RP029604

Standard

Recommended Practice

Guidelines for Detection, Repair, and Mitigation of Cracking of Existing Petroleum Refinery Pressure

Vessels in Wet H 2S Environments

This NACE International standard represents a consensus of those individual members who have reviewed this document, its scope, and provisions. Its acceptance does not in any respect preclude anyone, whether he has adopted the standard or not, from manufacturing, marketing, purchasing, or using products, processes, or procedures not in conformance with this standard. Nothing contained in this NACE standard is to be construed as granting any right, by implication or otherwise, to manufacture, sell, or use in connection with any method, apparatus, or product covered by Letters Patent, or as indemnifying or protecting anyone against liability for infringement of Letters Patent. This standard represents minimum requirements and should in no way be interpreted as a restriction on the use of better procedures or materials. Neither is this standard intended to apply in all cases relating to the subject. Unpredictable circumstances may negate the usefulness of this standard in specific instances. NACE assumes no responsibility for the interpretation or use of this standard by other parties and accepts responsibility for only those official NACE interpretations issued by NACE in accordance with its governing procedures and policies which preclude the issuance of interpretations by individual volunteers.

Users of this NACE standard are responsible for reviewing appropriate health, safety, environmental, and regulatory documents and for determining their applicability in relation to this standard prior to its use. This NACE standard may not necessarily address all potential health and safety problems or environmental hazards associated with the use of materials, equipment, and/or operations detailed or referred to within this standard. Users of this NACE standard are also responsible for establishing appropriate health, safety, and environmental protection practices, in consultation with appropriate regulatory authorities if necessary, to achieve compliance with any existing applicable regulatory requirements prior to the use of this standard.

CAUTIONARY NOTICE: NACE standards are subject to periodic review, and may be revised or withdrawn at any time without prior notice. NACE requires that action be taken to reaffirm, revise, or withdraw this standard no later than five years from the date of initial publication. The user is cautioned to obtain the latest edition. Purchasers of NACE standards may receive current information on all standards and other NACE publications by contacting the NACE Membership Services Department, 1440 South Creek Dr., Houston, Texas 77084-4906 (telephone +1 281/228-6200).

Revised 2004-02-12 Reaffirmed 2000-09-13 Approved 1996-03-30 NACE International 1440 South Creek Dr. Houston, Texas 77084-4906

+1 281/228-6200

ISBN 1-57590-013-0 NACE Standard RP0296-2004

Item No. 21078

RP0296-2004

________________________________________________________________________

Foreword

NACE International Task Group T-8-16 on Cracking in Wet H2S Environments was formed in 1988

to conduct an organized study on the incidence and mechanisms of cracking in pressure vessels

operating in refinery wet hydrogen sulfide (H2S) environments. Specific objectives were to (a)

define the nature and extent of the problem by means of an industry survey; (b) define

mechanisms for the type of cracking found, to be accomplished primarily through a literature

survey; (c) establish inspection guidelines for existing vessels; and (d) develop repair and

mitigation guidelines for cracked vessels. Four work groups were formed to address these tasks.

In 1990, a fifth work group was formed with a fifth objective, (e) to investigate material

specifications and fabrication practices for new pressure vessels.

This standard recommended practice summarizes objectives (a), (c), and (d) listed above. A

technical committee report (NACE Publication 8X294)1 has been issued to address objective (b).

Finally, objective (e) has been handled by another technical committee report (NACE Publication

8X194).2

This standard is intended for use primarily by refinery corrosion and materials engineers and

inspection, operations, and maintenance personnel. Information and guidance presented in this

standard reflect the work of many individuals representing numerous companies worldwide.

The titles and source information of the codes, specifications, and standards referred to or

discussed in this standard are provided in Appendix A rather than listed in footnotes throughout the

standard. Confining this information to one appendix should help readers who have any interest in

further research. This standard was originally prepared in 1996 by former Task Group T-8-16 on

Cracking in Wet H2S Environments. It was reaffirmed in 2000 by Group Committee T-8, and

revised in 2004 by Task Group (TG) 268 on Wet H2S Cracking in Petroleum Refinery Pressure

Vessels. TG 268 is administered by Specific Technology Group (STG) 34 on Petroleum Refining

and Gas Processing. This standard is issued by NACE International under the auspices of STG

34.

In NACE standards, the terms shall, must, should, and may are used in accordance with the

definitions of these terms in the NACE Publications Style Manual, 4th ed., Paragraph 7.4.1.9. Shall

and must are used to state mandatory requirements. The term should is used to state something

considered good and is recommended but is not mandatory. The term may is used to state

something considered optional.

________________________________________________________________________

RP0296-2004

________________________________________________________________________

NACE International

Standard

Recommended Practice

Guidelines for Detection, Repair, and Mitigation of Cracking of Existing Petroleum Refinery

Pressure Vessels in Wet H2S Environments

Contents

1. General (1)

of Cracking (1)

2. Mechanisms

3. Inspection

for Cracking (3)

4. Repair of Cracked or Blistered Vessels (7)

Considerations for Operation (9)

5. Mitigation

References (10)

Bibliography (10)

Appendix A—Cited Codes, Specifications, and Standards (11)

Appendix B—Nature and Extent of Problem—Survey Results from 1990 Industry Survey (12)

Appendix C—Typical Cracks Found in Wet H2S Environments (17)

Figure C1: SSC in HAZ of head-to-shell weld of FCCU absorber tower (17)

Figure C2: Hydrogen blister in A 516-70 amine contactor/water wash tower (18)

Figure C3: Upper photo: Blisters on ID surface of amine contactor/water wash tower (19)

Figure C3: Lower photo: Cross-section of plate shown in upper photo illustrating HIC

(“stepwise” cracking) (19)

Figure C4: SOHIC in soft base metal extending from the tip of SSC in a hard HAZ of a

repair weld in the shell of a primary absorber (deethanizer) column in an FCCU gas

plant. (20)

Figure C5: ASCC (carbonate cracking) of non-PWHT A 285-C steel shell of FCCU main

fractionator overhead accumulator. (21)

Table B1: Overall Summary (12)

Table B2: Company Breakdown (13)

Table B3: Cracking by Process Unit (13)

Table B4: Cracking vs. Operating Temperature (14)

Table B5: Cracking vs. H2S Concentration (14)

Table B6: Cracking vs. Steel Specification (14)

Table B7: Cracking vs. Steel Grade (15)

Table B8: Cracking vs. PWHT (15)

Table B9: Cracking vs. Blistering History (15)

Table B10 Cracking vs. Weld Repairs (15)

Table B11: Depth of Cracking (16)

Table B12: Crack Penetration (16)

Table B13: Disposition of Cracked Pressure Vessels (16)

________________________________________________________________________

RP0296-2004 ________________________________________________________________________

Section 1: General

1.1 This standard is intended to be a primary source of information on cracking in wet H2S petroleum refinery environments and provides guidelines on the detection, repair, and mitigation of cracking of existing carbon steel refinery pressure vessels in wet H2S environments. Refinery pressure vessels include items such as, but not limited to, columns, heat exchangers, drums, reboilers, and separators. All repairs should be conducted in accordance with API(1) 5103 or another applicable industry code or standard. Limited cracking has been noted in piping and therefore, piping is not included in the scope of this standard. Information on inspection practices for new pressure vessels (never in service) is given in NACE Publication 8X194.

1.2 For the purposes of this standard, wet H2S service is defined as refinery process environments containing free water as a liquid phase and in which

(a) the concentration of H2S is >50 mg/L (50 ppmw) in the free water, or

(b) free water pH is <4 with some dissolved H2S, or

(c) free water pH is >7.6 with at least 20 mg/L (20 ppmw) dissolved hydrogen cyanide (HCN) in the water with some dissolved H2S, or

(d) >0.0003 MPa abs (0.05 psia) partial pressure of H2S is present in the gas in processes with a gas phase. However, the threshold concentration of H2S in the aqueous phase required for cracking to occur has not been clearly established. Therefore, selective application of this standard may be appropriate when past experience has indicated the presence of cracking or blistering in comparable service, regardless of H2S concentration.

1.3 Increased industry attention to the potential for cracking of carbon steel pressure vessels began in 1984 with the rupture of a monoethanolamine (MEA) absorber tower at a Lemont, Illinois, refinery. The ensuing explosion and fire resulted in fatalities and extensive damage to the facility.4 In response to this incident, NACE Task Group T-8-14 on Stress Corrosion Cracking of Carbon Steel in Amine Solutions was formed in the fall of 1984. An industry survey to determine the nature and extent of the cracking problem was conducted by T-8-14. Appendix B presents a summary of the survey findings. The results of the T-8-14 effort have been reported separately.5

1.4 In 1988, some new results on vessel inspections and the cracking found were reported to the industry.6 Among the significant findings was the observation that cracking problems were occurring in other wet H2S environments, not just in MEA. It was further reported that inspection techniques commonly used at the time (visual, liquid penetrant, and dry magnetic particle testing) were not sensitive enough to find these cracks. In response to this new information, NACE Task Group T-8-16 on Cracking in Wet H2S Environments was formed in the spring of 1988.

________________________________________________________________________

Section 2: Mechanisms of Cracking

2.1 The objective of this section is to define the terms used to describe cracks that occur because of exposure to wet H2S environments and describe the mechanisms of cracking. Photographs of typical cracks found in wet H2S environments are shown in Appendix C.

2.2 Definitions

2.2.1 Sulfide Stress Cracking (SSC): Cracking of a

metal under the combined action of tensile stress and

corrosion in the presence of water and H2S. SSC is a

form of hydrogen stress cracking resulting from absorption of atomic hydrogen that is produced by the

sulfide corrosion process on the metal surface. SSC

usually occurs more readily in high-strength steels or in

hard weld zones of steels. (See Figure C1.)

2.2.2 Hydrogen Blistering: The formation of subsurface planar cavities, called hydrogen blisters, in a metal resulting from excessive internal hydrogen pressure. Growth of near-surface blisters in low-strength metals usually results in surface bulges. Hydrogen blistering in steel involves the absorption and diffusion of atomic hydrogen produced on the metal surface by the sulfide corrosion process. The development of hydrogen blisters in steels is caused by the accumulation of hydrogen that recombines to form molecular hydrogen at internal sites in the metal. In its molecular state, hydrogen is too large to diffuse through the steel. Typical sites for the formation of hydrogen blisters are large nonmetallic inclusions, laminations, or other discontinuities in the steel. This differs from the voids, blisters, and cracking associated with high-temperature hydrogen attack. (See Figure C2.)

___________________________

(1) American Petroleum Institute (API), 1220 L St. NW, Washington, DC 20005.

RP0296-2004

2.2.3 Hydrogen-Induced Cracking (HIC): Stepwise

internal cracks that connect adjacent hydrogen blisters

on different planes in the metal, or to the metal surface

(also known as stepwise cracking). No externally applied stress is needed for the formation of HIC. In

steels, internal cracks that may develop (sometimes

referred to as blister cracks) tend to link with other

cracks by a transgranular plastic shear mechanism.

This occurs due to internal pressure resulting from the

accumulation of hydrogen. The link-up of these cracks

on different planes in steels has been referred to as

stepwise cracking to characterize the nature of the

crack appearance. HIC is commonly found in steels

with (a) high impurity levels that have a high density of

large planar inclusions, and/or (b) regions of anomalous microstructure produced by segregation of

impurities and alloying elements in the steel. (See

Figure C3.)

2.2.4 Stress-Oriented Hydrogen-Induced Cracking

(SOHIC): Arrays of cracks, aligned nearly perpendicular to the stress, that are formed by the link-

up of small HIC cracks in steel. Tensile stress (residual

or applied) is required to produce SOHIC. SOHIC is

commonly observed in the base metal adjacent to the

heat-affected zone (HAZ) of a weld, oriented in the

through-thickness direction. SOHIC may also be produced in susceptible steels at other high stress points such as from the tip of mechanical cracks and

defects, or from the interaction of hydrogen blisters on

different planes in the steel. (See Figure C4.)

2.2.5 Alkaline Stress Corrosion Cracking (ASCC):

Cracking of a metal produced by the combined action

of corrosion in an aqueous alkaline environment containing H2S, CO2, and tensile stress (residual or

applied). The cracking is branched and intergranular in

nature, and typically occurs in nonstress-relieved carbon steels. This form of cracking has often been

referred to as carbonate cracking when associated with

alkaline sour waters, and as amine cracking when associated with alkanolamine treating solutions. (See

Figure C5.)

2.3 Environmental Parameters Affecting Cracking

2.3.1 Several cracking mechanisms in wet H2S

environments, including SSC, hydrogen blistering, HIC,

and SOHIC, are related to the absorption and permeation of hydrogen in steels. The key variables

involved in hydrogen permeation in steels are pH and

the composition of the service environment. Typically,

the hydrogen permeation flux in steels has been found

to be minimal in neutral solutions (pH 7), with increasing flux at both lower and higher pH values.

Corrosion at low-pH values is caused by H2S, whereas

corrosion at high-pH values is caused by increasing

concentrations of ammonium bisulfide (higher ammonia levels in H2S-dominated environments).

2.3.2 Hydrogen permeation has also been found to

increase with increasing H2S partial pressure and with

the presence of cyanide at alkaline pH levels.

2.3.3 SSC susceptibility increases with increasing H2S

partial pressure. Based on investigations in oil and gas

production environments, 0.0003 MPa abs (0.05 psia) and greater partial pressure H2S in the presence of free

water may produce SSC in susceptible steels.7

2.3.4 ASCC can occur over a wide range of

temperatures, but susceptibility appears to increase with increasing temperature. ASCC generally occurs in

alkaline solutions with a pH in the 8 to 11 range, but its

occurrence is highly dependent on the solution composition. This form of cracking has occurred in refinery services such as sour water streams and alkanolamine solutions containing H2S and CO2.

ASCC is promoted by carbonates in the presence of weak sulfiding agents such as thiosulfate and thiocyanate. The mode of cracking involves local anodic dissolution of iron at breaks in the normally protective corrosion product film on the metal surface.

Laboratory tests have shown that cracking occurs in a

relatively narrow range of electrochemical potential that

corresponds to a destabilized condition of the protective film. This film destabilization occurs at very

low ratios of the sulfide concentration to the carbonate/bicarbonate concentration in the solution, and is possibly affected by a number of contaminants in the solution. This form of cracking is not directly associated with the above-mentioned forms of hydrogen-related damage. However, in sour waters and alkanolamine services containing H2S, cracking due to HIC, SOHIC, and SSC is possible, in addition to

ASCC.

2.4 Material Parameters Affecting Cracking of Carbon Steels in Wet H2S Environments

2.4.1 Sulfide Stress Cracking

2.4.1.1 SSC has not generally been a concern in

the carbon steel base metals typically used for

pressure vessels in refinery wet H2S environments

because these steels have generally been below

620 MPa (90 ksi) tensile strength.

2.4.1.2 Carbon steel weld metal is generally

considered resistant to SSC if its hardness is

limited to 200 HBW maximum in corrosive

petroleum refining environments in accordance

with NACE Standard RP0472.8However,

weldments (weld metal, HAZ, and adjacent base

metal zones subject to residual stresses from

welding) may contain localized zones of high

hardness. SSC in carbon steel weldments

frequently is limited to hard HAZs of the last weld

pass, which are not tempered by subsequent weld

passes. Data show that, depending on the

severity of the service environment, small hard

regions of up to 248 HV (237 HBW) can be

RP0296-2004

tolerated without the occurrence of SSC. The

Rockwell Superficial Hardness equivalent to 248

HV is 70.5 HR15N. These values are a direct

conversion from the 22 HRC maximum specified

in NACE Standard MR01037 for ferritic materials to

be used in petroleum refining environments.

2.4.2 Hydrogen Blistering and Hydrogen-Induced Cracking

2.4.2.1 Hydrogen blistering and HIC have been

encountered in the lower-strength carbon steels

typically used in refinery wet H2S environments.

2.4.2.2 These cracking mechanisms are

associated with the formation of blisters caused by

an accumulation of molecular hydrogen at internal

laminations, nonmetallic inclusions, or other discontinuities in the steel. Reducing the inclusion

level of the steel by lowering the sulfur content

increases the resistance to hydrogen blistering

and HIC. In addition, control of sulfide inclusion

morphology by calcium or rare earth metal

additions to produce a spheroidal sulfide shape, in

conjunction with use of lower-sulfur steels, has

been found to increase resistance to hydrogen

blistering and HIC.

2.4.2.3 Base metal heat treatments, such as

normalizing or quenching and tempering above

593°C (1,100°F), increase resistance to crack

growth.

2.4.3 Stress-Oriented Hydrogen-Induced Cracking

2.4.

3.1 Generally, the material parameters

affecting hydrogen blistering and HIC are expected

to apply to SOHIC.

2.4.

3.2 Susceptibility to SOHIC is increased by

increasing local tensile stresses. Notch-like weld

discontinuities and/or local differences in microstructure present in the area of a weldment

may increase the localized stresses. Postweld

heat treatment (PWHT) is expected to reduce the

susceptibility to SOHIC when it is influenced by

residual stress. PWHT can also reduce local HAZ

hardness, thereby reducing the possibility for SSC,

which can initiate SOHIC.

2.4.

3.3 SOHIC has been found in pressure

vessels constructed with conventional steels in

refinery wet H2S environments. In laboratory tests,

SOHIC has been found in a variety of steels. In

severe hydrogen-charging laboratory tests, SOHIC

has also been found in steels processed to

optimize resistance to HIC.

2.4.4 Alkaline Stress Corrosion Cracking

2.4.4.1 ASCC has occurred in a variety of steels.

Field experience to date has not indicated any

significant correlation between susceptibility to

ASCC and steel properties.

2.4.4.2 Susceptibility to ASCC increases with

increasing tensile stress level. Areas of deformation resulting from cold forming or localized high residual stresses in weldments are

more prone to ASCC. Surface discontinuities,

especially in areas adjacent to welds, often serve

as initiation sites for ASCC because they act as

localized stress raisers. ASCC can be effectively

controlled by PWHT and proper heat treatment

after cold forming.

________________________________________________________________________

Section 3: Inspection for Cracking

3.1 The objective of this section is to provide guidelines on inspection for cracking of existing carbon steel pressure vessels in petroleum refinery wet H2S environments.

3.2 The scope of this section is the inspection of weldments. This includes pressure-retaining circumferential, longitudinal, and nozzle welds, and internal attachment welds to the pressure boundary.

3.3 Inspection requirements for new pressure vessels (never in service) are beyond the scope of this standard. However, initial inspection of new vessels during fabrication with methods of comparable sensitivity to anticipated in-service inspection methods is of significant value in assessing subsequent inspection results.

3.4 These guidelines incorporate risk-based principles to determine the need and frequency for inspection. (Risk is defined as the likelihood [or probability] of failure times the consequence of failure.) See API RP 5809 and API Publication 581.10 Also included are recommendations for inspection personnel qualifications, nondestructive examination (NDE) procedures, areas of inspection, surface preparation, inspection techniques, acceptance criteria, reporting of results, and reinspection.

3.5 Application of these inspection guidelines shall be made by engineers and/or inspection personnel who are knowledgeable in the technical aspects of this section.

3.6 Inspection Priorities and Intervals

3.6.1 Each refinery should prioritize equipment in wet

H2S service. The priority ranking for equipment should

consider the consequences of a leak or a failure on the

surrounding area, operating conditions (temperature,

RP0296-2004

pressure, and contents), criticality of the equipment, and the fabrication, inspection, and repair history. Priorities can be established by assessing the risk that cracking represents to the refinery. Evaluation of risk may utilize industry-approved approaches such as found in API RP 580, API Publication 581, or similar procedures/methodologies unique to the refiner. In either case, the risk assessment process should address the likelihood of cracking and the consequence of failure.

3.6.2 Some of the factors that should be considered when assessing the likelihood of wet H2S cracking and blistering are the following. These guidelines are based on survey data, literature information, and industry experience.

(a) History of cracking and blistering. Vessels with a history of blistering are more likely to be cracked. Also, vessels in service comparable to that of other vessels that have cracked are more likely to be cracked.

(b) Materials, fabrication, and repair history. Vessels without PWHT or those with nonpostweld heat-treated repairs should be given higher priority when setting inspection requirements. NACE Publication 8X194 provides some background information on materials and fabrication practices typically used for vessels in wet H2S service.

(c) Type of vessel. Trayed columns or drums in which an aqueous phase can condense, splash, or accumulate are more susceptible to cracking and blistering. Vapor spaces where condensation occurs or where sections are intermittently wetted are often the most severely damaged.

(d) Severity of the process environment. Pressure vessels in the following environments should be considered more susceptible to HIC, SOHIC, or blistering.

A process temperature between ambient and 149°C (300°F), and:

? high concentration of H2S (generally >2,000 mg/L [2,000 ppmw]) and pH above 7.8;

? H2S (generally >50 mg/L [50 ppmw]) and pH below 5.0;

? presence of hydrogen cyanide (HCN) (generally >20 mg/L [20 ppmw] in the aqueous phase); or

? other environments with high potential for hydrogen activity as a result of aqueous corrosion.

(e) Type of process unit. The data presented in Table B3 of Appendix B may be useful in prioritizing inspection of vessels in various process units. In addition, the following list highlights specific areas within certain process units in which significant wet H2S cracking has been found:

? catalytic cracking unit fractionation and light ends

recovery sections containing separated overhead vapor streams;

? hydrocracking and hydrotreating unit separation

and fractionation sections;

? coker fractionation and light ends recovery sections containing separated overhead vapor streams;

? sour water stripping unit overhead systems; and

? alkanolamine acid gas removal unit contactor

(absorber), rich amine flash drum, stripper (regenerator), and overhead systems.

3.6.3 Some of the factors that should be considered

when assessing the consequences of failure or leakage are as follows:

(a) Nature of the process fluid, e.g., tendency to form

a vapor cloud, flammability, combustibility, toxicity, etc.;

(b) Total release inventory;

liquefied petroleum gas [LPG]), which could result in

brittle fracture;

(d) Potential impact on plant operations and/or

surrounding community;

(e) Total pressure on the system; and

(f) Leak scenario versus a rupture scenario.

3.7 Extent of Inspection

3.7.1 The extent of initial inspection shall be sufficient

to provide a representative sample of the various areas

of concern. The areas of concern include longitudinal,

circumferential, and nozzle welds, and internal attachment welds to the pressure boundary. For those

pressure vessels warranting inspection based on prioritization, the intent of inspection should be based

on the risk the vessel represents to the refiner. If

environmental cracks are found, the inspection coverage should be increased as necessary to adequately define the extent of cracking.

3.7.2 Areas to be inspected should specifically include

repair or vessel alteration welds.

3.7.3 Areas to be inspected should also include

portions of the vessel that exhibit visible blistering or

significant corrosion.

3.7.4 In pressure vessels such as columns or towers

in which there is a variation of environments, the

inspection should focus on areas considered more

susceptible to cracking, e.g., cooler areas in which an

aqueous phase may be present.

RP0296-2004

3.8 Inspection Methods

3.8.1 Several NDE techniques can be used to detect

cracks and blisters in pressure vessels. These include

visual methods (VT), ultrasonic testing (UT) (shear

wave, longitudinal wave, and time-of-flight diffraction

[TOFD]), acoustic emission (AE) testing, liquid penetrant testing (PT), eddy current testing (ECT), wet

or dry magnetic particle testing (MT), wet fluorescent

magnetic particle testing (WFMT), and alternating current field measurement (ACFM). The usefulness of

these methods is dependent on the tightness, severity,

and location of cracks, as well as proper application of

the method, which includes a reasonable understanding of its benefits and limitations. The

results of all of these techniques are technician-

dependent. The following additional guidelines are

provided on NDE techniques used for detecting cracks

in pressure vessels exposed to wet H2S environments.

3.8.2 Wet Fluorescent Magnetic Particle Testing

(WFMT)

3.8.2.1 For surface-breaking cracks, WFMT is

sensitive, demonstrates reproducible results, and

is one of the most commonly used methods for

internal pressure vessel inspection.

3.8.2.2 Surfaces to be inspected shall be

prepared to a finish that will facilitate inspection

and not mask indications. In order to perform a

satisfactory WFMT inspection, the surface of the

weld and adjacent base metal for a distance of

about 150 mm (6 in.) on both sides should be

cleaned of all scale and residue. Care should be

taken to ensure that the surface preparation

method does not deform the metal surface and

mask indications. Abrasive blasting to a near-

white finish in accordance with NACE No.

2/SSPC(2)-SP 1011 should be performed. Other

methods, such as high-pressure water or CO2

blasting, may be used if they provide a suitable

surface for inspection. In some instances, the use

of flapper disc polishing has been necessary to

enhance the detection sensitivity of fine, tight

cracks.

3.8.2.3 Alternating current (AC) yoke WFMT

should be used instead of direct current (DC) or

prod methods. DC methods are not as sensitive

for surface-breaking cracks and prod methods

may leave arc strikes that, if not ground out, can

serve as crack initiators.

3.8.2.4 AC yoke WFMT is a sensitive technique

that may detect discontinuities not detected by

other NDE methods. Some indications may be

irrelevant. A representative number and type of

indications shall be evaluated to determine their

relevance and severity.

3.8.2.5 Surface preparation, magnetic particle

materials, magnetic testing equipment, processing

techniques, sequence of operation, levels of magnetizing fields, etc., should be monitored periodically to assure proper inspection. Methods

for checking system performance and sensitivity

are detailed in ASME(3) SE-70912 or equivalent.

3.8.2.6 Based on limited laboratory data and field

experience, there is concern that in certain instances, removal of protective scales associated

with surface preparation for WFMT may increase

the likelihood of cracking when the vessel is

returned to service.13 Depending on the severity

of the environment and specific startup conditions,

a short period of higher-than-normal hydrogen flux

that could lead to cracking in a susceptible base

metal or weldment may occur.

3.8.2.7 There are limitations to the use of WFMT

for the detection of cracking in wet H2S

environments. These include:

(a) WFMT requires internal vessel access, and

some areas of the vessel may be inaccessible.

(b) WFMT may not detect subsurface cracks

scales and requires cleanup.

(d) WFMT can reveal many small irrelevant

indications.

(e) Costs and time constraints are associated

with removal of internals (e.g., trays).

3.8.3 Ultrasonic Testing (UT)

3.8.3.1 UT methods (manual or automated) may

be used to detect subsurface cracking and blistering, for inspection on-stream, and for nonintrusive inspection from the external surface.

UT methods used include shear wave, longitudinal

wave, and time-of-flight diffraction. UT can be

used to evaluate blister size and depth and detect

deeper surface-connected defects (greater than

3.18 mm [0.125 in.] deep). Other than destructive

grinding of cracks, UT is the most frequently used

method for sizing cracks for fitness-for-service

evaluations. The use of external UT can alleviate

potential future hydrogen-charging concerns associated with cleaning ID surfaces for WFMT,

as stated in Paragraph 3.8.2.6.

___________________________

(2) Society for Protective Coatings (SSPC), 40 24th St., 6th Floor, Pittsburgh, PA 15222-4656.

(3) ASME International (ASME), Three Park Avenue, New York, NY 10016-5990.

RP0296-2004

3.8.3.2 There are limitations to the use of UT

methods for the detection of cracking and

blistering in wet H2S environments. It is difficult to

achieve consistently reliable interpretation of

results because of weld geometry, joint design,

shadowing effect of multiple defects, and the need

for more highly qualified NDE personnel experienced in the detection of wet H2S cracking.

Potential cost and time constraints include external

scaffolding, insulation removal and replacement,

surface preparation (e.g., grinding weld crowns),

and slow production rates.

3.8.3.3 Carefully planned and executed

automated UT procedures can provide some

advantages in mapping specific areas, compared

to manual UT, especially when follow-up inspection for crack growth is anticipated.

3.8.4 Acoustic Emission (AE) Testing

3.8.

4.1 AE testing is a global inspection method

that may be used to detect surface-breaking

cracks, subsurface cracks, and blisters. AE

testing is typically conducted when the equipment

is subjected to higher than normal tensile loading,

normally accomplished by hydrostatic testing,

pneumatic testing, or in-service pressurization

above normal operating pressure. AE testing can

be used both for inspection on-stream and for

nonintrusive inspection from outside the equipment during a shutdown.

3.8.

4.2 During AE testing, defect growth is

detected by an acoustic emission sensor array

attached to the external surface of the equipment.

The AE sensors transmit signals to a central

computerized data collection system. The data

are evaluated using software developed for this

purpose.

3.8.

4.3 There are limitations to the use of AE

testing for the detection of damage in wet H2S

environment. These include:

(a) AE testing detects only cracks that are active

during the conditions of the test. Therefore, the

absence of AE indications does not ensure that

the equipment is free of flaws.

(b) AE testing methods currently used in the

refining industry cannot discriminate the type or

nature of the defect and cannot determine the

defect size or exact location.

relatively high occurrence of false indications (or

“over-calls”). These can result from rain hitting the

sensors, mechanical rubbing/squeaking of equipment internals or attachments, flange leaks,

etc. AE testing personnel must be aware of and

take into account potential extraneous influences

and their effect on test results.

(d) AE testing requires considerable skill and

experience on the part of the personnel conducting

the test and evaluating the data. The availability of

both AE testing hardware and qualified personnel

can be limited.

(e) A stress analysis may need to be conducted

to ensure that the components of interest are

adequately stressed during the test.

3.8.

4.4 Because of the limitations stated above,

AE testing should not be used as a stand-alone

inspection method for the detection of cracking in

wet H2S environments. Follow-up inspection with

other appropriate NDE techniques should be performed on any significant AE source area that

potentially represents a location of cracking. AE

testing may be used as a global screening technique in conjunction with other NDE methods. 3.8.5 Alternating Current Field Measurement (ACFM)

3.8.5.1 ACFM is an electromagnetic technique

that can be used to detect and size surface-

breaking cracks in ferrous materials. The method

can be applied through thin coatings and does not

require extensive surface preparation.

3.8.5.2 ACFM is best used as a screening tool

for rapid detection of cracking along welds and/or

HAZs with little or no surface preparation. It can

be used in lieu of WFMT.

3.8.5.3 The sensitivity of ACFM to cracks

decreases with an increase in the coating thickness and loose scale on the examination

surface. ACFM can size crack length reliably. It

can also assess the depths of non-branched through-wall oriented cracks. However, its crack-

depth sizing can yield erroneous values when

ACFM is applied on highly branched, closely spaced, or tilted (i.e., not exactly in the through-

wall direction) cracks, such as amine stress corrosion cracks.

3.8.5.4 ACFM data interpretation is much more

complicated than WFMT. Highly skilled, experienced operators are essential to the success of ACFM inspection.

3.8.6 Eddy Current Testing (ECT)

3.8.6.1 ECT can be used to detect surface-

breaking cracks. The method can be applied

through thin coatings and does not require extensive surface preparation.

3.8.6.2 ECT is best used as a screening tool. It

can be used in lieu of WFMT. It is not effective in

RP0296-2004

finding very shallow cracks (less than about 1.5

mm [0.06 in.] deep).

3.8.6.3 Increasing coating or scale thickness

decreases the sensitivity of ECT.

3.8.6.4 ECT data interpretation is simpler than

interpreting ACFM results. However, skilled

operators are required to get accurate results.

3.9 NDE Personnel Qualifications

3.9.1 NDE personnel performing nondestructive

examinations shall be those recognized by the owner/user as having been trained in accordance with

ASNT(4) SNT-TC-1A14 or equivalent, to a minimum of

Level I. Interpretation of indications detected by NDE

methods should be made by personnel trained to a

minimum of Level II or equivalent. Refinery inspectors

interpreting results and following up on repair procedures should be certified to API 510,3 NB-23,15 or

other applicable industry code or standard.

3.9.2 Personnel interpreting results, especially

characterization and sizing, should be familiar with the

features of these cracking mechanisms to minimize

errors in interpretation.

3.10 NDE Procedures

3.10.1 NDE procedures for crack detection by

methods outlined in Paragraph 3.8.1 should be in

accordance with the appropriate article in Section V of

the ASME Boiler and Pressure Vessel Code16 (e.g.,

Article 5 for UT, Article 7 for MT), or other applicable

industry code or standard. In addition, special procedures may be required for detection and sizing of

environmental cracking.

3.10.2 NDE procedures should be developed and

approved by personnel with a demonstrated understanding of potential damage morphologies and

with certification to ASNT Level III, or other qualified

personnel.

3.11 Determining the Extent and Magnitude of Cracking and Blistering

3.11.1 A representative number and type of linear

indications shall be explored for length and depth, using appropriate methods such as grinding, arc gouging followed by grinding, or ultrasonic sizing techniques.

3.11.2 A UT survey of areas with blisters or cracks

should be made to determine the extent of subsurface

blistering, HIC, and/or SOHIC. The area adjacent to welds should be targeted for this inspection.

3.12 Analysis of Inspection Results

3.12.1 Based on the extent and magnitude of

cracking and blistering, an evaluation of the need for repair, which may include a fitness-for-service analysis

in accordance with a recognized methodology such as

API RP 57917 or equivalent, shall be made by engineers or inspection personnel who are recognized

by the owner/user as qualified to make such evaluations. Flaws judged to be allowable by such an

evaluation may remain in the vessel with no repairs required. Increased monitoring or mitigation may be necessary.

3.13 Records

3.13.1 Permanent records of inspection results

should be maintained for the life of the vessel. The location, orientation, length, and depth of significant indications, blisters, and cracks should be documented.

3.14 Reinspection

3.1

4.1 Reinspection intervals should be based on

the risk that the pressure vessel represents to the refiner, recognizing prior inspection results, disposition

of indications, weld repairs or alterations, changing process conditions, or requirements of API 510 or other

applicable industry code or standard. In general, if the

risks are such that reinspection is warranted, the reinspection should be done using techniques discussed in this standard.

3.1

4.2 In assessing risk, other issues to be

considered include possible growth of subsurface damage, possible accelerated hydrogen flux due to surface cleaning prior to inspection, and changes to the

process environment that may change the hydrogen-

charging rate.

________________________________________________________________________

Section 4: Repair of Cracked or Blistered Vessels

4.1 The objective of this section is to provide guidelines for the repair of existing carbon steel pressure vessels that have experienced cracking and/or blistering when exposed to a petroleum refinery wet H2S environment. Decisions on the type of repair and procedure shall be made by engineers or inspection personnel who are recognized by the owner/user as qualified to make such evaluations.

___________________________

(4) American Society for Nondestructive Testing (ASNT), 4153 Arlingate Lane, Columbus, OH 43228.

RP0296-2004

4.2 All standards for repairs shall, as a minimum, not be less than those specified in the API 510 inspection code, or other applicable industry code or standard. All welding procedure specifications, procedure qualifications, and welder performance qualifications shall be in accordance with the requirements of the ASME Boiler and Pressure Vessel Code, Section IX,18 or other applicable industry code or standard.

4.3 In some cases, grinding or welding operations can cause cracks to initiate or propagate because of the hydrogen-charged nature of the steel. In such instances, a hydrogen bake-out procedure involving heating the area to a temperature above 204°C (400°F) and holding for up to four hours may be employed to try to improve weldability. Bake-out temperatures up to those required for full PWHT may be used for holding times shorter than specified for PWHT.

4.4 Repair of Blisters

4.4.1 Blisters may be evaluated in accordance with the

provisions of API RP 579, an equivalent fitness-for-

service document, or applicable industry code or standard. If it is determined that blisters are present to the extent that repairs are necessary, the following options may be utilized.

4.4.1.1 Surface blisters less than 50 mm (2 in.) in

diameter may be drilled to relieve the internal

pressure. Appropriate caution shall be taken to

protect the operator from injury during hydrogen

venting. An engineering analysis should be

performed prior to drilling blisters larger than 50

mm (2 in.) in diameter to ensure that the remaining

net section of metal will hold the internal pressure.

4.4.1.2 Blistered steel plates may be removed

from the vessel and replaced with new steel.

Hydrogen bake-out in accordance with Paragraph

4.3 may be required prior to thermal cutting or

welding.

CAUTIONARY NOTE: Blisters are typically filled

with molecular hydrogen, which will not diffuse

during the bake-out described in Paragraph 4.3 or

during PWHT. As a result, the blisters may grow

or rupture during the bake-out or PWHT. In

addition, molecular hydrogen remaining in blisters

after the bake-out may cause cracking during

subsequent repair or PWHT. High-temperature

hydrogen attack may also result from PWHT.19

4.5 Removal of Cracks

4.5.1 Cracks may be evaluated in accordance with the

provisions of API RP 579, an equivalent fitness-for-

service document, or an applicable industry code or standard. When it is determined that crack removal is necessary, cracks may be removed by any suitable method, such as grinding, arc gouging, etc. If arc gouging or another method that will heat the steel

above its lower critical temperature is used, subsequent grinding should be employed to remove all

heat-affected material.

4.5.2 The excavated area should be reinspected with

WFMT to ensure that all cracks have been removed.

4.5.3 Once the cracks have been removed, the need

for weld repair shall be determined based on the minimum required wall thickness or an engineering fitness-for-service analysis. Local areas thinned beyond the corrosion allowance may be acceptable under some conditions, such as those outlined in API

510, API RP 579, or other applicable industry code or

standard.

4.6 Blend Grinding Repairs

4.6.1 The cavities formed by removing the cracks that

are not subsequently weld repaired should be contoured to eliminate notches in accordance with the

provisions of API RP 579 or other applicable industry

code or standard. An appropriate taper or radius is recommended to avoid sharp edges that could act as

stress raisers and lead to further cracking, or confuse

interpretation of subsequent UT inspections.

4.7 Weld Repairs

4.7.1 If weld repairs are determined to be necessary,

they shall be made in accordance with a recognized code such as API 510, or other applicable industry code or standard.

4.7.2 Preheat should be applied to the repair area

when deemed necessary. When possible, the preheat

should be applied from the outside of the vessel and

measured on the inside. This method of applying preheat ensures that the required temperature has been achieved through the full material thickness.

4.7.3 Low-hydrogen welding electrodes should be

used and handled in accordance with ASME SFA-

5.1,20 or other applicable industry code or standard,

and the electrode manufacturer’s recommendations to

minimize the potential for delayed hydrogen (cold) cracking.

4.7.4 Arc strikes should be removed by blend grinding.

4.7.5 The repair area should be given a PWHT in

accordance with the ASME Boiler and Pressure Vessel

Code, Section VIII21 or other applicable industry code

or standard after welding, especially if PWHT was done

in original fabrication. Heat treatment at lower temperatures (below 593°C [1,100°F]) for longer times,

as allowed by the ASME code, should not be used.

4.7.

5.1 Certain microalloying elements that can

be present in pressure vessel steels can retard the

softening effect of PWHT. When microalloying

elements are known to be present, consideration

RP0296-2004

should be given to increasing the PWHT temperature such that suitable softening of the weld and HAZ is accomplished.

4.7.

5.2 PWHT of vessels containing blisters or

HIC can result in additional cracking. See Cautionary Note in Paragraph 4.4.1.2.

4.7.

5.3 As an alternative to conventional PWHT,

welding techniques that soften the HAZ, such as

temper bead welding in accordance with API 510,

may be used. This may control HAZ hardness, but

has no significant impact on residual stress levels.

Consideration may also be given to the mitigation

techniques outlined in Section 5.

4.7.6 Repair weld hardness control shall be in accordance with NACE Standard RP0472.

4.7.7 The repair area should be reinspected after welding and PWHT. When delayed hydrogen (cold) cracking is a concern, a minimum interval of 48 hours

should be provided between welding and final inspection. Any new cracks or defects found should be

repaired according to the steps outlined above. If a subsequent welded repair is required (repair of a repair), other remedial steps such as hydrogen bake-

out, higher preheat temperature, etc., should be considered.

4.7.8 Radiographic testing (RT) or UT according to

ASME Boiler and Pressure Vessel Code, Section VIII should be considered for major repairs.

4.7.9 Permanent records of repairs should be

maintained for the life of the vessel. All relevant information, such as location, size, and depth of repair,

repair method, preheat temperature (if any), and PWHT temperature and hold time (if any), should be recorded.

4.8 Replacing a section of the vessel or replacing the entire vessel are acceptable options to performing repairs.

________________________________________________________________________

Section 5: Mitigation Considerations for Operation

5.1 The objective of this section is to outline several methods that may be employed to decrease the likelihood or severity of cracking in wet H2S environments.

5.1.1 To the extent that modifications are acceptable

from a process standpoint, process changes that may decrease the likelihood of cracking include control of water carry-over into downstream pressure vessels, dilution or removal of corrosive constituents by water washing, or use of additives. The use of additives, such as polysulfide or other corrosion inhibitors, or the use of water washing to reduce concentrations of ammonium bisulfide and cyanide in the aqueous phase, have been shown to reduce hydrogen permeation at alkaline pH levels. Polysulfide provides

a resistant corrosion product film and converts

cyanides in the aqueous phase into thiocyanates.

(Note: Polysulfide does not react with cyanides in the gas phase.)

5.1.2 Corrosion inhibitors injected into the process

stream may decrease the corrosion reaction, which tends to lower the cathodic evolution of atomic hydrogen and hence lower the potential for hydrogen entry into the steel and subsequent blistering and cracking.

5.1.3 Organic or inorganic coatings may be used as a

barrier to corrosion. Care should be taken to select a suitable coating that will perform in the process environment and during shutdown operations such as depressurizing and steam-out. Periodic inspection and maintenance of the coating should be performed over the life of the vessel to ensure continuing protection. If there is evidence of coating deterioration, consideration should be given to inspecting the internal steel surfaces periodically for cracking.

5.1.4 A corrosion-resistant alloy in the form of cladding, weld overlay, or strip lining can be applied to the vessel interior as a permanent corrosion barrier.

5.1.4.1 Plate that is clad by the hot-rolling or

explosive-bonding process can be used for replacement parts of existing pressure vessels.

5.1.4.2 Weld overlay can be applied in situ or

installed as a replacement part.

5.1.4.3 Attachment of strip lining by welding can

also be used to cover existing areas of equipment.

Periodic inspection and maintenance of strip lining should be performed over the life of the vessel to ensure protection. Cracking of the strip-

lining attachment welds due to issues such as differential expansion may result in process fluids

entering the gap between the lining and the vessel. Subsequent cracking of the base metal beneath the lining can occur due to exposure to

wet H2S. Additionally, cracking into the base metal

at the termination point of the strip lining has been

shown to occur in laboratory testing.22 This latter

issue has not been shown to be a significant problem in actual service, however.

5.1.5 An on-stream corrosion-monitoring program may be employed to detect corrosion activity that may produce conditions leading to cracking. This also helps

RP0296-2004

to determine whether certain corrosion control methods, such as chemical injection, e.g., polysulfide injection, or water washing, are effective. The corrosion-monitoring program may include corrosion coupons and/or corrosion probes, hydrogen probes, and ultrasonic measurements.

________________________________________________________________________

References

1. NACE Publication 8X294 (latest revision), “Review of Published Literature on Wet H2S Cracking of Steels Through 1989” (Houston, TX: NACE).

2. NACE Publication 8X194 (latest revision), “Materials and Fabrication Practices for New Pressure Vessels Used in Wet H2S Refinery Service” (Houston, TX: NACE).

3. API 510 (latest revision), “Pressure Vessel Inspection Code: Maintenance Inspection, Rating, Repair, and Alteration,” (Washington, DC: API).

4. H.I. McHenry, D.T. Read, T.R. Shieves, “Failure Analysis of an Amine-Absorber Pressure Vessel,” Materials Performance 26, 8 (1987): p. 18.

5. J.P. Richert, A.J. Bagdasarian, C.A. Shargay, “Stress Corrosion Cracking of Carbon Steel in Amine Systems,” Materials Performance 27, 1 (1988): p. 9.

6. R.D. Merrick, “Refinery Experiences With Cracking in Wet H2S Environments,” Materials Performance 27, 1 (1988): p. 30

7. NACE Standard MR0103 (latest revision), “Materials Resistant to Sulfide Stress Cracking in Corrosive Petroleum Refining Environments” (Houston, TX: NACE).

8. NACE Standard RP0472 (latest revision), “Methods and Controls to Prevent In-Service Environmental Cracking of Carbon Steel Weldments in Corrosive Petroleum Refining Environments” (Houston, TX: NACE).

9. API RP 580 (latest revision), “Risk-Based Inspection” (Washington, DC: API).

10. API Publication 581 (latest revision), “Base Resource Document—Risk-Based Inspection” (Washington, DC: API).

11. NACE No. 2/SSPC-SP10 (latest revision), “Near-White Metal Blast Cleaning” (Houston, TX: NACE).12. ASME SE-709 (latest revision), “Standard Practice for Magnetic Particle Examination” (New York, NY: ASME).

13. API Publication 939-A, “Research Report on Characterization and Monitoring of Cracking in Wet H2S Service,” (Washington, DC: API).

14. ASNT SNT-TC-1A (latest revision), “Personnel Qualification and Certification in Nondestructive Testing,” (Columbus, OH: ASNT).

15. NBBPVI(5) NB-23 (latest revision), “National Board Inspection Code” (Columbus, OH: NBBPVI).

16. ASME Boiler and Pressure Vessel Code, Section V (latest revision), “Nondestructive Examination” (New York, NY: ASME).

17. API RP 579 (latest revision), “Fitness for Service” (Washington, DC: API).

18. ASME Boiler and Pressure Vessel Code, Section IX (latest revision), “Welding and Brazing Qualifications” (New York, NY: ASME).

19. J.L. Hau, C.H. Molina, “Hydrogen Damage Inspection and Evaluation of H2S Absorber Column,” CORROSION/92, paper no. 446 (Houston, TX: NACE, 1992).

20. ASME SFA-5.1 (latest revision), “Specification for Carbon Steel Electrodes for Shielded Metal Arc Welding” (New York, NY: ASME).

21. ASME Boiler and Pressure Vessel Code, Section VIII (latest revision), “Design and Fabrication of Pressure Vessels” (New York, NY: ASME).

22. API Publication 939-B (latest revision), “Repair and Remediation Strategies for Equipment Operating in Wet H2S Service” (Washington, DC: API).

________________________________________________________________________

Bibliography

Bartz, M.H., and C.E. Rawlins. “Effects of Hydrogen

Generated by Corrosion of Steel.” Corrosion 4, 5

(1948): p. 187.

___________________________

(5) National Board of Boiler and Pressure Vessel Inspectors (NBBPVI), 1055 Crupper Avenue, Columbus, OH 43229-1108.

RP0296-2004

Berkowitz, B.J., and H.H. Horowitz. “The Role of H2S in the Corrosion and Hydrogen Embrittlement of Steel.”

Journal of Electrochemical Society 129, 3 (1982): p.

468.

Bulla, J.T., and J.T. Chikos. “Case History—FCCU Absorber Deethanizer Tower Hydrogen Blistering and

Stepwise Cracking.” CORROSION/89, paper no. 264.

Houston, TX: NACE, 1989.

Cayard, M.S., R.D. Kane, L. Kaley, and M. Prager.

“Research Report on Characterization and Monitoring

of Cracking in Wet H2S Service.” API Publication 939.

Washington, DC: API, October 1994.

Gutzeit, J. “Process Changes for Reducing Pressure Vessel Cracking Caused by Aqueous Sulfide Corrosion.” Materials Performance 31, 5 (1992): p. 60. Hildebrand, E.L. Aqueous Phase H2S Cracking of Hard Carbon Steel Weldments—A Case History, Proceedings API, vol. 50 (III). Washington, DC: API,

1970: p. 593.

Kotecki, D.J., and D.G. Howden. Weld Cracking in a Wet Sulfide Environment, Proceedings API, vol. 53 (III).

Washington, DC: API, 1973: p. 573.

Kotecki, D.J., and D.G. Howden. Wet Sulfide Cracking of Submerged Arc Weldments, Proceedings API, vol. 52

(III). Washington, DC: API, 1972: p. 631.

Merrick, R.D. “Refinery Experiences with Cracking in Wet H2S Environments.” Materials Performance 27, 1 (1988): p. 30.

Merrick, R.D., and M.L. Bullen. “Prevention of Cracking in Wet H2S Environments.” CORROSION/89, paper no.

269. Houston, TX: NACE, 1989.

Neill, W.J. Jr. “Prevention of In-Service Cracking of Carbon Steel Welds in Corrosive Environments.” Materials Protection and Performance 10, 8 (1971): p. 33. Schuetz, A.E., and W.D. Robertson. “Hydrogen Absorption, Embrittlement, and Fracture of Steel.” Corrosion 13, 7

(1957): p. 437t.

Schutt, H.U. “Intergranular Wet Hydrogen Sulfide Cracking.” Materials Performance 32, 11 (1993): p. 55. Schutt, H.U. “New Aspects of Stress Corrosion Cracking in Monoethanolamine Solutions.” Materials Performance

27, 12 (1988): p. 53.

Van Gelder, K., M.J.J. Simon Thomas, and C.J. Kroese.

“Hydrogen Induced Cracking: Determination of Maximum Allowed H2S Partial Pressures.” Corrosion

42, 1 (1986): p. 36.

________________________________________________________________________

Appendix A

Cited Codes, Specifications, and Standards

ASME International Boiler and Pressure Vessel Code

SE-709 Standard Practice for Magnetic Particle Examination (latest revision)

Section II, Part C Specifications for Welding Rods, Electrodes, and Filler Metals

Section V Nondestructive Examination

Section VIII Rules for Construction of Pressure Vessels

Section IX Qualification Standard for Welding and Brazing Procedures, Welders, Brazers, and Welding and Brazing Operators

ASTM International(6)

A 53/A 53M Standard Specification for Pipe, Steel, Black and Hot-Dipped, Zinc-Coated, Welded and Seamless (latest revision)

___________________________A 70 Specification for Low and Intermediate Tensile Strength Carbon Steel (discontinued 1947—replaced by A 285)

A 106 Standard Specification for Seamless Carbon Steel Pipe for High-Temperature Service (latest revision)

A 201 Specification for Carbon-Silicon Steel Plates of Intermediate Tensile Ranges for Fusion-Welded Boilers and Other Pressure Vessels (discontinued 1967—replaced by A 515, A 516)

A 212 Specification for High Tensile Strength Carbon-Silicon Steel Plates for Boilers and Other Pressure Vessels (discontinued 1967—replaced by A 515, A 516)

A 285/A 285M Standard Specification for Pressure Vessel Plates, Carbon Steel, Low- and Intermediate-Tensile Strength (latest revision)

(6) ASTM International, 100 Barr Harbor Dr., West Conshohocken, PA 19248-2959.

RP0296-2004

A 515/A 515M Standard Specification for Pressure Vessel Plates, Carbon Steel, for Intermediate- and Higher-Temperature Service (latest revision)

A 516/A 516M Standard Specification for Pressure Vessel Plates, Carbon Steel, for Moderate- and Lower-Temperature Service (latest revision)

NACE International

MR0103 Standard Material Requirements, “Materials Resistant to Sulfide Stress Cracking in Corrosive Petroleum Refining Environments” (latest revision)

NACE No. 2/

SSPC-SP 10 “Near-White Metal Blast Cleaning” (latest revision)

RP0472 Standard Recommended Practice, “Methods and Controls to Prevent In-Service Environmental Cracking of Carbon Steel Weldments in Corrosive Petroleum Refining Environments” (latest revision)

American Society for Nondestructive Testing (ASNT)

SNT-TC-1A Recommended

Practice,

Personnel Qualification and Certification in Nondestructive Testing (latest revision)

American Petroleum Institute (API)

API 510 Pressure Vessel Inspection Code—Maintenance Inspection, Rating, Repair, and Alteration (latest revision).

API RP 579 Fitness-for-Service (latest revision)

API RP 580 Risk-Based Inspection (latest revision)

API

Publication 581 Base Resource Document—Risk-Based Inspection (latest revision)

________________________________________________________________________

Appendix B

Nature and Extent of Problem—Results from 1990 Industry Survey

B1 The objective of this section is to report the frequency and severity of cracking of existing carbon steel pressure vessels in petroleum refinery wet H2S environments.

B2 Survey of Inspection Results

B2.1 A survey was conducted in 1990 by Work Group T-8-16a to determine the nature and extent of cracking problems in wet H2S environments in the petroleum refining industry. There was insufficient information reported about the type of cracking found to correlate cracking incidence with cracking mechanism. In addition to asking for crack inspection results, the survey requested information about original fabrication details, service environment, prior inspection history, and disposition of cracked vessels (e.g., type of repairs, replacement).

B2.2 The use of various inspection techniques, such as visual inspection for hydrogen blisters and magnetic particle testing and liquid penetrant testing for crack detection, was reported. However, most of the inspections for cracks were conducted using WFMT, which is a very sensitive inspection technique for detection of surface discontinuities. Therefore, in addition to detecting service-related cracks, a number of linear indications that may have been flaws present from original fabrication, repair, or alteration of the pressure vessels were found. Subsurface cracks may not be detected by this method. There was no uniformity in reporting of discontinuities; some companies reported all discontinuities, some excluded obvious fabrication flaws, and others excluded very shallow indications that could be easily ground out. In the context of this section of the standard, the terms cracks and cracking refer to all linear indications reported by the survey respondents.

B2.3 Survey responses covering inspection results for almost 5,000 pressure vessels were received. Overall, cracking was reported in 26% of the inspected pressure vessels, as shown in Table B1.

TABLE B1

Overall Summary

Number of Pressure Vessels Inspected 4,987 Number of Pressure Vessels Cracked 1,285 Cracking Incidence 26%

B2.4 Cracking incidence reported by different companies varied from a low of 10% to a high of 73%. The cracking incidence reported by each company is shown in Table B2.

RP0296-2004 TABLE B2

Company Breakdown

Company Number Inspected % Cracked

A 45 47

B 1,358 13

C 968 11

D 55 15

E 179 20

F 104 43

G 85 21

H 129 34

I 172 16

J 71 10

K 141 73

L + Misc. 1,680 41

B2.5 The task group believes that a number of factors have influenced the wide disparity in cracking incidence reported by various companies, although not all were factors considered in the survey. These include, but are not limited to, differences in (1) surface preparation prior to inspection; (2) extent of inspection; (3) reporting of cracks, e.g., whether fabrication flaws or shallow indications were excluded; (4) process units inspected;

(5) crude feed compositions; and (6) original fabrication practices. B2.6 Cracking was reported in pressure vessels in essentially all refinery process units with wet H2S environments. Table B3 shows the cracking incidence in each of the common refinery process units.

B2.7 Cracking incidence varied from a low of 18 to 19% in crude units and coker fractionation units to a high of 45% in fluid catalytic cracking unit (FCCU) light-ends sections. Other process units also experiencing high cracking incidence include FCCU fractionation (41%), liquefied petroleum gas (LPG) (41%), and atmospheric light ends (38%).

TABLE B3

Cracking by Process Unit

Process Unit Number Inspected % Cracked

Crude Coker Fractionation 300

Vacuum Amine Other

574

364

Hydrotreating Sulfur Recovery

Hydrocracking Sour Water Stripper Amine/Caustic 368

156

132

811

Coker Light Ends

Flare Catalytic Reformer

134

Atmospheric Light Ends 140 38

FCCU Fractionation

LPG 252

FCCU Light Ends 704 45

B2.8 Some companies reported that a significant percentage of cracking detected in FCCU fractionation was carbonate cracking, a form of ASCC in alkaline sour waters with high carbonate/bicarbonate concentration. Cracking was also prevalent in amine and caustic services. B2.9 There is not a strong correlation in the survey data between cracking incidence and operating temperature. Throughout the entire range of operating temperatures, the cracking incidence only varied between 23% and 37%, as shown in Table B4. The highest cracking incidence occurred in the 65 to 93°C (150 to 200°F) operating temperature range.

RP0296-2004

TABLE B4

Cracking vs. Operating Temperature

Operating Temperature

°C °F

Number Inspected % Cracked

<38 <100 284 23

38-65 100-150 926 34

65-93 150-200 385 37

93-121 200-250 312 33

121-149 250-300 237 27

>149 >300 356 29

B2.10 In general, cracking incidence increased with increasing H2S concentration in the water phase, as shown in Table B5. The most noteworthy observation is the 17% cracking incidence for pressure vessels in services containing less than 50 mg/L (50 ppmw) H2S dissolved in an aqueous phase. This is considered a high rate for a service environment previously thought not to be a concern. However, inclusion of fabrication-related flaws in some of the survey responses probably had an impact on this cracking incidence. The practical difficulty of measuring actual concentration of H2S in the aqueous phase, especially at low concentrations, also might have had an impact.

TABLE B5

Cracking vs. H2S Concentration

H2S Concentration

(mg/L [ppmw])

Number Inspected % Cracked

<50 94

50-250 309

250-500 35

500-1,000 76

1,000-2,500 134

2,500-5,000 83

5,000-10,000 137

>10,000 378

B2.11 The cracking incidences for refinery pressure vessels fabricated from the most commonly used ASTM specification steel materials are listed in Table B6. There was no apparent correlation between cracking incidence and the specification of the steel plates used for fabrication of pressure vessels in wet H2S service. It was evident that the cracking incidence in pipe steels, such as ASTM A 53 and A 106, was much lower than that in plate steels.

TABLE B6

Cracking vs. Steel Specification

ASTM Steel Specification Number Inspected % Cracked

A 70 230 20

A 201 102 32

A 212 277 30

A 285 907 29

A 515 293 36

A 516 681 27

A 53 129 9

A 106 103 12

B2.12 The cracking incidences for refinery pressure vessels fabricated from steel plates of the commonly used steel grades (with corresponding minimum tensile strength) are listed in Table B7. There was no apparent trend between cracking incidence and plate steel grade. The lowest cracking incidence was experienced with grade 60 materials, but this was based on far less data than those for grades 55 and 70.

RP0296-2004

TABLE B7

Cracking vs. Steel Grade

Steel Grade(A)Number Inspected % Cracked

Grade 55 1,187 28

Grade 60 202 22

Grade 65 35 31

Grade 70 1,085 31

(A)Steel grade levels correspond to minimum tensile strength requirements; e.g., ASTM A 285 grade C is included in grade 55.

B2.13 The cracking incidences for the two most common plate steel materials were 29% for A 285 grade C, and 27% for A 516 grade 70. Among the steel plate materials for which there were at least 100 inspection results reported, the highest cracking incidence (36%) was experienced by pressure vessels fabricated with ASTM A 515 grade 70 steel.B2.14 Table B8 lists the cracking incidence for pressure vessels with and without PWHT. The cracking incidence for pressure vessels with PWHT (25%) was only marginally lower than that for pressure vessels without PWHT (30%). The survey data included some fabrication flaws, as well as hydrogen blistering and HIC that would not be expected to benefit from PWHT. PWHT would be expected to be beneficial for resistance to SSC, SOHIC, and ASCC.

TABLE B8

Cracking vs. PWHT

Pressure Vessel Condition Number Inspected % Cracked

Non-PWHT 2,325

PWHT 1,132

B2.15 A strong correlation between cracking incidence and the hydrogen blistering history of the pressure vessel was evident, as shown in Table B9. Pressure vessels with a history of hydrogen blistering had about twice the cracking incidence (54%) of pressure vessels with no prior history of blistering (25%).

TABLE B9

Cracking vs. Blistering History

Pressure Vessel Condition Number Inspected % Cracked

No Blisters 2,256 25

Blisters 216

B2.16 The presence of weld repairs in pressure vessels did not appear to have a significant effect on cracking incidence, as shown in Table B10. Pressure vessels with prior weld repairs had only a marginally higher cracking incidence (30%) than pressure vessels without weld repairs (27%).

TABLE B10

Cracking vs. Weld Repairs

Pressure Vessel Condition Number Inspected % Cracked No Weld Repairs 2,022 27

Weld Repairs 506 30

B2.17 The maximum depth of cracking reported is shown in Table B11. Only 38% of the cracked pressure vessels experienced cracking with a maximum depth of less than 3.18 mm (0.125 in.). Conversely, over 60% of the cracked pressure vessels experienced cracking deeper than 3.18 mm (0.125 in.). About 20% of the cracked pressure vessels had cracking deeper than 9.53 mm (0.375 in.).

RP0296-2004

TABLE B11

Depth of Cracking

Crack Depth

mm in.

Number of Vessels Percent

<1.59 <0.0625 83 12

1.59 to 3.18 0.0625 to 0.125 185 26

3.18 to

4.78 0.125 to 0.188 67 10

4.78 to 6.35 0.188 to 0.250 124 18

6.35 to 9.53 0.250 to 0.375 99 14

9.53 to 12.7 0.375 to 0.500 45 6

12.7 to 19.1 0.500 to 0.750 77 11

19.1 to 25.4 0.750 to 1.00 10 1

>25.4 >1.00 14 2

B2.18 Crack penetration, a ratio calculated by dividing the maximum depth of cracking by the wall thickness of the pressure vessel, is shown in Table B12. About 40% of the cracked pressure vessels experienced less than one-quarter penetration through the wall thickness. About 40% of the cracked pressure vessels experienced cracking more than halfway through the wall thickness.

TABLE B12

Crack Penetration

(% of Wall

Thickness)

Number of Vessels Percent

<10 59

10 to 24 120 26

25 to 49 106 23

50 to 74 99 22

75 to 99 37 8

100 36

B2.19 Table B13 summarizes the reported disposition of the pressure vessels that were found to contain cracks. In 43% of the cracked pressure vessels, the cracks were sufficiently shallow that they could be ground out and weld repairs were not required to restore vessel integrity. About 38% required weld repairs to restore vessel integrity. One of five was replaced after the inspection or a replacement was planned for the near future.

TABLE B13

Disposition of Cracked Pressure Vessels Pressure Vessel Disposition Number Percent Cracks Ground Out 476 43

Weld Repaired 426 38 Replaced (Done/Planned) 214 19

B3 Summary of Observations

B3.1 The extent and magnitude of cracking of pressure vessels in wet H2S environments is a significant concern in the petroleum refining industry. B3.2 All process units containing wet H2S environments appear to be affected, but to varying degrees.

B3.3 There is a strong correlation between cracking incidence and a history of hydrogen blistering in pressure vessels.

RP0296-2004

B3.4 There is not a strong correlation between cracking incidence and any singular process parameter, such as operating temperature or stream chemistry, in this particular survey. Nevertheless, recommendations in Paragraph 3.6.2 (d), which also take industry experience into account, are valid.

B3.5 There does not appear to be a strong correlation between cracking incidence and material and fabrication factors such as specification and grade of steel commonly used for pressure vessels, PWHT, or prior weld repairs in this particular survey. However, it is generally recognized that PWHT is beneficial for prevention of certain types of cracking mechanisms (see Paragraph 2.4.3.2 and recommendations in Paragraphs 3.6.2 [b] and 4.7.5).

B3.6 A significant (17%) incidence of cracking was reported in pressure vessels exposed to process environments with less than 50 mg/L (50 ppmw) H2S in the aqueous phase. But, inclusion of fabrication-related flaws by some respondents or difficulty in measuring low concentrations of H2S in the aqueous phase probably had an impact on this result.

B3.7 The number of pressure vessels involved and the required corrective action demonstrates the potential impact of this problem on refinery production losses and maintenance costs. However, despite the high incidence of cracking reported in this survey, few in-service failures of carbon steel vessels in wet H2S environments have been reported.

________________________________________________________________________

Appendix C

Typical Cracks Found in Wet H2S Environments

Figure C1: SSC in HAZ of head-to-shell weld of FCCU absorber tower. The crack is on the A 516-70 shell side. The numbers in the photograph are Knoop hardness values. (Nital etch)