Preamble

Vol. XX, No. X, XXX 20XX, NUCLEAR TECHNIQUES

Preparation of Ni-Cr Composite Coatings on Tungsten by Molten Salt Thermal Diffusion Method

CAO Xueshan¹,², DAI Jianxing¹, ZHANG Wei¹, YU Guojun¹
¹Shanghai Institute of Applied Physics, Chinese Academy of Sciences, Shanghai 201800, China
²University of Chinese Academy of Sciences, Beijing 100049, China

Abstract

[Background]: Tungsten is commonly employed as a shielding material in Sr-90 radioisotope thermoelectric generators (RTGs), but its high-temperature oxidation resistance is relatively poor. [Purpose]: This study aims to enhance the oxidation resistance of tungsten under high-temperature operating conditions. [Methods]: A nickel-chromium composite coating was prepared on tungsten substrates using the molten salt thermal diffusion method. Coatings fabricated at various molten salt temperatures were characterized via scanning electron microscopy (SEM), energy dispersive spectroscopy (EDS), and X-ray diffraction (XRD). Cyclic oxidation tests and isothermal oxidation tests were conducted to evaluate the coating's oxidation and spallation resistance, while indentation methods were used to assess coating-substrate adhesion. [Results]: The results indicate that the primary phase composition of the coatings is γ-(Ni,Cr). At a molten salt temperature of 950 ℃, the coating surface exhibited a blocky morphology, impurity phosphorus elements gradually disappeared, and the resulting Ni-Cr coating was purer with good oxidation and spallation resistance. Investigation revealed that a dense Cr₂O₃ layer formed on the oxidized sample surface, and this sample demonstrated the strongest adhesion with a fine-grained coating structure. [Conclusions]: Therefore, 950 ℃ is an optimal temperature for preparing Ni-Cr coatings via the molten salt thermal diffusion method.

Keywords: Molten salt diffusion, Nickel-chromium coating, High-temperature oxidation resistance, Adhesion

Introduction

Radioisotope thermoelectric generators (RTGs) are power generation devices that convert the decay heat of radioactive isotopes into electrical energy using the Seebeck effect. They are commonly employed in extreme environments such as polar regions, deep space, and deep sea to power detectors, sensors, and coastal warning lights. Sr-90 is a suitable nuclide for these batteries; however, its decay produces numerous electrons that generate bremsstrahlung X-rays, requiring high-density materials for shielding. Tungsten is a high-melting-point, high-density material with excellent radiation resistance, making it ideal for RTG applications. Unfortunately, tungsten exhibits poor oxidation resistance at elevated temperatures, with significant oxidation occurring above 600 ℃.

The SENTINEL 100F is a 100-watt-class RTG using Sr-90 as nuclear fuel, and its shielding layer temperature serves as a key reference parameter in this study. During normal operation, the shielding layer temperature typically ranges from 593–643 ℃, while under accident conditions, it can reach up to 852 ℃. Although RTG systems are evacuated, complete vacuum cannot be achieved, and vacuum levels may not remain stable during accident scenarios or long-term operation. Consequently, antioxidant measures are necessary.

One of the most effective approaches is depositing an antioxidant coating on tungsten surfaces. Chromium coatings are typical antioxidant coatings with high melting points that form a dense Cr₂O₃ layer during oxidation, preventing oxygen diffusion into the substrate. The SENTINEL 100F employs hard chromium plating on tungsten for oxidation protection, but hard chromium coatings suffer from poor substrate adhesion and brittleness, potentially leading to cracking and spallation under stress at high temperatures. In contrast, Ni-Cr coatings exhibit superior mechanical properties and adhesion, making Ni-Cr binary alloy coatings a promising alternative for high-temperature oxidation protection of tungsten.

During initial oxidation stages of Ni-Cr alloy coatings, Cr₂O₃, NiO, and NiCr₂O₄ form on the surface. Whether chromium oxides completely replace nickel oxides depends on chromium content; with sufficient chromium, selective oxidation occurs, generating only Cr₂O₃ for antioxidant protection. In the Ni-Cr system, the critical chromium concentration for selective oxidation is approximately 20 wt.%.

Various methods exist for preparing Ni-Cr alloy coatings, with co-electrodeposition in aqueous solutions being widely studied. Zhang et al. prepared a 39 nm Ni-9.6 wt.% Cr nanocomposite film via co-electrodeposition, demonstrating good oxidation resistance at 800 ℃ due to its nanocrystalline structure facilitating chromium selective oxidation. Liu et al. successfully fabricated nanocrystalline Ni-Cr coatings with excellent mechanical properties on 30CrNiMo steel substrates using pulsed current co-electrodeposition. Firouzi-Nerbin et al. deposited Ni-Cr coatings on copper substrates using sulfate-chloride electrolytes, investigating the effects of DC and pulsed current on cathode efficiency, alloy composition, grain size, microhardness, morphology, and corrosion performance. However, co-electrodeposition faces challenges primarily due to the different deposition potentials required for Cr and Ni co-deposition, plus issues with coating-substrate adhesion.

To address these issues, this study proposes a novel two-step approach: first depositing an electroless nickel coating on tungsten substrates, then chromizing the nickel coating via molten salt thermal diffusion. This method avoids the complex operation of matching Cr and Ni deposition potentials. Molten salt thermal diffusion is a technique where samples are immersed in inorganic molten salts, allowing active elements to diffuse into the sample surface to form controllable coatings. This technique is essentially a form of chemical plating that produces dense, uniform coatings with enhanced high-temperature adhesion to substrates compared to electrodeposition.

1.1 Materials

In this study, fluoride salts were selected as the base salt due to their good fluidity, high thermal stability, and high solubility for active elements. To avoid the toxicity of hexavalent chromium, trivalent chromium salts were chosen as the solute. The molten salt formulation consisted of FLiNaK (a eutectic mixture of LiF, NaF, and KF), CrF₃, and chromium powder. Tungsten samples were circular specimens measuring 10×10×3 mm. All chemicals were reagent-grade: sodium fluoride, chromium powder, and ethanol were provided by Sinopharm Chemical Reagent Co., Ltd.; lithium fluoride and chromium fluoride were provided by Shanghai Zhongli Industrial Co., Ltd. and Shanghai Macklin Biochemical Co., Ltd., respectively; potassium fluoride was provided by Xinxiang Yellow River Fine Chemical Co., Ltd.; tungsten specimens (purity >99.99%) were provided by Qinghe County Tengfeng Metal Materials Co., Ltd. and wire-cut to size.

1.2 Preparation of Ni-Cr Composite Coatings

Before experiments, sample surfaces were polished with sandpaper to remove oxide scale and achieve a mirror finish. Samples were then ultrasonically cleaned in distilled water and ethanol, followed by electroless nickel plating at Shanghai Paka Company, resulting in nickel coatings approximately 10 μm thick. During electroless nickel plating, trace phosphorus elements were inevitably incorporated, with P content in the coatings measuring approximately 1–3 wt.%.

The molten salt chromizing process was conducted in a glove box under argon atmosphere with O₂ and H₂O contents below 1 ppm, as moisture and oxygen accelerate fluoride salt corrosion. The molten salt consisted of FLiNaK, CrF₃, and chromium powder, with FLiNaK and CrF₃ in a 95 wt.%-5 wt.% ratio and chromium powder added in appropriate amounts. In operation, the molten salt mixture was placed in a stainless steel crucible and transferred to a resistance furnace. The furnace temperature was adjusted to 750–950 ℃, and electroless nickel-plated tungsten samples were immersed in the molten salt for 4 hours using stainless steel wire. After treatment, samples were removed and cleaned with distilled water to remove residual salt.

1.3 Testing and Characterization

A D8 Advance X-ray diffractometer (XRD) was used to analyze phase composition of the coatings and post-oxidation coatings, operating at a scanning rate of 8°/min over a 10–90° range with Cu Kα radiation (λ = 1.5418 Å). Surface and cross-sectional morphologies were examined using a LEO 1530vp scanning electron microscope (SEM). Local elemental analysis was performed using energy dispersive spectroscopy (EDS) attached to the SEM. High-temperature cyclic oxidation and isothermal oxidation tests were conducted in air using an SXL-1200 box furnace. Cyclic oxidation was performed at 850 ℃, with each cycle comprising 1 hour heating at 850 ℃ followed by cooling to room temperature, for a total of 6 cycles. Before testing, alumina crucibles were preheated to constant weight to hold samples, and each sample (crucible + sample) was weighed using an electronic balance with 10⁻⁴ g precision. Samples were weighed after each cycle, and weight gain per unit area versus time curves were plotted to evaluate high-temperature oxidation and spallation resistance.

Isothermal oxidation testing was conducted at 650 ℃ for 300 hours, with samples weighed at 50, 100, 150, and 300 hours to plot weight gain curves. Coating adhesion was evaluated using an 8150 LK Rockwell hardness tester with a 120° diamond cone indenter at loads of 30, 45, 60, 100, and 150 kg. The minimum load causing coating cracking or spallation was determined through SEM observation.

2. Results and Discussion

2.1 Microstructure of Coatings

Figure 1 [FIGURE:1] shows surface SEM images of electroless nickel coatings and Ni-Cr coatings prepared in molten salt at 750–950 ℃ for 4 hours. The untreated electroless nickel coating (Fig. 1(a)) exhibited a dense, uniform surface composed of blocky particles. Ni-Cr coatings prepared at 750 ℃ (Fig. 1(b)) and 850 ℃ (Fig. 1(c)) showed similar granular morphologies with some pores. In contrast, the coating prepared at 950 ℃ (Fig. 1(d)) displayed a distinct blocky structure resembling a molten state, suggesting elemental dissolution occurred at the coating surface under high temperature.

XRD patterns of Ni-Cr coatings prepared at 750–950 ℃ for 4 hours are shown in Figure 2 [FIGURE:2], revealing four phases: Ni₃P, Cr₃P, γ-(Ni,Cr), and substrate W, with the γ-phase being predominant. The presence of phosphorus-containing phases stems from trace P in the electroless nickel coating. At 750 ℃, Ni₃P diffraction peaks were observed. As temperature increased to 850 ℃ and 950 ℃, Ni₃P gradually transformed into Cr₃P because Cr has a lower standard electrode potential than Ni, making it more reactive with P to form stable compounds. Meanwhile, W diffraction peak intensity decreased with increasing temperature, indicating more chromium diffused into the electroless nickel coating, increasing coating density and reducing X-ray penetration to the substrate.

Cross-sectional SEM morphologies and EDS elemental analysis of electroless nickel and Ni-Cr coatings prepared at 750–950 ℃ for 4 hours are presented in Figure 3 [FIGURE:3]. All coatings were approximately 10 μm thick. The electroless nickel coating cross-section (Fig. 3(a)) was uniform and dense with relatively uniform Ni and P distribution. At 750 ℃ (Fig. 3(b)), a crack filled with granular material was observed in the coating middle. EDS mapping revealed these granular fillings were phosphorus-rich. This occurred because surface P reacted with diffusing Cr to form Cr₃P, creating a driving force for P diffusion outward. At this stage, P had diffused to the coating mid-plane. At 850 ℃ (Fig. 3(c)) and 950 ℃ (Fig. 3(d)), coatings were dense and uniform without through-thickness cracks or spallation. Cr completely diffused into the Ni coating, while P diffused to the surface, with partial P loss at 950 ℃. This corresponds to the elemental dissolution observed in Fig. 1(d).

This phenomenon can be explained as follows: electroless nickel coatings are considered amorphous at room temperature, an inherently unstable structure. At 300 ℃, they transform into a crystalline mixture of Ni and Ni₃P. When P content reaches 11 wt.%, this mixture melts at 880 ℃. Therefore, at 950 ℃, before chromium fully forms stable compounds with the electroless nickel, P content at the coating surface likely reached 11 wt.%, causing melting of the Ni-Ni₃P crystalline mixture. This is beneficial for oxidation resistance as it reduces P content, which is advantageous since studies show Ni-P coatings oxidize 100 times faster than pure nickel at 800–1000 ℃.

2.2 High-Temperature Oxidation Testing and Characterization

Sample appearance after cyclic oxidation in air is shown in Figure 4 [FIGURE:4], where yellow regions indicate tungsten oxide. Pure tungsten was completely oxidized, while electroless nickel and Ni-Cr coatings prepared at 750 ℃ and 850 ℃ exhibited varying degrees of spallation. The coating prepared at 950 ℃ remained relatively intact, showing the best spallation resistance. This enhanced resistance likely stems from increased coating-substrate adhesion at high temperature, which was systematically investigated subsequently.

Weight gain curves after cyclic oxidation in air are presented in Figure 5 [FIGURE:5]. Faster weight gain indicates more rapid oxidation. The electroless nickel coating provided slight oxidation protection compared to pure tungsten, but the effect was modest. Chromizing significantly improved oxidation resistance, with performance increasing as molten salt temperature rose from 750 ℃ to 950 ℃, the latter showing optimal results. Two factors explain this: first, reduced surface P content at 950 ℃ enhanced oxidation resistance; second, samples prepared at 750 ℃ and 850 ℃ showed poorer spallation resistance, and once spallation occurred, oxidation resistance naturally decreased. As shown in Figure 5, initial oxidation stages showed slow weight gain across all temperatures before spallation initiated in the 750 ℃ and 850 ℃ samples, causing rapid weight increase.

Since the coating prepared at 950 ℃ exhibited the best performance, detailed characterization was performed on this oxidized sample. Figure 6 [FIGURE:6] shows SEM images, XRD patterns, and EDS line scan results for the Ni-Cr coating prepared at 950 ℃ for 4 hours after cyclic oxidation. The oxidized surface (Fig. 6(a)) appeared granular, with EDS analysis revealing 54.43 at% O and 45.57 at% Cr, and no detectable Ni or P, suggesting Cr₂O₃ formation. The post-oxidation XRD pattern (Fig. 6(b)) showed γ-(Ni,Cr) and Cr₂O₃ phases, confirming Cr₂O₃ generation. Absence of NiO and NiCr₂O₄ relates to chromium selective oxidation. According to Wagner's theory, when a binary alloy A-B is exposed to oxygen, with A being the more noble element and B the less noble, only B's oxide forms if B concentration is sufficiently high, while A diffuses away from the alloy/oxide interface. In Ni-Cr systems, when chromium concentration exceeds approximately 20 wt.%, selective oxidation occurs. Cross-sectional EDS line scan analysis (Fig. 6(d)) shows chromium content clearly surpassing 20 wt.%. Elevated Cr and O near the coating surface confirm Cr₂O₃ formation, while the intermediate region containing only Ni and Cr corresponds to the γ-phase in Fig. 6(b). Oxygen was confined to the coating surface without diffusing into the substrate, demonstrating excellent oxidation resistance. The cross-sectional morphology (Fig. 6(c)) shows no through-thickness cracks.

Given that nuclear batteries are designed for long-term operation, isothermal oxidation testing at 650 ℃ for 300 hours was conducted on the optimal coating. Sample appearance after this test is shown in Figure 7 [FIGURE:7], where pure tungsten was completely oxidized while the Ni-Cr coated sample appeared black without tungsten oxide formation, indicating coating integrity. Figure 8 [FIGURE:8] shows weight gain during isothermal oxidation at 650 ℃. At all measurement points (50, 100, 150, and 300 hours), the Ni-Cr coated sample exhibited far lower weight gain than pure tungsten, with a relatively slow increase rate, confirming sustained oxidation protection during long-term testing.

2.3 Coating Adhesion Testing

To verify the effect of high temperature on Ni-Cr coating adhesion, indentation tests were performed. Adhesion strength can be evaluated through critical load—the minimum load causing coating cracking or spallation—with higher critical loads indicating stronger coating-substrate bonding. Indentation morphologies of electroless nickel and Ni-Cr coatings prepared at 750–950 ℃ for 4 hours are shown in Figure 9 [FIGURE:9]. At 60 kg load, the electroless nickel coating developed obvious cracks, establishing its critical load between 45–60 kg. Similarly, critical loads for all samples were determined and summarized in Table 1 [TABLE:1]: Sample A (750 ℃) had a critical load below 30 kg, Sample B (850 ℃) between 100–150 kg, and Sample C (950 ℃) above 150 kg. This demonstrates significant adhesion improvement after high-temperature molten salt chromizing. Sample A's poor adhesion primarily resulted from P remaining in the coating interior, forming a crack filled with granular material (Fig. 3(b)). Sample C, processed at the highest temperature, showed the strongest adhesion, explaining its excellent spallation resistance.

Spallation resistance also relates to grain size. Using the Scherrer equation applied to XRD patterns in Figure 2, Sample C's average grain size was estimated at approximately 29 nm, indicating a fine-grained structure. Fine-grained Ni-Cr alloy coatings produce oxide layers with finer grain structures, which exhibit better high-temperature plasticity and creep properties. Stresses at high temperatures can be relieved through creep rather than cracking or spallation, making the fine-grained structure another reason for the coating's strong spallation resistance.

Conclusion

This study successfully prepared Ni-Cr composite coatings on tungsten using molten salt thermal diffusion. Results show that coatings prepared at 750 ℃ and 850 ℃ exhibited granular surface morphologies, while the 950 ℃ coating displayed a molten blocky structure. Four phases were identified: Ni₃P, Cr₃P, γ-(Ni,Cr), and substrate W, with the γ solid solution as the main component. As temperature increased from 750 ℃ to 950 ℃, Ni₃P gradually transformed into Cr₃P while P diffused to the coating surface and disappeared at 950 ℃. The coating prepared at 950 ℃ demonstrated optimal oxidation and spallation resistance, forming a dense Cr₂O₃ layer upon oxidation with the strongest adhesion and fine-grained structure. Therefore, 950 ℃ is an appropriate temperature for preparing Ni-Cr coatings via molten salt thermal diffusion.

Author Contributions

CAO Xueshan: Data curation, manuscript writing; DAI Jianxing: Review and editing; ZHANG Wei: Investigation; YU Guojun: Conceptualization, methodology. All authors have read and approved the final manuscript.

References

1. Wang X, Chan W, Stelmakh V, et al. Radioisotope thermophotovoltaic generator design and performance estimates for terrestrial applications[C]//2017 25th International Conference on Nuclear Engineering, American Society of Mechanical Engineers, Shanghai, China, 2017: V003T013A008.

2. Bennett G L. Mission interplanetary: Using radioisotope power to explore the solar system[J]. Energy Conversion and Management, 2008, 49(3): 382-392. DOI: 10.1016/j.enconman.2007.06.051.

3. Prelas M A, Weaver C L, Watermann M L, et al. A review of nuclear batteries[J]. Progress in Nuclear Energy, 2014, 75: 117-148. DOI: 10.1016/j.pnucene.2014.04.007.

4. Jing H, Xiang Q, Ze R, et al. A skutterudite thermoelectric module with high aspect ratio applied to milliwatt radioisotope thermoelectric generator[J]. Applied Energy, 2023, 350: 121776. DOI: 10.1016/j.apenergy.2023.121776.

5. Ying H, Wen A, Zhou S, et al. Molecular dynamics analysis of primary radiation damage evolution in nickel, iron, and tungsten[J]. Nuclear Techniques, 2023, 46(12): 120301. DOI: 10.11889/j.0253-3219.2023.hjs.46.120301.

6. Kellett E A, Rogers S E. The structure of oxide layers on tungsten[J]. Journal of the Electrochemical Society, 1963, 110(6): 502. DOI: 10.1149/1.2425800.

7. TES. Structural and thermal evaluation of sentinel 100F: INSD-3080[R]. Timonium, Maryland: Teledyne Energy Systems.

8. Liu Z, Zhang Q, Zhang X, et al. Electrodeposition of nanocrystalline Ni and NiCr alloy coatings: Effects of Cr content on microhardness and wear resistance improvement[J]. Journal of Materials Research and Technology, 2024, 30: 3584-3593. DOI: 10.1016/j.jmrt.2024.04.100.

9. Sidelev D V, Kashkarov E B, Syrtanov M S, et al. Nickel-chromium (Ni–Cr) coatings deposited by magnetron sputtering for accident tolerant nuclear claddings[J]. Surface and Coatings Technology, 2019, 69-78. DOI: 10.1016/j.surfcoat.2019.04.057.

10. Quan C, He Y, Zhang J. High temperature oxidation behavior of a novel Ni−Cr binary alloy coating prepared by cathode plasma electrolytic deposition[J]. Surface and Coatings Technology, 2016, 292: 11-19. DOI: 10.1016/j.surfcoat.2016.03.012.

11. Lin K L, Hsu C J, Hsu I M, et al. Electroplating of Ni-Cr on steel with pulse plating[J]. Journal of Materials Engineering and Performance, 1992, 1(3): 359-361. DOI: 10.1007/BF02652390.

12. Conde A, Zubiri F, De Damborenea Y J. Cladding of Ni–Cr–B–Si coatings with a high power diode laser[J]. Materials Science and Engineering: A, 2002, 334(1-2): 233-238. DOI: 10.1016/S0921-5093(01)01808-1.

13. Pawlowski L. The science and engineering of thermal spray coatings[M]. England: John Wiley & Sons Ltd., 2008.

14. Zhang Y, Peng X, Wang F. Development and oxidation at 800 °C of a novel electrodeposited Ni–Cr nanocomposite film[J]. Materials Letters, 2004, 58(6): 1134-1138. DOI: 10.1016/j.matlet.2003.09.006.

15. Firouzi-Nerbin H, Nasirpouri F, Moslehifard E. Pulse electrodeposition and corrosion properties of nanocrystalline nickel-chromium alloy coatings on copper substrate[J]. Journal of Alloys and Compounds, 2020, 822: 153712. DOI: 10.1016/j.jallcom.2020.153712.

16. Kang J C, Lalvani S B, Melendres C A. Electrodeposition and characterization of amorphous Fe-Ni-Cr-based alloys[J]. Journal of Applied Electrochemistry, 1995, 25(4): 376-383. DOI: 10.1007/BF00249658.

17. Zhou Y B, Zhao G G, Zhang H J. Fabrication and wear properties of co-deposited Ni-Cr nanocomposite coatings[J]. Transactions of Nonferrous Metals Society of China, 2010, 20(1): 104-109. DOI: 10.1016/S1003-6326(09)60104-7.

18. Su X, Zhao S, Hou J, et al. Formation of chromium carbide coatings on HT250 steel by thermal diffusion processes in fluoride molten salt bath[J]. Vacuum, 2018, 155: 219-223. DOI: 10.1016/j.vacuum.2018.06.015.

19. European Commission. Restriction of the use of certain hazardous substances (RoHS): Directive 2011/65/EU[S]. Belgium: European Standardisation Organisations, 2011.

20. Sethi R S. Electrocoating from molten salts[J]. Journal of Applied Electrochemistry, 1979, 9(4): 411-426. DOI: 10.1007/BF00617552.

21. Jiang F, Huang W, She C, et al. Electrochemical properties and separation of uranium from lanthanide (Eu, Sm, Yb, La, Ce, and Ho) fluorides in FLiBe molten salt[J]. Nuclear Techniques, 2024, 47(06): 060301. DOI: 10.11889/j.0253-3219.2024.hjs.47.060301.

22. Tomlinson W J, Wilson G R. The oxidation of electroless Ni-B and Ni-P coatings in air at 800 to 1000°C[J]. Journal of Materials Science, 1986, 21(1): 97-102. DOI: 10.1007/BF01144705.

23. Tomlinson W J, Randall S C. Oxidation kinetics of electroless Ni plating in air[J]. Corrosion Science, 1978, 18(6): 573-574. DOI: 10.1016/S0010-938X(78)80030-4.

24. Wagner C. Oxidation of Alloys Involving Noble Metals[J]. Journal of the Electrochemical Society, 1956, 103(10): 571. DOI: 10.1149/1.2430159.

25. Wu Z, Zhang C. Experimental study of Rockwell indentation method for evaluation of coating-matrix bond[J]. Experimental Technology and Management, 2011, 28(6): 294-296. DOI: 10.16791/j.cnki.sjg.2011.06.090.

26. Seal S, Kuiry S C, Bracho L A. Studies on the surface chemistry of oxide films formed on IN-738LC superalloy at elevated temperatures in dry air[J]. Oxidation of Metals, 2001, 56(5/6): 583-603. DOI: 10.1023/A:1012569803467.

27. Ul-Hamid A. TEM study of scale microstructures formed on Ni–10Cr and Ni–10Cr–5Al alloys with and without Y addition[J]. Oxidation of Metals, 2002, 58(1/2): 41-56. DOI: 10.1023/A:1016060423612.