Prototype characterization

Grant Agreement number: 101058371 Project acronym: ESTELLA Project title: Design of bio-based thermoset polymer with recycling capability by dynamic bonds for bio-composite manufacturing Grant Agreement number: 101058371 Project acronym: ESTELLA Project title: Design of bio-based thermoset polymer with recycling capability by dynamic bonds for bio-composite manufacturing Prototype Characterization DELIVERABLE 3.4 Contractual Date of Delivery: 31-05-2025 Actual Date of Delivery: 04-07-2025 Lead contractor for this deliverable: Author(s): CID Participants(s): CID, WK+ WP contributing to the deliverable: WP 3 Nature: PU Version V. 2 Grant Agreement 101058371 – Project ESTELLA Deliverable T3.4 ©ESTELLA - This is the property of ESTELLA Parties: shall not be distributed/reproduced without formal approval of ESTELLA General Assembly. This reflects only the author’s views. The Community is not liable for any use that may be made of the information contained therein. REVISION TABLE Document version Date Modified sections - Details V1 04.07.2025 First version written by CID and WK+ V2 21.07.2025 Second version reviewed by ICSO Grant Agreement 101058371 – Project ESTELLA Deliverable T3.4 ©ESTELLA - This is the property of ESTELLA Parties: shall not be distributed/reproduced without formal approval of ESTELLA General Assembly. This reflects only the author’s views. The Community is not liable for any use that may be made of the information contained therein. Table of Contents TABLE OF CONTENTS .................................................................................................................... 3 ABBREVIATIONS ........................................................................................................................... 4 EXECUTIVE SUMMARY .................................................................................................................. 5 1. INTRODUCTION .................................................................................................................... 6 2. PROTOTYPES MANUFACTURED ........................................................................................... 7 3. METROLOGICAL EVALUATION ........................................................................................... 9 3.1. PROTOTYPE 1: SCOOTER PLATFORM ................................................................................... 9 3.2. PROTOTYPE 2: WINDOW PROFILE ..................................................................................... 13 4. MECHANICAL TESTS .......................................................................................................... 18 4.1. PROTOTYPE 1: SCOOTER PLATFORM ................................................................................. 19 4.1.1. Static Test ......................................................................................................... 20 4.1.2. Dynamic Test ................................................................................................... 23 4.1.3. Fatigue Test ...................................................................................................... 24 4.2. PROTOTYPE 2: WINDOW PROFILE ..................................................................................... 26 4.2.1. Static Testing .................................................................................................... 27 4.2.2. Dynamic Test ................................................................................................... 30 4.2.3. Fatigue Testing................................................................................................. 33 5. NON-DESTRUCTIVE TESTING ............................................................................................. 35 6. POST-PROCESSED MICROSTRUCTURAL CHARACTERISATION ........................................ 38 7. LONG-TERM BEHAVIOR EVALUATION.............................................................................. 42 7.1. PROTOTYPE 1: SCOOTER PLATFORM ................................................................................. 43 7.2. PROTOTYPE 2: WINDOW PROFILE ..................................................................................... 51 8. CONCLUSIONS .................................................................................................................. 58 BIBLIOGRAPHY ........................................................................................................................... 59 Grant Agreement 101058371 – Project ESTELLA Deliverable T3.4 ©ESTELLA - This is the property of ESTELLA Parties: shall not be distributed/reproduced without formal approval of ESTELLA General Assembly. This reflects only the author’s views. The Community is not liable for any use that may be made of the information contained therein. Abbreviations CAD: Computer-Aided Design RTM: Resin Transfer Moulding 3D: Three-Dimensional EN: European Norm (Norma Europea) CTE: Spanish Technical Code of Building Regulations FEM: Finite Element Method MPa: Megapascal N·mm: Newton per Millimetre mm: Millimetre g: Grame s: Second Grant Agreement 101058371 – Project ESTELLA Deliverable T3.4 ©ESTELLA - This is the property of ESTELLA Parties: shall not be distributed/reproduced without formal approval of ESTELLA General Assembly. This reflects only the author’s views. The Community is not liable for any use that may be made of the information contained therein. Executive Summary This deliverable presents the results of the experimental validation carried out on two prototypes manufactured with bio-based resins and natural reinforcements: a scooter platform (ICSO resin and WK+ hemp fibre) and a window profile (ICSO resin and Celabor nanocellulose). The objective of this task is to assess the structural integrity, dimensional stability, and environmental ageing of the developed composites, as part of the ESTELLA project. Three main categories of testing were performed: • Dimensional evaluation, including 3D scanning and local thickness measurements to verify manufacturing consistency and identify geometrical deviations. • Mechanical testing, including static, dynamic, and fatigue adapted to each prototype and application. • Environmental ageing to evaluate the influence on tensile properties after chemical absorption. The results show that the scooter platform meets the mechanical performance requirements, including under fatigue and impact loading. The window profile, while working well under static loads, is more sensitive to dynamic loads and fatigue, especially due to its fragility under impact. The aging study confirms a higher absorption sensitivity in the hempreinforced composite, whereas the nanocellulose-reinforced material exhibits higher dimensional and mechanical stability over time due to the lower absorption. Overall, the mechanical and environmental behaviour of both prototypes confirms the potential of bio-based composites for lightweight applications, although further optimization is required especially for composites exposed to dynamic or wet conditions. Grant Agreement 101058371 – Project ESTELLA Deliverable T3.4 ©ESTELLA - This is the property of ESTELLA Parties: shall not be distributed/reproduced without formal approval of ESTELLA General Assembly. This reflects only the author’s views. The Community is not liable for any use that may be made of the information contained therein. 1. Introduction This deliverable presents the characterisation of the prototypes manufactured in Task T3.3, with the aim of evaluating their mechanical performance, structural integrity, and durability under environmental ageing conditions. The prototypes (scooter platform and window profile) were developed using the bio-based resin selected in Task T3.2 with continuous fibre reinforcement (hemp) and short fibre reinforcement (nanocellulose). Both prototypes have been designed with potential application in the fields of personal mobility and the construction sector, respectively. The tests carried out in this task cover three main areas: • Metrological: to ensure dimensional stability and identify deviations related to the manufacturing process. • Mechanical: to evaluate resistance to static, dynamic and repetitive impact loads. • Environmental ageing: to evaluate the effect of absorption on material properties following prolonged water immersion. This deliverable is connected directly on the work carried out in the following previous tasks: • T3.1: Design of the scooter platform and window profile prototypes. • T3.2: Mould design and validation of the design through simulation. • T3.3: Manufacturing prototypes using RTM (scooter platform) and open-mould injection (window profile). Due to the type of composites and manufacturing processes it was defined a testing plan for the characterization. The analysis of the characterization weas established in Deliverable T3.1 (Table 2), including static strength, fatigue durability, weight reduction, and acceptable deflections under service loads. These steps were defined prior to prototype manufacturing and guided the design and material selection of the project. The characterisation carried out in Task T3.4 provides an experimental evaluation of whether the manufactured prototypes meet those initial targets. The results obtained also allow assessing the suitability of the selected bio-based materials for applications and contribute to a partial validation of the structural requirements defined in earlier project phases. Grant Agreement 101058371 – Project ESTELLA Deliverable T3.4 ©ESTELLA - This is the property of ESTELLA Parties: shall not be distributed/reproduced without formal approval of ESTELLA General Assembly. This reflects only the author’s views. The Community is not liable for any use that may be made of the information contained therein. 2. Prototypes manufactured As described in the Deliverable T3.3, scooter platform prototypes were manufactured by RTM process and window profile prototypes were manufactured by open mould injection. Once manufactured the scooter platform, this composite adopts a rectangular plate shape, as shown in the Figure 1, with approximate dimensions of 660 × 210 × 4 mm. On the other hand, the window profiles were obtained with a constant cross-sectional geometry along their longitudinal axis and are shown in Figure 1. The dimensions of the manufactured window profiles were 78x57x220 mm. Before performing the test characterisations, both composites were machined to obtain their final testing geometry as shown in Figure 2. The scooter platforms were machined according to the geometry defined in Task T3.1, aiming to replicate the shape of the Xiaomi scooter platform used as a reference in that task. In contrast, the window profiles were machined to eliminate irregularities at the ends and to ensure that all profiles had the same final length. Figure 1. Manufactured composites Grant Agreement 101058371 – Project ESTELLA Deliverable T3.4 ©ESTELLA - This is the property of ESTELLA Parties: shall not be distributed/reproduced without formal approval of ESTELLA General Assembly. This reflects only the author’s views. The Community is not liable for any use that may be made of the information contained therein. As stated in Deliverable T3.3, a total of 11 scooter platforms and 11 window profiles were selected for mechanical testing. Figure 2. Machined prototypes after the manufacturing process Grant Agreement 101058371 – Project ESTELLA Deliverable T3.4 ©ESTELLA - This is the property of ESTELLA Parties: shall not be distributed/reproduced without formal approval of ESTELLA General Assembly. This reflects only the author’s views. The Community is not liable for any use that may be made of the information contained therein. 3. Metrological evaluation A metrological evaluation of the prototypes was carried out to verify the thickness homogeneity. This is essential to ensure that the mechanical properties evaluated are not influenced by significant undesired dimensional variations. To this end, representative zones were defined on the surface of each prototype, where thickness measurements were taken using a calibrated instrument, KROEPLIN DIGITAL 0–20 mm model. The results obtained will confirm the dimensional quality of the parts selected for testing and helping to identify potential irregularities from the manufacturing process. Additionally, a 3D scan was performed on one prototype of each type of composite using the Handy Scan Black Elyte device and the VX Elements scanning and measurement software. The aim was to compare the actual dimensions of the manufactured parts with those of the CAD models developed during Task T3.1. 3.1. Prototype 1: Scooter platform A detailed thickness analysis was carried out to assess the regularity of the plate obtained via the RTM process. For this purpose, 9 scooter platform were selected and divided into 8 zones along its main axis, as shown in Figure 3. In each of these zones, several thickness measurements were taken using the instrument, and a representative value was calculated per zone. Figure 3. Zones defined on the scooter platform for metrological evaluation Grant Agreement 101058371 – Project ESTELLA Deliverable T3.4 ©ESTELLA - This is the property of ESTELLA Parties: shall not be distributed/reproduced without formal approval of ESTELLA General Assembly. This reflects only the author’s views. The Community is not liable for any use that may be made of the information contained therein. Additionally, a longitudinal thickness analysis was performed by comparing values obtained in the same zone but at two different sections of the profile: the initial section (Section 1) and the final section (Section 5), as shown in Figure 9. The aim of this analysis is to determine whether significant thickness variations occur along the longitudinal axis, which could be due to variable resin flow or thermal effects during curing. Individual graphs were generated for each zone, showing thickness measurements from both sections (Section 1 and Section 5) for each of the four profiles. With Figure 9 is possible to compare the thickness values for each zone between the two longitudinal sections, providing the stability of wall thickness along the profile length. Generally, differences between sections are minor for most zones, suggesting a relatively uniform thickness distribution along the profile. Nonetheless, some exceptions were noted, such as in Zones 1 and 4, where more significant deviations were detected between the two sections in certain prototypes. In particular, profile 3 exhibited a maximum deviation of 0.189 mm between Zones 1 and 4, potentially indicating a slightly asymmetric material distribution in those walls. This analysis confirms a longitudinal thickness stability, with small variations that do not damage the dimensional integrity of the window profile. Finally, a statistical analysis of the thickness values for each manufactured profile was carried out (¡Error! No se encuentra el origen de la referencia.) to assess the manufacturing process and the homogeneity between different profiles. The analysis includes mean value, standard deviation, minimum and maximum values, and the 25th, 50th and 75th percentiles. Figure 9. Longitudinal thickness analysis of the tested window profiles Grant Agreement 101058371 – Project ESTELLA Deliverable T3.4 ©ESTELLA - This is the property of ESTELLA Parties: shall not be distributed/reproduced without formal approval of ESTELLA General Assembly. This reflects only the author’s views. The Community is not liable for any use that may be made of the information contained therein. The results show that the profiles exhibit good dimensional validity, with moderate standard deviations and values clustered around the mean. Notably, profile 1 exhibited the lowest dispersion, while profile 3 showed the highest variability, possibly due to slight differences during curing or resin flow during injection. Nevertheless, all profiles remain within a reasonable range (between 3.645-4.158mm), without significant outliers, confirming the repeatability of the process. Most of the data used for the metrological assessment of the window profiles were obtained from 3D scans of the four measured profiles. Additionally, this scanning process allows the comparison between the manufactured geometries and the reference CAD file used to design the injection mould. Figure 10 shows selected images from the dimensional report generated using the Handy Scan Black Elyte device and VX Elements scanning and measurement software. Figure 10. 3D scan of the window profiles Table 6.Statistical analysis of the measured thickness of the window profiles Grant Agreement 101058371 – Project ESTELLA Deliverable T3.4 ©ESTELLA - This is the property of ESTELLA Parties: shall not be distributed/reproduced without formal approval of ESTELLA General Assembly. This reflects only the author’s views. The Community is not liable for any use that may be made of the information contained therein. 4. Mechanical tests Mechanical test characterisation has been carried out on the scooter platforms and window profiles in order to evaluate the structural behaviour of the prototypes under different loading conditions. The tests performed are divided into three main categories: • Static tests, which allow the determination of deformation, maximum load capacity, and stiffness of the prototypes under quasi-static conditions. • Dynamic tests, aimed to analyse the behaviour under impacts or short-duration loads. • Fatigue tests, used to study the resistance of the prototypes under repeated cyclic loads. A total of three units were manufactured for the static and dynamic tests, and five units were produced for the fatigue tests for each of the two prototypes. Each type of test is described in detail in the following sections, including the procedure adopted and the analysis of the results obtained. Grant Agreement 101058371 – Project ESTELLA Deliverable T3.4 ©ESTELLA - This is the property of ESTELLA Parties: shall not be distributed/reproduced without formal approval of ESTELLA General Assembly. This reflects only the author’s views. The Community is not liable for any use that may be made of the information contained therein. 4.1. Prototype 1: Scooter platform The tests conducted on the scooter platforms were intended to assess their structural performance under various loading conditions representative of actual urban use. To this end, eleven manufactured platforms are used. To validate the design developed in Task T3.1 and to inform the design of the test fixture, a 3D model (.stl format) of a commercial Xiaomi scooter was downloaded and used as a geometric reference for the platform development in Task T3.1. Figure 11 shows a direct comparison between the CAD model of the platform and the scanned model of the original scooter, confirming dimensional compatibility and ensuring that the designed component fits within the actual mounting and operating conditions. Moreover, the 3D model of the reference scooter was analysed to understand how the surrounding components such as the chassis and wheels contribute structurally during operation. Based on this analysis, a test was designed to replicate the real loading and support conditions by the scooter when someone will use. The fixture consists of a structure made from rectangular-section aluminium tubes that simulates the scooter chassis and allows the platform to be fixed by bolts. Solid wheels, similar to the original ones and compatible with the reference model, were also integrated into the fixture to better replicate Figure 11. Scooter platform designed in the T3.1 task and overlaid with the 3D model of the Xiaomi scooter taken as a reference. Grant Agreement 101058371 – Project ESTELLA Deliverable T3.4 ©ESTELLA - This is the property of ESTELLA Parties: shall not be distributed/reproduced without formal approval of ESTELLA General Assembly. This reflects only the author’s views. The Community is not liable for any use that may be made of the information contained therein. the dynamic response of the full system during testing. Figure 12 shows the custom fixture used to secure the platforms during all mechanical tests. The platforms were subjected to three types of tests: static, dynamic and fatigue. Each test type was used to investigate different aspects of the mechanical behaviour of the prototype, such as strength, energy absorption capacity, cumulative deformation and damage tolerance. The test procedures were adapted to the specific geometry of the platform and to the typical loading conditions encountered in real use, ensuring a representative characterisation of the product. 4.1.1. Static Test The aim of the static test on the scooter platform is to assess its ability to withstand a maximum vertical load without undergoing permanent deformations that could compromise its functionality. The test was defined taking as reference the EN 17128 standard, which specifies strength requirements for personal mobility devices. Specifically, the standard states that a load equivalent to the maximum payload of the vehicle, multiplied by a safety factor of 2.5, must be applied over a contact area of 100 mm × 100 mm for a specified duration. The platform is placed into a rigid fixture simulating the actual installation on the scooter chassis, using the same attachment points defined during the design phase based on the Xiaomi reference model, as shown in Figure 12. In this case, and in accordance with the standard's recommendations, a load of 250 kg was applied to the centre of the platform and maintained for 10 minutes. The load was applied using a pneumatic working with a maximum force of 25 kN and a load cell with the same capacity. The scooter platforms 1, 2 and 3 were tested in the static test. After removing the load, the residual deformation was measured. To be considered valid, the residual displacement must be less than 3 mm. Figure 12. Fixings on the scooter platform for all mechanical tests Grant Agreement 101058371 – Project ESTELLA Deliverable T3.4 ©ESTELLA - This is the property of ESTELLA Parties: shall not be distributed/reproduced without formal approval of ESTELLA General Assembly. This reflects only the author’s views. The Community is not liable for any use that may be made of the information contained therein. This procedure allows verification of the static strength of the manufactured composites when working with the rest of the structural elements, and validates its performance under a loading condition representative of real-world use. Figure 13 shows the setup used for the static test on the scooter platform. The results of the static tests are presented in Figure 14, which shows the force–displacement curves obtained for the three tested platforms. All specimens supported the applied load of 2500 N (equivalent to 250 kg) without exhibiting any visible damage or residual deformation, reaching the criteria defined by the EN 17128 standard. The maximum displacement recorded by the ranged from 11 mm to 12 mm, with minor variations between tests. This displacement includes the combined deformation of the platform, the aluminium support structure, and any compliance in the system. After unloading, no visible permanent deformation was observed in the tested platforms, suggesting fully elastic behaviour under the applied load. Figure 13. Static test setup for the scooter platform Figure 14. Results of the static test of the scooter platform Grant Agreement 101058371 – Project ESTELLA Deliverable T3.4 ©ESTELLA - This is the property of ESTELLA Parties: shall not be distributed/reproduced without formal approval of ESTELLA General Assembly. This reflects only the author’s views. The Community is not liable for any use that may be made of the information contained therein. The force–displacement curves show a consistent and reproducible mechanical response across all tests, with an initially linear trend indicating good structural stiffness. No signs of damage such as cracking, delamination, or surface degradation were detected upon visual inspection. These results confirm the static load-bearing capacity of the hemp-reinforced resin platform and its suitability for urban mobility conditions. Grant Agreement 101058371 – Project ESTELLA Deliverable T3.4 ©ESTELLA - This is the property of ESTELLA Parties: shall not be distributed/reproduced without formal approval of ESTELLA General Assembly. This reflects only the author’s views. The Community is not liable for any use that may be made of the information contained therein. 4.1.2. Dynamic Test The purpose of this test is to simulate the dynamic loads experienced by the platform during regular scooter use, such as riding over bumps, descending kerbs, or impacts resulting from abrupt manoeuvres. Its definition is based on the EN 17128 standard, which establishes safety criteria for personal mobility devices. The procedure consists of performing a controlled vertical impact on the platform, which is placed on a test reproducing the actual installation conditions and allowing rotation about the rear axle. First, a static mass of 50 kg is secured at the centre of the platform to simulate the weight of an user. The assembly is then fixed in such a way that the rear wheel can pivot around its central axis. Finally, the front wheel is raised to a specific height of 200 ± 1 mm and dropped onto a rigid steel frame. Figure 15 shows the configuration used for the dynamic test on the scooter platform. Following the test, the platform is visually inspected for signs of damage, such as cracks, delaminations, or residual deformations. The platform is consider having passed the test if no structural damage or permanent deformation exceeding the 3 mm limit (as defined by the standard) is detected. Finally, all platforms supported the test without permanent deformation, cracks or fractures. The scooter platforms 4, 5 and 6 were tested in the dynamic test. Grant Agreement 101058371 – Project ESTELLA Deliverable T3.4 ©ESTELLA - This is the property of ESTELLA Parties: shall not be distributed/reproduced without formal approval of ESTELLA General Assembly. This reflects only the author’s views. The Community is not liable for any use that may be made of the information contained therein. This test provides an assessment of the impact resistance of the hempreinforced resin material used in the platform and its ability to absorb energy without compromising structural integrity. The dynamic impact test was performed on three different scooter platforms, each of which underwent a single drop test as described. In all three test, the platforms successfully withstood the impact without exhibiting visible damage, cracks, delamination, or permanent deformation. These results confirm that the platforms meet the acceptance criteria defined in the EN 17128 standard, where residual deformation must remain below 3 mm and no structural failure must occur. Although the vertical deformation was not quantitatively recorded during the test, visual inspection confirmed that all specimens retained their geometry and structural functionality after impact. No signs of damage or detachment were observed at the assembly points or along the platform surface. Overall, this test confirms the capacity of the hemp-reinforced resin composite to withstand dynamic impacts representative of real-world scooter use. The material demonstrated adequate energy absorption and structural integrity, supporting its potential application in urban mobility devices. 4.1.3. Fatigue Test The fatigue test was designed to evaluate the long-term mechanical durability of the scooter platform when subjected to repeated loading, simulating typical usage conditions over its service life. The procedure was Figure 15. Dynamic test setup on the scooter platform Grant Agreement 101058371 – Project ESTELLA Deliverable T3.4 ©ESTELLA - This is the property of ESTELLA Parties: shall not be distributed/reproduced without formal approval of ESTELLA General Assembly. This reflects only the author’s views. The Community is not liable for any use that may be made of the information contained therein. defined based on the EN 17128 standard, adapted to the available testing resources and tailored to the geometry of the fabricated prototypes. The scooter platforms 7, 8, 9, 10 and 11 were tested in the fatigue test (platforms 10 and 11 were not measured). The test consisted of applying an oscillating vertical load at the centre of the platform, which was placed on a rigid support replicating real conditions, like shown in Figure 12. The load applied corresponded to approximately 80% of the estimated maximum user weight (80 kg) and was delivered using the same pneumatic and load cell employed in the static test. The loading frequency was set to 0.67 Hz (one cycle every 1.5 seconds), completing a total of 15,000 cycles. This corresponds to a highusage scenario equivalent to approximately 5 impacts per day over an eight-year period. Five untested platforms were subjected to this fatigue test. The acceptance criterion was defined as the absence of visible damage (cracks, fractures, or delamination) and no significant residual deformation upon completion of the test. Grant Agreement 101058371 – Project ESTELLA Deliverable T3.4 ©ESTELLA - This is the property of ESTELLA Parties: shall not be distributed/reproduced without formal approval of ESTELLA General Assembly. This reflects only the author’s views. The Community is not liable for any use that may be made of the information contained therein. To rigidly secure the profile during the test, the custom fixture shown in Figure 18 was employed. The full test setup is illustrated in Figure 22. The impactor was released freely from various heights onto the centre of one of the profile’s outer faces, starting from the minimum drop height specified in the standard (100 mm, corresponding to Class 0). If the profile withstood the impact without exhibiting any signs of permanent deformation, cracking, fracture or other failure indicators, the drop height was progressively increased. Otherwise, the profile was classified according to the last drop height sustained without visible failure. Figure 23 shows the classification table defined in the UNE-EN 13049 standard for soft and heavy body impact resistance. This test was conducted on three of the window profiles manufactured within the project. In two of the three cases, the profile underwent complete fracture upon release of the impactor from the initial 100 mm height specified by the UNE-EN 13049 standard. As a result, no further tests or height increments were required. In the third case, the profile developed a visible crack but did not completely fracture. Figure 23. Table of impact heights extracted from the UNE EN 13049 standard. Figure 22. Dynamic test on the window profile Grant Agreement 101058371 – Project ESTELLA Deliverable T3.4 ©ESTELLA - This is the property of ESTELLA Parties: shall not be distributed/reproduced without formal approval of ESTELLA General Assembly. This reflects only the author’s views. The Community is not liable for any use that may be made of the information contained therein. Figure 24 displays the results of the three impact tests, illustrating the fractures and cracks observed on the profiles after testing. In all cases, the profiles failed under the impact from the lowest standard-defined height (100 mm). According to the adapted classification criteria from UNE-EN 13049, the tested profiles fall within Class 0, indicating very limited resistance to localised impacts. This behaviour may be influenced by the profile geometry, the type of reinforcement used, and the inherent brittleness of the bio-based materials selected for their manufacture. However, it is important to note that this classification is not directly equivalent to that defined in the original standard, as the testing procedure was extrapolated and adapted to suit the specific context of the project. In the standardised procedure, the impactor strikes the glazing panel mounted within the window, transmitting the load indirectly to the frame profiles. In contrast, in this adapted test, the impact was applied directly onto the profile itself, resulting in a much more severe and localised loading condition that does not fully represent the behaviour of a complete window assembly. Moreover, the outcome of this test may be further affected by multiple factors, such as the simplified geometry of the profile, the characteristics of the reinforcement material, and the fragility associated with the biobased composite used. Overall, the test provides an initial comparative reference regarding the impact resistance of such profiles, although it cannot be considered a strict classification within the original terms defined by UNE-EN 13049. Figure 24. Window profiles after impact test Grant Agreement 101058371 – Project ESTELLA Deliverable T3.4 ©ESTELLA - This is the property of ESTELLA Parties: shall not be distributed/reproduced without formal approval of ESTELLA General Assembly. This reflects only the author’s views. The Community is not liable for any use that may be made of the information contained therein. 4.2.3. Fatigue Testing The fatigue test carried out on the window profile aimed to evaluate its resistance to repeated loads, simulating the long-term effects of windinduced stresses. The methodology was based on the EN 12211:2017 standard, which defines cyclic wind load testing for complete windows, and the dynamic pressure was calculated in accordance with Eurocode 1 (EN 1991-1-4). The window profiles 7, 8, 9, 10 and 11 were tested (none of them were measured). The profile was rigidly secured using the clamping system shown in Figure 18, and the general test setup is illustrated in Figure 25. A load of 3000 N was applied to one of the profile’s outer faces using a steel plate connected to a pneumatic actuator. The load was applied for 5,000 cycles at a frequency of 1 Hz, following a high-demand protocol designed to generate significant mechanical stresses capable of producing progressive damage phenomena without the need for testing over millions of cycles. Figure 26 shows the loading and unloading sequence during the test. Figure 25. Window profile fatigue test setup Unload Load Figure 26. Loading and unloading moment of the fatigue test of the window profile Grant Agreement 101058371 – Project ESTELLA Deliverable T3.4 ©ESTELLA - This is the property of ESTELLA Parties: shall not be distributed/reproduced without formal approval of ESTELLA General Assembly. This reflects only the author’s views. The Community is not liable for any use that may be made of the information contained therein. In Figure 27Figure 27. Fatigue load cycle on window profiles, the maximum displacement of the tested prototypes is plotted for selected cycles, showing that all profiles experienced a peak deformation of approximately 5 mm at the maximum load in each cycle. Of the five profiles tested, Profile 7 withstood the full 5,000 loading cycles, although a visible surface crack developed across a significant section of the component. In contrast, all others teste profiles failed before completing all load cycles and obtaining brittle fractures. The post-test appearance of the tested profiles are shown in Figure 28Figure 28. Results of fatigue tested profiles. Figure 27. Fatigue load cycle on window profiles 3000 cycles 1500 cycles 5000 cycles Figure 28. Results of fatigue tested profiles 2300 cycles 1700 cycles Grant Agreement 101058371 – Project ESTELLA Deliverable T3.4 ©ESTELLA - This is the property of ESTELLA Parties: shall not be distributed/reproduced without formal approval of ESTELLA General Assembly. This reflects only the author’s views. The Community is not liable for any use that may be made of the information contained therein. 5. Non-destructive testing The objective of this experiment was to detect defects in composite materials using thermographic non-destructive testing (NDT). The tested samples were scooter platforms parts fabricated with vitrimer-based resin and hemp natural fiber reinforcement. Thermographic NDT relies on the detection of infrared radiation emitted from external heat source to the surface of a material. Variations in surface temperature can indicate subsurface defects, inconsistencies, or structural anomalies, allowing for inspection without causing any damage to the sample. All the measurements were performed using an infrared camera. To evaluate the presence and repair of defects, the following steps were conducted: 1. A baseline thermographic image was captured from the front surface of a defect-free sample. 2. An artificial surface crack was then introduced to the sample. 3. The sample was re-imaged to observe changes in the thermal profile. 4. The damaged area was treated with a hot air gun and subsequently heated in a temperature-controlled oven at 70 °C to initiate the vitrimer’s self-healing behavior. 5. A final thermographic measurement was taken after the thermal treatment. Figure 29 shows the thermographic image of the original, undamaged sample. No anomalies are visible in the inspected area. Grant Agreement 101058371 – Project ESTELLA Deliverable T3.4 ©ESTELLA - This is the property of ESTELLA Parties: shall not be distributed/reproduced without formal approval of ESTELLA General Assembly. This reflects only the author’s views. The Community is not liable for any use that may be made of the information contained therein. In Figure 30, a distinct linear anomaly appears in the thermographic image after introducing the crack, confirming the presence of the defect. Figure 30. Thermographic image after inducing a surface crack. A clear linear anomaly is visible in the marked area, indicating the presence of the defect. Figure 29. Thermographic image of the composite sample in its initial, defect-free state. No anomalies are visible in the marked region Grant Agreement 101058371 – Project ESTELLA Deliverable T3.4 ©ESTELLA - This is the property of ESTELLA Parties: shall not be distributed/reproduced without formal approval of ESTELLA General Assembly. This reflects only the author’s views. The Community is not liable for any use that may be made of the information contained therein. Finally, Figure 31reveals a significant change in the thermal image after healing: the white line that previously indicated the crack has greatly diminished or disappeared. While the complete vitrimer-based self-healing behaviour was not conclusively observed, the experiment demonstrated that thermographic imaging is an effective method for identifying and monitoring defects in natural fiber-reinforced vitrimer composites. As a conclusion, thermographic non-destructive testing proved to be a reliable technique for detecting and tracking defects in vitrimer-based composite materials. Although full self-healing behavior was not clearly confirmed through thermal imaging, the results indicate the potential of this method for monitoring damage and repair in future vitrimer systems. Figure 31. Thermographic image following thermal treatment at 70 °C. The previously visible white line is significantly reduced, suggesting partial healing of the crack. Grant Agreement 101058371 – Project ESTELLA Deliverable T3.4 ©ESTELLA - This is the property of ESTELLA Parties: shall not be distributed/reproduced without formal approval of ESTELLA General Assembly. This reflects only the author’s views. The Community is not liable for any use that may be made of the information contained therein. 6. Post-processed microstructural characterisation Both prototypes were characterized studying the microstructure postprocessing. For this aim, the characterization results were compared with the initial composites. The characterization used was micrography and fractography. For the micrography, it was studied the composites superficially using an optical microscope and an electronic microscope. In the case of the fractography, both composites were studied with the same techniques but analyzed after the fracture during the mechanical test. Firstly, scooter platform material was compared with optical microscope previous and after mechanical test (Figure 32 and Figure 33Figure 34. Transversal section of the scooter platform material after mechanical test with optical microscope). It is observed that in both cases, there is a small lack of resin. In the case of the specimen after the tensile test, a symmetrical fracture across the section it is observed. In the case of the image of the transversal section (Figure 34Figure 34. Transversal section of the scooter platform material after mechanical test with optical microscope), it can be observed the five layers of the fibres. Moreover, a good adhesion of the fibres and resin is seen in the section. Figure 32. Initial scooter platform material with optical microscope Grant Agreement 101058371 – Project ESTELLA Deliverable T3.4 ©ESTELLA - This is the property of ESTELLA Parties: shall not be distributed/reproduced without formal approval of ESTELLA General Assembly. This reflects only the author’s views. The Community is not liable for any use that may be made of the information contained therein. Figure 33. Scooter platform material after mechanical test with optical microscope Figure 34. Transversal section of the scooter platform material after mechanical test with optical microscope Grant Agreement 101058371 – Project ESTELLA Deliverable T3.4 ©ESTELLA - This is the property of ESTELLA Parties: shall not be distributed/reproduced without formal approval of ESTELLA General Assembly. This reflects only the author’s views. The Community is not liable for any use that may be made of the information contained therein. In addition, with the electronic microscopy, Figure 35 shows that the fibres were torn during the fracture in the tensile test. Furthermore, window profile material was compared also with optical microscope previous and after tensile test (Figure 36 and Figure 37). It is observed that in both cases, there is good dispersion of the nanocellulose in the resin. In the case of the specimen after the tensile test, a symmetrical fracture across the section it is also observed. Figure 35. Specimen of the scooter platform material after mechanical test with electronic microscope. Figure 36. Initial window profile material with optical microscope Grant Agreement 101058371 – Project ESTELLA Deliverable T3.4 ©ESTELLA - This is the property of ESTELLA Parties: shall not be distributed/reproduced without formal approval of ESTELLA General Assembly. This reflects only the author’s views. The Community is not liable for any use that may be made of the information contained therein. When comparing the stress–strain curves (Figure 41. True stress–strain curve of the reference hemp–resin specimens to Figure 44. Comparison between reference specimens (dark grey), 1 day (light blue), 7 days (dark blue), and 14 days (black) immersed specimens), a clear trend can be observed in the progressive degradation of mechanical properties as the immersion time increases. The reference (dry) specimens exhibit the Figure 43. Comparison between reference specimens (dark grey), 1 day immersed specimens (light blue), and 7 days immersed specimens (dark blue) / Elastic modulus in 7 day specimens. Grant Agreement 101058371 – Project ESTELLA Deliverable T3.4 ©ESTELLA - This is the property of ESTELLA Parties: shall not be distributed/reproduced without formal approval of ESTELLA General Assembly. This reflects only the author’s views. The Community is not liable for any use that may be made of the information contained therein. highest maximum stress values, as well as a steeper initial slope, indicating greater stiffness. After 1 day of immersion, a noticeable reduction in both maximum strength and elastic modulus is already apparent. This trend continues in the 7 days specimens, where the loss of stiffness and strength becomes more pronounced. Additionally, increased strain at failure is observed in the aged specimens, suggesting a reduction in structural integrity and potential plasticisation of the material due to the presence of moisture. The specimens immersed for 14 days also show a decrease in mechanical properties compared to the initial condition. However, the deterioration is not as severe or systematic as might be expected. Some curves even surpass those of the 7 days or 1 day groups in terms of maximum stress, suggesting that the material may reach a degree of stabilisation or saturation in its mechanical behaviour once a certain level of water uptake is reached. The tensile test results for specimens immersed in water for 7 days and 14 days show very similar behaviour, with minimal differences in stiffness and strength. This suggests that the material may have reached a saturation point around one week of exposure, beyond which the mechanical degradation does not significantly progress. The loss in mechanical performance correlates directly with the increase in moisture content over the different immersion periods. The observed decline in both stiffness and strength follows a clear trend with prolonged water exposure, confirming the negative impact of water absorption on the structural integrity of the material. The relationship between moisture uptake and mechanical degradation clearly highlights the high susceptibility of the hemp-reinforced material to humid environments. This behaviour underlines the need to consider protective treatments, coatings, or physical barriers if this material is to be used in outdoor or moisture-exposed applications. Grant Agreement 101058371 – Project ESTELLA Deliverable T3.4 ©ESTELLA - This is the property of ESTELLA Parties: shall not be distributed/reproduced without formal approval of ESTELLA General Assembly. This reflects only the author’s views. The Community is not liable for any use that may be made of the information contained therein. Figure 45 shows the comparison of the average elastic modulus of specimens manufactured with hemp fibre and resin, after being subjected to water immersion for different periods of time. It is observed that the initial value of the elastic modulus is 1,761.52 MPa, which represents the original condition of the specimens before any exposure to water. After one day of immersion, the elastic modulus decreases significantly to 1,049.10 MPa, indicating a considerable loss of stiffness Figure 44. Comparison between reference specimens (dark grey), 1 day (light blue), 7 days (dark blue), and 14 days (black) immersed specimens / Elastic modulus in 14 day specimens. Grant Agreement 101058371 – Project ESTELLA Deliverable T3.4 ©ESTELLA - This is the property of ESTELLA Parties: shall not be distributed/reproduced without formal approval of ESTELLA General Assembly. This reflects only the author’s views. The Community is not liable for any use that may be made of the information contained therein. within the first 24 hours. After seven days, this decrease continues more sharply, reaching a value of 587.74 MPa, which corresponds to approximately one third of the initial value. However, after fourteen days of immersion, a slight recovery in the elastic modulus is observed, increasing to 624.00 MPa, although it remains noticeably lower than the initial value. This behaviour suggests that prolonged exposure to water causes a progressive degradation of the mechanical properties of the material, possibly due to moisture absorption by the hemp fibres and partial deterioration of the resin matrix. The slight recovery at fourteen days could be related to a swelling process or internal rearrangement of the fibres, although the material does not manage to regain its original properties. In summary, hemp fibre and resin specimens show a marked loss of elastic modulus when immersed in water, particularly during the first few days, which is an important factor to consider in applications where this type of material is exposed to humid environments. 1761,52 1049,10 587,74 624,00 INITIAL 1 DAY 7 DAYS 14 DAYS ELASTIC MODULUS [MPA] ID COMPARASION BETWEEN SAMPLES (AVG) Figure 45. Comparison between reference specimens, 1 day, 7 days, and 14 days immersed specimens Grant Agreement 101058371 – Project ESTELLA Deliverable T3.4 ©ESTELLA - This is the property of ESTELLA Parties: shall not be distributed/reproduced without formal approval of ESTELLA General Assembly. This reflects only the author’s views. The Community is not liable for any use that may be made of the information contained therein. 7.2. Prototype 2: Window profile This section presents the results obtained from the tensile tests performed on specimens extracted from flat plates manufactured using the same material as the window profiles (resin reinforced with nanocellulose, Figure 39b). These specimens were subjected to different water immersion periods prior to testing, with the aim of studying the evolution of their mechanical properties under prolonged moisture exposure. Table 11 to Table 13 show the initial and final weights of the specimens after 1 day, 7 days, and 14 days of water immersion. As can be seen, the increase in mass is significantly lower compared to the other material analysed, with absorption values around 0.4% after 1 day, 0.8% after one week, and 1.1% after two weeks. This confirms that the nanocellulosereinforced material exhibits low susceptibility to moisture uptake under these exposure conditions. All specimens are internally traceable to the plate from which they were taken, irrespective of the duration of immersion. For example, several samples from plate 1 were tested initially, after 1 day, 7 days, 14 days of immersion. Table 11. Resin + Nanocellulose. Water absorption after 1 day of immersion Weight (g) Water absorbed (g) Water absorbed (%) Initial 1 day-Water SPECIMEN-1 10.3691 10.4034 0.0343 0.3 SPECIMEN-2 10.0479 10.0854 0.0375 0.4 SPECIMEN-3 10.2024 10.2402 0.0378 0.4 Table 12. Resin + Nanocellulose. Water absorption after 7 days of immersion Weight (g) Water absorbed (g) Water absorbed (%) Initial 7 days-Water SPECIMEN-4 10.1929 10.274 0.0811 0.8 SPECIMEN-5 9.9758 10.0536 0.0778 0.8 SPECIMEN-6 10.1081 10.2001 0.0920 0.9 Table 13. Resin + Nanocellulose. Water absorption after 14 days of immersion Weight (g) Water absorbed (g) Water absorbed (%) Initial 14 days-Water SPECIMEN-7 10.0525 10.1638 0.1113 1.1 SPECIMEN-8 10.2597 10.3801 0.1204 1.2 SPECIMEN-9 10.1239 10.2479 0.1240 1.2 Grant Agreement 101058371 – Project ESTELLA Deliverable T3.4 ©ESTELLA - This is the property of ESTELLA Parties: shall not be distributed/reproduced without formal approval of ESTELLA General Assembly. This reflects only the author’s views. The Community is not liable for any use that may be made of the information contained therein. Figure 46 shows some of the specimens after the tensile test. Figure 47 to Figure 50 present the true stress–strain curves obtained from the tensile tests conducted on the nanocellulose-reinforced specimens. In addition, the corresponding elastic modulus values for each specimen are also provided. Figure 46.Tensile test specimens made of resin and nanocellulose after testing Grant Agreement 101058371 – Project ESTELLA Deliverable T3.4 ©ESTELLA - This is the property of ESTELLA Parties: shall not be distributed/reproduced without formal approval of ESTELLA General Assembly. This reflects only the author’s views. The Community is not liable for any use that may be made of the information contained therein. Figure 47. True stress–strain curve of the reference specimens made of resin + nanocellulose / Elastic modulus in initial specimens Grant Agreement 101058371 – Project ESTELLA Deliverable T3.4 ©ESTELLA - This is the property of ESTELLA Parties: shall not be distributed/reproduced without formal approval of ESTELLA General Assembly. This reflects only the author’s views. The Community is not liable for any use that may be made of the information contained therein. Figure 48. Comparison of the true stress–strain curves of the reference specimens (dark grey) and the specimens immersed in water for 1 day (light blue) / Elastic modulus in 1 day specimens Grant Agreement 101058371 – Project ESTELLA Deliverable T3.4 ©ESTELLA - This is the property of ESTELLA Parties: shall not be distributed/reproduced without formal approval of ESTELLA General Assembly. This reflects only the author’s views. The Community is not liable for any use that may be made of the information contained therein. Figure 49. Comparison of the true stress–strain curves of the reference specimens (dark grey) and the specimens immersed in water for 1 day (light blue) and 7 days (dark blue) / Elastic modulus in 7 days specimens Grant Agreement 101058371 – Project ESTELLA Deliverable T3.4 ©ESTELLA - This is the property of ESTELLA Parties: shall not be distributed/reproduced without formal approval of ESTELLA General Assembly. This reflects only the author’s views. The Community is not liable for any use that may be made of the information contained therein. Figure 50. Comparison of the true stress–strain curves of the reference specimens and the specimens immersed in water for 1 day, 7 days, and 14 days / Elastic modulus in 14 days specimens