Mechanical Properties of
Structural Steels in Hydrogen
B.P. Somerday, K.A. Nibur, C. San Marchi, and M. Yip Sandia National Laboratories
Livermore, CA
DOE Hydrogen Pipeline Working Group Meeting Aiken, SC
September 25-26, 2007
HHH
d/dt 0
H2H2
Methods for measuring mechanical properties
of structural steels in hydrogen
d/dt > 0 d/dt > 0 strength of materials: UTS, YS, f, RA
H2HH H H2H H2H H2H HH2H H2
fracture mechanics: KIH, KTH
HH H
H H
H H
H
H H
d/dt 0
H
H H
H
H
H HH
H
H
HH H
H
H2H2 H2 H2 H2
H2
Tensile Testing Carbon Steel in H2
Extrusion Direction/ L-C Orientation
T
T
Base HA
Z
Weld
Wel
d
Alloys: 106 Grade B Multi-pass SMAW w/out stress
relief Specimens machined in 3
conditions: Base metal, Weldand HAZ
Orientation: L-C Atmosphere: 10.3 MPa H2,
ambient pressure Air Strain Rate: 10-4 /sec # of samples per matrix point:
6 Argon purge followed by H2
pressurization Soak (30 min.) at pressure Test to failure
Tensile Properties
Mechanical Properties Average Properties
0.2%Yield (MPa)
Dev. UTS (MPa)
Dev. Elong. at
failure
Dev. Reduction of
Area (%) Dev.
Air 355.3 15.8 484.6 8.6 0.29 0.04 68.6 1.5
Bas
e
H2 357.1 32.9 486.4 17.8 0.19 0.04 30.8 5.0
Air 343.0 20.0 490.4 9.0 0.28 0.01 74.9 2.2
Wel
d
H2 350.0 16.1 480.9 12.0 0.21 0.02 30.5 6.4
Air 349.3 20.8 482.3 7.6 0.27 0.04 71.0 1.7
HA
Z
H2 338.2 18.5 475.5 9.6 0.19 0.04 30.4 6.8
Fractography
Fracture Surface
Area
Air Hydrogen HAZ Samples
Reduction of Area reduced when tested in hydrogen
Crack propagation resistance of X100 steel
Alloy composition C Mn Si P S Ti V Ni Cu Mo Cr
0.073 1.86 0.11 0.009
Measurement of sustained-load cracking thresholds
wedge opening load (WOL) cracking threshold specimen
Loading bolt
Load cell
13 mm
strain gage leads (Excitation and DAQ)
Load
(P)
Po Ko
PTH KTH
incubation time= f(environment, Ko)
Time in H2
Specimen loaded to Ko>KTH using bolt while contained in glove box (Ar with ~1 ppm O2)
Loaded specimen exposed to H2, crack extends after incubation time
Crack arrests at K=KTH
High-pressure H2 gas severely degrades crack
propagation resistance of X100 steel
Stre
ss in
tens
ity fa
ctor
, K (
ksii
n1/2)
X100 in air H2 gas pressure (MPa)
0 25 50 75 100 125 150 400
150
KJIc Ko KTH
X100 Steel C-L orientation 25 oC
no crack extension after >1000 hrs
150
100 100
50 50
0 0
1/2 )
Stre
ss in
tens
ity fa
ctor
, K (
MP
am400 350
350 300
300 250
250 200
200
X100 in 15 kpsi H2 gas
0 5 10 15 20 25 H2 gas pressure (kpsi)
Images of fracture surfaces from X100 tested in
H2 show long crack lengths and delaminations
H2-assisted crack extension from precrack
a/W=0.94
http:a/W=0.94
FEM calculations verify K solution and demonstrate
K dominance for long cracks in WOL specimen
a / W = 0.941 KI
VmE / W1/ 2
0.18
0.16
0.14
0.12
0.1
0.08
0.06
0.04
0.02
0 0.5 0.6 0.7 0.8 0.9 1
a / W ASTM E1681: KI = 57.5 MPa m FEM elastic: KI = 63.8 MPa m FEM plastic: J = 16.0 kJ/m2
Calculations by M. Dadfarnia and P. Sofronis JEKIJ = 2 KIJ = 62.2 MPa m1
ASTM E1681
FEM
Incubation time for crack extension depends on Ko
Initial stress intensity factor, Ko (MPam1/2)
0 30 60 90 120 150 2700
0 30 60 90 120 150
cracking in H2 no cracking in H2
X100 Steel WOL specimens 15 kpsi H2 gas 25 oC
no crack extension after >2500 hr
KTH In
cuba
tion
time
(hr)
2400
2100
1800
1500
1200
900
600
Load
(P)
Po Ko
PTH KTH
incubation time
Time in H2 300 0
Initial stress intensity factor, Ko (ksi in1/2)
Measurement of sustained-load cracking threshold in H2 cannot be assured unless crack propagates
Crack branching may account for no crack extension
in specimens with Ko>KTH
Incu
batio
n tim
e (h
r)
SA 372 Gr. J in 15 kpsi H2
Initial stress intensity factor, Ko (MPam1/2)
0 30 60 90 120 150 2700
2400
2100
1800
1500
1200
900 X100 in 15 kpsi H2 600
300
0 0 30 60 90 120 150
Initial stress intensity factor, Ko (ksi in1/2)
precrack
cracks propagated in H2
cracking in H2 no cracking in H2
X100 Steel WOL specimens 15 kpsi H2 gas 25 oC
no crack extension after >2500 hr
KTH
Low-alloy steels for pressure vessels:
sustained-load cracking thresholds in H2 gas
Yield strength, Sy (ksi) 87 92 97 102 107 112 117 122 127 132 137
160
600 650 700 750 800 850 900 950
Cr-Mo Steels WOL specimens
4130 steel
100 MPa H2 gas
25 oC DOT 3AAX
Loginow and Phelps, Corrosion, 1975
4145 steel 4147 steel
SA 372 Grade J
DOT 3T
80 80
60
60
140
140 120
120
1/2 )
(M
Pa
mK
TH
100 100
(ksi
in1/
2 )K
TH
4040
2020
0 0
Yield strength, Sy (MPa)
Thresholds in modern Cr-Mo steels are higher than thresholds in older steels
Stainless steels for piping: tensile fracture results
for specimens precharged in hydrogen gas
316 stainless steels with ~140 wppm hydrogen
Tensile fracture resistance of H-precharged specimens is sensitive to nickel content but not strain hardening
Stainless steels tested in hydrogen gas vs precharged show different tensile fracture behavior
H2 H2
H2
H2 H2
H2
H
H H H
HH
H H
0
20
40
60
80
100R
educ
tion
of A
rea,
RA
(%)
1 1 2
2 3
4
3
4
3
4
Internal Hydrogen 3 - air 4 - H-precharged
External Hydrogen 1 - 69 MPa He 2 - 69 MPa H 2
A-286 San Marchi et al., SAE World Congress, 2007
S (MPa) = 720 850 790 810 810 y
Stable stainless steels may not exhibit loss in RA when strained in H2 due to limited H uptake
System for measuring fatigue crack growth
in high-pressure H2 nearly complete
vessel on mechanical test frame
gas manifold
Pressure vessel can contain H2 gas up to 20 kpsi (138 MPa)
What is metric for qualifying steels for H2 pipelines?
Tech. Ref. Hydrogen Compatibility of Materials, www.ca.sandia.gov/matlsTechRef
10-1
10-2
10-3 X70 (6.9 MPa H2) X42 (6.9 MPa H2) X60 (6.9 MPa H2) A516 (6.9 MPa H2)
10-4 1020 (7.0 MPa H2) SA 105 (6.9 MPa H2) X70 (6.9 MPa N2)
10-5 X42 (6.9 MPa N2) X60 (air) A516 (air) 1020 (air)
10-6 SA 105 (34.5 MPa He)
10-7
Stress intensity factor range, K (MPam1/2) 1 10 100
Fatig
ue c
rack
gro
wth
rate
, da/
dN (
mm
/cyc
le)
Carbon Steels frequency = 0.1 to 1 Hz load ratio = 0.1 to 0.15 Fatigue crack growth rates in
H2 greater than growth rates in air or inert gas
Fatigue crack growth rates in H2 vary by factor of 10
What is acceptable fatigue crack growth rate?
Assess fatigue crack growth data by conducting
fracture mechanics analysis of pipeline
Assume pipeline subjected to pressure cycling
p
Ri Ro
aot
p pR=pmin/pmax~0.1 R=pmin/pmax~0.8 pmax pmax
pmin
pmin time time
Calculate number of pressure cycles for crack to reach critical depth
calculated from threshold for sustained-load cracking, KTH, or fracture toughness, KIH ac
a
ao number of pressure cycles, N Nc
X
calculated from da/dN vs K relationship
Assume steel, dimensions, and maximum
pressure are similar to current H2 pipelines
p
Ri Ro
aot
Data for Air Products H2 pipelines (ASME, Hydrogen Standardization Interim Report, 2005)
API 5L X42 steel, manufactured with ERW Inner radius, Ri = 6.0 in Wall thickness, t, from 0.219 to 0.500 in
Maximum operating pressure, p = 1500 psi
Assume existing defect with depth ao and length parallel to pipeline axis Simulates defect along manufacturing
(seam) weld
Select appropriate material property data Fa
tigue
cra
ck g
row
th ra
te, d
a/dN
(in
/cyc
le)
10-2
Cialone et al., Met Trans A, 1985
10-3
10-4
10-6
10-5 1000 psi H2, R=0.1 1000 psi N2, R=0.1 1000 psi H2, R=0.8 1000 psi N2, R=0.8
10-7
Stress intensity factor range, K (ksiin1/2)
10-8
1 10 100
API 5L X42 Steel frequency = 1 Hz
da/dN=(3.55x10-11)K3.83
da/dN=(2.51x10-12)K6.56
Need da/dN vs K for X42 ERW in 1500 psi H2 gas assume data for 1000 psi H2
gas representative assume data for weld and base
metal are similar Need da/dN vs K at relevant
R ratio and frequency pressure cycling parameters
for pipeline unknown use data at R=0.1 and
frequency=1 Hz Need fracture toughness for
X42 ERW in 1500 psi H2 gas KIH=44 ksiin1/2 for weld HAZ in
1000 psi H2 (Cialone et al., ASTM STP 962, 1988)
Fatigue crack growth rates similar in base
metal, fusion zone, and HAZ for X60 steel
Fatig
ue c
rack
gro
wth
rate
, da/
dN (
mm
/cyc
le)
Wachob, NASA-CR-166334, 1981
10-2
10-3 base metal (6.9 MPa H2) SA weld FZ (6.9 MPa H2) SA weld HAZ (6.9 MPa H2) base metal (air) SA weld FZ (air) SA weld HAZ (air)
10-4
10-5
1/2)Stress intensity factor range, K (MPam
X60 Steel frequency = 1 Hz load ratio = 0.15
1 10 100
Calculate number of pressure cycles to reach
critical crack depth for three cases
cycle 1 cycle 2
p
Ri Ro
aot
K1=f(ao,p,Ro,Ri,t) K2=f(a1,p,Ro,Ri,t) p
1500 psi
150 psi cycles
a=(2.51x10-12)K1 6.56 a1=ao+a
a=(2.51x10-12)K2 6.56 a2=a1+a
ao a1 a2
Case 1: t=0.330 in (h=65% SMYS) and ao/t=0.10
Case 2: t=0.500 in (h=43% SMYS) and ao/t=0.10
Case 3: t=0.500 in (h=43% SMYS) and ao/t=0.05
http:ao/t=0.05http:ao/t=0.10http:ao/t=0.10
0.3
0.7
0.8
Number of pressure cycles vs crack depth
relationships: H2 compared to inert gas
crac
k de
pth
/ wal
l thi
ckne
ss, a
/t
0.6
0.5
0.4
0.2
0.1
0.0 0 50000 100000 150000 200000 250000
hydrogen t=0.50 in h=43% SMYS
nitrogen t=0.50 in h=43% SMYS
X42 Pipeline hydrogen or nitrogen gas pressure cycle = 150 - 1500 psi inner diameter = 12 in
X
X critical crack depth calculated from KIc
critical crack depth calculated from KIH
10-2
Fatig
ue c
rack
gro
wth
rate
, da/
dN (
in/c
ycle
)
10-3
10-4
10-5
10-6
10-7
1 10 100
API 5L X42 Steel frequency = 1 Hz
da/dN=(3.55x10-11)K3.83
da/dN=(2.51x10-12)K6.56
1000 psi H2, R=0.1 1000 psi N2, R=0.1
10-8
number of pressure cycles Stress intensity factor range, K (ksiin1/2)
Number of cycles to critical crack depth reduced
by about factor of 40 in H2 compared to inert gas
0.3
0.7
0.8
Number of pressure cycles vs crack depth
relationships in H2 for varying dimensions
crac
k de
pth
/ wal
l thi
ckne
ss, a
/t
0.6
0.5
0.4
0.2
0.1
0.0 0 5000 10000 15000 20000 25000 30000 35000 40000 45000
case 1 t=0.33 in h=65% SMYS
case 2 t=0.50 in h=43% SMYS
case 3 t=0.50 in h=43% SMYS
X42 Pipeline hydrogen gas pressure cycle = 150 - 1500 psi inner diameter = 12 in
X
X X
critical crack depths calculated from KIH
10-2 API 5L X42 Steel frequency = 1 Hz
1000 psi H2, R=0.1
case 1 initial K
case 2 initial K
case 3 initial K
da/dN=(2.51x10-12)K6.56
10-8
1 10 100 Stress intensity factor range, K (ksiin1/2)number of pressure cycles
Fatig
ue c
rack
gro
wth
rate
, da/
dN (
in/c
ycle
)
10-3
10-4
10-5
10-6
10-7
Number of cycles to critical crack depth is sensitive to applied stress and initial crack depth
http:da/dN=(2.51x10-12)K6.56
ASME applies safety factors on critical crack depth
and number of cycles to critical crack depth
a/ac
1.0
0.25
critical crack depth critical crack depth
M.D. Rana et al., ASME PVP, 2007 cycles to
Safety factors: a/ac = 0.25 N/Nc = 0.5
0.5 1.0 N/Nc
0.500
0.500
0.500
0.330
t (in)
43% SMYS
43% SMYS
43% SMYS
65% SMYS
h
112,545125,2600.10inert
20,69038,7900.05H2
2,6602,7350.10H2
5351700.10H2
0.5 x Nccycles to 0.25 x ac
ao/tEnvironment
Lifetime of steel hydrogen pipeline depends on
both material properties and structural design
Fatigue crack growth law da/dN=CKn in stage II
Fracture toughness or threshold for sustained-load cracking does not significantly affect cycles to critical crack depth
Wall thickness affects stress (K)
Initial crack depth depends on detection limit of NDE
Operating pressure affects stress (K) and material properties
Fatigue crack growth rates depend on
load cycle profile
da/d
N (
mm
/cyc
le)
Walter and Chandler, Effect of Hydrogen on Behavior of Materials, 1976
0.04 SA 105 Steel H2 pressure = 100 MPa K = 44 MPam
0.03 load ratio = 0.1
0.02
0.00 0 200 400 600 800 1000 1200 1400
Seconds Per Cycle
0.01
0.5 sec ramp 100 sec ramp
Mechanical Properties ofStructural Steels in HydrogenMethods for measuring mechanical propertiesof structural steels in hydrogenTensile Testing Carbon Steel in H2Tensile PropertiesFractographyCrack propagation resistance of X100 steelMeasurement of sustained-load cracking thresholdsHigh-pressure H2 gas severely degrades crackpropagation resistance of X100 steelImages of fracture surfaces from X100 tested inH2 show long crack lengths and delaminationsFEM calculations verify K solution and demonstrateK dominance for long cracks in WOL specimenIncubation time for crack extension depends on KoCrack branching may account for no crack extensionin specimens with Ko>KTHLow-alloy steels for pressure vessels:sustained-load cracking thresholds in H2 gasStainless steels for piping: tensile fracture resultsfor specimens precharged in hydrogen gasStainless steels tested in hydrogen gas vs prechargedshow different tensile fracture behaviorSystem for measuring fatigue crack growthin high-pressure H2 nearly completeWhat is metric for qualifying steels for H2 pipelines?Assess fatigue crack growth data by conductingfracture mechanics analysis of pipelineAssume steel, dimensions, and maximumpressure are similar to current H2 pipelSelect appropriate material property dataFatigue crack growth rates similar in basemetal, fusion zone, and HAZ for X60 steelCalculate number of pressure cycles to reachcritical crack depth for three casesNumber of pressure cycles vs crack depthrelationships: H2 compared to inert gasNumber of pressure cycles vs crack depthrelationships in H2 for varying dimensionsASME applies safety factors on critical crack depthand number of cycles to critical crack depthLifetime of steel hydrogen pipeline depends onboth material properties and structural designFatigue crack growth rates depend onload cycle profileFatigue crack growth rates depend onload cycle profile
Button26: