Corrosion behavior of sialon ceramics in supercritical water

Article abstract of DOI:10.1111/j.1551-2916.2006.01243.x

The corrosion behavior of sialon ceramics was investigated in supercritical water at 450 °C under 45 MPa for 2-50 h. α-sialon exhibited better corrosion resistance than β-sialon and α/β-sialon. Pitting corrosion with the formation of corrosion products wa

Full text of DOI:10.1111/j.1551-2916.2006.01243.x

J. Am. Ceram. Soc., 89 [11] 3550–3553 (2006)
DOI: 10.1111/j.1551-2916.2006.01243.x
r 2006 The American Ceramic Society
ournal
J
Corrosion Behavior of Sialon Ceramics in Supercritical Water
w
z
y
Masahiro Nagae, Yasunori Koyama, Seita Yasutake, and Tetsuo Yoshio
Graduate School of Environmental Science, Okayama University, Okayama 700-8530, Japan
Kohei Oda
Department of Materials Science, Yonago National College of Technology, Yonago 683-8502, Japan
The corrosion behavior of sialon ceramics was investigated in
supercritical water at 4501C under 45 MPa for 2–50 h. a-sialon
exhibited better corrosion resistance than b-sialon and a/b-si-
alon. Pitting corrosion with the formation of corrosion products
was observed in the case of b-sialon and a/b-sialon. By contrast,
the corrosion behavior of a-sialon was characterized by uniform
corrosion with the formation of corrosion products. The degree
of strength deterioration was strongly dependent on the corro-
sion morphology. The bending strength of a-sialon after corro-
sion for 30 h was about 90% of its initial strength, while the
strength of b-sialon decreased to 65% of its original strength.
a-sialon and sintering aids without any remaining grain bound-
ary phase. The aim of this study is to investigate the influence of
the presence of the grain boundary phase in sialon ceramics on
the corrosion behavior and the degree of strength deterioration.
II. Experimental Procedure
Three kinds of sialon, a-, a/b-, and b-sialon, were used in this
study. The chemical compositions of the specimens are listed in
Table I. a-sialon is an interstitial solid solution with Y
does not have a grain boundary phase. a/b-sialon is a composite
of the a- and b-phases. b-sialon contains 5 wt% Y as
a sintering aid. The specimens were prepared for bending tests
2 3
O and
2 3
O
I. Introduction
12
in accordance with JIS R1601.
Rectangular specimens
L
W
T
UPERCRITICAL WATER OXIDATION (SCWO) is a promising
method for decomposing dioxin, PCB, and other materials,
which cannot be easily decomposed by conventional processes,
(7.5 ꢀ 4 ꢀ 3 mm) were cut out from the sialon, ultrasonic-
ated in acetone, and dried in a desiccator before weight meas-
urement. The corrosion tests were performed in a chamber made
of Hastelloy C-22 (TAS-01, Taiatsu Techno Corporation,
Tokyo, Japan) at 4501C under 45 MPa for 2–50 h. The eluate
was redistilled water (pH 5 5.6 as measured with a pH meter).
The filling ratio of redistilled water was 34.33% to attain the
pressure of 45 MPa at 4501C on the basis of the relationship
S
1
,2
as well as for chemical recycling of waste plastic products. Ni-
based super alloys, such as Inconel and Hastelloy, have been
considered as reactor materials. However, to date, no acceptable
reactor materials have been developed that can withstand the
severe supercritical water (SCW) conditions containing the cor-
rosive HCl and HF generated by decomposition reactions. In
the middle of the 1990s, ceramics began to be considered as
candidates for SCWO reactors instead of Ni-based super al-
1
3
between temperature and pressure by Kennedy. The V/A ratio
was set at 52 cm (V, volume of the chamber; A, geometrical
surface area of the test specimen). After corrosion testing, the
specimens were dried at 1101C and then cooled naturally in a
desiccator. Once the specimens cooled down to ambient tem-
perature, their weight change and dimensional change were
measured. X-ray diffraction (XRD; Rigaku RINT 2000,
Tokyo, Japan) was performed with CuKa radiation for phase
identification. Corrosion behavior was examined by scanning
electron microscopy (SEM; JEOL JSM-6330F, Tokyo, Japan).
In order to evaluate the degree of strength deterioration, three-
point bending tests were performed at room temperature at a
crosshead speed of 0.5 mm/min. The span of the jig was 30 mm.
3
,4
5
loys. Hara and Sugimoto carried out a screening test of 18
kinds of ceramics and demonstrated that oxide ceramics had
better corrosion resistance than non-oxide ceramics in an SCW
environment. However, the details of the corrosion mechanism
of each ceramic in an SCW environment were not fully clarified.
We have conducted a series of fundamental studies on the
corrosion behavior of various ceramics in a high-temperature
water environment under equilibrium vapor pressure, simulat-
ing a light-water reactor, to investigate the influence of sintering
6–9
methods and sintering aids on corrosion behavior. Our recent
study on the corrosion behavior of representative structural ce-
3 4 2 3
ramics, such as Si N , SiC, and Al O , in an SCW environment
has revealed that the presence of grain boundary phases was
III. Results and Discussion
1
0
detrimental to the corrosion resistance of these ceramics.
In this paper, we report the corrosion behavior of sialon ce-
ramics in an SCW environment. Sialon has two different crys-
Figure 1 shows the time dependence of the weight loss after
corrosion in SCW. The weight loss of all the specimens tended to
saturate with increasing corrosion time. As can be seen, the
weight loss of a-sialon, which has no grain boundary phase, was
lower than those of a/b- and b-sialon. A similar trend was ob-
served in the case of high-temperature high-pressure water at
1
1
tal structures, a- and b-sialon, which have the same structure as
3 4
a- and b-Si N , respectively. Grain boundary phases, such
as Y Al 12, remain in b-sialon after sintering, whereas it is
3
5
O
theoretically possible to form an interstitial solid solution of
1
4
3
ance of sialon is strongly affected by the presence of the grain
001C under 8.6 MPa. This suggests that the corrosion resist-
N. Jacobson—contributing editor
boundary phase.
Figure 2 shows an SEM image of the surface of b-sialon after
corrosion for 30 h. The specimen surface was covered with crys-
talline corrosion products. Similar corrosion products were
identified on the surface of a- and a/b-sialon. In order to anal-
yze the corrosion products, fracture surfaces of specimens after
corrosion were examined with an energy dispersive X-ray
Manuscript No. 21756. Received May 1, 2006; approved June 3, 2006.
Author to whom correspondence should be addressed. e-mail: nagae@cc.okayama-
w
u.ac.jp
z
y
Present address: Mate Co. Ltd., 526-3 Saeki, Saeki-cho, Wake 709-0514, Japan.
Present address: Sanyo Special Steel Co. Ltd., Hyogo 672-8677, Japan.
3
550

November 2006
Communications of the American Ceramic Society
3551
20
15
10
5
0
:
:
:
α-sialon
α/β-sialon
β-sialon
0
10
20
30
40
50
60
Time (h)
Fig. 1. Time evolution of the weight loss after corrosion in supercritical
water.
(
EDX) spectrometer. Figure 3 shows a SEM image and EDX
line analysis profiles of the fracture surface of b-sialon after cor-
rosion for 30 h. The thickness of the corroded layer increased
with corrosion time, showing a diminishing rate of growth. Sim-
ilar results were obtained for a- and a/b-sialon. Thus, the cor-
roded layer provided a certain degree of protection against
corrosion of the sialon substrate by SCW. EDX line analysis
showed that the concentration of Si and N in the corroded layer
was lowered, whereas O and Al were enriched. A similar trend
was observed for a- and a/b-sialon. Although XRD results
showed no evidence of any compounds other than sialon, on
the basis of the composition of sialon, the crystalline corrosion
2 2 3
products were identified as hydrates of the SiO –Al O system,
such as kaolinite (Al Si (OH) ). (The possibility of the exist-
2
O
2 5
4
ence of a small amount of amorphous compounds should not be
ruled out, although no harrow-like pattern was observed.)
Figure 4 shows SEM images of the surface of a-sialon (a),
a/b-sialon (b), and b-sialon (c) after removal of surface corrosion
products by ultrasonication in redistilled water. The corrosion
1
0
time was 30 h. As observed previously for Si N containing
4
3
Y
O
2 3
and Al O
2 3
as sintering aids, a number of corrosion pits
Fig. 3. Scanning electron microscopy image and energy dispersive X-
ray line analysis profiles of the fracture surface of b-sialon after corro-
sion for 30 h.
were clearly observed in the case of a/b- and b-sialon. The di-
ameter and number of the pits depended on the corrosion time.
Thus, the corrosion morphology of a/b- and b-sialon was char-
acterized by pitting corrosion. The unevenness of the interface
between the corroded layer and b-sialon substrate (see Fig. 3)
resulted from the pitting corrosion. By contrast, the surface of
the substrate of a-sialon was smooth, and no corrosion pits were
observed. This means that a-sialon itself corroded uniformly in
SCW. It is known that Si N corrodes in high-temperature high-
3
4
15,16
pressure water by the following reaction
:
Si3N4 þ 6H2O ! 3SiO2 þ 4NH3
(1)
(2)
SiO2 þ OHꢁ ! HSiOꢁ
3
Sialon is basically corroded by a similar reaction in SCW. In the
case of a/b- and b-sialon, corrosion pits are formed by leaching
of the grain boundary phase. Corrosion of sialon in the pits pre-
sumably increases the pH locally, resulting in the accelerative
dissolution of SiO by reaction (2). The weight loss of a/b- and b-
2
sialon was larger than that of a-sialon because of such local cor-
rosion.
In order to investigate the influence of the difference in cor-
rosion morphology on the degree of strength deterioration,
three-point bending tests were performed at room temperature.
Figure 5 shows the average flexural strength S after corrosion in
Fig. 2. Scanning electron microscopy image of the surface of b-sialon
after corrosion for 30 h.
SCW for 30 h. The ratio of S to the initial strength S was also
0
shown in the figure. The initial strength increased in the order

3
552
Communications of the American Ceramic Society
Vol. 89, No. 11
8
7
6
5
4
3
2
1
00
00
00
00
00
00
00
00
0
α-sialon
α/β-sialon
β-sialon
Fig. 5. Average flexural strength of a-, a/b-, and b-sialon after corro-
sion for 30 h.
using SEM showed that the origin of fracture was a corrosion
pit. These results indicate that corrosion pits generated by
corrosion of b-sialon are fatal defects that cause large strength
deterioration. The strength deterioration of a-sialon, which
corroded uniformly, was small because no new defects were
formed as corrosion proceeded. As the dimples due to pitting
corrosion were smaller for a/b-sialon than for b-sialon (see
Fig. 4), a/b-sialon showed intermediate strength deterioration
during corrosion, between a- and b-sialon. However, if the cor-
rosion pits grow as a result of extended corrosion, the strength
deterioration of a/b-sialon will increase. The initial strength of
a/b- and b-sialon, which consist of columnar crystals, is higher
than that of a-sialon, which consists of equiaxed crystals. How-
ever, the strength deterioration due to pitting corrosion and the
variation in the strength of a/b- and b-sialon make their life
prediction difficult. Thus, it is concluded that a-sialon, in which
pitting corrosion due to grain boundary phases does not occur,
is more advantageous in SCW environment.
IV. Conclusion
The corrosion behavior of sialon ceramics in supercritical water
(4501C, 45 MPa) was studied. The results can be summarized as
follows:
(
1) a-sialon showed lower weight loss compared with a/b-
and b-sialon.
(
consisting of crystalline SiO –Al O system compounds.
2) All three sialons were covered with a corroded layer
2
2
3
(
roded uniformly without generating new defects on the substrate
surface.
(4) In the case of a/b- and b-sialon, corrosion pits that
caused serious strength deterioration were formed by leaching of
3) a-sialon, which has no grain boundary phases, was cor-
Fig. 4. Scanning electron microscopy images of the surface of a-sialon
a), a/b-sialon (b), and b-sialon (c) after corrosion for 30 h. The corro-
sion products were removed by ultrasonication in redistilled water.
(
the grain boundary phase.
a-sialonob-sialonoa/b-sialon as shown in Table I, whereas the
strength of a-sialon after corrosion was higher than that of b-
sialon after corrosion. The strength deterioration of b-sialon
(5) On the basis of these results, it is concluded that a-sialon
is preferable to the other sialons as, in the case of a-sialon, su-
percritical water corrosion forms no corrosion pits, which act as
sources of fracture.
(
35%) was the largest among the three sialons. Fractography
Table I. Chemical Composition and Initial Strength of the Specimens Used in This Study
Specimen
a-sialon
12ꢁ4.5X
a/b-sialon
b-sialonw
Composition
Y (Si
X 5 0.5
617
, Al4.5X)(O1.5X, N
)
Si Al O N
Z
X
16ꢁ1.5X
6ꢁZ
Z 5 0.5
794
Z
8ꢁZ
X 5 0.2
902
z
Initial strength, S (MPa)
0
wDoped with 5 wt% Y
z
2 3
O . At room temperature.

November 2006
Communications of the American Ceramic Society
3553
9
T. Yoshio, T Oda, K. Suemasu, and A. Kohno, ‘‘Corrosion Behavior of
Mullite Ceramics in Water at 3001C,’’ J. Ceram. Soc. Jpn., 100, 668–74
(1992).
References
1
D. B. Mitton, J.-H. Yoon, and R. M. Latanision, ‘‘An Overview of Corrosion
Phenomena in SCWO Systems for Hazardous Waste Destruction,’’ Zairyo-
to-kanhyo, 49, 130–7 (2000).
T. Sako and T. Sugeta, ‘‘Decomposition of Environmentally Toxic Substances
with Supercritical Water Oxidation Process,’’ Materia Jpn., 39, 314–9 (2000).
N. Boukis, N. Claussen, K. Ebert, R. Anssen, and M. Schacht, ‘‘Corrosion
Screening Tests of High-Performance Ceramics in Supercritical Water Containing
Oxygen and Hydrochloric Acid,’’ J. Eur. Ceram. Soc., 17, 71–6 (1997).
E. Proverbio and F. Carassiti, ‘‘Low-Temperature Oxidation of Silicon Nitride
10
M. Nagae, S. Kuwae, T. Yoshio, and K. Oda, ‘‘Corrosion Behavior of
Structural Ceramics in Supercritical Water,’’ Corrosion Eng., 53, 601–9
(2004).
2
11
K. H. Jack, ‘‘Sialons: Ceramic Alloys for Engineering Applications’’;
pp. 465–81 in Ceramics: toward the 21st Century, Edited by N. Soga and A.
3
Kato. The Ceramic Society of Japan, 1991.
Japan Industrial Standard ‘‘Testing Method for Flexural Strength (Modulus
12
4
of Rupture) of Fine Ceramics’’; pp. 288–9 in JIS Handbook Ceramics-2005. Japan
Standards Association, Tokyo, 2005.
G. C. Kennedy, ‘‘A Portion of the System Silica-Water,’’ Econ. Geol., 45, 629–
by Water in Supercritical Condition,’’ J. Eur. Ceram. Soc., 16, 1121–6 (1996).
N. Hara and K. Sugimoto, ‘‘Corrosion of Ceramics in Supercritical Water
5
13
Environments,’’ Materia Jpn., 39, 325–30 (2000).
T. Yoshio, K. Oda, and K. Oda, ‘‘Effect of Grain Morphology in Pressureless-
53 (1950).
6
14
T. Yoshio and K. Oda, ‘‘Hydrothermal Corrosion and Strength Degradation
of Sialon Ceramics’’; p. 692 in Extended Abstracts, 184th Electrochemical Society,
Sintered Si N –Y O /Al O
3 4 2 3 2 3
on Leaching Behavior under Hydrothermal Condi-
tions,’’ J. Ceram. Soc. Jpn., 95, 435–41 (1987).
T. Yoshio and K. Oda, ‘‘Leaching Characteristics of MgO and Al
Vol. 93-2. The Electrochemical Society, New Orleans, LA, 1993.
M. Yoshimura, J. Kase, and S. Somiya, ‘‘Hydrothermal Oxidation of Si N
7
15
O
2 3
from
3
4
Glass-Matrix as a Simulated Grain Boundary Phase of Ceramics,’’ J. Ceram. Soc.
Jpn., 96, 159–64 (1988).
K. Oda and T. Yoshio, ‘‘Hydrothermal Corrosion of Alumina Ceramics,’’ J.
Powder,’’ J. Ceram. Soc. Jpn., 94, 129–34 (1986).
16
K. Oda, T. Yoshio, Y. Miyamoto, and M. Koizumi, ‘‘Hydrothermal Corro-
sion of Pure, Hot Isostatically Pressed Silicon Nitride,’’ J. Am. Ceram. Soc., 76,
1365–8 (1993).
8
Am. Ceram. Soc., 80, 3233–6 (1997).
&