Sonia Ladeira

Gold–Silane and Gold–Stannane Complexes: Saturated Molecules as σ‐Acceptor Ligands

The discovery that saturated molecules may form s complexes by side-on coordination of a s bond to a transition metal represents a major breakthrough in transition-metal chemistry. Over the years, considerable progress has been made in the understanding of this bonding situation. The coordination and activation of s bonds involving Group 14 elements (E = C, Si, Ge, Sn, Pb) is at the forefront of developments in this area. An increasing variety of complexes A and B 7] (Scheme 1) featuring side-on coordinated s(E H) and s(E E) bonds have been isolated, and the key factors governing the delicate balance between dissociation and oxidative addition have been progressively identified. The common feature and believed prerequisite for the coordination of saturated molecules free of lone pairs to transition metals is the superposition of ligand!metal donation (from a filled s orbital of the ligand to an empty d orbital of the metal) and metal!ligand back-donation (from a filled d orbital of the metal to an empty s* orbital of the ligand). Aiming at identifying new types of metal–ligand interactions, and stimulated by our work on Group 13 Lewis acids as acceptor ligands, we recently became interested in complexes of type C, in which a saturated Group 14 element could behave as an end-on, s-acceptor ligand toward a transition metal. Heavier Group 14 elements such as silicon and tin are known to readily form hypervalent compounds through donor!acceptor interactions with organic Lewis bases. A related situation is envisioned in complexes C, with a transition metal acting as a Lewis base. Such donor! acceptor interactions between transition metals and silanes or stannanes have been invoked in a few highly strained complexes on the basis of relatively short M E distances. Furthermore, the presence of a Pd!Sn dative bond was recently evidenced structurally and theoretically in a palladastannatrane cage complex supported by four methimazolyl groups. Herein, we report the straightforward synthesis and complete characterization of three gold complexes supported by diphosphino silane and stannane ligands. The presence of metal!silane and metal!stannane interactions in these complexes has been substantiated spectroscopically, structurally, and theoretically, thus providing unambiguous evidence for the existence of complexes of type C. From our previous studies on Group 13 Lewis acids, the use of two phosphine buttresses ligated by ortho-phenylene spacers was considered as a good compromise in order to support, but not impose, the coordination of the Group 14 element. We thus targeted the two complexes [o{(iPr2P)C6H4}2E(Ph)FAuCl] 2 (E = Si) and 4 (E = Sn). The Scheme 1. Complexes A–C featuring Group 14 saturated molecules coordinated to transition metals (E= C, Si, Ge, Sn, Pb).

Facile oxidative addition of aryl iodides to gold(I) by ligand design: bending turns on reactivity.

Thanks to rational ligand design, the first gold(I) complexes to undergo oxidative addition of aryl iodides were discovered. The reaction proceeds under mild conditions and is general. The ensuing aryl gold(III) complexes have been characterized by spectroscopic and crystallographic means. DFT calculations indicate that the bending induced by the diphosphine ligand plays a key role in this process.

Copper(I) complexes derived from mono- and diphosphino-boranes: Cu-->B interactions supported by arene coordination.

The monophosphino-boranes o-iPr(2)P(C(6)H(4))BR(2) (1: R = Ph and 3: R = Cy) and diphosphino-boranes [o-R(2)P(C(6)H(4))](2)BPh (5: R = Ph and 6: R = iPr) readily react with CuCl to afford the corresponding complexes {[o-iPr(2)P(C(6)H(4))BPh(2)]Cu(mu-Cl)}(2) 2, {[o-iPr(2)P(C(6)H(4))BCy(2)]Cu(mu-Cl)}(2) 4, {[o-Ph(2)P(C(6)H(4))](2)BPh}CuCl 7, and {[o-iPr(2)P(C(6)H(4))](2)BPh}CuCl 8. The presence of Cu-->B interactions supported by arene coordination within complexes 2, 7, and 8 has been unambiguously evidenced by NMR spectroscopy and X-ray diffraction studies. The unique eta(2)-BC coordination mode adopted by complexes 7 and 8 has been thoroughly analyzed by density-functional theory (DFT) calculations.

Spontaneous Oxidative Addition of σ‐SiSi Bonds at Gold

The past decade has witnessed tremendous development in homogeneous gold catalysis. Thanks to unique soft and tunable Lewis acidity, gold(I) and gold(III) complexes have been shown to efficiently activate most types of p-bonded substrates. Besides these synthetic achievements, mechanistic issues have attracted considerable attention, and much progress has been accomplished recently thanks to the characterization of several key intermediates. With the aim of further extending the scope of gold catalysis, major interest has been devoted over the last few years to gold-mediated processes involving two-electron redox cycles. The isolobal analogy between Au(I/III) and Pd(0/II) complexes is stimulating the search for “gold versions” of palladium-catalyzed transformations, and in particular cross-couplings. Accordingly, a variety of intraand intermolecular coupling reactions have been shown to be efficiently catalyzed by gold complexes in the presence of strong exogeneous oxidants. A few studies have also suggested the ability of gold complexes to promote crosscouplings in a similar way to palladium catalysts, but possible palladium contamination has recently cast doubt on the involvement of gold complexes in these transformations. These recent contributions as a whole open a new facet of gold catalysis, and at the same time, raise fundamental questions about the properties and behavior of gold complexes. In particular, better knowledge on the basic elementary steps involved in redox cycles (oxidative addition, transmetallation, migratory insertion, b elimination, reductive elimination) is highly desirable to help further developments into this emerging area of gold catalysis. In this regard, it is striking to note that, despite the stability and availability of both Au and Au complexes, oxidative addition reactions at gold remain very elusive. In fact, the reluctance of gold to undergo oxidative addition is presumed to be an intrinsic limitation to achieve Au/Au redox cycles akin to those typically operating with Pd. Rare evidence for oxidative addition at gold dates back to the pioneering contribution of Kochi on the reaction of highly polarized sp C X bonds with alkyl gold(I) complexes. Furthermore, Bachman et al. reported in 2008 the first and to date only example of activation of an apolar s bond with gold. Fluorinated disulfides were shown to oxidatively add and reductively eliminate at gold in a reversible fashion. Seeking to gain more insight into s-bond activation and oxidative addition at gold, we became interested in extrapolating the strategy we have been developing for some years to study unusual metal–ligand interactions. Our approach consists in the use of anchoring phosphine sites, and this has been applied to support the coordination of Lewis acids and most recently s-Si Si bonds. Accordingly, the coordination of the diphosphine–disilane ligand 1 to copper and silver has been explored (Scheme 1), leading to the first structural characterization of a s complex with a coinage metal. As a further extension of this work, we were intrigued by the coordination of 1 to gold, considering three possible scenarios: 1) Formation of a s-complex upon side-on coordination of the disilane moiety, as observed with copper; 2) no participation of the s-Si Si bond to the coordination, as observed with silver; or 3) oxidative addition of the s-Si Si bond to give a bis(silyl)gold(III) complex. Oxidative addition

Hypervalent Silicon Compounds by Coordination of Diphosphine–Silanes to Gold

Coordination of ambiphilic diphosphine-silane ligands [o-(iPr(2)P)C(6)H(4)](2)Si(R)F (R=F, Ph, Me) to AuCl affords pentacoordinate neutral silicon compounds in which the metal atom acts as a Lewis base. X-ray diffraction analyses, NMR spectroscopy, and DFT calculations substantiate the presence of Au→Si interactions in these complexes, which result in trigonal-bipyramidal geometries around silicon. The presence of a single electron-withdrawing fluorine atom is sufficient to observe coordination of the silane as a σ-acceptor ligand, provided it is positioned trans to gold. The nature of the second substituent at silicon (R=F, Ph, Me) has very little influence on the magnitude of the Au→Si interaction, in marked contrast to N→Si adducts. According to variable-temperature and 2D EXSY NMR experiments, the apical/equatorial positions around silicon exchange in the slow regime of the NMR timescale. The two forms, with the fluorine atom in trans or cis position to gold, were characterized spectroscopically and the activation barrier for their interconversion was estimated. The bonding and relative stability of the two isomeric structures were assessed by DFT calculations.

Stable N‐Heterocyclic Carbene Complexes of Hypermetallyl Germanium(II) and Tin(II) Compounds

The chemistry of the heavier Group 14 element carbene analogues has received wide interest because of their special properties and reactivity. Recently, an unexpected application of cyclopentadienyl, amido, or alkoxy germylenes and cyclic diazastannylene as precursors of nanomaterials has been described, thus opening a wide and promising field for investigation. Germanium nanowires have also been obtained by decomposition of hexakis(trimethylsilyl)digermane, highlighting the labile character of trimethylsilyl groups. Thus, the hypermetallyl germylenes or stannylenes, which contain both a low-coordinate Group 14 atom and a good leaving substituent, might be suitable candidates for nanomaterial alloys preparations. However, to the best of our knowledge, germylenes or stannylenes having electropositive silyl or germyl substituents have only been postulated as transient species; for example, in the reaction of Cl2Ge·dioxane with (Me3Si)3ELi (E = Si, Ge), chloro(silyl)germylenes rapidly oligomerize with the formation of cyclotetragermanes [(Me3Si)3EGeCl]4 or rearrange leading to cyclotrimetallanes [(Me3Si)2E{Ge(SiMe3)2}E(SiMe3)2]. To date, only one bis(hypersilyl)stannylene has been reported, but in its dimeric form in equilibrium with the monomeric form in solution and as a dimer in the solid state. Recently, a thermally unstable bis(hypergermyl)stannylene was synthesized (requiring preparation and handling below 30 8C), but like its hypersilyl substituted analogue, the X-ray structural analysis revealed the presence of dimers in the solid state. These results show the limitations of the steric shielding of the hypersilyl or hypergermyl ligands for the stabilization of low-coordinate species. Among the stabilization strategies of germylenes or stannylenes, the intermolecular coordination has aroused a great interest in the last decades, particularly with the use of N-heterocyclic carbenes (NHC) as stabilizing co-ligand. The first examples of carbene–germanium(II) adducts were described by Arduengo et al. ((NHC)GeI2) [8] and then by Lappert et al. (NHC–heterocyclic germylene). Later, NHCs were successfully employed for the stabilization of transient germanium(II) species. In contrast, there are few examples of carbene–stannylene adducts. In all of these cases, the carbene coordination is one of the key factors to obtain divalent species in their monomeric state. Herein we describe the synthesis of hypermetallyl germylenes and stannylenes that are stabilized by complexation with carbene units. We investigated the reactivity of the carbene–germylene adduct 1 and of the carbene–stannylene adduct 3, which were obtained as previously reported from the known carbene [DC{N(iPr)C(Me)}2] [13] and Cl2Ge·dioxane or Cl2Sn, respectively, towards various sources of hypermetallyl units. All attempts to displace the chloride from germylenes with hypersilyl salts (Me3Si)3SiLi [14] or (Me3Si)3SiK [15] were unsuccessful and led to complex mixtures in which only some amounts of disilagermirane [(Me3Si)2Si{Ge(SiMe3)2}Si(SiMe3)2] could be identified. [4a] The latter has been previously obtained by mixing (Me3Si)3SiLi and Cl2Ge·dioxane. By contrast, treatment of 1 with one half equivalent of [(Me3Si)3Si]2Mg [16] in THF solution at room temperature gave the complex 2 a as the only product in a good yield (71%) (Scheme 1). Compound 2a is the first example of a donor-stabilized hypersilyl(chloro)germylene that could be isolated by the combination of both steric hindrance of the hypersilyl ligand and strong Lewis base coordination. Similarly, the addition of a stoichiometric amount of digermylmagnesium [(Me3Si)3Ge]2Mg [17] to 1 produces the hypergermyl-

A Crystalline σ Complex of Copper

Over the last decades, our understanding of σ-bond activation at transition metals has progressed considerably from both fundamental and synthetic points of view thanks to the preparation and characterization of a variety of σ complexes. Here we report the synthesis and structural analysis of the first σ complex involving a coinage metal. The copper(I) complex 2 derived from the diphosphine-disilane [Ph(2)P(C(6)H(4))Me(2)Si-SiMe(2)(C(6)H(4))PPh(2)] (1) has been isolated and crystallographically characterized. The coordination of the Si-Si σ bond to copper was thoroughly analyzed by quantum-chemical methods.

Coordination of phosphinoboranes R2PB(C6F5)2 to platinum: an alkene-type behavior.

The paucity of boron-containing heteroalkene complexes prompted us to explore the coordination of phosphinoboranes. The complexes {[R(2)PB(C(6)F(5))(2)]Pt(PPh(3))(2)} (R = Cy, t-Bu) were obtained by ethylene displacement. Spectroscopic and crystallographic data indicated symmetric side-on coordination of the phosphinoborane to Pt. Thorough analysis of the bonding situation by computational means revealed important similarities but also significant differences between the phosphinoborane and ethylene complexes.

Multicomponent reactions in ionic liquids: convenient and ecocompatible access to the 2,6-DABCO core

Herein we describe the use of ionic liquids as complementary new media for multicomponent reactions leading to the 2,6-diazabicyclo[2.2.2]octane core. The reaction took place efficiently in ionic liquid solvents instead of toluene, giving overall better results (faster rate, higher concentration) and allowing generalization to hitherto unreactive substituted Michael acceptors. In these cases, the use of molecular sieves, acting as a heterogeneous catalyst in the preliminary study in molecular solvents, could be totally excluded thanks to the promoting properties of the ionic liquid solvent, which could be recycled and reused up to five times. Complete diastereoselectivity was observed at one of the bridge positions, as assigned by X-ray analysis.

The reactivity of phosphagermaallene Tip(t-Bu)Ge=C=PMes* with doubly and triply bonded nitrogen compounds.

Phosphagermaallene Tip(t-Bu)Ge=C=PMes* (1; Mes* = 2,4,6-tri-tert-butylphenyl, Tip = 2,4,6-triisopropylphenyl) gives, with N-benzylidenemethylamine and pivalonitrile, [2+2] cycloadditions between the Ge=C double bond and the C=N and C≡N unsaturations, leading to the formation of the corresponding four-membered heterocycles 2 and 9. With N-tert-butyl-α-phenylnitrone and benzonitrile oxide, [2+3] cycloadditions occur to form the five-membered ring derivatives 6 and 7. By treatment of 1 with derivatives which possess weak acidic hydrogens in α of the C=N or C≡N multiple bond, two types of reactions were observed: an ene reaction with methyl(benzylideneamino)acetate and a 1,2 addition with acetonitrile to afford azadienyl(germyl)ether (4) and 3-germa-1-phosphapropene (8), respectively. In the case of benzonitrile, phosphagermaallene 1 behaves as a 1,3-dipole, to give, via a cyclic phosphagermacarbene intermediate, the tricyclic derivative 10.