A neutral low-coordinate heterocyclic bismuth-tin species

Abstract

The reaction of distannadiazane bearing bulky ^RAr*-groups (^RAr* = C₆H₂{C(H)Ph₂}₂R-2,6,4; R = iPr, tBu) with ECl₃ (E = Sb, Bi) was studied resulting in the isolation of previously unknown N,N-bis(dichloropnictino)amines (3) and a novel heterocyclic carbenoid bismuth species (4) bearing a Bi^(III) and a Sn^(IV) center. The structure and bonding was investigated by means of X-ray structure elucidations and DFT calculations.

---

Pnictogen–nitrogen heterocycles of the type [XE(μ-NR)]₂ (E = P, As, Sb, Bi; species I in Fig. 1) are valuable starting materials for preparative E–N chemistry.¹ Usually, [ClE(μ-NR)]₂ (E = P, As) is prepared from RN(ECl₂)H in a base-assisted (e.g. NEt₃) cyclization,² however, for the heavier analogs this strategy works poorly. For example, [ClBi(μ-NTer)]₂ (Ter = terphenyl = 2,6-bis-(2,4,6-trimethylphenyl)phenyl) was initially obtained in moderate yields of 45% besides large amounts of ClBi(N(H)Ter)₂.³ In analogy to Veith's synthesis of [Me₂SiE(μ-NtBu)₂]⁺ (II in Fig. 1),⁴ our group succeeded in establishing a straightforward route towards the synthesis of [ClE(μ-NTer)]₂ (E = Sb, Bi), based on the trans-metalation of the respective tin precursor.⁵ Now highly reactive cyclo-1,3-dipnicta-2,4-diazenium salts of the type [E(ClE)(μ-NTer)₂]⁺ (E = P, As,⁶ Sb, Bi;⁵ III in Fig. 1) can be obtained by chloride abstraction from [ClE(μ-NTer)]₂ by means of Lewis acids such as GaCl₃. A new area of research opened up with the isolation of thermally stable biradicaloids of the type [E(μ-NTer)]₂ (E = P, As; IV in Fig. 1) which can easily be accessed by reduction of [ClE(μ-NTer)]₂ with activated magnesium chips.⁷

Fig. 1 Selected known four-membered E–N heterocycles.^4–7

Just recently, we described the synthesis of stable acyclic chloropnictenium ion salts, with an exceedingly bulky ^RAr*-group (Ar* = C₆H₂{C(H)Ph₂}₂R-2,6,4; R = Me, tBu) attached to the nitrogen atom.⁸ This sterically demanding moiety offers two flanking phenyl groups for arene-interactions with the low-coordinate reactive site of the molecules. Jones and co-workers realized new bonding situations with the aid of the ^RAr*-moiety,⁹ such as mono-coordinate Ge or Sn cations,¹⁰ singly bonded distannyene and Ge and Sn hydride complexes,^11,12 that showed magnificent activity as a catalyst in hydroboration reactions.¹³ Just recently, the first example of an amido-distibene in [^iPrAr*N(SiiPr₃)Sb]₂ was reported.¹⁴ Herein we describe the synthesis of an unprecedented distannadiazane [Sn(μ-N^RAr*)]₂ with a planar N₂Sn₂-core and its trans-metalation with ECl₃ (E = Sb, Bi), resulting in the isolation of the first N,N-bis(dichlorostibino)amine and an elusive four-membered ring system with a N₂Bi^(III)Sn^(IV) unit.

In analogy to a procedure described by Power et al., leading to the first isolable distannadiazane [Sn(μ-NTer)]₂,¹⁵ the exceedingly bulky amine ^tBuAr*NH₂ and Sn{N(SiMe₃)₂}₂ were combined in a Schlenk flask without solvent and heated to 160 °C over a period of 45 min, affording a deep red solid. HN(SiMe₃)₂ and excess Sn{N(SiMe₃)₂}₂ were removed in vacuo and the crude product was recrystallized from C₆H₅F to obtain red crystals of [Sn(μ-N^tBuAr*)]₂ (1R, R = tBu) in moderate yields (64%). The synthesis of 1Me and 1iPr suffered from low solubility of the products in common organic solvents, however, minimal amounts of X-ray quality crystals of 1iPr were obtained from C₆H₆. In the ¹³C and ¹H NMR spectrum 1iPr and 1tBu can be easily identified by the signals of the para-substituent of the inner phenyl group and their diagnostic ¹¹⁹Sn NMR shifts (1iPr 783.1 ppm, 1tBu 789.2 ppm; cf. [Sn(μ-NTer)]₂ 738.9 ppm). 1iPr and 1tBu crystallize as solvates of C₆H₆ or C₆H₅F (see Fig. S1 and S4 in the ESI†), respectively, in the triclinic space group P1̄ with one molecule in the asymmetric unit, which lies on a crystallographically imposed centre of inversion. In contrast to [Sn(μ-NTer)]₂, in which the Sn₂N₂ ring is characterized by a folding about the Sn⋯Sn axis of 148°, the Sn₂N₂-core is planar with slightly different N1–Sn1 and N1′–Sn1′ distances (1iPr 2.076(2), 2.086(2); 1tBu 2.075(2), 2.090(2) Å; cf. [Sn(μ-NTer)]₂ 2.09, 2.11 Å), a transannular Sn1⋯Sn1′ separation of 3.2304(4) (1iPr) and 3.2318(3) Å (1tBu) and rather acute angles at the tin center (1iPr 78.27(7), 1tBu 78.22(6)°, cf. [Sn(μ-NTer)]₂ 77.6°).¹⁵ The nitrogen atoms are in a planar environment as expected for a formal sp²-hybridized center with a p-type lone pair (LP) of electrons. Hence, the planarity of the core is imposed by the increasing bulkiness of the tBuAr*-moieties, as a bend core would result in pyramidalization about the N atoms to fit both ^RAr*-groups in. Just recently, the bonding in [E(μ-NTer)]₂ (E = Ge, Sn, Pb) was studied in detail by Ziegler et al., who analysed the interaction of the monomeric units E(μ-NTer) in the dimeric structure, with the result that the dimer is kept together by two σ- and π-bonds.¹⁶

Combining red 1tBu with two equivalents of SbCl₃ in CH₂Cl₂ resulted in an immediate decolourisation, accompanied by a colourless precipitate (Scheme 1, reaction (ii)), which was removed by filtration and from the filtrate X-ray quality crystals of trans-[ClSb(μ-NTer)]₂ (2) were grown overnight at room temperature. This metathesis route gives 2 reproducibly in good yields, while using the elimination of SnCl₂ as the driving force, which dates back to the seminal work of Veith,¹⁷ who established this route to prepare [Me₂SiECl(μ-NtBu)₂] ring systems (vide supra, Fig. 1 species II).¹⁸

Scheme 1 Preparation of 1R-4: (i) 2 ^RAr*NH₂, –2 HN(SiMe₃)₂, (ii) 2 SbCl₃, –2 SnCl₂, (iii) 4 SbCl₃, –2 SnCl₂, (iv) 2 SbCl₃, and (v) BiCl₃, –Sn.

Pale yellow crystals of 2 are moisture-sensitive, but indefinitely stable in an inert gas atmosphere and can be heated above 270 °C without decomposition. 2 crystallizes solvent-free in the triclinic space group P1̄ with one molecule in the unit cell and displays a trans-substituted centrosymmetric dimer with a planar Sb₂N₂ core protected by two bulky ^tBuAr* groups similar to the molecular structures of [XSb(μ-NMes*)]₂ X = F, Cl, Br, I; trans-[ClSb(μ-NtBu)]₂.^19,20 As expected the Sb atoms are trigonal pyramidally coordinated, with an s-type LP located on Sb and a trigonal planar coordination environment about the N atom. Additionally, one rather weak dipolar interaction between Sb and a flanking phenyl group (Sb⋯C_Ct = 3.29 Å, C_Ct = centroid) is detected (Fig. 2, left).²¹ The formation of 2 can be reproduced, however, if an excess of SbCl₃ is used, a new product ^tBuAr*N(SbCl₂)₂ (3) was isolated. Consequently, we reasoned that 3 was accessible directly from 1tBu (reaction (iii) in Scheme 1) when combined with four equiv. of SbCl₃, which yielded pure 3. Moreover, treatment of 2 with two additional equiv. of SbCl₃ also afforded (reaction (iv) in Scheme 1) 3 in good yields (78%). 3 is thermally stable and melts without decomposition at 236 °C and also shows distinct ¹H NMR shifts for the p-tBu, the CHPh₂ and the inner phenyl H atoms. Furthermore, 3 belongs to the family of N,N-bis(dichloropnictino)amines, which are well documented for phosphorus (RN(PCl₂)₂, R = Dipp, Trip, Ph).² Compound 3 was found to be monoclinic (P2₁/n) with one molecule of 3 and two disordered C₆H₅F solvents molecules in the asymmetric unit. The Sb–N distances of 2.030(2) and 2.039(2) Å are shorter than the sum of the covalent radii for Sb and N (cf. ∑r_cov(N–Sb) = 2.11 Å)²² representing highly polarized Sb–N single bonds. The trigonal planar N atom lies between both pyramidal SbCl₂ units, which adopt a trans configuration with respect to the SbCl₂ moieties (Fig. 2 right). Interestingly, two intramolecular Sb⋯Cl contacts (Sb1⋯Cl4, Sb2⋯Cl2 ca. 3.35 Å; cf. ∑r_vdW(N–Sb) = 3.81 Å),²³ stabilizing this trans configuration, but no intermolecular contacts are observed.

Fig. 2 Molecular structures of 1tBu (left), 2 (middle) and 3 (right). Thermal ellipsoids drawn at 50% probability and −100 °C. ^tBuAr* substituents rendered as wire-frame and H atoms omitted for clarity. Selected bond lengths (Å) and angles (°) of 1tBu: Sn1–N1 2.0752(16), 2.0897(16); N1–Sn1–N1′ 78.22(6); 2: Sb1–N1 2.033(2), Sb1–N1′ 2.034(2), Sb1–Cl1 2.4327(7), Sb1Sb1 3.1749(3), N1–C1 1.430(3) Å, ∑(<Sb) 273.05; ∑(<N) 359.83, C1–C2–N1–Sb1 77.6(2); 3: Sb1–N1 2.030(2), Sb1–Cl1 2.3709(7), Sb1–Cl2 2.4338(7), Sb2–N1 2.039(2), Sb2–Cl3 2.3731(7), Sb2–Cl4 2.4199(7), N1–C1 1.434(3), ∑(<Sb1) 280.08, ∑(<Sb2) 281.47, Sb1–N1–C1–C6 80.0(2).

In addition, the reaction of 1tBu with two equiv. of BiCl₃ was studied in CH₂Cl₂, resulting in a black reaction mixture (reaction (v) in Scheme 1). After multiple filtrations a clear orange solution was obtained. Recrystallization yielded small amounts of orange crystals that were identified as the hitherto unknown [BiSnCl₃(μ-N^tBuAr*)₂] (4). The black residue could not be conclusively identified and we assume that elemental tin is formed in a complex redox process that might also involve the formation of elemental bismuth (vide infra). It has been shown before that the Sn(II) center in [Me₂SiSn(μ-NtBu)₂] acts as a chloride acceptor in the coupling of phosphaalkenes²⁴ and in the reaction with chlorophosphanes.²⁵

Revision of the reaction conditions prompted us to repeat the experiment in C₆H₅F with one equivalent of BiCl₃ (with respect to 1tBu), to exclude a chloride-shift from CH₂Cl₂. This again resulted after filtration over a celite-padded frit and concentration of the filtrate in the deposition of orange crystals of 4 as a C₆H₅F solvate. Only small amounts of pure 4 could be isolated, therefore we cannot provide a comprehensive characterization. Nevertheless, the ¹¹⁹Sn NMR spectrum of these isolated crystals revealed a signal at 115.5 ppm (Fig. S13, ESI†), which is in the expected range for a hypercoordinate N₂Sn^(IV)Cl₃ moiety (cf. Me₃SnCl₂⁻: 47.7, Me₂SnCl₃⁻: 128 ppm, MeSnCl₄⁻: 274 ppm).²⁶ It should be noted that ¹¹⁹Sn NMR data strongly depend on substitution, coordination number and solvent giving rise to large chemical shift differences (cf. [SnCl₃{κ²-DippN(H)C₂H₄N(Dipp)}] −303 ppm).²⁷ According to MO and NBO analyses of the truncated model [BiSnCl₃(μ-NPh)₂], 4 can either be described as zwitterionic bismaallyl species (Lewis representation A/C in Fig. 4), as a bismuthenium species (E) or as an iminobismutane (B and D), and therefore represents the first neutral compound with a 4e–3c double bond delocalized along N–Bi–N (Fig. 4). In addition, an s-type lone pair (93%, see Fig. S14 and S15, ESI†) is located at the Bi center. Lewis representations A/C represent the best Lewis structures according to NBO analysis. Along with structures of type E/F, which also possess a rather large weight, since the π bonds are dominantly located at the N atoms (81%), this situation resembles that of N-heterocyclic carbenes (NHC),²⁸ which are stabilized by intramolecular π-donor–π-acceptor interactions (population of the p_z(Bi) = 0.47e) to stabilize the dicoordinate carbene C atom. It should be noted that also Bi–N σ bonds (78%) are highly polar, as well as the Sn–Cl or Sn–N bonds (N, Cl: ca. 80%). The computed large positive charges at the Bi and Sn centers are very similar with values of +1.67 and 1.77e supporting the picture of highly polarized Bi–N and Sn–Y (Y = Cl, N) bonds.

4 crystallizes as CH₂Cl₂ solvate (4·(CH₂Cl₂)₂) in the triclinic space group P1̄ with two molecules of 4 and four CH₂Cl₂ molecules (disordered on their positions) in the cell. Moreover, from C₆H₅F species 4 crystallizes as a solvate of fluorobenzene solvate (4·C₆H₅F) in the orthorhombic space group Pna2₁ (the discussion is led for 4·CH₂Cl₂). The most prominent structural feature is the planar 4-membered Sn–N–Bi–N heterocycle featuring two different heavy main group metals (deviation from planarity < 2.3°, Fig. 3). Both Bi–N bond lengths are rather short with 2.106(3) and 2.108(3) Å (cf. ∑r_cov(N–Bi) = 2.22, (N=Bi) = 2.01 Å;²² [Me₂SiBi(μ-NtBu)₂]⁺ 2.08 Å, [Bi(IBi)(μ-NTer)₂]⁺ 2.13 Å, and [Me₂SiBi(μ-NDipp)₂] 2.12 Å, where Dipp = 2,6-iPrC₆H₃)^4,5,29 clearly displaying some Bi–N double bond character in accord with our computation (Fig. 4). Interestingly, both Sn–N bond lengths (2.094(3) and 2.107(3) Å, cf. ∑r_cov(N–Sn) = 2.11, (N=Sn) = 1.90 Å) are in the similar range like the Bi–N distances, however, describing typical highly polarized Sn^(IV)–N single bonds. Both the N–Bi–N angle and N–Sn–N angles are rather acute with ca. 78° (cf. [Me₂SiBi(μ-NtBu)₂]⁺ 72.9, [Bi(IBi)(μ-NTer)₂]⁺ 77.4°, and [Me₂SiBi(μ-NDipp)₂]⁺ 73.7),^4,5,29 while the two Bi–N–Sn angles are much larger with 101–102°. A closer look at the secondary interactions revealed that the Sn–N–Bi–N heterocycle is well protected inside the pocket formed by the two ^tBuAr*-phenyl substituents. However, the dicoordinate bismuth is stabilized by strong secondary interactions (Menshutkin type π complexes)²¹ with two phenyl groups as indicated by very short Bi⋯centroid distances (2.891/2.978 Å; cf. [^MeAr*N(SiMe₃)BiCl][Al(OR^F)₄]⁺ 2.86/2.94 Å)⁸ which are well within the range of van-der-Waals radii (∑r_vdW(C⋯Bi) = 3.77 Å).²³

Fig. 3 Molecular structures of 4. Thermal ellipsoids drawn at 50% probability and −100 °C. ^tBuAr* substituents rendered as wire-frame and H atoms omitted for clarity. Selected bond lengths (Å) and angles (°) of 4: Sn1–N1 2.094(3), Sn1–N2 2.107(3), Sn1–Cl1 2.353(1), Sn1–Cl3 2.387(1), Sn1–Cl2 2.403(1), Sn1⋯Bi1 3.2631(4), Bi1–N1 2.106(3), Bi1–N2 2.108(3), N1–C37 1.425(5), N2–C1 1.426(5), N1–Sn1–N2 78.41(12), N1–Bi1–N2 78.10(12), ∑(<N1) 358.0, ∑(<N2) 353.4, Bi1–C_Ct1 2.891, Bi1–C_Ct2 2.978 Å.

Fig. 4 Selected Lewis representations of 4.

In conclusion, we succeeded in the preparation of the first N,N′-bis(dichlorostibinino)amine and an unusual heterocycle containing Sn^(IV) and a dicoordinate Bi-center, which is protected by arene-interactions to flanking phenyl groups of the bulky Ar* moiety. These species might be useful starting materials for the preparation of pnictadiazonium salts of Sb and Bi. In comparison to stable N-heterocyclic carbenes,²⁸ the dicoordinated Bi species 4 can be regarded as a heavy atom analog of NHCs.

DFG (SCHU 1170/11-1) is gratefully acknowledged for financial support. C. H.-J. thanks the Fonds der chemischen Industrie for financial support. The authors thank MSc Jonas Bresien for setting up and maintaining Gaussian and NBO software on the cluster computer.

Notes and references

1. (a) G. He, O. Shynkaruk, M. W. Lui and E. Rivard, Chem. Rev., 2014, 44, 7815–7880; (b) M. S. Balakrishna, D. J. Eisler and T. Chivers, Chem. Soc. Rev., 2007, 36, 650–664.
2. (a) F. Reiß, A. Schulz, A. Villinger and N. Weding, Dalton Trans., 2010, 39, 9962; (b) C. Ganesamoorthy, M. S. Balakrishna, J. T. Mague and H. M. Tuononen, Inorg. Chem., 2008, 47, 7035–7047; (c) N. Burford, C. T. Stanley, K. D. Conroy, B. Ellis, C. L. B. MacDonald, R. Ovans, A. D. Phillips, P. Ragogna and D. Walsh, Can. J. Chem., 2002, 80, 1404–1409; (d) V. D. Romanenko, A. B. Drapailo, A. N. Chernega and L. N. Markovskii, Zh. Obshch. Khim., 1991, 61, 2434–2441; (e) S. Goldschmidt and H.-L. Krauß, Liebigs Ann. Chem., 1955, 595, 193–202.
3. D. Michalik, A. Schulz and A. Villinger, Angew. Chem., Int. Ed., 2010, 46, 7575–7577 ( Angew. Chem. , 2010 , 122 , 7737–7740 ).
4. M. Veith, B. Bertsch and V. Huch, Z. Anorg. Allg. Chem., 1988, 559, 73–88.
5. M. Lehmann, A. Schulz and A. Villinger, Angew. Chem., Int. Ed., 2012, 51, 8087–8091 ( Angew. Chem. , 2012 , 124 , 8211–8215 ).
6. (a) D. Michalik, A. Schulz, A. Villinger and N. Weding, Angew. Chem., Int. Ed., 2008, 47, 6465–6468 ( Angew. Chem. , 2008 , 120 , 6565–6568 ); (b) A. Schulz and A. Villinger, Inorg. Chem., 2009, 48, 7359–7367.
7. (a) T. Beweries, R. Kuzora, U. Rosenthal, A. Schulz and A. Villinger, Angew. Chem., Int. Ed., 2011, 50, 8974–8978 ( Angew. Chem. , 2011 , 123 , 9136–9140 ); (b) S. Demeshko, C. Godemann, R. Kuzora, A. Schulz and A. Villinger, Angew. Chem., Int. Ed., 2013, 52, 2105–2108 ( Angew. Chem. , 2013 , 125 , 2159–2162 ); (c) A. Hinz, A. Schulz and A. Villinger, Chem. – Eur. J., 2014, 20, 3913–3916; (d) A. Hinz, R. Kuzora, A. Schulz and A. Villinger, Chem. – Eur. J., 2014, 20, 14659–16673; (e) A. Hinz, A. Schulz and A. Villinger, Chem. Commun., 2015, 51, 1363–1366.
8. C. Hering-Junghans, M. Thomas, A. Schulz and A. Villinger, Chem. – Eur. J., 2015, 21, 6713–6717.
9. J. Li, A. Stasch, C. Schenk and C. Jones, Dalton Trans., 2011, 40, 10448–10456.
10. J. Li, C. Schenk, F. Winter, H. Scherer, N. Trapp, A. Higelin, S. Keller, R. Pöttgen, I. Krossing and C. Jones, Angew. Chem., Int. Ed., 2012, 51, 9557–9561 ( Angew. Chem. , 2012 , 124 , 9695–9699 ).
11. J. Li, C. Schenk, C. Goedecke, G. Frenking and C. Jones, J. Am. Chem. Soc., 2011, 133, 18622–18625.
12. T. J. Hadlington and C. Jones, Chem. Commun., 2014, 50, 2321.
13. (a) T. J. Hadlington, M. Hermann, J. Li, G. Frenking and C. Jones, Angew. Chem., Int. Ed., 2013, 52, 10199–10203 ( Angew. Chem. , 2013 , 125 , 10389–10393 ); (b) T. J. Hadlington, M. Hermann, G. Frenking and C. Jones, J. Am. Chem. Soc., 2014, 136, 3028–3031.
14. D. Dange, A. Davey, J. A. B. Abdalla, S. Aldridge and C. Jones, Chem. Commun., 2015, 51, 7128–7131.
15. W. A. Merrill, R. J. Wright, C. S. Stanciu, M. M. Olmstead, J. C. Fettinger and P. P. Power, Inorg. Chem., 2010, 49, 7097–7105.
16. M. Brela, A. Michalak, P. P. Power and T. Ziegler, Inorg. Chem., 2014, 53, 2325–2332.
17. M. Veith, Angew. Chem., Int. Ed. Engl., 1975, 14, 265–266 ( Angew. Chem. , 1975 , 87 , 287–288 ).
18. M. Veith and B. Bertsch, Z. Anorg. Allg. Chem., 1988, 557, 7–22.
19. M. Lehmann, A. Schulz and A. Villinger, Eur. J. Inorg. Chem., 2010, 5501–5508.
20. D. J. Eisler and T. Chivers, Inorg. Chem., 2006, 45, 10734–10742.
21. H. Schmidbaur and A. Schier, Organometallics, 2008, 27, 2361–2395.
22. P. Pyykkö and M. Atsumi, Chem. – Eur. J., 2009, 15, 12770–12779.
23. M. Mantina, A. C. Chamberlin, R. Valero, C. J. Cramer and D. G. Truhlar, J. Phys. Chem. A, 2009, 113, 5806–5812.
24. E. Niecke, H. J. Metternich and R. Streubel, Eur. J. Inorg. Chem., 1990, 67–69.
25. J. K. West and L. Stahl, Organometallics, 2012, 31, 2042–2052.
26. (a) P. J. Smith and A. P. Tupciauskas, Annu. Rep. NMR Spectrosc., 1978, 8, 291; (b) G. F. Hewitson, Master thesis, Durham University, 1980; (c) J. Ortera, J. Org. Chem., 1981, 221, 57.
27. S. M. Mansell, C. A. Russell and D. F. Wass, Dalton Trans., 2015, 44, 9756–9765.
28. (a) D. Bourissou, O. Guerret, F. P. Gabbaï and G. Bertrand, Chem. Rev., 2000, 100, 39–92; (b) N. Marion and S. P. Nolan, Acc. Chem. Res., 2008, 41, 1440–1449; (c) D. Bézier, J.-B. Sortais and C. Darcel, Adv. Synth. Catal., 2013, 355, 19–33.
29. R. J. Schwamm, B. M. Day, M. P. Coles and C. M. Fitchett, Inorg. Chem., 2014, 53, 3778.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental and computational details and information on X-ray structure elucidation. CCDC 1403993–1403998. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c5cc04516a

---