http://www.hydrocarbonprocessing.com/Article/3395041/Is-your-plant-vulnerable-to-a-brittle-fracture.html?

Print=true

Copying and distributing are prohibited without permission of the

publisher
Is your plant vulnerable to a brittle fracture?
11.01.2014  |  Macejko, B. ,  The Equity Engineering Group, Inc., Shaker Heights, Ohio

Keywords:

Much of the process equipment operating today was designed to construction codes that did not
require a formal evaluation for low-temperature considerations. Metal temperature highly
influences the fracture toughness of construction materials of plant equipment. At low
temperatures, some materials tend to behave in a brittle manner, making it much more
susceptible to fracture. The author discusses methods to identify potential brittle-fracture
conditions in process equipment before failures occur.

BACKGROUND

The majority of pressure equipment used in the refining and petrochemical industry is
constructed of carbon or low-alloy steel. Metal temperature highly influences the fracture
toughness of these materials. At low temperatures, construction materials can behave in a brittle
manner (i.e., like glass), making them more susceptible to fracture. At high temperatures, the
materials tend to behave in a ductile fashion. Pressure vessels and piping may experience low
temperatures from the ambient environment or from operating and upset conditions.

The potential for auto-refrigeration from depressurization of liquefied compressed gases can be
particularly concerning due to the extremely low equipment metal temperatures (–55°F and
below). An engineering evaluation is typically required to assess whether low-temperature
operating or upset conditions could result in a brittle-fracture failure of pressure equipment.

Brittle fracture

This is the sudden and rapid propagation of a crack-like flaw under stress (residual or applied)
where the material exhibits little or no evidence of ductility or plastic deformation.1 Although
rare, the consequences of a brittle fracture are typically catastrophic. Brittle-fracture failures
experienced within industry have resulted in costly unplanned repairs, extensive production
downtime and loss of life (Fig. 1).

  Fig. 1. Ductile vs. brittle fracture failures.a

There can be a large variation in the way owner-users assess the risk for brittle fracture. Owner-
users that have previously experienced such a failure may have detailed and comprehensive
programs in place to evaluate susceptibility to future issues. Alternatively, owner-users fortunate
enough to have avoided such a failures tend to be less cognizant of potential risks. Just because a
failure has not occurred yet, it does not mean it will not happen tomorrow.

WHY EVALUATE BRITTLE FRACTURE

Much of the plant equipment operating today was designed to codes of construction that did not
require a formal evaluation for low-temperature considerations. However, catastrophic failures
and subsequent root-cause investigations have revealed deficiencies in code requirements.
Present process equipment design and post-construction codes and standards have taken action to
address the industry need to assess for brittle fracture.

OSHA 1910.119 process safety management requirements do not allow owner-users to ignore
deficiencies in the original code of construction:2

 1910.119(d)(3)(ii)—The employer shall document that equipment complies with

recognized and generally accepted good engineering practices.2
 1910.119(d)(3)(iii)—For existing equipment designed and constructed in accordance
with codes, standards or practices that are no longer in general use, the employer shall
determine and document that the equipment is designed, maintained, inspected, tested,
and operating in a safe manner.2

Owner-users should consider a brittle fracture evaluation for any pressure-retaining equipment
item where:

 Original design did not consider susceptibility to brittle fracture

 A change in process operating conditions increases the possibility that low-metal
temperatures has occurred
 A PHA or HAZOP identifies process or ambient temperatures lower than anticipated in
the original design, i.e., depressurization/auto-refrigeration potential
 The owner-user desires to rerate using a lower design margin
 The owner-user desires to optimize the timing of startups or shutdowns
 The owner-user intends to complete a hydrostatic pressure test at a temperature lower
than the original
test temperature.

KEY COMPONENTS

There are three key components that drive a brittle-fracture failure:

 Stress (residual and/or applied)

 Material toughness
 Crack-like defect/flaws.
A combination of metallurgical structure, residual stress and inherent defects that act as stress
intensifiers (lack of fusion, lack of penetration, porosity, and slag inclusions make welds and
weld-heat-affected zones critical locations for potential brittle fracture.

Stress. Stress provides the energy necessary to drive a defect to fracture. Typical stress sources
include pressure, weight and thermal loads, in addition to residual stress from the welding
processes. A post-weld-heat treatment (PWHT) operation will significantly reduce weld-residual
stress.

Material toughness

Metal temperature greatly influences material toughness in carbon steel (CS) and low-alloy
steels. At low temperatures, the construction materials can behave in a brittle manner, and they
have a high susceptibility to fracture. Conversely, the materials act in a ductile fashion at high
temperatures. The fracture-appearance-transition temperature (FATT) is defined as the
temperature corresponding to 50% shear. In Fig. 2, the FATT may be approximated as point D in
the plot of Charpy impact energy as a function of temperature. Additionally, the material
chemistry, grain size and heat treatment all affect toughness. The brittle-to-ductile-transition
temperature will decrease (i.e., improve) with:

 Decrease in material carbon content

 Increase in manganese to carbon ratio
 Decrease in sulfur content
 Decrease in material grain size
 Normalization heat treatment of material through rapid cooling.

 
  Fig. 2. Plot of Charpy impact energy FATT.

Partly due to chemistry control issues, older (or dirty) steels often have a moderate brittle-to-
ductile-transition slope relative to clean steels (Fig. 3). Most of the common steels used to
construct pressure equipment in the 1970s, 1960s and earlier (such as A70, SA-212, SA-201,
SA-283, SA-285 and SA-515-70) have exhibited relatively high FATT (i.e., low toughness). A
number of environmental issues exist that can degrade material toughness or result in
embrittlement. The adverse effects of any such potential damage mechanisms should be
considered in a brittle-fracture evaluation.

 
  Fig. 3. Brittle-to-ductile transition
  for “old” steels.

Crack-like defect

Brittle fracture typically initiates at a crack-like defect. These defects can result from
environmental damage (such as exposure to wet hydrogen sulfide or caustic), mechanical
damage (such as gouges or dents), or from original fabrication (such as laminations, lack of
fusion, lack of penetration, slag inclusions, and porosity). Performing a detailed inspection
including both surface examination techniques (dye penetrant or magnetic particle examinations)
and volumetric examination techniques (angled-beam ultrasonic methods) can be used to detect
and categorize any crack-like defects present in pressure equipment.

ASSESSING SUSCEPTIBILITY TO BRITTLE FRACTURE IN PRESSURE

VESSELS

Industry codes and standards assess susceptibility to brittle fracture by comparing a critical
exposure temperature (CET) to a minimum allowable temperature (MAT). The CET represents
the driving force for fracture and consists of the lowest-potential metal temperature from all
operating, upset or atmospheric conditions. The MAT represents the material resistance to
fracture, either calculated through engineering evaluation techniques or assigned based on
destructive testing (i.e., Charpy impact testing). Both the CET and the MAT can consist of a
single temperature and coincident pressure or an envelope of temperature and pressure
combinations.

ASME Section VIII Division 1

Most pressure vessels at petrochemical facilities were designed to ASME Section VIII Division
1 (ASME S8D1).3 The 1987 Addenda to the 1986 Edition introduced drastic changes to Part
UCS-66 due to brittle fracture concerns with CS and low-alloy steels. Prior to this, ASME S8D1
permitted operation of CS equipment to a temperature of –20°F without testing. Considering that
no new refinery has been built in the US since the 1970s, the majority of pressure equipment
operating today did not receive a formal assessment for susceptibility to brittle fracture.
Currently, ASME S8D1 provides a series of impact-test exemption curves to assign a baseline
MAT based on the material of construction and weld-joint governing thickness. ASME S8D1 also
permits application of a temperature reduction if the stress levels are below the design allowable
stress.

API 579-1/ASME FFS-1 Fitness-For-Service

The API 579-1/ASME FFS-1 (API-579) Fitness-For-Service (FFS) standard provides procedures
and criteria to evaluate equipment in post-construction service.4 The Part 3 Level 1 and Level 2
methods used to screen for the propensity for brittle fracture are consistent with the ASME S8D1
design philosophy. The Level 3 method (which references Part 9 of API-579) includes a detailed
fracture mechanics evaluation. Characteristics of the three levels include:

Level 1:

 Typically completed by an inspector or a plant engineer

 Table or chart exemption curve lookup based on material of construction and joint
governing thickness.

Level 2:

 Typically completed by a plant engineer

 Screening method extension of the Level 1 approach with more detailed and prescriptive
calculations
 Three analysis method options are provided and described here. However, typically,
Method A is the
only practical and reasonable option:

o Method A. Calculation of a safe envelope based on an evaluation of actual stress
compared to design allowable stress (i.e., trade lower stress levels for lower
permissible temperatures).
o Method B. Determine the MAT based on the metal temperature at time of
hydrotest. This method is not practical or feasible for many instances. Substantial
risks exist for fracture during the hydrotest.
 o Method C. Grandfathering approach based on a proof-test argument. This method
assumes that if it can be shown that the combination of worst-case applied stress,
temperature and flaw size has already occurred and did not result in a failure, then
other less-severe conditions will not result in a failure. This method requires
detailed historical operating, inspection and repair information that is not
available in many instances.

Level 3:

 Completed by an engineering specialist with extensive experience in FFS.

 Most reliable and accurate method for establishing a MAT because it involves a detailed
fracture mechanics evaluation
 Required when the equipment operates in a service where brittle fracture is a legitimate
concern or if a crack-like flaw is known to exist

If the equipment has undergone PWHT at original construction and all subsequent repairs or
alterations have not adversely compromised the effectiveness of the PWHT, then the FFS
analysis calculation procedures allow credit for the positive benefit from the stress-relief
operation on the calculated MAT. PWHT is often critical for low temperature (below –55°F)
acceptability.

CURRENT CODES AND STANDARDS DEFICIENCIES

A number of deficiencies and inconsistencies exist with the ASME S8D1 and API-579 Level 2
FFS methods. All of these items enforce the requirement to complete a Level 3 Part 9 fracture
mechanics evaluation if brittle fracture is a legitimate concern.

Applied stress threshold and impact testing of weldments

Both ASME S8D1 and the Part 3 Level 2 procedure in API-579 assign a MAT of –155°F to a
component if the calculated stress ratio falls below a certain threshold (that depends upon the
original code of construction factor for safety). However, Part UCS-67(c)(3) of ASME S8D1
requires mandatory impact testing of welds to qualify metal temperature below –55°F. This
weldment impact test requirement applies regardless of the stress ratio. It could be argued that
the API-579 Level 2 procedure should also require the weldment impact test.

Additionally, API-579 implies that, for a component thickness less than 2 in. and a general
primary membrane tensile stress less than 8 ksi, brittle fracture will not occur. Although this may
be true for equipment that has had PWHT performed, failures of non-PWHT weld joints due to
weld-residual stress alone have occurred. Therefore, the 8-ksi primary stress threshold may not
be appropriate in all instances.

PWHT temperature credit

For P-1 Group 1 and P-1 Group 2 materials, both ASME S8D1 and API-579 (Level 1 or 2) allow
a temperature reduction of 30°F from the component MAT if PWHT was completed and the
component thickness does not exceed 1.5 in. The origin and engineering justification for the
30°F reduction are unknown. As noted in WRC 528, estimates of the weld-residual stress were
directly considered in the ASME Section VIII Division 2 (ASME S8D2) exemption curves
development.5,6 Thus ASME S8D2 provides different exemption curves for PWHT and non-
PWHT materials. The ASME S8D2 approach appears more technically appropriate when
compared to the general 30°F reduction.

NBIC alternative weld methods

The National Board Inspection Code (NBIC) provides a series of alternative welding methods
without PWHT.7 These include various combinations of preheat temperatures, weld-interpass
temperatures and temper-bead procedures. Although these welding methods may help to slightly
temper the material, they do not significantly reduce weld-residual stress when compared to a
proper PWHT stress-relief operation. Therefore, without further study, it is difficult to justify
credit for a full PWHT in a brittle-fracture evaluation if one of these alternative welding methods
were used.

CASE STUDY 1

This example illustrates the use of the Level 2 FFS methodology for brittle-fracture screening of
typical “pots and pans” pressure vessels that meet these limitations:

 Do not operate in cyclic service

 Not susceptible to any shock-chilling event
 Not susceptible to any environmental cracking
 Do not operate in a service that may result in loss of material toughness
 Have no known crack-like flaws
 Where excessive low-temperature operation (below –55°F) is not feasible.

In this case history, a service review identified that the knockout (KO) drum has the potential
under ambient conditions to reach a minimum metal temperature of –10°F. Because the Level 1
analysis for the vessel indicated a baseline MAT of 53°F, a Level 2 FFS evaluation was
completed. Table 1 summarizes the construction criteria for this KO drum.

 
Analysis

Level 2 FFS calculations were performed in accordance with Part 3, Method A of API-579 using
proprietary software. The Level 1 MAT for each component was obtained using the component
governing thickness and respective material curve. In the Level 2 calculations, credit was taken
for any additional plate thickness, above the minimum required thickness, through the use of a
stress ratio. The evaluation then assigned a temperature reduction based on the calculated stress
ratio. The final MAT was determined as the Level 1 MAT minus the temperature reduction.

Results

A Level 2 FFS evaluation established the maximum permissible pressure of 174 psig at the
client-specified CET of –10°F. Additionally, the evaluation provided an envelope of permissible
pressure-temperature combinations that the owner-user could compare against all operating and
upset scenarios (Fig. 4). This enabled the owner-user to develop operational procedures to ensure
satisfactory protection from a potential brittle-fracture event.

  Fig. 4. API-579 Part 9 FAD diagram.

  Fig. 5. Level 2 MAT curve results
  for Case Study 1.

CASE STUDY 2

This example illustrates the use of the Level 3 FFS methodology for brittle-fracture evaluation of
a deethanizer column that operates in a service with auto-refrigeration potential. Table 2
summarizes the operating conditions for the deethanizer column.

Analysis

A rigorous, fracture mechanics-based assessment in accordance with the Level 3 procedures and
criteria in Part 9 of API-579 was performed to establish the allowable pressure-temperature curve
using proprietary software. The analysis performed assumed a detectible reference flaw size that
could be compared to the results of detailed inspection for cracking. A comprehensive, 3D finite-
element (FE) model was used to characterize operating stresses (Fig. 6).
  Fig. 6. FEA stress results plot at local
discontinuity.

The failure assessment diagram (FAD) fracture mechanics methodology presented in Part 9 of
API-579 (Fig. 4) was used to calculate the allowable pressure-temperature curve. A postulated
semi-elliptical surface-breaking flaw with a depth of 1⁄4 of the nominal shell thickness and a
length of six times the depth (i.e., 6:1 aspect ratio) was analyzed in various orientations at critical
locations. The presence of such large flaws, while unlikely, should be reliably detected by
modern inspection techniques (including surface and ultrasonic inspection techniques). The
calculated operating and weld-residual stress, along with the material strength and material
fracture toughness, were used to determine a toughness ratio and a load ratio. These two
quantities represent the coordinates of a point that was plotted on a 2D FAD to determine
acceptability.

Results

A Level 3 FFS evaluation established the envelope of permissible pressure vs. temperature
curves, as shown in Fig. 7. This enabled the owner-user to develop operational procedures to
ensure satisfactory protection from a potential brittle-fracture event. The bottom curve in this
figure shows the boiling point curve for the process fluid, ethane.

  Fig. 7. Level 3 MAT curve results

  for Case Study 2.
CASE STUDY 3

This example illustrates the difference in results from a Level 2 FFS vs. a Level 3 FFS for brittle-
fracture evaluation of a pressure vessel that did not receive a PWHT, but operates in a service
with an auto-refrigeration potential. Table 3 summarizes details for a pressure vessel subject to
auto-refrigeration with no PWHT.

Analysis

The assessment included a Part 3 Level 2 FFS brittle-fracture screening evaluation in addition to
a detailed Part 9 Level 3 FFS fracture mechanics evaluation using proprietary software. The
methodologies for the Level 2 and Level 3 evaluations were consistent with those detailed in
Case Studies 1 and 2, respectively.

Results

The results from the Level 2 FFS evaluation and the Level 3 FFS evaluation (for the
recommended postulated flaw depth of 1⁄4 of nominal thickness) both indicate that the vessel
does not satisfy the requirements for protection against brittle fracture, as shown in Fig. 8.
However, the evaluation shows that the Level 2 results are more favorable than the Level 3
results for pressures below 350 psig. Even if the Level 3 analysis uses a flaw depth of 1⁄8 of
nominal thickness, the Level 2 results remain more favorable than the Level 3 results below 230
psig. This apparent contradiction occurs because weld residual stress greatly influences the
permissible MAT, and it completely drives the MAT at lower applied stresses (i.e., low
pressure). These results highlight the importance of completing a detailed Level 3 evaluation
when brittle fracture is a legitimate concern.
  Fig. 8. Level 2 and Level 3 MAT curve
  results for Case Study 3.

However, if the system was re-evaluated to the Level 3 FFS using a postulated flaw depth of 1⁄16
of nominal thickness, as opposed to the 1⁄4 or 1⁄8 of nominal thickness, then the calculated MAT
curve is above the process boiling point curve, thus indicating the acceptability of the 1⁄16
thickness flaw depth. Therefore, with a much more extensive and comprehensive inspection
plan, qualification for protection against brittle fracture may be justifiable so long as the
inspection techniques are sensitive and thorough enough to ensure that no defects greater than
1⁄16 of the nominal shell thickness exist.

CASE STUDY 4

This example illustrates the use of a Level 3 FFS brittle-fracture evaluation to optimize the cool-
down rate of a hydrodesulfurization (HDS) reactor.

Background

To expedite the cool-down process and limit downtime for an HDS reactor, owner-users often
inject cold nitrogen into the reactor upstream piping system. To determine the rate limits for the
cool-down process, a formal FFS evaluation was conducted.

Analysis

The study of the cooling process used a rigorous Part 9 Level 3 fracture mechanics assessment to
establish the allowable pressure-temperature curve. The analysis was performed for a reference
flaw size that could be readily detected in detailed inspections for cracking. A comprehensive,
3D FE model determined the stresses due to internal pressure and thermal transient effects during
the cooling process, as shown in Fig. 9. 
  Fig. 9. Thermal and stress plots at reactor
  quench nozzle.

Results

The results of the Level 3 FFS evaluation (Fig. 10) were used to optimize the cool-down
procedure and enable the owner-user to minimize downtime without adversely affecting the
mechanical integrity of the equipment. 

  Fig. 10. Level 3 MAT curve results

  for Case Study 4.

Overview

There have been a number of catastrophic brittle-fracture failures in the petrochemical industry.
Deficiencies in historical codes of construction and discrepancies in present codes and standards
have been identified. Pressure equipment must be properly assessed to qualify for low-
temperature service. In many instances, a Level 2 FFS evaluation may not be sufficient or
appropriate. If a legitimate concern exists for brittle fracture due to the potential for cracking, or
if metal temperatures below –55°F are achievable, a detailed Level 3 Part 9 FFS fracture
mechanics evaluation should be completed. The Part 9 Level 3 evaluation, coupled with PWHT
and a detailed inspection plan, can be used successfully to qualify low-temperature acceptability.
HP

ACKNOWLEDGMENT
a
Source for Fig. 1 is: Callister, W. D. and D. G. Rethwisch, Fundamentals of Materials Science
and Engineering: An Integrated Approach, 4th Ed., and Callister, W. D., Fundamentals of
Materials Science and Engineering, 5th Ed., pg. 257, Fig 9.3.

LITERATURE CITED
1
API Recommended Practice 571, Damage Mechanisms Affecting Fixed Equipment in the
Refining Industry, Second Ed., April 2011.
2
Occupational Safety and Health Administration, Part Number 1910, Subpart H, Standard
Number 1910.119, Process Safety Management of Highly Hazardous Chemicals.
3
ASME B&PV Code Section VIII, Division 1, Rules for construction of pressure vessels, ASME,
July 2013.
4
API 579-1/ASME FFS-1, Fitness-For-Service, June 5, 2007 (API 579 Second Ed.).
5
WRC Bulletin 528, “Development of Material Fracture Toughness Rules for the ASME B&PV
Code, Section VIII Division 2.”
6
ASME B&PV Code Section VIII, Division 2, Alternative Rules for Construction of Pressure
Vessels, ASME, July 2013.
7
National Board Inspection Code , Part 3 Repairs and Alterations, 2013.

The author

 
Brian Macejko is the head of the pressure vessel group within the mechanical
engineering business unit of The Equity Engineering Group, Inc. (E2G). He is also a
member of the ASME/API Joint Committee on Fitness-For-Service. He has
experience as both an owner-user and, as a consultant providing engineering support
  to oil and gas and petrochemical facilities. The primary focus of his experience has
been in the design, maintenance/repair, failure analysis, and fitness-for-service
activities for fixed equipment. 

IIT Delhi
04.01.2015

It should be perpetual excess