1 Introduction

Lignocellulosic fibers have gained significant attention due to their renewable and sustainable characteristics, making them desirable as reinforcing elements in composite materials. This interest has led to their widespread adoption across various sectors, including automotive [1], building materials, furniture, packaging, sports [2], ballistic protection [3], and marine applications [4]. However, the inherent polarity and hydrophilicity of these fibers, in contrast to the nonpolar characteristics of polyolefins, present challenges in achieving effective wetting and adhesion, often resulting in suboptimal composite performance [5,6,7,8].

To enhance the compatibility between these disparate materials, various fiber pretreatment methods have been explored, including nano-silica deposition [9,10,11], ionizing radiation [12, 13], graft copolymerization [14, 15], alkaline treatments [16, 17], and the application of silanes [18,19,20,21], isocyanates [22, 23], or maleic anhydride [24, 25]. These pretreatments modify the fibers, resulting in reduced water absorption by the lignocellulosic material when exposed to high humidity, improved dimensional stability [26], and enhanced fiber-matrix interfacial bonding, consequently improving the mechanical properties of composites [27, 28].

However, the search for pretreatment methods that are environmentally friendly and generate minimal pollution has driven research into alternative techniques, such as corona discharge stabilization [26, 29] and hornification [30, 31]. Corona discharge is a clean and environmentally friendly technology that is adaptable to continuous processes [32] and offers economic advantages [33]. Corona discharge alters the surface of plant fibers [26, 34] by activating the polymers [35, 36], improving wettability, and enhancing adhesion strength [24, 37, 38].

This study aims to evaluate how the surface activation of jute fibers with hot water and corona discharge affects the fiber properties and the fiber-matrix interface of a polyester-based system. The effects of the pretreatments on the fibers’ properties have been investigated through thermogravimetric analysis (TGA), differential scanning calorimetry (DSC), Fourier transform infrared (FTIR) spectroscopy, and scanning electron microscopy (SEM). The fiber-matrix interface behavior has been investigated through pullout tests.

2 Materials and methods

2.1 Materials

The jute fibers (Corchorus capsularis) used in this study were sourced from the Amazonas state in Brazil. Morphological details of these fibers are available in a previous study [39]. Orthophthalic unsaturated polyester resin served as the matrix, and 5% methyl-ethyl-ketone by weight was used as a catalyst; both materials were provided by Fibrasil Indústria e Comércio Ltda., Brazil. The resin had a relative density of 1.1-\(-\)1.2 g/cm\(^{3}\), a boiling point of approximately 145\(^{\circ }\)C, a solids content of 56–62%, and a gel time of 10 s according to the supplier’s specifications.

2.2 Jute fiber pretreatments

The fibers were divided into six groups: (i) raw fibers, (ii) fibers subjected to hot water pretreatment only, (iii) fibers pretreated only with corona discharge for 5 min, (iv) fibers pretreated only with corona discharge for 10 min, and (v) fibers pretreated with hot water followed by corona discharge (referred to as hybrid pretreatment) for 5 min and (vi) 10 min.

Hot water pretreatment involved immersing the fibers in boiling water at 100° C for 4 h, followed by drying at 60° C for 24 h. The corona discharge treatment was conducted using a Plasma-Tech device (model P-1, Corona Brasil Ltda.), following the procedure outlined by Mesquita et al. [26]. 50 mm long fibers were treated in bundles of approximately 2 mm in diameter. The distance between the electric discharge and the fibers was 2 cm, and the treatment durations were 5 and 10 min.

2.3 Thermal analyses: thermogravimetric analysis (TGA) and differential scanning calorimetry (DSC)

TGA and DSC analyses were performed separately using Shimadzu DTG-50 and DSC-60 H equipment. Samples weighing 4 mg of raw or pretreated jute fibers were placed in platinum crucibles and heated up to 550° C. One sample per treatment was tested for TGA and DSC analyses. The heating rate and nitrogen flow conditions followed the parameters specified by Ferreira et al. [30].

Fig. 1
figure 1

Pullout test casting procedure: fiber alignment process

Full size image

2.4 Fourier transform infrared (FTIR) spectroscopy

FTIR spectra of the untreated and pretreated fibers were obtained using an IRAffinity-1 spectrometer (Shimadzu Corp.). Attenuated Total Reflection (ATR) technique was employed with 64 scans and a resolution of 4 cm\(^{-1}\), covering the spectral range from 4400 to 600 cm\(^{-1}\). One sample per treatment was tested.

2.5 Scanning electron microscopy (SEM)

Surface characteristics of the raw and pretreated jute fibers were examined using a Zeiss Leo Evo 40 microscope operating at an acceleration voltage of 20.0 kV and a working distance of 6.40 mm. No tilting was applied, and samples were not coated. Images were obtained from one sample of each treatment. The images obtained were processed with ImageJ software (version 1.48v).

2.6 Pullout tests

For the pullout tests, sample preparations followed the procedure outlined in Ferreira et al. [40], as illustrated in Fig. 1. Ten samples were prepared for each of the four specified fiber embedment lengths in the resin (5, 10, 20, and 30 mm). The pullout tests were performed using a Tytron 250 MTS machine with pneumatic action claws, as shown in Fig. 2. The equipment operated at a speed of 0.3 mm/min at room temperature and with a 50N load cell.

Fig. 2
figure 2

Pullout and tensile test system: a general view of the machine, b the pullout test setup, c pullout specimen

Full size image

3 Results and discussion

3.1 Thermal analyses: thermogravimetry analysis (TGA) and differential scanning calorimetry (DSC)

The thermogravimetric behavior of both raw and pretreated fibers showed no significant differences (Fig. 3). The thermal stability of natural fibers mainly depends on the primary components of the fiber cell wall, which are not degraded by the pretreatments proposed herein. Maintaining thermal stability despite the pretreatments is advantageous for composite applications.

All samples exhibited three decomposition stages. The first stage, occurring between 25 and 100° C, corresponds to the loss of moisture from the material. The second stage, between 270 and 320 \(^{\circ }\)C, is related to the decomposition of low molecular weight hemicellulose [41, 42]. Lignin degradation initiates around 250 \(^{\circ }\)C and continues up to 600 \(^{\circ }\)C [30]. The third stage, occurring between 340 and 400 \(^{\circ }\)C, represents the final mass loss, characteristic of the thermal degradation of cellulose [42,43,44]. The dominant thermogravimetric peak (DTG) of the control sample manifested around 360 \(^{\circ }\)C.

In the DSC curves, the initial endothermic peak corresponds to moisture evaporation from the fibers [45], consistent with the observations in the TGA/DTG curves. The exothermic peak around 370° C is typical of the degradation of native cellulose [46]. Lignin and other non-cellulosic compounds continue to decompose by absorbing more heat at temperatures above 400° C [47].

Fig. 3
figure 3

Thermal analyses of the raw and pretreated jute fibers: TG, DTG, and DSC (up-bottom graphics)

Full size image

3.2 Fourier transform infrared spectroscopy (FTIR)

The FTIR spectra display bands corresponding to the chemical groups inherent to the primary constituents of lignocellulosic materials, such as cellulose, lignin, and hemicellulose, between 2000 and 800 cm\(^{-1}\) (Fig. 4) [30]. Certain components, such as hemicelluloses’ ester and carboxylic acids, are assigned to the 1740 cm\(^{-1}\) band corresponding to the C=O vibrations [44, 48]. The C–O–C and C–OH bonds of the constituent structures are attributed to IR bands found in the range between 1200 and 1000 cm\(^{-1}\), arising from C–O stretching vibrations, as reported by [49].

There were no structural changes in the chemical group compositions of jute fibers after the pretreatments. However, the intensity of the peaks related to the oxygen-containing groups (C=O, C–OH, and O–C=O; indicated by arrows in Fig. 4) increased with corona discharge [49]. Possibly, this pretreatment increased the surface energy of the lignocellulosic fibers by oxidation, forming hydroxyl, carbonyl, and other functional groups by first rupturing the C–C bonds and subsequently causing oxygen reaction by corona discharge [50, 51].

Fig. 4
figure 4

Fourier transform infrared spectroscopy (FTIR) of the raw and pretreated jute fibers

Full size image

3.3 Scanning electron microscopy (SEM)

SEM micrographs revealed that raw jute fibers had a smooth surface. Following hot water and corona discharge pretreatments, changes in surface morphology were observed, characterized by an increase in surface roughness. This change can be attributed to the removal of waxes and mucilage during hot water pretreatment and the erosive effect induced by surface activation during corona discharge treatment (Fig. 5) [52,53,54]. According to Ward et al. [34], pretreatment with plasma on the fiber’s surface results in the formation of cellulosic radicals by two distinct mechanisms. The first one concerns the rupture in the bond between C1 and ring oxygen, while the second mechanism involves the possible breakage of the connection between C1 and glycosidic-binding oxygen (Fig. 6).

The effect of corona discharge on the surface of lignocellulosic fibers increases oxygenated groups such as carboxylic acid (–COOH) and hydroxyl (–OH), raising the activated sites along the polymer chain of the fiber [26]. Such modifications increase anchorage points, providing better interaction between the fiber and matrix [26, 27, 32, 43, 55].

Fig. 5
figure 5

SEM images of jute fiber surface: a raw jute fiber (untreated), b corona discharge 5 min, c corona discharge 10 min, d hot water, e hot water + corona discharge 5 min, and f hot water + corona discharge 10 min

Full size image
Fig. 6
figure 6

Scheme of the possible effect of surface activation of jute fibers by corona discharge. Adapted from [34]

Full size image

3.4 Pullout tests

As described by Silva et al. [56], the ideal pullout test presents four distinct regions, showing a rapid linear elevation of the load in the first region, followed by a loss of linearity due to the beginning of fiber interfacial detachment in the second region. The peak corresponds to the third region, where the fiber is partially separated from the matrix, reaching the maximum value (P\(_{\textrm{ad}}\)), where the shear strength is defined as the adhesion stress (\(\sigma _{\textrm{ad}}\)). The fourth region corresponds to low and constant values, characteristic of interface friction shear strength (P\(_{\textrm{fr}}\)), up to the point of total fiber removal, defined as friction stress (\(\sigma _{\textrm{fr}}\)) (see Fig. 7). Variations in embedding and fiber pretreatments significantly affected the typical pullout curves (Fig. 8) and the curves’ parameters (Table 1).

Table 1 Average and standard deviation values of the pullout test for untreated and treated fibers
Full size table

For all fiber types, an increase in the embedment length promoted an increase in the adhesional load. For embedding lengths of 20 and 30 mm, fiber fracture was observed in the majority of specimens. A relevant variation occurred for a 10 mm embedding, whose curves closely resembled the ideal curve. The modifications indicated that hot water pretreatment increased adhesion to the matrix, possibly by removing waxes and other extractives from the jute surface [52]. The pretreatment may also reduce dimensional variation, decreasing the interface zone [30].

Fig. 7
figure 7

Typical pullout curve of the jute fibers at embedment length of 20 mm. Region I: elastic linear zone, Region II: beginning interfacial debonding, Region III: total debonding of the fiber from the matrix, Region IV: final slip removal of the fiber from the matrix

Full size image

Improved adhesion also occurred for the corona discharge and hybrid (corona + hot water) pretreatments. The pretreatments with corona for 5 and 10 min increased stiffness, adhesion, and friction load since the 5 mm test. Ruptures were observed for 20 and 30 mm embedding (see Fig. 8). It is possible to assume that the improvement in adhesion was promoted by the modifications caused on the fiber’s surface by the corona discharge (see Fig. 6).

The hybrid pretreatments (hot water and corona discharge for 5 and 10 min) showed higher adhesion and frictional phases. The combination of dimensional stability promoted by hot water washing and surface activation provided better fiber interaction with the polymer matrix. The results indicated that the longer the exposure time (10 min) increased the adhesion.

Fig. 8
figure 8

Influence of the pretreatments on the pullout behavior of jute fibers embedded in different lengths (5, 10, 20, and 30 mm)

Full size image

Mesquita et al. [26] reported in their study that the time of corona discharge on the surface of the material influences the adhesion between the materials. At an embedment length of 20 mm, there is a noticeable increase in frictional force, indicating an adequate bond to the polyester matrix. However, at the 30 mm embedment length, fiber fracture occurred. A comparison was provided since the embedment length of 20 mm indicated more similarity with an ideal pullout curve for all pretreated fibers. Considering the typical pullout curves of all pretreatments (Fig. 9), it is possible to observe that the hybrid one was quite effective in increasing the jute fibers’ adhesional and frictional behavior.

Fig. 9
figure 9

Typical pullout curves of jute fibers: raw, pretreated with hot water, pretreated with corona discharge for 10 min, and sequentially pretreated with both techniques

Full size image

4 Conclusions

The activation of the surface by corona discharge was effective in enhancing the interaction between the jute fibers and the polyester matrix within the composite materials. This improvement resulted from both chemical and physical modifications of the treated fiber surfaces.

The thermogravimetric analysis (TGA) and differential scanning calorimetry (DSC) revealed that the thermal properties of the jute fibers were not significantly affected by the pretreatments. This indicates that the chemical composition of the fibers remained largely unchanged, which is advantageous for maintaining the integrity of the fibers in composite materials.

Fourier transform infrared spectroscopy (FTIR) indicated that corona discharge pretreatment modified the surface chemistry of the fibers, providing more active sites for interaction with the polyester matrix.

Pullout tests demonstrated that fibers subjected to hot water immersion followed by a 10-minute corona discharge exhibited an increase in adhesion strength compared to untreated fibers. This indicates that the combined pretreatment effectively promoted modifications to the functional groups on the fiber surface, thereby improving interfacial bonding sites capable of interacting with the polyester matrix.

This study contributes knowledge on applying corona discharge under atmospheric conditions (25 ± 3° C and 70 ± 5% RH) to activate jute fiber surfaces. Further studies should be conducted to provide a better understanding of the effects of corona discharge treatment, particularly when combined with hot water pretreatment, on fiber-matrix interactions and composite performance. This could include more detailed investigations into the chemical changes induced by corona discharge, as well as comprehensive mechanical testing to assess the overall performance of the composite materials.