Measurement of oxide properties for numerical evaluation of their failure under hot rolling conditions

https://doi.org/10.1016/S0924-0136(02)00290-XGet rights and content

Abstract

In this work a method based on combination of hot tensile measurements and finite element (FE) modelling is discussed—a method of direct measurement of oxide scale properties that are crucial for numerical characterisation of oxide scale failure in metal forming. This requires a combination of experiments under appropriate operating conditions and computer-based models for interpretation of test results and implementation of physical insight into predictions for technological operations. Separation loads for scale failure in tension have been measured using a modified hot tensile test technique. The separation loads seem to be critical mechanical parameters characterising scale failure and they depend on morphology of the particular oxide scale, scale growth temperature and are also very sensitive to the chemical composition of the underlying steel. Application of the micro-scale FE model to provide numerical analysis of experimental results significantly improves accuracy of determination of the separation loads. The method has been demonstrated for low carbon steel oxides.

Introduction

Studies of the oxide scale on a metal surface during deformation at elevated temperatures have mostly been motivated by a desire to understand better the micro events at the tool–workpiece interface that could influence heat transfer and friction during hot rolling. Between successive rolling passes a secondary scale is inevitably formed, which is further removed by high pressure water jets before the subsequent passes during reversing rolling or before the strip enters the tandem finishing mill. It is important to know whether the complex state of stresses at the roll gap results in oxide failure. This is because the thermal conductivity of oxide scale is significantly less than that of the steel [1]; because fractured scale can enable direct contact of hot metal with the cold tool [2]; because sliding of oxide raft is possible due to weakness of the scale/metal interface at high temperature [3]; and also because the location of the plane of sliding is determined by the cohesive strength at the different interfaces within the metal/inhomogeneous oxide and by the stress distribution when delamination within the scale takes place [4]. The phenomena will produce sharp changes in heat transfer and friction in the roll gap.

Irregularly removed scale defects are often formed on hot rolled steel strips, causing an inhomogeneous or dirty appearance as a result. In spite of measures having been undertaken, the reason why a conventional descaler cannot always remove the scale before rolling is not clear always. As has been shown elsewhere [5], the crack patterns formed in scale during hot rolling to a great extent depend on the process parameters such as temperature, scale thickness and reduction. Oxide scale on carbon steels just after the hot rolling pass when its temperature is still high cannot be assumed to be a continuous layer perfectly adhering to the metal surface and to be fully brittle. Understanding the scale removal mechanism is important for optimisation of industrial descaling conditions. However, investigation of the descaling process is far from easy, even under laboratory conditions. Additionally, the chemical content of the steel can significantly influence the state of the scale after the deformation [6]. There are various theories that have been proposed to account for the effect of alloying elements, including enhanced scale plasticity, oxide growth process and chemical bonding at the interface [7], [8], [9], [10]. No one theory can satisfactorily explain all the reported experimental observations. In this work a method based on combination of hot tensile measurements and finite element (FE) modelling is discussed. A method for direct measurement of oxide properties is critical for quantitative characterisation of oxide scale failure in metal forming.

Section snippets

Experimental

The aim of the hot tensile tests is twofold: determination of the temperature ranges for modes of oxide scale failure and evaluation of separation loads for scale failure in tension. Round tensile specimens for determination of failure modes have Ø6.5mm×20mm gauge section (Fig. 1(a)) and are ground to a 1000 grit surface finish with SiC paper. Each specimen has a hole from one end for a thermocouple allowing temperature measurement during the tests. Specimens for evaluation of separation loads

Mathematical model

The mathematical model for oxide scale, based on application of the FE method, has been proposed for analysis of oxide scale failure in hot rolling and descaling operations [5], [11] and is used here coupled with the hot tensile testing to determine the critical parameters for characterising of scale failure. In this work, the macro-parts of the FE model that compute the temperature, strain, strain rate and stress in the tensile specimen during testing are adjusted according to the

Results and discussion

Fig. 6 shows the final states of the scale after testing. The first mode corresponds to the strong interface between the oxide scale and metal relative to the oxide scale and failure occurs by through-thickness cracking. In this case, the separation force within the oxide scale is registered. The second mode relates to the interface being weaker than the oxide scale, which results in sliding of the oxide scale raft along the oxide/metal interface. The tangential separation force at the

Conclusions

The objective of this paper to pinpoint the necessity of application of different techniques, namely, a combination of hot tensile measurements and FE modelling for determination of oxide scale properties. These properties seems to be crucial for numerical characterisation of oxide scale failure in metal forming. In a case study, the reasons why making the measurement is particularly difficult are demonstrated. This is mainly because the separation loads are relatively small, and oxide scale

Acknowledgements

The authors are grateful to the Engineering and Physical Sciences Research Council, UK, for grant GR/L50198, which supported this work. Thanks are also due to Mr. J.V. Goodliffe for assistance in the course of this work.

References (18)

  • G.I Kolchenko et al.

    Izv VUZ Chernaja Metall.

    (1984)
  • Y.H. Li, C.M. Sellars, in: J.H. Beynon, P. Ingham, H. Teichert, K. Waterson (Eds.), Proceedings of the Second...
  • M Krzyzanowski et al.

    Steel Res.

    (1999)
  • K.S Tan et al.

    Steel Res.

    (2001)
  • M Krzyzanowski et al.

    Metall. Mater. Trans. B

    (2000)
  • M Krzyzanowski et al.

    Modell. Simul. Mater. Sci. Eng.

    (2000)
  • D.P Whittle et al.

    Phil. Trans. Roy. Soc. Lond. A

    (1980)
  • A Strawbridge et al.

    Mater. High Temp.

    (1994)
  • D.G Lees

    Oxid. Met.

    (1987)
There are more references available in the full text version of this article.

Cited by (31)

  • Reduction of friction of steel covered with oxide scale in hot forging

    2014, Journal of Materials Processing Technology
    Citation Excerpt :

    Torres and Colás (2000) have examined the heat conduction properties of a scale, and Guerrero et al. (1999) have proposed the heat transfer model with consideration of oxide layer in hot rolling. Krzyzanowski and Beynon (2002) have measured oxide properties for numerical analysis under hot rolling condition. Garza-Montes-de-Oca and Rainforth (2009) have investigated the influence of oxidation of the roll surface on the wear in hot rolling.

  • A time-dependent direct current potential drop method to evaluate thickness of an oxide layer formed naturally and thermally on a large surface of carbon steel

    2012, Thin Solid Films
    Citation Excerpt :

    Yu and Lenard [3] and Zhang et al. [2] have shown that this stress is a function of the thickness of the oxide layer. Therefore, according to Krzyzanowski and Beynon [4], it is important to measure the thickness of the oxide layer to observe the complex state of stresses at the roll gap that results in oxide failure. Sun et al. [5] reported that the oxide layer can act as a lubricant if it is thick and ductile or as an abrasive if it is thin, hard, and brittle.

  • Oxide scales on hot-rolled steel strips

    2008, Developments in High-Temperature Corrosion and Protection of Materials
View all citing articles on Scopus
View full text