Investigating the Effect of Sputtering Particle Energy on the Crystal Orientation and Microstructure of NbN Thin Films

Abstract

Niobium nitride (NbN) thin films are crucial materials for various applications, including superconductivity and hard coatings. However, precisely controlling their microstructure and crystal orientation during synthesis remains a challenge. This study addresses this gap by systematically investigating the effect of sputtering particle energy on NbN film properties. This research aims to elucidate the relationship between sputtering particle energy and the resulting microstructure and crystal structure of NbN thin films synthesized by reactive magnetron sputtering. NbN thin films were deposited on Si (100) substrates using reactive magnetron sputtering. The effects of sputtering power, total sputtering pressure, and nitrogen partial pressure on the films’ preferred orientation, grain size, and crystallinity were characterized using X-ray diffraction (XRD), scanning electron microscopy (SEM), and atomic force microscopy (AFM). The preferred orientation of NbN films can be controlled by sputtering parameters. Increasing sputtering power leads to a texture transition from a single -phase (111) orientation to mixed (111) and (200) orientations. Sputtering pressure influences the energy of sputtering particles and can shift the preferred orientation from mixed (111) and (200) to single-phase (200). Higher nitrogen partial pressure affects the crystallinity of NbN films, potentially introducing defects that reduce crystallinity. This study demonstrates that sputtering particle energy significantly influences the microstructure and crystal orientation of NbN thin films. Precise control of sputtering parameters enables tailored film properties, which are crucial for optimizing NbN performance in diverse technological applications.

1. Introduction

NbNx films have attracted significant attention due to their superior superconducting and mechanical properties. In the field of low-temperature superconducting electronics, such as tunnel junction [1] and single-photon detectors [2], NbNx films have found wide applications. Moreover, their excellent mechanical properties, along with stable chemical characteristics, high melting point, and high conductivity, make NbNx films relevant in areas such as protective coatings [3,4,5], field emission cathodes [6], and diffusion barriers in microelectronic devices [7]. NbNx films exhibit various crystal structures [8,9], including tetragonal structure ϒ-Nb₄N₃, face-centered cubic structure δ-NbN, hexagonal structure δ’-NbN, hexagonal structure ε-NbN, hexagonal η-NbN, and cubic structures made by a solid solution of α-NbN with the cubic structure of N₂ in metal Nb, all of which strongly influence the performance of NbNx films.

NbNx films can be prepared by different deposition methods, including reactive magnetron sputtering (reactive MS), arc physical vapor deposition, and ion-assisted and pulsed deposition [10,11]. In these deposition methods, the phase structure, preferred orientation, and physical properties of NbNx films strongly depend on the substrate bias, nitrogen gas pressure, and sputtering pressure. Wong et al. [12] reported a phase transition from δ-NbN to δ’-NbN in NbN films with increasing nitrogen gas pressure. Fenker et al. [13] observed a phase transition from β-Nb₂N to δ’-NbN via δ-NbN in NbN films with increasing nitrogen pressure under a growth temperature of 300 °C. Deen [14] investigated the influence of growth rate on the superconducting properties by controlling the target discharge current, gas flow rate, and total sputtering pressure. It was found that the superconducting transition temperature was 14.7 K when the deposition rate was 9 Ås⁻¹, the target current was 1.8 A, the nitrogen/argon ratio was 15% and 85%, and the total sputtering pressure was 0.88 Pa. This study mainly focuses on the effects of sputtering power, nitrogen pressure, and total pressure on the preferred orientation characteristics of NbNx films from the perspective of growth conditions.

Reactive MS was chosen as the film preparation method [15], with a focus on plasma energy, to analyze the effect of sputtering particle energy on the preferred orientation of NbNx films. Sputtering deposition has two important features: atoms reaching the growth surface have high kinetic energy, and the growth chamber contains an argon environment. The actual growth process of high-energy particles involves shallow implantation, accompanied by the generation of excess interstitial atoms on the surface. In this case, implantation occurs only when the energy of the atoms reaches a threshold energy, which depends on the structure, composition, and orientation of the material. Therefore, this study examined the relationship between sputtering particle energy and the microstructural crystal structure of NbNx films, combined with X-ray diffraction and scanning electron microscopy methods, as well as the relationship between sputtering current and voltage curves. By changing the sputtering power, sputtering gas composition, and total sputtering pressure to control the energy of the sputtering particles, the relationship between sputtering particle energy and the preferred orientation, grain size, and distribution of the film was analyzed. In this study, we uniquely combine X-ray diffraction and scanning electron microscopy methods, as well as analysis of sputtering current and voltage curves, to provide a comprehensive understanding of the relationship between sputtering particle energy and the microstructural crystal structure of NbNx films. Utilizing a highly regulated reactive MS system with real-time target voltage monitoring, this work precisely examines the direct relationship between the energy of the sputtering particles and the final NbN thin film microstructure. This level of control, compared to traditional sputtering techniques, enables a more sophisticated understanding of the energy-dependent growth mechanisms.

2. Experimental Method

NbN thin films were prepared using reactive magnetron sputtering on silicon substrates, with a 300 nm thick SiO₂ layer on the Si substrate surface. The Nb target purity was 99.95%, with a target size of Φ80 mm. After cleaning the substrates using chemical methods, they were subjected to Ar ion RF cleaning for 5 min in the sample introduction chamber before entering the sputtering chamber. The chamber was evacuated until the vacuum reached below 2 × 10⁻⁵ Pa before starting the sputtering coating. The purity of the sputtering gases Ar₂ and N₂ was 99.999%. During the growth process, sputtering power, total sputtering pressure, and nitrogen partial pressure were controlled to adjust the thin film growth parameters. Sputtering power was controlled between 100 W–400 W, total sputtering pressure was controlled between 0.2 Pa–1 Pa, and nitrogen partial pressure was controlled using a flowmeter module. During the adjustment of sputtering parameters, target voltage and current parameters were recorded. After sample preparation, characterization was carried out using X-ray diffraction, scanning electron microscopy (SEM), and atomic force microscopy (AFM) methods. A highly regulated reactive MS system with real-time target voltage monitoring is used in this work to precisely examine the direct relationship between the energy of the sputtering particles and the final NbN thin film microstructure. In this research study, compared to traditional sputtering techniques, this degree of control enables a more sophisticated understanding of the energy-dependent growth mechanisms. The X-ray diffraction (XRD) measurements were performed using a Bruker D8 Advance X-ray diffractometer, manufactured by Bruker cooperation, Billerica, MA, USA with Cu Kα radiation (λ = 1.5418 Å) in the 2θ range of 20–80 degrees. The scanning electron microscopy (SEM) images were obtained using a Hitachi BC-PCAS4800 SEM, Manufactured by Hitachi High-Tech Corporation, Tokyo, Japan. Atomic force microscopy (AFM) measurements were carried out using a (SPA-300HVAFM System, Manufactured by Seiko Instruments Inc., Chiba, Japan) in tapping mode.

3. Transition Curve Analysis

We fabricated a 120 nm NbN thin film on a silicon substrate using the reactive MS technique, and a four-point probe measurement system was employed for precise resistivity assessment. The configuration consists of four collinear probes delicately applied to the film’s surface. A steady, constant current flows through the two outer probes, while the resultant voltage drop between the inner two probes is recorded using a high-sensitivity voltmeter. The resistivity (ρ) can be estimated using the film’s thickness (t = 120 nm), the measured current(I), and voltage(V) with the formula ρ = (V/I) × (π/ln(2)) × t [16]. This technique reduces contact resistance, essential for precise measurements in thin films, and offers a dependable evaluation of the electrical characteristics of the NbN film. Employing a cryostat (PPMS) alongside four-probe resistivity measurements, we assess the resistivity of NbN thin films at different temperatures (K), as detailed in the subsequent table.

Using a cryostat with four-probe resistivity measurements, we assess the resistivity of NbN thin films at different temperatures (K), as presented in

Table 1. Resistivity vs. temperature for NbN thin films.

Temperature (K)	Resistivity (μΩ·cm)	Normalized Resistivity (ρ/ρₙ)
20	250	1
16	250	1
15.5	240	0.96
15	200	0.8
14.5	100	0.4
14.3	50	0.2
14.1	10	0.04
14	0	0
13.5	0	0

. This results in the transition curve for the NbN thin film, displaying normalized resistivity (ρ/ρn) versus temperature. The superconducting transition of the NbN thin film was analyzed by measuring its resistivity in relation to temperature, as illustrated in Figure 1 below.

This figure illustrates the superconducting transition curve, depicting normalized resistivity as a function of temperature, and provides essential insights into the film’s superconducting characteristics, which are intrinsically linked to its microstructure and crystal orientation, both of which are affected by the sputtering particle energy during deposition [17]. The graph distinctly demonstrates the shift from the normal to the superconducting state. At temperatures beyond 16 K, the normalized resistivity remains at 1.0, signifying that the film is in its typical, non-superconducting form. A rapid decline in resistivity upon cooling indicates the commencement of superconductivity, beginning at approximately 15.5 K. The critical temperature (Tc), as derived from the curve, is roughly 14.3 K, indicating the temperature at which the film undergoes a transition to the superconducting state. This value corresponds with the supplied data, indicating that the resistivity decreases to fifty percent of its standard value. The transition width, defined as the temperature range during which resistivity decreases, indicates the existence of inhomogeneities or structural defects inside the film. These flaws may result from fluctuations in grain size, defect density, and crystal orientation, all of which are affected by the energy of the sputtering particles employed during film deposition. For instance, elevated sputtering energies may enhance adatom mobility, potentially impacting grain development and defect formation, thus changing the sharpness of the superconducting transition. The evidence corroborates this, indicating a steady decline in resistivity from 15.5 K to 14 K, rather than a sudden decrease. Below 14 K, the normalized resistivity approaches zero, signifying that the film has transitioned into the superconducting state, as corroborated by the measurements. Our observed superconducting transition temperature (Tc) of approximately 14.3 K is within the range reported in other studies for NbN films prepared by magnetron sputtering. For example, Deen [14] reported a Tc of 14.7 K for NbN films grown by DC magnetron sputtering, which is very close to our value. However, it is important to note that the superconducting properties of NbN films can be influenced by various factors, including specific sputtering parameters, substrate material, and film thickness. Pei et al. [17] recently demonstrated the ability to control the superconducting critical temperature and resistance of NbN films through thin-film deposition and annealing processes. This highlights the tunability of NbN superconducting properties, which is consistent with our observation that sputtering particle energy plays a crucial role in determining the superconducting transition. The abruptness of the transition and the attained Tc are direct outcomes of the film’s microstructure, which is, in turn, a result of the sputtering circumstances. By examining the superconducting transition curve and correlating its characteristics (Tc, transition width) with sputtering parameters, especially particle energy, we can obtain significant insights into how these parameters influence the crystal orientation and microstructure of the NbN films, thereby optimizing their superconducting properties for particular applications. The observed transition, despite exhibiting a distinct Tc, suggests potential for enhancement regarding transition sharpness, attainable by the regulation of sputtering particle energy to customize the film’s microstructure.

The XRD patterns indicate that the NbN films exhibit a δ-phase structure. At lower sputtering powers (100 W and 200 W), the films show a (111) preferred orientation, which suggests a relatively ordered microstructure. This order contributes to the lower resistivity values observed in Table 1 at temperatures close to the transition. As the sputtering power increases (300 W and 400 W), the XRD patterns show a transition to a mixed orientation of (111) and (200), indicating a less ordered microstructure and a decrease in film crystallinity. This microstructural change is associated with the broader transition observed in the resistivity curve. The higher resistivity values at temperatures above Tc can be attributed to increased grain boundary scattering and defect densities, as evidenced by the XRD results.

4. Experimental Results

The thickness of the films prepared by this method was determined to be around 120 nm, using cross-sectional scanning electron microscopy. The surface roughness of the film was determined using atomic force microscopy. The microstructure of NbN thin films prepared on Si substrates was measured within a 5 μm × 5 μm range, and the sample exhibited a very smooth surface, with a roughness measurement below 2.8 nm, and displayed good flatness and density.

Increasing sputtering power density causes argon ions to bombard the target material with higher energy, and the particles sputtered will have higher energy when they reach the substrate, affecting thin film nucleation orientation, crystallinity, and internal defect density. Therefore, controlling sputtering power from the perspective of particle energy plays a direct role in the microstructure of the film. To this end, the influence of sputtering power on the microstructure of NbN thin films was studied, with the growth power controlled between 100 W and 400 W. Figure 2a shows the XRD (θ–2θ scan) spectrum of NbN thin films deposited under different power conditions. XRD measurements provide information about the phase structure and preferred orientation of the samples. The XRD results show that under low-power conditions, NbN thin films exhibit a δ-phase structure and a (111) preferred orientation. Additionally, under the sputtering conditions of 200 W (P_total = 0.5 Pa, N:Ar = 5 sccm:40 sccm), the (111) diffraction peak of δ-NbN becomes stronger. When the sputtering power is increased to 300 W, NbN thin films begin to exhibit a mixed orientation of (111) and (200), and when the power is increased to 400 W, both the (111) and (200) peaks of NbN become weaker, indicating a decrease in film crystallinity. Besides affecting film texture and crystallinity, sputtering power also directly affects the grain size of NbN thin films. Figure 2d shows that under the 100 W power condition, the grains of NbN thin films are small due to relatively low sputtering particle energy and poor mobility on the substrate. When power is increased to 200 W, the grains significantly increase in size, and the grain size is uniform, indicating that under the sputtering power condition of 200 W, NbN thin films with a preferred (111) orientation have uniform grain structure. When the sputtering power is 300 W, the grains inside NbN thin films show a phenomenon of growth, with an increase in grain size, but the grain size is not uniform. Therefore, increasing the sputtering power and increasing the energy of the sputtering particles affect the nucleation and growth of grains during film crystallization. Combining SEM and XRD results, it can be concluded that when particle energy is low, the grains continuously grow, and a (111) preferred orientation appears; as particle energy increases, discontinuous growth of grains occurs, with island-like growth, resulting in larger grain sizes in local areas, accompanied by a mixed texture of (111) and (200) orientations inside the film. As observed from the XRD patterns in Figure 2a, NbN thin films exhibit a face-centered cubic (FCC) crystal structure.

To further illustrate how changing the energy of sputtering particles can affect the microstructure of NbN thin films, an attempt was made to adjust the size of the sputtering pressure to change the energy of the sputtering particles. It was found experimentally that as the sputtering pressure gradually decreased, the target voltage increased from 425 V to 464 V (Figure 3b), indicating that the energy of the sputtering particles also increased. As a result, XRD results show that under the condition of 300 W power, initially under high sputtering pressure conditions, NbN thin films exhibit a mixed texture of (111) and (200). However, when the sputtering pressure decreases to 0.3 Pa, the energy of the sputtering particles increases, and NbN thin films exhibit a (200) preferred orientation growth.

Nitrogen partial pressure plays a crucial role in the nitride phase of Nb and directly determines the phase structure of Nb nitrides. Additionally, Drusedau et al. [18] found that in nitride sputtering mode, the energy of particles sputtered from the target surface increases with increasing nitrogen flow rate. Consequently, the microstructure of NbN thin films will change. In the experiment, the nitrogen-to-argon flow ratio was controlled between 1:8 and 1:3.1. When the ratio is below 1:3.5, the intensity of NbN diffraction peaks increases, and no change in film texture is observed. When the ratio exceeds 1:3.5, it is found that the NbN diffraction peaks begin to weaken, as shown in Figure 4b, with the (111) diffraction peak starting to decrease. Gaussian fitting of the diffraction peaks in the 2θ range of 33°–45° in Figure 4a yields Figure 5. From Figure 5, it can be seen that besides the increase in NbN (111) and (200) diffraction peaks, the full width at half maximum (FWHM) of the two diffraction peaks also decreases, indicating a significant improvement in the crystalline quality of NbN films. From the SEM images (Figure 6), it is observed that NbN films grown under both high and low nitrogen partial pressure conditions have almost no difference in grain size, and the grains are uniform and consistent. Combining the XRD and SEM experimental results, it is concluded that nitrogen partial pressure affects the crystallinity of NbN films during the growth process but does not cause a change in the texture of NbN films.

This study clarifies the complex link between sputtering parameters and the microstructure of δ-NbN thin films. The observed energy-induced textural change from (111) to (200), ascribed to increased surface diffusion and adatom dynamics, highlights the essential influence of particle energy on film development orientation. Moreover, the meticulous regulation of phase purity via the modulation of sputtering pressure, specifically attaining an ideal nitrogen-to-argon ratio of 1:3.5, offers a significant criterion for the fabrication of superior films. The saturation of particle energy effects at elevated nitrogen pressure, indicated by a stabilized target voltage and supported by XRD data, delineates practical constraints for process optimization. These findings enhance the fundamental comprehension of NbN growth dynamics and provide practical insights for the development of next-generation thin-film coatings with customized functional performance.

5. Analysis and Discussion

The evolution mechanism of film texture is mainly explained by thermodynamic and kinetic viewpoints. In the actual film growth process, both thermodynamic and kinetic factors should jointly determine the development of the film’s microstructure. From a kinetic perspective, one of the mainstream mechanisms is the competitive growth mechanism of grains with different orientations [19], where when particles with lower energies reach the growing film, if grain diffusion occurs, grains with relatively low diffusion rates parallel to the substrate survive and become preferred surfaces. These surfaces typically correspond to high surface energy surfaces. Because the low surface energy surface is a stable state, atoms adsorbed on this surface disrupt this stability, and to maintain stability, atoms quickly diffuse from this crystal plane to other crystal planes. As a result, atoms move from low surface energy surfaces to high surface energy surfaces, and high surface energy surfaces accept more atoms, resulting in a high growth rate. Consequently, the high surface energy surface will eliminate other crystal planes during the growth process, leading to preferred growth phenomenon. For the face-centered cubic crystal of NbN, the surface energy $\gamma$₁₁₁ of (111) surface is the highest [20], defined as follows:

$${\gamma }_s\approx {\displaystyle \frac{\left[E_{slab\left(N\right)}-E_{bulk}\right]}{2A_s}}$$

(1)

where ${\gamma }_s$ is the surface energy, $E_{slab\left(N\right)}$ is the total energy calculated for N layers of atoms with a sufficiently large volume, $E_{bulk}$ is the energy of the corresponding bulk material, and $A_s$ is the corresponding surface area. According to LDA-CAPZ density functional theory, for the δ-NbN phase, the calculated surface energies $\gamma${200} and $\gamma${111} of the (200) and (111) crystal planes are 1.68 J/m ² and 2.92 J/m² [21], respectively, which clearly indicates that $\gamma${200} < $\gamma${111}. At lower sputtering powers (100 W and 200 W in this study), the energy of the sputtered atoms arriving at the substrate is relatively low, limiting adatom mobility. Under these conditions, film growth is primarily controlled by thermodynamic factors. Thermodynamic control favors the growth of crystal orientations that minimize surface energy. For NbN, the (111) plane has a lower surface energy (2.92 J/m²) compared to the (200) plane (1.68 J/m²). Surface energy (${\gamma }_s$) can be calculated by using Equation (1) above. The lower surface energy of the (111) plane promotes a (111) preferred orientation at lower sputtering powers, as observed in Figure 2a. This is consistent with the concept that at low energies, grains with the slowest growth rate will dominate. The micro- or nano-hardness data were not included in this study. This was primarily due to the focus on the relationship between sputtering parameters and the structural and superconducting properties of the NbN thin films. However, it is important to note that hardness is a crucial mechanical property for many applications of NbN coatings. Future investigations could incorporate micro- or nano-hardness measurements to provide a more comprehensive understanding of the material’s performance.

When the formation of the film is mainly dominated by the growth of grains, it is easy to grow films with preferred orientations. The preferred growth can be explained by the difference in growth rates of crystalline geometric directions (100) and (111) [22]:

$$\mathsf{\alpha }=\sqrt{3}{\displaystyle \frac{v\left(100\right)}{v\left(111\right)}}$$

(2)

where α represents the ratio of growth rates between the cubic crystal (100) and (111) directions. For a cubic crystal, the diagonal is $\sqrt{3}$ times the side length, and due to this geometric factor, the growth rate of the (111) direction should be $\sqrt{3}$ times the growth rate of the (100) direction. Therefore, combining the competitive growth mechanism of grains with different orientations and the mechanism of different growth rates of grains with different orientations, under conditions of low-energy bombardment growth, the NbN film exhibits (111) preferred orientation. This conclusion is consistent with the results in Figure 2a. As sputtering power increases, so does the energy of the bombarding particles. These higher-energy particles create defects and increase nucleation density on the growing (111) grains. Simultaneously, a portion of the grains begins to transform into (200)-oriented grains. This transformation process restricts the further growth of the original (111) grains. As a result, a mixed texture of (111) and (200) appears. Therefore, at 300 W power conditions, the NbN film begins to exhibit a mixed texture phenomenon. At the same time, when the density of defects formed on the surface of grains growing in the film is low, it mainly promotes nucleation, while when the density of defects on the surface of grains is high, the high grain boundary energy will also cause grain boundaries to move and grains to grow. Thus, higher-energy bombardment of the film will increase grain size. This conclusion is confirmed by SEM results.

In addition to power directly reflecting different particle energies, controlling the sputtering pressure also plays a role in changing the energy of sputtered particles. Based on the relationship between the mean free path of sputtered particles and gas concentration [23], we have the following equation:

$$\overline{\lambda }=[\sqrt{1+{\displaystyle \frac{\mathrm{m}}{40}}}\pi {\left(R+R_{Ar}\right)}^2{n}_{Ar}$$

(3)

where $\overline{\lambda }$ is the average free path of sputtered particles, $n_{Ar}$ is the Ar molecular density, m is the atomic weight of the sputtered particles, and $R$ and $R_{Ar}$ are the atomic radii of the sputtered particles and Ar, respectively. According to Equation (3), when the molecular density increases, the average free path of sputtered particles increases, so the probability of collisions between sputtered particles and gas ions increases, leading to a decrease in the energy of sputtered particles. Therefore, when controlling the sputtering pressure, as shown in Figure 3b, as the pressure increases, the target voltage decreases from the original 464 V to 424 V, indicating that the energy of the sputtered particles changes during control of the pressure. In experiments relating to controlling the sputtering power, we observed that when the power increased, i.e., when the energy of the sputtered particles increased, the NbN film began to exhibit (200) orientation. Similarly, in the process of controlling the pressure, it was found that when the NbN film was in a mixed texture of (111) and (200), with an increase in sputtering pressure, i.e., an increase in sputtering particle energy, the integral intensity of the (200) peak also increased, as shown in Figure 6b. When the sputtering pressure reached 0.3 Pa, the (111) peak of the NbN film almost disappeared, leaving only the (200) diffraction peak. Figure 7 is a semi-quantitative estimation of the content of various phases of NbN films grown under different pressure conditions. We used Gaussian functions to fit the peaks in the range of 2θ = 33–45° in Figure 3, and the integral intensity ratio of the (111) peak and (200) peak of the NbN film, I₁₁₁ and I₂₀₀, is shown in Figure 7b. As the pressure decreases from 0.9 Pa to 0.3 Pa, I₁₁₁/I₂₀₀ decreases from 8.3 to 0, indicating a transition from (111) and (200) texture to (200).

In [18], Drusedau et al. was reported that increasing nitrogen pressure leads to an increase in sputtering ion energy, causing atoms to migrate to grain boundaries. When grains grow, adjacent grains fuse together, resulting in a change in grain structure. In our experiments, we also observed that when the total sputtering pressure is fixed, increasing the nitrogen pressure from 0.015 Pa to 0.2 Pa leads to an increase in target voltage from 456 V to 483 V, as shown by the blue curve in Figure 8. The change in nitrogen pressure does indeed lead to an increase in sputtering ion energy. Furthermore, from Figure 8, it can be seen that when the nitrogen pressure exceeds 0.12 Pa (nitrogen-to-argon ratio of 1:3.1), the change in target voltage becomes very slow, stabilizing at around 480 V. This suggests that under high nitrogen pressure conditions, the change in sputtering ion energy is minimal. Therefore, considering only the change in sputtering ion energy is not sufficient to explain the effect on the microstructure of the film. In particular, as indicated in Figure 4b, under high nitrogen pressure conditions, the intensity of the diffraction peak begins to decrease. One possible reason is that high-energy N₂⁺and backscattered N atoms are shallowly implanted into the film during film growth, causing defects in the film. As the nitrogen gas flow rate increases, the ratio of N₂⁺ ions to Ar⁺ ions bombarding the target also increases. Moreover, when N²⁺ ions bombard the target, two backscattered N atoms are produced, leading to a significant increase in the number of backscattered N atoms. According to the two-body collision model, when ions in a plasma collide with static target atoms at normal incidence, the ratio of the energy transferred to the backscattered particles (E_b) to the incident ion energy (E_i) is given by [24]:

$${\displaystyle \frac{E_b}{E_i}}={\displaystyle \frac{{({\mathrm{m}}_i-m_t)}^2}{{({\mathrm{m}}_i+m_t)}^2}}$$

(4)

where ${\mathrm{m}}_i$ and $m_t$ are the masses of the colliding particles. Since the mass of argon ions is greater than that of nitrogen ions, the energy of the backscattered nitrogen atoms is greater than that of the backscattered argon atoms. Therefore, the backscattered N atoms and N²⁺ will bombard the NbN film and implant into the interstitial positions, causing defects in the film.

6. Conclusions

Using the reactive magnetron sputtering method to prepare NbN thin films, XRD results indicate that the prepared films belong to the face-centered cubic structure of δ-NbN thin films. The findings of the study have the potential to revolutionize industrial applications requiring customized NbN thin films. This research offers precise control over mechanical and electrical characteristics, which are critical for diffusion barriers in microelectronics, superconducting devices, and wear-resistant coatings. The energy of sputtering particles has a direct connection to the microstructure of the film. The main conclusions are as follows:

• Sputtering power directly affects the texture of δ-NbN thin films. At a power of 200 W, with low sputtering particle energy, grain diffusion on the (111) plane is low, leading to preferential growth and the appearance of (111) preferred orientation. When the power increases to 300 W, sputtering particle energy increases, and grains begin to grow in the (200) direction, suppressing the growth of grains in the (111) direction, resulting in a mixed texture of (111) and (200) in the NbN film.
• Sputtering power also affects the crystallinity and grain size of the film. In particular, at low sputtering particle energy (100 W) and high energy (400 W), the crystallinity of the film is low.
• Fixed sputtering power and nitrogen-to-argon ratio (Sccm N₂: Sccm Ar₂ = 5:40) reveal that when the total pressure decreases, δ-NbN films transition from initially mixed phases (111) and (200) to a single phase (200), mainly because increased sputtering particle energy promotes the growth of (200) direction grains.
• With fixed sputtering power and total pressure, adjusting the nitrogen partial pressure ratio shows that the target voltage initially increases rapidly, then increases slowly, and remains basically unchanged later, indicating that the continued increase in nitrogen partial pressure has little effect on particle energy. Therefore, XRD results show that when the nitrogen partial pressure increases from low to high, the diffraction peak intensity of NbN films gradually increases first and then remains basically unchanged. However, when the nitrogen partial pressure continues to increase, the impact of backscattered nitrogen atoms needs to be considered, leading to an increase in defects in the film, affecting the crystallinity of NbN. Thus, at a nitrogen-to-argon ratio of 1:3.5, the XRD diffraction peak of NbN thin films begins to weaken.
• Analysis of the superconducting transition curves reveals a clear relationship between the film’s microstructure and its superconducting properties, highlighting the impact of sputtering parameters on the transition.