MPWT19-14468

The Detrimental Effect of CP, CO2 & H2S or Other Environmental Parameters
on Fracture Toughness and Fatigue

Andy Barron MSc, MIMMM, CEng

Global Fatigue and Fracture Mechanics Specialist
Element Materials Technology
10 Lower Grosvenor Road
London, SW1W 0EN
United Kingdom

Olivier Ravier MSc, MIMMM

Technology Development Specialist
Element Materials Technology
Hareness Circle, Altens Industrial Estate
Aberdeen, AB12 3LY
United Kingdom

ABSTRACT

The challenge that the oil and gas industry is currently facing is that pipelines are exposed to increasingly
complex environmental conditions which degrade the fracture toughness and fatigue crack growth
properties of the materials. There is a lack of standardized test techniques and consequently limited
specific materials related data and knowledge available which is affecting the reliably of engineering
critical assessments (ECA) and fitness for service (FFS) evaluations.

This paper discusses some factors influencing the test methods and provides some outline on the
degradation in the material properties expected and then considers their impact on ECA and FFS
analysis.

Keywords: Fracture toughness, fatigue crack growth rate, engineering critical assessment, fitness for
service, sour service, in-situ testing, cathodic protection, CO2, H2S.
 INTRODUCTION

The qualitative “workmanship” approach to weld flaw acceptance has been increasingly replaced by well-
established alternative weld flaw acceptance criteria based on quantitative fracture mechanics
procedures such as those outlined in BS 79101 and API 11042. The accuracy of these procedures is
reliant on key input data which adequately characterize physical properties of the pipe material and the
welds used during pipe manufacture and pipe line construction.

There are many qualitative tests in use to evaluate the susceptibility of materials to cracking in sour
environments but there is a growing need in the oil and gas industry to generate accurate quantitative
data for use in Engineering Critical Assessments (ECA) which are used to derive weld flaw acceptance
criteria during construction. In addition, this data is then used for ongoing fitness for service evaluations.
Such data has to be generated using project specific material and environmental conditions.

Although often seen as an additional burden, the time, effort and expenditure required to perform the
necessary testing is essential as it underpins any structural analysis and without it any analysis can be
at best overly conservative or at worst unsafe.

There are standardized test methods for both fracture toughness and fatigue testing, however it is
increasing apparent that the test methods to establish the behavior of metallic materials and welds used
in extreme environments need to be adapted and are yet to be standardized.

The purpose of this paper is to initially outline the detrimental effect of three major environmental
contributing factors on either the fracture toughness or the fatigue crack growth rate behavior of carbon
steel materials used by the oil and gas industry.Then a brief review of the significance of the experimental
data generated when used within an ECA and how it influences the results of the analysis will be
presented.

INTRODUCTION - FRACTURE TOUGHNESS and FATIGUE CRACK GROWTH RATE

It is important to establish a few high-level definitions and provide an overview of key physical parameters
discussed in the later part of the document.

Fracture Toughness - Monotonic

Most materials can be defined and characterized by their strength and toughness. Strength is a measure
of how well a material can resist being deformed from its original shape whereas toughness is the ability
of a material (with a notch or crack/defect) to absorb energy and plastically deform without fracturing.

Crack or flaw extension by a fracture toughness mechanism has two basic failure criteria:

Brittle cleavage fracture (unstable cleavage fracture) and ductile tearing (stable tearing resulting in
plastic collapse). Depending on the behavior of the material this fracture toughness can be expressed in
terms of:
It is important that a separation is made between the term or designation of the fracture toughness
parameter and the fracture toughness test geometry used.

When performing a test or a series of tests the fracture toughness can be expressed as a single point
value (K, crack tip opening displacement (CTOD / δ) or J as presented in Figure 1) or a resistance curve
(CTOD-R or JR as presented in Figure 2).

Figure 1: Single Point Toughness Values

JR Curve
2.5

10023 WCL (110℃)

10023 WCL (RT)

2.0
Power (Manual Fit Lower Bound)

Power (Manual Fit Lower Bound)

y = 1.47x 0.76
1.5
J (MPa.m)

y = 1.33x 0.78

1.0

0.5

0.0
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6
Tearing (mm)

Figure 2: Resistance Curve Toughness Values

Fracture Toughness & Fatigue – Sample Geometry
The most common specimen configurations used in the oil and gas industry are single-edge notched
bend (SENB), compact tension (CT) and single-edge notched tension (SENT). However, there are many
other well-defined configurations and the selection of the specimen geometry is normally dictated by the
configuration of the component under test.

The logical fracture toughness and fatigue sample geometry to extract from a pipeline butt weld is a
SENB geometry sample as recommended in API 11042 Annex A and illustrated in Figure 3.

Figure 3: SENB Fracture Toughness Geometry Sample Extraction from a Pipe Butt Weld

There is an option to modify the sample geometry and notch depth selection based on attempting to more
closely represent/replicate the structural constraint. The use of the SENT geometry sample in certain
circumstances is an example of this as the relaxation of constraint allows for a more representative and
less conservative fracture toughness value compared to the more highly constrained and conservative
SENB and CT geometry sample configurations.

It should be noted that, as shown in Figure 4, the relaxation of constraint due to sample geometry is only
applicable for fracture toughness samples working in the plastic constraint regime. The elastic constraint
for all specimen geometries is the same and therefore for fatigue the test results are not affected by
sample geometry.

Figure 4: Fracture Toughness Expressed as a Function of the Sample Geometry/Constraint

Fatigue Crack Growth - Cyclic
Cyclic fatigue crack growth is defined as the phenomenon of crack propagation under repeated cyclic
loading. Although there are different fatigue loading regimes that can be investigated, Fatigue crack
growth rate is the regime most used for fracture-based assessments.

Fatigue crack growth rate or FCGR characterizes the ability of a material to “resist” the propagation of a
fatigue crack within the elastic domain. The testing usually focuses on two main areas of interest:

Fatigue Threshold below which a fatigue crack will not propagate.

Paris’ Law (or Paris II) where a logarithmic relationship exists expressed in terms of crack growth rate
for a given stress intensity factor at the tip of the crack as shown in Figure 5.

Figure 5: FCGR Represented as a Function of Crack Growth (da/dN) Versus Stress Intensity ΔK

ENVIRONMENTAL TESTING - Methodology

The behavior of most metallic materials in air at room temperature (RT), elevated or sub ambient
temperatures have been extensively investigated and consequently generic data or pass/fail criteria can
be used for design.

Although some data is available for materials in air and seawater a significant challenge faced by the oil
and gas industry is the lack of specific material data and knowledge regarding materials subjected to
complex environment and in particular sour service conditions.

The following sections outline some specific information related to testing in an environment focusing on
two key environmental tests (in-situ Fracture Toughness and FCGR) which provide critical input data to
pipeline ECA and FFS analyses.
Testing in Environment
It should be noted that the main focus in the following sections is to provide information related to those
environmental conditions most associated with line pipe and girth welds used in the oil and gas industry.

The key environmental parameters associated with line pipe and girth welds used in the oil and gas
industry are pH, H2S and/or CO2 partial pressures, cathodic protection, temperature, salt concentration
and inhibitors. However, the combination and concentration of these parameters changes on a project to
project basis.

To replicate some of the environmental conditions detailed in the paragraph above testing can be
performed in various solutions for example 5%wt NaCl, ASTM D11413, NACE TM01774 or EFC 165.
As stated in DNVGL-RP-F1086 Appendix C, it is recommended to perform the fracture toughness tests
at room temperature (typically worst case) and the FCGR test should be tested at either RT or operating
temperature.

The specimen geometry frequently used when testing in environment is SENB. The samples are notched
and pre-cracked in “air”, to generate a sharp crack as the starting point of the test and can be coated (5
of the 6 sides to simulate one sided diffusion) with or without the notch face exposed.

Post pre-cracking and coating the sample is exposed to the test environment for a fixed number of days
in order to achieve full saturation. Four days is commonly used for this “pre-soaking” / “pre-conditioning”
however depending on the coating configuration, specimen size and environment separate testing may
be required to validate this period.

Test Methods – in-situ Fracture Toughness and FCGR

There are a number of international standards governing the testing of Fracture Toughness and FCGR
samples in “air”. However, these documents do not address the specifics related to testing in-situ in an
environment.

Fracture Toughness
The aim of the fracture toughness testing is to determine a critical toughness and, if appropriate, an
initiation value for crack extension by tearing in the specific environment. The test method uses a single
specimen approach with DCPD to measure crack extension.

The K-rates targeted are between 0.1 and 0.005 N.mm-3/2/s (this is ~500 to 10,000 times slower than a
comparable test performed in air). The fracture toughness test output is expressed as J or CTOD at
maximum load (Critical), 0.2mm or 0.05mm of crack extension, a J or CTOD-R curve may also be
produced.

Fatigue Crack Growth Rate (FCGR)

The speed of the application of the cyclic elastic load (test frequency) is a key parameter and will influence
the test result obtained. Ideally the FCGR testing would be performed at the frequency which would
represent the worst case for the specific test environment.

Due to the typical time constraints when performing project work using the lowest frequency is not
possible. Consequently, the FCGR is determined at a fixed frequency (0.2Hz for example) and a
frequency-based scaling factor generated from a set of Frequency scan tests (performed at constant ΔK
over a range of frequencies: 1Hz down to 0.01Hz or lower) is applied.
The actual FCGR test is carried out using either increasing or decreasing loading performed over a range
of ΔK. The output obtained from the FCGR testing is in the form of a one or two stages Paris Law curve.

ENVIRONMENTAL TESTING – Experimental results

The following sections contain a brief overview of the detrimental effect of Cathodic Protection (CP), CO2
and H2S or other environmental parameters on fracture toughness and fatigue crack growth rate.

Fracture Toughness – Effect of Cathodic Protection

Pipe material was tested in-situ in seawater with and without CP. A photograph of the testing setup is
presented in Figure 6.

Figure 6: In-Situ Fracture Toughness Testing in Seawater with CP

A reference sample was tested without CP to provide the baseline information for the material. Tests
were performed to investigate the influence of CP and test speed on the fracture toughness of both
carbon steel parent metal and girth weld material.

As shown in the graphs contained in Figure 7 the application of CP resulted in a reduction of the CTOD
by a factor of ~6. In addition, as illustrated by the two graphs in Figure 8 a reduction in the test speed
resulted in a further decrease in the CTOD by a factor of ~2.

Figure 7: Effect of CP on CTOD Results

 Figure 8: Effect of CP and Speed on CTOD Results

Fracture Toughness – Effect of H2S/CO2

Pipe material was tested in-situ in EFC165 solution with an H2S/CO2 gas mix at elevated pressure. A
photograph of the setup is presented in Figure 9.

Figure 9: in-situ FT testing at Elevated Pressure with H2S/CO2

Carbon steel parent metal and girth weld material was tested in air and in-situ at elevated pressure with
H2S/CO2 to generate J-R curves and an environmental knock down factor (KDF).

It was found that under these environmental conditions the parent metal experienced a KDF of 10 and
for the weld metal a KDF of 6. The experimental J-R curves are presented in Figure 10 below.
 Figure 10: J-R Curve “Air” Versus “Sour”

Fatigue Crack Growth Rate – Effect of H2S/CO2

Pipe girth weld material was tested in-situ in chloride solution with H2S/CO2 gas mix at ambient pressure.

Increasing ΔK tests were performed in air to generate a baseline for comparison against the BS 79101
curves, which are reassuringly conservative.

An in-situ frequency scan was performed at constant ΔK=16 MPa√m at 1Hz, 0.1Hz and 0.01Hz. The
worst crack growth rate for this specific combination of material and environment was found to be at
0.1Hz.

As a result, the subsequent in-situ increasing ΔK tests were performed at this worst-case frequency, and
under these environmental conditions the project specific acceleration factor was 27.

The experimental data is presented in Figure 11 below.

Figure 11: FCGR in Chloride with H2S/CO2

 RESULTING IMPACT ON THE ECA & FFS ANALYSIS

Compared to the equivalent testing in air, the results from the testing in environment show a significant
reduction in fracture toughness and increase in the fatigue crack growth rate. Although better than some
of the generic “knock down” factors used in lieu of testing, factors 10 on toughness and 27 on fatigue
crack growth are significant.

It should also be noted that the shape of the J-R curve has changed resulting in the portion of the J-R
curve above the initiation value being flatter. This potentially changes the mechanism of further crack
extension by tearing from an increase in driving force to potentially a more time dependent mechanism.

Depending on the other installation and operational demands this degradation of the material properties
maybe accommodated within the ECA or FFS evaluation. However, this may not be the case resulting in
either a reduction in the initial weld flaw acceptance criteria, alteration of the operational conditions of the
pipeline or ultimately a significant reduction in service life.

Decreasing the weld flaw acceptance criteria can lead to more reliance on weld repairs which, depending
on the process, can exhibit inferior material properties compared to those achieved from an automatic
welding process leading to additional cost and delays without any tangible increase in the overall integrity
of the pipeline.

From an ongoing fitness for service perspective, additional analysis can be performed utilizing actual
operational conditions and in-service inspections. However, without the initial test data generated at the
pipe procurement and pipeline manufacturing stage using project environmental conditions, such
analysis will have less ability to demonstrate ongoing integrity.

CONCLUSIONS

It is accepted in the oil and gas industry that sour service testing is the subject of ongoing research and
although there is some information in the public domain no standardized methods or data exists. This is
partly due to the complexity of the environmental conditions and the variability it has on the material
properties.

In order to accommodate the more severe environmental service conditions a “workmanship” approach
to weld flaw acceptance criteria is no longer adequate. A fitness for service approach needs to be applied.
However, it is also clear that the use of generic knockdown factors combined with a series of other
conservative input parameters or additional global safety factors can lead to an overly pessimistic and
unworkable weld flaw acceptance criterion.

A more pragmatic approach needs to be employed, testing and analysis using more realistic (project
specific) but conservative parameters with the aim to achieve the balance between safety, reliability and
cost i.e. fitness for service.

ACKNOWLEDGEMENTS

The authors acknowledge the colleagues in element materials technology who contributed to the work
described in this paper and especially Louise Barron, who helped with the content.
 REFERENCES

1. BS 7910 (latest revision): “Guide to methods for assessing the acceptability of flaws in metallic
structures”.
2. API 1104 Twenty First Edition, September 2013 “Welding of pipelines and related facilities”
3. ASTM D1141 (latest revision), “Standard Practice for the Preparation of Substitute Ocean Water”.
4. NACE TM0177 (latest revision), “Laboratory Testing of Metals for Resistance to Sulfide Stress
Cracking and Stress Corrosion Cracking in H2S Environments”.
5. EFC 16 (latest revision), “Guidelines on Materials Requirements for Carbon and Low Alloy Steels
for H 2 S-Containing Environments in Oil and Gas Production”.
6. DNVGL-RP-F108 (October 2017) Appendix C, “Sour service testing guidelines for the fatigue and
fracture limit state”.