ML20244E128
Text
_ _ _ _ _
r i
Y
~
. q p 9, 6 n,
ng pc T' c-. -+
n d
Corrosion Evaluation - Nine Mile Point Reactor Vessel
'1 i
d e
~=1 by R,,
0.X:'
to g@rg Craig F. Cheng g
'j Argonne National Laboratory March 24, 1969 l
/
% d5 The mechanisms of in urgranular stress corrosion and corrosion fatigue of
- chwy k
sensitizedTyy.304stainlesssteelinboilingwaterreactor(BWR)environmentare(
J/
not clearly understood. Hopefully the " Reactor Primary Coolant System Rupture Study - Task K (PA37-Contract AT(04-3)-189" at Dresden I will contribute to I
the understanding of these mechanisms. However, after reviewing the 6th Sup-f pienent to Final Safety Analysis Report - Docket 50-220 and other literatures, I believe the seuitized components (e.g. stub tubes) in Nine Mile Point 3
Reactor Vessel can be operated with reasonable assurance for the next few 1
{
years under normal reactor operating conditions, provided an in-service sur-
[
l ve111ance program with static loaded specimens be augmented.
z 1
The recommended locations for these specimens are!
steam space in the vessel head to evaluate oxygen and solid carry a.
over effects, b.
recirculating area (where the return feed water mixes with the coolant) to evaluate the flowing water environment, and t
4 recirculation supply line adjacent to the reactor vessel to evaluate c.
i 1
the return water environment.
These specimens should be similar in design and geometry as those static and cyclic loaded specimens to be used in Dresden I. so that the proper correla-tion and comparison between reactor environments can be evaluated in terms of fracture me anics (stress intensity f actor, fatigue crack growth, flaw,.
p
(? ) ~ ' R depth, etc.)..y f
,f N d i.C.
[
U g, a
e OA san
.. _ ~
.3
~,
~j In addition, I do not agree with the applicant's conclusions in two l
areas:
l, j a.
reported failure of sensitized Type 304 steel components occurred l
only with acid pickled (HNO -HF) surface.
3 4
l' '{-
b.
Dresden I experience is applicable to all BWR.
i There are reported cases of intergranular corrosion cracking in reactor i
l 1 vater service with Type 304 stainless steel components whose surfac.s have not been previously acid pickled (HNO -HF).
These include:
3 1.
thin walled (0.008 to 0.028 inch) fuel cladding in solution annealed condition,(l' )
i preheater pipes in solution annealed condition,( } and J
2.
pres cre vessel liners in sensitized condition.(4,5)
(
3.
(
The last case refers to EBWR vessel cladding.
It was concluded the initial F
j cracking within the cladding was probably caused by thermally induced stress i
(
rupture during vessel fabrication. Those surface cracks propagated by low-
?
cycle corrosion fatigue during reactor operation. Remember, there are two 1
j consecutive steps in the mechanism of stress corrosion cracking. The first l,
step is initiation, which can be triggered by chemical, electrochemical and/or l
~
mechanical means. The second step is propagation under press and the propaga-f tion continues if the crack tip is not blunted or arrested. In GE's 100 ppm oxygenated water test, the marked difference in time to failure between as 1
sensitized and as sensitized, plus scid pickled (HNO -HF) stainless steel may 3
[
easily be due to the difference in the incubation period during the initiation step.
I Furthermore, BNWL(6) recently reported stress corrosion cracking of sen-sitized 304 SS piping, which had been previously mill pickled (HNO -HF), in 3
_____.__.i.._.___._.___
m 1
j
' )
3-
]
sulfite treated watet at 164 C.
Soditsn sulfite (100 ppm) as added to the demineralized water ( ( 1 ppm chloride) to remove residual oxygen. The oxygen content in the water analyzed 0.2 ppm and the pH of the water ranged 6
I 6.5 to 7.0.
Failures were observed in all three phases in descending e
i order of susceptibility, liquid, liquid-vapor interface, and vapor. Thus i
there are other unknown contaminants in addition to oxygen and chloride that promote cracking. The reactor environments in Nine Mile Point may not be identical to Dresden I, particularly as to the utsaber and duration of off-chemistry periods as well as unknown contaminants. In which case, satisfactory performance of sensitized specimens with respect to stress corrosion and corrosion fatigue in Dresden I does not guarantee the same rescit in Nine Mile Poi' t..
n 8
(
i 1
I f
(1)
W. H. Arlt and S. R. Vanderberg, Gen. Elec. Co. Rpt. CEAP-4360 (1963).
(2)
C. N. Spalaris, Nucleonics, 21, 41 (1963).
(3) Combustion Engineering Co., Report CEND 265 (1965).
t (4)
N. Balai et al., Report ANL-7117 (1965).
,I (5)
N. Balai et al., Report ANL-7117 Supplement (Septembe'r '1968').
(6)
A. P. Larrick, BNWL-753 (April 1968).
1
- e e.-ees e e ser
,p me p.ee a
_.. _ _ _. _ - _ _ _ _ _ _. _ _ _ _ _ _ _ _ _ _ _. _ _ _ _ _ _ _ _ - - - - _ - - - - _