1 Introduction

Since the spread of coronavirus pandemic in 2019, healthy life has attracted more people's attention, especially the demand for effective antibacterial materials [1]. At the same time, with the advent of the intelligent society and the popularization of electronic products, billions of touch screens have been popularized in people’s lives. However, the species and quantities of bacteria on the surface of them are affecting people's health [2, 3]. At present, traditional antibacterial methods include antibiotics, inorganic antibacterial agents (such as Cu, Ag, Au) [4,5,6], and organic antibacterial agents (quaternary ammonium salts, etc.) [7]. Since the discovery of penicillin by Alexander Fleming in the early twentieth century, antibiotics have been widely utilized to treat diseases caused by bacterial infections. However, the overuse and abuse of antibiotics have led to the emergence of bacterial resistance, even the birth of "super bacteria" [8]. Metal ions on the carrier through physical or chemical means can kill the microbes by destroying the cell membrane, which has strong bactericidal ability, low toxicity, and high safety. However, the expensive cost and unpromising durability limit its application [9]. Further, organic antibacterial agents have fast sterilization speed and high stability, but they are prone to the development of drug resistance, and the decomposition products are toxic to human body [10].

In the post-antibiotic era, photocatalytic sterilization is an effective antibacterial method. Photoactivated sterilization refers to the use of light with appropriate wavelengths, ranging from ultraviolet (UV) to near-infrared (NIR), to activate photoresponsive materials [11,12,13,14,15,16]. Photoresponsive materials can absorb light energy to effectively kill pathogens in a short time through the synergy of heat and reactive oxygen species (ROS) such as O2−, ·OH, 1O2, generated by photocatalysts [17,18,19,20,1,22]. Among various semiconductor-based photocatalysts, TiO2 has appeared as the leading candidate due to its high photoactivity, superior chemical stability, and broad-spectrum antibacterial property. However, the higher energy band gap limits the application of TiO2 under visible light [23]. At the same time, the low rate of electron transfer to oxygen and the high electron–hole recombination rate of TiO2 also inhibit the production of ROS [24]. Therefore, many efforts have been devoted to improving visible light absorbance by modifying TiO2 nanoparticles (NPs), including element doping, heterojunction structure, noble metal deposition, and so on [25,26,27,28,29,30].

Among the commonly used preparation methods of TiO2 coating, atomic layer deposition (ALD) has great advantages such as accurate thickness control, outstanding atomic-scale uniformity, and saturated surface reactions compared with other deposition methods [31,32,33,34]. Further, as one of the effective methods, noble metals can act as electron traps which are close to conduction bands of semiconductors or improve the excitation of surface electrons through surface plasmon resonance effects, thereby reducing the recombination of photoinduced carriers in noble metal/semiconductor [35,36,37]. On the other hand, the cell membranes of microorganisms are mostly negatively charged, while metal ions are positively charged, so they can firmly adhere to the cell membrane by electrostatic interaction and further penetrate the cell wall into the bacterial cell membrane [38]. Thus, TiO2 deposited silver or other noble metals, which can not only effectively prevent recombination of electron–hole pairs, but also enhance the photocatalytic effect of TiO2 under visible light, thereby enhancing the antibacterial efficiency of composite materials [39, 40].

In this work, we prepared a TiO2 nanofilm by ALD, then Ag particles were deposited on the surface by photo-reduction. Due to the different Fermi levels of TiO2 and Ag, they can form Schottky barrier which can efficiently improve the separation of electrons and holes. Therefore, more ROS can be generated under visible light. The platform showed great antibacterial efficiency against Staphylococcus aureus (S. aureus, 98.2%) and Escherichia coli (E. coli, 98.6%) under the conditions of simulated sunlight for 5 min, respectively. The excellent antimicrobial effects of TiO2/Ag nanofilm under visible light illumination open up new possibilities, such as continuous visible light-powered disinfection during daytime and at night, for a broad range of surface disinfection application.

2 Experimental

2.1 Preparation of TiO2 nanofilms

Firstly, glass slides (2 cm × 2 cm) were ultrasonically cleaned with acetone, ethanol, and distilled water for 30 min to remove stains on their surface. Then TiO2 nanofilm was deposited on glass slides by plasma-enhanced ALD (MNT-P-100–43, Micro and Nanotech Co, LTD, Wuxi, China). Tetrakis(diethylamino)titanium (TDMAT) as titanium precursor and H2O as oxygen precursor were kept at 75 °C and 25 °C, respectively. The reaction temperature was 200 °C and the base pressure was 25 Pa with high purity N2 during deposition. Each cycle consisted of precursor exposure and N2 purging following a sequence of H2O:N2:TDEAT:N2 with a corresponding duration of 0.1:15:0.1:20 s. The reaction was repeated for 200 cycles to produce TiO2 nanofilms.

2.2 Synthesis of TiO2/Ag nanofilms

Silver nitrate aqueous solution (0.05 mol·L−1, 400 μl) was dropped onto surface of TiO2 nanofilm, then it was placed in a 60 °C oven. To spread the silver ions uniformly on the surface of TiO2, a xenon lamp was used to irradiate it for 50 min. The ultraviolet light could quickly reduce the silver ions into silver nanoparticles, with the color of the surface changed from transparent to light gray. The Ag nanofilm was also fabricated in the same way.

2.3 Morphological and structural characterization

The morphology of TiO2 and TiO2/Ag was observed by scanning electron microscopy (SEM, S-4800, Hitachi, Japan). Water contact angle (JC2000D Contact Angle system, POWER EACH) measurement was conducted to analyze the hydrophilic property of surfaces of TiO2 and TiO2/Ag. The surface elemental composition of the samples was obtained using X-ray photoelectron spectroscopy (XPS, ESCALAB 250Xi). Ultraviolet–visible–NIR (UV–Vis–NIR) spectrometer (UV-2700, Shimadzu, Japan) was used to obtain samples' optical properties.

2.4 Photoelectrochemical measurement

A three-electrode system in quartz glass cell in 0.5 mol·L−1 Na2SO4 aqueous solution was performed to measure the photocurrent response. Pt plate as counter electrode, an Ag/AgCl electrode as reference electrode, and an experimental sample as working electrode was used to analyze electrochemical impedance spectroscopy (EIS). The photoinduced current densities of the photocurrent response with time (i-t curve) were measured at a 0 V bias potential under xenon lamp (PLS-SXE300, Beijing Changming Technology Co., Ltd, China) irradiation. EIS tests were recorded over the frequency range from 1 × 104 to 1 × 10−1 Hz with an alternating current voltage magnitude of 5 mV under simulated visible light irradiation. The experimental samples (TiO2 and TiO2/Ag) were prepared on ITO conductive glasses (2 cm × 2 cm).

2.5 Detection of ROS

Electron spin resonance (ESR) spectra were recorded on a JES-FA200 spectrometer. 5,5-dimethyl-1-pyrroline-N-oxide (DMPO) (Sigma) was used to trap ·OH and ·O2− under xenon lamp irradiation. The glass slides, TiO2, and TiO2/Ag samples were immersed in 100 mmol·L−1 DMPO at ambient temperature under xenon lamp irradiation (PLS-SXE300/300UV).

2.6 Evaluation of antibacterial activity

The spread plate method was used to study the antibacterial activity of the samples (including pure glass slide, TiO2, Ag, and TiO2/Ag) under stimulated visible light irradiation. E. coli (Gram-negative, ATCC 25,922) and S. aureus (Gram-positive, ATCC 29,213) were used for antibacterial experiments. All bacteria were cultured in Luria–Bertani (LB) medium. The pure glass slide was set as the control group and TiO2, Ag and TiO2/Ag constituted the experimental groups. The details were as follows: 40 μl of bacterial suspension (~ 1 × 104 CFU·ml−1) was dropped onto the surface of the samples (2 mm × 2 mm glass slide). After 5 min with or without stimulated visible light irradiation, the samples were put upside down on the LB agar plates, then incubated at 37 °C for 1 h. Next, the glass slide was removed from the LB plates, and the LB plates were still incubated at 37 °C for 24 h for bacterial counts. Three parallel samples from each group were used in the antibacterial test. The antibacterial efficiency of each sample was calculated by the number of bacterial colonies on the glass slide using Eq. (1):

$${\text{Antibacterial}}\;{\text{ ratio}}\; = \;\frac{{{\text{CFU}}_{{{\text{Control}}}} - {\text{CFU}}_{{{\text{Sample}}}} }}{{{\text{CFU}}_{{{\text{Control}}}} }} \times 100\%$$
(1)

where CFUcontrol indicates the growth of bacteria without treatment and CFUsample indicates the growth of bacteria with sample cotreatment.

To better analyze the bacterial morphologies, the bacterial morphology study was evaluated using SEM. After 5 min stimulated visible light irradiation, the E. coli and S. aureus on the samples were fixed with a 2.5 wt% glutaraldehyde solution for 2 h, then washed with phosphate buffer saline (PBS, pH = 7.0). Further, the bacteria were dehydrated using graded ethanol solutions (20, 40, 60, 80, and 100 wt%) for 15 min. After drying, the morphologies and microstructures were observed by SEM.

2.7 Cytotoxicity assay

L929 fibroblast cells were cultured in Roswell Park Memorial Institute 1640 (RPMI 1640, Meilunbio) including 10% (v/v) fetal bovine serum and 1 wt% penicillin–streptomycin. The cells were placed in an incubator at 37 °C with 95% humidity and 5% CO2. The cell culture medium was replaced regularly every other day. To investigate the cytotoxicity of TiO2, Ag, and TiO2/Ag, the sterilized samples (Φ8 mm) were put into 48-well plates, and 400 µl L929 (1 × 105 cells·ml−1) were added to the plates. Then each type of sample was treated with or without light irradiation. The cells were cultured in samples for 24 h under the same conditions. After that, the culture medium was removed and 200 μl 0.05 mg·ml−1 MTT (dissolved MTT powder into pH 7.4 PBS solution) solution was added to each well. Next, the culture medium was incubated for 4 h at 37 °C in an atmosphere of 5% CO2 and 95% air until purple precipitate appeared. Then, MTT solution was removed, then 200 μl of dimethyl sulfoxide (DMSO) was added to each well with continuous shaking for 15 min to dissolve the purple precipitate. Finally, the samples were taken out, and the optical density (OD) of liquid was tested at 490 nm with a microplate reader. The cell viability was calculated using Eq. (2):

$${\text{Cell}}\;{\text{viability}}\;=\;\frac{{{\text{OD}}\;{\text{in}}\;{\text{experiment}}\;{\text{group}}}}{{{\text{OD}} \;{\text{in}}\;{\text{control}} \;{\text{group}}}} \times 100\%$$
(2)

2.8 Statistical analysis

All the quantitative data were analyzed by one-way analysis of variance (ANOVA) and expressed as mean values ± standard deviations with n = 3 (3 biologically independent samples). A student t-test was performed to evaluate the statistical significance of the variance. Values of *p < 0.05, **p < 0.01, ***p < 0.001 ****p < 0.0001 were considered statistically significant.

3 Results and discussion

3.1 Morphology and structural characterization

As schematically illustrated in Scheme 1, the TiO2 nanofilm on glass substrate was prepared by ALD. From the thermodynamic and kinetic point of view, the reaction of TiO2 by TDMAT and H2O can be separated into two half-reactions, which are described as follows [41, 42]:

$${\text{Ti}}\left( {{\text{N}}\left( {{\text{CH}}_{{3}} } \right)_{{2}} } \right)_{{4}} + {\text{ TiO}}_{{2}} {-}{\text{OH}}* \, \to {\text{NH}}\left( {{\text{CH}}_{{3}} } \right)_{{2}} + {\text{ TiO}}_{{2}} {-}{\text{O}}{-}{\text{Ti}}\left( {{\text{N}}\left( {{\text{CH}}_{{3}} } \right)_{{2}} } \right)_{{3}} *$$
(3)
$${\text{TiO}}_{{2}} {-}{\text{O}}{-}{\text{Ti}}\left( {{\text{N}}\left( {{\text{CH}}_{{3}} } \right)_{{2}} } \right)_{{3}} * \, + {\text{ 2H}}_{{2}} {\text{O }} \to {\text{ TiO}}_{{2}} {-}{\text{TiO}}_{{2}} {-}{\text{OH}}* \, + { 3}\left( {{\text{NH}}\left( {{\text{CH}}_{{3}} } \right)_{{2}} } \right)$$
(4)

where the asterisks denote the functional groups absorbed on the surface of the glass substrate. The growth mechanism of TiO2 nanofilm can be explained in the following steps:

Scheme 1
scheme 1

Illustration of synthesis process of TiO2/Ag coating

Full size image
  1. (1)

    The vapor pulse of the precursor TDMAT entered the reaction formula, then chemisorption reaction occurred on the exposed glass substrate surface.

  2. (2)

    The cleaning gas N2 brought the excess precursor TDMAT vapor which was not adsorbed by the substrate surface, and the reaction product methylamine (NH(CH3)2) out of the reaction chamber.

  3. (3)

    Water vapor entered the reaction chamber and reacted with the TDMAT precursor adsorbed on the surface of glass substrate.

  4. (4)

    The cleaning gas N2 brought the excess water vapor and the by-product methylamine out of the reaction chamber.

The first stage was the chemisorption of TDMAT molecules by active sites of surface and the exchange of ligand. It usually happened easily even at very low operating temperatures, which was attributed to high reactivity of both TDMAT and OH groups. The subsequent step was the purge of remained TDMAT and reaction products (NH (CH3)2) from the reaction chamber. The next step was to introduce H2O into the ALD chamber and perform the oxidation reaction. After the reaction process was completed, an atomically-layered TiO2 film was formed on the surface in which all Ti atoms were connected to each other by O atoms. After several cycles of reaction, TiO2 nanofilm was formed on the surface of glass substrate. Because the two half-reactions were carried out through a cycle, the film thickness could be adjusted with atomic-level accuracy and the film uniformity could be guaranteed. Subsequently, the silver nitrate solution was dropped onto the surface of the film. When TiO2 nanofilm was irradiated with UV light, holes and electrons were generated as shown in Eq. (5), where hν is the energy absorbed:

$${\text{TiO}}_{2} \xrightarrow{{{{hv}}}}{\text{h}}^{ + } {\text{ + e}}^{ - }$$
(5)

There were silver ions (Ag+) and nitrate ions (NO3−) in the AgNO3 aqueous solution. The silver ions gained a generated electron and, consequently, silver metal was deposited onto the TiO2 surface, as shown in Eq. (6):

$$ {\text{AgNO}}_{3} \rightarrow{{{{\text H_2\text O}}}}{\text{Ag}}^{ + } + {\text{NO}}_3^- \rightarrow{{{\text{e }^- }}}{\text{Ag}}\left( {\text{s}} \right) $$
(6)

SEM images exhibited the morphology of the samples. As shown in Fig. 1a, b, TiO2 deposited by ALD formed a uniform and flat coating. Further, in Fig. 1c, the SEM image of TiO2/Ag nanofilm exhibited that the size of photo-reduced silver nanoparticles was 200–800 nm, with slight agglomeration. The main reason for the agglomeration may due to the longer light irradiation. Elemental mappings further confirmed the successful deposition of silver nanoparticles on TiO2 nanofilm (Fig. 1d–f).

Fig. 1
figure 1

Morphologies and characterization of synthesized materials: SEM images of a glass slide, b TiO2 and c TiO2/Ag; d–f SEM element mapping images of TiO2/Ag; g–i light transmittance of TiO2/Ag film; j water contact angles of glass, TiO2, Ag and TiO2/Ag, in which data represent mean ± standard deviation (n = 3 independent experiments per group). Significance was assessed using a one-way ANOVA with Dunnett’s multiple comparisons (*p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001)

Full size image

Considering the subsequent antibacterial application of the coating on the touch screen, the transmittance of the sample was studied. The visible light transmittance of the TiO2 nanofilm reached 83.2%. Even after adding photo-reduced silver particles, the light transmittance could still be maintained at about 76.4%, which satisfied the standard of the screen surface (Fig. 1g). The thickness of the nanofilm was proportional to the number of ALD cycles, thus the transmittance of the samples can be controlled and adjusted (Fig. 1h, i). The hydrophilicity and hydrophobicity of the coating would affect its antibacterial effect, so water contact angle was also measured. The contact angles of glass slide, TiO2 film, Ag film, and TiO2/Ag film were 46.0°, 52.7°, 71.7°, and 84.7°, respectively (Fig. 1j). The result demonstrated the hydrophilicity of the samples, which were not conducive to the adhesion of bacteria.

The UV–Vis spectra (Fig. 2a) could reflect the optical property of TiO2, Ag, and TiO2/Ag nanofilms. Obviously, the optical absorbance of TiO2 nanofilm mainly focused on the ultraviolet region, while Ag particles could absorb partial visible light. After Ag nanoparticles were deposited on TiO2 nanofilm, the absorption edge of TiO2/Ag had a red shift compared with separate TiO2 or Ag. It may be based on the local plasmon resonance effect of Ag nanoparticles [43]. The red shift of the absorption curve led to a decrease in the band gap energy and recombination rate, thereby increasing the effect of photocatalytic activity.

Fig. 2
figure 2

Light absorption and XPS spectra of synthesized materials: a UV–Vis spectra (200–800 nm); b band gap of TiO2 and TiO2/Ag; c XPS spectrum of TiO2/Ag; corresponding narrow scan of d Ti 2p, e O 1s, and f Ag 3d

Full size image

The calculation of the semiconductor band gap was expressed by the following equation:

$$\left( {{\alpha} hv} \right)^{n} = C (hv - {E_{\text{g}}} )$$
(7)

where α is the absorption coefficient, hν is the absorption energy, C is a parameter associated with the valence and conduction band, and Eg is the band gap. The digit of n depends on the nature of the transition. In our case, for an indirect band gap, the value of n is 1/2. The variation of (αhν)1/2 with photon energy is shown in Fig. 2b. The band gaps were determined to be about 3.44 eV of TiO2 and 2.61 eV of TiO2/Ag, respectively, by extrapolation of the linear portion of the absorption coefficient α to zero for indirect-band-gap nanoparticles. When the band gap is reduced, the spectral absorption range of TiO2 will be broadened, from only absorbing ultraviolet light originally to visible light. Besides, the yield of photogenerated electron–hole pair will also increase because of the shortening of band gap [44].

XPS measurements were conducted to verify the surface components and valence states of TiO2/Ag nanofilm in Fig. 2c. The survey spectra of TiO2/Ag nanofilm indicated the signal peaks corresponding to Ag, Ti, O and C elements. The adventitious C 1s peak might lead to the contamination because the samples were exposed to air atmosphere. By fitting the high-resolution spectra, the binding energies centered at 457.91 and 463.69 eV corresponded to the 2p3/2 and 2p1/2 core levels of Ti4+ (Fig. 2d) [45, 46]. Figure 2e displays the O 1s spectra with peaks at 529.62 and 531.41 eV. The former was attributed to oxygen lattice, while the latter was correlated to surface hydroxyl groups [47]. 367.66 and 373.73 eV in the high resolution XPS (HRXPS) spectrum corresponded to Ag 3d5/2 and Ag 3d3/2, respectively; and the splitting of the 3d doublet was about 6 eV, indicating the silver was of metallic nature (Fig. 2f) [48].

3.2 Photocatalytic performance

Photocurrent response and EIS were utilized to characterize the photo-electrical properties of photocatalysts. It is widely accepted that the photocurrent intensity is decided by the separation efficiency of photogenerated carriers. As shown in Fig. 3a, the TiO2 nanofilm had a weak current response under dark conditions but exhibited a stable current when exposed to xenon lamp irradiation. What’s more, the composite film had a larger photocurrent response than the pure TiO2 film. This indicated that the recombination of Ag particles could improve the separation of electron–hole pairs and transition efficiency of carriers. As shown in Fig. 3b, the diameter of Nyquist semicircle of TiO2 was obviously shortened after the deposition of Ag particles, suggesting that the carriers transfer resistance was smaller.

Fig. 3
figure 3

Photoelectrochemical performance of samples: a photocurrent responses and b EIS plots of TiO2 and TiO2/Ag, where Z′ is real part of impedance and Z′′ is imaginary part of impedance

Full size image

Further, electron spin resonance (ESR) measurement was used to study generation process of ROS (Fig. 4). It can be seen that ·OH and ·O2− were detected under stimulated sunlight, while no signals for these two species were detected in the dark. In addition, it is indicated that the TiO2/Ag heterojunction can provide more photogenerated holes and electrons than TiO2 nanofilm to generate more ROS, which further supported the photocurrent and EIS test results. This was due to the localized plasmon resonance (LSPR) effect on the surface of Ag nanoparticles. After being excited under the irradiation of visible light, Ag nanoparticles can induce the transition of electron from the valence band (VB) of TiO2 to the conduction band (CB), leaving holes in the VB. The photo-generated electrons reacted with O2 adsorbed on the surface of TiO2 to generate ·O2−, while the holes in the CB of TiO2 have more positive potential to oxidize OH− and produce ·OH. Therefore, the TiO2/Ag nanofilm can generate more ROS [49]. As the product of photocatalysis, ROS bears the important responsibility of killing bacteria, which is exactly the antibacterial mechanism of this study. This will be further elaborated in the following text.

Fig. 4
figure 4

ESR spectra of TiO2 and TiO2/Ag with and without light irradiation: a ·OH and b ·O2−

Full size image

3.3 In vitro antibacterial property

Two typical pathogenic bacteria, S. aureus and E. coli were chosen to evaluate the antibacterial performance of the coating. It can be seen from Fig. 5a–d that after 5 min of exposure to the simulated sunlight, the bacterial colonies of the TiO2 group and the Ag group were slightly reduced compared to the control group. However, the doping of silver particles improved the photocatalytic activity. The bacterial colony reduction of the TiO2/Ag group was the most significant, whose antibacterial efficiency against S. aureus and E. coli were 98.6% and 98.2%, respectively. The results of antibacterial activities also agreed with photocatalytic results. That is to say, better photocatalytic performance means more ROS generated during visible light irradiation, thus providing better antibacterial property [50].

Fig. 5
figure 5

Antibacterial activity of TiO2, Ag and TiO2/Ag: a spread plate of S. aureus and c corresponding antibacterial efficiency; spread plate of b E. coli and d corresponding antibacterial efficiency. Data represents mean ± standard deviation (n = 3 independent experiments per group; *p < 0.05, **p < 0.01, and ***p < 0.001, ****p < 0.0001)

Full size image

The bacterial morphologies of representative samples of different groups under xenon lamp irradiation were examined by SEM. It can be seen from Fig. 6 that the bacteria in the control group were smooth, complete, and representative. In contrast, the bacterial membranes of the TiO2 group and the Ag group showed slight wrinkles and ruptures. However, the bacterial membrane of the TiO2/Ag group exhibited obvious deformation and cracking, whether it was S. aureus or E. coli. It has been verified that ROS could react with membranes through multiple mechanisms, causing them to rupture, inducing protein leakage, and even entering bacteria to damage DNA and organelles [51]. In addition to the effect of ROS, the release of Ag+ also played a certain role in the damage of the bacterial membrane. The release of Ag+ which interacted with the thiol group of a bacterial enzyme interrupted the respiratory mechanism by having a lethal effect on the bacteria [52].

Fig. 6
figure 6

SEM images of bacteria with TiO2, Ag, and TiO2/Ag under 5 min stimulated solar irradiation: a S. aureus and b E. coli

Full size image

The mechanism of the photocatalytic disinfection was schematically illustrated in Fig. 7. As one of traditional n-type semiconductors, when TiO2 was irradiated by stimulated sunlight, the electrons in its VB would be excited to the CB. These electrons would be removed by O2, so that ·O2− was produced. At the same time, the holes on the VB trapped the H2O on the surface to oxidize them to ·OH. However, when Ag particles were deposited on the surface of TiO2 and in contact with each other, since the Fermi level of TiO2 was higher than Ag, electrons will transfer from TiO2 to Ag until the Fermi levels of them were equal. Therefore, in the space charge layer formed after the electrical contact, the surface of Ag got excessive quantity of negative charges, while the TiO2 surface left excessive quantity of holes, which formed a Schottky heterojunction at the interface of TiO2 and Ag. The Ag particles acted as electron traps to leave electrons and holes in two different phases, inhibiting the recombination of electron–hole pairs, thereby improving the efficiency of photocatalysis.

Fig. 7
figure 7

Schematic illustration of mechanism of enhanced photocatalysis of TiO2/Ag, where Φm is work function of metal, Φs is work function of semiconductor, Φb is Schottky barrier, Ex is electron affinity, Ef is Fermi level

Full size image

3.4 Cytotoxicity evaluation

The MTT assay was used to examine the effects of different samples on the growth of L929 cells. Since the film was used in vitro, the cells were only cultured for one day. It can be seen from Fig. 8 that the effect of light on cell viability was almost negligible. The cell survival rate of the TiO2 group was about 90%, while the cell viability of final group could still be maintained at about 85%, indicating that the material will not have a health impact on the human body when applied in vitro.

Fig. 8
figure 8

L929 assay of cell viabilities with different samples after co-cultured for 24 h. Data are presented as mean ± standard deviations from a representative experiment. Error bar represents standard deviation (n = 3 independent experiments per group, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001)

Full size image

4 Conclusion

In this work, TiO2 nanofilm was prepared by atomic layer deposition, and then Ag particles were deposited on the film by photo-reduction to prepare photocatalytic TiO2/Ag coating. Because the Schottky heterojunction was formed at the interface between TiO2 and Ag, the electron–hole recombination rate of TiO2/Ag was suppressed. Besides, the photocatalytic efficiency of TiO2/Ag was greatly increased by generating more ROS. Its visible light catalytic antibacterial effect was better than that of untreated TiO2 nanofilm. The antibacterial efficiency against S. aureus and E. coli in vitro under the irradiation of stimulated sunlight reached 98.2% and 98.6%, respectively. Meanwhile, the deposited film had great light transmittance and biocompatibility with/without light irradiation, with a good prospect not only for the traditional antibacterial mode but also in the field of surface antibacterial application.