How Does Substitutional Doping Affect Visible Light Absorption in a Series of Homodisperse Ti₁₁ Polyoxotitanate Nanoparticles?

Abstract

Homodisperse doped polyoxotitanate nanoclusters with formulae Ti11 (MX)O14 (OiPr)17 (M=Mn, Fe or Co; X=Cl, Br or I, OiPr=isopropoxide) display strongly dopant-dependent properties. Spectroscopic solution and reflectance measurements backed up by density of states and time-dependent DFT calculations based on the determined structures, show the prominent effect of FeX substitution by decreasing the HOMO-LUMO gap of the particles. The effect is attributed to the presence of an occupied Fe β orbital halfway up the bandgap, leading to long-wavelength absorption with electron transfer to the titanium atoms of the cluster. Whereas the light absorption varies significantly with variation of the transition metal dopant, its dependency on the nature of the halogen atom or the change in dipole moment across the series is minor.

How Does Substitutional Doping Affect Visible Light Absorption in a Series of Homodisperse Ti11 Polyoxotitanate Nanoparticles—A Comment on the Band Gap Determination of the FeII Cages (Chem. Eur. J. 2015, 21, 11538)

“…The headline result in this paper, based on DFT calculations, is the reduction of the HOMO–LUMO gap resulting from FeX (X=Cl, Br, I) substitution of the cages [Ti 11 (MX)O 14 (O i Pr) 17 ] (M=Fe II compared to Mn II , Co II ). According to the authors “The effect is attributed to the presence of an occupied Fe β state half‐way up the bandgap, which leads to long‐wavelength [charge‐transfer] absorption with electron transfer to the titanium atoms of the cluster.” The experimentally‐determined direct allowed (HOMO–LUMO) band gaps from diffuse reflectance of the three FeX cages are in the range 1.43–1.59 eV (Table 4 of reference ), a result which would revolutionize not just this area, but possibly also semiconductor physics. Putting these results into context, the direct allowed band gap in the solid‐state for Fe‐doped TiO 2 is significantly higher (2.67 eV for the bulk material and 2.97–3.14 eV for nanorods) and the reported values are on a par with the band gap of the semiconductor GaAs (1.43 eV) .…”

“…Figure shows an overlay of the solution UV/Vis spectrum of the FeBr complex in chloroform (in brown, see Figure 2 b of reference ) and the predicted UV/Vis spectrum (in black, see Figure 3 b of reference ). The theoretically‐predicted charge‐transfer transition for the cage is highlighted in green.…”

“…The theoretically‐predicted charge‐transfer transition for the cage is highlighted in green. In order to facilitate direct comparison of the data we have selected one experimental UV spectrum from reference (i.e., Figure 2 b, 2 g L −1 ) so that the intensity can be normalized as far as possible with respect to the theoretical data. Unfortunately, Figure S2 of the Supporting Information does not contain the key experimental UV/Vis spectra of the FeCl and FeI cages so that the same analysis cannot be made in these cases (otherwise we would show this here also) .…”

“…In order to facilitate direct comparison of the data we have selected one experimental UV spectrum from reference (i.e., Figure 2 b, 2 g L −1 ) so that the intensity can be normalized as far as possible with respect to the theoretical data. Unfortunately, Figure S2 of the Supporting Information does not contain the key experimental UV/Vis spectra of the FeCl and FeI cages so that the same analysis cannot be made in these cases (otherwise we would show this here also) . However, Figure S4 in the Supporting Information shows that the theoretical spectra of these cages are almost identical to those for the FeBr cage and the implication in the text of the paper is that the general features of the UV/Vis spectra of the FeCl and FeI cages are the same as the FeBr cage.…”

“… Overlay of the experimental (brown) and theoretical (black) UV/Vis spectra of the FeBr cage (data has been taken from Figures 2 b and 3b of reference ).…”