Optimization of Deposition Temperature and Gyroid Infill to Improve Flexural Performance of PLA and PLA–Flax Fiber Composite Sandwich Structures

Abstract

This research investigates the optimization of 3D-printed sandwich structures fabricated using fused filament fabrication (FFF) with polylactic acid (PLA) and PLA reinforced with flax fibers. The core of the sandwich structure features a gyroid infill pattern, which is known for its mechanical efficiency. The study delves into the effects of deposition temperature on the adhesion between the core and skin layers, as well as the impact of infill density on the overall mechanical properties. Three-point bending tests are conducted to assess the flexural performance of the structures. The objective is to identify the optimal processing parameters to enhance the performance of PLA-based composite sandwich structures. Potential applications for these structures include lightweight components for automotive interiors, sustainable packaging solutions, and architectural elements requiring a balance of strength and environmental sustainability.

1. Introduction

Additive manufacturing (AM), also referred to as 3D printing, has rapidly gained recognition in recent years as a groundbreaking technology poised to improve and optimize several industrial production contexts [1]. Among its diverse applications, a notable area of impact in the research field is in the development of lightweight sandwich structures characterized by specific performance attributes [2]. These structures represent a composite design concept where two thin outer layers are spaced apart by a lightweight core material, resulting in a composite construction that outperforms solid components of equivalent weight across a spectrum of mechanical criteria [3].

Furthermore, the versatility of additive manufacturing technology extends beyond conventional manufacturing methods by enabling the complex fabrication and customization of these sandwich structures to suit specific applications across sectors such as aerospace [4], automotive [5], and even civil architecture [6,7]. Traditional manufacturing methods for sandwich structures often involve complex and time-consuming processes, such as mold injection or vacuum bagging [8]. 3D printing, on the other hand, offers a more flexible and efficient approach, allowing for the creation of multifaceted geometries and customized designs [9]. By directly depositing material layer by layer, 3D printing eliminates the need for tooling and assembly, reducing production costs and lead times [10]. This allows for the optimization of material distribution and structural efficiency, ultimately leading to the development of more sustainable, resource-efficient products that meet the evolving demands of modern industries [9,11,12].

In recent years, there has been a growing interest in sustainable and eco-friendly materials for 3D printing [13,14]. Polylactic acid (PLA), a biodegradable thermoplastic derived from renewable resources such as corn starch and sugarcane, has emerged as a popular choice for AM applications [15]. PLA offers several advantages, including low toxicity, good mechanical properties, and ease of processing. However, the mechanical performance of PLA can be limited, particularly under high-stress conditions.

To enhance the mechanical properties of PLA-based structures while preserving an environmentally sustainable design, researchers have explored the use of reinforcement materials, such as natural fibers [16]. Flax fibers are a promising option due to their high strength-to-weight ratio, low density, and excellent thermal and acoustic insulation properties. By incorporating flax fibers into PLA, it is possible to create composite materials with superior mechanical performance and improved sustainability [16,17,18].

A comprehensive review by Ilyas et al. [19] discussed the processing and applications of PLA biocomposites reinforced with natural fibers, including flax. The study highlighted that incorporating natural fibers into PLA could improve mechanical properties, making these biocomposites suitable for various applications.

Aliotta et al. [20] investigated PLA/short flax fiber composites with varying fiber content (10–40 wt.%) produced via extrusion. Their findings indicated that the addition of flax fibers improved the tensile strength and stiffness of the composites, with optimal performance being observed at 20 wt.% fiber content. This enhancement was attributed to the effective stress transfer between the matrix and fibers, despite the absence of surface treatments to improve interfacial adhesion.

Paulo et al. [21] developed PLA composites reinforced with flax fibers using a 3D printing process to evaluate improvements in tensile and flexural strength. Their experimental design varied parameters such as extruder temperature, number of fiber strands, infill percentage, and the application of surface chemical treatments to the fibers. The study found that while the NaOH surface treatment of fibers did not significantly influence the mechanical properties, the infill density had a substantial impact on the mechanical strength. The maximum tensile and bending stresses achieved were 50 MPa and 73 MPa, respectively.

Priya Muthe et al. [22] explored the production of high-performance 3D printing composite filaments by pre-impregnating bleached flax yarns with PLA. They compared solution (solvent-based) and emulsion (water-based) impregnation techniques to enhance fiber wetting and distribution within the PLA matrix. The study concluded that solution impregnation resulted in the highest tensile strength for PLA/bleached flax filaments, reaching 356 MPa, due to the improved fiber wetting and distribution.

These studies collectively demonstrate that integrating flax fibers into PLA matrices can significantly improve mechanical properties such as tensile strength and stiffness. However, factors like fiber content, alignment, and environmental exposure critically influence the performance and longevity of the resulting biocomposites.

Furthermore, the core of a sandwich structure plays a crucial role in determining the overall mechanical properties of the composite structure. The infill pattern, which defines the arrangement of material within the core, significantly impacts the structure’s stiffness, strength, and weight. Various infill patterns have been investigated in 3D printing, including honeycomb, grid, and gyroid [7,23].

The gyroid infill pattern offers a high degree of structural integrity, excellent mechanical performance, and low weight. By optimizing the gyroid infill pattern, it is possible to tailor the properties of sandwich structures to specific applications.

A study by Adam Khan et al. [24] examined the tensile and impact strengths of gyroid structures fabricated using additive manufacturing. The research compared gyroid structures to conventional solid models, revealing that while solid structures exhibited superior ultimate and yield strengths, gyroid structures demonstrated better elongation and impact strength, highlighting their potential for lightweight applications.

Hao et al. [25] investigated the mechanical properties and energy absorption capabilities of gyroid sandwich structures with different gradient designs. Their study revealed that introducing density gradients, particularly those based on trigonometric functions, can significantly enhance the elastic modulus and energy absorption of gyroid structures, making them suitable for applications requiring high energy absorption.

This study aims to optimize the manufacturing parameters of 3D-printed sandwich structures using fused filament fabrication (FFF) with PLA and PLA reinforced with flax fibers. By investigating the influence of deposition temperature (200 °C and 220 °C) on the interfacial adhesion between the core and skins and the impact of infill density (20% and 30% infill density) on the overall mechanical properties, this research seeks to identify optimal processing conditions for enhanced performance. The flexural performance of the fabricated sandwich structures will be evaluated through three-point bending tests.

The focus on the gyroid infill pattern, renowned for its high strength-to-weight ratio, aligns with current trends in lightweight design and material optimization. By examining the adhesion between the core and skin layers, the study addresses a critical aspect of sandwich structural integrity that has often been overlooked in previous research. Moreover, the utilization of PLA and PLA–flax fiber composites will be applied to assess a commitment to sustainable and environmentally friendly materials, expanding the application potential of 3D printed structures in areas such as lightweight automotive components, eco-friendly packaging, and sustainable architectural elements. The findings of this research are expected to provide valuable insights into the design and fabrication of high-performance 3D-printed sandwich structures with improved mechanical properties and enhanced sustainability.

2. Materials and Methods

2.1. Sandwich Structure Design

In this work, novel 3D-printed sandwich panels with gyroid core geometries were produced by changing the extrusion temperature (i.e., 200 °C and 220 °C) for subsequent mechanical characterization through three-point bending tests, using polylactic acid (PLA) provided by Filoalfa (Ciceri de Model Srl, Ozzero, Italy) and polylactic acid reinforced with flax fibers (PLA–flax; type Starflax 3d) provided by Nanovia (Nanovia, Louargat, France). These core patterns were printed at two different infill densities: 20 and 30%. Infill density refers to the percentage of filled material within the core, with 0% being completely hollow and 100% being fully solid. The infill density values of 20% and 30% were selected based on prior studies [26], which demonstrated these levels as optimal for balancing lightweight design and mechanical performance. These values aligned with the study’s focus on developing sustainable and lightweight structures for potential industrial applications.

In producing specimens, additive manufacturing based on “design-driven manufacturing” was used. The initial parametric modeling was performed using Autodesk Inventor 2022, while nTop (formerly nTopology) was employed to generate and refine the gyroid geometry. This advanced software allowed precise control over lattice structures, enabling the creation of gyroid cores with predefined wall thicknesses based on the selected infill density (20% or 30%). Specimens were created based on D790 standard [27] dimensions. Then, they were weighed, and the results are summarized in

Table 1. Properties of samples.

Material	T [°C]	Infill [%]	Dimension [mm × mm × mm]	Weight [g]
PLA	200	20	150 × 40 × 12	31.5 ± 0.06
		30		37.5 ± 0.14
	220	20		30.1 ± 0.04
		30		35.9 ± 0.10
PLA–flax	200	20		28.1 ± 0.04
		30		35.5 ± 0.02
	220	20		28.4 ± 0.04
		30		35.9 ± 0.01

.

Then, the CAD file was exported in STL format. This file represented a solid with triangle-based surfaces. Each triangle in the STL file included X, Y, and Z coordinates for its vertices, along with a surface normal vector. The STL file was then imported into slicing software, which converted digital 3D models into instructions for 3D printers, enabling layer-by-layer physical object creation. The slicing process involved dividing the 3D model into layers and determining printing paths with customizable settings like infill density and print speed. This study utilized Bambu Studio, enabling modification of cell configuration and infill settings.

The TPMS Gyroid structure is a geometric model with exceptional properties used in 3D printing and engineering. It is a mathematically defined triply periodic minimal surface featuring a continuous, intersection-free configuration that repeats in 3D space. The surface has zero mean curvature, resulting in equal tensile forces across all points [28]. Its design maximizes the strength-to-weight ratio by efficiently distributing material [29,30]. Components benefit from their optimized internal structures, reducing required material while maintaining strength [28]. Similar to other triply periodic minimal surfaces, the gyroid can be approximated trigonometrically with a simple equation:

$$\mathrm{s}\mathrm{i}\mathrm{n} \left(x\right)·\mathrm{c}\mathrm{o}\mathrm{s} \left(y\right)+\mathrm{s}\mathrm{i}\mathrm{n} \left(y\right)·\mathrm{c}\mathrm{o}\mathrm{s} \left(z\right)+\mathrm{s}\mathrm{i}\mathrm{n} \left(z\right)·\mathrm{c}\mathrm{o}\mathrm{s} \left(x\right)=0$$

(1)

The infill density (20% or 30%) determines the wall thickness of the gyroid structure when generated as a solid CAD model. During slicing, the software was set to use 100% infill, meaning that the internal regions of the gyroid walls were completely filled with material. This ensured that the printed model accurately replicated the predefined geometry, as the wall thickness was directly controlled by the infill density chosen during the design stage (20% or 30%). By using 100% infill for the walls, the structural integrity and mechanical properties of the final components were preserved, faithfully reflecting the intended design parameters.

Figure 1 presents a visual representation of the TPMS gyroid structure. Subfigure (a) depicts the structure with a 20% infill density, while subfigure (b) showcases the structure with a 30% infill density.

After slicing the model, it was transferred to the 3D printer for production using additive manufacturing. The test specimen was positioned with the smallest dimension aligned along the Z-axis and horizontal layers at 90°. No post-production treatments (e.g., chemical, thermal, or finishing processes) were applied to the produced samples.

2.2. Additive Manufacturing of Test Specimens

In this study, Fused Deposition Modeling (FDM) was employed using Bambulab X1-Carbon (Bambu Lab, Shenzhen, China) for PLA samples and Creality K1 Max (Creality, Shenzhen, China) for PLA–flax samples. Both printers were used in their default configurations, with no modifications to the extruder or nozzle.

Table 2. FDM parameters.

Parameter	PLA	PLA–flax
3D printer	Bambulab X1-Carbon	Creality K1 Max
Extrusion temperature	200 °C|220 °C
Bed temperature	60 °C
Layer height	0.2 mm
Default line width	0.42 mm
First layer line width	0.5 mm
Inner wall line width	0.45 mm
First layer speed	50 mm/s
First layer fill speed	105 mm/s
Outer/inner wall speed	200/300 mm/s
Fill speed	270 mm/s
Resolution	0.012 mm
Infill patters	Gyroid (100% wall infill)

provides detailed printing parameters, ensuring full replicability of the process.

The 3D printer, equipped with a Titan extruder and Volcano nozzle, could reach a maximum extrusion temperature of 240 °C. For this study, extrusion temperatures of 200 °C and 220 °C were compared. Furthermore, a bed temperature of 60 °C was used to ensure optimal printing conditions. The selection of deposition temperatures (200 °C and 220 °C) was based on the manufacturer’s recommendations for the materials used. The extrusion temperature range for PLA was 170–210 °C, while for PLA–flax it was 200–230 °C. To ensure compatibility and proper adhesion during the manufacturing process, the minimum temperature was set to 200 °C, aligning with the lower limit for PLA–flax, while the maximum temperature of 220 °C was chosen as an intermediate value between the upper limits of the two materials. This choice was further supported by thermogravimetric analysis (TGA) of the materials (see Figure 2), which confirmed their thermal stability up to 300 °C, validating the selected extrusion temperatures.

Polylactic acid (PLA) is a widely used biodegradable polymer derived from renewable resources like corn starch and sugarcane. Known for its ease of use in 3D printing, PLA offers good mechanical properties, including a tensile strength of 53 MPa and a tensile modulus of 3.6 GPa. It has a density of 1.24 g/cm³ and a glass transition temperature of approximately 55–60 °C. PLA is FDA and ROHS compliant, making it suitable for non-critical applications such as prototypes, educational models, and decorative items. PLA reinforced with flax fibers combines the advantages of PLA with enhanced mechanical and sustainability properties. This biocomposite material features a tensile modulus of 3.4 GPa and exhibits a wood-like appearance when printed. The incorporation of flax fibers increases the rigidity and impact resistance of the material while maintaining compatibility with standard 3D printers. It is particularly suitable for lightweight, rigid parts and design-oriented applications. The addition of flax fibers also improves the environmental footprint of the composite.

In Figure 3, we can observe the novel 3D printed TPMS gyroid sandwich structures featuring a 20% infill density and 200 °C deposition temperature. These structures were created using two different materials—pure PLA, as depicted in Figure 3a, and PLA reinforced with flax fibers, as shown in Figure 3b—highlighting the versatility and potential for enhancing material properties through reinforcement techniques. The quality observed, including the excellent surface finish and dimensional accuracy, was consistent with results reported in recent studies on parameter optimization [31,32].

All samples were assigned a unique code. This code consisted of a prefix: “P” for PLA polymer-based sandwiches and “PF” for PLA–flax composite-based sandwiches. Following the prefix, two numbers separated by a hyphen were included. The first number represented the extrusion temperature used during the additive manufacturing process, while the second number indicated the infill density of the core. For example, the code “PF-220-30” would denote a PLA–flax composite sandwich with a core infill density of 30% manufactured using an extrusion temperature of 220 °C. Analogously, P-200-20 code would reference a PLA based sandwich characterized by 20% infill density and produced by using an extrusion temperature of 200 °C.

2.3. Three-Point Bending Tests

The flexural tests followed ASTM D790 standards [27] to optimize flexural strength while considering practical infill densities and cell geometries. This standardized approach enabled comparison with existing research and real-world applications.

Mechanical characterization of the samples was performed using a ZwichRoell testing machine (Zwick Roell, Ulm, Germany) equipped with a 2.5 kN load cell, adhering to the ASTM D790 standard [27]. To conduct bending tests, 12 gyroid-based sandwiches were fabricated by FFF. Three specimens were produced for each of the two specified filling densities (20%, 30%) and extrusion temperatures (200 °C, 220 °C). Bending tests were carried out at a strain rate of 0.1 mm/min. A Hirox Digital Microscope KH 8700 (Hirox, Tokyo, Japan) was employed to analyze the samples and assess failure modes.

3. Results and Discussion

To begin our investigation and assess the mechanical performance of the sandwich structures under bending stress, Figure 4 shows the observed trends in flexural strength (plotted on the primary axis) and flexural modulus (plotted on the secondary axis) as a function of varying strain levels. These data were obtained for a specific reference P-220-20 sandwich structure to establish a baseline for comparison with other configurations.

Three distinct regions can be identified within the material’s response to deformation.

Stage I: Firstly, at low levels of deformation, a clear linear relationship exists between applied deformation and the resulting stress. This indicates an elastic behavior of the sandwich structure, where the material returns to its original shape upon stress removal. The modulus of elasticity, representing the stiffness of the structure, remains relatively constant throughout this stage, reaching a maximum value of 1023 MPa. Based on this consideration, this initial stage can be identified as a linear elastic regime.

Stage II: Secondly, as deformation increases progressively, a noticeable deviation from this linear stress-strain relationship emerges. In the Figure 4, a threshold of 1.5% deformation has been identified as the point where this deviation becomes apparent and where stage II starts. Subsequently, the rate of stress increase diminishes, and concurrently, the flexural modulus that represents the tangent to the stress–strain curve exhibits a progressive decline. During this stage, the sandwich structure undergoes increasing deflection, leading to a concentration of stress within the layers and amplifying local deformation at the interfaces. This ultimately culminates in the tensile fracture of the lower skin of the sandwich, marking the point of maximum load capacity (point 1 in Figure 5). At this critical point, the flexural modulus effectively reduces to zero, indicating a relevant loss of structural integrity.

Stage III: During Stage III, following the initiation of a crack in the lower skin, the applied load exhibits a consistent and continuous decline. This downward trend directly correlates with a modulus value that consistently remains below zero. This type of damage progression is characterized by a gradual and non-sudden deterioration of stress levels, indicating a progressive rather than catastrophic failure mode. Notably, the observed fluctuations in the applied load, closely mirroring the local valleys and peaks in the modulus trend, serve as indicators of distinct damage evolution phases that synergistically contribute to the diminishing performance of the sandwich structure.

In particular, as the deformation continues to increase progressively, the crack in the lower skin (point 1 in Figure 5) undergoes further extension and widening along the skin layers growing from the external side toward the neutral axis along the load direction. Concurrently, delamination phenomena are observed in close proximity to the crack at the interface between the skin and the gyroid core (point 2 in Figure 5). This delamination arises from the weakening of the adhesive bonds between the complex gyroid geometry and the external skin. However, a significant observation is the absence of any buckling phenomena of the walls of the gyroid core (point 3 in Figure 5). This significant finding strongly suggests that the core structure possesses the inherent capacity to function effectively as a stress transfer element, facilitating the efficient distribution of loads between the upper and lower skins of the sandwich structure.

In order to comprehensively assess and understand the impacts of different variables, specifically the material type (whether mono-material or composite reinforced with natural fibers) and the variations in infill and core percentages, on the mechanical behaviors of sandwich structures, Figure 6 provides a detailed overview of the stress–strain relationship observed during a standard reference test conducted across all batches. For a more focused analysis, Figure 6a specifically delves into the performance of PLA-based composites (identified as batches P), highlighting their unique characteristics and responses under different conditions. On the other hand, Figure 6b captures the performance data for composite sandwiches (batches PF), allowing for a comparative examination of the mechanical properties exhibited by these specific structures.

In particular, when evaluating Figure 6a regarding PLA sandwiches, it becomes evident that the infill density does not lead to a significant change in the flexural strength, which remains nearly constant. However, it does have a notable impact on the damage modification, as evidenced by the larger strain observed at failure for batches with high-density infill values (P-xx-30 batches). Furthermore, it is worth noting that the utilization of a high extrusion temperature (P-220-xx batches) results in a slight decrease in flexural strength, regardless of the chosen infill density. Additionally, a subtle shift in the curve towards higher strain levels can be observed when a higher extrusion temperature is employed. This observation suggests that opting for a higher extrusion temperature triggers the occurrence of the initial failure and subsequent damage propagation in the sandwich at greater strain levels. This indicates that the relationship between infill density and extrusion temperature plays a significant role in determining both the flexural strength and the strain at which failure and damage propagation occur.

Analogously, in the composite sandwich (Figure 6b), both infill density and extrusion temperatures exhibit effects that closely mirror those seen in PLA-based sandwiches. However, in this scenario, these effects are heightened and more pronounced, emphasizing the impacts of these factors on the overall structure and characteristics of the sandwich.

At the same time, the composite sandwich exhibits lower maximum flexural strength values compared to pure PLA, specifically around 15–19 MPa for the former and 23–25 Mpa for the latter. In parallel, once the maximum load is reached, the performance degradation (stage III of the stress–strain curve) is significantly different. The PLA system shows a progressive damage evolution. Conversely, the composite sandwich undergoes a sudden and abrupt load drop associated with the catastrophic failure of the sandwich. This is attributable to the different failure modes observed. As previously shown in Figure 5, in the PLA-based sandwich, damage evolution occurs through the progressive advancement of a tensile crack in the lower skin, associated with delamination phenomena at the interface between the skin and the core in the tensile zone of the sandwich. Instead, during testing, the composite sandwich structure experiences a unique failure scenario where a single catastrophic fracture occurs in the sandwich.

The observed failure mode In the composite sandwiches Involves a critical flaw: the simultaneous rupture of the lower skin (point 1 in Figure 7) occurs independently of any debonding between the skin and the core (point 2 in Figure 7).

This rupture is specifically manifested within the region of highest stress concentration. Contrary to expectations, the primary failure mechanism does not involve a separation between the skin layers and the core. Instead, a distinct collapse of the gyroid core’s internal wall structure is clearly evident (point 3 in Figure 7). This collapse is initiated by the formation of a significant crack in close proximity to the lower skin’s crack. The presence of a high stress concentration at this point acts as a catalyst, rapidly driving crack propagation through the core wall along the direction of applied load.

The rupture pattern in Figure 7 highlights the strong interlayer bonding, likely achieved through optimized printing parameters such as extrusion temperature, print speed, and layer height. These findings align with studies that emphasize the critical role of parameter selection in achieving superior interlayer adhesion and mechanical performance [31,32].

This critical observation strongly suggests a limitation of the PLA–flax filament material system within the context of 3D printing that needs to be managed. Specifically, the observed failure behavior indicates that the material is less effective at establishing high interlayer adhesion between successive filament layers during the 3D printing process. This inherent weakness in interlayer bonding ultimately results in the significantly reduced overall structural integrity of the sandwich structure, ultimately leading to premature and catastrophic fracture under load.

It should be noted that the choice of a higher extrusion temperature does not result in material degradation or lead to a beneficial joining improvement (as evidenced by the constant flexural strength observed). Indeed, this adjustment does not yield any noticeable mechanical improvements in the composite’s performance, as the strength remains almost constant when comparing PF-200 and PF-220 batches.

To more effectively discern and quantify the subtle variations in the mechanical performance exhibited by sandwich structures when 3D printed under diverse manufacturing conditions, Figure 8 compares the flexural strength (a) and the strain at maximum stress (b) by varying the infills and the extrusion temperatures for the monolithic (P batches) and composite (PF batches) sandwiches.

The flexural strength of the composite sandwich incorporating flax fibers consistently exhibits lower values compared to its pure polylactic acid (PLA) counterpart. This disparity becomes more pronounced at lower infill densities, with a notable 35% strength reduction observed at a 20% infill density and a 22% reduction at a 30% infill density. Concurrently, the deformation at maximum stress in the composite sandwich is slightly elevated by approximately 0.3% compared to the pure PLA structure.

This observed behavior can be primarily attributed to the superior interlayer adhesion achieved within the pure PLA filament during additive manufacturing. The introduction of flax fibers into the composite material necessarily diminishes the volume fraction of the polymer matrix. This reduction in the binding polymer matrix material directly impacts the achievable interlayer adhesion within the composite sandwich. Furthermore, the presence of flax fiber reinforcement influences the stress transfer at crucial junction zones between the skin and core layers of the sandwich structure. This disruption in stress transfer at the interfaces between the constituent elements of the sandwich contributes to the observed lower strength values and increased deformability within the eco-sustainable composite sandwich.

Compared to similar studies [21,33], the lower mechanical properties observed in this study can be attributed to several factors. Firstly, the flax fibers in our work were not subjected to chemical treatments or pre-impregnation techniques, which are known to improve fiber–matrix bonding. Secondly, the chosen additive manufacturing process (FFF) inherently limits interlayer adhesion compared to other techniques like injection molding or filament extrusion, leading to reduced mechanical performance. Lastly, the infill densities of 20% and 30% were deliberately selected to prioritize weight reduction, which inherently compromises ultimate strength.

These factors underscore the challenges of utilizing natural fibers in additive manufacturing and highlight areas for future improvement. For instance, chemical treatments or pre-impregnation of flax fibers can enhance the mechanical performance of PLA–flax composites. Moreover, optimizing processing parameters and exploring alternative fabrication techniques might help achieve higher interlayer adhesion and improved structural integrity.

To investigate whether the cell configuration significantly impacts the flexural strengths of the samples, an analysis of variance (ANOVA) was conducted using MINITAB software 22.1.0. This analysis considered three factors with two levels each:

• Material: PLA and PLA–flax.
• Extrusion Temperature: 200 °C and 220 °C.
• Infill: 20% and 30%.

Before performing the ANOVA, some assumptions were checked to ensure the validity of the results: (i) normality of residuals; (ii) homoscedasticity (equal variance); and (iii) independence of observations. A residual analysis was undertaken to assess the accuracy of the underlying assumptions.

Figure 9 shows that the errors are evenly spread around zero, suggesting the data follow a normal distribution. The consistent scatter of errors across predicted values confirms that the variance is constant (homoscedasticity). No significant patterns in the errors can be identified, indicating that the data points are independent. These findings support the reliability of the ANOVA analysis and confirm that the observed significant effects are meaningful within the scope of this study.

Table 3. Analysis of variance for flexural strength.

Source	DF	Adj SS	Adj MS	F-Value	p-Value
Model	9	307.420	34.158	141.19	0.000
Blocks	2	0.318	0.159	0.66	0.534
Linear	3	290.816	96.939	400.70	0.000
Material	1	262.497	262.497	1085.04	0.000
Temperature	1	9.950	9.950	41.13	0.000
Infill	1	18.369	18.369	75.93	0.000
2-Way Interactions	3	16.086	5.362	22.16	0.000
Material*Temperature	1	4.386	4.386	18.13	0.001
Material*Infill	1	11.651	11.651	48.16	0.000
Temperature*Infill	1	0.050	0.050	0.21	0.656
3-Way Interactions	1	0.199	0.199	0.82	0.380
Material*Temperature*Infill	1	0.199	0.199	0.82	0.380
Error	14	3.387	0.242
Total	23	310.806
S 0.491857|R-sq 98.91%|R-sq(adj) 98.21%|R-sq(pred) 96.80%

presents the key findings from the statistical analysis (ANOVA), which aimed to identify if the studied factors had a significant impact on the load. The table includes the degrees of freedom (DF), a crucial value used to compute the mean square (MS). Essentially, these statistical measures assess the amount of independent information used to calculate the sum of squares (SS), which represents the overall variability in the data. This variability is further broken down into components: the variability within each factor, the variability due to the interactions between factors, and the residual variability not explained by the factors. The mean square (MS) is calculated by dividing the sum of squares by the degrees of freedom. The mean squared error estimates the remaining variance in the data after considering the effects of the factors and their interactions. The F-statistic is used to determine the significance of each term in the model. A p-value less than or equal to a predetermined threshold (e.g., 0.05) indicates that the effect of that particular term is statistically significant.

The ANOVA results highlight the significant impacts of the main factors (material, temperature, and infill) on the response variable, with the material showing the most pronounced effect (F = 1085.04, p < 0.001), followed by the infill (F = 75.97, p < 0.001) and temperature (F = 41.13, p < 0.001). Among the two-way interactions, material*temperature (F = 16.13, p < 0.001) and material*infill (F = 48.16, p < 0.001) are statistically significant, indicating that the combined effects of these factors play a crucial role in determining the response. However, the interaction between temperature and infill is not significant (F = 0.21, p = 0.656), suggesting that these two factors do not meaningfully influence the response when combined. Similarly, the three-way interaction (material*temperature*infill) is not significant (F = 0.82, p = 0.380), indicating that the combined influences of all three factors do not substantially contribute to the variability in the response.

Overall, the model explains 98.91% of the variability in the response variable (R-squared), with a strong adjusted R-squared of 98.21% and a predicted R-squared of 96.80%, confirming the model’s robustness and predictive capability. These results suggest that the main effects and the significant two-way interactions should be prioritized when optimizing the response variable, as they are the primary drivers of variation.

The effects of the interactions are also evidenced in Figure 10.

The interaction plots provide key insights into how the factors material, temperature, and infill interact to influence the response variable. The material*temperature interaction reveals a marked difference between PLA and PLA–flax. While PLA exhibits a decrease in response as the temperature increases, PLA–flax maintains a relatively stable performance across the tested temperatures. This indicates that PLA is more sensitive to temperature changes, whereas PLA–flax demonstrates a robustness that makes it less influenced by this parameter.

The material*infill interaction further highlights the distinct behaviors of the two materials. PLA shows a significant decrease in response as the infill level increases, suggesting that its performance is more dependent on infill configuration. In contrast, PLA–flax is less affected by changes in infill, exhibiting a more consistent performance across the levels. This suggests that PLA–flax might be more suitable for applications where infill variability is expected or necessary.

Conversely, the temperature*infill interaction is negligible, as indicated by the near-parallel lines in the corresponding plot. This implies that the combined effect of temperature and infill on the response variable is minimal, particularly for PLA–flax, which maintains stable performance regardless of these factors.

The significant interactions observed for material*temperature and material*infill highlight the importance of material selection when optimizing for temperature and infill. PLA demonstrates greater sensitivity to these parameters, while PLA–flax offers more stability, making it a promising candidate for applications requiring consistent performance under varying conditions. The minimal interaction of temperature*infill further simplifies the optimization process, focusing attention on the more impactful material-driven interactions.

In summary, this research investigated the mechanical behaviors of 3D-printed sandwich structures with gyroid cores fabricated using PLA and PLA–flax composites, revealing the significant influences of material type, extrusion temperature, and infill density on flexural strength and failure modes, with PLA–flax demonstrating potential for consistent performance across varying processing conditions despite the lower overall strength.

The amplified effects noted in the composite sandwich highlight the need for the precise control and understanding of these variables to achieve desired outcomes in terms of strength, stiffness, and structural integrity. By delving deeper into the relationship between infill density, extrusion temperature, and the resulting characteristics of the composite sandwich, researchers can optimize the manufacturing process and tailor the performance of the sustainable material to specific applications. Therefore, thorough exploration and analysis of these factors are crucial for maximizing the potential benefits offered by composite sandwiches, ultimately leading to advancements in material science and engineering practices.

In this regard, a future step will be taken to explore 3D printing parameters and strategies able to enhance interlayer adhesion and optimize fiber distribution within PLA–flax composites, leading to improved mechanical properties and expanded applications in sustainable lightweighting and structural engineering.

This can open a valid applicative scenario of 3D-printed sandwich structures, incorporating bio-based composite materials like PLA–flax, considering the increasing interest in sustainable and lightweight structures for industrial applications. In fact, by optimizing processing parameters, addressing the limitations of interlayer adhesion, and further exploring the influences of fiber orientation and distribution, it may be possible to better exalt the engineering potential of these innovative materials. This will lead to the development of high-performance, eco-friendly components for diverse sectors, including automotive, aerospace, and construction, where light weight and sustainability are paramount.

4. Conclusions

This study investigated the mechanical behaviors of 3D-printed sandwich structures with gyroid core geometries fabricated using PLA and PLA–flax composites. Three-point bending tests were conducted to evaluate the influences of extrusion temperature (200 °C and 220 °C) and infill density (20% and 30%) on the flexural strength and stiffness.

The results demonstrate that the material type significantly impacts the mechanical performance. PLA exhibits higher flexural strength (23–25 MPa) compared to PLA–flax (15–19 MPa), attributed to superior interlayer adhesion within the pure PLA filament.

Analysis of variance (ANOVA) revealed the significant main effects of material, temperature, and infill on flexural strength. Notably, the material and temperature and the material and infill interactions are also highly significant, indicating that the material choice significantly influences the impacts of temperature and infill on the final performance.

The enhancement provided by flax fibers in this study is limited due to the absence of surface treatments and the inherent challenges of achieving strong interlayer adhesion in 3D-printed composites. These findings underscore the need for the further exploration of fiber–matrix interactions, surface modification techniques, and alternative fabrication strategies to fully leverage the potential of flax fibers in additive manufacturing. While the mechanical performance of PLA–flax composites is lower than that of pure PLA, their more consistent behavior under varying processing conditions highlights their potential for applications requiring reliability and sustainability.

Furthermore, the failure modes are differed between the two materials. PLA-based sandwiches exhibit mainly a progressive failure characterized by tensile cracking in the lower skin and delamination at the core–skin interface. In contrast, PLA–flax composites experience catastrophic failure due to the combination of lower skin rupture and the sudden collapse of the gyroid core’s internal wall structure, suggesting limitations in interlayer adhesion within the composite material. While higher extrusion temperatures do not significantly improve the mechanical properties of either material, they influence the strain at maximum stress, suggesting a shift in the failure mechanism.

Overall, this research provides valuable insights into the influences of processing parameters and material selection on the mechanical behaviors of 3D-printed sandwich structures with gyroid cores. The findings can guide the optimization of manufacturing processes and material choices for achieving the desired mechanical properties and enhancing the sustainability of 3D-printed components for various applications.