Acrylonitrile Butadiene Styrene-Based Composites with Permalloy with Tailored Magnetic Response

This work reports on tailoring the magnetic properties of acrylonitrile butadiene styrene (ABS)-based composites for their application in magnetoactive systems, such as magnetic sensors and actuators. The magnetic properties of the composites are provided by the inclusion of varying permalloy (Py—Ni75Fe20Mo5) nanoparticle content within the ABS matrix. Composites with Py nanoparticle content up to 80 wt% were prepared and their morphological, mechanical, thermal, dielectric and magnetic properties were evaluated. It was found that ABS shows the capability to include high loads of the filler without negatively influencing its thermal and mechanical properties. In fact, the thermal properties of the ABS matrix are basically unaltered with the inclusion of the Py nanoparticles, with the glass transition temperatures of pristine ABS and its composites remaining around 105 °C. The mechanical properties of the composites depend on filler content, with the Young’s modulus ranging from 1.16 GPa for the pristine ABS up to 1.98 GPa for the sample with 60 wt% filler content. Regarding the magnetic properties, the saturation magnetization of the composites increased linearly with increasing Py content up to a value of 50.9 emu/g for the samples with 80 wt% of Py content. A numerical model has been developed to support the findings about the magnetic behavior of the NP within the ABS. Overall, the slight improvement in the mechanical properties and the magnetic properties provides the ABS composites new possibilities for applications in magnetoactive systems, including magnetic sensors, actuators and magnetic field shielding.


Introduction
In the last few decades, focus on magnetic composites based on a soft polymer matrix with embedded magnetic fillers, including hard and soft ferromagnetic particles [1,2], superparamagnetic particles [3] and particles with permanent magnetization [4,5], has increased due to their technological relevance [6]. Magnetically responsive polymer composites are relevant for applications in next-generation printable and moldable functional devices [7], thanks to their capability to combine the magnetic responsiveness of the fillers with the mechanical properties of the polymer matrix, leading to applications in areas such as magnetic field sensors, energy harvesting systems, motors, inductors, tunable transformers, memory devices and resonators, among others [6,8]. Improving the functional response and processability demands appropriate relationships between functional magnetic fillers In this scope, one interesting magnetic particle that has been scarcely addressed in the context of ABS-based composites is NiFe-based permalloy (Py-Ni x Fe 100-x ). Py, with typically about 80% nickel (x = 80) and 17-20% iron content, is characterized by a very high magnetic permeability and saturation magnetization, low coercivity, near-to-zero magnetostriction and significant anisotropic magnetoresistance [40,41]. The relative permeability of commercial Py is around 1·10 5 , which is almost two orders of magnitude higher than steel [42]. These properties make Py useful as a magnetic core material in electrical and electronic equipment [43,44] and in magnetic shielding to block magnetic fields [45,46], as well as in a variety of magnetic devices [47], including sensors, actuators [48,49] and magnetic recording heads [50,51], among others. One drawback of Py is that it is not very ductile and, therefore, one of the main advantages of its polymer composites is that they allow for the improvement of processability and implementation in specific applications. Magnetic particle polymer composites open a wide window of opportunities in printed electronics [52] and for the development of magnetic devices to implement in flexible substrates [53,54].
The aim of the present work is (i) to provide experimental and numerical tools to better understand the dispersion and the magnetic interactions of the Py-NPs within the ABS, which is essential for the application of the composites as magnetic devices; and (ii) to analyze the effect of nanoparticle inclusion in the thermal and mechanical properties of the composites. The morphological, mechanical, thermal, dielectric and magnetic properties of permalloy nanoparticles (Ni 75 Fe 20 Mo 5 -Py-NP) embedded in ABS have been characterized and numerical simulations have been performed to explore and analyze the magnetic behavior of the composites with different nanoparticle content.

Materials and Sample Preparation
Acrylonitrile butadiene styrene (ABS) thermoplastic polymer from Elix Polymers was selected as the polymeric matrix (Elix ABS DP, melt volume rate (MVR-220 • C/10 kg) = 20.34; impact strength (23 • C, ISO 180-1A) = 24.57 KJ/m 2 ), and acetone from Scharlab (Barcelona, Spain) as the solvent. The magnetic nanoparticles (NP) are Ni 80 Fe 17 Mo 3 (permalloy-Py), with reference 9288 HW, a purity of 99+% and average particle size of 70 nm (as indicated by the provider, Nanostructures & Amorphous Materials Inc). The ratio of the elements lays in between the supermalloy (Ni 75 Fe 20 Mo 5 ) and permalloy (Ni 80 Fe 17 ), both of which maintain the soft-magnetic properties. The prepared ABS/Py-nanoparticles (Py-NP) composites with varying filler content are presented in Table 1, together with the nomenclature used in the rest of the work. Table 1. Pristine as composite samples processed with varying Py content and the corresponding nomenclature.

Sample Name
Pure ABS ABS ABS + Py 10 wt% ABS-Py10 ABS + Py 20 wt% ABS-Py20 ABS + Py 40 wt% ABS-Py40 ABS + Py 60 wt% ABS-Py60 ABS + Py 80 wt% ABS-Py80 Pure Py-NP (100%) Py-NP All samples were prepared following the same procedure, as described in Figure 1. First, 2 g of ABS were dissolved in 10 mL of acetone under magnetic stirring at room temperature. The corresponding mass of permalloy was weighed, depending on the desired filler concentration of the sample, and added to the dissolved polymer within an ad-hoc plastic container and mixed for 10 min at 2000 rpm with a Thinky ARE-250 (THINKY CORPORATION, Nordson EFD, Tokyo, Japan) planetary mixer.
All samples were prepared following the same procedure, as described in Figure 1. First, 2 g of ABS were dissolved in 10 mL of acetone under magnetic stirring at room temperature. The corresponding mass of permalloy was weighed, depending on the desired filler concentration of the sample, and added to the dissolved polymer within an ad-hoc plastic container and mixed for 10 min at 2000 rpm with a Thinky ARE-250 (THINKY CORPORATION, Nordson EFD, Tokyo, Japan) planetary mixer. The samples were labelled according to their content in permalloy as presented in Table 1, (ABS-PyX indicates a X wt% content of the magnetic filler). To prepare the films, the opaque viscous liquid obtained from mixing was coated on a clean glass substrate by the solvent-casting technique. The films were dried overnight at ambient temperature. Finally, the samples were detached from the glass using water and stored for future characterization. Films with an average thickness of 50 μm were obtained with a Mitutoyo MDC-25PX micrometer gauge. Additionally, neat ABS samples were prepared by dissolving the polymer, casting onto the glass substrate and drying overnight. The Py-NPs were also magnetically characterized.

Sample Characterization
The magnetic filler dispersion and distribution within the polymer composites were evaluated by scanning electron microscopy (SEM), Hitachi TM300 Tabletop microscope (Tokyo, Japan). Surface and cross-sectional measurements were performed together with energy-dispersive X-ray spectroscopy (EDX) using a Hitachi S-3400 microscope at an accelerating voltage of 20 kV and different magnifications of 500×, 2500× and 10,000×. Before the measurements, the samples were coated with a 20 nm gold layer via sputtering with a Polaron SC502 apparatus.
To characterize the shape and size distribution of the nanoparticles, images were obtained using transmission electron microscopy (TEM). The equipment used was a JEOL JEM 1400 Plus (JEOL Ltd., Tokyo, Japan) with a tungsten filament, accelerating voltage of 120 kV and equipped with a sCMOS digital camera for image acquisition. Before imaging, the particles were dispersed in ethanol with a concentration of 0.5 mg/mL using an ultrasound bath. About 3 microliters of this dispersion were deposited onto a carbon grid and left at room temperature until the ethanol was evaporated. The images were obtained at different magnifications, and then the diameter of the particles was measured using Fiji (ImageJ) software.
The thermal behavior of the samples was determined by differential scanning calorimetry (DSC) and thermogravimetric (TGA) curves. The DSC measurements were carried out in a Perkin-Elmeer DSC 8000 (Waltham, MA, USA) apparatus with a sample The samples were labelled according to their content in permalloy as presented in Table 1, (ABS-PyX indicates a X wt% content of the magnetic filler). To prepare the films, the opaque viscous liquid obtained from mixing was coated on a clean glass substrate by the solvent-casting technique. The films were dried overnight at ambient temperature. Finally, the samples were detached from the glass using water and stored for future characterization. Films with an average thickness of 50 µm were obtained with a Mitutoyo MDC-25PX micrometer gauge. Additionally, neat ABS samples were prepared by dissolving the polymer, casting onto the glass substrate and drying overnight. The Py-NPs were also magnetically characterized.

Sample Characterization
The magnetic filler dispersion and distribution within the polymer composites were evaluated by scanning electron microscopy (SEM), Hitachi TM300 Tabletop microscope (Tokyo, Japan). Surface and cross-sectional measurements were performed together with energy-dispersive X-ray spectroscopy (EDX) using a Hitachi S-3400 microscope at an accelerating voltage of 20 kV and different magnifications of 500×, 2500× and 10,000×. Before the measurements, the samples were coated with a 20 nm gold layer via sputtering with a Polaron SC502 apparatus.
To characterize the shape and size distribution of the nanoparticles, images were obtained using transmission electron microscopy (TEM). The equipment used was a JEOL JEM 1400 Plus (JEOL Ltd., Tokyo, Japan) with a tungsten filament, accelerating voltage of 120 kV and equipped with a sCMOS digital camera for image acquisition. Before imaging, the particles were dispersed in ethanol with a concentration of 0.5 mg/mL using an ultrasound bath. About 3 microliters of this dispersion were deposited onto a carbon grid and left at room temperature until the ethanol was evaporated. The images were obtained at different magnifications, and then the diameter of the particles was measured using Fiji (ImageJ) software.
The thermal behavior of the samples was determined by differential scanning calorimetry (DSC) and thermogravimetric (TGA) curves. The DSC measurements were carried out in a Perkin-Elmeer DSC 8000 (Waltham, MA, USA) apparatus with a sample robot, between 25 • C and 350 • C at a heating rate of 10 • C min −1 under nitrogen purge (50 mL min −1 ) in 40 µL aluminum cans with perforated lids. The TGA was performed with a thermal gravimetric analyzer (TGA) METTLER TGA/DSC1, Switzerland, apparatus in the temperature range from 30 to 900 • C in nitrogen atmosphere (50 mL/min) at a heating rate of 10 • C/min.
The mechanical properties were evaluated using a universal testing machine Shimadzu model AG-IS with a load cell of 1 kN. The films were cut into rectangular probes 50 mm in length and 10 mm wide (cut with an Epilog Laser Mini-18 30 W-Epilog Corporation  (Table Mountain Pkwy, Golden, CO, USA), with an average thickness of 50 µm. Four different probes for each sample were tested at room temperature in the tensile mode, with a deformation velocity of 3.0 mm s −1 . The Young's modulus or elastic modulus (E) was obtained by calculating the slope of the linear region. Furthermore, the strain at break (εb) and stress at break (σb) were also obtained.
The electrical properties of the samples were evaluated measuring the dc conductivity from the characteristic current-voltage (I-V) tests measured by applying voltage steps and measuring the current using an automated picoammeter/voltage source Keithley 487 (steps of 10 V between −100 V to +100 V for samples with low wt%, ABS-Py10, ABS-Py 20 and ABS-Py 40; steps of 1 V between −10 V to +10 V for ABS-Py60 and steps of 0.01 V between −0.1 V to +0.1 V for ABS-Py80). Gold electrodes 5 mm in diameter were previously deposited (Polaron SC502 sputter coater) on both sides of the samples. The electrical resistance (R) was calculated (considering the geometry of the samples, thickness (d) and area (A)) from the slope of the obtained I−V curves. The electrical conductivity (σ) was determined as the inverse of the resistivity (ρ), as presented in equation 1.
The dielectric properties were measured with a Quadtech 1920 LCR precision meter. The capacity (C) and the dielectric losses (tan δ) were obtained at room temperature in the frequency range from 20 Hz to 1 MHz with an applied voltage of 1 V. The electrodes were placed similarly, as indicated in the electrical measurements. The dielectric constant (ε') was determined, taking into consideration the geometrical characteristics of the sample (Equation (2), where the permittivity of the vacuum is ε 0 = 8.85 × 10 −12 F m −1 .
The magnetic properties of the composites were evaluated by measuring the magnetic hysteresis loops (HL) at room temperature using a MicroSense EZ7 vibrating sample magnetometer (VSM). The experimental concentrations of the magnetic filler in the samples were obtained by comparing the saturation magnetization (Ms) of the composites with the saturation magnetization of the nanoparticles. The coercive field (Hc) and remanent magnetization (Mr) were also obtained from the hysteresis loops.

Morphological Features
The shape and size distribution of the nanoparticles was obtained, as these parameters determine the mechanical and magnetic properties of the composites. The determination of the dispersion of NP in the polymeric matrixes is of significant relevance [55], since the homogeneity of the dispersion is essential to maintain a uniform functional behavior and mechanical properties, among others, which strongly depend on particle distribution [56]. In this scope, the magnetic NP represents an extra challenge due to the magnetic interaction, which promotes agglomeration. Figure 2 shows two representative transmission electron microscopy (TEM) images of pure Py-NP. The NP show a spherical morphology with a wide distribution of diameters, ranging from 18 nm to 117 nm ( Figure 2a). Figure 2b shows a closer detail of the NP. From the TEM images, an average particle diameter of 60 nm with a standard deviation of ± 20 nm was obtained. From the TEM images, an average particle diameter of 60 nm with a standard deviation of ± 20 nm was obtained. Representative surface and cross-section SEM images of the composites with different filler weight percentage (wt%) are shown in Figure 3c-l, and pure ABS in Figure 3a and b. The surface and cross-section images are shown with different scale/magnification to appreciate the overall surface morphology and particles dispersion in the first case, and the wettability of the NPs by the polymer and NP dispersion across the cross-section in the second. The Py-NP are homogeneously dispersed in the form of clusters, independently of the filler content. The clusters and voids are indicated with light-blue circles and green-light diamonds, respectively, in Figure 3c for the ABS-Py10 sample and Figure  3h for the ABS-Py40. Finally, there is a good wettability of the fillers by the polymer and no cracks or patterns are observed.  Representative surface and cross-section SEM images of the composites with different filler weight percentage (wt%) are shown in Figure 3c-l, and pure ABS in Figure 3a and b. The surface and cross-section images are shown with different scale/magnification to appreciate the overall surface morphology and particles dispersion in the first case, and the wettability of the NPs by the polymer and NP dispersion across the cross-section in the second. The Py-NP are homogeneously dispersed in the form of clusters, independently of the filler content. The clusters and voids are indicated with light-blue circles and greenlight diamonds, respectively, in Figure 3c for the ABS-Py10 sample and Figure 3h for the ABS-Py40. Finally, there is a good wettability of the fillers by the polymer and no cracks or patterns are observed. From the TEM images, an average particle diameter of 60 nm with a standard deviation of ± 20 nm was obtained. Representative surface and cross-section SEM images of the composites with different filler weight percentage (wt%) are shown in Figure 3c-l, and pure ABS in Figure 3a and b. The surface and cross-section images are shown with different scale/magnification to appreciate the overall surface morphology and particles dispersion in the first case, and the wettability of the NPs by the polymer and NP dispersion across the cross-section in the second. The Py-NP are homogeneously dispersed in the form of clusters, independently of the filler content. The clusters and voids are indicated with light-blue circles and green-light diamonds, respectively, in Figure 3c for the ABS-Py10 sample and Figure  3h for the ABS-Py40. Finally, there is a good wettability of the fillers by the polymer and no cracks or patterns are observed.  This behavior has been previously described [57,58] and the formation of clusters has been explained based on two main stages or regimes: the first stage is dominated by Brownian motion, where the NP gather first in micro-dimensions clusters or micronclusters. As the concentration increases [39], the second stage is based on the collisions and adhesions effect, leading to bigger clusters (or macro-clusters). In the present case, the same behavior is observed, where micro-clusters (first regime) are present in samples with lower wt% (e.g., ABS-Py10 sample- Figure 3c) and macro-clusters (the second regime) are in the samples with larger filler contents (e.g., ABS-Py60 and ABS-Py80 samples- Figure 3j,l).
Energy-dispersive X-ray spectroscopy (EDS) images of two representative samples are shown in Figure  This behavior has been previously described [57,58] and the formation of clusters has been explained based on two main stages or regimes: the first stage is dominated by Brownian motion, where the NP gather first in micro-dimensions clusters or micron-clusters. As the concentration increases [39], the second stage is based on the collisions and adhesions effect, leading to bigger clusters (or macro-clusters). In the present case, the same behavior is observed, where micro-clusters (first regime) are present in samples with lower wt% (e.g., ABS-Py10 sample- Figure 3c) and macro-clusters (the second regime) are in the samples with larger filler contents (e.g., ABS-Py60 and ABS-Py80 samples- Figure 3j and Figure 3l).
Energy-dispersive X-ray spectroscopy (EDS) images of two representative samples are shown in Figure 4, where the surface images are shown (Figure 4a

Thermal Properties
In order to evaluate how the addition of the Py-NP affects the glass transition temperature of the polymer, DSC measurements were performed in all samples. Figure 5a depicts the DSC thermograms of the most representative composites, as well as the values of the glass transition temperatures (Tg).

Thermal Properties
In order to evaluate how the addition of the Py-NP affects the glass transition temperature of the polymer, DSC measurements were performed in all samples. Figure 5a depicts the DSC thermograms of the most representative composites, as well as the values of the glass transition temperatures (Tg).
Brownian motion, where the NP gather first in micro-dimensions clusters or micro ters. As the concentration increases [39], the second stage is based on the collisio adhesions effect, leading to bigger clusters (or macro-clusters). In the present ca same behavior is observed, where micro-clusters (first regime) are present in sampl lower wt% (e.g., ABS-Py10 sample- Figure 3c) and macro-clusters (the second regi in the samples with larger filler contents (e.g., ABS-Py60 and ABS-Py80 samples-F and Figure 3l).
Energy-dispersive X-ray spectroscopy (EDS) images of two representative s are shown in Figure 4, where the surface images are shown (Figure 4a

Thermal Properties
In order to evaluate how the addition of the Py-NP affects the glass transitio perature of the polymer, DSC measurements were performed in all samples. Fi depicts the DSC thermograms of the most representative composites, as well as the of the glass transition temperatures (Tg).  The Tg of the sample is obtained from the peak of the calorimetry curve and it is found at around 105 • C for ABS, Figure 5a, which is consistent with values from the literature [59]. The inclusion of the magnetic filler does not affect this transition temperature for filler concentrations up to 60 wt%. An increase of Tg is observed for the sample containing 80 wt% of particles, which is attributed to the confinement of the polymer within the fillers and the corresponding clamping effect, leading to a slowdown of the local dynamics of the polymer and thus an increase in Tg [60].
TGA thermograms for the ABS-Py composites are shown in Figure 5b, and the quantitative values of the extrapolated onset temperature (T o ), which correspond the temperature at which the weight loss begins, the temperature at 10% weight loss (T 10 ) and the residual weight at the final test are summarized in Table 2. The results reveal that the thermal degradation of all the samples is characterized by one main stage. This step that occurs in the temperature range of approximately 350-450 • C is associated with ABS degradation [61]. In addition, significant changes in the onset temperature are observed when the concentration of Py is increased. The onset temperature T o increases from 350 for 10 wt% of Py to 390 • C for 80 wt% of Py. Based on this observation, it was concluded that the addition of Py has a significant effect in the thermal stability of the composites, indicating an improvement in the thermal stability of the ABS-based composites with Py. Table 2. Glass transition temperature, Tg, for the samples obtained from the DSC measurements and thermal degradation parameters of the ABS-Py composites, obtained from the TGA measurements. Thus, with respect to the thermal properties, it was concluded that the Tg of the composites is mainly independent of the filler loading and that increasing the Py concentration in the ABS matrix leads to a corresponding increase in the final residual weight after polymer degradation, directly related to the amount of Py in the composites.

Electrical Conductivity
The electric properties of the ABS and ABS-Py composites are presented in Figure 6a. The electrical conductivity of the ABS polymer is about 4 × 10 −12 S m −1 , corresponding to an insulator material. Regarding the Py-NP, the electrical resistivity of bulk Py (Ni 80 Fe 20 ) has been reported to be 30 µΩ cm [62], showing a conductive character. In the case of particles of Py (Ni 45 Fe 55 ) with an average diameter of 2.53 µm used in composite materials based on polyphenylene sulfide (PPS) resin, the AC conductivity depends on the surface oxidation state, which is typically formed as a thin layer on the surface of the particles. Thus, the electrical resistivity at 10 KHz increases in four orders of magnitude from 1.25 Ω cm for the non-oxidized samples to 2.4 × 10 4 Ω cm for the ones with an oxide layer. In our case, the pure Py-NP compacted in a pill shape and showed an electrical resistivity of 18 Ω cm, which agrees with the surface oxidation that occurs naturally once the particles are under environmental conditions. The electrical conductivity of the composite films increases with increasing Py content. For composite ABS-Py10 and ABS-Py20, the electrical conductivity is similar to pristine ABS, increasing almost 4 orders of magnitude for the composite with 80 wt% of Py embedded into the ABS host matrix. Although the electrical conductivity increases with increasing Py content, the low conductivity of the filler prevents a percolative increase of the electrical conductivity typical of conductive fillers within an insulator matrix [63,64]; therefore, the increase of the electrical conductivity is more related to interfacial effects and ionic conductivity [65,66]. The volume electrical conductivity for the samples with the lager filler concentrations are in the order of 10 −8 S m −1 , which are compatible with static dissipative materials [67].
Antistatic or dissipative materials can work as electrostatic discharge protection for electronic components and devices [68,69]. In this sense, the ABS-Py composites might have an interesting application when integrated with electronic devices.

Dielectric Response
The dielectric properties of the ABS-Py composites are presented in Figure 6b. The behavior of the dielectric constant with frequency (from 200 Hz to 1 MHz) is similar for all materials, decreasing for the initial frequencies and stabilizing for higher ones. The value of the dielectric constant increases with increasing the Py-NP content in the composites. At 1 kHz, the ε' ≈ 3.5 for ABS increases up to ε' ≈ 5.7 for the ABS-Py60 composite. The dielectric behavior obtained is similar to that presented in the literature for composites using ABS as a matrix (in the range 3 < ε'< 10) [70]. The increase of the dielectric constant is related to the dielectric contribution of the filler, as well as to the interfacial effects [71,72]. The dielectric losses present similar behavior when compared to dielectric constant as a function of frequency and as a function of the Py content in the composites. The tan δ ≈ 4.0 × 10 −3 for ABS, increasing two orders of magnitude for the ABS-Py60 composite (tan δ ≈ 1.7 × 10 −1 ) due to larger Py-NP filler content [73]. Figure 7a shows representative stress-strain mechanical curves of the neat polymer and the corresponding composites; and Table 3 shows the values of the mechanical characteristics of the ABS and the composites with different content of Py-NP. Figure 7a displays the neat ABS, which presents the characteristic stress-strain curve for thermoplastics, characterized by an elastic region, a yielding region (denoted by a circle) and breaking at higher strains [74]. The Young's modulus values are presented in Figure 7b with the The electrical conductivity of the composite films increases with increasing Py content. For composite ABS-Py10 and ABS-Py20, the electrical conductivity is similar to pristine ABS, increasing almost 4 orders of magnitude for the composite with 80 wt% of Py embedded into the ABS host matrix. Although the electrical conductivity increases with increasing Py content, the low conductivity of the filler prevents a percolative increase of the electrical conductivity typical of conductive fillers within an insulator matrix [63,64]; therefore, the increase of the electrical conductivity is more related to interfacial effects and ionic conductivity [65,66]. The volume electrical conductivity for the samples with the lager filler concentrations are in the order of 10 −8 S m −1 , which are compatible with static dissipative materials [67].

Mechanical Properties
Antistatic or dissipative materials can work as electrostatic discharge protection for electronic components and devices [68,69]. In this sense, the ABS-Py composites might have an interesting application when integrated with electronic devices.

Dielectric Response
The dielectric properties of the ABS-Py composites are presented in Figure 6b. The behavior of the dielectric constant with frequency (from 200 Hz to 1 MHz) is similar for all materials, decreasing for the initial frequencies and stabilizing for higher ones. The value of the dielectric constant increases with increasing the Py-NP content in the composites. At 1 kHz, the ε' ≈ 3.5 for ABS increases up to ε' ≈ 5.7 for the ABS-Py60 composite. The dielectric behavior obtained is similar to that presented in the literature for composites using ABS as a matrix (in the range 3 < ε'< 10) [70]. The increase of the dielectric constant is related to the dielectric contribution of the filler, as well as to the interfacial effects [71,72]. The dielectric losses present similar behavior when compared to dielectric constant as a function of frequency and as a function of the Py content in the composites. The tan δ ≈ 4.0 × 10 −3 for ABS, increasing two orders of magnitude for the ABS-Py60 composite (tan δ ≈ 1.7 × 10 −1 ) due to larger Py-NP filler content [73]. Figure 7a shows representative stress-strain mechanical curves of the neat polymer and the corresponding composites; and Table 3 shows the values of the mechanical characteristics of the ABS and the composites with different content of Py-NP. Figure 7a displays the neat ABS, which presents the characteristic stress-strain curve for thermoplastics, characterized by an elastic region, a yielding region (denoted by a circle) and breaking at higher strains [74]. The Young's modulus values are presented in Figure 7b with the exception of the ABS-Py80 sample, which is very brittle.   Table 3 show that the presence of the Py-NP changes the mec behavior of the polymer. First, the stress-strain curves show a completely vanished ing region, retaining only the elastic behavior. Second, the Young's modulus de slightly when the Py-NP are first introduced with a lower wt% (from 1.2 GPa for 0.7 GPa for the ABS-Py20 composite), although it increases later with increasin Thus, for small filler concentrations, the inclusions act as a defect within the polym trix, whereas for larger concentrations, they act as a reinforcement material [52]. It be noticed that the ABS-Py60 sample exhibits an elastic modulus higher than th polymer, with breaking stress also remarkably close to the bare ABS. However, du significant content of nanoparticles in this sample, the composite also presents the est elongation at the break.

Mechanical Properties
Composites of NP embedded into polymer matrices have been under study their mechanical properties, since many parameters must be considered when the are added to polymer materials, such as particle/matrix interfacial adhesion, NP co dimensions and dispersion [75,76]. In many of these studies, the Young's modu proves by the presence of the NP, since the polymer matrices have much lower s than the inorganic particles; however, it also depends on the interfacial adhesion a stress transferred between the matrix and the NP [76]. In the present case, it can b cluded that the Py-NP are well bonded to the ABS, since the presence of the NP im the Young´s modulus, mainly for higher filler contents [77,78] up to 60 wt%.
Increasing filler content leads to a decrease of the breaking strain to values c the yielding strain of ABS. As the wt% increases, the particles lead to discontinuity the polymer, increasing the brittleness of the composite [79]. The elongation at the does not present a linear behavior with increasing wt%, and all the composite s show lower elongation at the break than ABS, decreasing with the addition of the (from pure ABS to the composites with 60 wt%, with average values of 3.29% ± 0.55 1.07% ± 0.31%, respectively, Table 3). An overall increase of the breaking stress w filler concentration is observed, which is related to the formation of clusters, sin particle-polymer interfaces may act as breaking sites. Since the clusters grow when wt% is introduced, the surface-to-volume ratio decreases, decreasing these breakin reaching to the values of the ABS. The average magnitude of the breaking stress o ABS and a composite with 60 wt% are 19.86 MPa ± 6.73 MPa and 19.26 MPa ± 9.1 respectively.
The deterioration of the mechanical properties of thermoplastics polymer w   Figure 7 and Table 3 show that the presence of the Py-NP changes the mechanical behavior of the polymer. First, the stress-strain curves show a completely vanished yielding region, retaining only the elastic behavior. Second, the Young's modulus decreases slightly when the Py-NP are first introduced with a lower wt% (from 1.2 GPa for ABS to 0.7 GPa for the ABS-Py20 composite), although it increases later with increasing wt%. Thus, for small filler concentrations, the inclusions act as a defect within the polymer matrix, whereas for larger concentrations, they act as a reinforcement material [52]. It should be noticed that the ABS-Py60 sample exhibits an elastic modulus higher than the neat polymer, with breaking stress also remarkably close to the bare ABS. However, due to the significant content of nanoparticles in this sample, the composite also presents the smallest elongation at the break.
Composites of NP embedded into polymer matrices have been under study due to their mechanical properties, since many parameters must be considered when the fillers are added to polymer materials, such as particle/matrix interfacial adhesion, NP contents, dimensions and dispersion [75,76]. In many of these studies, the Young's modulus improves by the presence of the NP, since the polymer matrices have much lower stiffness than the inorganic particles; however, it also depends on the interfacial adhesion and the stress transferred between the matrix and the NP [76]. In the present case, it can be concluded that the Py-NP are well bonded to the ABS, since the presence of the NP improves the Young's modulus, mainly for higher filler contents [77,78] up to 60 wt%.
Increasing filler content leads to a decrease of the breaking strain to values close to the yielding strain of ABS. As the wt% increases, the particles lead to discontinuity within the polymer, increasing the brittleness of the composite [79]. The elongation at the break does not present a linear behavior with increasing wt%, and all the composite samples show lower elongation at the break than ABS, decreasing with the addition of the Py-NP (from pure ABS to the composites with 60 wt%, with average values of 3.29% ± 0.55%, and 1.07% ± 0.31%, respectively, Table 3). An overall increase of the breaking stress with the filler concentration is observed, which is related to the formation of clusters, since the particle-polymer interfaces may act as breaking sites. Since the clusters grow when higher wt% is introduced, the surface-to-volume ratio decreases, decreasing these breaking sites, reaching to the values of the ABS. The average magnitude of the breaking stress of pure ABS and a composite with 60 wt% are 19.86 MPa ± 6.73 MPa and 19.26 MPa ± 9.18 MPa, respectively.
The deterioration of the mechanical properties of thermoplastics polymer with different fillers has also been observed in composites of ABS with nanosized barium ferrite (BaFe 12 O 19 ) [1], with a decrease of the tensile strength and the Young's modulus when the filler amount increases.
For many applications, it is essential to maintain the mechanical properties of the thermoplastic ABS, since it is one of its most striking properties, such as high impact resistance (even at low temperatures), toughness, high dimensional stability (mechanically strong and stable over time) and rigidity compared to other related polymers. Overall, the ABS-based composites with embedded Py-NP show an elastic behavior with suitable mechanical properties up to 60 wt%, where an increases in the Young's modulus and a breaking stress similar to the pristine ABS provides good possibilities to being implemented in magnetic devices.

Magnetic Response
The understanding of the magnetic properties of the composites and the interaction/agglomeration of the Py-NPs within the polymer will allow for engineering of the composites for their application as magnetic sensors and actuators. Figure 8a displays the hysteresis loops (HL) of the Py composite samples with increasing wt% and the representative magnetic properties extracted from the HL, such as Hc in Figure 8b, Mr in Figure 8c and Ms in Figure 8a-inset. Table 4 shows the magnetic properties, Hc, Mr and Ms, of all the samples of ABS-Py. Regarding the Py-NP [80,81], the magnetic behavior of magnetic-NP is not the same as the bulk material due to different contributions that depend on the low-dimensions and the shape of the nanoparticles, such as shape anisotropy and the surface effect. The latter is one of the most important contributions, which gives place to the "spin canting" at the surface of the NP as a response of the different magnetic contributions. The spins at the surface are slightly disoriented with respect to the core, decreasing the Ms.
The diameter of the NP also plays an important role, with the NP ranging from superparamagnetic at very low dimensions to a single-domain regime, and then to a multi-domain one at the critical diameter (Dc) (which it is translated into a variation in the mechanism of the magnetic switching shown in the HL) [80,82]. A series of numerical simulations were performed by using Mumax3 Code [83], which allows for the evaluation of the hysteresis loops of Py-NP with different diameters, and to prove the critical diameter where the NP magnetic behavior changes from magnetic monodomain to multidomain. The values used for the Py are: Ms = 8,0 e +5 A/m; A exc = 1.9 e −11 J/m; K 1 = 0.0 J/m 3 ; and alpha = 1.0 (such value is typically used on the calculation of HL). The boundary conditions, the mathematical model and the model validation conditions are detailed in Appendix A. The temperature was set at 10K, which is enough to observe a small thermal activation in the spin's configurations.
According to the calculations (HL in Figure 8e), the Py-NP with diameters smaller or equal to 40 nm show a squared-shaped HL, indicating a coherent rotation switching mechanism characteristic of the monodomain regime. On the other hand, when the diameters of the spherical Py-NP increases, the shape of the HL near the Hc changes, showing a smooth and curved switching of the magnetization, which is characteristic of the domain wall formation and propagation of a multidomain regime.  The diameter of the NP also plays an important role, with the NP ranging from superparamagnetic at very low dimensions to a single-domain regime, and then to a multidomain one at the critical diameter (Dc) (which it is translated into a variation in the mechanism of the magnetic switching shown in the HL) [80,82]. A series of numerical  It was also corroborated how the Hc increases with the diameter of the NP at lower dimensions (lower than 40 nmAt higher dimensions (diameters higher than 40 nm), the Hc decreases considerably (from 8.5 Oe to 1.5 Oe, for the 40 nm and the 60 nm diameter, respectively); confirming that D = 40 nm is the Dc.These results agree with the experimental results, where the mono-multidomain transition in the Py-NP occurs in-between 35 and 45 nm [84]. Nevertheless, the NPs with D = 80 nm show a higher coercivity, which can be related to the higher energy needed to reverse the magnetization by the production and propagation of the domain wall (a competition between the magnetostatic and exchange energy) [81]. The switching mechanism transition is also corroborated by the spin configurations at remanence in Figure 8f; where the magnetization after saturation shows a full remanence in the NP with D = 40 nm, which is demonstrative of a square HL, while the NP with D = 50 nm presents a spin canting due to the thermal activation and a mechanism of the domain wall propagation.
Experimentally, the Py-NP of this work is a mix of NPs with different diameters, in which the switching magnetization is a mix of single-domain and multidomain regimes (Dc is 40 nm, and the average diameter in the commercial Py-NP is around 60-70 nm; therefore, the composites show magnetic behavior from both regimes). The HL of the pristine NP in Figure 8a show that the reversal magnetizations (the HL) are not square, with Mr one order of magnitude lower than the Ms; making evident the predominance of multidomain regimes. In this case, the Ms of pure Py-NP is 65.78 emu/g, which, as it is expected, it is lower than the reported value for the bulk Py (around 80 emu/g [85]), due to the spin canting and the multidomain regime. The higher coercivity obtained experimentally in the pristine NP is due to the interactions between them (magnetostatic interactions), therefore the spherical symmetry is broken, increasing the effective anisotropy and, therefore, the coercive field.
Regarding the ABS-Py composites, it is essential to evaluate if they maintain their ferromagnetic properties when they are embedded in the ABS. This can be observed in the values of the Ms in Figure 8a-inset. In this case, the linear and proportional increases with the wt% (based on the Ms of the pure Py: Ms = 65.78 emu/g) is evidence that they retain their ferromagnetic behavior.
An important aspect of the composite films is the homogeneity in the concentration of the NP, since their dispersion within the polymer will dictate the magnetic and mechanical behavior of the films. This space-dependence is led by the amount of NP present in specific areas, as indicated by the morphological characteristics of the samples, where it was observed that there are clusters distributed along the sample that can slightly alter their properties. As an example, in the films of ABS-Py80, the values of Ms and Hc measured at different places shows a standard deviation (SD) of around 4.7 emu/g (i.e., which corresponds to a deviation of 13% of amount of material) and 1.9 Oe, respectively. In the former, it is caused by the dispersion of the clusters, which change along the sample. In the latter, the SD is very small, and it is more related to the interparticle-interaction (I-I), which depends on the interparticle distance. In this case, the distance is shorter and the I-I is stronger due to the higher concentration of Py-NP.
Regarding the samples with lower wt%, the SD of the Ms is smaller than the one of ABS-Py80. Nevertheless, there is not a linear trend with increasing Py. In the case of the SD of the Hc, it decreases with increasing Py, which can be explained by the fact that, as the amount of Py increases, the distances between the NP decrease, decreasing the I-I and its deviation.
Regarding the relationship of the Ms and the diameter of the NP, it has been shown [85] that Ms increases with increasing diameter (as can be observed as well in Figure 8d), ranging from 28.7 emu/g to 80.8 emu/g for the NPs with 20 nm and 440 nm of diameter, respectively. In the present work, the Py-NP present a distribution of diameters around 60 nm ± 20 nm, with an Ms of 65.78 emu/g, resulting in a dispersion of Ms from different diameters; hence there is a high contribution from the larger diameters, showing higher values of Ms. This effect is also evidenced in the Hc, laying around 87-90 Oe for low wt% and 81-82 Oe for both ABS-Py80 and pure Py-NP. In another study [85] particles with 76 nm of diameter show a coercivity of 27 Oe, which is much lower than in the present study; therefore, we attribute the high values of coercivity in the Py-NP of this study not to the dimensions of the NP (since there is a mix of dimensions), but rather to a high interaction between particles and between the clusters (as explained below), which creates a higher needed energy for their magnetic switching. The decreasing values of Hc with filler concentration is explained through the packing effect [82]. The packing effect works in the following manner: a magnetized NP (e.g., NP-1) exerts a field, which affects the neighbor's NPs (e.g., NP-2); therefore, when the magnetic field reverses to the opposite direction, the field of NP-1 assists in the switching of the field of NP-2, reversing the magnetization at the lower field. All the NPs are interacting with their neighbors, and the interaction is stronger at lower distance. The magnetizations of the NPs at zero field are randomly oriented and when the field increases, the NP that are already aligned to the field help the others to switch sooner, decreasing the Hc. These interactions also explain the increasing values of the Mr with the wt% (Figure 8c) since they help each other to maintain their magnetic configuration after saturation.
There is also another effect resulting from the clusters (tightly bound NPs forming a larger virtual particle), which is that they may behave magnetically within the polymer as a bigger particle instead of behaving like individual NPs. This conclusion can be based in the results presented in [86], where magnetic studies by simulation of individual and groups of Py-NP were conducted. It was shown that a large NP behaves the same as a cluster of several smaller ones (the cluster + the space between the NP has the same dimensions as the large particle). This leads us to conclude that the magnetization of the micro-clusters within the composites behaves as one big particle, and that the reversal mechanisms of the cluster need higher fields to switch their magnetization. This contrasts with the individual NP that needs lower fields to reverse the magnetization. These findings contributes in explaining the higher coercivities of the composites in this study in comparison to other studies; and in comparison to the simulations of individual NPs.

Conclusions
ABS composites with Py-NP as fillers up to 80 wt% have been prepared and evaluated. After the addition of the Py-NP to the polymer matrix, the glass transition temperature is preserved, while the thermogravimetric analysis shows that the addition of the Py nanoparticles leads to a significant increase in the final residual weight, which corresponds to the amount of NPs added. The dielectric constant of the composites increases with increasing filler content up to a maximum value of 4.4 at 10 kHz for the sample with 60 wt% Py content. Furthermore, the electrical conductivity of the different composites is in the order of 10 −8 -10 −10 S/m, increasing with filler content, which opens possibilities to integrate these composites into electronic devices based on the dissipative characteristics of some composites.
Regarding the mechanical properties, the presence of the Py-NP leads to an increase of the elastic modulus, or Young's modulus, for the samples with the larger filler contents, going from 1.16 GPa in the ABS, to almost 2 GPa in the composite with 60 wt%; while the ones with less than 20 wt% have a decrease in the Young's modulus, with 7.6 GPa. Additionally, the elongation at the break decreases with the load of Py-NP and the breaking stress increases with the Py content, reaching almost the same values as the ABS, 19.5 GPa and 19.3 GPa for composites of ABS-Py40 and ABS-Py60, respectively (the value is 19.8 GPa for the ABS). These values are important because they not only demonstrate an improvement in the Young's modulus, but also the capability of the ABS in taking a high load of magnetic nanoparticles.
Regarding the magnetic properties, the saturation magnetization of the composites increases linearly with increasing Py content up to a value of 50.9 emu/g for the samples with 80 wt% of Py content, and the interparticle filler interactions determine the coercivity and remanence of the composites, which has been supported by the numerical modeling. This study has gone deeper in the understanding of the magnetic behavior of the NP within the polymer since their dispersion and agglomeration dictates the behavior of the composites. The numerical modeling that is presented in this work gives support in the findings about the magnetic behavior of the NP, providing an excellent tool to best understand their behavior within the ABS when they are agglomerated. As a conclusion about the composite's properties, the unchanged transition temperature, the slight improvement in the mechanical properties and the added value of the magnetic properties demonstrates their suitability for new possibilities of applications as magnetic devices, such as magnetic sensors, actuators and magnetic field shielding.  Data Availability Statement: Data available on request due to restrictions of privacy. The data presented in this study are available on request from the corresponding author. The data are not publicly available due to restrictions of privacy.