Microstructural Characterization of As-Cast V-Si Alloys and Reevaluation of the Invariant Reactions Involving the Liquid Phase of the V-Si System

Abstract

Alloys containing Me-Si-B (Me - Refractory Metal) are of great interest for high temperature structural applications and accurate knowledge of its low order systems (binaries and ternaries) is important to predict the phase relations in the multicomponent alloys. Recent reevaluations of binaries containing Me-Si found out that the accuracy of the description of this type of systems could be improved. Knowing this, a reevaluation of the invariant reactions in the V-Si system via microstructural characterization of as-cast alloys is presented. Alloys of key compositions were prepared by arc melting pure V (min. 99.75%) and Si (min. 99.998%) and characterized via scanning electron microscopy (SEM) using back-scattered electron (BSE) mode, energy-dispersive X-ray spectroscopy (EDS) and X-ray diffraction (XRD). The results of this study confirmed the solid phases, the nature of the invariant reactions and determined new compositions for the liquid participating in certain invariant reactions.

Keywords
V-Si system; phase diagrams; silicides; solidification

1. Introduction

Accurate description of binaries and ternaries phase diagrams is of fundamental importance for the development of thermodynamic databases to predict phase relations and to define processing conditions for multicomponent alloys.

Reviews of the V-Si system were published by Smith^1,2 and Schlesinger³, consolidating the information available in the literature up to 1981, 1985 and 1990, respectively. All reviews are in good agreement, the following solid phases are considered stable: (V), V₃Si, V₅Si₃, V₆Si₅, VSi₂ and (Si). The (V) phase dissolves important amount of Si while the solubility of V in (Si) is negligible. The phase V₃Si has an important range of solubility whereas the V₅Si₃, V₆Si₅ and VSi₂ phases are stoichiometric. Smith¹ initially considered the V₃Si phase with peritectic formation but the investigation of Jorda and Muller⁴ indicated it as congruent and also established its range of solubility. This new information was taken into account in the latter reviews^2,3. The compounds V₅Si₃, VSi₂ are also formed via congruent transformations while V₆Si₅ is a product of a peritectic reaction (L + V₅Si₃ ↔ V₆Si₅), decomposing eutectoidically at lower temperature (V₆Si₅ ↔ V₅Si₃ + VSi₂). Four eutectics reactions are found in this system: (1) L ↔ (V) + V₃Si; (2) L ↔ V₃Si + V₅Si₃; (3) L ↔ V₆Si₅ + VSi₂; (4) L ↔ VSi₂ + (Si). Table 1 shows the crystallographic data for the stable solid phases of this system. Zhang et al.⁵ assessed the V-Si system taking into account the available data for phase equilibria and thermodynamic properties and also performed differential thermal analysis (DTA) experiments. Their optimization is in good agreement with the phase diagram proposed by Smith². Figure 1 shows the superposition of the V-Si phase diagram proposed by Smith² and that calculated by Zhang et al.⁵ where the vertical lines indicate the nominal compositions of the alloys prepared in this work.

Table 1
Crystallographyc data¹¹ of the stable solid phases in the V-Si system.

Phase	Pearson symbol	Space Group	Strukturbericht designation	Prototype
(V)	cI2	Im-3m	A2	W
V3Si	cP8	Pm-3m	A15	Cr3Si
V5Si3	tI32	I4/mcm	D8m	W5Si3
V6Si5	oI44	Immm	…	V6Si5
VSi2	hP9	P6222	C40	CrSi2
(Si)	cF8	Fm-3m	A4	C(Diamond)

Figure 1
V-Si Phase Diagram proposed by Smith² (solid lines) and Zhang et al.⁵ (dashed lines). Vertical lines indicate the assumed compositions of the alloys prepared in this work.

Recent investigations carried out in our group^6-10 have contributed to better description of phase diagrams of several Me-Si (Me-metal) binaries, revealing the need to reevaluate the phase relations in these systems. Thus, in this work the invariant reactions involving the liquid phase of the V-Si system have been reevaluated via microstructural characterization of as-cast alloys.

2. Experimental Procedure

V-Si alloys with masses between 2 and 8 g were prepared by arc melting V (min. 99.8 wt.%) and Si (min. 99.998 wt.%) under argon (min. 99.995%) in a water-cooled copper hearth using non-consumable tungsten electrode and Ti getter. Five melting steps were carried out for each alloy to produce chemically homogeneous samples. After melting the alloys were weighted to evaluate possible mass losses during arc melting.

The alloys were characterized through scanning electron microscopy (SEM) in the back-scattered electron mode (BSE), Energy-Dispersive X-ray Spectroscopy (EDS) and X-ray diffraction (XRD).

For the SEM/EDS analyses the alloys were hot mounted in resin, ground with SiC abrasive paper (#120 → #4000) and polished with colloidal silica suspension (OP-S). The images were obtained in a 1450VP (LEO) and TM3000 (Hitachi) SEM instruments. The EDS analyses were carried out using a Swift ED3000, Oxford Instruments.

For the XRD experiments, the as-cast alloys were mechanically ground and sieved to below 80 Mesh (177 µm). The measurements were carried out at room temperature using Ni-filtered Cu-Kα radiation in an Empyrean (Panalytical) diffractometer. The measurement conditions were 20°<2θ<90°, 0.02° step. The phases were identified based on Villars and Calvert crystallographic data¹¹ and the Powder Cell software¹².

The liquid compositions of the invariant reactions were estimated based on the microstructure of the alloys: either according to the transitions of primary precipitation between two consecutive samples or when the microstructure of the sample was 100% eutectic. In the first case the estimated composition for the liquid is the average between the calculated compositions attributing the mass loss to Si in the two consecutive samples. In the second case it is the average between the nominal composition and the calculated one attributing mass loss to Si.

3. Results and Discussion

The results will be presented according to the Si contents of the alloys, from the lowest to the highest. Table 2 shows the chemical compositions of the alloys prepared in this work, i.e. their nominal composition and their composition range based on the mass losses occurred during the arc melting (attributing all the mass losses to either V or Si). In the following text as well as in Figure 1, the alloys are referred to their calculated compositions attributing all the mass losses to Si vaporization (column 4 of Table 2). Table 3 indicates the phases present in each alloys, as determined via XRD.

Table 2
Compositions of the alloys produced in this work.

Alloy Composition (a)	Nominal composition (at. % Si) (b)	Mass loss during arc-melting(%)	Composition Range (at. % Si)
			(c)	(d)
V11.9Si	11.96	0.02	11.93	11.97
V12.6Si	13.00	0.28	12.57	13.03
V23.9Si	24.00	0.09	23.89	24.02
V27.6Si	28.02	0.36	27.61	28.11
V37.2Si	37.50	0.33	37.19	37.60
V55.7Si	56.01	0.54	55.68	56.24
V57.7Si	58.01	0.60	57.67	58.27
V58.7Si	58.97	0.42	58.74	59.15
V65.8Si	66.70	1.96	65.85	67.63
V95.0Si	94.98	2.55	94.84	96.39

• (a) Adopted composition; (b) composition based on the weighted masses of V and Si, before arc melting; (c) calculated composition attributing mass losses to Si; (d) calculated composition attributing mass losses to V.

Table 3
XRD identification of the phases present in the alloys produced in this work.

Alloy	Phase 1	Phase 2	Phase 3
V11.9Si	(V)	V3Si
V12.6Si	(V)	V3Si
V23.9Si	V3Si
V27.6Si	V3Si	V5Si3
V37.2Si	V5Si3
V55.7Si	V5Si3	V6Si5	VSi2
V57.7Si	V6Si5	VSi2
V58.7Si	V6Si5	VSi2
V65.8Si	VSi2
V95.0Si	VSi2	(Si)

The XRD results of V11.9Si and V12.6Si alloys have indicated the presence of (V) and V₃Si in their microstructure. Figure 2 (a, b) shows SEM/BSE micrograph of these alloys where an eutectic formed by (V) and V₃Si is observed in both of them. However, the V11.9Si alloy presents primary precipitation of (V) while V12.6Si alloy shows primary precipitation of V₃Si. The presence of (V) primary precipitates in V11.9Si alloy and of V₃Si in V12.6Si alloy indicates that the liquid eutectic composition is in the 11.9 - 12.6 at.% Si (i.e. 12.3 ± 0.3 at.% Si) interval, which is between 10.8 assessed by Zhang et al.⁵ and 13 at.% proposed by Smith².

Figure 2
SEM/BSE micrograph of: (a) V11.9Si; (b) V12.6Si as-cast alloys.

The XRD results of V23.9Si alloy have shown only V₃Si phase in the microstructure of this alloy, confirmed by single-phase observation in the SEM/BSE analysis. This result confirms the congruent formation of V₃Si phase, in agreement with the proposal of Smith² and Zhang et al.⁵.

The XRD results of the V27.6Si alloy have indicated the presence of V₃Si and V₅Si₃ phases in its microstructure. Figure 3 shows a SEM/BSE micrograph of this alloy indicating a full eutectic microstructure formed by V₃Si and V₅Si₃. The presence of a 100% eutectic microstructure in this alloy indicates that the composition of the eutectic liquid is between 27.6 and 28.0 at.% Si (i.e. 27.8 ± 0.2 at.% Si), a value slightly lower than that found by Smith² but equal to that proposed by Zhang et al.⁵.

Figure 3
SEM/BSE micrograph of the V27.6Si as-cast alloy.

The XRD results of V37.2Si alloy have indicated only V₅Si₃ phase in its microstructure, supported by the SEM/BSE analysis. This result confirms the congruent formation of the V₅Si₃ phases, in agreement with Smith² and Zhang et al.⁵.

Figure 4 shows SEM/BSE micrographs of the V55.7Si (a, b), V57.7Si (c) and V58.7Si (d) alloys. XRD results have indicated the presence of V₆Si₅ and VSi₂ in all these alloys as well as V₅Si₃ in the V55.7Si alloy. SEM/BSE micrographs from all these alloys presented a eutectic formed by V₆Si₅ and VSi₂ in the last regions to solidify. Alloy V55.7Si micrograph shows small fractions of V₅Si₃ primary precipitation and pro-eutectic V₆Si₅. The microstructure of V57.7Si alloy presents near fully V₆Si₅ + VSi₂ eutectic microstructure and few V₆Si₅ primary precipitates. The micrograph of V58.7Si alloy shows primary precipitation of VSi₂. The presence of V₅Si₃ primary precipitates in the V55.7Si alloy and V₆Si₅ primary precipitates in the V57.7Si alloy suggests that the composition of the liquid in the peritectic reaction involving V₆Si₅ is between 55.7 and 57.7 at.% Si (i.e. 56.7 ± 1.0 at.% Si) in accordance (considering the estimated error) with the 57 at.% proposed by Smith² and 56.2 at.% proposed by Zhang et al.⁵. The presence of V₆Si₅ primary precipitates in V57.7Si alloy and VSi₂ primary precipitates in V58.7Si alloy indicates that the composition of the liquid in the V₆Si₅ + VSi₂ eutectic is between 57.7 and 58.7 at.% Si (i.e. 58.2 ± 0.5 at.% Si). Our proposed composition for the eutectic liquid is between 57.6 (assessed by Zhang et al.⁵) and 59 at.% (proposed by Smith²).

Figure 4
SEM/BSE micrographs of the: (a,b) V55.7Si, (c) V57.7Si and (d) V58.7Si as-cast alloys.

The XRD results of V65.8Si alloy have indicated only the VSi₂ phase. The SEM/BSE micrographs of this alloy have shown a major VSi₂ microstructure, confirming the congruent formation of VSi₂ as proposed by Smith² and Zhang et al.⁵.

The XRD results of the V95.0Si alloy have indicated the presence of VSi₂ and (Si) in its microstructure. Figure 5 shows a SEM/BSE micrograph of this alloy with a fully eutectic microstructure formed by VSi₂ and (Si). The presence of a 100% eutectic microstructure indicates that the composition of the liquid in the VSi₂ + (Si) eutectic is between 94.8 and 95.0 at.% Si (i.e. 94.9 ± 0.1 at.% Si). Our proposal for the liquid composition is richer in V than the 97 at.% proposed by Smith² but in agreement with the assessment of Zhang et al.⁵.

Figure 5
SEM/BSE micrograph of the V95.0Si as-cast alloy.

The microstructural analysis of these alloys, therefore, confirms all the solid phases and invariant reactions in the V-Si phase diagram. The proposal for the liquid phase compositions in the invariant reactions is summarized in Table 4.

Table 4
Liquid compositions at invariant reactions in the V-Si system.

Invariant reaction	Liquid Composition, at.% Si	Liquid Composition, at.% Si	Liquid Composition, at.% Si
	Zhang et al.5	Smith2	This work
L ↔ (V) + V3Si	10.8	13	12.3 ± 0.3
L ↔ V3Si	25	~24 (a)	Not evaluated
L ↔ V3Si + V5Si3	27.8	~29	27.8 ± 0.2
L ↔ V5Si3	37.5	37.5	Not evaluated
L + V5Si3 ↔ V6Si5	56.2	57	56.7 ± 1.0
L ↔ V6Si5 + VSi2	57.6	~59	58.2 ± 0.5
L ↔ VSi2	66.7	66.7	Not evaluated
L ↔ VSi2 + (Si)	94.7	97	94.9 ± 0.1

• (a) Approximate value obtained by reading the phase diagram

4. Conclusions

The results of microstructural characterization (SEM/BSE + EDS + XDR) of the as-cast V-Si alloys investigated in this study confirmed the solid phases, the nature of the invariant reactions and determined new compositions for the liquid participating in certain invariant reactions of this system. The compositions of the liquid phase that participates in the invariant reactions are suggested as: (1) In the (V) + V₃Si eutectic the liquid composition is 12.3 at.% Si; (2) In the V₃Si + V₅Si₃ eutectic the liquid composition is 27.8 at.% Si; (3) In the V₅Si₃ + L peritectic the liquid composition is 56.7 at.% Si; (4) In the V₃Si + V₅Si₃ eutectic the liquid composition is 58.2 at.% Si; and finally in (5) In the VSi₂ + (Si) eutectic the liquid composition is 94.9 at.% Si.

5. Acknowledgements

Belmira Benedita de Lima-Kühn acknowledges CAPES for the financial support. The authors also acknowledge Wah Chang (Albany, OR, USA) for providing the vanadium used in this work.

6. References

• ¹
  Smith JF. The Si-V (Silicon-Vanadium) System. Bulletin of Alloy Phase Diagrams 1981;2(21):42-48.
• ²
  Smith JF. The Si−V (Silicon-Vanadium) system: Addendum. Bulletin of Alloy Phase Diagrams 1985;6(3):266-271.
• ³
  Schlesinger ME. Thermodynamics of solid transition-metal silicides. Chemical Reviews 1990;90(4):607-628.
• ⁴
  Jorda JL, Muller J. The V₃Si phase: Type of formation and homogeneity range. Journal of the Less Common Metals. 1982;84:39-48.
• ⁵
  Zhang C, Du Y, Xiong W, Xu H, Nash P, Ouyang Y, et al. Thermodynamic modeling of the V-Si system supported by key experiments. Calphad 2008;32(2):320-325.
• ⁶
  da Silva AAAP, Ramos ECT, Faria MIST, Coelho GC, Nunes CA. The Ta-Si System: Reevaluation of the Liquid Compositions in the Invariant Reactions and Determination of the Invariant Reaction Involving Both βTa₅Si₃ and αTa₅Si₃ Phases. Journal of Phase Equilibria and Diffusion 2015;36(3):209-217.
• ⁷
  Gigolotti JCJ, Nunes CA, Suzuki PA, Coelho GC. Evaluation of Phase Equilibria Involving the Liquid Phase in the Hf-Si System. Journal of Phase Equilibria and Diffusion 2014;35(5):622-630.
• ⁸
  Chad VM, Faria MIST, Coelho GC, Nunes CA, Suzuki PA. Microstructural characterization of as-cast Cr-Si alloys. Materials Characterization 2008;59(1):74-78.
• ⁹
  Baldan R, Faria MIST, Nunes CA, Coelho GC, Chad VM, De Avillez RR. Microstructural Evidence of βCo₂Si-phase Stability in the Co-Si System. Journal of Phase Equilibria and Diffusion 2008;29(6):477-481.
• ¹⁰
  Nunes CA, Coelho GC, Ramos AS. On the invariant reactions in the Mo-rich portion of the Mo-Si system. Journal of Phase Equilibria 2001;22(5):556-559.
• ¹¹
  Villars P, Calvert LD. Pearson's handbook of crystallographic data for intermetallic phases 2nd ed. Materials Park: ASM International; 1991.
• ¹²
  Kraus W, Nolze G. PowderCell for Windows (version 2.3) Berlin: Federal Institute for Materials Research and Testing; 1999.

Publication Dates

• Publication in this collection
  05 Sept 2016
• Date of issue
  Sep-Oct 2016

History

• Received
  02 Jan 2016
• Reviewed
  27 June 2016
• Accepted
  10 Aug 2016