Characterization of the alumina–zirconia ceramic system by ultrasonic velocity measurements

M A TE RI A L S CH A RACT ER IZ A TI O N 60 ( 20 0 9 ) 8 7 5 –8 8 1

Characterization of the alumina–zirconia ceramic system by ultrasonic velocity measurements Hector Carreon⁎, Alberto Ruiz, Ariosto Medina, Gerardo Barrera, Juan Zarate Instituto de Investigaciones Metalurgicas, Edif. “U” Ciudad Universitaria, Morelia, Mich. 58000-888, Mexico

AR TIC LE D ATA

ABSTR ACT

Article history:

In this work an alumina–zirconia ceramic composites have been prepared with α-Al2O3

Received 24 May 2006

contents from 10 to 95 wt.%. The alumina–zirconia ceramic system was characterized by

Received in revised form

means of precise ultrasonic velocity measurements. In order to find out the factors affecting

27 October 2007

the variation in wave velocity, the ceramic composite have been examined by X-ray

Accepted 15 February 2009

diffraction (XRD) and (SEM) scanning electron microscopy. It was found that the ultrasonic velocity measurements changed considerably with respect to the ceramic composite

Keywords:

composition. In particular, we studied the behavior of the physical material property

Alumina–zirconia ceramic

hardness, an important parameter of the ceramic composite mechanical properties, with

Indentation hardness

respect to the variation in the longitudinal and shear wave velocities. Shear wave velocities

Ultrasonic velocity

exhibited a stronger interaction with microstructural and sub-structural features as

Density

compared to that of longitudinal waves. In particular, this phenomena was observed for

Inverse correlation

the highest α-Al2O3 content composite. Interestingly, an excellent correlation between ultrasonic velocity measurements and ceramic composite hardness was observed. © 2009 Elsevier Inc. All rights reserved.

1.

Introduction

The study of ultrasonic wave propagation in metals and composites gives information on the microstructure, mechanical and physical properties of the material. The quantitative assessment of microstructural changes and properties can be carried out through the measurement of ultrasonic parameters such as attenuation and velocity [1]. The ultrasonic wave velocity depends on the elastic constants and density of the body while ultrasonic wave attenuation depends on microstructure and crystalline defects [2]. The wave propagation in a composite is affected by different material parameters such as density, stiffness, chemical composition and microstructural features [2,3]. Hing et al. determined the elastic properties of ZrO2–Al2O3 ceramic composite system from ultrasonic velocity measurements. They reported that the wave velocity increases up to a maximum at 3 wt.% of

unstabilized ZrO2 dispersed in Al2O3 matrix and decreases gradually thereafter [4]. The increase in elastic modulus, shown by an increased in the ultrasonic velocity, is attributed to phase transformation of the unstabilized ZrO2 from tetragonal to a monoclinic phase, which leads to a toughening and strengthening effect [4]. Ceramic composites are particularly suited for structural applications because they combine the individual physical properties of two or more materials. In alumina–zirconia ceramic composite system, a small addition of ZrO2 into ceramic increases significantly material strength and toughness. Several theories explaining the increase in elastic modulus due to the amount of alumina dispersed within the zirconia matrix have been reported [4–6]. Among them, the crack-tip transformation toughening and stress induced transformation toughening are the most widely investigated mechanisms. The ZrO2–Al2O3 ceramic composites display different properties depending on the raw materials, chemi-

⁎ Corresponding author. Tel.: +52 443 117 2658; fax: +52 443 316 7414. E-mail address: [email protected] (H. Carreon). 1044-5803/$ – see front matter © 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.matchar.2009.02.008

876

MA TE RI A L S CH A R A CT ER IZ A TI O N 60 ( 20 0 9 ) 8 7 5– 8 8 1

cal composition and preparation style. The development of diverse methods for fabricating transformation-toughness ceramics such as ZrO2–Al2O3, mullite– ZrO2, Si3N4– ZrO2 and others, have received considerable interest recently. The ZrO2–Al2O3 ceramic composites have been studied most widely among them. The microstructure of the matrix material and the zirconia particles dispersed in the alumina matrix are thus important in order to produce optimally tough transformation-toughened composite materials that increase the mechanical and thermal properties of the composite [4–6]. In this work, we investigate the influence of α-Al2O3 seeding in the ZrO2–Al2O3 and monitor the process by precise ultrasonic velocity measurements. The ceramic composite is characterized by X-ray diffraction and scanning electron microscopy in order to find out the factors affecting the variation of the longitudinal and shear wave velocity.

2. 2.1.

Materials and methods

Homogeneous sols of pseudoboehmite and ZrO2(Y2O3) were prepared by a mechanochemical treatment employing HNO3 as a peptisizing agent. Pseudoboehmite with average formula Al4O3(OH)6 and a specific surface area of 227 m2/g was obtained from a basic aluminum sulfate derived from alunite mineral (U.G. Process) [7]. The basic salts of such process were completely hydrolyzed in an aqueous ammonia medium at ~70–80 °C and mantaining the pH at ~9. Suspensions were prepared with pseudoboehmite sols seeded with 2.5 wt.% of α-Al2O3 particles of 0.20 µm (Taimicron, TM10). Tetragonal zirconia powders (TOSOH, TZ-3YS) with average particle size of 0.26 µm were added in adequate proportions for each composition. The suspensions were ultrasonically stirred and thereafter spray dried using a YAMATO Mini-spray dryer ADL31. In this way, mixtures of TZ-3YS and α-Al2O3 seeded pseudoboehmite with compositions of 100, 90, 70, 50, 30, 15 and 0 mass % of ZrO2 were prepared. The mixed powder was uniaxially compacted into half cylindrical shaped of 40 × 20× 10 mm samples by isostatic compression at 200 MPa. The samples were presintered at 1250 °C and finally sintered at 1550 °C for 1 h [8]. The samples were identified with the letters SDI followed by a number that indicates the Al2O3 content except the sample that contains 100% wt of ZrO2 as shown in Table 1. Table 1 – Sample identification of the alumina–zirconia ceramic system.

ZTY SDI10 SDI30 SDI50 SDI70 SDI85 SDI95 SDI100

Assessment methods

2.2.1.

Density

The bulk density of the sintered samples was measured by the Archimedes immersion method using a pyknometer [9]. The density of the specimens was determined with the following relation: q=

Wa q Wa − W b b

Composition wt.% ZrO2–Y2O3/Al2O3 100/0 90/10 70/30 50/50 30/70 15/85 5/95 0/100

ð1Þ

where Wa and Wb are the weights of the specimen in the air and water, respectively. ρb is the density of water. All the weight measurements were made using a digital balance (AND, Model-HR 200). The error in the measurement of weight is ± 0.1 mg. The experiment was repeated for four times for each of the specimen to get the reliable density value. The percentage error in the density measurement is ± 0.05%.

2.2.2.

Sample preparation

Sample ID

2.2.

Hardness

Hardness was measured on the ceramic composite samples using a Vickers indentation hardness tester. A four-side diamond indenter was used with a 1 kg load. A small indentation on the polished surface of the sample was made by the system using the standard procedure. For each specimen, at least six indentations were made with similar conditions and the average value of the diagonal length of the indentations was used for the hardness measurements [10,11]. Hardness measurements were made in “line” direction (along sample axis) on all the ceramic composite material samples. The following equation was used for estimation of the hardness: Hv =

1:854 F d2

ð2Þ

where d is the diagonal length of the indentation (mm)and F the applied load (kg). The accuracy in the measurement of d is ±2 µm.

2.2.3.

XRD and SEM

The microstructure and phases identification of the sintered samples were determinated by scanning electron microscopy (SEM) and X-ray diffraction (XRD) respectively. XRD was carried out on a Philips X5000 X-ray diffractometer with CuKα radiation (λ = 1.5418 Å) and nickel filter. The parameters for the XRD analysis were 45 keV and 40 mA, 0.02°/step and 5 s/ step with a diffraction angle (2θ) from 20 to 80°. SEM was carried out on a JEOL JSM-6400 SEM operated at 10 kV. The parameters for the SEM analysis were magnification ≈6000×, <0.1 nA, data acquisition 500 s, working distance (WD) 7 mm. The sintered samples were prepared for SEM by polishing and applying a gold-coating.

2.3.

Ultrasound experiments

Ultrasonic p-wave and s-wave velocities were determined by measuring the time taken for the ultrasonic waves to travel through the material. The ultrasonic velocity measurements were performed using a commercial focused broadband

877

M A TE RI A L S CH A RACT ER IZ A TI O N 60 ( 20 0 9 ) 8 7 5 –8 8 1

Fig. 1 – Experimental set up for the ultrasonic measurements.

immersion ultrasonic transducer (Panametrics V309) with a nominal frequency of 5 MHz, 10 mm diameter piezoelectric disc and bandwidth of 100% (for a 6 dB drop). Half cylinder shaped samples allow us to generate the two bulk modes (longitudinal and transversal) by insonifying a longitudinal wave with different angles of incidence [12]. The time-of-flight (TOF) of the p-waves was determined at normal beam incidence. And the time-of-flight (TOF) of the s-waves was determined at oblique beam incidence by using the Snell´s refraction law. The experimental arrangement for the ultrasonic measurements in the ceramic composite samples with different Al2O3 content is sketched in Fig. 1. The ultrasonic transmitter-receiver has adjustable gain on both emitting and receiving signal paths [13]. Axial movement of the waterimmersed ultrasound probe is computer controlled using step motors. Composite ceramic samples were mounted on a goniometer stage for better position and movement control. The ultrasound signal was recorded with a computer controlled oscilloscope with adjustable time window, enabling digital post-processing.

3.

Experimental results

The measured bulk density, ultrasonic longitudinal and shear velocities (p-wave and s-wave) and hardness of the seeded

sol–gel-derived Al2O3–ZrO2 composites are shown in Table 2. Fig. 2 shows the measured Vickers hardness VH of the ceramic composite as a function of Al2O3 wt.%. Vickerss hardness VH of the alumina–zirconia composite increases in parallel with increasing alumina content in wt.%, from ~ 1285 VH in the ZrO2 phase to ~1792 VH in the Al2O3 phase. Published values for both substances are 1300 VH and 1750 VH, respectively [14]. Fig. 3 shows the X-ray diffraction (XRD) data of the two-phases Al2O3–ZrO2 composite. The main intensity peaks of tetragonal (t) and alfa (α) phases of zirconia (30-1468 ICCD) and alumina

Table 2 – Ultrasonic longitudinal (p-wave) and shear velocities (s-wave), density and hardness (Hv) in the alumina–zirconia ceramic samples. Sample ID ZTY SDI10 SDI30 SDI50 SDI70 SDI85 SDI95 SDI100

p-wave [mm μs− 1]

s-wave [mm μs− 1]

ρ (g cm− 3)

Hv

7.222 7.671 8.184 8.976 9.544 10.125 10.490 10.833

3.748 3.934 4.263 4.599 4.932 5.366 5.871 6.681

6.040 5.660 5.051 4.754 4.352 4.142 4.015 3.901

1286 1380 1475 1560 1619 1730 1770 1790

878

MA TE RI A L S CH A R A CT ER IZ A TI O N 60 ( 20 0 9 ) 8 7 5– 8 8 1

Fig. 2 – Vickers hardness for ZrO2–Al2O3 ceramic composite system.

Fig. 3 – XRD patterns of ZrO2–Al2O3 ceramic composite system with different Al2O3 content: (a) SDI100, (b) SDI95, (c) SDI85, (d) SDI70, (e) SDI50, (f) SDI30, (g) SDI10 and (h) ZTY.

(10-0173 ICCD) are depicted respectively. It is observed that as Al2O3 content increases, the intensity peaks of zirconia tetragonal (t) phase decreases, while the intensity peaks of α-Al2O3 phase increases. The t-ZrO2 peaks gradually increased due to a volume expansion of the ZrO2 particles that promotes a tetragonal phase stabilization during the sintering process as ZrO2(Y2O3) wt.% increases. These phenomena can also be observed in Fig. 4, in which the variation of the material sintered density (ρ) of the composite as a function of the Al2O3 wt.% is presented. The pure ZrO2(Y2O3) attained a density of 6.0400 g/ cm3 after sintering. Addition of alumina into the Al2O3 – ZrO2 composite promoted a decrease of the sintered density until it reaches a value of 3.9010 g/cm3 for pure Al2O3. Fig. 5a–h show scanning electron micrographs of polished surfaces of the composites containing 0, 10, 30, 50, 70, 85, 95 and 100 Al 2O3 wt.%. Samples containing only zirconia present equiaxial grains with grain size ranging from submicron up to several microns as shown in Fig. 5a. In addition, ZTY sample was found to have sintered significantly with a density of about 85% theoretical. Fig. 5b–c shows an homogeneous microstructure distribution of the two com-

Fig. 4 – Variation of the reciprocal sintered density of the samples as a function of Al2O3 content.

M A TE RI A L S CH A RACT ER IZ A TI O N 60 ( 20 0 9 ) 8 7 5 –8 8 1

879

Fig. 5 – SEM micrographs of the sintered samples for the different Al2O3 content: a) ZTY, b) SDI10, c) SDI30, d) SDI50, e) SDI70, f) SDI85, g) SDI95 and h) SDI100.

ponent phases: the Al2O3 particles (gray grains) around the ZrO2 matrix (clear grains) and some residual porosity (black regions). At around 50 wt.% Al2O3 is difficult to distinguish which of the two-phases is the matrix due to the same alumina and zirconia content wt.% (Fig. 5d). The SDI50 sample presents fine grains in an equiangular fabric, ranging in size from less than one micron to several microns across, but the alumina grain sizes are bigger than those of the zirconia. Fig. 5e–g presents an homogeneous phase distribu-

tion of zirconia particles in the alumina matrix with some porosity. The well-dispersed zirconia particles hinder the extensive coarsening of the alumina grains that was observed in pure alumina (Fig. 5h). Fine-grained microstructures with small zirconia particles uniformly distributed at the alumina grain junctions are evident through Fig. 5e–g. These results indicate that ZrO2 particles inhibited alumina grain growth [15]. Finally, Fig. 5h shows an increase of grain size of the alumina phase up to 2–3 µm. In general, SEM

880

MA TE RI A L S CH A R A CT ER IZ A TI O N 60 ( 20 0 9 ) 8 7 5– 8 8 1

Fig. 6 – Longitudinal wave velocity (p-wave) and shear wave velocity (s-wave) measurements along the sample axis as a function of Al2O3 content.

micrographs allow direct observation of some micro-porosity present at the different Al2O3–ZrO2 samples. Depending on grain size, this could bear an effect on Vickers hardness as well as on ultrasound velocity measurements. On the other hand, the measured longitudinal (p-wave) and shear (s-wave) velocities plotted against Al2O3 wt.% are shown in Fig. 6. The longitudinal wave velocity (p-wave) measurements varied between 7.22 and 10.83 mm/ms while the shear wave velocity (s-wave) measurements varied between 3.74 and 6.68 mm/ms as a function of Al2O3 wt.%. In all cases the ratio of s-wave by p-wave velocities was ~ 0.5. It can also be observed that the longitudinal and shear velocities increase with ZrO2 with a maximum at about 5, 15 and 30 wt.%.

4.

Discussion

From Fig. 2, it was observed that Vickers hardness VH of the alumina–zirconia composite increases in parallel with increasing alumina content in wt.%. This can be explained by the fact that small additions of ZrO2 to alumina enhance the mechanical properties of alumina–zirconia composites [4,8]. The improvement in properties has been linked to the polymorphic transformation of ZrO2 from the tetragonal to the monoclinic phase which is accompanied by a volume expansion of about 3% to 5%. Two of the mechanisms by which toughness and hardness are achieved in Al2O3–ZrO2 composite, are attributed to stress induced transformation toughening [16], and resistance to crack propagation by the ZrO2 particles [17,18]. From SEM micrographs, fine-grained microstructures with small zirconia particles uniformly distributed at the alumina grain junctions are evident particularly in alumina–zirconia samples containing small additions of ZrO2. Fig. 6 shows a variation in the p-wave velocity and s-wave velocities along the sample axis as a function of Al2O3 content wt.%. This data scattering can be explained by diffraction due to the probe and the plane parallelism of the faces of samples which must be improved. This fact is more evident in the s-wave velocity measurements (R2 = 0.9631)

compared to p-wave velocity measurements (R2 = 0.9894) as shown in Fig. 7. Because s-wave velocity measurements required a higher precision, in particular in the positioning of the sample relatively to its rotation axis. By analyzing Fig. 7, the p-wave and s-wave velocities as a function of Al2O3 wt.%, it is clearly observed a linear trend through the ultrasonic velocity measurements. The p-wave and s-wave velocity data successfully fit to a linear equation until the zirconia phase is reached, exactly as hardness data (Fig. 4). Hence, the ultrasonic velocity is a good parameter to obtain the material qualitative hardness of an alumina–zirconia composite. The increase in hardness, as shown by an increase in ultrasonic velocity, has been attributed to phase transformation of the ZrO2 from tetragonal to a monoclinic phase, which presumably leads to a toughening and strengthing effect and also due to the action of ZrO2 in stopping grain growth of Al2O3 after sintering.

Fig. 7 – Longitudinal wave velocity (p-wave) and shear wave velocity (s-wave) as a function of Al2O3.

M A TE RI A L S CH A RACT ER IZ A TI O N 60 ( 20 0 9 ) 8 7 5 –8 8 1

5.

Conclusions

This experimental work shows the possibility to estimate the material qualitative hardness of ZrO2–Al2O3 composite from ultrasonic velocity measurements. The results were found to be dependent upon the content of Al2O3 and ZrO2 phases (changes in material chemical composition and density). In both cases, the p-wave and s-wave velocity data successfully fit to a linear equation as a function of Al2O3 content wt.%. The increase in hardness, as shown by an increase in ultrasonic velocity, has been attributed to phase transformation of the ZrO2 which leads to a toughening and strengthing effect and also prevents large grain growth of Al2O3 after densification. The longitudinal and shear velocities increase in ZrO2 samples with a maximum of about 5, 15 and 30 wt.% in which SEM micrographs show fine-grained microstructures with small zirconia particles uniformly distributed at the alumina grain junctions. The effect of micro-porosity on wave velocities in Al2O3–ZrO2 composites needs to be studied experimentally to quantify any corrections. Thus, it is quite possible to determine the concentration of Al2O3–ZrO2 by measuring the ultrasonic velocities.

Acknowledgment This work was supported by CONACYT-COECYT (MEXICO). The authors would like to acknowledge Nagy PB for his valuable contribution of the experimental set up for the ultrasonic measurements.

REFERENCES [1] Vary A. Material property characterization. In: McIntire P, editor. Non-destructive testing handbook, ultrasonic testing. Columbus, OH: ASNT; 1991. [2] Rokhlin SI, Matikas TE. Ultrasonic characterization of surfaces and interphases. MRS Bull 1996;2:22–9.

881

[3] Nayfeh AH, Carreon H, Nagy PB. Role of anisotropy in noncontacting thermoelectric materials characterization. J Appl Phys 2002;91:225–31. [4] Tan KS, Hing P. Ultrasonic through-transmission method of evaluating the modulus of elasticity of Al2O3–ZrO2 composite. J Mat Sci 1997;32:6633–8. [5] Li C, Chen YW, Yen TM. The effects of preparation method on the characteristics of alumina–zirconia powders. J Sol–Gel Sci Technol 1995;4:205–10. [6] Basu D, Sarkar BK. Effect of zirconia addition on the fatigue behaviour of fine grained alumina. Bull Mater Sci Indian Acad Sci 2001;24:2101–6. [7] Wistrom KW, Clark DE. Processing of sol–gel derived doped powders and substrates in science of ceramic chemical processing. New York: John Wiley & Sons; 1986. p. 259–61. [8] Srdic V, Radonjic L. Transformation toughening in sol–gel derived alumina–zirconia composites. J Am Ceram Soc 1997;80:2056–60. [9] Kinger WD, Bowen HK, Uhlmann DR. Introduction to ceramics. New York: John Wiley & Sons; 1976. p. 530–2. [10] Nishara A, Rajendran V, Jayakumar T, Palanichamy P, Priyadharsini N, Aravindan S, et al. On-line ultrasonic velocity measurements for characterisation of microstructural evaluation during thermal aging of β-quenched zircaloy-2. Mater Charact 2007;58(6):563–70. [11] Mak VK, Steinfl RB. Ultrasonic velocity measurements of a Au–W composite. Nondestr Test Eval 1989;5:39–48. [12] Bouda BA, Benchaala A, Alem K. Ultrasonic characterization of material hardness. Ultrasonics 2000;38:224–7. [13] Badidi Bouda A, Lebaili S, Benchaala A. Grain size influence on ultrasonic velocities and attenuation. NDT&E Int 2003;36:1–5. [14] Gong J. On the energy balance model for conventional Vickers microhardness testing of brittle ceramics. J Mater Sci Lett 2000;19:515–7. [15] Zhang W, Glasser FP. The structure and decomposition of Al2O3–ZrO2 gels. J Europ Ceram Soc 1993;11:149–54. [16] Ruhle M, Evans AG. High toughness ceramics and ceramic composites. Prog Mater Sci 1994;38:215–32. [17] Ruhle M, Claussen N, Heuer AH. Transformation and microcrack toughening as complementary processes in ZrO2 —toughened Al2O3. J Am Ceram Soc 1986;69:195–7. [18] Green DJ. Critical microstructures for microcracking in Al2O3–ZrO2 composites. J Am Ceram Soc 1982;65:610–4.