Comparative characterization of FDM structures with electrically-conductive sensing elements under static, dynamic and thermal loads

Comparative characterization of FDM structures with electricallyconductive sensing elements under static, dynamic and thermal loads Ferdinando Ursi & Giorgio De Pasquale Fused filament fabrication (FFF), or fused deposition modeling (FDM), is one of the most widely accessible additive manufacturing (AM) processes. Recent advancements in this technology have expanded its material portfolio to include conductive composites with electromechanical properties, enabling new applications. The thermal melting of the filament, required for material extrusion, introduces variability in the final component properties, which are difficult to predict due to the influence of several process-related parameters. In particular, for applications where mechanical and electrical properties are critical, it is essential to optimize the process to control both the mechanical performance and electrical conductivity of the material in static and dynamic conditions. Postprocess thermal treatments can significantly alter these electromechanical transduction properties. In this study, we investigate the static, dynamic, and thermal behavior of two composite filaments. The microstructure of the feedstock materials was analyzed using scanning electron microscopy (SEM) to establish a correlation between material composition and component behavior. The results demonstrate that the inclusion of specific fillers, such as black carbon, enhances electrical resistance and improves electromechanical stability under static and dynamic conditions. In contrast, graphene additives increase electromechanical sensitivity but result in a degradation of electrical properties during thermal treatment. Keywords Conductive filament, FDM, FFF, Additive manufacturing, Sensors, IoT Fused filament fabrication (FFF), originally called fused deposition modelling (FDM) gained widespread popularity among additive manufacturing (AM) processes due to its accessibility and versatility. In this process, a thermoplastic filament is heated up to the melting point and extruded through a nozzle to build, layer by layer, the three-dimensional object. The main advantages of FDM are the ability to produce complex geometries impossible to achieve with subtractive processes and the availability of machines relatively affordable, easy to use and with reduced production costs compared to other AM processes. Materials commonly used in FDM are polylactic acid (PLA), acrylonitrile butadiene styrene (ABS), and polyethylene terephthalate glycol (PETG), each offering different mechanical properties and suitability for various applications. PLA is biodegradable and relatively easy to print, ABS is more durable and heat-resistant, making it ideal for functional parts and end-use products, PETG combines the ease of printing of PLA with the strength of ABS, offering a balance of flexibility and durability1,2. The layer-by-layer building strategy results in anisotropic properties, meaning the strength of the printed object varies depending on the load direction. Post-processing techniques such as sanding, acetone smoothing, or painting can improve the surface finish and mechanical properties of the printed parts. Despite some limitations, such as slower production speed and lower resolution compared to other AM methods like stereolithography (SLA) and selective laser sintering (SLS), FDM achieved a consolidated level in several industrial sectors involving fast prototyping and small-medium scale production based on cost-effectiveness and ease of use1,3. Among these sectors, there are automotive, aerospace, and healthcare. Moreover, the beneficial environmental footprint of FDM materials is an area of active research, because this process can operate on sustainable, recyclable and lowimpact options. The advancement of technology is opening to new needing and challenges, then improving print quality, speed, and material properties are being addressed, paving the way for future innovations4. Smart Structures and Systems Lab, Politecnico Di Torino, Corso Duca Degli Abruzzi 24, 10129 Torino, Italy. email: [email protected] OPEN Scientific Reports | (2025) 15:26877 1 | https://doi.org/10.1038/s41598-025-11234-0 www.nature.com/scientificreports The advancement of the FDM process also lead to the integration of new filler materials in the process resulting in polymer composites. Conductive polymer composite (CPC)5 is a class of material in which the filler materials exhibit electrical conductive properties6. Filler materials employed in CPC are mainly classified as metal7 or carbon-based8. Anyway, due to problems related to abrasion and clogging affecting the metal-based, carbon-based CPC are the most recommended when aiming to the production of smart and functionalized components6. For example, long fiber filaments are involved in the FDM process as reinforcement material in the so called CFR-FDM (composite fiber reinforcement—FDM) process9. When using conductive materials such as long carbon fibers, in addition to improved mechanical properties, the final components will also exhibit enhanced electrical properties. However, long fibers suffer damages during the deposition process itself due to the inherent stresses and mechanical forces involved10. This presents a significant challenge in the use of advanced composite materials for 3D printing, necessitating appropriate printing techniques and parameters to minimize such issues. However, nanomaterials are the most widespread fillers employed to provide the feedstock with electrically conductive properties. Among these, black carbon (CB), carbon nanotubes (CNTs), and graphene particles (GP) are the most promising ones11,12. The physical phenomenon that allows the electrical conduction through the nanoparticles composite feedstoks is the percolation theory that describes the electrical conduction in the macroscale as result of conduction mechanisms between filler particles. The percolation threshold is the minimum amount of conductive filler to introduce in the polymer matrix to make it able to exhibit conductive properties12,13. As reported by Pejak Simunec et al.14 the conduction is due to electron tunnelling that consists of electrons “jump” from one filler to the next one across distances through the polymer matrix and/or direct contact of conductive fillers, meaning that electrons transfer happens via conductive filler pathway. However, other recent studies proved that CPCs can find large application in the manufacturing of smart structures. Several studies demonstrated that such materials are strain-sensitive, thus suitable as strain or pressure sensors15–18, while others studies highlighted temperature sensitivity and thus potential application as thermometers19,20. Maurizi et al. have proposed the application of sensors manufactured via FDM for the dynamic measurements21 while other studies have investigated the application of these composites in the manufacturing of electrochemical sensors22,23. The combination of AM processes and conductive materials has already demonstrated its potential for producing functional, end-use devices. For example, Hook et al. developed a frequency selective surface (FSS) using a conductive material printed via FDM24, Podsiadły et al. demonstrated the feasibility of manufacturing human interface device (HID)25 and Jayanth et al. confirmed the suitability of ABS-BC composites for electromagnetic shielding applications26. Flowers et al. proposed 3D printed inductors27, while Hampel et al. manufactured a capacitor with a commercial electrically conductive filament and developed a model of the electrical resistance of 3D printed structures manufactured with FDM28. As the FDM process is sensitive to the printing parameters, the use of functional materials exacerbates this criticality as reported by different studies29–31. Other cons highlighted by Yu et al. include the shift of the electrical properties when cyclical quasi-statics loads are applied32. Several researchers have explored the use of customized filaments to enhance electrical performance. However, this approach makes it more challenging to replicate their experiments and demands greater resources to initiate the process. In a recent study, Aloqalaa reported the current status about the commercially available filament identifying up to twelve different electrically conductive filaments34. Many of the aforementioned studies highlight the dependence of the electrical conductivity of CPCs on a single factor (e.g., mechanical stress or temperature) or propose end-use applications while neglecting aspects related to real environmental conditions, making them applicable only under specific circumstances. In contrast to existing studies, the present work adopts a targeted approach to assess the feasibility of smart components equipped with embedded strain sensors. As reported in the literature, temperature is an environmental factor that can influence the performance of strain gauges, along with vibrations33. The innovative contribution of the present study consists in a comparative analysis of two commercial CPC for sensing applications. The proposed investigation involves the static and dynamic characterization of the electro-mechanical and electro-thermal behavior of two CPCs, followed by a Scanning Electron Microscopy (SEM) analysis of the tested materials. Similar experimental approaches have been previously applied by the authors in earlier studies35–39. Finally, conclusions and remarks are provided regarding the observed behavior of the two materials, evaluating their potential for use in the manufacturing of smart components. Description of samples This section describes the samples design, the manufacturing parameters setup and the embedding of electrodes. The samples are specifically fabricated for three test typologies: tensile tests, thermal tests and dynamic tests. All the samples are composed of a base material, with electrically insulating properties, and of a conductive material that is deposited in shape of wires or meander. In total, 18 samples for tensile tests, 6 samples for thermal tests and 2 samples for dynamic tests were fabricated. Samples geometry For the tensile and thermal tests, the same sample shape is employed, to use the same printing path and reduce the variability of the process. The geometry of the base material is in agreement with the tensile test standard shape ASTM D638-22 Type V40, referred to experimental characterization of polymers. Shape and nominal dimensions used for the samples fabrication, through CAD and STL file formats, are reported in Fig.1a. The conductive material is deposited on the top surface of the sample with a wire shape in variable widths of 0.4, 0.8, and 1.6mm. The width of each extruded filament is 0.4mm, than the three conductive wire variants are obtained with one, two and four parallel depositions respectively. The thickness of each layer is imposed at Scientific Reports | (2025) 15:26877 2 | https://doi.org/10.1038/s41598-025-11234-0 www.nature.com/scientificreports/ 0.2mm for both insulating and conductive materials. At the ends of the conductive path, two connection pads are also fabricated to allow the fastening of copper electrodes used for measurements in the experimental setup. The shape of one conductive stripe (0.8mm) is represented in Fig.1b as an example. The samples for dynamic tests consist in a beam with dimensions 125 × 12x3 mm3, with a conductive meander printed on the top surface with geometry and dimensions reported in Fig.1c. This regular shape is used to simplify the modelling of the beam response. The conductive meander amplifies the electric output signal and reduces the noise respect to the single wire, which allows dynamic measurements. Similarly to the previous conductors, connection pads are fabricated at the ends of the meander to fasten the electrodes. Materials and fabrication process The samples are fabricated with the UltiMaker S5 multi-material system, equipped with the dual-extruder device. The material used for the structural part of the samples is the PLA Pro1-white, provided by BASF. The matrix polymer selected for the study is the thermoplastic polyurethane (TPU), a polymer with considerable elongation at break, able to accommodate the structure during the deformation without occurring in fracture. Fig. 1. Specimen shape and dimensions in agreement with ASTM D638-22 Type V for polymers tensile test (a), conductive deposited filament (0.8mm width) and connection pads (b), conductive meander for dynamic tests samples (c). Scientific Reports | (2025) 15:26877 3 | https://doi.org/10.1038/s41598-025-11234-0 www.nature.com/scientificreports/ The two CPC considered for the experimental characterization are “Fili”, provided by AIMPLAS Technological Institute of Plastics, and the other one is called “Filaflex Conductive”, provided by RECREUS. According to the information provided by the manufacturers, the “Fili” material, using graphite particles as conductive filler, has 27.44 Ω·cm electrical resistivity. The “Filaflex Conductive” material, including carbon black conductive filler, has 3.9 Ω·cm resistivity41. The filler concentration as volume percentage is defined by the producer by considering several constraints, as the shape and typology of the same filler, and the expected physical properties of the filament. In particular, higher concentrations led to higher conductivity but also higher viscosity and brittleness, causing extrusion and homogeneity issues of the filament. The printing of the samples is preceded by the optimization of the printing parameters. Due to the presence of fillers, the filaments viscosity and their propensity to the extrusion are sensitively different from their filler-free versions. For this reason, compared to the standard parameters recommended in the material datasheet, further improvements of the printability are achieved by reducing the printing speed and by increasing the flow rate. The Table 1 summarizes the most relevant parameter resulting from the optimization of the printing process. The fabrication process is driven by the slicer software UltiMaker Cura Enterprise (v. 5.6.0), that is used to perform the optimization process and to manage the printing. The selected layer height is 0.2mm to achieve good dimensional resolution and reasonable printing time. Through the multi-material functions of the software, it is possible to assign different materials to specific geometries imported with separate STL files. By using the merging functions, the dual-extrusion process is optimized to make a continuous transition at the borders. Moreover, through the programming of the G-CODE embedded into the slicer software, it is possible to interrupt the deposition process at a specific layer for a given time and restart the process after the manual operations on samples are completed. Electrodes coupling The physical connection between the conductive material and the instrumentation must provide good adhesion and low concentrated resistance to prevent measurement errors and achieve stable electric output. The electrical resistance of the deposited filaments is measured in the following tests. In other experiments11,19,21, this connection is obtained by means of conductive inks applied between the conductive material under investigation and external electrodes. However, the conductive ink shows generally low mechanical reliability and properties variation over time. Then, in this study, two copper plates are embedded into the sample during the growth process and then used as electrodes for connecting the instrumentation. The growth process is interrupted after the deposition of the first conductive layer on the top surface of the base material. At this point, the copper plates are positioned on the corresponding contacting pads at the ends of the conductor. By using a jig and tape, the copper plates are secured in that position and the printing process is completed with two additional layers of conductive material. The jig and tape protect the copper plates against undesired movements induced, for example, by the nozzle motion. The sample with embedded electrode is represented in the drawing of Fig.2a and the fabrication steps are reported in Fig.2b. Before the embedding process, the edges of the copper plates are shaped in a saw-teeth profile and folded. This shape of the plate edges provides a stronger anchoring with the polymer and an interface more stable mechanically. After the samples fabrication, the electro-mechanical properties of the electrodes connection are validated. The mechanical strength is evaluated manually by ensuring that relative movement between the copper plate and the polymer is present. The electric resistance of each connection is measured, obtaining for all the samples values below 0.1 Ω. Experimental tests This section describes the three tests setup for tensile, dynamic and thermal characterization, whose adopted experimental setup are reported in Fig.3a–c, respectively. Each test has specific goal in relation to the validation of the electrical characteristics of samples. In particular, the correlations between the stress–strain field and the electric output is investigated in view of sensing applications of conductive materials. The effect of heat treatments after the printing and their influence of the mentioned mechanical-electric response is also investigated. Finally, the ability to measure the dynamic response of one structure through the electric output of conductive materials is tested. Tensile tests The samples with integrated conductive wires are subjected to axial tensile force; the force and strain instant values are measured together with the instant values of the electric resistance of the conductive material. The goal of the tensile tests is to evaluate the behaviour of the conductive materials in the elastic field. Three samples were fabricated and tested per each conductive material (two variants) and wire width (three variants). Printing parameter Standard recommended Fili (AIMPLAS) Filaflex conductive (RECREUS) Printing temperature [°C] 225 265 230 Flow rate [%] 100 120 110 Printing speed [mm/s] 25 10 6 Line width [mm] 0.4 0.3 0.3 Table 1. Result of the printing parameter optimization of the conductive filaments, in comparison with standard recommended values for the filler-free filament. Scientific Reports | (2025) 15:26877 4 | https://doi.org/10.1038/s41598-025-11234-0 www.nature.com/scientificreports/ The tensile tests are performed with the INSTRON 8801 universal testing machine with displacement control mode at 1mm/min velocity. The applied force varies in the range 0–400 N. The effective strain rate on the specimen is measured with an uniaxial dynamic extensometer (INSTRON 2620–604) referred to 22.5mm gauge length. The testing machine is controlled with the software Bluehill. The acquisition of the data generated by the testing machine, the extensometer and the conductive material is provided with a 4-channels DAQ system (National Instruments CompactRIO-9054 and NI-9219 module). The experimental setup is reported in Fig.3a. The configuration of the DAQ system is developed in LabVIEW environment and allows the recording of the load, the strain returned by the extensometer and the electrical resistance of the conductive wire over the time domain, with 100Hz sampling frequency. The post-processing of the recorded data is performed with MATLAB. The voltage divider is applied to the electrical resistance signal measured on the conductive wire, to respect the measurement range of the NI-9219 module. Prior the tensile test, the volume resistance of the specimen is recorded with a Fluke 114 multimeter and then the resistivity is estimated as: ρ =R0 · A L (1) where ρ [Ω cm] is the electrical resistivity, R0 [Ω] is the electrical resistance measured at the start of the test, A [cm2] and L [cm] represent the cross-section and the length of the conductive track, respectively. The results are presented graphically, and their suitability for application as strain gauges is evaluated using the gauge factor and the coefficient of determination. The gauge factor is defined as: GF = δR / R0 ε (2) where GF [dimensionless] is the gauge factor, ε [dimensionless] is the mechanical strain, δR [kΩ] represents the change in electrical resistance between the initial resistance and the resistance at a given ε . The coefficient of determination is defined as: R 2=1− ∑ n i=1 ( Yi− ˆ Yi )2 ∑ n i=1 ( Yi − Y ) 2 (3) where R2 [dimensionless] is the coefficient of determination, Yi [dimensionless] represents the i-th of n observed values of the analysed phenomenon,  Yi [dimensionless] is the predicted value of the i-th observation provided by the estimation model, Y [dimensionless] is the mean of all observed values. Fig. 2. Sample layer composition after the embedding of the conductive layer (a). Creation of the interface between the conductive material and copper electrodes (b): the printing process is interrupted at a certain layer (1), the copper plates are applied (2), securing of the electrodes with a jig and tapes (3), final shape of the sample, after the printing re-start, with copper plates embedded (4). Scientific Reports | (2025) 15:26877 5 | https://doi.org/10.1038/s41598-025-11234-0 www.nature.com/scientificreports/ Thermal tests The primary goal of the thermal tests is to determine if the value of electrical resistance of the embedded conductive wire is sensitive to the environmental temperature. Another goal is to identify eventual beneficial effects of heat treatments applied to the conductor, leading to an increase of the resistance-strain sensitivity for transducers applications. The tests consider the temperature variation on the component after the filament deposition, while the filament heating during the extrusion is considered as a constraint of the optimized process setup. Fig. 3. Experimental setup: tensile test (a), dynamic test (b) and thermal test (c). Scientific Reports | (2025) 15:26877 6 | https://doi.org/10.1038/s41598-025-11234-0 www.nature.com/scientificreports/ The sample with embedded conductive wire is heated by thermal conduction through direct contact with a hot surface for long time (five hours followed by one hour for cooling down). The instant electric resistance is measured at regular time intervals during the heating process, with sampling frequency of 100Hz. The setup of the thermal tests is reported in Fig.3c. The values of electrical resistance and temperature are measure with the same DAQ system described before. The same voltage divider circuit is also used to measure the electrical resistance of the conductive wire. One sample per combination of conductive material (two variants) and wire width (three variants) was measured. A type-K thermocouple, calibrated with Fluke 123 equipped with the 80TK thermocouple module, is the sensor used to measure the instant temperature of the hot plate. The temperature range imposed to the hot plate is 0–140°C, then the maximum temperature is kept constant for five hours. After this time, the heating system is turned off and one hour time is waited for the cooling down of the sample, by continuing the measurements. Dynamic tests In the dynamic tests, a beam sample with integrated conductive material is excited in the frequency domain and its vibrational response is measured. The vibration measurements obtained with a laser and with the conductive material used as a strain sensor are compared, with the goal to understand their agreement. The ultimate goal is to demonstrate the feasibility of dynamic measurements on FDM components, in terms of response amplitude and frequency, with an embedded sensor made of conductive material. The experimental setup, represented in Fig. 3b, consists of a mechanical shaker (TV 51,120 TIRAvib) connected to a power amplifier (BAA 500 TIRAvib). The sample is a cantilever beam, with a conductive pattern on the top, connected to one end to the shaker. An ICP accelerometer (353B34 PCB) is fastened to the shaker stage to measure the excitation. The dynamic response of the free end of the sample is measured with a laser doppler vibrometer (VibroONE VIO130 Polytec). The ADC/DAC (analog-to-digital / digital-to-analog controller) cRIO-9054 (National Instruments) is used to manage all the tests data. The sinusoidal signal used to actuate the shaker, through the amplifier, is generated with the NI-9263 module. The accelerometer output signal and the electric resistance of the conductive meander are measured with the NI-9234 module with a sampling frequency of 1652Hz. This module is compatible with IEPE accelerometers, then no further instrumentation is needed. The same voltage divider circuit already described before is used to measure the conductor resistance. The ADC/DAC is managed by a control interface self-developed in LabVIEW environment. The excitation is represented by a sequence of sinusoidal signals with a frequency increasing of a given step. The interface allows the user to set the frequency range and the step of frequency increment; the excitation voltage generated by the DAC is also imposed through the interface. When a new frequency is applied, the measurements starts with some time delay to wait the extinction of the transitory. The signals generated by the accelerometer (excitation), by the laser (tip response) and by the conductive meander are measured at each frequency and stored. Before to start the sample characterization, the response is preliminary measured across a wide frequency range to identify the resonances, using DAC excitation voltages of 0.5, 1 and 1.5V in the range 10–100Hz, with step of 1Hz. Then, the test is repeated on a narrower frequency range centred on the first resonance with step of 0.2Hz and the same amplitudes. The post-processing of the measured data is operated with MATLAB. The results are represented by means of two frequency response functions (FRFs), or gain functions, in the frequency domain, showing the structural response and the electro-mechanical response given by the conductive meander. The FRF of the beam structure is:   G(f)struct   = ¨ Ytip ¨ Xbase (4) where. G(f)struct [adim] is the FRF of the structure (i.e. cantilever beam) ,¨ Ytip [m/s2] is the acceleration amplitude of the system response measured at the free tip of the beam with the laser vibrometer, and ¨ Xbase [m/s2] is the acceleration amplitude of the excitation measured on the shaker stage with the ICP accelerometer. The FRF of the electro-mechanical system composed by the beam with coupled conductive meander is:   G(f)electro − mech   = R(f) meander ¨ Xbase (5) where. G(f)electro−mech [kΩ/ m/s2] is the FRF of the electro-mechanical system composed by the beam and the embedded conductive meander [kΩ/ m/s2] and Rmeander [kΩ] is the electrical resistance amplitude at given frequency f . Results This Section reports the results of the experimental tensile, thermal and dynamic tests on the samples described before. Tensile tests results All tensile tests were performed at ambient temperature on samples produced directly through the FDM process, without any additional heat treatment. Three variants of conductive wires, with widths of 0.4mm, 0.8mm, and Scientific Reports | (2025) 15:26877 7 | https://doi.org/10.1038/s41598-025-11234-0 www.nature.com/scientificreports/ 1.6mm, were tested with the equipment previously described at 1mm/min strain rate. Three samples have been tested for each wire width variant. Table 2 reports the values of electrical resistance and resistivity measured at variable applied strain. The answer to the discrepancy with the supplier’s data can be found in the model adopted to evaluate the resistivity reported in Eq. 1. This model doesn’t consider additional conduction mechanisms such as electrical contact between adjacent nanoparticles or tunnelling effect33. The tensile test results for Fili—AIMPLAS and Filaflex Conductive—RECREUS materials are reported in Figs. 4 and 5, respectively. For both materials, the load-strain curves are shown in Figs. 4a and 5a, while the normalized electrical resistance variation (δR/R₀) under applied strain for different widths (0.4, 0.8, and 1.6mm) is presented in Figs. 4b–d and 5b–d, respectively. The signals were acquired with a sampling frequency of 100Hz. Over the tensile test, both the strain, the load and the resistance signals are affected by noise over the time Fig. 4. Tensile test results on Fili – AIMPLAS material: load-strain curve (a), normalized electrical resistance variation (δR/R₀) under applied strain for conductive paths of varying widths: 0.4mm (b), 0.8mm (c), and 1.6mm (d). Wire width [mm] Initial electrical resistance (R0) [kΩ] Initial electrical resistivity ( ρ0 ) [Ω • cm] Fili – AIMPLAS Filaflex conductive— RECREUS Fili—AIMPLAS Filaflex conductive—RECREUS Mean value St. Dev Mean value St. Dev Mean value Mean value 0.4 31.50 1.92 327.69 45.61 13.75 142.99 0.8 17.26 2.32 170.56 33.41 15.06 148.85 1.6 10.86 2.09 89.54 19.56 18.96 156.29 Table 2. Average values of the electrical resistances (R0) and resistivity ( ρ0 ) over the applied strain range (0–0.012) for the two materials with different widths of the conductor. Scientific Reports | (2025) 15:26877 8 | https://doi.org/10.1038/s41598-025-11234-0 www.nature.com/scientificreports/ domain. The reason why Figs. 4 and 5 for a certain resistance correspond multiple values for the strains is due to the fact that the variability introduced by the noise affecting the strain is larger than the variability exhibited by the resistance because the more the values of R₀ is larger, the more the effect of noise on the value of δR/R₀ is irrelevant. Table 3 reports the gauge factor (GF) and the coefficient of determination (R2) for the electrical resistance in wires with different widths. The range of the reported values for Fili–AIMPLAS and their low variability are in agreement with the literature33. On the other hand, in some samples of Filaflex Conductive—RECREUS, since the behaviour of the material is variable, the linear regression does not represent faithfully the measured response. This is testified by the values of the coefficient of determination (up to 0.717). The poor quality of the linear regression, together with low gauge factor suggests that the CB is not reliable when meant for strain sensing applications. Wire width [mm] Gauge factor (GF) [adim] Coefficient of determination (R2) [adim] Fili—AIMPLAS Filaflex conductive— RECREUS Fili—AIMPLAS Filaflex conductive— RECREUS Mean value St. Dev Mean value St. Dev Mean value St. Dev Mean value St. Dev 0.4 117.03 25.46 6.09 5.20 0.960 0.016 0.717 0.266 0.8 121.97 17.08 4.33 2.97 0.956 0.018 0.686 0.533 1.6 129.59 15.73 2.02 4.96 0.970 0.006 0.161 0.081 Table 3. Gauge Factor (GF) and Coefficient of Determination (R2) for electrical resistance vs. applied strain in conductive wires of different widths. Fig. 5. Tensile test results on Filaflex Conductive—RECREUS material: load-strain curve (a), normalized electrical resistance variation (δR/R₀) under applied strain for conductive paths of varying widths: 0.4mm (b), 0.8mm (c), and 1.6mm (d). Scientific Reports | (2025) 15:26877 9 | https://doi.org/10.1038/s41598-025-11234-0 www.nature.com/scientificreports/ 11. Horst, D. J., de Andrade Junior, P. P., Duvoisin, C. A. & Vieira, R. D. Fabrication of conductive filaments for 3d-printing: Polymer nanocomposites. Biointerface Res. Appl. Chem. 10, 6577–6586 (2020). 12. Zhou, X., Wu, L. & Wang, J. Recent developments in conductive polymer composites for fused deposition modeling. Compos. Part A Appl. Sci. Manuf. 174, 107739 (2023). 13. Bunde, A. & Dieterich, W. Percolation in composites. J. Electroceram. 5, 81–92 (2000). 14. Pejak Simunec, D. & Sola, A. Emerging research in conductive materials for fused filament fabrication: A critical review. Adv. Eng. Mater. 24, 2101476 (2022). 15. Ren, H. et al. Smart structures with embedded flexible sensors fabricated by fused deposition modeling-based multimaterial 3D printing. Int. J. Smart Nano Mater. 2022, 447–464 (2022). 16. Musenich, L. et al. Anisotropic mechanical and sensing properties of carbon black-polylactic acid nanocomposites produced by fused filament fabrication. Smart Mater. Struct. 33, 095010 (2024). 17. Kouvatsos, T., Pagonis, D. N., Iakovidis, I. & Kaltsas, G. Towards a 3D printed strain sensor employing additive manufacturing technology for the marine industry. Appl. Sci. 14, 6490 (2024). 18. Gackowski, B. M., Goh, G. D., Sharma, M. & Idapalapati, S. Additive manufacturing of nylon composites with embedded multimaterial piezoresistive strain sensors for structural health monitoring. Compos. Part B Eng. h t t p s : / / d o i . o r g / 1 0 . 1 0 1 6 / j . c o m p o s i t e s b . 2 0 2 3 . 1 1 0 7 9 6 (2023). 19. Kwok, S. W. et al. Electrically conductive filament for 3D-printed circuits and sensors. Appl Mater Today 9, 167–175 (2017). 20. Marasso, S. L. et al. PLA conductive filament for 3D printed smart sensing applications. Rapid Prototyp J 24, 739–743 (2018). 21. Maurizi, M. et al. Dynamic measurements using FDM 3D-printed embedded strain sensors. Sensors h t t p s : / / d o i . o r g / 1 0 . 3 3 9 0 / s 1 9 1 2 2 6 6 1 (2019). 22. Omar, M. H., Razak, K. A., Ab Wahab, M. N. & Hamzah, H. H. Recent progress of conductive 3D-printed electrodes based upon polymers/carbon nanomaterials using a fused deposition modelling (FDM) method as emerging electrochemical sensing devices. RSC Adv 11, 16557–16571 (2021). 23. João, A. F. et al. 3D printing pen using conductive filaments to fabricate affordable electrochemical sensors for trace metal monitoring. J. Electroanal. Chem. 876, 114701 (2020). 24. Hook, A., Motes, D. T., Morrow, C. & Donnell, K. M. Additively Manufactured Aperture-based FSS. Conference Record–IEEE Instrumentation and Measurement Technology Conference (2024) https://doi.org/10.1109/I2MTC60896.2024.10560971 25. Podsiadły, B., Skalski, A., Wałpuski, B. & Słoma, M. Heterophase materials for fused filament fabrication of structural electronics. J. Mater. Sci.: Mater. Electron. 30, 1236–1245 (2019). 26. Jayanth, N., Senthil, P. & Mallikarjuna, B. Experimental investigation on the application of FDM 3D printed conductive ABS-CB composite in EMI shielding. Radiat. Phys. Chem. 198, 110263 (2022). 27. Flowers, P. F., Reyes, C., Ye, S., Kim, M. J. & Wiley, B. J. 3D printing electronic components and circuits with conductive thermoplastic filament. Addit. Manuf. 18, 156–163 (2017). 28. Hampel, B., Monshausen, S. & Schilling, M. Properties and applications of electrically conductive thermoplastics for additive manufacturing of sensors. Tech. Mess. 84, 593–599 (2017). 29. Dembek, K., Podsiadły, B. & Słoma, M. Influence of process parameters on the resistivity of 3D printed electrically conductive structures. Micromachines 13, 1203 (2022). 30. Palmić, T. B., Slavič, J. & Boltežar, M. Process parameters for FFF 3D-printed conductors for applications in sensors. Sensors 20, 4542 (2020). 31. Stankevich, S. et al. Electrical resistivity of 3D-printed polymer elements. Polymers 15, 2988 (2023). 32. Yu, G. et al. Thermoplastic elastomer composite filaments for strain sensing applications extruded with a fused deposition modelling 3D printer. Flex. Printed Electron. 5, 035002 (2020). 33. del Bosque, A., Sánchez-Romate, X. F., Sánchez, M. & Ureña, A. Toward flexible piezoresistive strain sensors based on polymer nanocomposites: A review on fundamentals, performance, and applications. Nanotechnology 35, 292003 (2024). 34. Aloqalaa, Z. Electrically conductive fused deposition modeling filaments: Current status and medical applications. Crystals 12, 1055 (2022). 35. Coluccia, A. & De Pasquale, G. Strain-based method for fatigue failure prediction of additively manufactured lattice structures. Sci. Rep. 13, 1–16 (2023). 36. De Pasquale, G. & Coluccia, A. Modeling and experimental validation of CFRP–metal joints utilizing 3D additively manufactured anchors. J. Manuf. Sci. Eng. 145(11), 111004 (2023). 37. De Pasquale, G., Enes Altunok, F. & Ursi, F. Investigation of the mechanical strength of CFRP with co-cured additively manufactured metal inserts. Proce. Struct. Integr. 47, 573–578 (2023). 38. De Pasquale, G. & Sibona, S. Hybrid materials based on polymers-filled AM steel lattices with energy absorption capabilities. Mech. Adv. Mater. Struct. 29, 2570–2580 (2022). 39. De Pasquale, G. & Ursi, F. Multi-material fittings with carbon fiber reinforcement from AM process: design and characterization. In 3rd Int. Symp. on Industrial Engineering and Automation (ISIEA) (Bolzano (Italy), 2024). 40. D638 standard test method for tensile properties of plastics. https://www.astm.org/standards/d638. 41. Nowka, M., Katja Ruge, L. S., Hilbig, K. & Thomas, V. Characterization of the anisotropic electrical properties of additively manufactured structures made from electrically conductive composites by material extrusion. Polymers h t t p s : / / d o i . o r g / 1 0 . 3 3 9 0 / p o l y m 1 6 2 0 2 8 9 1 (2024). 42. Alessandro Ragolia, M. et al. Thermal characterization of electrical resistance of 3D printed sensors. h t t p s : / / d o i . o r g / 1 0 . 1 1 0 9 / I 2 M T C 5 0 3 6 4 . 2 0 2 1 . 9 4 5 9 9 6 8 . 43. Fernández Sánchez-Romate, X. X., del Bosque, G. A., Sánchez, M. & Ureña, A. Electrical transport mechanisms in graphene nanoplatelet doped polydimethylsiloxane and application to ultrasensitive temperature sensors. ACS Appl. Mater. Interfaces 15(18), 22377–22394 (2023). 44. Chu, K. et al. Smart conducting polymer composites having zero temperature coefficient of resistance. Nanoscale 7, 471–478 (2014). 45. Solangi, N. H., Karri, R. R., Mubarak, N. M. & Mazari, S. A. Mechanism of polymer composite-based nanomaterial for biomedical applications. Adv. Ind. Eng. Polym. Res. 7, 1–19 (2024). 46. Dhinesh, S. K. et al. Recent trends in additive manufacturing of electronics devices. Mater. Today Proc. 66, 928–941 (2022). Author contributions Ferdinando Ursi: data acquisition, experimental tests, data interpretation, writing. Giorgio De Pasquale: conceptualization, design of the work, data interpretation, work financement, writing and revising. Funding This paper was funded by the Horizon Europe grant “HORIZON-CL4-2022-RESILIENCE-01”, Project MIMOSA (Project 101091826). Scientific Reports | (2025) 15:26877 16 | https://doi.org/10.1038/s41598-025-11234-0 www.nature.com/scientificreports/ Declarations Competing interests Authors declare they do not have any competing financial and/or non-financial interests in relation to this work. Additional information Correspondence and requests for materials should be addressed to G.P. Reprints and permissions information is available at www.nature.com/reprints. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit h t t p : / / c r e a t i v e c o m m o n s . o r g / l i c e n s e s / b y - n c - n d / 4 . 0 / . © The Author(s) 2025 Scientific Reports | (2025) 15:26877 17 | https://doi.org/10.1038/s41598-025-11234-0 www.nature.com/scientificreports/