Innovative Elaboration of Polyvinylidene Fluoride Thin Films via Dip-Coating: Beta Phase Optimization, Humidity Control, Nanoparticles Addition, and Topographic Analysis

Abstract

Polyvinylidene fluoride (PVDF) is a multifunctional polymer renowned for its unique electrical, mechanical, and piezoelectric properties, making it an attractive candidate for various applications. Although the spin-coating method has been the conventional method for fabricating PVDF thin films, this work is the first to apply the dip-coating technique with humidity control, which is a largely unexplored method in the literature on PVDF thin films. This novel approach offers great prospects for improved control and performance adjustments, as well as expanding the range of film deposition procedures. Here, we examine the phase composition of PVDF thin films; adjust different parameters to optimize the electroactive phases fraction, especially the Beta phase; and examine how relative humidity affects the properties of the film. Moreover, we test the impact of different nanoparticles’ addition on the phases fraction and characteristics of the film. Furthermore, we analyze the topography of the resultant films using several approaches, providing fresh insights into their structural features.

1. Introduction

Research on piezoelectric materials has taken a significant turn towards improving their properties and searching for new application leads. Developments from the creation of polymer-based types of piezoelectrics, like polyvinylidene fluoride (PVDF) [1], possess flexibility and, therefore, can be readily fit into any device, compared to traditional types made out of ceramic materials, such as lead zirconate titanate (PZT).

Polyvinylidene fluoride (PVDF) is a highly nonreactive, pure thermoplastic fluoropolymer that has attracted much industrial interest due to its piezoelectric behavior [2]. It is versatile and robust for various applications, from industrial to medical fields [3].

PVDF is a vinylidene fluoride polymer with the chemical formula (CH2-CF2)n. The polymer has high resistance to solvents, acids, and bases, which makes it chemically resilient. Excellent stability under conditions of high temperature and UV light is exhibited by the polymer, too. PVDF usually possesses a melting point of about 170 °C and a density of about 1.78 g/cm³. The material is characterized by unique properties, which emanate from its molecular structure: predominantly strong carbon–fluorine bonds and high crystallinity [4]. PVDF can exist in several polymorphic forms, as diplayed in Figure 1 [5], each displaying distinct properties:

• Alpha ($\alpha$) phase: This is the most common and least polar phase, characterized by a TGTG’ conformation where T stands for trans (180°) and G for gauche (±120°). This phase is non-polar and does not exhibit piezoelectric properties until it is mechanically or electrically manipulated.
• Beta ($\beta$) phase: The most relevant phase for piezoelectric applications, the $\beta$-phase, exhibits all-trans planar zigzag chains. It is induced from the alpha phase via mechanical stretching or poling under an electric field. The $\beta$-phase is highly polar and exhibits strong piezoelectric and pyroelectric properties [6].
• Gamma ($\gamma$) phase: Characterized by a T3GTT3G’ conformation, this phase also exhibits piezoelectric properties but to a lesser extent than the beta phase.
• Delta ($\delta$) and Epsilon ($ϵ$) phases: These are rare phases and can be obtained under special mechanical and thermal conditions.

The applications of PVDF are numerous and highly diversified, mainly because of its piezoelectric, ferroelectric, and pyroelectric features, and they include energy harvesting [7], sensors and actuators [4], medical [3,8], and filtration membranes [9]. Advantages of using PVDF include the following:

• Chemical Resistance: PVDF has excellent resistance to chemicals and solvents, making it suitable for use in harsh chemical environments [10,11].
• Thermal Stability: It has high thermal stability, capable of operating continuously at temperatures up to 170 °C [12].
• Mechanical Strength and Toughness: PVDF offers a good balance of mechanical properties, including strength, toughness, and flexibility [13].
• Electrical Properties: PVDF exhibits excellent dielectric properties and piezoelectric response [14].

2. Experimental Section

The solutions used in our studies were mainly made by dissolving PVDF in Dimethylformamide (DMF). We varied the mass concentration percentages of PVDF each time based on the needed outcome and to modify certain properties of our thin film, especially thicknesses and phases composition, as shown in

Table 1. Physico-chemical properties and quantities of DMF and PVDF utilized to make X% PVDF films.

	M (g/mol)	d (g/mL)	V (mL)	m (g)
DMF (Liquid)	73.09	0.95	10	9.50
PVDF (Powder)	534,000	—	—	C (m)\%	Mass added (g)
				5	0.500
				10	1.055
				15	1.676

.

2.1. Deposition Techniques

In the literature, the use of the spin-coating technique for the elaboration of nanometric PVDF thin films is well documented [15,16]. Meanwhile, in our work, we focused on the use of another technique that has rarely been used for this kind of application: the dip-coating technique [17,18,19,20].

2.1.1. Substrate Used and Preparation

The wafers on which the PVDF solution was deposited were 250 μm thick one side, and they were polished silicon wafers (orientation (100) provided by Siltronix), which were IR-Transparent. Before they were used, the silicon wafers were thoroughly cleaned to remove any pollution or other traces that could affect the quality of our films. The first cleaning step consisted of immersing them in an ultrasonic bath of ethanol, followed by a drying step using a regulated-pressure blower. The same procedure was repeated in acetone. The silicon substrates were then placed in the UV/Ozone Cleaner (Model ProCleane Plus from BioForce Nanosciences) for a minimum of 30 min. This is a very important step, as it provides a totally hydrophilic surface and eliminates organic traces, which significantly improves the quality of the thin film after deposition.

2.1.2. Spin Coating

Spin coating is one of the most common methods for depositing thin films on substrates and is widely applied in the semiconductor and electronics industries [21], as well as in many other application areas; for example, it is used in manufacturing optical and protective coatings and in fields of research concerned with experimental physics and materials science [22]. This process, due to its simplicity, repeatability, and the possibility of producing uniform films over large areas, is one of the most valuable methods. The process includes the application of a small quantity of the solution, which contains a solute in a solvent, to the flat substrate center. The wafer is then spun at a high speed, whereby the liquid spreads outward due to the centrifugal force and forms a thin and even film as the solvent evaporates. The spin coater used was Delta6 from SUSS MicroTec SE (Sternenfels, Germany) with a maximum rotation speed of 3000 rotations per minute (RPM) and an acceleration ramp between 0.5 s (noted RO) and 5 s (R9). The rotation time at the speed set in this study at 500, 1000, and 1500 RPM was set at 60 s for an acceleration ramp of R0 and R9.

2.1.3. Dip Coating

In materials science, dip coating is a simple yet effective method for depositing thin films on substrates. It is widely utilized in the creation of optical coatings, protective layers, and other kinds of sensors [23]. This process entails dipping a substrate into a coating solution, pulling it out at a predetermined pace, and then waiting for a chemical reaction or solvent evaporation to form the film [24]. Dip coating (Figure 2) involves a few crucial steps [25]:

• Immersion: A liquid solution containing the intended film material suspended or dissolved in a solvent is applied to the substrate.
• Dwell Time: The solution stabilizes and interacts equally with the surface while the substrate is submerged for a certain amount of time.
• Withdrawal: A liquid film is dragged behind the substrate as it is gradually removed at a regulated pace.
• Drainage and Evaporation: A thin film is left behind when the liquid layer thins as a result of gravity and solvent evaporation.
• Curing: In order to attain the required mechanical and chemical qualities, the film may go through further procedures like heat or UV curing.

Several physical forces and fluid dynamics play crucial roles in determining the quality and characteristics of the film, most importantly [26] viscous force, gravity, surface tension, and capillary force.

Moreover, several process parameters have significant impacts on the film formed by dip coating [24], especially the withdrawal speed, the concentration and composition of the solution, the temperature and humidity, and the properties of the substrate.

2.1.4. Film Thickness Measurement

In this study, a Dektak Pro (Bruker, Durham, UK) stylus profilometer (contact measurement) was used to measure the thickness of the PVDF films. To enhance precision, we applied multiple scratches on each film’s surface and measured the thickness in different zones. This approach allowed us to obtain a more accurate estimation of the thickness and the associated measurement error range.

2.1.5. PVDF Phase Fraction Calculation and Formulas

One of the most important characterization techniques for the identification of the different phases of PVDF is FTIR spectroscopy. The FTIR used was a commercial Nicolet 8700 FT-IR from Thermo Fisher Scientific (Illkirch-Graffenstaden, France) In an earlier work, we based our calculations and analysis of the phases on the works of Gregorio et al. [27] by mainly calculating the fraction of the $\beta$ phase using the following formula:

$$F\left(\beta \right)={\displaystyle \frac{X_{\beta }}{X_{\alpha }+X_{\beta }}}={\displaystyle \frac{A_{\beta }}{\left(K_{\alpha }/\phantom{K_{\beta }{K}_{\alpha }}{K}_{\beta }\right)A_{\alpha }+A_{\beta }}}$$

(1)

$X_{\alpha }$ and $X_{\beta }$ are the fractions of $\alpha$ and $\beta$, respectively; $K_{\alpha }$ and $K_{\beta }$ have values of $6.1\times {10}^4$ cm²/mol and 7.7 × 10⁴ cm²/mol, respectively; $A_{\alpha }$ and $A_{\beta }$ are the absorbance at 763 cm⁻¹ and 840 cm⁻¹ [27], respectively.

In general, various vibration bands in the FTIR spectra of PVDF signify the presence of one or multiple different phases, but based on the mentioned study, we should focus on the vibration band at 763 cm⁻¹ (which characterizes the $\alpha$ phase) and the band at 840 cm⁻¹ (which characterizes the $\beta$ phase). These two bands can be used to calculate the fraction of each phase using the formula mentioned above.

At the early stages of the research, we exclusively used this formula for calculations, but as we delved deeper into the literature and after analyzing some of the results we had obtained, we noticed that this formula was incomplete, as it considered only the presence of the $\alpha$ and $\beta$ phases, while in reality, the ($\gamma$) phase was also present in the films. To this end, the vibration band at 840 cm⁻¹ represented, in fact, the totality of the electroactive phases ($\beta +\gamma$) [28] and not exclusively the ($\beta$) phase. Thus, we can conclude that the previous formula does not give us F($\beta$), but more specifically F($\beta +\gamma$), which is also noted as FEA (electroactive fraction), and the new formula becomes the following:

$$F(\beta +\gamma )={\displaystyle \frac{X_{\beta +\gamma }}{X_{\alpha }+X_{\beta +\gamma }}}={\displaystyle \frac{A_{\beta }+\gamma }{\left(K_{\beta +\gamma }/\phantom{K_{\beta +\gamma }{K}_{\alpha }}{K}_{\alpha }\right)A_{\alpha }+A_{\beta +\gamma }}}$$

(2)

$X_{\alpha }$ and $X_{\beta +\gamma }$ are the fractions of $\alpha$ and $\beta +\gamma$, respectively; $K_{\alpha }$ and $K_{\beta +\gamma }$ have values of 6.1 × 10⁴ cm²/mol and 7.7 × 10⁴ cm²/mol, respectively; $A_{\alpha }$ and $A_{\beta +\gamma }$ are the absorbance at 763 cm⁻¹ and 840 cm⁻¹, respectively.

$A_{\alpha }$ and $A_{\beta +\gamma }$ can be obtained by integrating and calculating the surface area underneath the vibration bands at 763 cm⁻¹ and 840 cm⁻¹, respectively.

The challenge now lies in accurately identifying and quantifying each of the three distinct phases. This requires examining a second particular section of the FTIR spectra, specifically around the 1200 cm⁻¹ region, as detailed by Cai et al. [28]. More precisely, the band at 1210 cm⁻¹ represents the $\alpha$ phase, the band at 1234 cm⁻¹ represents the $\gamma$ phase, and the band at 1275 cm⁻¹ represents the $\beta$ phase.

It is important to note that this approach is largely qualitative and not quantitative, but the different phases can be roughly estimated by applying the following formulas to obtain the new and final F($\beta$) and F($\gamma$):

$$\begin{array}{c}F\left(\beta \right)=F_{EA}\times \left({\displaystyle \frac{\Delta H_{{\beta }^{\prime }}}{\Delta H_{{\beta }^{\prime }}+\Delta H_{{\gamma }^{\prime }}}}\right)\times 100\%\hfill \\ F\left(\gamma \right)=F_{EA}\times \left({\displaystyle \frac{\Delta H_{{\gamma }^{\prime }}}{\Delta H_{{\beta }^{\prime }}+\Delta H_{{\gamma }^{\prime }}}}\right)\times 100\%\hfill \end{array}$$

(3)

where $\Delta H_{{\beta }^{\prime }}$ and $\Delta H_{{\gamma }^{\prime }}$ are the height differences (absorbance differences) between the peak at 1275 cm⁻¹ and the nearest valley at 1260 cm⁻¹, and the peak at 1234 cm⁻¹ and the nearest valley at 1225 cm⁻¹, respectively (see Figure 3) [28].

In the totality of this work, the calculations and estimations obtained for the different phases are based on this approach.

On the other hand, Raman shift can also be used for the identification of the different phases of PVDF, especially as described and reported by Pavlović et al. [29]. The sharp Raman band at 795 cm⁻¹ indicates the presence of the $\alpha$ phase, the band at 812 cm⁻¹ indicates the presence of the $\gamma$ phase, and the band at 839 cm⁻¹ indicates the presence of both the $\beta$ and $\gamma$ phases, but by looking at the 812 cm⁻¹ band at the same time as the 839 cm⁻¹ band, we can deduce if there is a coexistence of both phases or not, and which phase is the dominant one. Other bands can also indicate the presence of the different phases, but we usually focus on the three stated bands to prove the existence of each.

Furthermore, the XRD technique can be used to identify the different PVDF phases, which is also based on multiple studies [28]. The presence of a small wide peak at 18.4° and an intense sharp peak at 19.9° indicates the presence of the $\alpha$ phase. Meanwhile, a very small wide peak at 18.5° along with a larger peak at 20.2° indicates the presence of the $\gamma$ phase. Finally, a strong peak at 20.6° indicates the presence of the $\beta$ phase.

In a lot of cases, when there is a coexistence of multiple phases, the identification becomes harder due to the peaks being very close to each other, which leads to a unified wider peak representing multiple phases at once (with a small 2 Theta(°) deviation).

2.2. Equipment Used

The dip-coater used in our work (see Figure 4a) is a state-of-the-art dip-coater (Model ACEdip 2.0 from Solgelway-France) which consists mainly of 3 parts: a small compartment in which we put the dipping solution, a chamber inside which we fix our wafer, and a mobile platform that moves upwards and downwards.

The ascent and descent speed of the platform can be finely tuned down to 0.0001 mm/s, which enables us to obtain a wide range of film thicknesses, from multiple micrometers-thick all the way to below 10 nm thin films. The temperature of the chamber can be controlled, along with the relative humidity, % using an independent system (Figure 4b).

2.3. NPs Addition

In the pursuit to increase the electroactive fraction, especially the $\beta$ phase in our films, we decided to add different types of nanoparticles. Based on earlier studies in the literature, the addition of certain nanoparticles to PVDF can negatively or positively impact the piezoelectric properties of the thin film [14,30,31]. To this end, we tested the following nanoparticles, which differ in nature, structure, and properties.

A well-known ferroelectric ceramic material with superior dielectric, piezoelectric, and electro-optic characteristics is barium titanate (BaTiO₃) [32]. Because of these qualities, it is an essential part of many different applications, including transducers, actuators, sensors, and capacitors. BaTiO₃ nanoparticles in particular, with their larger surface area and smaller size, provide better characteristics than their bulk counterparts.

In previous studies seen in the literature, barium titanate had a positive impact on PVDF [33], as it enhanced, to some extent, the electroactive fraction, which is why we chose to utilize it.

Another well-known ferroelectric material with exceptional piezoelectric, electro-optic, and nonlinear optical capabilities is lithium niobate (LiNbO₃) [34]. Because of these qualities, it is an essential part of many applications, including photonics, telecommunications, and sensor technologies.

LiNbO₃ nanoparticles represent a promising material with unique properties that are highly advantageous for a variety of applications. To this end, we experimented with the addition of LiNbO₃ NPs to PVDF [35].

Yttrium iron garnet (YIG) is a ferrimagnetic material possessing distinct optical, magnetic, and microwave properties [36]. Its chemical formula is Y₃Fe₅O₁₂. Because of their improved magnetic properties—which are essential for applications in microwave devices, magnetic sensors, and spintronics—YIG nanoparticles in particular have attracted a lot of attention. YIG NPs had been chosen due to the fact that magnetic nanoparticles had been proven to increase the electroactive fraction (especially the Beta phase) when added to PVDF [37].

The nanoparticles used were purchased from Alfa Aesar. Their size and properties are given in

Table 2. Characteristics of the nanoparticles (NPs) used in this work.

Nanoparticles	Size- NP Diameter (nm)	MW (g/mL)	d (g/mL)	Provider
Barium Titanate BaTiO3	50 to 70	233.19	5.85	Alfa Aesar
Lithium Niobate LiNbO3	<80	147.85	4.30	Alfa Aesar
Yttrium Iron Garnet (YIG) Y3Fe5O12	<100	737.94	5.11	Alfa Aesar

.

To achieve a homogeneous dispersion of magnetic nanoparticles in the PVDF solution, a multi-step approach was implemented. Initially, the magnetic nanoparticles were dried to eliminate any residual moisture, and then they were accurately weighed before being gradually added to N,N-Dimethylformamide (DMF). Given the tendency of magnetic nanoparticles to cluster when exposed to magnetic fields, mechanical stirring was avoided. Instead, the dispersion was first homogenized using a vortex mixer for a few minutes, followed by ultrasonication in a bath sonicator for 60 min. The temperature was controlled to prevent solvent evaporation or nanoparticle degradation.

3. Results and Discussion

As already mentioned, the dip-coating technique is rarely reported in the literature for the elaboration of PVDF thin films [38]. It would, thus, be quite interesting to study and test this technique, as it could lead to favorable results in multiple areas. In this section, we first begin by examining the impact of dipping and withdrawal speeds on the films’ thickness and composition; then, we detail the results obtained by controlling the relative humidity percentage, a step that has not been previously taken with dip-coated PVDF thin films.

We also discuss the outcomes of adding nanoparticles to the PVDF solution, especially the magnetic ones. This leads us to the results obtained by simultaneously combining humidity control and NP addition, and the influence this combination has on the film.

A study of the topography and roughness of the surface is conducted for each section using multiple imaging techniques.

3.1. Dipping and Withdrawal Speed Impact

We start by presenting the results of the thickness and electroactive phase fractions of PVDF films at varying PVDF concentrations and withdrawal speeds. These results are compared with those obtained with spin coating.

As can be seen in Figure 5, we can deduce that by using the dip-coater, we can obtain a wider range of film thicknesses in comparison with the spin coater. We can deposit both thicker and thinner films as a function of withdrawal speeds. In dip-coating, the withdrawal speed significantly influences the final film thickness. This effect becomes more pronounced at higher PVDF concentrations due to several interrelated factors, as can be seen in Figure 5a. The influence of withdrawal speed on film thickness is more pronounced at higher PVDF concentrations due to increased viscosity, leading to stronger capillary effects, enhanced shear effects, and slowed-down liquid drainage leading to higher film retention as speed increases. The Landau–Levich model predicts this strong dependence of thickness on speed for viscous solutions. According to the Landau–Levich model, film thickness t scales with withdrawal speed $w_s$ following a power-law relationship, with viscosity $\eta$ playing a crucial role:

$$t\propto {\eta }^{2/3}{w}_s^{2/3}$$

(4)

At higher PVDF concentrations, the increased viscosity amplifies this effect, leading to a steeper increase in thickness with speed. Additionally, rapid solvent evaporation further locks in thickness variations, preventing fluid redistribution. These combined effects create a stronger correlation between speed and final film thickness compared to low-concentration solutions, where drainage occurs more freely. This enhanced sensitivity of thickness to speed is a well-documented phenomenon in dip-coating processes [24].

An RPM higher than 1500 can be used to obtain very thin films via spin coating, but then they become very heterogeneous and cracked, thus useless for any application.

Meanwhile, using the dip-coater, we obtain compact and homogeneous films on both ends of the spectrum (Figure 6), which is a first clear advantage over using the spin coater. It is also important to note that we tested the impact of dipping speed and the dwelling duration inside the solution, and we found that there was no significant influence on the films whatsoever.

Figure 7 shows the electroactive fraction of different spin-coated films at different RPMs, accelerations, and PVDF concentrations. For 5% PVDF films, the maximum that can be obtained is ≈52%; meanwhile, for the 10% film, it is ≈77%. In both cases, the films obtained are quite thick (multiple micrometers), with a rough heterogeneous surface. As for Figure 8, showing dip-coated films at different withdrawal speeds, we can see that the overall electroactive fraction is very high (>90%) in a specific zone of the withdrawal speed range (different for each concentration of PVDF).

For 5% PVDF, in the zone going from 0.025 mm/s to 0.1 mm/s, fractions of the electroactive phases in the excess of 90% are obtained throughout. For 10% PVDF, this zone is between 0.001 mm/s and 0.03 mm/s, where the fraction is also constantly higher than 90%.

Figure 7 further shows the evolution of different phase contents, as the electroactive phases increase with decreasing withdrawal speed, and the inverse can be said for the Alpha phase, which decreases by lowering the speeds. So, how can we explain this phenomenon?

In general, a substrate is dipped into a PVDF solution and removed at a predetermined speed during the dip-coating process. As the substrate is pulled out, a thin liquid film forms and undergoes solvent evaporation, leading to the formation of a solid film. The speed at which the withdrawal occurs and the transversal forces that are applied during withdrawal have a significant impact on the thickness of the resulting film and the crystallinity of PVDF.

• Slow Withdrawal Speeds:
  • Thin Film Formation: The slow movement enables a more uniform solvent evaporation, which gives polymer chains longer time to rearrange [23].
  • High $\beta$-phase and $\gamma$-phase Fractions: The slow evaporation rate enhances the formation of both $\beta$ and $\gamma$ phases in PVDF. This slower solvent loss facilitates the necessary planar zigzag conformation for $\beta$-phase formation and supports the distinct packing needed for $\gamma$-phase. The extended withdrawal time allows PVDF chains to align and pack efficiently into these electroactive structures.
  • Enhanced Electroactivity: Higher $\beta$- and $\gamma$-phase contents enhance electroactive properties. Slow solvent evaporation leads to fewer defects and better molecular ordering, improving the piezoelectric performance of PVDF.
• Intermediate Withdrawal Speeds:
  • Moderate Film Thickness: At intermediate speeds, the solvent evaporation rate is faster, leading to less time for molecular rearrangement and thicker films than the ones obtained at slow speeds.
  • Mixed Phases: The PVDF films obtained at these speeds typically contain a mixture of $\alpha$-, $\beta$-, and $\gamma$-phases. The relatively rapid solvent evaporation forms the kinetically favored $\alpha$-phase. This rapid evaporation prevents the complete alignment of the polymer chains into the $\beta$- and $\gamma$-phases.
  • Balanced Properties: These films exhibit balanced properties but lower electroactivity compared to films formed at slower speeds.
• Fast Withdrawal Speeds:
  • Thick Films: A very fast solvent evaporation rate occurs as the high withdrawal speeds lead to significantly thicker films [25].
  • Predominantly $\alpha$-phase: The quick solvent loss favors the formation of the thermodynamically stable but non-polar $\alpha$-phase while hindering the alignment of PVDF chains into the $\beta$- and $\gamma$-phases. The high rate of solvent evaporation produces more amorphous but less ordered structures.
  • Reduced Electroactivity: The films are mechanically robust, but there is the predominance of the $\alpha$-phase results.

Transversal forces (perpendicular to the direction of withdrawal) can significantly influence the film formation and phase composition during the dip-coating process. These forces, acting on the polymer solution, arise from fluid dynamics, especially drag and shear forces.

Transversal forces promote the formation of the $\beta$- and $\gamma$-phases by aiding in the alignment of the polymer chains in the direction of the applied force.

The strain-induced alignment of polymer chains, which facilitates the transition to more ordered and electroactive phases, is created by transversal forces [39]. This mechanical stress helps the phase transformation from the $\alpha$-phase to the $\beta$- or $\gamma$-phases.

Explanation for High Electroactive Phase Content in Thin Films:

• Solvent Evaporation Kinetics: Slow withdrawal speeds where the solvent evaporation occurs gradually allow PVDF chains to attain the necessary planar zigzag conformation for the $\beta$-phase and the distinct packing associated with the $\gamma$-phase.
• Chain Mobility: Sufficient mobility for the polymer chains to orient and crystallize is provided by the extended drying time, leading to the $\beta$- and $\gamma$-phases’ formation.
• Surface Effects: Thin films have a higher surface-to-volume ratio, and due to the preferential alignment of PVDF chains near the interface, the surface interactions with the substrate can promote the $\beta$- and $\gamma$-phase formation.
• Transversal Forces: By promoting chain alignment and phase transformation, the application of transversal forces during slow withdrawal can further enhance the formation of electroactive phases.

Figure 9a shows the XRD spectrum of four different PVDF films at 5% and at 10% mass concentrations. For the 10% PVDF films, the signal is more intense, and the increase in signal intensity is due to the greater thickness of the film at a PVDF concentration of 10% compared to 5%. This is explained by the fact that thicker films lead to greater absorption and scattering effects, resulting in the greater intensity of the measured signal. After analyzing these spectra, we identify the presence of the Alpha and Gamma phases, represented by strong sharp peaks at 17.9° and 20.2°, respectively; meanwhile, there is little to no presence of the Beta phase. Based on FTIR analysis, the Alpha and Gamma phases were also the dominant ones for those same films. It is understandable that the Beta phase cannot be identified in this XRD study because the 2Theta(°) at which we usually identify the Beta phase is really close to the ones representing the two other phases. Thus, the three peaks could theoretically convolute, making a clear identification of each phase independently somewhat difficult.

We also utilized a third characterization technique to further examine our films and to guarantee that our reasoning and conclusions were, in fact, accurate. This technique is Raman shift spectroscopy [40]. In Figure 9, we analyze a PVDF film annealed at 80 °C, and as can be seen, it is Alpha-dominated, indicated by the sharp peak at 794 cm⁻¹, alongside a minimal presence of the other two phases. This result aligns completely with the one obtained by FTIR [29].

Meanwhile the second film’s analysis showed that it had no heat treatment after deposition. Based on the Raman shift, there was a coexistence of both the Beta and Gamma phases, which also aligns with the FTIR results (which indicated a Gamma-phase dominance alongside a secondary presence of the Beta phase).

3.2. Imaging and Topography Analysis

The microscopic images shown in Figure 10 illustrate the evolution of dip-coated 5% PVDF thin films as a function of withdrawal speeds. For speeds below 0.1 mm/s, we have an integrity collapse, as the films become porous and even shredded. When the speed reaches 0.1 mm/s, we begin to have a homogeneous film with a granular surface, and as we increase the speed, the formation of spherulite shapes starts to appear. As will be explained further in the following sections, spherulite shapes usually indicate the Alpha phase [41]. Meanwhile, a granular surface is indicative of either the Beta or Gamma phase or a combination of both [6]. The study of these images aligns with the results obtained for the electroactive phases of 5% PVDF dip-coated films.

The microscopic images in Figure 10 show the evolution of dip-coated 10% PVDF thin films as a function of withdrawal speeds. In this case, we can see that the films’ surfaces are homogeneous, and the integrity is preserved throughout the withdrawal speeds. Also, all surfaces are granular, indicating a high electroactive fraction. There is no presence of the spherulite shapes.

These results are consistent with the previous outcomes, pointing to a dominance of the electroactive phases, especially for low speeds (0.5 mm/s and below).

At a polarized light mode (Figure 11), we can clearly see and better identify the spherulite shapes of the 5% PVDF thin films, further proving the dominance of the Alpha phase in this case. Spherulites in the Alpha phase of PVDF can be easily observed under polarized light microscopy due to their distinctive radial birefringence patterns [42].

The spherulites have well-defined borders separating each one of them, and this is further noticeable in the AFM and SEM images.

Spherulites indicating the Alpha phase presence form through crystallization processes, where the polymer nucleates and grows in a radial pattern from a central point.

• Crystallization Behavior: The $\alpha$-phase of PVDF crystallizes more easily compared to other phases during the cooling process from the melt. This leads to the formation of spherulites, as the phase tends to nucleate and grow in a radial pattern [43].
• Molecular Conformation: The $\alpha$-phase consists of a non-polar TGTG’ (trans-gauche-trans-gauche’) molecular conformation. The radial growth necessary for spherulite formation is facilitated by this specific arrangement.
• Thermodynamic Stability: At lower temperatures, the $\alpha$-phase is more thermodynamically stable than other phases, making it more likely to form and grow into spherulites during the cooling process [44].

The electroactive phases, Beta and Gamma, are found in structures different than that of Alpha. More specifically, the Beta phase forms highly oriented fibrous and lamellar structures due to the chain alignment from stretching, while the Gamma phase exhibits complex morphologies, including less uniform lamellar and fibrillar structures, often with intermixed phases.

By observing the AFM images in Figure 12, we can see that at 0.033 mm/s (Figure 12a), we have more granular and fibrous shapes, indicating a high presence of the electroactive phases. At 0.125 mm/s (Figure 12b), we begin to observe the spherulite structures alongside the fibrous shapes, indicating that a transition to the Alpha phase has begun. At 0.17 mm/s (Figure 12c), we see a full transition to spherulite structures, indicating the dominance of the Alpha phase. The same trend continues up to 2 mm/s (Figure 12e).

This analysis fully aligns with the results of the characterization methods mentioned previously in this chapter and the percentages of the electroactive phases obtained at each speed.

In Figure 13, we can see AFM images of 10% PVDF films withdrawn at different speeds: at 0.033 mm/s (Figure 13a), we have a granular/lamellar structure, indicating a high electroactive fraction. At 0.05 mm/s (Figure 13b), we begin to see spherulite structures, and as the topography becomes increasingly dominated by these spherulites, an increase in the Alpha phase can be observed. This is consistent with previous results, indicating a minimum electroactive fraction at this speed. Increasing the speed to 0.5 mm/s (Figure 13d), the fibrous shapes re-dominate, indicating a transition to the electroactive phases. This tendency continues with the increasing withdrawal speed, reaching up to 2 mm/s (Figure 13e). These trends and the overall analysis fully support the results obtained previously using the different characterization techniques.

Figure 14 presents SEM images of 5% PVDF thin films withdrawn at different speeds (the first set of three figures have a double magnification in comparison with the second set, and the scales are at 10 microns and 20 microns, respectively). In these images, one can see the evolution of the topography function of the dominant phase. For both cases, the first image is Alpha-dominant (Figure 14a), indicated by the spherulite shapes, while the second image (Figure 14b) is a film in the midst of a transition from the Alpha phase to the electroactive phases, indicated by the mixed topography between spherulite and granular shapes. The third image (Figure 14c) is a full transition to the electroactive phases, with a clear granular topography, noticeably different from the signature spherulite shapes.

3.3. Relative Humidity Percentage Control

We aimed to study the impact of controlling the humidity inside the dipping chamber on the PVDF films, especially on the different phases’ fractions and on the topography [45]. To this end, we conducted multiple tests on films withdrawn at various speeds and controlled the humidity percentage inside the chamber, adjusting it for each test. Figure 15 represents the results obtained for the electroactive fraction as a function of the 10% PVDF films’ thickness and the different controlled humidity percentages inside the chamber.

At ambient humidity, there is a fluctuation of the electroactive phase percentage due to the humidity variation inside the room. Humidity can vary based on the time of day, weather conditions, and the number of people in the room, leading to fluctuations. These fluctuations affect the solvent evaporation rate and the mobility of PVDF chains.

Optimized results can be obtained at 45% and 50% humidity, with electroactive fractions exceeding the 90% threshold consistently, for all thicknesses. For higher humidity percentages, at 60% and 70%, the results become unstable because at such high percentages, the withdrawn solution becomes too fluid and does not stick correctly on the silicon wafer, severely impacting both the topography and phase composition.

The 5% PVDF films exhibit the same behavior regarding humidity control, demonstrating similar trends in phase composition and thickness variation as observed for the 10% PVDF films.

To further detail the findings, we can see some FTIR spectra results for 10% PVDF films withdrawn at 40% humidity and 50% humidity (Figure 16b,c).

The electroactive phase is fully dominant in both cases, but we can see a specific increase in the Beta fraction with an increase in the humidity (vibration band at 1275 cm⁻¹).

On the other hand, we performed some tests at very low humidity percentages (10% and 20%) to study the impact on PVDF films. As can be deduced from Figure 16a, the electroactive fraction decreases sharply with decreasing humidity; meanwhile, there is a sharp increase in the non-polar Alpha phase.

It is important to note that the roughness of the films increases with increasing humidity. So, at high humidity percentages, we have rough surfaces, while at low percentages, the surface becomes quite smooth. Going back as to why the humidity percentage inside the chamber impacts the electroactive fractions obtained, the explanation is detailed as follows:

• High Humidity Conditions:

  –

  Slower Solvent Evaporation: Due to the increased presence of water vapor in the air, a high humidity level slows down the evaporation rate of the solvent. More controlled and gradual crystallization of PVDF is allowed.

  –

  Enhanced Chain Mobility: The time needed for the rearrangement into the planar zigzag conformation required for the formation of the electroactive $\beta$-phase and the distinct molecular packing needed for the $\gamma$-phase can be achieved through the slower evaporation rate [46].

  –

  High $\beta$- and $\gamma$-phase Content: The formation of $\beta$ and $\gamma$ phases is favored by the increased chain mobility and prolonged crystallization time under high humidity levels.

• Low Humidity Conditions:

  –

  Rapid Solvent Evaporation: The time available for PVDF chain mobility and alignment is limited due to the low humidity level (reduced presence of water vapor), which leads to faster evaporation of the solvent.

  –

  Reduced Chain Mobility: The chains are locked in a more disordered state due to the quick solvent loss, which hinders their movement, preventing them from achieving the necessary conformation for $\beta$- and $\gamma$-phase formation [47].

  –

  Dominance of $\alpha$-phase: The $\alpha$-phase is kinetically favored when there is insufficient time for the chains to rearrange, favoring the formation of this thermodynamically stable phase [45].

The AFM images in Figure 17 show the evolution of PVDF films’ topography as a function of controlled humidity percentage. In the first set, we have 5% PVDF films withdrawn at 0.1 mm/s. At 40%H, the topography is granular, indicating a high percentage of the electroactive phase; this topography gradually transforms into the spherulite structure, indicating a transition to the Alpha phase at 20%H. The topography becomes fully spherulitic at 10%H, signifying the dominance of the Alpha phase at this low humidity level.

Meanwhile, the 10% PVDF films were withdrawn at 0.025 mm/s. Similar to 5% PVDF, at 40%H, the topography was highly granular, indicating the dominance of the electroactive phases. At 20%H, we had the apparition of spherulites, and this presence further increased at 10%H, indicating an increase in the Alpha phase. The dominance of Alpha for 10% PVDF was less than in the case of 5% PVDF films.

These results align with the analysis obtained by the different characterization techniques, also pointing to a transition to the Alpha phase when the humidity is controlled at a low level. The root mean square (RMS) was measured for each film to quantify the roughness of the surface. We saw a decrease in the RMS with a decrease in the controlled humidity percentage, reaching 14 nm for 5% PVDF and 80 nm for 10% PVDF films.

Figure 18 represents SEM images of 5% PVDF films, withdrawn at 0.1 mm/s, at different controlled humidity percentages. Similar to the AFM images, we can clearly see the evolution of the topography function of the humidity percentage.

At 40%H, we have the granular, rough surface, indicating a high electroactive fraction. At 20%H, we have a mixture of granular and fibrous shapes, indicating a coexistence of the different phases, alongside a smoother surface.

At 10%H, we have the signature spherulite structures, indicating a dominance of the Alpha phase, alongside a very smooth surface with an RMS of 14 nm.

3.4. Nanoparticles Addition

We wanted to test the impact of adding different types of NPs to dip-coated PVDF films in the hopes of increasing the electroactive fraction, especially the Beta phase.

To this end, in this section, we examine the results of adding multiple NPs to the solution, and we conclude on which one is best suited for our objectives.

Figure 19 shows the impact of each NP’s addition on the electroactive fraction of dip-coated PVDF films as a function of the film’s thickness for measurements conducted at ambient humidity without humidity control. As can be deduced, firstly, the films’ thicknesses increase with the addition of NPs, which is an expected outcome. More importantly, concerning the electroactive phase, we can see that the results obtained with the addition of BaTiO₃ [31] and LiNbO₃ [33] are quite similar to the results found with pure PVDF films; meanwhile, a slight decrease in the electroactive fraction can be found in the case of LiNbO₃. On the other hand, with the addition of the magnetic YIG NPs [35], the electroactive fraction is persistently high (>90%) for all thicknesses. To this end, we continued our studies for dip-coated films using these magnetic nanoparticles.

Figure 19. The evolution of the electroactive fraction in dip-coated 10% PVDF films as a function of film thickness and the addition of different nanoparticles (5% mass) without humidity control. The numbers next to the symbols and in the same color correspond to the withdrawal speed (in mm/s).

Figure 19. The evolution of the electroactive fraction in dip-coated 10% PVDF films as a function of film thickness and the addition of different nanoparticles (5% mass) without humidity control. The numbers next to the symbols and in the same color correspond to the withdrawal speed (in mm/s).

Mechanisms of Phase Transformation Induced by YIG Nanoparticles:

• Nucleation and Crystallization Dynamics:
  • Heterogeneous Nucleation: Due to their surface energy and interaction with PVDF, YIG nanoparticles act as heterogeneous nucleation sites [36]. This favors the formation of the $\beta$ phase by lowering the energy barrier for nucleation and, thus, providing a template for the alignment of PVDF chains.
  • Enhanced Nucleation Density: The number of nucleation sites is increased by the high surface area of well-dispersed YIG NPs, promoting finer and more uniform crystalline structures, which facilitates the formation of the $\beta$ phase.
• Interfacial Interactions and Magnetic Influence:
  • Dipole-Magnetic Interactions: The magnetic properties influence the dipolar interactions within the PVDF matrix, and these interactions can enhance the alignment of polymer chains into the higher dipolar nature phase, the $\beta$ phase.
  • Electrostatic and Magnetic Effects: The $\beta$ phase is promoted by the presence of magnetic fields and the intrinsic properties of YIG, which can induce local magnetic fields affecting the molecular orientation of PVDF chains.
• Mechanical Stress and Morphological Effects:
  • Stress-Induced Phase Transformation: The inclusion of YIG nanoparticles introduces mechanical stress within the PVDF matrix. This stress can facilitate the phase transformation from the $\alpha$ phase to the $\beta$ phase by aligning the polymer chains under the influence of internal stress fields.
  • Nanoparticle Dispersion: Uniform dispersion of YIG nanoparticles ensures a consistent stress distribution and maximizes the impact on phase transformation, promoting the formation of the $\beta$ phase.
• Thermal and Dielectric Properties:
  • Enhanced Thermal Stability: By increasing a PVDF film’s thermal stability, YIG nanoparticles may have an impact on the phase stability and transformation kinetics. A higher $\beta$-phase content can result from the presence of YIG due to the increased thermal energy that promotes chain mobility and reorganization.
  • Dielectric Enhancement: The $\beta$ phase, which has a higher dielectric constant compared to the $\alpha$ phase, is stabilized by the enhanced dielectric properties of the composite material by the presence of YIG nanoparticles.

Based on our results, we deduce that the addition of YIG nanoparticles primarily increases the overall electroactive fraction and significantly enhances the $\beta$ phase.

Figure 20 shows that the addition of 5%YIG to PVDF dip-coated films not only increases the electroactive phases fraction, but more specifically enhances the Beta phase, as can be seen in the 1275 cm⁻¹ vibration band.

All of these tests were performed at ambient humidity, so the surface had a relatively high roughness. In this context, we wanted to perform the same tests but with a controlled humidity at a very low level, aiming to obtain a smooth surface and to figure out if the addition of magnetic NPs can eliminate the negative effects of low humidity levels on the electroactive phases.

Figure 21 presents SEM images of 10% PVDF films withdrawn at 0.02 mm/s, with the addition of 5%YIG NPs. As can be seen, we have a rough granular structure representing the PVDF matrix, and small bright spots spread throughout the surface, indicating the YIG NPs’ presence.

3.5. Addition of YIG Combined with a Low Humidity % Control

In the following section, we detail the results obtained by combining the addition of YIG NPs, alongside a control of the humidity at ≈10%. Both the surface topography and phase composition are examined, with modifications made throughout the process to achieve optimal results.

Table 3. Electroactive fraction of doped 5% PVDF and 10% PVDF films as a function of withdrawal speed and humidity control.

	$\mathit{F}(\mathit{\beta }+\mathit{\gamma })$ (%)
Dip-Coating Solution	Low Speed	Medium Speed	High Speed
PVDF 5\% + YIG \5%	98	97	92
PVDF 5\% + YIG \5% and H10\%	97	96	90
PVDF 10\% + YIG \5%	97	94	89
PVDF 10\% + YIG \5% and H10\%	95	91	89

shows the results obtained for both 5% PVDF and 10% PVDF. It is evident that we obtained nearly identical results (for the two PVDF concentrations), both with and without humidity control.

The humidity at 10% previously caused a significant decrease in the electroactive phases for pure PVDF films. Now, this unwanted effect was eliminated with the addition of 5%YIG NPs, proven by the Beta + Gamma-dominated films obtained. We further delved into studying this outcome using different characterization techniques (focusing on the film composition) and different imaging techniques (focusing on the topography and surface roughness).

From Figure 22a, showing the FTIR spectra of different 5% PVDF films withdrawn at multiple speeds, it can be deduced that the Beta phase is the dominant one in all of the cases (sharp vibration band at 1275 cm⁻¹), both with controlled humidity and non-controlled humidity.

We analyzed the same kind of films (10% PVDF+5%YIG) using Raman shift spectroscopy (Figure 22b,c). The first film was deposited at ambient humidity and the second one at 10% humidity.

Identical results for the two films were obtained, showing a sharp Beta + Gamma dominance, and more importantly, an emphasis on the strong Beta-phase presence. Meanwhile, little-to-no Alpha presence was detected.

Finally, within the domain of characterization techniques, we performed a detailed XRD study on various types of PVDF films. These ranged from Alpha-dominated to Gamma-dominated films, culminating in Beta-dominated films being obtained in the end at 10% humidity with the addition of 5% YIG NPs.

Figure 23a is an XRD spectrum of an Alpha-dominated, pure PVDF film, alongside a small presence of the Gamma phase. This film was dip-coated at normal conditions, and we used, in this case, a withdrawal speed that usually favors the formation of the Alpha phase.

The XRD results confirm the dominance of the Alpha phase, indicated by the sharp peak at 17.9°. A secondary presence of the Gamma phase is also observed, highlighted by the peak at 20.1°, which represents both the Alpha and Gamma phases, as it falls between their characteristic angles.

Figure 23b is an XRD of a Gamma-dominated pure PVDF thin film; it was dip-coated at normal conditions, with the withdrawal speed specifically chosen to achieve a Gamma-phase majority.

This Figure 23b clearly shows the dominance of the gamma phase, indicated by the signature peak at 20.3°. However, the peak is quite broad, suggesting it may represent all three phases, as their signature angles are very close to each other. The absence of the peak at 17.9° suggests a near absence of the Alpha phase. While the Beta phase may be present, its presence cannot be confirmed with certainty. Finally, Figure 23c shows XRD spectra of two PVDF films at 10% humidity, doped with 5%YIG NPs, and withdrawn at two different speeds.

We have a stronger signal for the thick film (≈1 micron) in comparison with the thin one (≈300 nm), which is normal. In both cases, we see a dominance of the Beta phase (signature peak at 20.6°), alongside a smaller presence of the Gamma phase. We face the same problem mentioned before concerning the peaks’ convolution, but the overall tendencies are clear.

Overall, these results align with those obtained from Raman shift and FTIR analyses, indicating that we successfully optimized the films. The final films feature a smooth surface with low roughness and a dominant Beta-phase fraction, offering strong piezoelectric properties.

Figure 24a,b present AFM images of dip-coated 5% PVDF films doped with 5%YIG; the first was deposited at ambient humidity, while the second was deposited at 10% controlled humidity. In both images, the granular topography is dominant, indicating a high fraction of the electroactive phases. The roughness of the film decreases significantly from an RMS of 81 nm to 23 nm for the controlled humidity case. Defects and heterogeneous grains can be detected (bright white spots), caused by agglomerations of the YIG NPs.

Furthermore, Figure 24c,d show the AFM results for dip-coated 10% PVDF films doped with 5%YIG, the first deposited at ambient humidity and the second deposited at 10% humidity.

In this case, the granular topography is clearly dominant, with the film deposited at ambient humidity showing a very high RMS of 112 nm. This RMS decreases sharply to 48 nm under controlled humidity conditions, further proving the positive impact of humidity control. Similar to the 5% PVDF case, defects with varying grain sizes are detected, which are caused by the non-ideal distribution of YIG nanoparticles.

From both cases, we can deduce two key points: firstly, the dominance of electroactive phases, and secondly, the enhanced and smoother surfaces achieved by maintaining low humidity levels. The primary challenge remains the difficulty in homogeneously breaking and spreading the YIG nanoparticles in the solution and during film deposition. Addressing this issue could eliminate the defects observed in the figures discussed.

The combination of humidity control along with the addition of MNPs proved to be a great approach to be adopted, as it improved the piezo properties and topography of the PVDF films.

4. Conclusions

This work systematically dissects the comprehensive results achieved through innovative deposition techniques and a meticulous optimization of piezoelectric polyvinylidene fluoride (PVDF) films.

As we delved into the dip-coating method, we ventured into relatively uncharted territories of film fabrication. Dip-coating, seldom utilized in piezoelectric research for PVDF, revealed its potential to significantly improve film homogeneity and surface smoothness compared to the traditional methods. This technique’s ability to precisely control film characteristics by adjusting withdrawal speeds and environmental conditions marked a pivotal advancement in our experimental repertoire. In our exploration, the manipulation of RH% emerged as a crucial factor, profoundly influencing the phase transitions and stability within the films. By optimizing humidity levels, we were able to consistently achieve higher electroactive phase contents (at ≈45% to ≈50%), as well as obtain a smoother and more homogeneous surface of the film using low levels of humidity, demonstrating the sensitive interplay between environmental conditions and material properties. At low levels of humidity, the topography was optimal, but the electroactive fraction decreased significantly.

Moreover, the strategic addition of various nanoparticles tailored the piezoelectric properties of the films. These additions were pivotal in promoting desirable phase compositions and enhancing the films’ overall piezoelectric responses. YIG nanoparticles, in particular, significantly increased the $\beta$-phase content, aligning with our objectives to maximize the material’s piezoelectric capabilities.

Finally, the combination of low-humidity control along with the addition of YIG NPs proved to be the optimal approach, as we obtained a PVDF layer with a very high $\beta$-phase percentage, along with a smooth film surface having minimal roughness. Characterization techniques such as FTIR, Raman shift, and XRD provided deep insights into the microstructural changes and phase dynamics within the films. These methodologies confirmed the effectiveness of our experimental modifications and underscored the complex relationships between processing conditions and the resulting material properties.

As we conclude this work, we not only reflect on the successful application of novel techniques and environmental controls but also pave the way for future investigations aimed at harnessing and further enhancing the piezoelectric properties of PVDF films. The groundwork laid here is instrumental in advancing towards new areas of research, where these optimized layers can be integrated into more complex systems.