Compressive creep testing of refractories at elevated loads—Device, material law and evaluation techniques

doi:10.1016/j.jeurceramsoc.2014.05.034

Journal of the European Ceramic Society

Volume 34, Issue 15, December 2014, Pages 4037-4042

https://doi.org/10.1016/j.jeurceramsoc.2014.05.034 Get rights and content

Abstract

In order to cost-effectively characterize the high temperature compressive creep behaviour of refractories a testing device was designed for application at elevated loads. Special measures have been taken necessary to enable an even stress distribution within the specimen. To identify Norton-Bailey strain hardening creep law parameters a general inverse procedure using a Levenberg–Marquardt algorithm was developed. Satisfying experimental results could be received from the creep measurement in a wide range of temperatures and loads for both shaped and unshaped materials. By fitting the strain/time curves the creep law parameters of refractories under various temperatures can be precisely identified. The measurements also reveal that at elevated loads all three creep stages can be observed.

Introduction

During service refractories are frequently exposed to intense thermomechanical loads resulting from the collaborative effects of severe thermal shock, thermal expansion of refractories and external mechanical constraints.1, 2 Under these conditions refractories may respond in elastic or inelastic manner which depends on the magnitudes and types of loads as well as on the material behaviour at high temperatures. Both material failure under Mode I–III conditions and creep may account for this inelastic behaviour and bring about an irreversible deformation of refractories.3, 4, 5, 6, 7

To characterize the creep behaviour of refractories, conventionally two types of testing procedures are employed. One is so called creep-in-compression (abbreviated as CIC) which applies the same equipment as the testing procedure for refractoriness-under-load (abbreviated as RUL). The standards of CIC in Europe, USA, Japan and China are more or less similar and mainly differ in specimen dimension, testing load and heating rate.⁸ According to the European standard EN 993-9,⁹ a fixed compressive stress of 0.2 MPa is loaded on the specimen during the heating-up and dwell periods. The change in length of the specimen is measured from the differential displacement of two corundum tubes through linear variable differential transducers (LVDTs). The deformation at certain dwell time relative to the maximum expansion occurring during the heating-up period is determined. This method is applied to compare the creep resistance capacity of refractories and to receive guidance for material selection and development in a rather phenomenological manner.10, 11, 12, 13 However, a deeper understanding of the creep mechanisms of refractories is demanded as well as extracting creep data for further thermomechanical modelling. For this purpose various loading conditions were also considered.14, 15, 16, 17, 18, 19 Formulating the secondary creep rate in dependence of temperature and stress offers one way to reveal the creep mechanisms of refractories, from which measures for material improvement may follow.

What are the tasks a satisfactory creep testing equipment should fulfil? The CIC measurement according to the European standard EN 993-9 is not suitable for determination of creep laws. One reason is that the onset of creep is not defined. Creep will start during the heating up procedure before the beginning of the dwell time because the specimen is heated up under load. Moreover the restriction of the maximum load to 0.2 MPa does not allow investigation of load levels significant for practical use of refractories in many cases. A newly designed testing equipment should especially allow the following features of the measurement: The start of the loading under isothermal conditions must be defined exactly. Moreover, displacement has to be measured on the specimen surface itself, preferably at least at two diametrical locations, in order to assess even creep of the material. Further alignment of the specimen and the testing machine is of high importance: the line of action of the imposed load has to coincide with the specimen axis. Otherwise a bending moment would be caused and result in an uneven stress distribution within the specimen. Moreover a homogenous load distribution will depend on the quality of the specimen preparation; especially parallel end faces perpendicular to the specimen axis are desired. Furthermore the application of sufficiently high stresses related to service conditions is necessary. Only sufficiently high load levels will allow observing secondary and tertiary creep in reasonable testing times.

To fulfil these missions a testing device for high temperature compressive creep application was newly developed and is presented in this paper. Efficient inverse procedure to identify the creep law parameters is also introduced with two case studies of shaped and unshaped materials.

Section snippets

Construction and main features of the testing machine

A schematic diagram of the device for high temperature compressive creep testing is shown in Fig. 1. The whole setup is based on a spindle-driven universal testing machine of sufficient stiffness. An electrical furnace is inserted and equipped with silicon carbide heating elements. The furnace is composed of two symmetrical parts, held together by means of hinges in the rear and can be closed tightly with two buckles in the front. Two platinum-rhodium thermocouples are used to measure the

Norton-Bailey creep law and inverse estimation procedure

Generally a complete creep strain/time curve under a constant load and temperature can be divided into three stages. As illustrated in Fig. 3, the first stage is called primary creep and characterized by time-dependent creep strain rate ( ${\dot{ε}}_{c r}$ ) which decreases with time ( ${\ddot{ε}}_{c r} < 0$ ). Following it the secondary creep shows a constant creep strain rate which is the minimum one of all three stages. Tertiary creep is characterized by a rising strain rate ( ${\ddot{ε}}_{c r} > 0$ ) and leads to failure.

Many suggestions

Creep of burnt refractory bricks

Cylinder specimens as defined above have been prepared from commercially available burnt magnesia-chromite bricks (56.6 wt% MgO, 25.5 wt% Cr₂O₃). During the heating-up period, 100 N (equivalent to 0.1 MPa) was loaded axially on the specimen in order to fix it. Once the target temperature was reached, the extensometers were attached to the specimen before the loading procedure started. A total strain/time curve under constant load was obtained after 5 h. The measurements were carried out at 1100–1550

Conclusions

The new high temperature compressive creep testing device can be used to characterize the creep behaviour of shaped and unshaped materials at elevated loads. A preload of 0.2 MPa as applied for CIC measurements cannot be used during the heating up procedure for unshaped materials since these materials show inferior creep resistance especially at intermediate temperatures if they have not been fired before. Creep law parameters inversely identified by Levenberg–Marquardt method imply that both

Acknowledgements

Financial support by the Austrian Federal Government and the Styrian Provincial Government (Grant no. A4.15) (Österreichische Forschungsförderungsgesellschaft and Steirische Wirtschaftsförderungs-gesellschaft) within the K2 Competence Centre on Integrated Research in Materials, Processing and Product Engineering (MCL Leoben) in the framework of the Austrian COMET Competence Centre Programme is gratefully acknowledged.

References (26)

D. Gruber et al.
FEM simulation of the thermomechanical behaviour of the refractory lining of a blast furnace
J Mater Process Technol
(2004)
K. Andreev et al.
FEM simulation of the thermo-mechanical behaviour and failure of refractories-a case study
J Mater Process Technol
(2003)
L.A. Diaz et al.
Hot bending strength and creep behaviour at 1000–1400 °C of high alumina refractory castables with spinel, periclase and dolomite additions
J Eur Ceram Soc
(2009)
S. Jin et al.
Classification of thermomechanical impact factors and prediction model for ladle preheating
J Wuhan Univ Sci Technol
(2011)
S. Jin et al.
(2012)
S. Jin et al.
Thermomechanical steel ladle simulation including a Mohr–Coulomb plasticity failure model
RHI Bull
(2012)
E. Chen et al.
Methodology for thermomechanical analysis of brittle system
Am Ceram Soc Bull
(1985)
E. Chen et al.
Thermomechanical behavior and design of refractory linings for slagging gasifiers
Ceram Bull
(1985)
I. Matsumura et al.
Refractoriness under load and hot creep measurements
Taikabutsu Overseas
(1990)
EN 993-9
Methods of test for dense shaped refractory products. Determination of creep in compression
(1973)

A. Terzic et al.

Influence of the phase composition of refractory materials on creep

Sci Sinter

(2006)

V. Jokanovic et al.

Creep and microstructure in refractory materials

Am Ceram Soc Bull

(1998)

Y. Chien et al.

Effect of Cr₂O₃ on creep resistance of high alumina bauxite refractories

Ceram Bull

(1984)

Cited by (57)

Investigation of the role of impurities typical in secondary raw materials on the behaviour of high alumina castables - Part II: Influence on thermomechanical behaviour
2023, Open Ceramics
The present work investigated the effect of relatively small additions of impurities originating in secondary raw materials on the thermomechanical properties of alumina castables bonded with active alumina. Castable compositions containing impurities for testing were selected on the basis of predictions from thermochemical calculations of their phase composition and the formation of large amounts of liquid phase at the temperature of 1600 °C. Viscosity of the emerging liquid phase was also calculated. The selected alumina castables with impurities from the systems: CaO–SiO₂. Na₂O–SiO₂. Fe₂O₃–CaO. Fe₂O₃–CaO–SiO₂ and TiO₂–CaO–SiO₂ were tested for thermomechanical properties by analysing the effect of impurities on refractoriness under load (RUL) creep in compression (CiC) and resonant frequencies and damping changes versus temperature (RFDA). Real composition of tested material via high temperature X-ray diffraction (HT-XRD) were verified. The obtained results indicate that impurities introduced into the refractory castables in a total amount of 0,18 wt% (2% in the fraction 0–45 μm) and in the most unfavourable ratio from the point of view of phase equilibria have an observable effect on the properties of the material at elevated temperature. The obtained results show that in the case of introducing impurities into the material with mutual proportions of shares limiting the formation of the liquid phase. It will be possible to maintain the thermomechanical properties of alumina castables at a satisfactory level (compared to the pure one) from the industrial process perspective in which it will be used. The obtained results shed new light on the possibilities of the conscious use of secondary raw materials in refractory monolithics technology.
Numerical study on the nonlinear thermomechanical behaviour of refractory masonry with dry joints
2023, Engineering Structures
Refractory masonry with dry joints is widely employed as a protective lining in industrial applications requiring high-temperature treatments. The thermal and mechanical behaviour of alumina spinel refractory masonry is investigated for a wide range of mechanical loading conditions at ambient and high temperature up to 1500 °C within the framework of the ATHOR project. This paper discusses the different numerical analysis approaches for the simulation of the experimental results. Micro and macro modelling approaches show good agreement with the large scale uniaxial and biaxial compression tests for loading and unloading at the ambient temperature. Simulations carried out for large scale uniaxial and biaxial creep tests as well as biaxial relaxation tests at 1500 °C show good agreement. The numerical results indicate the ability of these modelling approaches to represent the complex thermomechanical behaviour of the refractory masonry. Both methods demonstrate an orthotropic and highly nonlinear behaviour of the refractory masonry as observed in the experimental campaign. The numerical outcome, validated with experimental results demonstrate compatibility between micro and macro modelling approach that can be employed to evaluate local and global behaviour of large industrial installations.
High temperature creep behavior of fireclay refractory and its masonry
2023, Construction and Building Materials
The creep behavior of refractory lining is crucial, which is determined by both material and masonry structure. In this paper, the effect of temperature and stress on the creep behavior of fireclay refractory and brick–mortar couplets were studied by using advanced uniaxial compression creep test. The creep behavior of fireclay refractory is sensitive to temperature due to formation of liquid phase. Meanwhile, the closure of discontinuous interfaces contributes to the increase of compressive creep in couplets. The masonry specimen with two mortar layers deforms less than that with one mortar layer, as thermal stresses are moderated by the additional mortar layer.
Experimental characterization of the nonlinear thermomechanical behaviour of refractory masonry with dry joints
2023, Construction and Building Materials
Refractory linings can withstand severe working conditions and are widely used for the linings of many high temperature applications, such as steel ladles. Some of these linings are built with dry joints. These structures undergo very complex responses in service conditions, making the prediction of its behaviour a challenging task. To develop robust numerical prediction tools, experimental validation is necessary. The present paper gathers a considerable database on the mechanical and thermomechanical characterization of refractory masonry with dry joints (up to 1500 °C). The test setups, measurement techniques, specimens, test procedures, and results of the large-scale experimental campaign are presented. The results of the tests help in understanding the complex thermomechanical behaviour of refractory linings. Aspect such as joints closure and reopening, dimensional and shape tolerances of the bricks, creep and stress relaxation are investigated. The results of these tests are essential for developing, calibrating, and validating efficient numerical models for the design and optimization of refractory masonry linings.
Research on creep damage model of high alumina bricks
2022, Ceramics International
Citation Excerpt :
The typical creep behavior is measured under constant temperature and constant stress conditions, including three stages. The first stage is characterized by time-dependent creep strain rate which decrease with time; and in the second stage, the creep strain rate maintains constant, so it is also called the steady state creep stage; tertiary creep is characterized by a rising strain rate until the material fails [14]. In order to describe the creep behavior of refractory materials, many creep models have been developed [9,13,15–19], among which the most widely used is the norton-bailey model.
In order to understand the mechanism of lining failure caused by creep, it is significant to evaluate the three creep stages of lining materials. In this paper, the three stages creep curves of 3 MPa, 3.25 MPa and 3.5 MPa for high alumina bricks at 1350 °C were obtained by high temperature creep testing equipment. Then the K-R model, the Sinh model and the Liu-Murakami model combined with the Garofalo formula are used to fit the measured creep curves. The results show that all the three models are suitable for the description of the creep behavior and the creep life prediction of high alumina bricks, and the Sinh-Garofalo model has the best fitting effect. Therefore, the Sinh-Garofalo model is selected for finite element modeling. The reliability of the model is verified by comparing the simulated and analytical values. Finally, by comparing the model with/without considering the creep damage, it is found that the creep damage model can be used to characterize the creep damage evolution process and predict the life of structures, which can be further used in the failure study of furnace lining.
Thermomechanical finite element modeling of steel ladle containing alumina spinel refractory lining
2022, Finite Elements in Analysis and Design
Refractory linings are used in steel ladles in iron and steel industry, to protect the vessel structure from the molten steel (with temperature above 1600 °C). To increase the durability of the refractory lining, researching the possible failure causes is of importance. In recent decades, alumina spinel refractories have become a common material in the barrel zone of steel ladles in direct contact with steel. In this regard, the current study employed unit-cell finite element modeling technique to investigate the irreversible material behavior of alumina spinel bricks. At first, a study on the joint size and friction effect was conducted. Then, three distinct constitutive material models were assigned to the working lining, each corresponding to an irreversible deformation mechanism. The Norton-Bailey creep model was used to simulate creep behavior, the Drucker-Prager yield criterion was used to simulate shear failure, and concrete damaged plasticity was used to describe tensile failure. The findings of the three models were compared to understand how each phenomenon affected the lining's and steel shell's stress-strain response. The simulations showed the occurrence location and time of each irreversible behavior. The effect of considering plasticity for the steel shell on the mechanical behavior of the refractory lining was also investigated, which showed a decrease of irreversible strains at the working lining.

View all citing articles on Scopus

View full text

Compressive creep testing of refractories at elevated loads—Device, material law and evaluation techniques

Abstract

Introduction

Section snippets

Construction and main features of the testing machine

Norton-Bailey creep law and inverse estimation procedure

Creep of burnt refractory bricks

Conclusions

Acknowledgements

J Mater Process Technol

J Mater Process Technol

J Eur Ceram Soc

Classification of thermomechanical impact factors and prediction model for ladle preheating

J Wuhan Univ Sci Technol

Thermomechanical steel ladle simulation including a Mohr–Coulomb plasticity failure model

RHI Bull

Methodology for thermomechanical analysis of brittle system

Am Ceram Soc Bull

Thermomechanical behavior and design of refractory linings for slagging gasifiers

Ceram Bull

Refractoriness under load and hot creep measurements

Taikabutsu Overseas

Methods of test for dense shaped refractory products. Determination of creep in compression

Influence of the phase composition of refractory materials on creep

Sci Sinter

Creep and microstructure in refractory materials

Am Ceram Soc Bull

Effect of Cr2O3 on creep resistance of high alumina bauxite refractories

Ceram Bull

Effect of Cr₂O₃ on creep resistance of high alumina bauxite refractories