Mesoporous
anatase TiO2, mixed anatase/brookite TiO2, and
rutile-type SnO2 nanocrystalline films were
coated with ultrathin (<5 nm) TiO
x
by
the atomic layer deposition (ALD) of tetrakisdimethylamidotitanium(IV)
and water at 150 °C to fabricate SnO2/TiO
x
and TiO2/TiO
x
core/shell materials. The core/shell materials were either left
as-deposited or underwent a postdeposition heat treatment at varying
temperatures. Unsensitized films were characterized by diffuse reflectance,
X-ray photoelectron, and Raman spectroscopies in conjunction with
high-resolution transmission electron microscopy, powder X-ray diffraction,
and reductive electrochemistry. Physical characterization of the unsensitized
TiO2/TiO
x
and SnO2/TiO
x
films indicates that as-deposited
TiO
x
coatings or those annealed at temperatures
<300 °C are disordered and quasi-amorphous, and reductive
electrochemistry indicates the presence of a broad distribution of
trap states within the band gap. When annealed at temperatures >300
°C, both the physical and electronic structures of the TiO
x
coatings on TiO2 and SnO2 cores more closely resemble crystalline anatase and rutile TiO2, respectively. Interfacial charge transfer in dye-sensitized
[Ru(bpy)2(4,4′-(PO3H2)2bpy)]2+ (bpy = 2,2′-bipyridine; 4,4′-(PO3H2)2bpy = 4,4′-bis(phosphonic
acid)-2,2′-bipyridine) core/shell films was investigated using
UV–vis, photoluminescence, and transient absorption spectroscopies.
Transient absorption experiments demonstrate that charge recombination
to the surface-bound, oxidized ruthenium(III) chromophore is intricately
connected to the crystallinity of the TiO
x
coating for both TiO2/TiO
x
and SnO2/TiO
x
films. Surprisingly,
charge transfer through the as-deposited TiO
x
coatings and those that were annealed at low temperatures <300
°C on both TiO2 and SnO2 proceeds through
a tunneling-type mechanism evidenced by an exponentially decaying
charge recombination rate constant with ALD cycles. An inflection
from a tunneling-type mechanism to a thermally activated mechanism
varies with the shell thickness and the annealing temperature. The
results question the validity of fixed rectangular barrier tunneling
models frequently used to explain charge recombination in core/shell
materials. Instead, a different model is proposed to explain the charge
recombination inflection from tunneling to thermally activated as
a function of shell thickness and annealing temperature. Finally,
the relevance of the findings to devices and the “TiCl4 treatment”, a commonly used method to create TiO2/TiO
x
core/shell materials to
improve the efficiency of dye-sensitized solar cells, is discussed.