semi合集-English.pdf - 第4196页

SEMI F72-1102 © SEMI 2002 7 Theoret i cal 0 10 20 30 40 50 60 70 80 0 5 1 0 15 20 25 30 35 40 45 Angstr oms Atomic Concentrati on Cr Fe Ni Mo O C Figure A1-4 Composition Versus Depth of a Pure Cr 2 O 3 Layer on Stainless…

100%1 / 7923
SEMI F72-1102 © SEMI 2002 6
in Figure A1-2. Although Auger and ESCA are
considered surface analytical tools, their depth of
analysis is on the order of 50 angstroms, or about one to
three times the depth of the oxide film found on a
typical passivated stainless steel surface.
Figure A1-2
The Electron Escape Depth of Low Energy
Electrons in Some Metals
A1-3 Depth of Analysis Function
A1-3.1 There has been some discussion with respect
to the “Depth of Analysis Function”. There is one
school that believes that it is 5 IMFPs and another that
uses 3 IMFPs. These two functions will be referred to
as 5λ and 3λ respectively. It will be seen that it makes
little difference which function is used even though the
depth of the oxide films is of the order of the functions.
These functions are seen in Figure A1-3, which shows
the proportion of the total signal generated from each
depth below the surface. The Y-axis is in decimal and
the X-axis is in angstroms from the surface. The first
point of the 5λ curve is 0.16 at the 1
st
angstrom level.
That means that 16% of the signal comes from the 1
st
angstrom, 14% from the 2
nd
angstrom level and so on
down to 1% from the 25
th
angstrom level. Thus
elements that do not appear on the surface or in the first
24 angstroms in from the surface, but do exist at the 25
angstrom level will be detected.
5 Lambda
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
1 2 3 4 5 6 7 8 9 10111213141516171819202122232425
Angstroms
3 Lambda
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
1 2 3 4 5 6 7 8 9 10111213141516171819202122232425
Angstroms
Figure A1-3
These Two Graphs Represent the “Depth of
Analysis” That Will be Seen When Analyzing 1 keV
Electrons. The Top Curve is the 5λ Curve and
Bottom the 3 λ.
A1-4 Modeling
A1-4.1 Figure A1-4 is the composition profile of a
“perfect” oxide. It represents 25 angstroms of pure
Cr
2
O
3
on an atomically flat 316L stainless steel surface,
with no surface or interfacial contamination present.
Figures A1-5a and b are the theoretical depth profiles
derived by assuming a depth of analysis of 5λ and 3λ
respectively. The first factor to note is that Fe, Ni and
Mo appear most or all of the way through the depth
profile of the oxide due to the contribution of the metal
substrate to the detected signal, the “depth of analysis”
effect as introduced above. 1 keV electrons can escape
from as deep as 25 angstroms. This must be
appreciated when interpreting these depth profiles.
SEMI F72-1102 © SEMI 2002 7
Theoretical
0
10
20
30
40
50
60
70
80
0 5 10 15 20 25 30 35 40 45
Angstroms
Atomic Concentration
Cr Fe Ni Mo O C
Figure A1-4
Composition Versus Depth of a Pure Cr
2
O
3
Layer
on Stainless Steel
A1-4.2 The second factor to notice is the thickness of
the oxide. Historically, the FWHM (Full Width Half
Maximum; the width of a peak measured at half its
maximum height) of the oxygen profile has been used
as the measure of oxide thickness. It is apparent in
Figures A1-4 and A1-5 that this underestimates the
oxide thickness by 15 to 20 %.
A1-4.3 It should also be noted that the Cr:Fe ratio
measured from Figures A1-4 and A1-5 has its
maximum value at the initial surface, decreasing as the
depth profile progresses, whereas the actual Cr:Fe ratio
in Figure A1-4 is infinite down to 25 angstroms. Most
actual depth profiles exhibit a maximum of the Cr:Fe
ratio at some depth below the initial surface, referred to
as the depth of maximum enrichment. This is a
consequence of variation in the actual composition of
the oxide, possibly having a higher concentration of Fe
near the surface, a phenomenon commonly termed a
“detached iron oxide layer”, or of contamination on the
surface, generally hydrocarbons. These are discussed
in the next section
A1-5 Effects of Detached Iron Oxide Layer
and Surface Contamination
A1-5.1 The model composition profile of Figure A1-6
shows 3 angstroms of pure Fe
2
O
3
over 22 angstroms of
pure Cr
2
O
3
on 316L stainless steel. This is a model of a
detached iron oxide layer. Figure A1-7 is the
theoretical depth profile of this model derived assuming
a depth of analysis of 5λ. Note that the initial Cr value
in the profile is higher than the Fe value, even though
the surface is pure Fe
2
O
3
. This is a consequence of the
depth of analysis detecting the Cr from levels beneath
the surface. The Cr:Fe ratio for this model has its
maximum value at 3 angstroms below the initial
surface. The theoretical depth profile using a depth of
analysis of 3λ is very similar
Convoluted 5 Lambda
0
10
20
30
40
50
60
70
80
90
100
0 5 10 15 20 25 30 35 40 45
Angstroms
Atomic Concentration
0
20
40
60
80
100
120
140
Cr Fe Ni Mo O C Cr:Fe
Figure A1-5a
Theoretical Depth Profile of Figure 5 Assuming 5λ
Depth of Analysis
Convoluted 3 Lambda
0
10
20
30
40
50
60
70
80
90
100
0 5 10 15 20 25 30 35 40 45
Angstroms
Atomic Concentration
0
100
200
300
400
500
600
700
Cr Fe Ni Mo O C Cr:Fe
Figure A1-5b
Theoretical Depth Profile of Figure 5 Assuming 3λ
Depth of Analysis
A1-6 Effects of Hydrocarbon Contamination
A1-6.1 The model composition profile of Figure A1-8
shows 3 angstroms of pure carbon over 22 angstroms of
pure Cr
2
O
3
on 316L stainless steel, representing an
idealized model of the adsorbed hydrocarbon
contamination generally found on stainless steel
surfaces exposed to the atmosphere. Figure A1-9 is the
theoretical depth profile of this model assuming a depth
of analysis of 5λ. In this case the O, Cr and Fe atomic
concentration values are reduced by the presence of the
carbon until the depth profiling proceeds past the
carbon, but they have the same relative values (ie: same
Cr:Fe ratio) versus depth as derived in the
uncontaminated model. The maximum of the Cr:Fe
ratio is seen to be at the initial surface. Although the
actual oxide thickness is less in this model, the FWHM
measure of the oxide thickness from the depth profile is
the same due to the presence of the carbon layer. Note
SEMI F72-1102 © SEMI 2002 8
that the Oxygen concentration profile initially increases
to a maximum, then decreases. This is typical of
Oxygen concentration profiles seen on actual samples,
which will generally have some adsorbed hydrocarbon
contamination on the surface.
Theoretical
0
10
20
30
40
50
60
70
80
90
100
0 5 10 15 20 25 30 35 40 45
Angstroms
Atomic Concentration
Cr Fe Ni Mo O C
Figure A1-6
Composition Versus Depth of a Fe
2
O
3
Layer Over a
Cr
2
O
3
Layer on Stainless Steel
Convoluted 5 Lambda
0
10
20
30
40
50
60
70
80
90
100
0 5 10 15 20 25 30 35 40 45
Angstroms
Atomic Concentration
0
5
10
15
20
25
Cr Fe Ni Mo O C Cr:Fe
Figure A1-7
Theoretical Depth Profile of Figure A1-6 Assuming
5λ Depth of Analysis
Theoretical
0
10
20
30
40
50
60
70
80
90
100
0 5 10 15 20 25 30 35 40 45
Angstroms
Atomic Concentration
Cr Fe Ni Mo O C
Figure A1-8
Composition Versus Depth of a Carbon Layer Over
a Cr
2
O
3
Layer on Stainless Steel
Convoluted 5 Lambda
0
10
20
30
40
50
60
70
80
90
100
0 5 10 15 20 25 30 35 40 45
Angstroms
Atomic Concentration
0
10
20
30
40
50
60
70
80
90
Cr Fe Ni Mo O C Cr :Fe
Figure A1-9
Theoretical Depth Profile of Figure A1-8 Assuming
5λ Depth of Analysis
A1-7 Effects of Hydrocarbon Contamination
over Detached Iron Oxide Layer
A1-7.1 The model composition profile of Figure A1-
10 shows a 3 angstrom carbon layer over 3 angstroms
of pure Fe
2
O
3
over 19 angstroms of pure Cr
2
O
3
on 316L
stainless steel. Figure A1-11 is the theoretical depth
profile of this model assuming a depth of analysis of 5
λ. This profile is seen to be similar to the depth profiles
generally observed for passivated stainless steel. The
Cr:Fe ratio maximum occurs below the initial surface,
and the initial surface atomic concentrations of the
elements of interest are diluted by the presence of the
carbon layer on the surface. The oxide thickness, as
measured by the FWHM technique, is affected by both