Microstructural and Morphological Characterization of the Cobalt-Nickel Thin Films Deposited by the Laser-Induced Thermionic Vacuum Arc Method

Abstract

Laser Induced-Thermionic Vacuum Arc (LTVA) technology was used for depositing uniform intermetallic CoNi thin films of 100 nm thickness. LTVA is an original deposition method using a combination of the typical Thermionic Vacuum Arc (TVA) system and a laser beam provided by a QUANTEL Q-Smart 850 Nd:YAG compact Q-switched laser with a second harmonic module. The novelty is related to the simultaneous deposition of a bi-component metallic thin film using photonic processes of the laser over the plasma deposition, which improves the roughness but also triggers the composition of the deposited thin film. Structural analysis of the deposited thin films confirms the formation of face-centered cubic (fcc) as the main phase CoNi and hexagonal Co₃Ni as the minority phase, observed mainly using high-resolution transmission electron microscopy. The magneto-optical measurements suggest an isotropic distribution of the CoNi alloy thin films for the in-plan angular rotation. From the low coercive field of H_c = 40 Oe and a saturation field at 900 Oe, the CoNi thin films obtained by LTVA are considered semi-hard magnetic materials. Magnetic force microscopy reveals spherical magnetic nanoparticles with mean size of about 40–50 nm. The resistivity was estimated at $\rho$ = 34.16 μΩ cm, which is higher than the values for bulk Co and Ni.

1. Introduction

Cobalt-nickel alloys were recently used as a metal-organic framework (MOF) in an efficient bio-sensing platform for a series of organic building blocks for sensitive detection of miRNA, acting as metal centers or clusters [1]. Several attempts have been made using these CoNi alloys as electro-catalysts for boosting oxygen evolution; in addition, reduction reactions have been obtained due to their strong coupling properties [2]. Among the carbon nanostructures, coated ultrafine CoNi alloy nanocomposites can be used as stable trifunctional electrocatalysts for energy conversion and storage devices [3]. When CoNi alloy nanoparticles were functionalized with nitrogen-doped carbon nanostructures, the obtained nanoparticles acted as a bifunctional electrocatalyst for water splitting technologies [4].

Nevertheless, the most important application of CoNi alloys is in the field of media for magnetic or optical recording and, lately, for magnetography and MEMS applications; this is primarily due to their semi-hard magnetic properties [5]. These materials, called glass metallic or intermetallic compounds, exhibit intermediate magnetic properties between soft and hard magnets. As primarily characteristics, these alloys exhibit low coercive fields, ranging from 40 to 1000 Oe, and relatively high magnetization at saturation M_s, medium magnetization remanence M_r, and medium saturation field. These properties only require a low energy cost to magnetize different materials when compared with hard magnets, making them very useful in the magnetic recording media, but they depend on the preparation conditions, including deposition, [6] concentration [7], film thickness [8], gain size [9], and thermally annealing after deposition [10].

The coercive field is considered a nonintrinsic property, which is highly dependent on the obtained method [11,12]. More than that, the existence of two crystalline phases, the hexagonal one (P63/mmc) and the cubic one (Fm-3m), but also the interplay between the amorphous and crystalline structures, may influence the magnetic properties, including magnetization at saturation and anisotropy.

Concerning production methods, many cited studies deal with electrochemical deposition films with thicknesses varying from a few hundred nanometers up to a few microns [13,14,15]. This relatively high thickness is necessary because physical deposition techniques have largely shown poor magnetic properties for the perpendicular magnetic recording application [16,17,18].

In this context, an optimal method to obtain controlled CoNi thin films with semi-hard magnetic properties is the Laser-induced Thermionic vacuum arc (LTVA) method [19]. Using this improved technology, based on the classical Thermionic vacuum arc (TVA) method [20], the laser-induced plasma modifications provide uniform intermetallic thin films with thickness up to 100 nm with smooth surface, high purity, and tailorable surface properties.

The interplay between Thermionic vacuum arc (TVA), which produces coatings with high roughness, and the addition of the photonic processes enhances the control between the primary compounds, enabling alloys with different ratios of initial compounds to be obtained.

This paper describes a versatile method for producing CoNi thin films with semi-hard magnetic properties and controllable thickness. The interplay between hexagonal and cubic structures of Co and Ni combinations results in uniform CoNi thin films with relatively lower electrical resistivity and isotropic magnetic orientation. The deposited CoNi thin films were structurally analyzed using X-ray diffraction (XRD) and X-ray reflectometry (XRR) techniques for crystallography, thickness and density of the metallic thin films, scanning electron microscopy (SEM) and energy dispersive spectrometer (EDX) for compositional analysis, high resolution transmission electron microscope (HRTEM) for identification of crystalline phases below 5% concentration, and atomic force microscopy (AFM) and magnetic force microscopy (MFM) for the local magnetism. In addition, the magnetic and electrical properties were also investigated by the magneto-optical Kerr and four-point probe techniques.

2. Materials and Methods

2.1. The LTVA System

A schematic overview of the LTVA system is depicted in Figure 1. The LTVA configuration consists of a classical TVA device upgraded with an adjustable power laser beam. The configuration of the TVA system is presented elsewhere [21].

Briefly, TVA consists of a tungsten-heated cathode surrounded by a Wehnelt cylinder that concentrates the electron beam on the anode material. In this way, continuous evaporation of the anode material occurs and ensures the vapor density of the deposited material in the pre-breakdown state. By varying the applied high voltage between the electrodes at a certain value, bright plasma ignites. The discharge sustaining gas is simply the evaporating atoms in a vacuum. The base pressure is typically p = 4 × 10^–5 Pa and the work pressure is two orders of magnitude higher throughout the deposition process. Therefore, thin film deposition occurs under high purity conditions [22]. Multicomponent thin films and single thin films were deposited using the Thermionic Vacuum Arc (TVA) method [23,24,25,26,27]. During the last decades, this technology has been successfully used for the synthesis of a large range of metals, oxides, refractory materials, and complex composites [28,29,30,31].

In the Laser-induced Thermionic Vacuum Arc–LTVA, the laser beam is provided by a QUANTEL Q-Smart 850 Nd:YAG compact Q-switched laser with a second harmonic module. The laser operates at a repetition rate of 10 Hz and the pulse-width is 5 ns for a wavelength of λ = 532 nm. The laser beam was incident on the front window of the coating chamber, exactly in the middle of the interelectrode gap. Using a pulsed laser for discrete photoexcitation of the element results in significant perturbation of the thermal energy-level distribution of those atoms.

The materials used for deposition in our experiment were Co granules (99.95%) and Ni granules (99.996%, metal basis), provided by the Alfa Aesar company, Kandel, Germany. The samples were deposited onto Si wafers (single-side silicon oxide substrates of 1 × 1 cm size). Before their introduction into the chamber, the substrates were cleaned to remove surface contamination. An ultrasonic bath with a highly effective cleaner (acetone) is the typical cleaning method. The substrate holder was grounded, and no substrate bias was applied.

The main experimental parameters are summarized in

Table 1. Main deposition conditions for CoNi thin films deposited using the LTVA method.

Experimental Conditions	Ua [V]	Ia [A]	E [mJ]	IL [GW/cm2]
	900	0.205	50	10
If [A]	55
Ub [V]	3000
p [Pa]	3.4 × 10–5
tdep [s]	420
d [nm]	95

, as follows: U_a is the applied high voltage over the electrodes; I_a is the arc current, E is the energy/pulse of the laser, I_L is the irradiance of the laser, I_f is the current of the filament heating; U_b is the breakdown voltage of the plasma; and p is the pressure inside the deposition chamber. The parameter t_dep represents the deposition time and d is the thickness of the thin film measured in situ by a film thickness monitor (Cressington MTM).

To avoid other oxidative processes after deposition, the samples were allowed to cool in the deposition chamber for 24 h before any further characterization.

2.2. Analysis of the CoNi Thin Films

The diffraction patterns and thin-film reflectivity were measured using a BRUKER D8 ADVANCE type X-ray diffractometer (XRD), using CuK_α1 radiation (λ = 1.54056 Å) at 40 kV and 40 mA. The analysis of the obtained patterns, such as the peak width, was performed after subtraction of K_α1 lines from the K_α1–K_α2 doublet; this was conducted with the Rachinger algorithm using the Bruker Difracplus Basic Evaluation program package [32]. The main XRD parameters were refined using Crystal Sleuth software [33] to identify the crystallographic behavior of the deposited metallic layers. The 2θ scan range was set at 30–80° with a step size of 0.05° and 0.01° resolution. Currently, Z-scan, theta rocking, and 2θ/ω steps were performed for the reflectivity of each sample.

The study of film microstructure was conducted using a Phillips CM 120 ST transmission electron microscope (TEM) equipped with HR-TEM and Selected Area Electron Diffraction (SAED) accessories. The distribution over dimensions was fitted with a log–normal function in SciDAVIS software, which gives the grain size distribution. The SAED results were used for indexing the main diffraction peaks and determining the lattice parameters. Furthermore, iTEM software (Radius 2.0 version, Emsis, Muenster, Germany) was used for imaging connected with MegaView III on the CM120ST microscope. The scratch method was used to prepare the samples for TEM analysis, and was conducted using a diamond knife and formvar-covered grids [34].

Scanning electron microscopy (SEM) and Energy-dispersive X-ray (EDX) were performed with an Evo 50 XVP with EDAX attachment (Carl Zeiss NTS) equipped with LaB6 cathode; this component works in a low vacuum mode when paired with an EDX accessory, down to 133 eV resolution.

Atomic Force Microscope (AFM) measurements were performed with an MFP-3D Stand Alone (MFP-3D-SA) microscope, enabling a scan range of XY > 90 μm and Z > 15 microns. The microscope was used in two modes: topological investigation-AFM mode and magnetic measurements-Magnetic Force Microscopy (MFM) mode. In the second mode, the phase shift Δφ of the oscillating cantilever at fixed lift height Δz was recorded. Typical dimensions of a cantilever length of 200 μm have a tip length of 4 μm, a diameter of 50 nm, and a distance of 30 nm from the surface. For MFM measurements, the cantilever tip was covered with CoCr magnetic alloy. The force on the magnetic tip is detected by measuring the displacement at the end of the cantilever, usually by optical means. An image of the magnetic stray field is obtained by slowly scanning the cantilever over the sample surface, in a raster-like fashion.

The longitudinal Magneto-Optical Kerr Effect (MOKE) was obtained using a home-made device. A beam of He-Ne laser (wavelength λ = 633 nm) through a polarizer was incident on the area around A = 3 mm² of each magnetizing cobalt-nickel film. The reflected beam passed through the other polarizer before reaching the photodiode detector. The second polarizer, acting as an analyzer of polarization, was aligned at the angle θ with respect to the first polarizer. The variation in the reflected beam intensity (ΔI_out) is directly proportional to the Kerr rotation (Δθ) and to the magnetization (M):

$$\Delta {\mathrm{I}}_{\mathrm{out}}=-{\mathrm{I}}_{\mathrm{in}}\times \sin \left(2\mathsf{\theta }\right)\times \Delta \mathsf{\theta }=-{\mathrm{I}}_{\mathrm{in}}\times \sin \left(2\mathsf{\theta }\right)\times K\times \mathrm{M}$$

(1)

where K is a constant of experimental set-up and M is the magnetization. Typically, the MOKE curves are given as a variation of the angle between the incident and reflected light as a function of the applied magnetic field. To avoid the usage of the K constant, the hysteresis loops can be expressed in terms of magnetization normalized to the saturation magnetization (M/M_S) as a function of the applied magnetic field [35].

Sheet resistances and electrical conductivities were measured using Four-Point probe equipment (Ossila) on the rectangular samples. The method is based on a current source applied on the metallic thin film between two point-like electrodes, leading to a logarithmic potential, according to the following formula:

$$\mathrm{V}=-\frac{I\times R_s}{2\mathsf{\pi }}\times \ln r$$

(2)

where V represents the difference in the potential between the two points situated at 1.27 mm in this study, I is the applied current, R_s is the sheet resistance, and r is the distance from the current source [36]. For a rectangular sample of 15 × 15 mm, the resistivity $\mathbf{\rho }$ of a thin layer material is given by:

$$\rho =R_s\times \mathrm{d}=\frac{\Delta \mathrm{V}}{\Delta \mathrm{I}}\times \frac{\mathsf{\pi }}{\ln 2}\times \mathrm{C}\times \mathrm{d}$$

(3)

where d is the thickness of the thin film and C is a correction coefficient that considers the geometry of each sample [37].

3. Results

3.1. Microstructural Analysis

3.1.1. X-ray Diffraction

Structural analysis of the CoNi thin films obtained using LTVA method was characterized using X-ray diffraction methods. Two peaks are observed in 2 θ scanning at 44.28°, which can be indexed as (111) plane of face-centered cubic (fcc) phase CoNi (JCPDS no. 15–0806 for fcc Co, JCPDS no. 04–0850 for fcc Ni, respectively) [38]. As shown in Figure 2, another small peak can be revealed at 51.58°, indexed as (200).

The zone between 65° and 75° degrees was arbitrarily removed to hide the XRD peaks of the Si substrate, centered at 69°. The peak from 44.28° can vary between 44.2° for a pure fcc Co structure to 44.28° in the case of Co₃Ni and at 44.4° for CoNi₃ [39,40].

Interestingly, the second phase, marked by diffraction on the (102) plane, was observed at 62.55° as well as a possible one at 47.66°, indexed as (101). All peaks are indexed in

Table 2. X-ray diffraction peak-fitting parameters.

No.	Peak Location (°)	(hkl)	Peak Width	d-Spacing (Å)	Cell Parameter (nm)	Symmetry
1	44.28	(111)	0.1157	2.0438	3.4084	Fm-3m cubic
2	51.58	(200)	-	1.7885

.

3.1.2. HRTEM and SAED Measurements

The XRD results were verified using High-Resolution Transmission Electron Microscopy (HRTEM) with the aid of Selected Area Electron Diffraction (SAED), a more sensitive method that allows a better estimation of the crystallite dimensions.

The HRTEM image shows a uniform thin film (Figure 3) with a crystallite dimension of about L_c = 5 nm, estimated using a lognormal distribution of Feret’s diameter, the algorithm implemented in the iTEM platform (right inset Figure 3). The left inset shows the electron diffraction and Miller indices for Ni or Co cubic system [41].

The profile extracted from electron diffraction pattern was used to evaluate the lattice parameters of cubic Ni, Co, and hexagonal Co₃Ni.

Table 3. SAED data analysis.

Peak	dhkl (Å)	2θ (°)	I/I0	hkl	FWHM	LDS (nm)
1	2.0376	1.0405	10,000	111	0.11	3.3943
2	1.7988	1.1786	1734	200	0.09	3.2335
3	1.2471	1.7001	1265	220	0.04	3.4980
4	1.0656	1.9897	833	311	0.03	3.4053
						3.3828

shows the peaks identified from the profile and Scherrer analysis of the crystalline size [42,43].

The values of crystalline sizes reported from XRD analysis are very well correlated with crystallite sizes estimated from TEM images.

The ratio of lattice parameters a/c is revealed in

Table 4. Lattice parameters of the CoNi thin films.

Structure	SG Symmetry Group	a|c (Å)	a|c Cohen (Å)	Errors (%)
Cubic Ni	Fm-3m	3.45 [42]	3.4951	1.3108
Cubic Co	Fm-3m	3.42 [43]	3.4951	2.1969
Hexagonal Co0.75Ni0.25	P63/mmc	2.504|4.065 [44]	2.6162|4.3834	4.4802|7.8331

, according to the cubic structure of Ni and Co. Additionally, the lattice parameter was refined using a modified Cohen method with a Nelson–Riley function applied to the electron diffraction (cosθ ≈ 1 and sinθ ≈ θ) [44]. This ratio, estimated from the Cohen method, agrees well with the measured value a/c = 3.4951 Å, with a 1.31% error for Ni and a 2.19% error for Co.

3.2. X-ray Reflectometry

The thickness and roughness of the CoNi thin films were obtained by X-ray reflectometry (XRR) with good resolution, using 0.2 mm slits for both X-ray tube and detector. The 2 θ/omega patterns can be seen in Figure 4.

The XRR patterns were analyzed with REFLEX software, considering a multilayer structure formed from silicon/SiO₂/Co-Ni/Co-Ni oxide [45]. The curve exhibits two main critical angles at 0.32° and 0.67°. The most important one (0.67°) is associated with the density of Co-Ni thin film. This critical angle is the same for both metals (Co and Ni) because they have similar densities (Co: density = 8.9 g cm⁻³; Ni: density = 8.908 g cm⁻³) [46]. The critical angle from 0.32° was assigned to cobalt-nickel oxide layer with a mean density of 6.44 g cm⁻³. A thin oxide layer gives a broad peak between 1.5° and 3° with a thickness of 7 nm; this was obtained from the fitting procedure.

The maxima/minima between 1° to 1.75° are assigned to the CoNi thin film; based on the Bragg formula, the distances between two maxima/minima give the thickness of the layer, estimated at 95 ± 5 nm. The roughness of the CoNi thin film is around R = 1.8–2 nm; this was confirmed by the AFM measurements.

3.3. SEM and EDX Analysis

The SEM image of the surface morphology of CoNi thin films grown on silicon substrate is shown in Figure 4. The surface of the CoNi thin films on an area of about 3 × 3 μm looks smooth, proving the superior characteristics of deposition by the LTVA method (Figure 5a).

In the 3D image with a length of 1 µm, the CoNi surface has an average roughness of about R_a = 4 nm, resulting in a good agreement with the fitting procedure of XRR patterns (Figure 5b).

More interesting is the EDX analysis, which gives information about the composition of the deposited thin films. It also shows where the main peaks are assigned to Co, Ni, and larger Si substrate, and shows that they are accompanied by traces of oxygen and carbon K-lines (Figure 6).

If the line 0.776 keV of Co is superimposed on 0.851 keV line of Ni, the line 6.929 keV of Co is well separated by one at 7.477 keV of Ni on the wide spectrum. The traces of carbon are due to the carbon crucible used during LTVA deposition. The concentrations measured at three points of the CoNi thin film are given in

Table 5. EDX compositional analysis.

Element	Atomic No.	Point 1 (at.%)	Point 2 (at.%)	Point 3 (at.%)	Mean Error
Co	27	1.35	1.40	1.32	0.2
Ni	28	3.20	3.26	3.24	0.1
Si	14	77.54	78.42	77.88	3.2
O	8	7.01	6.07	6.81	0.9
C	6	10.89	10.86	10.75	1.1
Total		100	100	100

.

As can be seen, the Ni/Co concentrations in all three points have a mean value of 2.3–2.5; this ratio that suggests an amorphous Co/Ni, but with partial crystalline structures, as can be seen in XRD and TEM.

3.4. Surface and Magnetic Properties

3.4.1. Atomic and Magnetic Force Microscopy

Atomic force microscopy (AFM) and magnetic force microscopy (MFM) were investigated at different scaling square zones: 20 × 20 μm (a, c) and 1 × 1 μm (b, d) (Figure 7).

At a large scale, the roughness was confirmed to be R_a = 6.68 nm (Figure 7a); at a small scale, the roughness is reduced to 4.5 nm (Figure 7b). This is the most interesting image because it shows some spherical particles with a mean size of about 40–50 nm.

The morphology and the shape of these particles were confirmed by magnetic force microscopy (MFM) measurements. In magnetic force microscopy (MFM), the magnetic field is detected by placing a small magnetic tip, usually covered with CoCr alloys, on a cantilever spring, close to the surface of the thin film. The scan areas follow AFM images from 20 × 20 μm to 1 × 1 μm, with a recording time between 5 and 30 min (Figure 7c,d).

In this study, we used CoCr cantilevers with a radius of less than 40 nm and a scan rate of 1 Hz. In standard Magnetic Force Microscopy measurements, the phase shift Δφ of the oscillating cantilever is recorded at a fixed lift height Δz [47].

The phase shift Δφ marks the magnetic interaction between the tip and the CoNi thin film, where this magnetic interaction is related to the gradient ∂F/∂z along the z-direction perpendicular to the surface plane of the component along z of the long-range interaction force F_z through the relation:

$$\Delta \mathsf{\phi }=-\frac{{\mathrm{Q}}_{\mathrm{c}}}{{\mathrm{k}}_{\mathrm{c}}}\times \frac{\partial {\mathrm{F}}_{\mathrm{z}}}{\partial \mathrm{z}}$$

(4)

where Q_c and k_c are the quality factors of the cantilever first resonance in air and the cantilever spring constant, respectively [48].

In Figure 7d, the same 40–50 nm magnetic particles can be easily identified; this confirms the magnetic characteristics of CoNi thin films. These behaviors have been further studied using the longitudinal magneto-optical Kerr effect, as shown in the next section.

3.4.2. Longitudinal MOKE Method

The hysteresis loops from the longitudinal magneto-optic Kerr effect (MOKE) magnetometry were recorded with the magnetic field oriented parallel with the CoNi surface of the sample (in-plane analysis). Even if the hysteresis loops differ from the typical characteristics of classical magnetic measurements, this effect can be used to study the local magnetism of magnetic thin films. Here, a hysteresis loop is a plot of the magnetization as a function of the applied magnetic field on the CoNi thin film. The magnetic field was established between H = −2000 Oe and H = 2000 Oe in order to determine the saturation of the magnetization effect.

The magnetic behavior observed in the magnetic force microscopy was confirmed in the MOKE measurements presented in Figure 8a,b.

3.5. Electrical Measurements

Electrical measurements follow two aspects: (a) measurements of the sheet resistance and, based on the estimated thickness and geometry of the sample, determination of the resistivity for the CoNi thin layer, and (b) dependence of the resistivity with heating of samples in the range T = 295−329 K.

The sheet resistance was measured between the applied current and the inner voltage of the four-point probe equipment (Figure 9).

Using the four-point probe technique, the applied voltage on the outer electrodes combined with the high internal resistance of the equipment allows only the current injection in the metallic thin films, while the measured voltage on the inner electrodes avoids the contact resistance between the electrodes and the metallic thin film. In this way, an accurate measurement of the sheet resistance can be controlled in thin films with thicknesses up to 1 mm.

The slope of the linear dependence between the inner voltage and applied current gives the sheet resistance and, by applying Formula (3), the resistivity was estimated at $\rho$ = 34.16 μΩ cm.

When the sample is heated, the resistivity varies linearly with temperature between T = 295 K and T = 329 K (Figure 10).

The resistivity versus applied temperature follows a linear dependence just over room temperature, given by:

$$\rho ={\rho }_0\left(1+\alpha \times \Delta T\right)$$

(5)

where α is the temperature coefficient expressed as 1/°C and $\rho$_o is the resistivity at room temperature. From the linear slope, the temperature coefficient is α = 11.99 × 10⁻³ C⁻¹.

4. Discussions

For Co-Ni alloys, it is important to discuss their phase diagram because solidification occurs in a temperature interval of only a few degrees [49]. In this context, the solid equilibrium between Co and Ni exhibits a complete solid solution in the α (Fm-3m) phase at temperatures between the solidus and the allotropic transformation temperature of the α to ε (P63/mmc) phase. The phase appears in the whole range of Ni dissolution in Co from 0 to 100 at.% Ni, but the ε (P63/mmc) appears between 0 and 35 at.% of Ni.

Several papers describe the Co-Ni alloys as a combination of cubic Ni (Fm-3m) and cubic Co (Fm-3m), proving the presence of Ni atoms by the shift of the (111) peak position of Co from 44.201° in pure Co nanoparticles [50,51] to 44.469° in the case of pure Ni nanoparticles [38]. For the Co:Ni (3:1) composition, the peak shift is up to 44.280°. However, the intensities of (111) for the cubic structure and (102) for the hexagonal one, together with the lower intensities for the (101) and (200), confirm a preferential orientation of crystallites in the amorphous matrix.

The hexagonal structure of the Co₃Ni system (Co₀.₇₅Ni_0.25) has bigger errors, so the probability of finding such a system in the sample is low. However, the intensity of the (102) peak reflection justified the presence of the oriented hexagonal structure.

The grain size of the crystalline structure for the Co or Ni metals can be evaluated using the Debye–Scherrer formula, accounting for the broadening of the equipment based on the Si (100) from 69.3°. The mean value of crystallites is around 3–4 nm.

From the XRR, the thickness and density of metallic thin films can be estimated. The film thickness, around 95 nm, is in good agreement with the measured one with the quartz balance during thin film processing. The film density was estimated to 6.44 g cm⁻³, compared with 8.9 g cm⁻³ for Co and 8.91 g cm⁻³ for Ni. The estimation of the relative porosity is the ratio (d_metal − d_film)/d_metal, where d_metal is the density of the pure Co or Ni crystalline film and d_film is the density of the Co₃Ni film [52]. This fact leads to a relative porosity of 27.5%.

The cell parameter, determined from the X-ray diffraction patterns for the CoNi composition of about 3.4084 nm, was confirmed by the HRTEM and SAED images, where the profile extracted from the electron diffraction pattern was used to evaluate the lattice parameters of cubic Ni, with Co giving a closed value of about 3.38 nm. However, the HRTEM images confirm the presence of hexagonal Co₃Ni with a more accurate precision when compared with the XRD measurements.

The surface images obtained using the SEM technique reveal a smooth surface in the 2D dimension, while the roughness from 3D images varies between 2 and 4 nm. This fact explains the advantages of the LTVA technique over the classical TVA method, where the roughness covers tens of nanometers.

EDX analysis at three points of the CoNi film confirms the ratio of Co:Ni around 3:1 and justified the presence of Co₃Ni hexagonal structure observed in SAED and XRD measurements.

The AFM measurements confirm the low roughness of the CoNi thin films on the extended area of about 20 × 20 µm (around 7 nm); in the small area of 1 × 1 µm, the roughness is 4.5 nm in a good agreement with the X-ray reflectivity measurements.

Magnetic force microscopy (MFM) reveals large domains of magnetization of about 40–50 nm; these are probably a mix between the crystalline structures with the amorphous Co and Ni.

The measurement of magnetization (M) is connected with the rotation of the polarization plane of light as it is reflected at the sample surface. This can be phenomenologically described by:

$$\overrightarrow{\mathrm{D}}=\overrightarrow{\mathsf{ϵ}}\times \left(\overrightarrow{\mathrm{E}}+\mathrm{i}\times \mathrm{Q}\times \overrightarrow{\mathrm{M}}\times \overrightarrow{\mathrm{E}}\right)$$

(6)

where D is the induced displacement vector, ε is the dielectric constant, E is the electrical vector of the incoming plane light wave, i is the imaginary part, and Q is a material parameter. The cross-product M × E links the magnetization with the change in the displacement vector and thus the electrical vector of the reflected light wave [53]. Commonly, the magnetic signal is measured as a variation of the angle between the incident and reflected light as a function of the applied magnetic field, θ = f(M(H)).

From the longitudinal MOKE measurements, the CoNi thin films with a mean thickness of about 100 nm have a coercive field of about H_c = 38–42 Oe, depending on the orientation in-plane of the sample (Figure 8b) and a saturation field of around M_s = 900 Oe. The angular dependence of the magnetization with the rotational angle indicates an isotropic magnetic thin film without specific magnetization domains, supported by the amorphous structure of the CoNi thin films.

Considering only the coercive field, the obtained CoNi thin films should be seen as soft materials, but the saturation field is higher than similar cobalt thin films obtained by thermal evaporation, which has a saturation field of around Ms = 130–150 Oe [36].

If we consider the squareness of the in-plane field hysteresis loop defined as the ratio between the remanent magnetization and saturation (M_r/M_s), the obtained value is around 0.15. This value is smaller than the 0.4 obtained in the Co_1−xNi_x with x = 25%, 1 μm-thick film of CoNi alloys deposited using the sputtering technique [54]. The coercive field is slightly lower than that in the case of Co_1−xNi_x for all Ni at.% concentrations for which the coercive field is over H_c = 100 Oe. In this context, our CoNi thin films can be seen as semi-hard magnetic materials with a coercive field ranging from 50 to 1000 Oe [5].

In most cases, the thin film resistivity ($\rho$_f) is highly dependent on the film thickness. The increase in resistivity, defined as Δ = $\rho$_f-$\rho$_bulk, is proportional to 1/d, where $\rho$_bulk is the bulk resistivity and d is the film thickness [55]. Many theories have been proposed to analyze experimental thin-film resistivity as a function of the thickness, but the most common is the one based on the grain boundary scattering model, developed by Mayadas et al. [56], which was successfully applied to cobalt and nickel thin films. This model can be applied considering thin polycrystalline layers; the amount of surface scattering is relatively small. In this context, thin-film resistivity depends on the bulk resistivity and a constant γ, which considers the size effects:

$${\rho }_{\mathrm{film}}=\frac{1}{\mathrm{d}}\left({\rho }_{\mathrm{bulk}}\times \mathrm{d}+\mathsf{\gamma }\right)$$

(7)

where γ depends on the ratio between the grain size diameter D and film thickness d.

The value of resistivity ($\rho$ = 34.16 μΩ cm for the CoNi thin films obtained by the LTVA method) agrees well with similar cobalt thin films thermally evaporated on the silicon substrate. For these films, the resistivity varies from $\rho$ = 98 μΩ cm (thickness d = 10 nm) up to $\rho$ = 43.14 μΩ cm for d = 100 nm thin film, having 1/d dependence [57]. These values are higher than the bulk resistivity for bulk Co ($\rho$ = 5.88 μΩ cm) or Ni ($\rho$ = 7.14 μΩ cm) [47].

The effective resistivity of a metal follows Matthiessen’s rule and the change in the resistivity, due to a small increase in ΔT, is temperature independent, enabling the linear approximation [58]. For the range above 100 K up to 1000 K, all metals exhibit linear resistivity versus temperature [59]. The temperature coefficient is larger when compared with the individual bulk metals, which are α = 6.6 × 10⁻³ (K⁻¹) for Co and α = 6.5 × 10⁻³(K⁻¹) for Ni. This fact can be explained by the same dependence of resistivity versus film thickness for metallic thin films.

5. Conclusions

The LTVA method is a new and reliable method for production of thin films with tailorable properties, in this case, intermetallic Co-Ni compound with a crystalline structure, due to the enhanced rate of collisional ionization of the resulting excited state atoms relative to the ground state. The obtained films have semi hard magnetic structures with low coercive field and medium magnetization characteristics.

This novel method allows simultaneous deposition of a bi-component metallic thin film using photonic processes of the laser over the plasma deposition, which improves the roughness but also the composition of the deposited thin film. The resulting CoNi thin films exhibit a mixture of crystalline phases with a large amorphous compound that influences the electrical and magnetic properties of these films. The roughness of these films is lower than those deposited with the TVA method or other thermal evaporation methods.

The electrical resistivity is higher than the bulk CoNi alloys, justified by the grain boundary scattering model with fcc cubic structures for pure Co and Ni nanocrystals with a small amount of hexagonal Co₃Ni nanocrystals. Just above room temperature, the electrical resistivity shows a higher thermal coefficient influenced by the relatively small thin film thickness.

The magnetic images based on the phase shift Δφ from the MFM measurements showed spherical magnetic particles with a mean size of about 40–50 nm. The Moke measurements suggest an isotropic distribution of the magnetic nanoparticles for the in-plan angular rotation.