CA measurements of hydrophobic/hydrophilic ZnO, SiOx, TiO2, and DLC nanostructures used for medical applications.
There are many studies in literature concerning contact angle measurements on different materials/substrates. It is documented that textiles can be coated with multifunctional materials in form of thin films or nanoparticles to acquire characteristics that can improve the protection and comfort of the wearer. The capacity of oxide nanostructures to inhibit fungal development and neutralize bacteria is a direct consequence of their wetting behavior [1–6]. Moreover, the radical modification of wetting behavior of nanostructures from hydrophilic to hydrophobic when changing the pulsed laser deposition (PLD) ambient will be thoroughly discussed.
- contact angle measurements
- oxide and diamond like carbon nanostructures
- textile functionalization
- medical applications
- pulsed laser deposition
- combined radio frequency plasma enhanced chemical vapor deposition
- magnetron sputtering
Wetting is the ability of liquids to keep in contact with solid surfaces. It is a direct result of intermolecular interactions, which occur when two media (liquid and solid) are brought together. Wettability studies usually involve the measurements of contact angle (CA), which indicates the degree of wetting when a solid and liquid interact. A low CA (<90°) corresponds to high wettability, and the fluid will spread over a large area of the surface. A high CA (>90°) corresponds to low wettability, and the fluid will minimize contact with the surface and form a compact liquid droplet. CA>150° indicates minimal contact between the liquid droplet and the surface and corresponds to a superhydrophobic behavior.
In the case of a liquid droplet on an ideal solid surface (which is flat, rigid, perfectly smooth, chemically homogeneous, and has zero CA hysteresis), which forms a CA (θ), the general formula of the well-known Young’s equation (which assumes a perfectly flat and rigid surface)  that describes the balance (Figure 1a) between the surface tension of the liquid/vapor and that of the liquid/vapor and the interfacial tension of the solid/liquid is as follows:
In reality, only a few solid surfaces are actually flat. The surface roughness is therefore one important parameter that should be taken into consideration when assessing the wetting behavior of a surface [8, 9]. This influence can prove significant for static and dynamic wetting.
Besides Young’s theory, the Wenzel (Figure 1b) and Cassie (Figure 1c) regimes [9, 10], or an intermediate state between these two models (Figure 1d) , are generally used to correlate the roughness of the surface with the apparent CA of a liquid.
Several authors modeled the effect of surface roughness over CA [12–14]. The basic idea was to account for roughness through r, which is the ratio of the actual to projected area. Thus, and where ALS and ASV are the liquid-solid and solid-vapor areas, ALSapp and ASVapp are the liquid-solid and solid-vapor areas. In this case, Eq. (1) becomes
Due to surface roughness, for CA <90°, the apparent CA will decrease, while for CA>90°, its value will increase. This corresponds to the Wenzel case, as illustrated in Figure 1b, where the liquid completely fills the air pockets of the rough surface at the contact site. If the CA is large and the surface rough, the liquid may trap air. As a consequence, a composite surface effect will appear, as illustrated in Figure 1c.
In the Cassie model , it is presumed that in the grooves the air is trapped under the liquid droplet. This determines the appearance of a composite surface (Figure 1c). The chemical heterogeneity of a rough surface can be related, in this model, to the apparent CA, θapp, through the following formula:
with fS and fV as the area fractions of the solid and vapor on the surface, respectively. Since , θS = θ, and θV = 180°, Eq. (3) can be written as follows:
where is the CA on a smooth surface .
From Eq. (4), it follows that in case of a true value of CA >90°, the surface roughness will determine an increase of CA’s apparent value. In contradiction to Wenzel regime, in the Cassie mode, the CA value will increase even for CA values <90º due to the air pockets trapped under the liquid droplet . It was shown that, when applying a physical deformation to a spherical droplet, a variation from the Cassie to Wenzel state can be achieved [16, 17] (Figure 1d). In some cases, a transition between these two modes can also occur [18–20]. Therefore, the droplet will fill the air pockets of the rough substrate resulting in a decrease of the apparent value of CA. In these cases, for the estimation of CA values, Eqs. (2) and (4) can be used before and after the transition, respectively. Taking into account these results, the following equation can be written :
where θth represents a threshold value between Wenzel and Cassie states.
1.1. CA measurements of nanostructured surfaces
Immediately after an implant is introduced inside the human body, the first events that occur are the wetting of the material by the physiological fluids, followed by attachment of cells to the implant surface . In order to evaluate the wetting behavior of a system, plenty of quantitative (CA, imbibition, and forced displacement, and electrical resistivity wettability) and qualitative (imbibition rates, microscope examination, flotation, glass slide, relative permeability curves, permeability/saturation relationships, capillary pressure curves, capillarimetric method, displacement capillary pressure, reservoir logs, nuclear magnetic resonance, and dye adsorption) methods have been developed . Among these, CA measurement is probably the most adopted technique to investigate the average wettability of a surface . Moreover, this type of investigation has been extensively applied to assess the wetting behavior of different nanostructured surfaces, used for various medical applications. Some relevant literature examples limited to oxides (ZnO, TiO2, SiOx) and diamond-like carbon (DLC) structures are briefly summarized in Table 1.
|Material||Envisaged application||Cited reference|
|ZnO||Self-cleaning coatings and antifogging materials|||
|Antibacterial properties||[25, 26]|
|SiO x||Antibacterial properties|||
|Cells migration on artificial surfaces|||
|Microbial fuel cells and bioremediations|||
|DLC||Temporary blood-contacting medical devices (cardiovascular and interventional devices, artificial organs, pacemakers)||[35, 36]|
|Femoral head and the acetabulum hip joint components|||
|Resistance to corrosion|||
|Resistance to microbial adhesion|||
CA can be classified into static or dynamic. Static CA is measured when liquid droplet is standing alone on the surface, without needle insertion, and the solid/liquid/air boundary is not moving. These measurements are used in quality control and research and product development. One can measure the dynamic CA when the solid/liquid/air boundary is moving. In this way, advancing and receding CA are measured. CA hysteresis, which represents the difference between these two angles, comes from surface chemical and topographical heterogeneities, solution impurities absorbing on the surface, or swelling, rearrangement or alteration of the surface by the solvent [41, 42].
The hydrophobic behavior of a surface is generally assessed by the apparent water CA, in static measurements. Moreover, when evaluating a surface repellency, one should take into consideration the sliding-down (which is evaluated by measuring the sliding angle, α, at which a liquid droplet begins to slide down an inclined plate) and rolling-off behaviors of liquid droplets . Due to the CA hysteresis [43, 44], the liquid droplets do not slide off easily on a surface presenting a high value of static CA. Eq. (6) [43, 45] quantitatively describes the relationship between the hysteresis and the sliding angle:
where θA and θR are the advancing and receding CAs, respectively (Figure 2), g is the gravitational force, m is the mass, and w is the width of the droplet.
Advancing and receding CA represent the maximum and minimum values that can be measured on the surface for the static CA. Due to the increasing interest on smart materials (self-cleaning and superhydrophobic), the dynamic CAs and CA hysteresis are highly applied [46, 47]. For self-cleaning applications, it is important that sliding angles (angle of the substrate which has to be tilted in order to move the droplet) to present small values.
From Eq. (6), it can be inferred that a lower droplet mass and smaller difference between the advancing and receding CAs will result in a smaller angle α. It is worthy to note that the surface roughness has a strong effect on the CA hysteresis .
Zisman observed for the first time that increases linearly as the surface tension of the liquid () decreases [48, 49]. He investigated the wettability of solids by determining the critical surface tension using CA. This method is used to determine the so-called critical surface free energy (), that differs from the solid surface free energy, . According to his method, the value of of a solid is equal to the value of of a liquid, which is in contact with the solid and for which the CA is zero. The value for γc is determined from empirical investigations, consisting of the CA measurements for the studied solid and the liquids of a homologous series of organic compounds like n-alkanes. The values are plotted with the y-axis corresponding to the cosine values of the CA () and the x-axis relating to the values for the studied liquids. The values of for the liquids of a series of n-alkanes form approximately a straight line. Extrapolation of this line to the point of yields the value of equal to .
Despite the fact that is not the solid surface free energy, the critical surface tension has been shown to correlate with the known surface chemistry of several solids.
The Zisman method has been widely used to assess the critical surface tension of different organic films or low-energy solids deposited on high-energy solids (e.g., metals, glass [50, 51]). In this approach, by using series of homologous nonpolar liquids (e.g., n-alkanes), one can obtain the total solid surface energy of a nonpolar solid and the dispersion component () of the total surface energy of a polar solid. We note that, when using polar liquids on polar and nonpolar solids, one can obtain the deviation from rectilinear relation. Also by using polar liquids, the determination of any component of the solid free energy it is not possible.
2. Alternative deposition techniques employed for the synthesis of hydrophobic/hydrophilic nanostructured surfaces (thin films or nanoparticles)
Many methods were employed to synthesize hydrophobic or hydrophilic nanostructures (thin films, TFs, and nanoparticles, NPs), and some literature examples limited to ZnO, SiOx, TiO2, and DLC are summarized in Table 2.
Among these methods, pulsed laser deposition (PLD), sol-gel (SG), thermal evaporation (TE), solution based on chemical approaches, sputtering, and plasma enhanced chemical vapor deposition (PECVD) will be briefly described hereinafter. They are easy to use, low cost, and yield high throughput of micro- and nanostructures.
|Material||Structure type||Deposition technique||Cited reference|
|TFs||Metal-organic vapor deposition|||
|TFs||Magnetron sputtering, MS|||
|TFs||Atomic layer deposition|||
|TFs||Ion beam evaporation, IBE|||
|NPs||Impregnating TiO2 particles in a resin and this composite deposited into the yarn-array|||
|TFs||Combined PECVD and MS|||
|TFs||Filtered pulsed arc discharge||[85, 86]|
|TFs||Ion beam deposition|||
In the field of TFs growth, PLD has proven to be among the most versatile methods , with features superior to conventional deposition techniques (fast processing, reliability and low production cost). In this technique, high power laser energies are used. They are focused onto a target in order to evaporate its surface under vacuum or different gas ambient atmospheres. The vaporized material consisting of ions, atoms, or molecules is subsequently deposited onto a generally parallel substrate. Repeated laser pulses will result in the deposition of the TFs in form of a coating on the substrate.
One important advantage of PLD method is the stoichiometric transfer of different materials from the targets in the deposited films [59, 90, 91]. This represents a direct consequence of the high ablation rate that allows all elements to evaporate simultaneously . This technique ensures an excellent adherence of the deposited structure to substrates, the high accuracy control of the growth rate (10–2–10–1 Å/pulse), the absence of contamination, the simplified growth of materials and combinations of materials of technological interest , and the good control of the final crystalline state of the coatings [59, 94].
The SG process is a synthesis route consisting in the preparation of a sol and successive gelation and solvent removal. This technique represents one of the simplest approaches to produce TFs. It presents many advantages in comparison with traditional deposition techniques, such as low working temperature, possibility to cover large surfaces, and high purity of the working conditions.
Compared to the physical route where harsh conditions such as high temperature or special equipment are usually required and consequently generating high costs, the solution based on chemical approaches [95–97] presents several advantages, including the simplicity in operation, low fabrication costs, low process temperatures (below 90 °C), and ambient pressure processing.
Thermal vacuum deposition or TE method is used to fabricate TFs under a high vacuum environment. In this method, an electron beam (e-beam) or resistive heating is usually used to evaporate the desired material inside the vacuum chamber, which then adheres to a substrate positioned above it.
The uniformity, high quality, and adherence of the deposited materials on large areas; the high deposition rate; and the versatility of sputtering techniques have made them attractive for the production of TFs [98–101]. In plasma sputtering deposition, plasma is used as the source of ions. These ions bombard a solid material, commonly known as the cathode or the target, with a typical kinetic energy of several hundreds electron volts. The ion bombardment produces the emission and acceleration of the secondary electrons, which play an important role in maintaining the plasma around the cathode . The ionizing energetic electrons are confined close to the cathode allowing operation at high plasma densities and low pressures.
3. Synthesis of hydrophobic/hydrophilic oxides and DLC nanostructures onto textiles and metallic medical substrates
A brief literature review on results regarding oxide (ZnO, TiO2, SiOx) and DLC nanostructures synthesized by PLD, SG, TE, solution based on chemical approaches, sputtering, and PECVD onto textile or metallic substrates will be presented hereinafter. Polyester, cotton/polyester, and poly(lactic acid) woven fabrics can be coated with multifunctional oxide materials in form of TFs or NPs to get properties that increase the protection and comfort of the wearer. When covering the surface with NPs, a new roughness is added leading thus to a dual-size surface roughness. Therefore, the study of wettability properties is a tool to test the surface functionalization . It is well known that wetting of a surface by a liquid is affected by surface roughness . In the case of textile materials, the roughness is related to the geometry which is very complex . Due to the fiber topography, the construction of the yarn, and the construction of the fabric, polymer, natural, and synthetic fibers might be made from porous materials that can absorb water from the environment. Fabrics have thus pronounced texture, porosity, and also (oriented) in-plane capillarity along the threads . CAs on textile substrates can be useful quantities for comparative measurements in order to characterize the effects of surface modification, especially if the textile is distinctly hydrophobic . Titanium (Ti) stands for the most used metallic material for medical applications due to its unique properties such as biocompatibility, excellent mechanical properties in bulk, relative to the low mass density, and high corrosion and ductility resistance .
ZnO is an n-type metal oxide semiconductor having a wide band gap, high electron mobility, and thermal conductivity. It mainly crystallizes in the wurtzite phase, being intrinsically polar, and thus exhibiting interesting piezoelectric properties. In addition, in the form of TFs or NPs, ZnO possesses promising antibacterial and antifungal, photocatalytic, electrical, electronic, and optical properties [107–115]. Recently, combinations ZnO-organic were tested for various applications requiring antimicrobial properties [116, 117]. Also, ZnO has probably the richest family of structures’ morphology including rods, prisms, wires, whiskers, or tubes [95–97, 118–123]. Moreover, morphology influences other properties such as wettability, another significant characteristic of ZnO covered surfaces bringing great advantages in a wide variety of applications in industry and daily life [124–127]. For example, wettability is critical in cosmetics and textile fields where ZnO can be used due to its biocompatibility property.
Hydrophobins are a class of small-size cysteine-rich proteins synthesized by filamentous fungi . They form ~5–10 nm thick self-assembled monolayers  on different substrates, changing their surface wetting properties. Namely, hydrophobic surfaces can be turned to hydrophilic, while hydrophilic materials become hydrophobic  after immersion in an aqueous solution of hydrophobin. Textile materials can be finished with various functionalization agents, such as chitosan microcomposites  or nanocomposites [132, 133], medicinal herbs , nisin , polyhexamethylene biguanide , or PMMA nanocomposites , in order to obtain new surface properties like antimicrobial, hydrophobicity, resistance to laundering, or protection against decoloration. Due to exceptional surface properties and to the tuning opportunities, their use is envisaged in cosmetic industry, polymer emulsion synthesis, and biosensing .
3.1.1. ZnO nanostructures synthesized by PLD onto cotton/polyester textiles
Yang et al.  and Papadopoulou et al.  demonstrated that the structures synthesized by PLD can be controlled in terms of wetting behavior. Therefore, ZnO structures showed a hydrophilic behavior after exposure to UV and were converted to hydrophobic after thermal treatment or storage in complete darkness. In this respect, a one-step PLD procedure to obtain either hydrophobic or hydrophilic ZnO structures (TFs or NPs), without any complementary post-deposition treatments of the surface, was recently proposed . Depending on the number of applied laser pulses, well-separated NPs (for 10 pulses) or compact TFs (for 100 pulses) were synthesized. By varying the ambient gas nature and pressure inside reaction chamber, hydrophilic or hydrophobic surfaces were obtained. The expected properties of the textiles coated with ZnO were evaluated at room temperature (RT) by static CA measurements.
The TFs deposited on textiles (Figure 3) in a flux of 13 Pa oxygen were highly transparent and had a hydrophilic behavior (Figure 3a), while those obtained in vacuum were opaque and showed a hydrophobic behavior (Figure 3b).
A CA of 157° (Figure 4) was measured, which qualified these films as superhydrophobic.
In the case of NP samples, eye examination confirmed a hydrophilic behavior for the structures deposited in the oxygen flux and a hydrophobic one after deposition in vacuum (Figure 3c).
The macroscopic and microscopic observations have revealed a smoother surface in case of TFs deposited in vacuum characterized by a six times smaller RMS and negative values for surface skewness (Ssk) and kurtosis (Sku) (Table 3).
|Amplitude parameters||Sample type/scanning area|
|TFs oxygen/10×10 μm2||TFs oxygen/2×2 μm2||TFs vacuum/10×10 μm2||TFs vacuum/2×2 μm2|
Figure 5 shows two-dimensional AFM images of the TFs deposited in 13 Pa oxygen flux and vacuum. The grains (of ~140 nm) visualized by AFM (Figure 5b, d) were in fact consisting of very small crystallites (of ≤10 nm), as proved by the XRD patterns.
In order to account for the significant difference observed in the wetting behavior of the TFs and NPs deposited in a flux of oxygen and in vacuum, a model was proposed for surface wetting. The numerous gaps between crystallites are filled with air acting as a support “buffer” for the water droplet, in contact to the surface in a few small nanometric sites only. Conversely, the TFs deposited in an oxygen flux (Figure 5a, c) consist of larger crystallites and a few intergranular pores only. Thus, the air “buffer” is rarefied, so the contact between the water droplet and the ZnO surface is extending over a larger area (Figure 6). The droplet weight prevails over the counter pressure exerted by the ZnO surface and eventually collapses under its own weight. Figure 6 shows schematically the water droplet in contact with ZnO structures synthesized in vacuum (Figure 6a) and oxygen flux (Figure 6b).
The NP depositions in vacuum consist of a large number of small crystallites, which include a huge amount of vapor pockets. Their action cumulates with the effect of the air, which is present in the space between NPs to more efficiently support the droplet weight. This model is in accordance with other studies on hydrophobic plant leaf surfaces . Accordingly, the largest contact area between the water droplet and the leaf surface corresponds to flat and microstructured surfaces but is generated in case of nanostructures as an effect of vapor pockets entrapment.
The electric charging of the surface should be considered when explaining the affinity or repellency to water of ZnO structures. XRD investigations demonstrated that the ambience in the interaction chamber also showed the combinations between Zn and O atoms in the crystalline lattice . In case of structures deposited in vacuum, there is a mix in each crystalline plane of positive and negative charges. One should note that the water droplet is neutral from the electrical point of view. Accordingly, the deposited structures do not interact electrically with the water droplet. Oppositely, the structures deposited in an oxygen flux have only one type of atoms per plane that induce a positive (Zn) or negative (O) charging of surface . The synthesized structures interact electrically with the droplet to reach the neutral status, thus attracting the water bubble toward the ZnO surface causing supplementary stress that contributes to the collapse of the bubble.
The intercalation of a hydrophobin nanolayer between substrate and ZnO film, which can boost the oxide efficiency against microorganisms with a higher natural resistance, was recently studied and an explanation of the observed phenomena was proposed . In case of ZnO TFs deposited on bare textiles, the adhesion is governed by physical mechanisms only (as e.g., mechanical or dispersion adhesion ), while in case of a buffer layer of hydrophobin interposed between textile and ZnO, chemical bonding occurs, the fastening between the ZnO and the textile substrate becoming much stronger. When used alone, the hydrophobin had no effect on both Candida albicans colonies and six strains of filamentous fungi. In case of simple finishing with ZnO, the reduction rate was of 50% and 70% of the colonies in 24 h (Figure 7a, b).
In order to improve ZnO efficiency against resistant fungi, the oxygen concentration on films’ surface was increased by covering the textile fibers with hydrophobin and then adding an upper layer of ZnO. As an effect, the orientation and shape of ZnO crystallites were changed, the (001) film texturing becoming more pronounced and nanocrystallites elongated, with more polar planes (001) parallel to the surface (Figure 8a). Depending on the orientation of the c-axis, these planes may contain oxygen atoms only (Figure 8b). The ZnO film deposited on hydrophobin proved in this case 100% efficient in reducing colonies of both C. albicans and a mold mix of filamentous fungi (Figure 7a, b). This significant enhancement was attributed to the higher texturing of the oxide film when growing on hydrophobin interlayer, resulting in an increased presence of oxygen species on surface.
ZnO is recognized to possess antibacterial and antifungal properties. Nair et al.  assessed the microbiological activity of ZnO against a mold mix of microbes and associated the high reduction ratio to the generation of surface oxygen species. Sawai et al.  and Premanathan et al.  suggested that these oxide species form in wet media hydroxyl radicals and hydrogen peroxide. As known, the hydroxyl radical is the most reactive one, able to interact with almost every type of molecule of the living cells of bacteria and fungi, causing irreversible damage to cellular components and eventual apoptosis. Applerot et al.  advanced a mechanism for the reactive oxygen species formation on ZnO surface. The oxygen atoms present on surface interact with water molecules, forming OH‒ radicals. A chain reaction occurs, resulting in exponential multiplication of these radicals on surface.
We note that no negative side effects of hydrophobins when in contact with human tissue were reported , and to the benefit of biomedical applications, they were able to form, in specific cases, resistant monolayers with antimicrobial activity . Moreover, the proposed antimicrobial finishing procedure of fabrics with a conjunction of a thin layer of hydrophobin and a ZnO layer can find applications in the medical field, where solutions are constantly required for elimination of microbial contamination, thus reducing the risks of infections during surgery.
3.1.2. ZnO nanostructures synthesized by solution based on chemical approaches onto solid (glass) substrates
In the synthesis process of ZnO nanostructures using solution based on chemical approaches, a zinc salt and a basic compound are brought together. The involved chemical reactions can be described as follows:
Using a weak base ((CH2)6N4)
Using a strong base (NaOH)
Using a reducing agent ((CH3)2NHBH3) 
The ZnO structures were examined by XRD (Figure 9). The diffraction peaks observed at 2θ = (31.8°, 34.5°, 36.3°, 47.5°, 56.6°, 63.0°, 66.4°, 68.0°, and 69.1°) are characteristic to ZnO hexagonal wurtzite phase (JCPDS file no. 36-1451), with corresponding Miller indexes at (100), (002), (101), (102), (110), (103), (200), (112), and (201). The strong and sharp diffraction patterns suggest that the as-obtained structures are well crystallized.
SEM images of the samples (Figure 10) revealed the following morphologies for the ZnO micro/nanostructured TFs: rods (4.5 µm in length and 330 nm in diameter; Figure 10a, b), flowers (1–2 µm in dimension; Figure 10c, d), and hexagonal prisms (400 nm in length and 200 nm in diameter; Figure 10e, f). Insets to Figure 10 show the influence of the ZnO surface morphology on wetting behavior. The corresponding CA values of the ZnO samples are 164.8° (rods), 94.3° (flowers), and 79.4° (prisms). An explanation for the different values of CA can be related to the numerous gaps between the ZnO structures filled with air. For this reason, the film containing a higher volume of air trapped between the ZnO structures at the solid/water interface has a superhydrophobic behavior. The CA results were confirmed by AFM measurements (Figure 11). The RMS values were as follows: 390 nm (rods), 120 nm (flowers), and 50 nm (prisms).
Due to their morphology, the ZnO structures present different degrees of compactness, trapping more or less air in-between. In this way, the CA value is linked to the RMS value of the sample.
Titanium dioxide (TiO2) is a transition metal oxide with UV absorbing properties with many technological applications [148, 149]. High photocatalytic efficiency, great stability, and low cost of production are in favor of TiO2’s photocatalytic properties . In addition to bulk applications, TiO2 TFs were obtained for UV blocking, antibacterial or/and photocatalytic properties .
3.2.1. TiO2 structures synthesized by SG and sputtering onto textile substrates
Some properties of the substrates used in the experiments are summarized in Table 4.
|Sample code||Textile 2D element/thread||Nature of fibers||Color||Thickness (mm)||Density (g/cm3)|
XRD and SEM investigations  indicated, for both deposition techniques, that TiO2 NPs were amorphous. Sputtered layers consisted of aggregates randomly distributed on substrate, while the SG layers showed a uniform distribution of NPs, with a mosaic-like structure. SEM images (Figure 12) suggest the formation of NPs, which are not singularly distinguishable. The sputtered layers consist of NP aggregates (in coalescence) with less than 20 nm diameter, randomly scattered on substrate. In the case of the SG layer, there are bridge-aggregated NPs leading both to a mosaic-like structure and to cracks and interfiber bonds .
A highly polar liquid–water was recommended  as testing liquid in CA measurements, for estimating the wettability of polar solids as polyester materials. The water repellency was thus regarded as indicating the performances of the coated layers and was evaluated by measuring static (equilibrium) CAs at RT . In order to have a general idea of the samples’ wetting behavior, different measurement points on each sample were thus considered (Figure 13).
The mean CA values of the raw or coated samples are summarized in Table 5.
|Sample code||CA (°)||f=1+cosθc1+cosθ0 *|
From Table 5, one can see that the CAs increase more by fabric modification (after Titania deposition). CAs were influenced by air, water droplet, and surface of fabric, which formed a nanorough substrate. One sample (P28) is hydrophilic meaning that water passes through it; this sample has voids large enough, and the margins become hydrophilic by deposition of hydrophilic particles. The behavior might be approximated by the Cassie–Baxter equation:
Here, θC is the composite CA formed on the treated fabric and θ0 is the CA formed on untreated fabric . The parameter f represents the fraction of the surface in contact with the water droplet. Knowing the corresponding CAs, its values can be calculated for each raw-treated pair of samples, using the following equation:
These values are summarized in Table 5. However, it seems that Cassie–Baxter equation (or Wenzel equation) should be applied to superhydrophobic surfaces with caution . TiO2 can be used to obtain hydrophobic surfaces by producing artificial roughness via micro structuring .
Silicon oxide was deposited onto polymeric substrates as a viable alternative to metallic depositions used for packing materials due to their transparency, recyclability, microwave use, and impressive barrier properties , to produce textiles with hydrophobic properties . In addition, silica NPs immobilized on textiles can lead to flame retardant properties . SG-deposited layer can be compared to the one deposited in vacuum at low angle because in both cases the deposit is awaited (super) hydrophilization evidence since SiO2 brings its OH groups which print to the media hydrophilic properties. However, due to the columnary nanostructured relief of deposited layer, it was expected that the roughness of the textile surface would be increased. In addition, information was acquired on vacuum deposition at small angle [159–162]. Thus, we preferred this technique for a SiOx deposition onto textile materials .
3.3.1. SiOx structures synthesized by thermal evaporation at small angles onto polyester (P), polyamide (PA), poly(lactic acid) (PLA), and natural cellulosic hemp (H) substrates
|Sample code||Textile 2D element/thread||Nature of the fibers||Color||CA (°)|
|P4||Fabric/Nm 70/2 + Nm 40/2||136.9||139.2|
The XRD diffractograms pointed to an amorphous phase of the SiOx deposited layers .
SEM morphologies of SiOx particles synthesized on fabrics are presented in Figure 14. The raw material images showed defects like kink bands, dislocations, nodes, and slip planes, which are common characteristics of hemp materials . SEM images of synthesized samples showed that SiOx particles were grown on the fiber surfaces in a continuous and noncolumnar layer (Figure 14). Apparently, each individual fiber of samples looks uniformly covered by an amorphous layer .
The wettability properties were evaluated by measuring static (equilibrium) CAs. The measurements were carried out at RT . The images were processed using specific programs to fit the profile with the Young–Laplace equation in order to obtain the value of static CA.
In Figure 15, the image of the water droplet onto the deposited P2 sample and the corresponding CA is represented. The measurements were performed for a direction parallel to the privileged one of the knitted matter (vertically advance geometry). When following a direction perpendicular to the privileged one, the measurements evidence differences of few degrees only.
From Figure 15, one can observe that the investigated surfaces are not flat, smooth, or homogeneous. These characteristics of the samples make it difficult to apply a specific model algorithm. Moreover, wetting of fabric surfaces is complicated by the heterogeneity, the diffusion of liquid into the fiber, and the capillary action of the fiber assembly. Under these conditions, the experimentally measured CA is an apparent one and can differ considerably from the actual value .
The mean CA values of the raw and deposited samples are summarized in Table 6. One can observe that the CA generally decreases after surface functionalization with SiOx, in a range of few up to several tens of degrees. As expected, the presence of SiOx NPs onto the fiber surface confers to the textiles a hydrophilic behavior (see Table 6).
3.3.2. Surface free energy of SiO2 (quartz) inferred from CA measurements
Starting from known values of the dispersive and polar parts of the probe liquids’ surface tension and obtained values of the CAs, the dispersive and polar parts of the surface tension of the solid (fused quartz) were estimated either by minimization of the equation system using the least square method or by solving the equations taken for combinations of two probe liquids .
Eq. (9) is a relation between the dispersive and polar parts of the solid substrate’s surface tension and the same quantities of the surface tension of the wetting liquid and the corresponding CA:
The values and were obtained by averaging the dispersive and polar components of resulted from solving Eq. (9) for all pairs of liquids that have the condition number of system matrix low enough (as defined in ).
CAs of water on fused silica can vary in a large interval. This behavior is in agreement with the one described in the literature, for example, with a 20º to 80º range obtained on quartz dehydroxylated by heating, slightly contaminated, or deliberately methylated . The values we found can be interpreted in terms of the dependence of water CAs on sample purity; the presence of amorphous materials, chemicals, heating, and other pretreatments; and contamination by adsorption of substances from laboratory ambient. All these factors could have an influence over the increasing values of the CA. The obtained values were supposed to depend on the amounts of silanol groups and physically adsorbed water molecules on the quartz/silica surface. The investigation of “cleaner” surfaces obtained by a thermal treatment removing the hydroxyl groups at temperatures of the beginning and ending of the dehydroxylation process  was carried out.
The components of surface free energy of fused silica were determined by CA measurements of several liquids (see Table 7).
|Treatment temperature (°C)||CA for different liquids (°)|
|Water||Glycerol||NP5||Ethylene glycol||Dimethyl sulfoxide|
The fused silica plate samples were heated in atmosphere in order to remove water adsorbed on surface and most of the silanol groups. Measurements of CA on solid substrate were performed by analysis of the profile images of symmetric static liquid drops using the Drop Shape Analysis System (model DSA 100, from Krüss) [141, 152]. The samples were placed on a stage, under the tip of liquid-dispensing disposable blunt-end stainless steel needle with an outer diameter of 0.5 mm. The fixed needle was attached to a syringe pump, which was controlled by the computer for drop delivery. The volume of the drops was of ~ 2–3 µl. The CAs were determined by fitting the shape of the sessile drop with a smooth curve and then calculating the slope of the tangent to the drop at the liquid–solid–vapor interface. Low CAs (θ < 30º) were determined by fitting the shape of the sessile drop with a circle, whereas larger CAs were estimated by fitting the drop shape with a polynomial equation of second degree or a circle equation. The camera was positioned to observe the droplet under an angle of about 2°–3° in respect to the sample surface supporting the droplet. The tests were carried out at RT. CAs were obtained with an uncertainty of ±2° due to combined effects of drop asymmetry, surface heterogeneity, and variation in drop position on the plate.
Representative images of the observed water droplets on plates are given in Figure 16.
Based on literature values [169–171] of the polar and dispersion parts of the liquid surface tension (see Table 8) and using the methods of geometric or harmonic mean for the interaction term, the calculation of the two components for fused silica (see Table 9) gave some differences, but their sum did not differ much. Moreover, our tests seem to indicate the method of harmonic mean as better than that one of Owens–Wendt .
|Sample liquid||Dipole moment (D)||γL (mN/m)||γLd (mN/m)||γLp (mN/m)|
|Treatment temperature (°C)||Owens-Wendt/least squares (mN/m)||Owens-Wendt/|
least squares (mN/m)
The polar part of the surface free energy of fused silica thermally treated is higher than the dispersive part as resulting from both methods (geometrical and harmonic mean). This might be an indication that, at the measurement moment, the plate surface was not (totally) covered by water vapors from environment.
In agreement with the decreasing number of silanol groups by the thermal treatment, the polar part of the surface tension shows a decreasing trend when increasing the pretreatment temperature.
The indirect method of CA measurements applied for the set of liquids chosen to have complementary interactions with quartz surface, allowed for obtaining values for the components of the surface free energy.
3.4. Effects of proteins from blood plasma on the hydrophobicity of DLC films
The amorphous phase of sp3 bonded C atoms is known as DLC [82, 172, 173]. Beside high wear resistance coatings for metallic parts, DLC also proved useful in coating implants due to specific surface properties (low surface energy values and chemical inertness) that prevent blood coagulation and favor osteoblasts adhesion [90, 172]. In the biomedical field, the main necessity for DLC coatings comes from vascular prostheses. In the case of interaction with blood, it seems that DLC quality has a major influence upon clotting time. During the blood flow through these tubes, the erythrocytes and thrombocytes (platelets) aggregate in certain spots and may eventually block the blood passage. To compensate for this general weakness of vascular prostheses, DLC films can bind albumin molecules from the sanguine plasma forming a passive layer that makes the surfaces less adhesive for blood platelets .
The blood compatibility with carbon-based films is extremely complex and for the moment there is no relation found between hemocompatibility and surface properties such as surface energy, atomic bond structure of carbon, or composition of material. Contradictory data have been reported regarding the behavior of the material in terms of blood clotting, the adherence of platelets, or protein adsorption to surfaces. The relationship between the sp3 bonds content of DLC and its antithrombogenicity properties is still not well understood. In vitro  and in vivo [88, 175] studies indicate that better results can be obtained for a higher sp3 content.
Kwok et al.  pointed out that a higher surface energy of phosphorous doped a-C:H films is associated with a low adsorption of proteins, among them the albumin being the preferential one. Similar findings in terms of protein adsorption were presented by Ma et al. , who reported a higher albumin to fibrinogen adsorption ratios on surfaces with higher surface energy.
Jones et al.  explored platelet attachment on Ti, TiN, TiC, and DLC surfaces and reported that the more hydrophilic surfaces present a greater platelet spreading and fibrinogen adsorption. They suggested that the better hemocompatibility of DLC surface is linked to its low surface energy and thus high hydrophobicity. Okpalugo et al.  also noted that improved blood compatibility can be obtained when surface energy is lowered in silicon doped a-C:H films.
Recently, the correlation between activated partial thromboplastin time (aPTT) and surface energy of DLC structures with different sp3/sp2 bonds ratio was studied. Attention was paid to the investigation of protein adsorption and platelets adherence to the surface, both acting as crucial factors for material hemocompatibility .
3.4.1. Types of bonds in the films
XPS analysis, indicating the C 1s core level variation, was used in order to assess the amount of sp2 and sp3 bonded C in three types of samples (D20, D60, and D100; see Table 10).
From the XPS analysis, the amount of sp3-bonded C and sp2-C, as the ratio between the integral intensities of each component, could be extracted. The XPS spectra exhibited a very complex shape pointing to the existence of different chemical states for C 1s (Figure 17).
Three components were needed in order to assure a good fit, associated with the sp3-C (286 eV) and sp2-C (284.3 eV) contributions, as well as to C–O, C=O, and/or O–C=O bonds (287.5–289.9 eV) owing most probably to the contamination of the sample surface [180–182]. The deconvolution studies of the C 1s spectra generally reveal two main distinct peaks assignable to sp2- and sp3-C hybridization . The peak placed at a higher binding energy (BE) is assigned to sp3-bonded carbon (C–C and C–H), and that at lower BE corresponds to the sp2 hybridization state of carbon. From the analysis of the main components of C 1s core level spectra, one could assume that the amount of sp2 bonded C decreases from 36% in D20 sample to about 10% in D100. When the methane dilution is increased (D60 and D100), the sp3-C concentration strongly increases (to ~78 and 87%, respectively).
The increase in the sp3 content with the augmentation of the methane concentration has been confirmed both by Raman and XPS. A significant sp3 content augmentation from sample D20 to D100 was measured. This could be the effect of the initial sp3 hybridization of carbon in the methane molecule. Bugaev et al.  also reported that high-quality DLC films can be obtained from pure methane, their results pointing that most probably methyl mechanism is favoring diamond-like bonds formation. It is known that CH3 are the most abundant species in pure methane discharges, while carbon dimer C2 is the most abundant in methane highly diluted in argon discharges [184, 185].
3.4.2. Surface energy
Using deionized water and formamide as standard solvents, solid surface energy calculations based on CA measurements were performed. The measurements of the prepared DLC structures were carried out using the goniometric method, the two solvents being dropped onto the surface and the CA estimated. The drop size and the drip distance were kept constant in all cases. The CA values were determined by the evaluation of the tangent angle of a sessile liquid drop on the DLC solid surface. The surface energy was calculated using the Owens–Wendt approximation [186, 187].
The surface energy values recorded for DLC/Ti structures were lower than those of the bare medical grade Ti and PMMA control substrates (see Table 11).
|Sample||Deposition atmosphere composition||Surface energy (mJ/m2)|
|Bare Ti||N/A||37.85 ± 0.94|
|PMMA||N/A||36.35 ± 0.78|
|D20||20% CH4 + 80% Ar||32.09 ± 0.73|
|D60||60% CH4 + 40% Ar||30.50 ± 0.70|
|D100||100% Ar||28.71 ± 0.34|
One notices a decrease of the surface energy with the increase of methane dilution in the reactor chamber (Table 11). The two tailed t-testing showed statistically significant differences (p < 0.05) between the surface energy values recorded for all samples. An important decrease (with ~25%) of the surface energy was obtained when applying the DLC coating: from 37.85 ± 0.94 mJ/m2 for the bare Ti substrate down to 28.7 ± 0.34 mJ/m2 for the D100 structure.
3.4.3. DLC films interaction with blood
Platelets were obtained by centrifugation of whole blood and their adherence to the DLC films surface was investigated by Western blot method. The detailed procedures for platelets isolation and for the Western blot technique are described in Ref. .
The obtained signal is proportional to the amount of beta-actin, a structural protein present in all cells and, therefore, to the number of platelets adhered on the sample surface at the moment of lysis. As visible from Figure 18, there was almost the same number of platelets present on the surface of bare titanium and D20 samples.
The number of platelets adhered on D60 and D100 was significantly lower. The DLC coatings ensure conditions for a weaker platelet–surface interaction, which in vivo can conduct to a lower platelet activation and subsequently a prolonged time of coagulation. One can assert that this effect derives from the fact that all cells have a negatively charged cellular membrane, which tends to interact/adhere to positively charged surfaces (hydrophilic surfaces) rather than to hydrophobic ones.
The polyvinylidene fluoride membranes were also probed with aprotinin (a protease inhibitor with proteic structure and a mass of ~6 kDa), which was present in the same concentration in all samples, since it was added to the lysis buffer formulation. This is an internal quality control which ensures that all steps of the technique are properly done.
The results of protein adsorption on the DLC surfaces are shown in Figure 19.
Figure 19a shows that serum albumin was adsorbed in greater quantities on all DLC surfaces than on the bare titanium surface [albumin molecular weight (MW) ~66,483 Da]. Other proteins (G immunoglobulins) presented a roughly similar pattern (G immunoglobulins MW ~134,350 Da). Another important peak is that of 28,900 Da, which can be assigned to the factor XIIa light chain and is more prominent on the titanium sample (Figure 19b). The factor XII, the activator of surface contact coagulation cascade, could not be identified because it had a mass similar to that of albumin (factor XII MW ~67,792 Da).
Albumin is a protein that has hydrophobic moieties, being a blood carrier for many hydrophobic molecules. Since our DLC surfaces tend to be more hydrophobic, it is expected to find more adsorbed albumin than on titanium as confirmed by mass spectroscopy spectra. The vast majority of proteins in blood are glycosylated, which makes them more hydrophilic and more susceptible to polar interactions. The quantity of albumin adsorbed on the surface shields the surface of the sample, making it difficult for the different proteins and coagulation factors to reach the sample and activate the coagulation cascade (Figure 19b). These findings are in line with Liu et al. , who showed that the albumin adsorption on DLC inactivates the surface for blood clotting. One can state that the coagulation time for each material is in line with surface energy data, with the platelet–surface adherence properties and protein adsorption profiles, and so advocates for a cause–effect relationship between these factors.
Wettability of solid substrates represents an important phenomenon for many natural systems and can play a key role in a wide range of applications such as coatings, tunable surfaces, design of hydrophobic/superhydrophobic, or hydrophilic surfaces. It is well known that the wettability of a solid surface is governed by both surface structure and chemistry. After a brief introduction on wettability of nanostructures and the possibility to investigate it by contact angle (CA) measurements, this chapter focused on hydrophobic and hydrophilic structures (oxide and DLC TFs or NPs) synthesized by various deposition techniques (PLD, SG, TE, solution based on chemical approaches, sputtering, and PECVD).
The possibility of tuning the wetting behavior of textile materials by their functionalization with oxide TFs or NPs was reviewed. Depending on the deposition ambience, the TFs can change their behavior from hydrophilic when obtained in an oxygen flux to superhydrophobic when deposited in vacuum. The hydrophobicity was found consistent with the organization of the deposits in vacuum consisting of nanometric crystallites. The subsequent treatment with a TF of a fusion hydrophobin, deposited by soaking in solution, and a ZnO TF finishing in vacuum boosted the antifungal efficiency of the structure by 100%. This significant enhancement was attributed to the higher texturing of the oxide film when growing on hydrophobin interlayer, resulting in an increased presence of oxygen species on surface. In complementary studies, fabrics functionalized with oxide layers showed improved UV protective performances. These results might offer guidance for laser manufacturing in one technological step of stable superhydrophobic and antifungal textile surfaces, used for everyday garments and medical clothing.
ZnO structures can present different degrees of compactness, and as a consequence, they can trap more or less air. This result can be explained by the Cassie–Baxter model. Due to the morphology of the deposited ZnO structure, which is made of a large number of small prisms, the roughness presents high values. The apparent CA is therefore enhanced as compared to the one measured on a similar smooth surface. When the space between the ZnO structures is large enough, the water droplet can penetrate, and an explanation of the phenomenon can be based on the Wenzel model. There exists also the possibility to obtain a transition between these two regimes, and the apparent CA could be different than the one inferred for a smooth surface.
CA measurements confirmed that the presence of SiOx particles on fiber surfaces can change the wetting behavior of the structure. Since it brings OH groups to the surface, the deposition of SiOx is therefore expected to provide hydrophilic properties to the textiles.
Although bulk polyester is hydrophobic, water droplets can be sucked into the fibers due to high porosity (void areas) of the material. The void areas were drastically reduced by the addition of TiO2 particles. They decrease the voids and concomitantly increase the sample hydrophobicity. Under these complex conditions, one cannot use the traditional equations like Cassie–Baxter or Wenzel to model the wettability behavior of the heterogeneous and rough samples.
Protein adsorption using fresh blood plasma from healthy patients was also studied. In the case of DLC films with the highest sp3 content, albumin was preferentially adsorbed (due to the affinity between the surface and the hydrophobic moieties of the protein), thus shielding the surface and preventing the immobilization of coagulation factors.
The results reviewed in this chapter are devoted to improve the understanding of the wettability of nanostructured surfaces. Understanding the importance of surface wettability and succeeding to control this phenomenon at nanometric scale will hopefully facilitate the fabrication of devices with improved characteristics for top applications, especially in nanotechnology.
LD and INM acknowledge project no. 7-083/2014 (CARLA). IZ acknowledges the financial support of the Romanian Ministry of Education and Research under the Project IDEI 281/2011. ACP acknowledges the funding of this research by the Romanian National Authority for scientific research through PNII-RU-TE-2012-3-0379 (TE 16/2013).