Volume 2021, Issue 35 p. 3528-3539
Full Paper
Open Access

Ligand Effects in Carbon−Boron Coupling Processes Mediated by σ-BH Platinum Complexes

Dr. Pablo Ríos

Dr. Pablo Ríos

Instituto de Investigaciones Químicas (IIQ), Departamento de Química Inorgánica, CSIC and, Universidad de Sevilla, Centro de Innovación en Química Avanzada (ORFEO-CINQA), C/Américo Vespucio 49, 41092 Sevilla, Spain

Search for more papers by this author
Dr. Francisco José Fernández-de-Córdova

Dr. Francisco José Fernández-de-Córdova

Instituto de Investigaciones Químicas (IIQ), Departamento de Química Inorgánica, CSIC and, Universidad de Sevilla, Centro de Innovación en Química Avanzada (ORFEO-CINQA), C/Américo Vespucio 49, 41092 Sevilla, Spain

Search for more papers by this author
Dr. Javier Borge

Dr. Javier Borge

Departamento de Quimica Física y Analítica, Centro de Innovación en Química Avanzada (ORFEO-CINQA), Facultad de Química, Universidad de Oviedo, C/Julián Clavería 8, 33006 Oviedo, Spain

Search for more papers by this author
Dr. Natalia Curado

Dr. Natalia Curado

Instituto de Investigaciones Químicas (IIQ), Departamento de Química Inorgánica, CSIC and, Universidad de Sevilla, Centro de Innovación en Química Avanzada (ORFEO-CINQA), C/Américo Vespucio 49, 41092 Sevilla, Spain

Search for more papers by this author
Prof. Agustí Lledós

Corresponding Author

Prof. Agustí Lledós

Departament de Química, Centro de Innovación en Química Avanzada (ORFEO-CINQA), Universitat Autònoma de Barcelona, Edifici Cn, 08193 Cerdanyola del Vallés, Spain

Search for more papers by this author
Dr. Salvador Conejero

Corresponding Author

Dr. Salvador Conejero

Instituto de Investigaciones Químicas (IIQ), Departamento de Química Inorgánica, CSIC and, Universidad de Sevilla, Centro de Innovación en Química Avanzada (ORFEO-CINQA), C/Américo Vespucio 49, 41092 Sevilla, Spain

Search for more papers by this author
First published: 21 June 2021
Citations: 4

Graphical Abstract

Cyclometallated low-electron count Pt(II) species stabilized by N-heterocyclic carbene ligands (NHCs) react with tri-coordinated boranes to form σ-BH complexes, some of which exhibit a η1-coordination mode. These compounds can undergo competitive C−B/C−H coupling processes that depending on the nature of the NHC can be reversible or not. The effects of the substituents on the borane have been also investigated.

Abstract

The reaction of tri-coordinated boranes (derived from dioxaborolanes and diazaborolanes) with cyclometalated low-electron count platinum complexes [Pt(NHC’)(NHC)][BArF] (NHC=ItBuiPr, IMes, IMes*) led, at low temperature, to the formation of the corresponding σ-BH species. Some of these species have been characterized by X-Ray diffraction methods showing a rare η1-coordination mode. These compounds are thermally unstable and undergo a carbon-boron coupling process whose reversibility depends on the NHC ligand. DFT calculations indicate that the energy barriers required for C−B bond formation events (together with Pt−H bonds) are lower than the competitive reactions leading to C−H bond formation (and Pt−B bonds). However, the C−B coupling products appear to be formed under kinetic control with ItBuiPr ligands, whereas the relative low energy barrier leading to C−H bond formation is sufficiently low to form the thermodynamically more stable platinum boryl complexes at rt. The latter energy barrier is, nevertheless, too high for the systems bearing IMes and IMes* ligands.

Introduction

The interaction of tri-coordinated boranes, HBR2, with transition metal complexes to form σ-BH species is a key step to induce the cleavage of the B−H bond on the way to form new B−X bonds (X=B, C, N).1 σ-borane complexes are a rather well established family of compounds,1k, 2 although considerably more limited in number in comparison with the parent dihydrogen and σ-silane derivatives.3 In most of the cases, the isolated σ-BH species are only involved in processes of cleavage of the B−H bond and/or exchange of the H atom with hydrides in metal compounds.1k, 2, 4 However, this type of compounds can be engaged as intermediates in the formation of carbon-boron bonds, in particular in hydroboration processes,1k, 5 in the borylation of alkanes as described by Hartwig et al.1j, 6 or in other processes involving at some point the reaction of a borane and an alkyl complex.7 In this sense, it is of particular relevance to understand the factors determining the formation of a carbon-boron bond when a metal-alkyl complex and a σ-BH borane are part of the same molecule. In this chemical situation, two possible scenarios can be foreseen: either the molecule evolves towards the formation of a metal-boryl (M−B) complex and an alkane (C−H) or to a metal-hydride (M−H) with concomitant formation of a carbon-boron bond (C−B) as exemplified in Scheme 1 for a σ-CAM (Complex Assisted Metathesis)8 mechanism. The factors that direct the reaction towards one of those specific pathways are, therefore, particularly relevant when designing a catalytic process for the borylation of metal-alkyl bonds. Nevertheless, there is little information about these competitive reaction processes, due, in part, to the inherent difficulty in isolating (or detecting) σ-BH complexes bearing alkyl fragments in their structure.1h, 9 Very recently, we have been able to isolate and characterize some rare platinum examples of this kind (Scheme 2, complexes 2.1 ac).10 Intriguingly, these compounds were unexpectedly involved in a reversible carbon-boron bond formation/cleavage process in which the formation of the carbon−boron bond (and a platinum hydride bond) was kinetically preferred whereas generation of a carbon−hydrogen bond (together with a platinum−boron bond) was thermodynamically favored, according to DFT calculations and low temperature NMR experiments. Thus, we set out to explore how the nature of the borane and NHC substituents influence the C−H/C−B bond forming reactions.

Details are in the caption following the image

Possible evolution pathways in the reaction of metal-alkyl complexes with boranes through a σ-CAM type mechanism.

Details are in the caption following the image

Reversible C−B/C−H process observed for complexes 2.1 ac.10

In this article we describe the reactivity of boranes with different sterereo-electronic properties (HBpin, HBcat, and substituted benzodiazaborolanes, with benzodiazaborolanes being more sterically demanding and having a less acidic boron atom and HBcat being the less sterically hindered and more acidic) with low-electron count cyclometalated Pt(II) complexes bearing bulky N-heterocyclic carbene ligands IMes, IMes* or less sterically congested ItBuiPr ligands.

Results and Discussion

Reactions of complex 1.1 with diaminoboranes

In our previous contribution we explored the reactivity of complex [Pt(ItBuiPr’)(ItBuiPr)[BArF] (1.1)11 toward HBpin (pinacolborane), HBcat (catecholborane) and HBdab (1,3,2-benzodiazaborolane).10 The reaction proceeded with the initial formation of the corresponding σ-BH complexes that are stable at temperatures ranging from −15 to +10 °C, being the diamino borane HBdab derivative (2.1 c) the most stable (Scheme 2). Similarly, the reaction of HBMedab (1-methyl-1,3,2-benzodiazaborolane) or HBMe2dab (1,3-dimethyl-1,3,2-benzodiazaborolane) with complex [Pt(ItBuiPr’)(ItBuiPr)[BArF] at low temperature leads to the initial formation of the corresponding σ-BH complexes 2.1 d, e (Scheme 3). These compounds have been characterized spectroscopically by low temperature NMR. The most significant feature

Details are in the caption following the image

Reaction of complex 1.1 with diamino boranes HBMedab and HBMe2dab at low temperature.

in the 1H NMR is a broad signal of the bridging hydride (Pt−H−B) that resonates at −2.30 (2.1 d) and −2.11 ppm (2.1 e) with coupling constants to 195Pt of 328.3 (2.1 d) and 320.0 Hz (2.1 e), comparable to those observed for complexes 2.1 ac, and clearly distinct to those observed for derivatives 3.1 ac (ca. 2200–2500 Hz).10 These signals sharpen upon 11B decoupling (See ESI Figure S12 and Figure S17). The 11B{1H} NMR spectra for 2.1 d, e show a very broad signal at 19.5 and 20.3 ppm, respectively. Interestingly, some of the 1H NMR signals for complex 2.1 d appear rather broad at −20 °C, particularly that of the N–CH3 group. Lowering the temperature to −40 °C indicated the presence of two species in a 3 : 1 ratio with similar chemical shifts for most of the protons, but clearly distinguishable by the bridging hydride, that shows two broad resonances at −2.00 and −2.40 ppm with 1JPt,H of ca. 290 and 305 Hz, respectively. In addition, two different signals are observable for the NH, NCH3 and the methinic protons of the iso-propyl fragments of the NHC in the same 3 : 1 ratio (see Figure S13 in the SI). All these signals merge into a single set of resonances (for each fragment) at about −20 °C. Further lowering of the temperature to −60 °C does not change significantly the ratio of the two species. According to theoretical calculations these two species are rotamers adopting different orientations of the HBMedab ligand, with similar energies (Figure S64). Complexes 2.1 d and 2.1 e are rather thermally stable and could be characterized by X-Ray diffraction studies of crystals grown at −20 °C. The X-ray representation of these two compounds is shown in Figure 1.

Details are in the caption following the image

ORTEP-type representation of complexes 2.1 d and 2.1 e (ellipsoids at 30 % probability). All hydrogen atoms except that for the hydride ligand have been omitted for clarity.

The structural parameters of complexes 2.1 d, e show some deviations from those of the parent complex 2.1 c. The angles defined by the C−Pt−H atoms are 171.1(9)° and 162.6(8)°, whereas those of the Pt−H−B fragment are 98(2)° (2.1 d) and 107(1)° (2.1 e). The Pt⋅⋅⋅B distances are 2.243(2) Å (2.1 d) and 2.272(3) Å (2.1 e), above of the sum of covalent radii for Pt and B (2.2 Å) and longer than those observed for Pt-boryl complexes.10, 12 The B−H bond distances of 1.22(3) Å (2.1 d) and 1.21(3) Å (2.1 e) are only marginally different to that in complex 2.1 c (1.19(3) Å) and are slightly elongated with respect to those in free boranes.13 These structural parameters indicate that the replacement of the NH groups by NCH3 fragments has not an important steric effect in the interaction of the borane with the Pt center and that, likely, the energies for bending the Pt−H−B angle leading to either an η1 or η2 type interaction are very low, as previously described.10 In order to have a better picture of the bonding situation for complexes 2.1 d, e a topological analysis of the electron density and its laplacian was carried out (see Figures S65–S67 and Figure 2). No bond critical point (BCP) between Pt and B was observed in any of the complexes, which suggests a very weak interaction between these two atoms (if any) and a η1 coordination mode.14 It is worth mentioning that in 2.1 d two different orientations of the diamino borane were considered, based on the unsymmetrical character of the molecule and in line with the aforementioned low-temperature NMR experiments.

Details are in the caption following the image

Plot of the Laplacian of the electronic density (∇2ρ) of one of the rotamers of complex 2.1 d.

Complexes 2.1 d, e are stable at temperatures below 15–20 °C. At higher temperatures a rearrangement takes place leading to the platinum hydride derivatives [Pt(H)(ItBuiPr-BX2)(ItBuiPr)][BArF] (BX2=BMedab, 3.1 d; BMe2dab, 3.1 e) after a carbon-boron bond coupling event (Scheme 4). Unfortunately, at the same temperature these hydride species evolve to the platinum boryl complexes [Pt(BX2)(ItBuiPr)2][BArF] (4.1 d, e), hampering the full characterization of these intermediates. Nevertheless, the presence of a hydride signal at δ −21.58 (1JPt,H=1883.2 Hz; 3.1 d) and δ −22.76 (1JPt,H=1983.9 Hz; 3.1 e) as well as the observation of broad resonances at ca. 30 ppm in the 11B NMR spectra are signatures in line with our previous results using HBpin, HBcat and HBdab.10 These species evolve slowly at room temperature over a period of 72 h leading to the platinum boryl species [Pt(BX2)(ItBuiPr)2][BArF] after a C−B bond cleavage reaction. The 1H NMR spectra for these new compounds show a highly symmetric environment consistent with their formulation, whereas the 11B NMR spectra show signals at 11.7 and 13.6 ppm for 4.1 d and 4.1 e,15 respectively, exhibiting a large coupling to 195Pt (∼1400 Hz), as expected for a direct interaction between the two atoms. Crystals suitable for X-ray diffraction studies were grown by slow diffusion of a concentrated solution of 4.1 e into pentane (Scheme 4 bottom right shows the cationic fragment of this compound). The complex exhibits a characteristic T-shaped structure with a formally vacant site16 in trans to the boryl ligand. The Pt−B bond distance of 2.000(4) Å lies at the lower end region for related systems.10, 12d, 15b

Details are in the caption following the image

Evolution of the σ-BH complexes 2.1 d,e and ORTEP-type representation of the cationic fragment of complex 4.1 e (hydrogen atoms have been omitted for clarity).

As inferred from previous theoretical calculations, the platinum boryl species are thermodynamically more stable than the platinum hydrides 3.1 d, e, but formation of the latter is kinetically preferred. The reactivity observed with boranes HBMedab and HBMe2dab does not differ significantly from that of HBdab, except that formation of the boryl complexes 4.1 d, e requires longer reaction times (72 h vs 15 h for HBdab), likely due as a consequence of the steric constrains induced by the methyl groups on the borane fragments.

Reaction of complexes 1.2 and 1.3 with HBpin and HBcat

We then turned our attention to a platinum system bearing NHCs with different steric properties. To this aim we focused on complexes [Pt(IMes’)(IMes)][BArF] (1.2) and [Pt(IMes*’)(IMes*)][BArF] (1.3)17 that have been also useful in the determination of the mechanism in the formation of carbon-silicon bonds by their reaction with silanes.18 The reaction of these complexes with HBpin at rt led to a fast process (less than 15 min) that forms exclusively the platinum hydrides 3.2 a and 3.3 a arising from a carbon-boron bond coupling reaction (Scheme 5). Low temperature NMR experiments carried out at −60 °C of a mixture of complex 1.2 and HBpin allowed us to identify the σ-BH complex [Pt(HBpin)(IMes’)(IMes)][BArF], 2.2 a (Scheme 5). A rather complex 1H NMR spectrum is observed for this complex at −60 °C (see ESI for details, Figure S46–48), but it clearly shows a broad resonance at −5.22 ppm with satellites arising from coupling to 195Pt (JH,Pt=355.0 Hz). These values are similar to those observed for complexes 2.1 ae. The 11B{1H} NMR spectrum for 2.2 a only shows a broad signal for the excess of HBpin (ca 28 ppm) used in the experiment and the boron atom of the BArF anion (−6.7 ppm). Notably, the C−B coupling process leading to 3.3 a starts to take place even at low temperatures (−25 °C).

Details are in the caption following the image

Reactivity of complexes 1.2 and 1.3 toward HBpin and Lewis bases. Bottom: ORTEP-type view of the cationic fragment of complex 3.2 a⋅H2O (most of the carbon and hydrogen atoms of the molecule have been omitted for clarity).

The 1H NMR of 3.2 a and 3.2 b reveals the highly unsymmetrical environment, but most particularly the presence of a broad resonance in the hydride region (−24.30, 3.2 a; −25.10 ppm, 3.3 a) exhibiting a coupling to 195Pt of about 2200 Hz, together with a broad resonance in the 11B NMR spectra at about 32 ppm are compatible with the proposed structure.19 Complexes 3.2 a and 3.3 a are very sensitive towards trace amounts of adventitious water (present in the solvents used for purification) precluding their isolation in their pure form.20 The water molecule binds the metal center leading to the corresponding 16-electron water adducts 3.2 a⋅H2O and 3.3 a⋅H2O. If pyridine is used as nucleophile, the corresponding adducts 3.2 a⋅py and 3.3 a⋅py can be isolated and fully characterized (see ESI for details). The structure of the water adduct 3.2 a⋅H2O was ascertained by X-ray diffraction studies (see Scheme 5, bottom). The molecule contains the two NHC ligands in the expected trans arrangement, one of them functionalized with the Bpin fragment with the new formed C−B bond. A hydride ligand and a water molecule complete the coordination sphere around the platinum center. One of the oxygen atoms of the Bpin moiety (O2), is oriented towards one of the hydrogen atoms of the water molecule and the close distances (d(O2⋅⋅⋅H3b)=1.76 Å; d(O2⋅⋅⋅O3)=2.63(1) Å) suggest hydrogen bonding character.

The reaction of 1.2 and 1.3 with HBcat was also performed. Again, a very fast reaction takes place that according to their 1H and 11B NMR spectra appear to give the C−B coupling products 3.2 b and 3.3 b (hydride signal ca. −25.0 ppm with 1JPt,H ∼2220 Hz and 11B NMR at ca. 34 ppm). However, the reaction is not as clean as that with HBpin thwarting their full characterization and obtention in pure form. Although the crude reaction mixtures of these reactions do not show signs of formation of the products arising from C−H bond coupling (i. e. the boryl species [Pt(Bcat)(IMes)2][BArF], 4.2 b, and [Pt(Bcat)(IMes*)2][BArF], 4.3 b, see below), apparently the hydride complexes 3.2 b and 3.3 b are able to further react with HBcat releasing hydrogen and giving rise to new species of unknown composition.

Very interestingly, compounds 3.2 a and 3.3 a are perfectly stable and remain unaltered after prolonged periods of time, even under heating in methylene chloride (Scheme 6). These results clearly contrast with those observed when using complex 1.1, for which the formation of the C−B bond is a reversible process that ultimately yields the platinum boryl species 4.1 ae. The first question that arises is if the platinum boryl complexes with IMes and IMes* are sufficiently stable to be isolable. In this sense we envisioned an alternative route to prepare them by reaction of the platinum hydrides [Pt(H)(IMes)2][BArF] (5.2) and [Pt(H)(IMes*)2][BArF] (5.3) with HBpin (Scheme 6). The reaction takes place smoothly under mild heating (60 °C), leading (quantitatively by NMR spectroscopy) to the boryl complexes [Pt(Bpin)(IMes)2][BArF] (4.2 a) and [Pt(Bpin)(IMes*)2][BArF] (4.3 a) in moderate isolated yields. The compounds are stable and can be stored in the absence of moisture indefinitely. The structure of 4.2 a has been ascertained by X-ray diffraction studies (Figure 3). It shows structural metric parameters in line with complexes 4.1 c, e (Pt−B bond length=2.029(4)Å), with the two IMes ligands in a trans position and the Bpin fragment in trans to the formally vacant site (the closest H atom to platinum lies at 2.93 Å; note that these boryl compounds do not bind water in contrast to hydride derivatives 3.2 a and 3.3 a, possibly due to the higher trans influence of the boryl ligand in comparison to the hydride).21 HBcat also reacts with platinum hydride complexes 5.2 and 5.3 generating the corresponding boryl complexes 4.2 b and 4.3 b (Scheme 6), but at variance with HBpin the reaction is very fast at 25 °C. The structure of 4.3 b was confirmed by X-ray diffraction studies (Figure 3, right). This difference in reactivity might explain the difficulties in isolation of 3.2 b and 3.3 b (as mentioned above these hydrides also react with HBcat leading to mixtures of products).

Details are in the caption following the image

Potential synthetic routes to obtain complexes 4.2 a, b and 4.3 a, b.

Details are in the caption following the image

ORTEP-type view of the cationic fragment of complexes 4.2 a (left) and 4.3 b (right). Some carbon atoms and the hydrogen atoms of the molecule have been omitted for clarity.

Thus, the fact that the C−B coupling products 3.2 a and 3.3 a do not evolve to the boryl species 4.2 a and 4.3 a might be related to either a higher energy reaction pathway for a reversible C−B bond cleavage process (as in Scheme 1) or a different mechanistic route in which formation of the C−B bond is favored over the C−H one. To clarify this point we carried out DFT calculations (M06 functional) in dichloromethane solvent (SMD continuum model) on the different possible reaction pathways (see Supporting Information for details). Firstly, the reaction between cyclometalated complex 1.2 and HBpin was explored to compare the resulting energy surface to that previously observed when using complex 1.1.10 Thus, the influence of the NHC ligands on the C−B bond formation/cleavage can be evaluated on both systems. Therefore, C−H and C−B bond formation pathways (Scheme 1) were considered in complexes possessing both NHC ligands in trans and cis geometries (See Figures S68-S71 for every energy profile). We started analyzing the possible reaction pathways leading to complex 3.2 a. A C−B bond formation pathway with a trans arrangement of the NHC ligands yielded too energy demanding transition states (>30 kcal mol−1) to be experimentally feasible. However, as previously found for the reaction of 1.1,10 considering a cis orientation of the IMes groups afforded an energy landscape that agrees well with the results summarized in Scheme 5 (Figure 4). The first step of this mechanism is the isomerization of cyclometalated complex 1.2 via transition state TS1 (16.0 kcal mol−1). The steric bulk of the IMes ligands makes this isomerization more difficult for 1.2 than for 1.1 (barriers of 16.0 and 12.8 kcal mol−1, respectively), but reachable at the reaction conditions. The stabilities of the cis- isomers of 1.2 and 1.1 are, however, alike (relative Gibbs energies of 13.0 and 11.1 kcal mol−1, respectively). This step yields species 1.2-cis, which is 13.0 kcal mol−1 above its trans isomer. The CNHC−Pt−CNHC angle in this compound is 119.2°, which is slightly wider than that observed for 1.1 (112.6°) as a consequence of the increased steric bulk of the NHC ligands. Coordination of HBpin is almost isoenergetic (2.2 a-cis, 13.2 kcal mol−1), after which the C−B and Pt−H bonds are forged in a concerted manner in TS2. This transition state is located 22.0 kcal mol−1 above the origin and yields species Int1-cis (15.5 kcal mol−1), which still bears both NHC ligands in a cis geometry and it exhibits an agostic interaction through the methylene unit (Pt−H=2.06 Å, C−H−Pt=113°). However, the steric hindrance of the IMes groups is evident once again by observing the CNHC−Pt−H angle of 169.5°, which is more acute than that observed for its ItBuiPr analogue (174.9°). Orienting the carbene ligands in a trans geometry gives 3.2 a, the second most stable species of the energy profile (−3.3 kcal mol−1). Coordination of adventitious water to the vacant position of this complex gives experimentally observed 3.2 a⋅H2O, only 0.5 kcal mol−1 above.22 Overall, the process needs to overcome energy barriers comparable to those observed for the ItBuiPr system10 (ItBuiPr: 15.6 kcal mol−1 vs IMes: 22.0 kcal mol−1). From here, cleaving the C−B bond to give boryl species 4.2 a would require overcoming TS3, but this barrier is too high in energy (39.8 kcal mol−1) to be reached under the experimental conditions. This proposed mechanism fits well with 3.2 a⋅H2O being the only observed species at the end of the reaction.

Details are in the caption following the image

Gibbs energy profile in dichloromethane for the C−B coupling reaction between 1.2 and HBpin (NHC ligands in cis). Gibbs energies at 298 K in kcal mol−1. The Gibbs energy of 1.2+HBpin has been taken as zero-energy. Energy values in grey correspond to the analogous species using complex 1.1 instead of 1.2 (for comparison purposes).

Then, we explored the energy costs for the competitive reaction pathway leading to a C−H bond coupling process. Expectedly, too high energy barriers (41.8 kcal mol−1, Figure S71 in the ESI) have been calculated for a reaction pathway in which the IMes ligands are in a trans disposition. Importantly, C−H bond forming scenarios involving a cis rearrangement of the IMes ligands also gave very high energy barriers (>35 kcal mol−1, Figure 5), in agreement with the lack of experimental observation of 4.2 a by NMR. This value is considerably higher than that observed for the platinum system bearing the ItBuiPr ligand (complex 1.1), for which a barrier of 22.1 kcal mol−1 was calculated.10 This latter barrier can be, however, surpassed under the experimental reaction conditions (25 °C), which explains the reversibility of the process for the C−B bond formation/cleavage observed for complex 1.1.

Details are in the caption following the image

Gibbs energy profile in dichloromethane for the C−H coupling reaction between 1.2 and HBpin (NHC ligands in cis). Gibbs energies at 298 K in kcal mol−1. The Gibbs energy of 1.2+HBpin has been taken as zero-energy. Energy values in grey correspond to the analogous species using complex 1.1 instead of 1.2 (for comparison purposes).

Next, the formation of boryl complexes 4.2 a and 4.2 b from hydride complex 5.2 was computed. Contrary to that observed for the previous energy profiles, mechanisms involving complexes in a cis geometry gave the highest barriers (Figures S74 and S75), making σ-CAM mechanisms unfeasible for this transformation. The potential energy surface for the reaction between 5.2 and HBcat is depicted in Figure 6. Since this mechanism only involves species in a trans geometry, it is much simpler; coordination of the borane is slightly exergonic, giving 6.2 b −1.9 kcal mol−1 below the energy reference. Upon binding (Pt−H−B=113.0°), the B−H bond slightly weakens based on its elongation (1.219 Å in the σ complex vs. 1.178 Å in the free borane). Then, the H (from Bcat) fragment migrates in an orbiting fashion23 assisted by the Pt atom toward the hydride ligand in TS7 (26.3 kcal mol−1). This transition state has a dihydride nature, pointing out a mechanism involving concerted B−H oxidative addition/H2 reductive elimination for the formation of the boryl species. Interestingly, the Pt−H distance of the migrating hydride is shorter (1.520 Å) than that of the H atom trans to the Bcat fragment (1.681 Å) due to the trans influence of the latter. After the transition state, complex 4.2 b⋅H2 is formed (−2.8 kcal mol−1), which features a dihydrogen molecule bound to the metal trans to the Bcat moiety. Release of H2 from the complex affords boryl species 4.2 b, the most thermodynamically stable compound of the profile, in agreement with the experimental observations. Upon H2 release, the Pt−B bond distance decreases from 1.988 to 1.968 Å. The reaction between complex 5.2 and HBpin leading to boryl complex 4.2 a follows the same reaction pathway, although with higher energetic barriers in agreement with experimental observations (see Figure S73 in the ESI).

Details are in the caption following the image

Gibbs energy profile in dichloromethane for the reaction between 5.2 and HBcat. Gibbs energies at 298 K in kcal mol−1. The Gibbs energy of 5.2+HBcat has been taken as zero-energy.

On the other hand, the reaction of complexes 1.2 and 1.3 with diamino boranes HBdab and HBMedab led to complex mixtures of products from which no clear conclusions could be drawn. No reaction was observed between complex 1.2 and HBMe2dab, even under heating at 60 °C. Nevertheless, HBMe2dab reacts slowly with complex 1.3 over a period of 40 h at rt but, interestingly, the major product observed when monitoring this reaction by NMR spectroscopy appeared to be platinum boryl complex 4.3 e (Scheme 7).24 However, another species which seem to have a rather unsymmetrical environment is formed at the same time in small amounts precluding isolation of complex 4.3 e in its pure form. The 1H NMR of complex 4.3 e reveals a highly symmetric environment in line with complexes 4.2 a, b and 4.3 a, b. The 11B NMR spectrum shows, besides the signal for the BArF anion, a broad resonance at ca. 10 ppm for the BMe2dab fragment in the expected region for this type of platinum boryl complexes.

Details are in the caption following the image

Reaction of complex 1.3 with HBMe2dab.

The reasons underlying this reaction pathway are not entirely understood, but the increased steric hindrance of HBMe2dab in comparison with HBpin and HBcat (at the boron atom environment) might hamper the coupling of the carbon and boron atoms. With respect to the minor product, it is difficult to establish its nature based on NMR spectroscopy, and we speculate that it might be related to the competitive formation of the C−B coupling product. Unfortunately, we have not been able to ascertain its chemical nature. Investigations along these lines are currently undertaken and will be reported in due course.

Conclusion

The reaction of the 14-electron Pt(II) complexes of general formula [Pt(NHC’)(NHC)][BArF] with a series of tri-coordinated boranes at low temperature led to the initial formation of the corresponding σ-BH complexes [Pt(HBR2)(NHC’)(NHC)][BArF] some of which are sufficiently stable to be characterized by X-ray diffraction studies or to be detected by NMR spectroscopy. These compounds exhibit an uncommon η1-coordination mode. At certain range of temperatures these systems evolve through a process that involves the formation of carbon-boron bonds, in which the nature of the borane does not appear to have an important impact other than their thermal stability. Thus, for platinum systems bearing the NHC ligand ItBuiPr the formation of the C−B bond is a reversible process that, according to our previous observations,10 can lead back to the σ-BH complexes that, at higher temperatures, evolve towards a competitive, energetically accessible, process leading to a thermodynamically more stable platinum-boryl complex and formation of a C−H bond, regardless of the nature of the borane. However, it has been found that the nature of the NHC ligand has an important effect on the energetic barriers that direct the σ-BH complexes in one direction or the other. For the IMes-based platinum system, formation of the C−B bond is energetically more favorable, requiring only 22 kcal mol−1 in a process that involves trans to cis isomerization of the NHC ligands. It is worth noting that in spite of the more thermodynamically stable platinum boryl complexes (4.2 a) compared to their platinum hydride isomers (arising from a C−B coupling process, 3.2 a) the energetic barrier needed to form them is too high (35.5 kcal mol−1 in the best case scenario), therefore explaining why these systems are not able to undergo a reversible C−B process. These results indicate that productive formation of C−B bonds can be controlled by modifying the nature of the NHC ligands.

Experimental section

General: All manipulations were carried out using standard Schlenk and glovebox techniques, under an atmosphere of argon and of high purity nitrogen, respectively. All solvents were dried and degassed prior to use. n-Pentane was distilled over sodium and stored under Na/K alloy. Dichloromethane-d2 (CD2Cl2) was heated under reflux over calcium hydride, distilled under Argon and stored under 3 Å activated molecular sieves. NMR spectra were recorded on Bruker DRX-500, DRX-400 and DPX-300 spectrometers, and they were referenced to external SiMe4 (δ 0 ppm) using the residual protio solvent peaks as internal standard (1H NMR experiments) or the characteristic resonances of the solvent nuclei (13C NMR experiments). 11B NMR spectra were referenced to an external standard of BF3⋅Et2O. Spectral assignments were made by routine one- and two-dimensional NMR experiments where appropriate. Elemental analysis was carried out with a LECO TruSpec CHN elementary analyser. SMe2⋅BH3, HBpin and HBcat were purchased from Aldrich or TCI, stored under argon and used as received. Boranes HBdab, HBMedab and HBMe2dab were prepared from SMe2⋅BH3 and 1,2-phenylenediamine, N-Methyl-1,2-phenylenediamine and N,N′-Dimethyl-1,2-diaminobenzene, respectively, as previously described.25 Complexes [Pt(ItBuiPr’)(ItBuiPr)][BArF], 1.1, [Pt(IMes’)(Mes)][BArF], 1.2, and [Pt(IMes*’)(Mes*)][BArF], 1.3, were obtained accordingly to published procedures.11, 17

Synthesis of HMedab and HBMe2dab: In a flame dried Schlenk tube N-Methyl-1,2-phenylenediamine (0.79 mL, 7.4 mmol) or N,N′-Dimethyl-1,2-diaminobenzene were dissolved, under argon in 20 mL of CH2Cl2. Then, SMe2⋅BH3 (0.7 mL, 7.3 mmol) was added and the mixture was heated at 50 °C until no further H2 evolution is observed (aprox. 2 h). The solvent was then removed under vacuum (at 0 °C in the case of HBMe2dab) and the resulting solid (HBMedab) was washed with 10 mL of pentane whereas the oil arising from HBMe2dab was extracted with pentane (2×20 ml). HBMedab was obtained as an off-white solid in 89 % yield, and HBMe2dab was obtained as a light brownish oil in 51 % yield. HBMedab: 1H-NMR (300 MHz, CD2Cl2, 25 °C): δ=7.11–6.90 (m, 4H, CH), 6.74 (br, 1H, NH), 4.56 (br q, 1H, JB,H=150 Hz, BH), 3.43 (s, 3H, NCH3). 13C{1H} NMR (100 MHz, CD2Cl2, 25 °C): δ=138.5 and 136.7 (Cq-N), 119.4, 119.2, 111.2 and 108.8 (CH). 11B{1H}(128 MHz, CD2Cl2, 25 °C): δ=25.0. HBMe2dab: 1H-NMR (300 MHz, CD2Cl2, 25 °C): δ=7.05 (br, 4H, CH), 4.56 (br q, 1H, JB,H=150 Hz, BH), 3.43 (s, 6H, NCH3). 3C{1H} NMR (100 MHz, CD2Cl2, 25 °C): δ=139.0 (Cq-N), 119.0 and 108.5 (CH), 31.0 (N−CH3). 11B{1H}(128 MHz, CD2Cl2, 25 °C): δ=26.0.

Synthesis of complexes 2.1 d, e and 2.2 a. General procedure: Complex 1.1 (70 mg, 0.050 mmol) was dissolved in 0.4 mL of dry CD2Cl2 in a screw-cap NMR tube and cooled to −78 °C. Then, a borane (HBMedab, HBMe2dab or HBpin) solution (0.060 mmol in 0.2 mL of CD2Cl2) was transferred via cannula into the NMR tube, and the mixture was shaken at −78 °C. The sample was analyzed by NMR in a pre-cooled (−30 °C) NMR apparatus. The 1H NMR signals of complex 2.2 a were too complex to be assigned. Complex 2.1 d: 1H{11B}-NMR (400 MHz, CD2Cl2, −20 °C): δ=7.75 (br, 8H, Hortho-BArF), 7.57 (br, 4H, Hpara-BArF), 7.26 and 7.13 (br s, 1H each, =CH), 7.10 (d, 1H, JH,H=1.5 Hz, =CH), 7.06–6.91 (m, 5 H, =CH and CH-Bdan), 6.33 (br s, 1H, NH), 4.75 and 4.63 (br, 1H each, CH(CH3)2), 2.71 (br, 3H, N−CH3), 2.36 (d+d, 1H, JH,H=12.0 Hz, JPt,H=86.6 Hz, CH2−Pt), 2.27 (d+d, 1H, JH,H=12.0 Hz, JPt,H=71.0 Hz, CH2−Pt), 1.81 (s, 9H, tBu), 1.55 (br s, 3H, Pt−CH2(CH3)2), 1.51 (d, 3H, JH,H=6.5 Hz, CH(CH3)2), 1.41 (br, 6 H, Pt−CH2(CH3)2 and CH(CH3)2), 1.25 (d, 3H, JH,H=6.2 Hz, CH(CH3)2), 0.61 (br, 3H, CH(CH3)2), −2.30 (br s+d, 1H, JPt,H=328.3 Hz, Pt−H−B). 13C{1H} NMR (100 MHz, CD2Cl2, −20 °C): δ=168.3 and 165.0 (Pt=C), 162.2 (q, JC,B=50 Hz, Cipso-BArF), 138.2 and 135.9 (Cq-Bdan), 134.8 (br, Cortho-BArF), 128.8 (br q, JC,F=30 Hz, Cmeta-BArF), 124.5 (q, JC,F=272 Hz, CF3-BArF), 120.9 (=CH), 120.2 (2×CH-Bdab), 117.6 (br, Cpara-BArF and =CH), 116.7 and 116.3 (=CH), 111.1 and 108.9 (CH-Bdan), 66.0 (Pt−CH2C(CH3)2), 59.0 (Cq-tBu), 53.1 (CH(CH3)2), 33.7 (Pt−CH2), 32.1 (CH3), 31.3 (tBu), 29.7 (N−CH3), 28.4 (CH3), 24.9, 24.3, 21.7 and 21.2 (CH(CH3)2). 11B{1H}(160 MHz, CD2Cl2, −20 °C): δ=19.5 (broad, Bdan), −6.7 (s, BArF). Complex 2.1 e: 1H{11B}-NMR (400 MHz, CD2Cl2, −30 °C): δ=7.75 (br, 8H, Hortho-BArF), 7.58 (br, 4H, Hpara-BArF), 7.23, 7.12 and 7.10 (br s, 1H each, =CH), 7.04 (br, 5 H, =CH and CH-Bdan), 7.10 (br s, 1H, =CH), 4.62 (br, 1H, CH(CH3)2), 4.51 (sept, JH,H=6.3 Hz, 1H, CH(CH3)2), 3.06 and 2.81 (s, 3H each, N−CH3), 2.41 (d+d, 1H, JH,H=11.2 Hz, JPt,H=88.0 Hz, CH2−Pt), 2.23 (d+d, 1H, JH,H=12.0 Hz, JPt,H ∼70 Hz, CH2−Pt), 1.81 (s, 9H, tBu), 1.57 (br s, 3H, Pt−CH2(CH3)2), 1.51–137 (m, 9H, Pt−CH2(CH3)2 and CH(CH3)2), 1.20 and 0.34 (br, 3H, CH(CH3)2), 1.25 (d, 3H, JH,H=6.2 Hz, CH(CH3)2), 0.61 (br, 3H, CH(CH3)2), −2.11 (br s+d, 1H, JPt,H=320.0 Hz, Pt−H−B). 13C{1H} NMR (100 MHz, CD2Cl2, −30 °C): δ=169.0 and 165.2 (Pt=C), 161.8 (q, JC,B=50 Hz, Cipso-BArF), 138.4 (Cq-Bdan), 134.8 (br, Cortho-BArF), 128.8 (br q, JC,F=30 Hz, Cmeta-BArF), 124.5 (q, JC,F=272 Hz, CF3-BArF), 121.1 (=CH), 120.1 (CH-Bdab), 117.6 (br, Cpara-BArF and =CH), 117.3, 116.8 and 116.5 (=CH), 108.4 (CH-Bdan), 65.9 (Pt−CH2C(CH3)2), 59.2 (Cq-tBu), 53.9 and 53.1 (CH(CH3)2), 32.9 (CH3), 32.6 (Pt−CH2), 31.5 (tBu), 31.1 and 29.9 (N−CH3), 28.4 (CH3), 24.8, 24.2, 21.4 and 21.2 (CH3 and CH(CH3)2). 11B{1H}(160 MHz, CD2Cl2, −30 °C): δ=20.3 (broad, Bdan), −6.7 (s, BArF). Complex 2.2 a: 1H{11B}-NMR (400 MHz, CD2Cl2, −60 °C): δ=7.75 (br, 8H, Hortho-BArF), 7.55 (br, 4H, Hpara-BArF), 7.43–6.71 (m, 12H, =CH), 2.721 (d+d, 1H, JH,H=9.5 Hz, JPt,H=97.3 Hz, CH2−Pt), 2.53 (d, 1H, JH,H=9.5 Hz, JPt,H not distinguible, CH2−Pt), 2.46-1.68 (m, 33H, CH3), 1.19 (s, 12H, CH3-HBpin), −5.22 (br s+d, 1H, JPt,H=355.0 Hz, Pt−H−B). 166.7 and 166.4 (Pt=C), 162.2 (q, JC,B=50 Hz, Cipso-BArF), 139.1, 139.0, 136.7, 135.6, 135.5, 135.3, 134.9, 134.8, 134.1 and 133.8 (Cq, Mes and CH-Mes), 134.4 (br, Cortho-BArF), 129.8-128.8. 124.2, 120.5 and 120.2 (m, CH-Mes, =CH), 129.2 (br q, JC,F=30 Hz, Cmeta-BArF), 125.1 (q, JC,F=272 Hz, CF3-BArF), 117.39 (br, Cpara-BArF), 83.2 (Cq-Bpin), 24.4 (CH3-Bpin), 22.7, 21.2, 20.9, 20.8, 19.2, 18.6, 18.3,18.2, 17.4 and 17.2 (CH3-Mes), 14.4 (br, CH2−Pt). 11B{1H}(128 MHz, CD2Cl2, −60 °C): δ=−6.7 (s, BArF) (the signal for the coordinated HBpin was not observed).

Synthesis of complexes 4.1 d, e. General procedure: Complex 1.1 (80 mg, 0.057 mmol) was dissolved in 2 mL of dry CH2Cl2 and mixed via cannula with a solution of the corresponding borane (0.063 mmol) in 0.5 mL of CH2Cl2. The mixture was stirred at rt for 72 h and the solvent was removed under reduced pressure. Complexes 4.1 d and 4.1 e were washed with pentane (2×5 mL) and dried under vacuum. Compounds 4.1 d and 4.1 e can be crystallized by slow diffusion of a concentrated solution of CH2Cl2 into hexamethyldisiloxane (4.1 d) or pentane (4.1 e). (Yields: 4.1 d: 60 %; 4.1 e: 54 %). Complex 4.1 d: 1H-NMR (400 MHz, CD2Cl2, 25 °C): δ=7.73 (br, 8H, Hortho-BArF), 7.57 (br, 4H, Hpara-BArF), 7.23 (d, 2H, JH,H=2.0 Hz, =CH), 6.97 (d, 2H, JH,H=2.0 Hz, =CH), 6.91-6.78 (m, 4H, CH-Bdab), 5.78 (br s, 1H, NH), 5.01 (sept, 2H, JH,H=6.7 Hz, CH(CH3)2), 2.96 (s, 3H, N−CH3), 1.95 (s, 18H, tBu), 1.08 (d, 12 H, JH,H=6.7 Hz, CH(CH3)2). 13C{1H} NMR (100 MHz, CD2Cl2, 25 °C): δ=179.3 (Pt=C), 162.2 (q, JC,B=50 Hz, Cipso-BArF), 138.4 and 136.3 (Cq-Bdan), 135.2 (br, Cortho-BArF), 129.3 (br q, JC,F=30 Hz, Cmeta-BArF), 125.1 (q, JC,F=272 Hz, CF3-BArF), 120.5 (=CH), 119.24 and 119.18 (CH-Bdab), 117.9 (br, Cpara-BArF), 116.1 (=CH), 110.1 and 108.2 (CH-Bdan), 59.1 (Cq-tBu), 55.2 (CH(CH3)2), 33.0 (tBu), 30.9 (N−CH3), 22.8 (CH(CH3)2). 11B{1H}(160 MHz, CD2Cl2, 25 °C): δ=11.7 (s+d, broad, Bdan, JPt,B ∼1370 Hz), −6.6 (s, BArF). Elemental analysis calcd (%) for C59H56B2F24N6Pt: C, 46.57; H, 3.71; N, 5.52; Found: C, 46.4; H, 3.7; N, 5.6. Complex 4.1 e: 1H-NMR (400 MHz, CD2Cl2, 25 °C): δ=7.74 (br, 8H, Hortho-BArF), 7.57 (br, 4H, Hpara-BArF), 7.23 (d, 2H, JH,H=1.2 Hz, =CH), 6.99 (d, 2H, JH,H=1.2 Hz, =CH), 6.93–6.85 (m, 4H, CH-Bdab), 4.85 (sept, 2H, JH,H=6.3 Hz, CH(CH3)2), 2.95 (s, 6H, N−CH3), 1.93 (s, 18H, tBu), 1.01 (br, 12H, CH(CH3)2). 13C{1H} NMR (100 MHz, CD2Cl2, 25 °C): δ=178.2 (Pt=C), 162.2 (q, JC,B=50 Hz, Cipso-BArF), 138.4 (Cq-Bdan), 135.2 (br, Cortho-BArF), 129.3 (br q, JC,F=30 Hz, Cmeta-BArF), 125.1 (q, JC,F=272 Hz, CF3-BArF), 120.8 (=CH), 119.0 (CH-Bdab), 117.9 (br, Cpara-BArF), 116.6 (=CH), 108.7 (CH-Bdan), 59.3 (Cq-tBu), 55.3 (CH(CH3)2), 33.2 (tBu), 30.8 (N−CH3), 22.9 (br, CH(CH3)2). 11B{1H}(160 MHz, CD2Cl2, 25 °C): δ=13.6 (s+d, broad, Bdan, JPt,B ∼1420 Hz), −6.6 (s, BArF). Elemental analysis calcd (%) for C60H58B2F24N6Pt: C, 46.92; H, 3.81; N, 5.47; Found: C, 47.1; H, 3.6; N, 5.4.

Synthesis of complexes 3.2 a, 3.3 a, 3.2 a⋅H2O, 3.3 a⋅H2O, 3.2 a⋅py and 3.3 a⋅py. General procedure: Complexes 1.2 or 1.3 (0.048 mmol) were dissolved in 0.7 mL of CD2Cl2 in a vial in the glovebox. The, HBpin (0.057 mmol, 8.3 μL) were added o this solution and transferred to an NMR tube. After 15 min at rt, complexes 3.2 a and 3.3 a were formed quantitatively according to 1H NMR spectroscopy (see Figures S34–S36 and Figures S43–S45 in the ESI). The corresponding water and pyridine adducts were formed by addition of an excess of water (0.096 mmol, 1.7 μL) or pyridine (0.072 mmol, 5.8 μL). The reaction is instantaneous. The solvent was the removed under vacuum and the resulting solids washed with pentane (2×5 mL). The solids were dried under vacuum leading to white or off-white solids. (Yields: 3.2 a⋅H2O: 65 %; 3.2 a⋅py: 63 %; 3.3 a⋅H2O: 58 %; 3.3 a⋅py: 53 %). Complex 3.2 a⋅H2O: 1H-NMR (400 MHz, CD2Cl2, 25 °C): δ=7.73 (br, 8H, Hortho-BArF), 7.56 (br, 4H, Hpara-BArF), 6.98–6.90 (m, 10 H, CH-Mes+=CH), 6.84 and 6–78 (s, 1H, each =CH or CH-Mes), 6.91–6.80 (m, 13H, Mes, =CH), 2.45, 2.43, 2.41 and 2.38 (s, 1 : 2 : 1 : 1 ratio, 12H, CH3-Mes), 2.10 (d, 1H, JH,H=16.0 Hz, CH2-Bpin), 1.90 (s, 6H, CH3-Mes), 1.82 (s, 3H, CH3-Mes), 1.77 (s, 6H, CH3-Mes), 1.71 (s, 3H, CH3-Mes), 1.72 (H2O), 1.68 (br s, 4H, 1.82 (s, 3H, CH3-Mes+CH2-Bpin), 1.14 (s, 12H, CH3-Bpin), −27.96 (s+d, 1H, JPt,H=1899.2 Hz, Pt−H). 13C{1H} NMR (100 MHz, CD2Cl2, 25 °C): δ=176.9 and 176.5 (Pt=C), 162.2 (q, JC,B=50 Hz, Cipso-BArF), 139.7, 139.6, 139.4, 136.6, 136.5, 135.8, 135.7, 135.4, 135.0 and 134.9 (Cq, Mes), 136.2, 136.1, 135.9, 135.7, 135.59, 135.57, 135.5 (Cq-Mes and CH-Mes), 135.2 (br, Cortho-BArF), 129.7, 129.5, 129.4 and 129.3 (CH-Mes), 129.2 (br q, JC,F=30 Hz, Cmeta-BArF), 128.3 (CH-Mes), 125.1 (q, JC,F=272 Hz, CF3-BArF), 122.9, 122.6 and 122.4 (1 : 2 : 1 ratio, =CH), 117.9 (br, Cpara-BArF), 85.0 (Cq-Bpin), 24.7 and 24.5 (CH3-Bpin), 21.34, 21.27, 21.24, 18.7, 18.0, 17.83. 17.75 and 17.6 (CH3-Mes), 15.0 (br, CH2-Bpin). 11B{1H}(128 MHz, CD2Cl2, 25 °C): δ=32.3 (br, Bpin), −6.6 (s, BArF). Elemental analysis calcd (%) for C80H74B2F24N5O3Pt: C, 53.02; H, 4.12; N, 3.09; Found: C, 53.1; H, 4.3; N, 3.1. Complex 3.2 a⋅py: 1H-NMR (400 MHz, CD2Cl2, 25 °C): δ=7.75 (br, 8H, Hortho-BArF), 7.55 (br, 4H, Hpara-BArF), 7.52 (t, 1H, JH,H=7.6 Hz, CH-py), 7.05 (d, 2H, JH,H=7.6 Hz, CH-py), 6.97 (s, 4H, =CH), 6.91–6.80 (m, 13H, Mes, =CH and CH-py), 2.41 (s, 12H, CH3-Mes), 1.70–1.58 (m, 24H, CH3-Mes), 1.56 and 1.50 (d, 1H each, JH,H=16.1 Hz, CH2-Bpin), 1.14 and 1.11 (s, 6H each, CH3-Bpin), −20.3 (s+d, 1H, JPt,H=1381.0 Hz, Pt−H). 13C{1H} NMR (100 MHz, CD2Cl2, 25 °C): δ=173.8 and 173.9 (Pt=C), 162.2 (q, JC,B=50 Hz, Cipso-BArF), 151.3 (CH-py), 139.38, 139.36, 137.3 and 139.3 (Cq, Mes), 136.9 (CH-py), 136.2, 136.1, 135.9, 135.7, 135.59, 135.57, 135.5 (Cq-Mes), 135.2 (br, Cortho-BArF), 129.43 and 129.37 (CH-Mes), 129.3 (br q, JC,F=30 Hz, Cmeta-BArF), 129.0 (CH-Mes), 128.9 (CH-py), 125.2 (CH-Mes), 125.1 (q, JC,F=272 Hz, CF3-BArF), 123.8, 122.6 and 122.0 (1 : 2 : 1 ratio, =CH), 117.9 (br, Cpara-BArF), 84.1 (Cq-Bpin), 25.0 and 24.8 (CH3-Bpin), 21.31, 21.26, 18.2 and 18.0 (CH3-Mes), 17.1 (br, CH2-Bpin). 11B{1H}(128 MHz, CD2Cl2, 25 °C): δ=32.7 (very broad, Bpin), −6.6 (s, BArF). Elemental analysis calcd (%) for C85H77B2F24N5O2Pt: C, 54.50; H, 4.14; N, 3.74; Found: C, 54.8; H, 4.0; N, 3.7. Complex 3.3 a⋅H2O: 1H-NMR (400 MHz, CD2Cl2, 25 °C): δ=7.73 (br, 8H, Hortho-BArF), 7.57 (br, 4H, Hpara-BArF), 6.96–6.88 (m, 6 H, CH-Mes), 6.83 and 6.78 (s, 1H each, CH-Mes), 2.45, 2.43, 2.41 and 2.38 (s, 12H global, CH3), 2.05 (d, 1H, JH,H=16.5 Hz, CH2-Bpin), 1.83–1.61 (s, 33H global, CH3), 1.49 (d, 1H, JH,H=16.5, CH2-Bpin), 1.15 (s, 12H, CH3-Bpin), −28.3 (s+d, 1H, JPt,H=1907.4 Hz, Pt−H). 13C{1H} NMR (100 MHz, CD2Cl2, 25 °C): δ=173.9 and 173.4 (Pt=C), 162.2 (q, JC,B=50 Hz, Cipso-BArF), 139.3, 139.2, 139.0, 138.8, 137.2, 136.22. 136.16, 135.8, 135.47, 135.44, 134.5 and 133.6 (Cq, Mes), 135.2 (br, Cortho-BArF), 129.7, 129.6, 129.5, 129.3 and 128.1 (br, CH-Mes), 129.3 (br q, JC,F=30 Hz, Cmeta-BArF), 125.8, 125.7, 125.6 and 125.5 (=C−CH3), 125.1 (q, JC,F=272 Hz, CF3-BArF), 117.9 (br, Cpara-BArF), 85.0 (Cq-Bpin), 24.8 and 24.4 (CH3-Bpin), 21.4, 21.3, 18.5, 17.8, 17.7 and 17.6 (CH3-Mes), 14.6 (very broad, CH2-Bpin), 9.4, 9.2 and 9.1 (1 : 1 : 2 ratio, =C−CH3). 11B{1H}(128 MHz, CD2Cl2, 25 °C): δ=32.5 (very broad, Bpin), −6.6 (s, BArF). Elemental analysis calcd (%) for C85H72B2F24N4O3Pt: C, 55.41; H, 4.44; N, 3.63; Found: C, 55.6; H, 4.6; N, 3.7. Complex 3.3 a⋅py: 1H-NMR (400 MHz, CD2Cl2, 25 °C): δ=7.74 (br, 8H, Hortho-BArF), 7.58 (br, 4H, Hpara-BArF), 7.52 (t, 1H, JH,H=7.5 Hz, CH-py), 7.04 (d, 2H, JH,H=7.6 Hz, CH-py), 6.91–6.78 (m, 10 H, CH-Mes and CH-py), 2.43 (br, 12H, CH3-Mes), 1.74 (s, 3H, CH3), 1.70 (d, 1H, JH,H=16.1, CH2-Bpin), 1.66 (br, 9H, CH3), 1.57 (br, 3H, CH3), 1.53 and 1.50 s, 3H each, CH3), 1.38 (d, 1H, JH,H=16.1, CH2-Bpin), 1.16 and 1.13 (s, 6H each, CH3-Bpin), −20.5 (s+d, 1H, JPt,H=1386.0 Hz, Pt−H). 13C{1H} NMR (100 MHz, CD2Cl2, 25 °C): δ=170.4 and 170.3 (Pt=C), 162.2 (q, JC,B=50 Hz, Cipso-BArF), 151.6 (CH-py), 139.05, 139.01, 138.96, 138.7 and 137.7 (2 : 1 : 1 : 1 : 1 ratio, Cq, Mes), 136.6 (CH-py), 136.13, 136.08, 136.0 (2 : 2 : 2 ratio, Cq, Mes), 135.2 (br, Cortho-BArF), 134.8, 134.7 and 134.3 (1 : 2 : 1 ratio, Cq, Mes), 129.43 (br, CH-Mes), 129.3 (br q, JC,F=30 Hz, Cmeta-BArF), 128.8 (CH-py), 125.5, 125.3 and 125.2 (=C−CH3), 125.1 (q, JC,F=272 Hz, CF3-BArF), 117.9 (br, Cpara-BArF), 83.8 (Cq-Bpin), 25.0 and 24.7 (CH3-Bpin), 21.4, 21.3, 17.9, 17.8 and 17.7 (CH3-Mes), 16.2 (br, CH2-Bpin), 9.5 and 9.2 (=C−CH3). 11B{1H}(128 MHz, CD2Cl2, 25 °C): δ=32.7 (very broad, Bpin), −6.6 (s, BArF). Elemental analysis calcd (%) for C89H75B2F24N5O2Pt: C, 55.41; H, 4.44; N, 3.63; Found: C, 55.6; H, 4.6; N, 3.7.

Synthesis of complexes 4.2 a, b and 4.3 a, b. General procedure: Cyclometallated complexes 1.2 or 1.3 (0.048 mmol) were dissolved in 0.4 mL of CH2Cl2 (or CD2Cl2 for monitoring the reactions) and transferred to a J. Young NMR tube with a screw cap. Thereafter, H2 was introduced at 3 bar to form in situ the corresponding platinum hydride derivatives [Pt(H)(IMes)2][BArF], 5.2, or [Pt(H)(IMes*)2][BArF], 5.3.26 After 30 min, this solution was transferred via cannula to a solution of the corresponding borane (0.144 mmol for HBpin or 0.072 mmol for HBcat) in 0.3 mL of CH2Cl2 (or CD2Cl2). The reactions with HBcat leading to complexes 4.2 b and 4.3 b take place at rt with full conversion after 30 min (gas evolution with a color change from colorless to yellow is observed). Reactions with HBpin to yield complexes 4.2 a and 4.3 a required heating at 60 °C over a period of 24 h. The solvent was then removed under vacuum and the yellow solids washed with pentane (3×4 mL). Although spectroscopic yields were high, the isolated yields decreased due to partial solubility of the final products in pentane. Crystals suitable for X-ray diffraction studies were grown from concentrated CH2Cl2 solution of the complexes into pentane or hexamethyldisiloxane. (Yields: 4.2 a: 65 %; 4.3 a: 54 %; 4.2 b: 76 %; 4.3 b: 60 %). Complex 4.2 a: 1H-NMR (400 MHz, CD2Cl2, 25 °C): δ=7.73 (br, 8H, Hortho-BArF), 7.56 (br, 4H, Hpara-BArF), 6.97 (br s, 4H, =CH), 6.91 (s, 8H, Mes), 2.42 (s, 12H, CH3-Mes), 1.85 (s, 24H, CH3-Mes), 0.76 (s, 12H, CH3-Bpin). 13C{1H} NMR (100 MHz, CD2Cl2, 25 °C): δ=186.7 (Pt=C), 162.2 (q, JC,B=50 Hz, Cipso-BArF), 139.7 (Cq, Mes), 135.4 (br, Cortho-BArF), 135.1 (Cq-N), 129.8 (CH-Mes), 129.3 (br q, JC,F=30 Hz, Cmeta-BArF), 125.1 (q, JC,F=272 Hz, CF3-BArF), 123.7 (=CH), 117.9 (br, Cpara-BArF), 84.6 (Cq-Bpin), 24.2 (CH3-Bpin), 21.3 and 18.6 (1 : 2 ratio, CH3-Mes). 11B{1H}(128 MHz, CD2Cl2, 25 °C): δ=−6.6 (s, BArF) (the signal for the Bpin fragment is too broad to be observable). Elemental analysis calcd (%) for C80H72B2F24N4O2Pt: C, 53.56; H, 4.04; N, 3.12; Found: C, 53.5; H, 3.9; N, 3.4. Complex 4.3 a: 1H-NMR (400 MHz, CD2Cl2, 25 °C): δ=7.73 (br, 8H, Hortho-BArF), 7.57 (br, 4H, Hpara-BArF), 6.88 (s, 8H, Mes), 2.43 (s, 12H, CH3-Mes), 1.78 (s. 24H, CH3-Mes), 1.71 (=C−CH3), 0.85 (s, 12H, CH3-Bpin). 13C{1H} NMR (100 MHz, CD2Cl2, 25 °C): δ=184.4 (Pt=C), 162.2 (q, JC,B=50 Hz, Cipso-BArF), 139.2 (Cq, Mes), 135.5 (Cq-Mes), 135.2 (br, Cortho-BArF), 134.0 (Cq-N), 129.7 (CH-Mes), 129.1 (br q, JC,F=30 Hz, Cmeta-BArF), 127.0 (=C−CH3), 125.0 (q, JC,F=272 Hz, CF3-BArF), 117.9 (br, Cpara-BArF), 84.3 (Cq-Bpin), 24.4 (CH3-Bpin), 21.4 and 18.5 (1 : 2 ratio, CH3-Mes), 9.2 (=C−CH3). 11B{1H}(128 MHz, CD2Cl2, 25 °C): δ=4.5 (very broad, Bpin), −6.6 (s, BArF). Elemental analysis calcd (%) for C84H80B2F24N4O2Pt: C, 54.53; H, 4.36; N, 3.03; Found: C, 54.4; H, 4.2; N, 3.1. Complex 4.2 b: 1H-NMR (400 MHz, CD2Cl2, 25 °C): δ=7.75 (br, 8H, Hortho-BArF), 7.58 (br, 4H, Hpara-BArF), 7.00 (br s, 4H, =CH), 6.90 (m, 2H, Bcat), 6.78 (s, 8H, CH-Mes), 6.70 (m, 2H, Bcat), 2.25 (s, 12H, CH3-Mes), 1.82 (s, 24H, CH3-Mes). 13C{1H} NMR (100 MHz, CD2Cl2, 25 °C): δ=183.4 (Pt=C), 162.2 (q, JC,B=50 Hz, Cipso-BArF), 149.0 (Cq-Bcat), 140.5 (Cq, Mes), 135.2 (br, Cortho-BArF), 133.7 (Cq-N), 129.8 (CH-Mes), 129.3 (br q, JC,F=30 Hz, Cmeta-BArF), 125.1 (q, JC,F=272 Hz, CF3-BArF), 123.5 (=CH), 121.9 (CH-Bcat), 117.9 (br, Cpara-BArF), 111.4 (CH-Bcat), 21.2 and 18.0 (1 : 2 ratio, CH3-Mes). 11B{1H}(160 MHz, CD2Cl2, 25 °C): δ=5.9 (s+d, very broad, Bcat, JPt,B ∼1700 Hz), −6.6 (s, BArF). Elemental analysis calcd (%) for C80H60B2F24N4O2Pt: C, 53.80; H, 3.61; N, 3.14; Found: C, 53.5; H, 4.1; N, 3.1. Complex 4.3 b: 1H-NMR (400 MHz, CD2Cl2, 25 °C): δ=7.74 (br, 8H, Hortho-BArF), 7.57 (br, 4H, Hpara-BArF), 6.90 (m, 2H, CH-Bcat), 6.79 (s, 8H, Mes), 6.72 (m, 2H, CH-Bcat) 2.27 (s, 12H, CH3-Mes), 1.75 (s, 12H, =C−CH3), 1.70 (s. 24H, CH3-Mes). 13C{1H} NMR (100 MHz, CD2Cl2, 25 °C): δ=181.1 (Pt=C), 162.2 (q, JC,B=50 Hz, Cipso-BArF), 149.1 (Cq-Bcat), 140.2 (Cq-Mes), 135.7 (Cq, Mes), 135.2 (br, Cortho-BArF), 132.4 (Cq-N), 129.7 (CH-Mes), 129.1 (br q, JC,F=30 Hz, Cmeta-BArF), 127.0 (=C−CH3), 125.0 (q, JC,F=272 Hz, CF3-BArF), 121.8 (CH-Bcat), 117.9 (br, Cpara-BArF), 111.4 (CH-Bcat), 21.2 and 17.8 (1 : 2 ratio, CH3-Mes), 9.0 (=C−CH3). 11B{1H}(128 MHz, CD2Cl2, 25 °C): δ=6.2 (very broad, Bcat, JPt,B ∼1700 Hz)), −6.6 (s, BArF). Elemental analysis calcd (%) for C84H72B2F24N4O2Pt: C, 54.77; H, 3.94; N, 3.04; Found: C, 53.5; H, 4.1; N, 3.1.

Synthesis of complex 4.3 e. Cyclometallated complex 1.3 (80 mg, 0.046 mmol) was dissolved in 0.7 mL of CH2Cl2 (or CD2Cl2 for monitoring the reactions) and HBMe2Bdab (20 mg, 0.14 mmol) was added. The solution stirred for 40 h a rt until all 1.3 complex was consumed. The solvent was removed under vacuum and the light yellow residue washed with pentane to remove the excess HBMe2Bdab. Attempts to purify this complex by crystallization were unsuccessful and side products remained together with complex 4.3 e. Complex 4.3 e: 1H-NMR (400 MHz, CD2Cl2, 25 °C): δ=7.73 (br, 8H, Hortho-BArF), 7.57 (br, 4H, Hpara-BArF), 6.82 (s, 8H, Mes), 6.81 (m, 2H, CH-Bdab), 6.49 (m, 2H, CH-Bdab) 2.33 (s, 6 H, N−CH3), 2.31 (s, 12H, CH3-Mes), 1.66 (s, 12 H, =C−CH3), 1.63 (s, 24H, CH3-Mes). 13C{1H} NMR (100 MHz, CD2Cl2, 25 °C): δ=180.1 (Pt=C), 162.2 (q, JC,B=50 Hz, Cipso-BArF), 139.9 (Cq-Bdab), 140.2 (Cq-Mes), 135.6 (Cq, Mes), 135.2 (br, Cortho-BArF), 133.1 (Cq-N), 129.8 (CH-Mes), 129.1 (br q, JC,F=30 Hz, Cmeta-BArF), 127.3 (=C−CH3), 125.0 (q, JC,F=272 Hz, CF3-BArF), 117.8 (CH-Bdab), 117.9 (br, Cpara-BArF), 106.5 (CH-Bdab), 31.4 (N−CH3), 21.2 and 18.1 (1 : 2 ratio, CH3-Mes), 9.1 (=C−CH3). 11B{1H}(128 MHz, CD2Cl2, 25 °C): δ=10.1 (very broad, Bcat, JPt,B ∼1300 Hz)), −6.6 (s, BArF).

The x-ray crystallographic coordinates for structures 2.1 d, 2.1 e, 3.2 a⋅H2O, 4.1 e, 4.2 a and 4.3 b have been deposited under deposition numbers 2083230–2083235.

Deposition Numbers 2083230 (for 4.1e), 2083231 (for 4.2a), 2083232 (for 4.3b), 2083233 (for 3.2a), 2083234 (for 2.1e), and 2083235 (for 2.1d) contain the supplementary crystallographic data for this paper. These data are provided free of charge by the joint Cambridge Crystallographic Data Centre and Fachinformationszentrum Karlsruhe Access Structures service www.ccdc.cam.ac.uk/structures.

Acknowledgements

Financial support (FEDER contribution) from the Ministerio de Ciencia e Innovación (Projects PID-2020-116861GB-I00, PID2019-109312GB-I00 and RED2018-102387-T), the CSIC (AEPP_CTQ2016-76267-P) and the use of computational facilities of the Supercomputing Center of Galicia (CESGA) is gratefully acknowledged.

    Conflict of interest

    The authors declare no conflict of interest.