Antibacterial Properties of Pure Silver Films with Nanoparticles Induced by Pulsed-Laser Dewetting Process

Silver particles are prepared by dewetting Ag ﬁlms coated on glass using a fiber laser. The size of the particles is controlled in the range of 92 nm ~ 1.2 μm by adjusting the thickness of the Ag film. The structural properties and surface roughness of the particles are evaluated by means of scanning electron microscopy. In addition, the antifungal activity of the Ag particles is examined using spore suspensions of Colletotrichum gloeosporioides. It is shown that the particles with a size of 1.2 μm achieve 100% inhibition of the conidia growth of the Colletotrichum gloeosporioides after a contact time of just 5 min. Furthermore, the smaller particles also achieve a good antibacterial activity given a longer contact time. Similar results are observed in spore germination and pathogenicity tests performed on mango fruit and leaves. Overall, the results confirm that the Ag particles have an excellent antifungal effect on Colletotrichum gloeosporioides.

a metallic film is deposited on a glass substrate and is then annealed; resulting in a dewetting phenomenon which minimizes the total energy of the free surfaces of the film and substrate.
Compared to other methods, dewetting has the advantages of large area coverage, low cost, and the production of ordered NPs. Furthermore, the rapid thermal treatment of thin Ag/Cu bilayers also enables the production of composite Ag-Cu nanoparticles [22]. However, t h e high melting temperatures of most metals are not easily achievable using conventional heating methods. Therefore, the dewetting process is generally performed using high-energy sources such as focused ion beams and lasers [ 23] .
The feasibility of pulsed laser-induced dewetting, especially that using nanosecond pulses, has attracted significant attention in the literature [24][25][26][27] . However, only scant information is available regarding the synthesis of metallic NPs using a high repetition rate laser. Accordingly, the present study prepares Ag particles with various sizes using a high repetition rate pulsed laser and an Ag film with a carefully controlled thickness. The antifungal activity of the Ag particles is investigated by examining the conidia growth and germination rate of C. gloeosporioides spores in both in vitro experiments and in vivo experiments performed using mango fruit and leaves.

Silver particle preparation
Glass substrates (AGC G2, 150 mm×150 mm x 0.7 mm) were cleaned ultrasonically in deionized water for 15 min. The substrates were dried with hot air and then transferred to an SSI-T500-1 thermal evaporator system. The base pressure of the evaporation chamber was set as 6×10 −6 Torr prior to the thermal evaporation process. Silver pellets (Ultimate Materilas Technology Co., Taiwan) were loaded into a tungsten boat in the evaporation chamber and vaporized by a resistive heat source. The growth conditions of the Ag film on the glass substrate were controlled by adjusting the current (IAg) and time of the evaporation process. In particular, IAg was fixed at 60 A and the time was adjusted as required to produce Ag thin films with thicknesses of 10 nm and 50 nm, respectively. The as-deposited films were dewetted using a fiber laser (SPI-12, UK, wavelength 1064 nm) with repetition rates ranging from 100 ~ 400 kHz and irradiation powers in the range of 8 ~ 12.5 W. For all of the samples, the laser spot size and pulse duration were set as 40 m and 30 ns, respectively. Moreover, the laser scanning speed was varied in the range of 25 ~ 1200 mm/s. The pulse energy (E) was computed as [28] where PAVG is the average power of the pulse laser and rep is the laser repetition rate. From Eq. (1), it is seen that the pulse energy reduces as the repetition rate increases. For the irradiation powers and repetition rates considered in the present study, the pulse energy varied from 5 to 250 J. The shape, size and cross-sectional height of the Ag particles produced in the laser dewetting process were observed by an optical microscope (OM, HRM-300) and a scanning electron microscope (SEM, JSM-7600F). The wettability of the different samples was evaluated using a contact angle

Spore germination inhibition assay
The possible antifungal mechanism of the NPs was investigated by means of spore germination inhibition assays. Aliquots of C. gloeosporioides spore suspension were contacted with

Morphology analysis
Disks of C. gloeosporioides (1 cm) and spores treated with Ag particles and glass were cultured in PDA medium and distilled water, respectively, in darkness at 28°C for every day. The colony count, hyphal morphology and mycelial radial growth rate were examined every day. In addition, the appressorium formation of each sample was observed at 0, 2, 4, 6 and 8 h after inoculation.

Virulence assay
Semi-ripe fruits of the mango cultivar 'Irwin' were prepared with weights of approximately 300 g. The fruits were washed in tap water, sterilized with 75% alcohol, and then washed twice with sterile distilled water. The fruits were then placed in a laminar air flow and dried under UV light for 10 minutes. Following the drying process, the samples were punctured on the same side to a depth of 2 mm with a sterilized needle. 20-μl C. gloeosporioides spore suspensions (2*10 4 conidia) were prepared and contacted with the Ag particle and glass (control) samples for 15 min. The wound and non-wound sites of the samples were then inoculated with 5-μl spore suspensions (5*10 3 conidia).
The fruits were placed in a sterilized plastic box (40 x 20 x 45 cm) with 100% relative humidity (RH) and incubated at 28°C for 7 days. Finally, the lesion lengths at each inoculation site were measured using an optical microscope.

RT-PCR analysis of gene expression
The effect of the Ag particles on the melanin synthesis behavior of the C. gloeosporioides spores was evaluated by performing the RT-PCT analysis of five related genes, namely polyketide synthase (PKS), tetra-HN reductase scytalone (THR), scytalone dehydratase (SCD), exoglucocanase (Ecg) and pectate lyase (PEL). Seven primers were designed for the RT-PCR analysis, and 18S rRNA was selected as the internal control [29]. The RNA of the spores contacted with the different Ag particles and glass (control) sample for 15 min were extracted using an RNA purification kit (BIO-GENESIS, Germany). cDNA synthesis was then performed using a commercial reverse transcription system (Magic RT Mastermix cDNA synthesis kit, BIO-GENESIS, Germany). Finally, the cDNA was used for real-time quantitative PCR tests with the following program: 50 cycles at 95°C for 10 s, 60°C for 20 s, and 97°C for 60 s.

Morphology of C. gloeosporioides conidia
To observe the effect of the Ag particles on the morphology of the C. gloeosporioides conidia, conidia samples (109 conidia/ml) were contacted with the Ag particles with a size of 1.2 μm for 3 h and were then centrifuged at 13,300 rpm for 2 min. The resulting pellets were washed with sterile water and pre-fixed with 2.5% glutaraldehyde for 2 h. The pre-fixed cells were washed with phosphate buffered saline (PBS) two times and post-fixed with 1% osmium tetroxide for 1 h. After washing with PBS three times, a dehydration process was conducted with 30, 50, 70, 80, 90 and 100% of ethanol. Finally, the fixed cells were dried and gold-coated using an ion sputtering system (HITACHI HUS-SGB, Japan) and their morphology was observed by SEM (HITACHI S-3000N, Japan). Three replications were performed for each sample. All data were statistical analysis using analysis of variance (ANOVA) test. In addition, the means of the various samples were compared by a Tukey-HSD test (P = 0.05)and Fisher's Least Significant Difference (LSD) test(P < 0.01). All of the analyses were performed using commercial statistical software (SPSS V. 20, IBM). Figure 1 presents SEM images of the synthesized Ag particles. It is seen that the particle size increases with an increasing Ag film thickness and laser repetition rate. Moreover, from Fig. 2, the corresponding particle heights are 76 nm, 288 nm and 515 nm, respectively. Figure 3 shows the relationship between the scanning speed and the laser input energy for initial Ag film thicknesses of 10 nm and 50 nm, respectively. As shown, both figures contain three distinct regions. In Region I, characterized by a low laser energy and high scanning speed, the energy is insufficient to prompt dewetting of the Ag film. As the scanning speed and laser energy increase, the input energy also increases and hence dewetting occurs; resulting in the formation of Ag particles (Region II). However, at excessive laser energies and low scanning speeds (Region III), the input energy leads to the ablation of the Ag film rather than dewetting, and consequently, no particles are formed. In general, the results confirm that the dewetting process is strongly dependent on the laser power and scanning speed. In particular, for a given scanning speed, a high laser energy results in the ablation of the Ag film, while a low laser energy results in an Ag film with many isolated island structures. For a constant laser energy, the particles tend to become larger with an increasing scanning speed due to a corresponding reduction in the thermal energy accumulation. It is noted that this finding is consistent with that reported in previous studies [30,31]. The results presented in Fig. 3 confirm that the size of the particles produced in the dewetting process also depends on the thickness of the original Ag film. For a thicker film, the dewetting process can be performed at a higher energy, and particles with a micrometer scale are produced at all the corresponding scanning speeds. For a thinner film, particles can be obtained at a lower energy, and the particle size (nanometer) depends on the particular scanning speed applied [32]. Overall, the results show that the particle size decreases with an increasing laser energy and a decreasing film thickness. Table 1 shows the surface roughness (Ra) and contact angle measurements of the various Ag samples. The surface roughness increases with an increasing particle size as a result of the greater particle height (see Fig. 2). The contact angle also increases with an increasing particle size. For example, the particles with a size of 92 nm and 1.2 m have contact angles of 43 o and 73 o , respectively. The surface roughness is another factor to affect the biological activities, suggesting that microbes are difficult to attach and move on an ultra-smooth surface [14]. For the sample with an Ag particle size of 570 nm, 100% bacteriostasis is also achieved. However, the contact time increases to 10 min. For the smallest particle size of 92 nm, the sample contains 22

Results and Discussion
colonies after 15 min. In other words, full bacteriostasis is not achieved. However, overall, the results confirm that the conidia growth suppression ability of all the samples increases with an increasing particle size and contact time. The present results are consistent with those of [33][34][35], which showed that metallic Ag has an excellent inhibitory ability against both bacteria (escherichia coli, staphylococcus aureus and pseudomonas aeruginosa) and fungi (fusarium oxysporum and botrytis cinerea). Figure 5 shows the spore germination rates of the C. gloeosporioides contacted with the different Ag samples for periods of 1 min and 15 min, respectively. For both contact periods, the spore germination rates of the treated samples are significantly lower than those of the untreated sample. Furthermore, for the Ag samples, the germination rate decreases by around 20% as the contact time is increased from 1 to 15 min. The germination rate also decreases with an increasing particle size. For example, given a contact time of 15 min, the germination rate reduces from around 29% to 15% as the particle size increases from 92 nm to 1.2 μm. It is noted that this trend is consistent with the plate test results (see Fig. 4). Anthrax appressorium formation is one of the most important structures invading the host [29]. Figure 6   observed that the melanin precipitation in the treated samples is greatly reduced compared to that in the untreated (control) sample. Among the treated samples, the sample with a 1.2-μm particle size has a particularly low melanin precipitation. Wei [29] showed that anthrax bacteria produce melanin to accumulate the swell pressure inside the press. In other words, the amount of melanin is one of the key factors determining the pathogenic ability of anthrax bacteria. Overall, the results presented in Fig. 7 suggest that Ag particle treatment is effective in suppressing melanin expression and hence in limiting the activity of anthrax bacteria. Figure 8 shows the relative expressions of the LAC1, PKS, THR, SCD, Ecg and PEL genes following contact with the treated and un-treated samples.
The results confirm that exposure to the Ag particles results in a significant reduction in the melanin synthesis compared to that for the untreated sample. Moreover, for each gene, the melanin synthesis reduces with an increasing particle size.  Table 2 show that the lesion size reduces from 13.50 mm to 11.00 nm as the Ag particle size increases from 92 nm to 1.2 m. In other words, the results are consistent with those of the plate bacteriostatic and spore germination tests, which show that the Ag sample with a particle size of 1.2 m has particularly good bacteriostatic properties and a superior suppression ability of C. gloeosporioides.

Conclusion
Silver particles have been prepared on glass slides using a combined thermal evaporation and laser dewetting process. It has been shown that the size of the particles can be controlled in the range of 92 nm ~ 1.2 μm by adjusting the thickness of the Ag film deposited on the glass slide and the laser energy applied during the dewetting process. The antibacterial performance of the Ag samples has been investigated by examining the conidia growth of C. gloeosporioides; a fungus spore associated with the anthracnose disease of mango. The results have shown that the conidia growth suppression effect increases with an increasing Ag particle size and contact time. For the largest particle size of 1.2 μm, 100% bacteriostasis is achieved within a contact time of 5 min. Furthermore, even for the smallest particle size (92 nm), a strong antibacterial performance is also observed after a contact time of 15 min. A similar size-dependent effect is also observed in the spore germination and pathogenicity tests. Overall, the present results confirm that Ag particles have an excellent inhibitory ability on C. gloeosporioides.  * significant differences at P = 0.05 (ANOVA, post-hoc Tukey-HSD test).