Designing SiC/IrSi₃ Composites for Aggressive Environments: Wetting Characteristics of the Liquid Si-Ir Eutectics in Contact with SiC and C-Materials

Abstract

The design and fabrication of metal matrix materials (MMCs), as well as the densification and joining of ceramic matrix composites (CMC), are still very challenging. For SiC- and C-based composites, liquid-assisted processing routes, such as the spontaneous infiltration process, emerge among the most cost-effective processes. To succeed in Ir-Si/SiC refractory composite fabrication by spontaneous infiltration, the wetting characteristics of the Ir-Si/SiC system, the surface and transport properties (surface tension and viscosity) of liquid Ir-Si alloys, and microstructural evolution at the interfaces formed between solid SiC (or C) with Ir-Si melt, have been carefully examined. Specifically, the wettability and interaction phenomena occurring at the Si-Ir eutectics/SiC interface as a function of temperature were investigated in the temperature range of T = 1350–1400 °C by the sessile drop method under an inert atmosphere with reduced oxygen content, and the results are presented and discussed in this paper. Taking into account the thermodynamics of the Si-C-Ir system, the interfacial phenomena and subsequent microstructural evolution are well-related to the process parameters, and the properties and characteristics of the as-produced interfaces may be predicted accordingly. The experimental conditions and results of wetting experiments, together with thermodynamic-based models predicting thermophysical property values of liquid Ir-Si alloys, are valuable key input data that are now available for the numerical study of infiltration processes.

1. Introduction

Material design is presently focused on developing high-performance materials combining high-temperature structural stability, long-term resistance to corrosive environments, and self-healing capability. These are key requirements for several highly demanding applications, such as aerospace, energy power plants, nuclear reactors, advanced electronics, catalysis, etc. In such cases, based on available datasets of experimental and theoretical properties, the crucial step consists of ad hoc material selection. Indeed, the exposure of materials to high temperatures and chemically aggressive environments for extended periods of time negatively impacts their service life by progressively decreasing their performance, leading to unexpected material failure.

To date, refractory MMCs and CMC materials, particularly SiC-based composites, meet all the requisites [1,2] thanks to their excellent combination of thermo-mechanical properties, high wear resistance, and remarkable chemical stability when exposed to oxidizing/corrosive environments [3,4]. Their overall performance can be further improved by combining SiC with a Si-based metal phase containing Ir as an alloying element. In fact, Ir exhibits a unique set of properties such as high melting temperature (2446 °C) [5], high temperature-specific strength, and remarkable oxidation and corrosion resistance, resulting in them being suitable for a wide range of applications [6]. Additionally, Ir-silicides succeed as a Si-replacement phase, and as recently documented by the authors of [7,8], a substantial improvement against the oxidation of the produced composites via reactive infiltration was observed. Specifically, highly dense, nearly shaped SiC/IrSi₃ composites were successfully produced in the temperature range of T = 1250–1350 °C under a vacuum by reactive melt infiltration of liquid Si–62 wt% Ir eutectic alloy (Si–19.5 at% Ir) into bimodal SiCp-C preforms with C content of 4 wt% [7,8].

Taking into account the observed infiltration kinetics and activation energy of about 330 kJ/mol, researchers deduced that the process was mainly controlled by solid–liquid interfacial reactions [9]. On the other hand, the infiltration rate is higher than that observed in similar experiments performed by infiltrating Si and Si-based alloys into porous graphite [10,11,12], together with an infiltration front deviating from the classical “truncated cone” shape [8], suggesting that capillarity and viscous forces may play a concomitant role.

To clarify the interfacial phenomena observed during the infiltration process used for manufacturing SiC/IrSi₃ composites, it is necessary to take a step back.

A preliminary investigation into the reactivity of the solid state at Ir/SiC interfaces has been performed by researchers as a function of SiC crystallinity in the temperature range of 1200–1450 °C under a vacuum and with reduced oxygen content in the surrounding atmosphere [13]. In agreement with thermodynamics, interfacial melting occurred by the formation of lower-melting-point Ir-silicides, and all the diffusion phenomena were driven by temperature and SiC crystallinity, as expected. Interfacial melting was also observed in [14], and similarly, the findings demonstrated the capability of Ir-silicides to join monolithic SiC ceramic parts, even at temperatures lower than 1400 °C.

As already demonstrated [11,12,13,14,15,16,17], the relevant phenomena occurring during the melt infiltration process can be successfully predicted and then controlled by performing preliminary experimental and theoretical investigations on the inherent properties of the liquid Si-infiltrating phase, in terms of the thermodynamics and thermophysical properties of the melt [18,19,20], as well as the wetting characteristics and reactivity at the interfaces of Si-based alloys in contact with SiC- or C-based materials [21,22,23].

To the best of our knowledge, neither thermophysical nor wetting property data of the Si-Ir/(SiC, C) system are available in the literature. Accordingly, the model predicted values of the enthalpy of mixing, molar volume [24,25], surface tension [26], and viscosity [27] of Ir-Si melts, calculated for T = 1400 °C, are used to substantiate the experimental data of the present work. In order to elucidate the wetting behavior of liquid Ir-Si alloys in contact with SiC and C, the calculated property isotherms of the Ir-Si system were analyzed. Based on this, after imposing theoretical and experimental constraints related to the wetting tests and spreading kinetics of a liquid Si-Ir eutectic alloy in contact with SiC and C, it is possible to define and control the interaction mechanisms taking place during the formation of the interface between the liquid phase and solid substrate, as well as the microstructural evolution. In addition, reliable thermophysical property data, such as viscosity, density and surface tension, are valuable inputs for modeling and optimizing the reactive infiltration process [7,8,18,22,23].

Moreover, interaction studies performed by wetting tests under well-defined operating conditions allow gaining insights into the more relevant phenomenon affecting the reactive infiltration mechanism, i.e., the pore-narrowing/closure effect given by the growth of SiC crystals at the pore walls. To this aim, wettability and spreading kinetics observed in the Si-Ir eutectics/SiC system at T = 1400 °C were compared with the results obtained at the Si-Ir alloy/SiC single-crystal triple line. Finally, to analyze the role of C in the preforms (as a SiCp-binder phase) during the infiltration process, wetting tests on the Si-Ir eutectic alloy/C system were also performed, and the preliminary results are shown here. In particular, to understand the influence of the C reactivity on the infiltration process, wetting tests on highly oriented graphite (C-CVD) and amorphous graphite (GC-Glassy Carbon) were performed and the results compared. Indeed, different C crystallinities play a critical role in controlling the reactivity and diffusivity of C into molten Si [28]. All the interaction phenomena observed by analyzing wetting behavior as a function of operating conditions (i.e., time and temperature) are related to the microstructural evolution at the interface analyzed post-mortem by light and SEM/EDS microscopies, as well as by the XRD technique.

2. Materials and Methods

2.1. Materials and Sample Preparation

Two types of SiC substrates were selected: pieces (with an area 1 cm²) of hot-pressed polycrystal SiC (HP-SiC-Goodfellows©) and 6H-SiC single crystal (Nitride Crystals©). The HP-SiC consists of α-SiC with zero free Si (as confirmed by X-Ray diffraction analysis) and an absence of sintering aids. SiC substrates were metallographically polished (Diamond UltraPrep 30-0.25 µm by Struers©) with a final roughness of Ra ≤ 0.25 µm.

Glassy carbon (GC) and CVD-graphite (C-CVD), provided by Alfa-Aesar© and Advanced Carbon Technologies©, respectively, were used as C-based substrates. Surface roughness values of Ra ≈ 20 nm (GC) and Ra ≈ 9.15 μm (C-CVD) (Figure 1) were detected, measured by an optical confocal-interferometric profilometer (Sensofar S-neox).

Figure 1. 3D surface reconstruction of the C-based substrates at different magnifications: (a) C-CVD 20× and (b) GC 100×.

The Si–62 wt% Ir (Si–19.5 at% Ir) eutectic alloy samples of about 50 mg were prepared by mixing nominal weights of high-purity Ir and Si (both 99.98%—Goodfellows©). To avoid loss of material, alloy samples were arc-melted under an Ar (N60, O₂ < 0.1 ppm) atmosphere. Prior to arc-melting the metal samples, the residual oxygen content inside the chamber was reduced by melting a Zr drop. To ensure the homogeneity of the alloy composition, each sample was arc-melted 3 times. As-cast Si-Ir composition was checked at the cross-sectioned sample by both SEM (to reveal the eutectic microstructure) and XRD (to identify the phase composition), as shown in Figure 2.

Figure 2. (a) As-produced Si-Ir eutectic alloy drop by arc melting. (b) SEM image at the cross-section. (c) XRD pattern.

2.2. Method, Procedure and Device

Before the wetting tests, both sample and substrate were ultrasonically rinsed and dried by compressed air. At room temperature, the alloy/substrate assembly was placed onto a graphite plate/sample holder, introduced into the furnace and located at the central part of the heater, as shown in Figure 3a,b. Once the sample was leveled at a horizontal plane, in order to remove any contaminant, the chamber was degassed under a vacuum (down to 10⁻⁴ Pa) for two hours. Subsequently, the metal/substrate couple was heated under a vacuum by an 800 kHz high frequency generator coupled to a graphite tube ensuring an experimental surrounding atmosphere with reduced oxygen content (PO₂ < 10⁻⁶ Pa). In order to avoid evaporation and material loss, the wetting experiments were performed under a static Ar atmosphere. During all the experiments, the temperature value was continuously monitored by a pyrometer (CYCLOPS 52-Minolta©) and an S thermocouple was placed near the sample.

Figure 3. Schematic layout of (a) the experimental setup used with (b) the internal view of the alloy/substrate couple during wetting experiments, (c) recorded in real time by ASTRAview^® software.

The wetting experiments were carried out by the contact heating sessile drop method. Except for the Si-Ir eutectics/6H-SiC couple, the temperature was increased up to the testing temperature with a rate of 5 °C/s and kept constant for 15 min. For the wetting test performed on the Si-Ir eutectics/6H-SiC sample, the temperature was raised with different heating rates. More precisely, the sample was firstly heated up to the melting point of T = 1251 °C, then the temperature was further increased up to T = 1400 °C and finally to T = 1450 °C, with a heating rate of 2, 1.2 and 0.15 °C/s, respectively.

At the end of the imposed dwell times, the samples were cooled down to room temperature (cooling rate 20 °C/s) to preserve the obtained interface microstructures.

The evolution of the contact angle was monitored and recorded (10 frames/s) by a high-resolution camera (CMOS—The Imaging Source). The images collected were processed in real time by image analysis software developed ad hoc (ASTRAVIEW^®) [29]. The software allows the real-time acquisition of contact angle and drop geometric variables (R—base radius and H—height). Spreading kinetics can be analyzed off-line by measuring the geometric variables of every single frame with great accuracy.

At the end of the wetting experiments, the solidified couples were removed from the experimental device, embedded into epoxy resin (Struers©), cross-sectioned and metallographically mirror-polished for microstructural characterization by a SEM/EDS (ZEISS-Merlin VP Compact) instrument paired with a BRUKER-Quantax 400 system. Images were acquired in both backscattered electron (BSE) and secondary electron (SE) modes with an X-ray diffractometer (Bruker D8 Advance) with a Goebel mirror and a Kristalloflex K 760-80F X-Ray generation system, fitted with a Cu cathode and a Ni filter. Spectra were registered between 5° and 80° with a step of 0.05° and a step time of 3s.

2.3. Thermodynamic Modeling

Thermodynamic modeling is the easiest way to calculate thermophysical properties, such as the surface tension and viscosity, that, in the framework of different models, can be evaluated from the mixing thermodynamic functions (e.g., the enthalpy and the Gibbs free energies of mixing and activities) of alloy melts [30,31]. Until now, there was no such experimental data for the Ir-Si liquid phase. Using Miedema’s model [24,32], Takeuchi and Inoue [25] have calculated the enthalpy of mixing of 2628 binary systems, including that of liquid Ir-Si alloys, and the most recent was taken as the input data for the calculations of the molar volume, surface tension and viscosity. The surface tension and viscosity isotherms were obtained by the Quasi-Chemical Approximation (QCA) for a regular solution [26,30] and the Moelwyn–Hughes (MH) model [27], respectively. Detailed mathematical formalisms of all models used in the present work can be found elsewhere [18]. In the following, each model is briefly described, and only the relevant equations used for the calculations are provided.

2.3.1. Enthalpy of Mixing and Molar Volume

In the framework of Miedema’s model [24,32], Takeuchi and Inoue [25] have introduced a relatively simple description for the enthalpy of mixing ∆H^mix of liquid binary alloys. Indeed, a cubic polynomial approximates a sub-regular solution model with interaction parameter Q taking into account the values at the equiatomic compositions, and ∆H^mix can be calculated as a function of the composition, as follows.

$$∆{\mathrm{H}}^{\mathrm{m}\mathrm{i}\mathrm{x}}=4\left({\sum }_{\mathrm{k}=0}^3{\mathrm{Q}}_{\mathrm{k}}({\mathrm{c}}_{\mathrm{A}}{-{\mathrm{c}}_{\mathrm{B}})}^{\mathrm{k}}\right){\mathrm{c}}_{\mathrm{A}}{\mathrm{c}}_{\mathrm{B}}$$

(1)

where c_A and c_B are the compositions of A and B components of an A-B alloy.

The calculations of ∆H^mix involve the molar Volume V_Alloy, expressed by

$${\mathit{V}}_{\mathit{A}\mathit{l}\mathit{l}\mathit{o}\mathit{y}}={\sum }_{\begin{array}{c}i,j=1\\ i\ne j\end{array}}^2{\mathit{c}}_{\mathit{i}}·{\left\{\left[1+\mathit{a}·\left(1-{\displaystyle \frac{{\mathit{C}}_{\mathit{i}}{\mathit{V}}_{\mathit{i}}^{2/3}}{{\mathit{C}}_{\mathit{i}}{\mathit{V}}_{\mathit{i}}^{2/3}+{\mathit{C}}_{\mathit{j}}{\mathit{V}}_{\mathit{j}}^{2/3}}}\right)·\left({\mathit{\phi }}_{\mathit{i}}-{\mathit{\phi }}_{\mathit{j}}\right)\right]{\mathit{V}}_{\mathit{i}}^{2/3}\right\}}^{3/2}$$

(2)

where $c_i$, $c_j$ $V_i$, $V_j$, ${\phi }_i$ and ${\phi }_j$ are the compositions, molar volumes and work functions of the A and B alloy components, and a is a constant.

2.3.2. Surface Tension

The surface tension of liquid Ir-Si alloys was analyzed by the Quasi-Chemical Approximation (QCA) for a regular solution. The basic hypotheses of the QCA model are a layered interfacial structure and the energetics expressed by the thermodynamic functions of mixing. Many details on the QCA or the first approximation are given in [26,30]. The QCA for a regular solution is characterized by one interaction parameter ω that, depending on the availability of thermodynamic data of the system investigated, can be expressed either as a number or as a function of temperature. In the framework of QCA, the surface tension is calculated by

$$\sigma ={\sigma }_{\mathrm{A}}+{\displaystyle \frac{{\mathrm{k}}_{\mathrm{B}}\mathrm{T}(2-\mathrm{p}\mathrm{Z})}{2\alpha }}\mathrm{l}\mathrm{n}{\displaystyle \frac{{\mathrm{C}}^{\mathrm{s}}}{\mathrm{C}}}+{\displaystyle \frac{\mathrm{Z}{\mathrm{k}}_{\mathrm{B}}\mathrm{T}}{2\alpha }}\left[\mathrm{p}\mathrm{l}\mathrm{n}{\displaystyle \frac{({\beta }^{\mathrm{s}}-1+2{\mathrm{C}}^{\mathrm{s}})(1+\beta )}{(\beta -1+2\mathrm{C})(1+{\beta }^{\mathrm{s}})}}-\mathrm{q}\mathrm{l}\mathrm{n}{\displaystyle \frac{(\beta -1+2\mathrm{C})}{(1+\beta )\mathrm{C}}}\right]$$

(3)

$$\sigma ={\sigma }_{\mathrm{B}}+{\displaystyle \frac{{\mathrm{k}}_{\mathrm{B}}\mathrm{T}(2-\mathrm{p}\mathrm{Z})}{2\alpha }}\mathrm{l}\mathrm{n}{\displaystyle \frac{{(1-\mathrm{C}}^{\mathrm{s}})}{(1-\mathrm{C})}}+{\displaystyle \frac{\mathrm{Z}{\mathrm{k}}_{\mathrm{B}}\mathrm{T}}{2\alpha }}\left[\mathrm{p}\mathrm{l}\mathrm{n}{\displaystyle \frac{({\beta }^{\mathrm{s}}+1-2{\mathrm{C}}^{\mathrm{s}})(1+\beta )}{(\beta +1-2\mathrm{C})(1+{\beta }^{\mathrm{s}})}}-\mathrm{q}\mathrm{l}\mathrm{n}{\displaystyle \frac{(\beta +1-2\mathrm{C})}{(1+\beta )(1-\mathrm{C})}}\right]$$

(4)

where C, C^s, β and β^s are the compositions of the B component and auxiliary variables containing the interaction energy parameter of the bulk and surface phase, respectively. T is the temperature in K, Z is the coordination number, and p and q are the surface coordination fractions, defined as the fractions of the total number of nearest neighbors of an atom in its own layer and that in the adjoining layer. Therefore, p + q = 1. For a close-packed structure, the values of these parameters are usually taken as ½ and ¼, respectively [26].

2.3.3. Viscosity

There are many empirical and semi-empirical viscosity models [31]. Taking into account the data on the enthalpy of mixing of Ir-Si melts [25], the Moelwyn–Hughes (MH) model [27] is the most appropriate to predict the viscosity of liquid Ir-Si alloys. The MH viscosity isotherm is described by

$$\eta =\left({\mathrm{x}}_1{\eta }_1+{\mathrm{x}}_2{\eta }_2\right)\left(1-2{\mathrm{x}}_1{\mathrm{x}}_2{\displaystyle \frac{\Omega }{\mathrm{R}\mathrm{T}}}\right)$$

(5)

where ${\eta }_1$ and ${\eta }_2$ are the viscosities of the pure components and Ω is the mixing enthalpy parameter of A-B alloys.

3. Results

3.1. Wettability in the Liquid Si-Ir Eutectics/C-Base Material System

The contact angle behavior of a liquid Si-Ir eutectic alloy in contact with C-CVD, over time and at two testing temperatures of T = 1350 °C and 1400 °C, are shown in Figure 4. For these testing temperatures, after approximately 345 and 185 s, respectively, the steady state conditions were achieved, resulting in contact angle values of θ ≈ 11–14°, measured at the triple lines. At the highest temperature, the equilibrium contact angle value of θ = 11° was kept constant for 30 s. After t = 225 s from the start of the isothermal condition, a complete spreading (i.e., θ ≈ 0°) of the liquid Si-Ir eutectic alloy on C-CVD substrate occurs, as shown in Figure 4a,b, where the time sequence of the images is also displayed.

Figure 4. Liquid Si-Ir eutectics in contact with C-CVD: (a) wetting tests at T = 1350 °C (◇) and T = 1400 °C (◆) under an Ar atmosphere; (b) time sequence images of alloy samples spreading on C-CVD substrate at two testing temperatures.

On the contrary, at T = 1350 °C, the equilibrium contact angle value of θ ≈ 14° was kept constant until the end of the experiment (t = 400 s). Additionally, the threshold contact angle value of θ = 90°, determining the transition from non-wetting to wetting behavior, was reached at the liquid Si-Ir eutectics/C-CVD, with a double wetting rate by increasing the temperature from 1350 to 1400 °C.

BSE/EDS analyses performed at the top drops and triple lines of the Si-Ir/C-CVD samples processed at T = 1350 °C and T = 1400 °C are shown in Figure 5. In both cases, at the top drops (Figure 5a,b), three phases were detected. Specifically, at the surface of the sample processed at T = 1350 °C, two phases, Si + IrSi₃, over the grains of pure Si, were found (Figure 5a). Additionally, the IrSi₃ phase was detected as a rejected phase at the Si + IrSi₃ phase contour. The same phases were found after the wetting test performed at T = 1400 °C (Figure 5b). At a higher temperature, the elongated grains of the IrSi₃ phase were randomly distributed over the surface.

Figure 5. BSE/EDS analyses performed on the Si-Ir/C-CVD samples at different magnifications after the wetting tests at T = 1350 °C and T = 1400 °C: (a,b) top of the solidified drops; (c,d) triple lines; and (e) images of the two samples after the wetting tests.

At the triple lines of both Si-Ir eutectics/C-CVD samples (Figure 5c,d), a final microstructure consisting of Si + IrSi₃ and IrSi₃ phases was detected. In particular, after the wetting test performed at T = 1350 °C, a fracture was found at the “apparent” triple line, which most probably occurred during the fast cooling of the sample. Indeed, a thin layer of Si-Ir alloy spread for ≈200 μm even beyond the line (see Figure 5c,e) was found, resulting from Si-Ir/SiC dewetting forced by the solidification process. As already mentioned, at T = 1400 °C, a complete spreading of the alloy drop on the C-CVD substrate was observed (Figure 5e). At the triple line of the solidified sample, a microstructure with elongated grown IrSi₃ crystals was detected laying on a SiC thin layer. Such a microstructure is easily visible by eye at the surfaces of the two samples after wetting tests (Figure 5e).

As can be seen, a two-phase microstructure consisting of Si + IrSi₃ is also revealed at the Si-Ir eutectic/C-CVD triple line formed at T = 1400 °C, together with a phase consisting of Si + IrSi₃ with dispersed SiC particles (Figure 5d).

Figure 6 shows BSE/EDS analyses carried out at the cross-sectioned Si-Ir eutectics/C-CVD samples after the wetting tests performed at T = 1350 °C and T = 1400 °C.

Figure 6. BSE/EDS analyses performed at three different magnifications at the triple lines and in the middle of the cross-sectioned Si-Ir eutectics/C-CVD samples after the wetting tests at (a,b) T = 1350 °C and (c–f) T = 1400 °C under an Ar atmosphere.

At first glance, at both the triple lines (Figure 6a,d), the alloy phase exhibits a concave contour, and the complete drop detachment from the substrate was observed. In addition, between the solidified Si-Ir alloy and the C-CVD substrate, an intermediate thick layer (≈50 μm) below both alloy drops was detected (Figure 6b,e) with an alternating change in their microstructures consisting of SiC and Si-Ir silicides (Figure 6c,f).

As introduced before, aiming at studying the interfacial phenomena occurring between the liquid Si-Ir eutectic alloy and C-based materials as a function of C reactivity, a wetting test was performed on GC at T = 1400 °C under the same working atmosphere and applying the same experimental procedure.

During the wetting experiment, the contact angle behavior of the liquid Si-Ir eutectics/GC couple as a function of time was recorded. The more relevant images related to the evolution of the spreading process are inserted into Figure 7.

Figure 7. Contact angle behavior as a function of time observed at the triple line of liquid Si-Ir eutectics/GC and relevant images recorded during the wetting test performed at T = 1400 °C under an Ar atmosphere.

Three different wetting stages are well-distinguished. Specifically, the contact angle decreased linearly down to a value of 40° in 100 s and, similarly to the liquid Si-Ir eutectics/C-CVD system, the threshold value discriminating between no wetting and wetting conditions was achieved in 50 s. After 100 s, a strong decrease in the wetting rate was observed. Indeed, the contact angle value further decreased from 40° to 32° in 50 s and, following this, an asymptotic behavior (dashed line) was achieved and retained for the next 70 s. As evidenced by the images inserted, during the spreading, the alloy drop drifted close to one of the edges of the GC substrate, and immediately after, a part of the material was pulled out from the plate. For this reason, the sample was quickly cooled down to room temperature.

BSE/EDS analyses performed at both the top drop, i.e., near the triple line, and at the cross-sectioned Si-Ir eutectic alloy/GC sample processed at T = 1400 °C, are shown in Figure 8. As can be seen, due to the mismatch in the CTE values, a fracture at the interface was detected (Figure 8a). Additionally, at the surface close to the triple line, two phases consisting of pure Si and Si + IrSi₃ (bright) were observed. Moreover, beyond the drop perimeter, a thin film of SiC over the layered circular area of the Si + IrSi₃ mixed phase, probably formed during the solidification process and consequent alloy/substrate dewetting, was revealed.

Figure 8. BSE/EDS analyses performed at different magnifications at (a) the top drop near the triple line and at (b) cross-section of the Si-Ir eutectics/GC couple after the wetting test carried out at T = 1400 °C under an Ar atmosphere.

At the cross-section of the sample, close to the Si-Ir alloy drop, several SiC crystals were detected (Figure 8b). At higher magnification, it was possible to identify the starting interface with C-dissolution pockets formed on the substrate side containing the IrSi₃ + SiC mixed phase. On the contrary, SiC crystals with sizes ranging between 0.5 and 7 μm were grown on the alloy side.

3.2. Wettability in the Liquid Si-Ir Eutectics/SiC System

The contact angle behaviors at the Si-Ir/HP-SiC triple lines, observed over time during the wetting tests performed by the contact heating sessile drop method at T = 1350 °C and T = 1400 °C under a static Ar atmosphere, are shown in Figure 9.

Figure 9. Evolution of contact angle behavior at the liquid Si-Ir eutectics/HP-SiC over time, observed during the wetting tests performed under an Ar atmosphere, at (☐) T = 1350 °C and (■) T = 1400 °C.

Despite the testing temperature differing by only 50 °C, the wetting behavior is evidenced as strongly dependent on the temperature. In particular, at T = 1350 °C, through three well-distinguished step-wise trends, after 700 s from the start of the isothermal conditions, the final contact angle value of θ ≈ 25° was achieved.

However, at the Si-Ir eutectics/SiC triple line processed at T = 1400 °C, the contact angle rapidly decreased to the value of θ ≈ 25° in 30 s, and this value remained constant until the end of the wetting experiment.

BSE/EDS analyses performed at both the top drops and at the cross-section of the Si-Ir eutectics/SiC couples are shown in Figure 10. In particular, the dewetting phenomenon was observed on both alloy drops (Figure 10a,b), most probably caused by the shrinkage of the alloy drops during the solidification. For both samples, a dewetted circular area with a radius of 200 μm was observed, and residual alloy material consisting of the IrSi₃ phase (bright) was retained on the SiC surface.

Figure 10. BSE/EDS analyses performed at different magnifications on Si-Ir eutectics/HP-SiC couples after wetting tests: top drop near the triple line, delimited by the dashed line at (a) T = 1350 °C and (b) T = 1400 °C; (c,d) show the cross-section at T = 1350 °C.

At the cross-sectioned sample processed at T = 1350 °C (Figure 10c,d), there was no evidence of reaction phenomena (Figure 10d) except for the presence of few SiC crystals grown at the interface. Inside the bulk alloy, pure Si and Si + IrSi₃ mixed phases were found (Figure 10c). On the other hand, nearby the Si-Ir alloy/SiC interface, the eutectic microstructure was also found (Figure 10d).

The wettability of a SiC single crystal (hereafter referred to as 6H-SiC) by the liquid Si-Ir eutectic alloy was also examined, as shown in Figure 11a–c. The wetting experiment was carried out under quasi-dynamic conditions by multi-stage increasing of the temperature, and the corresponding contact angle values were measured and analyzed post-test as functions of time and temperature, as shown in Figure 11b. As can be seen, the melting of the Si-Ir alloy sample was observed at T = 1251 °C, i.e., 29 °C above the eutectic melting point temperature of T = 1222 °C [5]. Increasing the temperature up to 1400 °C, the contact angle decreased down to θ ≈ 30° in 150 s through different “wetting stages”. In fact, following the time/temperature sequence images (Figure 11a), the alloy melting evolved progressively in the temperature range of 1251–1300 °C, and the spherical drop shape (complete melting of the Si-Ir eutectic alloy) was achieved at T > 1300 °C. Increasing the temperature from T = 1400 °C to T = 1450 °C, a further decrease in contact angle of 7–10° was observed and, under isothermal conditions, the value of θ ≈ 20° was kept constant for 30 s. During the last part of the experiment, the complete spreading (θ ≈ 6°) of the alloy drop on the SiC single crystal was achieved. Concomitantly, the appearance of a solid phase at the top of the Ir-Si eutectics/6H-SiC sample was observed, and two different microstructures were also evident by visual inspection. Specifically, a continuous layer in the central part of the drop, and a fragmented circular area at the contour, were observed. Moreover, on the sample holder (a porous graphite disk), a “shadow” was present. Microstructural characterization of the Si-Ir eutectic alloy/6H-SiC sample performed by BSE/EDS confirms the presence of a quasi-continuous microstructure at the top, and the resulting alloy was completely covered by SiC crystals with sizes of 2–20 μm (Figure 12).

Figure 11. Wettability of liquid Si-Ir eutectics in contact with 6H-SiC (☐) observed from T = 1251 °C to T = 1450 °C, under an Ar atmosphere for 15 min: (a) time sequence of alloy spreading; (b) contact angle behavior as functions of time and temperature; and (c) Si-Ir eutectics/6H-SiC sample after the wetting experiment.

Figure 12. BSE/EDS analyses performed at different magnifications on the Si-Ir eutectics/6H-SiC sample obtained after the wetting test at T = 1251–1450 °C under an Ar atmosphere: (a,b) top view at the triple line; (c) Si-Ir alloy/6H-SiC interface; and (d) Si-Ir alloy/6H-SiC/C-porous system.A circular area with a radius of 500 μm delimiting the alloy drop (a) shows the eutectic microstructure (b), while the substrate surrounded by metal material exhibits platelet fragmentation. The 6H-SiC platelets are also evident at the interface of the cross-sectioned sample (c). A part of the alloy material was found under the substrate and even infiltrated into the underlying C-porous plate (sample holder), as shown in (d).

3.3. Thermodynamics and Thermophysical Properties of Si-Ir Melts

Material design and the microstructure tailoring of metal/ceramic composites by infiltration processes mainly include three steps, i.e., the preform preparation, the permeation of a melt into the perform, and the solidification of liquid metal throughout the preform. The last two steps are closely related to the wetting and thermophysical properties of the liquid phase. Theoretical analysis and modeling of infiltration processes take into account Young’s and Darcy’s laws (the last one is derived from the Navier–Stokes equations), and therefore the contact angle, surface tension, and viscosity and density values are requested as input data. To the best of our knowledge, the thermophysical property data of liquid Ir-Si alloys are not yet experimentally determined, and thus, in the present work, only the model predicted values are reported. Taking into account that Ir and Co belong to Group 9 of the chemical elements in the Periodic table, it is expected that the mixing behavior in Co-Si and Ir-Si melts will be affected by some similarities in the physico-chemical properties of alloy constituents. Accordingly, considering our previous studies on the thermophysical and wetting properties of liquid Co-Si alloys [16,19], a comparison of the model predicted property values of the two systems was performed. To this aim, the aforementioned properties of liquid Ir-Si and Co-Si were calculated for T = 1400 °C.

Based on Miedema’s semi-empirical model [24,32] and its refinement [25], the enthalpy of mixing $∆H^{mix}$ of liquid Ir-Si alloys (Figure 13; curve 1) can be calculated by

$$∆H_{A-B}^{mix} (\mathrm{k}\mathrm{J}·{\mathrm{m}\mathrm{o}\mathrm{l}}^{-1})=4c_A{c}_B\left[-26.2-0.8(c_A-c_B)\right]$$

(6)

Figure 13. Concentration dependence of $H_M/RT$ (curve 1), $G_M/RT$ (curve 2), $a_{Ir}$ (curve 3a) and a_Si (curve 3b) for liquid Ir-Si alloys calculated by the QCA for T = 1400 °C. Dashed lines represent the ideal mixture.

Adding the ideal mixture term to Equation (6) and assuming $∆S_{A-B}^{mix}$ = 0, the Gibbs free energy of mixing of Ir-Si melts was calculated. Its normalized form, G_M/RT, indicates the type of interactions between the constituent atoms, and the value of -2.98 is characteristic for strongly interacting systems (Figure 13; curve 2), such as Co-Si [16], Al-Ni [33], Au-Sn [34], and others. The calculated values of Ir and Si activities exhibit strong negative deviation from ideality, comparable to that observed for liquid Co-Si alloys [16].

From a theoretical point of view, the density affects the atomic structure and short- range ordering of metallic melts. For all technological processes that include the presence of the liquid phase, the density is implicitly contained in a Rayleigh number that characterizes fluid flow. Therefore, the density or molar volume of alloy melts is an input necessary for material design. The molar volume of liquid Ir-Si and Co-Si was calculated for T = 1400 °C by using Equation (2). A lower density of liquid Ir in comparison to Co implies a higher molar volume of Ir-Si alloys (Figure 14, curve 1) with respect to that of Co-Si (Figure 14, curve 2).

Figure 14. Concentration dependence of the molar volume of liquid Ir-Si (curve 1) and Co-Si (curve 2) alloys calculated for T = 1400 °C. (------) the ideal solution model.

From our previous work on the thermophysical properties of liquid Co-Si alloys [16], their molar volume $V_M$ has been calculated from the density data obtained at T = 1550 °C and 1800 °C over the whole composition range [35]. Negative deviation from the ideal behavior was observed [16], and the same trend is expected for liquid Ir-Si alloys. Due to the lack of experimental data, a comparison with model predicted values is not possible. Among the thermophysical properties, the surface energy (surface tension and interfacial tension) and viscosity of metallic melts are key parameters for solidification simulations as a part of casting process design. The surface tension of liquid Ir-Si alloys was calculated by the QCA for a regular solution [26].

To this aim, the Gibbs free energy of the liquid phase obtained using the thermodynamic data [25] predicted by Miedema’s model [24,32], together with the molar volume, structural data [36], and the surface tension reference data of liquid Ir [37] and Si [38] alloy components, were used as input. The surface tension isotherms of Ir-Si melts calculated for T = 1400 °C are shown in Figure 15. The surface tension values obtained by QCA (Figure 15, curve 1a) are significantly higher with respect to those indicating the ideal behavior (Figure 15, curve 1b), as was already observed in the case of liquid Co-Si alloys [16]. For comparison, the Co-Si surface tension isotherms obtained for the same temperature are shown (Figure 15, curves 2a and 2b), indicating similar behavior, and moreover, in Si-rich parts of the two alloy systems (up to 50% Si), the corresponding curves overlap.

Figure 15. Surface tension of liquid Ir-Si alloys at T = 1400 °C (curve 1a—QCA; curve 1b—ideal behavior). For comparison, the Co-Si isotherms (curve 2a—QCA; curve 2b—ideal behavior) are shown.

The viscosity of liquid Ir-Si alloys was predicted using Moelwyn-Hughes’s model [27] for T = 1400 °C. The enthalpy of mixing [25] and the viscosity reference data of liquid Ir [37] and Si [39] were taken as input in the present calculations. The Ir-Si viscosity isotherm was calculated for T = 1400 °C and compared with that of Co-Si melts (Figure 16). The last one was calculated using the thermodynamic data [40], and the viscosity reference data of liquid Co [41] and Si [39], together with the Co-Si experimental datasets [42]. As already shown for the above-mentioned thermophysical properties (Figure 13, Figure 14 and Figure 15), the viscosity of two alloy systems, including the experimental data of liquid Co-Si alloys [42], exhibit similar behavior, and the main differences arise from the different viscosity data of liquid Ir [37] and Co [41].

Figure 16. The viscosity isotherm (curve 1) of liquid Ir-Si alloys calculated by Moelwyn-Hughes’s model for T = 1400 °C. For comparison, the Co-Si viscosity isotherm (curve 2) and datasets (●) [42] are shown.

4. Discussion

The present investigation was performed for better understanding of the interfacial phenomena observed during the fabrication of SiC/IrSi₃ composites via reactive infiltration [7,8] in the temperature range of T = 1250–1350 °C. Aiming at considering the local overheating phenomenon caused by the exothermic reaction between liquid Si and solid C, the wetting studies of the present work were performed in the range of T = 1350–1400 °C. In fact, the chemical reaction did not proceed entirely to completion (a so-called “limited reaction”), and in the SiC matrix around the SiC particles, residual C was found (4.7 wt%) [7], indicating that such preforms may be previously thermally treated for improving the graphitization of the residual C.

It is well known that liquid Si and Si-based alloys readily react with C, and at the interface, a SiC reaction product is formed [43]. Therefore, it is not surprising that the interaction phenomena, in terms of the wettability and spreading kinetics observed in the Si-Ir eutectics/C-CVD system, are strongly temperature-dependent. In fact, as observed in Figure 4, at the highest testing temperature (T = 1400 °C), the same steady-state contact angle value of θ = 11–14° was achieved faster than the experiment conducted at T = 1350 °C. Moreover, the drop rapidly “disappeared” in around 30 s.

At the same temperature (T = 1400 °C), different behavior was observed at the Si-Ir eutectics/GC triple line, as shown in Figure 17, in agreement with the results [44] for pure Si; in 0 < t < 50 s, θ_GC > θ_C-CVD. It can be explained by the different C crystallinity and by the lower reactivity of C in its amorphous state. On the other hand, the transition between no-wetting to the wetting phenomenon was observed on both the substrates at the same time (t ≈ 50 s). At the interface, a thin layer of SiC, promoting further advancing of the triple line, was already detected [45]. At the starting condition, the different surface roughness of the two substrates, specifically 20 nm for GC and 9.18 μm for C-CVD, may be counted as an affecting factor of the spreading kinetics within the no-wetting stage (after 50 s) [45]. In fact, the surface of GC already covered by the thin layer of SiC resulted in being more wettable than C-CVD by the liquid Si-Ir alloy.

Figure 17. Contact angle behavior of the liquid Si-Ir eutectic alloy in contact with (◇) C-CVD and (●) GC as a function of time observed at T = 1400 °C.

After 100 s, at the liquid Si-Ir eutectics/GC triple line, a decrease in the wetting rate was observed. The decrease in the slope is caused by the epitaxial growth of SiC crystals at the interface and their progressive packaging, as documented for other Si-based alloys/GC systems [19,20,21,28,43,44]. Conversely, at the Si-Ir eutectics/C-CVD, a continuous decrease in the contact angle value down to the sudden disappearance of the molten phase was observed. It is obviously due to the collapsing alloy drop during its infiltration/percolation into the C-CVD substrate, which is promoted by the reaction of the molten phase with the graphitic layers.

As introduced, the microstructural characterization performed at the cross-sectioned Si-Ir eutectics/C-CVD samples after the wetting tests confirmed such a multilayered microstructure (Figure 18a). As shown in Figure 18b, a metal phase consisting of Ir-silicides, namely IrSi₃ and Ir₃Si₄, inserted between thin layers of SiC, was identified by XRD analysis. The appearance of the Ir₃Si₄ phase, as a consequence of an ongoing reaction between enriched Ir-silicides with non-reacted underlying C during the percolation process, may be identified as the limiting phenomenon/factor affecting Si-Ir/C-CVD, mainly at the lowest testing temperature (T = 1350 °C). Indeed, the solidification temperature of the Ir₃Si₄ phase starts at T = 1408 °C [5]. Consequently, as soon as the Ir₃Si₄ was formed at the interface, the percolation of molten material was not further permitted, as it was in the case of the sample processed at higher temperature.

Figure 18. Cross-sectioned sample of Si-Ir eutectic alloy/C-CVD after the wetting test performed at T = 1400 °C: (a) HR-BSE image and (b) XRD pattern.

The present experimental observations are in very good agreement with the results reported on wettability investigated for other Si-based alloys, such as Si-Co [19] and Au-Si [46,47]. In particular, at T = 1450 °C (≈200 °C of overheating with respect to the alloy melting temperature [48]), the liquid Si–22.5 at% Co alloy in contact with GC underwent quasi-complete spreading (θ < 15°) in 800 s, as observed by [19]. In addition, the complete spreading (θ ≈ 0°) in 230 s was achieved when, under the same operating conditions, the same alloy composition was in contact with highly oriented pyrolytic graphite (HOPG). Similarly to what was observed at the cross-sectioned Si-Ir eutectics/C-CVD sample processed at T = Tm + 180 °C (e.g., T = 1400 °C), a thicker reaction layer was developed at the interface consisting of SiC layers alternating with unreacted Si-Co alloy and C. The decrease in surface tension with temperature [40,41,42], combined with a decrease in the viscosity of the melt [48], are probably the factors making the apparent equilibrium at 36–40° observed at lower testing temperatures unstable. Conversely, a set of more favorable inherent thermophysical properties, mainly viscosity and molar volume, allowed the drop stability even at high temperatures for other Si-based alloys, such as Si-Zr and Si-Ti eutectic alloys in contact with GC and SiC, as documented by [19,20,21,49]. On the other hand, it should be pointed out that the final stage of the wetting experiment performed on the liquid Si-Ir eutectics in contact with GC at T = 1400 °C was affected by the contact that occurred between the molten material and the border of the substrate. Indeed, as shown in Figure 19, the anomalous trend observed at the Si-Ir eutectic alloy/GC triple line, in comparison with the behavior observed under the same operating conditions when the same alloy is in contact with HP-SiC, is evident.

Figure 19. Contact angle behavior as a function of time observed at T = 1400 °C for the liquid Si-Ir eutectic alloy in contact with (☐) HP-SiC and (●) GC.

From previous experiences [19,20,21,28,43,44,45,46,47], the interaction phenomena occurring at the interface between liquid Si/Si-based alloys in contact with GC are mostly controlled by the appearance of SiC as a reaction product, according to the following exothermic reaction reported in Equation (7):

Si_(l) + C_(s) → SiC_(s)

(7)

Moreover, when the steady state conditions are achieved at the interface, the value of the final contact angle is almost comparable with the value measured at the Si-based alloy/SiC triple line. It lets us conclude that the first contact with the border of the GC substrate presumably occurred in 220 s from the start of the isothermal conditions, and θ ≈ 32° may be considered as the final contact angle value for the Si-Ir eutectics/GC system at T = 1400 °C. In addition, it results 6° higher than the contact angle measured under the same operating conditions at the Si-Ir eutectics/HP-SiC triple line.

Although the contact angle values are almost comparable, the slight difference may be caused by the rougher interface at the alloy/GC couple caused by the growth of SiC crystals, as shown in Figure 8. Additionally, the evident dissolution of a large amount of C at the interface, resulting in the presence of the typical C-dissolution pockets, should be highlighted, mainly taking into account the short time of contact between the alloy and the substrate, in comparison with other similar investigations carried out by the authors [19,20,21,23].

Such a phenomenon is worth being thermodynamically investigated, mainly by analyzing the effect of Ir as an alloying element on C solubility, dissolution and reactivity in the liquid state, which may be a subject for another paper. However, the absence of concomitant reactions between C and Ir was already documented in [13,14].

As mentioned in the previous section, the wetting kinetics of liquid Si-Ir eutectics in contact with HP-SiC are temperature-dependent (Figure 9). Reaching the equilibrium conditions at the interface, the temperature strongly affects both the wettability rate and the mechanism of interfacial interaction. Indeed, a decrease in the contact angle value evolved by different steps at T = 1350 °C.

As documented in [19,20,21,28,43,44,45,46,47], two phenomena are already experimentally proven when liquid Si and Si-based alloys are processed at high temperatures, namely the unavoidable presence of SiO₂ primary oxide, both at the alloy drop surface and at the surface of the SiC substrate, and the evaporation/condensation of Si droplets beyond the triple line.

The native oxide should be decomposed by its reaction with the underlying liquid Si, forming the volatile Si monoxide [43]:

SiO_2(s) + Si_(l) → 2SiO_(g) ↑

(8)

and at the alloy/SiC interface by the reaction (9)

2SiO_2(s) + SiC_(s) → CO_(g) ↑ + 3SiO_(g) ↑

(9)

When at the interface, the layer of native oxide completely disappears, and liquid Si starts to interact with the substrate. On the other hand, the presence of SiO₂ as a layer covering the alloy drop prevents the evaporation of Si.

As reported in [19,20,21], the evaporation/condensation of Si beyond the triple line favors the advancing contact angle on both the C-based and SiC substrates. The condensation of Si onto the GC substrate close to the triple line allows the appearance of a thin layer of SiC (Equation (7)), and the advancing of the triple line proceeds by the presence of a formed well-wettable substrate (SiC). On the contrary, the Si condensation over the SiC surface promotes the SiO₂ layer decomposition (Equation (8)) and consequently the wettability by liquid Si/Si-based alloys.

Under an inert atmosphere, liquid Si typically condenses in a circular area close to the alloy perimeter, with a radius depending on the process parameters. Liquid Si may condense up to a distance of 20–30 μm from the triple line, as reported for similar investigations performed on Si-Co/GC [19], Si-Ti/GC [20] and Si-Zr/GC [21] systems. In the present case, the circular area of condensed Si onto the HP-SiC substrate progressively favored the disappearance of the SiO₂ layer. As a consequence, new interfaces of Si-Ir alloy/HP-SiC were gradually formed. However, the continuous advancing of contact angle may be impeded (i.e., “pinning” of the triple line) by the presence of grain boundaries, typical of sintered refractory materials.

For this reason, the wettability of liquid Si-Ir eutectics in contact with 6H-SiC (i.e., the single crystal selected as a reference material) was investigated under the same operating conditions, and considering Van der Waals forces as the only existing interaction process at the interface. As usual, the experiment was performed by following the change in contact angle values as functions of time and temperature.

As can be seen in Figure 11a, the first evidence of alloy melting was observed 30 °C higher than the melting temperature reported in [5]. In addition, the complete melting was achieved at T = 1300 °C. As widely discussed above and in agreement with [19,20,21,43], the delay in the complete melting can be caused by the presence of SiO₂ as a native oxide at the alloy drop surface and even at the Si-Ir eutectics/SiC interface. Indeed, by analyzing the wetting kinetics, three different wetting “stages” can be recognized. Within the first segment, the SiO₂ layer at the interface is probably decomposed by the reactions detailed in Equations (8) and (9). By increasing the temperature, a decrease in the slope of the wetting curve is observed, and it may be due to the oxide-free alloy interacting with the SiC surface itself, progressively free from the SiO₂ layer. By futher heating, a decrease in the wetting kinetics was observed. It can be followed by a stepwise advancing of the contact angle from θ ≈ 36° to θ ≈ 24°, measured at T = 1350 °C and T = 1450 °C, respectively.

Under conditions far from equilibrium, and taking into account different SiC crystallinities, such contact angles are comparable with the values measured at the Si-Ir eutectics/HP-SiC triple lines within the wetting tests performed at T = 1350 °C and T = 1400 °C. In addition, as observed in the Si-Ir eutectics/HP-SiC system processed at T = 1350 °C, the stepwise behavior can be caused by the progressive oxide decomposition by the reaction detailed in Equation (9), promoted by the evaporation/condensation phenomenon of Si beyond the triple line.

At T = 1450 °C, as already introduced in the previous section, a sudden decrease in the contact angle value was observed.

As confirmed by the BSD/EDS analysis performed at the cross-sectioned sample, part of the material infiltrated into the C-porous disk located beneath the 6H-SiC plate, as shown in Figure 12. In particular, alloy material was found under the 6H-SiC plate, and SiC crystals were detected at the interface of the C-pores combined with an IrSi₃ phase. SiC crystals were even found at the top of the alloy surface and close to the triple line, as a result of the reaction between the alloy and C-disk. Most probably, the SiC crystals were dragged up because of the dewetting phenomenon driven by the solidification. Additionally, at the surface of the 6H-SiC substrate, several cracks generated by the fast cooling of the sample at the end of the experiment were detected (Figure 12b,c).

Crack formation in ceramic materials or ceramic-based composites can be due to two factors. The first is the cooling rate, which is controlled by the thermal shock parameter S. As described in [50], the S parameter is linearly dependent on the thermal conductivity and critical strength, as well as inversely dependent on the coefficient of thermal expantion and Young’s modulus. In particular, the critical strength can be derived from Griffith’s equation, as reported in [51]. The other factor is the mismatch in the thermal conductivity, which generates stress at the interface that may lead to phase separation. This mismatch is primarily governed by the product of ΔαΔT [52]. To calculate the stress produced at the interface, Lee et al. [53] derived the following expression for an ideal single-fiber situation in an infinite matrix:

$$\psi ={\displaystyle \frac{4{\mathrm{G}}_{\mathrm{m}}\frac{(1+{\epsilon }_{\mathrm{m}})}{3(1-{\epsilon }_{\mathrm{m})}}\frac{{\mathrm{K}}_{\mathrm{r}}}{{\mathrm{K}}_{\mathrm{m}}}\Delta \alpha \Delta \mathrm{T}}{1+\frac{\left(1+{\epsilon }_{\mathrm{m}}\right)}{\left(1-{\epsilon }_{\mathrm{m}}\right)}(\frac{{\mathrm{K}}_{\mathrm{r}}}{{\mathrm{K}}_{\mathrm{m}}}-1)}}$$

(10)

where the subscripts m and r denote the matrix (eutectic alloy) and reinforcement (SiC), G is the shear modulus, K is the bulk modulus, and ε is Poisson’s ratio.

The data for the calculations were taken from [8,53]. The shear stress was found to be 500 MPa, which is quite high, but similar to the results reported by [53]. In addition, considering the results reported in [14], where large amounts of Ir silicides were produced as reaction products within the joining process of SiC monoliths by interposed Ir-sheets, similar microcracks were experimentally observed, even by measuring a value of bond strength less than 200 MPa. Moreover, cracks grow intergranularly, mainly at the SiC/IrSi_x interfaces, in agreement with the results presented here.

5. Conclusions

To better understand the mechanism of infiltration of liquid Si-Ir alloys into SiC-based porous materials, a parallel investigation on the interactions occurring both at Si-Ir alloy/SiC-based and Si-Ir/C-based material interfaces was performed by studying the wettability and analyzing spreading kinetics as a function of the operating conditions. The Si-Ir eutectics/6H-SiC system was selected as a case study to establish the factors affecting spreading kinetics. Such results were compared with the experimental findings obtained on Si-Ir eutectics/HP-SiC. The liquid Si-Ir alloy quickly spread on both the substrates, achieving a complete wetting on the SiC-single crystal substrate, as expected. To evaluate the influence of C reactivity (C-CVD vs. C-Glassy), reference wetting measurements in the liquid Si-Ir eutectic alloy/C-CVD system were also performed at two different temperatures. Wetting behavior strongly depends on the temperature. If the Si-Ir eutectic alloy/C-CVD and Si-Ir eutectic alloy/GC triple line behavior are compared in terms of spreading kinetics, the transition between no-wetting to the wetting phenomenon was observed on both the substrates at the same time in very good agreement with the literature. Furthermore, evidence on how surface roughness may affect the spreading kinetics is provided. Finally, it is clear that liquid Si-Ir eutectics/HP-SiC triple line behavior differs from Si-Ir eutectics/GC—the wetting mechanism by Van der Waals and by reaction phenomena are the two driving phenomena, respectively.

Finally, a set of thermophysical properties are provided here. Namely, surface tension, surface segregation, molar volume and viscosity were calculated by applying thermodynamic models. Although calculated, such data may be helpful for optimizing liquid-assisted processes, from casting to liquid metal infiltration.