Effect of surface grinding on the strength of NiAl and Al2O3/NiAl composites

(1)

Effect of surface grinding on the strength of

NiAl and Al

2

O

3

/NiAl composites

S.T. Chang, W.H. Tuan

*

_{, H.C. You, I.C. Lin}

Institute of Materials Science and Engineering, National Taiwan University, Taipei, Taiwan, 10764 R.O.C.

Received 24 July 1998; received in revised form 11 January 1999; accepted 23 February 1999

Abstract

For structural applications, dimensional tolerance has to be controlled tightly. Surface grinding is thus frequently applied to match the requirement. For brittle material, improper surface grinding can degrade their strength considerably. In the present study, an alumina wheel and a diamond wheel were used to grind NiAl and Al2O3/NiAl composites. The surface quality, strength and toughness after grinding are

investigated. Cracks are formed on the surface of NiAl ground with the alumina wheel. However, there is no crack found on the surface of NiAl ground with the diamond wheel. The strength of the NiAl machined with diamond wheel is three times that of the NiAl machined with alumina wheel. As Al2O3particles are added into NiAl, the presence of weak Al2O3/NiAl interfaces limits the formation of large ¯aws. The

strength of the Al2O3/NiAl composites is thus less sensitive to the grinding conditions. The present study demonstrates that the Al2O3/NiAl

Keywords: Grinding; NiAl; Composite

1. Introduction

The b-NiAl is a potential material for high-temperature applications [1,2]. Previous studies suggested that the tough-ness and strength of NiAl at room-temperature can be enhanced by adding ceramics inclusions [3,4]. For the Al2O3±NiAl system [3], the presence of Al2O3inclusions

limits the grain growth of NiAl. The strength of NiAl is thus increased. The Al2O3/NiAl interface in the composites is

weak. The weak interfaces deviate the propagation of cracks. The toughness of NiAl is enhanced [3]. Therefore, the Al2O3/NiAl composites are potential candidates for

struc-tural applications.

During the manufacturing of structural components, sur-face grinding is frequently applied to match dimensional tolerance. However, improper grinding can result is surface cracks and sub-surface ¯aws [5,6]. For brittle materials, the presence of sub-surface ¯aws degrades their strength sig-ni®cantly. In the present study, Al2O3particles are added

into NiAl. The NiAl and Al2O3/NiAl composites are

pre-pared by hot-pressing. The cost of alumina wheel is rela-tively low. Alumina wheel is thus frequently used to machine ductile metals. However, when the material to be

machined is hard, such as ceramics, a diamond wheel is used. The intermetallic, such as NiAl, is not as hard as Al2O3; however, NiAl is brittle [1,2]. The machining

beha-vior or NiAl has not been reported in the literature. There-fore, alumina and diamond wheels are both used in the present study. The effect of the grinding on the mechanical properties of NiAl and Al2O3/NiAl composites is

investi-gated.

2. Experimental procedures

Nickel aluminide (b-Nial, Xform, New York) and 0±40 vol% alumina (a-Al2O3, TM-DR, Taimei Chem., Tokyo,

Japan) were attrition milled together in ethyl alcohol. The milling time was 1 h. The grinding media used was zirconia balls. The slurry of powder mixtures was dried with a rotary evaporator. The dried lumps were crushed and passed through a plastic sieve with an aperture size of 74 mm. The sintering was performed by hot-pressing at 14508C with a graphite die for 1 h. The pressure applied was 24.5 MPa. The vapor pressure during hot-pressing was kept below 5 10ÿ3_{torr. The dimensions of the hot-pressed}

specimen were 50 mm in diameter and roughly 4.5 mm in thickness.

*Corresponding author.

(2)

The hot-pressed specimens were cut into rectangular bars with a diamond saw. The surfaces of the rectangular bars were then ground longitudinally with a 120-mesh alumina wheel or a 325-mesh diamond wheel. The cutting depths were 5 mm per pass. The ®nal dimensions of the specimens were 4 3 34 mm3_{. The strength of the specimens was}

determined by the four-point bending technique at ambient conditions. The inner and outer spans were 10 mm and 30 mm, respectively. The loading rate was 0.5 mm/min. The fracture toughness was determined by the single-edge-notched-beam (SENB) technique. The notch was generated by cutting with a diamond saw. The size of the notch was roughly 1 mm in depth, which corresponds to one-third-to-one half of the thickness. Three to four specimens were used to determine the strength and toughness for each composi-tion. The density was determined by the water displacement method. The polished surface was prepared by grinding with diamond slurry to 6 mm and polishing with silica suspension to 0.05 mm. The phase identi®cation was performed by X-ray powder diffractometry (XRD). The microstructure was observed with optical microscopy (OM) and scanning elec-tron microscopy (SEM). The size of NiAl grains was determined with the lineal intercept technique.

3. Results and discussion

The X-ray diffraction patterns indicate that no phases other than a-Al2O3 and û-NiAl are produced after

hot-pressing. The relative density and the grain size of NiAl of the specimens are shown in Table 1. The relative density of the composites is >98%. The microstructures of the NiAl and Al2O3/NiAl composites are shown in Fig. 1. Alumina

particles are attached to the surface of NiAl particles after attrition milling [7]. The alumina grains are thus located at the grain boundaries of NiAl after hot-pressing [3]. The Al2O3particles exert pinning force on the grain boundaries

of NiAl. The grain growth of NiAl is thus prohibited. The electrical resistance of the Al2O3/NiAl composites has been

measured by Tuan et al [3]. They suggested that NiAl grains are still interconnected, despite NiAl grains being sur-rounded by Al2O3particles.

The ground surface of NiAl and Al2O3/NiAl composites

are shown in Fig. 2. The specimens in Fig. 2 are ground with a diamond wheel. For the NiAl specimens, scratches can be observed on the surface. It indicates that plastic

deformation and subsequent tearing are the dominant mate-rial-removing mechanisms. As Al2O3 is added into NiAl,

scratches are decreased. The surfaces of NiAl and Al2O3/

NiAl composites, after grinding with alumina wheel, are shown in Fig. 3. Beside scratches, large cracks can be observed on the surface of the NiAl specimen. As 20 vol% Al2O3is added int NiAl, the width of the large cracks

is decreased. For the 40% Al2O3/NiAl composite, no large

crack is observed.

The arithmetical average surface roughness height, Ra,

and the maximum height of the surface pro®le, Rmax, are

shown as a function of Al2O3content in Fig. 4. Although the

grit sizes in the alumina and diamond wheels are different, the resulting average roughness is very close to one another. The maximum roughness of the 40% Al2O3composite is

close to the grain size of NiAl, Table 1. It indicates that the NiAl grains in the composite are pulled out from the surface

Table 1

The relative density and grain size of NiAl of the hot-pressed specimens Composition Relative density/% Grain size of NiAl/ mm

NiAl 99.2 11.6

5% Al2O3/NiAl 98.9 9.5

10% Al2O3/NiAl 99.3 8.8

20% Al2O3 99.0 8.5

40% Al2O3/NiAl 98.9 8.2

Fig. 1. Microstructures of the Al2O3/NiAl composites containing (a) 0, (b)

20 and (c) 40 vol% Al2O3.

(3)

during grinding. Many pull-outs are observed on the surface of the composites, Fig. 3. Expect the 40% Al2O3composite,

the maximum roughness is smaller than the size of NiAl grains. From Figs. 2 and 3, it is seen that scratches are limited to the NiAl grains. Since Al2O3is brittle, the Al2O3

grains are mainly removed by brittle fracture during grind-ing. It suggests that the material removing mechanism for composites is the combination of tearing of NiAl and brittle fracture of Al2O3.

The strength of the Al2O3/NiAl composites is shown as a

function of Al2O3content in Fig. 5. Fig. 5 suggests that the

strength of NiAl depends strongly on the grinding condi-tions employed. For example, the strength of the NiAl machined with a diamond wheel is threefold that of the NiAl machined with an alumina wheel. As for the compo-sites ground with alumina wheel, the strength is increased with the increase of Al2O3content. As Al2O3is added into

NiAl, the strength difference between the composites

machined with diamond and alumina wheels is reduced. For the 40 vol% Al2O3composite, the grinding condition

has little effect on the resulting strength. The toughness of the Al2O3/NiAl composites is shown as a function of Al2O3

content in Fig. 6. The toughness of the specimens is less sensitive to the grinding condition.

For brittle materials, the failure is originated from their critical ¯aw. The size of critical ¯aw, C, can be estimated by using the Grif®th law as [8]

C p

K_YIC (1)

where is the strength, KICthe toughness and Y a

dimen-sionless constant. By knowing the values of strength and toughness, the size of critical ¯aw can be calculated from Eq. (1). The critical ¯aw sizes of NiAl and Al2O3/NiAl

composites are shown as functions of Al2O3 content in

Fig. 7. The size of critical ¯aw is much bigger than the surface roughness. The strength is thus determined by the sub-surface ¯aws. The ¯aws within the specimens are either

Fig. 2. The ground surface of the composites after grinding with 325-mesh diamond wheel. There are (a) 0, (b) 20 and (c) 40 vol% Al2O3in the

composites.

Fig. 3. The ground surface of the composites after grinding with 120-mesh alumina wheel. There are (a) 0, (b) 20 and (c) 40 vol% Al2O3in the

(4)

originated from the processing stage or from the grinding stages. In the present study, the specimens are prepared by hot-pressing. As the Al2O3content is the same, the size of

the ¯aws formed during the processing stage should be very similar. For the 40% Al2O3/NiAl composite, the size

of critical ¯aw is independent of grinding conditions. This suggests that the critical ¯aw in the 40% Al2O3

composites is formed during the processing stage. The critical ¯aw size of the NiAl ground with an alumina wheel is much larger than that of the NiAl ground with a diamond wheel. It indicates that the critical ¯aw in the NiAl ground with alumina wheel is formed during the grinding stage. The size of the critical ¯aw in the composites ground with alumina wheel decreases with the increase of Al2O3content. The strength of the

compo-sites is thus increased with the increase of Al2O3content.

For specimens ground with a diamond wheel, the ¯aw size is independent of the Al2O3 content. It indicates that the

critical ¯aw in the specimens ground with diamond wheel is formed during the specimen preparation stage. No large ¯aw is generated during the grinding with diamond wheel.

The hardness of Al2O3 is higher than that of NiAl [3].

Therefore, the alumina wheel should be able to grind the surface of NiAl. However, the size of alumina grit is larger than that of diamond grits. Furthermore, the hard-ness of alumina is lower than that of diamond. The grinding force during machining while using the alumina wheel is thus larger than that while using the diamond wheel. Since NiAl is a brittle material, a larger grinding force can induce a larger sub-surface ¯aw [9], as shown in Fig. 8(a). The presence of ¯aws reduces the strength of NiAl signi®cantly. As Al2O3 is added into NiAl, weak

Al2O3/NiAl interfaces are presented in the composites

[3]. During machining, the sub-surface ¯aw is de¯ected along the weak interfaces. The propagation of crack is thus twisted along the interface, Fig. 8(b). No long, straight crack is formed. The strength is determined by the largest ¯aw. As no large ¯aw is formed during grinding, the strength of Al2O3/NiAl composites is less affected by the

grinding conditions employed.

The notch in the SENB specimens is around 1000 mm. The notch is much larger than the size of critical ¯aw. The toughness should thus be independent of the grinding

Fig. 4. The average surface toughness (Ra) and maximum roughness

(Rmax) of NiAl and Al2O3/NiAl composites as functions of Al2O3content.

Fig. 5. The flexural strength of NiAl and Al2O3/NiAl composites as

functions of Al2O3content.

Fig. 6. The fracture toughnesses of NiAl and Al2O3/NiAl composites as

Fig. 7. The sizes of critical flaws in NiAl and Al2O3/NiAl composites as

(5)

condition. However, the toughness of the specimens ground with the diamond wheel is slightly higher than that of the specimens ground with the diamond wheel. The grinding process can induce residual stresses near the surface region [10,11]. A higher contact force during grinding can induce higher residual stresses. The four surfaces of the testing bars were all ground before the notch is introduced. The presence of residual stresses affects the resulting values of fracture toughness. The toughness of the specimens ground with the alumina wheel is thus lower than that of specimens ground with the diamond wheel.

4. Conclusions

In the present study, the effect of surface grinding on NiAl and Al2O3/NiAl composites is investigated. Since the

grind-ing force applied by the alumina wheel is larger than that applied by the diamond wheel, sub-surface ¯aws are formed in NiAl specimens after machining with the alumina wheel. Nevertheless, no cracks are formed in the NiAl specimens ground with the diamond wheel. The strength of the NiAl specimen ground with the diamond wheel is, therefore, three times that of the NiAl specimens ground with the alumina wheel. As Al2O3is added into NiAl, the formation of a long

crack is limited by the presence of the weak interfaces. The strength of Al2O3/NiAl composites is thus less sensitive to

the machining conditions. Acknowledgements

The present study is supported by the National Science Council, R.O.C., through the contract numbers of NSC83-0405-E002-006 and NSC84-2216-E002-030.

References

[1] R. Darolia, J. Metal 43 (1991) 44.

[2] R.D. Noebe, R.R. Bowaman, M.V. Nathal, Inter. Mater. Rev. 38 (1993) 192.

[3] W.H. Tuan, W.B. Chou, H.C. You, S.T. Chang, Mater. Chem. Phys. 56 (1998) 157 (in press).

[4] S.M. Barinov, V.Yu. Evdokimov, Acta Metall. Mater. 41 (1993) 801. [5] D.B. Marshall, B.R. Lawn, R.F. Cook, J. Am. Ceram. Soc. 70 (1987)

139.

[6] H.H.K. Xu, S. Jahanmir, L.K. Ives, J. Mater. Res. 11 (1996) 1717. [7] W.H.Tuan, J. Mater. Eng. Performance 7 (1998) 613 (in press). [8] A.A. Griffith, Phil. Trans. R. Soc. London. A221 (1920) 163. [9] K.A. Kibble, L.A. Phelps, British Ceram. Trans. 94 (1995) 209. [10] S. Srinivasan, P.J. Blau, J.L. Bjerke, J. Mater. Res. 10 (1995) 95. [11] S. Samuel, S. Chandrasekar, T.N. Farris, R.H. Licht, J. Am. Ceram.

Soc. 72 (1989) 1960. Fig. 8. Schematic diagrams for the formation of flaws during the grinding