Home \ Browse Journal \ 2020 \ 2020 Issue 1 \ Polyhydride Complexes of Rare-Earth Metals

2020 Volume 3 Issue 1

инэос-open

INEOS OPEN, 2020, 3 (1), 1–19 

Journal of Nesmeyanov Institute of Organoelement Compounds
of the Russian Academy of Sciences

Download PDF

DOI: 10.32931/io2001r

issue_cover_html_3-1      

Polyhydride Complexes of Rare-Earth Metals

D. M. Lyubov a and A. A. Trifonov *a,b

a Razuvaev Institute of Organometallic Chemistry, Russian Academy of Sciences, ul. Tropinina 49, Nizhny Novgorod, 603950 Russia
b Nesmeyanov Institute of Organoelement Compounds, Russian Academy of Sciences, ul. Vavilova 28, Moscow, 119991 Russia
 

Corresponding author:  A. A. Trifonov, e-mail: trif@iomc.ras.ru
Received 13 January 2020; accepted 11 February 2020

Abstract

grab

The current review is devoted to a unique class of compounds, namely, polyhdride complexes of rare-earth metals bearing both cyclopentadienyl and multidentate noncyclopentadienyl ligands as stabilizing environments. Different synthetic aspects, structural features as well as reactivity of the polyhydride derivatives are highlighted.

Key words: rare-earth metals, lanthanides, hydride complexes, ligands, synthesis, structure, reactivity.

 

1. Introduction

Metal hydrides are the main intermediates in a variety of stoichiometric and catalytic reactions and their significance in inorganic and organic chemistry cannot be overemphasized [1–7]. Nowadays, the most explored hydride complexes among the rare-earth metal compounds are monohydride derivatives of a general formula [(L)2MH] or [(L)(L')MH] stabilized by two monoanionic (or one dianionic) auxiliary ligands [8–15]. The first example of hydride derivatives of sandwich rare-earth metal complexes, [(C5H4Me)2Y(μ-H)(THF)]2, was described over 40 years ago [16]. Although dihydride complexes of rare-earth metals [(L)MH2] stabilized by monoanionic ligands are of particular interest owing to their potential application in catalytic and stoichiometric transformations of a wide range of substrates and possibility of elucidation of their structural features and the factors that affect the reactivity of Ln–H bonds, they have been long unknown. The main challenges in the synthesis of dihydride derivatives [(L)MH2] are their extremely high reactivity and, as a consequence, low stability. The isolation of these complexes in individual forms is also associated with certain difficulties.

The main methods for the synthesis of rare-earth metal hydride complexes are the metatheses of Ln−C σ-bonds in bis(alkyl) derivatives [(L)LnR2] under the action of molecular hydrogen H2 [17, 18] or phenylsilane PhSiH3 [19]. The advances in the field of synthesis of bis(alkyl) complexes of rare-earth metals stabilized by both cyclopentadienyl and multidentate noncyclopentadienyl ligands [20–23] enabled the synthesis of the corresponding dihydrides [(L)MH2].

It should be noted that not only the synthesis but also the elucidation of structures of polyhydride complexes of rare-earth metals is a nontrivial task. The investigation of hydride metal complexes (including those of rare-earth elements) by X-ray crystallography, which is based on the diffraction of X-rays on electrons, faces the problem of localization of the hydride ligands in close proximity to the heavy metal ions. Therefore, XRD studies require high-quality single crystals to enable the objective localization of the hydride ligands. The positions of the hydride ligands in hydride metal complexes can be assigned unambiguously by neutron diffraction [24], which is based on neutron scattering on atom nuclei, including protons. However, this method is even more sensitive to the quality and sizes of single crystals. Furthermore, neutron diffraction requires longer durations of experiments than XRD.

In the case of complexes with diamagnetic rare-earth metal ions (Sc, Y, La, and Lu), the structures of polyhydride clusters in solution can be assigned by NMR spectroscopy. The most useful complexes are yttrium derivatives since yttrium is monoisotopic (89Y, 100%) and features the nuclear spin of ½, which provides an opportunity for investigation of the direct interaction of the hydride ligands with the metal ion in solution.

2.1. Polyhydride complexes stabilized by cyclopentadienyl ligands

The first examples of polyhydride complexes of rare-earth metals, namely, tetranuclear octahydride yttrium complexes [Cp'Y(μ-H)2]4(THF)x (x = 2 (1), x = 1 (2); Scheme 1) stabilized by bulky cyclopentadienyl ligands C5Me4SiMe3 (Cp') were obtained by Okuda et al. [25] in 2003 upon hydrogenolysis of bis(alkyl) half-sandwich yttrium complex Cp'Y(CH2SiMe3)2(THF) with molecular H2. It was revealed that the structures of the resulting polyhydride derivatives are defined by the nature of a solvent used for their synthesis (hexane or toluene). The hydrogenolysis of bis(alkyl) complex Cp'Y(CH2SiMe3)2(THF) in hexane led to tetranuclear complex 1 [Cp'Y(μ-H)2]4(THF)2 which contains two metal centers coordinated by the THF molecules. In contrast, the reaction in toluene afforded complex 2 [Cp'Y(μ-H)2]4(THF) bearing only one THF molecule. Luo et al. showed [26] that THF can be completely removed from the yttrium coordination sphere in 2 by recrystallization from hexane, which leads to non-solvate tetranuclear octahydride cluster [Cp'Y(μ-H)2]4 (3; Scheme 1).

sch1

Scheme 1

At the same time, Hou et al. showed by the example of bis(alkyl) lutetium derivative Cp'Lu(CH2SiMe3)2(THF) that the hydrogenolysis of Lu−C bonds (Scheme 2) leads to the formation of a mixture of tetranuclear complex [Cp'Lu(μ-H)2]4(THF) (4) [27], analogous to yttrium complex 2, and trinuclear pentahydride cluster [Cp'Lu]2[Cp'−HLu](μ-H)5(THF)2 (5) [28], bearing a dianionic moiety (µ-CH2SiMe2C5Me4). The dianionic moiety in complex 5 results from the activation of the Csp3−H bond of one of the SiMe3 groups of cyclopentadienyl ligand C5Me4SiMe3. The hydrogenolysis of Cp'Lu(CH2SiMe3)2(THF) in hexane leads to tetranuclear cluster 4 as a main reaction product (86%), whereas the reaction in THF affords trinuclear pentahydride derivative 5 (up to 90%).

sch2

Scheme 2

Polyhydride half-sandwich rare-earth metal complexes can be obtained also by Ln−C σ-bond metathesis under the action of phenylsilane [27]. The sequential addition of an equimolar amount and, then, an excess of PhSiH3 to bis(alkyl) derivative Cp'Lu(CH2SiMe3)2(THF) selectively leads at the first step to dimeric alkyl-hydride complex [Cp'Lu(CH2SiMe3)(µ-H)(THF)]2 (6) and, then, to octahydride cluster [Cp'Lu(μ-H)2]4 (7) which does not contain coordinated THF molecules (Scheme 2).

Subsequently, the hydrogenolysis of bis(alkyl) complexes Cp'Ln(CH2SiMe3)2(THF) afforded the corresponding tetranuclear octahydride derivatives of scandium as well as medium and late lanthanides [Cp'Ln(μ-H)2]4(THF)x (x = 0, Ln = Sc (8); x = 1, Ln = Gd (9), Dy (10), Ho (11), Er (12), Tm (13); Scheme 3) [28]. The structures of the resulting polyhydrides depended on the metal ion size. Thus, in the case of scandium, the reaction gave rise to tetranuclear octahydride complex 7 which does not contain the coordinated Lewis bases. For the larger ions Ln3+, THF adducts 9–13 were formed, in which one of the metal centers is coordinated by the THF molecule. Using the example of THF monoadduct 10, Cui et al. [29] showed that the recrystallization from THF/hexane furnishes the corresponding bis(tetrahydrofuranate) [Cp'Dy(μ-H)2]4(THF)2 (14; Scheme 3).

sch3

Scheme 3

Since the synthesis of bis(alkyl) derivatives of rare-earth metals Cp'LnR2(THF) with the large ionic radii (La, Ce, Pr, Nd, Sm) bearing CH2SiMe3 groups as alkyl ligands failed due to their instability, the synthesis of lanthanum and early lanthanide polyhydrides was carried out using the corresponding bis(о-dimethylaminobenzyl) complexes Cp'Ln(CH2C6H4NMe2-o)2. The hydrogenolysis of Cp'Ln(CH2C6H4NMe2-o)2 with molecular H2 (1 bar, THF, 12 h) led to tetranuclear octahydride complexes [Cp'Ln(µ-H)2]4(THF)2 (Ln = La (15), Ce (16), Pr (17), Nd (18), Sm (19); Scheme 4) which were isolated as the adducts with two THF molecules [28]. Noteworthy, in the case of bis(о-dimethylaminobenzyl) yttrium complex Cp'Y(CH2C6H4NMe2-o)2, which does not contain the coordinated THF molecules unlike Cp'Ln(CH2SiMe3)2(THF), the hydrogenolysis of the Y−C bonds with molecular Н2 in benzene resulted in pentanuclear decahydride [Cp'Y(μ-H)2]5 (20; Scheme 5). An analogous reaction of Cp'Y(CH2C6H4NMe2-o)2 with H2 in THF afforded, as expected, tetranuclear complex 1 [Cp'Y(μ-H)2]4(THF)2 in the form of an adduct with two molecules of THF [30].

sch4

Scheme 4

sch5

Scheme 5

To study the effect of the bulkiness of a cyclopentadienyl ligand on the nuclearity of polyhydride clusters [CpY(μ-H)2]x(THF)n, Shima et al. [30] explored bis(o-dimethylaminobenzyl) yttrium complexes CpY(CH2C6H4NMe2-o)2 bearing different cyclopentadienyl ligands (Cp = C5Me4H, C5Me5, C5Me4Et) as the starting bis(alkyl) derivatives in hydrogenolysis. The hydrogenolysis of (C5Me4H)Y(CH2C6H4NMe2-o)2, having the least bulky cyclopentadienyl ligand C5Me4H, in THF gave rise to tetranuclear octahydride cluster [(C5Me4H)Y(μ-H)2]4(THF)4 (21; Scheme 5), which contains one THF molecule per each yttrium atom. The reactions of complexes (C5Me4R)Y(CH2C6H4NMe2-o)2 (R = Me, Et) with H2 appeared to strongly depend on the nature of a solvent. Thus, in toluene the formation of mixed ligand benzyl-hydride complex [(Cp*Y)2(μ-CH2C6H4NMe2-o)2(μ-H)3]2 (22, Cp* = C5Me5; Scheme 5) was observed. This complex is a product of partial hydrogenation of the Y−C bond of the starting bis(alkyl) derivative. However, in THF the reaction resulted in pentanuclear decahydride clusters [(C5Me4R)Y(μ-H)2]5(THF)2 (R = Me (23), Et (24); Scheme 5). Furthermore, the Y−C σ-bond metathesis in complex Cp*Y(CH2C6H4NMe2-o)2 under the action of PhSiH3 in benzene afforded hexanuclear cluster [Cp*Y(μ-H)2]6 (25, Scheme 5) having twelve bridging hydride ligands.

Li et al. [31] used polyhydride yttrium clusters 1–3 [Cp'Y(μ-H)2]4(THF)n (n = 0, 1, 2) stabilized by bulky cyclopentadienyl ligand Cp' to obtain the corresponding cationic polyhydrides via selective protonolysis of one of the hydride ligands under the action of an equimolar amount of trityl borate [Ph3C][B(C6F5)4]. It was shown that the structures of the resulting tetranuclear cationic polyhydride clusters depend on the presence of coordinated THF molecules in the initial complexes. The use of the THF adducts of 1 and 2 led to the formation of tetranuclear cationic polyhydrides in the form of separated ion pairs [(Cp'Y)4(μ-H)7(THF)n]+[B(C6F5)4]− (n = 2 (26), 1 (27); Scheme 6). In the case of solvent-free derivative 3, the reaction with [Ph3C][B(C6F5)4] gave rise to contact ion pair [(Cp'Y)4(μ-H)7]+[[B(C6F5)4]− (28; Scheme 6) owing to a contact between one Y atom and F atom at the meta-position of one of the phenyl rings of triphenylborate anion [B(C6F5)4]− (Y···F 2.405(4) Å).

sch6

Scheme 6

Hence, the cyclopentadienyl ligands can serve as a convenient environment for the synthesis of polyhydride complexes of rare-earth metals. The variation of reaction conditions, ionic radii of Ln3+ ions, nature of the leaving group in the initial bis(alkyl) derivatives (CH2SiMe3 vs CH2C6H4NMe2-o), and the bulkiness of the cyclopentadienyl anion allows one to obtain half-sandwich complexes [CpLn(μ-H)2]n(THF)x in the form of tri-, tetra-, penta- and hexanuclear clusters. The polynuclear structures of dihydrides [CpLn(μ-H)2]n(THF)x were unambiguously confirmed by XRD. According to the results of XRD analysis, the binding of the metal centers in the clusters is accomplished by the bridging hydride ligands. The coordination mode of the latter in complexes [CpLn(μ-H)2]n(THF)x is defined by the degree of compound association, ionic radius of the Ln3+ ion as well as the presence of coordinated THF molecules (Table 1).

Table 1. Ln−Ln and Ln−H distances in polyhydride complexes [CpLn(μ-H)2]n(THF)x stabilized by cyclopentadienyl ligands (Å)

Trinuclear complexes

Ln−Ln

Ln−μ2-H

Ln−μ3-H

Ln−μ4-H

Ln−μ5-H

Ln−μ6-H

[Cp'Lu]2[Cp'−HLu](μ-H)5(THF)2 (5)

3.271(1)–3.387(1)

2.02–2.13

2.07–2.48

–

–

–

Tetranuclear complexes

Ln−Ln

Ln−μ2-H

Ln−μ3-H

Ln−μ4-H

Ln−μ5-H

Ln−μ6-H

[Cp'Sc(μ-H)2]4 (8)

3.211(1)–3.278(1)

1.82–1.96

2.11–2.54

1.96–1.99

–

–

[Cp'Y(μ-H)2]4 (3)

3.460(3)–3.620(3)

2.11–2.23

2.34–2.42

2.10–2.26

–

–

[Cp'Lu(μ-H)2]4 (6)

3.276(1)–3.463(1)

2.51–2.70

1.97–2.46

2.02–2.11

–

–

[Cp'Y(μ-H)2]4(THF) (2)

3.412(1)–3.688(1)

2.08–2.33

2.32–2.36

2.14–2.23

–

–

[Cp'Gd(μ-H)2]4(THF) (9)

3.289(1)–4.220(1)

2.09–2.27

2.12–2.30

2.17–2.36

–

–

[Cp'Dy(μ-H)2]4(THF) (10)

3.287(1)–4.194(1)

2.01–2.23

2.08–2.35

2.13–2.38

–

–

[Cp'Er(μ-H)2]4(THF) (12)

3.375(1)–3.654(1)

1.81–2.29

2.23–2.53

2.06–2.35

–

–

[Cp'Lu(μ-H)2]4(THF) (4)

3.302(1)–3.615(1)

1.85–2.49

2.03–2.08

1.94–2.33

–

–

[Cp'Y(μ-H)2]4(THF)2 (1)

3.329(2)–3.922(2)

1.98–2.31

2.07–2.39

–

–

–

[Cp'La(μ-H)2]4(THF)2 (15)

3.734(1)–4.126(1)

2.32–2.56

2.28–2.54

–

–

–

[Cp'Ce(μ-H)2]4(THF)2 (16)

3.580(1)–4.060(1)

2.29–2.68

2.28–2.42

–

–

–

[Cp'Pr(μ-H)2]4(THF)2 (17)

3.606(1)–4.032(1)

2.18–2.31

2.29–2.39

–

–

–

[Cp'Nd(μ-H)2]4(THF)2 (18)

3.567(1)–3.994(1)

2.12–2.39

2.18–2.39

–

–

–

[Cp'Sm(μ-H)2]4(THF)2 (19)

3.509(1)–3.942(1)

2.14–2.32

2.20–2.31

–

–

–

[Cp'Dy(μ-H)2]4(THF)2 (14)

3.422(1)–3.867(1)

2.05–2.29

2.08–2.44

–

–

–

[(C5Me4H)Y(μ-H)2]4(THF)4 (21)

3.361(1)–3.986(1)

2.12–2.17

2.22–2.27

–

–

–

Pentanuclear complexes

Ln−Ln

Ln−μ2-H

Ln−μ3-H

Ln−μ4-H

Ln−μ5-H

Ln−μ6-H

[Cp'Y(μ-H)2]5 (20)

3.392(1)–3.686(1)

1.95–2.12

2.14–2.32

–

2.37–2.65

–

[(C5Me4Et)Y(μ-H)2]5(THF)2 (24)

3.314(1)–3.756(1)

2.02–2.23

2.03–2.50

–

2.11–2.74

–

Hexanuclear complexes

Ln−Ln

Ln−μ2-H

Ln−μ3-H

Ln−μ4-H

Ln−μ5-H

Ln−μ6-H

[Cp*Y(μ-H)2]6 (25)

3.524(3)–3.714(3)

2.26–2.59

2.00–2.40

–

–

2.34–2.66

Thus, for example, in trinuclear complex 5 [Cp'Lu]2[Cp'−HLu](μ-H)5(THF)2 the binding of three (Cp'Lu) moieties is realized through five bridging hydride ligands: three of them are μ2-bridging ligands and are located in an Lu3 plane, and the other ligands are μ3-bridging and are located above and below the Lu3 plane, respectively.

In the case of tetranuclear complexes 3, 6, and 8 [Cp'Ln(μ-H)2]4 (Ln = Sc, Y, and Lu) as well as their counterparts 2, 4, 9, 10, and 12 [Cp'Ln(μ-H)2]4(THF) (Ln = Y, Gd, Dy, Er, Lu), bearing one coordinated THF molecule, four atoms of the rare-earth element are bound by eight hydride ligands and form distorted octahedra. The center of the Ln4 octahedron is occupied by one μ4-bridging hydride ligand. The remaining seven hydrides serve as μ2- and μ3-bridging ligands and are located above edges or faces of the Ln4 tetrahedra, respectively. The DFT calculations performed for non-solvate complexes [Cp'Ln(μ-H)2]4 3 and 6 (Ln = Y, Lu) showed [26] that optimized core Ln4H8 has a pseudo-C3v symmetrical structure with one µ4-bridging hydride ligand in the center of the Ln4 tetrahedron. According to the computations, the binding of the metal centers is realized through both the bridging hydride ligands and the direct overlap of the metal orbitals with each other. Yousufuddin et al. [24] confirmed the presence of the central μ4-bridging hydride ligand in complex 2 experimentally by neutron diffraction.

The presence of the second coordinated THF molecule in clusters 1 and 14–19 [Cp'Ln(μ-H)2]4(THF)2 (Ln = Y, La, Ce, Pr, Nd, Sm) led to the absence of a μ4-bridging hydride ligand in the center of the Ln4 cluster. In this case, the rare-earth metal binding in the Ln4 octahedral moiety is accomplished through four μ2- and four μ3-bridging hydride ligands, which, as well as in the case of monoadducts 2, 4, 9, 10, and 12 [Cp'Ln(μ-H)2]4(THF), are located above edges and faces of the Ln4 octahedron, respectively. The analogous structure of Ln4(μ-H)8 moiety was observed also in complex 21 [(C5Me4H)Y(μ-H)2]4(THF)4.

Cationic polyhydride derivatives 26–28 [(Cp'Y)4(μ-H)7(THF)n]+[B(C6F5)4]− demonstrate analogous structures of the Y4H7 cationic moiety. Four yttrium atoms form a slightly distorted Y4 octahedron, which center is occupied by one μ4-bridging ligand. Two μ3-bridging hydride ligands are located above two edges, and the remaining four ligands adopt a μ2-bridging mode.

Pentanuclear decahydride complexes 20 [Cp'Y(μ-H)2]5 and 24 [(C5Me4Et)Y(μ-H)2]5(THF)2 contain a pentacoordinated hydride atom, which is located inside a distorted Y5 tetragonal pyramid formed by five yttrium atoms. Interestingly, the μ5-bridging hydride ligand in complexes 20 and 24 is located not in the center of the Y5 moiety but almost in one plane with the pyramid basis. Complexes 20 and 24 have two μ3-bridging hydride ligands which are located above two adjacent side faces of the Y5 pyramid; the rest seven hydride ligands are bound to the yttrium atom in a μ2-mode.

In the case of hexanuclear complex 25 [Cp*Y(μ-H)2]6, the yttrium atoms form an Y6 octahedron with the distances of 3.524(3)–3.714(3) Å between the atoms on edges and 4.682(3)–5.325(3) Å between the yttrium atoms at opposite vertices, respectively. The yttrium atoms in the octahedron are connected with each other by twelve bridging hydride ligands. Eight hydride ligands are μ3-bridges and are located above edges of the Ln6 octahedron. Three hydrides serve as μ2-bridging ligands and are located above the octahedron edges. The center of the Ln6 octahedron is occupied by one μ6-bridging hydride ligand.

Half-sandwich polyhydride complexes of rare-earth metals [CpLn(μ-H)2]n(THF)x display unique reactivity in stoichiometric and catalytic transformations of unsaturated substrates [31–35]. Half-sandwich polyhydride complexes can add across double C=C and triple C≡C bonds as well as across multiple C=O and C≡N bonds.

The stoichiometric reactions of tetranuclear octahydride complex 2 [Cp'Y(μ-H)2]4(THF) with styrene, 1,3-cyclohexadiene, and 1,4-bis(trimethylsilyl)butadiyne-1,3 afforded the products of insertion of unsaturated hydrocarbons across Ln−H bonds in individual forms. The reaction of 2 with styrene resulted in benzyl-pentahydride (Cp'Y)4(μ-H)7(μ-CH(Me)Ph) (29; Scheme 7) [36], in which the bridging benzyl ligand (μ-CH(Me)Ph) is bound with one Y atom in a η3-allyl mode through the benzyl carbon atom and ipso- and ortho-carbon atoms of the phenyl ring as well as with the second Y atom in a η2-mode through the carbon atoms at the meta- and para-positions. The insertion of the second equivalent of styrene across the Y−H bond was not observed even in the presence of an excess of the alkene. The hydrogenolysis of benzyl derivative 29 with molecular Н2 (1 bar) in benzene is accompanied by the hydrogenation of the bridging benzyl ligand (μ-CH(Me)Ph) with elimination of ethylbenzene and formation of octahydride yttrium complex 3 [Cp'Y(μ-H)2]4 which does not contain THF. The reaction of 2 with 1,3-cyclohexadiene proceeded as 1,4-addition across the Y−H bond along a system of the conjugated double bonds of the substrate. The product was complex (Cp'Y)4(μ-H)7(μ-cyclo-C6H9) (30; Scheme 7) containing a bridging monoanionic allyl ligand (μ-cyclo-C6H9), which is bound with two Y atoms in a η1-mode through the terminal carbon atoms [30]. As well as in the case of styrene, the use of an excess of 1,3-cyclohexadiene does not lead to further reaction. The interaction of 2 with 1,4-bis(trimethylsilyl)-1,3-butadiyne gave tetranuclear tetrahydride complex (Cp'Y)4(μ-H)4(η4-Me3SiCCHCHCSiMe3) (31; Scheme 7) [36, 37], which contains bridging tetraanionic moiety Me3SiCC(H)C(H)CSiMe3 connected simultaneously with two Y atoms in a η4-mode.

sch7

Scheme 7

The interaction of octahydride lutetium complex 4 [Cp'(μ-H)2](THF) with γ-butyrolactone led to tetranuclear mixed ligand alkoxide dihydride derivative (Cp'Lu)4(μ-H)4(μ-O(CH2)4O)4 (32; Scheme 8) [36].

sch8

Scheme 8

γ-Butyrolactone is known to be a stable cyclic ester that does not undergo polymerization; however, in this reaction three molecules of γ-butyrolactone underwent complete reduction with ring opening due to addition of six out of eight hydride ligands in 4, giving rise to −O(CH2)4O− diolate moieties. There was no further interaction of complex 32 with an excess of γ-butyrolactone. However, it was mentioned that complexes 4 and 32 exhibit high activity in ring-opening polymerization of ε-caprolactone.

Polyhydride complexes [Cp'Ln(μ-H)2]4(THF) display unprecedented reactivity towards carbon monoxide CO, promoting its reduction to ethylene CH2=CH2. Polyhydride complexes 2 and 4 [Cp'Ln(μ-H)2]4(THF) (Ln = Y, Lu) readily react with CO (toluene, −10 °С, 3 min), giving rise to tetranuclear hexahydrido-oxymethylene derivatives (Cp'Ln)4(μ-H)6(μ-CH2O) (Ln = Y (33), Lu (34); Scheme 9) [38] which contain a bridging dianionic oxomethylene ligand. Using yttrium complex 33 as an example, it was shown that it is able to react with one more equivalent of CO. Thus, the dissolution of 33 in THF under CO atmosphere (20 °С, 2 h) or the addition of CO to complex 2 in THF (20 °С, 2 h) results in dioxo-tetrahydride complex (Cp'Y)4(µ3-O)2(µ-H)4(THF) (36; Scheme 9) bearing two μ3-bridging oxo ligands. The formation of 36 was accompanied by the release of ethylene. The NMR spectroscopic monitoring of the reaction of 33 with СО in THF-d8 at low temperatures evidenced the formation of enolate intermediate (Cp'Y)4(µ-O)(µ-H)5(OCH=CH2)(THF) (35; Scheme 9). The subsequent interaction of dioxo complex 36 with СО afforded the corresponding tetraoxo complex [Cp'Y(μ-O)]4 (37; Scheme 9) and was accompanied by elimination of the second ethylene molecule. Analogous lutetium complex [Cp'Lu(μ-O)]4 (38; Scheme 9) was obtained upon treatment of octahydride 4 with an excess of СО (1 bar, THF, 20 °С).

Nishiura et al. [28] showed that polyhydride complexes 2, 4, and 10 [Cp'Ln(μ-H)2]4(THF) (Ln = Y, Dy, Lu) can serve as single-component catalysts for homopolymerization of cyclohexene oxide as well as copolymerization of cyclohexene oxide with CO2. The homopolymerization of cyclohexene oxide initiated by complexes 2, 4, and 10 affords the corresponding ethers (Mn = (60–80)·103; Mw/Mn ≈ 2) in high yields, while the copolymerization of cyclohexene oxide and CO2 (70–110 °C, 12 bar) furnishes polycarbonates (Mn = (20–40)·103, Mw/Mn = 4–5) with the content of carbonate units up to 95–99%. Presumably, a step that initiates copolymerization is the formation of methylene-diolate intermediates (−OCH2O−) due to the insertion of a molecule of CO2 into two Ln−H bonds. It was noted that polyhydride complex 2 [Cp'Y(μ-H)2]4(THF) readily reacts with CO2; however, the authors failed to isolate any individual product from the reaction mixture.

sch9

Scheme 9

The interaction of dihydrides 2 and 4 [Cp'Ln(μ-H)2]4(THF) (Ln = Y, Lu) with four equivalents of nitriles R−C≡N (R = Ph, Me) afforded cubane-like tetranuclear imide derivatives (R = Ph, Ln = Y (39), Lu (40) [36]; R = Me, Ln = Y (41) [39]; Scheme 10), the analogs of oxo complexes 37 and 38 [Cp'Ln(μ-O)]4 [Cp'Ln(µ3-NCH2R)]4. During the reactions, triple C≡N bonds of the nitriles underwent complete hydrogenation as a result of double insertion into Ln−H bonds.

sch10

Scheme 10

Shima and Hou [40] described an interesting example of selective protonolysis of the Ln−H bonds in tetranuclear octahydride complexes 2 and 4 [Cp'Ln(μ-H)2]4(THF) (Ln = Y, Lu) under the action of ammonia. Treatment of complexes 2 and 4 with ammonia (1 bar, C6D6, 30 min) led to the formation of heptaamido-monohydride derivatives (Cp'Ln)4(μ-NH2)7(μ-H) (Ln = Y (42), Lu (43); Scheme 11). Complexes 42 and 43 retained the Ln4 tetranuclear structures analogous to those of starting octahydrides 2 and 4; only μ2- and μ3-bridging hydride ligands underwent protonolysis, whereas the central μ4-bridging hydride ligand remained unreacted. This evidences that the reactivity of the μ4-bridging hydride atom inside the Ln4 tetrahedron is essentially lower. However, an increase in the duration of a reaction with NH3 from 30 min to 2 days enabled the protonolysis of even μ4-bridging hydride ligand, which resulted in tetranuclear octoamide complex [Cp'Lu(μ-NH2)2]4 (44; Scheme 11) in which four Lu atoms are connected with each other by eight bridging amide (NH2)− ligands.

sch11

Scheme 11

2.2. Polyhydride complexes stabilized by bidentate ligands

The first example of polyhydride derivatives of rare-earth metals in a noncyclopentadienyl ligand environment was obtained accidentally [41]. The hydrogenolysis of monoalkyl complex [ArN(CH2)2NAr]Y[CH(SiMe3)2](THF) (Ar = C6H3iPr2-2,6), bearing a dianionic amide ligand, with molecular H2 (1.5 bar) in toluene proceeded as the partial hydrogenolysis of the diamide ligand Y−N bond along with the expected hydrogenation of the Y−C bonds. The reaction product appeared to be trinuclear polyhydride complex [Y3{ArNH(CH2)2NAr}2{ArN(CH2)2NAr}(m-H)3(m3-H)2(THF)] (45; Scheme 12) containing two yttrium ions coordinated by new monoanionic amido-amine ligand {ArNH(CH2)2NAr} and one yttrium ion bound with the initial diamide ligand {ArN(CH2)2NAr}. However, the yield of this compound was only 9% and the authors failed to reproduce the synthesis.

sch12

Scheme 12

According to the results of XRD analysis, three metal centers in complex 45 are bound with five hydride ligands, three of which serve as µ2-bridging ligands, providing almost planar six-membered ring Y3(µ2-H)3, and the remaining two hydrides serve as µ3-bridging ligands and are located on either side of the Y3(µ2-H)3 plane. The distances Y-(µ2-H) in 45 are within 1.98(6)–2.18(6) Å and, as it was expected, are shorter than those between the yttrium atoms and µ3-bridging hydrides (1.94(12)–2.39(8) Å).

It should be noted that the hydrogenolysis of Ln−N bonds was recently detected while attempting to obtain a dihydride yttrium complex stabilized by monoanionic bidentate amido-imine ligand [ArNC(=CH2)‒C(Me)=NAr] (Ar = C6H3iPr2-2,6) [42]. Treatment of bis(alkyl) complex [ArNC(=CH2)‒C(Me)=NAr]Y(CH2SiMe3)2(THF) (Ar=C6H3-iPr2-2,6) with molecular H2 resulted in the hydrogenation of the double C=N and C=C bonds of the amido-imine ligand along with the hydrogenolysis of the Y−C bonds. The reaction was also accompanied by the redistribution of the nitrogen-containing ligands and gave rise to a mixture of complexes [ArNCH(Me)CH(Me)NAr]Y[ArNC(=CH2)CH(Me)N(H)Ar](THF) and [ArNCH(Me)C(=CH2)NAr]Y[ArNC(=CH2)CH(Me)N(H)Ar](THF) (46MeMe and 46MeСH2) bearing, along with dianionic diamide [ArNCH(Me)CH(Me)NAr]2− or [ArNCH(Me)C(=CH2)NAr]2− ligands, monoanionic en-amido-amine moiety [ArNC(=CH2)CH(Me)N(H)Ar]− (Scheme 13). The en-amido-amine ligand [ArN−C(=CH2)−CH(Me)−N(H)Ar]− was likely to result from the hydrogenation of the C=N bond followed by the cleavage of the Y−N bond under the action of H2, which led to the formation of an amino group.

sch13

Scheme 13

The first reliably characterized polyhydride complex of rare-earth metals in a noncyclopentadienyl environment was obtained upon application of bulky bidentate amidopyridinate ligands Ap (Ap* = N-(2,6-diisopropylphenyl)-6-(2,4,6-triisopropylphenyl)pyridine-2-amide; Ap9Me = N-mesityl-6-(2,4,6-triisopropylphenyl)pyridine-2-amide). They afforded a series of trinuclear alkyl-hydride clusters of rare-earth metals [(ApLn)3(μ-H)5(CH2SiMe3)(THF)2] which were structurally characterized [43–45].

It was shown that the interaction of bis(alkyl) complexes Ap*Ln(CH2SiMe3)2(THF) (Ln = Y, Er, Yb, Lu) with molecular H2 (5 bar, 24 h) or PhSiH3 (Ln:PhSiH3 = 1:2) results selectively in the formation of unusual trinuclear alkyl-hydride complexes [(Ap*Ln)3(μ-H)5(CH2SiMe3)(THF)2] (Ln = Y (47), Er (48), Yb (49), Lu (50); Scheme 14), in which three ApLn moieties are connected with each other through five bridging hydride atoms [43, 44]. One out of three rare-earth metal atoms remains covalently bound with −CH2SiMe3 alkyl group, which was not substituted even in the presence of an excess of PhSiH3. It should be noted that bis(alkyl) scandium complex Ap*Sc(CH2SiMe3)2(THF), bearing the metal atom with the smallest ionic radius in the series (0.745 Å) [47], appeared to be absolutely inert towards phenylsilane [46]. Such a difference in the reactivities was explained by the highly sterically crowded coordination sphere of the scandium atom, which, in turn, hampered the approach of a molecule of PhSiH3 to the metal ion.

sch14

Scheme 14

In the case of the less sterically hindered amidopyridinate ligand Ap9Me (N-mesityl-6-(2,4,6-triisopropylphenyl)pyridine-2-amide) bearing a mesityl substituent instead of the diisopropyl-phenyl group at the anilide nitrogen atom, the formation of analogous trinuclear alkyl-hydride complexes was observed only for Lu. Complex [(Ap9MeLu)3(μ-H)5(CH2SiMe3)(THF)2] (53; Scheme 15) was synthesized in high yield both upon treatment of the corresponding bis(alkyl) complex Ap9MeLu(CH2SiMe3)2(THF) with two equivalents of PhSiH3 and with molecular hydrogen H2 (3 bar, 36 h) [45]. For the metals with slightly larger ionic radii, such as yttrium and ytterbium (ionic radius for CN 7: Y3+ 0.96 Å; Yb3+ 0.93 Å; Lu3+ 0.89 Å [47]), the interaction of bis(alkyl) complexes Ap9MeLn(CH2SiMe3)2(THF) (Ln = Y, Yb) proceeded in the more complicated manner and afforded a mixture of trinuclear polyhydride clusters [(Ap9MeLn)3(μ-H)5(CH2R)(THF)2] (Ln = Y (51), Yb (52); Scheme 15), which differ in the nature of the remaining alkyl group (CH2SiMe3 or CH2SiH2Ph) covalently bound with one of three metal centers [45]. The formation of Ln−CH2SiH2Ph moiety was explained by the unusual metathesis of the Ln−CH2SiMe3 σ-bond under the action of phenylsilane (Scheme 15). Presumably, this process was accompanied by the cleavage of the silicon–carbon bond in the LnCH2−SiMe3 moiety, which led to the elimination of trimethylsilane HSiMe3. The presence of the latter in the reaction mixture was confirmed by GC-MS and NMR spectroscopy.

sch15

Scheme 15

According to the XRD data, in crystals of complexes [(ApLn)3(μ-H)5(CH2SiMe3)(THF)2] three hydride ligands are μ2-bridging and lie in the plane formed by three metal centers, whereas the remaining two hydrides are μ3-bridging and are located above and below the Ln3 plane, respectively (Table 2). The detailed investigation of the alkyl-hydride complexes of diamagnetic metals Y (47, 51) and Lu (50, 53) in solution using multinuclear and two-dimensional NMR spectroscopy enabled the description of dynamics of the hydride ligands inside trinuclear moiety Ln3(μ-H)5. It was shown that in solution two µ2-bridging hydride ligands, which bind pairwise a rare-earth metal atom and alkyl group ApLn(CH2SiMe3), and rare-earth metals coordinated by THF ApLn(THF) do not participate in the dynamic process. At the same time, the third μ2-bridging hydride ligand, located between ApLn(THF) moieties, and μ3-bridging hydride atoms undergo an exchange process which is rapid within the NMR timescale (Scheme 16).

sch16

Scheme 16

Table 2. Ln−Ln and Ln−H bond distances in polyhydride complexes [(ApLn)3(μ-H)5(CH2SiMe3)(THF)2]
and [(Ap*Y)3(μ-H)5(THF)3][B(C6F5)4] stabilized by amidopyridinate ligands (Å)

Complex

Ln−Ln

Ln−μ2-H

Ln−μ3-H

Ln−C

[(Ap*Y)3(μ-H)5(CH2SiMe3)(THF)2] (47)

3.441(1)–3.515(1)

2.08–2.16

2.19–2.42

2.402(5)

[(Ap9MeY)3(μ-H)5(CH2SiMe3)(THF)2] (51)

3.455(1)–3.506(1)

2.03–2.15

2.22–2.30

2.446(6)

[(Ap*Y)3(μ-H)5(THF)3]+[B(C6F5)4]− (54)

3.482(1)–3.487(1)

1.95–2.20

2.22–2.32

−

[(Ap*Er)3(μ-H)5(CH2SiMe3)(THF)2] (48)

3.416(3)–3.470(3)

2.02–2.15

2.12–2.31

2.445(6)

[(Ap*Yb)3(μ-H)5(CH2SiMe3)(THF)2] (49)

3.362(1)–3.413(1)

1.90–2.11

2.15–2.40

2.362(9)

[(Ap9MeYb)3(μ-H)5(CH2SiMe3)(THF)2] (52)

3.349(1)–3.423(1)

2.05

2.22

2.341(4)

Despite the fact that the remaining alkyl group at one of the rare-earth metal atoms in clusters 47–53 [(ApLn)3(μ-H)5(CH2SiMe3)(THF)2] could not be replaced even under the action of an excess of PhSiH3, the DFT calculations demonstrated [44, 45] that HUMO (−4.24 eV) and HUMO-1 (−4.35 eV) orbitals are localized just on the ApYR alkyl moiety. Therefore, the Y−C bond is the most reactive one. This was confirmed experimentally by the selective protonolysis of the Y−CH2SiMe3 bond upon treatment of alkyl-hydride complex [(Ap*Y)3(μ-H)5(CH2SiMe3)(THF)2] 47 with the Broensted acid ([HNMe2Ph][B(C6F5)4]) (Scheme 17). The reaction product appeared to be cationic polyhydride yttrium complex [(Ap*Y)3(μ-H)5(THF)3][B(C6F5)4] (54) [44]. In the crystalline state, the structure of trinuclear polyhydride moiety Y3(μ-H)5 is analogous to that of the initial alkyl-hydride derivative; however, the empty coordination site of the released alkyl group, CH2SiMe3, is occupied by the additional molecule of THF.

sch17  

Scheme 17

It was shown that alkyl-hydride complexes 47–53 and cationic hydride complex 54 are efficient single-component catalysts for ethylene polymerization [45]. As it was expected, for a series of the Ap*-containing alkyl-hydride complexes, the catalytic activity was found to depend on the ionic radius of the central metal. The highest activity was observed for the yttrium derivatives: alkyl-hydride complex 47 (560 g∙mmol−1∙bar−1∙h−1) and cationic polyhydride complex 54 (465 g∙mmol−1∙bar−1∙h−1). Ytterbium 49 and lutetium 50 analogs exhibited considerably lower activity (165 and 168 g∙mmol−1∙bar−1∙h−1, respectively), whereas erbium derivative 48 with the metal ionic radius close to the that of the yttrium counterpart (Y3+ 0.96 Å vs Er3+ 0.95 Å [47]) appeared to be almost inactive (12 g∙mmol−1∙bar−1∙h−1). A transition from Ap*- to the less bulky Ap9Me-containing derivative in the case of ytterbium complexes led to a drastic increase (49) in the catalytic activity to 880 g∙mmol−1∙bar−1∙h−1 (52). Alkyl-hydride complexes 47–53 and cationic polyhydride complex 54 were not active in styrene polymerization but exhibited low activity in propylene polymerization. Ap*-Containing alkyl-hydride derivatives 47 and 50 demonstrated the efficiencies of 64 and 37 g∙mmol−1∙bar−1∙h−1 for propylene, respectively (toluene, 0 ° C, 0.5 bar), but both catalysts lost their activity in 5 h. Our research group [45] also reported on the possibility of isoprene polymerization using polyhydride derivatives 47 and 54 as single-component catalysts. Alkyl-hydride complex 47 appeared to be inactive, while cationic polyhydride 54 at [IP]/[Cat] = 1000/1 over 24 h afforded the polymer in 45% yield. The resulting polymers featured a bimodal molar mass distribution (Mw/Mn = 1.57, Mw = 11.6×105; Mw/Mn = 1.97, Mw = 7.3 105) and contained predominantly 1,4-cis-units (up to 63%; 1,4-trans-units 4%; 3,4-units 33%).

The bulky benzamidinate ligands with steric properties and geometry of a coordination site similar to those of the amidopyridinate derivatives also afford polyhydride complexes of rare-earth metals. Cheng et al. [48] synthesized yttrium polyhydride complexes via hydrogenolysis of bis(alkyl) complex (Amd)Ln(CH2SiMe3)2(THF) (Amd = PhC(NC6H3iPr2-2,6)2, Ln = Y, Lu) with molecular H2 (10 bar, THF, 0 °C), which gave rise to dimeric polyhydrides [(Amd)Ln(THF)2(μ-H)3Ln(Amd)(H)(THF)] (Ln = Y (55), Lu (56); Scheme 18). Moreover, the hydrogenolysis of cationic alkyl complex [(Amd)Ln(CH2SiMe3)(THF)3][BPh4] under the action of molecular H2 (10 bar, THF, 0 °C) enabled the synthesis of the rare examples of cationic hydride derivatives of rare-earth metals in the form of monomeric [(Amd)LnH(THF)3][BPh4] (Ln = Y (57), Lu (58)) or dimeric [(Amd)Ln(μ-H)(THF)2]2[BPh4]2 (Ln = Y (59), Lu (60); Scheme 18) complexes [49].

sch18

Scheme 18

In the crystalline state, polyhydride complexes of Y (55) and Lu (56) are dimers, in which two (Amd)Ln moieties are bound by three μ2-bridging hydride ligands and one of the Ln atoms in the dimer is attached to the terminal hydride ligand. The distances between the Lu atoms and the μ2-bridging hydride ligands in 56 are within 2.00–2.32 Å (Table 3); surprisingly, the length of the Lu−H bond with the terminal hydride atom is significantly longer (2.40 Å). In solution, the dimeric structures of complexes 55 and 56 are preserved. This is confirmed by the presence of a single triplet signal in the 1H NMR spectrum (293 K , THF-d8) at 6.34 ppm (1JYH= 25.2 Hz), which can be attributed to the hydride ligands connected simultaneously with two equivalent 89Y nuclei (I = ½, 100%).

Table 3. Ln−Ln and Ln−H distances in polyhydride and cationic hydride complexes stabilized by amidinate ligands (Amd) (Å)

Complex

Ln−Ln

Ln−HTerminal

Ln−μ2-H

[(Amd)Y(THF)2(μ-H)3Y(Amd)(H)(THF)] (55)

3.361(2)

2.40

2.01–2.31

[(Amd)Y(μ-H)(THF)2]2[BPh4]2 (59)

3.642(2)

−

2.09–2.11

[(Amd)Y(THF)2(μ-H)3Y(Amd)(OBu)(THF)] (61)

3.390(2)

−

1.95–2.30

[(Amd)Y(THF)(μ-H)]2(μ-PhCH−C≡C−CHPh) (64)

3.340(4)

−

2.14–2.17

[(Amd)Y(μ-H)]2(μ-ArN−CH2) (65)

3.270(4)

−

2.01–2.26

[(Amd)Lu(THF)2(μ-H)3Lu(Amd)(H)(THF)] (56)

3.236(2)

2.29

2.04–2.18

[(Amd)LuH(THF)3][BPh4] (58)

−

2.00

−

[(Amd)Lu(μ-H)(THF)2]2[BPh4]2 (60)

3.526(2)

−

2.02–2.15

[(Amd)Lu(μ2-H)3Lu(CPh=CHPh)(Amd)](THF)3 (62)

3.281(2)

−

1.99–2.26

[(Amd)Lu(THF)]2(μ2-H)2(μ-PhCH−CHPh) (63)

3.205(3)

−

1.93–2.28

In spite of the fact that all four hydride ligands in complexes 55 and 56 are equivalent in solution according to the NMR spectroscopic data, they display different reactivities. For example, yttrium complex 55 appeared to be unstable in THF and gradually reacted with the solvent, yielding quantitatively mixed butoxide-hydride derivative [(Amd)Y(THF)2(μ-H)3Y(Amd)(OBu)(THF)] (61, Scheme 18) as a result of the cleavage of the C−O bond in the THF molecule. Note that only the terminal hydride ligand reacted with THF, whereas μ2-bridging hydrides appeared to be inert to THF even at an elevated temperature (70 °С).

The investigation of reactivity of dihydride complexes of Y 55 and Lu 56 towards tolane, 1,4-diphenyl-1,3-butadiyne, and 2,6-dimethylphenylisocyanide revealed the possibility of addition of the Ln−H bond across multiple C≡C and C≡N bonds [48].

Thus, the interaction of Lu complex 56 with tolane PhC≡CPh was accompanied by the addition of the terminal Lu−H bond across the multiple C≡C bond, which led to trihydrido-1,2-diphenylethenyl derivative [(Amd)Lu(μ-H)3Lu(CPh=CHPh)(Amd)](THF)3 (62, Scheme 19). The subsequent transformation of 1,2-diphenylethenyl ligand in the coordination sphere of the rare-earth metal involved one of μ2-hydride ligands and led to hydrogenation of the С=С bond of 1,2-diphenylethenyl moiety. The product of this transformation was complex [(Amd)Lu(THF)]2(μ-H)2(μ-PhCH−CHPh) (63; Scheme 19), which contains the bridging dianionic ligand PhCH−CHPh bound with one of the Ln atoms in a η2-mode through the ethane-1,2-diyl carbon atoms and with the other metal atom in a η3-mode through one benzyl and ipso- and ortho-carbon atoms of one of the phenyl rings.

sch19

Scheme 19

The interaction of yttrium complex 55 with 1,4-diphenyl-1,3-butadiyne Ph−C≡C−C≡C−Ph also proceeded as the addition of Ln−H bonds across multiple C≡C bonds and gave rise to complex [(Amd)Y(THF)(μ-H)]2(μ-PhCH−C≡C−CHPh) (64; Scheme 19) which has a bridging 1,4-diphenyl-2-butyne-1,4-diyl dianionic moiety. Presumably, at the first step the terminal Y−H bond adds across one of the multiple C≡C bonds of 1,4-diphenyl-1,3-butadiyne Ph−C≡C−C≡C−Ph. At the second stage, the μ2-bridging hydride ligand adds across the remaining C≡C bond, which is accompanied by the formation of 64b and its following isomerization into 64 due to redistribution of multiple bonds (Scheme 19).

The interaction of yttrium complex 55 with 2,6-dimethylphenylisocyanide was also accompanied by the double hydrogenation of the triple C≡N bond. The product of an equimolar reaction of 55 and 2,6-Me2C6H3N≡C was binuclear dihydride complex [(Amd)Y(μ-H)]2(μ-ArN−CH2) (65, Ar = C6H3Me2-2,6; Scheme 19), which has a bridging dianionic methylene-amide moiety (−N(Ar)CH2−) along with two μ2-bridging hydride ligands.

2.3. Polyhydride complexes stabilized by tridentate ligands

Tris(pyrazolyl)borate ligands (TpR,R') found wide application in coordination and organometallic chemistry of rare-earth metals [50] and enabled the synthesis of a series of principally new classes of compounds, such as terminal dimethyl complexes [51] and hydride derivatives of low-valent lanthanides (Yb(II)) [52]. One of the apparent advantages of tris(pyrazolyl)borate ligands is an opportunity of their steric modification via variation of substituents at the third position of pyrazole rings, which allows one to change the steric properties of TpR,R ligands over a wide range. The tris(pyrazolyl)borate ligands of variable bulkiness provided a large series of dihydride rare-earth metal complexes in the form of tri-, tetra- and even hexanuclear clusters.

The hydrogenolysis of bis(alkyl) complexes (TpR2)Ln(CH2SiMe3)2(THF) (TpR,R' = TpH2, TpMe2, TpiPr2) with molecular hydrogen (75 bar, 20 °С) leads to the corresponding dihydride derivatives [(TpR2)Ln(μ-H)2]n (Scheme 20). Cheng et al. established that in the case of the least bulky ligand TpH2, which does not contain substituents in the pyrazole rings, the dihydride derivatives represent hexanuclear clusters [(TpH2)Ln(μ-H)2]6 (Ln = Y (66), Yb (67), Lu (68)) [53, 54]. In the case of bis(alkyl) complexes bearing dimethyl-substituted tris(pyrazolyl)borate ligands (TpMe2), the nuclearity of the resulting clusters reduced, and the dihydride complexes were isolated as tetramers [(TpMe2)LnH2]4 (Ln = Y (69), Sm (70), Yb (71), Lu (72), Nd (73); Scheme 20). A further increase in the tris(pyrazolyl)borate ligand bulkiness owing to the introduction of isopropyl groups (TpiPr2) enabled the synthesis of trinuclear complexes [(TpiPr2)LnH2]3 (Ln = Y (74), Lu (75); Scheme 20) [55]. The nature of the solvent in use also affected the nuclearity of the resulting derivatives. Thus, in the case of (TpMe2)Y(CH2SiMe3)2(THF) the hydrogenolysis in toluene led to tetramer 69, whereas THF as a solvent gave rise to trinuclear hexahydride complex [(TpMe2)YH2]3(THF)2-3 (74; Scheme 20), in which the yttrium atoms are additionally coordinated by the THF molecules [53, 54].

sch20

Scheme 20

It should be noted that, in the case of tris(pyrazolyl)methanide analog (Tpm)Lu(CH2SiMe3)2(THF) (Tpm = tris(3,5-dimethylpyrazolyl)methanide), the corresponding polyhydride derivative was not obtained. The interaction of bis(alkyl) complex (Tpm)Lu(CH2SiMe3)2(THF) with PhSiH3 afforded an unexpected product that represents binuclear complex [(Tpm)Lu]2(μ-C3HN2Me2)2(μ-CHSiMe3) (77, Scheme 21), in which two (Tpm)Lu moieties are connected by bridging pyrazolyl anions and one bridging dianionic methylidene ligand (μ-CHSiMe3)2− [56]. The analysis of liquid reaction products revealed the presence of PhSiH2CH2SiMe3 in the reaction mixture, which resulted from the Lu−C bond σ-metathesis under the action of phenylsilane. This fact confirmed the formation of hydride particles (Tpm)LnH, which are likely to be unstable and undergo decomposition due to elimination of the pyrazole rings from the ligand and activation of the Csp3−H bond of the alkyl group.

sch21

Scheme 21

The molecular structures of rare-earth metal polyhydride complexes 66–73, 75, and 76, stabilized by tris(pyrazolyl)borate ligands, were assigned by XRD. The Ln−Ln and Ln−H distances are listed in Table 4. In the case of hexanuclear complexes 66–68 [(TpH2)Ln(μ-H)2]6 (Ln = Y, Yb, Lu), the rare-earth metal ions form octahedra. In yttrium compound 66, the distances between the metal ions and the equatorial plane comprise 3.287(3)–3.742(3) Å and 5.073(3) Å between the metal ions in the apical positions. Six metal ions in the Ln6 octahedron are bound with each other through twelve bridging hydride ligands. Eight hydride ligands are μ3-bridging and are located above faces of the Ln6 octahedron at the distances of 0.91–0.96 Å. Three hydride ligands are μ2-bridging and are located above edges of the octahedron. The center of the Ln6 moiety is occupied by one μ6-bridging hydride ligand. In general, the Ln−(μ-H) distances observed in hexanuclear complexes [(TpH2)Ln(μ-H)2]6 are defined by the ionic radius of the metal and reduce in the following series: Y, Yb, and Lu (according to a reduction in their ionic radii), as well as the degree of association of the hydride ligands (Ln−μ2-H < Ln−μ3-H < Ln−μ6-H).

Table 4. Ln−Ln and Ln−H distances in complexes [(TpR2)Ln(μ-H)2]n (Å)

Trinuclear complexes

Ln−Ln

Ln−μ2-H

Ln−μ3-H

Ln−μ4-H

Ln−μ6-H

[(TpiPr2)Y(μ-H)2]3 (75)

3.342(3)–3.684(3)

2.01–2.29

2.11–2.19

−

−

[(TpiPr2)Lu(μ-H)2]3 (76)

3.250(3)–3.575(3)

2.00–2.25

2.05–2.16

−

−

Tetranuclear complexes

Ln−Ln

Ln−μ2-H

Ln−μ3-H

Ln−μ4-H

Ln−μ6-H

[(TpMe2)Y(μ-H)2]4 (69)

3.533(3)–3.711(3)

1.95–2.36

2.18–2.49

2.18–2.28

−

[(TpMe2)Nd(μ-H)2]4 (73)

3.713(3)–3.948(4)

2.18–2.47

2.26–2.60

2.24–2.40

−

[(TpMe2)Sm(μ-H)2]4 (70)

3.661(4)–3.893(5)

2.12–2.38

2.32–2.47

2.16–2.36

−

[(TpMe2)Yb(μ-H)2]4 (71)

3.456(4)–3.620(4)

1.97–2.13

2.16–2.33

2.10–2.21

−

[(TpMe2)Lu(μ-H)2]4 (72)

3.433(3)–3.593(3)

1.87–2.15

2.18–2.38

2.06–2.27

−

Hexanuclear complexes

Ln−Ln

Ln−μ2-H

Ln−μ3-H

Ln−μ4-H

Ln−μ6-H

[(TpH2)Y(μ-H)2]6 (66)

3.287(3)–3.742(3)

2.06

2.23–2.34

−

2.55

[(TpH2)Yb(μ-H)2]6 (67)

3.240(4)–3.636(4)

1.99

2.18–2.36

−

2.49

[(TpH2)Lu(μ-H)2]6 (68)

3.201(3)–3.641(3)

2.08

2.15–2.28

−

2.48

The structures of tetranuclear octahydride derivatives 69–72 [(TpMe2)LnH2]4 (Ln = Y, Sm, Yb, Lu) were analogous to those of cyclopentadienyl Sc and Y compounds [Cp'Ln(μ-H)2]4, complexes 8 and 3, respectively. Four metal centers in complexes 69–72 are bound by one μ4-, one μ3- and six μ2-bridging hydride ligands. Analogous Nd complex 73 features, along with a μ4-bridging hydride ligand located in the center of a Nd4 tetrahedron, two μ3- and five μ2-bridging hydride atoms.

In trinuclear complexes 75 and 76 [(TpiPr2)Ln(μ-H)2]3 (Ln = Y, Lu), two hydride ligands are located in the same plane as three Ln ions. One of them serves as a μ3-bridging ligand, while the second one adopts a μ2-bridging mode. The remaining four hydride ligands are bound with the Ln ions in a μ2-mode and are pairwise located on either side of the Ln3 plane.

The use of bulky tridentate phosphazene ligands (NNN = (2,6-iPr2C6H3N=PPh2)2N−) as a stabilizing environment afforded the rare examples of dimeric dihydride complexes of rare-earth metals [57]. The interaction of bis(alkyl) derivatives (NNN)Ln(CH2SiMe3)2 (Ln = Y, Lu) with PhSiH3 (2.5 eq.) or H2 (1 bar) in toluene led to the formation of the corresponding dimeric tetrahydride complexes [(NNN)Y(μ-H)2]2 (Ln = Y (78), Lu (79); Scheme 22). Rong et al. noted that the interaction of the bis(alkyl) derivative of yttrium with PhSiH3 or H2 readily proceeded at room temperature in toluene. A lutetium analog, (NNN)Lu(CH2SiMe3)2, appeared to be less reactive: its reactions both with PhSiH3 and H2 required heating to 60 °C. It was also detected that dissolution of yttrium complex 78 in THF initiates the intramolecular activation of the ortho-Csp2−H bond of one of the phenyl rings at the phosphorus atoms, resulting in binuclear aryl-hydride complex [(NNN)Y(THF)(μ-H)3Y(NNN−H)(THF)]2 (80, Scheme 22). Two metal centers in 80 are bound with each other by three μ2-bridging hydride ligands. One of the yttrium ions inside the dimer is coordinated by the initial phosphazene ligand, whereas the second one is connected with the new dianionic ligand (NNN−H) by one covalent and one coordination Y−N bonds and one covalent Y−CPh bond (2.513(7) Å).

sch22

Scheme 22

In the crystalline state, two rare-earth metal ions in complexes 78 and 79 are bound with each other by four μ2-bridging hydride ligands. The Ln−Ln distances in 78 and 79 compose 3.219(3) and 3.104(3) Å, respectively; the Ln−(μ2-H) distances fall within 2.02–2.38 Å (78) and 2.06–2.11 Å (79) (Table 5). On passing to aryl-hydride derivative 80, the Y−Y distance significantly increases (3.368(2) Å) compared to that in initial dihydride 78, whereas the Y−(μ2-H) bond distances in both complexes appear to be comparable (2.06–2.30 Å for 80 and 2.02–2.38 Å for 78).

Table 5. Ln−Ln and Ln−H distances in polyhydride complexes stabilized by tridentate pincer ligands (Å)

Binuclear complexes

Ln−Ln

Ln−μ2-H

Ln−μ3-H

[(NNN)Y(μ-H)2]2 (78)

3.219(3)

2.02–2.38 

−

[(NNN)Y(THF)(μ-H)3Y(NNN−H)(THF)]2 (80)

3.368(2)

2.06–2.30

−

[(NNN)Lu(μ-H)2]2 (79)

3.104(3)

2.06–2.11

−

[{(PNP)Y}2(μ-H)3][BPh4] (86)

3.330(3)

2.08–2.21

−

[{(PNP)Lu}2(μ-H)3][BPh4] (87)

3.217(3)

2.02–2.54

−

Trinuclear complexes

Ln−Ln

Ln−μ2-H

Ln−μ3-H

[(PNP)Sc(μ-H)2]3 (81)

2.848(4)–3.217(4)

1.90–2.06

1.94–2.24

[(PNP)Y(μ-H)2]3 (82)

3.165(5)–3.468(5)

2.07–2.27

2.06–2.37

[(PNP)Lu(μ-H)2]3 (83)

3.091(4)–3.407(4)

2.01–2.17

2.08–2.42

[{(PNP)Y}3(μ-H)5][BPh4] (84)

3.441(3)–3.481(3)

2.04–2.11

2.30–2.43

[{(PNP)Lu}3(μ-H)5][BPh4] (85)

3.359(3)–3.369(3)

1.90–2.27

2.11–2.40

{[(CzPziPr)Lu]2[(CzPziPr−H)Lu](µ-H)5} (88)

3.341(3)–3.367(3)

1.91–2.14

1.86–2.32

The tridentate ligands that enable the synthesis of dihydride complexes of rare-earth metals include pincer ligands based on ortho-disubstituted diphenylamine and 1,8-disubstituted carbazole.

The hydrogenolysis of bis(alkyl) complexes (PNP)Ln(CH2SiMe3)2 (PNP = 2-iPr2P-4-Me-C6H3)2N, bis(2-(diisopropylphosphino)-4-methylphenyl)amide, Ln = Sc, Y, Lu) in toluene with molecular H2 (10 bar) gave rise to the corresponding trinuclear hexahydride complexes [(PNP)Ln(μ-H)2]3 (Ln = Sc (81), Y (82), Lu (83); Scheme 23) [58, 59]. The subsequent protonolysis of dihydride complexes of yttrium (82) and lutetium (83) with the Broensted acid ([HNEt3][BPh4]) was accompanied by the cleavage of one of the hydride ligands and led to the formation of the corresponding trinuclear cationic polyhydride derivatives [{(PNP)Ln}3(μ-H)5][BPh4] (Ln = Y (84), Lu (85); Scheme 23). Furthermore, the hydrogenolysis of bis(alkyl) complexes (PNP)Ln(CH2SiMe3)2 (Ln = Y, Lu) in the presence of cationizing agent [HNEt3][BPh4] (0.5 eq.) in THF afforded binuclear cationic trihydride complexes [{(PNP)Ln}2(μ-H)3][BPh4] (Ln = Y (86), Lu (87); Scheme 23) [58].

sch23

Scheme 23

According to the results of XRD analysis, three (PNP)Ln moieties in complexes [(PNP)Ln(μ-H)2]3 81–83 are bound by six bridging hydride ligands. Two out of six hydride ligands are µ3-bridging and are located above and below the Ln3 plane. The remaining four hydrides serve as µ2-bridging ligands and are pairwise bound only with two metal centers. Two μ2-bridging hydride ligands lie in the Ln3 plane, whereas two remaining ligands are located above and below the Ln3 plane. The ionic radius of the central metal atom affects the binding mode of the substituted diphenylamide ligand (PNP). Thus, in the case of Y (82) and Lu (83) complexes, all three ligands are tridentate and are bound with the rare-earth metal atoms by one covalent Ln−N and two coordination Ln−P bonds. In the case of the complex of scandium (81), which ionic radius is shorter than those of yttrium and lutetium, in one of three (PNP)Ln moieties the ligand is bidentate and is connected with the Sc atom by one covalent Sc−N and one coordination Sc−P bond. The second phosphorus atom of the (PNP)-ligand does not participate in binding with the metal center.

In the case of cationic trinuclear polyhydride complexes [{(PNP)Ln}3(μ-H)5][BPh4] 84 and 85, three metal centers are bound by five bridging hydride ligands. Analogously to the earlier reported neutral alkyl-hydride [(ApLn)3(μ-H)5(CH2SiMe3)(THF)2] and cationic hydride [(Ap*Y)3(μ-H)5(THF)3][B(C6F5)4] complexes with amidopyridinate ligands, three μ2-bridging hydrides are arranged in the Ln3 plane, and two μ3-bridging ligands are located above and below this plane. On passing to dimeric cationic hydride derivatives [{(PNP)Ln}2(μ-H)3][BPh4] 86 and 87, the binding of two metal centers is accomplished by three μ2-bridging hydrides.

The investigation of the neutral trinuclear polyhydride derivatives and their cationic analogs in solution by NMR spectroscopy revealed different behaviors of the hydride ligands inside the Ln3(μ-H)6 moieties. In the case of neutral complexes [(PNP)Ln(μ-H)2]3 81–83, all six hydride ligands appear as a single signal—a quadruplet at 6.08 ppm for yttrium complex 82 (1JYH = 16.9 Hz) and a singlet at 9.70 ppm for lutetium compound 83 (C6D6, 293 K). This fact testifies the presence of an exchange process between the hydride ligands in the cluster, which is rapid within the NMR timescale. In cationic polyhydride complexes 84 and 85 [{(PNP)Ln}3(μ-H)5][BPh4], the exchange between μ2- and μ3-bridging hydride ligands at room temperature was not observed, and the hydride atoms became nonequivalent. Thus, in the case of yttrium complex, μ2-hydride ligands appeared as a triplet at 6.52 ppm (1JYH = 31.7 Hz) and μ3-bridging ligands appeared as a broadened singlet at 5.53 ppm. For dimeric cationic yttrium complex 86, the signals of three μ2-bridging hydrides appeared as a single triplet at 5.72 ppm (1JYH = 23.1 Hz), which unambiguously testifies the retention of its dimeric structure even in a coordinating solvent (THF-d8).

Trinuclear alkyl-hydride complex {[(CzPziPr)Lu]2[(CzPziPr−H)Lu](µ-H)5} (88; Scheme 24) [60], which has a structure analogous to those of trinuclear alkyl-hydride clusters with cyclopentadienyl Cp (5) and amidopyridinate Ap (47–53) ligands, was obtained upon treatment of lutetium bis(alkyl) complex (CzPziPr)Lu(CH2SiMe3)2 stabilized by the pincer ligand based on 1,8-disubstituted carbazole (CzPziPr = 1,8-bis(3-isoppropyl-1H-pyrazolyl)-3,6-dimethyl-9H-carbazolide) with molecular H2 (4 bar, 50 °C, 24 h). Complex 88 represents a trinuclear cluster, in which two Lu atoms are coordinated by the monoanionic carbazolyl ligands, whereas the third one is connected with the dianionic ligand via the covalent Lu−N and Lu−CPyrazol bonds. The formation of the covalent Lu−CPyrazol bond results from the activation of the Сsp2−Н bond of one of the pyrazole rings.

sch24

Scheme 24

According to the XRD data, the Lu−Lu, Lu−(μ2-H), and Lu−(μ3-H) distances in trinuclear moiety Lu3(μ-H)5 (Table 5) are close to those in alkyl-hydride and cationic polyhydride derivatives with Cp (5) and Ap (47–53) ligands. Three Lu atoms are pairwise bound with each other by three μ2-bridging hydrides (Lu−(μ2-H) 1.91–2.14 Å) which are arranged in a Lu3 plane. Each metal center is connected with two μ3-H ligands Lu−(μ3-H) 1.86–2.32 Å) located above and below the Lu3 plane. In THF-d8, five hydride ligands of complex {[(CzPziPr)Lu]2[(CzPziPr−H)Lu](µ-H)5} appear as a sole broadened singlet at 10.3 ppm, which indicates the existence of a rapid exchange between the hydride ligands inside the Lu3(μ-H)5 trinuclear moiety.

2.4. Polyhydride complexes stabilized by tetradentate ligands

Tetradentate macrocyclic ligands based on tetraazacyclododecane (Me2TACD)H2 (1,7-dimethyl-1,4,7,10-tetraazocyclododecane), (Me3TACD)H (1,4,7-trimethyl-1,4,7,10-tetraazocyclododecane), and (Me4TACD) (1,4,7,10-tetramethyl-1,4,7,10-tetraazocyclododecane), being the nitrogen-containing analogs of crown ethers, have been successfully used in organometallic chemistry of rare-earth metals. The variation of the number of methyl groups at the nitrogen atoms in a twelve-membered macrocycle enabled their application as dianionic ((Me2TACD)2−), monoanionic ((Me3TACD)−) and neutral ligands (Me4TACD) [61–63].

The reactions of bis(alkyl) complexes (Me3TACD)Ln(CH2SiMe3)2 (Ln = Y, Ho, Lu) with PhSiH3 smoothly proceeded in pentane at room temperature, resulting in trinuclear hexahydride clusters [(Me3TACD)Ln(μ-H)2]3 (Ln = Y (89), Ho (90), Lu (91); Scheme 25) [62]. Arndt et al. [63] showed that cationic alkyl derivatives of yttrium [(Me3TACD)Y(CH2SiMe3)(THF)][BR4] (R = C6H3Cl2-3,5, C6H3(CF3)2-3,5), generated in situ upon treatment of bis(alkyl) complex (Me3TACD)Y(CH2SiMe3)2 with the Broensted acid ([HNR3][BR4]), used as the starting compounds in reactions with PhSiH3 (1 eq.) or H2 (1 bar) led to the formation of dimeric dicationic hydride complexes [(Me3TACD)Y(THF)(μ-H)]2[BR4]2 (R = C6H3Cl2-3,5 (92), C6H3(CF3)2-3,5 (93); Scheme 25).

sch25

Scheme 25

In complex 89 [(Me3TACD)Y(μ-H)2]3, all six hydrides serve as μ2-bridging ligands and are located on either side and beyond a Y3 moiety (Table 6). It should be noted that complex 89 is the only one example of trinuclear polyhydride rare-earth complexes, in which the metal centers are bound only by the μ2-bridging hydride atoms, whereas in the other cases trinuclear structures Ln3Hn feature one or two μ3-bridging hydride ligands located inside the Ln3 moiety (Table 6). In solution, all six hydride ligands in complexes 89 and 91 are equivalent and appear at 6.37 ppm (a multiplet for yttrium compound 89) and 9.81 ppm (a singlet for lutetium compound 91). The investigation of a mixture of complexes 89 and 91 in C6D6 did not reveal any sign of the exchange of metal centers, resulting in heterobimetallic species Y2Lu and YLu2, which suggests the preservation of trinuclear structures in solution.

Table 6. Ln−Ln and Ln−H distances in the polyhydride complexes stabilized by tetradentate ligands (Å)

Binuclear complexes

Ln−Ln

Ln−μ2-H

Ln−μ3-H

Ln−μ4-H

[(Me3TACD)Nd(η3-C3H5)(μ-H)]2 (97)

3.784(1)

2.23–2.29

−

−

[(Me3TACD)Sm(η3-C3H5)(μ-H)]2 (98)

3.734(1)

2.13–2.30

−

−

[(Me3TACD)Y(THF)(μ-H)]2++[B(C6H3(CF3)2)4]−2 (93)

3.620(1)

2.17–2.24

−

−

[(Me4TACD)Y(μ-H)2]2++[B(C6H3(CF3)2-3,5)4]−2 (99)

3.062(1)

2.15–2.23

−

−

[(Me4TACD)Lu(μ-H)2]2++[B(C6H3(CF3)2-3,5)4]−2 (100)

2.927(1)

1.95–2.21

−

−

[(Me4TACD)Lu(μ-H)2]2++[B(C6F5)4]−2 (101)

2.944(1)

2.00–2.21

−

−

[(Me4TACD)Lu(μ-H)3Lu(CH2TACDMe3)]++[B(C6H3(CF3)2-3,5)4]−2 (102)

3.175(1)

2.16–2.17

−

−

[(Me6TREN)Y(μ-H)3Y(CH2TACDMe3)]++[B(C6H3(CF3)2-3,5)4]−2 (103)

3.343(3)

1.96–2.18

−

−

[(Me6TREN)Gd(μ-H)3Gd(CH2TACDMe3)]++[B(C6H3(CF3)2-3,5)4]−2 (104)

3.414(4)

1.92–2.40

−

−

[(Me6TREN)Dy(μ-H)3Dy(CH2TACDMe3)]++[B(C6H3(CF3)2-3,5)4]−2 (105)

3.338(2)

1.84–2.25

−

−

[(Me6TREN)Lu(μ-H)3Lu(CH2TACDMe3)]++[B(C6H3(CF3)2-3,5)4]−2 (106)

3.278(3)

1.83–2.32

−

−

[(Me5TRENCH2)Lu(μ-H)]2 (107)

3.515(2)

2.01–2.12

−

−

Trinuclear complexes

Ln−Ln

Ln−μ2-H

Ln−μ3-H

Ln−μ4-H

[(Me3TACD)Y(μ-H)2]3 (89)

3.516(1)

2.04–2.35

-

 

Tetranuclear complexes

Ln−Ln

Ln−μ2-H

Ln−μ3-H

Ln−μ4-H

[(Me3TACD)La(μ-H)2]4 (94)

3.892(1)–4.169(1)

2.27–2.52

2.61–2.70

2.41–2.53

[(Me3TACD)Ce(μ-H)2]4 (95)

3.833(1)–4.228(1)

2.08–2.61

2.56–2.65

2.27–2.74

[(Me3TACD)Pr(μ-H)2]4 (96)

3.797(1)–4.177(1)

2.09–2.60

2.39–2.63

2.25–2.47

 Subsequently, the counterparts of these complexes with the metals with larger ionic radii were obtained. In these cases, the allyl derivatives were used as precursors, since they are more stable than trimethylsilyl complexes. Treatment of bis(allyl) complexes (Me3TACD)Ln(η3-C3H5)2 (Ln = La, Ce, Pr) with molecular H2 (5 bar, 50 °C, 72 h) or PhSiH3 (2.2 eq., 20 °C, 1 h) enabled the synthesis of the corresponding polyhydrides [(Me3TACD)Ln(μ-H)2]4 (Ln = La (94), Ce (95), Pr (96); Scheme 26) [64, 65]. In the case of neodymium and samarium, the authors failed to obtain the corresponding polyhydrides: the interaction of bis(allyl) derivatives (Me3TACD)Ln(η3-C3H5)2(Ln = Nd, Sm) with an excess of PhSiH3 (2.2 eq.) occurred only by one allyl group, giving rise to dimeric heteroleptic hydride-allyl complexes [(Me3TACD)Ln(η3-C3H5)(μ-H)]2 (Ln = Nd (97), Sm (98); Scheme 26) [65].

sch26

Scheme 26

A transition to the metals with the larger ionic radii reflected on the nuclearity of the resulting polyhydride derivatives and, according to the XRD data, polyhydride complexes 94–96 [(Me3TACD)Ln(μ-H)2]4 (Ln = La, Ce and Pr) represent tetranuclear octahydride clusters. Four metal centers form a distorted Ln4 tetrahedron with the Ln−Ln distances of 3.892(1)–4.169(1) Å (for La), 3.833(1)–4.228(1) Å (for Ce), and 3.797(1)–4.177(1) Å (for Pr). The binding of four metal centers in complexes 94–96 was accomplished by eight bridging hydride ligands. Six hydrides act as μ2-bridging ligands and are located above edges of an Ln4 tetrahedron (Table 6). Two additional hydride ligands adopt μ3- and μ4-bridging modes and are located either above one of the Ln4 tetrahedron faces or inside it.

Ohashi et al. [62] revealed that complex 89 [(Me3TACD)Y(μ-H)2]3 can serve as a catalyst for intermolecular hydrosilylation of olefins. The hydrosilylation of 1-hexene with phenylsilane in the presence of 2.5 mol % of 89 (C6D6, 60 °C, 7 h) afforded a mixture of the products of single ((n-Hex)PhSiH2, 85%) and double ((n-Hex)2PhSiH, 10%) alkylation of the SiH bonds in phenylsilane. The hydrosilylation of 1,5-hexadiene with phenylsilane catalyzed by complex 89 (C6D6, 60 °C, 17 h) proceeded with a quantitative conversion of the substrates and led to 1,6-bis(phenylsilyl)hexane (90–95%) and phenylsilacycloheptane (5–10%). It was shown that complex 89 catalyzes the hydrosilylation of multiple C=O bonds of furfal with phenylsilane and provides a quantitative conversion of 40 eq. of the aldehyde already in 7 min, affording a mixture of PhSiH(OCH2-C4H3O)2 and PhSi(OCH2-C4H3O)3 (18:82), being the products of insertion of the aldehyde groups in two or three SiH bonds of phenylsilane, respectively [64].

A series of cationic polyhydride derivatives of rare-earth metals [(Me4TACD)Ln(μ-H)2][BR4]2 (R = C6H3(CF3)3-3,5, Ln = Y (99), Lu (100); R = C6F5, Ln = Lu (101); Scheme 27) and [(Me4TACD)Lu(μ-H)3Lu(CH2TACDMe3)][BR4]2 (R = C6H3(CF3)2-3,5; 102; Scheme 27) [63, 66] were obtained using cationic bis(alkyl) complexes [(Me4TACD)Ln(CH2SiMe3)2][BR4] bearing neutral (Me4TACD) ligand as starting compounds. Depending on the reagents (H2 vs PhSiH3) used for the Ln−C σ-bond metathesis, the reactions afforded either dimeric tetrahydride complexes 99–101, which contain two metal centers coordinated by the neutral ligand (Me4TACD), or dimeric trihydride complex 102, in which one of the Lu atoms is bound with the monoanionic ligand (CH2TACDMe3)−. The latter resulted from the activation of the Csp3−H bond of one of the methyl groups of the NMe moieties. Fegler et al. [66] showed that lutetium complexes 100 and 102, bearing neutral (Me4TACD) and monoanionic (CH2TACDMe3)− ligands, can be reversibly converted to each other under the action of H2.

sch27

Scheme 27

A series of binuclear dicationic trihydride complexes of rare-earth metals [(Me6TREN)Ln(μ-H)3Ln(CH2TACDMe3)][BR4]2 (Ar = C6H3(CF3)2-3,5; Ln = Y (103), Gd (104), Dy (105), Lu (106); Scheme 28), analogous to dicationic Lu complex 102, were obtained upon hydrogenolysis of the corresponding bis(alkyl) complexes (Me6TREN)Ln(CH2SiMe3)2, bearing neutral tris(dimethylaminoethyl)amine (N(CH2CH2NMe2)3 = Me6TREN) as a stabilizing ligand, with molecular H2 [67, 68]. The interaction of bis(alkyl) complex (Me6TREN)Ln(CH2SiMe3)2 with H2 was accompanied by the activation of the Сsp3−H bond of one of the methyl groups of the NMe2 moieties, giving rise to binuclear complexes 103–106 in which one of the metal centers is coordinated by the neutral ligand (Me6TREN), whereas the second one is bound with the monoanionic ligand (Me5TRENCH2)−. Me6TREN in complexes 103–106 serve as a tetradentate ligand and is bound with the rare-earth metal atoms by four Ln−N coordination bonds. The anionic ligand (Me5TRENCH2)− affords, along with the coordination Ln−N bonds, the covalent Ln−C bond and becomes pentadentate.

In contrast, the hydrogenolysis of cationic alkyl lutetium complex (Me5TRENCH2)Lu(CH2SiMe3)][BR4], initially bearing the monoanionic ligand (Me5TRENCH2)−, with molecular H2 afforded, along with the hydrogenolysis of the Lu−CH2SiMe3 bond, the selective hydrogenation of the Lu−C bond in one of Lu(CH2TACDMe3) moieties, producing unsymmetrical binuclear trihydride complex 106. Treatment of [(Me5TRENCH2)Lu(CH2SiMe3)][BR4] with phenylsilane did not lead to the cleavage of the covalent Lu−C bond in Lu(CH2TACDMe3) moieties, and the corresponding dimeric dihydride complex {[(Me5TRENCH2)Lu(μ-H)]2}{BR4}2 was obtained (107; Scheme 28).

sch28

Scheme 28

In order to synthesize polyhydride yttrium complex stabilized by tetradentate β-diketiminate ligands with an additional −(CH2)2NMe(CH2)NMe2 donor moiety at one of the nitrogen atoms, Zhou et al. [69] studied the reaction of dimethyl complex [MeC(NC6H3iPr2-2,6)CHC(Me)N(CH2)2N(Me)(CH2)2NMe2]YMe2 with phenylsilane PhSiH3 (2.5 eq.). However, the authors failed to obtain the corresponding dihydride derivative. XRD studies showed that the reaction product was binuclear complex 108 (Scheme 29), in which two yttrium atoms are coordinated by new dianionic tetradentate ligands, resulting from the initial β- diketiminate owing to the addition of the Y−H bond across the C=N bond. Two metal centers in complex 108 are bound by two μ2-bridging hydride ligands (2.12(3)–2.24(3) Å) and one bridging PhSiH3 molecule. The binding of the latter with dimeric moiety [LY(μ-H)YL]2 occurs owing to the coordination of one of the hydride atoms of PhSiH3 molecule to two metal centers (Y−HSi 2.09(2) Å) as well as owing to the interaction of the silicon atom with two nitrogen atoms of the adjacent ligands (Si···N 1.916(2) and 1.925(2) Å).

sch29

Scheme 29

3. Conclusions

The achievements of the last 15 years in the chemistry of polyhydride complexes of rare-earth metals show that these compounds are no longer exotic. This breakthrough appeared to be possible owing to the enhanced understanding of the structure–stability–reactivity relationships and the synthesis of a multitude of new ligand systems that are able to stabilize these highly reactive derivatives. The progress in this area offers major opportunities to control the reactivity of polyhydride complexes of rare-earth metals via variation of the ligand systems and design of the metal center coordination sphere. The promising routes for the development of the chemistry of polyhydride complexes of rare-earth metals are, undoubtedly, their application in catalysis, where the greatest progress can be expected in metal-catalyzed reactions of unsaturated substrates, hydrofunctionalization of alkenes and acetylenes, activation of CH bonds, and polymerization of olefins and dienes.

Acknowledgements

This work was performed within the state assignment.

References

  1. G. S. McGrady, G. Guilera, Chem. Soc. Rev., 2003, 32, 383–392. DOI: 10.1039/B207999M
  2. H. Suzuki, Eur. J. Inorg. Chem., 2002, 1009–1023. DOI: 10.1002/1099-0682(200205)2002:5<1009::AID-EJIC1009>3.0.CO;2-0
  3. F. Maseras, A. Lledόs, E. Clot, O. Eisenstein, Chem. Rev., 2000, 100, 601–636. DOI: 10.1021/cr980397d
  4. S. Sabo-Etienne, B. Chaudret, Chem. Rev., 1998, 98, 2077–2092. DOI: 10.1021/cr9601066
  5. Z. Lin, M. B. Hall, Coord. Chem. Rev., 1994, 135–136, 845–879. DOI: 10.1016/0010-8545(94)80084-7
  6. G. G. Hlatky, R. H. Crabtree, Coord. Chem. Rev., 1985, 65, 1–48. DOI: 10.1016/0010-8545(85)85020-7
  7. L. M. Venanzi, Coord. Chem. Rev., 1982, 43, 251–274. DOI: 10.1016/S0010-8545(00)82099-8
  8. J. Okuda, Dalton Trans., 2003, 2367–2378. DOI: 10.1039/B300818E
  9. S. Arndt, J. Okuda, Chem. Rev., 2002, 102, 1953–1976. DOI: 10.1021/cr010313s
  10. G. A. Molander, J. A. C. Romero, Chem. Rev., 2002, 102, 2161–2186. DOI: 10.1021/cr010291+
  11. Z. Hou, Y. Wakatsuki, Coord. Chem. Rev., 2002, 231, 1–22. DOI: 10.1016/S0010-8545(02)00111-X
  12. A. J. Hoskin, D. W. Stephan, Coord. Chem. Rev., 2002, 233–234, 107–129. DOI: 10.1016/S0010-8545(02)00030-9
  13. M. Ephritikhine, Chem. Rev., 1997, 97, 2193–2242. DOI: 10.1021/cr960366n
  14. H. Schumann, J. A. Meese-Marktscheffel, L. Esser, Chem. Rev., 1995, 95, 865–986. DOI: 10.1021/cr00036a004
  15. C. J. Schaverien, Adv. Organomet. Chem., 1994, 36, 283–362. DOI: 10.1016/S0065-3055(08)60393-7
  16. W. J. Evans, J. H. Meadows, A. L. Wayda, W. E. Hunter, J. L. Atwood, J. Am. Chem. Soc., 1982, 104, 2008–2014. DOI: 10.1021/ja00371a035
  17. G. Jeske, H. Lauke, H. Mauermann, P. N. Swepston, H. Schumann, T. J. Marks, J. Am. Chem. Soc., 1985, 107, 8091–8103. DOI: 10.1021/ja00312a050
  18. G. Jeske, L. E. Schock, P. N. Swepston, H. Schumann, T. J. Marks, J. Am. Chem. Soc., 1985, 107, 8103–8110. DOI: 10.1021/ja00312a051
  19. A. Z. Voskoboynikov, I. N. Parshina, A. K. Shestakova, K. P Butin, I. P. Beletskaya, L. G. Kuz'mina, J. A. K. Howard, Organometallics, 1997, 16, 4041–4055. DOI: 10.1021/om970363g
  20. M. Konkol, J. Okuda, Coord. Chem. Rev., 2008, 252, 1577–1591. DOI: 10.1016/j.ccr.2007.09.019
  21. W. Fegler, A. Venugopal, M. Kramer, J. Okuda, Angew. Chem., Int. Ed., 2015, 54, 1724–1736. DOI: 10.1002/anie.201406677
  22. A. A. Trifonov, D. M. Lyubov, Coord. Chem. Rev., 2017, 340, 10–61. DOI: 10.1016/j.ccr.2016.09.013
  23. J. Okuda, Coord. Chem. Rev., 2017, 340, 2–9. DOI: 10.1016/j.ccr.2016.09.009
  24. M. Yousufuddin, M. J. Gutmann, J. Baldamus, O. Tardif, Z. Hou, S. A. Mason, G. J. McIntyre, R. Bau, J. Am. Chem. Soc., 2008, 130, 3888–3891. DOI: 10.1021/ja076520c
  25. K. C. Hultzsch, P. Voth, T. P. Spaniol, J. Okuda, Z. Anorg. Allg. Chem., 2003, 629, 1272–1276. DOI: 10.1002/zaac.200300063
  26. Y. Luo, J. Baldamus, O. Tardif, Z. Hou, Organometallics, 2005, 24, 4362–4366. DOI: 10.1021/om0502382
  27. O. Tardif, M. Nishiura, Z. Hou, Organometallics, 2003, 22, 1171–1173. DOI: 10.1021/om021014b
  28. M. Nishiura, J. Baldamus, T. Shima, K. Mori, Z. Hou, Chem. Eur. J., 2011, 17, 5033–5044. DOI: 10.1002/chem.201002998
  29. D. Cui, M. Nishiura, Z. Hou, Macromolecules, 2005, 38, 4089–4095. DOI: 10.1021/ma050256v
  30. T. Shima, M. Nishiura, Z. Hou, Organometallics, 2011, 30, 2513–2524. DOI: 10.1021/om1012055
  31. X. Li, J. Baldamus, M. Nishiura, O. Tardif, Z. Hou, Angew. Chem., Int. Ed., 2006, 45, 8184–8188. DOI: 10.1002/anie.200603450
  32. Z. Hou, M. Nishiura, T. Shima, Eur. J. Inorg. Chem., 2007, 2535–2545. DOI: 10.1002/ejic.200700085
  33. M. Nishiura, Z. Hou, Bull. Chem. Soc. Jpn., 2010, 83, 595–608. DOI: 10.1246/bcsj.20100003
  34. M. Nishiura, Z. Hou, Nat. Chem., 2010, 2, 257–268. DOI: 10.1038/nchem.595
  35. T. Shima, Y. Luo, T. Stewart, R. Bau, G. J. McIntyre, S. A. Mason, Z. Hou, Nature Chem., 2011, 3, 814–820. DOI: 10.1038/nchem.1147
  36. D. Cui, O. Tardif, Z. Hou, J. Am. Chem. Soc., 2004, 126, 1312–1313. DOI: 10.1021/ja039324o
  37. O. Tardif, D. Hashizume, Z. Hou, J. Am. Chem. Soc., 2004, 126, 8080–8081. DOI: 10.1021/ja047889u
  38. T. Shima, Z. Hou, J. Am. Chem. Soc., 2006, 128, 8124–8125. DOI: 10.1021/ja062348l
  39. D. Cui, M. Nishiura, Z. Hou, Angew. Chem., Int. Ed., 2005, 44, 959–962. DOI: 10.1002/anie.200461939
  40. T. Shima, Z. Hou, Dalton Trans., 2010, 39, 6858–6863. DOI: 10.1039/c0dt00006j
  41. A. G. Avent, F. G. N. Cloke, B. R. Elvidge, P. B. Hitchcock, Dalton Trans., 2004, 1083–1096. DOI: 10.1039/b400149d
  42. A. A. Kisel', D. M. Lyubov, A. V. Cherkasov, A. A. Trifonov, Russ. J. Coord. Chem., 2019, 45, 266–272. DOI: 10.1134/S1070328419040043
  43. D. M. Lyubov, C. Döring, G. K. Fukin, A. V. Cherkasov, A. S. Shavyrin, R. Kempe, A. A. Trifonov, Organometallics, 2008, 27, 2905–2907. DOI: 10.1021/om800364b
  44. D. M. Lyubov, C. Döring, S. Yu. Ketkov, R. Kempe, A. A. Trifonov, Chem. Eur. J., 2011, 17, 3824–3826. DOI: 10.1002/chem.201003616
  45. D. M. Lyubov, A. V. Cherkasov, G. K. Fukin, S. Yu. Ketkov, A. S. Shavyrin, A. A. Trifonov, Dalton Trans., 2014, 43, 14450–14460. DOI: 10.1039/c4dt00806e
  46. D. M. Lyubov, V. Yu. Rad'kov, A. V. Cherkasov, G. K. Fukin, A. A. Trifonov, Russ. Chem. Bull., 2016, 65, 2594–2600. DOI: 10.1007/s11172-016-1623-5
  47. R. D. Shannon, Acta Crystallogr., Sect. A: Cryst. Phys., Diffr., Theor. Gen. Crystallogr., 1976, 32, 751–767. DOI: 10.1107/S0567739476001551
  48. J. Cheng, H. Wang, M. Nishiura, Z. Hou, Chem. Sci., 2012, 3, 2230–2233. DOI: 10.1039/c2sc20300f
  49. J. Cheng, Z. Hou, Chem. Commun., 2012, 48, 814–816. DOI: 10.1039/c1cc14949k
  50. N. Marques, A. Sella, J. Takats, Chem. Rev., 2002, 102, 2137−2160. DOI: 10.1021/cr010327y
  51. M. Zimmermann, R. Litlabø, K. W. Törnroos, R. Anwander, Organometallics, 2009, 28, 6646–6649. DOI: 10.1021/om900838h
  52. G. M. Ferrence, R. McDonald, J. Takats, Angew. Chem., Int. Ed., 1999, 38, 2233–2237. DOI: 10.1002/(sici)1521-3773(19990802)38:15<2233::aid-anie2233>3.0.co;2-x
  53. J. Cheng, K. Saliu, G. Y. Kiel, M. J. Ferguson, R. McDonald, J. Takats, Angew. Chem., Int. Ed., 2008, 47, 4910–4913. DOI: 10.1002/anie.200705977
  54. J. Cheng, K. Saliu, M. J. Ferguson, R. McDonald, J. Takats, J. Organomet. Chem., 2010, 695, 2696–2702. DOI: 10.1016/j.jorganchem.2010.08.020
  55. J. Cheng, M. F. Ferguson, J. Takats, J. Am. Chem. Soc., 2010, 132, 2–3. DOI: 10.1021/ja905679k
  56. T. Li, G. Zhang, J. Guo, S. Wang, X. Leng, Y. Chen, Organometallics, 2016, 35, 1565–1572. DOI: 10.1021/acs.organomet.6b00166
  57. W. Rong, D. He, M. Wang, Z. Mou, J. Cheng, C. Yao, S. Li, A. A. Trifonov, D. M. Lyubov, D. Cui, Chem. Commun., 2015, 51, 5063–5065. DOI: 10.1039/c4cc07262f
  58. J. Cheng, T. Shima, Z. Hou, Angew. Chem., Int. Ed., 2011, 50, 1857–1860. DOI: 10.1002/anie.201006812
  59. T. Kurogi, M. Kamitani, P. J. Carroll, D. J. Mindiola, Isr. J. Chem., 2017, 57, 999–1009. DOI: 10.1002/ijch.201700062
  60. K. R. D. Johnson, B. L. Kamenz, P. G. Hayes, Organometallics, 2014, 33, 3005–3011. DOI: 10.1021/om500226t
  61. P. Cui, T. P. Spaniol, L. Maron, J. Okuda, Chem. Commun., 2014, 50, 424–426. DOI: 10.1039/c3cc47805j
  62. M. Ohashi, M. Konkol, I. Del Rosal, R. Poteau, L. Maron, J. Okuda, J. Am. Chem. Soc., 2008, 130, 6920–6921. DOI: 10.1021/ja801771u
  63. S. Arndt, M. U. Kramer, W. Fegler, Y. Nakajima, I. Del Rosal, R. Poteau, T. P. Spaniol, L. Maron, J. Okuda, Organometallics, 2015, 34, 3739–3747. DOI: 10.1021/acs.organomet.5b00422
  64. E. Abinet, D. Martin, S. Standfuss, H. Kulinna, T. P. Spaniol, J. Okuda, Chem. Eur. J., 2011, 17, 15014–15026. DOI: 10.1002/chem.201102145
  65. D. Martin, J. Kleemann, E. Abinet, T. P. Spaniol, L. Maron, J. Okuda, Eur. J. Inorg. Chem., 2013, 3987–3992. DOI: 10.1002/ejic.201300167
  66. W. Fegler, A. Venugopal, T. P. Spaniol, L. Maron, J. Okuda, Angew. Chem., Int. Ed., 2013, 52, 7976–7980. DOI: 10.1002/anie.201303958
  67. A. Venugopal, W. Fegler, T. P. Spaniol, L. Maron, J. Okuda, J. Am. Chem. Soc., 2011, 133, 17574–17577. DOI: 10.1021/ja207180x
  68. A. Venugopal, F. Tuna, T. P. Spaniol, L. Ungur, L. F. Chibotaru, J. Okuda, R. A. Layfield, Chem. Commun., 2013, 49, 901–903. DOI: 10.1039/c2cc38036f
  69. J. Zhou, J. Chu, Y. Zhang, G. Yang, X. Leng, Y. Chen, Angew. Chem., Int. Ed., 2013, 52, 4243–4246. DOI: 10.1002/anie.201300638