Introduction

The urgency to combat climate change, prevent environmental damage, and reduce pollution has significantly heightened the focus on the need for more durable and environmentally friendly materials. In this context, scientific research has heavily invested in developing biodegradable and eco-friendly materials capable of effectively replacing non-renewable materials that threaten the ecosystem. Biocomposites utilising renewable resources such as polysaccharide materials, flax, or agricultural waste1,2, have emerged as a major area of interest due to their ability to degrade naturally. That is, they offer a sustainable alternative to traditional materials3 and open up new prospects for various applications, by combining the environmental benefits of biological materials with the technical performance needed to meet industry standards4,5 while preserving opportunities for future generations6.

In response to this challenge, numerous studies have focused on exploring and improving the properties of composites, with the aim of developing more efficient materials from natural resources. Among these materials, natural rubber Primarily derived from Hevea brasiliensis (or Gutta-percha, a natural rubber originating from South America: Brazil, Venezuela, etc.), stands out for its renewability, biodegradability, and remarkable elasticity7,8,9. Cellulose is the most plentiful natural polymer found on Earth, known for its outstanding mechanical strength, strong chemical stability, significant biodegradability, and non-toxic nature10,11.

Blanchard et al.12 conducted a study focused on enhancing the physical properties of natural rubber (NR) latex by adding cellulose nanocrystals (CNCs) alongside vulcanization agents and dispersed zinc oxide (ZnO) as an activator. The findings revealed that CNCs promoted uniform dispersion of ZnO throughout the matrix by forming Zn(II)-cellulose complexes, which boosted cross-linking efficiency and led to a steady increase in cross-linking density as the CNC content rose. The improved structural organization caused by increasing CNC content positively influenced the mechanical properties, as shown by DMA measurements of the elastic modulus, which revealed a gradual stiffening of the films. This increased stiffness was attributed to the restricted mobility of natural rubber chains resulting from the incorporation of CNCs. In films containing 0.5 phr of CNC, the glass transition temperature (Tg) of the NR chains rose from − 55 °C to − 52 °C, reflecting better filler dispersion at low concentrations. However, at higher CNC loadings, a slight decrease in Tg was detected, possibly due to a phase inversion occurring between the CNCs and rubber once the filler concentration surpassed the percolation threshold12.

Similar results were pointed out by Visakh et al.13 who reported an increase in the storage modulus of a natural rubber-based composite reinforced with 10% (w/w) of cellulose nanofibers derived from bamboo. They also observed a shift in the tan δ peak temperature from − 47 °C to − 42 °C. Oboh et al.14 investigated the dependence of the dynamic mechanical properties of these composites on the type of cellulose nanoparticles used (coconut husk, bamboo, and cotton linters), as well as the effect of incorporating carbon black into a vulcanized rubber matrix. The results showed that the composite made from natural rubber and cellulose particles from coconut husk (NR-CHNC) exhibited the highest storage modulus at − 70 °C (3619 MPa). Additionally, at + 20 °C, the cellulose-based composites showed a higher storage modulus than both pure natural rubber (neat NR) and standard rubber with carbon black (NR-CB) (2.08 MPa and 2.78 MPa, respectively). Regarding the glass transition temperature (Tg), composites containing cellulose particles showed better performance than the reference sample (pure NR) and outperformed NR-CB composites. The authors suggested that this improvement is due to increased stiffness of the rubber main chains through interactions with the cellulose particles.

Other studies15,16 have focused on the preparation of composites made from natural polymer without the use of crosslinking agents, concentrating on interfacial interactions. Sinclair and his collaborators17 carried out esterification of the surfaces of cellulose nanofibers (CNFs) and then compared the mechanical performance of natural rubber (NR) nanocomposites incorporating either raw or functionalized CNFs. While unmodified CNFs improved the mechanical properties of the material, the functionalized CNFs exhibited even greater reinforcing effects. This improvement is attributed to enhanced interfacial compatibility between the rubber matrix and the modified cellulose nanofibers (CNFs), leading to more efficient stress transfer between the phases. However, although effective, these chemical modification methods involve costly reagents and complex procedures, which can restrict their large-scale application. A simpler approach is to incorporate functional groups directly into the rubber itself. For example, epoxidized natural rubber (ENR), produced by epoxidizing NR with hydrogen peroxide and formic acid, contains epoxy groups along its polymer chain that facilitate new chemical interactions with reinforcing fillers. Tanpichai et al.18 demonstrated that the interaction between epoxidized natural rubber (ENR) and cellulose nanofibers (CNFs) primarily takes place through interfacial bonding. This improved compatibility resulted in a substantial enhancement of mechanical properties, with tensile strength increasing by up to 80% compared to pure ENR. However, the addition of CNFs had little impact on the composites’ elongation at break and thermal stability.

Soheilmoghaddam et al.19 studied films made from regenerated cellulose and epoxidized natural rubber (ENR-50) prepared by solution casting using the ionic liquid solvent 1-butyl-3-methylimidazolium chloride (BMIMCl). They observed a uniform dispersion of cellulose within the ENR matrix and noted a notable improvement in thermal stability compared to regenerated cellulose alone. This enhancement was attributed to hydrogen bonding between the epoxy groups in ENR and the hydroxyl groups in regenerated cellulose. Additionally, films containing 20% (w/w) ENR exhibited a significant elongation at break of approximately 39%.

Cao et al.20 examined the interfacial interactions between marine biomass (tunicate cellulose) and rubber in ENR-based nanocomposites with an epoxidation degree of about 40%. The authors attributed the enhanced mechanical properties to hydrogen bonding between ENR and tunicate cellulose nanocrystals (t-CNC), which facilitated a more uniform dispersion of CNCs compared to natural rubber (NR). This led to a remarkable improvement in mechanical performance, including a sevenfold increase in stress at 300% strain, a 1.7-fold rise in fracture energy, and a 57% boost in tensile strength.

In this context, the present study explores the incorporation of hydroxyethyl cellulose (HEC) into epoxidized natural rubber (ENR) to strengthen interfacial interactions between the two polymers without resorting to crosslinking or vulcanizing agents. Mechanistically, in HEC the hydroxyl groups are predominantly more mobile and more accessible than many of the –OH sites in native cellulose (often secondary and strongly engaged in intra/interchain hydrogen-bond networks). NMR/structural studies show that hydroxyethyl substitution disrupts the ordered organization of cellulose and increases functional accessibility, promoting stronger interfacial interactions with other polymers21. Consequently, HEC –OH groups interact more readily with ENR epoxide groups forming denser hydrogen-bond networks and, under appropriate conditions, even promoting limited epoxide ring opening. In parallel, hydroxyethyl side chains reduce crystallinity and improve HEC solvation/compatibility, increasing the frequency of HEC/ENR contacts. At the materials scale, these effects consolidate the interphase, improve stress transfer, and raise Tg, without relying on costly grafting steps often required for CNC/CNF systems (which, being very rigid, can reduce elongation at break if compatibilization is insufficient). The morphology, mechanical properties, and thermal stability of the resulting composites are investigated using complementary characterization techniques, and the effect of ENR content on these properties is assessed.

Materials and methods

Materials: Hydroxyethylcellulose, was supplied by Sigmaaldrich. Epoxidized natural rubber (ENR, with an epoxidation degree of 25%) was used after purification. The organic solvent tetrahydrofuran (THF), also supplied by Sigma Aldrich, and dimethylformamide (DMF) were obtained from Fluka.

Films preparation: To prepare composite films of hydroxyethylcellulose (HEC) and epoxy natural rubber (ENR), HEC is first dissolved in dimethylformamide (DMF) at a concentration of approximately 6% by weight under mechanical stirring at 60 °C, until a homogeneous solution without lumps is obtained, which takes about 3 h of stirring. In parallel, ENR is dissolved in tetrahydrofuran (THF) at 40 °C under mechanical stirring until complete dissolution for 1 h. Next, the ENR/THF solution is slowly added to the HEC/DMF solution under continuous stirring at temperature 60 °C, and the mixture is stirred for 3 to 4 h to ensure homogeneous dispersion. The homogeneous solution is then cast onto a flat glass or Teflon plate, spread evenly to achieve the desired film thickness (0.6 to 1 mm), and degassed to remove air bubbles. The film is first dried at room temperature for 48 h to allow gradual solvent evaporation, followed by heating in an oven at 80 °C for 3 h to eliminate any remaining solvent residues. After complete drying, the film is carefully removed from the mold and stored in a dry place.

Blend samples with cellulose to ENR weight ratios of 50/50, 60/40, 70/30, and 80/20 were prepared and labeled as HEC/ENR-1, HEC/ENR-2, HEC/ENR-3, and HEC/ENR-4, respectively.

Instruments

FTIR spectra were recorded at room temperature using a JASCO Fourier Transform Infrared spectrometer equipped with an attenuated total reflection accessory (JASCO-FT/IR-4700-ATR). Each spectrum was collected over 32 scans with a resolution of 4 cm−1, covering the range from 400 to 4000 cm−1. Attenuated total reflection Fourier transform infrared spectroscopy (FTIR-ATR) was employed to examine and analyze the probable chemical structures of the samples.

X-ray diffraction (XRD): measurements were taken using a Shimadzu XRD 6000 diffractometer system with Cu radiation (λ = 0.154 Å) throughout a 2θ range of 2 to 80°.

Thermogravimetric analysis (TGA): The thermal behaviour of the nanocomposite films was investigated using a TA Instruments TGA METTLER thermogravimetric analyzer at a heating rate of 10 °C/min under nitrogen. Temperatures varied from 25 to 900 °C.

Differential Scanning Calorimetry (DSC): The glass transition temperature (Tg) of pure ENR, pure HEC, and their blends was determined using a TA Instruments Q100 calorimeter. Samples of approximately 5 to 10 mg, pre-dried, were accurately weighed and sealed in standard aluminum pans. All analyses were carried out under a nitrogen atmosphere to avoid oxidation. The thermal program ranged from − 70 °C to 150 °C.

Scanning Electron Microscopy (SEM): The surface morphology of the samples was analyzed using a Hitachi S-4800 scanning electron microscope (SEM) in high vacuum (HighVac) with a secondary electron (SE) detector, at an accelerating voltage between 5 and 10 kV and a working distance of 5–8 mm. Images were acquired at magnifications of ×750 (scale bar: 50 μm) and ×2,500 (scale bar: 20 μm).

The cross-linking density of ENR/HEC composites was measured by first weighing five sample pieces (mo) and immersing them in toluene at 23 °C for 72 h to reach swelling equilibrium. The swollen samples were periodically removed, blotted with absorbent paper to eliminate excess solvent, and then weighed using an analytical balance. The cross-linking density was subsequently calculated using the Flory-Rehner equation22. :

$$\:\text{V}\text{e}\:=\:-\frac{\text{ln}\left(1-{V}_{r}\right)+{V}_{r}+x{V}_{r}^{2}}{{V}_{1}({V}_{r}^{\frac{1}{3}}-\frac{{V}_{r}}{2})}$$

where

$$\:{V}_{r}=\:\frac{\frac{{m}_{r}}{{\rho\:}_{r}}}{\frac{{m}_{r}}{{\rho\:}_{r}}+\frac{{m}_{s}}{{\rho\:}_{s}}}$$

V1: The molar volume of the solvent Toluene (V1 = 106.23 cm3/mol)20,23; X: Interaction parameter of the solvent Toluene (X = 0.393); mr and ρr are the weight and density of the rubber (for ENR, ρr = 0.94 g/cm3).

ms and ρs are the weight and density of the solvent (for toluene, ρs = 0.865 g/cm3).

Contact angle measurements: The contact angles of the water drop on the HEC/ENR films were measured using an Ossila contact angle goniometer. Uniform drops of water (5 µL) were carefully deposited on the film surface using a 5 µL syringe. Contact angles were measured as a function of time at 60-second intervals from the time the water came into contact with the film surface. Measurements for each percentage were carried out using a five-measurer.

Dynamic mechanical analysis DMA: is a fundamental tool for studying the viscoelastic properties of polymer materials18. It works by applying a sinusoidal stress to a sample and measuring the resulting strain, allowing for the calculation of values such as the storage modulus, loss modulus, and damping factor. Additionally, quasi-static monotonic tensile and compression tests are often conducted to evaluate the strength and stiffness of materials under uniaxial loads. Furthermore, fatigue tests enable the analysis of material durability when subjected to repeated loading cycles, evaluating how materials degrade over time. By examining how these parameters change with temperature, frequency, and dynamic displacement amplitude, we acquire critical insights into the thermo-viscoelastic properties of materials, such as thermodynamic transitions and relaxation effects.

Dynamic mechanical analysis (DMA) was carried out using a Metravib DMA + 300 instruments. Specimens were cut to nominal dimensions of 20 mm (height) × 8 mm (width) × 1.00 mm (thickness). Before testing, actual dimensions were measured with a digital caliper (0.01 mm resolution). For thickness, 5 points per film were taken (center + four quadrants); for height and width, 3 points per film (edges and center). Results are reported as mean ± SD as follows: thickness (t) = 1.00 ± 0.03 mm; height (H) = 20.0 ± 0.1 mm; width (W) = 8.00 ± 0.05 mm. Initially, a Freundlich-Dubinin (FD) isotherm test was performed to identify the displacement and frequency parameters relevant to the materials. We scanned a frequency range from 0.1 Hz to 10 Hz and a displacement ranging from 10− 6 m to 4 × 10− 5 m, with a static/dynamic ratio of 1.1, reflecting the viscoelastic nature of the material. This ratio helps prevent buckling in elastomeric films by balancing static and dynamic loads, ensuring the material remains stable under compressive forces and avoids instability. This analysis was conducted by applying a certain number of cycles before measurement to ensure that the sample reached a state where its mechanical properties were constant.

Subsequently, we performed temperature sweep tests for each sample with a dynamic displacement of 1 × 10− 6 m and a frequency of 1 Hz, in a temperature range from − 50 °C to 40 °C, using air chiller system for cooling. Additionally, tensile, and fatigue tests were conducted under the same conditions at ambient temperature.

Results and discussion

FTIR-ATR analysis

The FTIR spectra of HEC, ENR, and their blends are shown in the figure. In the FTIR spectrum of HEC, a broad band was observed at 3380 cm− 1, attributed to the stretching vibrations of the hydroxyl group in the HEC structure24,25. Additionally, a medium absorption band, ranging from 2927 to 2873 cm− 1, is attributed to the C-H stretching vibration26. The band at 1640 cm− 1 is associated with the deformation vibration characterizing the water naturally absorbed in the HEC structure23,27. As for the absorption bands of the O-H deformation of the primary alcohol and the symmetric bending of C-H in -CHOH-, they are located at 1458 and 1408 cm− 1, respectively28. The absorption peak at 1120 cm− 1 is attributed to the antisymmetric vibration of C-O. The C-O-C stretching vibration in glucopyranose is situated at 1062 cm− 1, and the absorption band of the β-(1,4) glycosidic linkage is observed at 887 cm− 123,25. For ENR, the FTIR spectrum exhibits distinctive peaks consistent with those reported in various publications9,29. The significant absorption peaks observed at 2923 cm− 1 and 2856 cm− 1 are respectively assigned to the asymmetric and symmetric stretching vibrations of the methyl group (–CH3) and the methylene group (–CH2). Additionally, the peaks located at 1659 and 836 cm− 1 correspond to the carbon-carbon double bond (C = C)30 and the bending of = C-H, respectively. The bending peaks of the CH group are found at 1377 and 1453 cm− 1. As for the characteristic peaks at 872 cm− 1 and 830 cm− 1, they correspond respectively to the asymmetric and symmetric vibrations of the epoxy ring. The FTIR analysis of the HEC/ENR blends reveals the simultaneous presence of characteristic bands from both HEC and ENR. Notably, in the region around 3400 cm−1, associated with the stretching vibrations of O–H bonds (Fig. 1a), a shift toward lower wavenumbers are observed, along with significant broadening of the band as reported in Table 1. This phenomenon indicates an intensification of hydrogen bonding, reflecting a strong affinity between the hydroxyl groups of HEC and the epoxy rings of ENR. A magnified view of this spectral region (Fig. 1b) further shows that as the ENR content increases, the O–H band shifts more noticeably, reinforcing the hypothesis of increasing interaction between the two polymers. The poxide-ring region (≈ 870–850 cm−1). In Fig. 1c the circled band assigned to ENR epoxide-ring vibrations shows a gradual decrease in intensity from pure ENR to the HEC/ENR-2 and HEC/ENR-4 blends. Figure 1d shows an enlarged view of the 900–820 cm−1 range (epoxide-ring vibration domain) to improve the readability of the changes highlighted in Fig. 1c. This attenuation indicates increasing involvement of epoxide groups in specific intermolecular interactions with HEC –OH groups. Specifically, the hydroxyl groups can form hydrogen bonds with the oxygen atoms of the epoxy rings. Although these interactions do not necessarily lead to complete ring opening, they facilitate the formation of intermolecular bridges between polymer chains, thereby enhancing the compatibility and cohesion of the HEC/ENR system.

Table 1 Quantitative analysis of the O–H stretching band for HEC/ENR blends.
Full size table
Fig. 1
figure 1

FTIR spectrum of HEC, ENR and HEC/ENR.

Full size image

Structure of HEC/ENR composites

Based on the previous analyses, interfacial interactions between HEC and the ENR matrix are mainly formed through hydrogen bonding. These interactions can occur via several pathways, including the formation of hydrogen bonds between the hydroxyl groups (–OH) of HEC and the epoxy groups of ENR, or through a chemical reaction where the epoxy ring opens20,30. In this reaction, the –OH group acts as a nucleophile, attacking an electrophilic carbon atom of the oxirane ring, resulting in a covalent bond between HEC and ENR, as depicted in Fig. 2.

Fig. 2
figure 2

Illustration of the proposed structure of the HEC/ENR composite.

Full size image

These hydrogen bond networks enhance structural cohesion and improve interfacial adhesion between the two materials. Additionally, chemical interactions between the functional groups of ENR and HEC can induce structural modifications within the polymer matrix, directly influencing its mechanical and thermal properties. Finally, the homogeneous dispersion of HEC in the ENR matrix, facilitated by these interfacial interactions, significantly contributes to the overall performance enhancement of the composite18.

XRD analysis

The interaction between HEC and 25% ENR was evaluated through structural changes observed in the HEC matrix using X-ray diffraction (XRD). According to Fig. 3, which displays the diffractograms of pure HEC and HEC films blended with 25% ENR, the diffractogram of pure HEC confirms its predominantly amorphous structure, characterized by a broad halo centered around 2θ ≈ 20°, with low-intensity features near 2θ = 8.39° and 38.4°. These are sometimes attributed to short-range ordered regions corresponding to the (110), (020), and (004) planes31,32. Upon blending HEC with 25% ENR, the diffractograms show noticeable changes in peak intensity. Specifically, the intensities at 2θ = 8.39° and 20.75° decrease compared to those of pure HEC, suggesting a reduction in the degree of structural ordering. This indicates that the presence of ENR—an amorphous elastomer—disturbs the local packing of HEC chains and hinders the formation of ordered domains.

The decreased crystallinity may be attributed to two main factors: the inherent amorphous nature of ENR disrupting any semi-ordered regions in the HEC matrix, and the formation of intermolecular hydrogen bonds between HEC and ENR, which interfere with the native intermolecular interactions responsible for ordering in pure HEC14.

Fig. 3
figure 3

X-ray diffractogram of HEC, ENR and HEC/ENR.

Full size image

Scanning electron microscopy (SEM)

Scanning electron microscopy (SEM) was used to examine the surface morphology of pure epoxidized natural rubber (ENR) (Fig. 4a) and the HEC/ENR composite film with a 60/40 mass ratio (Fig. 4b). The SEM image of pure ENR reveals a generally rough and disordered surface, characterized by protrusions and poorly defined amorphous domains, which are typical of elastomeric polymers lacking reinforcement or significant intermolecular interactions. In contrast, the image of the HEC/ENR film displays a much smoother and more homogeneous surface, with no visible phase separation at the microscopic scale. This uniformity suggests good compatibility between the two polymers, likely facilitated by hydrogen bonding between the epoxy groups of ENR and the hydroxyl groups of hydroxyethylcellulose. The fine and dense microstructure observed in the composite film indicates a favorable interaction between the phases. These findings confirm the structuring effect of HEC within the elastomer matrix and support the conclusions of L. Cao et al.20, who demonstrated that t-CNCs disperse effectively within the ENR matrix, forming a homogeneous structure without significant aggregates due to strong hydrogen bond interactions. Such uniform dispersion is crucial for enhancing the properties of the resulting nanocomposites.

Fig. 4
figure 4

SEM images of ENR (a); HEC/ENR-2 (b).

Full size image

TGA

Fig. 5
figure 5

Courbes TGA pour l’HEC, l’ENR et les composites HEC/ENR.

Full size image

The grafting of hydroxyethyl cellulose (HEC) onto epoxidized natural rubber (ENR) notably affects the thermal stability of the resulting composites. This effect was examined using thermogravimetric analysis (TGA). The thermograms shown in Fig. 5 display the thermal stability profiles of pure ENR, pure HEC, and their composite materials.

During the first stage, between room temperature and 150 °C, a slight mass loss is observed for all samples. This is attributed to the evaporation of water physically absorbed on the surface and residual solvents. For HEC alone, a second stage, corresponding to a more significant mass loss, occurs between 200 °C and 300 °C. This primary degradation is due to the decomposition of the polymer chains of HEC18 and ends between 400 °C and 450 °C with complete combustion.

ENR alone exhibits a primary degradation between 300 °C and 350 °C, followed by final combustion around 450–500 °C, demonstrating better thermal stability compared to HEC. The composites formed by the association of ENR and HEC exhibit an intermediate behavior, combining the properties of both materials. HEC is stabilized by ENR20,33, as reflected by a similar initial water loss (room temperature − 150 °C), but with a primary degradation temperature ranging from 250 °C to 350 °C, depending on the respective proportions of HEC and ENR Table 2.

In conclusion, ENR is the most thermally stable, while the ENR/HEC composites offer a balance between rigidity (provided by HEC) and thermal stability (contributed by ENR).

Table 2 TGA—temperatures at 5% and 50% mass loss for the samples (HEC, ENR, HEC/ENR).
Full size table

Crosslinking density

In order to further demonstrate the crosslinking of ENR with HEC, the crosslinking density was evaluated by analyzing the swelling of ENR/HEC films in toluene solvent, using the Flory-Rehner Eqs18,34. based on the proportions of HEC. Figure 6 illustrates the significant swelling of ENR/HEC nanocomposites in toluene, as well as an increase in crosslinking density compared to pure ENR. This increase is slightly amplified as the HEC percentage rises. For instance, in the HEC/ENR-4 formulation (80%/20%), the crosslinking density reaches 0.047 mmol/cm³.

This increase in crosslinking density is attributed to strong interfacial interactions, particularly hydrogen bonding between the hydroxyl groups of HEC and the epoxy groups present in ENR. These interactions provide greater rigidity to the material, as also demonstrated by Cao.L et al.20 in their work, which confirmed that cellulose associated with ENR exhibits an increased crosslinking density due to hydrogen bond interactions.

Furthermore, this effective interaction restricts the expansion of ENR/HEC composites in toluene by reducing their swelling capacity. It also ensures a strong bond between the polymer chains, resulting in improved structural cohesion.

It is noteworthy that crosslinking density plays a crucial role in determining the mechanical and thermal properties of materials. A higher crosslinking density can enhance tensile strength, reduce film permeability, and improve thermal stability35,36. These properties open up potential applications in fields such as sustainable packaging, selective membranes, and functional coatings, where materials that are both durable and adaptable are required.

Fig. 6
figure 6

Crosslinking density neat ENR and HEC/ENR composites.

Full size image

Dynamic mechanical analyses

Dynamic mechanical analyses were conducted on pure ENR 25% and on nanocomposites incorporating HEC at various mass fractions. The curves obtained for the storage modulus (E′) and loss tangent (tan δ) as a function of temperature (Fig. 7) reveal that, at low temperatures, ENR 25% remains in a glassy state. In this state, the storage modulus (E′) is relatively stable at approximately (0.4 ± 0,08 GPa), reflecting a rigid structure where molecular mobility is confined to short-range vibrations and rotations.

Fig. 7
figure 7

The storage modulus of ENR and HEC/ENR composites.

Full size image

This stability arises from the fact that, in the glassy state, the molecules do not possess sufficient energy to undergo large-amplitude motions. As a result, molecular movements are confined to localized vibrations and rotations, without inducing significant structural changes. Consequently, the material retains its rigidity and exhibits resistance to deformation under mechanical stress, thereby maintaining the stability of the storage modulus (E′).

Incorporating HEC into ENR 25% leads to a significant rise in the storage modulus (E′), HEC/ENR-1, HEC/ENR-2, HEC/ENR-3, and HEC/ENR-4 composites exhibit mean storage moduli of (2.5 ± 0.08GPa), (2.6 ± 0.08 GPa), (2.7 ± 0.09 GPa), and (2.8 ± 0.10 GPa), respectively. Each value is the average of three independent tests. Of all the composites tested, the one with an 80/20 weight ratio of hydroxyethyl cellulose to ENR showed the highest storage modulus, as illustrated in Fig. 7, exceeding the storage modulus values of the other compositions.

This significant increase in E’ can be attributed to the reinforcing effect of HEC within the ENR 25% matrix. The incorporation of HEC enhances the rigidity of the composite material, even below the glass transition temperature. HEC forms physical interactions, such as hydrogen bonds and reinforcement networks, within the polymer matrix. These interactions strengthen the structural cohesion of the material and restrict the mobility of the ENR chains, leading to improved rigidity and enhanced resistance to deformation. As a result, the composite exhibits superior mechanical properties and better performance under mechanical stresses.

All samples showed a significant drop in storage modulus around − 30 °C (Fig. 7), which corresponds to the primary energy dissipation event near the glass transition temperature (Tg). This decrease in storage modulus (E′) reflects the shift from a rigid, glassy state to a more flexible, rubbery state. In the rubbery phase, the cooperative movement of long polymer chain segments becomes more noticeable. This behavior is attributed to the development of a three-dimensional network between cellulose particles and polymer chains, which limits their mobility and thereby increases the overall stiffness of the composite, as confirmed by multiple studies20,37.

The reinforcing ability of HEC particles enhances rigidity up to the glass transition temperature, beyond which a considerable drop in rigidity is observed, regardless of the mixture composition. A similar, low level of rigidity is then observed for all batches. This phenomenon is explained by the fact that the material softens as it heats up, making the mixture less rigid.

When examining the loss factor (tan δ) versus temperature, pure ENR 25% displays a peak at (− 47 ± 2 °C) which corresponds to the glass-to-rubber transition of the matrix, also known as the glass transition temperature (Tg). In contrast, for the samples containing HEC in combination with ENR, the Tg shifts to slightly higher temperatures, as illustrated in Fig. 8. DMA measurements were performed in triplicate for each samples. (Table 3).

Fig. 8
figure 8

Tan δ curve of ENR and HEC/ENR composites.

Full size image
Table 3 Glass transition temperature (Tg) determined by DMA (tan δ peak) for ENR and HEC/ENR composites.
Full size table

Furthermore, a gradual yet moderate decrease in tan δ and a reduction in the amplitude of the damping peak associated with the glass transition of ENR are observed as the HEC content increases, as shown in Fig. 8 for the HEC/ENR composite. This suggests a decrease in energy dissipation. The effect is likely due to hydrogen bonding networks forming between HEC and ENR, which strengthen the matrix cohesion and limit the internal mobility of polymer chains, thereby enhancing the composite’s mechanical properties.

Similar studies20,38,39 have corroborated these observations and attribute the Tg​ shift to a mechanical coupling effect. These studies further explain that this shift is associated with a phase inversion phenomenon that occurs when the filler content reaches a percolation threshold. In general, the tan δ peak reflects the fraction of the matrix material capable of molecular relaxation, which is influenced by both the concentration of the matrix phase and by the restricted mobility of polymer chains at the interface between the matrix and the reinforcing material.

However, some studies20 suggest that the incorporation of cellulose into polymer matrices does not consistently affect the glass-to-rubber transition temperature. This observation appears to hold true regardless of the cellulose origin, matrix nature, or processing conditions, suggesting that the impact of cellulose on Tg is primarily influenced by the strength of specific interactions between the filler and the matrix.

Differential scanning calorimetry (DSC)

Differential Scanning Calorimetry (DSC) was used to explore possible interactions between epoxidized natural rubber (ENR 25%) and hydroxyethyl cellulose (HEC). The study aimed to assess whether intermolecular interactions, particularly between the epoxy groups of ENR and the hydroxyl (–OH) groups of HEC, influence the system’s glass transition temperature (Tg).

The thermal curves shown in (Fig. 9) display a single glass transition for the ENR/HEC-2 blend, indicating good compatibility and partial miscibility between the two polymers. Moreover, the Tg shifted from − 50 °C for pure ENR to − 37.4 °C for the ENR/HEC blend, reflecting a modest increase in polymer network rigidity. This rise in Tg is typically linked to decreased mobility of macromolecular chains, likely due to specific interactions such as hydrogen bonding between the functional groups of both polymers.

These findings are consistent with data obtained from Dynamic Mechanical Analysis (DMA), particularly the tan δ curves, which display a similar trend (Table 4). Previous studies, such as those by Nie et al.40, reported a comparable increase in Tg in ENR systems modified with functional fillers (e.g., nanochitin, carboxymethyl chitosan), attributed to interactions that restrict molecular motion. Similarly, Supramaniam et al.41. observed a Tg increase when cellulose nanofibers (CNF) were incorporated into ENR, emphasizing the structuring effect of chain interactions on the thermomechanical properties of the material.

Fig. 9
figure 9

DSC curve of ENR and HEC/ENR composites.

Full size image
Table 4 Glass transition temperature (Tg) measured by DMA (tan δ peak) for ENR and HEC/ENR composites.
Full size table

Gordon–Taylor equation

The glass transition temperature (Tg) is a key property of amorphous materials, particularly polymers and their blends. In the case of binary polymer mixtures, the variation of Tg with composition generally does not follow a simple rule of proportionality42. To model this behavior, the Gordon–Taylor equation is widely used. It is expressed as follows:

$$\:Tg=\:\frac{{w}_{1}{Tg}_{1}+K{w}_{2}{Tg}_{2}}{{w}_{1}+K{w}_{2}}$$

where \(\:{w}_{1}\)​ and \(\:{w}_{2}\:\)are the mass fractions of components 1 and 2, Tg1 ​ and Tg2 are the respective glass transition temperatures of the pure HEC (Tg1 = 137 °C) and ENR (Tg2= -50 °C), and k is an empirical parameter that reflects the strength of intermolecular interactions between the two components.

In this study, the Gordon–Taylor equation was applied to analyse the glass transition behavior (Tg) of ENR/HEC blends. To determine the interaction parameter k, a linearised form of the equation was used, in which the term \(\:{Tg}_{1}-{Tg}_{exp}\) was plotted as a function of \(\:\frac{{w}_{2}({Tg}_{exp}-{Tg}_{2})}{{w}_{1}}\)41.

The slope of the resulting straight line corresponds to the value of the parameter k. As shown in Fig. 10, the linear regression of the experimental data yielded a slope of 4.93, corresponding to a k value of 4.93. Such a high value suggests the presence of significant intermolecular interactions between the two polymers, likely due to hydrogen bonding between the epoxy groups of ENR and the hydroxyl groups of HEC, or possibly a chemical compatibility that enhances interfacial cohesion. This result supports the hypothesis43 that values of k > 0 are generally associated with positive deviations from the additivity of the glass transition temperature, indicating chain rigidification effects induced by attractive interactions between polymer segments. Conversely, values of k < 0 would reflect negative deviations, typically attributed to interchain repulsions that promote increased mobility of macromolecular segments.

Fig. 10
figure 10

Linearized Gordon–Taylor plot.

Full size image

Mechanical properties

To analyze the mechanical properties of HEC/ENR films, tensile tests were conducted using a dynamic mechanical Analysis (DMA). These tests, limited to low and maximum deformations of 3% due to the device’s capabilities, allowed for the characterization of the initial elastic modulus of the materials without reaching their breaking point. The results are presented as curves in the Fig. 11.

Fig. 11
figure 11

Stress–strain curves of ENR and HEC/ENR composites.

Full size image
Table 5 Tensile modulus (MPa) of ENR and HEC/ENR composites.
Full size table

For the pure ENR sample, represented by the black curve, a lower slope is observed, which indicates reduced mechanical strength. This behavior is consistent with findings from other studies20,44, which demonstrate that pure ENR exhibits limited mechanical properties, primarily due to the lack of structural reinforcement within the polymer matrix.

In contrast, with the incorporation of hydroxyethyl cellulose (HEC) into the ENR/HEC blends (ENR/HEC1 to ENR/HEC4), a progressive increase in tensile modulus is clearly observed in the histogram in Fig. 11 and Table 5. The ENR/HEC4 blend demonstrates the highest mechanical stiffness under low deformations, emphasizing the reinforcing effect of HEC. This enhancement can be attributed to the ability of HEC to function as a rigid structural component within the ENR matrix, promoting stronger intermolecular interactions and improving the overall mechanical cohesion of the composite.

The results also indicate that the elastic modulus increases proportionally with the HEC content in the blends, enhancing the material’s resistance to deformation even under low applied stresses. This property makes ENR/HEC films particularly suitable for applications requiring both flexibility and mechanical stiffness, such as membranes for food packaging, biomedical films for controlled drug release, protective films for electronic devices requiring resistance to minor mechanical stresses, or functional layers in advanced technical composites.

In conclusion, these tests underscore the critical role of HEC as a reinforcing agent, which markedly enhances the mechanical properties of ENR/HEC blends while maintaining a degree of flexibility an essential characteristic for a wide range of industrial and technological applications.

Fig. 12
figure 12

Fatigue curves of ENR and HEC/ENR composites.

Full size image

The curve obtained from the fatigue test (Fig. 12), performed using dynamic mechanical analysis (DMA), illustrates the variation of storage modulus as a function of the number of cycles for pure epoxidized natural rubber (ENR) and its composites with hydroxyethyl cellulose (HEC) at different HEC proportions. The storage modulus (E′), which indicates the material’s elastic rigidity, decreases progressively with the number of cycles, reflecting the onset of material fatigue. The pure ENR sample shows a sharp decline in E′ after a certain number of cycles, indicating a rapid degradation of mechanical properties under repeated stress. In contrast, the ENR/HEC composites demonstrate superior fatigue resistance, maintaining higher storage modulus values over an extended number of cycles before a sharp decline. This behavior can be attributed to strong interfacial interactions, particularly hydrogen bonds between the hydroxyl groups of HEC and the epoxides of ENR, which enhance the structural cohesion of the composites38. Additionally, the intrinsic rigidity of HEC helps to mitigate crack propagation and plastic deformations under cyclic stress. Increasing the mass fraction of HEC further improves fatigue resistance, as indicated by the shift of the curves toward a higher number of cycles before failure. However, an excessive high HEC content may lead to a reduction in flexibility, requiring a compromise depending on the intended applications. These findings demonstrate that incorporating HEC significantly improves the mechanical properties of ENR/HEC composites, making them suitable for applications where both elasticity and fatigue resistance are crucial, such as coatings, seals, or dampers45.

Contact angle measurements

The study of the contact angle of hydroxyethylcellulose (HEC) and epoxidized natural rubber (ENR) composite films provides insights into their wettability and surface interaction with liquids such as water46. For the film consisting solely of HEC, the contact angle is relatively low, around 50°, which is expected since HEC is a hydrophilic polysaccharide due to the presence of polar hydroxyl groups on its surface, providing excellent wettability47. In contrast, the composite HEC/ENR films have higher contact angles with a higher percentage of ENR increases the contact angle, attributed to the presence of ENR, a naturally hydrophobic polymer (Fig. 13)48. These composite films thus exhibit contact angles intermediate between those of HEC and ENR, resulting from the modification of the hydroxyl groups of HEC through interaction with ENR. The results show that these composite films, with their adjustable contact angles, offer tunable surface properties, Measurements were performed in quintuplicate (n = 5 per sample), and results are reported as mean ± standard deviation. Combining hydrophilicity and hydrophobicity depending on the ratios used. These characteristics open up promising applications, particularly in biodegradable packaging where partial water barrier properties are desirable, smart coatings that control wettability, and filtration membranes requiring a hydrophilic/hydrophobic balance.

Fig. 13
figure 13

Contact angle of HEC and HEC/ENR composites.

Full size image

Conclusion

Composite films based on hydroxyethyl cellulose (HEC) and epoxidized natural rubber (ENR) were successfully fabricated. FTIR analysis revealed the formation of hydrogen bonds between HEC hydroxyl groups and ENR epoxide groups, evidencing strong interfacial interactions between the two components. These hydrogen bonds increase the effective crosslink density and raise the glass-transition temperature (Tg), with more pronounced effects as the HEC content increases, thereby enhancing the overall performance of the material.

DMA results show an increase in the storage modulus in both the glassy and rubbery regimes, confirming the reinforcing role of HEC within the ENR matrix. In addition, the tan δ peaks shift to higher temperatures as the HEC fraction increases, reflecting altered thermomechanical behavior and more constrained segmental mobility.

HEC addition also improves tensile strength and fatigue resistance, with more significant gains at higher HEC loadings. Regarding surface properties, blending ENR with HEC increases hydrophobicity, thereby improving resistance to moisture and harsh environments. Interfacial interactions—particularly hydrogen bonding between HEC and ENR—strengthen structural cohesion and adhesion, contributing to the overall enhancement of composite performance.

These composites are suitable for a wide range of applications, including sustainable packaging films, selective membranes, functional coatings, biomedical films for controlled drug release, and protective layers for electronic devices. Their resistance to aging and mechanical fatigue also makes them relevant for seals, coatings, and vibration-damping elements. Moreover, these films are appropriate for uses requiring tunable wettability (e.g., biodegradable packaging and filtration membranes balancing hydrophilic/hydrophobic character).

Going forward, it will be valuable to broaden the formulation space by varying the HEC degree of substitution (DS) and molar mass, comparing other modified celluloses (HPC, CMC, HPMC, etc.), and finely adjusting HEC/ENR ratios to optimize the stiffness–damping–durability trade-off. In parallel, interfacial optimization with ENR should be deepened (polarity control, mild surface treatments, solution pH/salt adjustment, mixing/evaporation sequences), with dedicated characterization (XPS, AFM-IR, solid-state NMR) to link interfacial chemistry to macroscopic properties. From a processing standpoint, more “green” and scalable routes should be explored, notably aqueous or hydro-alcoholic media (with low-toxicity co-solvents). Finally, to validate application potential, we propose pilot-scale transfer and in-use testing of demonstrators for seals and gaskets, bushings and vibration isolators, functional coatings (packaging, protective, controlled friction), and membranes (separation, humidity control), integrating accelerated aging (UV, humidity, thermal cycling), extended fatigue testing, and compliance with sector standards (automotive, packaging). These avenues should consolidate the formulation–process–interface–performance linkage and facilitate translation from laboratory to industrial applications.