Open Access Article
Rana
Marghalani
and
Eric S.
Cueny
*
Department of Chemistry, Boston University, 590 Commonwealth Ave., Boston, MA 02215, USA. E-mail: ecueny@bu.edu
First published on 27th July 2024
In this contribution, we prepare the dinuclear complex [(CNCF)(PPh3)Pt–Au(PPh3)]+ (2-F) supported by an electron deficient derivative of 2,6-diphenylpyridine (CNC), 2,6-di(4-fluorophenyl)pyridine (CNCF). Solution state spectroscopic data and solid-state structural data reveals formation of the desired dinuclear complex occurs and that it remains intact in solution. The solid state structure of 2-F, compared to [(CNC)(PPh3)Pt–Au(PPh3)]+ (2), reveals a substantial change in the C–Au–P bond angle. We postulated that this change in bond angle arises due to a weaker interaction between [(PPh3)Au]+ and (CNCF)Pt(PPh3) (1-F) vs. (CNC)Pt(PPh3) (1). Through pyridine titration experiments, we demonstrate that the interaction is indeed weaker between [(PPh3)Au]+ and 1-Fvs. 1. Cyclic voltammetry (CV) experiments confirm that 1-F is less electron rich than 1. DFT calculations demonstrate that the HOMO of 1 and 1-F is not dz2, helping explain the differences in electrochemical behavior of 1 and 1-F and bonding between 1 and 1-F with [(PPh3)Au]+.
By contrast, comparatively little is understood about the mechanism and thermodynamic driving force for transmetallation between two transition metal complexes,7 though it is known in the Sonogashira reaction8 and has been proposed in other cooperative catalytic reactions.9–13 Stahl and coworkers convincingly demonstrate transmetallation of an aryl group between two palladium species upon C–H activation via kinetic isotope effect (KIE) studies.14 Other research groups have explored stoichiometric transmetallation between two organometallic complexes.15–20 However, one of the challenges in developing a deeper understanding of transmetallation is the stability of the proposed intermediate along the pathway of transmetallation. Often, the proposed intermediate in transmetallation has a low kinetic barrier to formation of the final exchange product making it challenging to detect and directly probe its structure.20 Despite this challenge, several research groups have examined dinuclear complexes that may play a role in understanding transmetallation reactions.21–36
Martin and coworkers used the CNC ligand (CNC = 2,6-diphenylpyridine) bound to Pt (1) to study transmetallation intermediates.37 One such intermediate is generated by reaction of 1 with [(PPh3)Au]+ to yield 2 where the Au binds to the Pt and the ipso carbon of the phenyl ring. While metallophilic interactions between Pt and Au likely play a role in the stability of this complex, the simultaneous binding to the Pt–phenyl ring suggests that the electronics of the arene ring also play a role in stabilizing such a dinuclear complex.
We chose to study a fluorinated derivative of the CNC ligand (CNCF) to establish the role of electron withdrawing groups in the stability of dinuclear Pt/Au complexes. For efficient cooperative catalysis with transmetallation as the key step, a delicate interplay of kinetic barriers between and thermodynamic stabilities of reaction intermediates is at work (Fig. 1). If the intermediate of transmetallation (e.g.2 or 2-F) is highly stable, this will decrease the concentration of active catalyst in a cooperative catalytic reaction. An analogy can be drawn to the use of chain transfer reagents in alkene polymerization where highly stable bimetallic complexes between the polymerization catalyst and chain transfer reagent inhibit propagation as intermediate formation decreases the concentration of the propagating catalyst species.38–44 Thus, understanding the strength of such dinuclear complexes as a function of metal identity (as Martin and coworkers have previously described)31,37,45,46 as well as electronics of the bridging arene ligand (reported here) are vital to the development of cooperative catalytic reactions.
Herein, we describe the synthesis of 2-F and compare its structure and stability to the known complex 2. First, we describe the differences between the solid-state structure of 2 and 2-F. Then, we use pyridine titration experiments to probe the thermodynamic stability of these two complexes. We also use cyclic voltammetry (CV) to determine the anodic peak potential of both (CNCR)Pt(PPh3), when R = H and F. We correlate the oxidation potential determined by CV experiments to the thermodynamic stability of the dinuclear [(CNCR)(PPh3)Pt–Au(PPh3)]+ complexes.
First, we set out to synthesize 2-F by reaction of the known 1-F complex with the in situ generated [(PPh3)Au]+ (Scheme 1).47 We observed two peaks in the 31P NMR spectrum of this new complex at 33.8 ppm with Pt satellites (2JP–Pt = 267 Hz) and 21.5 ppm with Pt satellites (1JP–Pt = 3568 Hz) corresponding to the PPh3 ligands bound to Au and Pt respectively. Similar to complex 2, the 2JP–Pt of the PPh3 ligated to Au supports that a dinuclear complex remains intact in solution as a two bond coupling between Pt and the P of PPh3 bound to Au is observed. The 1H NMR spectrum of 2-F reveals an expected upfield shift of the ortho C–H (relative to Pt) of 0.18 ppm upon coordination of the cationic Au species. A similar shift upfield was observed for 2 of 0.21 ppm upon coordination of [(PPh3)Au]+.37
The solid-state structure of 2-F (Fig. 2), obtained by single crystal X-ray diffraction, revealed differences between 1-F and 1 upon binding the [(PPh3)Au]+.37 Most prominently, the bond angles between the P–Au–C for the two complexes are substantially different; for 2-F, P–Au–C = 173.5°, whereas for 2, P–Au–C = 153.2°. Interestingly, despite this change in the P–Au–C bond angle, the degree of arene group transfer (as indicated by the Cpara–Cipso–Au angle) does not change substantially between 2 and 2-F. For 2-F, Cpara–Cipso–Au = 120.9° whereas for 2 Cpara–Cipso–Au = 120.1°. If the arene fully transferred to Au, the Cpara–Cipso–Au would reach ∼180°, with no arene Au interaction Cpara–Cipso–Au would be ∼90°.46 Other bond metrics change minimally between the proteo and fluoro substituted arenes including the Pt–Au distances, 2.7430(5) Å for 2-F and 2.7222(2) Å for 2. The Pd derivative of 2 (2-Pd) has been previously characterized by single crystal X-ray diffraction.46
![]() | ||
| Fig. 2 (Top) Molecular structure of the cation portion of complex 2-F. Hydrogens, DCM solvent, and triflate anion omitted for clarity. (Bottom) Molecular structure of the cation portion of complex 2.37 Hydrogens and perchlorate anion omitted for clarity. For both complexes, thermal ellipsoids are shown at the 50% probability level. | ||
Interestingly, 2-Pd shows some structural features related to both 2 and 2-F; for example, Pd–Au = 2.7422(3) Å, similar to the Pt–Au distances in 2 and 2-F. In 2-Pd, the P–Au–C = 173.8°, which is quite similar to that of 2-F (P–Au–C = 173.5°). However, the degree of arene transfer in 2-Pd is much higher than 2 and 2-F as indicated by Cpara–Cipso–Au = 143.1° vs. 120.1° and 120.9° for 2 and 2-F respectively. The higher degree of arene transfer in 2-Pd is reflective of generally stronger Pt–C bonds relative to Pd–C bonds. Clearly, there are multiple structural factors to consider when trying to understand the bonding interactions between such dinuclear, frustrated transmetallation complexes.
Previously, Martin and coworkers have used Energy Decomposition Analysis (EDA) to understand the strength of the interactions in dinuclear complexes such as 2, 2-Pd, and other related complexes.37,45 Herein, we chose to use experimental data to qualitatively determine the relative stabilities of 2vs. 2-F. We postulated that the C–Au–P bond angle difference between 2 and 2-F (with similar degrees of arene group transfer) may arise from an overall decrease in stability of the interaction between the [(PPh3)Au]+ and 1-F relative to 1. We chose to evaluate the thermodynamic strength of these interactions using pyridine titrations. We dissolved [(CNCR)(PPh3)Pt–Au(PPh3)]+ in dcm–d2 and added various equivalents of pyridine or 2-fluoropyridine. In these titration experiments, an equilibrium may be established between the dinuclear complex plus pyridine and the monometallic Pt complex and a pyridine ligated [(PPh3)Au]+ (Fig. 3).
![]() | ||
| Fig. 3 (Top) Titrations of 2 (R = H) and 2-F (R = F) with pyridine (R1 = H) and 2-fluoropyridine (R1 = F). Titrations were monitored by 1H NMR spectroscopy (see ESI† for details). We used the peak associated with the proton alpha to Pt in 2 & 2-F and monitored the change in ppm based on the equivalents of pyridine or 2-fluoropyridine added. We normalized the ppm change upon pyridine (or 2-fluoropyridine addition) plotting the normalized Δppm on the y-axis. Here, the value of zero on the y-axis corresponds to no change in the ppm shift (i.e. the chemical shift of the starting 2 or 2-F). The value of one on the y-axis corresponds to complete dissociation of [(PPh3)Au]+ from the Pt (i.e. the chemical shift of 1 or 1-F). (Bottom left) Titration of 2 or 2-F with pyridine. (Bottom right) Titration of 2 or 2-F with 2-fluoropyridine. | ||
Based upon a qualitative comparison of these two complexes, it is clear that [(PPh3)Au]+ binds less strongly to the more electron deficient 1-F than 1. While one equivalent of pyridine completely displaces [(PPh3)Au]+ from 2-F, nearly 10 equivalents of pyridine are required to completely displace [(PPh3)Au]+ from 2 based upon 1H NMR spectroscopic analysis. Using 2-fluoropyridine as the titrant, 2 is only minimally affected; however, an equilibrium is established where [(PPh3)Au]+ is partially displaced from 2-F by 2-fluoropyridine. It is worth noting that 2-fluoropyridine is a weaker base than pyridine by several orders of magnitude revealing how weak of an interaction exists between [(PPh3)Au]+ and 1-F. Importantly, control reactions (see ESI† for details) demonstrate that pyridine and 2-fluoropyridine do not react with 1 or 1-F and appear to bind readily to [(PPh3)Au]+. Based on the ppm shift change upon addition of pyridine and 2-fluoropyridine, we can calculate the Keq for these titration experiments (see ESI† for details).48,49 The affinity of pyridine and 2-fluoropyridine for the [(PPh3)Au]+ fragment of 2-F is higher than for 2 by two orders of magnitude (Fig. 3).
Both titration results were initially surprising. We anticipated that pyridine would more easily displace [(PPh3)Au]+ from both 2 and 2-F. The stability of 2, however, suggests that the Pt–C bond of 1 serves as a reasonably good ligand for the cationic gold species as it competes with pyridine for ligation with [(PPh3)Au]+. While 2 has been studied computationally and through variable temperature NMR spectroscopy,37 a more direct experimental measurement of the stability of 2 has not been reported previously. Based on the stability of 2, it was surprising that 2-F was readily displaced by pyridine and equilibrium established with 2-fluoropyridine (a substantially weaker base than pyridine) as the only difference between these complexes is the meta (relative to Pt) fluoro groups of 2-F.
Based on the electron withdrawing nature of fluorine, we expect that the Pt–C bond of 1-F is likely less electron rich than 1, leading to a dinculear complex 2-F with lower overall stability than 2. Using CV, we established that 1-F is less electron rich than 1 as judged by the relative oxidation potentials of these two complexes. The onset of oxidation of 1-F occurs ∼200 mV more anodic relative to 1. However, neither 1 nor 1-F exhibit fully reversible electrochemical behavior as determined by scan rate dependent CV studies precluding a more detailed analysis. These electrochemical results support our hypothesis that the electron deficient nature of the CNCF ligand plays a substantial role in the decreased stability of the dinuclear complex 2-Fvs. 2.
Given that both 1 and 1-F are d8 complexes and nearly square planar in geometry, one might expect that the HOMO is the dz2 orbital. The dz2 should be minimally impacted by the electronics of the arene rings. The results from CV studies suggest, however, that the HOMO is impacted by the electronics of the arene ring. The expectation that the dz2 orbital is the HOMO arises from a sigma only view of bonding in transition metal complexes, whereas the extended π structures of CNC and CNCF may alter the HOMO of complexes 1 and 1-F. DFT calculations (see ESI† for details) performed at various levels of theory (B3LYP, PBE, PBE0, and M06, see ESI† for details) reveal that the HOMO is a dπ* orbital for both 1 and 1-F with significant contributions from the ligand π orbitals, which likely explains why the oxidation potential of 1 and 1-F are different by ∼200 mV.
Complex 2-F and its comparison to 2, help shed light on the proposed intermediates involved in transmetallation. Martin and coworkers have established that the metal identity in these dinuclear complexes plays a major role in their structure and stability. Here, we show that the electronics of the arene ring involved in this arrested transmetallation state impacts the stability of the dinuclear complex but not the degree of arene group transfer. As the development of cooperative catalytic reactions continues where transmetallation is the key step, it is important to not only consider the metals involved in transmetallation but also the steric and electronic parameters of the ligands that undergo exchange during transmetallation.
Footnote |
| † Electronic supplementary information (ESI) available. CCDC 2364269. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d4dt01828a |
| This journal is © The Royal Society of Chemistry 2024 |