Conducting EVA/GNP Composite Films with Multifunctional Applications: Effect of the Phosphonium-Based Ionic Liquid

Abstract

The application of graphene nanoplatelets (GNPs) in polymer composites is a challenge due to their high tendency to agglomerate and restack during processing. In this work, alkyl phosphonium-based ionic liquid was used to assist the dispersion of GNP in an ethylene-vinyl acetate (EVA) matrix, through a melt-mixing procedure. The mechanical properties and creep resistance of the films prepared by the film extrusion process were evaluated. The results demonstrated that the noncovalent treatment of GNP with the ionic liquid (IL) enhanced the electrical conductivity and creep stability of the EVA composites. The microwave absorbing properties were studied in the X-band and Ku-band. A reflection loss (RL) of −15 dB for EVA containing 0.5 wt% of GNP and 1:1 wt% of GNP/IL was achieved. The use of a multi-layered structure containing thin film layers was efficient for enhancing the microwave absorbing performance, with a minimum RL of −24.6 dB and effective absorption bandwidth of 4.3 GHz. This result is attributed to the internal reflection and scattering of the radiation between layers. The use of simple, low-cost materials and procedures, combined with the system’s excellent mechanical and electrical properties, makes it a promising candidate for multifunctional applications as electrostatic dissipative and microwave absorbing materials for electronic packaging and other electronic devices.

1. Introduction

Ethylene–vinyl acetate (EVA) copolymers are a versatile, commercially available and cost-effective polymer. Their characteristics range from thermoplastic to elastomer, depending on the vinyl acetate (VA) content. Due to their excellent processability, flexibility, low density, and reasonable mechanical properties, EVA copolymers are widely used in various sectors, including electrical cables, footwear, hot-melt adhesives, and packaging. However, their insulating nature limits applications such as electrostatic dissipating packaging films for the protection of electronic equipment, and electromagnetic interference (EMI) shielding materials. The incorporation of conducting fillers, primarily carbonaceous fillers, can enhance the electrical and thermal conductivity of EVA films, broadening their applications. Graphene and its derivative, such as exfoliated graphite (EG), graphene nanoplatelets (GNP), graphene oxide (GO), and reduced graphene oxide (RGO), have been extensively studied for polymer composites due to their superior mechanical, electrical, and thermal properties [1,2]. Among these, GNPs have gained popularity because they are commercially available, with a relatively low cost compared to carbon nanotube, and suitable for large-scale production. For instance, Sabet et al. [3] reported enhancements in thermal stability, tensile properties and modulus with the addition of 1–3% of GNP in an EVA matrix using the melt-mixing procedure. A similar process was employed to prepare EVA composites loaded with thermally reduced graphite oxide (TrGO), achieving improved thermal and electrical conductivity, as well as modulus, compared to conductive graphite, particularly in EVA containing 18% of VA [4,5]. Azizi et al. [6] utilized the solution casting approach, achieving an electrical conductivity of around 10⁻⁷ S/cm with the addition of 15% GNP in EVA copolymer with 28% VA. Similarly, Soheilmoghaddam et al. [7] prepared composite films by solvent casting EVA (18% VA) and 3% GNP, resulting in improved thermal stability, mechanical properties, and modulus. The differences in these results can be attributed to variations in the number of layers, average thickness, and lateral dimensions of the graphene materials supplied by different manufacturers, as reported in previous studies [8,9,10].

Despite their benefits, GNPs tend to agglomerate in polymer matrices due to strong van der Waals and π–π interactions between the nanosheets. To mitigate these issues, the surface modification of graphene and its derivatives has been extensively studied. While covalent functionalization is well documented, it often involves harsh chemistry with strong acids [11], and introduces defects that can negatively impact electrical properties. Therefore, non-covalent functionalization via physical interactions with specific surfactants has emerged as a promising alternative, preserving the conjugated structure of the graphene surface [12]. In this context, ionic liquids (ILs) have been considered effective additives for dispersing GNPs through van der Waals, π–π, and cation–π interactions, in analogy to that already documented in the case of carbon nanotube/IL [13,14,15]. For example, RGO modified with imidazolium-based IL has shown improved mechanical properties in epoxy network [16]. RGO modified with triphenylphosphonium-based IL and dispersed in a poly(vinylidene fluoride) (PVDF) solution resulted in cast films with enhanced electric and dielectric properties [17]. Additionally, Xu et al. [18] dispersed commercial GNP with imidazolium-based IL and mixed them with PVDF in solution, achieving better dielectric properties. Joseph and Francis [19] reported the modification of GNP with imidazolium IL to improve the mechanical properties of styrene–butadiene rubber (SBR). Furthermore, polyimide matrices modified with amino imidazolium-based IL and GNP exhibited outstanding electromagnetic wave absorption properties [20].

Over a decade ago, GO was modified with octadecylamine followed by reduction, and mixed with EVA using the solvent casting approach, leading to an increased tensile strength and thermal stability, as well as decreased electrical resistivity [21].

Conducting polymer composite films have garnered significant interest due to the demand for flexible and smart materials in wearable electrical devices [22,23,24], stretchable electronics [25], flexible capacitors [26], sensors [27], and other flexible devices [28]. Most studies employ solution casting or the spin-coating method to prepare these films. However, EVA/GNP composite films prepared by melt mixing and the effect of phosphonium-based IL on their properties have not been extensively explored.

Therefore, the motivation and novelty of this work lie in preparing EVA/GNP composite films and evaluating the influence of IL on the dispersion of GNP in the EVA matrix. The melt-mixing procedure was chosen for preparing the mixture and the films due to its scalability and low-cost process. The mechanical, dynamic mechanical, and creep properties were investigated as a function of GNP content. The non-covalent functionalization of GNP by IL through a solvent-free grinding process was also chosen to facilitate environmentally friendly scalability. Finally, this study highlights the development of a cost-effective processing method with commercially available and low-cost materials to develop flexible composites with outstanding microwave absorbing properties, particularly in the Ku-band frequency range, which is valuable for both civilian and military applications.

2. Materials and Methods

2.1. Materials

Ethylene–vinyl acetate copolymer (EVA) with 28% vinyl acetate content (trade name = EVA HM728F, melt flow index = 6.0 g/10 min at 190 °C/2.16 kg; and density = 0.95 g cm⁻³) was kindly supplied by Braskem, Rio Grande do Sul, Brazil. Graphene nanoplatelet (GNP) (commercial name = xGNP^® grade C750; particle size < 2 μm; surface area = 750 m²·g⁻¹; and relative gravity: 2.0–2.25 g·cm⁻³) was purchased from Sigma Aldrich, São Paulo, Brazil. Ethyl-(tributyl) phosphonium diethyl phosphate as the ionic liquid (trade name = CYPHOS^® IL 169) was kindly donated by Solvay, São Paulo, Brazil.

2.2. Composites Preparation

EVA was dried overnight at 50 °C before blending to avoid moisture. The composites were prepared by melt blending using an Xplore MC15 HT conical twin-screw micro extruder (Xplore Instruments BV, Sittard, The Netherlands) with co-rotating screws and an internal volume of 15 mL. For the samples containing ionic liquid (IL), a 1:1 weight ratio of GNP to IL 169 was used. The GNP and IL were manually ground using a mortar and pestle for 10 min to ensure proper mixing. The resulting mixture was then added to the EVA and directly fed into the extruder. The blend was processed at 160 °C and 200 rpm for 3 min with the extruder in a closed configuration. The films were prepared directly from the extruder using the micro cast film line, according to Figure 1.

2.3. Characterization

An electrical conductivity analysis was carried out in an alternating current (AC) mode, with the assistance of a potentiostat VersaSTAT 3 from Ametek Scientific Instruments (Oak Ridge, TN, USA) and a circular sample holder. The conductivity was analyzed from 10⁻² to 10⁵ Hz at 0.1 V, using circular samples with a diameter of 25 mm and thickness of 1 mm. The compression-molded samples were covered with a thin layer of gold to improve the contact with the electrodes.

The rheological behavior of the materials was analyzed with a Discovery DHR1 rheometer from TA Instrument, Inc. (New Castle, DE, USA). The analysis was carried out under the oscillatory mode, using parallel plates with a 25 mm diameter and gap of 1.0 mm, at a temperature of 160 °C, with a frequency sweep from 0.1 to 100 Hz and 0.1% strain.

Mechanical properties were analyzed using an Autograph AGS-X tensile tester from Shimadzu (Tokyo, Japan). The tests were carried out, using thin films with a thickness of 0.2 mm and width of 22 mm, at 23 °C with a crosshead speed of 500 mm/min.

Creep behavior was evaluated in a DMA Q800 from TA Instruments (New Castle, DE, USA), in the film mode. Thin films with a thickness of 0.2 mm and width of 6 mm were tested at a constant tension of 0.1 MPa for 60 min, followed by a recuperation time of 30 min.

Electromagnetic properties were carried out with a VNA Vector Network Analyzer from Keysight Technologies (Santa Rosa, CA, USA), model E5080B, using a rectangular wave guide in the X-band (8.2–12.4 GHz) and the Ku-band (12.4–18 GHz) microwave range. The scattering parameters, S₁₁ and S₂₁, were acquired to calculate the complex relative permittivity (ε_r = ε′ − jε″) and permeability (μ_r = μ′ − jμ″). A metal-backed configuration, consisting of fixing a metal plate on the back end of the measured sample, was used to obtain the reflection loss (RL) of the composites.

3. Results and Discussion

3.1. Electrical Conductivity

The influence of GNP content and the incorporation of IL on the electrical conductivity of EVA-based composites is depicted in Figure 2. Composites containing only GNP exhibited typical insulating behavior (Figure 2a), as evidenced by a linear relationship between σ_AC and frequency, even at a GNP concentration as high as 10 wt%. However, the addition of IL significantly enhanced the σ_AC of the system. At lower frequencies, the conductivity increased progressively with the GNP content. Moreover, all the GNP/IL-containing samples displayed a distinct DC conductivity plateau at low frequencies. This plateau shifted toward higher frequencies with increasing GNP and IL concentration, indicating a clear transition from insulating to conductive behavior. This feature is attributed to the interaction between IL molecules and the surface of GNP, particularly through cation–π and van der Waals forces, which reduce the restacking and agglomeration of GNP platelets, thereby improving their dispersion within the polymer matrix. As a result, a conductive network can be achieved with a lower filler content. Notably, the composite containing 10 wt% GNP and 10 wt% IL exhibited a low-frequency conductivity (at 0.1 Hz) of 2.4 × 10⁻⁷ S/m, with the plateau extending up to 50 Hz.

3.2. Rheological Properties

The effect of GNP and IL incorporation on the rheological behavior of EVA composites are presented in Figure 3, in terms of complex viscosity versus frequency. The addition of a small amount of GNP (0.5 wt%) led to an increase in viscosity, which is indicative of improved filler dispersion within the matrix. Interestingly, increasing the GNP content to 1 and 2 wt% caused a decrease in viscosity. In particular, the 2 wt% GNP composite exhibited lower viscosity than neat EVA, likely due to the lubricating action of GNP. This trend is consistent with findings by Genoyer et al. [29], who observed a similar viscosity reduction in polystyrene (PS) composites containing 2 wt% GNP. They attributed this to a processing aid effect, where GNPs act as lubricants depending on the type, dispersion quality, and aggregate size of GNPs. Lima et al. [30] reported a viscosity drop in polypropylene (PP)/GNP composites with 1 wt% filler, also ascribed to the interlayer structure of GNPs imparting a lubricant effect.

When the GNP concentration was increased to 5 wt% and 10 wt%, the formation of a physical network began to dominate, outweighing the lubricating effect and resulting in a viscosity increase. The 10 wt% GNP composite demonstrated the highest viscosity among all the samples.

The inclusion of IL in the composites further influenced the viscosity. The GNP/IL (0.5:0.5 wt%) composite showed a viscosity similar to or slightly higher than that of neat EVA. As the GNP/IL content increased, the viscosity initially decreased due to the combined lubricating effects of both components. However, beyond a concentration of GNP:IL = 2:2 wt%, the viscosity began to rise again, driven by the increasing of the filler and the onset of a physical filler network. An interesting observation was that the composite with GNP/IL = 0.5/0.5 wt% had a higher viscosity than that with 5:5 wt%. This suggests the superior dispersion of the filler in the low-GNP system, resulting in a larger polymer-filler interfacial area, which increased flow resistance and offset the lubricating effects. This enhanced interface likely contributed to the higher viscosity observed at the lower filler loading.

The correlation between GNP concentration and viscoelastic behavior, as well as the structural changes in the filler-loaded polymeric matrices, are better estimated from the Han Plots [31], as shown in Figure 4a,b for the composites without and with IL. The diagonal line in the graph represents the points where G′ = G″, and it is known as the equimodular line. Above it, the elastic component dominates the material’s behavior, while below it the viscous mechanisms have a greater influence on the sample’s properties. All the curves are linear in nature, indicating no phase separation [32]. The curve of neat EVA as well as those corresponding to EVA/GNP with 0.5 and 2 wt% GNP stayed below the equimodular line, indicating that the mechanism is completely dominated by the viscous component. It is important to emphasize that the composites containing GNP modified with ionic liquid exhibited a viscous-like-to-solid-like transition, which occurred at slightly lower G′/G″ value as the amount of GNP increased up to 2 wt%. This behavior indicates an increased reinforcing action of the filler, caused by its improved dispersion assisted by the ionic liquid, and confirms the improved dispersion of GNP by the ionic liquid.

3.3. Morphology

The effect of the IL incorporation on the morphology of the EVA composite containing 5 wt% of GNP is demonstrated in the SEM images shown in Figure 5. As discussed previously with regard to electrical conductivity and rheological properties, the presence of IL significantly enhances the dispersion of GNP within the polymer matrix. This improvement is evident when comparing Figure 5a and Figure 5c, which depict the morphology of the composite without and with IL, respectively. In these images, two large GNP agglomerates are clearly visible (indicated by arrows), underscoring the inadequate dispersion of the filler in the absence of IL. In contrast, the composite containing the same GNP content (5 wt%) but modified with IL (Figure 5c) displays a much more improved dispersion. The GNPs are more uniformly distributed throughout the matrix, with only minor clustering observed (highlighted by the circle in the image), indicating a substantial reduction in agglomeration due to the presence of IL.

3.4. Mechanical Properties

Using GNP as filler can result in a higher mechanical performance, especially in stiffness and resistance, owing to the high elastic modulus and intrinsic strength assigned to graphene materials. In addition, the use of 2D particles has the potential for reinforcing polymer composites due to their large surface, leading to more efficient load-transfer points between the matrix and the filler [33]. Figure 6 presents the stress vs. strain curves of the EVA composite films, whose main data are also summarized in

Table 1. The main data obtained from the stress x strain curves of EVA composite.

EVA/GNP	Young’s Modulus (MPa)		Strain at Break (%)		Max Stress (MPa)
GNP (wt%)	Without IL	With IL	Without IL	With IL	Without IL	With IL
0	19.0 ± 0.8	- a	671 ± 55	- a	12.2 ± 1.5	- a
0.5	17.9 ± 1.1	16.2 ± 0.8	793 ± 35	922 ± 96	12.1 ± 0.7	12.9 ± 0.8
1	17.1 ± 1.1	14.8 ± 0.9	858 ± 55	1139 ± 182	12.2 ± 1.9	11.5 ± 2.3
2	20.0 ± 0.6	15.9 ± 0.7	741 ± 33	1074 ± 126	12.5 ± 0.5	13.6 ± 1.1
5	21.3 ± 0.5	16.3 ± 1.2	825 ± 166	676 ± 70	17.2 ± 2.4	14.1 ± 2.6
10	22.7 ± 0.9	14.8 ± 1.4	832 ± 121	853 ± 111	19.3 ± 0.9	13.5 ± 1.3

^a IL was not added to the sample with 0 wt% of GNP.

. The addition of 2 wt% of GNP or higher resulted in an increase in modulus and tensile strength, indicating the reinforcement behavior of the GNP in the composites. As indicated in Table 1, the maximum stress shows an increase of around 41% and 58% for the composite with 5 and 10 wt% of GNP, respectively, when compared to the neat EVA. Similarly, for the elastic modulus, an increase of 10 and 21% was observed. All the systems presented a higher elongation at break than the neat EVA.

The plasticizing effect of the IL is evident in the mechanical properties of the composites. As a general assessment, it is observed that all the samples with IL showed a higher total deformation and smaller Young’s modulus when compared with the samples with only GNP.

3.5. Creep Behavior

During the tensile creep testing, the polymeric material undergoes an instantaneous elastic deformation followed by a creep deformation [34]. Figure 7 illustrates the typical creep/recovery curves of EVA composites as a function of GNP and GNP/IL content, and

Table 2. Main creep data related to the EVA composites containing different amounts of GNP and GNP/IL (1:1).

GNP (wt%)	Strain at 3600 s (%)				Instant Strain Recovery (%)
	Without IL		With IL		Without IL	With IL
	Value	Variation (%)	Value	Variation (%)
EVA	2.04	- a	- a	- a	31
0.5	1.35	34	1.38	32	57	57
1	1.44	29	1.22	40	40	31
2	2.02	1	2.37	−16	33	25
5	1.26	38	1.07	48	34	41
10	1.20	35	1.20	41	36	39

^a IL was not added to the sample with 0 wt% of GNP.

summarizes the main results. Except for the EVA/GNP composite with 2 wt% GNP, all other composites exhibited a lower creep strain and higher instant creep recovery than neat EVA, indicating that the addition of a proper GNP content enhances the creep stability due to the reinforcing action of the filler. The GNP reduces the mobility of the polymer chains and acts as blocking sites [35]. The presence of 0.5 wt% of GNP reduced the creep strain by around 34% due to the good filler dispersion and the increase in the filler–matrix interface. However, increasing the amount of GNP to 2 w% increased the creep strain, probably due to a growth in filler agglomeration. Nevertheless, the composite with 5 wt% of GNP resulted in the best creep stability because the higher amount of GNP resulted in greater reinforcing action, surpassing the agglomeration effect. The addition of IL resulted in a better creep response for all systems except that containing GNP/IL = 2:2 wt%. This behavior may be attributed to the improved distribution of the filler within the EVA matrix, as previously discussed. A more uniform filler dispersion strengthens the interactions between the EVA chains and the GNP, promoting more effective load transfer and restricting molecular mobility. The highest strain recovery was observed for the composites with 0.5:0.5 wt% and 5:5 wt% of GNP/IL. The IL exerts a plasticizing effect on the polymer matrix and can increase the mobility of polymer chains, thus decreasing the creep stability. However, at the same time, IL can enhance the dispersion of GNP, improving filler–matrix interaction. Thus, the last phenomenon surpasses the plasticizing effect of IL for the system containing GNP/IL = 5:5 wt%.

The creep compliance illustrated in Figure 8 is another way to present the creep resistance of the EVA composites. For the system without IL, the highest creep compliance was observed for the neat EVA and the composite with 2 wt% of GNP, whereas the lower values were observed for the composites with higher amount of GNP (5 and 10 wt%). The presence of IL in the composites contributed to an increase in the creep compliance in most situations, which is attributed to the easiness of polymeric chain movement caused by the plasticizing effect of the IL. Nevertheless, the composite containing GNP/IL = 5:5 wt% exhibited the lowest creep compliance. As discussed earlier, several factors contribute to changes in creep resistance. As the GNP content increases, the tendency for agglomeration also rises, reducing the interfacial area and diminishing the filler’s influence on polymer chain mobility. However, GNPs can simultaneously act as reinforcing agents, which may restrict chain mobility. Additionally, a higher GNP content leads to an increased amount of IL, which can enhance polymer chain mobility and improve filler dispersion, thereby increasing the interfacial area and strengthening the filler–matrix interaction. The creep response of the composite containing GNP/IL = 5:5 wt% indicates that the filler dispersion promoted by the IL improves the reinforcing action of the filler and surpasses the plasticizing influence of the IL

The main parameters of the creep testing can be estimated using the classic Burges model, which can give a good description of the composite’s performance within the range of the first two, out of three, characteristic stages of the creep behavior [36]. Creep compliance (J = ε(t)/σ) from the Burger’s model can be calculated according to Equation (1):

$$J\left(t\right)={\displaystyle \frac{1}{E_1}}+{\displaystyle \frac{1}{E_2}}\left(1-e^{-{\displaystyle \frac{t{E}_2}{{\eta }_2}}}\right)+{\displaystyle \frac{t}{{\eta }_1}}$$

(1)

where E₁ is the instant component’s modulus of elasticity, E₂ is the modulus of elasticity of relaxation response, η₁ and η₂ are the coefficients of dynamic viscosity, and τ (=η₂/E₂) is the relaxation time. The indices 1 and 2 represent the Maxwell and Kelvin–Voigt elements from the model, respectively.

As indicated in Figure S1 of the Supporting Information, the experimental data were well fitted with the Burger’s model (R² = 0.99). Thus, the main creep parameters could be estimated from the model, whose results are summarized in

Table 3. Burger’s model parameter for tensile creep behavior of EVA composites.

GNP (wt%)	E1 (MPa)		E2 (MPa)		η1 (MPa∙s)		η2 (MPa∙s)		τ = η2/E2 (s)
	A	B	A	B	A	B	A	B	A	B
EVA	12.3 ± 0.2		13.6 ± 0.6		(70 ± 7) × 103		(45 ± 3) × 102		328.8
0.5	11.9 ± 0.1	12.0 ± 0.1	24.9 ± 0.7	36.7 ± 1.2	(276 ± 37) × 103	(125 ± 7) × 103	(19 ± 2) × 102	(52 ± 4) × 102	212.5	141.5
1	13.5 ± 0.2	11.4 ± 0.2	20.9 ± 0.6	13.5 ± 0.4	(132 ± 12) × 103	(102 ± 10) × 103	(25 ± 2) × 102	(15 ± 1) × 102	121.7	110.9
2	12.0 ± 0.2	9.5 ± 0.1	14.3 ± 0.5	12.4 ± 0.4	(65 ± 5) × 103	(58 ± 4) × 103	(19 ± 2) × 102	(15 ± 1) × 102	134.1	124.4
5	18.7 ± 0.3	18.8 ± 0.3	23.3 ± 0.6	30.4 ± 0.9	(110 ± 6) × 103	(140 ± 9) × 103	(25 ± 2) × 102	(20 ± 2) × 102	105.4	67.4
10	16.5 ± 0.3	13.7 ± 0.2	25.0 ± 0.8	27.5 ± 1.0	(134 ± 11) × 103	(222 ± 31) × 103	(29 ± 3) × 102	(35 ± 3) × 102	117.7	129.0

A = without ionic liquid; B = with ionic liquid.

. In accordance with the above-mentioned mechanical and rheological properties, the creep behavior of the composites follows a similar trend, where the composite with 2 wt% of GNP shows a higher ductility, which translates to a higher creep compliance.

Typically, η₁ can be compared to the composite’s viscosity and, as it was seen from the rheological analysis, the composite with 2 wt% of GNP (with and without IL) shows the smaller values, while those containing GNP 0.5 wt% and GNP/IL (10:10 wt%) show the highest. E₁ can be correlated to the Young’s modulus, and in general the obtained values show comparable behavior to those achieved by tensile testing. The modulus tends to increase with an increasing amount of GNP added.

Similar to what was seen for the material’s viscosity, the composites with the IL show more distinction between the creep compliance values, whereas the samples with only GNP are more comparable between each other. Additionally, in accordance to the rheological behavior, the composite with 2 wt% of GNP shows a higher creep compliance, which can be attributed to the easiness of polymeric chain movement caused by the GNP’s lubricant effect. The composite with GNP/IL = 5:5 wt% presented the lowest deformation at 3600 s, a good strain recovery, and the lowest retardation time, indicating a very good response of creep resistance.

3.6. Electromagnetic Absorption Properties

An important application of conducting polymeric composites is as microwave absorbing materials. This subject has become of great interest due to the rapid development of sophisticated communication equipment and wireless devices that work in the high frequency range. The CPCs offer the advantages of being lightweight, their ease processability, and the possibility of tunable conductivity and design [37,38,39,40]. Regarding EVA-based composite films, the addition of a conducting particle can extend the range of applications as electronic packaging for electronic devices. The electromagnetic absorption properties of the composites were evaluated through reflection loss (RL). Figure 9 evaluates the influence of the incorporation of GNP and IL in EVA matrix on the RL values in the X-band and Ku-band frequency range. The 2 mm thickness samples displayed a better EM attenuation response at the Ku-band range, characterized by a minimum RL lower than −10 dB, which means an EM attenuation higher than 90%. For the composites without IL, the best result was presented by the composite with 0.5 wt% GNP, with RL reaching −15 dB at 15.8 GHz. For the systems with IL, the lowest RL peak was obtained by the sample with 1 wt% of GNP, reaching values of −15.3 dB. It is important to point out that the RL parameter describes the attenuation of the EM by absorption. In these systems, increasing the GNP content enhances the likelihood of interactions between electromagnetic radiation and the free charge carriers of the conductive particles, thereby increasing the potential for EM attenuation through reflection. This hypothesis justifies the higher values of RL (lower attenuation by absorption).

Multi-layered structures usually provide better absorbing properties and a broad effective absorbing bandwidth (EAB) (with RL < −10 dB), due to the presence of several interfaces that increase the multi-reflection and scattering of the radiation inside the material [41,42]. Furthermore, the absorption properties can be enhanced by introducing a layer of low dielectric property in between the conductive layer, due to the gradual decrease in impedance and phase cancelation [43,44]. Reflections occurring at the submillimeter scale can significantly affect microwave absorption when they result from impedance mismatches at interfaces between different materials, particularly in multilayered structures. Huynen [45] evaluates the mechanisms of microwave absorption in foamed composites at the Tera hertz frequency, providing valuable insights into how electromagnetic waves interact with layered media through impedance contrasts, showing that interfaces with distinct impedance (Z) can cause partial reflections. Similar conclusions may be reached for non-foamed multilayer materials. In those systems, submillimeter-thick layers can produce multiple internal reflections due to impedance mismatch and interface polarization, enhancing microwave absorption by increasing the electromagnetic field interaction with the lossy layers through destructive interference and scattering energy [46,47].

In this work, multilayer samples were fabricated by stacking thin films (around 0.2 mm thick) of EVA containing 0.5 wt% and 10 wt% of GNP or GNP/IL intercalated with neat EVA layer. The RL results are shown in Figure 10 and summarized in

Table 4. Main EM absorbing parameters of EVA composites containing GNP (0.5 and 10 wt%) and GNP/IL (0.5:0.5 and 10:10 wt%).

Amount of Filler	0.5 GNP		0.5 GNP IL		10 GNP		10 GNP IL
Thickness (mm)	RL (dB)	EAB (GHz)	RL (dB)	EAB (GHz)	RL (dB)	EAB (GHz)	RL (dB)	EAB (GHz)
AS3	−10.3	0.28	−14.2	1.9	−23.3	1.5	−10.8	0.5
AS4	−9.7	− a	−10.9	0.4	−13.2	1.2	−13.6	1.2
AS5	−12.3	1.2	−13.0	1.3	−19.3	2.4	−19.2	1.8
NAS2.5	−20.1	3.5	−14.4	2.6	−12.4	0.4	−24.6	4.3

AS = alternated structure; NAS = non-alternated structure. ^a The sample didn’t present EAB.

. For the samples with an alternated structure (AS), the same amount of layers of the composite and the neat EVA was used, until it reached the desired thickness. For example, AS₂ corresponds to an alternated multilayer structure at 2 mm thick. The measurement of RL was also performed with the non-alternated structure (NAS). For this experiment, 12 layers of around 0.2 mm were stacked together without any other layer in between (total of around 2.5 mm). Regarding the AS structures, the increasing number of layers resulted in a decreasing minimum RL (enhanced EM attenuation) in the X-band, for all the systems studied. However, for each sample, the best response was observed with 5 mm in the X-band and with 3 mm in the Ku-band. Systems at 5 mm thick and with 10 wt% of GNP or GNP/IL resulted in higher EM attenuation.

The multilayer NAS structures presented outstanding EM attenuation for the system containing 0.5 wt% of GNP and 10:10 wt% of GNP/IL, both in the Ku-band frequency range. Overall, the lowest values of RL (better absorption effectiveness) were obtained by the components with 10 wt% of GNP and IL, built with only a thin layer of composites and thickness of 2.5 mm. For this component, the RL reached values of −24 dB (at 15.7 GHz) and an EAB of 4.3 GHz. However, the component with only 0.5 wt% GNP thin films also showed an RL of around −20 dB and an EAB 3.5 GHz.

Table 5. Comparison between the RL parameter of some thermoplastic-based composites loaded with GNP with those involving EVA/GNP and EVA/GNP/IL developed in this study.

Matrix	Filler	Concentration	Processing	RL	Ref
PET/PU	Pani-DBSA	15 wt%	Nonwoven substrate impregnation	−23 dB	[43]
PLA/TPU	CNT	4 wt%	Extrusion + 3D printing	−31 dB	[48]
PMMA	GNP + IL	Multilayer: 1.9 wt%	Solution mixture + compression molding	−13 dB	[49]
EVA	GNP	0.8 wt%	Solution mixture + compression molding	~−20 dB	[50]
PLA	GNP	15 wt%	Solution mixture + compression molding	−19 dB *	[51]
EVA	GNP + IL	Multilayer: 10 wt%	Melt mixing + film extrusion	−24 dB	This work

* value of RL was calculated.

presents a comparison between the microwave absorption performance, in terms of RL, obtained in this study and those reported in the literature. Most other works employ solvent mixture as a processing method to produce the composites, which, although not ideal, often yield good results. However, in this work, even with the use of a more scalable and environmentally friendly method (melt mixing), the evaluated composites demonstrated RL values comparable to those found in the existing literature.

Overall, the samples (bulk and multilayered) showed better results for the Ku-band range. Multilayered samples exhibited higher RL values compared to bulk samples, a phenomenon that can be attributed to the presence of multiple interfaces. These interfaces induce interfacial polarization, which enhances the probability of electromagnetic wave absorption by creating multiple reflections inside the components’ structure, thereby improving the material’s overall attenuation performance [52].

4. Conclusions

Conducting polymer composite films based on an EVA copolymer containing different amounts of GNP were prepared by a melt-mixing procedure using a mini-extruder equipped with a microfilm line accessory. The phosphonium-based ionic liquid, in a proportion GNP/IL = 1:1, was employed to disperse GNP. The addition of IL resulted in a significant increase in AC conductivity due to the improvement of the GNP dispersion, which helps the formation of the conducting pathway. The viscosity of the molten polymer decreased with the addition of IL due to the plasticizing effect of this component. The tensile properties, such as Young’s modulus, tensile strength, and elongation at break presented an improvement with the presence of GNP, mainly with a proportion of 5 and 10 wt%. However, the addition of IL decreased the Young’s modulus. The effect of GNP and GNP/IL on the creep/recovery response of EVA composite films was evaluated. The presence of 5 and 10 wt% of GNP improved the creep resistance and the best response was observed when IL was combined with GNP in a proportion of 5:5 wt%. All these characteristics confirm the effect of ionic liquid as a dispersing agent for the GNP. Finally, the microwave absorbing properties of these composites were investigated with the aim of enlarging the field of application in the electro-electronic industry and stealth technology. In fact, the multilayered structure with 12 thin layers of EVA with GNP/IL = 10:10 wt% stacked together resulted in material with a minimum RL of −24.6 dB, which represents more than 99% of electromagnetic attenuation, and a large frequency range with attenuation lower than −10 dB (4.3 GHz). In summary, the GNP in combination with ionic liquid in an appropriate amount can provide good conductivity, excellent microwave absorbing properties, and creep resistance to the EVA copolymer films. These characteristics make these systems promising candidates for developing reinforcing electronic packaging with applications as electrostatic dissipative and microwave absorbing materials.