Introduction

The growing recognition of environmental concerns and the enforcement of environmental regulations have spurred a shift toward the use of eco-friendly materials in both industrial and domestic settings. This shift has led to increased interest in utilizing natural cellulose fibers as reinforcement materials in polymer matrix-based composites [1,2,3,4]. These cellulose fibers, primarily sourced from animals and plants, present unique challenges and advantages. The animal-derived fibers are scarce and difficult to extract, while plant-based fibers offer abundant availability and numerous benefits, including higher specific strength, biodegradability, renewability, affordability, lower energy consumption during processing, favorable mechanical properties, and carbon dioxide absorption capabilities [5, 6]. Additionally, plant-based fibers require significantly less energy—about 60% less—to produce compared to synthetic fibers such as glass fibers [7, 8].

Over the past decade, there has been a notable focus on developing composites reinforced with natural cellulose fibers and understanding their behavior [9, 10]. Research suggests the potential for a significant reduction in automotive weight (about 15%) by incorporating these fibers into composite materials due to their lower density compared to synthetic alternatives [11, 12]. Among the various polymer matrices, epoxy stands out as a preferred choice for researchers due to its exceptional properties, including damage tolerance, mechanical strength, dimensional stability, chemical resistance, electrical and thermal insulation, corrosion resistance, and low toxicity [6]. Consequently, a range of plant-based fibers such as piassava, Arundo donax, sisal, jute, kenaf, sugar palm, coir, palmyra palm, and Calotropis gigantea have been utilized in fabricating epoxy composites, with their properties extensively studied and documented [13,14,15,16,17]. However, the implementation of natural fibers might be hindered due to poor mechanical strength, weak adhesion with polymer-reinforced epoxy composites (PEC), improper sizing (length and diameter), high moisture content, etc. [14, 15]. Hence, studies on the alkali treatment of natural fibers have become increasingly significant in the past decade, which has helped to enhance their applicability in PECs for various fields. Alkali treatment, typically involving sodium hydroxide (NaOH), enhances interfacial adhesion by removing impurities and waxes from the fiber surface, leading to better stress transfer. Studies have shown that alkali treatment can improve tensile strength by 10–30%, depending on the concentration and duration of treatment. The flexural strength and modulus may also see improvements ranging from 5 to 20%, while impact strength could increase by 15–25% due to better fiber-matrix bonding [18,19,20,21,22,23,24].

Orkhis and Ettaqi [18] examined polypropylene composites reinforced with kenaf fibers, specifically focusing on comparing the effects of untreated kenaf fibers in nonwoven form with those treated with NaOH. This approach was taken to evaluate the impact of NaOH treatment on the mechanical properties and overall performance of the composites. Various NaOH solutions ranging from 1 to 8% by mass were used for alkali treatments [19]. The investigation revealed that the composite properties, including tensile strength, tensile modulus, flexural strength, and modulus, were optimized when using 5% alkali-treated kenaf fibers. In a separate experiment, abaca fibers treated with 5 wt% NaOH demonstrated improved crystallinity, tensile strength, and Young’s modulus compared to untreated fibers [20]. This treatment also strengthened the interfacial shear strength with epoxy. However, higher NaOH concentrations (10% and 15%) resulted in the complete removal of fiber-binding materials and the breakdown or fibrillation of abaca fibers, leading to poor adhesion to the epoxy matrix. Similarly, the effects of alkaline treatment (4%, 6%, and 8%) on the mechanical properties of bamboo fiber PEC were investigated [21]. Composites treated with 6% NaOH solutions exhibited significant enhancements in bending, tensile, and compressive strength, and stiffness compared to untreated composites. In another investigation [22], kenaf fibers immersed in a 6% NaOH solution for 3 h displayed higher flexural strength and modulus, with notable increases compared to untreated kenaf PECs. Furthermore, the utilization of alkali treatment (5% NaOH) on life-cycling of Jatropha Curcas L. filler was explored [23]. Their examination showed up to a 41% enhancement in adhesive bonding with alkali treatment. Moreover, the use of alkali treatment with NaOH (1 wt%) in a grape-cane reinforced composite exhibited excellent flexural and tensile properties [24]. These promising results have spurred further exploration of alkali treatment on natural fibers and its effects on the mechanical and dynamic properties of developed composites. In summary, alkali treatment is a crucial step in enhancing the performance of natural fibers in composite materials. It improves fiber-matrix adhesion, enhances mechanical properties, and increases the thermal stability and durability of the composites, making them more suitable for a wide range of applications.

Among the various fibers, Areca fibers (AF) have emerged as a promising candidate due to their abundance and unique properties. These fibers are primarily produced from the stalk leaf, frond, and fruit of the Areca palm [25]. AFs are categorized under the Areca catechu Linnaeus family, which is mostly found in the Pacific and Southeast Asian regions [25]. Compared to other types of palm fibers, the utilization of AFs is still underdeveloped despite their comparable properties to coir, jute, and kenaf [26]. More importantly, AFs have a biodegradable nature and superior availability. It has been reported that the length, diameter, and density of AFs vary in the ranges of 18–46 mm, 0.285–0.89 mm, and 1.05–1.25 g/cm³, respectively [27]. However, a major issue with AFs is their weak interfacial bonding with the hydrophobic epoxy matrix and increased moisture content. As a result, Desai et al. [28] emphasized the need for surface treatment to improve the applicability of these AFs. However, studies on alkali surface treatment and optimization of AFs are very limited [28, 29]. Shanmugasundaram et al. [29] conducted NaOH surface treatment with concentrations ranging from 5 to 15% on AFs and elucidated their chemical, physical, and mechanical properties. Their results revealed an increase in cellulose content with NaOH treatment despite a decrease in hemicellulose and lignin concentrations. Moreover, FTIR analysis confirmed the presence of chemical groups after NaOH treatment. Additionally, 5% NaOH treatment exhibited peak tensile properties compared to other concentrations. However, their analysis used only a single AF to study the changes in physicochemical and mechanical properties after NaOH treatment. Therefore, further studies are needed to confirm the actual changes in these properties by developing PECs with and without NaOH treatment. Hence, the novelty of the present work is to develop a new PEC using NaOH-treated AFs by varying NaOH concentrations (5%, 10%, and 15%). This will be followed by an investigation into the impact of NaOH treatment on the static and dynamic mechanical and thermal properties of these PEC + AF composites for the first time.

Materials and methods

Materials

A stalk of an Areca palm leaf was harvested from a local farm in Udumalai, Tamil Nadu, India [29]. The fibers were meticulously detached from the stalks and submerged in water for 21 days. Later, the fibers were extracted using a wooden hammer to strike and separate them [30, 31]. To remove excess moisture from the fibers, they underwent a series of rinsing and cleansing procedures conducted under continuous water flow. Subsequently, the fibers were left to dry for two days at 32 °C [32]. Figure 1 illustrates the process of AF extraction from the stalks of Areca palm leaves. Covai Seenu and Company, located in Coimbatore, India, offers a wide range of materials appropriate for the manufacturing of composites. The supplier provides methyl ethyl ketone peroxide (a catalyst), cobalt naphthenate (a curing agent), and isophthalic polyester resin (IPR), which are readily available for commercial use. Table 1 provides information on the characteristics of isophthalic polyester resin.

Fig. 1
figure 1

Photographic view of an AF

Table 1 Epoxy resin and hardener properties [33]

Alkali Treatment of the Fibers

The AF was immersed in NaOH solutions with weight percentages of 5%, 10%, and 15% for a duration of 30 min at 32 °C. Subsequently, the fibers underwent a washing procedure in an aqueous medium to remove the excess NaOH previously attached to the fiber surface. The remaining NaOH groups on the fiber surface were eliminated through the neutralization of the fibers using a highly diluted hydrochloric acid (HCl) solution with a concentration of 3% by weight. The fibers were then sanitized using distilled water, followed by air drying at 32 °C for the next two days. Table 2 displays the chemical, physical, and mechanical characteristics of four distinct categories of AF, including untreated AF (UTAF) and alkali-treated AFs (5% ATAF, 10% ATAF, and 15% ATAF).

Table 2 Properties of the considered UTAF, 5% ATAF, 10%ATAF and 15% ATAF

Preparation of the Composite Specimens

Fig. 2
figure 2

Photographic view of the prepared composite materials

The present study involved the fabrication of composites using a rectangular mild steel plate measuring 200 × 150 × 10 mm³. To fabricate the composite, a top and bottom plate were utilized in the mold. The longitudinal arrangement of UTAF, 5% ATAF, 10% ATAF, and 15% ATAF was achieved by placing them separately on the bottom plate. Following this, the upper plate was positioned perfectly and subjected to 900 s of compression. The objective of this process was to create a fiber mat with a thickness of 3 mm, which was subsequently removed from the mold and stored in a separate location. The mold was cleaned with pure yarn fabric, and a release agent was sprayed onto the bottom and upper plates. To prepare the matrix, an accelerator and a catalyst were mixed into the isophthalic polyester resin at a ratio of 1.5%, and the mixture was continuously agitated to ensure homogeneity. The prepressed fiber mat was then positioned within the lower plate of the mold, followed by the addition of the matrix material into the mold cavity. The weight ratio between the fiber and polymer matrix was consistently maintained at 30:70. To ensure a solid binding between the fiber and matrix, a pressure corresponding to 1 tonne was applied to the upper and bottom plates for 720 min. The prepared composite was then post-cured for another 720 min at 32 °C. Figures 2(a-d) shows the preparation of composite materials.

Static Mechanical Characteristics of the Composites

The study investigated the static mechanical properties of untreated Areca Palm Leaf Stalk Fiber Composite (UTAFC) and alkali-treated Areca Palm Leaf Stalk Fiber Composite (ATAFC) with varying levels of alkali treatment (5%, 10%, and 15%). These were denoted as 5% ATAFC, 10% ATAFC, and 15% ATAFC, respectively. Specimens for impact, flexural, and tensile tests were obtained by cutting them from a manufactured composite plate. Tensile properties were assessed following ASTM D 638-03 guidelines, using dog bone-shaped specimens with specified dimensions, tested on an Instron tensile testing machine. Tests were conducted at ambient temperature, with a crosshead speed of 5 mm/min and a gauge length of 50 mm. Impact strength evaluation followed ASTM D256 standards using an Izod impact tester, taking into account the sample dimensions.

Flexural properties were evaluated according to ASTM D 790-03 standards, as followed in [23]. The flexural strength and modulus were determined using a Lloyd instrument LR with a maximum load capacity of 100 kN, at a crosshead speed of 2 mm/min. Five samples were analyzed, and the average results for tensile, flexural, and impact tests were recorded. Equations 1 and 2 were used to calculate the flexural strength (σf) and flexural modulus (Ef) of the composites, providing a detailed explanation of the empirical methodologies employed.

$$\:{\sigma\:}_{f}=\frac{3P{L}_{s}}{2b{h}^{2}}$$
(1)
$$\:{E}_{f}=\frac{{L}_{s}^{3}m}{4b{h}^{3}}$$
(2)

Where P denotes peak ultimate failure load (N); Ls represents support span length (mm); h and b symbolizes composites thickness and width (mm); m refers to the slope of load-deflection curves.

Water Absorption Test

Specimens for water absorption testing were cut from composite plates comprised of UTAFC, and 5%, 10%, and 15% ATAFC. The measurements were taken in line with the ASTM D 570 − 99 standard, taking into account the specimens’ dimensions [23]. Initially, an electronic scale with a resolution of 0.00001 mg was employed to weigh the specimens after they had been dried. The specimens were immersed in distilled water and then removed at predetermined intervals. They were gently dried with tissue paper to eliminate any lingering moisture from their surfaces, and the weight was recorded. This process was repeated at regular intervals, and weight measurements were taken. The water absorption percentage in composites was estimated using Eq. 3.

$$\:{W}_{c}=\frac{{W}_{o}-{W}_{t}}{{W}_{o}}\times\:100$$
(3)

Where, Wc represents the sample’s water absorption percentage, Wo denotes the sample’s initial dry weight, and Wt symbolizes the sample’s wet weight after the specific period.

Dynamic Mechanical Analysis (DMA) and Scanning Electron Microscopy (SEM)

The DMA was employed to investigate the temperature-dependent properties, namely the damping factor (Tan δ), loss modulus (E”), and storage modulus (E’) of the considered UTAFC and ATAFCs. Rectangular specimens obtained from the produced composites were prepared in accordance with the guidelines outlined in ASTM D5418-15. The SII-Inkarp (DMS 600) machine was employed to perform the DMA experiment. The analysis was conducted within a frequency range of 1 Hz and a heating rate of 3.8 °C per minute from 30 °C to 200 °C. The JEOL 6390 type SEM, which operates at 10 kV, was utilized to analyze the morphological alterations of crushed composite specimens. The fracture zones of the composite specimens were meticulously dissected and securely affixed to aluminum stubs within the SEM apparatus in order to collect micrographs at different magnification settings.

Results and Discussion

This section elaborates on the changes in the mechanical properties of the composites under consideration (UTAFC and ATAFC), including stress-strain curves, tensile strength/modulus, flexural strength/modulus, and impact strength. Results from water absorption tests, DMA, and SEM analysis are also presented.

Mechanical Properties

The properties related to the composites’ mechanical aspects can be investigated by analyzing the interfacial interactions among the hydrophilic AFs and the hydrophobic matrix. The present study aims to examine the impact of different concentrations of alkali treatment on the tensile properties of PECs reinforced with AF. Figure 3 illustrates the stress-strain curves for UTAFC, 5% ATAFC, 10% ATAFC, and 15% ATAFC.

Fig. 3
figure 3

Tensile stress-strain characteristics of the various composites

Fig. 4
figure 4

Changes in the tensile strength and modulus of the considered UTAFC and ATAFCs

The application of an alkali treatment to fiber-reinforced PECs resulted in significant enhancements in their tensile properties, including both tensile strength and modulus, as displayed in Fig. 4a and b, respectively. Notably, the 10% ATAFC exhibited a remarkable improvement, with a 53% increase in tensile strength compared to UTAFC, while tensile modulus increased by 28%. In contrast, the improvements in tensile strength for the 5% ATAFC and 15% ATAFC were 39% and 17%, respectively, while the corresponding improvements in tensile modulus were 22% and 10%, respectively. Moreover, the 10% ATAFC demonstrated an 18% increase in tensile strength compared to the 15% ATAFC, with the tensile modulus increasing by 12%. The UTAFC demonstrated the lowest tensile strength value of 41.83 ± 1.29 MPa, with a tensile modulus of 1.07 ± 0.15 GPa. However, the 10% ATAFC material exhibited the highest tensile strength among the tested samples, measuring 88.16 ± 2.01 MPa, with a tensile modulus of 1.48 ± 0.31 GPa. The tensile properties of UTAFC were observed to decrease, ultimately resulting in failure due to the weak bond between AF and the matrix. However, the alkali treatment of the AFs led to the separation of strands into smaller sizes and a consequent alteration in their crystalline structure [22, 23]. This modification facilitated better interlocking of the cellulose chains and consequently improved the bond between the AF and the matrix [23]. The observed enhancement in bonding can be attributed to the elimination of undesirable substances from the AF surface and the prevention of cementing materials from adhering to it.

Fig. 5
figure 5

Changes in the flexural strength and modulus for the considered UTAFC and ATAFCs

The ability of PECs to withstand distortion under a bending load is commonly known as flexural strength or bending strength. The quantification of a material’s ability to resist bending deflection is referred to as flexural modulus or bending modulus. Figure 5a and b illustrate the flexural characteristics of various samples, namely UTAFC, 5% ATAFC, 10% ATAFC, and 15% ATAFC. The results indicate minimal flexural strength and modulus of 65.39 ± 2.13 MPa and 1.82 ± 0.39 GPa, respectively, for UTAFC. In contrast, the 10% ATAFC displayed peak flexural strength and modulus of 97.21 ± 3.28 MPa and 2.39 ± 0.32 GPa, respectively. Compared to UTAFC, the flexural strength of 5% ATAFC exhibited a 21% increase, while the 10% and 15% ATAFC demonstrated a 32% and 28% increase, respectively. Additionally, the flexural modulus of 5% ATAFC showed a 7% increase, whereas the 10% and 15% ATAFC exhibited a 24% and 16% increase, respectively. When a force perpendicular to the composites’ longitudinal alignment is applied, the flexural properties of PECs reinforced with AF are vulnerable to the influence of inter-laminar forces. The observation of heightened levels of alkali (10%) has been found to have a beneficial effect on the surface roughness of AF, thereby resulting in an increase in wettability and an enhancement in adhesion between the AF and the matrix. As a result, the bending strength and rigidity of the AF-reinforced PEC were improved.

The fracture of a composite specimen is a critical factor in determining its impact strength, which is quantified by the energy required to induce fracture. To enhance the impact strength of AF-reinforced PEC, it is imperative to ensure the uniform distribution of stress during the application of sudden loads. The impact behavior of AF reinforcement in a PEC is influenced by various factors, including the properties of the fibers and matrix, as well as the adhesion between them.

Fig. 6
figure 6

Impact strength of the various composites

Figure 6 depicts the impact strength of different composites, specifically UTAFC, 5% ATAFC, 10% ATAFC, and 15% ATAFC, and Table 3 shows the comparison of the mechanical properties of epoxy with 10% ATAFC. The presented data unequivocally indicate that fiber composites treated with alkali exhibit superior impact strength compared to UTAFC. Specifically, the alkali-treated fiber composites demonstrate a notable increase of 33%, 51%, and 42% in impact energy absorption, which corresponds to a 31%, 48%, and 37% increase in impact strength for 5% ATAFC, 10% ATAFC, and 15% ATAFC, respectively, when compared to UTAFC. The 10% ATAFC exhibits superior performance in terms of impact energy (0.88 J) and impact strength (22.56 kJ/m²) when compared to other composites in its classification. The study demonstrates a notable improvement in the impact strength of ATAFC compared to UTAFC. This enhancement can be attributed to the improved interfacial adhesion between the AF and the matrix, which is a direct consequence of the alkali treatment. The treatment has been found to facilitate superior energy absorption and effectively hinder the propagation of cracks within the composites. Consequently, a substantial increase in impact strength has been achieved. Table 4 shows the comparison of mechanical properties with other natural fiber composites.

Table 3 Comparison of Mechanical properties of Epoxy with 10%ATAFC [34, 35]
Table 4 Comparison of Mechanical properties with other natural Fiber composites [34, 35]

Water Absorption Properties

The incorporation of natural AFs as reinforcement in PEC is associated with significant drawbacks, particularly their high susceptibility to water absorption. This phenomenon causes a reduction in the mechanical properties of the PECs and an increase in dimensional instability. The diminished mechanical properties and inconsistent dimensions observed in composites can be attributed to the inadequate transmission of stress between the AF and the matrix, which arises from the expansion and degradation of fibers within the polymer matrix. The heightened water absorption in these composites is collectively contributed by the hydrophilic nature of the AF, the existence of internal voids and micro-cracks within the AF and PEC, and the interfacial bonding between the AF and the matrix.

Fig. 7
figure 7

Percentage of water absorption for the various UTAFC and ATAFCs

Figure 7 illustrates the water absorption behavior of the AF-reinforced PECs over 24 h, with measurements taken at 2-hour intervals. The results demonstrate that ATAFC exhibits a decreased water absorption rate compared to UTAFC. This phenomenon is attributed to the presence of wax, lignin, hemicellulose, and other impurities on the AF surface. The application of an alkali treatment effectively eliminates these impurities, subsequently reducing the hydrophilic nature of AF. Moreover, the utilization of an alkali treatment has been shown to increase the AF’s surface roughness, leading to the formation of indentations and pores on its surface [23]. This alteration has been found to enhance the bonding between the AF and the matrix and confer hydrophobic properties to the fiber [35, 36]. The improved bonding between the AF and the matrix has been observed to reduce the rate of water molecule diffusion. After conducting experiments over 24 h, the rates of water absorption were determined to be 9.2% for UTAFC, 7.7% for ATAFC, 6.3% for 10% ATAFC, and 5.4% for 15% ATAFC. It is important to note that 15% ATAFC exhibited lower water absorption rates compared to UTAFC, 5% ATAFC, and 10% ATAFC due to the increased concentration of alkali on the AF surface, which reduces hydrophilicity. Table 5 compares the water absorption performance of the present PEC+AF with other natural fiber composites and the results indicate that the observed water absorption values of the UTAFC and ATAFCs were in line with existing studies.

Table 5 Comparison of water absorption performance with other natural fiber composites

DMA Results

The influence of temperature on the storage modulus (E’) of UTAFC, 5% ATAFC, 10% ATAFC, and 15% ATAFC is illustrated in Fig. 8. There was a decline in E’ with increasing temperature for the alkali-treated AF-reinforced PECs, and a similar trend was observed in [30] for Kenaf-Epoxy with alkali modification. Specifically, in the region of plastic deformation, it was observed that 10% ATAFC had a storage modulus of 2487 MPa, whereas it was 34.9% lower in the case of UTAFC. Similarly, the storage modulus of 5% and 15% ATAFCs was 12.8% and 6.7% lower, respectively, than that of 10% ATAFC. These findings suggest that modifying the alkali treatment concentration significantly influenced the physical and chemical properties of the AF, leading to improved bonding between the AF and the matrix. The storage modulus exhibited a consistent increase in all ATAFCs compared to UTAFC, indicating a positive trend throughout the transition region. Figure 8a showed a gradual decline in the storage modulus with increasing temperature for all AF-reinforced PECs. The stiffness of the AFs decreased as the temperature increased, resulting in a decline in the storage modulus of the composites. The pure resin displayed a storage modulus of 145 MPa, whereas the incorporation of 10% ATAFC increased it to 685 MPa at 160 °C. This improvement can be attributed to enhanced interfacial adhesion between the AF and the matrix, consequently augmenting the composite’s storage modulus.

Fig. 8
figure 8

Changes in the storage modulus (E’) and Loss modulus (E’’) for the considered UTAFC and ATAFCs

Changes in the loss modulus with respect to temperature for UTAFC, 5% ATAFC, 10% ATAFC, and 15% ATAFC are displayed in Fig. 8b. It is noteworthy that the pure resin material exhibited the lowest loss modulus, measuring 183 MPa. This can be attributed to the limited dissipation of heat within the material. Conversely, in the glassy transition region, the 10% ATAFC material demonstrated the highest loss modulus, measuring 366 MPa. This notable increase in loss modulus can be attributed to the alkali treatment applied to the material, which effectively removed impurities from the fiber surface and transformed its smooth texture into a rough one.

Fig. 9
figure 9

Impact of the temperature on the damping factor of the various composites

In addition to these modulus properties, the damping factor (Tan δ) is also an important phenomenon that pertains to energy dissipation within an AF-reinforced PEC when it is subjected to cyclic loading. Tan δ can be estimated by the ratio of E’’ to E’ of the AF-reinforced PEC. Figure 9 depicts the temperature-dependent Tan δ for UTAFC, 5% ATAFC, 10% ATAFC, and 15% ATAFC. The Tan δ of all the AF-reinforced PECs increased until approximately 70 °C, beyond which it declined and entered the rubbery region [30]. The damping characteristics of AF-reinforced PECs are influenced by the shear stress experienced by the AF and the viscoelastic dissipation of the PEC. The adhesion strength between the AF and the matrix is a critical determinant of the Tan δ of composites. A weak bond between these two components leads to an increase in Tan δ, while strong adhesion restricts the polymer chains’ movement within the composites, resulting in a decrease in Tan δ [26]. As depicted in Fig. 9, the 10% ATAFC demonstrates a lower Tan δ compared to UTAFC, 5% ATAFC, and 15% ATAFC. This observation implies that the effective interfacial bonding between 10% ATAFC and epoxy is responsible for the reduction in Tan δ of the AF-reinforced PEC.

SEM Analysis

Fig. 10
figure 10

Surface morphology of the Tensile Fractured Specimen (a) UTAFC (b) 5% ATAFC (c) 15% ATAFC (d) 15% ATAFC

The surface morphology of fractured samples subjected to tensile loads for UTAFC, 5% ATAFC, 10% ATAFC, and 15% ATAFC is depicted in Fig. 10. In the case of UTAFC (Fig. 10a), a higher incidence of fiber pullout and fracture was observed compared to the ATAFCs. The presence of wax, lignin, hemicellulose, and moisture in the AF could be the major factors behind the observed changes. The inadequate adhesion between the AF and PEC was evident in UTAFC, leading to weakened mechanical properties of the AF-reinforced PECs [31]. Figure 10b presents the SEM image of 5% ATAFC, where the application of tensile force resulted in the observation of partial fiber pullout and breakage. This phenomenon can be attributed to the presence of certain quantities of wax, lignin, hemicellulose, and moisture on the AF surface. In contrast, Fig. 10c displays the SEM image of 10% ATAFC, which exhibited a reduced occurrence of fiber breakage and pullout. This suggests favorable adhesion between the AF and matrix. Consequently, the transfer of stress between the AF and matrix experienced a significant increase, leading to enhanced mechanical properties compared to UTAFC, 5% ATAFC, and 15% ATAFC. The SEM image presented in Fig. 10d shows a higher incidence of fiber breakage and pullout in 15% ATAFC compared to 10% ATAFC. The application of alkali treatment resulted in the defibrillation of the AF surface, as well as the formation of a rough texture with an increased number of pits and pores [23]. These changes facilitated improved adhesion between the AF and matrix, ultimately leading to enhanced mechanical properties of 10% ATAFC compared to 15% ATAFC.

Thermal Degradation of the Composites

Fig. 11
figure 11

TGA curves for the UTAFC and ATAFCs

Fig. 12
figure 12

DTG curves for the UTAFC and ATAFCs

The thermal degradation of PECs reinforced with AF is primarily influenced by the chemical properties of AF, including its cellulose, hemicellulose, lignin, and wax contents [32]. In this context, Thermogravimetric Analysis (TGA) and Differential Thermogravimetric (DTG) curves of the AF-reinforced PECs under consideration are presented in Figs. 11 and 12, respectively. The analysis reveals four distinct stages in this degradation process. Figure 11 indicates that the decomposition initiation temperature of the ATAFC samples is higher than that of the UTAFC sample, and, as expected, the peak temperature was observed at higher NaOH concentrations. Increased resistance of AFs to elevated temperatures and enhanced compatibility of PEC + AF with NaOH treatment were the major causes behind the improved performance. A similar variation was also observed by Dhakal et al. [46] after NaOH treatment of hemp/UP PEC. On the other hand, the DTG results in Fig. 12 showed three distinct stages of degradation, with weakened performance of UTAFCs and improved performance of ATAFC samples. The DTG results of Dhakal et al. [46] after NaOH treatment of hemp/UP PEC also supported the present trend. The first stage, occurring between 220 °C and 250 °C, can be attributed to the elimination of solvents from the PECs. The subsequent phase of degradation occurs between 330 °C and 360 °C, leading to the rapid disintegration of lignin, hemicellulose, and cellulose within the composite materials. The thermal degradation thresholds for UTAFC, 5%, 10%, and 15% ATAFC are documented as 334 °C, 339 °C, 346 °C, and 351 °C, respectively. The third phase of thermal degradation, experienced between 430 °C and 480 °C, is primarily attributed to the decomposition of the soft segment and volatilization within the PECs. Specifically, the thermal decomposition temperatures associated with this stage are recorded as 441 °C, 450 °C, 458 °C, and 466 °C, respectively. The decomposition process reaches its final stage between 540 °C and 620 °C, wherein both the epoxy matrix and the AF undergo degradation. The thermal decomposition temperatures for UTAFC, 5% ATAFC, 10% ATAFC, and 15% ATAFC are measured as 543 °C, 558 °C, 565 °C, and 572 °C, respectively. These findings suggest an enhancement in the thermal stability of the composites and improved diffusion between the AF and matrix due to the elevated alkali concentration in the AF Fig. 12 illustrates the DTG curve, which provides insight into the weight losses during the degradation of UTAFC, 5% ATAFC, 10% ATAFC, and 15% ATAFC. The initial stage shows a weight loss of 20–25% in the composites at 220 °C to 250 °C. In the second stage, a weight loss of 60% occurs at 330 °C to 360 °C, while the third phase reveals a weight loss of 80% between 430 °C and 480 °C. Finally, in the last stage, a significant weight loss of 90% is observed beyond 540 °C in the composites. In addition, Table 6 shows a summary of the TGA results of AF composites (UTAFC, and ATAFCs).

Table 6 Summarized TGA results

Conclusions

To create lightweight composites, cellulosic fibers from the petioles of Areca palm leaf stalks were effectively extracted using the water-retting process. These fibers were then treated with different concentrations of NaOH—5%, 10%, and 15%—and subsequently compressed. This study followed standard ASTM procedures to investigate the composites’ static and dynamic mechanical properties. Through SEM and single-fiber pull-out tests, the results showed that 10% ATAFC exhibited better interfacial bonding with the epoxy matrix and improved static and dynamic mechanical properties compared to UTAFC. Among the AF-reinforced PECs, 10% ATAFC showed the highest tensile and flexural properties. However, the 15% ATAFC exhibited lower water absorption rates, likely due to the increased concentration of alkali on the AF surface, which reduced the AF’s hydrophilic nature. Moreover, the storage modulus of 10% ATAFC was 34.9% higher than that of UTAFC, with the lowest Tan δ. The thermal decomposition temperatures for UTAFC, 5% ATAFC, 10% ATAFC, and 15% ATAFC were measured as 543 °C, 558 °C, 565 °C, and 572 °C, respectively. The outcomes of this study indicated that 10% NaOH-treated AF was the best composition, and it is recommended for the development of AF-based epoxy composites. The properties of this composite material were found to be the best natural option for automotive panels and other industrial applications.

Future Scope

Following this research, future studies could explore the long-term durability and environmental impact of these ATAFCs in various applications, particularly under different working temperatures, humidity, and other conditions. Additionally, investigating the use of other natural fibers in combination with AF or in hybrid composites could provide insights into further improving material properties. Finally, studying the recyclability and end-of-life disposal of these ATAFCs would contribute to the development of more sustainable composite materials.