Recycling Cork/PLA Bio-Composites Through Dissolution–Precipitation Method

Abstract

Composites can offer superior properties and versatility but raise environmental concerns due to disposal challenges, even when made from bio-based materials. Hence, in this study, cork/PLA bio-based composites were recycled using dissolution–precipitation principles. First, virgin cork and PLA were extruded to produce cork/PLA bio-composites which were then recycled using dichloromethane to separate the biomass filler from the biopolymer matrix. It was found that 80.9% ± 2.4 of cork and 85.9% ± 5.9 of PLA were successfully recovered, with the recovered materials retaining the same chemical structure as the virgin counterparts. The cork maintained its honeycomb structure after extrusion and recycling, indicating its resistance to the process. As expected, adding cork to PLA reduced the composite’s mechanical performance, but the recovered PLA showed similar mechanical properties to the virgin PLA. Both virgin PLA and composite filaments displayed similar glass transition (Tg) and cold crystallization (Tcrist) temperatures, but the recovered PLA presented slightly lower values, likely due to some PLA degradation. Despite this, all recovered materials exhibited similar thermal stability to their virgin counterparts. Cork is primarily used in the production of cork stoppers, and, hence, its recycling efforts mainly focus on reusing cork from stoppers rather than from composites. Therefore, the recycling process proposed successfully separated cork from PLA composites, with the recovered materials maintaining comparable properties, highlighting the potential for improving the eco-efficiency of composites.

1. Introduction

Lignocellulosic polymer composites, comprising a polymer matrix and lignocellulosic fillers, are increasingly used as replacements for metal and alloy components in various industries, including automotive, aerospace, and space applications [1]. Hence, they are used to produce insulating panels, flooring, furniture, window panels, doors, and tiles, among others. Their adaptability to various environmental conditions and complex design requirements makes them well-suited for modern, sustainable engineering applications. Furthermore, they provide significant advantages over synthetic fillers, including biodegradability, affordability, CO₂ neutrality, ease of processing, and minimal health risks [2,3,4,5,6]. These bio-based composites are obtained from natural sources, including abaca, kenaf, flax, hemp, sisal, coir, bagasse, jute, cork, bamboo, and many more [7,8,9].

Cork, the outer bark of Quercus suber L., is a versatile lignocellulosic material predominantly sourced from southern Mediterranean regions, with Portugal being the world’s largest producer [5]. Cork is primarily composed of suberin (30–60%), lignin (19–22%), polysaccharides (12–20%), and extractives (9–20%). In addition, compared to other fillers, like hemp or bamboo, cork offers distinct advantages in certain applications. For instance, due to its higher hydrophobicity cork has superior moisture resistance. Additionally, cork exhibits very low density, resulting in light composites with improved processing characteristics. In a similar manner, due to its unique cellular structure, cork has low thermal conductivity [10].

Increasingly, there is interest in the sustainable valorization of cork residues, especially in the development of cork-based polymer composites. These composites, made by binding cork granules with synthetic or natural matrices, have applications in thermal and acoustic insulation, footwear soles, flooring, or sandwich composites for construction. Rezaieyan et al. [4] produced micro-perforated panels based on cork and polylactic acid (PLA) with enhanced acoustic properties, Ghonjizade-Samani et al. [5] produced acrylonitrile–butadiene–styrene (ABS) composites where cork acted as a flame retardant, or Svetlana et al. [11] produced cork–plaster composites where the addition of cork to the plaster matrix reduced the density and thermal conductivity of composites. Overall, cork’s inherent properties make it an ideal filler material for green composites, aligning with the broader goals of environmental sustainability and resource efficiency [12].

In turn, examples of polymers used in the production of bio-based composites include PLA, poly (butylene succinate) (PBS), poly (ethylene furanoate) (PEF), and polyhydroxyalkanoates (PHAs). The total global production capacity of biobased plastics reached 2.2 Mt in 2022, which is less than 1% of global plastics production [13]. PLA, one of the leading bio-based polymers, is derived from starch through fermentation. As a renewable, microbially produced, and compostable polyester, PLA can be processed in a manner similar to many conventional commodity plastics [14,15]. In addition, PLA has been utilized in high-value applications, including tissue engineering, drug delivery, orthopedic regenerative engineering, and shape memory hydrogels.

Despite their bio-based composition, composites are seldom reused or recycled. In fact, their hybrid structures compromise end-of-life processing [16,17,18]. The bond between the fibers and the matrix is designed for durability, which makes it difficult to separate the two components during recycling. As a result, attempts to extract the fibers may break or damage them, making them unsuitable for reuse. Consequently, composite materials are typically only reused to produce the same type of material. Hence, effective recycling requires separating the reinforcement from the matrix, which is essential for enhancing recyclability. One promising approach is solvolysis, a dissolution–precipitation process often used for recycling plastic waste [19]. In solvolysis, polymer wastes are dissolved in a solvent, followed by filtration to remove undissolved contaminants such as dirt, fillers, and reinforcements within composites. The dissolved polymer can be subsequently recovered as a pure resin through three methods: (1) the addition of an anti-solvent; (2) evaporating the solvent from the polymer solution; (3) reducing the temperature of the polymer solution to recover the polymer as a precipitate [20,21]. This method allows for the recovery of cleaner, high-quality polymers, supporting more sustainable recycling practices for composite materials.

Cork is a fully recyclable material, primarily used in the production of cork stoppers. Consequently, the literature predominantly focuses on recycling and reusing cork from cork stoppers rather than from composites [22,23]. Hence, in this work, a dissolution–precipitation methodology was applied to separate the reinforcement from the polymer matrix in cork/PLA composites. Both recovered fractions were characterized and found to have properties comparable to those of the original materials. Thus, this study successfully demonstrates composite recycling through the dissolution–precipitation process, contributing to the sustainability of an already eco-friendly material.

2. Results

In this study, cork/PLA bio-based composites were recycled using dissolution–precipitation principles. Initially, both cork powder and virgin PLA were characterized. Subsequently, cork/PLA composites (containing 5 wt% cork) were prepared through extrusion, and the resulting filaments were analyzed. Next, the recycling process involved dissolving the cork/PLA composites in dichloromethane at ambient temperature to separate the biomass filler from the biopolymer matrix. Both the recovered cork and PLA were thoroughly characterized and compared to their virgin counterparts. In Figure 1, the scheme of the methodology adopted is presented, where it can be seen that 80.9% ± 2.4 of the initial cork and 85.9% ± 5.9 of the PLA were successfully recovered. These values are lower than those reported by López et al. [24] or Li et al. [25], who achieved yields close to 100% when recycling polymers through a dissolution–precipitation process. However, in those studies, the entire composition of the material to be recycled is soluble in a specific solvent. In contrast, cork is insoluble in any solvent, which makes its separation from the polymer matrix difficult.

In addition to the recovery yield, a full assessment of the potential of the recycling process and the quality of the recovered materials requires comprehensive characterization to determine whether they retain their original properties. In that sense, the morphology of virgin and recovered cork are presented in Figure 2.

From the SEM micrographs in Figure 2, it is evident that both virgin and recovered samples exhibit the characteristic honeycomb structure of cork. In a previous study [26], cork underwent chemical modification through a two-step reaction to enhance its compatibility with styrene–ethylene–butylenestyrene (SEBS). Notably, the cork retained its original morphology following the chemical treatment. This suggests that cork may be resistant to mild chemical processes, thus keeping its most relevant properties preserved. Nevertheless, it is known that the composition of cork can be affected when treated with dichloromethane, as shown by Simões et al. [27]. Nonetheless, further characterization was performed to compare virgin materials, composites, and recovered materials, focusing on their chemical properties. Figure 3 presents the FTIR and ¹³C NMR spectra of the virgin and recycled raw materials, as well as the composite, respectively.

As can be seen in Figure 3a, the normalized FTIR spectra of cork samples are identical and present the characteristic bands of cork PMMA from Cork [28,29]. The band observed at 3620–3110 cm⁻¹ corresponds to the stretching vibrations of hydroxyl (OH) groups. The peaks between 2920 and 2850 cm⁻¹ are associated with the stretching vibrations of aliphatic C-H bonds. A distinct peak at 1738 cm⁻¹ indicates the stretching vibrations of ester carbonyl (C=O) groups. Peaks at 1610 and 1512 cm⁻¹ are linked to the stretching vibrations of aromatic C-H bonds. The bending vibrations of aliphatic C-H bonds are represented by the peak at 1460 cm⁻¹, while the peaks at 1260, 1160, and 1035 cm⁻¹ are indicative of C-O stretching vibrations in alcohol and ether groups [30,31]. The only difference between the virgin and recovered cork appears in the 1260–1160 cm⁻¹ region, where the peaks are barely visible in the recovered cork. This may be attributed to traces of PLA masking the peaks. For PLA, it can be observed that both virgin and recovered samples exhibit identical FTIR spectra [32,33], showing the typical features. The band in the 3600–3000 cm⁻¹ range, though barely noticed, corresponds to the stretching vibrations of O–H groups. Peaks between 2920 and 2850 cm⁻¹ are associated with the stretching vibrations of aliphatic C–H groups, while the peak at 1780 cm⁻¹ is attributed to the stretching vibrations of C=O bonds. The bending vibrations of aliphatic C–H groups are represented by the peaks at 1455 and 1380 cm⁻¹. Peaks at 1180 and 1080 cm⁻¹ correspond to the C–O stretching vibrations in C–C(O)–O groups. For the composite, which is predominantly composed of PLA, its FTIR spectrum largely mirrors that of the matrix. Notably, the aliphatic C–H peaks at 2920–2850 cm⁻¹ are particularly prominent, reflecting the contribution of the cork content.

Similarly, both virgin and recovered cork samples, as well as virgin and recovered PLA, displayed identical ¹³C NMR spectra. The peaks observed between 22 and 40 ppm in the cork samples are attributed to aliphatic methylene groups. A peak at 56 ppm is associated with the methoxyl groups of lignin. Signals in the range of 60–105 ppm correspond to the carbons of cellulose and hemicellulose. The peaks between 112 and 156 ppm are assigned to aromatic carbons in lignin, cellulose, and hemicellulose, while the peak at 172 ppm is linked to the carbonyl carbons in suberin and lignin components [34,35]. In the PLA samples, the peak at 170 ppm is attributed to the carbons of the ester bond, while the peak at 70 ppm corresponds to the carbons adjacent to the ester linkages. The peak at 16 ppm is assigned to the methyl groups attached to the carbonyl groups. Since the composite is mainly composed of PLA, its ¹³C NMR spectrum closely resembles that of the matrix.

As noted earlier, some components, namely extractives, can be removed from the cork when subjected to dichloromethane. These compounds should theoretically transition to the liquid phase and, following solvent evaporation, be incorporated into the recycled PLA. However, a comparison of the FTIR and ¹³C NMR spectra of virgin and recovered raw materials reveals no significant differences. Specifically, if certain species are removed from the cork, their absence is not evident in the recycled cork. Similarly, if any species are transferred to the recycled PLA, their presence remains undetectable compared to neat PLA. These findings suggest that either no component was removed from the cork during the recycling process or that their removal occurred to such a minimal extent that it is undetectable.

After producing and recycling the composites, the virgin PLA, composite, and recovered PLA samples were characterized for their mechanical and thermal properties; the results are presented in

Table 1. Properties of virgin and recovered PLA as well as of composite.

Sample	Young’s Modulus (MPa)	Elongation at Break (%)	Tg (°C)	Tcrist (°C)
VPLA	884.4 ± 52.2	46.1 ± 0.8	57.1	96.3
Composite	490.3 ± 35.4	29.4 ± 1.5	56.9	98.0
RPLA	826.0 ± 78.5	50.8 ± 1.4	53.2	92.8

.

The results in Table 1 indicate that incorporating cork into PLA leads to a composite with a lower Young’s modulus and reduced elongation at break. It is widely recognized that adding cork to a polymer matrix can have an adverse effect on the mechanical performance of the resulting composite. This is primarily due to the flexible nature of cork. Yet, other properties, including its spherical morphology and hydrophobic character, must not be neglected. As previously reported [28], cork and pine fibers were used to produce polyurethane-based composites. This difference in shape (spherical in the case of cork and cylindrical in the fibers), together with their distinct hydrophobicity, affected their compatibility with the polymer matrix, contributing to the lower mechanical performance of cork-based composites. Similar results were observed by Fabijanski [36], Martins et al. [37], and Daver et al. [38]. Nevertheless, considering that, in the present case, the matrix is distinct, other issues may be associated with changes in the mechanical properties, namely, the reduction in the percentage of crystallinity of the PLA matrix as well as the presence in this matrix of some minor quantities of low molar components from the cork [39].

Furthermore, the recovered PLA exhibits similar mechanical performance when compared to the virgin counterpart, proving that it retains its mechanical properties after the recycling process. The maintenance of its mechanical performance is crucial. If these properties are compromised during recycling, the material may not meet the performance standards required for its original applications, reducing its value in the market. This finding contrasts with the work reported by Wang et al. [40], who mechanically recycled composites and observed a decline in the properties of the recycled materials due to thermo-mechanical degradation.

The thermal properties of all filaments were analyzed by DSC; the results are presented in Table 1 and Figure 4.

Upon analyzing the glass transition temperatures (Tgs) and cold crystallization temperatures (Tcrists) (see Table 1 and Figure 4, it is evident that all filaments exhibit similar values. Mazur et al. [41] investigated the mechanical, thermal, and microstructural properties of 3D-printed PLA composites with wood, bamboo, and cork. They reported a peak around 67–69 °C in the DSC thermogram, corresponding to Tg, which is associated with the relaxation of the PLA amorphous domains. Another notable temperature on the thermogram presented in Figure 4 was Tcrist, observed in the range of 102–103 °C, related to the crystallization of the material during heating for aliphatic polyesters. The addition of natural fillers did not significantly affect these temperatures. When comparing virgin PLA to the recovered counterparts, lower values were observed for the latter. This can be attributed to the increased molecular chain mobility in virgin PLA, while the lower values for the recovered PLA might be due to shorter molecular chains resulting from some PLA degradation, as suggested by Chaitanya et al. [42] in their study on the recyclability of PLA/Sisal fiber biocomposites. Additionally, as previously mentioned, some extractives and other low molar mass components, such as fatty acids from cork, may be present in the RPLA, which, despite not being detected by FTIR or NMR, can affect chain mobility. Indeed, it is well established, as reported by Donato et al. [43], that these moieties can act as plasticizers, enhancing molecular mobility and reducing Tg values. Moreover, their presence could contribute to a decrease in the Young’s modulus and an increase in elongation at break, consistent with prior observations.

Regarding the thermal stability of the samples, their properties were evaluated using TGA, with the results presented in Figure 5.

The TGA results of both cork and PLA samples reveal identical degradation profiles. The cork samples show a slight weight loss at around 100 °C, attributed to moisture evaporation. Following this, both cork and PLA samples exhibit a multistep weight loss pattern. The maximum decomposition temperatures for cork occur in the range of 330–420 °C, corresponding to the breakdown of hemicellulose and cellulose, and 475–575 °C, associated with the degradation of lignin [44]. In the case of PLA samples, all materials show a primary degradation step between 250 and 350 °C, corresponding to the thermal decomposition of the polymer chain, primarily due to the cleavage of ester linkages and breakdown of the polymer backbone [45]. The composite, being primarily composed of PLA, displays a thermal degradation profile similar to that of the PLA matrix. Ngaowthong et al. [29] investigated the recycling of sisal fiber-reinforced polypropylene (PP) and PLA composites through injection molding. They found that the thermal stability of the PP/sisal fiber composites was unaffected by repeated recycling, while the thermal stability of PLA-based composites decreased with the recycling process.

3. Materials and Methods

3.1. Materials

Cork powder, a byproduct of cork stopper production, sourced from the outer bark of Quercus suber L., was kindly provided by Corticeira Amorim. PLA, with a molecular weight (Mn) of approximately 100,000 g.mol⁻¹, as determined by SEC against PS standards, and a dispersity (Ð) of around 1.3, was supplied by Corbion Purac. Dichloromethane, PA grade, was supplied by Fisher Scientific (Hampton, NH, USA).

3.2. Production of Composites

Filaments of virgin PLA, recycled PLA, and cork/PLA composites (5% wt.wt⁻¹ of cork) were extruded using a 3Devo Composer 350 extruder (Utrecht, The Netherlands) at a speed of 3.5 rpm. The extrusion process followed a temperature profile of 170 °C, 190 °C, 185 °C, and 170 °C.

3.3. Recycling of Composites

Cork/PLA composites were dissolved in dichloromethane at ambient temperature, using a solvent/polymer ratio (v.wt⁻¹) of 8, for 30 min, under magnetic stirring. After the dissolution, the solid fraction was separated from the liquid fraction via decantation. Next, the recovered cork was washed using dichloromethane at ambient temperature, using a polymer ratio (v.wt⁻¹) of 8, for 30 min, under magnetic stirring. This procedure was repeated three times, then the solvent was recovered and reused. The yields of solubility/extraction of PLA (%_{yield_PLA}) and cork (%_{yield_cork}) were determined using Equation (1) and Equation (2), respectively.

$${\%}_{yield\_PLA}={\displaystyle \frac{{m_{initial PLA}-m}_{dissolved PLA}}{m_{initial PLA}}}\times 100$$

(1)

$${\%}_{yield\_cork}={\displaystyle \frac{m_{extracted cork}}{m_{initial cork}}}\times 100$$

(2)

where m_{dissolved PLA} is the mass of PLA present in the liquid fraction, m_{initial PLA} is the initial mass of PLA, m_{extracted cork} is the mass of cork extracted from the liquid fraction, and m_{initial cork} is the initial mass of cork.

3.4. Characterization

SEM analyses were conducted using a Hitachi SU-70 scanning electron microscope (Hitach, Tokyo, Japan), after vacuum-coating the samples with gold to prevent electrostatic charging during examination, with an accelerating voltage of 15.0 kV.

FTIR spectra were collected on a Perkin Elmer FTIR System Spectrum BX Spectrometer (Perkin Elmer, Springfield, IL, USA), equipped with a single horizontal Golden Gate ATR cell. Data were recorded at room temperature within the range of from 4000 to 500 cm⁻¹, accumulating 32 scans with a resolution of 4 cm⁻¹.

¹³C solid-state Cross Polarization—Magic Angle Spinning Nuclear Magnetic Resonance (¹³C CPMAS NMR) spectra were recorded on a Bruker Avance 400 spectrometer (Bruker, Singapore). Samples were packed into zirconia rotors sealed with Kel-FTM caps and spun at 12 kHz. Acquisition parameters included approximately 7000 scans, a 90° proton pulse, a cross-polarization contact time of 1 ms, and a recovery delay of 2.5 s.

Tensile tests of dog-bone specimens were performed using a Hegewald & Peschke (Nossen, Germany) universal testing machine (Inspekt solo) equipped with a 2.5 kN load cell, following the DIN EN ISO 527 standard. Tests were conducted at a speed of 10 mm.min⁻¹ up to the breaking point to determine the ultimate tensile strength, elongation at break, and Young’s modulus.

Differential scanning calorimetric (DSC) analyses were carried out using a Netzsch DSC 204F1 Phoenix (Exton, PA 19341, USA) at a heating rate of 10 °C.min⁻¹.

Thermogravimetric analysis (TGA) was conducted using a SETSYS Evolution 1750 thermogravimetric analyzer (Setaram, Lyon, France), from room temperature to 800 °C, at a heating rate of 10 °C.min⁻¹ under a nitrogen flux of 200 mL.min⁻¹.

4. Conclusions

In this study, cork/PLA bio-composites were recycled using dissolution–precipitation. It was observed that 80.9% ± 2.4 of cork and 85.9% ± 5.9 of PLA were recovered, with the materials retaining their original chemical structure and having cork maintained its honeycomb structure after both extrusion and recycling. While adding cork to PLA reduced both the Young’s modulus (from 884.4 MPa ± 52.2 to 490.3 MPa ± 35.4) and the elongation at break (from 46.1% ± 0.8 to 29.4% ± 1.5), the recovered PLA exhibited similar or even better properties to the virgin PLA (Young’s modulus of 826.0 MPa ± 78.5 and elongation at break 50.8% ± 1.4). Both virgin PLA and the composite showed Tg around 70 °C and Tcrist around 97 °C, although the recovered PLA had slightly lower values (53.2 °C and 92.8 °C, respectively), likely due to some degradation. Despite this, all recovered materials maintained similar thermal stability.

Cork can be depolymerized by dichloromethane, which extracts polysaccharides and other components, potentially transferring them into RPLA after solvent evaporation. However, spectroscopy analyses show no significant differences between virgin and recovered materials, suggesting that either no polysaccharides are removed during recycling or their removal is negligible. If extractives are present in the RPLA, they could act as plasticizers, contributing to reducing Tg, decreasing Young’s modulus, and increasing elongation at break.

Overall, the recycling process successfully separated cork from PLA, with the recovered materials retaining comparable properties, supporting the eco-efficiency of the composites. Hence, this work represents significant progress in the recycling of cork-based composites, where further research on multiple recycling cycles is still necessary to fully address the reuse potential of recycled cork and PLA.