Strips 5 cm wide and 30 cm long were obtained from four-year-old bamboo culm stems of the species Guadua angustifolia Kunth. Subsequently, the strips were crushed by mechanical means in order to obtain short fibers (between 1 and 2 cm long). Ozone treatment was carried out in order to eliminate lignin, wax, oil, and some impurities. A Clearwater Tech LLC generator was used for the treatment. The ozone was added at a rate of 230 mg O₃/h. The fibers were treated for two hours. After treatment, the fibers were dried, pulverized, and sieved through a 45 µm sieve. The size of the filler was selected based on the characteristics of the extruder used (single-screw extruder). With the use of particles retained in the #325 sieve (45 μm), the dispersion capacity of the reinforcement within the PLA increases. On the other hand, larger particles can cause clogging in the extruder nozzle and affect the surface quality of the extruded filament.

For this study, an Ingeo biopolymer 4032D was used. Following the supplier’s suggestions, the material was dried for 4 h at 80°C to prevent loss of viscosity. After drying, the material was stored in a sealed package in order to avoid contact with atmospheric conditions.

1.75 mm filaments were obtained by applying the extrusion method. Pure PLA filaments (PLA-F) and filaments obtained from the combination of PLA and pulverized Guadua fibers (PLA-G-F) were constructed using a single-screw extruder operating at 50 mm/s. The extrusion temperature was 180°C. A commercial biodegradable plasticizer was used to improve the mixing and extrusion process. For the PLA-G-F filaments, a composition of 90% granulated PLA and 10% pulverized Guadua fibers was used.

The dimensions of the tensile specimens were established in accordance with the recommendations of ASTM D638-Type V [33]. For the compression tests, the samples were designed with a diameter of 25 mm and a length of 50 mm. To define the number of specimens, a three-factor complete general factorial design was used. For each factor, three levels and five replicates were defined. Based on the experimental design, the number of probes to be tested was defined. A total of 270 samples were printed (135 for tensile and 135 for compression tests). The models were prepared using Ultimarker Cura 5.5.0 software and were printed using a Creality Ender 5 S1 3D Printer. The printing parameters are presented in Table 1.

The surface roughness was determined following the procedure outlined in Section 2.2.5. A measuring platform was employed to adjust the position between the specimen and the instrument. To enhance the accuracy of the results, ten measurements were taken for each specimen. HV hardness was evaluated according to the procedure described in Section 2.2.6. For these tests, five specimens with a diameter of 15 mm and a thickness of 3 mm were designed for each of the combinations defined during the printing process. The absorption capacity after 24 h of immersion in water at room temperature was assessed in accordance with ASTM 570-22 [34]. Test specimens with a diameter of 50.8 mm and a thickness of 3.2 mm were prepared for this evaluation. The density of the specimens was determined following the procedures outlined in ASTM D792-20 [35], using square samples measuring 2 cm on each side.

The tribological properties were assessed by measuring the coefficient of friction (COF) and calculating the wear rate. For the tests, specimens were printed with a diameter of 15 mm and a thickness of 5 mm. The “pin-on-disk” technique was employed using a Nanovea tribometer. Testing was conducted at room temperature with a Cr6 steel ball 6 mm in diameter. The linear speed was set at 600 mm/min with an amplitude of 10 mm, and the test was carried out for 1000 cycles. A normal load of 5 N was applied, resulting in a total sliding distance of 20 m. The wear rate was calculated using Archard’s law [36].

For the mechanical characterization, tensile and axial compression tests were performed. The tensile tests of the 3D-printed biocomposites were conducted in accordance with ASTM D638 specifications [33]. The compression tests were carried out following the recommendations of ASTM D695 [37]. The mechanical tests were carried out on the WP 300 universal testing machine, which has a 20 kN load capacity. For the tensile tests, the strain rate was 1 mm/min. For the compression tests, the strain rate was 1.3 mm/min. The test setup is shown in Fig. 2. The calculated properties include the tensile and compressive strength, the elastic modulus under tensile and compressive loads, and the maximum strain percentages.

The SEM micrographs for PLA-F and PLA-G-F are shown in Fig. 3a,b. Comparing the micrographs, some differences can be seen in the surface texture of the filaments. In Fig. 3b, the presence of microvoids and the randomly distributed Guadua fibers is evident. The microvoids could be associated with the non-homogeneous distribution of pulverized Guadua fibers and PLA granules during the extrusion process. On the other hand, some irregularities and the presence of some protuberances can be observed in PLA-G-F filaments (see Fig. 3b). Similar results were presented by Karis et al. [38], and these are aspects that must be considered, since they could cause clogging of the nozzle due to the agglomeration of fibers and the irregular diameter of the filament.

EDS was used to analyze the elemental composition of the filaments. Fig. 4 presents the spectra. From the figure, it can be observed that more than 98.5% of the elemental composition of the fibers corresponds to carbon and oxygen elements, while the small concentrations of Ba, Si, Fe, and Ti may be associated with impurities and/or contamination during the extrusion process. A small percentage of Ca (0.27%) in PLA-G-F filaments could be related to the lignocellulosic nature of the vegetal fiber. The results presented in Table 2 show the elemental composition of filaments, as well as the atomic C/O ratio. An increase of 14.5% can be observed in the C/O ratio of the PLA-G-F. Similar results were reported by Mondragón-Herrera et al. [39] and can be attributed to the addition of vegetable fiber as filament filler.

XRD spectra of the PLA-F and PLA-G-F filaments were analyzed. The deconvolution and the identification of crystallographic planes are shown in Fig. 5a,b. The results suggest the semi-crystalline nature of the filaments analyzed. The diffractograms of PLA-F and PLA-G-F display characteristic crystalline peaks corresponding to the planes (020) at 13.54°, (010) at 16.22°, (021) at 19.87°, (121) at 25.2°, (040) at 27.14°, and (002) at 28.97°. By contrast, the PLA-G-F diffractograms show distinctive peaks for the planes (020) at 13.75°, (110) at 16.58°, (111) at 22.13°, and (040) at 27.47°. The observed crystalline peaks may be attributable to the α structure of PLA [40].

The crystallinity index (C.I) was calculated using Eq. (1) (see Table 3). The results show an increase of 13.50% in the crystallinity index of the PLA-G-F. According to Liao et al., because of the semi-crystalline nature of PLA it is important to evaluate its crystallinity index and how this value influences the processing of materials obtained through additive manufacturing processes [41]. Although increasing the crystallinity index can improve the mechanical performance and strength against thermal degradation of the material, this increase can affect the adhesion between its successive layers during the printing process. Additionally, depending on the diameter of the nozzle, it may be necessary to increase its temperature in order to avoid clogging during the deposition of the filament on the printing bed [42]. An increase in temperature facilitates the flow of the filament, promoting greater adhesion between successive printing layers, thus counteracting the effect of the increase in the crystallinity index on the adhesion between the printing layers.

The ATR-FTIR technique was employed to identify the principal absorption bands corresponding to the functional groups present in the filaments, as shown in Fig. 6. The results indicate that both PLA-F and PLA-G-F exhibit the same functional groups. The bands observed between 2995 and 2851 cm⁻¹ are associated with the symmetric and asymmetric stretching of C–H in CH₃. The band near 1745 cm⁻¹ suggests carbonyl group stretching. The bands between 1453 and 1360 cm⁻¹ are related to the symmetric and asymmetric bending of the CH₃ bond, while those between 1266 and 1081 cm⁻¹ correspond to the ether group (C-O-C). Additionally, the bands from 1043 to 956 cm⁻¹ correspond to the rocking of the C–CH₃ bond, and the bands near 867 and 754 cm⁻¹ represent the stretching of the O-CH-CH₃ group and the wagging of CH₃, respectively [40–43]. On analyzing the ATR-FTIR spectra, variations in the intensity of the bands between 2995 and 2851 cm⁻¹ can be observed. Similar results were found by Macedo et al. [44] and may be related to changes in the structure of CH₃. According to Kljun et al., the lateral order index (LOI) can be used to interpret qualitative changes in crystallinity [45]. This index is based on establishing the relationship between the absorbance bands located at 1430 and 893 cm⁻¹. Generally, as the LOI increases, the crystallinity also increases. The LOI was determined for the filaments studied, and values of 1.09 and 1.15 were found for the PLA-F and PLA-G-F, respectively. An increase of 5% can be seen in this value. These results are consistent with the increase in the crystallinity index value of the PLAF-G presented in Table 3.

A combined DSC-TGA thermal analysis was performed to evaluate the thermal stability of the PLA-F and PLA-G-F filaments. The results are shown in Figs. 7 and 8. The findings indicated that adding pulverized Guadua fibers did not significantly affect the thermal stability of the filaments. The first thermal transition observed was the glass transition temperature (T_g), which occurred between 60°C and 65°C for PLA-F and close to 72°C for PLA-G-F. On the other hand, an exothermic peak near 100°C in the PLA-G-F filaments can be associated with the cold crystallization of PLA (T_c), indicating the crystallization of the polymer with increasing temperature. The main endothermic peak can be observed near 175°C and corresponds to the melting temperature T_m. Differences between the values obtained for PLA-F and PLA-G-F may be related to changes in the crystallinity of the filaments because of the incorporation of plant fibers. Finally, at temperatures above 350°C, an exothermic peak can be seen, which is related to the weight loss detected in the TGA curves, confirming the thermal degradation of the material. Similar results were reported by Cuifo et al. and show typical behavior for semicrystalline thermoplastic materials [46].

The stages of the thermal degradation of the filaments can be observed in Figs. 7 and 8. At a temperature close to 150°C, a slight decrease in mass (less than 2%) in both filaments can be related to the loss of initial humidity of the samples and the elimination of volatile compounds present in the PLA-G-F filament. In Fig. 7, the PLA-F degradation process was divided into two stages. The first, located between 315.63°C and 353.86°C, can be associated with the main degradation of the polymer. At this stage, a mass loss of 96.51% and a maximum degradation rate of 0.9998%/min can be seen. Additionally, an endothermic peak at 335.12°C could be related to the thermal decomposition of PLA. A second degradation stage can be seen between 376.50°C and 497.32°C, suggesting the presence of carbonaceous residues in the sample. Finally, the endothermic peak at 597.53°C could be related to the carbonization of these residues.

Fig. 8 shows the thermal degradation stages of the PLA-G-F filaments. Similar results were presented by Khan et al. [47]. In this case, it is possible to identify three stages in the degradation process. The first stage begins at a temperature of 308.43°C and corresponds to the initial degradation of hemicellulosic compounds. The second stage can be observed between 351.44°C and 371.61°C. This is the fundamental phase of the mass loss of the filament and could be associated with the thermal degradation of the cellulose. In the third stage, the degradation of PLA begins. This occurs at temperatures above 350°C.

Table 4 presents the TGA results for the PLA-F and PLA-G-F specimens. It can be seen that the addition of Guadua filler results in a reduction of less than 3% in the values of T_oi, T_i, T_max, and T_b temperatures. However, a 12% reduction in the value of the end of combustion temperature can be observed in the PLA-G-F. According to Daver et al., this can be attributed to the decomposition of low molecular weight substances present in the plant fibers used as filler. These substances can decompose at lower temperatures, thus influencing the overall thermal behavior of the biocomposite [48].

Table 5 presents the physical and mechanical properties of the PLA-F and PLA-G-F filaments. The results indicate that the incorporation of vegetal filler did not significantly reduce the density of the filaments, with differences of less than 3%. Additionally, an increase in the roughness and hardness of the PLA-G-F was observed, with values rising by 15% and 20%, respectively. According to Kariz et al., the non-homogeneous distribution of the filler during the extrusion process can lead to irregularities and protuberances that contribute to increased roughness of the filaments [38]. On the other hand, the vegetable filler acts as a reinforcement for the thermoplastic resin, enhancing both the hardness and the tensile strength of the biofilament.

On analyzing the mechanical properties, an increase of up to 10% in the strength and stiffness of PLA-G-F was seen. Similar findings were reported by Lau et al. and can be attributed to the amount of fiber used as filler (10%) [49]. Despite the increase in the strength and stiffness values of the filament under tension, a reduction of approximately 22% in the maximum strain percentage demonstrates the loss of ductility of the filaments with the addition of the filler material. The nonhomogeneous dispersion of the reinforcement and the presence of voids in the filament can increase the brittleness of the filament, affecting their quality and the mechanical performance of the manufactured products [50].

The effect of the printing parameters on the hardness and roughness of the specimens printed with PA-G-F was analyzed. The results are shown in Fig. 9a,b.

Fig. 9a shows the effect of the printing parameters (infill density, build orientation, and layer thickness) on the hardness of the specimens printed with PLA-G-F. By increasing the infill density from 25% to 100%, an increase of between 10% and 40% in the hardness of the specimens can be observed. According to Maguluri et al., this increase is due to the fact that a higher infill density provides a more solid and homogeneous internal structure, which contributes to improving the strength against penetration [51]. Analysis of the influence of printing layer thickness reveals a reduction of up to 40% in the hardness of the specimens when the layer thickness is increased from 0.20 to 0.30 mm. The increase in layer thickness affects the cohesion between successive layers due to the greater volume of material deposited with each layer [52]. Furthermore, specimens printed at 0° exhibited the highest hardness values, ranging from 18.83 to 30.89 HV, while those printed at a 90° displayed the lowest hardness values, between 16.21 and 22.56 HV.

On observing Fig. 9b, it can be verified that increasing the infill density from 25% to 100% reduces the roughness by up to 30%. On the other hand, on modifying the build orientation from 0° to 90° an increase in roughness of 30% to 50% can be seen. Finally, increasing the layer thickness to 0.3 mm leads to a significant rise in the roughness of the specimens, exceeding 50%. Specimens printed at 0° with 100% infill density and a layer thickness of 0.2 mm exhibited the lowest roughness values (4.88 +/− 0.21 µm), while those printed with 25% infill density at 90° and a layer thickness of 0.3 mm exhibited the highest values (22.75 +/− 2.05 µm).

The effect of the printing parameters on the density and the absorption capacity is illustrated in Fig. 10a,b. Fig. 10a shows a significant increase in density as the infill density rises from 25% to 100%. Specimens with 25% infill density exhibited values between 420 and 730 kg/m³, while those with 100% infill density reached values between 970 and 1225 kg/m³. Additionally, a 20% to 40% reduction in density occurs when the printing direction changes from 0° to 90°. There is also a slight decrease (less than 10%) in density when increasing the layer thickness from 0.20 to 0.30 mm.

On analyzing Fig. 10b, a reduction of up to 38% in the absorption rate is evident in specimens printed with 100% infill density. Specifically, specimens with 100% infill density, a layer thickness of 0.2 mm, and a 0° orientation exhibited the lowest absorption values (4.17%), whereas those printed with 25% infill density, a layer thickness of 0.30 mm, and a 90° orientation exhibited the highest absorption values (6.87%). These results indicate that on increasing the layer thickness and the build orientation, a greater number of micropores appear on the surface of the specimens, resulting in higher water absorption.

According to Dawoud et al., abrasive wear appears to be the main wear mechanism of polymers that slide against hard surfaces [53]. In the present study, a preliminary investigation of the effect of printing parameters (infill density, layer thickness, and built orientation) on the friction coefficient and the abrasive wear rate of biocomposites printed using FDM was carried out. Fig. 11a,b shows the coefficient of friction (COF) as a function of the sliding distance for specimens printed at 0°. In Fig. 11a, a reduction of up to 10% in the COF value on increasing the infill density from 20% to 100% can be seen. From the results, it is evident that for the same layer thickness (0.20 mm), specimens printed with the highest infill density (100%) exhibited the lowest COF values (0.63), while those printed with the lowest infill density (20%) exhibited higher COF values (0.72). Similar results were reported by Akıncıoğlu et al. when they analyzed the tribological characteristics of ABS structures with different infill densities [54]. As can be seen, the use of a higher infill density results in a smoother and less porous surface, which contributes to reducing the coefficient of friction. On the other hand, a slight increase (less than 5%) in COF values can be observed when the layer thickness is increased from 0.20 to 0.30 mm. Although increasing the thickness of the printing layer did not significantly affect the COF values, it should be taken into consideration that when thicker layers of material are deposited, they can increase the surface roughness, contributing to an increase in the COF.

The effect of the print orientation on the COF values of specimens printed with a specific infill density is shown in Fig. 11b. The results reveal an increase of up to 12% in COF values when the print direction is changed from 0° to 90°. As can be seen, specimens printed along the horizontal axis (0°) had COF values between 0.63 and 0.65, while those printed in the vertical direction exhibited higher values, ranging from 0.67 to 0.73.

The wear rate was calculated using Archard’s law, where the volume of material removed in the crater can be computed using the applied load, the tribometer’s sphere radius, and the length of interaction between the ball and the surface. The results are presented in Fig. 12. The specimens printed at 0° with 100% infill density and a layer thickness of 0.20 mm exhibited the lowest wear rate (4.26 × 10⁻³ mm³/N·m), while those printed at 90° with 20% infill density and a layer thickness of 0.30 mm had the highest wear rate (5.91 × 10⁻³ mm³/N·m). Increasing the infill density resulted in a reduction of up to 17% in the wear rate of all the analyzed samples. Additionally, with an increase in the print orientation and the layer thickness, the wear rate varied by a factor of between 5% and 10%.

The influence of the printing parameters (infill density, printing orientation, and layer thickness) on the strength and stiffness of specimens subjected to axial loads (tensile and compression) was analyzed. The results are presented in this section.