Synergistic Effects of Furfurylated Natural Fibers and Nanoclays on the Properties of Fiber–Cement Composites

Abstract

Fiber–cement composites have been increasingly studied for sustainable construction applications, but durability issues—particularly fiber degradation in alkaline environments—remain a challenge. This study aimed to evaluate the individual and combined effects of furfurylated sisal fibers and nanoclay additions on the physical and mechanical performance of autoclaved fiber–cement composites, seeking to enhance fiber durability and matrix compatibility. All the composites were formulated with CPV-ARI cement and partially replaced with agricultural limestone to reduce the environmental impact and production costs. Sisal fibers (2 wt.%) were chemically modified using furfuryl alcohol, and nanoclays—both hydrophilic and surface-functionalized—were incorporated at 1% and 5% of cement weight. The composites were characterized for physical properties (density, water absorption, and apparent porosity) and mechanical performance (flexural and compressive strength, toughness, and modulus). Furfurylation significantly improved fiber–matrix interaction, leading to higher flexural strength and up to 100% gain in toughness. Nanoclay additions reduced porosity and increased stiffness, particularly at 5%, though excessive content showed diminishing returns. The combination of furfurylated fibers and functionalized nanoclay provided the best results in maintaining a compact microstructure, reducing water absorption, and improving mechanical resilience. Optical microscopy confirmed improved fiber dispersion and interfacial bonding in composites containing furfurylated fibers and functionalized nanoclay. These findings highlight the effectiveness of integrating surface-treated natural fibers with pozzolanic additives to enhance the performance and longevity of fiber–cement composites.

1. Introduction

The construction industry is widely recognized as a major contributor to environmental degradation, from raw material extraction to waste generation. According to Munaro et al. [1], construction accounts for nearly one-third of global energy consumption. Among construction materials, cementitious products represent a significant environmental burden, with Portland cement production alone responsible for approximately 7% of global CO₂ emissions [2]. Reducing greenhouse gas emissions, limiting the use of non-renewable resources, and minimizing the generation of non-biodegradable waste in the cement industry are critical steps toward achieving sustainability in this economically and socially vital sector. Cement-based materials, due to their inorganic and non-metallic nature, are conventionally classified as ceramic materials within the field of materials science. This classification reinforces their relevance in discussions on ceramic materials for sustainable development.

One approach to making cement-based materials more sustainable involves reinforcing them with natural fibers. The use of plant fibers, such as jute and sisal, has historical roots but became prominent in the 1940s with the production of non-structural precast components [3]. Jaber et al. [4] highlight that natural fibers help address several inherent weaknesses in cementitious matrices, including brittleness, poor crack resistance, limited flexural deformability, and low flexural strength relative to compressive strength. Numerous studies report improvements in mechanical performance—such as increased flexural and compressive strength, enhanced ductility, reduced cracking, and improved toughness—following the incorporation of natural fibers [5,6]. These fibers may also reduce the density and thermal conductivity of the composite as their content increases [7].

However, several limitations still hinder the performance of fiber-reinforced cement composites. These include increased setting time, reduced workability, and most notably, fiber degradation in the highly alkaline environment (pH > 12) produced by cement hydration products like Ca(OH)₂ [3]. Alkaline degradation primarily affects lignin and hemicellulose, exposing holocellulose components (cellulose and hemicellulose) to further deterioration. Over time, even the cellulose breaks down, leaving the fiber ineffective as reinforcement. This degradation accelerates at elevated temperatures (above 75 °C) [8]. As highlighted by Pickering et al. [9], fiber degradation also compromises durability, increases permeability, and weakens the fiber–matrix bond. These effects reduce mechanical strength, alter fiber morphology (shorter and thinner fibers), and disrupt the chemical composition, resulting in reduced adherence between fibers and the cement matrix [10]. To mitigate these issues, pozzolanic additions and surface treatments are commonly used.

Pozzolans help lower matrix alkalinity by reacting with the alkaline ions released during cement hydration, such as NaOH and KOH, forming non-expansive compounds that densify the matrix and reduce its aggressiveness toward fibers [11]. This reduction in alkalinity is particularly important for preserving the integrity of lignocellulosic fibers, which are highly susceptible to degradation in high-pH environments. By consuming calcium hydroxide (Ca(OH)₂), pozzolanic materials contribute to the additional formation of calcium silicate hydrate (C–S–H) gel, thus enhancing the mechanical strength and dimensional stability of the composite [11,12]. Furthermore, pozzolans reduce porosity and refine the pore structure, which limits water ingress and mitigates the leaching of alkaline components over time [13].

Among the available pozzolans, nanoclays are particularly promising due to their high silica and alumina content—approximately 60% of total composition—and their ability to react with calcium hydroxide to form additional C–S–H gel and calcium aluminate hydrates. This reaction improves both early and long-term strength and durability. Nanoclays also enhance microstructural refinement through their filler effect and hydration acceleration, making them highly effective for improving cement-based materials. Recent studies have supported these findings. Hakamy et al. [12,14] demonstrated that the addition of 1 wt.% of organomodified nanoclay (Cloisite 30B—Sigma-Aldrich, St. Louis, MO, USA) to cement composites reinforced with hemp fabric significantly improved flexural strength, fracture toughness, and matrix density while reducing porosity. The nanoclay not only acted as a filler but also facilitated a more effective pozzolanic reaction, enhancing the bond between the fiber and the matrix. Wei and Meyer [15] explored the combined effect of metakaolin and nanoclay in sisal fiber–cement composites. Their findings showed that a partial replacement of cement with 3 wt.% nanoclay and 27 wt.% metakaolin reduced the CH content by more than 64%, suppressed ettringite formation, and increased the interfacial bond strength and pull-out energy of the fibers by 131% and 196%, respectively.

In parallel, chemical fiber treatments are used to limit Ca(OH)₂ deposition on fiber surfaces. The treatment of natural fibers before their incorporation into fiber–cement composites is a well-established strategy to enhance their compatibility with the cementitious matrix and improve composite durability. Methods like acetylation, mercerization, and silanization are commonly reported in the literature [16]. For instance, Ajouguim et al. [17] demonstrated that chemically treated alfa fibers (e.g., alkaline and hydrothermal treatments) increased the flexural strength of cement mortars by up to 44% and improved fiber–matrix adhesion, confirmed through shear stress measurements. Similarly, Biskri et al. [18] investigated the incorporation of chemically treated sisal fibers using urea and ethylenediaminetetraacetic acid and observed a 17–23.4% increase in flexural strength along with denser interfacial microstructures and higher cement hydrate development. Fonseca et al. [19] studied the hybrid reinforcement of fiber–cement composites using jute fibers and cellulose nanofibrils, reporting synergistic improvements in mechanical strength, porosity reduction (up to 75%), and enhanced post-weathering performance, particularly for compositions combining both reinforcement scales.

One environmentally friendly treatment option is furfurylation, a process in which furfuryl alcohol (FAL) reacts with a catalyst within the fiber structure, leading to the formation of poly(furfuryl alcohol). This treatment, widely used for wood modification, enhances fiber dimensional stability, hydrophobicity, and chemical resistance—key properties for protecting fibers in alkaline cementitious environments [20]. In a recent study by Aramburu et al. [20], microfibrillated cellulose treated with furfuryl alcohol was incorporated into cementitious pastes, resulting in significant improvements in fresh and hardened properties. The furfurylated MFC enhanced early-age compressive strength by up to 149%, mitigated setting time delays, and increased durability under wet–dry cycles.

Based on that, the combination of pozzolanic additions and fiber surface treatments in fiber–cement composites represents a promising approach, with several encouraging results already reported in the literature, to enhance both mechanical performance and long-term durability. While pozzolanic materials contribute to matrix densification, a reduction in free calcium hydroxide, and decreased alkalinity, fiber treatments improve dimensional stability, reduce water absorption, and strengthen fiber–matrix interfacial bonding. This dual approach may lead to composites with higher flexural strength, improved toughness, and reduced porosity. Recent studies support these findings. Mejia-Ballesteros et al. [21] observed that thermally treated eucalyptus fibers combined with a pozzolanic matrix containing activated coal waste led to increased modulus of rupture and durability after aging cycles. Hakamy et al. [22] demonstrated that alkali-treated hemp fabrics reinforced with calcined nanoclay yielded substantial gains in fracture toughness and flexural strength due to improved matrix compactness and fiber adhesion. Thepruttana et al. [23] further showed that coating sisal fibers with natural rubber latex and expanded perlite, a pozzolanic additive, significantly enhanced flexural performance and reduced porosity, highlighting the synergy between fiber modification and matrix refinement.

Therefore, despite its potential, furfurylation has been scarcely investigated in the context of natural fiber reinforcement for fiber–cement systems. Additionally, there is limited literature exploring the combined use of fiber surface treatments and pozzolanic additions as a strategy to prevent fiber mineralization and enhance the mechanical performance of fiber cements. This study deals with novel fiber–cement composites reinforced with chemically treated sisal fibers using furfurylation and further modified by incorporating two types of nanoclays—one hydrophilic and the other surface-functionalized. Physical properties such as density, water absorption, and porosity, as well as mechanical performance including flexural and compressive strength, toughness, and modulus, are systematically assessed. Fracture toughness is also measured using notched specimens. Through this approach, the study investigates the potential synergistic interaction between fiber treatment and pozzolanic addition to improve fiber durability and overall composite performance under autoclave curing conditions.

2. Materials and Methods

2.1. Materials

The cement used in this study was a high-early-strength Portland cement (CPV-ARI) from the Cauê brand (InterCement Brasil S.A, Cezarina, Goiás, Brazil), chosen due to its purity and absence of mineral additives that could interfere with the pozzolanic reactions or fiber interaction. Agricultural limestone from Dagoberto Barcellos (Dagoberto Barcellos S.A, Caçapava do Sul, Rio Grande do Sul, Brazil) was used as a partial replacement for the cement to reduce the environmental impact and material cost, a practice already common in industrial fiber–cement production.

Two nanoclays were used: (i) a hydrophilic bentonite nanoclay (Nanomer^® PGV), consisting mainly of unmodified montmorillonite, with a maximum particle size of 25 µm, and (ii) a surface-functionalized nanoclay (Nanomer^® I.31PS), based on montmorillonite modified with octadecylamine (15–35 wt.%) and aminopropyltriethoxysilane (0.5–5 wt.%), with particles up to 20 µm. Both were acquired from Sigma-Aldrich (Laboratory and Industrial Chemical—St. Louis, MO, USA) and incorporated into the cement matrix using the same dry mixing method.

The sisal fibers were commercially obtained from Compel, a supplier located in Arujá, São Paulo. They were delivered as twisted yarns and manually cut to 2 cm in length before use. Their selection was based on their low cost, native availability in Brazil, and good reinforcement potential. The fibers were characterized morphologically using an Opton TNB-04T-PL optical microscope (Opton—São Paulo, Brazil) and analyzed using ImageJ software (Version 1.54d) for dimensional measurements of 30 samples. The fibers were also chemically characterized using standard wet-chemistry methods for quantifying moisture, ash, extractives, acid-insoluble lignin, and holocellulose content (by difference). Fourier Transform Infrared Spectroscopy (FTIR) analysis was conducted on dried fiber powder compacted into KBr pellets, with 32 scans per sample in the 4000–1000 cm⁻¹ range, at 4 cm⁻¹ resolution. Thermal stability was assessed via thermogravimetric analysis (TGA) under inert nitrogen atmosphere from 50 °C to 250 °C at a heating rate of 10 °C/min, using a TA Instruments Q50 device (TA Instrutments—New Castle, DE, USA). Particle size distribution was determined using a CILAS 1180 laser (CILAS—Aubagne, France) diffraction analyzer, with isopropyl alcohol as the dispersant and continuous stirring at 200 rpm.

2.2. Simulation of Alkaline Fiber Degradation

To simulate the degradation process of sisal fibers caused by the alkaline environment of the cementitious matrix, an induced degradation reaction was carried out using a sodium hydroxide (NaOH) solution (98% purity; Sigma-Aldrich). NaOH was added to distilled water until it reached a pH of 12 (approximately 1.7 wt.% NaOH relative to water), which is representative of the alkalinity typically found in cementitious systems. The following material mixtures, including fibers, were immersed in the NaOH solution: (i) fibers only; (ii) fibers + nanoclay; and (iii) fibers + nanoclay + acrylic acid (AF) + maleic anhydride (MA). The MA (98% purity, Sigma-Aldrich) was used as a catalyst for the AF (98% purity, Sigma-Aldrich). After 60 days of exposure to the induced alkaline degradation, the samples were analyzed by FTIR, using the same equipment and parameters previously described.

2.3. Fabrication of Fiber–Cement Composites

The fiber–cement formulations were developed based on the mass proportions established by Tonoli et al. [24]. A fiber content of 2 wt.% was adopted, as supported by prior analyses and recommendations from the literature [3,5,24]. All fiber–cement pastes were prepared with a fixed water-to-cement (w/c) ratio of 0.40, which was determined using a standard consistency test in accordance with ASTM C187 [25]. These parameters ensured consistency across all the formulations and were selected to align with established practices in fiber–cement composite research. To prevent excessive water absorption during mixing, all fibers were pre-moistened. The amount of water used for moistening was measured and subtracted from the total water added during paste preparation to ensure consistent water/cement ratios.

Table 1. Formulation of the fiber–cement composites.

Nomenclature	Cement (wt.%)	Limestone (wt.%)	Sisal (wt.%)	Water (wt.%)	Nanoclay (wt.%)	FAL (wt.%)	MA (wt.%)
CON	62.91	10.43	1.5	25.16	0.0	0.0	0.0
NC1	62.52	10.37	1.49	25.01	0.63	0.0	0.0
NC5	60.99	10.11	1.45	24.4	3.05	0.0	0.0
FAL50	62.42	10.35	1.49	24.97	0.0	0.74	0.04
NC1FAL50	62.03	10.28	1.48	24.81	0.62	0.74	0.04
NC1FAL50P	62.03	10.28	1.48	24.81	0.62	0.74	0.04
NC5FAL50	60.53	10.04	1.44	24.21	3.03	0.72	0.04
FNC1FAL50	62.03	10.28	1.48	24.81	0.62	0.74	0.04

Where FAL and MA are furfuryl alcohol and maleic anhydride, respectively; CON = control (no fibers or additives); NC1 = 1 wt.% nanoclay; FAL50 = furfurylated fibers (50 wt.%); NC1FAL50 = combination of 1 wt.% nanoclay and furfurylated fibers (50 wt.%); NC1FAL50P = fibers pre-treated and with 1 wt.% nanoclay and furfuryl alcohol (50 wt.%); NC5FAL50 = combination of 5 wt.% nanoclay and furfuryl alcohol (50 wt.%); FNC1FAL50 = combination of 1 wt.% surface-functionalized nanoclay and furfuryl alcohol (50 wt.%).

presents detailed information on the nomenclature and mass proportions applied for all the fiber–cement composites from this study.

The materials were first manually homogenized and then mixed using a planetary mortar mixer (Electromechanical planetary mortar mixer—Fortminas—Sao Paulo, Brazil): 3 min at 140 rpm, followed by 1 min at 220 rpm. In mixtures containing furfuryl alcohol (FAL), the fibers were blended with FAL, maleic anhydride (MA), and water before being added to the cement paste. In one specific group (NC1FAL50P), the fibers were pre-treated with FAL and MA, followed by oven curing at 80 °C for 24 h. The color change to reddish brown indicated successful in situ polymerization of the FAL into poly(furfuryl alcohol), which increases fiber hydrophobicity and resistance to alkaline degradation. These treated fibers were then incorporated into the cement paste as in the other groups.

All pastes were cast into prismatic molds (40 × 40 × 160 mm) in two layers and compacted on a vibrating table. In specimens used for fracture toughness testing, a cylindrical metal rod (8 mm diameter) was gently pressed into the center of the fresh paste to form a consistent notch, which remained after hardening. After molding, the specimens were demolded following 5 days of air curing and then subjected to hydrothermal curing in an autoclave at 0.15 MPa and 100 °C for 8 h, as described elsewhere [6]. This also ensured the polymerization of any remaining FAL within the fiber structure.

2.4. Characterization of Fiber–Cement Composites

The final setting time was determined using a Vicat apparatus (ASTM C191 [26]). The initial setting time could not be reliably measured due to the interference of fibers with the needle. Workability was assessed using the flow table method in accordance with ASTM C230/C230M [27]. A truncated cone mold with a base diameter of 10 cm, top diameter of 7 cm, and height of 6 cm was placed at the center of the flow table and filled with the fresh paste in a single lift without compaction. The mold was then lifted vertically, and the table was dropped 25 times within 15 s. The spread diameter was measured in three directions, and the average was taken as the flow value for each mixture.

Bulk density was determined by measuring the oven-dry mass of the specimen and its geometric volume. To assess apparent porosity and water absorption, the specimens were immersed in water for 24 h. Their mass was recorded while submerged, saturated surface-dry, and oven-dry. These values allowed the estimation of the open pore volume and the amount of water absorbed. The calculations followed the ASTM C948 [28] procedure, which relates these masses to determine the accessible voids and water-holding capacity of the composite, as also described by Gutiérrez et al. [29].

Three-point bending and compression tests were conducted to evaluate the mechanical behavior of the composites, following the procedures defined by ASTM C348 [30]. Flexural tests were performed using a universal testing machine (EMIC DL 30000—Instron Brasil, São José dos Pinhais, Paraná, Brazil) with a span of 100 mm and a loading rate of 50 N/s. Rectangular prismatic specimens were positioned on two lower supports, and a central load was applied until failure. During the test, the equipment continuously recorded the applied force (F) and the vertical displacement (δ) at the upper face of the specimen. These raw data were used to calculate the flexural stress (σ_x), strain (ε_x), and flexural toughness. The flexural stress at a given load was calculated using Equation (1). The corresponding flexural strain was determined using Equation (2), which relates the specimen’s height, span length, and measured displacement. The flexural strength (σ_f), defined as the maximum stress sustained by the specimen before failure, was obtained using the maximum applied load (F_max) in Equation (3). Finally, the flexural toughness was computed as the area under the stress–strain curve using OriginPro 2022 software (version 9.9.0.225):

$${\mathsf{\sigma }}_{\mathrm{x}}={\displaystyle \frac{3\mathrm{F}\mathrm{L}}{{2\mathrm{b}\mathrm{h}}^2}}$$

(1)

$${\mathsf{\epsilon }}_{\mathrm{x}}={\displaystyle \frac{6\mathrm{h}\mathsf{\delta }}{{\mathrm{L}}^2}}$$

(2)

$${\mathsf{\sigma }}_{\mathrm{f}}={\displaystyle \frac{3{\mathrm{F}}_{\mathrm{m}\mathrm{a}\mathrm{x}}\mathrm{L}}{{2\mathrm{b}\mathrm{h}}^2}}$$

(3)

where σ_x = flexural stress at a given load (MPa); ε_x = flexural strain (dimensionless); σ_f = flexural strength (MPa); F = applied load (N); F_max = maximum applied load (N); L = span length (mm); b = specimen width (mm); h = specimen height (mm); δ = vertical displacement at the midpoint (mm).

After the flexural test, one side of each fractured specimen was subjected to a uniaxial compression test, also following ASTM C348 [30]. The compression test was carried out using the same universal testing machine, with prismatic platens (40 mm × 40 mm) and a loading rate of 500 N/s. During the test, load and displacement were recorded and converted into stress and strain. The compressive strength (σ_c) was determined by Equation (4). The compressive modulus of elasticity (E_c) was calculated as the slope of the linear portion of the stress–strain curve, defined between 0.5 MPa and 30% of σ_c, using Equation (5):

$${\mathsf{\sigma }}_{\mathrm{c}}={\displaystyle \frac{{\mathrm{F}}_{\mathrm{m}\mathrm{a}\mathrm{x}}}{\mathrm{A}}}$$

(4)

$${\mathrm{E}}_{\mathrm{c}}={\displaystyle \frac{\mathsf{\Delta }\mathsf{\sigma }}{\mathsf{\Delta }\mathsf{\epsilon }}}$$

(5)

where σ_c = compressive strength (MPa); E_c = compressive modulus of elasticity (MPa); F_max = maximum applied load (N); A = cross-sectional area of the specimen under compression (mm²); Δσ = variation in stress within the linear region (MPa); Δε = corresponding variation in strain (dimensionless).

Fracture toughness was measured using notched specimens in flexural configuration, as per ISO 12135 [31]. The test span was set to 120 mm. The critical load required to initiate crack propagation was used in conjunction with specimen geometry to calculate fracture toughness. Although originally developed for metallic materials, this standard has also been applied to cementitious composites with appropriate adaptations, as seen in previous studies [7,32]. All calculations were performed using OriginPro 2022 (version 9.9.0.225), which was also used to apply correction factors and determine notch depth and geometry parameters. Following specimen fracture, selected samples were subjected to morphological evaluation using optical microscopy (OM). The analysis was conducted with a Dino-Lite PRO series microscope (AnMO Eletronics Corporation—Hsinchu City, Taiwan).

2.5. Characteristics of the Raw Materials

Figure 1 shows the morphological, chemical, and structural characterization of the sisal fiber used as reinforcement in this study. In Figure 1a, the optical micrograph reveals a rough and irregular surface, which can improve mechanical interlocking with the cementitious matrix. This topography is characteristic of natural lignocellulosic fibers and contributes to better anchoring within the matrix. In Figure 1b, the FTIR spectrum highlights the presence of functional groups typical of cellulose, hemicellulose, and lignin. The broad absorption band around 3340 cm⁻¹ corresponds to O–H stretching from hydroxyl groups, confirming the hydrophilic nature of the fiber [33]. The bands near 2900 cm⁻¹ refer to C–H stretching in methyl and methylene groups, while the peak around 1735 cm⁻¹ is associated with C=O stretching, mainly from hemicellulose [34]. Bands in the region between 1030 and 1160 cm⁻¹ are related to C–O stretching and C–O–C linkages in polysaccharides, confirming the presence of holocellulose [34]. The presence of these functional groups corroborates the fiber’s hydrophilic nature, which affects the water demand of cementitious mixtures [35]. These chemical groups are directly related to the high water-absorption capacity observed for the fiber–cement composites, as discussed later.

Figure 1c presents the semi-quantitative chemical composition of the sisal fiber, showing a high holocellulose content (71.03%), moderate lignin (13.01%), and low ash content (1.34%). These values are in agreement with those reported in the literature for sisal fibers and justify their selection in the present work due to their balance between strength, availability, and sustainability. To better understand the role of each component used in the formulation of the composites,

Table 2. Summary of raw material characteristics.

	Density (g/cm3)	Main Compounds	Size	Function in Composite
Cement	~3.15	Ca, Si, Al and Fe	D50 ≈ 13 µm	Main binder and source of C–S–H
Limestone	~2.70	CaCO3 with high Ca content	D50 ≈ 25 µm	Filler that reduces cement use and overall cost
Hydrophilic nanoclay	≤2.60	Montmorillonite	≤25 µm	Pozzolanic additive
Functionalized nanoclay	≤2.60	Montmorillonite + OA/APTES	≤20 µm	Pozzolanic additive
Untreated sisal fiber	~1.45	Cellulose, hemicelluloses, and lignin	2 cm length and 100–300 µm diameter	Reinforcement

summarizes the main physical and chemical characteristics of the raw materials, which were reported by each respective supplier.

However, the presence of the hemicellulose and hydroxyl groups increases the susceptibility of the fibers to alkaline degradation, especially under autoclave curing conditions. This explains the need for chemical treatments such as furfurylation, which are intended to reduce hydroxyl group activity and improve durability within the cementitious matrix.

Figure 2 presents the thermal and particle size characterization of the cement and agricultural limestone used in the formulations. In Figure 2a, the thermogravimetric analysis (TGA) curves of the cement CPV-ARI display mass loss stages corresponding to the thermal decomposition of each component. The cement sample exhibits three main decomposition regions: the first below 150 °C, associated with the evaporation of free water and the dehydration of C–S–H gels; the second between 400 and 500 °C, attributed to the decomposition of calcium hydroxide (Ca(OH)₂); and the third between 600 and 750 °C due to the decarbonation of calcium carbonate (CaCO₃). These thermal events are typical of Portland cement and confirm its purity and hydration potential. The limestone, in turn, shows a predominant weight loss starting around 650 °C, indicating the decarbonation of CaCO₃ as its main thermal event. These observations are consistent with findings by Tawfik and Abd-El-Razik [36]. This confirms its high calcium content and its potential as a partial replacement for cement in the mixtures.

Figure 2b shows the particle size distribution obtained by laser diffraction. The cement exhibited a finer distribution with a median particle diameter (D50) of approximately 13 µm, while the agricultural limestone showed a broader distribution and larger particles, with a D50 of approximately 25 µm. The smaller particle size of the cement contributes to a higher surface area for hydration reactions, which is essential for early-age strength development. On the other hand, the coarser limestone particles provide a filler effect, improving the packing density of the paste and promoting matrix densification. The difference in granulometry between the two materials justifies the observed changes in rheology and physical properties of the composites, especially regarding flowability and porosity, as discussed in later sections. Moreover, the particle size contrast plays a significant role in the mechanical synergy between the binder components. When compared to the nanoclays used, both with particle sizes ≤ 25 µm and ≤20 µm, it is clear that the nanoclays fall within or below the size range of cement and limestone. Their ultrafine nature and high specific surface area allow them not only to fill voids between larger particles but also to participate in pozzolanic reactions, especially in the case of the functionalized nanoclay.

3. Results and Discussion

3.1. Physicochemical Evaluation of the Treated Sisal Fibers

The resistance of sisal fibers to alkaline hydrolysis was evaluated for untreated fibers, fibers treated with FAL, and fibers exposed to nanoclay, preserving the proportions used in the composites. The FTIR results are shown in Figure 3. For comparative purposes, the spectra of the untreated fiber and pure nanoclay were also included.

In lignocellulosic materials subjected to alkaline hydrolysis, structural alterations often manifest through variations in the intensity and shape of the C–H deformation bands between 1300 and 1500 cm⁻¹. This degradation process can promote the transformation of aldehyde and ketone groups into carboxylic acids (–COOH), giving rise to a new C=O stretching band around 1600 cm⁻¹. Simultaneously, shifts in the C–O (1000–1300 cm⁻¹) and C=C (1500–1600 cm⁻¹) regions further signal chemical modifications in the fiber’s structure [37].

The introduction of aliphatic groups from FAL results in increased intensity of the C–H stretching bands (2800–3000 cm⁻¹). A distinctive absorption band also emerges near 1600 cm⁻¹, attributed to ketone groups from FAL and maleic anhydride. Alterations in the C=C stretching band in this region further confirm the fiber’s chemical transformation. The polymerization of FAL is supported by the appearance of a pronounced carbonyl peak near 1600 cm⁻¹ [38].

Absorption bands around 2900 cm⁻¹ are associated with methyl (C–H) stretching, while those near 1350 cm⁻¹ reflect the bending vibrations of the C–H bonds in natural fibers. The literature reports that these bands typically intensify after alkaline treatment due to lignin and hemicellulose removal [39]. In the present study, however, such intensification was not observed, suggesting that both the furfurylation treatment and nanoclay incorporation were effective in mitigating chemical degradation and preserving the fiber structure. Additionally, bands attributed to the aromatic vibrations of lignin and hemicellulose—around 1500 and 1450 cm⁻¹—were observed with reduced intensity [39]. The broad absorption between 3000 and 3700 cm⁻¹, characteristic of hydrogen bonding, confirmed the widespread presence of hydroxyl groups in all compositions.

The presence of polymerized FAL is further evidenced by the formation of diketone structures, as indicated by the distinct carbonyl absorption near 1700 cm⁻¹, consistent with the findings of Pranger et al. [38]. An intensified peak at 1600 cm⁻¹ also supports the presence of 2,5-disubstituted furan rings, reinforcing the successful incorporation of FAL into the fiber matrix [40].

Alkali-treated fibers exhibited spectra with similar peak positions but significantly reduced intensities, especially around 3000 cm⁻¹, 1500 cm⁻¹, and 1625 cm⁻¹. The near disappearance of these peaks suggests substantial degradation of aromatic compounds such as lignin, hemicellulose, and pectins [37]. Compared to the untreated fiber, the reduction in O–H and C–H stretching intensities implies the cleavage of hydroxyl bonds. The broad band from 4000 to 3000 cm⁻¹ further confirms the presence of hydroxyl groups in carbohydrate structures [41]. The peak observed at 1800 cm⁻¹ corresponds to C–O stretching in acetyl groups from hemicellulose. Interestingly, its intensity remained unchanged after NaOH treatment, indicating that the alkali treatment did not significantly reduce the hemicellulose content in the sisal fibers. Peaks between 1000 and 1050 cm⁻¹ are related to hydroxyl group stretching in cellulose and β-glycosidic bonds—features that persisted even after chemical modification, suggesting that such treatments did not compromise these components [42].

For nanoclay, prominent peaks at 3500 and 3200 cm⁻¹—associated with hydroxyl groups and hydrogen bonding—were detected in the composite spectra. This suggests that nanoclay interacted with the cementitious matrix, possibly forming hydrogen bonds with water or other components, which could enhance adhesion and improve the composite’s mechanical performance [41]. Altogether, the chemical treatments appear to have protected the fiber structure effectively. However, when comparing all spectra, signs of degradation due to the alkaline solution were evident, and no clear protective effect could be attributed to either the nanoclay or furfurylation under the specific test conditions.

3.2. Characteristics of the Fiber–Cement Composites

Figure 4 presents the flow diameter of the various fiber–cement pastes measured on the flow table, allowing the evaluation of their workability under a constant water-to-cement ratio. The reference CPV-ARI cement paste (CON) shows a flow diameter of 264.67 mm, while the incorporation of 2 wt.% untreated sisal fiber (Control) led to an increase of approximately 9% in flow diameter. This unexpected gain in workability may be related to the slight deflocculation of particles due to fiber interaction, as also observed by Aramburu et al. [20].

The addition of 50% furfurylated fiber (FAL50) resulted in the highest spread, with a flow diameter of 282.50 mm, representing a 6.7% increase compared to the control paste. This suggests that furfurylation improved fiber dispersion and compatibility with the matrix, likely due to the reduction in polar hydroxyl groups and increased hydrophobicity. However, when 1 wt.% of nanoclay was added to the FAL50 system (NC1FAL50), the flow diameter slightly decreased to 266.50 mm, indicating minimal interference of nanoclay in this context. This aligns with Hisseine et al. [43], who reported reduced fluidity in cement pastes containing nano-additives. The PGV nanoclay likely absorbs free water, reducing workability.

In contrast, pastes with 5 wt.% nanoclay (NC5 and NC5FA50) showed reductions of 21% and 23%, respectively, compared to the control paste, evidencing a marked loss of fluidity. This behavior is associated with the high specific surface area of the nanoclay and its strong water-absorption capacity, which increases the internal friction and reduces paste mobility—a phenomenon also reported by Kunchariyakun et al. [44].

Interestingly, the paste with functionalized nanoclay (FNC1FAL50) also showed a reduced flow diameter compared to the FAL50 and NC1FAL50 groups, though the decrease was less pronounced than in NC5. This may be due to better particle dispersion promoted by the organosilane groups in the modified clay, which mitigates some of the water demand and agglomeration effects.

The control paste with fibers delayed the setting, a behavior corroborated by Onuaguluchi et al. [45], who reported that fibers slow hydration initially but can accelerate it later. Jiao et al. [46] explained that hydroxyl and carboxyl groups in fibers form complexes with Ca²⁺, decreasing the number of active hydration sites and delaying early C–S–H formation.

Figure 5 shows the physical properties of the composites after autoclave curing, including water absorption, apparent porosity, and bulk density. In general, the incorporation of untreated sisal fibers increased both water absorption and porosity compared to the cement-only reference (CON). Specifically, the control paste containing 2 wt.% untreated fiber showed a water absorption increase of 11.2% and porosity increase of 10.3%, which is consistent with the intrinsic hydrophilicity and irregular geometry of the fiber, which induces void formation during mixing and curing.

The incorporation of 1 wt.% nanoclay (NC1) further increased water absorption and porosity, reaching values 13% and 23% higher than for the control, respectively. This behavior may be attributed to partial agglomeration and inefficient dispersion of the nanoclay, which limited its ability to fill the micropores effectively and instead contributed to capillary porosity. However, the group with 5 wt.% nanoclay (NC5) exhibited reduced porosity and water absorption compared to NC1, indicating a saturation effect in which the increased nanoclay content led to better pore filling and matrix densification. This phenomenon aligns with the findings of Hisseine et al. [43], who reported that an optimal threshold exists for nanoclay addition, beyond which its pozzolanic activity becomes more effective.

The FAL50 group showed 25% and 33% increases in absorption and porosity, respectively, consistent with findings in fiber–cements with untreated natural fibers [47,48]. This result suggests that the furfurylation process, although beneficial for chemical stability, may have increased the surface roughness or micro-cracking of the fiber during curing, which contributed to higher moisture uptake. Nonetheless, when nanoclay was added to the furfurylated systems (NC1FAL50 and NC5FAL50), these values decreased and approached those of the control group. The NC5FAL50 group, in particular, exhibited a 22% reduction in porosity compared to FAL50, evidencing the synergistic sealing effect of nanoclay on the porous matrix.

Interestingly, the FNC1FAL50 composite—which combines furfurylated fibers and functionalized nanoclay—showed one of the lowest water-absorption values among all the modified groups, 7.8% lower than those of the control. This suggests that the functionalized nanoclay not only reacted pozzolanically but also promoted superior particle dispersion and fiber encapsulation, contributing to a more impermeable and compact matrix.

As for bulk density, most modified composites showed slightly lower values compared to the control, particularly the groups with isolated fiber or nanoclay addition. The exceptions were NC5 and NC5FAL50, which maintained densities similar to the control group due to better packing and fewer voids. The FNC1FAL50 group also demonstrated a relatively high density among the modified composites, confirming that functionalization improved the filler efficiency of the nanoclay and its integration into the matrix.

Figure 6 presents the stress–strain curves obtained from the three-point bending tests for the different fiber–cement composites. All curves display the expected behavior for fiber-reinforced cementitious materials: an initial linear elastic region, a peak stress point, and a post-peak softening region associated with progressive fiber pull-out and rupture. The control composite (CON) showed a modest post-peak response, indicative of limited ductility and energy absorption, despite the presence of untreated sisal fibers. This matches the mechanisms described by Claramunt et al. [49], where fiber bridging delays crack propagation.

The addition of 1 wt.% nanoclay (NC1) did not significantly alter the shape of the stress–strain curve but resulted in a slightly higher stress level in the linear region, suggesting improved stiffness. In contrast, the NC5 composite exhibited a more brittle response, with a sharper drop in stress after the peak. This behavior is consistent with the saturation mechanism discussed earlier, in which excess nanoclay may lead to agglomeration and ineffective stress transfer across the matrix.

The incorporation of furfurylated fibers (FAL50) resulted in a broader and more pronounced post-peak region, with higher strain values compared to CON, indicating an enhancement in toughness and ductility. This effect became even more significant in the NC1FAL50 composite, which showed the most extended plastic region among all the samples. The curve of NC1FAL50 suggests that the chemical treatment improved fiber–matrix interfacial bonding, enabling better stress redistribution and delaying catastrophic failure.

On the other hand, the pre-treatment group (NC1FAL50P) displayed a much steeper drop in stress after the peak, with reduced strain capacity, reflecting that furfurylation conducted before the composite mixing stage may have negatively affected fiber dispersion or flexibility. This result highlights the importance of the timing and method in applying chemical treatments to natural fibers.

The NC5FAL50 group also exhibited an extended post-peak response, though slightly less than that of NC1FAL50. This composite still benefited from the combined effects of pozzolanic activity and fiber modification, indicating synergy between the two strategies. Most notably, the FNC1FAL50 composite revealed one of the most favorable profiles, combining moderate peak stress with a prolonged post-peak response, which is characteristic of materials capable of sustaining loads beyond initial cracking.

These findings are consistent with studies by Claramunt and Ardanuy [5,49], which reported similar ductile behavior in composites with properly treated fibers, and by Lima et al. [50], who observed extended strain capacity due to gradual fiber pull-out. Overall, the results from Figure 6 demonstrate that the combination of furfurylation and nanoclay incorporation—particularly when functionalized—effectively enhances the ductile performance of fiber–cement composites under flexural loading.

Figure 7 presents the flexural performance of the composites through three key parameters: (a) flexural strength, (b) maximum deflection, and (c) flexural toughness. Together, these data allow a comprehensive evaluation of the influence of nanoclay content and fiber chemical treatment on mechanical behavior under bending loads.

In Figure 7a, the flexural strength results show that the composites NC5, NC5FAL50, and NC1FAL50 exhibited significant improvements compared to the control (CON). NC5 achieved an increase of 19.3%, NC5FAL50 increased by 12.84%, and NC1FAL50 by 15.72%. These gains are attributed to the pozzolanic and filler effects of nanoclays, especially when used at 5%, as they enhance matrix densification and improve interfacial adhesion. The superior performance of NC1FAL50, despite having only 1% nanoclay, reinforces the role of the furfurylation treatment in strengthening fiber–matrix bonding. Conversely, the NC1FAL50P composite—where the fiber was furfurylated prior to incorporation—did not show the same gains, indicating that premature treatment may have affected fiber integrity or interaction with the cement matrix.

In Figure 7b, the maximum deflection results highlight the ductility of the composites. The NC1FAL50 sample presented the highest deflection, approximately 17% greater than that of the control, demonstrating enhanced deformability prior to rupture. This behavior reflects the synergistic effect of nanoclay and in situ furfurylation on fiber preservation and stress redistribution. In contrast, the NC5 and NC1FAL50P groups exhibited the lowest deflections, both showing reductions around 20% compared to the control. These results align with the hypothesis of nanoclay saturation and premature stiffening caused by excess additive or pre-treatment, respectively. The FNC1FAL50 sample displayed a high deflection value, similar to that of NC1FAL50, confirming the efficacy of combining functionalized nanoclay with furfurylated fibers.

In Figure 7c, the toughness values—calculated from the area under the stress–strain curve—show the clearest distinction between groups. The NC5FAL50 composite demonstrated a remarkable increase in toughness, nearly doubling the control value (approximately +100%), indicating an exceptional capacity for energy absorption prior to failure. This result underscores the synergistic role of nanoclay and fiber modification when both are properly dosed. NC1FAL50 showed a moderate gain, while FAL50 alone and NC5 showed toughness losses, confirming that neither furfurylation nor nanoclay addition alone was sufficient to significantly enhance this property. The FNC1FAL50 sample again exhibited a robust response, with toughness values around 45% higher than those of the control, validating the strategy of combining pozzolanic and chemical interventions.

Figure 8 shows the stress–deflection curves obtained from the compressive strength tests of the fiber–cement composites. All the curves follow the expected pattern for cementitious materials reinforced with natural fibers, displaying an initial linear elastic region followed by a peak stress, a softening phase, and, finally, failure. In fact, this shape is common in fiber–cements with short fibers, as described by Pereira et al. [51]. However, distinct differences in curve shapes and slopes allow for the qualitative assessment of stiffness and ductility across the experimental groups.

The control (CON) composite exhibited a typical sharp peak followed by a steep stress drop, indicating brittle behavior and low energy absorption after the yield point. In contrast, composites containing furfurylated fibers, particularly NC1FAL50 and FNC1FAL50, showed broader post-peak regions and more gradual stress reductions, suggesting enhanced plastic deformation and increased resistance to crack propagation. These groups also exhibited a more extended plateau, indicative of better stress redistribution provided by the treated fibers.

The NC1 composite presented the highest initial slope among all the curves, reflecting its superior compressive modulus and stiffness, likely a result of the optimal pozzolanic activity of 1% nanoclay. Conversely, NC5 showed a less pronounced slope and a more abrupt failure, likely associated with excess nanoclay saturation, agglomeration, and possible interference with proper cement hydration.

Notably, the NC1FAL50P composite showed a relatively brittle curve, with reduced ductility and abrupt fracture after the stress peak. This behavior indicates that the pre-treatment of the fiber—although chemically protective—may have compromised its structural integrity or interaction with the cement matrix. FAL50 and NC5FAL50 also showed moderately ductile behavior, although less pronounced than NC1FAL50 and FNC1FAL50, reinforcing the role of functionalized nanoclay in preserving fiber efficiency during compressive loading.

Overall, the stress–strain curves from the compressive testing suggest that the best combination of stiffness and deformability was achieved by composites containing 1% nanoclay and furfurylated fibers added in situ, with the FNC1FAL50 and NC1FAL50 samples showing the most favorable mechanical response. These results are consistent with prior findings in the literature on natural fiber–cement systems and confirm the beneficial synergistic effect of combining surface treatment with nanostructured pozzolanic additives.

Figure 9a displays the compressive strength results of the fiber–cement composites. The control and NC1 groups achieved the highest average values, around 23 MPa, indicating that the addition of 1% nanoclay contributed positively to compressive strength. This matches findings by Melo Filho et al. [52] and Krishna et al. [53], who observed strength reductions with poorly dispersed or degraded fibers. In contrast, the FAL50 and NC5 groups experienced reductions of approximately 13% and 17%, respectively, compared to the control. This reduction can be associated with either the partial substitution of cement by furfurylated fiber content (in FAL50) or the excess nanoclay in NC5, which may have caused particle agglomeration and impaired cement hydration due to a saturation effect. These phenomena lead to poor dispersion and inefficient pozzolanic reactions, corroborating the literature that warns of mechanical impairments in systems with excessive nanostructured pozzolanic materials.

Interestingly, the NC1FAL50 composite maintained compressive strength values statistically similar to those of the control, suggesting that the negative effect of fiber addition was mitigated by the synergistic interaction between the 1% nanoclay and the pozzolanic filler effect. However, the pre-treated NC1FAL50P group exhibited a strength drop of approximately 21%, likely due to damage caused to the fiber structure during furfurylation prior to mixing. The NC5FAL50 group also showed a decline, reinforcing the idea that 5% nanoclay oversaturates the matrix, limiting hydration and weakening the mechanical integrity of the composite. The FNC1FAL50 sample, incorporating functionalized nanoclay and in situ-treated fibers, presented a balanced result close to that of the control, indicating that this combination may prevent significant degradation of compressive performance.

Several studies corroborate the present compressive strength range. For instance, Kunchariyakun et al. [44] achieved similar strength values using autoclaved composites reinforced with untreated natural fibers. Similarly, Choi and Choi [54] reported compressive strengths around 25 MPa in cement matrices incorporating 2% natural fiber content. Pereira et al. [51] observed a slight strength reduction (~3 MPa) with short sisal fibers (25 mm), yet the overall structural integrity was preserved across a broader deformation range. These comparisons validate the consistency of the current results and the viability of these composites for structural applications.

Figure 9b shows the compressive modulus (modulus of elasticity in compression) of the same composites. The NC1 sample stood out with the highest modulus, showing a 174% increase compared to the control, confirming the efficiency of 1% nanoclay in enhancing rigidity through pore refinement and better packing. The NC1FAL50 composite followed with the second-highest modulus, suggesting that the nanoclay remained effective even in the presence of furfurylated fibers. Although the FAL50 sample displayed higher deformability, its stiffness was lower than that of the control, likely due to the reduction in the cement fraction and the lower modulus of the fiber material.

In contrast, the NC5FAL50 and NC1FAL50P composites showed the lowest stiffness among all the groups, suggesting that either excess nanoclay or aggressive fiber pre-treatment impaired the structural integrity of the matrix. This decrease in modulus reinforces the concept of an optimal nanoclay dosage threshold, beyond which the expected filler and pozzolanic effects are no longer beneficial. Overall, the data demonstrate that the judicious combination of nanoclay and fiber treatments can significantly affect compressive behavior, especially stiffness, which is critical for load-bearing applications.

Figure 10 illustrates the fracture toughness results of the fiber–cement composites, which reflect the material’s ability to resist crack propagation under stress. Among all groups, the FAL50 composite exhibited the highest toughness value, surpassing the control by approximately 23%. This indicates that the presence of furfurylated fibers—despite partially replacing cement—was effective in increasing energy absorption during crack initiation and propagation. The enhanced performance may be attributed to the improved fiber–matrix interface achieved through furfurylation, which likely limited the interfacial debonding and allowed greater fiber bridging during fracture.

On the other hand, composites containing only nanoclay (NC1 and NC5) did not exhibit significant improvements in fracture toughness compared to the control, suggesting that nanoclay alone, whether at 1% or 5%, did not meaningfully contribute to energy dissipation under fracture. This result highlights that while nanoclay may refine the microstructure and enhance stiffness, it does not directly enhance fracture energy unless paired with mechanisms like fiber bridging.

Notably, the NC1FAL50 and NC1FAL50P composites showed lower toughness values than FAL50, and even lower than the control in the case of NC1FAL50P, indicating that the expected synergistic benefits of combining nanoclay and furfurylated fiber were not fully realized in this property. In the pre-treated group (NC1FAL50P), the decreased toughness may be due to over-hardening or embrittlement of the fibers caused by the ex situ furfurylation process, which could have compromised their ability to deform and arrest cracks. Similarly, the NC5FAL50 composite exhibited a modest toughness gain, suggesting that excessive nanoclay may hinder effective fiber–matrix interaction by saturating the system or causing particle agglomeration, thus limiting fiber activation during crack propagation.

Interestingly, the FNC1FAL50 composite, which incorporated functionalized nanoclay and in situ-treated fibers, demonstrated a toughness level close to that of the control. This suggests that while the treatment and filler may have preserved the fiber structure and microstructural integrity, the chemical modifications might have reduced the number of active sites for energy absorption mechanisms (e.g., fiber pull-out or rupture). Therefore, although this formulation proved promising in other mechanical aspects, its contribution to fracture resistance was neutral.

A possible explanation, supported by Melo Filho et al. [52], is that excessive pozzolanic material can reduce matrix deformability, limiting fiber movement during fracture and negating toughness gains. Despite these observations, all values fall within ranges reported by other studies involving natural fiber composites. For example, Choi and Choi [54] also noted increased toughness through fiber reinforcement, though they stressed the need for careful control of the interface chemistry. Therefore, the fracture behavior observed here aligns well with the literature findings and further supports the potential of fiber treatments like furfurylation when properly integrated into the composite system.

Figure 11 illustrates the microstructural differences among selected fiber–cement composites, emphasizing the effects of fiber treatment and nanoclay incorporation on fiber dispersion, interfacial adhesion, and matrix densification. In the control composite (CON), the image reveals poorly distributed fibers with visible interfacial gaps and voids, suggesting limited fiber–matrix interaction and inadequate encapsulation. These morphological features are consistent with the mechanical results, where the control group displayed lower toughness and limited post-peak deformation capacity. For the NC1FAL50 composite, the micrograph shows a more homogeneous fiber dispersion and improved fiber alignment, along with fewer interfacial voids. This reflects the synergistic effect of 1 wt.% nanoclay and furfurylated fiber treatment, which promoted better fiber wetting and interfacial bonding. These observations align with the mechanical findings, where this group exhibited enhanced flexural strength, ductility, and a moderate increase in toughness. The FNC1FAL50 composite stands out, with a compact microstructure and fibers well integrated within the cement matrix and minimal voids or microcracks observed. The improved interaction is attributed to the surface-functionalized nanoclay, which enhances compatibility and dispersion, as well as the in situ furfurylation, which preserves fiber integrity. These microstructural characteristics corroborate the superior mechanical performance and reduced porosity noted for this formulation, reaffirming the benefits of combining targeted chemical modifications with pozzolanic nanomaterials.

4. Conclusions

This study aimed to improve the performance and durability of autoclaved fiber–cement composites by combining furfurylated sisal fibers and pozzolanic nanoclays. The incorporation of 2 wt.% furfurylated sisal fibers increased flexural strength by 15.7%, maximum deflection by 17%, and toughness by 45% compared to the untreated fiber reference. These improvements were attributed to enhanced interfacial bonding and the fibers’ greater resistance to alkaline degradation.

Among the pozzolanic additives, the functionalized nanoclay modified with octadecylamine and aminopropyltriethoxysilane (OA/APTES) yielded superior performance relative to unmodified bentonite. When used at 1 wt.% in combination with furfurylated fibers (FNC1FAL50), it resulted in one of the lowest water-absorption values (7.8% below the control), reduced porosity by 22%, and improved flexural strength and compressive modulus without compromising workability or density. Morphological analysis by optical microscopy revealed that the combination of furfurylated fibers and functionalized nanoclay led to a more compact matrix and improved fiber–matrix integration, corroborating the mechanical and physical property results. These outcomes confirm the synergistic effect between the fiber treatment and the enhanced pozzolanic and filler properties of the nanoclay.

Furthermore, the partial replacement of Portland cement by agricultural limestone—at a ratio of approximately 14%—reduced the environmental impact of the formulations without significantly affecting physical or mechanical performance. The findings support the feasibility of combining sustainable additives and fiber treatments to overcome the limitations typically associated with natural fiber–cement systems. Nevertheless, the study has some limitations, such as the absence of microstructural analysis via SEM and the restriction of curing and aging conditions to a single autoclave regime. Future investigations should focus on the following:

(i)

optimization of the furfurylation process for other lignocellulosic fibers;

(ii)

evaluation of long-term durability;

(iii)

exploration of additional pozzolanic nanomaterials to enhance interfacial compatibility;

(iv)

evaluation of workability retention over time.