A New Area in Main-Group Chemistry: Zerovalent Monoatomic Silicon

Jul 19, 2017 - His current research interests include coordination chemistry of main-group elements and transition metals in unusual coordination and ...
0 downloads 20 Views 4MB Size
Article pubs.acs.org/accounts

A New Area in Main-Group Chemistry: Zerovalent Monoatomic Silicon Compounds and Their Analogues Shenglai Yao, Yun Xiong, and Matthias Driess* Technische Universität Berlin, Department of Chemistry, Metalorganics and Inorganic Materials, Sekr. C2, Strasse des 17. Juni 135, 10623 Berlin, Germany S Supporting Information *

CONSPECTUS: Monoatomic zerovalent main-group element complexes emerged very recently and attracted increasing attention of both theoretical and experimental chemists. In particular, zerovalent silicon complexes and their congeners (metallylones) stabilized by neutral Lewis donors are of significant importance not only because of their intriguing electronic structure but also because they can serve as useful building blocks for novel chemical species. Featuring four valence electrons as two lone pairs at the central atoms, such complexes may form donor−acceptor adducts with Lewis acids. More interestingly, with the central atoms in the oxidation state of zero, they could pave a way to new classes of compounds and functional groups that are otherwise difficult to realize. In this Account, we mainly describe our contributions in the chemistry of monatomic zerovalent silicon (silylone) and germanium (germylone) supported by a chelate bis-Nheterocyclic carbene (bis-NHC) ligand in the context of related species developed by other groups in the meantime. Utilizing the bis-NHC stabilized chlorosilyliumylidene [:SiCl]+ and chlorogermyliumylidene [:GeCl]+ as suitable starting materials, we successfully isolated silylone (bis-NHC)Si and germylone (bis-NHC)Ge, respectively. The electronic structures of the latter complexes established by theoretical calculations and spectroscopic data revealed that they are genuine metallylone species with electron-rich silicon(0) and germanium(0) centers. Accordingly, they can react with 1 molar equiv of GaCl3 to form Lewis adducts (bis-NHC)E(GaCl3) (E = Si, Ge) and with 2 molar equiv of ZnCl2 to furnish (bis-NHC)Si(ZnCl2)2. Conversion of the metallylones with elemental chalcogens affords isolable monomeric silicon(II) and germanium(II) monochalcogenides (bis-NHC)EX(GaCl3) (X = Se, Te), representing molecular heavier congeners of CO. Moreover, their reaction with elemental chalcogens can also yield monomeric silicon(IV) and germanium(IV) dichalcogenides (bis-NHC)EX2 (X = S, Se, Te) as the first isolable complexes of the molecular congeners of CO2. Moreover, (bis-NHC)Si could even activate CO2 to afford the monomolecular silicon dicarbonate complex (bisNHC)Si(CO3)2 via the formation of SiO and SiO2 complexes as intermediates. Furthermore, starting with a chelate bis-Nheterocyclic silylene supported [:GeCl]+, we developed two bis-N-heterocyclic silylene stabilized germylone→Fe(CO)4 complexes. Our achievements in the chemistry of metallylones demonstrate that the characteristic of monatomic zerovalent silicon and its analogues can provide novel reaction patterns for access to unprecedented species and even extends the series of functional groups of these elements. With this, we can envision that more interesting zerovalent complexes of the main-group elements with unprecedented reactivity will follow in the near future.

1. INTRODUCTION Molecular complexes of zerovalent elements are of fundamental importance due to their interesting electronic structures and their roles as indispensable building blocks in synthetic chemistry. Although the chemistry of zerovalent transitionmetal complexes featuring neutral ligands such as carbonyl, phosphanes, and N-heterocyclic carbenes (NHCs) has been well developed and enjoys numerous applications in synthesis and catalysis throughout academia and industries,1 that of zerovalent main-group elements is much less explored owing to the difficulty of stabilizing such highly reactive species. Taking advantage of thermodynamic and kinetic stabilization by using strongly σ-donating and sterically shielding DNHC (DNHC = :C[N(Dipp)CH]2, Dipp = 2,6-iPr2C6H3), the striking example of zerovalent dinuclear main-group-element complex A (DNHC→SiSi←DNHC, Scheme 1),2a emerged in 2008 © 2017 American Chemical Society

from Robinson and co-workers. Since then, other related types of compounds could be realized by several research groups, leading to a number of isolable dinuclear group 13, 14, and 15 element complexes in the form of L→E02←L (L = donor ligand, E = main-group element).2,3 The chemistry of monoatomic zerovalent main-groupelement complexes began with the discussion concerning the electronic structure of carbodiphosphorane C(PPh3)2 B, known since 1961,4 which is best described in terms of donor− acceptor interaction between a “bare” carbon atom and two phosphines as σ-donors Ph3P→C←PPh3.5 The term “carbone” was suggested by Frenking for this type of compound with a carbon atom in its zero oxidation state (L→C0←L), which Received: June 7, 2017 Published: July 19, 2017 2026

DOI: 10.1021/acs.accounts.7b00285 Acc. Chem. Res. 2017, 50, 2026−2037

Article

Accounts of Chemical Research Scheme 1. Known Examples of Zerovalent Dinuclear Main-Group Complex A and Monoatomic Complexes B−K

both a Ge0 species and a mesoionic germylene.18 Moreover, the monoatomic zerovalent tin compounds I19 and J20 supported by a diiminopyridine and a butadiene, respectively, emerged recently. Remarkably, apart from these zerovalent p-block element complexes, the fascinating s-block complex K of beryllium(0) with strong multiple bonding has been reported by Braunschweig et al. using the cAAC ligands.21 In this Account, we wish to present our contribution in this research area by summarizing the synthesis and striking reactivity of a chelate bis-NHC ligand supported monatomic Si0 and Ge0 complexes with an electron-rich central atom. In addition, two bis-N-heterocyclic silylene (NHSi) supported germynium(0) complexes will be described.

show a different chemical reactivity from carbenes with a carbon(II) center.6 Moreover, the Lewis-donor supported monoatomic Si(0) species L→Si0←L, termed as “silylones”, and its heavier homologues (germylones, stannylones) have been predicted by theoretical calculations to have a bent molecular geometry with the central group 14 atom bearing two lone pairs of electrons.7 Accordingly, the previously reported trisilaallene and trigermaallene C8−10 with bent E− E−E angles should rather be considered as silylone and germylone with a central Si0 or Ge0 atom, which is stabilized by two silylene or germylene moieties. A similar description as a stannylone may also be applied to the “tristannaallene” D reported by Wiberg et al. in 1999, which was described as an adduct of R2SnSn: and stannylene R2Sn:.11 In contrast, the bipyridine silicon complexes [Si(bpy)3] (bpy = 2,2′-bipyridine) and [Si(bpy)2] were documented as Si0 complexes in the original reports but recently turned out to be silicon(IV) species with noninnocent bpy-ligand based on the results of Density Functional Theory (DFT) investigations.12 Following the inspiring results from theoretical calculations by Frenking and co-workers,7,13 several groups of experimentalists started to synthesize other types of monoatomic maingroup element complexes and examined their fascinating chemical behavior. In 2008, Bertrand and co-workers synthesized an extremely bent acyclic allene E, which can act as a strong donor ligand to bind RhCl(CO)2, indicating its carbone character.14 In 2013, utilizing two cyclic alkyl amino carbenes (cAAC), Roesky and co-workers reported the siladicarbene (silylone) and the germadicarbene (germylone) F.15,16 Shortly after that, a Ge0 complex G supported by a diiminopyridine was described by Nikonov et al.17 In addition, Kinjo and co-workers described the imino-NHC ligand stabilized germanium complex H, which may be viewed as

2. SYNTHETIC ROUTES TO BIS-NHC STABILIZED ZEROVALENT SILICON AND GERMANIUM SPECIES For the access to monoatomic zerovalent silicon and germanium species, the corresponding precursors are crucial. The cAAC supported acylic silylone (cAAC)2Si, reported by Roesky et al. in 2013, resulted from reduction of the Si(IV) biradical species (cAAC)2SiCl2.15 It can also be obtained from reduction of the Si(I) radical (cAAC)2SiI.22 Our investigations on realizing zerovalent Si and Ge complexes began with the successful isolation of the first stable chlorosilyliumylidene [:SiCl]+ complex 1a supported by a bis(iminophosphorane) in 2012 (Scheme 2).23 Starting from the silicon(II) dichloride complex DNHC→SiCl2,24 the DNHC ligand exchange with 1,8bis(tributylphosphazenyl)naphthalene afforded 1a in 46% isolated yields (Figure 1, left). According to DFT calculations, however, the lowest unoccupied molecular orbital (LUMO) of 1a is mainly localized on the naphthalene moiety of the supporting ligand, indicating that the reduction of this 2027

DOI: 10.1021/acs.accounts.7b00285 Acc. Chem. Res. 2017, 50, 2026−2037

Article

Accounts of Chemical Research Scheme 2. Synthesis of Bis(iminophosphorane) Supported Chlorosilyliumylidene (1a), Chlorogermyliumylidene (1b) and Their Unsuccessful Reduction to the Corresponding E(0) Complexes (E = Si, Ge)

Scheme 3. Synthesis of Chlorosilyliumylidene and Chlorogermyliumylidene Complexes 2a and 2b and Their Dehalogenation To Give Silylone 3a and Germylone 3b, Respectively

compound may not take place at the silicon atom to furnish the desired zerovalent silicon species but at the ligand backbone. In line with this, treatment of 1a with sodium naphthalenide only gave an unidentified mixture with no indication for the formation of the desired silylone. Similarly, we obtained the chlorogermyliumylidene [:GeCl]+ complex 1b in 60% isolated yields from the reaction of 1,8-bis(tributylphosphazenyl)naphthalene with GeCl2-dioxane (Figure 1, right).25 In analogy to its silicon homologue 1a, the LUMO of 1b is also localized on the supporting ligand. Thus, this compound is also unsuitable for the desired reduction to give a germylone (Scheme 2). Utilizing a far more redox innocent chelating bis-NHC ligand, the exchange reaction with DNHC→SiCl224 in THF yielded the desired bis-NHC stabilized chlorosilyliumylidene [:SiCl]+ complex 2a in 57% isolated yields (Scheme 3).26 Complex 2a exhibits a drastic upfield shift in the 29Si NMR spectrum at δ = −58.4 ppm in CD3CN compared to that of 1a (δ = −3.30 ppm in CD2Cl2), indicating a very strong electron donation effect of the bis-NHC chelate ligand toward the Si(II) atom. The molecular structure of 2a established by X-ray single crystal diffraction analysis (XRD) reveals a bis-NHC coordinated Si(II) atom in a trigonal-pyramidal environment (Figure 2, left). The two Si−C distances (average 1.961(4) Å) fall in the normal range for dative C→Si bonds. In contrast to that of 1a, both HOMO and LUMO of 2a are mainly localized on the Si center (Figure 3). The latter confirms that 2a is indeed a divalent silicon compound with a lone pair of electrons located

at the silicon atom, which is in contrast to the situation in the biradical silicon(IV) species (cAAC)2SiCl2 reported by Roesky and co-workers.27 This drastic difference can be understood taking into account that NHCs have two nitrogen atoms adjacent to the carbene-carbon atom while cAACs have only one. As a result, the singlet−triplet gap of cAACs is significantly smaller than that of NHCs and the σ C−Si electron-sharing bonds, other than C→Si dative bonds, are energetically favorable in (cAAC)2SiCl2.28 In analogy to the approach used for chlorosilyliumylidene 2a, the reaction of the same bis-NHC ligand with GeCl2-dioxane furnished the bis-NHC supported chlorogermyliumylidene 2b in 95% isolated yields (Scheme 3).29 The latter possesses a molecular structure similar to that of 2a (Figure 2, right). Moreover, the electronic structure of 2b resembles that of 2a with both HOMO and LUMO mainly located on the Ge(II) atom. As their LUMOs are mainly located at the central atoms, both 2a and 2b could serve as long-sought precursors for the corresponding zerovalent complexes under dechlorination conditions. As expected, the reduction of 2a with sodium naphthalenide at −60 °C led to the formation of the desired silylone 3a as a dark red powder in 68% isolated yields (Scheme 3).26 The molecular structure of 3a established by XRD (Figure 4, left) bears significantly shorter Si−C bond distances (average 1.869 Å) than those observed in 2a (average 1.962 Å). These

Figure 1. Molecular structures of the cations of 1a and 1b. 2028

DOI: 10.1021/acs.accounts.7b00285 Acc. Chem. Res. 2017, 50, 2026−2037

Article

Accounts of Chemical Research

Figure 2. Molecular structures of the cations of 2a and 2b.

Figure 3. HOMO and LUMO of 2a.

Figure 5. HOMO and HOMO − 1 of 3a.

Si−C bonds in 3a are slightly longer than the corresponding values observed for the acyclic silylone (cAAC)2Si (1.841(2) Å),15 in which the cAAC ligands are stronger π electron acceptors than NHCs. Consequently, the Si atom in 3a is more electron-rich than that in (cAAC)2Si as indicated also by the upfield shift of the 29Si NMR chemical shift (δ = −80.1 ppm for 3a vs δ = 66.7 ppm for (cAAC)2Si). In line with that, 3a shows larger first and second proton affinities (PA(1) = 283.4, PA(2) = 168.3 kcal mol−1) in comparison to those of (cAAC)2Si (268.8, 155.3 kcal mol−1) as suggested by DFT calculations based on molecular models.26 The relatively large PAs also imply that two lone pairs of electrons are present at the silicon center of silylone 3a. Accordingly, the HOMO of 3a (Figure 5, left) represents the silicon π-orbital with slight Si−C π bonding, whereas the HOMO − 1 shows the silicon σ-lone pair orbital (Figure 5, right). This is in line with theoretical predictions provided by Frenking and co-workers.30 Moreover, silylone 3a shows a deep red color in toluene solutions, and the UV−vis spectrum exhibits an absorption maximum at λ 547 nm (ε = 7.5 × 103), which can be assigned to the electronic HOMO → LUMO transition. Akin to the silicon analogue, dechlorination of [:GeCl]+ complex 2b with sodium naphthalenide afforded germylone 3b as a dark red powder in 45% isolated yields (Scheme 3).29 Its molecular structure established by XRD is practically identical with that of the silicon analogue 3a (Figure 4, right). The Ge− C distances of 1.967(2) and 1.962(2) Å in 3b are significantly longer than those in the acyclic (cAAC)2Ge complexes

reported by Roesky and co-workers (1.9386(16)−1.954(2) Å),16 indicating again the strong electronic differences of NHCs vs cAACs as supporting ligands. The HOMO of 3b consists of a π-type orbital located at the Ge center with appreciable Ge−C π bonding character (Figure 6, left), while the HOMO − 1

Figure 6. HOMO and HOMO − 1 of 3b.

represents a σ lone pair orbital at the Ge center (Figure 6, right). The proton affinities of compound 3b (PA(1) = 279.6, PA(2) = 175.0 kcal mol−1) agree with the notion of two lone pairs at the Ge(0) center as expected for a genuine germylone.

3. REACTIVITY TOWARD LEWIS ACIDS (METAL HALIDES) Featuring two lone pairs at the silicon and germanium central atoms, silylone 3a and 3b are expected to form stable donor−

Figure 4. Molecular structures of 3a and 3b. 2029

DOI: 10.1021/acs.accounts.7b00285 Acc. Chem. Res. 2017, 50, 2026−2037

Article

Accounts of Chemical Research acceptor adducts with appropriate Lewis acids. This has been proven by reacting 3a and 3b with GaCl3 to give complex 4a and 4b, respectively (Scheme 4). The silylone→GaCl3 complex Scheme 4. Formation of GaCl3 Adducts 4a and 4b and Germinal bis-ZnCl2 Complex 5

Figure 8. Molecular structure of 5.

silicon atom clearly distinguishes a silylone compound from a silylene and silaallene species. Silylone 3a and germylone 3b not only feature two lone pairs at the central atoms for coordination, but to our surprise, they can also reduce germanium(II) and silicon(II) dichloride complexes to give elemental germanium and silicon, respectively (Scheme 5). During the reactions with DNHC→

4a was isolated as a yellow solid in 58% yields,31 while the germanium analogue 4b resulted as a white solid in 65% isolated yields.32 In their molecular structures, the threecoordinate silicon and germanium atoms adopt a pseudotetrahedral coordination geometry with a lone pair of electrons occupying the vertex (Figure 7). This differs significantly from the situation of H (Scheme 1), which forms complexes with M(CO)5 (M = Cr, Mo, W), but the germanium centers therein feature a trigonal planar geometry without any indication for a lone pair.33 Complexes 4a and 4b represent the first isolable Lewis adducts of silylone and germylone with a lone pair left at the central atom. The central silicon(0) atom of 3a can also serve as a 2-fold donor toward Lewis acids. For instance, 3a reacts with 2 molar equiv of ZnCl2 in THF at room temperature to afford the germinal dimetalation complex 5 (Scheme 4).34 The latter has been obtained as colorless crystals and characterized by XRD (Figure 8). Its molecular structure bears a tetrahedral coordinate silicon center attached to the chelate bis-NHC and two ZnCl2 molecules. The coordination environment of the two Zn atoms is different: one is trigonal planar with a Si··· Zn distance of 2.3735(8) Å, while the other adopts a tetrahedral geometry due to the coordination by a THF molecule (Si···Zn = 2.4396(9) Å). Compound 5 represents the first example of a silylone−metal complex with two Lewis acceptors. This germinal dimetalation is the most characteristic attribute of a silicon(0) complex. Such reactivity of a neutral

Scheme 5. Reducing Ability of 3a and 3b

SiCl2,24 the dinuclear complex A2 is also formed along with elemental silicon. In both cases, 3a and 3b converted back to 2a and 2b as corresponding divalent species. This demonstrates experimentally the unique electron-rich nature and enormous potential of the silylone 3a and germylone 3b to act as molecular reducing agents.

Figure 7. Molecular structures 4a and 4b. 2030

DOI: 10.1021/acs.accounts.7b00285 Acc. Chem. Res. 2017, 50, 2026−2037

Article

Accounts of Chemical Research

4. REACTIVITY OF SILYLONE TOWARD ELEMENTAL CHALCOGENS Since genuine silylones bear a silicon atom in the oxidation state of zero, they may undergo facile oxidation reaction at the silicon atom with elemental chalcogens to afford the corresponding silicon(II) mono- and silicon(IV) dichalcogenides. In fact, the oxidation of 3a with elemental sulfur led to the novel bis-NHC supported SiS2 complex 6 in 89% isolated yields as a colorless powder (Scheme 6).31 Its solid-state CP/

at room temperature for 3 h to afford 9 in 65% isolated yields (Scheme 6). On the other hand, the reaction of 4a with red selenium in THF furnished 8a in 56% isolated yields. The 29 Si{1H}-NMR spectrum of 9 displays a signal at δ = −56.6 ppm, while that of 8a resonates at δ = −69.9 ppm. Compound 8a is isostructural with the sulfur homologue 7a, featuring a strongly bent SiSe2 moiety and significantly different Si−Se bonds (Si−Se1 = 2.129(2) and Si−Se2 = 2.241(6) Å, Figure 10, left). In the molecular structure of 9 (Figure 10, right), the “silicon end” of :SiSe is coordinated by both bis-NHC and GaCl3, leading to the tetrahedral coordination geometry of the silicon atom. The silicon−selenium distances of 2.135(1) and 2.1439(9) Å in the two independent molecules of 9 suggest appreciable Si−Se double bond character. Thus, compound 9 could be formally regarded as a monomolecular complex of silicon(II) monoselenide. It is worth mentioning that complexes containing heavier analogues of CO are rare. Known examples comprise the complexes of SnO,36 PbO,36 and PbSe37 stabilized by donor−acceptor interactions at both ends of the heavy CO analogues. Compound 9 represents the first example of a heavy homologue of CO complexes without any support at the chalcogen atom. In contrast to the reaction of 4a with elemental sulfur and selenium, the reaction of 4a with elemental tellurium did not furnish any isolable product.35 However, the reaction with 2 molar equiv of TeP(nBu)3 as a more reactive tellurium source led to the isolation of (bis-NHC)SiTe2 10 as orange crystals in 63% yields (Scheme 6). Its 29Si{1H}-NMR spectrum displays an upfield-shifted resonance at δ = −143.9 ppm, while the 125Te{1H}-NMR spectrum shows a singlet resonance signal at δ = −1120 ppm. XRD study revealed a distorted tetrahedral coordination for the silicon center, which is ligated to the bisNHC ligand and two terminal tellurium atoms (Figure 11). The two Si−Te distances are slightly different (2.389(4) and 2.436(2) Å), and the Te−Si−Te angle amounts to 128.4°. Based on natural resonance theory (NRT), the predominant resonance structures of the SiSe and SiX2 (X = S, Se, Te) complexes with evaluated contributions are depicted in Scheme 7.35 In general, semipolar Si−X bonds and delocalization of positive charge into the bis-NHC ligand framework are expected for these species, which leads to a considerable number of corresponding resonance structures. In the case of 8a with a terminal :SiSe moiety, moderate π-bonding contribution to the relatively polar Si−Se bond is realized by no-bond/double-bond resonance structures. Similarly, certain π-bonding contribution can also be expected for the terminal SiX (X = S, Se, Te) moieties in the SiX2 complexes but decreasing going down in the group. The coordination of GaCl3 to one of the two X atoms quenches much of the Si−X π-bonding character but enhances the other one without GaCl3 coordination. It should be mentioned, all the reported structures of pristine SiX2 (X = O, S, Se, Te) are polymeric by nature. Complexes 6, 7a, 8a, and 10 represent the first examples of isolable monomeric SiX2 complexes. Recently, molecular Si2X4 (X = O, S, Se) complexes, representing dimers of SiX2 supported by NHCs or cAACs, have been reported by the Robinson and Roesky groups, starting from zerovalent disilicon complexes.38−40

Scheme 6. Reactivity of Silylone 3a and Its GaCl3 Adduct 4a towards Elemental Chalcogens

MAS 29Si{1H} NMR spectrum shows a resonance at δ = −32.5 ppm. Compound 6 can further react with GaCl3 to form Lewis adduct 7a, which has also been isolated in 91% yields as colorless crystals (Scheme 6). Alternatively, the latter 7a can be obtained by treating 4a with elemental sulfur. Its molecular structure features a strongly bent silicon disulfide moiety with GaCl3 coordinated to one of the sulfide atoms (Figure 9). The two Si−S distances (Si−S1 = 2.006(2) and Si−S2 = 2.106(2) Å) are thus different. The reaction of 4a with red selenium (Se8), however, resulted in the formation of two products. Apart from the silicon(IV) diselenide complex 8a, the silicon(II) monoselenide complex 9 could also be obtained.35 The latter reaction was solvent dependent. In acetonitrile, 4a reacted with red selenium

Figure 9. Molecular structure of 7a. 2031

DOI: 10.1021/acs.accounts.7b00285 Acc. Chem. Res. 2017, 50, 2026−2037

Article

Accounts of Chemical Research

Figure 10. Molecular structures of 8a and 9.

(Scheme 8).32 Although attempts to synthesize the germanium(II) monosulfide complex from 4b failed, the reaction of 4b with red selenium in acetonitrile afforded the desired germanium(II) monoselenide complex 12a in 64% yields as colorless crystals. Moreover, the tellurium analogue 12b could be obtained in 71% yields as a pale yellow solid. Furthermore, the germanium(II) compound 12a could react further with red selenium in THF to yield the germanium(IV) species 13 as the final product. Unexpectedly, both 12a and 13 reacted with elemental sulfur to give the GeS2 complex 11 with liberation of elemental selenium and tellurium, respectively. The molecular structures of the germanium(IV) dichalcogenides 11 and 13 are isostructural with those of 7a and 8a mentioned above (Figure 12). In contrast, the germanium(II) complexes 12a and 12b possess a molecular structure completely different from that of the SiSe complex 9. The former features a lone pair at the germanium center with GaCl3 coordinated to the Se and Te atoms (Figure 13), while the latter prefers having the GaCl3 coordinated to the silicon site. This corresponds to the significantly stronger donor ability of

Figure 11. Molecular structure of 10.

5. REACTIVITY OF GERMYLONE TOWARD ELEMENTAL CHALCOGENS By analogy with the silicon analogue, 4b reacts smoothly with elemental sulfur to yield the bis-NHC and GaCl3 supported GeS2 complex 11 in 90% isolated yields as a colorless solid

Scheme 7. Predominant Electronic Structure Situation of 6−10 and No-Bond/Double-Bond Resonance Structures with Evaluated Contributions

2032

DOI: 10.1021/acs.accounts.7b00285 Acc. Chem. Res. 2017, 50, 2026−2037

Article

Accounts of Chemical Research

exergonically by only −88 kJ mol−1, with 14 even being formed endergonically (12 kJ mol−1). The extremely polar Si−O bond in 15 is reflected by the free-energy gain of −81 kJ mol−1 upon reaction with two additional molecules of CO2 to afford 16. It is worth mentioning that the calculated values for the reaction of 3a with O2 are −236 and −583 kJ mol−1 for the mono- and dioxygenation, respectively, accounting for the uncontrollable reaction with dioxygen as mentioned above. The silicon dicarbonate complex 16 crystallized as a monomer with a six-coordinate silicon center attached by a bis-NHC and two carbonato chelate ligands (Figure 14). It represents the first isolable molecular silicon dicarbonate complex. It should be mentioned that germylone 3b also reacted with CO2 but led merely to the formation of Lewis adduct (bis-NHC)(CO2)2 along with a precipitate of GeO2.

Scheme 8. Reactivity of Germylone GaCl3 Adduct 4b towards Elemental Chalcogen

7. BIS-NHSi SUPPORTED ZEROVALENT GERMANIUM COMPLEXES N-Heterocyclic silylenes (NHSis) are divalent silicon species featuring a lone pair at the silicon(II) center and considered to be more electron-rich ligands compared with NHCs.42 Recently, by utilizing the amidinate-based silylene [PhC(NtBu)2]SiN(SiMe3)2, So and co-workers synthesized a digermanium(0) complex in the form of NHSi→:GeGe:← NHSi.43 However, NHSis supported monoatomic complexes in the form of NHSi→E0←NHSi are currently unknown. Employing the pincer-type, tridentate bis(silylene)pyridine 17 developed by our group originally for metal-mediated catalytic transformations, we successfully synthesized the germanium(II) complex 18, which was isolated as a yellow powder in 62% yields (Scheme 10).44 The latter crystallized as a separated ion-pair with the chlorogermyliumylidene cation coordinated by the bis(silylene)pyridine ligand through the two silylene-donor sites (Si···Ge = 2.3503(6) and 2.3513(6) Å, Figure 15). Although the synthesis of bis(silylene)pyridene-germanium(0) through reductive dechlorination of 18 failed, the reaction with Collman’s reagent, K2Fe(CO)4, afforded 19 as an iron complex of the desired bis(NHSi)pyridene-germanium(0) species in 49% isolated yields (Scheme 10).44 The latter crystallized as a monomer with the germanium(0) center being coordinated by the two NHSi-silicon atoms and the iron(0) site, adopting a trigonal-pyramidal geometry similar to that of

divalent silicon vs germanium if exposed to Lewis-acid centers due to the lower electronegativity of silicon. It is interesting to note that theoretical calculations demonstrate that the presence of GaCl3 is essential for the stabilization of all these monomeric EX (E = Si, Ge, X = Se, Te) species bearing highly polar EX bonds. In these complexes, the donor−acceptor interaction with GaCl3 is very strong. The same holds also for the EX2 complexes with GaCl3. In fact, attempts to remove the coordinated GaCl3 using strong external Lewis bases such as NHC failed.

6. ACTIVATION OF CARBON DIOXIDE Due to the presence of electron-rich silicon and germanium centers, both 3a and 3b are extremely sensitive toward dioxygen, and the reaction is uncontrollable, leading to the formation of SiO2 and GeO2 under liberation of the “free” bisNHC ligand. However, the dioxygenation of 3a can be controlled through activation of CO2 at low temperature, affording the silicon(IV) dicarbonate complex (bis-NHC)Si(CO3)2 16 in 75% yields as a colorless solid (Scheme 9).41 Quantum chemical calculations proposed (bis-NHC)SiO, 14, and (bis-NHC)SiO2, 15, as important intermediates for the latter reaction and showed that the formation of 15 proceeds

Figure 12. Molecular structures of 11 and 13. 2033

DOI: 10.1021/acs.accounts.7b00285 Acc. Chem. Res. 2017, 50, 2026−2037

Article

Accounts of Chemical Research

Figure 13. Molecular structures of 12a and 12b.

Scheme 9. Formation of 16 via 14 and 15 from 3a with Calculated Reaction Free Energies ΔG (in kJ mol−1)

Figure 15. Molecular structure of the cation of 18.

Figure 14. Molecular structure of the silicon(IV) dicarbonate complex 16.

4b (Figure 16). The coordination between the Fe(CO)4 moiety and the germanium center (Ge−Fe distance = 2.499 Å) is

Scheme 10. Formation of the Bis-NHSi Supported Chlorogermyliumylidene 18, Germylone→Iron Complex 19 and Its GeCl2 Adduct 20, Utilizing the Bis(NHSi) Pyridine Ligand 17

Figure 16. Molecular structure of 19.

dictated by a very strong donor−acceptor interaction. Insertion of GeCl2 into the latter Ge→Fe donor−acceptor bond occurred when 19 was treated with GeCl2-dioxane, leading to the formation of compound 20 in 65% isolated yields (Scheme 10). The Ge−Ge distance of 2.4784(7) Å in the molecular structure of 20 lies in the range of a Ge−Ge single bond 2034

DOI: 10.1021/acs.accounts.7b00285 Acc. Chem. Res. 2017, 50, 2026−2037

Article

Accounts of Chemical Research

may give access to more unusual molecular building blocks even suitable for materials synthesis.

(Figure 17), while the Ge−Fe distance of 2.359(1) Å is shorter than the Ge−Fe bond in 19. This is in agreement with the



ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.accounts.7b00285. Crystallographic data for compound 5 (CIF)



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Figure 17. Molecular structure of 20.

Matthias Driess: 0000-0002-9873-4103 Notes

push−pull Ge(0)→Ge(II)→Fe(CO)4 electronic structure of 20 predicted by NBO.

The authors declare no competing financial interest. Biographies

8. CONCLUSION AND OUTLOOK In this Account, we have summarized the synthesis and striking reactivity of reactivity of monatomic zerovalent silicon and zerovalent silicon and germanium complexes, so-called silylones and germylones (“metallylones”). As predicted by theory, these isolable metallylones stabilized by a chelate bis-NHC ligand feature two lone pairs of electrons and two dative electron pairs from the supporting bis-NHC donors. Accordingly, they can readily react with Lewis acids to form respective isolable 1:1 and even 1:2 Lewis adducts with the lone pairs of electrons located at the central atoms. The single silicon(0) and germanium(0) atoms in the latter can serve as intriguing building blocks for the stabilization of elusive monomeric silicon and germanium compounds and functional groups. This has been demonstrated by the reaction with elemental chalcogens, which allowed access to a series of complexes containing monomeric heavy CO and CO2 homologues including the complexes of SiSe, SiS2, and SiTe2. The reaction of the silylone with CO2 afforded the first isolable molecular silicon(IV) dicarbonate complex via the corresponding silicon(II) monoxide and silicon(IV) dioxide complexes as intermediates under cleavage of a CO bond in the CO2 molecule and liberation of CO. The successful reactivity studies with our silylone and germylone may be owing to both the strong donor ability and the chelate effect of the bis-NHC ligand, which makes the central atoms electron-rich and tightly holds the silicon or germanium center during the reaction. Moreover, using a chelate N-heterocyclic silylene pyridine ligand, a zerovalent germanium species has been isolated as Fe(CO)4 complexes, which represent the first example of donor−acceptor stabilized germylone complexes supported by NHSi instead of NHC donor ligands. The chemistry of monoatomic zerovalent main-group elements is still in its infancy. The current progress in synthesizing metallylones and the investigation of the wealth reactivity is very encouraging to extend the series of realizable monomeric zerovalent main-group elements in the presence of suitable supporting ligands. More striking examples of such complexes with intriguing reactivity may be expected in the near future. Our research in this area is currently directed toward the development of monoatomic E(0) species of the group 12, 13, and 15 metals and their oxidation reaction, which

Shenglai Yao obtained his Ph.D. degree in 2005 at the Johannes Gutenberg University of Mainz under supervision of Professor Karl W. Klinkhammer. Afterwards he joined the Driess group at the Technische Universität at Berlin and continues to work in the group as a senior researcher. He has a strong interest in coordination chemistry of low-valent main-group elements and transition metals. Yun Xiong obtained her Ph.D. degree in 2004 at the Johannes Gutenberg University of Mainz under supervision of Professor Karl W. Klinkhammer. In 2006, she joined the Driess group at the Technische Universität at Berlin and continues to work in the group as a senior researcher. Her interest is low-valent and low-coordination-number main-group-element chemistry. Matthias Driess completed his Ph.D. degree in 1988 at the University of Heidelberg under the supervision of Professor Walter Siebert. He then worked from 1988 to 1989 as a postdoctoral fellow at the University of Wisconsin at Madison with Professor Robert West. He returned to the University of Heidelberg and finished his Habilitation in 1993. In 1996, he accepted a position as a full-professor of Inorganic Chemistry at the Ruhr-University of Bochum (Germany) before moving to a full professorship at the Institute of Chemistry (Metalorganic Chemistry and Inorganic Materials) of the Technische Universität Berlin (2004). Since 2007, he serves as the chairman of the Cluster of Excellence “Unifying Concepts in Catalysis” (UniCat) in the Berlin-Potsdam area. His current research interests include coordination chemistry of main-group elements and transition metals in unusual coordination and oxidation states and synthesis of functional inorganic materials, for example, heterobimetal oxide nanoparticles, employing molecular architecture for catalysis.

■ ■

ACKNOWLEDGMENTS We are grateful to the Deutsche Forschungsgemeinschaft (CRC 1109 and DR 226 17-2) for financial support. REFERENCES

(1) Hartwig, J. Organotransition Metal Chemistry: From Bonding to Catalysis; University Science Books, 2000. (2) (a) Wang, Y.; Xie, Y.; Wei, P.; King, R. B.; Schaefer, H. F., III; Schleyer, P. v. R.; Robinson, G. H. A stable silicon(0) compound with a SiSi double bond. Science 2008, 321, 1069−1071. (b) Sidiropoulos, A.; Jones, C.; Stasch, A.; Klein, S.; Frenking, G. N-Heterocyclic Carbene Stabilized Digermanium(0). Angew. Chem., Int. Ed. 2009, 48, 2035

DOI: 10.1021/acs.accounts.7b00285 Acc. Chem. Res. 2017, 50, 2026−2037

Article

Accounts of Chemical Research 9701−9704. (c) Jones, C.; Sidiropoulos, A.; Holzmann, N.; Frenking, G.; Stasch, A. An N-heterocyclic carbene adduct of diatomic tin,:Sn Sn. Chem. Commun. 2012, 48, 9855−9857. (3) For reviews see: (a) Wilson, D. J. D.; Dutton, J. L. Recent advances in the field of main-group mono- and diatomic ‘allotropes’ stabilized by neutral ligands. Chem. Chem. - Eur. J. 2013, 19, 13626− 13637. (b) Zhao, L.; Hermann, M.; Holzmann, N.; Frenking, G. Dative bonding in main group compounds. Coord. Chem. Rev. 2017, 344, 163−204 and references therein. (4) Ramirez, F.; Desai, N. B.; Hansen, B.; McKelvie, N. Hexaphenylcarbodiphosphorane, (C6H5)3PCP(C6H5)3. J. Am. Chem. Soc. 1961, 83, 3539−3540. (5) Tonner, R.; Ö xler, F.; Neumüller, B.; Petz, W.; Frenking, G. Carbodiphosphoranes: The Chemistry of Divalent Carbon(0). Angew. Chem., Int. Ed. 2006, 45, 8038−8042. (6) Frenking, G.; Tonner, R. Divalent Carbon(0) Compounds. Pure Appl. Chem. 2009, 81, 597−614. (7) Takagi, N.; Shimizu, T.; Frenking, G. Divalent E(0) Compounds (E = Si-Sn). Chem. - Eur. J. 2009, 15, 8593−8604. (8) Ishida, S.; Iwamoto, T.; Kabuto, C.; Kira, M. A Stable Siliconbased Allene Analogue with a Formally sp-Hybridized Silicon Atom. Nature 2003, 421, 725−727. (9) Iwamoto, T.; Masuda, H.; Kabuto, C.; Kira, M. Trigermaallene and 1,3-Digermasilaallene. Organometallics 2005, 24, 197−199. (10) Kira, M.; Iwamoto, T.; Ishida, S.; Masuda, H.; Abe, T.; Kabuto, C. Unusual Bonding in Trisilaallene and Related Heavy Allenes. J. Am. Chem. Soc. 2009, 131, 17135−17144. (11) Wiberg, N.; Lerner, H. − W.; Vasisht, S. − K.; Wagner, S.; Karaghiosoff, K.; Nöth, H.; Ponikwar, W. Tetrasupersily-tristannaallene and − tristannacyclopropene (tBu3Si)4Sn3-Isomers with the Shortest SnSn Double Bonds to Date. Eur. J. Inorg. Chem. 1999, 1999, 1211−1218. (12) England, J.; Wieghardt, K. 2,2′-Bipyridine Compounds of Group 14 Elements: A Density Functional Theory Study. Inorg. Chem. 2013, 52, 10067−10079. (13) Tonner, R.; Frenking, G. C(NHC)2: Divalent Carbon(0) Compounds with N-Heterocyclic Carbene Ligands-Theoretical Evidence for a Class of Molecules with Promising Chemical Properties. Angew. Chem., Int. Ed. 2007, 46, 8695−8698. (14) Dyker, C.; Lavallo, V.; Donnadieu, B.; Bertrand, G. Synthesis of an Extremely Bent Acyclic Allene (“Carbodicarbene“): A Strong Donor Ligand. Angew. Chem., Int. Ed. 2008, 47, 3206−3209. (15) Mondal, K. C.; Roesky, H. W.; Schwarzer, M. C.; Frenking, G.; Niepötter, B.; Wolf, H.; Herbst-Irmer, R.; Stalke, D. A Stable Singlet Biradicaloid Siladicarbene: (L:)2Si. Angew. Chem., Int. Ed. 2013, 52, 2963−2967. (16) Li, Y.; Mondal, K. C.; Roesky, H.; Zhu, P.; Stollberg, P.; HerbstIrmer, R.; Stalke, D.; Andrada, D. M. Acyclic Germylones: Congeners of Allenes with a Central Germanium Atom. J. Am. Chem. Soc. 2013, 135, 12422−12428. (17) Chu, T.; Belding, L.; van der Est, A.; Dudding, T.; Korobkov, I.; Nikonov, G. I. A Coordination Compound of Ge0 Stabilized by a Diiminopyridine Ligand. Angew. Chem., Int. Ed. 2014, 53, 2711−2715. (18) Su, B.; Ganguly, R.; Li, Y.; Kinjo, R. Isolation of an Imino-Nheterocyclic Carbene/Germanium(0) Adduct: A Mesoionic Germylene Equivalent. Angew. Chem., Int. Ed. 2014, 53, 13106−13109. (19) Flock, J.; Suljanovic, A.; Torvisco, A.; Schoefberger, W.; Gerke, B.; Pöttgen, R.; Fischer, R. C.; Flock, M. The Role of 2,6Diaminopyridine Ligands in the Isolation of an Unprecedented, Low-Valent Tin Complex. Chem. - Eur. J. 2013, 19, 15504−15517. (20) Kuwabara, T.; Nakada, M.; Hamada, J.; Guo, J. D.; Nagase, S.; Saito, M. (η4-Butadiene)Sn(0) Complexes: A New Approach for ZeroValent p-Block Elements Utilizing a Butadiene as a 4π-Electron Donor. J. Am. Chem. Soc. 2016, 138, 11378−11382. (21) Arrowsmith, M.; Braunschweig, H.; Celik, M. A.; Dellermann, T.; Dewhurst, R. D.; Ewing, W. C.; Hammond, K.; Kramer, T.; Krummenacher, I.; Mies, J.; Radacki, K.; Schuster, J. K. Neutral ZeroValent s-Block Complexes with Strong Multiple Bonding. Nat. Chem. 2016, 8, 890−894.

(22) Li, Y.; Chan, Y.-C.; Leong, B.-X.; Li, Y.; Richards, E.; Purushothaman, I.; De, S.; Parameswaran, P.; So, C.-W. Trapping a Silicon(I) Radical with Carbenes: A Cationic cAAC-Silicon(I) Radical and an NHC-Parent-Silyliumylidene Cation. Angew. Chem., Int. Ed. 2017, 56, 7573−7578. (23) Xiong, Y.; Yao, S.; Inoue, S.; Irran, E.; Driess, M. The Elusive Silyliumylidene [ClSi:]+ and Silathionium [ClSiS]+ Cations Stabilized by Bis(Iminophosphorane) Chelate Ligand. Angew. Chem., Int. Ed. 2012, 51, 10074−10077. (24) Ghadwal, R. S.; Roesky, H. W.; Merkel, S.; Henn, J.; Stalke, D. Lewis Base Stabilized Dichlorosilylene. Angew. Chem., Int. Ed. 2009, 48, 5683−5686. (25) Xiong, Y.; Yao, S.; Driess, M.; et al. Taming the Germyliumylidene [CIGe:]+ and Germathionium [CIGeS]+ Catio n s b y Dono r−Acceptor Stabilization Using 1,8-Bis(tributylphosphazenyl)naphthalene. Chem. Commun. 2012, 48, 12198−12200. (26) Xiong, Y.; Yao, S.; Inoue, S.; Epping, J. D.; Driess, M. A Cyclic Silylone (‘Siladicarbene’) with an Electron-Rich Silicon(0) Atom. Angew. Chem., Int. Ed. 2013, 52, 7147−7150. (27) Mondal, K. C.; Roesky, H. W.; Schwarzer, M. C.; Frenking, G.; Tkach, I.; Wolf, H.; Kratzert, D.; Herbst-Irmer, R.; Niepötter, B.; Stalke, D. Conversion of a Singlet Silylene to a Stable Biradical. Angew. Chem., Int. Ed. 2013, 52, 1801−1805. (28) Holzmann, N.; Andrada, D. M.; Frenking, G. Bonding Situation in Silicon Complexes [(L)2(Si)2] and [(L)2Si] with NHC and cAAC Ligands. J. Organomet. Chem. 2015, 792, 139−148. (29) Xiong, Y.; Yao, S.; Tan, G.; Inoue, S.; Driess, M. A Cyclic Germadicarbene (“Germylone“) from Germyliumylidene. J. Am. Chem. Soc. 2013, 135, 5004−5007. (30) Takagi, N.; Shimizu, T.; Frenking, G. Divalent Silicon(0) Compounds. Chem. - Eur. J. 2009, 15, 3448−3456. (31) Xiong, Y.; Yao, S.; Müller, R.; Kaupp, M.; Driess, M. From Silylone to an Isolable Monomeric Silicon Disulfide Complex. Angew. Chem., Int. Ed. 2015, 54, 10254−10257. (32) Xiong, Y.; Yao, S.; Karni, M.; Kostenko, A.; Burchert, A.; Apeloig, Y.; Driess, M. Heavier Congeners of CO and CO2 as Ligands: ́ From Zero-Valent Germanium (Germyloné ) to Isolable Monomeric GeX and GeX2 Complexes (X = S, Se, Te). Chem. Sci. 2016, 7, 5462− 5469. (33) Su, B.; Ganguly, R.; Li, Y.; Kinjo, R. Synthesis, Characterization, and Electronic Structures of a Methyl Germyliumylidene Ion and Germylone-Group VI Metal Complexes. Chem. Commun. 2016, 52, 613−616. (34) Yao, S.; Xiong, Y.; Driess, M. unpublished results. The CIF of compound 5 is available in the Electronic Supporting Information. (35) Burchert, A.; Müller, R.; Yao, S.; Schattenberg, C.; Xiong, Y.; Kaupp, M.; Driess, M. Taming Silicon Congeners of CO and CO2: Synthesis of Monomeric SiII and SiIV Chalcogenide Complexes. Angew. Chem., Int. Ed. 2017, 56, 6298−6301. (36) Zabula, A. V.; Pape, T.; Hepp, A.; Schappacher, F. M.; Rodewald, U. Ch.; Pöttgen, R.; Hahn, F. E. Trapping of Tin(II) and Lead(II) Homologues of Carbon Monoxide by a Benzannulated Lutidine-Bridged Bisstannylene. J. Am. Chem. Soc. 2008, 130, 5648− 5649. (37) Thiele, G.; Franzke, Y.; Weigend, F.; Dehnen, S. {μ-PbSe}: A Heavy CO Homologue as an Unexpected Ligand. Angew. Chem., Int. Ed. 2015, 54, 11283−11288. (38) Wang, Y.; Chen, M.; Xie, Y.; Wei, P.; Schaefer, H. F., III; Schleyer, P. v. R.; Robinson, G. H. Stabilization of Elusive Silicon Oxides. Nat. Chem. 2015, 7, 509−513. (39) Mohapatra, C.; Mondal, K. C.; Samuel, P. P.; Keil, H.; Niepötter, B.; Herbst-Irmer, R.; Stalke, D.; Dutta, S.; Koley, D.; Roesky, H. W. A Stable Dimer of SiS2 Arranged between Two Carbene Molecules. Chem. - Eur. J. 2015, 21, 12572−12576. (40) Chandra Mondal, K.; Roy, S.; Dittrich, B.; Maity, B.; Dutta, S.; Koley, D.; Vasa, S. K.; Linser, R.; Dechert, S.; Roesky, H. W. A Soluble Molecular Variant of the Semiconducting Silicondiselenide. Chem. Sci. 2015, 6, 5230−5234. 2036

DOI: 10.1021/acs.accounts.7b00285 Acc. Chem. Res. 2017, 50, 2026−2037

Article

Accounts of Chemical Research (41) Burchert, A.; Yao, S.; Müller, R.; Schattenberg, C.; Xiong, Y.; Kaupp, M.; Driess, M. An Isolable Silicon Dicarbonate Complex from Carbon Dioxide Activation with a Silylone. Angew. Chem., Int. Ed. 2017, 56, 1894−1897. (42) Asay, M.; Jones, C.; Driess, M. N-Heterocyclic Carbene Analogues with Low-Valent Group 13 and Group 14 Elements: Syntheses, Structures, and Reactivities of a New Generation of Multitalented Ligands. Chem. Rev. 2011, 111, 354−396. (43) Shan, Y.-L.; Yim, W.-L.; So, C.-W. An N-Heterocyclic SilyleneStabilized Digermanium(0) Complex. Angew. Chem., Int. Ed. 2014, 53, 13155−13158. (44) Zhou, Y. − P.; Karni, M.; Yao, S.; Apeloig, Y.; Driess, M. A Bis(silylenyl)pyridine Zero-Valent Germanium Complex and Its Remarkable Reactivity. Angew. Chem., Int. Ed. 2016, 55, 15096−15099.

2037

DOI: 10.1021/acs.accounts.7b00285 Acc. Chem. Res. 2017, 50, 2026−2037