3d metals are typically more reluctant towards oxidative addition reactions. This is due to the fact that the 3d orbitals are nodeless and thus particularly small compared to the 4d or even 5d orbitals (the latter additionally experience an expansion due to relativistic effects). Hence, 3d orbitals generally overlap less efficiently with ligand orbitals and typically form  less stable bonds. This is the reason why Ni does not reach the +IV oxidation state that easily and why Ni-H bonds are less stable.

I hope this helps; All the best

Viktoria

I agree with the above. There are two related factors implicit in Viktoria's answer: the M(II) to M(IV) oxidation is easier for 2nd and 3rd transition series than for 1st (high oxidation numbers are in general more stable lower down - MnO4- vs ReO4- is an extreme example) and because of improved d orbital-ligand overlap, the higher oxidation number does not imply so high a real charge density on the metal as you go down.

Dear Viktoria, thanks a lot! Any reference (review article) for this?

At the same time (and this is the background of my question), this would mean that expanding the d orbitals of Ni(II) (through the influence of the ligands) or reducing Ni(II) to Ni(I) should enhance the ability to activate C-H bonds, right?

Is the chance for Ni(0) even better? Or will the fully filled d10 shell be unreactive?

Fully agree; the same general trend is also found in studies about Ni(0) and Pt(0) with regard to C-F vs. C-H bond activation, in which Pt(0) easily activates C-H bonds, but Ni(0) does not. Again, the explanation is the better overlap of Pt d orbitals with the ligand orbitals, and the missing thermodynamic driving force for Ni-H bonds. Although this is found for OS 0 going to II, I would assume the same arguments hold for OS change from II to IV just as Viktoria said.

Check out this really nice combined theoretical/experimental paper by McGrady and Perutz:

http://pubs.acs.org/doi/pdf/10.1021/ja0396908

Cheers,

Andreas

Regarding the part b) of your question, I doubt that M-H complex formation is absolutely mandatory for C-H activation in terms of catalytic coupling reactions etc. A viable mechanism could also include sigma-bond metathesis as described by Robin Perutz and William D. Jones; some mechanisms also show signs of radical participation depending on the specific substrate. My gut feeling is that for most coupling reactions we have an idea how it works, but proof is available for only a small number of systems, and that maybe different mechanisms are operative depending on substrate, solvent etc. Who would have thought a few years back, that 3d metals start to play an important role in homogenous catalysis, and that e.g. Zn can catalyse borylation reactions??

Thanks Andreas,

as you can probably infer from my question, we are trying currently to activate C-H using Ni(0), Ni(I), Ni(II), Ni(III) ... . This is why I wanted some "external input". You are right, many other mechanisms might be thinkable. Lets see.

Dear, all. Thank you very much for your contributions. I hope you find this discussion as interesting as I find it.

Thanks for this very interesting question and replies. I found it fascinating bringing back the overlap of 3d orbitals of the transition metals, I taught many years at the University of Montreal, where you could find also my Notes( University Press) in French! It is true that the obvious difference between Ni, Pd and Pt is in their 3d orbitals. As you go down the Periodic Table the atomic Numbers increase: 28, 46 and 78 for Ni, Pd and Pt, respectively. Thus, the overlap of the 3d orbitals with the H orbitals increases and the M-H bond forms better weak wth Ni and stronger with Pd and Pt. the activation of the C-H bond maybe due to the the easier formation of stable Pt-C bonds by activating the C-H . The reference is given above. There are several references of inorganic books dealing with the 3d orbitals.

I'd just like to add a really interesting theoretical review article dealing with this problem:  "The Role of Radial Nodes of Atomic Orbitals for Chemical Bonding and the Periodic Table" by Martin Kaupp.

J. Comput. Chem. 2006, 28, 320–325.

I agree with the argument that nickel has less diffuse d obitals than Pd and Pt, hence it is less efficient in any C-H oxidative addition process. But importantly, nickel is the most electronegative of the three metals, hence it has more difficulty in alienating its electrons, espiacially four of them in the formal oxidation state IV. On the other hand,  a +4 metal is not fundamental for C-H activation, which more likely spans the 0 and 2 oxidation states.  What is important is that a Ni-H linkage hardly implies a limiting hydride ligand for mentioned electronegativity reasons. Rather, the linkage can be thought as having acidic character similarly to that of the complexes HCo(CO)4 and H2Fe(CO). Along this line and to push the picture to its limits, the R-H activation over Ni may be imagined to be heterolytic with formation of the R- + H+ specioes at the  Ni-C and Ni-H bonds. This description is clearly exaggerated but helps to understand intrinsically different reactivity behaviors with respect to the Pd or Pt metals..