Nano Materials

NANOMATERIALS
Edited by
dr hab. in. Lucyna Jaworska

Krakw 2010

Reviewers: Prof. dr hab. in. Stanisaw Wierzbiski
dr hab. in. Andrzej Dziado, prof Pk

Editorial board: dr hab. in. Lucyna Jaworska

Project of cover: dr in. Pawe Kurtyka

Text arrangement: dr in. Pawe Figiel

ISBN 978-83-912887-9-5

The monograph was supported by the Polish Ministry of
Science and Higher Education as the Project
DPN/N111/BIAORU/2009

Copyright by The Institute of Advanced Manufacturing
Technology, Krakow 2010

Printed and cover from supplied materials by:
Zakad Usug Poligraficznych, Krakow, J. Lea 114 Street
Edition 100 copies

The main aim of this Nanomaterials monograph is to bring
science and application together in nanoscale and
nanostructured materials, with emphasis on modeling of new
phenomena, synthesis, processing, characterization and
application of materials containing nano-sized particles or
nanostructures enabling novel properties and/or functions.
Nanotechnology may be thought of as an extension of
traditional disciplines. In addition, traditional disciplines can
be re-interpreted as specific applications of nanotechnology.
A basic understanding of nanoscale phenomena and
mechanisms in nanomaterials is still lacking. Nanomaterials
are evolving and constantly appearing in new applications.
Each chapter of Nanomaterials presents the results
of research conducted by institutions in Poland and abroad.
Readers can become familiar with technologies and research
methods which constitute original solutions. The chapters of
this monograph emphasize original results relating to
experimental, theoretical and computational applications of
nanomaterials.

Editorial Board
dr hab in Lucyna Jaworska

4
CONTENTS

MATHEMATICAL MODEL OF ELECTROKINETIC
PHENOMENA IN TWODIMENSIONAL CHANNELS 5
Vladimir V. Mityushev Natalia Rylko

MECHANICAL PROPERTIES OF Si
3
N
4
/SiC COMPOSITES
WITH VARIOUS ADDITIONS 21
Piotr. Klimczyk, Lucyna Jaworska, Vladimir Urbanovich

FRACTURE TOUGHNESS BEHAVIOUR OF TICNANO
REINFORCED ALUMINA-ZIRCONIA TOOL COMPOSITES
AT ELEVATED TEMPERATURES 45
Magdalena Szutkowska, Barbara Smuk, Marek Boniecki

NANOCOMPOSITES IN THE ZrO
2
-Al
2
O
3
-WC SYSTEM 55
Zbigniew Pdzich, Wojciech Maziarz

TEM INVESTIGATION OF MATRIX SAFFIL
TM
FIBER
INTERFACES IN ALUMINIUM ALLOYS BASED
COMPOSITES 63
Jerzy Morgiel, Jacek Kaczmar, Magorzata Pomorska, Krzysztof Naplocha

TRIBOLOGICAL PROPERTIES OF SUPERSONIC
SPRAYED NANOSTRUCTURED WC12CO COATINGS 74
Wojciech rawski

AMORPHOUS Ti-Si-C THIN FILM DEPOSITED ON AISI
316L IN LOW TEMPERATURE 84
Agnieszka Twardowska

IONIC METHODS OF COATINGS FORMATION FOR A
SPECIAL APPLICATIONS 93
Bogusaw Rajchel

Mathematical model of electrokinetic phenomena in twodimensional channels

5
MATHEMATICAL MODEL OF
ELECTROKINETIC PHENOMENA IN
TWODIMENSIONAL CHANNELS

VLADIMIR V. MITYUSHEV
1
NATALIA RYLKO
1

ABSTRACT

Electroosmotic flows for a binary dilute electrolyte are theoretically studied in two
dimensional channels by expanding the solution into a series in terms of a
dimensionless parameter related to the DebyeHuckel length. Analytical
approximate formulae for the coupling coefficient are deduced.
1 Introduction
The study of electroosmotic flows through porous or dispersed
media is important from a fundamental standpoint and for industrial
applications. The external electric field E and the pressure gradient
p generate a macroscopic current density I and a Darcy seepage
velocity u which for small E and p are given by the linear
formulae:

(1.1)
where is the gradient operator, K permeability, viscosity, is
the electrical conductivity, is the electroosmotic coupling
coefficient. Theoretical and experimental study of the electrokinetic
phenomena is given in [1][5]. In the present paper we pay attention
to analytical formulas for the coupling coefficient in the two-
dimensional case.
Consider a two-dimensional curvilinear channel bounded by the
smooth dimension walls:

1
Department of Mathematics, Krakow Pedagogical University, ul. Podchorazych 2,
30-084 Krakow, Poland, E-mail: vmityu@yahoo.com d


6

(1.2)
The values of S
(x) are dimensionless and of order 1, i.e., S
(x)~1.
For simplicity, sometimes a symmetric channel (S
(x)=S(x)) is
considered.
Introduce the dimension parameter and the dimensionless
parameter . Introduce two different dimensionless
length scales:
1) Distorted coordinates

, (1.3)
Then (1.2) becomes

, - < x < (1.4)
where
2) Proportional coordinates
,
(1.5)
Then (0.2) for symmetric channels becomes

, - < x < (1.6)
where S
*
(x
*
) = S(x). Further, the first dimensionless scale is used up
to Sec. 4.
The electrokinetic phenomena in the curvilinear 2D channel S
-
(x) <
z < S
+
(x) is governed by the following PDE [1]

(1.7)

(1.8)

(1.9)
with the boundary conditions

7

(1.10)

(1.11)

(1.12)

(1.13)
Here, u = (u;w), is the outward normal derivative to the curves
(1.2). Instead of the first equation (1.9) we consider equation

(1.14)
where . It is assumed that the functions
S
(x) are twice continuously differentiable.

2 Dimensionless equations
Introduce the following dimensionless variables and parameters

(2.1)

where the most important in this paper the dimensionless parameter
is introduced via the relation

(2.2)
Then

,
(2.3)
Using (1.3)(2.3) we rewrite the problem (1.9) in the dimensionless
form. For shortness, primes are omitted

(2.4)

8

(2.5)

(2.6)
Hereafter, the subscripts x and z are used for the corresponding
partial derivatives. The dimensionless parameter
i
are introduced as

(i=1 ,2) (2.7)
Up to the end of this section dimension and dimensionless variables
are again distinguished by primes and the boundary conditions
(1.10)(1.13) are rewritten in the dimensionless form. Consider
equation of the walls in the dimensionless form (1.6). Then (1.10)
becomes

(2.8)
u(x, S
(x)) = 0, w(x, S
(x)) = 0 (2.9)
g(, z) g 9 (, z) 2g
0
(2.10)
i.e., the constant g
0
has the form g
0
= . Consider the boundary
condition (0.13)

(i = 1, 2) (2.11)
where the normal vector n is determined by formula
. Then (1.11) can be written in the form
on

(i = 1, 2) (2.12)
Using (1.6) and (2.3) we have . Then (2.12)
becomes

9

on z = S
(x) (i = 1, 2) (2.13)
3 Cascade on
2

We solve equations (2.4)(2.6) with the boundary conditions (3.2)
(4.8) by expansion on
2
. Therefore, we actually assume that the
parameter is small, i.e., the ratio of the boundary layer
1

to the length of the periodicity cell L is small. This enables us to
find the unknown functions in the form

(3.1)
Further in this paper, the zero order approximations from (3.1) are
explicitly written. Omitting primes further we write the boundary
condition (1.8) in the form

(3.2)
Substituting the first equation (3.1) into (2.4) and (3.2) and selecting
the coefficients on the same powers of
2
we obtain a cascade of the
boundary value problem for .
The zeroth approximation of the problem for has the form

(3.3)
Solution to the problem (3.3) has the form

(3.4)
For symmetric channels when S
+
(x) = S(x) and S
(x) = S(x), (3.4)

takes the form

(3.5)
We have at the jstep ,

.
If is known, is easily obtained by the formula

10

(3.6)
where
(j)
(x; z) is a partial solution to the ordinary differential
equation (3.6) on z. The functions are found from the
boundary conditions (3.6). For instance, the first order
approximation for symmetric channels has the form

(3.7)
It is possible to write an exact formula for for general channels,
but it is too long.
4 Cascade for u
First, we introduce the auxiliary functions

(4.1)
Adding and subtracting equations (2.5) we obtain equations

(4.2)

(4.3)

(4.4)
where the dimensionless parameters
i
are introduced via
i

determined by (2.7)

(4.5)
Omitting primes further we write the boundary conditions (2.9),
(2.10), (2.13) in the form

(4.6)

(4.7)

on z = S
(x) (i = 1, 2) (4.8)

11
The problem (4.2)(4.4), (4.6)(.8) in the zeroth approximation has
the form

(4.9)

(4.10)

(4.11)

(4.12)

(4.13)

(4.14)

(4.15)

on z = S
(x) (i = 1, 2) (4.16)
Here, the function is known and has the form (3.4). One can
consider (4.9)(4.10) as a system of linear ordinary differential
equations on the variable z of first order with respect to the functions
with zero boundary conditions (4.16). It is easily seen that this
system has only trivial solutions. This implies that the functions
i depend only on x. Then it follows from (4.12) that , hence
g
(0)
also depends only on x. Using (4.15) we obtain

(4.17)
Substitution of (4.17) into (4.11) yields

(4.18)
where is an undetermined function on x. Integrating (4.18)
on z and using the relation (3.3) we obtain

(4.19)
where C
j(x)
are undetermined functions (constant on z). Using the
boundary conditions (4.14) for u
(0)
we obtain

12

(4.20)
Integrate equation (4.13) on z. Applying (4.19) we have

(4.21)
where C
3
(x) is undetermined function (constant on z),

Using the boundary conditions (4.14) for w
(0)
we have a formula for
C
3
(x) and a linear differential equation of first order for .
Further constructive calculations are possible. However, the ultimate
explicit form of C
3
(x) and 1x is too huge. For symmetric
channels when S
+
(x) = S(x) and S
(x) = S(x), these formulas take a

simpler form. The following formulas are valid for symmetric
channels, hence, C3(x) 0. The function satisfies equation

(4.22)
Its solution has the form

(4.23)
where C is an arbitrary constant. Hereafter, for shortness we write S
instead of S(x). The function has to be periodic with the
period 2. The latter condition yields

(4.24)
Therefore, the velocity in symmetric channels is properly defined in
the zeroth approximation by formulas (4.19) and (4.21), where

13

(4.25)
has the form (3.23)(3.24). Ultimately,

(3.26)

(3.27)
5 Verification of approximations
In the present section different dimensionless length scales noted by
primes and stars are used. Consider the approximate solution (3.5)
constructed by equation . This equation is considered as
an approximation of equation which is valid
for sufficiently small . Therefore in order to check directly the
approximation (3.5), the value should be evaluated.
However, it is not correct to estimate the ratio in the distorted
variables (x; z). So, the value

(5.1)
is estimated in the coordinates (x*, z*) introduced by (1.5). We have

(5.2)
Then =
1
+
2
, where


14

(5.3)
Therefore, (3.5) (or equivalently (5.2)) provides acceptable
approximation if
(
1
< 1, (
2
< 1 (5.4)
Inequalities (5.4) mean that the amplitude of the walls expressed by
the first derivative and their curvature expressed by the second
derivative are sufficiently small.
Consider an example to clarify the conditions (5.4). Let the channel
is restricted by the dimension walls

(5.5)
with small positive dimensionless parameter and given
dimensionless coefficient which describes deviations of the walls.
Equations (4.5) can be written in the dimensionless form

(5.6)
Using (5.3) calculate up to O(
2
)

(5.7)
Consider the cases:
a) b = 1 (or equivalently b
= 1). Then

(5.8)
b) b >> 1 (or equivalently b
>> 1). Then

(5.9)
The relations (5.8)(5.9) demonstrate geometric restrictions. For
instance, if b ~
1
, (5.8) implies that has to be of order 1.

15
If b >>
1
, (5.9) implies that despite b >> 1, it can be compensated
by small or small .
The proper expansion of by
2
yields the proper order for u,
hence the proper order for the coupling coefficient discussed in the
next section.
6 Coupling coefficient
The dimension coupling coefficient is calculated by the integral
[1][4]

(6.1)
where < f(x; z) > means the average of the function f(x; z) over the
periodicity cell. Following [2][4] introduce the dimensionless
coefficient , where

(6.2)
The averaged of the dimensionless values can be calculated by
formula

(6.3)
Here, is the dimensionless area of the periodicity cell

(6.4)
Applying the calculated zeroth approximations (3.4) and (4.19) for
and u for symmetric channels we have in the zeroth
approximation

16

(6.5)
where C has the form (4.24). In the case of the straight channel
S(x) = b formula (6.5) becomes the wellknown formula [1][4]

(6.6)
Consider the permeability of the channel .
Following [1][4] introduce the dimensionless permeability K =
2
K, which can be written in the form

(6.7)
Using (4.26) and (4.24) we calculate the zero approximation for the
symmetric channels< u
(0)
>= . Dividing (6.2) by (6.7) we obtain

(6.8)
The following relation were discussed in [5]

(6.9)
where the dimension length scale and the dimensionless length
scale =
1
are introduced by the following three ways:

(6.10)
Here, is the local electric field in the channel filled by an
electrolyte of constant conductivity and submitted to a macroscopic
potential difference along the xaxis.

17

(6.11)
where F is the formation factor. F
1
is equal to the macroscopic
conductivity of the channel filled by electrolyte of the normalized
unit conductivity, K is the permeability.

(6.12)
where has the form (6.4), |S is the length of the curve (1.6)

(6.13)
Consider the symmetric channel restricted by the walls

(6.14)
We are interested in the coupling coefficient and the permeability to
O(
2
). It is convenient to find them in the form
,

, where and are the dimensionless coupling
coefficient and the dimensionless permeability for the straight
channel , the dimensionless constants
1
and k
1
have to
be found.
First, we calculate and . For the straight channel, equations
(3.5) and (4.19) become , .
Then ; . Using (6.8) we get the
relation

(6.15)
which coincides with (6.9) for = b. Recall that the normalized
case = 1, g
0
= 1 is considered.
The constant k
1
were exactly calculated in [6] (formula (82) from [6]
contains a slight error which is corrected below)

18

(6.16)
Equation (6.5) for the channel (6.14) becomes up to O(
4
)

(6.17)
Therefore,

(6.18)
Then (6.8) up to O(
4
) become

(6.19)

Figure 1: Results presented as functions on b for = 0:1 < b < 10. Data are for:
calculated by (6.8) dashed,

with
1
calculated by (6.10) solid,
with
2
calculated by (6.11) dotted, with
3

calculated by (6.12) thin line coincided to (6.8).

19
Let denote the coefficient on
2
of . It follows from (6.19) that

(6.20)
The corresponding coefficients on
2
of the function

when takes the values (5.10)(5.12) have the form

(6.21)

(6.22)

(6.23)
The value from (5.20) does not coincide symbolically with the
values given by (5.21)(5.23). However, numerically one can see in
the figure good agreement except the parameter
2
given by (5.11).

References

[1] A. E. Malevich, V. V. Mityushev and P. M. Adler, Electrokinetic
phenomena in wavy channels. Journal of Colloid Interface Sci. 2010,
345(1), 72-87.
[2] D.Coelho, M.Shapiro, J.E.Thovert, P.M.Adler, Electroosmotic Phenomena
in Porous Media, Journal of Colloid and Interface Science, 1996, 181,
169190.
[3] P.M.Adler, D.Coelho, J.-F.Thovert, M.Shapiro, Electrokinetic Phenomena
in Porous Media and Around Aggregates, in Surfaces of nanoparticles and
porous materials, ed. J.A. Schwarz, C.I. Contescu, Surfactant Science
Series, (M. Dekker), 1999, 78, 211257
[4] P.M.Adler, Macroscopic Electroosmotic Coupling Coefficient in Random
Porous Media Mathematical Geology, 2001, 33(1), 6393.

20
[5] A. Gupta , D. Coelho , P.M. Adler, Universal electroosmotic formulae for
porous media, Journalof Colloid Interface Sci. 2008 319(2), 549-554.
[6] A. E. Malevich, V. V. Mityushev and P. M. Adler, Stokes flow through a
channel with wavy walls, Acta Mechanica, 2006, 182, 3-4, 151-182.

Mechanical properties of Si
3
N
4
/SiC composites with various additions

21
MECHANICAL PROPERTIES OF SI
3
N
4
/SIC
COMPOSITES WITH VARIOUS ADDITIONS

PIOTR KLIMCZYK
1
, LUCYNA JAWORSKA
1,2
,
VLADIMIR URBANOVICH
3

Abstract

Nano or submicro 30vol.%.Si
3
N
4
/70vol.%.SiC composites, with additions of Ti,
TiB
2
and cBN, were obtained by High Pressure - High Temperature (HPHT)
sintering. Density, Youngs modulus, hardness, fracture toughness and coefficient
of friction were measured. Microstructural (SEM) investigations and phase analysis
(XRD) were also conducted for selected samples. Composites obtained from
30vol.% Si
3
N
4
/70vol.% SiC submicro powders with the addition of 30 vol.% cBN
micro powder were characterized by the best mechanical properties.
1. Introduction
Silicon nitride has a favorable combination of properties that
includes high strength over a broad temperature range, high
hardness, moderate thermal conductivity, a low coefficient of
thermal expansion, moderately high elastic modulus, and relatively
high fracture toughness for a ceramic. Silicon nitride ceramics have
reached large-scale production for cutting tools, bearings,
turbocharger rotors and a variety of custom wear parts. Silicon
carbide has many of the same applications as silicon nitride. Most
silicon carbide materials have very high hardness (harder than
alumina and silicon nitride) and thus have superior wear resistance
[1, 2, 3, 4].
A major disadvantage of SiC ceramic materials is their low fracture
toughness, which usually does not exceed about 3.5 MPam
1/2
[5, 6].

1
Mat. Eng. Dept., Inst. of Advanced Manuf. Techn., 37a Wroclawska St., 30-011
Krakow, Poland.
2 Dep. Technology, Pedagogical University, 2 Podchorazych St., 30-084 Krakow,
Poland.
3
Scientific-Practical Materials Research Center of NAS of Belarus 19, P. Brovka
St., Minsk 220072, Belarus.
3
N
4

22
In previous studies, attempts have been made to obtain an Si
3
N
4
-
SiC composite characterized by both high hardness and high
fracture toughness [7, 8]. The purpose of the presented work is to
study the effect of the addition of third-phase particles selected from
metals (Ti) or ceramics (TiB
2
, cBN) to an Si
3
N
4
-70vol%SiC system
on the mechanical properties of Si
3
N
4
-SiC composites obtained by
the HPHT method. The main goal was to improve the fracture
toughness of the investigated materials.
When a particulate second phase is introduced into a brittle matrix,
there are several toughening mechanisms that may be operative.
E.g. metallic particles are capable of plastic deformation and thus
absorption of energy and bridging of a growing crack, resulting in
increased strengthening. (Fig. 1a) [9].
The addition of titanium particles into an Si
3
N
4
SiC system may
cause chemical reaction, and favorable, plastic, ceramic phase type
titanium silicon nitride may appear. On the other hand, hard ceramic
particles can introduce a favorable stress state, which can cause a
toughening effect by crack deflection and crack bifurcation (Fig.
1b). Hard particles also improve the hardness and other mechanical
properties of the composite as a whole.
a) b)

Fig. 1. Strengthening mechanisms in ceramic matrix composites with dispersed
metallic (a) and ceramic (b) particles.
2. Experimental
2.1. Materials preparation
Powders used for the preparation of mixtures are listed in
Table 1.
3
N
4

23
Table 1. Mixture composition.
Powder
Granulation
[m]
Manufacturer Description
Si
3
N
4
(1) 0.6 H.C. Starck, Germany M11-grade (alpha>90%)
Si
3
N
4
(2) 1 5 AEE, US (alpha>85%)
Si
3
N
4
(08) 0.1 0.8 Goodfellow, UK (alpha)
SiC(2) 0.1 1 Goodfellow, UK (alpha)
SiC(3) - AGH, Poland specific surface 24 m
2
/g
TiH
2
(1) <44 Fluka, Switzerland
TiB
2
(4) 2.5 3.5 H.C. Starck, Germany F-grade
BN(M010) 0 0.01
Element6, South
Africa
Micron+ABN (cubic)
BN(M36) 3 6
Element6, South
Africa
Micron+ABN (cubic)

The following mixtures were prepared by mixing the appropriate
powders in an isopropanol environment using a Fritsch Pulverisette
6 planetary mill.
A. 30Si
3
N
4
/70SiC composite:
A1) 30vol.% Si
3
N
4
(1) + 70 vol.% SiC(2) (submicro:
Starck + Goodfellow)
A2) 30 vol.% Si
3
N
4
(08) + 70 vol.% SiC(2) (submicro:
Goodfellow)
A3) 30 vol.% Si
3
N
4
(2) + 70 vol.% SiC(3) (micro: AEE +
AGH)
A4) 30 vol.% Si
3
N
4
(2) + 70 vol.% SiC(2) (micro: AEE +
submicro: Goodfellow)
B. 30 Si
3
N
4
/70SiC composite + Ti:
3
N
4

24
B1) 28 vol.% Si
3
N
4
(1) + 64 vol.% SiC(2) + 8 vol.% Ti -
from TiH
2
(1) (submicro: Starck + Goodfellow +
micro: Fluka)
C. 30 Si
3
N
4
/70SiC composite + TiB
2
:
C1) 28 vol.% Si
3
N
4
(1) + 64 vol.% SiC(2) + 8 vol.%
TiB
2
(4) (submicro: Starck + Goodfellow + micro:
Starck)
C2) 21 vol.% Si
3
N
4
(1)+49 vol.% SiC(2)+30 vol.%
TiB
2
(4) (submicro: Starck + Goodfellow + micro:
Starck)
D. 30 Si
3
N
4
/70SiC composite + cBN:
D1) 28 vol.% Si
3
N
4
(1) + 64 vol.% SiC(2) + 8 vol.%
cBN(M36) (submicro: Starck + Goodfellow +
micro: E6)
D2) 28 vol.% Si
3
N
4
(08) + 64 vol.% SiC(2) + 8 vol.%
cBN(M010) (submicro: Goodfellow + nano: E6)
D3) 21 vol.% Si
3
N
4
(1) + 49 vol.% SiC(2) + 30 vol.%
cBN(M36) (submicro: Starck + Goodfellow +
micro: E6)
After drying, the mixtures were preliminarily compressed into
pellets of diameter 15 mm and height 5 mm under pressure of
~200 MPa. The green bodies with the addition of TiH
2
were
additionally annealed in a vacuum at a temperature of 800
o
C for 1h
in order to remove the hydrogen and obtain pure metallic titanium.
The composites were obtained at high pressure (6 GPa) in the
temperature range of 790 2030 C using a Bridgman-type toroidal
apparatus (Fig. 2). The sintering temperatures were established
experimentally for each composite to obtain crack-free samples with
the highest values of density and mechanical properties. Duration of
the sintering process was 60 s.
3
N
4

25

Fig. 2. Sintering process in a Bridgman-type HPHT system. Quasi-hydrostatic
compression of the preliminary consolidated powders (sample - 1) is achieved as a
result of plastic deformation of the gasket material (2) between anvils (3); electrical
heating is provided by a high-power transformer (4) and graphite resistive heater
(5).
The sintered compacts were subsequently ground to remove remains
of graphite after the technological process of sintering and to obtain
the required quality and surface parallelism for physical and
mechanical studies.
2.2. Methods of investigation
Densities of the sintered samples were measured by the hydrostatic
method. The uncertainty of the measurements was below 0.02
g/cm
3
, which gave a relative error value of below 0.5 % (excluding
measurements of small pieces of broken samples, where error was
up to 0.1 g/cm
3
, due to their insufficient volume and mass).
Youngs modulus of the samples obtained by HPHT sintering was
measured based on the transmission velocity of ultrasonic waves
through the sample, using a Panametrics Epoch III ultrasonic flaw
detector. Calculations were carried out according to the following
formula:
3
N
4

26

2 2
2 2
2
4 3
=
T L
T L
T
C C
C C
C E (1)
where: E - Youngs modulus, C
L
- velocity of the longitudinal wave,
C
T
- velocity of the transversal wave, - density of the material.
The velocities of transverse and longitudinal waves were determined
as a ratio of sample thickness and relevant transition time. The
accuracy of calculated Youngs modulus (Eq. 1) was estimated to be
below 2 %.
Hardness of selected samples was determined by the Vickers
method using a digital Vickers Hardness Tester (FUTURE-TECH
FV-700). Five hardness measurements, with indentation loads of
2.94, 9.81 and 98.1 N, were carried out for each sample. Standard
deviations of HV values were relatively high but usually no more
than 5 % of the average values.
Indentation fracture toughness was calculated from the length of
cracks which developed in a Vickers indentation test (with
indentation load - 98.1 N) using Niiharas equation (Eq. 2):

2
3
5
2
129 0
\
|
|
|
\
|
a
c
. =
E
H
a H
K
IC
(2)
where: K
IC
- critical stress intensity factor, - constrain factor, H -
Vickers hardness, E - Youngs modulus, a - half of indent diagonal,
c - length of crack.
Microstructural observations were carried out on the densified
materials using a JEOL JXA-50A Scanning electron Microscope
equipped with back scattering electron (BSE) imaging. In ball-on-
disc tests, the coefficients of friction for contact with an Si
3
N
4
ball
were determined using a CETR UMT-2MT (USA) universal
mechanical tester. In the ball-on-disc method, sliding contact is
produced by pushing a ball specimen onto a rotating disc specimen
under a constant load. Tests were carried out without lubricant. The
loading mechanism applied a controlled load F
n
to the ball holder
3
N
4

27
(Fig. 3) and the friction force was measured continuously during the
test using an extensometer. For each test, a new ball was used or the
ball was rotated such that a new surface was in contact with the disc.
After mounting of the ball and sample, materials were washed in
ethyl alcohol and dried.

Fig. 3. Material pair for the ball-
on-disc method: 1 Si
3
N
4
ball; 2
sample (disc)

The size of the disc-shaped samples was ~13.5 x 3.8 mm; the
surface of the discs flat and parallel to within 0.02 mm; and the
roughness of the test surface not more than 0.1 m R
a
. The test
samples were ground using diamond wheels and polished using
diamond slurries. The following test conditions were established:
ball diameter: 2 mm,
applied load: 4 N,
sliding speed: 0.1 m/s,
diameter of the sliding circle: 2 - 5 mm,
sliding distance: 100 m,
calculated duration of the test: 1000 s.
Friction coefficient was calculated from the following equation:

Fn
F
f
= (3)
where: F
f
is the measured friction force, and F
n
is the applied normal
force.
3
N
4

28
4. Results and discussion
Density and Youngs modulus of various 30 vol.% Si
3
N
4
/70 vol.%
SiC composites sintered at different temperatures with and without
additions of Ti, TiB
2
and cBN phases are presented in Figs. 4-15 and
in Table 2.

2,90
2,92
2,94
2,96
2,98
3,00
3,02
3,04
3,06
3,08
3,10
3,12
3,14
3,16
3,18
3,20
3,22
1
4
0
0
1
5
0
0
1
6
0
0
1
7
0
0
1
8
0
0
1
9
0
0
2
0
0
0
2
1
0
0
Sintering temperature,
o
C
D
e
n
s
i
t
y
,

g
/
c
c
m
Theoretical 30Si3N4/70SiC
A1) 30Si3N4(1)+70SiC(2)
(submicro:Starck+Goodfellow)
A2) 30Si3N4(08)+70SiC(2)
(submicro:Goodfellow)
A3) 30Si3N4(2)+70SiC(3)
(micro:AEE+AGH)
A4) 30Si3N4(2)+70SiC(2)
(micro:AEE+submicro:Goodfe
llow)

Fig. 4. Density of various kinds of 30 Si
3
N
4
/70 SiC composites sintered at different
temperatures. Dark symbols samples without cracks; white symbols samples
with cracks; white symbols lying on temperature axis broken samples.
3
N
4

29
200
220
240
260
280
300
320
340
360
380
400
420
1
4
0
0
1
5
0
0
1
6
0
0
1
7
0
0
1
8
0
0
1
9
0
0
2
0
0
0
2
1
0
0
o
C
Y
o
u
n
g

m
o
d
u
l
u
s
,

G
P
a
A1) 30Si3N4(1)+70SiC(2)
(submicro:Starck+Goodfellow)
A2) 30Si3N4(08)+70SiC(2)
(submicro:Goodfellow)
A3) 30Si3N4(2)+70SiC(3)
(micro:AEE+AGH)
A4) 30Si3N4(2)+70SiC(2)
(micro:AEE+submicro:Goodfe
llow)

Fig. 5. Youngs modulus of various kinds of 30 Si
3
N
4
/70 SiC composites sintered at
different temperatures. Dark symbols samples without cracks; white symbols
samples with cracks; white symbols lying on temperature axis broken samples.

3,10
3,12
3,14
3,16
3,18
3,20
3,22
3,24
3,26
3,28
3,30
3,32
3,34
7
0
0
8
0
0
9
0
0
1
0
0
0
1
1
0
0
1
2
0
0
1
3
0
0
1
4
0
0
1
5
0
0
1
6
0
0
1
7
0
0
1
8
0
0
1
9
0
0
2
0
0
0
o
C
D
e
n
s
i
t
y
,

g
/
c
c
m
Theoretical 30Si3N4/70SiC +
8Ti
B1)
28Si3N4(1)+64SiC(2)+8Ti(1)
(submicro:Starck+Goodfellow
+micro:Fluka)

Fig. 6. Density of Si
3
N
4
/70 SiC + 8 vol.% Ti composites sintered at different
3
N
4

30
100
120
140
160
180
200
220
240
260
280
300
320
340
360
380
400
7
0
0
8
0
0
9
0
0
1
0
0
0
1
1
0
0
1
2
0
0
1
3
0
0
1
4
0
0
1
5
0
0
1
6
0
0
1
7
0
0
1
8
0
0
1
9
0
0
2
0
0
0
o
C
Y
o
u
n
g

m
o
d
u
l
u
s
,

G
P
a
8Ti
B1)
28Si3N4(1)+64SiC(2)+8Ti(1)
+micro:Fluka)

Fig. 7. Youngs modulus of Si
3
N
4
/70SiC + 8 vol.% Ti composites sintered at

3,16
3,18
3,20
3,22
3,24
3,26
3,28
3,30
3,32
1
6
0
0
1
7
0
0
1
8
0
0
1
9
0
0
o
C
D
e
n
s
i
t
y
,

g
/
c
c
m
8TiB2
C1)
28Si3N4(1)+64SiC(2)+8TiB2(4)
(submicro:Starck+Goodfellow+
micro:Starck)

3
N
4
/70SiC + 8 vol.% TiB
2
composites sintered at different
3
N
4

31
340
350
360
370
380
390
400
410
420
430
1
6
0
0
1
7
0
0
1
8
0
0
1
9
0
0
o
C
Y
o
u
n
g

m
o
d
u
l
u
s
,

G
P
a
8TiB2
C1)
28Si3N4(1)+64SiC(2)+8TiB2(
4)
+micro:Starck)

3
N
4
2
composites sintered at

3,40
3,42
3,44
3,46
3,48
3,50
3,52
3,54
3,56
3,58
3,60
3,62
1
6
0
0
1
7
0
0
1
8
0
0
1
9
0
0
2
0
0
0
2
1
0
0
2
2
0
0
o
C
D
e
n
s
i
t
y
,

g
/
c
c
m
30TiB2
C2)
21Si3N4(1)+49SiC(2)+30TiB
2(4)
+micro:Starck)

3
N
4
2
composites sintered at different
3
N
4

32
320
340
360
380
400
420
440
460
1
6
0
0
1
7
0
0
1
8
0
0
1
9
0
0
2
0
0
0
2
1
0
0
2
2
0
0
o
C
Y
o
u
n
g

m
o
d
u
l
u
s
,

G
P
a
+ 30TiB2
C2)
21Si3N4(1)+49SiC(2)+30Ti
B2(4)
(submicro:Starck+Goodfello
w+micro:Starck)

3
N
4
/SiC + 30 vol.% TiB
2
composites sintered at

2,90
2,95
3,00
3,05
3,10
3,15
3,20
3,25
1
6
0
0
1
7
0
0
1
8
0
0
1
9
0
0
2
0
0
0
o
C
D
e
n
s
i
t
y
,

g
/
c
c
m
8BN
D1)
28Si3N4(1)+64SiC(2)+8BN(M
36)
+micro:E6)
D2)
28Si3N4(08)+64SiC(2)+8BN(
M010)
(submicro:Goodfellow+nano:E
6)

3
N
4
/70 SiC + 8 vol.% cBN composites sintered at different
temperaturesDark symbols samples without cracks; white symbols samples with
cracks; white symbols lying on temperature axis broken samples.
3
N
4

33
320
340
360
380
400
420
440
460
1
6
0
0
1
7
0
0
1
8
0
0
1
9
0
0
2
0
0
0
o
C
Y
o
u
n
g

m
o
d
u
l
u
s
,

G
P
a
+ 8BN
D1)
28Si3N4(1)+64SiC(2)+8BN(
M36)
w+micro:E6)
D2)
28Si3N4(08)+64SiC(2)+8BN
(M010)
(submicro:Goodfellow+nano:
E6)

3
N
4
/70SiC + 8 vol.% cBN composites sintered at
samples with cracks; white symbols lying on temperature axis broken sample.

3,25
3,30
3,35
3,40
3,45
3,50
3,55
3,60
3,65
3,70
3,75
3,80
3,85
3,90
3,95
1
6
0
0
1
7
0
0
1
8
0
0
1
9
0
0
2
0
0
0
o
C
D
e
n
s
i
t
y
,

g
/
c
c
m
30BN
D3)
21Si3N4(1)+49SiC(2)+30BN(
M36)
+micro:E6)

3
N
4
/70SiC + 30 vol.% cBN composites sintered at different
with cracks; white symbols lying on temperature axis broken sample.
3
N
4

34
420
440
460
480
500
520
540
560
580
1
6
0
0
1
7
0
0
1
8
0
0
1
9
0
0
2
0
0
0
o
C
Y
o
u
n
g

m
o
d
u
l
u
s
,

G
P
a
+ 30BN
D3)
21Si3N4(1)+49SiC(2)+30BN
(M36)
w+micro:E6)

3
N
4
/70SiC + 30 vol.% cBN composites sintered at
samples with cracks; white symbols lying on temperature axis broken sample.
3
N
4

35
Table 2 /part 1. Physical-mechanical properties of the best samples selected from
different modifications of Si
3
N
4
/SiC composite.

3
N
4

36
Table 2 /part 2. Physical-mechanical properties of the best samples selected from
different modifications of Si
3
N
4
/SiC composites.

3
N
4

37
Among the composites sintered without additional phases, the
highest degree of densification and best mechanical properties were
demonstrated by composite A1, obtained from submicron powders
30 vol.% Si
3
N
4
(1) + 70 vol.% SiC(2) (submicro: Starck +
Goodfellow) (Figs. 4, 5 and Table 2). This composite was selected
for modification by the addition of third phase particles.
Modification of the 30 vol.% Si
3
N
4
/70 vol.% SiC composite by the
addition of Ti was not successful.
Samples from series B1, with the addition of 8% Ti introduced in
the form of TiH
2
, sintered at low temperatures, were characterized
by a very low Young's modulus, whilst all the samples sintered at
temperatures above ~1200 C were cracked.
A decrease in density was observed with increasing sintering
temperature, whilst Young's modulus showed an upward trend (Figs.
6 and 7).
Composites with the addition of TiB
2
were characterized by a high
degree of densification, a high Youngs modulus and improved K
IC

as compared to the unmodified composite. No improvement in
hardness was observed (Table 2).
In the case of material C1, with the addition of 8% TiB
2
, there
appears to be some increase in density and Youngs modulus with
increasing temperature, (Figs. 8-9).
Composite C2, with the addition of 30 vol.% TiB
2
, shows an
increase in density with sintering temperature up to a maximum
value, and then its stabilization. Further increase of the sintering
temperature results in cracking of the samples. This is typical
behavior in the sintering of advanced ceramics by the HPHT method
(Fig. 10). The Young's modulus of this material decreases with
increasing sintering temperature, which is not typical (Fig. 11).
Composites from series D3, modified by the addition of 30 vol.%
cBN micropowder, showed the highest degree of densification,
equal to 118% of theoretical density - EDS analysis showed a high
content of tungsten carbide and zirconium dioxide from the vessel
and grinding media used to prepare the mixtures - the best
mechanical properties (Table 2) and the lowest percentage of
cracked samples. (Figs. 14 and 15).
3
N
4

38
SEM microstructures of 30Si
3
N
4
/70SiC with and without the
addition of TiB
2
and cBN are presented in Fig. 16.

A1) 30 vol.% Si
3
N
4
(1) + 70SiC(2) (submicro: Starck + Goodfellow).

A3) 30 vol.% Si
3
N
4
(2) + 70 vol.% SiC(3) (micro: AEE + AGH).

3
N
4

39

C1) 28 vol.% Si
3
N
4
(1) + 64 vol.% SiC(2) + 8 vol.% TiB
2
(4)
(submicro: Starck+Goodfellow + micro: Starck).

C2) 21 vol.% Si
3
N
4
(1)+49 vol.% SiC(2)+30 vol.% TiB
2
(4)
(submicro: Starck + Goodfellow + micro: Starck.

3
N
4

40

D1) 28 vol.% Si
3
N
4
(1) + 64 vol.% SiC(2) + 8 vol.% cBN(M36)
(submicro: Starck + Goodfellow + micro: E6).

D3) 21 vol.% Si
3
N
4
(1) + 49 vol.% SiC(2) + 30 vol.% cBN(M36)
(submicro: Starck + Goodfellow + micro: E6).
Fig. 16. SEM microstructures of selected 30Si
3
N
4
/70SiC composites with and
without the addition of TiB
2
and cBN.
The microstructures of investigated samples are compact and dense,
with the ingredients uniformly distributed in the volume of
composite. This demonstrates successful blending, using a planetary
mill; EDS analysis, however, showed a high content of tungsten
carbide and zirconium dioxide from the vessel and grinding media
3
N
4

41
used to prepare the mixtures (white areas visible in the
microstructures (Fig. 16).
Three exemplary curves representing measurements of coefficient of
friction for sample C1: 30 vol.% Si
3
N
4
/70 vol.% SiC + 8 vol.%
TiB
2
, and the mean curve, are presented in Fig. 17. A comparison of
mean curves of friction coefficient for various kinds of 30 vol.%
Si
3
N
4
/70 vol.% SiC composites, with and without the addition of
cBN and TiB
2
, is presented in Fig. 18.
0 100 200 300 400 500 600 700 800 900 1000
0,0
0,1
0,2
0,3
0,4
0,5
0,6
0,7
0,8
0,9
1,0
C1 30Si3N4/70SiC (submicro) + 8TiB2_Test1
C1 30Si3N4/70SiC (submicro) + 8TiB2_Mean

[
-
]
Time [s]

Fig. 17. Coefficient of friction of 30 vol.% Si
3
N
4
/70 vol.% SiC + vol.% 8TiB
2

composite.

3
N
4

42
0 100 200 300 400 500 600 700 800 900 1000
0,0
0,1
0,2
0,3
0,4
0,5
0,6
0,7
0,8
0,9
1,0
A1 30Si3N4/70SiC (submicro)
A4 30Si3N4/70SiC (micro/submicro)
C1 30Si3N4/70SiC (submicro) + 8TiB2
C2 30Si3N4/70SiC (submicro) + 30TiB2
D3 30Si3N4/70SiC (submicro) + 30cBN

[
-
]
Time [s]

Fig. 18. Coefficient of friction of selected 30 vol.% Si
3
N
4
/70 vol.% SiC composites
with and without the addition of TiB
2
and cBN (mean curves).
High coefficients of friction generate thermal stress, which is
detrimental to the wear behavior of materials. Hard ceramic bodies
possessing high fracture toughness and low coefficients of friction
used in mechanical systems that involve high loads, velocities and
temperatures, will reduce costs and be less harmful to the
environment. Obtaining coefficients of friction below 0.2 is unlikely
since, under non-lubricated conditions, current dual-phase ceramics
typically have coefficients of friction of 0.50.8 [10].
Unmodified composites had the highest coefficients of friction.
Average values of friction coefficient for materials A1 and A4 were
0.60 and 0.56 respectively. Composites modified by the addition of
a TiB
2
phase were characterized by intermediate values of friction
coefficient. Average values of friction coefficient for materials C1
and C2 were 0.48 and 0.46 respectively. The composite with the
addition of 30% cBN was characterized by the lowest average
coefficient of friction, at only 0.36.

3
N
4

43
Summary
Various modifications of 30Si
3
N
4
/70SiC (volume ratio) composites,
with additions of Ti, TiB
2
and cBN, obtained by HPHT sintering,
were investigated.
Composites modified by the addition of Ti show deterioration of
properties in comparison with the unmodified composites.
Composites modified by the addition of TiB
2
show only a partial
improvement of properties in comparison with the unmodified
composites.
Composites modified by the addition of 30% cBN micropowder are
characterized by the best combination of Youngs modulus,
hardness, fracture toughness and coefficient of friction. Such
properties predispose 30Si
3
N
4
/70SiC + 30% cBN composites to
various advanced engineering applications. Wear tests and cutting
tests (intended in the future) will show the range of applications of
this material in machining.
Acknowledgements
This study was carried out within the framework of the project
funded by the Polish Ministry of Science and Higher Education
(Project number: DPN/N111/BIALORUS/2009).

REFERENCES

[1] Peng H., Spark Plasma Sintering of Si
3
N
4
-Based Ceramics, Doctoral
Dissertation, Department of Inorganic Chemistry, Stockholm University
2004.
[2] Pierson H.O., Handbook of refractory carbides and nitrides, Noyes
Publications, Westwood, New Jersey 1996.
[3] Richerson D.W., Advanced ceramic materials, in J. K. Wessel (ed.),
Handbook of advanced materials, John Wiley & Sons, Inc., Hoboken, New
Jersey 2004.
[4] Eblagon F., Ehrle B., Graule T., Kuebler J., Development of silicon
nitride/silicon carbide composites for wood-cutting tools, Journal of the
European Ceramic Society, 2007, 27, s.419428.
[5] Lee S.M., Kim T.W., Lim H.J., Kim C., Kim Y.W., Lee K.S., Mechanical
Properties and Contact Damages of Nanostructured Silicon Carbide
3
N
4

44
Ceramics, Journal of the Ceramic Society of Japan, 2007, 15 (5), s. 304-
309.
[6] Suyama S., Kameda T., Itoh Y., Development of high-strength reaction-
sintered silicon carbide, Diamond and Related Materials, 2003, 12, s. 1201-
1204.
[7] Klimczyk P., Urbanovich V.S., Micro-, submicro- and nano-Si
3
N
4
SiC
composites sintered by the HPHT method, Archives of Materials Science
and Engineering, 2009, 39/2, s. 92-96.
[8] Klimczyk P., Mechanical properties of Si
3
N
4
-SiC composites sintered by
the HPHT method, Proceedings of CIMTEC 2010 - 12th International
Ceramics Congress in press.
[9] Yeomans J.A., Ductile particle ceramic matrix composites - Scientific
curiosities or engineering materials?, Journal of the European Ceramic
Society, 2008, 28, s.15431550.
[10] Kerkwijk B., Garcia M., Zyl W.E., Winnubst L., Mulder E.J., Schipper D.
J., Verweij H., Friction behaviour of solid oxide lubricants as second phase
in -Al
2
O
3
and stabilised ZrO
2
composites, Wear 2004, 256, s.182189.

Fracture toughness behaviour of TiC
nano
reinforced alumina-zirconia tool composites at
elevated temperatures

45
FRACTURE TOUGHNESS BEHAVIOUR OF
TIC
NANO
REINFORCED ALUMINA-ZIRCONIA
TOOL COMPOSITES AT ELEVATED
TEMPERATURES

MAGDALENA SZUTKOWSKA
1
, BARBARA SMUK
1
,
MAREK BONIECKI
2

Abstract

Recent advances in high-speed cutting materials have focused on reinforcing
alumina with different carbides, oxides and nitrides in order to improve hardness,
fracture toughness and wear resistance. The potential use of this type of material for
cutting tool applications has not yet been determined. The present study reports
selected properties obtained by reinforcing Al
2
O
3
10 mass% ZrO
2
(partially
stabilized with Y
2
O
3
-Y5) composite with TiC in amount of 5 mass%. Specimens
were prepared based on submicro- and nano-scale trade powders. Vickers hardness
(H
V
) and fracture toughness (K
IC
) at room and elevated temperatures characteristic
for tool work were evaluated. Wear resistance (Vn), a very important property for
tool ceramics, was determined by speed of mass loss. Microstructural observations
of the specimens were carried out using a scanning electron microscopy (SEM).
The addition of TiC nanopowders does not result in a significant improvement in
fracture toughness. Preliminary industrial tests confirm the higher cutting
performance of TiC
submicro
reinforced alumina-zirconia tool composites in
comparison to the same composites reinforced with TiC
nano
.
1. Introduction
Engineered ceramics are used in thermal and structural applications
requiring high temperature resistance, high hardness and chemical
inertness. Applications that exploit the thermal structural properties
of ceramics commonly include cutting tool inserts, wear resistant
components, ballistic armour, heat exchangers, burner tubes,

1
Mat. Eng. Dept., Inst. of Advanced Manuf. Techn., 37a Wroclawska St., 30-011
Krakow, Poland.
2
Dep. of Cer. and Joints., Inst. of Elec. Mat. Techn., 133 Wolczynska St., 01-919
Warsaw, Poland
3
N
4

46
prosthetics, dental implants, heat engine components and thermal
barrier coatings. The main characteristics limiting these ceramics in
the above and other engineering applications are their lack of plastic
behaviour at room temperature and their low tolerance to flaws, i.e.
low fracture toughness, which lead to catastrophic failure.
Reinforcing ceramics with particles, whiskers, platelets, continuous
fibres and discontinuous fibres significantly improves their strength,
toughness and apparent ductility [1]. In ceramic composites,
reinforcement usually increases strength indirectly by increasing the
toughness of the matrix. Composite design can also be used to tailor
other important properties such as high-temperature strength and
thermal shock resistance, wear resistance, friction, thermal and
electrical conductivity and thermoelastic properties. The need for
high-temperature, oxidation-resistant materials has driven research
into oxide-oxide composites. By an appropriate adjustment of the
concentration of its components, alumina matrix ceramic composites
(AMCC) can offer a high level of strength properties, toughness and
hardness [2]. Previous studies have indicated that the toughening
improvement by the addition of a dispersed second phase is based
on several mechanisms, such as transformation toughening, crack
deflection, thermal residual stresses, microcrack formation,
debonding, pull-out, and crack bridging [3]. In an Al
2
O
3
matrix, t-
ZrO
2
grains undergo the tm transformation (stress induced phase
transformation) and microcracks form around pretransformed m-
ZrO
2
grains (microcrack formation), these are the major toughening
mechanisms in Al
2
O
3
-ZrO
2
composites. The mechanism of this
process based on the polymorphic martensitic transformation of the
t-ZrO
2

tetragonal phase into an m-ZrO
2
monoclinic phase during
cooling from sintering temperature to room temperature, - allows an
increase in the strength and/or fracture toughness of Al
2
O
3
-ZrO
2
composites [4]. Optimization of the stability of tetragonal zirconia
can lead to improved mechanical properties in zirconia-toughened
alumina composites (ZTA). It is suggested that the highest
toughness can be obtained without a reduction in hardness if the
additives are adjusted to optimize the mixture of zirconia phases of
different sizes (nano and submicro scales). On the other hand, pure
titanium carbide (TiC) has many attractive properties, such as high
3
N
4

47
hardness, low density, and relatively high thermal and electrical
conductivity [5]. These ceramic composites, to their great
thermodynamic stability, high hardness, and compression strength
of the TiC additive, have a better cutting properties in comparison to
oxide ceramics, and can be used for precision machining of hard
work-pieces. The advantages of such tools are the possibility of high
duty working, obtaining of surfaces with very low roughness (in
many cases grinding is eliminated) and the possibility of ecological
dry cutting without the use of cutting fluids. A comparison of
cutting properties for different tool ceramics is presented in Fig.1
[6].

Fig.1. Comparison of cutting properties for different tool ceramics [6]

Thus the main aim of this study is to obtain new titanium carbide-
reinforced alumina-based composite tool ceramics that combine
excellent toughness with increased hardness. These properties result
from the addition to zirconia-toughened alumina ceramics of TiC
particles in submicro and nano scale.
2. Experimental Procedure
Alumina, zirconia partially stabilized with yttria, non-stabilized
zirconia and titanium carbides were used as raw materials to
manufacture the titanium carbide-reinforced alumina-based
composites. A commercial Alcoa alumina powder (containing
85.0% -phase, of 99.8% purity) type A16SG with a submicron
3
N
4

48
particle size of below 0.5 m and a nano particle size of 40 nm, with
average size of agglomerates of 150 nm produced by Inframat
Advanced Materials, USA were used to prepare the tested
composites. Experiments were carried out on alumina-based
composites with the addition of: 10 mass% ZrO
2
with modification
of the zirconia phase (partially stabilized with 5% mass Y
2
O
3
PYT05.0 (ZY5) produced by Unitec Ceramics, England, with an

average particle size of 0.9-1.1 m; ZrO
2
stabilized with 3% mol.
Y
2
O
3
produced by Inframat Advanced Materials, USA (marked as
YSZ
nano
) and a monoclinic phase of zirconia in submicron

(m-ZrO
2
)
and nano scale (m-ZrO
2nano
) produced by Fluka, Germany. A
commercial TiC powder with different particle sizes as a standard
TiC (grade ST120 produced by H.C. Starck, Germany) with an
average particle size of 1.0- 4.0 m and TiC
nano
with agglomerates
of : 130 nm and 200 nm produced by Aldrich Sigma, Germany.
Sintering additives, such as MgO
nano
(0.3 mass%), were introduced
to inhibit grain growth.
The initial compositions of compounds selected for testing are
presented in Table 1.

Table 1. Composition of selected compounds
Compound composition [mass%]
Al
2
O
3
ZY5 ZrO
2
(m)
TiC
Compound
[m] [nm] [m] [nm] [m] [nm] [m]
<130
nm
<200
nm
B 85.0 5.0 5.0 5.0
B1 42.5 42.5 5.0 5.0 5.0
B2 42.5 42.5 5.0 5.0 5.0

Components were mixed for approximately thirty hours in alumina
mills with zirconia balls, with the addition of a plasticizer.
Materials, uniformly set, after plasticizing and drying, were
granulated. Green compacts with dimensions of 5.5 3.0 35 mm
and 16.5x16.5x10 mm were uniaxially pressed at 100 MPa and then
cold isostatically pressed at 200 MPa. Ceramic composites were
sintered at a maximum temperature of 1973 K at constant heating
and cooling rates of the furnace. Sintering of the specimens was
carried out in a high-temperature following measurements were
3
N
4

49
performed on the tested specimensVickers hardness HV1, fracture
toughness at room temperature K
IC
(based on 3PB), fracture
toughness at an elevated

temperature of 873K K
IC(ET)
, wear
resistance V
n
(determined by speed of mass loss), apparent density
p
, and porosity P
c
(determined by the hydrostatic method).
Single Edge Notched Beam (SENB) specimens (mechanically
notched) with dimensions 1.54.035.00.1mm were used to
determine fracture toughness by means of a conventional method
based on three-point bending of specimens (3PB). An initial 0.9 mm
deep notch was produced a diamond saw (thickness 0.20 mm) and
then the notch tip was pre-cracked with a thin diamond saw
(thickness 0.025 mm). The total initial notch length was
approximately 1.1 mm. The relationship K
IC
= f (c) is given by the
equations (1, 2) [7]:

2 / 1
2
5 . 1 Yc
B W
S P
K
c
IC
= (1)
( ) (
(
\
|
+
=

=
4
0 ,
) 1 ( 3738 . 0
1
3
2
j i
j
i
ij
S
W
A Y
(2)

where: P
c
= critical load, S = support span, W= width, B = specimen
thickness, Y = a geometric function, c = crack length, = c/W and
A
ij
is

the coefficients given by Fett [8]. Measurements of fracture
toughness at an elevated temperature of 873K were carried out on a
ZWICK 1446 instrument with mounted electrical furnace (Fig. 2).
Modulus of elasticity (Youngs modulus) of the tested composites
was determined by a measuring the transmission velocity of
longitudinal and transversal ultrasonic waves through the sample.

3
N
4

50

Fig.2. Measuring position for determination of fracture toughness at elevated
temperatures.

Probe sets work together with a Panametrics Epoch III ultrasonic
flaw detector connected to a controlling computer. Calculations
were made using the following formula:

2 2
2 2
2
4 3
T L
T L
T
C C
C C
C E
= (4)

where: E-Youngs modulus, C
L
-velocity of the longitudinal wave,
C
T
-velocity of the transversal wave, - density of the material.
The velocities of transversal and longitudinal waves were
determined as a ratio of sample thickness and relevant transition
time. The accuracy of calculated Youngs modulus from equation
(4) was estimated to be below 2 %.
Wear resistance was determined by a method based on the
measurement of mass decrement rate during wear of the specimen
against an SiC80 abrasive cloth. The following formula was used to
calculate mass decrement rate (Vn):

Vn=
T F
m
P

1000
[m/h] (5)

3
N
4

51
where: m-absolute mass wear,
p
-apparent density, F-specimen
contact surface, T-time.
Microstructural observations of the specimen were carried out using
a JEOL JSM-6460LV scanning electron microscope. X-ray
diffraction was used both to identify phases and to assess the ratio of
tetragonal zirconia phase (t) to monoclinic zirconia phase (m).
Preliminary cutting tests in the turning of constructional alloy steel
(40H) of hardness 252 HB were carried out for SNGN 12 04 08 type
inserts. Tool life (T
mean
) of the insert cutting edge was measured for
tested composite inserts at the following cutting parameters: speed
(v
c
) = 150 m/min, feed (f) = 0.15 mm/rev., depth (a
p
) = 0.5 mm,
wear criterion (width of flank wear land) VB
B
= 0.30 mm. Cutting
tests were performed without cooling lubricant.
The results obtained from the tests concerning: mechanical
properties, apparent density
p
, and porosity P
c
of tested ceramic
composites of various compositions are presented in Table 2.

Table 2. Physical and mechanical properties of tested ceramic composites: apparent
density
p
, porosity P
c
, Youngs modulus E, Vickers hardness HV1, fracture
toughness at room temperature K
IC
and elevated temperature (873K) K
IC(ET)
, wear
resistance V
n
.
Mat
eria
ls
Appa-
rent
density
p
[g/cm
3
]
Poro-
sity
P
c
[%]
Vickers
hardness
HV1
[GPa]
Youngs
modulus
E
[GPa]
Critical
stress
intensity
factor
K
IC

[MPa m
1/2
]
Critical
stress
intensity
factor
K
IC(ET)

[MPa m
1/2
]
Wear
resistance
V
n

[m / h]
B 4.13 2.3 18.8 378 4.6 (0.1) 4.3 (0.04) 5.7
B1 4.11 2.7 18.4 376 4.5 (0.2) 4.4 (0.4) 6.8
B2 4.09 3.5 18.3 358 4.2 (0.2) 4.0 (0.3) 7.8

Tested titanium carbide-reinforced alumina-based composite tool
ceramics exhibited: high Vickers hardness (in the range 18.3-18.8
GPa), similar values of critical stress intensity factor K
IC
at room
temperature (in the range 4.2-4.6 MPa m
1/2
), K
IC(ET)
at an elevated
temperatures of 873K (in the range 4.0-4.4 MPa m
1/2
), similar values
3
N
4

52
of elastic modulus (378-358 GPa) and a lower mass decrement
rateV
n
(5.7-7.8 m/h) in relation to pure alumina.

Analysis of results
indicates a somewhat higher fracture toughness (by approx.10%) for
ceramic composites containing TiC submicro powders than for those
containing TiC nano powders (<130 m).
The microstructure of the titanium carbide-reinforced ceramic
composites with an alumina matrix is presented in Figures 3 and 4.

a)

b)

Fig.3. SEM micrographs of a surface of the ceramic composites: a) Al
2
O
3
-ZrO
2
-
TiC
submicro
(B), b) Al
2
O
3
-ZrO
2
-TiC
nano
(B1).

a)

b)

Fig.4. SEM micrographs of a fracture surface of the tested composites: a) Al
2
O
3
-
ZrO
2
-TiC
submicro
(B), b) Al
2
O
3
-ZrO
2
-TiC
nano
(B1).

The values of K
IC(ET)
at 873K are somewhat lower in comparison to
those of K
IC
at room temperature. A small reduction in fracture
toughness is observed at elevated temperature (873K) in comparison
to that at room temperature in the range of 2-6% for both the Al
2
O
3
-
ZrO
2
-TiC
nano
and the Al
2
O
3
-ZrO
2
-TiC
submicro
composites. Fine
particles of zirconia and of titanium carbides, homogeneously
ZrO
2

3
N
4

53
distributed in the alumina matrix, are visible in the micrograph of
the Al
2
O
3
-ZrO
2
-TiC
nano
composites (Fig.4). Results of a cutting test
carried out on the titanium carbide-reinforced ceramic composites
with alumina matrix reveal a more than two-fold increase in tool life
(T
mean
=24 min) for the Al
2
O
3
-ZrO
2
-TiC
submicro
composites (B) at the
accepted wear criterion VB
B
= 0.30 mm in comparison to the Al
2
O
3
-
ZrO
2
-TiC
nano
composites B1 and B2 (for which tool life achieves
values of 9 min and 11 min respectively). The results of the
presented investigations allow rational use of existing ceramic tools.
4. Conclusions
Titanium carbide-reinforced alumina-zirconia composites
exhibit: high Vickers hardness, sufficient fracture toughness
(critical stress intensity factor K
IC
is increased to 4.6 MPa
m
1/2
), a high elastic modulus and a higher wear resistance in
comparison to pure alumina.
Analysis of the results indicates a somewhat higher fracture
toughness (by approx. 10%) for ceramic composites
cotaining TiC submicro powders than for those containing
nano powders (<130 m).
A small reduction in fracture toughness is observed at
elevated temperature (873K) in comparison to that at room
temperature in the range of 2-6% for both Al
2
O
3
-ZrO
2
-
TiC
nano
and Al
2
O
3
-ZrO
2
-TiC
submicro
composites.
Preliminary industrial tests confirm the high cutting
performance of titanium carbide-reinforced alumina-based
composite tool ceramics.
Cutting test results indicate a more than two-fold increase in
tool life (T
mean
) for Al
2
O
3
-ZrO
2
-TiC
submicro
composites in
comparison to Al
2
O
3
-ZrO
2
-TiC
nano
composites.
Acknowledgement
This work was supported by the 2007-2013 Innovative Economy
Programme under the EU National Strategic Reference Framework,
priority axis 1, section 1.1.3 No UDA-POIG.01.03.01-12-024/08-00.
3
N
4

54
REFERENCES
[1] Handbook of Ceramic Composites. Edited by Narottam E Bansal, NASA
Glenn Research Center, Kluwer Academic Publishers, Boston / Dordrecht /
London, 2005.
[2] Fortulan C.A., Souza P.D.F., Mat. Research, 1999,2,3, p.205.
[3] Y.G. Liu, Y. Zhou, D.C. Jia, Q.C. Meng, Y.H. Chen. Scripta Materialia,
2002, 27, p. 63.
[4] Szutkowska M., Journal of Materials Processing Technology, 2004, 153-
154, p. 868.
[5] Kee-Do Woo, Byung-Ryang Kim, Eui-Pyo Kwon, Duck-Soo Kang, In-Jin
Shon, Ceramics International, 2010, 36, p.351.
[6] Dobrzaski L., Materiay inynierskie i projektowanie materiaowe. WNT
Warszawa, 2006.
[7] Fett T., Munz D., J. Am. Ceram. Soc., Vol. 75, No 4 (1992), p.958.
[8] Fett T., Eng. Fract. Mech., Vol. 40, No 3 (1991), p.683.

Nanocomposites in the ZrO
2
-Al
2
O
3
-WC system

55
NANOCOMPOSITES IN THE ZrO
2
-Al
2
O
3
-WC
SYSTEM

ZBIGNIEW PDZICH
1
, WOJCIECH MAZIARZ
2

Abstract

Particulate composite in the Al
2
O
3
-ZrO
2
-WC carbide system shows promising
properties in structural applications especially as wear resistant parts. Despite
of mechanical properties of constituent phases, its properties depends on
the residual stress state, the microstructure and the interphase grain boundaries
state. Paper presents the results of investigation of high-resolution transmission
electron microscopy (HREM) on the different type grain boundaries structure and
its chemical composition.
1. Introduction
Fabrication of the dense composite materials using typical powder
compacting technology and following sintering process meets many
barriers connected with each step of this technology. Application
of submicro-or nanosized powders increases difficulties occurring
during manufacturing.
The first task for applied technology is to assure the uniformity
of different phase distribution in prepared composite powder.
The process must eliminate the possibility of one phase grains
aggregation. If it is performed with success, sintering could provide
to homogeneous composite body.
Usually it is profitable for sintering process to avoid of chemical
reactions, nevertheless sintering must be provided at temperatures,
which assure the effective mass transport in sintered phases.

1
AGH University of Science and Technology, Faculty of Materials Science and
Ceramics, Department of Advanced Ceramics, 30 Mickiewicza Av., 30-059
Krakow, Poland, e-mail: pedzich@agh.edu.pl
2
Institute of Metallurgy and Materials Science of Polish Academy of Sciences,
25 Reymonta Str., 30-058 Krakow, Poland
2
-Al
2
O
3
-WC system

56
Additionaly, after sintering process, during cooling, in dense
multiphase bodies residual stresses could be created due to thermal
expansion coefficient mismatch [1, 2].
This work shows the results of detail investigation of the fine-
grained microstructure of three-phase composite containing
comparable amount of each phase. The special care was put to
describe grain boundaries between the grains of different phases
(i. e. alumina-carbide or zirconia-carbide).
2. Experimental
Al
2
O
3
-ZrO
2
-WC composite powder was prepared from commercial
powders (-alumina: TM-DAR, Taimei Chemicals; yttria stabilized
zirconia: 3Y-TZ, Tosoh and tungsten carbide: Baildonit) by wet
(ethanol) mixing in attritor. The volume proportion of each phase
was 1:1:1. Sintering of composite bodies was conducted in
the Thermal Technology hot-press device at 1650C (with 45 min.
soaking time) in argon atmosphere under 25 MPa pressure. Density
of sintered bodies was 8.46 g/cm
3
(98.5% of theoretical value).
The structure investigations were performed in Tecnai G2 FEG
microscope dedicated for high resolution (HREM) observations.
The thin foils were prepared by cutting thin slices of compacts using
diamond sow and followed dimpling and ion milling using Gatan
equipments. Finally the samples were covered by carbon thin layer
in order to assure their conductivity.
The chemical analyses both point and line scans were performed in
the HAADF/STEM mode using a energy dispersive X-ray
spectrometry (EDS).
3. Results
Fig. 1 presents a set of high angel angular dark field (HAADF)
image and corresponding elemental mapping images recorded in
the STEM mode on a marked area.

2
-Al
2
O
3
-WC system

57
1 m
HAADF
1 m 1 m
HAADF

AlK AlK

ZrK ZrK

WL WL

Fig. 1. HAADF image and corresponding elemental mapping images recorded
along marked area.

The obtained contrast is due to a different atomic number
of elements (Z
2
-contrast) and it allows to analyze a particulate
phases. Taking to the account a distribution of Al, Zr and W, we can
calculate the mean grain size of Al
2
O
3
, ZrO
2
, WC phases,
respectively. In this purpose the image analyze software Gatan
Digital Micrograph 3.10.0 was used. The procedure of analyze
was as follow: the elemental mapping images of Al, Zr and W were
filtered applying maximum filter, then from obtained binary
images the histograms of gray level were created and determination
the threshold for maximum intensity of particulate phases were
performed. Finally, we obtained the images of particulate phases
2
-Al
2
O
3
-WC system

58
and measured values of their mean grain size and filled area (Fig. 2).
The mean grain size of investigated phases were as follow: Al
2
O
3

330 nm, ZrO
2
280 nm and WC 340 nm, and relative area
friction: 35.8, 30.7, and 33.5%, respectively. The obtained data were
in a good agreement with these used as initial conditions for
preparing of compacts (size of powders and volume proportion
of each phase). Calculated mean grain size values for composite
components are comparable with these calculated using much
smaller image magnifications presented in [3].

Al
2
O
3
- 35,8% ZrO
2
- 30,7% WC - 33,5%
Fig. 2. Results of images analyze showing distribution of particulate phases in the
area indicated in Figure 1.

In order to analyze the grain boundaries between particulate phases
a line scan analyses were performed for several grains. Fig. 3
presents result of line scan analyse between WC and Al
2
O
3
grains
performed along the marked line.
One can see that both Al and W concentration is changing very
sharp close the boundary, it means that no diffusion of this elements
occurs. Concerning concentration changes of O, and C, is difficult to
say about their diffusion, due to contamination effect, limit
of detection of light elements by EDS method, thickness of sample
and finally necessity of covering by the carbon nonconductive
ceramic sample. Additionally, there is no information about
the thickness of grains, therefore the overlapping of signal from
different grains can occurs.
2
-Al
2
O
3
-WC system

59

Fig. 3. HAADF image of composite microstructure and result of line scan analyze
across the WC and Al
2
O
3
grains boundary.

The observations of grain boundaries were performed in atomic
scale also. Fig. 4 presents the high-resolution transmission electron
microscopy (HRTEM) showing grains boundary between WC and
Al
2
O
3
.

5 nm
HREM
Al
2
O
3
WC

Fig. 4. HRTEM image and corresponding FFT and IFTT images showing the
grains boundary between WC and Al
2
O
3
.

One can see the sharp boundary between [001] and [451] oriented
grains of WC and Al
2
O
3
phases, respectively. The inverse fast
Fourier transform (IFFT) images showed that Al
2
O
3
is almost free
from structural defects, whereas plenty of dislocations and stacking
faults on {010} planes can be distinguish in WC grain.
5 nm
90
FFT
IFFT
WC [001]
010
010
d=3.0
d=3.0
5 nm
87
FFT
IFFT
d
-110
=4.1
d
0-14
=2.6
Al
2
O
3
[451]
2
-Al
2
O
3
-WC system

60
Fig. 5 presents a result of line scan analyse between WC and ZrO
2

grains performed along the marked line.

100 nm
HAADF
100 nm 100 nm
HAADF

Fig. 5. HAADF image of composite microstructure and result of line scan analyze
across the WC and ZrO
2
grains boundary.

One can see that signal originating from Zr increases in
measurement range form 25 nm up to about 110 nm reaching
maximum inside the ZrO
2
grain. This behaviour can be explain as
both overlapping of the WC and ZrO
2
grains or some diffusion of Zr
into WC grain. However taking to the account character of signal
originating form O (almost the same as in case of Zr) most probably
this first one explanation is correct. The above assumption that the
grain boundaries in Al
2
O
3
-ZrO
2
-WC composite are free from
additional phases was proofed by high resolution observation.
Despite of chemical reaction, during high temperature treatment,
some chemical elements segregation could occurs. In the WC grain
the W/C and W/O ratio was changing continuously. It suggests that
during sintering oxygen diffuse into the carbide structure. Paralelly,
carbon changes its content in carbide grain what has to change
locally WC stoichiometry. It was previously observed in [4].
WC
ZrO
2

2
-Al
2
O
3
-WC system

61
5 nm
5 nm 5 nm
5 nm
110
110
100
WC [001]
ZrO
2
[111]
011
101
Al
2
O
3
[451]
104
111
hex-Al
2
O
3
t-ZrO
2
L1
2
-WC

Fig. 6. HRETM image showing a triple point between three phases,
corresponding FFT and IFFT images.

Fig. 6 presents HRETM image showing a triple point between three
phases. Each type of interphase boundaries is represented Al
2
O
3
-
ZrO
2
, Al
2
O
3
-WC and ZrO
2
-WC. The corresponding fast Fourier
transforms (FFT) allowed to identify the crystallographic structure
of coexistence phases. The tetragonal ZrO
2
with [111] orientation,
L1
2
WC with [001] orientation and hexagonal Al
2
O
3
with [451]
orientation were identified respectively by measurement of distances
and angles on corresponded FFTs images. The phase boundaries
areas did not show additional phases inside.
4. Summary
Investigations showed that the composite microstructure consisted
of sub-micrometric grains of each phase randomly and
homogeneously distributed. The grains boundaries were distinct and
clear. This confirms high purity of used starting powders and a lack
2
-Al
2
O
3
-WC system

62
of chemical reactions between constituent phases at applied
sintering conditions. HREM observations of interphase boundaries
and the corresponding fast Fourier transforms (FFT) and inverse fast
Fourier transforms (IFFT) allowed to identify the crystallographic
structure of coexistence phases in boundaries area.
Local changing of the chemical composition of WC phase in contact
with zirconia grains was detected. However, its nature should be
elaborate in details, distinct changes of O/W and C/W ratios suggest
that in the area of WC/zirconia interphase boundary chemical
changes occurred.
WC in contact with alumina was stable. There was no distinct
chemical changes on the WC/alumina interphase boundary.
Acknowledgements
The work was supported by the Polish Ministry of Science and
Higher Education under project number 11.11.160.364.

REFERENCES

[1] Z. Pdzich, G. Grabowski, W. Maziarz, Wybrane kompozyty ceramiczne o
fazach cigych analiza stanu napre i ich wpyw na waciwoci,
Materiay Ceramiczne/Ceramic Materials, 60 [4] (2008) 258.
[2] Z.Pdzich, G.Grabowski, Wpyw rodzaju wtrce na stan napre,
wytrzymao i niezawodno kompozytw ziarnistych na osnowie tlenku
glinu, Kompozyty (Composites), 9 [2] (2009) 149.
[3] Z.Pdzich, Zuycie cierne materiaw kompozytowych na osnowach
tlenkw glinu i cyrkonu w rnych rodowiskach pracy, Materiay
Ceramiczne (Ceramic Materials) 61 [2] (2009) 15.
[4] K. Haberko, Z. Pdzich, J. Piekarczyk, M.M. Buko, W. Suchanek,
Tetragonal Zirconia Polycrystals Under Reducing Conditions, in Third
Euro-Ceramics, vol. 1, Processing of Ceramics, P.Duran, J.F.Fernandez
(Ed.), Faenza Editrice Iberica S. L., Spain, 1993, 967.

TEM investigation of matrix SAFFIL
TM
fiber interfaces in aluminium alloys based composites

63
TEM INVESTIGATION OF MATRIX -
SAFFIL FIBER INTERFACES IN
ALUMINIUM ALLOYS BASED COMPOSITES

JERZY MORGIEL
1
, JACEK KACZMAR
2
,
MAGORZATA POMORSKA
1
, KRZYSZTOF NAPLOCHA
2

Abstract

The microstructure of metal matrix composites of AA2024, AA 6061 and EN AC
44300 alloys with 10 vol. % Saffil fibers were investigated using transmission
electron microscopy. The fibers in the preform were stabilized using liquid glass
binder and fired at 800
o
C resulting in formation of amorphous silicon oxide joints.
The preforms infiltration with AA2024 leads to reaction of liquid metal with SiO
2

binder substituting it with fine-crystalline mixture of MgO, and silicon. Similar
operation with AA6061 leads to binder substitution with porous amorphous Al
2
O
3

and MgO. Only squeeze casting of EN AC 44200 with Mg<0.6 % left most of the
binder and allowed to obtain practically clean Saffil/ matrix interfaces.
1. Introduction
The aluminium alloys found already wide application in automotive
industry for production of engine blocks, piston heads and others,
but stronger and stiffer materials would be very much welcomed.
The experiments with reinforcing of high strength aluminium alloys,
like AA6061, 2024, 7075 or AlMgCuAg with up to 20 vol. % of
Al
2
O
3
Saffil fibers showed roughly 20% of UTS increase already
at ambient temperatures [1-4]. On the other hand, a composite based
on AC 8B casting alloys just reproduced the matrix maximum
hardness [2]. Simultaneously, both groups of composites retain their
ambient temperature strength by ~100
o
C higher, than their matrixes.

1
Institute of Metallurgy and Materials Scince, Polish Academy of Scinces, 25
Reymonta St., 30-059 Krakow,
2
Institute of Production Engineering and Automation, Wrocaw Uniwersity of
Technology, ukasiewicza 3/5, 50-371 Wrocaw
TM

64
The reason of different response of matrixes to the same reinforcing
Saffil fibers, as well as further optimization of properties of these
composites, could be obtained by finding the detail nature of
metal/ceramic interfaces in these materials.
The problems with production of such composites were overcome
partly by pressure infiltration and partly by wetting enhancement
with alloying additions, like magnesium, routinely introduced into
commercial aluminium alloys. However, switching from the Al
Al
2
O
3
system in to AlMg Al
2
O
3
causes that reactive infiltration
takes over. In that case, at the metal/ceramic interface an MgO oxide
and at 4<Mg<8 wt. % an even more brittle MgAl
2
O
4
spinel may
form [1]. The situation in that area is worsened by abundance of
oxygen from SiO
2
binder easily dissolved in contact with liquid
metal containing magnesium.
The observations of Al4Cu1Mg0.5Ag matrix with Saffil fibers
bonded with SiO
2
showed, that in multi-component alloy the
MgAl
2
O
4
may form along the MgO at very low magnesium
additions (~1 wt.%) [3,5]. Similarly, the MgAl
2
O
4
was found on -
Al
2
O
3
particles strengthening Al1Mg0.6Si0.3Cu (AA6262) alloy
[6]. The type of phases formed at the Saffil fiber/matrix interface
are important, but their microstructure, as well as presence of voids
easily nucleated during such phase transformation might be even
more detrimental on l properties of such composites. However, the
information on that problem is limited.
This project was aimed at investigation of the type and
microstructure of phases formed at the Saffil/ metal matrix interface
in a series of aluminium alloys used in car industry applications.
2. Experimental procedure
Performs made of Safill fibers (10 vol. %) were stabilized with
liquid glass (hydrated SiO
2
), dried and fired at 800
o
C. Next, they
were squeeze cast with an AA2024, 6061 or EN AC 44200 alloys
TM

65
(Table I). Finally, composite materials were solution heat treated
and aged to peak hardness.
The microstructure was investigated using Tecnai SuperTWIN FEG
200kV transmission electron microscope with EDAX microanalysis
system. Thin foils f were cut using Quanta 3D focused ion beam
equipped with Omniprobe lift-out attachment.
Table 1. Additions (wt.%) to alloys used for infiltrating the preforms
Alloys Mg Si Cu Mn Fe other
AA2024 1,2-1,8 0,4-0,6 3,8-4,9 0,3-0,9 <0,5 <0,1
1)

AA6061 0,8-1,2 0,4-0,8 0,1-0,4 <0,15 <0,7 <0,4
2)

EN AC 44200 0,2-0,6 6,5-7,5 <0,1 <0,3 <0,5 <0,2
3)

1)
Cr <0,1, Ti <0,15, Zn <0,25,
2)
Cr 0,04-0,35, Ti <0,15, Zn <0,25,
3)
Ni<0,1, Zn<0,1, Pb<0,1
3. Results
Immersing of Saffil fibers into liquid glass covers them with
thin layer of binder forming bridges between individual fibers
during subsequent forming and firing (Fig.1a). The firing of such
preform at high temperature fixes the bridges into permanent form
and allows to keep the preform shape during squeeze casting with
liquid aluminium alloy. The microstructure observations showed,
that rough fiber surface is well covered with the binder, which
solidifies in an amorphous form (Fig. 1b).
a)

b)

Fig.1. Scanning (Secondary Electrons) image of preform after firing process (a) and
transmission image of SiO
2
bridge in-between adjacent Saffil fibers
TM

66
The squeeze casting of such preform with AA2024 alloy caused
reaction of liquid metal with the binder substituting it with dense
nano-crystalline material (Fig.2a). The local chemical analysis using
EDS system indicated, that at areas with bigger amount of binder,
i.e. so called bridges the surface of the Saffil fibers covers copper-
aluminium crystallites, i.e. probably phase (Fig. 3). The centre of
such post-binder area is filled with the mixture of the and silicon
crystallites immersed in remnants of SiO
2
binder phase, while at
their outskirts the MgO oxide forms. However, outside the joints,
i.e. where Suffil were covered with thinner layer of binder the
latter was totally dissolved and fine crystallites of the MgO
precipitated directly on the fibers surface.
The infiltration of preform with AA6061 alloy resulted in
substitution of SiO
2
binder with an amorphous highly porous
material (Fig. 4). Aside of porosity, larger voids at the Saffil
/metal matrix were also observed. The maps presenting local
chemical composition helped to determine, that the amorphous
material consist of alumina with traces of silicon oxide (Fig.5).
Small amount of SiO
2
present inside Al
2
O
3
fiber was introduced into
them already at their production stage, i.e. the R.F. grade contain up
to 4 at. % Si preventing grain growth. The manganese oxide was
located in small amount and only at the border of amorphous areas
and matrix. The copper containing type precipitates are usually
nucleated on these amorphous areas like the one visible in the centre
of HAADF image in Fig.5, or directly on Saffil surface.
The same experiments were also performed with EN AC 44200
alloy. In that case, most of the binder bridges withstand contact with
liquid metal preserving connections in-between Saffil fibers (Fig.
6). More detailed observations of the binder showed that it contain
small petal-type voids, which may indicate either start of dissolution
of amorphous SiO
2
or just represent the effect of sample preparation
with high energetic Ga+ beam (Fig.7). The precipitation of silicon
was noted both within matrix and at the fiber surface some of the
within the binder area (Fig.8, 9).
TM

67

Fig.2. Transmission image of products of reaction between SiO
2
binder and liquid
AA2024 alloy.

Fig.3. Scanning-transmission (HAADF) image and series of maps presenting
distribution of O, Al, Si, Cu and Mg obtain from area of reaction between SiO
2

binder and liquid AA2024.
TM

68

Fig.4. Transmission image of products of reaction between SiO
2
binder and liquid
AA6061 alloy.

distribution of O, Al, Si and Mg obtain from area of reaction between SiO
2
binder
and liquid AA6061.

TM

69

Fig.6. Transmission image of SiO
2
binder bridges between Saffil in perform force
infiltrated with AC 44200 alloy.

Fig.7. Transmission image of Si precipitates at Saffil boundaries within binder in
perform force infiltrated with AC 44200 alloy.
TM

70

Fig.8. Transmission image of Si precipitates at Saffil boundaries outside binder in
perform force infiltrated with AC 44200 alloy.

distribution of O, Al, Si and Mg obtain from Saffil/ AC 44200 matrix interface.
4. Discussion and summary
The squeeze casting of ceramic preforms made of Saffil fibers
using liquid AA2024, 6061 or EN AC44200 alloys produced dense
TM

71
metal matrix composites. However, an improvement in mechanical
properties of these composites could be gained only after
ascertaining a proper connection between these high strength (~2
GPa) nano-crystalline slightly porous (5 10 %) of 3 -5 m
diameter Al
2
O
3
Saffil fibers and their matrix. The Vaucher and
Beffort [1] indicate, that a proper bonding between Al
2
O
3
fibers and
aluminum alloy depends to large extent on amount of magnesium
additions deciding on type of oxide phases, like MgO or MgAl
2
O
4

formed as a consequence of the production process. They pointed
also, that presence of the spinel forming from 4<Mg< 8 wt. %
content is the most detrimental for composite mechanical properties.
Even, as this amount is higher than usual magnesium additions in
commercial alloys, though presence of SiO
2
bonding phase might
cause nucleation of MgAl
2
O
4
also in case of such alloys, i.e. like
Al1Mg2Cu0.5Ag [3] or Al12SiCuMgNi [5]. Present experiments
with AA2024 and AA6061 alloys containing 1.6 and 1 wt. % Mg
respectively showed however only presence of MgO precipitates.
The EN AC44200 alloy with even lower magnesium was free from
any magnesium oxides. The observed differences between literature
and experiment are probably connected with varying squeeze
casting conditions, like infiltration time, pressure and other.
The lack of MgAl
2
O
4
spinel at Saffil fiber/matrix interface,
generally considered as a necessary condition for obtaining proper
composite strength improvement, is not an only requirement. The
microcrystalline mixture of -Al
2
Cu, Si and MgO material found in
Saffil/AA2024 composite has close similarity to the one with
spinel obtained after infiltration Saffil with Al4Cu1Mg0.5Ag [3].
Even, as such a mixed microstructure seems of dubious value for
load transfer, though the measurement of the latter indeed showed
significant strength increase. However, the porous mixture of
amorphous Al
2
O
3
and MgO substituting bonding phase in
Saffil/AA6061 composite definitely stand no chance at all at
keeping fiber-matrix together. What is more, the voids also present
in this area might serve, as good crack initiators. In this situation,
TM

72
the Saffil/AC44200 composite with preserved amorphous SiO
2

bonding phase at fiber surface and therefore with clean interfaces
should present the highest relative property improvement, i.e. as
compared with its matrix.
The performed microstructure observation aimed at characterization
of Saffil fiber/ aluminum matrix interfaces clearly indicated, that
the composite mechanical properties are strongly depended not only
on type of phase formed over there, but also on defects like porosity
or voids. The decreasing magnesium content in the AA2024, 6061
or EN AC44200 alloys used as matrixes in investigated composites
resulted in diminishing or even eliminating formation of MgO at
fiber interfaces. However, the above alloys differ also in amount of
other alloying additions and consequently in melting temperature, so
the magnesium content was just one of parameters influencing
above changes at Saffil/matrix interfaces.
Acknowledgements
The work was supported through KomCerMet project (contract
no. POIG.01.03.01-14-013/08-00 funded by Polish Ministry of
Science and Higher Education) in the framework of the Operational
Programme for Innovative Economy 2007-2013.

REFERENCES

[1] Vaucher S. and Beffort O., Bonding and interface formation in metal matrix
composites, MMC-Assess Thematic Network, EMPA, TU Wien 1998-2001
[2] Automotive Data Sheet Overview, Honda Motor Company,
www.saffil.com/pdfs/automotive/data/mmcprop.pdf
[3] Cayron C, TEM studies of interfacial reactions and precipitation mechanisms
in Al
2
O
3
short fiber or high volume fraction SiC particle reinforced
Al4Cu1Mg0.5Ag squeeze cast composites, Ph. D thesis, Lausane 2000
[4] Long S., Beffort O., Moret and Thevoz Ph., Processing of Albased MMCs by
Indirect Sqeez Infiltration of Ceramic Preforms on a Shot-Control High
Pressure Die casting machine, Aluminium 76(2000)82-89
TM

73
[5] Dudek H.J., Wang S., Fibre-matrix interaction in the - Al2O3 fibre
reinforced aluminium piston alloy, Materials Science and Engineering,
A205(1996)180-186
[6] Veeresh Kumar G.B., Rao C.S.P., Selvaraj N., Bhagyashekar M.S., Studies on
AA6061-SiC and AA7075-Al
2
O
3
Metal matrix Composites, Journal of
Minerals and Materials Characterization and Engineering, 9(2010)43-55

Tribological properties of supersonic sprayed nanostructured WC12Co coatings

74
TRIBOLOGICAL PROPERTIES
OF SUPERSONIC SPRAYED
NANOSTRUCTURED WC12CO COATINGS

WOJCIECH RAWSKI
1

Abstract

WC12Co nanostructured composite was prepared by means of the hypersonic spray
process (HVOF). The microstructure and composition of tungsten carbide
nanopowder were analyzed by scanning electron microscope (SEM) and
transmission electron microscope (TEM). Investigations revealed the size of the
original particles to be approximately 50-500 nm. The nanostructured sprayed
coating was observed by SEM analysis and investigated by X-ray diffractometer
(XRD). A denser coating structure with higher hardness was observed compared to
conventional coating with a small amount of W
2
C, WC
1-x
, W and some amorphous
phase. The coefficient of friction for the HVOLF-sprayed WC-12Co nanostructured
coating was four times lower than for the conventionally sprayed WC-12Co
coating.
1. Introduction
The study of nanostructured materials has been extended to include
coatings processed using the thermal spray technique. The
possibility of producing coatings with superior properties (e.g. wear
resistance) when compared to the conventional thermal spray
coatings currently available opens a wide range of research
opportunities for different materials. In thermal spray processes,
materials in wire or powder form are fed into a flame, arc or plasma
spray gun, where they are fully or partially melted and accelerated in
a gas stream toward a substrate to be coated. Nanostructured
materials cannot be fed directly into the spray guns. They are
deposited in the form of suspensions, or powders whose grains are
agglomerated nanocrystals [1-3]. Although the coatings produced by

1
Kielce University of Technology, The Centre for Laser Technologies of Metals,
Al. 1000-lecia PP7, 25-314 Kielce, Poland, e-mail: ktrwz@tu.kielce.pl

75
thermal spraying possess numerous defects, such as pores and
microcracks, the process efficiency is much higher and the costs are
considerably lower, compared to those of other processes. It is,
therefore, possible to use this technique on an industrial scale.
Tungsten carbide is widely used in industry to improve the wear
resistance of machine parts. WC-Co thermally sprayed coatings are
well established as materials highly resistant to wear in a wide
variety of situations [4-6]. Therefore nanostructured WC-Co
coatings are of particular scientific interest because of their physical
and mechanical properties, which are superior to those of
conventional materials.
Tungsten carbide coatings were deposited using the atmospheric
plasma spray (APS) or D-gun process or the high velocity oxy-fuel
(HVOF) process. The high temperature of the APS process caused
decarburization of the tungsten monocarbide to W
2
C and the
occurrence of metallic tungsten. These phenomena can be described
by the following reactions [7]:
2WC W
2
C + C (1)
W
2
C 2W + C (2)
There was a considerable reduction in the amount of carbon, due to
oxidation of the coating in the flame, which can be described by the
following reaction:
2C + O
2
2CO (gas) (3)
All these processes are dependent on the thermal spray parameters,
which have a significant influence on the coatings wear resistance.
The D-gun process is a method that yields coatings with excellent
properties; it is, however, expensive to operate. The high velocity
oxy-fuel (HVOF) process is one of the most popular thermal spray
technologies. The HVOF process possesses advantages over APS,
such as high particle velocities and a lower temperature of the spray
stream, producing a denser coating with reduced formation of
detrimental reaction products. Due to its flexibility and superior
coating quality, the HVOF process has numerous applications in
different branches of industry. Investigations carried out in this
paper concern the properties of nanostructured and conventional

76
WC-12Co cermet coatings deposited by the HVOLF method,
including microhardness, microstructure, phase composition,
coefficient of friction and wear resistance.
2. Experiment
The powder and sprayed coating structure was analyzed using
tungsten carbide, WC12Co. The nanostructured tungsten carbide
powder (Infralloy S7412) and sprayed coating was denoted as
WC12Co-N, whilst the conventional powder (Amperit 519.074) and
sprayed structure was denoted as WC12Co-A. In both cases, the
grain size ranged from 15 to 45 m. The grain size and surface
morphology of the two powders are shown in Figs. 1 and
2, respectively. These materials are designed to be used in thermal
spraying. They are produced by agglomeration and sintering of fine
powder grains. They contain a very large number of spherical
grains, which results in loose powder grains compared to materials
with irregularly shaped grains.

a)

b)

Fig. 1. WC12Co-N powder, morphology of; a) grains, b) grain surface


77
a)

b)

Fig. 2. WC12Co-A powder, morphology of; a) grains, b) grain surface.
Figs. 1b and 2b show a distinct difference in grain surface
morphology between the two materials. This is a result of the
different dimensions of the materials applied to produce the
powders. The nanostructured powder grains consist of nanometric
and submicron WC crystals in the range of 50-500 nm (Fig. 3)
whilst in the production of the conventional WC12Co-A powder,
1 m grains were applied.
a)

b)

Fig. 3. Nanometer and submicron WC12Co-N powder particles; a) on the powder
grain surface, b) in the cross-section of the powder grain
The coatings were deposited using a kerosene-fuelled HVOLF
thermal spray gun (TAFA JP-5000).
The structure and chemical composition of the powders and the
coatings were analyzed using the following scanning microscopes:
JSM5400 with an ISIS 300 Oxford (EDS) microprobe and FEI

78
Nova' NanoSEM 200. Their phase composition was studied using
a Bruker D8 Advance diffractometer with Co-K radiation of
wavelength = 1.78897 .
A CSM Instruments ball-on-disc type tribotester was used to
determine the coefficient of friction under unlubricated conditions
for the sprayed carbide coatings. The ball, made of bearing steel
100Cr6, had a diameter of 6 mm. The testing involved the use of
a computer to aid in controlling and registering the action of the
friction force in the function of time. The parameters of the
tribotester were as follows: radius 10 mm, load 1 N, linear
velocity v = 0.063 m/s, number of laps 10,000, total distance - 754
m. A schematic diagram of the operation of the tester is presented in
Fig. 4. The coatings of the disc samples were sprayed and
subsequently ground and polished for one hour. Roughness of the
WC12Co-N coating was R
a
= 0.043 m, and that of the WC12Co-A
coating was R
a
= 0.015 m. Microhardness, measured with
a Vickers indenter with a load of 5 N, was 1159 HV and 966 HV,
respectively, as the average of 10 measurements carried out using
a CSM Micro Identation Tester.
The microstructures of the HVOLF-sprayed coatings are shown in
Fig. 4.
a)

b)

Fig. 4. Cross-section of hypersonic sprayed coating: a) WC12Co-A, b) WC12Co-N
Metallographic images of the two WC12Co coatings showed that
there were some small undeformed tungsten carbide grains

79
embedded in the cobalt matrix. From the EDS microanalysis, it was
clear that the coating content was different in each zone. The light-
coloured grains in the WC12Co coatings testify to a high amount of
tungsten, whilst the dark-coloured matrix is an area with a high
content of cobalt and a low content of tungsten. Different sizes of
tungsten carbide grains were visible in both coatings. The coating
deposited using the nanostructured powder had a finer-grained
structure with unmodified nanocrystals. It exhibited lower porosity,
despite the fact that the nanostructured powder contained larger
grains. The coating produced from the conventional powder, on the
other hand, had higher porosity, resulting from the higher porosity
of the grains, which can be seen in the metallographic images.
Microstructure in bright field (BF) and related electron diffractions
from an area with grains of different Co content are shown in Fig. 5.
In the area with enhanced Co content, an amorphous structure
occurs, and crystallite grains possess a hexagonal WC structure.

Fig. 5. Microstructure of WC12Co-N coating
The microstructure of two WC grains in the WC12Co-N coating and
related electron diffraction (BF) are shown in Fig. 6. Grains are
surrounded by an amorphous cobalt matrix. The dimensions of WC

80
grains are in the range of 200-500 nm. Electron diffraction (SADP)
is in conformity with the hexagonal structure and orientation of WC
[102]. On the basis of analysis of the microstructure of WC12Co-N
sprayed coating (Fig. 6), it can be concluded that nanocrystal grains,
present in the powder, are also present in the sprayed coating. This
confirms that the process of hypersonic spraying, with a relatively
low spray stream temperature and short dwell time of particles in the

Fig. 6. WC grains in WC12Co-N coating.
stream, does not cause decomposition of WC nanocrystals, despite
their very small dimensions. Significant changes in the phase
composition of WC, which occur particularly during plasma
spraying, substantially reduce the properties of coatings. [8, 9]. The
influence of heat on sprayed WC12Co-N powder is well seen in the
case of cobalt, in which amorphous structures form around
nanocrystal grains of WC.
Analysis of the powder diffraction patterns showed that the WC
grains in the nanostructured powder were half the size of those in
the conventional powder. The diffraction patterns revealed the
presence of WC and Co both in the WC-Co powder and in the
deposited coating. The new phases that appeared in both coatings,
i.e. W
2
C, WC
1-x
and W, were attributable to the prior disintegration
of WC in the spray stream. The diffraction lines of the phases after
spraying were considerably wider. This testified to a significant
degree of elastic and plastic deformation, i.e. a high level of energy
stored in the form of network defects; this particularly in the case of
the Co phase [10].
Figure 7 shows time-dependent changes in the coefficient of friction
for both coatings. As can be seen in the two diagrams, the

81
coefficient of friction for the sprayed WC12Co-N stabilized after the
first 50 m, with a slight tendency for growth. The changes in the
coefficient of friction for the WC12Co-A coating revealed a much
higher level of growth and significant oscillations.

Fig. 7. Coefficient of friction for hypersonic sprayed coating: a) WC12Co-N
coating, b) WC12Co-A coating.
Conclusions
1. The nanostructured WC-12Co coating possessed nanocrystal
grains which were present in the sprayed powder.
2. The influence of heat on sprayed WC12Co-N powder forms
amorphous structures of Co which surround nanocrystal grains
of WC.
3. The microhardness of the nanostructured WC-12Co coating was
higher than that of the conventional WC-12Co coating; average
microhardness exceeded 1159 HV 20% more than that of the
conventional coating.
4. The coefficient of friction for the HVOLF-sprayed WC-12Co
nanostructured coating was relatively stable and was four times
lower than for the conventionally sprayed WC-12Co coating.


82
Acknowledgements
This work was funded by the Polish Ministry of Science and Higher
Education Grant No. N503 015 32/2296.

REFERENCES

[1] Fauchais, P., Montavon, G., Bertrand, G.: From powders to thermally
sprayed coatings. Journal of Thermal Spray Technology 2010, 19 (1-2), pp.
56-80.
[2] Lima, R.S.; Khor, K.A.; Li, H.; Cheang, P.; Marple, B.R.: HVOF spraying
of nanostructured hydroxyapatite for biomedical applications. Materials
Science & Engineering A, Volume: 396, Issue: 1-2, April 15, pp. 181-187
(2005).
[3] Bolelli, G.; Cannillo, V.; Gadow, R.; Killinger, A.; Lusvarghi, L.; Rauch, J.;
Romagnoli, M.: Effect of the suspension composition on the microstructural
properties of high velocity suspension flame sprayed (HVSFS) Al2O3
coatings. Surface & Coatings Technology, Volume: 204, Issue: 8, January
15, 2010, pp. 1163-1179.
[4] Fang, Z. Zak; Wang, Xu; Ryu, Taegong; Hwang, Kyu Sup; Sohn, H.Y.:
Synthesis, sintering, and mechanical properties of nanocrystalline cemented
tungsten carbide A review. International Journal of Refractory Metals and
Hard Materials, Volume: 27, Issue: 2, March, 2009, pp. 288-299.
[5] Chivavibul, Pornthep; Watanabe, Makoto; Kuroda, Seiji; Shinoda, Centavo:
Effects of carbide size and Co content on the microstructure and mechanical
properties of HVOF-sprayed WCCo coatings. Surface & Coatings
Technology, Volume: 202, Issue: 3, December 5, 2007, pp. 509-521.
[6] Chen, Hui; Gou, Guoqing; Tu, Mingjing; Liu, Yen: Research on the Friction
and Wear Behavior at Elevated Temperature of Plasma-Sprayed
Nanostructured WC-Co Coatings. Journal of Materials Engineering and
Performance, Volume: 19, Issue: 1, February 2010, pp. 1 6.
[7] Liao H., Normand B., Coddet C.: Influence of coating microstructure on the
abrasive wear resistance of WC/Co cermet coatings. Surface & Coatings
Technology 124 (2000), pp. 235-242.
[8] Bartuli, Cecilia; Valente, Teodoro; Cipri, Fabio; Bemporad, Edoardo; Tului,
Mario: Parametric study of an HVOF process for the deposition of
nanostructured WC-Co coatings. Journal of Thermal Spray Technology,
Volume: 14, Issue: 2, June 2005, pp. 187 195.
[9] Hewitt, Stephen A.; Laoui, Tahar; Kibble, Kevin K.: Effect of milling
temperature on the synthesis and consolidation of nanocomposite WC10Co
powders. International Journal of Refractory Metals and Hard Materials,
Volume: 27, Issue: 1, January, 2009, pp. 66-73.

83
[10] rawski W., Burakowski T.: Mikrostruktura nanostrukturalnych powok
wglikowych natryskanych naddwikowo. Inynieria Materiaowa 6/2008,
pp.608-610.
Amorphous Ti-Si-C thin film deposited on AISI 316L in low temperature

84
AMORPHOUS Ti-Si-C THIN FILM
DEPOSITED ON AISI 316L IN LOW
TEMPERATURE

AGNIESZKA TWARDOWSKA
1

Abstract

Dual beam IBAD technique was used for TiSiC films formation on AISI 316L
austenitic steel substrates in room temperature, from a single Ti
3
SiC
2
compound
target. As evidenced by X-ray diffraction, scanning electron microscopy, and
transmission electron microscopy and nanoindentation test, resultant Ti-Si-C films
were soft, uniform and dense, amorphous or multiphase, with presence of Ti
5
Si
3
(or
Ti
5
Si
3
C
x
), TiSi and TiSi
2
traces in amorphous Ti- and Si rich matrix.
1. Introduction
To obtain desirable properties of multilayer coatings the selection
and sequence of layers is of primary importance, because
parameters, structure and function of each one give individual
contributions and/ or synergistic effect in total properties of resultant
composite coating. Both chemical vapor deposition (CVD) and
physical deposition techniques (PVD) are used for multilayer
coatings synthesis. CVD techniques are very effective, but usually
require high substrate temperature, which is not acceptable for metal
or polymer substrates preferred for industrial applications. PVD
techniques enable low substrate temperature deposition and in
general rely on material transfer from condensed phase evaporant or
sputtered target. In ion beam techniques ion bombardment is
employed to modify the structure and composition of the substrate
subsurface (ion implantation) or to densify and modify the structure
and composition of previously deposited films (ion mixing) or to
eject atoms from the target surface with subsequent deposition on
the substrate surface in film form. The main difference in above

1
Dep. Technology, Pedagogical University, 2 Podchorazych St., 30-084 Krakow,
Poland

85
mentioned methods is that the primary interest in implantation and
ion mixing is the surface composition, structure and properties
modification by atoms remain, while in sputtering- atoms remove
from the target surface. Ion beam assisted deposition (IBAD)
combines mentioned above processes in one, as shown in Fig.1.

a)

b)

c)

d)

Fig.1. Ion beam techniques in surface engineering:
a) ion implantation, b) ion mixing, c) ion sputtering, d) ion beam assisted
deposition BAD

Titanium-silicon carbide Ti
3
SiC
2
is the most widely researched
representant of ternary, inherently nanolaminated compounds, which
stoichiometry is written after Barsoum as M
n+1
AX
n
, where M is
early transition metal, A- IIIA or IVA group element, X- carbon or

86
nitrogen, n=1, 2, or 3. For simplification,
due to n value, the stoichiometry of MAX
phases is often given by numbers alone, and
noted as 211 (n=1), 312 (n=2), and 413
(n=3). The crystal structure of MAX phases
is hexagonal, related to the interstitial
transition metal carbides and nitrides (MX)
with insertion of A-element layers (the unit
cell of Ti
3
SiC
2
is shown on Figure 2),
typical c/a ratio for MAX phases is 58 [1].
Ti
3
SiC
2
is a light material, resistant to
oxidation and thermal shocks as ceramic,
but relatively soft, damage tolerant,
machinable, thermal and electrical
conductive, as metal [2]. Atypical for
ceramic characteristic of this carbide
(Tab.1) make it desirable for numerous
practical application: as high temperature constructions, protective
coatings, low friction surfaces, petrochemical industry, electrical
contacts or optics.

Table 1. Selected physical and mechanical properties of Ti
3
SiC
2

Density
[g/cm
3
]
Melting
Point
[C]
Thermal
Conductivity
at 20C
[Wm
-1
K
-1
]
Coefficient of
Thermal
Expansion
[10
-6
K
-1
]
Young
Modulus
[GPa]
Vickers
Hardness
[GPa]
4,53 >3000 30-45 8-12 300-320 4-6

For titanium-silicon carbide thin films synthesis, chemical vapour
deposition methods are mostly used but such process requires
temperature of the substrate above 1000C, which is too high to use
most metal or polymer substrates. To lower Ti
3
SiC
2
synthesis
temperature to 300C or less, different physical vapor deposition
methods (PVD) are applied [3,4], but unsuccessfully so the problem
is still unresolved. The aim of this work is to apply IBAD method to
form in low temperature Ti-Si-C thin film on AISI 316L substrate,
from single Ti
3
SiC
2
target. IBAD was chosen as an effective low

Fig.2. Ti
3
SiC
2
unit cell:
a= 3.07 , c= 17.67 [1]

87
temperature thin film deposition method, its versatility, possibility
of process control and excellent film to substrate connection.
2. Experimental Procedure
2.1. Ti-Si-C thin films deposition parameters
The parameters of IBAD process i.e. type of ions and ions initial
energy, target composition, etc. was determined by the SRIM 2008
computer code. In result, Ti-Si-C coatings were formed with two
Ar
+
ion beams of the initial energy ~15 keV. For The angle between
sputtering ion beam axis and normal to the target surface was
fixed at 67
0
. TiSiC thin films were grown on AISI 316L austenitic
steel substrates (Fe, 0.02% wt. C, 18% wt. Cr, 10% wt. Ni,
2% wt.Mo). A single Ti
3
SiC
2
compound target was used, produced
by the SHS method at the Department of Advanced Ceramics AGH
(Krakow, Poland) from stoichiometric mixtures of Ti, Si and C.
Steel substrates were prepared by cutting from 2 mm thick plate
samples (Goodfellow, UK) using a Struers Minitom cut-off wheel
and cBN cutting disc, then mechanically ground and polished using
Struers MD system working with diamond grinding and polishing
suspensions.
2.2. Films characterization
The microstructure, chemical and phase composition of coated
substrates were examined by scanning (SEM, EDS) and
transmission (TEM, SAED) electron microscopy methods using a
JSM-6460LV with EDS INCAX-act 350 Oxford Instrument and
Tecnai G2 F20 (200kV). For TEM analysis, cross-sections of coated
substrates were prepared by FIB method. Phase composition of as-
deposited samples was analyzed by XRD using Siemens D500
Kristalloflex diffractometer, operated with monochromatic Cu K

88
(=1,54) radiation in grazing incident configuration (=1 and 3)
and XRD patterns were collected using a step size of 0,02 over 2
angular range of 20 to 125. The hardness and reduced elastic
modulus of coated and uncoated substrates were determined by
nanoindentation test. CSM with OM imaging system was used
working with a Berkovich-type diamond indenter under load of
10 mN. Five indentations were made in three regions of each
sample. Load- displacement curves were recorded and hardness H
IT
and reduced elastic modulus E
IT
were calculated using the Oliver &
Pharr method.
3.1. Chemical and phase composition
The chemical and phase characterization of formed coatings was not
an easy task due to films thickness and limitation of EDS and XRD
methods. EDS analysis accompanied both SEM and TEM
observation from the coating surface and cross-section respectively.
EDS analysis results taken from the surface of coated substrate at
area marked in Fig.4 are presented in Tab.2.

Table 2. EDS analysis results taken from marked area (Fig.4) of the surface of Ti-
Si-C coating, formed by IBAD method on AISI 316L substrate.
area C [%wt.] Si [%wt.] Ti [%wt.] total
spectrum 1 10.37 21.84 67.79 100
spectrum 2 55.19 19.67 25.14 100

In EDS (EDX) analysis accompanying TEM examination of the
coating cross-section Ti to Si concentration ratio was determined as
70 to 30 (in %wt.). Low temperature deposition of TiSiC films
(temperature less than 300C) from compound Ti
3
SiC
2
target effects
usually in a nanocomposite or nanocrystalline structure, with the
presence of TiC
x
phase in amorphous matrix, enriched in carbon

89
content [5, 6]. In our films, formed by ion beam assisted deposition,
carbon content is lowered. XRD analysis did identify neither TiC
nor TiC
x
. The main crystalline phase was Ti
5
Si
3
which coexisted
with TiSi and TiSi
2
in amorphous TiSi matrix. The presence of
Ti
3
SiC
2
was not evident, due to partial correspondence of registered
diffraction patterns with characteristic data for that compound.
3.2. Microstructure
Scanning electron microscopy was used to examine the surface of
coated substrates. As shown in Fig.4a, IBAD formed Ti-Si-C
coating surface was dense, flat and smooth. Some increase in the
roughness of the surface was observed with film thickness. TEM
observation of formed films were provided in plain view and on
coated substrate cross-section prepared by FIB method, cut
perpendicular to the coating surface. TEM examination confirmed
that coatings were continuous and uniform (Fig.4b). The thickness
of the coating was in the range 150 nm to ~1 m, depending on
IBAD process parameters. Film to substrate boundary was
continuous and smooth. Selected area electron diffraction pattern,
presented in Fig.4c, was taken from the film area and confirmed that
examined coating was amorphous.

90
3.3. Mechanical Properties
Vickers hardness and reduced elastic modulus were determined by
nanoindentation tests. Berkovich-type diamond, indenter was used
working under load of 10mN. Load- displacement curves were
collected for coated and uncoated substrates. The Oliver & Pharr
method was used for calculation of the hardness H
IT
and elastic
modulus E
IT
. The hardness and reduced elastic modulus for
uncoated substrates were H
IT
= 4.4 GPa and E
IT
= 250 GPa,
respectively. For coated substrates, calculated hardness was 4 GPa,
and calculated reduced elastic modulus E
IT
was 160 GPa. In
comparison to CVD derived Ti
3
SiC
2
Vickers hardness of about 6
a)

b)

Fig.4. Microstructure of Ti-Si-C coating
formed by IBAD technique:
a) SEM image of coating surface with
marked area for EDS analysis,
b) TEM image (plain view) with SAED
of the coating,
c) TEM image of coated AISI 316L
substrate cross-section
c)


91
GPa in the 0.989.8 N load [7], for polycrystalline bulk Ti
3
SiC
2
with
10%vol. TiC
x
Vickers hardness was 12.7 GPa, for indentation load
of 49 mN and 4 GPa for 9.8 N [8], while calculated Youngs
modulus were referred as 283 GPa [8] or even 326 GPa for bulk
Ti
3
SiC
2
were referred [9]. Such decrease in the hardness and elastic
modulus values was reasonable for amorphous state of formed
coatings and was often referred as 50 to 60% of values that
respective crystalline state [10].
Summary
Ion beam assisted deposition process parameters allowed to form,
in low temperature, continuous, dense and flat Ti-Si-C films on
AISI 316L steel substrates, by use of single Ti
3
SiC
2
target.
Low temperature ion beam assisted deposition of Ti-Si-C results in
full or partial film amorphisation, which is desirable effect for
possible film application as electrical contacts [11].
Amorphous Ti-Si-C films formed by IBAD are soft: calculated
hardness H
IT
2.7 GPa to 5.3 GPa and the average value of
calculated reduced elastic modulus E
IT
160 GPa.
Acknowledgements
The work was supported by the Polish Ministry of Science and
Higher Education under project No. N507 451434.

REFERENCES

[1] Jeitschko W., Nowotny H., Monatsh. Chem. 98 (1967) 329.
[2] Barsoum M.W., Prog. Solid State Chem. 28, 201 (2000).
[3] Palmquist J.P., Jansson U., Seppanen T., Persson P.O.A., Birch J., Hultman
L., Isberg P., Appl. Phys. Lett. 81 (2002) 835.
[4] Emmerlich J., Palmquist J.P., Hogberg H., Molina-Aldareguia J.M.,
Czigany Zs., Sasvari Sz., Persson P.O.A., Jansson U., Hultman L., J. Appl.
Phys. 96 (2004) 4817.

92
[5] Koutzaki S. H., Krzanowski J. E., Nainaparampril J. J., J. Vac. Sci.Technol.,
A 19 (2001) 1912.
[6] Zehnder T., Matthey J., Schwaller P., Klein A., Steinmann P.A., Patscheider
J., Surface & Coatings Technology 163164 (2003) 238.
[7] Goto T., Hirai T., Mat. Res. Bull. 22 (1987) 1195.
[8] Gao N. F., Miyamoto Y., Zhang D.: Journal of Materials Science 34 (1999)
4385 4392.
[9] Pampuch R., Lis J ., Piekarczk J ., Stobierski L., Ceram. Inter. 19 (1993)
219.
[10] Wachtman J. B., Haber R. A.: Ceramic Films And Coatings, Noyes
Publications Mill Road, Park Ridge, New Jersey, USA, 1993.
[11] Eklund P. et al. Thin Solid Films 518 (2010) 1851187.

Ionic methods of coatings formation for a special applications

93
IONIC METHODS OF COATINGS
FORMATION FOR A SPECIAL
APPLICATIONS

BOGUSAW RAJCHEL
1

Abstract

Carbon based coatings are frequently applied to improve of the physicochemical and
mechanical properties of covered tools, parts of machines or artificial medical
implants. For all those applications the adhesion of coatings to covered substrate
must be especially good. The physicochemical properties of carbon coatings are a
function of the sp2/sp3 bonds concentrations. The physicochemical properties and
adhesion of carbon coatings are strongly defined by method used to their formation.
Especially two ionic methods, IBSD Ion Beam Sputter Deposition and IBAD Ion
Beam Assisted Deposition, can be applied to formation of carbon coatings with
good adhesion and with controlled sp2/sp3 bonds concentrations.
1. Introduction
Beam of ions, can be used for modification of a thin surface layer or
for formation of the flux of energetic atoms and ions. For this
reasons, are existing two fundamental ionic methods. First is the ion
implantation and the second is the IBSD (Ion Beam Sputter
Deposition). The first one is used for modification of the surface
layer and second one is used for creation of coatings. All another,
such as the IBAD Ion Beam Assisted Deposition, are based on
those methods. The main idea of the IBSD and the IBAD techniques
is shown on Fig. 1.

1
Institute of Nuclear Physics Polish Academy of Sciences., 152 Radzikowskiego
Str, 31-342 Krakow, Poland.


94
IBSD IBAD

Fig. 1. The IBSD (left) and the IBAD (right) techniques for formation of complex
coatings.
In the method IBSD, the primary problem is to obtain a flux of
atoms and ions, with optimal energetic and angular distributions for
forming the coating with expected nanostructure. In this method, a
well-defined ion beam is used for bombarding a material with a
suitably chosen elemental composition. As a result of binary
collisions of ions with nuclei the flux of atoms and ions is formed.
This process is called as ion sputtering. The sputtering process was
described with details by Eckstein [1].
2. Sputtering
Sputtering is one of the processes occurring during ion beam
bombardment of solid surfaces. The sputtering process, taking
place during the bombarding at an angle by beam of ions
with atomic number Z
1
, mass M
1
and energy E
0
of an
amorphous material composed from atoms with atomic
number Z
2
and mass M
2
is shown in Fig.1.
The coating is formed from atoms and ions, sputtered by ions beam
into a solid angle d defined by the surface to be covered.
Nanostructure and physicochemical properties of coatings, depend
on distributions of energies and angles of sputtered products. The
thickness of the coating depends on the number of atoms emitted

95
into the solid angle. which is defined by the geometry of this
process. This number is determined, by the number of bombarding
ions and by the so-called sputtering coefficient. This coefficient
depends on ions beam parameters and composition of the sputtered
material.
Sputtering coefficient Y(E
0
, ) can be expressed as [1]:

( ) ( ) ( ) f E Y E Y 0 , ,
0 0
= = (1)

where
( )
f
c
f
(
(
|
|
\
|
2
cos
0

(2)
The Y(E
0
, = 0) can be calculated as

( ) ( )
th
E E F E Y , 0 ,
0 0
= =
(3)
Where the E
th
is a threshold energy for sputtering.
The sputtering coefficient can be calculated by formulae [1] [3],
from tables [2] or by computer codes used to simulation of
interaction of energetic ions and solid materials. This interaction can
be simulated by the Monte Carlo methods or by the molecular
dynamics techniques. Frequently, to simulation of interaction of ions
and amorphous material, the SRIM computer code [3] is used. The
SRIM 2008 code, based on the Monte Carlo mechanism, can be
used to calculate of ion ranges and sputtering coefficients.
In this paper the SRIM 2008 code was applied to calculate of
sputtering coefficients of carbon, titanium and silicon from
amorphous Ti
3
SiC
2
bombarded by Ar
+
or Xe
+
ions at energy of 15
keV. All obtained coefficients are shown in the Table 1.


96
Table 1. Sputtering coefficients [atoms/ions] of carbon, titanium and silicon
calculated for amorphous Ti
3
SiC
2
sample bombarded by Ar
+
or Xe
+
ions at energy
of 15 keV.
Ar
+
Xe
+
Impact
angle Carbon Titanium Silicon Carbon Titanium Silicon
0
0
0.3739 1.08 0.2956
10
0
0.331 0.9742 0.2999 0.4081 1.21 0.3757
20
0
0.3737 1.09 0.3397
30
0
0.4784 1.37 0.4275
40
0
0.6566 1.88 0.5807
50
0
0.9509 2.72 0.837
60
0
1.55 4.34 1.15
70
0
2.03 5.86 1.80
80
0
2.50 7.06 2.20

The nanostructure of the formed coating is determined by
distributions of energies and angles of atoms after sputtering. Those
distributions can be calculated as product of two independent
distributions:
( ) ( ) ( ) g E g E g
s s
= , (4)
Distributions of energies and angles can be also calculated in
simulation of interaction of ions with materials.
3. Experimental results
The IBSD and the IBAD techniques were used for creation of thin
Ti
X
SiC
Y
and carbon coatings. The Ti
X
SiC
Y
were formed on the
316L steel. The carbon coatings were formed on the polyurethane
substrates.
For formation of the Ti
X
SiC
Y
the beam of Ar
+
ions at energy of 15
keV was used to sputtering of the Ti
3
SiC
2
target. The impact angle
was 67
0
.
For formation of carbon coatings as sputtering target the graphite or
aluminium plate were used.
All formed coating were investigated by the confocal dispersive
Raman microspectroscopy. The Raman spectrum obtained for the

97
Ti
X
SiC
Y
sample is shown on the Fig. 2a. The Raman spectrum
obtained for the carbon coating is shown in the Fig. 2b.
Ti
X
SiC
Y
Carbon

Fig. 2. Raman spectra obtained for the Ti
X
SiC
Y
and carbon coatings. For
experiment the 532nm laser light was used.
The 3D depth profile of the carbon coating formed on the Al
substrate is shown on Fig.3.

Fig. 3. The 2D depth profile of the amorphous carbon (a-C) coatings formed on the
Al substrate.
Numerical analysis of the Raman spectrum, collected for the
Ti
X
SiC
Y
coating, shown peaks characteristic for the Ti
3
SiC
2
phase.
The depth profiles, obtained for the carbon coatings, shown a thin
interface sublayer.
substrate
coating
G and D peaks

98
4. Discussion
Sputtering is one of the processes occurring during ion beam
bombardment of solid surfaces. The sputtering process, taking place
during the bombarding at an angle by beam of ions with atomic
number Z
1
, mass M
1
and energy E
0
of an amorphous material
composed from atoms with atomic number Z
2
and mass M
2
is shown
in Fig.1.
5. Summary
The IBSD and the IBAD techniques are suitable for formation of
complex, multi-elemental and multi-layer coatings on any type
substrate. All coatings formed by those ionic methods have very
good adhesion to protected surface.
6. Acknowledgements
The work was supported by the MNiSW project, contract no. Nn507
451434 and by the project SPO WKP 1.4.3, Nanotechnology for
special applications in industry and medicine.
REFERENCES
[1] Eckstein W., Sputtering yields, Vacuum, 2008, 82, 930-934
[2] Yamamura Y, Tawara H., At. Data Nucl. Data Tables 1996, 62, 149
[3] www.srim.org

Nano Materials

Hochgeladen von

Dokumentinformationen

Originalbeschreibung:

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Nano Materials

Hochgeladen von

Copyright:

Verfügbare Formate

NANOMATERIALS

(x) are dimensionless and of order 1, i.e., S

(x) are twice continuously differentiable.

(x) = S(x), (3.4)

(x) = S(x), these formulas take a

>> 1). Then

PYT05.0 (ZY5) produced by Unitec Ceramics, England, with an

Das könnte Ihnen auch gefallen