ML053070384

Degradation of Reactor Coolant System Pressure Boundary Resulting from Boric Acid Corrosion

ML053070384
Person / Time
Site:	Davis Besse
Issue date:	12/29/1986
From:	Jordan E NRC/IE
To:
References
SSINS 6835 IN-86-108, NUDOCS 8612230091
Download: ML053070384 (4)

Text

SSINS No.: 6835 RECEIVED IN 86-108 1987 UNITED STATES JC%N NUCLEAR REGULATORY COMMISSION TOLFIX, W-wM OFFICE OF INSPECTION AND ENFORCEMENT WASHINGTON, 0. C. 20555 December 29, 1986 -

I INFORMATION NOTICE NO. 86-108: DEGRADATION OF REACTOR COOLANT SYSTEM PRESSURE BOUNDARY RESULTING FROM BORIC ACID CORROSION

> Addressees:

1 A1 1 pressurized water reactor (PWR) faci 1i t i es holding an operating 1icense or a construction permit.

Purpose:

This notice is t o alert recipients of a severe instance o f boric acid induced corrosion of ferritic steel components i n the reactor coolant system (RCS).

,Recipients are expected t o review the information for applicability t o their facilities and consider actions, i f appropriate, to preclude similar problems occurring a t their facilities. However, suggestions contained i n this information notice do not constitute NRC requirements; therefore, no specific action or written response i s required.

Description of Circumstances:

I n October 1986, the Arkansas Nuclear One, U n i t 1 (ANO-1) Plant was i n cold shutdown and was performing nondestructive testing of the high ressure injec-P tion (HPI) nozzle thermal sleeves. An HPI nozzle is attached d rectly on the side of each of the four RCS cold legs. The metallic insulation was removed from the "A1' HPI nozzle t o allow radiographic examination. Removal o f this insulation revealed severe corrosion wastage on the exterior of the HPI .nozzle and some wastage on the RCS cold leg pipe. The corrosion apparently was caused by reactor coolant leakage from an HPT isolation valve located about 8 feet above the nozzle as shown i n the attached Figure 1.

The wastage began adjacent t o where the 3-1/2 inch OD stainless steel safe-end i s welded t o the carbon steel HPI nozzle. The safe-end is located between the stainless steel HPI line and the carbon steel HPI nozzle. The wastage was approximately 1/2 inch a t i t s deepest location (adjacent t o the stainless-to-carbon steel weld). The HPI nozzle (including cladding) is approximately 3/4 inch thick a t this point. A t the transition weld between the safe-end and the carbon steel nozzle, the wastage extended approximately 20 percent around the circumference of the nozzle i n the form of several trenches. From this point, the wastage narrowed t o two separate trenches that became shallower as they progressed more.than 10 inches along the bottom of the HPI nozzle towards the RCS cold leg. The two trenches then continued down the cold leg for approx-imately 6 inches. The depth of t h e trenches on the cold leg were less than V4 inch.

8612230091

I N 86-108 D e c e e ? 29, 1986 Page 2 o f 5 - .

The HPI nozzle i s constructed o f f e r r i t i c (ASTM, A-105, grade 2) s t e 1, The c o l d l e g also i s constructed o f f e r r i t i c (ASTM, A-106, grade C) s t e e b the' HPI nozzle and the cold l e g are clad on the inside with stainless steel o f 3/16 -

- i n c h nominal thickness.

Leakage from the HPI i s o l a t i o n valve was f i r s t noted i n August 1985, through RCS leak detection methods. The measured leakage was approximately 0.08 gal-lons per minute (gpm). The leakage was a t t r i b u t e d t o a valve body-to-bonnet leak. The valve's seal r i n g and yoke clamp were replaced i n September 1985.

A leakage of 0.09 gpm was again detected from t h i s valve 8 days l a t e r during p l a n t startup. This leak r a t e continued . u n t i l subsequent repair o f the valve i n February 1986. The insulation was not removed a t the t i m e of these repairs.

After the damaged HPI nozzle was discovered, a reddish stain, resulting from the leaching out o f i r o n oxide corrosion products, was found on the exterior o f the insulation near the damaged area.

D1scussion:

There have been a number o f reported incidents o f b o r i c acid corrosion wastage o f f e r r i t i c steels. I n 1981 Calvert C l i f f s , Unit 2, experienced boric acid

- corrosion wastage on an RCS cold l e g near the suction t o a reactor coolant pump (RCP). This corrosion wastage was from lf8 t o l/4 inch i n depth and extended about 20 percent around the circumference o f the RCS pipe. This RCS piping i s fabricated from f e r r i t i c (ASME, SA 516, grade 70) steel.

Most incidents, however, have been wastage o f threaded fasteners. I n June 1982,

. IE B u l l e t i n 82-02, "Degradation o f Threaded Fasteners I n The Reactor Coolant' Pressure Boundary OF WR Plants," was issued. The closeout o f t h i s b u l l e t i n was addressed i n NUREG-1095, May 1985. The affected threaded fasteners were of a low alloy, hfgh strength, f e r r i t i c steel. A generic issue, "Bolting Degradation o r Failures i n Nuclear Power Plants, i s currently under review by the NRC s t a f f t o determine i f additional actions are necessary. One of the main concerns i n t h i s issue i s boric acid corrosion. The 1983 e d i t i o n o f the ASME Code,Section XI, was revised t o provide f o r more r e s t r i c t i v e requirements for visual examinations o f systems containing borated water. Part of these require-ments i s an inspection o f insulation a t the j o i n t s for evidence of leaks. This r e v i sion i s contai ned in Section IWA-S242(a) , Insul ated Components Boric acid corrosion has been found t o be most active where the metal surface i s cool enough so t h a t i t i s wetted. Ifthe metal i s s u f f i c i e n t l y hot, then the surface w i l l stay dry and t h i s loss of e l e c t r o l y t e w i l l slow the corrosion rate. A t ANO-1, borated water leaked from the HPI i s o l a t i o n valve i n the form o f a l i q u i d and then ran down the HPI piping on the inside o f the insulation t o t h e HPI nozzle. As the leakage approached the c o l d leg, the increased piping temperatures caused evaporation of the water, thus increasin the boric acid concentration and lowering the PH o f the solution. It i s be ieved t h a t theB close tolerance between the H P I nozzle and the insulation, aided by boric acid c r y s t a l l i z a t i o n , caused pooling of the solution a t the nozzle. This pool of highly acidic solution wetted the nozzle and resulted i n accelerated corrosive

0 0 Q 0 0 4 0 I 9 . : i ;

IN 86-108 December 29, 1986 Page 3 of 3 attack. Experience has shown that even relatively hot metal can be sufficiently cooled on the surface by the flow of the leaka e SO that the surface stays P

wetted and boric acid corrosion is promoted. n addition, periods during which a metal surface is below normal operating temperature may allow corrosion i n -

areas that would not otherwise be expected. Boric acid corrosion rates i n excess of 1 inch depth per year i n f e r r i t i c steels have been experienced i n plants and duplicated i n laboratory tests where low quality steam from borated reactor coolant impinged upon a surface and kept i t wetted.

Additional information is contained i n EPRI-NP-3784, "A Survey of the Literature on Low Alloy Steel Fastener Corrosion i n PWR Power Plants," December 1984, and NUREG/CR-2827, "Boric Acid Corrosion of Ferritic Reactor Components , I t July 1982.

Fol1owue:

The damaged HPI nozzle has been repaired by grinding out all indications of corrosion and rebuilding by welding i n those areas w i t h less than the minimum required wall thickness. Repair t o the cold leg required only grinding out the corrosion. A l l repairs were i n accordance w i t h ASME codes. The other HPI nozzles were inspected and no evidence of corroston wastage was found. The 1icensee continues t o evaluate methods and procedures t o minimize recurrence of t h i s type of event. The primary defense i s t o minimize leaks, detect and stop leaks soon after they start, and promptly clean up any boric acid residue.

Detection o f leaks will be enhanced by an evaluation of any iron oxide stains on insulation.

No specific action o r written response is required by t h i s information. notice.

If you have questions about t h i s matter, please contact the Regional Admini-strator. o f the appropriate NRC regional office or t h i s office.

Divisib(; of, Qnergency Preparedness and Engineering Response Office of Inspection and Enforcement Technical

Contact:

Hen A. Bailey, IE

( 3 0 8 492-9006 Attachments:

1. Figure 1: ANO-1 H P I Line/Nozzle Configuration

2. List of Recently Issued IE Information Notices

Attachment 1 IN 86-108

. December 29, 1986 .

E 0

2 U

\

0, E

I-4 3 f3 v)

W cn