Synthesis and structures of molecular beryllium Grignard analogues featuring terminal and bridging pseudohalides

Abstract

The carbene-stabilised beryllium Grignards [(CAAC)BeBrR] (R = CAACH 1a, Dur 1b; CAAC/H = 1-(2,6-diisopropylphenyl)-2,2,4,4-tetramethylpyrrolidin-2-yl/idene; Dur = 2,3,5,6-tetramethylphenyl) undergo salt metathesis with various pseudohalide salt precursors. Whereas with [NaNCS] the thiocyanato Grignards [(CAAC)Be(NCS)R] (R = CAACH 2a, Dur 2b) are obtained selectively, salt metatheses with [Na(OCP)(dioxane)_2.3] and [K(OCN)] are fraught with side reactions, in particular scrambling of both neutral and anionic ligands, leading to complex product mixtures, from which the first examples of beryllium phosphaethynolate Grignards [(thf)₂(CAACH)Be(OCP)] (3) and [(CAAC)Be(OCP)R] (R = CAACH 4a, Dur 4b), as well as the isocyanate-bridged hexamer [(CAAC)BrBe(1,3-μ-OCN)]₆ (7) were determined as the main products. The complexity of possible side reactions is seen in complex 5, a byproduct of the salt metathesis of 1b with [Na(OCP)(dioxane)_2.3], which hints at radical redox processes, OCP homocoupling, OCP coupling with CAAC, as well as OCP insertion into the Be–R bond. Finally, the unstable, tetrameric cyano-bridged beryllium Grignard [(thf)(CAACH)Be(1,2-μ-CN)] (8) was obtained by salt metathesis of 1a with [Na/KSeCN] alongside one equiv. CAAC=Se. The new complexes were characterised by heteronuclear NMR and IR spectroscopy, as well X-ray crystallography.

---

Introduction

Named a century ago after their chemical resemblance with the halides,¹ pseudohalides have since found applications ranging from organic cross-coupling or carboxylation reagents to ionic liquids and solar cell materials.^2–7 Linear pseudohalogen anions, such as CN⁻, N₃⁻ or NCO⁻, are ambiphilic and thus capable of coordinating via both ends. As a result, the homoleptic alkaline earth pseudohalides, [AeY₂], are usually oligomers featuring bridging pseudohalides.⁸

Whereas the chemistry of classical and heavier Grignard reagents, [RAeX] (R = organyl, X = halide), is one of the mainstays of organic and inorganic chemistry,^9–13 that of their pseudohalide derivatives is virtually unknown. The difficulties in isolating well-defined molecular group 2 Grignard analogues of the form [L_nAeRY] (L = neutral donor), featuring a terminal unsymmetrical pseudohalide ligand (e.g. CN⁻, NCO⁻, NCS⁻, OCP⁻) are threefold: (1) the ambidentate nature of unsymmetrical pseudohalides, frequently resulting in the formation of mixtures of linkage isomers or oligomers,¹⁴ (2) the tendency of heteroleptic group 2 complexes to undergo Schlenk-type ligand scrambling,¹⁵ which results in the often irreversible formation of homoleptic compounds, and (3) the propensity of some unsymmetrical pseudohalides, like CN or NCO, to undergo unwanted nucleophilic addition reactions.¹⁶

In 1974, Klopsch and Dehnicke were the first to synthesise dianionic organomagnesium pseudohalides of the form [{Me₂Mg}₂(μ-Y)]²⁻ and [{Me₂Mg(η¹-Y)}₂]²⁻ (Y = CN, N₃, NCO), as determined by IR spectroscopy.¹⁷ In 1998 Böhland reported the first neutral thiocyanate and selenocyanate Grignards [(thf)_nMg(NCE)R] (I^R-E, R = Et, Ph, E = S, Se, Fig. 1A), which undergo Schlenk equilibration to the homoleptic species in solution.¹⁸ Much more recently, Gilliard isolated the first phosphaethynolate Grignard, II, stabilised against ligand redistribution by two bulky N-heterocyclic carbenes (NHCs).¹⁹ To date, complex II is the only structurally characterised pseudohalide Grignard analogue.

Fig. 1 Examples of pseudohalide Grignards and molecular beryllium pseudohalides. Dip = 2,6-diisopropylphenyl, Dur = 2,3,5,6-tetramethylphenyl (= duryl), CAAC = 1-(2,6-diisopropylphenyl)-2,2,4,4-tetramethylpyrrolidin-5-ylidene.

Long neglected due to the high toxicity of its compounds,‡ the last decade has seen a revival of organoberyllium chemistry,^20–22 not least because of the landmark syntheses of the first Be(0) and Be(I) complexes.^23–27 As the least polarisable and smallest of the group 2 dications, Be²⁺ forms largely covalent bonds with carbon. As a result, beryllium-based Grignard complexes tend to be more stable than their magnesium analogues towards Schlenk equilibration.^28,29 Our group recently reported the first beryllium-based pseudohalide Grignards, the cyclic alkyl(amino)carbene (CAAC)-stabilised terminal azide complexes III^R and the imine-stabilised azide-bridged dimer IV (Fig. 1B).³⁰ Beyond this, only a handful of molecular beryllium pseudohalides have been reported, including tetrahedral homoleptic terminal cyanide, azide and thiocyanate complexes of the form [L₂BeY₂] (L = pyridine (V-Y), amine, imine, NHC (VI)),^30–32 as well as a heteroleptic diamine-supported chloroberyllium azide (VII),³³ and the tris(pyrazolyl)borate (Tp) complexes [(Tp)BeY] (VIII-Y, Y = CN, N₃, NCS).^34,35

In this study we report the synthesis, spectroscopic and structural characterisation of a range of monomeric and oligomeric Lewis-base-stabilised beryllium pseudohalide Grignards, including the first structurally characterised beryllium phosphaethynolates and isocyanates.

Results and discussion

Synthesis of beryllium pseudohalides

The dropwise addition of 1 equiv. [NaNCS] in THF to 1a²⁶ and 1b³⁶ in THF at −78 °C, followed by stirring for 3 h at room temperature yielded, after workup, the corresponding beryllium thiocyanates 2a and 2b as yellow (75%) and colourless (58%) solids, respectively (Scheme 1a). Their ⁹Be NMR spectra show a broad resonance at 15.5 (ω_1/2 (full width at half maximum) ≈ 420 Hz) and 14.8 (ω_1/2 ≈ 210 Hz) ppm, respectively, ca. 5 ppm upfield-shifted from that of their respective precursors (δ_⁹Be = 20.0 (1a), 18.7 (1b) ppm). Their ¹³C NMR C_carbene resonances at 249.6 and 240.1 ppm, respectively, are virtually identical to that of their respective precursors (δ_¹³C = 249.7 (1a), 240.1 (1b) ppm). The ¹³C NMR resonance of the NCS moiety could not be detected, presumably owing to excessive broadening caused by strong coupling with the nearby quadrupolar ⁹Be and ¹⁴N nuclei. Complexes 1a and 1b represent the first examples of tricoordinate beryllium thiocyanates, previous examples being limited to tetracoordinate neutral³⁵ or ionic complexes.³⁷ The solid-state IR spectra of 2a and 2b show a single, very intense and relatively broad N=C stretching band at 2100 and 2089 cm⁻¹, respectively, in the same range as other molecular and ionic terminal beryllium thiocyanates (ν(N=C) = 2065–2117 cm⁻¹).^35,37 Compounds 2a/b were stable in benzene up to 80 °C.

Scheme 1 Synthesis of beryllium isothiocyanates and phosphaethynolates. Isolated yields in parentheses.

The ⁹Be and ³¹P NMR spectra of the reaction mixture of 1a with [Na(OCP)(dioxane)_2.3] showed the formation of two main products in a 96 : 4 ratio. Workup revealed the major one of these to be the colourless tetracoordinate complex [(thf)₂(CAACH)Be(OCP)], present as a mixture of two isomers, 3 and 3′, in an 88 : 12 ratio, respectively (δ_⁹Be = 7.1 (s, ω_1/2 ≈ 150 Hz, 3′), 4.7 (s, ω_1/2 ≈ 145 Hz, 3) ppm; δ_³¹P = −342.6 (s) ppm, Scheme 1b).§ Recrystallisation from pentane also yielded a few crystals of the expected tricoordinate CAAC-stabilised species 4a (δ_³¹P = −339.4 ppm) as determined by X-ray crystallography (vide infra).¶ The ¹H NMR spectrum of the major rotamer, 3, confirms the presence of two THF ligands, with 8H multiplets at 3.41 and 1.90 ppm, as well as the chiral CAACH ligand with its characteristic 1H singlet at 2.68 ppm for the BeCH resonance. The terminal nature and Be–O bonding of the OCP ligand is attested by a broad ¹³C{¹H} NMR doublet at 156.6 (¹J_C–P = 13.3 Hz, OCP) and a ³¹P{¹H} singlet at −342.6 (OCP) ppm. These are upfield- and downfield-shifted, respectively, from the range of ¹³C{¹H} and ³¹P{¹H} NMR OCP resonances observed for molecular terminal magnesium phospaethynolates (δ_¹³C = 161–169 ppm, δ_³¹P = −365–(−386) ppm).^19,38,39 This may be rationalised by the much higher charge density of the Be²⁺ dication, which draws electron density away from the terminal phosphorus atom, thus favouring the O–C≡P over the O⁺=C=P⁻ resonance structure. Furthermore, magnesium phospaethynolates display significantly larger ¹³C–³¹P coupling constants (¹J_C–P = 25–60 Hz),^19,38,39 the smallest one being observed for complex II (see Fig. 1A).¹⁹ Similarly small OCP ¹³C–³¹P coupling constants to that of 3 have been observed in iPr₃Si(OCP),⁴⁰ a β-diketiminate-stabilised scandium OCP complex (both ¹J_C–P = 10 Hz),⁴¹ and an OCP-substituted 1,3-dihydrodiazaborole (¹J_C–P = 17 Hz).⁴² Smaller ¹³C–³¹P coupling constants are associated with a higher degree of phosphaethynolate O–C≡P bonding, whereas larger ones are indicative of increased phosphaketene O=C=P bonding,⁴³ thereby confirming that the Be–O–C≡P resonance structure is highly dominant in 3. This was also confirmed by IR spectroscopy of 3/3′, which shows an intense and relatively broad asymmetric OCP stretching band at 1728 cm⁻¹, in the lower range of those observed for terminal magnesium phosphaethynolates (ν(OCP)_asym = 1727–1770 cm⁻¹)^19,38,39 and similar to that of [(18-crown-6)K(OCP)] (ν(OCP)_asym = 1730 cm⁻¹).⁴⁴ Compound 3/3′ was stable in benzene up to 60 °C but decomposed unselectively upon UV irradiation.

The reaction of 1b with [Na(OCP)(dioxane)_2.3] proved much less selective. The ⁹Be NMR spectrum of the crude reaction mixture after 30 min at room temperature evidenced the formation of a least two tricoordinate (δ_⁹Be = 17.4 (br), 14.1 (br) ppm) and two tetracoordinate (δ_⁹Be = 4.7 (s), 1.8 (s) ppm) beryllium complexes.⁴⁵ The two main products observed by ³¹P NMR spectroscopy feature a sharp singlet at −328.6 (ca. 46%) ppm, which fits a phosphaethynolate complex, and a very broad resonance at −343.1 ppm (ω_1/2 ≈ 320 Hz, ca. 36%), suggesting Be–P bonding, respectively, alongside several other minor unidentified products. Furthermore, the unexpected intense colouration of the reaction mixture (first dark purple, then dark brown), usually the hallmark of radical CAAC complexes,^23,26,27,46 prompted us to measure a solution EPR spectrum. The latter indeed showed the presence of at least three radical species, none of which could be identified further, but which highlight that this seemingly simple salt metathesis is accompanied by complex redox processes.

Recrystallisation of the crude product mixture by diffusion of hexane into a saturated benzene solution yielded a few crystals of the tricoordinate beryllium phosphaethynolate 4b, which was identified as the major reaction product (δ_⁹Be = 14.1 (br, ω_1/2 ≈ 470 Hz) ppm; δ_³¹P = −328.6 (s) ppm).||4b systematically cocrystallised with 30% of another tetracoordinate beryllium complex (6, δ_⁹Be = 1.4 (br, ω_1/2 ≈ 280 Hz) ppm; δ_³¹P = −339.6 (s) ppm), containing a 1 : 1 ratio of Dur and terminal OCP substituents, as determined by ¹H and ³¹P NMR spectroscopy. Unfortunately, the constitution of this complex could not be elucidated further. Further crystallisation yielded a few crystals suitable for X-ray diffraction analysis of an unexpected side-product, complex 5 (see Fig. S30 in the ESI†), which evidences the insertion of an OCP fragment into the Be–Dur bond, coupling of two PCO ligands via P–C bond formation, as well as C_CAAC–C_PCO and C_CAACH–P_PCO bond formation, the CAACH residue suggesting the involvement of a CAAC radical and hydrogen atom abstraction.

The lack of selectivity of the reaction between 1b and [Na(OCP)(dioxane)_2.3] reflects the multitude of possible side-reactions when attempting to synthesise main group phosphaethynolates, including ether cleavage of the dioxane ligands of [Na(OCP)(dioxane)_2.3],¹⁹ OCP coupling,^42,47 CAAC–PCO coupling,^48,49 aryl transfer to the phosphorus atom from an s-block aryl reagent,^42,50 photoinduced CO extrusion with radical P–P coupling,^42,51,52 or photoinduced [PCO]˙ extrusion.⁵¹ Similar light-induced redox processes might be at play here and responsible for the radical species observed by EPR spectroscopy.

Attempts to obtain an isocyanate derivative of 1a by salt metathesis with [KOCN] at 60 °C yielded a mixture of four tetracoordinate beryllium complexes (δ_⁹Be = 8.8, 7.0, 4.0, 2.0 ppm) in various ratios depending on the reaction time. Multiple attempts to improve the selectivity of the reaction by changing equivalents, solvent, temperature and reaction time failed. Recrystallisation of the crude product mixture from benzene yielded a few colourless single crystals of the hexameric complex 7, which is comprised of six [(CAAC)BeBr(OCN)] units, the OCN ligands bridging between two adjacent beryllium centres, thus forming a 24-membered (BeOCN)₆ ring (Scheme 2a). It is noteworthy that 7 has lost the CAACH substituent but still contains an unreacted bromide substituent, suggesting the involvement of temperature- and/or sterics-induced Schlenk-type equilibria, which have been observed in beryllium Grignard analogues, albeit to a lesser degree than their magnesium counterparts.^28,29 Attempts to obtain 7 more selectively by salt metathesis with (CAAC)BeBr₂, however, proved unsuccessful. Taking into account potential ligand exchange processes and steric effects, the other products observed by ⁹Be NMR spectroscopy are likely to be either similarly OCN-bridged hexamers of the form [LBeY(1,3-μ-OCN)]₆ (L = CAAC, Y = η¹-OCN; L = THF, Y = CAACH), or dimeric [(THF)Be(OCN)]₂, by analogy to Buchner's [(Me₂S)Be(η¹-OCN)(μ-OCN)]₂ (δ_⁹Be = 7.0 ppm),³⁵ or mononuclear terminal beryllium isocyanates of the form [(THF)₂BeY(η¹-OCN)] (Y = CAACH, η¹-OCN), akin to complex 3 and VIII-OCN (see Fig. 1B).³⁵

Scheme 2 Synthesis of beryllium isocyanate and cyanoberyllium oligomers.

Finally, the salt metathesis of 1a with the heavier isocyanate analogues [NaSeCN] and [KSeCN] proceeded selectively at room temperature in THF under elimination of CAAC=Se,⁵³ to yield a single beryllium-containing product with a broad ⁹Be NMR resonance at 3.4 ppm (ω_1/2 ≈ 440 Hz). Recrystallisation from pentane at −30 °C yielded light brown crystals of the tetramer 8, which is comprised of four [(THF)Be(CAACH)(CN)] units, the CN ligands bridging between two adjacent beryllium centres, thus forming a 12-membered (BeCN)₄ ring (Scheme 2b). The ⁹Be NMR shift of 8 is slightly downfield-shifted from those observed for the coordination isomers VIII-CN and VIII-NC at 1.9 and 1.6 ppm, respectively.³⁵ Complex 8 proved extremely sensitive, even decomposing on the diffractometer at −173 °C. A full in-situ NMR-spectroscopic characterisation of the reaction mixture was marred by the fact that all four CN linkers of the (BeCN)₄ ring are flip-disordered, yielding a complex mixture of isomers. Attempts to obtain a cyanoberyllium complex more directly by salt metathesis of 1a with [NaCN] or [KCN] required higher reaction temperatures (60–80 °C) and led to intractable product mixtures.

X-ray crystallographic analyses

X-ray diffraction analyses provided the solid-state structures of 2a, 2b, 3, 4a, 4b, 5, 7 and 8. Due to excessive twinning, poor diffraction and rapid crystal decomposition the data for 8 is of insufficient quality to be discussed but provide conclusive proof of connectivity (Fig. S32 in the ESI†). Selected structural data for 2a, 2b, 3, 4a, 4b and 7 are provided in Table 1, together with relevant NMR- and IR-spectroscopic data.

Table 1 ⁹Be NMR shifts (δ_⁹Be, ppm) with their full widths at half-maximum (ω_1/2, Hz), pseudohalide IR stretching frequencies (ν_ps, cm⁻¹), as well as selected bond lengths (Å), bond angles (°) and absolute values of torsion angles (°) for the beryllium complexes presented herein

	1a ^,	1b ^,	3 ^, ^,	4a ^, ^,	4b ^,	7 ^,
a YCE = NCS. b Z = N3. c Z = C4. d YCE = OCP. e Data for the only molecule of 3 in the asymmetric unit (out of 3), which does not display a disorder in the BeOCP moiety. f 4a cocrystallised with 4% of leftover starting material 1a, both molecules overlapping fully in the asymmetric unit, except for their Br/OCP ligands, which were freely refined, leading to relatively high esds. g YCE = OCN. h Structural data ranges provided for all three (CAAC)BBr(OCN) units present in the asymmetric unit. i n.d. = not determined.
δ ^9Be	15.5	14.8	4.7	n.d.^i	17.0	n.d.
ω 1/2	420	210	150	n.d.	340	n.d.
νps	2100	2089	1727	n.d.	1691	n.d.
N1–C1	1.3011(15)	1.2967(16)	—	1.302(2)	1.2998(18)	1.292(4)–1.296(4)
N2–C3	1.4833(15)	—	1.5151(17)	1.486(2)	—	—
Be–C1	1.811(2)	1.823(2)	—	1.813(3)	1.810(2)	1.827(5)–1.836(5)
Be–C3	1.7655(19)	1.743(2)	1.776(2)	1.763(3)	1.745(2)	—
Be–Y	1.6186(19)	1.6197(19)	1.626(2)	1.534(3)	1.557(2)	1.665(4)–1.667(4)
Y–C2	1.1674(17)	1.1740(17)	1.224(2)	1.273(5)	1.2626(19)	1.231(3)–1.234(3)
C2–E	1.6080(14)	1.6100(14)	1.5841(17)	1.531(5)	1.5575(16)	1.102(4)–1.138(4)
E–Be	—	—	—	—	—	1.693(4)–1.712(5)
max. (YCE)⋯Beplane	ca. 0.28	ca. 0.40	—	ca. 0.38	ca. 0.41
Be–Y–C2	170.42(12)	169.84(12)	144.73(13)	159.8(4)	132.99(13)	130.1(2)–131.6(2)
Y–C2–E	179.29(12)	178.73(13)	179.45(18)	177.5(6)	177.45(14)	177.3(3)–178.0(3)
C2–E–Be	—	—	—	—	—	167.1(3)–168.2(3)
Σ∠Be	359.99(11)	359.90(11)	—	359.94(16)	359.98(13)	—
|N1–C1–Be–Y|	110.1(1)	15.40(18)	—	113.43(19)	28.9(2)	16.9(4)–20.1(4)
|Z–C3–Be–Y|	63.8(2)	78.2(2)	51.3(2)	62.1(2)	114.3(2)	—
|N–C–Be–E|	—	—	—	—	—	96.7(3)–99.8(3)

---

Complexes 2a, 2b, 4a and 4b (Fig. 2) are trigonal-planar (Σ∠Be ca. 360°) and display a terminal pseudohalide ligand. The Be–C1 bond length to the neutral CAAC ligand (1.810(2)–1.823(2) Å) is similar to that in the precursors 1a and 1b (1.798(2), 1.829(2) Å).^26,36 The CAAC and CAACH ligands in 2a and 4a adopt an anti conformation relative to each other so as to minimise steric repulsion. The duryl substituent in 2b is near-orthogonal to the beryllium plane (|C4–C3–Be1–N2| 78.2(2)°), but less so in 4b (|C4–C3–Be1–O1| 62.1(2)°), perhaps to minimise repulsion between the π systems of the duryl ring and the C≡P moiety. The NCS and OCP moieties in 2a/b and 4a/b, respectively, approach linearity (N–C–S/O–C–P 177.45(14)–179.45(18)°) and are quasi-coplanar with the beryllium plane (S1/P1⋯Be_plane < 0.41 Å).

Fig. 2 Crystallographically determined solid-state structures of 1a, 1b, 3, 4a and 4b. Atomic displacement ellipsoids at 50% probability. Minor disordered parts, lattice solvent molecules, ellipsoids of ligand periphery, and hydrogen atoms omitted for clarity, except for CAACH protons.

The steric constraints imposed by the anionic CAACH/Dur substituents have a significant impact on the orientation of the pseudohalide substituent relative to the CAAC ligand. In the duryl derivatives the NCS/OCP ligand tends towards coplanarity with the trigonal-planar CAAC carbene centre (2b |N1–C1–Be1–N2| 15.40(18); 4b |N1–C1–Be1–O1| 28.9(2)°) in order to maximise overlap between the pseudohalide and CAAC π systems. In the CAACH derivatives, however, the NCS/OCP ligand tends more towards orthogonality with the C1 plane as the much higher steric constraints override electronic effects (2a |N1–C1–Be1–N2| 110.1(1); 4a |N1–C1–Be1–O1| 113.43(19)°).

Complexes 2a/b are the first solid-state structures of tricoordinate beryllium thiocyanates. Accordingly, their Be1–N1 bonds (1.3011(15), 1.2967(16) Å) are necessarily shorter than those in the known tetracoordinate complexes (ca. 1.65–1.69 Å).^35,37 The Be–N2–C2 angle in 2a/b deviates only slightly from linearity (ca. 170°), the NCS residue bending away from the CAAC and towards the CAACH/Dur ligand. The N2–C2 (1.1674(17), 1.1740(17) Å) and C2–S1 (1.6080(14), 1.6100(14) Å) bond lengths are typical for terminal beryllium thiocyanates (N–C ca. 1.14–1.17; C–S ca. 1.59–1.62 Å).^35,37

Complexes 3, 4a and 4b are the first examples of beryllium phosphaethynolates. The Be1–O1 bond lengths range from 1.534(3) to 1.626(2) Å, being longer in the tetrahedral complex 3 than in the trigonal-planar complexes 4a/b. Furthermore, as Be1–O1 lengthens, O1–C2 shortens from 1.273(5) to 1.224(2) Å, while C2–P1 lengthens from 1.531(5) to 1.5841(17) Å, indicating small fluctuation in the degree of O=C=P character in the predominantly O–C≡P resonance form. In the only four structurally characterised terminal s-block phosphaethynolates the O–C bonds (1.196(3)–1.207(4) Å) are significantly shorter, while the C–P bonds (1.555(4)–1.582(3) Å) fall within the same range as those in 3 and 4a/b.^37,54,55 Whereas in 4a/b the OCP substituent bends away from the CAAC ligand, the asymmetric unit of 3 contains two distinct topoisomers, one in which the OCP ligand points away from the CAAC ligand, the other towards. This is in agreement with the NMR data, which shows the presence of two rotamers, 3 and 3′, in solution. The Be–O1–C2 angle in 3 and 4a/b is much larger and varies more (132.99(13)–159.8(4)°) than the corresponding Be–N2–C2 angle in 2a/b. In the literature-known s-block phosphaethynolates the M–O–C angle varies even more, from highly bent in [Na(OCP)(dibenzo-18-crown-6)] (138.1(2)°)³⁷ to nearly linear in [Na(OCP)(dme)₂] (107.7(3)°, dme = dimethyoxyethane),⁵⁵ but without apparent correlation with the O–C and C–P bond lengths and thereby the degree of O–C≡P character.

The centrosymmetric hexameric complex 7 crystallises in the chiral monoclinic space group I2/a (Fig. 3). The OCN ligands bridge between the six beryllium centres, forming a central 24-membered (BeOCN)₆ ring, within which the beryllium stereocentres alternate between the R and S configurations. Complex 7 is the first structurally characterised beryllium isocyanate and, to our knowledge, the largest known molecular (MOCN)_n ring structure. The magnesium aluminate trimer [Me₂Al{μ-O(SiMe₃)}₂Mg(μ-OCN)]₃ presents a central 12-membered (MgOCN)₃ ring,⁵⁶** while [(Cp^iPr)U(μ-OCN)]₄ features a 16-membered (UOCN)₄ ring.⁵⁷ The Be–O (ca. 1.67 Å) and Be–N (ca. 1.70 Å) bond lengths are slightly longer than those in the tetrahedral phosphaethynolate 3 (1.626(2) Å) and known terminal beryllium thiocyanates (1.65–1.69 Å),^31–35 respectively. Each O–C–N–Be fragment is nearly linear (O–C–N ca. 178, C–N–Be ca. 168°), whereas the Be–O–C moieties are strongly bent (ca. 131°). The O–C (ca. 1.23 Å) and C–N (ca. 1.12 Å) bond lengths indicate a predominance of the O–C≡N resonance form. The only other structurally characterised s-block complex with an OCN unit bridging between two metal centres is an anionic magnesium porphyrin complex featuring an Mg–O–C–N–K linkage to the K⁺ countercation, which shows significantly more pronounced O=C=N character (O–C 1.175(5), C–N 1.195(5) Å), resulting in much larger Mg–O–C (157.2(3)°) and much smaller C–N–K (123.1(3)°) angles.⁵⁸

Fig. 3 Top: crystallographically determined solid-state structures of 7 (centrosymmetric hexamer). Atomic displacement ellipsoids at 50% probability. Ellipsoids of ligand periphery and hydrogen atoms omitted for clarity. Bottom: structure of the central (BeOCN)₆ ring of 7 (side view).

Finally, the tetrameric complex 8 also crystallises in the chiral monoclinic space group I2/a (Fig. S32 in the ESI†). The CN ligands bridge between the four beryllium centres, forming a central 16-membered (BeCN)₄ ring. This structural motif is relatively common, in particular in group 13 cyanide complexes,^59–62 but has never been reported for s-block complexes. Due to poor diffraction data quality and extensive structural disorders, structural parameters may not be discussed, but the data show that all four bridging cyano ligands are flip-disordered, as was already apparent from the solution NMR data (vide supra). This coordinative CN/NC isomerism is also observed in other structurally characterised cyanoberyllium complexes.^32,35

Conclusions

The synthesis of well-defined beryllium-based pseudohalide Grignards from the CAAC-stabilised beryllium Grignards [(CAAC)(CAACH)BeBr] (1a) and [(CAAC)DurBeBr] (1b) has proven less straightforward than expected. With the exception of the first tricoordinate thiocyanate derivatives, [(CAAC)(CAACH)Be(η¹-NCS)] (2a) and [(CAAC)DurBe(η¹-NCS)] (2b), which are readily accessible by salt metathesis with [NaNCS], the syntheses of phosphaethynolate, isocyanate and cyanide derivatives are complicated by Schlenk-type ligand exchange and other undesirable side reactions. The syntheses of the OCP derivatives, in particular, are fraught with numerous side reactions, including scrambling of both the neutral and anionic ligands, OCP homocoupling, OCP coupling with the CAAC ligand, insertion of OCP into the Be–organyl bond, and ill-defined radical-generating redox processes. Nonetheless, the first examples of beryllium phosphaethynolates, tetracoordinate [(thf)₂(CAACH)Be(η¹-OCP)] (3) and tricoordinate [(CAAC)RBe(η¹-OCP)] (R = CAACH 4a, Dur 4b) were obtained and characterised by NMR and IR spectroscopy, as well as single-crystal X-ray crystallography. While attempts to obtain OCN derivatives also proved very unselective, fractional crystallisation yielded the first solid-state structure of a beryllium isocyanate, a hexameric complex featuring a unique (BeOCN)₆ ring, [(CAAC)BrBe(1,3-μ-OCN)]₆. Finally, attempts to synthesise a beryllium selenocyanate Grignard by salt metathesis with [Na/KSeCN] resulted instead in selenium abstraction by the CAAC ligand and the relatively selective generation of the tetrameric, self-stabilising mixed cyano/isocyanoberyllium Grignard [(thf)(CAACH)Be(1,3-μ-CN/NC)]₄. Overall, this study shows that CAAC ligands, which have enabled the isolation of numerous, otherwise inaccessible low-valent main group complexes,^46,63,64 are ill suited to the stabilisation of pseudohaloberyllium Grignards as they are easily displaced by THF, which is often required to solubilise the pseudohalide salt precursor, and may undergo side reactions with the pseudohalide substituents.

Data availability

Electronic Supplementary Information (ESI) Available: Synthetic procedures, NMR spectra and X-ray crystallographic details.

Crystallographic data have been deposited with the Cambridge Crystallographic Data Center as supplementary publication no. 2379134 (5), 2379135 (2a), 2379136 (3), 2379137 (7), 2379138 (2b), 2379139 (8), 2379140 (4b), 2379141 (4a).

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This project was funded by the Julius-Maximilians-Universität Würzburg and Deutsche Forschungsgemeinschaft (DFG, German Research Foundation; project number 466754611).

References

1. L. Birckenbach and K. Kellermann, Ber. Dtsch. Chem. Ges. B, 1925, 58, 786.
2. F. Bellina and R. Rossi, Chem. Rev., 2010, 110, 1082.
3. M. R. Netherton and G. C. Fu, Adv. Synth. Catal., 2004, 346, 1525.
4. Y.-G. Chen, X.-T. Xu, K. Zhang, Y.-Q. Li, L.-P. Zhang, P. Fang and T.-S. Mei, Synthesis, 2018, 50, 35.
5. S. Arlt, K. Bläsing, J. Harloff, K. C. Laatz, D. Michalik, S. Nier, A. Schulz, P. Stoer, A. Stoffers and A. Villinger, ChemistryOpen, 2021, 10, 62.
6. P.-Y. Lin, A. Loganathan, I. Raifuku, M.-H. Li, Y.-Y. Chiu, S.-T. Chang, A. Fakharuddin, C.-F. Lin, T.-F. Guo, L. Schmidt-Mende and P. Chen, Adv. Energy Mater., 2021, 11, 2100818.
7. F. Cheng, J. Zhang and T. Pauporté, ChemSusChem, 2021, 14, 3665.
8. C. Glidewell, Inorg. Chim. Acta, 1974, 11, 257.
9. S. Harder and J. Langer, Nat. Rev. Chem., 2023, 7, 843.
10. A. Guérinot and J. Cossy, Acc. Chem. Res., 2020, 53, 1351.
11. Y. Nassar, F. Rodier, V. Ferey and J. Cossy, ACS Catal., 2021, 11, 5736.
12. C. E. I. Knappkea and A. Jacobi von Wangelin, Chem. Soc. Rev., 2011, 40, 4948.
13. S. R. Harutyunyan, T. den Hartog, K. Geurts, A. J. Minnaard and B. L. Feringa, Chem. Rev., 2008, 108, 2824.
14. J. L. Burmeister, Coord. Chem. Rev., 1990, 105, 77.
15. A. Tuulmets, M. Mikk and D. Panov, Main Group Met. Chem., 1997, 10, 1.
16. M. Hvastijová, J. Kohout, J. W. Buchler, R. Boča, J. Kožíšek and L. Jäger, Coord. Chem. Rev., 1998, 175, 17.
17. A. Klopsch and K. Dehnicke, Chem. Ber., 1975, 108, 420.
18. H. Böhland, F. R. Hofmann, W. Hanay and H. J. Berner, Z. Anorg. Allg. Chem., 1989, 577, 53.
19. A. D. Obi, H. R. Machost, D. A. Dickie and R. J. Gilliard, Jr., Inorg. Chem., 2021, 60, 12481.
20. D. Parveen, R. K. Yadav and D. K. Roy, Chem. Commun., 2024, 60, 1663.
21. M. R. Buchner, Chem. – Eur. J., 2019, 25, 12018.
22. K. J. Iversen, S. A. Couchman, D. J. D. Wilson and J. L. Dutton, Coord. Chem. Rev., 2015, 297–298, 40.
23. M. Arrowsmith, H. Braunschweig, M. A. Celik, T. Dellermann, R. D. Dewhurst, W. C. Ewing, K. Hammond, T. Kramer, I. Krummenacher, J. Mies, K. Radacki and J. K. Schuster, Nat. Chem., 2016, 8, 890.
24. J. T. Boronski, A. E. Crumpton, L. L. Wales and S. Aldridge, Science, 2023, 380, 1147.
25. J. T. Boronski, A. E. Crumpton, A. F. Roper and S. Aldridge, Nat. Chem., 2024, 16, 1295.
26. C. Czernetzki, M. Arrowsmith, F. Fantuzzi, A. Gärtner, T. Tröster, I. Krummenacher, F. Schorr and H. Braunschweig, Angew. Chem., Int. Ed., 2021, 60, 20776.
27. G. Wang, J. Walley, D. Dickie, A. Molino, D. Wilson and R. J. Gilliard, J. Am. Chem. Soc., 2020, 142, 4560.
28. M. R. Buchner, L. R. Thomas-Hargreaves, C. Berthold, D. F. Bekiş and S. I. Ivlev, Chem. – Eur. J., 2023, 29, e202302495.
29. C. Helling and C. Jones, Chem. – Eur. J., 2023, 29, e202302222.
30. C. Czernetzki, T. Kunz, S. Huynh, A. Lamprecht, J. Sprenger, M. Finze, M. Arrowsmith and H. Braunschweig, Angew. Chem., Int. Ed., 2024, 63, e202401279.
31. H. Böhland, W. Hanay, M. Noltemeyer, A. Meller and H. G. Schmidt, Fresenius’ J. Anal. Chem., 1998, 361, 725.
32. A. V. G. Chizmeshya, C. J. Ritter, T. L. Groy, J. B. Tice and J. Kouvetakis, Chem. Mater., 2007, 19, 5890.
33. B. Neumüller and K. Dehnicke, Z. Anorg. Allg. Chem., 2007, 633, 2262.
34. D. Naglav, D. Bläser, C. Wölper and S. Schulz, Inorg. Chem., 2014, 53, 1241.
35. C. Berthold, M. Müller, S. I. Ivlev, D. M. Andrada and M. R. Buchner, Dalton Trans., 2023, 52, 13547.
36. C. Czernetzki, M. Arrowsmith, L. Endres, I. Krummenacher and H. Braunschweig, Inorg. Chem., 2024, 63, 2670.
37. B. Neumüller and K. Dehnicke, Z. Anorg. Allg. Chem., 2010, 636, 1206.
38. R. J. Gilliard, Jr., D. Heift, J. M. Keiser, Z. Benko, A. L. Rheingold, H. Grützmacher and J. D. Protasiewicz, Dalton Trans., 2018, 47, 666.
39. Z. Li, X. Chen, M. Bergeler, M. Reiher, C.-Y. Su and H. Grützmacher, Dalton Trans., 2015, 44, 6431.
40. D. Heift, Z. Benko and H. Grützmacher, Dalton Trans., 2014, 43, 5920.
41. L. N. Grant, B. Pinter, B. C. Manor, H. Grützmacher and D. J. Mindiola, Angew. Chem., Int. Ed., 2018, 57, 1049.
42. D. W. N. Wilson, A. Hinz and J. M. Goicoechea, Angew. Chem., Int. Ed., 2018, 57, 2188.
43. J. M. Goicoechea and H. Grützmacher, Angew. Chem., Int. Ed., 2018, 57, 16968.
44. A. R. Jupp and J. M. Goicoechea, Angew. Chem., Int. Ed., 2013, 52, 10064.
45. J. K. Buchanan and P. G. Plieger, Z. Naturforsch., B: J. Chem. Sci., 2020, 75, 459.
46. S. Kundu, S. Sinhababu, V. Chandrasekhar and H. W. Roesky, Chem. Sci., 2019, 10, 4727.
47. G. Becker, G. Heckmann, K. Hubler and W. Schwarz, Z. Anorg. Allg. Chem., 1995, 621, 34.
48. W. Yang, K. E. Krantz, D. A. Dickie, A. Molino, D. J. D. Wilson and R. J. Gilliard, Jr., Angew. Chem., Int. Ed., 2020, 59, 3971.
49. B. Li, C. Wölper, H. M. Weinert, H. Siera, G. Haberhauer and S. Schulz, Eur. J. Inorg. Chem., 2024, 27, e202300622.
50. S. J. Urwin and J. M. Goicoechea, Chem. – Eur. J., 2023, 29, e202203081.
51. Y. Xiong, S. Yao, T. Szilvási, E. Ballestero-Martínez, H. Grützmacher and M. Driess, Angew. Chem., Int. Ed., 2017, 56, 4333.
52. D. W. N. Wilson, W. K. Myers and J. M. Goicoechea, Dalton Trans., 2020, 49, 15249.
53. M. Tretiakov, Y. G. Shermolovich, A. P. Singh, A. P. Samuel, H. W. Roesky, B. Niepötter, A. Visschera and D. Stalke, Dalton Trans., 2013, 42, 12940.
54. M. Westerhausen, S. Schneiderbauer, H. Piotrowski, M. Suter and H. Nöth, J. Organomet. Chem., 2002, 189, 643.
55. G. Becker, W. Schwarz, N. Seidler and M. Westerhausen, Z. Anorg. Allg. Chem., 1992, 612, 72.
56. H. Phull, D. Alberti, I. Korobkov, S. Gambarotta and P. H. M. Budzelaar, Angew. Chem., Int. Ed., 2006, 45, 5331.
57. M. A. Boreen, K. N. McCabe, T. D. Lohrey, F. A. Watt, L. Maron, S. Hohloch and J. Arnold, Inorg. Chem., 2020, 59, 8580.
58. K. Ezzayani, A. Ben Khelifa, E. Saint-Aman, F. Loiseau and H. Nasri, Polyhedron, 2016, 117, 817.
59. M. Arrowsmith, D. Auerhammer, R. Bertermann, H. Braunschweig, G. Bringmann, M. A. Celik, R. D. Dewhurst, M. Finze, M. Grüne, M. Hailmann, T. Hertle and I. Krummenacher, Angew. Chem., Int. Ed., 2016, 55, 14464.
60. M. Molon, K. Dilchert, C. Gemel, R. W. Seidel, J. Schaumann and R. A. Fischer, Inorg. Chem., 2013, 52, 14275.
61. W. Uhl, T. Abel, E. Hagemeier, A. Hepp, M. Layh, B. Rezaeirad and H. Luftmann, Inorg. Chem., 2011, 50, 325.
62. W. Uhl and M. Matar, Z. Naturforsch., B: J. Chem. Sci., 2004, 59, 1214.
63. S. K. Kushvaha, A. Mishra, H. W. Roesky and K. Chandra Mondal, Chem. – Asian J., 2022, 17, e202101301.
64. M. Melaimi, R. Jazzar, M. Soleilhavoup and G. Bertrand, Angew. Chem., Int. Ed., 2017, 56, 10046.
65. D. Naglav, M. R. Buchner, G. Bendt, F. Kraus and S. Schulz, Angew. Chem., Int. Ed., 2016, 55, 10562.
66. M. R. Buchner and M. Müller, ACS Chem. Health Saf., 2023, 30, 36.

---

Footnotes

† Electronic supplementary information (ESI) available: Synthetic procedures, NMR spectra, and X-ray crystallographic details. CCDC 2379134–2379141. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d4dt02457e

‡ Caution! Beryllium and its compounds are regarded as toxic and carcinogenic. Because the biochemical mechanisms that cause beryllium-associated diseases are still unknown, special (safety) precautions are strongly advised.^65,66

§ The same reaction carried out in benzene instead of THF to avoid CAAC-THF ligand scrambling proved much less selective. The low isolated yield of 3 of 33% is due to the difficulty in separating 3 from the CAAC ligand released during the reaction, as both have very similar solubility in pentane.

¶ The calculated ⁹BeNMR shift of 4a at the B3LYP-D3(BJ)-def2-SVP level of theory using 2b (δ_⁹Be = 14.8 ppm) as a reference is 14.6 ppm, therefore 4a is unlikely to be the minor product observed at δ_⁹Be = 7.1 ppm.

|| Compound 4b slowly decomposed within the reaction mixture upon exposure to light.

** The structural data for [Me₂Al{μ-O(SiMe₃)}₂Mg(μ-OCN)]₃ are of insufficient quality to allow comparison with 7.

---