Deposition of Iridium Coating on Pure Tungsten and High-Temperature Oxidation Behavior at 1300 K

Abstract

Iridium (Ir) coating was electrodeposited on tungsten (W) substrate for resistance to high-temperature oxidation. The reduction of iridium was studied using an electrochemical cyclic voltammetry (CV) measurement technique. The structure characterization and performance testing were carried out by scanning electron microscope (SEM), transmission electron microscopy (TEM), X-ray photoelectron spectroscopy (XPS) and a high-temperature oxidation test. The results showed that the reduction of tetravalent iridium ions to metallic iridium included three reduction steps. The deposited iridium coating had a fine polycrystalline + amorphous structure, no grain orientation phenomenon was observed after electrodeposition, and the microstructure was isotropic. The iridium-coated tungsten metal had excellent resistance to high temperatures at 1300 K, which was attributed to the better chemical stability of the oxide film generated on the surface of the iridium coating.

1. Introduction

Tungsten is the most heat-resistant metal on earth, with high specific gravity; high hardness; good thermal and electrical conductivity, heat resistance, wear resistance, and corrosion resistance; and low expansion and dimensional stability at high temperatures. Among all metals, tungsten has the highest melting point (up to 3380 °C), the lowest vapor pressure, the highest tensile strength (at 1650 °C), and very good corrosion resistance, and most inorganic acids have little erosion effect on it. Thus, it has a wide application value in the aerospace, machinery, mineral processing, metallurgy, nuclear industry, electronics and other industries [1].

Although the melting point of tungsten is high, tungsten and its alloys are highly susceptible to oxidation at high temperatures. High-temperature devices prepared with pure tungsten, such as tungsten crucibles, tungsten heating filaments, etc., have very serious oxidation/volatilization phenomena even when they are protected by inert atmospheres such as nitrogen and argon, due to the high volatility of the oxide phase (W-O system) generated by tungsten at high temperatures, which cannot effectively prevent the further oxidation of tungsten [2,3].

Protective coatings on tungsten surfaces can effectively enhance the high-temperature service performance of tungsten, but tungsten is a high-melting-point metal and the choice of materials for high-temperature protective coatings is very limited. Iridium also has a high melting point (2713 K), and it has better chemical stability and higher temperature-oxidation resistance than tungsten, with very low oxygen permeation and oxide volatilization rates [4], which can effectively prevent the reaction of oxygen with tungsten and oxide volatilizations, enhance the mechanical properties and high-temperature stability of tungsten, and finally, extend the service life of tungsten devices in high-temperature extreme environments.

The main technologies that have been successfully used for the preparation of iridium coatings include the chemical-vapor deposition of metal organics, molten salt electrodeposition, magnetron sputtering, the double-glow plasma method, etc. Metal-organic chemical-avapor deposition (MOCVD) is the most effective method for preparing iridium coatings. Baklanova et al. [4] studied the iridium coatings deposited on ex-PAN carbon fibers using a low-temperature MOCVD approach from iridium (III) acetylacetonate and found that iridium appears to be bound to carbon fiber only by van der Waals forces, so they tried to increase the tensile strength of the carbon fiber through subsequent heat treatment [5]. Vikulova et al. [6] demonstrates the application of iridium coatings onto the pole tips of endocardial electrodes by MOCVD. Yang et al. [7] studied the growth kinetics and microstructure of MOCVD iridium coating from iridium (III) acetylacetonate with hydrogen.

Electrodeposition is also an important method for preparing iridium coatings [8,9,10,11]. The advantage of electrodeposition is that it can be deposited on a variety of complex matrices, and the thickness, chemical composition and structure of the deposited layer can be precisely controlled by controlling the process conditions, so it is suitable for matrix materials of various shapes, especially for heterotypic structural parts. Vot et al. [12] investigated the mechanism for the electrodeposition of iridium onto glassy carbon and platinum substrates. Qian et al. [13] studied metal Ir electrodeposited in the composite ionic liquid BMIC-BMIBF4. Huang et al. [14,15,16,17] prepared iridium coatings on rhenium-coated graphite by electrodeposition and investigated the growth mechanism, mechanical properties and oxidation resistance of the iridium coatings. Zhu et al. [18,19,20] prepared iridium coatings on molybdenum, rhenium and C/C composite substrates by electrodeposition in molten salt in the air atmosphere, respectively.

Magnetron sputtering is also a fast and efficient coating-deposition method. Probst et al. [21] produced iridium coatings within a radio-frequency (rf) magnetron sputtering process with an oblique angle deposition. In order to enhance the properties of iridium coatings for X-ray mirrors, they investigated the influence of total sputtering pressure on the coating properties, especially the correlation between X-ray reflectivity and coating morphology. The double-glow plasma method is a coating-deposition technology developed on the basis of ion nitriding technology. Wu et al. [22] investigated (110)-oriented iridium coating and prepared it by performing the double-glow plasma technique on (200)-oriented Nb substrate.

A great deal of work has also been conducted on the mechanism of iridium coating deposition and oxidation resistance at high temperatures. An et al. [23] built the heat transfer model of iridium coating materials deduced by the heat transfer principle. Gao et al. [24] proposed the reaction process of the deposition of iridium coating on a Mo (110) surface by atomic layer deposition (ALD), and this was studied theoretically by using the density functional theory and periodical slab model. Cui et al. [25] proposed a uniform and continuous iridium–hafnium diffusion coating fabricated using the metallizing process in a molten salt of LiF-HfF4 at relatively low temperatures to improve the oxidation resistance of pure iridium under extreme operating conditions, while Zhang et al. [26] considered a novel Ir-Hf diffusion coating, preparing it on iridium through modified pack cementation to improve the non-equilibrium ablation property while maintaining the oxygen-diffusion resistance.

Iridium is the preferred material for oxidation-resistance coating at present. The difficulty of iridium coating preparation is that it is difficult to obtain an iridium coating with high density and excellent binding strength. In particular, tungsten is a chemically active element, which is quickly passivated in oxygen-containing media. A dense passivating film will be formed on the surface of tungsten, which will hinder the close binding of the deposited film/coating on the tungsten matrix. Therefore, it is very difficult and challenging to obtain an iridium coating on tungsten with an excellent binding force.

In this study, we first developed an aqueous iridium electroplating process, and then adopted a special pre-plating treatment method for tungsten to solve the problem of the binding force between the iridium and tungsten. Finally, we carried out a high-temperature test on the iridium-coated tungsten to investigate the protective effect of the iridium coating on tungsten at high temperatures.

2. Experimental

2.1. Substrate and Pre-Treatment

Pure tungsten (Si < 0.001, Sb < 0.001, Mo < 0.001, P < 0.001, Other impurities < 0.001, W balance, in mass%) with a dimension of 15 mm × 10 mm × 5 mm was selected as substrate. These specimens were sanded smooth with 800# SiC-sandpaper, then degreased with acetone, rinsed with distilled water, and dried in air.

2.2. Electrolytes and Electrochemical Measurements

Ammonium hexachloroiridate (99.9%) purchased from Aldrich (Shanghai, China) was used as 4-valent iridium salt for the electrodeposition of iridium (Ir). Sulfamic acid and boric acid were used to stabilize the pH of the electrolyte, and sodium malonate was added as a brightening agent to obtain a fine crystalline iridium coating. Considering that the oxide film that spontaneously formed on tungsten is very stable at room temperature, a hydrofluoric acid + nitric acid solution was used for surface activation in order to enhance the binding force of the iridium coating. The activation solution and electrolyte composition and operating conditions are shown in

Table 1. The solution components and operating conditions using in electroplating iridium.

Categories	Chemical Designation	Formula	Concentration	Operating Condition
Activation	Hydrofluoric acid	HF	0.9 mol/L	T = 298 K T = 2 min Da = 2 A/dm2
	Nitric acid	HNO3	3.2 mol/L
Electrolyte	Ammonium hexachloroiridate(IV)	H8Cl6IrN2	0.05 mol/L	T = 358 K T = 8 h pH = 0.5~1.5 Dc = 0.2 A/dm2
	Boric acid	H3BO3	0.5 mol/L
	Sulfamic acid	H3NO3S	0.4 mol/L
	Sodium malonate	C3H2Na2O4	0.02 mol/L

.

Cyclic voltammetry (CV) curves were measured to explore the electrochemical reduction of iridium ions in the electrolyte. To obtain electrochemical data with good repeatability, a Pt electrode (S = 0.07 cm²) was employed as a working electrode, a platinum sheet was used as a counter electrode, and a saturated calomel electrode (SCE) was used as a reference electrode.

2.3. Testing and Characterization

In order to investigate the high-temperature performance of the iridium coating, the specimens were tested in a muffle furnace at 1300 K with heating rate of 10 K/min and a holding time of 8 h. The microscopic morphologies were characterized by scanning electron microscopy (SEM, Nova NanoSEM 450, FEI, Hillsboro, OR, USA) and transmission electron microscopy (TEM, JEM2100F, JEOL, Tokyo, Japan) equipped with Energy Dispersive Spectrometer(EDS, XLT SDD, EDAX, Mahwah, NJ, USA). Thin films were cut from as-deposited coating specimens for TEM observation and then were thinned by using in-situ-focused ion beam (FIB, Barcelona, Spain) lift-out (Quanta 200 3D Dual Beam, FEI, Hillsboro, OR, USA). The surface analyses of the passive film were conducted by X-ray photoelectron spectroscopy (XPS, ESCALAB 250, VG Company, St Helier, Jersey) with an Mg Ka source.

3. Results and Discussion

3.1. Electrochemical Reduction of 4-Valent Iridium Ions

Figure 1 shows a family of CVs with the scan rate ranging from 50 to 500 mV/s, obtained on the Pt electrode at 358 K in the 4-valent iridium electrolyte. The main feature of these CVs is that they exhibit three pairs of redox peaks. The first reduction peak (A) appears at around 789.5 mV, and the corresponding oxidation peak (A’) appears at around 828.0 mV. The second peak (B) appears at around 605.3 mV, and the corresponding oxidation peak (B’) appears at around 676.2 mV. The third reduction peak (C) is not too obvious, and the position of the reduction peak can be judged by careful identification to be around 375.7 mV, while the corresponding oxidation peak (C’) is at around 446.1 mV. With the scanning speed from 50 to 500 mV/s, the position of each oxidation or reduction peak does not change much, indicating that these redox reactions are reversible or quasi-reversible reactions. As each pair of redox peaks represents an electrochemical reaction course on a CV curve, the presence of three pairs of redox peaks on the CV curves indicates that the reduction of the 4-valent iridium ion is not a one-step process but may go through intermediate valence states before being reduced to the metallic state. Unfortunately, no detailed study has been reported on the reduction kinetics of iridium (IV). According to the cyclic voltammetry curve-analysis method of reversible electrode processes [27], if the reduction of iridium ions is considered as a redox reaction in a reversible system, the potential difference ($\Delta E_P$) between the reduction peak and oxidation peak is as follows:

$$\left|\Delta E_P\right|=E_{P_a}-E_{P_c}\approx \frac{2.3RT}{nF}$$

(1)

$E_{P_a}$—Anode peak potential, V

$E_{P_a}$—Cathode peak potential, V

n—number of electrons transferred

T—temperature, K

R—gas constant, 8.31451 J/(K·mol);

F—Faraday constant, 9.6485 × 10⁴ C/mol

Adding the constants into Equation (1) yields:

$$\left|\Delta E_P\right|=E_{P_a}-E_{P_c}\approx \frac{2.3\times 8.31451\times 358}{n\times 96485}\approx \frac{71mv}{n}$$

Therefore, the number of electron transfers (n) for each redox reaction can be estimated by the above equation, and the results are shown in

Table 2. Calculation of electron transfer number (n) for each redox reaction.

Redox Reaction	${\mathit{E}}_{{\mathit{P}}_{\mathit{a}}}$, mV	${\mathit{E}}_{{\mathit{P}}_{\mathit{c}}}$, mV	$\left|\Delta {\mathit{E}}_{\mathit{P}}\right|$, mV	$\frac{71mV}{\left|\Delta {\mathit{E}}_{\mathit{P}}\right|}$	n
A-A’	828.0	789.5	38.5	1.84	≈2
B-B’	676.2	605.3	70.9	1.00	1
C-C’	446.1	375.5	70.6	1.01	1

.

As can be seen from Table 2, the number of electrons transferred for the A-A’ reaction is about two, while the number of electrons transferred for the B-B’ and C-C’ reactions is about one. In the aqueous chloride iridate system, iridium ions are likely to be present as ${\mathrm{IrCl}}_6^{2-}$, and its reduction to an iridium metal can be written as [28]

$${\mathrm{IrCl}}_6^{2-}+2e^{-}\to {\mathrm{Ir}}^{2+}+6{\mathrm{Cl}}^{-}$$

(2)

$${\mathrm{Ir}}^{2+}+e^{-}\to {\mathrm{Ir}}^{+}$$

(3)

$${\mathrm{Ir}}^{+}+e^{-}\to \mathrm{Ir}$$

(4)

Actually, the oxidation state of iridium is multivalent, with +1, +2, +3 and +4 valence states all possible, and the electrochemical course of the reduction of iridium from the +4 valence to the 0 valence iridium is very complex. There are many factors that affect the valence stability of iridium ions, including the solution temperature, pH, current density, etc. More in-depth research will be carried out in the follow-up work.

3.2. Electrodeposition of Iridium Coating on Tungsten Substrate

Iridium could not be directly deposited on tungsten substrate in the iridium chloride (IV) solution when the tungsten specimen was immersed in the electrolyte. At the same time, the surface of the tungsten specimen had slightly corroded after a long period of plating, as shown in Figure 2a. As the tungsten metal was extremely easy to passivate, a stable passive film was instantly formed on its surface at room temperature. This passive film, although very thin, prevented the discharge and deposition of iridium on the tungsten surface, which also had a detrimental effect on the bonding of the iridium layer to the substrate. Therefore, it was necessary to choose an appropriate activation method for the tungsten metal before plating, and we used a strongly corrosive nitric acid–hydrofluoric acid mixed solution to activate the tungsten specimens. The energy spectrum analysis in Figure 2b shows that iridium was detected on the surface of the tungsten specimen after chemical activation treatment in nitric acid–hydrofluoric acid mixed solution, implying that the passive film on the tungsten surface was removed to some extent after the tungsten by chemical activation, resulting in the discharge reduction of iridium ions and adsorption of iridium atoms on the tungsten substrate. The surface morphology of the specimen in Figure 2b also shows that the deposited iridium coating was very uniform. The iridium coating deposited on tungsten by chemical activation was very thin since there was still a lot of W information from the substrate in the energy spectrum. It was considered that the chemical activation was not completely effective in removing the passive film from the tungsten before iridium plating and that the iridium may only be deposited at some of the active sites, resulting in the low deposition efficiency of the iridium coating. To further effectively remove the passive film from the tungsten substrate, the tungsten specimens were electrochemically activated in a nitric acid–hydrofluoric acid mixed solution. It was found that a very complete and uniform iridium coating with a tuberous morphology was finally deposited on the surface of the tungsten specimen after this treatment, as shown in Figure 2c.

The microstructure of the iridium coating was observed by transmission electron microscopy (TEM), as shown in Figure 3. It can be observed in Figure 3a that the iridium atoms are arranged in a regular dotted structure in the local microscopic regions, but the microscopic crystalline surfaces in each region do not grow in a consistent direction and the crystalline surface orientation is random, indicating that the deposition process of the iridium atoms is not selectively oriented and the microstructure is polycrystalline. The iridium coating has very small grains, there is no obvious grain boundary between the individual grains, and there are some amorphous regions between the various crystalline regions, so the overall structure of the iridium coating is fine polycrystalline + amorphous. It is due to the fine polycrystalline + amorphous structure of the microstructure of the iridium coating that the electron diffraction pattern of the plating shows a circular characteristic. The crystal structure diffraction card (Ir-PDF#01-1212) of iridium metal can be consulted to calibrate the crystal surface index for the different crystal surface spacing of the Ir coating, as shown in Figure 3a,b.

3.3. Oxidation Behavior and Mechanisms of Iridium Coating

Oxidation tests were performed on pure W specimens and iridium-coated tungsten specimens at 1300 K, and the oxide appearance is shown in Figure 4. It can be seen that a thick oxide skin was formed on the surface of pure tungsten after 8 h of oxidation, and the oxide skin was dark green and very loose (Figure 4a). The tungsten specimen after iridium coating still maintained a metallic luster on the surface after oxidation, and there was no obvious green oxide formation (Figure 4b). The gravimetric analysis of the oxidized specimen showed that the weight-gain rate of the pure tungsten specimen after oxidation was 0.48 ± 0.04 mg/cm², while the weight-gain rate of the iridium-coated tungsten specimen after oxidation was 1.35 ± 0.03 mg/cm². After tungsten was plated with iridium, the oxidation weight-gain rate was increased about 2.8 times (Figure 5). It was generally believed that after the high-temperature oxidation of metal materials, the smaller the weight-gain rate, the more stable the material properties, and the more resistance to high-temperature oxidation. However, the oxidation rate of the pure tungsten mentioned above was lower than that of the iridium-plated specimens, which does not mean that the oxidation resistance of tungsten was better than that of the iridium coating. The reason for the low weight-gain rate of tungsten specimens after oxidation was that the oxides of tungsten were volatile and continuously volatilized and peeled off after being formed on the surface of the specimens, so the measured weight-gain rate was not the actual oxidation loss of the tungsten metal. In contrast, iridium oxides were relatively stable and had low volatility, which enabled the iridium coatings to maintain a metallic luster.

To further investigate the oxidation mechanism of W metal and iridium coatings, the elemental composition and chemical state of the oxide film were investigated by XPS analysis. In order to remove the contaminants on the surface of the sample and obtain the elemental distribution of the passive film profile, the surface of the sample was etched by argon ions sputtering at 0.1 nm/s for 300 s. All the XPS data results were processed by XPSPEAK41 software.

As shown in Figure 6a, it can be determined that the main constituent elements of the oxide were W and O. Comparing the binding energy of the W4f element in Figure 6b with the standard spectrum database [29], it can be determined that the oxide film was mainly composed of W monomers at 31.5 and 33.7 eV and compounds WO₃ at 36.0 and 38.1 eV in the binding energy. Therefore, the main component of the green oxide film in Figure 4a is WO₃, so the dark-green oxide film in Figure 4a is mainly composed of WO₃, while the oxide layer also contained part of the volatile W element.

Similarly, the results of the XPS analysis of the oxide film on the iridium-coated tungsten is shown in Figure 7. The survey spectrum in Figure 7a shows that the main constituent elements of the oxide film produced on Ir coating are W, Ir, and O. It is confirmed in Figure 7b that the characteristic peaks of W4f appeared at 32.0 and 34.1 eV in the binding energy spectrum, corresponding to the characteristic peaks of the W metal, while the characteristic peaks appeared at 36.1 and 38.1 eV, corresponding to the characteristic peaks of WO₃. The characteristic peaks of Ir_4f appeared at 61.7 and 64.9, corresponding to the characteristic peaks of IO₂ [30]. Compared with Figure 6b, the binding energy of W metal shifted by about +0.5 eV, but the peak strength was stronger than that of tungsten oxide, indicating that the oxidation resistance of tungsten was improved.

In order to better understand the generating mechanism of oxidation products at high temperatures, the thermodynamic data book was consulted to obtain thermodynamic data for the reactions of each relevant substance at 1300 K, as shown in

Table 3. Thermodynamic reaction equilibrium constants of various substances at 1300 K [31].

Substance	$\Delta {\mathit{G}}_{\mathit{f}}$ (kJ/mol)	$log{\mathit{K}}_{\mathit{f}}$
${\mathrm{W}}_{\left(\mathrm{g}\right)}$	664.772	−26.711
${\mathrm{Ir}}_{\left(\mathrm{g}\right)}$	468.526	−18.826
${\mathrm{IrO}}_2{}_{\left(\mathrm{S}\right)}$	−18.880	0.759
${\mathrm{IrO}}_2{}_{\left(\mathrm{g}\right)}$	187.869	−7.549
${\mathrm{WO}}_3{}_{\left(\mathrm{s}\right)}$	−513.405	20.629
${\mathrm{WO}}_3{}_{\left(\mathrm{g}\right)}$	−221.035	8.881

.

For pure tungsten specimens, the process of oxidation products can be represented by the following reaction equation:

$$\frac{2}{3}\mathrm{W}\left(\mathrm{s}\right)+{\mathrm{O}}_2=\frac{2}{3}{\mathrm{WO}}_3\left(\mathrm{s}\right)$$

(5)

Gibbs energy of reaction: $\Delta G_f=-513.405$ kJ/mol. The equilibrium oxygen partial pressures for oxide production at 1300 K are as follows:

$${\mathrm{logP}}_{{\mathrm{O}}_2}=-\frac{2}{3}\log K_f^{{\mathrm{WO}}_3}=-\frac{2}{3}\times 20.629$$

$${\mathrm{P}}_{{\mathrm{O}}_2}=1.77\times {10}^{-14} \mathrm{bar}$$

For pure iridium metal, oxidation can be represented by the following reaction equation:

$$\mathrm{Ir}\left(\mathrm{s}\right)+{\mathrm{O}}_2\to {\mathrm{IrO}}_2\left(\mathrm{s}\right)$$

(6)

Gibbs energy of reaction: $\Delta G_f=-18.88$ kJ/mol. The equilibrium oxygen partial pressures for oxide production at 1300 K are as follows:

$${\mathrm{logP}}_{{\mathrm{O}}_2}=-\log K_f^{{\mathrm{IrO}}_2}=-0.759$$

$${\mathrm{P}}_{{\mathrm{O}}_2}=1.75\times {10}^{-1} \mathrm{bar}$$

As can be seen from the above calculations, the equilibrium oxygen partial pressure of iridium oxide was much higher than that of tungsten oxide when oxidized at 1300 K. This means that oxidation reactions started at a very low oxygen pressure (1.77 × 10⁻¹⁴ bar) for tungsten, while a higher oxygen partial pressure (1.75 × 10⁻¹ bar) was required for the generation of iridium oxide, so iridium was significantly more resistant to high-temperature oxidation than tungsten.

In addition to the rate of oxide production, the volatility of the oxide had an important influence on the oxidation resistance of metallic materials. The results of Gulbransen et.al [30] showed that WO₃ was an oxide with significant volatility. It can also be seen from Table 2 that the Gibbs reaction energy for the gaseous WO₃ was from −221.035 kJ/mol < 0. Therefore, from a thermodynamic point of view, W can be directly formed into gaseous WO₃; in other words, WO₃ had a high volatility. So, the oxidation of tungsten at high temperatures, with extremely high rates of oxidation products and volatilization, could be catastrophic to the tungsten substrate.

In contrast, iridium generated gaseous IrO₂ with a Gibbs reaction energy of 187.869 kJ/mol > 0. Therefore, with iridium it was more difficult to generate volatile IrO₂, and the stability of the IrO₂ film was excellent, resulting in a very small loss of iridium coating in the high-temperature oxidation process, which can provide excellent protection for the substrate.

4. Conclusions

Corrosion-resistant and high-temperature-resistant iridium coatings can be prepared by electrodeposition from tetravalent iridium salt solutions. The reduction of tetravalent iridium ions to metallic iridium consisted of three reduction steps, i.e., in the first step, tetravalent iridium ions were reduced to 2-valent iridium after gaining two electrons; in the second step, 2-valent iridium ions were reduced to 1-valent iridium after gaining one electron; and finally, 1-valent iridium was reduced to metallic iridium after gaining one electron.

Intact and adhesive iridium coating was deposited on a tungsten substrate by the electrochemical activation of tungsten in nitric acid–hydrofluoric acid mixed solution. The deposited iridium coating had a fine polycrystalline + amorphous structure, no grain orientation phenomenon was observed after electrodeposition, and the microstructure was isotropic.

The oxidation of pure tungsten at 1300 K was very serious, the surface quickly generated a WO₃ film, and the WO₃ film was strongly volatile at high temperatures, which could not have a protective effect on the tungsten substrate. The iridium coating was chemically very stable at high temperatures, and only a thin IrO₂ oxide film was generated, which provided good high-temperature protection for the tungsten metal.