848 S. Bueno et al. /Journal of the European Ceramic Society 25 (2005)847-850 response, 2 in agreement with a simplified model6., 3 pro- jar and balls during 4h. These conditions were selected from osed for biphasic materials. Therefore a combination of a previous work layers with different aluminium titanate contents might lead Rheological characterisation was carried out using a ro- to simultaneous high strength and flaw tolerance. once the tational rheom tational rheometer(Haake, Rs50, Germany) with a double residual stresses due to the thermal expansion mismatch of cone/plate sensor system layers with different composition are controlled Solid discs with 20 mm in diameter were slip cast in In order to obtain the desired strength-flaw tolerance plaster of Paris moulds in order to determine the casting behaviour, the properties of the layer materials as well as rate of each suspension by measurement of the dry wal the green processing and sintering conditions of the lami- thickness(Mitutoyo, JDU25, Japan)after different casting nates need to be carefully adjusted. First, the composition times(1-16 min). For mechanical characterisation, plates and microstructure of the different layer materials should with 70 mm x 70 mm x 10 mm dimensions were also ob- be optimised to achieve the suitable mechanical behaviour. tained by slip casting for every composition. The cast bodies Second, compatible processing conditions, in particular, were carefully removed from the moulds and dried in air at sintering schedule, should be established to maintain the room temperature for at least 24h properties of the layers in the layered structure and impede The reaction sintering behaviour of the specimens was the failure of the laminate during fabrication due to incom- studied with a differential dilatometer(Adamel Lhomargy patible shrinkage of the layers. Last, residual stresses in the DI24, France)to 1550C. To obtain the monolithic mate- layers, originated by thermal expansion mismatch, have to rials, the dried blocks were sintered in air in an electrical be controlled to avoid fracture box furnace(Termiber, Spain) at heating and cooling rates In this work, the processing parameters to obtain flaw tol- of 2C min, with 4-h dwell at 1200C and 3-h dwell at erant and high strength laminates in the alumina-aluminium the maximum temperature, 1550C titanate system are investigated. Slip casting of aqueous alt The densities of the green and sintered compacts we mina and titania mixtures with high solids contents allows to determined by the Archimedes method using mercury and obtain composite materials with homogeneous microstruc- water, respectively. The crystalline phases present were de- tures and is a simple way to fabricate laminates constituted termined by X-ray diffraction(Siemens AG, D5000, Ger by relatively thick( 200-1000 um)layers. Accurate con- many)after grinding, and results were processed using the trol of the layer thickness can be reached by the control of ASTM Files for corundum(42-1468), anatase(21-1272), the wall thickness formation rate and the sintering shrinkage rutile(21-1276)and B-aluminium titanate(26-0040) of each slip formulation The sintered blocks were machined into bars of 25 mm x e First, the influence of the volume fraction of aluminium 2 mm x 2.5mm(referred to as small bars) for bend strength tanate on the stress-strain response of the composites was tests(three point bending, 20 mm span, 0.5 mm min-,Mi- studied,and from these results, characteristic layered struc- crotest, Spain)and dynamic Youngs modulus( Grindosonic, tures with external layers of sufficient strength were de- Belgium). Nominal stress-apparent strain curves were cal- signed. Second, the green processing and sintering condi- culated from the load values and the displacement of the different layers were selected on the basis of those for the and apparent Youngs modulus was determined from the lin- monoliths, and recalculated with experimental results of sin- ear part of the curves. Reported bend strength and Youngs tered samples. Last, fracture of the laminates was charac- modulus values are the average of five measurements and terised to check whether the desired mechanical behaviour To determine the thermal expansion curves of the mono- was attained pieces of 10 mm x 5mm x 5mm were tested in a differential dilatometer(402 EP, Netzsch, Germany) using es of 5oC 2. Experimental curves the average thermal expansion coefficients between 25 and 850C were calculated. Reported values are the av- The starting materials were commercial a-AlO3( Con- erage of three measurements and errors are the standard de- dea, HPAO5, USA)and anatase-TiO2(Merck, 808, Ger- rations many)powders. Al2O3/TiO2 mixtures with relative TiO2 Two layered composites of five layers were fabricated contents of 0, 5, 15 and 20 wt were prepared to obtain by alternately casting each suspension. Casting times were VoI Sites with Al2 TiOs concentrations of fixed to reach the desired layer thickness considering the 0.10. 30 and 40 vol. after reaction sintering sting kinetics and sintering shrinkage of each composi- The single oxides and the mixtures were dispersed in tion. One laminate, A10A40, had the central and outer layers deionised water by adding 0.5 wt %(on a dry solids ba- (1200 um) made of AlO(A+T) and the two inner layers sis)of a carbonic acid based polyelectrolyte(Dolapix CE64, (300 um)of A40(A+T). In the other system, AAlO, the Zschimmer-Schwarz, Germany). Suspensions were prepare central and outer layers were made of alumina and the two to a solids loading of 50 vol. and ball milled with Al2O3 inner layers of AlO(A+T). Due to the eometry and848 S. Bueno et al. / Journal of the European Ceramic Society 25 (2005) 847–856 response,12 in agreement with a simplified model6,13 pro￾posed for biphasic materials. Therefore, a combination of layers with different aluminium titanate contents might lead to simultaneous high strength and flaw tolerance, once the residual stresses due to the thermal expansion mismatch of layers with different composition are controlled. In order to obtain the desired strength-flaw tolerance behaviour, the properties of the layer materials as well as the green processing and sintering conditions of the lami￾nates need to be carefully adjusted. First, the composition and microstructure of the different layer materials should be optimised to achieve the suitable mechanical behaviour. Second, compatible processing conditions, in particular, sintering schedule, should be established to maintain the properties of the layers in the layered structure and impede the failure of the laminate during fabrication due to incom￾patible shrinkage of the layers. Last, residual stresses in the layers, originated by thermal expansion mismatch, have to be controlled to avoid fracture. In this work, the processing parameters to obtain flaw tol￾erant and high strength laminates in the alumina–aluminium titanate system are investigated. Slip casting of aqueous alu￾mina and titania mixtures with high solids contents allows to obtain composite materials with homogeneous microstruc￾tures and is a simple way to fabricate laminates constituted by relatively thick (∼=200–1000
m) layers. Accurate con￾trol of the layer thickness can be reached by the control of the wall thickness formation rate and the sintering shrinkage of each slip formulation. First, the influence of the volume fraction of aluminium titanate on the stress–strain response of the composites was studied, and from these results, characteristic layered struc￾tures with external layers of sufficient strength were de￾signed. Second, the green processing and sintering condi￾tions to fabricate laminates with controlled thickness of the different layers were selected on the basis of those for the monoliths, and recalculated with experimental results of sin￾tered samples. Last, fracture of the laminates was charac￾terised to check whether the desired mechanical behaviour was attained. 2. Experimental The starting materials were commercial -Al2O3 (Con￾dea, HPA05, USA) and anatase-TiO2 (Merck, 808, Ger￾many) powders. Al2O3/TiO2 mixtures with relative TiO2 contents of 0, 5, 15 and 20 wt.% were prepared to obtain Al2O3/Al2TiO5 composites with Al2TiO5 concentrations of 0, 10, 30 and 40 vol.% after reaction sintering. The single oxides and the mixtures were dispersed in deionised water by adding 0.5 wt.% (on a dry solids ba￾sis) of a carbonic acid based polyelectrolyte (Dolapix CE64, Zschimmer-Schwarz, Germany). Suspensions were prepared to a solids loading of 50 vol.% and ball milled with Al2O3 jar and balls during 4 h. These conditions were selected from a previous work.11 Rheological characterisation was carried out using a ro￾tational rheometer (Haake, RS50, Germany) with a double cone/plate sensor system. Solid discs with 20 mm in diameter were slip cast in plaster of Paris moulds in order to determine the casting rate of each suspension by measurement of the dry wall thickness (Mitutoyo, JDU25, Japan) after different casting times (1–16 min). For mechanical characterisation, plates with 70 mm × 70 mm × 10 mm dimensions were also ob￾tained by slip casting for every composition. The cast bodies were carefully removed from the moulds and dried in air at room temperature for at least 24 h. The reaction sintering behaviour of the specimens was studied with a differential dilatometer (Adamel Lhomargy, DI24, France) to 1550 ◦C. To obtain the monolithic mate￾rials, the dried blocks were sintered in air in an electrical box furnace (Termiber, Spain) at heating and cooling rates of 2 ◦C min−1, with 4-h dwell at 1200 ◦C and 3-h dwell at the maximum temperature, 1550 ◦C. The densities of the green and sintered compacts were determined by the Archimedes method using mercury and water, respectively. The crystalline phases present were de￾termined by X-ray diffraction (Siemens AG, D5000, Ger￾many) after grinding, and results were processed using the ASTM Files for corundum (42-1468), anatase (21-1272), rutile (21-1276) and -aluminium titanate (26-0040). The sintered blocks were machined into bars of 25 mm × 2 mm × 2.5 mm (referred to as small bars) for bend strength tests (three point bending, 20 mm span, 0.5 mm min−1; Mi￾crotest, Spain) and dynamic Young’s modulus (Grindosonic, Belgium). Nominal stress–apparent strain curves were cal￾culated from the load values and the displacement of the central part of the samples recorded during the bend tests, and apparent Young’s modulus was determined from the lin￾ear part of the curves. Reported bend strength and Young’s modulus values are the average of five measurements and errors are the standard deviations. To determine the thermal expansion curves of the mono￾liths, pieces of 10 mm × 5 mm × 5 mm were tested in a differential dilatometer (402 EP, Netzsch, Germany) using heating and cooling rates of 5 ◦C min−1. From the recorded curves the average thermal expansion coefficients between 25 and 850 ◦C were calculated. Reported values are the av￾erage of three measurements and errors are the standard de￾viations. Two layered composites of five layers were fabricated by alternately casting each suspension. Casting times were fixed to reach the desired layer thickness considering the casting kinetics and sintering shrinkage of each composi￾tion. One laminate, A10A40, had the central and outer layers (∼=1200
m) made of A10(A+T) and the two inner layers (∼=300
m) of A40(A+T). In the other system, AA10, the central and outer layers were made of alumina and the two inner layers of A10(A+T). Due to the geometry and dimen-