Ayda Ghary
Haghighat
abc,
Eider
Matxinandiarena
d,
Manuela
Zubitur
de,
Agurtzane
Mugica
d,
Fulvio
Bellato
f,
Anna M.
Ferretti
f,
Alessandro
Ponti
f,
Souad
Ammar
g,
Maryam
Abdolrahimi
ab,
Gaspare
Varvaro
b,
Pierfrancesco
Maltoni
ab,
Dario
Cavallo
a,
Alexander
Omelyanchik
*ab,
Alejandro J.
Müller
*dh and
Davide
Peddis
*ab
aUniversità degli Studi di Genova, Dipartimento di Chimica e Chimica Industriale, nM2-Lab, Via Dodecaneso 31, 1-16146 Genova, Italy. E-mail: aleksander.omelianchik@ext.unige.it; davide.peddis@unige.it
bCNR, Istituto di Struttura della Materia, nM2-Lab, Monterotondo Scalo (Roma), 00015, Italy
cUniversità degli Studi Roma Tre, Dipartimento di Scienze, Via Ostiense 159, 00146 Rome, Italy
dPOLYMAT and Department of Advanced Polymers and Materials: Physics, Chemistry and Technology, Faculty of Chemistry, University of the Basque Country UPV/EHU, Paseo Manuel de Lardizabal 3, Donostia, San Sebastián 20018, Spain. E-mail: alejandrojesus.muller@ehu.es
eDepartment of Chemical and Environmental Engineering, University of the Basque Country, UPV/EHU, Plaza Europa 1, 20018 Donostia-San Sebastián, Spain
fConsiglio Nazionale delle Ricerche, Istituto di Scienze e Tecnologie Chimiche “Giulio Natta”, Via G. Fantoli 16/15, 20138 Milano, Italy
gUniversité Paris Cite, CNRS, ITODYS (UMR-7086), 15 rue JA de Baïf, 75205 Paris, France
hIKERBASQUE, Basque Foundation for Science, Plaza Euskadi 5, 48009 Bilbao, Spain
First published on 15th July 2025
Biodegradable polymer matrices, poly(ε-caprolactone) (PCL), and poly(butylene succinate-ran-butylene adipate) (PBSA) were used to fabricate magnetic composites with recycled NdFeB and rare earth-free lab-synthesized ferrite fillers (SrFe12O19 and SrFe12O19–CoFe2O4) across a wide filling range (1–90%). Results obtained by differential scanning calorimetry, polarized light optical microscopy, and phase contrast microscopy, indicated that the magnetic particles tend to aggregate, leading to bimodality in the crystallization process, which can be attributed to distinct regions of the composites with well-dispersed and aggregated particles. Notably, ferrite fillers exhibited lower magnetic anisotropy compared to NdFeB, enabling magnetic saturation at lower fields. These results demonstrate the potential of combining biodegradable polymers with sustainable magnetic fillers for eco-friendly circular economy applications.
Sustainability spotlightThis work addresses critical sustainability challenges by developing magnetic polymer composites that combine biodegradable matrices (PCL and PBSA) with either recycled neodymium magnets or rare-earth-free magnetic fillers. The research demonstrates a significant advancement toward circular economy principles in functional materials design. By successfully incorporating recycled NdFeB magnets and developing rare-earth-free alternatives, we present viable pathways to minimize environmental impacts associated with rare earth mining while maintaining functional magnetic properties. The integration of these fillers with biodegradable polymer matrices enhances the sustainability profile of these composites, as both PCL and PBSA can be produced from renewable resources. The systematic investigation of filler loading effects on thermal and magnetic properties provides essential knowledge for optimizing these materials for specific applications without compromising their environmental benefits. This research contributes to several UN Sustainable Development Goals: Responsible Consumption and Production (SDG 12) through the recycling of end-of-life permanent magnets; Industry, Innovation, and Infrastructure (SDG 9) through the development of new sustainable functional materials; and Climate Action (SDG 13) by promoting circular economy approaches that reduce environmental impacts associated with material extraction and processing. |
PCL, a semicrystalline polymer with a low melting temperature (Tm = 56–65 °C),15 exhibits high flexibility and compatibility with various processing methods, including 3D printing. Its controlled biodegradation rate and biocompatibility make it a sustainable material for environmentally friendly applications.15,19 PBSA is a copolymer of poly(butylene succinate) (PBS) with a melting temperature (Tm ≈ 84 °C)20 lower than that of PBS but with comparable flexibility and processability.17 PBSA benefits from improved enzymatic degradability due to its lower crystallinity compared with PBS.17 Moreover, the PBS-based copolymers can be produced using bio-based renewable resources like sugarcane and corn.18 Both PCL and PBS-based polymers are frequently blended with other biodegradable materials, such as PLA, to optimize mechanical and thermal properties, broadening their applicability in advanced composite designs.9,18
Incorporating magnetic micro- or nanoparticles (M(N)Ps) into polymer matrices significantly modifies the structural and thermal properties of the polymer matrix.21 These effects are highly dependent on the composite processing method, the molecular mass of the polymer, and the surface functional groups of magnetic particles, leading to contradictory results in the literature. For instance, in ref. 9 it was reported that both Tm and the crystallization temperature (Tc) of PCL are not affected by the presence of Fe3O4 magnetic nanoparticles, while in ref. 22 the authors observed an opposite effect, with both Tm and Tc decreasing with increasing Fe3O4 nanoparticle content, likely due to a decreased chain mobility and crystallization ability caused by interactions between polymer chains and Fe3O4via purposely introduced groups in PCL. Additional effects observed in polymer magnetic composites include a reduction in the size of PCL spherulites, accompanied by an increase in Tm when Fe3O4@graphene oxide nanocomposites are incorporated,23 or changes in thermal stability, often accelerating degradation, driven by interactions between particles and polymer chains.23,24 These findings highlight the complexity of such systems, underscoring the need for a systematic investigation, particularly in PCL and PBSA-based systems, to optimize their properties for advanced applications.
Beyond structural modifications, M(N)Ps impart unique functionalities, such as magnetic responsiveness.21 By tailoring the type and concentration of M(N)Ps, these composites can be optimized for advanced applications, including 3D-printed magnets and intelligent devices.9–13,25–28 PCL has been studied as electrospun fibrous24 or 3D-printed9 scaffolds for hyperthermia applications and magnetic-triggered actuators.22 Fe3O4/PBSA composites have been investigated as magnetic actuators in soft robotics.20 Research on these polymers remains limited, presenting an opportunity for novel investigations. Exploring biodegradable polymers in PMCs is crucial, as they offer unique attributes that complement those of more commonly studied materials.
It is worth mentioning that, to the best of our knowledge, in most cases involving PCL and PBSA, the magnetic properties of PMCs were controlled primarily by varying the content of simple iron oxides (magnetite Fe3O4 or maghemite γ-Fe2O3) nanoparticles. In contrast, PLA-based composites have been explored with a wider range of magnetic fillers, including cobalt ferrite CoFe2O4 (CFO),6,8,13,29 zinc-substituted cobalt ferrite Zn0.3Co0.7Fe2O4 and CoFe2O4 mixture,6 and barium hexaferrites BaFe12O19.6 This variability in composition enables the tuning of magnetic properties for targeted applications.
NdFeB stands out among various magnetic materials due to its exceptional magnetic properties, such as high saturation magnetization (MS) and coercivity (HC), making it the preferred choice for manufacturing high-performance permanent magnets.30 However, the scarcity of rare earth elements (REEs), like neodymium, along with the environmental challenges associated with their extraction, has led to a growing demand for sustainable alternatives, and recycling of end-of-life permanent magnets emerged as a promising solution with significant potential for both the short and long term.31–34 Although recycled magnetic materials generally underperform compared to solid NdFeB magnets in terms of energy product, PMCs offer significant advantages in terms of mechanical properties, corrosion resistance, and the ability to produce complex geometries, including micro-textured patterns.21,26,27,35–38
Another important category of magnetically hard materials is M-type hexaferrites (MFe12O19, where M = Ba or Sr).39 These materials, which do not contain REEs, are known for their excellent chemical and oxidation resistance, low cost, high coercivity, and Curie temperature despite having relatively lower MS. Their unique properties make hexaferrites attractive for applications where cost-effectiveness and moderate performance are desired, particularly in high-temperature environments.39 Additionally, there has been an increasing interest in hexaferrite-based hard/soft exchange-coupled structures, combining the high anisotropy of the hard phase with the high saturation magnetization of the soft phase, resulting in enhanced magnetic performance for a variety of applications.34,40,41
In this variegated scenario, this work represents the first step to address the existing gap by selecting less commonly used, biodegradable PCL and PBSA polymers and incorporating three different types of magnetically hard particles: recycled NdFeB microparticles, REE-free lab-synthesized SrFe12O19 (SFO) nanoparticles and hard/soft SrFe12O19–CoFe2O4 (SFO-CFO) nanocomposites. The key novelty of this work lies in the use of biodegradable polymers and REE-free or recycled materials, aligning with the principles of circular economy and sustainable material design. The effects of incorporating these M(N)Ps into the polymer matrices, assessing mainly their influence on thermal properties were investigated. Additionally, the tunability of the composite magnetic properties by varying the type and concentration of magnetic fillers was demonstrated.
NdFeB microparticles employed in this work were donated by Dr Nerea Burgos from CEIT (University of Navarra, Spain). These microparticles were produced from different industrial waste sources by mixing NdFeB powder with an epoxy resin. Morpho-structural characterization of NdFeB microparticles is shown in the (ESI, Fig. S1–S3†). The X-ray diffraction pattern corresponds to the hexagonal Nd2Fe14B crystalline structure. According to scanning electron microscopy, the NdFeB particles have spherical shapes with diameters in the range of 2–20 μm. Energy-dispersive X-ray spectroscopy (EDX) shows the copresence of B, Fe, and Nd.
A simultaneous bi-phasic approach was employed for the SFO-CFO synthesis. Precursors for CFO and SFO were separately prepared following the same initial steps as the SFO synthesis, then mixed in stoichiometric ratios to achieve 40 wt% SFO in the final composite. The combined solution underwent the same gel formation and combustion process, with the as-burnt powders annealed at 950 °C.
Structural and magnetic characterization of bare SFO and SFO-CFO magnetic nanoparticles are discussed in ref. 40. These samples are composed of interconnected particles forming aggregates: specifically, the morphology of SFO particles is characterized by a plate-like shape with a polycrystalline nature. The SFO-CFO system is characterized by an oriented growth relationship between SFO and CFO, with dislocations forming at the interface, indicating strong structural interactions.43
Morphological properties were studied by an Olympus BX53M polarized light optical microscope (PLOM). It is equipped with a THMS600 Linkam hot stage, a liquid nitrogen cooling system for temperature control, and an SC50 Olympus camera for recording micrographs. Morphological changes were determined employing 20 °C min−1 as cooling and heating rates, in which samples were crystallized and melted on a glass slide with a thin glass coverslip on top. In addition, an Olympus BX53 phase-contrast microscope (PCM) was used to observe the morphology of samples at room temperature.
Field-dependent magnetization M(H) loops were measured at 300 K using a vibrating sample magnetometer (VSM Model 10 – Microsense) in the magnetic field range ±2 T. The measurements were conducted to evaluate the magnetic response of the composites, including key parameters, such as coercivity (μ0HC), remanent magnetization (MR), and magnetization measured at 2 T (M2T).
![]() | ||
Fig. 1 DSC cooling curves at 20 °C min−1 of PCL composites with (a) NdFeB, (b) SFO, and (c) SFO-CFO magnetic micro- and nanoparticles in the composition range of 1–10%. |
![]() | ||
Fig. 2 DSC cooling curves at 20 °C min−1 of PBSA composites with (a) NdFeB, (b) SFO, and (c) SFO-CFO magnetic micro- and nanoparticles in the composition range of 1–10%. |
Fig. 1 shows DSC scans from the melt of selected samples. A main exothermic event corresponding to the PCL PMCs' crystallization is observed. The crystallization peaks contain a high-temperature shoulder at temperatures close to 30 °C or above, followed by the main peak. This is considered a bimodal crystallization, as two crystallization exotherms are overlapped. Fig. 2 shows the DSC cooling scans from the melt of PBSA composites. In this case, the bimodality of the crystallization exotherm is much more pronounced (with a high-temperature peak at approximately 40 °C and a low-temperature peak at approximately 10 °C). The exact values of all the calorimetric transitions recorded are reported in Tables S1–S12.† The higher temperature peaks (or shoulder) correspond to regions with better M(N)P dispersion, where the M(N)Ps effectively nucleate the polymer matrix. In contrast, the lower temperature peaks are associated with areas of M(N)P aggregation, where reduced surface contact with the polymer matrix hinders their nucleating effect.
Polymers crystallize by nucleation (or primary nucleation) and growth (or secondary nucleation, as crystals grow by successive nucleation on the previously formed primary nuclei) mechanism, where primary nuclei form and crystals can grow on them through secondary nucleation mechanisms. The nucleation of neat polymers is normally triggered by heterogeneities or impurities, like catalytic debris and others, i.e., the so-called heterogeneous nucleation. The crystallization temperature upon cooling from the melt is proportional to the active heterogeneous nucleation density. The higher the crystallization temperature, the higher the nucleation density, which implies a more efficient nucleation process. Therefore, if a foreign substance, like a filler, is added to a polymer matrix, the crystallization temperature will only increase if the filler has a higher nucleating efficiency than the heterogeneities available within the polymer.44,45
The bimodality in the DSC crystallization exotherms is related to different primary nucleation effects. The nucleation of the filler occurs by interactions of the polymer chains and the exposed surface area of the fillers. When the dispersion is heterogeneous, the well-dispersed particles can cause the maximum nucleation effects, as their surface area is much higher in comparison with aggregates. The fact that the bimodality is due to nucleation effects and not to other possible reasons (like the crystallization of two polymorphs) is also proven by the DSC subsequent heating scans (Fig. S8 and S9†), as they all show unimodal melting endotherms. In the case of PBSA, a small cold crystallization exotherm is also observed during the DSC heating scans, which correlates with a reorganization process during the heating scan.46
The melting temperature (Tm) in polymers depends on the lamellar thickness, and because of the metastable nature of polymeric crystals (for kinetic reasons, usually polymers crystallize in thinner lamellae, which are considered metastable from a thermodynamic point of view, as the equilibrium crystalline structures should be lamellae with thickness equivalent to extended chain crystals), large changes in crystallization temperature (Tc) would be needed to modify the Tm values. So, it is usual to find that samples with or without nucleating agents exhibit similar Tm or with minimal changes. The observed Tm of ∼56 °C for PCL and ∼88 °C for PBSA homopolymers are consistent with the values reported in the literature.15,20 Our PCL-based composites show relatively stable melting temperature values of 54.7–56.8 °C and PBSA-based composites of 84.8–87.6 °C largely independently of the type of filler particles, demonstrating that used M(N)Ps do not modify this parameter. The crystallinity degree (Xc) of the polymer in PMCs was evaluated, and the results are collected in Tables S1–S12.† As expected, the crystallinity degree did not change beyond the experimental error of the measurements. This suggests that the morphological structural features of the fillers employed here do not influence the crystallinity of the matrix in PMCs.
As explained above, it is worth underlining that M(N)Ps can act as nucleating agents. In particular, when the nucleation of added M(N)Ps is effective, a typical increase in crystallization temperature is detected, as the nucleation density is increased beyond that of the nucleating heterogeneities that the polymer already contains. To characterize this nucleating effect, the initial or “onset” crystallization temperature values (Tc,onset) was used. As shown in Fig. 3a, the Tc,onset of the PMCs increases compared to the PCL homopolymer (in black), although it remains almost constant beyond 1% loading. In contrast, Fig. 3b shows a nearly linear increase in the Tc,onset of PBSA-based PMCs with SFO (blue) and SFO-CFO (green) nanoparticles as their content increases. These variations are relatively low, and for C > 10%, no significant changes in Tc,onset were observed with increasing particle content (ESI, section S2†).
![]() | ||
Fig. 3 T c,onset values of PMCs of (a) PCL and (b) PBSA with three types of M(N)Ps (NdFeB in red, SFO in blue, and SFO-CFO in green) in the filler content range of 1–10%. |
Micrographs of neat PBSA and PBSA/NdFeB (C = 1%) composites (Fig. 5), obtained during cooling at 20 °C min−1 from the melt, reveal a similar morphology. There is little evidence of a nucleating effect from the M(N)Ps, as shown by the PBSA homopolymer micrograph (Fig. 5 upper panel), where the formation of some spherulites at 40 °C is observed. However, in selected areas (shown by black boxes in Fig. 5), smaller spherulites can be observed in the composite versus the neat sample at the same crystallization temperature. In this case, the presence of M(N)Ps is also evident (see large black aggregates of M(N)Ps in the top left-hand corner of the image), and the dispersion is poor and non-homogenous, negatively affecting the nucleating effect of the M(N)Ps.
The morphology of selected samples was also examined using PCM at room temperature. Fig. 6 presents micrographs of PCL composites containing NdFeB, SFO, and SFO-CFO M(N)Ps at C = 1%, 5% and 50%. At 1% loading, the M(N)Ps form aggregates, indicating poor dispersion within the polymeric matrix. At 5% loading, many M(N)Ps are dispersed throughout the PCL matrix; however, aggregates are still clearly present. At 50% loading, the excessive M(N)P content leads to the formation of large aggregates within the polymer matrix, which likely diminishes the nucleating effect of the M(N)Ps. Similar observations were made using PCM for PBSA composites (Fig. S10†) with the same M(N)P loadings as those in the PCL composites. In the PBSA matrix, aggregates were also visible at low M(N)P contents, and the size and dispersion issues became more pronounced as the filler content increased.
![]() | ||
Fig. 6 PCM micrographs at room temperature of PCL-based composites: (1) PCL/NdFeB, (2) PCL/SFO, and (3) PCL/SFO-CFO, each at (a) 1%, (b) 5%, and (c) 50% filler content. |
SFO nanoparticles, with their highest μ0HC (∼0.57 T), moderate M2T (∼66 Am2 kg−1), and high MR/M2T (∼0.53), provide excellent magnetic remanence, making them suitable for patterned magnetic structures or permanent magnetic components in polymer matrices.29,55,56 In contrast, SFO-CFO nanoparticles offer a balance between hard and soft magnetic behaviors, with slightly higher M2T (∼71 Am2 kg−1), moderate MR/M2T (∼0.47), and low μ0HC (∼0.22 T). This combination makes SFO-CFO ideal for biomedical applications, such as magnetic hyperthermia or flexible magnetic sensors, where moderate magnetization and reconfigurability are critical.21,22,57,58 Integrating these magnetic M(N)Ps into polymer matrices enables the creation of flexible, lightweight, and magnetically responsive composites.
To investigate the magnetic behavior in greater detail within the low-composition range (C = 1–10%), where the most significant changes (specifically step-like increase in Tc,onset for PCL) in polymer properties were observed, field-dependent magnetization measurements were performed (Fig. 9). It should be underlined that the error on magnetization measurements is higher in this range due to errors in measuring the low masses of M(N)Ps. The SFO and SFO-CFO composites exhibit major magnetic hysteresis loops that saturate below 2 T. This behavior is consistent with the lower anisotropy of these ferrite-based materials compared to NdFeB-based composites, which have higher anisotropy and thus require stronger fields for saturation. For PBSA/SFO composites, the MR/M2T ratio ranges from 0.54 to 0.59, while μ0HC remains steady at ∼0.53 T across the 1–10% filler loading range. In PBSA/SFO-CFO composites, the MR/M2T decreases from 0.56 at 1% to 0.51 at 10% and μ0HC remains relatively stable at ∼0.21 T. Overall, the magnetic properties of the PMCs are determined by the characteristics of the loaded M(N)Ps, so for NdFeB, which has the highest M2T among the powders, the resulting composites also display the highest overall magnetization.
This preliminary study offers a promising framework to develop these materials further, opening perspectives for applications in sustainable technologies. Refining particle dispersion strategies appears to be a key point to improve the material's performance. Finally, investigating multi-component systems could further enhance their performance and expand their functionality.
PMCs | Polymer-based magnetic composites |
M(N)Ps | Micro- or nanoparticles |
PCL | Poly(ε-caprolactone) |
PBSA | Poly(butylene succinate-ran-butylene adipate) |
PLA | Poly(lactic acid) |
REE | Rare earth elements |
NdFeB | Neodymium–iron–boron alloy |
SFO | Strontium ferrite |
SFO-CFO | Strontium ferrite/cobalt ferrite |
DSC | Differential scanning calorimetry |
PLOM | Polarized light optical microscopy |
PCM | Phase contrast microscopy |
VSM | Vibrating sample magnetometry |
Footnote |
† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5su00222b |
This journal is © The Royal Society of Chemistry 2025 |