Hydriding Induced Corrosion Failures in BWR Fuel Dan Lutz 1 , Yang-Pi Lin 2 , Randy Dunavant 2 , Rob Schneider 2 , Hartney Yeager 2 , Aylin Kucuk 3 , Bo Cheng 3 , Jim Lemons 4 1 Global Nuclear Fuel – Americas, Sunol, CA. 2 Global Nuclear Fuel – Americas, Wilmington, NC 3 Electric Power Research Institute, Palo Alto, CA 4 Tennessee Valley Authority, Chattanooga, TN ASTM 17th International Symposium on Zirconium in the Nuclear Industry, Hyderabad, India
20
Embed
Hydriding Induced Corrosion Failures in BWR Fuel · Corrosion Features • Elevated, coalesced nodular corrosion and oxide spallation • Excessive corrosion, spallation, hydriding
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Hydriding Induced Corrosion Failures in BWR Fuel Dan Lutz1, Yang-Pi Lin2, Randy Dunavant2, Rob Schneider2, Hartney Yeager2, Aylin Kucuk3, Bo Cheng3, Jim Lemons4
1 Global Nuclear Fuel – Americas, Sunol, CA. 2 Global Nuclear Fuel – Americas, Wilmington, NC 3 Electric Power Research Institute, Palo Alto, CA 4 Tennessee Valley Authority, Chattanooga, TN
ASTM 17th International Symposium on Zirconium in the Nuclear Industry, Hyderabad, India
Outline
• Key features of corrosion failures • Summary of failure mechanism • Unresolved issues
– Environmental factor contribution – Hydride localization under BWR conditions
• Results • Implications of environmental factor to the
BWR fleet
2 ASTM 17th International Symposium on Zirconium in the Nuclear Industry, Hyderabad, India
– 63 assemblies (9x9), starting at ~30 GWd/MTU, 7 month into 24 month cycle
– Only second cycle fuel failed – Cladding type: Zircaloy-2 with barrier, over 3 million fuel rods
experience base – Failed rods from multiple ingots; initial failures involved low
alloy content, common lot material operated successfully in 12 other reactors to end-of-life
– Two rounds of hot cell investigations for root cause and failure mechanism
• Browns Ferry 3, Cycle 11 (2002 – 2003) – 3 assemblies, near end of 3 cycle life – 1 assembly discharged after 1 cycle, not corrosion related
3 ASTM 17th International Symposium on Zirconium in the Nuclear Industry, Hyderabad, India
Corrosion Features • Elevated, coalesced nodular corrosion and oxide spallation
• Excessive corrosion, spallation, hydriding and the primary failure perforations near 2.4 m (95 in) and above
• Low corrosion at assembly spacer (Zircaloy-2 ferrule with Inconel springs) locations
• Large azimuthal variation in corrosion Spacer
Location
4 ASTM 17th International Symposium on Zirconium in the Nuclear Industry, Hyderabad, India
Oxide + Crud thickness Mostly oxide
Failure Mechanism • Schardt et al, “Fuel Corrosion Failures in the Browns Ferry Nuclear Plant,” 2004 International Meeting on LWRFP, Orlando, FL, September
2004. • Lutz et al, “Investigation of BWR Fuel Failures,” TopFuel 2012, Manchester, UK, September 2012.
• Elevated corrosion on cladding with inadequate corrosion resistance for the exposed environment resulting in high hydrogen pickup
• Spallation of oxide locally • Hydrogen localization leading to hydride lens near cladding OD • Cladding stressed from cladding creep down and pellet expansion • Failure from cracks initiated in brittle hydride
• Shared similarity in corrosion characteristics with previous CILC from 1970 - 1980s: failure involved crud intrusion into oxide, thermal insulation and autocatalytic corrosion
5 ASTM 17th International Symposium on Zirconium in the Nuclear Industry, Hyderabad, India
Sound Rod
Failed Rod
Unresolved Issues
• Environmental factor for elevated corrosion – Material factor alone cannot explain elevated
corrosion; expect coolant water chemistry effect – Coolant’s role in the failures not readily
identifiable from review of water chemistry data • Hydrogen localization under BWR conditions
– Are local thermal gradients at discontinuities in oxide thickness sufficient to cause localization?
6 ASTM 17th International Symposium on Zirconium in the Nuclear Industry, Hyderabad, India
Hydride Localization
7 ASTM 17th International Symposium on Zirconium in the Nuclear Industry, Hyderabad, India
Oxide
Cladding
Crud
Oxide spalling
Oxide plateau
Objectives and Scope
• Environmental – SIMS analyses of oxide (Failed vs. sound, plants, elevation
and common material)
– Assess hideout return investigation data (water chemistry during reactor shutdown)
• Hydrogen localization under BWR conditions – Finite element simulation
8 ASTM 17th International Symposium on Zirconium in the Nuclear Industry, Hyderabad, India
Samples Analyzed Using SIMS Sample Assembly Rod Condition
Rod Average Exposure
(GWd/MTU) Reactor
Insertion Date
Residence Time
(Days)
Sample Elevation (mm/inch)
Average Oxide
Thickness (microns)
A YJS734 H2 Failed 47.3 BF2 May 1999 1211 ~2286 / 90 n/a*
B YJK363 B3 Sound 35.1 BF2 Oct 1997 1751 ~3023 / 119 21.1
C YJ1380 D1 Sound 68.9 L Jul 1992 2708 ~2413 / 95 21.5
D YJS614 G9 Sound 34.5 BF2 May 1999 1040 ~3023 / 119 21.9
E YJS616 B8 Sound 41.1 BF2 May 1999 1040
724 / 28.5 28
F 2311 / 91 23*
G YJN587 E9 Sound 17.0 BF3 Oct 1998 548
762 / 30 18
H 2362 / 93 17
I YJP354 E8 Sound 44.0 H Nov 1998 1498
699 / 27.5 18
J 2286 / 90 10
9 ASTM 17th International Symposium on Zirconium in the Nuclear Industry, Hyderabad, India
* Severely Spalled
Sound Rod, Heavy corrosion
Earlier Reload
Another BWR Early discharge Light corrosion
Failed Rod
1 Cycle Rod
Common Lot (to E and F)
Li in Oxide: SIMS Line Scans D, Plant BF2
C, Plant L
METAL
METAL
METAL
• High Li level in failed BF2 rod • Lower Li level in sound, low corrosion BF2 rod • Low Li in sound rod from plant L • 6Li implies natural source
10 ASTM 17th International Symposium on Zirconium in the Nuclear Industry, Hyderabad, India
Failed BF2 H2
Line Scans • High [Li] with 6Li in sound
BF2 rod at upper elevation confirmed.
• 1 cycle BF3 rod shows lower [Li], higher and with 6Li at upper elevation
• 3 cycle Plant H rod shows low [Li]; trend with elevation reversed compared with BF2 and BF3 rods
11 ASTM 17th International Symposium on Zirconium in the Nuclear Industry, Hyderabad, India
E: BF2 B8 ~2 ppm
F: BF2 B8 ~300 ppm
G: BF3 E9 <1 ppm
H: BF3 E9 ~3 ppm
I: H E8 ~3 ppm
J: H E8 <1 ppm
Lower elevation Upper elevation
SIMS Imaging
Association of Li and B implies similar trapping site and entry path from coolant
BF2 B8 Upper elevation
BF3 E9 Upper elevation
12 ASTM 17th International Symposium on Zirconium in the Nuclear Industry, Hyderabad, India
OXIDE CRUD
METAL OXIDE
Hideout Return, BF2 and H, Feb 2003
After: Shutdown Hideout Return Chemistry at Browns Ferry-2 and Hatch-2. EPRI, Palo Alto, CA: 2004. 1009448
Reactor Hot Full power
Reactor Hot Zero power
Reactor Cold Zero power
• Plant H had cladding common to failed BF2 rods
• No corrosion failures at Plant H
• Li below detection limit at Plant H in hideout return study (detected in SIMS)
• Sulfate, chloride, Ca return also higher at BF2 relative to Plant H
• Steady state sulfate and chloride reflect hideout return results
• Concentration factor 103 - 104 predicted for BWR; possible ppm level of Li in BF2
• Together with SIMS results indicative of environmental factor effect
13 ASTM 17th International Symposium on Zirconium in the Nuclear Industry, Hyderabad, India
Hydrogen Localization Model
14 ASTM 17th International Symposium on Zirconium in the Nuclear Industry, Hyderabad, India
Fuel Cladding Thickness, TH 0.7 mm
Fuel Cladding Section Axial Height Modeled, HT 25 mm
Thick Oxide Thickness, OXTHICK 60
30 microns
Spall Depth or Oxide Plateau Height, OXTHIN 45/15
15 microns
Axial Height of Spall or Thin Oxide Plateau,
OXHT � 2.5 mm, Spall0.25 mm, Plateau mm
LHGR 250 (7.6) W/cm (kW/ft)
𝑄𝑄𝛼𝛼∗
𝑅𝑅 3015𝐾𝐾 Ref. [3]
𝑄𝑄𝛿𝛿∗
𝑅𝑅 653𝐾𝐾 Ref. [3]
𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇 1.28 × 105𝑒𝑒−36540𝑅𝑅𝑇𝑇� 𝑝𝑝𝑝𝑝𝑝𝑝 Ref. [4]
𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇 � 1.07 × 104𝑒𝑒−21028
𝑅𝑅𝑇𝑇� 𝑝𝑝𝑝𝑝𝑝𝑝 𝑇𝑇 ≤ 533𝐾𝐾 5.26 × 104𝑒𝑒−28068
𝑅𝑅𝑇𝑇� 𝑝𝑝𝑝𝑝𝑝𝑝 𝑇𝑇 > 533𝐾𝐾
Ref. [4]
ID O/M Interface
𝑇𝑇𝛼𝛼(0) 2.17 × 10−3 𝑐𝑐𝑝𝑝
2
𝑠𝑠 Ref. [3]
𝑇𝑇𝛿𝛿(0) 1.09 × 10−3 𝑐𝑐𝑝𝑝
2
𝑠𝑠 Ref. [3]
𝑄𝑄𝛼𝛼𝑅𝑅
4170𝐾𝐾 Ref. [3] 𝑄𝑄𝛿𝛿𝑅𝑅
5730𝐾𝐾 Ref. [3] 𝑄𝑄∗
Thermal Gradient
ASTM 17th International Symposium on Zirconium in the Nuclear Industry, Hyderabad, India 15
temp at ID and O/M interface • ∆T not strongly sensitive to oxide
thickness (OXTHICK)
Clad Thickness, TH (mm)
Axi
al H
eigh
t, H
T/2
(mm
)
OD ∆T = 5ID ∆T = 5
250 W/cm60 µm oxide layer15 µm discontinuity
0 0.2 0.4 0.60
5
10
320330340
Clad Thickness, TH (mm)
Axi
al H
eigh
t, H
T/2
(mm
)
OD ∆T = 16ID ∆T = 15
250 W/cm60 µm oxide layer45 µm discontinuity
0 0.2 0.4 0.60
5
10
310320330340
317°C 349°C 317°C 349°C
Clad Thickness, TH (mm)
Axi
al H
eigh
t, H
T/2
(mm
)
OD ∆T = 3ID ∆T = 1
250 W/cm60 µm oxide layer15 µm discontinuity
0 0.2 0.4 0.60
5
10
320330340
Clad Thickness, TH (mm)
Axi
al H
eigh
t, H
T/2
(mm
)
OD ∆T = 10ID ∆T = 5
250 W/cm60 µm oxide layer45 µm discontinuity
0 0.2 0.4 0.60
5
10
310320330340
317°C 349°C 317°C 349°C
Oxide Plateau: • Similar trend as Oxide Spalling
cases • ∆T less than corresponding Oxide
Spalling cases
ID O/M Interface
Hydrogen Diffusion
ASTM 17th International Symposium on Zirconium in the Nuclear Industry, Hyderabad, India 16
Agreement with data from Sawatzky and Vogt (Mathematics of Thermal Diffusion of Hydrogen in Zircaloy-2, AECL-1411, 1971.)
Oxide Discontinuity and Hydride Localization
ASTM 17th International Symposium on Zirconium in the Nuclear Industry, Hyderabad, India 17
Oxide Spalling: • Localization increases with H
content and ∆T • 5 oC ∆T can cause localization
with 200 ppm H • No significant localization at 100
ppm H, even at 16 oC ∆T • Localization not sensitive to oxide
thickness (OXTHICK), similar ∆T
Oxide Plateau: • Similar trend as Oxide Spalling
cases • Less localization due to lower ∆T • 3 oC ∆T insufficient to cause
localization with 200 ppm H
60/15 micron oxide/plateau
60/45 micron oxide/plateau
60/15 micron oxide/spall
60/45 micron oxide/spall
1000 hrs
1000 hrs
Hydride Localization
ASTM 17th International Symposium on Zirconium in the Nuclear Industry, Hyderabad, India 18
• Hydride localization possible if H is high (above 100 ppm) and with
sufficient ∆T from elevated corrosion and spalling (and high heat flux) – all needed for failure mechanism to operate
Hydride Localization
Implications / Discussion
ASTM 17th International Symposium on Zirconium in the Nuclear Industry, Hyderabad, India 19
• Corrosion failures result from inadequate cladding corrosion resistance for the environment
• Li in oxide associated with corrosion failures – Control blade absorber tube leakage (B4C) can be a source of 7Li,
consistent with Li and B association but B4C leakage is common; presence of 6Li implies natural source
• Evidence of underperforming water chemistry cleanup system, e.g. sulfate spikes indicative of degraded cation resin
• Since the corrosion failures, cladding corrosion resistance and water chemistry control improved (Fe & Zn control, filter efficiency, monitoring)
• Confirmation of environmental factor provides added assurance for preventive measures
• No recurrence in ~10 years
Summary • Elevated corrosion associated with Li in oxide, indicating
environmental factor; Li as initiator not proven • Hideout return study supported higher concentration of
multiple species, including Li, than in reference plant • Finite element modeling confirm significant thermal gradient
associated with oxide spall or plateau can cause hydride localization if sufficient hydrogen is present
• Failure mechanism: cracking of brittle hydride under stress; hydride localization from thermal gradient at discontinuities of thick oxide; elevated corrosion from inadequate cladding corrosion resistance under demanding environment
ASTM 17th International Symposium on Zirconium in the Nuclear Industry, Hyderabad, India 20