EDP Sciences logo
Subscriber Authentication Point
Search
Menu
All issues Volume 98 (2023) Eur. Phys. J. Appl. Phys., 98 (2023) 14 Full HTML
Issue
Eur. Phys. J. Appl. Phys.
Volume 98, 2023
Article Number 14
Number of page(s) 13
Section Thin Films
DOI https://doi.org/10.1051/epjap/2023220262
Published online 07 March 2023

© EDP Sciences, 2023

1 Introduction

In the last decade, many studies have been realized as far as the various devices of magnetic films and compounds are concerned. Several research groups investigated ferromagnetic films specially based on Ni and Fe transition metal elements for their large used applications such as storage technology, sensor development, magnetoelectronics, spintronic, biomedical applications, electric motors and giant magnetoresistance [1–10].

Nowadays, ferromagnetic Ni-Fe alloys are very important magnetic materials due to the possibility of substitution up to 25% of iron atoms instead of nickel atoms. The obtained Ni-Fe alloy is much cheaper than pure Ni and the competing contribution of Ni and Fe atoms can generate a physical phenomenon in Ni-Fe thin films such as magnetic anisotropy [11,12]. Indeed, the crystalline structure of NiFe films can also exhibit exceptional magnetic properties such as high permeability, small coercivity, negligible magnetostriction and distinct anisotropy magnetoresistance. Permalloy Ni80Fe20, hereinafter referred to as Py, is an important and attractive material widely used in magnetic industry due to its remarkable magnetic properties of high permeability, low coercivity and small magnetic anisotropy. These magnetic properties have made this type of Permalloy films a good candidate for many devices and technological applications such as magnetoresistive sensor [13], microelectromechanical systems (MEMS) [14], magnetic thin film inductors [15], shields and spintronic area [13]. On another hand, many Py thin films at different stoichiometry of nickel and iron atoms and different substrate deposition were studied using diverse elaboration methods including RF-sputtering [16], thermal evaporation [17] and electro-deposition [18]. These techniques are very useful for the elaboration of materials due to their many advantages: low economic costs such as low energy consumption and low operating temperature. Trukhanov et al. [8] have studied the multilayered Ni80Fe20/Cu film shields, using the electrochemistry method with diverse thicknesses and number of magnetic layers in different measurement of DC and AC shields. The authors observed the maximal value of DC-shielding effectiveness for multilayered films at small number of magnetic layers and large thickness, and the minimal value of DC-shielding effectiveness for the multilayered films at minimum 5µm thickness and maximum number of magnetic layers (80 layers). A study of microstructures and magnetic properties of Py thin films, grown by high power impulse magnetron sputtering compared with films grown by dc magnetron sputtering techniques, at different pressure and substrate temperature, have been reported by Movaffaq et al. [19]. The authors remarked a drop in film density for samples deposited at room temperature with increasing pressure, accompanied by a rise in both coercive and anisotropy fields. On the other hand, Belyaev et al. [20] investigated the ferromagnetic resonance (FMR) spectrometry of magnetic properties of nanocrystalline thin films of permalloy targets of various composition NixFe1-x (x = 0.6–0.85). They reported a fluctuation of the uniaxial magnetic anisotropy field near the film edges, as well as a decrease in the effective saturation magnetization and. a drastic widening of the FMR line. Wang et al. [21] studied the physical properties of NiFe/FeMn film obtained by flash annealing in reversing field, and report values of magnetoresistance ratio as high as 2.41% and a maximum of square resistance up to 1.48 Ω.

Zhang et al. reported that films grown at 300 °C exhibit smaller sheet resistance Rs, down to ∼2 Ω/sq and smaller resistivity ρ, down to ∼4 Ω, whereas films grown at room temperature exhibit higher sheet resistance up to ∼39 Ω /sq, and higher resistivity up to ∼78 Ω, for their NiFe films prepared by DC magnetron sputtering [22].

In this present work, we focus on the investigation of the influence of thickness on the physical properties of Py thin films, deposited by thermal evaporation from the metallic powders of Ni and Fe on GaAs (100) substrates. A short description of the experimental details, relating to the deposition and the different characterization methods used in this study, is exposed in Section 2. In Section 3, the obtained experimental results and their discussions are developed. The results, obtained from the analysis of the microstructural and structural properties of the films by X-ray diffraction (XRD) method, are studied in Section 3.1. In Section 3.2, we present the analysis (SEM-EDX), carried out to study the surface morphology and chemical compositions of the Py layers. The observations by atomic force microscopy (AFM) technique are presented in Section 3.3. Finally, the evolution curves of electrical resistivity, magnetoresistance and mobility as functions of Py layer thickness, obtained by the experimental Hall Effect technique and Van der Pauw method, are studied in Section 3.4. A conclusion is given in Section 4.

2 Experimental procedures

2.1 Preparation of Py thin films

We thermally evaporated Py thin films, of different thicknesses, onto GaAs (100) substrates, under vacuum, and at room temperature, from a mixture that consists of two commercial powders (Fe, Ni) with purity 99.99%, produced and supplied by Good Fellow GmbH. In the working chamber, GaAs substrates are placed on a substrate holder suspended at a height equal to 7 cm perpendicularly above a tungsten crucible covered with alumina (Al2O3) and containing a quantity of the mixed desired powders. A metal plate, playing the role of a mask, is placed above the crucible. This plate is secured to a rod that can be moved manually from the outside of the evaporation chamber. It is used to improve the quality of the produced layers by protecting the substrate before and after evaporation and thus to limit the deposition duration. When the pressure and electrical heating conditions are met, the evaporation process of the mixture will take place. We used a primary pallet pump to create a primary vacuum. Afterwards, we used a secondary oil diffusion pump, equipped with a cooling system, to reach a secondary vacuum: the ultimate base pressure was 10−7 mbar. The working pressure was of the order of 10−6 mbar. The current intensity used for the evaporation of our powders was around 280 A and the deposition rate has been stabilized around 2 Å/s.

2.2 Characterization methods

In order to measure the thicknesses of the samples, t (nm), we used a Veeco Dektak 150 profilometer apparatus. The structural properties were performed by X'PERT-PRO Diffractometer System, Holland, with measurement program PANalytical of the Bragg-Brentano geometry in the “θ-2θ” scan mode, equipped with a copper anode as the source of X-rays and with Ni filter. The diffractometer is operating at 40 mA and 45 kV, using Cu Kα radiation at λCuKα = 1.54060 Å collected with PIXcel1D detector at a sweep angle 2θ ranging from 20° to 80°. A continuous scan type of all samples is carried out in a step size of 0.013° with a scan time of 18.87 s per step. The surface morphology and chemical compositions were studied by means of a scanning electron microscope (SEM, JSM-7001F from JEOL, Japan) coupled to energy dispersive X-ray analyzer facility (EDS, E2V Scientific Instruments, model 8700-2014A). The operating voltage was fixed at 15 kV. Likewise, in order to get more information about the surface topography and surface roughness measurement, we used a MFP-3D™ (SPM) classic atomic force microscope (AFM, HERZAN TS-150, ASYLUM Research, Oxford Instruments Company) in no-contact mode (AC-mode), with a scanning speed of 1Hz at room temperature. The electrical resistivity was derived from the Ecopia model No AMP55T Hall effect measurement system (HMS-5300) based on the Van der Pauw method under an excitation magnetic field (B = 0.56T) and an intensity current (I = 1 mA).

3 Results and discussions

3.1 Structural properties

We performed X-ray diffraction measurements on all Py/GaAs samples. The analysis of the diffraction diagrams with X'Pert High Score Plus software allows an estimation of the intensity and the position of the peaks. After stripping analytically the contribution of Kα2 line and the instrumental contribution (βinstrumental) from all profiles in order to identify the present phases and the preferential orientation, we studied the crystalline structure and the microstructure (lattice parameters, grain sizes, etc.).

In order to understand the change in structural properties with increasing thickness of thin Py layers deposited on GaAs substrates, X-ray diffractogram patterns for thin films are presented in Figure 1, in the diffraction angle reduced range 40–60°. The prepared samples show high intensity peaks at 2θ ≈ 44.05°, which is corresponding to (111) plane diffraction with an fcc FeNi3 crystal structure and associated to space group Pm-3m. In addition to this peak, it was observed the appearance of another peak located at 2θ ≈ 52.36° identified as FeNi3(200) phase, while thicker films, t = 263 nm and t = 277 nm, present another peak (220) of very low intensity(not shown here) situated at 2θ ≈ 75.45°. The peaks associated with substrate GaAs (220) and (222) are also present in Figure 1. The diffraction peaks of FeNi3 and GaAs were fully indexed here using the standard database PDF-2 release 2004 designed for analyzes of inorganic materials, in the reported (ICOD card number: 03-065-3244) [23] and (ICOD card number: 00-032-0389) [24], respectively. The change in thickness is accompanied by an increase in the height of the Bragg (111) diffraction peak and the appearance of new diffraction planes, indicating a change in crystal structure and an improvement in crystallinity. These results indicate that the films are polycrystalline and grow according to a normal deposition geometry in which the dominant peak is (111) and the preferred orientation in all cases is along <111> direction.

It is worth noting that our results are in agreement with other studies cited in the literature. Indeed, Sierra et al. [25] found that the Py films are highly (111) textured from an fcc structure for thicknesses between 25 and 30 nm in case of the A-SL (amorphous single-layer samples) sputtered in a high vacuum chamber at a base pressure of 10−9 Torr on quartz wafers. As for the epitaxial structure prepared by molecular beams epitaxy system at base pressure of 10−10 Torr on Si substrates, they reported that the polycrystalline structure was confirmed by the clear appearance of peak from the Py (111) plane as well as peaks from the Py (200) and Py (220) planes. Czerwinski et al. [26] observed a polycrystalline structure with the presence of a preferred orientation along the <111> direction for samples of nanocrystalline Ni-20%Fe Permalloy films with thicknesses ranging from 122 to 150 µm deposited at a current density of 100 mA cm−2 onto titanium substrates by electrodeposition. Prieto et al. [11] synthesized thin films of Py obtained by nitrogen ion beam assisted deposition (IBAD) of FeNi at 200 °C on Si (100) substrates. They found that the non-assisted Py films present only peaks of fcc Ni3Fe phase assigned to (111), (200), and (220) planes which shows a clear preferential growth of the (220) planes. As the ions assistance increases up to 50%, they observed a broadening of the diffraction peaks but this has no significant influence on the structure of the films. While for concentrations ≥75% during growth they noticed that this induces the creation of a new phase identified as an fcc γ׳-NiFe phase with a preferential orientation (111).

Zhang et al. [27] reported that the FeNi3 intermetallic compound, synthesized in situ by selective laser melting (SLM) using Fe-80%Ni alloy, present a phase with fcc structure in samples with low laser scanning velocity 0.1–0.3 m/s. When the laser scanning speed reached 0.4 m/s, the sample shows a completely different phase composition where the main phase is Fe7Ni3 with a presence of Ni single phase. Chen et al. [28] found an fcc <111> preferred orientation for Ni80Fe20/glass deposited at room temperature (RT) or post-annealed at distinct temperatures during 1 h for thicknesses between 300 and 1500 Å. On the other hand, Lottis et al. [29] reported similar diffraction peaks of Ni3Fe in their rf-sputtered (Ni80Fe20/Cu/Co/Cu) multilayers for total thickness of 2000 Å on Si (100) substrates. They further indicated that the samples were polycrystalline with [111] texturing and predominantly fcc structure.

The grain size is a parameter that ensures a quantitative appreciation of the films crystallinity. The nanocrystalline grains size of our Py/GaAs samples were determined from the X-ray diffraction spectra using the following Debye-Scherrer formula [30–32]:

(1)

were D is the crystallite size; k is the Scherrer constant, which value is close to 1; λ is the monochromatic wavelength of line Kα1 (in the present study we used λCu = 1.54060 Å). β is the angular full width at half height (FWHM in radians) of the most intense diffraction peak (111). θ is the Bragg diffraction angle at the top of the peak. The experimental measured values of D are presented in Table 1. The variation of D with thickness, t, is shown in Figure 2. We observe that the crystallite size, inversely proportional to β, estimated from the (111) peaks, displays a monotonous increase with t. Effectively, D increases promptly from 8.5976 nm for t = 132 nm to 13.7482 nm for t = 155 nm. Afterwards, it increases slightly to 13.9490 nm for t = 237 nm, and 14.7669 nm for t = 263 nm, and finally jumps to its highest value 19.6194 nm, for t = 277 nm. These results indicate that the thickest Py layer has a smaller FWHM and a better crystallinity compared to the thinnest layer. This implies that nanocrystalline crystallites tend to grow in an agglomerated manner with increasing thickness, i.e., they change from a columnar grain structure to a compact particle structure. Therefore, it is obvious that the form of the diffraction peak it represents may confirm the phase crystallinity information: an intense and finer peak (a smaller FWHM) indicates a well-crystallized phase (large grain size) without defect. On the contrary, if the diffraction peak presents a low intensity and an important enlargement (a bigger β) the diffracted phase is disordered and present a very small grain size, i.e., poorer crystalline quality. In the present study, the behavior and the measured values of D versus thickness are in accordance with the data reported in previous works. Indeed, Arab Pour Yazdi et al. [15] found that the grain size increases with the thickness from 9 to 16 nm for Ni3Fe coatings deposited by DC-magnetron sputtering on glass slides and alumina pellets. Also, Chen et al. [28] observed an increase in grain sizes from 10 to 18 nm, and from 15 to 38.6 nm when a post annealing treatment was carried out at TA = 150 °Ċ and TA = 250 °C for 1h, respectively. On another hand, they reported that grain size increases with the thickness of the film since the thicker Ni80Fe20 layer corresponds to the larger NiFe grains, and so, an increase in temperature and thickness favors crystallite growth. Chen et al. [33] found that the grains size are between 124 and 158 Å for Ni80Fe20 thin films of (300–1000) Å thickness deposited on a glass substrate by a DC magnetron sputtering system. They also indicated that the annealed NiFe film is more crystallized than the as-deposited film. Abdel-Karim et al. [34] found that the average grains size produced by the electrodeposition of the Ni-Fe nanocrystalline layer are between 20 and 30 nm for thicknesses between 140 and 230 µm and that these grains size decrease with the increase of the iron content. Svalov et al. [35] obtained values of grains size between 9 and 12 nm for Fe20Ni80 thin films in the 20–300 nm thickness range prepared by dc-sputtering at different argon pressures on glass and Si (100) substrates. They also observed that the increase of Ar pressure helps the formation of a columnar structure, which leads to a deterioration of the crystallinity of these films. On another hand, Dumpich et al. [36] reported that NixFe1-x thin films, deposited under vacuum on quartz-crystal substrates at room temperature by electron beam evaporation, present an average crystallite size D = 20 nm for layers 200 nm thick, comparable to our results.

On another hand, the lattice parametera was derived from the XRD spectra according to the well-known Bragg formula for the cubic system, equation (2) [37,38]:

(2)

where dhkl is the interatomic spacing between (hkl) planes; h, k, and l are the Miller indices of the most intense (111) peak . The computed values of the lattice constant are displayed in Table 1. In Figure 3, we show the evolution of the lattice constant, a, as a function of Py films thickness. We observed that a monotonously increases with increasing thin film thickness. Its value varies from a = 3.5224 Ǻ for 132 nm to a = 3.5573 Ǻ for 277 nm. Moreover, we note that for the samples with thickness 132, 155 and 237 nm, the obtained values of a are somewhat lower than the bulk one but for the thicker films with thickness 263 and 277 nm, we found that a is slightly greater than the bulk value, the bulk lattice parameter value of Ni3Fe being equal to 3.5523 Ǻ [23]. These results infer that the thinner films are under a compressive stress while the thicker ones are under a tensile stress. The lattice parameter measured in this paper is smaller than that of Permalloy produced by electron-beam co-evaporation a= 3.56 and 4.08 Ǻ, reported by Pazukha et al. [39] and, a = 3.573 Ǻ, reported by Zubar et al. [40], in their thin films obtained by pulsed electrodeposition technique. Furthermore, using the measured values of the cell parameter of the thin films and the mass lattice parameter value of the bulk material, we computed the out-of-plane stress, ε, defined as . The obtained values of ε (%) for all samples are summarized in Table 1. These experimental measured values of ε specify that the samples be thus submitted to a very low stress due probably to the mismatch lattice parameter between Py thin film and GaAs substrate. For the thinner films, we also observe that the values of the ε microstrains are negative and start from −0.8417 for t = 132 nm to −0.1463% for t = 237 nm. This indicates that these films are under a compressive stress. For the samples thicker than 237 nm, the micro-deformation has again increased further to become positive for samples 263 and 277 nm thick, where ε = +0.0957 and ε = +0.1407%, respectively. For these thicker films, we can deduce that the samples are under a tensile stress. These two results corroborate the conclusions found previously using the cell parameter values. Figure 2 shows the evolution of the micro-stress as a function of Py films thickness. It is also important to note the proportional correlation between the evolution of the stress and the crystallites size: the increase of <D> is accompanied by an increase in strain ε. The magnetic film thickness and the surface state of the substrate may be the cause for these variations.

thumbnail Fig. 1

X-rays diffractometry patterns of Py thin films with different thicknesses: (a) 132 nm, (b) 155 nm, (c) 237 nm, (d) 263 nm and (e) 277 nm.
Table 1

Diffraction angle, 2θ, crystallite size D, lattice parameter a, microstrain ε (%) and root mean square (rms) for different Py/GaAs thin film thickness,.

thumbnail Fig. 2

Variations of crystallite size <D> (nm) and microstrain ε (%) as functions of film thickness.
thumbnail Fig. 3

Lattice parameter a as a function of film thickness.

3.2 Scanning electron microscopy (SEM) and energy dispersive X-ray (EDX) spectroscopy

The effect of thickness on the morphology of Py deposits has been evaluated using a scanning electron microscope equipped with energy dispersive X-ray (SEM-EDXS) measurements tool. The analysis was performed with respect to the uniformity of the films, their grain size and the presence/absence of microvoid and microcrack.

For the quantitative analysis, in order to estimate the average elemental and the chemical composition with a good accuracy, we have performed standardless ZAF simulation of the EDX spectra for all Py samples. The estimated compositional analysis show that the weight percentage (wt.%) ratio of Ni and Fe is close to the starting powder with a deviation not exceeding 3% to that of Permalloy (Ni81Fe19%). These results corroborate the stoichiometry of the prepared NiFe film. A typical experimentally specimen of EDX spectra, for the 132 nm thick film, is presented in Figure 4, where the weight percentage ratio is 83.145%Ni and 16.855%Fe. It is noted that EDX results indicate that only peaks pertaining to Ni and Fe elements are observed with substrate peaks belonging to GaAs(100) and that no carbon contamination was observed.

Figure 5 shows SEM micrographs of Py films versus thickness. The micrographs under SEM observations clearly revealed a very smooth uniform structure, with regular, homogeneous surface morphology and without cracks. Moreover, we see the formation of clusters or islands not uniformly distributed. These islands are separated on the surface by grain boundaries that appear to be very dense, with many fine grains of similar size and generally having the shape of spherical balls. These growths may be explained by the coalescence and agglomeration of nanoparticles with various forms, having average grain size ranging from 22 to 39 nm. However, it seems to be difficult, with this magnification, to distinguish easily the effect of the increase in thickness on the surface morphology. Hence, complementary observations have been performed using Atomic Force Microscopy (AFM) technique (see Sect. 3.3). Similar observations have been reported by other researchers. Balachandran et al. [41] studied NiFe films, electrodeposited on Cu substrates and prepared by galvanostatic mode with and without the presence of an ultrasonic field at different pulse current. The SEM images taken at a current of 40 mA and a duty cycle of 50.00% under ultrasonic treatment reveal that the surface roughness is reduced significantly from 39.01 nm to 6.96 nm and the spherical shape of the particle in the range from [579.40–623.30] to [29.00–49.90] nm. Saeed et al. [42] reported grains of NiFe/Cu of spherical shape and an average size of 66–93 nm in their electrodeposited thin films. The average surface roughness is found to be in the range of 4.82–8.73 nm and it decreased with the deposition time decreasing. On another hand, Moniruzzaman et al. [43] studied the effect of Ni/Fe ratio (of 1 and 12) in simple and complex baths, as well as the variation of the current density (from 30 to 100 mA /cm2) on the structure of Ni/Fe. They report that the increase in Ni/Fe ratio in the bath as well as current density results in decreasing grain size of the deposits. They obtained a different morphology characteristic of the Fe-Ni films, such as the non-smooth surface with the presence of microcracks from simple bath; these microcracks being due probably to the stress developed during electrodeposition.

thumbnail Fig. 4

Example of EDS spectrum: Py/GaAs (1 0 0) sample, 132 nm thick.
thumbnail Fig. 5

SEM micrographs of Py films deposited on GaAs (100) for different thicknesses: (a) 132 nm, (b) 155 nm, (c) 237 nm, (d) 263 nm, and (e) 277 nm.

3.3 Atomic force microscopy (AFM) observations

Atomic force microscopy is a non-destructive method for exploring the topography of surfaces as well as the individual position of atoms and possibly their nature. The observations of the overall surface morphology using 3D and 2D views, as well as the average root mean square (rms) roughness of our Py thin films, have been achieved by means of MFP-3D™ atomic force microscope, under the experimental conditions detailed in Section 2.2. Specimen of typical 2D and 3D AFM images, scan area 8 μm × 8 μm, of Py films, are shown in Figure 6. The rms roughness values, measured for all films, are mentioned in Table 1.

The analysis of the AFM surface images provided the following results. For the 132 nm thick film, the image shows several islands in the form of large peaks dispersed on a very smooth, homogeneous surface whose average roughness is 10.552 nm. For the 155 nm thick film, the AFM image presents a smooth surface with undulations and some intense peaks whose average roughness increases to be 14.358 nm. For the other samples with thickness ranging from 237 to 263 nm, we observe an improvement of the structure, where the islands are practically identical and agglomerated in a collective way with an average roughness approximately equal to 0.719 nm, which reflects a better crystallinity. Finally, the thickest sample, 277 nm thick, displays a slightly non-homogeneous distribution of peaks on the entire surface and shows unclear grain boundaries with an agglomerated structure, composed of cluster grains, with roughness equal to 3.145 nm.

Other researchers have reported similar findings. Indeed, Zhang et al. [22] observed that Permalloy thin films, prepared by DC magnetron sputtering technique on SiO2 substrates, showed that with increasing temperature, the film grows better on islands on the substrate surface, and they observed that with increasing time, the film clusters become uniform. Chaudhari et al. [44] studied the impact of the incorporation of ZrO2 particles on the properties of Ni–Fe alloy electrodeposited from ethylene glycol bath. They reported that the Ni-Fe/ZrO2 nanocomposite coatings exhibit a hill-valley morphology composed of many small grains with a spherical structure and uniform distribution. They explained that this morphology is a result of incomplete nucleation and irregular grain formation. In addition, they found that the Ni-Fe/ZrO2 nanocomposite was smoother than the Ni-Fe alloy that has more nodular and granular surface morphology, with a high marginal rms greater than 77 nm. Yi et al. [45] investigated the effect of the sputtered Cu substrate on the growth of Permalloy films. They reported that the Ni79Fe21 thin films electrodeposited on 100 °C-sputtered Cu layer showed the most uniform surface with the smallest particle size than other films and have the smoothest surface with an rms near 2 nm. In addition, they indicated that the surface roughness and the grain size of the films increase monotonically with temperatures higher than 100 °C.

thumbnail Fig. 6

Specimen of 2D and 3D AFM topographic surface images for 237 and 263 nm thick Py films. The size of the scanning area is 8 μm × 8μm.

3.4 Electrical properties

Measurements of electrical resistivity, magnetoresistance and mobility of the Py thin films were carried out with an Ecopia HMS 5300 Hall Effect Measurement System and Van der Pauw geometry, at room temperature under the experimental conditions described in Section 2. The numerical data of the mean values of the electrical resistivity, magneto-resistance and mobility, for the Py samples, at room temperature, are given as function of film thickness ranging from 132 to 277 nm in Table 2. The electrical resistivity ρ is obtained by the multiplication of the sheet resistance R and the film thickness t. The variation of the computed resistivity ρ of the deposited Py material as a function of film thickness ranging from 132 to 277 nm, at room temperature, with an accuracy equal to 5%, is depicted in Figure 7. The bold squares are the experimental data while the continuous line is a fitted curve. As we can see from this figure, the resistivity shows a pronounced decrease with increasing film thickness. This decrease follows the well-known Fuchs/Sondheimer Model [46]. This might be sustained by the diminution of continuity of electron percolation due to the decrease of metallic nanogranular dimensions, as reported by Ma et al. [47]. We observe that the resistivity of our thin films decreases by seven orders approximately, from 85.40 to 12.32 μΩ cm, when the film thickness increases from 132 to 277 nm. Relatively higher resistance for a thinner film demonstrates the effects of substrate-film interface. The value of the resistivity for the thinnest film (t = 132 nm) is 85 μΩ.cm, which is relatively equal to that of FeTaN films (∼80 μΩ.cm) [48], whereas that of a thicker film (t = 237 nm) is about 31 μΩ.cm and similar to that of Co–Al–O materials [49], Permalloy [50], and FexNi100-x films [17]. Hence, such a monotonous decrease of ρ with the thickness may be attributed to the contributions of the scattering of the conduction electrons relating to the grain boundaries scattering due to the increase of the crystallite size, and therefore, the decrease of the number of the grain boundaries, as can be seen in Figure 2. The high resistivity is undoubtedly an advantage for the applications of the soft magnetic films since the eddy current loss can be reduced by an increasing electrical resistance due to an increasing skin depth [51]. Other studies relating to the dependence of the resistivity ρ on film thicknesses are described in the literature [24,52].

The variations of the magnetoresistance of our films as a function of the thickness are shown in Figure 8. Briefly and concisely, the magnetoresistance of a material is defined as the change in its resistivity when subjected to an applied magnetic field. The experiment shows that the thickness, the grains size, the morphology and the conditions of elaboration of the samples, influence in a significant way the rate of magnetoresistance.

When scrutinizing Figure 8, we clearly observe that the magnetoresistance decreases monotonously with the increase of film thickness. The MR decreases from 8.263× 10−3 Ω, for the 132 nm thick sample, to 2.164 × 10−3 Ω (four times smaller), for the 277 nm thick sample. This decrease in magnetoresistance could be related to the surface morphology, since the crystallite size increases with increasing thickness, as highlighted previously in Table 1. Indeed, when the film thickness increases, the increased crystallite size, and hence the reduction in the number of crystallite boundaries, will provide a longer mean free path for the conduction electrons, resulting in a reduction of the magnetoresistance. The same trend has been reported in the literature [53].

The evolution of mobility as a function of film thickness is displayed in Figure 9. This figure reveals a linear growth of the mobility with the increase of the film thickness. Subsequently, these results may be due to the improved crystallinity of all deposited thin films, since the formation of large grain sizes with increasing thickness is accompanied by a reduction in the number of grain boundaries as mentioned above, and the mobility of electrons is then enhanced. Similar descriptions are reported by Pazukha et al. in their Ni80Fe20 (Py) and Ag prepared by electron-beam co-evaporation technique [39].

Table 2

Values of electrical resistivity ρ, magnetoresistance ΔR, and mobility μ as functions of Py thickness.

thumbnail Fig 7

Electrical resistivity ρ as a function of thickness t, for Py thin films.

thumbnail Fig. 8

Magnetoresistance as a function of thickness t, for Py thin films.
thumbnail Fig. 9

Evolution of the mobility μ as a function of thickness t, for Py thin films.

4 Conclusions

In this work, we have investigated the effect of an increasing thickness on the properties of Permalloy thin films prepared by thermal evaporation under vacuum onto GaAs(100) substrates at room temperature. The XRD results indicate that the obtained Py samples are polycrystalline and exhibit the face-centered cubic Ni3Fe structure, with a preferential orientation <111> and without reflections corresponding neither to Ni nor to Fe. The crystallite size and the lattice parameter increase with the increase in the thickness. This is accompanied with small microstrains, due probably to the mismatch of lattice parameters between Py thin film and GaAs substrate. The SEM observations show a very dense surface with coalescing and agglomerating nanoparticles with a diameter of about [20–28] nm and without any presence of micro cracks on the surface. From AFM observations, most of the layers are smooth with roughness not exceeding ∼14 nm, and the best surface quality, reflecting good crystallinity, was observed for the 263 nm thick film with an average rms of 0.641 nm. Both the electrical resistivity and magnetoresistance decrease with increasing thickness, this evolution being due to the surface morphology of the films, while the mobility gradually increases as the Py film thickness increases.

Accepted principles of ethical and professional conduct have been followed by the authors.

This work is funded under the grant of Algerian Directorate for Scientific Research and Technological Development(DGRSDT).

The authors have no relevant financial or non-financial interests to disclose.

The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.

The authors warmly thank the Director of LCIMN laboratory, Prof. Amor AZIZI and his team, for alloying Mrs. Ounissa CHERRAD to carry out the AFM images in their laboratory.

Author contribution statement

All authors contributed to the study conception and design. Material preparation were performed by A. Kharmouche. Data collection and characterizations were carried out by O. Cherrad. First analysis were performed by O. Cherrad. The final manuscript was written by A. Kharmouche. All authors read and approved the final manuscript.

References

  1. V. Torabinejad, M. Aliofkhazraei, S. Assareh, M.H. Allahyarzadeh, A. Sabour Rouhaghdam, J. Alloys Compd. 691, 841 (2017) [CrossRef] [Google Scholar]
  2. S.M. Cherif, Y. Roussigné, A. Kharmouche, T. Chauveau, D. Billet, Eur. Phys. J. B 45, 305 (2005) [CrossRef] [Google Scholar]
  3. A.L.R. Souza, M.R. Araujo, W. Acchar, R.D. Della Pace, A.S. Melo, F. Bohn, M.A. Correa, Appl. Phys. A 125, 236 (2019) [CrossRef] [Google Scholar]
  4. A. Melloul, A. Kharmouche, J. Mater. Sci.- Mater. Electron 30, 13144 (2019) [Google Scholar]
  5. X. Zhao, Y. Dang, H. Yin, Y. Yuan, J. Lu, Z. Yang, Y. Gu, J. Alloys Compd. 644, 66 (2015) [CrossRef] [Google Scholar]
  6. Y. Liu, Z. Lan, Z. Yu, R. Guo, X. Jiang, C. Wu, K. Sun, Appl. Phys. A 126, 792 (2020) [CrossRef] [Google Scholar]
  7. P. Laha, B.K. Mahato, R. Gayen, S. Datta, R. Singh Rawat, Appl. Phys. A 128, 394 (2022) [CrossRef] [Google Scholar]
  8. A.V. Trukhanov, S.S. Grabchikov, A.A. Solobai, D.I. Tishkevich, S.V. Trukhanov, E.L. Trukhanov, J. Magn. Magn. Mater. 443, 142 (2017) [Google Scholar]
  9. H. Han, Y. Li, W. Su, L. Chen, J. Zhang, Appl. Phys. A 113, 385 (2013) [CrossRef] [Google Scholar]
  10. M. Tinouche, A. Kharmouche, B. Aktas, F. Yildiz, A.N. Kocbay, J. Supercond. Nov. Magn. 28, 921 (2015) [Google Scholar]
  11. P. Prieto, J. Camarero, J.F. Marco, E. Jiménez, J.M. Benayas, J.M. Sanz, IEEE Trans. Magn. 44, 3913 (2008) [CrossRef] [Google Scholar]
  12. J.M. Wood, C.I. Oseghale, O. Cespedes, M. Grell, D.A. Allwood, J. Appl. Phys. 124, 085304 (2018) [CrossRef] [Google Scholar]
  13. V.K. Belyaev, D.V. Murzin, N.N. Perova, A.A. Grunin, A.A. Fedyanin, V.V. Rodionova, J. Magn. Magn. Mater. 482, 292 (2019) [Google Scholar]
  14. S. Guan, B.J. Nelson, J. Electrochem. Soc. 152, C190 (2005) [CrossRef] [Google Scholar]
  15. M. Arab Pour Yazdi, N. Fenineche, E. Aubry, A. Kaibi, A. Billard, J. Alloys Compd. 550, 252 (2013) [CrossRef] [Google Scholar]
  16. X. Zhu, Z. Wang, Y. Zhang, L. Xi, J. Wang, Q. Liu, J. Magn. Magn. Mater. 324, 2899 (2012) [Google Scholar]
  17. N. Guechi, A. Bourzami, A. Guittoum, A. Kharmouche, S. Colis, N. Meni, Physica B 441, 47 (2014) [Google Scholar]
  18. S. Riemer, J. Gong, M. Sun, I. Tabakovic, J. Electrochem. Soc. 156, 439 (2009) [Google Scholar]
  19. M. Kateb, H. Hajihoseini, J.T. Gudmundsson, S. Ingvarsson, J. Phys. D: Appl. Phys. 51, 285005 (2018) [Google Scholar]
  20. B.A. Belyaev, N.M. Boev, A.V. Izotov, G.V. Skomorokhov, P.N. Solovev, Russ. Phys. J. 63, 16 (2020) [Google Scholar]
  21. Z. Wang, B. Dai, Y. Ren, S. Tan, J. Ni, J. Li, J. Mater. Sci.- Mater. Electron 30, 18328 (2019) [Google Scholar]
  22. M. Zhang, C. Deng, J. Mater. Sci.- Mater. Electron 32, 4949 (2021) [Google Scholar]
  23. R.J. Wakelin, E.L. Yates, Proc. R. Soc. London Ser. B B66, 221 (1953) [Google Scholar]
  24. P. Gong, Polytechnic Inst. of Brooklyn, NY, USA, ICCD Grant-in-Aid (1981) [Google Scholar]
  25. J.F. Sierra, V.V. Pryadun S.E. Russek, M. García-Hernández, F. Mompean, R. Rozada, O. Chubykalo-Fesenko, E. Snoeck, G.X. Miao, J.S. Moodera, F.G. Aliev, J. Nanosci. Nanotechnol. 11, 1 (2011) [Google Scholar]
  26. F. Czerwinski, J.A. Szpunar, U. Erb, J. Mater. Sci.- Mater. Electron 11, 243 (2000) [Google Scholar]
  27. B. Zhang, N.-E. Fenineche, H. Liao, C. Coddet, J. Magn. Magn. Mater. 336, 49 (2013) [Google Scholar]
  28. Y.-T. Chen, C.-W. Chen, T.-H. Wu, Appl. Surf. Sci. 351, 946 (2015) [CrossRef] [Google Scholar]
  29. D. Lottis, A. Fert, R. Morel, L.G. Pereira, J.C. Jacquet, P. Galtier, J.M. Coutellier, T. Valet, J. Appl. Phys. 73, 5515 (1993) [CrossRef] [Google Scholar]
  30. A. Bourezg, A. Kharmouche, Vacuum 155, 612 (2018) [Google Scholar]
  31. A. Kharmouche, J. Nanosci. Nanotechnol. 11, 4757 (2011) [Google Scholar]
  32. B.D. Cullity, Elements of X-Ray Diffraction (Addison-Wesley, Reading, MA, 1978) [Google Scholar]
  33. Y.-T. Chen, J.-Y. Tseng, T.-S. Sheu, Y.C. Lin, S.H. Lin, Thin Solid Films 544, 602 (2013) [Google Scholar]
  34. R. Abdel-Karim, Y. Reda, M. Muhammed, S. El-Raghy, M. Shoeib, H. Ahmed, J. Nanomater 2011, 519274 (2011) [Google Scholar]
  35. A.V. Svalov, I.R. Aseguinolaza, A. Garcia-Arribas, I. Orue, J.M. Barandiaran, J. Alonso, M.L. Fernández-Gubieda, G.V. Kurlyandskaya, IEEE Trans. Magn. 46, 333 (2010) [CrossRef] [Google Scholar]
  36. G. Dumpich, E.F. Wassermann, V. Manns, W. Keune, S. Murayama, Y. Miyako, J. Magn. Magn. Mater. 67, 55 (1987) [Google Scholar]
  37. A. Kharmouche, A. Melloul, J. Mater. Sci.- Mater. Electron 31, 19680 (2020) [Google Scholar]
  38. A. Clearfield, J. Reibenspies, N. Bhuvanesh, Principles and Applications of Powder Diffraction (Blackwell Publishing Ltd., 2008) [Google Scholar]
  39. I.M. Pazukha, О.V. Pylypenko, L.V. Odnodvorets, Mater. Res. Express 5, 106409 (2018) [Google Scholar]
  40. T. Zubar, V. Fedosyuk, D. Tishkevich, O. Kanafyev, K. Astapovich, A. Kozlovskiy, M. Zdorovets, D. Vinnik, S. Gudkova, E. Kaniukov, A.S.B. Sombra, D. Zhou, R.B. Jotania, C. Singh, S. Trukhanov, A. Trukhanov, J. Nanomater. 10, 1077 (2020) [Google Scholar]
  41. R. Balachandran, H.K. Yow, B.H. Ong, K.B. Tan, K. Anuar, H.Y. Wong, Int. J. Electrochem. Sci. 6, 3564 (2011) [Google Scholar]
  42. A. Saeed, B. Ruthramurthy, W.H. Yong, O.B. Hoong, T.K. Ban, Y.H. Kwang, S. Sreekantan, Russ. J. Electrochem. 52, 788 (2016) [Google Scholar]
  43. M. Moniruzzaman, K.M. Shorowordi, A. Azam, M.F.N. Taufique, J. Mech. Eng. 44, 51 (2014) [Google Scholar]
  44. A.K. Chaudhari, V.B. Singh, Funct. Compos. Struct. 1, 045006 (2019) [CrossRef] [Google Scholar]
  45. J.B. Yi, X.P. Li, J. Ding, H.L. Seet, J. Alloys Compd. 428, 230 (2007) [CrossRef] [Google Scholar]
  46. E.H. Sondheimer, Adv. Phys. 1, 1 (1952) [CrossRef] [Google Scholar]
  47. Y.G. Ma, Y. Liu, C.Y. Tan, Z.W. Liu, C.K. Ong, J. Appl. Phys. 100, 054307 (2006) [CrossRef] [Google Scholar]
  48. V. Bekker, K. Seemann, H. Leiste, J. Magn. Magn. Mater. 296, 37 (2006) [Google Scholar]
  49. Z.W. Liu, Y. Liu, C.Y. Tan, Y.G. Ma, C.K. Ong, J. Magn. Magn. Mater. 313, 37 (2007) [Google Scholar]
  50. A.F. Mayadas, J.F. Janak, A. Gangulee, J. Appl. Phys. 45, 2780 (1974) [CrossRef] [Google Scholar]
  51. B. Viala, M.K. Minor, J.A. Barnard, J. Appl. Phys. 80, 3941 (1996) [CrossRef] [Google Scholar]
  52. M. Kateb, E. Jacobsen, S. Ingvarsson, J. Phys. D: Appl. Phys. 52, 075002 (2019) [Google Scholar]
  53. M.A. Shazni, M. Haniff, H.W. Lee, D.C.S. Bien, I.H. Abd. Azid, M.W. Lee, S.S. Embong, Thin Solid Films 550, 22 (2014) [Google Scholar]

Cite this article as: Ounissa Cherrad, Ahmed Kharmouche, Thickness dependent physical properties of evaporated permalloy/GaAs(100) thin films, Eur. Phys. J. Appl. Phys. 98, 14 (2023)

All Tables

Table 1

Diffraction angle, 2θ, crystallite size D, lattice parameter a, microstrain ε (%) and root mean square (rms) for different Py/GaAs thin film thickness,.

In the text
Table 2

Values of electrical resistivity ρ, magnetoresistance ΔR, and mobility μ as functions of Py thickness.

In the text

All Figures

thumbnail Fig. 1

X-rays diffractometry patterns of Py thin films with different thicknesses: (a) 132 nm, (b) 155 nm, (c) 237 nm, (d) 263 nm and (e) 277 nm.
In the text
thumbnail Fig. 2

Variations of crystallite size <D> (nm) and microstrain ε (%) as functions of film thickness.
In the text
thumbnail Fig. 3

Lattice parameter a as a function of film thickness.
In the text
thumbnail Fig. 4

Example of EDS spectrum: Py/GaAs (1 0 0) sample, 132 nm thick.
In the text
thumbnail Fig. 5

SEM micrographs of Py films deposited on GaAs (100) for different thicknesses: (a) 132 nm, (b) 155 nm, (c) 237 nm, (d) 263 nm, and (e) 277 nm.
In the text
thumbnail Fig. 6

Specimen of 2D and 3D AFM topographic surface images for 237 and 263 nm thick Py films. The size of the scanning area is 8 μm × 8μm.
In the text
thumbnail Fig 7

Electrical resistivity ρ as a function of thickness t, for Py thin films.

In the text
thumbnail Fig. 8

Magnetoresistance as a function of thickness t, for Py thin films.
In the text
thumbnail Fig. 9

Evolution of the mobility μ as a function of thickness t, for Py thin films.
In the text

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.