Introduction to an approach based on the (α+β ... · dependent influence factors, e.g ......

4
Available online at www.sciencedirect.com Procedia Engineering Mesomechanics 2009 Introduction to an approach based on the (α+β) microstructure of elements of alloy Ti-6Al-4V M. Stoschka a , W. Tan a , W. Eichlseder a , M. Stockinger b, * a University of Leoben, Department of Mechanical Engineering, Franz-Josef-Straße 18, 8700 Leoben, Austria b Böhler Schmiedetechnik GmbH&Co KG, Mariazellerstraße 25, 8605 Kapfenberg, Austria Received 28 February 2009; revised 17 April 2009; accepted 20 April 2009 Abstract The goal of linking a fatigue life approach with different forging processes enforces the consideration of the main process dependent influence factors, e.g. effective strain, effective strain rate, anisotropy and morphology, alpha-grain size, alpha- contiguity, distance between secondary alpha and beta-lamellae, etc. Linking the specimen lifetime results to these microstructural features, new design parameters for forged titanium parts can be identified. These forecasting parameters can than be used at design stage to optimise the (α+β)-titanium alloy forgings with regard to lifetime. Keywords: Fatigue analysis; Aerospace components; Microstructural evaluation 1. Introduction In the last years the main objective of the research done has been the development of a method to predict the lifetime of Ni-base forged superalloy IN718 aircraft parts 1,2 . As the final result, a microstructure based fatigue approach was achieved introducing two new parameters. First, the value of the ASTM grain size is substituted by the microstructural energy parameter e, which is more sensitive to fine microstructures. Second, the grade of bimodality or the amount of ‘as-large-as’ grains in a finer matrix is evaluated by the factor of heterogeneity b. By use of this two new parameters e and b, an alternative and more meaningful characterisation of the microstructural image can be done for Ni-base forged superalloy 718. The current and further work deals with the build-up of fracture mechanical and fatigue test based properties for the (α+β)-titanium alloy, the expansion of the investigated microstructural energy approach to characterise the dual microstructure of Ti-6Al-4V and the validation of the modified microstructural energy approach regarding lifetime. The alternative microstructural evaluation supports a multi-parametric characterization of the forging and heat- treatment process. The new parameters e and b offer a complementary link between local microstructure and fatigue life. The microstructural dependent local S/N-curves can be used to calculate the fatigue life of components using * Corresponding author. Tel.: +43-3842-402-1455; fax: +43-3842-402-1402. E-mail address: [email protected]. Procedia Engineering 1 (2009) 31–34 www.elsevier.com/locate/procedia doi:10.1016/j.proeng.2009.06.009

Transcript of Introduction to an approach based on the (α+β ... · dependent influence factors, e.g ......

Available online at www.sciencedirect.com

Procedia Engineering 01 (2009) 000–000

Procedia Engineering

www.elsevier.com/locate/procedia

Mesomechanics 2009

Introduction to an approach based on the (α+β) microstructure of elements of alloy Ti-6Al-4V

M. Stoschkaa, W. Tana, W. Eichlsedera, M. Stockingerb,* aUniversity of Leoben, Department of Mechanical Engineering, Franz-Josef-Straße 18, 8700 Leoben, Austria

bBöhler Schmiedetechnik GmbH&Co KG, Mariazellerstraße 25, 8605 Kapfenberg, Austria

Received 28 February 2009; revised 17 April 2009; accepted 20 April 2009

Abstract

The goal of linking a fatigue life approach with different forging processes enforces the consideration of the main process dependent influence factors, e.g. effective strain, effective strain rate, anisotropy and morphology, alpha-grain size, alpha-contiguity, distance between secondary alpha and beta-lamellae, etc. Linking the specimen lifetime results to these microstructural features, new design parameters for forged titanium parts can be identified. These forecasting parameters can than be used at design stage to optimise the (α+β)-titanium alloy forgings with regard to lifetime. © 2009 Elsevier B.V. All rights reserved

Keywords: Fatigue analysis; Aerospace components; Microstructural evaluation

1. Introduction

In the last years the main objective of the research done has been the development of a method to predict the lifetime of Ni-base forged superalloy IN718 aircraft parts1,2. As the final result, a microstructure based fatigue approach was achieved introducing two new parameters. First, the value of the ASTM grain size is substituted by the microstructural energy parameter e, which is more sensitive to fine microstructures. Second, the grade of bimodality or the amount of ‘as-large-as’ grains in a finer matrix is evaluated by the factor of heterogeneity b. By use of this two new parameters e and b, an alternative and more meaningful characterisation of the microstructural image can be done for Ni-base forged superalloy 718. The current and further work deals with the build-up of fracture mechanical and fatigue test based properties for the (α+β)-titanium alloy, the expansion of the investigated microstructural energy approach to characterise the dual microstructure of Ti-6Al-4V and the validation of the modified microstructural energy approach regarding lifetime.

The alternative microstructural evaluation supports a multi-parametric characterization of the forging and heat-treatment process. The new parameters e and b offer a complementary link between local microstructure and fatigue life. The microstructural dependent local S/N-curves can be used to calculate the fatigue life of components using

* Corresponding author. Tel.: +43-3842-402-1455; fax: +43-3842-402-1402. E-mail address: [email protected].

Procedia Engineering 1 (2009) 31–34

www.elsevier.com/locate/procedia

doi:10.1016/j.proeng.2009.06.009

Stoschka M. et al./ Procedia Engineering 01 (2009) 000–000

damage concepts like the notch stress approach. Since the mechanical properties vary as a function of a wide range of parameters such as the α-content, α/β-grain morphology, beta-lamella distance, etc. it is a comprehensive task to assess the local fatigue strength.

Fig. 1. General flowchart of a microstructural based Ti-6Al-4V fatigue approach

To assess the lifetime of forged and heat-treated titanium parts, a local fatigue approach has to be used. Therefore, the manufacturing based knowledge of the local macroscopic material properties in conjunction with the local fatigue strength provides the basis for numerical fatigue life calculation using local stress concepts3. In the case of superalloys and their turbine disc applications4, the microstructure influences the low-cycle-fatigue and the creep-rupture lifetime in an opposite way.

2. Microstructural evaluation

To characterise the microstructure of plane metallographic sections, standard test methods such as the average grain size can be used5. The average grain size G is determined by intersection counting or planimetric methods. In case of non-uniformly distributed grains, the Abrams-procedure6 delivers the most suitable average grain size results. In addition, to describe the properties of individual planar grains, geometric features like the area, the length of the boundary or the integral of curvature can be used. These geometric properties also support a spatial description of the microstructure using Poisson-Voronoi tessellations. The mathematical link between plane to spatial microstructure and the corresponding stereographic formulations is summarized by Ohser and Mücklich7.

The introduced microstructural based energy approach1 offers an alternative way to describe the microstructural properties of plane metallographic sections. The microstructural based energy approach was originally developed to link the microstructural properties of Ni-base forged superalloy IN718 to its fatigue lifetime2.

2.1. Microstructural based energy approach

The shortened development circle of a microstructural based fatigue approach including experimental as well as accompanying modelling work concerning forged (α+β)-titanium parts is illustrated in figure 1. Different hot-forging technologies allow the adjustment of specific microstructures ranging from mill- or recrystallisation annealed over solution treated to beta-forged. As first step, the fatigue behaviour was achieved by specimen S/N-curves using distinct forging- and heat-treatment manufacturing parameters. Multiple orthogonal sections of each specimen were metallographically investigated using light optical microscopy. This build-up of an extensive metallographic database is labelled “1” in figure 1; and the accompanying fatigue lifetime results as “2”. The manufacturing processes create different shape and connectivity of the α-grains. Mill annealing, for example, causes

32 M. Stoschka et al. / Procedia Engineering 1 (2009) 31–34

Stoschka M. et al./ Procedia Enginerting 01 (2009) 000–000

incomplete recrystallisation and leads therefore to a distinct texture of the primary α-grain shapes in respect to the forging process. Due to the hot-forming temperature no major crystallographic texture occurs in the microstructure8. The alternative microstructural based energy approach uses therefore the shape and connectivity of the individual α/β-grains as current Ti-6Al-4V approach. The build-up of the homologous model of individual α/β-grains is labelled as “3” in figure 1. The interior procedure is identical to the original microstructural energy approach of alloy IN718 and proceeds as follows. The calculation of the major radius of the synthetic ellipsis per α/β-grain is based on the unbiased average of interior extent, outer diameter and cord length. The minor radius is build up as the sum of two parts; on one hand, based on area equivalency, and on maximum interior extent on the other. The summarising equation is multiplicative weighted by convexity and elongation. Adjacent grains are connected by beam elements using a Voronoi-triangulation and possessing a unique material card each. The metallographic image is transformed to a synthetic microstructure. In the next step, the plane structure is loaded by a uniform displacement vector. Dependent of the investigated load angle, a characteristic strain energy value is obtained. The process of transforming the metallographic image to synthetic microstructure and its evaluation using the original alloy IN718 concept9 is shown in figure 2. The elastic strain energy as a characteristic response function is linked to the stiffness of the underlying particle-based microstructure. The evaluation of this aggregated energy curve leads to the introduced microstructural energy based parameters e and b.

Fig. 2. Microstructural based energy approach applied to alloy IN718; (a): Metallographic section with grain particle based shape results (inserts); (b): Synthetic microstructure; (c): Energetic response function

2.2. Extension to Ti-6Al-4V

Dependent on the heat treatment, the (α+β) titanium microstructure varies widely; see figure 3. Lamellar microstructures show continuous arrangement of the α-regions, whereas bimodal or equiaxed microstructures show a more insular distribution. The fatigue life behaviour depends on the size of the β-grains as well as the fine-lamellar (α+β)-distance8. In addition, the overall content of the α-phase depends on the forging process and heat treatment parameters. Under uniaxial tension, the crystallographic orientation of the α-phase (001)-basal plane has a major influence on the local fatigue life10.

As the next step in the development of the titanium alloy model, crack growth properties are included as follows. Analysing the crack initiation and stable propagation region according to the modified Klesnil-Lukáš relation; the fine-lamellar microstructure shows an increased stress intensity threshold ΔKth of about sixty percent compared to the bimodal one (ΔKth,lamellar~4.9 MPa√m vs. ΔKth,bimodal 3.1 MPa√m). This comparison is only valid for stress ratios of R = 0.3. Both crack growth curves join near the unstable fracture region. For this reason, the modified titanium alloy analysis uses for the synthetic microstructure, build up with triangulated beam elements, an inhomogeneous element property approach. In case of connected fine-lamellar beta-grains the element property value is increased by sixty percent. In addition, the synthetic microstructural properties of connected alpha-grains are kept unchanged. Although the stress intensity threshold is only valid for long crack propagation, it is used to distinguish between α/β-grains in the current model. This milestone is labelled “4” in figure 1. In addition, the microstructural dependency of short crack growth behaviour is currently being investigated11. Results from this ongoing fracture mechanical research will be used to improve the accuracy of the model. The statistical evaluation of the microstructural energy e – describing chiefly the size of α-grains, and the factor of heterogeneity b – indicating the adjacency of individual grains, is under further development; compare to labels “5” and “6” in figure 1. The combined use of the newly introduced parameters e and b in conjunction with the amount of α-phase offers

M. Stoschka et al. / Procedia Engineering 1 (2009) 31–34 33

Stoschka M. et al./ Procedia Engineering 01 (2009) 000–000

additional identification parameters to characterise the complex field of Ti-6Al-4V microstructures instead of textual denomination as lamellar, bimodal or globular. As an example, figure 3 shows proprietary results of the combined and α/β-weighted microstructural based energy approach.

Fig. 3. Results of microstructural Ti-6Al-4V classification (lamellar; mill-annealed; beta-forged; recrystallisation annealed; solution treated)

3. Conclusion

Using the presented approach, the microstructural energy parameter e and the factor of heterogeneity b offer alternative paths in characterising both the microstructure and the morphology of metallographic Ti-6Al-4V sections. In addition, the extreme values of the curve representing the summarised energy transfer function indicate the orientation of the texture. These new parameters allow a broad characterisation of the dual microstructure. The forging process simulation chain will be closed by integrating this microstructural approach to specimen fatigue life and to microstructural forecasting parameters at the design stage. The investigations show that based on this knowledge a tailored interdisciplinary lifetime optimisation and a useful definition of the demanded specifications can be worked out, together with the aerospace component designers in order to deliver economic and safe parts.

References

1. Riedler M, Stockinger M, Stoschka M, Eichlseder W. Fatigue Analysis of Forged Aerospace Components based on Microstructural

Parameters. Key Engineering Materials 2007; 348-349:185-3.

2. Stoschka M, Stockinger M, Leitner H, Riedler M. Assessment of lifetime calculation of forged IN718 aerospace components based on a

multi-parametric microstructural evaluation. In:. Proc. of the 11th International Symposium on Superalloys 2008, TMS, 2008:573-10.

3. Eichlseder W, Unger B. Prediction of the fatigue life with the finite element method. SAE Paper 1994; 940245.

4. Reed RC. The Superalloys Fundamentals and Application. Cambridge University Press; 2008.

5. ASTM Standard E 112, 1996e2. Standard Test Methods for Determing Average Grain Size. In: ASTM Internat., West Conshohocken, PA.

6. Abrams H. Grain Size Measurement by the Interception Method. Metallography 1971; 4:59-78.

7. Ohser J, Mücklich F. Statistical Analysis of Microstructures in Material Science. John Wiley & Sons; 2000.

8. Lütjering G. Influenece of processing on microstructure and mechanical properties of (α+β) titanium alloys. Materials and Science

Eng. A, 1998; A243:32-14.

9. Wagner L, Lütjering G. Influence of shot peening treatment on the fatigue limit of Ti-6Al-4V. Shot Peening Society 1984; 201-7.

10. Nalla RK, Boyce BL, Campbell JP, Peters JO, Ritchie RO. Influence of Microstructure on High-Cycle-Fatigue of Ti-6Al-4V. Bimodal vs.

Lamellar Structures. Metallurgical and Materials Transactions A 2002; 33A: 899-20.

11. Oberwinkler B, Oberwinkler C, Redik S, Leitner H. Analysis of Short Crack Growth for Two Representative Light Metals. Accepted for

Proceedings of the 12th International Conference on Fracture; 2009.

34 M. Stoschka et al. / Procedia Engineering 1 (2009) 31–34