UV-light promoted C–H bond activation of benzene and fluorobenzenes by an iridium(I) pincer complex

Abstract

Iridium(I) carbonyl complex [Ir(2,6-(P^tBu₂CH₂)₂C₆H₃)(CO)] undergoes reversible C–H bond activation of benzene and a series of fluorobenzenes on UV irradiation. Exclusive ortho-selectivity is observed in reactions of fluorobenzene and 1,2-difluorobenzene.

---

Epitomised by applications in the catalytic dehydrogenation of alkanes, iridium complexes of phosphine-based pincer ligands are widely recognised for their capacity to activate C–H bonds.¹ With fluoroaryls representing valuable synthons in organic chemistry,^2,3 we have targeted use of these iridium compounds for carrying out selective C–H bond activation reactions of partially fluorinated benzenes (C₆H_6−nF_n, n ≤ 3).³ The presence of fluorine substituents results in significantly stronger C–H bonds than benzene and, correspondingly, fluorobenzenes represent challenging substrates.^3,4 Previous work by Milstein and Ozerov employing neutral (PNP) and anionic (PNP*) pincer ligands has highlighted the potential of iridium pincers, although under moderate temperature regimes (<100 °C) these systems showed poor regioselectivity in the activation of fluorobenzene (Scheme 1).⁵ As C–H bond activation is thermodynamically favoured ortho to the fluorine substituents,⁴ indiscriminate and irreversible oxidative addition reactions of transient 14 VE Ir(I) intermediates {Ir(PNP)}⁺/{Ir(PNP*)} are implicated.

Scheme 1 C–H bond activation of fluorobenzene using Ir-pincers. PNP = 2,6-(P^tBu₂CH₂)₂C₅H₃N, PNP* = (4-Me-2-(ⁱPr₂P)C₆H₃)₂N⁻, COE = cyclooctene, NBE = norbornene, NBA = norbornane.

We postulated that use of anionic pincer ligands bearing central aryl donors could promote selective activation of fluorobenzene substrates through increased reaction reversibility imparted by the high trans-influence aryl donor.^6,7 With a view to testing this hypothesis we selected [Ir(PCP)(CO)] 1 (PCP = 2,6-(P^tBu₂CH₂)₂C₆H₃⁻)⁸ as a well-defined precursor for the low coordinate and formally 14 VE Ir(I) fragment {Ir(PCP)}, through photochemically promoted dissociation of the carbonyl ligand. In this way, subsequent products of C–H bond activation would be trapped on re-coordination of the carbonyl ligand (in a closed system). Initial experiments using benzene as the substrate supported this reasoning, with [Ir(PCP)(C₆D₅)D(CO)] 2-d₆⁹ generated on irradiation of a 20 mM C₆D₆ solution of 1 at RT using a 100 W Hg arc lamp (quartz J. Young's NMR tube, Scheme 2). Following this reaction by periodic analysis using ³¹P NMR spectroscopy, however, indicated that conversion of 1 (δ_31P 82.0) to 2-d₆ (δ_31P 52.3) plateaued at 62% after ca. 2 h total irradiation. On the same timeframe, irradiation of independently synthesised 2 resulted in an equivalent reaction composition. In contrast, both 1 and 2 are thermally stable on extended heating at 80 °C in C₆D₆ solution (8 h) and no isotope exchange was observed for 2 (to 2-d₆).¹⁰ Together these results indicate establishment of a photostationary mixture of 1 and 2-d₆, mediated through light induced carbonyl dissociation from both species.¹¹

Scheme 2 UV-promoted interconversion between 1 and 2 ([Ir] = 20 mM, RT).

To gain deeper understanding of the photolysis experiments, a series of DFT and TD-DFT calculations were performed (see ESI† for full details). In line with expectation, the computed free energy for carbonyl dissociation to form {Ir(PCP)} is a significantly endergonic process (ΔG_298K = +194.4 kJ mol⁻¹). While the subsequent C–H bond activation of benzene is exothermic (ΔH = −15.3 kJ mol⁻¹), the Ir(III) product lies thermodynamically further uphill from 1 (ΔG_298K = +238.0 kJ mol⁻¹). Such energetics are characteristic of an unfavourable equilibrium reaction, although one that would be offset by the use of the substrate as the solvent.¹² Re-coordination of the carbonyl ligand counteracts the unfavourable thermodynamics (ΔG_298K = −95.0 kJ mol⁻¹), however, 2 is still calculated to be +141.3 kJ mol⁻¹ higher in free energy than 1 + C₆H₆. Together these results are consistent with the lack of any thermal reaction observed for either 1 or 2 in C₆D₆ and highlight the important promoting role of UV-irradiation in the formation of 2 (and reformation of 1). In this context, analysis of 1 by TD-DFT identified a number of singlet–singlet electronic transitions between 195 and 235 nm (i.e. UV) that can be attributed to carbonyl dissociation. A representative example is shown in Fig. 1 (full details provided in ESI†). Similar excitations are also calculated for 2 between 190 and 260 nm, suggesting that it would be difficult to selectively enact the UV-promoted dissociation of carbonyl from either 1 or 2.¹³

Fig. 1 Representative electronic excitation related to carbonyl dissociation from 1; wavelength, oscillator strength (f) and % contribution of the represented natural transition orbitals (rendered with an orbital isosurface value of 0.02).

To further explore this C–H bond activation chemistry, 20 mM solutions of 1 in fluorobenzene, 1,2-difluorobenzene, and 1,3,5-trifluorobenzene were irradiated for a total of 8 h at RT (Scheme 3). Similar to that observed in benzene, analysis by ¹H and ³¹P NMR spectroscopy indicated formation of photostationary mixtures composed of 1 and Ir(III) products of C–H bond activation 3; as the minor and major components, respectively. Supporting our hypothesis, and contrasting with reactions of analogous PNP and PNP* systems (vide supra), C–H bond activation of fluorobenzene (3a) and 1,2-difluorobenzene (3b) proceeded with exclusive ortho-selectivity. Both possible rotamers of each isomer are formed, although in disparate proportions (Scheme 3).

Scheme 3 UV-promoted C–H bond activation of fluorobenzenes (top; [Ir] = 20 mM, 8 h, RT). Major rotamers depicted in red, minor rotamers depicted in blue (bottom).

Consistent with these observations and manifestation of the “ortho fluorine effect”,⁴ the alternative regioisomers of 3a (>10 kJ mol⁻¹) and 3b (>18 kJ mol⁻¹), and their respective 5-coordinate precursors [Ir(PCP)(2-FC₆H₄)H] (>24 kJ mol⁻¹) and [Ir(PCP)(2,3-F₂C₆H₃)H] (>26 kJ mol⁻¹), are calculated to be significantly higher in free energy. As for the formation of 2 from 1, computed reaction free energies suggest the formation of 3 occur outside of thermally accessible regimes and instead are presumably driven through nuanced differences in the photophysical properties of 1 and 3. Moreover, increasingly favourable overall reactions correlate with number of fluorine substituents: ΔG_298K/kJ mol⁻¹ = +141.3 (2), +128.1 (3a), +121.0 (3b), +119.1 (3c).

With the above in mind and to independently verify their structures, analytically pure samples of 3 were isolated though dehydrohalogenation of [Ir(PCP)HCl]¹⁴ using K[O^tBu] in the respective fluorobenzene at 75 °C, followed by reaction with carbon monoxide (yield = 32–54%). The structures of these new compounds, and for comparison 2,⁹ were fully verified in solution by ¹H, ¹³C, ¹⁹F and ³¹P NMR, IR and UV-vis spectroscopy, and in the solid-state by X-ray crystallography (3a, 3b – Fig. 2; 2, 3c – ESI†). Notable spectroscopic markers include low frequency hydride resonances at δ −9.07 (2), −8.86/−9̲.̲6̲8̲ (3a), −8.88/−9̲.̲6̲6̲ (3b), −9.46 (3c) that show ³J_PH coupling of ca. 17 Hz, and ³¹P resonances at δ 51.1 (2), 54.5/5̲1̲.̲7̲ (3a), 54.7/5̲2̲.̲2̲ (3b), 54.0 (3c) in C₆D₁₂ solution (data for major rotamers underlined). The structures of major rotamers of 3a and 3b, bearing the fluorine atoms proximal to the carbonyl ligand, were definitively established in solution through NOESY experiments and corroborated in the solid-state (Fig. 2). Likewise, the hydride resonances of the minor rotamers show characteristic ^1hJ_FH coupling of ca. 10 Hz. The observed relative conformational preferences are also reproduced in silico (see ESI†). Complexes 3 are thermally stable on extended heating at 80 °C in C₆D₆ solution: after 8 h, <5% 1 is formed and for 3a and 3b the ratio of rotamers was unchanged, as determined in situ by ³¹P NMR spectroscopy.

Fig. 2 Solid-state structures of 3a (occupancy of disordered fluorine substituents in parenthesis) and 3b. Thermal ellipsoids at 50% probability level; hydrogen atoms omitted for clarity. Selected bond lengths [Å] and angles [°]: 3a, Ir1–C2, 1.931(6); Ir1–C4, 2.159(5); Ir1–C12, 2.115(5); Ir1–P10, 2.3394(15); Ir1–P11, 2.3474(15); C2–Ir1–C12, 83.8(2); C4–Ir1–C12, 178.2(2); P10–Ir1–P11, 155.84(5). 3b, Ir1–C2, 1.928(2); Ir1–C4, 2.150(2); Ir1–C12, 2.111(2); Ir1–P10, 2.3511(6); Ir1–P11, 2.3397(6); C2–Ir1–C12, 89.07(9); C4–Ir1–C12, 173.63(8); P10–Ir1–P11, 157.41(2).

In conclusion, we have demonstrated that selective C–H bond activation reactions of fluorobenzenes can be achieved under mild conditions using photolysis of [Ir(2,6-(P^tBu₂CH₂)₂C₆H₃)(CO)] 1 as a means to generate the reactive 14 VE Ir(I) fragment {Ir(2,6-(P^tBu₂CH₂)₂C₆H₃)} in solution. This work not only showcases the ability of iridium pincer complexes to mediate challenging C–H bond activations, but more generally highlights a potentially useful catalyst design principle that we hope will stimulate the development of new organic transformations employing partially fluorinated benzenes as building blocks.

We gratefully acknowledge financial support from the Swiss National Science Foundation (S. A. H.), EPSRC (J. E.-K.), Royal Society (A. B. C.) and University of Warwick. Crystallographic data was collected using a diffractometer purchased through support from Advantage West Midlands and the European Regional Development Fund. We also thank the Scientific Computing Research Technology Platform at the University of Warwick for providing high-performance computing facilities.

Notes and references

1. (a) K. Krogh-Jespersen, M. Czerw, K. Zhu, B. Singh, M. Kanzelberger, N. Darji, P. D. Achord, K. B. Renkema and A. S. Goldman, J. Am. Chem. Soc., 2002, 124, 10797–10809; (b) R. Ghosh, X. Zhang, P. Achord, T. J. Emge, K. Krogh-Jespersen and A. S. Goldman, J. Am. Chem. Soc., 2007, 129, 853–866; (c) W. H. Bernskoetter, P. S. White, K. I. Goldberg and M. Brookhart, J. Am. Chem. Soc., 2009, 131, 8603–8613; (d) J. Choi, A. H. R. MacArthur, M. Brookhart and A. S. Goldman, Chem. Rev., 2011, 111, 1761–1779; (e) J. Choi, D. Y. Wang, S. Kundu, Y. Choliy, T. J. Emge, K. Krogh-Jespersen and A. S. Goldman, Science, 2011, 332, 1545–1548.
2. For representative examples see: (a) M. Lafrance, C. N. Rowley, T. K. Woo and K. Fagnou, J. Am. Chem. Soc., 2006, 128, 8754–8756; (b) H.-Q. Do and O. Daugulis, J. Am. Chem. Soc., 2008, 130, 1128–1129; (c) Y. Nakao, N. Kashihara, K. S. Kanyiva and T. Hiyama, J. Am. Chem. Soc., 2008, 130, 16170–16171; (d) X. C. Cambeiro, T. C. Boorman, P. Lu and I. Larrosa, Angew. Chem., Int. Ed., 2012, 52, 1781–1784; (e) P. Ricci, K. Krämer, X. C. Cambeiro and I. Larrosa, J. Am. Chem. Soc., 2013, 135, 13258–13261; (f) X. C. Cambeiro, N. Ahlsten and I. Larrosa, J. Am. Chem. Soc., 2015, 137, 15636–15639; (g) J. Takaya, S. Ito, H. Nomoto, N. Saito, N. Kirai and N. Iwasawa, Chem. Commun., 2015, 51, 17662–17665; (h) M. Simonetti, G. J. P. Perry, X. C. Cambeiro, F. Juliá-Hernández, J. N. Arokianathar and I. Larrosa, J. Am. Chem. Soc., 2016, 138, 3596–3606.
3. (a) E. Clot, O. Eisenstein, N. Jasim, S. A. Macgregor, J. E. McGrady and R. N. Perutz, Acc. Chem. Res., 2011, 44, 333–348; (b) S. D. Pike, M. R. Crimmin and A. B. Chaplin, Chem. Commun., 2017, 53 10.1039/c6cc09575e.
4. (a) M. E. Evans, C. L. Burke, S. Yaibuathes, E. Clot, O. Eisenstein and W. D. Jones, J. Am. Chem. Soc., 2009, 131, 13464–13473; (b) E. Clot, C. Mégret, O. Eisenstein and R. N. Perutz, J. Am. Chem. Soc., 2009, 131, 7817–7827; (c) E. Clot, M. Besora, F. Maseras, C. Mégret, O. Eisenstein, B. Oelckers and R. N. Perutz, Chem. Commun., 2003, 490–491.
5. (a) L. Fan, S. Parkin and O. V. Ozerov, J. Am. Chem. Soc., 2005, 127, 16772–16773; (b) E. Ben-Ari, M. Gandelman, H. Rozenberg, L. J. W. Shimon and D. Milstein, J. Am. Chem. Soc., 2003, 125, 4714–4715; (c) E. Ben-Ari, R. Cohen, M. Gandelman, L. J. W. Shimon, J. M. L. Martin and D. Milstein, Organometallics, 2006, 25, 3190–3210.
6. D. Y. Wang, Y. Choliy, M. C. Haibach, J. F. Hartwig, K. Krogh-Jespersen and A. S. Goldman, J. Am. Chem. Soc., 2016, 138, 149–163.
7. During the preparation of this manuscript the C–H bond activation of fluorobenzene using related Ir-POCOP pincer complexes has been described: L. P. Press, A. J. Kosanovich, B. J. McCulloch and O. V. Ozerov, J. Am. Chem. Soc., 2016, 138, 9487–9497.
8. (a) K. Krogh-Jespersen, M. Czerw, K. Zhu, B. Singh, M. Kanzelberger, N. Darji, P. D. Achord, K. B. Renkema and A. S. Goldman, J. Am. Chem. Soc., 2002, 124, 10797–10809; (b) D. Morales-Morales, R. Redón, Z. Wang, Do W. Lee, C. Yung, K. Magnuson and C. M. Jensen, Can. J. Chem., 2001, 79, 823–829.
9. M. Kanzelberger, B. Singh, M. Czerw, K. Krogh-Jespersen and A. S. Goldman, J. Am. Chem. Soc., 2000, 122, 11017–11018.
10. Acid catalysed oxidative addition of C(sp)–H bonds to 1 have, however, been reported: J. D. Hackenberg, S. Kundu, T. J. Emge, K. Krogh-Jespersen and A. S. Goldman, J. Am. Chem. Soc., 2014, 136, 8891–8894.
11. Consistent with this suggestion, photolysis of 20 mM solutions of 1 and 2 in C₆D₆ under 1 atm of CO at RT resulted in prolonged (>6 h) establishment of similar photostationary mixtures of 1 : 2-d₆ (1: 1.4, corresponding to 58% 2-d₆) to those obtained under an argon atmosphere (after ca. 2 h). For discussion of mechanistic pathways in related photochemical reactions of [Rh(PMe₃)₂(CO)Cl] see: G. P. Rosini, W. T. Boese and A. S. Goldman, J. Am. Chem. Soc., 1994, 116, 9498–9505.
12. J. E. Wheatley, C. A. Ohlin and A. B. Chaplin, Chem. Commun., 2014, 50, 685–687.
13. Notably using a long pass (glass) filter (<330 nm): no reaction was observed when a 20 mM solution of 1 was irradiated in C₆D₆ for 2 h at RT, while <5% formation of 1 was observed when a 20 mM solution of 2 was irradiated in C₆D₆ for 2 h at RT.
14. C. J. Moulton and B. L. Shaw, J. Chem. Soc., Dalton Trans., 1976, 1020.

---

Footnote

† Electronic supplementary information (ESI) available: Full experimental and computational details, NMR, IR and UV-vis spectra, TD-DFT analysis of 1 and 2, and optimised geometries in xyz format. CCDC 1520522–1520525. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c6cc09807j

---