1. Introduction

The hydroxides of transition metals are known to catalyze the water oxidation.[1] Nowadays, investigations are focused mostly on mixed (Ni,Fe) hydroxide which is extremely effective in this process. The iron moiety of the hydroxide is commonly considered as responsible for overall activity of this material.[2,3] One of the major open questions in this field is the electron state of “active” iron cation and the detailed mechanism of the O-O coupling. The up-to-date experimental and DFT-computation results are the following. On base of operando Mössbauer spectroscopic studies the FeIV=O ferryl site of the (Ni,Fe) hydroxide was suggested to be responsible for the water oxidation.[4] The ferryl site can appear via proton-coupled electron-transfer (PCET) from the FeIII−OH species.[4] The ferryl model contradicts conclusion by Freibel, Bell, and Nørskov with coauthors[5] that in Ni1-xFexOOH the actual active site for oxidation of water is FeIII. The latter conclusion was put forward on base of operando X-ray absorption spectroscopy (XAS) combined with high energy resolution fluorescence detection (HERFD) and DFT modeling.[5] The ferric iron is claimed to occupy under-coordinated octahedral positions appeared on high-index surfaces (\(01\overset{\overline{}}{1}2\)) or (\(01\overset{\overline{}}{1}4\)) of NiOOH.[5]Alternatively, Goddard with coauthors suggested that the FeIV-O• species is a key intermediate which determines the activity of the (Ni,Fe) hydroxide in the water splitting.[6] However, in the latter work the FeIV-O• species only assisted the O-O coupling which actually proceeds on the NiIV-O• site. Anyway, the idea that the oxyl intermediate is an indispensable participant of the water splitting by hydroxides nicely agrees with suggestion by Siegbahn, that the O-O bond association on natural photosynthetic center necessarily involves an endergonic formation of oxygen radical.[7] The only question arises how this intermediate can appear in the hydroxide, taking into account obvious metastability of such radical species.
In our previous work on the O-O coupling, with the use of tetramer cluster model Fe4O4(OH)4of the FeIII hydroxide, the ferryl FeIV=O species was formed from FeIII-OH group via the first PCET.[8] So-formed ferryl oxygen attacks on the cubane edge forming metastable FeIII-O• oxyl group with negative spin density on oxygen. It is the spin polarization of the oxyl that makes FeIII-O• group responsible for further O-O coupling with a low barrier of 12 kcal/mol.[8] This route happens to be energetically favorable. A competing process blocking this scenario is the hydroxylation of the ferryl center by water molecule to form the HO-FeIV-OH moiety. However, the second PCET restores a bare oxo site (having the HO-FeIV-O• or HO-FeV=O configuration) capable of the O-O coupling with neighboring cubane corner oxygen. In addition, the presence of hydroxo group in the neighborhood of terminal oxo site creates a possibility of the oxo-hydroxo association at the same Fe center to form the OOH species. So-obtained O-O coupling on a single iron site is though less probable due to a relatively high barrier of 18 kcal/mol.[8]In this work the O-O coupling is investigated for water nucleophilic attack on the HO-FeIV-O• oxyl site.
As far as the above described formation of the -O• or =O terminal oxo sites are concerned, a question arises whether unavoidable hydroxylation process can deactivate these sites in water solution. One may guess that electrophilic attack of water molecule on oxo sites resulting in the FeIV=OH and FeV=OH formation competes with the nucleophilic water attack on the same sites to form hydroperoxo species Fe-OOH. To answer this question, the DFT comparative modeling of the hydroxylation and oxidation was performed using simple cubane cluster Fe4O4(OH)4 treated in our previous works.

Model

Active sites of water-oxidation catalysts based on iron hydroxides are believed to have much in common with the structure of gamma-FeOOH hydroxide. The latter consists of the double chain of edge-sharing Fe(O,OH)6 octahedra. Major structural motif of this Fe-hydroxide is a trimer consisting of iron-centered octahedra (Figure 1). The Fe-O-Fe angle is about 100 degrees, the oxo bridge sites are always three-coordinated, while hydroxo groups couple two or one Fe cation. Monocoordinated hydroxyl can appear only on the vertex of the terminating octahedron and is most probably the subject of the first PCET step initiating various scenarios for the O-O coupling process.
The cubane tetramer (FeOOH)4O (Figure 2) was chosen for modeling of the reactive center. Previously this model was proved to be quite useful to study O-O coupling since it allows to simulate oxidation acts and the O-O coupling on the vertex of terminal octahedron in the edge-sharing M(O,OH)6 (M=Co,Fe) octahedra chain.[8–10]