Insertion chemistry of an acyclic–(hydrido)silylene ruthium complex towards alkenes, and subsequent metathesis chemsitry with a silane.

Among the group 9 elements, iridium dominates the field of base-free acyclic silylene chemistry. These complexes, besides a single scorpionate-silylene cobalt example with chelating phosphine arms (Scheme 59), represent the only stable and structurally characterised examples. The formation of the aforementioned cobalt complex involved the combination of phosphine-functionalised bis(aryl)silane 3.259 with [Co₂(CO)₈], leading to the loss of H₂ and CO. Subsequent hydride abstraction from the resulting silyl complex yielded the cationic cobalt silylene complex 3.260.¹⁷⁴ Notably, the Si^II center in this species exhibited Lewis acidity, enabling the binding of nucleophiles and facilitating synergistic bond activation across the SiCo bond. Substrate coordination (EtOH, H₂O) at silicon resulted in the formation of cationic silyl complexes 3.261 and 3.262.

	Scheme 59 Synthesis and addition chemistry of a cobalt complex bearing an acyclic silylene ligand.

The initial example of an acyclic–silylene iridium species involved the deallylation of a tris(phosphino)phenyl borate-stabilised Ir^III allyl complex with bis(organyl)silanes, accompanied by Si–H bond migration, leading to alkyl- (3.263 and 3.264) and aryl-silylene (3.265 and 3.266) complexes (Scheme 60).^175,176 It was also demonstrated that employing TripSiH₃ as the silane (Trip = 2,4,6-ⁱPr₃C₆H₂) allowed for the synthesis of the (aryl)(hydrido)silylene derivative 3.267, which undergoes alkene insertion at the Si–H bond, akin to the earlier described catalytically active ruthenium complex 3.254, in forming (alkyl)(aryl)silylene complex 3.268. Subsequently, a range of iridium silylene complexes, 3.269, 3.270, and 3.271, were accessible via hydride abstraction from iridium silyl complexes (Scheme 61). The hydride-substituted silylene complexes 3.269 and 3.271 exhibited catalytic activity in alkene hydrosilylation, suggesting a second sphere mechanism where alkene insertion into the Si–H bond plays a pivotal role. One such postulated mechanism is outlined in Scheme 61.¹⁷⁷

	Scheme 60 Synthesis of iridium complexes bearing acyclic bis(organyl)- and hydrido-silylene ligands, and the insertion chemistry of the latter towards alkenes.

	Scheme 61 Synthesis of osmium complexes bearing acyclic silylene ligands, via salt elimination, and σ-metathesis and hydride migration (above), the one possible alkene hydrosilylation mechanism which involves the Ir and Si centres in 3.269 (below).

One of the earliest reported instances of a TM–silylene complex was the platinum species 3.272, featuring the bis(thionyl)silylene ligand, and was accessed through triflate abstraction from its silyl complex (Scheme 62).⁵² Despite this initial discovery, only a limited number of well-defined examples of acyclic–silylene complexes involving group 10 metals have been reported. Subsequent to the 1993 publication, the neutral platinum silylene complex 3.273 was reported, generated through the in situ formation of bis(mesityl)silylene in the presence of [(R₃P)₂Pt] (R = ⁱPr, Cy).¹⁷⁸ The resulting PCy₃-stabilised complex exhibited reactivity at the PtSi bond, leading to H₂ cleavage and the formation of silyl complex 3.274, as well as alcohol cleavage to yield Pt⁰ species and alkoxy silanes. A closely related (aryl)(bromo)silylene complex of the [(Cy₃P)₂Pt] fragment (3.275) was also obtained via the addition of a stable 1,2-diaryl-1,2-dibromo disilene to [(Cy₃P)₂Pt].¹⁷⁹ However, further investigations into the reactivity of this species were not pursued.

	Scheme 62 Synthesis of platinum complexes bearing acylic silylene ligands.

The acyclic silylene Ni⁰ complex 3.276 was subsequently reported (Scheme 63), synthesised through the combination of NHC-stabilised (amido)(chloro)silylene [{(Dip)(Me₃Si)N}(Cl)Si·^iPrNHC] with [Ni(cod)₂] in the presence of an additional equivalent of ^iPrNHC.¹⁸⁰ The resulting 16-electron Ni⁰ complex displayed high reactivity in synergistic bond activation processes. Notably, it demonstrated the capacity to cleave H₂, similar to platinum complex 3.273, resulting in the formation of silyl complex 3.277. More remarkably, 3.276 cleaves the catechol ligand upon reaction with catechol borane, leading to the formation of the unique Ni^II hydroborylene complex 3.278 (Scheme 73). Further studies demonstrated that the chloride ligand in 3.276 could be exchanged through salt-metathesis, yielding a “half parent” bis(amido) silylene complex and various sila-phosphene, -arsene, -phosphinidene, and -arsinidene species (vide infra).¹⁸¹ Additionally, it was shown that 3.276 could activate ammonia across the SiNi bond, resulting in the formation of 3.279, a rare reaction for TM species. Furthermore, this complex also demonstrated the cycloaddition of alkenes (3.280), alkynes (3.281 and 3.282), aldehydes (3.283), and imines (3.284), in all cases forming novel metallacyclic complexes (Scheme 64).⁴⁹ Notably, the reaction with ethylene was found to be reversible. Related chemistry has recently been reported for (amido)(bromo)silylene-Pt⁰ complex 3.285, synthesised via the addition of the free (amido)(bromo)silylene [(Cy₃P)₂Pt]. This complex reacts with an excess of ethylene in forming the [2+2+2] product 3.286. Alternatively, the free silylene also reacts with [(Ph₃P)₂Pt(C₂H₄)₂] in generating the formal [2+2] reaction product, 3.287, which reacts with one addition equivalent of ethylene to form 3.288. Here, ethylene insertion chemistry is not reversible, but, overall, both mono- and bis-ethylene [2+2] cycloadditions are observed at the Si–Pt interface (Scheme 65).¹⁸²

	Scheme 63 Synthesis and small-molecule activation chemistry of an acyclic–silylene nickel(0) complex.

	Scheme 64 Cycloaddition chemistry of an acyclic–silylene nickel(0) complex.

	Scheme 65 Cycloaddition chemistry of an acyclic–silylene platinum(0) complex.

In all silylene systems discussed to this point, the TM centre has borne a ‘classical’ geometry. Recently, Kato and co-workers reported that (amido)(chloro)silylene ligand 3.289 reacts with [Ni(cod)₂], in the presence of an additional NHC, to form a chelating silylene-ligated Ni⁰ complex 3.290. Though structurally very similar to the ‘non-chelating’ (amido)(chloro)silylene complex 3.276 described above, by being built into a chelating ligand scaffold the Lewis acidic character of the Si^II centre becomes more apparent, leading to Z-type ligand behaviour (Scheme 66).¹⁸³ The Ni⁰ centre thus behaves as an L-type ligand towards silicon, this electronic situation borne out by the T-shaped geometry at nickel. Prior to that report, only one T-shaped Ni⁰ complex had been described, featuring the [GeCl₂] ligand (vide infra). Following the initial publication of neutral 3.290, the cationic derivative 3.291 was also reported, accessed through chloride abstraction from 3.290 with Na[BAr₄^F].¹⁸⁴ Cationic 3.291 also holds a Z-type silylene centre, and thus a T-shaped geometry at nickel. Both described complexes are highly reactive, but show rather different characteristics. Neutral complex 3.290 readily activates H₂, leading to the 1,2-dihydride complex 3.292 through ‘dual-centred’ activation. In contrast, cationic 3.291 behaves akin to a metallosilylene, with H₂ oxidative addition only at the Si-centre, in forming 3.293. A similar case is observed for MeOTf, whilst the unsaturated species diphenylacetylene and 2,3-dimethyl-1,3-butadiene undergo [2+1] and [4+1] cycloadditions at the silicon centre (e.g. 3.294 and 3.295, Scheme 66), respectively, in both neutral and cationic systems. This opens the question as to whether cationic 3.291 is better described as a cationic Ni^II metallosilylene, which may become more clear through future studies.

	Scheme 66 Synthesis and reactivity of neutral and cationic acyclic–silylene nickel(0) featuring a T-shaped nickel centre.

To date, no base-free acyclic silylene complexes of group 11 or 12 metals have been reported.

	Fig. 7 A variety of the known chelating bis(silylene) ligands which are known, utilising the Roesky silylene.

The first instance of a chelating bis(silylene) ligand derived from silylene 3.1 did not involve the chloro–silylene itself. Instead, it was synthesised by dehydrochlorination of its disiloxane derivative with Li[N(SiMe₃)₂], resulting in the formation of “disilylenoxane” 3.303. This compound was found to readily act as a chelating ligand toward Ni⁰ when reacted with [Ni(cod)₂], leading to the formation of complex 3.304 (Scheme 67).¹⁸⁶ Shortly after this initial discovery, it was demonstrated that chloro–silylene 3.1 could be used to easily access chelating bis(silylene) compounds. When reacted with dilithio-resorcinolate, compound 3.296 is efficiently formed. Furthermore, the reaction of this chelating ligand with [Pd(PPh₃)₄] resulted in the insertion of the metal into the central aryl C–H bond, followed by hydrogen migration to one of the silylene centers, ultimately generating Pd^II complex 3.305. This complex was further stabilised by one silylene center from an additional equivalent of 3.296 (Scheme 68).¹⁸⁷

	Scheme 67 Synthesis of an oxo-bridged bis(silylene) ligand, and its coorindation to nickel(0).

	Scheme 68 Coordination of a resorcinol-based bis(silylene) ligand to palladium(II).

The methodology described above was also expanded to include complexes of Ir and Rh (Scheme 69), as well as complexes incorporating closely related chelating Ge^II (vide infra) and phosphine ligands featuring the resourcinol backbone.¹⁸⁸ In all cases, no hydrogen migration was observed, thereby preserving the E^II character (E = Si, Ge) of the donor centers in the ligands (for Si: 3.306, 3.307, and 3.308). These complexes were found to be highly efficient catalysts for the dehydrogenative borylation of arenes using pinacol borane. This provided the initial evidence that chelating silylene (and germylene) ligands exhibit significantly stronger σ-donating capabilities compared to their phosphine counterparts. The same ligands derived from resorcinol were employed in Ni-catalysed Sonogashira cross-coupling reactions, utilising Ni^II bis(silylene) complex 3.309 (Scheme 70).¹⁸⁹ It is worthy of note that the heterobimetallic complex 3.310 was isolated through introduction of a copper acetylide to 3.309, suggesting its crucial role as an intermediate in the catalytic reaction mechanism. More recently, the m-phenylenediamine derivative, 3.311, was reported, and Co^III monohydride complexes accessed through oxidative addition of Co^I into the C–H bond of the central arene (3.312, Scheme 71).¹⁹⁰ Though the bis(silylene) ligated complex was an active catalyst for alkene hydrosilylation, it was in fact less active in this chemistry than the related bis(phosphine) ligated complexes 3.313 and 3.314.

	Scheme 69 Synthesis of iridium and rhodium complexes bearing a resorcinol-derived bis(silylene) ligand.

	Scheme 70 Reactivity of a resorcinol-derived bis(silylene) ligand towards nickel(II) bromide (inset: Identified bimetallic intermediate in the catalytic Sonogashira coupling).

	Scheme 71 Synthesis of a cobalt complex utilising a m-diaminophenyl-derived bis(silylene) ligand, and corresponding phosphine complexes.

The ferrocene-bridged bis-silylene 3.297 was successfully synthesised by reacting di(lithio)ferrocene with chloro–silylene 3.1. In the initial publication on this ligand, it was demonstrated that the reaction with in situ generated [CpCo] resulted in the formation of the Co^I complex 3.315 (Scheme 72).¹⁹¹ This complex was utilised as a catalyst for the cyclo-trimerisation of phenylacetylene and acetonitrile, expanding the scope of transformations achievable with this ligand class. Subsequently, the same bis-silylene ligand was employed to stabilise novel Fe⁰ arene complexes 3.316 and 3.317, as well as Fe^II dihalide complexes 3.318 and 3.319. The Fe^II complex 3.316 proved to be an effective catalyst for the hydrogenation of ketones.¹⁹²

	Scheme 72 Formation of iron(II), iron(0), and cobalt(I) complexes bearing a ferrocene-derived bis(silylene) ligand.

The pyridine-derived pincer ligand 3.298, which is closely related to the resorcinol derivative 2.396, can be readily synthesised via a salt-metathesis reaction involving chloro–silylene 3.1. In the initial publication regarding Fe^II and Fe⁰ complexes, it was observed that the coordination of the pyridine moiety to iron depended on the oxidation state of the metal center. The ligand exhibited tridentate coordination in Fe⁰ complexes 3.320 and 3.321, while no N → Fe donation was observed in Fe^II complex 3.322 (Scheme 73).¹⁹³ Bis(trimethylphosphine) complex 3.320 served as a catalyst for the hydrosilylation of ketones, with lower catalyst loadings required compared to ferrocene-derived bis-silylene complex 3.316 for comparable transformations (vide infra). Pyridine-supported bis(silylene) 2.298 was also later shown to stabilised Mn⁰ in complex 3.323, as well as Mn^I species 3.324, carbonyl-free examples of which are rare. The former open-shell species was shown to be an active catalyst for the selective 1,2-hydroboration of pyridines (Scheme 74).¹⁹⁴

	Scheme 73 Formaition of iron(II) and iron(0) complexes bearing a bis(amino)pyridine-derived bis(silylene) ligand.

	Scheme 74 Formaition of managnese(II), manganese(I), and manganese(0) complexes bearing a bis(amino)pyridine-derived bis(silylene) ligand.

A subsequent publication revealed that the Fe⁰ complex 3.320 undergoes oxidative addition of silanes to yield octahedral complexes 3.325, 3.326, and 3.327, which was proposed as the initial step in hydrosilylation catalysis (Scheme 75).¹⁹⁵ Further investigations using density functional theory (DFT) calculations indicated that a “peripheral” mechanism, where the incoming ketone attacks the silyl-Si center followed by the formation of a 4-membered transition complex at the activated ketone with a second equivalent of silane, was energetically favorable for hydrosilylation.

	Scheme 75 The reactivity of an iron(0) bis(silylene) complex towards silanes, and proposed intermediates in the catalytic hydrosilylation of ketones.

More recently, further examples of chelating bis(silylene)-complexed iron species have been reported. The first involved the development of a novel pyrollyl-derived bis(silylene) ligand, 3.328, which reacts with [Fe(PMe₃)₄] to yield hydrido-Fe^II complex 3.329. This species was shown to undergo PMe₃-N₂ exchange (viz. 3.330), and as such was also investigated for the catalytic silylation of N₂ (Scheme 76).¹⁹⁶ Here, up to 74 turnovers could be observed, being comparable to the relatively small number of molecular catalysts which can achieve this reaction.⁹⁵ An additional example of an Fe⁰ complex has also been reported, accessed rather uniquely via small-molecule activation by a ‘disilyne’ complex of iron (Scheme 77). Addition of Me₃SiN₃ leads to scission of the Si–Si bond, and formation of azide-bridged 3.331. This complex eliminates N₂, finally forming [(Me₃Si)N]-bridged 3.332,¹⁹⁷ the bis(silylene) ligand backbone of which is not dissimilar to that recently reported by Driess and co-workers.¹⁹⁸ Indeed, it was recently reported by Roesky and co-workers that this bis(silyenyl)aniline (viz. 3.333) reacts with [Fe(CO)₅].^198,199 Given the tight bite angle in this bis(silylene) ligand, the resulting complex (viz. 3.334) contains the [Fe₂(CO)₇] unit, featuring a formal Fe–Fe interaction.

	Scheme 76 Synthesis and iron complexation of a bis(pyrrole)-derived bis(silylene).

	Scheme 77 Iron complexes of RN-appended bis(silylene) ligands.

Cobalt complexes of pyridine-derived bis-silylene ligand 3.298 have been developed and successfully applied in the C–H borylation of arenes and heterocycles. In the initial report, the cobalt dibromide complex 3.335 was utilised as the catalyst, and an additional hydride source, Na[BEt₃H], was employed to generate the active cobalt hydride catalyst in situ (Scheme 78).²⁰⁰ A subsequent publication demonstrated that the dichloro-complex 3.336 could be utilised for the generation of stable hydride complex 3.337, while the (hydrido)(boryl)cobalt complex 3.338 could be subsequently generated through reaction with HBpin.²⁰¹ A detailed study was conducted to explore the potential reaction mechanism, which goes beyond the scope of this review. However, it was proposed that a Co^I–Co^III redox cycle may be involved, especially as no evidence of non-innocent ligand behavior was observed.

	Scheme 78 Formaition of cobalt(II) complexes bearing bis(amino)pyridine-derived bis(silylene) ligands.

The carborane-derived bis-silylene ligand 3.299, similar to examples described earlier, was synthesised through a straightforward salt-metathesis reaction and was utilised in the nickel-catalysed Buchwald-Hartwig amination of halo-arenes. The high donor strength of the ligand, imparted by the electron-rich carborane moiety, contributed to impressive reaction rates. Manganese dichloride complexes of this bis-silylene ligand, as well as the ligands discussed above (viz. 3.339, 3.340, 3.341, and 3.342, Scheme 79), were also reported and used in the transfer semi-hydrogenation of alkynes to E-alkenes.²⁰²

	Scheme 79 Synthesis of known manganese(II) dichloride complexes of bis(silylene) ligands.

Utilising novel xanthene-derived ligand 3.300, Ni⁰ complexes could be readily accessed which were shown to be catalytically active in alkene hydrogenation.²⁰³ The wide bite-angle of this ligand resulted in remarkable catalytic activity, similar to the well-known XantPhos ligand.^204,205 In this case, the Ni⁰ complex 3.343 was synthesised and shown to catalyse the hydrogenation of unactivated alkenes, including the challenging tetramethylethylene, at ambient temperature and 1 bar H₂. It was suggested that the ligand exhibits non-innocent behavior, with the 3.343 forming the silyl complex 3.344 upon reaction with H₂. Notably, the related bis(trimethylphosphine) complex 3.345 undergoes reversible reaction with H₂, leading to the mixed-valence complex 3.346, incorporating one silyl and one silylene center (Scheme 80).

	Scheme 80 Synthesis of nickel(0) complexes supported by a xanthene-derived bis(silylene) ligand, and their reactivity towards H2.

As an adjunct to the established xanthene derived bis(silylene), Mo and co-workers have developed the 9,10-dihydroacridine derived bis(silylene) 3.347, which was utilised in the formation of Fe^II complex 3.348 (Scheme 81).²⁰⁶ This complex was shown to undergo oxidation with N₂O, yielding 3.349, referred to as an iron-silanone complex, with one oxygen atom bridging the Fe and a single Si centre. Further, 3.348 was found to be an active catalyst for the reduction of N₂O with HBpin, in the formation of N₂, H₂, and (pinB)₂O.

	Scheme 81 Synthesis of iron(II) complexes supported by and anionic 9,9-dimethylacridine-derived bis(silylene) ligand, and subsequent oxidation of one silylene centre.

Three additional examples of Ni⁰ complexes incorporating chelating bis-silylene ligands have been reported. The first, namely acenaphthene-derivived 3.350, is obtained through the addition of disilene 3.301 to [Ni(cod)₂], resulting in insertion into the Si–Si bond (Scheme 82).²⁰⁷ The second example was obtained through reaction of the bis(silylene)-stabilised germylone 3.351 with [Ni(cod)₂], leading to a unique complex in which two bis-silylene ligands stabilise a [Ge₂Ni] three-membered ring (viz. 3.352, Scheme 83).²⁰⁸ Most recently, the terphenyl bis(silylene) ligand 3.353 was developed, and its Ni⁰ complex accessed.²⁰⁹ Here, the central aryl ring of the ligand backbone acts as an additional ligand towards nickel (Scheme 84). Complex 3.354 was employed in the catalytic hydrogenation of alkenes, with reversible H₂ activation shown to occur at one Ni–Si interface in the formation of 3.354·H₂.

	Scheme 82 Formation of a bis(silylene) nickel(0) complex through the addition of a acenapthene-derived disilene to [Ni(cod)2].

	Scheme 83 Formation of a unique [NiGe2] moiety through the addition of a bis(silylene)-supported germylone to [Ni(cod)2].

	Scheme 84 Formation of a nickel(0) complex of a terphenyl-derived bis(silylene) ligand.

Finally, the only reported group 6 complex bearing a bis(silylene) ligand was recently reported by Jones and co-workers, through the reaction of bis(silylene) 3.25 with [Mo(CO)₆], under blue light irradiation, forming 3.26.⁸² The ‘free’ bis(silylene) could also be accessed by activation of gaseous CO, which subsequently reacts with [Mo(CO)₆] (Scheme 85).

	Scheme 85 The addition of an Si–Si bonded bis(silylene) to [Mo(CO)6], leading to CO-activation and Mo-chelation.

The diverse range of complexes obtained using chelating bis-silylene ligands, which includes both chemical curiosities and active catalysts, suggests that this area of research will continue to flourish in the coming years. This should further enhance our understanding of bonding in these complexes and enable the development of tailored catalysts for key catalytic processes. Particularly noteworthy are complexes where the low-valent silicon centre exhibits non-innocent behaviour, a concept that is just beginning to be explored in the broader context of tetrylene ligation.

	Scheme 142 Synthesis of group 4 complexes stabilised by cyclic 4- and 5-membered bis(silyl)germylenes.

Later, it was shown by Lee, Sekiguchi and co-workers that the bicyclic dianion 3.579 reacts with [Cp₂TiCl₂] in forming the bicyclic-germylene adduct of [Cp₂Ti], 3.580, as its THF adduct. This species could be crystallised as their PMe₃ (3.581) and XylNC (3.582) adducts, revealing a short Ge–Ti bond marked as a formal double bond (Scheme 143).²⁷⁷ Moreover, this linkage is negatively polarised towards Ge, in contrast to the vast majority of tetrylene complexes discussed in this review, and as such is best described as a Schrock germylene complex. The same publication reported the W and Mo derivatives, though no structural data was provided.

	Scheme 143 Synthesis of bicyclic bis(silyl)germylene complexes of titanium, which are described as Schrock germylene complexes.

Later, the hafnium-based bicyclo[2.1.1.]hexene germylene (BCHGe) systems 3.583 and 3.584 were reported, accessed via the salt-metathesis of the corresponding dipotassium germole ligands with [Cp₂HfCl₂], which proceeds through the anionic germylene complex 3.585 (Scheme 144).^278,279 These BCHGe ligands are stabilised through a significant interaction with the unsaturated C–C backbone.²⁷⁹ Nevertheless, they were shown to react with TM fragments as typical Lewis basic ligands, as shown through the addition of 3.583 to [Fe₂(CO)₉] and [W(CO)₅(THF)]. In these species (viz. 3.586 and 3.587) the interaction of the Ge centre with the CC π-system is maintained, indicating negligible M → Ge (M = Fe, W) back-bonding. In contrast, reaction with [Ni(cod)₂] led to an equilibrium mixture of 3.583 and bis(germylene) complex 3.588, in which the Ge⋯CC interaction is cleaved. This is indicative of a greater Ni → Ge back-donation when compared to the above complexes, which competes with the described π-interaction. As such, isolated complex 3.586 is not stable in solution, rapidly decomposing upon dissolution.

	Scheme 144 Transient formation of an unsaturated 5-membered germylene-haffnium complex, rearrangement to a bicyclo[2.1.1.]hexene germylene complex, and its utility in the formation of heterometallic species.

Finally, Kira and co-workers reported the synthesis of a cyclicalkylgermylene-stabilised di-goldgermane complexes, generated through the initial insertion of bis(alkyl)germylene 3.589 into the Au–Cl bond of [AuCl(PR₃)] (R = Me (3.590), R = Et (3.591)), followed by KC₈ reduction forming 3.592 and 3.593 (Scheme 145).²⁸⁰ The core of these species can be described as a dimetallated germane, with halogen-like Au centres which bear a degree of negative charge localisation, each supported by a dative ligand (i.e. germylene or phosphine).

	Scheme 145 Formation of mono- and bis-gold complexes utilising a cylic bis(alkyl)germylene ligand.

No such complexes are known for group 3. Regarding group 4, addition of the dilithiogermanide 3.594 to a hafnocene dichloride derivative led to Hf complex 3.595 (Scheme 146).²⁸³ Given the instability of this initially formed product above −50 °C, the germylene complex is additionally stabilised by PMe₃, which binds at Hf, forming 3.596. Complex 3.595 bears a formal Hf–Ge double bond, reflected clearly by the calculated frontier orbitals for this species which represent π- and π*-bonds for the HOMO and LUMO, respectively. 3.595 was assigned as a Schrock-type germylene complex, on the basis of the highly polarised Ge–Hf bond, with partial charges on Ge and Hf of −0.32 and +0.74, respectively.

	Scheme 146 Synthesis of an acyclic bis(silyl)germylene haffnium complex.

Numerous examples of acyclic–germylene complexes of the group 6 metals are known, the first reported by Jutzi and co-workers in 1976, with the first structurally characterised example appearing the following year, reported by Lappert and co-workers.^51,282 We note that the X-ray crystal structure of the Jutzi example was reported two years later.²⁸⁴ The former report first involved the synthesis of the base-stabilised [GeCl₂] adduct of [Cr(CO)₅] (viz. 3.597, Scheme 147),²⁸⁵ which reacts with thiosilanes RSSiMe₃ (R = Me, Mes) to yield di(thio)germylene complexes 3.598 and 3.599.²⁸² These systems also further react with BCl₃ in thiol-chloride exchange, forming base-free dichlorogermylene complex 3.600. Later, alongside the crystallographically determined structure of 3.599, the same group reported that addition of bases to the germylene complexes led to binding at Ge in forming complexes 3.601, 3.602, and 3.603.²⁸⁴ This has implications in cooperative bond activation via ligand centred nucleophile binding. Soon after, Lappert and co-workers demonstrated the coordination ability of their earlier reported acylic bis(alkyl)- and bis(amido)germylenes (3.604 and 3.605, respectively; Scheme 148).⁵¹ Bis(amido) system 3.605 was shown to undergo mono-carbonyl substitution with [W(CO)₆] under irradiation, yielding 3.606, whilst reaction with [W(CO)₄(norbornadiene)] formed bis-germylene complex 3.607. For bis(alkyl) germylene 3.604, the corresponding bis-germylene was also reported (viz. 3.608), in addition to the [Co(CO)₅] complex 3.609, generated by addition of 3.604 to [Co(CO)₆] under irradiation.

	Scheme 147 Early examples of chromium complexes bearing acyclic germylene ligands.

	Scheme 148 Formation of chromium(0) and tungsten(0) complexes bearing acyclic germylene ligands.

Much later, further examples of group 6 complexes were reported. In 1993, it was demonstrated that terminal digermyl-tungsten complex 3.610 undergoes germyl migration upon irradiation, forming (germyl)(germylene)tungsten complex 3.611, with a terminal dimethylgermylene ligand (Scheme 149).²⁸⁶ This approach to such a species is of significant importance, given that [Me₂Ge:], the smallest alkyl germylene, is not stable in its own right. Notably, Ogino and co-workers reported spectroscopic evidence for such a process some years earlier, in the irradiation of [CpFe(CO)₂(Ge₂Me₃)] and derivatives.²⁸⁷ The electrophilicity of the germanium centre in 3.611, as per examples from Jutzi, was also demonstrated through the addition of pyridine, which selectively coordinates at Ge (viz. 3.612). Soon after, the first examples of base-free bis(aryl)germylene TM complexes were reported (viz. 3.613 and 3.614), through addition of the isolable bis(aryl)germylene 3.615 to [M(CO)₅·THF] (M = W, Cr; Scheme 150).²⁸⁸ En route to a germylidyne complex (vide infra), Filippou and co-workers reported the synthesis of phosphine-appended complex 3.616, through addition of (aryl)(chloro)germylene 3.617 to hexakis-trimethylphosphino tungsten (Scheme 151). Here, the Ge–Cl bond is activated, in addition to the Ge centre inserting into the W–C bond of an activated PMe₃ ligand.²⁸⁹ Acyclic bis(alkyl)germylene and bis(aryl)germylene complexes of molybdenum were reported by Tilley, accessed through the activation of germanes by their previously reported benzyl molybdenum(II) complex [Cp*Mo(pmpe)(CH₂Ph)] (dmpe = 1,2-(PMe₂)C₂H₄; Scheme 152).²⁹⁰ An initial Ge–H bond activation leads to the elimination of toluene, with subsequent hydride migration from Ge to Mo generating target bis(alkyl)- and bis(aryl)-germylene complexes 3.618 and 3.619, respectively. Although one might assume that bis(aryl) system 3.619 bears a more electrophilic Ge centre, only the bis(alkyl) system features a bridging Mo–H⋯Ge interaction. This chemistry is particularly reminiscent of silane/silylene chemistry from the same group, which has implications in hydrosilylation catalysis which may proceed via TM silylene complexes.

	Scheme 149 Formation and coordinative behaviour of dimethylgermylene in the coordination sphere of tungsten.

	Scheme 150 Formation of bis(aryl)germylene complexes of tungsten and chromium. Tip = 2,4,6-iPr3-C6H2.

	Scheme 151 Formation of a phosphine-appended germylene complex of tungsten.

	Scheme 152 Formation of bis(organyl)germylene complexes of molybdenum through dehydrogenation of a germanes.

Three further examples of acyclic–germylene complexes of group 6 metals are known, accessed via metal germylidyne complexes which are discussed subsequently (Scheme 153). First, Filippou and co-workers reported that Mo and W germylidyne complexes 3.620 and 3.621 react with the small NHC [(MeCNMe)₂C:], with binding exclusively at Ge.²⁹¹ This leads to a formal imidazolium salt, with a negative charge residing at Mo/W, and a formal Ge–C bond. As such, products 3.622 (Mo) and 3.623 (W) are best described as germylene complexes, borne out by the considerable lengthening of their Ge–M bonds (i.e. Δd_M–Ge for 3.622:0.1199 Å; for 3.623:0.1173 Å). Finally, the closely related Cp*-tungsten germylidyne complex 3.624, reported by Hashimoto and Tobita and co-workers, reacts with both alcohols and aldehydes in the formation of germylene complexes.²⁹² In the former reactions, 1,2-addition across the W–Ge triple bond is observed, with the alkoxide moiety bound at Ge indicative of the electrophilicity of this centre, forming novel germylene species 3.625, 3.626, and 3.627. The situation is somewhat more complex for aldehydes; two equivalents react, the first in C–H activation across the W–Ge bond in forming an (alkyl)(hydrido)germylene complex, and the second through insertion into the newly formed Ge–H moiety. This was observed for a variety of aryl aldehydes, yielding compounds 3.628, 3.629, and 3.630, in all cases forming chiral Ge centres, though none demonstrating notable enantioselectivity.

	Scheme 153 Formation of acyclic germylene complexes of molybdenum and tungsten through addition reactions to germylidyne complexes.

Three true examples of group 7 TM complexes bearing acyclic germylenes are known, namely Re complexes 3.631, 3.632, and 3.633, the initial complex accessed through the direct addition of (aryl)(chloro)germylene 3.617 to [ReCl(PMe₃)₅] (Scheme 154).²⁹³ Subsequently this species loses an additional PMe₃ ligand in forming a germylidyne complex (vide infra), which reacts with CO or MeNC in the regeneration of germylene complexes, 3.632 and 3.633.

	Scheme 154 Formation of rhenium complexes bearing acyclic germylene ligands.

A number of Fe and Ru complexes with acyclic germylene ligands have been reported (Scheme 155), accessed either via direct germylene addition to Fe species, or through Ge–H activation of germanes. For the former, the [Fe(CO)₄] complexes 3.634 and 3.635 are known, the initial example reported by Lappert and co-workers as early as 1986, as well as a bis(germylene) [Ru(CO)₃] complex reported by Cabeza, García-Álvarez and co-workers (viz. 3.636).^242,294,295

	Scheme 155 Generation of iron(0) and ruthenium(0) complexes bearing acyclic germylene ligands.

The first observation of terminal germylene formation through germane activation was reported by Peters and co-workers, in the addition of Ph₂GeH₂ to Ru complex 3.637, in which one Ph group of the ligand is activated. Addition of the germane reprotonates this ligand moiety, with the second Ge–H fragment transferred to the Ru centre in forming germylene complex 3.638 (Scheme 156).¹⁷² In a similar methodology, Tilley and co-workers soon after reported that the addition of (Trip)GeH₃ to [CpRu(PMeⁱPr₂)(CH₂Ph)], which led to terminal (aryl)(hydrido)germylene complex 3.639, again through protonation, in this case of the benzyl fragment, and hydride migration to Ru (Scheme 157).¹⁷⁰ This is also akin to earlier described silylene chemistry reported in the same submission, which has implications in alkene hydrosilylation catalysis.⁵⁹ As with earlier examples, the addition of H₂O to germylene complex 3.639 leads to addition across with Ru–Ge bond, with hydroxyl group at Ge, forming 3.640. This further indicates the Lewis acidity of such germylene ligands, and feeds into potential applications in cooperativity. Later, Tilley reported a further example of a hydrido germylene complex, 3.641, which forms on addition of alkyl germane ^tBuGeH₃ to [{PhB(CH₂PPh₂)₃}Ru(CNAr)(Mes)].²⁹⁶ A final example of a Ru system has been reported, generated via addition of Ph₂GeH₂ to Ru–hydrogen complex [Ru(PCy₃)₂(H₂)₂(H)₂], leading to bis(phenyl)germylene complex 3.642, demonstrating that bulky aryl ligands aren’t a requirement for the synthesis of such systems.²⁹⁷

	Scheme 156 Formation of a diphenylgermylene complex of ruthenium via germane dehydrogenation.

	Scheme 157 Further examples of ruthenium germylene complex formation through H2 extrusion.

Moving to iron, Tobita and co-workers extended the described methodologies to this lighter metal in accessing hydrido germylene complex 3.643, through deprotonation of the bulky alkyl germane [(Me₃Si)₃C]GeH₃ with methyl iron(II) complex [Cp*Fe(CO)(Py)Me] (Py = pyridine; Scheme 158).²⁹⁸ Addition of ligands at Fe led to migration of the Fe–H fragment back to Ge, forming germyl complex 3.644, 3.645, and 3.646, whilst addition of aldehydes and ketones led to insertion into the Ge–H fragment (viz 3.647, 3.648, 3.649, and 3.650). Most recently, our group has demonstrated ready access to carbonyl- and Cp-free iron(0) complexes (Scheme 159). The addition of cationic germylene 3.651 to [IPr·Fe·{η²-(vtms)}₂] leads to alkene elimination and NHC migration to Ge, in the formation of 3.652. This is best described as a 2-coordinate acyclic germylene complex, given that the cationic charge likely resides on the NHC ligand (i.e. as the imidazolium), which features a formal Ge–Fe double bond.²⁹⁹ Despite this, addition of NH₃ to 3.652 leads exclusively to binding at Ge, and proton transfer the carbene–carbon, generating bis(amido)germylene complex 3.653.

	Scheme 158 Formation of an iron complexes bearing an (aryl)(hydrido)germylene, and reactivity towards insertion and base coordination.

	Scheme 159 Synthesis of a cationic germylene iron(0) complex, and its cooperative activation of ammonia.

As per above described Ru chemistry, Tilley and co-workers have shown that addition of diaryl germane to allyl iridum complex [{PhB(CH₂PPh₂)₃}IrH(C₈H₁₃)] leads to the formation of germylene species 3.654 (Scheme 160).¹⁷⁶ Whilst that reaction is high yielding, it is interesting to note that Lappert's attempt to generate an iridium germylene complex some years earlier led to C-H activation of the ligand, when utilising the [(Me₃Si)₂N]⁻ scaffold.³⁰⁰ Closely related Ir germylene complexes were recently described by Wesemann and co-workers, via the cooperative activation of a range or protic small-molecules (NH₃, N₂H₄, H₂O, H₂, HCl) in pseuso-germylyne complex 3.655, formed through bromide abstraction from (aryl)(bromo)germylene complex 3.656 (Scheme 161). We describe complex 3.655 as such due to the 2-coordiante Ge-centre, its bent geometry, and positive charge localisation at Ge (i.e. this is not a ‘true’ germylidyne, nor a metallogermylene). In all cases, the nucleophile binds at Ge, leading the proton transfer to the Ir centre, in the formation of 3.657, 3.658, 3.659, 3.660, and 3.661.³⁰¹

	Scheme 160 Formation of a bis(aryl)germylene iridium complex via H2 extrusion.

	Scheme 161 Formation of neutral and cationic iridium(I) complexes bearing gerlymene ligands, and the cooperative activation of numerous protic species across the Ir–Ge bond.

A single example of a rhodium complex bearing an acyclic germylene ligand is known, that is [GeCl₂] complex 3.662, simply accessed through addition of [dioxane·GeCl₂] to [Cp*Rh(PMe₃)₂], where interestingly the [GeCl₂] moiety behaves as a Z-type ligand, receiving electron density from Rh (Scheme 162).³⁰² Two examples of related cobalt complexes are known. The first involves the direct addition of bis(amido)germylene 3.605 to [Co₂(CO)₈], leading to end-on coordination of the germylene ligands at each Co centre (3.663, Scheme 163).³⁰³ The second such example sits on the boundary between a formal tetrylene Co complex, and a metallotetrylene, but it is nevertheless described here. Addition of (aryl)(chloro)germylene 3.617 to the magnesium cobaltate 3.664 leads to metathesis in the formation of 3.665, in which the germanium centre forms a close contact with on P-atom diphosphacyclobutadiene ligand, and a long Ge–Co contact is observed (Scheme 164).³⁰⁴ It is likely, then, that the Ge^II centre in this species is formally ligated by the aryl and diphosphacyclobutadiene ligands, and forms a dative interaction with the Co centre.

	Scheme 162 Germanium(II) chloride behaving as a Z-type ligand in the coordination sphere of rhodium.

	Scheme 163 Formation of a dimeric cobalt(0) complex bearing two acyclic bis(amido)germylene ligands.

	Scheme 164 Formation of a cobalt complex bearing a diphosphacyclobutadiene-appended acyclic gerlymene ligand.

Given the ready accessibility of M⁰ species of the group 10 metals, a considerable number of acyclic germylene complexes are known for these elements. This is made more interesting but the combined capacities of germylenes to behave as Lewis acids (i.e. Z-type ligands), and the group 10 metals to behave as Lewis bases (i.e. L-type ligands). Initial reports regarding this compound class came from the group of Holl, with a series of three publications giving key insights into both the bonding and reactivity of such species (Scheme 165). In the first submission, the Pt⁰ germylene complex 3.666 was described, bearing the bis(amido)germylene developed by Lappert, 3.605.⁴⁷ This remarkable species, accessed through CO₂ elimination from a Pt^II oxalate, demonstrated both the reversible [2+2] cycloaddition of CO₂ and reversible H₂ addition across the Ge–Pt bond (viz. 3.667 and 3.668). This was an early demonstration of the cooperative bond activation capacities of such heteroelemental bonds, and remains a central point of investigation in the implementation of low-valent p-block ligands to this day. The described [2+2] cycloaddition chemistry was later extended to arylnitroso species.³⁰⁵ Subsequently, the same group showed that the closely related Ni⁰ complex (viz. 3.669) can be readily accessed, and behaves as an efficient catalyst for the hydrogenation of the germylenes 3.604 and 3.605.³⁰⁶ It's important to note here that addition of H₂ to isolated Ni complex 3.669 did not lead to an isolable H₂ activation product, highlighting both the differences in reactivity between Ni and the heavier group 10 metals, as well as the importance of ligand design in developing such chemistry for the more abundant base metals. In their third submission on the topic, Holl and co-workers extended their Ni⁰ system to the (fluroaryl)germylene 3.670, in the synthesis of complex 3.671 (Scheme 166).³⁰⁷ This report did not describe further reactivity studies, but rather focused on the electronic structure and bonding in this species when compared with more electron rich germylene systems. The key finding was a significant degree of π-character in the Ge–Ni bond, with the strongly electron withdrawing nature of the [2,4,6-(CF₃)₃C₆H₂] increasing Ni → Ge back-bonding relative to the bis(amido) derivative 3.669. This bonding nature is additionally borne out by the planarity of the Ge centre in these systems.

	Scheme 165 Synthesis and reactivity of an acyclic bis(amido)germylene–platinum(0) complex.

	Scheme 166 Synthesis and reactivity of various acyclic germylene–nickel(0) complexes.

It was later shown by a number of groups that acyclic–germylene complexes of group 10 metals can feature pyramidalised Ge centres, which is indicative of a predominant M → Ge donation, and little or no ‘classic’ L-type ligand behaviour of the germylene. This bonding motif is also thought be brought about by ‘double-σ’ bonding, with simultaneous σ-donation from M → Ge and Ge → M,³⁰⁸ reminiscent of the bonding model leading to trans-bending in the heavier alkene analogues.⁵⁵ The first such example for Ge was reported by Braunschweig and co-workers, who showed that both [(Cy₃P)₂Pt] and [(Cy₃P)(IMes)Pt] react with [dioxane·GeCl₂], to yield germylene complexes 3.672 and 3.673 in which the Ge centre is paramidalised (Scheme 167).³⁰⁹ Similarly, it was later shown that the phosphine-functionalised bis(aryl)germylene 3.674 reacts with [Ni(cod)₂] in the formation of 3.675, with a pyramidalised Ge centre (Scheme 168).³¹⁰ Here, accompanying computational analysis located both Ge → Ni and Ni → Ge σ-bonding, forming the so-called σ-double-bond.

	Scheme 167 Dichlorogermylene complexes of platinum(0), and the suggested double-σ-bonding model.

	Scheme 168 Formation of an acyclic–germylene nickel(0) complex, featuring a Lewis basic germanium centre.

In contrast, complex 3.676, synthesised through the reduction of a bis(chloro-germylene) Ni⁰ complex 3.677, also bears a pyramidalised Ge centre, but rather with a Z-type ligand character (Scheme 169).²²⁹ That is, the Ni⁰ centre in this species is Lewis basic, and is thus an L-type donor towards Ge. This bonding character is particularly rare for Ni⁰, and was first formally demonstrated by Figueora and co-workers in 2011.³¹¹ More recently, T-shaped Ni⁰ species featuring Z-type germylene ligands have been realised, in which this geometry at Ni⁰ is strongly indicative of Lewis basic character, as per numerous known Lewis basic Pd and Pt systems.^312–314 The first such example, reported by Roesler and co-workers, utilises the chelating bis(NHC) stabilised [Ni(cod)] species 3.678, which reacts with [dioxane·GeCl₂] in forming the Ni-T-shaped dichlorogermylene complex 3.679 (Scheme 170).²⁵⁰ Here, the Ge-centred lone pair of electrons was found by computational analysis to not partake in a bonding interaction with Ni, with the only distinct bonding orbitals resembling Ni → Ge donation.

	Scheme 169 Nickel complexation of a chelating amidinato-germylene, leading a Z-type acyclic–germylene complex.

	Scheme 170 Formation of a T-shaped geometry at nickel(0), driven by the Lewis acidic nature of a dichlorogermylene ligand.

In our own work, we recently showed that chelating cationic germylene ligands also drive a similar bonding mode in complexation with Ni⁰ (Scheme 171).³¹⁵ Addition of phosphine-chelated cationic germylenes 3.651, 3.680, and 3.681 to [IPr·Ni·{η²-(vtms)}₂] gave ready access to complexes 3.682, 3.683, and 3.684 through alkene displacement. All complexes were found by structural analysis to feature a T-shaped geometry at Ni⁰. This was shown to be highly dependent on the Lewis acidity of the Ge centres. Addition of N-bases to these systems led to site-selective binding at Ge, e.g. in DMAP adduct 3.685. Typical π-acceptor ligands (CO (3.686), CyNC (3.687)) were shown to bind at Ni⁰, forming tetryhedral complexes. In the former systems, the now more electron rich germylene ligands switch the geometry at Ni from T-shaped to trigonal planar. This thus demonstrates a switch from Z-type to L-type ligand character for Ge^II, with clear structural effects resulting from this change. Moreover, all described T-shaped complexes were shown to be active catalysis for the hydrogenation of alkenes, whist trigonal planar complexes were not, opening up a new vista for tetrylene ligands in switchable catalysis. The closely related 18-electron Ni⁰ species 3.688 and 3.689, also bearing our developed phosphine-funcationalised cationic germylene ligands are also potent in catalysis, in this case for the hydrosilylation of alkenes.³¹⁶ These complexes were shown to be accessible by differing routes (Scheme 172): (i) the free cationic germylenes can be directly combined with a mixture of [Ni(cod)₂] and PPh₃; (ii) the hydride, hydroxide, or amide complexes 3.690, 3.691, and 3.692 can also be treated with the oxonium species [(Et₂O)₂H][BAr^F₄], in the elimination of H₂, H₂O, or NH₃, respectively; or (iii) the chloride ligand can be abstracted from chloro-germylene complexes 3.693 and 3.694.

	Scheme 171 Formation and reactivity of a T-shaped nickel(0) complex, enforced by both geometric constraint and high Lewis acidity of the germylene ligand.

	Scheme 172 Formation of cationic germylene complexes of nickel(0) via various routes.

In addition, the neutral chlorogermylene complex 3.693 and 3.694 undergoes non-redoxactive σ-metathesis of the Ge–Cl bond with the N–H bond in ammonia, a reaction in which the added ammonia simultaneously acts as a base in abstraction of HCl, forming bis(amido)germylene complexes 3.692 and 3.695 (Scheme 173).³¹⁷ This reaction is reversible, highlighting the potential for such germylene ligands to behave as bifunctional ligands which can bind a metal centre, whilst also reacting as a Lewis acidic binding site.

	Scheme 173 Germylene-centred reversible ammonia activation in an acyclic–germylene nickel(0) complex.

It was additionally recently shown by Campos and co-workers that the NHC-bound (aryl)(chloro)germylene complex 3.696 reacts with [Ni(cod)₂], leading to NHC-migration to and arene coordination at Ni, in complex 3.697 (Scheme 174).³¹⁸

	Scheme 174 Formation of an acyclic–germylene complex of nickel(0) through Ge–to–Ni NHC migration.

Gold complexes which may be described as cationic germylene complexes have been reported by Campos and co-workers, as the only such species for the coinage metals. These were generated by first addition of (aryl)(chlor)germylene 3.698 to bulky phosphino-gold chloride 3.699, leading to germyl-gold complex 3.700 via gold-to-germanium chloride migration (Scheme 175). A simple chloride-abstraction from these (chloro-germyl)gold(I) complexes led to the target cationic complex 3.701 and 3.702.³¹⁹ The Lewis acidic nature of the Ge centre these species was demonstrated by Ge-centre DMAP binding, leading to adducts 3.703 and 3.704.

	Scheme 175 Generation of cationic gold(I) complexes bearing (aryl)(chloro)germylene ligands, and their Ge-centred Lewis base binding chemistry.

	Scheme 176 Synthesis of chelating bis(germylene) ligands featuring 5-membered NHGe moieties.

	Scheme 177 Formation of molybdenum complexes of chelating bis(NHGe) ligands.

Some years later, Driess and co-workers reported the bis(germylene) ligand 3.714 built upon the 4,6-di-tert-butylresorcinol scaffold, analogous to the earlier described Si system 3.296. This germylene ligand was initially utilised in generating iridium complex 3.715, and the catalytic borylation of arenes studied, the reaction rate for the germylene system being slightly slower than that of the silicon derivative (Scheme 178).¹⁸⁸ Later, the same ligand system was utilised in forming Ni complex 3.716, which was employed in the Sonogashira cross-coupling reaction.¹⁸⁹ Here, the intermediary copper acetelides were shown to form a bimetallic complex with the catalyst, supported by coordination to one Ge^II centre. This was also achieved with the earlier described silylene complex 3.310. This incites that the tetrylene centre is these systems may play a non-innocent role in catalysis, allowing close spatial arrangement of substrates. Further extending the library of chelating bis(germylenes) which utilise the amidinato ligand at Ge^II, the bis(amino) pyridine-derived ligand 3.717 was developed, reported along-side the silicon analogue 3.311. The bis(germylene), as for the Si system, was employed in accessing complexes with [FeCl₂], [Fe(PMe₃)₂], and [Fe(CO)₂], though catalytic efficacy was not described (viz. 3.718, 3.719, and 3.720; Scheme 179).¹⁹³

	Scheme 178 Formation of iridium and nickel complexes bearing a resorcinol-derived chelating bis(germylene) ligand.

	Scheme 179 Formation of iron(0) and iron(II) complexes bearing a bis(amino)pyridine-derived chelating bis(germylene) ligand.

The above described systems represent the only well-defined transition metal complexes of chelating bis(germylenes). Though these are few, additional interest in bis(germylene) ligands may lead to an expansion of this number.³²²

	Scheme 197 Divergent behaviour of bis(amido)- and bis(alkyl)-stannylene ligands towards tungsten ligation.

	Scheme 198 The first examples of crystallographically characterised acyclic stannylene complexes if iron(0), palladium(0), and platinum(0).

	Scheme 199 Formation of palladium and platinum trimers supported by acyclic bis(amido)stannylene ligands.

No acyclic stannylene complexes of the group 3 elements are known, but a handful of examples have been accessed for the group 4 elements (Scheme 200). The first of these, complex 3.805, was synthesised through the in situ generation of zirconocene (i.e. [Cp₂Zr]), which is trapped by two equiv. of Lappert's bis(alkyl)stannylene.³⁵⁴ 3.804 features slightly contracted Zr–Sn interactions when compared with the sum of the covalent radii for these elements, indicative of a degree of Zr → Sn π-backbonding. The amine-functionalised stannylene 3.806 was later utilised in accessing the closely related bis(stannylene) complex 3.807, using a similar synthetic route.³⁵⁵ Wesemann and co-workers later reported hydrido-stannylene group 4 complexes, accessed using a different synthetic strategy. Here, hydridostannyl lithium species 3.808 was shown to react with metallocene-dichloride complexes [Cp₂MCl₂] (M = Ti–Hf), leading to salt-metathesis and H₂ elimination, forming bis(stannylene) complexes 3.809, 3.810, and 3.811.³⁵⁶ All complexes exhibit low-field shifted Sn–H signals in their ¹H NMR spectra (12.63–13.27 ppm) when compared with the free hydrido-stannylene (7.87 ppm),³⁵⁷ but in keeping with other reported hydrido-stannylenes.²¹ It was also demonstrated that one hydride ligand could be abstracted from titanium complex 3.801, leading to cationic 3.812, with one Ti-Sn-L angle now close to linearity indicative of amplified Ti → Sn π-back bonding.

	Scheme 200 Synthesis of bis(stannylene) complexes of group 4 metallocenes.

The same group later reported the closely related tantalum complex 3.813, being the only representative example of a group 5 acyclic stannylene complex (Scheme 201).³⁵⁸ This complex was synthesised via the addition of cationic stannylene 3.814 to [CpTaH₃], leading to hydride migration from Ta to Sn. Here, the observed ¹H NMR shift for the Sn–H (15.55 ppm) is considerably down field when compared with described group 4 derivatives.

	Scheme 201 Synthesis of a (aryl)(hydrido)stannylene adduct of tantalum.

The first examples of group 6 complexes featuring acyclic stannylenes were reported by Lappert and co-workers, utilising their bis(amido)stannylene ligand 3.797. The addition of two equiv. of this ligand to [Cr(CO)₆] led to the bis(stannylene) complex 3.815, with stannylene ligands in a cis-arrangement (Scheme 202).³⁵⁹ Soon after this report, it was shown that bis(aryl)stannylene 3.816 also reacts with the group 6 metal(0) hexacarbonyl complexes. This situation is slightly more complex, however. 3.813 exists as the distannene form, which dissociates in solution leading to H-migration/rearrangement in one [2,4,6-^tBuC₆H₂] group, forming an (alkyl)(aryl)stannylene which reacts with [M(CO)₆] to form complexes 3.817, 3.818, and 3.819.^360,361 Considerably later, in 2021, Power and co-workers showed that their (aryl)(hydrido)stannylene 3.820 reacts with [Mo(CO)₅·THF] in forming complex 3.821 (Scheme 203).³²⁹ This species undergoes CO₂ insertion into the Sn–H bond under non-forcing conditions forming earlier described 3.714, which was extended to the catalytic hydroboration of CO₂ with HBpin. The directly analgous hydrido-stannylene tungsten complex (viz. 3.822) was also accessed through the same synthetic methodology, whereby insertion of alkenes into the Sn–H bond was demonstrated, forming 3.823 and 3.824.³⁶² Attempts were made to extend this to alkene hydrosilylation catalysis, but it was found computationally that the barrier to Sn–C σ-metathesis was to great, at more than 70 kcal mol⁻¹. A similar (aryl)(hydrido)stannylene complex of tungsten was also reported by Wesemenn and co-workers, 3.825, which was accessed through the addition of stannylene cation 3.813 to the tungsten dihydride species [Cp₂WH₂], followed by reaction with EtMe₂N.³⁵⁸ The hydride moiety in 3.823 inserts alkenes in forming 3.826, as per the above described complex reported by Power and co-workers, and can also be abstracted by an NHC in forming a neutral metallostannylene (vide infra).

	Scheme 202 Formation of group 6 complexes bearing acyclic stannylene ligands.

	Scheme 203 Synthesis of molybdenum and tungsten complexes bearing (aryl)(hydrido)stannylene ligands, and subsequent insertion chemistry of the Sn–H bond.

The initial example of a group 7 acyclic stannylene complex involved the rearranged bis(aryl)stannylene 3.815; two equiv. of the isomerised form react with [HMn(CO)₅], leading to complex 3.827 in which one stannylene ligand binds Mn, and the second undergoes C–H activation, presumably in loss of H₂, forming a novel stannyl ligand (Scheme 204).³⁶³ The somewhat related phosphine-stabilised Mn^I species 3.828 has also been used in accessing the chloro-stannylene complex 3.829, which was utilised as a precursor to a heavier alkylidyne derivative (vide infra).³⁶⁴

	Scheme 204 Formation of a acyclic stannylene ligated manganese complexes.

Numerous group 8 complexes in this class are known, the initial example being the bis(aryloxy)stannylene complex 3.798 describe above (Scheme 198). Following this, two examples were reported featuring the isomerised form of bis(aryl)stannylene 3.808, namely 3.830 and 3.831, with [Fe(CO)₄] and [Fe(CO)(NO)₂], synthesised through direct addition of the stannylene ligand to [Fe₂(CO)₉] and [Fe(CO)₂(NO)₂], respectively (Scheme 205).^363,365 Schneider and co-workers later reported on the synthesis of toluene-bound iron(0) complexes of acyclic stannylene ligands, employing the metal-vapour synthesised iron(0) complexes (Scheme 206). In an initial report, it was shown that the bis-ethylene derivative 3.832 reacts with one equiv. of bis(alkyl) or bis(aryl) stannylenes to yield mono-stannylene complexes 3.833 and 3.834.³⁶⁶ Further substitution of the remaining ethylene ligand apparently is not possible. However, it was later reported that the direct reaction of the bis(toluene) complex 3.835 with two equiv. of the same stannylene ligands did lead to the bis-ligated complexes, 3.836 and 3.837.³⁶⁷ Addition of carbon monoxide to these various species led to a range of mono- and bis-stannylene complexes of [Fe(CO)_n] (n = 1 (3.838); 3 (3.839 and 3.840); 4 (3.841 and 3.842)).

	Scheme 205 Formation of iron complexes of a rearranged bis(aryl)stannylene ligand.

	Scheme 206 Ready formation of acyclic-stannylene iron(0) complexe utilising toluene-stabilised iron(0) synthons.

Fe, Ru, and Os complexes of acyclic stannylenes have all been reported by the group of Tilley, generally featuring a Cp*M (M = Fe–Os) fragment at their core. The first report described the synthesis of the osmium system, which was achieved through the reaction of TripSnH₃ to the allyl osmium complex [Cp*Os(ⁱPr₃P)CH₂Ph)], leading to elimination of toluene, and H-migration to Os in forming the (aryl)(hydrido)stannylene complex 3.843 (Scheme 207).³⁶⁸ Reactivity studies indicated a Lewis acidic Sn centre, indicated by both base coordination (3.844) and MeOH(D) addition across the Os–Sn bond (3.845 and 3.846) as well as tautomerisation to the metallostannylene form via a further H-migration (vide infra). This was later followed by the synthesis of the (aryl)(chloro)stannylene ruthenium complex 3.847, generated by direct addition of the stannylene ligand to the N₂ complex [Cp*Ru(ⁱPr₂MeP)(H)(N₂)], in loss of gaseous N₂ (Scheme 208).³⁶⁹ Here, it was found that the hydrido-stannylene complex cannot be isolated, given the significantly greater stability of the metallostannylene tautomer. A similar case was later observed for iron, where the (aryl)(chloro)stannylene complex 3.848 could be isolated, but the hydrido derivative also readily rearranges to the metallostannylene form (vide infra).³⁷⁰ Oxidation of this species led to H₂ elimination, which undergoes ligand C–H activation when re-reduced in the absence H₂, forming novel chelating (alkyl)(aryl)stannylene complex 3.849.

	Scheme 207 Formation of an osmium complex bearing an (aryl)(hydrido)stannylene ligand, and its subsequent reactivity.

	Scheme 208 Synthesis of ruthenium and iron complexes bearing an (aryl)(chloro)stannylene ligand, and the further reactivity of the iron species leading to a novel phosphine-appended stannylene ligand.

More recently, our group recently reported a cationic stannylene iron(0) complex and its revesible H₂ activation, featuring our developed chealting phosphine functionalised amido scaffold (Scheme 209). This complex was accessed through the addition of phosphine-stabilised cationic stannylene 3.850 to the iron(0) complex [IPr·Fe·{η²-(vtms)}₂], leading to loss of the two alkene ligands as per the earlier described germanium system and forming 3.851.²⁹⁹ This complex demonstrates the reversible activation of H₂ across the Sn–Fe bond, leading to bridged hydride complex 3.852, under 1.5 atm of H₂ at room temperature. 3.850 can also be reduced using a further equivalent of the Fe⁰ precursor, leading to the first example of a covalently bound iron(−I) compound (vide infra).

	Scheme 209 Formation of a cationic stannylene iron(0) complex, and its reversible H2 activation.

Similar to toluene-bound iron complexes 3.832 and 3.833, Schneider and co-workers demonstrated that reaction of [CpCo(η²-C₂H₄)₂] with bis(alkyl) and bis(aryl) stannylenes leads to mono-substitution of ethylene at ambient temperature, forming complexes 3.853 and 3.854 (Scheme 210).^371,372 Heating the bis(aryl)stannylene complex 3.853 under mass spectrometric conditions, above 75 °C, leads to ethylene elimination in generating dimeric [(Cp*CoSnR₂)₂] (R = CH(SiMe₃)₂), though this was only observed in the mass spectrum. Some years later, Campos and co-workers reported the [SnCl₂] adduct 3.855, in demonstrating the broad Lewis basic properties of the [Cp*Rh(PMe₃)₂] (Scheme 211).³⁰² Wesemann and co-workers also described the reaction of the stannyl lithium species 3.814 with Rh^I chloride [(Ph₃P)₃RhCl], leading to double H-migration to Rh, in forming hydride-bridged metallostannylene 3.856, which can also be described as having some degree of (aryl)(hydrido)stannylene character (Scheme 212).³⁷³ Similar to the germanium(II) system previously described, the (aryl)(diphosphabutadienyl)stannylene complex 3.857 can be accessed by reaction of the magnesium colbaltate complex with the corresponding chloro-stannylene (Scheme 213).

	Scheme 210 Formation of a cobalt(I) complex bearing an acyclic stannylene ligand.