This study investigates the application of multimaterial additive manufacturing to design auxetic metastructures with tunable mechanical properties and energy absorption (EA) capabilities. By integrating stiff and compliant polymers within a single architected lattice, the aim is to understand how localized variations in material distribution influence deformation behavior, structural adaptability and reconfigurability. The overarching goal is to demonstrate how multimaterial auxetic designs can be leveraged to create adaptive and multifunctional structures.
An auxetic lattice exhibiting a negative Poisson’s ratio was designed in multiple multimaterial configurations and fabricated using a dual-nozzle fused deposition modeling (FDM) process. Polylactide (PLA) and polycaprolactone (PCL) were selected as the hard and soft phases, respectively, providing mechanical contrast while enabling thermally induced shape reconfiguration through shape memory behavior. Compression testing was conducted to characterize the force–deformation behavior, densification response and EA of the material. The influence of material arrangement within each unit cell was evaluated by comparing circular, line and cross-distributions of PLA and PCL.
An auxetic lattice exhibiting a negative Poisson’s ratio was designed in multiple multimaterial configurations and fabricated using a dual-nozzle FDM process. PLA and PCL were selected as the hard and soft phases, respectively, providing mechanical contrast while enabling thermally induced shape reconfiguration through shape memory behavior. Compression testing was conducted to characterize the force–deformation behavior, densification response and EA of the material. The influence of material arrangement within each unit cell was evaluated by comparing circular, line and cross-distributions of PLA and PCL.
This work provides an experimentally validated demonstration of geometry–material coupling as a design strategy for creating adaptive, reconfigurable metastructures. By exploiting the contrasting properties of soft and hard materials within an auxetic architecture, the study highlights how multimaterial 4D printing can be used to achieve tailored mechanical performance and functional programmability. The presented approach offers broad potential for future applications in soft robotics, impact mitigation systems and thermally responsive devices.
1. Introduction
In recent decades, additive manufacturing (AM), commonly known as 3D printing, has emerged as a key technology for fabricating complex, customized and lightweight structures (Gibson et al., 2021). This capability represents a significant advantage over conventional manufacturing processes, which often struggle or are entirely unable to produce such intricate geometries. Moreover, AM enables the integration of functional design into structures that can perform beyond simple load-bearing functions, such as energy absorption (EA), adaptability and environmental responsiveness (Llanos et al., 2025; Zolfagharian et al., 2025).
A relatively recent advancement in AM is multimaterial 3D printing, where two or more distinct materials (e.g. hard and soft polymers) are integrated within a single printed part (Mirzaali et al., 2020; Ghimire and Chen, 2024). This approach allows the creation of multifunctional components with spatially varied mechanical or functional properties, eliminating the need for post-assembly and enabling innovative design solutions tailored to specific performance requirements.
Another important evolution in AM is 4D printing, which introduces a dynamic aspect to 3D-printed parts, allowing them to change shape, functionality or properties over time in response to external stimuli such as temperature, magnetic fields or light (Khoo et al., 2015, Mehrpouya and Huang, 2022; Ghalayaniesfahani et al., 2024a, b). Central to this technology are shape memory polymers (SMPs), which enable such transformations through their inherent shape memory behavior. Several studies have explored common SMPs such as polylactide (PLA), polycaprolactone (PCL) and thermoplastic polyurethane (TPU) for 4D printing applications (Mehrpouya et al., 2021; Yarali et al., 2024). For instance, Barletta et al. demonstrated that PLA exhibits excellent shape recovery characteristics, with printing parameters significantly influencing its mechanical and functional performance (Barletta et al., 2021). Similarly, Zhou et al. investigated 4D printing using PCL-based SMPs, reporting not only good mechanical strength and elasticity but also biocompatibility suitable for biomedical applications such as stents (Zhou et al., 2021).
In addition to single-material SMPs, blending different SMPs has been used to tailor mechanical and functional performance. For example, Rahmatabadi et al. studied 4D printing of PLA–TPU blends with varying PLA contents (50, 70 and 90 wt%), achieving tunable shape recovery between 90% and 96% depending on composition and thermal programming conditions (Rahmatabadi et al., 2023). Roudbarian et al. combined TPE and PLA in multimaterial structures using both layered and blended configurations, finding that multilayer lattice designs exhibited superior shape recovery (Roudbarian et al., 2021). Similarly, Ma et al. reported a PLA–PCL composition that achieved rapid shape recovery of 1.2 s and a recovery rate of 92%, further demonstrating the potential of polymer blending for performance optimization (Ma et al., 2021).
Multimaterial 4D printing of SMPs, where two or more materials are strategically distributed within a single structure, offers even greater potential for developing adaptive and reconfigurable designs. These structures can alter geometry or stiffness in response to stimuli, enabling programmable, spatially controlled behavior. Ge et al. demonstrated this concept using microstereolithography (PμSL) to fabricate 3D architected structures from multiple photo-curable SMPs, achieving time-dependent sequential shape recovery by tuning the dynamic properties of each material (Ge et al., 2016). Similarly, Zolfagharian et al. developed a multimaterial 4D printed composite combining an SMP and a flexible elastomer (Vero and Tango) via PolyJet printing; the resulting structures exhibited self-bending behavior strongly dependent on material ratios and thickness (Zolfagharian et al., 2023). Yuan et al. explored hydrogel–elastomer bilayer composites, where dehydration induced large volumetric shrinkage (up to 60%) and stiffness increase (from 100 kPa to 4 GPa), allowing initially flat patterns to transform into complex 3D configurations (Yuan et al., 2021). Their study revealed a relationship between bending curvature and structural stiffness, demonstrating potential applications in load-bearing and deployable 4D printed mechanisms.
Among the various structures fabricated from SMPs, auxetic structures, which exhibit a negative Poisson’s ratio, stand out for their exceptional EA, indentation resistance and flexibility. These properties make them highly suitable for applications requiring enhanced mechanical performance. Bodaghi et al. investigated a re-entrant auxetic structure fabricated from two SMPs distributed linearly, showing that stiffness and energy dissipation could be tuned through material arrangement (Bodaghi et al., 2020). Likewise, Mehrpouya et al. tailored the mechanical performance of multimaterial anti-chiral auxetic structures made from PLA and PCL, demonstrating that EA and strength could be precisely controlled by adjusting the material ratios within the structure (Mehrpouya et al., 2024).
Despite these advancements, limited research has focused on the multimaterial 4D printing of auxetic structures. In particular, there remains a lack of understanding of how material distribution influences deformation behavior and EA mechanisms. Therefore, this study aims to design and fabricate multimaterial 4D printed auxetic structures using PLA- and PCL-based shape memory biopolymers, with an emphasis on optimizing material distribution to achieve targeted mechanical and EA performance. The outcomes of this research could inform the design of next-generation adaptive structures for applications requiring controlled and tunable functional behavior.
2. Materials and methods
2.1 Materials
For multimaterial structures, it was necessary to identify a suitable combination of polymers that would ensure strong interfacial adhesion and mechanical compatibility. Therefore, PLA was selected as the “hard” material due to its rigidity, ease of printing and high tensile strength. Conversely, PCL was chosen as the “soft” material, as previous studies have reported promising adhesion between PLA and PCL (Mehrpouya et al., 2023, 2024).
Two polymers with distinct thermomechanical properties were therefore considered for material selection: PLA (3D4makers, The Netherlands) and PCL (Facilan™ PCL100, The Netherlands), both with a filament diameter of 2.85 ± 0.1 mm. The material properties of the printing filaments, as provided by the manufacturers, are summarized in Table 1.
Material properties of the printing filaments in this study
| Material | Density (g/cm3) | Tensile strength (MPa) | Young modulus (MPa) | Elongation at yield (−) | Melting temperature (°C) | Glass temperature (°C) |
|---|---|---|---|---|---|---|
| PLA | 1.24 | 66 | 3,027 | 0.07 | 145–165 | 57 |
| PCL | 1.1 | 45 | 350 | 0.15 | 58–60 | −60 |
| Material | Density (g/cm3) | Tensile strength (MPa) | Young modulus (MPa) | Elongation at yield (−) | Melting temperature (°C) | Glass temperature (°C) |
|---|---|---|---|---|---|---|
| PLA | 1.24 | 66 | 3,027 | 0.07 | 145–165 | 57 |
| PCL | 1.1 | 45 | 350 | 0.15 | 58–60 | −60 |
2.2 Printing process
The samples were fabricated using the Fused Deposition Modeling (FDM) technique with an Ultimaker S3 printer (Ultimaker B.V., The Netherlands). All structures were designed in SolidWorks (2020) (Dassault Systèmes SolidWorks Corporation, USA), and the resulting STL files were imported into Ultimaker Cura 4.11.0 (Ultimaker B.V., The Netherlands) to define the appropriate printing parameters.
Table 2 summarizes the printing parameters used in this study, which were selected based on the recommendations provided by the filament manufacturers. To improve adhesion between the first layer and the printing bed and to minimize warping, an adhesion spray and blue masking tape were applied to the build surface. Furthermore, all samples were printed in the same orientation to ensure consistency and reproducibility of the results.
2.3 Sample configuration
In this study, a re-entrant structure exhibiting a negative Poisson’s ratio was selected. The re-entrant geometry resembles a honeycomb in which two opposing sides are inclined inward rather than outward, resulting in an auxetic behavior that enhances EA. The re-entrant angle plays a key role in determining the structure’s mechanical response under compression, with auxetic behavior typically observed when the angle ranges between 25 and 75° (Masters and Evans, 1996; Mekala et al., 2019). Figure 1 illustrates the dimensions of the re-entrant element used in this study, including the unit cell geometry and the total sample size.
On the left, an hourglass-shaped unit has straight slanted sides that narrow at the center and widen at the top and bottom. A vertical dashed line marks the left edge, and a second vertical dashed line marks the central axis. A horizontal dimension at the bottom between these two dashed lines is labeled 4 millimeters. A horizontal reference line passes through the midpoint of the shape. At the lower left, an angle labeled thirty degrees is shown between the slanted edge and the vertical direction. On the right side of the unit, a vertical dimension labeled 5 millimeters marks the height of the inward notch at the center, indicated with double headed arrow. On the right, identical hourglass units repeat in aligned rows and columns to form a rectangular lattice panel. The pattern creates a continuous zigzag structure with narrow waists at the center of each unit. The total width across the top is labeled 42 millimeters, and the total height along the right side is labeled 36 millimeters. Thick horizontal boundary bars appear at the top and bottom edges..Geometry of the re-entrant element and sample dimensions
On the left, an hourglass-shaped unit has straight slanted sides that narrow at the center and widen at the top and bottom. A vertical dashed line marks the left edge, and a second vertical dashed line marks the central axis. A horizontal dimension at the bottom between these two dashed lines is labeled 4 millimeters. A horizontal reference line passes through the midpoint of the shape. At the lower left, an angle labeled thirty degrees is shown between the slanted edge and the vertical direction. On the right side of the unit, a vertical dimension labeled 5 millimeters marks the height of the inward notch at the center, indicated with double headed arrow. On the right, identical hourglass units repeat in aligned rows and columns to form a rectangular lattice panel. The pattern creates a continuous zigzag structure with narrow waists at the center of each unit. The total width across the top is labeled 42 millimeters, and the total height along the right side is labeled 36 millimeters. Thick horizontal boundary bars appear at the top and bottom edges..Geometry of the re-entrant element and sample dimensions
Various combinations of hard (black) and soft (white) materials were incorporated into the multi-material design configurations. In addition to the multi-material specimens, two single-material samples, PLA and PCL, were also fabricated for comparison purposes.
Each layout was designed with a gradual transition between the two materials throughout the structure. The specimens were designated according to their material distribution pattern and cell geometry, namely circular, cross-shaped and vertical line configurations, as illustrated in Figure 2(a–c).
The 3 D auxetic lattice structure consists of three panels. Each panel shows a thick rectangular block made of interlocking bowtie-shaped cells using hard black and soft white materials. Panel “(a)” displays a “Circular” arrangement where a dark center is surrounded by a ring of light cells and a dark outer edge. Panel “(b)” displays a “Cross” arrangement where the light and dark cells form a cross-like pattern through the grid. Panel “(c)” displays a “Lines” arrangement where the materials are organized in alternating vertical columns of light and dark cells. Each panel is labeled with its respective letter and description for the material distribution.Multimaterial configurations combining the hard material (PLA, shown in black) and the soft material (PCL, shown in white) in (a) circular, (b) cross and (c) lines arrangements
The 3 D auxetic lattice structure consists of three panels. Each panel shows a thick rectangular block made of interlocking bowtie-shaped cells using hard black and soft white materials. Panel “(a)” displays a “Circular” arrangement where a dark center is surrounded by a ring of light cells and a dark outer edge. Panel “(b)” displays a “Cross” arrangement where the light and dark cells form a cross-like pattern through the grid. Panel “(c)” displays a “Lines” arrangement where the materials are organized in alternating vertical columns of light and dark cells. Each panel is labeled with its respective letter and description for the material distribution.Multimaterial configurations combining the hard material (PLA, shown in black) and the soft material (PCL, shown in white) in (a) circular, (b) cross and (c) lines arrangements
2.4 Testing setup and equipment
Cyclic compression tests were performed to evaluate the mechanical properties of the printed structures, with particular attention to their energy dissipation capacity, in accordance with ISO 7743:2017 guidelines. The experiments were carried out using a ZwickiLine universal testing machine equipped with a 5-kN load cell (ZwickRoell, The Netherlands). The TestXpert3 software (ZwickRoell, The Netherlands) was used to record force–strain data and calculate the dissipated energy during compression. All tests were conducted at a compression speed of 4 mm/min up to a maximum strain of 50%, under ambient conditions. After reaching this limit, the force is unloaded. The initial pre-load was set at 0.5 N, and each test was repeated three times to ensure reproducibility of the results.
2.5 Shape recovery testing
A shape recovery test was conducted to evaluate the structural response under thermal stimulation. All multi-material samples, along with the single-material PCL specimens previously deformed during compression testing, were immersed in a temperature-controlled water bath (Julabo, CORIO CD-BT19, Germany) maintained at 53°C, a temperature below the melting point of PCL.
After thermal exposure, the samples were carefully removed from the water bath and allowed to recover under ambient conditions. The recovered height of each specimen was then measured. The experimental data were subsequently analyzed to determine the shape memory recovery ratio of the materials.
The shape recovery ratio, typically expressed as a percentage (Equation 1), is defined as the ratio between the recovered dimension and the original dimension. Here, the original dimension refers to the height prior to deformation, while the recovered dimension corresponds to the height after thermal stimulation. This parameter quantifies the ability of a shape memory material or structure to return to its original shape following deformation.
3. Results and discussion
All designed single-material and multi-material structures were successfully fabricated using a dual-nozzle FDM printer. The multi-material specimens were produced using a combination of PLA and PCL polymers, following the processing parameters described in the previous section. Figure 3a–c illustrates the 3D-printed circular, cross- and line-patterned multi-material samples.
The 3 D auxetic lattice structure consists of three panels. Each panel shows a rectangular block made of interlocking bowtie-shaped cells using hard black and soft white materials. Panel “(a)” displays a “Circular” arrangement where a dark center is surrounded by a white ring of cells and a dark outer edge. Panel “(b)” displays a “Cross” arrangement where the white cells form a vertical and horizontal cross shape through the dark grid. Panel “(c)” displays a “Lines” arrangement where the materials are organized in alternating vertical columns of white and black cells. Each panel is labeled with its respective letter and description for the specimen type.Schematic of 3D printed (a) circular, (b) cross- and (c) lines multimaterial specimens
The 3 D auxetic lattice structure consists of three panels. Each panel shows a rectangular block made of interlocking bowtie-shaped cells using hard black and soft white materials. Panel “(a)” displays a “Circular” arrangement where a dark center is surrounded by a white ring of cells and a dark outer edge. Panel “(b)” displays a “Cross” arrangement where the white cells form a vertical and horizontal cross shape through the dark grid. Panel “(c)” displays a “Lines” arrangement where the materials are organized in alternating vertical columns of white and black cells. Each panel is labeled with its respective letter and description for the specimen type.Schematic of 3D printed (a) circular, (b) cross- and (c) lines multimaterial specimens
Following fabrication, the specimens were subjected to cyclic compression testing. Each sample was compressed to a displacement corresponding to 50% of its original height (18 mm), representing the onset of the densification stage during deformation.
Figure 4 presents a comparison of the force–strain behavior of specimens composed of hard (PLA) and soft (PCL) materials arranged in three different infill patterns (lines, circular and cross), along with single-material PLA and PCL specimens.
The 3 D auxetic lattice structure is represented by a line graph and a sequence of three photos labeled “Before test”, “After compression”, and “After shape recovery”. The graph displays force-strain curves for five different structural types across a single cycle. The vertical axis is labeled “Force [Newton]” and ranges from 0 to 3000 in increments of 500 units. The horizontal axis is labeled “Strain [Percent]” and ranges from 0 to 50 in increments of 10 units. The graph includes five trend lines labeled in the legend as “P L A”, “Cross”, “Circular”, “Lines”, and “P C L”. Each line follows a general path of increasing force as strain increases to 50 percent, followed by a sharp unloading phase back to 0 Force. “P L A”: This line reaches a local peak of 1250 at 4 percentage strain, plateaus with minor fluctuations between 900 and 1100, then rises sharply to a maximum of 2900 at 50 percentage strain. “Cross”: This line rises to 500 at 5 percentage strain, maintains a steady plateau near 600 until 30 percentage strain, and then rises sharply to reach 2900 at 50 percentage strain. “Circular”: This line peaks at 550 at 5 percentage strain, gradually increases through fluctuations to 850 at 35 percentage strain, and reaches 2400 at 49 percentage strain. “Lines”: This line peaks at 950 at 5 percentage strain, dips to 500 by 20 percentage strain, stays near 500 until 40 percentage strain, and then rises to 1800 at 50 percentage strain. “P C L”: This line shows the lowest force values, peaking at 250 at 5 percentage strain, maintaining a plateau near 300 until 25 percentage strain, and reaching its maximum of 1200 at 50 percentage strain. Below the graph, three images connected by right-pointing arrows show the physical test. “Before test” shows a black and white bowtie-shaped lattice sitting uncompressed between two metal plates. “After compression” shows the same lattice flattened significantly under pressure. “After shape recovery” shows the lattice after the pressure is removed, having returned to nearly its original height and shape.The results of the compression test for single and multimaterial structures
The 3 D auxetic lattice structure is represented by a line graph and a sequence of three photos labeled “Before test”, “After compression”, and “After shape recovery”. The graph displays force-strain curves for five different structural types across a single cycle. The vertical axis is labeled “Force [Newton]” and ranges from 0 to 3000 in increments of 500 units. The horizontal axis is labeled “Strain [Percent]” and ranges from 0 to 50 in increments of 10 units. The graph includes five trend lines labeled in the legend as “P L A”, “Cross”, “Circular”, “Lines”, and “P C L”. Each line follows a general path of increasing force as strain increases to 50 percent, followed by a sharp unloading phase back to 0 Force. “P L A”: This line reaches a local peak of 1250 at 4 percentage strain, plateaus with minor fluctuations between 900 and 1100, then rises sharply to a maximum of 2900 at 50 percentage strain. “Cross”: This line rises to 500 at 5 percentage strain, maintains a steady plateau near 600 until 30 percentage strain, and then rises sharply to reach 2900 at 50 percentage strain. “Circular”: This line peaks at 550 at 5 percentage strain, gradually increases through fluctuations to 850 at 35 percentage strain, and reaches 2400 at 49 percentage strain. “Lines”: This line peaks at 950 at 5 percentage strain, dips to 500 by 20 percentage strain, stays near 500 until 40 percentage strain, and then rises to 1800 at 50 percentage strain. “P C L”: This line shows the lowest force values, peaking at 250 at 5 percentage strain, maintaining a plateau near 300 until 25 percentage strain, and reaching its maximum of 1200 at 50 percentage strain. Below the graph, three images connected by right-pointing arrows show the physical test. “Before test” shows a black and white bowtie-shaped lattice sitting uncompressed between two metal plates. “After compression” shows the same lattice flattened significantly under pressure. “After shape recovery” shows the lattice after the pressure is removed, having returned to nearly its original height and shape.The results of the compression test for single and multimaterial structures
All configurations exhibit an initial peak in force, followed by a plateau region occurring within the strain range of approximately 5–10%. Beyond this region, the force increases progressively as the structures undergo larger deformations, reaching a displacement of 18 mm, which corresponds to 50% of the original specimen height.
As shown in the figure, the single-material PLA structure generated the highest compressive force (2980 N), whereas the single-material PCL specimen exhibited the lowest value (1180 N). Among the multi-material configurations, the cross pattern demonstrated the highest compressive force, reaching approximately 2898 N, followed by the Circular pattern at about 2372 N. In contrast, the Lines pattern achieved only approximately 1940 N, representing the lowest compressive force among the three multi-material designs.
Figure 4 also illustrates the Cross specimen before and after the compression test, as well as its shape recovery behavior during thermal exposure in the water bath.
The ability of a structure to maintain higher forces after loading and unloading part of the cyclic compression test over the same range of deformation varies between patterns, and this behavior is directly reflected in the energy dissipation results. Dissipated energy can be interpreted as the area under the force–deformation curve in the loading-unloading diagram of the cyclic compression test; thus, patterns that sustain higher forces over the same deformation range (up to 50%) and then unload the force demonstrate greater dissipated energy capability. Equation 2 is used to calculate dissipated energy, where δd represents the maximum displacement reached during the compression test, and F denotes the corresponding compression force.
Figure 5a presents the dissipated energy values for all specimens obtained from the cyclic loading–unloading compression tests. Overall, the single-material PLA structures exhibit the highest dissipated energy (approximately 20,000 J). This behavior is attributed to the relatively high stiffness and strength of PLA, which enable it to withstand greater loads and absorb more energy prior to failure. In contrast, the single-material PCL specimens show the lowest dissipated energy (approximately 7,000 J), reflecting their lower stiffness and more flexible nature.
Panel (a): Vertical bar graph: The vertical axis for “Dissipated Energy [Joules]” ranges from 0 to 20000 in increments of 5000 units. The right vertical axis for “Maximum Force [Newton]” ranges from 0 to 3000 in increments of 500 units. The horizontal axis shows five categories: P L A, Cross, Circular, Lines, and P C L. The data values are as follows: “P L A”: “Dissipated Energy”: 20000, “Maximum Force”: 2950. “Cross”: “Dissipated Energy”: 14900, “Maximum Force”: 2950. “Circular”: “Dissipated Energy”: 15500, “Maximum Force”: 2300. “Lines”: “Dissipated Energy”: 10000, “Maximum Force”: 2000. “P C L”: “Dissipated Energy”: 7500, “Maximum Force”: 1250. Panel (b): Horizontal bar graph: The horizontal axis is labeled “Shape Recovery [Percent]” and ranges from 0 to 100 in increments of 25 units. The vertical axis shows four categories: P C L, Lines, Cross, and Circular. The data values are as follows: “P C L”: 91 percent. “Lines”: 91 percent. “Cross”: 83 percent. “Circular”: 86 percent. Note: All numerical data values are approximated.(a) Measured dissipated energy and maximum compressive force values for each specimen during cyclic compression testing; (b) Shape recovery ratios of all multi-material structures and the single-material PCL specimen
Panel (a): Vertical bar graph: The vertical axis for “Dissipated Energy [Joules]” ranges from 0 to 20000 in increments of 5000 units. The right vertical axis for “Maximum Force [Newton]” ranges from 0 to 3000 in increments of 500 units. The horizontal axis shows five categories: P L A, Cross, Circular, Lines, and P C L. The data values are as follows: “P L A”: “Dissipated Energy”: 20000, “Maximum Force”: 2950. “Cross”: “Dissipated Energy”: 14900, “Maximum Force”: 2950. “Circular”: “Dissipated Energy”: 15500, “Maximum Force”: 2300. “Lines”: “Dissipated Energy”: 10000, “Maximum Force”: 2000. “P C L”: “Dissipated Energy”: 7500, “Maximum Force”: 1250. Panel (b): Horizontal bar graph: The horizontal axis is labeled “Shape Recovery [Percent]” and ranges from 0 to 100 in increments of 25 units. The vertical axis shows four categories: P C L, Lines, Cross, and Circular. The data values are as follows: “P C L”: 91 percent. “Lines”: 91 percent. “Cross”: 83 percent. “Circular”: 86 percent. Note: All numerical data values are approximated.(a) Measured dissipated energy and maximum compressive force values for each specimen during cyclic compression testing; (b) Shape recovery ratios of all multi-material structures and the single-material PCL specimen
The multi-material structures demonstrate intermediate behavior, indicating that structural geometry plays a significant role in mechanical performance. Among these configurations, the Circular pattern exhibits higher dissipated energy compared to the Cross and Lines designs. Although the Cross structure achieves a relatively high maximum compressive force, comparable to that of PLA, it shows lower overall energy dissipation. The Lines configuration displays a noticeable reduction in both maximum force and EA capacity.
Notably, the trend in dissipated energy closely follows the trend in maximum compressive force, suggesting that higher load-bearing capacity during compression strongly influences the total energy dissipated during cyclic loading. However, the observed difference between the Cross and Circular configurations indicates that, beyond material stiffness, the architectural design and material distribution significantly affect both maximum force and energy dissipation. Overall, these results demonstrate that while material selection primarily governs mechanical performance, structural design can effectively tailor the balance among stiffness, strength, and energy dissipation. This finding highlights the importance of architectural optimization in enhancing mechanical functionality without altering the constituent materials.
Figure 5b presents the shape recovery ratios of all multi-material structures, along with the single-material PCL specimen. Both the single PCL sample and the Lines-pattern multi-material structure achieved the highest recovery ratio of 92%. The Cross configuration exhibited a recovery ratio of 86%, while the Circular design showed the lowest recovery ratio, approximately 83%.
4. Conclusion
This study investigated the mechanical performance, energy dissipation behavior, and shape recovery capability of multi-material auxetic structures fabricated from PLA (hard phase) and PCL (soft phase) using three distinct infill patterns: Circular, Lines, and Cross. The experimental results clearly demonstrate that both material distribution and geometric configuration play decisive roles in governing the overall structural response under cyclic compression.
The single-material PLA specimen exhibited the highest compressive force and energy dissipation, while the single-material PCL showed the lowest values, confirming the dominant influence of intrinsic material stiffness on load-bearing capacity. The multi-material structures displayed intermediate yet highly tunable behavior. Among them, the Cross configuration achieved the highest maximum compressive force, approaching that of pure PLA, whereas the Circular pattern exhibited the greatest energy dissipation among the multi-material designs. The Lines configuration showed comparatively lower force and EA, highlighting the sensitivity of mechanical performance to phase arrangement and load-transfer pathways within the architecture. The close correlation between maximum force and dissipated energy further indicates that load-bearing capacity strongly influences overall EA during cyclic deformation.
In addition to mechanical performance, shape recovery tests revealed that recovery behavior is strongly dependent on material composition and structural constraint. The single-material PCL and the Lines-pattern multi-material structure achieved the highest recovery ratios, while increased structural constraint from the rigid PLA phase reduced recovery in other configurations.
Overall, this study confirms that controlled integration of materials with contrasting mechanical properties enables programmable stiffness, strength, energy dissipation, and shape recovery within a single structure. These findings underscore the potential of multi-material auxetic architectures for applications requiring tailored impact mitigation, adaptive mechanical response, and functional gradient behavior. Furthermore, the results provide a foundation for future optimization of material distribution and geometric design to achieve enhanced multifunctional performance without altering the base materials.
