Competitive gold/nickel transmetalation

Abstract

Transmetalation is a key method for the construction of element–element bonds. Here, we disclose the reactivity of [Ni^II(Ar)(I)(diphosphine)] compounds with arylgold(I) transmetalating agents, which is directly relevant to cross-coupling catalysis. Both aryl-for-iodide and unexpected aryl-for-aryl transmetalation are witnessed. Despite the strong driving force expected for Au–I bond formation, aryl scrambling can occur during transmetalation and may complicate the outcomes of attempted catalytic cross-coupling reactions.

---

The forging of metal–element bonds via transmetalation is a pivotal elementary step in catalytic cross-coupling for access to a variety of functionalized molecules, to the extent that cross-coupling reactions are defined and categorised by their transmetalation event.¹ Frequently, this reaction follows substrate oxidative addition and precedes product reductive elimination.¹ Many such coupling reactions have emerged, relying on electropositive metals for functional group (R) delivery viz. Kumada (R—MgBr), Suzuki (R—BR₂), Negishi, (R—ZnX), Stille (R—SnR₃), and Hiyama (R—SiR₃), to name a few.^1,2 In 2009, Hashmi et al. introduced organogold(I) compounds ([R—Au(L_n)]; L_n = ligand) as competent partners for catalytic cross-coupling using palladium(0).³

Blum and co-workers subsequently reported the room temperature [NiCl₂(PCy₃)₂]-catalysed cross-coupling of [R—Au(PPh₃)] complexes with aryl bromides.⁴ This reaction is proposed to proceed via an open-shell Ni complex wherein [Ni^II(Cl)₂(PCy₃)₂] first undergoes transmetalation with 2 equivalents of [Ar—Au(PPh₃)] to give [Ni^II(Ar)₂(PCy₃)₂] and 2 equivalents of [Cl—Au(PPh₃)], presumably via [Ni^II(Ar)(Cl)(PCy₃)₂]; subsequent halide transfer and oxidation by Au(I) gives [Ni^III(Ar)₂(Cl)(PCy₃)₂] and Au(0). This Ni^III species would thus reductively eliminate Ar—Ar to provide [Ni^I(Cl)(PCy₃)₂] and enter the catalytic cycle, undergoing transmetalation, oxidative addition, and reductive elimination, giving a C—C coupled product. To buttress this proposal, the stoichiometric reaction of [NiCl₂(PCy₃)₂] with 10 equivalents of [(4-(MeO)C₆H₄)—Au(PPh₃)] resulted in the homocoupled biaryl product, and a paramagnetic nickel complex that was characterised by EPR spectroscopy (Chart 1A).⁴

Chart 1 Literature precedent and accessing a [Ni(diphosphine)(Ar)(I)] precursor and its reactivity with organogold(I) transmetalating reagents.

Organogold(I) compounds are competent for transmetalation to (e.g.) [CpFe(CO)₂(X)] (Cp = C₅H₅⁻, X = Cl, Br, I),⁵ [(p-Cym)Ru(X)₂(PPh₃)] (p-Cym = p-cymene, X = Br, I),⁵ and [Cp*Rh(Cl)₂(PPh₃)] (Cp* = C₅Me₅⁻) (Chart 1B).⁶ For the reaction of [CpFe(CO)₂(X)] with [(4-(NO₂)C₆H₄)—Au(PPh₃)], the rate of reaction followed the order I > Br > Cl, consistent with the affinity of gold(I) for soft halides in the [X—Au(PPh₃)] by-product.⁵ We refer interested readers to several reviews pertaining to [Au]—based transmetalations.⁷

We recently reported the fundamental reactivity of [Ni⁰(P₂B^Cy₄)₂] (P₂B^Cy₄ = 1,2-bis(di(3-dicyclohexylboraneyl)propylphosphino)ethane), which undergoes room temperature iodoarene activation to afford [Ni^II(P₂B^Cy₄)(Ar)(I)] and P₂B^Cy₄ (Chart 1C).⁸ This reaction is noteworthy as the treatment of alkyl/aryl-substituted diphosphine complexes of the form [Ni⁰(diphosphine)₂] with haloarenes generally does not result in C–X bond activation, due to the significant endergonicity associated with forming the [Ni⁰(κ¹-diphosphine)(κ²-diphosphine)] complex required for substrate activation;⁹ [Ni⁰(dnppe)₂] (dnppe = 1,2-bis(di-n-propylphosphino)ethane) does not undergo oxidative addition under similar conditions.

To extend the reactivity of such P₂B^Cy₄-ligated complexes, we elected to study the behaviour of [Ni^II(P₂B^Cy₄)(Ar)(I)] with a transmetalating reagent, as a means to explore the second elementary step in a viable cross-coupling sequence using [Ni⁰(P₂B^Cy₄)₂]. Our interests turned to [Au]—based reagents because these reagents are accurately weighed in small quantities, soluble in hydrocarbon solvents, readily tailored, and are unlikely to undergo transfer to the sp²-hybridized boranes present on the P₂B^Cy₄ ligand.⁸ Furthermore, these agents are bench-stable and provide [X—Au(PPh₃)] as a by-product, whose formation can be conveniently monitored by ³¹P NMR spectroscopy.¹⁰ In addition to probing the stepwise reactivity of “[Ni^0/II(P₂B^Cy₄)]”-type compounds, we were equally interested in contributing to the growing area of gold-to-metal transmetalations starting from a bona-fide [Ni^II(diphosphine)(Ar)(X)] complex, informing studies such as that noted for [Ni^II(Cl)₂(PCy₃)₂].⁴

[Ni^II(P₂B^Cy₄)(4-FC₆H₄)(I)] (2-F), generated from oxidative addition of 4-fluoroiodobenzene at [Ni(P₂B^Cy₄)₂] (1), was selected as the transmetalation partner, enabling reaction monitoring by ¹⁹F NMR spectroscopy.⁸ Transmetalation was examined using gold(I) aryl compounds with electron-donating or -withdrawing groups at the para-position, [(4-X-C₆H₄)—Au(PPh₃)] (X = H, OCH₃, CF₃) (Scheme 1). Initially, compound 2-F was combined with 1 equiv. of [(C₆H₅)—Au(PPh₃)], and the reaction was analysed by NMR spectroscopy. Given the literature precedent noted above, we hypothesized that this would represent an ideal pairing, owing to the favourable generation of [I—Au(PPh₃)]. However, monitoring the reaction by ³¹P NMR spectroscopy evidenced formation of [Ni(P₂B^Cy₄)(C₆H₅)(I)]⁸ (2-H) in 25% yield after 7 mins, resulting from ‘aryl-to-aryl’ transmetalation. The fate of the exchanged “4-FC₆H₄” unit was deduced by ¹⁹F NMR spectroscopy, matching data reported for [(4-FC₆H₄)—Au(PPh₃)].¹⁰ In addition to signals for 2-H, a signal at δ_P = 43 ppm was also noted for [(4-FC₆H₄)—Au(PPh₃)] in the ³¹P NMR spectrum. The formation of 4-fluorobiphenyl, from reductive elimination via [Ni(P₂B^Cy₄)(4-FC₆H₄)(C₆H₅)], was observed by ¹⁹F NMR spectroscopy, indicating that both Ar-for-Ar and Ar-for-I transmetalation processes operate. The treatment of 2-F with 1 equiv. of [(4-X-C₆H₄)—Au(PPh₃)] (X = OCH₃, CF₃) gave [Ni(P₂B^Cy₄)(4-X-C₆H₄)(I)] (2-X; X = OCH₃ (17%), CF₃ (38%)) after 7 minutes.

Scheme 1 Transmetalation of 2-F and 4-F using [Ar—Au(PPh₃)] reagents. Yields taken after 7 min. Inset of B shows the scXRD structure of 4-F (50% occupancy, hydrogens omitted for clarity). ** The maximum theoretical yield of 3 is 50%. dnppe = 1,2-bis(di-n-propylphosphino)ethane.

Compounds 2-X were highly reactive under the conditions studied. Up to 50% of [Ni]_total (after ca. 60 mins) in these elementary reactions is [Ni⁰(P₂B^Cy₄)(PPh₃)₂] (3) δ_P = 34.8 (t, ²J_P,P = 27 Hz; PPh₃), 19.7 (t, ²J_P,P = 27 Hz; P₂B^Cy₄), which forms via PPh₃ transfer from Au following C(Ar)—C(Ar) reductive elimination in yields from 14–26%, based on Ni (Scheme 1). The observation of some S = 0 Ni(0) product suggests reductive elimination from Ni(II). The identity of 3 was confirmed by its independent preparation from the reaction of [Ni⁰(P₂B^Cy₄)₂] with 2 equiv. of PPh₃. This reaction illustrates the lability of the P₂B^Cy₄ ligand scaffold; the related species [Ni⁰(dnppe)₂], which is devoid of pendant boranes, does not coordinate PPh₃.

To demonstrate the generality of our observations regarding transmetalation between gold and nickel, and to show that this is not a boron effect, [Ni(dnppe)(4-FC₆H₄)(I)] (4-F) (see ESI†) was prepared, characterized, and exposed to 1 equiv. of [(C₆H₅)—Au(PPh₃)]. Analysis of the reaction mixture after 7 min by NMR spectroscopy showed formation of [Ni(dnppe)(C₆H₅)(I)] (4-H) in 34% yield, consistent with previous results for P₂B^Cy₄.

The relevance of such compounds viz. [Ni(PP₂B^Cy₄)₂] (1) in cross-coupling was also confirmed with 10 mol% 1 enabling the cross-coupling of 4-fluoroiodobenzene and [(4-FC₆H₄)—Au(PPh₃)], delivering the corresponding biaryl in 86% conversion by ¹⁹F NMR spectroscopy.

To probe the thermodynamics associated with transmetalation, density functional theory (DFT)¹¹ and DLPNO-CCSD(T) calculations¹² were carried out on model reactions (Scheme 2). In all cases, there was excellent agreement between these methods. The data from DLPNO-CCSD(T) calculations are discussed here, but DFT-derived energies are reported in Scheme 2 and in the ESI.† The outcomes from I-for-Aryl and Ar-for-Ar transmetalation at 4-F are endergonic and energetically neutral, respectively. Somewhat surprisingly, the exchange of iodine for phenyl is endergonic (ΔG° = +2.6 kcal mol⁻¹). The exchange of para-fluorophenyl for phenyl (i.e. formation of 4-H from 4-F) is essentially energetically neutral (ΔG° = −0.1 kcal mol⁻¹) (similar trends are observed with gold reagents having p-F₃CC₆H₄, p-H₃COC₆H₄, and 4-pyridyl substituents (vide infra) – see ESI†). In the context of catalysis, the product of I-for-Ar transmetalation can undergo reductive elimination, driving catalysis forward if the overall reaction is exergonic; however, these experimental and computational results indicate the potential for aryl scrambling during cross-coupling reactions, eroding reaction selectivity and forming undesired homocoupling products.

Scheme 2 Thermodynamics of transmetalation (kcal mol⁻¹). ΔG°(CC) was calculated using DLPNO-CCSD(T), and ΔG°(DFT) using DFT.^11,12

Previously, we showed that [Ni(P₂B^Cy₄)₂] (1) coordinates Lewis bases, binding up to eight equiv. of 4-N,N-dimethylaminopyridine (DMAP).¹³ We wondered whether an appropriate gold(I)-pyridyl fragment (i.e. [Au]—(4-NC₅H₄)) might instead be used for coordination, with the P₂B^Cy₄ ligand serving in a directing capacity, luring transmetalating agents into close proximity with a Ni(II) fragment. A new compound, [(4-NC₅H₄)—Au(PPh₃)] (5) was prepared via the reaction of [Br—Au(PPh₃)] with 4-pyridylboronic acid (75% yield) (Scheme 3A).¹⁴ The formation of 5 is substantiated by NMR spectroscopy (δ_P = 43 ppm) and single crystal X-ray diffraction(Scheme 3A). Importantly, 5 boasts an accessible Lewis basic pyridine, allowing for possible interaction with the pendant boraneyl groups of [Ni(P₂B^Cy₄)₂] (1).

Scheme 3 A. Synthesis of a 4-pyridylgold(I) reagent 5. Inset shows the scXRD structure of 5 (50% occupancy, hydrogens omitted for clarity). B. Synthesis of octaaurated compound 6.

The reaction of 1 with 8 equiv. of 5 in THF¹⁵ provided an immediate colour change from light yellow to dark brown to give [Ni⁰(P₂B^Cy₄)₂{(4—NC₅H₄)—Au^I(PPh₃)}₈] (6) (Scheme 3B). Analysis by ³¹P NMR spectroscopy (THF-d₈) provided two broad signals (δ_P = 42, 36 ppm) that are shifted upfield compared to uncoordinated 5 (Δδ_P = −1 ppm) and 1 (Δδ_P = −3 ppm) (see ESI†).¹³ Consistent with an array of interacting borane–pyridine groups, the ³¹P NMR signature for the equivalent [Ni]—P fragments is similar to [Ni⁰(P₂B^Cy₄)₂(DMAP)₈] (δ_P = 35 ppm).¹³ No signal is detected by ¹¹B NMR spectroscopy at 298 K cf. δ_B = 84 ppm for free [Ni⁰(P₂B^Cy₄)₂] and δ_B = +4 ppm for [Ni⁰(P₂B^Cy₄)₂(DMAP)₈].¹³ To probe the possibility of fluxional solution behaviour, a variable temperature (VT) NMR study was undertaken. At 193 K, the ³¹P NMR spectrum is markedly different; the broad signal associated with [Au]—P groups decoalesces, while the signal for the [Ni]—P groups moves upfield and the baseline broadens (see ESI†). Fluxional behaviour is also witnessed by VT ¹¹B NMR spectroscopy, which at 263 K shows an averaged signal for the sp³-hybridized boraneyl groups at δ_B = −4 ppm; this sharpens considerably at 193 K. These data demonstrate the proclivity of the pendant boranes of the [Ni⁰(P₂B^Cy₄)₂] framework to support a metal-rich secondary coordination sphere, an attractive approach towards accessing multimetallic complexes that are “docked” via Lewis acid/base interactions.

Previously, we showed that reaction of [Ni⁰(P₂B^Cy₄)₂(DMAP)₈] with PhI resulted in the formation of [Ni^II(P₂B^Cy₄)₂(DMAP)₈(I)]I via iodine atom abstraction.⁸ We probed the reactivity of 6 with 4-fluoroiodobenzene but protection of the Ni(0) site by a “golden” secondary coordination sphere prevents oxidative addition. Nonetheless, switching the order of addition and allowing 1 to fully react with 4-fluoroiodobenzene to give [Ni^II(P₂B^Cy₄)(4-FC₆H₄)(I)] (2-F), followed by addition of 5, provided the pyridyl-linked oligomer, [Ni^II(P₂B^Cy₄)(4-NC₅H₄)(I)]_n⁸ in 24% yield after 7 mins by ³¹P NMR spectroscopy, which results from aryl-to-pyridyl transmetalation (Scheme 4).¹⁶ Under these conditions 2-Pyr was noted to be more robust than 2-X, producing less 3; 4-(4-fluorophenyl)pyridine was not detected by ¹⁹F NMR spectroscopy. Interestingly, the reaction between [Ni^II(dnppe)(4-FC₆H₄)(I)] (3-F) and 5 in THF provides a cloudy reaction mixture and [Ni^II(dnppe)(4-Pyr)(I)] is not observed.

Scheme 4 Transmetalation using a Au-based 4-pyridyl reagent. * = Yield after 60 min (the maximum theoretical yield of 3 is 50%).

In sum, we have confirmed that isolated [Ni^II(diphosphine)(Ar)(I)] compounds undergo aryl-for-aryl in addition to aryl-for-iodide transmetalation with arylgold(I) reagents. Furthermore, we have exploited the boron-rich secondary coordination sphere of [Ni(P₂B^Cy₄)₂] to host a metal (gold)-rich secondary coordination sphere via pyridine–borane dative interactions, a starting point toward achieving metal–metal cooperativity using such boron-rich ligand scaffolds in an elementary cross-coupling transformation. This study expands our knowledge of Au-to-metal transmetalation and provides insights into the initial steps associated with metal-to-metal functional group transfer relevant to carbon–carbon cross-coupling.

The authors are grateful to the University of Windsor, the Council of Ontario Universities, Compute Canada, and the Natural Sciences and Engineering Research Council of Canada (Discovery Grant, RGPIN-2020-04480, Discovery Launch Supplement, DGECR-2020-00183, and a graduate award (CGS-M) to J. A. Z.) for funding. This work was also made possible by the facilities of the Shared Hierarchical Academic Research Computing Network (SHARCNET: www.sharcnet.ca) and Compute/Calcul Canada. D. J. N. thanks the Carnegie Trust for the Universities of Scotland for a Research Incentive Grant (RIG008165). Some of the calculations were performed using the Archie-WEST High Performance Computer (http://www.archie-west.ac.uk) at the University of Strathclyde; we thank Mr J. Buzzard, Dr K. Kubiak-Ossowska, and Dr R. Martin for their assistance with this facility.

Conflicts of interest

There are no conflicts to declare.

Notes and references

1. J. F. Hartwig, Organotransition Metal Chemistry From Bonding to Catalysis. Chapter 7, University Science Books, Mill Valley, California. 2010.
2. (a) R. Jana, T. P. Pathak and M. S. Sigman, Chem. Rev., 2011, 111, 1417–1492; (b) N. Hazari, P. R. Melvin and M. M. Beromi, Nat. Rev. Chem., 2017, 1, 0025; (c) K. C. Nicolaou, P. G. Bulger and D. Sarlah, Angew. Chem., Int. Ed., 2005, 44, 4442–4489.
3. (a) A. S. K. Hashmi, C. Lothschütz, R. Döpp, M. Rudolph, T. D. Ramamurthi and F. Rominger, Angew. Chem., Int. Ed., 2009, 48, 8243–8246; (b) M. Hansmann, M. Pernpointner, R. Dopp and A. S. K. Hashmi, Chem. – Eur. J., 2013, 19, 15290–15303; (c) For a related study, see: S. Witzel, M. Hoffmann, M. Rudolph, F. Rominger, A. Dreuw and A. S. K. Hashmi, Adv. Synth. Catal., 2021 DOI:10.1002/adsc.202101113.
4. J. J. Hirner and S. A. Blum, Organometallics, 2011, 30, 1299–1302.
5. A. S. K. Hashmi and L. Molinari, Organometallics, 2011, 30, 3457–3460.
6. Y. Shi and S. A. Blum, Organometallics, 2011, 30, 1776–1779.
7. (a) J. J. Hirner, Y. Shi and S. A. Blum, Acc. Chem. Res., 2011, 44, 603–613; (b) L.-P. Piu and G. B. Hammond, Chem. Soc. Rev., 2012, 41, 3129–3139.
8. J. A. Zurakowski, B. J. H. Austen, M. C. Dufour, D. M. Spasyuk, D. J. Nelson and M. W. Drover, Chem. – Eur. J., 2021, 27(64), 16021–16027.
9. (a) C. Amatore, G. Broeker, A. Jutand and F. Khalil, J. Am. Chem. Soc., 1997, 119, 5176–5185; (b) P. Fitton and E. A. Rick, J. Organomet. Chem., 1971, 28, 287–291; (c) A. L. Clevenger, R. M. Stolley, N. D. Staudaher, N. Al, A. L. Rheingold, R. T. Vanderlinden and J. Louie, Organometallics, 2018, 37, 3259–3268; (d) G. Yin, I. Kalvet, U. Englert and F. Schoenebeck, J. Am. Chem. Soc., 2015, 137, 4164–4172.
10. C. Croix, A. Longeau-Balland, H. Allouchi, M. Giorgi, A. Duchêne and J. Thibonnet, J. Organomet. Chem., 2005, 690, 4835–4843.
11. DFT calculations were carried out using Gaussian16 Rev. C.01 (Gaussian 16, Revision C.01, M. J. Frisch, et al., Gaussian, Inc., Wallingford CT, 2016) at the M06/6-311 + G(d,p) + LANL2DZ(d,p) + LANL2DZ(f) + SMD(benzene)//B3LYP-D3/6-31G(d) + LANL2DZ(d,p) + LANL2TZ(f) level of theory. DLPNO-CCD(T) single points were obtained using ORCA 5.0.0 on DFT-optimised geometries using a cc-pVTZ-PP basis set and SK-MCDHF-RSC ECP on Au and I, and a cc-pVTZ basis set on all other atoms. For more details, see the ESI.
12. (a) C. Riplinger, P. Pinski, U. Becker, E. F. Valeev and F. Neese, J. Chem. Phys., 2016, 144, 024109; (b) C. Riplinger, B. Sandhoefer, A. Hansen and F. Neese, J. Chem. Phys., 2013, 139, 134101.
13. M. W. Drover, M. C. Dufour, L. A. Lesperance-Nantau, R. P. Noriega, K. Levin and R. W. Schurko, Chem. – Eur. J., 2020, 26, 11180–11186.
14. For examples of Au-arene synthesis, see: (a) N. Ahlsten, G. J. P. Perry, X. C. Cambeiro, T. C. Boorman and I. Larrosa, Catal. Sci. Technol., 2013, 3, 2892–2897; (b) N. V. Tzouras, M. Saab, W. Janssens, T. Cauwenbergh, K. Van Hecke, F. Nahra and S. P. Nolan, Chem. – Eur. J., 2020, 26, 5541–5551; (c) F. J. L. Ingner, Z. X. Giustra, S. Novosedlik, A. Orthaber, P. J. Gates, C. Dyrager and L. T. Pilarski, Green Chem., 2020, 22, 5648–5655; (d) D. V. Partyka, M. Zeller, A. D. Hunter and T. G. Gray, Inorg. Chem., 2012, 51, 8394–8401.
15. This reaction mixture precipitated from benzene or toluene, and required THF as solvent. Notably, both components (1 and 5) are soluble in benzene or toluene, indicating persistent adduct formation.
16. In addition, reaction of [Ni^II(P₂B^Cy₄)(4-CH₃OC₆H₄)(I)] (2-OCH₃) with 5 provided the pyridyl-linked oligomer, [Ni^II(P₂B^Cy₄)(4-NC₅H₄)(I)]_n in 48% yield.

---

Footnotes

† Electronic supplementary information (ESI) available: ¹H, ¹³C{¹H}, ³¹P{¹H}, and ¹¹B NMR spectra for all complexes. XYZ coordinates for DFT calculations. CCDC 2109283 and 2109284. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/d1cc06064c

‡ These authors contributed equally.

---