Skip to main content
NCBI home page
ACS AuthorChoice logoLink to ACS AuthorChoice
. 2025 Apr 11;147(16):13871–13884. doi: 10.1021/jacs.5c02178

Tripodal Silanolate Ligands Expand [MoX3] Chemistry Beyond Its Traditional Borders

Daniel Rütter 1, Nils Nöthling 1, Markus Leutzsch 1, Alexander A Auer 1, Alois Fürstner 1,*
PMCID: PMC12022994  PMID: 40214616

Abstract

graphic file with name ja5c02178_0019.jpg

Homodimeric complexes [X3Mo≡MoX3] are commonplace, but in no case is the corresponding monomeric [MoX3] species known; conversely, none of the very rare monomeric complexes [MoX3] has the respective homodimeric analogue. This mutual exclusivity ends with the present study; on top, an entirely unprecedented class of heterodimers of type [X3Mo≡MoY3] is reported. Key to success was the use of tripodal silanolates as ancillary ligands; the fence formed by properly chosen peripheral substituents shields the sensitive Mo(+3) center; homodimerization of the resulting [MoX3] complexes is then kinetically strongly disfavored, though possible. The monomers are able to cleave N2O and convert gem-dihalides into metal alkylidynes; they exist in different binding modes, in which the basal phenyl ring of the ligand backbone is either completely unengaged with the central metal or tightly bound to it, depending on whether the ligand sphere is complemented by solvent molecules or not. If the latter are sufficiently labile, a surprisingly facile heterodimerization of the d3 electron fragments will ensue; the resulting products [X3Mo≡MoY3] incorporate the intact Cummins complex [(tBu)(Ar)N]3Mo (Ar = 3,5-dimethylphenyl) as one of their constituents, which is famous for not engaging in metal–metal triple bonding otherwise. Heterodimerization was also observed with simple tert-butoxide ligands. The new type of heterodimers features unusually long yet robust Mo≡Mo bonds, which are notably polarized according to DFT. However, there is no direct correlation between the extreme Mo≡Mo bond lengths and the strikingly deshielded 95Mo NMR signals, since ligand-based orbitals can also markedly affect the shielding tensor.

Introduction

The chemistry of Mo(+3) is dominated by the massive bias of [MoX3] units to dimerize. It is the stability of the resulting σ2π4 metal–metal triple bond between the two d3 electron fragments that provides a formidable thermodynamic driving force for the formation of homobimetallic complexes of the general type [X3Mo≡MoX3] (X = alkyl, (pseudo)halide, alkoxy, silyloxy, dialkylamido, etc.). Ever since the first such complex had been described by Wilkinson and co-workers in 1971,1 the field flourished, leading to a plethora of homoleptic representatives; in no case has the corresponding monomer [MoX3] ever been detected.2,3 Some of the dimers were transformed into heteroleptic variants by partial ligand exchange via protonolysis or salt metathesis, whereby “symmetrical” ligand distributions ([X2YMo≡MoYX2], “1,2-pattern”)4,5 were far more commonly attained than “unsymmetrical” ones ([XY2Mo≡MoX3], “1,1-pattern”).6−9 Such isomers tend not to interconvert, suggesting that the barriers for group transfer between the two neighboring molybdenum centers under thermal conditions are high in most cases.7,10 Heteroleptic examples of the type [X2YMo≡MoX3] or [XY2Mo≡MoX3] are rare,11−16 and heterodimers of the general constitution [X3Mo≡MoY3] seem to be unknown.17

The challenge of obtaining a kinetically stable monomeric Mo(+3) complex was first met in 1995 by Cummins and co-workers, who disclosed the preparation, structure and stunning reactivity of the trigonal-planar complex [(tBu)(Ar)N]3Mo (Ar = 3,5-dimethylphenyl) (1) (Figure 1).18−22 This particular trisamidomolybdenum species and its siblings are capable of activating numerous small molecules under notably mild conditions, including substrates as unreactive as N2 or N2O. The serendipitous discovery in our laboratory that 1 also reacts with CH2Cl2 to generate alkyne metathesis catalysts in situ that proved highly chemoselective and therefore appropriate for applications to polyfunctionalized substrates opened an interface to advanced organic synthesis as well.23−27 In any case, the strongly π-donating and very bulky amido ligands safeguard the Mo(+3) center so effectively that dimerization of 1 has never been observed, not even as a side reaction in any of the applications that this complex has found so far in the literature.

Figure 1.

Figure 1

Open in a new tab

Top: the literature knows of either monomeric or homodimeric Mo(+3) complexes but in no case are both types supported by the same ligand set; the present study closes the gap and reports the first full ensemble, including an entirely new type of heterodimeric complexes; middle: known monomeric Mo(+3) complexes; bottom: envisaged use of tripodal ligands for the stabilization of monomeric Mo(+3); these ligands were previously used to craft “canopy catalysts” for alkyne metathesis and analogous Mo(+6) nitrido complexes; R = aryl, alkyl.

Very few other monomeric Mo(+3) complexes were described ever since. Arguably most prominent among them are complexes such as 2 comprising a trisamidoamine ligand framework,28−30 which served as entry point for the reductive cleavage of N2 with formation of NH3 in a catalytic mode.31 Though much less π-donating than amides, bulky silanolates were also shown to be adequate ligands, as witnessed by complex [(tBu3SiO)3Mo] (3), even though the structure of 3 proper is unknown and its monomeric nature was solely inferred from a crystalline phosphine adduct.32,33 Complex [[tBuO)3SiO]3Mo] (4) constitutes an interesting variation on that theme: by virtue of the additional lateral oxygen atoms, each siloxide engages in weak κ2 binding; 4 is hence a coordinatively saturated octahedral entity, yet retains the prototypical reactivity of a monomeric Mo(+3) species toward N2 and other small molecules.34 Complex 5 bearing overcrowded, strongly π-donating thiolate ligands gains additional stability – at the expense of reactivity – by coordination of one of the lateral arene rings onto the Mo(+3) center.35 An as yet higher level of coordinative and electronic saturation also helps stabilize complex 6, in which two of the three 2,5-dimethylpyrrolide ligands are η5-bound, thus rendering the bonding situation similar to that of a bent metallocene.36

Intrigued by these reports, we wondered whether tripodal silanolates or alkoxides of the types recently used to support high-valent molybdenum or tungsten alkylidynes could serve similar purposes. For the excellent synergy between these ligands and the respective high-valent metal centers, catalysts such as 7a and 8 excel in alkyne metathesis for reasons that are fairly clear by now.37−43 The isolobal terminal molybdenum(+6) nitrido complexes 7b are equally well behaved monomeric entities.44 In view of this encouraging precedent, we reasoned that such tripodal ligands might also be able to support Mo(+3): the phenyl ring forming the basal plane of the “canopy” ligand framework would effectively block one side of the targeted complex A (Figure 1), whereas the fence of upward-pointing substituents R on the tethers encircling the Mo(+3) atom potentially shields the opposite face; if properly chosen, dimerization might thus be prevented.

While this reasoning ultimately turned out to be correct, the new ligand platform actually proved far more auspicious, as it enabled the first “collective” synthesis of an entire panel of coordination compounds of this series. Specifically, it opened access to the corresponding monomeric Mo(+3) complex, the homoleptic dimer thereof with a tripodal silanolate cap on both ends, and even to heterodimeric complexes of type [X3Mo≡MoY3] that are without precedent in the literature. What makes this latter finding particularly stunning is the fact that these heterodimers comprise the intact Cummins complex 1 as one of the subunits, which is famous for not engaging in metal–metal triple bonding otherwise; the surprising ease of their formation is hence deemed remarkable, as are their structural and spectroscopic properties.

Results and Discussion

Monomeric Mo(+3) Complexes and a Homodimer with an Ultralong [Mo≡Mo] Bond

During our attempts at developing ever more efficient catalysts for alkyne metathesis,45,46 we noticed that the alkylidyne complex 7a (R = Me) is highly active but rather short-lived. The isolation of the corresponding tolane derivative 10 suggested that bimolecular coupling of 7a constitutes a (major) decomposition pathway; in this case, the corresponding homobimetallic dimer 11 should also be formed (Scheme 1).47 Although no X-ray crystal structure of this compound was obtained, the very diagnostic 95Mo NMR shift (δMo = 2631.5 ppm)48 in combination with matching HRMS and combustion analysis data allowed the assignment to be made with confidence. This result shows that methyl substituents on the silicon linkers of the tripodal ligand framework do not suffice to prevent metal–metal bonding from occurring. Interestingly though, reaction of 1 with the free ligand 9a failed to afford complex 11 but gave an ill-defined mixture of presumably oligomeric species.

Scheme 1. A Dumbbell Dimer with Two Tripodal Silanolate End-Caps.

Scheme 1

Open in a new tab

Therefore, we turned our attention to ligand 9b carrying sterically much more demanding 3,5-dimethylphenyl substituents on the lateral Si-atoms (Scheme 2). Addition of 9b to a solution of 1 in THF furnished a dark red solution; upon concentration, the newly formed pale-orange product [12·3thf] crystallized out. This allowed the constitution to be unambiguously determined as the targeted Mo(+3) monomer supported by the tripodal silanolate scaffold and three ancillary THF ligands (Figure 2). With a distance of 4.08 Å between the Mo center and the centroid of the basal phenyl ring, the metal does evidently not interact with the π-system underneath.

Scheme 2. A New Monomeric Mo(+3) Complex Supported by a Tripodal Silanolate Ligand Scaffold Exists in Three Distinctly Different Bonding Modes.

Scheme 2

Open in a new tab

Figure 2.

Figure 2

Open in a new tab

Truncated structure of complex [12·3thf] in the solid state; the 3,5-dimethylphenyl substituents on the silicon linkers and H-atoms are removed for clarity; the full structure is contained in the Supporting Information. Selected bond lengths (Å) and angles (°): Mo1–O1 2.0146(19), Mo1–O2 2.2407(19), Si1–O1–Mo1 163.16(17).

This information was all the more relevant as further analytical data of [12·3thf] could not be obtained. All attempts at drying the crystals in vacuo were accompanied by a drastic color change from orange to dark red; the same is true when [12·3thf] was dissolved in pentane: a deep red solution was instantly formed, from which red single crystals could be grown, leaving a colorless supernatant behind.

While the elemental analysis of the crystals agreed with the mono-THF adduct [12·thf], single crystal X-ray diffraction revealed a more intricate composition, rendering the analysis particularly challenging. Optically, the crystals looked good, but the diffraction pattern was of medium quality. Refinement revealed several disorders that could explain the lower intensities at high diffraction angles. We interpreted the structure model as a multicomponent crystal with three different chemical species. The by far major species with a refined occupancy of ≈90% was the mono-THF adduct [12·thf] (Figure 3). The entirely THF-free complex 12 accounting for an occupancy of ≈3% could also be identified from the residual electron density. The third component (ca. 7%) could not be adequately modeled, although the bis-THF adduct [12·2thf] seems possible and arguably likely (for details, see the Supporting Information). It is important to mention that this analysis was repeated with 9 different crystals from various batches at temperatures between 100 and 253 K; each of them was shown to comprise the constituents outlined above with similar ratios (for details, see the Supporting Information).

Figure 3.

Figure 3

Open in a new tab

Truncated structure of complex [12·thf] in the solid state; the 3,5-dimethylphenyl substituents on the silicon linkers and H-atoms are removed for clarity; the full structure is contained in the Supporting Information. Selected bond lengths (Å) and angles (°): Mo1A-O1 1.9162(14), Mo1A-O2 1.9134(14), Mo1A-O3A 1.8979(18), Mo1A-O4 2.2524(15), Mo1A-C1A 2.276(2), Mo1A-C2A 2.224(2), C1A-C2A 1.466(3), C2A-3 1.453(3), C3–C4 1.351(3), C4–C5 1.438(3), C5–C6A 1.375(4), C6A-C1A 1.429(4), C1A-C2A–C3–C4 19.7.

The fact that THF is bound to the Mo(+3) center is notable since the parent Cummins complex 1 had been explicitly classified as being non-Lewis acidic, incapable of ligating THF, pyridine or PEt3.21 The notion that silanolates gently upregulate the acidity of the central metal they are attached to had already surfaced in our previous studies.40,42 However, the THF ligands in [12·3thf] can only be weakly bound. Loss of two of them entails ligation of the basal phenyl moiety to the Mo-center, which is displaced from the central axis of the aryl ring underneath. Along comes a notable distortion of the aromatic π-system (Figure 3); this fact is manifested in the notable kink of the formerly planar ring toward the metal atom, resulting in fairly short Mo1A–C1A (2.276(2) Å) and Mo1A–C2A (2.224(2) Å) distances. These structural attributes indicate significant electron donation from the metal center into the antibonding π* orbital, corresponding to an η2-bonding mode of the arene.49 The ensuing partial loss of aromaticity of the basal plane illustrates the reactivity of the central metal atom.

As one might expect, the loss of THF is reversible (Scheme 3). Thus, addition of excess [D8]-THF to a solution of [12·thf] in [D8]-toluene allowed the reformation of labeled [12·3thf] to be monitored by 2H NMR spectroscopy despite the paramagnetic nature of the complexes (for details, see the SI); moreover, crystal structure analysis confirmed the constitution of the resulting adduct. In line with the literature precedent on monomeric Mo(+3) species,19 [12·thf] reacted with N2O to give the nitrido complex 7b, which was identified by comparison with an authentic sample;44 the expected nitrosyl complex, however, which should be concomitantly formed, has not been identified in the mixture. Likewise, treatment with 1,1-dichloropropane furnished the corresponding alkylidyne 7a′;23,24,27 although the (unoptimized) yield was low, the very characteristic spectroscopic fingerprints, most notably the deshielded alkylidyne C atom (δC = 322.8 ppm), cannot be missed; the structure was fully assigned from the mixture (for details, see the Supporting Information).

Scheme 3. Reactions of [12·thf].

Scheme 3

Open in a new tab

As mentioned above, [12·thf] is accompanied by the entirely THF-free species 12. Although the latter is only a very minor constituent with ≈3% occupancy in the unit cell, the diffraction data allowed its constitution to be deduced, although the atomic positions could not be determined very accurately and a detailed discussion is hence not possible. A comparison with the computed structure shows good agreement, thus confirming the tentative assignment (Figure 4). In any case, the Mo-atom is located on a 3-fold axis; it resides only 1.726 Å above the centroid of the basal phenyl group, which is almost planar. Therefore, an η6-interaction between the Mo(+3) center and the arene ring seems to be the most appropriate description of the bonding situation.49 The situation is reminiscent of what has been observed with the tris-thiolate complex 5.35

Figure 4.

Figure 4

Open in a new tab

Superposition of the truncated experimental structure (blue) of the THF-free complex 12 in the solid state with the truncated calculated structure (red); the 3,5-dimethylphenyl substituents R on the Si-linkers were removed for clarity; for the full structure and further details, see the Supporting Information.

Comparison of the structures displayed in Figures 2–4 showcases yet another remarkable attribute of the tripodal ligand architecture. In [12·3thf], the Mo-atom resides 1.05 Å above the plane defined by the three O atoms, in [12·thf] it lies roughly in this plane, whereas in 12 it comes to rest recognizably below the plane. The amazing floppiness of the silanol linkages is also manifested in the evidently adaptable bond lengths; thus, the average Mo–O distances vary considerably (2.034 Å in [12·3thf]; 1.909(6) Å in [12·thf]; 2.156 Å (computed) in 12). For these attributes, this type of ligand is broadly applicable: it can adapt to the needs of a single metal in different oxidation states, coordination environments and/or binding modes as shown herein, but can also bind to many different central metals of largely variable radii. A more comprehensive study sampling the periodic table is underway and will be published in due course.50

Homodimerization and Discovery of an Entirely New Type of Heterodimers

Attempts were made at stripping the remaining THF off in order to obtain ligand-free 12 in pure form. To this end, [12·thf] was dissolved in toluene and all volatile materials were evaporated in high vacuum; this procedure was repeated three times. When a solution of the residue in toluene was then kept at ambient temperature, the slow formation of large orange single crystals was observed, which proved highly insoluble in all common inert solvents. Therefore, analysis was essentially limited to HRMS, combustion data, and single crystal X-ray diffraction. Much to our surprise, the complex turned out not to be the expected THF-free monomer 12 but rather the corresponding homodimer 13 (Scheme 4), akin to homodimer 11 derived from alkylidyne 7a (R = Me) that had previously only been inferred from the analytical and spectral data (Scheme 1).47 The 3,5-dimethylphenyl substituents on the Si-atoms of 13 are innately interlocked, thus rendering the core region exceptionally crowded; the exceptionally long Mo≡Mo bond (2.2873(3) Å) is likely a derivative thereof (Figure 5). This truly encumbered situation may also explain why 13 was obtained in only very low yield.51

Scheme 4. Formation of an Overcrowded Homodimer.

Scheme 4

Open in a new tab

Figure 5.

Figure 5

Open in a new tab

Truncated structure of the homodimeric complex 13 in the solid state; the 3,5-dimethylphenyl substituents on the silicon linkers and H-atoms are removed for clarity; the full structure is contained in the Supporting Information. Selected bond lengths (Å) and angles (°): Mo1–Mo1′ 2.2873(3), Mo1–O1 1.9107(12), Mo1–O2 1.9043(17), Mo1–O3 1.9148(12), Si1–O1–Mo1 158.67(8), O1–Mo1–Mo1′ 103.23(4).

We conjectured that the use of a less good donor solvent could pave an alternative way to THF-free 12. To this end, the trisilanol ligand 9b was added to a suspension of 1 in Et2O at −78 °C and the resulting mixture was slowly warmed to ambient temperature. Surprisingly, NMR inspection of the crude material suggested that ca. 50% of the ligand remained unchanged even though 1 was fully consumed. To confirm this paradoxical result, the experiment was repeated with a 1:2 ratio of the reaction partners, which allowed the new complex 14 to be isolated in analytically pure form in 53% yield (Scheme 5). Its diamagnetic nature suggested that a Mo≡Mo bond might have been formed. ROESY cross peaks showed that two chemically different ligand spheres are present as part of one and the same molecule. The recorded 15N NMR shift (δN = −129 ppm) differs significantly from that of the free (tBu)(Ar)NH (δN = −289 ppm), thus indicating a metal-bound amide ligand. Overall, the complex has C3 symmetry about the Mo≡Mo axis on the NMR time scale, since only one set of signals was observed for each of the ligands. In contrast, high barriers to rotation about the N–Ar as well as the Si–R bonds render the H-atoms on the aryl rings diastereotopic.

Scheme 5. A Heterodimeric Complex Comprising an Intact “Cummins Complex” Entity as One of the Constituents.

Scheme 5

Open in a new tab

The structure of 14 in the solid state confirmed the conclusions drawn from the NMR data. To the best of our knowledge, this complex is the first example of a heterodimeric species of the type [X3Mo≡MoY3] (Figure 6). Truly perplexing is the fact that 14 incorporates an intact “Cummins fragment” as one of its constituents, even though complex 1 is famous for not forming metal–metal triple bonding otherwise. With 2.3440(3) Å, the Mo≡Mo bond of 14 is by far the longest triple bond between two unbridged Mo-centers with CN = 4 each (CN = coordination number) documented in the literature (see below);52 it is even longer than that of the overcrowded homodimer 13. Despite this striking attribute, which one could (mis)take for a sign of weak bonding, the monomeric complex 12 (with or without additional Et2O ligands) has neither been observed nor isolated; heterodimerization must hence be fast and facile. Along the same lines, we note that 14, once formed, is thermally and chemically robust; no indications for dissociation into the constituent monomeric units or for “scrambling” into the corresponding homodimer 13 and free 1 were noticed on prolonged warming of a solution in toluene or on treatment with two-electron donor ligands.

Figure 6.

Figure 6

Open in a new tab

Truncated molecular structure of complex 14 in the solid state; the 3,5-dimethylphenyl substituents on the silicon linkers, disordered parts, and H-atoms are not shown for clarity; the full structure is contained in the Supporting Information. Selected bond lengths (Å) and angles (°): Mo1–Mo2 2.3440(3); Mo1–O1 1.9305(15), Mo1–O2 1.9455(14), Mo1–O3 1.9409(14), Mo2–N1 1.9885(18), Mo2–N2 1.9810(18), Mo2–N3 1.9954(18), O1–Mo1–Mo2 105.10(5), N3–Mo2–Mo1 110.15(5), Si1–O1–Mo1 173.36(11).

Next, it was probed whether the formation of 14 is due to the peculiar trisilanolate ligand framework. To this end, we first resorted to the tripodal alcohol derivative 15, which had originally been designed to bolster the activity of tungsten alkylidyne catalysts of type 8.42 This particular ligand is similar to 9 in geometric terms, but its −OH groups are less acidic than the corresponding silanols in 9; once bound to the metal center, alkoxides are notably better π-donor ligands than siloxides. These nuances apart, addition of 1 to a solution of 15 cleanly furnished the corresponding heterodimeric complex 16; in this case, the best results were obtained in toluene as the solvent, which allowed 16 to be isolated in analytically pure form in 77% yield (Scheme 6). The recorded NMR data were analogous to those of 14; the diagnostic 15N NMR shift (δN = −147 ppm) was taken as a clear sign for the incorporation of an intact “Cummins fragment” into the newly formed Mo≡Mo bond. Variable temperature NMR spectra showed high rotational barriers about the N–Ar bonds of the amido ligand; one of the aromatic protons evidently resides in the anisotropy cone of a neighboring aryl ring and is therefore notably shielded (δH = 4.19 ppm).53 The constitution of 16 as a heterodimeric entity was ascertained by crystal structure analysis (Figure 7). With 2.2955(7) Å, the Mo≡Mo distance is slightly shorter than that of 14 (2.3440(3) Å). Another interesting structural feature is visible in the Newman-type projection, which shows that the ligands are not perfectly staggered about the Mo–Mo axis, despite their bulk; the smallest torsion (O1–Mo1–Mo2–N1) is only 39.6° (rather than 60°).

Scheme 6. Heterodimers by Alcoholysis of 1.

Scheme 6

Open in a new tab

Figure 7.

Figure 7

Open in a new tab

Molecular structure of the heterodimeric complex 16 in the solid state; top: side view; bottom: projection along the 3-fold crystallographic Mo–Mo axis; H-atoms, disordered parts, and solute solvent molecules in the unit cell not shown for clarity; the full structure is contained in the Supporting Information. Selected bond lengths (Å) and angles (°): Mo1–Mo2 2.2955(7), Mo1–O1 1.915(7), Mo2–N1 1.998(2), C1–O1–Mo1 144.7(7), O1–Mo1–Mo2 100.44(7), N1–Mo2–Mo1 106.29(8).

The astounding bias for heterodimerization manifested in these examples is not caused by the peculiar tripodal ligand framework of 9 and 15, in which the three arms carrying the protic substituents are tied back onto a rigidifying arene linker. This became clear when 1 was reacted with ordinary tert-butanol (Scheme 6). Once again, the corresponding heterodimer 17 was formed in respectable yield. It shows the same distinctive structural attributes, namely a very long Mo≡Mo bond (2.2944(2) Å) as well as small torsional angles (O–Mo–Mo–N: 32.57, 33.35, 34.72°) even further away from the ideal 60° of a staggered “ethane-like” conformer that one might intuitively expect to be optimal for such a crowded ligand environment (Figure 8). It is likely that these peculiar structural features as well as the striking stability of such overcrowded dimers are due, at least in part, to attractive forces such as “interligand” London dispersion interactions.54−56

Figure 8.

Figure 8

Open in a new tab

Molecular structure of complex 17 in the solid state; H-atoms and disordered solvent in the unit cell removed for clarity. Selected bond lengths (Å) and angles (°): Mo1–Mo2 2.2944(2), Mo1–N1 1.9893(14), Mo1–N2 1.9910(14), Mo1–N3 2.0010(14), Mo2–O1 1.9147(12), Mo2–O2 1.9122(12), Mo2–O3 1.9175(13), Mo1–Mo2–O1 99.49(4), Mo2–Mo1–N1 106.72(4).

Other structural parameters of the heterodimers are also best discussed with complex 17 as it lacks any distortion by a caged ligand framework. With an average Mo–Mo–O bond angle of 100.3°, the coordination geometry about Mo2 is halfway between planar and tetrahedral, which may indicate that the bending back of the ligands is more likely caused by ligand–ligand repulsion rather than rehybridization. The fact that the Mo–Mo–N angles are slightly larger (average: 106.1°) does not discount the argument, as the sterically more demanding −NAr(tBu) groups must tilt further away. The Mo2–OtBu distances in 17 (average: 1.9148 Å) are comparable to those in [(RO)3Mo≡Mo(OR)3] (R = CMe2Ph; average: 1.892 Å)57 and in the alkylidyne ArC≡Mo(OtBu)3 (average: 1.884 Å; Ar = 2,6-dimethylphenyl);58 as expected, they are shorter than the Mo–OSi bonds in complex 14 (average: 1.939 Å), which confirms that alkoxides are better π-donors than the corresponding silanolates.40,59,60 The Mo–N bonds in 17 (average: 1.994 Å) are longer than those in the parent Cummins complex 1 (1.967 Å);19 the barrier to rotation about the N–Ar bond was determined by NMR to be on the order of 12.5 kcal·mol–1 (for details, see the Supporting Information).

Perplexed by the ease with which these unorthodox heterodimers are formed, the alcoholysis of 1 was repeated with tert-butanol in excess. Although a mixture was obtained in this case, the heterodimer 17 was still the major species in the crude material, whereas the homoleptic dimer [Mo2(OtBu)6] (18) was minor (Scheme 6). Similarly instructive was another control experiment, in which purified 17 was treated with tert-butanol; the recorded spectra showed no sign of formation of 18 even when the temperature was raised and the solution stirred at 60 °C for 15 h.61 This result is striking if one considers that the alcoholysis of [Mo2(NMe2)6] (19) with tert-butanol in a hydrocarbon solvent is the standard method for the preparation of 18.62 Moreover, 18 is the final product of and thermodynamic sink in many different reactions starting from various molybdenum sources.5,12,63−67 The remarkable preference for the formation of the heterodimer 17 is hence most likely kinetic in origin: putative [(tBuO)3Mo] (or a mixed species [(tBuO)n[(tBu)(Ar)N]mMo], n + m = 3) in statu nascendi is evidently so “hot” that it instantly traps any remaining 1 with formation of 17.68 Once formed, the crowded ligand sphere precludes protonolysis of the remaining Mo–N bonds, at least by an encumbered reaction partner such as tert-butanol.

The notion that the Cummins complex 1 can be intercepted by a (transient) monomeric [MoX3] entity was experimentally confirmed. Thus, a 1:1 mixture of [12·thf] and 1 in toluene cleanly afforded the corresponding heterodimer 14 as proven by NMR spectroscopy (Scheme 7).

Scheme 7. Cross-Dimerization of Two Molecularly Defined Mo(+3) Complexes.

Scheme 7

Open in a new tab

95Mo NMR Data and Computational Analysis

A brief comment on the 95Mo NMR shifts of the heterodimeric complexes is warranted. Because this molybdenum isotope is a spin 5/2 nucleus with low natural abundance (≈15.9%), an unfavorable gyromagnetic ratio and a low quadrupole moment, 95Mo NMR is not without challenges.69−71 Early work, however, had shown that the nuclei in dimeric Mo≡Mo complexes are strongly deshielded and can therefore be easily distinguished from other molybdenum species.72 Although the problems caused by the quadrupolar nature of 95Mo are massively enhanced by lowering symmetry and increasing the molecular weight of the compounds, we managed to record two distinct 95Mo resonances for the new heterodimeric complex 17: the sharper signal at 3260 ppm is tentatively attributed to the [≡Mo(OtBu)3] unit, whereas a much broader signal at 3127 ppm likely stems from the [≡Mo[N(Ar)(tBu)]3] entity.73 For complex 14, however, only one very broad peak at 3314 ppm was detected even after 5 d of acquisition time.

Although the data set is currently small, some aspects are noteworthy (Table 1). The heterodimers 14 and 17 are massively deshielded (≈600–800 ppm) compared to homodimeric siblings carrying the same or similar alkoxide, siloxide or amido ligands (11, 18, 19); at the same time, they show much longer Mo–Mo distances compared to these (and all other) homodimers bearing heteroatom ligands, for which metric data are available (Figure 9). Therefore, one might speculate that both properties are manifestations of the peculiar electronic character of their polarized Mo≡Mo core. A look at the homoleptic complex [R3Mo≡MoR3] (20, R = CH2Me3) devoid of any π-donating ligands, however, teaches that a simple correlation between shift and bond length does not exist: 20 resonates at δMo = 3695 ppm,72 yet shows an extremely short Mo≡Mo bond (2.1654(7) Å).76

Table 1. 95Mo NMR Shifts ([D8]-Toluene, 333 K (unless stated otherwise)) and Mo–Mo Distances of Selected Dimeric Complexes.

graphic file with name ja5c02178_0018.jpg

Open in a new tab
a

Ref (72): 2430 ppm (C6D6, RT).

b

Two independent complexes in the asymmetric unit cell, ref (74).

c

Ref (47).

d

Ref (67).

e

Ref (72): 2645 ppm (C6D6, RT).

f

Ref (57).

g

Ref (75).

h

Ref (72).

i

Ref (76).

Figure 9.

Figure 9

Open in a new tab

Plot of the Mo–Mo bond distances of complexes comprising an unbridged Mo≡Mo core with a CN = 4 on both metal atoms found in the Cambridge Crystallographic Data Centre (black dots); comparison with the new heterodimeric complexes reported herein (for details including structures and accession codes of the literature-known complexes, see the Supporting Information).

As these experimental data are ostensibly difficult to reconcile, we resorted to a computational analysis. The computed 95Mo NMR shieldings/shifts (Figure 10) match the recorded values well, thus corroborating the validity of the chosen level of theory (TPSSH/def2-TZVPP (SARC-ZORA-TZVPP for Mo) and CPCM(toluene) with structures optimized at the B3LYP/def2-TZVP D3BJ CPCM(toluene) level of theory). In line with what the data of complex 20 had suggested, a meaningful correlation between any of the paramagnetic components of the shielding tensors and the [Mo≡Mo] bond distances has not been found.

Figure 10.

Figure 10

Open in a new tab

Comparison of the MO schemes of complexes 20, 19, 18 and the new heterodimeric complex 14; for full MO plots and further details, see the SI.

A more detailed orbital analysis provided some insights into why this is the case. Qualitatively, the bonding in [X3Mo≡MoX3] complexes is adequately described by assuming that the dz2 and the dxz and dyz orbitals of the individual metal atoms form one σ- and two π-bonds, which constitute the HOMO–2, HOMO–1 and the HOMO, respectively.2 Interestingly, however, it turns out that this rather intuitive interpretation is valid only for X = OR, that is for complexes of type 18 (Figure 10). In case of complex 19 endowed with amido ligands, the p-orbitals of the N-lone pairs lie in the HOMO–LUMO region; although they do not contribute to Mo–Mo binding, they affect the 95Mo NMR shifts. Actually, the orbitals forming the Mo≡Mo triple bond of 19 are HOMO–2, HOMO–3 and HOMO–4. Although formed by entirely different orbitals, the energies of the HOMOs of 18 and 19 are similar.

It is striking that the ordering of the σ- and the π-orbitals changes completely in the neopentyl complex 20 devoid of heteroatoms in the ligand sphere (entry 6):77 it is the σ-orbital derived from the dz2 orbitals of the individual Mo atoms which constitutes the HOMO rather than the HOMO–2, as one might expect and as has indeed been found in case of 18; in energetic terms, it is significantly lower-lying than the HOMOs of 18 and 19. These three examples suffice to demonstrate that one has to beware of generalizations even when analyzing the bonding situation in simple homodimeric and homoleptic [X3Mo≡MoX3] complexes: the HOMO of 18 has π-symmetry, that of 20 has σ-character, whereas the HOMO of 19 is largely centered on the heteroatoms of the ancillary amido ligands.

Therefore, it is unsurprising that heterodimer 14 is yet another special case. All frontier orbitals are strongly polarized toward the side of the Mo atom carrying the −NR2 ligands; the average energy of the three frontier orbitals is about −4.8 eV, which is higher than that of the homodimers (ca. −5.5 eV (20), −5.1 eV (19), −5.0 eV (18)) (obtained at the TPSS level of theory, see the Supporting Information for details). While this trend suggests that stretching of the [Mo≡Mo] bond comes along with the expected increase in the energies of the orbitals forming the Mo–Mo σ- and π-bonds, no simple correlation between this trend and the 95Mo NMR chemical shifts does emerge; the relationship is obviously more complex. An in depth analysis seems warranted once more experimental shift data is available.

Conclusions

Although tripodal silanols of type 9 have originally been designed to foster the catalytic activity and performance of high-valent Mo(+6)-alkylidyne complexes, they proved equally versatile in the realm of Mo(+3) chemistry; they are the first and, so far, only ligands capable of supporting highly reactive monomeric [MoX3] complexes, the corresponding homodimers, as well as a previously unknown type of heterodimers. Specifically, the resulting mononuclear trigonal-planar Mo(+3) complexes are rendered kinetically stable if the substituents on the peripheral Si-linkers of the tripodal ligand framework are sufficiently encumbered, even though such d3 electron species are exceptionally reactive and hence exceedingly rare otherwise. For the peculiar ligand architecture, the Mo(+3) complex resulting from ligand 9b was shown to exist in three distinctly different binding modes: it can form an adduct with three, one or no THF ligand. Along comes a massive change in the interaction with the basal plane of the ligand framework itself, which is either not ligated at all, η2-bound, or forced to bind in an η6-coordination mode, respectively. The adaptive character manifested in this data is expected to be enabling in other context too.

Homodimerization of the monomeric complexes is possible but kinetically handicapped. In striking contrast, they can intercept unreacted Cummins complex 1 with exceptional ease even though the latter is famous for not engaging in metal–metal bonding otherwise. The resulting unsymmetrical heterodimers are the first examples with a ligand pattern of the general type [X3Mo≡MoY3]. Their metal–metal bond is polarized and exceptionally long, though chemically and thermally rather stable. Moreover, this new type of heterodimerization is not limited to tripodal silanolates as the ligands but is clearly a more general transformation. The novel class of heterodimeric [X3Mo≡MoY3] complexes described herein shows strongly deshielded 95Mo NMR signals; however, a simple correlation between the chemical shift and the bond length does not exist because ligand-based orbitals also affect the shielding tensor to a significant extent.

Acknowledgments

Generous financial support by the Max-Planck-Gesellschaft is gratefully acknowledged. We thank A. Mateos Calbet for help with the N2O experiments, H. Schucht and L. Schulte-Zweckel for their perseverance in carrying out numerous XRD measurements, Dr. J. Hillenbrand for valuable discussions at the outset of the project, and all analytical departments of our Institute for excellent assistance.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/jacs.5c02178.

  • Experimental Section including characterization data and NMR spectra of new compounds (PDF)

  • Crystallographic Data and Discussion Computational Details (PDF)

  • Computational Study (PDF)

Open access funded by Max Planck Society.

The authors declare no competing financial interest.

Supplementary Material

ja5c02178_si_001.pdf (6.7MB, pdf)
ja5c02178_si_002.pdf (6.1MB, pdf)
ja5c02178_si_003.pdf (1.6MB, pdf)

References

  1. a Huq F.; Mowat W.; Shortland A.; Skapski A. C.; Wilkinson G. Crystal structure of hexakis(trimethylsilylmethyl)-dimolybdenum. J. Chem. Soc. D: Chem. Commun. 1971, 1079–1080. 10.1039/c29710001079. [DOI] [Google Scholar]; b Yagupsky G.; Mowat W.; Shortland A.; Wilkinson G. Trimethylsilylmethyl compounds of transition metals. J. Chem. Soc. D: Chem. Commun. 1970, 1369–1370. 10.1039/c29700001369. [DOI] [Google Scholar]; The compound had originally been misassigned as a monomeric Mo(+4) tetraalkyl species, see
  2. Cotton F. A.; Murillo C. A.; Walton R. A.. Multiple Bonds Between Metal Atoms., 3rd ed.; Springer: New York, 2005. [Google Scholar]
  3. Chisholm M. H. The coordination chemistry of dinuclear molybdenum(III) and tungsten(III): d3-d3 dimers. Acc. Chem. Res. 1990, 23 (12), 419–425. 10.1021/ar00180a004. [DOI] [Google Scholar]
  4. Chetcuti M. J.; Chisholm M. H.; Folting K.; Haitko D. A.; Huffman J. C.; Janos J. 1,2-Dibenzyl- and 1,2-diaryltetrakis-(dimethylamido)dimolybdenum and -ditungsten compounds: M2R2(NMe2)4 (M≡M). Structural effects of Me2N-to-M π-bonding. J. Am. Chem. Soc. 1983, 105 (5), 1163–1170. 10.1021/ja00343a015. [DOI] [Google Scholar]
  5. Krackl S.; Ma J.-G.; Aksu Y.; Driess M. Facile Access to Homo- and Heteroleptic, Triply Bonded Dimolybdenum Hexaalkoxides with Unsaturated Alkoxide Ligands. Eur. J. Inorg. Chem. 2011, 2011, 1725–1732. 10.1002/ejic.201001236. [DOI] [Google Scholar]
  6. Chisholm M. H.; Rothwell I. P. Formation and stability of 1,1- and 1,2-bis(dimethylamido)tetrakis(trimethylsiylmethyl)-dimolyb-denum (Mo≡Mo). J. Chem. Soc., Chem. Commun. 1980, 985–987. 10.1039/C39800000985. [DOI] [Google Scholar]
  7. Chisholm M. H.; Folting K.; Huffman J. C.; Rothwell I. P. The molybdenum-molybdenum triple bond. 11. 1,1- and 1,2-Disubstituted dimolybdenum compounds of formula Mo2X2(CH2SiMe3)4. Observation of rotation about the triple bond Mo≡Mo. Organometallics 1982, 1 (2), 251–259. 10.1021/om00062a005. [DOI] [Google Scholar]
  8. Chisholm M. H.; Click D. R.; Huffman J. C. Further Studies on the Substitutional Behavior of 1,2-Mo2Br2(CH2SiMe3)4. Alkyl, Amide, Phosphide, Alkoxide, and Thiolate for Bromide Exchange and Isomerizations of 1,1- and 1,2-Mo2X2(CH2SiMe3)4 Compounds. Organometallics 2000, 19 (19), 3916–3924. 10.1021/om0003584. [DOI] [Google Scholar]
  9. This also holds true for a special subset comprising ligands that bridge the two metal centers, see; Budzichowski T. A.; Chacon S. T.; Chisholm M. H.; Feher F. J.; Streib W. Dimolybdenum and ditungsten derivatives of the trisilanol, [(c-C6H11)7Si7O9(OH)3]: [(c-C6H11)7Si7O12]2Mo2 (M≡M) and [(c-C6H11)7Si7O12]2W2(μ-H)(O-tert-Bu). J. Am. Chem. Soc. 1991, 113 (2), 689–691. 10.1021/ja00002a056. [DOI] [Google Scholar]
  10. Complexes that exist in equilibrating ligand-bridged and dumbbell forms might be an exception; for an example, see ref (9); for a study into assisted 1,2-group transfer see ref (8)
  11. Akiyama M.; Chisholm M. H.; Cotton F. A.; Extine M. W.; Murillo C. A. The tungsten-to-tungsten triple bond. 7. Replacement of dimethylamido by chloro groups in hexakis(dimethylamido)ditungsten and -dimolybdenum compounds. Preparation, properties, and structure of dichloro-tetrakis(dimethylamido) ditungsten and -dimolybdenum. Inorg. Chem. 1977, 16 (9), 2407–2411. 10.1021/ic50175a049. [DOI] [Google Scholar]
  12. Chisholm M. H.; Huffman J. C.; Tatz R. J. β-Hydrogen effects in alkylalkoxides of dimolybdenum (M≡M). J. Am. Chem. Soc. 1983, 105 (7), 2075–2077. 10.1021/ja00345a073. [DOI] [Google Scholar]
  13. Chisholm M. H.; Tatz R. J. Metal alkoxides - models for metal oxides. 10. Synthesis and properties of alkyl alkoxides of formula 1,2-Mo2R2(OR’)4 and Mo2R(OR’)5 (M≡M). Organometallics 1986, 5 (8), 1590–1598. 10.1021/om00139a014. [DOI] [Google Scholar]
  14. Chisholm M. H.; Huffmann J. C.; Tatz R. J. Dimolybdenum compound (iso-PrO)3Mo≡Mo(CH2Ph)2(O-iso-Pr)(PMe3): the first example of a compound containing a triple bond uniting three- and four-coordinate molybdenum atoms and the observation of phosphine-promoted, reversible benzyl-alkoxy migrations at the (M≡M)6+ center. J. Am. Chem. Soc. 1984, 106 (18), 5385–5386. 10.1021/ja00330a077. [DOI] [Google Scholar]
  15. Schulz H.; Folting K.; Huffman J. C.; Streib W. E.; Chisholm M. H. Studies of mono substituted compounds of formula M2(NMe2)5X (M≡M), where M = molybdenum and tungsten and X = iodine, alkyl, aryl, and diphenylphosphido, and the bridged-1,1’-ferrocenyl complex W2(NMe2)(μ-(C5H4)2Fe). Inorg. Chem. 1993, 32 (26), 6056–6066. 10.1021/ic00078a025. [DOI] [Google Scholar]
  16. Abbott R. G.; Cotton F. A.; Falvello L. R. Multiply bonded dimetal fluoroalkoxides. 1. Pentafluorophenoxides of dimolybdenum and ditungsten. Inorg. Chem. 1990, 29 (3), 514–521. 10.1021/ic00328a036. [DOI] [Google Scholar]
  17. Our search showed a few complexes of the net formula [Mo2X3Y3], but none of them is of the unsymmetrical type [X3Mo≡MoY3], see:; a Chisholm M. H.; Davidson E. R.; Huffman J. C.; Quinlan K. B. J. Am. Chem. Soc. 2001, 123, 9652–9664. 10.1021/ja001208n. [DOI] [PubMed] [Google Scholar]; b Ahmed K. J.; Chisholm M. H.; Folting K.; Huffman J. C. Inorg. Chem. 1985, 24, 4039–4044. 10.1021/ic00218a017. [DOI] [Google Scholar]
  18. Laplaza C. E.; Cummins C. C. Dinitrogen Cleavage by a Three-Coordinate Molybdenum(III) Complex. Science 1995, 268 (5212), 861–863. 10.1126/science.268.5212.861. [DOI] [PubMed] [Google Scholar]
  19. Laplaza C. E.; Odom A. L.; Davis W. M.; Cummins C. C.; Protasiewicz J. D. Cleavage of the Nitrous Oxide NN Bond by a Tris(amido)molybdenum(III) Complex. J. Am. Chem. Soc. 1995, 117 (17), 4999–5000. 10.1021/ja00122a033. [DOI] [Google Scholar]
  20. Laplaza C. E.; Johnson M. J. A.; Peters J. C.; Odom A. L.; Kim E.; Cummins C. C.; George G. N.; Pickering I. J. Dinitrogen Cleavage by Three-Coordinate Molybdenum(III) Complexes: Mechanistic and Structural Data. J. Am. Chem. Soc. 1996, 118, 8623–8638. 10.1021/ja960574x. [DOI] [Google Scholar]
  21. Cummins C. C. Reductive cleavage and related reactions leading to molybdenum–element multiple bonds: new pathways offered by three-coordinate molybdenum(III). Chem. Commun. 1998, 1777–1786. 10.1039/a802402b. [DOI] [Google Scholar]
  22. Zhai D.-D.; Zhang S.-Q.; Xie S.-J.; Wu R.-K.; Liu F.; Xi Z.-F.; Hong X.; Shi Z.-J. n-Bu)4NBr-Promoted N2 Splitting to Molybdenum Nitride. J. Am. Chem. Soc. 2022, 144 (31), 14071–14078. 10.1021/jacs.2c01507. [DOI] [PubMed] [Google Scholar]
  23. Fürstner A.; Mathes C.; Lehmann C. W. Mo[N(tBu)(Ar)]3 Complexes as Catalyst Precursors: In Situ Activation and Application to Metathesis Reactions of Alkynes and Diynes. J. Am. Chem. Soc. 1999, 121, 9453–9454. 10.1021/ja991340r. [DOI] [Google Scholar]
  24. Fürstner A.; Mathes C.; Lehmann C. W. Alkyne Metathesis: Development of a Novel Molybdenum-Based Catalyst System and its Application to the Total Synthesis of Epothilone A and C. Chem. - Eur. J. 2001, 7, 5299–5315. . [DOI] [PubMed] [Google Scholar]
  25. a Fürstner A.; Radkowski K.; Grabowski J.; Wirtz C.; Mynott R. Ring-Closing Alkyne Metathesis. Application to the Total Synthesis of Sophorolipid Lactone. J. Org. Chem. 2000, 65 (25), 8758–8762. 10.1021/jo0012952. [DOI] [PubMed] [Google Scholar]; b Fürstner A.; Grela K.; Mathes C.; Lehmann C. W. Novel and Flexible Entries into Prostaglandins and Analogues Based on Ring Closing Alkyne Metathesis or Alkyne Cross Metathesis. J. Am. Chem. Soc. 2000, 122 (48), 11799–11805. 10.1021/ja003119g. [DOI] [Google Scholar]
  26. a Fürstner A.; De Souza D.; Turet L.; Fenster M. D. B.; Parra-Rapado L.; Wirtz C.; Mynott R.; Lehmann C. W. Total Syntheses of the Actin-Binding Macrolides Latrunculin A, B, C, M, S and 16-epi-Latrunculin B. Chem. - Eur. J. 2007, 13 (1), 115–134. [DOI] [PubMed] [Google Scholar]; b Fürstner A.; Larionov O.; Flügge S. What is Amphidinolide V? Report on a Likely Conquest. Angew. Chem., Int. Ed. 2007, 46, 5545–5548. 10.1002/anie.200701640. [DOI] [PubMed] [Google Scholar]; c Mailhol D.; Willwacher J.; Kausch-Busies N.; Rubitski E. E.; Sobol Z.; Schuler M.; Lam M.-H.; Musto S.; Loganzo F.; Maderna A.; Fürstner A. Synthesis, Molecular Editing, and Biological Assessment of the Potent Cytotoxin Leiodermatolide. J. Am. Chem. Soc. 2014, 136 (44), 15719–15729. 10.1021/ja508846g. [DOI] [PubMed] [Google Scholar]; d Schulthoff S.; Hamilton J.; Heinrich M.; Kwon Y.; Wirtz C.; Fürstner A. The Formosalides: Structure Determination by Total Synthesis. Angew. Chem., Int. Ed. 2021, 60, 446–454. 10.1002/anie.202011472. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. This approach to alkyne metathesis catalysts has later been refined, see:; a Zhang W.; Kraft S.; Moore J. S. Highly Active Trialkoxymolybdenum(VI) Alkylidyne Catalysts Synthesized by a Reductive Recycle Strategy. J. Am. Chem. Soc. 2004, 126 (1), 329–335. 10.1021/ja0379868. [DOI] [PubMed] [Google Scholar]; b Zhang W.; Lu Y.; Moore J. S. Preparation of a Triamidomolybdenum(VI) Propylidyne Complex – A Highly Active Catalyst Precursor for Alkyne Metathesis. Org. Synth. 2007, 84, 163. 10.1002/0471264229.os084.17. [DOI] [Google Scholar]
  28. Yandulov D. V.; Schrock R. R.; Rheingold A. L.; Ceccarelli C.; Davis W. M. Synthesis and Reactions of Molybdenum Triamidoamine Complexes Containing Hexaisopropylterphenyl Substituents. Inorg. Chem. 2003, 42 (3), 796–813. 10.1021/ic020505l. [DOI] [PubMed] [Google Scholar]
  29. Byrnes M. J.; Dai X.; Schrock R. R.; Hock A. S.; Müller P. Some Organometallic Chemistry of Molybdenum Complexes that Contain the [HIPTN3N]3- Triamidoamine Ligand, {[3,5-(2,4,6-i-Pr3C6H2)2C6H3NCH2CH2]3N}3. Organometallics 2005, 24 (18), 4437–4450. 10.1021/om050373e. [DOI] [Google Scholar]
  30. Schrock R. R. Transition Metal Complexes That Contain a Triamidoamine Ligand. Acc. Chem. Res. 1997, 30 (1), 9–16. 10.1021/ar950195t. [DOI] [Google Scholar]
  31. Yandulov D. V.; Schrock R. R. Catalytic Reduction of Dinitrogen to Ammonia at a Single Molybdenum Center. Science 2003, 301 (5629), 76–78. 10.1126/science.1085326. [DOI] [PubMed] [Google Scholar]
  32. Kuiper D. S.; Wolczanski P. T.; Lobkovsky E. B.; Cundari T. R. Low Coordinate, Monomeric Molybdenum and Tungsten (III) Complexes: Structure, Reactivity and Calculational Studies of (silox)3Mo and (silox)3ML (M = Mo, W; L = PMe3, CO; silox = tBu3SiO). J. Am. Chem. Soc. 2008, 130, 12931. 10.1021/ja802706u. [DOI] [PubMed] [Google Scholar]
  33. See also:; Kuiper D. S.; Douthwaite R. E.; Mayol A.-R.; Wolczanski P. T.; Lobkovsky E. B.; Cundari T. R.; Lam O. P.; Meyer K. Inorg. Chem. 2008, 47, 7139–7153. 10.1021/ic800139c. [DOI] [PubMed] [Google Scholar]
  34. Pucino M.; Allouche F.; Gordon C. P.; Wörle M.; Mougel V.; Copéret C. A reactive coordinatively saturated Mo(iii) complex: exploiting the hemi-lability of tris(tert-butoxy)silanolate ligands. Chem. Sci. 2019, 10 (25), 6362–6367. 10.1039/C9SC01955C. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Buyuktas B. S.; Power P. P. Synthesis and characterization of the monomeric molybdenum trithiolate Mo(η7-dmpt)(η1-dmpt)2 (dmpt = 2,6-dimesitylphenylthiolate). Chem. Commun. 1998, 1689–1690. 10.1039/a803064b. [DOI] [Google Scholar]
  36. Wampler K. M.; Schrock R. R. Molybdenum Tris(2,5-dimethylpyrrolide), a Rare Homoleptic Molybdenum(III) Monomer. Inorg. Chem. 2008, 47 (22), 10226–10228. 10.1021/ic801695j. [DOI] [PubMed] [Google Scholar]
  37. Hillenbrand J.; Leutzsch M.; Fürstner A. Molybdenum Alkylidyne Complexes with Tripodal Silanolate Ligands. The Next Generation of Alkyne Metathesis Catalysts. Angew. Chem., Int. Ed. 2019, 58, 15690–15696. 10.1002/anie.201908571. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Hillenbrand J.; Leutzsch M.; Yiannakas E.; Gordon C. P.; Wille C.; Nöthling N.; Copéret C.; Fürstner A. “Canopy Catalysts” for Alkyne Metathesis: Molybdenum Alkylidyne Complexes with a Tripodal Ligand Framework. J. Am. Chem. Soc. 2020, 142 (25), 11279–11294. 10.1021/jacs.0c04742. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Haack A.; Hillenbrand J.; Leutzsch M.; van Gastel M.; Neese F.; Fürstner A. Productive Alkyne Metathesis with “Canopy Catalysts” Mandates Pseudorotation. J. Am. Chem. Soc. 2021, 143, 5643–5648. 10.1021/jacs.1c01404. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Haack A.; Hillenbrand J.; van Gastel M.; Fürstner A.; Neese F. Spectroscopic and Theoretical Study on Siloxy-Based Molybdenum and Tungsten Alkylidyne Catalysts for Alkyne Metathesis. ACS Catal. 2021, 11, 9086–9101. 10.1021/acscatal.1c01587. [DOI] [Google Scholar]
  41. Berkson Z. J.; Lätsch L.; Hillenbrand J.; Fürstner A.; Copéret C. Classifying and Understanding the Reactivities of Mo-Based Alkyne Metathesis Catalysts from 95Mo NMR Chemical Shift Descriptors. J. Am. Chem. Soc. 2022, 144 (33), 15020–15025. 10.1021/jacs.2c06252. [DOI] [PubMed] [Google Scholar]
  42. Hillenbrand J.; Leutzsch M.; Gordon C. P.; Copéret C.; Fürstner A. 183W NMR Spectroscopy Guides the Search for Tungsten Alkylidyne Catalysts for Alkyne Metathesis. Angew. Chem., Int. Ed. 2020, 59, 21758–21768. 10.1002/anie.202009975. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Thompson R. R.; Rotella M. E.; Zhou X.; Fronczek F. R.; Gutierrez O.; Lee S. Impact of Ligands and Metals on the Formation of Metallacyclic Intermediates and a Nontraditional Mechanism for Group VI Alkyne Metathesis Catalysts. J. Am. Chem. Soc. 2021, 143 (24), 9026–9039. 10.1021/jacs.1c01843. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Rütter D.; van Gastel M.; Leutzsch M.; Nöthling N.; SantaLucia D.; Neese F.; Fürstner A. Molybdenum(VI) Nitrido Complexes with Tripodal Silanolate Ligands. Structure and Electronic Character of an Unsymmetrical Dimolybdenum μ-Nitrido Complex Formed by Incomplete Nitrogen Atom Transfer. Inorg. Chem. 2024, 63 (18), 8376–8389. 10.1021/acs.inorgchem.4c00762. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Fürstner A. Alkyne Metathesis on the Rise. Angew. Chem., Int. Ed. 2013, 52 (10), 2794–2819. 10.1002/anie.201204513. [DOI] [PubMed] [Google Scholar]
  46. Fürstner A. The Ascent of Alkyne Metathesis to Strategy-Level Status. J. Am. Chem. Soc. 2021, 143 (38), 15538–15555. 10.1021/jacs.1c08040. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Hillenbrand J.; Korber J. N.; Leutzsch M.; Nöthling N.; Fürstner A. Canopy Catalysts for Alkyne Metathesis: Investigations into a Bimolecular Decomposition Pathway and the Stability of the Podand Cap. Chem. - Eur. J. 2021, 27 (56), 14025–14033. 10.1002/chem.202102080. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. [(tBuO)3MoMo(OtBu)3] resonates at 2645 ppm, cf. ref (72)
  49. For a remotely related case in which low-valent Mo was shown to engage one phenyl ring of a para-terphenyl diphosphine in different binding modes (η2, η4, η6) depending on the formal oxidation state (+2, 0, −2, −3) and the chosen ancillary ligands, see:; a Buss J. A.; Edouard G. A.; Cheng C.; Shi J.; Agapie T. Molybdenum Catalyzed Ammonia Borane Dehydrogenation: Oxidation State Specific Mechanisms. J. Am. Chem. Soc. 2014, 136 (32), 11272–11275. 10.1021/ja5059923. [DOI] [PMC free article] [PubMed] [Google Scholar]; b Buss J. A.; Agapie T. Four-electron deoxygenative reductive coupling of carbon monoxide at a single metal site. Nature 2016, 529 (7584), 72–75. 10.1038/nature16154. [DOI] [PubMed] [Google Scholar]
  50. Examples from the lanthanide and actinide series have recently been disclosed by other laboratories, see:; a Keener M.; Shivaraam R. A. K.; Rajeshkumar T.; Tricoire M.; Scopelliti R.; Zivkovic I.; Chauvin A.-S.; Maron L.; Mazzanti M. Multielectron Redox Chemistry of Uranium by Accessing the + II Oxidation State and Enabling Reduction to a U(I) Synthon. J. Am. Chem. Soc. 2023, 145 (29), 16271–16283. 10.1021/jacs.3c05626. [DOI] [PubMed] [Google Scholar]; b Guo H.; Hong D.; Cui P. Tripodal tris(siloxide) ligand supported trivalent rare-earth metal complexes and redox reactivity. Dalton Trans. 2023, 52 (43), 15672–15676. 10.1039/D3DT02519E. [DOI] [PubMed] [Google Scholar]; c Hsueh F.-C.; Rajeshkumar T.; Maron L.; Scopelliti R.; Sienkiewicz A.; Mazzanti M. Isolation and redox reactivity of cerium complexes in four redox states. Chem. Sci. 2023, 14 (22), 6011–6021. 10.1039/D3SC01478A. [DOI] [PMC free article] [PubMed] [Google Scholar]; d Hsueh F.-C.; Chen D.; Rajeshkumar T.; Scopelliti R.; Maron L.; Mazzanti M. Two-Electron Redox Reactivity of Thorium Supported by Redox-Active Tripodal Frameworks. Angew. Chem., Int. Ed. 2024, 63 (6), e202317346 10.1002/anie.202317346. [DOI] [PubMed] [Google Scholar]
  51. Because of the paramagnetic nature of [12·thf], the course of the reaction could not be monitored by NMR. It is therefore not clear if the homodimerization is just exceptionally slow or if degradation of the highly sensitive [12·thf] is limiting
  52. Compare the computed bond lengths of hypothetical ″Cummins dimers″ of type [(Ar)(R)N]3Mo≡Mo[N(R)(Ar)]3: 2.367 Å for R = iPr; for R = tBu, however, the hypothetical distance (2.466 Å) is still much larger, see:; Christian G.; Stranger R.; Yates B. F.; Cummins C. C. Rationalizing the different products in the reaction of N2 with three-coordinate MoL3 complexes. Dalton Trans. 2007, 1939–1947. 10.1039/b701050h. [DOI] [PubMed] [Google Scholar]
  53. This interpretation is corroborated by the fact that the structure of 17 in the solid state shows such an orientation as evident form Figure S7 contained in the Supporting Information
  54. Liptrot D. J.; Power P. P. London Dispersion Forces in Sterically Crowded Inorganic and Organometallic Molecules. Nat. Rev. Chem. 2017, 1, 0004 10.1038/s41570-016-0004. [DOI] [Google Scholar]
  55. Wagner J. P.; Schreiner P. R. London Dispersion in Molecular Chemistry – Reconsidering Steric Effects. Angew. Chem., Int. Ed. 2015, 54, 12274–12296. 10.1002/anie.201503476. [DOI] [PubMed] [Google Scholar]
  56. For catalyst design by our group that makes proactive use of London dispersion, see:; a Singha S.; Buchsteiner M.; Bistoni G.; Goddard R.; Fürstner A. A New Ligand Design Based on London Dispersion Empowers Chiral Bismuth–Rhodium Paddlewheel Catalysts. J. Am. Chem. Soc. 2021, 143 (15), 5666–5673. 10.1021/jacs.1c01972. [DOI] [PMC free article] [PubMed] [Google Scholar]; b Buchsteiner M.; Singha S.; Decaens J.; Fürstner A. Chiral Bismuth-Rhodium Paddlewheel Complexes Empowered by London Dispersion: The C–H Functionalization Nexus. Angew. Chem., Int. Ed. 2022, 61 (45), e202212546 10.1002/anie.202212546. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Gilbert T. M.; Bauer C. B.; Bond A. H.; Rogers R. D. Structures of homoleptic triply bonded M2(OR)6 compounds where the alkoxide is tertiary: the effect of steric bulk and alkoxide conformation on structural parameters. Polyhedron 1999, 18 (8), 1293–1301. 10.1016/S0277-5387(98)00433-1. [DOI] [Google Scholar]
  58. Korber J. N.; Wille C.; Leutzsch M.; Fürstner A. From the Glovebox to the Benchtop: Air-Stable High Performance Molybdenum Alkylidyne Catalysts for Alkyne Metathesis. J. Am. Chem. Soc. 2023, 145, 26993–27009. 10.1021/jacs.3c10430. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Heppekausen J.; Stade R.; Goddard R.; Fürstner A. Practical New Silyloxy-Based Alkyne Metathesis Catalysts with Optimized Activity and Selectivity Profiles. J. Am. Chem. Soc. 2010, 132, 11045–11057. 10.1021/ja104800w. [DOI] [PubMed] [Google Scholar]
  60. Heppekausen J.; Stade R.; Kondoh A.; Seidel G.; Goddard R.; Fürstner A. Optimized Synthesis, Structural Investigations, Ligand Tuning and Synthetic Evaluation of Silyloxy-Based Alkyne Metathesis Catalysts. Chem. - Eur. J. 2012, 18, 10281–10299. 10.1002/chem.201200621. [DOI] [PubMed] [Google Scholar]
  61. Only traces of free aniline (tBu)(Ar)NH were detected by 1H NMR
  62. Chisholm M. H.; Reichert W. Hexakis(dimethyl-aminato)- and hexakis(tert-butoxy)dimolybdenum(III). J. Am. Chem. Soc. 1974, 96 (4), 1249–1251. 10.1021/ja00811a063. [DOI] [Google Scholar]
  63. Laplaza C. E.; Johnson A. R.; Cummins C. C. Nitrogen Atom Transfer Coupled with Dinitrogen Cleavage and Mo–Mo Triple Bond Formation. J. Am. Chem. Soc. 1996, 118 (3), 709–710. 10.1021/ja953573y. [DOI] [Google Scholar]
  64. Chisholm M. H.; Corning J. F.; Huffman J. C. The molybdenum-molybdenum triple bond. 12. Preparation and characterization of bis(tert-butyl mercaptido)tetrakis(dimethyl-amido)dimolybdenum(M≡M). Inorg. Chem. 1983, 22 (1), 38–40. 10.1021/ic00143a010. [DOI] [Google Scholar]
  65. Burroughs B. A.; Bursten B. E.; Chen S.; Chisholm M. H.; Kidwell A. R. Metathesis of Nitrogen Atoms within Triple Bonds Involving Carbon, Tungsten, and Molybdenum. Inorg. Chem. 2008, 47 (12), 5377–5385. 10.1021/ic8003917. [DOI] [PubMed] [Google Scholar]
  66. Chen J.-H.; Deng C.-H.; Fang S.; Ma J.-G.; Cheng P. Binuclear molybdenum alkoxide as the versatile catalyst for the conversion of carbon dioxide. Green Chem. 2018, 20 (5), 989–996. 10.1039/C7GC03372A. [DOI] [Google Scholar]
  67. Hillenbrand J.; van Gastel M.; Bill E.; Neese F.; Fürstner A. Isolation of a Homoleptic Non-oxo Mo(V) Alkoxide Complex: Synthesis, Structure, and Electronic Properties of Penta-tert-Butoxymolybdenum. J. Am. Chem. Soc. 2020, 142 (38), 16392–16402. 10.1021/jacs.0c07073. [DOI] [PMC free article] [PubMed] [Google Scholar]
  68. The exceptional reactivity of this putative intermediate has previously been inferred from the observation that its likely presence greatly enhances the rate of dinitrogen cleavage by complex 1, see ref (63)
  69. Malito J. Molybdenum-95 NMR Spectroscopy. Annu. Rep. NMR Spectrosc. 1996, 33, 151 10.1016/S0066-4103(08)60050-2. [DOI] [Google Scholar]
  70. Mann B. E. The Cinderella Nuclei. Annu. Rep. NMR Spectrosc. 1991, 23, 141–207. 10.1016/S0066-4103(08)60277-X. [DOI] [Google Scholar]
  71. For recent use of 95Mo NMR by our group as a sensitive tool for the study of molybdenum alkylidynes, see ref (37)(38)(41)(44),
  72. Young C. G.; Kober E. M.; Enemark J. H. 95Mo and 183W NMR Studies of Triply Bonded Dinuclear M(III) and Related M≡C (M = Mo or W) Complexes. Polyhedron 1987, 6, 255–259. 10.1016/S0277-5387(00)80794-9. [DOI] [Google Scholar]
  73. The attribution is tentaive and solely based on the comparison with analogous homoleptic complexes that show the same trends regarding the line widths; for instructive examples, see ref (44)
  74. Chisholm M. H.; Cotton F. A.; Frenz B. A.; Reichert W. W.; Shive L. W.; Stults B. R. The molybdenum-molybdenum triple bond. 1. Hexakis(dimethylamido)dimolybdenum and some homologs: preparation, structure, and properties. J. Am. Chem. Soc. 1976, 98 (15), 4469–4476. 10.1021/ja00431a024. [DOI] [Google Scholar]
  75. Bittner C.; Ehrhorn H.; Bockfeld D.; Brandhorst K.; Tamm M. Tuning the Catalytic Alkyne Metathesis Activity of Molybdenum and Tungsten 2,4,6-Trimethylbenzylidyne Complexes with Fluoroalkoxide Ligands OC(CF3)nMe3–n. Organometallics 2017, 36, 3398–3406. 10.1021/acs.organomet.7b00519. [DOI] [Google Scholar]
  76. Chapovetsky A.; Langeslay R. R.; Celik G.; Perras F. A.; Pruski M.; Ferrandon M. S.; Wegener E. C.; Kim H.; Dogan F.; Wen J.; Khetrapal N.; Sharma P.; White J.; Kropf A. J.; Sattelberger A. P.; Kaphan D. M.; Delferro M. Activation of Low-Valent, Multiply M–M Bonded Group VI Dimers toward Catalytic Olefin Metathesis via Surface Organometallic Chemistry. Organometallics 2020, 39 (7), 1035–1045. 10.1021/acs.organomet.9b00787. [DOI] [Google Scholar]
  77. For a detailed early analysis of truncated model complexes reaching similar conclusions, see:; Bursten B. E.; Cotton F. A.; Green J. C.; Seddon E. A.; Stanley G. G. Molecular orbital and spectro-scopic studies of triple bonds between transition-metal atoms. 1. The d3-d3 Mo2L6 compounds (L = OR, NR2, CH2R). J. Am. Chem. Soc. 1980, 102 (14), 4579–4588. 10.1021/ja00534a003. [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

ja5c02178_si_001.pdf (6.7MB, pdf)
ja5c02178_si_002.pdf (6.1MB, pdf)
ja5c02178_si_003.pdf (1.6MB, pdf)

Articles from Journal of the American Chemical Society are provided here courtesy of American Chemical Society

RESOURCES