zirconia ceramics and mechanical surface interactions · 6. tribology and ceramics 76 76 78 78 78...

Zirconia ceramics and mechanical surface interactions

Citation for published version (APA):van den Berg, P. H. J. (1992). Zirconia ceramics and mechanical surface interactions. Technische UniversiteitEindhoven. https://doi.org/10.6100/IR382859

DOI:10.6100/IR382859

Document status and date:Published: 01/01/1992

Document Version:Publisher’s PDF, also known as Version of Record (includes final page, issue and volume numbers)

Please check the document version of this publication:

• A submitted manuscript is the version of the article upon submission and before peer-review. There can beimportant differences between the submitted version and the official published version of record. Peopleinterested in the research are advised to contact the author for the final version of the publication, or visit theDOI to the publisher's website.• The final author version and the galley proof are versions of the publication after peer review.• The final published version features the final layout of the paper including the volume, issue and pagenumbers.Link to publication

General rightsCopyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright ownersand it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.

• Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal.

If the publication is distributed under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license above, pleasefollow below link for the End User Agreement:www.tue.nl/taverne

Take down policyIf you believe that this document breaches copyright please contact us at:[email protected] details and we will investigate your claim.

Download date: 08. Oct. 2020

https://doi.org/10.6100/IR382859

https://doi.org/10.6100/IR382859

https://research.tue.nl/en/publications/zirconia-ceramics-and-mechanical-surface-interactions(6fd632bc-3e85-456f-944b-203f888add2a).html

Zirconia Ceramics and

Mechanical Surface Interactions

Paul van den Berg

/



Proefschrift

ter verkrijging van de graad van doctor aan de

Technische Universiteit Eindhoven, op gezag van

de Rector Magnificus, prof. dr. J. H. van Lint,

voor een commissie aangewezen door het College

van Dekanen in het openbaar te verdedigen op

dinsdag 6 oktober 1992 om 16.00 uur

door

Paul Hendrik Jacob van den Berg

geboren te Haarlem

druk: wlbro dtsser ta l ledru KK.eriJ, n e 1mona.

Dit proefschrift is goedgekeurd

door de prornotoren

prof. dr. G. de With

en

prof. ir. A. W. J. de Gee

to my parents

Contents

2

3

Introduction

1.1 Materials

1.2 Zirconia ceramics

1.3 Mechanical surface interactions

1.4 Outline of this thesis

References

Materials and experimental methods

2.1 Experimental methods

2.2 Materials characterization

References

The phase transformation in zirconia

3.1 The phase transformation

page

1

2

2

5

6

7

7 10

15

16

18

3.2 Mechanical aspects of the phase transformation 23

3.3 Grinding 26

3.4 Fracture, residual stress and transformation zone depth 26

References 27

4. Residual stress and the stress-strain curve for Mg-PSZ 35

35

37

38

43

45

46

47

4.1 Introduction

4.2 Experimental

4.3 Results

4.4 Discussion

4.5 Considerations

4.6 Summarizing conclusions

References

5. Residual stress and strength of zirconia after grinding 48

5.1 Residual stress and strength of Mg-PSZ after grinding 48

5 .1.1 Introduction 48

5.1.2 Experimental 49

5.1.3

5.1.4

5.1.5

Results

Discussion

Conclusions

52

57

66

5.2 Residual stress and strength of Y-TZP after grinding 66

5.2.1 Introduction 66

5.2.2 Experimental 67

5.2.3 Results 67

5.2.4 Discussion

5.2.5 Summarizing remarks

References

72

74

74

6. Tribology and ceramics 76

76

78

78

78

6.1 Wear models

6.2 Wear of ceramics

6.3 Wear of zirconia

References

7. Tribology and zirconia 83

8.

7.1 Wear and strength of Mg-PSZ sliding against stavax 83

7.1.1 Introduction

7.1.2 Experimental

7.1.3 Results

7.1.4 Discussion

7 .1.5 Summarizing conclusions

7.2 Wear and strength of Mg-PSZ and Y-TZP

7 .2.1 Introduction

7.2.2 Experimental

7.2.3 Results

7 .2.4 Discussion


References

Pin-on-Plate and Pin-on-Disk measurements

8.1 Y-TZP reciprocating against sialons and AlONs

8.1.1 Introduction

8.1.2 Experimental

8.1.3 Results

8.1 .4

8.1.5

Discussion

Summarizing conclusions

83

84

87

97

103

104

104

105

105

109

109

110

112

112

112

113

116

140

144

8.2 Conventional Pin-on-Disk measurements

8.2.1 Introduction

8.2 .2 Experimental

8.2.3 Results and discussion


References

9. Summarizing discussion

Listing of symbols often used

Summary

Samenvatting

Acknowledgement

Curriculum Vitae

145 145 146

147 153

153

155

159

160

162

164

165

1. INTRODUCTION

1.1 MATERIALS

Why is materials research done and what reasons justify the effort in

a specific subject, are major questions to answer. Materials are used

everywhere and everyday. Characteristics like size, shape, composition,

surface finish, mechanical, chemical or electrical properties determine the

possible applications of a material. The choice of a certain material for a

certain function is not always obvious. Often there is a variety of

materials that can be used. Simplified marketing principles can be used to

illustrate the variety of arguments that determine the choice for a

material. These principles tell that the market determines which product is

required. There are many factors that influence the needs and requirements

of a market. Amongst these are economical, environmental, political,

aesthetical, sociological and many more. There is a lot of interaction

between these factors and they change continuously with time. Environmental

arguments, for instance, become increasingly important. These changes are

difficult to predict. To remain competitive and to have the ability to

react to changes in a market, fundamental research on materials is a

necessity. A large part of materials science research is concentrated on

the improvement of existing materials, like various kinds of alloys,

plastics and technical ceramics.

One of the most important aspects of materials research is the field

of application in which the material is foreseen. The investigated

phenomena are usually closely connected to the circumstances of the

possible application.

Classification within a group of materials, like technical ceramics,

can be based on properties. The group of technical ceramics can be divided

in functional and structural ceramics. Functional ceramics is a term used

2 Chapter 1. Introduction

for ceramics that are interesting because of distinctive properties like

electrical or magnetic properties. Structural ceramics are ceramics that

are mainly interesting for their mechanical properties. In this

project zirconia ceramics have been studied.

1.2 ZIRCONIA CERAMICS

Various types of zirconia ceramics exist. The type of materials which

are the subject of this thesis, transformation toughened zirconia ceramics,

are materials which came under attention after about 1975 [1]. They combine

favorable properties like high hardness and oxidation resistance with a

strength and fracture toughness which is high relative to the value for

other ceramics. The potential applications for the materials are

miscellaneous. They are for instance mentioned as biocompatible materials

[2, 3], as materials suitable as heating elements [4], applications in

armors were examined [5] and they were even mentioned as a dental compound

[6]. A major area of application for zirconia ceramics is in tribology

[7 -15] and research is done in this field in a variety of places [8].

1.3 MECHANICAL SURFACE INTERACTIONS

wear.

most

The two surface interactions investigated are grinding and sliding

Grinding is interesting because ceramic materials are machined for

applications and grinding is a reasonably defined and reproducible

interaction. Sliding wear, part of tribology, is interesting

the application of zirconia ceramics is often sought in wear

surface

because

resistance.

The study of friction, wear and lubrication, called tribology is a

relatively new area that combines aspects of physics, chemistry, mechanical

Chapter 1. Introduction 3

engineering and contact mechanics in order to understand the problems

involved. A recent review of tribological education is given in [16]

illustrating some of the difficulties in the education of this

interdisciplinary field. A major problem in tribology is the lack of a

lubricant which can be applied at temperatures above approximately 300 °C.

Research is done, aiming to solve this problem. New lubricants are

developed and materials like wear resistant structural ceramics, are tested

under dry sliding conditions at elevated temperatures.

The study of two materials sliding against each other contains many

aspects. This is illustrated with simplified descriptions of indenting,

single-point scratching and grinding, which are at this point only used to

introduce the problems of tribology.

The indentation is a relatively simple beginning. Indenting a material

means that one, hard, material makes an impression into another, relatively

soft, material. The impression left in the soft material after removal of

the indentor is the indentation. A common indentor is the Vickers diamond.

This is a diamond point with a pyramidal geometry, leaving an indentation

that is visible as a square at the surface of the indented material. One

characteristic of indenting is that one material, in most cases a diamond,

remains in its original state, and the other material deforms. This

deformation is plastic, sometimes combined with brittle fracture, depending

on the materials properties. The situation is approXimately static after

the full load bas been applied. Indenting a material bas been modelled but

there still is discussion about several aspects, as for instance, details

of the near field stress state.

One step further than indenting is a single-point scratch with, e.g.,

a diamond. A single-point scratch also means that there is only one

material deforming. The relative movement between a single point and the

scratched material results in various additional phenomena, like the


wedging of material in front of and beside the moving point . Fracture of

the material is, amongst others, dependent on the relative velocity.

Modelling the single scratch has also been done, but as can be imagined,

there are many questions left open to answer.

During grinding multiple, in principle non-deforming points, are in

sliding contact with a deforming material . Abrasion, plastic deformation,

and fracture interfere with eachother and are dependent on the chosen

grinding conditions. Parameters like the type of grinding wheel, size of

the grains, the grinding method, cooling liquid, feed and velocity are

important process parameters. It is clear that the process of grinding is

more difficult to describe than the single-point scratch.

A tribological test is even more complicated. During unlubricated

wear, initially two materials slide against eachother. During this test

wear debris can be formed and this debris can be trapped between the two

surfaces. This results in so-called three body wear, where the debris

interacts with the two initial materials. These two or three materials are

thus continuously deforming. The mechanisms of deformation will depend on

the test conditions, like atmosphere, load, velocity, temperature, surface

geometry, etc. There are a vast number of unknown factors during a wear

test and this makes modeling very difficult. A large part of the literature

on wear of ceramics is either about well developed theoretical models that

are difficult to verify experimentally, or about experimental data with

some theoretical considerations. This study was performed to obtain a

better understanding of the behaviour of zirconia ceramics in relation to

mechanical surface interactions and to correlate the tribological behaviour

with the characteristics of the materials.

Chapter 1. Introduction 5

1.4 OUTLINE OF THIS THESIS

As mentioned, this thesis is about zirconia ceramics. The specific

properties of zirconias are ascribed to a phase transformation in these

materials. One part of the present study is about fundamental

characteristics caused by this phase transformation at the surface of the

materials, like residual stress and the transformation zone. The other part

is about some of the tribological properties of the materials. Since wear

is a process acting at the surface of a material, there is a clear

connection between the investigated fundamental characteristics and the

tribological properties.

The materials and experimental techniques used are described in

Chapter 2. The subdivision . in two parts, fundamental characteristics

related to the phase transformation at the surface, and tribology, is used

for the rest of this thesis. The specific transformation related properties

are treated in Chapters 3, 4 and 5. In Chapter 3, a literature review about

the transformation, grinding, and related subjects is given. In Chapter 4

[17], and Chapter 5 [18], the experiments, results, considerations and

conclusions, are given for this part of the project. The tribological

properties are treated in Chapters 6, 7 and 8. Chapter 6 is again a chapter

containing a literature review, while Chapter 7 [19], gives the experiments

and results from tests performed to in:vestigate some of the tribological

properties of the materials . . In Chapter 8 [20, 21], Pin-on-Plate and

Pin-on-Disk tests are described.

The last Chapter, Chapter 9, summarizes the results and discusses some

aspects of the relation between the transformation and the tribological

properties of zirconia ceramics. Interesting subjects for future research

are indicated.


References

I. R. C. Garvie, R. H. Hannink and R. T. Pascoe, Ceramic steel? Nature 258 (1975) 703.

2. R. C. Garvie, C. Urbani, D. R. Kennedy and J . C. McNeuer, Biocompatibility of magnesia-partially stabilized zirconia (Mg-PSZ) ceramics. J. Mater. Sci. 19 (1984) 3224.

3. J. L. Drummond, In vitro aging of yttria-stabilized zirconia. J. Am. Ceram. Soc. 72 (1989) 675.

4. Zirconia heating elements. Adv. Ceram. Rep. nov. 1991 , 6. 5. D. J. Viechnicki, M. J. Slavin and M. I. Kliman, Development and

current status of armor. Ceram. Bull 70 (1991) 1035. 6. Zirconia may find use as dental compound. Adv. Ceram. rep. Jan. 1991,

4. 7. A. Hendry, Processing of engineering ceramics. Powder Metall. 31

(1988) 20. 8. M. Woydt and K.-H. Habig, Tribologie keramischer Werkstoffe in

geschlossenen Systemen. Trib. Schmier. Tech. 37 (1990) 124. 9. C. T. Sims, Non metallic materials for gas turbine engines: are they

real? Adv. Mater. Proc. 6 (1991) 32. 10. D. Zeus, How the use of advanCed ceramics as tribomaterial has

affected the evolution of mechanical seals. cfi/Ber. DKG 68 (1991). 11. J . G. Baldoni, S. F. Wayne and S. T. Buljan, Cutting tool materials: mechanical properties wear-resistant relationships. ASLE Trans. 29 (1986) 347.

12. M. Fripan, U. Dworak and D. Fingale, Friction and wear of ceramic sliding and sealing elements. cfi/Ber. DKG 617 (1987) 239.

13. G. Schuseil, Keramik im Motorenbau. cfi/Ber. DKG 617 (1987) 242. 14. I. Birkby, P. Harrison and R. Stevens, The effect of surface

transformation on the wear behaviour of zirconia TZP ceramics. J. Eur. Ceram. Soc. 5 (1989) 37.

15. S. Aiyoshizawa, A. Wakigawa, D. Konno and K. Takagi, A study of ceramic bearings for vertical pumps. JSME, Int. J. Series II, 33 (1990) 41.

16. S. Jahanmir and F. E. Kennedy, Tribological Education: present status and future challenges. J. Trib. 113 (1991) 229.

17. P. H. J. van den Berg and G. de With, Residual stress and the stress-strain curve for Mg-PSZ. J. Europ. Ceram. Soc. 9 (1992) 265.

18. P. H. J. van den Berg and G. de With, Residual stress and strength of Mg-PSZ after grinding. Subm. to Wear.

19. P. H. J. van den Berg and G. de With, Wear and strength of Mg-PSZ, worn on hardened steel. J. Europ. Ceram. Soc. 8 (1991) 123.

20. P. H. J. van den Ber~, G. de With, L. Dortmans, E. Kokmeijer and G.-Z. Cao, Wear and friction of Y-TZP spheres reciprocating against various sialon plates. Subm. to J. Mater. Sci.

21. P. H. J. van den Berg, H. X. Willems and G. de With, Wear and friction of Y-TZP spheres reciprocating against various AlON plates. In preparation.

2. MATERIALS AND EXPERIMENTAL METHODS

In this chapter an overview of the common properties of most of the

materials used and some of the standard experimental methods to determine

these properties are presented. The specific experimental set-ups and

procedures used for the residual stress analysis and the wear tests as well

as various additional measurements are discussed later in the relevant

chapters. Only the basic experimental procedures and the results of

measurements of properties like chemical composition, density,

three-point-bend strength, fracture toughness, Young's modulus, Poisson's

ratio, Vickers hardness and the microstructure are presented here.

The materials used were the commercially purchased zirconia varieties

Mg-Partially-Stabilized-Zirconia, Mg-PSZ (Nilcra), Mg-PSZ (Feldmtihle)

Y-Tetragonal-Zirconia-Polycrystalline, Y-TZP (Feldmuhle), and Y-TZP

(Dynamic Ceramic). Other ceramics used were alumina (SWIP), alsint alumina

(Haldenwanger), various sialons developed at the Centre for Technical

Ceramics (CTK), and various AlONs 8lso developed at the CTK. One type of

metal was used in part of the wear tests, namely stavax, a commercially

available steel (Uddeholm). Part of the data on the material properties was

taken from suppliers brochures and part of the data was determined

experimentally as indicated in the tables presenting the results which will

be given in section 2.3.

Some of the data and experimental · procedures given here will be

repeated in the main part of this thesis for reasons of convenience for the

reader, and to present a complete set of data in the following chapters.

2.1 EXPERIMENTAL METHODS

Samples were prepared ·by sawing and grinding in order to obtain

8 Chapter 2. Materials and experimental methods

samples of the approximately correct dimensions. Specimens were ground to

the required dimensions using a diamond wheel, containing diamonds with a

maximum diameter of 46 !Jm (D46) or 56 !Jm (D56) and with water cooling. The

grinding wheel rotated in the same horizontal plane as the surface which is

to be ground. The rotation axis of the grinding wheel was orientated

perpendicular to this surface.

Quite often mechanical polishing is the most suitable surface

treatment to use as a preparation of surface tests for a ceramic.

Mechanical polishing is better reproducible than grinding and the

introduction of unwanted phenomena during polishing is less than for other

surface treatments. A polished surface is, however, not expected to be

entirely free of damage. Polishing was usually done with a diluted soap

solution on a tin disk, initially with diamond powder of 4-7 !Jm, and in the

final steps with diamond powder of 2-4 !Jm.

The dimensions of samples were measured with a thickness gauge

(Heidenhain, MT30, VT103) at regular positions on the samples. The accuracy

of the instrument was ± 1 !Jm.

The density of the materials was determined from the weight and

dimensions of several samples.

Young's modulus, E, and Poisson's ratio, v, were measured with the

pulse-echo method. Longitudinal waves at 5 MHz and transverse waves at 20

MHz were used. No correction for damping was applied.

Vickers hardness was measured with a Leitz hardness tester usually at

a load of 2.0 N on polished surfaces.

Three-point-bend tests to determine strength and fracture toughness

were performed on samples of approximately 15x3x1 mm3 using a span width of

12 mm at a dew point of -40 °C. The precise dimensions of each sample were

measured with the thickness gauge. The crosshead speed was 0.1 mm/min which

corresponded to a strain rate of approximately 1.2 %/min [1]. The fracture

Chapter 2. Materials and experimental methods 9

toughness measurements were performed on the same kind of samples,

containing a notch of about 0.1 mm width and 0.4 mm depth positioned

halfway at the lxl5 mm2 surface. A Knoop indentation at a load of 20 N was

positioned at the end of the notch to localize the crack and to obtain a

straight crack front. This method is referred to as the Single Edge Notched

Beam test (SENB).

A sample shape of the zirconia materials often used was a shape with

ten three-point-bend samples in a row still attached to their common base

also denoted as a 'cam' and shown in Fig. 2 .1.1. This particular shape made

it possible to perform wear and grinding tests on the surfaces of the

material and to obtain afterwards ten samples of 15x3xl mm3 suitable for a

three-point-bend test with the worn or ground surfaces of lx15 mm2 as the

surfaces under tension.

!

f 3

-~1

Fig. 2.1.1: Schematic illustration of the 'cam' shape.

The microstructure of the zirconias was visualized mainly by etching

a polished surface with HF for about 1 h. Thermal etching, e.g. 6 h at 1100

°C, is also a possible method to examine the microstructure of zirconia and

has been used as well.


The stavax disks used for some of the wear tests were thermally

hardened at a maximum temperature of 1000 °C for one minute.

Profiles of wear tracks on worn plates or disks were measured with a

Talysurf 5 (Rank-Taylor-Hobson, Leicester). The surface roughness

measurements were performed with the same instrument.

Optical examination of materials was performed with optical microscopy

using Interference Contrast (InCo, Leitz), and with Scanning Electron

Microscopy (SEM, Philips). An indication of the elements present on a

sample was obtained by a qualitative analysis with Energy Dispersive

Element Analysis by X-rays (EDX). Scanning Acoustic Microscopy (SAM) was

incidentally used to obtain information about subsurface phenomena. The

resolution was high enough to distinguish grain boundaries. Determination

of the various phases in zirconia was performed with X-ray analysis using

Cu-Ka radiation. The required peak areas were measured from diffractograms

with the help of a digitizer coupled to a personal computer.

2.2 MATERIALS CHARACTERIZATION

The characteristics of the various materials, partly measured

experimentally and partly from suppliers brochures as indicated, are given

in Tables 2.2.1-2.2.3. As mentioned in the introduction the characteristics

are only the basic properties aimed to present an overview, while a

detailed discussion on the investigated mechanical behaviour is given in

the main part of this thesis.

In Table 2.2.1 the compositions and grain sizes of most of the used

materials are given. The zirconias were mainly analyzed for the amount of

stabilizer and the amount of Hf. The sialons are a-sialons, IJ-sialons or

composites, which are sialons containing both the a- and IJ-phase. The

relative amount of a!IJ is indicated in Table 2.2.1. The AlONs differed in

Chapter 2. Materials and experimental methods

alumina content as shown in Table 2.2.1.

Table 2.2.2 presents the data for density,

Poisson's ratio. Table 2.2.3 gives the strength,

11

Young's modulus and

fracture toughness and

Vickers hardness data. The strength and fracture toughness results for the

zirconias were obtained as described in this chapter. The strength data for

sialons Al-A7 were obtained from biaxial tests [2) and the fracture

toughness data with indentation techniques [2). The strength of sialon B6

was measured also with biaxial tests and the fracture toughness with SENB

tests [3]. The data for the AlONs were obtained from [4).

The observed microstructures for the zirconia& Mg-PSZ and Y-TZP are

characteristic for these types of materials and they are similar to the

micrographs given in literature. There are extreme differences in

microstructure between Mg-PSZ and Y-TZP. In Mg-PSZ there is a cubic matrix

with ellipsoidal tetragonal precipitates within grains with a mean maximum

diameter of about 60 JIM. In Y-TZP there are mainly tetragonal grains with a .

mean maximum diameter of about 1.2 JIM and no precipitates. This difference

in grain size and the corresponding difference in (grain boundary

area)/(volume) ratio is one of the explanations for differences in the

mechanical behaviour of these materials.


material characteristics grain size, !Jm

Mg-PSZ, 2.5 % Mg, 0.5 % Hf, Ni. balance Zr and 0 61

Mg-PSZ, 2.05 % Mg, 1.63 % Hf Feld. balance Zr and 0

Y-TZP, 4.18 % Y, 1.61 % Hf Dyn. balance Zr and 0 1.15

Y-TZP, 4.18 % Y, 1.66 % Hf Feld. balance Zr and 0

stavax * 0.38 % c, 0.8 % Si, 0.5 % Mn 13.6 % Cr 0.3% v and mainly Fe

alsint 99.7 % Alp3

sialont A1 alP: 1 sialon A2 alP: 1 a-P: 4-0 sialon A3 alP: 1 a-P: 4-0 sialon A4 alP: 1 a-P: 4-0 sialon A5 alP: 0.75 a-P: 3-14 sialon A6 alP: 0.40 sialon A7 alP: 0.35

sialon B6 alP: 0 a-P: 0-30

alon*1 67.5 mol% Al 0 33 alon 2 77.5 mol% Aeo3 56 alon 3 73.0 mol% Al~O! 28

Table 2.2.1: The compositions and grain sizes of the materials used. See

text for further explanation.

* These tklta were taken from the suppliers brochure.

t The tklta on the sialons Al-A7 are from [2] and the tklta from sialon B6

are taken from [3].

:t: The tklta on the AIONs are taken from [4].

Chapter 2. Materials and experimental methods

material

type, supplier

Mg-PSZ, Ni.

Mg-PSZ, Feld.

Y-TZP, Dyn.

Y-TZP, Feld.

stavax * alsint

sialon A1 sialon A2 sialon A3 sialon A4 sialon AS sialon A6 sialon A7

sialon B6

alon 1 * alon 2 alon 3

density

p, g/cm3

5.73 ± O.Ql (7)

5.70 ± 0,01 (8)

5.86 ± 0.03 (5)

6.03 ± 0.01 (7)

6.7 3.8

3.33 t 3.29 3.28 3.27 3.23 3.22 3.23

3.13

3.68 3.65 3.67

Young's modulus

E, GPa

195.0 ± 3.0 (7)

199.7 ± 0.8 (8)

208.0 ± 2.0 (5)

210.0 ± 4.0 (12)

215 380

317 327 322 316 332 324 313

230

333 306 314

13

Poisson's ratio

v

0.324 ± 0.005 (7)

0.345 ± 0.001 (8)

0.321 ± 0.006 (5)

0.291 ± 0.005 (12)

0.28 0.29 0.29 0.28 0.30 0.30 0.29

0.30

0.32 0.26 0.25

Table 2.2.2 Density, Young's modulus and Poisson's ratio of the

materials. See text for further explanation. Values presented as x ± S (n)

stand for the mean x ± the sample standard deviation S for n samples.

Single values in the table are either based on two or three measurements or

the statistics of the measurement are not known.

* These data were taken from the suppliers brochure.

t The density values for the sialons Al-A7 are from [2) and the data for

sialon B6 are taken from [3].

* The data for the AIONs are taken from [4].


material

type, supplier

Mg-PSZ, Ni.

Mg-PSZ, Feld.

Y-TZP, Dyn.

Y-TZP, Feld.

strength, 3pb

a3

pb. MPa

761 ± 33 (10)

522 ± 39 (15)

966 ± 78 (28)

1130 ± 150(23)

fracture toughness

K , MPa.m 1n lc

11.5 ± 1.1 (20)

6.81 ±

7.90 ± 0.6 (10)

9.59 ± 1.07 (20)

alsint * 340

sialon AI sialon A2 sialon A3 sialon A4 sialon A5 sialon A6 sialon A7

sialon B6

alon 1 * alon 2 alon 3

517 ± 36 (~ 12) 5.7 ± 0.5 (~ 15) 519 ±55 (~12)5.6 ± 0.3 (~15) 570 ± 40 (~ 12) 5.4 ± 0.4 (~ 15) 553 ± 101(~ 12) 5.3 ± 0.3 (~ 15) 734 ± 36 (:2::: 12) 6.0 ± 0.7 (:2::: 15) 680 ± 50 (~ 12) 5.8 ± 0.6 (:2::: 15) 598 ± 73 (:2:::12)5.7 ± 0.7 (:2:::15)

480 ± 26 (9)

376 ± 45 (10) 408 ± 59 (10) 413 ± 42 (10)

2.8

2.3 ± 0.1 (5) 2.3 ± 0 .3 (5) 2.2 ± 0.3 (5)

Vickers hardness

HV, GPa

12

10.7 ± 1.5 (10)

13.4 ± 0.8 (6)

13.4 ± 0.8 (10)

21.0 ± 1.6 (~ 15) 19.6 ± 0.6 (:2:::15) 19.4 ± 0.8 (~ 15) 18.9 ± 1.0 (~ 15) 17.2 ± 1.5 (~ 15) 16.8 ± 1.0 (~ 15) 14.5 ± 1.0 (~ 15)

14.5 (2.0 N)

17.7 16.1 17.7

Table 2.2.3: Strength , fracture toughness and Vickers hardness (2 N load)

of the materials. Values presented as x ± S (n) stand for the mean x ± the

sample staruklrd deviation S for n samples. Single values in the table are

either based on two or three measurements or the the statistics of the test

is not known. See text for further explanation of the table.

* These data were taken from the suppliers brochure.

t The data for the sialons Al-Al are from [2] and the data for sialon B6

are taken from [3].

:t: The data for the AlONs are taken from {4].

Chapter 2. Materials and experimental methods 15

References

1.

2.

3.

4.

G. W. Hollenber~, G. R. Terwelliger and R. S. Gordon, Calculation of stresses and strains in four-point bending creep tests. J. Am. Ceram. Soc. 54 (1971) 196. G.-Z. Cao, Preparation and characterization of a-sialon ceramics. Ph.D. Thesis, Eindhoven Universitr of Technology, 1991. E. Kokmeijer, Sintering behaviOur and properties of P-Si3Al03N3 ceramics. Ph.D. Thesis, Eindhoven University of Technology, 1990. H. X. Willems, Preparation and properties of translucent )'-Aluminum Oxynitride. Ph.D. Thesis in preparation, Eindhoven University of Technology, 1992.

3. THE PHASE TRANSFORMATION IN ZIRCONIA

This chapter presents a general overview of some of the theory about

zirconia ceramics. There exists a vast amount of literature on this subject

in a variety of journals. This review is therefore not intended to be

complete. Most of the literature discussed is about subjects that are

relevant to the phenomena investigated. Various other phenomena are also

mentioned because their relevance to the investigated principles bas to be

recognized. One cannot state that connected subjects are irrelevant without

knowing anything about them. They are, however, only simply mentioned. A

detailed discussion on these subjects can be found in literature.

Zirconia ceramics is the name for a group of materials based on Zr01

.

A basic property that makes zirconia ceramics different from other ceramics

is the occurrence of a phase transformation of tetragonal zirconia to

monoclinic zirconia. This phase transformation is roughly spoken equivalent

to the martensitic phase change. The transformation from tetragonal to

monoclinic zirconia occurs upon cooling and one of the major features is

the accompanying increase in volume. This associated lattice dilatation

causes a relatively high toughness and strength. In Chapter 3.1 this phase

transformation is described in more detail. The energetic aspects of the

phase transformation are mentioned, and it is noted that there are some

severe problems to be solved in this area. The nucleation of the

transformation is a subject that bas not been clarified yet. There are two

different approaches and it is not clear which one is the right one.

It is important to know the phase content of these materials

quantitatively and some of the known equations to determine the phase

content are presented. Attention bas to be given to the possibility of the

occurrence of a reverse transformation which has been mentioned in

literature. Some remarks are made on this subject. Another aspect of the

material closely connected to the phase determination and the reversible

Chapter 3. The phase transformation in zirconia 17

transformation, is domain switching, which is discussed as well. Some

remarks are made about superplasticity and degradation, both important

phenomena for the TZP's.

The transformation has a major influence on the mechanical properties

of the materials. The transformation is the cause of a relatively high

toughness of zirconia ceramics compared to non-transformable zirconia

ceramics. There has thus been extensive attention to the mechanisms

underlying this so-called transformation toughening, Chapter 3.2. A related

aspect is the process that occurs at the crack tip of a transformable

material. Plastic deformation at the crack tip is illustrated with R-curve

behaviour. Microcracking is another concept that is discussed in literature

possibly contributing to the toughness. The transformation can be described

as plastic deformation and this plastic deformation can be presented in a

stress-strain relation.

The present study can be summarized as a study of the influence of

mechanical surface interactions on zirconia ceramics. One of the surface

interactions studied is grinding. Grinding causes the transformation at the

surface of the materials and the process of grinding is understood better

than sliding wear. Some references and considerations about grinding are

therefore presented in 3.3.

Fracture of structural ceramics is one of the most important aspects

of this type of materials and is discussed in 3.4. The specific

consequences of grinding on zirconia ceramics can be described with the

transformation zone and residual stress. These two concepts will especially

influence the strength and fracture of the materials and are also discussed

in section 3.4.

18 Chapter 3. The phase transformation in zirconia

3.1 THE PHASE TRANSFORMATION

Pure Zr02

at 2500 °C is cubic. Upon cooling, this cubic phase will

become tetragonal, and upon further cooling this tetragonal phase will

become monoclinic. A sample of pure ZrO cooled down to room-temperature is 2

completely monoclinic and cracked. This cracking of the samples is caused

by the dimensional changes that occur during the phase changes. Especially

the change from the tetragonal to the monoclinic phase is accompanied by

large strains and this results in cracking of the material.

In structural zirconia ceramics this phase transformation is used to

the benefit to develop materials with improved mechanical properties. This

has been accomplished by adding a stabilizer, for instance Mg, Y, Ce or Ca,

to Zr02

, thus stabilizing the cubic phase at lower temperatures. The phase

diagram for the system with Mg is shown as an example in Fig. 3.1.1. After

applying a suitable heat treatment, a material is obtained consisting at

room-temperature of a majot amount of tetragonal zirconia, and a small

amount of cubic and monoclinic zirconia. This tetragonal zirconia can

transform to monoclinic zirconia if energy, like thermal or mechanical

energy, is provided . This phase change is referred to as "the phase

transformation", and the specific properties of zirconia are ascribed

mainly to the occurrence of this phase transformation. Extensive attention

has been given in literature to these materials . Crystallographic aspects

are given in e.g. [1-16], in which especially Transmission Electron

Microscopy (TEM) and Selected Area Diffraction (SAD) are mentioned as tools

to examine the phases in detail. There are many more interesting aspects ,

like the orthorombic phase [3], the P-phase at the grain boundaries of

Mg-PSZ [4], the reverse transformation [5], and of course the mechanism of

the transformation , e.g. [2 , 11-13]. These aspects are important but most

of them have no significance to the subjects of this study.


3000r---------------------------~

~, ___ __

2500- ' ' --' ........................ _ ........ -

' --'' Cubic ss + liquid • •

-u . - t', \ '\

2000 ~ '\ Cubic; ss >-------I \ I 1 Cubic ss\ 1

1

I + \ I Cubic ss + MgO Tet.ss v

1500 yTet. ss r.J.4£2:£. ---------

...... _12~:£.. ___ _!e~g~a!..:,s...!., ~ _

. 1 1

Monoclinic1 ss + Mr;O 1000 ._ ____ __._ ______ _._ __ _

Zr01 10 20 30 40

MgO (mol .,.)

Fig. 3.1.1: An example of a phase diagram of Zr02

with MgO from ref. [9] as

usually found in literature.

The main consequence of the transformation is the dimensional change.

Comparison of the tetragonal and monoclinic unit cells results for example

in the following strains for the transformation:

til = 0.0041

£22 = 0.0128

£33 0.0217

£13 0.0772

with a positive strain indicating the larger dimensions of the monoclinic

structure relative to the tetragonal structure.

There is thus a substantial amount of shear and a volume dilatation of

about 4 %. A rotation of the lattice is also required to compare the unit

cells in the same orientation. The shear and the rotation are not

considered to be relevant for the investigated principles. The volume

20 Chapter 3. The phase transform£Jtion in zirconia

dilatation is the main consequence of the transformation.

An important aspect of the transformation is the energy balance.

Considerations on the energy balance are often presented in literature but

there is no clarity upon this subject [17-19]. It is a difficult problem

because there are many possible mechanisms that can contribute

significantly to the energy equations, but it is not possible to isolate

them to investigate the exact influence of each mechanism.

The nucleation of the transformation is also not without discussion

[20-23]. It is not clear whether the nucleation is homogeneous [20], or

heterogeneous [21-23]. But this discussion falls again outside the scope of

our present study.

The main tool used to investigate the phase content of samples was

X-ray diffraction as described in Chapter 2, although other methods can be

used as well, e.g. [24]. The interpretation of the resulting diffractograms

is important for the investigated phenomena and will be discussed in this

chapter because it is completely taken from literature. In Fig. 3.1.2 an

example is shown of an X-ray diffractogram of Mg-PSZ. It was interpreted

with the help of literature [25-31]. The main peaks, (11 I) at 28 = 28° and m

( 111) at 28 = 30° are clearly distinguishable. There is no debate about c+t

the qualitative part of the interpretation.


20.00 25.00 30.00 35.00 40.00

Fig. 3.1.2: An example of a dilfractogram as measured with Cu-Ka.

There still is some debate on the quantitative interpretation of the

diffractograms [25-31]. The relations mentioned are usually based on

calibration curves. The procedures used in this work to obtain a

quantitative estimate are described in [30]. In [31] the same equation is

improved by using the correct value for one of the constants used [25], and

this results in the following equation:

f 2.3741(111)

m

2.3741(111) + 1(111) m c

where f is the volume fraction of monoclinic zirconia and l(hkl) is the y

intensity, area under the peak, of reflection hkl of the y-phase

The required relation for the ternary system is even less certain

[29] . The best approximation is given by the above presented formula from

the monoclinic-cubic system, replacing the 1(111) value with the summation c

of 1(111) and 1(111) . I c

The formula used for Y-TZP was taken from [27]:


+ 3.llX f = --------=m=---- with

+0.311X m

I(lll) + I(III) X = -----=m _____ _:m:__ __

m I( Ill) + I ( II I) + I( Ill) m m t+c

Two other phenomena that could be relevant, namely domain switching

and the reverse transformation, are presented in literature [32-38]. Both

of these are mentioned as a possible explanation for observed

peak-intensity reversals in diffractograms of zirconia ceramics. These

peak-intensity reversals are visible on diffractograms that were, for

example, made on polished samples and on ground samples. Normally, in the

as-fired state, the (200)t intensity at 28 = 35° is larger than the (002)t

intensity at 28 34°. The same idea applies to the (131)t intensity at 28

= 60° and the (113)t intensity at 28 = 59°. Shifts in these 28 values are

possible, for instance, because of the presence of different stabilizers.

After processes like grinding or after heating, the relative

peak-intensities are reversed; that is, the (200)t and (131)t intensities

are decreased relative to the (002\ and (113)t intensities. The

explanation is that initially randomly oriented tetragonal zirconia gets

the opportunity to re-orientate into an energetically more stable

configuration with the c-axis perpendicular to the surface.

This observed reversal is modelled in two ways: first, with domain

switching and second, with re-transformation monoclinic ~ tetragonal. Both

of these possible mechanisms can be triggered by an increase in

temperature, which occurs at the surface of the material during grinding.

Experimental evidence in favor of one of these mechanisms is difficult to

obtain.

The consequences of domain switching are not clear. The mechanism is

only of importance for the deviatoric part of the strain tensor. There is


no volume change involved which is the same for the shear aspect of the

transformation that is considered irrelevant.

Two other phenomena have to be mentioned also, namely degradation

[39-46], and superplasticity [47-59]. These are important for Y-TZP and not

for Mg-PSZ. Degradation is the spontaneous transformation, and thus total

breakdown of the mechanical properties, of Y-TZP if heated to temperatures

between about 100 and 300 °C. Superplasticity means that the material is

capable to withstand extremely large deformations, at least more than 100

%, without failure. It is a property that can be used profitably for

shaping procedures.

3.2 MECHANICAL ASPECTS OF THE PHASE TRANSFORMATION

The influence of the transformation on the mechanical properties of

zirconia has been widely investigated [60-95], because the transformation

is seen as one of the possible solutions to the problem of the brittleness

of ceramics. The fracture toughness of materials containing transformable

tetragonal zirconia is namely significantly higher than the fracture

toughness of, for instance, cubic zirconia. It is well known that the

development of a crack causes the transformation of tetragonal zirconia in

a zone near the crack. The main idea is that the transformation caused by a

crack results in crack propagation retardation, either through energy

dissipation due to shear, twinning, domain switching, microcracking,

dilation, or through the direct closure of cracks.

The principles for the calculation of the increase in fracture

toughness due to the transformation are usually taken from [65]. The

transformation is modelled in simplified terms as follows. An inclusion,

initially free of stresses, is taken out of the bulk of material. This

inclusion 'transforms' without constraints and thus expands. Then it is

forced back into its original shape, causing internal stresses, after which


it is placed back in its original position in the bulk material. The

artificial constraints keeping the inclusion in its original shape are then

released, and the inclusion will expand into the material.

The description of the transformation in mechanical terms is quite

relevant to the present study. The main aspect of the transformation that

is of interest is the dilatation. Shear, twinning, domain switching,

rotation, the reverse transformation and microcracking, are important, but

not in particular relevant to the specific investigated properties. The

correspondence between the models derived later on using this

simplification, and the experimental results will illustrate that this

point of view is reasonable. The description of the transformation in

mechanical terms is possible with a stress-strain curve [76-80]. A

stress-strain curve is a one-dimensional graph. A problem is thus

encountered, because deformation is always three-dimensional. The

translation from three to one dimension involves major simplifications. A

common stress-strain curve begins with an elastic part determined by the

Young's Modulus E. The first part of the curve is defined by a Ee.

Extremely brittle materials will fail during elastically deformation.

Ideally plastically deforming materials will deform elastic until the flow

stress, a, y

is reached and will then deform at a constant stress, a. y

The

plastic part of deformation is given by a horizontal line in the

stress-strain curve. The transformation in zirconia, restricted to the

dilatation, is described in mechanical terms as transformation plasticity.

The use of this term is not unambiguous because the definition of plastic

deformation is not clear, but it is the most suitable. Drawing a

stress-strain curve for zirconia will thus result in a graph beginning with

the elastic part, until the transformation occurs. The second part of the

curve, describing the transformation plasticity, has been a subject of

interest. The slope of this part can be negative, zero or positive. A

negative slope is in most cases referred to as a supercritical

Chapter 3. The phase transfomwtion in zirconia 25

transformation. Assuming a negative slope simplifies the modelling of the

toughening significantly, but direct experimental measurements of

stress-strain curves on zirconia only indicate a positive slope [76-79].

The present study follows a different experimental approach to the

determination of the second part of the curve, and the results are

consistent with the results presented in [76-79].

The relation between the phase transformation and fracture toughness

bas been quantified by various equations. These equations depend on chosen

boundary conditions. One formula often presented is based on the assumption

of a supercritical transformation, and on the assumption of a constant

amount of transformation along b, the transformation zone depth. The

results from the present study as presented in Chapter 4 do not support

these assumptions. This formula often quoted is given by:

LIK = 0.22EfeT Vh I (1-v) lc li

where LIK is the increase in fracture toughness, E is Young's modulus, f Ic

is the fraction of transformed zirconia, e~1 is the dilatation associated

with the transformation, h is the transformation zone depth and v is

Poisson's ratio.

A different approach to describe the effect of the transformation on

cracks in a material, is through crack resistance curves, R-curves [81-87].

The fracture-energy for materials exhibiting R-curve behaviour depends on

crack length for relatively small cracks. Typical R-curve behaviour is

connected with plastic deformation at the crack tip.

Another concept associated with the transformation that might have a

significant influence on toughness is microcracking [88-95]. The derivation

of the amount of toughening due to microcracking is analogous to the

calculation of the toughening due to the volume dilatation.

Most of the calculations for the amount of toughening due to each

mechanism, like shear, twinning, domain switching, microcracking,

dilatation, are based on energy dissipation. A problem arises, however,

=2::..6 ____________ fhapter 3. The phase transformation in zirconia

when all the possible energy dissipative mechanisms are used to calculate

the fracture toughness. This results in a value that is far too high.

Difficulties with the interpretation of the energy equations were also

illustrated in the earlier mentioned references [17-19].

3.3 GRINDING

This study was done to investigate the influence of mechanical surface

interactions on zirconia ceramics. The two surface interactions chosen were

grinding and wear. Grinding is a method used often to obtain samples of the

required dimensions. It could also present information about processes that

could be important during wear. Grinding comprises a major field of

research, but in this case it is only used as a reproducible method to

introduce the transformation at the surface and only a few remarks will be

made about grinding itself. Some fundamental characteristics are given in

[96-104]. Especially the relation between diamond grain shape and the

forces on the ground material, as well as the abrasive processes occurring

during grinding are of interest [98-104].

Grinding will not only cause the transformation in zirconia but could

have a lot more influence. It is therefore important to know something

about the influence of grinding on materials [105-115].

3.4 FRACTURE, RESIDUAL STRESS AND

TRANSFORMATION ZONE DEPTH

One of the major advantages of zirconia is its high fracture

toughness, although it is not yet comparable to the toughness of steel.

Fracture of zirconia remains important. Brittle fracture is characteristic

for glasses and most ceramics [116-119], and contains aspects that are also

relevant to zirconia. Failure of zirconia ceramics cannot be described by


brittle fracture and not by plastic failure. Intermediate descriptions as

used by the principle of the J-integral [120], should be used. These

problems, together with usual concepts as slow crack growth, fatigue,

microcracks and fractography makes the study of fracture of zirconia

interesting and difficult [121-127].

One of the major subjects of interest for this study is the shape and

dimension of the transformation zone in zirconia after surface interactions

like grinding and wear. Experimental determination of this transformation

zone has been presented in literature using various methods and with

various results [128-136]. Raman microprobe has been used [129-131], X-ray

diffraction [132, 136], synchroton radiation [134], and MoirE!

interferometry [135]. The method used for this study, polishing material

from the surface in small steps, is far more laborious, but it does give

direct information.

The consequence of a transformation zone caused by surface

interactions, is a residual stress zone. Residual stress can be measured

with various techniques [137-142]. The most practical consequence of

residual stress is the influence of this residual stress on strength, and

thus on fracture [143-152]. There has been theoretical research on this

subject and the present study gives some results of experiments on the

relation between residual stress and strength. It is this relation that

also combines the wear tests with the grinding tests, because both the wom

surfaces and ground surfaces were tested in strength tests.

References 1. R. Chaim, M. Ruhle and A. H. Heuer, Microstructural evolution in a

Zr02-12wt% Yz03 ceramic. J. Am. Ceram. Soc. 68 (1985) 427. 2. M. L. Mccartney and M. Ruhle, In situ transmission electron microscopy

observations of the monoclinic to tetragonal phase transformation in tetragonal Zr02. Acta Metall. 37 (1989) 1859.

3. D. B. Marshall and M. R. James, Structural and mechanical property changes in toughened magnesia-partially-stabilized zirconia at low temperature. J. Am. Ceram. Soc. 72 (1989) 218.

4. F.-C. Wu and S.-C. Yu, The effect of b-phase Mgzzr,Otz, on the stabilization of the tetragonal phase in MgO-PSZ. Mat. Res. Bull. 23

28

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

Chapter 3. The phase transformation in zirconia

(1988) 467. M. Yoshimura, Phase stability of zirconia. Ceram. Bull. 67 (1988) 1950. B. C. Muddle and P. M. Kelly, Stress-activated martensitic transformation and transformation plasticity. Mater. Forum 11 ( 1988) 182. M. V. Swain and R. H. J. Hannink, Metastability of the martensitic transformation in a 12 mol% ceria-zirconia alloy: II, grinding studies. J. Am. Ceram. Soc. 72 (1989) 1358. M.Shibata-Yanagisawa, M. Kato, H. Seto, N. Ishizawa, N. Mizutani and M. Kato, Crystallographic analysis of the cubic-to-tetragonal phase transformation in the Zr02-Y203 system. J. Am. Ceram. Soc. 70 (1987) 503. R. H. J. Hannink, Microstructural development of sub-eutectoid aged Mg0-Zr02 alloys. J. Mater. Sci. 18 (1983) 457. R. Chaim and D. G. Brandon, Microstructure evolution and ordering in commercial Mg-PSZ. J. Mater. Sci. 19 (1984) 2934. R. R. Hugbhan and R. H. J . Hannink, Precipitation during controlled cooling of magnesia-partially-stabilized zirconia. J. Am. Ceram. Soc. 69 (1986) 556. G. K. Bansal and A. H. Heuer, Precipitation in partially stabilized zirconia. J. Am. Ceram. Soc. 58 (1975) 235. P. M. Kelly and C. J. Ball, Crystallo~aphy of stress-induced martensitic transformations in partially stabilized zirconia. J. Am. Ceram. Soc. 69 (1986) 259. G. K. Bansal and A. H. Heuer, On a martensitic phase transformation in zirconia (ZrOl)-1. metallographic evidence. Acta Metall. 20 (1972) 1281. S. Chen and P. Shen, Polymorphic transformation of t' phase in yttria partially stabilized zirconia. Mater. Sci. Eng. A123 (1990) 145. N. M. Ghoneim and S. B. Hanna, Sintering and microstructure of ultrafine yttria-zirconia compacts. J. Mater. Sci. 25 (1990) 5192. F. Guiu and R. N. Stevens, Physical interpretation of fracture-toughening mechanisms. J . Mater. Sci. 26 (1991) 4375. W. Kreher and W. Pompe, Increased fracture toughness of ceramics by energy-dissipative mechanisms. J. Mater. Sci. 16 (1981) 694. F. F. Lange, Transformation toughening part 1 size effects associated with the thermodynamics of constrained transformations . J. Mater. Sci. 17 (1982) 225. A. H. Heuer and M. Ruhle, On the nucleation of the martensitic transformation in zirconia. Acta Met. 33 (1985) 2101. 1.-W. Chen and Y.-H. Chiao, Martensitic nucleation in Zr02. Acta Metall. 31 (1983) 1627. 1.-W. Chen and Y.-H. Chiao, Theory and experiment of martensitic nucleation in Zr02 containing ceramics and ferrous alloys. Acta Metall. 33 (1985) 1827. 1.-W. Chen, Y.-H. Chiao and K. Tsuzaki , Statistics of martensitic nucleation. Acta Metall. 33 (1985) 1847. R. Srinivasan, R. J. De Angelis, G. Ice and B. H. Davis, Identification of tetragonal and cubic structures of zirconia using synchroton x-radiation source. J. Mater. Res. 6 (1991) 1287. J. D. McCullough and K. N. Trueblood, The crystal structure of baddeleyite (monoclinic Zr02). Acta Cryst. 12 (1959) 507. D. K. Smith and H. W. Newkirk, The crystal structure of baddeleyite (monoclinic Zr02) and its relation to the polymorphism of Zr02. Acta Cryst. 18 (1965) 983. H. Toraya, M. Yoshimura and S. Somiya, Calibration curve for quantitative analysis of the monoclinic-tetragonal Zr02 system by X-ray diffraction. Comm. Am. Ceram. Soc. (1984) C-119 . . R. C. Garvie and P. S. Nicholson, Phase analysis in zirconia systems.


29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

J. Am. Ceram. Soc. 55 (1972) 303. H. K. J. Schmid, Quantitative analysis of polymorphic mixes of zirconia by X-ray diffraction. J. Am. Ceram. Soc. 70, (1987) 376. D. L. Porter and A. H. Heuer, Microstructural development in MgO-Partially Stabilized Zirconia (Mg-PSZ). J. Am. Ceram. Soc. 62 (1979) 298. P. A. Evans, R. Stevens and J. G. P. Binner, Quantitative X-ray diffraction analysis of polymorphic mixes of pure zirconia. Br. Ceram. Trans. J. 83 (1984) 39. D. B. Marshall and M. R. James, Reversible stress-induced martensitic transformation in Zr<n. J. Am. Ceram. Soc. 69 (1986) 215. C. L. Lin, P. Shen and D. Gan, Reverse martensitic transformation in yttria-partially stabilized zirconia with Ti02 addition. Mater. Sci. Eng. A128 (1990) Ll. C.-J. Chan, F. F. Lange, M. Ruhle J.-F. Jue and A. V. Virkar, Ferroelastic domain switching in tetragonal zirconia single crystals -microstructural aspects. J. Am. Ceram. Soc. 74 (1991) 807. B.-S. Li, J.-S. Cherng, K. J. Bowman and 1.-W. Chen, Domain switching as a toughening mechanism in tetragonal zirconia. J. Am. Ceram. Soc. 71 (1988) C-362. A. V. Virkar and R. L. K. Matsumoto, Ferroelastic domain switching as a toughening mechanism in tetragonal zirconia. J. Am. Ceram. Soc. 69 (1986) C-224. G. V. Srinivasan, J.-F. Jue, S.-Y. Kuo and A. V. Virkar, Ferroelastic domain switching in polydomain tetragonal zirconia single crystals. J. Am. Ceram. Soc. 72 (1989) 2098. M. G. Cain and M. H. Lewis, Evidence of ferroelasticity in Y-tetragonal zirconia polycrystals. Mat. L. 9 (1990) 309. J. F. Jue, J. Chen and A. V. Virkar, Low-temperature aging of t'-zirconia: the role of microstructure on phase stability. J. Am. Ceram. Soc. 74 (1991) 1811. J. Nowotny, M. Sloma and W. Weppner, Surface reactivity of yttria-doped zirconia with oxygen. Solid State Ionics 32/33 (1989) 709. S. Schmauder and the martensitic zirconia polycrystals 534.

H. Schubert, Significance of internal stresses for transformation in yttria-stabilized tetragonal

during degradation. J. Am. Ceram. Soc. 69 (1986)

T. Sato and M. Shimada, Transformation of yttria-doped tetragonal Zr02 polycrystals by annealing in water. J. Am. Ceram. Soc. 68 (1985) 356. T. Masaki, Mechanical properties of Y203-stabilized tetragonal Zr02 polycrystals after ageing at high temperature. J. Am. Ceram. Soc. 69 (1986) 519. M. Yoshimura, T. Noma, K. Kawabata and S. SOmiya, Role of H20 on the degradation process of Y-TZP. J. Mater. Sci. L. 6 (1987) 465. T. Sato and M. Shimada, Control of the tetragonal-to-monoclinic phase transformation of yttria partially stabilized zirconia in hot water. J. Mater. Sci. 20 (1985) 3988. N. L. Hecht, D. E. McCullum, G. A. Graves and S. D. Jang, Environmental effects in toughened ceramics. Ceram. Eng. Sci. Proc. 8 (1987) 892. T.-G. Nieh and J. Wadsworth, Dynamic grain growth during superplastic deformation of yttria-stabilized tetragonal zirconia polycrystals. J. Am. Ceram. Soc. 72 (1989) 1469. Y. Kitano, Y. Mori, A. Ishitani and T. Masaki, Structural chan~es by compressive stresses of 2.0-mol %-yttria-stabilized tetragonal ztrconia polycrystals. J. Am. Ceram. Soc. 72 (1989) 854. A. Dominguez-Rodriguez, K. P. D. Lagerlf and A. H. Heuer, Plastic deformation and solid-solution hardening of Y203-stabilized Zr02. J. Am. Ceram. Soc. 69 (1986) 281.

~--~--.


50. A. H. Heuer, V. Lanter and A. Dominguez-Rodriguez, High-temperature precipitation hardening of Y 203 partially-stabilized ZrOl (Y-TZP) single crystals. Acta Metall. 37 (1989) 559.

51. M Nauer and C. Carry, Creep parameters of yttria doped zirconia materials and superplastic deformation mechanisms. Scripta Metall. Mater. 24 (1990) 1459.

52. T. Masaki, Mechanical properties of toughened Zr02-Yz03 ceramics. J. Am. Ceram. Soc. 69 (1986) 638.

53. J. Lankford, Deformation and fracture of yttria-stabilized zirconia single crystals. J. Mater. Sci. 21 (1986) 1981.

54. R. Duclos, J. Crampon and B. Amana, Structural and topological study of superplasticity in zirconia polycrystals. Acta Metall. 37 (1989) 877.

55. T. G. Nieh and J. Wadsworth, Superplastic behaviour of a fme-grained yttria-stabilized, tetragonal zirconia polycrystal (Y-TZP). Acta Metall. Mater. 38 (1990) 1121.

56. J. Lankford, The influence of temperature and loading rate on flow and fracture of partially stabilized zirconia. J. Mater. Sci. 20 (1985) 53.

57. T. Masaki and K. Sinjo, Mechanical properties of highly toughened Zr02-Yz03. Ceram. Int. 13 (1987) 109.

58. 1.-W. Chen and L. A. Xue, Development of superplastic structural ceramics. J . Am. Ceram. Soc. 73 (1990) 2585.

59. J. Lankford, R. A. Page and L. Rabenberg, Deformation mechanisms in yttria-stabilized zirconia. J. Mater. Sci. 23 (1988) 4144.

60. R. H. J. Hannink and M. V. Swain, A mode of deformation in partially stabilized zirconia. J. Mater. Sci. L. 16 (1981) 1428.

61. 1.-W. Chen and P. E. Reyes Morel, Implications of transformation plasticity in ZrOl-containing ceramics: I, shear and dilatation effects. J. Am. Ceram. Soc. 69 (1986) 181.

62. N. Claussen, Umwandlungsverstrkte keramische Werkstoffe. Z. Werkstofftech. 13 (1982) 138.

63. F. F. Lange, Transformation toughening part 2 contribution to fracture toughness. J. Mater. Sci. 17 (1982) 235.

64. A. G. Evans and R. M. Cannon, Toughening of brittle solids by martensitic transformations. Acta Met. 34 (1986) 761.

65. J. D. Eshelby, The determination of the elastic field of an ellipsoidal inclusion, and related problems. Proc. R. Soc. London, Ser. A, A241, (1957) 376.

66. A. G. Evans, Perspective on the development of high-toughness ceramics. J . Am. Ceram. Soc., 73 (1990) 187.

67. R. M. McMeeking and A. G. Evans, Mechanics of transformation-toughening in brittle materials. J. Am. Ceram. Soc. 65 (1982) 242.

68. D. L. Porter, A. G. Evans and A. H. Heuer, Transformation-toughening in partially-stabilized zirconia (PSZ). Acta Metall. 27 (1979) 1649.

69. A. G. Evans and A. H. J. Heuer, Review - transformation toughening in ceramics: martensitic transformations in crack-tip stress fields. J. Am. Ceram. Soc., 63 (1980) 241.

70. B. Budiansk:y, J. W. Hutchinson and J. G. Lambropoulos, Continuum theory of dilatant transformation toughening in ceramics. Int. J. Solids Structure 19 (1983) 337.

71. A. H. Heuer, M. Ruhle and D. B. Marshall , On the thermoelastic martensitic transformation in tetragonal zirconia. J, Am. Ceram. Soc. 73 (1990) 1084.

72. P. F. Becker, M. V. Swain and M. K. Ferber, Relation of transformation temperature to the fracture toughness of transformation-toughened ceramics. J. Mater. Sci. 22 (1987) 76.

73. M. J. Readey, A. H. Heuer and R. W. Steinbrech, Annealing of test specimens of high-toughness magnesia partially stabilized zirconia. J.


74.

75.

76.

77.

78.

79.

80.

81.

82.

83.

84.

85.

86.

87.

88.

89.

90.

91.

92.

93.

94.

95.

96.

Am. Ceram. Soc. 71 (1988) C-2. T. K. Gupta, F. F. Lange and J. H. ~htold, Effect of stress-induced phase transformation on the properties of polycrystalline zirconia containing metastable tetragonal phase. J. Mater. Sci. 13 (1978) 1464. Ingels, E., Heuer, A, H. & Steinbrech, R. W., Fracture mechanics of high-toughness Magnesia-Partially- Stabilized-Zirconia. J. Am. Ceram. Soc. 73 (1990) 2023. 1.-W. Chen, Implications of transformation plasticity in ZrOl-containing ceramics: II, elastic-plastic indentations. J. Am. Ceram. Soc. 69 (1986) 189. 1.-W. Chen and P. E. Reyes-Morel, Transformation plasticity and transformation toughening in Mg-PSZ and Ce-TZP. In 'Mat Res. Soc. Symp. Proc. 78 (1987) 75, Advanced structural Ceramics', eds. Becker, P. F., Swain, M. V. & Somiya, S. E. Inghels, A. H. Heuer and R. W. Steinbrech, Fracture mechanics of high-toughness magnesia-partially-stabilized zirconia. J. Am. Ceram. Soc. 73 (1990) 2023. I.-W. Chen, Model of transformation toughening in brittle materials. J. Am. Ceram. Soc. 74 (1991) 2564. N. Ramakrishnan, H. okada and S. Atluri, Computer simulation of transformation induced plasticity using finite element method. Acta Metall. Mater. 39 (1991) 1297. D. B. Marshall, Strength characteristics of transformation-toughened zirconia. J. Am. Ceram. Soc. 69 (1986) 173. L. R. F. Rose and M. V. Swain, Two R curves for partially stabilized zirconia. J. Am. Ceram. Soc. 69 (1986) 203. S. J. Bums and M. V. Swain, Fracture toughness of MgO-partially-stabilized ZrOl specimens with KR-curve behavior from transformation toughening. J. Am. Ceram. Soc. 69 (1986) 226. M. V. Swain, Inelastic deformation of Mg-PSZ and its significance for strength-toughness relationship of zirconia toughened ceramics. Acta Met. 33 (1985) 2083. D. B. Marshall and M. V. Swain, Crack resistance curves in magnesia-partially-stabilized zirconia. J. Am. Ceram. Soc. 71 (1988) 399. S. Srinivasan and R. 0 . Scattergood, R-curve measurements in PSZ ceramics. J, Mater. Res. 5 (1990) 1490. A. H. Heuer, M. J. Readey and R. Steinbrecb, Resistance curve of supertough MgO-partially-stabilized ZrOl. Mater. Sci. Eng. A105/106 (1988) 83. P. E. Reyes-Morel and 1.-W. Chen, Stress-biased anisotropic microcracking in zirconia polycrystals. J. Am. Ceram. Soc. 73 (1990) 1026. W. Pompe, H.-A. Bahr, G. Gille and W. Kreher, Increased fracture toughness of brittle materials by microcracking in an energy dissipative zone at the crack tip. J. Mater. Sci. L. 13 (1978) 2720. A. G. Evans, On the formation of a crack tip microcrack zone. Scripta Metall. 10 (1976) 93. Y. Fu and A. G. Evans, Some effects of microcracks on the mechanical properties of brittle solids 1. stress, strain relations. Acta Metall. 33 (1985) 1515. A. G. Evans and K. T. Faber, Crack-growth resistance of microcracking brittle materials. J. Am. Ceram. Soc. 67 (1984)255. D. R. Clarke, Microfracture in brittle solids resulting from anisotropic shape changes. Acta Metall. 28 (1980) 913. N. Claussen, J. Steeb and R. F. Pabst, Effect of induced microcracking on the fracture toughness of ceramics. Ceram. Bull. 56 (1977) 559. J. W. Hutchinson, Crack tip shielding by micro-cracking in brittle solids. Acta Metall. 35 (1987) 1605. S. Y. Chen, T. N. Farris and S. Chandrasekar, Sliding microindentation

32

97.

98.

99.

100.

101.

102.

103.

104.

105.

106.

107.

108.

109.

110.

111.

112.

113.

114.

115.

116.

117.

118.

119.

120.

Clwpter 3. The plwse transformation in zirconia

fracture of brittle materials. Trib. Trans. 34 (1991) 161. A. Broese van Groenou, Wat gebeurt er bij slijpen van glas en keramiek? Mikroniek 6 ( 1977) 121. A. A. Torrance, An approximate model of abrasive cutting. Wear 118 (1987) 217. M. Abebe and F. C. Appl, Theoretical analysis of the basic mechanics of abrasive processes part 1: general model. Wear 126 (1988) 251. M. Abebe and F. C. Appl, Theoretical analysis of the basic mechanics of abrasive processes part 2: model of the ploughing process. Wear 126 (1988) 267. P. Gilormini and E. Felder, Theoretical and experimental study of the ploughing of a rigid-plastic semi-infinite body by a rigid pyramidal mdenter. Wear 88 (1983) 195. A. A. Torrance, A new approach to the mechanics of abrasion. Wear 67 (1981) 233. M. de Yathaire, F. Delamare and E. Felder, An upper bound model of ploughing by a pyramidal indenter. Wear 66 (1981) 55. A. A. Torrance, The correlation of process parameters in grinding. Wear 139 (1990) 383. T. J. Larchuk, J. C. Conway and H. P. Kirchner, Crushing as a mechanism of material removal during abrasive machining. J . Am. Ceram. Soc. 68 (1985) 209. J. J. Mecholsky, Jr., S. W. Freiman and R. W. Rice, Effect of grinding on flaw geometry and fracture of glass. J. Am. Ceram. Soc. 60 (1977) 114. H. P. Kirchner, Damage penetration at elongated machining grooves in bot-pressed Si3N4. J. Am. Ceram. Soc. 67 (1984) 127. H. P. Kirchner and E. D. Isaacson, Residual stresses in hot-pressed Si3N4 grooved by single-point grinding. J. Am. Ceram. Soc. 65 (1982) 55. H. P. Kirchner, Comparison of single-point and multipoint grinding damage in glass. J. Am. Ceram. Soc. 67 (1983) 347. R. M. Hooper and J. E. Morgan, Wear processes in the diamond grinding of zirconia ceramics. Wear 130 (1989) 353. R. W. Rice and B. K. Speronello, Effect of micromicrostructure on rate of machining of ceramics. J. Am. Ceram. Soc. 59 (1976) 330. W. Kanematsu, S. Sakai, Y. Yamauchi, S. Ito and K. Kubo, Effects of grinding on dynamic fatigue behavior and strength degradation of bot pressed silicon nitride ceramics at room temperature. J. Ceram. Soc. Jap. Int. Ed. 98 (1991) 1296. J. C. Conway, Jr. and H. P. Kirchner, Crack branching as a mechanism of crushing during grinding. J. Am. Ceram. Soc. 69 (1986) 603. J. S. Reed and A.-M. Lejus, Affect of grinding and polishing on near-surface phase transformations in zirconia. Mat. Res. Bull. 12 (1977) 949. R. 0 . Scattergood, S. Srinivasan, T. G. Bifano and T . A. Dow, R-curve effects for machining and wear of ceramics. Cer. Acta 3 (1991) 52. C. Cm. Wu, S. W. Freiman, R. W. Rice and J. J . Mecholsky, Microstructural aspects of crack propagation in ceramics. J. Mater. Sci. 13 (1978) 2659. J. J. Mecholsky, Jr. , S. W. Freiman and R. W. Rice, Fracture surface analysis of ceramics. J. Mater. Sci. 11 (1976) 1310. J. J. Mecholsky, Jr., R. W. Rice and S. W. Freiman, Prediction of fracture energy and flaw size in glasses from measurements of mirror size. J. Am.Ceram. Soc. 57 (1974) 440. S. W. Freiman, Brittle fracture behavior of ceramics. Ceram. Bull. 67 (1988) 1. J. R. Rice, A path independent integral and the approximate analysis of strain concentration by notches and cracks. Appl.Mech. 35 (1968) 379.

Chapter 3. The phase transfortrWtion in zirconia 33

121. P. F. Becher, Subcritical crack growth in partially stabilized ZrOz (MgO). J. Mater. Sci. 21 (1986) 297.

122. R. H. Dauskardt, W. Yu and R. 0. Ritchie, Fatigue crack propagation in transformation toughening zirconia ceramics. J. Am. Ceram. Soc. 70 (1987) C-248.

123. R. H. Dauskardt, D. B. Marshall and R. 0. Ritchie, Cyclic fatigue-crack propagation in magnesia-partially-stabilized zirconia ceramics. J. Am. Ceram. Soc. 73 (1990) 893.

124. D. G. Jensen, V. Zelizko and M. V. Swain, Small flaw static fatigue crack growth in Mg-PSZ. J. Mater. Sci. L. 8 (1989) 1154.

125. K. No~chi, Y. Matsuda, M. Oishi, T. Masaki, S. Nakayama and M. Mizushina, Strength analysis of yttria-stabilized-tetragonal zirconia polycrystals. J. Am. Ceram. Soc. 73 (1990) 2667.

126. V. Zelizko and M. V. Swain, Influence of surface preparation on the rotating flexural fatigue of Mg-PSZ. J. Mater. Sci. 23 (1988) 1077.

127. J. Sung and P. S. Nicholson, Strength improvement of yttria-partially-stabilized zirconia by flaw elimination. J. Am. Ceram. Soc. 71 (1988) 788.

128. R. W. Steinbrech, E. Inghels and A. H. heuer, Residual displacement effects during crack propagation in high-toughness magnesia-partially-stabilized zirconia. J. Am. Ceram. Soc. 73 (1990) 2016.

129. D. B. Marshall, M. C. Shaw, R. H. Dauskardt, R. 0. Ritchie, M. J. Readey and A. H. Heuer, Crack tip transformation zones in toughened zirconia. J. Am. Ceram. Soc. 73 (1990) 2659.

130. C. A. Leach, A Raman microprobe study of a magnesia partially stabilized zirconia fracture surface. J. Mater. Sci. 24 (1989) 1380.

131. D. R. Clarke and F. Adar, measurement of the crystallographically transformed zone produced by fracture in ceramics containing tetragonal zirconia. J. Am. Ceram. Soc. 65 (1982) 284.

132. Y. Mori, Y. Kitano and A. Ishitani, X-ray determination of tranformed zone size in toughened zirconia ceramics. C. J. Am. Ceram. Soc. 71 (1988) C-322.

133. C.-H. Hsueh and P.F. Becher, Some considerations of nonideal transformation zone profile. J. Am. Ceram. Soc. 71 (1988) 494.

134. K. D. Keefer and T. A. Michalske, Determination of phase transformation depth profiles with synchrotron radiation. J. Am.Ceram. Soc. 70 (1987) 227.

135. M. S. Dadkhah, D. B. Marshall, W. L. Morris and B. N. Cox, Direct measurement of transformation zone strains in toughened zirconia. J. Am. Ceram. Soc. 74 (1991) 584.

136. R. C. Garvie, R. H. J. Hannink and M. V. Swain, X-ray analysis of the transformed zone in partially stabilized zirconia (PSZ). J. Mater. Sci. L. 1 (1982) 437.

137. A. B. Brenner and S. Senderoff, Calculation of stress in electrodeposits from the curvature of a plated strip. J. Research 42 (1949) 105.

138. R. Tandon and D. J. Green, Residual stress determination with strain gage measurements. J. Am. Ceram. Soc. 73 (1990) 2628.

139. A. V. Virkar, Determination of residual stress profile using a strain gage technique. J. Am. Ceram. Soc. 73 (1990) 2100.

140. C.-C. Chiu, Determination of the elastic modulus and residual stresses in ceramic coatings using a strain gage. J. Am. Ceram. Soc. 73 (1990) 1999.

141. J. H. Root, J. D. Sullivan and B. R. Marple, Residual stress in alumina-mullite composites. J. Am. Ceram. Soc. 74 (1991) 579.

142. R. G . . Treuting and W. T. Read Jr., A mechanical determination of biaxial residual stress in sheet metals. J. Appl. Phys. 22 (1951) 130.

143. R. Tandon, D. J. Green and R. F. Cook, Surface stress effects on indentation fracture sequences. J. Am. Ceram. Soc. 73 (1990) 2619.

4. RESIDUAL STRESS AND THE STRESS-STRAIN CURVE FOR MG-PSZ

4.1 INTRODUCTION

Some information presented here therefore has already been given in

Chapter 2 or 3. The results and conclusions from the experiments described

in this Chapter are used in Chapter 5. The material that is examined here

is Mg-PSZ (Nilcra).

The phase transformation in zirconia ceramics has been described

thoroughly in crystallographic terms, Chapter 3 and for instance [1-3], but

the description in mechanical terms is not yet clear, as discussed in

Chapter 3. In general [4, 5], a stress-strain curve as shown in Fig. 4.1 is

presented.

c; pp

Fig. 4.1: Illustration of . the three possible stress-strain curves for

Mg-PSZ, where u stands for mean stress, £ for the dikltational strain, B m W

the eklstic bulk modulus and IJ for the transformation bulk modulus.

The first part of the curve is determined by the elastic properties of

the Mg-PSZ. In a hydrostatic stress state, the elastic deformation is given

36 Chapter 4. Residual stress and the stress-strain curve for Mg-PSZ

by the slope of the curve, also referred to as the bulk modulus B. The

transformation to the monoclinic phase starts after a threshold stress [6,

7] has been reached. This threshold is referred to as the critical

transformation stress, a. c

The description of this a in terms of the

principal stresses is, however, not clear. Its value depends on the stress

state, but the quantitative relation is not known.

The slope of the second part of the curve, given by the transformation

modulus :S in a hydrostatic stress state (or more generally by the

work-hardening coefficient in other stress states), is still uncertain.

Some authors [8, 9] have determined part of the curve experimentally for

different stress states. Their results clearly indicate work hardening with

a work-hardening coefficient that depends on the stress state.

The third part of the curve, starting when all transformable

tetragonal zirconia has been transformed into monoclinic, is determined by

the elastic properties of mainly monoclinic zirconia. Failure of the

material occurs, either by fracture, or by plastic deformation of the

monoclinic zirconia.

This study presents a different approach to the determination of the

slope of the second part of the curve. The stress-strain curve for a

biaxial stress state is determined from experimental data. The stress

profile and the amount of monoclinic zirconia are measured. Combining the

stress data and phase data gives a quantitative relation between the cause,

the transformation of tetragonal zirconia to monoclinic zirconia, and the

consequence, the residual stress which is equivalent to the flow stress.

The

four

dilatation, t , pp

associated with the complete transformation is about

percent. The flow stress and t pp

can thus be related. A flow law from

literature [7], relates flow stress to hydrostatic pressure, p, so tPP and

p can be related and an estimate of :S can be given. The resulting curves

Chapter 4. Residual stress and the stress-strain curve for Mg-PSZ 37

are compared with data given iii the literature [8, 9].

4.2 EXPERIMENTAL

Commercially available Mg-PSZ was delivered in tiles of about

lOOxlOOxlO mm~ Samples were sawn and ground to the required size. All

grinding was done as carefully as possible under the same conditions with

the D46 grinding wheel as described in Chapter 2. This grinding was

performed with water cooling. Polishing was done as described in Chapter 2.

The material characteristics were determined as described in Chapter

2. The fracture toughness of the material was measured with three-point

bend tests on twenty single notched samples and with the Double Cantilever

Beam (DCB) method on three samples. The strength of the Mg-PSZ was

determined from 36 samples of lx3x15 mm3 with ground surfaces.

Vickers indentations were made with a Leitz hardness tester at a load

of 20 N. The indentations and the surrounding areas were observed with

optical microscopy

and the radii of

using Interference Contrast.

the uplifted areas around

rumpling) were measured from photographs.

The

the

indentation

indentations

diagonal

(surface

The residual stress analysis was performed on bend-strip samples.

These strips were sawn and ground to dimensions of about 40xl0xl mm3• One

surface of 40x10 mm2 was polished until at least 30 Jim had been removed

from the surface. This side is assumed to represent the bulk material. The

strips were glued with the polished side on a 10 mm thick Alp3

base. The

surface opposite to the polished surface was ground until the strip had a

total thickness of somewhere between 0.15 and 0.4 mm. Some of these strips

were released from their substrate, after which they curved. Residual glue

and dirt were removed in boiling ethanol and/or in a ultrasonic bath with

38 Chapter 4. ResidUJJl stress and the stress-strain curve for Mg-PSZ

acetone. After this, the curvature of the strip was measured optically.

Other strips were used to determine the stress as a function of depth,

measured from the surface. The ground surfaces were polished until a

certain amount of surface had been removed (e.g. 5 Jim, 10 Jim, etc.). Then

these strips were also removed from their base, cleaned and measured.

The thickness of a strip was measured with the thickness gauge

described in Chapter 2 at regular positions on the strip. The phase content

of these bend-strip samples was determined from diffractograms and

calculated as described in Chapter 3.

4.3 RESULTS

The results are summarized in Table 4.1. The bulk modulus B was

calculated from:

E B = (1)

3(1-2v)

The values obtained are typical of Mg-PSZ [7, 10, 11].

Chapter 4. Residual stress and the stress-strain curve for Mg-PSZ

E

v

B

material properties:

elastic constants:

(GPa)

(-)

(GPa)

195.0 ± 3.0

0.324 ± 0.005

185 ± 8

· strength, fracture toughness and indentation dimensions:

u Klc DCB 3pb

2b d

depth interval (.urn)

0-2

2-10

10-22.5

(MPa) (MPa.m112

) 915 ± 56

10.3 ± 0.2 11.5 ± 1.1

(urn) 15.3 ± 2.7 (urn) 5

bend strip results:

calculated stress average (GPa)

3.2

0.25

0.1

39

depth interval (.urn) averaged monoclinic zirconia content (%)

0-2

2-2.5

2.5-6

6-8

8-10

92

47

29

22

20

Table 4.1: Results, 2b is the diameter of the area of surface rumpling, and

d is the indentation diagonal. The values given as X ± S stand for the

average X with the sample standord deviation S.

The critical transformation stress, u, c

was determined with two

different methods. First, according to the method given in [7], the

indentation diagonal, the Vickers hardness and the radius of surface

40 Clu:lpter 4. Residual stress and the stress-strain curve for Mg-PSZ

rumpling were used to determine a . c

Second, the strength and fracture

toughness were used to determine a according to the method given in [11]. c

Both methods gave a value for a of about 1.1 GPa. This value of the e

critical transformation stress depends on the stress state, but the

relationship is not clear. The value derived here applies to an

approximately hydrostatic stress state.

The residual stress is calculated from the curvature of the

bend-strips. There are slight variations in the formulas used [12-15].

Initially equation (2) from [14] was used, derived for situations in which

the strip is clamped to a base to inhibit bending when the residual stress

is developed. Afterwards, the strip is released from its base and allowed

to curve.

s E(t +d) 3

6rd t (1-v) ' (2)

In this equation, t is the thickness of the substrate, d is the thickness

of the residual stress layer, r is the radius of the strip and S is the

stress. This formula assumes a constant compressive stress along d and a

constant tensile stress along t. The arm of the moment causing the bending

of the strip is therefore taken as (t +d)/2 . The factor 1-v is added to

account for the biaxial stress state.

The value for d, about 22.5 pm, was determined from the fact that a

sample had no clear curvature after 25 pm had been removed by polishing,

and the fact that a sample was still curved after 20 pm was removed.

The stress profile for Mg-PSZ calculated from the measurements using

equation (2) gives a profile with a very high stress at the surface which

decreases very fast within the first two microns. However, a problem

occurred because applying formula (2) results in stress values that depend

on the thickness of the strip. A dependence of the calculated stress on the


ratio of d/t is also mentioned in [15]. This problem was solved by

adjusting equation (2) to equation (3) which takes approximately into

account the exponential shape of the stress profile:

_ E(t+dr S - 12rd((t I )+d)(l-v) (3)

The arm of the moment causing the bending is now taken as (t/2)+d. The

error caused by the assumption that the stress is concentrated precisely at

the surface instead · of about 1.5 Jim below the surface, is minor and

neglected. Applying (3) gives stress data which are not a function of the

thickness of the strip but show a normal statistical scatter around an

average. The results calculated with (3) are integrated values. The

stress-depth results were derived from these values and are shown in Table

4.1. The profile is illustrated in Fig. 4.2.

The determination of the amount of monoclinic zirconia in the ternary

system (cubic, tetragonal and monoclinic) has been performed as described

in Chapters 2 and 3. The formula used is given by:

f

2.3741(111) m (4)

2.3741(111) + 1(111) m e+t

where f is the volume fraction of monoclinic zirconia and I stands for the

area under the peak in question.

The values calculated with (4) are also the integrated values. The

effective penetration depth of Cu-K is about 10 Jim. No difference was a

measured in the amount of monoclinic zirconia, between a sample from which

30 Jim was removed from its surface through polishing, and a sample from

which 10 Jim was removed. The resulting values are given in Table 4.1 and

the profile is shown in Fig. 4.3, where V = 100f. m


.. a. (!)

or-------~========~~·r------------l

· 1

·3 -

·4L--------L--------~------~

0 10 20 30

depth; pm

100 .----------------------------,

-

80

60

40

20 ' ------ -- ----- ---------

o~-------L--------~------~

0 10 20 30

depth. f.rTl

Fig. 4.2 (top): Results of the residual stress analysis. The stress is

determined from bending in a plane-stress situation caused by the grinding

process.

Fig. 4.3 (bottom): Results from the phase analysis. The amount of

monoclinic zirconia, as derived from the measurements done on the

bend-strip samples is given as a function of depth.


4.4 DISCUSSION

The residual stress is directly related to the flow stress of the

transformation plasticity and referred to as a r" Comparing the stress and

phase profile as shown in Figs. 4.2 and 4.3, results after a correction for

the apparent amount of cubic zirconia and non-transformable tetragonal

zirconia (8%), in:

a = f

where at is in GPa.

3.5f, (5)

The dilatation associated with complete transformation is 4 %. This

means: f = 25e . Combining with equation (5) and adding a a of 1.1 GPa ~ c

gives the constitutive equation:

(6)

In this case, a a of 1.1 GPa can be used because the stress state c

considered is hydrostatic.

In literature [7], the flow law:

(7)

is proposed, where Y is the flow stress, Y is the initial flow stress, a 0

is a constant and p is the confining pressure. The predictions about the

radius of surface rumpling relative to the indentation size, made by the

model associated with equation (7), correspond to the observations on the

Mg-PSZ. The constant a was determined from the graphs given in [8] and the

properties of the Mg-PSZ. This resulted in a value for a of 2. Thus

relating equations (6) and (7) gives:

p = 44e (8) ~·

which describes the second part of the hydrostatic stress-strain curve.

After complete transformation, with a strain of 0.04, the material

will deform elastically. It is assumed that the bulk modulus of mainly

44 Chapter 4. Resid~ml stress and the stress-strain curve for Mg-PSZ

monoclinic zirconia is not significantly different from 185 GPa. The

complete curve resulting from this analysis is shown in Fig. 4.4a. This

figure shows the curve in the top-right quadrant, as it is normally given

in literature. In Fig. 4.4b. the position of the curve relative to the

stress-strain axes is given when the formal sign conventions are used.

5,--------------,

4

Oc___ _ _._ __ ..__ _ __._ __ _.__ _ __J

0 2 3 4 5

strain. %

tension

I I I I \

compression

4%

I I I I I I

', I , , I

I ', I

, I I

, , I

'"

Fig. 4.4a (left): The stress-strain curve as derived from the resid~ml

stress analysis and the phase analysis in the us~ml form. The elastic bulk

modulus in the third part, which is the modulus of almost pure monoclinic

zirconia , is also taken arbitrarily as about 185 GPa.

Fig. 4.4b (right): Using the formal sign conventions means that the

stress-strain curve in Fig. 4.4a would have to be positioned beneath the

strain axis.


4.5 CONSIDERATIONS

The result of a stress-strain curve showing a significant amount of

strain hardening seems to be in contradiction with the observed spontaneous

transformation in TEM samples. Phase analysis of HF-etched samples also

shows the spontaneous transformation of tetragonal zirconia into monoclinic

zirconia. The amount of monoclinic zirconia increases with increasing

etching time.

Both these observations of spontaneous transformation are made on

samples practically without mechanical constraints. A TEM sample is

extremely thin and HF etching removes material from the surface, leaving

unconstrained grains. The Mg-PSZ as examined in this study is a completely

dense material. Assume, for example, a sample which has 50 % monoclinic

zirconia at its surface. There will thus be a compressive stress of 1.8 GPa

at this surface according to equation (5), assuming equibiaxial stress. To

transform more zirconia, the critical transformation stress is required to

nucleate the transformation, and the compressive residual stress of 1.8 GPa

has to be overcome, resulting in a transformation stress of ac + 1.8 GPa.

This corresponds to significant work hardening.

The experimentally measured stress-strain curves in [9] and [10]

illustrate the dependence of B on the stress state. The curve given in [9]

is derived from bending in a DCB geometry. Extrapolating the second part of

the curve gives a stress of 3.6 GPa at a strain of 4 %. This corresponds

very well with the value of 3.5 GPa derived from the bend-strip analysis.

Both stresses are derived from bending caused by comparable stress states

(plane stress). From [10], a differential stress of 2.3 GPa is determined

at 4 % axial compressive strain at a confining pressure of 120 MPa. This

value is larger than the calculated stress from p=44e , which at 4 % pp


strain is 1.8 GPa. This hydrostatic stress state could give a lower limit

to the stress. These numbers illustrate the importance of a clear

definition of the stress state, which is also relevant to the value of the

critical transformation stress. The hydrostatic stress state could

represent one limiting situation, while the biaxial stress state could

represent another limiting situation.

The consistent results of this study justify the use of the phase

relation (4). The direct relation between residual stress and amount of

monoclinic zirconia means that a phase analysis gives enough information to

describe the residual stress. Chapter 5.1 describes the experiments and

results using this concept, that were performed to investigate whether

there are significant differences in the residual stress profiles due to

different grinding methods as well as additional concepts like strength and

fracture.

4.6 SUMMARIZING CONCLUSIONS

- The results from the X-ray analysis and the bend-strip measurements

indicate a clear relationship between the phase transformation and the

residual stress.

The present results lead to the stress-strain curve shown in Fig. 4.4,

which illustrates the importance of strain hardening for the deformation

of this material.

- The values of the critical transformation stress, a , the transformation c

bulk modulus, :B, and the work-hardening coefficient depend on the stress

state.

- The results of a phase analysis can be directly related to the residual

stress for the grinding process applied.

Chapter 4. ResidZUJI stress and the stress-strain curve for Mg-PSZ 47

References

1. A. G. Evans and A. H. J. Heuer, Review - transformation toughening in ceramics: martensitic transformations in crack-tip stress fields. J. Am. Ceram. Soc., 63 (1980) 241.

2. R. R. Lee and A. H. J. Heuer, In situ martensitic transformation in a ternary MgO-Y203-Zr02 alloy: I, transformation in tetragonal Zr02 grains. J. Am. Ceram. Soc., 71 (1988) 694.

3. R. Chaim and D. G. Brandon, Microstructure evolution and ordering in commercial Mg-PSZ. J. Mater. Sci. 19 (1984) 2934.

4. B. Budiansky, J. W. Hutchinson and J. G. Lambropoulos, Continuum theory of dilatant transformation toughening in ceramics. Int. J. Solids Structure 19 (1983) 337.

5. A. G. Evans and R. M. Cannon, Toughening of brittle solids by martensitic transformations. Acta Met. 34 (1986) 761.

6. A. H. Heuer and M. Rhle, On the nucleation of the martensitic transformation in zirconia. Acta Met. 33 (1985) 2101.

7. 1.-W. Chen, Implications of transformation plasticity in Zr02-containing ceramics: II, elastic-plastic indentations. J. Am.

· Ceram. Soc. 69 (1986) 181. 8. E. Ingels, A. H. Heuer and R. W. Steinbrech, Fracture mechanics of

high-toughness Magnesia-Partially- Stabilized-Zirconia. J. Am. Ceram. Soc. 73 (1990) 2023.

9. I.-W. Chen P. E. R. Morel, Transformation plasticity and transformation toughening in Mg-PSZ and Ce-TZP. In 'Mat Res. Soc. Symp. Proc. 78 (1987) 75, Advanced structural Ceramics', eds. P. F. Becker, M. V. Swain and S. Somiya.

10. M. V. Swain, R. C. Garvie and H. J . Hannink, Influence of thermal decomposition on the mechanical properties of magnesia-stabilized cubic zirconia. J. Am. Ceram. Soc. 66 (1983) 358.

11. M. V. Swain, Inelastic deformation of Mg-PSZ and its significance for strength-toughness relationship of zirconia toughened ceramics. Acta Met. 33 (1985) 2083.

12. R. Samuel and S. Chandrasekar, Effect of residual stresses on the fracture of ground ceramics. J. Am. Ceram. Soc. 72 (1989) 1960.

13. R. G. Treuting and W. T. Read Jr., A mechanical determination of biaxial residual stress in sheet metals. J. Appl. Phys. 22 (1951) 130.


15. C.-C. Chiu, Determination of the elastic modulus and residual stresses in ceramic coatings using a strain gage. J . Am. Ceram. Soc. 73 (1990) 1999.

5. RESIDUAL STRESS AND STRENGTH OF ZIRCONIA AFTER GRINDING

5.1 RESIDUAL STRESS AND STRENGTH OF MG-PSZ AFTER GRINDING

5.1.1 Introduction

Grinding is a reasonably reproducible surface treatment which is often

required and can therefore be chosen as the surface treatment to be varied.

A polished surface can be used as a reference, i.e. polishing is modelled

as the treatment which removes material from the surface without causing

the transformation or microfracture, although it is known that this is only

approximately true.

The influence of microfracture on toughness is described in literature

[1-5]. These relations are, however, based on toughening due to the

nucleation and propagation of microcracks caused by the growing crack.

Microcracks caused by a surface treatment like grinding are already present

in the material. The influence of this kind of microfracture on strength is

largely unknown. An experimental problem in this area is the difficulty of

characterizing microfracture at the surface and,

beneath the surface. Interesting considerations are

more important still,

presented in [6] of a

relation between microfracture and the result of strength measurements on

two sets of samples containing an increasing amount of stabilized zirconia

for one set and unstabilized zirconia for the other set.

The experiments described in this Chapter were carried out on Mg-PSZ,

Nilcra, to investigate the influence of grinding with two diamond wheels,

differing mainly in diamond-grain size, on phase content, residual stress,

strength and fracture behaviour of the material. In particular, the

dependence of strength on surface treatment is interesting for obvious

reasons. In literature [7-10], the relation between residual stress and

strength is predicted from a theoretical point of view. This chapter aims

Chapter 5. Resid1«1l stress and strength of zirconia after grinding 49

to give some experimental data on this subject and aims to present a model

explaining an experimentally obtained inverse relation between residual

stress and strength.

5.1.2 Experimental

The material used was delivered in tiles of 100x100x10 mm3• The main

characteristics of the material are given in Table 5.1.1 and Chapter 2. The

samples were roughly sawn to the approximate size.

Mg-PSZ

strength 3pb (polished), MPa

fracture toughness (MPa.m 112)

DCB SENB (3pb)

roughness, Ra (urn)

polished ground, A ground, B

761 ± 33 (10)

10.3 ± 0.2 (3) 11.5 ± 1.1 (20)

0.025 0.51 ± 0.08 (4) 0.69 ± 0.28 (5

Table 5.1.1: Characteristics of the Mg-PSZ. The numbers given as x ± S (n)

stand for the average x and the sample standard deviation S for n

measurements. The abbreviations SENB and DCB denote Single Edge Notched

Beam and Double Cantilever Beam respectively.

The characteristics of the chosen grinding methods are given in Table

5.1.2. The two methods are further denoted by A and B, A being the method

with relatively fine grains and B that with relatively coarse grains. All

other adjustable variables and procedures like wheel dressing and the

bronze bonding, were equal for A and B. The concentration is different, but

this cannot be avoided. This difference in concentration is assumed to be

insignificant, These grinding procedures were continued until at least 200

52 Chapter 5. Residual stress and strength of zirconia after grinding

interference from a surface-removing process . The resolution was high

enough to visualize grain boundaries.

5.1.3 Results

The results from the phase analysis were used to derive the phase

content profile. Three X-ray measurements were performed on surfaces that

were not polished afterwards. The value for A is 53.3 ± 2.2 % and for

method B 58.5 ± 1 %, where x ± y stands for the average x and the sample

standard deviation y for three measurements. These values illustrate the

slight but significant difference between the two grinding methods. The

measurements performed on samples from which the ground surface was removed

through polishing to obtain depth information, were single measurements.

The data of the phase analysis are integrated values. These values are

used to estimate the monoclinic zirconia-depth curve as described in

chapter 4. This calculation results in a shift of the original data as

shown in Fig. 5.1.1.

The calculated points between 0 and 4 J,lm for method B were higher than

100 %. Since this is physically impossible, there must have been some

experimental error in one of the measurements. The interdependence of the

calculated values is a consequence of the method used and it cannot be

excluded . The amount of monoclinic zirconia in the bulk material is

initially estimated as 0-15 %. This is illustrated in Fig. 5 .1.1 by the

shaded area. A more precise estimate based on some of the experimental

results will be given later. The results in Fig. 5.1.1 thus show a minimum

depth of 20 J.lm of the transformation zone for method A and 25 J.lm for method

B. These are minimum depths because the precise amount of transformation

caused by polishing is not known. All points, except the measurements at 4

J.lm, indicate more transformation at a certain depth for method B.

Chapter 5. Residual stress and strength of zirconia after grinding 53

The residual stress measurements resulted in a compressive surface

stress of 1.02 ± 0.09 GPa for method A and of 1.22 ± 0.10 GPa for method B

as determined from three measurements. The average residual stress

introduced by method B is thus higher than the stress introduced by method

A.

The results of the strength measurements for the first 20 pm are shown

in Fig. 5.1.2. This figure illustrates the slight but consistent difference

in strength after grinding by method A compared with B. Each point on the

curve is the average of ten measurements. The strength of samples ground by

method A is higher for every depth than that obtained with method B. The

steep decrease at 2-4 pm which is present in the phase-content-depth curve,

is absent.

In Figs. 5.1.3a and 5.1.3b the results of additional measurements done

to extend the strength-depth profile to 100 pm are added to the results of

the initial measurements. The experimental setup for these additional

measurements was such that it was not possible to thoroughly control the

amount of material polished off the surface. This explains the difference

in density of points shown in Figs. 5.1.3a and 5.1.3b. It is difficult to

estimate the depth at which the strength has decreased to the reference

strength, which is the average strength of 10 samples with polished

surfaces from which at least 100 pm has been removed. This strength is 761

MPa with a sample standard deviation of 33 MPa. This depth appears to be

higher for method B then for method A. The depths for both A and B appear

to be higher then the 20 pm for A and 25 pm for respectively, which were

estimated as minimal values from the phase analysis. The lack of accuracy

is unfortunately inherent to strength measurements on this material.


100~~-------------------------,

80

60

40 +-I=J+ ------$--·6

20 +-- 1 I

· +.-·~·~··

10 2 0 3 0 40 50

dept h. I-"'

-+ - ginding method A

- · 6 - · ginding method B

Fig. 5.1.1: The calculated amount of monoclinic zirconia as a function of

depth. The amount of monoclinic zirconia is given by V in percentage. The m

shaded area indicates the possible amount of monoclinic zirconia in the

bulk.

950

900

.. ~ ~ 850

~ "'

··-·~ - - · ll, -+- g inding

r!" ', +--- method A ' , , +

- · 6 -- ginding ' 6- - --- - - method B ---- -6

800

750 0 10 2 0 30

depth. I-"'

Fig. 5.1.2: The strength-depth curve for methods A and B for the first 20

IJm which illustrates the consistent difference in strength between the two

methods.

Chapter 5. Residuol stress and strength of zirconia after grinding 55

950

900 ~6

6 ca 850 t + one marker

~ one sarrple + +

t 6 one marker

800 -to!- + ten sarrples

"""++ ++

~ + reference

750 ------T-t---- ---------------- strength

+ + 700 +

650 0 20 40 60 80 100

depth. I-"'

950

900

850 ~6

+

~ one marker

6 one sarrple 6+

t + + 6 one marker 800

+ + ++ + ten sarrples ++

----------------·---+~------ reference "!j.+ -lj. 1/) 750 + +++ strength

+ +

700 + +

650 0 20 40 60 80 100

depth.!-""

Fig. S.1.3a (top): The strength-depth curve for method A extended to about

50 !Jm.

Fig. S.l.3b (bottom): The strength-depth curve for method B extended to

about 100 !Jm.

Examples of critical flaws are shown in Figs. 5.1.4a and 5.1.4b. The

size of these flaws, a maximum length of about 60 !Jm, is found to be


independent of the surface treatment of the sample. This value is

comparable to the maximum grain size in this material.

Scanning Acoustic Microscopy visualized only the grain boundaries and

showed no features of microfracture.

Figs. 5.1.4a and 5.1.4b: Examples of two fracture surfaces showing a

critical flaw of about 60 ~Jm.

Chapter 5. ResidUDI stress and strength of zirconia after grinding 57

5.1.4 Discussion

There are two extreme approaches in describing aspects of the

influence of grinding on a material. One is restricted to the grinding of

relatively ductile materials. The deformation of the material is described

mainly in terms of plastic deformation [14-17]. Metals are usually used to

illustrate principles of this approach. The other is restricted to the

grinding of brittle materials. In this case the deformation of the material

is mainly described in terms of brittle fracture, and ceramics or glasses

are used to illustrate the characteristics of this approach. The material

investigated, Mg-PSZ, is a ceramic with an unusually high toughness

compared to most other ceramics. Therefore a relatively large influence of

plastic deformation can be expected during the grinding of this material.

There is a difference in diamond grain size and concentration between

method A and B. The influence of the difference in concentration on the

investigated properties is assumed to be minor compared to the influence

caused by the difference in grain size. This difference in size is modelled

by a difference in grain shape. The smaller grain of A is assumed to be

sharper than the grain of B. The shape of the abrasive is often used as a

variable related to the forces working on a sample. A blunt shape results

in plastic deformation of materials at higher normal forces. This principle

will be used to explain the measured differences in material properties for

the two grinding methods.

The correspondence between residual stress and phase content was

indicated in Chapter 4. The residual stress profile has the same shape as

the corresponding phase profile. The absolute stress profile can be derived

from the average stress value and the phase profile. The exact amount of

monoclinic zirconia in the bulk material is not known because polishing


does cause the transformation of some tetragonal zirconia. The derived

residual stress profiles shown in Figs. 5.1.5a and 5.1.5b are thus

dependent on a correction for the amount of monoclinic zirconia in the

bulk. An amount of 5 - 10 % · monoclinic zirconia for the bulk can be

estimated both from Figs. 5.1.5a and 5.1.5b and from the result given in

Chapter 4 and [11], which shows a residual stress at the surface of 3.6

GPa. Grinding method B introduces a deeper residual stress profile and at

most depths a higher value for the residual stress. The main difference

between methods A and B is in diamond grain size. A larger grain thus

causes a deeper and higher transformation-zone in the material. This is

consistent with the general ideas found in literature [14-17] · of a higher

normal force during grinding with a larger grain.

The possibility of fracture originating from the edges of the samples

bas to be mentioned. Fracture originating from the edges could be

controlled by flaws introduced by the sawing of the samples. However,

observations of critical flaws on fracture surfaces gave no reason for the

suspicion of a significant influence of failure from the edges.

Chapter 5. Residual stress and strength of zirconia after grinding

"' f!l

j If)

~ -~

"' f!l lZ !!' iii

~ ·u; !!'

-5

-4

' I

-3 I I I

-2

-1 ~~~1

0 0

-5

--- ' -4

I I I I I

-3

-2 -1

A "' ·----------· 5 10 15 20 25 30

-1 6:---- ·t ~--1>

0 0 5

I

·------ --- -·---- ·• 10 15 20 25 30

depth. f-IT'

-I>- method A. 0 % corr.

-·•-· method A. 15 o/o corr.

-I>- method B. 0 % corr.

-·•- . method B. 15 % corr.

59

Fig. S.l.Sa (top): The derived residual stress profile as a function of

various corrections for the amount of monoclinic zirconia in the bulk for

method A.

Fig. S.l.Sb bottom): The derived residual stress profile as a function of

various corrections for the amount of monoclinic zirconia in the bulk for

method B.

60 CluJpter 5. Residual stress and strength of zirconia after grinding

N'o

monocl.l --i I + I 'vcY\/\1\P/\/\/"~V\.f\../'ov

I I I I I ----o 0 0 o-

tetr. 1 practicallr no sfess 1 ----o 0 0 o-

1 I I I

2 gra1n boundary

~o~o------- ~o/

~l~L :::--L - _l_ __ _ L \

I ~ I -0\-~ tetr.

--9- -- --?- - --9 --

0-

1 --o-1

Fig. 5.1.6: Schematic model of a transformed surface layer and the

tetragonal bulk material. The plastic strain corresponding to the

transformation results in elastic compressive stresses in the transformed

part of the material. The compensating tensile stresses are spread out over

the bulk material. The compressive stresses can be relaxed near grain

boundaries as shown in 5.1.6.2. This will cause relatively high localized

tensile stresses in the tetragonal structure which can promote fracture at

these locations. Further explanation is given in the text.


The interpretation of the strength data presented in Figs. 5.1.2,

5 .1. 3a and 5 .1. 3b is less clear. The strength after grinding by method B is

less than the strength after grinding by method A down to a depth of about

25 !Jm, despite the higher residual stress and thicker transformation zone

for method B. This means that the direct influence of the compressive

residual stress is not the only phenomenon influencing strength of this

material. The apparent discrepancy can be explained with the model

illustrated in Fig. 5.1.6. Shear stresses, twinning, preferred orientations

and other possible phenomena are considered irrelevant to the qualitative

aspects of the principles which are used. In this model, grinding

introduces the transformation in a surface layer of the material and thus

compressive stresses in this layer. This residual stress is developed

because the unit cell in the monoclinic structure is larger than the unit

cell in the tetragonal structure as schematically illustrated by the

spring-like bindings in Fig. 5.1.6. A continuum which is transformed at the

surface will contain elastic compressive stresses. The compensating tensile

stresses are averaged to the bulk of the material. The main inhomogeneities

in Mg-PSZ in this simplified model are pores and grain boundaries. These

inhomogeneities are locations where the compressive stresses can be

somewhat relaxed by strain of the bindings. Beneath these areas of

relaxation, the tetragonal structure will be stretched to an elastic

tensile stress as illustrated in Fig. 5.1.6.2 and such tensile stresses

will promote fracture of the material at these locations. Grain boundaries

especially are thus seen as areas where the likelihood of fracture is

increased.

Grinding according to the defined procedures results in more residual

stress in samples ground according to method B and higher strength of


samples ground according to method A. The lower strength of samples ground

by B can be explained on the assumption that greater compressive stresses

give rise to more areas of tensile stresses and also higher tensile

stresses at these locations. These higher tensile stresses will cause

larger microcracks beneath the compressive stress layer. These ideas are in

agreement with the results presented in [18]. In this article the influence

of Young's modulus of a thin coating on the stress state due to a ball

indentor at the interface between the coating and substrate is calculated.

It is shown in [18] that a stiffer coating, resulting in less strain under

a ball indentor, gives smaller tensile stresses at the interface. A stiffer

coating in [18], causing less strain, relates to less compressive residual

stress in the present study. A smaller compressive residual stress results

thus in smaller stress relaxations, which are equivalent to smaller

strains, smaller tensile stresses beneath the compressive layer and thus a

higher strength.


100

l

l

1.0

~ 0.8

·x 0.6

~ 0.4

~ 0.2

B 0.0

·~ -0.2

"' ~ -0.4

~ -0.6

~ -0.8

1

}.5

~

1

1J~

10 -I

:100

: 10 ~ I I I

~X I I I

0

"""\ •' . '\ .

'\' \\ I'· .·· I I \,

-1.0 L_ _ __. __ ~--~-~~----'>....J

500 -

-- no notch, S 1

no notch. S2

notch, S1

notch. S2

75 80 85 90 95 100

x. relative t.r1it

63

Fig. 5.1. 7 (top): Geometry used for the Finite Element Analysis. All

dimensions are indicated by a relative number. The vertical line with the x

is the line along which stresses are plotted in Fig. 5.1.8. See text for

further information.

Fig. 5.1.8 (bottom): The two principal stresses S1 and S2 along x from 75

to 100 both for a situation with a notch and without a notch. See text for

further explanation.

64 Chapter 5. Residual stress and strength pf zirconia after grinding

To obtain a better indication about the stress state beneath a

compressive stress layer near inhomogeneities, a Finite Element Analysis

was performed. The analysis used 300 quadrilateral, 8-node, plane stress

elements. The geometry used is illustrated in Fig. 5.1.7. The dimensions

are such that the compressive layer is far thinner than the total

thickness, and that the width of the model is large enough to be without

influence. The bottom of the model contains also a layer of compressive

stresses to avoid curvature of the whole model, without using an extremely

large thickness. Stresses were determined along a vertical line for a

situation with and without a notch and they were quantified relative to the

maximum compressive stress. The notch represents a possible location for

compressive stress relaxation, as discussed before. Such a stress

relaxation, through a notch or shear or other phenomena, is quite realistic

due to a process like grinding. The differences between shear-strain as

proposed in Fig. 5 .1.6.2 and compressive-strain as in this model are not

relevant to the principle. The main idea is that there is a possibility of

strain, relaxing the compressive stresses, and that larger compressive

stresses result in larger strains. The values for the two principal

stresses in situations with and without a notch are presented in Fig.

5.1.8. It is clearly shown in Fig. 5.1.8 that the situation with a notch

results in a maximum tensile stress at the interface between the

compressive stress and the bulk of the material, of about one third of the

maximum compressive stress. This means a tensile stress of more than 1 GPa

for the Mg-PSZ. The precise magnitude of this stress is not meant to be

representative for the modelled situation. It depends on shape of the

notch, ratio of thicknesses , etc. The modelling was done to illustrate that

a notch in a residual stress layer gives a concentration of tensile


stresses beneath this residual" stress layer. The calculation does indicate

that during a fracture test, fracture is likely to originate from this

tensile stressed area.

The presence of residual stress results in a strength higher then the

strength in a situation without residual stress. And it is seen that a

larger residual stress results in a lower strength compared to the strength

in a situation with less residual stress. This concept can be illustrated

with the principle of a threshold. Starting at the point of no residual

stress and no strengthening, the strength is increased with increasing

residual stress. After a maximum in the strengthening for a certain

residual stress, the strengthening decreases with increasing residual

stress. This is observed for the two grinding wheels A and B.

Additional arguments for the importance of the stress state beneath

the compressive residual stress layer can be given. The residual stress of

about 3.6 GPa at the surface is significantly more than the failure stress

in a three-point-bend · test of less than 1 GPa. The absence of a steep

decrease in strength at a depth of 2-4 J.lm, which is observed in the

phase-content-depth curve is also an indication of the importance of the

subsurface stress state. It is known from observations that fracture of

this material is usually intergranular. These considerations are consistent

with the concept of areas near grain boundaries with tensile stresses

beneath the residual stress layer as one of the strength-determining

factors. The bulk material contains flaws varying in shape and size, but

the maximum size is the critical flaw size of about 60-70 J.lm. This size is

comparable to the grain size of the material. A logical conclusion is thus

that the strength of the grain boundaries in the material, combined with

the dual influence of the residual stress, determine the strength of the

Mg-PSZ.


5.1.5 Conclusions

Grinding with coarse diamond grains results in a deeper residual stress

layer and a higher residual stress at a certain depth but a lower

strength.

The residual stress is modelled to cause localized tensile stresses

beneath the compressive stress layer near inhomogeneities like grain

boundaries and pores.

- Higher residual compressive stresses at the surface are assumed to result

in higher tensile stresses near grain boundaries beneath this compressive

stress layer and this will encourage the development of a· crack along

grain boundaries.

5.2 RESIDUAL STRESS AND STRENGTH OF Y-TZP AFTER GRINDING

5.2.1 Introduction

The differences between Mg-PSZ and Y-TZP are of a microstructural,

chemical and mechanical nature. Degradation and superplasticity, for

instance, are relevant to Y-TZP but not to Mg-PSZ. The microstructure of

both materials is entirely different and there are major differences in

wear resistance, as will be discussed in Chapter 7.2. Another aspect, the

re-transformation as discussed in Chapter 3 .1, can be relevant to Y-TZP.

This chapter deals with the influence of grinding on the residual stress

and strength of Y-TZP (Feldmiihle), using the same procedures as used for

Mg-PSZ. The results of the experiments can be compared to the results

presented in Chapter 5 .1 .

Chapter 5. ResidZUJI stress and strength of zirconia after grinding 67

5.2.2 Experimental

The characteristics of the Y-TZP are given in Chapter 2. Exactly the

same experimental procedures as described in Chapter 5 .I were used for the

Y-TZP samples, apart from the use of an extra grinding wheel with diamond

grains of about 160-179 !Jm (D180). The three different grinding wheels are

now indicated as flne, (D64), intermediate (D107) and coarse (D180).

Polishing was again used as the method to obtain samples that modelled the

material without residual stress, although it is known that this is an

approximation. The measurements as described in Chapter 5.1, of properties

like strength, phase content and residual stress, were done according to

the same procedures. The residual stress analysis was only performed on the

samples ground with the coarse grinding wheel. X-ray diffraction was used

to examine the possibility of the re-transformation.

5.2.3 Results

The results are summarized in Table 5.2.1 and Figs. 5.2.1-5.2.4. The

original data of the phase analysis are presented in Fig. 5.2.1. The amount

of monoclinic zirconia after grinding with a fine grain is less than after

grinding with the other two grains. The largest amount of monoclinic

zirconia is found after grinding with the coarse grain, but there is not

much difference in the total amount between the intermediate and the coarse

grain. The results for the coarse grain show a single maximum at a depth of

about 2 !Jm. The presence of a maximum is not clear for the materials ground

with the fine and intermediate grain size.

The residual stress for the samples ground with the coarse grain, was

calculated with the formula presented in Chapter 4 using a value for d, the

depth of the residual stress layer, of 12 !Jm. The curve has the same shape

as the results of the phase analysis with a maximum at 2 !Jm as shown in

68 Chopter 5 . Residual stress and strength of zirconia after grinding

Fig. 5.2.2. The original values of the phase content and the residual

stress are averages over a certain depth and an approximation of the

profile can be calculated as discussed in Chapter 4. The results of these

calculations are shown in Figs. 5.2.3a and 5 .2.3b. Both figures show a

sharp subsurface maximum and both the amount of monoclinic zirconia and the

residual stress at the surface are practically zero.

surface fune intermediate coarse removed, Jlm

0 1 2 4 6 10

0 1 2 4 6 10

the amount of monoclinic zirconia, %

8.95 6.65 8.02 1.49 1.79 0.00

10.07 8.87 10.00 9.08 4.55 1.99

strength, GPa

1.38 ±0.12 (20) 1.35 ±0.10 (20) 1.38±0.13 (10) 1.32±0.16 (10) 1.35±0.19 (10) 1.42±0.10 (10) 1.34±0.12 (10) 1.28±0.11 (10) 1.31 ±0.15 (10) 1.38±0.14 (10) 1.42±0.14 (20) 1.45±0.14 (20)

7.36 11.35 12.98 5.09 6.12 1.29

1.40±0.16 (20) 1.41 ±0.06 (10) 1.35±0.06 (10) 1.28±0.12 (10) 1.30±0.17 (9) 1.42±0.10 (20)

the strength after polishing is 1.13 ± 0.15 GPa (23)

residual stress, GPa

1.23 (4) 1.34 (2) 1.46 (2) 1.29 (2) 0.53 (2) 0.51 (4)

Table 5.2.1: The amount of monoclinic zirconia and the strength as a

function of depth for fine, intermediate and coarse grinding. The residual

stress was only determined after coarse grinding. Results given as x ± y

(n) stand for mean x ± the sample starukJrd deviation y for n measurements.

The results of the strength measurements are shown in Fig. 5.2.4. The


scatter in the results is unfortunately too large to derive anything from

the data.

The results of the X-ray analysis, which was performed to examine the

possible occurrence of the re-transformation, are presented in Fig. 5.2.5.

The intensities of the peaks as mentioned in Chapter 3 were measured and

the peak ratio's were calculated. The data were normalized to the peak

ratio at 10 J.lm below the surface. As shown in Fig. 5.2.5, there is a

continuous decrease in ratio towards the surface.

20 r-------,

16

~ -- fine 12 •'

~ tl I I -a- intermediate

E I

> 8

--o-- coarse

depth. 1-nl

Fig. 5.2.1: The original data for the amount of monoclinic zirconia after

fine, intermediate and coarse grinding.

70 Chapter 5. Residl«ll stress and strength of zirconia after grinding

- 1.50

"' (!,

~ Q)

~ -1.00

~ 'iii ~

-0.50

0.00 .__ _____ ___. ______ __J

0 10 20

Fig. 5.2.2: The original data for the residl«ll stress after coarse

grindi(lg.

50 -5

40 -4

t-

& 30

~ -3

* ~ £: ;; >

~ 20 - 2 iii 1?

10 -1

__r---t---------~ olr

-------- --+ 0

0 10 20 30 0 10 20 30

depth. fill del>th.fP>

Fig. 5.2.3a (left): The derived profile for the amount of monoclinic

zirconia after coarse grinding.

Fig. 5.2.3b (right): The derived profile for the residl«ll stress after

coarse grinding.


2.00 r---------------,

1.80

I'll

~ 1.60

L

f 1.40 f. 0 ~ "' o•+

•* 1.20

-------------------------------+

1.00 '------'---~--~-~----' 0 10 20 30 40 50

+ tine

• intermediate

o coarse

71

Fig. 5.2.4: The strength profile after fine, intermediate and coarse

grinding

1.00 ......--------...------,

0 .80

0 ·~

"'0.60

~ --+-- (200)/(002lt

~ ---6- (131)/(113)! N

~ 0.40

~

020

0.00 '-------'--------' 0 10 20

Fig. 5.2.5: The change in peak-intensity ratio's of (200)/(002)1

and

(131) 1(113) normalized to the peak ratio measured at 10 J.lm below the I I

surface. The principles concerned are explained in Chapter 3.1.

72 Chapter 5. ResidUJJI stress and strength of zirconia after grinding

5.2.4 Discussion

The smaller amount of monoclinic zirconia after grinding with a fine

grain is as was expected, Chapter 5 .1. A fine grain results in lower forces

on the material than a coarse grain. The integration of the data for the

fme and intermediate grain from Fig. 5.2.1 does not give reliable results

due to the the presence of the dip at a depth of 1 JJm. Smoothing the data

would be possible but this would not provide reliable information and has

therefore been omitted. The general shape of the three curves is, however,

similar.

The subsurface maximum in Figs. 5.2.3a and 5.2.3b ·is the main

phenomenon extracted from the results . A relaxation of stresses at the

surface is not hard to explain. The grain size of Y-TZP is about 1 JJm and

grain boundary sliding for instance could relax stresses at the surface.

The amount of monoclinic zirconia is also minimized at the surface. A

possible explanation can be given by the re-transformation. The

re-transformation has been reported for TZP's and it would explain both the

absence of monoclinic zirconia and of residual stress at the surface. The

results from the X-ray analysis do show a continuous change in orientation

of the tetragonal phase which could be an indication of this

re-transformation. However, as explained in Chapter 3, the influence of

domain switching on the peak ratio's cannot be distinguished from the

influence of the re-transformation.

The results of the strength measurements show a large scatter. This

means that grinding with different grinding wheels does not result in

significant differences in strength. It should be mentioned that no

information is obtained about possible internal defects .

The differences with Mg-PSZ can be illustrated with the differences in


microstructure. Grinding a material containing a maximum grain size of 60

1-1m with a feed of 10 1-1m is different from grinding a material containing a

maximum grain size of about 1 1-1m at the same feed . In the case of Mg-PSZ

only part of a grain is removed. In the case of Y-TZP about 8 or 9 grains

are removed during passing of the grinding wheel. The residual stress layer

for Mg-PSZ is about 25 1-1m thick, and about 10 1-1m for Y-TZP. Mg-PSZ is far

more of a continuum during the chosen grinding procedures then Y-TZP.

A second difference between the results on Mg-PSZ and Y-TZP, is the

difference in the results of the strength measurements. The measurements on

Mg-PSZ showed a clear dependence on the type of grinding, while Fig. 5.2.4

shows no relation at all. The strength data on Mg-PSZ are related to the

residual stress, Chapter 5.1 , and the correlation between strength and

residual stress is clear. It is assumed that the amount of monoclinic

zirconia is equivalent to the amount of residual stress although the only

support for this argument is given by the similarity between the graphs in

Figs. 5.2.3a and 5.2.3b. The scatter in Fig. 5.2.4 combined with the

concept that grinding with a fine grain results in less monoclinic

zirconia, Fig. 5 .2.1, can be seen as an indication that differences in

residual stress are not significant to differences in strength. The

strength of Y-TZP is, however, significantly increased by grinding as

compared to the reference strength. This implies that residual stress does

strengthen the material, but the exact shape of the residual stress profile

is not important to the strength level.

A third difference with Mg-PSZ is the quantitative relation between

residual stress and the amount of monoclinic zirconia. About 100 %

monoclinic zirconia in Mg-PSZ results in about 3.6 GPa residual stress.

This value of 3.6 GPa is consistent with data from literature as shown in

Chapter 4. For Y-TZP a value of 3.6 GPa is obtained at about 40 %


monoclinic zirconia as shown in Figs. 5 .2.3a and 5.2.3b. This means that

the coefficient of transformation plasticity for Y-TZP could be about twice

as large as for Mg-PSZ. The value of 40 % is, however, not very reliable.

These differences imply also that the fracture behaviour of Y-TZP is

different from that of Mg-PSZ. The variety of additional phenomena assumed

to occur in Y-TZP, like the re-transformation, create more problems in

modelling the behaviour of the material. More research is thus necessary to

identify the role of individual phenomena.


- The residual stress profiles for Mg-PSZ and Y-TZP are different. The

maximum stress for Mg-PSZ is located at the surface and the profile is 20

to 25 J.lm thick, while for Y-TZP the maximum in stress is located beneath

the surface and the thickness is about 10 J.lm. The maximum amount of

stress for Y-TZP and Mg-PSZ is about equal.

Grinding with a fine diamond grain results in Y-TZP in less monoclinic

zirconia then grinding with an intermediate or coarse grain.

- There is no significant difference in strength between fine, intermediate

and coarse grinding of Y-TZP.

References

1. W. Pompe, H.-A., Bahr, G. , Gille and W. Kreher, Increased fracture toughness of brittle materials by microcracking in an energy dissipative zone at the crack tip. J. Mater. Sci. Letters 13 (1978) 2720.

2. F. Guiu and R. N. Stevens, Physical interpretation of fracture-toughening mechanisms. J. Mater. Sci. 26 (1991) 4375.

3. W. Kreher and W. Pompe, Increased fracture toughness of ceramics by energy-dissipative mechanisms. J. Mater. Sci. 16 (1981) 694.

4. Y. Fu and A. G. Evans, Some effects of microcracks on the mechanical properties of brittle solids-11. microcrack toughening. Acta Metal!. 33 (1985) 1525.

5. A. G. Evans and K. T. Faber, Crack-growth resistance of microcracking brittle materials. J . Am. Ceram. Soc. 67 (1984) 255.

6. T. Ono, K. Nagata, M. Hashiba, E. Miura, Y. Nurishi and T. Shimida, Internal friction, crack length of fracture origin and fracture surface energy in alumina-zirconia composites. J. Mater. Sci. 24


(1989) 1974. 7. D. J. Green, Compressive surface strengthening of brittle materials.

J. Mater. Sci. 19 (1984) 2165. 8. D. J. Green, Compressive surface strengthening of brittle materials by

a residual stress distribution. J. Am. Ceram. Soc. 66 (1983) 807. 9. Y. W. Mai On 'the effect of residual stresses in quasi-static cracking

of materials. Int. J. Mech. Sci. 19 (1977) 325. 10. M. Bakioglu, F. Erdogan and D. P. H. Hasselman, Fracture mechanical

analysis of self-fatigue in surface compression strengthened glass plates. J. Mater. Sci. 11 (1976) 1826.

11. P. H. J. van den Berg and G. de With, Residual stress and the stress-strain relation for Mg-PSZ. J. Europ. Ceram. Soc. 9 (1992) 265.

12. R. G. Treuting and W. T. Read Jr., Mechanical determination of biaxial residual stress in sheet metals. J. Appl. Phys. 22 (1951) 130.


14. A. A. Torrance, An approximate model of abrasive cutting, Wear, 118 (1987) 217.

15. M. Abebe and F. C. Appl, Theoretical analysis of the basic mechanics of abrasive processes part II: Model of the ploughing process, Wear, 126 (1988) 267.

16. P. Gilormini and E. Felder, Theoretical and experimental study of the ploughing of a rigid-plastic semi-infinite body by a rigid pyramidal indenter, Wear, 88 (1983) 195.

17. A. A. Torrance, The correlation of process parameters in grinding, Wear, 139 (1990) 383.

18. S. van der Zwaag and J. E. Field, The effect of thin hard coatings on the Hertzian stress field. Phil. Mag. 46 (1982) 133.

6. TRIBOLOGY AND CERAMICS

The study of friction, wear and lubrication is an extensive one. Many

aspects from a variety of disciplines have to be considered. Surfaces

interacting have practically always been important, and basic concepts like

adhesion from 1953 [1, 2], are still used today. Tribology is a field that

has dealt largely with metals. Most engines and other designs with moving,

interacting parts, were usually constructed with metals. New materials,

plastics, new alloys and ceramics, resulted sometimes in better

tribological performances, but also in more variables and fundamental

problems to be solved.

Presenting experimental data on wear of ceramics means first of all

that data obtained from different laboratories should be comparable.

Therefore, multilaboratory tests were done [3, 4], and these show which

variables are important to obtain results that can be compared with the

results of other authors.

A wear system can be described in many ways, on a variety of scales

[5], and many features can be examined, like vibrations, friction noises,

friction forces, etc. [6-10]. Temperature is an important variable as well,

and testing instruments are developed to examine wear at high temperatures

[11]. Some of the literature summarizes the concepts encountered so far

and presents a view for the future [12].

Research on wear is often closely connected to an application [13-19].

This is logical and in most cases practically the best. The number of

choices of variables for a wear test is huge, and some kind of limitation

provided by the application is useful.

6.1 WEAR MODELS

A variety of models has been developed to explain the various

phenomena encountered during wear. These models are usually based on the

Chapter 6. Tribology and ceramics 77

dominance of one wear mechanism like adhesion, delamination or abrasion.

Adhesion [20-22], is often important. Pieces of one material remain

attached to the counter material which gives a contact between the same

materials . Sliding between the same materials is usually a situation that

gives a relatively large amount of wear. Chemical aspects, like oxidation

[23], are also likely to influence the wear-process. The local temperature

during sliding, the flash temperature, can be much higher then the

temperature of the bulk. It is, however, difficult to obtain information

about this flab temperature. Phase changes at the surface of materials, or

sliding with sapphire can give information [24], but this information is

restricted to specific materials. Theoretical analysis is another method to

derive values for the flash temperature [25, 26].

Abrasion is a term often used to indicate a variety of processes that

involves a relatively hard material, that scratches and cracks a relatively

soft material. Fracture during wear [27-30], is quite relevant for ceramics

due to the brittle nature of these materials. One specific type of fracture

is delamination [31-34]. This is the loss of material in sheets that

originated from lateral cracks running parallel to the contact surface. The

development of lateral cracks is consistent with the stress state that has

been modelled for sliding materials.

The relative importance of one mechanism depends on conditions like

temperature, load, velocity, etc. The transitions between the occurrence of

the various mechanisms is also often a transition between regions of

various amounts of wear [35, 36]. Such a boundary between two regions

depends thus on the conditions. The relation between these boundaries and

the conditions can be presented in a wear map [37-41]. Such a wear-map

gives an overview of the amounts of wear under several conditions.

78 Chapter 6. Tribology and ceramics

6.2 WEAR OF CERAMICS

There has been a lot of interest in the wear behaviour of ceramics

[42-66]. This interest is usually focused on the four common structural

ceramics, coatings not included, alumina, siliconnitride, siliconcarbide

and zirconia. Most of the aspects related to wear like the effect of

elevated temperature, the influence of the atmosphere, lubrication and the

importance of wear debris, have been examined. But there still is the

common problem in tribology, that the presentation of laws and rules making

the right predictions, is hazardous. Part of the literature on wear of

ceramics is thus based on experiments and part of it is based on

theoretical models.

6.3 WEAR OF ZIRCONIA

Zirconia ceramics are as mentioned, interesting as wear resistant

materials [67-82]. The chemical, mechanical and thermal properties of these

materials are such that research after their tribological properties both

within the fundamental field as well as within the field of applications,

is justified. Zirconia ceramics are, for instance, investigated in systems

of zirconia-metal combinations at various temperatures, in self-mated

systems, and often in unlubricated sliding.

References

1. R. F. King and D. Tabor, The strength properties and frictional behaviour of brittle solids. Proc. Roy. Soc. A 223 (1953) 225.

2. J. F. Archard , Contact and rubbing of flat surfaces. J . Appl. Physics 24 (1953) 981.

3. E. A. Almond and M. G. Gee, Results from a U. K. interlaboratory project on dry sliding wear . Wear 120 (1987) 101.

4. H. Czichos, S. Becker and J. Lexow, International multilaboratory sliding wear tests with ceramics and steel. Wear 135 (1989) 171.

5. P. J. Blau, Scale effects in steady-state friction . Trib. Trans. 34 (1991) 335.

6. H. L. Price and V. M. Desai, A statistical analysis of forces at sliding interface. Trib. Trans. 33 (1990) 149.


7.

8.

9. 10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

M. Nakai and M. Yokoi,- Generation mechanism of friction noises in dry friction. Jap. J. Trib. 35 (1990) 513. H. Shimura and T. Kitahara, Effects of vibration characteristics of measuring system on friction and wear. Jap. J. Trib. 35 (1990) 577. N. P. Sub and H.-C. Sin, The genesis of friction, Wear 69 (1981) 91. M. G. Gee and E. A. Almond, Effects from vibrations in wear testing of ceramics. Mat. Sci. and Techn. 4 (1988) 655. M. G. Gee and C. S. Matharu, The measurement of sliding friction and wear at high temperature. Int. J. High Techn. Ceram. 4 (1988) 319. D. C. Cranmer, Ceramic tribology-needs and opportunities. Trib. Trans. 31 (1987) 164. A. N. Grearson, J. Aucote and H. Engstrm, Wear of ceramics in grit blasting. Br. Ceram. Trans. J. 88 (1989) 213 . K. F. Dufrane, Wear performance of ceramics in ring/cylinder applications. J. Am. Ceram. Soc. 72 (1989) 691. R. A. Mayville, Abrasive concentration effects on wear under reciprocating conditions. Wear 140 (1990) 359. M. Fripan, U. Dworak and D. Fingerle, Friction and wear of ceramic sliding and sealing elements. cfi/Ber. DKG 617 (1987) 239. G. Schuseil, Ceramics in engine manufacture. cfi!Ber. DKG 617 (1987) 242. S. Aiyoshizawa, A. Wakigawa, D. Konno and K. Takagi, A study of ceramic bearin~ for vertical pumps (friction coefficients of submersible beanngs in air and water). JSME Int. J. Series II, Vol. 33 (1990) 41. J. G. Baldoni, S. F. Wayne and S. T. Buljan, Cutting tool materials: mechanical properties wear resistance relationships. ASLE Trans. 29 (1986) 347. 0. 0. Ajayi and K. C. Ludema, Mechanism of transfer film formation during repeat pass sliding of ceramic materials. Wear 140 (1990) 191. K. Hiratsuka, H. Saito and T. Sasada, Oxygen chemisorption onto metal surfaces during adhesive wear. Jap. J. Trib. 35 (1990) 817. K. Miyoshi, Fundamental considerations in adhesion, friction and wear for ceramic-metal contacts. Wear 141 (1990) 35. A. Enomoto, K.. Hiratsuka and T. Sasada, Effect of oxidation activity of metal and of atmospheric oxygen on friction and wear between metal and alumina. Jap. J. Trib. 36 (1991) 61. J . A. Griffioen, S. Blair and W. 0. Winer, Infrared surface temperature measurementsin a sliding ceramic-ceramic contact. Proc. 12th Leeds-Lyon symp. Trib., D. Dowson ed. (1985) 238. M. F. Ashby, J. Abulawi amd H. S. Kong, Temperature maps for frictional heating in dry sliding. Trib. Trans. 34 (1991) 577. B. Gecim, Transient hot spot temperatures at a sliding ceramic contact including surface coating effects. Wear 123 (1988) 59. S. Wada, Micro-fracture of ceramics and tribology. Jap. J. Trib. 35 (1990) 859. Y. Kimura and M. Shima, Longitudinal contact-point model for calculating stress-intensity factors at surface cracks in sliding wear. Wear 141 (1991) 335. A. G. Evans and D. B. Marshall, Wear mechanisms in ceramics, in Fundamentals of friction and wear of materials 1980, 439, ed. D. A. Rigney A. R. Rosenfield , Fracture of brittle materials under a simulated wear stress system. J. Am. Ceram. Soc. 72 (1989) 2117. N. P. Suh, An overview of the delamination theory of wear. Wear 44 (1977) 1. S. Jahanmir and N. P. Sub, Mechanics of subsurface void nucleation in delamination wear. Wear 44 (1977) 17. J. R. Fleming and N. P. Sub, Mechanics of crack propagation in delamination wear. Wear 44 (1977) 39.


34. J. R. Fleming and N. P. Suh, The relationship between crack propagation rate and wear rates . Wear 44 (1977) 61.

35 . A. W. J. de Gee, A. Begelinger and G. Salomon, Failure mechanisms in sliding lubricated concentrated contacts. Proc. 11th Leeds , 4-7 sept. 1984: Butterworth 1985 , 105.

36. D. Landheer and A. W. J. de Gee, Adhesion, friction and wear. MRS Bull. 16 (1991) 36.

37. S. C. Lim and M. F. Ashby, Overview no. 55 wear-mechanism maps. Acta metall. 35 (1987) I.

38. S. C. Lim, M. F. Ashby and J. H. Brunton, Wear-rate transitions and their relationship to wear mechanisms. Acta metall. 35 (1987) 1343.

39. M. Woydt , D. Klaffke, K.-H. Habig and H. Czichos, Tribological transition phenomena of ceramic materials. Wear 136 (1990) 373.

40. H. Kitsunai, K. Kato, K. Hokkirigawa and H. Inoue, The transitions between microscopic wear modes during repeated sliding fiction observed by a scanning electron microscope tribosystem. Wear 135 (1990) 237.

41. S. M. Hsu, Y. S. Wang and R. G. Munro, Quantitative wear maps as a visualization of wear mechanism transitions in ceramic materials. Wear 134 (1989) 1.

42. M. Woydt and K.-H. Habig, High temperature tribology of ceramics. Trib. Int. 22 (1989) 75 .

43. D. Klaffke, Fretting wear of ceramics. Trib . Int. 22 (1989) 89. 44. P. Studt, Boundary lubrication : adsorption of oil additives on steel

and ceramic surfaces and its influence on friction and wear. Trib. Int. 22 (1989) 111.

45. T. E. Fischer and H. Tomizawa, Interaction of tribochemistry and microfracture in the friction and wear of silicon nitride. Wear 105 (1985) 29.

46. Z. Wang and C. Ruiz, Effect of wear damage on the strength of hot isostatically pressed silicon nitride. J . Am. Ceram. Soc. 72 (1989) 1076.

47. J. Denape and J. Lamon, Sliding friction of ceramics: mechanical action of the wear debris . J. Mater . Sci. 25 (1990) 3592.

48. S. Jahanmir and T. E. Fischer, Friction and wear of silicon nitride lubricated by humid air, water, hexadecane and hexadecane + 0.5 percent stearic acid. STLE Trans. 31 (1987) 32.

49. M. G. Gee, C. S. Matharu, E. A. Almond and T. S. Eyre, The measurement of sliding friction and wear of ceramics at high temperature. Wear 138 (1990) 169.

50. S. Sasaki, The effects of the surrounding atmosphere on the friction and wear of alumina, zirconia , silicon carbide and silicon nitride. Wear 134 (1989) 185.

51. K. Kato, Tribology of ceramics. Wear 136 (1990) 117. 52. Y. Nakamura and S. Hirayama, Wear tests of grey cast iron against

ceramics. Wear 132 (1989) 337. 53. M. Akazawa and K. Kato, Wear properties of SbN4 in rolling-sliding

contact. Wear 124 (1988) 123. 54. H. Ishigaki , R. Nagata and M. Iwasa, Effect of adsorbed water on

friction of hot-pressed silicon nitride and silicon carbide at slow speed sliding. Wear 121 (1988) 107.

55. M. Takatsu, H. Kamiya , K. Kubota and K. Seki, Effect of sliding condition on wear mechanism of SiC. J. Ceram. Soc. Jap., Int. Ed. 98 (1990) 968.

56. V. Aronov and T. Mesyef, Wear in ceramic/ceramic and ceramic/metal reciprocating sliding contact. part 1. Trans. ASME 108 (1986) 16.

57. C. Cm. Wu, R. W. Rice, D. Johnson and B. A. Platt, Grain size dependence of wear in ceramics . Ceram. Eng. Sci. Proc. 6 (1985) 995.

58. D. H. Buckley and K. Miyoshi , Tribological properties of structural ceramics. In 'Materials science and technology, vol. 29 (1989) 293,


59.

60.

61.

62.

63.

64.

65 .

66.

67.

68 .

69.

70.

71.

72.

73.

74.

75.

76.

77.

78.

79.

80.

81.

Academic Press, Inc. M. Woydt and K.-H. Habig, On the tribology of ceramic materials in closed systems. cfi/Ber. DKG 66 (1989) 198. M. A. Moore and F. S. King, Abrasive wear of brittle solids. Wear 60 (1980) 123. J. Breznak, E. Breval and N. H. McMillan, Sliding friction and wear of structural ceramics part 1 room-temperature behaviour. J. Mater. Sci. 20 (1985) 4657. U. Dworak, H. Olapinski and W. Stannek, VersleiPverhalten keramischer Werkstoffe. Ber. DKG 54 (1977) 416. J. Breznak, E. Breval and N. H. McMillan, Sliding friction and wear of structural ceramics part 2 analysis of room-temperature wear debris. J. Mater. Sci. 21 (1986) 931. Y. Nakamura and S. Hirayama, Wear tests of grey cast iron against ceramics. Wear 132 (1989) 337. P. Andersson and 0. Ylstalo, The influence of lubrication on ceramic and steel sliding contacts. Mater. Sci. Eng. A109 (1989) 407. C. S. Yust and F. J. Carignan, Observations on the sliding wear of ceramics. ASLE Trans. 28 (1984) 245. G. M. Carter, R. M. Hooper, J. L. Hensall and M. 0. Guillou, Friction of metal sliders on toughened zirconia ceramic between 298 and 973 K. Wear 148 (1991) 147. M. Woydt and K. H. Habi~, Influence of temperature and sliding speed on friction and wear of SiS1C and MgO-Zr<h. Ceram. Eng. Sci. Proc. 9 (1988) 1419. I. Birkby, P. Harrison and R. Stevens, The effect of surface transformations on the wear behavior of zirconia (TZP) ceramics. Ceram. Eng. Sci. Proc. 9 (1988) 1431. P. C. Becker, T. A. Libsch and S. K. Rhee, Wear mechanisms of toughened zirconias. Ceram. Eng. Sci. Proc. 6 (1985) 1040. R. W. Rice and C. Cm Wu, Wear and related .evaluations of partially stabilized Zr02. Ceram. Eng. Sci. Proc. 6 (1985) 1012. B. Hwang, C. R. Houska, G. E. Ice and A. Habenschuss, X-ray analysis of the near-surface phase distribution applied to wear on a PSZ disk. Adv. Ceram. Mater. 3 (1988) 180. G. W. Stachowiak and G. B. Stachowiak, Unlubricated friction and wear behaviour of toughened zirconia ceramics. Wear 132 (1989) 151. R. H. J. Hannink, M. J. Murray and H. G. Scott, Friction and wear of partially stabilized zirconia: basic science and practical applications. Wear 100 (1984) 355. T. E. Fischer, M. P. Anderson, S. Jahanmir and R. Salher, Friction and wear of tough and brittle zirconia in nitrogen, air, water, hexadecane and hexadecane containing stearic acid. Wear 124 (1988) 133. R. H. J. Hannink, M. J. Murray and M. Marmach, Magnesia-partially stabilized zirconias (Mg-PSZ) as wear resistant resistant materials. In "Proc. Int. Conf. on Wear of Materials", Reston, VA, ASME, NY, (1983) 181. M. Dzimko and M. Uemura, EinfluP der Warmebehandlun~ und der Oberflachenaufbereitung von Zirkoniaproben auf ihr tnbologisches Verhalten im Kontakt mit Graphit. Trib. und Schmier. Techn. 37 (1990) 40. J. F. Braza, H. S. Cheng and M. E. Fine, Wear of partially stabilized zirconia: sliding vs. rolling contact. Scr. Metall. 21 (1987) 1705. V. Aranov, Wear resistance anomaly of magnesia partially stabilized zirconia. ASLE Trans. 30 (1987) 100. G. W. Stachowiak and G. B. Stachowiak, Unlubricated wear and friction of toughened zirconia ceramics at elevated temperatures. Wear 143 (1991) 277. M. Woydt, J. Kadoori, K.-H. Habig and H. Hausner, Unlubricated sliding behaviour of various zirconia-based ceramics. J. Bur. Ceram. Soc. 7


(1991) 135. 82. T, E. Fischer, M. P. Anderson and S. Jahanmir, Influence of fracture

toughness on the wear resistance of yttria-doped zirconium oxide. J. Am. Ceram. 72 (1989) 252.

7 TRIBOLOGY AND ZIRCONIA

7.1 WEAR AND STRENGTH OF MG-PSZ SLIDING AGAINST STAVAX

7.1.1 Introduction

Mg-PSZ is as mentioned a potentially interesting wear resistant

material, Chapter and 6, references [1-6]. The relative brittleness of

ceramics makes the strength behaviour of the material during wear of

importance. No information whatsoever is available on this point.

Therefore, tests were done on Mg-PSZ (Nilcra), to relate strength and wear

conditions.

The phase transformation tetragonal ~ monoclinic in zirconia, which is

of crucial importance to the mechanical behaviour of Mg-PSZ as described in

e.g. [7-10], also influences its wear behaviour significantly. Again,

hardly any specific information with regard to wear is available. The

dilatation associated with the phase transformation results in a layer of

compressive stresses at a ground or worn surface. These compressive

stresses will have a closing effect on surface flaws, thus increasing the

strength.

To investigate the influence of wear on strength, wear tests were

performed using a 'cam'-on-disk configuration. The disk used for the

complete test series as described in this chapter was stavax.

Two types of wear-test series were performed: one series under ambient

conditions with polished Mg-PSZ surfaces, and one series with water as a

lubricant with ground Mg-PSZ surfaces. Additional tests were done with

ground Mg-PSZ under ambient conditions and with polished Mg-PSZ with water

as a lubricant. A wear mechanism, derived from the observations on the worn

surfaces and the wear debris, is proposed. The results from this study are

compared to the results from earlier studies mentioned in literature [2-6,

11].

84 Chapter 7. Tribology and zirconia

7 .1.2 Experimental

The material characteristics were determined as described in Chapter

2. The fracture toughness of the material was measured under dry conditions

(at a dew point of -40 °C) with three-point bend tests on twenty single

notched samples and with the Double Cantilever Beam (DCB) method on three

samples. The fracture toughness was also determined from seven single

notched samples which were broken in tap water. Ten single notched samples

which were aged in tap water for 200 h at 30 °C were taken out of the water

and the fracture toughness of ·these samples was measured at a dew point of

-40 °C. The strength of the Mg-PSZ was determined from three-point bend

tests both on 10 samples with polished surfaces and on 36 samples with

ground surfaces. The Vickers hardness of the stavax disk, Chapter 2, was

measured with the Leitz hardness tester at a load of 5.0 N.

The Mg-PSZ 'cams' were prepared by sawing and grinding. The worn

surfaces were either initially polished or initially ground. A layer of at

least 30 Jim was removed by polishing from the surface of the samples to

remove the residual stress layer caused by the grinding, Chapter 4 and

[12-15]. The ten surfaces of 15x1 mm2 of the 'cam' were worn tangentially

to the rotation direction of the steel disk. The grinding marks on the

ground samples used in the wear tests were orientated perpendicular to the

length direction of the three-point bend samples.

The steel disk was polished after each measurement. This assured that

every test started with the same surface conditions for the disk. The

hardness of the disk, about 6.5 GPa, was checked with Vickers hardness

measurements at a normal load of 5.0 N on an unworn part of the disk before

or after each test.

A schematic drawing of the wear testing apparatus is shown in Fig.

7 .1.1. It is based on a common polishing table, with the steel disk

Chapter 7. Tribology and zirconia 85

attached to it. The Mg-PSZ is· glued to the centre of a steel block of about

50x20x20 mm3• A brass cylinder, and when needed extra weights to vary the

normal pressure, are screwed on top of the centre of this block. The line

through the center of gravity of the cylinder and the center of gravity of

the Mg-PSZ is normal to the steel disk. This pile of Mg-PSZ, steel and

brass is kept in position by two strips of metal, which enable the pile to

move only in the vertical direction.

/,.- ........ , a ( I

I' 1 1 I '- -- _.,.."' I I I

u==+===F=F===!=: ===jl!-------- _t_

c -------rb e :~ ----8

d

· Fig. 7.1.1: Schematic drawing of the equipment; a is optioMl extra weight,

b is a metal strip only able to move in the vertical direction, c is the

steel-holder to which the Mg-PSZ cam, d, is attached, e indicates the

sliding direction of the disk and f is the tube used to remove residual

lubricant.

The wear tests were performed in air at room temperature and ambient

conditions. The sliding velocity was kept constant at 1.2 m/s for the inner


specimen and 1.5 m/s for the outer specimen. Before testing, the Mg-PSZ and

the disk were cleaned ultrasonically in ethanol, washed in ethanol and

dried in blown hot air. Improvement of this procedure is limited by

practical problems, especially because of the large size of the steel disk.

Also, the testing in air makes a more complicated cleaning procedure less

useful. Three different loads were used, 14 N, 19 N and 34 N, resulting in

a normal stress of 0.09 MPa, 0.13 MPa and 0.23 MPa respectively. At each

load, tests were done at three different time intervals, 20 h, 70 h and 200

h. Three tests were done twice to investigate the reproducibility of the

measurements.

Tests were performed under ambient conditions and with water as a

lubricant. The lubricant was applied to the disk at a constant velocity of

2 ml/min, in front of the contact between the steel disk and the Mg-PSZ.

The worn surfaces of two or three Mg-PSZ samples for each test

condition were examined in the sliding direction with a Taylor-Hobson

Talysurf. The results were digitally processed, and presented as Ra values.

The Ra values of the steel were measured after three tests with the same

equipment.

After each wear test, the ten samples of 15x3xl mm 3 were removed from

their basis by sawing. The dimensions of each sample were measured and they

were broken in a three-point bend test with the worn surface under tension.

The wear testing instrument was primarily designed to relate wear

conditions to strength which explains the 'cam' shape, Chapter 2, of the

Mg-PSZ sample. Wear volumes were, however, also determined. Probably

because of the relatively long total edge length in comparison with the

area loaded, it is difficult to make a comparison with other measurements

done on more conventional instruments. The weight of a Mg-PSZ sample was

measured before and after each test. From the weight difference and the

density, the wear volume of Mg-PSZ was calculated. Also, the height of each


sample at various positions was measured before and after each wear test.

This also gives an estimate for the wear volume of Mg-PSZ. With these wear

volumes, a wear coefficient, K, can be calculated according to: K = W /Px, v

where Wv is the wear volume, P is the normal load and x is the sliding

distance.

The wear track of the steel disk was measured with a profilometer.

Profiles were measured at various enlargements, depending on the size of

the tracks, at intervals of 30° and/or 90°. The wear volume was calculated

from the radii of the tracks and the average area from the profiles.

The worn surfaces were observed with optical microscopy using

Interference Contrast (!nCo), and with Scanning Electron Microscopy (SEM).

Some of the surfaces were qualitatively analyzed with Energy Dispersive

Element Analysis by X-rays (EDX). A polished and a worn Mg-PSZ surface were

observed with Scanning Acoustic Microscopy (SAM). The wear debris from

various tests was collected, and observed with SEM and qualitatively

analyzed with EDX.

7.1.3 Results

The characteristics of the Mg-PSZ and the steel are given in Table

7:1.1.


material Mg-PSZ

Young' s modulus (GPa) 195 ± 3.0

Poisson' s modulus (-) 0.324 ± 0.005

Vickers hardness (GPa) 12

P=2N

strength 3pb, MPa

polished

ground, 046

761 ± 33

915 ± 56

fracture toughness, MPa.mm

OCB three-point bend (3pb) 3pb measured in water 3pb after 200 h in water

10.3 ± 0.2 11.5 ± 1.1 7.5 ± 0.2

11.9 ± 0.9

roughness R , JJm •

polished ground 046

0.025 0.21

stavax

215

5.8-6.8

P = 5 N

0.001

Table 7.1.1: Characteristics of the Mg-PSZ and the steel used. The latter

is also known as stavax. All results given as X ± S stand for the average X

and the sample standard deviation S.

The quantitative results of the various tests are summarized in Tables

7.1.2, 7.1.3 and 7.1.4. The material loss after testing for 200 hat P = 34

N under ambient conditions was too large to do the required measurements.

It seems reasonable to conclude from the data that the test condition, and

not the initial surface preparation is the main cause of differences

between the results of the two complete test series.


time, load, u 3pb ± s, R1

, pm Wv Mg-PSZ, W stavax v

h N MPa Mg-PSZ stavax mm3 ; K, m2/N mm3

; K, m2/N

ground D46 915 ± 56 0.21 polished 761 ± 33 0.025

20 14 797 ± 22 0.099 3 2.1E-15 7 5.0E-15 70 14 779 ± 7* 0.161 16 ; 3.2E-15 13 2.6E-15 200 14 795 ± 31 0.100 50 ; 3.5E-15 28 2.0E-15 20 19 774 ± 22 0.194 24 ; 1.3E-15 70 19 770 ± 38 0.275 10 ; 1.5E-15 126 ; 1.9E-14 200 19 768 ± 24 0.059 240 ; 1.3E-14 20 34 802 ± 19 0.209 20 34 830 ± 19 0.327 0.047 125; 3.6E-14 117 ; 3.4E-14 70 34 795 ± 51 0.064 118 ; 9.8E-15 70 34 807 ± 27 0.213 0.064 115 ; 9.6E-15 59 ; 4.9E-15

Table 7 .1.2: Results of the wear tests performed under ambient conditions

on initially polished surfaces for Mg-PSZ. All results given as X ± S stand

for the average X and the sample standard deviation S. The results written

in boldface are the results of the reproducibility tests.

*one sample was not included for the calculation of the average, because

the strength of this sample deviated extremely from the other samples,

probably due to an exceptional large inherited flaw.

90 CIUJpter 7. Tribology and zirconia

time, load, a3pb ± S, R •• .um Wv Mg-PSZ, Wv stavax

h N MPa Mg-PSZ stavax mm3 ; K, m2/N mm3

; K, m2/N

ground D46 polished

20 14 70 14 200 14

20 70 200

20 70 200 200

19 19 19

34 34 34 34

915 ± 56 0.21 761 ± 33 0.025

838 ± 50 0.11 784 ± 26 0.094 790 ± 107 0.005

767 ± 19 0.17 740 ± 23 0.046 747 ± 19 0.017

715 ± 45 0.11 718 ± 29 0.075 719 ± 26 0.034 733 ± 26 0.041 0.059

0.2 3 6

2 5 4

4 12 20 15

1.4E-16 6.1E-16 4.3E-16

l.OE-15 7.5E-16 2.1E-16

2.1E-15 l.OE-15 5.8E-16 4.4E-16

1 53 183

7.1E-16 l.lE-14 1.3E-14

26 1.4E-14 94 ; 1.4E-14 120 ; 6.3E-15

21 ; 6.1E-15 108 ; 9.0E-15 163 ; 4.8E-15 101 ; 2.9E-15

Table 7.1.3: Results of the wear tests performed with water as a lubricant

on initially ground surfaces for Mg-PSZ. All results given as X ± S stand

for the average X and the sample standard deviation S. The results written

in boldface are the results of the reproducibility tests.

time, load,

h N

172 200 70

200 70 70 200

14 19 34

14 19 34 34

a 3pb ± S, R •• .um MPa Mg-PSZ

W Mg-PSZ, v 3 2

mm ;K, miN

ambient conditions, ground Mg-PSZ

812 ± 31 0.304 777 ± 42 0.314 813 ± 28 0.128

26 ; 2E-15 38 ; 2E-15 122 ; 1E-14

W stavax, v

mm3 ; K, m2/N

242; 2E-14 350; 2E-14 80 ; 7E-15

with water as a lubricant, polished Mg-PSZ

791 ± 46 0.004 774 ± 43 0.019 792 ± 30 0.011 766 ± 41 O.Ql5

2 ; lE-16 3 ; 4E-16 8 ; 7E-16 26 ; 8E-16

3 2E-16 33 5E-15 75 ; 6E-15 243; 7E-15

Table 7.1.4: Results of the wear tests done to compare the different test

conditions and the different initial surface preparations. It is concluded

that the test condition is the dominant factor and not the initial surface

preparation.


The tests done with water as a lubricant show a decrease in Ra with

increasing time after 200 h, where the average Ra is 0.02 Jlm. The scatter

around the average values for one time interval at various loads decreases

with increasing time. The data from the tests done under ambient conditions

show that the Ra values at various loads approximate an average of about

0.1 Jlm. The scatter around this average decreases with increasing time. The

Ra of surfaces worn under ambient conditions is higher than the Ra of

surfaces worn with water as a lubricant. No relation is found between Ra

and normal load.

After a period of running-in, the strength of the Mg-PSZ does not vary

significantly with time, at least until 200 h of testing. The strength

after wear with water as a lubricant is less than the strength after

testing under ambient conditions at P = 34 N.

The relations between load and strength are illustrated in Figs. 7 .1.2

and 7.1.3. The data from the tests done with water as a lubricant are shown

in Fig. 7 .1.2. Starting with a ground surface, which has a strength of

about 915 MPa, the strength decreases with increasing load. Starting with a

polished surface, there is a slight increase in strength at P = 14 N and a

decrease in strength at higher loads. The strength at P = 34 N of initially

ground Mg-PSZ is significantly less after wear with water as a lubricant

than the strength of initially polished or ground Mg-PSZ worn under ambient

conditions. The strength of initially polished Mg-PSZ after wear with water

as a lubricant lies between the results of the tests done under these two

conditions.

m a..

92

950.-----------------------~

m a..

Chapter 7. Tribology and zirconia

950.-------------------------~ t:. p, Oh

;:::;: 900

6. p , Oh 0 p, 20h v p, 70h 0 p, 200h • g, Oh

;:::;: 900

0 p, 20h v p, ?Oh 0 p, 200h j. g,Oh ..c ..c

c;, c:

c;, 1' g , 70h • g , 200h

2: c: 2:

Ui 850

1' g, 70h • g, 200h Ui 850

800

750

•

• 1'

v • • 1'

v

0

• '

800

750

• 0

v

0

700oL-----1~0----~20------3~0----~40

Load, N

10 20 30 Load, N

40

Fig. 7.1.2 (left): Strength-load curve, showing the three-point bend

strength of Mg-PSZ after testing with water as a lubricant. See text for

explanation.

Fig. 7.1.3 (right): Strength-load curve, showing the

strength of Mg-PSZ after testing under ambient conditions.

P = 34 N is higher than the strength at P = 19 N and

strength of a polished surface. See text for further explanation.

three-point bend

The strength at

higher than the

The data from the tests done under ambient conditions are shown in

Fig. 7 .1.3. The strength at P = 19 N is clearly less than the strength at P

= 34 N. All strength values are larger than the strength of polished Mg-PSZ

and less than the strength of ground Mg-PSZ.

The relation between wear volume and time, excluding some extreme

results explained by the anomalous sample shape, is approximately linear,

according to W v c t, with W as wear volume, c as a constant and t as I I

time. The relation between wear volume and load is also approximately

linear, according to Wv = c2(P-P c), with P as load and c

2 and P c as

constants. The values of these constants depend on testing conditions. A

critical load P c is suspected, below which there is hardly any wear.


The most essential obserVations from the wear surfaces and the wear

debris are given in Figs. 7.1.4-7.1.10. The observed phenomena occur more

or less on most of the worn surfaces. The wear condition, ambient or with

water as a lubricant, determines the relative frequency of the

characteristics of the wear surface. No influence of load or time interval

is observed.

Fig. 7.1.4: Wear surface of initially polished Mg-PSZ after testing for 200

h, at a normal load of 19 N, equivalent to 0.13 MPa, under ambient

conditions. A transition between the two characteristic wear surfaces is

visible. Adhesion is the dominant mechanism during testing under ambient

conditions. Photograph taken with Interference Contrast (!nCo).

94 Chapter .7. Tribology and zirconia

Fig. 7.1.5 (top): Wear surface of initially polished Mg-PSZ after testing

for 200 h, at a normal load of 14 N, equivalent to 0.09 MPa, under ambient

conditions. The shown surface is characteristic for the adhesion dominating

during wear under ambient conditions. Photograph taken with /nCo.

Fig. 7.1.6 (bottom): Wear surface of initially ground Mg-PSZ after testing

for 200 h, at a normal load of 19 N, equivalent to 0.13 MPa, with water as

a lubricant. The adhered patches of steel and the 'slip-stream' marks are

characteristic for wear with water as a lubricant. Sliding from top to

bottom. Photograph taken with /nCo.


Fig. 7.1.7 (top): Wear surface of initially ground Mg-PSZ after testing for

200 h, at a normal load of 14 N, equivalent to 0.09 MPa, with water as a

lubricant. The 'slip-stream ' marks, the clearly visible grain boundaries

and the steel patches are characteristic for a surface worn with water as a

lubricant. Sliding from right to left. Photograph taken with /nCo.

Fig. 7.1.8 (bottom): Wear surface of initially ground Mg-PSZ after testing

for 70 h, at a normal load of 34 N, equivalent to 0.23 MPa, with water as a

lubricant. This surface shows clearly the perpendicularly oriented

microcracks. Sliding from right to left. Photograph taken with SEM.


Fig. 7.1.9 (top): Wear surface of initially ground Mg-PSZ after testing for

70 h, at a normal load of 14 N, equivalent to 0.09 MPa, with water as a

lubricant. This surface shows the delamination and fracture of the Mg-PSZ.

Sliding from right to left. Photograph taken with SEM.

Fig. 7.1.10 (bottom): Wear debris, Mg-PSZ and steel. The figure shows a

sheet, containing Mg-PSZ on the visible surface (determined with EDX)

surrounded by oxidized steel spheres.


Two mechanisms (adhesion and abrasion) are found on samples worn under

each condition. Gradual transitions between these mechanisms, as shown in

Fig. 7 .1.4, are found, but under each condition one mechanism dominates.

The wear surface of the Mg-PSZ after wear under ambient conditions is

mainly covered with steel (adhesion). Metallic film transfer is usually

observed in literature about wear of ceramic/steel couples [3, 4]. An

example of such a surface is shown in Fig. 7.1.5. The wear surface after

wear with water as · a lubricant is formed mainly due to the process of

abrasive wear, with patches of steel adhered to the ceramic. This steel is

mainly located at the boundary of a sheet of Mg-PSZ which is still attached

to the bulk material. Observing a surface worn with water as a lubricant

with Interference Contrast, shows the preferential wearing of grain

boundaries under these conditions (Fig. 7 .1. 7). This is also observed after

testing a Mg-PSZ/Mg-PSZ couple in an acetic acid buffer solution [S]. The

'slip-stream' marks in Figs. 7 .1.6 and 7 .1. 7 are characteristic for

surfaces worn with water as a lubricant. Microcracks, shown in Fig. 7.1.8,

are often associated with adhered steel, and preferentially located at

grain boundaries. Most of the surfaces show the delamination of sheets, as

illustrated in Fig. 7.1.9.

The wear debris consists of oxidized steel spheres and thin sheets,

Fig. 7 .1.10. The composition of the sheets, examined with EDX, is found to

. be either mainly steel, mainly Mg-PSZ, or steel and Mg-PSZ both in

significant quantities .

7.1.4 Discussion

The average Ra value and the scatter around the average for the tests

done with water as a lubricant, and the scatter around the average for the

tests done under ambient conditions are time-dependent. This means there is

still a time effect after 200 hours which influences the local contact


situation. A variation in load, resulting in an increase in the number of

contact points rather than a change in shape and size of the contact points

explains why the Ra is independent of load. The Ra values for the tests

done with water as a lubricant on ground and polished Mg-PSZ are

approximately equal, .. but they are significantly lower than the Ra values

from the tests done under ambient conditions. This shows that the

environmental test conditions, and not the initial surface preparation,

will eventually determine the local contact situation. The R values give an •

indication about the geometry of the two surfaces that have been in

contact. This provides additional information on the investig~ted wear

systems.

The independence of the strength of Mg-PSZ from testing time indicates

the continuous development of a residual stress layer during steady-state

wear. After a period of running-in, equilibrium is reached between the

removal of material from the surface and the continuous development of a

residual stress layer.

The strength data at P = 14 N and 19 N are about equal after wear

under the two different conditions. At P = 34 N the strength of Mg-PSZ is

significantly less after sliding with water as a lubricant compared to the

strength after sliding under ambient conditions. This is explained by

assuming that the residual stress layer . developed due to sliding with water

as a lubricant is smaller than the residual stress layer . developed during

sliding under ambient conditions. This difference is caused by a difference

in friction coefficient or because of the higher temperature under ambient

conditions. The closure effect on flaws is thus greater under ambient

conditions at a load of 34 N than under conditions with water as a

lubricant at the same load. Whether there is a difference in flaw size due

to the different test conditions is still to be investigated. The fracture

toughness measurements show that the fracture toughness is smaller when


tested in water, but this doos not result in a measurable decrease in

strength. The strength after testing with water as a lubricant at P = 34 N

is less than the strength of polished Mg-PSZ. This is consistent with SAM

observations, which show a thinner residual stress layer after the

mentioned wear test than after polishing.

The strength after testing under ambient conditions at P = 34 N is

higher than the strength at P = 19 N. This indicates that the residual

stress layer is enhanced by the stronger mechanical interaction. Under more

severe conditions a higher strength should be reached.

The calculated value for the wear coefficient K is in most cases

higher after testing under ambient conditions than after testing with water

as a lubricant. The relative difference in these values of K are comparable

to the differences given in [2]. The influence of a decrease in fracture

toughness due to the presence of water is not measurable. The calculated K

is usually higher for the steel disk than for the Mg-PSZ, also as in [2],

except for a few tests done under ambient conditions.

The relation between wear volume and time is approximately linear.

This is in accordance with e.g. Archard's law [17]. The wear volume-load

relation shows a threshold point, P c' above which the wear volume suddenly

increases as discussed earlier. Since wear of Mg-PSZ is associated with

lateral cracks, a change in wear mechanism at P = P c from wear by plastic

deformation to wear by fracture is likely.

In Fig. 7 .1 .11, the characteristics of the proposed wear mechanism are

illustrated. The mechanism is based on the results of the tests done under

ambient conditions but, with adjustments, it is also applicable to wear

with water as a lubricant. The difference in wear rate, expressed in the

volume data, for these two conditions is probably explained by the

different friction coefficients. More friction under ambient conditions

results in more stress, and thus in higher stress concentrations.


1 Stavax

_/ --....

Mg-PSZ local differences in heig,t

2 Stavax

a~i~~ abrasi~

=oo..JlJ/ ' ul

Mg-PSZ

...::::.{({I II Q

~ v cracks

ad>esion of stavax + abrasion by debris

3 Stavax

-=-z "-~~~~~~~ ~Z"O:: ' ·: 1 ----- -Mg-PSZ

delaminati~ + transformati~ of U'lderlying zirconia

Fig. 7.1.11: Schematic illustration of the wear mechanism. See text for

explaMtion.

At steady-state wear, there will be a certain surface topography on

the Mg-PSZ (Fig. 7 .1.11.1). Plastically deforming steel will adhere to the

Mg-PSZ, mainly in 'valleys' (Fig. 7 .1.11.2). The 'hills' are abraded by

debris. This debris causes the 'slip-stream' marks located behind the

adhered patches. The stress pattern developed due to the sliding causes

lateral and vertical cracks. In the third step, Fig. 7 .1.11.3, the cracks

grow, and sheets of Mg-PSZ delaminate. The sheets resulting from this

delamination are found in the wear debris collected afterwards.

These observations correspond to the observations given in [6], where

surface-initiated subsurface microcracks and plate-like debris particles

are mentioned. The Mg-PSZ underneath the delaminated Mg-PSZ will transform

to monoclinic zirconia. At steady-state wear, there will thus be a


continuous cycle of the removal of transformed material from the surface,

and the transformation of the material which had been underneath this

sheet. This transformation and the accompanying residual stress is thus

restricted to thin layers at the surface. This will enhance the development

of lateral cracks. The thickness of the sheets found in the wear debris is

of the same order of magnitude as the thickness of the part of the residual

stress layer which contains the largest stress [12] . Also, the development

of subsurface lateral cracks is predicted in [11], where a mixed mode

(shear and compression) stress state is used to model the sliding of

brittle materials. In [18-21] the stress state during sliding is used to

predict delamination wear.

During wear with water as a lubricant, the Ra of the Mg-PSZ is of the

same order of magnitude as the Ra of the steel. There will thus be less

adhesion of steel to the Mg-PSZ, which corresponds with the above mentioned

observations.

The differences in hardness and yield stress for Mg-PSZ and the used

steel have interesting consequences. A polished Mg-PSZ surface models a

situation free of residual stress. The hardness of polished Mg-PSZ is about

12 GPa, giving an estimate for the overall flow stress of 4 GPa. The

critical transformation stress which is the onset of plastic deformation,

however, is about 1.1 GPa [12, 22, 23] in a triaxial stress state. This

value could be less in a biaxial stress state, but the exact relation

between stress state and critical transformation stress is not yet

clarified [12]. The hardened steel has a Vickers hardness of about 6.5 GPa.

A rough estimate for the yield stress of the steel is given by one third of

the hardness, although it could become significantly less under ambient

conditions due to the increase in temperature. In [6], wear tests are

described between PSZ and steel under ambient conditions at loads of 5 - 40

kg and at velocities of 1 m/s and 4 m/s. The observations done on the

surface of the PSZ are comparable to the observations described in this


report. In [6], the high temperature and high plastic deformation at the

PSZ/metal interface are mentioned.

The polished Mg-PSZ is thus the first material to yield due to the

phase transformation. This plastic deformation influences the initial

stages of the wear process, in particular the number and shape of contacts

points. This results in a change in local pressure.

Gradually, as a consequence of the phase transformation, the Mg-PSZ is

strain hardened. The critical transformation stress, responsible for

initial yield, is 1.1 GPa, but for complete transformation a stress of

probably more than 3.5 GPa is required in an equibiaxial stress state [12].

If thermal effects are neglected, the strain hardening will raise the yield

stress of the Mg-PSZ, until it is equal to the yield stress of the steel.

After this point, the local pressure is assumed to remain constant, at a

level close to the yield stress of the steel. This will also hold for other

wear couples with Mg-PSZ. This consequence has to be verified

experimentally.

Finally, in a steady-state situation, the partly transformed surface

of Mg-PSZ will deform the steel surface plastically. The patches of steel

adhered on the Mg-PSZ are examples of plastically deformed steel.

An initially ground Mg-PSZ surface will then initially plow through

the steel, until the residual stress layer on the Mg-PSZ caused by grinding

is removed. The wear conditions will then determine the residual stress

profile as described above.

A problem connected to the relation between residual stress and wear

is the relation between the phase transformation and the process which

causes the transformation. The transformation can occur due to the

temperature increase, due to the normal load or due to the transverse

force. It is assumed that the influence of thermal energy during lubricated

tests can be neglected. The lubricants are supposed to be active as cooling


liquid. The temperature at the contact points during tests under ambient

conditions seems to be high enough to cause the transformation, but it is

not known what the temperature rise is quantitatively, and more

importantly, it is not known how far the temperature rise goes into the

heat-insulating material.

The normal force component of the load is supposed to have no

significant contribution to the transformation. The overall pressure is far

less than the critical transformation stress. Incidental interactions

between the Mg-PSZ surface and loose debris particles are assumed to be

negligible, although they could well be the cause of the 'slip-stream'

marks shown in Figs. 7.1.5 and 7.1.6.

The tangential forces caused by the lateral movement between the two

materials arc supposed to be the main cause of the phase transformation. It

is thus interesting to investigate the relation between friction and

residual stress.

7.1.5 Summarizing conclusions

The roughness of worn Mg-PSZ is independent of the load. This indicates

an increase in the number of contact points with increasing load, and not

a change in shape and size of contact points.

- The Ra of Mg-PSZ worn with water as a lubricant decreases to 0.02 J,lm

after 200 hours. The Ra of Mg-PSZ worn under ambient conditions

approaches an average of 0.1 Jim after 200 hours. The scatter around

average Ra values decreases with increasing time under both conditions.

- The environmental conditions determine the wear behaviour of the Mg-PSZ

at the local contact points after a period of running-in. The influence

of the initial surface preparation is negligible after a period of

running-in.

The strength of Mg-PSZ is independent of wear time after a period of

running-in.


The residual stress layer is the strength-dominating factor after

polishing, grinding and wear, all three as described in this paper.

The wear volume is approximately linear in time. There is a threshold

load in the wear volume - load relation. This threshold load could

indicate the transition of wear by plastic deformation to wear by

fracture.

- The following wear mechanism is proposed for the steady state situation:

1. adhesion of steel on the partly transformed surface of the Mg-PSZ, in

the lower parts of the surface topography of the Mg-PSZ, either in

large quantities (ambient conditions), or in smaller amounts (with

water as a lubricant).

2. Abrasion of the higher parts of the surface topography of the Mg-PSZ

caused by the wear debris.

3. The development of perpendicular and lateral cracks in the Mg-PSZ.

4. The delamination of sheets of either Mg-PSZ, steel or Mg-PSZ plus

adhered steel and the transformation of underlying zirconia.

7.2 WEAR AND STRENGTH OF MG-PSZ AND Y-TZP

7 .2.1 Introduction

The concepts presented in Chapter 7.1 are based on elaborate test

series between Mg-PSZ and stavax. the conclusions and considerations given

in Chapter 7.1 can be extended to comparable systems by changing part of

the wear system. It is interesting to examine whether there are differences

with Mg-PSZ from a different supplier, whether Y-TZP shows the same

behaviour, and what happens if the disk is changed. Additional measurements

were therefore performed to obtain some answers on the above mentioned

questions.


7 .2.2 Experimental

The same instrument and procedures were used as described in Chapter

7.1, only this time with different materials. The 'cams' used were Mg-PSZ

(Feldmuhle) and Y-TZP (Dynamic Ceramic). The disks used were stavax and

alsint alumina. The material characteristics are given in Chapter 2. The

same tests as described in Chapter 7.1 were performed on various material

combinations. Severe wear limits the duration of some tests and this

resulted in some tests of 5 h, some of 24 h and some of 200 h. Both

unlubricated tests were performed and tests with water as a lubricant,

applied as described in Chapter 7 .1. The various test conditions and

combinations are shown in the tables with the results that will follow.

7 .2.3 ResuJts

Tables 7 .2.1 and 7 .2.2 summarize the results for Mg-PSZ and Y-TZP

respectively. The wear volumes are represented by the wear coefficient. The

wear coefficient, K, of Mg-PSZ for the Mg-PSZ against stavax tests, Table

7.2.1, under ambient conditions is 2.10-1~ m2/N on the average, and 5.10-16

m2/N with water as a lubricant. These values as well as the K values of the

stavax, are comparable to the values given for Mg-PSZ (Nilcra), sliding

against stavax, Chapter 7 .1. The wear coefficients for the tests with

alumina alsint are about 1.10-13 m2/N

substantially higher.

The wear coefficients for Y-TZP are in all cases higher than for

Mg-PSZ, Table 7.2.2. Tests of 200 h could thus not be performed because the

whole 'cam' would be worn down. The wear coefficients of the stavax disk

are also higher compared to the tests with Mg-PSZ.

The results· of the strength measurements are visualized in Figs.

7.2.1, Mg-PSZ, and 7.2.2, Y-TZP. The remaining strength after the tests

with stavax shows the same dependence on load as described in Chapter 7 .1.

At ambient conditions a strength higher then the strength of a polished


sample, due to the transformation, and with water as a lubricant a strength

less then the strength of a polished sample. The large amounts of wear with

alsint has limited the number of tests.

time,

h

200 200 200

W disk v

load, a3pb ± S, Wv Mg-PSZ,

N MPa

ambient conditions, ground Mg-PSZ I polished stavax

14 19 34

599 ± 32 1.6 ; l.lE-16 74 ; 5.2E-15 581 ± 46 17 ; 8.7E-16 232 ; 1.2E-14 571 ± 23 187 ; 5.5E-15 333 ; 9.7E-15

water as a lubricant, ground Mg-PSZ I polished stavax

200 200 200

5 24 5

200 170 24 200

14 19 34

584 ± 57 4 469 ± 30 8.6 463 ± 21 22

2.8E-16 30 4.5E-16 58 6.3E-16 141

2.1E-15 3.0E-15 4.1E-15

ambient conditions, ground Mg-PSZ I polished alsint

14 14 34

562 ± 62 17 ; 4.8E-14 503 ; 3.0E-13

489 ± 66 234 ; 2.7E-13

water as a lubricant, ground Mg-PSZ I polished alsint

14 19 34 34

560 ± 33 1.1 ; S.OE-17 514 ± 34 247 ; 1.5E-14

> 600; > 2E-14 > 600; > 2E-13

Table 7.2.1: The results of the wear-strength measurements on Mg-PSZ

(Feld.).


time, load, a3pb ± S, Wv Y-TZP, W disk

v

h N MPa

ambient conditions, polished Y-TZP I polished stavax

5 14 794 ± 52 25 ; 7.1E-14 4.7 ; 1.3E-14 24 14 801 ± 58 107 ; 6.3E-14 115; 6.8E-14 200 14 559 ; 4.0E-14 5 19 785 ± 49 22 ; 4.6E-14 3.8 7.9E-15 24 19 835 ± 32 153 ; 6.7E-14 26 l.lE-14 5 34 804 ± 50 57 ; 6.7E-14 8.1 9.5E-15 5 34 751 ± 35 72 ; 8.4E-14 5.3 6.2E-15 24 34 831 ± 56 339 ; 8.2E-14 43 l.lE-14

water as a lubricant, polished Y-TZP I polished stavax

24 14 1051 ± 112 2.3 1.4E-15 20 1.2E-14 24 19 1012± 152 2.0 8.7E-16 31 1.4E-14 5 34 902 ±127 10 l.2E-14 4.5 5.3E-15 24 34 868 ± 119 65 1.6E-14 26 6.3E-15

ambient conditions, polished Y-TZP I polished alsint

24 14 913 ± 68 154 ; 9.1E-14

water as a lubricant, ground Y-TZP I polished alsint

5 14 >600; >2E-12

Table 7.2.2: The results of the wear-strength measurements on Y-1ZP.

The results of the strength tests with Y-TZP are presented in Fig.

7 .2.2. The lowest strength, independent of load, is obtained after tests

with stavax at ambient conditions. The tests with water as a lubricant show

approximately the same behaviour as Mg-PSZ, a decreasing strength with

increasing load.


650

+ poliShed 600 6

• 6 0 goo..rd

~ ' 550 6 amb. stav

t • • water. stav

~ 500

• • amb. Al203 • • 450 • water. Al203

400 0 10 20 30 40

load. N

Fig. 7.2.1: Results of the strength measurements of the worn Mg-PSZ after

testing against stavax (stav), and alsint (Al20~, under ambient conditions

(amb), and with water as a lubricant.

"' ~

t "'

1200

1100

1000

900

800

700 0

•

•

A

10

+ polished

0 goo..rd • 6 amb. stav

• water. stav

• 6 6 • amb. Al203

6 6

6

20 30 40

load. N

Fig. 7 .2.2: Results of the strength measurements of the worn Y-1ZP after

testing against stavax (stav), and alsint (AI 0 ), under ambient conditions 2 3

(amb), and with water as a lubricant.


7 .2.4 Discussion

The wear behaviour of the two Mg-PSZ's from the two different

suppliers is comparable, although the absolute values for the strength are

significantly different. The amount of wear for Mg-PSZ sliding against

alsint is larger then for Mg-PSZ sliding against stavax. The alsint disk

shows no loss of material after the tests, and there is some adhesion. This

results in all possible combinations of Mg-PSZ I Mg-PSZ contacts, alumina I

alumina contacts and Mg-PSZ I alumina contacts.

The wear behaviour of Y-TZP is different from Mg-PSZ. The amount of

wear is higher, thus limiting the number and duration of tests. The concept

of degradation, Chapter 3, could be the cause of this, because the tests

were performed under ambient conditions and with water as a lubricant.

There was thus enough water available to influence the material. The stavax

shows also more wear sliding against Y-TZP. Tests of Y-TZP against alsint

are practically not possible under the chosen test-conditions. The strength

values for Y-TZP sliding against stavax. with water as a lubricant are

comparable to the results for Mg-PSZ. The strength after tests at ambient

conditions is independent of load and less then the strength of polished

material. The influence of the phase transformation and residual stress in

Y-TZP is different as explained in Chapter 5.2. The introduction of damage

during testing under ambient conditions is apparently far more important

then the increase in strength due to residual stress. More research has to

be done to examine the behaviour of Y-TZP that is different from Mg-PSZ.

7 .2.5 Summarizing remarks

- The wear behaviour of two Mg-PSZ's from different suppliers is

comparable, although the absolute values of the mechanical properties of

both materials are different.

The wear resistance for Mg-PSZ is better then for Y-TZP under the studied


circumstances.

The wear resistant for zirconias sliding against stavax is better then

for zirconias sliding against alumina.

References

1. D. C. Cranmer, Ceramic tribology - needs and opportunities. Trib. Trans. 31, (1987) 1642.

2. K. H. Z. Gahr, Sliding wear of ceramic-ceramic, ceramic-steel and steel-steel pairs in lubricated and unlubricated contact. Wear 133 (1989) 1.

3. G. W. Stachiowiak and G. B. Stacbiowiak, Unlubricated friction and wear behaviour of toughened zirconia ceramics. Wear 132 (1989) 151.

4. G. W. Stachiowiak G. B. Stachiowiak and A. W. Batchelor, Metallic film transfer during metal-ceramic unlubricated sliding. Wear 132 (1989) 361.

5. R. H. J. Hannink, M. J. Murray and H. G. Scott, Friction and wear of partially stabilized zirconia basic science and practical applications. Wear 18 (1984) 355.

6. T. A. Libsch, P. C. Becker and S. K. Rhee, Friction and wear of toughened ceramics against steel. Proc. JSLE Int. Trib. Conf. (1985) 185.

7. A. G. Evans, Perspective on the development of high-toughness ceramics. J. Am. Ceram. Soc. 73 (1990) 187.

8. R. M. McMeeking and A. G. Evans, Mechanics of transformation-toughening in brittle material. J. Am. Ceram. Soc. 65 (1982) 242.

9. B. Budiansky J. W. Hutchinson and J. C. Lambropoulos, Continuum theory of dilatant transformation toughening in ceramics. Int. J. Solids Structure 19 (1983) 337.

10. A. G. Evans and R. M. Cannon, Toughening of brittle solids by martensitic transformations. Acta Metall. 34 (1986) 761.

11. A. R. Rosenfield, Fracture of brittle materials under a simulated wear system. J. Am. Ceram. Soc. 72 (1989) 2117.

12. P. H. J. van den Berg and G. de With, Residual stress and the stress-strain relation for Mg-PSZ. J. Europ. Ceram. Soc. 9 (1992) 165.

13. D. J. Green, F. F. Lange and M. R. James, Factors influencing residual surface stresses due to a stress-induced phase transformation. J. Am.Ceram. Soc. 66 (1983) 623.

14. D. J. Green, Compressive surface strengthening of brittle materials. J. Mater. Sc. 19 (1984) 2165.

15. R. Samuel and S. Chandrasekar, Effect of residual surface stresses on the fracture of ground surfaces. J. Am. Ceram. Soc. 72 (1989) 1960.

16. R. R. Hughan R. H. J. and Hannink, Precipitation during controlled cooling of Magnesia - Partially - Stabilized - Zirconia. J. Am. Ceram. Soc. 69 (1986) 556.

17. J. F. Archard, Contact and rubbing of flat surfaces. J. Appl. Phys. 24 (1953) 981.

18. N. P. Sub, An overview of the delamination theory of wear. Wear 44 (1977) 1.

19. S. Jahanmir and N. P. Sub, Mechanics of subsurface void nucleation in delamination wear. Wear 44 (1977) 17.

20. J. R. Fleming and N. P. Sub, Mechanics of crack propagation in delamination wear. Wear 44 (1977) 39.

21. J. R. Fleming and N. P. Sub, The relationship between crack

Chapter 7. Tribology and zirconia Ill

propagation rate and wear ·rates. Wear 44 (1977) 61. 22. 1.-W. Chen and P. E. R. Morel, Implications of transformation

plasticity in ZrOz- containing ceramics: I, shear and dilatation effects. J . Am. Ceram. Soc. 69 (1986) 181.

23. M. V. Swain, Inelastic deformation of M~-PSZ and its significance for strength-toughness relationships of zircoma toughened ceramics. Acta metall. 33 (1985) 2083.

8 PIN-ON-PLATE AND PIN-ON-DISK MEASUREMENTS

8.1 Y-TZP RECIPROCATING AGAINST SIALONS AND ALONS

8.1.1 Introduction

For this study wear systems of a Y-Tetragonal-Zirconia-Polycrystalline

(Y-TZP) sphere sliding reciprocative against various sialon plates and

three different types of AlON plates were investigated. Sialons are an

interesting kind of structural ceramic [1-3], often mentioned in literature

as a wear-resistant material [4-12] and in some cases specifically as a

material suitable as a cutting tool. The three AlONs, that were processed

with different amounts of Al20

3, are a new kind of translucent ceramic

[13-15]. Wear tests between Y-TZP and AlON could provide interesting

information about the wear behaviour of systems like the system of

Y-TZP/sialon or more general, systems of TZP against harder and more

brittle ceramics.

The wear tests on the sialons were performed in two test series. The

first series was done to investigate the influence of variations in the

phase content or the composition of the sialons on the friction and wear

characteristics of the systems. The second series, performed on a single

type of o:-sialon, was done to study the influence of load and velocity and

to examine the reproducibility of the measurements. Earlier tests on

various P-sialons, as described in [16], were performed under slightly

different conditions than the first series of tests for the o:-sialons. Some

additional measurements were therefore performed on one P-sialon to compare

the tribological properties of a- and P-sialons under the same conditions.

The AlONs were tested to examine tribological differences between the

three types and to examine the influence of frequency and load.

The friction coefficient and the total vertical displacement were

continuously measured during the tests and sampled at certain time

Chapter 8. Pin-on-Plate and Pin-on-Disk measurements 113

intervals. This continuous measuring of the vertical displacement is not

recommended in the ASTM standard [17] because of the effects of debris and

film formation, but it does give relevant information about the system

during the test. Both film formation and the role of debris can be

incorporated in a model describing the wear behaviour.

After each test various geometrical measurements were performed and

the worn surfaces were visually examined with optical microscopy (OM) and

scanning electron microscopy (SEM). Some surfaces were examined with energy

dispersive analysis of X-ray (EDX). The data and observations were used to

derive a wear mechanism.

8.1.2 Experimental

The pin material used was in all tests a commercially available Y-TZP

sphere with a radius of 2 mm and a polished surface. In Chapter 2, the

properties of Y-TZP, measured on a plate of the same material from the same

supplier, are given. These values are assumed to be comparable to the

properties of the spheres. The sialons and the AlONs were fabricated at the

Centre for Technical Ceramics. A summary of some of their properties is

given in Chapter 2. Most tests with the sialons were performed on a-sialons

with different compositions, code A1, A2, A3, A4 and on composites, code

A5, A6 and A7. Additional tests were done with a P-sialon, B6. Details

about the processing and characteristics of these sialons are given in

[16]. Details about the AlONs are given in [14, 15].

The wear tests were performed on a Pin-on-Plate tribometer (Central

Technical Workshop, Eindhoven University of Technology). In this setup, a

spherical Y-TZP pin reciprocates with a sinusoidal velocity on the flat

plate. The track length for the tests was chosen as 10 mm with pin

frequencies, f, for the tests with the sialons of 1, 4 and 8 Hz

114 Chapter 8. Pin-on-Plate and Pin-on-Disk measurements

corresponding to maximum velocities of respectively 0.02 m/s, 0.08 m/s and

0.16 m/s. The normal loads, P, used were 2 and 8 N. The AlONs were tested

at 1 and 4 Hz also at normal loads of 2 and 8 N. The duration was varied

from 8 to 72 h for the sialons and the AlONs were tested for 24 and 72 h.

The experiments were performed in a flow of dry nitrogen at room

temperature. This gives an approximately constant environment of less then

1% relative humidity.

The total vertical displacement of the pin during sliding was measured

by an extensometer (Sangamo DG1). The friction force was measured by a

force transducer. The displacement and friction force were simultaneously

sampled under external triggering control with a computerized data

acquisition system. After each external trigger pulse 2560 data points were

taken in 5 series of 512 points both for the vertical displacement and for

the force signal with a certain sampling frequency. For a pin frequency of

1 Hz, a sampling frequency of 50 Hz was used and for frequencies of 4 and 8

Hz, the sampling frequency was 100 Hz. The vertical displacement values

were averaged. The auto-power spectrum of the force signal was calculated

with the Fast Fourier Transformation (FFT) [18]. This FFT was performed on

the 5 time records of 512 points whose individual power spectra were

subsequently averaged. The total energy was approximated by integrating the

first, third and fifth order harmonic of the auto power spectrum. Higher

order harmonics can be neglected. The friction coefficient

calculated from these values according to:

f= (Q +Q +Q )1/2

I 3 5

c

where Q. is the power of the ilh harmonic and P the normal load. l

f was c

Various additional geometrical measurements were performed. The weight

loss of the pin was measured when possible and the track width on the plate


was measured with the device · mounted on a Leitz microhardness tester. The

decrease in diameter of the worn Y-TZP sphere was measured with the

thickness gauge, Chapter 2, and profLles of some wear tracks on the plate

were measured with the profLlometer, Chapter 2. The worn surfaces were

observed with OM and SEM. Some surfaces of worn spheres were examined with

EDX.

Two test series were performed with the sialons as described in Table

8.I.I. The first series was · performed to determine whether there are

significant differences in the tribological properties of the various

sialons and composites. Various P-sialons were tested at a load of 4 N and

at a frequency of I Hz in an earlier study [16] , whereas in the present

study the various a-sialons and the composites were tested at a load of 8 N

and at frequencies of I and 4 Hz. The second series was performed on one

a-sialon, A2 for no particular reason, to investigate the reproducibility

of the measurements, to study the influence of load and frequency and to

derive a wear mechanism for this system. This a-sialon was tested at normal

loads of 2 and 8 N, at frequencies of I, 4 and 8 Hz and at times of 8, 24

and 72 h as shown in Table 8.I.l. Four additional tests were done on a

P-sialon, B6, under similar conditions as the tests performed on tbe

a-sialon to compare the wear behaviour of the P-sialon and the a-sialon.


material P, N f, Hz t, h

first series

A1-A7 8 1 72 A1-A7 8 4 24 B1-B5 4 1 50

second series

A2 2 & 8 1, 4 & 8 8, 24 & 72 B6 2 & 8 1 & 4 24 & 72

Table 8.1.1: Measurement scheme in which the first series was performed to

examine the tribological properties of various sialons and the second

series to examine the influence of P, f and t.

8.1.3 Results

In the following we present first some considerations about the

geometry of the contact surface between a worn sphere and plate. This is

done because the interpretation of the geometrical measurements is not

straightforward if both pin and plate show significant amounts of wear. The

contact surface between the plate and the pin has a complex geometry.

Therefore two models, which are illustrated in Fig. 8.1 .1, were used to

approximate the shape of the contact surface from the measured geometrical

data. A listing of the symbols used is included in the text of Fig. 8.1.1.


A

Fig. 8.1.1: Two models to describe the shape of the contact surface between

a sphere and a plate if both materials show wear. The models are explained

in the text. The initial radius of the sphere is given by r, a is the

radius as measured from the track width, y is the decrease in diameter of

the sphere as measured, z is the maximum vertical wear-depth as derived

from a and r, Al is the worn area of the plate in cross section, A2 is the

worn area of the sphere in cross section, r' is the new sphere radius used

in model A, and b is the radius of the circle which can be imagined on the

assumption of a completely flat pin used in model B .

According to model A, the surface of the sphere, which is in contact

with the plate, will develop into a flatter shape which can be seen as part

of a new sphere with a larger radius r'. This r' can be calculated from y,

r and a, and can be used to calculate the wear volumes of the sphere, using


the area of A2, and plate, using the area of AI. In model B, the sphere is

flat at the bottom and keeps the original radius at the side which has also

been in contact with the plate. This represents an upper boundary with a

minimum amount of wear for the sphere. The radius of the bottom circle, b,

can be calculated from again r, y and a and the wear volumes of the pin and

plate can be calculated using respectively area A2 and area A 1. Model A

results in a maximum amount of wear for the sphere and a minimum amount for

the plate, while model B results in the opposite. Both models assume that

the profile of the sphere in the direction of the wear track is the same as

the proflle at a right angle to the wear track.

In Figs. 8.1.2a and 8.1.2b two examples are given of profiles of wear

tracks. The radius of curvature and the depth of the wear track can be

measured from these graphs. The shape of the profiles clearly resembles the

two models A and B from Fig. 8.1.1.

The width of the wear track, 2a, can be measured directly. The

decrease in diameter of the sphere, y, can be measured only if enough

material has been removed during the wear process. The weight measurements

of the pin give the true volume losses which can be used to determine

whether model A or model B is more appropriate. A factor A,

A = W -W

w I B

WA- WB , will be used further on. The term W

wt

defined as:

is the volumetric

amount of wear as calculated from the measured weight loss and W A and W 8

are the theoretical wear volumes based on respectively model A and B. A

value of 0 for A means wear according to model B and a value of 1

indicates wear according to model A.

The value of a can also be used to calculate z, which is the summation

of the decrease in diameter of the sphere and the depth of material removed

from the plate. The value of z can be compared with h, although they are

not supposed to be equal.

Chapter 8. Pin-on-PlaJe and Pin-on-Disk measurements

20

E :>.

-20 - 2 0 2

mm

500.----------------------------------------

E :>.

-5001-.~2------~----------~0--------~------~2 mm

119

Fig. 8.1.2a (top): The profile of a wear track which was caused by a test

at 2 N and 4 Hz for 72 h on sample A2. This profile shows a clear

correspondence to model A from Fig. 8.1.1.

Fig. 8.1.2b (bottom): The profile of a wear track which was caused by a

test at 2 N and 8 Hz for 24 h on sample A2. This profile shows a clear

correspondence to model B from Fig. 8.1.1. The scales in Figs. 8.1.2a and

8.1.2b. are entirely different.

First the results of the test series with the sialons will be


discussed. The results of the first series of tests performed on various

P-sialoos are presented in [16). These tests show no differences in

tribological properties between the various P-sialons.

The results of the first series of tests on the a-sialons and the a-P

composites are given

first in Table 8.1.2

in Table 8.1.2. The friction coefficients f (0) listed c

are the values at the beginning of the test and the

values f (t) are the steady state friction coefficients. The tests c

performed at a frequency of 1 Hz resulted in hardly any wear. Therefore no

y values are given. The only geometric variables which could be measured

reliably were a and the continuously measured vertical displacement, h.

Some of the tests performed at 4 Hz did show enough wear to determine all

the other variables. The results of this analysis are given in Table 8.1.3.

The average values given in Tables 8.1.2 and 8.1.3 show a small, but

consistent, difference in wear between the a-sialons and the composites.

The values for a, z and h in Table 8.1.2 show the difference for the whole

system of Y-TZP/sialon while Table 8.1.3 shows the separate values for the

sphere and the plate. It is shown in Table 8.1.3 that the wear of the Y-TZP

spheres depends more on the sialoo type than the wear of the sialon plates

themselves. The steady state friction coefficient f (t) c

is less for tests

with a composite than with tests with a pure a-sialon.

The value for Ll in Table 8.1.3 depends on the sialon type, pure a or a

composite, and thus on the amount of wear. Less wear, as for the

composites, gives a low value for Ll and thus a preference for model B.


code a y z h r' b f (0) f (t) c c

results of the measurements performed at 1 Hz for 72 h at 8 N

a-stalons

A1 0.40 0.040 0.60 0.50 A1d 0.32 0.026 0.063 0.34 0.50 A2 0.61 0.100 0.078 0.65 0.50 A3 0.58 0.086 0.090 0.52 0.40 A 3d 0.045 0.41 0.41 A4 0.55 0.077 0.066

average ----u-:49" ~

composites

A5 0.50 0.064 0.038 A5d 0.35 0.030 0.45 0.34 A6 0.100 0.30 0.47 A6d 0.42 0.044 0.047 0.48 0.44 A7 0.40 0.040 0.041 0.31 0.29

average --u:4Z ""U:"Oi3

results of the measurements at 4 H for 2 h at 8 N

a-sialons At 1.70 0.43 0.95 >0.80 3.01 1.24 0.25 0.90 Ald 0.26 0.39 0.60 A2 1.47 0.38 0.64 >0.50 4.16 1.17 0.37 0.74 A3 1.53 0.36 0.71 >0.50 3.47 1.14 0.15 0.67 A 3d 0.24 0.24 0.52 A4 1.58 0.41 0.78 0.62 3.60 1.21 0.13 0.70

average 1.37 {}.ff

composites

AS 1.36 0.35 0.53 >0.30 5.13 1.13 0.14 0.50 A6 0.12 0.50 A6d 1.40 0.28 0.57 0.45 3.51 1.02 0.31 0.53 A7 1.53 0.31 0.72 0.52 3.08 1.06 0.16 0.67

average T.U o:or

results of the measurements at 1 H for 5 h at 4 N, P-sialons

B1 0.027 0.52 B2 O.Dl8 0.44 B2d 0.014 0.40 B3 0.010 0.37 B3d 0.025 0.42 B4 0.025 0.48 B5 0.012 0.39


Table 8.1.2 (previous page): Results of the first series of wear tests. All

tests were performed at a load of 8 N. The codes in the first column give

the sia/on type (A1-A7). A 'd' in the first column stands for a duplicate

measurement. The geometrical variables a to b are given in mm and the

meanings of the geometrical variables and the friction coefficients in the

last two columns are given in the text. The codes B1-B5 stand for various

P-sia/ons which were processed along different routes. The density of B1-B5

varies from 3.08 to 3.21 g/cm3, HV (2.0 N) from 14.5 to 17.3 GPa, E from

218 to 249 GPa, v from 0.285 to 0.299, K from 2.51 to 3.63 MPa.m 112 and lc

ubi from 383 to 467 MPa.

code WA WB w L1 w WB wcalc wt A sphere sphere weight plate plate plate

a-sialons

A1 2.33 1.07 1.45 0.30 12.15 15.73 14.65 A2 1.41 0.83 1.15 0.55 5.30 7.14 6.13 A3 1.49 0.76 0.85 0.14 7.35 9.64 9.32 A4 1.83 0.98 1.43 0.53 7.79 10.3 8.99

average r.n- "CJ":7I

composites

A5 1.09 0.72 0.71 IV 0 3.34 4.59 4.59 A6d 0.93 0.46 0.56 0.21 5.43 7.07 6.73 A7 1.29 0.56 0.79 0.31 8.50 10.86 10.13

average 0":09""" ~

Table 8.1.3: Calculation of the wear volumes (W) in mm3 according to

models A (W ) and B (W ) for sphere and plate for the tests performed A B

at a frequency of 4 Hz at a load of 8 N. The calculated wear volumes

for the spheres are compared with the wear volumes derived from weight

losses (W ...J and the relative importance of the models is expressed in

Lt. With this L1 an estimation for the wear volume of the plate can be

calculated. This calculated value is given in the last column as wcalc

Chapter 8. Pin-on-Plate and Pin-on-Disk measurements

t, (h) a y h z f (t) b c

a-sialon A2, P= 2 N, f = 1 Hz

8 0.17 0.003 0.007 0.45 24 0.21 0.007 O.ot1 0.44 72 0.21 0.013 0.012 0.37

a-sialon A2, P = 2 N, f = 4 Hz

24 0.39 0.034 0.038 0.54 0.18 72 0.51 0.045 0.065 0.46 0.24

a-sialon A2, P = 2 N, f = 8Hz

8 1.11 0.21 0.180 0.334 0.80 0.89 24 1.33 0.28 0.280 0.507 0.80 0.91


8 0.27 0.013 0.019 0.39 24 0.34 0.022 0.028 0.44 72 0.61 0.080 0.096 0.50


8 0.32 0.033 0.026 0.27 24 1.47 0.38 0.600 0.645 0.65 1.17

a-sialon A2, P = 8 N, f = 8Hz

4 1.11 0.23 0.280 0.333 0.47 0.94

P-sialon B6, P = 2 N, f = I Hz

24 0.19 0.007 0.009 0.39 72 0.23 0,015 0.014 0.40

P-sialon B6, P = 2 N, f = 4 Hz

72 0.34 0.080 0.075 0.47

P-sialon B6, P = 8 N, f = I Hz

72 0.54 0.028 0.029 0.37

Table 8.1.4: Results from the second series of wear tests. The symbols

are explained in the text and the geometrical variables are in mm.

123

124

-0.01

E E

i ~ -0.02

~ ~ ;;

~ - 0 .0 3

:

-0.04

- 0 .0 5

-0.00~

-0.0 1

E

: ,:-0 .015

~ ~ - 0.02

0. ~0.025

~ .:: -O.O.l

: - 0 .0.}5

- 0 .0 4

-0.045


0 0 liD

' ' ' 20 40 60

time, h

0

\

-0.05 L-....L. __ ..J.._ __ J...._ __ .__ _ __. __ __._ __ -L __ _.___j

20 40 60 time, h

Fig. 8.1.3a (top): An example of a vertical displacement graph. The test

was performed at 2 Nand 4 Hz for about 24 h on sample A2.

Fig. 8.1.3b (bottom): A graph of a test of about 70 h performed under the

same conditions as in the test given in Fig. 8.1. 3a. The first 24 h in this

graph compare well with Fig. 8.1.3a.


-0.02 00

- O.Oo4

-0.06

- 0 .08

-0.1

-0.1 2

-0. 1.4

-0. 16

- 0 . 18

-0.2

-0.22

-0.2 .4

-0.26

-0. 28

0

0

0

0

0

0 0 0

-0.3 L___[_ _ _L...___JL___J.__.....l.... _ _L...___j _ _J.__.....l.... _ _L...___jL____l_ _ _,___J

-0.02

-0.0"

- 0 .06

~ - 0 .0 8

i - 0.1 . ~-0 . 1 2

~-0. 14

:C-0. 16

~-0. 18 : -0.2

-0.22

-0. 2<4

-0.26

- 0.28

0

0

0

0

0

0

0

12 16 20

time, h

0

0

0 0

DO

-0.3 '---"--~~--'----'-----'--~--'----'-----'-----'--'---'----'

12

time, h

16 20 2 4

125

Fig. 8.1.4a (top): An example of a vertical displacement graph of a test

performed at 2 Nand 8 Hz for about 8 h on sample A2.

Fig. 8.1.4b (bottom): A graph of a test of about 24 h performed under the

same conditions as in the test given in Fig. 8.1.4a. Not only the amount of

vertical displacement is comparable with Fig. 8.1.4a, but also the

occurrence of an event as in Fig. 8.1.4a after 3 h and in Fig. 8.1.4b after

5 h.

126

i 0

~ u

8 ~ i

i 0

~ u

8 ~ .E


1.00 .-----------,

0.80 +

+ 0.60

+

• A

t 0.40 e

v

0.20

0 .00 L........&-..1..--1...-'--'--'----'---.L......l

·o 1 2 3 4 5 6 7 8 9

freQ.Jeney, Hz

1.00 .----------,

0.80 +

0 .60

+

Gl

0.40 0 v

0 .20

0.00 L...JL.........L--'---'--'--'--'--'--'

0 1 2 3 4 5 6 7 8 9

freQ.Jency. Hz

+ sialon A2 P=2 N

A sial on A2 P=8 N

0 sialon 86 P=2 N

v sialon 86 P"8 N

+ sialon A2 P"2 N

0 sialon 86 P"2 N

v sialon 86 P"8 N

Fig. 8.1.5a (top): The friction coefficient fc (t) as a function of

frequency after 1. 73 km sliding distance.

Fig. 8.1.5b (bottom): As in Fig. 8.1.5a, but now after 5.18 km sliding

distance.


0.10 .----------,

0 .08

+ sial on A2 P=2 N

0.06 lJ. sialon A2

P=8 N

+ 0 sialon 86

0.04 P=2 N

+ v sialon 86 lJ. P=8 N

0.02

* 0 .00 .__!_.....__.'-'-_.lil'-'-_._..__,__.

0 1 2 3 4 5 6 7 8 9

frequency, Hz

Fig. 8.1.6: The total vertical displacement divided by load in mm/N after

1. 73 km sliding distance as a function of frequency.

The relevant interdependencies of vertical displacement, friction

coefficient, load and frequency are presented in Figs. 8.1.5a, 8.1.5b and

8.1.6. The preferable method to obtain comparable data is to present the

data at a certain total sliding distance. The results for the friction

coefficients are given in Figs. 8.1.5a and 8.1.5b. The tests performed with

a load of 2 N show an increase in friction coefficient with increasing

frequency. The friction coefficients at a load of 8 N are less than the

friction coefficients at a load of 2 N at frequencies of 4 and 8 Hz. The

(vertical displacement/load) data after 1.73 k:m sliding distances versus

frequency are given in Fig. 8.1.6. The data are directly taken from the

vertical displacement graphs from which time and frequency are used to

calculate the sliding distance. The dependence of vertical displacement on


load is approximately linear as shown by the

representation of (vertical displacement/load) as a

after a certain sliding distance thus presents a

wear-enhancing influence of increasing frequency.

clusters of points. A

function of frequency

clear picture of the

The results of the four additional tests on the P-sialon are similar

to the results of the tests on the a-sialon as shown in Fig. 8.1.6.

In Figs. 8.1.7, 8.1.8, 8.1.9 and 8.1.10 pictures from

are shown. The characteristics of these surfaces depend

worn surfaces

on the test

conditions and not on the sialon type. The surface of the plate as shown in

Fig. 8.1. 7 is typical for tests at a low load and a low frequency. A

surface with 'holes', elongated particles and an occasional straight dark

band is visible. After tests at intermediate conditions for load and

frequency a wear surface as shown in Fig. 8.1.8 is usually observed with

straight dark bands alternating with bands of a 'polished' surface. At a

high load and frequency a surface is observed as in Fig. 8.1.9. The surface

is covered with a dark layer without structure. These observations are also

shown in [16]. The notion of dark bands or a dark layer on surfaces

observed with OM is an indication of areas which reflect less light than

accompanying bright areas. A heavily distorted, ploughed area next to a

relatively flat area will appear as a dark area. Samples observed with SEM

show parts with adhered and distorted material next to areas with material

which resembles the unworn part of the plate. In Fig. 8.1.10 the surface of

a Y-TZP sphere as observed with SEM is shown with various dark parts fixed

on or into the Y-TZP. These parts were identified with EDX as sialon. These

observations and the test data were used in combination with the results

from the tests on the AlONs to derive a wear mechanism for this system.


Fig. 8.1.7 (top): An example of a wear surface of an o:-sialon, A8, after a

test at 8 N and 1 Hz for 72 h. This picture was taken with normal light

microscopy. The surface is characteristic for wear by plastic deformation

under mild conditions.

Fig. 8.1.8 (bottom): An example of a wear surface of an o:-sialon, A2, after

a test at 2 N and 4 Hz for 72 h. This picture was taken with normal light

microscopy. The surface is characteristic for wear by both plastic

deformation and abrasion under intermediate conditions.


Fig. 8.1.9 (top): An example of a wear surface of an a-sialon, A9, after a

test at 8 N and 4 Hz for 24 h. This picture was taken with normal light

microscopy. The worn surface is the dark part of the figure, the bright

part is the original surface. This wear surface is characteristic for wear

by abrasion under severe conditions.

Fig. 8.1.10 (bottom): An example of a worn Y-1ZP surface after a test

against a composite, A9, at 8 N and 4 Hz for 24 h. The dark pits are

identified as sialon parts. This picture was taken with scanning electron

microscopy.


The results of the tests with the AlONs are presented with examples

that are characteristic for the conditions at which the tests were

performed. The other graphs of the tests performed under comparable

conditions were similar in shape and reasonably similar in quantitative

aspects.

Figs. 8 .1.11 a and 8 .1.11 b are examples of a vertical displacement

graph and a friction coefficient graph for a test performed at 8 N, 1 Hz,

for 72 h. All samples tested under these conditions showed relatively mild

wear characteristic for a polishing mechanism. The wave pattern with a

period of 24 h is caused by the environmental temperature changes and has

nothing to do with the wear behaviour.

The tests performed at 4 Hz all showed a transition of relatively mild

wear to severe wear with a corresponding increase in friction coefficient

as illustrated in Figs. 8.1.13a and 8.1.13b after about 18 h. This

transition for each experiments can be represented by a line in a

displacement time plot as shown in Figs. 8.1.13a and 8.1.13b. The begin

point of the line is the point at which the regime of severe wear starts,

as indicated by the arrows in Figs. 8.1.12a and 8.1.12b. The end point is

either at the end of the test at 25 h or it is at the point of total

failure of the sample, which occurred in two cases. The point at which the

transition starts shows some scatter, although it can be seen in Figs.

8.1.13a and 8.1.13b that the average time before it occurs is larger for

tests at a load of 2 N than for tests at a load of 8 N.

132 Clulpter 8. Pin-on-Plate and Pin-on-Disk measurements

-0.01 0

-0.02 \ -O.OJ

-O.Oo4

-0.05 L__j0l...._ _ _._ _ ___J2LO _ ___J~---'.0----'-----'60----'-----'80:-----'

urM,h

0 . •6 ,----,-----------------------,

o .. u

0 . 42

o.• O.J8

O.l 6 0

O.J. B 0 .32 ~

i O.J u = 0 .28

~ 0. 26 c 0 0.24 1i E o.22

0.2

0.18

0 .16

0 . ,. <--...Jll----"'-----'-- ___J----'---'----'---'----'-___J 20 •o 60 80

time, h

Fig. 8.1.1la (top): A clulracteristic example of a vertical displacement

graph of a test performed at 8 N, 1 Hz for 72 h on AION 2.

Fig. 8.1.1lb (bottom): A clulracteristic example of a friction coefficient



-0.02

-0.04

-0.06

~ -0.08

~ -0.1

i :i! -0.12

~ :.0 -0.14

~ -0.16

~ -0. 18

-0.2

-0.22

-0.24

oooaoa / 0 oo lt 00Dooooooaoooooooooooaooa

0

0

0

0

0

0 0

0

oo 0

oo 0

0

0

-0.26 '-----'--- -'--'---'-- --'--"---'---'--.J._---'- ---'----'--'---'----' 12

time, h

16 20 24

1.1 r--,- -------------------------------.

0.9

0.8

! 0.7

:l! ; 0.0

8

~ 0.5

:: 0 . 4

0.3

0 . 2

0

0

ooooaaoo o

0 o Doooaoooooooooooooo0000

0 0 0

0 0 0 0

0

p \

12

time, h

16 20

133

Fig. 8.1.12a (top): A characteristic example of a vertical displacement


Fig. 8.1.12b (bottom): A characteristic example of a friction coefficient


134 Clwpter 8. Pin-on-Plate and Pin-on-Disk measurements

1.00 .------------,

0.80

~

i 0.60

0

"' ~ '6 -0.40 "' 0

·~

020

i i

• i 1' / l i ' : I .. I 1<1 • : , I

I ; I / i / ,' i i : ,' i 'I/

• i

~I 0.00 L_-=._..___.J.4L-.........:Io----L---'

0 10 20 30 40 50

time, h

1.50 .------------,

120 lil I

I I

I

~ I

I I ;g 0.90 I

I I

] I Jl I 0 I

I /

.§ 0 .60 I

I I ./

:E I -~·

1!1~/

,t 0.30 ' j, 0 .00 '----'------'---'-----L---'

0 10 20 30 40 50

time, h

--6-- A ION 1 P=2 N

~ A ION 2 P=2 N

--G- A ION 3 P=2 N

··· • ··· A ION 1 P=8 N

---+--· A ION 2 P=8 N -- A ION 3 P=8 N

--6-- A ION 1 P=2 N

~ A ION 2 P=2 N

--a-- A ION 3 P=2 N

··· • ··· A ION 1 P=8 N

---+--· A ION 2 P=8 N -- A ION 3 P=8 N

Fig. 8.1.13a (top): A graphical representation of the discontinuities in

the vertical displacement indicated by the beginning and endpoint.

Fig. 8.1.13b (bottom): A graphical representation of the discontinuities in

the friction coefficient indicated by the beginning and endpoint.


0 .05 r-----------,

t 0.04 6 AION 1 P=2N

-o 0 AION 2 ~ P=2N 5 0.03

0 AION 3

~ • P=2 N

0 • AION 1 IU

g0.02 P=8N

'6 • AION 2 • ~ P=8N

6

~ 0.01

0 • AION 3 P=8N

6

i i

" 6 0 .00 0 2 3 4 5

frequency, Hz

025

t 0.20 [J

6 AION 1 P=2 N

-o 0 AION 2

~ P=2N 5 0.15 [J AION 3

! P=2 N

0 • AION 1 IU P=8 N g0.10 '6 • AION 2

~ P=8 N

• • AION 3

~ 0.05 P=8N

6

0 .00 I 2

0 2 3 4 5

frequency, Hz

Fig. 8.1.14a (top): A summary of the results from the vertical displacement

graphs after a sliding distance of 1. 73 km.

Fig. 8.1.14b (bottom): A summary of the results from the vertical

displacement graphs after a sliding distance of 5.18 km.


1.00 .------------,

0.80 6. A ION 1 P=2 N

6. 0 AION 2

~ • P=2 N g 0.60 0 AION 3

~ • P=2 N

1!1 6. • A ION 1 .~ 0.40 ~ 8 P=8 N

~ • AION 2 P=8 N

• AION 3

0.20 P=8 N

0 .00 c..____._ _ _.___...__..____J 0 2 3 4 5

freQ.Jeney, Hz

1.00 .------------,

0.80 6. A ION 1 P=2 N

~ 0 AION 2

P=2 N ij 0.60

0 AION 3

~ 0 P=2 N

0 i a • A ION 1 • 8 0.40 • • P=8 N

~ 6. • AION 2 £ P=8 N

• AION 3

0.20 P=8 N

0.00 '----'-----''----'---'---' 0 2 3 4 5

freQ.Jeney, Hz

Fig. 8.1.15a (top): A summary of the results from the friction coefficient

graphs after a sliding distance of 1. 73 km.

Fig. 8.1.15b (bottom): A summary of the results from the friction

coefficient graphs after a sliding distance of 5.18 km.

Chopter 8. Pin-on-Plate and Pin-on-Disk measurements 137

In Figs. 8.1.14a and 8.1.14b, the results of the vertical displacement

measurements are summarized in vertical displacement/load graphs after 1. 73

km. and 5.18 km. The clusters of points at Hz illustrates the

approximately linear dependence of vertical displacement on load.

Differences in running-in are minor. The figures also show the independence

on the AlON type. The values at 4 Hz show quite some scatter, but this is

caused by the mentioned scatter in the transition.

The results of ·the friction coefficient measurements are summarized in

Figs. 8.1.15a and 8.1.15b. The data for tests at 1 Hz are well within a

range of 0.35 to 0.55 and independent of load and AION type. The scatter

for the data at 4 Hz is again caused by the transition.

Fig. 8.1.16: The worn surface of AION I after a test at 8 N, I Hz for 72 h.

This picture was taken with OM.


Fig. 8.1.17a (top): The worn surface of AWN 3 after a test at 8 N, 1 Hz

for 72 h. This picture was taken with SEM.

Fig. 8.1.17b (bottom): The worn surface of AION 3 after a test at 8 N, 1 Hz

for 72 h. This picture was taken with SEM.


Fig. 8.1.18a (top): The worn surface of A/ON 2 after a test at 8 N, 4 Hz

for 24 h. This picture was taken with OM.

Fig. 8.1.18b (bottom): The worn surface of the Y-1ZP sphere after a test at

8 N, 4 Hz for 24 h on A/ON 2. This picture was taken with SEM.


An example of a worn surface after a test under intermediate

conditions, 8 N and 1 Hz, is shown in Fig. 8.1.16. This figure shows the

microstructure of the AION due to polishing. Visible is also some banding.

In Figs. 8.1.17a and 8.1.17b this banding is examined in detail with SEM.

These figures clearly show that the banding is formed by an alternation of

bands with high and low concentrations of grains pulled-out or parts of

grains which are pulled-out. Figs. 8.1.18a and 8.1.18b present

characteristic examples of respectively the plate and the sphere after a

test at 4 Hz. The plate is covered with a layer of what appears to be

debris without structure, and the sphere is covered with dark particles

that are identified with EDX as AlON particles.

8.1.4 Discussion

The results of the sialon tests will be discussed first again,

followed by a discussion on the results with the AlON tests.

The slight difference in wear resistance for the sphere and the plate

between tests with sialon composites and the a-sialons can possibly be

related to the higher fracture toughness and the lower hardness of the

composites, which are shown in Table 8.1.1. This is logical but not

supported by additional arguments.

The geometrical changes during a wear test in the contact between the

sphere and the plate are not entirely clear but there is some

correspondence between the models and the experimental data. In the initial

stages of wear, the surface of the sphere is flattened to a sphere with a

larger radius than the original radius as shown in Fig. 8.1.2a because the

plate as well as the sphere is worn. During continued wear, the area of

contact will increase because an increased area of contact reduces local

pressures. This means that model B is now more appropriate since the area


of contact is larger for model · B than for model A. Another relevant concept

is the local pressure at the angles between the flat part and the curved

part of the contact surface. If model B is followed, this angle will become

sharper with increasing wear and the local pressure at this angle will

increase. If the amount of wear is further increased, this corner will be

flattened and this will result in a wear surface which can be described as

a sphere according to model A. Table 8.1.3, which gives the results after

tests performed under constant, rather severe conditions, shows the

increasing importance of model A with increasing wear.

The reproducibility as shown in Figs. 8.1.3a - 8.1.4b is an indication

of the reliability of the tests. These tests were performed on one

a-sialon. The .large scatter between the data in Table 8.1.2 is abscribed to

the use of different materials. Small differences between material

properties which appear to be irrelevant for the usually measured

characteristics could well cause large differences in wear behaviour.

The friction coefficient data given in Figs. 8.1.5a and 8.1.5b

indicate a load and frequency dependence of friction coefficient. An

increase in friction coefficient with increasing frequency could be

explained by the increasing importance of abrasion at higher frequencies .

The lower friction coefficient at a load of 8 N relative to the friction

coefficient at a load of 2 N is restricted to frequencies of 4 and 8 Hz and

is not understood.

The relations between the total vertical displacement, and frequency

and load are illustrated in Fig. 8.1.6. The clustering of points in Fig.

8.1.6 is an indication of an approximately linear relation between load and

vertical displacement since values of the (vertical displacement/load) are

plotted along the vertical axis. The severe increase in the total amount of

wear with increasing frequency corresponds to the relative importance of


abrasion, which will be discussed later.

There is some correspondence between the graphs of the vertical

displacement and the friction coefficient graphs. Points in the friction

coefficient graph which are clearly higher or lower than the average

friction coefficient usually correspond with discontinuities in the graph

of the vertical displacement.

These considerations, the observations from the worn surfaces and the

various graphs are now used to derive a model for the wear mechanism. Wear

in the Y-TZP/sialon system is possible because of plastic deformation of

both materials. The difference in hardness between Y-TZP and the sialons is

not large enough to prevent wear of either one of the materials. Severe

wear by abrasion is possible due to the following process. Adhesion is

concentrated at the turning points on the plate. At these turning points,

sialon parts break loose from the plate and are incidentally attached to

the Y-TZP sphere as shown by EDX. This can be caused either by adhesion or

because the sialon part is indented into the sphere. These sialon parts

will cause abrasion of the plate in a straight line, resulting in a dark

band on the plate. Occasionally a larger sialon part can cause an

incidental large upward movement corresponding to an incidental change in

friction coefficient. A higher frequency will enhance abrasion. Many sialon

parts attached to the sphere will give a surface which consists practically

of abrading sialon particles. Frequency is thus the main variable which

determines the contribution of abrasion to the total amount of wear. These

ideas are subscribed by the results on the AIONs that will be discussed

next.

A number of observations are made on the wear system with AlONs in

order to illustrate some of the relevant characteristics.


The first part of the 'wear tests is determined by the process of

running-in. The contact between the sphere and the plate has to be

flattened to reduce the local pressure.

The coherence of the grain boundaries and material bonds of the AlONs.

appear to be very low. This is consistent with the relatively low fracture

energy of 8 J/m2 as calculated from fracture toughness and Young's modulus.

A ground surface is full of pits and holes and shows no grinding marks.

This means that the initial wear surface of the AlONs can loose many AlON

grains or parts of grains of various shape and size.

The hardness of the AlONs is about 1.3 times the hardness of the

Y-TZP. This means that the Y-TZP will not scratch or abrade the AlON plate.

The banding as shown in Figs. 8.1.17a and 8.1.17b thus has to be caused by

AlON particles localized at the sphere, either adhered or indented.

The observed phenomena and obtained data were used to develop a model.

The experiments performed under intermediate conditions, 8 N and 1 Hz,

provide information which explains most of the data. The worn surfaces

after tests at these conditions show bands with a high concentration of

grain pull-out. These bands are caused by AlON particles which are

positioned at the sphere. The areas in-between these bands support the

load. These areas are thus supposed to be the wear determining parts. The

main mechanism of AlON removal in these areas is polishing. The amount of

wear is thus simply linear with load which is confirmed by Figs. 8.1.14a

and 8.1.14b. The friction coefficients are about equal for tests at 1 Hz, 2

N and 1 Hz, 8 N. Running-in effects appear to have a minor influence in

this. respect.

At a frequency of 4 Hz the whole process is severely enhanced. More

AlON particles are accumulated to the sphere until a threshold is reached

and the contact between sphere and plate is practically an AlON-AlON


contact with regular grain pull-out in the plate and the continuous removal

of Y-TZP grains together with AlON particles from the sphere. A comparison

with the results of the Y-TZP/sialon systems under approximately the same

conditions shows that there are quite some features found in both systems.

The amount of wear at a frequency of 1 Hz is about equal, the banding is

found in both systems and a transition to severe wear was also observed in

the Y-TZP/sialon system. The wear resistance of the AlONs at frequencies

above 1 Hz is, however, much less than that of the sialons. The

correspondence in observed phenomena does provide information which is

useful in explaining features in the Y-TZP/sialon system and possibly in

other systems with Y-TZP sliding against a harder ceramic.

8.1.5. Summarizing conclusions

- The tests performed in the wear system on the same sialon type are

reproducible. The scatter in results for measurements on different

materials is large.

- Wear tests with a composite results in less wear of both the Y-TZP sphere

and the plate compared to the tests with a pure a- or P-sialon. Wear of

the AlONs is independent of the AlON type.

- The system Y-TZP/AlON shows at a frequency of 1 Hz a comparable

wear-resistance as was observed in the Y-TZP/sialon system. The

wear resistance at higher frequencies of the Y-TZP/AlON system is however

larger.

- The total vertical displacement is approximately linear with load both

for the tests on the sialons as for the AlONs.

The total vertical displacement increases severely with increasing

frequency at loads of 2 and 8 N. The friction coefficient increases with

increasing frequency at a load of 2 N for the tests on the sialons.


- Polishing is the main wear determining mechanism at a frequency of 1 Hz.

- At 8 N and 1 Hz AlON I sialon grains or parts of AION I sialon grains are

pulled out in bands but this does not influence the vertical

displacement.

- At 4 Hz a transition occurs, at 2 N on the average later than at 8 N, to

severe wear characterized by many AlON particles attached to the Y-TZP

sphere resulting in mainly AlON-AlON contacts.

8.2 PIN-ON-DISK MEASUREMENTS

8.2.1 Introduction

The Pin-on-Disk configuration is one of the most common wear testing

methods. In this test two sliding surfaces rotate relatively to each other

as shown in Fig. 8.2.1. Various variables determine the test conditions,

e.g. velocity, load, temperature, lubrication, surface preparation and the

shape of the pin-surface. A hemispherical pin has been used in most cases.

This has the advantage that the contact-surface is not disturbed by

alignment of the pin and small amounts of wear can be measured from the

remaining surface radius. A disadvantage is that the overall contact

surface changes continuously during a test.


Fig. 8.2.1: A schematic representation of the Pin-on-Disk. The pin is

rotating while the disk remains remains rigid.

The measurements performed and described in this Chapter are of

various combinations between Mg-PSZ, Y-TZP, aluminumoxide and stavax under

unlubricated conditions in a nitrogen controlled atmosphere. No tests were

performed of zirconia against zirconia.

8.2.2 Experimental

The materials used are described in Chapter 2. All the disks were

polished and the surface preparation and shape of the pins are indicated in

Table 8.2.1. The aluminumoxide and Y-TZP pins were purchased as polished

spheres with a radius of 2 mm. The stavax and Mg-PSZ pins had a cylindrical

shape with a diameter of 5 mm. Most of the stavax pins had a ground surface

with a sphere radius at the top of 4 mm and most of the Mg-PSZ pins have a

polished surface with a sphere radius of 4mm. The experimental set-up was a


configuration with a horizontai clamped disk and a rotating pin above the

disk. The normal load was applied with dead-weight loading. The loads used

were 2 and 6 N. The radius of the rotation was 20 mm, 30 mm or 40 mm. The

frequencies were chosen such that the velocity was fixed at three different

levels. The same velocity using a smaller radius thus implies a higher

frequency. Mainly velocities of 0.0565 m/s, 0.188 m/s and 0.565 m/s were

used. The test duration was usually 72 h. The atmosphere was controlled

with a nitrogen flow, ensuring that the humidity during a test was below 1

%.

The friction coefficient and vertical displacement were continuously

measured as described in Chapter 8.1. Only the calculation of the friction

coefficient was different from the Pin-on-Plate set-up. The friction forces

on the disk were measured with three force transducers positioned

tangentially at 120 ° around the disk. The data from the transducers were

used to calculate the friction force with the least-squares method in the

time-domain. This results in a value for the friction force and a standard

deviation for every sampled point. The result of a wear test was thus a

vertical displacement graph, a friction coefficient graph, the worn

surfaces and debris. The various combinations of materials used are

presented in the table with the results. Some of the materials were only

available as pins or only as disks. Measurements were done to find possible

differences in wear or friction between the zirconia ceramics, the various

disk materials or the interchanging of pin and disk material.

8.2.3 Results and discussion

The results and measurement scheme are given in Tables 8.2.1 and

8.2.2. The value for the vertical displacement, h, and the original sphere

radius, r, were used to calculate the maximum wear volume for the tests.


These values were used to calculate the wear-coefficients, K, presented in

the last column of Tables 8.2.1 8.2.2.

Figs. 8.2.1a and 8.2.lb give the friction coefficients for the tests

with Mg-PSZ for a load of respectively 2 and 6 N, and Figs. 8.2.2a and

8.2.2b give the same data for Y-TZP. It is clear from these figures that

there is a higher friction coefficient at a load of 2 N than at a load of 6

N. It is also clear that there are no differences in friction coefficient

between the various materials. The data in Figs. 8.2.1a and 8.2.2a are for

instance not much different and there is no material combination with a

significantly different friction coefficient. There is also no clear

dependence on velocity within the range of 0.0565 m/s to 0.565 m/s except

for the combination Y-TZP against stavax.

ChaP.ter 8. Pin-on-Plate and Pin-on-Disk measurements 149

pin v, m/s P, Nx, km h, mm f c

K, m2 /N

pin: Al p3

, disk: Mg-PSZ (Feld.) p

r2, p 0.0565 2 14.6 0.07 0.65 1.03E-15 r2, p 0.188 2 48.7 0.27 0.91 4.52E-15 r2, p 0.188 6 48.7 0.34 0.55 2.36E-15

pin: A12 q , disk: Mg-PSZ (Nilcra) p

r2, p 0.188 2 48.7 0.40 0.80 9.65E-15

pin: Mg-PSZ (Nilcra) disk: Al2 q , alsint, p

r4, p 0.57 4 143.6 0.45 0.32 4.27E-15 r4, p 0.0565 6 14.6 0.13 0.30 2.40E-15 r4, p 0.25 6 63.0 0.34 0.27 3.73E-15

pin: stavax, disk: Mg-PSZ (Feld.) p

r4, g 0.0565 2 14.6 0.32 0.50 4.28E-14 r4, g 0.188 2 48.7 0.60 0.60 3.99E-14 r4, g 0.565 2 146.4 0.42 0.40 6.32E-15 r4, g 0.0565 6 14.6 0.33 0.32 1.52E-14 r4, g 0.188 6 48.7 0.50 0.26 9.51E-15 r4, g 0.565 6 146.4 0.60 0.23 4.13E-15

pin: stavax, disk: Mg-PSZ (Nilcra) p

r4, g 0.0565 2 14.6 0.35 0.82 5.10E-14 r4, g 0.0565 6 14.6 0.41 0.26 2.33E-14 r4, g 0.188 6 48.7 0.66 0.32 1.77E-14

pin: Mg-PSZ (Nilcra) disk: stavax, p

f, g 0.188 2 48.7 0.65 f, g 0.565 2 146.4 0.12 0.75 r4, p 0.0565 6 14.6 0.67 0.33 6.08E-14 r4, p 0.25 6 63.0 0.36 0.29 4.18E-15

Table 8.2.1: Results of the Pin-on-Disk measurements with Mg-PSZ. The p and

g stand for the surface preparations polishing and grinding respectivi/y.

ISO Chapter 8. Pin-on-Plate and Pin-on-Disk measurements

pin v, m/s P, Nx, km b, mm f c

K, m2 /N

pin: Alp3

, disk: Y-TZP (Dyn.) p

r2, p 0.0565 2 14.6 0.10 0.62 2.12E-15 r2, p 0.565 2 3.7 0.25 5.14E-14 r2, p 0.0565 6 14.6 0.2 0.21 2.74E-15 r2, p 0.188 6 48.7 0.63 0.27 7.43E-15 r2, p 0.565 6 2.0 0.3 0.21 4.50E-14

pin: Y-TZP, disk: Al2 q , alsint, p

r2, p 0.188 6 0.42 0.15 l.llE-14

pin: stavax, disk: Y-TZP (Dyn.) p

r4, g 0.0565 2 14.6 0.26 0.50 2.84E-14 r4, g 0.0565 2 14.6 0.25 0.52 2.43E-14 r4, g 0.188 2 48.7 0.47 0.60 2.85E-14 r4, g 0.377 2 97.7 0.30 0.42 5.27E-15 r4, g 0.565 2 146.4 0.50 0.77 1.03E-14 r4, g 0.0565 6 14.6 0.30 0.32 1.26E-14 r4, g 0.188 6 16.2 0.28 0.31 9.88E-14 r4, g 0.188 6 48.7 0.26 0.32 r4, g 0.565 6 146.4 0.81 0.20 8.55E-15 r4, g 0.188 4 43.3 0.39 0.40 1.12E-14 f, p 0.188 6 48.7 0.25 0.32

pin: Y-TZP, disk: stavax, p

r2, p 0.0565 2 14.6 0.35 0.44 2.64E-14 r2, p 0.188 2 48.7 0.22 0.30 2.98E-15 r2, p 0.188 2 48.7 0.29 0.37 5.13E-15 r2, p 0.377 2 97.7 0.42 0.35 5.53E-15 r2, p 0.565 2 146.4 0.65 0.40 8.09E-15 r2, p 0.0565 6 4.88 0.32 0.21 1.95E-14 r2, p 0.188 6 16.2 0.22 0.20 2.78E-15

Table 8.2.2: results of the Pin-on-Disk measurements with Y-1ZP. The p and

g stand for the surface preparations polishing and grinding respectivily.


c ., ·c; :: ., 0 0

c: .!2 '0 E

c ., ·c; :: ., 0 0 c: 0

~

1.00

0 .90

0 .80 • "' 0

0 .70 + 0

0 .60 " 0 .50 " 0 .40

0 .30

0 .20

0 .10

0 . 00 \.___.J.__..____,__..___, 0 .00 0.200.40 0 .60 0 .801 .00

velocity, m/s

1.00

0 .90

0 .80 -

0 .70

0 .60 +

0 .50

0 .40

0 .30 ~. .. ~ " 0 .20

0 .10

0 .00 0 .00 0.200 .400 .600.801 .00

velocity, m/s

+ Al203 Mg ·PSZ, Feld

"' Al203 Mg-PSZ, Nl

" stav Mg -PSZ, Feld

• stav Mg-PSZ, Nl

0 Mg-PSZ, Nl slav

+ Al203 Mg -PSZ, Feld

+ Mg -PSZ, Ni Al203

" stav Mg-PSZ, Feld

• slav Mg -PSZ, Nl

0 Mg -PSZ, Nl slav

151

Fig. 8.2.1a (top): The friction coefficient for various combinations with

Mg-PSZ at a load of 2 N. Fig. 8.2.1b (bottom): The friction coefficient for various combinations

with Mg-PSZ at a load of 6 N.

152

c: "' ··o ~ 0 0

c: 0

~

c: "' ·u

= .. 0 0

c: .2 ;:; :E


1.00

0 .90

0 .80

0 .70

0.60 + •

0.50 t v

0.40 • v

0.30 v

0 .20

0.10

0.00 L__.__..__--L _ __.____J

0.00 0.20 0 .40 0 .60 0 .80 1.00

velocity. m/s

1.00

0.90

0 .80

0.70

0 .60

0.50

0.40

0.30 • ' +

0.20 9 v i 0

0.10

0 .00 0.00 0 .20 0.40 0 .60 0 .80 1.00

velocity. m/s

+ AI203/Y-TZP

A stav/Y-TZP

V Y-TZP/stav

+ AI203/Y-TZP

0 Y-TZP/A1203

• stav/Y-TZP

v Y-TZP/stav

Fig. 8.2.2a (top): The friction coefficient for various combinations with

Y-1ZP at a load of 2 N.

Fig. 8.2.2b (bottom): The friction coefficient for various combinations

with Y-1ZP at a load of 6 N.


The data from the vertical displacement measurements can be presented

in graphs as well, but an examination of the wear coefficients gives the

same information. The wear coefficients, and thus the amount of wear, is

not significantly different for any combination. There is also no clear

dependence on velocity and the load dependence appears to be linear

although the variation of load is only a factor 3 which is small compared

to the variation in the wear coefficients. The magnitude of the wear

coefficients is approximately equal to the wear coefficients for Mg-PSZ

given in Chapter 7 for tests under ambient conditions. These values were

also from a sliding and rotating test, but with a different pin geometry,

at a higher velocity, 1.4 m/s, and at higher loads. The difference between

Y-TZP and Mg-PSZ mentioned in Chapter 7.2 is not found in the results from

this Chapter. This could be due to the lower velocity and the lower loads.


The results of the tests are an indication that the examined zirconia

ceramics show approximately the same friction and wear at low loads and

velocities during unlubricated rotating sliding. At velocities between

0.0565 m/s and 0.565 m/s there is no dependence on velocity under the

mentioned conditions. There is a dependence of the friction coefficient on

load that remains unexplained.

References 1. K. Breder, T. Andersson and K. Schlin, Fracture strength of a

and P- SiAlON measured by biaxial and four-point bending. J. Am. Ceram. Soc. 73 (1990) 2128.

2. A. K. Mukhopadhyay, S. K. Datta and D. Chakraborty, Hardness of silicon nitride and sialon. Ceram. Int. 17 (1991) 121.

3. K. Kishi, S. Umebayashi and E. Tani, Influence of microstructure on strength and fracture toughness of P-sialon. J. Mater. Sci. 25 (1990) 2180.

4. Z. P. Wang and C. Ruiz, Characterization of contact damage of sialon in contact with waspaloy. Wear 140 (1990) 107.

154

5.

6.

7.

8.

9.

10.

11.

12.

13. 14.

15.

16.

17.

18.


Y. Nakamura and S. Hirayama, Effect of liquid lubricants on the wear of grey cast iron against Si-Al-0-N ceramic. Wear 37 (1990) 91. C. Yin-Qian, D. Xiang-Dong, W. Fu-Xing, C. Qi-Gong and Z. Zhang-Xiao, On wear mechanisms of sialon and metal in dry sliding. Wear 137 (1990) 175. J. Aucote and S. R. Foster, Performance of sialon cutting tools when machining nickel-base aerospace alloys. Mater. Sci. Techn. 2 (1986) 700. S. K. Bhattacharyya, A. Jawaid, M. H. Lewis and J. Wallbank, Wear mechanisms of Syalon ceramic tools when machining nickel-based materials. Metals Techn. 10 (1983) 482. . S. A. Horton, J . Denape, D. Broussaud, D. Dowson, F. L. Riley and N. Wallbridge, The wear behaviour of sialon and silicon carbide ceramics in sliding contact. Non-Oxide Tech. Eng. Ceram., Proc. Int. Conf. (1985) 281, ed. Hampshire, Stuart. Elsevier Appl. Sci. London. J. C. Conway, Jr., R. N. Pangborn, P. H. Cohen D. A. Love, Dry sliding wear behaviour of an Si-Al-0-N ceramic. Wear 126 (1988) 79. Z. P. Wang and C. Ruiz, Characterization of contact damage of Syalon in contact with Waspaloy. Wear 140 (1990) 107. Y. Nakamura and S. Hirayama, Effect of liquid lubricants on the wear of grey cast iron against Si-Al-0-N ceramic. Wear 137 (1990) 91. J. W. McCauley and N. D. Corbin, J. Am. Ceram. Soc. 62 (1979) 476. H. X. Willems, M. M. R. M. Hendrix, G. de With and R. Metselaar, Ace. J. Eur. Ceram. Soc. H. X. Willems, M. M. R. M. Hendrix, G. de With and R. Metselaar, presented at the 2nd ECSC, 11-14 september 1991, Augsburg, FRG. E. Kokmeijer, Sintering behaviour and properties of b ' -ShAh03N' ceramics, PhD Thesis, Eindhoven University of Technology, 1990. ASTM G-99, Standard test method for wear testing with a Pin-on-disk apparatus. D. E. Newland, Random vibrations and spectral analysis, ed. Longman, London 1975.

9. SUMMARIZING DISCUSSION

In this chapter some general remarks are given about the work

described and some possible future research is mentioned. The statements

presented in this chapter are meant as discussion, and to provide working

hypotheses for future research.

This study was aimed to obtain some understanding about the influence

of mechanical surface interactions, grinding and wear, on zirconia

ceramics. Most of the study was performed on Mg-PSZ (Nilcra). Other

zirconia ceramics investigated and compared to Mg-PSZ (Nilcra), were Mg-PSZ

(Feldmtihle), Y-TZP (Feldmtihle) and Y-TZP (Dynamic Ceramic).

The analysis on Mg-PSZ showed that the dilatation accompanying the

phase transformation caused by grinding results in an amount of residual

stress which is proportional to the amount of monoclinic zirconia. The

scale of the grain size, about 60 f.Jm, compared to the depth at which the

influence of the grinding process is present, about 20 to 30 f.Jm, indicates

that Mg-PSZ can be regarded as a continuum as far as the residual stress is

concerned.

Y-TZP on the other hand, has a grain size of about 1 f.Jm. The depth at

which the influence of grinding is noticed, is estimated at 10 f.Jm. This

means that the relatively large amounts of grain boundaries are likely to

play a significant role. There are other phenomena important for Y-TZP,

like degradation, superplasticity and the re-transformation, that influence

the mechanical behaviour of the material.

Brittle failure of a material is usually not described with continuum

mechanics, but with fracture mechanics on the scale of microfracture. The

fracture behaviour of Mg-PSZ is partly described in Chapter 5.2. The most

likely location for failure is indicated and the indirect influence of

residual stress on fracture stress is illustrated. The precise failure mode

and the dynamic aspects of fracture remain unclear. It is not known whether

156 Chapter 9. Summarizing discussion

there is any slow crack growth, or whether there is any preferred direction

for crack opening. These aspects are interesting future research areas.

The results of the tribology tests show that there is some correlation

between the wear conditions and the phase transformation for Mg-PSZ but not

for Y-TZP. The remaining strength of worn samples of Mg-PSZ can be related

directly to the influence of the test conditions on the phase

transformation and on the residual stress. The behaviour of Y-TZP is less

understood and is again a possible subject for future research.

A different kind of study was performed with conventional Pin-on-Plate

and Pin-on-Disk instruments. One of the sliding materials was always a

zirconia ceramic, but the behaviour of the counter materials was examined

in more detail. The counter materials for the tests with the Pin-on-Plate

were various sialons and AlONs with different phase contents and

compositions, respectively. The results showed that the differences in

phase content and compositions have only a minor influence on the

wear-behaviour of the systems. The other system parameters, like velocity,

load and counter material, testing geometry, are altogether more important.

Wear mechanisms for the wear systems were derived, indicating the

importance of grain pull-out, the influence of velocity and the importance

of abrasive particles from the hard material attached to the soft material.

The tests with the Pin-on-Disk were performed with various

combinations between Al20

3, stavax, Mg-PSZ and Y-TZP, unlubricated at room

temperature for about 70 h. Testing at elevated temperatures could be

interesting. A more analytical, chemical approach to the investigation of

worn surfaces after for instance a single pass is also interesting. The

concept of adhesion, important for most systems with ceramics, could be

treated from a chemical point of view in a more fundamental way.

Tribochemistry is a field of research where much work remains to be done.

The amounts of wear clearly showed that under the tested conditions,

Mg-PSZ showed superior wear behaviour relative to the wear behaviour of

Chapter 9. Summarizing discussion 157

Y-TZP. The absolute amounts of wear that were measured after the tests with

Mg-PSZ also showed that the wear coefficients of 10·" to 10"16 m2/N are

still quite high as compared with wear systems with metals.

The first part of this thesis was about the transformation related .

characteristics caused by a reproducible surface treatment, grinding. The

second part was about tribological characteristics. During a wear test a

mechanical surface interaction occurs which is more complicated than the

surface interaction during grinding. The results of two types of

measurements, strength and surface roughness measurements, can be related

to the results of other surface treatments.

A comparison between the results obtained from wear tests and the

results from other surface treatments, can only be made if major

simplifications are accepted. The overall results from the tests are used

in the following considerations and no attention is given to the detailed

results. Adhesion is present in all wear systems examined. This means that

during a wear test, the , possible contacts are zirconia-zirconia,

zirconia-counter material and counter material-counter material. There is

thus always the situation of a hard material sliding against a soft

material. Each material can be abraded by itself.

Wear testing with water as a lubricant can be compared with polishing.

Unlubricated testing at normal loads of more than about 10 N can be

compared with grinding. Unlubricated sliding at loads of less than about 10

N normal load can result in significant contributions both from polishing

and from abrasion. The considered difference between polishing and

grinding, or abrasion, is scratching without fracture and scratching with

fracture. Scratching without fracture means that either the normal load is

less than the critical load, or that the contact is partly or completely

lubricated. Abrasion means that a material is not only plastically

deformed, but also cracked. Two types of fracture on a worn surface were

observed, delamination and grain pull-out. Delamination is caused by

158 Chapter 9. Summarizing discussion

lateral cracks and it is the major mechanism of material removal for the

relatively 'coarse' grained, strong and high-toughness Mg-PSZ. Grain

pull-out is observed for the 'fine' grained Y-TZP and sialon, and for the

'coarse' grained, weak and brittle AlON.

The lubricated tests on Mg-PSZ resulted in surface roughness and

strength . values, Chapter 7, that can be related to the values for a

polished surface. Worn surfaces at loads of 2 and 8 N show alternating

bands of abrasion and polishing, Chapter 8 indicating the transitional

character of these two mechanisms. They both occur under intermediate

conditions on one sample.

The surface roughness and strength values of Mg-PSZ after unlubricated

sliding against stavax, can be related to grinding. Surface roughness is

determined by the mechanism of material removal. Strength is influenced by

residual stress. There is thus some correlation between lubricated as well

as unlubricated wear at low loads, with polishing, and between unlubricated

wear at high loads with grinding. These types of comparisons between

grinding, polishing and wear could be extended by future research.

Summarizing it can be stated that there are significant differences in

surface mechanical behaviour between Mg-PSZ and Y-TZP. The residual stress

profile and strength behaviour of worn samples are different. During wear

testing at low loads, adhesion and plastic deformation play a major role.

At high loads, fracture becomes more important. Band formation is observed

for some wear systems with Y-TZP.

Items for future research include subcritical crack growth, preferred

crack-opening directions during microfracture, a more detailed insight in

the behaviour of Y-TZP, tribochemistry and more knowledge about the

influence of the counter material.

159

LISTING OF SYMBOLS OFfEN USED

symb~l meaning unit

E Young's modulus GPa

v Poisson's ratio

p Density g/cm3

HV Vickers hardness GPa

B Bulk modulus GPa

:B Transformation bulk modulus GPa

a3pb Three-point-bend strength MPa

K Ic

Fracture toughness MPa.m112

a Critical transformation stress GPa c

p Hydrostatic pressure Pa

at Residual stress GPa

a andY Flow stress GPa y

y Initial flow stress GPa 0

eli Strain

eT li

Dilatational transformation strain

e Dilatational strain PI'

f Fraction monoclinic zirconia

v Percentage monoclinic zirconia % m

p Normal load N

p Critical load N

K Wear coefficient m2/N

R Surface roughness, C. L.A. pm

f Friction coefficient

SUMMARY

This thesis is about the influence of mechanical surface interactions

on zirconia ceramics. The studied surface interactions were grinding and

sliding wear. The main investigated properties of the zirconias were the

phase content, residual stress and strength, all three as a function of

depth in the material.

Analysis of Magnesium Partially Stabilized Zirconia (Mg-PSZ) was done

after grinding with a 046 diamond grinding wheel. The analysis resulted in

profiles for the amount of monoclinic zirconia and the amount of residual

stress. These two profiles were identical in shape; a high value in the

first two micrometers, a sharp decrease beneath these first two

micrometers, and a low value until a depth of about 22 micrometers. This

shape similarity was combined with a flow law from literature to a

stress-strain curve showing a transformation modulus of 44 GPa, indicating

a significant amount of work hardening.

Additional grinding tests were done on Mg-PSZ with two grinding wheels

containing grains that were different in size. Measured were the phase

content profile, the strength as a function of depth and the residual

stress at the surface. The results showed a correlation between the diamond

grain size, the amount of residual stress, the transformation and the

strength. A larger diamond grain, assumed to exert higher forces on the

material due to a blunter shape, resulted in more transformation, a higher

residual stress and a lower strength compared to a smaller diamond grain.

This correlation was combined with additional experimental information to

derive a model for the fracture behaviour of the material which is based on

fracture originating from a concentration of tensile stresses located

Summary 161

beneath the residual stress layer near inhomogeneities like grain

boundaries. The same type of analysis was performed on Yttria Tetragonal

Zirconia Polycristalline (Y-TZP). The results showed that the

transformation and residual stress have a subsurface maximum at a depth

from about two to four micrometers. The amount of transformation and the

residual stress at the surface are much lower as compared to Mg-PSZ. The

strength was independent of three different diamond grain sizes, and

grinding with the smallest diamond grain resulted in the least amount of

transformation. These results illustrate the differences between Mg-PSZ and

Y-TZP. More research is required to understand the behaviour of Y-TZP.

Wear tests in rotational sliding were done to relate strength of

zirconia ceramics to wear conditions. The results from tests between Mg-PSZ

and stavax showed a correlation between wear conditions, ambient or

lubricated, and strength, as well as a correlation between normal load and

strength. These correlations were explained with the concept of residual

stress developed due to the , surface interactions during sliding. The

observations and data were used to derive a wear mechanism, indicating the

importance of delamination for Mg-PSZ as a wear mechanism. The same type of

tests were performed on Y-TZP. The results showed that the wear resistance

of Y-TZP was far less then the wear resistance of Mg-PSZ under the test

conditions. The behaviour of Y-TZP was explained with the concept of

degradation, the spontaneous transformation of the material under the

influence of elevated temperatures and the presence of moisture or water.

Pin-on-Plate measurements were performed between Y-TZP and various

sialon and AlON plates. This resulted in the modelling of the wear

behaviour of such systems.

SAMBNV A TIING

Dit proefscbrift bebandelt de invloed van mecbaniscbe oppervlakte

bewerkingen op de eigenscbappen van zirconia keramieken. Deze oppervlakte

bewerkingen waren slijpen en slijten. De bestudeerde eigenscbappen waren

voomamelijk de fase samenstelling, de restspanning en de sterkte, alledrie

als functie van diepte in bet materiaal.

Magnesium Partieel Gestabiliseerd Zirconia (Mg-PSZ) is onderzocbt na

slijpen met een D46 diamant scbijf. Deze analyse leidde tot diepte

profielen voor de fase samenstelling en de restspanning, die dezelfde vorm

hebben. Ben hoge waarde in de eerste twee micrometer, daama een sterke

afname, en vervolgens een lage waarde tot een diepte van ongeveer 22

micrometer. Deze gelijkbeid in vorm is gecombineerd met een vloeiwet uit de

literatuur tot een spannings-rek relatie, waaruit een transformatie

coefficient van 44 GPa volgt. Dit houdt .in dat Mg-PSZ een 'verstevigend'

materiaal is.

Verdere slijptesten op Mg-PSZ zijn gedaan met twee slijpschijven die

voomamelijk in diamant korrelgrootte verscbilden. Het profiel voor de fase

samenstelling, de sterkte als functie van diepte en de restspanning aan bet

oppervlak zijn gemeten. De resultaten lieten zien dat er een correlatie

bestaat tussen de diamant korrelgrootte, de restspanning, de transformatie

en de sterkte. Ben grotere diamant korrel, geassocieerd met grotere

kracbten door een bottere vorm, geeft meer transformatie, meer

restspanning, en een ·lagere sterkte dan een kleinere korrel. Deze verbanden

zijn gecombineerd met verdere experimentele resultaten tot een model voor

bet bezwijken van dit materiaal. Dit model is gebaseerd op een concentratie

van trekspanning gelocaliseerd onder de restspanningslaag nabij

beterogeniteiten zoals korrelgrenzen. Benzelfde soort analyse is ook

Samenvatting 163

uitgevoerd op Yttrium Tetragonaal Zirconia Polykristallijn (Y-TZP). De

resultaten tonen aan dat de fase samenstelling en de restspanning een

maximum bebben dat op ongeveer twee tot vier micrometer onder bet oppervlak

ligt. De transformatie en de restspanning aan bet oppervlak zijn veel

minder. De sterkte bleek onafhankelijk te zijn van drie verscbillende

diamant korrelgroottes en slijpen met de kleinste diamant korrel gaf de

minste transformatie. Deze resultaten tonen de verscbillen tussen Mg-PSZ en

Y-TZP en geven aan dat meer onderzoek aan Y-TZP nodig is.

Slijtage testen, · roterend glijden, zijn uitgevoerd om sterkte van

zirconia keramiek te relateren aan slijtcondities. De resultaten van de

testen tussen Mg-PSZ en stavax geven aan dat er een correlatie bestaat

tussen slijtomstandigheden, in Iucht of met water als smeermiddel, en

sterkte na slijten. Tevens is er een correlatie tussen de normaalbelasting

en sterkte. Deze correlaties kunnen met bet restspanningsprincipe verklaard

worden. De waamemingen en meetresultaten zijn gebruikt om een slijtage

mecbanisme af te leiden waaruit bet belang van delaminatie voor Mg-PSZ

blijkt. Dezelfde soort testen zijn ook uitgevoerd met Y-TZP. Hieruit bleek

dat de slijtage bestendigheid van Y-TZP veel minder is dan die van Mg-PSZ

onder de gebruikte test condities. Het gedrag van Y-TZP is verklaard met

bet verscbijnsel degradatie, de spontane transformatie van bet materiaal

onder de invloed van bogere temperaturen en de aanwezigbeid van vocbt.

Verder zijn er Pin-on-Plate metingen verricbt tussen Y~TZP en

verscbeidene sialon en AlON plaatjes. Dit resulteerde in bet modeleren van

bet slijtage gedrag van deze systemen.

164

ACKNOWLEDGEMENT

First of all I would like to express my gratitude to prof. dr. G. de

With who gave me the opportunity to work on the presented study within the

Philips Research Laboratories as well as within the Centre for Technical

Ceramics (CTK), and I would like to thank him for his consistent coaching.

Most of the work was done at the Philips Research Laboratories within the

group 'Inorganic Materials and Processing', which was a truly stimulative

environment. Many people from Philips helped me with advice, experiments,

measurements and comments, too many to name them all. However, I would like

to mention in this context W. Horden, 1. van den Berg, W. Gijsbers, C.

Geenen, 1. van Oijen, P. Rommers, C. Alting, P. Bouten, M. Buijs, A. Broese

van Groenou, H. Veenvliet, A. Corbijn and W. Mesman. I would like to thank

the Commission for the Innovative Research Program Technical Ceramics

(lOP-TK) of the Ministry of Economic Affairs in the Netherlands who partly

supported this work, from september 1990 to september 1992 (IOP-TK grant

90A211). The Pin-on-Plate and Pin-on-Disk instruments at the CTK became

operational thanks to, amongst others, W. de Maijer, H. de Laat, E.

Ridderhof and L. . Dortmans. The latter also performed the Finite Element

Analysis. Several X-ray measurements were done by H. de 1onge Baas and M.

Hendrix. I also would like to express my gratitude to the people who

commented on this thesis, prof. dr. G. de With, prof. ir. A.W.1. de Gee,

prof. dr. ir. M. 1. W. Schouten, prof. dr. R. Metselaar and prof. dr. ir.

H. Verweij.

During my stay at Eindhoven these four years I've really enjoyed being

a member of the Eindhovense Studenten Atletiek Vereniging Asterix. Finally,

I would like to thank my family, my parents and three sisters, Ankie,

Evelien and Sandra, for their continuous support, attention and care.

165

CURRICULUM VITAE

Paul van den Berg was born on the lOth of March, 1964, at Haarlem. He

started at the 'Eerste Chrystelijk Lyceum' in Haarlem in 1976 and passed

his 'Atheneum B' examinations in 1982. He studied geology at the University

of Amsterdam from 1982 untill november 1986 when he received his

'doctoraal' in Structural Geology. After this he served. in the army as a

sergeant for the artillery. In september 1988 he began with the 'korte

onderzoekers opleiding Chemische Technologic' at the Eindhoven University

of Technology. Most of the practical work was done at the Philips Research

Laboratories. This 'second phase' of his study was succesfully ended in

september 1990. The research was continued for two years, again mainly at

Philips, and it resulted in this thesis.

STELLING EN

1. Het is onjuist de wet van Hooke toe te passen om een bovengrens te

bepalen voor de restspanning t.g.v. de transformatie in zirconia.

D. J. Green, F. F. Lange and M. R. James, J. Am. Ceram. Soc. 66 (1983)

623.

2. Een diepte van 250 JJm voor de transformatie zone vanaf een breukvlak

van Mg-PSZ zoals gegeven door Steinbrech et al. lijkt eerder bet gevolg

van latere bewerkingen zoals 'renotching' en temperatuursverhoging, in

plaats van bet gevolg van breuk.

R. W. Steinbrech, E. lnghels and A. H. Heuer, J. Am. Ceram. Soc. 73,

(1990) 2016.

3. De vertaai1ng van zirconia zoals beschreven door de formule:

LIK = 0.22Efe (1-v)v'h c pp

is gebaseerd op de onjuiste veronderstellingen van een superkritische

transformatie en de transformatie van al bet tetragonale zirconia over

de transformatiediepte h.

A. G. Evans, J. Am. Ceram. Soc. 73 (1990) 187.

4. Eindige Elementen modellering van de deformatie van Mg-PSZ rondom een

5.

scheur geeft wel qualitatieve informatie over de hoeveelheid

getransformeerd zirconia als functie van diepte, maar onjuiste

quantitatieve informatie.

H. Okada, T. Tamura, N. Ramakrishnan, S. N. Atluri and J . S. Epstein,

Acta Metall. Mater. 40 (1992) 1421.

Het feit dat er vijftien gelijkwaardige en verschillende formules

bekend zijn om de K Jc

te be pal en uit de radiele scheurlengte bij een

indentatie betekent dat deze methode als absolute meting nog niet

bruikbaar is.

C. B. Ponton and R. D. Rawlings, Mater. Sci. Techn. 5 (1989) 865.

6. Het is niet juist om mediane scbeuren in aluminiumoxide, die mogelijk

ontstaan tengevolge van slijpen, te gebruiken om resultaten van

sterktemetingen te verklaren zonder deze scbeuren experimenteel a an te

tonen.

Y. Motsuto, T. Ogosowaro, S. Kimura, s. Soto and E. Yasuda. J . Cerom.

Soc. Japan Int. Ed. 99 (1991) 371.

7. Slijtage-onderzoek aan materiaalsoorten waar geen economiscb baalbare

toepassing voor bestaat is gerecbtvaardigd wanneer deze materialen nog

volop in ontwikkeling zijn.

8. Het is een illusie om te denken dat in de huidige situatie de continue

verslechtering van bet milieu gestopt kan worden; pas wanneer bet

milieu dezelfde prioriteiten krijgt als politieke en militaire belangen

bestaat er een kans op verbetering.

9. Er is geen belangrijk Oost-West contact in de Sierra de Almagro zoals

beschreven in bet proefscbrift van 0. J. Simon.

0. J. Simon, proefschrift 'Geologie von de Sierra de Almogro',

Universiteit von Amsterdam (1962).

10. Er zijn slechts weinig mensen die bet belang van de uitspraak 'een

niet-onafhankelijk gecontroleerde meting is geen meting'• inzien.

• De owi von Diessen AOC Breda.

11. Over een slordige bonderd miljoen jaar, een relatief korte tijd op de

geologiscbe tijdsscbaal, zal de periode waarin de menselijke bescbaving

tot op beden beeft bestaan gekarakteriseerd worden door een laagje met

een gemiddelde dikte van minder dan 1 mm.

zirconia ceramics and mechanical surface interactions · 6. tribology and ceramics 76 76 78 78 78...

Documents