Open Access Article
This Open Access Article is licensed under a Creative Commons Attribution-Non Commercial 3.0 Unported Licence

Reversible P–P bond cleavage at an iridium(III) metal centre

Simon J. Coles a, Peter N. Horton a, Patrick Kimber b, Wim T. Klooster a, Pingchuan Liu b, Felix Plasser b, Martin B. Smith *b and Graham J. Tizzard a
aUK National Crystallography Service, School of Chemistry, University of Southampton, Southampton, SO17 1BJ, UK
bDepartment of Chemistry, Loughborough University, Loughborough, Leics LE11 3TU, UK. E-mail: m.b.smith@lboro.ac.uk

Received 4th February 2022 , Accepted 7th April 2022

First published on 19th April 2022


Abstract

Treatment of a κ1-P-monodentate bicyclic diphosphane iridium(III) complex with a labile gold(I) precursor afforded an unusual IrIII/AuI complex in which the P–P single bond has been cleaved. This reaction was cleanly reversed upon addition of tertiary phosphine. Carbon–carbon bond activation, across neighbouring P2C2N rings of the coordinated bicyclic diphosphane, occurred upon thermolysis of the IrIII/AuI complex.


Phosphorus based compounds underpin many significant advances that have been achieved in the fields of synthesis, reactivity and mechanistic understanding, and in catalysis. In this context, the ubiquitous tertiary phosphines stand out as excellent examples of highly tuneable ligands for various transition metals. By contrast, the coordination chemistry of P-based compounds such as terminal/bridging phosphides (PR2)1 and, to a lesser extent, diphosphanes (R2P–PR2),2 have witnessed relatively few studies. In spite of this, there is a growing interest3 in new synthetic routes to R2P–PR2 compounds given also the importance of P–P bonded species in organophosphorus chemistry. Metal bound phosphide ligands can undergo P–P radical coupling leading to dimeric products,4 in a few instances this process is reversible (Chart 1).5 Furthermore, diphosphanes have been shown to undergo homolytic P–P bond cleavage affording R2P˙ radicals.6 The reactivity of R2P–PR2 towards soft transition metals has been shown to result in rapid P–P bond cleavage and formation of terminal or bridging phosphido ligands (Chart 1). A terminal phosphido RhV intermediate has been implicated in the dehydrocoupling of HPR2 affording R2P–PR2.7 Tsipis and co-workers also reported the reversible transformation of two bridging PPh2 ligands to a neutral, μ-Ph2P–PPh2 ligand at a trinuclear Pt2Pd core.8 Herein we describe the intramolecular P–P cleavage of a well-defined IrIII1-R2P–PR2) complex upon treatment with a labile gold(I) source and, moreover, shown this to be a reversible process. Under thermal conditions we also observed an unexpected C–C bond activation across both P2C2N rings of a coordinated bicyclic diphosphane.
image file: d2cc00706a-c1.tif
Chart 1 Interplay between PR2 and R2P–PR2 (R = alkyl, aryl) metal chemistry via oxidative P–P coupling/cleavage reactions.

Various groups2,3,6,9–12 have reported the synthesis and reactivity of several types of R2P–PR2, with acyclic variants being most common. We reasoned a “rigid” R2P–PR2 would reduce P–P bond rotation/dissociation yet still function as a bridging ligand. Accordingly, the new air stable bicyclic diphosphane, P–P(OMe), was readily synthesised in 44% yield by a one-step reaction between [P(CH2OH)4]Cl and 4-H2NC6H4OMe (see ESI for details).13,14 The 31P{1H} NMR spectrum of a CDCl3 solution of P–P(OMe) displayed a singlet at δ −34.7 ppm and the single crystal X-ray structure (Fig. 1) confirmed a bicyclic five-membered arrangement with a typical P–P bond length of 2.2042(7) Å.12 The nitrogen atoms are nearly planar (sum of angles approx. 350°) whilst the C(1)–P(1)–P(1′)–C(1′) torsion angle between both five-membered rings is 118.65(6)°. Conformational restriction imposed by the two adjacent, five-membered, P2C2N rings dictates the orientation of the two phosphorus lone pairs are syn disposed.


image file: d2cc00706a-f1.tif
Fig. 1 Molecular structure of P–P(OMe) in the solid state. Thermal ellipsoids are drawn at the 50% probability level. All hydrogen atoms have been omitted for clarity. Selected bond length [Å]: P(1)–P(1′) 2.2042(7) Å. Symmetry operator: 1 − x, 1 y, +z.

Reaction of 2 equiv. of P–P(OMe) and [IrCl(μ-Cl)(η5-C5Me5)]2 in CH2Cl2 gave 1 in excellent yield (88%) as a yellow solid (Scheme 1). The 31P{1H} NMR spectrum showed the expected AX spectrum consistent with κ1-mode (IrIII–Pcoord, δ 15.8 ppm; Pfree, δ −56.3 ppm, JPP = 230 Hz) and was further supported by the appearance of two 13C signals for the methylene carbon atoms (δ 54.5 ppm, JPC = 28 Hz; δ 52.1 ppm, JPC = 24 Hz). Reaction of 1 equiv. of P–P(OMe) and [IrCl(μ-Cl)(η5-C5Me5)]2 in CH2Cl2 gave the dinuclear IrIII complex 1′ which showed a singlet at δ −7.2 ppm consistent with a highly symmetric bridged structure and two P-bound {IrCl25-C5Me5)} groups. The X-ray structure of 1 (Fig. 2) shows a slightly contracted P–P bond length of 2.1836(19) Å, with respect to P–P(OMe), whilst the nitrogen atoms in 1 are nearly planar [sum of angles 351° around N(1); 345° around N(2)].


image file: d2cc00706a-s1.tif
Scheme 1 Synthesis of P–P(OMe), 1, 1′ and 2. Reaction conditions: (i) 2 equiv. P–P(OMe)/[IrCl(μ-Cl)(η5-C5Me5)] (ii) 1 equiv. P–P(OMe)/[IrCl(μ-Cl)(η5-C5Me5)] (iii) AuCl(tht) (iv) 1 equiv. PPh3, – AuCl(PPh3). R = 4-C6H4OMe; Cp* = η5-C5Me5.

image file: d2cc00706a-f2.tif
Fig. 2 Molecular structure of 1 in the solid state. Thermal ellipsoids drawn are at the 50% probability level. All hydrogen atoms have been omitted for clarity. Selected bond lengths [Å] and angles [°]: Ir(1)–P(1) 2.2705(13), Ir(1)–Cl(1) 2.4034(13), Ir(1)–Cl(2) 2.4155(15), P(1)–P(2) 2.1836(19); Cl(1)–Ir(1)–P(1) 89.25(5), Cl(2)–Ir(1)–P(1) 82.33(5), Cl(1)–Ir(1)–Cl(2) 89.49(5).

Given the available free P-coordination site in 1 we sought to explore whether a second, different, soft metal centre could be supported on a R2P–PR2 ligand. Hence, reaction of 1 with AuCl(tht) (tht = tetrahydrothiophene) in CH2Cl2 at r.t. gave, instead, compound 2 in excellent yield (83%) as a pale yellow solid. The 31P NMR spectrum of 2 showed an AX pattern, with a significantly reduced JPP coupling of 53.4 Hz, consistent with two distinct, non-bonded, P centres and further supported by the 1H NMR spectrum which showed two sets of aromatic resonances consistent with non-equivalent –C6H4OMe rings.

The most significant feature of the X-ray structure of 2 (Fig. 3) is insertion15 of an “Ir(η5-C5Me5)Cl” fragment, coupled with P–P bond cleavage [P⋯P interatom separation is 2.8839(18)/2.8941(18) Å] and migration of Cl to P(1)16 and a terminal AuCl group on P(2). The nitrogen atoms are significantly more distorted (sum of angles 340–346°) and there is a strong aurophilic Au⋯Au interaction [2.9278(3) Å] between neighbouring molecules.17 The bond lengths for Ir(1)–P(2)/P(2)–Au(2)/Au(1)–Cl(3) are similar to reported examples18,19 whilst the Ir(1)–P(1)/Ir(1)–P(2) bond distances differ reflecting the bonding arrangements of both phosphorus donors [P(2) bridges both AuI and IrIII centres].


image file: d2cc00706a-f3.tif
Fig. 3 Molecular structure of 2·CDCl3 in the solid state. Thermal ellipsoids are drawn at the 50% probability level. All hydrogen atoms and solvent have been omitted for clarity. Selected bond lengths [Å] and angles [°] (values in parentheses are for the second independent molecule): Ir(1)–P(1) 2.2111(13) [2.2095(14)], Ir(1)–Cl(2) 2.3965(12) [2.3967(13)], Ir(1)–P(2) 2.3318(13) [2.3381(13)], Au(1)–P(2) 2.2429(12) [2.2525(13)]; Cl(3)–Au(1) 2.3252(12) [2.3273(13)], P(1)–Cl(1) 2.0389(17) [2.0510(18)]; P(1)–Ir(1)–P(2) 78.76(5) [78.99(5)], Cl(2)–Ir(1)–P(1) 92.21(5) [95.04(5)], Cl(2)–Ir(1)–P(2) 88.83(5) [87.70(5)], Cl(3)–Au(1)–P(2) 172.32(5) [174.09(5)].

In order to gain more insight into the reaction pathway leading to 2 from 1via2′, we undertook density functional theory calculations (see ESI for details). An accessible transition state between 2′ and 2 (TS, 18.6 kcal mol−1), following initial migration of a chlorine atom to P(1), with the P–P bond intact was located and optimised (Fig. 4). The formation of 2 from 2′ is exothermic by −3.5 kcal mol−1. It is unlikely steric factors are solely important here given that P–P(OMe) can accommodate two, bulky, {IrCl(η5-C5Me5)} groups (as in 1′) across the P–P single bond vector suggesting electronic effects contribute. Natural bond orbital (NBO) analysis for 2 (see ESI for details) confirmed a near perfect sp3 hybridisation of P(2) as a phosphido with a dative bond to AuCl. Conversely, in 2′ enhanced p orbital character is associated to the P–P single bond, leaving the P(2) lone pair sp hybridised and unable to form a strong Au–P bond.


image file: d2cc00706a-f4.tif
Fig. 4 Energy profile (kcal mol−1) for the P–P cleavage of 2.

When 2 was treated with 1 equiv. of PPh3, in CDCl3, clean conversion to 1 occurred (as monitored by in situ31P{1H} NMR, Fig. 5), accompanied by concomitant formation of AuCl(PPh3) (δ 33.8 ppm). No reaction took place between 1 and AuCl(PPh3) acting as the source of “AuCl”.


image file: d2cc00706a-f5.tif
Fig. 5 31P{1H} NMR spectra of (a) complex 1 (b) complex 2 (c) in situ addition of 1 equiv. PPh3 to 2 (d) AuCl(PPh3).

Whilst complex 1 shows good thermal stability (C7H8, reflux, 24 h), heating a CDCl3 solution of 2 for 4 d at ca. 50 °C did not result in any observable formation of 1. Instead, in situ monitoring by 31P{1H} NMR (Scheme 2) revealed the appearance of a new AX pattern (δ 50.5 ppm, −30.2 ppm, JPP = 55.2 Hz), along with a second minor species and significant amounts of decomposition products. The 1H NMR spectrum of 3, after fractional crystallisation (∼20% isolated yield) from CDCl3 solution, showed seven unique aliphatic hydrogens suggesting formation of a new C–C bond across neighbouring P2C2N rings. The X-ray structure of 3 (Fig. 6) again confirmed the absence of a P–P bond [P⋯P separation 2.943(3) Å] and on P(1) a chlorine atom and P(2) a terminal AuCl group. The nitrogen atoms are significantly more distorted [sum of angles 343° around N(1); 346° around N(2)]. Most significantly a new C–C single bond [1.511(11) Å] leads to a unusual bicyclic ligand framework. The closest Au⋯Au interaction between neighbouring molecules is approx. 10 Å, significantly different form that observed for 2. Treatment of 3 with 1 equiv. of PPh3 afforded cleanly 4, as two diastereomers (ca. 50[thin space (1/6-em)]:[thin space (1/6-em)]50), as shown by their diagnostic AX patterns (in situ31P{1H} NMR studies, see ESI for details).


image file: d2cc00706a-s2.tif
Scheme 2 Thermal conversion of 2 to 3 and reversible P–P coupling to generate 4 as a pair of diastereomers. Reaction conditions: (i) heat, CDCl3 (ii) 1 equiv. PPh3, – AuCl(PPh3). Cp* = η5-C5Me5.

image file: d2cc00706a-f6.tif
Fig. 6 Molecular structure of 3·CDCl3 in the solid state. Thermal ellipsoids are drawn at the 50% probability level. All hydrogen atoms and solvent have been omitted for clarity. Selected bond lengths [Å] and angles [°]: Ir(1)–P(1) 2.233(2), Ir(1)–Cl(2) 2.403(2), Ir(1)–P(2) 2.328(2), Au(1)–P(2) 2.245(2), Au(1)–Cl(3) 2.302(2) P(1)–Cl(1) 2.063(3); P(1)–Ir(1)–P(2) 80.34(8), Cl(2)–Ir(1)–P(1) 91.32(8), Cl(2)–Ir(1)–P(2) 90.75(8), Cl(3)–Au(1)–P(2) 177.60(8).

In conclusion, this study highlights the reversible, intramolecular insertion of a late transition metal entity across a P–P single bond of a coordinated diphosphane. By virtue of flexible, neighbouring arene substituents on the diphosphane, we observed an unprecedented Csp2–Csp3 coupling affording a highly functionalised P-ligand. Further studies will be directed towards exploring the ligand scope, reactivity and P–P/C–C mechanistic aspects of these transformations.

We thank Johnson Matthey for their donation of IrCl3·nH2O, Solvay for kindly providing tetrahydroxymethylphosphonium chloride (THPC) and the UK National Crystallography Service at the University of Southampton for all four data collections and processing. We acknowledge the Lovelace high-performance computing cluster at Loughborough University.

Conflicts of interest

There are no conflicts of interest to declare.

Notes and references

  1. (a) C. Fortuño, A. Martin, P. Mastrorilli, M. Latronico, V. Petrelli and S. Todisco, Dalton Trans., 2020, 49, 4935–4955 RSC; (b) J. Yang, S. Langis-Barsetti, H. C. Parkin, R. McDonald and L. Rosenberg, Organometallics, 2019, 38, 3257–3266 CrossRef CAS; (c) K. Kaniewska, A. Dragulescu-Andrasi, L. Ponikiewski, J. Pikies, S. A. Stoian and R. Grubba, Eur. J. Inorg. Chem., 2018, 4298–4308 CrossRef CAS; (d) U. Fischbach, M. Trincado and H. Grützmacher, Dalton Trans., 2017, 46, 3443–3448 RSC.
  2. (a) S. Molitor, C. Mahler and V. H. Gessner, New J. Chem., 2016, 40, 6467–6474 RSC; (b) D. Tofan and C. C. Cummins, Chem. Sci., 2012, 3, 2474–2478 RSC.
  3. For a recent example, see: I. Elser, R. J. Andrews and D. W. Stephan, Chem. Commun., 2022, 58, 1740–1743 RSC.
  4. M. W. Bezpalko, B. M. Foxman and C. M. Thomas, Inorg. Chem., 2013, 52, 12329–12331 CrossRef CAS PubMed.
  5. (a) S. Kim and Y. Lee, Inorg. Chem. Front., 2020, 7, 1172–1181 RSC; (b) Y.-E. Kin and Y. Lee, Angew. Chem., Int. Ed., 2018, 57, 14159–14163 CrossRef PubMed; (c) A. Tohmé, G. Grelaud, G. Argouarch, T. Roisnel, S. Labouille, D. Carmichael and F. Paul, Angew. Chem., Int. Ed., 2013, 52, 4445–4448 CrossRef PubMed.
  6. (a) C. Branfoot, T. A. Young, D. F. Wass and P. G. Pringle, Dalton Trans., 2021, 50, 7094–7104 RSC; (b) M. Blum, O. Puntigam, S. Plebst, F. Ehret, J. Bender, M. Nieger and D. Gudat, Dalton Trans., 2016, 45, 1987–1997 RSC; (c) J.-D. Guo, S. Nagase and P. P. Power, Organometallics, 2015, 34, 2028–2033 CrossRef CAS; (d) N. A. Griffin, A. D. Hendsbee, T. L. Roemmele, M. D. Lumsden, C. C. Pye and J. D. Masuda, Inorg. Chem., 2012, 51, 11837–11850 CrossRef PubMed.
  7. V. P. W. Böhm and M. Brookhart, Angew. Chem., Int. Ed., 2001, 40, 4694–4696 CrossRef.
  8. J. Forniés, C. Fortuño, S. Ibáñez, A. Martín, A. C. Tsipis and C. A. Tsipis, Angew. Chem., Int. Ed., 2005, 44, 2407–2410 CrossRef PubMed.
  9. S. Molitor, J. Becker and V. H. Gessner, J. Am. Chem. Soc., 2014, 136, 15517–15520 CrossRef CAS PubMed.
  10. D. M. C. Ould, T. T. P. Tran, J. M. Rawson and R. L. Melen, Dalton Trans., 2019, 48, 16922–16935 RSC.
  11. A. K. King, A. Buchard, M. F. Mahon and R. L. Webster, Chem. – Eur. J., 2015, 21, 15960–15963 CrossRef CAS PubMed.
  12. N. Szynkiewicz, L. Ponikiewski and R. Grubba, Dalton Trans., 2018, 47, 16885–16894 RSC.
  13. A. W. Frank and G. L. Drake, Jr., J. Org. Chem., 1977, 42, 4125–4127 CrossRef CAS . We have shown, by 31P NMR studies, that this reaction proceeds via the intermediate P{CH2N(H)C6H4OMe}3 and we also observe the phosphine oxide. Due to the low solubility of P-P(OMe) in methanol, this compound readily separates from other methanol soluble P-based compounds. This reaction has previously been observed only for C6H5NH2 (see ref. 14).
  14. A. W. Frank and G. L. Drake, Jr., J. Org. Chem., 1972, 37, 2752–2755 CrossRef CAS.
  15. R. Franz, S. Nasemann, C. Bruhn, Z. Kelemen and R. Pietschnig, Chem. – Eur. J., 2021, 27, 641–648 CrossRef CAS PubMed.
  16. M. J. Ray, M. Bühl, L. J. Taylor, K. S. Athukorala Arachchige, A. M. Z. Slawin and P. Kilian, Inorg. Chem., 2014, 53, 8538–8547 CrossRef CAS PubMed.
  17. P. Pyykko, Chem. Soc. Rev., 2008, 37, 1967–1997 RSC.
  18. Y. Gloaguen, W. Jacobs, B. de Bruin, M. Lutz and J. I. van der Vlugt, Inorg. Chem., 2013, 52, 1682–1684 CrossRef CAS PubMed.
  19. M. Brym, C. Jones and J. D. E. T. Wilton-Ely, Inorg. Chem., 2005, 44, 3275–3282 CrossRef CAS PubMed.

Footnote

Electronic supplementary information (ESI) available: Synthetic procedures, spectroscopic/analytical data, single crystal X-ray crystallographic data, and computational details. CCDC 2143419, 2143421, 2143424 and 2143445. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d2cc00706a

This journal is © The Royal Society of Chemistry 2022
Click here to see how this site uses Cookies. View our privacy policy here.