TiO₂ Nanocoatings on Natural Fibers by DC Reactive Magnetron Sputtering

Abstract

The surface functionalization of natural fibers, mainly using TiO2 films, shows a growing interest in its application as yarns in fabrics that require advanced properties, allowing the use of their excellent physical and chemical properties in the textile area. The DC magnetron sputtering technique is a potential method for depositing TiO2 films onto natural fibers, allowing for the creation of advanced and competitive properties compared to synthetic fibers. Different crystalline phases of TiO2 have been shown to be effective in photocatalytic applications. Reactive discharges like the Ar/O2 gas mixture can be used to deposit TiO2 films with desired characteristics, and controlling deposition parameters can further manipulate the properties of the coatings. Analytical techniques such as XRD, XPS, and SEM/EDS can be used to study the surface properties of TiO2 films. XRD determines crystal structure, XPS provides information on chemical composition, and SEM/EDS examines morphology and elemental composition.

1. Introduction

In the last few years natural fibers of natural origin, especially those of lignocellulosic composition [1], attracted a great attention due its interesting properties as high tensile strength and rigidity, and to the advantages of being recyclable, lightweight, biocompatible, and extremely low costing, when compared to the synthetic ones. Among others, ramie, jute, kenaf, etc. reinforced composites have been highly emphasized [2]. The use of natural fibers fit into the concept of circular economy, which seeks to reduce, reuse, recover, and recycle products. In addition, there are economic and functional advantages in the use of natural fibers compared to the most common artificial fibers, made of carbon, glass, or polymeric resins [3], namely due to their low production cost and high abundance.

Sustainability is becoming a concern in the development of new materials, mainly due to problems related to the use of scarce resources and waste management. On the other hand, there is a kind of activation energy for the creation of new products and functionalities, enabling new commercial paradigms or complementing the existing ones.

Cellulose fibers can be obtained from many plants and represent one of the most abundant organic materials on earth. Invasive plants, such as ginger lily (Hedychium gardnerianum) [4], are abundant in various countries and are therefore an ecological source of fibers for scientific and industrial applications as an alternative to the traditional glass, polymer, or carbon fibers [5]. Invasive species are a threat to ecosystems and the survival of many endemic species, and their monitoring, control, or eradication is crucial, to prevent the modification of ecological processes and the loss of biodiversity. The ginger lily plant can be found in large quantities at Azores Islands (Portugal), and they are mainly considered as waste. However, its biological nature gives them specific characteristics less good for high-tech applications, namely the ease way since they absorb water. This causes dimensional changes and swelling on the fibers, mainly because its main composition of cellulose, which is structurally a linear polymeric chain with OH groups, and thus highly hydrophilic.

Nowadays, titanium dioxide (TiO₂) is one of the most effective photocatalysis [6] demonstrating high efficiency of decomposition and detoxification of several toxins and pollutants [7]. However, there is a huge disadvantage that involves the removal of TiO₂ catalysts after their applications, in the case of catalysts based on particles in suspension. In general, water purification reactors employ photocatalyst particles (powder type) that have higher photocatalytic activity due to less mass transfer limitations between the treated contaminants and the photocatalyst. However, powdered photocatalysts need to be filtered and separated after water treatment, which is a tedious and expensive process. So, to commercialize the process as a full-scale technology, it is critical to increase the photocatalytic activity of TiO₂ and manufacture devices with TiO₂ immobilized on a specific support. This strategy can bring a great benefit. Therefore, several attempts have been already used to immobilize TiO₂ on different supporting materials [8] and shapes. From these, glass substrates [9, 10], glass spheres [11], fiberglass [12], activated carbon, zeolite, and ceramics [13] stainless steel [14] and polyamide fibers [15], can be emphasized. Moreover, photocatalytic fiber is an emerging solution to immobilize catalyst powders [16, 17]. The natural fibers are nowadays preferable due to their multiple advantages in terms of environmental sustainability. Inspired by these remarkable characteristics, fibers have found a great interest as supporting substrates [18]. Cotton and ginger lily fibers, have booth cellulose in its main composition and so have abundant hydroxyl groups (OH) [18, 19] to link photocatalysts through hydrogen bonds and van der Waals forces. Wool fiber, instead, possesses plenty of disulfide bonds (-S-S-), carboxyl groups (-COOH), and amino groups (-NH₂) [16]. Moreover, natural fibers are considered desirable for TiO₂ immobilization platforms on account of their intrinsic porous structure [20], large specific surface area, and flexibility [16]. Its flexible form can be adapted to different spaces and purification devices. In addition, they can be cut to any size, rolled up, etc., to meet the function’s requirements. For example, cotton fibers were proved to be easily installed inside photoreactors [21].

Since TiO₂ can impart antibacterial [22] and self-cleaning [23] properties to the fibers it becomes clear that is of great interest for the textile industry. The multiple fiber-related substrates involving fibers, yarns, and fabrics with different structures can be used as support substrates for photocatalyst proposes [16]. Important fiber properties are good adhesion of TiO₂ which demands improvement of binding efficiency with fibers to keep the necessary high specific surface area to enhance the absorption affinity.

There are many methods used for the synthesis of TiO₂ [24]. TiO₂ nanoparticles/films immobilization was prepared successfully in a variety of natural fibers, by the sol-gel method [25], microwave-assisted liquid phase deposition process [26], and DC-reactive magnetron sputtering [18].

A promising method to prepare photocatalytic immobilized TiO₂-based thin films is by DC-reactive magnetron sputtering. This technique enables large-area deposition with high uniformity, yet it is essential to understand why the film properties exhibit after deposition. This is because the properties of the coatings obtained are highly dependent on the selected parameters, and so it is necessary to establish the ideal ones that satisfy any film application and understand the basic processes that control the properties of that films. These include, for example, the type of species deposited, their energy, and consequently the effects that their bombardment will have on the surface of a growing film, etc. In addition, the influence of substrate temperature on the nature of the film is also to be considered, among other factors, namely, for instance, the substrate position relative to the target, discharge pressure, and the gas mixture. In fact, bombardment can result in a series of surface effects, namely displacement of lattice atoms, creation of defects that can lead to increased atomic mobility, surface heating that can promote crystallization of nanoparticles, etc. These effects will consequently affect the internal stress, crystal size, morphology, and roughness of the deposited films. Quite often the physical structure of the thin film is directly responsible for the expected film property. For instance, in photocatalytic TiO₂ films, deposition of crystalline or amorphous TiO₂ is of crucial importance for their functionality [22]. So, understanding the relationship between deposition parameters that will affect film properties is therefore important for defining procedures.

2. Basics of DC reactive magnetron sputtering

2.1 Sputtering

Sputter deposition is a physical vapor deposition (PVD) process [27] based on the ejection of atoms from a solid target as a result of collisions with energetic particles. However, if the collisions are due to the impact of positive ions, the process in known as cathodic sputtering.

Sputtering allows the deposition of thin films of a variety of materials, including metals and certain compounds such as oxides and nitrides and any type of substrate can be applied for deposition. Simultaneous deposition from various sources permits to develop complex compositions.

It has advantages over other deposition methods when the intended film is a compound (e.g., oxide) or an alloy, avoiding non-stoichiometric films, separation of phases of the constituent elements, and even differences in the desired composition. Coatings deposited generally have good adhesion and exceptional coverage.

The objective is to remove material from a target and bring it to the substrate (e.g., the fiber to be coated). This is achieved by means of ion bombardment in the plasma, usually by Ar⁺ ions. Further, the ions that reach the target with enough energy can eject atoms from the target that are then dropped on the fiber (or other substrate) surface placed nearby to the target. The process basically consists of three distinct steps that occur altogether (Figure 1):

Figure 1.

Schematic diagram of magnetron sputtering. Adapted from [28].

1. Production of atoms to deposit from a target material source,

2. Transference of the atoms through the plasma to the substrate,

3. Deposition of atoms on the substrate and film growth.

The widespread use of sputtering is explained by the many advantages of this technique, mainly due to its simplicity of operate and the quality of the thin films through the stoichiometry control in complex compositions, excellent film adhesion to the substrate, uniform deposition over a large area and tailor of the film thickness. Moreover, by the change of deposition parameters such as oxygen partial pressure, working pressure, and sputtering power is possible to achieve desired film parameters, for example, microstructure, composition, step coverage, among others.

2.2 Chamber preparation and DC reactive plasma

The process begins by creating vacuum inside the chamber and thus the air is pumped out. The chamber is then filled with argon, an inert gas, reaching a pressure between 1 and 10 Pa. When a DC voltage is applied between the electrodes (with a gas in between them), a plasma is formed. The applied voltage is high enough to enable that a large quantity of inert gas atoms turns into ions; electrons acquire enough kinetic energy to ionize gas atoms (break gas atoms) and thus, the plasma is formed [29]. The ions and electrons are then accelerated towards opposite electrodes. Plasma is thus a partially ionized gas. Depending on the mean free path in the gas, the accelerated particles can collide with inert gas atoms and give rise to scattering occurrences at a rate that can change with the pressure and nature of the gas. Moreover, these scattering occurrences can lead to ionization of further gas atoms. The probability of ionization (α) occurs will depend basically on the threshold voltage to initiate the breakdown of the gas (trigger the gas discharge) which must surpass the ionization potential of the gas species and can be calculated by [30]:

$$\alpha =\frac{1}{\lambda }exp\left(-\frac{V_i}{\mathit{eE\lambda }}\right)$$E1

where λ is the mean free path of the sputtering ion, V_i is the ionization potential of the gas in electron volts, e is the electron charge, and E = V/d is the electric field between electrodes [30].

The breakdown voltage of a gas, which is the voltage required break a sustained plasma, is established by Paschen’s law, which is a function of the product of electrode gap spacing and chamber pressure, according to:

$$V_b=A\frac{\mathit{pd}}{\mathit{ln}\left(\mathit{pd}\right)+B}$$E2

where V_b is the breakdown voltage, d is the gap electrode distance (cm), p is the pressure (torr), and A and B constants depending on the gas mixture inside the chamber. Paschen’s law relationship the breakdown voltage versus the product of the pressure and the gap electrode distance (pd) as shown in Figure 2 [31] and predict a minimum breakdown voltage for any gas.

Figure 2.

The breakdown voltage versus gas pressure curve [31].

Once the V_b is achieved plasma becomes self-sustaining and plasma reaches a steady state, exhibit enough energy to be used in sputtering.

2.3 Principle of sputter deposition

Typical gases used in the sputtering process are from the group of noble gases because they tend not to react with the target material. Argon (Ar) gas is the most common one in this process. Positively charged argon ions from the plasma (Ar⁺ ions) are accelerated by an electrical potential difference toward the negatively biased target (cathode), where the target material, for example, Ti, is placed and hits it.

With the impact energy, atoms are ejected from the target and diffuse through the vacuum chamber until they are deposited on the substrate to form a thin film (Figure 1). This atom ejection is known as sputtering. From a physical point of view, the principles of sputtering are based on a simple momentum transfer model, which allows understanding how atoms are ejected from the surface of a material due to successive collisions. The collision of particles and the transfer of momentum are important aspects of the DC sputtering process. In a plasma, there are various types of particles, such as electrons, ions, and neutral atoms or molecules. When these particles collide with each other or with the target material, momentum is transferred between them.

Because of the bombardment of the target, beyond ejected or sputtered atoms, additional events can occur as shown in Figure 3, including the followed briefly underlined: secondary electrons, reflected ions at the target surface, ion implantation in the structural atomic network, lattice defects, and structural rearrangement by trapping ion species.

Figure 3.

Events that may occur on the target surface being bombarded with energetic ions. Adapted from [27].

The mass of the energetic ions is key to the energy and momentum transferred to the film atom during the collision. From the physics laws of the conservation (of energy and momentum), the energy transferred in a collision of an incident particle (i) and a target particle (t) is given by:

$$\frac{E_t}{E_i}=\frac{4m_i{m}_t}{{\left(m_i+m_t\right)}^2}{cos}^2\theta$$E3

where E and m, are, respectively, the energy and the mass. θ is the angle of incidence as measured from a line across the two centers of masses, as shown in Figure 4. When the ejected particles reach the substrate, they deposit onto its surface due to the momentum transfer that occurs during the collision. The amount of momentum transferred during the collision depends on the mass and velocity of the particles involved. In general, heavier particles transfer more momentum than lighter particles, and faster particles transfer more momentum than slower particles.

Figure 4.

Collision of particles and the transfer of momentum. Adapted from [32].

The transfer of momentum is an important factor in determining the quality and properties of the deposited thin film. If the momentum transfer is too low, the deposited film may be porous and have a low density. On the other hand, if the momentum transfer is too high, the deposited film may be dense but have high levels of residual stress. Therefore, it is important to carefully control the parameters of the DC sputtering process, such as the gas pressure, target material, and substrate temperature, to optimize the momentum transfer and achieve the desired properties of the deposited thin film. The efficiency of the momentum transfer is the highest, ${\left(\frac{{\mathit{E}}_{\mathit{t}}}{{\mathit{E}}_{\mathit{i}}}\right)}_{\mathit{max}}\mathit{,}$when$\mathbf{cos}\mathbf{\theta }\mathbf{=}\mathbf{1}$ and m_i = m_t, that is, is desired that the atomic weight of the sputtering gas could be identical to that of one of the target.

Ejected atoms must be able to diffuse freely toward the substrate with desirable little opposition to their movement, which explains the necessity of the sputtering to be done in vacuum conditions. To achieve this, a low pressure within the chamber and a suitable large DC voltage applied between the electrodes, in other words, between the target and the substrate, give rise to a glow discharge that allow accelerate the positive ions to the target. Therefore, ions can retain their high energies. Besides atom-gas collisions can be prevent after ejection from the target. Still, the initial kinetic energy of the atoms transported through the plasma can be lost by collisions within the plasma, failing the energy needed to deposit themselves on the substrate. Thereby, not all atoms ejected from the target reach the substrate, many are projected in different directions and deposit on any surface they encounter. The atoms that can reach the substrate thereby form a layer called a thin film. So, sputtering is also described by its yield, which is the ratio of the number of atoms ejected to the number of incident energetic ions and depends on the chemical bonding of the target atoms and the energy transferred by impact [27].

The sputtering yield (Y) is an important parameter that characterizes the efficiency of the sputtering process since it determines the rate at which atoms are ejected from the target material and deposited onto the substrate. Therefore, the sputtering yield plays a crucial role in the fabrication and processing of thin films, coatings, and surface modifications using sputtering techniques. Y is defined as the number of atoms or molecules sputtered from the target per incident particle. Y is zero for ion energies below the threshold energy of sputtering, Φ. This means that particles with energy below this threshold are not able to cause sputtering. Mathematically, we can express this as Y = 0 for E < Φ, where Y is the sputtering yield and E is the ion energy. Y is a function of the ion energy and the target material properties, with a threshold energy below which sputtering does not occur and a power law relationship above the threshold energy.

The sputtering yield depends on various factors, such as the energy and flux of the incident ions, the target material properties, and the surface conditions. By controlling these factors, the sputtering yield can be optimized to achieve desired properties, such as film thickness, composition, morphology, and adhesion. The sputtering yield also affects the overall efficiency and quality of the sputtering process, as well as the cost and environmental impact.

Reactive sputtering is a widely used technique for depositing compound films on substrates. In reactive sputtering, a target material is bombarded with ions in the presence of a reactive gas such as oxygen, nitrogen, or hydrogen. The sputtered species react with the reactive gas to form a compound film on the substrate surface.

One of the advantages of reactive sputtering is that it allows for precise control of the stoichiometry of the deposited film by adjusting the flow rate of the reactive gas. This makes it possible to deposit films with desired properties such as optical, electronic, magnetic, or mechanical properties.

Another advantage of reactive sputtering is that it can be used to deposit films on complex substrates, fibers, nanoparticles, films and materials with irregular surfaces, porous materials, etc. This is because the sputtered species have high kinetic energies, which enable them to penetrate the pores and irregularities of the substrate surface. As a result, the deposited film can conformally coat the entire surface of the substrate, including its complex features, such as corners, edges, and high aspect ratio structures.

2.4 Magnetron sputtering

The sputtering process is a relatively simple technology, but it still requires additional support systems, such as efficient cooling of the substrate because the electrons that are repelled by the negative cathode can reach the substrate heating it; and the use of magnets to confine the electron paths towards the cathode surface (magnetron sputtering) to increase the plasma efficiency and therefore the deposition rate. This allows the plasma thus located/confined to improve deposition rates due to the greater number of ions colliding with the target and reduces the temperature of the substrate as less electrons collide with it.

The presence of magnets behind the cathode creates a magnetic field close to the surface of the target. These magnets are positioned to produce a magnetic field near the target is a such way that magnetic field lines are parallel to the cathode surface and perpendicular to the electric field lines (Figure 5). This arrangement allows to concentrate the electrons close to the target, as shown in Figure 5a, instead of them circulating randomly dispersed around it, while the ion trajectories are not influenced by the deflection due to their greater mass. The combined action of the electric (E) and magnetic (B) fields near the target generates the E × B drift phenomenon. The trajectories of electrons, of charge q and velocity v, captured in this drift are forced to bend and follow helical trajectories around the magnetic field lines (Figure 5b), and to follow them because of the Lorenz force (F_L). The Lorenz force acting on a charged particle is given by the following equation:

Figure 5.

(a) Layout of the DC magnetron sputtering system near the target; (b) helical electron trajectory around the magnetic field line due to the Lorenz force.

$${\overrightarrow{F}}_L=q(\overrightarrow{E}+\overrightarrow{v}\mathrm{x}\overrightarrow{B})$$E4

where q is the charge of the particle, $\overrightarrow{E}$ is the electric field, $\overrightarrow{v}$ is the velocity of the particle, and $\overrightarrow{B}$ is the magnetic field. The term $\overrightarrow{v}$ x $\overrightarrow{B}$ represents the cross-product of the velocity and magnetic field vectors. In the case of DC sputtering, the magnetic field is typically generated by a permanent magnet. The electric field is created by the potential difference between the cathode and anode. As the electrons move toward the anode, they experience a Lorentz force that causes them to follow a curved path.

This curved path increases the path length of the electrons to the anode, which means they have a larger number of collisions with the argon atoms in the plasma. This, in turn, significantly improves the ionization probability because the collisions between the electrons and argon atoms result in the formation of more ions.

This additionally acting Lorenz force restricts the trajectories of the electrons. Therefore, the path of the electrons to the anode increases which significantly improves the ionization probability because of the larger number of collisions between argon atoms and electrons.

3. TiO₂: brief structure and properties

TiO₂ naturally exists in three polymorphs such as rutile (tetragonal), anatase (tetragonal), and brookite (rhombohedral) [33]. Figure 6 shows the three polymorphic structures of TiO₂, which can be described based on their cell structure consisting of a TiO6 distorted octahedron. Each structure has a different degree of distortion of this octahedron, resulting in the characteristic differences observed between the polymorphs.

Figure 6.

TiO₂ main polymorphs—anatase, rutile, and brookite [34].

TiO₂ is an n-type semiconductor material that exhibits a bandgap ranging from 3.0 eV (for rutile) to 3.2 eV (for anatase), which corresponds to a light absorption edge in the UV range. One of the advantages of TiO₂ is that it is non-toxic, biocompatible, and has a high chemical stability. These properties, along with its exceptional electronic properties, make it an ideal material for use in photovoltaic applications, which convert solar energy into electricity. The discovery by Fujishima and Honda in 1972, which demonstrated the photocatalytic splitting of water in TiO₂ electrodes [35], has led to a growing interest in using TiO₂ as a photocatalyst for environmental purification [7], detoxification, and self-cleaning applications.

Anatase phase has proved to be a more efficient photocatalyst than rutile phase [36], that might be explained by the reduction ability of O₂ higher that rutile, despite the larger band gap of anatase [36]. Recently, amorphous TiO₂ has been considered very effective in antibacterial disinfection [22]. Is well known that TiO₂ photoactivity is hampered by the narrow range of UV wavelengths for photoactivation. Its energy gap is only sensitive to radiation in the ultraviolet (UV) region of the solar spectrum, which represents only 4% of the global solar radiation [37].

Almost 20 years ago it has been reported that the TiO₂ lattice doping with non-metallic atoms like N [38] can shift the absorption edge from UV to lower energies and thus increase visible absorption. Recently, a photonic band gap of 3.18 eV (390 nm) was measured for amorphous TiO₂ [22] which is slightly lower than the value reported to anatase (3.2 eV). TiO₂ absorbs photons and acquires enough energy (hν) to allow an electron in the valence band to jump to the conduction band.

This process (photocatalysis) gives rise to an electron (e⁻)-hole (h⁺) pair, in accordance with the reaction TiO₂ + hν↔ h⁺ + e⁻, further responsible for the elimination of water toxic components by active species (•OH, •O₂, and H₂O₂) generated by redox reactions on the TiO₂ surface.

Rutile and anatase are stable phases at normal conditions an comprise identical TiO₆ octahedron building unit but with diversely sharing corners and edges giving rise to different configurations [22]. The TiO₆ in anatase are arranged in zigzag chains along {221}, sharing four edges, while in rutile, TiO₆ share two edges and link up in linear chains along {001} [39]. These structural differences give rise to different densities and electronic band structures between these two phases [22].

Moreover, the number of shared edges is related to the “energy of the structure” (and thus its stability). Rutile is more stable that anatase (metastable) being the number of shared edges per octahedron, respectively, two and four [22]. The distance Ti-Ti between the center of edge-sharing octahedra being smaller with the decrease of the number of shared edges which provided shorter Ti-Ti distances and a more closely packed crystal structure of rutile. Thus, there is a strong interaction in the Ti-Ti bond of rutile which has only two Ti atoms at the shortest distance. On the other hand, in anatase, the Ti-Ti interaction, instead, depends on four Ti atoms, which allows for a Ti-Ti distance greater than that of rutile. Therefore, rutile exhibits less blockage around each TiO₆ unit, leading to a more stable phase [40].

As in anatase there are four octahedrons, at a distance between them of 3.04 Å, while in rutile, despite its higher density, only two octahedrons are present at 2.96 Å [40], the distinct arrangement of TiO₆ octahedrons gives rise to different structures packaging that will condition the anatase-rutile transition.

In many synthetic routes, amorphous TiO₂ is often the first phase to form. The transformation of amorphous TiO₂ into anatase and/or rutile, usually occurs by effect of temperature, both in wet chemical methods, such as sol-gel [41] and also in DC reactive magnetron sputtering [42].

X-ray diffraction (XRD) and Raman spectroscopy are commonly used techniques for analyzing the crystallization process in materials science.

X-ray diffraction is a technique that involves shining X-rays onto a crystalline material and observing the resulting diffraction pattern. The diffraction pattern is characteristic of the crystal structure and provides information about the crystal lattice parameters, the orientation of the crystal grains, and the degree of crystallinity.

The XRD patterns of anatase (JCPDS card No. 96-900-9087) and rutile (JCPDS card No. 96-900-9084) phases shown in Figure 7 provide important information about the crystal structures of these two polymorphs of TiO₂. In the XRD pattern of anatase, the main peak at 25.3° corresponds to the (101) plane of the crystal structure. This peak is relatively sharp and intense, indicating a high degree of crystallinity and a well-defined crystal structure. The presence of other peaks at lower angles also indicates the presence of other crystallographic planes in the anatase structure.

Figure 7.

XRD patterns for anatase and rutile TiO₂ phases. The insets are Miller indices of anatase and rutile phases. Adapted from [43].

In the XRD pattern of rutile, the main peak at 27.4° corresponds to the (110) plane of the crystal structure. This peak is also relatively sharp and intense, indicating a well-defined crystal structure and a high degree of crystallinity. The presence of other peaks at higher angles also indicates the presence of other crystallographic planes in the rutile structure.

Raman spectroscopy, on the other hand, involves shining laser light onto a material and measuring the scattered light as a function of wavelength. The scattered light provides information about the vibrational modes of the atoms in the material, which are characteristic of the crystal structure.

By combining XRD and Raman spectroscopy, a more comprehensive understanding of the crystallization process in a material is obtained. XRD provides information about the long-range order and crystal structure, while Raman spectroscopy provides information about the short-range order and local structure.

Anatase has a tetragonal crystal structure and is characterized by Raman peaks at around 144, 399, and 519 cm⁻¹. The peak at 144 cm⁻¹ is due to the symmetric stretching vibration of the Ti-O bond, while the peaks at 399 and 519 cm⁻¹ are due to the bending modes of the Ti-O-Ti bond. The Raman spectrum of anatase is also characterized by a broad peak at around 639 cm⁻¹, which is due to the lattice vibrations of the TiO₆ octahedra. Figure 8 shows the Raman spectrum of a pure anatase film.

Figure 8.

Raman spectra of anatase TiO₂ phase.

Rutile, on the other hand, has a tetragonal crystal structure and is characterized by Raman peaks at around 143, 445, 610, and 880 cm⁻¹. The peak at 143 cm⁻¹ is due to the symmetric stretching vibration of the Ti-O bond, while the peaks at 445, 610, and 880 cm⁻¹ are due to the bending modes of the Ti-O-Ti bond. The Raman spectrum of rutile is also characterized by a sharp peak at around 237 cm⁻¹, which is due to the lattice vibrations of the TiO₆ octahedra.

When a material undergoes a phase transformation from the amorphous or liquid phase to a crystalline phase, the specific crystalline phase that forms depend on a number of factors, including the thermodynamic stability of the different phases and the kinetics of nucleation and growth. In the case of TiO₂, the initial crystalline phase that forms is generally anatase, rather than rutile, because of its lower surface free energy compared to the rutile structure. Surface-free energy is a measure of the amount of energy required to create a unit area of a material’s surface. Materials with lower surface free energy are typically more stable, because they have a lower tendency to form new surfaces or interfaces. In the case of TiO₂, the anatase structure has a lower surface free energy than the rutile structure, which means that it is more thermodynamically stable.

This difference in surface free energy is due to the different crystal structures of anatase and rutile. The anatase structure has a higher percentage of exposed (001) surfaces, which have a lower surface free energy compared to the (110) and (100) surfaces that are more prevalent in the rutile structure. As a result, the anatase structure is more stable and more likely to form during the crystallization process.

The surface roughness and microstructure can significantly influence the performance and hence the purpose of TiO₂ thin films. These characteristics depend on the deposition process, type of substrate, and chosen deposition parameters.

Liang et al. have produced TiO₂ films by the sol-gel method [44], highly compacts, continuous and smooth (Figure 9), exhibiting excellent self-cleaning properties. Figure 10 shows SEM images of TiO₂ films (on glass substrates) prepared by reactive magnetron sputtering under different deposition conditions, namely plasma O₂ concentration (50% and 75%) and used power (500 and 1000 W) [18]. It can be seen clearly the differences in the morphology of the surface of TiO₂ coatings, as a function of different powers and concentrations of O₂. In general, the morphology is typically constituted by several agglomerates of nanoparticles (or grains) in the shape of a cauliflower but of different sizes, which are distributed over the surface of the substrate [18] in accordance with what was reported by Sério et al. [42]. There is a variation in the size of the agglomerates in the morphology of the films dependent on the O₂/(Ar + O₂) ratio.

Figure 9.

SEM image of the surface morphology of TiO₂ film deposited onto glass substrates by dipping-based sol-gel method. Adapted from [44].

Figure 10.

SEM images of surface morphology of TiO₂ films deposited onto glass substrates by reactive magnetron sputtering [18].

The value of the thickness (t_h) of the films allows estimating the deposition rate (v_d), in nanometers per minute.

$$v_d=\frac{t_h}{t}$$

where t is the deposition time. The thickness measurement is performed on SEM images in cross-section (as exemplified in the insert of Figure 10a). Regardless of the geometry, the surface is covered evenly.

4. Characteristics of TiO₂ thin films deposited on natural fiber surfaces by DC reactive magnetron sputtering

The surface is the boundary that separates the material and the external environment. All alterations that occur in the sense of changing the surface structure will condition and tailor the material for certain applications. A wide range of surface modification processes, such as reactive magnetron sputtering, thermal oxidation, thermal evaporation, molecular-beam-epitaxy, chemical vapor deposition, sol-gel-assisted by dip-and/or-spin coating, spray pyrolysis, and electrodeposition have been used for used to produce thin film and tailor surfaces. However, in many of them, adhesion of the coating to the supports is still an important issue, which deserves special attention, especially in some types of substrates, such as natural fibers, which main composition is cellulose.

Cellulose is a polymeric chain with abundant hydroxyls (–OH) groups and other oxygen-containing functional groups –C=O, –C–O–C–, –CHO, and –COOH which makes the fiber surface potentially reactive [18]. These functional groups are available to bond to desired molecules and provide new properties and new applications for natural fibers [18]. The deposition of suitable coatings, such as TiO₂ films, allows the optimization of natural fibers by creation of new tailored properties of their surfaces, such as the photocatalytic ones, which are independent of that exhibited by the bulk fiber. Recently, nanostructured TiO₂ films successfully deposited on ginger lily fiber surfaces have been created by DC reactive magnetron sputtering [18].

The efficiency of the DC magnetron sputtering process to functionalize natural fibers (Figure 11) depends not only on the quantity cellulose reactive accessible groups but also on sputtering conditions, such as the operating pressure, discharge power, O2 gas partial pressure, and deposition time.

Figure 11.

SEM images of ginger lily fibers: (a) pristine; (b) after TiO₂ sputtered at 50% O₂–1000 W [45].

The morphology of the films can be tailored by change of the partial pressure of the reactive gas (Ar/O₂) and the sputtering power [10]. Moreover, different film typology, namely dense and porous, can be obtained, as well as amorphous or crystalline (anatase and/or rutile) nanofilms. The influence of the O₂% in the discharge and the sputtering power on the amorphous/anatase phase transition, surface stoichiometry, and surface roughness of the films can be tailored.

4.1 Fibers preparation

Ginger lily fibers are obtained by mechanical extraction from the stems of the plant, after removing the leaves, as shown in Figure 12a–c. Before TiO₂ deposition by reactive magnetron sputtering (Figure 12d), fibers are cleaned successively in acetone, isopropanol, and deionized water to remove any organic contamination and further dried at low temperature (about 30°C).

Figure 12.

Sequence of the ginger lily fiber preparation process for deposition of TiO₂ films by reactive magnetron sputtering. (a) Plant harvest; (b) stem preparation; (c) extraction of long fibers; (d) film deposition apparatus.

4.2 Scanning electron microscopy (SEM) and energy dispersive X-ray analysis (EDX)

The EDX analysis (Figure 13) was used to determine the elemental composition of the TiO₂ films. The results show that as the power used during the reactive magnetron sputtering process increases, there is a corresponding increase in the intensity of the Ti peak. This finding supports the expected result that an increase in the number of Ti-O bonds contributes to the growth of the TiO₂ film. In other words, the higher the power used during the sputtering process, the greater the amount of titanium present in the film.

Figure 13.

SEM/EDX of the TiO₂ films deposited by DC reactive magnetron sputtering: (a) 50% O2–500 W, (b)75% O2–500 W, (c)50% O2–1000 W, (d)75% O2–1000 W.

As the power and oxygen percentage during the reactive magnetron sputtering process are increased, the resulting TiO₂ films exhibit a nanostructured morphology in certain areas, similar to that seen in Zone 1 of Thornton et al.’s model [46]. This morphology is primarily due to the adatoms on the surface of the growing film having low mobility and the “shadow” effect. A nanostructured thin film exhibits nano-scale surface features, typically ranging in size from a few nanometers to several hundred nanometers. These features can include nanopores, nanocrystals, nanotubes, or other nano-items that are engineered into the film’s surface by adjusting the deposition parameters, such as temperature, pressure, and substrate morphology. The structural and physical properties of the thin film can be controlled at the nanoscale. The nanostructured surface greatly increases the surface area of the coating and is well suited to photocatalytic applications due to its large surface area-to-volume ratio. The presence of pores, for example, increases the density of active sites with high accessibility of photons, but also facilitates diffusion and increases the adsorption capacity of pollutants. The size, shape, and distribution of the pores can be precisely controlled, allowing for the customization of the film’s properties for specific applications.

In sputtering, the bombardment of the substrate surface with high-energy ions causes atoms to be ejected from the target and deposited onto the substrate surface. If the bombardment intensity is insufficient for film densification, the presence of pores can dominate the film’s structure. When the deposited atoms do not have enough kinetic energy to overcome the surface diffusion and adhesion forces, they can accumulate on the substrate surface and form islands. These islands can coalesce and form a continuous film, but the presence of voids and pores between the islands can significantly affect the film’s properties.

The presence of pores in sputtered films can have both positive and negative effects on their properties. For example, in some applications, such as sensing or photocatalysis, the large surface area-to-volume ratio provided by the pores can enhance the film’s activity. On the other hand, in other applications, such as barrier coatings or electronic devices, the presence of pores can reduce the film’s performance and durability.

The greater availability of oxygen (75%) in the chamber during sputtering causes more ions to be generated, leading to an increase in the number of atoms bombarding the surface of the growing film. This results in a denser film which can lead to the formation of ridges and depressions on the film’s surface.

Overall, the effect of the increased oxygen concentration in the chamber during sputtering is to create films with a high roughness topography, which can be advantageous or disadvantageous depending on the intended application. The ability to control the oxygen concentration and other process parameters during sputtering is therefore important for achieving the desired film properties.

4.3 X-ray photoelectron spectroscopy (XPS)

Figure 14a is shown the XPS survey spectra of the film 75% O2–1000 W. Carbon and oxygen lines dominate as expected because of the organic nature of the fiber. Intense Ti lines are also observed due to the TiO₂ film on the fiber surface. Typically, the fiber surface area can be divided into two kinds of regions: those covered with TiO₂ and those covered with organic material. These two regions are on different potentials, so that their reference binding energies are different. Nevertheless, the analysis can be performed by using the charge reference Ti 2p3/2 assumed to be at 458.5 eV, which is characteristic for TiO₂ phase [47] and the C 1 s line, with the smallest binding energy corresponding to adventitious carbon at 284.8 eV. It is believed that Ti at the surface of a “TiO_x material” is generally present as TiO₂. Since in this case of study, the measured Ti 2p lines clearly show only a single phase, as can be seen from Figure 14b, which confirms that only TiO₂ phase is present.

Figure 14.

(a) XPS survey spectrum from coated fiber 75% O2–1000 W, (b) high-resolution XPS spectrum of the line Ti 2p, and (c) high-resolution XPS spectrum of the line C 1 s taken from the pristine fiber. Adapted from [18].

Deposition of TiO₂ increase the amount of oxygen at the fiber surface. This fact can be interpreted in three ways: (a) the reactive atmosphere during the deposition process contribute to significant surface oxidation of the surface; (b) the reactive atmosphere in the magnetron chamber etches (probably chemically) the surface and “opens” oxygen-rich phases laying below the carbon-rich surface layer; and (c) after the magnetron sputtering the samples are able to adsorb more water which is bound strongly so that it remains at the surface in vacuum [18].

During the deposition process, the increase in oxygen content mainly occurs for two reasons: (1) deposition of the TiO₂ film and (2) oxidation of the organic material. The latter occurs due to the presence of O–C–O and COOH groups in the fiber [18]. The XPS analysis in this study is related to the fitting of the C 1 s line, which have four contributions related to (a) C-C and C-H bonds, (b) C-OH and C-O-C bonds, (c) O-C-O bonds, and (d) COOH group [18]. From Figure 14c, C1 is attributed to the saturated C-C and C-H bonds. C2 at 287.0 eV is attributed to oxygen bound to two neighboring carbon atoms, forming a triangle. C3 at 288.7 eV can be attributed to carboxyl group (C=O)-OH, and C4 can be only attributed to –O–(C=O)–O– group [18].

4.4 Fourier transform infrared (FTIR) spectroscopy

The FTIR spectra observed in Figure 15 show the presence of TiO₂ on the surface of the fibers. The peak observed between 800 and 450 cm⁻¹, at 670 cm⁻¹, is quite intense in the 75% O2–1000 W sample, being attributed to the Ti–O elongation, which is one of the characteristic peaks of the FTIR spectrum of TiO₂. Švagelj et al. [48] in a study of TiO₂ deposition on Al₂O₃ substrates, they reported the presence of the Ti–O elongation band, in the range of 640–700 cm⁻¹. This peak is associated with the presence of O–Ti bonds in the TiO₂ film, which, in turn, bond to the surface of natural fibers, possibly by hydrogen bonding or van der Waals forces.

Figure 15.

FTIR spectra of the TiO₂ films deposited by DC reactive magnetron sputtering in the 900–500 cm⁻¹. Adapted from [18].

4.5 X-ray diffraction (XRD)

The structure of the deposited films is influenced by various deposition parameters, such as sputtering power, pressure, target-substrate distance, and the amount of reactive gases present in the deposition chamber.

The formation of a solid film during the sputtering process is affected by two factors: the heat generated by the substrate and the energy of the sputtered particles hitting the substrate. In situations where the substrate is not intentionally heated, it can still reach temperatures between 60 and 100°C due to the energy transfer from the sputtered particles. Normally, amorphous TiO₂ films require annealing at temperatures above 300°C to crystallize. However, Sério et al. [10] observed that crystallization occurred in as-sputtered TiO₂ thin films not because of the thermal energy, but rather due to the energy of the sputtered particles.

The sputtered particles could be from the target, such as atomic Ti, molecular TiO, molecular TiO₂, and TiO₂ clusters, as well as energetic electrons, negative ions (O⁻), and neutrals reflected from the target (e.g., atoms of argon and oxygen) [10]. The films prepared with a sputtering power of 1000 W were found to be crystalline, likely due to the enhancement of plasma density in front of the substrate and an increase in the cluster growth rate with an augment in the sputtering power (Figure 16) [10].

Figure 16.

XRD patterns of as-sputtered TiO₂ thin films deposited at 5% O2–500 W, 20% O2–1000 W, and 50% O2–1000 W [10].

5. Conclusion

Ginger lily fibers are a new sustainable resource for new product development. Its surface can be coated with a thin film of TiO₂ by reactive magnetron sputtering whose topography can be tailored depending on the power used and the percentage of oxygen in the chamber. Regarding the structure, the application of 500 W power allows obtaining a TiO₂ film with an amorphous structure, while the samples with twice the power showed polycrystalline structures. Anatase is the dominant phase in the films deposited at 1000 W. XPS and FTIR analyses revealed that ginger lily fibers can serve as a new sustainable resource for developing novel products, and the topography of the TiO₂-coated fibers can be tailored by adjusting the power and oxygen percentage during reactive magnetron sputtering.

Conflict of interest

The authors declare no conflict of interest.