Superconducting quantum interference device (SQUID) analysis for (α-Fe₂O₃) nanoparticles with tailored physical and magnetic properties Authors: Syed M. Hussain, Syed Mustansar Hussain, Hijaz Ahmad, Mustafa Bayram, Mohamed R. Eid, Assmaa Abd-Elmonem, Kamel Guedri Published: 2026 ABSTRACT This study presents the hydrothermal synthesis of hematite (α-Fe₂O₃) nanoparticles with controlled morphological variations, including rhombohedral, irregular, and Nano plate structures. The phase purity and crystallographic structure of the synthesized samples were validated using X-ray diffraction, confirming the successful formation of hematite. Detailed transmission electron microscopy analysis revealed pronounced differences in particle shape and size, emphasizing the effectiveness of the synthesis method in tailoring morphology. Magnetic properties were systematically investigated using a superconducting quantum interference device magnetometer, demonstrating a clear correlation between nanoparticle morphology and magnetic response. The results collectively show that particle morphology plays a substantial role in determining both the structural and magnetic behavior of hematite nanoparticles. These findings contribute to the rational design and engineering of nanomaterials for advanced technological applications. Introduction Magnetic iron oxide nanoparticles have attracted sustained interest because their magnetic response is highly sensitive to crystal phase, size and morphology. Magnetite (Fe₃O₄) has been the focus of a lot of research because of its strong magnetism and potential in the field of medicine. However, hematite (α-Fe₂O₃) is a different type of iron oxide that is not only very chemically stable but also low in toxicity and cheap, which makes it quite suitable for catalysis, sensing, energy conversion, as well as magnetic applications. When hematite is at the nanoscale level, it shows size- and shape-dependent magnetic properties, the Morin transition being suppressed or changed and weak ferromagnetism or superparamagnetic appearing. Being at the forefront of materials science and nanotechnology, magnetic nanoparticles synthesized with high accuracy possess the potential to revolutionize a broad range of applications. Examples of these are drug delivery, bio imaging, and therapeutic technologies for medical applications, purification of wastewater and removal of pollutants for the environment, and data storage via advanced technologies. It has been reported in recent work that significant achievements have been made in the development of nanoparticle-based biomedical devices, magnetic nanostructures for rapid purification of wastewater, and nanofiber magnetic materials for data storage and information transmission. On-going progress in the development of controlled synthesis methods facilitates the fine adjustment of nanoparticle size, shape, and magnetic performance, which are crucial aspects to be considered for maximizing their efficiency in these new technological ‍‌‍‍‌‍‌‍‍‌applications. Chen‍‌‍‍‌‍‌‍‍‌ et al. [1] fabricated α-Fe₂O₃ nanowires by nano casting and studied how the different modes of their assembly affect the magnetic interactions and the gas-sensing performance. Akhavan and Azimirad [2] found that Fe₂O₃ nanograin chains coated with a TiO₂ nanolayer exhibited a high photocatalytic activity for visible-light. Sadeghi-Niaraki et al. [3] created Fe₂O₃@TiO₂ nanostructured pigments which are characterized with high near-infrared reflectance and excellent photocatalytic performance. Zhang et al. [4] illustrated that porous Fe₂O₃ nanoparticles have the potential to act as high-performance electrochemical anode materials for lithium-ion batteries. Moradi et al. [5] prepared magnetite/dextran-functionalized graphene oxide nanosheets for improved positive contrast in in-vivo magnetic resonance imaging. Tadić et al. [6] have made a comprehensive study of hydrothermal synthesis of α-Fe₂O₃ nanoparticles and the relationship among morphology, magnetic properties, and cytotoxicity. Colombo et al. [7] studied how the hydrothermal treatment factors affect the particle size, shape, and colloidal stability of hematite nanoparticles. Ge et al. [8] achieved morphology-controlled hydrothermal synthesis of α-Fe₂O₃ and demonstrated its improved photocatalytic activity for Cr(VI) reduction. Hematite nanoparticles have excellent air stability, strong corrosion resistance, low toxicity, low cost and are environmentally friendly. Hematite nanoparticles are less hazardous during synthesis and more stable than other iron polymorphs. They are used in potential applications, such as gas sensors, catalysis, pigments, lithium-ion batteries, solar cells, magnetic and antibacterial materials. To further improve the physical properties of hematite nanoparticles, it is mandatory to have a uniform shape of nanoparticles and a very narrow particle size distribution [9]. There are some studies found on the synthesis of hematite nanoparticles done by other groups: Direct metal oxidation, iron precursor hydrolysis, and phase transition from other iron oxide phases to the hematite phase are the three main methods used to create hematite nanoparticles. The processes of oxidation, dihydroxylation (loss of OH), and dehydration (loss of H2O) may result in the phase transition of any other iron phase to the hematite phase [10]. Several techniques have been devised to create different types of nanoscale hematite structures, including methodologies such as precipitation, sol-gel, mechanochemical, thermal decomposition, solvothermal, and hydrothermal [11–14]. One of the most sensitive methods for detecting very weak magnetic signals is magnetic measurement with a SQUID magnetometer. The principle of operation for SQUID is the phenomenon of quantum interference occurring in superconducting loops with Josephson junctions, which allows for the detection of very small changes in magnetic flux. The extremely high sensitivity of SQUID magnetometers makes them an ideal tool for studying the magnetic characteristics of nanoscale materials, such as saturation magnetization, magnetic phase transitions, and temperature-dependent magnetic behavior. In the field of nanoparticles, magnetic characterization by SQUID is particularly useful for: differentiating the superparamagnetic, ferromagnetic, and weak ferromagnetic behaviors, determining transition temperatures quite accurately, like the Morin transition in hematite nanoparticles. Zhu et al. organized a cantaloupe-like hematite framework through hydrothermal fabrication, which exhibited high coercivity at room temperature with an HC value of 2279 [15]. Additionally, they also produced a silkworm pod-like hematite superstructure with high coercivity through hydrothermal synthesis, with an HC value of 3315.5 Oe [15]. Jayanthi et al. [16] described weak ferromagnetic characteristics of hematite nanoparticles with a size of approximately 50 nm, which exhibited a saturation magnetization of Ms = 1.899 emu/g, a remanent magnetization of Mr = 0.659 emu/g, and a coercivity of HC = 414.98 Oe at a temperature of 300 K. By decreasing the particle size, the Morin temperature of hematite nanoparticles can also be reduced, eventually disappearing for particles smaller than approximately 10 nm [12, 13, 17, 18]. For small hematite nanoparticles, specifically those below 10 nm, superparamagnetic behaviour properties have been observed [7, 19, 20]. Lee and Kwak [21] acquired tubular superstructures composed of superparamagnetic hematite nanoparticles (around 10 nm) from the pyrolysis of metal-organic frameworks in a confined space. Li et al. [22] stated that the high feature ratio improves shape anisotropy, which brings large magnetic coercivity. In recent studies, it has been investigated how inter-particle interaction affects the magnetic characteristics of hematite nanoparticles. It has been found that the strength of interaction between the particles plays a crucial role in the superparamagnetic relaxation of the particles [23, 24]. Long-range dipole-dipole interactions may also affect the magnetic properties in ferromagnetic or ferri-magnetic systems, even if the magnetic particles are dispersed in a non-magnetic solid matrix [25]. However, for hematite nanoparticles, the dipole-dipole interaction is generally weak, and it is believed that exchange interactions between surface atoms of the nanoparticles in direct contact have a significant influence on their magnetic characteristics [26]. Ganure et al. [27] prepared Ni²⁺– and In³⁺-doped cobalt ferrites and structurally, magnetically and catalytically tested them for Bis-(indolyl)methane derivative ‍‌‍‍‌‍‌‍‍‌synthesis.‍ Ganure et al. [28] found a green method of making pyrimidine derivatives using a magnetically recoverable and reusable catalyst. Dhale‍‌‍‍‌‍‌‍‍‌ et al. [29] have revealed that reusable La³⁺-doped mixed ferrite nanocrystals could be GE magnetic catalysts for the green synthesis of triaryl imidazole derivatives in one-pot and without any solvent. Singh and Sharma [30] studied the magnetic behavior improvements of Er³⁺-doped Co–Mn ferrite nanoparticles for magnetic device applications. Singh and Sharma [31] conducted a study on the Improved magnetic features of Gd3+ doped Co-Mn ferrite nanoparticles for data storage applications requiring high density. They disclosed the Gd³⁺-doped Co–Mn ferrite nanoparticles with superior magnetic properties for high-density data storage ‍‌‍‍‌‍‌‍‍‌applications. Previous studies have demonstrated that hematite can be synthesized in diverse morphologies such as nanoplates, rhombohedral particles and irregular structures using hydrothermal and related methods. These works have largely focused on either morphological control or qualitative magnetic characterization. Nevertheless, while individual shapes have been reported extensively, there is no systematic and comparative study explaining how different morphologies, under similar synthesis conditions, affect the Morin transition temperature, thermal hysteresis, and saturation magnetization. Especially, the extent to which particle anisotropy and dimensional constraints determine the magnetic transitions measured through the same SQUID protocols has not been clearly understood. This research therefore seeks to fill this void by performing a controlled hydrothermal synthesis of α-Fe₂O₃ nanoparticles with clearly rhombohedral, nanoplate, and irregular shapes, followed by a direct and morphology-resolved SQUID magnetic comparison. By connecting the in-depth TEM-based structural data with the field-cooled/zero-field-cooled magnetic measurements, the present article tries to elucidate in what way shape anisotropy and particle size together influence the magnetic properties of hematite nanoparticles. Novelty of the study The key original contributions of this work are as follows: A detailed, morphology-based comparison of rhombohedral, nanoplate, and irregular α-Fe₂O₃ nanoparticles synthesized through hydrothermal methods under similar conditions.Establishment of a direct link between particle morphology and Morin transition characteristics, accompanied by a quantitative assessment of transition temperature variations and thermal hysteresis through SQUID experiments.Validation through experiments is that the thickness of the nanoplate is the leading factor in the increase of thermal hysteresis and the stabilization of weak ferromagnetism, aside from the effects of particle size.Evidence through experiments that the effects of shape anisotropy and dimensional confinement can be so strong as to negate the simple size-dependent trends in saturation magnetization for hematite nanoparticles in a resin ‍‌‍‍‌‍‌‍‍‌matrix. A detailed, morphology-based comparison of rhombohedral, nanoplate, and irregular α-Fe₂O₃ nanoparticles synthesized through hydrothermal methods under similar conditions. A detailed, morphology-based comparison of rhombohedral, nanoplate, and irregular α-Fe₂O₃ nanoparticles synthesized through hydrothermal methods under similar conditions. Establishment of a direct link between particle morphology and Morin transition characteristics, accompanied by a quantitative assessment of transition temperature variations and thermal hysteresis through SQUID experiments. Establishment of a direct link between particle morphology and Morin transition characteristics, accompanied by a quantitative assessment of transition temperature variations and thermal hysteresis through SQUID experiments. Validation through experiments is that the thickness of the nanoplate is the leading factor in the increase of thermal hysteresis and the stabilization of weak ferromagnetism, aside from the effects of particle size. Validation through experiments is that the thickness of the nanoplate is the leading factor in the increase of thermal hysteresis and the stabilization of weak ferromagnetism, aside from the effects of particle size. Evidence through experiments that the effects of shape anisotropy and dimensional confinement can be so strong as to negate the simple size-dependent trends in saturation magnetization for hematite nanoparticles in a resin ‍‌‍‍‌‍‌‍‍‌matrix. Evidence through experiments that the effects of shape anisotropy and dimensional confinement can be so strong as to negate the simple size-dependent trends in saturation magnetization for hematite nanoparticles in a resin ‍‌‍‍‌‍‌‍‍‌matrix. Materials and methods Synthesis of hematite (α-Fe₂O₃) nanoparticles Hematite nanoparticles were synthesized using a hydrothermal method with controlled variation of precursor chemistry, solvent composition, and additive type to obtain three different morphologies. Three representative samples were selected for detailed structural and magnetic characterization: sm_14, sm_24, and sm_41. The synthesis parameters are summarized in Table 1. For each synthesis, the required iron precursor, nucleating agent, and morphology-directing additive were dissolved in the selected solvent system under magnetic stirring until a homogeneous precursor solution was obtained. The prepared solution was then transferred into a 50 mL Teflon-lined stainless-steel autoclave. The autoclave was sealed and heated under hydrothermal conditions according to the temperature and reaction time listed in Table 1. After completion of the reaction, the autoclave was allowed to cool naturally to room temperature. The obtained reddish-orange precipitate was collected by centrifugation and washed repeatedly to remove residual ions, unreacted precursors, and organic additives. The washing process was carried out using distilled water and ethanol, followed by acetone when required. Each washing step was performed by centrifugation at 6500 rpm for 15 min. The purified product was then dried and stored for subsequent TEM, XRD, and SQUID characterization. The dried product masses recovered from the selected hydrothermal synthesis batches are listed in Table 2. Table 1Key hydrothermal synthesis parameters for selected α-Fe₂O₃ samplesSampleMorphologyPrecursorAdditive systemSolventConditionsm_14IrregularFeCl₃·6 H₂O = 4 mmolNaAc = 40 mmol; PVP = 1 gH₂O = 30 mL200 °C, 18 hsm_24RhombohedralFe(acac)₃ = 3.2 mmolNaOH = 9.6 mmol; oleic acid = 20 mLH₂O = 10 mL180 °C, 42 hEtOH = 22 mLsm_41NanoplateFeCl₃·6 H₂O = 1 mmolNaAc = 10 mmolEtOH = 13 mL180 °C, 24 hNaAc = sodium acetate; Fe(acac)₃ = iron (III) acetylacetonate; EtOH = ethanol. For sm_41_B2, no additional water was added; the limited water content originated mainly from the hydrated iron precursor Table 1Key hydrothermal synthesis parameters for selected α-Fe₂O₃ samplesSampleMorphologyPrecursorAdditive systemSolventConditionsm_14IrregularFeCl₃·6 H₂O = 4 mmolNaAc = 40 mmol; PVP = 1 gH₂O = 30 mL200 °C, 18 hsm_24RhombohedralFe(acac)₃ = 3.2 mmolNaOH = 9.6 mmol; oleic acid = 20 mLH₂O = 10 mL180 °C, 42 hEtOH = 22 mLsm_41NanoplateFeCl₃·6 H₂O = 1 mmolNaAc = 10 mmolEtOH = 13 mL180 °C, 24 hNaAc = sodium acetate; Fe(acac)₃ = iron (III) acetylacetonate; EtOH = ethanol. For sm_41_B2, no additional water was added; the limited water content originated mainly from the hydrated iron precursor Key hydrothermal synthesis parameters for selected α-Fe₂O₃ samples NaAc = sodium acetate; Fe(acac)₃ = iron (III) acetylacetonate; EtOH = ethanol. For sm_41_B2, no additional water was added; the limited water content originated mainly from the hydrated iron precursor Table 2Obtained dried product mass from selected hydrothermal synthesis batchesSampleMorphologyObtained dried product masssm_14Irregular nanoparticles46 mgsm_24Rhombohedral nanoparticles28 mgsm_41Nanoplates39 mg Table 2Obtained dried product mass from selected hydrothermal synthesis batchesSampleMorphologyObtained dried product masssm_14Irregular nanoparticles46 mgsm_24Rhombohedral nanoparticles28 mgsm_41Nanoplates39 mg Obtained dried product mass from selected hydrothermal synthesis batches Product mass refers to the recovered dried powder obtained from a single hydrothermal synthesis batch after washing and drying. A schematic representation of hydrothermal synthesis, product recovery, washing, and sample-preparation procedure is shown in Fig. 1. Transmission electron microscopy was used to investigate the morphology, particle-size distribution, and crystallinity of the selected α-Fe₂O₃ nanoparticles. Three representative samples were analyzed: sm_24 with rhombohedral morphology, sm_41 with nanoplate morphology, and sm_14 with irregular morphology. Particle-size analysis was performed using Gatan Digital Micrograph software by measuring representative particles from TEM micrographs. The results are reported as mean ± standard deviation. Fig. 1Schematic workflow of the hydrothermal synthesis, purification, and sample preparation procedure for α-Fe₂O₃ nanoparticles. The authors using Adobe Illustrator (version 29.0, Adobe Inc., https://www.adobe.com/products/illustrator.html) created this original schematic Schematic workflow of the hydrothermal synthesis, purification, and sample preparation procedure for α-Fe₂O₃ nanoparticles. The authors using Adobe Illustrator (version 29.0, Adobe Inc., https://www.adobe.com/products/illustrator.html) created this original schematic Figure 2 shows the TEM characterization of the rhombohedral α-Fe₂O₃ nanoparticles obtained from sample sm_24. The low-resolution TEM image, Fig. 2(a), shows that the particles mainly exhibit well-defined rhombohedral morphology with moderate dispersion. Although the sample is not completely mono-dispersed, most particles retain a similar faceted shape. The HRTEM image in Fig. 2(b) confirms the crystalline nature of the particles, while the magnified HRTEM image in Fig. 2(c) shows clear lattice fringes. The measured interplanar spacings are approximately 0.25 nm and 0.36 nm, corresponding to the hematite lattice planes indicated in the figure. The FFT pattern in Fig. 2(d) further confirms the crystalline structure of the rhombohedral particles. The particle-size distribution shown in Fig. 2(e) gives an average particle size of 32 ± 4 nm, indicating a relatively narrow distribution for sm_24. Figure‍‌‍‍‌‍‌‍‍‌ 2 illustrates the TEM investigation of the rhombohedral α-Fe₂O₃ nanoparticles that were isolated from sm_24 sample. The low-resolution TEM in Fig. 2(a) indicates that the nanoparticles exhibit distinctly rhombohedral forms and are moderately distributed. Even though the sample is not completely mono-dispersed, majority of the particles possess a similar faceted shape. The HRTEM image, Fig. 2(b), verifies that the particles are crystalline. The brightly focused HRTEM shown in Fig. 2(c) reveals the lattice fringes very distinctly. The interplanar spacings that have been determined are around 0.25 nm and 0.36 nm, which align with the hematite lattice planes that are identified in the figure. Figure 2(d) FFT pattern adds another layer of support for the crystallinity of the rhombohedral particles. Particle size distribution in Fig. 2(e) indicates an average particle size of 32 ± 4 nm, suggesting a fairly narrow size distribution of sm_24 sample. Sample sm_41 is a nanoplate α-Fe₂O₃ sample, and its TEM analysis is presented in Fig. 3. The LRTEM image, Fig. 3(a), illustrates the morphology of the plate-like particles. These nanoplates’ thickness and lateral size have been estimated from their face-on orientation, edge-on, and the partially stacked ones that show the particles’ two dimensions (size of the plane and thickness). The hematite structure is well supported by the observation of clearly resolved lattice fringes made evident by a HRTEM image in Fig. 3(b) and the lattice image whose magnification is provided in Fig. 3(c). The indexed diffraction spots in the FFT pattern in Fig. 3(d) provide additional evidence for the nanoplates’ crystallinity. Statistic evaluation shows an average thickness of the nanoplates of 22 ± 5 nm and average lateral diameter of 109 ± 12 nm as indicated in Figs. 4(e) and 4(f), respectively. These findings verify that sm_41 is made of anisotropic nanoplates that have a large lateral-size-to-thickness ratio. Figure 4 displays TEM results of irregular α-Fe₂O₃ nanoparticles from sample sm_14. Low magnification LRTEM image, Fig. 4(a), discloses irregularly shaped particles possessing a wider range of morphologies than sm_24 and sm_41. Internal contrast variations and non-uniform outlines of some particles may be indicative of local thickness differences, structural inhomogeneity, or surface features. The HRTEM image in Fig. 4(b) is a signal that the particles are single crystals, and the magnified HRTEM image in Fig. 4(c) at the same time shows lattice fringes with an interplanar distance of approximately 0.27 nm. The FFT image Fig. 5(d) consists of reflections that are indexed and correspond to hematite planes, thereby the sample has a crystalline phase. Particle size distribution by histogram in Fig. 4(e) results in a mean particle size of 53 ± 4 nm. It is true that majority of particles are centered around this size, but it should also be noted that there are some bigger ones, resulting in a less uniform size distribution compared to the rhombohedral ‍‌‍‍‌‍‌‍‍‌sample. Fig. 2TEM characterization of rhombohedral α-Fe₂O₃ nanoparticles from sample sm_24: (a) LRTEM image, (b) HRTEM image, (c) zoomed HRTEM image with indexed lattice spacings, (d) FFT pattern, (e) particle-size distribution histogram, and schematic representation of the indexed rhombohedral morphology TEM characterization of rhombohedral α-Fe₂O₃ nanoparticles from sample sm_24: (a) LRTEM image, (b) HRTEM image, (c) zoomed HRTEM image with indexed lattice spacings, (d) FFT pattern, (e) particle-size distribution histogram, and schematic representation of the indexed rhombohedral morphology Fig. 3TEM characterization of α-Fe₂O₃ nanoplates from sample sm_41: (a) LRTEM image, (b) HRTEM image, (c) zoomed HRTEM image showing lattice spacing, (d) FFT pattern with indexed crystallographic planes, (e) nanoplate thickness distribution histogram, and (f) nanoplate diameter distribution histogram TEM characterization of α-Fe₂O₃ nanoplates from sample sm_41: (a) LRTEM image, (b) HRTEM image, (c) zoomed HRTEM image showing lattice spacing, (d) FFT pattern with indexed crystallographic planes, (e) nanoplate thickness distribution histogram, and (f) nanoplate diameter distribution histogram Fig. 4TEM characterization of irregular α-Fe₂O₃ nanoparticles from sample sm_14: (a) LRTEM image, (b) HRTEM image, (c) zoomed HRTEM image showing lattice spacing, (d) FFT pattern with indexed crystallographic planes, and (e) particle-size distribution histogram TEM characterization of irregular α-Fe₂O₃ nanoparticles from sample sm_14: (a) LRTEM image, (b) HRTEM image, (c) zoomed HRTEM image showing lattice spacing, (d) FFT pattern with indexed crystallographic planes, and (e) particle-size distribution histogram XRD X-ray diffraction was used to identify the crystalline phase and estimate the crystallite size of the selected α-Fe₂O₃ nanoparticle samples. The XRD patterns of sm_14, sm_41, and sm_24 are shown in Fig. 5. These samples correspond to irregular nanoparticles, nanoplates, and rhombohedral nanoparticles, respectively. The diffraction peaks of all three samples are mainly indexed to the hematite α-Fe₂O₃ phase. The characteristic diffraction peaks observed at approximately 2θ = 24.12°, 33.11°, 35.61°, 40.82°, 49.41°, 54.00°, 57.50°, 62.38°, and 63.96° correspond to the (012), (104), (110), (113), (024), (116), (018), (214), and (300) crystallographic planes of hematite, respectively. The three most intense hematite reflections are located near 33.11°, 35.61°, and 49.41°, corresponding to the (104), (110), and (024) planes. These results confirm the successful formation of crystalline α-Fe₂O₃ in the selected samples. For samples sm_24 and sm_41, no additional impurity peaks were detected within the measured range, indicating that these samples are predominantly composed of the hematite phase. In contrast, sample sm_14 shows weak additional peaks marked with the symbol (*) in Fig. 5. These additional reflections are assigned to sodium chloride, with peaks observed at approximately 2θ = 31.70°, 45.45°, and 75.28°. The presence of NaCl in sm_14 is most likely related to residual salt remaining after the washing process. Therefore, sm_14 is described as hematite containing a minor NaCl impurity rather than a completely predominantly hematite phase sample. The crystallinity values calculated from the XRD data were 71% for sm_24, 69% for sm_41, and 69% for sm_14, as summarized in Table 3. Among the three selected samples, sm_24 showed the highest crystallinity. However, because the XRD samples were prepared by drop-casting nanoparticle dispersions onto glass substrates, the calculated crystallinity values should be interpreted as comparative estimates rather than absolute crystallinity values. Film thickness, particle packing, preferred orientation, and background contribution from the glass substrate may influence the calculated values. Fig. 5XRD patterns of α-Fe₂O₃ nanoparticles with different morphologies: sm_14 irregular nanoparticles, sm_41 nanoplates, and sm_24 rhombohedral nanoparticles. The peaks marked with (*) correspond to the hematite α-Fe₂O₃ phase, while the peaks marked with (#) indicate the NaCl impurity phase observed in sm_14 XRD patterns of α-Fe₂O₃ nanoparticles with different morphologies: sm_14 irregular nanoparticles, sm_41 nanoplates, and sm_24 rhombohedral nanoparticles. The peaks marked with (*) correspond to the hematite α-Fe₂O₃ phase, while the peaks marked with (#) indicate the NaCl impurity phase observed in sm_14 Table 3TEM-based morphological parameters and average particle dimensions of selected α-Fe₂O₃ nanoparticles, including rhombohedral nanoparticles, nanoplates and irregular nanoparticlesSampleMorphologyMeasured parameterAverage sizesm_24Rhombohedral nanoparticlesParticle diameter32 ± 4 nmsm_41NanoplatesPlate thickness22 ± 5 nmsm_41NanoplatesPlate diameter109 ± 12 nmsm_14Irregular nanoparticlesParticle diameter53 ± 4 nm Table 3TEM-based morphological parameters and average particle dimensions of selected α-Fe₂O₃ nanoparticles, including rhombohedral nanoparticles, nanoplates and irregular nanoparticlesSampleMorphologyMeasured parameterAverage sizesm_24Rhombohedral nanoparticlesParticle diameter32 ± 4 nmsm_41NanoplatesPlate thickness22 ± 5 nmsm_41NanoplatesPlate diameter109 ± 12 nmsm_14Irregular nanoparticlesParticle diameter53 ± 4 nm TEM-based morphological parameters and average particle dimensions of selected α-Fe₂O₃ nanoparticles, including rhombohedral nanoparticles, nanoplates and irregular nanoparticles For sm_24 and sm_14, crystallite size was estimated using the Scherrer equation: \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$D=~\frac{{K\lambda }}{{\beta \sin \theta }},$$\end{document} where D is the crystallite size, K is the Scherrer shape factor, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\lambda$$\end{document} is the X-ray wavelength, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\beta$$\end{document} is the full width at half maximum, and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\theta$$\end{document} is the Bragg angle. In this work, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$K=0.9$$\end{document} K and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\lambda =0.154$$\end{document} nm was used for the calculation. For sm_24, the crystallite size calculated from the hematite (104) peak near 2θ = 33.18° was approximately 28 nm, which is slightly smaller than the TEM particle size of 32 ± 4 nm. For sm_14, the crystallite size calculated from the hematite (104) reflection was approximately 36 nm, while the TEM particle size was 53 ± 4 nm (Table 4). The smaller crystallite sizes obtained from XRD compared with TEM are expected because XRD estimates the size of coherently diffracting crystalline domains, whereas TEM measures the physical particle dimensions. Surface disorder, internal defects, strain, and the presence of multiple crystalline domains within a particle can reduce the apparent crystallite size measured by XRD. For the nanoplate sample sm_41, the crystallite size was not reported. Because nanoplates have strong anisotropic geometry, the peak broadening can be affected by preferred orientation, direction-dependent coherent domain size, and drop-casting-induced texture. Therefore, applying a single Scherrer crystallite-size value to sm_41 may be misleading. The nanoplate dimensions were instead evaluated directly from TEM analysis, which gave a plate thickness of 22 ± 5 nm and a lateral diameter of 109 ± 12 nm. Overall, the XRD results confirm that sm_24 and sm_41 are predominantly crystalline α-Fe₂O₃, while sm_14 consists mainly of α-Fe₂O₃ with a minor NaCl impurity. For sm_24 and sm_14, the XRD crystallite sizes are smaller than the corresponding TEM particle dimensions, consistent with the difference between coherent crystalline domain size and physical particle size. For sm_41, TEM analysis provides a more reliable description of the anisotropic nanoplate dimensions. These structural results support the TEM observations and provide the crystallographic basis for comparing the magnetic behavior of the three selected morphologies. Table 4Showing the hematite phase planes, d-spacing values and peak position 2θHematite plane (hkl)Angle (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$2\theta$$\end{document} in degrees)d-spacing (Å)(012)24.123.68(104)33.112.70(110)35.612.51(006)39.212.29(113)40.822.20(202)43.482.07(024)49.411.84(116)541.69(018)57.51.60(214)62.381.48(300)63.961.45(208)69.51.35(1010)71.821.31(220)75.41.25(306)77.661.22 Table 4Showing the hematite phase planes, d-spacing values and peak position 2θHematite plane (hkl)Angle (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$2\theta$$\end{document} in degrees)d-spacing (Å)(012)24.123.68(104)33.112.70(110)35.612.51(006)39.212.29(113)40.822.20(202)43.482.07(024)49.411.84(116)541.69(018)57.51.60(214)62.381.48(300)63.961.45(208)69.51.35(1010)71.821.31(220)75.41.25(306)77.661.22 Showing the hematite phase planes, d-spacing values and peak position 2θ Superconducting quantum interference device (SQUID) technique SQUID magnetometry was used to investigate the magnetic response and Morin transition behavior of the selected α-Fe₂O₃ nanoparticle samples. Three representative samples were measured: sm_24 rhombohedral nanoparticles, sm_41 nanoplates, and sm_14 irregular nanoparticles. A pure resin sample was also prepared and measured under the same conditions to evaluate the diamagnetic contribution from the resin matrix and sample holder. For SQUID sample preparation, the dried nanoparticles were first dispersed in ethanol to improve particle separation and handling. The dispersion was placed in a Petri dish and left overnight to allow complete evaporation of ethanol. After drying, the nanoparticle powder was collected and mixed with a polymer resin and hardener. The hardener-to-polymer ratio was maintained at 25:75. The nanoparticles were added to the polymer before curing, and the total mass of the nanoparticle–polymer mixture was recorded. The mixture was homogenized using a Miccra D1 homogenizer to improve particle distribution inside the resin matrix. After homogenization, the nanoparticle–resin mixture was transferred into a plastic capsule using a dropper. The filled capsule was weighed, and the nanoparticle mass inside the capsule was calculated from the nanoparticle-to-polymer mixing ratio and the total filled capsule mass. The capsule was then left to dry and cure overnight before SQUID measurement. The magnetization values reported in this work were normalized with respect to the nanoparticle mass in the capsule. The capsule compositions used for SQUID measurements are summarized in Table 5. The use of a resin matrix allowed the nanoparticles to be mechanically stabilized inside the capsule during SQUID measurement. However, because the nanoparticle loading was low compared with the total resin mass, the measured magnetic signal contained both the nanoparticle contribution and a diamagnetic background contribution from the polymer and capsule. Therefore, the pure resin reference sample was measured separately and used to assess the background response. Table 5Capsule composition used for SQUID magnetometrySampleTotal capsule mass (g)Nanoparticle mass in capsule (mg)Polymer mass in capsule (g)sm_410.15222.080.1501sm_240.18412.380.1817sm_140.13433.750.1305Resin reference0.1825—0.1825 Table 5Capsule composition used for SQUID magnetometrySampleTotal capsule mass (g)Nanoparticle mass in capsule (mg)Polymer mass in capsule (g)sm_410.15222.080.1501sm_240.18412.380.1817sm_140.13433.750.1305Resin reference0.1825—0.1825 Capsule composition used for SQUID magnetometry SQUID data measurement method Magnetic measurements were performed using a SQUID magnetometer. Magnetization versus magnetic field, M–H, measurements were carried out at 300 K over a magnetic-field range of − 7 T to + 7 T. These measurements were used to examine the room-temperature magnetic response and estimate the saturation magnetization, Ms, of each sample. Temperature-dependent magnetization measurements were carried out from 10 K to 350 K using three measurement protocols: zero-field-cooled warming (ZFC), field-cooled warming (FCW), and field-cooled cooling (FCC). All temperature-dependent measurements were performed under an applied magnetic field of 50 Oe. For the ZFC measurement, the sample was first heated to 350 K in zero magnetic field and held at this temperature for 5 min to stabilize the thermal state. The sample was then cooled to 10 K without applying an external magnetic field. After reaching 10 K, a magnetic field of 50 Oe was applied, and the magnetization was recorded during warming from 10 K to 350 K. For the FCW measurement, the sample was cooled to 10 K under an applied field of 50 Oe, and the magnetization was recorded during warming from 10 K to 350 K under the same field. For the FCC measurement, the sample was cooled from 350 K to 10 K under 50 Oe, and the magnetization was recorded during cooling. The FCW and FCC measurements were used to examine the thermal hysteresis around the Morin transition. The Morin transition temperature \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${T_m}$$\end{document}, was identified from the abrupt change in the temperature-dependent magnetization curves. The transition temperature during warming is denoted as \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${T_m}\left( W \right)$$\end{document}, and the transition temperature during cooling is denoted as \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${T_m}\left( C \right)$$\end{document}. The thermal hysteresis, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{{\Delta}}}{T_m}$$\end{document}, was calculated using: \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${T_m}={T_m}\left( W \right) - {T_m}\left( C \right)$$\end{document} where \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${T_m}\left( W \right)$$\end{document} is the Morin transition temperature obtained from the field-cooled warming curve and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${T_m}\left( C \right)$$\end{document} is the Morin transition temperature obtained from the field-cooled cooling curve. SQUID data evaluation method The M–H curves were used to evaluate the room-temperature magnetic behavior of the nanoparticles. The measured magnetic response included two contributions: the magnetic contribution from α-Fe2O3 nanoparticles and the diamagnetic contribution from the resin matrix, capsule, and sample holder. The pure resin reference sample was used to identify the background response. The high-field linear component in the M–H curves was attributed mainly to the diamagnetic background, while the nonlinear low-field part was associated with the nanoparticle magnetic response. The saturation magnetization, Ms, was estimated from the M–H curves after considering the high-field background contribution. The obtained values were normalized to the nanoparticle mass inside each capsule and are therefore reported in emu g⁻¹ of nanoparticles. The Morin transition temperature was evaluated from the ZFC, FCW, and FCC magnetization curves. First, the temperature range containing the sharp magnetic transition was identified from the magnetization curve. Then, the first derivative of the magnetization with respect to temperature was calculated to locate the transition more accurately. The derivative peak was fitted using a Gaussian function, and the temperature corresponding to the maximum of the fitted peak was assigned as the Morin transition temperature. This derivative-based approach was used because the Morin transition appears as a sharp slope change in the magnetization curve rather than as a simple maximum or minimum. The Gaussian fit was used to reduce uncertainty in selecting the peak position from the derivative curve. The transition temperatures obtained from ZFC, FCW, and FCC measurements were then compared for each sample. SQUID results Figure 6 illustrates the SQUID magnetometry analysis of rhombohedral α-Fe₂O₃ particles from sample sm_24. The magnetization vs. magnetic-field (M–H) curve at 300 K is presented in Fig. 6(a). This curve has the general shape of the letter S but has a small high-field linear component. The nonlinear low-field part is believed to represent the magnetic behavior of the hematite nanoparticles, while the high-field linear component is mostly linked to the diamagnetic background from the resin matrix and sample holder. After adjusting to the nanoparticle mass, the saturation magnetization of sm_24 was roughly in the order of 0.24 emu g⁻¹. Figure 6(b) displays ZFC and FCW magnetization curves of sm_24. Magnetization slightly below and above Morin transition is distinctly indicated by both curves. The transition is close to about 214 K. FCW and FCC at 50 Oe are illustrated in Fig. 6(c). FCW first peak corresponds to transition at around 214 K while FCC shows nearly 212 K transition, the difference being a thermal hysteresis of about 2 K. Morin transition temperature was also estimated from the derivative curves and Gaussian fits in Fig. 6(d) to be around 214 K for sm_24. Figure 7 shows the SQUID magnetometry results of α-Fe₂O₃ nanoplates from sample sm_41. The M–H graph in Fig. 7(a) displays a small ferromagnetic-like signal and a minor hysteresis loop. Additionally, a high-field linear response is present, and this is thought to be caused by the diamagnetic signal of the resin and sample holder. After adjusting for the nanoparticle weight, the saturation magnetization of sm_41 comes out to be around 0.25 emu g⁻¹. Shown in Fig. 7(b) are the ZFC and FCW magnetization curves of sm_41. There is a rather evident transition in the Morin transition region. Based on the derivatives, it was roughly about 241 K from ZFC and about 239 K from FCW. FCW and FCC plots are shown in Fig. 7(c). The transition is indicated by FCW at about 239 K and by FCC at around 232 K. Hence, thermal hysteresis of sm_41 being about 7 K turns out to be the highest amongst the three samples considered. Determination of the transition temperatures via derivative curves and Gaussian fits is presented in Figure ‍‌‍‍‌‍‌‍‍‌7(d). Figure‍‌‍‍‌‍‌‍‍‌ 8 presents the results from SQUID magnetometry analysis of irregular α-Fe₂O₃ nanoparticles from sample sm_14. A magnetization vs. magnetic field (M–H) curve at 300 K is illustrated in Fig. 8(a). The plot reveals only a weak magnetic response, also displaying a diamagnetic background component. In fact, the low field region reveals a faintly ferromagnetic-like contribution attributed to the hematite nanoparticles. The saturation magnetization of sm_14 was evaluated at ~ 0.10 emu g⁻¹, a value which is even less than those for sm_24 and sm_41. Figure 8(b) displays the ZFC and FCW magnetization curves for sm_14. The ZFC magnetization indicates the transition point around 239 K, whereas the FCW curve exhibits a transition point around 238 K. Figure 8(c) shows the FCW and FCC magnetization curves taken at 50 Oe. The transition in the FCW curve occurs at approximately 238 K, while the transition in the FCC curve occurs at approximately 237 K. The thermal hysteresis obtained is about 1 K. The derivative plots and their Gaussian approximations displayed in Fig. 8(d) substantiate the Morin transition temperature of sm_14 to be around 239 ‍‌‍‍‌‍‌‍‍‌K. Fig. 6SQUID magnetometry results for rhombohedral α-Fe₂O₃ nanocrystals of sample sm_24: (a) 300 K M–H curve, (b) ZFC and FCW curves under 50 Oe from 10 K to 350 K, (c) FCW and FCC curves indicating thermal hysteresis near the Morin transition, and (d) Morin transition temperature determination by Gaussian fitting of derivative curves SQUID magnetometry results for rhombohedral α-Fe₂O₃ nanocrystals of sample sm_24: (a) 300 K M–H curve, (b) ZFC and FCW curves under 50 Oe from 10 K to 350 K, (c) FCW and FCC curves indicating thermal hysteresis near the Morin transition, and (d) Morin transition temperature determination by Gaussian fitting of derivative curves Fig. 7SQUID magnetometry results for α-Fe₂O₃ nanoplates from sample sm_41: (a) M–H curve measured at 300 K, (b) ZFC and FCW curves measured from 10 K to 350 K under 50 Oe, (c) FCW and FCC curves showing thermal hysteresis near the Morin transition, and (d) derivative curves with Gaussian fitting used to determine the Morin transition temperature SQUID magnetometry results for α-Fe₂O₃ nanoplates from sample sm_41: (a) M–H curve measured at 300 K, (b) ZFC and FCW curves measured from 10 K to 350 K under 50 Oe, (c) FCW and FCC curves showing thermal hysteresis near the Morin transition, and (d) derivative curves with Gaussian fitting used to determine the Morin transition temperature Fig. 8SQUID magnetometry results for irregular α-Fe₂O₃ nanoparticles from sample sm_14: (a) M–H curve measured at 300 K, (b) ZFC and FCW curves measured from 10 K to 350 K under 50 Oe, (c) FCW and FCC curves showing thermal hysteresis near the Morin transition, and (d) derivative curves with Gaussian fitting used to determine the Morin transition temperature SQUID magnetometry results for irregular α-Fe₂O₃ nanoparticles from sample sm_14: (a) M–H curve measured at 300 K, (b) ZFC and FCW curves measured from 10 K to 350 K under 50 Oe, (c) FCW and FCC curves showing thermal hysteresis near the Morin transition, and (d) derivative curves with Gaussian fitting used to determine the Morin transition temperature The magnetic parameters extracted from the M–H curves are summarized in Table 6, while the Morin transition temperatures and thermal hysteresis values obtained from temperature-dependent SQUID measurements are summarized in Table 7. The thermal hysteresis values obtained from the difference between the FCW and FCC Morin transition temperatures are compared in Fig. 9. The SQUID measurements show that the three α-Fe₂O₃ nanoparticle samples display different magnetic responses at 300 K and different Morin transition behavior. Sample sm_24 shows an S-shaped M–H curve and the lowest Morin transition temperature of approximately 214 K. Sample sm_41 shows a weak ferromagnetic-like M–H response and the largest thermal hysteresis of approximately 7 K. Sample sm_14 shows the lowest saturation magnetization, approximately 0.10 emu g⁻¹, and a Morin transition temperature of approximately 239 K. These results provide the magnetic parameters required in comparison with the structural and morphological results obtained from TEM and XRD. Table 6Magnetic properties obtained from M–H measurementsSampleMorphologyTEM sizeMs (emu g⁻¹)Magnetic response at 300 Ksm_14Irregular nanoparticles53 ± 4 nm0.10Weak ferromagnetic-likesm_24Rhombohedral nanoparticles32 ± 4 nm0.24Superparamagnetic-likesm_41NanoplatesThickness: 22 ± 5 nm; diameter: 109 ± 12 nm0.25Weak ferromagnetic-like Table 6Magnetic properties obtained from M–H measurementsSampleMorphologyTEM sizeMs (emu g⁻¹)Magnetic response at 300 Ksm_14Irregular nanoparticles53 ± 4 nm0.10Weak ferromagnetic-likesm_24Rhombohedral nanoparticles32 ± 4 nm0.24Superparamagnetic-likesm_41NanoplatesThickness: 22 ± 5 nm; diameter: 109 ± 12 nm0.25Weak ferromagnetic-like Magnetic properties obtained from M–H measurements Table 7Morin transition parameters obtained from temperature-dependent SQUID measurementsSampleMorphology\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${T_m}$$\end{document} (K)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${T_m}\left( W \right)$$\end{document} (K)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${T_m}\left( C \right)$$\end{document} (K)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{{\Delta}}}{T_m}$$\end{document} (K)sm_14Irregular nanoparticles239238237~ 1sm_24Rhombohedral nanoparticles214214212~ 2sm_41Nanoplates241239232~ 7 Table 7Morin transition parameters obtained from temperature-dependent SQUID measurementsSampleMorphology\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${T_m}$$\end{document} (K)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${T_m}\left( W \right)$$\end{document} (K)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${T_m}\left( C \right)$$\end{document} (K)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{{\Delta}}}{T_m}$$\end{document} (K)sm_14Irregular nanoparticles239238237~ 1sm_24Rhombohedral nanoparticles214214212~ 2sm_41Nanoplates241239232~ 7 Morin transition parameters obtained from temperature-dependent SQUID measurements Fig. 9Comparison of thermal hysteresis values, ΔTm, obtained from FCW and FCC magnetization curves for irregular sm_14 nanoparticles, rhombohedral sm_24 nanoparticles, and sm_41 nanoplates Comparison of thermal hysteresis values, ΔTm, obtained from FCW and FCC magnetization curves for irregular sm_14 nanoparticles, rhombohedral sm_24 nanoparticles, and sm_41 nanoplates Discussion The study aimed to investigate the magnetic characteristics of a range of hematite nanoparticles, such as the magnetic behaviour and Morin temperature. This was accomplished by analyzing a sample group composed of sm_14, sm_24, sm_41_B2, and pure resin, using Superconducting Quantum Interference Device (SQUID) measurements. Magnetization versus magnetic field (M vs. H) measurements were taken at 300 K for all samples. The‍‌‍‍‌‍‌‍‍‌ magnetic properties of the synthesized α-Fe₂O₃ nanoparticles have a clear and uniform dependence on particle shapes and characteristic sizes, as confirmed through SQUID measurements described in the Results section. Rhombohedral particles (mean size 32 ± 4 nm) behave superparamagnetic at room temperature, having a saturation magnetization of about 0.24 emu g⁻¹ and a Morin transition temperature near 214 K. On the other hand, irregular particles (53 ± 4 nm) and nanoplates (thickness 20 ± 5 nm, lateral dimension ≈ 123 ± 11 nm) show very weak ferromagnetic behavior, with their Morin transition temperatures being around 238 K and 241 K, respectively. The extent of thermal hysteresis, described as \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\Delta {T_m}={T_m}\left( W \right) - {T_m}\left( C \right)$$\end{document}, progressively grows from irregular particles (≈ 1 K) to rhombohedral particles (≈ 2 K) and reaches the peak for nanoplates (≈ 7 K), thus supporting the notion that shape anisotropy and dimensional confinement have been influential. Quantitative results like these are invariably detected in M–H, ZFC/FC, and FCC experiments and indicate that hematite nanoparticle thickness and morphological anisotropy have a stronger impact on magnetic transitions than the overall particle size. The Discussion and Conclusions sections have been merged so that they emphasize these quantitative correlations and thus are directly supported by the experiment. Before‍‌‍‍‌‍‌‍‍‌ determining the intrinsic magnetic parameters of the nanoparticles, the diamagnetic signal from the resin matrix and the sample holder were corrected very precisely. A reference sample with pure resin in the same kind of capsule was measured in the same way (− 7 T to + 7 T at 300 K) in the SQUID. The reference showed a completely linear response to the field without any hysteresis, thus confirming that it was diamagnetic. The total measured magnetization of the composite samples was then written as Mtotal(H) = MNP(H)+χdiaH is the linear diamagnetic background. At the high-field region \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$(\left| H \right| \geqslant 5~T$$\end{document}), where the magnetization of the nanoparticles nears the saturation, a linear function (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M\left( H \right)=Ms+\chi diaH$$\end{document}) was used to fit the data, thus getting both the diamagnetic susceptibility and the real saturation magnetization (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$Ms$$\end{document}). Moreover, the subtraction method of the mass-normalized resin reference curve was carried out as a consistency check. It gave totally consistent corrected curves with the level of uncertainty of 3% or less. Post background subtraction, M–H curves showed good symmetry and field independence at high fields, and the application of Langevin fitting resulted in a low residual error with no evident systematic deviations. The obtained values of diamagnetic susceptibilities agreed with typical polymer character values (~ 10⁻⁷–10⁻⁶ emu g⁻¹ Oe⁻¹), thus confirming that the correction was done correctly. The saturation magnetization values being referred to here are 100% from the background-corrected nanoparticle ‍‌‍‍‌‍‌‍‍‌contribution. To‍‌‍‍‌‍‌‍‍‌ pick out a reliable Morin transition temperature (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${T_m}$$\end{document}) from the magnetic measurements, the first derivatives of the ZFC and FCW magnetization plots (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$dM/dT$$\end{document}) were taken and the resulting curves were fitted with a Gaussian function. Tm uncertainties were obtained by considering the Gaussian peak fitting error, the instrumental temperature resolution (± 1 K), and the difference in transition temperatures derived from the ZFC and FCW measurements. According to this evaluation, the error in \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${T_m}$$\end{document} is ± 1–2 K. The rhombohedral particles (sm_24) have \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${T_m}$$\end{document} = 214 ± 1 K, the irregular ones (sm_14) − \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${T_m}$$\end{document} = 239 ± 1 K, and the nanoplates (sm_41_B2) indicate \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${T_m}$$\end{document} = 241 ± 2 K. The slightly higher error for the nanoplates is due to the broader derivative peak and bigger thermal hysteresis. Both FC and ZFC magnetization measurements were carried out in an external magnetic field of 50 Oe. Since this field magnitude is negligible in comparison with the intrinsic anisotropy and exchange fields of hematite, its effect on Tm should be almost non-existent. Still, a magnetic field can promote the weak ferromagnetic spin-canted state above Tm and thus lower the Morin transition temperature slightly. At 50 Oe, such a shifting effect can be expected to be not more than 1–2 K and hence, it is covered by the measurement uncertainty of the present work. Therefore, the experimentally determined \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${T_m}$$\end{document} values are to be considered as the inherent ones within the error range ‍‌‍‍‌‍‌‍‍‌given. The field cool and zero field cool warming measurement was performed on samples sm_14, sm_24 and sm_41_B2, using a 50 Oe magnetic field in a temperature range from 10 K to 350 K. Based on TEM analysis, the sm_41_B2 sample comprises nanoplates with a thickness of 20 ± 5 nm and a length of 123 ± 11 nm. In contrast, according to TEM analysis, the sm_14 samples contain irregularly shaped nanoparticles, with sizes of 53 ± 4 nm. According to TEM analysis, the sm_24 sample consists of cube-shaped nanoparticles that are 32 ± 4 nm. The curve plotting magnetization (M) against the magnetic field (H), commonly called the M vs. H curve, can reveal insights into nanoparticle characteristics and saturation magnetization value. Upon analyzing the magnetization vs. field measurements, it was discovered that at room temperature, the particles sm_41_B2, and sm_14 exhibit behaviour similar to weakly ferromagnetic particles. Conversely, the particles sm_24 display superparamagnetic properties at room temperature, as evidenced by the presence of an “s” shaped M vs. H curve in both samples. Ozdemir and Dunlop conducted a comprehensive study on the magnetic hysteresis characteristics of natural and synthetic hematite with varying grain sizes. They suggested that the magnetic properties of hematite nanoparticles, such as the morin transition, saturation magnetization, and magnetic coercivity, were greatly influenced by anisotropy. The authors also examined the relationship between grain size and coercivity and determined that HC peaked between 0.1 and 3 μm [32]. The literature indicates that combining small and oriented sub-particles into the superstructure transformed the single-domain structure into a multi-domain-like structure, resulting in higher remanent magnetization and coercivity [33–35]. The results from samples sm_14, sm_24 and sm_41_B2 are very much in agreement with the literature. No remanent magnetization and coercive field were seen in the M vs. H curve of sm_14, sm_24, and sm_41_B2. The nanoparticles probably have a signal crystal structure, no strain and defect free. The M vs. H curve was used to determine the magnetic saturation (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${M_s}$$\end{document}) of samples sm_14, sm_24 and sm_41_B2, which had values of 0.1 emu/g, 0.24 emu/g and 0.25 emu/g, respectively. Typically, the magnetic saturation of hematite nanoparticles increases with particle size [36]. However, this trend was not observed in these samples despite having different sizes and shapes. The magnetic saturation value is affected by particle size, surface coating (for shape control), and magnetic anisotropy. Sample sm_24 had rhombohedral-shaped particles with a size of 32 ± 4 nm coated with oleate group, while sample sm_41_B2 had nanoplates with a thickness of 20 ± 5 nm and a length of 123 ± 11 nm coated with acetate group. The irregularly shaped particles in sample sm_14 and sm_38 were coated with the PVP group. These samples’ lack of size-dependent magnetic saturation is likely due to their different shapes, sizes, magnetic anisotropy, and surface coating agents. Tadic et al. [35] investigated the nanoplates and rhombohedral-shaped nanoparticles. In his work, the size of the rhombohedral-shaped particles was 30–70 nm. He observes the saturation magnetization \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$Ms=0.87$$\end{document} emu/g at 300 K for rhombohedral-shaped particles [35]. For nanoplates, he achieved the saturation magnetization \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${M_s}=0.87$$\end{document} emu/g, a diameter of 60–140 nm and a thickness of 20–40 nm [35]. The magnetic saturation \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${M_s}$$\end{document} value was observed less in rhombohedral shapes and nanoplates than those reported by the Tadic et al. [35] literature. The less particle-particle interaction was the most probable reason for getting a lesser value for Saturation magnetization. The lesser particle-particle interaction can only be achieved using the ultrasonication probe during sample preparation and embedding the particle into the solid resin matrix. Sun et al. [36] employed a hydrothermal method to synthesise hematite nanoplatelets with 100–140 nm diameters and 40–70 nm thickness. For nanoplates, he achieved magnetic saturation \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${M_s}=0.91$$\end{document} em/g. He observed the weak ferromagnetic behaviour of the nanoplates [37]. Hao et al. [37] reported hematite nanoplates with a diameter of 180 nm and thickness of 30 nm. He noted the weak ferromagnetic behaviour from hematite nanoplates. Zhu et al. [38] synthesized the irregularly shaped hematite nanoparticles. The size was in the range of 30–40 nm. He reported the weak ferromagnetic behaviour of nanoparticles at room temperature. Zhao et al. [39] found that synthesised nanoplates exhibited ferromagnetic behaviour at room temperature and showed that magnetization saturation was a function of the field strength. The temperature at which the morin transition occurs in the bulk hematite phase was measured to be \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${T_m}=263$$\end{document} K; hematite exhibits weak ferromagnetism above this temperature, up to 963 K, and anti-ferromagnetism below this temperature. Previous studies suggest that particle size primarily influences the morin transition temperature and is less affected by the shape of the particles. The zero-field cool warming measurement was used to determine the morin transition in samples sm_14, sm_24 and sm_41_B2. The point at which there was a sudden increase in the slope of the curve at a specific temperature was identified as the morin transition temperature. The morin transition temperature was found to be 238 K for sample sm_14, 214 K for sample sm_24 and 241 K for sample sm_41_B2. Tadic et al. [35] observed the morin transition temperature for nanoplates at 242 K. He reported the morin transition temperature around 221 K for rhombohedral-shaped particles. The morin transition temperature for nanoplate from sample sm_41_B2 and rhombohedrons particles from sm_24 remarkably agree with Tadic et al. [35]. Zhu et al. [38] synthesised the irregularly shaped hematite nanoparticles through the hydrothermal process. He reported the morin temperature of 259 K for the irregularly shaped particles. Contrarily the irregularly shaped hematite nanoparticles obtained from sample sm_14 show the morin temperature at 238 K. A considerable difference was seen in the reported literature and experimental values for morin temperature (Table 8). The difference in morin temperature was most likely because of higher particle–particle interaction. Zhu et al. [38] highlighted the higher particle–particle interaction in their work. Table 8Comparison of experimental and literature values for saturation magnetization and morin temperatureSample nameTEM particle size (nm)Saturation magnetization (emu/g)Morin temperature (K)ExperimentalLiteratureExperimentalLiteraturesm_41 B2Plate thickness = 20 ± 5 length = 123 ± 110.250.87 [37]241242 [37]sm_24Rhombohedral = 32 ± 40.240.87 [37]214221 [41]sm_14Irregular = 53 ± 40.10.53 [39]238259 [41] Table 8Comparison of experimental and literature values for saturation magnetization and morin temperatureSample nameTEM particle size (nm)Saturation magnetization (emu/g)Morin temperature (K)ExperimentalLiteratureExperimentalLiteraturesm_41 B2Plate thickness = 20 ± 5 length = 123 ± 110.250.87 [37]241242 [37]sm_24Rhombohedral = 32 ± 40.240.87 [37]214221 [41]sm_14Irregular = 53 ± 40.10.53 [39]238259 [41] Comparison of experimental and literature values for saturation magnetization and morin temperature It was observed as the size goes down the hematite nanoparticles changed their behavior to superparamagnetic nanoparticles. It occurs when nanoparticles are small enough to experience rapid thermal energy fluctuations in their magnetic moments. The superparamagnetic phenomena can be more explanatory by Eq. 4 [41]: \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\tau _0}={\tau _p}\exp \frac{{KV}}{{kT}},$$\end{document} where \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\tau _0}$$\end{document} is the relaxation time of the particle at zero field, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\tau _p}$$\end{document} is the attempt time (the time it takes for the magnetic moment to reverse direction due to thermal fluctuations), \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$KV$$\end{document} is the magnetic anisotropy energy density, and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$kT$$\end{document} is the thermal energy [41]. The equation describes the dependence of relaxation time on the magnetic anisotropy energy density and the thermal energy. In particular, the equation predicts that the relaxation time of a superparamagnetic nanoparticle decreases exponentially as the magnetic anisotropy energy density decreases or as the temperature increases. It can be seen in nanoplates and rhombohedral- shaped particles as the size of the nanoparticles decreases, the energy density of the nanoparticles decreases, and hematite nanoparticles start behaving like the superparamagnetic nanoparticles at 32 ± 4 nm. One more theory explains the ferromagnetic to superparamagnetic transformation. Nanoplates consist of two large dimensions and one small dimension, so they have a higher level of morphological anisotropy than rhombohedrons. The nanoplates have higher anisotropy energy (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$KV$$\end{document}) than the rhombohedrons. Because of nanoplates’ higher anisotropy energy, the critical temperature for the spin-flip was higher than the critical temperature of rhombohedrons. Differences in magnetization were observed in samples sm_14, sm_24 and sm_41_B2 between the field cool warming and zero field cool warming. The field cool warming curve shifted to higher magnetization at lower temperatures below the morin transition. In contrast, the zero-field cool warming curve was observed at lower magnetization due to uncompensated surface spins. Sample sm_24 showed a more significant gap than sm_14 and sm_41_B2 because it had the smallest particle size. The decrease in particle size increased the number of uncompensated surface spins due to increased surface area. It led to a more significant gap between the field cool and zero field cool warming curve, which is evident in sample sm_24. The magnetic field of 50 Oe was used to measure the field cool warming (FCW) and cooling (FCC) for several samples, namely sm_14, sm_24, and sm_41_B2. The FCW and FCC curves for sm_14 overlap until 240 K, whereas for sm_24, the curves overlap until 200 K. The curves for sm_41_B2 overlap until 210 K, but there is hysteresis due to the non-overlapping of the FCW and FCC curves. The difference between the two temperatures (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${T_m}\left( W \right)~ -$$\end{document} \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${T_m}\left( C \right)$$\end{document}) varies based on the shape of hematite nanoparticles. A thermal hysteresis can be seen near the morin transition temperature in samples sm_14, sm_24, and sm_41_B2. For irregularly shaped hematite nanoparticles in samples sm_14, the difference of morin temperature during warming and cooling (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${T_m}\left( W \right)~ -$$\end{document} \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${T_m}\left( C \right)$$\end{document}) is only 1 K. The TEM results revealed the presence of rhombohedral-shaped particles with a size range of 32 ± 4 nm in sample sm_24. Compared to irregularly shaped particles, the rhombohedral-shaped particles showed wider thermal hysteresis. These rhombohedral-shaped particles were smaller in size and more isotropic in shape than the irregularly shaped particles, which led to wider thermal hysteresis. The difference between the morin temperature during cooling and warming (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${T_m}\left( W \right)~ -$$\end{document} \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${T_m}\left( C \right)$$\end{document}) was observed to be 2 K for the rhombohedral-shaped particles. TEM analysis of sample sm_41_B2 revealed the presence of nanoplates with a thickness of 20 ± 5 nm and a diameter of 123 ± 11 nm. These nanoplates exhibited wider thermal hysteresis compared to the rhombohedral-shaped particles. The smaller thickness of the nanoplates contributed to the broader thermal hysteresis. Compared to rhombohedral-shaped particles, nanoplates are less isotropic and have a smaller thickness. The thickness of nanoplates is the most critical factor contributing to the broader thermal hysteresis observed in them. The difference between the morin temperature during cooling and warming (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${T_m}\left( W \right)~ -$$\end{document} \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${T_m}\left( C \right)$$\end{document}) was found to be 7 K for the nanoplates. López-Sánchez et al. [41], Goya et al. [42], Chow and Keffer [43] and Morrish et al. [44] have reported varying values of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${T_m}$$\end{document} in heating and cooling. They suggested that a metastable state of the antiferromagnetic (AF) or weakly ferromagnetic (WF) phase beyond the Morin transition temperature (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${T_m}$$\end{document}) could only be observed in crystals with few defects, as nucleation is difficult in these nearly defect-free crystals and lead to thermal hysteresis near the morin transition temperature. The findings from samples sm_14, sm_24 and sm_41_B2 align with previous research by Goya et al. [38], Chow et al. [39] and Morrish et al. [40] suggesting that these nanoparticles have few defects and are small in size. During the synthesis process, defects move from the core of the nanoparticles to the surface and disappear. The difference in morin temperature (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${T_m}\left( W \right)~ - {T_m}\left( C \right)$$\end{document}) during heating and cooling is associated with surface magnons, which become harder as particle size decreases and shape isotropy increases. Irregularly shaped particles have larger particle sizes and lower shape isotropy, resulting in softer surface magnons and only a 1 °C difference in morin temperature during heating and cooling. Rhombohedral-shaped particles are more isotropic in shape than nanoplates but still have softer surface magnons. The thickness of the nanoplates (20 nm) is the most crucial factor determining surface magnon hardness, as thinner plates have harder surface magnons. Overall, nanoparticle size is the most critical factor affecting the change in morin temperature during field cool cooling and warming, with shape isotropy playing a limited role. Conclusions This study focuses on the synthesis, structural characterization and magnetic analysis of hematite (α-Fe₂O₃) nanoparticles with different morphologies rhombohedral, irregular and nanoplate using the hydrothermal method. Each morphology was achieved by varying the chemical precursors and reaction conditions. TEM analysis revealed distinct particle sizes: rhombohedral (32 ± 4 nm), irregular (53 ± 4 nm), and nanoplates (20 ± 5 nm thick, 123 ± 11 nm in diameter). Size distribution was narrow for rhombohedral and irregular shapes, while nanoplates showed narrow diameter but broader thickness variation. XRD confirmed the predominantly pure hematite phase, with the most intense peaks corresponding to the (110) and (104) planes. Crystallite sizes from XRD (using the Scherrer formula) were smaller than TEM sizes, indicating single-crystal nature. A minor presence of the magnetite phase was also detected in one sample. SQUID magnetometry revealed distinct morphology-dependent magnetic behaviors among the synthesized hematite nanoparticles. Rhombohedral particles exhibited superparamagnetic characteristics at room temperature, with a saturation magnetization of 0.24 emu/g and a Morin transition temperature of 221 K. In contrast, both nanoplate and irregular-shaped particles displayed weak ferromagnetism, with Morin transitions occurring at 241 K and 238 K, respectively. Field-cooled warming/cooling measurements indicated the presence of thermal hysteresis, the magnitude of which increased with decreasing particle isotropy and size largest for nanoplates, followed by rhombohedral particles, and smallest for irregular morphologies. Additionally, the magnetite reference sample exhibited the characteristic Verse transition at 106 K, confirming its phase identity. Limitations of the work and future perspectives This investigation was primarily a comparative study of structural and magnetic features of α-Fe₂O₃ nanoparticles of various morphologies prepared by hydrothermal synthesis. Nevertheless, few limitations should be considered. The first one is that in order to obtain three different nanoparticle morphologies, quite different hydrothermal synthesis conditions were used, for example, different precursors, additives, solvent compositions, temperatures, and reaction times. As a result, although there is a strong correlation between morphology and magnetic behavior, the effect of an individual synthesis parameter cannot be entirely determined. So, it will be better if in future experiments, a single-variable experimental design is used to determine the effect of particle size, morphology, precursor chemistry, pH, capping agent, and reaction time on magnetic properties. Secondly, the structural characterization mainly relied on XRD, TEM, HRTEM, and FFT observations. Even though these techniques have shown that crystalline α-Fe₂O₃ nanoparticles are formed, a higher level of crystallographic analysis could provide a lot more information. For instance, Rietveld refinement with powder samples that are randomly oriented, preferably those contained in a capillary or back-loaded holder, should be considered in future work to get quantitative phase fractions, lattice parameters, crystallite size, microstrain, and perhaps orientation effects due to the preferred- orientation. Spectroscopic techniques that are more specific to the surface, like X-ray photoelectron spectroscopy or Mössbauer spectroscopy, will be very useful in determining the influence of the surface oxidation state, spin disorder, and maybe the magnetic contributions caused by defects. Third, the magnetic analysis was confined to SQUID magnetometry at a few measurement conditions. The work presented here was primarily based on M–H curves at 300 K along with ZFC/FC measurements from 10 to 350 K at 50 Oe. Probably more detailed temperature-dependent M–H and field-dependent ZFC/FC measurements, repeated thermal cycling, and frequency-dependent AC susceptibility should be done later on to effectively differentiate weak ferromagnetism, superparamagnetic relaxation, surface-spin freezing, and interparticle interaction effects. Lastly, the nanoparticles were enclosed in a resin matrix during SQUID measurements. Even though this method ensured mechanical stability and lessened particle aggregation, at the same time, it could have limited interparticle magnetic interactions and changed the saturation magnetization obtained. In the future, similar investigations should be made that check the results of the resin-embedded samples against those obtained with pressed powder, dispersed powder, and magnetically diluted samples to figure out more clearly the influences of sample preparation on the measured magnetic response. In general, research in future should be targeted to linking controlled synthesis, quantitative structural characterization, surface/defect analysis together with extending magnetic measurements in order to get a much clearer picture of the connection between the α-Fe₂O₃ nanoparticle morphology, surface structure, and the Morin transition ‍‌‍‍‌‍‌‍‍‌behavior.