close

Вход

Забыли?

вход по аккаунту

?

1.4035331

код для вставкиСкачать
D. Artymowicz
Department of Chemical Engineering
and Applied Chemistry,
University of Toronto,
200 College Street,
Toronto, ON M5S 3E5, Canada
e-mail: [email protected]
C. Bradley
Department of Chemical Engineering
and Applied Chemistry,
University of Toronto,
200 College Street,
Toronto, ON M5S 3E5, Canada
B. Xing
Department of Chemical Engineering
and Applied Chemistry,
University of Toronto,
200 College Street,
Toronto, ON M5S 3E5, Canada
R. C. Newman
Department of Chemical Engineering
and Applied Chemistry,
University of Toronto,
200 College Street,
Toronto, ON M5S 3E5, Canada
Adhesion of Oxides Grown in
Supercritical Water on Selected
Austenitic and Ferritic/
Martensitic Alloys
A series of austenitic alloys (800H, H214, I625, 310S, and 347) with different surface finishes were exposed to supercritical water (SCW) at 550 C and 2.5 107 Pa for 120 h,
260 h, and 450 h in a static autoclave with an initial level of dissolved oxygen of 8 ppm.
Indentation with a hardness indenter was used for assessment of oxide adhesion. This
was compared with the results of a similar test on SCW-oxidized ferritic alloys. Delamination in all the tested ferritic alloys was insufficient for quantification of the results but
allowed for qualitative comparison within this group. In the set of austenitic alloys, oxide
on stainless steel (SS) 347 exfoliated during cooling from 550 C, and from the remaining
four alloys, only oxide on H214 delaminated, which made the qualitative comparison
across the whole group impossible. Energy dispersive X-ray spectroscopy (EDX) revealed
that under delaminated external Cr2O3 on H214 alloy, there was a submicron thick layer
of Al-rich oxide. To investigate a possible oxide spallation on austenitic samples during
exposure, mass loss obtained through descaling was compared with mass gain due to
SCW exposure. The results indicated that the applied descaling procedure did not, in
most cases, fully remove the scale. Apart from one case (SS 347 with alumina surface
finish), there was no clear indication of oxide spallation.
[DOI: 10.1115/1.4035331]
Introduction
The global energy demand is rapidly increasing, with some predictions of up to a 40% increase by 2040 [1]. Burning fossil fuels
for energy production has become untenable due to environmental
side effects. A focus on developing new, carbon neutral technologies is crucial for maintaining this balancing act between a healthy
environment and quality of life for future generations. Many carbon neutral technologies are in development, such as solar or
wind, but nuclear energy remains the most reliable and welldeveloped solution. Currently, 16% of Canada’s electricity is produced using nuclear power plants [2]. It is expected that nuclear
power will play a key role in the energy mix of the future.
The supercritical water reactor (SCWR) design is the focus of
Canada’s contribution to the development of generation-IV
nuclear technology. It is expected to benefit from a huge jump in
thermal efficiency up to 45% from current technology outputs of
33% for light-water reactors or 35% for Canada deuteriumuranium-6 (CANDU-6) reactors [3]. Along with this efficiency
increase, the SCWR is also attractive compared to current systems
because of its simpler “once-through” design, eliminating the
need for costly pieces of equipment, such as steam generators.
The higher heat capacity of supercritical water (SCW) translates
into a smaller required flow rate and reduced coolant tube size
along with other equipment reductions [4].
There are many challenges that must be overcome before the
application of SCWR technology, especially from a materials’
perspective. Of key interest is the integrity of the interface
between the protective corrosion layer and substrate of the proposed material in SCW environments. As materials corrode, it is
desirable to have a protective oxide layer develop on the surface
of the material which acts to inhibit further corrosion [5]. It is
important that this layer not only exists during steady-state
operation but is also mechanically stable on the surface. Significant spallation due to stresses within the oxide is potentially catastrophic, resulting in greatly increased corrosion rate. Spallation
may be induced by residual stress from the growth of oxide due to
differences in density between oxide and substrate, thermal
cycling, and the difference in between the coefficient of thermal
expansion of the scale and substrate, or even external loads. This
challenge is being explored in the second phase of Canada’s
SCWR materials research program.
The Rockwell indentation method is a technique for measuring
the interface toughness of brittle films on ductile substrates using
common laboratory equipment [6]. A Rockwell indenter with a
Brale indenter tip (120 deg diamond, conical) can be used to
induce delamination of the brittle film on the sample. By modeling the elastic–plastic response due to the indent, it is possible to
relate the interface toughness of the coating to the radius of
delamination caused by the indent [7]. The test is more likely to
be effective when the residual stress in the coating is compressive
and applies to coatings with thicknesses on the order of 105 m. A
qualitative evaluation of film adhesion is still possible on thinner
metallic substrates [8,9]. It was the purpose of this work to evaluate the applicability of the indentation to test the adhesion of
oxides formed in SCW.
Mass gain is often used as a measure of corrosion resistance but
the results are only valid if the oxide scale is not spalling. On the
other hand, mass loss measured with descaling is independent
from spallation. The descaling procedure has been successfully
used for quantification of oxidation in stainless steels [10]. In this
research, we intended to use the comparison of mass gain with
mass loss as a potential measure of oxide spallation on austenitic
alloys exposed to SCW similar to the work reported by Otoguro
et al. [11].
Experimental
Manuscript received May 30, 2015; final manuscript received October 28, 2016;
published online March 1, 2017. Assoc. Editor: Thomas Schulenberg.
Materials. The five austenitic alloys being evaluated for the
generation-IV program were purchased from Metal Samples Co.
Journal of Nuclear Engineering and Radiation Science
C 2017 by ASME
Copyright V
APRIL 2017, Vol. 3 / 021006-1
Downloaded From: http://nuclearengineering.asmedigitalcollection.asme.org/ on 10/26/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use
(Munford, AL) as rolled plates with a 120 grit surface finish. The
elemental composition of each alloy is shown in Table 1. H214
and I625 are austenitic nickel based alloys designed for superior
high-temperature oxidation resistance, and H214 also has a significant aluminum component. Incoloy 800 H is an
iron–nickel–chromium alloy designed for high-temperature applications; the “H” refers to the carbon content which lies in the
upper range of the 800 series requirements for improved creep
properties. Alloy 310 S is a low carbon version of the 310 series
known for ease of fabrication with high-temperature oxidation
resistance, while alloy 347 is a niobium stabilized stainless steel
(SS) with good oxidation properties and resistance to intergranular
corrosion. The ferritic/martensitic (F/M) alloys were provided by
CanmetMATERIALS as hot-rolled plates. Elemental composition
of F/M alloys is shown in Table 2.
SCW Exposure
Coupon Preparation. Coupons were cut from the supplied
sheets and plates to the nominal dimensions of
0.01 0.02 0.002 m with a 0.002 m diameter hole drilled near
one edge for mounting. For the austenitic alloys, three surface finishes were applied—120 grit (referred to as AR later in the text),
600 grit and fine polishing with 0.05 lm alumina (referred to as
ALU). Coupons cut from the F/M steels were all fine polished.
After polishing, coupons were degreased in acetone and sonicated
in ethanol.
SCW Exposure of Austenitic Alloy. The austenitic alloys were
exposed to the SCW conditions in the static-autoclave facilities at
CanmetMATERIALS. Three exposures have been done with
durations of 120 h, 260 h, and 450 h at 2.5 107 Pa and 550 C.
The uncertainties of time, temperature, and pressure were 0.5 h,
5%, and 5 C, respectively. The initial level of dissolved oxygen
(DO) was nominally 8 ppm and was achieved by saturating water
with air at room temperature. DO level was not monitored during
or after the exposure. The mass of each coupon was measured
three times immediately before and after exposure for weight gain
results using an analytical balance of 6106 g precision. The
exact dimensions of each sample were measured using a pair of
digital callipers with a precision of 6105 m.
SCW Exposure of F/M Alloy. The F/M coupons were exposed
to the SCW at 500 6 1 C and 2.5 107 Pa 60.5%, for
500 6 0.5 h in the flow-through SCW loop at the University of
New Brunswick, Fredericton, NB, with DO concentration maintained relatively low at 200 ppb. The DO was monitored with a
Hach EC Oxygen Orbisphere with a precision better than 1 ppb
DO. The mass of each coupon was measured three times immediately before and after exposure for weight gain results using an
analytical balance of 6106 g precision. The exact dimensions of
each sample were measured using a pair of digital callipers with a
precision of 6105 m.
Indentation. The oxide integrity was measured on 600 grit and
alumina finished samples by the Rockwell indentation method
using the indenter in the Materials Science and Engineering
Department at the University of Toronto, Toronto, ON. A 120 deg
diamond conical indenter (Brale) was employed. Indents with
loadings of 589 N (60 kgf), 981 N (100 kgf), and 1472 N (150 kgf)
were performed on each exposed sample of a given material and
surface finish type. Loading uncertainty is not known. It is, however, of no importance as indent dimension rather than the applied
force is used for estimation of the interface toughness.
The extent of delamination was determined on indented samples from scanning electron microscope (SEM) micrographs. The
micrograph of the indented surface is shown in Fig. 1(a). The two
features—the delamination zone and the indent—are outlined in
Figs. 1(b) and 1(c), respectively. Since the delamination was not
of uniform radius from the center of the indent, the area of the
delamination was measured and then an equivalent disk radius
was determined. The equivalent disk is an outer circle in Fig. 1(d),
while the inner circle corresponds to the indent.
Descaling. After determining the extent of delamination on the
indented samples, a descaling procedure was performed to remove
the oxide from the surface of the coupons to measure the total
weight lost due to corrosion. This corrosion indicator is considered preferable to simple weight gain because it circumvents
issues, such as deposition and exfoliation.
The descaling procedure [12] consists of submerging the samples in one of the two solutions at 90 C for 30 or 60 min at a time
until the oxide has been completely removed; solution compositions can be found in Table 3. Between each submersion, the coupons are sonicated in methanol for 15 min and weighed three
times using a balance with 6105 g of precision. An unexposed
coupon of each material type is included in the descaling procedure to act as a blank; the descaling is complete once the weight
lost between submersions for a coupon approaches that of the corresponding blank. Including the blank allows for a correction to
be applied to the result, which compensates for the amount of
bulk alloy dissolved into solution as a consequence of the descaling procedure. Initially, the samples being descaled are submerged in solution (a) for 60 min. Next, they are submerged in
solution (b) for 60 min. Then, the samples are submerged in solution (a) for 30 min at a time until completion.
Results and Discussion
Indentation of F/M Alloys. The F/M alloys were submitted to
the Rockwell indentation method for measuring oxide integrity.
The extent of delamination resulting from these tests is shown in
Fig. 2; r is the radius of delamination, and R is the radius of the
plastic region of the indent.
Delamination ratio (dr) calculated according to Eq. (1) was
evaluated with 1% of uncertainty
dr ¼
r
R
(1)
The lower limit of delamination ratio for the applicability of a
quantified measurement of interface toughness as calculated by
the Rockwell indentation method is 2.0 [7]. None of the F/M
alloys had a delamination ratio above this limit, thus only the
lower bound of the interface toughness can be estimated for the
oxide grown on these materials, that is, the interface toughness is
calculated for a delamination ratio of 2. The results are
Table 1 Candidate austenitic alloys composition (wt.%)
Alloy
Fe
Ni
Cr
C
Si
Mn
Mo
Al
Ti
N
P
S
H214
347
I625
800 H
310 S
3.58
Bal
4.07
46.3
Bal
Bal
9.08
61.1
31.0
19
16.24
17.24
22.04
20.0
24.5
0.04
0.05
0.04
0.07
0.08
0.05
0.61
0.2
0.30
0.75
0.21
1.71
0.18
0.80
2.0
<0.1
0.47
8.35
—
—
4.15
—
0.19
0.44
—
<0.0027
—
0.21
0.52
—
—
—
—
0.01
0.11
0.002
0.028
0.008
0.012
0.045
<0.002
0.0001
0.0001
0.001
0.015
021006-2 / Vol. 3, APRIL 2017
Transactions of the ASME
Downloaded From: http://nuclearengineering.asmedigitalcollection.asme.org/ on 10/26/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use
Table 2 Ferritic/martensitic alloy composition (wt.%)
Alloy
Fe
Cr
C
Si
Mn
Al
V
Ti
N
O
P
S
Fe–9Cr–1.5Al
Fe–9Cr–1.5Si
Fe–9Cr
Fe–13Cr–1.5Si
Fe–9Cr–1.5V
Fe–13Cr
Bal
Bal
Bal
Bal
Bal
Bal
9.33
9.65
8.65
14.15
8.68
13.9
0.09
0.1
0.14
0.09
0.17
0.13
<0.2
1.22
<0.2
1.49
<0.19
<0.2
0.2
<0.05
<0.05
<0.05
<0.05
<0.05
1.25
0.12
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
1.71
<0.01
0.01
0.01
n/a
0
0.01
0
0.09
0.06
0.04
0.09
0.11
0.17
0.031
0.127
n/a
0.058
2.492
0.21
0.005
0.007
0.002
0.005
<0.009
0.007
0.002
0.002
0.004
0.002
0.003
0.002
Fig. 1 Example of the extent of delamination determination, Fe–13Cr–1.5Si ferritic/martensitic alloy. The inner and outer
circles on the micrograph (d) represent the plastic zone of radius R and the equivalent delamination zone of radius r,
respectively.
Fig. 2 Extent of delamination of ferritic/martensitic alloys (dashed line separates 9% Cr from
13% Cr alloys)
summarized in Table 4 together with model and material parameters that were used in estimation.
A relative interface toughness can still be inferred from the
results of the indentation tests; however, it is important to note
that the mode with which the oxide delaminates differs between
materials and different modes of delamination are expected to
have an impact on the protection from further corrosion provided
by the delaminated region.
The morphology of the oxides grown on F/M alloys typically
shows a bilayer structure with the original metal surface acting as
the interface between the inner and outer oxides [15]. Oxide scale
on Fe-13Cr-1.5Si delaminated along inner/outer oxide interface;
only the outer oxide was detached, leaving the inner oxide to offer
some protection from further general corrosion. This delamination
mode is depicted in Fig. 3(a).
Journal of Nuclear Engineering and Radiation Science
Scales on all the remaining F/M alloys delaminated along both
interfaces. An example of this delamination mode is shown in Fig.
3(b).
The Fe-13Cr-1.5Si in the only F/M alloy that delaminated in
mode depicted in Fig. 3(a). It has also the highest extent of the
Table 3
Descaling solutions
Solution (a)
2.0%
5.0%
0.5%
Solution (b)
Citric acid
Dibasic ammonium citrate
Disodium EDTA
10.0%
4.0%
Potassium permanganate
Caustic soda
APRIL 2017, Vol. 3 / 021006-3
Downloaded From: http://nuclearengineering.asmedigitalcollection.asme.org/ on 10/26/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use
Table 4 List of model and material parameters and the resulting lower bound interface toughness G. eo is the residual stress in
the substrate before indentation, ry/Ey is the yield stress to Young’s modulus ratio for the substrate, N is the strain-hardening coefficient of the substrate, and Eo and ho are Young’s modulus and thickness of the oxide, respectively.
Model
Substrate
Oxide
Interface toughness
Alloy type
e0
ry/Ey
N
Type
Eo (Pa)
ho (m)
G (J/m2)
F/M
Austenitic
0
0
0.005
0.005
0.1
0.1
Magnetite
Chromia
2.5 1011 [13]
2.73 1011 [14]
1 105
2 106
150
32
Fig. 3 Delamination modes observed in (a) Fe–13Cr–1.5Si and (b) Fe–9Cr. The capital letters
designate: A—the outer, Fe-rich oxide; B—outer/inner oxide interface (this is the original
surface of the coupon); C—the inner mixed Cr-rich oxide; and D—inner oxide/metal interface.
Fe-13Cr-1.5Si was the only F/M alloy with delamination mode depicted in Fig. 3(a). All remaining F/M alloys delaminated as shown in Fig. 3(b). The scale bar is 50 lm.
delamination which is probably due to delamination progressing
along one interface only.
Indentation of Austenitic Alloys. The candidate alloys 800 H,
I625, and 310 S after the 450 h exposure showed almost zero
delamination. It was therefore not possible to extract even relative
oxide integrity from the indentation results for these alloys, and a
more sensitive evaluation of oxide integrity will be needed to differentiate the oxide integrity of these candidate materials.
A closer inspection of the surface of the indented 310 S sample
showed very thin cracks extending radially from the indent but no
delamination of the oxide from the surface, as shown in Fig. 4.
The radial cracks are indicated with the short white arrows.
A coupon of alloy 347 with alumina finish showed partial oxide
exfoliation (Fig. 5) even when not subject to indentation. The
exfoliation happened most probably during cooling rather than
during exposure as no visible oxide was observed on the uncovered metal surface.
Alumina and 600 grit finished H214 samples delaminated under
three applied loads with a delamination ratio of just under 2.0 (Fig.
6). Estimated interface toughness and a list of model and material
parameters used in estimation are summarized in Table 4. It should
Fig. 4 Indented 310S
delamination
Fig. 5 SEM micrograph of alumina finished coupon of SS 347
after 450 h exposure to SCW showing exfoliation
sample showing
021006-4 / Vol. 3, APRIL 2017
cracks without
Transactions of the ASME
Downloaded From: http://nuclearengineering.asmedigitalcollection.asme.org/ on 10/26/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use
Fig. 6 SEM micrograph of indented exposed H214 sample with
alumina fine polished surface. The white line shows the delamination extent.
be noted that Young’s modulus for polycrystalline chromia (isotropic) may not represent the elastic properties of the chromia layer
composed of plates.
The delamination in H214 progressed along the interface
between an externally grown layer of chromia plates of approximate thickness of 5 106 m and a metal substrate. This is clearly
visible in Fig. 7 showing a piece of oxide peeled off by delamination and rotated about 90 deg out of the plane of the page.
The cross-sectional view of the exposed H214 coupon with 600
grit surface finish is shown in Fig. 8. Starting at the upper edge of
the image, four regions can be distinguished as follows: externally
grown Cr2O3 chromia plates, a submicron thick layer of Al-rich
oxide—probably Al2O3, a Cr depletion zone extending less than
2 106 m into the sample bulk, and the unoxidized metal. As
the interface between the unoxidized metal and Cr depletion zone
is difficult to detect, a fragment of this interface is indicated with
solid black arrows.
EDX signals from Cr, Al, and O along the line normal to the
surface are shown in Fig. 9.
Weight Gain Versus Weight Loss in Austenitic Alloys. A
schematic representation of an oxidized metal surface is shown in
Fig. 10. With variables introduced in Fig. 10, mass gain (mg) and
mass loss (ml) were defined by Eqs. (2) and (3), respectively.
Additionally, coefficients a, b, and c were defined using Eqs.
(4)–(6). The values for a for Fe, Cr, Ni, and Al oxides are listed in
Table 5.
mg ¼ m1 m0
(2)
ml ¼ m0 m2
(3)
a¼
b¼
Fig. 7 Partially delaminated oxide on indented H214 alumina
polished coupon
o
Mox
m
Mox
(4)
ex
Mox
o
þ Mox
(5)
mg
ml
(6)
m
Mox
c¼
m
)
Under assumption of full oxide removal by descaling (ml ¼ Mox
and allowing for spallation, c can be expressed as follows:
Fig. 8 SEM micrograph of a cross section of 600 grit finished H214 after 450 h SCW exposure.
A segment of the interface between the Cr-depleted zone and the unoxidized metal is indicated
with a set of black arrows.
Journal of Nuclear Engineering and Radiation Science
APRIL 2017, Vol. 3 / 021006-5
Downloaded From: http://nuclearengineering.asmedigitalcollection.asme.org/ on 10/26/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use
Fig. 9
3 lm.
EDX line scan taken on a cross section of the 600 grit finished H214 coupon exposed to SCW for 4150 h. The size bar is
Fig. 10 A schematic representation of an oxidized surface of a metal. The mass of the oxide
m
o
was separated into that of metallic elements (Mox
) and oxygen (Mox
). m0, m1, and m2 represent
the mass of an unexposed, exposed, and descaled coupon, respectively. A section of the oxide
ex
) indicates the part of oxide that exfoliated during the
marked with gray net pattern (Mox
exposure.
c¼
o
o
m
Mox
bðMox
þ Mox
Þ
¼ að1bÞb
m
Mox
(7)
The possible values of c for different oxide behaviors are summarized in Table 6. The values of c for different exposure time/alloy/
surface finish combinations are plotted in Fig. 11.
The majority of data points in Fig. 11 are located above the
upper bound of a indicating an incomplete oxide removal by the
descaling procedure. One data point (347_AR_240 h) is within
the upper and lower bounds of a indicating some or no exfoliation
depending on whether the lower bound a is that of NiO or magnetite. One data point—347_ALU_450 h—is below the lowest possible value of a. This is a clear sign of exfoliation, in agreement
with Fig. 5.
It is worth noticing that incomplete descaling, although
observed in all the alloys, is most pronounced in alloy H214.
Table 5 Stoichiometry of common oxides of Cr, Fe, and Ni. a is
the ratio of oxygen mass to metal mass.
Oxide
Fe2O3
Fe3O4
FeCr2O4
Cr2O3
NiO
Al2O3
021006-6 / Vol. 3, APRIL 2017
a
0.43
0.38
0.40
0.46
0.27
0.89
Alloy H214 relies on both Cr and Al, to form a protective scale,
the expected phase sequence being Cr2O3–Al2O3–metal similar to
the one forming in high-temperature oxidation [16]. The descaling
solution used in this study has been originally designed for SS and
is known to dissolve crystalline alumina very slowly.
Conclusions
Indentation with a Rockwell hardness indenter was used to estimate oxide adhesion on a series of F/M and austenitic alloys
exposed to SCW. All the F/M alloys delaminated around the
indents although the delamination zone was not large enough to
allow for a quantification of the oxide/metal interface toughness.
The austenitic alloys show stronger oxide adhesion as measured
by the Rockwell indentation test compared to the F/M alloys. The
exception is the H214 with an alumina-polished surface finish
which performed about the same as the F/M alloys, and SS 347
with alumina finish which exfoliated during SCW exposure. Alloy
310 S showed radial cracks around the indent but no delamination
Table 6 Possible values of c for different oxide behaviors
Value of c
Oxide behavior
1
1 < c < a
c¼a
c>a
Full exfoliation
Partial exfoliation
No exfoliation
Incomplete descaling
Transactions of the ASME
Downloaded From: http://nuclearengineering.asmedigitalcollection.asme.org/ on 10/26/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use
the SCW exposures. The authors greatly appreciate the financial
support from the Natural Science and Engineering Research
Council (NSERC).
Nomenclature
dr ¼
Eo ¼
Ey ¼
G¼
h¼
ho ¼
m0 ¼
m1 ¼
m2 ¼
mg ¼
ml ¼
m
¼
Mox
delamination ratio ðr=RÞ
Young’s modulus of the oxide, Pa
Young’s modulus of the substrate, Pa
interface toughness, J/m2
exposure time, h
oxide thickness, m
mass of the unoxidized coupon, g
mass of the oxidized coupon, g
mass of the descaled coupon, g
mass gain ðm1 m0 Þ, g
mass loss ðm0 m2 Þ, g
mass of metal ions in the oxide, g
o
Mox
¼ mass of oxygen ions in the oxide, g
ex
¼
Mox
N¼
r¼
R¼
mass of the exfoliated oxide, g
strain-hardening factor
equivalent delamination radius, m
indent radius, m
Greek Symbols
Fig. 11 Mass gain to mass loss ratio for austenitic alloys
exposed to SCW. Upper and lower graphs show data for coupons with as-received and alumina surface finish, respectively.
The exposure numbers 1, 2, and 3 correspond to exposure
times 120 h, 260 h, and 450 h, respectively, and align with the
vertical grid lines: dotted for 120 h, dashed–dotted for 260 h,
and dashed for 450 h. The two continuous gray horizontal lines
indicate the upper (Cr2O3) and lower (NiO) bounds of a excluding Al2O3. The horizontal-dashed line indicates the value of a
for Al2O3. Majority of measurements were performed on three
samples, and error bars represent the standard deviation of the
mean. In four cases, the measurements were performed on a
single sample and no error could be associated with the
reported results. These four cases are S347_AR_120 h,
S347_AR_450 h, S347_ALU _450 h, and I625_ALU_240 h.
of the oxide, which suggest that the interface between the oxide
and substrate is tougher than the oxide itself in this case. A more
sensitive adhesion test should be performed on the candidate
alloys to determine their relative oxide integrity.
The oxide scale on H214 has a bilayered structure with an
external Cr2O3 and an internally grown submicron layer of Alrich oxide (most likely Al2O3). The delamination progressed
along the interface between these two oxides. Directly under the
oxide scale, there is a Cr depletion zone extending to a maximum
of 2 106 m into the bulk and a narrower Al depletion zone.
Comparison between the weight gain and descaled weight loss
results of the austenitic alloys shows that in most cases, the descaling fails to remove all the oxides. This is most obvious for the
H214 alloy and is related to the presence of Al-rich oxide.
For the studied exposure time, there was no clear indication of
oxide spallation for any austenitic samples except the alumina finished SS 347.
Acknowledgment
We would like to thank the Department of Chemical Engineering of the University of New Brunswick and CanmetMaterials
Laboratory of Natural Resources Canada (NRCan) for performing
Journal of Nuclear Engineering and Radiation Science
a ¼ ratio of mass of oxygen ions to mass of metal ions in
o
m
=Mox
Þ
oxide ðMox
ex
m
o
=Mox
þ Mox
Þ
b ¼ mass fraction of exfoliated oxide ðMox
c ¼ mass gain to mass loss ratio ðmg=mlÞ
e0 ¼ residual stress in the substrate before indentation, Pa
ry ¼ yield stress of the substrate, Pa
Acronyms and Abbreviations
ALU ¼
AR ¼
CANDU ¼
DO ¼
EDX ¼
F/M ¼
NRCan ¼
NSERC ¼
SCW ¼
SCWR ¼
SEM ¼
SS ¼
surface fine polished with 0.05 lm alumina powder
surface left in the as-received state
Canada deuterium uranium reactor
dissolved oxygen
energy dispersive X-ray spectroscopy
ferritic martensitic
Natural Resources Canada
Natural Science and Engineering Research Council
supercritical water
supercritical water reactor
scanning electron microscope/microscopy
stainless steel
References
[1] OECD/IEA, 2015, “World Energy Outlook,” International Energy Agency,
Paris, accessed Feb. 1, 2016, https://www.iea.org/Textbase/npsum/
WEO2015SUM.pdf
[2] Statistics Canada, 2016, “CANSIM Table.127-0002,” Ottawa, ON, Canada,
accessed Mar. 1, 2016, http://www5.statcan.gc.ca/cansim/a26?lang=eng&id
=1270002
[3] Parent, E., 2003, “Nuclear Fuel Cycles for Mid-Century Deployment,” M.A.Sc.
thesis, Massachusetts Institute of Technology, Cambridge, MA.
[4] Wolf, J., 2011, “Supercritical Water Reactor,” Nuclear Energy Encyclopedia,
Wiley, Hoboken, NJ, pp. 305–308.
[5] Fontana, M. G., 1986, Corrosion Engineering, McGraw-Hill, New York.
[6] Monteiro, M. J., Saunders, S. R. J., and Rizzo, F. C., 2011, “The Effect of
Water Vapour on the Oxidation of High Speed Steel, Kinetics and Scale
Adhesion,” Oxid. Met., 75(1–2), pp. 57–76.
[7] Drory, M. D., and Hutchinson, J. W., 1996, “Measurement of the Adhesion of a
Brittle Film on a Ductile Substrate by Indentation,” Proc. Math. Phys. Eng.
Sci., 452(1953), pp. 2319–2341.
[8] Ramirez, M. A. R., Silva, P. C., Corat, E. J., and Trava-Airoldi, V. J., 2015,
“An Evaluation of the Tribological Characteristics of DLC Films Grown on
Inconel Alloy 718 Using the Active Screen Plasma Technique in a Pulsed-DC
PECVD System,” Surf. Coatings Technol., 284, pp. 235–239.
[9] Bhattacherjee, S., Niakan, H., Yang, Q., Hu, Y., and Dynes, J., 2015,
“Enhancement of Adhesion and Corrosion Resistance of Diamond-Like Carbon
Thin Films on Ti-6Al-4V Alloy by Nitrogen Doping and Incorporation of
Nanodiamond Particles,” Surf. Coatings Technol., 284, pp. 153–158.
APRIL 2017, Vol. 3 / 021006-7
Downloaded From: http://nuclearengineering.asmedigitalcollection.asme.org/ on 10/26/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use
[10] Guzonas, D. A., and Cook, W. G., 2012, “Cycle Chemistry and Its Effect on
Materials in a Supercritical Water-Cooled Reactor: A Synthesis of Current
Understanding,” Corros. Sci., 65, pp. 48–66.
[11] Otoguro, Y., Sakakibara, M., Saito, T., Ito, H., and Inoue, Y., 1988, “Behavior
of Austenitic and High Heat-Resisting Steam Steels in a Temperature Pressure
Environment,” Trans. ISIJ, 28(9), pp. 761–768.
[12] Guzonas, D. A., 2014, personal communication.
[13] Chicot, D., Mendoza, J., Zaoui, A., Louis, G., Lepingle, V., Roudet, F.,
and Lesage, J., 2011, “Mechanical Properties of Magnetite (Fe3O4), Hematite (a-Fe2O3) and Goethite (a-FeOOH) by Instrumented Indentation
021006-8 / Vol. 3, APRIL 2017
and Molecular Dynamics Analysis,” Mater. Chem. Phys., 129(3),
pp. 862–870.
[14] Tortorelli, P. F., 1993, “Mechanical Properties of Chromia Scales,” J. Phys. IV,
3(C9), pp. 943–949.
[15] Li, W., 2012, “Oxidation Resistance and Nanoscale Oxidation Mechanism in
Model Binary and Ternary Alloys Exposed to Supercritical Water,” M.A.Sc.,
thesis, University of Toronto, Toronto, ON.
[16] Wood, G. C., Richardson, J. A., Hobby, M. G., and Boustead, J., 1969, “The
Identification of the Thin Healing Layers at the Base of Oxide Scales on Fe-Cr
Base Alloys,” Corros. Sci., 9(9), pp. 659–671.
Transactions of the ASME
Downloaded From: http://nuclearengineering.asmedigitalcollection.asme.org/ on 10/26/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use
Документ
Категория
Без категории
Просмотров
2
Размер файла
3 635 Кб
Теги
4035331
1/--страниц
Пожаловаться на содержимое документа