COO- 2 576 3 *I

HYDROGEN SULFIDE STRESS CORROSION CRACKING

IN MATERIALS FOR GEOTHERMAL POWER

by R. F. Hehemann", A. R. Troiano**, B. Abu-Khaterf and S. Ferrignot

Presented at

The Geothermal Power Conference The Electrochemical Society

October 1976

, !

Sponsored by ERDA Under Contract EY-76-5-02-2576.000

NOTICE

I

* f

R. F. Hehemann, Professor of Physical Metallurgy

B. Abu-Khater and S. Ferrigno, Graduate Assistants

7 . x$-- ** A. R. Troiano, Republic Steel Professor of Metallurgy

t

Department of Metallurgy and Materials Science Case Western Reserve University

1

f

i

L

ON OF THIS DOCUMENT _-. IS UNLlMlTE - - .l,

DISCLAIMER

This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency Thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.

DISCLAIMER Portions of this document may be illegible in electronic image products. Images are produced from the best available original document.

HYDROGEN SULFIDE STRESS CORROSION CRACKING IN MATERIALS FOR GEOTHERMAL POWER

R. F. Hehemapn and A. R. Troiano Department of Metallurgy and Materials Science

Case Western Reserve University

strophic brittle delayed failure can occur in many

environments, some of which may not appear to be very aggres- \

sive from a straight corrosion point of view. This type of

environmentally induced failure, often occurring well below

d conservative design stresses has generally been cate-

gorized as stress corrosion cracking (SCC). It is well recog- - nized that in many instances hydrogen is

brittling agent leading to this type of f

ing a definite limitation on the use o f ava

The highly developed petroleum industry presents some of

the most critical examples of these problems. The ultimate

success of a greatly expanded source of geothermal power en-

counters, in many cases, much the same type of problems.

deed, a major limitation in the growing energy problem resides

in materials to contain, process, and generally handle the

various aggressive environments that are encountered

In-

(1-4).

It has been realized for many years that rezatively high

strength constructional type steels will crack in tfsourl' envi-

ronments.

that contains damaging amounts of

:Such an environment has been characterized as one

H2S which can be as low

as .001 atmosphere.'

cracking of normally employed oil well casing and tubing

steels in the 80,000 psi yield strength range (N-80) and

above (P110) (5-10). Indeed, as little as 0.1 ppm H2S has

produced failure in laboratory tests. The H2S acts as a

cathodic poison, dissociating to allow hydrogen absorption

into the metal and the formation of FeS

Accompanying C 0 2 results in acidification and pH levels of

3 to 4 are not uncommon.

This is based on t--e susceptibility to :+

(11). at the surface

a1 holes much the same type and even more onments are often encountered (12) and the

res have been recognized for * some of the older geothermal p'ower fields.

1961, wells in Waireki, New Zealand experienced brittle

layed fracture in well casing at 80,000 psi yield strength

and below (13).' At this same time it also was indicated that

a form of thermal corrosion fatigue was encountered due to

the "unclean" nature of geothermal steam, particularly damag-

ing to pumps, turbines, and ancillary equipment.

P

Other experience with geothermal wells indicated that

55,000 psi yield strength was the maximum safe strength allow-

in terms of HZS induced failures. However, at these low strengths, tubing collapse was encountered and limited

the strength of the joints required to compensate for non-axial r*

e * At ambient temperature this i s equivalent to approximately 3 ppm in solution.

*

loading, thermal stresses, lack of complete cementing, etc. (14). Higher strength steels are also clearly necessary for

bursting and collapse resistance where even thermal stresses

alone can exceed 60,000 psi

(9

(15)

Even at the levels generally considered to be a safe

maximum Y . S . , greater reliability would be most valuable and

help alleviate the drill pipe, casing, and tubing problems.

The consistency of performance is not only a material strength

level variable but also is dependent upon environmental fac-

tors, some o f which may change with time or the manner in

which the wells are operated.

strength is imperative in much of the ancillary equipment such

as well head fittings, valves, pump components, bolts, sub-

surface equipment, such as down hole pumps for geothermal

wells, etc.

Reliability at still higher

. *

L

Indeed, similar limitations on useful strength

level exist even in the higher alloyed classes of materials

such as stainless steels, Inconels, Monel, etc. as indicated

in a number of recent conferences. However, in these alloys,

the specificity of the environment may not completely corre-

spond to that involved in the constructional type steels.

Classical hydrogen embrittlement delayed failure exhibits

many characteristics' of environmentally induced stress corro-

sion cracking (16-18).

cracking in sour environments involves hydrogen embrittlement,

Since it is generally accepted that c(

particularly for the constructional type steels, it may be

?* . c 4

instructive to outline some of the parallelism of behavior.

Both will exhibit an incubation time for crack initiation,

discontinuous crack propagation, preferred crack nucleation

at crevices, notches, pits, etc., similar temperature ranges

of Sensitivity, similar fracture appearance, a critical

threshold stress for failure, a generally low stress inten-

sity factor (K1 ) (19,20), and increased sensitivity to SCC

failure as the yield strength is raised.

Of particular interest to the problem at hand are the

temperature range of sensitivity and the threshold stress.

The temperature range is generally considered to span the

region from approximately minus 50°F to plus 300°F for the

constructional type (body-centered cubic) alloys. It must

be appreciated that this range is sensitive to many other (21,231 parameters such as the environment and the strength level

erature limits correspond with those at wh

gen can be mobile and yet not so high as to preclude the ac- ( 2 4 ) . cumulation of the critical concentration for failure

Also of significance is the relation between the threshold

stress for failure and the Y.S. This critical stress rises

as the Y.S. is reduced until eventually it becomes equal to

the Y.S. Below this strength, brittle delayed failure will

not occur (”). This, of course, accounts for t

limitations of 80,000-100,000 psi yield strengt

mperatures. This threshold stress is also sensitive to’

temperature and increases with rising temperature.

The incubation time is of some relevance to the behavior

of geothermal materials especially with above-ground instal-

lations. The incubatioh time marks the line between revers-

ible and permanent damage and as such provides the basis for

dehydrogenation cycles for equipment involving hydrogeneous

environments.

Environmental induced failure (SCC) as compared to hydro-

gen pick-up in processing usually involves the additional com-

plications of an inexhaustible supply of hydrogen and the com-

plexities of changing surface and environment in terms of

hydrogen absorption,

take many different directions.

Proposed solutions to this problem can

Obviously one approach is

to systematically establish the limits of confidence for a

given set of circumstances in terms of material and environ- Y

mental conditions for established or recent developed alloys*

Clearly another approach involves attempts to develop mater-

ials with higher threshold stress values, either static or

dynamic, for the specific environmental conditions related

to geothermal holes.

For the available alloy constructional type steels, the

behavior in sour oil well and similar environments has been

quite thoroughly explored.

environment f in oil-gas and geothermal wells, potential

Although there are many similari-

geothermal holes present a much wider spectrum of variation

in the environment; for example, the very high chloride con-

tents and broad range of pH in these environments.

The development of materials resistant bt higher yield T

strengths and threshold stress values is of pfiime concern.

There is an increasing awareness of the role of metallurgical

structure.

tures with uniform carbide distributions such as tempered

martensite are more resistant to hydrogenous environments

than the more coarse structures such as ferrite-pearlite mix-

It is now generally appreciated that fine struc-

types such as austenitic stainless steel, Inconel, Stellites,

R-Monel, A-286 , MP35N, etc., the limits of confidence in

terms of heat treatment (e.g., overageing) , threshold stress, maximum safe strength level, etc., have not been systemati-

cally determined particularly with respect to the relevant

geothermal environments.- c For the most part, these more highly alloyed materials

have face-centered cbbic structures as compared to the l 0

tures (normalized structures) in constructional type steels,

all other things equal . (26-28)

I

-

Many different types of materials are necessary to meet

the requirements of a complete geothermal power system. For

1 the most part, steels suitable for casing, tubing, etc., will

i

not serve for much of the rest of the installation, such as

1 well head fittings, valves, pumps, bolts, and ancilliary

7 .*'

body-centered cubic structure of the constructional steels. i

There are several vital differences that should be appreciated

in assessing the response to a geothermal environment.

F.C.C. alloys are not notch sensitive, generally more sub-

The

ject to chloride induced pitting and most significantly ex-

hibit a substantially slower rate of hydrogen diffusion which

raises the temperature range for failure which is dependent

upon the influence of diffusible (mobile) hydrogen.

lly, this communication represents a progress

report on studies to evaluate performance of alloys in geo-

thermal power systems. This includes not only commercially

available alloys with standard treatments but also manipula-

tion of the metallurgical structures and/or designed composi- S

*

c tion modifications. The petroleum industry has accumulated

and made available a vast storehouse of data involving per-

formance of virtually all commercially available alloys in

a sour environment. Much but not all of these data are rele-

vant t o geothermal environments which apparently exhibit a

wider range of varying aggressive situations.

Experimental

The test specimens were mostly of the two point bend

c type, usually loaded to 90-100 percent of the design yield

strength.

Fig. 1.

The use of this type of test is ill ted in t

One advantage of this test aside from its simplicity

nd the relatively low cost type of specimen rests in the.

** 4

D

I

a

1

8

fact that there is virtually no critical crevice corrosion.

On the other hand, extreme care must be exercised in loading

the specimen since any plastic flow will reduce the calcu-

lated applied stress. Of course, any permanent set on re-

moval from the test jig will indicate some overstressing.

The specimens' dimensions were generally 4" x .5" x .035"

,

and in all cases cut with the long directionfparallel to the

rolling direction.

The 'tself-stressedtt type of bend test specimen was em-

ployed for some of the early tests as illustrated in Fig. 2

he difficulty in circumventing crevice corrosion

caused us to drop its use, at least during'the early screen-

ing procedures; However, it is partic

will allow a semi-quantitativ

if any, of environment exposure for specimens that

do not fail in the NACE adopted standard of 30 days exposure.

Essentially two different environments have been em- I ployed.

ments which is a deaerated aqueous solution of

0.5 acetic acid saturated with H2S. This solution 3 has been

One is the standard NACE solution for sour environ- 5 % NaC1,

. t.

widely used for evaluating materials in sour environments

and thus there i s a vast storehouse o f information on almost

all commercially available alloys with standard treatments.

Geothermal environments vary widely and no one

serve for total evaluation. However, those material situations

' utim can ' t

that appear promising with the standard solution are then

subjected to a modified NACE solution with 20 percent instead

of 5% NaC1, which will better match the high chloride contents

of many geothermal holes.

.

For both environments, the recommended tentative proce- .

dure of continuous purging with H2S for the full 30 day

tests was adopted (30). For the higher temperatures employing

a pressure vessel, continuous purging with H2S was not pos-

sible. However, this does not pose a problem, since it is

a closed system, previously deaerated, and substantial con-

centration of H2S is evident in the environment after

30 days. a The standard available tank H2S has approximately

t 300 ppm oxygen according to the supplier. In those tests

re a continuous flow of H2S was employed, the influence

this oxygen was quite apparent in the relative

definite corrosion in the 30 day tests at ambient t

and seriously affected tests near the boiling temperature of

the solution. Consequently, tests at the boiling point and

higher were conducted in autoclaves. Steps are being taken

tt? reduce this oxygen concentration.

A wide range of alloys are under examination both with

ard commercial treatments and metallurgical structures L omposition variations which ma$ potentially enhance

. performance: Their nominal compositions are listed in

Table I . The beam specimen has yielded data for commercial

I

f 10 0 .

alloys in the standard NACE solution at ambient temperatures

that are completely consistent with the literature. For ex-

ample, steels 410, 431, and 440, which represent indus-

try attempts to achieve improved corrosion resistance with

strengths above the present general limits, all failed at

the higher strength levels just as they had with many other

types of test specimens, including simple tensile, notch

bend, bent tubing, etc. These particular steels will not be

further considered in this report.

All heat treatments were performed in a purified

nitrogen atmosphere at controlled temperature.

8 structional steels in particular, extra precautions were

For the COII-

taken to avoid decarburization during both the normalizing

and austenitizing treatment. All constructional steels were

quenched for martensite and given a single temper a t the ap-

propriate temperature to attain the desired strength as in-

dicated in the tables. The age-hardenable alloys all required

prior cold work to attain their optimum yield strength levels.

Generally, these materials were received in the cold worked

condition, Our ageing treatments were designed to produce

under-aged, fully aged,

I j

and over-aged conditions

t

t' b I

F 11 e .

Results and Discussion

It will be convenient to consider the performance of

the constructional steels and of the more corrosion resis-

tant materials separately.

Constructional Steels

e superior resistance to SSC of tempered martensitic

with pearlitic structures has placed a premium on

hardenability. This has favored Cr/Mo steels over the more

economical Mn or Mn/Mo grades. I his program, 4130

with its well-documented behavior in r environments has

been chosen as a reference material and its resistance to SSC

is shown in Table 11.

1

I

%

With the unnotched beam samples employed in this study,

4130 is resistant at ambient temperatures SSC at a yield

strength level of 110 KSI and lower, Variable performance

is exhibited at intermediate strength levels and failure

consistently occurs at strengths of 128 KSI and

It is well known that the resistance of the BCC struc-

drogen embrittlement improves signi

5 e is raised. Thus, at 212'F and ab

not been observed at yield strength levels as high as 1

Higher strength levels are currently under study in order to *

t

?* . f. * i

a

t

12

determine the maximum strength for resistance to SSC as a

function of temperature.

The performance of the modified 4135 steel is presented

in Table 111. As has been shown by others, this composition

is somewhat more resistant to SSC than the standard 4130

and exhibits a yield strength advantage of the order o f

10 KSI (31y32).

per se, may have a greater significance. Specifically, it

This yield strength advantage while of interest

should provide greater reliability at a given strength level

such as 110 KSI compared t o that of standard

As for the standard 4130,

modified 4135 steel improves significantly as the test tem-

perature is increased.

4130 steels.

the resistance to cracking of this

In general, the resistance to SSC of both the standard

4130 and the modified 4135 steels in the 20% NaCl solution

was comparable or perhaps slightly superior to that in the B NaCl solution, The possibility that this may resu

similar or slightly reduced corrosion rates in the 20% NaCl

solution is being further examined, since the chloride ion is

reported t o act as a corrosion inhibitor.

Despite the dominant influence of strength level on re-

sistance tcj

that microstructural and compositional details also exert a

significant affect and can be manipulated t o optimize useable

strengths and reliability.

ydrogen embrittlement and SSC, it is evident

Lower carbon content is known to

enhance toughness significantly, and its influence on sulphide

SCC is currently under study.

a Q

) k

*rl 1 d

0.

k

v)

d

VI

Q)

Q

)a

k

c,

3

v)

0

. k

0

c, V

cd w

h

k

.rl k

n

1

d

a, Q

) +J

v)

Q)

no

E

a&

0

d .

Q)

5 fi. k

U

a

P

M

U

d

0

1 . M

rl

in cd

;

ko

PI

, 4

s

Q)

P

4 . 4

b

.

d

d

e .4

at

4

I l.

a 4

a

c, *rl

rlr

l

0

mk

0

rl

ut

4

0

NW

cd

Ei .rl

cd

c, v3

'* d

i,

z k

k

cd

cd rl 4

c, F: 7

L)

d

(d

0

c, x rl c,

u Q)

k

developed non-ferrous, age-hardenable alloy - MP35N (33) This alloy exhibits excellent resistance to general and to

pitting corrosion and, as shown in Table V I I I , has been re-

sistant to SSC in both the NACE and the 20% chloride environ-

ments. In fact, an additional sample, not reported in

Table VI11 was exposed for a total of 90 days in the NACE

C

I

solution (30 days at 325'F and 60 days at 425'F) with no

apparent co

This outsta

. in the lite

ion and complete resistance to cracking.

g performance generally confirms that

K Monel 500, treated to 140 and to 170 KSI strength

levels has resisted cracking in both the standard and the

modified NACE solutions (Table IX) . This agrees with the

rather limited literature on this alloy indicati resistance

to SSC in the NACE solution; however, there are several un-

officially reported service failures i

While generally not considered a s high Strength alloys,

the 300 grade austenitic stainless steels also are of po-

tential interest for use in geothermal environments. As indi-

cated in Table X several of these have resisted SSC at 325'F

in both the annealed and higher strength conditions achieved

by cold work, In general, susceptibility of these austenitic alloys, .

* As a word of cautiop, however, it has been shown recently that cracking can be induced in ing to a more active metal such as iron

c MP35N ( j l f , galvanic coupl-

r* . O * *

16

including A-286, to chloride SCC increases as the concentra-

tion of the chloride ion is increased. The oxygen content

of the solution also appears to be of significance with re-

gard to the cracking resistance in these solutions.

the influence of temperature, salt concentration and oxygen

concentration on the cracking resistance of the austenitic

Thus,

steels is under further study.

r

i. . 17

*

Conclusions

1. For the constructional steels of interest for casing,

tubing and other applications, the modified 4135

analysis exhibits a strength advantage of approximately

10 KSI over the standard 4130 analysis. This should

also improve reliability when employed at strength

levels below its maximum (approximately 120 KSI) for

2 .

1

resistance to SSC.

For a specified strength level, resistance to SCC is

improved at higher tempering temperatures. This ap-

pears to provide a lower limit on the carbon level for

optimum resistance to SSC.

3; Increasing salt concentration from 5% to 20% does not

appear to increase susceptibility t o SSC.

4 . Of the higher strength, corrosion resistant materials,

MP35N has shown outstanding resistance to general cor-

rosion and has not suffered SSC in these tests. Recent

reports, however, indicate that certain specific condi-

tions may cause SCC. .I

*

18

5. An age hardenable austenitic alloy, A-286 , is suscep-

tible to SSC in the fully aged condition (190 KSI) at

325OF and 425OF.

over-ageing to a strength level of 155 KSI but not by

under-ageing to a comparable strength level.

1 Susceptibility is greatly reduced by

,

6. For the constructional steels, increasing test tempera-

ture reduces the susceptibility to SSC substantially

thereby increasing the strength level that is resistant

to cracking.

terials, on the other hand, increases with increasing

4

Susceptibility of the highly alloyed ma-

Acknowledgements

This program constitutes a cooperative effort

between the Armco Steel Corporation, Middletown,

Ohio and the Metallurgy Department, Case Western

Reserve University.

tion of Armco Steel Corporation is gratefully ac-

knowledged.

The very substantial contribu-

The entire program has been supported

1.

20

References

R. N. Tuttle, "Deep Drilling - A Materials Engineering Challenge", Shell Development Company, Bellair Research Center Publication, #BRC-EP-630, Houston, Texas, 1973. See also Materials Performance, February, 1974.

2. T. Marshall and A. Tombs, "Delayed Fracture of Geothermal

3. Basil Wood, I'Basis of Selection of Materials for Geo-

4. K, C * Youngblut, "Materials Selection - Coal Gasifica-

Bore Casing Steels", Aus. Corros. Eng'g, 13, 9, 2-8, 1969.

thermal Schemes", UNESCO, Geothermal Energy, Paris, 1973.

tion Pilot p. 33, Dec

Corrosion Cracking of Steel under Sulphide Conditions", Corrosion, Vol. 8 , p. 333, 1952.

Corrosion Cracking, Corrosion, 24, 31, 1968.

Steels in Sour Crude Oilstt, Materials Protection, p. 33, Jan. 1969.

Strength Steels", Proc. 7th World Petroleum Congress, Elsevier Publishing Company, Ltd., p. 201, 1967.

9. J. L. Battle, T. V. Miller, and M. E. True, 'IResistance of Commercially Available High Strength Tubular Goods to Sulphide Stress Cracking", Materials Performance, June, 1975,

lant", Materials Protection and Performance,

5. C, N. Bowers, W. J. McGuire and A. E. Wiehe, "Stress

1

6. R. S. Treseder and T. M. Swanson, ltFactors in Sulphide

a 7. J. F. Bates, "Sulphide Cracking of High

8. E. H. Phelps, "Stress Corrosion Behavior of H

10. D. S, Burns., "Laboratory Test for Evaluating A110

11. P. Bastien and Amiot, "Mechanism of the Effect o f Ionized Solutions of H S on Iron and Steel", Compte Rendue, Vol. 235, p. 1051, 1952.

istics of the Salton Sea Geothermal System", Amer. Jour. of Sci., - 266, p. 129, 1968.

Servicert, Materials Performance, Sari, 1976.

- b

12. H. C. Helgeson, Veologic and Thermodynamic Character-

1' b .

1

L

21

13. P. K. Forster, T. Marshall, and A, Tombs, U.N. Conf. on New Sources of Energy, Rome, 1961. See also Aust. Corrosion Eng'g., Vol. 6, p. 306, 1962.

14. J. H. Smith, "Casing Failures in Geothermal Bores at Wairekei", Geothermal Energy, Vol. 3, U.N. Conf., p, 254, 1961

1 5 . A. C. L. Fooks, "The Development of Casings for Geo- thermal Boreholes", Ibid.

16. A. R. Troiano and J. P. Fidelle, "Hydrogen Embrittle- ment in Stress Corrosion Cracking'?, Magistrale Lecture, International Conference Hydrogen in.Metals, Paris, France, June 1972. No. 18.

ttHydrogen Permeability and Delayed Failure of Marten- sitic Steels", Corrosion, Vol. 25, p. 353, 1969.

18. J. Papp, R. F. Hehemann, and A. R. Troiano, "Hydrogen Embrittlement of High Strength FCC Alloystt, ASM, CMU Conference, Seven Springs, Pennsylvania, September 1973, ASM Confer. Vol. 1973.

Steel in Liquid Environments", Proc. Am. SOC. Testing Materials, Vol. 60, p. 750, 1960.

of Steels", Corrosion, Vol. 26, p . 177, 1970.

See also ART Keynote Lecture, Ref.

17, C. F. Barth, E. A. Steigerwald and A. R. Troiano,

19. E. A. Steigerwald, "Delayed Failure of High Strength

20. L. M. Dvoracek, ItSulphide Stress Corrosion Cracking

21. W. Beck, E. J. Jankowsky and P. Fischer, "Hydrogen 8t;ress Cracking of High Strength Steels", Naval Air Development Center, Rept. No. NADC-MA-7140, 1972.

22. J. B, Greer, "Factors Affecting the Sulphide Cracking of High Strength Steels", Materials Performance,

23. R. D, Kane and J, B. Greer, 'lSulphide Stress Cracking Vol. 14, 11-22, 1975.

of High-Strength Steels in Laboratory and Oil Field Environmentstt, Jour, Petroleum Tech., AIME, Oct. 1976. Preprint SPE 6144.

I ,

i

*

L

. 22

24. A. R. Troiano, "The Role of Hydrogen and Other Inter- stitials in the Mechanical Behavior of Metals", Edward DeMille Campbell Memorial Lecture, Trans. ASM, Vol. 52, p. 54, 1960.

"Stress Corrosion of Ferritic and Martensitic Stain- less Steelsft, Corrosion, Vol. 29, p. 268, 1973.

25. M. F. McGuire, A. R. Troiano and R. F. Hehemann,

26. W. M. Cain and A. R. Troiano, "Steel Structure and Hydrogen Embrittlement", Petroleum Engineer, Vol. 37, p. 5, May 1965.

27. E. Snape, Yloles of Composition and Microstructure in Sulphide Cracking of Steel", Corrosion, Vol. 24, p. 261, 1968.

and Low Alloy Steels", Corrosion, Vol. 23, p. 1S4, 1967.

and Standards", ASTM, January, 1965.

Cracking at Ambient Temperature", Proposed NACE Stan- dard T-11-9 Procedure, January 15, 1975.

28. E. Snape, "Sulphide Stress Corrosion of Some Medium

29. G. J, Geimerl and D. N. Braski, "Materials Research

30. "Evaluation of Metals for Resistance to Sulphide Stress

31. J. B. Grobner, D, L. Sponseller, and W. W. Cias, "Development of Higher Strength HZS Resistant Steels f o r Oil Field Applicationstt, Materials Performance, June, 1975.

32. L. M. Dvoracek, "High Strength Steels for H2S Service",

33. J. P. Stroup, A. H. Bauman, and A. Simkovich, "Multiphase

Materials Performance, May, 1976.

MP35N - An Alloy for Deep Sour Well Servicett, Materials Performance, June, 1976.

Co., Houston, Texas. 34, Private Communication, P. R. Rhodes, Shell Development

a i

c

4 .

23

The nominal analyses of the alloys considered in this report are given in Table I below:

TABLE I

Cr - C Designa tion - Mo - Cb - - 0.20 0.50 0.15 4118

4135 mod.* 0.35 1.0 0.75 - 4130 0630 1.0 0.20 .04

03 2 Cr-0.75 Mo 0.20 2.0 0.75 0.10 3.0 1.0 3 Cr-1.00 Mo

- L - - - - - - - - - - - - - - l - - - - - - - - - - - - - -

v - Mo - _. Ni - Ti - Mn

* . . * * . 24

TABLE I1

Sulfide S t ress Cracking of 4130 Stee l

Tempering No. Failed Test TW.-OF T-.-OF* Y.S. KSI No. Tested

5% N a C l (NACE Solution)

Ambient 1100 138 1/1 1150 128 4/4 1200 123 213 1250 119 5 2/3 1300 107.5 0/2

212 1100 138 o/ 1 1500 128 0/3 1200 123 o/ 2 1250 119 5 o/ 2

325 1150 128 O / l 1200 123 o/ 1 1250 119.5 o/ 2

4 . e

* * 25

TABLE 111

,

Sulfide Stress Cracking of Modified 4135 Steel

Tempering T e s t Te.mp.-OF T=P.-OF* Y.S. KSI

5% N a C l (NACE Solution)

Ambient 1200 138 1250 125 1275 116 1300 110

212 1150 150 1200 138 1250 125 1300 ) 110

325 1200 138 1250 125

20% NaCl

Ambtent 5200 138 1250 125 1390 110 '

212 1250 125 1300 110

325 1150 150

425 1250 125

* 1750°F, 1 H r , Temper 2 Hrs,

A.C.: A.C.

1650°F,

No. Failed No. Tested

5f 5 5f 8 0f4 0/4

1/1 0f1 1f3 of 2

0f3 0/2

2f 2 2f 3 0f1

0f1 0/1

0/1

'Of 1

26

.

TABLE I V

Sulfide Stress Cracking of 4118 Steel

Tempering No. Failed Test Temp.-OF T~IQP ,-OF* Y.S. KSI No. Tested

Sg NaC1 (NACE Solution)

Ambient 1000 120 1/1

1050 I08 112

1100 95 01 3

* 1/2 Hr, 1650°F, A.C.; 1/2 Hr, 1550°F, W . Q . ; Temper 1 H r , A.C.

I

4

I

27

TABLE V

Sulfide Stress Cracking of 2 Cr-0.75 Mo S t e e l

Tempering No. Failed T e s t Temp.-'F Temp. -' F* Y.S. KSI No. Tested

5% NaCl (NACE Solution)

Ambient 1200 130 212

1250 118 11 1

1300 105 0/4

20% NaCl

W i e n t 1300 105 O/l

20% NaCl

W i e n t 1300 105 O/l

* 10 Min. 1750°F, A.C.: T a p e r 1 Hr, A.C.

10 Min. 1650'F, W.Q.;

2 8

29

TABLE V I 1

Sulfide Stress Cracking of A-286 Steel

Ageing No. Failed Test Temp .-OF Treatment Y.S. KSI No. Tested

5% NaCl (NACE Solution)

212 16 Hrs-1200°F 190 011

325 16 Hrs-1200°F 190 111 16 Hrs-1300°F 155 011

I

I

8

20% NaCl

425 16 Brs-1200°F 190 314 1 Hr -1150OF 165 111 16 Hrs-1300°F 155 0/2*

* An additional sample tested without deaeration of the solution exhibited small cracks.

i

T

30

a .

TABLE VI11

SULFIDE STRESS CRACKING OF MP35N

No. Failed Test Temp . - O F Y.S. KST No. Tested

5% NaCl (NACE Solution)

Ambient 220 0/1

425 220 0/1

20% NaCl - 425 220 0/2*

* One sample tested without deaeration.

I

TABLE IX

Sulfide,Stress Cracking of K Monel

Ageing No. Failed Test T=p.-OF Treatment Y.S. KSI No. Tested

5X NaCl (NACE Solution)

Ambient 8 Hr, 1000°F/F.C. to 900°F/W.Q. 170 O/l

1 325 16 Hr, llOO°F/F.C. to 900°F/A.C. 14 0 0/1 i

425 16 FIr, llOO°F/F.C, to 900°F/A.C. 140 o/ 2 z

20% NaCl

425 8 Hr, 100O0F/F.C. to 900°F/W.Q. 170 0/1

16 Hr, llOO°F/F.C, to 900°F/A.C. 140 0/2

.- - IL

I

i

0

c

1

33

Fig. 1. A286 Standard NACE solution. 34 days, 32S°F, Stressed at 100% of Y.S.

A. B.

Overaged to 135,000 psi Y.S. Fully aged to 190,000 psi Y.S.

-

Note:

Fig. 2. A 286 Standard NACE solution. 190,000 psi Y.S. Stressed at 90% Y.S. A. Not exposed. Compressed to initial cracking B.

Reduced ability to be compressed before cracking (~40%)

32 days ambient temperature

Exposed to environment but no failure