Microstructural Evolution and Compressive Properties of Two-Phase Nb-Fe Alloys Containing the C14 Laves Phase NbFe₂ Intermetallic Compound

Introduction

The increasing demands for high-performance aero-engines and land-based gas turbines urge the development of new materials that can withstand higher operational temperatures. However, the service temperature of the most used Ni-based superalloy has been increased up to 1,423 K, which reached about 80–90% of its melting point [1, 2]. Thus, further enhancement of the service temperature is limited, since even the most advanced single-crystalline Ni-based superalloy starts to soften beyond 1,423 K and melts at about 1,623 K [3].

Recently, eutectic alloys formed by high melting point intermetallic phases are identified as potentially important materials to be utilized as high-temperature structural materials as their phases are thermodynamically stable and problems normally associated with composites, such as interfacial reactions, can be avoided [4–7]. The AB₂ Laves phases with high melting temperatures are candidates for this application. Thus, the Laves phase-based eutectic alloys are attractive for high-temperature structural material applications owing to their large quantities, and the AB₂ Laves phases usually process very high melting temperatures.

Among the Laves phases, the NbFe₂ Laves phase is stable with a C14 (MgZn₂-type) hexagonal structure built up by Kagomé planes of Fe atoms separated by sites of mixed occupation of both Fe and Nb atoms [8]. In recent years, the NbFe₂ Laves phase attracts increasing attention as continuous efforts are underway to develop novel austenitic heat-resistant steels strengthened by NbFe₂ Laves phases, which can offer considerable strength and creep resistance at elevated temperatures [9–11]. However, detailed experimental investigations on the solidification behavior of the NbFe₂ Laves phase in the off-eutectic Fe-Nb alloys have not been reported previously, which can be used to guide the design of new structural materials with higher melting temperatures.

This investigation of two-phase Nb-Fe alloys explores the basic microstructural characteristics, such as microstructure morphology, precipitate growth and growth of the NbFe₂ Laves phase, as part of ongoing attempts to develop Laves phases as practical materials. The emphasis of this paper is to understand how the microstructure of two-phase Nb-Fe Laves phase alloys may be controlled, which is essential in developing the Laves phases as useful structures. The compressive properties of the two-phase Nb-Fe alloys are also tested with the aim of revealing the toughing mechanism.

Experimental

The alloys with nominal compositions of Fe-6%Nb and Fe-10%Nb were prepared as small buttons (ϕ = 5 mm) by melting high-purity Nb (99.99 wt%) and Fe (99.5 wt%) using a non-consumable arc melting furnace with a water-cooled copper hearth in argon gas atmosphere. Before melting each ingot, a titanium getter was melted in order to eliminate the residual oxygen and nitrogen impurities in the furnace. The button ingots were flipped and re-melted at least five times to avoid macro-segregation in the as-cast specimens. The compositions of the as-cast ingots were analyzed by the inductively coupled plasma mass spectrometer (ICP-MS). The results indicated that the compositions in different position of the ingot were close to the nominal value for the element with the deviation less than 0.51 at.% for Cr, indicating negligible losses during processing.

Since the solidification rate varied at different parts of the arc-melted ingots, the effect of solidification rate on the microstructure evolutions of Nb-Fe alloys can be analyzed qualitatively. Thus, the specimens taken for this study were all taken from the center line of each ingot, corresponding to different solidification rates, as described in our previous paper [12]. Then, three samples were cut longitudinally, and then ground, polished and etched for microstructure analysis. An optical microscopy (OM, Leica DM4000M) and scanning electron microscopy (SEM, JSM-6390A) equipped with an energy dispersive spectroscopy (EDS) were employed to identify the phases and characterize the microstructure. The quantitative image analysis was conducted by means of SISCIAS V8.0 metallographic image analysis software.

The compressive specimens with the size of ϕ 5 mm × 5 mm were cut from the arc-melted ingots by electro-discharge machining and all surfaces were mechanically ground with 800-grit SiC abrasive prior to compression test. The compression tests were conducted at room temperature with an initial strain of 1.0 × 10⁻³ s⁻¹. In our study, three to four specimens were tested and the average value was employed for the compressive strength of the alloy.

Results and discussion

The typical optical microstructures of samples taken from different positions of the as-cast Fe-6%Nb alloy are shown in Figure 1. It can be seen that microstructure consisting of fully granulate eutectic consisting of Fe solid solution and NbFe₂ Laves phases was observed at the bottom of the ingot, corresponding to the largest solidification rates, as shown in Figure 1(a). With increasing the solidification distance, primary Fe dendrites occurred in the microstructure as indicated in Figure 1(b) and 1(c). Meanwhile, the volume fraction and size of the primary Fe dendrites were observed to increase gradually. Once the solidification distance increased to the top part of the arc-melted ingot, the microstructure developed into coarsened Fe primary dendrites surrounded by eutectic two-phase regions as shown in Figure 1(d). Moreover, some particles of NbFe₂ were observed to precipitate in the Fe primary phase (Figure 1(d)–1(f)), which was caused by the decrease of solid solubility of α-Fe phase under small solidification rates.

Figure 1:

Optical images in different positions of the as-cast Fe-6%Nb alloy: (a) the bottom part of the ingot; (b) and (c) the lower part of the ingot; (d) and (e) the middle part of the ingot; (f) the top part of the ingot.

As for the as-cast Fe-10%Nb alloy, the typical optical micrographs of samples taken from different positions of the as-cast ingot are presented in Figure 2. Dendritic eutectics were observed at the bottom of the ingot as shown in Figure 2(a). With the increase of solidification distance, the volume fraction of NbFe₂ phase in the eutectic increased as shown in Figure 1(b). Then, primary Fe dendrites occurred in the microstructure as indicated in Figure 1(c). Similarity, the volume fraction and size of primary NbFe₂ dendrites increase gradually (Figure 2(d)–2(f)). However, no particles were observed to precipitate in the NbFe₂ primary phase (Figure 1(d)–1(f)).

Figure 2:

Optical images in different positions of the as-cast Fe-10%Nb alloy: (a) the bottom part of the ingot; (b) and (c) the lower part of the ingot; (d) and (e) the middle part of the ingot; (f) the top part of the ingot.

As indicated in Figures 1 and 2, the Fe primary phase and Fe₂Nb primary phase appear in the microstructures of Fe-6%Nb and Fe-10%Nb once the solidification distance achieves critical positions. In this case, the emphasis is focused on the competitive growth relation between the primary phase and eutectic.

By the calculation and comparison of the interface growth temperatures of different phases in the off-eutectic alloy and applying the maximum growth temperature criterion [13, 14], we can determine the precipitation sequence of involved phases. In such case, the Trivedi-Magnin-Kurz (TMK) [15] and Boettinger-Coriell-Trivedi (BCT) models [16], which have been proved to be the most successful models for describing the eutectic and dendrite growth, are employed to calculate the interface growth temperatures of eutectic and primary phase.

For eutectic alloy solidification, Trivedi and Magnin [15] developed a model to calculate the growth temperature of coupled eutectic in free solidification, which is given by

where α^L represents the capillarity constant, Q^L the activation energy for solute diffusion, m the equilibrium liquidus slope and P the infinite series.

Also, the dendritic growth can be described with the BCT model by the following equation [16]:

where ΔT_t represents the thermal undercooling, ΔT_c the solute undercooling, ΔT_r the curvature undercooling and ΔT_k the kinetic undercooling. ΔH is the heat of fusion, C_P the specific heat of the liquid phase, σ* the stability constant, P_t the thermal Peclet number, P_c the solute Peclet number, IvPc=PcexpPcEiPc the Ivantsov function of solute Peclet number, k the actual solute partition coefficient, C₀ the alloy composition, ξ_c the solute stability function and ξ_t the thermal stability function.

Substituting the physical parameters listed in Table 1 into the TMK model can obtain the relationships of V–λ in the as-cast Fe-10%Nb alloy as shown in Figure 3. Combining with the experimentally measured eutectic spacing, λ, the solidification rate of the as-cast ingot was determined in the range of 0.001 m/s–0.3 m/s. In addition, as affected by the shape of water-cooled copper hearth, the solidification rate of the ingot decreased gradually from the bottom to the top. In this case, the competitive growth between the primary phase and eutectic may change in different position of the as-cast ingot, which results in the formation of various kinds of microstructures.

Figure 3:

Calculated solidification rate V versus eutectic spacing λ in Fe-10%Nb.

Figure 4 presents the calculated interface growth temperatures of the eutectic and primary phase versus solidification rate in Fe-6%Nb and Fe-10%Nb alloys. By equating the growth temperatures of the primary phase and eutectic, the critical solidification rates for the precipitation of Fe and Fe₂Nb primary phases are determined as 0.16 m/s and 0.05 m/s, respectively. It can be seen that, at the bottom of the as-cast ingots, the growth temperature of the coupled eutectics is higher than that of the primary Fe and Fe₂Nb phases. Thus, fully eutectics rather than primary Fe₂Nb or Fe phase were selected in these regions as shown in Figures 1(a) and 2(a).

Figure 4:

Calculated interface growth temperature T versus solidification rate V in Fe-6%Nb (a) and Fe-10%Nb (b).

With the decrease of solidification rate, the interface growth temperatures of primary Fe and Fe₂Nb dendrites will exceed that of the coupled eutectics at certain solidification distance. Based on the maximum interface growth temperature criterion [13, 14], the Fe and Fe₂Nb primary dendrites will overgrow the eutectics and appear in the microstructures, which is in agreement with the experimental results as shown in Figures 1(c) and 2(c).

The introduction of ductile Fe phase may help to improve the brittleness of the Fe₂Nb Laves phase. Thus, the compressive properties of the as-cast Fe-6%Nb and Fe-10%Nb alloys were tested at room temperature, and the typical stress–strain curves were depicted in Figure 5. Their compressive properties are summarized in Table 2.

Figure 5:

Stress–strain curves of the Fe-6%Nb and Fe-10%Nb alloys.

As illustrated in Table 2, the elongation of Fe-Nb alloys increased remarkably and exhibited extremely high compressive strength. Enhancements in the compression properties of the Fe-Nb alloys result from second Fe phase and halo toughening mechanisms.

In the Fe-6%Nb hypoeutectic alloy, the Fe₂Nb C14 Laves phase exhibits as the strengthening phase rather than matrix phase in the Fe-10%Nb hypereutectic alloy. The introduced α-Fe phase with bcc crystal structure owns good deformation behavior and can undertake plenty of deformation, which leads to the substantial increase in the compression properties. Meanwhile, abundant Fe/Fe₂Nb eutectic interfaces were introduced in the microstructure by the existence of Fe phase, which can inhibit the crack propagation effectively, and are helpful in improving the compression properties for they are thermodynamically stable.

Further, the compression properties of the Fe-10%Nb alloy are also affected by microstructure distribution. As the primary Fe₂Nb dendrites exhibit completely dendritic morphology, blunt cracks propagate originating in the Fe₂Nb Laves phase. In this situation, the compression ratio of the sample increased remarkably even in the Fe₂Nb-based Laves phase alloy, which may offer insights into alloy modifications improving the ductility of other intermetallic compounds.

Conclusions

1. The microstructures of Fe-6%Nb hypoeutectic and Fe-10%Nb hypereutectic alloys consist of fully eutectics in the bottom of the as-cast ingots, and then develop into coarse Fe or Fe₂Nb primary dendrites plus interdendritic eutectic containing Fe₂Nb and Fe phases in the middle and top parts of the ingots.

2. The microstructure evolutions of the two alloys were successfully explained by means of the competitive growth between the primary phase and eutectic based on the maximum interface growth temperature criteria.

3. The remarkable enhancements in compressive properties of Fe₂Nb Laves phase were explained in terms of the presences of the ductile Fe phase and Fe/NbFe₂ halo.