Read 8-tech-assessment-rpv.pdf text version

Technical Assessment of Davis-Besse Degradation

Prepared for Meeting With NRC Technical Staff May 22, 2002 Prepared by: G. White C. Marks S. Hunt Dominion Engineering, Inc.

Contents

Purpose and Approach Material Loss Mechanisms

· · · Corrosion mechanisms Erosion mechanisms Flow accelerated corrosion

Degradation Progression Boric Acid Corrosion Tests Simulating Nozzle Leakage Thermal-Hydraulic Environment

· · · Leak rate Expansion cooling Velocity and wall shear stress Volume of boric acid deposits produced Boric acid morphology and properties Concentration of primary water pH Electrochemistry

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002 2

Chemical Environment

· · · · ·

Purpose and Approach

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002

3

Purpose

The purpose of the technical assessments is to complement plant experience in answering the following questions:

· If a significant amount of RPV head material loss occurs, will it be detectable visually from above the head (either directly or through the presence of deposits)? · Could significant material loss occur during a single cycle?

In addition, the technical assessments also address current questions regarding the progression of material loss mechanisms (i.e., understanding of degradation progression)

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002

4

Approach

The basic approach is to examine how the various potential material loss mechanisms vary as the leak rate is increased from 10-6 to 1.0 gpm and the initial tight nozzle annulus becomes a large cavity through material loss. Evaluations focus on:

· · · · Thermal-hydraulic environment Chemical environment Properties of boric acid and boron compounds Relevant experimental results and plant experience

The leak rate is expected to be the key parameter:

· Expansion cooling increases with leak rate, potentially permitting a liquid film to reach the top head surface · Increasing leak rates result in higher velocities and potentially erosion or flow accelerated corrosion

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002 5

Approach (continued)

The leak rate also determines the amount of boric acid deposits that exit the pressure boundary The results of corrosion and erosion rate evaluations are used to bound:

· The timeframe for significant degradation · The volume of low alloy steel material loss versus the volume of deposits produced

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002

6

Material Loss Mechanisms

· · · Corrosion mechanisms Erosion mechanisms Flow accelerated corrosion

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002

7

Material Loss Mechanisms Overview

Chemical Mechanisms

· Low-oxygen, boric acid corrosion (deaerated, concentrated boric acid solutions) · Dry boric acid or boric oxide crystal corrosion · Classic crevice corrosion (conductive liquid in the crevice forms an ionic path to allow dissolution deep in crevice remote from oxygen at crevice mouth) · Galvanic corrosion (driving corrosion potential due to dissimilar metal couple between Alloy 600 nozzle and low-alloy-steel (LAS) head) · "Classic" boric acid corrosion (aerated, concentrated boric acid solutions) · Molten boric acid corrosion

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002

8

Material Loss Mechanisms Overview (continued)

Flow-Enhanced Chemical Mechanisms

· Two-phase flow accelerated corrosion (FAC) (low oxygen; boric acid not required)

Mechanical Mechanisms

· · · · Droplet or solid particle impingement erosion Flashing-induced erosion Steam cutting erosion Single-phase erosion

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002

9

Material Loss Mechanisms Matrix

PRELIMINARY

Deaerated Boric Acid Corrosion

Conc. Boric Acid Corrosion but DO2 0-10 ppb

Extent of Wastage Initial Tight Annulus Enlarged Annulus Small Cavity Low rates Low rates Possible for high Less likely than for tight annulus leak rates Possible if liquid velocities high enough and temperature low enough

Large flow area precludes high velocities

Large Cavity

Dry BA or Boric Oxide Crystal Corrosion Possible Material Loss Mechanisms

Corrosion in Contact with Dry Crystals and Humidity

Single-Phase Erosion

Potential Erosion if High Steam Velocities

Flow Accelerated Corrosion (FAC)

Low-Oxygen Dissolution through Surface Oxides

Unlikely as oxygen stabilizes

Impingement / Flashing-Induced Erosion

Droplet and Particle Impact Opposite Crack Outlet

Possible if droplets right size and momentum Believed not to be likely because low alloy steel does not passivate in an aerated, concentrated boric acid

Not possible because no crevice geometry

Crevice Corrosion

Liquid Ionic Path from Top Head Surface

"Occluded Region" Galvanic Corrosion

Driven by Potential Difference Btw Dissimilar Metals

Possible at locations where liquid solution exists Possible but rate expected to be lower than for aerated BAC

Not possible due to low oxygen deep in crevice

"Molten" Boric Acid Corrosion

Corrosion in Pure or Nearly Pure Melted BA Crystals

Aerated Boric Acid Corrosion (BAC)

Concentrated Boric Acid Solution with Oxygen

Unlikely

Possibly

Up to 1-5 inches per year

10

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002

Chemical Mechanisms Classic Crevice Corrosion

Source: F.P. IJsseling. Survey of Literature on Crevice Corrosion (1979-1998), IOM Communications Ltd., London, 2000.

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002

11

Flow Accelerated Corrosion Effect of Velocity on FAC Rate

Source: B. Vyas, Treatise on Materials Science and Technology, vol. 16, 1979, p. 357.

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002 12

Flow Accelerated Corrosion Time Dependencies of FAC Processes

Stagnant Liquid

Low Velocities

Velocity > Breakaway

High Velocities (pure erosion)

Source: B. Chexal, et al., Flow-Accelerated Corrosion in Power Plants, TR-106611, EPRI, Palo Alto, CA, 1996.

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002

13

Flow Accelerated Corrosion Effect of Temperature for Two-Phase Flows

Temperature Dependence of Two-Phase FAC From Keller, H., VGB Kraftwerkstechnik, 54, (1974), p. 292.

Temperature Dependence of Two-Phase FAC with a Steam Quality of 65% and a Velocity of 185 ft/s From Bouchacourt, M., EDF Internal Report, (1982), Ref.: HT-PVD. XXX MAT/T.42.

Temperature Dependence of Two-Phase FAC From Izumiya, M., Water Chemistry and Corrosion Products in Nuclear Power Plants, International Atomic Energy Agency, Vienna (1983), p. 61.

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002

14

Flow Accelerated Corrosion Effect of Alloy Content on Erosion / Corrosion Rate

1.200 Relative Erosion/Corrosion Rate 1.000 0.800 0.600 Mo 0.400 0.200 Cr 0.000 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 Material Content (%)

Source: EPRI CHECWORKS

Cu

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002

15

Flow Accelerated Corrosion EPRI CHECWORKS FAC Predictions

Maximum Material Loss Rate (in/yr)

Predictions for saturated twophase water flow through a 2-inch Sch 80 90° elbow with R/D = 1.5 No Cr assumed but 0.5% Mo Dissolved O2 = 0 pHRT = 7

1.4

T = 212°F, x = 0.446

1.2 1.0 0.8 0.6 0.4 0.2 0.0 0.1 1. Max rate at 0.1 ft/s is ~0.02 inches/year (250°F)

T = 250°F, x = 0.425 T = 300°F, x = 0.383 T = 350°F, x = 0.343 T = 375°F, x = 0.321 T = 400°F, x = 0.296 T = 600°F, x = 0.007

10.

100.

Liquid Velocity (ft/s)

NOTE: CHECWORKS is intended to be used to model FAC in secondary cycle piping systems and not in situations such as leaking crevices. These calculations show the rough effects of liquid velocity and temperature that may be expected for leaking CRDM nozzles.

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002

16

Degradation Progression

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002

17

Degradation Progression Leak Rate is Main Controlling Parameter

Increasing Leak Rate

likely crack in nozzle wall reaches above top of weld on OD and ID Nozzle/Weld Condition leak path to annulus leak path to annulus leak path to annulus

PRELIMINARY

crack in nozzle wall reaches relatively high above top of weld on OD and ID

Annulus Condition

hypothetical clogged annulus

possibly some opening up of annulus

likely some opening up of annulus

may require some opening up of annulus

likely requires some opening up of annulus

Leak Rate

hypothetical zero leak rate

< on the order of 0.001 gpm

roughly between 0.001 and 0.01 gpm

roughly between 0.01 and 0.1 gpm

> on the order of 0.1 gpm

Liquid Velocity Exiting Crack

0 ft/s

less than on the order of 0.01 ft/s

roughly between 0.01 and 0.1 ft/s

roughly between 0.1 and 1 ft/s

> on the order of 1 ft/s

Local Temperature

600°F

Close to 600°F

at least roughly 500°F

roughly between 212 and 500°F

Close to 212°F

Liquid Location

fills annulus up to hypothetical blockage

all liquid vaporizes close to bottom of annulus

liquid film unlikely to exist high in annulus

liquid film may cover much of annulus walls

liquid film covers local top surface of head

Possible Significant Mechanisms

· none

· possibly very

minor galvanic

· possibly some

galvanic corrosion; · erosion until annulus opens slightly

· likely some galvanic

· aerated BAC on top of corrosion; head; · minor erosion and FAC; · possibly molten BAC, · possibly aerated BAC if galvanic corrosion, erosion, annulus is opened enough or FAC

roughly 70 to 700 lbs > on the order of 700 lbs

Pounds of Boric Acid Deposits Released in 2 years

at least small amount extruded

< on the order of 7 lbs all or most other leaking CRDM nozzles

roughly 7 to 70 lbs

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002 EPRI & CE Annulus Tests Davis-Besse Nozzle #3

18

Degradation Progression

Condition 1a. If--contrary to plant experience--a leak path crack forms in the absence of leakage to the top surface of the head

· There will be low oxygen, zero velocity, and no vaporization-driven concentration mechanism, so material loss rates will be small

Condition 1b. For tight nozzle cracks that allow a leak path

· The leak rate will be limited and the annulus downstream of the crack will boil dry within a short distance · Erosion and FAC will not be active due to very low liquid velocities · Small amounts of boric acid or boric oxide crystals will accumulate on the top head surface

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002

19

Degradation Progression (continued)

Condition 2. As the crack widens and the minimum leak path flow area increases

· Flashing-induced erosion or FAC may initiate the material loss process · Galvanic corrosion may be important if cooling is sufficient to allow liquid to exist over a significant height in the annulus · These mechanisms could be expected to produce greater relative material loss deep in the annulus, consistent with Davis-Besse Nozzle #2 and the EPRI BAC leaking annulus tests

Condition 3. As the leak rate increases and the wastage area grows from a small cavity to a large, open cavity

· Aerated boric acid corrosion (up to 1-5 inches per year) may occur

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002

20

Degradation Progression (continued)

The geometry of the Davis-Besse Nozzle #3 cavity may indicate that aerated BAC removing material from the top surface down toward the cladding replaced corrosion and/or erosion deep down in the annulus as the dominant degradation mode

· The slope of the walls of the cavity change with distance from the top head surface · Heat transfer calculations show considerable local cooling of the head for the range of leak rates believed to apply to this nozzle, indicating an aerated, concentrated liquid boric acid solution film on the top head surface adjacent to this nozzle · Laboratory tests and plant experience indicate relatively high corrosion rates for low alloy steel exposed to aerated, concentrated liquid boric acid solution in comparison to other material loss mechanisms · Gravity-driven flow of this liquid film would tend to produce the observed oblong shape of the Nozzle #3 cavity

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002 21

Degradation Progression Relating Linear Loss Rate to Volume Loss (Example Calcs)

40

W

Material Loss in 2 years

co

35

r a corrosion direction L

corrosion d ire

30 Volume Corroded (in3)

Wall and Top Attack (3 inches past the nozzle, radially) Maximum Rate Observed in Deaerated Solutions 0.072 in/year (EPRI BAC Guidebook) Angled Attack

25

20

15

r a corrosion direction L corrosion dire

10 Bore Wall Attack 5

r a corrosion direction L corrosion d

Material loss assumed to occur over a 45° (1.57 inch) arc of the nozzle circumference

0 0 0.5 1 Corrosion Rate (inches/year)

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002 22

1.5

2

Boric Acid Corrosion Tests Simulating Nozzle Leakage

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002

23

BAC Tests Simulating Nozzle Leakage Overview

Aerated Borated Water

An extensive set of experimental data has been compiled and reported in the EPRI Boric Acid Corrosion Guidebook, Revision 1

· · Tests by several organizations prior to 1995 Tests of a range of conditions

Deaerated water Aerated water Dripping Impingement Leakage into annulus

Borated Water Steel Part or Defect in Clad Surface

Steel Part in Contact With Borated Water

Immersion in Deaerated Water

Immersion in Aerated Water

Dripping Borated Water Boric Acid Crystals Borated Water Impinging On Hot Metal Surface

Increasing Concentration and Corrosion

Dripping onto Hot Metal Surfaces

Impingement onto Hot Metal Surfaces

·

Tests performed by EPRI at Southwest Research Institute in 1996/97

Crack

Results of additional tests performed by CEA in France have been made available to EPRI

Corrosion at Point Where Leakage Exits Annulus

Leakage into Annular Gap

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002

24

BAC Tests Simulating Nozzle Leakage EPRI Annulus Test Matrix

Test Number 4a 4b 5a 5b 6a 6b

Temperature (F) 600 600 600 600 600 600

Flow Rate (gpm) 0.01 0.10 0.01 0.10 0.01 0.10

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002

25

BAC Tests Simulating Nozzle Leakage Typical Sectioned EPRI Test Specimen

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002

26

BAC Tests Simulating Nozzle Leakage Test Conclusions

The maximum corrosion rates in both the EPRI and CE tests were about 2.0 ­ 2.5 in/yr The maximum corrosion rates occurred at leak rates of about 0.01 gpm with decreasing corrosion rate as leak rate was increased above 0.01 gpm

· However, one test by CE at a low leak rate (0.002 gpm) showed a very low corrosion rate

While the tests may not represent the initial conditions of a very tight fit, they are considered to represent anticipated conditions once the annulus opens up to about 0.005" While the corrosion depth can be greater below the exposed surface than at the surface, the tests showed relatively large amounts of boric acid deposits for the range of flow rates tested

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002 27

Thermal-Hydraulic Environment

· · · Leak rate Expansion cooling Velocity and wall shear stress

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002

28

Leak Rate Calculations Method

Calculate axial crack length and opening area above the top of the weld using welding residual stress FEA or an available analytical expression from fracture mechanics Calculate the leak rate based on industry correlations for choked flow through a crack in a steam generator tube Consider the potential additional flow resistance of a tight annulus downstream of the crack

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002

29

Leak Rate Calculations Crack Opening Displacement and Area

Crack opening displacement and area determined using finite element models with welding residual and operating stresses

Head Material Removed 90 Around Nozzle From Symmetry Plane

1

Nozzle Midplane (90 From Downhill)

Symmetry Displacement Restraints

U

2401

Top of Weld

Nodes Spaced Axially at 0.125"

1.25"

Axial Crack Half Width

1401 Weld Top

Crack Bottom

Axial Cracking Region

DB CRDM(8d,48.5k,4/2.765,)- Ax. Crack to 1.25 in. Above Weld

101 1 5

105

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002

30

Leak Rate Calculations Effect of Actual Crack Front Profile

Crack opening displacement calculations have assumed crack cuts completely through the nozzle wall, and J-groove weld, from the reported crack bottom to top Subsequent to initial leak rate calculations, the actual crack profiles at Davis-Besse have been determined from top-down UT data

Crack Profile for Nozzle 3, Flaw #1 - Downhill

0.6 Crack Depth from OD 0.5 0.4 0.3 0.2 0.1 0 26.0 26.5 27.0 27.5 28.0 28.5 29.0 29.5 30.0 30.5 Crack Elevation 31.0

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002

31

Leak Rate Calculations Typical Results

Actual unidentified leak rate is bounded by leak rates calculated using

· Crack opening area for a through-wall axial crack in a pipe with length equal to the length that the axial crack extends above the top of the Jgroove weld Crack opening area determined using the finite element method for an ideal through-wall crack

ANSYS Model -Head Material Intact Zahoor Analytical Model 10

ANSYS Model - Head Material Corroded Davis-Besse Nozzle N-3

1

Leak Rate (gpm)

0.1

0.01

·

0.001

Calculations show leak rate increases quickly with crack length above the top of the Jgroove weld

0.0001 0.00

0.20

0.40

0.60

0.80

1.00

1.20

1.40

1.60

Crack Length Above Weld (inches)

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002

32

Leak Rate Variation with Crack Length

Radial Annular Gap of: 0.40

Calculation methodology per EPRI NP-6864-L Rev. 2 August 1993 Curves for annulus gaps of 0.005" and larger lie on top of the infinite gap curve

Infinite 0.001" (360°) 0.003" (43°)

0.30 Leak Rate (gpm)

Infinite, corroded head

0.20

0.10

Est. for Davis-Besse Nozzle #2 0.0003" (360°)

0.0001" (360°)

0.00 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 Through-Wall Crack Length Above Weld (in)

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002 33

Leak Rate Variation with Annular Gap Width

0.40 All curves shown for a 1.5" long annulus (43°) 0.30 Leak Rate (gpm)

1.25" crack

1.00" crack

0.20

0.75" crack

0.10

0.50" crack

0.25" crack

0.00 0.0001 0.0010 0.0100 Radial Annular Gap (in)

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002 34

0.1000

1.0000

Expansion Cooling Modeling Overview

Approach is to determine extent of cooling along the leak path as a function of leak rate using

· Heat required to vaporize all leaking liquid is the leak rate times the enthalpy increase (from primary water at 613 Btu/lb to saturated steam at atmospheric pressure at 1150 Btu/lb) · FEA heat transfer model of conduction within head materials with convection boundary conditions from primary coolant and to space above · Correlations for two-phase and single-phase heat transfer coefficients along the leak path

Extent of cooling affects important parameters including

· Location of concentrated liquid · pH · FAC susceptibility

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002 35

Expansion Cooling Modeling Magnitude of Heat Sink

1,000,000. 100,000. Magnitude of Heat Sink (Btu/h) 10,000. 1,000. 100. 10. 75% outlet steam quality 1. 50% outlet steam quality 0.1 0.000001 0.00001 0.0001 0.001 Leak Rate (gpm)

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002 36

NOTE: Inlet state assumed to be subcooled water at 600°F and 2250 psia and outlet assumed at 1 atm.

600°F superheated steam 1 gallon per year = 1.9E-6 gpm saturated steam at outlet

0.01

0.1

1

Expansion Cooling Modeling Finite Element Analysis of Head Heat Transfer

ANSYS 5.7 APR 2 2002 11:34:11 PLOT NO. 1 ELEMENTS MAT NUM XV =-1.573 YV =.700326 ZV =-.189006 *DIST=16.541 *XF =107.147 *YF =32.286 A-ZS=6.591 PRECISE HIDDEN

FEA Model

Y X

Z

Y X Z

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002

37

Expansion Cooling Modeling Finite Element Analysis of Head Heat Transfer

ANSYS 5.7 APR 2 2002 12:00:37 PLOT NO. 3 ELEMENTS HGEN RATES QMIN=-.568E-05 QMAX=0 XV =-1.573 YV =.700326 ZV =-.189006 *DIST=16.541 *XF =107.147 *YF =32.286 A-ZS=6.591 PRECISE HIDDEN -.568E-05 -.504E-05 -.441E-05 -.378E-05 -.315E-05 Y -.252E-05 X Z -.189E-05 -.126E-05 -.631E-06 0 Davis Besse Nozzle 2 Leak - 22.5 deg - 2.5832E-05 BTU/s heat removal

Uniform Surface Heat Sink Along the Leak Path Assumed

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002

38

Expansion Cooling Modeling Finite Element Analysis of Head Heat Transfer

MN

Y X Z

TEMP SMN =603.37 SMX =604.996 603.37 603.551 603.731 603.912 604.093 604.273 604.454 604.635 604.815 604.996

MX

MN

Example Calculation for Low Leak Rate (18.6 Btu/h Heat Sink: complete vaporization of 7×10-5 gpm leak)

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002 39

Expansion Cooling Modeling Finite Element Analysis of Head Heat Transfer

Y X

MN

Z

TEMP SMN =514.122 SMX =604.939 514.122 524.212 534.303 544.394 554.485 564.576 574.667 584.758 594.849 604.939

MX

MN

Example Calculation for Moderate Leak Rate (1860 Btu/h Heat Sink: complete vaporization of 0.007 gpm leak)

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002

40

Expansion Cooling Modeling Finite Element Analysis of Head Heat Transfer

650

Average Metal Surface Temperature Along the Leak Path

Average Metal Temperature on Surface of Cavity, T (°F)

45 Deg Half Arc (h = 600)

600

45 Deg Half Arc (h =110) 22.5 Deg Half Arc (h = 600) 22.5 Deg Half Arc (h = 110)

550

500

T = -0.02537Q + 604.55678 Sink imposed on total 90° arc surface 2 h on inside head = 600 Btu/h-ft -°F

450

400

T = -0.02670Q + 604.55110 Sink imposed on total 90° arc surface 2 h on inside head = 110 Btu/h-ft -°F T = -0.03505Q + 604.54874 Sink imposed on total 45° arc surface 2 h on inside head = 600 Btu/h-ft -°F T = -0.03647Q + 604.39390 Sink imposed on total 45° arc surface 2 h on inside head = 110 Btu/h-ft -°F

350

300

250

200 0 2,000 4,000 6,000 8,000 10,000 12,000 14,000 16,000 Magnitude of Heat Sink, Q (Btu/hr)

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002

41

Effluent Velocity Average Velocities Up Through a 1.5-inch Wide Cavity

Average Upward Flow Velocity (ft/s)

Calculations for a uniform cavity with 1.5-inch circumferential extent

100,000. 10,000. Speed of sound in saturated steam at 1 atm = 1549 ft/s 1,000. 100. 10. 1. 0.1 0.01 0.001 0.0001 0.000001 0.00001 0.0001 0.001 Leak Rate (gpm) 0.01 0.1 1 NOTE: Assuming leak flow through a 1.5" wide cavity.

steam at 1 atm; 0.025" radial gap steam at 1 atm; 0.075" radial gap steam at 1 atm; 0.150" radial gap steam at 1 atm; 0.250" radial gap primary liquid; 0.025" radial gap primary liquid; 0.075" radial gap primary liquid; 0.150" radial gap primary liquid; 0.250" radial gap x=44.6%; 1 atm; 0.025" radial gap x=44.6%; 1 atm; 0.075" radial gap x=44.6%; 1 atm; 0.150" radial gap x=44.6%; 1 atm; 0.250" radial gap

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002

42

Effluent Velocity Single-Phase Steam Critical (Choked) Velocity

Critical Leak Rate Resulting in a Steam Velocity at the Speed of Sound (gpm)

Figure shows the gap size resulting in sonic steam velocities at the annulus/cavity exit for

· · · 360° uniform annulus 3-inch wide cavity 1.5-inch wide cavity

100. 360° annulus with inside diameter of 4.0"

NOTE: Steam velocity calculated based on the density of saturated steam at atmospheric pressure (212°F).

10.

1.

0.1 Cavity with circumferential extent of 3.00" (85.9°) on outside of 4.0" diameter nozzle Cavity with circumferential extent of 1.50" (43.0°) on outside of 4.0" diameter nozzle 0.001 0.01 0.1 1.

0.01

0.001

0.0001 0.0001 Radial Gap of the Annulus or Cavity (inches)

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002

43

Effluent Velocity Liquid Velocity Estimates at Exit of Crack

100 HEM Model Moody Model Fauske Model Calculated liquid velocities based on a flow area equal to 1.8 in2 based on a crack length above the weld of 1.2 inches and a crevice width of 1.5 inches.

10

Liquid Velocity (ft/s)

1

0.1

0.01

0.001

Two-phase water flow at atmospheric pressure and 212°F with enthalpy that of primary water (quality = 44.6%)

0.0001 0.000001

0.00001

0.0001

0.001

0.01

0.1

1

Leak Rate (gpm)

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002 44

Wall Shear Stress Calculation

(Single-Phase Steam,1.25-inch Crack Length Above Top of Weld)

Annular Gap (mils) 0.00 0.90 Bounding Shear Stress in Annulus (psi) 0.80 0.70 0.60 0.50 0.40 0.30

85° 360° Single-phase steam at : ­ 600°F ­ 15 psi ­ 1946 ft/s (sonic)

0.18

0.35

0.53

0.70

0.88

1.06

Annulus Covering:

PRELIMINARY

0.20 0.10 0.00 0.000

43°

0.005

0.010

0.015

0.020

0.025

0.030

Leak Rate at Critical Steam Velocity (gpm)

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002 45

Single-Phase Erosion in Steam Experimental Data

Data available from testing of turbine materials in 1950s (Trans. ASME, v. 80, 1958) Erosion tests carried out for a number of materials:

· · · · 430°F / 350 psia 9% moisture 460 ft/s steam velocity 1000 h duration

Key result: 3­4 mils erosion in carbon and ½-Mo steels

· Represents a rate of 0.025­0.035 inches per year · Erosion could be due principally or partly to presence of liquid (9%)

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002

46

Chemical Environment

· · · · · Volume of boric acid deposits produced Boric acid morphology and properties Concentration of primary water pH Electrochemistry

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002

47

Volume of Boric Acid Deposits on the Vessel Head Methodology

Integrate the leaking boron mass over the fuel cycle Calculate the volume of leaked boron based on the density of boric acid (H3BO3) or boric oxide (B2O3) crystals, conservatively assuming no porosity The fraction of precipitated boron compounds that deposits on the head adjacent to the leaking nozzle may be affected by

· Droplet entrainment into the steam flow · Boric acid volatility (10% or less)

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002

48

Volume of Boric Acid Deposits on the Vessel Head Example Integration of Boron Mass

25,000

Volume of Boric Acid Crystals Exiting Annulus (in3)

203 lbs of crystals

Leak Rate = 0.01 gpm

20,000

15,000

10,000

start of leak

5,000

0

0

1

2

3

4

5

6

EFPYs After Start of First Cycle

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002 49

Boric Acid Morphology and Properties Boron Phases

Boric acid solutions and dry crystals

· · During evaporative concentration, boric acid solutions precipitate boric acid crystals The end results depend upon the rate of concentration and drying

If drying is fast, boric acid powder will result If drying is slow, a single irregularly shaped mass is likely

Molten boric acid

· · When heated above 340-365°F, solid boric acid melts to form a highly viscous liquid that will fuse into a single mass and flow under the influence of gravity Molten boric acid can contain 8-14% water by weight and is known to be corrosive

Solid boric oxide

· · Above 302°F boric acid is subject to a dehydration reaction to form boric oxide The resultant crystalline mass is an anhydrous, white, opaque, non-glasslike, stony solid

Molten boric oxide

· Above 617°F boric oxide begins to soften and at about 842°F becomes a highly viscous liquid

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002 50

Boric Acid Morphology and Properties Key Temperature Behavior

100%

Boric Acid Content (wt%)

80%

Melting Point of Boric Acid Crystals Solubility at Saturation Pressure Dryout of Mixture of Molten Boric Acid and Water at 1 atm

60%

40% Solubility at 1 atm 20%

0%

0

100

200

300

400

500

600

Temperature (°F)

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002 51

Boric Acid Morphology and Properties Partial Vapor Pressure

300

Partial Vapor Pressure of Water over a Saturated Boron Solution Partial Vapor Pressure of Water (psia)

250

200

Calculated Using Raoult's Law

sat pH2O = PH2 O x H2 O

150

100

50

0 200

300

400

500

600

700

800

900

52

Temperature (°F)

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002

Boric Acid Morphology and Properties General pH Effects without Large Local Cooling

For low concentration factors, the solution becomes slightly alkaline, having a small effect on crack growth rates For high concentration factors, the solution becomes acidic with a high-temperature pH of 4.5 according to MULTEQ calculations The initial high ratio of crevice surface area to volume may allow some buffering by the iron in the head material Precipitation of complex lithium and boron compounds occurs and tends to limit pH swings

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002

53

MULTEQ Modeling Three Main Flow Models Available

Step 1: Equilibrium Calculated Using Constant M ass Variable Volum e Constant M ass Vapor Phase, Rem ains Solid P hases, Rem ain

Variable Volume

Step 2: Vapor and/or Solids Rem oved Variable Volum e

Con stant L iquid W ater Mass

Equ ilibriu m Vap or Pha se Flow O ut

Constant M ass Vapor Phase, Rem ains Solid Phases, Rem ain

New Control M ass Vapor Phase, Rem ains

Solution Flow In

Solid Phas es, Re m ain W ater Mass Flow In (Solu tion) Eq uals W ater Ma ss Flo w O u t (Vap or)

Static

Static with Removal

Flowing

54

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002

MULTEQ Modeling Available Control Volumes

O nly Vapor Flow Out Control Volume with Constant Liquid M ass Only Solution Flow In

Only Vapor Flow Out

C ontrol Mass a t H ighe r C once ntra tio n Factor

Control Volum e with Constant Liquid Mass

C ontrol M ass a t Lo we r C once ntra tio n Factor

Only Solution Flow In

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002

55

Example MULTEQ Calculation pH in a Flowing System at 100°C

7 Concentration of Boron Dominates Evaporation of Boron Dominates 0.08

6

0.07

0.06 5 0.05 4

pH Li/B Ratio (in solution)

0.04 3 0.03 2 Initial Li/B = 2.2/1000 Flowing System 100°C

0.02

1 Evaporation of Boron Dominates 0 1.E+00 1.E+01 1.E+02 1.E+03 Concentration Factor 1.E+04 1.E+05 Lithium Decreases Boron Volatility

0.01

0 1.E+06

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002

56

Molten Boric Acid

Orthoboric Acid-H3BO3 Metaboric Acid-HBO2 Boric Oxide B2O3

Corrosion in molten boric acid largely unstudied Degradation:

· Melting point above the degradation point

Orthoboric acid: melts at 170.9°C (340°F); degrades to metaboric acid at 169.6°C (337°F) Metaboric acid: melts at 236°C (457°F); degrades to boric oxide at 235°C (455°F)

· Degradation reaction is slow · Effect of degradation products on corrosion largely unknown

(degradation probably lower in boric oxide, B2O3, than in either acid) Analysis of deposits not likely to indicate their at-temperature composition

· Degradation products highly hygroscopic

-

Solubility issues largely unstudied

· Miscibility limits unknown · For pH calculations, molten boric acid could be an additional precipitate · Degradation products not included in MULTEQ

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002 57

Molten Boric Acid Molten Salt Corrosion

Molten salt corrosion is electrochemically very similar to aqueous corrosion, depending on a reaction couple:

· · · · Fe Å Fe2+ anodic reaction O2 Å OH- or H+ Å H2 cathodic reaction Additional cathodic reactions unlikely in molten boric acid Typical molten salt corrosion occurs through de-passivation

Not relevant since LAS and CS are not passive in acidic media

Acceleration possible due to high conductivity of molten salts

· Unlikely to lead to a qualitative difference relative to highly concentrated solutions

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002

58

Molten Boric Acid Issues Molten Salt Corrosion (continued)

Solubility of corrosion products likely to be less in molten boric acid than in water

· Leads to lower corrosion rates

Molten boric acid corrosion likely to be significantly slower than corrosion in aqueous solution

· Lower O2 and H+ concentrations (slower cathodic reactions) · Possibly lower conductivity · Likely lower corrosion product solubility (slower anodic reactions)

Corrosion in molten boric acid is a particular case of corrosion in boric acid solutions, not a separate phenomenon

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002

59

Crevice Corrosion Mechanism Classic Crevice Corrosion is Not Believed to be Active

Crevice corrosion typically requires a passivating material in order to allow separation of cathodic and anodic zones Carbon and low alloy steels generally do not passivate in acidic media Corrosion testing in boric acid solutions indicates that general corrosion is much greater in aerated environments--i.e., there is no passivation

Iron Corrosion Rates in Various Solutions

Makar and Tromans, Corrosion 52:4 p.250, 1996

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002 60

Electrochemistry of Corrosion

Galvanic Corrosion Electrochemistry for a Non-Passivating Metal

1800 1600 1400 O2ÅHO1200 1000 800 600 400 200 0 -7 -6 -5 -4 H+ÅH2 FeÅFe2+ Potential (mV SHE)

O2ÅHO-

H+ÅH2

Current Density log(i )-1 (log(A/cm2)) -3 -2 0

1

2

3

4

Increasing Corrosion Rate

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002

61

Electrochemistry of Corrosion

Galvanic Corrosion Electrochemistry for a Non-Passivating Metal

1400

IR Drop Effect

1200 H ÅH2

+

1000 Potential (mV SHE)

800

DV =

Li x

FeÅFe2+

600

L = distance between metals (cm) 2 i = current (A/cm ) x = specific conductivity (S/cm)

400

200

0 -7 -6 -5 -4

Current Density log(i)-1 (log(A/cm2)) -3 -2 0

1

2

3

4

Increasing Corrosion Rate

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002

62

Degradation Progression Leak Rate is Main Controlling Parameter

Increasing Leak Rate

likely crack in nozzle wall reaches above top of weld on OD and ID Nozzle/Weld Condition leak path to annulus leak path to annulus leak path to annulus

PRELIMINARY

crack in nozzle wall reaches relatively high above top of weld on OD and ID

Annulus Condition

hypothetical clogged annulus

possibly some opening up of annulus

likely some opening up of annulus

may require some opening up of annulus

likely requires some opening up of annulus

Leak Rate

hypothetical zero leak rate

< on the order of 0.001 gpm

roughly between 0.001 and 0.01 gpm

roughly between 0.01 and 0.1 gpm

> on the order of 0.1 gpm

Liquid Velocity Exiting Crack

0 ft/s

less than on the order of 0.01 ft/s

roughly between 0.01 and 0.1 ft/s

roughly between 0.1 and 1 ft/s

> on the order of 1 ft/s

Local Temperature

600°F

Close to 600°F

at least roughly 500°F

roughly between 212 and 500°F

Close to 212°F

Liquid Location

fills annulus up to hypothetical blockage

all liquid vaporizes close to bottom of annulus

liquid film unlikely to exist high in annulus

liquid film may cover much of annulus walls

liquid film covers local top surface of head

Possible Significant Mechanisms

· none

· possibly very

minor galvanic

· possibly some

galvanic corrosion; · erosion until annulus opens slightly

· likely some galvanic

· aerated BAC on top of corrosion; head; · minor erosion and FAC; · possibly molten BAC, · possibly aerated BAC if galvanic corrosion, erosion, annulus is opened enough or FAC

roughly 70 to 700 lbs > on the order of 700 lbs

Pounds of Boric Acid Deposits Released in 2 years

at least small amount extruded

< on the order of 7 lbs all or most other leaking CRDM nozzles

roughly 7 to 70 lbs

Technical Assessment of Davis-Besse Degradation ­ May 22, 2002 EPRI & CE Annulus Tests Davis-Besse Nozzle #3

63

Information

63 pages

Find more like this

Report File (DMCA)

Our content is added by our users. We aim to remove reported files within 1 working day. Please use this link to notify us:

Report this file as copyright or inappropriate

932084