-
5/27/2018 Ac Corroosion Phd Thesis
1/168
Politecnico di Milano
Dipartimento di Chimica, Materiali e Ingegneria Chimica, G.
Natta
INFLUENCE OF ALTERNATING CURRENT
ON METALS CORROSION
Sara Goidanich
Advisor: Prof.Luciano Lazzari
Tutor: Prof. Pietro Pedeferri
Coordinator: Prof. Giuseppe Zerbi
Ph.D. Thesis in Materials Engineering XVII Course
2002 2005
-
5/27/2018 Ac Corroosion Phd Thesis
2/168
-
5/27/2018 Ac Corroosion Phd Thesis
3/168
0 INTRODUCTION
..................................................................................................................1
1
AC-INDUCED
CORROSION................................................................................................3
1.1 DEFINITIONS
.......................................................................................................................3
1.2 DIRECT CURRENT (DC)
INTERFERENCE...............................................................................4
1.2.1 Metallic Structures Interfered by
DC........................................................................6
1.2.2 Maximum Allowable
Interference.............................................................................8
1.3 CASE HISTORIES OFAC-INDUCED CORROSION
....................................................................8
1.4 MORPHOLOGY CHARACTERISTICS OFAC CORROSION
.......................................................10
1.5 ALTERNATING CURRENT INTERFERENCE
............................................................................11
1.5.1 Conductive
interference.........................................................................................12
1.5.2 Magnetic induction (coupling)
................................................................................13
Evaluation of induced peak voltage
.................................................................................14
1.5.3 Steady-State or Load
Conditions...........................................................................15
1.5.4 Fault
Conditions.....................................................................................................16
Conductive
coupling.........................................................................................................16
Inductive
coupling.............................................................................................................16
Combined effect
...............................................................................................................16
Transferred voltages
........................................................................................................17
1.6 CHARACTERISATION OFAC-INDUCED CORROSION
.............................................................171.7
CORROSION RATE
.............................................................................................................18
1.8 EVALUATION OF CORROSION RISKS AND CATHODIC PROTECTION
MONITORING ...................20
1.8.1 DC Potential Measurements in the Presence of
AC..............................................21
1.8.2 AC voltage
.............................................................................................................24
1.8.3 AC current
densities...............................................................................................25
1.8.4 AC Density/DC Density Ratio
................................................................................25
1.8.5 Cathodic Protection Criteria Based on AC and DC Densities
Measured by
Coupons 26
1.8.6 Size of the Coating Defects
...................................................................................27
Probabilistic
approach......................................................................................................27
1.8.7 ER corrosivity
probes.............................................................................................30
ER based AC corrosivity probe
........................................................................................30
1.9 DAMAGE TO PIPELINE EXTERNAL
COATING.........................................................................31
1.10 SAFETY FOR
PERSONNEL..............................................................................................32
1.11 AC INFLUENCE ON CATHODIC
PROTECTION...................................................................32
1.12 HOWAC INFLUENCES CORROSION PROCESSES
............................................................33
1.12.1
Formation of a Non Protective Oxide
Layer...........................................................34
1.12.2 Faradic Rectification Effect
....................................................................................34
I
-
5/27/2018 Ac Corroosion Phd Thesis
4/168
Limitations of the faradic rectification
models...................................................................38
A model including double layer capacitance
....................................................................38
1.12.3 Effect on
Overpotentials.........................................................................................39
1.12.4 Effect of the Anodic and Cathodic
Half-Cycles......................................................40
Positive half-cycle rectified AV
signal...............................................................................41Negative
half-cycle rectified AV signal
.............................................................................42
Full-wave sinusoidal AV
...................................................................................................43
1.12.5 AC-Induced Fluid
Flow...........................................................................................43
1.13 EFFECT OF THEAC FREQUENCY
...................................................................................44
Effect of the
waveform......................................................................................................45
1.14 AC EFFECTS ON THE CORROSION OF THE DIFFERENT METALS
.......................................46
1.14.1 DC Potential Variation in the Presence of AC
.......................................................46
1.15 PASSIVE
METALS..........................................................................................................47
1.15.1 Influence of Stray Currents on Corrosion of Steel in
Concrete..............................47
1.16 EFFECT OFAC ON CORROSION IN SOILS
.......................................................................48
1.17 INTERFERENCE
MITIGATION...........................................................................................49
1.17.1 DC
Mitigation..........................................................................................................49
Stationary interference
.....................................................................................................49
Non stationary interference
..............................................................................................50
1.17.2 Mitigation of AC Interference
.................................................................................51
Lumped
grounding............................................................................................................51
Cancellation wires
............................................................................................................51Gradient
control
wires.......................................................................................................52
2 EXPERIMENTAL
................................................................................................................53
2.1 SETTING UP OF THE ELECTRICAL CIRCUIT
..........................................................................54
2.1.1 AC mesh
................................................................................................................55
Characteristics of the capacitor
........................................................................................55
2.1.2 DC mesh
................................................................................................................56
Characteristics of the inductor (L)
....................................................................................56
2.1.3 Common Branch
....................................................................................................562.1.4
Efficiency of the Circuit
..........................................................................................56
2.2 TESTS ON GALVANIC CORROSION
......................................................................................57
2.2.1 Experimental
Conditions........................................................................................57
Solution
Composition........................................................................................................57
Cell, samples and Electrodes
Characteristics..................................................................57
2.2.2 Experimental Procedure
........................................................................................58
2.3 TESTS WITH SEPARATED CIRCUITS: ACAND DC OVERLAPPED ON THE
SAME SAMPLE ........59
2.3.1 Experimental
Conditions........................................................................................59
Solution
Composition........................................................................................................59
Cell, Samples and Electrodes Characteristics
.................................................................59
II
-
5/27/2018 Ac Corroosion Phd Thesis
5/168
2.3.2 Experimental Procedure
........................................................................................60
2.4 DETERMINATION OF THEANODIC AND CATHODIC POLARISATION
CURVES............................61
2.4.1 Measurement Procedure
.......................................................................................61
2.4.2 Experimental
conditions.........................................................................................62
Solution Composition
.......................................................................................................64Cell
and Electrodes
..........................................................................................................65
2.4.3 Experimental Procedure
........................................................................................65
2.5 WEIGHT LOSS TESTS
........................................................................................................67
2.5.1 Weight Loss
Determination....................................................................................68
2.5.2 Test in Simulated Seawater Chlorinity: 35 g/L
NaCl..............................................69
2.5.3 Test in Simulated Soil Solution
..............................................................................69
Specimens........................................................................................................................69
2.6 CHARACTERISATION OF THE CORROSION MORPHOLOGY OF THEAC
CORROSION ................71
2.7 EFFECT OFAC ON PASSIVE METALS
.................................................................................72
2.7.1 Samples
geometry.................................................................................................72
2.7.2 Solution
composition..............................................................................................72
2.7.3 Cell and
Electrodes................................................................................................73
Potentiodynamic tests
......................................................................................................73
Immersion
Tests...............................................................................................................73
2.7.4 Potentiodynamic Tests
Procedure.........................................................................73
Polarisation
Curve............................................................................................................73
Experimental
Procedure...................................................................................................742.7.5
Immersion Tests
Procedure...................................................................................75
Preliminary
tests...............................................................................................................75
Second set of
tests...........................................................................................................75
2.8 CATHODIC PROTECTION
MONITORING................................................................................75
2.8.1 First set of measurements
.....................................................................................76
2.8.2 Second set of measurements
................................................................................77
2.9 OSCILLOSCOPE
MEASUREMENTS.......................................................................................79
3 RESULTS
...........................................................................................................................81
3.1 TESTS ON GALVANIC CORROSION
......................................................................................81
3.2 TESTS WITH SEPARATED CIRCUITS: ACAND DC OVERLAPPED ON THE
SAME SAMPLE ........84
3.2.1 Cathodic Polarisation (CP)
Tests...........................................................................84
3.2.2 Anodic Polarisation (AP) Tests
..............................................................................86
3.2.3 Free Corrosion Tests
.............................................................................................87
3.3 ANODIC AND CATHODIC POLARISATION
CURVES.................................................................89
3.3.1 Tafel
Equation........................................................................................................89
3.3.2 Carbon Steel
..........................................................................................................90
0.4 g/L Sodium Sulphate
Solution....................................................................................90
Simulated soil
solution......................................................................................................92
III
-
5/27/2018 Ac Corroosion Phd Thesis
6/168
Simulated seawater solution: 35 g/L NaCl
.......................................................................95
3.3.3 Galvanised
Steel....................................................................................................96
3.3.4 Carbon Steel, Zinc and Copper in 1 M Solution of their
Ions ................................98
3.4 WEIGHT LOSS TESTS
......................................................................................................104
3.4.1 Simulated Seawater Solution: 35 g/L
NaCl..........................................................1043.4.2
Simulated Soil Solution
........................................................................................105
Aerated solution without
thermostating..........................................................................105
Aerated solution - thermostated at 20C
........................................................................110
De-aerated solution - thermostated at
20C...................................................................111
3.5 CHARACTERISATION OF THE CORROSION MORPHOLOGY IN THE PRESENCE
OFAC
INTERFERENCE
............................................................................................................................113
Neither AC nor DC interference
.....................................................................................113
AC = 100 and 300
A/m2..................................................................................................115
0.3 A/m2DC anodic
interference....................................................................................119
Potential
measurements.................................................................................................120
3.6 EFFECT OFAC ON CARBON STEEL REBAR IN CONCRETE
.................................................121
Second set of
measurements.........................................................................................124
3.7 EFFECT ON CATHODIC PROTECTION MONITORING
............................................................127
3.7.1 First set of measurements
...................................................................................127
3.7.2 Second set of measurements
..............................................................................129
3.8 MEASUREMENTS BY OSCILLOSCOPE
................................................................................131
4
DISCUSSION....................................................................................................................133
4.1 AC DENSITY THRESHOLD
................................................................................................133
4.2 AC INFLUENCE ON CORROSION
PROCESS........................................................................134
4.2.1 Corrosion Rate and AC Current Efficiency
(CE%)...............................................134
4.2.2 Influence of AC on Corrosion
Kinetics.................................................................136
4.2.3 Corrosion
Potential...............................................................................................137
4.2.4 Analysis of the Faradaic Rectification Models
.....................................................138
Oscilloscope measurements
..........................................................................................140
Results of the
models.....................................................................................................1404.2.5
Proposed
Mechanism..........................................................................................145
4.3 CORROSION MORPHOLOGY IN THE PRESENCE
OFAC.......................................................146
4.4 CARBON STEEL INALKALINE
SOLUTION............................................................................146
4.5 CATHODIC PROTECTION
MONITORING..............................................................................148
4.5.1 Field Monitoring Approach
...................................................................................148
5
CONCLUSIONS................................................................................................................151
AC density threshold
......................................................................................................151
Corrosion rate and AC current efficiency
(CE%)............................................................151Influence
of AC on Corrosion
Kinetics............................................................................152
IV
-
5/27/2018 Ac Corroosion Phd Thesis
7/168
Proposed
mechanism.....................................................................................................152
Carbon steel in alkaline
solution.....................................................................................153
Cathodic protection
monitoring.......................................................................................153
6
ACKNOWLEDGMENTS...................................................................................................155
7 REFERENCES
.................................................................................................................157
V
-
5/27/2018 Ac Corroosion Phd Thesis
8/168
VI
-
5/27/2018 Ac Corroosion Phd Thesis
9/168
0 INTRODUCTION
Alternating current enhanced corrosion has been considered since
the beginning of the XX
century[ ]1 . However, for many years, corrosion experts did not
recognise corrosion due to
alternating currents (AC) on metallic structures as an important
phenomenon as corrosion has
been normally attributed to direct current.
During the last 20 years pipelines failures in the USA, Canada
and Europe have been ascribed toAC interference. The presence of AC
caused serious damages on interfered metallic structures
even when cathodic protection was applied. In some cases,
failures also occurred on pipelines
with OFFpotential lower than 850 mV CSE, that is normally
considered a safe condition. A
number of studies carried out to investigate the effect of AC on
corrosion have concluded that
metallic structures corrode at an accelerated corrosion rate in
the presence of stray AC [2-9]. The
scientific community has slowly recognised that, under certain
conditions, alternating currents can
cause corrosion although at a lower rate than direct current.
The number of scientific publications
on this topic has therefore increased during the last 15 years.[
, , , , , , , ]2 3 4 5 6 7 8 9
What is clear is that the increased number of AC related
failures is associated to the more and
more frequent parallelism between buried pipelines and AC high
tension transmission lines which
use the same right of way. A particularly harmful situation is
when parallelism is associated with
the use of high dielectric coatings, like extruded polyethylene
or polypropylene, which leads to high
induced AC current densities at coating defects.
In the near future, at least in Italy, a new interference
condition is expected because of the
overlapping of high current AC traction systems (25 kV and 50
Hz) and the traditional direct current
(DC) traction systems that have been operating for about a
century.
However, while for DC interference corrosion on buried
structures is a well known problem and
there is a large agreement on protection criteria to be used for
corrosion mitigation and
international standards are available since many years[ ]10 ,AC
induced corrosion continues to be a
controversial subject and many aspects need to be clarified,
especially the relationship between
AC density and corrosion rate.
Different theories have been proposed about the mechanism by
which AC produces and
enhances corrosion, although no one is able to fully explain the
phenomenon. Furthermore it is not
yet clear how to estimate the risk of AC-induced corrosion by
means of AC density or alternating
-
5/27/2018 Ac Corroosion Phd Thesis
10/168
Introduction
2
voltage. Another important aspect that needs further
investigations is related to cathodic protection
monitoring in the presence of AC.
The experimental work dealt with the following:
design of an electric circuit to allow independent control and
measurement of AC and DC
contributions in order to guarantee reliable measurements;
study of the effect of AC current on galvanic corrosion;
study of AC and DC overlapping through the designed circuit;
weight loss test in order to quantify the damage caused by AC
and estimation of AC
corrosion threshold parameters;
determination of AC effects on passive metals;
cathodic protection monitoring in the presence of AC and DC
stray currents;
determination of anodic and cathodic polarisation curves in the
presence of AC. The aim
of these experiments was the evaluation of AC on the corrosion
mechanism as far as
corrosion kinetics is concerned;
analysis of the mathematical models proposed in literature and
comparison with
experimental data.
The analysis of the experimental data allowed to draw
conclusions about the following topics:
AC density threshold;
mechanism of AC-induced corrosion;
corrosion morphology in the presence of AC
AC induced corrosion of carbon steel in alkaline solution
(concrete);
Cathodic protection monitoring.
The first two years of this experimental research were co-fund
by MIUR (Italian Ministry of
School, University and Research).
-
5/27/2018 Ac Corroosion Phd Thesis
11/168
1 AC-INDUCED CORROSION
Before discussing the AC induced corrosion, it is important to
emphasise that very often
corrosion attacks are carelessly attributed to non-existent
interference corrosion conditions, simply
because the corrosion conditions are ignored. This is typical
for internal corrosion attacks in water
heaters or leakage of piping embedded in apartment floors, most
often attributed to ghost stray
current, through nobody knows its origins.
Interference corrosion greatly worries owners of buried
structures, because of the severe
damages it causes. The interference can be the result of DC or
AC, flowing predominantly with a
frequency of 50 Hz (public electricity supply).
A pipeline which shares a common path with AC transmission lines
becomes energized by the
magnetic and electric fields surrounding the power system in the
air and soil. This AC interference
can result in an electrical shock hazard for people touching the
pipeline or metallic structures
connected to the pipeline; furthermore, corrosion of the
pipeline, damages to the pipelines coating,
insulating flanges and rectifiers can occur. The induced
AC-voltages can lead to an increase of thecorrosion rate, the
breakdown of a passive layer or increasing pitting corrosion.
In this chapter a brief summary of the DC interference
characteristics and the state of art of AC
induced corrosion phenomena are reported.
1.1 DEFINITIONSAs a general definition, interference is any
alteration in the electric field caused by a foreign
structure[ ]11 . Figure 1.1 illustrates a general scheme of
interference in an electrodic system where
two electrodes exchange current in the presence of a foreign
body: when it is a conductor, the
current is intercepted, when it is an insulator, the current is
withdrawn. In both cases, there is a
redistribution of current and equi-potential surfaces within the
altered electric field.
The first instance, when the interfered structure has ionic or
electronic conductivity, leads to the
classic and most well known interference conditions, the
so-called stray current. When the
structure is far from the electrodes, the influence on the
electric field is usually negligible[11].
-
5/27/2018 Ac Corroosion Phd Thesis
12/168
AC-Induced Corrosion
4
IsolatorConductor
Equipotential lines
Current lines
M N M N
a) b)
Figure 1.1 - General scheme of electrical interference between
two electrodes on a bodya) conductor; b) isolator[11]
1.2 DIRECT CURRENT DC) INTERFERENCEInterference can be of two
types: stationaryand non-stationary.
Stationary interference is verified when the structure is
immersed in a stationary electric field
generated, for example, by a cathodic protection (CP) system,
and the effect is greater as the
structure is closer to the groundbed. In Figures 1.2 and 1.3,two
general cases are reported. In the
first case, the interfered pipeline crosses the protected one
and the zone close to the groundbed
tends to gather current from the soil, which is released at the
crossing zone, causing localcorrosion. In the second case, the two
pipelines are almost parallel and the releasing current zone
is usually more extensive than in the previous case, and
typically occurs in low resistivity soil. In
both cases, if the interfered structure is provided with an
integral coating, interference cannot take
place, but when the coating has a number of faults, corrosion is
very severe since, current
concentrates in them.
Non-stationary interferencetakes place when the electric field
is variable, as in the typical case
of stray currents dispersed by traction systems, as illustrated
in Figure 1.4. Interference takes
place only during the trains transit, and often, in spite of the
limited duration, a few minutes, theeffects may be severe due to
high circulating current.
The mechanism of direct stray current corrosion is simple: the
DC traction system has a positive
aerial conductor and, for safety reasons, a grounded negative
pole, so the return current path is
double: track and soil, with over 50% in soil even when all
precautions to reduce current dispersion
in the ground have been taken[11]. A pipeline, parallel to rail,
buried between the track and the
substation is easily interfered, with corrosion taking place
where current leaves the structure near
the substation.
-
5/27/2018 Ac Corroosion Phd Thesis
13/168
AC-Induced Corrosion
5
Groundbed
Interfered Pipeline
T/R
Corrosion attacks
Interference current
Figure 1.2 - Scheme of stationary interference between two
crossing pipelines[11]
T/R
Low resistivity
Groundbed
Interfered pipeline Corrosion attacks
Interference current
soil
Figure 1.3 - Scheme of stationary interference between two
almost parallel pipelines[11]
E
protection potential
time
Substation
cathodic zone anodic zone
Figure 1.4 - Scheme of non-stationary interference caused by
stray current dispersed by a DC transit
system[11]
Rail
Pipeline
Substation
R
R
R Ri iI
i
1 2rail
pipe
Figure 1.5 - Electrical scheme of non-stationary
interference
-
5/27/2018 Ac Corroosion Phd Thesis
14/168
AC-Induced Corrosion
6
From an electrical point of view, the interference currentmay be
evaluated from the balance of
the ohmic drop through the circuit, when overpotential
contributions on both pipeline and rails are
neglected.
With reference to Figure 1.5,where Iis the current that passes
through the rail (estimated to be
about 50% of the total current tension) and I*is the
interference current that passes through the
pipeline, then:
2pipe1
r
RRR
IR*I
++= Eq. 1.1
where, Rr, Rpipe indicate the track and pipe resistance
respectively, R1 and R2 the ground
resistance. Corrosion damages, quantified through I*, decrease
as the track resistance Rr
decreases (for instance by using welding joints to reduce the
ohmic drop to about 1 mV/m) and by
increasing the parallel path resistance by applying protective
coatings, especially in the cathodic
zone (increase in R1+ R2) and by installing insulating joints on
the pipelines (increase in Rpipe).
1.2 1 Metallic Structures Interfered by DCDC interference occurs
when a metallic structure is placed in a DC electric field. The
reaction
that takes place over the surfaces receiving the current
(cathodic zones) is primarily oxygen
reduction and then hydrogen evolution at more negative
potentials, by the following reactions:
O2+ 2 H2O + 4e-= 4 OH- Eq. 1.2
2H2O + 4e-= H2+ 2OH- Eq. 1.3
On anodic zones, where DC current leaves the metallic
structures, the anodic reaction is metal
(Me) dissolution when the metal is active:
Me = Mez++ ze- Eq. 1.4
When the metal is passive, the anodic reaction depends on the
type of metal and environment. In
alkaline media, such as pristine concrete where both stainless
steel and carbon steel rebars are in
passive condition, the anodic reaction is oxygen evolution, by
the following reaction:
2 H2O = O2+ 4 H++ 4e- Eq. 1.5
Because acidity is produced, passivity may be destroyed at high
currents and long time of
interference, still provoking iron dissolution. For stainless
steel in a neutral environment, the
corrosion reaction is mainly metal dissolution, so corrosion
effects are as harmful as in the case of
carbon steel.
According to Faradays law, the amount of metal that dissolves by
reaction 1.4 is directly
proportional to current and time. A flow of 1 A dissolves about
9 kg of iron in a year, equal to about
25 g/day. A current density of 1 A/m2provokes a metal thickness
loss at a rate of 1 mm/year.
-
5/27/2018 Ac Corroosion Phd Thesis
15/168
AC-Induced Corrosion
7
Figure 1.6 - Interference on metallic structures[11]
With reference to Figure 1.6, ignoring the electrode
overpotential at M and N surfaces, path 1
includes only the electrolyte ohmic drop, while path 2 includes
the overpotential of the cathodic
oxygen reduction and the anodic iron dissolution on the
interfered structure. Therefore, introducingthe IRdrop within the
structure, we obtain[11]:
++
++
=
S
LI
S
LI
S
LI
S
LI 2a2strc21 Eq. 1.6
or when the structures ohmic drop is ignored:
ES
LI
S
LI
S
LI ac221 +
Eq. 1.7
where indicates the electrode overpotential with respect to the
free corrosion potential(= E Ecorr) localised at anodic (a) and
cathodic (c) surfaces of the interfered structure. Terms
and str refer to environment and metal resistivity,
respectively. Assuming negligible the IRdrop
in the interfered metallic structure the following result is
achieved: the overpotential sum E, which
represents the driving voltage for current circulation in the
interfered structure, is equal to the
ohmic drop saving within the environment through the
interference path. The term in brackets
can be approximated to:
strerfint LICE Eq. 1.8
where C is a geometric constant, Iinterf is interference
current, Lstr is structure length. It follows
that, for a fixed geometry, interference effects, determined by
the available E, are increasing as
structure length increases.
The available driving voltage, E, is also given by:
strLEEdLE = Eq. 1.9
where E is the average electric field gradient. From Eq. 1.8 and
1.9 it follows:
-
5/27/2018 Ac Corroosion Phd Thesis
16/168
AC-Induced Corrosion
8
E
'CI erfint Eq. 1.10
where C is a geometric constant. Then, Iinterfincreases as
electric field gradient E increases
and environment resistivity decreases.
The interference current Iwhich gives the corrosion rate, is
generally not easy to calculate since
the electric field solution is required.
1.2.2 Maximum Allowable InterferenceInterference is as
negligible as the interference current density is low. From a
general viewpoint,
it may be reasonable to assume the value at which carbon steel
passivates in concrete, that is
1 mA/m2, or even double that, as an engineering compromise [11].
Unfortunately, because in
practice the circulating current is not measurable, interference
acceptance criteria refer to thepotential measurement, to
distinguish between freely corroding structures and those under CP.
For
the latter, potential variation should remain within the
protection interval, while for the former, the
maximum potential variation in the positive direction should not
exceed 20-50 mV with respect to
the free corrosion potential.
1.3 CASE HISTORIES OF AC-INDUCED CORROSIONHigh alternating
current density can cause corrosion attacks on steel in soil,
despite the steel
surface being provided with proper cathodic protection[ , ]1213
.
Actual cases of AC corrosion on buried steel pipelines have been
reported since 1986[ ]14 . The
cause of two corrosion anomalies on polyethylene coated
pipelines in Germany, that were installed
in 1980 paralleling a 15 kV AC powered rail transit system
operating at the frequency of 16-2/3 Hz,
was attributed to AC corrosion, although the instant-OFF
potentials were more negative than
1.0 V CSE at those sites.
Several cases of AC corrosion were reported in France where
ONpotentials were at an
adequately low level between 2.0 V CSE and 2.5 V CSE[ ]15 Five
cases of AC corrosion in North
America were reported even if ONpotentials were sufficiently
negative in all cases[ ]16 .
A field survey using coupons on a pipeline buried parallel to a
66 kV overhead AC power
transmission line showed that a high level of AC current can
cause corrosion on coupons even
though the instantOFF potentials are more negative than 1.0 V
CSE[ ]17 .
In Oregon (USA) the failure of a pipeline that had been laid in
high voltage AC utility corridors
was investigated. Severe AC corrosion occurred resulting in four
holes in the pipeline after only
five months life and before operation had begun. The pipeline
was left buried in place withoutcathodic protection while the
construction of the plant was completed. There appeared to be
some
-
5/27/2018 Ac Corroosion Phd Thesis
17/168
AC-Induced Corrosion
9
coating disbondment. The pits typically had a small aerial
extent and were very deep. It was
verified that there was no DC interference at the leak
locations[ ]18 .
In Figure 1.7 the measured AC voltages taken during a survey of
a pipeline which had reported
AC-induced corrosion are shown[18].
0
20
40
60
80
100
0 2 4 6 8 10 12 1
Distance - Miles
ACVoltage
4
Figure 1.7 - Induced voltage on a AC-interfered pipeline
[18]
Also in Texas a careful investigation of a pipeline failure was
carried out. The review of the
annual survey data, bimonthly rectifier readings and an ON/OFF
close interval survey data showed
apparent satisfaction of industry criteria for effective
corrosion mitigation. Nothing in the review of
the cathodic protection history suggested a plausible
explanation for the rapid rate of corrosion
experienced on this pipeline[ ]19 . The pipe-to-soil AC
potentials and the number of corrosion
anomalies as a function of the location were evaluated. The plot
of the AC potentials and defects
are shown in Figure 1.8. This data indicated that the highest
population of corrosion anomalies
were located in the first five miles where the AC potentials
typically exceeded 4 V. The excessive
AC current densities observed on the coupon test stations and
the physical and chemical analysis
indicated that the likely cause of the observed corrosion
anomalies was AC corrosion. This
conclusion was supported by the correlation of higher defect
occurrences within areas of higher
AC potentials.
It can be observed by the analysis of the reported case
histories that AC corrosion is possible
despite satisfying the conventional cathodic protection criteria
such as the 0.85 V CSE potential
criterion.
-
5/27/2018 Ac Corroosion Phd Thesis
18/168
AC-Induced Corrosion
10
0
2
4
6
8
10
12
14
290 295 300 305 310
Mile Post
ACPotentia
l(VCSE)
0.000
0.003
0.006
0.009
0.012
0.015
0.018
0.021
Defectspe
rFoot
AC Potential
Anomalies Per Foot
Figure 1.8 - AC potential and defects distribution [19]
1.4 MORPHOLOGY CHARACTERISTICS OF AC CORROSIONIn DC stray
current interference, the corrosion products are soluble due to the
low pH at the
discharge location and the corroded area is generally found to
be free of corrosion products. In
case of AC interference the corrosion products are generally
present in the pits.
Measurements performed on cathodically protected coupons and on
corroded pipelines
suggested a mechanism of AC induced corrosion which involves
local alkalization of the soil
environment close to a coating defect[7,19,20,[ , ]2122 . This
has probably to be related to the enhanced
cathodic reaction during the AC interference.
The enhanced corrosion due to the influence of AC voltages tends
to be of a localised nature.
The corrosion morphology is reported to be of a shallow
saucer-shaped pit of a several millimetre
diameter that is overlaid by a layer of hard encrusted earth[
]23 .
AC corrosion can take on the appearance of generalised corrosion
when the defect surface is
large. These large areas are in fact several pits that have
coalesced over a period of time. InFigure 1.9 the corrosion of a
coupon which was placed in an Italian situ and was subjected to
AC
interference is shown.
-
5/27/2018 Ac Corroosion Phd Thesis
19/168
AC-Induced Corrosion
11
Figure 1.9 Corroded coupon placed in an area in Italy
characterised by high AC interference
The appearance of the corrosion may therefore vary depending on
the duration of the
interference from small point-shaped attacks distributed across
the steel surface to large and deeplocal attacks with smooth round
corrosion morphology.
Generally the coating in the area surrounding the pits is
damaged, often brittle and has poor
adhesion[19].Coating disbondment is often associated to AC
interference.
A typical aspect of failures occurred on cathodically protected
pipelines is summarised in
Figure 1.10.
Corrosion product placed on top of an hard oxide layerand
corrosion diameter is larger than the coating
defect diameter ( = a few cm )Coating
Possibility of coatingdisbondment
Metal
Corrosion product placed on top of an hard oxide layerand
corrosion diameter is larger than the coating
defect diameter ( = a few cm )Coating
Possibility of coatingdisbondment
Metal
Figure 1.10 - Specific characteristics of AC corrosion [12]
1.5 ALTERNATING CURRENT INTERFERENCEAs the DC interference, also
the AC interference can be either stationary or non-stationary.
Alternating current interference is generated by AC traction
systems or high voltage transmission
lines, through conduction and induction mechanisms. The effects
of power system interference on
pipelines are due to the relative voltage difference which is
created between the pipeline metal and
the local soil.
Where a pipeline runs closely parallel to and/or crosses an AC
power line, steady state and fault
AC voltages can be induced on the pipeline. Induced AC voltages,
which are respect to groundpotential, arise as a result of
electromagnetic capacitive and inductive effects. The magnitude
of
-
5/27/2018 Ac Corroosion Phd Thesis
20/168
AC-Induced Corrosion
12
the induced voltage depends on the AC transmission line
voltage/current, the pylon dimensions,
the separation distance of the pipeline from the AC transmission
line, the distance over which the
pipeline are in parallel with and crosses the AC transmission
line, soil resistivity, coating
conductivity of the pipeline and whether the AC transmission
line is experiencing a fault or large
transient surge associated with a switch on[23]
.
The AC power lines induce a secondary alternating current in the
steel pipes buried nearby: field
measurements indicated that this secondary AC field has a
voltage drop on the order of
kilovolts/km along insulated pipes[ ]24 . The potential drop
between the bare pipe and the
surrounding soil was reported in the range of 5 to 70 V.
An example of voltage profile induced along a pipeline by a
380kV50 Hz power line carrying
balanced currents of 630 A is shown in Figure 1.11.
0
10
20
30
40
50
60
0 5 10 15 20 25 30
Distance (km)
ACVoltage(V)
Figure 1.11 - Example of voltage profile induced along a
pipeline by a nearby 380kV-50Hz power l ine[ ]25
1.5 1 Conductive interferenceConductive interference due to
currents flowing in the soil is of particular concern at
locations
where the pipeline is close to transmission line structures
which may inject large currents into the
soil during power line fault conditions. Such structures include
transmission line tower or pile
foundations and substation grounding systems[ ]26 .
Interference by conduction takes place when a buried structure
spreads AC current. This is the
case of grounding networks and high voltage transmission lines
with a phase grounded. Stray
currents choose the path of least resistance, therefore they
affect buried metallic structures such
as buried steel pipelines and tanks[11].
There are two typical cases of conductive interference:
dispersion from a grounding network
(plant earthing systems) and from AC traction systems (rails and
their grounding conductor). In
-
5/27/2018 Ac Corroosion Phd Thesis
21/168
AC-Induced Corrosion
13
these cases, AC voltage on interfered structures in soil can be
calculated when geometry, soil
resistivity and AC voltage of dispersing sources are known.
1.5.2 Magnetic induction (coupling)Magnetic induction (coupling)
acts along the entire length of pipeline which is approximately
parallel to the power line and can result in significant
pipeline potentials even at relatively large
separation distances. Inductive coupling takes place when an
high voltage transmission line with
its neutral grounded is parallel to a buried well coated
pipeline. In such a way, an alternating
magnetic flux is linked between high voltage transmission line
and soil, which induces an
alternating current in the coated pipeline. In fact, the
pipeline and the soil behave as the second
coil of a transformer, where the first coil is the high voltage
transmission line and the soil. In other
words, this is an auto-transformer, with a 1:1 transformation
ratio. [11].
Figure 1.12 is a schematic representation of the inductive
coupling on a steel pipelines located
near a high voltage AC transmission systems. The voltages are
induced in the pipeline due to
imbalances in the distance of the various phase wires and in the
current carried in each phase.
The resulting voltages in the pipeline are a function of the
distance of the pipeline from the phase
wires, the voltage in the system, the current carried by the
system and the phase imbalance in the
current.
Da = Db = Dc
Da Db Dc
Da
= Db = Dc
Da Db Dc
Figure 1.12 - Induced AC interference on a pipeline [18]
Inductive coupling causes voltage peaks on the pipeline at
locations where the pipeline and
power line separation changes, such as when the pipeline
approaches or recedes from the power
line right-of way. The crossing of the pipeline from one side of
the power line to the other will also
cause an induced voltage peak. For example if a pipeline and
transmission line are parallel to each
other and one of the two changes direction, the voltages will
build-up at the point of divergence.Pipelines that pass under a
transmission line at an angle of 90 do not have any AC current
or
-
5/27/2018 Ac Corroosion Phd Thesis
22/168
AC-Induced Corrosion
14
voltage induced. Although these potential and current changes
are physical in nature, they are
generally considered as electrical discontinuitiesbecause they
cause a change in the longitudinal
electric field impressed upon the pipeline[18, ]27 .
The AC-induced voltages can even reach values that can cause
serious danger to people
maintaining pipelines that are subjected to AC-induced corrosion
[ ]28 .
The AC circulating within the pipeline is exchanged with soil at
coating defects.
Evaluation of induced peak voltage
The following formula has been proposed to evaluate the induced
pick voltage in
correspondence of discontinuities[27]
I2
fpEV
= Eq. 1.11
where, V = induced voltage at discontinuity (V);
E = coupling coefficient (V/km/A);
p = voltage coupling coefficient (km);
f = multiplying factor;
I = power line load current (A).
Equation 1.11 and the coupling parameters (Tab. 1.1)are based
upon a nominal 10-meter power
line to pipeline separation or less. At greater distances, the
induced voltage will be reduced by
approximately the inverse ratio of the pipeline distance
relative to the 10-meter reference[27].
-
5/27/2018 Ac Corroosion Phd Thesis
23/168
AC-Induced Corrosion
15
Tab. 1.1 Values of the coefficients in Equation 1.11[27]
Steady state powerline coupling
coefficientPipeline induced voltage coefficient
Geometrical/electricalfactors
Power lineconfiguration
CouplingcoefficientE (V/km/A)
Coatingdescription
Resistance
(m2)
Voltagecoefficient
p (km)
Type ofdiscontinuity
Multiplyingfactor f
Single circuithorizontal
0.05 Excellent(thick)
100000 10.6 Pipelineapproach/recede
1
Single circuitvertical
0.05Excellent
(thin
-
5/27/2018 Ac Corroosion Phd Thesis
24/168
AC-Induced Corrosion
16
effect of a fault on the pipe. Induced steady-state pipeline
potentials are more severe when the
pipeline coating has a high electrical resistance. However, a
high coating resistance is desirable
from a cathodic protection standpoint, so reducing the coating
resistance is not usually considered
a viable solution.
In spite of the relatively low magnetic field levels during
steady-state conditions, induced voltages
on an unprotected pipeline can reach hundreds of volts[26]. Even
with extensive grounding systems
connected to the pipeline, pipeline potentials can be on the
order of tens of volts. This constitutes
primarily a shock hazard, which can be transferred miles away
from the parallel corridor.
1.5.4 Fault ConditionsIn cases where the overhead line has an
earthed neutral, unbalanced or fault conditions can
result in substantial currents flowing in the soil. For an
unbalanced single-phase supply, e.g. an
electrified railway, or earth faults on the AC transmission
line, hazardous voltages may be induced
even for short parallel lengths[23]. Also a fault at a pylon or
a lightning discharge or where a
severed AC phase wire comes into contact with the ground may
induce high voltages.
Conductive coupling
When a single-phase-to-ground fault occurs at a power line
structure, the large fault current
injected into the soil by the structure raises the local soil
potential. A pipeline located nearby,
however, will typically remain at a relatively low potential,
due to the resistance of its coating and
grounding at points distant from the fault location. The
pipeline potential rise will be particularly
small for a pipeline with a high resistance coating. Therefore,
the earth around the pipeline will be
at a relatively high potential with respect to the pipeline
steel potential[26].
During a fault, the worst affected part of a pipeline will
usually be that part which is the closest to
the faulted power line[23]. However, for a well-coated pipeline,
considerable voltages can be
transferred many kilometres beyond the end of a shared
corridor.
Inductive coupling
During single-phase-to-ground fault conditions on the power
line, induced potentials in a pipeline
with no mitigation system can reach thousands of volts, due to
the intense magnetic field causedby the large current which may
flow in the faulted wire[26].
Combined effect
Unless the pipeline is perpendicular to the power line, the
pipeline will be simultaneously subject
to both inductive and conductive interference. In most cases the
inductive and conductive
components are addictive and this may give rise to a very large
voltage difference. In fact the
change in pipeline steel potential due to induction is
essentially opposite in sign to the soil potential
change due to conduction, therefore inductive and conductive
effects add, producing even more
severe coating stress voltages and touch voltages[23,26].
-
5/27/2018 Ac Corroosion Phd Thesis
25/168
AC-Induced Corrosion
17
Therefore for fault conditions, the AC voltages are a
combination of both the inductive and
conductive voltages arising from fault currents flowing in the
phase wires and into the soil from
each faulted pylon. In general, during a fault condition, the
induced voltage will tend to increase
with [23,26]:
an increase in soil resistivity;
a decrease in the distance between the faulted structure and the
pipeline;
a decrease in the conductivity of the pipeline external
coating;
close proximity to a pylon earthing assembly.
The effect of a fault on the predicted induced AC voltage can
extend several kilometres away
from he crossing point.
Transferred voltagesDue to inductive and conductive coupling,
considerable voltages can be transferred many
kilometres beyond the end of a common corridor. As a result, it
is necessary to properly study the
entire pipeline system, as mitigation may be required, for
instance, to protect sites which are far
away from the transmission line exposure location[26].
1.6 CHARACTERISATION OF AC-INDUCED CORROSIONIn the corrosion
field it is fundamental to ascertain the causes of corrosive
phenomena. This may
be done by in situ measurements or by using samples that can be
subsequently analysed in the
laboratory.
The following are some aspects that can be considered in order
to identify a corrosion failure as
induced by AC interference[ ]29 :
presence of AC voltage on the pipeline;
protection voltage on the pipeline (negative OFF-potential
greater than -850 mV CSE);
presence of a defect in the coating (usually an open small
defect) often associated withcoating disbondment and
degradation;
elevated pH values of the soil in the vicinity of the corrosion
(pH >10);
the corrosion form is usually rounded (advanced stage) or
penetrating pits (initial stage);
the corroded area is larger than the coating defect;
corrosion products are very bulky and could be removed as a
block;
high AC and high DC cathodic protection densities.
-
5/27/2018 Ac Corroosion Phd Thesis
26/168
AC-Induced Corrosion
18
1.7 CORROSION RATEThe principal factors that affect AC corrosion
can be summarised as follows[18,12,29, , ],3031
corrosion increases with increased AC density or alternating
voltage (AV) (Figure 1.13);
cathodic protection decreases AC corrosion, but the magnitude of
the DC currents
required is high;
corrosion decreases with increased AC frequency;
the chemical composition and resistivity of the soil have a
strong influence on the
corrosion rate: corrosion increases in the presence of
carbonates or bicarbonates and
decreases in the presence of chlorides;
corrosion increases with increased temperature;
the average corrosion rate decreases rapidly with time, then
less slowly with the
exposure time and finally stabilizes to a steady value (Figure
1.14);
corrosion rates are higher when the area of exposed steel is
approximately 1 cm2.
0.0
0.2
0.4
0.6
0.8
1.0
0 300 600 900 1200 1500 1800
Peak cell voltage (mV)
Corrosionrate(g/cm
2y)
Figure 1.13 - Corrosion rate as a function of peak alternating
voltage. Experiments were conducted at room
temperature under nitrogen purge for 24 h. The 60 Hz AV was
applied between a Cu-Ni alloy and acounter platinum mesh[31]
-
5/27/2018 Ac Corroosion Phd Thesis
27/168
AC-Induced Corrosion
19
0.0
0.2
0.4
0.6
0.8
1.0
0 20 40 60 80 100 120 140 160
Time (h)
Corrosionrate(g/cm
2y
)
Figure 1.14 - Corrosion rate as a function of exposure time (AV
= 1000 mV)[31]
Studies on carbon steel specimens exposed to soil under the
influence of AC and CP
polarisation showed that corrosion rates due to AC corrosion are
not necessarily high, but that the
rates are certainly multiples of the normal corrosion rates of
steel in soil in the absence of AC[ ]32 .
The results of this work are summarised in Figure 1.15.It can be
noticed that AC current density in
the order of 20 A/m2could significantly increase corrosion; CP
can be effective at "low" AC current
densities but not at "high" AC current densities. The data also
suggest that even at "Low AC + No
CP" conditions, at AC densities which are considered 'safe' by
many publications (20 A/m 2), the
observed increase in corrosion rate, relative to a control, is
by a factor of 2. For "High AC + HighCP" results showed that
protection could not be achieved. This last result is quite
unexpected. It
does not seems logical to have higher corrosion rates whit
higher cathodic protection. However
such behaviour has been reported also by other authors [29].In
any case more analysis are needed
in order to verify such experimental evidence and to explain the
reason of it. Most probably the
high corrosion rate reported can be explained considering the
enhancement of the cathodic
process that can lead to coating disbondment[32].
-
5/27/2018 Ac Corroosion Phd Thesis
28/168
AC-Induced Corrosion
20
0
100
200
300
400
500
600
Low AC +No CP
High AC +No CP
Low AC +Low CP
Low AC +High CP
High AC +Low CP
High AC +High CP
Test category
R
elative(tocontrolspecimens)cor
rosion
rate >150>300High AC + High CP
300High AC + Low CP
>150150
-
5/27/2018 Ac Corroosion Phd Thesis
29/168
AC-Induced Corrosion
21
The European approach is the specification of the maximum
allowable AC density to be
measured by means of corrosion coupons: the threshold value so
far proposed is 30 A/m2[ ]35 .
The German standard, DIN 50 925 states that, for coated steel
pipelines, AC current density
should be less than about 30 A/m2while DC current density should
be maintained about 1 A/m2[ ]36 .
Others criteria are based on the maximum AC voltage. The NACE
RP0177-2000[ ]37 suggests
15 V, but this value has been chosen mostly considering safety
criteria for the persons than AC-
induced corrosion. The minimum voltage level below which AC
enhanced corrosion cannot occur
has not yet been defined. It has been reported that it may occur
at induced AC voltage levels of
only a few volts[23], therefore mitigation of AC potentials to
values below 15 V may not be sufficient
with respect to AC current densities at the coating holidays.
Application of increased cathodic
protection current may be required to control AC corrosion
[32].
Evaluation of corrosion risk is achieved by several
measurements. The association of several of
them makes it possible to consider a situation as potentially at
risk or not.
The parameters influencing AC corrosion risks are[12,33]:
the real level of polarisation of the structure by cathodic
protection;
the level of AC current density;
the extent of the metal surface exposed at coating defects;
the electrolyte (soil composition, resistivity, etc.). Soil
composition strongly affects the
AC-corrosion risk. Identical electrical parameters can result in
significantly different
ACcurrent densities;
climatic conditions, such as surface temperature, seem strongly
affect the AC-current
density on coupons.
AC corrosion is most likely to occur only on well-coated
structures where the AC current is shown
to transfer from very small defects in the coating.
1.8.1 DC Potential Measurements in the Presence of ACNatural gas
transmission pipelines are usually supplied with cathodic
protection (CP) to protect
coating defects against corrosion. The potential measurement is
easy to carry out, but its
interpretation may present some concerns when stray currents
cause huge ohmic drop
contributions which may lead to misinterpretation[11].
The scope of the cathodic protection monitoring is to verify the
protection level of the structure by
checking that the potential matches the protection criteria. In
the presence of DC or AC stray
currents the potential measurement may lead to misleading
interpretations.
The potential reading is a function of the reference electrode
position with respect to the structure
and is the sum of three terms (Figure 1.16):
-
5/27/2018 Ac Corroosion Phd Thesis
30/168
AC-Induced Corrosion
22
E = Eeq+ + IR Eq. 1.12
Eeqis the thermodynamic contribution and represents the
potential value of the electrode versus
the standard hydrogen electrode; is the overpotential
contribution, i.e. the polarisation due to
cathodic protection or interfering currents; IRis the ohmic drop
in the electrolyte, depending upon
the reference electrode position with respect to the structure,
on the electrolyte resistivity and on
the circulating current within the electrolyte. E, which
includes all terms, is often called
ONpotential.
The term EIR= ET+ = Etrueis called true potentialor
OFFpotentialand gives the value of
the overpotential due to the current flux.
E
Soil (or electrolyte)
Voltmeter
RI
I
I'
Eeq
Reference electrode
Pipeline
Reference electrode
Figure 1.16 - Potential measurement
Therefore stray currents strongly influence the potential
reading because ohmic drop contribution
is often high. In case of DC stray currents, the measured
potential is more negative than the true
potential at cathodic zones and more positive at anodic ones. To
know the true polarisation level
the ohmic drop contribution has to be eliminated.
The ON/OFF technique is commonly used to determine the true
potential of protected structures
provided a correct interpretation. For continuous monitoring, a
corrosion coupon is often used with
a reference electrode located close to the coupon. This device
is also called true potential probe.
The case of AC interference seems to be particularly critical:
depolarisation time lower than 1 ms
and corrosion attacks on samples with instantaneous OFFpotential
below of 0.85 V CSE have
been reported[33]. In literature there is lack of agreement
about the correct procedure to measure
the true potential in the presence of AC interference [33,39,21,
,38 39]. It has also been claimed that the
potential measurement misinterpretations caused by the AC could
be also more severe than those
related to the DC. In the specific case of coating defects in
cathodically protected buried pipelines
influenced by induced AC current, the CP evaluation becomes
complicated, and the usual
potential criteria are no longer necessarily valid.
-
5/27/2018 Ac Corroosion Phd Thesis
31/168
AC-Induced Corrosion
23
It has been claimed that the instantaneous OFF-potential
measurement is a non suitable
technique for the characterisation of the AC-corrosion risk of
cathodically protected structures
because the obtained data are strongly influenced by the soil
composition causing a very fast
depolarisation of the metal surface[33].
In Figure 1.17 the instantaneous potential of a sample
interfered by AC is represented. The curve
arepresent the case of a non protected pipeline, while band
crepresent cathodic protection. The
form and the amplitude of the sinusoidal curves remain, but
their position with respect to the
reference axis is changed[12].
Time (s)
E SCE a
c
b
-850
(mV)
Figure 1.17 - Evolution of the potentials in the absence (a) and
the presence (b and c) of cathodic protection[
The samples potential curve is a sinusoidal curve whose
displacement towards the more
electronegative potentials is proportional to the level of
cathodic protection. Consequently, if the
level of cathodic protection is increased in such a way that
there is no more positive current the
potential becomes more and more electronegative[12]. The case c
is when total compensation of
the positive AC. However, to achieve this shift in potential
high DC current are required, which
would result in the production of large amount of hydrogen which
may be disadvantageous
because of all the problems relate to overprotection as hydrogen
stress cracking or coating
disbondment.
Unfortunately the measurement of the true potential do not give
information about the
instantaneous IR free pick potential.
Recently a measuring technique was developed to measure the AC
induced effects on metallic
structures. The system, with a switch OFF method on a coupon,
provide IR-free potential, the
current density and the phase angle at every instant of the AC
period [39]. Using this measuring
technique, the pipeline is considered protected from AC
corrosion if the coupon-to-soil
EOFFpotential is at any moment more negative than the protective
potential range.
-
5/27/2018 Ac Corroosion Phd Thesis
32/168
AC-Induced Corrosion
24
1.8.2 AC voltageAC in soil or AC intercepted by buried metallic
structures is surveyed by measuring the potential
with an electronic multimeter that measures AC voltage for
frequencies between 15 and 100 Hz. A
mobile oscilloscope is useful for verifying the type of waves,
their amplitude and frequency.
For safety reason, it is generally considered that the AC
voltage should be lower than 15 V rms
(rms = root mean square)[37].
Unfortunately, unlike the case of DC stray currents, there is no
correlation between corrosion rate
and AC voltage and therefore, even today there are no criteria
based on AC voltage
measurements. It is however possible to estimate AC density
flowing through a coating defect of a
certain size by means of the following equation:
r
4
Vi F = Eq. 1.13
when soil resistivity () , defect radius (r), and voltage drop
to remote earth (VF) are known[11,16, 40].
The AC voltage level alone does not allow the evaluation of AC
corrosion risk. Two separate
pipelines exposed to the same AC voltage may or may not be
subject to AC corrosion.
Nonetheless, pipelines exposed to low AC voltage seem to be
subject to very little risk[12].
1
10
100
1000
10000
0 5 10 15 20 25 30VAC
Vcorr(m/year)
Initial Vcorr
Steady State Vcorr
Figure 1.18 - Corrosion rate measured by mean of ER corrosivity
probes as a function of AC voltage (UAC) ona Danish site[7]
Field data showed that a relatively small-sized defect (6 to13
cm2) in the coating on the pipe in a
relatively low-resistivity soil could have AC current density in
excess of 100 A/m2, while the AC
voltage could be as low as 6 V[32]. Others field
surveys[7,13]showed that very high corrosion rates
were associated to AC potentials lower than 10 V (Figure 1.18).
It has also been reported[13]that
coupons exposed to the same alternating voltage showed AC
density which varied with time and
-
5/27/2018 Ac Corroosion Phd Thesis
33/168
AC-Induced Corrosion
25
location. This was probably due to some changes in soil
resistivity and in chemical conditions
closest to the steel surfaces.
1.8.3 AC current densitiesThe determination of AC current
densities is a very important parameter in order to evaluate AC
corrosion risks: the higher the density, the greater the risks.
Current measurements on steel
coupons must be carried out with a shunt whose resistance is
less than that of the pipeline, of the
potential coating defects and of the coupon. If this is not the
case, the values will be significantly
erroneous.
As previously said the threshold value suggested by standards
and laboratory tests[34,35,36]ranges
between 20 and 30 A/m2. Though it has been reported that an AC
current density of 20 A/m2could
increase corrosion nearly two-fold relative to a control
specimen; suggesting a threshold current
density lower than 20 A/m2[32].
It seems that there is not a theoretical 'safe' AC current
density, i.e., a threshold below which AC
does not enhance corrosion; however, a practical one exists for
which the increase in corrosion
due to AC is not appreciably greater than the free-corrosion
rate for a particular soil condition [32].
1.8.4 AC Density/DC Density RatioGenerally higher is the AC
higher is the DC cathodic protection required[32, ]41 . Therefore
recently
in France theAC/DC current densityratiowas proposed as AC
induced corrosion risk indicator[12]
:when the ratio is above 10, AC induced corrosion is almost
certain; above 3 the risk is serious;
below 2, the risk is less, although not necessarily totally
absent.
This approach has been proposed considering the following
experimental data and field
measurement[12]:
measurements of DC and AC current carried on field showed that,
at the corrosion sites,
the ratios were always higher than in those areas where no
corrosion had been
observed;
laboratory measurements showed that in the presence of AC
currents was necessary a
quantity of DC current greater than that is normally applied in
order to obtain 850 mV;
through the calculation of the AC/DC current density ratio,
errors due to individual
measurements of DC and AC intensities are partially eliminated;
therefore the most
significant errors of measurement with regard to the internal
resistance of
instrumentation and measurement circuits are eliminated;
even if the current densities increase in magnitude as the
defects surface areas
decrease, the ratio remains constant whatever the surface areas
of the coupons used to
perform the measurements;
-
5/27/2018 Ac Corroosion Phd Thesis
34/168
AC-Induced Corrosion
26
this ratio also makes it possible to adjust the level of DC
current with regard to the AC
current, allowing therefore the definition of the quantity of DC
current sufficient for
improved polarisation of a defect compared to a level of AC
influence.
It should be noted that the value of the ratio provides no means
of quantifying AC corrosion and
that this approach alone cannot be applied since there is no
mention on the potential value. In fact
even if DC true potential measurements have been proven to be of
difficult interpretation in the
presence of AC, their values remain a fundamental parameter
during corrosion monitoring and risk
evaluations.
Furthermore considering the experimental evidence that high
alternate current with high cathodic
protection levels may be associated with very high corrosion
rates[32](Figure 1.15), it is possible to
have a not too high AC/DC ratio whereas the actual corrosion
rate is very high.
1.8.5 Cathodic Protection Criteria Based on AC and DC
DensitiesMeasured by Coupons
New cathodic protection (CP) criteria have been developed based
on AC and DC current
densities measured using coupons[17]. The numerical expression
of the criteria is reported in
Tab. 1.2, while the graphic expressions of the current
density-based criteria are illustrated in
Figure 1.7.DC current density is determined as the average
current density on a coupon (i.e. CP
conditions). Positive values of DC current density indicate that
the current is flowing from the soil to
the coupon. Cathodic protection area is indicated in grey.
Corrosion activity can be divided roughly
into three areas corresponding to the possible main risks of DC
corrosion, AC corrosion andoverprotection [17].
DC corrosion is defined as the corrosion due to the lack of CP
current, stray currents from DC
powered transit systems or other DC interference. The lowest
limit of DC current density to prevent
DC corrosion was determined at 0.1 A/m2[17].On the other side
the excessive DC current density
may lead to hydrogen stress cracking or coating disbondment
[17].
The AC corrosion area corresponds either to low DC cathodic
protection and low AC densities or
to high AC densities.
Tab. 1.2 - Numerical expression of the cathodic protection
criteria for buried steel pipelines based on DCand AC current
densities measured using coupons[17]
0.1 A/m2iDC< 1.0 A/m2then iAC< 25 iDC
1.0 A/m2iDC< 20 A/m2then iAC< 70 Arms/m
2
where:
iDC= DC density (A/m2)
iAC= AC density (A/m2)
-
5/27/2018 Ac Corroosion Phd Thesis
35/168
AC-Induced Corrosion
27
AC Corr osion
DC Corros ion
Overprotection
10070
25
10
25
1
0.10.01 0.1 1 10 20 100
Protection
iDC (A/m2)
iAC
(Arms
/m2)
AC Corr osion
DC Corros ion
Overprotection
10070
25
10
25
1
0.10.01 0.1 1 10 20 100
Protection
iDC (A/m2)
iAC
(Arms
/m2)
Figure 1.19 - Graphic illustration of the cathodic protection
criteria for buried steel pipelines based on DC andAC current
densities measured using coupons[17]
1.8.6 Size of the Coating DefectsSmall rather than large coating
defects are susceptible to AC corrosion since the spread
resistance(in m2) associated with the defects increases with
increasing area:
Rs= K sd (m2) Eq. 1.14
where Kis a constant depending on the geometry of the defect,
dis a measure of the extensionof the defect, and sis the specific
resistivity of the local soil environment adjacent to the
defect
[7].
Experience[12] showed that the diameter of coating defects on a
polyethylene coating is often
closer to a few millimetres than it is to a centimetre. In
addition, cases of corrosion have been
observed on defects of very small areas (around 1 mm2).
Probabilistic approach
The idea that the current density flowing through an holiday in
the insulating coating is a
meaningful parameter related to the risk of corrosion is
commonly accepted within the community
of experts in AC corrosion; on the other side, the assessment of
corrosion conditions is only
possible when the pipeline has already been laid down in the
trench and the current density can be
really measured on simulated holidays. These measurements in the
field are affected by a
possible wide variation of the spread resistance of these
simulated holidays, which is connected to
complex and not yet completely understood electrochemical
reactions deriving from DC and AC
current effects and the chemical composition of the soil in
contact with the bare steel. It has been
demonstrated by laboratory tests that this spread resistance
may, during time, increase by as
much as 100 times or decrease by as much 60 times
[25].Simulation tools able to assess the level
of current density exchanged between pipe and soil through the
insulating coating holidays havebeen proposed. The method consists
in the integration of deterministiccalculations, necessary for
-
5/27/2018 Ac Corroosion Phd Thesis
36/168
AC-Induced Corrosion
28
the determination of the induced voltage profile along the
pipeline route, with probabilistic
calculations necessary for the evaluation of the holidays size
and consequently of the current
density.
The parameters and the characteristics relevant to the random
quantities (i.e. the holidays area
distribution, the number of holidays per unit length) have been
inferred starting from experimental
data coming from the field. The probability levels are directly
related to:
the levels of induced voltage on the pipeline;
the number of holidays per unit length besides the other
statistical characteristics of the
holidays distribution;
the length of the pipeline section considered.
Deterministic approaches are based on specific assumptions
concerning the holiday size and
location and are normally focused on the worst case study,
whereas probabilistic approaches takeinto account of the random
nature of some significant parameters like the holidays size
and
location.
The proposed algorithm[25]allows the assessment of the
probability associated to the exceeding
of a certain threshold current density value (J*) in a generic
section of pipeline; the advantage of
such an approach is the possibility to individuate those
pipeline sections which are more exposed
to the AC corrosion risk. The algorithm is still deterministic
as far as the calculation of electric
quantities (voltage, current, current density) is concerned
while it is statistic as far as the holidays
data are concerned; therefore the results of the algorithm are
given in terms of probability ofexceeding a certain value of
current density for a generic section (J*), of given length, along
the
pipeline.
Thus, if the pipe under the electromagnetic influence of an AC
power or of an electrified railway
line assumes an alternating voltage
V at the holiday location, the current density J exchanged
between pipe and soil through the holiday itself is given
by:
A
4
d
V
AR
VJ
sh
h
+== Eq. 1.15
where Rh is the resistance to remote earth of the pipeline
through the holiday, known as spread
resistance. The spread resistance is given by the sum of two
contributions (Figure 1.20): Rch,
which represents the resistance relevant to the small cylinder
inside the coating, and Reh
, which
represents the earth resistance (with respect to the remote
earth) of the holiday; for points inside
the holiday the resistivity is h while in all the other points
is s; A is the area of the holiday.
-
5/27/2018 Ac Corroosion Phd Thesis
37/168
AC-Induced Corrosion
29
d
soil s
pipeline axis
insulating coating
pipeline wall
holiday hA
Rch
Reh
= Rh
d
soil s
pipeline axis
insulating coating
pipeline wall
holiday hA
Rch
Reh
= Rh
Figure 1.20 Spread resistance or holiday resistance Rhto remote
earth
Once the threshold current density value (J*) is established the
upper limit (in function of the pipe
voltage V) for the holiday area A in order to have the exceeding
of the limit value J *
for the current
density through the holiday itself can be evaluated.
2
h2s
d
*J
V16*A
Eq. 1.16
where A*(V) is the limit area.
The probability (P%) of exceeding a certain value for the
current density is mainly related to two
important quantities[25]:
the length of the pipeline section considered (the longer is the
section, the higher is the
probability);
the mean value of the voltage assumed by the pipeline in the
same section (the higher is
the mean voltage the higher is the probability).
It is therefore possible to obtain set of curves relating the
probability of exceeding the limit value
of current density J* versus the pipeline section length for a
given value of the mean voltage V. In
Figure 1.21 an example is shown by considering a limit value of
J*=30 A/m2.
-
5/27/2018 Ac Corroosion Phd Thesis
38/168
AC-Induced Corrosion
30
0
10
20
30
40
50
10 100 1000
Pipeline section length (m)
P%(J>J
*)
0
10
20
30
40
50
10 100 1000
Pipeline section length (m)
P%(J>J
*)
Figure 1.21 - Per cent probability (P%) of exceeding the current
density limit J*=30A/m
2versus the pipeline
section length for different values of the mean voltage; the
curves from top to bottom correspond toV = 50, 40, 30, 20, 10V
respectively; n
h=0.7 holidays/km
[25]
1.8.7 ER corrosivity probesIn the specific case of CP of coating
defects in buried pipelines influenced by induced AC
current, the CP evaluation becomes complicated, and the usual
potential criteria are no longer
necessarily valid. In Denmark, the connection of high
sensitivity electrical resistance (ER) sensor
for evaluation of CP and corrosion rate measurements has become
usual practice, particularly
when studying risks of AC corrosion[7,20,21].
ER based AC corrosivity probe
The AC corrosivity probe with associated instrumentation is
based on the high-resolution
differential electrical resistance. The technique utilises the
fact that the electrical resistance of a
metal probe element increases when corrosion diminishes the
thickness of the element.
Considering a metal element with dimensions L (length), W
(width), and (thickness), the
electrical resistance of the element can be expressed
by[7,20,21]
:
( )
=W
LTR Eq. 1.17
where (T) is the specific resistivity of the element material,
which depend on the temperature.
Re-arranging the equation, the thickness of element can be
obtained as a function of element
resistance[7,20,21]
:
( )R
1
W
LT = Eq. 1.18
Differentiation of the equation gives the corrosion rate (vcorr)
of the element as:
-
5/27/2018 Ac Corroosion Phd Thesis
39/168
AC-Induced Corrosion
31
( )TLW
dt
dR
dt
dv
2
corr
=
= Eq. 1.19
Hence, once the time change of the electrical resistance of the
element, the element dimensions
and resistivity are known the corrosion rate can be evaluated.
The period of time needed to
quantify a certain corrosion rate can be calculated as follows
[7,20,21]:
( )TLW
v
Rt
2
corr
= Eq. 1.20
The equation predicts that the necessary period of time increase
with increasing W/L ratio, and
increases with the square of the element thickness. Hence, high
sensitivity ER probes can be
produced by using thin elements, but this decreases probe
lifetime as well. R can be regarded as
the minimum resistance change needed to provide a reliable
measurement[7,20,21].
In Figure 1.22 an examples of corrosion rate measured by mean of
ER corrosivity is given. It can
be noticed that, as previously said, the corrosion rate can be
very high even when the AC potential
is lower than 10 V.
Figure 1.22 - Corrosion rate measured by mean of ER corrosivity
probes and AC voltage throughout time ona Danish site[7]
1.9 DAMAGE TO PIPELINE EXTERNAL COATINGDuring an AC power fault
condition, current flowing from the AC power lines to earth can
give
rise to instantaneous very high conductive and inductive AC
voltage and current on a pipeline.
Fault duration and magnitude are controlled by the AC circuit
protection equipment installed by the
local AC power authority. Voltage levels of several thousand
volts are possible and can damageexternal coating. To prevent undue
damage to an external coating, a maximum short-term
-
5/27/2018 Ac Corroosion Phd Thesis
40/168
AC-Induced Corrosion
32
cumulative AC voltage level of 3 kV to 5 kV [23] is usually
considered acceptable for fusion bonded
epoxy and polyethylene coated pipelines that have high integrity
field joint coating. For tape-wrap,
coal tar enamel and asphalt coating systems 2 kV is suggested as
threshold value.
Also in case of AC-induced corrosion occurred during
steady-state conditions, damages of the
coating in the area surrounding the pits are often reported[19].
Coating disbondment is often
associated to AC interference due to the enhanced cathodic
process.
1.10 SAFETY FOR PERSONNELIn terms of safety for personnel and
the public, a potential shock hazard exists when someone
touches an exposed part of the pipeline (such as a valve) while
standing on soil which is at a
different potential. The touch voltage to which a person would
be subjected when standing nearthe pipeline and touching an exposed
part of the pipeline is defined as the difference in potential
between the pipeline metal and the earth surface above the
pipeline. Similarly, the step voltageis
the difference in potential between a persons feet, and is
defined as the difference in earth surface
potential between two points spaced 1 m apart[23,26].
The following AC touch voltage levels are typically considered
acceptable
Less than 50 V AC steady-state for buried pipeline sections that
may be exposed
periodically by operator personnel, who are aware of the
possible voltage levels and are
suitably prepared for the problem;
Less than 15 V AC steady-state for aboveground appurtenances to
minimize the
danger to personnel and livestock.
The 50 V level is based on the fibrillation current threshold
(50 mA) for an infinitely long fault
duration[ ]42 , for a body resistance of 1000 and assuming a
person is bare-foot while touching the
pipe with his/her bare hand. The 15 V level is as set by NACE
RP0177-2000[37]and has become
the standard value.
1.11 AC INFLUENCE ON CATHODIC PROTECTIONAs previously said the
AC-interference causes perturbation of cathodic protection and
AC
corrosion risks. One of the most serious consequences of steady
state induced AC is that
corrosion can occur even if cathodic protection levels satisfy
the standards and despite the
induced AC has been reduced to less than 15V as required by
standards.
The presence of alternating voltage and current on buried
metallic pipelines can cause
malfunctioning or even damage to the Transformer/Rectifier (T/R)
units of impressed current
-
5/27/2018 Ac Corroosion Phd Thesis
41/168
AC-Induced Corrosion
33
cathodic protection. This is particularly valid for T/R units
that are not equipped with suitable AC
filter and surge protection.
1.12 HOW AC INFLUENCES CORROSION PROCESSESThe mechanism by which
AC affects the corrosion process is not yet understood:
different
theories and hypotheses have been proposed, but none of them is
able to fully explain and
describe such a phenomenon.
An important aspect to consider is whether at the frequency of
the interfering signal, the
electrochemical processes are reversible or not[ ]43 . In fact
in the presence of alternating voltage the
reversibility of the electrochemical processes is related to the
frequency of the signal: if the
frequency is not high enough and the chemical reaction is not
reversed, then the metal coulddegrade, for example, because the
metal ions migrate and precipitate into the solution. The
irreversibility of the chemical reactions occurring at the
interface may causes changes on the
double layer compo