Cold work effects on sulfide stress cracking of pipeline steel exposed to sour environments

Cold work effects on sulfide stress cracking of pipeline steel exposed to sour environments

Corrosion Science, Vol. 34. No. 1, pp. 61-78, 1993 Printed in Great Britain. 0010-938X/93 $5.00 + 0.00 O 1992 Pergamon Press Ltd C O L D WORK EFFECT...

1013KB Sizes 4 Downloads 11 Views

Corrosion Science, Vol. 34. No. 1, pp. 61-78, 1993 Printed in Great Britain.

0010-938X/93 $5.00 + 0.00 O 1992 Pergamon Press Ltd

C O L D WORK EFFECTS ON SULFIDE STRESS CRACKING OF PIPELINE STEEL EXPOSED TO SOUR ENVIRONMENTS H. HUANG and W. J. D. SHAW Department of Mechanical Engineering, The University of Calgary, Calgary, Alberta, Canada T2N 1N4 Abstract--Cold work effects on sulfide stress cracking of a pipeline steel were examined for exposure to a sour gas environment. Cold worked steel was found to be sensitive to hydrogen embrittlement. The fracture toughness decreases with increasing cold work. Sufficiently low values of fracture toughness may be achieved to promote plane strain fracture even in relatively thin laboratory specimens after a steady state level of hydrogen has occurred in the material. Sulfide stress cracking failure of steel with low amounts of cold work (<30%) is by transgranular cleavage fracture, while heavily cold worked steel shows secondary cracks and microvoid coalescence. The results indicate that sulfide stress cracking of the steel is a mixture of anodic stress corrosion cracking (SCC) and hydrogen embrittlement. The anodic SCC mechanism is mainly promoted by carbon dioxide and a high level of chloride. INTRODUCTION UNSUCCESSFUL application of tubular steels, A P I N80 and P l 1 0 grades, to sour gas wells in the early 1950s I has led to extensive investigations concerning sulfide stress cracking (SSC). These investigations have b e e n c o n d u c t e d to evaluate the p e r f o r m ance of steel products in sour gas environments. 2-1t It is generally accepted that SSC of high strength steel results from h y d r o g e n a t o m accumulation in the steel, although some earlier experiments show that anodic polarization p r o m o t e d cracking. 12-13 T h e r e is still considerable confusion with respect to understanding failure of low strength steel in sour gas environments. For low strength, low nickel alloys, D u n l o p 14 concluded that the mechanism of SSC is typical anodic stress corrosion cracking (SCC) failure. M o t o d a et al. 15 s h o w e d that SSC is caused by anodic SCC in dilute H2S solutions regardless of the presence of 1% nickel, and that the addition of nickel m a k e s the steel m o r e sensitive. Wilhelm and K a n e 16 showed that the failure m e c h a n i s m of duplex stainless steel in H2S/Cl- e n v i r o n m e n t s shifted f r o m h y d r o g e n e m b r i t t l e m e n t to anodic S C C when the t e m p e r a t u r e rose to 177°C. O n the o t h e r hand, T u r n et al. 11 indicated that the h y d r o g e n embrittlement m e c h a n i s m is responsible for SSC of nickel-free alloy, which has been accepted by other investigators. 2,7,10 Some w o r k had been c o n d u c t e d with respect to corrosion b e h a v i o r of steel affected by cold w o r k conditions. 17-2o T h e r e is very little information available in the literature regarding the effects of cold w o r k on SSC behavior of steel. 9'21-25 A l t h o u g h this area has a significant degree of practical importance, very little concrete information has been d o c u m e n t e d . Generally the m e c h a n i s m for the effect of cold work on the failure b e h a v i o r is not yet clear. H y d r o g e n adsorption on the electrode surface is an intermediate step during Manuscript received 25 November 1991; in amended form 8 May 1992. 61

62

H. HUANGand W. J. D. SHAW TABLE1. CHEMICALCOMPOSITIONOFAISI 1020 STEEL(wt%)

C

0.19

Si

Mn

P

S

Ni

Cr

Mo

Cu

0.21

0.93

0.01

0.04

0.17

0.13

0.01

0.14

AI

Fe

0.03 Balance

h y d r o g e n e v o l u t i o n w h e n steel is exposed to acid e n v i r o n m e n t s . 26'27 H y d r o g e n atoms a d s o r b e d o n the electrodic surface can either react with o n e a n o t h e r to form h y d r o g e n gas, or diffuse into the steel a n d p r o m o t e e m b r i t t l e m e n t . Highly localized, fine scale plastic d e f o r m a t i o n has b e e n directly associated with h y d r o g e n i n d u c e d brittle fracture .28 T h e p r e s e n c e of i n t e r n a l h y d r o g e n leads to strain localization in the form of d e n s e dislocation structures which in t u r n e n h a n c e brittle fracture of steel. 29 Cold work is expected to e n h a n c e h y d r o g e n - i n d u c e d failures by facilitating the particle matrix s e p a r a t i o n , void i n i t i a t i o n a n d growth, local stress c o n c e n t r a t i o n caused by plastic i n c o m p a t i b i l i t y a n d particle cracking. 24 S i m u l t a n e o u s tensile straining a n d charging with various fugacities of h y d r o g e n were f o u n d to e n h a n c e void n u c l e a t i o n in spheroidized 104521'22 a n d 109523 steels. Some d e f o r m a t i o n m o d e s have b e e n f o u n d to exert beneficial effects o n the resistance of steel to h y d r o g e n e m b r i t t l e m e n t . F o r e x a m p l e , m o d e r a t e cold work (up to 15% r e d u c t i o n of thickness for a 0.16% C steel a n d 7 % tensile straining for a 0.15% C low C r - N i - C o steel 24 were f o u n d to decrease the susceptibility to h y d r o g e n i n d u c e d d e g r a d a t i o n , while larger d e f o r m a t i o n causes void f o r m a t i o n a n d growth. I n c r e a s e d resistance to h y d r o g e n e m b r i t t l e m e n t was r e p o r t e d 24 for a low alloy steel with cold w o r k of 50% a n d 90%. This beneficial effect was a t t r i b u t e d to a decrease in corrosion a n d h y d r o g e n p e r m e a t i o n rates as a result of compressive stresses in the subsurface layer of the steel. T h e m a i n p u r p o s e of this study is to investigate the effects of cold work o n SSC b e h a v i o r of p i p e l i n e steel in a sour gas e n v i r o n m e n t . A n a t t e m p t is m a d e to d e t e r m i n e the r e l a t i o n s h i p of a n o d i c SCC a n d h y d r o g e n e m b r i t t l e m e n t of cold w o r k e d steel. A d d i t i o n a l l y this work provides some q u a n t i t a t i v e i n f o r m a t i o n that will help in o b t a i n i n g a b e t t e r u n d e r s t a n d i n g d u r i n g the design process. EXPERIMENTAL METHOD The material used in this study was a commercial ferritic/pearlitic AISI 1020 steel in the hot rolled condition and is referred to as 0% cold worked material. The composition of the steel is listed in Table 1. The grain size of the hot rolled material was measured to be ASTM No. 8.38 according to ASTM E 112.3o The cold rolled conditions were produced by reducing various thickness of the hot rolled material to a finished thickness of 6.35 mm in percentages from 10% to 70% occurring in steps of 10%. The tensile specimens were subsized being 200 mm in overall length with threaded ends and a reduced section of 3.2 mm in diameter. The gage length was 27 mm. The elongation data were adjusted to a standard gage length of 50 mm according to Barba's law.31 Compact fracture toughness specimens 25.4 mm in width and 6.35 mm in thickness were taken from the L-T orientation. All compact fracture toughness specimens were fatigue precracked below Kma× = 30 MPa mI/2 in ambient air, according to ASTM E-399.32The configurations of the tensile specimen and the compact fracture toughness specimen are shown in Fig. 1. The aqueous and gaseous environment used for this study was based upon analysis of some problem wells in Alberta. 33 The aqueous components (in mg 1-1) are as follows: 48,500 Na÷ , 14,250 Ca2÷ , 1045 Mg2÷, 91,500 CI , 180 HCO3 and 150 SO4z-. The gaseous components consist of 34% (vol) hydrogen sulfide, 10% (vol) carbon dioxide with the balance being methane at 1 atm pressure. This simulated salt content and gaseous components are somewhat simpler than that found in the field but the major constituents affecting the material have been duplicated. This environment has been used for investigating

Sulfide stress cracking of pipeline steel

63

corrosion mechanisms in previous studies 33 and is referred to as the sour gas environment in this paper. The tests were performed in the brine environment saturated with the H2S/CO2/CH 4 mixture, which was bubbled through at a flow rate of 0.4 1m i n - 1, at 1 atm pressure and 25°C temperature. A widely used laboratory environment for determining material resistance to SSC is that recommended by N A C E , 34 being 5% NaCI, 0.5% acetic acid and saturated with hydrogen sulfide. Comparisons of SSC results after testing in both the sour environment and the N A C E solution are made.

101.6

30.4 16.0 - ~

I_

19.6

25.4

/

_1

!!!!!!!H!!!l!!/!/!!!A AlL DIMENSIONS ARE IN

¢6.55

mm

(a)

L

31.75

i-

"--=""

"='--" 6.35

()

c 7-00

30.50

4.70

L

¢6.35

THICKNESS= 6.35 ALL DIMENSIONS ARE IN mm

(b) FIG. 1.

The configuration of specimens: (a) tensile specimen: (b) compact fracture toughness specimen.

64

H. HUANG and W. J. D. SHAW

A batch of tensile and fracture toughness specimens was prepared for later testing by first immersing, unloaded, in a reaction kettle that contained the sour environment. Selected immersion times of 0, 24, 72 and 216 h were chosen to determine the time required for a steady state condition of hydrogen to occur in 0% and 30% cold worked tensile specimens while in an unloaded condition. An immersion time of 72 h was chosen as it was found that a steady state hydrogen level was achieved for all conditions of this material. At the end of the 72 h time period the specimens were quickly removed and tested by loading in a regular manner according to tensile and fracture mechanics test methods. Slow strain rate (SSR) tests were conducted using tensile type specimens. These specimens were mounted in a plastic cell with the sour environment surrounding them and sealed at either end. The loading grips were external to the cell. Tests were performed at displacement rates ranging from 2.7 x 10 -7 to 2.7 × 10-4 cm s- 1. This produced engineering strain rates of 10-7 to 10-4 s- 1 for the smooth tensile specimens (prior to necking). A strain rate of 1 × 10-Ss - l was chosen to investigate the sensitivity of the pipeline steel to SSC with respect to the various cold work conditions. Recording of load versus specimen extension was conducted for the tensile and SSR tests. In SSC tests the loading pins were covered with plastic which prevented any galvanic contact between the grips and the specimen. The specimen was allowed to reside at its rest potential. In order to investigate the contribution of gaseous environmental components to SSC of cold worked steel, the following gaseous environments were used in conjunction with the brine solution: 34% H2S/10% CO2/56% CH4; 10% CO2/90% CH 4 and 100% HzS. In addition the NACE solution was also included. This environmental comparison was undertaken using SSR tests in order to try and distinguish between the mechanisms of anodic SCC and hydrogen embrittlement. Testing was similarly conducted on tensile specimens that were first immersed unloaded for 72 h prior to SSR testing. Two specimens of both 20% and 40% cold work were used for this evaluation. Once the prior immersion period was completed the specimens were quickly transferred from the reaction kettle to the environmental cell and subsequently tested under SSR loading. Compact fracture toughness specimens were used for determining the change of fracture toughness properties for cold worked steels using a slow crack opening displacement (COD) rate technique. The change in fracture toughness properties reflects the susceptibility of cold worked steel to SSC. Crack length during testing was determined by the back face strain method which previously was modified for SCC testing in aqueous solutions. 35 The loads and the corresponding strains on the back face of the specimens were continually monitored using an x-y recorder. The stress intensity factor used for the compact fracture toughness specimens was calculated using equation (1),

K = ( B ~ ) f (a/w)

(1)

where K is the stress intensity factor, P is the applied stress load, B is the specimen thickness, W is the specimen width, a is the crack length which is measured from the load line, andf(a/w) is a geometric factor as defined in ASTM E-399. 32

EXPERIMENTAL

RESULTS

Mechanical properties of cold worked materials The tensile properties and hardness values of the material for various cold work c o n d i t i o n s a r e s h o w n i n T a b l e 2.

Unstressed prior immersion tests No apparent difference in the yield strength of tensile specimens after a period of unstressed prior immersion was found. The ultimate strength and uniform strain after unstressed prior immersion decreases slightly. These tensile tests were conducted in an air ambient according to the tensile test method, ASTM E-8. The strain to fracture or percentage elongation was greatly influenced by the sour environment. T h e c h a n g e i n p l a s t i c p r o p e r t i e s is o f t e n u s e d as a c r i t e r i o n f o r t h e s u s c e p t i b i l i t y o f

Sulfide stress cracking of pipeline steel TABLE 2.

Cold work (%) 0 10 20 30 40 50 60 70

65

MECHANICAL PROPERTIES AND HARDNESS OF COLD WORKED STEEL

oy (MPa)

tru (MPa)

e, (%)

ef (%)

Hardness (HRC)

382 601 726 808 850 889 926 985

581 696 740 820 855 896 930 1019

14.6 1.6 0.65 0.47 0.36 0.42 0.42 0.30

27.7 15.5 12.6 12.2 10.6 8.5 8.7 7.9

5.7 13.1 16.5 19.0 21.3 23.5 24.4 25.1

steel to SSC. 11 The strain to fracture decreases with an increase in immersion time (Fig. 2). The 30% cold worked steel shows a consistent drop in ductility with increasing immersion time. The 0% cold worked steel is not very sensitive to hydrogen embrittlement as shown in Fig. 2. It does lose some ductility but the loss is small compared to the 30% cold work condition. The loss in ductility is related to exposure time due to hydrogen diffusion and subsequent concentration of hydrogen within the steel. The results in Fig. 2 for both hot rolled steel and cold worked steel suggest that the immersion time for hydrogen steady state in the steel is approximately 35-70 h. The steady state concentration of hydrogen for various cold work conditions after 72 h immersion is shown in Table 3. 36 Figure 3 shows the loss in ductility of 72 h prior immersion on tensile specimens for various conditions of cold work. The loss of ductility of the 0% cold worked steel is 39.9%, while the loss of

Ambient Tensile Tests 35

I

'

I

,

i

'

,

I

i

i o

0 %

Cold

Worked

S~el

" i

30

#

"

50% Cold Worked Steel

25

~

zo

=

(5

o

n I

0

50

I

lO0

i

I

(50

i

I

200

i

250

Exposure Time, hrs

FIfi. 2. The relationship of ductility of tensile specimens vs unstressed prior immersion time.

66

H. HUAnGand W. J. D. SHAW TABLE 3.

MOBILE HYDROGEN CONCENTRATION IN COLD WORKED

STEEL AFTER 7 2 h UNSTRESSED PRIOR IMMERSION

Cold work (%)

CO×

mole Hcm - 3

10 - 6

0 10 20 30 40 60 70

11.8 13.7 18.9 22.7 29.1 34.7 37.9

ductility of 70% cold worked steel reaches 88.6%. The results indicate that the sensitivity of the cold worked steel to hydrogen embrittlement increases steadily with an increase in cold work. Figure 4 shows that the fracture toughness decreases after 72 h unstressed prior immersion as a result of the hydrogen steady state in the material. This figure shows that when the cold work reaches a magnitude of 20-30%, the decrease in fracture toughness becomes nearly constant, not changing significantly with further amounts of cold work. According to the requirements of A S T M E-39932 for plane strain fracture, the fracture m o d e of cold worked steel was determined to be plane strain. This was based on pop in or m a x i m u m load fracture conditions. The combination of cold work and hydrogen state causes the failure m o d e of fracture mechanics specimens of cold worked steel to change from plane stress to plane strain. The value of 96 MPa m 1/2 was measured in this study to be the stress intensity of 10% cold Ambient Tensile Tests I

1

I

I

I

I

I

Cold Worked, NO immersion.

o AS

z= P r i o r

Prior

Immersion

"=Z 20

o

IO

0

t0

~

D

30

40

50

60

70

80

Cold Work, %

FIG. 3. The effects of cold work on the ductility of tensile specimensand the effect of 72 h unstressed prior immersion.

Sulfide stress cracking of pipeline steel

67

worked steel for plane strain as affected by 72 h unstressed prior immersion. This value represented the initial position where valid plane strain conditions were found to occur. S S R tests

Figure 5 shows that the dutility of 0% cold worked steel under SSR testing decreased with a decrease in strain rate over the range of strain rates from 1 × 10 -7 to 1 × 10 -4 s -1. Strain rate was based upon specimen extension as a function of time c o m p a r e d to initial gage length. Examination of specimens showed that multiple cracks formed on the side surface of specimens tested under slow strain rates in the range from 1 × 10 - 6 t o 1 × 10 - 4 s - 1 . The m a x i m u m occurrence of multiple cracks occurred around 1 × 10 -5 s - t . Further decrease in the strain rate to 1 × 10 -7 s -1 resulted in an absence of multiple cracks forming on the sides of specimens. Also c o m p a r e d on Fig. 5 are results of tests conducted in ambient air. SSR tests were also conducted on tensile specimens with various cold work conditions at the strain rate of I × 10 -5 s -1. Figure 6 shows the ductility of the cold worked materials tested under SSR loading in various environmental conditions. The results for materials tested in ambient air are plotted on Fig. 6 for comparison. The loss in ductility is a result of the combined effect of both cold work and environment. The multiple cracks on the side surface of the specimens do not form when cold work reaches 30% or greater. The 100% H2S environment causes a higher loss of ductility than when tested in the 34% H2S sour gas mixture as shown in Fig. 6. There is an absence of multiple cracks on the side surface of specimens due to the presence of 100% H2S in the brine solution for all conditions of cold work. The

35o

Ambient Fracture Toughness Tests

'

'

'

D AS Cold Worked, No Prior Immersion,

3 0 0 F-

Prior immersion,

i

._~ i

E Z50 o o. =E 200

150

£ E I,I.

]00

50

0

FIG. 4.

I 0

I tO

I 20

i 30

I 40

I 50

I 60

I 70

80

Cold Work, % The effects of cold work on the fracture toughness and the effect of 72 h unstressed prior immersion.

68

H. HUANG and W. J. D. SHAW Slow e

Strain

I

Rate

'

I

Tensile

Tests

'

1

'

I

o

3O

o

./

# c

20 U)

A

e ~0 a Ambienf Air A Sour Gel Solution 008. --

'""

t

I

L

-7.0

I

i

I

-6.0

i

-5.0

I

i

-4.0

-:5.0

Slow Stroin Rote, Loo, ram/ram s

F=G. 5.

The relationship of SSR vs ductility of 0% cold worked steel.

Slow Strain Rate Tensile Tests

"

30

J

t0

"

011 0

. Ambienl Air

o

a

~

n

I

I

I

T

T

t0

20

30

40

50

60

70

80

Cold Work, %

FnG. 6.

The environmental effects on the ductility of cold worked steel al a strain nate of 1 × 10-Ss u

Sulfide stress cracking of pipeline steel

69

Tensile Test Comparison of Two Cold Worked Conditions

12.6 ~2

t0.6 Io c

9.4 Ambient Air, Regular Tensile Test, No Prior Immersion.

86.5

6-

SSR Test, No Prior Immersion.

.4

!,

U.

3.t

2-

17 14

72 hr Prior Immersion, then Reoulor Tensile Test in Ambient Air. 72 hr Prior Immersion * $SR Test.

0 Cold Work,

FI~. 7.

%

The effects of SSR loading and unstressed prior immersion on the ductility of cold worked steel.

results of the slow strain rate tests on cold worked steel exposed to the N A C E solution can also be found in Fig. 6. The results from the tests in the N A C E solution indicate that there is a greater decrease in ductility compared with those exposed to the 34% H2S sour gas mixture and the 100% H2S environment. Multiple cracks do not form on the side surface of specimens tested in the N A C E solution. The p H value of the N A C E solution is 3, 34 while the pH value of the HzS/COz/CH 4 brine environment was measured to be 4.8. The pH of the 100% H2S brine solution was measured to be 3.6. Testing was similarly conducted under SSR loading for tensile specimens with 20% and 40% cold work. These specimens underwent 72 h unstressed prior immersion before SSR tests were started. The results of these tests can be seen in Fig. 7. Figure 7 also shows results from materials not undergoing prior immersion as well as air environment tests. Similar behavior occurs as compared to that recorded for the other immersion tests and for the SSR tests. However, the ductility values for prior immersion followed by SSR testing show a greater loss than that of either the previous SSR tests or the 72 h prior immersion tensile tests. After unstressed prior immersion the steady state hydrogen level in the steel has been achieved, resulting in a high degree of hydrogen embrittlement. Multiple cracks on 20% cold worked components did not occur for these more stringent tests. This is in contrast to the results obtained for SSR loading without prior immersion. The results in Fig. 8 indicate that, in the absence of H2S, the brine environment with C O 2 / C H 4 c a u s e s less embrittlement for the 0% cold worked and the low cold worked steel compared with those tested in the 34% H2S sour gas environment. No apparent effect on the ductility of the heavily cold worked steel can be found as compared to those tested in ambient air as seen in Fig. 8. The observation of specimens tested in the C O 2 / C H 4 environment indicated that multiple cracks appear on the side surface of specimens for the 0% cold work and low cold work condition.

70

H. HUANGand W. J. D. SHAW Slow Strain Rate Tensile Tests

I

30,

I

I

I

I

o

I

0

I

Ambient Air

#

d P aO

0

~

D

D

t0

D

_

0

o I

0

t

40

I

I

I

I

I

I

ZO

50

40

50

60

70

80

Cold Work, %

FIG. 8. The effects of CO2/CH4 on the ductility of cold worked steel at a strain rate of 1 × 10-Ss -l

The results from the sour environment are also included on Fig. 8 for comparison. The p H of the CO2/CH 4 environment was measured to be 5.5.

Slow COD rate tests Figure 9 indicates the fracture toughness of cold worked steel under a constant slow C O D rate loading resulting from a pin displacement rate of 6.35 × 10 -5 m m s- 1. The change of fracture toughness of the 0% cold worked steel between the ambient environmental test condition and the sour environmental test condition is 11.1%, while the change of 10% cold worked steel is 39.4%. The prior immersion condition resulting in a steady state hydrogen level causes an ever larger reduction of fracture toughness. The results show that cold working and steady state hydrogen level within the steel have pronounced effects on the fracture toughness of this material. GENERAL DISCUSSION Hydrogen effect During exposure of the steel to the sour gas environment, atomic hydrogen is generated on the material surface due to the reaction of HaS reduction. 33'37 The generation of hydrogen depends upon the environment and surface condition of the material while diffusion of hydrogen is affected by the dislocation density, the microstructure of the material and the stress state. As the corrosion rate, or hydrogen reduction increases, more hydrogen is generated and the gradient for hydrogen diffusion increases. H y d r o g e n adsorption and subsequent absorption are enhanced when the surface

Sulfide stress cracking of pipeline steel

71

Compact Fracture Toughness Specimen Tests

350tr 3001-°

i

I

i

I

f

I

I

t~ Ambient Air, Sour Environment, NO Prior Immersion.

E =E

250

3 \ i ~

~

o 72 hr Prior Immersion. then Sour Environmen, Tes,ing.

tJ

3¢ = 200

==

II

o

o

.¢; c~ 150

g

t00 oo

~

~

__~

o

50 0

I 0

1 10

/ 20

I 50

[ 40

I 50

t 60

1 70

80

Cold Work, % FIG. 9.

The fracture toughness of cold worked steel at a slow pin displacement rate of 6.35 x 10 -5 mm s -I.

is free of corrosion products or films. Observations in previous work 36 indicated that increases in cold work cause a steady increase in hydrogen ion reduction. In the acid solution, the hydrogen adsorption process occurs due to the H + reduction. 26 This reaction progresses to hydrogen evolution, or absorption. The hydrogen evolution reaction on steel and iron in aqueous acid and alkaline solution occurs as a result of the coupled discharge electrodic desorption mechanism, 26 or discharge followed by electrodic desorption. 37 The following reaction sequence describes this process in the mixture H2S/CO2/CH 4 solution 36 rds

H2S + e --+ H S - + Had s

(2)

rds

Haas + H2S + e - -+ H2 + H S -

(3)

The hydrogen absorption process can be expressed as abs Had s ~ Habs des

(4)

where rds is the rate determining step, Hods is the hydrogen atom adsorbed on the electrodic surface and Hob S is the hydrogen atom absorbed in the steel. Cold work causes an increase in the interior stored energy and deformation slip steps on the surface which changes the surface energy distribution. These deformation slips on the surface are energetically favorable sites for hydrogen adsorption. With an increase in cold work the surface fraction of hydrogen adsorption increases. Electrochemical studies on cold worked steel have shown that the cathodic hydrogen reduction rate increases with hydrogen coverage which is affected by cold work. 36

72

H. HOANGand W. J. D. SHAW

From equations (2) and (3), it is apparent that hydrogen adsorption on the electrodic surface is a necessary intermediate step during the hydrogen evolution. Hydrogen atoms adsorbed on the electrode surface are a unique source for hydrogen entry into the steel during hydrogen evolution as described in equation (4). Therefore, an increase in cold work which increases hydrogen adsorption on the surface should cause an increase in the gradient of hydrogen entry into the steel, and the hydrogen embrittlement effect will increase. Scully and Moran 38 have indicated that a large hydrogen coverage is created on the bare metal surface under slow strain rate loading. As the dislocation density increases from 1 × 108 to 1 x 1012 cm -2 due to heavy cold working, significant hydrogen trap enhancement due to the interaction with dislocations can be expected at the entanglements of dislocations. 39 The above discussion suggests that an increase in the gradient of hydrogen entry into steel and hydrogen trap enhancement are promoted by cold work. The increase in the yield strength resulting from cold work also acts to enhance hydrogen embrittlement due to stress field amplification at crack tips as discussed by Akhurst and Baker. 4° Therefore it is expected that cold work causes an increase in hydrogen concentration in cold worked steel and makes the pipeline steel more sensitive to hydrogen embrittlement. The increase in hydrogen ion concentration or alternatively H2S concentration which is the major parameter enhances the hydrogen reduction. This is probably the reason why the NACE solution and the 100% H2S environment cause more severe embrittlement than the 34% H2S sour gas environment. As the exposure time increases, more hydrogen atoms accumulate in the steel. The condition of hydrogen charging into the steel is a result of the corrosion mechanism. The time needed for steady state hydrogen levels depends on the diffusion rate and the thickness of the component. 41 The time for a steady state hydrogen level to occur can be estimated by the following expression, assuming that steady state hydrogen concentration at the centre of specimens is greater than 80% of the surface concentration. 41 Dt r~- - 0.4.

(5)

Taking a diffusion coefficient D of 1 × 10 -7 cm 2 s -1 for hydrogen in AISI 1090 carbon steel with 30% cold work, 42 and a radius (r) of 1.6 mm for the tensile specimens, the time (t) to reach saturation is calculated to be approximately 1700 min (about 28 h). This value for the time of steady state hydrogen level in the tensile specimen is different from the experimental value of 35-70 h. Considering the formation of surface films during the unstressed prior immersion, 37 the corrosion rate decreases with an increase in exposure time. It is reasonably expected that the time needed for hydrogen saturation in steel from the experiments is longer than that calculated from the simple theoretical considerations. One possible reason for the slow strain effect on sulfide stress cracking is that hydrogen entry is enhanced by surface deformation during the SSR test once film rupture occurs due to SSR loading. Slow strain rates provide greater time for the diffusion of hydrogen to susceptible sites. Hydrogen embrittlement is observed only if sufficient hydrogen has entered and permeated to deleterious trapping sites within the specimen, as can be seen in Fig. 5. This graph shows that a decrease in SSR results

Sulfide stress cracking of pipeline steel

73

in an increase in embrittlement. This explanation is supported by SSR tests after 72 h unstressed prior immersion allowing a steady state hydrogen level to occur in 20% cold worked steel. These tests showed that an increase in hydrogen as a result of 72 h prior immersion causes the hydrogen embrittlement mechanism for SSC of 20% cold worked steel under SSR testing to become dominant. A n o t h e r possible reason for the SSR effect is the sensitivity of the deformed structure to hydrogen embrittlement, which will be discussed in the section on effects of deformed structure.

C O 2 and brine environment effects

Several cases of carbonate type cracking have recently been reported in equipment. 43'44 A n u m b e r of investigators have shown anodic SCC in low carbon and low strength steel in the presence of carbonate/bicarbonate or a m m o n i u m carbonate solution. 43-46 Figure 8 indicates that C O 2 / C H 4 brine solution with the absence of HzS results in a m o d e r a t e amount of embrittlement on the 0% cold worked steel and the low cold worked steel under SSR loading conditions. Hudgins et al. 46 have shown that anodic SCC of carbon steel can occur under a high CO2 pressure environment. The anodic stress corrosion cracking is normally associated with the presence of chloride and oxygen and is favored by acidic conditions. The brine solution in this study contains a very high level of chloride. The SSR technique is believed to be one of the most severe testing methods for stress corrosion cracking. This may be why the anodic SCC of the 0% cold work steel and the low cold worked steel is found to occur in the CO2/CH 4 brine solution. Multiple crack initiation on the side surface of specimens is believed to be related to the anodic SCC mechanism of sulfide stress cracking of low strength steel due to the combined contribution of CO2 components and brine solution in the sour gas environment. 47 Multiple cracks in general have been found to be associated with pits and trenches, s'14 Pits and trenches result from the interaction of applied stress and corrosion dissolution on the side surface of tensile specimens. These findings have shown that in the anodically polarized specimen, many shallow pits and deep trenches form which subsequently act as initiation sites for multiple cracks. H y d r o g e n atoms produced from the corrosion reaction on the surface tend to prefer to diffuse to the center of tensile specimens due to the presence of triaxial stress under SSR loading condition. As a result of this, crack initiation occurs at the center of the tensile specimen when the hydrogen embrittlement effect is dominant. These arguments are in agreement with the observation found here. There was an absence of multiple cracks on the side surface of the specimens tested in 100% HzS brine solution and the N A C E solution under the same loading conditions. This was a direct result of an increase in the hydrogen embrittlement effect. The occurrence of multiple cracks on the side surface of specimens of the 0% cold worked steel and low cold worked steel at a strain rate of 1 × 10-5 s - t and the combined effect of the CO2 c o m p o n e n t and a high level of chloride in the sour environment suggests that the anodic SCC failure mechanism is present. It can be concluded that C O 2 and a high level of chloride in the sour gas environment causes the 0% cold worked steel and low cold worked steel to be sensitive to anodic SCC, which is in addition to the hydrogen embrittlement effect. Heavily cold worked steel is more sensitive to hydrogen embrittlement.

74

H. HUANGand W. J. D. SHAW

The effects of deformed structures The internal microstructure of cold worked material has evolved using microbeam X-ray studies and electron microscopy of thin films. 48 Typically it is characterized by a series of cells where the walls are defined by dislocation tangles within grains. Recent results on materials with small amount of cold work ( < 5 % ) 49 have shown that the yield strength of cold worked steel has a more direct influence on the SSC as opposed to the amount of cold work itself. An increase in yield strength due to cold work depends on distortion of the microstructure. The distorted structure of cold worked steel is identified with dislocation piles, where mobile hydrogen is easily attracted .39 These tangled areas resist further movement of dislocations and cause an increase in strength along with a subsequent decrease in ductility. The combination of hydrogen accumulation and distorted structures caused by cold work may therefore be expected to be related to crack initiation and propagation. For the cold worked materials, the change of the SSC failure of the compact fracture toughness specimen from a plane stress to a plane strain condition has serious consequences with respect to field conditions. Since plane strain is an engineering brittle failure mechanism it means that once a crack has initiated catastrophic failure will easily follow at values well below the yield strength of the material. On the microscopic scale, the fracture appearance of moderately cold worked steel, 20% cold work, shows almost complete brittle cleavage failure, as can be seen in Fig. 10. A small amount of microvoid coalescence also occurs in a few specific locations. Under low cold work conditions, the hydrogen accumulation within the pile of dislocations due to stress gradients and the interaction with dislocations may induce cracks within the pile of dislocations. Once cracks initiate, they propagate along specific crystallographic planes by reducing the stress for lattice decohesion due to the effect of hydrogen. 39 The fracture morphology of 70% cold worked material (Fig. 11) shows secondary cracks in the cold work direction and a considerable amount of microvoid coalescence. The hydrogen in the cold worked steel tends to induce secondary cracks. These secondary cracks increase with increasing cold working. The increase in cold work results in a greater amount of deformation at grain boundaries and very likely enhancement of hydrogen accumulation at these deformed grain boundaries. The hydrogen at deformed grain boundaries will likely in turn make the boundaries more susceptible to crack initiation. Thus cracking will occur preferentially at grain boundaries in the cold work direction. As cracks propagate along the grain boundaries, a type of delamination occurs which in turn reduces the effective specimen thickness. This relaxation in thickness constraint may explain the fracture appearance of microvoid coalescence occurring between secondary cracks in the cold work direction for heavily cold worked conditions. CONCLUSIONS (1) Cold worked steel (0%) is sensitive to hydrogen embrittlement in a concentrated H2S environment and N A C E solution under slow strain rate loading. (2) COz and a high level of chloride in the sour environment may cause the 0% cold worked steel to become sensitive to anodic stress corrosion cracking. (3) Cold work and slow strain rate loading both promote hydrogen embrittlement.

FIG. 10.

Fractograph of SSC of 20% cold worked steel.

Fig. 11.

Fractograph of SSC of 70% cold worked steel.

75

Sulfide stress cracking of pipeline steel

77

(4) After hydrogen equilibrium occurs in the steel, the fracture toughness decreases with increasing cold work; sufficiently low values may be achieved to promote plane strain fracture even in relatively thin laboratory specimens. (5) Fractography indicates that sulfide stress cracking failure of low cold worked steel (<30%) is controlled by cleavage fracture. While sulfide stress cracking of heavily cold worked steel results primarily in microvoid coalescence along with secondary cracks aligned along the cold work direction and occurring at elongated grain boundaries. Acknowledgement--The authors wish to thank the Petroleum Graduate Research Program and AOSTRA for financial support of this work. Thanks are also given to Bob Konzuk of the Department of Mining, Metallurgy and Petroleum Engineering at the University of Alberta for his efforts in cold rolling the material.

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. ll. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24.

25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36.

REFERENCES A report of NACE Technical Practices Committee 1-G. Corrosion 8, 351 (1952). E. SNAPE,Corrosion 23, 154 (1967). F. F. LYLE, JR AND E. B. NORRIS, Corrosion 34, 193 (1978). B.J. BERKOWITZand R. D. KANE,Corrosion 36, 24 (1980). B. E. WILDE, C. D. KIM and J. C. TURN JR, Corrosion 38, 515 (1982). S. W. CIARALDI,Corrosion 40, 77 (1984). B.J. BERKOWITZand F. H. HEUBAUM,Corrosion 40, 240 (1984). Y. YOSH1NOand Y. MINOZAKI,Corrosion 42,222 (1986). D. C. LANGSTAFF,G. MEYR1CKand J. P. HIRTH, Corrosion 37, 429 (1981). H. ASAHI, Y. SOGO,M. UENOand H. HIGASHIYAMA,Corrosion 45,519 (1989). J. C. TURN,JR, B. E. WILDE and C. A. TROIANOS,Corrosion 39,364 (1983). J. P. FRASERand R. S. TRESEDER,Corrosion 8,342 (1952). L. M. VOLLMER,Corrosion 14, 324t (1958). A. K. DUNLOP, Corrosion 34, 88 (1978). K. MOTODA,Y. YAMANE,K. UESUGIand Y. NAKAI,Iron Steel Inst. Japan 67, s478 (1981). S. K. WILHELMand R. D. KANE,Corrosion 40,431 (1984). N. D. GREENEand G. A. SALXZMAN,Corrosion 20, 293t (1964). R. L. CHANCE, Corrosion 33, 108 (1977). Z. A. FOROULISand H. H. UHLlS, J. electrochem. Soc. 111,522 (1964). U. HUANGand W. J. D. SHAW,to be published. H. CIALONEand R. J. ASARO,Metall. Trans. 10A, 367 (1979). R. A. ORIANIand P. H. JOSEPHIC,Metall. Trans. IIA, 1809 (1980). M. W. JOOSXEN,T. D. LEE, T. GOLDENBERGand J. P. HIRTH, Hydrogen Effects in Materials (eds I. M. BERNSTEIN and W. W. THOMPSON),p. 235, AIME, Warrendale, PA (1980). E. LUNARSKA,Hydrogen induced degradation of low carbon steel, in Hydrogen Degradation of Ferrous Alloys (eds R. A. ORIANI, J. P. HIRTH and M. SMIALOWSKI),p. 712. Noyes, New Jersey (1985). D. A. RYDER,T. J. DAVIESand E. STRECKER,Second Int. Congr. Hydrogen in Metals, Paper 3B2. Pergamon Press, London (1977). J. O'M. BOCKRISand A. K. N. REDDY, Modern Electrochemistry, Vol. 2, p. 1240. Plenum, New York (1973). B. E. WILDE and C. D. KIM, Corrosion 42,243 (1986). C. D. BEACHEM,Metall. Trans. 3,437 (1972). T. D. LE and I. M. BERNSTEIN,Acta Metall. Mater. 39,363 (1991). ASTM Standard E-122-88 03.01,282 (1990). M. J. BARBA,Mem. Soc. Civils, Pt. 1,682 (1880). ASTM Standard E399-83 03.01,488 (1990). G. I. OGUNDELE,Ph.D. Thesis, University of Calgary (1984). NACE Standard TM0177-86, 1986 Revision. W. J. D. SHAW,Microstruct Sci. 13, 565 (1986). H. HUANGand W. J. D. SHAW,to be published.

78

H. HUANGand W. J. D. SHAW

37. G. I. OGUNDELEand W. E. WHITE, Corrosion 42, 398 (1986). 38. J. R. SCULLYand P. J. MORAN,Corrosion 44, 176 (1988). 39. H. H. BIRNBAUM,Hydrogen related failure mechanisms in metals, in Environment-sensitive Fracture of Engineering Materials (ed. Z. A. FOROULIS),p. 326. AIME, Warrendale, PA (1979). 40. K. N. AKHURSTand T. J. BAKER,Metall. Trans. 12A, 1059 (1981). 41. G.T. MURRAY,H. H. HONEGGERand T. MOUSEL, Corrosion 40, 146 (1984). 42. S. X. XIE and J. P. HIRTH, Corrosion 38,486 (1982). 43. E. MIRABALS. BHAYI'ACHARJEEand N. PAzos, Mater. Perform. 30, 41 (1991). 44. R. N. PARKINS,A. ALEXANDRIDOUand P. MAJUMDAR,Mater. Perform. 25, 20 (1986). 45. J. M. SUTCLIFFE,R. R. FESSLER,W. K. BOYDand R. N. PARKINS,Corrosion 28,313 (1972). 46. C. M. HUDGINS, R. L. MCGLASSON, P. MEHDIZADEHand W. M. ROSBOROUGH,Corrosion 22,238 (1966). 47. H. HUANGand W. J. D. SHAW,Microstruct. Sci. in press (1992). 48. P. GAY, P. B. HIRSCHand A. KELLY,Acta Crystall. 7, 41 (1954). 49. M. W. JOOSTEN, J. J. MURALIand J. L. HESS, Interactive effects of cold work, yield strength, and temperature on sulfide stress cracking, in Environmentally Assisted Cracking (eds T. W. CROOKERand B. N. LEIS), p. 136. ASTM, Baltimore, MD (1990).