ML20199K027

Feasibility of Underwater Welding of Highly Irradiated IN-VESSEL Components of Boiling Water Reactors.A Literature Review

ML20199K027
Person / Time
Issue date:	11/30/1997
From:	Lund A NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES)
To:
References
NUREG-1616, NUDOCS 9711280297
Download: ML20199K027 (46)

Text

- _-_ _

l 1

i NUREG-1616 Feasibility of Enderwater Welding of Highly Irradiated In-Vessel Components of Boiling-Water Reactors A Literature Review U. S. Nuclear Regulatory Commission Office of Nuclear llegulatory ltescareli A. L Lund y l; h ,

3gmfa.

f2 97 128 971130 l!I.lilliI!lllBIjlll,llillilli 1616 C PDR

AVAILABILITY NOTICE Availabibty of Referenco Materials Cited in NRC Publications Most documents cited in NRC public:itions will be available from one of the following sources:

1. The NRC Pubhc Document Room, 2120 L Stroot, NW., Lower Loyol, Washington, DC 20555-0001

2. 1ho Suponntendent of Documents, U.S. Governmont Printing Offico, P. O. Box 37082, Washington, DC 204b2-9328

3. Tho National Technical information Sorvice, Springfield, VA 22101-0002 Although the listing that follows represents the majority of documents cited in NRC publica-tions, it is not intended to be exhaustivo.

Br3fotonced docun.onts availablo for inspection end copying for a 106 from the NRC Public Documont Room iwiudo NRC correspondenco and internal NRC memoranda; NRC bulletins, circulars, inforrnation noticos, inspection und investigation notices; liconsco ovont reports; vendor roports and correspondence; Commission popors; and applicant and licensoo docu-ments and correspondenco.

T he following documents in iho NUREO serios are availablo for purchose from the Government Printing Offico: formal NRC staff and contractor reports, NRC sponsored conference pro-coodings, international agroomont reports, granteo reports, and NRC booklets and bro-churos. Also availablo are regulatory guidos, NRC regulations in the Code of Fodotal Aopula-tions, and Nuclear Regulatory Commission Issuancos.

Documents available from the National Technical Information Servico includo NUREG-series reports and technical reports prepared by other Fodoral agencios and reports propared by thu Atomic Enorgy Commitsion, fororunner agency to the Nuclear Regulatory Commission.

Documents availablo from pubhc and special technical hbrarios includo all opon literature items, such as books, journal articios, and transactions. Fodotal Register noticos Federal

- and Stato legislation, and congressional reports can usually be obtainod from those libraries.

Documerus such as thoses, dissortations, t.iroign reports and translations, and non-NRC con-forenco proccodin0s are availablo for purchaso from the organization sponsoring the publica-tion cited, l

Singio copios of NRC draft reports are availabio froo, to the extent of sup)ly, upon written request to tha Offico of Administration, Distnbution and Mall Services Secti >n, U.S. Nuclear Roguinic y comrrission, Washington DC 20555-0001.

Copies of industry codos and standards used in a substantive manner in the NRC regulatory process ato maintained at the NRC Library Two Whito Flint North 11545 Rockville Pike, Rock- j villo, MD 20852-2738. for use by the pubhe. Codos and standards aro usually copyrighted and may 60 purchased from the originating organization or, if they are Amoncan National Standards, from the American National Standards Instituto,1430 Broadway, New York, NY 10018-3308.

NUREG-1616 Feasibility of Underwater Welding of Highly Iin adiated In-Vessel Components of Boiling-Water Reactors A Literature. Review Manum:ript Completed 11ovemter 1997 Date Published: Novemter 1997 A. L 1.und Division of Engineering Technology Omce of Nuclear Regulatory Research U.S. Nuclear Regulatory Commission Washington, DC 20555-0001 p-'%

W.....),,

l l

ABSTRACT in Febmary 1997, the U. S. Nuclear Regulatory Commission (NRC), Office of Nuclear Regulatory Research (RES), initiated a literature review to assess the state of und:rwater welding technology.

In particular, the objective of this literature review was to evaluate the viability of undenvater welding in vessel components of boiling water reactor (BWR) in vessel components, especially those components fabricated from stainless steels that are subjected to high neutron fluences. This assessment was requested because of the recent increased level of activity in the commercial nuclear industry to address generic issues conceming the reactor vessel and intemals, especially those issues -

related to repair options. This literature review revealed a preponderance of general information about underwater welding technology, as a result of the active research in tl 1 field sponsored by the U. S. Navy and ofTshore oil and gas industry concems, llowever, the literature search yielded only a limited amount ofinformation about underwater welding of components in low fluence areas of BWR in vessel environments, and no information at all conceming underwater welding i

experiences in high fluence environments.

Research reponed by the staff of the U. S. Department of Energy (DOE) Savannah River Site and researchers from the DOE fusion reactor program proved more fruitful. This research documented relevant experience conceming velding of stainless steel materials in air environments exposed to high neutron fluences. It also addressed problems with welding highly irradiated materials, and primarily attributed those problems to helium induced cracking in the material. (llelium is produced from the neutron irra intion of boron, an impurity, and nickel.) The researchers found that the amount of helium induced cracking could be controlled, or even eliminated, by reducing the heat input into the weld and applying a compressive stress perpendicular to the weld path, ill NUREG 1616 I

CONTENTS .

ABSTRACT........................................................................................................................... iii LISTOFTABLES.................................................................................................................. vii EX E C U Tl V E S U M M A R Y . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . ... . . ... . ... . . . .. . . .. . . . . . ... .ix . . . . . .. . . . .

xiii A B B R EV I ATl O N S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

w A C KN O W L E D G M E N T S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1. I N T R O D U C TI O N . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1. . . . . . . . . . . . . . . . . . . .

2. PROBLEMS ASSOCIATED WITil WELDING 1RRADIATED MATERIALS IN I IIG 1 i FLU ENC E EN VI RON M ENTS .................. .................................................... 4

3. WELDING CODES USED FOR UNDERWATER WELD REPAIR ....................... 6

4. UNDERWATER WEl. DING: GENERAL DISCUSSION .................. .................... 7

5. IN-VESSEL WELD REPAIRS AT U.S. COMMERCIAL NUCLEAR POWER PLANTS.................................................................................................................. 9 5.1 In-Vessel Weld Repair at Susquehanna ..... ....................... ........... ..... ........ 10 5 .1.1 F i rst R e pai r . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 5.1.2 Second Repai r ...... .... ..... ...... .. ... . .... ... .. .. . . . ..... ...................... ... I1 5 .1 . 3 Thi rd R e pai r . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I2 5 .1. 4 Fo u n h R e pa i r . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 .......

5 .1 . 5 F i llh R epai r . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 .......

5.1.6 Summary of Susquehanna Weld Repairs .. .. ......... . . . ......... ...... ......... I3 5.2 in Vessel Weld Repair at Peach Bottom Atomic Power Station Unit 3... ..... .. 13 5.3 In Vessel Weld Repair at River Bend Station ......... ........ ..... ... . ................... 13

6. WELD REPAIRS ON 111G11 FLUENCE MATERIALS AT DOE SAVANNAll RIVERSITE..................................................................................................... I5

7. FUSION PROGRAM RESEARCil ON WELDING OF lilGilLY I R RA D I AT E D M AT E RI AL S . . . . .. . . . . . . .. . . .. . . . . .. . . . . . . .. . . . . . . . . . . .. . .. . . . . . . . . . . . . . . . . . . . . . . I8 v NUREG 1616 1

l)lS. CUSS 1ON............................................................................................................. ,4 8.

9. CONCLUSIONS.......................................................................................................... 27

- ~ ,o I O. Rr.r.

Ls L:KL:NLL:a3.......................................................................................................... o

(

.J NUREG 1616 vi

E . . . . . _ . _ _ _ _

i LIST OF TABLES I

1. In Vessel Weld Repairs in U.S. Commercial Nuclear Power Plants ............................. 9

2. Weld Repairs and Research at the DOE Savannah River Sit: ...................................... I6

3. International Fusion Program Research on Welding ofliighly irradiated Materials ..... 19

4. Composition of Materials Used in the Fusion Research Program .................................. 21

5. Summary of Welding Techniques Referenced in this Report ......................................... 25 vii NUREG-1616 l

EXECUTIVE

SUMMARY

in Febmary 1997, the U. S. Nuclear Regulatory Commission (NRC), Office of Nuclear Regulatory Research (RES), initiated a literature review to assess the state of underwater welding technology.

In particular, the objective of this literature review was to evaluate the viability of underwater welding in vessel components of boiling water reactor (IlWR) in vessel components, especially those components fabricated from stainless steels that are subjected to high neutron Guences. This assessment was requested because of the recent increased level of activity in the commercial nuclear industry to address generic issues concerning the reactor vessel and internals, especially those issues related to repair options.

Of pmticular concern to the commercial nuclear industry is the incidence of environmentally assisted cracking found during examinations of the llWR core internals fabricated from stainless steel and high nicke! alloys. In the near future, the industry will have to make decisions about repairing or replacing components for wh;ch continued structural integrity cannot be ensured. Nonetheless, licensees and vendors have found that development of repair methodologies for components in the high fluence areas of the core is not an easy task.

Development of welding technology for use in repairing in vessel components subjected to high neutron fluences has been impeded by numercus instances in which helium induced cracking resulted from welding highly irradiated materials. This situation is further complicated by the general lack of consensus in the available literature with regard to the absolute minimum level of helium that induces cracking during welding operations. This confusion results from the complexity and synergistic effects of the variables involved. In general, the most limiting helium values reported in the literature are in the range of 1 to 2 atomic parts per million (appm) helium. Although techniques exist for measuring the amount of helium present in a material in a laboratory environment,it is far more complex and expensive to determine (or predict) the amount of helium present in in vessel materials that have not been removed from senice Consequently, sesearchers, nuclear engineers, and others who routinely use this information generally predict the amount of helium in reactor in vessel materials using flux mapping techniques, retrospective dosimetry techniques, or similar calculated approaches, rather than directly measuring the content.

In developing welding technology for use in repairing in vessel components, researchers, licensees, and vendors must address many complicated issues. For example,in vessel comporients typically have limited access (less than 360*) for inspection or repair, and high radiation levels which typically reluire a remote repair approach or a well shielded work environment. Other common issues include the structural integrity of the in vessel components, material damage attributable the high-fluence environment, and the deleterious effects of helium present in materials in the high fluence environments, to name a few, liccause the industry has not yet entirely resolved these problems, the literature included reports ix NUREG-1616

regarding the use of underwater welding in commercial nuclear plant repairs only for components that are not exposed to any neutron fields, anu those exposed only to low neutron fluences.

Nonetheless this literature review revealed ample interest in developing successful welding techniques for use in the challenging high-fluence in vessel environment of the BWR. In addition, this literature review indicated active research intended to resolve the problems associated with welding materials exposed to high neutron fluences, as well as an extensive core of knowledge, substantiated by > ears of research, regarding undenvater welding technology, Some of this research was conducted during years of funding fmm the oil and gas industry to develop techniques for repairing underwater oil platforms. The purpose of this research was to produce better welds at greater underwater depths, predominantly for components fabricated from carbon steels and stainless steels.

By contiast, this literature review did not reveal any reports of undenvater welding of materials exposed to high neutron Ibences. Moreover, little research has been performed in the area of undenvater welding la commercial reactor in vessel cnvironments. In fact, most of the applicable literatare regarding cemmercial nucl~ar plants conveys anecdotal experience involving emergent outage repairs in low fluence areas inside the reactor vessel (as opposed to carefully controlled experimental work). Nonetheless, the published accounts of underwater weld repairs show great promise for this technique. Successful underwater weld repairs have been trade on austenitic stuinless steel components with high quality welds that have passed inspection during subsequent outages.

Because of the scarcity of data pertaining to welding highly irradiated materials in an underwater environment, Ri!S expanded the literature search to include relevant data on dry welding of highly irradiated stainless steel components from the international Fusion Reactor Program and Savannah River Site (SRS), U. S. Depanment oflinergy (dol!). This expanded scope revealed an efTort to understand the fundamental mechanisms and propose solutions to the probbms associated with helium entrapment. To achieve that objective, a significant portion of the fusion program research has been performed using tritium doped material, thereby simulating helium entrapment in the absence of the damage normally associated with irradiated materials. In addition, fusion program research into welding stainless steels exposed to high lluence levels has shown some promise as a result of altering the heat input into the weld and applying a compressive stress du ing the welding operation. Current research funded by the Doli Fusion Program is focused on retining the techniques that produce acceptable welds in the desired ranges of Iluence. Although all of this research has been perfonned with dry welds,it proves relevant as a comprehensive examination of possible techniques that can be applied to welding highly irradiated materials.

To a considerable degree, the fusion program research built upon the earlier research and field repair- '

l experiences at the SRS. Weld repairs were perfonned dry on one stainless steel production reactor tank that experienced stress corrosion cracking during service. In 1968, SRS welding staff welded stainless steel patches to the tank walls with gas tungsten are weld (GTAW) techniques to successfully repair the knuckle region. When the welding staff tried the same weld repair approach on ditrerent cracks in the same reactor tank in 1984, the welds immediately failed through what the NURl!G 1616 x

stafflater determined was helium induced cracking. SRS staff subsequently considered at least 30 different weld processes to mitigate the effects of the helium induced cracking during welding operations, but efforts were concentrated on the gas metal arc (GMA) overlay weld technique. This technique used very low heat inputs and showed the greatest initial promise. In 1996, the tritium production scactors were permanently removed from service, and DOE discontinued the research and development funding for development of welding techniques. Therefore, the SRS welding staff had very little opportunity to use the techniques in the field on highly irradiated components.

In summary, the promising research and actual plant experiences, that have been captured by the literature review, may enable the c techniques to be applied in commercial nuclear power plants in the future.

xi NUREG 1616

I i

l ABBREVIATIONS AN51 American National Sta,dards Institute ASME American Society of hicchanical Engineers ASTM American Society for Testing and Materials AWS American Welding Society ll&PV lloiler and Pressure Vessel (ASME Code)

IlWR boiling water reactor IlWRVIP Iloiling Water Reactor Vessel and Internals Project CW cold work DOE U. S. Depanment of Energy EPRI Electric Power Research Institute ICAW flux core are welding GE General Electric Company GMA gas metal arc GTAW gas tungsten arc welding ilAZ heat affected zone lilR iligh I lux Reactor IGSCC intergranular stress corrosion cracking MIT Massachusetts Institute of Technology NDEC Non Destructive Examination Center NRC U. S. Nuclear Regulatory Commission NRR Nuc . tr heactor Regulation Of6ce of(NRC)

ORR Oak Ridge Research Reactor PCA primary candidate alloy PECO Philadelphia Electric Company PP&L Pennsylvania Power & Light RES Nuclear Regulatory Research, Office of(NRC)

SA solution annealed SCC stress corrosion cracking SMAW shielded metal are welding SRS Savannah River Site TIG tungsten inert gas TWI The Welding Institute xiii NUREG 1616

I ACKNOWLEDGMENTS The author acknowledges the asristance provided by hir. C.E. Carpenter, the NRC Project hianager for the Iloiting Water Reactor Vessel and Internals Project (DWRVIP), who coordinated requests for information from the IlWRVIP and the Electric Power Research Institute (EPRI), and performed a pect review of the report. The author also appreciates the assistance of hir. Robin Dyle, fonnerly of the Scuthern Nuclear Operating Company, and hir. Ken Wolfe of EPRI, who participated in technical discussions about the content of this report.

i xv NUREG-1616

0

l.0 INTitODUCTION in 1 chruary 1997, the U. S. Nuclear Regulatory Commission (NRL), Office of Nuclear Regulatory Research (Riis), initiated a hterature review to assess the state of underwater welding technology, in particular, the objective of this literature review was to evaluate the viability of underwater welding in vessel components of boiling water reactor (IlWR) in vessel components, especially those components fabricated from stainless steels that are subjected to high neutron fluences. This assessment was requested because of the recent increased level of activity in the commercial nuclear industry to address generic issues concerning the reactor vessel and internals, especially those issues related to repair options.

Of particular concem to the commercial nuclear industry is the incidence of environmentally assisted cracking found during examinations of the llWR core internals fabricated from stainless steel and high nickel alloys. In the near future, the industry will have to make decisions about repairing or replacing components for which continued structural integrity cannot be ensured. Nonetheless, licensees and vendors have found that development of repair methodologies fbr components in the high fluence areas of the core is not an easy task.

In developing welding technology for use in repairing in vessel components, researchers, licensees, and vendors must address many complicated issues.1:or example, in-vessel components typically have limited access (less than 36(r) (br inspection or repair, and high radiation levels typically require a remote repair approach or a well-shielded work environment. Other common issues include the structural integrity ofin vessel components, material damage attributable to the high-11uence environment, and the deleterious effects of helium present in the materials in high Iluence en"ironments, to name a few.

Currently, licensees are assessing a variety of repair methodologies through the llWR Vessel and internals Project (llWRVIP), a technical committee of the llWR Owners Group. These methodologies may yield alternatives to replacing components, an option oflast resort because of the high costs and radiation dose rates involved, in particular, the options ihr inoitu repair include of mechanical devices (clamps, etc.), welding, or some combination of the two.

llecause of the pressing need for sound repair technologies, the logical tirst step (for both Rl!S and the industry) was to embark on a literature review to determine the extent of previous research in this area and the current state of knowledge on related issues. This repon documents the findings of the

, literature review conducted by Ri!S, which was limited to open literature source in particular. this literature review revealed an extensive core of knowledge, substantiated by years of research, regarding general underwater welding technology (International Institute of Welding,

~ 1953; Ogden and Joos,1990; Grubbs,1993; Tsai,1995). Some of this research was conducted during years of funding from the oil and gas industry to develop techniques for repairing underwater oil platibrms. The purpose of this research was to produce better welds at greater underwater depths, predominantly thr components fabricated from carlen steels and stainless steels. Research is 1 NURl!G-16 ? t; i

1 i

ongoing in this area at the Navy Joining Center, hiassachusetts Institute of Technology (hilt) Ocean

!!ngineering Department, I!. O. Paton lilectric Wi.. ding Institute, Sea Grant Program at Ohio State University,lidison Welding Institute, and 1he Welding institute (TWI).

Ily contrast, this literature review revealed that substantially less research has been perfonned in the area of undenvater welding in commercial reactor in veswl environments. In practice, the weld repair in reactor inaessel environments is restricted by limited access to the coiaponent to be welded, high radiation levels around the component, and complications associated with radiation damage to the component (Shah and hiacDonald,1993). In addition, this literatme review revealed that most of the applicable literature regarding commercial nuclear plants conveys anecdotal experience involving emergent outage repairs in low fluence areas inside the reactor vessel (as opposed to carefelly controlled experimental work)(Jenco,1990) llecause of the scarcity of data pertaining to welding highly irradiated materials in an underwater environment, IlliS expanded the literature search to include a second base of applicable experience drawn from reactor tank repairs in a plant at the Department oflinergy (Doli) Savannah lliver Site (Sits). In particular, failures of reactor tanka in service at the Sits led to the development of welding techniques that would work for repair welds in high fluence environments, howeser, the reactor tank repairs at the Sits did not involve the complicating factor of underwater welding.

In addition, the ill!S drew a third base of c> perience from the fusion reactor international research program, which y ielded substantial welding research involving tritium doped materials that mimic helium entrapment effects without the damage nonnally associated with irradiated materials. hiost fusion research ellbrts has e focused on trying out various techniques for producing successful welds in these surrogate high fluence materials without having the constraints of working with highly irradiated material, l.ike the Sits research, howes er, this research has been perfbnned under dry welding conditions, not undenvater. Together, these three sources of research data and experience provide a comprehensive picture of the problems associated with welding highly irradiated materials.

They also suggest a variety of possible solutions to the pmblems, llecently, for example, researekert. in the Netherlands, United States, and Japan have conducted limited research involving the welding of irradiated materials, and future plans for the fusion community include more extensive research of this type. In addition, research into welding stainless steels exposed to high fluence Evels has shown some promise as a result of altering the heat input into the weld and applying a compressive stress during the welding operation. Current research funded by the Doli l'usion Program is focused on refining the techniques that produce acceptable welds in the desired ranges of tluence. Although this rc ,: arch has been peribnned exclusively with dry welds, it proves relevant as a comprehensive examination of possible techniques that can be applied to welding highly irradiated materials.

4 To date, the published accounts show great promise for undenvater celd repairs perfonned in low fluence in vessel environments of commercial nuclear plants. in particular, successful undenvater weld repairs have been made on austenitic stainless steel components with high-quality welds that have passed inspection during subsequent outages at Susquehanna Units I and 2, Peach llottom Unit NUltliG-1616 2

3, and River llend. This research suggests that a very useful rnaintenance toal could feasibly be developed if researchers can find a way to reduce or even eliminate the problems associated with .

welding high fluence materials, including helium induced cracking and effects from welding material damaged by radiation. Thus, a primary purpose of this report is to provide background on .

the problems associated with welding high-fluence materials, and is to discuss some promising research work and actual plant experiences that may enable this technique to be applied in commercial nuclear power plants in the future.

To achieve this purpc.sc. Section 2 of the repon summarizes the problems associated with undenvater welding ofirradiated materials in high fluence environments. Section 3 then briefly presents the welding codes that have been used by the commercial nuclear industry in the United States as guidance for undenvater repairs. Section 4 builds on that foundation by discussing the current base of knowledge regarding undenvater welding in general (largely derived in the context of the oil and gas industry). This is followed by a detailed discussion in Sections 5 through ? regarding specific un6rwater welding experience and techniques developed by conducting repairs at U. S. commercial s nuclean power plants and the Savannah River Site, as well as experimental research perfomied under the auspices of the fusion reactor program. Sections 8 and 9 summarize the existing knowledge base from the literature review and the directions that future work will have to take to address areas of uneenainty regarding undenvater welding ihr repairs in high 11uence environments ofcommercial nuclear power p! ants. Finally, Section 10 lists the various literature reviewed and cited in this report.

b 3 NUREG 1616

2. PitollLEMS ASSOCIATED WITil WELDING lititADIATED MATEltlAL IN lilGil FLUENCE ENVillONMENTS The literature reviewed in this study includes extensive discussion of two problems specific to welding highly irradiated materials. Specifically, these problems include ciacking during the welding process as a result of helium entrapment in the material and cracking attributable to irradiation-damaged microstructure in the components. In an ellbrt to understand the fundamental mechanisms and propose solutions to the problems associated with beham entrapment, a signifi: ant portion of the fusion program research has focused on welding tritium doped materials, thereby simulating helium entrapment in the absence of the damage normally associated with irradiated materials.

As the fbliowing equations indicate, helium is ihmied by neutron reactions with alloy constituents, primarily boron and nickel (Goods and Karfs,1991):

"D 4 n " 'Li + 'lle "Ni + n " "Ni + y "Ni 4 n "Fe + dlic IIelium tends to have a low solubility in metals, and Ibnns small clusters with diameters on the order of a few tenths of a nanometer et low temperatures, < 200"C (< 392 F), and 2 to 10 nanometers at high temperatures,200"C to 600"C (392"F to 1112"F). Goods and Karfs (1991) have concluded that:

" Preferred nucleation sites fbr this helium induced damage are lattice inhomogeneities such as radiation induced defects, precipitate interfaces, dislocations, and most importantly, grain boundaries. At elevated temperatures, these clusters and bubbles can grow very rapidly and under applied stress can severely weaken grain boundaries. Because conventional, deep penetration, ibsion welding processes produce high stresses (above the yield strength) as well as high temperatures (above the melting point), entrapped helium can teverely affect the weldability and postweid properties ofirradiated materials. The severity of the observed clTects is thought to be daminated by helium content and by the extent and magnitude of the temperature arid stress fields that prevail near the fusion line during welding."

Hy contrast, the " tritium trick" technique ofimplanting helium in the material by tcitium charging (American Society tbr Testing and Materials,1983), yields profbundly different levels of cracking after the material is subjected to the welding process when compared to materials that have been irradiated (Kanne et al.,1995). Kanne et al. (1995) has explained this phenomenon in terms of the preferential locations of the alloy constituents before transmutation and the nature of the transmutation products in the microstructure. (irradiation of the material results in transmutation of certain elements such as boron or nickel, but tritium charging only implants helium in the material.) Since boron is often present in steels as an impurity, the helium produced from that NUltEG 1616 4 il

element is expected to fonn preferentially at the grain boundaries. Nickel, however, is present throughout the matrix, so its helium transmutation product is also distributed throughout the matrix.

The presence oflithium as a transmutation product may also be detrimental to the welding process, and researchers suspect some difficult to measure contribution to the enhanced cracking during the welding ofirradiated materials as a result of the radiation induced damage to the metal matrix. The material exhibits this radiation induced damage in the form of dislocation loops, dislocation networks, and voids (I.in and Chin,1991b).

The available literature contains numerous instances in which helium induced cracking occurred during the welding of irradiated materials (llall c/ al.,1978; Atkin,1981; Kanne et al.,1995; Robinson,198H; Goods and Karfs,1991; Lin and Chin,1991a; Yan Osch ct al., 1994; 17abritsiev and Van der Laan,1996; and Watanabe et al.,1996). Ilowever, these reports lack consensus regarding the absolute minimum level of helium that induces cracking during welding operations.

This confusion results from the complexity and synergistic effects of the variables involved, including the base mt. erial, weld heat input, degree of radiation damage to the material, temperature of the rnaterial, residual stresses in the material, material microstructure, and so forth, in general, the most limiting helium values reponed in the literature are in the range of 1 to 2 atomic parts per million (appm).

Although techniques exist for measuring the amount of helium present in a material in a laboratory environment, it is far more complex and expensive to determine (or predict) the amount of helium present in in vessel materials that have not tw i removed fmm service. Consequently, researchers, licensees, and vendors generally predict the amount of helium in the reactor in-vessel materials using flux mapping techniques, retrospective dosimetry techniques, or similar calculated approaches, rather than directly measuring the content. liccause of the imprecision inherent in estimating neutron flux from one location to another in the vessel, the calculations entail some degree ofinherent error that is dillicult to quantify.

One interesting exception is documented in a Japanese patent record, which describes an invention consisting of a laser beam mechanism and a mass analyzer to measure the atoms generated from laser irradiation of the material, thereby determining inoitu the amount of helium in the material (1:ukutani,1995). Ilov vver, the literature does not . .ite whether this device is presently being used in the nuclear industry in Japan, nor does it substantiate the reliability of such mer.surements.

5 NURI!G 1616

3. WEEDING CODES USED FOR UNDERWATER WEl.D REPAllt in many environments, licensees and vendors base performed undenvater welding according to American Welding Society (AWS) specification D3.6," Specification for Undenvater Welding,"

isstxd in 1983. This specification discus:es welding variables, as well as associated procedures and welder qualification requirements. The specification primarily related to undenvater welding of carbor and low alloy steels common to offshore oil structures (O'Sullivan,1988). In 1989, the original version of the specification was supeiseded by the American National Standards institute

( ANSI)/AWS D3.6. 't he latest edition of the standard is AWS D3,6-93. The underwater weld repairs in commercial nuclear power plants cited in the literature were all performed in accordance w.th either the original specification, AWS D3.6, or the latcr specifica%n, ANSI /AWS D3.6.

According to O'Sullivan (1988), weld repairs in nuclear power plants are customarily performed uccoruing to the requirements of American Society of hiechanical Engineers (AShiE) Boiler and Pressure Vessel (Il&PV) Code, Sections IX and XI; however, the il&PV Code does not address the wet environment, submergenec depth, and other pertinent variables. Also, the steam dryer (the component frequently mentioned in the open literature about repairs at Susquehanna), was not originally fabricated to the AShili B&PV Code or any other standard code or specification. By the time the River Bend Station was Ibnnulating repair strategies for a damaged feedwater sparger pipe, the industry had accepted the philosophy that ash 1E Section IX would be used for " position", and AWS D3.6 wauld be used for " depth range"(Niahan,1990). Then, in 1989, the Electric Power Rese,uch Institute (EPRI) Non Destructive Examination Center (NDliC) reviewed the Type 3081.

stainless ste:1 weld deposits that were currently considered acceptable for most types of repairs in accordance with AhSI/AWS D3.6. As a result of that resiew, the NDEC detennined that the welds would not be acceptable for ASN1H Code repairs (Findlan et al.,1992). Consequently, in 1989, EPld initiated a research program to develop weldments that would qualify under Section IX of the AShlE B&PV Code.

To further clarify the application of ash 1E Sections IX and XI for unArwater welding, the ash 1E drafted Code Case N-516, which was approved on August 9,1993, under ash 1E Section XI, Division 1. That Code Case cited ANSI /AWS D3.6 as a general requirement for undenvater welding, and also specified requ:rements for procedure qualification, performance qualitication, tiller metal qualitication, confirmation welds, and weld examination.

Just recently, the NRC has given conditional acceptance to Code Case N 516 (NRC,1997). This acceptance is conditional because of the problems that have been noted in welding materials subjected to a high fluence environment, and because Code Case N 516 fails to recognize the condition of the material to be welded. Therefore, to ensure that the welder has used adequate crack prevention measures when welding highly irradiated Class I materials, the conditional acceptance criteria require a demonstration of the weld technique using a material subjected to a similar neutron fluence in a weld rcockup.

NUREG-1616 6

4. UNDEllWATEll WELI) LNG: GENEllAL 1)lSCUSSION C.E. " Whitey" Grubbs wrote an exhaustive article about the " state-of the ait" of undenvater welding, which he submitted to the 12th International Conference On Offshore Mechanics and Arctic Engineering, as well as the 1994 Underwater Intervention Conference (Grubbs,1993 and 1994). Since that time, the underwater welding technology has matured, with many organizations funding current research into optimized techniques and equipment (1 sai,1995). Further background regarding the history o rundenvater welding and the development of specific techniques and electrodes appears in numerous comprehensive articles (Ogden and Joos,1990; and Tsai,1995) and conference proceedings (International Institute of Welding,1983; American Welding Society,1991; Undenvater intervention '94 Conference Committee,1994). In his review Grubbs (1993 and 1994) states that wet welding was used as early as 1917 to stop leaks from the seams and rivets in ship hulls, but these were not structural welds. Documented reports of undenvater structural wcids by skilled welders were first published around 1970, and a considerable number of repairs were perfomied in offshore structures subsea pipelines, and dock and harbor facilities in the subsequent 3 c .rs. Ily 1985, welders and procedures were being qualified to a depth of 100 meters (328 n).

Grubbs also discussed the many problems specific to underwater welding, including the rapid cooling resulting from the infinite her. Tink represented by the surrounding water, the hydrogen

% enriched gaseous envelope surrounding the weld pool, and the reduction of manganese and silicon as well as the increase in carbon and oxygen associated with the increased hyperbaric pressure.

These problems can lead to poor weld quality. Ilowever, ihr the special case of weldinb austenitic stainless steel using austenitic stainless steel electrodes, Grubbs stated that welders have made underwater weldments that are metallurgically superior in some respects to similar welds made above water. lie further noted that 40 or more such underwater weld repairs have been made in nuclear power plants since 1980, mostly using Type 3161, austenitic stainless steel welding electrodes to weld Type 304 austenitic stainless steel base metal. In addition, Grubbs stated that anstenitic stainless steel resists transibnnation into martensite, so the weldment is essentially rapid-quench annealed during the welding process and thereby minimizes heat-affected zone (llA7.)

sensitization. Although other materials may not be as ideally suited for underwater welding, adjustment of weld parameters and electrode materials has provided good-quality welds that approach the quality of dry welds for the same material.

It is interesting to note that although " wet" undenvater welding was the technique chosen for the commercial nuclear plant repairs, it is only one of several undenvater welding techniques available to the welder. According to White et al. (1997), there are six general categories of undenvater welding:

1) In wet welding, water separates the workpiece from the torch.

2) Dry-spot welding uses a slightly pressurized enclosure to surround the weld pool and are volume, in order to keep out the intervening water.

7 NUREG 1616

I

3) Dry box welding uses a box that encloses the head and shoulders of the welder, as well as the weld area.

4) in habitat welding, the weld areas are isolated from the water. and the welder does not have to wear diving equipment.

5) Chamber welding uses a chamber that is maintained at a pressure of 101 kPa (1 atm) to keep the water out.

6) Remote welding uocs not require continuous welder participation during the joining operation (that is, the procedure relies on automatic or robotic welding systems).

Of there six, welding technique 1 has been the preferred choice oflicensees when making repairs to in vessel reactor components because the surrounding water affords additional radiation shielding.

Nonetheless, welding techniques 2 through 5 offer considerable advantages t>ecause the weld is essentially fabricated in a dry environment. (This is especially important for carbon steel welds, since underwater wet welds in carbon steels can suffer from a rapid qtanching effect and a susceptibility to hydrogen embrittlement, which can lead to porosity, slag entrapment, lack of fusion, and lack of penetration in the weld.) llor:ver, application of techniques 2 through 5 can be limited by the accessibility of the weld to effect the repair, and some weld areas simply cannot accommodate an enclosure around the weld area. The relative case of setup and execution also makes technique I attractive, especially for the spe:ial case of welding internal HWR vessel components, as long as the weld quality will meet tlic requirements of ASMi!Section IX.

NUREG 1616 8

5. IN VESSEL WELD itEPAIRS AT U.S. COMMEllCIAL NUCLEAR POWER PLANTS

!)uring the course of this literature review, the RES identified detailed accounts of weld repairs performed at Susquehanna Units 1 and 2, Peach llottom Atomic Power Station Unit 3, and River Hend Station (O'Sullivan,1988 and 1990; Mahan,1990; The Nuclear Professional,1990; and Rey nolds et ul,1991). (Table 1 summarizes the available infonnation about the weld repairs.) The reader should note, however, that the weld locations addressed in the literature were all in low fluence areas inside the reactor vessel.

lable 1. In Vessel Weld Repairs in U.S. Commercial Nuclear Power Plants Daic Plant kepair M at'l  % cid Depth Wcld l lectrode Contractor 1cc hnique Name 1984 \t.squehanna %tcam Dr>tr tid 3D4 $% N/A OT AW Dr) l R30tl. I dier Dolity l' nit 1 (Dr) wcidi Metal 1987 husquchanna Nicam I)r> cr llood 344 % % 18 to 4 3 m  % MAW l.ow Carlen 3161. Global Dners &

Unil l (61o14 ft) w'proprictary waterpnor Contramts coat.ng 14AR %usquehanna l redwater %parner Aust lim  % MAW 3161. I aller Metal w/Al Utiht)

Unit 2  %% (4) ft) wnicrproof mai:ng 1989 %usquchanna steam Dr>cr Dram 344 %% $m \MA% 3161. I iller Metal w/Al Utiht)

Und l Channct ( l ? 11 1 w attipnol (natmg 19n9 Ausquc hanna \tcam Dr3 ct lic- 304 %% I R io 41 m  % MAW ll61. lillcr Metal w/Al Uitht)

Unit 2 flod Capturc Plates 16 to 14 ft) waterprtof tuating iMN Pens h Itottom  % tram Drytt Dram 104 %% 4 to 6 m %M%W l ow.( aston 3161. Gl Nuticar and Unit 3 ( hanncis. (1) in 2310 wiproprictary waterrnor Global t hsces &

Alignment! (oaimg Contr actors henmic 1 up IVM9 R nct ilend l ecdw ater %parper 3161 l) m %M%% l.ow Curtun 3161 Ulobal Dn cts &

\tation Pipe (4) f t) wirroprietary waterpnol Contractors matmg The RES lbund it difficult to ascertain the scope of repairs in the commercial nuclear industry solely on the basis of an open literature search. One report made reference to similar underwater repairs by Georgia Power Company (Rey nolds et al.,1991). Jim O'Sullivan, of Pennsylvania Power and Light (PP&l.), stated that at least a dozen plants have made, or plan to make, similar underwater weld repairs, both domestically and abroad (The Nuclear Professional,1990). lie also mentioned that the Japanese have shown interest in, and become converted to the idea of underwater welding.

This is supported by a Japanese pa:ent application in the open literature, which describes a device for measuring inoitu helium in reactor structural components to assess the feasibility of welding in-

% essel structural components (Fukutani,1995).

9 NUREG 1616

Allegedly, diver / welders using a shielded metal are welding (Sh1AW) process, have manually perfbnned additional undenvater weld repairs on manway covers and jet pump hold-down beams; however, the details of the repairs are not readily gleaned from the open literature (1 indlan ci al.,

1992). According to l'indlan et al., research has been undenvay at the !! Pit! NDliC since 1989, to study techniques that can successfully be applied Ihr weld repairs in high fluence in vessel areas of commercial nuclear peer plants. Ilecause of accessibility concerns for divers in the lower portions of the reactor vessel, where components generally are not removed for repair and radiation levels are prohibitively high, automated remote undenvater wet welding processes are desirable.

Initially, the liPitt built on the base of available knowledge about underwater welding, most of which was initially developed Ihr olishore applie,tions (Smith and Childs,1991). hianual Sh1AW was the most common process, and defects such as porosity and lack-of fusion plagued much of the undenvater welds. The development work resulted in some irnprovement of the SN1AW techniques, including the use of pulsed welding arcs with commercially available electrodes to maintain a lower us erage current, thereby reducing weld spatter and improving penetration, lhe !!Plti also initiated the develor' ment of cutomated weld processes Ihr applications in the lower portions of the reactor vessel, anJ performed a literature survey to identify past uses of remo;e undenvater welding (l'indlan et al 1992). The results of the literature survey indicated that a water-free, dry habitat was used in a majority of the cases. On the basis of the results from the literature survey, a s endor survey, and past experience, the I!Piti selected flux core are welding (1 CAW) as the topic of further study. This study revealed that better quality welds were perfbrmed using commercially available stainless steel self shielded wires (compared to welds using gas shielded wires). With the self shielded wires, a gas bubble was created around the molten weld bead and the bubble developed into an easily removed slag deposit. The finished welds were quite sensitive to the welding parameters, including torch angles, electrical polarity, wire site and type, and voltage.

The !!Pitt further developed this technique Ibr a member utility, in order to add weight to ajet pump riser bracket to change the vibrational hannonics uhich could othcnvise crack the bracket weld. The process was successibily demonstrated under water with porosity and slag-free welda in non-irradiated material, but the utility did not need to make the repair during the outage. No further information about the use of this technique appeared in the literature.

The ibliowing sections summarize the particulars of each nuclear plant weld repair documented in the open literature.

5.1 In-Vessel Weld itcoairs at Susquehantu in total, the open literature documented five in vessel weld repairs at PP&l.'s Susquehanna Steam lilectric Station, Units 1 and 2. Of these five, welders at Unit 1 perfbnued the first weld repair above water, aller preparing the weld under water to reduce exposure. The next time a weld repair was required for the steam dryer, the exposure rates were prohibitively high and welders perfbrmed repairs under water. Welders also perfbrmed the three subsequent repairs to the steam dryer and NUltliG-1616 10

feedwater sparger under water. Sections 5.1.1 through 5.1.5 discuss the details of the individual repairs, and Section 5.1.6 summarizes the lessons learned from these repairs.

5.1.1 First Repair in service visualinspection of Susquehanna Unit I during the first refueling outage in 1984 revealed a crack in the steam dryer (O'Sullivan,1988). This crack was approximately 168 cm (66 in.)in length, and was located in the weld metal and along the fusion zone of the weld between the hood

. and end panel of the steam dryer. This weld failure was subsequently identified as resulting from fatigue. 'lhe welder therefore prepared the failed region ihr weld repair under water, but performed the repair dry using manual gas tungsten are welding (OTAW) with ER308L filler metal. The repair consisted of re-v.ciding the joint, as well as welding a Type 304L plate oser the joint to provide additional reinfbreement against cyclic loading.

5.1.2 Second Repair .

During the third refueling outage (October 1987), inspection person's! <liscovered a similar crack in another weld joint. This crack was 137 cm (54 in.) in length, and was a so identified as resulting from fatigue. I or this r: pair, the eleveted steam dryer radiation levels made dry welding undesirable, so the licensee considered underwater weld repair to be the preferable alternative. At the time, however, the industry had little experience in welding stainless steel under water, since most of the enderwa er welding up to that time involved carbon and low-alloy steels common to otTshore oil structures. PP&L thercibre selected a contractor,' Global Divers and Contractors, Inc.

of Port ofIbr,ria, Louisiana, because the company had a considerable amount of underwater welding experience, hypeihark test facilities Ihr procedure and welder qualification, and involvement in the AWS subcommittee on underwater welding to develop underwater welding techniques to repair the steam dryer.

Global Divers and Contractors, Inc. developed procedures and qualifications for SMAW with direct current on the basis of AWS D3.6 ( American Weiding Sociuy,1983) Type 11 welds on A240 Type 304 and Type 3041 stainless steels. The contractor a. u specified low-carbon stainless steel electrodes with a minimum ferrite number of 8.0, as well as qualification esting standards including weld position, visual acceptance criteria, radiography requirements, metallography requirements, chemical analysis requirements, as welded ferrite determination, and tensile and shear mechanical

, test requirements ihr single V-groove butt joints (requiring tis % minimum penetration) and fillet welds. The final procedure that the welders used for qualitication specified 75% partial penetration 30 groove welds on 0,32 cm and 0,64-cm (%-in, and %-in.) groove welds, with perfomiance testing of two each root and face bends as well as three macro specimens for each thickness. Divers prepared the steam dryer surface with a water grinder, and used a pneumatic chipping gun for interpass and final cleaning. The final lit up resuhed in a 90-degree V-groove, and the welding sequence consisted of placing 10.2 cm (4 in.)long tacks on approximately 30.6-cm (12-in.) centers beginning near the botmm, ibliowed by the root and cover pass. Divers then welded reinforcing

> 11 NUREG 1616

plates to each of the three comers. Inspectors perfomied final acceptance by a remote visual (video) examination, and all of the welds passed. In addition, the weids were re inspected during the fourth refueling outage in Spring 1989, and the welds were still deemed intact (O'Sullivan,1990).

5.1.3 Third Repair During the second refueling outage in May 1988, misalignment of the separator guide lugs and the vessel internal guide rods led to damage of the Unit 2 fredwater sparger (O'Sullivan,1990). The damage resulted in a 3.8-cm (1%-in.)long open split on the upper face of a nozzle on the feedwater ,

sparger. The weld iepair consisted of a welded overlay over the horizontal indentation in the nozzle, and a built up closure patch applied to the vertically-oriented open split. The welders were qualified for groove welding P-8 (austenitic) base metals, using 316L filler metal with an aluminum waterp;oof coating, between 0.32 and 0.95 cm (%-in. and %-in.) thickness in horizontal and vertical positions at depths from 3 to 19 m (10 to 63 ft.). As before, insp;ctors perfonned final acceptance of the welds by remote visual (video) examination, and the welds passed. In addition, the welds were re inspected at the next refueling outag:(Fall 1989), and the repair welds were still deemed intact.

5.1.4 Founh itepair During remote video inspection of the st:am dryer during the fourth refueling outage of Unit 1 (Spring 1989), inspectors discovered a 46-cm (18-in.) long fatigue crack in the horizontal fillet weld between the top of one drain channel and the dryer support ring, and the crack advanced about 7.6 cm (3 in.)into the 0.32-cm (% in.) thick base metal of the drain channel (O'Su'.livan,1990). Divers i therefore drilled a 2.54-cm (1 in.) hole into the lower end of the crack in the drain channel to prevent further cuck propagation. The welder also used the sparger repait procedure cited in Section 5.1.3 to ensure welder qualification for overhead fillet (4F) welding of 0.32-cm to 1.27-cm (%-in. tn % in.)

and :hicker austenitic stainless steel bas: metal. In addition, the welder removed the cracked fillet weld, as well as some additional weld material at the end of the cracked weld, and restored the horizontal weld to the maximum size allowed by the configuration. Inspectors performed final weld acceptance by remote visual (video) inspection.

5.1.5 Fillh llepair Inservice inspection during the Fall 1989 outage revealed a crack-like indication perpendicular to ,

a venical drain channel to skirt weld in the steam dryer (O'Sullivan,1990), which prompted PP&L to devise a repair strategy for a possible crack. PP&L also had a plan for a contingency modification in which stainless steel capture plates would be welded over the openings of the steam dryer

• containing tie-rod end nuts which held the intemal baflies during original dryer assembly. (This was a contingency modification because tie-rod end nut cracking had been discovered at other plants, although in service inspection at Susquehanna had not revealed any indications of cracking in the end nuts.) Close visual inspection of the indication by a welder subsequently revealed that the indication was not a crack, so the welding stafT simply planned and executed the capture plate NUREG 1616 12

installation. He welder qualification required s;ngle pass fillet welds in the horizontal, vertical, and overhead positions (2F,3F, and 4F).

5.1.6 Summary of Susquehanna Weld Repairs Underwat weld repairs were successfully performed at Susquehanna, Units I and 2, and were documented carefully in the open literature. Jim O'Sullivan, Group Supervisor ofInstallation Engineering at Susquehanna (also the plant welding engineer for the period of the weld repairs) not only provided a detailed summary of the weld repairs, but also commented on the fine quality and integrity of the welds perfbrmed with this technique (O'Sullivan,1990). lie stated in the literature that a welJer can achieve acceptable results from welding carbon steel under water, but can attain execIIcnt results from welding under water with stainless steel (The Nuclear Professional,1990).

5.2 In-Vessel Weld Renair at Peach 130ttom Atomic Power Station Unit 3 During an outage at the Peach llottom Atomic Power Station Unit 3, operated by the Philadelphia Electric Company (PECO), the steam dryer was damaged as a result of misalignment with the reactor guide pin during reinstallation inside the reactor (Reynolds et al.,1991). As a result, the steam dryer required repair of structural damage at the 0 and 180 azimuth, cracks at the welds at the base of the drain chsmels, and an alignment / seismic lug located at the 270 azimuth. PECO therefore contracted with General idectric (GE) Nuclear Energy and Global Divers and Contractors to plan and execute the necessary repairs. Welders were qualified Ihr four different welding positions (20,30,2F, and 3F) on base metal thicknesses from 0.32 cm to 2.54 cm (% in. to 1-in.)

at working depths from 4 to 6 m (13 to 23 feet). The welding procedure and qualifications were in accordance with AWS D3.6 - 83. The filler metal was an E3161, SMAW electrode with a proprietary waterproof coating, and the base material of the steam dryer was SA240 Type 304 stainless steel. Qualification tests included ferrite detennination, chemical examination, visual examination, radiography, tensile tests, tillet weld shears, root / face / side bend tests, and macroexamination. Weld areas were prepared by using a water-powered grinder or an are water-gouge tool, and pre-fabricated assemblies (such as plates, a dryer skirt, a support ring, and a guide bracket) were welded into place. The welds were all inspected using underwater video cameras.

5.3 In Vessel Weld Renair at River llend Station Inservice inspection at the River llend Station (operated by Gulf States Utilities) revealed a damaged nonle on the feedwater sparger pipe (Mahar,,1990). The sparger was constructed from 15.2-cm (6-in.) Schedule 80 SA312, Type 3161. pipe, and specifically, the rersair locations were 13 m (43 ft) under water, at approximately 3 m (10 (I) above the vessel's loaded core. Further inspection J identified the cause of the damage as a misplaced sca' Told knuckle clamp and revealed a second misplaced scaftbid knuckle clamp in a separate sparger. In consultstion with GE, the licensee decided that the header design would still perfbrm its function with two a.ess nozzles per sparger, so the licensee des eloped a repair technique to repair the sparger and eliminate the alTected nozzles.

13 NUREG 1616

Gulf States Utihties then contracted with Global Divers and Contractors to assist with the repair.

The contractor provided mock-up training at their hyperbaric welding and testing facility. The expected repair activitics included cutting off the nonle, cleaning the outer surface of the pipe, removing the sc-ffold clamp, preparing the hole and replacement plug, and welding the plug into place. Welders were qualiCed in accardance with AWS D3.6-83 for type "O" welds, so that ASME )

Section IX could be used for all essential variables not afTected by the underwater environment. The welders were also specifically qualified in the 3G position on 0.95-cm (%-in.) groove welds with backing. In addition, because the contractors used this approach to qualify the welders for position under the ASME Code and depth range under the AWS D3.6 specification, the procedure was '

qualified foi all positions and depths from 1.5 to 16 m (5 to 53 ft). The contractor also used a diver's workstatior,10 shield the welders and to suppc,rt a loose parts dam cloth positioned to capture any .; '

cutting dross, parts, or tools. An electrically driven positive-displacement water pump provided wr.i., c ever for the water grinder and arc-water gouging unit. The contractor preferred water-driven power tt,18 over the pneumatic tools, because they eliminate the loss of visibility that otherwise resu!!s from euessive bubbles under water in addition, the utility used video equipment, including fiber optics, TV monitors, and underwater cameras, to record the process and to inspect the final

(

welds According to the published account of the repair (Mahar 1990), the process resulted in a good-quality, permanent repair of he t sparger pipe.

i O

NUREG-1616 14 j

6. WELD REPAIRS ON lilGil FLUENCE MATERIALS AT Tile DOE SAVANNAll RIVER SITE Since the 1950's, the U.S. Departn.ent of Energy (DOE) has built and operated five production teactors at the Savannah River Site (SRS) to produce radioactive materials for national defense and peacetime applications. The reactors were unpressurized, operated at temperatures below the boiling point of water, and used heavy water as a moderator. One reactor had a unique design that

] incorporated a curved knuckle to join the reactor tank sidewall to the reactor tank bottom. The tank

g. was fabricated from high carbon (-0.07%) Type 304 stainless steel, and the knuckle region was sensitized during fabrication and subsequently experienced stress corrosion cracking (SCC) during I,

E, service. In 1968, welders used stainless sw,.1 patches to successfully repair the knuckle region, and these patches were welded to the tank walls using GTAW , cnniques (Maloney,196W Kanne,1988).

These weld repairs were raade in air, net under water.

in 1984, the tank again began to leak, aad SRS staff planned and executed weld repairs that were R, similar in scope to those performed in 1968 (Kanne et al.,1987; Kanne,1988). Again, the weld repairs were made in air, not under water, and autogenous GTAW was used to install repair patches in the 10' R/h radiation tield of the reactor. During this process, SRS stafTnoted extensive cracking i in the ll AZ of the tank side of the welds, but the patch (new metal) side of the welds was free of g cracks. Extensive analysis eliminated many potential causes of the cracking, including incomplete fusion to the tank wall hydrogen embrittlement, pre-existing undetected intergranular stress

{= g corrosion cracking (IGSCC), intergranular attacks from pickling durin; fabrication, and radiation-c.,

induced segregation. SRS welding staff suspected helium induced cracking because of the known elTects of helium increasing low-temperature strength and decreasing ductility. The staff estimated that the reactor tank walls contained approximately 3 appm of helium, and calculated fluences to be 1 x 10" n/cm2 thermal and I x 10'* n'em: fast (Kanne,1993a). In addition, staff at Sandia National Latyratory purely computed fast lluences (greater than 0.1 MeV) and thermal values ..ere based on -

the measured helium content with derived corrections tbr boron contributions and extrapolations to the surface. (Iligh stresses and high temperatures introduced by welding were thought to contribute to the propensity for cracking.)

llefore this observation, only a few observations of cr.,cking or porosity had been attributed to welding of materials with entrapped heli im (llall et al.,1978: Atkin,1982; Schiller et al.,1987; Clark et al.,1986). Consequently, SRS welding stalT perfermed tests to confirm their suspicions about the . ole of helium in the ll AZ cracking (Kanne et al.,1993b). Specitically, the stalTperformed GTAW test welds identical to the tank repairs on trit um-doped 304L plates with preexisting GTAW (befoie the tritium deping). in addition, because a concentration gradient of helium was present on the plates as a result the tritium charging process, the stafT was able to make observations for a high-helium weld environment and a weld environment relatively free of helium on the same plate. As a result, the stalT observed cracking only in the llAZ region of the tests welds in the areas that contained helium, and the preexisting weld exhibited no cracks. Low heat input welds with base metal melting and spot GTAW also produced cracking, but resistance welds or a low heat GTAW pass with no base metal melting caused no cracking (Kanne,1988).

15 NUREG-1616  ;

1

I These findings led to a comprehensive research effort to develop welding techniques that would reduce the helium induced cracking associated with the welding cfirradiated mat: rials. (fable 2 summarizes the combination of repairs and research at the SRS.) During this research, the ulding staff considered at least 30 different processes, but concentrated on the gas metal arc (GMA) overlav weld technique utilizing very low heat inputs, because this technique showed the greatest in:ual promise (Franco-Ferreira and Kanne,1992; Kanne et al.,1991 and 1994; Kanne 1993a and 1995; and Perra,1990).

Table 2. Weld Repairs and Research at the DOE Savannah River Site ,

Date Item Welded Material Description am:=

1968 Tank knuckle region 304 SS GTAW repair of tank with SS patches; j no problems identified with repair f, 1984 Yank knuckle region 304 SS GTAW repair utank with SS patches attempted; resulted in extensive HAZ cracking 1984 to 1994 Specimens from tank wall 304 SS Reduced amount of cracking in ll AZ observed for GMA overlay technique; postwew crack lengths in irradiated materiais measured 28 times greater than for tritium-doped material Studies showed that the heat input to the irradiated materiai must be minimized, since growth of the helium bubbles (resulting from a diffusion creep mecitanism) from their initial nanomete: 3ize to a

approximately I micron in diameter occurred at temperatures above 50u C (932 F)(Franco-lierreira ,

and Kanne,1992; and Kanne et al.,1995). Th,: SRS welding staff therefore developed a weld overlay technique utilizing Type 308 tiller wire to produce a weld weave 2.6-cm (1-in.) wide and 0.9 mm (0.035-in) thick (Kanne,1993a). The extent of the weld penetration into the base metal was only 0.08-mm (0.003-in), which reduced the size of the II AZ. Welders performed test welds on 15-cm (5.9 in.) discs cut from the wall cf a SRS reactor tank, with measured helium concentrations or 10.4 appm on the inside surface and 5.0 appm on the outside surface. No large surface etacke were (

visible in the ll AZ around the edge of a weld bead, either by dye penetrant or rnetahographic

• analysis. Ilowever, approximately twice the ainount of underbead cracking was discovered for the 10.4 appm region as for the 5.0 appm region. In addition, some minor cracking was obserwd in the q llAZ, but far less than that observed for conventional penetration welds.

The SRS staff also used tne weld overlay technique on tritium enarged specimens, and compared the micrestructural features from these welds to thon found as a result of conventional weld techniques by Sandia National Laboratories (Goods and Kerfs,1991; and Perra,1990). Cracking in the llAZ i

NUREG-1616 16 1

)

I was found for all of the weld techniques (even the low heat input weld overlays), and the degree of cracking positively correlated to the helium concentration and heat input.

The SRS welding staff directly compared the post weld crack lengths measured in materials subject to helium implantation by tritium decay and irradiated materials. In comparison to the overlay test welds performed on tritium-charged and aged materials, the measured crack lengths from overlay welds on the irradiated ma*erial were up to 28 times greater. Kanne et al. (1993) concluded that the low heat input weld overlay technique showed promise, but needed additional development work on irradiated materials. Unfonunately, the last tritium production reactor at SRS was placed in cold-standby in 1993, and finally shut down permanently in 1996, so further development under the auspices of this welding research and develop.nent (R&D) program was suspended.

i 7

l 17 NUREG-1616

l.

l

7. INTERNATIONAL FUSION PROGRAM RESEARCil ON WELDING OF lilGill3 IRRADI ATED M ATERIALS /

The DOE sponsored fusion reactor research program has provided an especially active area of welding research into the pn3blems associated with helium induced cracking. Becau: of the expected severity of the materials environment, research has been ongoing for a number o fears, both domestically and intemationally, to develop repair techniques for components that are degraded by exposure to high fluences in future fusion reactors. In particular, the available literature includes extensive reports of collaborative efforts between the research stafTs at Sandia National Laboratories, .

Oak Ridge National Laboratory, and Auburn University.

Table 3 summarizes the significant fusion program research in the open literature regarding welding of highly irradiated materials, and Table 4 identities the materials used in the DOE-sponsored fusion research program. Lin et al. (1989) noted that intergranular cracking was found in the llAZ of welded Type 316 stainless steel doped to 27 and 105 appm helium. The weld technique was GTAW with the plates under full restraint to simulate structural restraint. Lin et al. (1990) continued this rescarch by observing the effects of changing the welding parameters and helium concentrations.

They tried both full and partial (30 to 50% depth) penetration GTAW welds under full constraint for materials doped with helium to concentranons of 0.18,2.5,27,105, and 256 appm. They noted that Ihr the full-penetration welds,intergranular ll AZ cracking was observed for 22.5 appm helium, and no cracking was observed for 0.18 appm helium. For the partial-penetration welds,intergranular il AZ cracking was observed for specimens with 105 appm helium.

Lin and Chin (1991a) continued this research by obsersing the effects that modifying the alloy metallurgical condition had on GTAW welding of highly irradiated materials under full constraint.

They used Type 316 stainless steel in the solution annealed (S A) condition, with a 20% cold work (CW) condition, and a titanium modified condition referred to as the " primary candidate alloy" (PCA). Again, the materials were doped to helium levels of 0.18 to 256 appm.1.in and Chin observed that for the 316SA material, severe, continuous llAZ cracking occurred for >2.5 appm helium, but the material was free of weld defectc at 0.18 appm helium. They also c'iserved that for the 20% CW material, cracking occurred in the llAZ at >l.9 appm helium, was less weld cracking occurred at 1.6 appm helium in the PCA material.

In collaboration with Wang, Lin et al. attempted a ditTerent techn:que to control the helium-induced cracking during welding (Wang et al.,1992) They used a hydraulic plate fixture to apply a .

controlled compressive stress perpendicular to the weld path (55 MPa (8 ksi),25% of room 1 temperature yield stress) during the GTAW process under full constraim. No visible cracks were ~

observed in the llAZ for Type 316 str.inless steel doped to 256 appm helium. They continued this research by varying the magnitude of compressive stiess and levels of helium in the material, in order to determine the minimum level of stress necessary to mitigate helium-induced cracking (Wang et al.,1993,1995), They used the GTAW process to perform full-penetration welds on fully constrained plates doped with 0.7,1.5,5.2, and e ppm a helium, subjected to compressive stresses between 55 and 165 MPa (8 and 24 ksi). They observed no cracking for the material doped to 5.2 NUREG 1616 18 1

1

Table 3. haternational Fusion Program Researen on Wekling of Highly hradiated Materials

. Researcher / Mat 1 Weld Tectumque Metimd oflie Ile Izvels I".neme Resums Year Producti<m Un er al,(1989) 316 Autogenous GTAW- Dayed by"tnanars 27 and 105 agyrn N/A IG !!AZ crackmg at tsoth Ile levels fall penetrationin5er trai" full constramt Un er al,(1990) 316 GTAW - fun and Dnped by "intaan 0 18,2 5,27,105 N/A 1 uD nenetraum welds- 10 IIAZ ancbng fw a 2.5 partial penetratsm irik" and 256 a,pn asyn lie, tuo crechng ebaersed for 0.18 asyn lie 00 to 50*4 deph) Partsal ocatration @ 10 IIAZ cW for 105 under full constraint asyn lie spec nens Un and Cim(1991) 316 SA GTAW - fn't Doped by "intann 0.!8 to 256 aspa N/A 111:0.18 appa free of weld ifects, severe.

316 20% CW penetration under full trit" cordansous IIAZ fw >2.5 asyn 316 PCA urstraint 29*v CW: cracking in IIAZ>l.9 agers Iga;less weld cradmg at 1.6 appa Wang er al (1992) 316 GTAW- fuH Doped by"intaan 256 appn N/A Ilydrauhc plane fixture used to apply comaroEed penetratzen trici " ccergpressive stress perpenhl- to sse meld push (25% ofYS -at RT1 no visible cracks in IIAZ Wang er af,(1995) 316 GTAW - fun Dryed by"tntman 0.7,1.5,5.2, and 10 N/A Ilydrambc plane 6xture used to asyly 55 to 165 MPs c penetrataar under fun I trui" asyn carryeauve stress; no crackmg found for 5.2 appe constraint Ile with fixtier, no cracking foisul for 0.7 appa lie w/out 2tture Wang eraL,(1996) 316 20% CW GTAW- fuH Irraduated in ORR 3Its. N appa target thsence Stresses of 0,55, t 10, and 165 MPs applied PCA 25% CW penetration PCA: 86 atyn 7 dpa perpsubcmaar to weld de.ctiest; IIT-9 IIT-9: 2 asyn for 0 streur crachng in 316 and PCA, no cradmg in IIT-9 for seresses 4 no cractung ebuerved en melds Goods and Karfs, 304 Cornersi<nal Doped by %iurn 2.7 and 85.0 appn N'A '

HA7, cradsng in aD specunens (segnaGcantly less in (1991) GMAW: deep trkk" bw heat irest overlay weldst pemtration GMAW cstriar Z nheDow pemtration C

y Goods and Yang, 304 GMAW overlays: Dcred by " int-an 22.5 and 35.0 arven N/A Degree ofcreding not sigraficarmly affected by tlee f"1 (1992) shallow penetratson trik" asylocation of sectual overlays counpared to sangle-9

~

m ==5d= 1

W l a 41 jrh-I.I 11 it 5 If'{

p ll $~i ia JJ]{1 41 .a.p}ig 1 i{ .n . j

!)I$ 3lIl !Aii

~

i 1 .,

l h

! di A i  : "aii 1 E sk a  ; as a$

E $'

T 4.n lj 'l {!f l i

2 .i .!i J -

.> P

8 sa is 151 J d lin s

i is 4

di E i a f e s =

3 1

D 1

a e[l j i > ,

NUREG-1616 20 E

1 1

1 1

3-i

{ l I

I s

. a.

8

- n. n - - . l I. 8

- l 5

. I. . l l l l B. I R

i - -

1 I I I  ! ! ! E  ! S 5 E I I E I E  ! S i I i l l I h

f& A b

-l m

I I I I E I E I B I I I I 5 1  ! ! D I I I I I I 4

.E 1

om

• E

] { i s

e s

e n

. E. E

- s. a. . E. E. l o. a. E. I.

a s

e i a

. n. x.

t

.8 1a J

j 2 I I I  ! l i i  : n 1 5 i E I  ! I  ! I I I 5 I I 1

s U E J I I I z  ! s E  :  :  ;  ! I I 3 i E

i 5; 1 I  ! $ I I

---

------------m--- ---

L-i I $ $$ $$ $ $ $a $r s$ e? 2$ai ri i:$.

a 2 4 x a a e a s

$, $, e$ $ $

o r $.

21 NUREG-1616 i

l l

appm helium using the hydraulic compressive stress fixture, whereas the lower limit for no observed cracking was 0.7 appm helium without using the fixture.

Because this technique of using compressive stress during welding seemed promising, Wang et al.

(1996) then tried the technique for Type 316 - 20% CW, PCA 25% CW, and HT-9 (fenitic steel heat treated to a tempered martensite microstructure) that had been irradiated in the Oak Ridge Research Reactor (ORR) to a target fluence of 7 dpa. Because of the different concentrations of Ni and B in the material, the helium concentrations were different for each alloy (75 appm for 316,86 appm for PCA, and 2 appm for HT-9). For tests with no applied compressive stress, no cracking was observed in the HT-9 material, but cracking was observed in the other two materials. For applied stresses of SS,110, and MS MPa (8,16, and 24 ksi) applied perpendicular to the weld direction, no a cracking was observed in the welds.

Concurrently at Sandia National Laboratories, Goods, Karfs, and Yang v ere evaluating weld overlay techniques to mitigate the helium-induced cracking caused by welding (Goods and Karfs,1991; and Goods and Yang,1992). W material they used was Type 304 stainless steel that was doped to 2.7 and 85 appm helium usin; the " tritium trick" method. Welding was performed using conventional GMAW (deep penetration) and GMAW overlay (shallow penetration), and significantly less HAZ cracking was observed for low heat input overlay welds. The efTect of adding a second overlay pass hs (compared to a single pass weld) was evaluated for Type 304 stainless steel doped to 22.5 and 85.0 appm helium. It was observed that the degree of cracking was not d;pificantly afTected by the application of second overlays.

Van Osch et al. (1994) in the Netherlands also published results of mitigating techniques for welding highly irradiated material. Specifically, they used a Nd/YAG laser welding technique that provided a low heat input to the material. Thev also used a 316L-SPH alloy, irradiated to 0.6 dpa in the high-flux reactor (HFR) in Petten. This level of irradiation yielded a level of 7*2 appm helium in the material. The 316L-SPH alloy is a designation given to a 316L(N) alloy used in the Superphenix fast breeder reactor. As a result of this research, Van Osch et al. observed no surface cracks after the welding proccu, although metallographic sections showed some weld cracks and porosities (mainly in the fusion zone). Since they found surface cracks in the un irradiated material, the cesearchers believed that the cracks were caused by insufTicient control oflaser weld parameters, and not necessarily by helium induced cracking.

i.

Fabritsiev et al. (1996) also performed welding research on the same material (316L-SPH) using automatic ciectron b:am welding. In this research, helium was implanted into the material by cyclotron, at levels of 50,100,300, and 860 appm. (A cyclotron ms used to incorporate defect ,

generation into the pre-welded material, because defect generation is not associated with materials doped with helium from the " tritium trick" procedure.) Fabritsiev et al. did not observe microcracks in the HAZ or in the weld for the material with helium levels from 50 to 300 appm, and noted only very small microcracks in the welis of specimens with 860 appin helium.

4 NUREG-1616 22 i

i Japnese researchers used yet another technique for welding irradiated Type 316 stainless steel.-

Watanabe et al. (1996) butt welded, with tungsten inert gas (TIG) are welds,10% cold-worked Joyo Plant (an experimental fast reactor) wrapper tubes irradiated to 7 to 22 dpa, which resulted in a ,

calculated 3 to 9 appm helium in the material. The tubes were irradiated at a fluence of(1.6 to 4.6) x 102'ivn 2 (E:t 0.1 MeV). Although HAZ cracking was not observed after welding, the irradiated weld joint specimens fractured at the HAZ during tensile testing after little plastic deformation.

23 . NUREG-1616

i

8. DISCUSSION Table 5 summarizes the various techniques that have been used for underwater welds in low-fluence areas of the reactor core, for welds involving highly irradiated materials in the fusion program, and for welds used on the DOE SRS reactor tank material. From the underwater welding experience in the nuciear plants, it is obvir us that stainless steel can successfully be welded in low fluence reactor core environments of BWRs. It also appears that reducing the heat input to the base metal can minim'.ze or possibly eliminate helium-induced cracking during the weld process. However, it must be emphasized that there are no repons available in the open literature that describe successful attempts to weld highly irradiated components in situ in the in-vessel environment of a BWR. In the absence of documented welding efforts invo' .ing highly irradiated reactor in-vessel components, assumptions about the welding behavior must be surmised on the basis of relevant fusion research and DOE SRS reactor welding experience.

Many of the techniques discussed in this rer, ort seem promis 5g in their ability to minimize the heat input to the weld, thereby reducing the size of the liAZ and minimizing helium bubble growth. The common variable among all of the successful welds on highly irradiated material noted in this review was the low heat input that was characteristic of weld process in addition, the use of a compressive stress to minimize helium bubble growth during welding in a laboratory environment has shown great promise as far as reducing the helium-induced cracking observed, but it is unclear how this technique could be applied uniformly to a component in a reactor in-vessel environment.

In order to compare the technical merit of th: various weld techniques, industry would need objective ctiteria. Researchers in the fusion research community listed the following welding process requirements that they were cor.sidering for fusion reactor applications; these requirements can also apply to the development of welding techniques for repairs of BWR reactor internals (Goods and Karfs,1991):

amenability to remote execution in hostile and nearly inaccessible reactor core environments ASME Code recognition and qualification

• the prospect of fabricating an actual load-c.aTying structural repair (rather than a superficial covering over a crack) chemicai and radiation compatibility of the repair material with the reactor .

• inspectability of the repair It seems reasonable that repair techniques for use in light-water reactors should be evaluated against the abne criteria to ensure successful and reliable implementation of repairs.

NUREG-1616 24 l

l l l ,l1 llJ iI1I?l I t s s O n s s S r e e g hs a o w o t r n d t

u s t l

s i

n l, s t l i

r d d e w b, d de fo e e iW4 e s = s r s r a n w d' ocr i

a -

ds er e l

p d e

lp a e r ,

n p p -

ct s d t

- r o

s o a e r  ? e c s e lea w n

l e

u c p s P s r a is a. 5a wf ye l

i r

w l

woA s Z A

e s M s a s -, i t

. se e r na a

wgn l l

a 5 s w nld! Z f

ri $ ev r e e e n eants A 4i 6i 9 n i

n gs e

s u

a re as f am d

n e

ew c

e ya J.

- W'o,ia d a

H d e

vt r

e e s 1

a me ap a d o fas pl s re p vn e d n

a t n e4 J- eh v o t 6 el idia m9 e 8 c a

r n w s a

va r

e kr rae l

lao ns h i sH ci e

fo u

s ei s wm 1 ci f o mnw ai wp i n e o -- u s t e  : o m mn t oe s wi r us s s on Wn gg g r

e g

n d s

r wno n,s i r

e

wwh@eosdinoMH n

w'_

i ifa e na s pd c e e k7 i o p c s uet e r

n %A a r n a o k e d s c a u a r pr l sd e a0 r rl c ,

ds u ne ra m

e MZt

_ i Ae k m a

- d l

e c ew n rr o

c a f 'alec a

Z=lr Ab nev a eh e r nrv a0 o

0 3

Z A

c M. .

M-c .

R WHw WD Nre Lam Hio n ae iap Nt o HJ e

e s. d e " e t H. p o

e f o . d d d e m m o

s,M *^r d d -. d-r a c e e e p d :.. s a

s n s e e m i bm m ed- a ai /. -

i d d n pa el m d A R t e . .

a r r a

r e

c o r be t a

r a

r C.

y =

r i

s Mw I r

I r

1 r

sdi r lH T I r

C I r

ht i

n 6

W- L 1

. 3 9 H) 16 P H))

,m a 3 6 W,T 1

S5)-

PN S(

i r & H. - L f

e s e

4 4 4 3,C.

65 A 4 4 L16 6 1 6 16 6 R M i

0 0 0 1 0C 0 0 1 3 1 1 3 3 3 32P 3 3 3( 3( 3 3 s

e u

q  ;

W- pa n hg r e

w o J

r. ))

1 2 ) l a

a r

n p 99 4 9 d e

id l

e m S e

e S

e e

r dn u

e -

s e

99 1 1

((

fs g 9

1

(

n a )

r e r r n V 6 9

Wf i

r e i v

e o

e s

m a

r C e

KY r

a a L m d n

9 1

R R Ji g t a

o r u ;c o E dd e (

e y dr h

a h

s p D nn ds i

b ,

r a

e

_ n n tr a. m N a a s s O R.9 )

6 a

n m e a a . r I dd n v v

. i R o o n 9 ta a

m e a a - s. P o o a '

i m R S S C F. E GG V F(

1 W

S g

. s s

5 d hr c r

n a

a )

Wa lv.

a g m )

y y

r n d T t a

i h k M d

(

r (

s r e d a .

l p w

. m e

) d d p a e w e

y r n e s p ) b c r

p d asl l e y n a

. (  ! e

! u e r o y n v p d r O m e fi e d. y ( t c

u .

a g e I

T d ) l r )ye)t ksl )) )l yr e y e v

ye tete r e m le d(r tsx r

T hi e d((

w m

( (w d v f o dn k N_

W

(

A T

W A

M WWAA TM Wd Al Tp p i

e WW A A Ao MC Ww Mtan l

w e

r e

s a

6.

Wm u

W O G OS Oa SF Gs l A B

)

s & t o t o o t to de t o

_ (

e s

e 84

%89 44 89 49 88 96 89 9 es 8 e 1 2 99 4

9 5

9 6

9 9 ri 99 99 99 99 9 9 9 D I 1 1 1 1 1 1 1 1 1 1 1 1 1 i

ytr. 2 k oi.a h r

lll

In addition, the industry has made substantial progress over the last 10 years in developing a thorough understanding of the physical processes that cause cracking in materials that contain helium. This progress has led to some promising techniques for solving the problems encountered while welding high fluence materials. Collaborative efforts involving those experienced in welding highly irradiated materials and those familiar with the demands of welding in the in-vessel reactor environment would seem a logic $tl next step toward developing viable weld repair strategies for degraded BWR core components.

b M

w b

1 r.

i 1

a 4

s o

=

f NUREG 1616 26

k 9.- CONCLUSIONS' Admittedly, an open literature review may not capture all of the pertinent information for a cenain

. topic, and conclusions based on the open literature for an active field ofresearch may be incomplete, at best. However, the following can be summarized by what was found in the open literature

- relevant to underwater weld repairs of highly irradiated materials.

- .- Repair techniques for BWR in-vessel coraponents may be required in the future as a strategy to manage aging in reactors. Underwater repair methodologies represent attractive options because the water shields workers from the high radiation levels expected from' aged in-venci components.

. Weld repairs, using conventional welding techniques, of BWR in-vessel components fabricated from stainless steel and exposed to high fluences may be unsuccessful because '

of helium-induced cracking and radiation damage in the matcrial.

. Underwater weld repairs have successfully been applied to BWR in-vessel components exposed to low fluences (i.e., steam dryers and feedwater spargers).

. Both field experience and research involving welding of highly irradiated materials suggest that helium-induced cracking can be found a.fler welding stainless steels containing as W.le

as 1 appm helium.

! = Researchers have developed techniques to mitigate the deleterious effects of the helium l bubbles contained in the highly irradiated materials during welding. These techniques limit the heat input to the welds (to limit the temperature-controlled growth of the helium bubbles) j and apply a compressive stress perpendicular to the weld path (which is thought to alter the helium bubble growth kinetics).

i . Wang et al. (1995) has performed successful welds on 20% cold worked Type 316L stainless

! steel with up to "5 appm helium using a compressive stress of 55 MPa (8 ksi).

Future work that could benefit the light water BWR community should concentrate on mitigating L the effects of helium in the concentrations expected for pressurized water reactors, as compared to

( the prior focus on helian concentrations expected from fusion reactors. Optimization of techniques that have shown success in the laboratory may provide techniques that can te used for future in-

- vessel repair needs.

1 t

5 27 NUREG-1616 T

10. REFERENCES Atkin, S.D., llanford Eng. Dev. Lab., Richland, Wash., Alloy Development for Irradiation Performance Semiannual Progress Reportfor Period Ending September 30,1981, DOEIER-0045/7, U. S. Department of Energy, Office of Fusion Energy, pp. I10 - 117, March 1982.

American National Standards Institute /American Welding Society," Specification for Underwater Welding," ANSI /AWS D3.6 - 89, Americe.n Welding Society, Miami, FL,1989.

American Society ibr Testing and Materials," Guide for Simulation of Helium Effects in irradiated Metals," ASTM Standards Vol.12.02, E 492-83, pp. 808 - 811,1983.

American Welding Society, Proceedingsfrom the international Conference on Underwater Welding, March 20-21,1991, New Orleans, LA,1991.

American Welding Society, " Specification for Underwater Welding," AWS D3.6, American Welding Society, Miami, FL 1983 and 1993.

American Society af Mechanical Engineers, Proceedings ofthe 12th International Conference on Ofshore Mechanics andArctic Engineering, Glasgow, Scotland, June 20 - 24,1993, ASME, New York, NY,1993.

American Society of Mechanical Engineers, " Case N 516: Underwater Weiding," Cases of ASME Boiler and Pressure Vessel Code,Section XI, Division 1, August 9,1993 Clark, D.E., C.J. Einerson and R.B. Loop,"Weldability of Rapidly Solidified Type 304 Stainless Steel," Advances in Welding Science and Technology, S.A. David, ed., pp. 817 - 820, 1986.

Findlan, S.J., M.K. Phillips and A.G. Peterson, Jr., Underwater Wet Weldingfor the Repair of Reactor Pressure VesselInternals: Interim Report, NP - 7481. Electric Power Research Institute, Palo Alto, CA, January 23,1992.

Franco-Ferreira, E.A. and W.R. Kanne, Jr.," Remote Reactor Repair: Avoidance of Helium-Induced Cracking Using GMA Welding," Welding Journal, pp. 43 - 52, February 1992.

Fabritsiev, S.A., and J.G. Van der Laan, " Helium Effects on the Reweldability and Low Cycle Fatigue Properties of Welded Joints for Type Cr16NillMo3Ti and 316L(N) Stainless Steels,"

Effects of Radia; ion on Materials: 17th Internationa! Symposium, ASTM STP 1270, David S.

Gelles et al., eds., American Society for Testing and Materials, pp. 980-994,1996.

Fukutani, K., " Analysis Method and Device for Reactor incore Structure," Toshiba Corporation, Kawasaki, Kanagawa, Japan, Patent Document 7-244190/A, September 19,1995.

NUREG-1616 28

- _ . _ _. _ - _ _ _ _ _ _ .. _ . _ _ _ _ _ _ . _ _ .~. _ _ _ .__ _

g t.

t Goods, S.H., and C.W.-Karfs,'" Helium Induced Weld Cracking in Low Heat Input GMA Weld Overlays," Welding Journal, Welding Research Supplement, pp; 123s - 132s, May 1991.

l Goods, S.H., and N.Y.C. Yang, " Microstructural Damage and Residual Mechanical Properties in Helium Bearing Gas Metal Arc Weldments," Metallurgical Transactions J, Vol. 23A, pp.1021 -

l ' 1032, March 1992.

Gmbbs, C.E. " Whitey," " Underwater Wet Welding (A State-of-the-Art Report)," Proceedings ofthe 12th International Coq 6-ence on Ofshore Mechanics and Arctic Engineering, Glasgow, Scotland,

June 20 - 24 1993, American Society of Mechanical Engineers, New York, NY, pp. I11 - 118, 1993.

Grubbs, C.E. " Whitey," " Underwater Wet Welding (A State-of the-Art Report)," Proceedings of Underwater intervention 1994. February 7 - 10,1994, San Diego, CA, Underwater Intervention '94 4

e Conference Committee, Washington, D.C., pp.93-103,1994.

Hall, M.M., Jr., et a!., " Fusion Welding of Irradiated AISI 304L- Stainless Steel Tubing,"

i Weldments: Physical Metallurgy and Failure Phenomena, Proceedings of the 5th Bolton Landing Conference, pp. 365 - 378,1978.

International Institute of Welding, " Underwater Welding", Proceedings of the International Conference held at Trondheim. Norway, June 27 - 28,1983, Pergamon Press, Oxford, Great Brite.in, 1983.

Jenco, J.," Underwater Maintenance Guide - A Guide to Diving and Remotely Operated Vehicle Operations for Nuclear Maintenance PersonncL" EPRI Report NP-7088, December 1990.

Kanne, W.R., Jr., C.L. Angerman, and B.J. Eberhard, "Weldability of Tritium-Charged 304L Stainless Steel, DP-1740, E.I. du Pont de Nemours & Co.,Inc., Savannah River Laboratory, Aiken, SC, February 1987.

Kanne, W.R., Jr., " Remote Reactor Repair: GTA Weld Cracking Caused by Entrapped Helium,"

WeldingJournal, pp. 33 - 39, August 1988.

Kanne,- W.R., Jr., et al., " Weld Repair of Helium Degraded Reactor Vessel Material," Proceedings ofthe Fiftn International Symposium on Environmental Degradation ofMaterials in Nuclear Power Systems - Water Reactors, August 25 - 29,1991, American Nuclear Society, pp. 390 - 402,- 1991.

Kanne, W.R., Jr. et al.t" Welding Irradiated Stainless Steel," Westinghouse Savannah River Co.

- (WSRC) Report MS-93-619,1993a.-

Kanne, W.R., Jr. et al., "Metallographic _ Analysis of Helium-Embrittlement Cracking of Repair Welds in Nuclear Reactor Tanks",' Materials Characterization, Vol. 30, pp. 23 - 34, January 1993b.

'29 NUREG-1616

Kanne, W.R., Jr., and G.T. Chandler, ' Analysis of Welds on Helium-Containing Restraint Test Plates,"WSRC-TR-94-0449, Westinghouse Savannah River Co., Aiken, SC, September 1994.

Kanne, W.R., Jr., G.T. Chandler, D.Z. Nelson, and E.A. Franco-Ferreira, " Welding Irradiated Stainless Steel", Journal ofNuclear Materials, Vol. 225, pp. 69 - 75,1995 .

Lin, H.T., 5.H. Goods. MJ . Grossbeck, and ll.A. Chin, " Helium-Induced Degradation in the Weldability of an Austenitic Stainle::s Steel," Efects ofRad'atlan ofMaterials: 14th Internationa Smposium, Vol. I, ASTM STP 1046, N.H. Packan, R.E. Stoller, and A.S. Kuraar, eds., American Society for Testing and Materials, Philadelphia, PA, pp. 301 - 314,1989.

Lin, H.T., M.L. Grossbeck, and B.A. Chin, " Cavity Microstructure and Kinetics During Gas Tungsten Arc Welding of Hellun,-Containing Stainless Steel," Metallurgical Transactions A, Vol.

21 A, pp. 2585 - 25%, September 1990.

Lin, H.T., and B.A. Chin, "The feasibility of Welding Irradiated Materials," Journal ofNuclear Materiae, Vol. 179 - 181, pp. I131 - 1134,1991a.

Lin, H.T., and B. A. Chin, " Helium-Induced Weld Cracking in Austenitic and Martensitic Steels,"

Journal ofMatenals Sc.'ence, Vol. 26, pp. 2063 - 2070,1991 b.

Mahan, G.D.,"Wr.t Underwater Weld Repair of Feedwater Sparger Pipe," Welding Journal, pp. 26 -

30, January 1990.

Maloney, J.P , et al., " Repair of a Nuclear Reactor Vessel," DP-1199, E.I. du Pont de Nemours &

Co., Inc., Savannan River Laboratory, Aiken, SC, June 1969.

Ogden, D., and T. Joos," Specification Stirs Underwater Electrode Development," Welding Journal, pp. 59 - 61, August 1990.

O'Sullivan, J.E.," Wet Underwater Weld Repair of Susquehanna Unit 1 Steam Dryer," Welding Journal,pp.19 23, June 1988.

O'Sullivan, J.E.," Wet Undenvater Weld Repairs of Susquehanna Reactor Internala", Excellent and Economic Nuclear Plant Performance (Combined ANS Power Division Topical Meeting and ASME .

Nuclear Energy Conference, September 16 - 19,1990), American Swiety of Mechanical Engineers, New York, NY,1990.

Perra,-M.W., ed.," Savannah River Reactor Tank Repair Program Progress Report: August 1,1989 -

February 28,1990," SAND 90-8213, Sandia National Laboratories, Albuquerque, NM, April 1990.

NUREG-1616 30

Reynolds, TJ., D.S. Fogarty, and M.A. Quirin," Peach Bottom Unit 3 Steam Dryer Repaired by Wet Welding," Proceedings offnternational Conference on Underwater Welding: March 20- 21,1991.

New Orleans, la, American Welding Society, Miami, FL, pp. 144 - 157,1991.

Robinson, S.L., " Tensile Properties of Welded llelium Charged 304L Stainless Steel," Journal of Nuclear Materials, Vol. 152, pp. 212 - 2.19,1988.

Schiller, P., et al.," Structure of Helium-Loaded Stainless Steel After Short Heat Pulses," Journal ofNuclear Materials, Vol.148, pp. 224 - 226,1987.

Shah, V.N., and P.E. MacDonald, eds.," Boiling Water Reactor Vessel Intemals," Aging and Life Euension of Major Light Water Reactor Components, Elsesler, New York, NY, pp. 750 - 786,1993.

Smith, R.E., and W.J. Childs, " Development Program for In Vessel Reactor Repair of Stress Corrosion Cracked Components," presented at the SMIRT 11 Post Conference Seminar No. 2:

Assuring StructuralIntegrity ofSteel Reactor Pressure Boundary Components. Taipel, Taiwan.

August 26 - 28,1991., Atomic Energy Society of Japan, Tokyo, Japan,1991.

The Nuclear Professional, pp.16 - 21, Winter 1990.

Tsai, C.L., " Underwater Welding, Cutting and Inspection," Welding Journal, pp. 55 - 62, February 1995.

Underwater Intervention '94 Conference Committee, Proceedingsfrom Underwater Intervention

'94, February 7 - 10,1994, San Diego, CA,1994.

U.S. Nuclear Regulatory Commission, " Inservice Inspection Code Case Acceptability, ASME Section XI Division 1," Regulatory Guide 1.147, Revision 12, May 1997.

Van Osch, E.V., D.3. Hulst, and J.G. Van der Laan, "First Result of Laser Welding of Neutron irradiated Stainless Steel," ECN Report Number ECN RX-94-074,1994.

Wang, C.A., H.T. Lin, M.L. Grossbeck, and B.A. Chin, " Suppression of HAZ Cracking During Welding of Helium-Containing Materials," Journal ofNuclear Materials, Vol.191 - IM, pp. 6% -

700,1992.

Wang, C.A., M.L. Grossbeck, and B.A. Chin, " Technique to Eliminate Helium-Induced Weld

- Cracking in Stainless Steels," Efects ofRadiation on Materials: 16th International Symposium, ASTM _ STP 1175, Kumar, Gelles, Narutad, and Little, eds., American Society for Testing and Materials, pp. 779 - 788,1993.

31 NUREG-1616

Wang, C.A., M.L. Grosslx:ck, and B.A. Chin, " Threshold liclium Concentration Required to initiate Cracking During Welding ofIrradiated Stainless Steel,".lournalofNuclear Materials, Vol. 225, pp.

59 - 68,1995.

Wang, C.A., M.L Grossbeck, H. Aglan, and B.A. Chin, "The Effect of an Applied Stress on the Welding ofirradiated Steels," Journal ofNuclear Materials, Vol. 239, pp. 85 - 89,1996.

Watanabe, K., S. Jitsukawa, S. Hamada, T. Kodaira, and A. Hishinuma, "Weldability of Neutron-Irradiate.1 Type 316 Stainless Steel," Fusion Engineering and Design, Vol. 31, Issue 1, pp. 9 - 15, April 1996.

White, R.A., R. Fusaro, M.G. Jones, R.R. Milian-Rodriguez and H.D. Solomon, " Underwater Cladding with Laser Beam and Plasma Arc Welding," Welding Journal, pp. 57-61, January 1997.

NUREG-1616 32 s

NaC Fosud 33s u.4. NuCEEAR REGULATORY cC ** 1. REPORT NL ABER

!',,*J ,,a sammeneo, Nac. Aas ves, swee, nov, l " " * " - p BIBLIOGRAPHIC DATA SHEET (sw meuceane sa n. ,e wi f

a. nTLt AND Su6 TITLE NUREG 1616 s Feesibility of Underwater We'. ding of Highly irradiated in-Vessel Components of Bolling Water 3,

Reactors: DATE REPORT PueUSMID

~

A Literature Review wowTH vtAR l

November 1997  :,

4. FIN OR GRANT NuMEiiR s AUTHOR (5) e TYPE OF REPORT A. L Lund Technical

r. PERIOo COVERED (lacww Desse) e PERFORMWG ORoANLZATiON . NAME AND AooRESS (r wec. poisse Dween, omas or Repon. u S. Nucsoar Amedesary conweseat, and meene einsees a contrecsor Mowe no"w and meent 00*ees1 Division of Engineering Technology Office of Nuclear Regulatory Research U.S. Nuclear Regulatory Commission Washington. D.C. 20555 0001

s. SPONSORdl0 ORGAPLZATION . NAME Aho ADORE SS iv NRC type 'seme es snow; ecoescsor, provsde N8tc Dweat. once or aspan, u 3 Nucuser Repusosary comanesort and meene ed*eeei Serve ac above

10. SUPPLEMENTARY NOTE 5

11. A8sTRACT r2co wones or *v in February 1997, the U. S. Nue' at Re ulatory Commission (NRC), Office of Nucice Regulatory Research (RES), initiated a literate o reviere to assess the E .te of u;nderwater welding technology. The objectiv this Irterature review was to evaluate the viability of underwater welding in-vessel components of boiling water reactor (BW. ., in-vessel components, especially those components fabricated from stainless steels that are subjected to hoh not. tron fluences. This literature review revealed a preponderance of general 'nformation about underwater welding technology, as a result of the active research in this field sponsored by the U. S. Navy and t'ffshore oil and gas industry corums. However, the literature search yielded only a limibd rm.iount of information about underwater welding of components in low-fluence areas of BWR in-vessel environments, and no information at all concoming underwate' welding experuces in high-fluence environments.

Research reported by the staff of the U. S. Department of Energy (DOE) Savannah River Site and researchers from the intomational fusion reactor regram documentM relevant exponence conceming welding of stainless steel materials in air omrironments exposed to h neutron fluences. It also addressed problems with welding highly irradiated r;.aterials, primarily helium-induced cracking in material, and suggested some solutions to these problems.

12. KEY WORDS/DESCRIP TORS ruer wense or parsess met est esser ressereners in mesang me soport ) 4 AVALA5W EWEMENT BWR Type Reactors, Underwater Welding Highly irradiated, Stainless Steel, Irt Vessel Components _

unlimited 14 4ECURITY CLAS$1 FICA > ON (Thre PeQel

% unclassified inn nesom unclassified MER OF PAGES

16. price MRC F0 mas 336 (240) Ttwo form wee esecawuceep produced try Ette Federal Forms, int I

l t

i..

Printed .

on recycled  :

paper ,

Federal Recycling Program a

.. j NUREG-1616 FEASIBILITY OF UNDERWATER WELDING OF IllGIILY IRRADIATED NOVEMEER 1997 IN-VESSEL COMPONENTS OF BOILING-WATER REACTORS

- FIRST CLASS MAIL UNITED STATES POSTAGE AND FEES PAID NUCLEAR REGULATORY COMMISSION WASHINGTON, DC 20555-0001 PERMIT NO G47 l

I OFFICIAL BUSINESS l PENALLY FOR PR!VATE USE 5300 l

120555154486 1 1ANIRilR5 US NAC-OIRM TPS-PDR-t,UREG 2WFN-6E7 WASHINGTON DC 20555 a

e

_ _ _ _