FRACTURE BEHAVIOUR OF

PRESSURE TUBES
A. Celovsky
Manager, Materials & Mechanics Branch
Presentation to USNRC and CNSC
2002 December

OUTLINE
Historical Overview
Cracking Mechanisms (Delayed Hydride Cracking)
Required conditions
Cracking characteristics

Effects of Aging
Fracture property changes due to irradiation
Hydrogen effects

Design and Operation to Prevent Fracture

Flaw Evaluation following FFSG
Defense in depth: Leak-before-break

Summary
Pg 2

PRESSURE TUBE HISTORY

1973/4 leaking tubes at Pickering A
Delayed hydride cracking (DHC) in the very high stress regions
of over-extended rolled joints during first use of Zr-2.5Nb
material
Cracking occurred during shutdowns

1982 leaks at Bruce A also in rolled joint zone

initiation occurred before stress relief
1983 Pickering A Zircaloy tube rupture due to hydride
blister formation at contact points with calandria tube
cracking occurred at power

Pg 3

HISTORY, cont
1986 Leaking Bruce A tube ruptured during leak
search with reactor cold long, manufacturing flaw
was the crack initiator
Since 1986 excellent performance
Including operation with original rolled joints with initially
high residual stress levels
Many pressure tubes in contact with calandria tubes
Operating and maintenance modified to avoid DHC and
hydride blister formation in pressure tubes
>100,000 pressure-tube-years of operation without any
failures
Pg 4

FRACTURE BY DELAYED HYDRIDE

CRACKING (DHC)
DHC is the only significant mechanism of crack initiation
and growth in pressure tubes
-no known environmental cracking mechanisms under normal
operating conditions (no SCC, IGA, EAC, etc.)
- fatigue not an issue

DHC is very well characterized through material

surveillance, experimental and theoretical work over
many years
Unlike SCC & other mechanisms, DHC is more
predictable and can be avoided
Pg 5

DELAYED HYDRIDE CRACKING

Calandria Tube
DHC growth
rings through
pressure tube wall
thickness

Leak

Leak Detection System

(Continuously monitored)
Pressure Tube

Coolant

Coolant

Pg 6

EXAMPLE OF DHC IN A Zr-2.5Nb

LAB SPECIMEN
An artificial defect in a lab specimen

Axial Crack
Velocity, VAxial
PLAN VIEW
Growth rings of DHC initiated from a artificial defect in a lab
specimen made from Zr-2.5Nb pressure tube material

Artificial
Defect
~25% of
wall

Radial
Crack
Velocity,
VRadial

CROSS-SECTION

DHC only significant mechanism for

cracking in Zr-2.5Nb
Pg 7

NECESSARY CONDITIONS REQUIRED

FOR DHC
1) Sufficient hydrogen concentration in the material to precipitate
hydride under conditions considered (temperature, stress)
AND
2a) At smooth surfaces very high stresses are required
(e.g. initial Pickering A rolled joints)
Does not exist in current design nor in ACR

OR
2b) Contact with calandria tube reducing local temperature
Does not exist in current design nor in ACR

OR
2c) A flaw that results in a high enough effective stress intensity
Ref: S. Sagat, S.Q. Shi & M.P. Puls, Crack Initiation Criterion at Notches in Zr-2.5Nb Alloys, Mats Sci and Eng, A176 (1994) 237-247.
Pg 8

HYDROGEN
180
160

TSSP

140

le

120

es
po
ss
ib

Solid Hydrides
Present

100

H
yd
rid

80
60

So
lid

Hydrogen Solubility (ppm)

Zr 2.5Nb

40

TSSD

DHC not
possible
Hydrides
dissolved

20
0
0

50

100

150

200

250

300

350

400

Temperature (C)

CANDU 6
Terminal Solid Solubility on Dissolution (TSSD)
Terminal Solid Solubility on Precipitation (TSSP)

ACR
Pg 9

DHC CHARACTERISTICS
When sufficient hydrogen is present to allow cracking to occur, the
principal features of the cracking are (similar to SCC):
Threshold stress intensity factor below which cracking ceases (KIH)
Depends on fluence

Crack grows by step-wise hydride precipitation, growth and fracture

Plateau for the crack growth rate (temperature-dependent)
Depends on rate of transport of hydrogen in solution in material to flaw tip
temperature dependence derives primarily from solubility and diffusion
coefficient of hydrogen in the material
Crack growth rates also depend on thermal history

For the FFSG database, testing is done to establish the most

conservative values
Pg 10

IRRADIATION EFFECTS ON
MATERIAL PROPERTIES
Both new and end-of-life properties are assessed
Lifetime fast-neutron (E> 1 MeV) irradiation fluence is high
Trend in properties is for an initial rapid change ( within first year
or so) followed by very small changes over remaining lifetime
Strength increases
DHC velocity increases (if sufficient hydrogen is present for DHC
to occur)
Fracture toughness decreases

Pg 11

EFFECT OF IRRADIATION ON
STRENGTH
1000
Unirradiated

900

Transverse UTS, MPa

800
At Low Fluence

700
600

Intial Transient

500
400

Mid-Life
Fluence

300
200

Linear
Regression

100
0
0

10

15

20

Fast Neutron Fluence >1MeV, x10

25
25

30

n/m

All blue data from same source material

Ref: R.R. Hosbons et. al., Effect of Long-Term Irradiation on the Fracture Properties of
Zr-2.5Nb Pressure Tubes, ASTM 1354, pp 122-138 (2000).

Pg 12

EFFECT OF IRRADIATION ON
DHC VELOCITY
At a test temp of 464 F (240C)
1.E-06

V, m/s

Low & mid

life Fluence

1.E-07

Log-Normal
Fit

1.E-08
0

10

12

14

16

18

20

22

24

26

28

30

Fast-Neutron Fluence >1MeV, x1025 n/m2

All blue data from same source material

Ref: R.R. Hosbons et. al., Effect of Long-Term Irradiation on the Fracture Properties of
Zr-2.5Nb Pressure Tubes, ASTM 1354, pp 122-138 (2000).

Pg 13

EFFECT OF IRRADIATION ON
FRACTURE TOUGHNESS
CHLORINE EFFECT

FLUENCE EFFECT

Initial Crack Growth Toughness from CCTS vs Fluence

Tube H737 tested at 240C

600

200

F r a c t u r e T o u g h n e s s ( d J /d A ) M P a

Unirradiated controls
NRUtest reactor/lowflux (Tirr = 265C)
OSIRIS test reactor/high flux (Tirr = 250C)

150

dJ/da, MPa

Linear regression to OSIRIS results

100

Older type tubes

Initial
transie nt

50

New improved tubes

500
400

Older type tubes

300

200
100
0

0
0

10

12

14

16

18

Fast-Neutron Fluence, x 10^25 n/m

0.00

2.00

4.00

6.00

Chlorine Conc. ppm(wt)

8.00

10.00

Fracture toughness of tubes

currently in service (produced
30 years ago) is sufficient to
demonstrate leak-before-break
for crack growth from small
initiating flaws.
R&D during past 15 years has
resulted in a large improvement
in fracture toughness of
irradiated material due to
removal during tube production
of chemical species associated
with low irradiated toughness
e.g. chlorine

Ref: R.R. Hosbons et. al., Effect of Long-Term

Irradiation on the Fracture Properties of Zr-2.5Nb
Pressure Tubes, ASTM 1354, pp 122-138 (2000).
Pg 14

EFFECTS OF HYDROGEN ON
TOUGHNESS
Hydrogen in solution has no significant effect
Quasi-planar hydride precipitates oriented
perpendicular to the plane of fracture have little
effect on toughness except at very high
concentration
Hydrides lying in the potential fracture plane
(perpendicular to hoop stress in the tube) can have
detrimental effect on toughness
Hydrides only exist at operating temperature if
solubility of hydrogen is exceeded (not predicted
to occur for ACR)
Produce and operate tubes to avoid formation of
radial hydride

Lab Specimen showing

solid hydrides (black)

Hoop Stress

Potential Fracture
Direction

Pg 15

DESIGN AND OPERATION TO

PREVENT FRACTURE
Design for leak-before-break
Good pressure tube fracture toughness and annulus gas
system leak detection
Manufacturing methods avoid long flaws and manufacturing
inspections to a very high standard
No hydrides present at operating conditions no DHC
No pressure tube to calandria tube contact no hydride
blisters

Pg 16

DESIGN AND OPERATION TO

PREVENT FRACTURE
Operation
Reduce pressure / stress at temperatures below operating
temperature - no crack initiation from any potential flaws
when hydrides present
Ensure foreign materials exclusion during maintenance
Monitor core state through inspection / surveillance
Assess any flaws and the core according to Fitness for Service
Guidelines (future CSA Standard N285.8)

Monitor annulus gas system for water ingress

Use appropriate operating procedures in response to leakage

Pg 17

SUMMARY
Fracture behaviour of Zr-2.5Nb pressure tubes is well
understood
Tubes with high fracture toughness after irradiation
can be specified
Delayed hydride cracking can be avoided during
operation by limiting hydrogen concentrations
Leak-before-break can be demonstrated

Pg 18

Pg 19