E L S E V I E R Wear 213 (tt)t.~7) 175-184

WEAR

The sliding wear resistance behavior of NiAI and SiC particles reinforced aluminum alloy matrix composites

Rang Chen a.., Akira lwabuchi b. Tomoharu Shimizu tL Hyung Seop Shin ~, Hidenobu Mifune " "~ Institute t!/" Compo.sire Materials..~h~tnghai Jhto 7",~Jg IIuiversity. Shallgh,i 200030. PeOl~le's Republh" of China

~' l h T , rtme,t ql Mechanhxd t:ngim'erhlg, bactdty ~[ l'Sngim.eiTng, lwate [ btiw'rsio; Morioka 020. Jap, l! • Dt'partntent ofMechttnical Engil~eertn.g, Atlchmg Natitm, tl Utlivt'r.~ity, At~tltntg, K)'ttngbllk 760-600. Sotllh Kt~rea

Received 27 Augu.st 1996; accepted 27 May 1q97

Abs t r ac t

The sliding wear resistance behavior of NiAI and SiC particles reinforced aluminum alloy matrix coml~sites against $45C steel was studied. Experiments were perfiwmed within a load range of 3.5 N to 82.7 N at a constant sliding velocity of I). 15 m s ~. Th,: sliding distance was i 0iX) m, Two types of composites, NiA lp/AI and S iCp/A! with different v~d ume fractions ( 5 col,% and I0 col.% ), ,,,. ere used. At low loads, where particles acted as load bearing constituents and prevented the aluminum matrix being directly involved in tile wear process, the wear resistance of the SiCp/AI and the NiAIp/AI composites was superior tt) that of unreinforced aluminum alloy. The wear rates of SiCp/AI and 10 pet NiAIp/AI composites at 3.5 N were about one factor of I0 lower than that of aluminum alloy. With increasing applied load. the wear rates of the composites increased to levels comparable to those of unreinfi~rced malrix alloys. At 9.4 N. the wear rates of the composites and aluminum alloy were almost the same. The wear rates of NiAIp/AI and SiCp/AI composites above 13.5 N were much lower than those of aluminum alloy, since the severe wear of aluminum alloy at higher loads was hampered by incorporating the SiC or NiAI particles into the matrix.

The wear rates of the counterface material. $45C steel, worn against aluminum alloy, were lower than those worn against the SiCp/AI composites at the entire applied load range. The wear rates were increased with the volume fraction of SiC particles. The NiAIp/AI composites wore the steel at the maximum wear rate at lower loads near 5 N. The NiA! particle was easily fractured when the applied load increased; as a result, the wear rates of steel against NiAlp/AI became smaller and were ahnost the same as those worn against aluminum. © 1997 Elsevier Science S.A.

K,9','ords: Metal matrix composiles: Particle rcinflm.-cd MMCs: Sliding wear: NiAI panicle: SiC" particle

!. Introduction

Particle reinforced metal matrix composites are recognized to have better wear resistance due to the presence of hard panicles. These materials can be used as a reinlbrced parts in pistons and in several wear resislance applications. The tri- bological behavior of MMCs depends on the type of MMCs, counterface materials and the contact situation.

An increase in the slidin~ wear resistance of particles or whisker reinforced aluminum alloys has been measured by many researchers in recent years 11-gl , The degree of improvement depends on the reinforcement, and the methods of manufacture of the composite materials which decide the microstructure and properties of the interface between pani- cle and matrix. Pramila Bat et al. 161 found the fact that the SiC particle reinforced A356 aluminum composites

* Correspnnding aulhor.

(I,[)43- 164R/~7/$17,00 ~O ! 997 Elsevier Science S.A. A|l rights rcxerved PII S 0 0 4 3- 1 6 4 8 ( 9 7 ) 0 0 1 2 3 - 3

improved the wear resistance was auributed to the presence

of SiC panicles which reduced the propensity for materials flow at the surface, and the formation of iron-rich layers on the surfaces of composite during sliding. However, results of sliding wear conducted by various researchers do not show a

consistent trend. AIpas and Embury [ 71 showed that, under

the conditions where SiC panicles promoted subsurface

cracking and materials removed by delamination, SiC rein- forcemenl did not contribute to the wear resistance of alu- minum alloys.

Recent sludies 17-91 revealed that wear re,;istance was largely affected by the strength of the interlace between the

panicle and the matrix as well as the mechanical properties

of the materials. Consequently, when panicles lose their abil- ity to support the applied load due to particle fracture, inter- face debonding and pull-out, panicle reinforcement may cause no improvemen! or even a deterioration in the wear

176 R. ('hen et al. / Weaw 213 (It~97) 17.~-1~4

properties. Alpas and el)workers 110,1 I I investigated the effect of SiC particles on the sliding wear resistance of alu- minum silicon alloys varied with the applied load, and Ibund that at low loads. SiC reinforced composites exhibited wear rates about an order of magnitude lower than those of the unreinfi~rced alloys, since SiC particles acted as load bearing elements and the iron-rich layers were transferred form the counterface.

The NiAI particle reinforced aluminunl alloy ( AI -Ni -Mg- Mn series alloy) composite was considered to have better interlace bonding between particle and matrix and betterwear resistance compared with unreinforced aluminum alloy I 12 I. Besides. since the hardness of the NiA! particle is lower than thai of the SiC particle, it is predicted that the NiAI particle reinforced aluminum composites give the counterface small wear damage. However. the research on the sliding wear resistance of the NiAIp/AI composites worn against the steel is less reported in the existing literature.

The purpose of this paper is to understand the sliding wear resistance of the SiCp/AI and NiAIp/AI composites with different volume fractions (5 vol.f~ and I(I vol.C,~) worn against $45C steel, and t() study the effects of the applied load and different types of particles (m the sliding wear mechanisms.

2. Experimental methods

"lhe experimental apparatus used was a pin-on-disk type tribometer, which was modilied in the previous works I 13 i.

Table I

A moving lower specimen (disk) and a fixed upper specimen ( pin ) were mounted at the ends of a driven shaft and a fixed shaft respectively. Normal load was applied to the specimens by a dead weight at the upper end of the fixed shaft supported by two sliding ball bearings. The rotating motion of the fixed shaft induced by frictional torque was restricted by a plate spring on which str;tin gauges were attached for measuring the frictional torque. The relative distance between the spec- imens was measured by a dial gauge type displacement trans- ducer placed on a fixed upper specimen, The frictional force and the relative distance were recorded continuously by a recorder. The dry sliding wear tests were carried out at a constant sliding velocity of 0.15 m s - t within the applied normal load range of 3.5-82.7 N, The sliding distance was up to 1000 m.

The disk specimens were made of an aluminum alloy (AI 91.4%, Ni 3,0%, Mg 4.8%, Mn 0.8%) with or without rein- forced particles of NiAI and SiC at volume tractions of 5 vol.% and l0 vol,%, respectively. These alloys and compos- ites were produced by the Ryobe Company in Tokyo, Japan, by means of the liquid aluminum alloy stirring method.

The dimensions of the disk specimen were a diameter of 19.5 mm and a height of 8 mm which was machined form the rods fabricated by directly die-casting. The average par- ticle size, Vickers hardness, density and mechanical proper- ties of the materials in the as-received condition are shown in Table I. The Vickers hardness of the NiAI particle was about 270 Hv. The counterface material used as a fixed pin

The average: particle si/e. Vickcrs hardnes,~, den.~ily and mcch:mical properties of Ih¢ materials

Materials Average particle .~i/e Vicker~ hardne,~.~ Density (p, lll} t l kg / lSs ) (gem ' )

Yield strength (MPa)

Elongalion (W)

AI alloy 94.6 2.73 149 3.5 NiAip/AI 15 vcdJ;~ ) 50 103.7 3.01 177 1.5 NiA|p/AI q IO vol.r;~ ) 50 1()9.7 3.29 SiCp/AI (5 vol.~ I I(] I(H).| 2.7*) 165 (l.3 SiCp/AI 1 IO v~,l.~ ) L0 102.8 2.85

Fig. I. The surface morphology of i0 voL% SiCpIAI (a) and 5 voi.% NiAi~JAi : (h) composites before sliding wear.

it. ('l,(,u el al. i Wear 213 (1997) 175-184 177

Aluminum alloy

N i A l p / A I I0 vol%

S i C p / A I lOvo l%

1 mm Fig. 2. The surface rough~,c.~.~ of aluminum alloy :rod COlnpt~.~itc~ hel't~rc sliding wear measured by a .~urface prolih,ueler

was 0.45%C steel, $45C, with the Vickers hardness of 225 Hv. The diameter t:ad the length were 4 mm and 8 mm respectively. The mean contacting dimneter was 13 mm on the disk specimen.

The wear surfaces of the disk specimen were polished by the SiC abrasive papers up to 1000 grits. The polished sur- faces were then cleaned ultrasonically in acetone for 15 rain. Fig. i shows scanning electron microscopy (SEM) photo- graphs of the prepared surfaces of SiC and NiAI composite disks. The surfaces of the pin were polished at the same consequence by the emery papers mounted on the apparatus in order to get a good contact with the disk. The roughness of the aluminum alloy and composites were measured by surface profilometer. The traces of the polished specimens are shown in Fig. 2. The center line average surface rough- ness, R,, was 0.2 p,m for aluminum alloy, 0.4 ~m for NiAlp/ AI composite. 0.6 I~m for SiCp/AI composite and 0.3 I,t,m for S45C steel respectively.

The weight losses of specimens were measured with an accuracy of 0.1 mg using an electronic analytical balance. The specific wear rates were calculated as the volume loss divided by the sliding distance and the applied load, The results were taken as the average from at least three tests.

Microstructural investigations and semi-quantilative chemical analyses on the worn surfaces were performed by SEM and energy dispersive X-ray analysis ( EDXA ).

3 . Results

3. I. The specilic wear rates of alumimmt alloy aml composites

The specific wear rate of the aluminum alloy and the NiAIp/AI and SiCp/A! composites at applied loads from 3.5

to 82.7 N within the sliding distance of 1000 m are shown in Fig. 3. The specilic wear rote of aluminum alloy is decreased with increasing the applied load between 3.5 and 9.4 N, How- ever, it is increased at the load from 9.4 to 13.5 N. Severe wear appears above 13.5 N. Large aluminum debris particles were formed at the initial s~age of sliding and ,severe surface damage occurred,

At the lowest load of 3.5 N, the wear rate of aluminum alloy is aboat an order of magnitude larger than those of lhe SiC and the l0 pet NiAI particle reinforced composites. It reveals that at small loads, aluminum alloy shows lower wear resistance than particle reinforced composites. The wear rate of the 5 pet N iAI particle reinforced composite is in the moderate region, about five times those of other composites. The specilic wear rate of composites is increased with the load, and at 9.4 N the specific wear rates of composites are almost close to that of unreinforced aluminum alloy, and then, no improvelnent of wear resistance of composites at this load level. However, above 9.4 N the specific wear rate of com- posites decreases with increasing applied load, even when the applied load is increased to 82.7 N. This reveals that particle reinforced composites have the ability to increase wear resistance at higher loads.

The effects of particle and volume fraction on the wear resistance of the composites is also found from Fig. 3, The NiAip/Ai composites seem to have a slightly better wear resistance than the SiCp/AI composites within load range IYom 5.5 to 82.7 N, even under the fact that the hardness of NiAI particle is smaller than that ofthe SiC particle. However, at the low load range below 4.5 N, the wear rate of 5 pet NiAIp/AI composite is higher than other composites. Increasing the volume fraction of particle decrea~s the wear rate of NiAIp/AI composites. On the contrary, the wear rate

][)-~ . . . . . . . ! - 7 - . . . . . . . I

Z

m G

~" I ( |-~ -

~ s 1,1

~ AI "~ ~ NiAIplAI Svel%

- - - * . . . . N i A I p / A I I O v o l % ~ v - - S i C p l A I 5 v o 1 % - - x - - S i C p l A I 1 0 v o l %

10-,~ . . . . . . . . . . . | . . . . . . . . . . I 5 tO 50 lOt)

L o a d (N) Fig. 3. The sl~:cilic wear rale of aluminum ahoy. 5 vol.e,~ and I0 vol.r~ NiAIp/AI and SiCp/AI composi|es ver.,,us the applied load alter a sliding di.~lanc¢ of I(XH) m,

178 R. Chert et al. / Wear 213 (1997) 175-184

0.X ~ " " '--" ' ' I . . . . . ' "1 : AI

g q~ ---am--- NIAIplAI Svol% • 11,7 \ - - 4~ - - N|AIplAI 1Ovol%

"-' \ ~ SICptAI Svol% k - - x ~ SICpfAI 10rot%

\ 0.fi -

~ f l . 5 °~

0 . 4 -

o= 0.3 , , , , ~ . . I . . . . . . . I

5 i t) 5O 100

L o a d ( N )

Fig. 4. The coefficient of friction versus the appl ied hind at the end tff a

sl iding dis tance of 10(X) m.

E

l (1-4

El 5

l 0 -5 4#

5

Q ,

!0-*

"' ' "' '""'i ' • , , , , • , , | , , , , . ,

=

--e--. AI ------- NIAIp/AI $vol%

/ , / ~ A - - * - - N I A I p / A I 1 0 v o l % StCp/At $vol%

- - x ~ S i C p / A I 1 0 v o l %

, . . . . . . . I . . . . . . . . I

5 i() 50 ]oo 5oo L o a d (N)

Fig. 5. The specilic wear rate of $45C steel worn against aluminum alloy and composi tes versus the appl ied load after a s l iding distance of I000 m.

of SiCp/AI composites is increased by increasing the volume fraction of particles over the load applied.

The coefficient of friction of aluminum alloy and compos- ites is shown in Fig, 4, It is obvious that the coefficient of friction between $45C steel and IO pet NiAIp/AI composite is the largest at the lowest load of 3.5 N, above 0.75 and decreasing with increasing load. and then. the coefficient of friction became |he same as those of other materials between 0.4 and 0.5 at loads above 9.4 N. Meanwhile, the 5 pet NiAIp/ AI composite had the largest coefficient of friction at 4.5 N: however, it dropped quickly with increasing load.

3.2. The wear rate of the ,'r,unterface materials, $45C steel

in practical use, it is important to know the total wear of the entire tribological system. The wear rates of the counter- face materials $45C steel worn against aluminum alloy, NiAIp/AI and SiCp/AI composites are shown in Fig. 5. The changes of relative distance between the surfaces of a pin and

' ' ' = " ' ' ' l "' ' ' " ' ' l

: A I ---o--. NiAIplAI Svol~,

,", ~ v ~ - NiAIpIAI 10vol% 11)o ; '* ~ S i C p I A i Svo l .a',,

, . , ] \ ~ . , ~ ~ x - - SiCp/AI 10vol%

5o .=_ ==

/ 0 ~ , -i r ' ' ' I . . . . . . . . i

5 ]. 5. It.) Load (N)

Fig. 6. The dis tance change of contact surfaces o f steel worn against alu- minum alloy and composi tes versus the appl ied load al•ler a s l iding dis tance o f I (XX) m .

l =

'°' t t A B

.i [ , _ , . . . .

0 250 500 750

S l i d i n g d i s t a n c e ( m )

,oo~ ~

l..--

(b)

I 4 6 . m i lille I

i't g

t j

2 0 500 ? ' ;0 1000

S l i d i n g d i s t a n c e I r a )

Fig . 7. T h e ¢oe l ' l i c i cn t o f f r i c t i o n and d is l :mce change b e t w e e n the conlat.-t

sur face versus the s l i d i ng d i s lance a! 3.5 N: ( a ) a l u m i n u m a l l oy , I b ) 5 vo l . rh

S i C p / A I c o m p o s i t e .

a disk after sliding a distance of 1000 m at different loads are shown in Fig. 6, which are related to the total height loss of the specimens.

The wear rate of the steel against aluminum alloy is smaller compared with those against SiCp/AI composites at entire loads and NiAIp/AI composites at load range below 9.4 N.

R. Chert et a L / Wear 213119t~7) 175-1,~4 1 7 9

The SiCp/A! composites wore the steel at higher wear rate. especially when the applied loads were above 9.4 N. The wear rates of the steel again.~t SiCp/AI composites ate about an order of magnitude larger than those against NiAlp/AI composites and aluminum alloys at 13.5 N. The wear rates

of the sleel are increased by the increase of volume fraction of the SiC particle.

The wear rates of steel against NiAIp/AI composites are affected seriously by the change of applied loads as shown in Figs. 5 and 6. Under the applied loads below 13.5 N. the

i 1 i I

I ! ,1 1

f . e l a

94:~, 9 : I.

Ft~.

t t

{

j ' - c t ~

~ s

i

i I

Fig+ 8. E D X A analysis dlala .~howin~ ulolni¢ PerCema£.e pre.nenl on the worn ! b ) area wi l lmu! pa~icle.s.

#.~F'~:: '?'61a8 :~',.'.,!.

I I I I I I l I

i

1

1 I t I I i t ...... i,

1

.~urfitce t~l Ille I0 vol,'/} S i q ~ f A I compo,~ile al 5.5 N: (a) area rich in parlicle,~,

180 R. Chen e! al. / Wear 213 (I 997) 175-184

NiAIp/A! composites wore the steel at a large wear rate. The wear rdte of steel reached the maximum at 4.5 N for the 5 pet composite and at 5.5 N for the l0 pet composite. The maxi- mum wear rate at relatively low loads is about 10-15 times those at high loads. When the load is above 4.5 N for the 5 pet composite and 5.5 N for the l0 pet composite, the wear rates began to drop and finally became the same as those of the aluminum alloy.

alloy directly involved in wear process. As a result, the wear rate of 5 pct NiAip/AI composite at 3.5 N is about five times that of the SiCp/AI composites. However, it is still lower than that of unreinforced aluminum alloy, as shown in Fig. 3. because the hardness of the NiAI panicle is larger than that of aluminum alloy and the particle can keep its structural integrity at low loads.

4.2. Wear resistance at middle and high loads

4. Discussions

4. I. Wear resistance at low loads

At the lowest applied load, the high wear resistance of SiCp/AI materials is attributed to the presence of SiC parti- cles in the matd.,:. The surface of a SiCp/AI composite in Fig. I shows that SiC particles stand proud of the polished surfaces. The particle protrusions are found as high as 6 ixm, as shown in Fig. 2. These protruded panicles act as load supporting elements and are useful to prevent the softer alu- minum matrix becoming directly involved in the wear process at low loads.

The presence of SiC particles is also useful to prevent the matrix aluminum alloy from early fracture at low loads. Fig. 7 shows the coefficient of friction and the relative distance between the surfaces of a pin and a disk for the aluminum alloy (a) and SiCp/AI composite (b) versus the sliding distance at 3.5 N. For the aluminum alloy, the coefficient of friction jumps to about two times the usual value at several sliding distances, meanwhile the relative distance is also changed abruptly as shown by points A and B in Fig. 7(a) , The increase of the coefficient of friction and the abrupt change in distance may be attributed to the formation of wear particles between surfaces due to the aluminum damaged at the contact surface. However, for the SiCp/AI composite at the same load, no such phenomenon could be found as shown

in Fig. 7(b) . The chemical composition of the worn surface of the disk

is also changed by the presence of SiC panicles. Energy dispersive X-ray analyses were performed to determine the relative atomic percentages of metallic elements ( AI, St, and Fe) on the worn surfaces of the SiCp/AI composites at low load. Two typical areas which are full with and without SiC panicles were selected. The results as shown in Fig. 8(a) and 8(b) reveal that the relative content of elemental Fe at the surface with SiC particles is larger than that of the surface without SiC particles at 3.5 N. This implies that the transfer of elemental Fe from steel to the surface of the composites is increased by the abrasive action of the presence of SiC par- ticles in the matrix.

The NiAI particle does not stand proud of the surface as shown in Figs. I and 2 because of its softer hardness (270 He) and the larger diameter ( 50 p,m) compared with that of the SiC particle ( 1000 Hv and 10 p,m). As ,~ result, the NiAI panicle is not able to prevent the sorer matrix aluminum

When the load is increased to reach the fracture strength of the particle, the particles begin to fracture and lose their ability to support the load. For panicle reinforced composites, the particle near the contact surface may induce the nucleation of cracks due to the interface debonding between particle and matrix than monolithic aluminum alloy. In the sliding wear process, these cracks may propagate and connect to form the subsurface cracks, the subsurface damage process is increased by the presence of particles. Fig. 9 shows that there are many cracks occurring in the wear surface of the I0 pet NiAlp/AI composites even at the lowest load of 3.5 N. The

iX:.

Fig. 9. The worn surf~e o f showing the cracks produced,

composite at 3.5 N

Fig. 10, The ¢ross-~ction perpendicular to the worn surface oflhe l0 vol.(; 5iCp/AI composite at 40 N showing ~hc subsurface crack occurs.

R. {'hen et al. / Wear 213 ( IO97.~ 175-184 l g l

Fig. I I. The worn surface of the 10 vol.% $iCp/AI composite (a) at 3.5 N showing the SiC

, ~ . . '

, , .

&,~r

(b) at 9.4 N.

Fig. 12. The worn, surface nf the I0 vol.% SiCp/AI compsite at 64 N (a) showing the triholayer fractured, (b) showing the broken p',miele fragmented and accumulalcd.

subsurface crack is also found for the IO pet SiCp/Ai corn- posite at 40 N as shown in Fig. I O. As a result, when the load increased up to 9.4 N, lhe wear rates of composites increased to levels comparable 1.o that of the unreinforced aluminum alloy. Fig. I I (a) shows the SiC particles in structural integ- rity on the wear surface of the SiCp/AI c~;mposite at 3.5 N, and reveals that the particles do not fracture ~,.t low loads. However, at loads higher than 9.4 N it is very difficult to discover these parlieles in structural inlegrity, as shown in Fig, I I(b).

When the load is increased above 13.5 N, the wear proce,~s of the aluminum alloy may be a deformation and damage accunmlation process. The deptll of the deformed layers of the aluminum alloy below the worn surfaces is more deeper than that of the composites since the yield strength of the composites is larger, as shown in Table 1. Besides, the surface temperature increased with the applied load and sliding dis- tance. Aluminum alloy easily undergoes thenna[ softening and recryslallization at high lemperature compared wilh the composites because the strength of Ihc composites at high temperature is greater. As a result, the wear rate of Ihe alu-

minum alloy is increased drastically at loads higher than 13.5 N. This transition to severe wear cannot be found for the composites. However. the deformation and damage of the worn surface is also very obvious. Some of the deformed layer is broken, and several broken particles become flag. mented and accumulated as shown in Fig. 12.

The NiAI particle is easily damaged when increasing the load. The IYactured NiA! particle near the cross-section of wear surface at 40 N can be found as shown in Fig, 13. The increased wear rate alter 3.5 N is due to the fracture of the parlicle. Fig. 14 shows that the deformed layer is seriously damaged at high load.

4..¢. 77w wear rate o[the counlerface material at different loads

The wear rate of the steel against the SiCp/AI composites is larger Ihan dmt against unreintbrced aluminum alloy over tile h~ad applied. The presence of high hardness or s i c par- titles in the aluminum matrix wears steel at a high wear rate.

182 R. Own et al. / Wear 213 (1997) 175-11¢4

. . . . . . ~ - ~ -

• .7

~, ;~il I ,~ •

. . ~ . ,

i

4d

~ 1 . 0 v~

• ~ 0,5

I

• J J

o 25o ~oo 7~0 looo S l i d i n g d i s t a n c e ( m )

Fig. 15. The coeflicienl of friction versus the :,liding dist:mce at 3.5 N I'~w the I0 vol.c~ NiAIp/A| composile.

Fitz. 13. The cross-~ction perpendicular to the worn surface of the I 0 wfl.CTf NiAIp/AI compasile at 40 N showing the tribolayer fraelured near the NiAI panicle.

/:it:. 14. The worn surfilce of the IO w~l.~ NiAIp/AI comr~site at 64 N sht~v,.ing lhe tribtllayer fractured.

It is interesting that even if the hardness ol'the NiAI particle is smaller than that of the SiC particle, the wear rate of steel against the NiAIp/AI composite is about twice that against the SiCp/AI composite at low loads near 5 N as shown in Figs. 5 and 6. Fig. i 5 shows the coeflicient of friction versus sliding distance of a NiAIp/AI composite at 3.5 N. Within the sliding distance of 250 m, the coefficient of friction is only 0,4, then the coefficient of friction is increased with the sliding distance and reached twice of the initial value. The

wear mechanism was changed alter the sliding distance of 250 m. As shown in Fig. I, the NiAI particle is not standing proud of the contact surface; however, at several meters slid- ing wear, the softer aluminum alloy was fractured and. as a result, the contact surface became the steel and the NiAI particle, the harder NiAI particle could abrade the $45C steel surface, and the NiAI particle wears the steel at very high wear rate. As a result, the coefficient of friclion and the weight loss of steel is increased. In Fig. 16(a) and 16(b), at low load, the higher intensity of Fe means that the transfer of steel to the surface of a NiAIp/AI composite is higher, meanwhile, in Fig. 16(c) and 16(d), the transfer of aluminum to the surface of steel ( pin ) is lower. At the applied load above 9.4 N. the NiAI particles are easy to fracture and become frag- mented, it can be seen that the coefficient of friction between steel and 10 p e t NiAIp/AI composite is decreased with the applied load as shown in Fig. 4. The broken NiAI particles could not abrade steel at high wear rates• The transfer of Fe to the surface of the NiAIp/AI composite is lower. As a result, the wear rate of steel at high loads becomes smaller.

5. Conclusions

I. At low loads, the wear resistance of SiCp/Ai and 10 pet NiAIp/AI composites are about an order of magnitude better than that of unreinforced aluminum alloy, which is attributed by the fact that the particles support the applied load. preventing the softer aluminum matrix to wear directly with steel.

2. The wear rates of the SiCp/AI and NiAIp/AI composites increased to a level comparable to that of unreinforced aluminum alloy at medium loads of 4.5 to 9.4 N because of particle fracture and damage of subsurface crack layers.

3. The transition to severe wear takes place for an unrein- forced aluminum alloy at 13.5 N. However, no transition could be found for the SiCp/AI and NiAIp/AI composites due to their better yield strength at high temperature.

R. Ch,'~t t'l al. / W e a r 2 1 3 f 19t)7~ 175-1, '¢4 1 8 3

"~...t" ,I "~: |

I

. . . . . . . !

I I

I 5 ! i i I

B - - - . - ~ g . g g

!

I

t i I I I

,' ! I ; i l F ~ . ' ,

i ,-,;

i i

!

i ! j

I q

l

[ i . .

L

: I ; I I I

~ " PI'~: I _ _ A _ A [ . . . . . . . . - .a

~ _ ~ 1 0 . - ' : ~ !

I

! : - I

I !

I

: t = =

k _ ~ ! -

t t i i , 4 ,

I I i

I !

"~d~ "~''~" " , "w,e ,~w ..r-[ i,'~ " " " "~ ~ ' , ' .

. ~,~.

' I l

' 1 ! i _ j

L'~ '~4' • F I ' ~ " ~ "1'~,,,

- i i !

I l

, 1

t u I

t I

! I I

L i

:,jr ? - . .

.~.

¢.

. ~ ~ •

. , "~ ' ~ . . . . . . . . . . . . . . - - ~ - " " ' : ~ . b ¢ ~

Fi,~. 16. EDXA analysis data showing alonlic percentage present oil tile wlJrn .,,urfa~:e of Ih¢ I(I vol.'/; NiAIp/AI com[x~site ( disk I al (a) 4.5 N and i h } 13.5 N. and the steel (pin) .'lgainsl I0 vol.gf NiAIp/AI composite ai ( c J 4.5 N a:',l I d I 13.5 N.

4. The S i C p / A I compos i te wore steel at the wear rote o f 5 -

10 times that of unreinforced aluminum alloy over the load applied because o f the presence of hard particles. A

h i s h volume fraction of SiC part icles wore against steel at high wear rales.

5, At low appl ied loads, Lhe N i A I p / A ! compos i t e wore

agains t sleel at a h igher wear rate than the S i C p / A I com-

posite, The wear tales o f steel againsl the N i A I p / A I com-

posi te become s imi lar to that against the a luminum al loy

when the NiAi part icle is fractured at high loads above 13.5 N.

A c k n o w l e d ~ e m e n t , ~

The aulhous would like to thank the Ryobe Connpany in T o k y o for supply o f the materials , and useful d i scuss ion with Professor K. Katagiri at Iwate Uni , ,ers i ty is great ly appreciated.

R e f e r e n c e s

I l l F.M. It~sking, F. Folgar Porlillo, R. Wunderlin, R. Mcltrabian, (',,~1111~o.",.,il,en i l l ' : _ | ] U l l l i t l t l l l l a l l o y . , , : I ' ; I h r i~c~ . l t i on ; . l l t d %~.,u,;.ti* h e h : t v i . r , r , J .

M a t , , , ' r . , q c i . 1 7 ~ 1 9 ~ 2 TI 4 7 7 - 4 0 8 .

184 R. C'hen et aL / W e a r 2 1 3 ~19971 175-184

121 M.K. Surappa, S.V. Prasad, P.K. Roh;itgi. Wear and ablasi~n of cast AI-alumina particle composites. Wear 77 t 1982l 2t)5-302.

[ 31 A.G, Wang, hM, Hatchings, Wear t)f alumina fiber-aluminum metal matrix composites by two body abrasion. Mater. Sci. T~.-chnol. 5 (1989) 71-76.

141 K.C. Ludema. A review of .scuffing and running in ~f luhrieated surfaces wilh asperities and oxides in perspective, Wear 100 119841 315-331,

151 A.T. Alpus, J. Zhang. Effect t~f SiC particulate reinfcwcement t~n the dry sliding wear of aluminum-silic~m all(ws t A356~. Wear 155 (19021 83-104.

161 B.N. Pramila Bai, B.S. Ramasesh, M.K. Surappa. Dry sliding we:lr of A356.--AI-SiCp composites, Wear 157 I 1992 ) 295-3(~.

171 A.T. Alpas, J.D. Embury. Sliding and abrasive wear resistunt:e t~t" nn aluminum ( 2014 D-SiC particle reinforced comfy)site. Scripta Metall. Mater. 24 (1990) 931-935.

181 S,J. Lin, K.S. Liu, Effecl (ff aging on abrasi()n rate ill an Al-Zn-Mg- SiCcomposile, Wear 121 (1988) 1-14.

191 J. Yang, D.D.L Chung, Wear t~l bauxite particle rcinli~rced aluminum alloys, Wear 135 ~ 1989) 53-65.

1101 A.T. Alpzs, J. Zhang. Effect of micnJstrtlclure ~ parliculale size and volume fraction) and counterfuce materials tm the sliding wear resistance of particulate-reinforced aluminunl matrix cLm~p~sites. Metall. Mater. Trans. A 25 ( 1094 ~ 969-983.

I I1 I W. Ames, A.T. Alpas. Wear mechanisms in hybrid compensates of graphite-2(I pet SiC in A35fi aluminum alloy I AI-7 pet Si-4L3 pet Mg). Metall. Mater. Tran-,. A 26 ~ 1995 ) 85-98.

1121 H.S. Shin. K. Katagiri, T. Satin. Y. Shoji. H. Kanemaru. H. ()umra.The mechanical properties of SiC and NiAI particles reinf~rced aluminum alloy come, site manufactured by alia-cast. Trans. of JSME, in pre.~s.

I I31 A. Iwabuchi, K. Hori, H. Kubo.sawa. The effect ~1 oxide particle~ supplied at the interface belbre sliding t~n the severe-mild wear transition. Wear 128 i ItJ881 123-137.

Biographies

Rong Chen, born in 1962, rece ived his Ph,D. in Engineer ing from Shanghai J iao Tong Univers i ty in 1989, Then, he became a lecturer in Shanghai Jiao Tong Universi ty. From 1993 to 1995, he was a Research Associa te at Tohoku Uni- versity and lwate Universi ty, now lie is an Associa te Profes-

sor at Shanghai Jiao Tong Univers i ty , and a special researcher

at lwate Universi ty. His major research areas are the manu-

facture and p e d b r m a n c e evaluat ion o f metal matrix compos- ites, inc luding cont inuous fibers, whiskers and particle re in |o rced a luminum matrix composi tes ,

Akira lwahuchi , professor o f lwate Universi ty, Ph.D. in Engi- neering, born in 1949. Majur research areas are t r ibology,

mechanica l work ing and mechanica l materials.

Tomoharu Shimizu, Associa te professor o f Iwate Universi ty, Ph,D, in Engineer ing , born in 1959. Major research areas are t r ibology and mechanics .

Hyung-seop Shin, horn in 1959, rece ived his Ph.D, in Engi-

neer ing from Tohoku Universi ty, He was once a lecturer at

lwate Universi ty, and now is an Associa te Professor o f A n d o n g National Universi ty. His major research areas are materials mechanics , materials s trength and mechanica l materials.

Hidenobu Mifune, horn in 1960, is a technician of lwate

Universi ty.