Elsevier

Materials & Design

Volume 112, 15 December 2016, Pages 131-139
Materials & Design

Combinatorial screening of the microstructure–property relationships for Fe–B–X stiff, light, strong and ductile steels

https://doi.org/10.1016/j.matdes.2016.09.065Get rights and content

Highlights

  • Systematic screening of mechanical, physical and microstructural properties of Fe–10B–5X alloys for lightweight design

  • Ti, Zr, Hf, V, Nb, Ta, Cr, Mo, W additions exhibit widely differing microstructures of different boride phases and mixes

  • Cr and Zr were identified as the most promising elements for design of high modulus steels

  • Fe–B–Cr alloys exhibit similar performance as the established Ti alloyed materials, but at greatly reduced alloying costs

  • Zr alloyed materials are softer and less ductile, but achieve extremely high stiffness/density ratio

Abstract

We systematically screened the mechanical, physical and microstructural properties of the alloy systems Fe–10 B–5 X (at.%; X = Ti, Zr, Hf, V, Nb, Ta, Cr, Mo, W), in order to identify novel metal matrix composite steels as next generation lightweight materials. The alloys were synthesised and processed by bulk liquid metallurgical techniques, and subsequently analysed for their mechanical and physical properties (i.e. Young's modulus, density, tensile strength and ductility) as well their microstructure and constitution. From the wide variety of observed boride phases and microstructures and resultant different properties, Cr and Zr additions were found to be most effective. Cr qualifies well as the high fraction of M2B borides of spherical morphology allows achieving a similar stiffness/density ratio and mechanical performance as the reference Ti alloyed materials, but at substantially reduced alloy costs. Zr blended composites on the other hand are softer and less ductile, but the alignment of spiky ZrB2 particles during swaging led to a much higher – though most probably anisotropic – specific modulus. Consequences and recommendations for future alloy and processing design are outlined and discussed.

Introduction

Weight reduction is a major technical challenge for structural material design. An ideal pathway is blending strong and ductile metallic matrices for optimised mechanical performance with stiff and low-density ceramic particles for improved physical properties, thus creating metal matrix composites [1], [2], [3], [4], [5], [6]. Key material parameters are a high Young's modulus (E) for improved stiffness, high yield strength (YS) to allow for higher loads to be transferred, satisfactory ductility (such as tensile elongation TE) for forming operations during manufacturing, and a low density (ρ). High modulus steels (HMS; as the most common acronym for iron (Fe)-based composites) are especially attractive, as Fe exhibits a similar specific modulus (E/ρ) as established lightweight materials such as for example aluminium, magnesium or titanium alloys (about 26 GPa g 1 cm3), but also a wide range of achievable mechanical properties due to its multitude of equilibrium and non-equilibrium phase transformations and low production costs [7].

Selecting suited particle phases however – from intrinsic properties alone – is difficult, as possible candidates are numerous and range from carbides to nitrides, oxides, intermetallics and borides [1]. Besides high efficiency (i.e. a high specific modulus), other aspects of critical importance are e.g. their thermodynamic stability (to prevent dissolution in the matrix), formation kinetics (possible floatation of low density particles forming in the liquid), interfacial properties (to ensure wetting of particles), as well as the availability and costs of the constituting elements. Furthermore, in-situ formation of such particles during synthesis greatly simplifies the production of such composite materials. All these factors are important for liquid metallurgy synthesis, and thus of key relevance for efficient mass production. When reviewing the wide range of stiff and low density particles under these engineering constraints, only a limited number of possible phases remain: Diamond for example has one of the highest specific moduli, but cannot be synthesised in-situ and will dissolve in Fe when added ex-situ. By contrast nitrides or oxides have high thermodynamic stability, but as they form rapidly in the melt and are typically of low mass density, they float and form a slag instead of rendering dispersed in the solidified material [8]. Carbides are typically less effective, and detrimental for both the mechanical properties (through preferential precipitation at grain boundaries) and the melt viscosity (formation in the liquid phase) [4].

Borides, on the other hand, fulfill most of the above listed criteria for HMS design. Specifically the Fe–Titanium diboride (TiB2) system has been intensely investigated [4], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], as TiB2 has a high specific modulus (~ 125 GPa g 1 cm3) [26] and can be precipitated in-situ from a homogeneous Fe–Ti–B melt in a pseudo-binary eutectic reaction [9], with excellent interfacial properties and sufficient mechanical compliance with the matrix [11], [18], [23]. An Fe–10 vol.% TiB2 alloy for example typically exhibits a specific modulus of about 30 GPa g 1 cm 3 at 400 MPa and ~ 30% TE [27], [28]. The increased costs associated with the comparatively large quantities of Ti required (about 5 wt.% for the above mentioned alloy) can be decreased by the in-situ reduction of Ti-oxides through aluminium additions in the melt [8]. More problematic is the pronounced embrittlement observed with increasing TiB2 fractions. This is caused by the unfavourable particle microstructure once the eutectic TiB2 concentration (about 12 vol.% [20], [25]) is exceeded and formation of large (several μm2) primary particles in addition to the already sharp edged eutectic TiB2 lamellae occurs. While the particle morphology can be successfully controlled by alloying additions and/or tailored solidification kinetics, leading to excellent property profiles [22], [24], [27], these measures further increase the associated efforts and costs, and thus may slow down the application of HMS as the next generation lightweight material on a broad scale.

However, as HMS still represent a relatively novel class of structural materials, the potential of other Fe–boride systems has not yet been investigated and exploited. Compared to the more thoroughly studied Fe–TiB2 based HMS, these alternative alloy systems may offer more effective boride particles (so that lower particle fractions are required for the same gain in properties), and/or improved boride microstructures even with established alloying and processing routes (allowing for improved mechanical performance at similar particle fractions). Identifying the most suitable Fe–B–X system for the design of HMS from existing literature alone is difficult though, for the following reasons: (i) The thermodynamics of the Fe-rich corners of ternary Fe–B–X systems are often not fully understood, thus making it difficult to reliably predict which phases are stable for a specific alloy composition and temperature [29]. (ii) Depending on the alloy system, a multitude of equilibrium phases occurs, with often closely spaced compositional ranges. In the Fe–B–Cr system for example, 9 binary and 6 ternary borides have been reported, not even counting metastable phases which are out of thermodynamic equilibrium [30]. (iii) Apart from the difficulties in predicting the formation of specific phases, data concerning their intrinsic properties (such as E and ρ) is rather scarce. Additionally, almost no information exists on what kind of particle microstructures (i.e. morphology, size and dispersion of the borides) will result for different processing conditions. Hence the prediction of the physical and mechanical performance of such an HMS alloy, i.e. its specific modulus, strength and ductility, is virtually impossible.

An experimental approach, i.e. screening the constitution, microstructure, mechanical and physical properties of ternary Fe–B–X systems, on the other hand, is extremely time consuming in view of the large number of possible alloy compositions and thermomechanical processing parameters, even with novel high throughput bulk metallurgical techniques [31]. Thin film combinatorial techniques may be substantially faster, but the correlation lengths of the property-dominating microstructural features (grain size, crystallographic texture, precipitate dispersion and topology etc.) typically exceed the dimensions accessible in thin films. It is therefore of high interest to use an alternative experimental strategy to efficiently obtain insight into which ternary Fe–B–X system has the highest potential for future HMS alloy design.

Section snippets

Objective and approach

The aim of this study is to evaluate ternary Fe–B–X alloy systems for the design of stiff, light, strong and ductile HMS. In view of the above listed difficulties with identifying the most suitable system exclusively from literature data, we follow here a property driven approach, i.e. first producing material based on the available data, then evaluate its mechanical and physical properties, followed by investigation of microstructure and constitution. This allows us to efficiently provide

Materials and methods

All alloys of this study are of the composition Fe–10 B–5 X (at.%; X = Ti, Zr, Hf, V, Nb, Ta, Cr, Mo, W). This represents a hypo-eutectic concentration for the reference alloy with Ti, which is an optimum benchmark for judging the performance of the other material systems, as it does not undergo strong embrittlement through the formation of coarse primary particles, and represents the current state of the art in HMS design. Charges of ~ 600 g were synthesised by Arc melting pure materials (> 99.99%)

Physical and mechanical properties

The E and ρ data obtained for the Fe–10 B–5 X alloys are plotted in Fig. 1. In the as-cast state (Fig. 1a), the Mo alloyed samples exhibited with 221 GPa the lowest E values (black rhombi), and the Nb alloyed material with 246 GPa the highest ones. The bulk mass density ρbulk (blue circles) was found to be lowest for the Ti alloyed materials (7.38 g cm 3) and highest for Ta-alloyed samples (8.28 g cm 3). The ρflakes values (empty red circles) were almost identical to the ρbulk data except for Nb, Ta

Microstructure–property relationships of the different Fe–B–X alloys

We successfully screened the stiffness, density, strength and ductility of Fe–10 B–5 X alloys (at.%; X = Ti, Zr, Hf, V, Nb, Ta, Cr, Mo, W ) after liquid metallurgy synthesis and thermo-mechanical processing. Our results allow for the first time to systematically evaluate and compare their microstructures and resultant properties, and thus to gain a first insight into their applicability for the alloy design of novel generations of lightweight structural materials. The reference Ti alloyed

Summary and conclusions

We systematically screened the mechanical, physical and microstructural properties of nine Fe–10 B–5 X alloy systems (at.%; X = Ti, Zr, Hf, V, Nb, Ta, Cr, Mo, W), all synthesised and processed by standard liquid metallurgical techniques. As precise knowledge about the forming boride phases and their intrinsic properties is rather incomplete, we chose to follow a property driven approach in order to identify the most suitable systems for the alloy design of metal matrix composite steels as next

Acknowledgements

M. Kulse, B. Breitbach and L. Christiansen are gratefully acknowledged for their support with synthesis, XRD analysis, data evaluation and metallographic preparation.

References (46)

  • H. Springer et al.

    Rapid alloy prototyping: compositional and thermo-mechanical high throughput bulk combinatorial design of structural materials based on the example of 30Mn–1.2C–xAl triplex steels

    Acta Materialia

    (2012)
  • H. Springer et al.

    Bulk combinatorial design of ductile martensitic stainless steels through confined martensite-to-austenite reversion

    Mater. Sci. Eng. A

    (2013)
  • H. Springer et al.

    Combinatorial design of transitory constitution steels: coupling high strength with inherent formability and weldability through sequenced austenite stability

    Mater. Des.

    (2016)
  • A. Antoni-Zdziobek et al.

    Solidification paths in the iron-rich part of the Fe–Ti–B ternary system

    J. Alloys Compd.

    (2016)
  • Y.H. Duan et al.

    Elastic constants of AlB2-type compounds from first-principles calculations

    Comput. Mater. Sci.

    (2012)
  • T.A. Sviridova et al.

    A new metastable phase in Fe–Nb–B system

    J. Alloys Compd.

    (2016)
  • R.F. Gibson

    A review of recent research on mechanics of multifunctional composite materials and structures

    Compos. Struct.

    (2010)
  • F. Bonnet et al.

    High modulus steels: new requirement of automotive market. How to take up challenge?

    Can. Metall. Q.

    (2014)
  • I.A. Ibrahim et al.

    Particulate reinforced metal matrix composites — a review

    J. Mater. Sci.

    (1991)
  • D.J. Lloyd

    Particle-reinforced aluminum and magnesium matrix composites

    Int. Mater. Rev.

    (1994)
  • Y. Feng

    Strengthening of Steels by Ceramic Phases

    (2013)
  • A. Evans, C. San Marchi, A. Mortensen, Metal Matrix Composites in Industry - An Introduction and a Survey, Springer...
  • N. Chawla, K.K. Chawla, Metal Matrix Composites, Springer...
  • Cited by (17)

    • Novel approach of alloy design and selection for additive manufacturing towards targeted applications

      2021, Journal of Alloys and Compounds
      Citation Excerpt :

      From an experimental standpoint, this preselection based on modelling is required, because a purely combinatorial approach to map a quinary system with one weight percent (wt%) step would require the manufacturing of >4,5 million alloys. Experimental alloy development strategies have recently been extended towards more accelerated screening methods, including combinatorial casting procedures [16–18] or by chemically graded compositions (CGC) made by directed energy deposition (DED) [19–23]. However, small casting, as well as CGC, are unsuitable to deliver an accurate assessment of the printability of an alloy for the PBF process since the thermal history is widely different, leading to substantially different microstructures [24].

    • On the effect of carbon content and tempering on mechanical properties and stiffness of martensitic Fe–18.8Cr–1.8B–xC high modulus steels

      2021, Materials Science and Engineering: A
      Citation Excerpt :

      The impact of C and tempering on materials' performance may be best evaluated by means of materials' specific moduli versus its tensile properties (cf. Fig. 4) and hardness (cf. Fig. 8). The E/ρ values of the Fe–B–Cr–C alloys (as-quenched state) in this study compared to values of Fe–B–Cr alloys [9,15,16] with similar particle content do not seem to differ considerably and are ranging between ~ 31.8 and 33 GPa g−1 cm3. However, the impact of C and tempering on tensile properties is significant, especially on YS and ultimate tensile strength (UTS) [9].

    • Additive manufacturing of high modulus steels: New possibilities for lightweight design

      2020, Additive Manufacturing
      Citation Excerpt :

      The particle morphology as well as the mechanical properties of the matrix can be adapted by additional alloying elements, for example to promote strengthening by solid solution and / or precipitations [15,17]. Even switching the HMS alloy system entirely, away from the most researched Fe-Ti-B alloys [3,9,17–21] to the more flexible and cost effective Fe-Cr-B based HMS [22–24], can be considered to further broaden the spectrum of possible industrial applications. Our findings also indicate that other powder metallurgy (PM) processes should be able to deliver similarly favourable microstructures and properties.

    • On the effect of Ni additions to Fe – Cr – B high modulus steels

      2019, Materials and Design
      Citation Excerpt :

      These limitations inspired the search for alternative alloy systems for HMS development. The Fe – Cr – B system was found to be especially attractive, which rely on the in-situ formation of borides of the M2B type, namely (Cr,Fe)2B [9,10]. The density of this phase of about 7 g cm−3 [11] renders (Cr,Fe)2B slightly less effective (lower E/ρ ratio) than TiB2, but in turn its slightly elevated density strongly reduces its tendency to float and agglomerate when precipitating during solidification.

    View all citing articles on Scopus
    View full text