Non-heteroatom-substituted alkylidene complexes of groups 10 and 11

1 Non-Heteroatom-Substituted Alkylidene Complexes of Groups 10 and 11 Riccardo Peloso,* Ernesto Carmona Instituto de Investigaciones Químicas (IIQ), Departamento de Química Inorgánica and Centro de Innovación en Química Avanzada (ORFEO-CINQA), Universidad de Sevilla and Consejo Superior de Investigaciones Científicas (CSIC) Avenida Américo Vespucio 49, 41092 Sevilla (Spain) E-mails: [email protected], [email protected] Keywords: carbenes; alkylidenes; organometallic chemistry; group 10 metals; group 11 metals. Abbreviations: Mes = 2,4,6-trimethylphenyl Xyl = 2,6-dimethylphenyl p-Tol = 4-methylphenyl Tf = trifluoromethylsulphonyl EDA = ethyl diazoacetate Highlights: • use of terms “carbene” and “alkylidene” in organometallic chemistry is discussed • unusual species included are cationic Ni(III), Pd(II), Pt(IV) and Ag(I) alkylidenes • group 10 and 11 alkylidene complexes with no heteroatom stabilization are reviewed 1. Abstract. This review aims to offer a detailed summary of the discoveries regarding nonheteroatom-substituted alkylidenes of the groups 10 and 11 metals that have appeared in the literature since the first representative of this class was reported in 2002. Special attention is focused on synthetic and bonding aspects of well-defined species. The importance of this family of compounds in catalytic organic transformations is often 2 mentioned and a number of references are provided. We stress, however, that it is not the main issue of this review, as this topic has received considerable attention in the current literature. 2. Introduction and Scope. Transition metal complexes with carbene ligands have expanded over the years to constitute one of the most important families of organometallic compounds. Since the development of N-heterocyclic carbenes (NHCs)1, and more recently of the also highly versatile cyclic(alkyl)(amino)carbenes (CAACs)2, carbenes have been employed as ancillary ligands to provide thermodynamic and kinetic stability to countless transition metal complexes and reaction intermediates. These carbenes have become indispensable partners of d transition elements and main group metals alike, and perform an invaluable function in homogeneous catalysis.1,2 Notwithstanding the importance of carbenes as spectator ligands, the vital role that transition metal carbenes play in modern chemistry rests largely on the extraordinarily rich and diverse reactivity of the M=C bond, that allows their successful applications in diverse organic transformations whose importance cannot be overstated, not least in homogeneous catalytic olefin metathesis.3-10 The explosive growth of the field has naturally given rise to extensive literature coverage. A comprehensive annual survey on the progress of metal-carbene chemistry is provided by Herndon in this Journal, with the latest, covering the year 2015, appearing in the second half of 2016.11 In this article we concentrate on carbenes of late transition metals. Complexes of Fe, Ru and Os with carbene ligands have been known for several decades. Similarly, a variety of carbene complexes of Group 9 elements have attracted considerable attention. In particular, we and other researchers have described a number of complexes of Ir(III) with carbenes of both the :CRR’ and :CRX types (X = heteroatom substituent).12-15 Accordingly, as detailed later, the aim of this review is to offer a concise summary of complexes of groups 10 and 11 metals with carbene ligands devoid of heteroatoms. Before we discuss this chemistry in due detail, a word is perhaps necessary on the terminology we use in this review. The comments we will make in this regard will smoothly lead us to some, necessarily brief, considerations on the reactivity modes exhibited by the metal-carbene bond and on the bonding models supporting this chemistry. 3 The terms “carbene” and “alkylidene” are often used as synonymous by coordination and organometallic chemists. Even though both indicate an organic fragment that binds to a metal centre through a divalent 6-electron carbon atom, :C(R)R’, this causes confusion and creates ambiguity. The origin of the above terms is as follows. In 1964, Fischer and Maasböl prepared the first transition metal carbene,16a the tungsten complex W(C(OMe)Me)(CO)5, that was soon followed by other related compounds.16b,c These complexes are 18-electron species in low oxidation states. They are reasonably stable and feature heteroatoms at the carbene atom, which generally exhibits electrophilic reactivity. Only four years later, Öfele reported a surprisingly stable chromium complex containing also a Cr(CO)5 fragment, but bound to a cyclopropylidene unit, i.e. a carbene ligand with no heteroatom.17 Subsequently, Schrock discovered a new class of metal carbenes, exemplified by the 10-electron species Ta(CHCMe3)(CH2CMe3)3. In the series of complexes made by Schrock and coworkers,18 an early transition metal in high oxidation state, with 10 to 16 valence electrons, is bonded to a nucleophilic carbene carrying C or H substituents at the carbon atom. It soon became customary to name the two classes of metal carbenes as Fischer and Schrock carbenes, respectively. In the following years many compounds of general formula LnM=CRR’, where R and R’ are H, alkyl or aryl, were prepared and named alkylidenes as derivatives of alkyls,19 regardless of their nucleophilic or electrophilic reactivity. In this way, it became common practice to use the term “alkylidene” when referring to complexes with H, alkyl or aryl substituents on the carbon, and “carbene” to refer to species with one or two heteroatom substituents.20a But this was not uniformly accepted by researchers in the field. Many, utilized carbene and alkylidene for Fischer and Schrock complexes, respectively, whereas others employed only carbene, or only alkylidene, for all metal carbenes, or even used the two terms interchangeably. The nomenclature committee of the IUPAC recommended that carbene should be reserved for describing the free ligand, naming all transition metal complexes of carbenes as substituted alkylidenes.21 As it is evident nowadays, these recommendations have not been standard practice among transition metal carbene chemists. However in his recent book22a Bochmann states that the systematic term “alkylidenes” is now applied to all types of carbenes. Moreover, the expression alkylidene complexes in low oxidation states is employed as equivalent to 4 Fischer carbenes, while alkylidene complexes in high oxidation states (dº) is meant to describe Schrock carbenes.22a The real distinction between the two families of carbene complexes resides doubtless in the nucleophilic or electrophilic reactivity they may display.22 As lucidly compiled and discussed by Roper some 30 years ago,20 similar to other chemical reactions, those of coordinated carbenes are influenced by frontier orbital-controlled and charge-controlled factors.23 Thus, despite the M+-C- polarity of the metal-carbene bond that may be suggestive of nucleophilic behaviour, the reactions of Fischer carbenes (e.g. Cr(=CXY)(CO)5) with nucleophiles are frontier-orbital controlled. Hence, electron density transfer from the HOMO of the nucleophile to the carbene complex LUMO, that resides largely on C, disrupts the usually reduced  interaction found in these complexes and converts the metal-carbene bond into a metal-alkyl one. Conversely, when the metal-carbene bond has high  character as in Schrock carbenes, electrophilic attack at the carbene carbon atom is expected, particularly when this atom is most negatively charged and when the  bonding orbital has strong carbon-p character.20b An alternative interpretation of carbene reactivity based on quantum-mechanic calculations was offered by Hall in 1984.24 In this approach the electrophilic or nucleophilic reactivity of the M=C bond can be explained with the aid of different bonding schemes (Figure 1). Using the isolobal analogy, 18-electron metal carbenes may be viewed as arising from the bonding of singlet metal and singlet carbene fragments, i.e. from donor-acceptor  and  interactions (Figure 1a).20,22 Clearly, the resulting dative covalent bonds25 would be akin those in M-CO linkages, such that for the singlet-bonding, donor-acceptor metal carbene scheme, electrophilic reactivity can reasonably be expected. Moreover, singlet electronic states would be favoured by strong  acceptors at the metal and by heteroatom substituents on the carbene, which increase the HOMO-LUMO gap in the carbene ligand.20 5 Figure 1. Bonding situations in metal complexes with carbene ligands.22 (a) Singlet metal and singlet carbene fragments proposed for low oxidation state Fischer carbenes. (b) The triplet metal and carbene building blocks in high oxidation state alkylidene complexes (Schrock carbenes). In contrast, nucleophilic, often electron-deficient metal carbenes, can be electronically associated with triplet electronic states of their metal and carbene building units (Figure 1b), resulting in the formation of two normal covalent bonds,25 and in olefinic-like, i.e. nucleophilic reactivity of the M=C bond. In this case, triplet states can be foreseen for metal fragments that lack strong  acceptor ligands, as well as for carbenes bearing only hydrogen or hydrocarbyl substituents, particularly alkyl groups. Naturally, intermediate cases appear when a metal fragment of one type combines with a carbene ligand of the other. These “hybrids” of Fischer and Schrock complexes26 have been known since the early days of transition metal carbene chemistry and continue at present to enrich our knowledge of the field.27 Some early examples were calculated to have reduced bond strength, reflecting the cost in energy of promoting one fragment to an excited state of the correct multiplicity.20a In addition, and in good agreement with experimental observations, these considerations support the expectation of a wide range of reactivity modes for the metal-carbene bond, rather than a clear-cut differentiation between electrophilic and nucleophilic patterns. To complete this brief lexical overview, it is worth mentioning that organic species formally bearing a positive charge and a carbon-metal bond at the same carbon atom are often called “carbenoids”.3a This definition is particularly useful for electrophilic late-transition alkylidene complexes for which Fürstner recommends to avoid the notation [M=C], as the presence of a metal-carbon double bond does not find any experimental evidence. Alkylidene complexes with no heteroatoms are particularly stable and easy to prepare for early-transition elements. By contrast, for late-transition metals the synthesis of alkylidene complexes is more difficult, particularly for cationic compounds, and isolating species of this kind remains a challenging goal in many cases. It is also worth mentioning the peculiar reactivity of late-transition metal alkylidene complexes which, in spite of their structural analogy with Schrock carbenes, often, albeit not always (vide infra), show strongly electrophilic behavior, typical of Fischer carbenes. 6 Figure 2. Schematic representations of metal-carbene bonds (R, R’ = H, hydrocarbyl; E, E’ = heteroatom) Figure 2 shows four possible representations of a metal carbene-bond. Type (a) depicts a metal-carbon double bond, which is typical of Schrock-carbenes. The second representation (b) is most commonly used by organic chemists in proposed catalytic intermediates and indicates a metal-stabilized carbocation (carbenoid), stressing its intrinsic electrophilicity. Model (c) emphasizes the  component of the M-C bond, typical of Fischer-carbenes, and is predominantly used for NHC carbenes, for which the  contribution to the bonding is small or even negligible. Finally, the last two representations (d) and (e) are quite uncommon and indicate nucleophilic nonheteroatom stabilized alkylidenes and carbenoids, respectively. Regrettably, no strict rules for the use of these representations have been established yet. As briefly mentioned earlier, in this article we present an overview of the chemistry of alkylidene complexes of late transition metals. Regarding the metal centre, we concentrate exclusively on the latest transition elements, Ni, Pd and Pt, and Cu, Ag and Au. Concerning the ligands, only non-heteroatom-substituted carbenes are considered, although R and R’ of the M=CRR’ unit might contain heteroatoms not directly bound to the carbene carbon atom and/or form part of a metallacyclic structure. Cycloheptatrienylidene or cylclopropylidene complexes, which have mainly been described for group 10 metals, are also excluded from this review in light of their special bonding situation dominated by the involvement of the carbon p orbital in the aromatic system. Although in most cases the coordinated carbene ligand features electrophilic reactivity, i.e. Fischer-type behavior, some highly relevant examples of nucleophilic alkylidenes (Schrock complexes) have been reported in the last two decades. Special emphasis is made on synthetic and reactivity aspects concerning isolated and well-characterized species. The first part of each section is dedicated to the first discoveries in the field, while the following discussions are centered on more recent findings. As the following sections will make clear, these complexes constitute a 7 young family of compounds, for the majority of its members have been discovered in the years elapsed since the turning of the century. Finally, as already evinced in the foregoing lines, the terminology used will follow the IUPAC recommendations summarized in the early literature,21 which are essentially coincident with the more recent Bochmann’s guidelines.22a 3. Group 10 metals. As mentioned before, in spite of the plethora of complexes of group 10 metals with carbene ligands described to date, only a few of them can be included in the restricted family of alkylidene compounds that lack of heteroatom stabilization. The first compound of this type was reported in 2002 in a landmark publication by Mindiola and Hillhouse.28 Taking advantage of the isolobality of the [NiL2] fragment with the CH2 group and of the steric protection provided by the 1,2-bis(di-tert-butylphosphino)ethane (dtbpe) ligand, the authors reacted the Ni(0) precursor Ni(dtbpe)(cod) (cod = 1,5- cyclooctadiene) with the diazocompound N2CPh2, to generate firstly the diphenyldiazomethane complex (dtbpe)Ni(N,N’:2-N2CPh2), which decomposed thermally to afford the alkylidene complex Ni(dtbpe)(CPh2) (Scheme 1). Scheme 1. The synthesis of the first Ni alkylidene (1), reported by Mindiola and Hillhouse in 2002. The Ni-C bond was described as a double bond based on spectroscopic and structural properties, which is reasonable for a metal in a low oxidation state that receives further electron-density from the strongly -donating phosphorus atoms. In the same contribution the authors described several reactions of complex 1 (vide infra). One year after this report by Hillhouse, Bröring and collaborators published the first nonheteroatom stabilized carbene complex of Pd(II), 2, also containing a CAr2 fragment, prepared by treatment of a tripyrrinate derivative of Pd(II) with p-tolyldiazomethane (p- TolCN2) in the presence of NaBArF (BArF = tetrakis[3,5- bis(trifluoromethyl)phenyl]borate).29 8 Scheme 2. The synthesis of the first Pd(II) alkylidene (2), reported by Bröring and coworkers in 2003. Finally, only five years ago we reported on the first Pt(II) alkylidene, which was characterized in solution at low temperature.30 The [Pt=CHR]+ unit was generated by - hydride abstraction from a Pt-CH2 group that forms part of a sterically protected platinacycle, obtained by cyclometallation of the bulky PiPr2Xyl phosphine (Xyl=2,6- Me2C6H3). Due to its enhanced reactivity and thermal instability, this compound could not be isolated in the solid state, but its alkylidene structure was demonstrated by reactivity studies and NMR spectroscopy. Scheme 3. Generation of the first Pt(II) alkylidene (3) to be characterized in solution. 3.1 Nickel Even in light of the intrinsic high reactivity of these species that makes difficult their isolation, it is quite surprising that after the Hillhouse’s report of 200228 no further examples of well-characterized Ni alkylidenes have appeared in the literature until 2013. It seems appropriate to discuss in this section the few reactions of complex 1 reported to date. The original contribution by Mindiola and Hillhouse disclosed the reactivity of the Ni alkylidene 1 towards a series of small molecules of different characteristics, namely CO, CO2, SO2, O=C=CPh2, and HNMe2Ph+ (Scheme 4). 9 Considering the products of the reactions of Scheme 4, it seems evident that carbene complex (dtbpe)Ni=CPh2 exhibits a clear nucleophilic character, unambiguously demonstrated, for instance, by the reaction with H+ which produces the alkyl complex 4. Scheme 4. Reactivity of the alkylidene complex 1. Further work from the same group centered on the reactions of ethylene with the carbene, imido and phosphinidene complexes (dtbpe)Ni(=X), where X represents CPh2 (i.e. complex 1), NAr, and PAr’. Transfer of the X group to C2H4 was observed, forming the respective cyclopropane, aziridine and phosphirane products.31 In a combined theoretical and experimental study,32 Hillhouse and Cundari described in 2009 some reactions of the Ni(II) alkylidene 1 with the energy-rich molecules N2O and RN3 (R = adamantyl (Ad), mesityl (Mes)). In both cases, the first step of the calculated reaction path is the 1,3-dipolar addition of the N2X (X = O, NMe) species to the nickel carbene with formation of a 5-membered ring intermediate, 5. With X = O, dinitrogen extrusion from the cyclometallated fragment affords the final 2-C,O- benzophenone adduct, which interestingly may be regarded as the product of an organometallic version of alkene epoxidation (Scheme 5, a). On the other hand, reactions with azides (X = NR) gave imido complexes as the ultimate product of series of reactions involving N2 liberation, formal nitrene-carbene coupling, coordination of a second equivalent of N3R to the (dtbpe)Ni(0) fragment and again N2 extrusion (Scheme 5, b). 16 Scheme 12. Formation of the alkylidene complex 12a and its reactions with E-Z species. DFT studies reveal the existence of three occupied molecular orbitals of  and  symmetry, two bonding and one antibonding, which leads to a Pd-C bond order of 1, as confirmed by NBO analysis. The reactivity of complexes 12a and 12b with B(C6F5)3 was also explored and some similarities between frustrated Lewis pairs and the nucleophilic palladium alkylidene recognized (Scheme 13).42 Thus, the crowded environment around the carbenic carbon atom, which would be the usual target of electrophilic attacks, led to unexpected products resulting from the activation of a C-F bond of the borane (Scheme 13, b) or the dearomatization of a phenyl ring of the ligand (Scheme 13, a). Accordingly, when R = H (12a) the accessibility of the para position (relative to the alkylidene) in combination with the enhanced electron density at the para carbon atom, resulted in the formation of the zwitterionic complex 13 with a dearomatized phenyl ring (Scheme 13, a). Scheme 13. Reactions of the alkylidene complexes 12 with B(C6F5)3. 17 Conversely, when R = tBu (12b), the carbene carbon atom is able to cleave a C—F of a C6F5 ring with formation of the ylide 14 (Scheme 13, b). It is noteworthy that the expected formation of a new C-B bond was not observed, which is to ascribe to steric reasons. The parallelism with frustrated Lewis pairs became also evident when zwitterionic complexes 13 and 14 were generated in situ in the presence of dihydrogen. Under these conditions products resulting from H2 splitting were detected. The reactivity of the nucleophilic palladium alkylidene containing coordinated PMe3 (PCcarbeneP)Pd (PMe3) toward C-H bonds of different acidity was also tested (Scheme 14).43 For instance, PhC≡CH, with pKa = 28.7, underwent C-H addition across the Pd-C bond, with formation of the alkenyl-(PCalkylP)-Pd(II) complex, [PC(H)alkylP]Pd (C≡CPh). Scheme 14. C-H activation reactions at the alkylidene complex 12a. In contrast, no observable reaction took place with diphenylmethane, which was attributed to its low Brönsted-Lowry acidity (pKa= 32.2). It was actually concluded that the observed reactivity was controlled by the pKa value, such that substrates with pKa higher than 29 do not react even after prolonged heating. Recently, these studies on the activation of C-H bonds by the ylid-type palladium alkylidene, have been extended to other E-H bonds (E= B, Si, Ge).44 In spite of the contrasted nucleophilicity of the carbene ligand in these complexes (Scheme 15), the regioselectivity of the formal addition of the E-H bond to the palladium-carbene bond is governed by the nature of the activated substrate and of the monodentate phosphine ligand (PMe3 or PPh3). 18 Scheme 15. B-H, Si-H, and Ge-H activation reactions at the alkylidene complexes 12a and 12c. Notably, reaction of the triphenylphosphine derivative 12c with Ph2SiH2 yields a mixture of the Pd-hydride [PC(SiHPh2)P]PdH and –silyl [PC(H)P]Pd(SiHPh2) isomers in a ratio that varies with temperature. Moreover, the reaction of GeH2Ph2 permitted the isolation of an unusual pentacoordinate Pd(II) complex, which can be envisaged as an intermediate preceding PMe3 dissociation (Scheme 15). Importantly, significant modifications of the (PCcarbeneP)Pd scaffold affecting the palladium-carbene bond, and therefore the electronic properties of the carbene ligand, were successfully introduced by means of chemical and electrochemical methods. As depicted in Scheme 16, in addition to the ylid-type carbene structure characteristic of the above PMe3 and PPh3 complexes, related compounds with different electronic carbene structures have been obtained. Scheme 16. Different types of palladium alkylidenes derived from the [PCcarbeneP] carbene ligand skeleton (refs. 40-47). 19 Thus, oxidation of the PMe3, ylid-like alkylidene complex 12a with Br2, I2 or with CH2X2 (X= Cl, Br, I) allowed the isolation of stable carbene radicals, 15a (Scheme 17).45 The new, four-coordinate complexes are monomeric and exhibit solution magnetic moments close to 1.8 B. EPR measurements revealed g-factors close to 2, in accordance with the radical formulation proposed for the carbene ligand skeleton. Scheme 17. Reactions of complex 12a with dihalogens and dihalomethanes yielding carbene radicals. In addition to the paramagnetic radical complexes, the dihalomethane oxidations yielded a diamagnetic, four-coordinate Pd(0) complex (16), exhibiting a (P(C=CH2)P) diphosphine-alkene tridentate ligand, resulting from the coupling of the methylene groups of the dihalomethane with the carbene carbon of the starting palladium alkylidene (Scheme 17b). 45 Another interesting observation based on a neutral, strongly nucleophilic alkylidene complex 12b was the possibility to effect successive, reversible one-electron oxidation reduction reactions.46 As shown in Scheme 18, a one-electron oxidation (0.5 equiv of I2) generated a stable carbene radical complex (15b) analogous to those already discussed (Scheme 17). Notice the tBu substitution on the aryl backbone of the diphosphine-carbene ligand in the present system, which is aimed to prevent the relatively facile radical dimerization observed for the non-substituted complex analogue.42 A second oxidation, this time promoted by Cp2Fe+, yielded the cationic, strongly electrophilic alkylidene complex 17. 20 Scheme 18. Reversible one-electron oxidations of complex 12b and reactivity of the oxidized 15b and 17 complexes. The three palladium alkylidenes of Scheme 18 feature square-planar geometry and sp2-hybridized carbene carbon, with Pd-C bond lengths which decrease regularly, from 2.076(3)Å in 12b42 to 2.022(3) and 1.968(3)Å in 15b and 17 respectively. As already mentioned, the Pd-C bond in the ylid-type complex 12b is better described as a single bond as a consequence of the antibonding -type palladium-carbene nature of the HOMO, that compensates the  bonding component residing in HOMO-10.41,42 Thus, the observed shrinkage of the Pd-C bond is consistent with an increase in the palladiumcarbene bond order. Since the LUMO of the cationic alkylidene complex 17 has a similar antibonding -type nature, successive addition of electrons should restore first the carbene radical and ultimately the ylid-type alkylidene. As can be seen in Scheme 18, these expectations were fulfilled chemically, employing KC8 as the reductant. Finally in this regard, the carbene reactivity illustrated in Scheme 18 (PhSSPh, NaH and PMe3) is in full agreement with predictions and constitutes a most remarkable example of a reversible change in the reactivity of a metal–bound carbene ligand from nucleophilic (Schemes 14 and 15) to electrophilic through radical-type reactivity (Scheme 18). Before closing this section there is yet one further contribution of the Iluc group that deserves attention.47 Metathesis of the iodide ligand of the radical alkylidene complex 15b of Scheme 19 by an amide or a benzyl group, followed by reduction with a suitable reagent, yielded unprecedented heterobimetallic palladium-potassium alkylidene complexes 18. The new compounds are diamagnetic and exhibit polymeric 21 structures in the solid state, although they have good solubility properties in ether solvents. Scheme 19. Unprecedented heterobimetallic palladium-potassium alkylidene complexes generated from the radical carbene 15b. 3.3 Platinum Our reports of 2012 and 2015 on the characterization and reactivity of a cationic platinum(II) alkylidene are to the best of our knowledge the first documented studies on a well-defined platinum species of this type.30,48 The coordination environment of the metal centre had to be accurately designed to prevent the facile decomposition of the incipient, highly reactive carbene. The bis-platinacyles shown in Scheme 20, based on the bulky phosphines PiPr2Xyl and PMeXyl2, were treated with a trityl salt in order to remove a hydride from the Pt—CH2 unit and form a cationic alkylidene complex. Not unexpectedly, the stereochemistry of the starting material was crucial for the success of this strategy, so that only the trans complex of Scheme 20 generated a detectable carbene complex, while the cis one decomposed to unidentified species under analogous reaction conditions. 22 Scheme 20. Bis-platinacycles used for the synthesis of a cationic platinum alkylidene. All attempts aimed to isolate pure samples of the Pt(II) alkylidene failed due to decomposition, although the stability of its dichloromethane solutions at low temperature permitted to carry out an extensive investigation on its reactivity towards a variety of organic and inorganic molecules (Scheme 21). The results demonstrated its carbenic nature and its expected electrophilicity. Computational studies were also developed.48 Scheme 21. Reactivity of the Pt(II) alkylidene complex 3. Reactions with phosphines and pyridines occurred with C-P and C-N bond formation leading to zwitterionic complexes with the formal positive charge on the group 15 element. In contrast, neutral C donor Lewis bases such as CO or isocyanides, induced a carbene-alkyl coupling in complex 3, probably as a consequence of nucleophilic attack of the Pt-Calkyl bond onto the electrophilic carbene carbon atom. In accordance with DFT calculations the formation of the new C-C bond occurred in a 23 pentacoordinated intermediate with trigonal bipyramidal geometry, resulting from CO coordination to platinum. A typical carbene-carbene coupling reaction was promoted when treating 3 with a diazocompound, resulting in the generation of a C=C double bond and formation of the corresponding alkene complex. DFT calculactions were also used to investigate the hydrogenation of complex 3 and revealed that H2 activation would take place at the carbenic carbon. Although our report of 2012 represented probably the first documented evidence of a well-defined Pt(II) alkylidene, Bercaw and Labinger postulated in 2003 the formation of a Pt=CH2 terminus in the catalytic alcoholysis of tetramethylsilane promoted by a metal-mediated C—H bond activation.49 Recently, Templeton and coworkers reported a detailed study on cationic Pt(IV) complexes [PtTp’(Me)2(=C(X)(Y)]+ (Tp’= hydrotris(3,5-dimethylpyrazolyl)borate) containing heteroatom stabilized carbene ligands.50 Investigation of their chemistry led to the neutral vinyl Tp’Pt(Me)2(CH=CH2) (Scheme 22) which upon protonation with HBArF gave a square-planar cationic Pt(II) complex containing an 2-CH2=CHCH3 olefinic ligand and a 2 Tp’ ligand protonated at one of the pyrazolyl arms. Evidently, the olefinic ligand resulted from a C-C bond formation reaction implicating one of the Pt- Me bonds. Low temperature NMR studies (-80ºC) evinced the formation of a shortlived ethylidene intermediate, [Tp’Pt(Me)2(=C(H)Me)]+ (19), which was characterized by NMR methods. The carbene carbon atom resonates at the striking  value of 500 ppm, seemingly the most deshielded carbene resonance observed, and features a 1JCH value of 165 Hz, consistent with sp2 1JCH couplings. The somewhat related carbene [Tp’Pt(Me)2(=C(OMe)(Me)]+ has a carbene resonance at 264 ppm, which implies a shift of over 240 ppm relative to the analogous ethylidene complex. The unusual ∆ value observed was attributed to the cationic character of the carbene carbon and to attenuated  backbonding from the Pt(IV) centre. Scheme 22. Generation and rearrangement of Templeton’s Pt(IV) alkylidene 19. 24 The first structurally characterized Pt alkylidene, 20, was described at the beginning of the current year by Iluc and Barret, in conjunction with the already mentioned lighter congener of palladium, 11 (Scheme 11).40 Scheme 23. Synthesis of the Pt(II) alkylidene 20. The ancillary ligand of choice is once again a sterically demanding diphoshine, now consisting of two PiPr2 units linked by a terphenyl diradical. The ligand is highly flexible and adapts readily to both cis and trans coordination geometries, such that it is capable to encompass P-M-P angles in the wide range of ca. 99 to 165º. The elasticity of the terphenyl-bridged diphosphine plays a crucial role in the synthetic strategy employed (Scheme 23) that involves transoid and cisoid complexes (e.g. cis-PtCl2(P^P) with P-Pt-P= 98.84(2)º; trans-Pt(P^P), P-Pt-P= 153.95(12)º and the tricoordinate Pt(=C(p-Tol)2)(P^P), P-Pt-P= 105.23(3)º). Similarly to the palladium analogue, the target alkylidene was generated from the two-coordinate Pt(P^P) platinum(0) complex and bis(p-tolyl)diazomethane. Interestingly, the analogous reaction between the diazomethane reagent and Pt(Cy3)2 afforded only the corresponding azine (p- Tol)2(C=N-N=C(p-Tol)2 and unaltered Pt(PCy3)2, suggesting the generation of catalytic, albeit spectroscopically undetectable, amounts of the alkylidene complex. X-ray crystallography and computational studies support a significant  character of the platinum-carbene bond, which is additionally endorsed by the observation by NMR spectroscopy of hindered rotation around the Pt= C(p-Tol)2 bond. The reactions of the alkylidene complex 20 with SiH2Ph2 and CH3I were also investigated. The first occurred with formal insertion of the carbene ligand into one of the Si-H bonds of the silane, followed by oxidative addition of a second equivalent of SiH2Ph2 to the resulting Pt(P^P) complex (Scheme 24, a). As for the second reaction, it is possible that methyl iodide attacks the Pt=C bond with formation of an undetected 25 iodo-alkyl Pt(II) complex that could undergo -hydrogen elimination yielding the observed products (Scheme 24, b). Scheme 24. Reactions of Pt alkylidene 20 with iodomethane and diphenylsilane. Interestingly, theoretical investigations support a significant  contribution to the Pt-C bonding in 20. On the other hand, no comments on the oxidation state of the metal are provided. 4. Group 11 metals Growing interest in copper catalyzed cyclopropanation reactions gave in the early 2000s a strong impetus to academic investigations on group 11 metal alkylidenes.51 To the best of our knowledge, the first report on a spectroscopically detected copper(I) nonheteroatom stabilized alkylidene dates back to 2001, only one year before Hillhouse’s first publication on the related nickel complex. In this contribution, Straub and Hofmann used an “extremely basic, sterically demanding, and highly symmetric iminophosphanamide ligand” to stabilize a Cu=C(Ph)(CO2Me) fragment, taking advantage of the enhanced metal-to-ligand  backdonation and the steric protection provided by the NPN ligand (Scheme 25).52 32 with the mononuclear terminal alkylidene 29a reveals a stronger backbonding contribution in the bimetallic species as a consequence of the three-centre, four-electron interaction existing in the CCu2 core. As described in preliminary form in 2008, Peters and Mankad used a sterically demanding bis(phosphino)borate ligand to generate a terminal Cu(I) alkylidene, 30 (Scheme 33).65 Its limited stability only permitted spectroscopic characterization (NMR, UV-vis) in solution, as decomposition with liberation of tertraphenyl ethylene was observed even at 30 ºC. Complex 30 is produced from the amino derivative 31 with excess diphenyldiazomethane, with concomitant formation of the unusual azine adduct 32. Scheme 33. Generation of the copper alkylidene 30 characterized by Peters. The carbenic nature of complex 30 was further demonstrated by its reaction with carbon monoxide, which yielded selectively the (PBP)Cu(CO) carbonyl complex and the ketene Ph2=C=CO, that is the product of formal carbene transfer from the metal to CO. The already mentioned communication by Straub and coworkers on the first isolated silver alkylidene of the Ag=CR2 type also reported the characterization in solution of its copper analogue, 33 (Scheme 34).54 Even under optimized reaction conditions only concentrated solutions of the copper carbene could be obtained, i.e. clean and selective conversion of the substrates was not achieved. Nevertheless, satisfactory NMR, UV-vis, and HR-ESI characterization was provided for complex 33. The original UV-vis and NMR spectroscopic studies performed by Straub on the three isostructural complexes [(IPr**)M(CPh2)] (M = Cu, Ag, Au), in conjunction with theoretical calculations, offer novel insights into the nature of the bonding in this class of cationic coinage metal alkylidenes. This discussion is postponed to the chapter dedicated to gold, as a final summary. 33 Scheme 34. Synthesis of the copper alkylidene 33. The last example of this family of complexes appeared very recently in a contribution by Wang and collaborators aimed to isolate possible intermediates generated in the course of catalytic amination of alkynes.66 Thus, copper and gold carbenes were obtained in a novel template synthesis in which the metal promoted the cyclization of a sterically demanding N-alkynyl formamidine and its transformation into an alkylidene ligand (Scheme 35). Nevertheless, similarly to its gold congener, the Cu(I) complex 34a presents a high degree of delocalization owing to the mesoionic imidazolium substituent, which implies that the vinyl mesomeric form shown in Scheme 35 provides a major contribution. Indeed, the double bond character of the C-O linkage is demonstrated by X-ray structural data. Scheme 35. Synthesis of the imidazolium-stabilized copper and gold alkylidenes 34. 4.2 Silver In spite of the great number of catalytic organic transformations in which they are supposed to be implied, silver carbene complexes without heteroatom substituents remained a mere hypothesis for decades, as their intrinsic high reactivity made impossible to provide any direct experimental evidence for their formation in the condensed phase. Only very recently,54 Straub and collaborators isolated the first 34 representative of this class (vide supra). Moreover, they provided satisfactory solution and solid-state characterization despite the impossibility to find a clean synthetic procedure owing to the extreme reactivity and thermal instability of the compound. Figure 3. ORTEP plot of the X-ray molecular structure of the silver alkylidene complex (22) reported by Straub et al. in 2015. The structural and spectroscopic data gathered for this compound deserve some additional comments, aimed to clarify the nature of the Ag-CMes2 bond. The Ag- CMes2 separation is ca. 6 pm longer than the other metal-carbon bond distance (i.e. Ag- C(IPr**)), which reflects the very limited or insignificant backbonding contribution. On the other hand, the negligible resonance stabilization of the divalent carbon in complex 22 is demonstrated by the fact that the mesityl substituents of the alkylidene are twisted out of the NHC-Ag-C plane (see Figure 3). Moreover, the 13C NMR signals due to the Ag-CMes2 carbon appear at very high frequency (332 ppm), consistent with highly reduced metal-to-carbon backdonation. Particularly informative is the comparison between the 13C resonances due to the two different silver-bonded carbon atoms, being the alkylidene one ca. 50 ppm down-field shifted with respect to the other. In addition, Ag-C coupling constants for silver bonded carbons reflect, accordingly to Bent’s rule, the bonding situation in complex 22 and in the precursor. To the best of our knowledge, this is the only example of non-heteroatom stabilized silver alkylidene isolated, or even characterized, to date. 4.3 Gold 35 The interest in alkylidene complexes of gold(I) has experienced a rapid and impressive growth in the last years, driven by the great importance of a variety of Au catalyzed organic transformations. Extensive data on well-defined species in the condensed phase were anticipated by Chen and co-workers, 67 who developed a method for the gas-phase generation of gold carbenes via collision-induced dissociation of ylide precursors (Scheme 36).68 Scheme 36. Gas-phase generation of gold alkylidenes. Theoretical work by Toste, Goddard and co-workers69 offers a useful tool for the estimation of the different contributions to the bonding in gold species of the general formula LAuI(CR2) depending on the carbene ligand substituents and the ligand L, with special reference to cationic phosphine complexes (Figure 4). Interestingly, the general conclusion is that the overall bond order of gold carbenes is “generally less than or equal to one”. Figure 4. Bonding in [(PR3)Au(CR2)]+, as described by Goddard and Toste. In the schematic representations of the  bonds, the orbitals of  symmetry that mainly influence the extent of the interaction are also depicted. As mentioned above, Fürstner and Seidel succeeded in the solid-state isolation of the first non-heteroarom substituted gold alkylidene (Scheme 27).55 Some comments will be dedicated to the synthesis, structural and spectroscopic properties, and reactivity of this compound (Scheme 37). 36 Scheme 37. Synthesis of the bis(p-methoxyphenyl)carbene complex of gold(I), 23, reported by Fürstner and Seidel. Alkylidene 23 was generated by transmetallation, along with the bimetallic species 35 in ca. 3:2 ratio. It is moderately stable in the crystalline form even at room temperature and can convert a substituted styrene into the corresponding cyclopropane via formal carbene transfer. To correlate the metal-carbon bond order with structural measurements, the authors compared the Au-C separation in 23 with those found in complexes of the type [Au=C(OMe)(Ar)]+. Besides, the Ccarbene-Carene distance in 23 was contrasted with those observed in a number of species containing the C(sp2)Ar2 unit, such as carbene complexes of ruthenium,, benzophenone, and diarylcarbocations. As a consequence of marginal backdonation and strong resonance stabilization promoted by the electron-donating methoxy group, the bond was described as a single Au-C bond. The barrier of rotation around the Ccarbene-Carene axis measured at low temperature of only of ca. 35 kJ/mol is also in agreement with this perspective. A dirhodium alkylidene source was latterly employed to prepare two further gold complexes of this class.70 Somehow related to Fürstner’s contribution and reminiscent of Jones’ early work on Pt(II),71 a relevant study on the synthesis and properties of a gold carbenoid was published by Widenhoefer and Harris in 2014,56 who prepared a stable cycloheptatrienylidene complex via hydride abstraction from the parent cycloheptatrienyl derivative. In the absence of heteroatoms linked to the carbene C atom, it is evident that the involvement of the empty 2pC orbital in the aromatic system of the tropylium cation plays a crucial role in the stabilization of this species. Bourissou and collaborators proposed a novel approach to the creation of isolable gold cationic alkylidenes,57 reasoning that the ancillary ligand L can be 37 specifically designed to enhance the  backdonation ability of the [LAu]+ moiety, with consequent improvement in the stability of the carbene. This was achieved by using the sterically demanding o-bis(aminophoshino)carborane, DPC, which is able to bend the two-coordinate (DPC)Au+ fragment and raise the energy of the filled 5d orbitals of the metal, therefore promoting the transfer of electron-density to the empty 2p orbital of the carbenic carbon atom. Interestingly, this approach also permitted the isolation of a rare example of cationic Au(I) carbonyl complex (36), the first of the classical type with a (CO) of 2113 cm1 (Scheme 38). Scheme 38. Synthesis of the stable alkylidene (37) and carbonyl (36) complexes of gold(I) described by Bourissou et al. in 2014. Some features of the alkylidene complex 37 deserve a brief discussion. The metal centre lies in a highly distorted trigonal-planar environment with P-Au-P angle of ca. 90º, whereas the CPh2 carbon is perfectly planar. The maximization of the  backbonding is reflected not only in the short Au-C bond distance of 1.984(4) Å, but also in the orthogonality of the two planes defined by the PAuP and CPh2 fragments. Moreover, in accordance with the lack of stabilization of the divalent carbon atom by resonance with the phenyl rings, the two aryl groups are twisted from the carbene plane by ca. 60º. Finally, DFT calculations offer a picture of the bonding interactions in 37, assigning to the HOMO a significant character of  backbonding involving the 5dxy orbital of gold and 2p of carbon. The LUMO coincides basically with the corresponding anti-bonding interaction. A nice comparison with Straub’s and Fürstner’s systems was also provided. As already cited, Straub and co-workers were able to generate and characterize the series of isostructural complexes of group 11 of the general formula [(IPr**)M(CMes2)]NTf2 (33, M = Cu; 22, M = Ag; 38, M = Au), which allows for a 38 stimulating comparison of the bonding in the three compounds. Structural data are available only for the two heavier congeners, 22 and 38.54,58 Table 2. Selected experimental data for group 11 metal alkylidenes 33, 22, and 38. (IPr**)Cu(CMes2) 33 (IPr**)Ag(CMes2) 22 (IPr**)Au(CMes2) 38 M—CMes2 - 205(2) 201.4(6) pm M—C(IPr**) - 199(3) 203.0(6) pm  (HOMO→LUMO) 730 nm 742 nm 642 nm M-CMes  (13C NMR) 332 ppm 359.3 ppm 321.3 ppm Color greenish yellow mint-green emerald-green In both species the negligible resonance stabilization furnished by the mesityl rings is demonstrated by the non-coplanarity the M(IPr**) fragment and the CMes2 one. Even though the M-C distances in the two species are almost identical within experimental error, the authors recognized a higher extent of metal-to-alkylidene  backbonding in the gold complex than in the silver one based on the differences in the M-CMes2 and M-C(IPr**) separations. In particular, they reasoned that, in agreement with Bent’s rule, in the absence of M-C  backdonation, the M-C(IPr**) distance should be shorter than the M-CMes2. This fact is actually observed for silver but not for gold. 13C NMR values for the resonances of the alkylidene carbon nuclei in the three species are correlated to the extent of carbenoid character, concluding that it increases in the order Au << Cu < Ag. 1 Comments already made on the copper(I) carbene published by Wang in 2017, apply essentially to its isostructural gold congener (34b, Scheme 34),66 which should therefore be considered as an atypical heteroatom-stabilized alkylidene. Table 3. M—CRR’ bond distances and 13C NMR chemical shifts for the carbenic carbon nucleus in the alkylidene complexes discussed in this review. Metal Compound M—CRR’ (Å) 13C NMR  for M-CRR’ (ppm) Ref. Ni 1 1.836(2) 222 28 6a 1.908(5) 182 33 1 We note that Straub used the term “carbenoid” in contrast to “carbenium” and “carbene”, see reference 58. 39 6b 1.927(4) - 33 7a 1.793(3) 206 35 7b 1.797(3) 216 35 Pd 2 1.978(7) 313 29 10 1.999(4) 190 39 11 - 296 40 12a 2.086(4) 2.076(1) - 41 12b 2.076(3) 136 42 15a 2.020(3) 2.005(2) - 45 15b 2.022(3) - 46 17 1.968(3) 284 46 15c 2.019(2) - 47 15d 2.024(2) - 47 18a 2.095(2) - 47 18b 2.043(2) - 47 Pt  - 209 30 19 - 500 50 20 1.942(3) 263 40 Cu 21 - 230 52 24 1.832(2) 240 60a 26a 1.822(4)a ~230b 60b 29a 1.834(3) 253 63 30 - 332 65 33 - 332 54 Ag 22 2.05(2) 359 54 Au 23 2.039(5) 285 55 37 1.984(4) 316 57 38 2.014(6) 321 58 a structural data available only for the closely related complex (NPN)Cu=C[p- OMeC6H4][CO2{CH(p-ClC6H5)2}]; b coupling constants available for several complexes of the same type. 5. Summary and Perspectives 40 As we attempted to demonstrate in this review, complexes of groups 10 and 11 metals with non-heteroatom stabilized carbene ligands constitute a young family of organometallic compounds characterized by varied bonding situations, intriguing reactivity, and stimulating spectroscopic and structural features. In light of the scarcity of satisfactorily characterized species existing in the literature, it appears that we are still far from a deep understanding of this class of compounds. For instance, besides the little number of stable non-heteroatom stabilized carbenes of coinage metals, their reactivity is almost unexplored with the exception of a few studies on the function they exert in catalytic cyclopropanation. In spite of the high reactivity of these compounds, recent findings demonstrate that with a suitable choice of ancillary ligands, species with fair thermal stability and therefore quite easy-to-handle can be prepared.57 As a final remark, we strongly encourage organometallic and coordination chemists to concentrate further efforts in this research area in view of the intrinsic structural value of these complexes and not least in light of their potential interest in catalysis. 6. Acknowledgements Financial support from the Ministerio de Economía, Industria y Competitividad of the Spanish Government (CTQ2016-75193-P, AEI/FEDER, UE) is gratefully acknowledged. 7. References 1. (a) Arduengo, J. A.; Bertrand, G. Chem. Rev. 2009, 109, 3209-3884 (special issue dedicated to carbenes). (b) Crabtree, R. H. J. Organomet. Chem. 2005, 690, 5451; (c) Díez-González, S., Nolan, S.P. Coord. Chem. Rev. 2007, 251, 874. 2. (a) Lavallo, V.; Canac, Y.; Prasang, C.; Donnadieu, B.; Bertrand, G. Angew. Chem. Int. Ed. 2005, 44, 5705; (b) Jazzar, R.; Dewhurst, R. D.; Bourg, J. B.; Donnadieu, B.;Canac, Y.; Bertrand, G. Angew. Chem. Int. Ed. 2007, 46, 2899; (c) Zeng, X.; Frey, D. G.; Kinjo, R.;Donnadieu, B.; Bertrand, G. J. Am. Chem. Soc. 2009, 131, 8690–8696; (d) Chu, J., Munz, D., Jazzar, R., Melaimi, M., Bertrand, G. J. Am. Chem. Soc. 2016, 138, 7884; (e) Roy, S.; Mondal, K. C.;Roesky, H. W. Acc. Chem. Res. 2016, 49, 357. 3. For some leading work, see the following citations and references therein: a) A. Fürstner, P. W. Davies, Angew. Chem. Int. Ed. 2007, 46, 3410; b) A. Fürstner, 41 Chem. Soc. Rev. 2009, 38, 3208; c) E. Jiménez- Núñez, A. M. Echavarren, Chem. Rev. 2008, 108, 3326; d) A. M. Echavarren, C. Nevado, Chem Soc. Rev. 2004, 33, 431; e) N. D. Shapiro, F. D. Toste, Synlett 2010, 675; f) D. J. Gorin, F. D. Toste, Nature 2007, 446, 395; g) E. Soriano, J. Marco-Contelles, Acc. Chem Res. 2009, 42, 1026; h) B. Trillo, F. López, S. Montserrat, G. Ujaque, L. Castedo, A. Lledós, J. L. Mascareñas, Chem. Eur. J. 2009, 15, 3336. 4. a) A. Fürstner, H. Szillat, B. Gabor, R. Mynott, J. Am. Chem. Soc. 1998, 120, 8305; b) A. Fürstner, F. Stelzer, H. Szillat, J. Am. Chem. Soc. 1998, 120, 8305; c) A. Fürstner, H. Szillat, F. Stelzer, J. Am. Chem. Soc. 2000, 122, 6785. 5. a) B. Martín-Matute, C. Nevado, D. J. Cárdenas, A. M. Echavarren, J. Am. Chem. Soc. 1998, 120, 8305; b) A. M. Echavarren, E. Jiménez-Núñez, Top. Catal. 2010, 53, 924; c) V. López-Carrillo, N. Huguet, A. Mosquera, A. M. Echavarren, Chem. Eur. J. 2011, 17, 10972. 6. K. H. Dötz, Metal Carbenes in Organic Synthesis, Springer, 2004. 7. a) Y. Chauvin, Angew. Chem. Int. Ed. 2006, 45, 3740; b) R. R. Schrock, Angew. Chem. Int. Ed. 2006, 45, 3748; c) R. H. Grubbs, Angew. Chem. Int. Ed. 2006, 45, 3760. 8. a) E. Poverenov, D. Milstein, Chem. Commun. 2007, 3189; b) R. J. Burford, W. E. Piers, M. Parvez, Organometallics 2012, 31, 2949; c) L. E. Doyle, W. E. Piers, J. Borau-Garcia, M. J. Sgro, D. M. Spasyuka, Chem. Sci. 2016, 7, 921; d) J. Yuan, R. P.Hughes, A. L. Rheingold, Dalton Trans. 2015, 44, 19518. 9. R. A. Moss, M. P. Doyle, Contemporary Carbene Chemistry; John Wiley & Sons, 2014. 10. F. Z. Drwald (Ed.), Metal Carbenes in Organic Synthesis; Wiley-VCH, Weinheim, Germany, 2008. 11. J. W. Herndon, Coord. Chem. Rev. 2016, 329, 53. 12. a) E. Carmona, M. Paneque, M. L. Poveda, Dalton Trans. 2003, 4022; b) S. Conejero, M. Paneque, M. L. Poveda, L. L. Santos, E. Carmona, Acc. Chem. Res. 2010, 43, 572. 13. a) J. Campos, R. Peloso, M. Brookhart, E. Carmona, Organometallics 2013, 32, 3423; b) J. Campos, E. Carmona, Organometallics 2015, 34, 2212; c) J. Campos, M. F. Espada, J. López-Serrano, E. Carmona, Inorg. Chem. 2013, 52, 6694; d) J. Campos, J. López-Serrano, E. Álvarez, E. Carmona, J. Am. Chem. Soc. 2012, 134, 7165.