Hydrogen Generation from Alcohols Catalyzed by Ruthenium−Triphenylphosphine Complexes: Multiple Reaction Pathways

Navarro R. M., 2007, Chem. Rev., 107, 3952, 10.1021/cr0501994

Cortright R. D., 2002, Nature, 418, 964, 10.1038/nature01009

Johnson T. C., 2010, Chem. Soc. Rev., 39, 81, 10.1039/B904495G

Charman H. B., 1970, J. Chem. Soc. B, 584, 10.1039/J29700000584

Delgado-Lieta E., 1982, Polyhedron, 1, 839, 10.1016/0277-5387(82)80020-X

Dobson A., 1977, Inorg. Chem., 16, 137, 10.1021/ic50167a029

Morton D., 1987, J. Chem. Soc., Chem. Commun., 248, 10.1039/C39870000248

Morton D., 1987, Polyhedron, 6, 2187, 10.1016/S0277-5387(00)84285-0

Rybak W. K., 1981, J. Mol. Catal., 11, 365, 10.1016/0304-5102(81)87024-1

Jung C. W., 2002, Organometallics, 1, 658, 10.1021/om00064a016

Ligthart G. B. W. L., 2003, Tetrahedron Lett., 44, 1507, 10.1016/S0040-4039(02)02842-3

Shinoda S., 1985, J. Chem. Soc., Chem. Commun., 860, 10.1039/c39850000860

Itagaki H., 1988, Bull. Chem. Soc. Jpn., 61, 2291, 10.1246/bcsj.61.2291

Fujii T., 1991, J. Mol. Catal., 67, 185, 10.1016/0304-5102(91)85045-4

Morton D., 1988, J. Chem. Soc., Chem. Commun., 1154, 10.1039/c39880001154

Morton D., 1989, J. Chem. Soc., Dalton Trans., 489, 10.1039/dt9890000489

Adair G. R. A., 2005, Tetrahedron Lett., 46, 8233, 10.1016/j.tetlet.2005.09.083

Zhanga J., 2007, Dalton Trans., 107, 10.1039/B613438F

Junge H., 2005, Tetrahedron Lett., 46, 1031, 10.1016/j.tetlet.2004.11.159

Junge H., 2007, Chem. Commun., 522, 10.1039/B613785G

Yang L.-C., 1996, J. Mol. Catal. A, 108, 87, 10.1016/1381-1169(96)00003-9

Fordyce W. A., 1985, J. Organomet. Chem., 296, 115, 10.1016/0022-328X(85)80343-0

Gusev D. G., 1991, Inorg. Chim. Acta, 179, 195, 10.1016/S0020-1693(00)85878-8

Halpern J., 1987, Pure Appl. Chem., 59, 173, 10.1351/pac198759020173

Linn D. E., 1987, J. Am. Chem. Soc., 109, 2969, 10.1021/ja00244a019

Itagaki H., 1993, Organometallics, 12, 1648, 10.1021/om00029a024

Joubert J., 2006, J. Organomet. Chem., 691, 1030, 10.1016/j.jorganchem.2005.11.037

Joubert J., 2006, Organometallics, 25, 854, 10.1021/om050782a

Rossin A., 2006, Organometallics, 25, 5010, 10.1021/om060353s

Alonso D. A., 1999, J. Am. Chem. Soc., 121, 9580, 10.1021/ja9906610

Yamakawa M., 2000, J. Am. Chem. Soc., 122, 1466, 10.1021/ja991638h

Handgraaf J.-W., 2007, J. Am. Chem. Soc., 129, 3099, 10.1021/ja062359e

Bacchi A., 2008, Eur. J. Inorg. Chem., 4462, 10.1002/ejic.200800509

Crabtree R. H., 1986, J. Am. Chem. Soc., 108, 3124, 10.1021/ja00271a063

We note that this study has been performed under different experimental conditions, since it involves the [RuCl2(PPh3)3] precursor and has been performed in the absence of base at a temperature of 64 °C, whereas the Cole-Hamilton system is based on [RuH2(X2)(PPh3)3] (X2= N2, PPh3) and involves a small concentration of NaOH at 150 °C.

Chaudret B. N., 1977, J. Chem. Soc., Dalton Trans., 1546, 10.1039/dt9770001546

Van der Sluys L. S., 1991, Organometallics, 10, 1033, 10.1021/om00050a039

Sieffert N., 2009, Inorg. Chem., 48, 4622, 10.1021/ic900347e

Clapham S. E., 2004, Coord. Chem. Rev., 248, 2201, 10.1016/j.ccr.2004.04.007

Grimme S., 2006, J. Comput. Chem., 27, 1787, 10.1002/jcc.20495

Zhao Y., 2007, Org. Lett., 9, 1967, 10.1021/ol0705548

Zhao Y., 2009, J. Chem. Theory Comput., 5, 324, 10.1021/ct800386d

Minenkov Y., 2009, J. Phys. Chem. A, 113, 11833, 10.1021/jp902940c

dSee also our previous study in ref16.

Becke A. D., 1988, Phys. Rev. A, 38, 3098, 10.1103/PhysRevA.38.3098

Perdew J. P., 1986, Phys. Rev. B, 33, 8822, 10.1103/PhysRevB.33.8822

Perdew J. P., 1986, Phys. Rev. B, 34, 7406, 10.1103/PhysRevB.34.7406

Andrae D., 1990, Theor. Chim. Acta, 77, 123, 10.1007/BF01114537

Structures have been determined without corrections for the basis set superposition error (BSSE), since the latter is found to have only a slight influence on geometries. This feature has been investigated in our previous study on a similar tris(triphenylphosphine) complex, and only small structural differences have been obtained when going from ECP1 to the larger ECP2 basis set, where BSSE is expected to be smaller.(16)On the other hand, we note that the influence of the BSSE on reaction energies has been carefully considered herein, and counterpoise energy corrections have been made for each investigated step.

Generated automatically according to the procedure implemented in Gaussian 03.

Hay P. J., 1998, J. Phys. Chem. A, 102, 3565, 10.1021/jp980229p

Gonzalez C., 1989, J. Chem. Phys., 90, 2154, 10.1063/1.456010

Gonzalez C., 1990, J. Phys. Chem., 94, 5523, 10.1021/j100377a021

Cowley R. A., 2005, Acta Crystallogr., Sect. E, 61, m1237, 10.1107/S1600536805016272

Costello J. F., 1998, J. Chem. Soc., Perkin. Trans. 2, 1683, 10.1039/a708014j

Ayscough A. P., 2001, Tetrahedron: Asymmetry, 12, 1621, 10.1016/S0957-4166(01)00253-1

aFollowing the comment of a reviewer, we tested the influence of the density functional on the initiation and activation energies of pathwaysA−Dwith the BP86, B3LYP, BP86-D, B3LYP-D, and M06-L functionals (seeTable S2). We found that B97-D and M06-L give similar activation barriers (within ±2.6 kcal/mol on the average), whereas the traditional BP86 and B3LYP functionals lead to inconsistent results due to a lack of description of noncovalent interactions. BP86-D and B3LYP-D indeed give more consistent results; however, our previous study(16)revealed that B97-D is the best suited functional to model the key PPh3dissociation step. More problematic is the case of H2dissociation, where B97-D leads to a lower dissociation energy compared to M06-L. Comparison with the experimental enthalpy of activation for this process reported by Halpern et al.(30b)shows that, for this particular step, M06-L would be more suited, since the B97-D value is underestimated by ca. 5.2 kcal/mol (ΔH⧧exp≈ 17.9 kcal/mol; ΔH⧧calc= 12.7 kcal/mol; computed by applying the enthalpy correction term δEH= −2.8 kcal/mol). The good agreement between B97-D and M06-L in overall rate-limiting steps shows that this discrepancy is not systematic to B97-D and therefore comforts us in the choice of this functional to model this system

Halpern J., 1991, J. Chem. Soc., Dalton Trans., 717, 10.1039/dt9910000717

Grimme S., 2004, J. Comput. Chem., 25, 1463, 10.1002/jcc.20078

Piacenza M., 2007, J. Comput. Chem., 28, 2275, 10.1002/jcc.20709

Boys S. F., 1970, Mol. Phys., 19, 553, 10.1080/00268977000101561

Klamt A., 1993, J. Chem. Soc., Perkin Trans. 2, 5, 799, 10.1039/P29930000799

Takano Y., 2005, J. Chem. Theor. Comput., 1, 70, 10.1021/ct049977a

Pople, J. A.Gaussian 03, Revision E.01;Gaussian, Inc.,Pittsburgh, PA, 2003.The full reference is given in theSupporting Information.

Ahlrichs R., 1989, Chem. Phys. Lett., 162, 165, 10.1016/0009-2614(89)85118-8

Treutler O., 1995, J. Chem. Phys., 102, 346, 10.1063/1.469408

Arnim M. v., 1998, J. Comput. Chem., 19, 1746, 10.1002/(SICI)1096-987X(19981130)19:15<1746::AID-JCC7>3.0.CO;2-N

Humphrey W., 1996, J. Mol. Graphics, 14, 33, 10.1016/0263-7855(96)00018-5

Seetharaman S. K., 2007, Inorg. Chem., 46, 561, 10.1021/ic061389f

Note that4is found as a result of the IRC calculations fromTS3−4and effectively corresponds to a stationary point on the potential energy surface. However, this complex is not found on the BSSE corrected free energy surface and is therefore expected to be rather unstable.

Chan A. S. C., 1985, J. Chem. Soc., Chem. Commun., 1379, 10.1039/c39850001379

Using the simple Eyring equation, at 423 K such a barrier would correspond to a unimolecular rate constant on the order of 10−2s−1and, thus, to a half-life of a few minutes.

These values are obtained by performing single points on the structures of1bandTS1b-2in which the H2moieties have been removed by hand. The resulting energies are then compared to the optimized energy of2. The same procedure has been applied to compute the deformation energies involved in the H2decoordination in the other pathways, i.e. in3HbandTS3Hb-7, relative to7for pathwayB, and in16HbandTS16Hb-17relative to17for pathwayD. These values are computed in the gas phase at the B97-D/ECP2 level (ΔE).

The MeOH coordination is thermodynamically unfavorable by 2.3 kcal/mol; however, given the large excess of MeOH (solvent) compared to PPh3,16Hbis expected to be accessible under reaction conditions.

The different proton transfer processes involved in pathwaysA-Dare considered to be non-activated. This assumption is based on the results of numerous computational studies on smaller systems, in which the corresponding free energy is found to evolve monotonically as proton transfer occurs. See e.g. the cases of the water self-hydrolysis reaction

Sprik M., 2000, Chem. Phys., 258, 139, 10.1016/S0301-0104(00)00129-4

the deprotonation of solvated formic acid.

Lee J.-G., 2006, J. Phys. Chem. B, 110, 2325, 10.1021/jp055809i

or the deprotonation of an aqueous uranyl metal complex

Bühl M., 2006, ChemPhysChem, 7, 2290, 10.1002/cphc.200600347

However, the presence of such transition states in the particular case of the ruthenium hydride complexes can not be fully ruled out, but their characterization seems hardly feasible by a static/continuum approach. Ab initio molecular dynamics simulations invloving explicit solvent molecules would be required for a meaningful investigation of proton transfers occuring between the solvent and Ru intermediates. Unfortunately, this approach is presently precluded by the size of the considered systems and the related large computational cost.

We note that Delbecq et al.(11b)and Joo et al.(11c)investigated a reverse analogue of pathwayBin which PH3is used as model ligand and in which the presence of the base is not considered. The reaction profiles found in these studies are in qualitative agreement with our results; however, qualitative comparisons are hardly possible, since they considered different phosphine ligands and different substrates.

For the less crowded complex11, we have also explored the possibility of a H transfer via a less strained six-membered ring involving the OH group of a second MeOH molecule. However, we could not find a feasible path for such a concerted, “outer sphere” mechanism (as had been suggested in the case of transfer hydrogenation(17)).

We note that the carbonylated [RuH3(CO)(PPh3)2]−complex has also been characterized, which may be involved in the decarbonylation side reaction occuring in the system under scrutiny.

Delbecq F., 1992, J. Am. Chem. Soc., 114, 2446, 10.1021/ja00033a021

Brown K. L., 1979, J. Am. Chem. Soc., 101, 503, 10.1021/ja00496a050

In a corresponding search in the Cambridge Structural Database considering the M(PC3)2(η2-HCHO) fragment (M being a group 8 metal), only one structure has been found, corresponding to [Os(PPh3)2(CO)2(HCHO)]. Another search considering M(PC3)2(η2-Me2CO) led to no hit, whereas searching the M(PC3)2(η1-Me2CO) fragment led to 30 hits.

In comparison to high-level CCSD(T) or to experimental data, the absolute reaction enthalpies for dehydrogenation of the alcohols are notably underestimated at the B97-D level, but the relative sequence is very well captured at the DFT level (seeTable S8).

Schneider N., 2009, Chem. Eur. J., 15, 11515, 10.1002/chem.200901594