- Email: [email protected]
Kubas complexes extended to four centers; a theoretical prediction of novel dihydrogen coordination in bimetallic tungsten and molybdenum compounds
Journal of Organometallic Chemistry 766 (2014) 67e72
Contents lists available at ScienceDirect
Journal of Organometallic Chemistry journal homepage: www.elsevier.com/locate/jorganchem
Kubas complexes extended to four centers; a theoretical prediction of novel dihydrogen coordination in bimetallic tungsten and molybdenum compounds Emmanuel D. Simandiras a, *, Dimitrios G. Liakos b, Nikolaos Psaroudakis c, Konstantinos Mertis c a b c
Theoretical and Physical Chemistry Institute, National Hellenic Research Foundation, 48 Vassileos Constantinou Avenue, 116 35 Athens, Greece Max-Planck Institute for Chemical Energy Conversion, Stiftstrasse 34-36, D-45470 Mülheim an der Ruhr, Germany Inorganic Chemistry Laboratory, University of Athens, Panepistimiopolis, 15771 Athens, Greece
a r t i c l e i n f o
a b s t r a c t
Article history: Received 31 March 2014 Received in revised form 6 May 2014 Accepted 9 May 2014
A number of molybdenum and tungsten bimetallic compounds with carbonyl and phosphine ligands are investigated, using theoretical methods, with respect to their ability to bind molecular hydrogen in a Kubas type coordination. Some are found to give novel complexes, containing a four center metaledihydrogen interaction. Ó 2014 Elsevier B.V. All rights reserved.
Keywords: Dihydrogen complexes Kubas complexes Theoretical prediction Molybdenum carbonyls Tungsten carbonyls
Introduction Transition metaledihydrogen complexes, also known as Kubas complexes, were first reported almost thirty years ago [1]. The significance, in terms of the beauty of the scientific breakthrough, is vividly described in a recent review [2]. The broader importance, both in the context of pure science, but also in the quest for efficient hydrogen storage, has been extensively reviewed [3e5]. The common features of Kubas complexes are: A three center interaction from the s orbital of H2 to an empty metal d-orbital A back-donation from occupied metal d orbitals towards the s* antibonding H2 orbital A lengthening of the HeH bond to some extent, but not to the point of breaking. Lengths of less than 1.0 A are generally accepted to constitute dihydrogen complexes, whereas for lengths of more that 1.3 A the structure is characterized as a dihydride.
* Corresponding author. Tel.: þ30 210 7273806; fax: þ30 210 7273794. E-mail address: [email protected] (E.D. Simandiras). http://dx.doi.org/10.1016/j.jorganchem.2014.05.007 0022-328X/Ó 2014 Elsevier B.V. All rights reserved.
A T-shaped structure and, generally speaking, low metal coordination The numerous dihydrogen complexes that have been synthesized and characterized, or predicted theoretically, share one common feature; a single metal atom is bound to dihydrogen. In a recent letter [6], we reported the first theoretical prediction of fourcenter Kubas complexes, where two transition metal atoms bind to a single H2 molecule simultaneously, with exactly the same characteristics as in the three-center Kubas complex. These will hereinafter be referred to as four-center Kubas (4cK) complexes. In the general field of transition metal hydrides, the area of bridging hydrides carries special interest due to its importance in various chemical reactions. Many aspects are not definitively elucidated, as for example the extent of direct vs indirect metale metal interaction. Numerous theoretical works address this and other issues; for example the works of Baik, Friesner and Parkin [7] and Richardson et al. [8] examine from a theoretical point of view a number of dibridging dihydrides of W, Re, Cr and Mo. Hoffmann and his coworkers [9] in a recent paper examine hydrogen binding to W clusters under pressure and mention that there are 85 reported polynuclear W complexes with Hydrogen bridges, and of these 14 involve two bridging hydrogens.
68
E.D. Simandiras et al. / Journal of Organometallic Chemistry 766 (2014) 67e72
Fig. 1. The structure of W2(CO)8(m-H2)and Mo2(CO)8(m-H2)(side and top views, bond lengths from PBE data).
Summarizing the literature, no theoretical or experimental evidence has up to now been presented, where in a bridging dihydride the distance between the hydrogen atoms is so short that a dihydrogen interaction (Kubas type) can be recognized. For completeness, it is worth mentioning that in a work of Krasnov et al. [10] on H2 adsorbed on Sc clustered on single wall carbon nanotubes, a similar four center interaction was predicted, as part of a number of hydrogen molecules attached to Sc atoms; this was however a part of a multi-metal cluster attached to a nanotube, and not an isolated molecule. A systematic theoretical study of a number of 4cK complexes is presented here. It is reported that both W and Mo compounds with carbonyl and phosphine ligands are predicted from theory to form such complexes, and their binding is analyzed. In the first part of the results section, the pure carbonyl compounds are presented. The second part shows the results of investigations with phosphine ligands, and the third reports on the possibility of both three center and four center dihydrogen binding existing on the same molecule. Computational details The density functional theory (DFT) approach was used throughout, as implemented in the GAUSSIAN 09 suite of programs [11]. From experience from our previous work and the overall literature in the field, two functionals were chosen that best describe transition metal hydride computations. These are the Perdew, Burke and Ernzerhof (PBE) [12,13] functional, and the M06 of Zhao and Truhlar [14]. The basis set used is the def2TZVPP of Weigend and Ahlrichs [15] including the relevant effective core potential for Mo and W [16] (basis sets retrieved from http://www.cosmologic.de/basis-sets/basissets.php). All structures were fully optimized using both functionals, and the potential energy minima were confirmed by a full second derivative calculation. Only geometrical data are reported here, but harmonic vibrational frequencies are also calculated; the stretching frequencies for H2 for some of the complexes are given in the Supplementary material. Coordinates for all optimized structures are also given.
Results and discussion Bimetallic carbonyls Tungsten and molybdenum octacarbonyls are found to give stable 4cK complexes, as shown in Fig. 1. Both W2(CO)8(m-H2) and Mo2(CO)8(m-H2) are given here simultaneously for comparison, but the latter was first published previously by us [6]. The basic bond lengths of the optimized structure, calculated at both functional levels, are given in Table 1. The metalemetal bond length, as also confirmed by the molecular orbital picture, corresponds to a single bond, in agreement with other similar compounds [17,18]. The dihydrogen bond length is elongated with respect to the free molecule (approx 0.75 A with both functionals), but remains below 1.0 A as is expected for a true Kubas complex. Also, a good agreement is seen between the two functionals, with M06 giving shorter HeH and longer metalemetal bonds, meaning that the backbonding interaction is underestimated compared to PBE. In any case, however, both functionals support the same finding, namely the stability of the 4cK complex. Attempts were made with other Mo and W complexes with different numbers of carbonyl ligands, but only the low coordination octacarbonyls are found to give dihydrogen complexes. Higher coordination compounds lead to dihydrides, in some cases unstable. The bonding picture, similar for both functionals, is revealing of the true Kubas type nature of the metal dihydrogen interaction. A diagram of the occupied orbitals of interest for W2(CO)8(m-H2) are given in Fig. 2.
Table 1 Geometries of 4cK complexes of W and Mo (in A). Complex
Functional
MeM (M ¼ W, Mo)
MeH
HeH
W2(CO)8(m-H2)
PBE M06 PBE M06
2.691 2.721 2.684 2.712
2.035e2.042 2.072e2.087 2.065e2.075 2.080e2.094
0.934 0.877 0.872 0.852
Mo2(CO)8(m-H2) [6]
E.D. Simandiras et al. / Journal of Organometallic Chemistry 766 (2014) 67e72
69
Fig. 2. The binding scheme of W2(CO)8(m-H2).
Orbitals 70, 69 and 68 are the three HOMO and have a s*, d* and p* character with respect to the metalemetal interaction, and at the same time bonding interaction in the metal-carbonyl bond. Orbital 66 is s for the metalemetal interaction, 67 is d, and 65 is p. The latter represents the backdonation factor of the Kubas interaction from a metalemetal orbital to the s* antibonding orbital of H2. Finally, the much lower in energy orbital 34 represents the main four-center interaction of d orbitals from both metals and dihydrogen. The extraordinary nature of the WeW (and MoeMo) bond, which can go up to quadruple, is responsible for the unique bonding in these four center dihydrogen complexes. As was found previously [19], a single protonation of a quadruple WeW bond breaks the p component. Theory predicts that a second protonation does not break the other p component, but due to the strong interaction with HeH interaction and the flexibility of the metals
gives rise to an inflated four center orbital as seen in Fig. 2 orbital 34. The second p component back-donates to dihydrogen, as in Fig. 2 orbital 65, softening and elongating the HeH bond.
Bimetallic carbonyl and phosphine compounds The replacement of carbonyls with one or more phosphine ligands in monometallic W and Mo compounds leads to the formation of dihydride rather than dihydrogen complexes [2,20]. This is due to the metal being progressively more electron rich, resulting in an increase of the backbonding interaction and the breaking (or lengthening to a great extent) of the HeH bond. The original Kubas complex contains two phosphines, mainly for stereochemical reasons, but for more than two phosphines there is a tendency towards dihydrides rather than dihydrogen.
Table 2 Structural data of carbonyl-phosphine 4cK complexes. Complex
Functional
MoeMo
MoeP
MoeH
HeH
Mo2(CO)6(PMe3)2(m-H2) (Fig. 3)
PBE M06 PBE M06 PBE
2.666 2.688 2.658 2.690 2.689
2.451 2.438 2.448e2.457 2.444e2.455 2.457e2.468
1.956e2.029 2.029e2.055 1.990e2.062 2.039e2.081 1.957e2.055
0.941 0.872 0.909 0.865 0.927
Mo2(CO)6(1,4dimethylphosphinobutane)(m-H2) (Fig. 4) Mo2(CO)6(1,6dimethylphosphinohexane)(m-H2)
70
E.D. Simandiras et al. / Journal of Organometallic Chemistry 766 (2014) 67e72
Fig. 3. Top and side view of Mo2(CO)6(PMe3)2(m-H2) (bond lengths PBE data).
Fig. 4. Top and side view of the 4cK complex Mo2(CO)6(1,4dimethylphosphinobutane)(m-H2) (bond lengths from PBE data).
The trend, when carbonyl ligands are replaced by phosphines in four center bimetallic dihydrogen complexes, is investigated here. A number of structures involving bimetallic W and Mo compounds with two (one on each metal) and four (two on each metal) PMe3 phosphine ligands were investigated at the DFT level with the same basis and functionals as in the previous computations reported here. The phosphines were positioned both cis and trans to the HeH dihydrogen/dihydride. The majority of structure optimizations leads to stable dihydride clusters. In particular, W2(CO)6(PMe3)2H2 is found to be a dihydride, with HeH distances between 2.1 and 2.3 A, depending on the position of the phosphines on each metal. Similarly, W2(CO)4(PMe3)4H2 is a dihydride with HeH distance at 2.25 A. This is in agreement with the experimental finding summarized by Kubas (Fig. 8 in Ref. [2] and Scheme 9 in Ref. [5])that stepwise replacement of carbonyl ligands with phosphines leads to breaking of the HeH bond and stable dihydrides. A notable exception is Mo2(CO)6(PMe3)2H2, which is found to be a stable 4cK complex, and the structure is shown in Fig. 3. The structural parameters found with both functionals are given in Table 2. Observation of the stable structure shown in Fig. 3 suggests that isomeric forms will exist, when the position of the PMe3 ligand on each metal atom changes. A theoretical investigation shows this to be true; the coordinates of an isomeric form, where the phosphines occupy different positions compared to those of Fig. 3, and the HeH distance is elongated to 2.253 A, are given in the Supplementary material. The energy differences are inconclusive as to the relative stability at this level of theory. To increase the stability and predict a possible 4cK cluster that could be synthesized in the laboratory, a number of bridging phosphines was investigated; in this way the positions of the ligands will not be allowed to vary. Attempts to replace two of the methyl groups of Mo2(CO)6(PMe3)2(m-H2) (one on each metal) and connect the two phosphorus atoms with a carbon chain revealed
the following: When two or three carbon atoms were used in the computation, the strain on the coordination was such that the 4cK dihydrogen complex breaks towards a classic three center complex on one of the metal atoms. For chains of four carbon atoms and longer, the strain on the metalemetal bond is reduced and novel 4cK complexes are predicted. The structure of the first such complex, Mo2(CO)6(1,4dimethylphosphinobutane)(m-H2) was optimized with both functionals and found to be stable with the structure given in Fig. 4. The basic geometrical parameters of both this complex and one with a six carbon chain, which is also predicted a stable 4cK, are given in Table 2.
Complexes involving both three center and four center Kubas interactions The positioning of the H2 molecule in the above described 4cK dihydrogen complexes leaves the possibility open for a further hydrogen molecule to be simultaneously attached to the metal center at a different ligand position. This is investigated here for both W and Mo carbonyls. Possible structures for W2(CO)6(m-H2)3 and Mo2(CO)6(m-H2)3, i.e. hexacarbonyls with three hydrogen molecules each, were
Table 3 Data for W and Mo complexes involving both 3c and 4c dihydrogen complexes. Complex
Functional
MeM (M ¼ W, Mo)
HeH four center
HeH three center
W2(CO)6(m-H2)3
PBE M06 PBE PBE M06 PBE
2.679 2.702 2.685 2.676 2.696 2.681
0.973 0.879 0.950 0.879 0.852 0.876
0.852 0.835 0.847 0.829 0.821 0.827
W2(CO)7(m-H2)2 (Fig. 6) Mo2(CO)6(m-H2)3 (Fig. 5) Mo2(CO)7(m-H2)2
E.D. Simandiras et al. / Journal of Organometallic Chemistry 766 (2014) 67e72
71
Fig. 5. Top and side view of Mo2(CO)6(m-H2)3 (bond lengths from PBE data).
Fig. 6. Top and side view of W2(CO)7(m-H2)2 (bond lengths from PBE data).
optimized at the same level of theory as above. In both cases, a Kubas complex involving three dihydrogen ligands, one in a four center approach and two in three center form, are found. The structure is shown in Fig. 5. All structural data are summarized in Table 3. For both complexes, further isomers were optimized and stable structures are found with the three center dihydrogen ligands attached at different positions of each metal atom. In all isomers of the molybdenum complex the four center ligand remains a
dihydrogen, whereas in some of the tungsten complexes it is extended to a dihydride. Overall, however, the Mo complex shown in Fig. 5 is energetically more stable (by more than 4 kcal/mol for PBE) compared to the other isomers, whereas the tungsten equivalent has isomers of similar energy (to within 2 kcal/mol) where the middle H2 distance is extended to about 2 A. For the latter, coordinates are given in Table S3 of the Supplementary material. It should be noted that for all cases where isomers of similar energies are found, the current approach can not give a
Fig. 7. The molecular orbitals of Mo2(CO)6(m-H2)3 that involve metaleH2 s interaction.
72
E.D. Simandiras et al. / Journal of Organometallic Chemistry 766 (2014) 67e72
definitive answer as to the relative stability. Many factors that affect the calculated result exist, such as the basis set, method and functional limitations, and lack of solvent and thermal effects to name just a few. The predictive nature of the calculations, however, is not affected. Further, the compounds W2(CO)7(m-H2)2 and Mo2(CO)7(m-H2)2 were investigated. As is seen in Table 3 these are also simultaneous 3ce4c Kubas complexes, and this is true for both molybdenum and tungsten with many possible configurations investigated. The optimized structure for the molybdenum complex can be seen in Fig. 6. The molecular orbital configuration of Mo2(CO)6(m-H2)3 (Fig. 5) is quite similar to that of the single dihydrogen 4cK complex. The six highest occupied orbitals of the complex are given in the Supplementary material. The orbitals that lie towards the top of the energy scale, account for the metalemetal interaction and for the backbonding towards all three dihydrogen molecules attached (orbitals 53, 55 and 56). The three direct interactions between metal d orbitals and each of the s bonding orbitals of dihydrogen molecules are found in three bonding orbitals that are much lower in the orbital energy scale and can be seen in Fig. 7. In particular, MO 32 is the direct four-center interaction between the two Mo atoms and dihydrogen, whereas MOs 33 and 34 represent the three center interactions between each Mo atom and the corresponding dihydrogen. Conclusions Novel dihydrogen complexes involving bimetallic compounds of tungsten and molybdenum are predicted by accurate theoretical calculations. The Kubas type interaction of H2 simultaneously with two metal centers that are already bound to each other, gives rise to a four center interaction. In this work, stable complexes of both pure and phosphinated carbonyls of Mo and W are predicted. Further, isolated molecule complexes where dihydrogen is bound (in a Kubas fashion) to both the individual metal center and to two metals simultaneously (three center and four center) are located from theoretical calculations. This suggests a possible increased capacity for metale dihydrogen coordination, to three H2 molecules per two metal atoms.
Finally, the novel Kubas type interaction of multiple metal centers reported here, could be possible in other transition metal compounds, and this is currently under investigation. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.jorganchem.2014.05.007. References [1] G.J. Kubas, R.R. Ryan, B.I. Swanson, P.J. Vergamini, H.J. Wasserman, J. Am. Chem. Soc. 106 (1984) 451e452. [2] G.J. Kubas, Comm. Inorg. Chem. 33 (2012) 102e121. [3] G.J. Kubas, Metal Dihydrogen and Sigma Bond Complexes, Kluwer Academic, New York, 2002. [4] G.J. Kubas, Chem. Rev. 107 (2007) 4152e4205. [5] G.J. Kubas, J. Organomet. Chem. 751 (2014) 33e49. [6] E.D. Simandiras, D.G. Liakos, Chem. Phys. Lett. 583 (2013) 18e22. [7] M.-H. Baik, R.A. Friesner, G. Parkin, Polyhedron 23 (2004) 2879e2900. [8] N.A. Richardson, Y. Xie, R.B. King, H.F. Schaefer, J. Phys. Chem. A 105 (2001) 11134e11143. [9] V. Labet, R. Hoffmann, N.W. Ashcroft, New J. Chem. 35 (2011) 2349e2355. [10] P.O. Krasnov, F. Ding, A.K. Singh, B.I. Yakobson, J. Phys. Chem. C 111 (2007) 17977e17980. [11] G.W.T.M.J. Frisch, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G.A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H.P. Hratchian, A.F. Izmaylov, J. Bloino, G. Zheng, J.L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J.A. Montgomery Jr., J.E. Peralta, F. Ogliaro, M. Bearpark, J.J. Heyd, E. Brothers, K.N. Kudin, V.N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J.C. Burant, S.S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J.M. Millam, M. Klene, J.E. Knox, J.B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli, J.W. Ochterski, R.L. Martin, K. Morokuma, V.G. Zakrzewski, G.A. Voth, P. Salvador, J.J. Dannenberg, S. Dapprich, A.D. Daniels, Ö. Farkas, J.B. Foresman, J.V. Ortiz, J. Cioslowski, D.J. Fox, Gaussian 09, Revision A.1, Gaussian, Inc., Wallingford CT, 2009. [12] J.P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 77 (1996) 3865e3868. [13] J.P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 78 (1997) 1396. [14] Y. Zhao, D. Truhlar, Theor. Chem. Acc. 120 (2008) 215e241. [15] F. Weigend, R. Ahlrichs, Phys. Chem. Chem. Phys. 7 (2005) 3297e3305. [16] D. Andrae, U. Haussermann, M. Dolg, H. Stoll, H. Preuss, Theor. Chim. Acta 77 (1990) 123e141. [17] A. Bino, F.A. Cotton, Z. Dori, J.C. Sekutowski, Inorg. Chem. 17 (1978) 2946e 2950. [18] F.A. Cotton, J.P. Donahue, M.B. Hall, C.A. Murillo, D. Villagrán, Inorg. Chem. 43 (2004) 6954e6964. [19] E.D. Simandiras, N. Psaroudakis, K. Mertis, Polyhedron 54 (2013) 173e179. [20] J. Tomàs, A. Lledós, Y. Jean, Organometallics 17 (1998) 4932e4939.
Contents lists available at ScienceDirect
Journal of Organometallic Chemistry journal homepage: www.elsevier.com/locate/jorganchem
Kubas complexes extended to four centers; a theoretical prediction of novel dihydrogen coordination in bimetallic tungsten and molybdenum compounds Emmanuel D. Simandiras a, *, Dimitrios G. Liakos b, Nikolaos Psaroudakis c, Konstantinos Mertis c a b c
Theoretical and Physical Chemistry Institute, National Hellenic Research Foundation, 48 Vassileos Constantinou Avenue, 116 35 Athens, Greece Max-Planck Institute for Chemical Energy Conversion, Stiftstrasse 34-36, D-45470 Mülheim an der Ruhr, Germany Inorganic Chemistry Laboratory, University of Athens, Panepistimiopolis, 15771 Athens, Greece
a r t i c l e i n f o
a b s t r a c t
Article history: Received 31 March 2014 Received in revised form 6 May 2014 Accepted 9 May 2014
A number of molybdenum and tungsten bimetallic compounds with carbonyl and phosphine ligands are investigated, using theoretical methods, with respect to their ability to bind molecular hydrogen in a Kubas type coordination. Some are found to give novel complexes, containing a four center metaledihydrogen interaction. Ó 2014 Elsevier B.V. All rights reserved.
Keywords: Dihydrogen complexes Kubas complexes Theoretical prediction Molybdenum carbonyls Tungsten carbonyls
Introduction Transition metaledihydrogen complexes, also known as Kubas complexes, were first reported almost thirty years ago [1]. The significance, in terms of the beauty of the scientific breakthrough, is vividly described in a recent review [2]. The broader importance, both in the context of pure science, but also in the quest for efficient hydrogen storage, has been extensively reviewed [3e5]. The common features of Kubas complexes are: A three center interaction from the s orbital of H2 to an empty metal d-orbital A back-donation from occupied metal d orbitals towards the s* antibonding H2 orbital A lengthening of the HeH bond to some extent, but not to the point of breaking. Lengths of less than 1.0 A are generally accepted to constitute dihydrogen complexes, whereas for lengths of more that 1.3 A the structure is characterized as a dihydride.
* Corresponding author. Tel.: þ30 210 7273806; fax: þ30 210 7273794. E-mail address: [email protected] (E.D. Simandiras). http://dx.doi.org/10.1016/j.jorganchem.2014.05.007 0022-328X/Ó 2014 Elsevier B.V. All rights reserved.
A T-shaped structure and, generally speaking, low metal coordination The numerous dihydrogen complexes that have been synthesized and characterized, or predicted theoretically, share one common feature; a single metal atom is bound to dihydrogen. In a recent letter [6], we reported the first theoretical prediction of fourcenter Kubas complexes, where two transition metal atoms bind to a single H2 molecule simultaneously, with exactly the same characteristics as in the three-center Kubas complex. These will hereinafter be referred to as four-center Kubas (4cK) complexes. In the general field of transition metal hydrides, the area of bridging hydrides carries special interest due to its importance in various chemical reactions. Many aspects are not definitively elucidated, as for example the extent of direct vs indirect metale metal interaction. Numerous theoretical works address this and other issues; for example the works of Baik, Friesner and Parkin [7] and Richardson et al. [8] examine from a theoretical point of view a number of dibridging dihydrides of W, Re, Cr and Mo. Hoffmann and his coworkers [9] in a recent paper examine hydrogen binding to W clusters under pressure and mention that there are 85 reported polynuclear W complexes with Hydrogen bridges, and of these 14 involve two bridging hydrogens.
68
E.D. Simandiras et al. / Journal of Organometallic Chemistry 766 (2014) 67e72
Fig. 1. The structure of W2(CO)8(m-H2)and Mo2(CO)8(m-H2)(side and top views, bond lengths from PBE data).
Summarizing the literature, no theoretical or experimental evidence has up to now been presented, where in a bridging dihydride the distance between the hydrogen atoms is so short that a dihydrogen interaction (Kubas type) can be recognized. For completeness, it is worth mentioning that in a work of Krasnov et al. [10] on H2 adsorbed on Sc clustered on single wall carbon nanotubes, a similar four center interaction was predicted, as part of a number of hydrogen molecules attached to Sc atoms; this was however a part of a multi-metal cluster attached to a nanotube, and not an isolated molecule. A systematic theoretical study of a number of 4cK complexes is presented here. It is reported that both W and Mo compounds with carbonyl and phosphine ligands are predicted from theory to form such complexes, and their binding is analyzed. In the first part of the results section, the pure carbonyl compounds are presented. The second part shows the results of investigations with phosphine ligands, and the third reports on the possibility of both three center and four center dihydrogen binding existing on the same molecule. Computational details The density functional theory (DFT) approach was used throughout, as implemented in the GAUSSIAN 09 suite of programs [11]. From experience from our previous work and the overall literature in the field, two functionals were chosen that best describe transition metal hydride computations. These are the Perdew, Burke and Ernzerhof (PBE) [12,13] functional, and the M06 of Zhao and Truhlar [14]. The basis set used is the def2TZVPP of Weigend and Ahlrichs [15] including the relevant effective core potential for Mo and W [16] (basis sets retrieved from http://www.cosmologic.de/basis-sets/basissets.php). All structures were fully optimized using both functionals, and the potential energy minima were confirmed by a full second derivative calculation. Only geometrical data are reported here, but harmonic vibrational frequencies are also calculated; the stretching frequencies for H2 for some of the complexes are given in the Supplementary material. Coordinates for all optimized structures are also given.
Results and discussion Bimetallic carbonyls Tungsten and molybdenum octacarbonyls are found to give stable 4cK complexes, as shown in Fig. 1. Both W2(CO)8(m-H2) and Mo2(CO)8(m-H2) are given here simultaneously for comparison, but the latter was first published previously by us [6]. The basic bond lengths of the optimized structure, calculated at both functional levels, are given in Table 1. The metalemetal bond length, as also confirmed by the molecular orbital picture, corresponds to a single bond, in agreement with other similar compounds [17,18]. The dihydrogen bond length is elongated with respect to the free molecule (approx 0.75 A with both functionals), but remains below 1.0 A as is expected for a true Kubas complex. Also, a good agreement is seen between the two functionals, with M06 giving shorter HeH and longer metalemetal bonds, meaning that the backbonding interaction is underestimated compared to PBE. In any case, however, both functionals support the same finding, namely the stability of the 4cK complex. Attempts were made with other Mo and W complexes with different numbers of carbonyl ligands, but only the low coordination octacarbonyls are found to give dihydrogen complexes. Higher coordination compounds lead to dihydrides, in some cases unstable. The bonding picture, similar for both functionals, is revealing of the true Kubas type nature of the metal dihydrogen interaction. A diagram of the occupied orbitals of interest for W2(CO)8(m-H2) are given in Fig. 2.
Table 1 Geometries of 4cK complexes of W and Mo (in A). Complex
Functional
MeM (M ¼ W, Mo)
MeH
HeH
W2(CO)8(m-H2)
PBE M06 PBE M06
2.691 2.721 2.684 2.712
2.035e2.042 2.072e2.087 2.065e2.075 2.080e2.094
0.934 0.877 0.872 0.852
Mo2(CO)8(m-H2) [6]
E.D. Simandiras et al. / Journal of Organometallic Chemistry 766 (2014) 67e72
69
Fig. 2. The binding scheme of W2(CO)8(m-H2).
Orbitals 70, 69 and 68 are the three HOMO and have a s*, d* and p* character with respect to the metalemetal interaction, and at the same time bonding interaction in the metal-carbonyl bond. Orbital 66 is s for the metalemetal interaction, 67 is d, and 65 is p. The latter represents the backdonation factor of the Kubas interaction from a metalemetal orbital to the s* antibonding orbital of H2. Finally, the much lower in energy orbital 34 represents the main four-center interaction of d orbitals from both metals and dihydrogen. The extraordinary nature of the WeW (and MoeMo) bond, which can go up to quadruple, is responsible for the unique bonding in these four center dihydrogen complexes. As was found previously [19], a single protonation of a quadruple WeW bond breaks the p component. Theory predicts that a second protonation does not break the other p component, but due to the strong interaction with HeH interaction and the flexibility of the metals
gives rise to an inflated four center orbital as seen in Fig. 2 orbital 34. The second p component back-donates to dihydrogen, as in Fig. 2 orbital 65, softening and elongating the HeH bond.
Bimetallic carbonyl and phosphine compounds The replacement of carbonyls with one or more phosphine ligands in monometallic W and Mo compounds leads to the formation of dihydride rather than dihydrogen complexes [2,20]. This is due to the metal being progressively more electron rich, resulting in an increase of the backbonding interaction and the breaking (or lengthening to a great extent) of the HeH bond. The original Kubas complex contains two phosphines, mainly for stereochemical reasons, but for more than two phosphines there is a tendency towards dihydrides rather than dihydrogen.
Table 2 Structural data of carbonyl-phosphine 4cK complexes. Complex
Functional
MoeMo
MoeP
MoeH
HeH
Mo2(CO)6(PMe3)2(m-H2) (Fig. 3)
PBE M06 PBE M06 PBE
2.666 2.688 2.658 2.690 2.689
2.451 2.438 2.448e2.457 2.444e2.455 2.457e2.468
1.956e2.029 2.029e2.055 1.990e2.062 2.039e2.081 1.957e2.055
0.941 0.872 0.909 0.865 0.927
Mo2(CO)6(1,4dimethylphosphinobutane)(m-H2) (Fig. 4) Mo2(CO)6(1,6dimethylphosphinohexane)(m-H2)
70
E.D. Simandiras et al. / Journal of Organometallic Chemistry 766 (2014) 67e72
Fig. 3. Top and side view of Mo2(CO)6(PMe3)2(m-H2) (bond lengths PBE data).
Fig. 4. Top and side view of the 4cK complex Mo2(CO)6(1,4dimethylphosphinobutane)(m-H2) (bond lengths from PBE data).
The trend, when carbonyl ligands are replaced by phosphines in four center bimetallic dihydrogen complexes, is investigated here. A number of structures involving bimetallic W and Mo compounds with two (one on each metal) and four (two on each metal) PMe3 phosphine ligands were investigated at the DFT level with the same basis and functionals as in the previous computations reported here. The phosphines were positioned both cis and trans to the HeH dihydrogen/dihydride. The majority of structure optimizations leads to stable dihydride clusters. In particular, W2(CO)6(PMe3)2H2 is found to be a dihydride, with HeH distances between 2.1 and 2.3 A, depending on the position of the phosphines on each metal. Similarly, W2(CO)4(PMe3)4H2 is a dihydride with HeH distance at 2.25 A. This is in agreement with the experimental finding summarized by Kubas (Fig. 8 in Ref. [2] and Scheme 9 in Ref. [5])that stepwise replacement of carbonyl ligands with phosphines leads to breaking of the HeH bond and stable dihydrides. A notable exception is Mo2(CO)6(PMe3)2H2, which is found to be a stable 4cK complex, and the structure is shown in Fig. 3. The structural parameters found with both functionals are given in Table 2. Observation of the stable structure shown in Fig. 3 suggests that isomeric forms will exist, when the position of the PMe3 ligand on each metal atom changes. A theoretical investigation shows this to be true; the coordinates of an isomeric form, where the phosphines occupy different positions compared to those of Fig. 3, and the HeH distance is elongated to 2.253 A, are given in the Supplementary material. The energy differences are inconclusive as to the relative stability at this level of theory. To increase the stability and predict a possible 4cK cluster that could be synthesized in the laboratory, a number of bridging phosphines was investigated; in this way the positions of the ligands will not be allowed to vary. Attempts to replace two of the methyl groups of Mo2(CO)6(PMe3)2(m-H2) (one on each metal) and connect the two phosphorus atoms with a carbon chain revealed
the following: When two or three carbon atoms were used in the computation, the strain on the coordination was such that the 4cK dihydrogen complex breaks towards a classic three center complex on one of the metal atoms. For chains of four carbon atoms and longer, the strain on the metalemetal bond is reduced and novel 4cK complexes are predicted. The structure of the first such complex, Mo2(CO)6(1,4dimethylphosphinobutane)(m-H2) was optimized with both functionals and found to be stable with the structure given in Fig. 4. The basic geometrical parameters of both this complex and one with a six carbon chain, which is also predicted a stable 4cK, are given in Table 2.
Complexes involving both three center and four center Kubas interactions The positioning of the H2 molecule in the above described 4cK dihydrogen complexes leaves the possibility open for a further hydrogen molecule to be simultaneously attached to the metal center at a different ligand position. This is investigated here for both W and Mo carbonyls. Possible structures for W2(CO)6(m-H2)3 and Mo2(CO)6(m-H2)3, i.e. hexacarbonyls with three hydrogen molecules each, were
Table 3 Data for W and Mo complexes involving both 3c and 4c dihydrogen complexes. Complex
Functional
MeM (M ¼ W, Mo)
HeH four center
HeH three center
W2(CO)6(m-H2)3
PBE M06 PBE PBE M06 PBE
2.679 2.702 2.685 2.676 2.696 2.681
0.973 0.879 0.950 0.879 0.852 0.876
0.852 0.835 0.847 0.829 0.821 0.827
W2(CO)7(m-H2)2 (Fig. 6) Mo2(CO)6(m-H2)3 (Fig. 5) Mo2(CO)7(m-H2)2
E.D. Simandiras et al. / Journal of Organometallic Chemistry 766 (2014) 67e72
71
Fig. 5. Top and side view of Mo2(CO)6(m-H2)3 (bond lengths from PBE data).
Fig. 6. Top and side view of W2(CO)7(m-H2)2 (bond lengths from PBE data).
optimized at the same level of theory as above. In both cases, a Kubas complex involving three dihydrogen ligands, one in a four center approach and two in three center form, are found. The structure is shown in Fig. 5. All structural data are summarized in Table 3. For both complexes, further isomers were optimized and stable structures are found with the three center dihydrogen ligands attached at different positions of each metal atom. In all isomers of the molybdenum complex the four center ligand remains a
dihydrogen, whereas in some of the tungsten complexes it is extended to a dihydride. Overall, however, the Mo complex shown in Fig. 5 is energetically more stable (by more than 4 kcal/mol for PBE) compared to the other isomers, whereas the tungsten equivalent has isomers of similar energy (to within 2 kcal/mol) where the middle H2 distance is extended to about 2 A. For the latter, coordinates are given in Table S3 of the Supplementary material. It should be noted that for all cases where isomers of similar energies are found, the current approach can not give a
Fig. 7. The molecular orbitals of Mo2(CO)6(m-H2)3 that involve metaleH2 s interaction.
72
E.D. Simandiras et al. / Journal of Organometallic Chemistry 766 (2014) 67e72
definitive answer as to the relative stability. Many factors that affect the calculated result exist, such as the basis set, method and functional limitations, and lack of solvent and thermal effects to name just a few. The predictive nature of the calculations, however, is not affected. Further, the compounds W2(CO)7(m-H2)2 and Mo2(CO)7(m-H2)2 were investigated. As is seen in Table 3 these are also simultaneous 3ce4c Kubas complexes, and this is true for both molybdenum and tungsten with many possible configurations investigated. The optimized structure for the molybdenum complex can be seen in Fig. 6. The molecular orbital configuration of Mo2(CO)6(m-H2)3 (Fig. 5) is quite similar to that of the single dihydrogen 4cK complex. The six highest occupied orbitals of the complex are given in the Supplementary material. The orbitals that lie towards the top of the energy scale, account for the metalemetal interaction and for the backbonding towards all three dihydrogen molecules attached (orbitals 53, 55 and 56). The three direct interactions between metal d orbitals and each of the s bonding orbitals of dihydrogen molecules are found in three bonding orbitals that are much lower in the orbital energy scale and can be seen in Fig. 7. In particular, MO 32 is the direct four-center interaction between the two Mo atoms and dihydrogen, whereas MOs 33 and 34 represent the three center interactions between each Mo atom and the corresponding dihydrogen. Conclusions Novel dihydrogen complexes involving bimetallic compounds of tungsten and molybdenum are predicted by accurate theoretical calculations. The Kubas type interaction of H2 simultaneously with two metal centers that are already bound to each other, gives rise to a four center interaction. In this work, stable complexes of both pure and phosphinated carbonyls of Mo and W are predicted. Further, isolated molecule complexes where dihydrogen is bound (in a Kubas fashion) to both the individual metal center and to two metals simultaneously (three center and four center) are located from theoretical calculations. This suggests a possible increased capacity for metale dihydrogen coordination, to three H2 molecules per two metal atoms.
Finally, the novel Kubas type interaction of multiple metal centers reported here, could be possible in other transition metal compounds, and this is currently under investigation. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.jorganchem.2014.05.007. References [1] G.J. Kubas, R.R. Ryan, B.I. Swanson, P.J. Vergamini, H.J. Wasserman, J. Am. Chem. Soc. 106 (1984) 451e452. [2] G.J. Kubas, Comm. Inorg. Chem. 33 (2012) 102e121. [3] G.J. Kubas, Metal Dihydrogen and Sigma Bond Complexes, Kluwer Academic, New York, 2002. [4] G.J. Kubas, Chem. Rev. 107 (2007) 4152e4205. [5] G.J. Kubas, J. Organomet. Chem. 751 (2014) 33e49. [6] E.D. Simandiras, D.G. Liakos, Chem. Phys. Lett. 583 (2013) 18e22. [7] M.-H. Baik, R.A. Friesner, G. Parkin, Polyhedron 23 (2004) 2879e2900. [8] N.A. Richardson, Y. Xie, R.B. King, H.F. Schaefer, J. Phys. Chem. A 105 (2001) 11134e11143. [9] V. Labet, R. Hoffmann, N.W. Ashcroft, New J. Chem. 35 (2011) 2349e2355. [10] P.O. Krasnov, F. Ding, A.K. Singh, B.I. Yakobson, J. Phys. Chem. C 111 (2007) 17977e17980. [11] G.W.T.M.J. Frisch, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G.A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H.P. Hratchian, A.F. Izmaylov, J. Bloino, G. Zheng, J.L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J.A. Montgomery Jr., J.E. Peralta, F. Ogliaro, M. Bearpark, J.J. Heyd, E. Brothers, K.N. Kudin, V.N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J.C. Burant, S.S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J.M. Millam, M. Klene, J.E. Knox, J.B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli, J.W. Ochterski, R.L. Martin, K. Morokuma, V.G. Zakrzewski, G.A. Voth, P. Salvador, J.J. Dannenberg, S. Dapprich, A.D. Daniels, Ö. Farkas, J.B. Foresman, J.V. Ortiz, J. Cioslowski, D.J. Fox, Gaussian 09, Revision A.1, Gaussian, Inc., Wallingford CT, 2009. [12] J.P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 77 (1996) 3865e3868. [13] J.P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 78 (1997) 1396. [14] Y. Zhao, D. Truhlar, Theor. Chem. Acc. 120 (2008) 215e241. [15] F. Weigend, R. Ahlrichs, Phys. Chem. Chem. Phys. 7 (2005) 3297e3305. [16] D. Andrae, U. Haussermann, M. Dolg, H. Stoll, H. Preuss, Theor. Chim. Acta 77 (1990) 123e141. [17] A. Bino, F.A. Cotton, Z. Dori, J.C. Sekutowski, Inorg. Chem. 17 (1978) 2946e 2950. [18] F.A. Cotton, J.P. Donahue, M.B. Hall, C.A. Murillo, D. Villagrán, Inorg. Chem. 43 (2004) 6954e6964. [19] E.D. Simandiras, N. Psaroudakis, K. Mertis, Polyhedron 54 (2013) 173e179. [20] J. Tomàs, A. Lledós, Y. Jean, Organometallics 17 (1998) 4932e4939.