Six high-entropy alloys of theorized chemical compositions were used for this research: WTaNbTiAlCr, WTaNbTiAlCrMo, WNbAlTiCr,. A high entropy refractory alloy (RHEA) is defined as a material composed of at least 5 metallic elements mixed in equal molar proportions [9]. For this reason, several high-entropy refractory alloys will be produced and their resistance to oxidation will be studied.
Oxidation of a high entropy alloy is observed when the material is kept at a very high temperature for a long time, at least 800 °C. The oxidation resistance of RHEA is due to the large number of atoms in the unit cell. This is the reason why a refractory high entropy single phase material is desirable.
Synthesizing these high-entropy refractory alloys can be very difficult due to the high melting points of the metallic elements. High-entropy refractory alloys are still a new research topic, so the literature has yet to look deeply into all possible compounds[4].
Synthesis
Two samples had to be taken at the same time to ensure a balance of the ball mill. Tungsten Carbide Jars with Tungsten Carbide Grinding Balls of various sizes grind and mix the powder to ensure uniform particle size and distribution. The PQ-N2 planetary ball mill (Across International, USA) was set to grind at 200 rpm for 2 h and would change direction every 30 min.
Spark plasma sintering dies were lined internally with carbon paper, and then two wafers were placed on each die. The carbon paper ensured that no sticking occurred between the metal powder and the graphite matrix. Each powder was sintered using a Spark plasma sintering furnace (FCT Systeme GmbH, Germany) set to press at 27 kN, speed 50 °C/min, total time 30 min, and nominal sintering temperature 1350 °C.
Sample Preparation
Analysis Methods
Oxidation
The typical materials used in the “hot side” of the turbine are the Ni-based superalloys[10]. In high-entropy refractory alloys, the oxygen will bind to various elements, causing significant phase changes and compromising the structural integrity of the material[8]. Another factor to which the slow oxidation of the material can be attributed is the lattice.
The changing atomic radii at the corners of the unit cells will cause a non-symmetrical unit cell. The molds were allowed to dry and then the cross-sectional surface of the samples was mirror polished. SEM and EDS images were then captured for each cross-sectional sample, allowing the oxide layer thickness to be accurately measured.
Surface calculations of the small specimen were done in Solidworks using the same procedure as before. The mass of the sample was initially recorded by the TGA instrument and then updated every 5.04 seconds. As seen in Figure 8, sample A shows excellent dispersion of elements throughout the mass of the material.
There is some titanium and aluminum oxide present in the material, but these are unavoidable by-products of the powder metallurgy method. The BSE image, as in Figure 11(b), most prominently shows a whitish area at the bottom right, extending upward toward the center of the sample. The EDS map of sample B shows that the whitish area in the BSE image is in fact a tungsten-rich area of the material.
After SE and BSE imaging of the surface, seen in Figure 13, sample D was in fact multiphase. As seen in Figure 20, the peaks of the oxide layer became more prominent as time progressed, but there were no shifts. The oxide layer appeared almost as large growths that did not form evenly on the surface of the sample.
The SEM images of sample E, shown in Figure 30, show complete oxidation of the material surface just like sample C. EDS maps obtained from different oxidized samples of sample E show a very detailed structure on the surface of the samples as seen in figures 31, 32 and 33.
XRD of the as-Fabricated RHEAs
SEM/EDS of the as-Fabricated RHEAs
The backscattered image in Figure 7(a) shows the presence of a small phase change, but as further analyzed with EDS this image will correlate with an oxide. The EDS image of sample B, as seen in Figure 10, shows the multiple phase regions seen in the backscattered image of sample B in Figure 9. Aluminum shows dark, Al-rich regions, but these correspond to known oxides that are present in the example.
Sample C, which showed a single-phase material after XRD analysis, is verified as such by examining its SEM images in Figure 11 and the EDS map in Figure 12. To distinguish the changes in material composition from the SE images, the sample should have very clearly distinct phases, and this is shown in Figure 13(b). Secondary (b) and scattered (a) images of sample D after sintering The EDS map of sample D, as seen in Figure 14, shows a non-uniform distribution of elements.
Sample E shows excellent single-phase material as seen in the SE and BSE images of Figure 15. The sample shows some whitish areas, Figure 15(a), as does sample C, but overall the sample is single-phase. These areas were initially believed to be tungsten, as in sample C, as confirmed by the EDS map shown in Figure 16.
The SE image in Figure 15(b) shows some voids, which are most likely caused by carbide formations in the sintering, but overall, a high density is also present in this material. In Figure 17(a), sample F, when viewed using the backscatter electron detector, clearly shows 4 different phases which are characterized by a change in color. The sample has regions of microcracks where it appears that the two phases have not fully bonded to each other and are marked with arrows in Figure 17.
The EDS data in Figure 18 clearly shows how the combination of elements in sample F does not form a single-phase or even fully bonded multiphase material.
Mechanical Property Testing
Oxidation Testing Results
Initially, a piece of each of the single phase samples would be examined at lower oxidation temperatures to visually study their oxidation response. The samples from the 2, 12 and 24 hour tests were placed in separate epoxy molds with the sample placed vertically and combined with epoxy/resin to cover the top of the sample. After sintering and polishing, the samples appeared as a button shape and with a shiny surface, as can be seen in Figure 4.
The XRD results used a solid sample holder and did not require further material preparation given the small engineered sample sizes. Sample A showed very favorable results with a large peak around 37°2θ as shown in Fig. 5(a) , which was found to have each element present and in phase equilibrium with each other, indicating a BCC structure. As HEAs are still a new classification, the PDF 4+ database does not yet contain specific cards for these materials.
As seen in Figure 5(b) , there is a distinct AlTaTi phase as marked by the blue square, which adds an additional phase to the primary BCC phase. Example D, in Figure 6(a), shows a mixture of subphases in addition to the main BCC phase. Sample E, seen in Figure 6(b) , contains the same elements at every peak with no variation except for the small titanium oxide showing.
Sample A, like every other sample to follow, was viewed using secondary electron imaging and backscattered electron imaging. SE showed the topography of the sample, which would reveal whether the sample was fully densified. A tungsten-rich area is present where the white spots were located in the material, but just as in sample C, this area is still part of the single phase that makes up the material. The cross-sectional surface, after cutting, was mirror polished, and the surfaces of the remaining three pieces of samples A, C and E were measured.
Ti, Nb, Cr, and Ta appear to have formed most of the oxide at all three oxidation times, as indicated by the three large peaks. In the 2-hour sample A, Figure 22, there is an area in the center of the BSE image where titanium appears to be dominant, however, this is part of the surface of the alloy showing through, so a better detection of the elements is present . Sample C (b) and Sample E (c) XRD scans after 2, 12 and 24 h of oxidation Sample C when viewed through SEM and EDS, showed complete oxidation of the surface at all different times as shown in Figure 26.
Cross Section Investigation
The BSE images seen in the EDS maps are the same as used in Figure 34, i.e. oxide layer and interfacial region. Since sample A was the only sample that showed oxidation resistance after each test, TGA testing was performed only on the small piece of this sample. Investigating compositional changes in refractory high-entropy alloys has proven to be an effective method in the search for a sample composition that has great oxidation resistance in air at 1000°C.
The equimolar composition of WTaNbTiAlCr proved to be an effective combination that had minimal oxide layer growth. The oxide layer took until the 12 hour mark to form a uniform shell around the material, and by 24 hours the material only saw an average oxide layer thickness of 139 μm. TGA testing of sample A also showed very favorable results for increasing the surface oxygen mass, providing results of 0.1125𝑚𝑚𝑚𝑔2 or 𝜇𝑚𝑚𝑔2.
The results of this thesis show that high entropy refractory alloys have the potential ability to resist oxidation and provide desired mechanical properties. The results seen in this thesis have shown that high entropy refractory alloys can be formed into a single phase and can provide resistance to oxidation while maintaining suitability. The powder metallurgy process used to make these high-entropy refractories can be quite expensive and time-intensive for high-volume materials, and several sintering methods should be explored.
Chopkar, “Microstructural and Mechanical Properties of AlCoCrCuFeNiSix (x = 0.3 and 0.6) High-Entropy Alloys Synthesized by Spark Plasma Sintering,” Journal of Alloys and Compounds. Liu, “A review of high-entropy basic alloys with promising high-temperature properties,” Journal of Alloys and Compounds. Wu, “Microstructure and compositional evolution of a molten silicide coating on MONBTATIW refractory high-entropy alloy in high-temperature oxidation environment,” Materials.
Hyun, “High-temperature oxidation behavior of low-entropy alloy to medium- and high-entropy alloys,” Journal of Thermal Analysis and Calorimetry.
