Magnetite, also known as Fe3O4, has been extensively researched as one of the most common half-metallic materials in the field of spintronics for a considerable period of time. Magnetoelectronic devices are possible because of the material’s high Curie temperature of 860 K , as well as its high spin polarization with only one spin at the Fermi level, even at room temperature . Fe3O4 thin films are an issue of interest and have extensive applications in Li-ion batteries, spin Seebeck devices, supercapacitors, spin Hall magnetoresistance, and the study of analog resistive switching of Fe3O4-based cross-cell memristive devices .

Fe3O4 thin films can be grown by many processes, including molecular beam epitaxy, which is employed for depositing single crystal films, and pulsed laser deposition, which is utilized to achieve epitaxial films . The RF magnetron sputtering technique is extensively utilized because of its cost-effectiveness, simplicity, effectiveness, and capacity to produce Fe3O4 films with remarkable uniformity. The qualities of the films can be modified by manipulating parameters throughout the growth process . The impact of substrate temperature, annealing temperature, gas flow rate, and thickness on enhancing the characteristics of Fe3O4 thin films has been examined . The substrates play a crucial role in directing the growth and enhancing the quality of the crystal, resulting in significant changes in the film’s characteristics .

Roy et al. conducted a study on polycrystalline Fe3O4 films on Si and SiO2/Si substrates. Their findings revealed that the value of the Gilbert damping parameter is significantly higher in Fe3O4/Si films compared to Fe3O4/SiO2/Si films . Hong and coworkers deposited Fe3O4 films on a MgO substrate, which exhibited a change in the direction of Fe3O4 crystal formation. The directions (222), (400), and (440) of the Fe3O4 peak matched, respectively, the (111), (100), and (110) orientations of the MgO substrate . In addition, Zhang et al. successfully applied a layer of Fe3O4(001) on a MgO(001) substrate. The resulting material exhibited saturation magnetization and magnetic moment values of 407 ± 5 emu/cm3 (3.26 ± 0.04 μB/(f.u.)) and 3.31 ± 0.15 μB/(f.u.), respectively .

This paper addresses the deposition of Fe3O4 thin films on three different types of substrates, namely an amorphous SiO2/Si(100) substrate, a single crystal MgO(100) substrate, and a buffer layer consisting of MgO/Ta/SiO2/Si(100). The properties of Fe3O4 thin films were analyzed using atomic force microscopy (AFM), X-ray diffractometry (XRD), and vibrating sample magnetometry (VSM). It is interesting to note that the saturation magnetization of the Fe3O4 films was significantly improved (278.9 emu/cm3) when utilizing a Ta interlayer located between MgO and SiO2, compared to films on SiO2 (136.3 emu/cm3) and MgO(100) (126.3 emu/cm3) substrates. This indicates the potential to facilitate the development of novel magnetic and spintronic architectures.

Figure 1: AFM images (1 × 1 µm2) of Fe3O4 thin films on different substrates. (a) SiO2, (b) MgO(100), and (c) MgO/Ta/SiO2.

Table 1: Surface properties obtained from the AFM scans of Fe3O4 samples.

Figure 2: XRD spectra of sample 1 (black), sample 2 (red), and sample 3 (blue) on SiO2, MgO(100) and MgO/Ta/SiO2, respectively.

XRD patterns provide further information about the structural properties of a material, such as lattice constant (a), dislocation density (δ), and microstrain (ε). Bragg’s law was used to calculate the d-spacing of the Fe3O4(311) and Fe3O4(400) peaks :

where n is the order of diffraction (n = 1) and λ is the X-ray wavelength (Cu Kα, λ = 1.5406 Å). The lattice constant a of the three Fe3O4 samples was determined by :

The microstrain in these samples can be calculated from the lattice constant a above by using the following relation :

where a0 is the lattice parameter of bulk Fe3O4 (a0 = 8.397 Å ). Microstrain is a crucial factor that helps to analyze the existence of strain and deformation in thin films .

The d-spacing values of the Fe3O4(311) and Fe3O4(400) peaks of sample 3 are 2.514 and 2.085 Å, respectively, which are smaller than those on SiO2 (2.527 Å) and MgO(100) (2.099 Å) substrates. These low d-spacing values can be caused by the microstrain in all Fe3O4 samples . The Fe3O4 film grown on the multilayer structure is under a higher compressive strain of −0.70% and −0.67%, corresponding to the Fe3O4(311) and Fe3O4(004) peaks, respectively, than samples 1 and 2 with values of −0.19% and −0.01%, respectively. Sample 3 exhibits a decrease in d-spacing for both the (311) and (400) peaks, in comparison to sample 1 and sample 2. The presence of compressive stress in the crystallites of the Fe3O4 thin films causes a shift in the peak observed in sample 3 . Our results reveal that the growth orientation of the Fe3O4 thin film depends on the lattice mismatch between the Fe3O4 thin film and the substrate or buffer layer. When the Fe3O4 thin film is deposited on the amorphous SiO2 substrate, the lattice mismatch between the amorphous substrate and the crystalline film is large. In this case, the growth orientation of Fe3O4 thin film is determined by the direction having the least internal energy, which is [111]. The energetically favored [111] direction also has the highest probability of occupying random dangling bonds from the amorphous substrate surface because it has the highest areal density . In contrast, the small lattice mismatch between Fe3O4 thin film and MgO(100) substrate (≈0.3% ) results in the growth orientation controlled by the substrate and leads to the appearance of the [100] direction in Fe3O4/MgO. In addition, the Fe3O4(400) and MgO(200) peaks are close because of the similarity in crystalline structure (cubic) and lattice constant (aFe3O4 = 8.397 Å, aMgO = 4.212 Å ). The growth orientation of the Fe3O4 thin film in sample 3 is also affected by the internal energy of the [111] direction in addition to effects from the buffer layer. This explains the highest microstrain value in sample 3. The difference in lattice constants between MgO and Ta (cubic, aTa = 3.3058 Å ) puts the MgO buffer layer under a higher strain and creates a larger lattice mismatch between the Fe3O4 thin film and the MgO layer compared to the Fe3O4 thin film and MgO substrate.

In addition, the dislocation density was calculated by the following relation :

Table 2: Structural parameters of Fe3O4 thin films on various substrates.

Figure 3: Magnetization of Fe3O4 samples in an external magnetic field in a range of (a) −10 kOe to 10 kOe and (b) −2 kOe to 2 kOe.

Table 3: Magnetic parameters of samples 1, 2, and 3.

Fe3O4 films were prepared on different substrates of SiO2/Si(100), MgO(100), and MgO/Ta/SiO2/Si(100) at room temperature using RF sputtering. Our finding highlights the role of the Ta buffer layer in the multilayered structure. Ta helps to decrease the crystallization temperature of the Fe3O4 film and prevents the diffusion of oxygen atoms from SiO2 to MgO, resulting in an enhancement in grain size and RMS roughness, and in the formation of a polycrystalline structure. Changes in grain size and structure have a strong impact on saturation magnetization and coercivity of the Fe3O4 thin films. Our results indicate that the combination of Ta and MgO buffer layers can influence the morphology and structure of Fe3O4 thin films and help to boost the magnetic properties.

RF magnetron sputtering was used at room temperature to grow magnetite films with 40 nm thickness on a variety of substrates, including SiO2, MgO(100), and the multilayer substrate MgO/Ta/SiO2/Si(100). The MgO(100) substrates were prepared by immersing them in a methanol bath at a temperature of 60 °C and drying them in N2 gas flow. Subsequently, the purified substrates were moved into an ultrahigh vacuum (UHV) chamber and underwent a pre-heating process at 600 °C for 30 min in order to eliminate any remaining impurities. The SiO2/Si(100) substrates were immersed in acetone and 2-propanol for a duration of 2 min in an ultrasonic bath. Subsequently, they were immersed in a solution of methanol at a temperature of 60 °C and then dried in N2 gas flow. A 5 nm thick layer of tantalum was deposited on a SiO2/Si(100) substrate using RF magnetron sputtering. This was followed by the formation of a 5 nm thick layer of MgO. The Fe3O4 layers were applied using RF magnetron sputtering at a base pressure of 10−8 Torr, employing a flow of 33 sccm of Ar gas to maintain a stable plasma. The initially deposited films were annealed at a temperature of 723 K for a duration of 2 h under a base pressure of 2.3 × 10−8 Torr. The Fe3O4 films were analyzed regarding their surface morphology, magnetic properties, and structural properties using atomic force microscopy (EasyScan2, Nanosurf), vibration sample magnetometry (Quantum Design magnetic property measurement system, MPMS-5XL), and X-ray diffractometry (Bruker Discover D8), respectively.