Nature of the heavy formal double bonds As=Ch, Sb=Ch and Bi=Ch (Ch = S, Se, Te) in NCN-pincer supported arsinidene, stibinidene and bismuthinidene chalcogenides

Abstract

The synthesis of arylpnictinidenes 2,6-(Ph₂PNMes)₂C₆H₃Pn based upon a novel bis(phosphine imine) NCN-pincer ligand is reported (Pn = As, Sb, Bi). The oxidation of 2,6-(Ph₂PNMes)₂C₆H₃Pn with sulfur, selenium and tellurium, respectively, afforded arylarsinidene chalcogenides 2,6-(Ph₂PNMes)₂C₆H₃AsCh, arylstibinidene chalcogenides 2,6-(Ph₂PNMes)₂C₆H₃SbCh and arylbismuthinidene chalcogenides 2,6-(Ph₂PNMes)₂C₆H₃BiCh, which can be formulated as containing terminal As=Ch, Sb=Ch and Bi=Ch double bonds. Based on the complementary bonding analysis, the bonding situation is best described in terms of bipolar ⁺Pn–Ch⁻ single bonds (Ch = S, Se, Te).

---

Introduction

Arylpnictinidenes RPn (Pn = N, P, As, Sb, Bi; R = aryl) are neutral carbene analogues of group 15^1,2 that have received tremendous attention in recent years owing to their low valency and arising opportunities for main group driven bond activation and catalysis.^3,4 In their native form, arylpnictinidenes, are fiercely reactive species due to their electron deficiency as well as their electronic triplet ground states and only very recently, it has been discovered that kinetic stabilization of triplet arylpnictinidenes can be achieved using extremely bulky and rigid M^SFluind substituents that effectively shield the pnictogen atoms.^5–9 Prior to this discovery, singlet arylpnictinidenes were obtained by electronic stabilization using intramolecularly coordinating aryl substituents comprising N-donor atoms, which are able to compensate the electron deficiency of the pnictogen atoms. In 2010, the first isolable arylstibinidene and arylbismuthinidene, 2,6-[RNC(R′)]₂C₆H₃Pn (I: Pn = Sb, Bi; R = 2,6-Me₃C₆H₃, R′ = Me) were reported using an adjustable bis(aldimine) and bis(ketimine)-based NCN-pincer ligand system (Scheme 1).¹⁰ Structural elucidation revealed that both N atoms are involved in the coordination of the pnictogens giving rise to formal hypervalent bonding. Since then, closely related arylstibinidenes and arylbismuthinidenes I have been reported with the same NCN-pincer ligand system, at which the positions of the organic substituents R and R′ were varied with the aim of fine tuning the reactivity.^11–14

Scheme 1 Electronically stabilized arylpnictinidenes containing NCN-pincer ligands.

The lighter group 15 analogues, namely arylnitrenes, -phosphinidenes and -arsinidenes II were reported with the same bis(imine)-based NCN-pincer ligand system.^14–19 Notably, within the lighter arylpnictindenes, one imine-group preferentially coordinates to the pnictogens due to the smaller size and aromatic stabilization of the resulting five-membered heterocycles, while a fluxional exchange (“bell-cappers”) was observed in solution. Asymmetric coordination modes may be locked in upon modification of the NCN-pincer system.^11,14,20,21

The arylpnictinidenes I and II were subject to extensive reactivity studies. The heavier arylpnictinidenes 2,6-[RNC(H)]₂C₆H₃Pn (Pn = As, Sb, Bi) were used as donor ligands for the preparation of numerous metal carbonyl complexes of Cr, Mo, W, Fe, Mn and Co^12,16 as well as Au²² or in Diels–Alder type cycloadditions with alkynes.^23,24 The stibinidene 2,6-[t-BuNC(H)]₂C₆H₃Sb was utilized as an effective main group redox catalyst in the hydroboration of disulfides²⁵ as well as in reversible addition reactions with N-alkyl/arylmaleimides.²⁶ The bismuthinidene 2,6-[RNC(H)]₂C₆H₃Bi undergoes (light assisted) oxidative addition with aryl iodides,^27,28 stemming from single-electron transfer (SET) processes.²⁹ Activation of dinitrogen oxide provided an asymmetric oxidation product,¹³ which was able to oxidize HBpin to HOBpin. Reaction of the bismuthinidene 2,6-[2′,6′-Ph₂C₆H₃NC(Me)]₂C₆H₃Bi with the 2,4,6-tri-tert-butylphenoxy radical (2,4,6-TTBP) gave rise to a Bi(III) species featuring an extremely labile Bi–O bond, which could be readily transformed into an arylbismuth(II) species able to activate protic species, such as water, ammonia, phenol and aniline.³⁰ Furthermore, upon single electron oxidation with [Cp₂Fe][BAr^F₄], the arylbismuthinidene could be converted into the related radical cations.³¹ The utility of the arylbismuthinidene 2,6-[t-BuNC(H)]₂C₆H₃Bi as a catalyst was showcased for transfer hydrogenation reactions of azobenzenes,³² which occur without metal–ligand cooperativity,²⁰ the hydrodefluorination of polyfluorinated aromatics,³³ and for the degradation of sulfur hexafluoride and phenylsulfur pentafluoride.³⁴ Most recent applications in catalysis include light-induced trifluoromethylation of heteroarenes,³⁵ light-induced reductive cyclopropanation of alkenes with diiodomethane and manganese powder,³⁶ and the intramolecular aminocyclization affording cyclic carbamates.³⁷

Given the rich diversity in the reactivity of stibinidenes and bismuthinidenes as well as the non-innocent character of the bis(aldimine) and bis(ketimine)-based NCN-pincer ligand system, we envisaged a novel bis(phosphine imine)-based NCN-pincer scaffold, derived from the 2,6-[bis(diphenylphosphinophenyl)]phenyl substituent, recently introduced by us and applied for the synthesis of transition metal complexes.^38–40 This novel bis(phosphine imine)-based NCN-pincer scaffold based upon PNMes groups allowed the preparation of a novel arsinidene, stibinidene and bismuthinidene, the oxidation of which with chalcogens led to the formation of electronically stabilized stibinidene chalcogenides and bismuthinidene chalcogenides comprising formal As=Ch, Sb=Ch and Bi=Ch double bonds (Ch = S, Se, Te) including the first structurally authenticated Bi=Te double bond. These compounds nicely complement the kinetically stabilized stibinidene chalcogenides reported very recently.⁴¹

Results & discussion

Synthetic aspects

The ligand precursor was synthesized by a Staudinger reaction of 2,6-(Ph₂P)₂C₆H₃Br⁴² with two equivalents of mesityl azide to provide 2,6-(Ph₂PNMes)₂C₆H₃Br (1) as colourless crystals in 84% yield (Scheme 2). The ³¹P NMR spectrum of 1 consists of a singlet at δ = −8.9 ppm, which is more deshielded compared to related pincer complexes reflecting the relative coordination strengths of the N atoms, vide infra.

Scheme 2 Synthesis of 1.

The metal halide exchange reaction of 1 with the Turbo-Grignard-reagent, i-PrMgCl·LiCl, produced the arylmagnesium chloride 2,6-(Ph₂PNMes)₂C₆H₃MgCl·THF·Et₂O (2) as pale-yellow crystals in 90% yield (Scheme 3).

The molecular structures of 1 and 2 are shown in Fig. 1. As anticipated, the N atoms of 1 are not involved in any interactions with the Br atom. The short, unaltered P–N bond lengths of 1 (1.546(2), 1.548(2) Å) may be regarded as reference values for the discussion of the relative coordination strengths in related pincer complexes, vide infra. In 2, both N atoms firmly coordinate to the Mg atom, leading to a tetrahedral spatial arrangement of the Mg atom in 2, defined by a CN₂O donor set. The Mg–N bond lengths (2.222(2), 2.248(2) Å) are marginally unequal. As the result of this coordination, the P–N bond lengths of 2 (1.596(2), 1.598(2) Å) are slightly longer than in 1. Additionally, 2 shows a ³¹P-NMR chemical shift of 10.0 ppm.

Scheme 3 Synthesis of 2.

Fig. 1 Molecular structures of 1 and 2 showing 50% probability ellipsoids and the essential atomic numbering scheme of the core region. Substituents are shown as wireframes for clarity. Selected bond lengths [Å] for 1: P1–N1 1.546(2), P2–N2 1.548(2); for 2: P1–N1 1.596(2), P2–N2 1.598(2), Mg1–N1 2.222(2), Mg1–N2 2.248(2), Mg1–C10 2.130(2), Mg1–Cl1 2.381(1), Mg1–O1 2.073(2).

The salt metathesis reaction of 2 with pnictogen trichlorides, PnCl₃ (Pn = P, As, Sb, Bi) afforded the aryldichloropnictogens, 2,6-(Ph₂PNMes)₂C₆H₃PnCl₂ (3Pn) as colourless crystalline solids in 61–94% yield (Scheme 4). Unlike 3As, 3Sb and 3Bi that are indefinitely stable when kept under inert conditions, 3P decomposes within a few days into ill-defined products even when stored as a solid under argon.

Scheme 4 Synthesis of 3Pn (Pn = P, As, Sb, Bi).

The molecular structures of 3Sb and 3Bi are shown in Fig. 2. The spatial arrangement of the Sb and Bi atoms is distorted square pyramidal due to the stereochemically active lone pairs and defined by CN₂Cl₂ donors sets. The Sb–N and Bi–N bond lengths of 3Sb (2.286(1), 2.326(1) Å) and 3Bi (2.395(2), 2.438(2) Å) are slightly unequal and significantly shorter than those of the arylantimony- and -bismuth dichlorides 2,6-(t-BuNCH)₂C₆H₃PnCl₂ (Sb: 2.415(2), 2.399(2) Å, Bi: 2.470(7), 2.499(6) Å).⁴³ In turn, the Sb–Cl and Bi–Cl bond lengths of 3Sb (2.605(1), 2.693(1) Å) and 3Bi (2.626(1), 2.849(1) Å) are slightly longer than those of 2,6-(t-BuNCH)₂C₆H₃PnCl₂ (Sb: 2.597(1), 2.583(1) Å; Bi: 2.662(2), 2.689(2) Å).⁴³ Based upon the comparison of the P–N bond lengths, the N-coordination in 3Sb (1.604(2), 1.600(1) Å) is somewhat stronger than in 3Bi (1.595(2), 1.596(2) Å), which reflects the greater Lewis acidity of the Sb compound. The phosphine imide moieties give rise to ³¹P NMR chemical shifts of δ = 19.8 (3P), 21.1 (3As), 22.9 (3Sb) and 37.6 ppm (3Bi), respectively. The central P-atom of 3P revealed a triplet centred at δ = 100.0 ppm with a J(³¹P–³¹P) coupling of 46 Hz. Despite arduous efforts, we failed to obtain single crystals of 3P and 3As. On one occasion, we obtained a small crop of crystals that was identified as the arylchloroarsenium ion, [2,6-(Ph₂PNMes)₂C₆H₃AsCl][As₂OCl₅] [4As][As₂OCl₅], which serendipitously formed from 3As, excess AsCl₃ and adventitious moisture.

Fig. 2 Molecular structures of 3Sb and 3Bi showing 50% probability ellipsoids and the essential atomic numbering scheme of the core region. Substituents are shown as wireframes for clarity. Selected bond lengths [Å] for 3Sb: P1–N1 1.604(2), P2–N2 1.600(1), Sb1–N1 2.286(1), Sb1–N2 2.326(1), Sb1–C10 2.129(2), Sb1–Cl1 2.605(1), Sb1–Cl2 2.693(1); for 3Bi: P1–N1 1.595(2), P2–N2 1.596(2), Bi1–N1 2.438(2), Bi1–N2 2.395(2), Bi1–C10 2.219(3), Bi1–Cl1 2.626(1), Bi1–Cl2 2.849(1).

In an effort to deliberately prepare arylchloropnictogenium ions, we reacted the aryldichloropnictogens 3Pn (Pn = P, As, Sb, Bi) with trimethylsilyl triflate, Me₃SiO₃SCF₃, or aluminium trichloride, AlCl₃ (Scheme 5). Both Lewis acids abstracted a chloride leading to the formation of the desired arylchloropnictogenium ions [2,6-(Ph₂PNMes)₂C₆H₃PnCl][A] ([4Pn]A, Pn = P, As, Sb, Bi; A^– = O₃SCF₃⁻, AlCl₄⁻) that were isolated as colourless crystalline solids in quantitative yields.

Scheme 5 Synthesis of [4Pn][A] (Pn = P, As, Sb, Bi; A = O₃SCF₃, AlCl₄).

The molecular structures of [4As]⁺ and [4Sb]⁺ are depicted in Fig. 3. The spatial arrangement of the As and Sb atoms is distorted disphenoidal due to the stereochemically active lone pairs and defined by CN₂Cl donor sets. The Sb–N bond lengths of [4Sb]⁺ (2.226(5), 2.272(6) Å) are slightly shorter than those of 3Sb (2.286(1), 2.326(1) Å) and the remaining Sb–Cl bond length [4Sb]⁺ (2.410(2) Å) is shorter than those in 3Sb (2.605(1), 2.693(1) Å), but longer than in the chlorostibenium cations [2,6-(t-BuNCH)₂C₆H₃SbCl]⁺ (2.360(1) Å) and [2,6-(2′,6′-Me₂C₆H₄NCH)₂C₆H₃SbCl]⁺ (2.361(1) Å).⁴⁴ Based upon the P–N bond lengths of [4As]⁺ (1.597(6), 1.601(6) Å) and [4Sb]⁺ (1.600(1), 1.601(6) Å), the N-coordination within these two cations is very similar, but slightly stronger than in 3Sb and 3Bi, vide supra. Interestingly, the two mesityl groups and the two P atoms of the complete series [4Pn]⁺ (Pn = P, As, Sb, Bi) are magnetically inequivalent in solution. The ³¹P NMR spectrum of [4P]⁺ gives rise to three equally intense signals at δ = 129.4 59.3 and 33.0 ppm. The first two signals comprise doublets with identical J(³¹P–³¹P) couplings of 62 Hz, while the last signal is a singlet, which suggest an unsymmetrical coordination mode in solution. Consistently, the remaining compounds show pairs of ³¹P NMR signals at δ = 60.0/33.7 ([4As]⁺), 60.0/30.0 ([4Sb]⁺) and 78.2/37.5 ppm for ([4Bi]⁺), respectively.

Fig. 3 Molecular structures of [4As]⁺ and [4Sb]⁺ showing 50% probability ellipsoids and the essential atomic numbering scheme of the core region. Substituents are shown as wireframes for clarity. Selected bond lengths [Å] for [4As]⁺: P1–N1 1.605(3), P2–N2 1.601(3), As1–N1 2.122(3), As1–N2 2.114(3), As1–C10 1.967(3), As1–Cl1 2.242(1); for [4Sb]⁺: P1–N1 1.601(6), P2–N2 1.597(6), Sb1–N1 2.226(5), Sb1–N2 2.272(6), Sb1–C10 2.158(6), Sb1–Cl1 2.410(2).

The reduction of the heavier aryldichloropnictogens 3Pn (Pn = As, Sb, Bi), with Li[Et₃BH] provided the arylpnictinidenes 2,6-(Ph₂PNMes)₂C₆H₃Pn (5Pn) as red solids in quantitative yields (Scheme 6). Alternatively, the arylstibinidene 2,6-(Ph₂PNMes)₂C₆H₃Sb (5Sb) was also obtained by the reduction of 3Sb with LiAlH₄ and isolated as 5Sb·(AlCl₃·THF)·THF. Notably, neither the donor acceptor complex AlCl₃·THF nor the free THF showed any coordination to the Sb atom, but facilitated crystallization. Unfortunately, the arylpnictinidenes are thermally unstable and start to decompose within a few hours under argon. While attempting to grow single crystals of the arylbismuthinidene 2,6-(Ph₂PNMes)₂C₆H₃Bi (5Bi), a small crop of a crystalline decomposition product was obtained, namely the arylalkylbismuthenium ion [2,6-(Ph₂PNMes)₂C₆H₃BiEt][Et₄B] (Fig. S128†), which formed via oxidative addition of the ethylboron species and ethyl group scrambling. Interestingly, all attempts to prepare 5Sb using NaBH₄ as the reducing agent gave the donor acceptor complex 5Sb·BH₃ as unstable red crystals. The reaction of 5Sb with BH₃·THF provided the same complex 5Sb·BH₃ that was isolated quantitatively, while 5Bi showed no reactivity towards BH₃·THF (Scheme 7). So far, all attempts to reduce 3P with various reducing agents, such as KC₈, lithium, potassium, Li[Et₃BH] or Cp₂Co, have resulted only in ill-defined mixtures of products.

Scheme 6 Synthesis of 5Pn (Pn = As, Sb, Bi).

Scheme 7 Synthesis of 5Sb·BH₃.

The molecular structures of 5As, 5Sb and 5Sb·BH₃ are shown in Fig. 4. The As–N bond lengths of 5As (2.242(3), 2.271(3) Å) are longer than those of the arylchloroarsenium ion [4As]⁺ (2.122(3), 2.114(3) Å). The Sb–N bond lengths of the arylstibinidene 5Sb (2.273, 2.272(6) Å) are between those of the arylantimony dichloride 3Sb (2.286(1), 2.326(1) Å) and the arylchlorostibenium ion [4Sb]⁺ (2.226(5), 2.272(6) Å). The Sb–N bond lengths of the donor acceptor complex 5Sb·BH₃ (2.337(3), 2.361(3) Å) are substantially longer, which suggests a lower Lewis acidity than in the aforementioned antimony compounds. This is also reflected in the short P–N bond lengths of 5Sb·BH₃ (1.585(3), 1.590(3) Å). Notably, the Sb–N bond lengths of the 2,6-bis[N-2′,6'-(dimethylphenyl)ketimino]stibinidene (2.352(3), 2.346(3) Å) are in between those of 5Sb and 5Sb·BH₃. The donor acceptor Sb–B bond of 5Sb·BH₃ (2.293(4) Å) compares well with the sum of covalence radii. The ³¹P NMR resonances of 5Sb, 5Sb·BH₃ and 5Bi show little variance at δ = 10.8, 16.0 and 8.8 ppm.

Fig. 4 Molecular structures of 5As, 5Sb and 5Sb·BH₃ showing 50% probability ellipsoids and the essential atomic numbering scheme. Selected bond lengths [Å] for 5As: P1–N1 1.592(3), P2–N2 1.586(3), As1–N1 2.242(3), As1–N2 2.271(3), As1–C10 1.900(4). Selected bond lengths [Å] for 5Sb: P1–N1 1.610(3), P2–N2 1.600(3), Sb1–N1 2.273(3), Sb1–N2 2.312(3), Sb1–C10 2.155(3). Selected bond lengths [Å] for 5Sb·BH₃: P1–N1 1.590(3), P2–N2 1.585(3), Sb1–N1 2.337(3), Sb1–N2 2.361(3), Sb1–C10 2.163(3), Sb1–B1 2.293(4).

The reaction of the arylarsinidene 5As, arylstibinidene 5Sb and the arylbismuthinidene 5Bi with sulfur, selenium and tellurium led to the formation of arylarsinidene chalcogenides, 2,6-(Ph₂PNMes)₂C₆H₃AsCh (6AsCh; Ch = S, Se), arylstibinidene chalcogenides, 2,6-(Ph₂PNMes)₂C₆H₃SbCh (6SbCh; Ch = S, Se, Te), and arylbismuthinidene chalcogenides, (Ph₂PNMes)₂C₆H₃BiCh (6BiCh; Ch = S, Se, Te), which were isolated as yellow (6AsS, 6AsSe, 6SbS, 6SbSe and 6BiS, 6BiSe) and red (6SbTe and 6BiTe) crystalline needles in nearly quantitative yield based on ³¹P-NMR (Scheme 8).

Scheme 8 Synthesis of 6PnCh (Pn = As, Sb, Bi; Ch = S, Se, Te).

The molecular structures of 6AsSe, 6SbSe and 6BiTe are shown in Fig. 5. The As–Se bond length of 6AsSe (2.310(1) Å) is marginally smaller than that of 2,6-(Me₂NCH₂)₂C₆H₃AsSe (2.351(3) Å).⁴⁵ The Sb–Se bond length of 6SbSe (2.466(1) Å) is marginally larger those in the electronically stabilized species 2-[(2′,4′-i-Pr₂C₆H₄)NC(H)]-6-[(2′′,4′′-i-Pr₂C₆H₄)NHC(H)]C₆H₃SbSe (2.437(1) Å),¹⁴, [2,6-(Me₂NCH₂)C₆H₃]SbSe (2.440(1) Å) and 2,6-[2′,6′-Ph₂C₆H₃NC(Me)]₂C₆H₃SbSe (2.433(1) Å)⁴⁶ and substantially longer than in the kinetically stabilized M^SFluind*SbSe (2.372(1) Å).⁴¹ The Bi–Te bond length of 6BiTe (2.7614(4) Å) is the longest formal double bond between different elements ever reported. It is considerably shorter than the single bond lengths of Et₂BiTeEt (2.912(1) Å)⁴⁷ and 2-[(2′,4′-i-Pr₂C₆H₄)NC(H)]C₆H₃Bi(TePh)₂ (2.895(1) Å).⁴⁸ The ³¹P NMR chemical shifts of 6AsCh, 6SbCh and 6BiCh are in the range from δ = 12.9 to 13.2 ppm, from δ = 14.6 to 15.6 ppm and from δ = 18.2 to 22.8 ppm, respectively. Despite all efforts, no ⁷⁷Se NMR signals were found for the selenides 6AsSe, 6SbSe and 6BiSe after scanning the range from −3000 to +6000 ppm for several hours. The ¹²⁵Te NMR spectra of the tellurides 6SbTe and 6BiTe gave rise to singlets at δ = −274.8 and −193.1 ppm.

Fig. 5 Molecular structures of 6AsSe, 6SbSe and 6BiTe showing 50% probability ellipsoids and the essential atomic numbering scheme of the core region. Substituents are shown as wireframes for clarity. Selected bond lengths [Å] for 6AsSe: P1–N1 1.569(2), P2–N2 1.605(2), As1–N1 2.376(2), As1–N2 2.126(2), As1–C10 2.000(2), Se1–As1 2.310(1). Selected bond lengths [Å] for 6SbSe: P1–N1 1.584(1), P2–N2 1.597(1), Sb1–N1 2.386(1), Sb1–N2 2.307(1), Sb1–C10 2.189(1), Sb1–Se1 2.4661(2). Selected bond lengths [Å] for 6BiTe: P1–N1 1.582(4), P2–N2 1.586(4), Bi1–N1 2.499(4), Bi1–N2 2.488(3), Bi1–C10 2.306(4), Bi1–Te1 2.7614(4).

DFT computations

In an effort to unravel the nature of the formal heavy As=Ch, Sb=Ch and Bi=Ch double bonds in the arylpnictinidene chalcogenides, 2,6-(Ph₂PNMes)₂C₆H₃AsCh (6AsCh; Ch = S, Se, Te), 2,6-(Ph₂PNMes)₂C₆H₃SbCh (6SbCh; Ch = S, Se, Te), (Ph₂PNMes)₂C₆H₃BiCh (6BiCh; Ch = S, Se, Te) we performed complementary bonding analyses.⁴⁹

The electron density-based Atoms-In-Molecule (AIM)⁵⁰ approach reveals decreasing electron density values (ρ_bcp) at the bond critical point with increasing size of the chalcogen of 0.86/0.70/0.65 e Å⁻³ (AsS/SbS/BiS), 0.72/0.61/0.57 e Å⁻³ (AsSe/SbSe/BiSe) and 0.56/0.50/0.47 e Å⁻³ (AsTe/SbTe/BiTe) with slightly higher values obtained for 6AsCh. Interestingly, the Laplacians (∇²ρ_bcp) are increasing from AsS to AsTe, while a steady decrease is observed for the heavier congeners SbS to SbTe as well as from BiS to BiTe indicating an increase in ionic contributions for 6AsCh and a decrease within 6SbCh and 6BiCh. However, the kinetic energy over electron density ratios (G/ρ_bcp) are slightly steadily decreasing also from AsS to AsTe indicating a slight decrease in ionic bonding contributions also for 6AsCh (Tables S2–S4†). On the other side, the total energy over electron density values (H/ρ_bcp) are getting closer to zero from S to Te, which in turn indicates a simultaneous slight increase in polarity/iconicity also for the heavier 6SbCh and 6BiCh. Such behavior has also been described before for Si–O bonds.⁵¹ Despite the rather low values of AIM derived parameters, the delocalization indices range from 1.19 to 1.26 for AsCh, 1.24 to 1.26 for SbCh and from 1.29 to 1.30 for BiCh, supporting the formulation of formal Ch=Pn double bonds (Table S5†). The Wiberg Bond indices (WBI, AsCh: 1.11–1.23, SbCh: 1.19–1.21; BiCh: 1.21–1.25) are lower compared to the recently reported kinetically stabilized stibinidene chalcogenides M^SFlunid*SbCh (ranging from 1.69 to 1.82)⁴¹ presumably owing to the electronic stabilization of the N-atoms in 6PnCh. Inspection of the Non-covalent interaction (NCI) index clearly indicates the covalent bonding contributions along the AsCh, SbCh and BiCh bond axis, which are visibly more pronounced within the AsS interactions already indicated by the AIM parameters (Fig. 6 and S129–S131†).⁵² Furthermore, the trend of the AIM parameters for the series 6PnCh augment the values reported for the heaviest members of SbO and BiO bonds being prepared only recently.^53,54 Following the trend, the SbO and BiO bcps show the highest values of ρ_bcp and ∇²ρ_bcp indicating a higher degree of polarity in the oxygen based SbO and BiO double bonds. Extending the analysis of ρ_bcp, ∇²ρ_bcp and the ellipticity (ε) along the entire PnCh bond lengths reveals a substantial difference between the PnS bonds and the heavier PnSe and PnTe bonds. The electron densities along the AsS, SbS and BiS bonds are asymmetric (Fig. S132†), reminiscent of the reported AsO, SbO and SbO bonds,⁵³ whereas the distribution of ρ becomes more symmetric for the heavier chalcogens Se and Te (Fig. S132†). The Laplacians of ρ show a clear minimum near the sulfur in all PnS bonds, again similar to the respective heavier PnO bonds,⁵³ which is absent in the PnSe and PnTe bonds. However, slight differences are observed between the AsSe and AsTe bonds compared to the heavier analogues. For AsSe, ∇²ρ_bcp is almost symmetrically distributed, whereas a small minimum is formed in the location of the selenium for SbSe and BiSe. For AsTe, ∇²ρ_bcp a small minimum is observed in the location of the arsenic, whereas for SbTe the distribution is rather symmetric and for BiTe a small minimum is observed in the direction of the tellurium (Fig. S132†). For the ellipticity, the highest values are observed close to the arsenic in the AsCh bonds, which decrease towards the bond critical point and reach a minimum after the bcp before increasing towards the chalcogens. For the SbCh and BiCh bonds, the overall ellipticities are substantially smaller, with the SbCh bonds showing the same trend as the AsCh bonds. For the BiCh bonds, the ellipticities reach a minimum before the bcp (closer to the bismuth) and increase towards the chalcogen (Fig. S132†).

Fig. 6 AIM molecular graphs of 6AsS (top left), 6SbSe (middle left) and 6BiTe (bottom left) with bond critical points as red spheres and bond paths in orange as well as NCI iso-surfaces at s(r) = 0.5 colour coded with sign(λ2)ρ in a. u. Blue surfaces refer to attractive forces and red to repulsive forces. Green indicates weak interactions. Respective HOMO–1 of 6AsS (top right), 6SbSe (middle right) and 6BiTe (bottom right) at iso-surfaces at s(r) = +/− 0.02 (blue/red).

NBO/NLMO analyses of the AsCh, SbCh and BiCh bonds revealed in each case only one As–Ch, Sb–Ch and Bi–Ch bonding orbital in line with the formulation of bipolar ⁺As–Ch^–, ⁺Sb–Ch^– and ⁺Bi–Ch⁻ bonds (Tables S7, S9 and S11†).⁵⁵ Second-order perturbation theory shows LP(Ch) → LV(As/Sb/Bi), LP(Ch) → LV(C_ipso) and LP(Ch) → σ*(As/Sb/Bi–C) donor–acceptor interactions summing up to a total of E2 = 63 to 73 kcal mol⁻¹ (6AsCh), E2 = 48 to 53 kcal mol⁻¹ (6SbCh) and E2 = 41 to 46 kcal mol⁻¹ (6BiCh), which are higher compared to values reported for the SbO and BiO bonds.⁵⁴ The NLMO analysis revealed that two LP_Ch NLMOs show reduced percentages of the parent NBO, indicating that the respective lone pairs are partially delocalized and make up 12.4 to 17.4% (6AsCh), 15.4 to 20.1% (6SbCh) and 20.8 to 25.5% (6BiCh) of the total AsCh/SbCh/BiCh NLMO bond orders (Tables S5, S8, S10 and S12†). The partial π-backdonation from the chalcogen is also visible in the respective HOMO–1 molecular orbitals (Fig. 6 and S136–S138†).

Additionally, EDA-NOCV⁵⁶ analyses of the two experimentally obtained molecular structures of 6SbSe and 6BiTe were carried out considering two different bonding situations: (i) for the interaction of the formal Sb=Se and Bi=Te double bond, two neutral fragments in their respective triplet states have been applied. (ii) For the bipolar bonding situation ⁺Sb–Se^– and ⁺Bi–Te^–, charged fragments (RSb⁺/RBi⁺ and Se⁻/Te⁻) in their respective doublet state have been used. The energy terms derived from the EDA-NOCV analysis are summarized in Table S13.† The bipolar bonding situation with a Sb–Se and Bi–Te single bond gives rise to ΔE_orb values which are closer to zero compared to the ΔE_orb values obtained for the double bond situation, indicating that the heavy formal double bonds are better described with a Lewis structure containing bipolar ⁺Pn–Ch⁻ bonds. As a last point, the iso-surface of the localized orbital locator⁵⁷ of the π-orbitals (LOL-π), exemplified for 6BiCh, shows the same trend (Fig. S140†), clearly indicating a charge separated formulation with π-electron residing at the chalcogens, respectively.

Conclusions

A novel NCN-pincer ligand based upon a bis(phosphine imine) substituted phenyl substituent was introduced and applied for the preparation of an electronically stabilized arylarsinidene 2,6-(Ph₂PNMes)₂C₆H₃As (5As), arylstibinidene 2,6-(Ph₂PNMes)₂C₆H₃Sb (5Sb) and arylbismuthinidene 2,6-(Ph₂PNMes)₂C₆H₃Bi (5Bi). The oxidation of 5As, 5Sb and 5Bi with sulfur, selenium and tellurium, respectively, produced the arylarsinidene chalcogenides 2,6-(Ph₂PNMes)₂C₆H₃AsCh (6AsCh, Ch = S, Se), arylstibinidene chalcogenides 2,6-(Ph₂PNMes)₂C₆H₃SbCh (6SbCh) and arylbismuthinidene chalcogenides 2,6-(Ph₂PNMes)₂C₆H₃BiCh (6BiCh) formally containing terminal As=Ch, Sb=Ch and Bi=Ch double bonds (Ch = S, Se, Te). The formal Bi=Te double bond of the arylbismuthinidene telluride 6BiTe (bond length 2.7614(4) Å) comprises the heteroatomic combination of the two heaviest main group elements excluding strongly radioactive ones. A bond analysis suggests these formal double bonds are best described as bipolar ⁺As–Ch⁻, ⁺Sb–Ch⁻ and ⁺Bi–Ch⁻ single bonds (Ch = S, Se, Te).

Data availability

Figures of NMR spectra as well as crystal and refinement data are given in the ESI.† Crystallographic information files (CIF) have been deposited with the Cambridge Crystallographic Data Centre, no. 2446710–2446720, 2457652 and 2457653. Additional results from quantum chemical calculations are given in the ESI.† The raw data that support the findings of this study are available from the corresponding authors upon reasonable request.

Author contributions

F. M. synthesized and isolated the compounds 1, 2, [4]A (A = AlCl₄, O₃SCF₃), 5Pn, 6PnCh and performed NMR and UV-vis measurements. A. S. synthesized and performed NMR and UV-vis measurements on 3Pn and [4As][As₂OCl₅]. J. B. conducted the X-ray diffraction measurements and structure refinement. J. B. and L. D. designed the project. E. H. performed all theoretical computations. Writing of the manuscript was done by F. M., L. D., J. B. and E. H.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

We gratefully acknowledge the Deutsche Forschungsgemeinschaft (DFG) for financial support (BE3716/14-1). We thank Prof. E. Rivard, University of Alberta, for providing access to the Amsterdam Modelling Suite provided by the Digital Research Alliance of Canada (https://alliancecan.ca/).

Notes and references

1. L. Dostal, Coord. Chem. Rev., 2017, 353, 142–158.
2. M. He, C. Hu, R. Wei, X.-F. Wang and L. L. Liu, Chem. Soc. Rev., 2024, 53, 3896–3951.
3. J. M. Lipshultz, G. Li and A. T. Radesevich, J. Am. Chem. Soc., 2021, 143, 1699–1721.
4. H. W. Moon and J. Cornella, ACS Catal., 2022, 12, 1382–1393.
5. Y. Pang, N. Nöthling, M. Leutzsch, L. Kang, E. Bill, M. van Gastel, E. Reijerse, R. Goddard, L. Wagner, D. SantaLucia, S. DeBeer, F. Neese and J. Cornella, Science, 2023, 380, 1043–1048.
6. M. Wu, H. Li, W. Chen, D. Wang, Y. Chen, S. Ye and G. Tan, Nat. Sci. Rev., 2023, 10, nwad169.
7. M. Wu, H. Li, W. Chen, D. Wang, Y. He, L. Xu, S. Ye and G. Tan, Chem, 2023, 9, 2573–2584.
8. M. Janssen, T. Frederichs, M. Olaru, E. Lork, E. Hupf and J. Beckmann, Science, 2024, 385, 318–321.
9. D. Wang, W. Chen, H. Chen, Y. Chen, S. Ye and G. Tan, Nat. Chem., 2025, 17, 38–43.
10. P. Šimon, F. de Proft, R. Jambor, A. Růžička and L. Dostál, Angew. Chem., Int. Ed., 2010, 49, 5468–5471.
11. I. Vránová, M. Alonso, R. Lo, R. Sedlák, R. Jambor, A. Růžička, F. de Proft, P. Hobza and L. Dostál, Chem.–Eur. J., 2015, 21, 16917–16928.
12. I. Vránová, M. Alonso, R. Jambor, A. Růžička, M. Erben and L. Dostál, Chem.–Eur. J., 2016, 22, 7376–7380.
13. Y. Pang, M. Leutzsch, N. Nöthling and J. Cornella, J. Am. Chem. Soc., 2020, 142, 19473–19479.
14. J. Zechovský, E. Kertész, V. Kremláček, M. Hejda, T. Mikysek, M. Erben, A. Růžička, R. Jambor, Z. Benkő and L. Dostál, Organometallics, 2022, 41, 2535–2550.
15. J. Hyvl, W. Y. Yoshida, A. L. Rheingold, R. P. Hughes and M. F. Cain, Chem.–Eur. J., 2016, 22, 17562–17565.
16. I. Vránová, V. Kremláček, M. Erben, J. Turek, R. Jambor, A. Růžička, M. Alonso and L. Dostál, Dalton Trans., 2017, 46, 3556–3568.
17. J. Hyvl, W. Y. Yoshida, C. E. Moore, A. L. Rheingold, R. P. Hughes and M. F. Cain, Polyhedron, 2018, 143, 99–104.
18. V. Kremláček, J. Hyvl, W. Y. Yoshida, A. Růžička, A. L. Rheingold, J. Turek, R. P. Hughes, L. Dostál and M. F. Cain, Organometallics, 2018, 37, 2481–2490.
19. M. T. Nguyen, B. Gabidullin and G. I. Nikonov, Dalton Trans., 2018, 47, 17011–17019.
20. H. W Moon, F. Wang, K. Bhattacharyya, O. Planas, M. Leutzsch, N. Nöthling, A. A. Auer and J. Cornella, Angew. Chem., Int. Ed., 2023, 62, e202313578.
21. I. Vránová, M. Alonso, R. Jambor, A. Růžička, J. Turek and L. Dostál, Chem.–Eur. J., 2017, 23, 2340–2349.
22. M. Kořenkova, V. Kremláček, M. Erben, R. Jirásko, F. de Proft, J. Turek, R. Jambor, A. Růžička, I. Cisařová and L. Dostál, Dalton Trans., 2018, 47, 14503–14514.
23. J. Zechovský, V. Kremláček, M. Erben, M. Hejda, E. Rychagova, R. Jambor, A. Růžička, S. Ketkov and L. Dostál, Dalton Trans., 2022, 51, 15933–15945.
24. L. Greb, F. Ebner, Y. Ginzburg and L. M. Sigmund, Eur. J. Inorg. Chem., 2020, 32, 3030–3047.
25. M. Huang, K. Li, Z. Zhang and J. Zhou, Antimony Redox Catalysis, J. Am. Chem. Soc., 2024, 146, 20432–20438.
26. M. Kořenkova, M. Hejda, M. Erben, R. Jirásko, R. Jambor, A. Růžička, E. Rychagova, S. Ketkov and L. Dostál, Chem.–Eur. J., 2019, 25, 12884–12888.
27. M. Hejda, R. Jirásko, A. Růžička, R. Jambor and L. Dostál, Organometallics, 2020, 39, 4320–4328.
28. M. Mato, P. C. Bruzzese, F. Takahashi, M. Leutzsch, E. J. Reijerse, A. Schnegg and J. Cornella, J. Am. Chem. Soc., 2023, 145, 18742–18747.
29. M. Mato, D. Spinnato, M. Leutzsch, H. W. Moon, E. J. Reijerse and J. Cornella, Nature Chem., 2023, 15, 1138–1145.
30. X. Yang, E. J. Reijerse, K. Bhattacharyya, M. Leutzsch, M. Kochius, N. Nöthing, J. Busch, A. Schnegg, A. A. Auer and J. Cornella, J. Am. Chem. Soc., 2022, 144, 16535–16544.
31. X. Yang, E. J. Reijerse, N. Nöthing, D. J. SantaLucia, M. Leutzsch, A. Schnegg and J. Cornella, J. Am. Chem. Soc., 2023, 145, 5618–5623.
32. F. Wang, O. Planas and J. Cornella, J. Am. Chem. Soc., 2019, 141, 4235–4240.
33. Y. Pang, M. Leutzsch, N. Nöthling, F. Katzenburg and J. Cornella, J. Am. Chem. Soc., 2021, 143, 12487–12493.
34. V. A. Béland, N. Nöthing, M. Leutzsch and J. Cornella, J. Am. Chem. Soc., 2024, 146, 25409–25415.
35. T. Tsuruta, D. Spinnato, H. W. Moon, M. Lentzsch and J. Cornella, J. Am. Chem. Soc., 2023, 145, 25538–25544.
36. S. Ni, D. Spinnato and J. Cornella, J. Am. Chem. Soc., 2024, 146, 22140–22144.
37. M. Mato, F. Wang and J. Cornella, Adv. Synth. Catal., 2024, 366, 740–744.
38. M. Olaru, J. F. Kögel, E. Lork, S. Mebs, M. Vogt and J. Beckmann, Chem.–Eur. J., 2020, 26, 275–284.
39. M. Olaru, E. Rychagova, S. Ketkov, Y. Shynkarenko, S. Yakunin, M. V. Kovalenko, A. Yablonsiky, B. Andreev, J. Beckmann and M. Vogt, J. Am. Chem. Soc., 2020, 142, 373–381.
40. F. Meyer, P. Puylaert, D. Duvinage, E. Hupf and J. Beckmann, Chem. Commun., 2024, 60, 12912–12915.
41. X. Li, Y. Chen, S. Dong, D. Wang, L. Xu, J. Zhu and G. Tan, J. Am. Chem. Soc., 2025, 147, 9858–9864.
42. F. Meyer, T. Kuzmera, E. Lork, M. Vogt and J. Beckmann, Z. Anorg. Allg. Chem., 2021, 647, 1890–1895.
43. I. Vránová, R. Jambor, A. Růžička, R. Jirásko and L. Dostál, Organometallics, 2015, 34, 534–541.
44. I. Vránová, M. Erben, R. Jambor, A. Růžička, R. Jirásko and L. Dostál, Z. Anorg. Allgem. Chem., 2016, 642, 1212–1217.
45. J. Vrána, R. Jambor, A. Růžička, A. Lyčka, F. De Proft and L. Dostál, J. Organomet. Chem., 2013, 723, 10–14.
46. (a) R. Jambor, A. Růžička, A. Lyčka, J. Brus, F. de Proft and L. Dostál, Organometallics, 2008, 27, 6059–6062; (b) P. Šimon, R. Jambor, A. Růžička, A. Lyčka, F. De Proft and L. Dostál, Dalton Trans., 2012, 41, 5140–5143.
47. S. Heimann, A. Kuczkowski, D. Bläser, C. Wölper, R. Haack, G. Jansen and S. Schulz, Eur. J. Inorg. Chem., 2014, 28, 4858–4864.
48. P. Šimon, R. Jambor, A. Růžička and L. Dostál, Organometallics, 2013, 32, 239–248.
49. Complementary Bonding Analysis, ed. Grabowsky, S., de Gruyter, Berlin, 2021.
50. R. W. F. Bader, Atoms in Molecules. A Quantum Theory, Cambridge University Press, Oxford U.K., 1991.
51. (a) F. Feige, L. A. Malaspina, F. Kleemiss, J. F. Kögel, S. Ketkov, E. Hupf, S. Grabowsky and J. Beckmann, Dalton Trans., 2023, 52, 5918–5925; (b) M. Fugel, M. F. Hesse, R. Pal, J. Beckmann, D. Jayatilaka, M. J. Turner, A. Karton, P. Bultinck, G. S. Chandler and S. Grabowsky, Chem.–Eur. J., 2018, 24, 15275–15286.
52. E. R. Johnson, S. Keinan, P. Mori-Sánchez, J. Contreras-García, A. J. Cohen and W. Yang, J. Am. Chem. Soc., 2010, 132, 6498–6506.
53. B. Lindquist-Kleissler, J. S. Wenger and T. C. Johnstone, Inorg. Chem., 2021, 60, 1846–1856.
54. (a) J. S. Wenger, M. Weng, G. N. George and T. C. Johnstone, Nat. Chem., 2023, 15, 633–640; (b) J. Kuziola, H. W. Moon, M. Leutzsch, N. Nothling, V. A. Beland and J. Cornella, Angew. Chem., Int. Ed., 2024, 64, e202415169.
55. E. D. Glendening, C. R. Landis and F. Weinhold, Wiley Interdiscip. Rev.: Comput. Mol. Sci., 2012, 2, 1–42.
56. M. P. Mitoraj, A. Michalak and T. Ziegler, J. Chem. Theory Comput., 2009, 5, 962–975.
57. (a) H. L. Schmider and A. D. Becke, J. Mol. Struct., 2000, 527, 51–61; (b) V. Tsirelson and A. Stash, Acta Cryst., 2002, B58, 780–785.

---

Footnote

† Electronic supplementary information (ESI) available: Complete experimental, spectroscopic, crystallographic and computational details. CCDC 2446710–2446720. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d5sc03320a

---