A Study on Tantalum Alloying Layer and Its Performance on the Surface of 316LSS in Harsh Environments

Abstract

Tantalum diffusion layers were fabricated on 316L stainless steel substrates using the double glow plasma surface alloying technology (DGPSAT). The optimization rules of the Fe-Ta diffusion layer under varying alloying times were investigated, focusing on the effects of processing parameters on the phase structure and microstructure. The results indicate that, as the alloying time increases, the surface wrinkled structure in the Fe-Ta alloy layer gradually transforms into a nanoscale acicular α-Ta structure, improving surface roughness and water contact angle. The surface microstructure influenced by the alloying time enhanced mechanical properties significantly, increasing Vickers hardness from 152 HV_0.2 to 970 HV_0.2, improving bonding strength, and reducing the friction coefficient to 0.5. Electrochemical testing showed that the corrosion rate of the tantalum diffusion layer was significantly reduced from 1.04 × 10⁻² mm/a to 2.83 × 10⁻⁴ mm/a, demonstrating the excellent corrosion resistance. The island growth pattern during the formation of alloy layers was simulated by molecular dynamics. Replacing bulk materials with tantalum diffusion layers can economize rare metals, reduce costs, and be of great significant for the special equipment applications in harsh environments.

1. Introduction

In special equipment, steel materials encounter issues that affect their performance stability, such as corrosion, wear, and high temperatures. These problems result in significant economic losses and even pose threats to life safety. Traditional iron and nickel-based alloys have proven inadequate in withstanding the intricate and harsh environments prevalent in aerospace, weaponry, renewable energy, and nuclear industries. The 316L stainless steel was selected as the substrate material due to its excellent corrosion resistance, good mechanical properties, and wide applications in harsh environments such as chemical processing, marine, and biomedical industries. However, its performance can be further enhanced through surface modification to meet more demanding service conditions. Simultaneously, rare metals have increasingly garnered attention as structural or functional materials. Among these, tantalum (Ta) stands out because of its distinctive physicochemical properties [1,2]. For instance, it can form a dense oxide layer on its surface at ambient or moderately elevated temperatures, offering exceptional protection to the underlying matrix. This renders tantalum highly effective in most complex and severe environments [3,4]. Furthermore, its high ductility, toughness, thermal stability at elevated temperatures, and low expansion coefficient make it an optimal choice for protective materials in the forthcoming generation of high-end equipment [5,6].

Tantalum and its alloys exhibit exceptional corrosion resistance, high density, dynamic elongation, superior biocompatibility, and resistance to bodily fluid corrosion [7]. Additionally, tantalum-based heat-resistant and high-temperature alloys exhibit high compressive strength and thermal stability at elevated temperatures [8,9,10].

However, the scarcity and extraction difficulty of tantalum metal result in high costs, limiting its large-scale application in high-tech industries such as military and aerospace. To address this, tantalum layers can be prepared on metal surfaces to replace pure tantalum, significantly reducing its use. Employing surface engineering techniques to treat or modify the substrate surface is a better choice [11], overcoming equipment shape and size limitations, reducing costs, and ensuring performance, offering tremendous development potential [12].

Research has shown that chemical vapor deposition, electroplating, magnetron sputtering, spraying, and ion implantation can enhance tantalum coatings’ corrosion resistance, wear resistance, and biocompatibility. However, these technologies have limitations for harsh environments in high-end equipment. Chemical vapor deposition requires long reaction times and can lead to oxygen infiltration [13]. Electroplated tantalum coatings have weak adhesion [14]. Magnetron sputtered tantalum layers typically have a thickness within 20 μm, with temperature and power significantly influencing grain size and phase. While power supply and duty cycle adjustments are feasible, these methods are varied and intricate, frequently resulting in two-phase mixtures within the coating that impair performance [15]. Spraying techniques often produce uneven, rough tantalum layers with high porosity [16]. Ion implantation, though faster, can induce surface defects if overdone, compromising corrosion resistance [17].

The double-glow plasma surface alloying technology (DGPSAT) yields tantalum and its alloy layers with superior performance. In biomedicine, Zhang et al. [18] prepared Ta coatings on Ti₆Al₄V substrates, finding that an 800 °C treatment for 0.5 h optimized corrosion resistance, mechanical properties, and osteoblast performance. Zhao et al. [19] deposited nano-TaC on the same substrate, achieving a coating hardness six times higher than the substrate and a wear rate reduced by two orders of magnitude. Electrochemical impedance and in vitro cell compatibility tests revealed superior corrosion resistance and biocompatibility. In tribology, tantalum-based alloy layers produced by double-glow technology also perform exceptionally well. Bing et al. [20] prepared a Ta transition layer on 304SS, significantly improving surface roughness and friction performance. The reduction in the friction coefficient and wear rate during friction wear was linked to the transition from metastable β-Ta to stable cubic α-Ta, guiding 304SS surface modification. Zang et al. [21] prepared Hf-Ta-N coatings under varying Ar/N flow rates, finding the highest nitride content at a 1:1 gas flow ratio, exhibiting optimal wear resistance. Yin et al. [22] prepared (Zr, Ta) N coatings on ZrN substrates and tested them under high-temperature tribological conditions. The coating showed a low and stable wear rate at 500 °C, just 32.9% of the wear rate exhibited by ZrN coatings under the same conditions, due to the formation of a dense oxide rich in (Zr, Ta) on the coating surface.

In summary, tantalum and its alloy layers produced by double-glow technology exhibit excellent stability and resistance in harsh environments. The strong metallurgical bonding and gradient elemental distribution enhance their application value. Optimizing the process parameters of double-glow technology to achieve desired surface microstructures and phase compositions is crucial for further improving mechanical properties, corrosion resistance, and high-temperature performance. This has significant implications for ensuring long-term stable service in the demanding environments of high-end equipment.

This study aims to fabricate Ta diffusion layers on 316L stainless steel (316LSS) using double-glow plasma surface alloying technology (DGPSAT) and systematically investigate the effects of alloying time on the microstructure, mechanical properties, and corrosion resistance. By optimizing the alloying parameters, we seek to achieve a dense α-Ta structure with enhanced surface roughness, hardness, and corrosion resistance, providing a cost-effective solution for high-performance coatings in harsh environments.

2. Materials and Methods

2.1. Experimental Materials and Structures

Placing the target material within a thermal insulation barrier elevates plasma density and enhances the infiltration effect of target material elements. As depicted in Figure 1, this schematic diagram illustrates the experimental setup for dual-glow plasma alloying. The sample was positioned on the sample stage, and the inter-electrode distance was meticulously regulated to induce a hollow cathode effect. Upon applying a voltage to the setup, a hollow cathode effect was generated between the planar target material and the planar sample stage, thereby intensifying the ionization of the metal target material. Under the influence of the electric field, the ionized target material elements uniformly coated the surface of the sample, ensuring continuous and stable deposition. Simultaneously, the current and sample temperature also increased, thereby accelerating the diffusion of the target metal elements.

In this experiment, the substrate selected is 316L stainless steel (316LSS), with its main composition shown in

Table 1. Chemical composition of 316LSS (wt.%).

Compositions	Fe	C	Si	Mn	P	S	Cr	Ni	Mo
316LSS	margin	≤0.03	≤1.00	≤2.00	≤0.04	≤0.03	16.50	2.40	14.30

. The material was processed into sheet form measuring 15 mm × 15 mm × 1 mm using laser cutting. Before heat treatment and metal diffusion, the target material, workpieces, and samples required pre-treatment. First, the 316LSS substrate was polished using sandpaper of varying grits (500#, 1000#, 1500#, 2000#). It was then finely polished to a mirror finish using a diamond grinding paste on a metallographic polishing machine. Following this, the substrate was ultrasonically cleaned in acetone and anhydrous ethanol solution for 10 min each and finally dried in an electric heating vacuum drying oven for later use. The Ta target materials and workpieces were polished smooth with a grinding wheel and sandpaper before each experiment, then ultrasonically cleaned in acetone and anhydrous ethanol solution for 10 min, and finally dried in an electric heating vacuum drying oven for later use.

The surface of 316L stainless steel (316LSS) was modified with tantalum (Ta) through double-glow plasma surface alloying technology (DGPSAT). Fe-Ta alloying treatments with varying durations (T1, T2, T3) were systematically studied. Under fixed process parameters (voltage: 950 V, pressure: 35 Pa, temperature: 700–800 °C), the influence of alloying time on the microstructure evolution, mechanical properties, and corrosion resistance of the samples was comprehensively analyzed. The experimental process parameters are shown in

Table 2. 316LSS surface Fe-Ta alloying process parameter table.

Sample	Working Hour/h	Working Pressure/Pa	Working Voltage/V	Working Temperature/°C
T1	3 h	35	800–900	700–800
T2	4 h
T3	5 h

.

2.2. Phase Analysis and Surface Morphology

The X-ray diffractometer utilizes the principle of diffraction to accurately determine the crystal structure, texture, and stress of materials and conduct phase analysis, qualitative analysis, and quantitative analysis [23]. In XRD, a high-energy electron beam bombarded a copper target to generate X-rays, which were diffracted by the sample. The resulting signals were collected and analyzed to determine the crystal structure and phase composition. In SEM, a high-energy electron beam was directed onto the sample, exciting various signals such as secondary electrons and X-rays. These signals were detected, amplified, and processed to obtain magnified images of the sample’s surface morphology and elemental composition.

Before examining the microstructure, pre-treatment is essential. For surface morphology observation, the sample was affixed to the sample stage using conductive adhesive tape. In the case of cross-sectional structure testing, the sample must be embedded in resin beforehand. After solidification, it was extracted, ground, and polished to a mirror-like finish. The sample was then secured onto the sample stage with conductive adhesive for observation.

This test involved placing the pre-treated sample and sample stage into a vacuum chamber and evacuating it to a high vacuum state. Then, the electron gun was turned on to emit electrons, and the image focus was adjusted. The operating voltage was set, and secondary electron imaging was selected to capture images at different magnifications. The built-in Bruker energy dispersive spectrometer of the SEM was used for elemental analysis. An appropriate voltage (10–20 kV) was selected for scanning, and the types, distribution, and content of the elements on the cross-section were observed and analyzed.

2.3. Mechanical Property

2.3.1. Vickers Hardness Test

The surface hardness was measured using the Vickers hardness tester with a pressure module set to 200 gf, applying a force of 1.96 N with the diamond indenter. The holding time was set to 20 s. After the pressure was released, the distance between the two diagonals of the indentation made by the indenter was measured at the observation port. The reading was then input into the instrument to obtain the specific Vickers hardness value. The average value from multiple experiments was calculated for analysis.

2.3.2. Friction and Wear Performance Test

Friction and wear tests were conducted by using a multifunctional reciprocating tribometer (model CFT-1, Gansu Lanzhou, China Zhongke Kaihua Technology Development Co., Ltd., Lanzhou, China). All tested parameters, including set parameters such as load and experimental temperature, as well as obtained experimental data such as the material’s friction coefficient and the wear volume of the coating, are displayed in real-time as data and graphs during this experiment.

In this test, the comprehensive surface performance testing machine was employed to assess the friction and wear performance of the material surface. Initially, the control box was powered on and preheated for 15 min. The sample was then affixed to the test platform. The appropriate sensor and friction mode were selected, and the setup was configured for reciprocating friction with a friction distance of 4 mm. A 30 g weight was utilized for loading, with each group undergoing a 30 min test. Subsequently, the suitable grinding head was mounted, and the GCr15 grinding head was adjusted to maintain a distance of 2–3 mm from the sample by rotating the lift counterclockwise. The position was fine-tuned to ensure observation of the friction track within a reasonable range. A zero adjustment was performed under no load, followed by lowering the grinding head until it contacted the sample, then raising the loading rod by 2–3 mm, placing the loaded weights, and initiating the test to analyze the data.

2.3.3. Adhesion Test Between Alloy Layer and Substrate

The WS-2005 coating adhesion scratch testing machine utilized acoustic emission detection technology. It employed an automatic loading mechanism to continuously apply load to a diamond indenter while simultaneously moving the sample. This process allows the indenter to scratch the coating surface and detect changes in acoustic signals, which were then processed by a computer to generate graphical representations of the measurement results. Ultimately, the bonding strength between the coating and the substrate was determined, specifically the critical load at the moment of coating failure.

In this experiment, the automatic scratch tester for coating adhesion was utilized to evaluate the bonding strength between the 316L stainless steel (316LSS) substrate and the Ta-containing layer. Initially, the computer and control box were powered on and preheated for 15 min. The sample was positioned on the test platform and secured using the built-in fixture. The sliding parameters were configured with a loading weight of 100 N, a loading rate of 50 N/min, and a scratch length of 6 mm, while the signal collected was acoustic emission. After zeroing under no load, the indenter was lowered to contact the sample surface. The loading rod knob was then rotated clockwise to adjust the dynamic load to approximately 0.2–0.35 N before initiating measurement and recording for analysis.

2.4. Electrochemical Measurement

The three-electrode system was widely used in electrochemical analysis, as depicted in Figure 2. It comprises three distinct electrodes and one or more circuits. These electrodes include the working electrode, reference electrode, and counter electrode. Two circuits were formed: one connecting the working electrode to the reference electrode and the other connecting the working electrode to the counter electrode.

The three-electrode system operates on the principle of measuring and analyzing the working electrode potential relative to a reference electrode while controlling the current flow between the working and counter electrodes or the voltage between the working and reference electrodes. This approach enables determining the relationship between potential and current, extracting crucial parameters that characterize electrochemical processes.

In this experiment, an electrochemical workstation with a three-electrode system was employed to assess the corrosion resistance of the sample surface. Before testing, the sample was encapsulated by attaching stainless steel wires to its back and sealing the other five sides with resin, exposing a 1 cm² area of the alloy layer. After the resin hardens, the sample was immersed in an acidic solution (pH = 3) prepared by diluting 98% sulfuric acid to a concentration of 0.5 M and allowed to soak for 24 h. Subsequently, the sample was used as the working electrode in a setup with an Ag/AgCl reference electrode and a platinum counter-electrode for electrochemical testing in the acidic solution. Measurements include the open-circuit voltage, potentiodynamic polarization curves, and impedance curves, which were then analyzed.

Before the experiment, open-circuit voltage testing was performed for 30 min. An initial voltage below the open-circuit voltage threshold was then selected for Tafel curve testing, with a voltage range of −0.5 V to 0.5 V and a scan rate of 10 mV/s. The results were analyzed using CorrView 3.5 software to determine the self-corrosion potential, corrosion current, and corrosion rate. Additionally, electrochemical impedance spectroscopy (EIS) tests were conducted, applying a 5 mV sine wave potential based on the open-circuit voltage over a frequency range of 10⁻² to 10⁶ Hz. The outcomes include Nyquist and Bode plots, which were further analyzed using ZSimpWin 3.60 software to create an equivalent circuit diagram.

2.5. Three-Dimensional Profile and Surface Contact Angle

In this experiment, the 3D profiler was used to measure the surface morphology and roughness of the sample. The sample was first placed flat on the sample stage, and the appropriate magnification was selected for the focusing lens to adjust the focal length for clear imaging of the object’s surface. After capturing the image, an appropriate template was selected for analysis.

The contact angle measurement instrument is commonly used to study surface properties and analyze surface wettability based on Young’s equation [24] describing liquid wetting on solid surfaces. The contact angle is the angle formed at the intersection of the droplet’s contact line with the solid surface. In this experiment, the instrument assessed the hydrophilicity and hydrophobicity of the sample surface. Following cleaning and purging, the syringe with deionized water, the light source, and camera were activated to ensure clear imaging. The needle was lowered in 2 mm increments until the droplet touched the sample surface. The image was captured at this point, and the baseline was adjusted. The software then calculated the contact angle, with multiple measurements taken per group for averaging and further analysis.

2.6. High-Temperature Oxidation Performance

Before the isothermal oxidation experiment, the alumina crucible was ultrasonically cleaned in anhydrous ethanol for 20 min, pre-dried in an electric heating oven, and then baked in a box furnace at 1000 °C for 24 h to remove residual chemical components, achieving a constant weight state. The tantalum plating sample with a size of 15 mm × 15 mm × 1 mm was put into the crucible (the thickness of the tantalum layer depends on the alloying time). The crucible was placed in the box furnace with a heating rate of 5 °C/min, maintained at 950 °C for 5 h, and then cooled with the furnace before removal for weighing. This cyclical experiment was repeated five times, totaling 25 h of oxidation, with data analyzed based on weight gain from oxidation.

2.7. Molecular Dynamic Simulation of the Growth Mechanism of Fe-Ta Alloy Layers

In this study, the molecular dynamics program LAMMPS was utilized to explore the effects of alloying time on the deposition of tantalum-based alloying diffusion layers on 316LSS substrates. The simulation temperature was set to 1023 K to align with experimental conditions. The EAM (Embedded Atom Method) potential [25] was employed to model the interactions between Fe-Fe, Fe-Ta, and Ta-Ta. The specific parameters for the simulation of Ta diffusion on the substrate surface are listed in

Table 3. 316LSS surface Ta diffusion model parameter table.

Model Name	Fe-Ta
Simulation Box Size (Å)	30 × 30 × 30
Fixed Layer Size (Å)	30 × 30 × 4
Thermostat Layer Size (Å)	30 × 30 × 6
Newtonian Layer Size (Å)	30 × 30 × 20
Potential Function	EAM
Time Step (fs)	1

.

Under the conditions of double-glow plasma surface alloying, target atoms were sputtered onto the substrate surface through argon ion bombardment. As the sputtered atoms traveled through the gas phase towards the substrate, they lost energy through collisions. Upon reaching the substrate surface, their final energy can be calculated based on references [26,27]. The initial energy distribution of sputtered atoms as they departed from the target can be determined using the Thompson formula [28], as provided in Equation (1). The final energy of sputtered particles upon their arrival at the substrate can be calculated using Equation (2) [29]:

$$f\left(E_0\right)\propto {\displaystyle \frac{1-\sqrt{\left(E_b+E_0\right)/\gamma E_{Ar}}}{E_0^2{\left(1+E_b/E_0\right)}^3}}$$

(1)

$$E_F=\left(E_0-E_g\right)exp\left[n ln\left(E_f/E_i\right)\right]+E_g$$

(2)

$$n=dp\tau /K_B{T}_g$$

(3)

$E_b$ is the binding energy of Ta, and $E_{Ar}$ is the energy of argon ions. The ratio concerning before and after the collision is $E_f/E_i$ = 1 − $\gamma$/2 [30], where $\gamma$ = 4$m_g{m}_s$/($m_g$ + $m_s$)², with $m_g$ and $m_s$ representing the masses of argon and sputtered metal atoms, respectively. $E_0$ is the initial energy of sputtered particles leaving the target, $E_F$ is their final energy upon reaching the substrate, and $E_g$ is the energy of gas atoms. $n$ is the number of gas-phase collisions, $d$ is the travel distance, $p$ is the sputtering gas pressure, $\tau$ is the hard-sphere collision cross-section, $K_B$ is the Boltzmann constant, and $T_g$ is the background gas temperature.

The specific parameters for the simulation of Ta diffusion on the substrate surface are listed in

Table 4. 316LSS surface tantalum diffusion simulation parameter table.

Model Name	Fe-Ta
Air Pressure (Pa)	35
Temperature (K)	1023
Pole Spacing (mm)	15
Voltage (V)	950
Deposition Rate (atom/ps)	1
Simulated Atom Count	20,000
Relaxation Time (ps)	100

. The initial energy $E_0$ was randomly sampled using the acceptance–rejection method from Equation (1). The energy $E_g$ of the gas atoms was selected based on collision probability, which is simply the product of $f\left(E_g\right)$ and the Maxwell–Boltzmann distribution value at $E_g$. The final energy $E_F$ was then evaluated according to Equation (2). This process was repeated 20,000 times to obtain the final energies of 20,000 sputtered Ta atoms. During the simulation, these Ta atoms were continuously released from a position of one lattice spacing above the growth surface every 1000 time steps, resulting in a deposition rate of 1 atom/ps. This deposition interval not only allows the system to reach thermal equilibrium, avoiding non-physical temperature increases, but also ensures satisfactory computational efficiency. The XY coordinates of the incident Ta atoms were randomly distributed within the XY surface of the simulation cell. The ejection angle of sputtered atoms leaving the target surface follows a cosine-type distribution [31]. During transport, the sputtered atoms were scattered and ultimately heated due to collisions with the background gas. Therefore, the distribution of impact angles of Ta atoms deposited on the substrate surface can be approximated as random. The polar and azimuthal angles of the incident Ta atoms were randomly sampled from uniform distributions U(−π/2, 0) and U(0, 2π), respectively. To enhance the comparability of simulation results at different temperatures, the initial XY coordinates and velocities of the incident Ta atoms were pre-stored in a file and repeatedly read through the LAMMPS Python 3.8.20 interface in all simulations. In each simulation, after releasing 20,000 Ta atoms, a 100 ps relaxation process was performed to allow the deposited film to reach complete thermal equilibrium.

3. Results and Discussion

3.1. Structure Characterization

Figure 3a–c depict the macro and micro-morphologies of Ta-diffused surfaces on a 316L stainless steel matrix at different alloying times, labeled T1, T2, and T3. Optical images reveal that all samples exhibit a deep gray surface with a complete structure and no apparent defects. At 2 × 10⁵ magnification, the T1, T2, and T3 samples show intact and uniformly dense diffusion layers. The pit-like wrinkled structure diminishes as alloying time increases, while the needle-like nanoscale tantalum structure proliferates. This suggests that prolonged alloying ionizes inert gases in the cavity, causing collisions that not only sputter target ions but also heat the cathodic workpiece and samples, converting kinetic energy into thermal energy. This heating activates the sample surface, promoting the deeper diffusion of target elements [32] and forming a unique acicular microstructure. The dense, rough surface may also result from intense ion bombardment during the double-glow surface metallurgy process, where high plasma density and deposition pressure weaken deposited atom energy, and plasma bombardment sputters away weakly adhered particles, shaping the observed morphology [33].

Figure 3d illustrates the phase structure of sample surfaces at different alloying times. The spectrum peaks and positions correspond to the body-centered cubic (BCC) α-Ta (PDF#04-0788), with no impurity phase peaks detected. Diffraction peaks at 2θ angles of 38.5°, 69.6°, and 82.5° correspond to the (110), (211), and (220) lattice planes of Ta, respectively, showing sharp shapes indicative of well-crystallized products. As alloying time increases, the diffraction peak intensity around the (110) lattice plane at 38.5° enhances, while the (211) lattice plane intensity around 69.6° weakens, and the (220) lattice plane intensity around 82.5° remains low. This increase in peak intensity is due to the enhanced preferred orientation of coating growth along the (110) lattice plane and grain refinement. The preferred orientation of the diffusion layer can be quantified using the texture coefficient formula. The texture coefficient formula [34] is as follows:

$$T_{hkl}={\displaystyle \frac{\raisebox{1ex}{$I_m\left(hkl\right)$}\!\left/ \!\raisebox{-1ex}{$I_0\left(hkl\right)$}\right.}{\frac{1}{n}{{\displaystyle \sum }}_1^n\left[\raisebox{1ex}{$I_m\left(hkl\right)$}\!\left/ \!\raisebox{-1ex}{$I_0\left(hkl\right)$}\right.\right]}}$$

(4)

In this context, $T_{hkl}$ represents the texture coefficient. $I_m\left(hkl\right)$ is the relative intensity of reflections from the indexed lattice plane. $I_0\left(hkl\right)$ is the value obtained from the standard powder sample for the same indexed lattice plane, and $n$ is the number of coating reflection peaks. If $T_{hkl}\approx 1$, the sample exhibits a random growth orientation similar to the $\left(hkl\right)$ lattice plane. When $T_{hkl}<1$, the coating grows poorly along the $\left(hkl\right)$ lattice plane. And, when $T_{hkl}>1$, the coating preferentially grows along the $\left(hkl\right)$ lattice plane. A larger $T_{hkl}$ value indicates greater preferential grain growth along the $\left(hkl\right)$ lattice plane [35]. All samples show the strongest diffraction peak at 38.5° for the (110) lattice plane, with texture coefficients greater than 1, while diffraction peaks for other lattice planes are relatively weak. This indicates that the pure Ta diffusion layer on the 316LSS substrate grows with a preferential orientation along the (110) lattice plane in this experiment.

Tantalum typically exists in two crystalline phases: α-Ta and β-Ta. The body-centered cubic α phase is common in bulk metals and is thermodynamically stable, exhibiting good ductility and excellent mechanical properties, making it widely favored. Conversely, β-Ta, a metastable tetragonal phase, is hard and brittle, often leading to adverse effects in applications, such as impacting coating adhesion, corrosion resistance, and wear resistance. Thus, there is significant interest in finding suitable deposition conditions for pure α-Ta coatings [36]. Optimized process parameters were determined by analyzing the Fe-Ta phase diagram and recent studies on tantalum phase transitions, confirming the substrate surface’s body-centered cubic α-Ta structure, which aligns with experimental expectations and is suitable for demanding environments.

Thickness testing and microstructural observation were performed on the sample cross-section. Figure 4 reveals that the alloy layer is complete, dense, and free of significant pores and defects, with a strong bond between the substrate and the Ta layer. The alloy layer thickness ranges from 25 to 55 μm, increasing with alloying time. Line scanning energy dispersive spectroscopy (EDS) analysis identified three distinct regions: the 316L stainless steel (316LSS) substrate, the diffusion layer, and the deposition layer. Ta concentration remains nearly constant in the deposition layer, gradually decreasing in the diffusion layer with increasing depth, while Fe concentration increases. The distribution of Fe and Ta elements is uniform. The deposition and diffusion layers are defined as the tantalum alloyed diffusion layer. The substrate region beneath the diffusion layer only detects iron elements, indicating minimal substrate elements adhering to the diffusion layer surface due to back sputtering. The diffusion layer forms an effective transition, transforming a simple two-phase bond into a reliable metallurgical bond, enhancing interfacial bonding strength and service life.

3.2. Research of Surface State

Figure 5 illustrates the three-dimensional surface profiles of the 316LSS substrate and tantalum-diffused samples at various alloying times. The surface roughness of the alloy layer correlates with alloying time, grain size, and crystal orientation [37,38,39,40], calculated using Equation (5):

$$S_q=\sqrt{{\displaystyle \frac{1}{A}}{{\displaystyle \iint }}_A{Z}^2\left(x, y\right)dxdy}$$

(5)

In this context, $S_q$ represents the root mean square deviation of the surface profile, A is the sampling area, and Z is the vertical coordinate value. The surface roughness $S_q$ of the substrate is 8.69 nm, significantly lower than that of tantalum-diffused samples produced by DGPSA at different times. As alloying time increases, the surface roughness of the samples gradually rises, and $S_q$ values for T1, T2, and T3 samples are 0.16, 0.19, and 0.24 μm, respectively. This enhancement is due to the intense plasma bombardment of the DGPSA process and the preferred orientation of the (110) lattice plane of α-Ta, creating a unique surface acicular micro-nanostructure that benefits surface roughness and improves specific properties.

The water contact angles of the substrate and tantalum-diffused samples were measured, as shown in Figure 6. The substrate surface exhibits a water contact angle of approximately 73.01°, indicating hydrophilicity. In contrast, the water contact angle significantly increased after the alloy layer was prepared. The T1 sample showed a contact angle of 88.18°, while the T2 and T3 alloyed samples reached hydrophobic levels at 90.58° and 101.72°, respectively. The extended alloying time led to an increased substrate temperature and a rougher surface due to high-energy bombardment during sputtering. Furthermore, the unique acicular microstructure of the T3 sample enhanced surface roughness, further improving hydrophobic performance [41].

Moreover, due to the high energy during the DGPSAT bombardment process, there is a particular etching effect while cleaning the surface of 316L stainless steel (316LSS). The difference in elastic modulus and thermal expansion coefficient between the Ta layer and the substrate induces residual stress on the surface, released by forming indentations or protrusions, resulting in a wrinkled structure similar to that shown in Figure 3. As the alloying time increases, the Ta layer preferentially grows at the wrinkled locations, enhancing surface roughness. This roughness allows more air to be trapped between the droplet and the surface during contact, improving the hydrophobic performance of the samples [42].

3.3. Research of Mechanical Performance

The hardness tests of the matrix and tantalum injection samples were analyzed and processed as shown in

Table 5. Vickers Hardness of 316LSS matrix and tantalum-implanted surface samples.

Samples	Vickers Hardness (HV0.2)	95% Confidence Interval
316LSS	152	±8
T1	587	±15
T2	820	±18
T3	970	±20

. As observed, the surface hardness of all tantalum-implanted samples has significantly increased, attributed to the higher hardness of the tantalum layer compared to the 316L stainless steel matrix.

In Figure 7a–d depict the indentation marks from hardness tests on the surfaces of the 316LSS matrix, T1, T2, and T3 samples, respectively. The matrix sample shows relatively softer material, with deeper and clearer indentations from the diamond indenter, indicating surface damage and severe plastic deformation exceeding its critical load. Conversely, the surface indentations on the double-phase heat-treated samples are minimal, suggesting that the tantalum diffusion layer effectively protects the underlying matrix.

Although the alloy layer on the matrix surface consists of the same crystalline structure as Ta, the depth and clarity of the indentations significantly decrease with longer alloying times. This is closely related to the surface microstructure influenced by the processing parameters, specifically the alloying time. Related research indicates that Zhang et al. [43] achieved a hardness of 11.6 GPa for a nanocrystalline α-Ta film with a grain size of 76.5 nm and a hardness of 14.29 GPa for a Ta-10 W film with a grain size of 35.9 nm, both significantly higher than the cast hardness of the metals themselves. Therefore, although the performance of the coatings is related to the preparation methods and parameters, the strengthening effect of the small grain size of Ta plays an important role [44] in the significant increase in hardness of the Fe-Ta coatings in this study. The Hall–Petch equation [45,46] describes grain size strengthening, indicating that smaller grain sizes correlate with increased hardness within a specific range. Moreover, as the alloying time increases, the formation of tantalum acicular nanostructures further enhances the diffusion layer hardness, thereby improving its resistance to plastic deformation and consequently enhancing its mechanical properties.

3.4. Research of Friction Properties

The bonding strength of the alloy layer on the Ta diffusion samples was tested using a scratch tester. As shown in Figure 8, the critical loads for fracture detection of the surface acoustic signals of samples T1, T2, and T3 were approximately 29 N, 36 N, and 64 N, respectively. The critical loads for fracture detection showed good repeatability with 95% confidence intervals of ±1.5 N, ±2 N, and ±2.5 N for T1, T2, and T3 samples, respectively. This indicates that, as Ta diffusion time increases, the bonding between the matrix and the diffusion layer becomes tighter, sufficient for engineering applications [47]. This improvement is attributed to the diffusion layer, which creates a concentration gradient and promotes the close mutual solubility of Fe and Ta, enhancing their mechanical properties.

For materials with an alloy diffusion layer on the surface, the shear stress caused by the movement of the diamond indenter can lead to the formation of cracks perpendicular to the scratch direction and the detachment of the alloy layer from the substrate surface. If the bonding strength is high, more cracks can be observed as they form and propagate to release the stress. Conversely, if the bonding strength is low, cracks form, and the alloy layer may peel off from the substrate surface, releasing stress in that manner [48]. The local magnified view of the scratch revealed inadequate bonding between the alloy layer and the substrate in the T1 sample, resulting in minor cracking of the diffusion layer immediately upon reaching the critical load, as indicated by the acoustic signal. This cracking could provide pathways for corrosive media or oxygen, leading to a reduction in material lifespan. However, with increasing alloying time, the bonding strength between the two layers improved significantly, eliminating this failure mode. Once cracks were initiated, the strong bonding of the alloy layer and the toughness of tantalum prevented the separation of the diffusion layer from the substrate. This demonstrates that the dual-phase technology effectively transitioned the simple interface bond into a superior metallurgical bond. This improvement is beneficial for increasing the bonding strength and extending the service life of the materials.

Finally, the friction performance of the substrate and tantalum-diffused samples at different durations was measured, as shown in Figure 9. The friction coefficient curve exhibits a small fluctuation stage (running-in stage) and a more stable stage [49] (steady wear stage). Within the initial friction stage, approximately the first 5 min, the friction coefficient sharply increases with the extension of sliding time, indicating the running-in stage. After some time, the friction coefficient tends to stabilize, entering the steady wear stage. The test results show that the friction coefficient of the 316LSS substrate is above 1.2, while the friction coefficients of samples T1 and T2 are similar, around 0.8. In contrast, the friction coefficient of sample T3 significantly decreases, averaging around 0.5. The friction coefficient measurements were highly repeatable, with variations less than 5% across multiple tests for each sample. The lower friction coefficient of the alloyed samples can be attributed to the presence of tantalum wear particles formed on the surface of the tantalum-diffused samples. The highly plastic tantalum particles reduce wear caused by friction and provide a good solid lubrication effect. The friction coefficient of 0.5 for T3 is lower than values reported for magnetron-sputtered Ta coatings (0.6–0.8) [15], highlighting the superior lubrication effect of the nanoscale α-Ta structure. Furthermore, good interfacial metallurgical bonding results in substantial load transfer enhancement. With the increase in alloying time, the particle size of Ta continues to decrease, leading to fine-grain strengthening. Consequently, the friction coefficient of the tantalum-diffused samples exhibits less fluctuation compared to the substrate samples. The reduction in the friction coefficient indicates that the wear resistance of the material significantly improves after processing with DGPSAT.

3.5. Research of Anti-Corrosion Performance

Figure 10 shows the potentiodynamic polarization curves of the 316LSS substrate, T1, T2, and T3 samples in an acidic solution with a pH of 3. The polarization curves reveal a significant passive region, indicating the formation of a stable passive film on the surface, which prevents further corrosion of the internal diffusion layer and the substrate, thereby extending its service life. The corrosion potential and corrosion current are important indicators of a material’s corrosion resistance. The corrosion potentials of the tantalum-diffused samples are −2.61 × 10⁻¹ V, −2.31 × 10⁻¹ V, and −1.88 × 10⁻¹ V, respectively, which are all more positive compared to the substrate’s corrosion potential of −3.77 × 10⁻¹ V, indicating better corrosion performance. Similarly, the corrosion currents of the tantalum-diffused samples are 1.07 × 10⁻⁷ A cm⁻², 5.84 × 10⁻⁸ A cm⁻², and 1.36 × 10⁻⁸ A cm⁻², which represent a one-order-of-magnitude decrease compared to the substrate’s corrosion current of 8.87 × 10⁻⁷ A cm⁻², further confirming that their corrosion resistance is superior to that of the substrate. Additionally, the corrosion rate is also an essential criterion for evaluating the corrosion resistance of the samples.

In this experiment, the polarization curves were further fitted using CorrView software 3.5, and the fitted corrosion performance parameters are shown in

Table 6. The surface corrosion current, corrosion potential, and corrosion rate of the 316LSS substrate and tantalum-diffused samples at different times.

Samples	Icorr (A cm−2)	Ecorr (V)	Corrosion Rate (mm/a)
316LSS	8.87 × 10−7	−3.77 × 10−1	1.04 × 10−2
T1	1.07 × 10−7	−2.61 × 10−1	1.26 × 10−3
T2	5.84 × 10−8	−2.31 × 10−1	6.87 × 10−4
T3	1.36 × 10−8	−1.88 × 10−1	1.60 ×10−4

. The corrosion rates of the tantalum-diffused samples are 1.26 × 10⁻³ mm/a, 6.87 × 10⁻⁴ mm/a, and 1.60 × 10⁻⁴ mm/a, respectively, representing a two-order-of-magnitude reduction compared to the 316LSS substrate’s corrosion rate of 1.04 × 10⁻² mm/a, indicating improved corrosion resistance after tantalum diffusion, beneficial for service in harsh acidic environments. This enhancement is due to the dense tantalum oxide passivation layer formed in the acidic solution, which protects the substrate and improves corrosion resistance. Additionally, the T₃ sample exhibits preferred orientation growth along the (110) lattice plane, with a significantly larger interplanar distance and higher atomic density compared to other planes. Given the constant number of atoms in a unit volume, a larger interplanar distance indicates a tighter atomic arrangement, resulting in lower surface energy and greater corrosion resistance under acidic conditions [50], aligning with experimental conclusions. The corrosion rate of 2.83 × 10⁻⁴ mm/a aligns with results for Ta coatings prepared by CVD [13] but with a significantly reduced processing time.

Figure 11 displays the impedance curves of the 316LSS substrate and tantalum-diffused samples. In Figure 11a, coated samples show greater overall impedance and superior electrochemical properties, enhancing corrosion resistance. In Figure 11b, surface-alloyed samples exhibit higher phase angles and stronger passive film capacitive responses, indicating better corrosion resistance [51]. The improved corrosion resistance of the 316LSS substrate after double-glow plasma tantalum diffusion is due to several factors. Firstly, the electrochemical behavior shifts from active dissolution to protective passive film control, with Ta⁵⁺ ions replacing Fe²⁺ ions, preventing direct contact with the acidic electrolyte. Secondly, DGPSAT provides a dense and uniform tantalum coating that results in corrosion-resistant α-Ta without columnar microstructures, in contrast to alternative techniques. The T3 sample, with its unique nanoscale needle-like tantalum structure, shows lower corrosion rates and higher impedance, indicating the significant impact of the surface microstructure on corrosion resistance. High surface roughness creates a hydrophobic surface, preventing prolonged droplet retention and facilitating liquid flow, reducing surface contact [52], and enhancing corrosion resistance.

In Figure 11c,d, the equivalent circuit model for the substrate consists of the solution resistance ($R_s$), coating resistance ($R_c$), charge transfer resistance ($R_{ct}$), coating capacitance ($Q_c$), and double-layer capacitance ($Q_{dl}$). In contrast, the alloyed samples exhibit three distinct time constants. In addition to $R_s$, the circuit includes the following: $R_1{Q}_1$ is the pore resistance and capacitance, $R_2{Q}_2$ is the alloy-layer electrons and capacitors, and $R_3{Q}_3$ is the electric double-layer resistance and capacitors. The total polarization resistance $R_p$ is made up of all the resistors.

The Nyquist plots were fitted using a Randles equivalent circuit ($R_s\left(Q\left(R_{ct}W\right)\right)$), where $R_s$ is the solution resistance, $Q$ is the constant phase element, $R_{ct}$ is the charge transfer resistance, and $W$ is the Warburg diffusion. The $R_{ct}$ values increased from 1.2 × 104 Ω⋅cm² (substrate) to 8.9 × 105 Ω⋅cm² (T3), confirming enhanced corrosion resistance due to the Ta layer.

Figure 12 shows the surface morphology and oxidation kinetics curves of the T1, T2, and T3 samples after isothermal oxidation at 950 °C for 25 h. The 316LSS substrate experienced a weight gain of approximately 16.04 mg cm⁻² during the entire oxidation process, while the tantalum-diffused samples exhibited weight gains of 4.06, 4.49, and 6.20mg cm⁻², respectively, significantly lower than that of the substrate, indicating superior hi gh-temperature oxidation resistance. The 316LSS substrate maintained a relatively low weight gain during the first 15 h, suggesting good short-term high-temperature resistance. However, after the third cycle, a sudden increase in weight indicated substantial oxygen infiltration into the substrate. The loosely formed surface oxide layer was insufficient to prevent oxygen diffusion, leading to the substantial oxidation of internal iron elements and sample failure. In contrast, the tantalum-diffused samples displayed stable, minimal weight gain throughout the oxidation cycles, and their weight gain curves tended to plateau with increasing oxidation time. This indicates that the oxide layer formed on their surfaces was relatively dense in the short term, effectively preventing further oxygen penetration into the substrate and thereby protecting it from oxidation.

Although the oxidation kinetic curves indicate that the weight gain due to oxidation of the Fe-Ta alloy layer is significantly lower than that of the substrate, the high reactivity of tantalum itself can lead to its rapid oxygen absorption and oxide formation when exposed to high-temperature air. The outermost layer, which is in adequate contact with oxygen, forms a high-valent oxide. Due to its increased volume compared to the substrate, this layer becomes powdery and collapses, resulting in the powdery substances observed on the surface in Figure 12a–c. Consequently, the bond between the alloy layer and the substrate continues to weaken, leading to the eventual spalling of the surface oxide layer and a gradual decline in its protective performance for the substrate. Further research is needed for long-term cyclic service in high-temperature environments.

3.6. Study on the Growth Mechanism of the Alloy Layer Surface

In this study, the tantalum diffusion process on 316LSS surfaces was simulated using LAMMPS-2Aug2023. Figure 13 shows the alloying process at different time steps. As alloying time increases, target elements are sputtered by high-energy argon ions, with most particles moving downward due to gravity, while some Ta ions migrate and aggregate on the 316LSS surface. These Ta particles follow the island growth mode [53], forming clusters at vacancies created by ion collisions. Under specific undercooling and energy conditions, these clusters nucleate and grow. Concurrently, argon ions bombard the substrate, heating it rapidly via the hollow cathode effect. This enables Ta atoms to migrate inward, forming a diffusion layer that enhances substrate bonding and mechanical properties. Over time, the diffusion layer develops, and the island growth mode continues above it, ultimately forming a controllable Fe-Ta alloy layer.

4. Conclusions

A tantalum (Ta) layer was fabricated on 316L stainless steel via double-glow plasma surface alloying. This study examined the effect of alloying time on process parameters, analyzed the surface morphology and phase structure, and evaluated the mechanical, corrosion, and high-temperature resistance properties of the Ta layer.

(1)

Dense and complete Fe-Ta diffusion layers with varying alloying times were prepared on a 316L stainless steel substrate using DGPSAT. The processing parameters significantly influenced the surface microstructure; as alloying time increased, the surface pit-like wrinkled structure gradually transformed into a nanoscale acicular α-Ta structure, exhibiting a preferred orientation of the (110) lattice plane.

(2)

Due to the influence of processing parameters, a unique needle-like nanoscale tantalum structure was formed on the surface. This structure significantly increased surface roughness, changing the water contact angle from 73.01° (hydrophilic) to 101.72° (hydrophobic). Additionally, hardness improved from 152 HV_0.2 to 970 HV_0.2. The nanoscale tantalum structure provided excellent solid lubrication effects, reducing the friction coefficient to 0.5.

(3)

The dense tantalum oxide passivation layer formed on the alloy layer surface significantly enhanced the corrosion resistance of the alloyed samples, reducing the corrosion rate from 1.04 × 10⁻² mm/a for the substrate to 2.83 × 10⁻⁴ mm/a. Isothermal oxidation experiments showed that the oxidation kinetic curves indicated a lower weight gain due to oxidation for the alloyed layer compared to the substrate, demonstrating a certain degree of resistance to high-temperature oxidation.

(4)

Molecular dynamic simulations revealed an island growth mode during the formation of the alloy layer, with a gradient distribution of Fe and Ta elements in the diffusion layer, significantly enhancing the bonding strength with the substrate.