Fabrication and characterization of a filament for 3D printing from polylactic acid with Cryptostegia grandiflora fiber

Since 1984, 3D printing, also referred to as additive manufacturing (AM), has been used for commercial purposes related to rapid prototyping, which has grown tremendously to reach consumer applications in the household [1]. Also, 3D printing technology has become a new industrial growth area in fields, such as medical research, aerospace, defence, precision goods, and jewelry manufacturing, due to its benefits of reducing manufacturing waste, simplifying assembly, and mass customization. The principal materials used for 3D printing filament production are high impact polystyrene (HIPS) [2], thermoplastic elastomer (PCTPE) [3], transparent polycarbonate (PC) [4], acrylonitrile butadiene styrene (ABS) [5], nylon (polyamide, PA) [6], and polylactic acid (PLA) has attracted particular interest being a bio-based material and, at least in principle, has been adapted to be coupled with other biomass materials [7]. By 2026, the filament market is expected to grow to $6.6 billion [8].

PLA, an important alternative polymer to replace petrochemical polymers, is mainly produced from agricultural products such as sugarcane and corn starch, which means it to be affected under appropriate conditions by hydrolytic degradation [9]. Presently, PLA is among the primary feedstock materials for desktop 3D printers because of its advantageous mechanical properties and environmentally friendly synthesis processes from renewable resources [10].

Considering the limits of PLA, including its limited resistance to shear, hardness, toughness, and not exceptional stiffness [11], filling it with particles that are able to offer some enhancement of these properties is a possible strategy to be pursued and that had some success in recent years.

In practice, the fillers for addition can be either ceramic or lignocellulosic, in both cases giving preference, in terms of circular economy, to secondary raw materials obtainable from waste/by products of an economic system. As regards the former, in a recent study, a composite PLA filament was fabricated by blending 10–30 wt% of hydroxyapatite (HA) powder with PLA by using a screw extruder for the fused deposition mode (FDM) process [12]. Due to the concern about the possible reduction of thermal degradation temperature, attempts to fill PLA filaments with carbon from agro-waste (coconut husk) have also been proposed [13]. Both ligneous (nutshell powder) (up to 2.5 wt%) and ceramic waste (eggshell powder) (up to 5 wt%) were used in the study of Lohar et al. [14], and the improvement in terms of tensile strength was found to be very significant for the former filler, while not for the latter.

In general terms, therefore, it appears that the addition of lignocellulosic materials to PLA 3D filaments would be beneficial. A number of examples do exist actually. In particular, various biomass-based wastes have been added to filaments for FDM, including hemp fibers (from 15 to 25%), hemp leaves (from 10 to 15%), tomato scraps (from 15 to 25%), carob flour (from 10 to 20%), and orange pruning waste (from 10 to 20%) [15]. Tomato and pruning waste offered substantial improvements toward pure PLA. Another possibility is the use of lignin and hemicellulose as filament fillers, following extraction from biomass waste, in a biorefinery concept: it is noteworthy that this involves further use of some chemicals, though ultimately proving to be effective [16]. More simply and immediately, though possibly with concerns over the results, PLA 3D filaments can be filled with different natural fibers, such as hemp, phormium, flax, coir, and bamboo; the amount of added fillers not to impair FDM functioning normally does not have to exceed 25 wt% [17].

In this research, a 3D printing composite filament was fabricated by blending the Cryptostegia grandiflora fiber (CGF) with PLA. The CGF was already investigated for potential introduction in polymer composites, as a fiber with cellulose content close to 80% and significant crystallinity, resulting in an average tensile strength as high as 791 MPa [18]. In this work, its introduction into the PLA filament has been attempted, and the relevant properties were investigated, such as tensile strength measurement, thermal analysis (TA), chemical investigation by Fourier transform infrared (FTIR) spectroscopy, crystallinity evaluation by X-ray diffraction (XRD), and surface morphology evaluation by microscopy. According to the results mentioned above regarding introducing agricultural waste, it is not deemed necessary to introduce more than a minimal amount of CGFs, which would be sufficient to prove the suitability of the process.

After being ground, particles with a maximum dimension of 80 µm and short fibers, spanning up to a few mm length, were selected for further analysis, as reported above; the CGF filler was further filtered using a 10 µm polyethylene filter to keep the powder within specific dimensions to be possibly used as the reinforcement for composite filament fabrication.

Following that, to verify the suitability of the CGF with this dimension, using the MFI value as the parameter, the flow ability of the polymer in the molten state was measured at 200°C. In practice, when introducing 10 wt% of fiber loading, compared to pure PLA, the MFI of the composite filament is decreased by 14%, from 6.1 (±0.1) g/10 min to 5.2 (±0.1) g/10 min. This appears suitable for FDM printing, considering that most commercial PLA filaments’ MFI range from 8.3 to 4.3 g/10 min [20].

To assist in the comparison with other natural fillers added to 3D printing filaments, some data are reported in Table 1. From these data, MFI values are very scattered and not very significant, though it can be recognised that the specific case examined yielded a sufficient value of MFI to allow 3D printing.

The morphological examination of the filament structure, as studied by optical microscopy, whose image is displayed in Figure 2, suggests sufficient adhesion of the CG particles, most of which are between 5 and 10 µm, to the PLA matrix. Despite this, the fracture may have occurred also partly by decohesion of the CG fibril from the matrix, as observable, or by differential deformation of the two, considering the bulging CG fibril. However, some of them produce evident damage, and the distribution of the particles might not always be considered uniform, though the diameter of the filament did not vary by more than ±3%, preserving the level of tolerance offered by pure PLA. This is promising, since in other cases more evident scattering is observed, due to the introduction of harder particles, such as corn cob fragments, in it [24]. The tensile properties, reported in Table 2, suggest an increase of strength and stiffness exceeding the proportion of the added filler, whilst on the other hand the strain to failure is greatly influenced.

Figure 2

Optical microscopy image of a section of PLA/10% CGF filament.

As far as FTIR spectroscopy is concerned (Figure 3), the distinct bands of pure PLA are observed at 768, 864, 1,033, 1,182, and 1,458 cm⁻¹. In particular, the transmittance peak at 768 cm⁻¹ is produced by the deformation of the CO and CH groups in polysaccharides connected to aromatic rings [25]. The additional absorption peak at 1,182 cm⁻¹ is associated with the CO bond [26]. The C–H group’s bending vibrations and the C–C group’s stretching vibrations are responsible for the peaks observed at 864 and 1,458 cm⁻¹, respectively. The pyranose chain vibrations of the C–H group are attributed to the absorption band seen at 1,033 cm⁻¹ [27].

Figure 3

FTIR spectra for pure PLA, and CGF powder, and their 90/10 composite.

In contrast, the characteristic peaks of the CGF spectra, mostly related to fiber cellulose, are found at 716, 898, 1,165, 1,431, and 1,633 cm⁻¹. More specifically, bending of the CH group of the anhydride ring is indicated by the peak at 716 cm⁻¹ [25]. The degradation of C–O–C groups attributed to amorphous cellulose is indicated by the peak observed at 898 cm⁻¹ [28]. The absorption peak at 1,165 cm⁻¹ is ascribed to the asymmetric C–O–C bridge stretching [29], whilst the peak at 1,431 cm⁻¹ is the result of hydroxide in-plane deformation [30]. The peak at 1,633 cm⁻¹ is attributed to water interaction and therefore to hydroxide vibration [31]. A specific peak is reported only for CGF at 3,196 cm⁻¹, due to the OH group vibration of carboxylic acid [32].

As far as the spectra of PLA/10% CGF filament are concerned, most of the peaks are repeated from either PLA or CGF. A specific peak that offers indications on the interaction of the two components is at 1,747 cm⁻¹: this has been attributed to the stretching vibration of the C&O group, most likely due to acetyl and carbonyl groups in hemicelluloses, the signal being normally concealed in the bulk of the fiber [33]. However, in general terms, the large similarity between the spectra of pure PLA and 10 wt% CGF composite indicates that the addition of CGF did not substantially modify the chemical composition of the pure PLA during blending and extrusion. As the consequence of substantially unaltered properties, it is largely expected that the mechanical properties would not decline by filler addition.

TGA curves of pure PLA and PLA + 10 wt% CGF composite filaments are shown in Figure 4. The moisture present in the sample evaporates and mechanically looser particles are gradually dissolved, which leads to a minor weight loss observed in both samples between 95 and approximately 250°C. Following this, the deterioration of the two filaments takes place, starting slightly earlier for pure PLA filament, as it is also revealed by the graphic measurement of the degradation peak, according to the method used, e.g., in the study of Johny et al. [34]. The situation appears different in the residue measured at 600°C, which as for PLA appears in alignment with the literature, suggesting a minimal char residue at that temperature [35], and even slightly lower in the case of presence of CGF, which is low in lignin (2.5%), therefore basically degrades only as the effect of the collapse of the crystalline cellulose structure [36]. The conclusion that can be obtained from TGA is that the filler slightly shifts up the degradation of the filament, which can be considered positive.

Figure 4

TGA curves for pure PLA and its 90/10 composite with CGF.

XRD patterns of pure PLA, pure CGF, and 10 wt% CGF/PLA composite filament are shown in Figure 5. The PLA polymer has a fully amorphous structure, according to the broad peak at 16.4° [37] and a smaller peak at 29.2°, observed also in the study of Revati et al. [38], yet not discussed there. The XRD pattern of cellulose I in CGF shows peaks at 2θ values of around 14.72, 16.64, 22.58, and 34.24° related to their respective reflection planes (101), (110), (200), and (004) [39]. The XRD pattern of CGF allowed measuring the crystallinity index (CI) from Segal’s law [40], according to the formula based on the corresponding crystallinity peak height I ₀₀₂ and amorphous local minimum I _am, namely 100 × (I ₀₀₂ − I _am)/I ₀₀₂, which yields a value of CI equal to 79.7%. Also, the smaller peaks suggest that CGF have relatively large crystallite size, so that (101) and (110) are not superposed. Calculation was done according to the Scherrer’s formula of the crystallite size D ₀₀₂ according to the diffraction peak broadening in the XRD curves, in practice, using equation (1): (1) D 002 = k λ / ( β 002 × cos θ ) , {D}_{002}=k\lambda /({\beta }_{002}\times \hspace{.25em}\cos \hspace{.25em}\theta ), where k is a constant, normally considered equal to unity, λ is the X ray wavelength used in XRD (in nm), and β ₀₀₂ is the true peak broadening at full-width half maximum (FWHM) in radians. The value obtained for the crystallite size was 41.6 nm.

Figure 5

XRD spectra of CGF, PLA, and its composite.

Although no specific calculations were performed, the notable appearance of the (110) peak in the 90% PLA/10% CGF filament XRD spectrum suggests that the crystallinity of the filament is enhanced by the introduction of CGF powder, even in small amounts. This is also well recognized with other plant fibers introduced in limited quantity into PLA composites, such as it is the case for Borassus flabellifer [41]. It is expected that this high value of crystallinity for the filament would enhance its performance; however, the possibility to introduce a higher amount of filler depends on whether it increases its brittleness or not, by representing an inclusion in the polymer structure.

Morphological investigation using SEM micrographs suggests that with respect to the bare PLA filament section shown in Figure 6(a) the composite filament is slightly rougher and less uniform, with obvious presence of some small bulges, obtained by the introduction of the filler, as reported in Figure 6(b). The composite filament section in Figure 7 with a larger magnification (300×) shows, on one hand, the visible formation of an effective interface between the matrix and the CGF, yet on the other hand slight disruptions (e.g., crack formations around CGF coalescing clusters) are also apparent.

Figure 6

50× SEM images representing the (a) neat PLA filament and (b) 90% PLA + 10% CGF filament.

Figure 7

Image (300×) of the composite filament section with CGF particles.

Adding agro-waste to 3D filaments produced from PLA appears to be an option, which is particularly of interest, whenever local biomass residue is available, as is the case for rubber vine, CGF. In this case, the addition of a limited amount of CGF powder to the PLA filament proved to be a success, in the sense that the crystallinity and tensile strength of the matrix were increased while maintaining a sufficiently viable MFI and not affecting thermal degradation. Thinking of an increase in the amount of filler introduced, which shows sufficient interfacial contact with the matrix, further verifications will be needed about the roughness of the filament. This will rule out the presence of differential stresses, which might lead to possible enhanced cracking, especially in view of clustering of the CGF particles. In the absence of roughness studies, microscopical investigations indicate the presence of some cracking as a result of filler presence, especially enhanced in the presence of air inclusions during polymer molding.

The authors acknowledge the funding from Researchers Supporting Project number (RSP2024R355), King Saud University, Riyadh, Saudi Arabia.

A. Udhayakumar, K. Mayandi, N. Rajini, and M. Murali – Developed the experimental work on novel fiber extraction and characterization studies. Faruq Mohammad, Hamad A. Al-Lohedan, and Kumar Krishnan – Optimized the process parameters for 3D printing. Carlo Santuli – Analysed the novel fiber characterization studies.

Authors state no conflict of interest.