Subtle Variations in Surface Properties of Black Silicon Surfaces Influence the Degree of Bactericidal Efficiency
- 1.5k Downloads
KeywordsBlack silicon Nanoarchitecture Bactericidal efficiency Deep reactive ion etching (DRIE) Neural network analysis
Three types of black silicon (bSi) surface were successfully fabricated using deep reactive ion etching with pillar heights (652.7–1063.2 nm) and density (8–11 tips per µm2).
Less bactericidal bSi surfaces were found to contain nanopillars of heights reaching 1000 nm that were not always well separated, lower pillar density (8 tips per µm2), and aspect ratios of 8.8.
The recent discovery that some nanostructured surfaces exhibit a mechano-bactericidal effect [1, 2, 3, 4, 5, 6] has prompted a number of extensive studies to be undertaken toward the fabrication of new nanopatterned materials that also possess antibacterial properties [7, 8, 9, 10, 11, 12, 13, 14]. Much of this inspiration has been drawn from naturally occurring bactericidal surfaces such as those possessed by cicada Psaltoda claripennis and dragonfly Diplacodes bipunctata wings [1, 5]. It has been noted in some studies that the efficiency with which the cells rupture by nanostructures may be dependent on the nanopattern of the surface [1, 2, 3, 15, 16], e.g., defined surface parameters of 280-nm pillars height, approximately 60-nm tip diameter and approximately 60-nm spacing have proven to be more effective at killing 85% and 89% of Staphylococcus aureus and Pseudomonas aeruginosa, respectively . It was also reported that approximately 99% of the Pseudomonas aeruginosa cells coming into contact with the surface of Psaltoda claripennis wings have been shown to be eliminated . Kelleher et al.  reported that the pillars present on the wings of some cicada species, with average height dimensions of 180–250 nm, were able to eliminate approximately 100% of Pseudomonas fluorescens cells coming into contact with the surface.
The black silicon (bSi) surface is the first synthetic analog of natural bactericidal surfaces of the Diplacodes bipunctata dragonfly wing . Previously bSi surfaces have been widely used in renewable energy applications such as photovoltaic and solar cells due to their property of having the low reflectivity [17, 18] and were successfully fabricated using a number of nanofabrication techniques including reactive ion etching , electrochemical etching , and laser treatment . Biomimetic bSi was demonstrated to be effective toward different types of bacteria, including common human pathogens such as Pseudomonas aeruginosa and Staphylococcus aureus .
Despite the striking similarity between the natural and synthetic nanostructured surfaces, some variations in bactericidal activity were observed among wings of different dragonfly species  and the biomimetic bSi surfaces . In light of these results, the aim of the current work was to investigate the relationship between the characteristics (nanopillar density, height, and interpillar distance) of the bSi nanopatterned surfaces and their bactericidal efficiencies. These data provide useful insights into way in which the design and fabrication of mechano-responsive antibacterial surfaces can be made more effective.
3 Materials and Methods
3.1 Reactive Ion Beam Etching
Reactive ion etching (RIE) using SF6 and O2 was performed for 5 min to produce the nanopillars on the surface of silicon wafers (WRS Wafers) using an Oxford PlasmaLab 100 ICP380 instrument. The nanopillars were fabricated with a high degree of precision in terms of accuracy in size and position, allowing for a systematic study of the surface topology [1, 19]. Details of reactive ion etching are found in Supplementary Data Section S1.1. Three fabricated bSi surfaces, designated bSi-1, bSi-2, and bSi-3, were investigated in this study for their surface characteristics and bactericidal efficiency.
3.2 Contact Angle Measurements
Static water contact angles were measured on the bSi surfaces using the sessile drop method [22, 23]. The contact angle measurements were carried out in air using an FTA1000c instrument equipped with a nanodispenser (First Ten Ångstroms, Inc., Portsmouth, VA, USA.). The volume of the droplets used for analysis was approximately 1.0 µL. The contact angles were measured by recording 50 images over 2 s with a Pelcomodel PCHM 575-4 camera and measuring the contact angles after the droplet had been rested on the surface for approximately 1 s. The surface wettability was determined on five different locations on the surface of three separate bSi samples of each type of the surfaces.
3.3 XPS Analysis
X-ray photoelectron spectroscopy (XPS) was performed using a Kratos Axis Ultra DLD X-ray photoelectron spectrometer (Kratos Analytical Ltd., UK) equipped with a monochromatic X-ray source (Al Kα, hυ = 1486.6 eV). Details of reactive ion etching are found in Supplementary Data Section S1.2. The relative atomic concentration of the elements detected by XPS was quantified on the basis of the peak area in the survey spectra, using the sensitivity factors appropriate for the Kratos instrument. High-resolution scans were performed across each of the C 1s, O 1s, F 1s, and Si 2s peaks.
3.4 Surface Characterization
The surface topography and architecture were analyzed using an Innova scanning probe microscope (Bruker, USA). Scans were performed in tapping mode under ambient temperature and pressure conditions, using silicon cantilevers (MPP-31120-10, Veeco, USA.) with a spring constant of 0.9 N m−1 and a resonance frequency of approximately 20 kHz. Scanning was performed perpendicular to the axis of the cantilever at a scan speed of 1 Hz. Initially, 2.5 × 2.5 µm2 areas were analyzed to evaluate the overall homogeneity of the surface, prior to generating topographical profiles at five different locations on each bSi surface.
High-resolution electron micrographs of the bSi surfaces were recorded using a field-emission scanning electron microscope (FE-SEM; ZEISS SUPRA 40 VP, Oberkochen, BW, Germany) at 3 kV under 10,000×, 30,000×, 70,000×, and 110,000× magnification using the method described in our previously published studies [1, 2]. The nanopillared patterns present on the bSi surfaces were analyzed using ImageJ® software package using a fast Fourier transform (FFT) algorithm [24, 25]. The average FFT images are obtained by averaging over FFT transformed tiles of 512 × 512 pixels fitting into SEM images at 10,000× magnification with displacements of 100 pixels to each other.
3.5 Neural Network Analysis for Pillar Tip Detection and Distribution
An extended training set of 504 pictures per species was generated by rotating each of the seven pictures by 24 equidistant angles, as well as rescaling to 90, 100, and 110% of the original size. Training was performed by (on average) 2000 forward passes per extended training set picture, with full error back propagations. Feed-forward and error-backward propagations employ well-known sigmoidal characteristics based on the logistic function. The target signals for the two output neurons were [1, 0] corresponding to E, and [0, 1] corresponding to P, respectively. The pillar tip positions in the SEM micrographs were detected by forward passing patches of 12 × 12 pixels through the network. Pillar tip positions have been identified by finding the regions of maximum response of the second output neuron (Fig. 1b).
3.6 Bacterial Strains, Growth Conditions and Sample Preparation
Two strains of pathogenic bacteria, Pseudomonas aeruginosa ATCC 9027, Staphylococcus aureus CIP 65.8T, which are responsible for a large number of postoperative infections, were obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA) and the Culture Collection of the Institute Pasteur (CIP, Paris, France) [1, 27]. The selected strains are representatives of two large prokaryotic lineages, namely Gram-negative and Gram-positive bacteria, whose responses on nanostructured surfaces will be typical for taxonomically related bacterial taxa. Prior to each experiment, bacterial cultures were refreshed on nutrient agar from stocks (BD, USA). Fresh bacterial suspensions were grown overnight at 37 °C in 5 mL of nutrient broth (BD, USA). Bacterial cells were collected at the logarithmic stage of growth (data not shown), and the density of bacterial suspensions was adjusted to OD600 = 0.1.
3.7 Cell Viability Analysis
Confocal laser scanning microscopy (CLSM) was used to visualize the proportions of live and dead cells using a LIVE/DEAD® BacLight™ Bacterial Viability Kit, L7012. A mixture of SYTO® 9 and propidium iodide fluorescent dyes (Molecular Probes™, Invitrogen, Grand Island, NY, USA). SYTO® 9 permeated both intact and damaged membranes of the cells, binding to nucleic acids and fluorescing green when excited by a 485-nm wavelength laser. Propidium iodide alone entered only the cells with significant levels of membrane damage, which are considered to be non-viable, and binds with higher affinity to nucleic acids than SYTO® 9. Bacterial suspensions were stained according to the manufacturer’s protocol and as described in our previously published articles [1, 5, 28] and imaged using a FluoView FV10i inverted microscope (Olympus, Tokyo, Japan).
To quantify the bactericidal efficiency of bSi surfaces, two bacterial strains were incubated on each type of the surfaces for 3 h. Viability assays of P. aeruginosa and S. aureus cells were performed using a standard plate count assay . Further details are given in Supplementary Data Section S1.3. The number of colony-forming units was regarded as being equivalent to the number of live cells in suspension . The bactericidal efficiency was estimated as the number of inactivated cells per square centimeter of surface, per minute of incubation time, relative to the control surfaces.
3.8 Statistical Analysis
Data were expressed as mean ± standard error of the mean. For comparison of two groups, p values were calculated by two-tailed paired Student’s t test. In all cases, p values < 0.05 was considered to be statistically insignificant.
4 Results and Discussion
4.1 Characterization of Black Silicon Surfaces
Summary of surface wettability, surface roughness analysis and geometrical parameters of nanopillars of the bSi surfaces
Water contact angle, θ (degrees)
130.8 ± 3.2
100.9 ± 1.6
8.1 ± 1.2
Average roughness (nm) Ra
82.3 ± 29.6
110.3 ± 27.6
124.7 ± 17.7
Root-mean-square roughness (nm) Rq
103.7 ± 37.3
136.5 ± 34.2
156.8 ± 22.2
836.8 ± 91.2
657.9 ± 74.3
1063.2 ± 159.5
Tip width (nm)
100.1 ± 36
110.3 ± 26.9
120.5 ± 17.1
Interpillar distance (nm)
153.1 ± 55.3
135.6 ± 33.9
197.4 ± 28.0
Density (number of tips per µm2)
11 ± 4
10 ± 3
8 ± 2
8.4 ± 2.9
6.0 ± 1.8
8.8 ± 2.0
Si (total) (At.%)
The bearing ratio of each bSi surface type was determined from the statistical AFM data, since the protrusion curvatures were present in both the x, y, and z (height) planes. The variation in the bearing ratio was dependent on the height of the nanopillars, as shown in Figs. S3 and S4. The bearing curve reflected the surface area at a specific depth with respect to the entire area being analyzed. For all samples shown in Fig. S4, at a height of about 200 nm, the bearing curve started to increase in the bearing area fraction and approaches a constant bearing area level within several 100 nm. The increase indicated the volume of everything present above the surface, which is the volume of the pillars. It appeared that the bSi-1 and bSi-2 samples reached a constant bearing area level at approximately 500 nm, while the bSi-3 samples reached a constant bearing area level at approximately 800 nm, highlighting the nanopillar height present on each surface.
A FFT analysis of the top view of the SEM images of the bSi surfaces confirmed the isotropy of the nanopillar pattern, showing a variation in the average distance between the pillars resulting in a broad halo ring in the FFT images (Fig. 2b and Fig. S1B). The halo ring peak position is found in a range of wave numbers (q) between 0.016 and 0.021 (nm–1) indicating a most pronounced regularity of surface structures on scale of approximately R = 330 and 390 nm. The radially averaged Fourier spectra are shown in Fig. S1B for all samples bSi-1 to bSi-3. No significant differences between the radial spectra of the samples bSi-1 to bSi-3 were detected. Note that the shown spectra convolute spatial relations between pillars as well as the SEM projected internal pillar structures.
4.2 Neural Network Analysis and Pillar Pair-Distributions
4.3 Bactericidal Activity
The results of this work provide evidence that despite bSi substrata having a nanoarchitecture with visual similarity, including undistinguishable by neural network and FFT analysis top views, the bactericidal efficiency of such substrata can vary. The three types of bSi surfaces investigated here were varied in particular in pillar height (652.7–1063.2 nm) and density (8–12 tips per µm2). It is suggested that while none of the individual surface nanotopographic parameters could be directly correlated with the variations in bactericidal activity, the highest bactericidal efficiency may be achieved through the combination of different parameters. Less bactericidal bSi surfaces were found to contain nanopillars of heights reaching 1000 nm that were not always well-separated, lower pillar density (8 tips per µm2) and aspect ratios of 8.8. The exact relationship between the nanopattern parameters and the bactericidal properties of the surface warrant a more rigorous investigation.
The authors acknowledge funding from Marie Curie Actions under EU FP7 Initial Training Network SNAL 608184. The authors gratefully acknowledge the RMIT Microscopy and Microanalysis Facility (RMMF) for providing access to the characterization instruments.
- 3.S. Kelleher, O. Habimana, J. Lawler, B. O’Reilly, S. Daniels, E. Casey, A. Cowley, Cicada wing surface topography: an investigation into the bactericidal properties of nanostructural features. ACS Appl. Mater. Interfaces 8(24), 14966–14974 (2016). https://doi.org/10.1021/acsami.5b08309 CrossRefGoogle Scholar
- 10.P.K. Sahoo, R. Janissen, M.P. Monteiro, A. Cavalli, D.M. Murillo et al., Nanowire arrays as cell force sensors to investigate adhesin-enhanced holdfast of single cell bacteria and biofilm stability. Nano Lett. 16(7), 4656–4664 (2016). https://doi.org/10.1021/acs.nanolett.6b01998 CrossRefGoogle Scholar
- 19.H. Jansen, M. De Boer, R. Legtenberg, M. Elwenspoek, The black silicon method: a universal method for determining the parameter setting of a fluorine-based reactive ion etcher in deep silicon trench etching with profile control. J. Micromech. Microeng. 5(2), 115–120 (1995). https://doi.org/10.1088/0960-1317/5/2/015 CrossRefGoogle Scholar
- 22.V.K. Truong, R. Lapovok, Y.S. Estrin, S. Rundell, J.Y. Wang, C.J. Fluke, R.J. Crawford, E.P. Ivanova, The influence of nanoscale surface roughness on bacterial adhesion to ultrafine-grained titanium. Biomaterials 31(13), 3674–3683 (2010). https://doi.org/10.1016/j.biomaterials.2010.01.071 CrossRefGoogle Scholar
- 25.S. Sasaoka, K. Saito, K. Higashi, W. Limwikrant, K. Moribe, S. Suzuki, K. Yamamoto, Design of one-dimensional power spectrum using two-dimensional fast Fourier transform for discrimination of paper-based kraft tapes. Forensic Sci. Int. 257, 329–336 (2015). https://doi.org/10.1016/j.forsciint.2015.09.016 CrossRefGoogle Scholar
- 27.D. Campoccia, L. Montanaro, C.R. Arciola, The significance of infection related to orthopedic devices and issues of antibiotic resistance. Biomaterials 27(11), 2331–2339 (2006). https://doi.org/10.1016/j.biomaterials.2005.11.044 CrossRefGoogle Scholar
- 28.L. Boulos, M. Prévost, B. Barbeau, J. Coallier, R. Desjardins, LIVE/DEAD® BacLight™: application of a new rapid staining method for direct enumeration of viable and total bacteria in drinking water. J. Microbiol. Methods 37(1), 77–86 (1999). https://doi.org/10.1016/S0167-7012(99)00048-2 CrossRefGoogle Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.