Thermal Shock Behaviour of Magnesia Hercynite-Spinel Composite Refractories

Size: px

Start display at page:

Download "Thermal Shock Behaviour of Magnesia Hercynite-Spinel Composite Refractories"

Clara Phillips
5 years ago
Views:

1 Journal of Multidisiplinary Engineering Siene and Tehnology (JMEST) Thermal Shok Behaviour of Magnesia Herynite-Spinel Composite Refratories Tuba (Aksoy) Bahtlı* Nemettin Erbakan University, Faulty of Engineering and Arhiteture, Department of Metallurgial and Materials Engineering Konya, Turkey C Aksel Anadolu University, Faulty of Engineering, Department of Material Siene and Engineering Eskisehir, Turkey Abstrat The effets of inorporations of spinel into MgO herynite (M H) ompositions to improve thermal shok behaviour were investigated. R, R, R parameters were alulated, and thermal shok tests were performed. The frature surfaes of materials before and after thermal shok test were examined using SEM. Parameters improving thermal shok behaviour of M H S materials at 500 and 1000 C are interlinking and arresting or deviation of miroraks when reahing the spinel and herynite grains or pores, oexistene of intergranular and transgranular fratures leading to longer servie life. Keywords MgO, herynite, FeAl 2 O 4, spinel, MgAl 2 O 4, omposite, refratories, thermal shok, Hasselman parameters I. INTRODUCTION In reent years the prodution of MgO-spinel briks and MgO-herynite briks as alternatives to MgO-hrome brik have beome widespread [1]. Magnesia based refratories have been widely used in ement rotary kilns and steel ladles due to their high melting point, no toxiity and good resistane to basi slags and linker phases. However, they have some disadvantages, suh as high thermal ondutivity, poor thermal shok resistane and infiltration resistane against hemial attak. In order to improve these properties, some metal oxides and ompounds suh as SiO 2, CaO, Al 2 O 3, MgAl 2 O 4, Cr 2 O 3, ZrO 2, CaZrO 3 and FeAl2O4 have been used as doping agent. These oxides may reat with MgO to form a seond phase whih improves the proess of sintering [2-4]. Magnesium aluminate spinel (MgAl 2 O 4 ) is an important omponent of magnesia based refratories [5]. MgO based spinel briks are used in the ooling zone and in the upper side of the sintering zone of ement rotary kilns [6, 7]. MgO spinel refratories ontaining a small amount of spinel display high resistane against thermo-mehanial stresses formed at high temperatures and resultant thermal shoks [8]. The resistane to sudden transient temperature hange, thermal shok, of refratory materials is a subjet of onsiderable pratial interest whih has been explored in detail over many years [9]. Magnesia-herynite briks have better oatability than magnesia-spinel briks. Also have better thermal shok resistane, good orrosion resistane as magnesia-spinel briks [10]. In this study, the thermal shok resistane and therefore the servie life MgO-herynite omposite refratories were intended to be improved by spinel addition. The relationships between the mehanisms ausing improvements in the thermal shok parameters of omposite refratories and strength and frature surfaes of the materials before and after thermal shok test were examined. Related thermal shok parameters (Hasselman parameters) expressing the ability of a material to resist these thermal stresses were defined. II. MATERIAL AND METHOD A. Material Preparation Reipes were prepared by inorporating 5, 10 and 20% MgAl 2 O 4 spinel (S) by weight into MgO (M) based ompositions ontaining 5, 10 and 20 30% FeAl 2 O 4 herynite (H) by weight. Those reipes were shaped as ~230 x 115 x 65 mm3 dimensions by applying ~127.5 MPa (300 bar) press pressure. Then, sintered nearly at 1550 C temperature. The materials were ut into 2.5 m 2.5 m 15 m bar aording to the standards for mehanial testing and thermal shok resistane test. B. Hasselman Parameters Hasselman parameters, whih determine the frature resistane of materials due to thermal stress/shok and used in estimation of maximum thermal resistane were alulated by following formulas: JMESTN

2 f 1 R E (1) where σ is the strength, α is the thermal expansion oeffiient, is the Poisson s ratio and E is the elasti modulus of materials the bar thikness. The thermal shok resistane fators (R) expressing the ability of a material to resist these thermal stresses, are defined in terms of these mehanial and physial properties. Journal of Multidisiplinary Engineering Siene and Tehnology (JMEST) E 1 R' ''. (2) 2 (1 ) f E WOF R' '''. (3) 2 (1 ) f where ɣ WOF is the work of frature of materials. A seond set of parameters (R and R ) is appliable to the ase of severe thermal environments, where thermal stress frature annot be avoided and the major requirement is to minimize the extent of rak propagation [11]. C. Mehanial Testing-Strength The materials were ut into 2.5 m 2.5 m 15 m bar aording to the standards. The strength(σ) were determined before and after thermal shok test by the 3-point bending method in Instron 5581 [12, 13]. D. Thermal Shok Test Thermal-shok was arried out using a standard quenh proedure. Bars were suspended using wire in a vertial tube furnae whih was heated to 500 and 1000 C temperatures. Thermal shok tests were performed by sudden ooling in water from 500 and C to 25 8C room temperature. Bars were dried in an oven at 110 C, and broken by 3-point bending test mentioned in C (Mehanial testing). In other words, strength values determined in onnetion with thermal shok temperatures. E. Mirostruture Analysis SEM (sanning eletron mirosope) studies arried out with Zeiss Evo 50 devie, the frature surfaes of the materials before and after thermal shok test were investigated. III. RESULTS AND DISCUSSION A. Thermal Shok Parameters (Hasselman Parameters) When the results of R parameter values were analyzed in Figure 1, generally R parameter values of MgO- herynite-spinel materials were higher than the MgO-herynite materials. The highest R parameter value was ahieved in M-10%H-20%S material. Fig. 1. R parameters of MgO-herynite-spinel Composite Refratoies. R parameter shows high resistane to the the rak propagation and the higher R values, the higher thermal shok resistane of material. MgO-herynite omposite materials had higher R thermal shok values than pure MgO (iron ontaining- MgO) material, and those values inreased with inreasing the amount of Herynite addition. R value for the omposite ontaining 20% Herynite was approximately 10 times more than pure material (Figure 2). Fig. 2. R parameters of MgO-herynite-spinel Composite Refratoies. Generally, R thermal shok parameter values of MgO-herynite-spinel omposite materials ontaining different ratios of spinel were higher than MgO-spinel materials. The highest R value was obtained M-20%H-20%S omposite material (Figure 2). R parameter is used to estimate the resistane to the rak propagation of the material when subjeted to thermal shok. MgO-herynite omposite materials had higher R thermal shok values than pure MgO material, and those values were signifiantly higher with inreasing the amount of herynite addition (Figure 3). For example, M-20%H omposite materials showed 20 times higher resistane to rak propagation than pure MgO. M-H-S materials had higher R thermal shok parameter values than MgO-Herynite materials. The highest R value was reahed in M-10%H-5%S JMESTN

3 material, approximately 4.6 times enhanement was observed ompared with M-5%H material. Fig. 3. R of MgO-herynite-spinel Composite Refratoies. B. Thermal Shok Test The strength and the strength ratio values of the produed iron ontaining-mgo, MgO-herynite and MgO-herynite-spinel omposite refratory materials were determined after the thermal shok test, aording to different thermal shok temperatures (Figure 4, 5). Fig. 4. Strength values of M H and M H S omposite refratories as a funtion of thermal shok temperatures. Journal of Multidisiplinary Engineering Siene and Tehnology (JMEST) In spite of the signifiantly redution in the strength value of MgO (~ 4.1-fold) was observed, the redution in the strength values of MgO-herynite refratory materials were muh less than MgO. During ooling from prodution temperatures, the large differene in the thermal expansion oeffiient between MgO and herynite ( C: α MgO = 13.6 x 10-6 C -1, α Herynite = 9.0 x 10-6 C -1 ) generates large tensile stress around herynite grains, resulted in interlinked miroraks that ould stop the rak propagation ourred by thermal shok. After thermal shok test at 1000 C, M-5%H, M-10%H and M-20%H refratory materials had higher thermal shok resistane than pure MgO. Those sudden derease in strength of MgO aused by the lak of spinel and herynite reinforements. This means that miroraks, ourred by mismath of thermal expansion oeffiients of MgO, herynite and spinel ( C: α spinel = 7.6 x 10-6 C -1 ), ouldn t been interlinked to eah other. So, the length and the amount of big miroraks in MgO inreased signifiantly. Either at 500 C or 1000 C, the strength values of M-H-S materials generally had less or lose to pure- MgO and MgO-herynite materials. M-5%H-5%S had the highest strength at 1000 C thermal shok test temperature among M-H-S materials. The mismath of thermal expansion oeffiients of MgO, herynite and spinel aused interlinked miroraks. The big miroraks, ourred after thermal shok test, progressed in short distane in M-H-S materials, then the redution in strength of those materials were limited. After thermal shok test at 500 C, although the strength ratio of iron ontaining-mgo was ~50%, the proteted strength ratio values of MgO-herynite omposite materials were >80%. After thermal shok test at 1000 C, %, the proteted strength ratio values of iron-ontaining MgO and MgO-herynite omposite materials were obtained as ~ 20% and >40% respetively. The highest proteted strength ratio value was reahed (~%50) in M-10%H refratory material among M-H materials. As the proteted strength ratio values of M-H-S omposite materials were examined in Figure 5, either at 500 C or 1000 C, all materials had less strength loss, higher strength ratio than pure MgO and M-H materials. The highest proteted strength ratio value was reahed (79%) in M-5%H-5%S material among M-H-S materials. In M-H and M-H-S materials whose initial strength values were lower, it wasn t observed big redution in strength values unlike pure MgO. The reasons of those onditions were thought that the length and the amounts of miroraks ourred as a funtion of thermal shok test in M-H materials were smaller than MgO and those miroraks progressed in shorter distane. JMESTN

Journal of Multidisiplinary Engineering Siene and Tehnology (JMEST) (a) Fig. 5. Strength ratio values of M H and M H S omposite refratories as a funtion of thermal shok temperatures. C.

In general, before thermal shok tests, the pure iron-ontaining MgO material was exposed to transgranular raks before thermal shok test.

4 Journal of Multidisiplinary Engineering Siene and Tehnology (JMEST) (a) Fig. 5. Strength ratio values of M H and M H S omposite refratories as a funtion of thermal shok temperatures. C. Eletron Mirosope Charaterizaton The frature surfae images of iron ontaining- MgO refratory material before and after thermal shok test were given in Figure 6. In general, before thermal shok tests, the pure iron-ontaining MgO material was exposed to transgranular raks before thermal shok test. In partiular, the raks of medium-large-sized rystal grains were transgranular raks type predominately. Due to this type of rak harater, the material had relatively high resistane against the rak beginning, but the resistane to rak propagation was low (Figure 6-a). After thermal shok test at 1000 C, transgranular raks type were predominately in iron ontaining- MgO (Figure 6-b). Beause of its high thermal expansion oeffiient, this material lost ~76% of its strength. The sudden derease in strength of MgO aused transgranular type raks in this material. The frature surfae image of the refratory omposite material formed by addition of 10% herynite to iron-ontaining MgO before and after thermal shok test were given in Figure 7. (b) Fig. 6. The frature surfae image of iron-ontaining MgO material a) before thermal shok test, b) after thermal shok test at 1000 C. Large tensile stresses and miroraks were ourred in the struture due to thermal expansion oeffiient differenes between the MgO and herynite grains during ooling after the sintering of omposite materials. Miroraks in the material were onneted to eah other, stopped when they herynite grains or hanged their diretion during the breaking of material. Aording to frature surfae image of MgO-herynite material, either transgranular raks or intergranular raks were observed before thermal shok tests (Figure 7-a). After thermal shok test at 1000 C, in M-10%H material, either intergranular (more than transgranular) or transgranular raks were observed (Figure 7-b). M-10%H material had smaller MgO grains than MgO material, then the big miroraks ourred after thermal shok tets progressed between those small MgO grains (intergranular raks). intergranular raks type aused higher thermal shok resistane of M-10 %H materials than pure-mgo. JMESTN

Journal of Multidisiplinary Engineering Siene and Tehnology (JMEST) (a) (a) (b) Fig. 7.

The frature surfae image of the M-5%H-5%S refratory omposite material, that had the highest proteted strength ratio value (79%), before and after thermal shok test were given in Figure 8.

After thermal shok test at 1000 C, transgranular raks type were predominant in M-5%H-5%S refratory material, but small amounts of intergranular raks were, also (Figure 8-b).

The frature surfae image of M-5%H-5%S refratory material a) before thermal shok test, b) after thermal shok test at 1000 C. IV.

5 Journal of Multidisiplinary Engineering Siene and Tehnology (JMEST) (a) (a) (b) Fig. 7. The frature surfae image of M-10%H refratory material a) before thermal shok test, b) after thermal shok test at 1000 C. The frature surfae image of the M-5%H-5%S refratory omposite material, that had the highest proteted strength ratio value (79%), before and after thermal shok test were given in Figure 8. Before thermal shok test, either intergranular or transgranular raks were observed in M-5%H-5%S refratory material (Figure 8-a). After thermal shok test at 1000 C, transgranular raks type were predominant in M-5%H-5%S refratory material, but small amounts of intergranular raks were, also (Figure 8-b). Those hange in raks type after thermal shok test was thought as an important parameter affeted the proteted strength ratio. (b) Fig. 8. The frature surfae image of M-5%H-5%S refratory material a) before thermal shok test, b) after thermal shok test at 1000 C. IV. CONCLUSIONS Either at 500 C or 1000 C, the strength values of M-H-S materials generally had less or lose to pure- MgO and MgO-herynite materials. M-5%H-5%S had the highest strength at 1000 C thermal shok test temperature among M-H-S materials. Also, either at 500 C or 1000 C, all materials M-H- S had less strength loss, higher strength ratio than pure MgO and M-H materials. M-5%H-5%S material had the highest proteted strength ratio value was reahed (79%) in among M-H-S materials. The mismath of thermal expansion oeffiients of MgO, herynite and spinel ( C: α MgO = 13.6x10-6 C- 1, α Herynite = 9.0x10-6 C -1, α spinel = 7.6x10-6 C -1 ) aused interlinked miroraks that ould avoid the propagation of big miroraks, ourred after thermal shok test, in M-H-S materials. After thermal shok test at 1000 C, transgranular raks type were predominant in M-5%H-5%S JMESTN

6 refratory material that was thought as an important parameter affeted the proteted strength ratio. As the results were ompared, it s observed that the thermal stress/shok parameters (Hasselman parameters) and results of thermal shok tests were ompatible. In other words, R ve R parameters indiating the resistane to propagation of rak, and R parameter expressing the ability of a material to resist these thermal stresses were indentified as reliable parameters ould be used to determine the thermal shok resistane of MgO, M-H and M-H-S omposite refratory materials. ACKNOWLEDGMENT This study was partly supported in part TUBITAK under projet no: 106M394. We would like to express our gratitude to all employees and personnel of Konya Seluklu Krom Magnezit Tugla A.S. for their support and the supplied equipments and raw materials. REFERENCES [1] Pablo M. Botta, Estaban F. Aglietti, José M. Porto López, Mehanohemial synthesis of herynite. Materials Chemistry and Physis, vol. 76, , [2] Peng C, Li N, Han B. Effet of ziron on sintering omposition and mirostruture of magnesia powder. Si Sinter 2009;41:11 7. [3] Han B, Li Y, Guo C, Li N, Chen F. Sintering of MgO refratories with added WO3. Ceram Int 2007;33: [4] Bartha P, Klishat HJ. Classifiation of magnesia briks in rotary ement kilns aording to Journal of Multidisiplinary Engineering Siene and Tehnology (JMEST) speifiation and servie-ability. ZKG Int 1994;47: [5] Pablo M. Botta, Estaban F. Aglietti, José M. Porto López, Mehanohemial synthesis of herynite. Materials Chemistry and Physis, vol. 76, , [6] Benbow J. Cement kiln refratories-down to basis. Ind Miner, 268, 37 45, [7] Aksel C, Riley FL. Magnesia spinel (MgAl 2 O 4 ) refratory erami omposites. In: Low IM, editor. Cerami matrix omposites: mirostruture, properties and appliations. USA: Woodhead Publishing Limited and CRC Press LLC; , [8] Eusner GR, Hubble DH. Tehnology of spinel bonded perilase brik. J Am Ceram So, 43:292 6, [9] Kingery, W. D., Fators affeting thermal stress resistane of erami materials, J. Am. Ceram. So., 1955; 38(1): [10] Lui Huillin, Properties of magnesiaherynite brik. China s Refratories, 17 [1], [11] Hasselman, D.P.H., Thermal stress resistane parameters for brittle refratory eramis: a ompendium, Am. Ceram. So. Bull., 49 [12], , [12] Standard Test Methods for flexural strength of advaned eramis at ambient temperature. In Annual Book of ASTM Standards, Designation: C , 15.01, , [13] Standard Test Methods for flexural properties of unreinfored and reinfored plastis and eletrial insulating materials. In Annual Book of ASTM Standards, Designation: D790M-86, 08.01, , JMESTN

The Effect Of Spinel Addition On The Mechanical Behaviours Of MgO-Hercynite Refractories

$The Effect Of Spinel Addition On The Mechanical Behaviours Of MgO-Hercynite Refractories$ The Effect Of Spinel Addition On The Mechanical Behaviours Of MgO-Hercynite Refractories Tuba (Aksoy) Bahtlı* Necmettin Erbakan University, Faculty of Engineering and Architecture, Department of Metallurgical