Nucleophilic Nickel and Palladium Pincer Hydroxides: A Study of Their Reactions with Dimethyl Carbonate and Other Non-Alkylating Organic Electrophiles

Nucleophilic Nickel and Palladium Pincer Hydroxides: A Study of Their Reactions with Dimethyl Carbonate and Other Non-Alkylating Organic Electrophiles Luis M. Martínez-Prieto,[a, b] Diego del Río,[a] Eleuterio Álvarez,[a] Pilar Palma,[a] and Juan Cámpora*[a] The reactions of the pincer hydroxide complexes [(iPrPCP)M(OH)] (M=Ni, Pd) with dimethyl carbonate (DMC), and a set of organic electrophiles including benzaldehyde have been investigated in the context of our ongoing investigation on the synthesis of alkyl carbonates from CO2and alcohols. The final outcome of such reactions is diverse, but for PhCHO and DMC the first step is a mechanistically similar addition of the [M]OH linkage across the carbonyl functionality, that leads to unstable adducts. DMC is cleaved irreversibly by both Ni and Pd hydroxides, affording the corresponding methylcarbonates [(iPrPCP)MOCO2Me] and methanol, whereas PhCHO affords benzoate complexes [[(iPrPCP)MOCO2Ph]. The main kinetic and thermodynamic features of these reactions were reproduced satisfactorily by computational DFT models. The calculations throw light on the true causes of irreversibility of DMC cleavage by nucleophilic hydroxide complexes, which is the primary cause that prevents methanol carboxylation from being catalysed by the Ni or Pd pincers. Introduction Pincer ligands and their transition metal complexes offer a privileged platform for the study of the chemical reactivity of complexes that are usually hard to prepare and manipulate.[1,2] Rigid pincers are perfectly adapted to the square planar geometry of Group 10 M(II) ions, to which they confer unusual robustness. For example, hydride complexes[3] with pincer ligands may combine MH and MC bonds, and yet they are very stable compounds. On the other hand, the fixed geometry imposed by pincer scaffolds imply that, for square-planar complexes, there is only one possible coordination site available for useful transformations. Though at first glance this might pose a severe restriction, rigid scaffolds like the well-known RPCP ligands (see Scheme 1 for R=iPr) enhance the chemical reactivity at the vacant site due to the strong trans labilizing effect of the central MC(pincer) bond.[4] It has been known for a long time that strongly basic ligands such as hydroxo, alkoxo or amido become highly reactive in the coordination sphere of late transition metals, due to the lack of suitable empty orbitals to accept π-electron donation from the exceedingly basic electron pairs sitting on the heteroatom.[5] Species containing such reactive MOH, MOR or MNR2linkages are key intermediates in various catalytic processes, ranging from commodity industrial processes (like the Wacker reaction)[6] to biocatalysis, e.g. in metalloenzymes like carbonic anhydrase or alcohol deshydrogenase.[7] Furthermore, the nucleophilicity of covalent metal hydroxides played an important role in the historic development Coordination Chemistry.[8] In contrast with the usually well-defined nature of MC bonds, the chemistry of MOH or MOR ensembles is often complicated by their [a] Dr. L. M. Martínez-Prieto, Dr. D. del Río, Dr. E. Álvarez, Dr. P. Palma, Prof. J. Cámpora Instituto de Investigaciones Químicas, CSIC-Universidad de Sevilla, C/Américo Vespucio, 49. 41092, Seville, Spain E-mail: [email protected] [b] Dr. L. M. Martínez-Prieto Instituto de Tecnología Química CSIC-Universidad Politécnica de Valencia, Avda. de Los Naranjos, s/n. 46022, Valencia, Spain Supporting information for this article is available on the WWW under https://doi.org/10.1002/ejic.202100400 Part of the “RSEQ-GEQO Prize Winners” Special Collection. © 2021 The Authors. European Journal of Inorganic Chemistry published by Wiley-VCH GmbH. This is an open access article under the terms of the Creative Commons Attribution Non-Commercial NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made. Scheme 1. Hypothetic mechanism for a catalytic alcohol carboxylation process. Full Papers doi.org/10.1002/ejic.202100400 2958Eur. J. Inorg. Chem. 2021, 2958–2975 © 2021 The Authors. European Journal of Inorganic Chemistry published by Wiley-VCH GmbH Wiley VCH Freitag, 30.07.2021 2129 / 211589 [S. 2958/2975] 1 10990682c, 2021, 29, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/ejic.202100400 by Universidad De Sevilla, Wiley Online Library on [14/10/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License tendency to cancel its basicity forming multinuclear aggregates, to experience irreversible hydrogen elimination processes, or simply by their facile hydrolysis with the slightest moisture traces. Pincer scaffolds offer unique opportunities to overcome these difficulties, opening new avenues to explore the full potential of terminal M-heteroatom linkages, combining classic nucleophilic/basic reactivity patterns with the unique ability of transition metals for the activation of unreactive bonds and small molecules.[9] One of the long-standing research topics in our group is the analogy between the chemical reactivity of metal-carbon bonds with similar metal-heteroatom bonds in monomeric Ni(II) and Pd(II) complexes with hydroxo, alkoxo or other similar nondative heteroatom ligands.[10] In our experience, the iPrPCP scaffold, which contains symmetrical CH2P(iPr2) arms, has an adequate balance of steric shielding to promote interesting chemical reactivity while allowing enough stability for isolation purposes. The iPrPCP pincer has other practical features that facilitate our task, like the stereochemical handle provided by the isopropyl substituents at phosphorus, or convenient monitoring of chemical reactions using the simple single-line 31P{1H} NMR spectra of most of their complexes. We found that hydroxide derivatives [(iPrPCP)M(OH)] (M=Ni or Pd) are excellent starting materials, easy to synthesize and handle,[11] and, although the isolation of pure samples of the corresponding alkoxides [(iPrPCP)M(OR)] is challenging,[12,13] their reactions can be readily monitored in situ, facilitating the investigation of important processes like β-hydrogen elimination from the alkoxide.[14] Our studies confirmed that hydrolysis of the MOR bonds is fully reversible. Accordingly, the hydroxides equilibrate with the corresponding alkoxides in the presence of wateralcohol mixtures.[13] Both the hydroxides and alkoxides are efficient nucleophiles. Their reactions with CO2bring a typical example and an opportunity for potential applications. These reactions are extremely fast and complete, yielding the corresponding bicarbonates [(iPrPCP)M(OCO2H)] and alkylcarbonates [(iPrPCP)M(OCO2R)], respectively. Even though CO2insertion is thermodynamically favorable, the products are kinetically labile, leading to partial decarboxylation to binuclear carbonates [{(iPrPCP)M}2(μ-CO3)], as soon as their solutions are manipulated.[15] However, the same chemical lability of the [M]OCO2linkage enables immediate reversion of the equilibrium mixtures if CO2and water or ROH are restored to the system. Furthermore, alkylcarbonates [(iPrPCP)M(OCO2R)] are thermodynamically favored under equilibrium conditions, in which they are always the prevalent species as long as enough ROH and CO2are available, even in the presence of substantial amounts of water. This led us to investigate the possible application of the pincer hydroxides and alkoxides to promote the synthesis of organic carbonates through direct alcohol carboxylation, as shown in Eq. 1 and Scheme 1.[15b,16] (1) Alcohol carboxylation has attracted much interest as a potential method for CO2fixation and valorization in the form of organic carbonates.[17] Organic carbonates (like dimethyl carbonate, DMC[18]) or polycarbonates are produced in large amounts from toxic or costly intermediates like phosgene or, more recently, epoxides.[17d–e,19] Therefore, direct carboxylation of alcohols has an enormous potential as a “green sink” to capture CO2into a variety of valuable chemicals and polymeric materials. Unfortunately, alcohols fail to react spontaneously with CO2, for both kinetic and thermodynamic reasons.[17a,20] For example, the synthesis of DMC from CO2and methanol is exothermic and only slightly endergonic at room temperature. However, in contrast with water, methanol does not react spontaneously with CO2to any extent. A number of metal alkoxides, including niobium methoxide or diorganotin methoxides, have been shown to catalyze the carboxylation of methanol, but in general their efficiency is rather low.[17b] After many attempts, we found that pincer Ni and Pd hydroxide or alkoxide complexes fail to promote the synthesis of DMC from methanol and CO2under various experimental conditions, ranging from room temperature NMR-tube studies to high pressure and temperature in neat MeOH or supercritical CO2.[15b,16] Catalytic carboxylation, even as a disfavored equilibrium, was definitively ruled out using isotope labeling experiments. Since processes (A) to (D), shown in the right side of Scheme 1, were demonstrated to be all facile and fully reversible, it follows that the pitfall that prevents the closing of the catalytic cycle and fixation CO2as DMC must lie in the final step (E) shown in the left side of the Scheme with a dashed arrow. To throw some light on the ultimate causes that prevent the catalytic synthesis of DMC as shown in Scheme 1, we decided to investigate the process (E) in the reverse direction, namely, studying the cleavage of DMC by nickel and palladium hydroxides. The reactions of organic esters with metal hydroxides and alkoxides are classic subjects of Organic Chemistry.[21] These are involved in such important transformations as ester hydrolysis or transesterification. In recent years, the use of metal alkoxides with well-defined coordination environments has improved the control over some of these reactions, e.g. the ring opening polymerization of cyclic esters.[22] However, in spite of the significance of metal hydroxides in the organic and bio-organic context, studies on the reactivity of well-defined pincer hydroxide complexes have been focused more on the activation of small electrophilic molecules like CO, CO2or dihydrogen,[11,15,23–27] than on typical carbonyl electrophiles like ketones, aldehydes or esters. As a part of studies on reduction of carbonyl compounds with pincer hydride species, Guan has reported relevant mechanistic data for some of the abovementioned reactions,[28] but, as far as we know, no study has been specifically devoted to investigate the interaction of welldefined pincer hydroxides with conventional carbonyl electrophiles. Therefore, to place our work in a broader context, we decided to complete our study on the nucleophilic cleavage of DMC by nickel and palladium pincer hydroxides [(iPrPCP)MOH], by examining their ability to react with some other typical organic electrophiles: benzaldehyde, ketones (acetone, acetophenone) and isocyanides. In addition, we modelled the mechanisms of the reactions with benzaldehyde and DMC by Full Papers doi.org/10.1002/ejic.202100400 2959Eur. J. Inorg. Chem. 2021, 2958–2975 www.eurjic.org © 2021 The Authors. European Journal of Inorganic Chemistry published by Wiley-VCH GmbH Wiley VCH Freitag, 30.07.2021 2129 / 211589 [S. 2959/2975] 1 10990682c, 2021, 29, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/ejic.202100400 by Universidad De Sevilla, Wiley Online Library on [14/10/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License means of DFT calculations, which we describe in detail in the last part of this contribution. Results and Discussion Reactions of nickel and palladium hydroxides [(iPrPCP)M(OH)] with several organic electrophiles Prior to undertaking our study of the DMC cleavage by pincer Ni and Pd hydroxides [(iPrPCP)M(OH)], we explored their reactions with a few selected electrophiles. The results are summarized in Scheme 2. From this point on, pincer-metal fragments are represented as [M] (with M=Ni or Pd) whenever it is advised for the sake of clarity. The reactions of the pincer hydroxide complexes with nearstoichiometric amounts of enolizable methylketones are controlled by the weak acidity of the α-methyl group, giving rise to the corresponding enolate complexes. This reaction is wellprecedented and there are several reports in the literature describing “neutralization” reactions of basic group 10 organometallic hydroxides with weakly acidic carbonyl compounds.[29,30] Furthermore, it is known that Ni(II) frequently favors the k-Ocoordination of enolate ligands, whereas Pd leads to k-C type coordination,[31,32] in line with the lower oxophilicity of the latter.[32b] Accordingly, the reaction of [(iPrPCP) Ni(OH)] with equimolar amounts of acetophenone in C6D6is incomplete and gives rise to equilibrium mixtures of the hydroxide and the corresponding O-enolate, unambiguously identified in the solution NMR spectra of the reaction mixtures on the basis of its characteristic 1H and 13C signals of the vinyl enolate fragment. The apparent equilibrium constant, Keq �2× 102, is very similar to that measured for the analogous reaction of acetophenone with the monomeric hydroxide complex [Ni(Me)(OH)(dippe)].[29a] Removing water with molecular sieves shifts the equilibrium to the enolate side until the hydroxide/ enolate ratio reaches 3:7, but the O-enolate product was reluctant to crystallize and could not be isolated pure (see Experimental Section). In contrast, [(iPrPCP)Pd(OH)] reacts clean and irreversibly with acetone, giving rise to the corresponding C-enolate (or ketonyl[32]) complex [(iPrPCP)Pd(CH2COCH3)], which was readily isolated as a stable white solid.[30] The Pd hydroxide reacts in THF solution with benzaldehyde which, lacking acidic hydrogens, can only undergo nucleophilic attack at the carbonyl functionality. 31P NMR monitoring reveals the formation of a main product characterized by a singlet resonance at 60.4 ppm. Full conversion requires an excess of reagent (up to 3:1 ratio). A sample of the main product, contaminated with a small amount of PhCHO, was isolated after solvent evaporation under vacuum (see experimental) which led to its identification as the benzoate [(iPrPCP)PdOCOPh]. The nickel hydroxide behaves similarly but the reaction with PhCHO is sluggish, requiring using neat PhCHO (reagent ratio �100:1) to complete the transformation within a reasonable time frame. However, the 31P{1H} NMR spectrum of the reaction mixture suggests that the reaction proceeds cleanly to yield a single product (singlet at δ57.7 ppm). In order to confirm the identity of the Ni and Pd benzoate complexes, their spectral properties were compared with those of authentic samples, independently synthesized by reacting equimolar amounts of the corresponding hydroxides with benzoic acid. For the sake of completeness, the reactions of the more reactive palladium hydroxide with pnitrobenzaldehyde and p-dimethylaminobenzaldehyde were carried out in NMR tubes, and the corresponding p-substituted benzoate complexes were prepared to confirm the identity of the products (see below). In a previous contribution, we have shown[11] that the reaction of Ni and Pd hydroxides [(iPrPCP)M(OH)] with CO leads to unstable hydroxycarbonyl complexes, [M]COOH. These evolve in solution, undergoing partial decarbonylation to yield isolable CO2-bridged species [M]COO[M].The mechanism of CO insertion into NiOH bonds has attracted much interest because it is involved in some catalytic processes and has also relevance in biological methanogenesis.[10] Further examples of carbonylation of Ni or Pd pincer hydroxide complexes were reported later.[23a,27] As in the case of CO2(see Scheme 1), the formation of [M]COO[M] from [M]COOH implies some degree of reversibility of CO insertion. DFT calculations suggest a migratory-type insertion mechanism (i.e., CO interacts first with the metal, and then the hydroxide “migrates” to the carbonyl C), albeit assisted by the nonbonding electron pairs localized on the hydroxyl O atom.[33] Thus, it is intriguing that the formally analogous reactions of well-defined hydroxide complexes with isocyanides (which are isoelectronic with CO), remain almost unknown.[34] To complete this study, we briefly examined the reaction of [(iPrPCP)M(OH)] hydroxides with t-BuNC. The reaction is fast and selective in both cases, yielding stable products that were isolated as yellow (M=Ni) or white (M=Pd) powdery solids, as shown in Scheme 2. The presence of a carbamoyl functionality (i.e., [M]C(=O)NHt-Bu), was deduced from their IR and solution NMR spectra. Strong IR absorptions at 1550 cm1and a low field 13C resonances (at 210 and 205 ppm for Ni and Pd, respectively, with characteristic triplet splitting by coupling with two equivalent 31P nuclei) demonstrate the presence of a C(=E)E’H fragment (E and E’ could be either O or N-tBu), carbonScheme 2. Reactions of the Ni and Pd pincer hydroxides with carbonyl-based electrophiles and isocyanides (here, [M] represents the pincer fragment [(iPrPCP)M)] with M=Ni or Pd, except where the metal is specified). Full Papers doi.org/10.1002/ejic.202100400 2960Eur. J. Inorg. Chem. 2021, 2958–2975 www.eurjic.org © 2021 The Authors. European Journal of Inorganic Chemistry published by Wiley-VCH GmbH Wiley VCH Freitag, 30.07.2021 2129 / 211589 [S. 2960/2975] 1 10990682c, 2021, 29, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/ejic.202100400 by Universidad De Sevilla, Wiley Online Library on [14/10/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License connected to the metal. In both products, the 1H signal corresponding to the NH was located in the proximity of 5.0 ppm. That the hydroxyl hydrogen atom migrates to the more basic nitrogen was deduced from the observation of NOE cross-peaks that correlate the latter signals with that of the t-Bu group (a singlet at ca. 1.5 ppm) in the 2D NOESY spectra of both products. Similar Pd and Pt carbamoyl complexes have been synthesized in a more straightforward manner by carbonylation of amido complexes.[35] Among the electrophiles whose reactions with the Ni and Pd hydroxides have been presented in this section, PhCHO is the one that bears a closer resemblance to organic carbonates. This is also a classic substrate for the investigation of nucleophilic attack on organic carbonyl functionalities. Thus, we examined more closely this reaction, focusing on the more reactive palladium hydroxide. The formation of the benzoate product [(iPrPCP)PdOC(O)Ph] from [(iPrPCP)PdOH] and PhCHO implies the formal loss of two hydrogen atoms. These could be either taken up by some hydrogen acceptor, or released free as H2(Equation 2 and Equation 3). (2) (3) Eq. 2 is directly related to the well-known Cannizaro reaction, whereby non-enolizable aldehydes react with aqueous alkali hydroxides to afford 1:1 mixtures of carboxylate and alcohol. In this reaction, the hydrogen acceptor is a second equivalent of the aldehyde, which ends up as benzyl alcohol. Indeed, GC analysis of the crude reaction mixtures of the Pd hydroxide with PhCHO reveals the presence of benzyl alcohol, but the amount is far too small to account for the equimolar ratio expected according to Eq. 2. This suggests that most of the remaining hydrogen could be escaping from the system as H2, as shown in Eq. 3. Figure 1 shows the course of the reaction of [(iPrPCP)PdOH] (0.05 M in THF) with 3 equiv. of PhCHO at 25°C, simultaneously monitored by GC and 31P{1H} NMR over a period of 24 h. The last measurement showed that about 90% of the starting hydroxide had been consumed, and 75% of the total 31P intensity corresponded to the signal of the benzoate complex. In contrast, benzyl alcohol (determined by GC) amounted to just 17% with regard to the starting hydroxide. In addition to the 31P resonances of the starting material [(iPrPCP) Pd(OH)] and the product [(iPrPCP)Pd(OCOPh)] (at ca. 59 and 61 ppm, respectively), only a low-intensity signal was observed at 71.7 ppm, whose chemical shift coincides with the characteristic low field-shifted signal of the pincer hydride [(iPrPCP) PdH].[36] The formation of detectable amounts of the Pd hydride complex is mechanistically significant. In contrast with the hydroxide and the benzoate complexes, which exhibit monotonous decay and rising trends, respectively, the concentration of the hydride reaches a maximum after 5 h (8%) and then stays quasi-stationary, decaying very slowly as the reaction advances towards completion in the following hours. This behavior suggests that the hydride complex is a reaction intermediate. From the fact that the hydride ligand can only originate from the aldehyde functionality, we deduce that the attack of the palladium hydroxide leads to an unstable insertion complex, most likely a classic “tetrahedral intermediate”,[37] [Pd]OC- (H)(Ph)(OH)], which then decomposes by β-H elimination, as proposed in Scheme 3. We have demonstrated previously that, despite the strong stabilizing effect of PCP pincer ligands, Ni and Pd methoxides [(iPrPCP)M(OMe)] undergo reversible β-H elimination above the room temperature, affording detectable amounts of the corresponding hydride, which slowly decomposes to afford M(0) species.[14] As shown also in Scheme 3, two parallel processes compete in trapping the palladium hydride. The main one is the direct reaction of the hydride with benzoic acid formed in the βelimination step. This route leads directly to the final benzoate complex, plus H2,i.e., it is responsible for the acceptorless stoichiometry described in Eq. 3. In line with this hypotheses, the NMR spectra of mixtures of [(iPrPCP)Pd(OH)] with substituted aldehydes p-ZC6H4CHO (Z=NMe2or NO2) showed visible differences in the intensity of the characteristic 31P signal of the hydride intermediate, that correlate with the strength of the Figure 1. Concentration profiles of the key species involved in the reaction of [(iPrPCP)PdOH] (0.05 M in THF) and PhCHO (3-fold excess) at room temperature, as determined using 31P NMR and GC. Scheme 3. Proposed mechanism for the reaction of [(iPrPCP)Pd(OH)] with PhCHO. Full Papers doi.org/10.1002/ejic.202100400 2961Eur. J. Inorg. Chem. 2021, 2958–2975 www.eurjic.org © 2021 The Authors. European Journal of Inorganic Chemistry published by Wiley-VCH GmbH Wiley VCH Freitag, 30.07.2021 2129 / 211589 [S. 2961/2975] 1 10990682c, 2021, 29, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/ejic.202100400 by Universidad De Sevilla, Wiley Online Library on [14/10/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License corresponding carboxylic acid. For Z=NMe2the weakness of the acid allowed the buildup of higher concentrations of the hydride, whereas with Z=NO2, its 31P signal was so weak that is hardly detectable. The benzyl alcohol detected by GC in the Pd reaction, characteristic of the Cannizaro-type stoichiometry (eq. 2), originates in the second pathway, that involves the capture of [Pd]H by benzaldehyde, followed by reaction of the resulting benzoxide with benzoic acid. Similar aldehyde hydrometallations play an important role in reactions catalyzed by pincer-stabilized hydrides.[28] Since the 31P signal for the resulting benzoxide intermediate [(iPrPCP)Pd(OCH2Ph)] was not observed (this compound is still unreported, but its 31P signal is expected to occur within a region 2–4 ppm upfield of the signal of the starting hydroxide[13]), it can be assumed that the subsequent neutralization of the benzoxide and benzoic acid outcompetes the hydrometallation reaction. In addition, the persistence of some [(iPrPCP)Pd(H)] at the end of the reaction suggests that some of the benzoic acid formed in the main process could be trapped also by the starting hydroxide, as suggested in Scheme 3. Direct evidence for this mechanism was provided by the identification of the unstable tetrahedral intermediate in the NMR spectra of mixtures of [(iPrPCP)Pd(OH)] and PhCHO at subambient temperature (see SI). When the 31P{1H} spectrum of a freshly prepared 0.1 M solution of the palladium hydroxide and PhCHO (1:1) in toluene-d8, was recorded at 30°C, it showed an additional resonance at 56.1 ppm that was not observed at room temperature. The new signal, with intensity 1/2.3 of that of the starting hydroxide, is only 0.6 ppm apart from the latter. The spectrum is quite clean since, other than these two resonances, only the incipient signals of the hydride and benzoate complexes were detected, at 71.7 and 60.3 ppm, respectively. The 30°C1H NMR spectrum (Figure S1) confirmed that the major P-containing species is still the hydroxide, unambiguously revealed by its characteristic high-field OH signal at 1.23 ppm. Two small resonances corresponding to a partially resolved AX spin system were observed in an otherwise clean region, at 6.36 (sharp doublet, 3JHH =7.5 Hz) and 5.35 ppm (broad). These are assigned to the methyne and OH protons, respectively, of the hemiacetal functionality (OCH(OH)) of the hydroxide-benzaldehyde adduct. This phenomenon is reversible and the signals of the labile adduct disappeared when the spectrum was recorded again at room temperature. Repeating the experiment with a three-fold excess of PhCHO causes the signals of the adduct to become of comparable intensity to those of the hydroxide complex, which enabled us to gather some key 13C data. For example, the 13C resonance corresponding to the hemiacetal methyne was located at 96.3 ppm in the 2D 1H-13C HSQC spectrum. The 1JCH coupling (153 Hz) is compatible with a hemiacetal-type sp3carbon. As expected, both 1H resonances at 6.36 and 5.35 ppm were further broadened by chemical exchange. The 2D phase-sensitive NOESY/EXSY spectrum (Figure S2) shows that each signal of free PhCHO is connected to another one in the adduct through exchange crosspeaks. Most evident is the characteristic aldehyde signal at 9.57 ppm with the acetal methyne at 6.36 ppm. In addition, the assignment of the broad 5.35 ppm signal to the hemiacetal hydroxyl was confirmed by an exchange crosspeak with a minor signal at 4.45 ppm, due to a small amount of water in the sample. These data conclusively demonstrate that the [(iPrPCP)Pd(OH)]/PhCHO mixture is in equilibrium with a labile addition product with hemiacetal structure. The relative intensities of the 31P resonances in the low temperature spectra allow an estimation of Keq �5 at 30°C, corresponding to a slightly negative ΔGo(ca. 0.7 Kcal·mol1at 243 K). Cleavage of DMC by Ni and Pd Pincer Hydroxides As mentioned in the introduction, our previous studies on the pincer alkoxide/CO2/MeOH system have shown that the pitfall of the catalytic process shown in Scheme 1 is within step (E). Admitting that this is a mechanistically simple process, the microscopic reversibility principle dictates that its reversal, namely, cleavage of DMC by the Ni and Pd pincer hydroxides should proceed through the same but inverse mechanistic sequence. In consequence, any thermodynamic or kinetic rate measurement pertaining to the process depicted by Eq. 4 would be also relevant to process (E). In an effort to unveil the causes, we decided to investigate the reactions of hydroxides [(iPrPCP)M(OH] whether the process shown Eq. 4 does in fact occur. (4) Preliminary experiments in C6D6solution showed that both pincer hydroxides do react very slowly with DMC, though appreciably faster with the Pd complex. Somewhat disappointingly, NMR analyses of the reaction mixtures revealed the formation of the expected alkylcarbonate [(iPrPCP)Pd(OCOOMe)] admixed with binuclear carbonate [{(iPrPCP)Pd}2(μ-CO3)] and unreacted hydroxide, whereas for M=Ni, most of the starting material hydroxide remained, and only a small amount of the corresponding carbonate [{(iPrPCP)Ni}2(μ-CO3)] was formed. Therefore, we decided to run both reactions in neat DMC, using 31P{1H} NMR to monitor the advance of the reaction. To avoid ambiguous assignments, the spectra were referenced with regard to external PPh3in C6D6(see the Experimental Section for details). Using gas-tight PTFE-valve sample tubes, both reactions proceed in a well-behaved manner, with no significant decay of the pincer framework over long reaction times. As anticipated by the preliminary experiments, the course of these reactions appear more complicated than suggested by Eq. 4. The reason is the lability of the [M]OCOOR linkage, which allows facile exchange of the OR unit, as previously reported.[15b] This is shown in Figure 2, which displays the 31P{1H} NMR monitoring of the reaction of [(iPrPCP)Ni(OH)] with DMC at 50°C (left) and 80°C (right). These experiments allow distinguishing two distinct stages in the reaction. The behavior of the Pd hydroxide is essentially the same, except that the reaction proceeds faster and the stages were not so clearly differentiated (see below). For convenience, the following discussion will focus Full Papers doi.org/10.1002/ejic.202100400 2962Eur. J. Inorg. Chem. 2021, 2958–2975 www.eurjic.org © 2021 The Authors. European Journal of Inorganic Chemistry published by Wiley-VCH GmbH Wiley VCH Freitag, 30.07.2021 2129 / 211589 [S. 2962/2975] 1 10990682c, 2021, 29, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/ejic.202100400 by Universidad De Sevilla, Wiley Online Library on [14/10/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License on the reaction with the nickel complex, but the same conclusions can be applied to the Pd system. As can be seen in the left side of Figure 2, the reaction appears to be almost complete after 11 h at 50°C, since very little hydroxide (signal tagged b) remains in the sample at this time. The signal of the main product, a, occurs at 55.9 ppm, which matches that of the binuclear carbonate [{(iPrPCP)Ni}2(μCO3)], as observed in the previous experiments. This was confirmed by comparison with an authentic sample[15a] of the carbonate recorded in the same conditions. This suggests that the reaction involves double cleavage of DMC, according to the stoichiometry shown in Eq. 5. (5) In addition to signal a, a low intensity signal, d, was observed at 53.0 ppm. The chemical shift for this resonance is distinct enough from those of aand b, allowing its assignment to the methoxide complex, [(iPrPCP)Ni(OMe)] (52.6 ppm in C6D6).[11] The latter can arise by methanol/water exchange with the remaining hydroxide complex, as shown in Eq. 6, see below.[13] Accordingly, signal dgrows broader as the exchange rate is accelerated by the water and methanol, and ultimately fades in the baseline when the hydroxide complex is nearly depleted by its reaction with DMC in the last few spectra shown on the top. Another effect of the rising concentration of methanol and water is the visible drift of the [(iPrPCP)Ni(OH)] signal, b. The position of this signal “senses” the concentration of acidic molecules because the basic hydroxide ligand is a powerful hydrogen bond acceptor. To verify this effect, we conducted an NMR titration of a 0.037 M solution of [Ni]OH with water in C6D6at room temperature, using both the 31P and 1H channels (see SI, Figures S3 and S4). We observed an approximately linear dependency of the chemical shift of the 31P resonance on the amount of added water, from 55.2 ppm in anhydrous benzene to 56.0 ppm in the presence of 3 moleequivalents of water. The magnitude and sign of this effect resembles the one observed in Figure 2. The hydroxyl 1H resonance retained its characteristic triplet structure (due to coupling to 31P) until the NiOH/H2O ratio reached 1:1, indicating that chemical exchange between the coordinated hydroxide ligand and free water is slow on the NMR timescale. The capacity of hydroxide complexes to accept strong hydrogen bonds is supported by the formation of isolable hydrates. The [(iPrPCP)Pd(OH)] was shown to crystallize as a binuclear hydrate containing a doubly hydrogen-bridged water molecule linking the OH ligands of each moiety.[11] We now report the X-ray diffraction structure of its nickel analogue (see SI). Crystals of this well-defined hemihydrate were reproducibly grown from concentrated solutions of [(iPrPCP)Ni(OH)] in moist diethyl ether. The O···O distances (2.819(4) Å) are significantly longer than in its Pd analogue (2.768(11) Å, avg.),[11] implying that H bond is stronger in the latter. Whereas the above observations apparently contradict the straightforward result anticipated in Eq. 4, we realized that longer reaction times cause the appearance of a new signal (c) at 56.8 ppm, which corresponds to the expected alkylcarbonate, [(iPrPCP)Ni(OCO2Me)] (56.5 ppm in C6D6).[15b] As can be seen in the right side of Figure 2, signal crapidly becomes prevalent when the reaction is carried out at 80°C. At this temperature, the conversion of the starting hydroxide complex into the binuclear carbonate takes just a few minutes (see spectra at the bottom of the right figure). From that point on, the signal of the carbonate complex (a) is gradually replaced by that of the methylcarbonate (c). Note that the methoxide (signal b), visible in the initial stages of the experiment, rapidly disappears, too. In conclusion, at 80°C the reaction is clean and complete within Figure 2. 31P{1H} spectra of a solution containing 25 μmol of [Ni]OH in neat DMC at two different temperatures: 50°C (left) and 80°C (right). ([Ni] stands for the [(iPrPCP)Ni] fragment). Full Papers doi.org/10.1002/ejic.202100400 2963Eur. J. Inorg. Chem. 2021, 2958–2975 www.eurjic.org © 2021 The Authors. European Journal of Inorganic Chemistry published by Wiley-VCH GmbH Wiley VCH Freitag, 30.07.2021 2129 / 211589 [S. 2963/2975] 1 10990682c, 2021, 29, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/ejic.202100400 by Universidad De Sevilla, Wiley Online Library on [14/10/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 24 h (top spectrum), the whole process being as expected according to Eq. 4. As mentioned above, the reaction of the Pd hydroxide in neat DMC proceeds much more readily than for Ni. The signal of the [(iPrPCP)PdOCO2Me] becomes noticeable shortly after mixing at room temperature, and the transformation is complete after 24 h at 50°C. Figure 3 shows the evolution of the 31P-containing species plotted against time in the Ni and Pd systems, at 50 and 25°C, respectively. The apparent complexity of the reactions of the hydroxides with DMC could have been foreseen on the basis of our previous investigations, summarized in Scheme 1. The binuclear carbonate and methoxide complexes arise as a consequence of rapid exchange equilibria which, within the range of temperature explored, are much faster than the cleavage of DMC. The whole process can be represented by a simple three-equation set: 4b, 6 and 7. We have shown before that the equilibria shown in Eq 7 are strongly temperature-dependent.[15] At room temperature (M=Pd) or above (M=Ni), these are shifted to the right side, reducing the amount of hydroxide complex available to react with DMC. This nearly stops the latter reaction. Increasing the temperature to 80°C allows overriding equilibrium 7, which gradually reverts, giving back the hydroxide complex sequestered in the binuclear carbonate. Eventually, the whole 31P intensity ends up in the single, intense peak c (Figure 2, right side). At this point, the overall stoichiometry of the process matches Eq. 4. Despite the complexity of the kinetic system, our results could be reasonably well modeled on the basis of equations 9b, 11 and 12, using numerical kinetic simulation software.[38] Our kinetic model describes the initial step as irreversible, and explicitly includes the forward and backward steps for Eqs. 11 and 12. The small deviation observed in the fitting of the palladium plot could be attributed to the difficult integration of the broad 31P resonances of the hydroxide and methoxide signals, which are lost at an early stage of the experiment. This simple model also avoids explicit consideration of bicarbonate complexes, ([(iPrPCP)M(OCO2H)], which should be part of the equilibrium (see Scheme 1), but were not detected. Independently of the metal, the concentration of the bicarbonate complex is expected to be fairly small. Fast exchange also contributes to the loss of these signals. Notwithstanding these shortcomings, the simplified model accounts for all the main features of the plots and provides reasonably good thermodynamic and kinetic data to be used as benchmarks for DFT calculations. (4b) (6) (7) DFT analysis of the nucleophilic reactions of Ni and Pd pincer hydroxides with PhCHO and DME In a previous work,[13] we showed that the GGA functional PBE, in combination with an implicit solvent model (CPCM), satisfactorily reproduces the experimental geometry and key vibrational frequencies of iPrPCP alkoxide complexes of Ni and Pd, at a low computational cost. Other authors concur in that PBE produces accurate geometries for transition metal complexes and is also good for approximate thermochemical data.[39] In this work, we have applied some additional refinements to the above methodology, which improve the quality of the results without increasing significantly their computational cost (see Computational Details in the Experimental Section). Even for a few relatively simple reactions, predicting the preferred configuration of the pincer iPr substituents in each intermediate, although feasible, increases very significantly the computational effort. Therefore, we removed conformational uncertainty by replacing i-propyl substituents by methyl groups (i.e.,MePCP ligand instead of iPrPCP). In our own experience,[14b,15a] and also for others,[26,28b] this simplification does not have excessive impact on the main conclusions. As shown below, the agreement between experimental and computed data proves satisfactory in general. Since the focus of this work is on the nucleophilic reactivity of the Ni and Pd complexes [(iprPCP)M(OH)], we limited our analysis to the attack of the hydroxides on the carbonyl functionality of PhCHO and DMC and the release of the primary products (benzoic acid and methanol, respectively), leaving aside the non-essential equilibria that complicate both systems. Scheme 4 and 5 display the models used in each calculation, and their results are summarized in the corresponding free energy profiles (Figure 4, Figure 5 and Figure 6). Relative values of free (ΔG) and ZPE-corrected electronic energies (ΔE; ZPE= Zero-Point Energy correction, in italics) are explicitly shown Figure 3. Evolution of the 31P-containing species involved in the reaction of hydroxides [M]OH with DMC at 50°C (M=Ni) and 25°C (M=Pd). Solid lines represent numerical fitting of the experimental data to the kinetic model represented by eqs 4b, 6 and 7. Full Papers doi.org/10.1002/ejic.202100400 2964Eur. J. Inorg. Chem. 2021, 2958–2975 www.eurjic.org © 2021 The Authors. European Journal of Inorganic Chemistry published by Wiley-VCH GmbH Wiley VCH Freitag, 30.07.2021 2129 / 211589 [S. 2964/2975] 1 10990682c, 2021, 29, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/ejic.202100400 by Universidad De Sevilla, Wiley Online Library on [14/10/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License next to each mark in the profiles. Full details including the breakdown of the different contributions to the energy values and atomic coordinates for each individual molecule can be found in the SI. Scheme 4 describes the initial stages of the reactions of the Ni and Pd hydroxides [(MePCP)M(OH)] with benzaldehyde, to afford equimolar mixtures of the corresponding hydride [(MePCP)M(H)] and benzoic acid. The sequence begins with the starting materials (jointly termed A0), namely, the hydroxide complex and benzaldehyde, and concludes at A2, an equimolar mixture of benzoic acid and the corresponding hydride complex. The energies of A0 and A2 are the sum of its individual components, i.e., two non-interacting molecules. A0 is conventionally taken as the origin (“zero”) of energies in each of the mechanistic pathways (M=Ni or Pd). On the whole, these are simple mechanisms involving a single intermediate, the “tetrahedral” adducts, A1 (M=Ni or Pd). These are connected to A0 and A2 through the corresponding transition states (TS), TSA01 and TSA12. To locate these transition states, we started from the optimized geometries of A1, imposing progressively shorter distances between the metal center to either the hydroxyl O atom (for TSA01), or to the acetal H atom (for TSA12). Intrinsic Reaction Coordinate (IRC) calculations on the optimized TS geometries confirmed that: i) the transition states do correspond to the intended processes; and ii) there are no additional stationary points along the reaction pathway. Both TSA01 and TSA12 can be seen as intramolecular σ-bond concerted exchanges whereby the [M]OC bond of A1 is displaced by the vicinal hydroxyl or hydrogen, leading to new [M]OH (in A0) or [M]H (in A2), respectively. As can be seen in Figure 4 (solid lines), the reactions of the Ni and Pd hydroxides with benzaldehyde in THF at 298 K show similar profiles. In either case, formation of the A1 adduct from A0 is endergonic, slightly more so for Ni (ΔGo= + 5.5 Kcal·mol1) than Pd (+3.4). The moderate energy barriers of less than 20 Kcal·mol1imply that the initial attacks on the carbonyl are effectively reversible processes. The next step, βhydrogen elimination, is exergonic and affords benzoic acid and hydride complex (Ni, 3.3; Pd, 7.7 Kcal·mol1). The energy barriers for β-elimination (TSA12) are 4–5 Kcal·mol1higher than for the nucleophilic attack (TSA01). In consequence, the energy of TSA12 (β-H elimination from A1), poses the main Scheme 4. Simplified mechanism of the reaction of Ni and Pd pincer hydroxides [(MePCP)M(OH)] with benzaldehyde, as used in DFT calculations. Scheme 5. Computational model used to explore different mechanistic pathways for the reaction of nickel and palladium hydroxide pincer complexes with DMC. The least energy pathway is highlighted. Full Papers doi.org/10.1002/ejic.202100400 2965Eur. J. Inorg. Chem. 2021, 2958–2975 www.eurjic.org © 2021 The Authors. European Journal of Inorganic Chemistry published by Wiley-VCH GmbH Wiley VCH Freitag, 30.07.2021 2129 / 211589 [S. 2965/2975] 1 10990682c, 2021, 29, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/ejic.202100400 by Universidad De Sevilla, Wiley Online Library on [14/10/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License solution at 25°C. Yield: 85%. IR (Nujol): ν(cm1)=1608 (ν(C=O). 1H NMR (300 MHz, C6D6, 25°C) δ=0.82 (dtv, 3JHH �*JHP �6.9 Hz, 12 H, CHMeMe), 1.17 (dtv, 3JHH �*JHP �7.3 Hz, 12 H, CHMeMe), 2.20 (m, 4 H, CHMe2), 2.24 (s, 3 H, CH3), 2.63 (dtv, 2JHP =4.5 Hz, 2 H, CH2COCH3), 2.99 (tv, *JHP =4.2 Hz, 4 H, CH2(iPrPCP)), 7.12 (m, 3H, CHarom) ppm. 13C{1H} (75 MHz, C6D6, 25°C) δ=17.8 (CHMeMe), 19.0 (tv, *JCP =2.7 Hz, CHMeMe), 23.3 (tv, *JCP =11.0 Hz, CHMe2), 27.2 (tv, *JCP =6.7 Hz, CH2COCH3), 30.8 (CH3), 36.7 (tv, *JCP =12.1 Hz, CH2 (iPrPCP)), 121.5 (tv, *JCP =10.0 Hz, m-CHarom (iPrPCP)), 124.8 (p-CHarom (iPrPCP)), 149.3 (tv, *JCP =10.2 Hz, o-Carom (iPrPCP)), 170.4 (i-Carom (iPrPCP)), 209.1 (C=O) ppm. 31P{1H} (121 MHz, C6D6, 25°C) δ= 58.6 ppm. Elemental analysis for C23H40OP2Pd (500.9): calcd. C 55.15, H 8.05; found C 54.80, H 8.20. Reaction of the hydroxide complexes [(iPrPCP)MOH] with tBuNC. Syntheses of the carbamate complexes [(iPrPCP)MC(=O)NHtBu] (M=Ni or Pd). These compounds were prepared according to the same procedure, here we describe the method to prepare the Pd derivative: 37 μL (0.33 mmol) of t-BuNC were added to a solution of 150 mg (0.33 mmol) of (iPrPCP)PdOH in 20 mL of THF at 80°C. The mixture was left to reach room temperature, stirred for 1 h and then was evaporated under reduced pressure. The solid residue was extracted with hexane, and after filtration, the solution was concentrated and cooled to 30°C. The product was separated by filtration and dried under vacuum. Yield, 181 mg, 60%. [(iPrPCP)Ni(CONHt-Bu): IR (Nujol): ν(cm1)=1547 (ν(C=O)), 3432 (ν(NH)) . 1H NMR (300 MHz, C6D6, 25°C) δ=1.00 (dtv, 3JHH �*JHP �6.8 Hz, 12 H, CHMeMe), 1.33 (dtv, 3JHH �*JHP �7.5 Hz, 12 H, CHMeMe), 1.44 (s, 9 H, CMe3), 2.06 (m, 4 H, CHMe2), 3.03 (tv, *JHP = 3.9 Hz, 4 H, CH2), 5.03 (sa, 1 H, NH), 7.14 (m, 3H, CHarom) ppm. 13C{1H} (75 MHz, C6D6, 25°C) δ=18.2 (CHMeMe), 18.6 (CHMeMe), 24.6 (tv, *JCP =11.6 Hz, CHMe2), 30.1 (CMe3), 37.0 (tv, *JCP =13.9 Hz, CH2), 50.7 (CMe3),121.0 (tv, *JCP =8.5 Hz, m-CHarom), 125.1 (p-CHarom), 150.9 (tv, *JCP =12.9 Hz, o-Carom), 174.4 (t, 2JCP =12.8 Hz, i-Carom), 210.1 (tv, *JCP =24.6 Hz, C=O) ppm. 31P{1H} (121 MHz, C6D6, 25°C) δ= 59.4 ppm. [(iPrPCP)Pd(CONHt-Bu)]: IR (Nujol): ν(cm1)=1558 cm1.1H NMR (300 MHz, C6D6, 25°C) δ=0.94 (dtv, 3JHH �*JHP �7.2 Hz, 12 H, CHMeMe), 1.27 (dtv, 3JHH �*JHP �7.8 Hz, 12 H, CHMeMe), 1.53 (s, 9 H, CMe3), 2.02 (m, 4 H, CHMe2), 3.13 (tv, *JHP =4.2 Hz, 4 H, CH2), 4.95 (sa, 1 H, NH), 7.18 (m, 3H, CHarom) ppm. 13C{1H} (75 MHz, C6D6, 25°C) δ=18.0 (CHMeMe), 18.7 (tv, *JCP =2.0 Hz, CHMeMe), 24.7 (tv, *JCP = 11.2 Hz, CHMe2), 30.2 (CMe3), 38.2 (tv, *JCP =12.6 Hz, CH2), 50.7 (CMe3),121.0 (tv, *JCP =9.3 Hz, m-CHarom), 124.8 (p-CHarom), 149.9 (tv, *JCP =10.5 Hz, o-Carom), 175.1 (t, 2JCP =5.9 Hz, i-Carom), 205.4 (tv, *JCP = 8.3 Hz, C=O) ppm. 31P{1H} (121 MHz, C6D6, 25°C) δ=57.6 ppm. Synthesis of [(iPrPCP)M(OCOp-C6H4R) (M=Ni, Pd; R=H, NO2, or NMe2)]. These compounds were prepared according to the same procedure, here we describe the method to prepare the Ni benzoate (iPrPCP)Ni(OCOPh): 61.1 mg (0.5 mmol) of benzoic acid were added to a solution of 207 mg (0.5 mmol) of (iPrPCP)NiOH in 20 mL of THF. The mixture was stirred at room temperature for 1 h and then was taken to dryness. The solid residue was extracted with hexane, and after filtration, the solution was concentrated and cooled to 30°C. The Ni compounds crystallized as yellow solids, and the Pd compounds as off-white or yellowish solids, in ca. 80% yield. (iPrPCP)Ni(OCOC6H5). IR (Nujol): ν(cm1)=1616 (ν(C=O)). 1H NMR (300 MHz, C6D6, 25°C) δ=1.23 (dtv, 3JHH �*JHP �6.9 Hz, 12 H, CHMeMe), 1.37 (dtv, 3JHH �*JHP �7.7 Hz, 12 H, CHMeMe), 2.16 (m, 4 H, CHMe2), 3.06 (tv, *JHP =4.1 Hz, 4 H, CH2), 6.84 (m, 3 H, CHarom (iPrPCP)), 7.32 (m, 3 H, m-, p-CHarom (Ph)), 7.90 (m, 2 H, o-CHarom (Ph)) ppm. 13C{1H} (75 MHz, C6D6, 25°C) δ=18.1 (CHMeMe), 18.8 (CHMeMe), 24.1 (tv, *JCP =10.0 Hz, CHMe2), 31.3 (tv, *JCP =13.4 Hz, CH2),122.3 (tv, *JCP =8.7 Hz, m-CHarom (iPrPCP)), 125.3 (p-CHarom (iPrPCP)), 127.8 (m-CHarom (Ph)), 129.5 (o-CHarom (Ph)), 129.8 (p-CHarom (Ph)), 137.6 (i-Carom (Ph)), 153.1 (tv, *JCP =12.9 Hz, o-Carom (iPrPCP)), 153.5 (t, 2JCP =17.2 Hz, i-Carom (iPrPCP)), 170.8 (C=O) ppm. 31P{1H} (121 MHz, C6D6, 25°C) δ=57.7 ppm. Elemental analysis for C27H40NiO2P2(517.3): calcd. C 62.70, H 7.79; found C 62.75, H 7.87. [(iPrPCP)Ni(OCOp-C6H4NO2)]. IR (Nujol): ν(cm1)=1628 (ν(C=O)). 1H NMR (300 MHz, C6D6, 25°C) δ=1.23 (dtv, 3JHH �*JHP �7.0 Hz, 12 H, CHMeMe), 1.35 (dtv, 3JHH �*JHP �7.7 Hz, 12 H, CHMeMe), 2.14 (m, 4 H, CHMe2), 3.07 (tv, *JHP =4.1 Hz, 4 H, CH2), 6.84 (m, 3 H, CHarom (iPrPCP)), 8.04 (d, 3JHH =8.6 Hz, 2 H, o-CHarom (PhNO2)), 8.15 (d, 3JHH = 8.7 Hz, 2 H, m-CHarom (PhNO2)), ppm. 13C{1H} (75 MHz, C6D6, 25°C) δ=18.1 (CHMeMe), 18.8 (CHMeMe), 24.1 (tv, *JCP =10.0 Hz, CHMe2), 31.2 (tv, *JCP =13.5 Hz, CH2), 122.5 (tv, *JCP =8.8 Hz, m-CHarom (iPrPCP)), 123.2 (m-CHarom (PhNO2)), 125.5 (p-Carom (iPrPCP)), 130.4 (oCHarom (PhNO2)), 142.9 (i-Carom (PhNO2)), 149.0 (p-Carom (PhNO2)), 152.6 (t, 2JCP =12.9 Hz, i-Carom (iPrPCP)), 153.0 (tv, *JCP =16.9 Hz, oCarom (iPrPCP)), 168.9 (C=O) ppm. 31P{1H} (121 MHz, C6D6, 25°C) δ= 58.1 ppm. Elemental analysis for C27H39NNiO4P2(562.2): calcd. C 57.68, H 6.99, N 2.49; found C 57.79, H 6.84, N 2.66. (iPrPCP)Ni(OCOp-C6H4NMe2)]. IR (Nujol): ν(cm1)=1602 cm1 (ν(C=O)). 1H NMR (300 MHz, C6D6, 25°C) δ=1.24 (dtv, 3JHH �*JHP �6.9 Hz, 12 H, CHMeMe), 1.37 (dtv, 3JHH �*JHP �7.7 Hz, 12 H, CHMeMe), 2.17 (m, 4 H, CHMe2), 2.97 (s, 6 H, NMe2), 3.05 (tv, *JHP = 4.0 Hz, 4 H, CH2), 6.82 (m, 3 H, CHarom (iPrPCP)), 6.61 (d, 3JHH =8.6 Hz, 2 H, o-CHarom (PhNMe2)), 7.76 (d, 3JHH =8.8 Hz, 2 H, m-CHarom (PhNMe2)), ppm. 13C{1H} (75 MHz, C6D6, 25°C) δ=18.1 (CHMeMe), 18.8 (CHMeMe), 24.1 (tv, *JCP =9.9 Hz, CHMe2), 31.4 (tv, *JCP = 13.3 Hz, CH2), 40.5 (NMe2), 110.9 (m-CHarom (PhNMe2)), 122.2 (tv, *JCP =8.6 Hz, m-CHarom (iPrPCP)), 125.1 (p-Carom (iPrPCP)), 125.8 (i-Carom (PhNMe2)), 130.8 (o-CHarom (PhNMe2)), 151.9 (p-Carom (PhNMe2)), 153.1 (tv, *JCP =13.1 Hz, o-Carom (iPrPCP)), 154.1 (t, 2JCP =17.3 Hz, iCarom (iPrPCP)), 171.5 (C=O) ppm. 31P{1H} (121 MHz, C6D6, 25°C) δ= 57.4 ppm. Elemental analysis for C29H45NNiO2P2(560.3): calcd. C 62.16, H 8.09, N 2.50; found C 62.18, H 7.78, N 2.75. (iPrPCP)Pd(OCOC6H5): IR (Nujol): ν(cm1)=1611 (ν(C=O)). 1H NMR (300 MHz, C6D6, 25°C) δ=1.20 (dtv, 3JHH �*JHP �7.2 Hz, 12 H, CHMeMe), 1.31 (dtv, 3JHH �*JHP �8.0 Hz, 12 H, CHMeMe), 2.30 (m, 4 H, CHMe2), 3.19 (tv, *JHP =4.3 Hz, 4 H, CH2), 6.92 (t, 3JHH =7.3 Hz, 1 H, p-CHarom (iPrPCP)), 7.00 (d, 3JHH =7.5 Hz, 2 H, m-CHarom (iPrPCP)), 7.34 (m, 3 H, m-, p-CHarom (Ph)), 8.00 (m, 2 H, o-CHarom (Ph)) ppm. 13C {1H} (75 MHz, C6D6, 25°C) δ=18.0 (CHMeMe), 18.9 (CHMeMe), 24.8 (tv, *JCP =10.9 Hz, CHMe2), 32.4 (tv, *JCP =11.7 Hz, CH2),122.8 (tv, *JCP =10.5 Hz, m-CHarom (iPrPCP)), 125.1 (p-CHarom (iPrPCP)), 127.7 (mCHarom (Ph)), 129.6 (p-CHarom (Ph)), 129.7 (o-CHarom (Ph)), 138.6 (i-Carom (Ph)), 151.4 (tv, *JCP =10.6 Hz, o-Carom (iPrPCP)), 155.4 (i-Carom (iPrPCP)), 170.8 (C=O) ppm. 31P{1H} (121 MHz, C6D6, 25°C) δ=60.4 ppm. Elemental analysis for C27H40O2P2Pd (565.0): calcd. C 57.40, H 7.14; found C 57.41, H 7.26. (iPrPCP)Pd(OCOp-C6H4NO2): IR (Nujol): ν(cm1)=1628 (ν(C=O)). 1H NMR (300 MHz, C6D6, 25°C) δ=1.19 (dtv, 3JHH �*JHP �7.3 Hz, 12 H, CHMeMe), 1.29 (dtv, 3JHH �*JHP �8.0 Hz, 12 H, CHMeMe), 2.29 (m, 4 H, CHMe2), 3.19 (tv, *JHP =4.3 Hz, 4 H, CH2), 6.92 (t, 3JHH =7.2 Hz, 1 H, p-CHarom (iPrPCP)), 6.99 (d, 3JHH =6.5 Hz, 2 H, m-CHarom (iPrPCP)), 8.15 (m, 4 H, CHarom (PhNO2)) ppm. 13C{1H} (75 MHz, C6D6, 25°C) δ= 18.0 (CHMeMe), 18.8 (CHMeMe), 24.8 (tv, *JCP =10.9 Hz, CHMe2), 32.2 (tv, *JCP =11.7 Hz, CH2), 122.9 (tv, *JCP =10.5 Hz, m-CHarom (iPrPCP)), 123.1 (m-CHarom (PhNO2)), 125.3 (p-Carom (iPrPCP)), 130.5 (o-CHarom (PhNO2)), 144.3 (i-Carom (PhNO2)), 148.9 (p-Carom (PhNO2)),151.4 (tv, *JCP =10.8 Hz, o-Carom (iPrPCP)), 154.7 (i-Carom (iPrPCP)), 168.8 (C=O) ppm. 31P{1H} (121 MHz, C6D6, 25°C) δ=60.7 ppm. Elemental Anal. for C27H39NO4P2Pd (610.0): calcd. C 53.16, H 6.44, N 2.30; found C 53.12, H 6.50, N 2.42. Full Papers doi.org/10.1002/ejic.202100400 2972Eur. J. Inorg. Chem. 2021, 2958–2975 www.eurjic.org © 2021 The Authors. European Journal of Inorganic Chemistry published by Wiley-VCH GmbH Wiley VCH Freitag, 30.07.2021 2129 / 211589 [S. 2972/2975] 1 10990682c, 2021, 29, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/ejic.202100400 by Universidad De Sevilla, Wiley Online Library on [14/10/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License (iPrPCP)Pd(OCOp-C6H4NMe2): IR (Nujol): ν(cm1)=1598 1H NMR (300 MHz, C6D6, 25°C) δ=1.20 (dtv, 3JHH �*JHP �7.2 Hz, 12 H, CHMeMe), 1.31 (dtv, 3JHH �*JHP �8.0 Hz, 12 H, CHMeMe), 2.32 (m, 4 H, CHMe2), 2.98 (s, 6 H, NMe2), 3.19 (tv, *JHP =4.3 Hz, 4 H, CH2), 6.65, 7.87 (AA’XX’, 4 H, o-m-CHarom (PhNMe2)),6.92 (t, 3JHH =7.4 Hz, 1 H, p-CHarom (iPrPCP)), 6.99 (d, 3JHH =7.7 Hz, 2 H, m-CHarom (iPrPCP)) ppm. 13C{1H} (75 MHz, C6D6, 25°C) δ=18.0 (CHMeMe), 18.9 (CHMeMe), 24.7 (tv, *JCP =10.9 Hz, CHMe2), 32.5 (tv, *JCP =11.5 Hz, CH2), 40.5 (NMe2), 111.1 (m-CHarom (PhNMe2)), 122.8 (tv, *JCP = 10.4 Hz, m-CHarom (iPrPCP)), 125.0 (p-Carom (iPrPCP)), 126.6 (i-Carom (PhNMe2)), 131.1 (o-CHarom (PhNMe2)), 151.5 (tv, *JCP =10.9 Hz, oCarom (iPrPCP)), 151.9 (p-Carom (PhNMe2)), 155.8 (i-Carom (iPrPCP)), 171.4 (C=O) ppm. 31P{1H} (121 MHz, C6D6, 25°C) δ=60.1 ppm. Elemental analysis for C29H45NO2P2Pd (680.0): calcd. C 57.28, H 7.46, N 2.30; found C 57.29, H 7.56, N 2.47. Simultaneous NMR and GC monitoring the reaction of [(iPrPCP) PdOH] with PhCHO at room temperature. A 50 mL glass ampoule equipped with a magnetic stirrer was charged with 10 mL of a THF solution containing 23.1 mg (0.05 mmol) of the palladium hydroxide complex, and 3.9 mg of biphenyl as GC internal standard. 15.3 mL of neat benzaldehyde were added to the solution. The solution was stirred for 24 h, withdrawing aliquots of 0.7 mL at different times, which were cannula-transferred to an NMR tube containing the external reference sealed in a borosilicate glass capillary tube. After recording the 31P{1H} spectrum, the solution was poured in a vial in the open air and subjected to GC analyses. Kinetics of the reactions of [(iPrPCP)MOH] (M=Ni, or Pd) with neat DMC. These reactions were carried out in NMR tubes with gas-tight PTFE screw valves, containing the external reference sealed in a borosilicate glass capillary tube. In each case, an amount corresponding to 0.025 mmol of the starting complex (Ni: 10.3 mg; Pd: 11.5 mg) were dissolved in 0.6 mL of neat DMC. The samples were transferred to the NMR probe, pre-heated and stabilized at the working temperature, and the reaction progress was monitored by integration of the 31P{1H} resonances. Once the reaction was complete, the solution was evaporated to dryness, and dissolved in C6D6to check the identity of the main products (methylcarbonate complexes, [(iPrPCP)M(OCO2Me)]. As reported before,[15] this procedure causes the essentially pure samples to become contaminated with some binuclear carbonate {[(iPrPCP)M]2(m-CO3)}, but the main species was still the methylcarbonate. The identity of the resonances of the binuclear carbonate complexes in DMC was verified by comparison to the spectra of authentic samples in the same solvent and temperatures (with regard to the PPh3external standard). In general, the chemical shifts of every signal were quite similar to their reported values in C6D6. The sum of the intensities of the 31P resonances was approximately constant with regard to the external standard, therefore for the kinetic calculations these were normalized giving the value 1 to the sum of all intensities. To deduce the actual values of rate constants, we used the molar concentrations of DMC, calculated using literature values of DMC at the experiment temperatures (in Kg·L1: 25°C, 1.063; 50°C, 1.030).[43] The data were fit to a suitable kinetic model using the “MacKinetics” program and the results checked with the software Dynafit.[38] Isolation of the hydrate [(iPrPCP)Ni(OH)]·1/2 H2O. Full details for the titration with water of the anhydrous hydroxide, IR data (nujol), molecular weight in solution (benzene, cryoscopy freezing point depression), elemental analysis and X-ray structure are given in the SI. Deposition Number 2078309 (for [Ni]OH·1/2H2O) contains 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. Computational details All DFT calculations were performed with the Spartan’18 software package.[44] Geometry optimizations and vibrational analyses were carried out with the PBE functional and the 6–31G* for the nickel complexes and LACVP* for palladium complexes. The latter comprises the 6–31G* functions for light elements and the LANL2DZ pseudopotential for palladium. The energy gradient for criterion for geometric convergence was tightened from Spartan’s default value 3×104to 5.5×105erg/bohr. Solvent effects were included in the optimization procedure with the CPCM implicit model, setting the Bondi-type PCM radii for Ni and Pd as 2.28 and 2.48 Å, respectively. These were estimated from X-ray diffraction structures of pincer hydroxide and alkoxide complexes, as described before.[13] The geometries of transition states were checked to have only one imaginary frequency and their identity was tested by an Internal Reaction Coordinate (IRC) calculation. Electronic (SCF) energies were refined with a single point calculation at the PBED3/6-311+ +G(3df,2p) level (SCF convergence criterion set to “HIGH”), which includes Grimme’s empirical correction for dispersive forces functional, and the largest of Popletype triple-ζquality basis function set available in Spartan. At this level of the theory, Spartan uses the all-electron def2-TZVP basis to describe the Pd atom. Single-point calculations were accelerated using the “dual” option, which specifies that SCF convergence is achieved at the 6–311G* level, and then corrected perturbatively for the effect of additional diffuse and polarization functions. Preliminary calculations showed that the introduction of the D3 parameter during the geometry optimization only results in minor improvements of the final geometries but hampers geometry convergence, particularly for transition states. For each stationary point in the potential energy surface, free energies (G°) at 298 K were estimating adding the SCF electronic energies (from the higher level single point) to the Thermal and Solvent Corrections determined at the geometry optimization level of theory, namely: G°(298 K)=E(SCF, PBE-D3/6-311+ +G(3df,2p))+TC (at 298 K)+ SC(298 K). The same procedure was extended to different temperatures. Thermal Corrections for a given temperature were computed at the same level for which the vibrational data were available (namely, the double-ζquality basis set used for geometry optimization) TC(at T)=G°(PBE/6-31G*, CPCM(T)) – E(SCF)(PBE/631G*, CPCM(T)). To compute the solvent correction, we performed an unexpensive additional gas phase single-point energy calculation on optimized geometries at the PBE/6-31*G level (i.e., omitting the solvent calculation), therefore SC(T)=E(SCF, PBE/631G*, CPCM(T))-E(SCF, PBE/6-31G*, gas phase). In addition to reducing the computational cost, this procedure has the advantage of computing the solvent effect at a similar level of the theory at which the CPCM model was originally parametrized.[45] CPCM calculations have their own temperature specification (independent from that of the thermal correction, but often neglected). For the calculations at temperatures different from 298 K, geometry optimization and vibrational analyses were updated with consistent use of temperature both for the thermal and solvent corrections, and single point calculations were performed over the re-optimized geometry. Full Papers doi.org/10.1002/ejic.202100400 2973Eur. J. Inorg. Chem. 2021, 2958–2975 www.eurjic.org © 2021 The Authors. European Journal of Inorganic Chemistry published by Wiley-VCH GmbH Wiley VCH Freitag, 30.07.2021 2129 / 211589 [S. 2973/2975] 1 10990682c, 2021, 29, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/ejic.202100400 by Universidad De Sevilla, Wiley Online Library on [14/10/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Acknowledgements This work was supported by the Spanish Research Agency (AEI), the European Union (Feder Funds) and Junta de Andalucía through projects PGC2018-095768-B-100 and PY20_0104. Conflict of Interest The authors declare no conflict of interest. Keywords: Pincer complexes ·Nickel ·Palladium ·Hydroxide complexes ·Reaction mechanisms [1] a) Pincer Compounds: Chemistry and Applications (Ed: D. MoralesMorales), Amsterdam, Elsevier, 2018; b) Pincer and Pincer-Type Complexes: Applications in Organic Synthesis and Catalysis (Eds: K. J. Szabó, O. F. Wendt), Weinheim, Wiley-VCH, 2014; c) The Chemistry of Pincer Compounds (Eds: D. Morales-Morales, C. Jensen), Amsterdam, Elsevier, 2007; d) Organometallic Pincer Chemistry (Eds: G. van Koten, D. Milstein), Heidelberg, Elsevier, 2013. [2] a) L. S. Merz, J. Ballmann, L. H. Gade, Eur. J. Inorg. Chem. 2020, 2023– 2042; b) A. Singh, D. Gelman, ACS Catal. 2020,10, 1246–1255; c) R. Arevalo, P. J. Chirik, J. Am. Chem. Soc. 2019,141, 9106–9123; d) H. Valdés, L. González-Sebastián, D. Morales-Morales, J. Organomet. Chem. 2017,845, 229–257; e) R. E. Andrew, L. González-Sebastián, A. B. Chaplin, Dalton Trans. 2016,45, 1299–1305; f) S. Murugesan, K. Kirchner, Dalton Trans. 2016,45, 416–439. [3] See, for example: a) A. Rossin, M. Peruzzini, Chem. Rev. 2016,116, 8848– 8872; b) N. A. Eberhardt, H. Guan, Chem. Rev. 2016,116, 8373–8426. [4] For some discussion of the trans influence and labilizing effects in pincer complexes, see: a) A. J. Canty, A. Ariafard, G. van Koten, Chem. Eur. J. 2020,26, 15629–15635; b) H.-W. Suh, T. J. Schmeier, N. Hazari, R. A. Kemp, M. K. Takase, Organometallics 2012,31, 8225–8236; c) Y. X. Jia, X. Y. Yang, W. S. Tay, Y. Li, S. A. Pullarkat, K. Xu, H. Hirao, P. H. Leung, Dalton Trans. 2016,45, 2095–2101; d) E. Sola, Silicon-Based Pincers: Trans Influence and Functionality Chapter 19 in Ref. 1a. [5] Some representative reviews: a) H. E. Bryndza, W. Tam, Chem. Rev. 1988, 88, 1163–1188; b) M. D. Fryzuk, C. D. Montgomery, Coord. Chem. Rev. 1989,95, 1–40; c) H. W. Roesky, S. Singh, K. K. M. Yusuff, J. A. Maguire, N. S. Hosmane, Chem. Rev. 2006,106, 3813–3843. [6] a) R. Jira, Angew. Chem. Int. Ed. 2009,48, 9034–9037; Angew. Chem. 2009,121, 9196–9199; b) A. Comas-Vives, A. Stirling, A. Lledos, G. Ujaque, Chem. Eur. J. 2010,16, 8738–8747. [7] a) H. Vahrenkamp, Dalton Trans. 2007, 4751–4759; b) A. Dolega, Coord. Chem. Rev. 2010,254, 916–937. [8] A. G. Blackman, L. R. Gahan, Z. Anorg. Allg. Chem. 2018,644, 616–629. [9] a) H. Valdés, M. A. García-Eleno, D. Canseco-González, D. MoralesMorales, ChemCatChem 2018,10, 3136–3172; b) V. Arora, H. Narjinari, P. G. Nandi, A. Kumar, Dalton Trans. 2021,50, 3394–3428. [10] L. M. Martínez-Prieto, J. Cámpora, Isr. J. Chem. 2020,60, 373–393. [11] J. Cámpora, P. Palma, D. del Río, E. Álvarez, Organometallics 2004,23, 1652–1655. [12] J. Cámpora, P. Palma, D. del Río, M. M. Conejo, E. Álvarez, Organometallics 2004,23, 5653–5655. [13] L. M. Martínez-Prieto, P. Palma, E. Álvarez, J. Cámpora, Inorg. Chem. 2017,56, 13086–13099. [14] a) C. Melero, L. M. Martinez-Prieto, P. Palma, D. del Rio, E. Alvarez, J. Campora, Chem. Commun. 2010,46, 8851–8853; b) L. M. MartinezPrieto, E. Avila, P. Palma, E. Alvarez, J. Cámpora, Chem. Eur. J. 2015,21, 9833–9849. [15] a) L. Miguel Martinez-Prieto, C. Real, E. Avila, E. Alvarez, P. Palma, J. Campora, Eur. J. Inorg. Chem. 2013,2013, 5555–5566; b) L. M. MartinezPrieto, P. Palma, J. Campora, Dalton Trans. 2019,48, 1351–1366. [16] L. M. Martínez-Prieto, Ph. D. Dissertation. Ni and Pd Pincer Complexes: A Model for the Study of Catalytic Alcohol Carboxylation in Homogeneous Phase, University of Seville, 2012. [17] a) M. Aresta, A. Dibenedetto, A. Dutta, Catal. Today 2017,281, 345–351; b) M. Aresta, A. Dibenedetto, A. Angelini, I. Papai, Top. Catal. 2015,58, 2–14; c) S. Huang, B. Yan, S. Wang, X. Ma, Chem. Soc. Rev. 2015,44, 3079–3116; d) M. Aresta, A. Dibenedetto, A. Angelini, Adv. Inorg. Chem. 2014,66, 259–288; e) T. Sakakura, K. Kohno, Chem. Commun. 2009, 1312–1330. [18] P. Tundo, M. Selva, Acc. Chem. Res. 2002,35, 705–716. [19] A.-A. G. Shaikh, S. Sivaram, Chem. Rev. 1996,96, 951–976. [20] Q. Cai, B. Lu, L. Guo, Y. Shan, Catal. Commun. 2009,10, 605–609. [21] See, for example: F. A. Carey, R. J. Sundberg, Advanced Organic Chemistry, 3rd Edition. Part A, Structure and Mechanisms. New York, Plenum Press, 1990. pp 465–470. [22] a) A. Arabaoui, C. Redshaw, Polym. Chem. 2010,1, 801–826; b) D. J. Walsh, M. G. Hyatt, S. A. Miller, D. Guironnet, ACS Catal. 2019,9, 11153– 11188. [23] a) R. Johansson, O. F. Wendt, Organometallics 2007,26, 2426–2430; b) K. J. Jonasson, A. H. Mousa, O. F. Wendt, Polyhedron 2018,143, 132– 137; c) A. H. Mousa, J. Bendix, O. F. Wendt, Organometallics 2018,37, 2581–2593. [24] a) G. R. Fulmer, R. P. Muller, R. A. Kemp, K. I. Goldberg, J. Am. Chem. Soc. 2009,131, 1346–1349; b) G. R. Fulmer, A. N. Herndon, W. Kaminsky, R. A. Kemp, K. I. Goldberg, J. Am. Chem. Soc. 2011,133, 17713–17726; c) G. R. Fulmer, W. Kaminsky, R. A. Kemp, K. I. Goldberg, Organometallics 2011, 30, 1627–1636; d) W. D. Bailey, A. S. Phearman, L. Luconi, A. Rossin, D. G. Yakhvarov, L. D’Accolti, S. E. Flowers, W. Kaminsky, R. A. Kemp, G. Giambastiani, K. I. Goldberg, Chem. Eur. J. 2019,25, 9920–9929. [25] a) D. Huang, O. V. Makhlynets, L. L. Tan, S. C. Lee, E. V. Rybak-Akimova, R. H. Holm, Proc. Nat. Acad. Sci. 2011,108, 1222–1227; b) D. Huang, O. V. Makhlynets, L. L. Tan, S. C. Lee, E. V. Rybak-Akimova, R. H. Holm, Inorg. Chem. 2011,50, 10070–10081. [26] T. J. Schmeier, A. Nova, N. Hazari, F. Maseras, Chem. Eur. J. 2012,18, 6915–6927. [27] C. Yoo, J. Kim, Y. Lee, Organometallics 2013,32, 7195–7203. [28] a) S. Chakraborty, P. Bhattacharya, H. G. Dai, H. R. Guan, Acc. Chem. Res. 2015,48, 1995–2003; b) N. A. Eberhardt, N. P. N. Wellala, Y. Z. Li, J. A. Krause, H. R. Guan, Organometallics 2019,38, 1468–1478. [29] a) J. Cámpora, I. Matas, P. Palma, E. Álvarez, C. Graiff, A. Tiripicchio, Organometallics 2007,26, 5712–5721; b) M. K. Samantaray, M. M. Shaikh, P. Ghosh, Organometallics 2009,28, 2267–2275; c) M. Kujime, S. Hikichi, M. Akita, Inorg. Chim. Acta 2003,350, 163–174. [30] a) M. Kujime, S. Hikichi, M. Akita, Organometallics 2001,20, 4049–4060; b) S. Kannan, A. J. James, P. R. Sharp, Inorg. Chim. Acta 2003,345, 8–14. [31] a) J. Cámpora, C. M. Maya, P. Palma, E. Carmona, E. Gutierrez-Puebla, C. Ruiz, J. Am. Chem. Soc. 2003,125, 1482–1483; b) J. Cámpora, C. M. Maya, P. Palma, E. Carmona, E. Gutierrez, C. Ruiz, C. Graiff, A. Tiripicchio, Chem. Eur. J. 2005,11, 6889–6904. [32] a) J. Vicente, A. Arcas, J. M. Fernandez-Hernandez, D. Bautista, P. G. Jones, Organometallics 2005,24, 2516–2527; b) K. P. Kepp, Inorg. Chem. 2016,55, 9461–9470. [33] S. A. Macgregor, G. W. Neave, Organometallics 2003,22, 4547–4556. [34] a) R. A. Michelin, R. Ros, J. Organomet. Chem. 1979,169, C42-C44; b) W.- S. Ojo, E. Paugam, F. Y. Petillon, P. Schollhammer, J. Talarmin, K. W. Muir, Organometallics 2006,25, 4009–4018. [35] a) L. C. Liang, C. W. Li, P. Y. Lee, C. H. Chang, H. M. Lee, Dalton Trans. 2011,40, 9004–9011; b) D. J. Mindiola, G. L. Hillhouse, Chem. Commun. 2002, 1840–1841; c) P. L. Holland, R. A. Andersen, R. G. Bergman, J. Am. Chem. Soc. 1996,118, 1092–1104. [36] L. M. Martinez-Prieto, C. Melero, D. del Rio, P. Palma, J. Cámpora, E. Alvarez, Organometallics 2012,31, 1425–1438. [37] see Ref 21, p. 439. [38] a) McKinetics v. 0.9.1b, by Walter Leipold III. http://members.dca.net./ leipold/mk/mk091.html; b) The results were checked with DynaFit v. 4.05.103 (Biokin, Ltd). [39] a) Y. Minenkov, A. Singstad, G. Occhipinti, V. R. Jensen, Dalton Trans. 2012,41, 5526–5541; b) Y. Minenkov, E. Chermak, L. Cavallo, J. Chem. Theory Comput. 2016,12, 1542–1560; c) S. Boonseng, G. W. Roffe, J. Spencer, H. Cox, Dalton Trans. 2015,44, 7570–7577; d) M. Poor Kalhor, H. Chermette, S. Chambrey, D. Ballivet-Tkatchenko, Phys. Chem. Chem. Phys. 2011,13, 2401–2408; e) See M. Swart’s “DFT poll” (http:// www.marcelswart.eu/dft-poll/), where PBE has remained within the first 4 positions for over the last 10 years. [40] a) A. Dibenedetto, M. Aresta, P. Giannoccaro, C. Pastore, I. Papai, G. Schubert, Eur. J. Inorg. Chem. 2006, 908–913; b) A. Y. Samuilov, Y. D. Samuilov, Theor. Chem. Acc. 2019,138. DOI: 10.1007/s00214-018-2411– 0. [41] Note that the hypothetical route through TSM12 poses an even more radical version of this problem, since the formation of M2 is an acidFull Papers doi.org/10.1002/ejic.202100400 2974Eur. J. Inorg. Chem. 2021, 2958–2975 www.eurjic.org © 2021 The Authors. European Journal of Inorganic Chemistry published by Wiley-VCH GmbH Wiley VCH Freitag, 30.07.2021 2129 / 211589 [S. 2974/2975] 1 10990682c, 2021, 29, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/ejic.202100400 by Universidad De Sevilla, Wiley Online Library on [14/10/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License base equilibrium involving the protonation of a methylcarbonate complex by methanol to generate methylcarbonic acid and a methoxide complex. [42] a) K. G. Caulton, New J. Chem. 1994,18, 25–41; b) J. T. Poulton, M. P. Sigalas, K. Folting, W. E. Streib, O. Eisenstein, K. G. Caulton, Inorg. Chem. 1994,33, 1476–1485. [43] E. Romano, J. L. Trenzado, E. Gonzalez, J. S. Matos, L. Segade, E. Jimenez, Fluid Phase Equilib. 2003,211, 219–240. [44] Spartan’18, Wavefuntion, Inc. Irvine, CA, USA. Version 1.4.4., 2019. [45] J. Ho, A. Klamt, M. L. Coote, J. Phys. Chem. A 2010,114, 13442–13444. Manuscript received: May 10, 2021 Revised manuscript received: June 18, 2021 Accepted manuscript online: June 23, 2021 Full Papers doi.org/10.1002/ejic.202100400 2975Eur. J. Inorg. Chem. 2021, 2958–2975 www.eurjic.org © 2021 The Authors. European Journal of Inorganic Chemistry published by Wiley-VCH GmbH Wiley VCH Freitag, 30.07.2021 2129 / 211589 [S. 2975/2975] 1 10990682c, 2021, 29, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/ejic.202100400 by Universidad De Sevilla, Wiley Online Library on [14/10/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License