Contaminant-Activated Visible Light Photocatalysis

Abstract

Pristine titanium dioxide (TiO₂) absorbs ultraviolet light and reflects the entire visible spectrum. This optical response of TiO₂ has found widespread application as white pigments in paper, paints, pharmaceuticals, foods and plastic industries; and as a UV absorber in cosmetics and photocatalysis. However, pristine TiO₂ is considered to be inert under visible light for these applications. Here we show for the first time that a bacterial contaminant (Staphylococcus aureus-a MRSA surrogate) in contact with TiO₂ activates its own photocatalytic degradation under visible light. The present study delineates the critical role of visible light absorption by contaminants and electronic interactions with anatase in photocatalytic degradation using two azo dyes (Mordant Orange and Procion Red) that are highly stable because of their aromaticity. An auxiliary light harvester, polyhydroxy fullerenes, was successfully used to accelerate photocatalytic degradation of contaminants. We designed a contaminant-activated, transparent, photocatalytic coating for common indoor surfaces and conducted a 12-month study that proved the efficacy of the coating in killing bacteria and holding bacterial concentrations generally below the benign threshold. Data collected in parallel with this study showed a substantial reduction in the incidence of infections.

Figure 1

Physical characterization of pristine anatase. (a) XRD spectrum of anatase used in this study. The black lines are data for standard anatase. (b) XPS spectrum for anatase used in this study. Adventitious carbon was used as reference. (c) Ground state absorption spectrum of anatase showing bandgap (3.2 eV) estimation. (d) XPS high resolution valence band spectrum for anatase showing the edge of valence band at 2.05 eV. (e) XPS high resolution Ti 2p spectrum for anatase showing the Ti 2p_3/2 peak at 458.1 eV and Ti 2p_1/2 peak at 463.8 eV. (f) XPS high resolution O 1 s spectrum for anatase. Peaks at 529.3 eV, 530.5 eV and 531.7 eV are attributed to oxygen in the TiO₂ lattice, surface Ti-OH and physisorbed water, respectively.

Figure 2

Electronic interactions of contaminant with anatase. (a) Ground state absorption spectra for Mordant Orange on anatase coating (MO@TiO₂) and Mordant Orange on silica coating (MO). The inset shows bandgap (2.34 eV for MO and 2.1 eV for MO@TiO₂) estimation. (b) Ground state absorption spectra for Procion Red on anatase coating (PR@TiO₂) and Procion Red on silica coating (PR). The inset shows bandgap estimation (2.03 eV for PR and 2.01 eV for PR@TiO₂). (c) XPS high resolution valence band spectrum for anatase in contact with MO shows valence band maximum at 1.8 eV with a tail at 1.3 eV. (d) XPS high resolution valence band spectrum for anatase in contact with PR shows valence band maximum at 2.05 eV with a tail at 1.6 eV. (e) XPS high resolution Ti 2p spectrum for anatase in contact with MO showing the Ti 2p_3/2 peak at 458.1 eV and Ti 2p_1/2 peak at 463.8 eV. (f) XPS high resolution Ti 2p spectrum for anatase in contact with PR showing the Ti 2p_3/2 peak at 458.1 eV and Ti 2p_1/2 peak at 463.8 eV. (g) XPS high resolution O 1 s spectrum for anatase in contact with MO. Peaks at 529.2 eV, 530.7 eV, 531.8 eV and 532.5 eV are attributed to oxygen in the TiO₂ lattice, surface Ti-OH, physisorbed water and oxygen in MO, respectively. (h) XPS high resolution O 1 s spectrum for anatase in contact with PR. Peaks at 529.2 eV, 530.5 eV, 531.5 eV and 532.3 eV are attributed to oxygen in the TiO₂ lattice, surface Ti-OH, physisorbed water and oxygen in PR, respectively.

Figure 3

Contaminant-activated visible light photocatalysis. (a) Hypothetical mechanism of contaminant-activated visible light photocatalysis. (b) and (c) Effect of light wavelength on photocatalytic degradation rate of Mordant Orange (MO) and Procion Red (PR) on anatase coatings. Light control uses silica as an inert coating. Relative contribution of wavelength bands to the overall degradation rate (0.018 hr⁻¹ for MO and 0.023 hr⁻¹ for PR). The wavelength bands evaluated were determined by the lower bound of emission (385 nm) from the fluorescent lamps used for illumination and the cutoff wavelengths (400, 495, 550 nm) of the longpass optical filters employed. The ratio of contaminants MO and PR was 1 µg contaminant per 10 µg of TiO₂ or SiO₂. N = 4.

Figure 4

Role of auxiliary light harvester in contaminant-activated photocatalysis. (a) Hypothetical mechanism of contaminant-activated visible light photocatalysis with auxiliary light harvester polyhydroxy fullerene. (b) Ground state absorption spectra for anatase (TiO₂), polyhydroxy fullerenes (PHF) on silica and anatase coating (PHF + TiO₂). (c) Pseudo first order rate coefficients for degradation of Mordant Orange dye on anatase (TiO₂) and anatase + 0.01 (w/w) PHF (TiO₂ + 0.01PHF) coatings. Dark control measures the ability of the photocatalytic coatings to degrade dye in the dark. N = 10.

Figure 5

Design of transparent coating. (a) Changes in appearance of tile surfaces achieved with application of TiO₂ coating at particle loadings of 1) 0 mg/cm²; 2) 0.128 mg/cm²; 3) 1.28 mg/cm²; and 4) 6.4 mg/cm². (b) Effect of different dispersants on particle size and zeta potential of TiO₂ formulation and performance of contaminant-activated photocatalysis with Procion Red. (c) and (d) Scanning electron micrographs of TiO₂ coatings prepared from formulations (c) without any dispersants and (d) stabilized with 0.01 M NaOH as dispersant.

Figure 6

Bacteria-activated photocatalysis. (a) Ground state absorption spectrum for Staphylococcus aureus calculated from reflectance of S. aureus deposited on a tile surface. The inset shows bandgap estimation. (b) Pseudo first order rate coefficients for inactivation of Staphylococcus aureus on various coatings. Dark control measures the ability of the rutile/anatase + 0.1 (w/w) PHF coatings to inactivate bacteria in the dark. Light control measures the ability of rutile/silica coatings to inactivate bacteria in light. The ratio of bacteria was ~80 CFU per 1 µg of TiO₂. N = 6.

Figure 7

Beta facility testing. Reduction in bacterial burden on surfaces with antimicrobial coating. For a given surface, the bars represent counts (n = 3) at times from 0 to 12 months. W = Wall; T = Thermostat; L = Locker; K = Knob; D = Soap Dispenser; R = Bathroom Rail; B = Bed Rail; C = Counter. The blue dashed line indicates the threshold of microbial counts for benign surfaces, and the yellow dashed line indicates the average microbial counts on copper surfaces in a clinical trial.