Evidence for Genuine Bimetallic Frustrated Lewis Pair Activation of Dihydrogen with Gold(I)/Platinum(0) Systems

1 Evidence for Genuine Bimetallic Frustrated Lewis Pair Activation of Dihydrogen with Gold(I)/Platinum(0) Systems Nereida Hidalgo, Juan José Moreno, Marina Pérez-Jiménez, Celia Maya, Joaquín López-Serrano*, Jesús Campos* Instituto de Investigaciones Químicas (IIQ), Departamento de Química Inorgánica and Centro de Innovación en Química Avanzada (ORFEO-CINQA). Consejo Superior de Investigaciones Científicas (CSIC) and University of Sevilla. Avenida Américo Vespucio 49, 41092 Sevilla (Spain). E-mail: jesus.cam[email protected] ; joaquin.[email protected].es Abstract: A joint experimental/computational effort to elucidate the mechanism of dihydrogen activation by a gold(I)/platinum(0) metal-only frustrated Lewis pairs is described herein. We have also investigated the drastic effects on H2 activation derived from subtle ligand modifications. The importance of the balance between bimetallic adduct formation and complete frustration has been interrogated, providing for the first time evidence for genuine metal-only FLP reactivity in solution. The origin of a strong inverse kinetic isotopic effect has also been clarified, offering further support for the proposed bimetallic FLP-type cleavage of dihydrogen. Introduction Frustrated Lewis pair (FLP) chemistry has become a continuously growing area of research owing to the ability of FLP systems to activate a wide range of small molecules.1 Moving from stoichiometric studies to catalysis has allowed the development of highly efficient hydrogenation processes that do not require the presence of a transition metal.2 Other related catalytic transformations such as hydrosilylation,3 hydroboration,4 hydroamination5 or dehydrogenation6 were soon disclosed for FLP systems and, more recently, the catalytic spectrum has spread with less conventional transformations,7 including C-H functionalization reactions. In most cases, the individual constituents of FLPs are based on main group elements, typically the lighter ones, being phosphine-borane pairs the foremost and most paradigmatic examples. However, the inherent limitations of these elements in terms of redox reactivity have precluded the discovery of more sophisticated catalytic transformations on a regular basis. The incorporation of transition metals and their toolkit of elementary reactions into FLP systems emerged as a promising approach to extend their catalytic usefulness.8 To this regard, pioneering work from the groups of Wass9 and Erker10 on zirconium-phosphine combinations was shortly followed by notable examples designed around a variety of metals,11 demonstrating the potential of transition metal frustrated Lewis pairs (TMFLPs) in small molecule activation and catalysis. In this context, the design of frustrated systems in which the two components are based on transition metals became an obvious target.12 Encouraged by this idea, we recently described the first transition metal-only FLP (TMOFLP) by combining Au(I) and Pt(0) compounds as the acidic and basic counterparts respectively (Scheme 1).13 This metallic pair is active towards the activation of dihydrogen to yield a heterobimetallic dihydride (3a). As discussed later in more detail, we observed a drastically different reactivity for the TMOFLP and for the individual Au(I) and Pt(0) species, demonstrating the cooperative effect imparted by the two metals. 2 Scheme 1. Dihydrogen activation by a Au(I)/Pt(0) transition metal-only frustrated Lewis pair (TMOFLP) (NTf2- (triflimide)= [N(SO2CF3)2]-).¡Error! Marcador no definido. The variety of potential chemical pathways to account for the H2 splitting depicted in Scheme 1 prevented us from providing a solid mechanistic picture during our preliminary investigations.13 In fact, the mechanisms by which TMFLPs operate remain underexplored compared to phosphine-borane and related main group designs. As an archetypical example, a vast amount of work has emerged to gain fundamental mechanistic understanding of the heterolytic cleavage of dihydrogen with traditional FLPs.14 Nonetheless, and despite being the simplest molecule to activate, the precise mechanism remains a matter of intense debate.15 Focusing on metalcontaining systems, most relevant information derives from the concept of metal-ligand cooperation, including prominent hydrogenation catalysts containing pendant Lewis acid or basic sites, whose reactivity resembles that of TMFLPs.8,16 In the same vein, the mechanistic knowledge regarding TMOFLPs is virtually nonexistent, albeit significant insights may be inferred from polar heterobimetallic complexes.17 An important aspect when studying mechanisms in FLP systems is the presence of acid-base interactions. Avoiding the formation of a dative bond between the Lewis acid and the Lewis base was once considered a sine qua non condition for FLP reactivity. However, a number of studies demonstrate that the so-called ‘thermally induced FLPs’,18 in which the resting state is a Lewis adduct, are competent for small molecule activation.19,20 In fact, these systems may even outperform their fully frustrated versions in catalytic applications.21 Thus, a precise control of the degree of frustration can be directly associated with catalytic efficiency (i.e. turnover frequency).22 With this in mind and in the light of the FLP-like reactivity exhibited by heterodinuclear complexes with an explicit M-M bond,17,23 we decided to investigate the effects derived from finely tuning the balance between metal-metal bond formation24 and complete metallic frustration in our gold(I)/platinum(0) system. To switch between the two extreme scenarios we have targeted herein three Lewis acidic gold complexes stabilized by terphenyl phosphines, PR2Ar’ (R = alkyl; Ar’ = C6H3-2,6-Ar2), whose steric parameters were recently investigated by our group (Figure 1).25 We provide a combined experimental and computational investigation that includes isolation and reactivity studies of competent intermediates, kinetic investigations and density functional theory (DFT) calculations with the aim of elucidating the mechanism of dihydrogen activation by Au(I)/Pt(0) TMOFLPs (Scheme 1), paying special attention to the effects of M-M interaction/frustration balance. 3 Figure 1. Lewis acidic gold complexes stabilized by terphenyl phosphines used in this work (Cyp = cyclopentyl). Results and Discussion Frustration versus Lewis adduct formation To modulate the equilibrium between the formation of a dative AuPt bond and complete bimetallic frustration we examined the interaction between Pt(0) compound 2 and Au(I) triflimide complexes 1a-c (Figure 1) bearing phosphines PMe2ArXyl2 (1a), PMe2ArDipp2 (1b) and PCyp2ArXyl2 (1c) (Cyp = cyclopentyl). Compounds 1a and 1c were readily prepared by the same procedure previously reported for 1b.26 The steric shielding provided by these phosphines follows the order PCyp2ArXyl2 > PMe2ArDipp2 > PMe2ArXyl2,25 which has a direct impact on the reactivity of their gold triflimide complexes with compound 2, as will be discussed in detail in the following sections. For instance, when compound 1a, based on the less congested phosphine, is combined with an equimolar mixture of 2, the corresponding bimetallic Lewis adduct 4a is readily formed (Scheme 2). In contrast, compound 1b bearing the intermediate size phosphine (PMe2ArDipp2), seems to be in equilibrium between adduct formation and monometallic fragments. This dynamic behavior is strongly affected by solvent, as previously investigated in detail for traditional FLP systems.27 Thus, in C6D6 solution, the 1H and 31P{1H} NMR spectra of compounds Scheme 2. Solution equilibria for adduct formation vs full frustration in Au(I)/Pt(0) bimetallic pairs as a function of ligand sterics and solvent conditions. Solvent 1a + 2  4a 1b + 2  4b 1c + 2  4c Adduct formation (by NMR) C6D6  X X CD2Cl2  Equilibrium X CD2Cl2/MeOH   X ΔG⁰ solvent (DFT, kcal·mol-1) C6H6 -2.5 +1.7 +17.5 CH2Cl2 -7.2 -2.4 +10.8 4 1b and 2 remained unaltered when mixed together, albeit using the more polar CD2Cl2 resulted in broadening of both of their 31P{1H} NMR signals, an observation that we attributed to the existence of the aforementioned equilibrium.13 Thus, Lewis adduct formation appears slightly favored under the more polar environment provided by CD2Cl2, as expected for the formation of two ionic species (4b and NTf2-). Moreover, we have now found that the addition of 10 equivalents of methanol results in rapid formation of 4b. This result parallels with the recently described reactivity of the 4b:2 pair with germanium dichloride, which also leads to the formation of 4b,28 suggesting that the role of methanol may be facilitating triflimide solvation, promoting the formation of the AuPt bond. Formation of the unsupported heterobimetallic compound 4b was confirmed on the basis of 1H and 31P{1H} NMR data. For the latter case, signals were recorded at 94.5 (1JPPt = 3159, 3JPP = 2 Hz) and -34.2 (1JPPt = 1984, 3JPP = 2 Hz) ppm, due to P(tBu)3 and PMe2ArDipp2, respectively.28 The analogous complex based on PMe2ArXyl2 (4a) revealed similar 1H and 31P{1H} NMR patterns, with characteristic 31P{1H} NMR resonances at 96.4 (1JPPt = 3140, 3JPP = 3 Hz) and -32.5 (1JPPt = 1933, 3JPP = 3 Hz) ppm. As in compound 4b, the resonance due to the terphenyl phosphine appears highly shifted to lower frequencies with respect to its gold precursor 1a (δ = -7.0 ppm). The molecular formulation of 4a was further validated by single crystal X-ray diffraction analysis (Figure 2), evincing a distorted T-shaped coordination at the platinum center. The P-Pt-P angle of 169.97(3)° and Au—Pt bond distance of 2.561(1) Å are slightly widened and shortened, respectively, compared to 4b (P-Pt-P = 167.59(5)°; dAuPt = 2.575(1) Å),28 as expected for the smaller xylyl-based terphenyl substituent on the phosphine. Figure 2. ORTEP diagram of compound 4a; for the sake of clarity hydrogen atoms and triflimide anion are excluded, while thermal ellipsoids are set at 50 % probability. Conversely, the analogous reaction based on 1c does not provide any spectroscopic sign for the formation of a new AuPt dative bond even under more forcing conditions (up to 80 °C with 100 equiv. of methanol). This contrasts with previous examples based on related [Pt(0)(PR3)]2 species, which in all cases led to the formation of metal-only Lewis pairs.29 Interestingly, while combining 1a or 1b with Pt(0) precursor 2 results in a distinct bright yellow color of the resulting solutions, the addition of 1c over benzene or chlorinated solutions of 2 did not alter the colorless appearance of the mixture, which we attribute to the absence of adduct 4c. Thus, the steric pressure exerted by the two cyclopentyl substituents in PCyp2ArXyl2 is likely responsible for quenching bimetallic Lewis adduct formation, which suggests a fully frustrated nature. These findings are in perfect agreement with our DFT computational studies (ωB97XD/6- 31G(d,p)[SDD], see Scheme 2), where the formation of the bimetallic adduct in CH2Cl2 (SMD model)30 is clearly exergonic for the systems based on PMe2ArXyl2 (ΔG = -7.2 kcal·mol-1) and PMe2ArDipp2 (ΔG = -2.4 kcal·mol-1), while for the latter the process turns to be endergonic when changing the solvent to benzene, as observed experimentally. In contrast, accessing compound 4c 5 based on PCyp2ArXyl2 is endergonic with respect to the monometallic fragments even in CH2Cl2 (10.8 kcal·mol-1) (see SI for details). Altogether, we have in hand a bimetallic pair where the M-M interaction/frustration balance can be rationally tuned from expeditious M-M dative bond formation in PMe2ArXyl2 to complete frustration in PCyp2ArXyl2, through an intermediate and adjustable equilibrium situation in PMe2ArDipp2. The effects of controlling this balance on the bimetallic reactivity and selectivity towards dihydrogen activation are discussed in the following section. Dihydrogen activation The activation of dihydrogen has continuously been examined as the prime benchmark reaction to demonstrate FLP behavior in main group systems. Similarly, we have investigated the activation of the H—H bond by TMOFLP 1b:2 to confirm its FLP-like reactivity.13¡Error! Marcador no definido. We now extend these studies to the metal-only Lewis pair 4a, as well as to the fully frustrated TMOFLP 1c:2, while providing a full mechanistic picture that highlights the relevance of M···M interactions. Although the H—H bond in dihydrogen is somehow the simplest covalent bond, we observe quite different product speciation for the three investigated Au(I)/Pt(0) systems (Table 1), despite the fact that they only differ on the substituents of the terphenyl phosphine. Entry Gold t Product speciation (Yields (%))a [Au] [Au:Pt] [Pt] 1 5 3 4 2 6 7 1 1a 5 min - - - 100 - - - 2 12 h - - 38 62 - - - 3 48 h - - 99 1 - - - 4 1b 5 min - 99 <1 - 50 50 - 5 12 h - <1 >95 - - <1 - 6 1c 5 min 89 11 - - 47 11 42 7 18 h - 95 - - 8b 14b 48b aYields were calculated by 1H and 31P{1H} NMR spectroscopy relative to each of the metals (Au or Pt) as appropriate; b Unknown Pt-containing species account for the remaining percentage of Pt in solution by 31P NMR 6 Table 1. Product speciation during FLP-like activation of H2 by 1:2 pairs It is important to note that neither complex 2 nor gold precursors 1 evolved when their C6D6 solutions were exposed to dihydrogen (1 bar) even under harsher reaction conditions (80 °C, up to 1 week) to those attempted with the metallic pairs. In stark contrast, TMOFLPs 1b:2 and 1c:2 readily react with H2 (1 bar, 25 °C), with the former pair exhibiting faster H—H cleavage. Complete consumption of 1b was observed by the time spectra were recorded rapidly after sample preparation (< 5min) to yield a 1:1:1 mixture of the hydride-bridged digold compound 5b, platinum hydride [Pt(PtBu3)2(H)]+ (6)31 and unreacted 2 (Table 1, entry 4). Low-temperature 1H and 31P{1H} NMR monitoring revealed the extreme reactivity of TMOFLP 1b:2 towards H2, since H—H cleavage is observed even at -20 °C, with a half-life for compound 1b of ca. 120 min at that temperature. The foregoing mixture evolved to the heterobimetallic dihydride 3b after 12 hours at 25 °C as the only discernible product by NMR spectroscopy (entry 5). Although proceeding at a lower rate, TMOFLP 1c:2 built on PCyp2ArXyl2 exhibited smooth reactivity towards H2 even after short reaction times, with conversion of one tenth of 1c and more than half of 2 (entry 6) by the time NMR spectra were recorded (ca. 5 min). The formation of platinum dihydride 7,32 which was not observed when combining 1b and 2, accounts for the dissimilar conversion rate of Pt(0) complex 2 compared to Au(I) species 1c. This finding suggests an unconventional catalytic role of gold for the hydrogenation of 2 towards 7.33 We demonstrated this by exposing C6D6 solutions of 2 to H2 atmosphere (0.5 bar) under variable catalytic amounts of 1c (Scheme 3). Intriguingly, best yields for the hydrogenated product (ca. 80% of 7) derived from using 10 mol% of gold catalyst 1c, whereas increasing its amount to 30% had a detrimental effect (only 17% of 7 formed) under otherwise identical conditions. This may result from a complex series of solution equilibrium processes existing after FLP-like H—H bond cleavage (see Scheme 4). For this system, the main gold containing species (ca. 95%) in the long term (18 hours) is the hydride-bridged digold compound 5c, while half of the platinum precursor was converted into dihydride 7 after that time, accompanied by smaller amounts (ca. 10 %) of unreacted 2 and monohydride 6, along with other unidentified Pt-containing species (entry 7). No signals of a presumed heterobimetallic dihydride 3c were detected, likely due steric reasons on account of PCyp2ArXyl2. Scheme 3. Catalytic hydrogenation of 2 mediated by 1c. Product speciation is considerably simplified with the less hindered system (1a:2) based on PMe2ArXyl2, which immediately yielded the metal-only Lewis pair 4a even in the presence of dihydrogen. H—H bond activation proceeds at a considerably slower pace, with full conversion to 3a reached after two days at 25 °C (entries 1-3), while no intermediates were detected by 1H or 31P{1H} NMR methods. Overall, the rate of H—H bond cleavage follows the trend 1b:2 (PMe2ArDipp2) > 1c:2 (PCyp2ArXyl2) >> 1a:2 (PMe2ArXyl2). As discussed later in deeper detail, this trend supports the notion of a genuine FLP-type H—H cleavage, since the existence of a AuPt dative bond (1a; PMe2ArXyl2) handicaps H2 activation, while forcing complete frustration by steric 7 clash (1c; PCyp2ArXyl2) somehow diminishes the rate of activation compared to the intermediate interaction/frustration situation found for the 1b:2 pair based on PMe2ArDipp2. Scheme 4. Proposed reaction pathways taking place after H-H cleavage by 1:2 pairs. As aforementioned, we assume metal speciation after H2 splitting to result from a series of equilibria that are likely controlled by steric reasons. We have not been able to spectroscopically observe a gold monohydride species A (Scheme 4) that would arise from heterolytic H—H cleavage, since it seems to be rapidly trapped by still unreacted gold triflimide 1 to yield digold hydrides 5, as it also occurs when treating compounds 1 with SiEt3H (see Experimental Section). Nevertheless, in the case of the less hindered 1a, direct access to heterobimetallic dihydride 3a is kinetically favored. The reduced size of PMe2ArXyl2 may facilitate the approach of platinum complex 6 to give the thermodynamic product, 3a. In turn, the reversible formation of compounds 5 would account for the slow formation (ca. 12 h) of 3b by trapping the continuously liberated monohydride A (from 5b) by 2. However, the more sterically congested system based on PCyp2ArXyl2 precludes the observation of 3c, likely due to steric reasons. Thus, we hypothesize an alternative pathway involving hydride abstraction by cationic platinum 6 to produce 7 and generate gold triflimide 1c. This would explain the catalytic role of gold for the hydrogenation of Pt(0) 2. Overall, the prevalence of each of the interrelated reactions seems to be controlled by steric reasons and is responsible for the product speciation observed with the three investigated terphenyl phosphines. The molecular formulation of compound 3a was ascertained by multinuclear NMR spectroscopy, where distinctive 1H NMR resonances at -1.75 (2JHP = 110 Hz, 1JHPt = 516 Hz) and -10.39 ppm, due to the bridging and terminal hydride, respectively, were recorded. Both low-frequency signals where flanked by 195Pt satellites with coupling constants of 516 and 1030 Hz, respectively. Besides, 195Pt satellites accompanying the corresponding 31P{1H} NMR signals at 91.5 (1JPPt = 8 2713 Hz, P(tBu)3) and 5.7 (2JPPt = 208 Hz, PMe2ArXyl2) evinced the presence of a Au···Pt interaction in 3a. These data compared well with its related compound 3b (δ1H = -1.67 (2JHP = 112 Hz, 1JHPt = 503 Hz), -11.39 (1JHPt = 1053 Hz); δ31P = 90.0 (1JPPt = 2744 Hz), 7.2 (2JPPt = 200 Hz)).13 With regards to compound 5b, the main gold containing species for the PMe2ArDipp2 system after short periods of time (Table 1, entry 4), we previously assumed a digold hydride-bridged formulation.13 We made this statement exclusively based on NMR data, particularly on a distinctive 1H NMR signal at 2.83 ppm that appears as a triplet (2JHP = 99.5 Hz) and was attributed to the bridging hydride. However, all attempts to isolate this compound in the past proved unsuccessful due to formation of [Au(PMe2ArDipp2)2]+ with concomitant release of hydrogen and appearance of gold nanoparticles (t1/2 at 25 °C ≈ 3 h). In stark contrast, the steric shrouding provided by PCyp2ArXyl2 allowed us isolate the first stable compound of type [Au2(µ-H)(PR3)2]+ (5c) authenticated by single-crystal X-ray diffraction (Figure 3). This compound was independently prepared by reacting 1c and SiEt3H (see Experimental Section).34 A short Au-Au distance of 2.748(1) Å suggests the existence of a strong aurophilic interaction,35 being comparable to other hydride-bridged digold complexes (2.70-2.78 Å).¡Error! Marcador no definido.b,36 In its 1H NMR spectrum, the bridging hydride appears as a triplet at 4.78 ppm and features a strong coupling with the 31P nuclei (2JHP = 90.9 Hz), comparable to 5b and to a related species based on P(tBu)2(o-biphenyl)34c and thus further supporting our prior formulation. Figure 3. ORTEP diagram of compound 5c; for the sake of clarity hydrogen atoms (except the gold hydride) and triflimide anion are excluded and cyclopentyl substituents have been represented in wireframe format, while thermal ellipsoids are set at 50 % probability. As discussed in the introduction, the mechanism of H-H bond cleavage for a variety of cooperative systems has been extensively investigated.15,16 In the context of heterobimetallic designs, three main scenarios can be envisaged. Those are represented in Scheme 5 with regard to our Au(I)/Pt(0) pairs. In the first, dihydrogen is added across the Au—Pt bond to yield two monometallic hydrides which, after rapid rearrangement, would lead to the final heterobimetallic dihydride 3 (Scheme 5a). Alternatively, the H—H bond could be cleaved by the cooperative action of the two independent metallic fragments, that is, by a truly FLP-type mechanism (Scheme 5b). As before, the resulting monohydrides could readily evolve towards 3. The third possible route involves the orthogonal reactivity of a metal fragment that initially activates the dihydrogen molecule and subsequently evolves to the final heterobimetallic compound in the presence of the second metal fragment (Scheme 5c). In principle, this mechanism may operate both in the presence or absence of a Au—Pt bond. 9 Scheme 5. Cooperative H2 activation involving (a) traditional bimetallic activation across a M- M bond; (b) transition metal-only FLP activation and (c) orthogonal reactivity. We had tentatively proposed a genuine FLP pathway as in Scheme 5b for the pair 1b:2,13 though we did not have enough data to unambiguously discern between the three alternatives described in Scheme 5. We have now investigated the potential role of these mechanisms in deeper detail by a combined experimental/computational approach. We have particularly focused on the pair 1a:2, which reacts with H2 (1 bar, 25 °C) at a considerably slower rate and thus facilitates kinetic analysis. It is important to remark that this is the only pair in which the existence of the AuPt bond is unequivocal under all experimental conditions. In fact, we observed that hydrogen activation with 1a:2 proceeds at a slower rate when using an excess of either the Au(I) or Pt(0) fragments (Figure 4), suggesting the need to access the independent Au(I) and Pt(0) fragments for the H—H cleavage to take place (as in Scheme 5b). When an excess of 0.5 equiv. of the platinum precursor 2 is added, the appearance of heterobimetallic dihydride 3a is decelerated (Figure 3, green line), while formation of trans-[Pt(PtBu3)2(H)2] (7) becomes noticeable. At variance, the latter species is absent when using a 1:1 mixture of 1a and 2 under the same conditions. Interestingly, after ca. 6 hours at room temperature the excess of Pt(0) precursor 2 is fully converted to compound 7, after which time the kinetic profile parallels the one that results from an equimolar ratio of Au(I)/Pt(0). Likewise, the kinetic profile derived from using an excess (0.5 equiv) of Au(I) evinces a detrimental effect on the H2 activation rate (orange line). These findings further support the notion of a genuine FLP mechanism which requires the initial cleavage of the AuPt bond in 4a. This would also explain why gold precursors 1b and 1c, for which no metallic Lewis adducts with [Pt(PtBu3)2] are observed under identical conditions –i. e. not involving an energetic penalty for M—M bond cleavage– do activate H2 considerably faster. 16 toluene solution (10 mL) of 1c (100 mg, 0.107 mmol) after stirring at room temperature for 30 minutes. Subsequent addition of pentane (10 mL) caused precipitation of a white solid that was washed with pentane and dried to provide compound 7c (147 mg, 87%). 1H NMR (400 MHz, C6D6, 25 °C) δ: 7.22 (m, 12 H, 3JHH = 7.6 Hz, Ha, Hb), 7.12 (td, 2 H, 3JHH = 7.6 Hz, 5JHP = 1.6 Hz, Hd), 6.66 (dd, 4 H, 3JHH = 7.6 Hz, 4JHP = 3.0 Hz, Hc), 4.78 (t, 1 H, 2JHP = 90.9 Hz, Au-H-Au), 2.36- 2.24 (m, 4 H, PCH), 2.01 (s, 24 H, MeXyl), 1.82-1.34 (m, 32 H, CH2). 13C{1H} NMR (100 MHz, C6D6, 25 °C) δ: 148.1 (d, 2JCP = 9 Hz, C3), 141.1 (d, 3JCP = 5 Hz, C2), 136.8 (C1), 136.5 (d, 3JCP = 7 Hz, CHc), 132.4 (d, 4JCP = 8 Hz, CHd), 131.9 (d, 1JCP = 40 Hz, C4), 129.1 (CHb), 128.7 (CHa), 120.4 (q, 1JCF = 323 Hz, CF3), 38.4 (d, 1JCP = 33 Hz, PCH), 35.6 (d, 2JCP = 10 Hz, CH2), 32.8 (d, 2JCP = 10 Hz, CH2), 25.4 (d, 2JCP = 10 Hz, CH2), 25.2 (d, 2JCP = 10 Hz, CH2), 21.5 (MeXyl). 31P{1H} NMR (160 MHz, C6D6, 25 °C) δ: 58.0. IR (Nujol): ν(Au-H) 1922 cm-1. Computational details. Geometry optimization of minima and transition states was carried out with the Gaussian software package.57 Optimizations were carried out without symmetry restrictions using the ωB97xD functional58 that includes empirical dispersion corrections.59 The 6-31g(d,p) basis set60 was used for non-metal atoms, Au and Pt atoms were described with the SDD basis and associated electron core potential (ECP).61 Bulk solvent effects (benzene and dichloromethane) were included during optimization with the SMD continuum model.62 Vibrational analysis was carried out on the stationary points to characterize them as minima or transition states as well as to calculate the zero-point corrections, and thermal corrections to enthalpy and free energy. Free energies were corrected (ΔGqh) to account for errors associated with the harmonic oscillator approximation. Thus, according to Truhlars’s quasi harmonic approximation, all vibrational frequencies below 100 cm-1 were set to this value so that the entropy contribution is not overestimated.63 These anharmonic corrections were calculated with the Goodvibes code.64 NBO analysis was performed with the NBO6.0 suite.65 Analysis of the electron density was performed within the Atoms In Molecules (AIM) theory of R. F. W. Bader66 using the Multiwfn program.67 The CYLview visualization software has been used to create some of the figures.68 Acknowledgements This work has been supported by the European Research Council (ERC Starting Grant, CoopCat, Project 756575) and by by the Spanish Ministry of Economy and Competitiveness (Project CTQ2016-75193-P [AEI/ FEDER, UE]). J. J. M. and M. P.-J. thank the University of Sevilla and the Ministry of Science, Innovation and Universities, respectively, for PhD fellowships. The use of computational facilities at the Supercomputing Centre of Galicia (CESGA) and the Centro de Servicios de Informática y Redes de Comunicaciones (CSIRC), Universidad de Granada are acknowledged. We are grateful to M. Roselló for valuable discussions. Keywords: Metal Frustrated Lewis Pair • Cooperative Chemistry • Hydrogen Splitting • Inverse Kinetic Isotopic Effect • Bimetallic Complexes References [1] a) D. W. Stephan and G. Erker, Angew. Chem. Int. Ed. 2015, 54, 6400 – 6441; b) D. W. Stephan and G. Erker, Top. Curr. Chem. 2013, 334, 1; c) D. W. Stephan and G. Erker, Top. Curr. Chem. 2013, 332, 1; d) D. W. Stephan and G. Erker, Angew. Chem. Int. Ed. 2010, 49, 46 – 76; e) D. W. Stephan, J. Am. Chem. Soc. 2015, 137, 10018 – 10032; f) D. W. Stephan, Science, 2016, 354, aaf7229; g) A. R. Jupp and D. W. Stephan, Trends in Chemistry 2019, 1, 35 –48. [2] D. W. Stephan, Dalton Trans. 2009, 3129 – 3136; b) D. W. Stephan, Org. Biomol. Chem. 2012, 10, 5740 – 5746; c) D. W. Stephan, Acc. Chem. Res. 2015, 48, 306 – 316. 17 [3] a) A. Y. Houghton, J. Hurmalainen, A. Mansikkamäki, W. E. Piers and H. M. Tuononen, Nat. Chem. 2014, 6, 983 – 988; b) X. Feng and H. Du, Tetrahedron Letters 2014, 55, 6959 – 6964. [4] a) M. A. Courtemanche, M. A. Légaré, L. Maron and F. G. Fontaine, J. Am. Chem. Soc. 2013, 135, 9326 – 9329; b) T. Wang and D. W. Stephan, Chem. Eur. J. 2014, 20, 3036 – 3039; c) P. Vasko, I. A. Zulkifly, M. A. Fuentes, Z. Mo, J. Hicks, P. C. J. Kamer and S. Aldridge, Chem. Eur. J. 2018, 24, 10531 – 10540; d) A. Jayaraman, L. C. M. Castro and F-G. Fontaine, Org. Process Res. Dev. 2018, 22, 1489 – 1499. [5] T. Mahdi and D. W. Stephan, Angew. Chem. Int. Ed. 2013, 52, 12418 – 12421. [6] a) Z. Mo, A. Rit, J. Campos, E. L. Kolychev and S. Aldridge, J. Am. Chem. Soc. 2016, 138, 3306 – 3309; b) G. Ma, G. Song and Z. H. Li, Chem. Eur. J. 2018, 24, 13238 – 13245. [7] See for example: a) M.-A. Légaré, M.-A. Courtemanche, E. Rochette and F.-G. Fontaine, Science 2015, 349, 513 – 516; b) M. Bakos, A. Gyömöre, A. Domján and T. Soós, Angew. Chem. 2017, 129, 5301 – 5305; c) Y. Han, S. Zhang, J. He and Y. Zhang, ACS Catal. 2018, 8, 8765 – 8773; d) J. L. Lavergne, A. Jayaraman, L. C. M. Castro, E. Rochette and F-G. Fontaine, J. Am. Chem. Soc. 2017, 139, 14714 – 14723. [8] S. R. Flynn and Wass, D. F. ACS Catal. 2013, 3, 2574 – 2581. [9] a) A. M. Chapman, M. F. Haddow and D. F. Wass, Eur. J. Inorg. Chem. 2012, 9, 1546 – 1554; b) O. J. Metters, S. J. K. Forrest, H. A. Sparkes, I. Manners and D. F. Wass, J. Am. Chem. Soc. 2016, 138, 1994 – 2003; c) A. M. Chapman, M. F. Haddow and D. F. Wass, J. Am. Chem. Soc. 2011, 133, 18463 – 18478; d) S. R. Flynn, O. J. Metters, I. Manners and D. F. Wass, Organometallics 2016, 35, 847 – 850; e) O. J. Metters, S. R. Flynn, C. K. Dowds, H. A. Sparkes, I. Manners and D. F. Wass, ACS Catal. 2016, 6, 6601 – 6611; f) H. B. Hamilton, A. M. King, H. A. Sparkes, N. E. Pridmore and D. F. Wass, Inorg. Chem. 2019, 58, 6399 – 6409. [10] a) G. Erker, Dalt. Trans. 2011, 40, 7475 – 7483; b) X. Xu, G. Kehr, C. G. Daniliuc and G. Erker, Angew. Chem. Int. Ed. 2013, 52, 13629 – 13632; c) X. Xu, G. Kehr, C. G. Daniliuc and G. Erker, J. Am. Chem. Soc. 2013, 135, 6465 – 6476; d) A. T. Normand, P. Richard, C. Balan, C. G. Daniliuc, G. Kehr, G. Erker and P. Le Gendre, Organometallics 2015, 34, 2000 – 2011; e) A. T. Normand, C. G. Daniliuc, B. Wibbeling, G. Kehr, P. Le Gendre and G. Erker, J. Am. Chem. Soc. 2015, 137, 10796 – 10808; f) Y. Hasegawa, G. Kehr, S. Ehrlich, S. Grimme, C. G. Daniliuc and G. Erker, Chem. Sci. 2014, 5, 797 – 803; g) X. Xu, G. Kehr, C. G. Daniliuc and G. Erker, Organometallics 2015, 34, 2655 – 2661; h) Z. Jian, C. G. Daniliuc, G. Kehra and G. Erker, Organometallics 2017, 36, 424−434. [11] See for example: a) Y. Jiang, O. Blacque, T. Fox and H. Berke, J. Am. Chem. Soc. 2013, 135, 7751−7760; b) A. Simonneau, R. Turrel, L. Vendier and M. Etienne, Angew. Chem. Int. Ed. 2017, 56, 12268 –12272; c) M. M. Rahman, M. D. Smith, J. A. Amaya, T. M. Makris and D. V. Peryshkov, Inorg. Chem. 2017, 56, 11798 – 11803; d) P. Cui, C. C. Comanescu and V. M. Iluc, Chem. Commun. 2015, 51, 6206 – 6209; e) K. Chang and X. Xu, Dalton Trans. 2017, 46, 4514 – 4517; f) Z. Wang, A. Ying, Z. Fan, C. Hervieu and L. Zhang, ACS Catal. 2017, 7, 3676 − 3680; g) B. R. Barnett, M. L. Neville, C. E. Moore, A. L. Rheingold and J. S. Figueroa, Angew. Chem. Int. Ed. 2017, 129, 7301 – 7305; h) S. Zhang, A. M.; Appel and R. M. Bullock, J. Am. Chem. Soc. 2017 139, 7376 – 7387; i) S. Arndt, M. Rudolph and A. S. K. Hashmi, Gold Bulletin, 2017, 50, 267 – 282; j) N. Zwettler, S. P. Walg, F. Belaj and N. C. Mösch‐Zanetti, Chemistry, 2018, 24, 7149 – 7160; k) K. Chang, X. Wang, Z. Fan and X. Xu, Inorg. Chem. 2018, 57, 8568 − 8580. [12] A. M. Chapman, S. R. Flynn and D. F. Wass, Inorg. Chem. 2016, 55, 1017 – 1021. [13] J. Campos, J. Am. Chem. Soc., 2017, 139, 2944 – 2947. 18 [14] a) T. A. Rokob and I. Ppai, Top. Curr. Chem. 2013, 332, 157−211; b) J. Paradies, Eur. J. Org. Chem. 2019, 283–294; c) L. Rocchigiani, Isr. J. Chem. 2015, 55, 134 – 149; d) L. Liu, B. Lukose, P. Jaque and B. Ensing, Green Energy Environ. 2019, 4, 20-28. [15] a) G. Skara, F. De Vleeschouwer, P. Geerlings, F. De Proft and B. Pinter, Sci. Rep. 2017, 7, 16024; b) J. Daru, I. Bak, A. Stirling and I. Ppai, ACS Catal. 2019, 9, 6049 − 6057; c) S. Grimme, H. Kruse, L. Goerigk and G. Erker, Angew. Chem., Int. Ed. 2010, 49, 1402 − 1405; d) D. Yepes, P. Jaque and I. Fernández, Chem. Eur. J., 2016, 22, 18801 – 18809; d) L. Liu, L. L. Cao, Y. Shao, G. Ménard and D. W. Stephan, Chem, 2017, 3, 259 – 267; e) H. B. Hamilton and D. F. Wass, Chem, 2017, 3, 198 – 210; f) T. A. Rokob, I. Bak, A. Stirling, A. Hamza and I. Ppai, J. Am. Chem. Soc. 2013, 135, 4425 – 4437; g) G. Bistoni, A. A. Auer and F. Neese, Chem. Eur. J. 2017, 23, 865 – 873. [16] a) R. M. Bullock and G. M. Chambers, Phil. Trans. R. Soc. A, 2017, 375, 20170002; b) M. K. Karunanan and N. P. Mankad, ACS Catal. 2017, 7, 6110 − 6119; c) E. R. M. Habraken, A. R. Jupp, M. B. Brands, M. Nieger, A. W. Ehlers and J. C. Slootweg, J. Inorg. Chem. 2019, 2436 – 2442. [17] a) N. P. Mankad, Chem. Commun. 2018, 54, 1291 – 1302; b) I. G. Powers and C. ACS Catal. 2017, 7, 936 − 958; c) R. C. Cammarota, L. J. Clouston and C. C. Lu, Coord. Chem. Rev. 2017, 334, 100 – 111; d) J. Berry and C. Lu, Inorg. Chem. 2017, 56, 7577 − 7581; (e) L. H. Gade, Angew. Chem. Int. Ed. 2000, 39, 2658 – 2678; e) N. P. Mankad, Chem. Eur. J. 2016, 22, 5822 – 5829. [18] T. A. Rokob, A. Hamza, A. Stirling and I. Pápai, J. Am. Chem. Soc. 2009, 131, 2029 − 2036. [19] F.-G. Fontaine and D. W. Stephan, Philos. Trans. R. Soc. A, 2017, 375, 20170004. [20] a) T. Mahdi and D. W. Stephan, J. Am. Chem. Soc. 2014, 136, 15809 – 15812; b) D. J. Scott, M. J. Fuchter and A. E. Ashley, J. Am. Chem. Soc. 2014, 136, 15813 – 15816; c) M. A. Légaré, E. Rochette, J. Légaré Lavergne, N. Bouchard and F. G. Fontaine, Chem. Commun. 2016, 52, 5387 – 5390. [21] J. Légaré Lavergne, A. Jayaraman, L. C. Misal Castro, E. Rochette and F.-G. Fontaine, J. Am. Chem. Soc. 2017, 139, 14714 − 14723. [22] M. V. Mane and K. Vanka, ChemCatChem, 2017, 9, 3013 – 3022. [23] See for example: a) M. K. Karunananda, S. R. Parmelee, G. W. Waldhart and N. P. Mankad, Organometallics, 2015, 34, 3857 − 3864; b) L.-J. Cheng and N. P. Mankad, J. Am. Chem. Soc. 2019, 141, 3710 – 3716; c) T. J. Mazzacano, N. J. Leon, G. W. Waldhart and N. P. Mankad, Dalton Trans. 2017, 46, 5518 – 5521; d) I. M. Riddlestone, N. A. Rajabi, J. P. Lowe, M. F. Mahon, S. A. Macgregor and M. K. Whittlesey, J. Am. Chem. Soc. 2016, 138, 11081 – 11084; e) K. M. Gramigna, D. A. Dickie, B. M. Foxman and C. M. Thomas, ACS Catalysis, 2019, 9, 3153 – 3164; f) J. Coombs, D. Perry, D.-H. Kwon, C. M. Thomas and D. E. Ess, Organometallics, 2018, 37, 4195 – 4203; g) A. M. Baranger and R. G. Bergman, J. Am. Chem. Soc. 1994, 116, 3822 – 3835; h) C. M. Thomas, J. W. Napoline, G. T. Rowe and B. M. Foxman, Chem. Commun. 2010, 46, 5790 – 5792. [24] J. Bauer, H. Braunschweig and R. D. Dewhurst, Chem. Rev. 2012, 112, 4329 - 4346. [25] M. Marín, J. J. Moreno, M. M. Alcaide, E. Álvarez, J. López-Serrano, J. Campos, M. C. Nicasio and E. Carmona, J. Organomet. Chem. 2019, 896, 120 - 128. [26] M. F. Espada, J. Campos, J. Lpez-Serrano, M. L. Poveda and E. Carmona, Angew. Chem., Int. Ed. 2015, 54, 15379 – 15384. 19 [27] L. X. Dang, G. K. Schenter, T.-M. Chang, S. M. Kathmann and T. Autrey, J. Phys. Chem. Lett. 2012, 3, 3312 − 3319. [28] N. Hidalgo, S. Bajo, J. J. Moreno, C. Navarro-Gilabert, B. Mercado, and J. Campos, Dalton Trans. 2019, 48, 9127 – 9138. [29] a) H. Braunschweig, K. Radacki and K. Schwab, Chem. Commun. 2010, 46, 913 – 915; b) J. Bauer, H. Braunschweig, P. Brenner, K. Kraft, K. Radacki and K. Schwab, Chem. - Eur. J. 2010, 16, 11985 – 11992; c) J. Bauer, H. Braunschweig, A. Damme and K. Radacki, Angew. Chem., Int. Ed. 2012, 51, 10030 – 10033; d) M. Ma, A. Sidiropoulos, L. Ralte, A. Stasch and C. Jones, Chem. Commun. 2013, 49, 48 – 50; e) H. Braunschweig, R. D. Dewhurst, F. Hupp and C. Schneider, Chem. Commun. 2014, 50, 15685 – 15688; f) B. R. Barnett, C. E. Moore, P. Chandrasekaran, S. Sproules, A. L. Rheingold, S. DeBeerde and J. S. Figueroa, Chem. Sci., 2015, 6, 7169 – 7178. [30] A. V. Marenich, C. J. Cramer and D. G. Truhlar, J. Phys. Chem. B 2009, 113, 6378 − 6396. [31] R. G. Goel and R. C. Srivastava, Can. J. Chem. 1983, 61, 1352 – 1359. [32] R. G. Goel, W. O. Ogiri and R. C. Srivastava, Organometallics, 1982, 1, 819 – 824. [33] a) C. González-Arellano, A. Corma, M. Iglesias and F. Sánchez, Chem. Commun. 2005, 3451 – 3453; b) A. Comas-Vives and G. Ujaque, J. Am. Chem. Soc. 2013, 135, 1295 − 1305. [34] a) E. Y. Tsui, P. Muller and J. P. Sadighi, Angew. Chem., Int. Ed. 2008, 47, 8937 – 8940; b) A. Escalle, G. Mora, F. Gagosz, N. Mezailles, X. F Le Goff, Y. Jean and P. Le Floch, Inorg. Chem. 2009, 48, 8415 – 8422; c) R. J. Harris and R. A. Widenhoefer, Angew. Chem., Int. Ed. 2014, 53, 9369 – 9369. [35] H. Schmidbaurab and A. Schiera, Chem. Soc. Rev. 2012, 41, 370 – 412. [36] N. Phillips, T. Dodson, R. Tirfoin, J. I. Bates and S. Aldridge, Chem. -Eur. J. 2014, 20, 16721 – 16731. [37] S. Tussing, L. Greb, S. Tamke, B. Schirmer, C. Muhle-Goll, B. Luy and J. Paradies, Chem. -Eur. J. 2015, 21, 8056 – 8059. [38] A. J. Esswein, A. S. Veige, P. M. B. Piccoli, A. J. Schultz and D. G. Nocera, Organometallics, 2008, 27, 1073 – 1083. [39] Y. Zhang, M. K. Karunananda, H.-C. Yu, K. J. Clark, W. Williams, N. P. Mankad and D. H. Ess, ACS Catal. 2019, 9, 2657 − 2663. [40] Computational studies at the ωB97xD/6-31g(d,p)+SDD level. Dispersion effects are included in the functional (D2 version of empirical dispersion by Grimme). Bulk solvent effects accounted by the SMD model (benzene and dichloromethane). Energies reported in the text are free energies with Truhlars’s anharmonic corrections. The calculations have been performed with the Gaussian09 package. Gaussian 09, Revision B.01. Gaussian, Inc., Wallingford CT, 2010. See the Suplementary information for further details. [41] Alternatively (see the Supporting information), the cis-Pt(II) dihydride forms a dimer with the Au fragment, [Au](μ-H)2[Pt], that can evolve to two monometallic hydrides. Whereas formation of the bridging complex is easy (ΔG⁰ = 20.0 kcal·mol-1), its transformation to the following intermediate featuring a dihydrogen bond, [Au]-H···H-[Pt] (A), has an overall barrier ΔG‡ = 30.7 kcal·mol-1. 20 [42] The calculations yield a geometry for 4a with a different conformation of the terphenyl phosphine that is 9.9 kcal·mol-1 above than the most stable conformation used as the origin of energies. See M. Marín, J. J. Moreno, C. Navarro-Gilabert, E. Álvarez, C. Maya, R. Peloso, M. C. Nicasio, E. Carmona, Chem. Eur. J. 2019, 25, 260 – 272. [43] R. F. W. Bader, Atoms in Molecules: A Quantum Theory; Oxford University Press: Oxford, UK, 1995. [44] L. Könczöl, E. Eszter Makkos, D. Bourissou and S. Szieberth, Angew. Chem. Int. Ed. 2012, 51, 9521 –9524. [45] a) N. J. Gunsalus, S. H. Park, B. G. Hashiguchi, A. Koppaka, S. J. Smith, D. H. Ess and R. A. Periana, Organometallics 2019, 38, 2319 – 2322; b) S. Das, S. Mondal and S. K. Pati, Chem. Eur. J. 2018, 24, 2575 – 2579. [46] We considered the role of NTf2- as a base in H2 activation. As expected, H—H cleavage between 1a(H2)+ and NTf2- was found to occur through a high energy TS lying 30.0 kcal·mol-1 above the reactants, 4a + H2 (see Figure S4). [47] J. J. Cabrera-Trujillo and I. Fernndez, Inorg. Chem. 2019, 58, 7828 − 7836. [48] While a correct analysis of KIEs requires full statistical mechanical treatment, it has been showed that at least for some cases KIEs and EIEs are controlled by ZPE differences: a) D. G. Churchill, K. E. Janak, J. S. Wittenberg and G. Parkin, J. Am. Chem. Soc. 2003, 125, 1403 –1420; b) F. Abu-Hasanayn, K. Krogh-Jespersen and A. S. Goldman, J. Am. Chem. Soc. 1993, 115, 8019 – 8023; c) F. Abu-Hasanayn, A. S. Goldman and K. Krogh-Jespersen, J. Phys. Chem. 1993, 97, 5890 − 5896; d) G. Parkin, Acc. Chem. Res. 2009, 42, 315 − 325; e) B. R. Bender, G. J. Kubas, L. H. Jones, B. I. Swanson, J. Eckert, K. B. Capps and C. D. Hoff, J. Am. Chem. Soc. 1997, 119, 9179 − 9190. [49] Focusing on ZPE changes, the bond strength (D) can be approximated as: 𝐷 = 𝐸 − 1 2 ⁄ℎ𝜈, where E is the potential energy and ν is the vibrational frequency, which equals (1 2𝜋 ⁄)√𝑘/𝜇. In this equation k is the force constant and μ(H/D) is the reduced mass of the E—H(D) bond. It can be assumed that E and k do not change in each set of isotopomers (e.g. H2 v D2). Therefore, the difference in zero-point energy for a pair of isotopomers (reactants, products or TSs) is: ∆𝑍𝑃𝐸~𝐷(𝐷)− 𝐷(𝐻)= 1 2 ⁄ℎ[𝜈(𝐷)− 𝜈(𝐻)]. i.e., the isotope effect depends on the differences in the vibrational frequencies of the isotopomers. The previous equations can also be written in terms force constants and reduced mass as: ∆𝑍𝑃𝐸 ∝ 𝑘1 2 ⁄[𝜇(𝐷)−1 2 ⁄− 𝜇(𝐻)−1 2 ⁄]. Thus, higher force constants at the TS than at the reactants are major contributors to an inverse primary KIE. In our case the difference in zero-point energy between the TS and the reactants can also be written in terms of vibrational frequencies: ∆∆𝑍𝑃𝐸 ∝ 1 2 ⁄ℎ[{∑𝜈𝐷 𝑇𝑆 − 𝜈𝑅(𝐷 − 𝐷)}−{∑𝜈𝐻 𝑇𝑆 − 𝜈𝑅(𝐻 − 𝐻}], where ∑𝜈𝑇𝑆represent the collective isotopically sensitive vibrational modes at the TS. [50] a) T.-Y. Cheng and R. M. Bullock, J. Am. Chem. Soc. 1999, 121, 3150 – 3155; b) H. C. Lo, A. Haskel, M. Kapon and E. Keinan, J. Am. Chem. Soc. 2002, 124, 3226 – 3228. [51] J. Bigeleisen, J. Pure. Appl. Chem. 1964, 8, 217 – 223. [52] M. Gómez-Gállego and M. A. Sierra, Chem. Rev. 2011, 111, 4857 –4963. [53] For some selected examples see: a) R. A. Periana and R. G. Bergman, J. Am. Chem. Soc. 1986, 108, 7332 – 7346 and reference 14 therein; b) D. G. Churchill, K. E. Janak, J. S. Wittenberg and G. Parkin, J. Am. Chem. Soc. 2003, 125, 1403 – 1420; c) T. O. Northcutt, D. D. Wick, A. J. Vetter and W. D. Jones, J. Am. Chem. Soc. 2001, 123, 7257 – 7270. 21 [54] L. Ortega-Moreno, M. Fernández-Espada, J. J. Moreno, C. Navarro, J. Campos, S. Conejero, J. López-Serrano, C. Maya, R. Peloso and E. Carmona, Polyhedron, 2016, 116, 170 – 181. [55] R. Uson, A. Laguna, M. Laguna, D. A. Briggs, H. H. Murray and J. P. Fackler, Inorg. Synth. 2007, 26, 85 – 91. [56] H. R. C. Jaw, W. R. Mason, Inorg. Chem. 1989, 28, 4370 – 4373. [57] Gaussian 09, Revision B.01, M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, T. Keith, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, and D. J. Fox, Gaussian, Inc., Wallingford CT, 2010. [58] J.-D. Chai and M. Head-Gordon, Phys. Chem. Chem. Phys. 2008, 10, 6615– 6620. [59] S. Grimme, J. Comp. Chem. 2006, 27, 1787-1799. [60] a) R. Ditchfield, W. J. Hehre and J. A. Pople. J. Chem. Phys. 1971, 54, 724 – 728; b) W. J. Hehre, R. Ditchfield and J. A. Pople. J. Chem. Phys. 1972, 56, 2257 – 2261; c) P. C. Hariharan and J. A. Pople. Theor. Chim. Acta 1973, 28, 213 – 222; d) M. M. Francl, W. J. Pietro, W. J. Hehre, J. S. Binkley, M. S. Gordon, D. J. DeFrees and J. A. Pople. J. Chem. Phys. 1982, 77, 3654 – 3665. [61] D. Andrae, U. Haeussermann, M. Dolg, H. Stoll and H. Preuss. Theor. Chim. Acta 1990, 77, 123 – 141. [62] A. V. Marenich, C. J. Cramer and D. G. Truhlar. J. Phys. Chem. B 2009, 113, 6378 – 6396. [63] R. F. Ribeiro, A. V. Marenich, C. J. Cramer and D. G. Truhlar, J. Phys. Chem. B 2011, 115, 14556 – 14562 [64] I. Funes-Ardoiz, R. S. Paton, (2016). Goodvibes: Goodvibes 2.0.2 http://doi.org/10.5281/zenodo.595246 [65] NBO 6.0. E. D. Glendening, J. K. Badenhoop, A. E. Reed, J. E. Carpenter, J. A. Bohmann, C. M. Morales, C. R. Landis, and F. Weinhold (Theoretical Chemistry Institute, University of Wisconsin, Madison, WI, 2013); http://nbo6.chem.wisc.edu/ [66] R. F. W. Bader, Atom in Molecules: A Quantum Theory; Oxford University Press: Oxford, UK, 1995. [67] a) Multiwfn: A multifunctional wavefunction analyser. T. Lu, F. Chen, J. Comput. Chem., 2012, 33, 580-592; b) Multiwfn 3.6, http://sobereva.com/multiwfn/ [68] CYLview, 1.0b; Legault, C. Y., Université de Sherbrooke, 2009 (http://www.cylview.org)