Articles
  • The effect of titanium and zirconium oxides additives on thermal properties of magnesium aluminate spinel
  • Fahad Albanumay*, Naif Alqahtani, Basheer Alshammari, Haytham Alodan, Turky Alopily and Mohammed Muhawes

  • Material Science Research Institute, King Abdulaziz City for Science and Technology (KACST), Riyadh 11442, Saudi Arabia

Abstract

Magnesium aluminate spinel (MgAl2O4) composites were prepared by mixing the commercial Alumina and Magnesium Oxide as starting raw materials. Titanium oxide (TiO2) and Zirconium oxide (ZrO2) were used as additives. The mixtures were prepared as by milling of five different combinations using zirconia balls for 1 hour each. Then Cold Isostatic Press (CIP) at 200 MPa pressed the batches. The material properties, such as porosity and density, and thermal expansion of the composites were characterized by X-ray diffraction (XRD), Scanning Electron Microscopy (SEM), and the dilatometer measurements. MgAl2O4 ceramic composites are composed of spinel and garnet structures. The thermal expansion coefficients (CTE) of MgAl2O4 composites with and without TiO2 and ZrO2 additives under different temperature condition (25 oC to 1,300 oC) were characterized for spinel thermal expansion study and it shows that the comparison between the five different sample combinations at 1,300 oC, adding TiO2 or ZrO2 by small percentage gives the lowest CTE (9.89E-06, 1.02E-05) respectively, but increasing ZrO2 increases the CTE.


Keywords: Magnesium aluminate spinel, Thermal expansion, Titanium oxide, Zirconium oxide

introduction

Magnesium aluminate spinel (MgAl2O4) has been known as a technologically vital material which has many applications in many different fields, such as in high temperature ceramics [1], fabricating transparent ceramics [2, 3], and catalyst support [4, 5], nuclear waste management applications [6], humidity sensors [7] and cement castables [8].
Magnesium aluminate spinel is an important refractory material because of its excellent properties such as high melting point that reach 2,105 oC, low thermal expansion, high thermal spalling, and corrosion resistance [9]. Synthesis MgAl2O4 is very challenging from the solid-state reaction route since it needs repetitive grinding and calcination steps. Some common methods, such as plasma spray decomposition of oxide and hydrothermal synthesis can be used to prepare high quality pure material. But these methods do not get a lot of attention in the commercial circle because of the use of expensive raw materials and the requirement of many processing steps [10].
It is important for the side walls, the checker work of glass tank furnace regenerators and the bottom of steel-teeming ladles. Thermal properties are important for all these applications. Increasing the temperature of a material increases the amplitude of vibration of the atoms and results in on overall increase in the volume of the material. This expansion is very critical for the structural integrity and spoiling property of the material [11]. The crystallographic structure of the MgAl2O4 spinel is simple cubic with eight formula units in one cubic unit cell [12]. The effect of particle size distribution of the spinel on the ceramic mechanical properties and thermal shock performance has been previously studied and mechanical properties of composites decreased significantly with increasing spinel content due to thermal expansion mismatch [13].
The crystallographic structure of the MgAl2O4 spinel has been reported by Kingrey [14] and illustrated in Fig. 1. The generic formula of this spinel group is AB2O4, which “A” represents a divalent metal ion such as magnesium, iron, nickel, manganese and zinc. The “B” represents trivalent metal ions such as aluminum, iron, chromium and/or manganese. In this study, the “A” is the magnesium and “B” is aluminum and the spinel structure is named after the mineral spinel (MgAl2O4) [15]. The positions of the A ions are nearly identical to the positions occupied by carbon atoms in the diamond structure. This could explain the relatively high hardness and high density typical of this group. The arrangements of the other ions in the structure conform to the symmetry of the diamond structure. However, they disrupt the cleavage as there are no cleavage directions in any member of this group.
The coefficient of thermal expansion (CTE) is a fundamental engineering material property that used to express the dimensional change (volume, length, etc.) of a material in response to temperature change. The thermal expansion/ contraction behavior due to daily and seasonal temperature changes plays an important role on the degree of opening/closing of transverse cracks in concrete structures [16].
The objective of this study is the investigation the influence of TiO2 and ZrO2 additives on the spinel material properties and the thermal expansion. The MgAl2O4 spinel without additives will be set as a baseline composite that will be compared with the spinel with TiO2 and ZrO2 additives. Spinel has high melting points that allow measurements to be made over wide temperature ranges. Five different spinel compositions, were tested for thermal expansion characteristics.

Fig. 1

Crystallographic structure of MgAl2O4 spinel (Redrawn from Kingrey [14]).

materials and experiment method

Commercial Alumina (Al2O3 with 94% purity) and Magnesium Oxide (MgO with 99% purity) were used as raw materials of the MgAl2O4 spinel. Titanium oxide (TiO2 with 99% purity) and Zirconium oxide (ZrO2 with 99% purity) were used as additives. Five samples were prepared for this study, sample A is a mixture of Al2O3 and MgO without any additives, sample B is a mixture of the MgAl2O4 spinel adding 2.44% Titanium oxide of the mixture weight, sample C is a mixture of the MgAl2O4 spinel adding 4.76% Titanium oxide of the mixture weight, sample D is a mixture of the MgAl2O4 spinel adding 2.44% Zirconium oxide of the mixture weight, and sample E is a mixture of the MgAl2O4 spinel adding 4.76% Zirconium oxide of the mixture weight. Table 1 shows the particle size analysis for MgO, Al2O3 and the mixture after milling for 1 h. Table 2 shows all samples with their specific weights. The five batches of 250 gm powder was prepared by milling of five different combinations using 10 small ZrO2 balls for one hour each. Then, the batches were pressed by Cold Isostatic Press (CIP) at 200 MPa.
All prepared samples then were slowly heated with the rate of 1 oC per minute to 100 oC with holding time 60 minutes, then to 1,750 oC with rate of 5 oC per minute and holding time 60 min.
For particle size analysis, MgO and Al2O3 powders were analyzed before and after milling the mixtures for one hour using “Shimadzu SALD-2300”. The porosity (ϕ) and density (ρ) of the sintered samples was determined by the Archimedes method with water as liquid media. The following equations used to calculate the porosity and the permeability for the spinel:
 

 
where:
WD = is the wt. of dried sample (g).
Ws = is the wt. of sample measured in water (g).
Ww= is the wt. of water soaked after 24 h (g).
 
Thermal expansion was investigated using dilatometer (DIL 402 PC, NETZSCH Geratebau GmbH, Germany) with temperature range 25-1,300 oC [17].
To calculate the Coefficient of Thermal Expansion (CTE) the following equation were used [18]:
 

 
where:
αl = thermal expansion coefficient for the parameter l, K-1
εl = dl/l is the strain for the parameter l, and,
T = is the temperature, K.
 
The effect of adding TiO2 and ZrO2 to the spinel mixture on the coefficient of thermal expansion can be calculated using equation (X) by setting sample A (when we have MgAl2O4 spinel without any additives) as a baseline to compare it with the other samples.
 

 
For thermal shock resistance study, the samples were exposed to a temperature of 1,000 oC for 15 min, each thermal shock cycle involved heating of the sample to 1,000 oC for 15 min in an electric furnace, the first two cycles quenched at room temperature for 15 min, and the third cycle quenched in cold water [19].
Finally, for the X-ray analysis, the sintered samples were ground into powder using milling. The powder was placed on a sample holder and was irradiated by a monochromatic X-ray beam from an X-ray tube. The samples were examined using scanning electron microscopy (SEM) (MiniFlex 600, Rigaku, Japan) operated at 20 KV. Cu-Kα radiation passed through nickel filter was used. The range of scanning angle (2θ) used was 0o-70o. XRD of 1,750 oC for one hour sintered sample was done to see the spinel phase.

Table 1

Results of particle size analysis

Table 2

Batch composition of samples (Raw materials and additives)

results and discussion

The results of XRD, porosity and density, thermal shock resistance, scanning electron microscopy (SEM), and the thermal expansion of the spinel mixtures are presented in this section. All the analysis and integration of the results obtained previously discussed also in this section.
 
Phase composition by X-Ray Diffraction (XRD)
Fig. 2 shows the XRD patterns of the samples after sintering at 1,750 oC for 1 h and milling for 1 h. In the first combination, Al2O3 was transformed completely to the magnesium aluminate spinel but there are still small patterns of MgO, and the same happened in the second combination when TiO2 added but the MgO patterns were smaller than what happened in the first combination.
X-Ray Diffraction (XRD) patterns of non-additive and additive contained batches are shown in Fig. 2. The XRD pattern of 1,750 oC sintered compositions shows the presence of spinel in both non-additive and additive contained batches. As the percentage of TiO2 was increased, amount of spinel formation was also increased in all batches, compared to that of no additives and ZrO2 contained batches. The highest spinel peak intensity was observed in TiO2 containing compositions. This clearly indicates a higher rate of spinel formation occurring in these compositions. The presence of only spinel phase was observed in 5% TiO2 containing sample, this observation of complete solid solubility of free MgO in spinel phase with TiO2 finds similarity with the work of Sarkar and Bannerjee [20]. However, small peaks of unreacted phases were detected in all the other samples, indicating incompletion of spinel formation reaction in the batch.
This is supported by comparing with the work of Quénard et al. [21]. In addition to the spinel and ZrO2 phases, the presence of a small amount of MgO. The relative intensity of the (200) MgO peak is similar whatever the ZrO2 content.
When TiO2 was increased in the third combination, MgO and Al2O3 were completely transformed to magnesium aluminate spinel and there was small pattern of TiO2. When ZrO2 was added, Al2O3 was transformed completely to the magnesium aluminate spinel but there are still small patterns of MgO and ZrO2, and when the ZrO2 percentage was increased in the fifth combination the patterns of MgO and ZrO2 were even bigger than what happened in the fourth combination.
 
Porosity and density
Porosity and Density of all the sintered samples at 1,750 oC for 1 h and 200 MPa are given in Table 3 and shown in Fig. 3. In first combination, which is without any additives, the porosity has the lowest value (0.36%). When TiO2 added to the mixture, the porosity increased (2.30%), but when the percentage of TiO2 increased the porosity decreased (1.23%). For ZrO2, the porosity increases with higher percentage of ZrO2 (0.66 at 2.5% ZrO2 and 0.92% at 5% ZrO2), but overall ZrO2 gives less porosity than TiO2 as additives.
For Density, there isn’t much different with additives at different percentages, but using TiO2 gives a lower value of density than using ZrO2.
 
Thermal shock resistance and thermal expansion
No visible cracks or damaged surface of the samples were found as a results of thermal shock resistance test. Thermal expansion coefficient depends mainly on its component materials. Spinel compositions (Table 4) shows no significant spinalization reaction up to 1,000 oC. Only a small, gradual increase in expansion values is observed with increasing temperature. Above 1,000 oC, expansion values improve sharply, which marks the starting of spinel formation reaction.
The addition of TiO2 and ZrO2 above 1,000 oC gives a better homogeneous spinel. On increase in its growth, spherical shape particles are formed which help to achieve the crystallization of those compositions at temperatures lower than that of model spinel without adding these oxides. The fluctuation in the thermal expansion coefficient value, when the temperature is below 1,000 oC, it believes to be a result of temperature variation and rearrangement of grains.
Fig. 4 shows the comparison between the five different sample combinations at 1,300 oC, adding TiO2 or ZrO2 by small percentage gives the lowest CTE (9.89E-06, 1.02E-05) respectively, but increasing ZrO2 increases the CTE.
 
Scanning Electron Microscopy (SEM) results
The morphology of the surface of the magnesium aluminate spinel - with and without additives -are analyzed by SEM observations as shown in Fig. 5. As can be seen from this figure, the baseline sample which is the MgAl2O4 spinel without additives have smallest crystals particles (see Fig. 5(a)) that maybe contributed to that milling grind the particles and made homogenous small size particle. With adding a TiO2 to the spinel (see Fig. 5(b)), the porosity increased because of the excessing of the fractures that induces which led to more connected accessible channels that led the fluid to occupied that space. In Fig. 5(c), the amount of TiO2 has been doubled and the porosity has been increased due to the induced fractured and air bubbles that is generated as a results of liberated gases from adding the TiO2 to the mixture similar behavior has been reported by Saleh and Hassen [22]. Not that far from adding TiO2, ZrO2 has been added to the spinel mixture to form the sample D as we can see in Fig. 5(d). It shows an increase in porosity and fractures was clearly presented in the structure. By doubling the amount of ZrO2 (Fig. 5(e)), the porosity has been increased allowing more isolation property and more fluid can be stored in this pours. Furthermore, it is clear that the adding TiO2 provided more homogenous dispersion and distribution of the metal oxide particle than ZrO2. This homogeneity agrees with thermal expansion values where the reduction has been noticed in the case of adding TiO2. At 2.44%, however, farther addition slightly improve the thermal expansion this could be due to agglomeration of the particle inside the spinel. In the other hand, this mechanism and behavior of agglomeration has been noticed in the case of ZrO2 even at 2.44% weigh. These results indicate the advantage of adding TiO2 and ZrO2 for improve the thermal properties of MgAl2O4 spinel. The optimal loading of TiO2 is 2.44% in this study.

Fig. 2

Phase analysis study (XRD) of (200 MPa, 1,750 oC-1 h) sintered samples: (a) First Combination: Sample A, (b) Second Combination: Sample B, (c) Third Combination: Sample C, (d) Fourth Combination: Sample D, (e) Fifth Combination: Sample E.

Fig. 3

Comparison between different combinations: (a) Porosity (b) Density.

Fig. 4

Comparison of CTE for the five combinations at 1,300 oC.

Fig. 5

SEM photograph of MgAl2O4 samples: (A) MgAl2O4 without additives (baseline sample); (B) MgAl2O4 with 2.44% TiO2; (C) MgAl2O4 with 4.76% TiO2; (D) MgAl2O4 with 2.44% ZrO2; (E) MgAl2O4 with 4.76% ZrO2.

Table 3

Experimental values of Porosity and Density with the theoretical density values

Table 4

Results of CTE at different temperatures and effect % of adding TiO2 and ZrO2

conclusion

In this study, different loading titanium oxide (TiO2) and zirconium oxide (ZrO2) were used as additives to fabricate magnesium aluminate spinel composites. The thermal expansion coefficients of these composites and porosity were analyzed. For pure spinel (without any additives), the porosity has the lowest value. However, adding TiO2 to the mixture increases the porosity, but when the percentage of TiO2 increased the porosity appeared to be decreasing. In other hand, increasing loading of ZrO2 leads to the increase in the porosity. Generally, ZrO2 gives less porosity than TiO2 when they added to the mixture.
In all compositions there are no significant spinalization reaction up to 1,000 oC, but above 1,000 oC expansion values improve sharply, which marks the starting of spinel formation reaction. The addition of TiO2 and ZrO2 above 1,000 oC gives a better homogeneous spinel. At 1,300 oC, adding TiO2 or ZrO2 by small percentage gives the lowest CTE.

References
  • 1. I. Ganesh, S. Bhattacharjee, B.P. Saha, R. Johnson, K. Rajeshwari, R. Sengupta, M.V. Ramana Rao, and Y.R. Mahajan, Ceramics International 28[3] (2002) 245-253.
  •  
  • 2. A.F. Dericioglu and Y. Kagawa, J. Euro. Ceram. Soc. 23[6] (2003) 951-959.
  •  
  • 3. M. Shimada, T. Endo, T. Saito, and T. Sato, Mater. Lett. 28[4-6] (1996) 413-415.
  •  
  • 4. J. Guo, H. Lou, H. Zhao, D. Chai, and X. Zheng, Appl. Catal. A: Gen 273[1-2] (2004) 75-82.
  •  
  • 5. J. Guo, H. Lou, H. Zhao, X. Wang, and X. Zheng, Mater. Lett. 58[12-13] (2004) 1920-1923.
  •  
  • 6. M. Beauvy, C. Dalmasso, C. Thiriet-Dodane, D. Simeone, and D. Gosset, Nucl. Instrum. Meth. Phys. Res. B 242[1-2] (2006) 557-561.
  •  
  • 7. G. Gusmano, G. Montesperelli, E. Traversa, A. Bearzotti, G. Petrocco, A. D’Amico, and C.D. Natale, Sens. Actuators B 7[1-3] (1992) 460-463.
  •  
  • 8. S. Mukhopadhyay, S. Ghosh, M.K. Mahapatra, R. Mazumder, P. Barick, S. Gupta, and S. Chakraborty, Ceramics International 28[7] (2002) 719-729.
  •  
  • 9. A. Ghosh, S.K. Das, J.R. Biswas, H.S. Tripathi, and G. Banerjee, Ceramics International 26[6] (2000) 605-608.
  •  
  • 10. I. Ganesh, R. Johnson, G.V.N. Rao, Y.R. Mahajan, S.S. Madavendra, and B.M. Reddy, Ceramics International 31[1] (2005) 67-74.
  •  
  • 11. R.D. Maschio, B. Fabbri, and C. Fiori, Inds. Ceramics 8[2] (1988) 121-126.
  •  
  • 12. W.H. Bragg, Philosophical Magazine and J. of Sci. 30[170] (1915) 305-315.
  •  
  • 13. C. Aksel and F.L. Riley, J. Euro. Ceram. Soc. 23[16] (2003) 3079-3087.
  •  
  • 14. W.D. Kingrey, in “Introduction to Ceramics” (John Wiley and Sons, 1960) p. 1.
  •  
  • 15. A.M. Pachpinde, in “Ferrite Catalysts” (Lulu Publication, 2017) p.1.
  •  
  • 16. Y.H. Huang, in “Pavement analysis and design” (Engle- wood Cliffs, 1993) p. 1.
  •  
  • 17. ASTM E228-17, ASTM International, West Conshohocken, PA, 2017
  •  
  • 18. R.I Belousov and S.K Filatov, Glass Phys. Chem. 33[3] (2007) 271-275
  •  
  • 19. ASTM C1525-18, ASTM International, West Conshohocken, PA, 2018
  •  
  • 20. R. Sarkar and G. Bannerjee, Euro. Ceram. Soc. 20[12] (2000) 2133-2141
  •  
  • 21. O. Quénard, C. Laurent, A. Peigney, and A. Rousset, Mater. Res. Bull. 35[12] (2000) 1967-1977
  •  
  • 22. Q. Saleh and B. Hassen, Iraqi Journal of Physics. 15[34] (2017) 114-122.
  •  

This Article

  • 2020; 21(6): 683-689

    Published on Dec 31, 2020

  • 10.36410/jcpr.2020.21.6.683
  • Received on Jul 8, 2020
  • Revised on Aug 23, 2020
  • Accepted on Sep 3, 2020

Correspondence to

  • Fahad Albanumay
  • Material Science Research Institute, King Abdulaziz City for Science and Technology (KACST), Riyadh 11442, Saudi Arabia
    Tel : +966114883555 ex. 2794 Fax: +966114813526

  • E-mail: falbanumay@kacst.edu.sa