Report comment

Please fill in the form to report an unsuitable comment. Please state which comment is of concern and why. It will be sent to our moderator for review.

Comment

The 2014 IUPAC Technical Report (TR) http://dx.doi.org/10.1515/pac-2013-0505 discusses the oxidation state (OS) of Fe nitrosyl complexes on p.1055–6. The summary IUPAC Recommendation (http://dx.doi.org/10.1515/pac-2015-1204) lists NO as one of the “non-innocent” ligands (referring to TR for details) that complicate the OS determination by their own redox ability that overlaps with that of the central atom. Evaluation of OS via the second OS algorithm (Recommendation, Section 4.2) may then lead to a fractional separation of the OS between the central atom and ligand.
The authors of the Ref. 1 paper, who isolated the brown-ring cation [Fe(H2O)5(NO)] (2+) in a crystalline salt, did not read/cite the TR. Yet they criticize the IUPAC approach to OS. For example, they claim on p.8569, 2nd column, that it ignores the back-donation: “CO is treated as a simple donor ligand in the examples”. It indeed is treated such, because the reversibly back-donated electron fraction (carbon nearly has an octet) to the M–CO bond of metal carbonyls is per definition assigned back to the Fe donor in the OS ionic approximation. This also follows from the caveat of reversible donations by a low-electronegativity Lewis base, exemplified in the Section 10.1 of the TR, or stated in Section 3 of the Recommendation.
Similarly, the back-donation from Fe to the Fe–NO bond would be neglected in the OS algorithm, because such reversibly back-donated electrons are assigned back to Fe upon the OS ionic approximation. However, the assumption in Ref.1 that this donation involves two electron pairs seems to exceed octet at the N atom. If some electron density is also “donated” to the π* MOs of the nitrogen–oxygen bond, as seems to be suggested in Ref.1 on p.8569, 1st column line 18, it no longer is back-donation, it is an internal redox equilibrium reaction. It moves the nitrosyl ligand closer to the |N=O| (−) anion end of the redox interval between the |N≡O| (+) cation and |N=O| (−) anion, thus expanding the nitrosyl NO bond distance. That distance is 1.143 Å in [Fe(H2O)5(NO)] (2+), as obtained by the single-crystal X-ray diffraction in Ref.1. Gaseous NO has 1.15 Å and the |N=O| (−) anion 1.16 Å. As these “donated” electrons might or might not affect the spin, the calculated spin in Ref.1 (wrapped up in a nonstandard terminology) is not entirely convincing in the argumentation about the OS of Fe in the brown-ring cation. Experimental data from magnetization or neutron diffraction would be better. Similarly, these electrons may or may not be affecting the reported Mössbauer isomer shift value of 0.655(3) mm/s (unclear whether or not corrected for the second-order Doppler shift), a value little lower than expected for Fe2+, approaching values for Fe2.5+.
The problem is that computational methods are imperfect due to imperfect basis sets and other circumstances, as discussed in TR. There is a limited success for MO approach towards OS in simple molecules, http://dx.doi.org/10.1002/anie.201407561. Yet Ref.1, instead of reporting the obtained experimental data, presents and discusses quantum-chemical calculations on the ions identified the crystal.
So, what is the OS in [Fe(H2O)5(NO)] (2+)? The Fe–N–O angle is 160.6° (X-ray diffraction dataset 1893483 of CDCC, not …482 listed in Supporting Info of Ref.1). This suggests that the ligand deviates from |N≡O| (+) of 180° bonding angle towards ideally 120° |N=O| (−). About half way on that path, there would be the neutral NO ligand. So, what the OS of Fe is? Close to 2+ (after all, NO is a non-innocent ligand :-) according to the bond lengths above (provided the 1.143 Å distance of the two light atoms by X-ray diffraction is reliable), as well as the angle. So, instead of defying the OS concept as the title of this Chemistry World article would have us believe, the [Fe(H2O)5(NO)] (2+) cation defies its quantum-chemical calculations; one suggests Fe(I) the other Fe(III), both incorrect. The OS follows from simple algorithms. Oxidation state is a heuristic term. AI does not need it; we do. We need simple rules, and that is no mania for classifying things, that is avoiding babel.

Your details

Cancel