Introduction

Current concerns regarding the environment and increasing awareness about sustainability are pushing forward the development of new composite materials, which incorporate renewed materials. Cork is a reliable and sustainable raw material.

Cork is the outer bark of the cork oak tree Quercus suber L. and presents tiny hollow cells of hexagonal shape in closed-cell foam. Its main chemical composition is 33–50% of suberin, 20–25% of lignin, 12–20% of polysaccharides and 14–18% of extractives [1]. Portugal is the world’s leading cork producer, and its main application is the production of stoppers. From stoppers production, a considerable amount of cork residues is generated (≈ 30 mass/%) [2]. The incorporation of these residues into polymeric matrices can be a suitable solution for the development of new materials solutions tailoring the needs of different applications.

Polylactic acid (PLA) is a biodegradable thermoplastic aliphatic polyester derived from renewable resources. It is a strong candidate to be used as a polymeric matrix due to its biodegradability and chemical compatibility. Different synthesis routes for PLA polymerization can be applied, and the most common one is the ring-opening lactide polymerization [3]. PLA stereochemical structure can be easily changed, depending on the amount of l- and d-isomers during polymerization [3]. This ratio of l- and d-isomers affects the mechanical and thermal properties of the obtained PLA. Regarding thermal properties, PLA can be a semi-crystalline or an amorphous material depending on its stereochemical structure and thermal history [4]. Moreover, PLA has a particular thermal behaviour in a way that multiphase transitions, such as glass transition, chain relaxation, cold crystallization and melting, are involved [5, 6]. The main drawback of PLA for commercial applications is its low crystallization rate. Several modifications have been proposed to improve the crystallization rate, mainly by the addition of nucleating agents [5, 7,8,9,10,11] or by the addition of plasticizers [12,13,14].

Additive manufacturing (AM) offers a set of techniques that opens a possibility of creating new products with a high level of design freedom. Fused filament fabrication (FFF), one of the AM techniques, is an extrusion-based process, in which a plastic filament is heated and selectively extruded via a nozzle layer by layer [15]. It is a disruptive technology on a constant growing market demanding for materials innovation. The combination of biodegradable polymer matrices with natural materials could be a sustainable solution for the development of new materials for AM. More specifically, considering the actual technological demand, cork–polymer composites (CPC) should be used to open new horizons in AM.

Cork–polymer composites (CPC) are a viable and a possible strategy for upgrading cork industrial residues based on sustainable development. The incorporation of lignocellulosic materials into polymeric matrices can bring several advantages. Some examples of these advantages are biodegradability, low density and hardness, low cost, easy availability, high levels of filling, good relation between strength and mass, good insulation/noise absorption properties and non-toxicity [16, 17]. Among other factors, CPC mechanical performance is influenced by the interaction of polymer and filler and by the crystallization behaviour of the matrix in the presence of the filler. Natural fibres and processing conditions can have an effect on polymer crystallization behaviour. Oliveira et al. [18, 19] showed that cork acted as a nucleating agent in a polypropylene (PP) matrix. To the better of our knowledge, few studies analysed the effect of cork on PLA crystallinity [20,21,22,23] and, so far, none of the studies are related to the crystallization kinetics. In this work, the physical/chemical changes in PLA during its melting, the non-isothermal cold crystallization kinetics of PLA and CPC and the nucleating ability of cork in a PLA matrix will be studied. The influence of cork on the nucleation and the crystal growth behaviour of PLA will be analysed by Avrami [24, 25] and Tobin [26,27,28] models. Crystallization activation energies (Ec) will be determined through Kissinger and Friedman methods [29, 30]. In addition, the morphology and crystalline structure of PLA and CPC were evaluated recurring to scanning electron microscopy (SEM), polarized optical microscopy (POM) and X-ray diffraction (XRD) techniques.

Experimental

Materials

Cork powder from a Portuguese cork producer was used. The material was fractionated through sieving (Retsch, Germany), and the fraction retained in the sieve of 20 μm was kept. Cork powder was dried in a vacuum oven (343 K) for 24 h. The average pore diameter, as well as the bulk density of cork powder, was measured by mercury intrusion porosimetry (MIP). The experiment was conducted in a Micromeritics AutoPore IV 9500 apparatus with pressure ranging from 0.3 to 227 MPa. The pore diameter was obtained by the Washburn equation [31], considering a surface tension of 0.485 N m−1 and a contact angle of 130° between cork and mercury.

The polymeric matrix used was an Ingeo™ Biopolymer PLA 4032D purchased from NatureWorks with a stereoisomer composition of 1.2–1.6% D-isomer lactide and a melting point between 428 and 443 K.

CPC compounding

Before compounding, the cork particles and the PLA were dried at 343 K for 24 h, in a vacuum oven (Carbolite AX60 model), to stabilize the moisture content. CPC formulation was made in a Brabender-type internal mixer. Initially, PLA pellets were charged and melted at 463 K, for 2 min at 40 rotations per minute (rpm), and then cork powder was added for additional 8 min. Previous studies reported that cork structure and composition do not alter at temperatures below 523 K [32, 33]. After compounding, the mixture was granulated in a Dynisco granulator into granules (0.5–1.0 mm). CPC formulation contains 85 mass/% of PLA and 15 mass/% of cork powder, in terms of volume of 45 volume/% and 55 volume/%, respectively [18, 34]. The selection of this mass ratio was to guarantee the highest amount of cork incorporated into PLA, in order to obtain a CPC filament with a non-plastic and natural touch, similar to cork.

Characterization by SEM

SEM Hitachi SU-70 equipment was used to morphologically analyse cork powder particles and cork dispersion into PLA matrix after compounding. Samples were mounted on aluminium stubs and sputter-coated (Polaron E5000) with Au/Pd target for 2 min at 12 mA.

Characterization by DSC

Thermal properties and non-isothermal crystallization behaviour of PLA and CPC were studied using Shimadzu DSC-60 equipment. All experiments were carried out in nitrogen atmosphere (50 mL min−1). Samples between 9.0 and 10.0 mg were hermetically sealed in aluminium pans. Each sample was heated from 293 to 473 K at a scan rate of 20 K min−1 and held for 5 min at this temperature to eliminate the thermal history. Then, the samples were cooled at 20 K min−1 to room temperature (≈ 293 K) and heated again up to 473 K. Four different heating rates were used, namely 1.25, 2.5, 5 and 7.5 K min−1. The selection of the cooling rates was based on the low kinetics of PLA. A study of Miyata and Masuko [35] showed that samples cooled at rates higher than 10 K min−1 did not crystallize and remained amorphous. Only the second run was considered for analysis. Glass transition temperature (Tg), cold crystallization temperature (Tcc), melting temperature (Tm), cold crystallization enthalpy (ΔHcc) and melting crystallization enthalpy (ΔHm) of PLA and CPC were determined from the DSC thermographs. The crystallinity degree (\(X_{\text{c}}\)) of samples was calculated using Eq. (1):

$$X_{\text{c}} = \frac{{\Delta H_{\text{cc}} }}{{\Delta H_{\text{m}}^{0} \left( {1 - m} \right)}} \times 100,$$
(1)

where \(m\) is the mass percentage of the filler and \(\Delta H_{\text{m}}^{0}\) is the melting crystallization enthalpy for 100% crystalline PLA (93 J g−1) [36].

Characterization by POM

POM was performed to evaluate the nucleation and spherulites growth in pure PLA and in PLA in the presence of cork powder. The observations were made on a Nikon Eclipse L150. Firstly, both samples were melted at 463 K for 5 min to remove thermal history. Then, the samples were cooled down to 393 K. Isothermal crystallization took place at 393 K for 15 min.

Characterization by XRD

Panalytical X’Pert Pro 3 equipment with a Cu Kα (λ = 1.5406 Å) radiation source was used. Diffraction intensities were measured from 10º to 40º with a step size of 0.02. XRD measurements were performed at 298 K.

Theoretical principles

Non-isothermal kinetic parameters were determined by Avrami [24, 25], Tobin [26,27,28], Kissinger [29] and Friedman [30] models. The relative degree of crystallinity at time t (\(X_{t}\)), as a function of temperature (T), was determined through Eq. (2):

$$X_{\text{t}} = \frac{{\mathop \smallint \nolimits_{{T_{0} }}^{T} \left( {\frac{{{\text{d}}H_{\text{cc}} }}{{{\text{d}}T}}} \right){\text{d}}T}}{{\mathop \smallint \nolimits_{{T_{0} }}^{{T_{\infty } }} \left( {\frac{{{\text{d}}H_{\text{cc}} }}{{{\text{d}}T}}} \right){\text{d}}T}},$$
(2)

where \(T_{0}\) and \(T_{\infty }\) are the temperatures corresponding to the onset and end of the crystallization process, respectively.

In the case of the non-isothermal cold crystallization process, the relationship between the crystallization time (\(t\)) and the corresponding temperature (T) can be obtained by Eq. (3):

$$t = \frac{{\left| {T_{0} - T} \right|}}{\varphi },$$
(3)

where ϕ is the heating rate.

Avrami model [24, 25] is commonly used to describe the isothermal crystallization kinetic behaviour and analyse the increase in relative crystallinity with time:

$$1 - X_{\text{t}} = \exp \left( { - Z_{\text{t}} t^{\text{n}} } \right),$$
(4)

where \(Z_{t}\) is the crystallization rate constant, containing the nucleation and growth rate parameters, and it is temperature dependent, and \(n\) is the Avrami index which is dependent on the nucleation type and on the growth geometry of the crystals. The linearized form can be written as follows:

$$\log \left[ { - \ln \left( {1 - X_{\text{t}} } \right)} \right] = \log \left( {Z_{\text{t}} } \right) - n{ \log }\left( t \right)$$
(5)

Mandelkern [37] considered that the primary non-isothermal crystallization stage can be described through Avrami model, based on a constant crystallization temperature assumption. The \(n\) and \(Z_{\text{t}}\) parameters do not have the same physical meaning as in the isothermal crystallization processes, since the temperature changes constantly during non-isothermal crystallization. Jeziorny [38] calibrated the \(Z_{t}\) parameter, considering the temperature dependence of the non-isothermal crystallization, correcting the crystallization rate constants by introducing heating rate:

$$\log Z_{\text{c}} = \frac{{\log (Z_{\text{t}} )}}{\varphi }$$
(6)

The time required to achieve 50% of crystallization is called half-time of crystallization (t1/2), and it is considered as a prime parameter for investigating the kinetics of crystallization process. The t1/2 for non-isothermal crystallization can be obtained from the following expression [39]:

$$t_{{{\raise0.7ex\hbox{$1$} \!\mathord{\left/ {\vphantom {1 2}}\right.\kern-0pt} \!\lower0.7ex\hbox{$2$}}}} = \left( {\frac{\ln 2}{{Z_{\text{t}} }}} \right)^{{{\raise0.7ex\hbox{$1$} \!\mathord{\left/ {\vphantom {1 n}}\right.\kern-0pt} \!\lower0.7ex\hbox{n}}}}$$
(7)

Tobin [26,27,28] model was utilized to study the phase transformation kinetics with growth site impingement on the non-isothermal crystallization kinetics of PLA and CPC (Eq. 8):

$$X_{\text{t}} = \frac{{K_{\text{T }} t^{{{\text{n}}_{\text{T}} }} }}{{1 + K_{\text{T}} t^{{\text{n}_{\text{T}} }} }},$$
(8)

where KT and nT correspond to Tobin crystallization rate constant and Tobin exponent, respectively. The nT parameter provides information about the type of nucleation and growth mechanism involved in the non-isothermal crystallization process. The linearized form of Eq. (9) is rewritten as follows:

$$\ln \left( {\frac{{X_{\text{t}} }}{{1 - X_{\text{t}} }}} \right) = \ln K_{\text{T}} + n_{\text{T }} \ln t$$
(9)

Kissinger [29] method was applied to determine Ec for non-isothermal cold crystallization of neat PLA and CPC. Ec is the energy required for the movement of PLA molecular segments from melt to the crystal growth surface. It is determined by the variation in Tcc with respect to \(\varphi\), as described in Eq. (10):

$$\frac{{ - E_{\text{C}} }}{R} = \frac{{{\text{d}}\left( {\ln \left( {\frac{\varphi }{{T_{\text{cc}}^{2} }}} \right)} \right)}}{{{\text{d}}\left( {\frac{1}{{T_{\text{cc}} }}} \right)}},$$
(10)

where \(R\) is the gas constant and \(T_{\text{cc}}\) is the cold crystallization temperature.

An isoconversional method developed by Friedman [30] was also used to determine Ec. Contrarily to the Kissinger method, where a single value of Ec is determined, regardless of the kinetics of the system, from the Friedman method it is evaluated the dependency of Ec with crystallinity and temperature. According to Friedman, different effective activation energies can be calculated for each \(X_{t}\) considering Eq. 11:

$$\ln \left( {\frac{{{\text{d}}X}}{{{\text{d}}t}}} \right)_{\text{X},\text{i}} = {\text{contant}} - \frac{{E_{\text{c}} }}{{RT_{\text{X},\text{i}} }},$$
(11)

where i represents each heating rate applied, (dX/dt) is the instantaneous crystallization rate as a function of time at a given conversion \(X\) and \(T_{{\text{X}},{\text{i}}}\) is the temperature associated with a given conversion \(X\) at different i’s. At a given \(X_{\text{t}}\), \(E_{\text{c}}\) can be calculated from the slope (\(E_{\text{c}} /R\)) of the correlation between ln(dX/dt) versus 1/\(T_{X}\).

Results and discussion

Cork powder characterization

MIP analysis was applied to characterize cork powder in terms of density and porosity. Table 1 shows the average value of pore diameter, the porosity and bulk and apparent densities.

Table 1 Cork powder characterization
Full size table

According to IUPAC [40], cork powder pores are classified as macropores, since they present an average diameter greater than 50 nm. Cork powder presents low density and high porosity, which is mainly due to its characteristic honeycomb structure, as shown in Fig. 1.

Fig. 1
figure 1

SEM images of cork powder

Full size image

Non-isothermal heating behaviour

Non-isothermal DSC curves of neat PLA and CPC from the second heating scan at different heating rates are represented in Fig. 2. Exothermic cold crystallization peaks appear for both, pure PLA and CPC. The PLA molecular chains become mobile above Tg, allowing crystallization. Above Tg, molecular chains possess the ability to be partly ordered and the cold crystallization takes place. These crystals melt upon further heating. For PLA, cold crystallization peaks become smoother as the heating rate increases. It suggests that lower scan rates during heating can promote PLA crystallization [6]. In Fig. 2, the Tm peaks for PLA and CPC are also visible.

Fig. 2
figure 2

DSC thermographs, in the second heating cycle, at different heating rates for a PLA and b CPC

Full size image

There are double melting peaks for both samples, which are usually ascribed to simultaneous occurrence of thinner lamellae melting and recrystallization. More specifically, the lower Tm peak is usually associated with the melting of some original crystals grown by primary crystallization, while the higher Tm peak is associated with the melting of crystals formed through a melt–recrystallization process during the heating process in DSC measurement [7, 41]. Other studies, involving biomaterials, allowed us to visualize the presence of double Tm peaks in the crystallization behaviour of PLA [20, 42]. The presence and reorganization of different types of crystals with different stabilities were the explanation given by the authors to the appearance of these double melting peaks. As the heating rate increases, these double melting peaks became weak for both samples. The same tendency was reported by Zhang and co-workers [41]. They appointed a decrease in the rate of crystallization based on the self-adjustment of the amorphous PLA chains. The non-isothermal crystallization parameters obtained from Fig. 2 and equations are listed in Table 2.

Table 2 Non-isothermal crystallization parameters of PLA and CPC materials
Full size table

The Tg values for PLA and CPC slightly increase with heating rate, which can suggest the confinement of the amorphous mobile phase [20]. In fact, for each material, the Tcc values also increase with heating rate, supporting the reduced ability of PLA to recrystallize on heating above Tg. In addition, as Tcc shifts to higher values with heating rate, it implies that the crystallization process occurs earlier for lower heating rates [41]. At a given heating rate, the values for Tcc of CPC were lower than for neat PLA. This indicates that cork can promote the initial cold crystallization of the PLA matrix due to the heterogeneous nucleation effect. Values of ∆Hcc present a downward trend as the heating rate increases. Considering that all samples were initially amorphous, as a consequence of the fast cooling rate applied, this trend indicates that, as the heating increases, polymer chains did not have sufficient time to form crystalline structures. In parallel, ∆Hm values exhibit the same trend as ∆Hcc. This can be an indication that PLA melting behaviour depends on the crystal structures created during cold crystallization phase. A study of Nofar et al. [5] reported the same behaviour. From the \(X_{\text{cc}}\) values, it is noticed that the amount of crystallinity developed in PLA is dependent on the heating rate. \(X_{\text{cc}}\) values increased with the addition of cork, indicating that cork facilitated the crystallization of PLA during heating. Cork may act as a heterogeneous nucleating agent. The ability of cork to modify the crystallization behaviour of PLA may be associated with the increased density of nucleating sites provided by cork particles. This nucleating ability of cork was also observed in other studies considering bio-based polyesters acting as matrices [20, 22, 43]. Figure 3 represents the non-isothermal DSC curves of neat PLA and CPC at 1.25 K min−1.

Fig. 3
figure 3

DSC thermographs of PLA and CPC, at a heating rate of 1.25 K min−1

Full size image

As referred above, the presence of cork shifted Tcc to lower values, showing the nucleating ability of cork. Its addition also led to a more well-defined and sharp cold crystallization peak, which usually corresponds to a well-defined crystal structure. This suggests that, somehow, the addition of cork resulted in a rearrangement of PLA molecules leading to a more defined crystal structure.

Non-isothermal cold crystallization kinetics

The relative crystallinity curves versus crystallization time for the PLA and CPC studied samples are shown in Fig. 4. At different heating rates, curves exhibit the same sigmoidal shape. The first nonlinear part is usually considered the nucleation step of the crystallization process [44]. The longest nucleation step occurs for PLA at 1.25 K min−1, resulting in the formation of more nuclei. The curvature of the second nonlinear part levelled off that is attributed to the impingement or crowding of spherulites at the later stage of the crystal growth [45].

Fig. 4
figure 4

Relative crystallinity (Xt) versus time (t) for a PLA and b CPC

Full size image

Higher heating rates have managed to complete the PLA crystallization process in a shorter time. This behaviour is also observed through t1/2 parameter (Table 3). A shorter t1/2 value means a faster crystallization process, where a shorter time is needed to achieve 50% of crystallization fraction. It is visible a decrease in t1/2 for all heating rates when PLA crystallizes in the presence of cork when compared to neat PLA. Once again, the nucleating effect of cork is observed. The higher t1/2 values observed for PLA and CPC at lower heating rates can be attributed to the secondary crystallization process.

Table 3 Half-time crystallization and Avrami kinetic parameters
Full size table

Avrami model

Avrami plots and kinetic parameters are shown in Fig. 5 and Table 3, respectively, obtained from the application of Eqs. (5) and (6). It was considered that \(X_{\text{t}}\) values range from 10 to 80%.

Fig. 5
figure 5

Avrami plots of a PLA and b CPC

Full size image

Avrami model was adequate to describe the non-isothermal cold crystallization kinetics of neat PLA and CPC. (R2 is equal to 1 for almost all heating scan rates.) It should be noted that Avrami model does not have the same physical meaning as in the isothermal crystallization, owing to the constant variation of temperature in non-isothermal crystallization.

This temperature dependence will have an effect on the rates of both nuclei formation and spherulite growth. \(Z_{\text{c}}\) values increased with heating rate, demonstrating that cold crystallization rate increases with heating rate. The n values for PLA and CPC are found in the range of ~ 2–3. Non-integer n values reveal a combination of thermal and athermal nucleation mechanisms [7]. The average n values for PLA and CPC were 2.78 and 3.02, respectively, indicating that the non-isothermal cold crystallization was initiated by an athermal nucleation and followed by a mixed two- and three-dimensional spherulite growth [7, 14].

Tobin model

Tobin model was considered to overcome the limitation of Avrami model in the study of the early stage of crystallization process. More specifically, Tobin model is able to describe the phase transformation kinetics with growth site impingement as well as secondary crystallization process [26,27,28]. Contrarily to Avrami exponent, Tobin exponent nT does not need to be integer, since it is controlled directly by different types of nucleation and growth mechanisms [10].

Tobin plots for PLA and CPC obtained at different heating rates are shown in Fig. 6, and the related parameters are presented in Table 4. An \(X_{\text{t}}\) range of 10–80% was used for calculations.

Fig. 6
figure 6

Tobin plots of a PLA and b CPC

Full size image
Table 4 Tobin kinetic parameters
Full size table

The nT average value of PLA is 3.60 and CPC is 3.93. For both cases, nT values are almost independent of the heating rate.

Concerning the KT parameter, it increases as the heating rate increases. The KT values for CPC are larger than for PLA, which indicates that the presence of cork accelerates the non-isothermal cold crystallization of PLA. Tobin parameters describe the same physical significance analogous to that of Avrami model. In this study, Tobin parameters exhibited a similar tendency to those obtained for Avrami analysis.

Comparison between Avrami and Tobin models

The efficiency of both models on describing the non-isothermal crystallization kinetics of PLA and CPC can be assessed from Fig. 7. For each heating rate, the curve that correlates Xt versus t (as in Fig. 4) was used as a reference. Considering the kinetic parameters displayed in Tables 3 and 4, as well as Eqs. (4) and (8), the fitting accuracy of Avrami and Tobin models was determined.

Fig. 7
figure 7

Comparison between Avrami and Tobin models—relative crystallinity (Xt) as a function of time (t) for a PLA and b CPC

Full size image

According to Fig. 7, a similar tendency was observed for PLA and CPC samples. The accuracy of Avrami model fitting becomes higher as the heating rates increase. On the contrary, for each heating rate, Tobin model fitting underpredicts the evolution of relative crystallinity. This tendency is more accentuated at higher relative crystallinity values (Xt ≥ 75%). Working with syndiotactic polypropylene, Supaphol [46] observed a deviation of the Tobin model fitting, which was attributed to the use of a simplified form of Eq. (8) or it could be associated with the overprediction of the impingement effect.

Activation energy

Kissinger plots are shown in Fig. 8, and the corresponding Ec values are listed in Table 5. The plots exhibit good linearity, and regression coefficients (R2) values are obtained with the values of ~ 0.990.

Fig. 8
figure 8

Kissinger plots for PLA and CPC

Full size image
Table 5 Activation energies for PLA and CPC according to Kissinger method
Full size table

The addition of cork resulted in a decrease in the Ec, from 141.5 to 63.3 kJ mol−1, suggesting that the energy barrier of cold crystallization is lower for CPC, causing an increase in crystallization ability for CPC as compared with PLA, an expected result since cork acted as a nucleating agent, as presented before. Comparable results have also been found for the cold crystallization activation energy of PLA in the presence of multiwalled carbon nanotubes [47], clay [6] and a combination of several nucleating agents [48].

The dependency of Ec as a function of Xt was evaluated according to the Friedman method. Xt values ranging from 0.1 to 0.8 were considered, and the value of R2 was obtained from 0.95 to 0.99. Figure 9 shows Ec versus Xt for both materials, and the corresponding average Ec values are listed in Table 6.

Fig. 9
figure 9

Dependency of Ec as a function of Xt for PLA and CPC

Full size image
Table 6 Average activation energies (\(\bar{E}_{\text{c}}\)) for PLA and CPC according to Friedman method
Full size table

The Ec values indicate that the energy barrier for the cold crystallization process is lower for the composite than for the unfilled PLA. The same tendency was observed in similar studies [49, 50]. The results are also in agreement with those obtained in the previous kinetic analysis, reinforcing the nucleating ability of cork. In fact, for Xt equal to 0.1, the Ec values acquired from Kissinger method are similar to Ec values from Friedman method. This can be indicative that the Kissinger method can represent one Ec value obtained by Friedman, as also observed by Li et al. [7] and Ries et al. [50].

Morphology and POM observations

The morphology of CPC was evaluated by SEM, and it is represented in Fig. 10.

Fig. 10
figure 10

SEM images of CPC after compounding

Full size image

It is seen that cork particles are homogeneously dispersed and imbibed within PLA matrix. Cork particles are completely covered by PLA (highlighted in orange).

The evaluation of cork powder as a nucleating agent in PLA matrix was performed through POM. Optical micrographs of pure PLA and CPC after 15 min at 393.2 K are shown in Fig. 11.

Fig. 11
figure 11

Optical micrographs of a–c PLA and b–d CPC (15 min at 393.2 K)

Full size image

It is revealed the nucleating ability of cork powder in PLA matrix, already determined by the non-isothermal crystallization kinetic analyses. Within 15 min, CPC exhibited higher nucleation density and smaller spherulite size when compared to pure PLA. The same tendency was visualized by Li et al. [51].

XRD

XRD analyses were performed to evaluate PLA and CPC crystalline structures. PLA exhibits several polymorphisms during melt or cold crystallization, such as α-, β- and γ-forms [52, 53]. The α-form is considered the most stable, and it is produced at high crystallization temperatures (> 393.2 K), while a disordered α′-form can be developed at lower temperatures [52, 54, 55]. A combination of these α forms is usually made in industrial processes to decrease PLA processing temperature. Figure 12 displays the XRD patterns of PLA and CPC measured at 298.2 K.

Fig. 12
figure 12

XRD patterns of PLA and CPC

Full size image

For PLA, two characteristic reflection planes of α-phase crystals can be identified, namely (200)/(110) and (203), with scattering angles (2θ) of 16.9° and 19.2°, respectively [52, 54]. As for CPC, stronger peaks at 16.9° and 19.2° can be observed, indicating that cork can improve the crystallization of PLA. In addition, the presence of cork during the cold crystallization of PLA alters its crystalline structure, which is reflected by the appearance of α-phase characteristic peaks, namely at 11.9° (110), 22.6° (015) and 27.4° (207) [52, 54, 56]. These findings are in agreement with those obtained by the non-isothermal crystallization kinetic analyses.

Conclusions

The non-isothermal cold crystallization behaviours of PLA and CPC were investigated at different heating rates (1.25, 2.5, 5 and 7.5 K min−1) through DSC measurements. In the presence of cork, Tcc shifted to lower values, revealing that cork facilitates the initial cold crystallization of the PLA matrix. Additionally, an increase of ~ 300% of Xcc was observed when PLA cold crystallizes in the presence of cork. Avrami and Tobin models were applied to analyse the non-isothermal kinetics of the samples. Both methods successfully described the non-isothermal cold crystallization processes of PLA and CPC. The same tendency was attained for both methods, where the addition of cork increased the crystallization rate of PLA matrix and n values indicated a mixed two- and three-dimensional crystallization growth. These results were supported by POM and XRD analyses. From the Kissinger and Friedman methods, lower ΔEC values for CPC were obtained, indicating a reduction in energy barrier for crystallization compared to neat PLA.

This study allows us to understand the cold crystallization behaviour of CPC filament during FFF, which is considered a fast local cooling process. As reported above, the addition of cork to PLA considerably increased the crystallization rate and Xcc, revealing the ability of PLA to crystallize rapidly in the presence of cork, as experienced during FFF. This will have an impact on the mechanical behaviour of printing parts. The presence of small spherulites will enhance ductility and impact strength. It will also help in defining printing settings and in evaluating layers adhesion and warpage.