**1. Introduction**

Nb-Si based alloys are considered to be one of the most promising high-temperature structural material, owing to the high melting points (1750 ◦C), medium density and excellent high-temperature strength [1–7]. However, the high-temperature oxidation resistance of Nb-Si based alloys are poor, which limits their application [8–10]. Adding elements such as Cr, B, Ta, Ti, Si in Nb-Si based alloys or preparing coatings on the Nb-Si based alloys can improve the oxidation resistance; however, alloying would compromise mechanical properties of alloys [11,12]. Therefore, more attention is paid to preparing coatings on Nb-Si based alloys.

MoSi2 is a promising coating material for Nb-Si based alloy [13–16]. It can produce SiO2 oxide film with excellent oxidation resistance at high temperatures. Besides, its thermal expansion coefficient (8.1 × 10−<sup>6</sup> ◦C−1) is close to that of Nb-Si based alloys (8.4 × 10−<sup>6</sup> ◦C−1) [17]. However, MoSi2 would suffer pest oxidation at 400–600 ◦C. The addition of B can effectively avoid this disadvantage [18]. The formation of borosilicate glass at high temperatures can effectively protect MoSi2 against pest oxidation at 400–600 ◦C [18]. Furthermore, borosilicate glass coating has a better self-repair ability than SiO2 due to its lower viscosity [19–21]. Fu et al. prepared B2O3 modified SiC-MoSi2 coating on C/C composites by a two-step pack cementation [22]. The coating could protect C/C composites from oxidation at 1500 ◦C in air for more than 242 h. Pang et al. prepared a Mo-Si-B coating on Nb-Si-based alloys by spraying Mo first and then co-deposition of Si and B [23]. The mass gain was 0.92 mg/cm<sup>2</sup> after oxidation at 1250 ◦C for 100 h. However, stresses caused by CTE mismatch between the MoSi2 coating and silica can produce cracks in the oxide film if they exceed the strength of the SiO2.

Atmospheric plasma spraying has attracted widespread attention due to the advantages such as high spraying temperature, high deposition efficiency and precise control of the composition and thickness of the coating [24,25]. Le et al. directly deposited the oxidation-resistant coating MoSi2 on Nb alloy substrate by supersonic air plasma spraying with pure agglomerated MoSi2 powder [25]. After oxidation at 1500 ◦C in air for 43 h, it showed excellent oxidation resistance with mass loss of 5.31 mg cm2. Some of the literature has mentioned that the addition of Zr to Mo-Si-B coating could effectively improve the mechanical properties of MoSi2 at a high temperature. Furthermore, the ZrSiO4 produced by the reaction of dispersive ZrO2 and SiO2 could minimize the CTE di fference between silica and MoSi2, as well as the consumption of SiO2 at a high temperature [26–28]. In this study, the MoSi2-ZrB2 coatings were prepared on Nb-Si based alloy by the atmospheric plasma spraying technology. The e ffects of spraying power on coating structure and the oxidation mechanism of MoSi2-ZrB2 coating were investigated.

#### **2. Materials and Methods**

#### *2.1. Preparation of MoSi2-ZrB2 Coating*

Substrates (Nb-15Si-24Ti-13Cr-2Al-2Hf (at.%) were fabricated by non-consumable arc-melting. The ingots were re-melted and inverted at least four times to guarantee the uniformity of the composition. Samples with a size of 10 × 10 × 8 mm<sup>3</sup> were cut from the ingots. All surfaces were mechanically ground on wet SiC paper to 800 grit, then cleaned ultrasonically with ethanol and dried at about 80 ◦C for 1 h. Commercially available MoSi2 with 95 wt.%, ZrB2 with 5 wt.% powders were selected as raw materials with the purity of 99.9 wt.% and the particle size between 45 and 65 μm. The powders were ground in a planetary ball mill for 2 h, to ensure their uniformity. The MoSi2-ZrB2 coatings were prepared by atmospheric plasma spraying, at the power of 40, 43 and 45 kW, respectively. The samples were designated as mz40, mz43 and mz45, according to the spraying power. The spraying distance was set as 100 mm. Argon was used as primary gas and carrier gas, and hydrogen was used as secondary gas. The detailed parameters are listed in Table 1.



#### *2.2. Isothermal Oxidation*

An isothermal oxidation test was carried out in an open tube furnace, in air, at 1250 ◦C. Each sample was placed in a separate alumina crucible. Samples were taken from the furnace at intervals of 10, 20, 40 and 60 h, and weighed with a crucible, using a precision analytical balance (model CPA225D, Sartorius, Göttingen, Germany) with an accuracy of 0.00001 g.

#### *2.3. Coating Characterization*

Phase composition of the coating and oxidation specimens were analyzed by X-ray di ffraction (XRD, CuK α-radiation, X'Pert Pro, Panalytical, Almelo, Holland) with Cu radiation. Morphology details and elemental distribution characteristics of the coated specimens were investigated by scanning electron microscope combined with energy dispersive spectroscopy (EDS) (Sigma 500, Zeiss, Oberkochen, Germany).
