Printability Optimization of LDPE-Based Composites for Tool Production in Crewed Space Missions: From Numerical Simulation to Additive Manufacturing

Abstract

Fused filament fabrication (FFF) is a 3D printing technology that has been successfully demonstrated aboard the International Space Station (ISS), proving its suitability for space applications. In this study, we aimed to apply FFF to the 3D printing of recycled space beverage packaging, made of LDPE and a PET-Aluminum-LDPE (PAL) trilaminate. To minimize material waste and optimize the experimental process, we first conducted numerical simulations of additive manufacturing. Using Digimat-AM 2021.1 software, we analyzed residual stresses and warpage in an LDPE/PAL composite with a 10 wt% filler content, processed through the FFF technique. Three key printing parameters, including printing speed and infill pattern, were varied across different levels to assess their impact. Once the optimal combination of parameters for minimizing residual stresses and warpage was identified, we proceeded with the experimental phase, printing objects of increasing complexity to validate the correlation between numerical predictions and the 3D-printed models. The successful fabrication of all geometries under optimized conditions confirmed the numerical predictions, particularly the reduction in warpage and residual stress, validating the material’s viability for additive manufacturing. These findings support the potential application of the LDPE/PAL composite for in situ resource utilization strategies in long-term space missions.

1. Introduction

The establishment of a permanent human presence beyond low Earth orbit (LEO) is becoming an increasingly ambitious goal, with space agencies and commercial partners focusing on long-duration missions to the Moon [1,2,3], Mars [4], and beyond [5].

A fundamental challenge in enabling such missions lies in making human exploration both affordable and sustainable over time, as shown by studies proposing multi-objective optimization models for mission planning [6], exploring hydrogen-based economies for Martian self-sufficiency [7], addressing radiation protection requirements [8], and assessing the long-term environmental impacts of future space activities through Life Cycle Assessment [9]. A key strategy in addressing this challenge is the development of technologies that enable in situ resource utilization (ISRU), minimizing reliance on materials transported from Earth, as reflected in NASA’s roadmap for sustainable exploration [10], recent advancements in ISRU technologies for life support and construction [11], studies on Martian CO₂ capture methods [12], and prospects for lunar-based solar power generation [13]. Among these resources, waste products generated by the crew are gaining attention as a potential asset, rather than merely a challenge to be managed [14,15]. Traditional waste management solutions, such as returning waste to Earth or incinerating it in orbit, become progressively impractical as missions extend in duration and distance. For instance, venting trash into space through small airlocks has been evaluated, but this method raises concerns due to the release of volatiles in a vacuum, which can recondense and contaminate sensitive spacecraft surfaces or systems. Furthermore, low-velocity disposal of solid waste in certain orbital trajectories may lead to uncontrolled debris or the risk of recontact with the spacecraft [16]. To overcome these limitations, researchers are investigating more sustainable waste management techniques, including in-suit urine filtration systems for water recovery [17], the Trash Compaction and Processing System (TCPS) for volume reduction [18], and the Multi-Bag Compaction System (MBCS) for efficient solid waste handling [19]. Among them, thermochemical conversion processes, such as pyrolysis, offer the ability to transform waste into useful gases. These gases can serve practical functions, including orbital station-keeping, or be further refined into propellants like oxygen and methane, which could enable lightweight scientific missions, such as small lunar landers, without requiring additional launch mass from Earth. These approaches not only improve waste handling, but also contribute directly to mission efficiency and operational autonomy [20]. Further innovation is found in the integration of Controlled Ecological Life Support Systems (CELSSs), which are designed to recover and recycle as much material as possible within a closed-loop environment. In a 180-day integrated experiment with a four-member crew, nearly 90% of renewable solid waste, including food packaging, was successfully recycled. This high recycling efficiency underscores the viability of more self-sufficient life support architectures for future exploration habitats [21].

Building on these concepts, one particularly promising direction is the repurposing of solid waste as feedstock for in-space manufacturing and construction. Many components of the mission waste stream, such as polymers and metals, are compatible with additive manufacturing (AM) technologies. This creates an opportunity to transform discarded materials into structural or functional elements needed for mission operations.

Among these materials, astronaut food packaging has emerged as a significant contributor to the waste stream, both due to its frequent usage and its substantial aluminum content. Specifically, this multilayer packaging is composed of approximately 29.5 wt% of aluminum alloy 1235, which could result in the recovery of over 75 kg of pure aluminum annually for a typical four-person crew on a long-duration mission. This aluminum-rich waste stream presents a promising opportunity for resource reutilization, as it can be repurposed as feedstock for metal-based additive manufacturing processes, enabling the in situ production of mission-critical components [22]. While this approach is conceptually attractive, its practical implementation in space remains at an early stage. A significant milestone was reached in 2024 with the deployment of the first metal-based 3D printer to the International Space Station (ISS), developed through a collaboration between the ESA, Airbus, and Space SAS. The system employs Directed Energy Deposition (DED) technology and successfully printed its first test specimen under sustained microgravity [23]. However, metal AM in space still faces key challenges, including high energy demands, complex thermal management, and ensuring reliable operation in microgravity.

Fateri et al. [24] proposed an alternative approach to support recyclability in space through the Solvent-Cast Direct-Write (SC-DW) method, which has been tested terrestrially using polyvinyl alcohol (PVA). This technique demonstrated that recycled material retains mechanical properties comparable to those of the original material. Nevertheless, its application in space is limited by practical constraints such as solvent handling, material compatibility, and integration into existing systems.

In contrast, fused filament fabrication (FFF) has already demonstrated high technology readiness and is actively being considered for integration into space-based manufacturing systems. FFF is one of the most widely used AM printing techniques for thermoplastic polymers [25] and has shown operational feasibility through successful testing aboard the ISS, where it was validated under microgravity conditions [26,27]. Further studies report on comparative experiments conducted in both microgravity and terrestrial environments, concluding that no engineering-significant effects of microgravity were observed on the FFF process or material outcomes. These findings highlight the robustness of the FFF process, confirming its suitability for the unique demands of space-based fabrication, and have spurred the development of advanced systems tailored to this context: Tethers Unlimited Inc. (TUI) introduced the ReFabricator, which combines polymer waste recycling and 3D printing in a single device [28], while NASA and Redwire are developing FabLab, a multimaterial fabrication laboratory aimed at providing next-generation in-space manufacturing capabilities [29].

This technological trajectory has sparked increasing interest in the use of FFF for recycling waste materials directly onboard spacecraft. However, despite these advancements, no studies to date have specifically addressed the disposal and subsequent recycling of space beverage packaging, a type of waste that constitutes a considerable portion of the total waste stream generated during space missions, especially those of extended duration. These multilayer packages, which are frequently used and composed of aluminum and polymeric materials, represent a critical and untapped resource for circular material strategies. Their high usage rate and material composition make them particularly relevant for future space missions, where the constraints of resupply logistics will require more efficient, closed-loop systems.

The present study seeks to fill this gap by proposing and validating a method for the recycling of astronaut beverage packaging waste through the FFF technique. This builds on our previous work [30,31], in which we developed and characterized a composite filament derived from beverage packaging, comprising 10 wt% of a PET–aluminum–LDPE multilayer (PAL) dispersed in an LDPE matrix. This composite exhibited optimal mechanical and thermal properties for FFF 3D printing, enabling the complete recycling of the packaging material (LDPE: 20 g, PAL: 2 g). Notably, the selected formulation enables the full reutilization of a single beverage package.

Expanding on these findings, the present study focuses on the 3D printing of the LDPE/PAL 10 wt% composite filament using FFF. To optimize printing parameters and conserve resources during the experimental phase, numerical simulations were conducted using Digimat-AM. These simulations aimed to predict thermal and mechanical behaviors during the printing process, ensuring the structural integrity and functionality of the printed components [32]. The integration of simulation tools into the additive manufacturing workflow is crucial for enhancing print quality and reliability, especially in the constrained environments of space missions. By precisely forecasting process-related phenomena, simulations facilitate the optimization of printing parameters, improving material performance and print quality. This approach supports sustainable manufacturing practices, ensuring the durability and efficiency of components essential for the success of long-duration space exploration missions.

2. Materials and Methods

2.1. Numerical Simulation

To perform additive manufacturing numerical simulations, the commercial software Digimat, version 2021.1 [33], was employed. Developed by Hexagon (Stockholm, Sweden) [34], Digimat-AM is a dedicated module within the broader Digimat suite, specifically designed to simulate additive manufacturing processes involving polymeric and composite materials.

2.1.1. Simulation Workflow

The Digimat-AM platform provides a structured six-step workflow that allows users to define process parameters, import geometries, and model the thermo-mechanical behavior during the build phase, while also evaluating the resulting distortions and residual stresses. A detailed description of the simulation procedure adopted in this study is presented in this section.

The process begins with the definition of the manufacturing scenario, where the user selects the printing technology, machine configuration, and type of analysis. In this work, simulations were set up for the fused filament fabrication (FFF) process, using a generic printer model provided by Digimat-AM. To ensure alignment between the virtual and experimental environments, the build chamber dimensions were manually adjusted to replicate those of the Ultimaker 3 (Ultimaker, Utrecht, Netherlands) [35], which was the printer used during the experimental phase of this study. Specifically, the chamber was set to 330 mm (X), 330 mm (Y), and 320 mm (Z). The selected analysis focused on evaluating the residual von Mises stresses (calculated according to Equation (1)) and warping of the geometric parts, aiming to quantify the deformation due to thermal gradients and shrinkage during and after material deposition.

$${\sigma }_{vM}=\sqrt{{\displaystyle \frac{1}{2}}{\left({\sigma }_{xx}-{\sigma }_{yy}\right)}^2+{\left({\sigma }_{yy}-{\sigma }_{zz}\right)}^2+{\left({\sigma }_{zz}-{\sigma }_{xx}\right)}^2+3\left({\tau }_{xy}^2+{\tau }_{yz}^2+{\tau }_{xz}^2\right)}$$

(1)

where

• ${\sigma }_{xx}$, ${\sigma }_{yy}$,${\sigma }_{zz}$ = normal stresses acting along the x, y, and z axes, respectively;
• ${\tau }_{xy}$, ${\tau }_{yz}$, ${\tau }_{zx}$ = shear stresses acting on the faces of the material element.

Distortion and stress prediction was carried out using the inherent strain method, which is the default approach in Digimat-AM due to its ability to provide accurate results while significantly reducing the computational time and resource consumption [33].

In the second step, the geometry of the component is imported into the simulation environment. Digimat-AM supports various Computer-Aided Design (CAD) formats; in this case, the parts were provided as Standard Tessellation Language (STL) files. During this phase, the user must also assign the material to be used in the simulation. This can be selected from Digimat’s internal database or manually defined if not available (a process described in detail in Section 2.1.3).

The third step involves defining the virtual manufacturing setup. This includes the part orientation on the build plate, the deposition path, and environmental boundary conditions. For all the simulated cases, the component was centered on the build platform and uniform thermal conditions were applied: the chamber temperature, ambient temperature, and final part temperature were all set to 23 °C. Heat transfer to the environment was modeled using a convection coefficient of 0.015 mW/mm² °C.

The fourth step entails the translation of the defined setup into a finite element model. Digimat-AM employs a voxel-based meshing approach, where the voxel size must not be smaller than the actual layer thickness used during the printing process. For the purposes of this study, a voxel resolution of 0.2 mm was adopted, offering a compromise between computational efficiency and geometric accuracy. The solver settings, including numerical methods and convergence criteria, were left at their default values.

Once the model is finalized, it is submitted for simulation in the fifth step.

The final step is dedicated to post-processing, where the simulation results are analyzed through advanced visualization tools. The software enables the assessment of temperature distribution, deformation fields, and residual stress patterns across the part. These outputs are critical for evaluating component manufacturability, as well as quantifying the influence of process parameters on both part quality and dimensional accuracy.

2.1.2. Description of Analyzed Geometries

To assess the influence of process parameters on part geometry and to evaluate the printability of the LDPE/PAL composite at 10 wt%, three CAD models with increasing geometric complexity were developed. The geometries were created using Autodesk Inventor Professional (version 2025) [36] and exported in STL format to ensure full compatibility with the simulation workflow implemented in Digimat-AM. Each model was designed to introduce progressively more challenging features, allowing for the evaluation of how printing conditions affect both manufacturability and final part quality.

Figure 1 illustrates the three geometries along with their respective dimensions. The first model is a simple square-shaped part, representative of basic prismatic forms. The second geometry features an elongated body with curved regions, introducing moderate complexity in terms of shape and contour. The third model presents multiple curvatures and fine details, aimed at testing the limits of resolution and deposition precision for the composite material.

2.1.3. Development of a Custom Material Model

The unavailability of the LDPE/PAL composite as a predefined material in Digimat-AM database required the implementation of a custom-defined model. This process involved modeling both the polymer matrix (LDPE) and the filler (PAL) to accurately represent the composite behavior within the simulation environment. The PAL filler, consisting of a multilayer laminate composed of approximately 68% LDPE, 16% PET, and 16% aluminum by mass, was modeled by integrating the behavior of each individual constituent phase according to their respective mass fractions. Specifically, the aluminum layer was represented using the AA1235 aluminum alloy, a material commonly employed in food packaging applications due to its excellent corrosion resistance and suitability for contact with food products [22].

To enable a reliable thermo-mechanical simulation, it was necessary to define a complete set of temperature-dependent material properties. These included both mechanical and thermal parameters essential for the accurate prediction of deformation and residual stresses during the additive manufacturing process. Specifically, data from the scientific literature were used to reconstruct the temperature-dependent behavior of the specific volume [38,39], coefficient of thermal expansion (CTE) [40,41], specific heat capacity [40,42], and thermal conductivity [43], as shown in Figure 2.

However, for the aluminum component of the PAL filler, the AA1235 alloy was modeled with constant property values across the studied temperature range, as its thermal properties exhibit negligible variation within this range [40,44]. The Poisson ratios of both the LDPE matrix and the PAL filler were also extracted from the literature [45,46] and subsequently incorporated into the model to estimate the effective Poisson ratio of the composite (

Table 1. The Poisson ratios of the LDPE matrix and PAL filler used for the composite model.

Component	Poisson Ratio (ν)
LDPE matrix	0.44 [45]
PAL filler	0.41 [45,46]

).

In parallel, experimental data were acquired to characterize the temperature-dependent Young’s modulus of the individual phases of the composite (Figure 3).

To evaluate the mechanical behavior of the LDPE/PAL 10 wt% composite, dynamic temperature scans were performed on three specimens using a Mettler Toledo DMA-1 instrument (Mettler Toledo, Columbus, OH, USA) in single-cantilever operational mode. The same test was also conducted on pristine LDPE samples, allowing for the estimation of the PAL filler’s contribution to the storage modulus by subtraction. The temperature was programmed to increase from −150 °C to 130 °C at a rate of 5 °C/min, with a fixed displacement amplitude of 35 μm and a frequency of 1 Hz. In particular, the selected temperature range was designed to replicate the extreme temperature conditions found on the lunar surface at the equator, corresponding to the minimum and maximum temperatures during nighttime and daytime [47,48]. These dynamic mechanical tests allowed for precise definition of the stiffness evolution with temperature, ensuring consistency between the simulated response and the experimental behavior of the composite system.

2.1.4. Investigation of Printing Parameter Variation in FFF Technique

The overall success of the fused filament fabrication process, as well as the dimensional accuracy and mechanical performance of the printed parts, is highly influenced by the selection and calibration of several key printing parameters [49,50,51]. These parameters can be classified into two main categories: structural parameters and manufacturing parameters. Structural parameters define the internal configuration of the printed object, and include the infill density, raster angle, build orientation, and infill pattern geometry. These factors critically affect the mechanical response and deformation behavior of the printed parts. Manufacturing parameters regulate the physical conditions under which the process takes place. These include layer thickness, printing temperature, printing speed, build plate temperature, chamber temperature, and ambient temperature. Their variation significantly impacts interlayer adhesion, thermal gradients, and the development of residual stresses and warpage [52].

In this study, the influence of three specific parameters was investigated: infill pattern (Figure 4), build plate temperature, and printing speed (

Table 2. Evaluated build plate temperatures and printing speeds.

Build Plate Temperature [°C]	Printing Speed [mm/s]
60	20
70	30
80	40
90	50

), with the aim of optimizing the printability of LDPE/PAL 10 wt% composite parts. The printing speed was found to significantly affect the material flow dynamics and the overall quality of the printed parts. At lower speeds, the simulations revealed an inadequate and non-uniform material deposition, which compromised the structural integrity of the components. These conditions also led to elevated residual stresses and greater warpage. Conversely, higher printing speeds promoted a more consistent material flow, resulting in enhanced surface quality, reduced deformation, and the successful production of defect-free parts. The infill pattern played a critical role in determining internal stress distribution and warpage behavior. The geometry of the infill influenced the way in which stresses were transferred and accumulated during the cooling phase, with certain patterns demonstrating greater efficacy in mitigating distortion and residual stresses. The build plate temperature was also a key factor in controlling the thermal gradient between the extruded filament and the print bed. An appropriate temperature setting improved first-layer adhesion and reduced warpage, particularly in the lower layers of the printed object.

The printing temperature was not included among the simulated variables, as its value was predefined based on processing conditions commonly reported in the literature for similar polymer-based materials [54,55,56,57]. Given the need for adequate melt flow and interlayer bonding, the composite was processed at a temperature well above its melting point [58,59].

To verify the material’s thermal stability throughout the processing window relevant to FFF, a Differential Scanning Calorimetry (DSC) analysis was performed on the LDPE/PAL 10 wt% composite. The objective of this analysis was to ensure the absence of degradation phenomena under the selected conditions. The analysis was carried out using a Pyris 8500 instrument (PerkinElmer, Waltham, MA, USA) on three powder specimens of the LDPE/PAL 10 wt% composite. A five-step heating and cooling cycle was employed, as shown in Figure 5.

The first step consisted of a heating ramp from 25 °C to 250 °C at 10 °C/min, aimed at evaluating the initial thermal behavior of the material. This was followed by an isothermal holding step at 250 °C for 1 min to stabilize the thermodynamic state. Subsequently, the samples were cooled down from 250 °C to 25 °C at the same rate (10 °C/min) to investigate the crystallization behavior of the composite; an additional isothermal hold at 25 °C for 1 min was included to ensure thermal equilibration. Finally, a second heating cycle from 25 °C to 250 °C at 10 °C/min was conducted to eliminate the material’s thermal history and to obtain accurate data regarding the melting transition. All analyses were performed under a nitrogen flow of 20 cc/min.

From the heating and cooling curves, the peak temperatures for both melting and crystallization transitions were extracted, along with the corresponding enthalpy values. Although the primary focus of this study was on the thermal degradation behavior of the material, melting and crystallization data and the relative enthalpy values are also reported for completeness. The extracted thermal data are summarized in

Table 3. Melting and crystallization temperatures and corresponding enthalpy values of LDPE/PAL 10 wt% composite powder from DSC analysis.

Transition	Temperature [°C]	Enthalpy [J/g]
Melting peak—1st heating	133.17 ± 0.68	164.93 ± 12.08
Melting peak—2nd heating	132.52 ± 0.44	191.42 ± 6.75
Crystallization peak	104.40 ± 0.56	185.49 ± 4.26

. The DSC results reveal that the composite remains stable up to 250 °C, thereby supporting the selection of 230 °C as a safe and effective printing temperature.

A range of values for each selected printing parameter was explored through numerical simulations, evaluating their effects on residual stress distribution and warpage.

For each of the three geometric configurations considered, the combination of parameters yielding the lowest values of stress and deformation was identified. These optimal settings were then used for the experimental FFF printing phase, ensuring coherence between the simulation predictions and the real behavior of the manufactured parts.

2.2. Three-Dimensional Printing of LDPE/PAL Composites

In the filament fused fabrication (FFF) process, a thermoplastic filament is continuously fed into a heated nozzle, melted and selectively deposited layer by layer to form the final 3D object. The quality and dimensional accuracy of the printed components are directly influenced by the consistency and integrity of the extruded filament, which therefore plays a critical role as the feedstock in this additive manufacturing technique [60]. In this study, the composite filament under investigation was derived entirely from recycled space beverage packaging, with LDPE from septum adapters and straws used as the matrix, and PET-aluminum-LDPE (PAL) trilaminate from the entire packaging acting as the reinforcing filler. Due to the limited availability of space beverage packaging, commercial coffee packaging, typically composed of PET, aluminum, and LDPE, was employed as a surrogate. The PAL filler was grated, sieved, and mechanically mixed with 500 μm LDPE powder (Thermo Fisher Scientific, Waltham, MA, USA), then thermally fused at 130 °C for 2 h to produce composite pellets. To produce suitable filaments for FFF, LDPE/PAL 10 wt% composite pellets were extruded using a Felfil Evo single-screw extruder (Felfil, Turin, Italy) equipped with a 1.75 mm diameter nozzle. The pellets were prepared by thermally consolidating the composite powder at 130 °C for 2 h in an oven, followed by mechanical fracturing to obtain irregular granules appropriate for extrusion. Extrusion parameters were optimized iteratively to achieve a uniform and circular cross-section of the filament, essential for stable feeding and dimensional fidelity during printing. The final selected conditions were a chamber temperature of 135 °C and a screw rotation speed of 6 rpm, which enabled the production of LDPE-based filaments containing 10 wt% PAL filler with consistent geometrical features. The dimensional quality of the extruded filament was assessed by measuring its diameter at multiple locations along the filament axis and across opposing directions of each cross-section. An average diameter of 1.74 ± 0.03 mm was recorded, confirming both roundness and longitudinal uniformity. Furthermore, the roundness of the filament was examined using a scanning electron microscope (SEM) VEGA LSH (Tescan, Brno, Czech Republic) operated at an accelerating voltage of 10 kV and a magnification of 119× (Figure 6).

The 3D printing of the filament was performed using an Ultimaker 3, a high-end FFF system equipped with dual extrusion capabilities, a heated build plate, and an open printing chamber (top and front). The printing process was managed through Ultimaker Cura (version 5.8.1) [53], a free open-source slicing software that supports various file formats, including the STL extension. Cura’s interface facilitates model positioning, orientation, and slicing, allowing for precise control over printing parameters (Figure 7).

3. Results

3.1. Numerical Simulation

The simulation campaign conducted in Digimat-AM aimed to assess the printability of the LDPE/PAL 10 wt% composite by evaluating the residual von Mises stress and total warpage across different geometrical structures and process parameters. As a baseline for comparison, a reference simulation was selected using concentric infill, a build plate temperature of 60 °C, and a printing speed of 20 mm/s. Under these conditions, the highest residual stress was recorded in the wheel geometry (6.205 MPa), followed by the wrench (5.340 MPa) and square (2.389 MPa) geometries. Regarding the total warpage, the wrench geometry showed the greatest deformation (3.120 mm), while the square and wheel geometries exhibited lower and comparable values (1.761 mm and 1.653 mm, respectively). These results underscore the influence of geometry on thermal stress accumulation and distortion behavior.

Focusing on the infill pattern, the concentric configuration consistently achieved the lowest residual von Mises stress across all geometries. In the square geometry, it also resulted in a total warpage of 1.74 mm, in line with the values observed for all other infill configurations. Given the significant stress reduction, concentric was identified as the optimal choice. For the wrench geometry, concentric again minimized residual stress (2.68 MPa) compared to zig zag (2.821 MPa), while the total warpage was slightly higher (3.07 mm vs. 3.044 mm, or 0.85% difference). Despite the small variation in deformation, the markedly lower stress makes concentric preferable. In the wheel geometry, concentric yielded the lowest residual stress, and no variation in warpage was observed across the tested infill patterns (1.624 mm constant), thus confirming concentric as the best option based on stress alone.

To provide a visual representation of the simulation results, Figure 8 shows the von Mises residual stress distribution and total warpage predicted for all geometries using a concentric infill pattern.

Regarding the printing speed, simulations revealed that increasing the speed from 20 mm/s to 50 mm/s led to a clear reduction in both residual stress and total warpage across all geometries (Figure 9).

The most pronounced effect was observed on von Mises stress, which decreased more significantly with higher speeds. This trend is attributed to the reduced thermal accumulation during faster deposition, limiting the development of internal stresses and thermal gradients.

Concerning the build plate temperature, which was varied in the simulations across 60 °C, 70 °C, 80 °C, and 90 °C, the results indicated no significant effect on either the residual von Mises stress or the total warpage for any of the tested geometries. This behavior aligns with findings reported in [52]. Consequently, the selection of the optimal build plate temperature was based on operational efficiency rather than mechanical performance. Specifically, the lowest tested temperature (60 °C) was chosen to minimize the heating time required for the build plate, thereby improving the overall process efficiency without compromising part quality.

The full set of simulation data, including detailed values of residual stress and total warpage for each geometry and configuration, is available in Appendix A (

Table A1. The Digimat-AM simulation results for the square specimen geometry: residual stress and warpage under varying printing conditions.

Square Specimen—Simulation Results
Run	Infill Pattern	Printing Speed [mm/s]	Bed Temperature [°C]	Residual Stress (von Mises) [MPa]	Total Warpage [mm]
1.	Zig Zag	50	60	1.204	1.741
2.	Triangles	50	60	1.205	1.74
3.	Grid	50	60	1.212	1.74
4.	Tri-Hexagon	50	60	1.205	1.741
5.	Concentric	50	60	1.198	1.74
6.	Concentric	20	60	2.389	1.761
7.	Concentric	30	60	1.802	1.751
8.	Concentric	40	60	1.452	1.744
9.	Concentric	50	70	1.198	1.74
10.	Concentric	50	80	1.198	1.74
11.	Concentric	50	90	1.198	1.74

,

Table A2. Digimat-AM simulation results for the wrench geometry: residual stress and warpage under varying printing conditions.

Wrench—Simulation Results
Run	Infill Pattern	Printing Speed [mm/s]	Bed Temperature [°C]	Residual Stress (von Mises) [MPa]	Total Warpage [mm]
1.	Zig Zag	50	60	2.821	3.044
2.	Triangles	50	60	2.898	3.106
3.	Grid	50	60	2.851	3.107
4.	Tri-Hexagon	50	60	2.984	3.106
5.	Concentric	50	60	2.68	3.07
6.	Concentric	20	60	5.34	3.12
7.	Concentric	30	60	3.997	3.095
8.	Concentric	40	60	3.205	3.08
9.	Concentric	50	70	2.68	3.07
10.	Concentric	50	80	2.68	3.07
11.	Concentric	50	90	2.68	3.07

and

Table A3. The Digimat-AM simulation results for the wheel geometry: residual stress and warpage under varying printing conditions.

Wheel—Simulation Results
Run	Infill Pattern	Printing Speed [mm/s]	Bed Temperature [°C]	Residual Stress (von Mises) [MPa]	Total Warpage [mm]
1.	Zig Zag	50	60	3.172	1.624
2.	Triangles	50	60	3.185	1.624
3.	Grid	50	60	3.227	1.624
4.	Tri-Hexagon	50	60	3.196	1.624
5.	Concentric	50	60	3.137	1.624
6.	Concentric	20	60	6.205	1.653
7.	Concentric	30	60	4.663	1.639
8.	Concentric	40	60	3.747	1.63
9.	Concentric	50	70	3.137	1.624
10.	Concentric	50	80	3.137	1.624
11.	Concentric	50	90	3.137	1.624

).

3.2. Three-Dimensional Printing of Parts

Following the identification of the optimal combination of printing parameters that minimized the von Mises residual stress and total warpage (

Table 4. The optimized FFF printing parameters selected for the fabrication of test geometries using the LDPE/PAL 10 wt% composite.

Process Parameter	Value
Printing temperature	230 °C
Build plate temperature	60 °C
Printing speed	50 mm/s
Infill pattern	Concentric
Infill density	10%
Layer thickness	0.2 mm

), which was determined through numerical simulations of the additive manufacturing process, three representative geometries were fabricated using the FFF technique: a square specimen, a wrench, and a wheel. All parts were printed with a 10% infill density, selected as a baseline condition for evaluating printability. This low infill density value was specifically selected to simulate the most demanding printing conditions, characterized by minimal internal support for upper layers and an increased risk of structural collapse.

The material employed in this study was a thermoplastic composite with the same composition as post-consumption beverage packaging used by astronauts, and the objective was to assess its processability via FFF, focusing exclusively on the ability to produce defect-free parts under space-relevant recycling scenarios (Figure 10).

To further validate the numerical predictions, experimental comparisons were performed by fabricating square specimens at two different printing speeds: a suboptimal speed of 20 mm/s and the optimized speed of 50 mm/s. At 20 mm/s, printing failure was observed, characterized by an insufficient and non-uniform material flow, which compromised the structural integrity of the printed parts. Additionally, these specimens exhibited higher residual stresses and increased warpage, as anticipated from the numerical simulations. In contrast, specimens printed at the optimized speed of 50 mm/s demonstrated improved surface quality, reduced warpage, and the successful fabrication of defect-free components. Figure 11 illustrates the visual differences between the specimens printed at the two speeds, confirming the strong correlation between the numerical predictions and experimental outcomes.

4. Conclusions

This study demonstrates the potential of fused filament fabrication for the 3D printing of recycled space beverage packaging, specifically an LDPE/PET-Aluminum-LDPE (PAL) composite, using a systematic approach involving numerical simulations and experimental validation. Through the use of Digimat-AM software, the effects of key printing parameters, such as the printing speed, infill pattern, and build plate temperature, on residual stresses and total warpage were thoroughly assessed.

The simulations revealed that varying the build plate temperature had no significant impact on the residual von Mises stress or total warpage for the tested geometries. This outcome supports the selection of the lower tested build plate temperature (60 °C) as the optimal choice for operational efficiency, reducing heating times while maintaining part quality. In contrast, the printing speed and infill pattern exhibited considerable effects.

Increasing the printing speed from 20 mm/s to 50 mm/s resulted in a notable reduction in both residual von Mises stress and warpage across all geometries, with the most evident effect observed for wheel geometry, where von Mises stress decreased from 6.205 MPa to 3.137 MPa. Simulations showed that the concentric infill pattern consistently resulted in the lowest residual von Mises stress across all geometries. In the square specimen, it achieved 1.198 MPa with a total warpage of 1.74 mm, while in the wrench, it reduced stress to 2.68 MPa, compared to 2.82 MPa for the zig zag pattern, with a negligible difference in warpage (3.07 mm vs. 3.04 mm). In the wheel geometry, concentric again minimized stress, and warpage remained constant (1.624 mm) across all patterns. These results identify concentric infill as the most effective configuration for stress reduction without compromising dimensional stability. The successful fabrication of square, wrench, and wheel geometries under the optimized printing conditions (60 °C build plate temperature, 50 mm/s printing speed, and concentric infill pattern) confirmed the numerical predictions, demonstrating reduced warpage and residual stress. High-quality defect-free parts were obtained, providing a preliminary validation of the LDPE/PAL composite’s suitability for additive manufacturing. To further confirm the correlation between numerical simulations and experiments, square specimens were printed at a suboptimal speed of 20 mm/s and an optimized speed of 50 mm/s. At 20 mm/s, printing failure occurred due to insufficient material flow, causing structural defects, higher residual stresses, and increased warpage, as predicted by the simulations. In contrast, specimens printed at 50 mm/s exhibited reduced warpage and flawless fabrication. These results validate the Digimat-AM numerical approach as an effective tool for optimizing process parameters in the fused filament fabrication (FFF) technique, and highlight the potential of recycled LDPE/PAL composites for in situ resource utilization during long-duration space missions.