Effect of surface pretreatment of TiO2 films on interfacial processes leading to bacterial inactivation in the dark and under light irradiation

Abstract

Evidence is presented for radio-frequency plasma pretreatment enhancing the amount and adhesion of TiO2 sputtered on polyester (PES) and on polyethylene (PE) films. Pretreatment is necessary to attain a suitable TiO2 loading leading to an acceptable Escherichia coli reduction kinetics in the dark or under light irradiation for PES-TiO2 and PE-TiO2 samples. The amount of TiO2 on the films was monitored by diffuse reflectance spectroscopy and X-ray fluorescence. X-ray electron spectroscopy shows the lack of accumulation of bacterial residues such as C, N and S during bacterial inactivation since they seem to be rapidly destroyed by TiO2 photocatalysis. Evidence was found for Ti(4+)/Ti(3+) redox catalysis occurring on PES-TiO2 and PE-TiO2 during the bacterial inactivation process. On PE-TiO2 surfaces, Fourier transform infrared spectroscopy (ATR-FTIR) provides evidence for a systematic shift of the na(CH2) stretching vibrations preceding bacterial inactivation within 60 min. The discontinuous IR-peak shifts reflect the increase in the C-H inter-bond distance leading to bond scission. The mechanism leading to E. coli loss of viability on PES-TiO2 was investigated in the dark up to complete bacterial inactivation by monitoring the damage in the bacterial outer cell by transmission electron microscopy. After 30 min, the critical step during the E. coli inactivation commences for dark disinfection on 0.1-5% wt PES-TiO2 samples. The interactions between the TiO2 aggregates and the outer lipopolysaccharide cell wall involve electrostatic effects competing with the van der Waals forces.

Keywords: TiO2 interface; aggregation; bacterial inactivation; dark disinfection; thin film; transmission electron microscopy.

Figure 1.

Escherichia coli inactivation kinetics of RF-plasma PES-pretreated samples irradiated by actinic light for different times: (1) PES alone, (2) PES–TiO₂ not RF-plasma treated sputtered for 8 min, (3) samples sputtered for 8 min and RF-plasma pretreated for 10 min, (4) RF-plasma pretreated for 20 min, (5) RF-plasma pretreated for 30 min and (6) RF-plasma pretreated for 120 min. (Online version in colour.)

Figure 2.

(a) Fluorescence intensity versus wavelength for PES–TiO₂ samples sputtered for 8 min and pretreated for: (1) zero, (2) 10 min, (3) 20 min and (4) 30 min. Samples were irradiated for 30 min by an actinic Osram Lumilux 18 W/827 lamp. For more details see text. (b) Spectral distribution of the actinic lamp Osram Lumilux 827/18 W (Winterthur, Switzerland). (Online version in colour.)

Figure 3.

Diffuse reflectance spectroscopy (DRS) of PES–TiO₂ samples sputtered for 8 min. The PES was RF-plasma pretreated for: (1) zero, (2) 10 min, (3) 20 min and (4) 30 min.

Figure 4.

(a) Fluorescence intensity versus wavelength for PES–TiO₂ sputtered for 8 min on RF-pretreated samples for: (1) zero, (2) 10 min, (3) 20 min and (4) 30 min. Samples were irradiated for 30 min by an actinic Osram Lumilux 18 W/827 lamp using a 400 nm filter as shown in the insert. (b) E. coli inactivation PES–TiO₂ samples sputtered for 8 min and irradiated by an actinic Osram Lumilux 827/18 W lamp in the presence of the 400 nm filter shown in figure 4a: (1) PES alone, (2) PES–TiO₂ non-pretreated (3) RF-plasma PES–TiO₂ samples pretreated for 10 min, (4) 20 min, (5) 30 min and (6) 120 min. (Online version in colour.)

Figure 5.

(a) TEM of (i) PES and (ii) PES–TiO₂ sputtered for 8 min after RF-plasma pretreatment for 30 min (E indicates the epoxide used in the preparation and cutting of the sample). (b) X-ray diffraction of PES (lower trace) and an RF-plasma pretreated PES–TiO₂ sample pretreated for 30 min and sputtered for 8 min with TiO₂. For more details see text. (c) XPS O1s spectra of PES–TiO₂ sputtered for 8 min and RF-pretreated showing the bands assigned to Ti–O, Ti–O–C and the OH_surf functionalities.

Figure 6.

PE–TiO₂ samples sputtered for 8 min after RF air plasma pretreatment for: (1) 15 min, (2) 20 min, (3) 30 min and (4) 5 min.

Figure 7.

(a) The hydrophobic–hydrophilic kinetics for a PE–TiO₂ sample sputtered for 8 min after being RF air plasma pretreated for 15 min under solar simulated light and (b) kinetics of the dark reverse reaction for the same sample.

Figure 8.

TiO₂ hydrophobic to hydrophilic conversion under solar simulated light showing the TiOH metastable state on the TiO₂ surface (for details see [29,30]).

Figure 9.

Atomic force microscopy topography of PE–TiO₂ samples: (a) PE–TiO₂ sputtered for 8 min no pretreatment, (b) PE–TiO₂ sputtered for 8 min and RF-plasma pretreated at 1 torr for 15 min, (c) PE–TiO₂ sputtered for 8 min and RF air plasma pretreated for 15 min, (d) PE–TiO₂ sputtered for 1 min and RF air plasma pretreated for 15 min. (Online version in colour.)

Figure 10.

(a) SEM of an PE–TiO₂ sample RF air plasma pretreated for 15 min and sputtered for 8 min (BF, bright field), (b) HAADF image taken on the same sample, (c) EDX mapping of Ti of this sample and (d) EDX mapping of O. For more details see text.

Figure 11.

(a,b) O1 s deconvolution of PE–TiO₂ XPS spectral peaks for samples sputtered for 8 min and: (a) no pretreatment, (b) after RF air plasma pretreatment for 15 min. (c,d) Ti2p peak deconvolution of a PE–TiO₂ sample sputtered for 8 min and pretreated with RF air plasma for 15 min (a) before and (b) after bacterial inactivation under solar simulated light involving Ti³⁺/Ti⁴⁺ oxidation states within 60 min, the bacterial inactivation period.

Figure 12.

TEM of TiO₂–PES 5% (hydrothermal) interacting with the E. coli cell wall in the dark at (a) time zero (b) time 30 min (c) time 120 min. For more details, see text.