Elsevier

Acta Materialia

Volume 58, Issue 11, June 2010, Pages 4067-4075
Acta Materialia

A new ultrahigh-strength stainless steel strengthened by various coexisting nanoprecipitates

https://doi.org/10.1016/j.actamat.2010.03.005Get rights and content

Abstract

A general computational alloy design approach based on thermodynamic and physical metallurgical principles and coupled with a genetic optimization scheme is presented. The model is applied to develop a new ultrahigh-strength maraging stainless steel. The alloy composition and heat treatment parameters are integrally optimized so as to achieve microstructures of fully lath martensite matrix strengthened by multiple precipitates of MC carbides, Cu particles and Ni3Ti intermetallics. The combined mechanical properties, corrosion resistance and identification of actual strengthening precipitates in the experimental prototype produced on the basic of the model predictions provide a strong justification for the alloy design approach.

Introduction

Steels combining properties of ultrahigh strength (UHS), good ductility and corrosion resistance are of great importance in the automotive, aerospace, nuclear, gear, bearing and other industries. They are the future key materials for lightweight engineering design strategies and corresponding CO2 savings [1]. Various alloying elements, such as C, Cr, Ni, Al, Ti, Mo, V, Mn, Nb, Co, Cu, W, Si, B and N, are combined in such alloy systems to obtain desirable microstructures and properties.

In UHS maraging steels, precipitate species may potentially cause the desired strengthening effect, depending on the nature of the precipitates and their distributions in terms of size, density and spatial distribution. Obtaining the most desirable combination depends greatly on the precipitation thermokinetics and can be tailored through alloy composition and heat treatment. From the list of potential precipitates, MC carbide is a very effective strengthening precipitate. It is usually composed of Ti, V and/or Nb, is stable and is expected even in low C steel grades. MC carbide precipitates show two main types of size distribution [2], [3]: (i) coarse (1–10 μm), ineffective primary particles formed during solidification and (ii) fine (5–500 nm), strengthening secondary precipitates. Under some conditions, a fraction of the primary MC carbides can dissolve during the austenitization/solution heat treatment and reprecipitate as fine secondary precipitates during the ageing treatment or when these materials are subjected to high-temperature applications. The significant strengthening effect of MC carbides and the effects of alloying elements have been reported for various steel grades [4], [5], [6]. On the other hand, steel grades based on intermetallic precipitates contain only a small amount of C in order to avoid the formation of carbides or carbonitrides but nevertheless reach a high strength level. Stiller et al. [7], [8] studied the precipitation sequence in Nanoflex 1RK91®. Their investigations showed a high density of Ni3(Ti, Al) precipitates, responsible for the significant strengthening effect when aged at 748 K. He et al. [9], [10] investigated the precipitation behaviour in 19Ni–4Mo–2Ti maraging steel and found that, for the optimal ageing temperature of 753 K, moderately sized Ni3Ti precipitates distribute uniformly in the martensite matrix, leading to an optimal combination of strength and fracture toughness. Moreover, the microstructure development in commercial PH15–5 stainless steel after different heat treatments was studied by Habibi-Bajguirani and Jenkins [11], [12]. Two stages of hardening were identified which are associated with the formation of Cu clusters aged at 723 K and another spherical Cu particle at higher ageing temperature (823–873 K). Isheim et al. [13] also reported a high number density (1024 m−3) of Cu-rich precipitates after ageing at 763 K for 100 min in a metastable martensitic matrix leading to near-peak hardness. The study of precipitation sequence in Nanoflex 1RK91® [8], [14] also suggested that, on ageing at 748 K, Cu is the only element that precipitates after just a few minutes. The fine Cu particles act as nuclei for Ni3(Ti, Al) precipitates during further ageing. Hättestrand et al. [14] also reported the co-existence of Cu and Ni3Ti precipitate in alloy C455: at the early ageing stage, small, densely distributed spherical precipitates rich in Cu, Ti and Ni were observed, after 1 h the precipitates grew into a rod shape and separation of Ni/Ti and Cu-rich precipitates was observed, and Ni3Ti were found to start coarsening after 2 h.

UHS stainless steels strengthened by multiple precipitates combining MC carbides and intermetallics are very attractive as they may achieve mechanical and corrosion properties beyond the current high level. However, the composition and ageing treatment conditions for fostering the precipitation of each species are very different and the various competitions/conflicts have to be compromised in the design of alloys utilizing multiple precipitates. The classical experimental trial-and-error approach or statistics-based artificial neural networks are no longer sufficient and a model more rooted in physical metallurgy is required. Recently, a theory-guided computational alloy design model has been presented by Xu et al. [15], [16], [17], [18] in which the alloy composition and corresponding heat treatment parameters for UHS stainless steels are optimized integrally so as to achieve desirable microstructures within a genetic optimization framework. The model was applied to design exercises of UHS stainless steels utilizing MC carbide, Cu clusters and/or Ni3Ti/NiAl precipitates. In the present work, the model is applied to design an alloy possessing multiple strengthening precipitates of MC carbides, Cu particles and Ni3Ti intermetallics. The characterizations of the prototype alloy, including mechanical properties, corrosion resistance and identifications of precipitates, provide experimental justification for the alloy design approach.

Section snippets

Model description

The combination of UHS and good toughness can be realized via an uniform dispersion of fine precipitates in a lath martensite matrix. This microstructure is obtained via a two-step heat treatment: solution treatment to achieve a homogenous and fully austenitic state followed by quench to room temperature so as to generate the martensitic matrix, with further ageing at a modest temperature to allow precipitation of desirable species in this finely grained matrix structure. A schematic diagram of

Application of the model

In order to incorporate complex interactions escaping intuitive analyses, steel compositions containing up to 13 alloying elements (C, Cr, Ni, Ti, Mo, Al, Cu, Co, Nb, N, V, Mn and Si) were considered. The concentrations of all alloying elements were allowed to vary except for Mn (set at 0.5 wt.%) and N (set at 0.01 wt.%), as recommended for the facilitation of alloy production on the industrial scale. The candidate solution is represented in the genetic model as a chromosome which concatenates 11

Experimental procedures

The alloys were prepared in a vacuum induction melting furnace as ingots of approximately 25 kg. The ingots were vacuum melted, using high-purity ingredients to achieve the best cleanliness. Phosphorus, sulphur, arsenic, boron and tin were kept at the lowest possible levels. The melting temperature was kept relatively low at about 1773 K and the chamber was flushed with argon. The ingots were ground to smoothen their surfaces, soaked at 1473 K for 2 h and subsequently forged into a slab at a

Microstructures

The microstructures on the cross-section perpendicular to the rolling direction were characterized after various solution treatment conditions. Typical optical micrographs after solution treatment at 1373 K for 15 min and 1423 K for 60 min are shown in Fig. 2a and b, respectively. The micrographs display homogeneous microstructures and indicate that prior austenite grains grow significantly with increasing solution treatment temperature and longer time, but invariably transform to lath martensite

Thermodynamic predictions

Thermodynamic equilibrium calculations were performed for both the design and the actual alloy composition of alloy CAR at various austenitization and ageing temperatures. The results are shown in Table 2. ThermoCalc calculations show that alloy CAR of the designed composition will yield 0.5% of primary carbide at the designed austenitization temperature of 1533 K, which amount is exactly at the upper limit of the design criterion. However, the C concentration of the experimental alloy is higher

Conclusions

The paper presents a general theory-guided computation alloy design methodology and characterizations of the designed prototype alloy. The composition and heat treatment parameters were designed by integrally applying a genetic design approach so as to achieve microstructures of a fully martensitic matrix strengthened by multiple desirable precipitates of MC carbides, Cu particle and Ni3Ti intermetallics.

The characterization of prototype alloy CAR showed that, after the austenitization

Acknowledgements

This research was carried out under the project number MC5.04192 in the framework of the Research Program of the Materials innovation institute M2i (www.m2i.nl), the former Netherlands Institute for Metals Research. The authors are grateful to Dr. V. Bliznuk (Ghent University, Belgium) for support in the TEM investigation and Dr. S.J. Garcia Espallargas (TUDelft, the Netherlands) for assistance in the corrosion potential measurement. Furthermore, we acknowledge the professional advice from Dr.

Reference (21)

  • D. Raabe et al.

    Scripta Mater

    (2009)
  • A.F. Padilha et al.

    J Nucl Mater

    (1982)
  • R.L. Klueh et al.

    Scripta Mater

    (2005)
  • K. Stiller et al.

    Mater Sci Eng A

    (1998)
  • K. Stiller et al.

    Acta Mater

    (1998)
  • H.R. Habibi-Bajguirani

    Mater Sci Eng A

    (2002)
  • D. Isheim et al.

    Acta Mater

    (2006)
  • M. Hättestrand et al.

    Acta Mater

    (2004)
  • W. Xu et al.

    Comput Mater Sci

    (2008)
  • W. Xu et al.

    Comput Mater Sci

    (2009)
There are more references available in the full text version of this article.

Cited by (0)

1

Present address: Department of Materials Science and Metallurgy, University of Cambridge, Pembroke Street, CB2 3QZ Cambridge, UK.

View full text