Analysis and characterization of surface oxides on intermetallic alloys of zirconium using auger
electron spectroscopy
by Rajesh G Mirpuri
A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in
Chemical Engineering
Montana State University
© Copyright by Rajesh G Mirpuri (1991)
Abstract:
The surface composition of oxidized zirconium and nickel and their oxidized intermetallics, ZrNi3,
ZrNi, ZrNi5 and Zr2Ni7 are evaluated. This was performed following the installation and set-up of an
Auger electron spectroscope and development of relevant software for both digital process control and
data analysis.
Surface oxide layers formed on the samples under two different conditions, room temperature and at
300°C for I hour were analyzed. Depth profiles were performed using Auger electron spectroscopy and
argon ion sputtering. The zirconium and nickel concentrations increased and the oxygen concentration
decreased with increase in sputtering time. Longer sputter times were noticed in the case of the samples
oxidized at 300°C indicating thicker oxide layers. The oxide layer formed at 300°C on zirconium was
thicker than the oxide layer on nickel at 300°C since the former required 90 minutes of sputtering
compared to 4 minutes of sputtering for the latter. The oxide layer formed on zirconium at 300°C is
shown to have a ZrO2 stoichiometry.
The sputter depth profiles on the intermetallics are consistent with earlier work reported in the
literature, with a slight discrepancy in the steady state final intensities of the zirconium and nickel
peaks noticed in the case of the intermetallics oxidized at 300°C. This can be attributed to a change in
the primary Auger electron beam energy from 3 keV to 2 keV which diminished the peak intensities
significantly. Surface cracks on the intermetallic surfaces which are laterally oxidized during high
temperature oxygen exposure may have also resulted in the attenuation of the nickel and zirconium
peak intensities and the increase in oxygen peak intensity. ANALYSIS AND CHARACTERIZATION OF SURFACE OXIDES ON
INTERMETALLIC ALLOYS OF ZIRCONIUM USING
AUGER ELECTRON SPECTROSCOPY
by
Rajesh G . Mirpuri
A thesis submitted in partial fulfillment
of the requirements for the degree
of
Master of Science
in
Chemical Engineering
MONTANA STATE UNIVERSITY
Bo zeman, Montana
May 1991
APPROVAL
of a thesis submitted by
Rajesh G. Mirpuri
This thesis has been read by each member of the thesis
committee and has been found to be satisfactory regarding
content, English usage,
format, citations, bibliographic
style, and consistency, and is ready for submission to the
College of Graduate Studies.
i L m /
Approved for the Major Department
V>7
-C-
' Llll/
Dhte
d, Major Department
V
Approved for the College of Graduate Studies
/f?/
Date
Graduate Dean
iii
STATEMENT OF PERMISSION TO USE
In presenting this thesis in partial fulfillment of the
requirements
University,
for
a
master's
degree
at
Montana
State
I agree that the Library shall make it available
to borrowers under rules of the Library.
Brief quotations
from this thesis are allowable without special permission,
provided that accurate acknowledgement of source is made.
Permission for extensive quotation from or reproduction
of this thesis may be granted by my major professor,
or in
his/her absence, by the Dean of Libraries when, in the opinion
of either, the proposed use of the material is for scholarly
purposes.
Any copying or use of the material in this thesis
for financial gain shall not be allowed without my written
permission.
Signature
Date
,UAjis I l ^ I cIaI I
iv
TABLE OF CONTENTS
Page
A P P R O V A L ........... ................................. . . ..
ii
STATEMENT OF PERMISSION TO U S E ..... , . . ................
iii
TABLE OF CONTENTS .......................................
iv
LIST
vi
LIST
OF TABLES ...................................
OF F I G U R E S ........................
A B S T R A C T .................................................
I .I N T R O D U C T I O N ..........................
vii
ix
I
2 .B A C K G R O U N D .............................................
Research Objectives ................................
6
10
3. OXIDATION OF METALS AND ALLOYS ...... ............ ;. . .
Low Temperature O x i d a t i o n .........
Diffusion in Oxides ............................. '..
Lattice Defects ...................................
Lattice Diffusion ................................
Diffusion Mechanisms ............................
High Temperature Oxidation o f Metals .............
Compact Oxide Scales ............................
Porous Oxide Scales .............................
High Temperature Oxidation of Metals & Alloys ....
H
12
13
14
14
16
17
17
21
22
4. EXPERIMENTAL M E T H O D .............. '...................
Auger Electron Spectroscopy .......................
Experimental S y s t e m ............ ................... '
Ultra-High Vacuum Systems .......................
Ion G u n ...........................................
Sample Mount System .............................
Energy A n a l y z e r .................................. Sample Preparation ...'...... <......................
Surface P r e p a r a t i o n ...............................
Sample Alignment ..............
Elastic Peak Procedure ............................
■
Acquiring AES D a t a ..............................
Shut Down P r o c e d u r e ................................
26
26
29
29
30
32
36
38
39
40
41
41
41
5. RESULTS .............. .................................
42
6. INTERPRETATION
68
V
TABLE OF CONTENTS-CONTINUED
Page
7. SUMMARY AND C O N C L U S I O N S .............................
80
8. RECOMMENDATIONS FOR FURTHER RESEARCH ................
82
R E F E R E N C E S ...............................................
84
vi
LIST OF TABLES
Table
1. Steady State Intensity Profile Data of Zrl peak for
the metal and intermetallie samples analyzed ..........
Page
71
2. Steady State Intensity Profile of Nil and Ni2 peaks for
the metal and intermetallie samples analyzed ........... 73
3. The sputter time required to remove the oxide layer
for the metal and intermetal lie s a m p l e s ................ 78
vii
LIST OF FIGURES
Figure
Page
1.
Wagner Schematics for Oxidation of Alloys ...........
19
2.
Wagner Schematics for Oxidation of Metal Alloys ......
25
3.
Energetics of the Auger p r o c e s s ......................
27
4.
AES Analytical E q u i p m e n t .............................
31
5.
Sample Holder for AES A n a l y s i s .......................
33
6.
Relative Orientation of Sample Mount
with Ion Gun and Electron G u n .........................
35
7.
The Electron Gun and CMA A r r a n g e m e n t ................
37
8.
AES Survey of Zirconium Foil .........................
43
9.
AES Survey of Sputter Cleaned Zirconium
44
............
10. Expanded energy region of Sputter
Cleaned Z i r c o n i u m ..................................
45
11. Multiplexed Zr2 Peak on Zirconium Foil ..............
47
12. Multiplexed Zr2 Peak on Sputter Cleaned Zirconium ...
48
13. Multiplexed Zrl Peak on Sputter Cleaned Zirconium ...
49
14. Multiplexed Oxygen Peak onZirconium .................
51
15. Sputter Depth Profile for Z r ...................... . . .
52
15.
53
Sputter Depth
Profile
for Z r (ox) ................
17. AES Survey of Nickel F o i l .......................
54
18. AES Survey of Sputter Cleaned Nickel ...........
56
19. Multiplexed Nil Peak on Sputter Cleaned Nickel .....
57
20. Multiplexed Ni2 Peak on Sputter Cleaned Nickel .....
58
21. Sputter Depth Profile for N i ..........................
59
22. Sputter Depth Profile for Ni (ox) ..........
60
viii
LIST OF FIGURES - CONTINUED
Figure
Page
23. Sputter Depth Profile for ZrNiS ......................
61
24. Sputter Depth Profile for Zr2Ni7 .........•...........
63
25. Sputter Depth Profile for ZrNi3 ......................
64
26. Sputter Depth Profile for Z r N i .......................
65
27. Sputter Depth Profile for ZrNiS(ox) .................
66
28. Sputter Depth Profile for ZrNi3(ox)
.................
67
29. Steady State Intensity Profile for Zr ...............
72
30. Steady State Intensity Profile for Ni ................
74
31. An Optical Microscope Photograph of Polished
ZrNi3 after Oxidation at 3 0 0 ° C .......................
76
ix
ABSTRACT
The surface composition of oxidized zirconium and nickel
and their oxidized intermetallics, ZrNi3, ZrNi, ZrNi5 and
Zr2Ni7 are evaluated.
This was performed following the
installation and set-up of an Auger electron spectroscope and
development of relevant software for both digital process
control and data analysis.
Surface oxide layers formed on the samples under two
different conditions, room temperature and at 3OO0C for I hour
were analyzed. Depth profiles were performed using Auger
electron spectroscopy and argon ion sputtering. The zirconium
and
nickel
concentrations
increased
and
the
oxygen
concentration decreased with increase in sputtering time.
Longer sputter times were noticed in the case of the samples
oxidized at 300°C indicating thicker oxide layers. The oxide
layer formed at 3OO0C on zirconium was thicker than the oxide
layer on nickel at 3OO0C since the former required 90 minutes
of sputtering compared to 4 minutes of sputtering for the
latter. The oxide layer formed on zirconium at 300°C is shown
to have a ZrO2 stoichiometry.
The sputter depth profiles on the intermetallics are
consistent with earlier work reported in the literature, with
a slight discrepancy in the steady state final intensities of
the zirconium and nickel peaks noticed in the case of the
intermetallics oxidized at 300°C. This can be attributed to a
change in the primary Auger electron beam energy from 3 keV to
2 keV which diminished the peak intensities significantly.
Surface cracks on the intermetallic surfaces which are
laterally oxidized during high temperature oxygen exposure may
have also resulted in the attenuation of the nickel and
zirconium peak intensities and the increase in oxygen peak
intensity.
I
CHAPTER I
INTRODUCTION
The
development
of
metal
alloys
resistance to surface oxidation,
cost, durability
and
utility
which
have
improved
has a major impact on the
of
equipment
and ,components
utilized in high temperature and other corrosive environments.
The
useful
structures
conditions
life
and
can
of
metallic
electrical
also
be
components
systems
incorporated
operated
significantly
under
extended
into
ambient
through
appropriate alloying to improve their resistance to oxidation.
Thus a more fundamental in-depth knowledge of improved metal
alloys and composites,
resistance
and
high
which demonstrate enhanced corrosion
temperature
stability,
would
greatly
benefit modern industry.
The reaction of oxygen with metal alloy surfaces is such
that it can lead to the formation of a cohesive surface oxide
layer,
which
protects
the
alloy
from
further
significant
oxidation. An in-depth study on this reaction would help in
the development of metal alloys especially suited to a large
range of future and current applications requiring oxidation
resistance.
Extensive
literature[1,2,3]
on
the
methods
of
relating surface oxidation mechanisms and oxide structures of
pure metals provides the basis for the design and conduct of
this investigation on the surface oxidation of metal alloys.
2
By enhancing the protective surface oxide layer,
which
can be either porous or compact, improved oxidation resistance
is usually realized in metals and alloys. Compact oxide layers
offer the best protection against high temperature oxidation
because they form a nonporous barrier between the unreacted
metal and oxidizing environment. Thus oxidation resistance in
alloys in which at least one of the components combine readily
with
which
oxygen
are
b u l k [4].
can be
effective
generally
barriers
attributed
to
to
further
surface
oxidation
layers
of
the
On oxidation of a metal alloy, where the base metal
oxidizes rather rapidly,
the base metal oxide forms on the
outer scale and the oxide of the alloying element forms on the
inner scale[2,3].
oxide
layer
can
The ratio of the components in a surface
be
different
from
surface oxide thickness keeps
that
in
the
increasing until
bulk.
The
it inhibits
further growth. Cracking or flaking of these oxidized layers
leads to the loss of their protective property.
If the oxide
layer cracks, then the metal alloy can return to its initial
stage and the oxidation rate increases. On the contrary,
if
the resulting oxide layer remains intact and is cohesive,
it
will
eventually
permeability,
any
resist,
further
by
virtue
significant
of
reduced
growth
in the
oxygen
oxide
layer thickness.
The
indicates
layer
Wagner
oxidation
model
that the growth rate
decreases
with
time
for
metal
of a compact
and
increasing
alloys[2,3]
surface
oxide
oxide
layer
3
thickness. When a clean metal alloy is exposed to an oxidizing
atmosphere,
initial reaction kinetics controls the extent of
oxide
formation of each of the metallic components.
oxide
layer
forms
on
the
surface
of
the
alloy,
Once a
further
oxidation is limited by the diffusion of the oxygen through
the oxide layer. Concentration gradients develop in the metal
near the metal/oxide interface. The rate of oxygen diffusion
into the alloy will depend in part on the concentrations of
the metal oxides resulting from the surface oxidation reaction
and
partly
developing
on
the
oxide
dif fusivities
layer
and
the
of
oxygen
metallic
in
both
elements
in
the
the
underlying reduced alloy.
The
surface
oxidation
characteristics
of
the
intermetallic alloys of zirconium make an excellent choice as
a subject for investigation because of zirconium's industrial
importance and the fact that it has received little previous
attention in this area. Zirconium is used extensively[5]
in
the nuclear power industry as a major constituent of various
fuel cladding alloys that separate the fuel from coolant water
in atomic reactors. The most widely used alloys, zircaloys, as
they are commonly known, consist of zirconium, together with
tin,
chromium,
better
iron, nickel etc. which are added to promote
corrosion
resistance.
An
important
property
of
zirconium is its low capture cross section for neutrons i.e.,
neutrons pass freely through the internal zirconium structure
without
an
appreciable
energy
loss
and
a
corresponding
4
temperature rise in the cladding material[6] . Zirconium based
fuel
cladding
alloys
are
resistant
to
the
corrosive
environment inside the atomic reactors.
The corrosive resistance of zirconium based alloys is due
to the formation of thin, protective, cohesive surface oxide
layers. Extensive w o r k [6] has been carried out on the kinetics
of bulk oxidation
temperatures
of zirconium and
under
different
its alloys
environments
at elevated
including
air,
water, steam and oxygen.
The primary method of analyzing the composition of oxide
surface layers in this study is Auger Electron Spectroscopy
(AES)
which
composition
alloys[7].
lends
itself
of
the
surface
One
of
the
to
investigating
strata
most
of
clean
interesting
the
and
atomic
oxidized
applications
of
quantitative AES analysis is the depth profiling technique, in
which surface atoms are removed by ion bombardment with A E S [8]
analysis
of
equipment
surface
used
composition
in AES
depth
at
sequential
profiling
is
depths.
equivalent
The
to
a
standard Auger spectrometer in combination with an ion gun.
The ion gun is used to bombard the surface atoms of the sample
at a relatively slow rate,
and remove surface atoms almost
atomic
a
carried
layer
out
by
layer.
during
If
quantitative
ion bombardment,
the
AES
analysis
results
yield
is
the
composition of the sample at different depths.
Single phase metals are a better choice for a surface
study of the present type even though most important alloys
5
consist primarily of polyphase composites of solid solutions
and intermetallic compounds( ! C s ) . The main drawback in using
polyphase alloys is that spectroscopic information gathered
after the oxidation of these alloys is difficult to interpret
due
to
the
probable
presence
of
more
than
one
surface
phase[l]. This complication can be circumvented by utilizing
alloy compositions corresponding to intermetallic compounds
which are multi-element single phase metals. The information
thus
developed
can
then
be
used
to
model
the
oxidation
resistance of composite alloys from an improved understanding
of the surface oxidation properties of their individual solid
phases.
6
CHAPTER 2
BACKGROUND
Techniques of Electron Spectroscopy,
Electron
Spectroscopy,
have
been
especially.
employed
to
Auger
study
the
interaction of oxygen with various metal surfaces. The types
of
metallic
surfaces
intermetallic
analyzed
compounds
in
include
both
pure
single
metals
crystal
and
and
polycrystalline form and composite alloys of intermetallics
and solid solutions. These materials have been subjected to
wide
ranges
of
temperatures,
pressures
and
exposures to various oxidizing atmospheres.
durations
of
Several surface
analytical methods have been used in combination with argon
ion milling
for depth profiles.
The most
extensively used
methods are the Secondary Ion Mass Spectroscopy (SIMS) and the
Electron Spectroscopy methods,
in particular Auger Electron
Spectroscopy(AES) and X-ray Photoelectron Spectroscopy (XPS).
The oxidation behavior of both single and polycrystalline
samples of zirconium have been investigated. It has been shown
that
a
cohesive
surface
oxide
layer,
which
stoichiometry is formed on bulk zirconium
[4].
has
a
ZrO2
The initial
exposure of zirconium to oxygen results in the.formation of a
nuclei
of
surface
completion at an O2
oxide
which
grow
to
coalescence
exposure of 3 0 to 60 L(1 L =
and
IO"6 Torr.s)
and has a thickness of approximately 2 nm at room temperature.
7
Since this oxidation proceeds slowly there is a transition
region of intermediate oxides between the ZrO2 layer and the
bulk Zr [I].
The
formation
protects
the
of
the
underlying
thin
metal
ZrO2 surface
from
layer
further
on
Zr
significant
oxidation at 300 K, since the diffusion of O2 through ZrO2 is
very slow. However at higher temperatures, oxygen permeation
into the bulk metal
measurable
rates
from the surface oxide layer occurs at
due
to
increased
oxygen .diffusivity
and
solubility in Zr.
J.S. Foord et a l ., studied the adsorption of D2, CO, N2,
NO and O2 on Zr at 300 K [9]. Work function, electron impact,
Auger,
and
adsorbed
diffusion
phases
data
leads
indicated that
to 'rapid
surface
heating
to
of these
bulk
oxygen
diffusion and very little desorption.
A
related
study
[10]
conducted
by
P.Sen
et
al.,
investigated the surface oxidation of three metglasses in the
copper-zirconium
system
by
employing
X-ray
Photoelectron
Spectroscopy and Auger Electron Spectroscopy to compare their
oxidation behavior with that of the corresponding crystalline
states
of the alloys.
The study was conducted over a wide
range
of
temperatures,
oxidation
pressures
and
alloy
compositions. It was shown that both glassy and crystalline
states of the Cu-Zr
alloys exhibit preferential oxidation at
intermediate oxygen exposures.
8
The surface oxidation of pure zirconium and its dilute
binary
alloys
investigated
by
with
tin,
XPS.
In
chromium
this
and
study
iron
Lalit
has
Kumar
et
been
a l .,
compared the oxidation behavior of zirconium and its alloys at
room temperature [11]. It was observed that mostly suboxides
of Zr are formed during the initial stages of oxidation at
oxygen exposures less than 10 L, while at higher exposures,
ZrO2 is
the
dominant
oxide
suboxides are present.
species
formed,
although
two
Pure Zr as well as its dilute alloys
exhibit a decreasing rate of oxidation with increasing oxygen
exposure.
Bertolini
et
al. ,
analyzed
the
surface
metallic
properties of Ni63 7Zr36 2 alloy by Auger electron spectroscopy
and photoelectron spectroscopy in combination with argon ion
bombardment [12]. The elemental composition as a function of
depth below the surface was analyzed. A thin ZrO2 oxide layer
was observed to extend up to 15-35 A below the outer surface
beyond which
an
abrupt
interface with
the Ni63 7Zr36 2 alloy-
existed.
Several
studies
of
the
surface
oxidation
behavior
of
nickel-zirconium intermetallic compounds have been conducted.
Wright,
Cocke and Owens conducted a detailed study
[13]
on
five Ni-Zr intermetallic compounds, Zr2Ni, ZrNi, ZrNi3, Zr2Ni7
and ZrNi5. The analysis included sputter cleaning of the metal
alloys by argon ion milling followed by exposure to sequential
treatments of hydrogen,
oxygen,
and hydrogen a second time.
9
These experiments were carried out at room temperatures and at
4OO0C.
The depth
concentration profiles were
then
analyzed
using high resolution AES. The results showed that the surface
was covered by ZrO2 after exposure to air at room temperature
and covered by NiO after high temperature oxidation at 4 OO0C .
The ratio at the surface of Ni to Zr was always greater after
oxidation than after reduction. This indicates a preferential
oxidation of Zr over Ni when both are present in a reduced
form and the outward diffusion of Ni cations through the oxide
strata at 4OO0C during oxidation.
Deibert and Wright investigated
[14]
two of the above
mentioned intermetallics, ZrNi3 and Zr2Ni7, by high resolution
Auger Spectroscopy and Ar ion sputtering. The intermetallics
were
subjected to sequential hydrogen,
oxygen and hydrogen
treatments. The results indicate an essentially pure reduced
Ni surface on a layer of oxidized zirconium, which in turn is
supported
on
constituted
a
the
unaltered
highly
active
alloy.
The
catalyst
reduced
surface
for
nickel
various
reactions of possible industrial relevance. The thickness of
the reduced Ni layer is estimated to be between 10 and 20 nm.
In another series of experiments by Deibert and Wright
[15], the initial surface oxidation of sputter cleaned nickelzirconium intermetallics (ZrNi5, Zr2Ni7, ZrNi3, ZrNi, ZrNi3) at
room
temperature
analysis
and
surfaces
of
was
argon
the
investigated
ion
alloys
by
sputtering.
were
shown
quantitative
The
to
be
Auger
sputter-cleaned
enriched
in
Zr
10
relative to their bulk,
relatively dilute
especially for the alloys which are
in Zr.
The experimental
results
indicate
that the ratio of the sputtering rate of Ni to that of Zr is
dependent on the composition of the alloy.
Research Objectives
*
Installation
Spectroscope
and
set-up
of
the
PHI
Auger
Electron
and development of relevant software
for both
digital process control and data analysis.
*
Study
intermetallic
of
the
surface
compounds
of
compositions
zirconium
and
of
the
nickel
oxidized
including
ZrNi3, ZrNi, ZrNi5 and Zr2Ni7 and oxidized zirconium and nickel
metal foils.
*
Develop an improved understanding of the surface oxidation
of the zirconium alloys at 3OO0C.
*
Compare the results of this study with recent studies of
Zr-Ni intermetallics, with particular emphasis on oxidation of
the intermetallics.
11
CHAPTER 3
OXIDATION OF METALS AND ALLOYS
The total chemical equation for the reaction between a
metal and oxygen gas is:
aMe+^
Though this appears to be a simple equation, the reaction path
and the oxidation behavior of a metal depends on a variety of
factors. The metal oxidation reaction mechanisms, as a result,
are complex.
The initial step in the metal-oxygen reaction involves
the adsorption of gas on the metal surface. As the reaction
proceeds,
oxygen may dissolve
in the metal;
then oxide
is
formed on the surface as an oxide layer. Both the adsorption
and
the
initial
orientation,
preparation,
oxide
crystal
and
formation
defects
impurities
are
at
functions
the
of .surface
surface,
in both the metal
surface
and the gas
[16] .
The surface oxide separates the metal and the gas. When
a
compact
film
covers
the
surface,
the
reaction
proceeds
through a solid-state diffusion of the reactants through the
film. For thin films the driving force for this transport of
reactants
is
mainly
due
to
electric
fields
(migration
of
charged imperfections) in or across the film. For thick films
12
or scales it is determined by the chemical potential gradient
across the scale [17].
Metals may also form porous oxide scales which as such do
not
serve
as
a
reactants. In
solid-state
such
cases
diffusion
the
barrier
reaction
may
between
be
the
limited
by
processes occurring at phase boundaries. At high temperatures,
oxides may also be liquid or volatile.
The overall driving force of metal-oxygen reactions is
the free energy change associated with the formation of the
oxide from the reactants. Thermodynamically the oxide will be
formed only if the ambient oxygen pressure is larger than the
dissociation pressure of the oxide
metal
in equilibrium with its
[16].
Low-Temperature Oxidation
The
initial
step
in the reaction between
a metal
oxygen is the adsorption of gas on the metal surface.
and
Then
isolated oxide nuclei form at random positions on the surface.
After
formation
of
the
random
oxide
nuclei
the
oxidation
proceeds through a growth of the individual crystallites until
the whole surface is covered with oxide. The rate of reaction
during the initial stage increases with time;
sticking
coefficients
of
oxygen
increases
that is,
as
oxygen
the
is
consumed. This increased activity is due to an initial oxide
nucleation
at
through
lateral
a
preferred
sites
surface
and
growth
the
reaction
of
the
proceeds
nuclei
[16].
13
Dislocations,
surface
defects,
impurities,
etc., serve
as
nucleation sites. The adsorbed oxygen diffuses on the surface
to the growing nuclei, causing an oxygen depletion in a zone
around each nucleus in which no other nuclei form. The size of
these zones is governed by the supply of adsorbed oxygen and
the rate of surface migration [18].
Diffusion in Oxides
When a dense, compact oxide film or scale has been formed
on a metal surface, the oxide separates the reactants. Thus,
reaction can only proceed through a solid-state diffusion of
reactants through the oxide
[19]. Diffusion in solids takes
place because of the occurrence of imperfections or defects in
solids [16] .
Imperfections in solids can be broadly divided into point
or
lattice
defects.
Vacancies,
interstitial
misplaced atoms come under point defects.
atoms,
and
Line and surface
defects include dislocations, grain boundaries and inner and
outer surfaces. Point or lattice defects are responsible for
lattice diffusion, also called volume or bulk diffusion [16].
In polycrystalline material's the relative contributions
of
the
different
types
of
diffusion
are
a
function
of
temperature, partial pressure of the components, grain size,
porosity,
activation
etc.
Grain
energy
than
boundary
volume
diffusion
diffusion
has
and
as
a
smaller
a
result
becomes increasingly important at lower temperatures [20].
14
Lattice Defects
In a perfect crystal both the entropy and internal energy
of the system increases with the creation of point defects.
The equilibrium concentration of the defects will be reached
when the free energy of the system is at a minimum.
given
crystal
all
types
of defects,
in principle,
In any
will be
formed although the free energies of formation of different
types
and
systems
of
defects
will
have
widely
different
values. Correspondingly one type of defect structure commonly
predominates
in
a
particular
solid
at
equilibrium.
The
relative concentrations of the different types of defects is
a function of temperature and other variables which influence
the state and the composition of the compound. Thus,
defect
equilibria with a large positive enthalpy of formation which
are not favored at low temperatures, may become important at
high
temperatures.
Lattice
defects
may
be
neutral
or
Lattice diffusion takes place through the movement
of
charged.
Lattice Diffusion
point
defects.
The presence
of 7 different
types
of defects
gives rise to different mechanisms of diffusion. Pick's laws
of diffusion give the mathematical relationship between the
concentration gradient and the diffusion coefficient [18] . The
rate
of
diffusion
is
expressed
in
terms
of
the
diffusion
coefficient D, which is defined by Pick's first law,
15
where J is the instantaneous flow rate per unit area of the
diffusing species across a plane,
the
diffusing
concentration
diffusion
species
at
gradient
constant
D
the
normal
is
the
c is the concentration of
plane,
to
and
the
flow
dc/dx
plane.
rate/unit
is
the
Thus,
the
area
at
unit
concentration gradient normal to the plane.
Pick's second law relates the concentration gradient to
the change in concentration with time. For the case in which
the diffusion coefficient is independent of concentration:
dt
These
equations
can be
Qx 2
solved
under
certain boundary
conditions which may be closely approximated experimentally.
For homogenous diffusion in a solid the concentration of the
diffusing species normal to the plane is given by:
exp (-■
2\JitDt
ADt
where c represents the concentration at a distance x from the
surface,
cQ is the concentration of the species
present on the surface,
originally
and t the time of diffusion anneal.
The typical experiment used to establish the constants in this
relationship
involves
deposition
of
a
very
thin
film
of
radioactive isotopes on a plane surface of a sample and, after
subsequent
diffusion
anneal,
determination
of
the
&
16
concentration of diffusing species as a function of distance
from the plane surface
[16].
If diffusion is nonhomogenous,
different relationships between concentration and penetration
distance may result.
The temperature dependence of the diffusion coefficient
is given by:
D-Z^exp(--^L)
where
Q
is the
activation
energy,
and the pre-exponential
factor D0, the frequency factor [20].
Diffusion Mechanisms
Diffusion is said to take place by a vacancy mechanism if
an
atom
on
a
normal
lattice
site
jumps
into
an
adjacent
unoccupied lattice site. If an atom moves from an interstitial
site
to
one
of
diffusion
occurs
movement,
or
its
by
jump,
neighboring
an
of
interstitial
interstitial
the
atom
sites,
mechanism.
involves
a
Such
the
a
considerable
distortion of the lattice, and this mechanism is probable only
when the interstitial atom is smaller than the atoms on the
normal
lattice position
[20].
jump becomes
too
probable, an
interstitial
neighbors
If the distortion during the
large to make
on normal
the
interstitial
atom .pushes
lattice site
one
of
into another
mechanism
its
nearest
interstitial
position and itself occupies the lattice site of the displaced
atom.
17
A
variant
of
the
interstitialcy
mechanism
is
the
crowdion[20] . In this case an extra atom is crowded •into a
line of atoms and thereby displaces several atoms along the
line from their equilibrium positions.
High Temperature Oxidation of Metals
High-temperature
oxidation
usually
results
in
the
formation of an oxide film. The mechanism of oxidation depends
on the nature of the scale,
that is, whether the oxide
is
solid or liquid or if it also partially evaporates. If solid
scales
are
formed,
the oxidation behavior
also
depends
on
whether the scales are compact or porous.
Compact Oxide Scales
A compact scale acts as a barrier separating the metal
and the oxygen gas. The rate of oxidation at high temperatures
is limited by diffusion through the compact scale, provided
there is sufficient oxygen available at the oxide surface. The
oxide grows in thickness leading to an increase in diffusion
distance. This results in a decrease in the rate of reaction
with
time.
Compact
scales
offer
the
best
protective
properties. Improving the protective properties of the oxide
scale
leads
directly
to
an
improvement
in
the
oxidation
resistance of metals and alloys [16].
Wagner [2,3] has provided a fundamental understanding of
the essential features of high temperature oxidation of pure
metals in his theory of high temperature parabolic oxidation.
*
18
Figure I shows the Wagner schematics of concentration gradient
of oxygen and metal ion vacancies and the transport processes
occurring in an oxide scale containing such vacancies = The
legend for Figure I is as follows:
VHP"
Vgq+
e
e"
M
MqOp
P
represents metal ion vacancies
represents oxygen ion vacancies
represents electron holes
represents electrons
represents electrons
represents the hypothetical metal oxide
represents the functional dependence
of V mP- on P0(d)
g
represents the functional dependence
of V0q" on Pp(s)
pQ<d) represents dissociation pressure of oxide
P0^9) represents ambient oxygen partial pressure
Wagner's theory applies to compact
products.
Volume
corresponding
point
diffusion
of
defects)
or
the
a
scales of reaction
reacting
transport
ions
of
(or
electrons
across the growing scale is assumed to be the rate-determining
process
of
considered
the
to
total
migrate
reaction.
Electrons
independently
of
and
each
ions
other.
are
Since
diffusion through the scale is rate-determining, reactions at
phase boundaries are considered rapid, and it is assumed that
thermodynamic equilibrium is established between the oxide and
the oxygen gas at the M0/02 interface and between the metal
and the oxide at the M/MO phase boundary [16].
The
driving
force
of the
reaction
is the
free-energy
change associated with the formation of the oxide MO from the
metal M and the oxygen gas. The partial pressure of oxygen at
the
M/MO
interface is equal to the
equilibrium dissociation
19
M
0?
IyITl*•const fptojtf#* i»
IyITI “ «n$t (pgp
Metal Ion Vacancies
[Va11+I - const f p $ -SPTi
-[V0^+I= Censtfpa^ r
Oxygen Ion Vacancies
F i g u r e I,
W a g n e r S c h e m a t i c s for O x i d a t i o n of Alloys
20
pressure of the oxide in contact with its met a l , p0(d), while
at
the
oxide/oxygen
interface
it
is
equal
to
the
oxygen
pressure in the gas phase, P0^95• For an oxide scale with metal
ion vacancies, the metal ions diffuse outward from the M/MO to
the M/M02 interface.
direction,
The vacancies migrate
in the opposite
and their equilibrium concentrations
at the gas
phase interface are given by the defect equilibrium:
When p0_
<9) > p0^d), the metal ion vacancies are continuously
produced
at
interface.
the
MO/O2 interface
Oxygen
ion
and
vacancies
opposite to the metal
consumed
migrate
ion vacancies,
in
at
the
the
M/MO
direction
and their equilibrium
concentration at the M/MO and MO/O2 interfaces is given by:
+?e-+-^-Og
Oxygen
ion
vacancies
are
correspondingly
created
continuously at the M/MO interface and consumed at the M0/02
phase boundary.
The rate of growth is, in these terms, determined by the
gradients and the rate of diffusion of the components. These
mechanisms lead to parabolic oxidation behavior.
The
Wagner
theory
permits
an
evaluation
of
the
rate
constant of high-temperature parabolic oxidation provided the
diffusion
coefficient
is known. In evaluating
experimental
results the parabolic rate constant is expressed as
21
-AE=A
dt
x
Here
kp is
the
parabolic
rate
constant
and
x,
the
oxide
thickness. On rearranging and integrating,
JL
X=Cfc 2
Porous Oxide Scales
During compact oxide scale formation the rate of reaction
is governed by the
solid-state
diffusion
of the
reactants
through the scale, as long as sufficient oxygen is available
at the oxide surface.
Deviations
from ideal behavior,
i.e.
compact and pore-free oxides, occur when porous scales form.
If the transport occurs by an outward migration of metal
ions by the vacancy mechanism, vacancies may collect to form
cavities and pores at the metal/oxide interface and produce
appreciable
porosity
in
the
oxide
scale
after
extended
reaction. The cavities then act as barriers to the solid-state
diffusion process. These oxide scales can rupture and breakup
becoming
non-barriers
to
gaseous
oxygen.
Also
when
phase
boundary processes become rate-determining instead of solidstate diffusion through the scale, then porous scales may be
formed.
When porous scales are formed,
the kinetics of the
reaction is commonly found to be linear with time. The rate of
initial
oxidation
indicating the
is often observed to decrease with
formation
of a protective
phase
before
time
the
oxidation transforms to a linear rate. Phase boundary process
22
limitations
occur when the linear oxidation for porous scales
is faster than the protective oxidation rate. Cracks can form
which rupture leading to the formation of porous scales.
Thus
porous
scales
are
non—protective. These
scales
usually occur during later stages of the reaction because of
rupture and
fragmentation of the
oxide
scale
after
it has
reached a critical thickness [16].
High Temperature Oxidation of Metal Alloys
Oxidation mechanisms for alloys are more complex than for
pure metals. The alloyed metal atoms do not diffuse at the
same
rates
either
in
the
oxide
or
alloy
phase
and
the
components of alloys have different affinities for oxygen. As
a result,
relative
oxide
amounts
scales
of
on alloys
the
alloy
do not
contain
constituents
as
the
does
same
the
metallic phase. As the oxidation proceeds, the composition and
structure of oxide scales often change.
kinetics
Thus the oxidation
often markedly deviate from ideal and simple rate
equations. Also, if oxygen dissolves in the alloy phase, the
least noble alloy component can form oxide inside the alloy.
Since more than one oxide is formed, Wagner classified
alloy oxidation in terms of different types of behavior.
1. S e l e c t i v e o x i d a t i o n in w h i c h t h e least n o b l e c o n s t i t u e n t is
s e l e c t i v e l y or p r e f e r e n t i a l l y oxidized.
2. Formation of composite scales.
3. Formation of scales with complex oxides.
23
Under1 any set of conditions
selective oxidation takes
place only above a critical concentration of the active alloy
component. Wagner [2,3] analyzed the conditions necessary for
oxidation
in
expression
extended
to
binary
alloys
this
critical
for
the
case
of
and
derived
a
mathematical
concentration.
composite
scales
This
and
can
scales
be
with
complex oxides.
Wagner considered an alloy A-B in which B is the less
noble metal and A and B do not react to form a double oxide or
spinel. Making the assumption that compact oxides are formed,
three main cases for the oxidation of the alloy A-B may be
considered.
1. At low concentrations of B, only the A oxide will be formed
and
B
will
diffuse
into
the
alloy
from
the
alloy/oxide
interface.
2.
For
sufficiently high concentrations
of B
in the alloy
only the B oxide will be formed and A will diffuse into the
alloy from the alloy/oxide interface.
3. At B concentrations ranging from the low concentrations in
case I to the high concentration in case 2, A oxide and the B
oxide will simultaneously be formed (composite scale).
Figure 2 shows the schematics
for the three different
oxidation cases. Assuming the formation of a compact,
free
oxide
scale,
Wagner
showed
that
the
pore-
critical
concentration N b (above which only the B oxide is formed)
given by :
is
24
U k r.
V
i
T
D
whsire V is the molar- volume of the alloy t Zg is the valence
of the B atoms, M0 is the atomic weight of oxygen,
diffusion
coefficient
of
B
in
the
alloy,
and
D is the
kp is
the
parabolic rate constant for the exclusive formation of the B
oxide.
The same type of assumptions can be made for composite
scales that are practically insoluble in each other. For the
case of formation of scales with complex oxides,
diffusion
rates are often appreciably smaller than in single oxides„ The
protective
scales
on
high-temperature
oxidation-resistant
alloys often consist of complex oxides (double oxides, spinels
etc.)
[16].
25
A-B alloy
A-B alloy
A
(A)
DO
"E
B
A
A oxide
B oxide
0,*
(B)
Diffusion Processes During Oxidation of A-B Alloys
(A)
Exclusive Formation of A Oxide
(B)
Selective Oxidation of B Oxide
(C)
Simultaneous Formation of A-Oxide (AO) and B Oxide (BO)
B Oxide Grows According to the Displacement Reaction
AO + B2+ = A2 + BO
F i g u r e 2. W a g n e r S c h e m a t i c s f o r O x i d a t i o n o f M e t a l A l l o y s
26
CHAPTER 4
EXPERIMENTAL METHOD
Auger Electron Spectroscopy
Auger-electron spectroscopy(AES) is based on the Auger
radiationless
process,
is
[21] . The
process
shown
in
Figure
energetics'
3.
This
of
process
the
Auger
involves
an
electron excitation, which leaves a hole in a core level of an
atom. This preliminary excitation process can be stimulated by
absorption
of
energy
excitation
process,
from
the
recombination process.
the core hole
is
a
core
primary
hole
electron.
is
filled
by
After
an
the
Auger
In the Auger recombination processes
filled by an electron occupying a higher
energy level, that is, either a shallower core electron or a
valence
electron.
transferred
to
The
another
energy
lost
electron.
by
This
this
last
receive enough energy to leave the system,
electron
electron
is
may
thus becoming an
Auger electron.
The whole process, excitation plus Auger recombination,
is
usually
notations
labeled
for
the
using
initial
the
conventional
states
of
the
spectroscopic
three
electrons
involved [22]. If the excitation involves a hole in a Is state
and this hole is filled by an electron in the 2s state that
transfers its energy to another electron in the 2s state, then
the
entire
process
is
labeled
KL1L2. In this notation the
27
Ee = Auger energy
FL = Fermi Level
F i g u r e 3. E n e r g e t i c s o f t h e A u g e r P r o c e s s
28
L
valence states are denoted by the letter V. Thus the final
energy of the Auger electrons is linked to the energies of the
three
electron
states.
Therefore
the
emission
of
Auger
electrons at a given energy reveals the presence of certain
energy
levels
characteristic
in
of
the
a
system.
given
Since
element,
those
levels
are
also
reveals
the
it
presence of that element in the system. This provides for the
analytical
capability
of AES.
The
exact
link
between
the
kinetic energy of the emitted Auger electrons and the corelevel energies is complicated by factors such as electronic
relaxation
on
formation
of
a
core
hole
and
final-state
multiplet interactions. Nevertheless, the uncorrected binding
energies of the three electrons are the most important factor
in determining the Auger-electron kinetic energy.
Neglecting all
energies,
factors
except the uncorrected binding
and calling these energies E1, E2 and E*3, for the
first excited electron, the electron that fills the core hole,
and the Auger electron,
respectively,
the kinetic energy of
the latter can be written as:
Ek = E*3 + E2 - E 1
This equation is derived by estimating the energy of the
two-hole final states in two subsequent steps, one creating a
hole in state 2 and the other creating a hole in state 3. The
energies in this equation are measured from the zero of the
kinetic-energy
scale,
that
is,
from the vacuum
level.
The
asterisk in the binding energy term E*3 means that this is not
29
the unperturbed binding energy but rather the binding energy
in the presence of another core hole.
The Auger electrons
[7] thus have unique energies
for
each atom and if the energy spectrum from about 0 to 2 keV is
analyzed, the energies of the Auger electron peaks allow for
I
the
identification of the surface elements present,
hydrogen
and
helium.
The
reason
that
AES
is
a
except
surface-
sensitive technique lies in the intense inelastic scattering
that occurs
for electrons
in this energy range.
Only Auger
electrons from the top few atomic layers of a solid survive to
be ejected and measured.
An
Auger
electron
spectroscopy
ultra-high vacuum test system,
excitation
and
an
energy
system
consists
of
an
an electron gun for specimen
analyzer
for
detection
of
Auger
electron peaks in the total secondary electron energy spectrum
[2 1 ] .
Experimental System
Ultra-High Vacuum Systems
A UHV system is required for AES for two reasons. First
and
foremost,
meet as
the electrons emitted from a specimen should
few gas molecules on their way to the analyzer as
possible so that they are not scattered [23] . More importantly
the reduction of contamination on the surfaces under analysis
by the residual gases in the vacuum establishes a necessary
base
pressure
of
about
I
x
io"9 torr
(I
torr=
I mm.
Hg
30
pressure). Such a pressure falls into the regime of ultra-high
vacuum (UHV). This greatly restricts the use of materials in
AES.
For
the
system
used
in
the
present
AES
study
most
materials used in fabrication are stainless steel with copper
gaskets used as a sealing material between conflat flanges
[7] .
The
system
assembly
including
the
two
turbomolecular
pumps, the electron gun, the ion gun and the vacuum chamber is
shown
in Figure
sections,
4.
a main
The vacuum chamber is divided
section
and
a
differential
into two
section.
The
turbomolecular pumps differentially pump the ion gun as well
as differentially pump a vacuum seal on the sample manipulator
in the main chamber.
Pumping speed is increased by using
separate pumps on the two chambers. The two pumping systems
each include a roughing pump and a turbomolecular pump. They
reduce
the
main
chamber
and
differential
pressures of about 1.5 x IO'9 and I x
after bakeout.
The
bakeout
involves
io ‘9
chamber
to
base
torr, respectively,
increasing
the
system
temperature to about ISO0C for a few hours in order to remove
most of the volatile contaminants.
Ion Gun
Argon ion bombardment is a method of producing a clean
surface and can be used in conjunction with surface analysis
to measure compositional information as a function of depth
[24].
Depth
simultaneous
profiling
ion
is
a
bombardment and
technique
analysis
which
involves
(AES analysis in
31
1. Sample Probe
2. Sample Mount
3. Main Chamber
4. Ion Gun
5. CMA & Electron Gun
6. CMA Control Hookup
7. Turbo Pump (Main Chamber)
8. Roughing line (Main Chamber)
9. Turbo Pump (Differential Chamber)
10. Roughing line (Differential Chamber)
11. Argon Supply line
12. Leak Valve
13. Butterfly Valve
14. Differential Pumping line
15. Vacuum Seal Pumping line
F i g u r e 4. A E S A n a l y t i c a l E q u i p m e n t
32
this case) to develop a continuous compositional profile. This
technique involves the use of an argon ion gun. The ion gun is
a
V.G.
Scientific
electron
devices
impact
source.
dual
Ion
excitation with
which
positive
are
emitted
ion
guns
in which the inert gas
collisional
energy
FAB61
atom
are
ions
gun
simple
which
has
an
electrostatic
(Ar+) are generated by
electrons
of
typically
100
eV
from a hot tungsten filament. The
ions are accelerated,
focused and rastered on the
sample, creating a sputtered area [7].
The primary Ar+ ions are produced from 99.9995% minimum
purity argon gas introduced in the differential pumping line.
These
ions
electronic
negotiate
fields
energy beam.
an
array
of
static
and leave the gun. in a
The primary beam
and
oscillating
columated,
is rastored to
high
scan an area
larger than the cross sectional area of the primary ion beam.
The FAB61 ion gun
is controlled by a V.G.
Scientific
400A
power supply. A Leybold-Heraeus 867 916 raster power supply
controls restoring of the primary ion beam. The energy of the
primary ion beam used is generally 3.0 keV with a ion gun
target current of about 950 nanoamps. The target current is
controlled by manually adjusting the ion gun emission current.
Sample Mount System
Samples, both thin foil metals and thick intermetallics,
are
mounted
on
a
sample
holder,
mounted
on
the
end
of
a
stainless steel rod in the vacuum chamber. A diagram of the
sample
holder
is
shown in
Figure 5.
It
has provision for
1. SCREWS TO HOLD SAMPLES
2. SCREWS TO HOLD MOUNT ON CLAMPS
3. CLEARENCE FOR SCREWS
4. TOP VIEW
5. SIDE VIEW
F i g u r e 5, S a m p l e H o l d e r f o r A E S A n a l y s i s
34
holding five samples of about I cm x I cm dimensions
each
including a LiF sample which is used for system alignment. Two
micrometers
connected
to
the
support
rod
establish
its
forward-backward and side-to-side position. Provision is also
made
for
rotational
and
z-directional
motion.
holder is grounded through a current meter.
The
sample
This is done to
prevent sample charging during the argon ion etching process
the
Auger
analysis
and also to
monitor the
target
current.
The
relative
spatial
orientation
of
the
samples
and
analytical equipment is critical. Figure 6 shows the relative
orientation of the sample mount, electron gun and ion gun. The
plane of the test sample surfaces is generally at an angle of
about 45° to longitudinal axis of both the Ar+ ion beam and the
electron gun, which have a 9O0 angle relative to one another.
During analysis the sample surface is about 10 mm from the
front end of the ion gun and about 8 mm from the end of the
electron gun.
A lithium fluoride foil is mounted on the sample holder
to facilitate alignment of the samples in the primary Auger
electron beam and the ion beam.
This foil emits a brilliant
red glow when sputtered. The lithium fluoride target foil is
aligned
relative
to
the
ion
gun
by
adjusting
the
probe
position while observing the brilliant red spot at the point
of impact of the ion beam. For this step the primary ion beam
is static rather than rastored.
35
8mm
10 mm
1. Ion Gun
2. CMA
3. Rotatable Sample Probe
4. Sample Mount
F i g u r e 6. R e l a t i v e O r i e n t a t i o n o f S a m p l e M o u n t
w i t h Ion Gun an d E l e c t r o n Gun
36
The spot size of the primary ion beam, and the dimensions
of the area sputtered are adjusted with the controls on the
gun power supply and raster power supply while observing the
sputter induced phosphorescence on the lithium fluoride coated
target foil. The sputter area is adjusted to about 4mm x 4mm.
The primary ion beam spot size is between 0.5 and Imm in dia.
Energy Analyzer
The energy analyzer used for the present study is the PHI
Model
10-155
designed
to
excited by
Cylindrical-Mirror
detect
an
Auger
Analyzer
electrons
electron beam.
(CMA)
ejected
The primary
which
from
is
specimens
electron beam
is
generated by an electrostatic electron gun mounted coaxially
inside the analyzer. The analyzer also contains an electron
multiplier which
amplifies
the detected
secondary
electron
signals.
The electron gun is controlled by the PHI Model 11-010
Electron
Gun
calibrated
Control.
steps
of
500
Beam
V
voltages
from
0 to
are
available
5 keV.
The
in
filament
current and emission voltage of the electron gun and the focus
of the electron gun are adjustable.
The electron gun is operated with a primary beam voltage
of 3 keV for most of the tests,
with
a
filament
current
of
and 2 keV for other tests,
about
I milliamp.
The
target
current on the sample is normally about 4000 nanoamps.
The electron gun and CMA arrangement are shown in Figure
7.
The
electron gun has one electrostatic lens. Beam current
--- OUTER
\CYLINDER
ELECTRON MULTIPLIER
0-4 kV
!— MAGNETIC
s h ie l d
MODULATION
CONTROL
Figure 7
The E l e c t r o n Gun and CMA A r r a n g e m e n t
38
is
varied
by
adjusting
extraction voltage
the
filament
temperature
(V1 - beam voltage)„ A
focused
and
beam
the
of
primary electrons is obtained from the gun when the ratio of
focus
voltage
(V2 - beam voltage)
to
the
beam
voltage
is
nominally 0.25.
Secondary
electrons
electron
beam
are
strikes
generated
the
at
specimen.
^-nc-*-u<^es an inner cylinder that is grounded,
the
point
The
the
analyzer
plus an outer
cylinder that is biased through terminal Vm. Negative voltages
applied to
the
outer
cylinder repel the secondary electrons
approaching the cylinder through openings in the side of the
inner cylinder,
forcing these electrons back through
other
openings in the inner cylinder so that they are collected by
the electron- multiplier. The number of resulting electrons is
multiplied, prior to striking the collector plate, by applying
a voltage bias of a few kiloelectron volts to the multiplier.
The
energy
analysis
of
the
secondary
electrdns
is
controlled digitally through a PHI 32-150 Digital AES Control
and a 96 V/f Preamplifier. The signal from the analyzer passes
through the preamplifier and the Digital AES Control which is
connected to an IBM PC AT.
Sample Preparation
The intermetallie test samples used in this study were
prepared by the argon arc melting of the appropriate amounts
of the pure metal powders at the Idaho National
Laboratory
39
Facility at EG&G,Inc.
The samples are cut and polished to a
final finish of 0.05 microns.
The pure metal samples are cut from research grade foils.
The pure zirconium metal samples are cut from 99.99% pure, 25
mm thick foil purchased from Johnson Mathey Inc. The pure Ni
samples
are cut
from high purity nickel
obtained
from the
Physics Department, Montana State University.
Surface Preparation
Prior to oxidation the intermetal^ic samples are polished
to a high lustre. This is accomplished using a motor driven
polishing wheel with the abrasive being a 0.05 micron gamma
alumina number 3 suspended in distilled water. The samples are
mounted on small aluminum blocks with a hot melt adhesive for
convenience
following
during
each
polishing.
oxidation
Polishing
and analysis
to
a
cycle
high
lustre
is assumed
to
remove the prior oxide scale to expose fresh bulk material.
Before oxidation the intermetallics are rinsed in bulk grade
acetone to dissolve the hot melt adhesive.
The samples are
then cleaned with bulk grade methanol.
The
pure
zirconium
and
pure
nickel
samples
are
acid
etched prior to oxidation with a 5% hydrofluoric acid solution
to produce a clean and reproducible surfaces. The reason for
this is their thickness and lack of flatness makes polishing
impractical.
The
acid
solution
is prepared
fresh
for each
etching. The samples are immersed in the acid solution for 15
40
seconds and are then rinsed in distilled water. This produces
a high metallic lustre on the samples.
Each sample is oxidized in air at SOO0C for I hour. The
oxidation is accomplished in a small metallurgical oven. The
temperature control and display of the oven were calibrated
prior to its use. For each run the oven is brought up to the
required
temperature
before
introducing
the
samples.
The
temperature drop caused by opening the oven door to introduce
the samples was observed to be negligible.
Sample Alignment
A sample
is loaded into the
system and the system
is
pumped down to a pressure of at least 2 x lb"9 torr prior to
aligning the samples. The current meter is activated, before
the electron or ion gun are operated,
to ensure the sample
mount is grounded. The electron beam target current is set as
desired.
The
LiF
sample
is
used
to
adjust
the
sample
position. The electron beam and the ion beam impact location
are
brought
close
to
each
other
by
observing
their
characteristic spots, while adjusting the probe position. The
elastic peak procedure is run on the LiF sample to ensure that
the elastic peak is at 2000 eV. The other samples are aligned
by adjusting the probe position to ensure the elastic peak is
at 2000 eV.
41
Elastic Peak Procedure
The electron gun control is activated and the voltage
adjusted
to
2
k e V . The
electron
beam
current
is
set
as
desired. The multiplier supply is adjusted to obtain an energy
trace of the secondary electrons on a CRT display,
case
a IBM PC AT.
manipulator
electron
until
gun
The x,
y and z-axis
the peak
focus
is
is
aligned
adjusted
to
in this
is adjusted on the
at
2 keV.
maximize
The n ■ the
the
signal
intensity at 2000 volts.
Acquiring AES Data
The AES
data
is acquired by setting the
energy
sweep
range, time per sweep and number of sweeps. The counts and the
multiplier voltage for each survey as well
as time between
surveys are recorded. The data is then acquired and stored in
a
.DAT
file
on
the
IBM
PC
AT
for
further
mathematical
manipulation such as smoothing, differentiation, multiplexing
at a later time.
Shut Down Procedure
Once the desired number of surveys have been taken, the
ion
gun
power
supply
is
shut
do w n . This
is
followed
by
switching off the Electron Gun Control, Digital AES Control,
and the Electron Multiplier Supply.
42
CHAPTER 5
RESULTS
The
Auger
zirconium
spectrum
foil
which
from
had
50
been
to
550
exposed
eV
of
to
unsputtered
air
at
room
temperature is shown in Figure 8. The ordinate in the spectrum
is the count rate, a measure of the intensity of the secondary
electrons,
in k counts/sec. The count rate at each energy is
the sum of the offset value shown plus the multiple of the
ordinate value times the scale factor specified on the figure.
Thus
an
absolute
value
of
2
on
the
ordinate
in
Figure
8
represents 19.860 kcts/sec (2 x 2.800 + 14.260), for example.
The survey shows oxygen and carbon auger peaks at about 505 eV
and 270 eV, respectively. Small zirconium peaks are observed
in the energy range from 80 to 180 eV. Figure 9 shows an Auger
survey of the same zirconium foil after 40 minutes of argon
ion sputtering. This survey shows strong zirconium peaks and
significantly diminished carbon and oxygen peaks.
The energy range between 80 and 180 eV from the survey
shown in Figure 9 is expanded and displayed in Figure 10. The
most important peaks for elemental Zr are the MNN peaks at 90
eV and 115 eV, MNV peaks at 123 eV and 145 eV and an M W
at
172
eV.
The
Auger
peaks
used
to
establish
Zr
peak
surface
abundance and oxidation state in the present study are those
at
90 eV
and
145 eV which are designated as the Zrl and Zr2
2.800 K cts/sec
Offset= 14.260 K cts/sec
Count Rate, arbitrary units
Scale factor=
50.0
100.0
150.0
Figure
200.0
250.0
300.0
Kinetic Energy, eV
8.
350.0
AES Sur v ey o f Z i r c o n i u m F o i l
400.0
450.0
500.0
550.0
Count Rate, arbitrary units
250
F i g u r e 9.
AES Sur v e y o f S p u t t e r
HOO .0
Cl eaned Z i r c o n i u m
450.0
500.0
550 .
Count Rate, arbitrary units
80 .0
90.0
100.0
Figure
130.0
Kinetic Energy, eV
110.0
10.
120.0
140.0
iso.o
iso .0
170.0
Expanded Ener gy Regi on o f S p u t t e r Cl eaned Z i r c o n i u m
180.0
46
peaks, respectively. The Auger surveys shown in Figures 8 and
9
have
too
broad
an
energy
range
of
display
to
be
used
accurately in the determination of Zr Auger peak intensities.
The Auger spectra are therefore expanded to smaller energy
regions to provide a better resolution of the elemental
Zr
surface abundance from the peak intensities measured by the
procedure described below.
Figures 11 and 12 show the multiplexed surveys of the Zr2
peak
on
oxidized
respectively.
and
sputter
cleaned
zirconium
foil,
On sputtering away the oxide layer a shift in
the Zr2 peak position from 140 eV to 145 eV is noticed. This
energy shift of the peak results from a change in the chemical
state of the metal. The Zr2 peak is useful in indicating the
oxidation state of zirconium and cannot be conveniently used
in quantitative Auger analysis of Zr surface abundance.
The
MNN Zrl peak does not shift significantly when the zirconium
is oxidized and therefore it can be reliably used for peak
intensity measurements, which are proportional to Zr surface
abundance.
A display of the multiplexed Zrl peak from 70 to H O
eV
on sputter cleaned Zr foil (from Figure 9) is shown in Figure
13. The Zrl peak for this survey is at 90 eV. The count rate
measured at the peak energy is 1901.200/sec and the base count
rate, which is the count at the high energy base of the peak
at 99 eV is 1693.629/sec. The peak height, the difference of
the
two, is
207.571/sec. This difference is divided into the
0.202 K cts/sec
Count Rate, arbitrary units
Scale factor =
120.0
128.0
Figure
132.0
136.0
ItO .0
Kinetic Energy, eV
11.
Multiplexed
148.0
152.0
Zr2 Peak on Z i r c o n i u m F o i l
156.0
160 .
actor =
Count Rate, arbitrary units
Scale
120.0
124.0
128.0
Figure
12.
132.0
140.0
Kinetic Energy, eV
M ultiplexed
144.0
152.0
Zr2 Peak on S p u t t e r Cl eaned Z i r c o n i u m
160 .
Count Rate, arbitrary units
70.0
78.0
Figure
13.
82.0
86.0
Kinetic Energy
Multiplexed
Zrl
90.0
98.0
102.0
106 .0
Peak on S p u t t e r Cl eaned Z i r c o n i u m
110.0
50
base count of 1693.629 to give a relative intensity of 0.1226
foT
the Zrl peak in this spectrum. Using the same procedure
the relative intensity for the Zrl peak for zirconium foil
exposed to air from Figure 8 is 0.0311.
A display of the oxygen Auger peak at 506 eV on oxidized
zirconium foil is shown in Figure 14. The relative intensity
for the oxygen peak at 506 eV is 0.1456.
The relative intensities of the oxygen peak and Zrl peak
at different sputter times for oxidized Zr foil are shown in
Figure 15 as a function of sputter time. This constitutes the
sputter
depth
temperature.
profile
The
for
zirconium
oxidized
at
Zrl peak intensity increases with
time while that of oxygen decreases.
room
sputter
This indicates a slow
sputtering of the oxide layer.
The sputter depth profile of zirconium foil oxidized at
3OO0C for I hour is shown in Figure 16. The oxide layer in
this case takes longer to sputter in comparison to Zr foil
exposed
to
air.
similar
trends
The
to
oxygen
those
of
and
Zrl
peak
the
room
intensities
temperature
show
oxidized
zirconium foil in Figure 15.
The Auger spectrum from 3 0 to 930 eV for nickel
foil
which had been exposed to air at room temperature is shown in
Figure 17. The survey shows carbon and oxygen peaks at about
270
eV
and
5 05
eV,
respectively.
Weak
nickel
peaks
are
displayed at about 55 eV, 708 ev,. 775 eV and 840 eV. The 55 eV
and 840 eV peaks are designated in this report as Nil and N i 2 ,
Scale factor
Offset
4*6.0
510.0
Kinetic Energy
Figure
14.
516.0
522.0
M u l t i p l e x e d Oxygen Peak on Z i r c o n i u m
534.0
Zr 1 int.
Relative Intensity
Ox int.
0.00
5.00
10.00
Figure
15.
15.00 20.00 25.00 30.00 35.00
Sputter Time (minutes)
Sputter
Depth P r o f i l e
for Zr
40.00
Zr 1 int.
Relative Intensity
Ox int.
20.00
40.00
60.00
80.00
100.00
Sputter Time (minutes)
Figure
16.
S p u t t e r Depth P r o f i l e
f o r Zr(ox)
120.00
Count Rate, arbitrary units
30 .O
120 .0
300.0
390.0
Kinetic Energy
210.0
Figure
17.
480.0
AES Sur v ey o f N i c k e l
570.0
Foil
750.0
840.0
930.0
55
respectively.
The Auger spectrum of the
same nickel
foil
after 10 minutes of argon ion sputtering is shown in Figure
18.
The
carbon
peak
is
significantly
diminished
with
an
increase in the intensity of the nickel peaks.
The Nil peak from Figure 18 is expanded between 45 and 75
eV and displayed in Figure 19. The relative intensity for this
peak
is
0.3770
from
the peak
intensity
at
56
eV
and
the
background intensity at 74 eV. The Ni2 peak on sputter cleaned
Ni foil is shown in Figure 20 in the energy range between 825
and 870 e V . The relative intensity for this peak is 0.1099
based
on
the
peak
intensity
at
840
eV and
the
background
intensity at 867 eV.
The relative peak intensities for the Nil, Ni2 and oxygen
peaks
for
the
room
temperature
exposed
nickel
foil
at
different sputter times are shown in Figure 21. This sputter
depth profile shows an increase in nickel peak intensities and
a decrease
in the
oxygen
peak
intensity
with
increase
in
sputter time. The sputter depth profile for the nickel foil
oxidized
at
300°C
for
I hour
is
shown
in
Figure
22.
The
profile shows a similar trend of peak intensities with sputter
time. The decrease in the oxygen peak intensity is not clearly
shown on the scale of these figures.
The sputter depth profile for the oxygen,
Zrl, Nil and
Ni2 peaks for the intermetallic ZrNi5 after exposure to air is
shown in Figure 23. With increase in sputter time,
in
the
Nil, Ni2 and
Zrl
increases
peak intensities are observed. The
Count Rate, arbitrary units
Scale factor=
30.0
120.0
210.0
300.0
480.0
570.0
Kinetic Energy, eV
Figure
13.
AES Sur v ey o f S p u t t e r
Cl eaned N i c k e l
750.0
9 3 0 .0
Offset = 30.373 K cts/sec
Count Rate, arbitrary units
Scale factor =
HS.O
H S .0
S I .0
S H .0
57.0
60.0
63.0
66.0
63.0
Kinetic Energy, eV
Figure
19.
M u l t i p l e x e d Ni I Peak on S p u t t e r Cl eaned N i c k e l
72.0
75
Count Rate, arbitrary units
825.0
843.0
847.5
Kinetic Energy, eV
F i g u r e 20.
M u l t i p l e x e d N i 2 Peak on S p u t t e r Cl eaned N i c k e l
870.0
Ox int.
Relative Iiitensity
Nil int.
Ni2 int.
0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 10.00
Sputter Time (minutes)
F i g u r e 21.
Sputter
Depth P r o f i l e
f o r Ni
Ox int.
Relative Intensity
Nil int.
Ni2 int.
0.00
2.00
4.00
6.00
8.00 10.00 12.00
Sputter Time (minutes)
F i g u r e 22.
Sputter
Depth P r o f i l e
14.00
f o r N i (ox)
16.00
Zr 1 int.
Relative Intensity
Ox int.
Nil int.
Ni2 int.
10.00
15.00
20.00
25.00
Sputter Time (minutes)
F i g u r e 23.
Sputter
Depth P r o f i l e
for
Zr Ni S
30.00
62
sputter depth profiles for air exposed Zr2Ni7, ZrNi3 and ZrNi
are
shown
in
Figures
24,
25
and
26,
respectively.
These
exhibit a similar trend to ZrNi5 in Figure 23.
The intermetallies ZrNi5 and ZrNi3 were oxidized at 3OO0C
for
I hour.
oxidized
The
sputter
intermetallics
depth
is
profiles
shown
in
for
these
Figures
27
surface
and
28,
respectively. The peak intensities after removal of the oxide
layers
are
different
than
those
of the
respective
sputter
cleaned room temperature oxidized intermetallics. The steady
state Zrl peak intensity is almost halved,
while an almost
order of magnitude difference is observed in the case of Nil
and Ni2 peak intensities.
used
for Auger
spectrum
The primary electron beam energy
of the
intermetallics
oxidized
at
300°C is 2 keV while that used for the other samples analyzed
is 3 keV.
Zr 1 int.
Relative Intensity
Ox int.
Nil int.
Ni2 int.
0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 10.00
Sputter Time (minutes)
F i g u r e 24.
Sputter
Depth P r o f i l e
for
ZrZNi7
Zrl int.
Relative Intensity
Ox ini.
Nil int.
Ni2 int.
0.00
5.00
10.00
15.00
20.00
25.00
30.00
Sputter Time (minutes)
F i g u r e 25.
Sputter
Depth P r o f i l e
f o r Zr Ni S
35.00
Zr 1 int.
Relative Intensity
Ox inf.
Nil int.
Ni2 int.
0.00
5.00
10.00
15.00
20.00
25.00
30.00
Sputter Time (minutes)
F i g u r e 26.
Sputter
Depth P r o f i l e
for
Zr Ni
35.00
LLUib
Zr 1 ini.
Relative Intensity
Ox ini.
0.025
Nil ini.
Ni2 ini.
0.015
0.005
10.00
F i g u r e 27.
20.00
30.00
40.00
Sputter Time (minutes)
Sputter
Depth P r o f i l e
50.00
f o r Zr Ni 5 ( o x )
60.00
0.045
Zr 1 inf.
Relative Intensity
0.035
Ox inf.
Nil inf.
0.025
Ni2 inf.
0.015
0.005
0.00
10.00
20.00
30.00
40.00
50.00
60.00
Sputter Time (minutes)
F i g u r e 28.
Sputter
Depth P r o f i l e
for
Z r N i 3( o x )
70.00
68
CHAPTER 6
INTERPRETATION
The sputter depth profiles shown in chapter 5 indicate
the time required to sputter the surface oxide layers. The
oxide layers are formed at two different temperatures a) room
temperature and b) heating at SOO0C for I hour.
The sputter
depth profile for room, temperature oxidized zirconium is shown
in Figure 15. The total sputter time to remove the oxide layer
is between thirty and forty minutes. The first three minutes
of sputtering is characterized by a steady removal of surface
carbon. The carbon is either deposited on the surface due to
handling prior to insertion in the analysis system or as a
result of outgassing from the chamber walls during the bakeout
of the vacuum system. The Zrl peak intensity increases rapidly
at the start of sputtering attaining a steady state value of
about 0.12 after 3 minutes. The oxygen peak intensity starts
decreasing steadily after three minutes of sputtering reaching
a minimum of about 0.04 after 30 to 40 minutes of sputtering.
This
indicates the presence of some residual
oxygen on the
surface even after long sputter times.
The sputter depth profile for zirconium foil oxidized at
300°C
is
shown
consistent
Zrl
in
Figure
relative
16.
The
intensity
profile
of
shows
about
0.06
an
almost
after
85
minutes of sputtering. The oxygen intensity up to this point
69
steadily
decreases.
There
is
a
increase ' in
the
Zrl
peak
intensity between 85 and 95 minutes of sputtering, indicating
the change from an oxidized zirconium surface to a sputter
cleaned surface. The longer sputter times observed in Figure
16 indicate a thicker surface oxide layer in comparison to
that
in
Figure
temperatures,
in
15.
This
this
case
shows . that
3 OO0C,
higher
induce
the
oxidation
formation
of
thicker oxide layers.
In previous
studies
[11]
it has been shown that lower
temperatures of oxidation leads to dissociative chemisorption
of oxygen on zirconium and formation of various suboxides. The
suboxides formed are converted to ZrO2 at higher temperatures
of oxygen exposure.
oxidized
at
The longer sputter times
3OO0C can be attributed to the
for zirconium
formation
of
a
thick ZrO2 layer on the zirconium surface.
The sputter depth profile for room temperature oxidized
nickel
is
shown
in
intensity profiles
Figure
21.
The
for the Nil,
figure
shows
relative
Ni2. and oxygen peaks.
The
sputter time required for the three profiles to attain steady
state intensities is about 3 minutes. The Nil and Ni2 peaks
increase with sputter time with a decrease in the oxygen peak
intensity.
The sputter depth profile for nickel oxidized at
3OO0C is shown in Figure 22. The sputter time for the three
profiles to attain steady state intensity values is about 10
to
14
minutes
indicating
temperature oxidation.
thicker
oxide
layer
due
to
high
70
Earlier studies
the
Nil
Auger
[15] on oxidized nickel
signal
attenuates
more
during the depth profiling of the
indicated that
than
the
surfaces
Ni2
signal
of both
single
metal foils and intermetallics. The sampling depth of the Ni2
Auger signal is about three times that of the Nil signal [25],
rendering the latter much more sensitive to changes in surface
Ni
atomic
Figures
abundance.
21
and
22
The
sputter
indicate
depth
that
the
profiles
Nil
and
shown
Ni2
in
signal
intensities increase with increase in sputter depths, but the
Nil peak intensity increases by a greater factor.
The
sputter
oxidized Zr-Ni
depth
profiles
intermetallics,
for
the
room
temperature
ZrNi5, Zr2Ni7, ZrNi3 and ZrNi
are shown in Figures 23, 24, 25 and 26, respectively. The Nil,
Ni2 and Zrl peak intensities increase with sputter time in all
four cases. There is a transition from an oxide layer on the
intermetallics to a sputter cleaned surface for sputter times
of between 7 and 20 minutes for ZrNi3, 8 and 15 minutes for
ZrNi5, 12 and 25 minutes
Zr2Ni7.
Prior
to
this
for ZrNi
the
and 3 and 6 minutes
surface
of
the
for
intermetallics
contains oxides of nickel and zirconium. Argon ion sputtering
leads to the removal of the oxide layers and a transition to
higher relative intensities for the Nil, Ni2 and Zrl peaks and
a diminishing of the oxygen peak intensity.
The
different
intensity
room
profile
data
temperature
of
the
exposed
Zrl
metal
peak
for
foils
the
and
intermetallics after sputtering to steady state intensities
71
are as shown in Table I. The Zrl peak intensity increases with
increase
in atomic
fraction
of Zr.
The
Zrl peak
intensity
decreases from 0.1226 for pure Zr to 0.0549 for ZrNi with an
atomic
fraction
of
0.5 Zr and to
0.0259
for ZrNi5 with
an
atomic fraction of 0.167 Zr. Thus a reduction in the atomic
fraction of Zr in the sample reduces the Zrl peak intensity
proportionally.
Though this
correlation
is not
followed as
closely in the case of ZrNi3 and Zr2Ni7, within experimental
tolerance
limits
this
provides
a
rough
indication
of
the
change in the relative Zrl peak intensity with atomic fraction
of Zr. This intensity profile for the Zrl peak intensity is
shown in Figure 29.
Table I. Steady State Intensity Profile Data of Zrl peak for
the metal and intermetallic samples analyzed.
NAME OF SAMPLE
ATOMIC FRACTION
OF ZIRCONIUM
STEADY STATE INTENSITY
OF ZRl PEAK
Ni
0
0
ZrNis
0.167
0.0259
Zr2Ni7
0.222
0.0389
ZrNi7
0.25
0.0403
ZrNi
0.5
0.0549
Zr
1.0
0.1226
The intensity profile data of Nil and Ni2 peaks for the room
temperature exposed metal foil and intermetallic samples after
sputtering to steady state intensities are shown in Table 2.
A correlation of the type discussed for the Zrl peak intensity
is
not
as
clear
for
the
Nil
and
Ni2
peak intensities.
A t o m i c fraction of Zirconium.
Zr 1 ini.
Zr2Ni7
ZrNiS
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
Relative Intensity
F i g u r e 29.
St e a d y S t a t e
Intensity
P rofile
f o r Zr
73
although the Nil and Ni2 peak intensities increase with atomic
fraction
of
Figure
30.
Figures 27 and 28 show the sputter depth profiles
for
ZrNi5 and
Ni.
These
ZrNi3 oxidized
results
at
are
plotted
300°C,
in
respectively.
The
total
sputter time for the Zrl, Nil and Ni2 peaks to attain steady
state intensities is about 45 minutes in both cases.
Table 2. Steady State Intensity Profile of Nil and Ni2 peaks
_________ for the metal and intermetallic samples analyzed.
NAME OF SAMPLE
ATOMIC FRACTION
OF NICKEL
Nil PEAK
INTENSITY
Ni2 PEAK
INTENSITY
Zr
0
0
0
ZrNi
0.5
0.0769
0.0485
ZrNi7
0.75
0.2988
0.0929
Zr7Ni7
0.777
0.3237
0.0989
ZrNi7
0.833
0.3521
0.0998
Ni
1.0
0.385
0.1136
On
comparing
temperature
the
oxidized
sputter
ZrNi3 to
depth
that
profiles
for
high
for
room
temperature
oxidized ZrNi3, the final steady state Zrl, Nil and Ni2 peak
intensities are
reduced from about 0.046,
0.3 and 0.09 to
about 0.024, 0.035 and 0.0195, respectively. The primary beam
energy used for the metal foils and intermetallics oxidized at
room temperature and the pure metal
foils oxidized at 300°C
was 3 keV. Excessive background noise in the Auger spectrums
while operating the electron gun at this energy prompted a
change in beam energy to 2 keV for the intermetallics oxidized
at
3 00 °C . This
change in operating conditions resulted
in
A t o m i c Fraction of Ni
Ni I int.
Ni2 int.
0.833
Zr2Ni7
0.777
0.05
0.1
0.15
0.2
0.25
0.3
0.35
Relative Intensity
Figure
30.
S t ea dy S t a t e
Intensity
P rofile
f o r Ni
0.4
75
a significant reduction in the background noise and background
intensities. The
software adopted
for determining the peak
intensities was not normalized for changes in the operating
conditions
which
resulted
in
a
significant
drop
in
the
observed peak intensities and hence the reduction in the Z r l ,
Nil and Ni2 peak intensitiy values, as noted above.
This
reduction
in
the
peak
intensities
may
also
be
partially attributed to cracks developed on the intermetallic
surfaces which are laterally oxidized during high temperature
oxygen exposure. An optical microscope photograph of the ZrNi3
surface after oxidation at 3OO0C and polishing
Figure
31.
The photograph
shows
surface. The darker portions
a lateral
removed
oxide
of
layer extending
is shown
in
a crack running along the
the
photograph
indicates
from a crack which
is not
after long polishing or sputtering times.
The steady state relative peak intensities for oxygen for
ZrNi3 exposed
0.0101,
to
air and
respectively.
oxidized
at
3OO0C are
0.0031
and
The relative degree of oxidation of a
metal alloy can be estimated from the ratio of the oxygen peak
intensity to the intensity of the major metal peaks. Since Ni
is
the
major
metallic
component
in
the
high
temperature
oxidized intermetallics, the Nil peak intensity is used here
for the oxygen to metal peak intensity ratio analysis.
This
ratio of the oxygen and Nil steady state peak intensities, for
ZrNi3 at the oxidizing conditions listed above are 0.0104 and
0.2837, respectively. There is thus an
increase in the degree
F i g u r e 31. A n O p t i c a l M i c r o s c o p e P h o t o g r a p h o f P o l i s h e d
Z r N i 3 a f t e r O x i d a t i o n at S O O 0 C
77
of oxidation after high temperature oxidation and extensive
sP^ttering,
probably
due
to
oxidized
surface
cracks.
Regardless of the length of sputtering the presence of oxygen
in the oxidized crack surface diminishes the Z rl, Nil and Ni2
peak intensities and increases the oxygen peak intensity.
Under high temperature conditions, it is known that bulk
Zr
can
dissolve
interstitial
Zr/Ni
up
to
28
atomic
percent
solute in its hep lattice
alloys
also,
therefore,
oxygen
[14].
as
an
The untreated
probably have
absorbed
some
oxygen in the process of their being prepared, which can lead
to an attenuation of the Zrl, Nil and Ni2 peak intensities.
Surface roughness before oxidation may have varied because of
slight
differences
in
the
way
the
samples
were
polished.
Contaminants may have found their way to the surface before
oxidation and affected the development of the surface oxide
layer. The surface contaminants could also cause a diminished
peak intensity after high temperature oxidation.
The sputter time required to remove the oxide layers,
formed
at both room temperature
and at 3OO0C,
for all
the
samples analyzed is shown in Table 3. Except in the case of
the Zr foil, this value of sputter time was evaluated as the
time
taken
to
attain
8 0%
of
the
steady
state
intensity.
The Nil profiles are used because
cases they
show a sharp transition
Nil
peak
in almost all
from a surface with an
absorbed oxide layer to a sputter cleaned surface. For the
room
temperature
oxidized
ZrNi3 the
Nil
steady
state
peak
78
intensity
is
0.2988,
80%
of
which
is
about
0.24.
This
indicates a sputter time of about 7 minutes from the sputter
depth profile in Figure 25. Since pure Zr does not contain any
Ni,
the
time
taken
to
attain
80%
of
the
final
Zrl
peak
intensity is chosen as the sputter time. In the case of room
temperature oxidized zirconium foil the sputter time is about
3 minutes.
Table 3. The sputter time required to remove the oxide layer
__________for the metal and intermetallic samples.
SAMPLE NAME
SAMPLE TYPE/TYPE OF
OXIDATION1"
TIME FOR SPUTTERING
THE OXIDE LAYER
Zr
METAL FOIL/R.T.
3
Zr,0
METAL FOIL/H.T.
90 MINUTES
Ni
METAL FOIL/R.T.
2
MINUTES
NiO
METAL FOIL/H.T.
6
MINUTES
ZrNis
INTERMETALLIC/R.T .
6
MINUTES
ZrNis(ox) *
INTERMETALLIC/H.T .
39 MINUTES
ZrpNi7
INTERMETALLIC/R.T .
2
MINUTES
ZrNix
INTERMETALLIC/R.T .
7
MINUTES
ZrNix (ox) *
INTERMETALLIC/H.T .
25 MINUTES
MINUTES
ZrNi
INTERMETALLIC/R.T .
9 MINUTES
" R.T. = ROOM TEMPERATURE ;H .T . = HIGH TEMPERATURE (300°C/lhr.)
t BEAM ENERGY = 2 keV
The sputter depth profiles for the pure metal Zr and Ni
foils shown in Figures 15 and 22, respectively indicate the
relative depth of the high temperature oxide layers formed.
Since
the
sputtering
conditions
were
consistent
for these
surfaces, and since the oxides should sputter at approximately
the same rate
[27],
this indicates a thicker ZrO2 layer
(90
79
Iflirvute sputter time)
nickel
relative to the oxide layer formed on
(4 minute sputter time)„
80
CHAPTER 7
SUMMARY AND CONCLUSIONS
The objectives of this research included the installation
and
set-up
of
development
of
control
data
and
the
PHI
relevant
Auger
electron
software
analysis
spectroscope
for both
together
with
digital
the
study
and
process
of
the
surface compositions of the oxidized intermetallic compounds
of zirconium and nickel including ZrNi3, ZrNi, ZrNi5 and Zr2Ni7
and
oxidized
zirconium
and
nickel
foils.
The
significant
findings of this investigation are as follows:
1)
Since the operating conditions were maintained constant
during the sputtering of the' oxide layers formed on the metal
foils at 3OO0C, this
indicates a thicker oxide layer on Zr
layer than the oxide layer on nickel.
2) The nickel and zirconium concentrations in both metal foils
and intermetallics increased with sputter time, regardless of
the oxidation temperature, with a subsequent decrease in the
oxygen concentration.
3)
The
sputter
depth
profiles
of
the
intermetallics
are
consistent with those of similar investigations in literature
[14,15]. There is a discrepancy in the sputter depth profiles
of the oxidized intermetallics
(3OO0C) with recent findings.
As reported in chapter 7 this could be due to a variety of
reasons such as differences in operating conditions, oxidized
81
surface
cracks
and
also
surface
contamination
during
polishing.
4) The Nil peak intensity diminished by almost an order of
magnitude
on
oxidizing
the
intermetallics
at
SOO0C
in
comparison to the room temperature oxidized intermetallics„
This is a direct result of changing the operating conditions
from 3 keV to 2 keV which caused a change in the background
intensity.
The software developed to calculate the relative
peak intensities is not equipped to normalize for such large
changes in operating conditions.
5)
Since
the
temperature,
surface
outlined.
samples
i.e.
oxidation
SOO0C,
were
an
oxidized
improved
characteristics
only
at
one
understanding
cannot
be
of
high
the
accurately
82
CHAPTER 8
RECOMMENDATIONS FOR FURTHER RESEARCH
Based
on
the
results
of
this
experimental
work
the
following recommendations are made:
1)
Develop
information
zirconium and nickel
work
using
method
Seat's
relates
on
the
concentrations
of
from the results of this experimental
quantitative
the
depth
surface
Auger
region
method
mole
[26].
fractions
Seat's
of
two
components to their Auger signal strength. This can be used to
evaluate the surface atomic ratios of the two components which
provides a measure of the surface concentrations of the two
components.
2)
Maintain
constant
operating
conditions
during
an
investigation of this nature so that the results of different
depth profiles are directly comparable. This would require the
elimination of the background noise developed in the system.
The background noise can be eliminated by ensuring that the
electronics
in
the
power
supplies
connected
to
the
Cylindrical-Mirror Analyzer is performing satisfactorily. This
was
beyond
the
scope
of
the
present
investigation
but
maintaining constant operating conditions would require the
rectification
of
this
problem
before
further
research
is
undertaken.
3) Perform Auger analysis of the intermetallics of zirconium
83
with
t i n , chromium,
and
iron to
develop
the basis
for
an
improved understanding of the zircaloys over a wide range of
compositions, temperatures and extents of oxygen exposures.
4)
Estimate the thickness of the oxide layers which can be
related to the rate of formation of these layers to give a
better understanding of the kinetics of the development of the
surface strata during oxygen exposure.
84
REFERENCES
1) Deibert, M.C., "A study of Surface Oxidation Reactions on
the Intermetallic Compounds of Zirconium", Proposal submitted
to the N S F , 1988.
2) Wagner, Carl, "Theoretical Analysis of
Processes Determining the Oxidation Rate
Electrochem S o c .. 99 (10) (1952) 369 - 380.
the Diffusion
of Alloys",j.
3) Wagner, Carl, "Formation of Composite Scales Consisting of
Oxides of Different Metals".J. Electrochem S o c . . 103 (11)
(1956) 627 - 633.
4) Tapping, R.L., "X-Ray Photoelectron and Ultraviolet Studies
of the Oxidation and Hydriding of Zirconium", Journal of
Nuclear Materials. 107 (1982) 151 - 158.
5) Bluementhal, W.B., The Chemical Behavior of Zirconium, D .
Van Nostrand Co. Inc, N.Y., 1958.
6) Cox, B . , Oxidation
Press, 1978.
of Zirconium and
its All o y s . Plenum
7) Briggs, D. and Seah, M.P., Practical Surface Analysis by
Auger and X-Ray Photoelectron Spectroscopy, John Wiley & Sons,
1983 .
8) Anderson, R.B. and Dawson, P.T., Experimental Methods in
Catalytic Research. Academic Press, 1976.
9) Foord, J .S ., Goddard, P.J. and Lambert, R . M . , "Adsorption
and Absorption of Diatomic Gases by Zirconium Studies of the
Dissociation and Diffusion of CO, NO, N2, O2 and D2", Surface
Science. 94 (1980) 339 - 354.
10) Sen, P., Sarma, D.D., Budhani, R.C., Chopra, K.L. and R a o ,
C.N.R.,
"An electron spectroscopic study of the surface
oxidation of glassy and crystalline Cu-Zr alloys", J. Phvs. F ;
Met. Phvs.. 14 (1984) 565 - 577.
11) Kumar, L., Sarma, D.D. and Krummacher, S., "XPS Studies of
the Room Temperature Surface Oxidation of Zirconium and its
Binary Alloys with Tin, Chromium and Iron", Applied Surface
Science. 32 (1988) 309 - 319.
85
12) Bertolini, j.c .■ Brissot, J., LeMogne, T., Montes, H„ ,
Calvayrac, Y and Bigot, J ., "Glassy
Ni63^
Alloy:. OSurface
— — WJ
W , , 7Zr36 3
_ fT.j_.Luy
U J . a _ciue
I
-
-
I
Composition and Catalytic Properties in Hydrogenation of 1.3
Butadiene", Applied Surface Science. 29 (1987) 29 - 39
13) Cocke, J.L., Owens, M.S . and Wright, R.B., " The Surface
Oxidation
and
Reduction
Chemistry
of
Zirconium-Nickel
Intermetallic Compounds Estimated by XPS", Applied Surface
Science, 31 (1988) 341 - 369.
~
14) Deibert, M.C. and Wright, R.B., "The Surface Composition
of
Nickel-Zirconium
Intermetallic
Compound
Catalysts",
Catalysts— 1987., Proceedings of the IOth North American
Catalysis Society Meeting, San Deigo, May 1987, Elsevier
Science, Amsterdam.
15)
Deibert, M.C. and Wright, R .B ., "The Surface Composition
Initial Oxidation of Zirconium—Nickel
Intermetallic
Compounds at Room Temperature, Applied Surface Sciencp.. 35
(1988-89) 93 - 109.
'
16) Kofstad, P., High Temperature Oxidation of Meta l s . John
Wiley & Sons, Inc., 1966.
17)
Kingerey, W.D.,
Bowen,
H.K.
and
Uhlmann7
Introduction to Ceramics, John Wiley & Sons, 1976.
18) ^ Van _ Vlack, ^ L.H., Elements of Materials
Engineering, Addison-Wesley Publishing Company,
19) Perry, R.H. and Green,
Handbook. 6th e d . , 1984.
20) Shewmon, P.G.,
Company, 1963.
D.W.,
Diffusion
D.R.
Science
1985.
and
Perry's Chemical Engineers
in
Solids. McGraw-Hill
Book
21) Davis, L.E., MacDonald, N.C., Palmberg, P.W., Riach, G „E „
and Weber, R. E . , Handbook of Auger Electron Spectroscopy. 2nd
ed., Physical Electronics Industries, Inc., 1978.
22) Margaritondo, G . and Rowe, J .E ., Treatise on Analytical
Chemistry, part I, vol.8, John Wiley & Sons, 1986.
23) Riviere, J.C., Surface
Science Publications, 1990.
24) Joshi,, A., Davis, L.E.
■
^ ur^ace— Analysis, Elsevier
25) Seah,
(1979) 2.
M.P.
and Dench,
Analytical
Technioues. . Oxford
and Palmberg, P.W. , Methods of
Scientific Publishing Company,
W.A.,
Surface Interface A n a l . . I
86
26) Seah, M.P., Scanning Electron Microscopy. p„521
Ed. 0. Johari (SEM, AMF O'Hare, TL, 1983).
V o l .2,
27) Tapping, R.L., Davidson, R.D. and Jackman, T.E., Surface
Interface A n a l .. 7(1985) 105.
MONTANA STATE UNIVERSITY LIBRARIES