Nickel doping supercharges zinc titanate spinel for dye cleanup under UV light

Researchers have synthesized a family of doped zinc orthotitanate spinels, Zn1.85M0.15TiO4 where M is Cu2+, Ni2+, Co2+, or Mg2+, and shown that swapping a small fraction of zinc for these metal ions substantially improves the materials’ ability to break down methylene blue dye in water. The nickel-substituted version emerged as the clear winner: under ultraviolet light it degraded 92.23 percent of a 10 mg per liter methylene blue solution within 180 minutes, compared with 80.52 percent for the undoped compound, at an alkaline pH of 10 with 90 mg of catalyst. The degradation followed pseudo-first-order kinetics, with the nickel-doped sample posting the highest rate constant of the series at 0.0135 per minute and a regression coefficient of 0.967.

Zinc orthotitanate, Zn2TiO4, has long interested materials scientists because it crystallizes as an inverse spinel in the cubic Fd-3m space group, with zinc ions occupying both tetrahedral and octahedral sites in the lattice. At the nanoscale it behaves as an n-type semiconductor with a wide band gap of roughly 3.22 electronvolts, and it is chemically stable, which has made it a candidate for water purification, photocatalytic water splitting, solar energy devices, and even lithium-ion battery anodes. But like many wide-gap oxides it suffers from a practical weakness: photogenerated electrons and holes recombine too quickly, and its quantum yield is low, both of which cap its usefulness as a photocatalyst.

The logic behind the new study was straightforward. Earlier work on Zn1.8Co0.2TiO4 suggested that metal substitution could help, so the team lowered the dopant level from 0.20 to 0.15 per formula unit and compared four dopants side by side. To make the powders, they used a two-step solid-state reaction: zinc oxide and titanium dioxide were mixed in a 2:1 molar ratio, ground in ethanol for three hours, pre-calcined at 500 degrees Celsius for four hours with a controlled 2-degree-per-minute heating ramp, reground, and then fired at 1000 degrees Celsius for three hours. The doped samples were prepared the same way, with the corresponding metal oxides added in stoichiometric amounts.

X-ray diffraction confirmed that every sample retained the cubic spinel structure, showing sharp diffraction peaks matching the reference pattern for Zn2TiO4. Weak extra peaks corresponding to residual ZnO also appeared, which the authors attribute to incomplete reaction during solid-state synthesis, a common artifact of this preparation route. Doping left the crystallinity essentially intact but produced a subtle rightward shift of the diffraction peaks. That shift is the fingerprint of lattice contraction: the dopant cations, ranging from 0.69 angstroms for Ni2+ to 0.73 angstroms for Cu2+, are all slightly smaller than the Zn2+ ion they replace at 0.74 angstroms, so substituting them shortens local cation-anion bond distances.

The numbers bear this out. The lattice parameter dropped from 8.511 angstroms for pristine Zn2TiO4 to 8.487 angstroms for the nickel-doped sample, with the unit-cell volume shrinking from 616.51 to 611.31 cubic angstroms; nickel produced the tightest lattice of the four dopants, consistent with its smallest ionic radius. Crystallite sizes calculated from the Scherrer formula ranged from 46 nanometers for the undoped material down to 40 nanometers for ZT-Ni, with the radius mismatch between nickel and zinc inducing the strongest lattice distortion and microstrain, which suppresses crystal growth during the high-temperature treatment. Dislocation density rose in parallel, from 0.48 per square nanometer for the pristine sample to 0.61 for ZT-Ni. Williamson-Hall analysis confirmed the trend, and its plots flipped from a positive slope, indicating tensile strain, in the pristine sample to negative slopes, indicating compressive strain, in the doped ones.

Electron microscopy told a consistent story. Field-emission scanning electron microscopy revealed predominantly spherical, agglomerated particles, an expected consequence of high-temperature calcination, with average grain sizes falling from 0.15 micrometers for pristine Zn2TiO4 to 0.09 micrometers for the nickel-doped sample. High-resolution transmission electron microscopy gave particle sizes of 143, 130, 119, 123, and 125 nanometers for the pristine, Cu-, Ni-, Co-, and Mg-doped samples respectively, and its fast Fourier transform patterns showed crisp lattice fringes with measured d-spacings of 0.253 down to 0.243 nanometers, corresponding to the (222) planes of the cubic phase. Energy-dispersive spectroscopy and elemental mapping confirmed that zinc, titanium, oxygen, and the respective dopants were distributed homogeneously, with no extraneous peaks signaling impurity phases.

Infrared spectroscopy reinforced the structural picture. All samples showed the characteristic metal-oxygen vibrations of the inverse spinel: a band near 569.7 per centimeter assigned to Ti-O stretching at tetrahedral positions, Zn-O stretching bands in the 400 to 500 per centimeter region, and Zn-O-Ti vibrations between 794.5 and 977.6 per centimeter. The absence of sharp peaks above 1000 per centimeter indicated well-calcined, high-purity powders free of residual organics or adsorbed moisture, and the retention of these bands after doping showed that the basic bonding framework of the spinel survived metal-ion incorporation.

The most consequential changes appeared in the optical measurements. In the undoped material, absorption was largely confined to the ultraviolet, but doping visibly redshifted the absorption edge and lifted light absorption across the 400 to 800 nanometer window. Tauc analysis put the band gap at 3.22 electronvolts for pristine Zn2TiO4 but only 2.81, 2.85, 2.70, and 3.12 electronvolts for the Cu-, Ni-, Co-, and Mg-doped samples. The transition-metal dopants, with partially filled 3d orbitals, introduce localized impurity states inside the band gap that enable lower-energy electronic transitions; magnesium, which has a closed-shell configuration and no d-electrons, narrows the gap only slightly through minor lattice distortion.

Photoluminescence spectroscopy added a second piece of the puzzle: charge separation. Emission peaks appeared at 387, 426, 486, 516, and 571 nanometers, spanning near-band-edge exciton recombination and defect-related blue, green, and yellow emission tied to oxygen vacancies, surface states, and interstitial zinc. Lower photoluminescence intensity generally means fewer electron-hole pairs recombining radiatively and more charge carriers available for chemistry. Here the cobalt-doped sample showed the weakest emission, while the pristine and magnesium-doped samples glowed brightest, pointing to abundant defect-mediated recombination in those lattices.

The photocatalytic tests distilled all of this into performance. Optimizing conditions for the undoped material first, the team found degradation efficiency climbing from 62.11 percent at a 50 mg catalyst dose to a peak of 80.52 percent at 90 mg before dipping to 74.65 percent at 110 mg, as particle agglomeration and turbidity limited light penetration. Alkaline conditions proved best, with removal efficiencies of 61.01, 70.43, and 80.52 percent at pH 3, 7, and 10, because hydroxide ions feed hydroxyl radical formation and the negatively charged surface attracts the cationic dye. Raising the dye concentration from 10 to 40 mg per liter cut efficiency from 80.52 to 65.25 percent as active sites saturated. In the head-to-head comparison, all four doped catalysts beat the pristine spinel, with efficiencies of 85.76, 92.23, 88.19, and 87.91 percent for Cu, Ni, Co, and Mg respectively after 180 minutes, and the nickel catalyst’s 92.23 percent outperformed previously reported figures such as 70.78 percent for undoped Zn2TiO4 and 78.12 percent for a samarium-doped variant under comparable UV conditions.

The authors propose that when ultraviolet photons exceed the band gap, electrons jump to the conduction band, whose edge at roughly minus 0.25 volts versus the normal hydrogen electrode can reduce dissolved oxygen to superoxide radicals, while holes in the valence band at about plus 2.97 volts oxidize water to hydroxyl radicals. Together these reactive oxygen species mineralize the dye into carbon dioxide and water. Doping also generates trap states that separate charges and slow recombination. Notably, although the cobalt-doped sample had the narrowest band gap and the lowest photoluminescence intensity, its degradation efficiency fell slightly below the nickel catalyst’s, a reminder that photocatalysis depends on the interplay of band gap, crystallite size, crystallinity, and defect concentration rather than any single parameter. The findings are laboratory results for methylene blue under ultraviolet irradiation, and the reported optimum dosage of 90 mg in the comparative experiments should be read against the study’s own optimization data.

Subject of Research: Synthesis, characterization, and UV photocatalytic degradation of methylene blue by doped Zn2TiO4 spinel nanomaterials

Article Title: Structural, morphological, and optical properties of Zn 1.85 M 0.15 TiO 4 (M = Cu 2+ , Ni 2+ , Co 2+ , and Mg 2+ ) spinel materials for photocatalytic applications

Article References: Nigussie, M. T., Mekonen, S., Liu, Y., & Taddesse, P. (2026). Structural, morphological, and optical properties of Zn1.85M0.15TiO4 (M = Cu2+, Ni2+, Co2+, and Mg2+) spinel materials for photocatalytic applications. Results in Physics, Article 108773. https://doi.org/10.1016/j.rinp.2026.108773

Image Credits: AI Generated

DOI: 10.1016/j.rinp.2026.108773

Keywords: Zn2TiO4, spinel, photocatalysis, nickel doping, methylene blue, band gap, X-ray diffraction, photoluminescence, semiconductor nanomaterials, water treatment, solid-state synthesis, reactive oxygen species