ML20079G631
| ML20079G631 | |
| Person / Time | |
|---|---|
| Site: | Crane  |
| Issue date: | 08/30/1982 |
| From: | Macdonald D - No Known Affiliation |
| To: | Benaroya V Office of Nuclear Reactor Regulation |
| Shared Package | |
| ML20079G498 | List:
|
| References | |
| FOIA-83-243, FOIA-83-A-18 NUDOCS 8209280427 | |
| Download: ML20079G631 (4) | |
Text
. _ - - - _ _ - - - _
M
,.]
, o, y u r n'.*. R.
1149 Regency Dr.
3 Columbus, Ohio 43220 August 30, 1982 Mr. Victor Benaraya, Chief Chemical Er.gineering Branch Division of Engineering United States Nuclear Reglatory Commission c
Washington, D.C.
20555
Subject:
Review of 8 & W Document No. 77-1135317. " Evaluation of Tube Samples from TMI-1."
Dear Mr. Senaroya,
I have now had an opportunity to review the recent report from B & W referred to above. My review and comuments are given below.
In general, the report describes a thorough and comprehensive evaluation of inconel-600 tubes removed from the TMI-1 steam generate-s subsequent to detection of extensive primary-to-secondary leaks in September 1981. However, there are a number of feature *. of the report that desenre comment, particularly with regard to findings which may have implications for future activities by GPU and B & W in repairing the damaged steam generators. The more important of these are as follows:
Page 11 - Several references han been made to " greenish" and " greenish-yellow" l
deposits in areat which gave eddy current indications. However, no positive identification of this deposit has been made. Compounds l
having this characteristic color include copper oxyhydroxy chlorides, l
nickel hydroxide, and various sulfur compe :.,ds such as elemental sulfur, iron disulfide (FeS. pyrite), and possibly some nickel 2
sulfides, (as indicated in the 8 & W report).
Although electron diffraction studies were performed in an effort to identify some of the deposits observed, these efforts were only partly successful.
I believe that efforts should continue to identify the deposits, because positive identification may provide unequivocal evidence i
for any given failure mechanism.
Page 26 - The report states the 26-3/4" location did not.show any significant sulfur peak., and that other spectra, not recorded, also did not show l
j the presence of sul fur. However, i t is di f ficul t to sucr~' this i
- '. t re e. ? from the s;;cctre s5cwn ir ~i a.
'.r.'.'"is--
1si.Je oackground signal in tt'e region ae.e sultur is c.
acccar, to me that tulfur species could very well be prev:rt, in wtitch case there is nt.*hing unique about t'iese particul Ae e M es.
Page 35 - The authors have made,1xtensive use of the sodium azide test for sul fur species. However, i t should be e rh n i zed tha t, d '.*cu<1m l }
the nan 3 test detects " reduced" for% of sulfur, it is not a r,.
3 XA Copy Has seen sent to pog
-6md
_ [
(
d; I
(&
i
-t-speci fic as to the form of the sul fur-containing species.
Frorn a l
chemistry viewpoint, the most important characteristic of the
" reduced" fortas of sul fur is that they contain elemental sul f ar, which can act as a powerful oxidant, particularly when deposited at the tip of a stress corrosion crack.
In s:y opinion, a better 7
trethod would have been to extract the zero-valent sulfur from the
^}
surface using cyclohexane followed by tJV absorption spectrometry f
to deter 1mine the concentration of sulfur.
I believe that this technique, which can also be applied to bulk water samples, would provide more reliable measures of the local concentrations ci aggressive sul fur.
11 Section 2.8.2 - The Auger depth proft".x data contained within this section are difficult to understand on a quantitative basis, since the data indicate that the films (and possibly the alloy atter For example, in sufficient sputtering) contain up to 101 carbon. I of sputtering
~1 Fig. 2.8.10 the oxygen signal is lost after sl400
+
i l
indicating that the surface oxide film is completely removed at l
this point. However, the carbon content remains constant from 400I L
l onwards at about 91.
Esen taking into account hydrocarbon contastn-l ation, and the fact that Ar+ spottering could result in " scrambling" l
of the underlying atomic layers, this figure appears to be far too high when,:oepared tC, the hulE alley carbes concentration of 0.0411.
The alogorith used for canapsting elemental composition was not described la detail, and beece it is not possible to comment on the viability of the particelar method used. The data do, however, demonstrate the very serious problems that can arise when using Auger Spectroscopy for questitative or seal-quantitative surface f
analysis.
Similar er===*nts may be levelled against the ESCA (or IPS) data, i.e., it is frequently difficult to give an unambiguous interpreta-
!s tion of the data without independent chemical evidence. Thus, on page 55 it is stated that the hydrocarbon is doubly and singly N
bonded to oxygen, presumeably as katone (R C=0), aldehyde (RCHO),
2 or acid (RCOOH) groups and alcohol (ROH) functions, respectively.
The authors also state that carbon-nitrogen bonds were detected;
,.j presumeably these indicate the existence of primary (RNH ).
1 2
secondary (RR'NH), or teniary (RR'R*N) amines or asides (including 4
s ubs t i tuted asi des o f the type RCOMR' R" ).
It was stated at a meeting of the Task Group at GPU that the primary circuit may have been contaminated with ar. oil of some type.
I f so, the oil may J
orovide an explanation for the ESCA observation, but this bas not j
heen demonstrated by the authors.
It is also clatw o
- bi ar*a, and **&ated aise#*r= in the ra ", th"
+
'"e sul fur seek are assoc'i ed with
~' --
t e vidence that a sul fur (+?) species of this type exists
.rA'
+n-condi tions which are relevant to IMI-1 g
c % - The SIs resul ts which shc., tr it si gni ficant arnounts o' 5v:rogen tre l
present in the alloy, part cu' irl y in t*e -egion o f t>e c rick
p, ar" very i n t e r e ". t i n g.
The a.
t,;
ha v e. men very Carr M rat ta i
- .~o
- 131, b
- 'a a
d i
'>pgp s n g e, in i
on' '1( !ce.
i * *.) C h ICL1 Mo(b Y
ii 'y,e "h,fr.3 gen pct. i e
- I pu,p i r o,,.., i r,
- g in, 1,,*
- L.,
e
- b+'
+
Ho.,ever, l'
"n her-
- e e
r i,
.9 2'
.p 3
o
[^,
that partially reduced sulfur species act to poison the hydrogen recombination reaction which occurs at the metal surface during free corrosion. therby accelerating. penetration of hydrogen into the bulk alloy phase.
It is emphasized that this is an area of great controversy in corrosion science (SCC vs HE) so that it would be premature to attach too much tignificance to this single observation.
p D
Soth the ESCA (page 55) and the electron diffraction (page 57) studies 7
on the metal surface is addition to varfous oxides,s"?) are present indicate that metal sulfides (e.g. Fe$ and *FeeNisSy 4.g.
Rio. Fe3 4.
0
- 1,g04*). It:should be recalled that "Fe0.7#IO 00N*. "F82 4 thermodyn.3amic e y useAts asented earlier fadicated that the prodvet 0
thiosulfate) reduced forms of sulfur" (e.g.for other iron l
of the reactica between l
and Rf-cr*Ps alleys. such as' Incenel sould be Fe5 sulfides) and spestbly micte11ssiffdes letter has not base a
mixed'selfles Fe sta$gs".
1 m
detected al.
er.ldelnce31s; citsi
,this censultant that these-e It uns al tapa~agrfite, exposure
~
iteld 1
(la ageoeus solution upon active sul y,
i4%t metal smitters have F
bees Aerfbees:
est M r d ignificant ties
- 4
'2h' C
- cf hete1-snifIde
- entside of en s
~
- amelyzed's a
90rl4R
- y. ears of. the
.t estdetton'
% y tan to the fers f
~
T f,. f d "
~ gn.n 'J L..
- 1, 9u!M@.
9 nn.
k w eh,k es.s, f,
j w
)
s
- s we w i thereby explafetag i
C1
.tispTy.'.the ftde may estdita ethe.ERAc 1.shryoti c?stids (Na ohich-e
&Tsai tse le
,the r
.,%sensensati
~
eMM :-
cad.ecetothe.
3 c,T h thes k 1secracen If se. the ted by,ths'
- va, the:
3
' a estdattoa
~
14 1 4Reensideredtobeveryreal.
a
- &ger of~
' ~4se lhe extensive *agb11'egraph[je[{spjte'e g m.v a
nst
~
l staa
(
l a'sonsitt2ed alcrestr,i Iow b.locoisel-600.1 particularly for that of dl' reS e
caecentratteifsferit the vfatage' esed,1n.W4. Tht' fact;that facenel-690 is much more j
~
susceptible to ~carbids-fadoced sensitizatfealthat is Type 30455 for i
l l
egstvalent carben.cencentratienti cten'be attributed to the much lower
);
a' carbon salubility in the former compared with the latter at the same I
I temperature. Indeed. as a point for future consideration, sansiti-zation-immune Tow' carbon gr.edes of Inconel-600 m not :
r Cic3' t
(j l
In the sense that the low carbon grades of austenitic stainless l
l steels are ( e.g., Type 304L).
Page 77 - Some effort was invested in this study to determine whether or not
/
i slow cract growth would occur in the tubes under static, axial, tensile loading.
The environments to which the inside surfaces of j)
I the tube were exposed included laboratory air, moist laboritory air,
(
"l M primary coolant. Apparently no attempt was made to Starmine d
,-4..
whether or not crock extension wecid occur if tha inside surfaces were exposed te thiosulfate (or polythfenic acid)-contaminated environments, prepared either Js1 t~ the addition of the contaminant to the costaat, or sne h the oxidation of metal sulffdes ca the metai serface.
ettal sulfides cow 1d easily D.
have been prepared by prior esposure of the tube to thiosulfate J;l i solutions la tkt absence of saygge, T talteve that this mould have
[f- (
1 been a cruclel esperleemt, siece it any, have ladicated whether or not the fears of subseessat caching la' MR primary systans which
. f; have not hees d,es,s.1forland are mell-Segefed.
i
.a-z.: m..
...,?
i ht this potat. I wish to rettetete that the t 4 5 study represeets a thorough r
4ad comprehenstwe analysis.ef. the fal1M115-1 SS tukee. partIcalarly tn vlew r
l of the very severe time fats. '
ch '
4 ethers had to operate.
Momever, I bellese.that ~
1;-
f
~
aslpela of thleselfate and matter of the 1RI failures has not been addressed:'.
ts, I
will the falleres costl
- occg> 17:-
ekevits are put tect late
[
l operetton withest
~
idessifbriaa '
fefog tenen, I as sors qubs t
M.. %e....a n.tIoss; W.
'"*
- J'" "8 97;< ~
,n eu e-M M III*
h IIM
$E
..:eGy_.g og, lW_.41EcMT,u.',
r
}k gb i-qulW. fr.Ne
.g com:ab
- fig '
% ; *v: %yqq '2% %:.
ee--
'egA% ; -
> 1, 1
Y c,
w
+ +
m;
)
, 7. y -
.c
. h=?.
+
^
r a-"
[~
. 1.)
\\*
h N
se s
f l
l; n o us.w m m m MJ v3L
(_m,,
v 4
O!
ON File RVollmer PCheck l
DEisenhut l
OCT 0 5 W JRoe HThompson Stinogue BSnyder DeYoung 5Hanauer Cunningham ED0 U320 ECase VBeneroya 19enten CMcCracken PPAS K. Johnson L. Miller Ms. Jane Me M. Bridgers EDO-12320 183 Valley Road E. Abbott, OCM Etters, Psensylvania U35 Q &f Deer Ms. Lae:
Your letter of Septeeber.1,1982. to Cassissioner Giliasty has been i
referred ta' es.for
. You raised ffye questions pertaining to the corresten. problems the Mt-1.staan rs. All of the concerns Mised tem are beita' eued by the staff as part of aer.
.eggloati dr..td e'spro~5a1" f restart for TRI-L e a(AmeiTiIIshdiidK 40.1heceggjetstdur.ingJosuary,1983.
j r
is eer f=11.'maineth o7 the corrosien l
(
preh1W this_.tige'fylly respetsd to each of
~
~~
year tejent,grellet
~dsta are avail ele l
la r of your concerns.
m g.m q f, y g t 4,r; :ty,plation to" 88 e astsees w ar w
?3 l
f;,;
wp-nt;- :
.y The que'stiensrelatetooncethrough
.c l
stees n
ftfzetfos and GPtR's determination thek; tta:eest is caus&tive apont. We i
have i#
1 and believe tut a
~
setes that sodium thiosulfate is cerreston. This determinatica is bened'est; t$1
' a' for the sodium thiosulfate to enter the~
end the presence of sodium thlessifate'7s at sufficient concentrations to have caused the cerrest. Additionally, laboratory tests have I
densestrated that the spacentration of su1Nr as sodium thiosulfate, found in the reactar coolent, was sufficient to have caused the I
corrosion. None of the other potential sources for sulfur would appear to have provided sufficient concentrations of sulfur or the necessary reduced sulfur chemical forms to cause the observed corrosion.
The word " sensitization", which you questioned, is a metallurgical tem used to describe the microstructural characteristics of the OTSG tube material.
S w ifically, for the OTSG tubes which are manufactured from INCOEL-600, sensitized microstructure refers to the preferential cart >ide precipitation along grain boundaries and the presence of lower chroalus content region, with respect to the bulk chromium content, adjacent to the grain boundaries in the tubing alloy.
If ~
}
. l...
. }.
.g.
OFFICIAL. RECORD COPY
.....om.
. R 9
P Ms. Jane Lee 2
During the manufacturing and fabrication process, thermal mechanical effects can lead to the precipitation of chromium carbides and the redistribution of chromium atoms along and adjacent to the grain boundaries which, in turn, change the mechanicai strength and the By contro11uw the heat treatment j
corrosion resistance of the teing.
conditions, it is possible to control the shape, amount, and distribution
.g of carbides along grain boundaries, and the degree of chreatum depletion -
EQ in surrounding areas.
In this manner, it is pessible to improve resistance Ly -
Conversely, a noterial structure which to specific ty ws of correstem.
b is controlled y heat treatment te leprove resistance to some specific types of corresten, can be sere susceptible to other types of corrosion.
When selecting a estarial for a specific application, the grain and grain boundary structure, cartida se14 slag and distributten regionchromiumcontentarecentrolledtoprovIderesistancetothe a.
types of cerrosten dich are. typically enticipated under the intended service conditions. The sensitized 1 41am used at THI-1 was selected to provide good strength and other eschaefcal properties, and resis-
.g tance to the twical Yeres of corresten that would be anticipated in g.;;
a nuclear steen generstar. 'The sedles thiessifate dich was introduced W
into the reacter coolant is met a twical corrodent in nuclear steas
@M generaurs. Therefer3, the:t elag eld met have a high degree of A
resistence to corresten by a redeced sulfur species.
The
- repairs for TIE-1 GT5G's do not include any changes in n.
treatment er grafa beendary structure.
Instead, the focus teing is on removing the corrodest and esserleG that it cannot be reintro-duced to the reacter coolant system.
The chloride dich was found during enseinetton of the tubing is
[
s normally present in trece geantities in unter and is typical y found during any corrosien examination.
TM chloride reported in the TM1-1 F
steam generator states report, in our opinion, was not a causative egent
~
in the cerrosion process.
Qusstion No. 2 This question r6lates to the presence and potential effect of sodium
'y thiosulfate on other system asterials (reacter internals).
A thorough j
examination of the reactor internals was conducted including removal
_t and destructive examination of some components. These exasinations were r-witnessed by MRC personnel and our consultants.
No evidence of sulfur-r:
induced corrosion wer found.
i Sulfur has been detected in the protective oxide layer on most system surfaces.
GPtJM is conducting an extensive program to determine if the j
concentrations of sulfur which have been found can cause furthe" h'
corrosion.
If a significant potential for future corrosion exists, the sulfur will be removed prior to restart.
The staff and our con-J sultants are following these progress closely.
j
~~
G NE
'OFFICikL R ECOED COPY RD-e--2TOPMP VE CF
{p [
.. { G 3: ~ ^ ll&?k&
Ms. Jane Lee 3
The most probable explanation of why corrosion was found on the steam generator tuces and not other system materials is that the tubes were under tension.
The degree of tension in the steam generator tubes was sufficient to fracture their protective oxide film and permit the sulfur to attack the base metal.
This concern will be addressed in dotati in our safety evaluation.
Question No. 4 The concern you expressed in the question relates to the potential for a "Ginna Type" tube ryture due to tube pullout potential in the tubesheet The 50*F temperature difference between and pressurized thereav shock.
the tube sheet and teing within the tubesheet is very conservative.
Under all normal plant operations this difference is only a few degrees.
For the worst case postulated accident (main steam line bicak), a maximum of approximately 30"F can be predicted.
Therefore, testing at a 50*F temperature differential is conservative. Additionally, GPUN will not expand the bottoe two inches of tubing within the tubesheet which provides an additional physical tubing restraint in the event of further corrosion.
Pressurized thermai shock is being reviewed by the staff as Generic Issus A-49. We are currently scheduled to havs A-49 resolved by late 1983. Preliminary information indicates that the potential for significant thermal shock due to a steam generator tube rupture is small.
I Questice No. 5 This questions the suitability and reliability of the kinetic expansion ir process.
An extensive amount of testing is being conducted by l
to qualify the proposed repair process for returning the steam generators to service.
In addition to monitoring this testing, the NRC is having independent testing of the repair process conducted by our contractors.
Preliminary inforestion indicates that kinetic expansion is the best repair method. Accelerated life cycle testing of sockups in the laboratory has shown no problaas in meeting technical spect-fications for steam generator t se leakage for up to a five year period.
These tests are continuing and will provide life cycle data for a period of 35 years.
We appreciate your interest and wish to assure you that we are l
thorougnly evaluating the corrosion problem which has occurred at TMI-1.
We will require whatever steps are necessary to protect the health and safety of the public.
5 0
DE:CMEB m Harold R. Denton, Director CMcCfa : ken: ao VBenarofa Office of Nuclear Reactor Regulation 9/ H /82 9/ p /,82 a37 g M u g)