1 Introduction

Lignosulfonate results from the sulfite pulping process and presents particular properties as water-soluble from the abundance of hydrophilic groups in its macromolecules, especially sulfonic groups [1, 2]. Figure 1 illustrates sodium lignosulfonate's complex functional chemical structure (SLig), composed of several groups, like aromatic hydroxyl, methoxyl, and aldehyde, which confers particular properties to lignosulfonate, making it versatile for application in diversified areas: as water reducers [3, 4], dispersion agent for dyes and pesticides [5,6,7] and dust depressors [8]. Furthermore, in the engineering field, lignosulfonate has been investigated as functional material due to compatible groups, hydrophilic and aromatic groups, to formulate phenolic resins to apply in the structured composites and adhesives [9,10,11,12,13,14,15], as a poliol in the synthesis of polyurethane-type polymers [14]. It finds application in various other fields, such as animal feed, stabilizer in colloidal suspensions, chelation, complexation, soil conditioning, and flotation [16, 17]. Efforts have been made to broaden the applications of lignosulfonate, including its use as a biosurfactant to aid in the enzymatic conversion of lignocellulosic fibers into fermentable sugars. [18].

Fig. 1
figure 1

Chemical structures of glutaraldehyde and formaldehyde; sodium lignosulfonate partial chemical structure

Full size image

The annual worldwide production of lignosulfonates stands at approximately 1.8 million tons, and these technical lignins constitute roughly 90% of the commercial lignin market [19]. Due to their wide range of applications and tendency to increase demand, lignosulfonates have become an important component of the wood biorefinery platform. They also can play a significant role in advancing the bioeconomy in various countries.

Despite being widely available almost worldwide, its use as a reagent in advanced value-added applications still needs to be on a large scale [20].

Thermosets, as the phenolic type, have been one of the most essential macromolecular materials used by industry worldwide, mainly due to their chemical and thermal properties. Biomass has the potential to replace fossil-based raw materials to meet society's growing demands for environmental sustainability [21, 22]. Chen et al. [1] reported promising results demonstrating that using SLig to prepare phenolic-type nanospheres contributed to a high-density loading of silver nanoparticles, as SLig enhanced the number of silver ions that could be adsorbed in the spheres while also preventing the aggregation of the nanoparticles.

Together with phenols, aldehydes are fundamental reagents to synthesize phenolic resins, being that formaldehyde (Fig. 1) has been used throughout the ages. Formaldehyde is in equilibrium with methylene glycol in an aqueous solution, which is its commercial form [23]. The reaction of formaldehyde with phenol forms hydroxymethyl phenols that can be later crosslinked through the formation of methylene and methyl ether bridges [24].

Glutaraldehyde (Fig. 1) has a much lower vapor pressure (16.4 mmHg at 20°C, 25% aqueous solution) than formaldehyde (33.2 mmHg at 20 °C, 37% aqueous solution) and is safer to handle. Furthermore, as a di-functional aldehyde, glutaraldehyde is an interesting candidate to substitute formaldehyde; its longer molecule can confer some flexibility to the formed networks [25].

Phenolic matrices are regularly reinforced with plant fibers, such as sisal, jute, curaua, mauve, coconut, and others [26,27,28]. Brazil is the leading sugarcane producer in the world. The country’s raw sugar production accounts for 22% of the global output; over 750 million tons of sugarcane are produced annually [29]. This large crop leads to a relatively large amount of lignocellulosic residue in the form of sugarcane bagasse. Besides being burned to make up sugar mills’ own energy needs, diverse applications have been proposed for sugarcane bagasse in the last decades, as in the production of liquid or gaseous fuels [30,31,32], chemicals [33,34,35], and materials [36, 37].

This study aimed to produce composites from a high content of renewable raw materials (Fig. 2), i.e., lignosulfonate-based phenolic matrices reinforced with sugarcane bagasse fibers (SBF) while evaluating the impact of the substitution of formaldehyde for glutaraldehyde on the properties of the thermoset matrix and the composite.

Fig. 2
figure 2

Biomass-based composite: sugarcane bagasse fibers as a reinforcement of lignosulfonate-based phenolic matrix

Full size image

2 Materials and methods

2.1 Materials

Sugarcane bagasse fibers (SBF) were supplied by the Santa Lucia sugarcane mill (Araras, Sao Paulo, Brazil), where they were burned and ground for sugarcane processing for ethanol production. First, waxes, terpenes, and fatty acids on the fiber's surface were removed by subjecting the fibers to a cyclohexane/ethanol mixture (1:1 v/v) reflux. Subsequently, the fibers were washed with distilled water and dried in an air-circulating stove at 105 °C until constant weight. The characterization of these fibers led to 56.50 ± 0.3%, 21.80 ± 0.30%, 20.90 ± 0.02% of cellulose, hemicelluloses, total Klason lignin, respectively, and 0.75 ± 0.04%, 5.50 ± 0.20%, and 47% of ashes, moisture, crystallinity, respectively, as described elsewhere [36].

The sodium lignosulfonate (SLig) used was a by-product obtained from the sulfite pulping processing of Pinus taeda wood, Vixilex SD-type (Mw approximately 6000 g mol−1) composed of sulfur (5.5 wt %), magnesium (1.7 wt %), calcium (0.2 wt %), and sugars (0.9 wt %), as informed by the supplier (Borregaard Group, LignoTech (Cambará do Sul, Rio Grande do Sul, Brazil).

All other reagents, formaldehyde (Synth, 37 %), glutaraldehyde (25%, Vetec), KOH (Synth), and HCl (Synth, 37 %), were used as purchased.

2.2 Methods

2.2.1 Prepolymer synthesis

All prepolymers were synthesized by heating on a mantle using a three-necked flask coupled to a condenser with ice water circulation, a mechanical stirrer, and a thermometer, as shown in Scheme 1. The reagent proportions, Table 1, were based on a previous study [36].

Scheme 1
scheme 1

Prepolymers synthesis

Full size image
Table 1 Proportions of the reagents in the prepolymer synthesis
Full size table

Upon reaching room temperature, the solution underwent pH adjustment using concentrated HCl until pH=7. Subsequently, a rotary evaporator was utilized to remove water under reduced pressure, with a bathwater temperature of approximately 45°C.

2.2.2 Unreinforced and reinforced thermoset synthesis

Scheme 2 shows the methods for synthesizing thermosets For-SLig (T) and Glu-SLig (T), using For-SLig (P) and Glu-SLig (P), respectively, and for their corresponding SBF-reinforced composites, namely, For-SLig (C) and Glu-SLig (C), respectively. The step cycles utilized in the process were established from a previous study [36]. The length of 15 mm was used due to the typical length of sugar cane bagasse fibers produced as agricultural waste. The percentage of 30wt%, Scheme 2, was chosen based on findings from previous research [36, 37].

Scheme 2
scheme 2

Unreinforced and reinforced thermoset synthesis

Full size image

2.2.3 Characterizations

Prepolymers, thermosets and composites were characterized to assess their main chemical groups, thermal behavior and mechanical properties, Scheme 3. In addition, the composites' dynamic-mechanical analysis (DMA) was conducted using the TA Instruments Q800 model. The specimens, measuring 64 mm x 12 mm x 3.2 mm, were subjected to bending mode with an oscillation amplitude of 20 mm and frequency of 1 Hz. The temperature range for the analysis was from 25 °C to 200 °C, with a heating rate of 2 °C/min. Also, to visualize the morphology of fractured surfaces, scanning electron microscopy (SEM) was conducted with a Leica Model 440 (Carl Zeiss Microscopy; Jena, Germany) under the same conditions described elsewhere [37].

Scheme 3
scheme 3

Characterizations of prepolymers, thermosets, and composites

Full size image

3 Results and discussion

3.1 FTIR

Figure 3 shows the infrared spectra of (a) sugarcane bagasse fibers (SBF) and sodium lignosulfonate (SLig); prepolymers, thermosets, and composites of (b) sodium lignosulfonate-formaldehyde (For-SLig) and (c) sodium lignosulfonate-glutaraldehyde (Glu-SLig) formulations. The large band at 3400 cm-1 can be attributed to the –OH group; methylene groups' C–H stretching was observed at around 2900–2980 cm-1. Both bands are present in cellulose, hemicelluloses, and lignin, therefore, appeared in all spectra.

Fig. 3
figure 3

FTIR spectra of (a) sugarcane bagasse fibers (SBF) and sodium lignosulfonate (SLig); prepolymers (P), thermosets (T), and composites (C) (b) sodium lignosulfonate-formaldehyde: For-SLig (P), For-SLig (T), For-SLig (C), and (c) sodium lignosulfonate-glutaraldehyde (Glu-SLig): Glu-SLig (P), Glu-SLig (T), Glu-SLig (C)

Full size image

The band at 1730 cm-1 in the SBF spectrum (Fig. 3a) can be associated with C=O stretch in lignin and/or hemicelluloses [38]. An intense band at around 1630 cm-1 in the SLig spectrum (Fig. 3a) is related to C=O stretching and C=C aromatic skeleton vibrations [39], which was also present in the spectra of prepolymers, thermosets, and composites (Fig. 3b,c), where SLig replaced phenol.

The low-intensity bands at 1512 cm-1 and 1427-1462 cm-1 in the SLig spectrum (Fig. 3a) are associated with aromatic ring stretching vibrations [40]. Both bands are present in the spectra of the synthesized materials. The bands at 1250–1200 cm-1 were attributed to C–C, C–O, and C=O stretching [39]. The strong band around 1040 cm-1 can be considered a consequence of vibrations in polysaccharides: C-OH bending, C-O, and C-C stretching [41]. The band around 655 cm-1 is related to the vibration of sulfonate groups.

Figure 4 shows possible structures of the thermosets For-SLig (T) and Glu-SLig (T), which are based on the reactivity of aldehydes with phenols in an alkaline medium.

Fig. 4
figure 4

Possible chemical structures of the thermosets (a) For-SLig (T) and (b) Glu-SLig (T)

Full size image

Figure 4 highlights possible bridges between the aromatic rings generated by the reaction with formaldehyde (Fig. 4a) or glutaraldehyde (Fig. 4b).

3.2 Thermal analysis

All raw materials and composites were analyzed by TGA; Fig. 5 shows the TG and DTG curves. As usually noticed for bio-based materials, there was a slight weight loss below 100 oC, corresponding to volatilization of absorbed or bound water [42]. Such mass loss was practically not observed for the composites For-SLig (C), Fig. 5b, and Glu-SLig (C), Fig. 5c. This suggests the surfaces of these composites are hydrophobic.

Fig. 5
figure 5

TG and DTG curves of (a) sugarcane bagasse fibers (SBF) and sodium lignosulfonate (SLig); prepolymers (P), thermosets (T), and composites (C): (b) sodium lignosulfonate-formaldehyde: For-SLig (P), For-SLig (T), For-SLig (C), and (c) sodium lignosulfonate-glutaraldehyde (Glu-SLig): Glu-SLig (P), Glu-SLig (T), Glu-SLig (C), under N2 atmosphere (flow 20 mL/min), and heating rate 10 °C/min)

Full size image

DTG curve of SLig (Fig. 5a) exhibited two intense peaks correlated to the thermal decomposition of sulfonate moieties and aromatic rings at 304 oC and 776 oC, respectively. The DTG curve of SBF (Fig. 5a) displayed a sharp peak at 364 oC and a shoulder at 331 oC, attributed to the decomposition of cellulose and hemicelluloses, respectively. SBF lignin decomposed above 530 oC.

The broad and intense peak between 120 oC and 150 oC in the DTG curves of prepolymers (Fig. 5b,c) resulted from the volatilization of the water generated as a by-product of the crosslinking reaction that occurred with the resols during scanning. However, there was no peak at this temperature interval in the curves of thermosets and composites (Fig. 5b,c) because the samples were crosslinked before analyses through a gap that reached 125 oC (Scheme 2).

Between 250 oC and 350 oC, the DTG curves of the thermosets (Fig. 5b,c) shows two peaks for For-SLig (T) and a broad peak for Glu-SLig (T), which can be attributed to a residual cure step during scanning, and the consequent released and volatilization of water. This appeared as much less intense peaks in the prepolymer´s DTG curves.

Some events might be overlapped regarding the composites, such as the matrix residual cure and the thermal decomposition of the reinforcement fibers between 250 oC and 350 oC, Fig. 5. The curves for both thermosets, For-SLig (T) and Glu-SLig (T), as well as composites, For-SLig (C) and Glu-SLig (C), showed no significant differences. This means that replacing formaldehyde with glutaraldehyde did not impact the thermal stability of these materials.The peaks around 760 oC are related to the last decomposition step of the prepolymers, thermosets, and composites.

3.3 Mechanical properties

Figure 6 shows the results of the unnotched Izod impact strength for each matrix type. It should be emphasized that the thermosets proposed in this study, synthesized from lignosulfonate, formaldehyde, or glutaraldehyde, resulted in a too-fragile material, and the tests could not be performed, Fig. 6.

Fig. 6
figure 6

(a) Unnotched Izod Impact test results of the lignophenolic thermosets (T) and composites (C) with sugarcane bagasse (30 wt%) as reinforcement of sodium lignosulfonate-formaldehyde (For-SLig) and sodium lignosulfonate-glutaraldehyde (Glu-SLig) matrices, (b) SEM micrographs of the fractured surfaces of For-Slig (C) and Glu-SLig (C)

Full size image

The phenolic-type thermosets are traditional polymeric matrices applied for a long in engineering. Its mechanical properties are usually improved by adding fibers to the matrix, and synthetic or natural fibers are commonly used to reinforce the thermoset, increasing their impact resistance [43,44,45], as observed for the SLig-based phenolic thermoset matrices reinforced with sugarcane bagasse fibers, Fig. 6.

The results suggested that sugarcane bagasse fibers from the Brazilian agro-industry, generated on a large scale, despite having a short length (≅1.5 mm) resulting from processing generation at the mill restraints, have potential application as reinforcement of phenolic-type composites. In addition to improving the impact strength, the bagasse residue fibers and using lignosulphonate as a reagent in the matrices synthesis resulted in composites formed from a high content of raw materials from plant sources. Besides excellent interface adhesion, the micrographs (Fig. 6 b,c) show that the matrices filled and recovered the fibers.

The substitution of formaldehyde for glutaraldehyde in the prepolymers syntheses increased the impact resistance values of the composite, reaching 82 ± 8 J m-1, Fig. 6a. Fig. 6b,c shows the fractured surface of the composites. Both composites, For-SLig (C) and Glu-SLig (C) showed similar characteristics when reinforced with sugar cane bagasse: fibers broke next to the fracture plane of the matrix, the fibers were filled with resin, and few pulled-out fibers were observed.

Due to the energy transferred to the fibers from the matrices, some fibers were detached from the matrices (Glu-SLig (C), For-SLig (C), during the impact test. As shown in Fig. 4, the chemical structure of thermosets contains polar groups like hydroxyls. These same polar groups can also be found in the primary components of sugarcane bagasse fibers (lignin, cellulose, and hemicelluloses). Thus, strong hydrogen bonding interactions may occur at the fiber and matrix interface. Furthermore, the chemical structure of the matrix and lignin that make up the fibers contain non-polar domains, such as aromatic rings. These non-polar domains facilitate hydrophobic intermolecular interactions. Combined with hydrogen bonding, these interactions create good adhesion at the fiber-matrix interface.

The fibers present at the interface are hydrophilic. According to the TGA results shown in Fig. 5, the composites contain mostly hydrophobic domains on their surfaces. This helps to protect the hydrophilic fibers from absorbing water. Therefore, it can be inferred that water did not substantially impact the adhesion between the fibers and matrix [46]. Figure 6b,c illustrates the adequate interface interaction, corroborating the results of impact tests.

In Glu-SLig (C), Fig. 6c, some microcracks appeared around the fiber, which indicated that the impact load on the matrix was transferred to the fiber during the test. In this way, the energy involved in the process was distributed through the crack and absorbed by the fiber. The fiber’s energy absorption partially disrupts the interactions at the fiber-matrix interface, thus causing its detachment. This mechanism is commonly related to thermoset composites due to the matrix properties [47].

The sugarcane bagasse used in the present study was produced through a process in which it is burned, as reported in 2.1. In a previous study, unburned sugarcane bagasse from mechanized harvesting and a phenolic matrix (instead of lignophenolic, as in the present study) were used. The mechanization/unburn technique enabled the incorporation of fibers of varying lengths (1/3/5 cm, 30 wt%). The impact resistance of phenolic composites reinforced with 1 cm and 3 cm fibers was around 50 J/m. Meanwhile, the composite reinforced with 3 cm fibers had an impact resistance of about 40 J/m [37]. The decrease in impact strength observed while using 5cm fibers can be attributed to the possibility of fiber bending during processing due to their length, which may negatively affect the influence of fiber length on the impact property. When comparing the best impact strength result of these phenolic composites (50 J/m, fibers measuring 1 cm and 3 cm) to the results of the present study for For-SLig (C) and GLu-SLig (C), Fig. 6, it was found that the impact strength of these lignophenolic composites were 38% and 65% higher, respectively. Based on the findings, it was observed that adding burnt fibers that are 1.5 cm in length improved the strength of lignophenolic matrices.

The composites were also tested for flexural properties, as shown in Fig. 7. The composite flexural strength of For-SLig and Glu-SLig (C) are similar; however, a noticeable increase in the flexural modulus occurred when formaldehyde was replaced by glutaraldehyde, which agrees with the higher impact strength (Fig. 6).

Fig. 7
figure 7

(a) Flexural strength; (b) flexural modulus of lignophenolic thermosets (T) and composites (C) based on sodium lignosulfonate-formaldehyde (For-SLig) and sodium lignosulfonate-glutaraldehyde (Glu-SLig) matrices reinforced with sugarcane bagasse (30 wt%)

Full size image

In previous studies on SLig phenolic thermosets [36, 48], the dispersive component of the free surface energy (which can be considered as an indication of the density of surfaces´ non-polar domains) of For-SLig thermoset, Glu-SLig thermoset, and sugarcane bagasse fibers were evaluated using inverse gas chromatography, which led to 31 mJ m-2, 42 mJ m-2 and 45 mJ m-2, respectively. It is possible to associate the better performance of Glu-SLig (C), compared to For-SLig (C), with better compatibility between the nonpolar domains of fibers and matrix, which were very close in value. Therefore, it stands out that interface plays a significant role when designing composites, and replacing formaldehyde with glutaraldehyde is beneficial when using reinforcements presenting surfaces with a higher density of nonpolar sites.

The DMA curves for the For-SLig (C) and Glu-SLig (C) composites are presented in Fig. 8. It is worth noting that the thermosets were not included in this evaluation process due to their fragile nature.

Fig. 8
figure 8

(a) Storage modulus of the composites based on sodium lignosulfonate-formaldehyde, For-SLig (C), and sodium lignosulfonate-glutaraldehyde, Glu-SLig (C) matrices reinforced with sugarcane bagasse (30 wt%); (b) and (c) Loss modulus and Tan δ curves of For-SLig and Glu-SLig, respectively

Full size image

Composites DMA curves are related to matrix properties, which are impacted by the presence of fibers [46, 49]. In the present study, the matrices differ by the chemical structure of the bridges between the aromatic rings, Fig. 4, and both have the same reinforcement. It is worth noting that the matrices are thermosets, which are crosslinked macromolecules. As a result, the changes in the storage modulus, loss modulus, and Tan δ curves are related to the movements of uncrosslinked short segments.

Throughout the temperature range examined in Fig. 8, the storage modulus of Glu-SLig (C) is slightly lower than that of For-SLig (C). This difference may be explained, at least in part, by the fact that the bridges between the aromatic rings of Glu-SLig (C) offer less resistance to movement (a consequence of the rotation of single bonds) when compared to the methylene bridges of For-SLig (C), Fig. 4.

The peaks in the loss modulus curves, Fig. 8b,c, can be related to the glass transition, Tg, associated with the movement of many uncrosslinked segments. The Glu-SLig (C) exhibits a higher peak temperature, 170°C, than For-SLig (C), which peaks at 135°C. This can be attributed to the stronger intermolecular interaction between Glu-SLig and lignocellulosic fibers of sugarcane bagasse, as already mentioned. The restricted movement of segments in the layers closest to the fibers raises the peak temperature. Figures 8c and b highlight that the Tan δ peak temperature of Glu-SLig (C) is 204°C, which is also higher than that of For-SLig (C) at 157°C. The second peak observed in the loss modulus Glu-SLig (C) curve at 220°C could be attributed to a crosslinking during scanning. Nonetheless, it is important to highlight that this temperature is proximate to the onset of the material's thermal decomposition, evident from its TG curve (Fig. 3c).

4 Conclusions

Reinforcing the lignosulfonate-based thermosets, synthesized from glutaraldehyde or formaldehyde, with sugarcane bagasse fibers increased the flexural and impact resistance compared to unreinforced thermosets. The composite formed from the prepolymer synthesized using glutaraldehyde showed ≅20% increase in impact resistance and ≅45% increase in flexural modulus compared to the formed using formaldehyde.

The results indicated that the interactions on fiber/matrix at the interface and a consequent good adhesion between them might arise not only using fiber modification, as primarily reported in the literature but can also be done by adjusting the matrix formulation.

A phenolic-type prepolymer was synthesized from a phenolic reagent derived from lignocellulosic biomass and an aldehyde with lower vapor pressure than formaldehyde, lignosulfonate, and glutaraldehyde, respectively. The prepolymer, previously mixed with sugarcane bagasse fibers, was used in the crosslinking reaction that led to the phenolic-type thermoset reinforced by such fibers. This process took place using molding under temperature and pressure. As a result, renewable raw materials available in various regions of the world generated composites through an uncomplicated process, facilitating the scaling up of such materials and promoting the circular bioeconomy.

Assessing the mechanical properties of composites is critical for identifying their potential applications. Based on the impact and flexural strength test results, the material can be considered suitable for use in applications involving moderate loads. In this scenario, given the excellent thermal stability at high temperatures, strong adhesion at the fiber-matrix interface, and high Tg values exhibited by the composites, they have the potential for use as internal panels in the automotive and aircraft industries, as well as in the furniture, civil construction, and rigid packaging sectors.