To evaluate the flow behavior of the materials during printing, dynamic rheological measurements were carried out on the neat resins (PLA and PBS), their PLA/PBS blend, and the corresponding composite systems containing 3% and 5% by weight of hemp shive. In particular, the rheological characterization was carried out at 190 °C, as this is the typical printing temperature for PLA, which serves as the benchmark in this study. Additionally, TGA analysis (Figure 1) revealed that significant thermal degradation of hemp shives begins above 200 °C.

Figure 4a,b compares the plots of complex viscosity (η*), storage modulus (G′), and loss modulus (G″) for the unfilled systems: PLA, PBS, and the PLA/PBS blend.

The complex viscosity curve of PLA exhibits a relatively broad Newtonian plateau, followed by a moderately shear-thinning region. In contrast, PBS shows a pronounced decrease in η* with increasing angular frequency (ω) across the entire frequency range analyzed. However, at frequencies below 1 rad/s, PBS displays higher complex viscosity values than PLA. The PLA/PBS blend exhibits lower viscosity values than both neat polymers up to 10 rad/s, while at higher frequencies, its viscosity falls between those of PLA and PBS (Figure 4a). As shown in Figure 4b, the storage modulus (G′) of PLA/PBS blend closely follows that of PBS, with neat PBS exhibiting the highest G′ values across the entire frequency range. Moreover, both PBS and the blend display a more pronounced elastic behavior than PLA in the low-frequency region.

These results provide useful insights for a preliminary assessment of the printability of PLA, PBS, and their blend via FDM technology. Specifically, a pronounced shear-thinning behavior is desirable to ensure good processability during both the extrusion and deposition stages. In fact, low viscosity inside the nozzle reduces melt pressure, allowing for lower processing temperatures, while moderately high viscosity at the nozzle exit prevents issues such as material dripping or poor layer definition [32,33,34,35,36,45]. On the other hand, during the deposition phase, the interlayer cohesion—which directly influences the final mechanical properties of the printed object—is primarily governed by the diffusion [46] and relaxation of polymer chains across adjacent layers. In particular, enhanced interfacial diffusion is favored in systems exhibiting predominantly viscous responses at the deposition temperature [32,33,34,35,36].

Thus, PBS is expected to exhibit favorable flow characteristics for extrusion through the printer nozzle, along with effective shape retention upon exiting the die. This behavior is mainly attributed to its marked shear-thinning properties. Nevertheless, its higher elastic component, compared to PLA, may introduce some drawbacks, such as an increased tendency for die-swell at the nozzle exit (as highlighted in the following) and reduced interlayer adhesion during deposition. However, some studies in the literature [12,45] evidenced that the main issue, affecting PBS printability, does not stem from its rheological behavior but from its rapid crystallization kinetics. In fact, if crystallization occurs too rapidly, it can lead to pronounced warping and shrinkage phenomena, resulting, in some cases, in the detachment of the printed material from the build platform, ultimately causing the failure of the printing process.

On the contrary, PLA is characterized by a more dominant viscous behavior, which is expected to promote improved interlayer diffusion and, consequently, enhanced cohesion in the printed structure. With regard to the PLA/PBS blend, the equal proportion of the two aliphatic polyesters results in a rheological response that combines the advantages of both components—exhibiting a more pronounced shear-thinning behavior than neat PLA and a more viscous character than neat PBS—potentially offering a favorable compromise for FDM processing.

Figure 5a–c presents the plots of complex viscosity, storage modulus, and loss modulus for PLA-based composites.

The complex viscosity curves (Figure 5a) for neat PLA and its composites exhibit similar overall trends. In particular, the filled systems display a Newtonian plateau at intermediate frequencies and a moderately shear-thinning behavior at higher angular frequencies, differing from the neat matrix essentially by the presence of a slight yield stress-like behavior at low ω. Moreover, across the entire frequency range analyzed, neat PLA exhibits higher complex viscosity than both the 3% and 5% hemp-filled composites. As supported by the thermogravimetric data in Figure 1, the filler undergoes partial degradation at the processing temperatures used during melt compounding with the polymer matrix (Section 2.1). In particular, according with the findings reported by Coppola et al. [47], the reduction in viscosity observed for the composites can be likely attributed to the thermal degradation of pectin, which therefore acts as a natural plasticizer inside the materials. Additionally, the lower viscosity of the composites compared to neat PLA can be partly attributed to the alignment of the anisotropic filler under flow.

The same behavior is observed for the storage modulus (Figure 5b), with the filled samples generally showing lower G′ values compared to the neat matrix, except in the low-frequency range. In this region, the composites also show a weaker dependence of G′ on frequency, suggesting a “pseudo solid-like” behavior. This feature, together with the observed yield stress, can be attributed to the development of a filler-induced network at low frequencies. Such structuring effect contributes to enhanced shape stability of the extrudate at the nozzle exit—where shear rates rapidly decrease—and during the subsequent deposition and solidification stages [32].

Figure 6a–c shows the viscoelastic properties of the PBS-based systems: neat PBS, PBS with 3 and 5 wt% of hemp shive.

As observed for PLA-based systems, the addition of hemp filler within the PBS matrix results in a reduction in the viscoelastic parameters compared to the neat polymer, with the most significant variations occurring at low frequencies. Moreover, the plots of the composites essentially overlap across the entire frequency range examined, indicating comparable rheological behavior despite the different filler contents. The absence of a storage modulus plateau at low frequencies in the PBS/hemp shive composites can be reasonably attributed to weaker interfacial interactions between PBS and the filler, likely due to PBS’s lower polarity compared to PLA.

Figure 7a–c compares the complex viscosity and the storage and loss moduli of PLA/PBS blend and its corresponding hemp-filled composites.

Comparably to PLA- and PBS-based composites, the presence of hemp shive inside the PLA/PBS blend leads to a slight but consistent increase in the rheological parameters compared to the neat matrix. This behavior can reasonably be attributed to the compatibilizing effect of the filler in the blend, likely promoted by esterification reactions between the hydroxyl groups of hemp shive and the carboxylic acid groups of polymer phases, as confirmed by FTIR analysis (Figure 3). Due to the improved interaction, the PLA/PBS-blend-based composites also exhibit a slight reduction in the slope of the storage modulus curves, at low frequencies, compared to the neat matrix.

Although the results of dynamic rheological tests have provided valuable insights into the printability of the systems under investigation, a more comprehensive understanding of the material’s flowability through the printer nozzle can be achieved by determining their viscosity values at shear rates that realistically occur within the nozzle during printing. To this purpose, an approximate estimation of the shear rates, corresponding to the selected printing parameters, was conducted following a procedure reported in the literature [11] and described in detail in the Appendix A (Figure A1).

Since the validity of the Cox–Merz rule has been demonstrated in the literature [32] for systems similar to the ones investigated in this study, dynamic rheological data were fitted using both the traditional Carreau model (Equation (1)) and a modified version (Equation (2)), the latter accounting for the occurrence of yield stress in the composite systems. where ƞ₀ represents the zero-shear viscosity, λ is the characteristic relaxation time, n is the power-law index describing the slope of the viscosity curve in the shear-thinning region, and σ₀ is the yield stress. In the Appendix A, Figure A2 shows the experimental η* values fitted using the Carreau models (solid lines) for PLA-, PBS-, and PLA/PBS-based systems, both neat and containing 3 wt% of hemp shives. The corresponding model parameters for all tested materials are reported in Table 5.

Analysis of the tabulated data reveals that the relaxation times (λ) consistently decrease upon the addition of the filler, regardless of the polymer matrix. On average, the composites exhibit relaxation times nearly halved compared to those of the corresponding neat polymers. The power-law index (n) decreases in neat systems containing PBS, reflecting its inherently more pronounced shear-thinning behavior compared to PLA. In contrast, the incorporation of hemp slightly increases n values in both PLA- and PBS-based matrices, whereas an opposite trend is observed for the filled blends. Yield stress was evaluated exclusively for PLA-based composites, whose viscosity profiles were fitted using the modified Carreau model. Notably, yield stress remained essentially unchanged with increasing filler content.

These models were therefore employed to extrapolate the viscosity values of the different systems at the shear rates estimated to occur within the nozzle during extrusion (Figure 8). The obtained results show that PBS-based materials exhibit the lowest viscosity values within the printer nozzle—in agreement with their more pronounced shear-thinning behavior—compared to PLA and PLA/PBS systems. Moreover, at the tested printing speeds, the viscosity of PBS-based systems does not change significantly with increasing filler content, whereas incorporating 3 wt% of hemp shive into PLA leads to a marked reduction in viscosity. In contrast, the filler addition inside PLA/PBS matrix results in a slight increase in viscosity.

In summary, all the analyzed systems exhibit viscosity values within the range of 100 to 1000 Pa·s, which has been previously identified in the literature as favorable for ensuring adequate flowability of polymer melts through the nozzle during FDM printing [12,34]. However, with the awareness that the absolute reliability of these extrapolated viscosities is limited, in future work, tests by means of capillary rheometer will be conducted to directly measure viscosity values in the shear rate range relevant to the FDM process.

Upon exiting the printer nozzle, the elastic recovery of stretched polymer chains may induce die-swell, which can affect the geometry of the printed objects. This phenomenon should be carefully controlled by adjusting the material’s formulation and/or the processing conditions. The goal is to achieve a certain degree of overlap between printed layers, resulting in a more compact structure—which is beneficial for the final mechanical performance—while still preserving the intended dimensions of the printed sample.

An approximative evaluation of die-swell behavior was carried out by extruding filaments of neat PLA, PBS, and their 50/50 blend, both unfilled and reinforced with 3 wt% hemp shives. The extrudates were collected directly at the nozzle exit and analyzed via optical microscopy (Figure 9). The corresponding die-swell (DS) values was calculated as the ratio between the diameter of the extruded filament (D_ext) and the nozzle diameter (D_nozzle).

Among the neat materials, PLA exhibits relatively low die-swell (DS = 1.1), while PBS and the PLA/PBS blend show a more pronounced expansion (DS = 1.5 and DS = 1.4, respectively). Conversely, the addition of hemp shive results in a reduction in die-swell for all the investigated systems, particularly pronounced for the PLA/PBS blend, with the value decreasing to DS = 0.8.

As well established in the literature [48], the extent of the die-swell phenomenon is strongly influenced by the molecular characteristics and/or morphology of the material, which affect its elastic response and, in turn, its characteristic relaxation times. In particular, PBS is characterized by longer relaxation times than PLA (Table 5), thus extending the overall relaxation behavior of PLA/PBS blend. Moreover, since the blend is immiscible (Figure 2), the interfacial elasticity, arising from the phases’ boundaries, further amplifies the die-swell effect [49].

The reduction in die-swell, due to the addition of hemp shive, can be attributed to two main mechanisms. Firstly, the filler exhibits a plasticizing effect (Figure 5 and Figure 6), resulting in reduced relaxation times during melt flow within the nozzle. Secondly, at the nozzle exit—where shear rates significantly decrease—the fillers’ microstructural network imposes a physical constraint on the polymer chains, thereby limiting their elastic recoil and further suppressing swelling behavior. Furthermore, hemp shives improve polymer phases’ compatibility within the PLA/PBS blend, as evidenced by the FTIR spectra in Figure 3. This results in a more homogeneous system and, from a rheological perspective, contributes to a further reduction in the swelling phenomenon for the filled blend [49].