ABSTRACT
We have computationally evaluated water as an active Lewis base (LB) and introduced the borohydride/hydronium intermediate in the mechanism of B(C6F5)3-catalyzed hydrogenation of carbonyl compounds with H2 in wet/moist ether. Our calculations extend the known frustrated Lewis pair mechanism of this reaction toward the inclusion of water as the active participant in all steps. Although the definition of the zero-energy point interweaves in comparison of the scenarios with and without water, we will be able to show that (i) water (hydrogen bonded to its molecular environment) can, in principle, act as a reasonably viable LB in cooperation with the borane Lewis acid such as B(C6F5)3 but relatively a strong borane-water complexation can be the hindering factor; (ii) the herein-proposed borohydride/hydronium intermediates with the hydronium cation having three OH···ether hydrogen bonds or a combination of the OH···ether/OH···ketone hydrogen bonds appear to be as valid as the previously considered borohydride/oxonium or borohydride/oxocarbenium intermediates; (iii) the proton-coupled hydride transfer from the borohydride/hydronium to a ketone (acetone) has a reasonably low barrier. Our findings could be useful for better mechanistic understanding and further development of the aforementioned reaction.
ABSTRACT
Knowing that the Papai's electron transfer (ET) and the Grimme's electric field (EF) models draw attention to somewhat different physical aspects, we are going to systematically (re)examine interactions in the transition states (TSs) of the heterolytic H2-cleavage by the Frustrated Lewis Pairs (FLPs). Our main vehicle is the quantitative energy decomposition analysis (EDA), a powerful method for elucidation of interactions, plus the analysis of molecular orbitals (MOs). Herein, the Lewis acid (LA) is B(C6F5)3 and the Lewis bases (LBs) are tBu3P, ( o-C6H4Me)3P, 2,6-lutidine, 2,4,6-lutidine, MeNâC(Ph)Me imine, MeN(H)-C(H)PhMe amine, THF, 1,4-dioxane, and acetone. For a series of the phosphorus-, nitrogen-, and oxygen-bearing LBs plus B(C6F5)3, we will show that (i) neither the electrostatic nor the orbital interactions dominate but instead both are essential alongside the Pauli repulsion and (ii) the frontier molecular orbitals (FMOs) of a TS can arise not only from the "push-pull" molecular orbital scheme by Papai et al., which directly involves the occupied σ and the empty σ* MOs of H2, but also from a more intricate but energetically more fitting orbital interactions which have escaped notice thus far.
ABSTRACT
We apply Born-Oppenheimer molecular dynamics to the practically significant [dioxane-H(+)-acetone][(C6F5)3B-H(-)] and [Et2O-H(+)-OCPr2][(C6F5)3B-H(-)] ion pair intermediates. Dynamics of hydride transfer in cation/anion ion pair takes place on the background of large-amplitude configurational changes. Geometry of oxocarbenium/borohydride ion pairs is flexible, meaning that we uncover significant actual structural disorder at a finite temperature. Therefore, although the starting structure can be fairly close to the configurational area of the hydride transfer transition state (TS) and despite a low potential energy barrier (ca. 1.5 kcal/mol, according to the literature), already at T ≈ 325 K the system can remain ignorant of the TS region and move round and about ("roam") in the configurational space for a period of time in the range between 10 and 100 ps. This indicates structural flexibility of oxocarbenium/borohydride ion pair on apparently a flat potential energy "landscape" of cation/anion interaction, and this has not been taken into consideration by the free energy estimations in static considerations made thus far. The difference between the dynamics-based representation of the system versus the static representation amounts to the difference between quasi-bimolecular versus unimolecular descriptions of the hydride transfer step.
ABSTRACT
We investigate the geometry of oxonium/borohydride ion pairs [ether-H(+)-ether][LA-H(-)] with dioxane, THF, and Et2O as ethers and B(C6F5)3 as the Lewis acid (LA). The question is about possible location of the disolvated proton, [ether-H(+)-ether], with respect to the hydride of the structurally complex [LA-H(-)] anion. Using Born-Oppenheimer molecular dynamics and a comparison of the potential and free energies of the optimized configurations, we show that herein considered ion pairs are much more flexible geometrically than previously thought. Conformers with different locations of cations with respect to anions are governed by a flat energy-landscape. We found a novel configuration in which oxonium is below [LA-H(-)], with respect to the direction of borane â hydride vector, and the proton-hydride distance is ca. 6 Å. With calculations of the vibrational spectra of [ether-H(+)-ether][(C6F5)3B-H(-)] for dioxane, THF, and Et2O as ethers, we investigate the manifestation of SSLB-type (short, strong, low-barrier) hydrogen bonding in the OHO motif of an oxonium cation.
ABSTRACT
As an extension of the reaction mechanism describing the base-catalyzed hydrogenation of ketones according to Berkessel etâ al., we use a standard methodology for transition-state (TS) calculations in order to check the possibility of heterolytic cleavage of H2 at the ketone's carbonyl carbon atom, yielding one-step hydrogenation path with involvement of carboxylic acid as a catalyst. As an extension of the catalyst scope in the base-catalyzed hydrogenation of ketones, our mechanism involves a molecule with a labile proton and a Lewis basic oxygen atom as a catalyst-for example, R-C(=O)OH carboxylic acids-so that the heterolytic cleavage of H2 could take place between the Lewis basic oxygen atom of a carboxylic acid and the electrophilic (Lewis acidic) carbonyl carbon of a ketone/aldehyde. According to our TS calculations, protonation of a ketone/aldehyde by a proton shuttle (hydrogen bond) facilitates the hydride-type attack on the ketone's carbonyl carbon atom in the process of the heterolytic cleavage of H2 . Ketones with electron-rich and electron-withdrawing substituents in combination with a few carboxylic and amino acids-in total, 41 substrate-catalyst couples-have been computationally evaluated in this article and the calculated reaction barriers are encouragingly moderate for many of the considered substrate-catalyst couples.
ABSTRACT
Using Born-Oppenheimer molecular dynamics (BOMD), we explore the nature of interactions between H2 and the activated carbonyl carbon, C(carbonyl), of the acetone-B(C6F5)3 adduct surrounded by an explicit solvent (1,4-dioxane). BOMD simulations at finite (non-zero) temperature with an explicit solvent produced long-lasting instances of significant vibrational perturbation of the H-H bond and H2-polarization at C(carbonyl). As far as the characteristics of H2 are concerned, the dynamical transient state approximates the transition-state of the heterolytic H2-cleavage. The culprit is the concerted interactions of H2 with C(carbonyl) and a number of Lewis basic solvent molecules-i.e., the concerted C(carbonyl)â¯H2â¯solvent interactions. On one hand, the results presented herein complement the mechanistic insight gained from our recent transition-state calculations, reported separately from this article. But on the other hand, we now indicate that an idea of the sufficiency of just one simple reaction coordinate in solution-phase reactions can be too simplistic and misleading. This article goes in the footsteps of the rapidly strengthening approach of investigating molecular interactions in large molecular systems via "computational experimentation" employing, primarily, ab initio molecular dynamics describing reactants-interaction without constraints of the preordained reaction coordinate and/or foreknowledge of the sampling order parameters.
ABSTRACT
Using Born-Oppenheimer molecular dynamics (BOMD) with density functional theory, transition-state (TS) calculations, and the quantitative energy decomposition analysis (EDA), we examined the mechanism of H2-liberation from LB-H(+) + (-)H-LA ion-pair, 1, in which the Lewis base (LB) is (o-C6H4Me)3P and the Lewis acid (LA) is B(p-C6F4H)3. BOMD simulations indicate that the path of H2 liberation from the ion-pair 1 goes via the short-lived transient species, LBâ¯H2â¯LA, which are structurally reminiscent of the TS-structure in the minimum-energy-path describing the reversible reaction between H2 and (o-C6H4Me)3P/B(p-C6F4H)3 frustrated Lewis pair (FLP). With electronic structure calculations performed on graphics processing units, our BOMD data-set covers more than 1 ns of evolution of the ion-pair 1 at temperature T ≈ 400 K. BOMD simulations produced H2-recombination events with various durations of H2 remaining fully recombined as a molecule within a LB/LA attractive "pocket"-from very short vibrational-time scale to time scales in the range of a few hundred femtoseconds. With the help of perturbational approach to trajectory-propagation over a saddle-area, we directly examined dynamics of H2-liberation. Using EDA, we elucidated interactions between the cationic and anionic fragments in the ion-pair 1 and between the molecular fragments in the TS-structure. We have also considered a model that qualitatively takes into account the potential energy characteristics of H-H recombination and H2-release plus inertia of molecular motion of the (o-C6H4Me)3P/B(p-C6F4H)3 FLP.
ABSTRACT
Brønsted acidification of water by Lewis acid (LA) complexation is one of the fundamental principles in chemistry. Using transition-state calculations (TS), herein we investigate the role that Brønsted acidification of the LA-bound water might play in the mechanism of the hydrogenation of carbonyl compounds in Lewis basic solvents under non-anhydrous conditions. The potential energy scans and TS calculations were carried out with a series of eight borane LAs as well as the commonly known strong LA AlCl3 in 1,4-dioxane or THF as Lewis basic solvents. Our molecular model consists of the dative LA-water adduct with hydrogen bonds to acetone and a solvent molecule plus one additional solvent molecule that participates is the TS structure describing the cleavage of H2 at acetone's carbonyl carbon atom. In all the molecular models applied here, acetone (O=CMe2 ) is the archetypical carbonyl substrate. We demonstrate that Brønsted acidification of the LA-bound water can indeed lower the barrier height of the solvent-involving H2 -cleavage at the acetone's carbonyl carbon atom. This is significant because at present it is believed that the mechanism of the herein considered reaction is described by the same mechanism regardless of whether the reaction conditions are strictly anhydrous or non-anhydrous. Our results offer an alternative to this belief that warrants consideration and further study.
ABSTRACT
By using transition-state (TS) calculations, we examined how Lewis acid (LA) complexation activates carbonyl compounds in the context of hydrogenation of carbonyl compounds by H2 in Lewis basic (ethereal) solvents containing borane LAs of the type (C6 F5 )3 B. According to our calculations, LA complexation does not activate a ketone sufficiently enough for the direct addition of H2 to the O=C unsaturated bond; but, calculations indicate a possibly facile heterolytic cleavage of H2 at the activated and thus sufficiently Lewis acidic carbonyl carbon atom with the assistance of the Lewis basic solvent (i.e., 1,4-dioxane or THF). For the solvent-assisted H2 splitting at the carbonyl carbon atom of (C6 F5 )3 B adducts with different ketones, a number of TSs are computed and the obtained results are related to insights from experiment. By using the Born-Oppenheimer molecular dynamics with the DFT for electronic structure calculations, the evolution of the (C6 F5 )3 B-alkoxide ionic intermediate and the proton transfer to the alkoxide oxygen atom were investigated. The results indicate a plausible hydrogenation mechanism with a LA, that is, (C6 F5 )3 B, as a catalyst, namely, 1)â the step of H2 cleavage that involves a Lewis basic solvent molecule plus the carbonyl carbon atom of thermodynamically stable and experimentally identifiable (C6 F5 )3 B-ketone adducts in which (C6 F5 )3 B is the "Lewis acid promoter", 2)â the transfer of the solvent-bound proton to the oxygen atom of the (C6 F5 )3 B-alkoxide intermediate giving the (C6 F5 )3 B-alcohol adduct, and 3)â the SN 2-style displacement of the alcohol by a ketone or a Lewis basic solvent molecule.
ABSTRACT
A ketone's carbonyl carbon is electrophilic and harbors a part of the lowest unoccupied molecular orbital of the carbonyl group, resembling a Lewis acidic center; under the right circumstances it exhibits very useful chemical reactivity, although the natural electrophilicity of the ketone's carbonyl carbon is often not strong enough on its own to produce such reactivity. Quantum chemical calculations predict that a proton shared between a ketone and the Lewis basic solvent molecule (dioxane or THF) activates carbonyl carbon to the point of enabling a facile heterolytic splitting of H2 . Proton-catalyzed hydrogenation of a ketone in Lewis basic solvent is the result. The mechanism involves the interaction of H2 with the enhanced Lewis acidity of a carbonyl carbon and the free Lewis basic solvent molecule polarizes H2 and enables the hydride-type attack on carbonyl carbon, which is very strongly influenced by the proton shared between a ketone and solvent. The hydride-type attack on carbon is reminiscent of the splitting of H2 by singlet carbenes except that, in this case, a Lewis base from the surrounding environment (solvent) is necessary for polarization of H2 and acceptance of the proton resulting from the heterolytic splitting of H2 .
ABSTRACT
The role solvent plays in reactions involving frustrated Lewis pairs (FLPs)-for example, the stoichiometric mixture of a bulky Lewis acid and a bulky Lewis base-still remains largely unexplored at the molecular level. For a reaction of the phosphorus/boron FLP and dissolved CO2 gas, first principles (Born-Oppenheimer) molecular dynamics with explicit solvent reveals a hitherto unknown two-step reaction pathway-one that complements the concerted (one-step) mechanism known from the minimum-energy-path calculations. The rationalization of the discovered reaction pathway-that is, the stepwise formation of PC and OB bonds-is that the environment (typical organic solvents) stabilizes an intermediate which results from nucleophilic attack of the phosphorus Lewis base on CO2 . This finding is significant because presently the concerted reaction-path paradigm predominates in the rationalization of FLP reactivity. Herein we point out how to attain experimental proof of our results.
ABSTRACT
We investigate the transition-state (TS) region of the potential energy surface (PES) of the reaction tBu3P + H2 + B(C6F5)3 â tBu3P-H(+) + (-)H-B(C6F5)3 and the dynamics of the TS passage at room temperature. Owing to the conformational inertia of the phosphaneâ â â borane pocket involving heavy tBu3P and B(C6F5)3 species and features of the PES E(Pâ â â H, Bâ â â H | Bâ â â P) as a function of Pâ â â H, Bâ â â H, and Bâ â â P distances, a typical reactive scenario for this reaction is a trajectory that is trapped in the TS region for a period of time (about 350 fs on average across all calculated trajectories) in a quasi-bound state (scattering resonance). The relationship between the timescale of the TS passage and the effective conformational inertia of the phosphaneâ â â borane pocket leads to a prediction that isotopically heavier Lewis base/Lewis acid pairs and normal counterparts could give measurably different reaction rates. Herein, the predicted quasi-bound state could be verified in molecular collision experiments involving femtosecond spectroscopy.
ABSTRACT
With the help of state-of-the-art ab initio molecular dynamics methods, we investigated the reaction pathway of the {tBu3 P + H2 + B(C6 F5 )3 } system at the mesoscopic level. It is shown that: i) the onset of H2 activation is at much larger boronâ â â phosphorus distances than previously thought; ii) the system evolves to the product in a roaming-like fashion because of quasi-periodic nuclear motion along the asymmetric normal mode of Pâ â â HHâ â â B fragment; iii) transient configurations of a certain type are present despite structural interference from the solvent; iv) transient-state configurations with sub-picosecond lifetime have potentially interesting infrared activity in the organic solvent (toluene) as well as in the gas phase. The presented results should be helpful for future experimental and theoretical studies of frustrated Lewis pair (FLP) activity.
ABSTRACT
The role of the intra- and intermolecular motion, i.e., molecular vibrations and the relative motion of reactants, remains largely unexplored in the frustrated Lewis acid/base chemistry. Here, we address the issue with the ab initio molecular dynamics (AIMD) study of CO2 binding by a Lewis acid (LA) and a Lewis base (LB), i.e., tBu3P + CO2 + B(C6F5)3 â tBu3P-C(O)O-B(C6F5)3 ([1]). Reasonably large ensemble of AIMD trajectories propagated at 300 K from structures in the saddle region as well as trajectories propagated directly from the reactants region revealed an effect arising from significant recrossing of the saddle area. The effect is that transient complexes composed of weakly interacting reactants nearly cease to progress along the segment of the minimum energy pathway (MEP) at the saddle region for a (subpicosecond) period of time during which the dominant factor is the light-to-heavy type of relative motion of the vibrating reactants, i.e., the "bouncing"-like movement of CO2 with respect to much heavier phosphine and borane as main contributor to the mode that is perpendicular to the MEP-direction. In terms of how P···C and B···O distances change with time, the roaming-like patterns of typical AIMD trajectories, reactive and nonreactive alike, extend far beyond the saddle region. In addition to the dynamical portrayal of [1], we provide the energy-landscape perspective that takes into account the hierarchy of time scales. The verifiable implication of the effect found here is that the isotopically substituted (heavier) LB/LA "pair" should be less reactive that the "normal" and thus lighter counterpart.
ABSTRACT
Activation of H2 by a "frustrated Lewis pair" (FLP) composed of B(C6F5)3 and P(tBu)3 species has been explored with high level direct ab initio molecular dynamics (AIMD) simulations at finite temperature (T = 300 K) in gas phase. The initial geometrical conditions for the AIMD trajectory calculations, i.e., the near attack conformations of FLP + H2, were devised using the host-guest model in which suitable FLP conformations were obtained from the dynamics of the B(C6F5)3∕P(tBu)3 pair in gas phase. AIMD trajectory calculations yielded microscopic insight into effects which originate from nuclear motion in the reacting complex, e.g., the alternating compression∕elongation of the boron-phosphorous distance and the change of the pyramidality of boron in B(C6F5)3. The ensemble averaged trajectory analysis has been compared with the minimum energy path (MEP) description of the reaction. Similar to MEP, AIMD shows that an attack of the acid∕base pair on the H-H bond gives rise to the polarization of the H2 molecule and as a consequence generates a large dipole moment of the reacting complex. The MEP and AIMD portrayals of the reaction are fundamentally different in terms of the magnitude of the motion of nuclei in B(C6F5)3 and P(tBu)3 during the H2 cleavage. In the AIMD trajectory simulations, geometries of B(C6F5)3 and P(tBu)3 appear as nearly "frozen" on the short time scale of the H2 cleavage. This is contrary to the MEP picture. Several of the concepts which arise from this work, e.g., separation of time scales of nuclear motion and the time-dependence of the donor-acceptor interactions in the reacting complex, are important for the understanding of chemical reactivity and catalysis.
ABSTRACT
The formation of an active 16-electron ruthenium sec-alkoxide complex via loss of the CO ligand is an important step in the mechanism of the racemization of sec-alcohols by (η(5)-Ph(5)C(5))Ru(CO)(2)X ruthenium complexes with X = Cl and O(t)Bu. Here we show with accurate DFT calculations the potential energy profile of the CO dissociation pathway for a series of relevant (η(5)-Ph(5)C(5))Ru(CO)(2)X complexes, where X = Cl, O(t)Bu, H and COO(t)Bu. We have found that the CO dissociation energy increases in the following order: O(t)Bu (lowest), Cl, COO(t)Bu and H (highest). Using the distance between ruthenium and C(CO), r = Ru-C(CO), as a constraint, and by optimizing all other degrees of freedom for a range of Ru-CO distances, we obtained relative energies, ΔE(r) and geometries of a sufficient number of transient structures with the elongated Ru-CO bond up to r = 3.4 Å. Our calculations provide a quantitative understanding of the CO ligand dissociation in (η(5)-Ph(5)C(5))Ru(CO)(2)Cl and (η(5)-Ph(5)C(5))Ru(CO)(2)(O(t)Bu) complexes, which is relevant to the mechanism of their catalytic activity in the racemization of alcohols. We recently reported that exchange of the CO ligand by isotopically labeled (13)CO in the Ru-O(t)Bu complex occurs twenty times faster than that in the Ru-Cl complex. This corresponds to a difference of 1.8 kcal mol(-1) in the CO dissociation energy (at room temperature). This is in very good agreement with the calculated difference between the two potential energy curves for Ru-O(t)Bu and Ru-Cl complexes, which is about 1.8-2 kcal mol(-1) around the corresponding transition states of the CO dissociation. The calculated difference in the total energy for CO dissociation in (η(5)-Ph(5)C(5))Ru(CO)(2)X complexes is related to the stabilization provided by the X group in the final 16-electron complexes, which are formed via product-like transition states. In addition to the calculated transition states of CO dissociation in Ru-O(t)Bu and Ru-Cl complexes, the calculated transient structures with the elongated Ru-CO bond provide insight into how the geometry of the ruthenium complex with a potent heteroatom donor group (X) gradually changes when one of the COs is dissociating.
Subject(s)
Alcohols/chemistry , Carbon Monoxide/chemistry , Coordination Complexes/chemistry , Ruthenium/chemistry , Catalysis , Models, Chemical , Molecular Conformation , ThermodynamicsABSTRACT
Using the combinations of imidazole and dimethyl sulfoxide (DMSO) as axial ligands and 2,2'-bipyridine-6,6'-dicarboxylate (bda) as the equatorial ligand, we have synthesized six novel ruthenium complexes with noticeably different activity as water oxidation catalysts (WOCs). In four C(s) symmetric Ru(II)(κ(3)-bda)(DMSO)L(2) complexes L = imidazole (1), N-methylimidazole (2), 5-methylimidazole (3), and 5-bromo-N-methylimidazole (4). Additionally, in two C(2v) symmetric Ru(II)(κ(4)-bda)L(2) complexes L = 5-nitroimidazole (5) and 5-bromo-N-methylimidazole (6), that is, fully equivalent axial imidazoles. A detailed characterization of all complexes and the mechanistic investigation of the catalytic water oxidation have been carried out with a number of experimental techniques, that is, kinetics, electrochemistry and high resolution mass spectrometry (HR-MS), and density functional theory (DFT) calculations. We have observed the in situ formation of a Ru(II)-complex with the accessible seventh coordination position. The measured catalytic activities and kinetics of complex 1-6 revealed details about an important structure-activity relation: the connection between the nature of axial ligands in the combination and either the increase or decrease of the catalytic activity. In particular, an axial DMSO group substantially increases the turnover frequency of WOCs reported in the article, with the ruthenium-complex having one axial 5-bromo-N-methyl-imidazole and one axial DMSO (4), we have obtained a high initial turnover frequency of â¼180 s(-1). DFT modeling of the binuclear reaction pathway of the O-O bond formation in catalytic water oxidation further corroborated the concept of the mechanistic significance of the axial ligands and rationalized the experimentally observed difference in the activity of complexes with imidazole/DMSO and imidazole/imidazole combinations of axial ligands.
Subject(s)
Dimethyl Sulfoxide/chemistry , Imidazoles/chemistry , Organometallic Compounds/chemistry , Ruthenium/chemistry , Water/chemistry , Catalysis , Crystallography, X-Ray , Ligands , Models, Molecular , Molecular Structure , Organometallic Compounds/chemical synthesis , Oxidation-ReductionABSTRACT
Across chemical disciplines, an interest in developing artificial water splitting to O(2) and H(2), driven by sunlight, has been motivated by the need for practical and environmentally friendly power generation without the consumption of fossil fuels. The central issue in light-driven water splitting is the efficiency of the water oxidation, which in the best-known catalysts falls short of the desired level by approximately two orders of magnitude. Here, we show that it is possible to close that 'two orders of magnitude' gap with a rationally designed molecular catalyst [Ru(bda)(isoq)(2)] (H(2)bda = 2,2'-bipyridine-6,6'-dicarboxylic acid; isoq = isoquinoline). This speeds up the water oxidation to an unprecedentedly high reaction rate with a turnover frequency of >300 s(-1). This value is, for the first time, moderately comparable with the reaction rate of 100-400 s(-1) of the oxygen-evolving complex of photosystem II in vivo.