Conclusions

We have performed static DFT calculations on the hydrated Fenton-like reagent in vacuo, [Fe$^{\rm {III}}$(H$_2$O)$_5$(H$_2$O$_2$)]$^{3+}$, and ab initio (DFT) molecular dynamics simulations of the Fenton-like reagent in aqueous solution, Fe$^{3+}$/H$_2$O$_2$(aq), to determine and characterize the active intermediates. The static DFT calculations on the hydrated iron(III) complexes in vacuo showed that the direct formation of active intermediates, such as the OH. radical or a high-valent iron oxo species ([Fe$^{\rm {V}}$O]$^{3+}$) are endothermic by as much as 61 and 57 kcal/mol, respectively. This is in agreement with the experimentally observed much lower reactivity of the Fenton-like reagent compared to Fenton's reagent (Fe$^{\rm {II}}$/H$_2$O$_2$), for which we found that the formation of the highly reactive ferryl ion ([Fe$^{\rm {IV}}$O]$^{2+}$) is exothermic by 8 kcal/mol in vacuo.

The question how the Fenton-like reagent can still be active, and what the important reaction intermediates are, has been answered by ab initio (DFT) molecular dynamics simulations of the Fenton-like reagent in aqueous solution. The solvent effects prove to play a crucial role in the two reaction steps that lead from the H$_2$O$_2$ coordination in a [(H$_2$O)$_5$Fe$^{\rm {III}}$(H$_2$O$_2$)]$^{3+}$ complex to both OH. radicals and ferryl ions, [Fe$^{\rm {IV}}$O]$^{2+}$ as active oxidative species. The first step is donation of the $\alpha$-proton of coordinated hydrogen peroxide to the solvent, as already suggested by the calculations in vacuo and confirmed by the AIMD simulations. The second step is suggested by our static DFT calculations to be O-O bond homolysis, producing the ferryl ion and a hydroxyl radical. This reaction step is uphill by 43 kcal/mol in vacuo, which is reduced to 26.1 kcal/mol upon hydrolysis of a water ligand. The AIMD simulations indicate that the solvent effects lower the barrier for O-O bond homolysis in [(H$_2$O)$_5$Fe$^{\rm {III}}$(OOH)]$^{2+}$ significantly, to a free energy barrier at $T=300$ K in aqueous solution of approximately $\Delta A^\ddagger = 21$ kcal/mol, with concomitant hydrolysis of a water ligand. The important iron(III)hydroperoxo(aq) intermediate has been investigated by comparing calculated vibrational properties with experimental data. Comparison of the calculated vibrations of the low-spin Fe(III)OOH confirms the influence of the spin-state of iron on the Fe-O and O-O bond strength, proposed in the literature. As the O-O bond strength is decreased and the Fe-O bond strength is increased in the low-spin Fe(III)OOH compared to the high-spin Fe(III)OOH, we expect the reaction free energy barrier for the O-O homolysis to be significantly lower for low-spin complexes.