First principles static and dynamic calculations for the transition metal hydride series MH4L3 (M = Fe, Ru and Os; L = NH3, PH3 and PF3)

We present a first principles static and dynamical study of the transition metal hydride series MH4L3 (M = Fe, Ru and Os; L = NH3, PH3 and PF3), with a view to arriving at an understanding of how the variation in the electronic properties of the metal sites and ligands can influence the dynamics of the resulting complexes. A broad range of behaviour was observed, encompassing stable classical minima (M = Os, L = NH3 and M = Ru, L = PH3) to stable η(2)-H2 non-classical minima (M = Fe, L = PF3 and M = Ru, L = PH3 or PF3), with the other structures exhibiting dynamical behaviour that spontaneously converted between the classical and non-classical states during the molecular dynamics simulations. The importance of a small L(axial)-M-L(axial) angle in stabilising the non-classical state is highlighted, as is a short η(2)-H2···H(cis) distance in non-classical complexes that spontaneously convert to the classical form. We also investigated the changes in the electronic structure of the complex FeH4(PH3)3 during a η(2)-H2 bond breaking/bond making reaction and observed direct evidence of the 'cis effect', whereby a neighbouring hydride ligand acts to stabilise the intermediate classical state.

Dihydrogen complexes as prototypes for the coordination chemistry of saturated molecules

The binding of a dihydrogen molecule (H(2)) to a transition metal center in an organometallic complex was a major discovery because it changed the way chemists think about the reactivity of molecules with chemically "inert" strong bonds such as H H and C H. Before the seminal finding of side-on bonded H(2) in W(CO)(3)(PR(3))(2)(H(2)), it was generally believed that H(2) could not bind to another atom in stable fashion and would split into two separate H atoms to form a metal dihydride before undergoing chemical reaction. Metal-bound saturated molecules such as H(2), silanes, and alkanes (sigma-complexes) have a chemistry of their own, with surprisingly varied structures, bonding, and dynamics. H(2) complexes are of increased relevance for H(2) production and storage in the hydrogen economy of the future.

Heterolytic Activation of H−X (X = H, Si, B, and C) Bonds: An Experimental and Theoretical Investigation

The highly electrophilic, coordinatively unsaturated, 16-electron [Ru(P(OH)3)(dppe)2][OTf]2 (dppe = Ph2PCH2CH2PPh2) complex 1 activates the H-H, the Si-H, and the B-H bonds, in H2(g), EtMe2SiH and Et3SiH, and H3B.L (L = PMe3, PPh3), respectively, in a heterolytic fashion. The heterolysis of H2 involves an eta2-H2 complex (observable at low temperatures), whereas the computations indicate that those of the Si-H and the B-H bonds proceed through unobserved eta1-species. The common ruthenium-containing product in these reactions is trans-[Ru(H)(P(OH)3)(dppe)2][OTf], 2. The [Ru(P(OH)3)(dppe)2][OTf]2 complex is unique with regard to activating the H-H, the Si-H, and the B-H bonds in a heterolytic manner. These reactions and the heterolytic activation of the C-H bond in methane by the model complex [Ru(POH)3)(H2PCH2CH2PH2)2][Cl][OTf], 4, have been investigated using computational methods as well, at the B3LYP/LANL2DZ level. While the model complex activates the H-H, the Si-H, and the B-H bonds in H2, SiH4, and H3B.L (L = PMe3, PPh3), respectively, with a low barrier, activation of the C-H bond in CH4 involves a transition state of 57.5 kcal/mol high in energy. The inability of the ruthenium complex to activate CH4 is due to the undue stretching of the C-H bond needed at the transition state, in comparison to the other substrates.

Mechanistic insight into hydrosilylation reactions catalyzed by high valent ReX (X = O, NAr, or N) complexes: the silane (Si-H) does not add across the metal-ligand multiple bond

Treatment of oxo and imido-rhenium(V) complexes Re(X)Cl3(PR3)2 (X = O, NAr, and R = Ph or Cy) (1-2) with Et3SiH affords Re(X)Cl2(H)(PR3)2 in high yields. Cycloaddition of silane across the ReX multiple bonds is not observed. Two rhenium(V) hydrides (X = O and R = Ph, 4a; X = NMes and R = Ph, 5a) have been structurally characterized by X-ray diffraction. The kinetics of the reaction of Re(O)Cl3(PPh3)2 (1a) with Et3SiH is characterized by phosphine inhibition and saturation in [Et3SiH]. Hence, formation of Re(O)Cl2(H)(PPh3)2 (4a) proceeds via a sigma-adduct followed by heterolytic cleavage of the Si-H bond and transfer of silylium (Et3Si+) to chloride. Oxo and imido complexes of rhenium(V) (1-2) as well as their nitrido analogues, Re(N)Cl2(PR3)2 (3), catalyze the hydrosilylation of PhCHO under ambient conditions, with the reactivity order imido > oxo > nitrido. The isolable oxorhenium(V) hydride 4a reacts with PhCHO to afford the alkoxide Re(O)Cl2(OCH2Ph)(PPh3)2 (6a) with kinetic dependencies that are consistent with aldehyde coordination followed by aldehyde insertion into the Re-H bond. The latter (6a) regenerates the rhenium hydride upon reaction with Et3SiH. These stoichiometric reactions furnish a possible catalytic cycle. However, quantitative kinetic analysis of the individual stoichiometric steps and their comparison to steady-state kinetics of the catalytic reaction reveal that the observed intermediates do not account for the predominant catalytic pathway. Furthermore, for Re(O)Cl2(H)(PCy3)2 and Re(NMes)Cl2(H)(PPh3)2 aldehyde insertion into the Re-H bond is not observed. Therefore, based on the kinetic dependencies under catalytic conditions, a consensus catalytic pathway is put forth in which silane is activated via sigma-adduct formation cis to the ReX bond followed by heterolytic cleavage at the electrophilic rhenium center. The findings presented here demonstrate the so-called Halpern axiom, the observation of "likely" intermediates in a catalytic cycle, generally, signals a nonproductive pathway.

Orbital landscapes for reductive 2e− activation of dihydrogen molecule

The various spatial arrangements of frontier orbitals that may lead to facile reductive splitting of the H2 molecule at mono- or binuclear catalysts containing s, p, d or f-block metals, and on surfaces of solids are briefly reviewed. The postulation is also made that binuclear divalent titanium (Ti(II)) and mononuclear silicon (Si(II)) species might serve as active sites for the H2 attachment reaction for hydridoalanates doped with Ti salts and hydridoborates doped with SiO2, respectively.

NMR Investigation of the Dihydrogen-Bonding and Proton-Transfer Equilibria between the Hydrido Carbonyl Anion [HRe2(CO)9]− and Fluorinated Alcohols

The interaction of fluorinated alcohols with the anionic hydrido complex [HRe2(CO)9]- (1) has been investigated by NMR spectroscopy. According to the acidic strength of the alcohols, the interaction may result not only in the formation of dihydrogen-bonded ROH...[HRe2(CO)9]- adducts 2, but also in proton transfer to give the neutral species [H2Re2(CO)9] (3). With the weaker acid trifluoroethanol (TFE) evidence for the occurrence of the dihydrogen-bonding equilibrium was obtained by 2D 1H NOESY. The dependence of the hydride chemical shift on TFE concentration at different temperatures provided values for the constants of this equilibrium, from which the thermodynamic parameters were evaluated as deltaH(degrees) = -2.6(2) kcal mol(-1), deltaS(degrees) = -9.3(2) cal mol(-1) K(-1). This corresponds to a rather low basicity factor (E(j) = 0.64). Variable-temperature T1 measurements allowed the proton-hydride distance in adduct 2 a to be estimated (1.80 angstroms). In the presence of hexafluoroisopropyl alcohol (HFIP) simultaneous occurrence of both dihydrogen-bonding and proton-transfer equilibria was observed, and the equilibria shifted versus the protonated product 3 with increasing HFIP concentration and decreasing temperature. Reversible proton transfer between the alcohol and the hydrido complex occurs on the NMR timescale, as revealed by a 2D 1H EXSY experiment at 240 K. For the more acidic perfluoro-tert-butyl alcohol (PFTB) the protonation equilibrium was further shifted to the right. Thermal instability of 3 prevented the acquisition of accurate thermodynamic data for these equilibria. The occurrence of the proton-transfer processes (in spite of the unfavorable pK(a) values) can be explained by the formation of homoconjugated RO...HOR- pairs which stabilize the alcoholate anions.

Electronic and Steric Effects on Molecular Dihydrogen Activation in [Cp*OsH4(L)]+ (L = PPh3, AsPh3, and PCy3)

Single-crystal neutron diffraction, inelastic neutron scattering, and density functional calculations provide experimental and theoretical analyses of the nature of the osmium-bound, "elongated" dihydrogen ligands in [Cp*OsH(4)(L)][BF(4)] complexes (L = PPh(3), AsPh(3), or PCy(3)). The PPh(3) and AsPh(3) complexes clearly contain one dihydrogen ligand and two terminal hydrides; the H(2) ligand is transoid to the Lewis base, and the H-H vector connecting the central two hydrogen atoms lies parallel to the Ct-Os-L plane (Ct = centroid of Cp* ring). In contrast, in the PCy(3) complex the H-H vector is perpendicular to the Ct-Os-L plane. Not only the orientation of the central two hydrogen atoms but also the H-H bond length between them depends significantly on the nature of L: the H...H distance determined from neutron diffraction is 1.01(1) and 1.08(1) A for L = PPh(3) and AsPh(3), respectively, but 1.31(3) A for L = PCy(3). Density functional calculations show that there is a delicate balance of electronic and steric influences created by the L ligand that change the molecular geometry (steric interactions between the Cp* and L groups most importantly change the Ct-Os-L angle), changing the relative energy of the Os 5d orbitals, which in turn govern the H-H distance, preferred H-H orientation, and rotational dynamics of the elongated dihydrogen ligand. The geometry of the dihydrogen ligand is further tuned by interactions with the BF(4)(-) counterion. The rotational barrier of the bound H(2) ligand in [Cp*OsH(4)(PPh(3))](+), determined experimentally (3.1 kcal mol(-)(1)) from inelastic neutron scattering experiments, is in reasonable agreement with the B3LYP calculated H(2) rotational barrier (2.5 kcal mol(-)(1)).

Highly Electrophilic, 16-Electron [Ru(P(OMe)(OH)2)(dppe)2]2+ Complex Turns H2(g) into a Strong Acid and Splits a Si−H Bond Heterolytically. Synthesis and Structure of the Novel Phosphorous Acid Complex [Ru(P(OH)3)(dppe)2]2+

The highly electrophilic, 16-electron, coordinatively unsaturated [Ru(P(OMe)(OH)(2))(dppe)(2)][OTf](2) complex brings about the heterolytic activation of H(2)(g) and spontaneously generates HOTf. In addition, trans-[Ru(H)(P(OMe)(OH)(2))(dppe)(2)](+) and an unprecedented example of a phosphorous acid complex, [Ru(P(OH)(3))(dppe)(2)](2+), are formed. The [Ru(P(OMe)(OH)(2))(dppe)(2)][OTf](2) complex also cleaves the Si-H bond in EtMe(2)SiH in a heterolytic fashion, resulting in the trans-[Ru(H)(P(OMe)(OH)(2))(dppe)(2)](+) derivative.

Effect of weak interactions on the H...H distance in stretched dihydrogen complexes

Computational and experimental results presented in this paper demonstrate that the H-H distance in stretched dihydrogen complexes can be hypersensitive to a variety of weak intra- and intermolecular interactions, including those with bulky ligands and solvent molecules, hydrogen-bonding interactions, or ion-pairing. Particularly, the complex IrH(H...H)Cl(2)(P(i)Pr(3))(2) which contains a stretched dihydrogen ligand in the crystalline form, as shown by neutron diffraction, is a trihydride in solution. The difference is due to the intermolecular Ir-Cl...H-Ir hydrogen bonding in the solid.

Trans --> cis isomerization of trans-[(dppm)2Ru(H)(L)][BF4] (L = P(OR)3) complexes: preparation of cis-[(dppm)2Ru(eta2-H2)(L)][BF4]2

A series of new dicationic dihydrogen complexes of ruthenium of the type cis-[(dppm)(2)Ru(eta(2)-H(2))(L)][BF(4)](2) (dppm = Ph(2)PCH(2)PPh(2); L = P(OMe)(3), P(OEt)(3), PF(O(i)Pr)(2)) have been prepared by protonating the precursor hydride complexes cis-[(dppm)(2)Ru(H)(L)][BF(4)] (L = P(OMe)(3), P(OEt)(3), P(O(i)Pr)(3)) using HBF(4).Et(2)O. The cis-[(dppm)(2)Ru(H)(L)][BF(4)] complexes were obtained from the trans hydrides via an isomerization reaction that is acid-accelerated. This isomerization reaction gives mixtures of cis and trans hydride complexes, the ratios of which depend on the cone angles of the phosphite ligands: the greater the cone angle, the greater is the amount of the cis isomer. The eta(2)-H(2) ligand in the dihydrogen complexes is labile, and the loss of H(2) was found to be reversible. The protonation reactions of the starting hydrides with trans PMe(3) or PMe(2)Ph yield mixtures of the cis and the trans hydride complexes; further addition of the acid, however, give trans-[(dppm)(2)Ru(BF(4))Cl]. The roles of the bite angles of the dppm ligand as well as the steric and the electronic properties of the monodentate phosphorus ligands in this series of complexes are discussed. X-ray crystal structures of trans-[(dppm)(2)Ru(H)(P(OMe)(3))][BF(4)], cis-[(dppm)(2)Ru(H)(P(OMe)(3))][BF(4)], and cis-[(dppm)(2)Ru(H)(P(O(i)Pr)(3))][BF(4)] complexes have been determined.

Mechanism of asymmetric hydrogenation of alpha-(acylamino)acrylic esters catalyzed by BINAP-ruthenium(II) diacetate

The mechanism of asymmetric hydrogenation of alpha-(acylamino)acrylic esters with Ru(CH(3)COO)(2)[(S)-binap] (BINAP = 2,2'-bis(diphenylphosphino)-1,1'-binaphthyl), giving the S saturated products in >90% ee, has been investigated by means of a kinetic study, deuterium labeling experiments, isotope effect measurements, and NMR and X-ray analysis of certain Ru complexes. The hydrogenation in methanol under a low H2 pressure proceeds via a monohydride-unsaturate mechanism that involves the initial RuH formation followed by a reaction with an olefinic substrate. The migratory insertion in the enamide-RuH chelate complex occurs reversibly and endergonically in an exo manner, giving a five-membered metallacycle intermediate. The cleavage of the Ru-C bond is achieved with either H2 (major) or CH3OH (minor). Both of the pathways result in overall cis hydrogenation products. The hydrogen at C3 is mainly from an H2 molecule, and the C2 hydrogen is from another H2 or protic CH3OH. The major S and minor R enantiomers are produced via the same mechanism involving diastereomeric intermediates. The turnover rate is limited by the step of hydrogenolysis of a half-hydrogenated metallacyclic intermediate. The participation of two different hydrogen donor molecules is in contrast to the pairwise dihydrogenation using a single H2 molecule in the RhI-catalyzed reaction which occurs via a dihydride mechanism. In addition, the sense of asymmetric induction is opposite to that observed with S-BINAP-RhI catalysts. The origin of this phenomenon is interpreted in terms of stereocomplementary models of the enamide/metal chelate complexes. A series of model stoichiometric reactions mimicking the catalytic steps has indicated that most NMR-observable Ru complexes are not directly involved in the catalytic hydrogenation but are reservoirs of real catalytic complexes or even side products that retard the reaction.

NMR and DFT analysis of [Re(2)H(2)(CO)(9)]: evidence of an eta(2)-H(2) intermediate in a new type of fast mutual exchange between terminal and bridging hydrides

Protonation of the anion [Re(2)H(CO)(9)](-) (1) with a strong acid at 193 K affords the neutral complex [Re(2)H(2)(CO)(9)] (2), that in THF above 253 K irreversibly loses H(2) to give [Re(2)(CO)(9)(THF)], previously obtained by room-temperature protonation of 1. Treatment of 2 with NEt(4)OH restores the starting anion 1. Variable temperature (1)H and (13)C NMR spectra as well as T(1) measurements agree with the formulation of 2 as a classical [Re(2)H(mu-H)(CO)(9)] complex, in which two dynamic processes takes place. The "windshield-wiper motion" observed in several related complexes equalizes the two carbonyls trans to the hydrides (E(a) = 44(1) kJ mol(-)(1)), while another much faster process equalizes bridging and terminal hydrides already at 172 K. The variable temperature behavior of the (1)H transverse relaxation times revealed also proton exchange between 2, water, and the parent anion 1 (due to the acidity of 2), but such a process is too slow to account for the fast hydrides exchange in 2. The nature of the latter process has been investigated both experimentally and theoretically. Kinetic data, obtained by the analysis of the variable temperature (1)H spectra (E(a) = 24.5(5) kJ mol(-1)), revealed a small normal kinetic isotope effect (ca. 1.5). The (2)H chemical shift of the fully deuterated isotopomer 2-d(2) was found isochronous with 2, thus ruling out the presence of a significant concentration of a nonclassical [Re(2)(eta(2)-H(2))(CO)(9)] tautomer, in fast exchange with the classical dihydride. Density functional theory (DFT) calculations, carried out at the B3LYP level, confirmed the formulation of [Re(2)H(2)(CO)(9)] as a classical complex. However, when DFT was used to obtain a detailed description of the dynamic behavior of 2 in solution, a new type of hydride fast exchange emerged, involving the nonclassical tautomer as a relatively high energy (12.7 kJ mol(-1)) intermediate. Isotopic perturbation of the equilibrium by partial deuteration of 2 indicated the preference of deuterium for the bridging sites, with Delta H degrees = -475(4) J mol(-1) and Delta S degrees = -0.80(2) J K(-1) mol(-1). The same preference was observed in the anion [Re(2)H(mu-H)Cl(CO)(8)](-).