under a Thermal Gradient and Two Phase Flow Regime. Helen Hulme 1 , - - PowerPoint PPT Presentation

woodplc.com

Investigating the Corrosion Behaviour of Zircaloy-4 in LiOH under a Thermal Gradient and Two Phase Flow Regime.

Helen Hulme1, Alexandra Panteli1, Felicity Baxter1, Mhairi Gass1, Aidan Cole-Baker1, Paul Binks1, Mark Fenwick1, Michael Waters1, James Smith2

• Part one – influence of two phase flow

– Introduction – Testing procedure – Results – Discussion

• Part two – relationship between [LiOH] & Temp.

– Introduction - Literature data – Testing procedure – Results & Discussion

• Summary & Industrial Context

Outline

Part one – influence of two phase flow on corrosion of Zircaloy-4 in LiOH

3

Objective: To understand the influence of two-phase flow (subcooled nucleate boiling) within thick oxide films under a thermal gradient on the corrosion properties of zirconium alloys in LiOH chemistry.

– Is there potential to cause a localised increase in LiOH concentration in cracks / pores, as a result of boiling in thick oxide films, leading to accelerated corrosion? – What is the critical LiOH concentration required to cause accelerated corrosion? – Could this be detrimental to the use of LiOH without boric acid additions, eg. For SMR design considerations?

Introduction

Heatin ting block ck Specim cimen n enclo close sed No specim cimen Water r flow

Testing Conditions

5

T esting under a thermal gradient was performed in Wood’s “heat flux rig”

• Recirculating autoclave loop
• Pressurised system
• Single coolant temperature
• Variable flow across

specimen

• ΔT across specimen (<90 °C)

Testing Conditions

6 A presentation by Wood.

Test Parameters Initial Oxide Thickness (µm) Water Chemistry Thermal Gradient Pre-Stressed Boiling 1 LiOH Yes No No 2 LiOH Yes No Yes 3 < 20 NH4OH Yes No Yes 4 LiOH Yes No No 5 LiOH Yes No Yes 6 LiOH No No No 7 > 20 LiOH Yes No No 8 NH4OH Yes No No 9 NH4OH Yes No Yes 10 NH4OH No No No 11 LiOH Yes No Yes 12 LiOH Yes Yes Yes

Testing Conditions

7 A presentation by Wood.

Test Parameters Initial Oxide Thickness (µm) Water Chemistry Thermal Gradient Pre-Stressed Boiling 1 LiOH Yes No No 2 LiOH Yes No Yes 3 < 20 NH4OH Yes No Yes 4 LiOH Yes No No 5 LiOH Yes No Yes 6 LiOH No No No 7 > 20 LiOH Yes No No 8 NH4OH Yes No No 9 NH4OH Yes No Yes 10 NH4OH No No No 11 LiOH Yes No Yes 12 LiOH Yes Yes Yes

Zircaloy-4 sheet specimens pre-filmed in 500 °C air to form relatively thick oxide films in a timely manner

Testing Conditions

8 A presentation by Wood.

Test Parameters Initial Oxide Thickness (µm) Water Chemistry Thermal Gradient Pre-Stressed Boiling 1 LiOH Yes No No 2 LiOH Yes No Yes 3 < 20 NH4OH Yes No Yes 4 LiOH Yes No No 5 LiOH Yes No Yes 6 LiOH No No No 7 > 20 LiOH Yes No No 8 NH4OH Yes No No 9 NH4OH Yes No Yes 10 NH4OH No No No 11 LiOH Yes No Yes 12 LiOH Yes Yes Yes

Corie ieu et al al, 1962 Corrosion behaviour of Zircaloy-4 in NH4OH similar to water (i.e. no accelerated corrosion even in extreme concentrations). Using NH4OH removes any effect of pH

Testing Conditions

9 A presentation by Wood.

Test Parameters Initial Oxide Thickness (µm) Water Chemistry Thermal Gradient Pre-Stressed Boiling 1 LiOH Yes No No 2 LiOH Yes No Yes 3 < 20 NH4OH Yes No Yes 4 LiOH Yes No No 5 LiOH Yes No Yes 6 LiOH No No No 7 > 20 LiOH Yes No No 8 NH4OH Yes No No 9 NH4OH Yes No Yes 10 NH4OH No No No 11 LiOH Yes No Yes 12 LiOH Yes Yes Yes

Testing Conditions

10 A presentation by Wood.

Test Parameters Initial Oxide Thickness (µm) Water Chemistry Thermal Gradient Pre-Stressed Boiling 1 LiOH Yes No No 2 LiOH Yes No Yes 3 < 20 NH4OH Yes No Yes 4 LiOH Yes No No 5 LiOH Yes No Yes 6 LiOH No No No 7 > 20 LiOH Yes No No 8 NH4OH Yes No No 9 NH4OH Yes No Yes 10 NH4OH No No No 11 LiOH Yes No Yes 12 LiOH Yes Yes Yes

Pre-stressed Oxides

11 A presentation by Wood.

Pre-filmed Oxide Zircaloy Specimen Attach to heater block and Heat-up  Pre-filmed Oxide Zircaloy Specimen Steel Heater Block

• Standard specimens are pre-filmed unrestrained.
• T

est set-up induces a stress on the Zircaloy specimen upon heating.

• T
• reduce this, specimen is pre-filmed attached to the heater block, conditioning it to

the test environment

Results

12

Results – influence of oxide thickness

15 A presentation by Wood.

Thick k Oxides es (>20 0 µm) ) re requi uired ed for r accelera rated ted corrosion

• sion to occur

Thick = >20 µm Thin = = <2 <20 µm

Coolant chem. 2 ppm LiOH Autoclave temp. 250 °C

Results – influence of boiling

16 A presentation by Wood.

Within thick k ox

• xid

ides es (>20 µm), boiling ing is re requi uired ed for r accelerate rated d corrosion

• sion to occur

ur

Coolant chem. 2 ppm LiOH Autoclave temp. 250 °C

Results – influence of chemistry

17 A presentation by Wood.

For thick ck ox

• xides

des (>20 µm) under boiling ing conditions, LiOH is re requir uired ed for r accelera rated ted corrosion sion to occur. Accelerated corrosion is not

• bserved in the presence of NH4OH

Coolant chem. 2 ppm LiOH Autoclave temp. 250 °C

Results – influence of stress

18 A presentation by Wood.

For thick ck oxides des (>20 µm) under boiling ing conditions in LiOH, stress ess in the ox

• xide

ide is re requi uired ed for accelerate rated d corrosion

• sion to occur

ur. Accelerated corrosion is not

• bserved if the oxide is formed in a

pre-stressed condition

Coolant chem. 2 ppm LiOH Autoclave temp. 250 °C

19

Discussion

• For accelerated corrosion to be observed during testing, the following criteria must be met:

– Thick >20 µm oxide film – Sub-cooled boiling – LiOH chemistry – Stress

Discussion – key observations

20 A presentation by Wood.

Key observations:

• Results from pre-filmed specimens under non-boiling conditions do not show

accelerated corrosion, indicating this is not a m memor

• ry

y effect ct.

• Comparable test conditions using NH4OH do not show accelerated corrosion,

demonstrating that LiOH does have e an effect ct.

Discussion – Hypothesised Mechanism

21 A presentation by Wood.

Effect of Stress Stress causes the pores & cracks present in the

• xide to open and

create a more accessible pathway for the coolant to penetrate nearer the metal / oxide interface Effect of Boiling Effect of LiOH

Discussion – Hypothesised Mechanism

22 A presentation by Wood.

Effect of Stress Effect of Boiling Stress causes the pores & cracks present in the

• xide to open and

create a more accessible pathway for the coolant to penetrate nearer the metal / oxide interface Boiling within the cracks & pores that have limited accessibility to the coolant causes localised concentration of LiOH solution to levels above that

• f the bulk coolant. SIMS

analysis agrees with this

• ccuring

Discussion – Hypothesised Mechanism

23 A presentation by Wood.

Effect of Stress Effect of Boiling Effect of LiOH Stress causes the pores & cracks present in the

• xide to open and

create a more accessible pathway for the coolant to penetrate nearer the metal / oxide interface Boiling within the cracks & pores that have limited accessibility to the coolant causes localised concentration of LiOH solution to levels above that

• f the bulk coolant. SIMS

analysis agrees with this

• bservation

Similar observations were seen by Jeong et al. 1999, following accelerated corrosion under a 70 ppm LiOH, 350 °C isothermal autoclave environment

Effect of boiling: SIMS results

24 A presentation by Wood.

Accel eler erate ated Accel eler erate ated Non-acce accele lerate ated Non-acce accele lerate ated Where accelerated corrosion was observed (as a result of boiling), lithium was leachable from cracks and pores indicating local accumulation

• f lithium in these regions.

All data from oxides >20 µm

Effect of boiling: SIMS results

25 A presentation by Wood.

Accel eler erate ated Accel eler erate ated Non-acce accele lerate ated Non-acce accele lerate ated Leachable lithium content was measured using ICP-OES. This equates to a lithium concentration of 25 ppm within the entire oxide film. Calculations estimate the metal /

• xide interface temperature,

under conditions for boiling, would be ~315 °C This is significantly lower than that expected for accelerated corrosion

ICPOES - Inductively Coupled Plasma Optical Emission Spectroscopy

Discussion – Hypothesised Mechanism

26 A presentation by Wood.

Effect of Stress Effect of Boiling Effect of LiOH Stress causes the pores & cracks present in the

• xide to open and

create a more accessible pathway for the coolant to penetrate nearer the metal / oxide interface Boiling within the cracks & pores that have limited accessibility to the coolant causes localised concentration of LiOH solution to levels above that

• f the bulk coolant. SIMS

analysis agrees with this

• ccurring

Billot et al. proposed that at high levels, lithium becomes incorporated into the oxide film on pore walls forming Li2ZrO3 within pores, which dissolves, further developing the porous network. Billot et al suggest an additional impact of oxide thickness, whereby the dimensions of the pores become sufficient to support rapid transport of lithium to the metal / oxide interface when the oxide thickness exceeds approximately 20 µm. Our data is consistent with this.

• Criteria for accelerated corrosion:

– Thick >20 µm oxide film – Sub-cooled boiling – LiOH chemistry – Stress

The above conditions cause LiOH to concentrate to a critical level resulting in accelerated corrosion.

• Q. What level of LiOH is required for accelerated corrosion to occur?

What does this mean for industry?

27 A presentation by Wood.

Part two – investigating the relationship between critical [LiOH] and temperature

28

Introduction - Understanding from Literature

29

Data taken from Murgatroyd et al., Pecheur et al., Ramasubramanian et al., Bramwell et al., Jeong et al. and McDonald et al. Grey lines indicate bounding conditions for acceleration based

• n these data.

All data are taken from post-transition (i.e. beyond 2 µm) corrosion rates of Zircaloy.

Testing Conditions

30

Specimen ID Steam Pre-Film (Demin, 400 °C) Aqueous Pre-film (2 ppm LiOH, 350 °C) Calculated time at temp. according to Hillner (days) Test Conditions Exposure (days) Oxide thickness (µm) Exposure (days) Oxide thickness (µm) [LiOH] (ppm) Temperature (°C) AC1 56 2.40 10 3.15 228 2 350 AC2 56 3.03 10 3.57 259 AC3 56 2.78 10 3.39 246 60 350 AC4 56 2.47 10 3.25 236 AC5 56 2.46 10 3.25 236 110 350 AC6 56 2.96 10 3.50 254 AC7 56 2.81 10 3.41 482 250 330 AC8 56 2.37 10 3.09 436

Pre-film in steam to

• btain post-

transition corrosion films in reasonable timeframe. Additional pre-film in water to remove any memory effects from steam environment Oxide thickness’ time estimated using Hillner 1977 corrosion equations

Testing Conditions

31

Test Condition Temp. (°C) [LiOH] (ppm) Exposure in LiOH (days) 1 350 2 215 2 350 60 177 3 350 110 120 4 330 250 76

Results

32

Results

33 A presentation by Wood.

No acceleration – cyclic corrosion observed Accelerated corrosion seen after an incubation period (oxide growth of ~2 µm) Despite almost double [LiOH], similar incubation period and accelerated corrosion rate

Results

34 A presentation by Wood.

No accelerated corrosion observed for 250 ppm LiOH, 330 °C; however, not yet achieved 2 µm

• xide in this

environment

35

Summary

Summary - What does all this information tell us?

36 A presentation by Wood. 100000 573 583 593 603 613 623 633 643 653

1

1 10 100 1000 10000 100000 300 310 320 330 340 350 360 370 380

Lithium Concentration (ppm) Temperature (°C)

Accelerated Accelerated (slower rate) Not accelerated Isothermal - Accelerated Isothermal - Not accelerated Thermal Gradient

Literature data Test data

Threshold reduction due to stress Li enhancement due to boiling

• Eq. 1

Temperature (K)

𝑀𝑗 = 8 x 1015 𝑓−0.095.𝑈 Equation 1: LiOH has been shown to affect the corrosion behaviour of thick >20 µm

• xide under boiling and stress conditions.

This is thought to be due to accumulation of LiOH in cracks and pores as a result of boiling which increases the concentration of LiOH locally. Stress impacts this mechanism; data suggests this is by lowering the critical [LiOH] required for accelerated corrosion to

• ccur from 800 ppm to 25 ppm @ 315 °C

?

• boiling

Summary - What does all this information tell us?

37 A presentation by Wood. 100000 573 583 593 603 613 623 633 643 653

1

1 10 100 1000 10000 100000 300 310 320 330 340 350 360 370 380

Lithium Concentration (ppm) Temperature (°C)

Accelerated Accelerated (slower rate) Not accelerated Isothermal - Accelerated Isothermal - Not accelerated Thermal Gradient

Literature data Test data

Threshold reduction due to stress Li enhancement due to boiling

• Eq. 1

Temperature (K)

Further studies have investigated the relationship between [LiOH] and

• temperature. Data obtained to date has

allowed us to reduce the threshold band for the critical [LiOH] – temperature relationship than that produced from literature alone. Key point to note is that this relates to the temperature of the metal / oxide interface, not the coolant! 𝑀𝑗 = 8 x 1015 𝑓−0.095.𝑈 Equation 1:

• boiling

?

• Data here suggest that the use of 2 ppm LiOH may not have an adverse effect on corrosion

where oxide films < 20 µm because the temperature of the metal / oxide interface would not be significantly higher than the coolant, therefore an extremely high level of LiOH would be required to cause accelerated corrosion – This argument assumes the absence of stress, which appears to lower the critical lithium concentration required for accelerated corrosion at a given temperature.

• Key considerations for the use of LiOH in the absence of boric acid (e.g. For future SMR

applications) include: – Could sub-cooled nucleate boiling influence corrosion later in life when films are thickest? – How does the metal / oxide interface temperature change during the core lifetime? Does the oxide film experience changes in stress during corrosion? – Read-across to other zirconium alloys? – Effect of irradiation on this mechanism?

Summary - What does this mean for Industry?

38 A presentation by Wood.

woodplc.com

• Coriou, H., Grall, L., Meunier, J., Pelras, M., Willermoz, H., Corrosion du Zircaloys dans Divers Milieux Alcalins a Haute Temperature, Journal
• f Nuclear Materials, 7(3), 1962: 320-327.
• Kandlikar, S.G., Heat Transfer Characteristics in Partial Boiling, Fully Developed Boiling, and Significant Void Flow Regions of Subcooled

Flow Boiling, Journal of Heat Transfer, 120 (2), (1998) 395 – 401.

• Abram, T.J., Modelling the Waterside Corrosion of PWR Fuel Rods, IAEA Technical committee Meeting on Water Reactor Fuel Element

Modelling at High Burnup, Windermere 1994, IAEA-TECDOC-957, p329.

• Kingery, W.D., Francl, J., Coble, R.L., & Vasilos, T., Thermal Conductivity: X, Data for Several Pure Oxide Materials Corrected to Zero

Porosity, Journal of the American Chemical Society, 37 (2), 1954, p107-111.

• Jeong, Y.H., Kim, K.H., Baek, J.H., Cation Incorporation into Zirconium Oxide in LiOH, NaOH, and KOH Solutions, Journal of Nuclear

Materials, 275 (2) 1999, pp 171-177.

• Murgatroyd R.A., Winton, J., Hydriding of Zircaloy-2 in Lithium Hydroxide Solutions, Journal of Nuclear Materials, 23 (1967) pp 249 –

256.

• Pecheur, D, Godlewski, J., Peybernes, J., Fayette, L., Noe, M.M Frichet, A., & Kerrec, O., Contribution to the Understanding of the Water

Chemistry on the Oxidation Kinetics of Zircaloy-4 Cladding, 12th Zirconium in the Nuclear Industry Symposium, 2000, STP 1354, pp 793.

• Ramasubramanian, N., Precoanin, N., Ling, V.C., Lithium Uptake and the Accelerated Corrosion of Zirconium Alloys, 8th Zirconium in the

Nuclear Industry Symposium, 1989, STP 1023, pp 187.

• Bramwell, I.L, Parsons, P.D., Tice, D.R, Corrosion of Zircaloy-4 PWR Fuel Cladding in Lithiated and Borated Water Environments, 9th

Zirconium in the Nuclear Industry Symposium, 1991, STP 1132, pp 628.

• McDonald, S.G., Sabol, G.P., & Sheppard, K.D., Effect of Lithium hydroxide on the Corrosion Behavior of Zircaloy-4, 6th Zirconium in the

Nuclear Industry, 1984, STP 824, p519.

• Hillner E., Corrosion of Zirconium Base Alloys – An Overview, Zirconium in the Nuclear Industry, ASTM STP 633, A.L. Lowe and G.W. Parry

(1977) pp 211 – 235

References

40 A presentation by Wood.