Corrosion Protection Underground Pipelines

SS SS S K o r r o s i o n s s c h u t z t ec ec h n i k  

Corrosion and Corrosion Protection of Underground Steel Pipelines

+

SSS Korrosionsschutztechnik Korrosionsschutztechnik GmbH & Co. KG Münchener Str. 69 D-45145 Essen

-

Telefon Telefax Internet E-mail

+49-201-17 55-702 +49-201-17 55-602 http://www.sss-kt.de [email protected]

Contents

Preface 1

2

2.1 2.1 2.2 2.3 2.3

2.4 2.4 2.5 2.5 2.5. 2.5.1 1 2.5 2.5.2 2.5. 2.5.3 3 2.5 2.5.4

3

3.1 3.1. 3.1.1 1 3.1. 3.1.2 2 3.1. 3.1.3 3 3.2 3.2 3.2. 3.2.1 1 3.2.1.1 3.2.1.2 3.2. 3.2.2 2 3.2.2.1 3.2.2.2 3.2.2.2 3.2. 3.2.3 3

Definitions: Corrosion, corrosion manifestation, corrosion damage, and corrosion protection Basics of underground corrosion and its conditions Part Partia iall reac reacti tion ons s in corr corros osio ion n chem chemis istr try y Influencing factors and consecutive chemical reactions Interactions between anodic and cathodic partial reactions mutual dependence Corr Co rros osio ion n with with unif unifor orm m me meta tall cons consum umpt ptio ion n Format rmatio ion n of elec lectroc troche hem mical ical cells ells Anod Anodic ic reg regio ion; n; loca locali lize zed d att attac ack k Cathodic region; formation of covering layers Inter nterac acti tion on betw betwee een n ano anodi dic c and cathodic regions Influence of of an anodic and cathodic area ratio Estimation of corrosion danger for pipelines not subject to electrical influence; Fundamentals of protection methods Aggressiveness of of th the so soil Unif Un ifor orm m oxyg oxygen en corr corros osio ion n Cell Ce ll form format atio ion n in the the soil soil Inten ntensi sity ty of corr corros osio ion n in different types of soil Methods of corrosion protection Coat Co atin ings gs:: „pas „passi sive ve““ cor corro rosi sion on pro prote tect ction ion Corrosion chemical interactions on a coated steel surface Required protective properties of coating materials Cathod hodic prote otection ion: “active“ corrosion protection Potential dependence of corrosion Applicat Application ion of cathodic cathodic protecti protection; on; verification of protection potential Acti Active ve prot protec ecti tion on + pas passi sive ve prot protec ecti tion on = complete protection

4 4.1 4.1

4.2 4.2 4.3 4.3 4.4 4.4 4.5 4.5 4.5.1 4.5 .1 4.5. 4.5.2 2

5

5.1 5.2 5.2 5.2. 5.2.1 1 5.2. 5.2.2 2 5.2. 5.2.3 3 5.2.4 5.3 5.3. 5.3.1 1 5.3. 5.3.2 2 6

6.1 6.2 6.2 6.3 6.3. 6.3.1 1 6.3.2 6.3.3 6.3. 6.3.4 4

Cathodic protection of pipelines Constructional requirements put by cathodic protection Cath Ca thod odic ic prot protec ecti tion on by usin using g gal galva vani nic c ano anode des s Cath Ca thod odic ic prot prote ectio ction n by by im impres press sed curre urrent nt Current feed test est and adjustment Side ide ef effect ects of of the pro protect ection ion cu curre rrent Inte Interf rfer eren ence ce with with extr extrane aneou ous s equi equipm pmen entt Inter nterna nall int inter erfferen erenc ce within water pipelines Corrosion danger caused by foreign cathodes; localized cathodic protection External cu currents Corr orrosion ca caused used by fo foreig eign ca cathode odes Char Ch arac acte teri rist stic ic of a for forei eign gn cath cathod ode e Exam Exampl ples es of of for forei eign gn cat catho hode des s Esti Estima mati tion on of of corr corros osio ion n dang danger er Coating influence Protection measures Galva alvani nic c separ eparat atio ion n Loca Locall cat catho hodi dic c prot protec ecti tion on,, hot spot protection Corrosion caused by stray currents; protection methods Origins of stray currents Char Ch arac actteris eristi tics cs of stra stray y curre urrent nt acti action on Protection measures against stray currents Immediate draina inage Rectified dra drainage Forced drainage Use Use of cont contro roll llin ing g rect rectif ifie iers rs for forced drainage

Contents

Preface 1

2

2.1 2.1 2.2 2.3 2.3

2.4 2.4 2.5 2.5 2.5. 2.5.1 1 2.5 2.5.2 2.5. 2.5.3 3 2.5 2.5.4

3

3.1 3.1. 3.1.1 1 3.1. 3.1.2 2 3.1. 3.1.3 3 3.2 3.2 3.2. 3.2.1 1 3.2.1.1 3.2.1.2 3.2. 3.2.2 2 3.2.2.1 3.2.2.2 3.2.2.2 3.2. 3.2.3 3

Definitions: Corrosion, corrosion manifestation, corrosion damage, and corrosion protection Basics of underground corrosion and its conditions Part Partia iall reac reacti tion ons s in corr corros osio ion n chem chemis istr try y Influencing factors and consecutive chemical reactions Interactions between anodic and cathodic partial reactions mutual dependence Corr Co rros osio ion n with with unif unifor orm m me meta tall cons consum umpt ptio ion n Format rmatio ion n of elec lectroc troche hem mical ical cells ells Anod Anodic ic reg regio ion; n; loca locali lize zed d att attac ack k Cathodic region; formation of covering layers Inter nterac acti tion on betw betwee een n ano anodi dic c and cathodic regions Influence of of an anodic and cathodic area ratio Estimation of corrosion danger for pipelines not subject to electrical influence; Fundamentals of protection methods Aggressiveness of of th the so soil Unif Un ifor orm m oxyg oxygen en corr corros osio ion n Cell Ce ll form format atio ion n in the the soil soil Inten ntensi sity ty of corr corros osio ion n in different types of soil Methods of corrosion protection Coat Co atin ings gs:: „pas „passi sive ve““ cor corro rosi sion on pro prote tect ction ion Corrosion chemical interactions on a coated steel surface Required protective properties of coating materials Cathod hodic prote otection ion: “active“ corrosion protection Potential dependence of corrosion Applicat Application ion of cathodic cathodic protecti protection; on; verification of protection potential Acti Active ve prot protec ecti tion on + pas passi sive ve prot protec ecti tion on = complete protection

4 4.1 4.1

4.2 4.2 4.3 4.3 4.4 4.4 4.5 4.5 4.5.1 4.5 .1 4.5. 4.5.2 2

5

5.1 5.2 5.2 5.2. 5.2.1 1 5.2. 5.2.2 2 5.2. 5.2.3 3 5.2.4 5.3 5.3. 5.3.1 1 5.3. 5.3.2 2 6

6.1 6.2 6.2 6.3 6.3. 6.3.1 1 6.3.2 6.3.3 6.3. 6.3.4 4

Cathodic protection of pipelines Constructional requirements put by cathodic protection Cath Ca thod odic ic prot protec ecti tion on by usin using g gal galva vani nic c ano anode des s Cath Ca thod odic ic prot prote ectio ction n by by im impres press sed curre urrent nt Current feed test est and adjustment Side ide ef effect ects of of the pro protect ection ion cu curre rrent Inte Interf rfer eren ence ce with with extr extrane aneou ous s equi equipm pmen entt Inter nterna nall int inter erfferen erenc ce within water pipelines Corrosion danger caused by foreign cathodes; localized cathodic protection External cu currents Corr orrosion ca caused used by fo foreig eign ca cathode odes Char Ch arac acte teri rist stic ic of a for forei eign gn cath cathod ode e Exam Exampl ples es of of for forei eign gn cat catho hode des s Esti Estima mati tion on of of corr corros osio ion n dang danger er Coating influence Protection measures Galva alvani nic c separ eparat atio ion n Loca Locall cat catho hodi dic c prot protec ecti tion on,, hot spot protection Corrosion caused by stray currents; protection methods Origins of stray currents Char Ch arac actteris eristi tics cs of stra stray y curre urrent nt acti action on Protection measures against stray currents Immediate draina inage Rectified dra drainage Forced drainage Use Use of cont contro roll llin ing g rect rectif ifie iers rs for forced drainage

7

7.1 7.1 7.2 7.2 7.2. 7.2.1 1 7.2.2 7.2 .2 7.3 8

8.1 8.2 8.2 8.3 8.3 8.3. 8.3.1 1 8.3. 8.3.2 2 8.3. 8.3.3 3 8.3.4 8.3. 8.3.5 5

9 9.1 9.2 9.2

High voltage interference, protective measures and effect to cathodic protection Types ypes and and caus causes es of high high volt voltag age e infl influe uenc nce e Protective measures against too high voltages in personal contact Shor Shortt tim time e int inter erfe fere renc nce e cau cause sed d by by sho short rt circuit currents Perm Perman anen entt int inter erfe fere renc nce e cau cause sed d by by ope opera rati ting ng currents Grounding the pipeline Conditions related to stress corrosion cracking; preventive measures Critical agents causing SCC Danger nger caus aused by cor corrrosion ion on on th the ins insiide surface of the pipes; protective measures Dange nger cau caus sed by by co corrosi osion on th the ou outside ide surface of the pipes; protective measures SCC ca cause used by by nit nitrrates SCC SC C caus caused ed by sodi sodium um hydr hydrox oxid ide e SCC SC C caus caused ed by sodi sodium um bica bicarb rbon onat ate e SCC in in ha hard sp spots SCC SC Cc cau ause sed d by by cat catho hodi dica call lly y pro produ duce ced d hydrogen

9.4 9.4 9.4.1 9.4. 9.4.2 2

Behaviour of of st stainless st steel in th the gr ground When to use stainless steel Stainl ainles ess s steel teel part parts s act acting ing as as for fore eign ign cathodes vs. unalloyed steel Overall co corrosion pr properties and influence of alien currents Cond Co ndit itio ions ns requ requir irin ing g prot protec ecti tive ve me meas asur ures es Sensitization Conce ncentra ntratted chlo chlorride ide

10

Symbols

9.3

Preface

In this paper corrosion and corrosion protection of underground steel pipelines are treated. Complementary to a number of existing technical regulations it describes causes of corrosion processes and possible protection measures based upon many years practical experience of our customers. It gives hints and explanations to engineers in charge of planning and operating steel pipelines in order to help them realize and assess corrosion dangers that may arise, and provide for suitable protection measures in time. The possible corrosion processes being manifold and complex, and the service conditions of the pipelines not always being thoroughly anticipated, completeness cannot be claimed, nor may any guarantee be given for security against corrosion damage. The planning, construction, and operation of active corrosion protection facilities are the tasks of pertinent expert contractors but are not the subject of this text.

As a coherent and compact treatise on corrosion-related problems and based upon the general principles of corrosion chemistry, this paper is intended to contribute to a more common understanding of these. This understanding is to promote the application of suitable corrosion protection, e.g. high grade passive protection by polyethylene coating and application of modern electrochemical protection methods so that steel pipelines can be put to use with optimal profit and operated securely. The paper has been subdivided into many paragraphs similar to a handbook so as to facilitate the locating of particular problems. The definitions of terms, and the electrochemical principles of underground corrosion have been described in great detail deliberately because for a comprehension of corrosion dangers arising in practice and of appropriate protective measures this knowledge is necessary.

1 Definations: Corrosion, corrosion manifestation, corrosion damage, and corrosion protection Steel pipelines are used for the transportation of gases, water, mineral oil, long-distance heating water and chemical products as well as for the hydraulic transportation of solid materials. In most cases the pipe material is unalloyed or low alloy steel. There is practically no difference in the chemical properties of these materials. High alloy steels - so-called stainless steels - used for special applications only, have totally different chemical properties varying considerably from type to type. The weldability of line pipe steels allows the construction of almost homo-geneous and mechanically durable pipeline systems. In this presentation mainly the corrosion behaviour of unalloyed and low alloy steels is covered. Minor differences in the chemical compositions of these may be ignored; so for simplification, in the following chapters this group of steels will just be called „ steel“. Corrosion is understood to be reactions of the material in question with chemical constituents of its environment. The changes resulting from these reactions are manifestations of corrosion. In the case of steel in water and humid soil the corrosion manifestation is always the transformation of iron into corrosion products, mostly solid, called rust. The question, whether or not this results in damage, is answered exclusively by the extent of the process in relation to the required performance of the construction element.

Essential for the assessment of corrosion resistance is always the definition of the requirements to be fulfilled by the construction element in question, and, consequently, the maximum tolerable rate of corrosion. With steel pipelines this may be assumed to be a few 0.01 mm per year. In case, however, corrosion rates amount to some 0.1 mm per year, this may result in later damage, depending on wall thickness, or have no consequences if, for instance, corrosion rates drop in the course of time due to the formation of surface layers. In projecting pipelines this consuming corrosion is usually allowed for by means of extra, compensatory wall thickness. Very dangerous are conditions that lead to corrosion rates of some millimeters per year, because in most such cases perforation results within a few years. Corrosion protection is understood to be methods that, correctly applied, ensure corrosion to be mitigated to a rate not exceeding the tolerable maximum. There is neither a technical necessity nor the possibility in common practice to achieve zero corrosion rate. That this is not possible by methods applicable in practice can be proved; as opposed to this, technically negligible corrosion rates below 0.01 mm per year may well be attained.

The different methods of corrosion protection for underground pipelines work in different ways. Which of them In case the construction element does no longer perform should be applied mainly depends on the relevant its task or if it may stop functioning within its projected conditions that may stimulate or mitigate corrosion, service life, then there is damage. In the case of a these not always being obvious and generally difficult pipeline the word „damage“ means the pipe wall is to be assessed. There are even borderline cases where perforated or does no longer fulfill the service a protection method, unsuitably applied, accelerates requirements. Generally corrosion damage may be corrosion. This is why pipeline protection, as a rule, is a taken to have occurred, if the wall thickness falls unduly  job for an expert. short of the specified minimum. In this way of differentiating the various corrosion related concepts from one another, as put down in DIN 50 900, the fact is taken into account that all the metals in common use generally corrode more or less rapidly without this causing damage in every case.

2

Basics of underground corrosion and its conditions

2.1

Partial reactions in corrosion chemistry

Steel is made up predominantly of iron atoms; their chemical symbol is Fe. Like all other metals steel is characterized by an exceptionally high electric conductivity in the order of magnitude of 105 Ω -1 cm-1. Responsible for this is a high concentration of free electrons in the metal lattice. These electrons being considered separate particles, for a realistic description it is more appropriate not to use the symbol Fe for denoting steel, but instead: Fe <---> (Fe 2+  + 2 e-)

(1)

This manner of writing is used in chemistry to denote mesomeric conditions of substances. The meaning is that steel behaves like being in a state between two limiting ones which are signified by the opposite sides of formula (1). This means steel can react in both these „modifications“: (Fe), or (Fe 2+ + 2 e-). This fact is the reason for the electrochemical nature of the corrosion of metals in electrolytic solutions. Both components of the metal may react with the environment independently of each other: (Fe2+)steel ---> (Fe 2+) electrolyte solution

(2a)

(e-) steel ---> (e-)electrolyte solution

(2b)

-

Electrons (e ) as such are not soluble in water, but they may react directly with oxidizing components of the electrolytic solution: 4e- + O2 + 2 H2O ---> 4OH -

(3)

2e- + 2 H2O ---> 2 OH - + H2

(4)

Eqs. (2), (3), and (4) may be described in this way: Eq. (2a) signifies a transfer of electrically positively  charged particles from the material into the soil; the transfer is accompanied by a loss of bulk material. This process is an anodic electrochemical reaction  immediately resulting in a corrosive metal consumption. This type of corrosion is called electrolytical corrosion, or better: anodic corrosion. According to Faraday’s law its velocity w is equivalent to an electric current I A or, more correctly, to the current density J A  defined by the relation of the current to the surface area of the material: J A J A w   = 11.6 -2 = 1.16 mA cm A m-2 mm/year

(5)

Eq. (2b) denotes a transfer of electrons - i.e. negative charges - from the material into the soil. The electron transfer is not accompanied by a loss of bulk material. This process is a cathodic electrochemical reaction , and it is possible only if a chemical consecutive reaction like those in eqs. (3) and (4) can occur. Obviously, for the reaction in eq. (3) to take place the access of oxygen (O2) to the steel surface within the soil is necessary. Consequently corrosion rendered possible this way is called oxygen corrosion. The oxygen reacting with electrons according to eq. (3) may be replaced by other oxidizing agents. However, these are rarely present in the soil except in the case of nitrates from fertilizers or hydrogen ions from humic and/or carbonic acid. In the absence of oxygen, i.e. in anaerobic regions of the soil, the action of sulfate reducing bacteria may lead to the following reaction: 8 e- + SO42- + 4 H2 O ---> 8 OH - + S2-

(6)

Characteristic of this reaction is the occurrence of sulfides, e.g. iron sulfide FeS, recognized by the well known odour of hydrogen sulfide upon addition of hydrochloric acid. Compared to the reduction of oxygen according to eq. (3) all these alternative reactions are of minor importance. The reaction according to eq. (4) may be totally neglected in the case of normal soil corrosion. This reaction is very slow; the equivalent corrosion rate stays well below 0.01 mm per year. But in case this reaction proceeds under electrical constraint it is of special importance as will be pointed out in section 2.5.2.

2.2 Influencing factors and consecutive chemical reactions All electrochemical reactions in eqs. (2a, b) depend on chemical and electrical influencing factors in a typical manner. • Electrolytic corrosion: Anodic reaction ace. to eq. (2a). Stimulation by: High dissolving power of the electrolytic solution for Fe 2+-ions (low pH, large content of dissolved salts, e.g. chlorides and sulfates); clean surface of the steel (no deposits, reaction products, and/or coatings); shifting of the voltage between the steel and the soil towards more positive values. Mitigation by: Immediate precipitation of Fe(II) compounds constituting solid corrosion products on the steel surface (high pH, small content of dissolved salts); covered steel surface (coatings or deposits of reaction products); shifting of the voltage between the steel and the soil towards more negative values.

• Oxygen reduction: Cathodic reaction following eq. (2b). Stimulation by: Large oxygen content and easy access to the steel surface (aerated soil); steel surface totally without a cover or with a coating of relatively low electric resistivity; shifting of the voltage between the steel and the soil towards more negative values. Mitigation by: Low oxygen content, and restricted access to the steel surface (unaerated soil); coating with a relatively high electrical resistivity; shifting of the voltage between the steel and the soil towards more positive values. These influencing factors may be summarized as follows: The voltage between the steel and the soil, commonly  called the pipe-to-soil potential, determines the electrochemical reactions in alternative ways. A shift  of the potential towards more positive values stimulates anodic corrosion. A shift of the potential towards more negative values stimulates the cathodic reaction.

2.3 Interactions between anodic and cathodic partial reactions mutual dependence It is true that the electrochemical reactions according to eqs. (2a, b) may be discussed independently of each other, but there are interactions that must be kept in mind: • Mutual influence via chemical consecutive reactions (e.g. impediment of the anodic reaction by products of cathodic reactions); • Conservation of electrical charges or balance of currents. Corrosion being electrochemical in character, current balance is of paramount importance for its influencing factors and its consequences. To achieve a fundamental comprehension, it is helpful to first consider a totally homogeneous steel surface in the ground. In this connection the word „homogeneous“ is to signify that the electrochemical reactions on this surface proceed at uniform velocity independent of the particular site on this surface. The reaction velocities are represented by currents, the equivalence given by eg. (5):  Anodic partial current  I A, and cathodic partial current  I K. Both of these are dependent on the pipe-tosoil potential, but with opposite tendencies. This potential is understood to be the voltage between the steel pipeline and a reference electrode at the surface of the ground. Commonly used as such is a copper/copper sulfate electrode (see figure 1).

Cable Lid

Gasket with filling hole and plug Housing

Reaction products and coatings on the steel surface impede the cathodic reaction much less than the anodic  one. As shown by eqs. (3), (4), and (6), the cathodic  reactions result in a formation of OH 2  -ions, i.e. in an increase of the pH on the steel surface. This  phenomenon is sometimes called surface alkalinity. Surface alkalinity promotes the formation of solid  reaction products on the steel surface, and thus impedes the anodic reaction.

Copper

Saturated copper sulfate solution Ceramic-Diaphragm

Figure 1 Copper/coppe r sulfate electrode Refernce electrode for measuring pipe-to-soil potentials (Potential values obtained with the Cu/CuSO4 electrode may be converted to the standard hydrogen scale by adding 0.32 V)

The dependence of the partial reactions on the potential is schematically shown in fig. 2. Superposition of the two partial current vs. potential curves results in a total current  vs. potential curve I(U): I  A  (U) - I K  (U) = I(U)

   t   n   e   r   r   u   c   e   v    i    t    i   s   o    P    t   n   e   r   r   u   c

(7)

Risk of anodic corrosion Cathodic protection

Ia IA Ik

IK

Potential

UR

  e   v    i    t   a   g   e    N

Figure 2 Partial and total current potential curves valid for a homogenoues steel-to-soil electrode (schematized) Anodic partial current - potential curve Cathodic potential current - potential curve I A anodic partial current; IK Cathodic partial current Total current - potential curve I A anodic partial current; IK Cathodic partial current

A

Current source

V

Reference electrode Capillary probe

Electrolyte solution

Figure 3 Measuring circuitry for determining total current - potential curves by means of a current sourc e, a voltage (V) and a current (A) meter  (Also suitable for the determination of corrosion rate - potential curves; in this case a potential controlling rectifier serves as the current source)

This total current - potential curve is the electrical characteristic of the homogeneous steel-to-soil electrode, i.e. it can be established by direct measurement. The procedure is shown in fig. 3. In this special case the reference electrode works with a capillary probe, which is necessary for  tapping the potential immediately in front of the steel surface. If this probe is omitted, the measured potential difference contains a deviation corresponding to ohmic voltage in the ground, and proportional to the total current.

2.4 Corrosion with uniform metal consumption If the steel surface is not electrically influenced from without the total current equals zero. This is the situation of  free corrosion; the pertinent potential is the rest potential. Following eq. (7) in the case of free corrosion we have I  A (U R  ) = I K  (U R  )

(8)

In this case the corrosion rate is directly equivalent to the velocity of the cathodic reaction, i.e. of oxygen reduction. In unaerated soils free corrosion is negligibly slow.  A total cur rent different from zero may either increase or decrease the corrosion velocity, depending on its direction. A negative total current signifies cathodic polarization. Corrosion protection achieved this way is called cathodic   protection. A positive total current establishes anodic  corrosion danger and may be due to various external origins. The corrosion rate which is possible in such a situation - in some cases it is very large - is almost independent of the nature of the surrounding soil.

2.5 Formation of electrochemical cells

2.5.1 Anodic region; localized attack

In the case of a homogeneous electrode eq. (8) is valid for every point of the surface, so this relation also holds good for current densities:

At the anodic site the anodic reaction prevails. This means the total current density is positive:

(9)

J  A (U R  ) = J K  (U R  )

As a consequence of soil corrosion usually solid corrosion products are formed, which differently affect different electrochemical reactions. So, in the course of corrosion, a homogeneous electrode is transformed into a heterogeneous electrode composed of different regions in each of which one of the electrochemical reactions prevails. Figure 4 schematically shows such a heterogeneous electrode with a central anodic area and peripheral cathodic regions. The electrochemical reactions in these areas combined with their chemical consecutive reactions are such that the heterogeneous situation of the electrode is stabilized. This will be discussed in the following sections.

O2 - diffusion

  +

½O2 + H2O + 2e-

2       C        l        -   

Fe (OH)+ 2 OHCathode

Fe Steel

Soil

Na+

2e-

Fe2+ + 2 Cl - + H2O Fe (OH) Cl + HCl

Fe2+

O2

(10)

Such a region may develop due to the fact that here the steel surface is loosely covered or otherwise less aerated. A positive total current corresponds to an anodic cell current which, according to eq. (2a),causes anodic corrosion. In the ground it is propagated in the form of an ionic current, partly represented by cations moving into the soil, partly by anions moving towards the steel surface. This „anion migration“ enables the positively charged corrosion products (Fe 2+ ) to be electrically neutralized in the soil. Together with a consecutive hydrolysis, this may be described as follows: Fe2+ + 2 Cl - + H2O ---> (Fe(OH) + + Cl -) + (H+ + Cl -) (11) On the left hand side of the equation we have a correct balance of currents (corrosion current minus migration current), on the right hand side we see two electrically neutralized ion pairs dissolved in the water. In this the hydrochloric acid (H+Cl-) lowers the pH and thus keeps the steel surface clear of deposits of corrosion products. At some distance from the steel surface the corrosion product may be further oxidized by oxygen to form insoluble rust:

Migrations of ions

   a     N    2

J a = J A - J K > 0 (Anode)

FeOOH

Rust nodule

Cathode

Anode

Figure 4 Corrosion cell with anode and cathode on heterogeneous steel surface in the soil

Cell circuit (arrows only denoting migration of particle in question) Surface layer on the cathode (formation favoured by surface alkalinity NaOH); due to HCl formed by hydrolysis, the anode remains free of surface layer within the rust nodule

4 Fe(OH)Cl + O2  + 2 H2O ---> 4 FeOOH + 4 Hl

(12)

This process causes the well known rust nodules to develop above the anodic spots of the steel surface. They cover the anode and, thus, stabilize it purely mechanically.

2.5.2 Cathodlc region; formation of covering layers

2.5.3 Interaction between anodic and cathodic regions

In the cathodlc region the cathodic reaction prevails. So the total current density is negative: J k = J A - J k < 0 (cathode).

(13)

Such a region may develop, e.g., owing to the steel surface being amply aerated or covered by coatings or reaction products with only minor electrical resistance. A negative total current corresponds to a cathodic cell current, enabling cathodic reactions to take place according to eq. (2b). Within the soil it is propagated in the form of an ionic current, partly represented by OH- anions moving off, partly by cations moving towards the steel surface. The latter process again is a migration current that covers the electrical neutralization of the cathodic reaction products according to eqs. (3), (4), and (6), namely the OH- -ions, in thesoil. Immigration of alkaline earth ions into the cathodic region leads to the formation of solid cover-layers, immigration of alkali ions to the formation of caustic alkali: CO2 2 OH- + Ca2+  ---> Ca(OH) 2 ------> CaCO3,

(14a)

OH- + Na+ = NaOH

(14b)

Eqs. (14a, b) should be understood the same way as eq. (11), showing the current balance on the left hand side (cathodic reduction current minus migration current), and the electrically neutralized ion pairs (metal hydroxides) on the right hand side. In addition to this, in eq. (14a) the formation of solid calcium carbonate with carbonic acid from the soil is indicated. These reactions are substantially promoted by cathodic protection because the negative total currents enforced by it speed up the reactions according to eqs. (3) and (4) on all bare parts of the steel surface.

In the formation of electrochemical cells it is important that there is an increase in pH at the cathode (surface alkalinity, eq. (14)).This decreases the solubility of corrosion products, as can be seen immediately from the solubility product: c (Fe2+ ) x c 2 (OH-) = 5 x 10-16  mole3 1-3 

(15)

Because of hydrolysis the pH at the anode is definitely lower than 6, so that c (OH-) < 10-8 mole/1. This allows c (Fe2+ ) > 5 mole/1 = 280 g/1. At the cathode pH is definitely above 8, corresponding to c (OH-) > 10-6  mole/1. The result is c (Fe2+ ) < 5 x 10-4 mole/1. In most cases pH is around 9. The possible iron concentration is then c (Fe2+ ) = 0.3 mg/1. This explains that there can be no solid corrosion products at the anode within a corrosion nodule, whereas the cathodic regions must be closely covered. Within the latter the anodic reaction is more impeded than the cathodic one. This consideration shows how the heterogeneous state of  the corrosion cell is stabilized by consecutive chemical reactions in both the anodic and cathodic regions. Salts dissolved in the moisture of the soil participate in these consecutive reactions, sodium and chloride ions being particularly effective. Without these salts neither anodes nor cathodes can stabilize themselves. Another  consequence of the absence of salts is the limitation of  electrolytic conductivity; this means that for merely electrical reasons no effective cells can develop. The existence of electrical space charges not being possible, the production of Fe2+   and OH-  ions according to eqs. (2a), (2b) and (3), respectively, must take place within the immediate neighbourhood of each other. This, however, does not lead to localized corrosion attack below rust nodules but to the formation of homogeneous rust layers:

O2 Fe2+ + 2 OH ---> Fe(OH) 2 ---> FeOOH

(16)

The presence of dissolved salts being essential for  localized corrosion to proceed the symbols of sodium and chloride ions were incorporated in fig. 4. Their omission for the sake of simplification would be incorrect. If they were omitted, a wrong concept of the action of saltfree water would be signalized, and the essential effect of the ions would be missed. To simplify the scheme in fig. 4, J K  was taken to be zero in the anodic region, the same holds good for J  A  in the cathodic one. This simplification is indeed acceptable, because it makes the picture more distinct without changing its meaning.

2.5.4 Influence of anodic and cathodic area ratio

In the case of a corrosion cell, i.e. any heterogeneous steel surface, neither are the current densities of the electrochemical reactions J A and J K locally uniform, nor does the total current density equal zero with free corrosion. It is true that taken over the complete steel surface the integral of current densities equals zero. Then, considering a simplified corrosion cell with one anode and one cathode like in fig. 4 we find: J a  x Sa + J k x Sk = 0

(17)

Sa and Sk stand for the areas of the anodic and cathodic regions of the steel surface respectively. The total current density J a of the anodic region increases proportionally to the ratio Sk / Sa , provided the total cathodic current density remains constant. This may be assumed to be the case with corrosion in the soil, J A in eq. (13) being neglected, and J k  being determined only by a uniform access of oxygen. Then, what follows from eqs. (13) and (17) is the law of areas: Sk

J a  = J K x

(18)

Sa

The danger of anodic corrosion caused by this cell formation is obvious from fig. 2, with a positive total current I a = J a x S a. Only in case I K in the anodic region may be neglected compared to I A, the total current density J a may be taken to equal the corrosion current density J A.   If exterior electric currents are active, corresponding to eq. (7), eq. (17) may be extended to:  = J  x S  + J  x S  I  a

a

k

k

(19)

J a, and J k   stand for the total current densities corresponding to eqs. (10), and (13), respectively. An exterior current may not only influence J a and J k but also via consecutive chemical reactions alter the areas Sa and Sk, provided its duration is sufficient. Sufficiently strong positive or negative currents totally convert a heterogeneous steel surface into one anode or one cathode respectively. Following eq. (19) a negative current J k x Sk may make J a = 0. Then, corresponding to eq. (10), the anodic region is no longer threatened by cell formation but only by the cathodic reaction proceeding within this very region at the velocity J K. = J A. Even this kind of a cathodic protective action may be very helpful in preventing early failure.

This consideration concerning the interaction of chemical and electrical parameters is fundamental for a comprehension of corrosion processes. It enables a semiquantitative description of steel corrosion in the soil to be given with reference to the conditions set in practice by the type of soil and the construction or type of installation. It also gives an insight into how to apply protective measures.

3

Estimation of corrosion danger for pipelines not subject to electrical influence; fundamentals of protection methods •

3.1 Aggressiveness of the soil 3.1.1 Uniform oxygen corrosion  As indicated by eq. (9), on a homogeneous steel surface the corrosion velocity corresponds to the cathodic current density J K. On a heterogeneous steel surface the local corrosion rate within the anodic region is higher by a factor Sk(/Sa - the ratio of cathodic and anodic areas as is indicated by eq. (18). So, in any case for assessing the corrosion rate, it is essential to know the cathodic current density J K, or, correspondingly, the rate of oxygen access. An approximation may be derived from the law of diffusion. By combining eqs. (5) and (9), we get a relation between corrosion rate and oxygen diffusion: ω

mm/year

= 1.6 x 10-2

c (O2)

mg/l

:

l D

mm

(20)

In the case of air saturation c (O2 ) may be taken equal to 10 mg/l.The quantity l D denotes the effective path of diffusion. In the case of a rapidly flowing underground water, which cannot form a covering layer on the steel surface, l D = 0.2 mm may be assumed. Air saturation given, this corresponds to a corrosion velocity of 0.8 mm/year. In soil being at rest, l D is bigger by a factor of 10, at least, corresponding to a corrosion velocity of 0.08 mm/year. Covering layers of corrosion products further increase l D, so corrosion slows down with time. Except in cases of  flowing ground water the corrosion velocity on a homogeneous steel surface in the soil equals a few 0.01 mm/year. So, as a rule, it may be neglected. 3.1.2 Cell formation In the soil But other considerations (in section 2.5) showed also that dissolved salts which are present in different concentrations in all types of soil, cause any homogeneous steel surface to become a heterogeneous one in the course of  time. The velocity of this process grows with the salt content, i.e. the electrolytic conductivity of the soil, and, above all, with the extension of the steel surface. Especially extended, however, are pipelines; moreover they may traverse areas with different types of soil. Then anodes and cathodes develop in a typical manner, as Is described in section 2.2, with respect to influencing factors: •

anodic region: rich in salts, aerated little or not at all, dense, and wet. Example: Clay; especially aggressive are sour soils (humus) and soils containing hydrogen sulfide.

cathodic region: well aerated, moderate moisture and salt content, light and porous soil. Example: Soils containing lime; sandy soils are little effective in the case of low conductivity, but they too favour the development of cathodic regions.

Experiments made with steel sheets in different types of soil verified that corrosion is never totally uniform. Fig. 5 shows samples that freely corroded for 6 years, one in sand and one in clay. The mean rates of metal consumption were 0.01 mm/year in sand and 0.03 mm/  year in clay. Local maxima were ten times that. Extensive investigations made by the National Bureau of Standards tor the U.S. Department of Commerce showed rates of metal con-sumption ranging from 0.01 to 0.09 mm/year, depending on the aggressiveness of the soil; in weakly aggressive soils there was a distinct drop in consumption rates after a few years.

Sand (ρ = 4 x 106 Ω  cm)

Clay (ρ = 7 x 106 Ω  cm)

Figure 5 Surface of steel samples (100 x 150 x 3 mm) after 6 years` corrosion in sand and clay

3.1.3 Intensity of corrosion in different types of soil Fig. 6 gives a systematic survey of the different types of  soil. The rates of metal consumption in aggressive soils may range from 0.05 to 0.10 mm/year. J K is quite large; the reaction of humus acids contributes to this. In soils conditionally aggressive corrosion velocities are frequently found to be negligibly small. In both these types of soil, though, anodic regions develop on traversing pipelines. Cathodic regions develop only in soils which are not aggressive. Humus 100%

Lime 100%

i         e m     e       m     i      l      i     L Lime    s   o   s    s    u   o     u    r  e     l    o    r   c  a   a     m     l   a    m Calcareous     u    C      H marl

Humus 100%

L    

Peaty soil

   y     d     l    n    r    S  a    m  a S      a     n     d      Sand 100%

H      u   m     o     u   s     s     o     i      l     

C      a     l      c   a     m     r     e   o     a     r     l      u   s    

Sandy marl Loess Sand

Clayey marl Loamy marl Loam Loam

Clay

Sand

Peaty soil

C      l        y     m     a     e    a       y     r     l         y     a     l    C Clay 100%

Clay

Not aggressive (cathodic regions

Humus 100%

The presence of coke and foreign electric influences excluded, unprotected steel pipelines can definitely be used without the danger of early failure, i.e. for several years. This is in accordance with experience. At some later time though, depending on the wall thickness, damage must be expected. This means corrosion  protection measures are indispensable.

3.2 Methods of corrosion protection

Humous soil

Peaty soil

Approximate assumptions of corrosion velocities within the anodic regions of pipelines may range over a few 0.1 mm per year. Local metal consumption rates above 1 mm/year are definitely due to other reasons to be detailed in sections 5 and 6. Corrosion danger imposed by soil containing coal (or coke!) belongs to this group. The reason is cell formation between the steel surface and the coke, the latter constituting the cathode, the former being the anode. In this case there is but one suitable protective measure: replacement of the soil along  the pipeline.

Conditionally aggressive

(anodic regions)

Aggressive

Figure 6 Type system of soils and their corrosion aggressiveness

The development of extended corrosion cells on a pipeline passing through various types of soil does not allow an estimation of corrosion velocity. As opposed to this the very positions of the anodic regions may be located quite well by referring to the groups in fig. 6. Corrosion velocity will be all the greater, the smaller the anodic and the greater  the cathodic regions are, compare eq. (18). This holds true provided the electric conductivity of the ground is sufficient; if it is not, for instance in dry sandy soils, even large cathodic regions remain inactive with respect to corrosion cell action. As opposed to this, intensively active cathodic regions afford fast corrosion of the anodic regions, then the properties of the soil near the anode are less influential. So, evidently, there is no absolute measure of corrosiveness of a soil. A rating of  aggressiveness in terms of grades may be gained based on a soil analysis.

3.2.1 Coatings: „passive“ corrosion protection

The most simple measure of corrosion protection is coating the steel surface with bitumen or plastics. To be sure there is a lot of literature and specifications concerning the required properties of the coating materials, but the purpose of each of these requirements with relation to corrosion protection is not always obvious. Essential are these two properties: •

The coating material must be stable for a long time under the relevant service conditions in the ground.

•

The applied coating must be mechanically as resistant as possible to minimize frequency and extent of  mechanical damage.

PE-coating, fully meets these requirements. In cases of  very rough service conditions, e.g. laying of the pipes in rocky soil with pointed stones and in soils containing slag, additional rock protection should be given.

3.2.1.1 Corrosion chemical interactions on a coated steel surface

In spite of top quality coatings, careful insulation of the weld joints, repair of gross mechanical coating damage, and in spite of scrupulous supervision of the laying procedures, the occurrence of damage exposing the steel surface to the soil cannot be prevented. Within these coating defects the electrochemical processes of corrosion take place as described in fig. 7. They must be considered in connection with the properties of the coating material. • Corrosion of the steel surface within a coating   damage The reactions taking place are those according to eqs. (2a, b and 3), the anodic reaction predominantly in the middle, the cathodic one - producing NaOH predominantly at the rim of the holiday. The alkali solution generated there may creep some way under the coating and thus loosen it. This effect may be observed in waters rich in dissolved salts. It is harmless, though, provided the mechanical properties of the coating ensure the coating’s unchangingly close fit to the pipe surface, thus preventing an open cleft from developing. • Oxygen and Ion permeation through the coating Due to water absorption, organic coating materials may become weak electric conductors. Then they are not insulators like, for instance, porcelain. Apart from that, organic coating materials are not gastight like, for instance, a metal. Accordingly, permeation of alkali ions (Na+) and oxygen (O2) is possible. This results in the coated surface behaving like a weakly effective cathode. The accompanying anode is the steel surface within the holiday. Thus the formation of cells accelerates the corro-

sion taking place below coating holidays. This corrosion danger is directly proportional to the electrical conductivity of the coating (to be measured in holiday-free areas!). Another result is the formation of NaOH on the cathodic steel surface below the coating. This may decrease the coating’s adhesion and, with thin coatings, even lead to blister formation. There are no such cell effects with coating resistivities large enough, say 109 Ω  m 2. As a consequence, one of the requirements to be put to coating materials is the display of very high resistivity, as previously said, this being measured in holiday-free coatings. T he resistivity of PE-coating even in long time service remains far above this value. • Permeation of corrosive substances through and rust formation under the coating Besides oxygen also water and carbon dioxide may permeate the intact coating, and, below this, give rise to corrosion of the steel surface. Generally, though, the diffusion velocity is so low as to result only in a minute rate of metal consumption. So, in the beginning, there is no corrosion danger. But as soon as corrosion products loosen and fracture the coating, i.e. force it open, the protection is lost. Such processes are possible, for instance, with thin coating layers used for the protection of structures in open air. Rust formation below these coatings is prevented for several years by using protecting pigments in a special primer. Long time or permanent protection though, as to be specified for underground pipelines, is not attainable this way. With thick plastic coatings sitting snugly on the pipe surface - sticking to it or not - such processes are of no concern. Velocities of rust formation below such coatings are equivalent to metal consumption rates less than 0.001 mm/year.

Figure 7 Corrosion chemical interactions in connection with coatings Undercutting   by alkaline moisture; cell formation including coated cathode surface; penetration and rust formation underneath the coating

3.2.1.2 Required protective properties of coating   materials

3.2.2 Cathodic protection: „active“ corrosion  protection

Following these lines the requirements to be fulfilled by a good pipeline coating may be summarized as follows:

3.2.2.1 Potential dependence of corrosion

•

Long time stability of the coating material under service conditions,

•

Good durability against mechanical influences and

•

High electric coating resistivity.

Passive corrosion protection given to pipelines by such coatings is sufficient, provided there are no external electrical influences, and the soil is „not aggressive“ according to fig. 6, because - within the few spots of coating damage - the steel surface corrodes almost homogeneously, and long line corrosion cells cannot develop. This statement is fully corroborated by practical experience provided they are embedded in sand and electrically disconnected from other parts of construction by insulating joints. Because of the good electric conductance of welds, with long pipelines in arbitrary soils cell formation between different bare parts of the steel surface in coating holidays may occur, even if these parts are separated by considerable distances. With a pipeline cover having a low coating resistivity - measured within an undamaged area of the coating, as already emphasized cathodic action of the coated area according to fig. 7, cannot be excluded either. It has to be kept in mind that the large surface of a pipeline even with a coating resistivity of 10 5 Ω m 2  brings about an electrical resistance in the cell current circuit which is quite poor. Anodic corrosion velocity depends on the extent of damage in the different soils and on the types of soil. It may only be assessed very roughly. Corrosion velocities of some 0.1 mm/year are possible. For this reason for all pipelines subject to special safety specifications an additional active protection, i.e. cathodic protection, is prescribed. In general this kind of protection is recommended for merely economic reasons, in spite of the fact that the probability of damage is much smaller with passively protected pipelines, compared to the situation with uncoated lines.

The basis of cathodic protection - „CP“ - was already given in fig. 2. By means of a cathodic current I k, the potential has to be shifted to more negative values, and that as far as to make the anodic partial current / A negligibly small. So a quantitative assessment necessitates the knowledge of the anodic partial current density - potential characteristic J A vs. U . Such relations may also be obtained by means of a measuring apparatus according to fig. 3, provided there is a controlling current source maintaining a constant potential for the duration of the test. By measuring the weight loss of the sample, the corrosion rate - potential characteristic ω (U), and, according to eq. (5), the partial current - potential characteristic J A vs. U , too, may be determined. Fig. 8 shows such results for saline electrolyte solutions. There it is obvious that with a potential U Cu/CuSO4 = -0.85 V in neutral waters the corrosion rate is less than 0.01 mm/  year. This is the cathodic protection potential, well known from practice. In waters containing much carbonic acid, that may flow from mineral springs but are not present in normal soil, the protection potential is slightly more negative. 10

3    )   r   a   e   y    /   m   m    (   e    t   a   r   n   o    i   s   o   r   r   o    C

1

0.3 0.1

0.03 0.01

0.003 -0.90

-0.85

-0.80

-0.75

-0.70

-0.65

-0.60

Potential (UCu/CuSO4, V) Figure 8 Corrosion rate/potential curve valid for steel in saline electrolyte solutions Saturated with oxygen Free of oxygen Free of oxygen, saturated with carbon dioxide Region of free corrosion (at more negative potentials the current is negative, at more positive potentials it is positive)

3.2.2.2 Application of cathodlc protection; verification of protection potential v = 50 cm/s

The application of CP in practice is achieved the same way principally as represented by the measurement arrangement in fig. 3. The proportions, however, are totally different. Generally it is not possible to use capillary probes of reference electrodes within the ground. The reference electrode according to fig. 1, is put onto the surface of the ground. The potential determined this way with the current flowing is called on-potential. By the Ohmic potential drop (IR) within the ground it is more negative than the true potential, i.e. the  potential free from IR, that would be measured with a capillary probe. In case the protection current is switched off for a short duration, the potential drop within the ground caused by this current collapses immediately. The change of potential due to chemical changes proceeds much slower, so the off-potential measured immediately after switching off the protection current may be taken as a very good approximation to the IR -free potential. There might still be errors due to Ohmic drops along stray currents from extraneous direct current systems or due to relaxation or cell currents between differently polarized regions of the pipeline including electrically connected foreign installations, see sections 5, and 6. Fig. 9 shows potential - time graphs obtained with different paper speeds. A registration effected too slowly may cause the off-potential to be found more positive than the true value. In general the potential change after the switching off essentially depends on the object being protected. As is shown in fig. 10, a high quality coating of a new pipeline as well as a long duration of CP will slow down the decay of the potential.

   )   v    (   e   m    i    t  ,    l   e   v   a   r    t   r   e   p   a    P

0

v = 250 cm/s

v = 1 cm/s v = 1 cm/min

IR

Ohmic voltage drop “off“

“on“

-0.2

-0.4

-0.6

-0.8

-1.0

-1.2

-1.4 V

Potential (UCU/SO4, V) Figure 9 Registration of the pipe-to-soil potential after switch-off  protection current (v = Recorder paper travel speed)

On-potential    ) -1.2    V  ,    4

Off-potential

   O    S    /    U    C

   U    ( -1.0    l   a    i    t   n   e    t   o-0.8   p    l    i   o   s     o    t     e-0.6   p    i    P

New pipeline (2 years`cathodic protection)

Old pipeline (10 years`cathodic protection)

Old pipeline (3 years`cathodic protection)

-0.4 0 20 40 Time, starting with current switch-off, h

60

Figure 10 Decay of pipe-to-soil potentials of different pipelines after switch-off protection current

3.2.3 Active protection + passive protection = complete protection Coating of pipe/coating of weld joints km

Coating and cathodic protection, when combined, make up a complete protection of a pipeline. An additional profit afforded by the method is the possibility of continuous supervision by monitoring the potential. In connection with this combined protection method the interaction between a coating and CP is sometimes also considered. This interaction may give rise to further requirements put to the coating. Cathodic currents accelerate the processes of blistering and of cathodic disbonding shown in fig. 7. Especially with a thin coating, a gaping cleft between the steel surface and the loosened coating may develop. An impairment of protection is the consequence, much less critical though with underground pipelines than, e.g., with ships and hydraulic structures. Thick pipeline coatings are not subject to cathodic blistering, comp. sec. 3.2.1.1. As opposed to this, cathodic disbonding starting from coating damage is inevitable. Because of the close fit of thick coatings, the lack of adhesion within the immediate neighbourhood of a spot of damage does not result in a gaping cleft. This means the exposed part of the steel surface does not grow through cathodic disbonding. The protection current  demand does not increase. Fig. 11 shows average coating resistivities of underground pipelines coated with bitumen and with PE. The slow decrease of r u with the bitumen coatings is due to the formation of very small pores in the bitumen during service, slightly enhancing its electric conductivity. Lines of PE coated pipes the  joints of which were field coated with bitumen show, less distinctly, the same effect, fig. 11.

3 x 105

   )    2   m

PE/PE 30 PE/PE 40 PE/PE 270 PE/PE 350 PE/B 150 PE/B 600

105

  m    h    O    (   r  u

3 x 104 B/B 300

B/B 450 104 1970

1975

1980

Figure 11 Coating resistivities of differently coated pipelines vs. time in service B = Bitumen; PE = Polyethylene

The effective coating resistivity of a pipeline is determined exclusively by number and sizes of holidays, inevitable in practice. So essentially the mechanical properties of a coating are of influence in this respect, not so the electrical properties of the coating material. The coating resistivity and the protection current density of a buried pipeline are related as follows:

J s

x -2

µA m

r u

= 3 x 105 2

Ωm

(21)

4

Cathodic protection of pipelines

The application of CP to pipelines is amply described in specific literature • Handbook of Cathodic Protection • Pocket Book of Cathodic Protection

In this section only the main features of this method of corrosion protection can be discussed which are of interest to those running pipelines.

4.1 Constructional requirements put by cathodic protection For CP to be applicable to a pipeline, some conditions must be fulfilled. The requirements are: • Continuous lengthwise electrical conductivity of the pipeline and its branches to be included into CP 

This means, pipelines with flange and socket joints having poor lengthwise conductivity cannot be cathodically protected without each joint being electrically bridged; this is quite costly. • Electrical separation of the pipeline from all units with low resistance groundings, and from extraneous installations

In other words: The ends of the pipeline and of its branches included into CP must be electrically separated from continuing lines by means of insulating flanges. In special cases, e.g. compulsory grounding for superordinate reasons (see sec. 7), this requirement may be given up. In case constructional reasons prohibit electrical disconnection, a special kind of CP, to be described in sec. 5, may be considered. •  A coating resistivity exceeding 104 Ω  m2

This refers to the apparent coating resistivity, described in sec. 3.2.3, and determined by size and number of holidays in the coating. According to eq. (21), the protection current density is inversely proportional to the coating resistivity. With PE coated pipelines, protection current densities of 1 µA m -2 may be achieved, see fig. 11. The smallness of protection current density is advantageous as it extends the range of protection and improves the distribution of current,

and decreases the overall current demand and thus reduces any interference with extraneous installations.

In cases of too great protection current densities only enlarged numbers of anodes - quite expensive - help to achieve good current distributions and to avoid detrimental influences on foreign installations. With pipelines covered with PE coating, protection ranges far exceeding 50 km are easily achieved, and interference with extraneous equipment is virtually excluded. •  A suitable site for the anode to be installed for feeding  in the protection current, and a power supply (public  utility)

The ground containing the anode should have the lowest possible resistivity so as to decrease the required rectifier voltage. Where public power supply is not available, the application of other sources of energy (e.g. solar, wind, gas) for the generation of current may be considered. Usually in these rare cases galvanic anodes rather than rectifiers will be the protection current sources.

4.2 Cathodic protection by using galvanic   anodes Galvanic anodes are applied for CP in those cases in which protection by current drawn from an external power supply is not possible or less economic. Galvanic anodes consist of light metal alloys. They are buried in a special bedding rich in electrolyte, called backfill, and their potential is substantially more negative than that of steel. This potential, negative compared to that of the pipeline, generates the protection current or, in other words, shifts the pipe-to-soil potential to the desired, more negative value. Because with galvanic anodes applied there is no way of controlling the current - except for decreasing it by Insertion of a resistor - the desired current distribution may only be achieved by a suitable distribution and sizing of the anodes. This layout must be given special care. For underground protection magnesium anodes may be used, in soils with very low resistivities also zinc or aluminium anodes. The maximum possible current delivered by a galvanic anode depends on the size of the anode and on soil resistivity. Its order of magnitude is 0.1 A. This is sufficient for the CP of not too large pipelines having a moderate protection current demand, for instance, pipelines covered with PE.

4.3 Cathodic protection by impressed current With large pipelines and also for reasons of easy control mainly externally powered protection facilities are installed. Fig. 12 shows such an installation consisting of embedded anodes and a current supply unit. Generally the anodes consist of ferrosilicon castings or magnetite embedded in coke. The electronic conductivity of the coke helps to cut down the anodic consumption of anode material. The main advantage of coke bedding an anode is a substantial reduction of its electrical resis-

Figure 12 Impressed current protection installation with anode bedding, protection current source, and connections to pipeline

tance in the soil. For economic reasons horizontal anodes - shown in fig. 12 - are preferred. Where there is not enough space for this type of anode and where foreign installations might be influenced due to an excessive field strength near the anode, vertical anodes may be burled 10 to 100 m deep (deep seated anodes).

Anode lead

Cathode lead

Measuring cable leading to the pipeline and the built-in reference electrode( helps to circumvent great IR -drops near the anode in the operation of potential controlling rectifiers)

Fig. 13 however, shows that the potential already did shift substantially from the starting point of free corrosion, thus stopping any corrosion cell action. When checked after about a year, the potential is supposed to have reached or gone beyond the protection potential. In case this cannot be achieved, even by increasing the protection current, the reason for this situation must be found. Aside from large damage of the coating, galvanic connections to extraneous installations may be mainly responsible by drawing off an unexpectedly large part of the protection current. Such connections may be discovered by electrical measurements and eliminated by constructional provisions. With pipes going through casings, crossing over other utility lines, and entering buildings, there is always the danger of galvanic contacts ensuing. Obviously this danger is especially imminent in urban areas.

4.4 Current drain test and adjustment For adjusting the protection current a current drain test is made. At the ends of the protection length and at other accessible places, but above all, at armatures, branch-offs, in the regions of casings and aggressive soils, on- and off- potentials are measured. Fig. 13 shows, as an example, the result of a current drain test made at a pipeline having a large protection current demand. Only then has the protection current been adjusted correctly if everywhere the on-potential reaches U Cu/CuSO4 < = 1 V. In fig. 13 this has been accomplished only for kilometers 5 to 9. In most cases an off-potential of U Cu/CuSO4 < = -0.85 V cannot be achieved during a current drain test. Only this off-potential indicates complete cathodic protection.

   4    O    S   u    C    /   u    C

   U    l   a    i    t   n   e    t   o   p    l    i   o   s     o    t     e   p    i    P

-1.5

-1.0

-0.5 0.3 A

1.03 A

9.6 A

1.17 A

0.36 A

0 0

1

2

3

4

5

6

7

8

9

10

11

12

13

Pipeline length (km)

Figure 13 Potentials measured in a current feed test with a bitumen coated pipeline of nomina l diameter 200 mm, after 4 h polarization

Pipe-to-soil potential under free corrosion conditions On-potential

Off-potential

Locations of current measurement with results; U s = Protection potential Location of feed-in and current value

4.5 Side effects of the protection current 4.5.1 Interference with extraneous equipment

It must always be kept in mind that an underground cathodic protection current may jeopardize extraneous installations which, from it, may suffer stray current corrosion. This is illustrated in fig. 14. Symbolically depicted is the potential and current field between an anode and a pipeline to be protected, the latter crossing another pipeline. Near the anode the extraneous pipeline receives protection current, i.e. here it is polarized cathodically. Close to the crossover, however, the current leaves this pipeline for the soil in order to reach the nearby pipeline to be protected. In the region of current exit the unprotected line is anodically polarized, and so, corresponding to fig. 2, there is also the danger of anodic

Today cathodic protection of all kinds of pipelines is a proven technique raising no unusual problems. CP may also be given to distribution systems in urban areas, provided the pipes are coated well; however, here it entails the heavy expenditure of numerous insulating  joints becoming necessary in the house connections. 4.5.2 Internal Interference within water pipelines

Naturally the goods conveyed within the pipeline are of no concern to external protection, provided they are not electrolyte solutions. But, if this is the case, special precautions have to be taken in the regions of insulating  joints. Protection current may enter an unprotected part of a pipeline situated beyond an insulating joint in the ground. From the internal pipe surface the current passes over

Current leaving the unprotected line to the protected line Figure 14 Influence of the protection current of a pipeline on a crossing unprotected pipeline

corrosion. In the situation depicted here, the extraneous pipeline may only be protected from anodic danger by being included into cathodic protection across a resistor (potential connection) . The conductance of the connection must be at least sufficient to cause some protection current to enter the foreign line in the crossover area; then for sure no current will leave. The easiest way to solve problems of interference is to fully include the foreign line in question into CP; this, though, may raise private law problems with the owners of the line. Whoever runs a cathodic protection unit is, in the legal sense, a producer of stray current and bound to observe the national standards and regulations. In practice, problems of interference should be kept to a minimum by the anodes being suitably placed and by a high coating resistivity of the pipeline.

Anodic region

Cathodic region

into the electrolyte within which it flows into the protected pipeline section. So danger arises for the internal surface of the unprotected pipeline section beyond the insulating  joint. With water pipelines an electrically insulating lining about half the length of a pipe on both sides of the insulating joint is sufficient for corrosion protection. In the case of a cement mortar lining there is no danger. With pipelines carrying salt waters, internal cathodic protection beside the insulating joint also proved useful. Since in these cases the danger is that of anodic corrosion, only the conductivity of the electrolyte but not its corrosion aggressiveness is of concern. In other words, with noncorrosive oxygen-free brines and waters used for long distance heating being transported, the danger of anodic corrosion behind the insulating joint is much more severe than the danger of free corrosion in aggressive waters.

5

Corrosion danger caused by foreign cathodes; localized cathodic protection

5.1 External currents In section 3.2.1 for an uninfluenced pipeline the corrosion danger was assessed, that may be countered by means of a coating and cathodic protection. An external influence is given, if current enters the pipeline from extraneous installations across a metallic contact, or if, caused by electric fields within the ground, stray currents enter and leave certain parts of the pipeline respectively, as was described in fig. 14. Stray current will be dealt with in chapter 6. Chapter 5 is concerned with those currents exclusively that enter the pipeline at places of  metallic contact. Normal cathodic protection is an example of this. The effect of the negative current fed in is the prevention of corrosion damage. This was pointed out in the beginning with relation to fig. 2. Conversely, there is a definite corrosion danger, if the current fed in is positive. So obviously great care has to be taken to make it  just  impossible for the rectifier of an external current   protection unit to be reversed.

5.2 Corrosion caused by foreign cathodes Positive current enters the pipeline at a place of metallic contact whenever the contacted extraneous installation has a rest potential more positive than that of the pipeline. However, some 10 mV need not yet be considered critical, a potential difference of 0.3 V is about the margin. Potential deviations of about 0.1 V are quite normal with pipelines and due to different properties of soil and coating. They signalize the formation of corrosion cells along the pipeline as described in section 3.1.2, compare also the rest potentials in fig. 13. 5.2.1 Characteristic of a foreign cathode

The rest potential of steel in the ground ranges between U Cu/CuSO4  = -0.7 and -0.5 V. A potential of -0.4 V and more positive values in the neighbourhood of a foreign contact raises the suspicion of corrosion danger caused by too positive a potential of the extraneous object. In this case we have a corrosion cell, the pipeline as a whole representing the anode and the extraneous object being the cathode.

Danger is evidenced by potentials around -0.2 V and above being found near the extraneous object. Indicative are measurements made after breaking the contact, if this is possible. The separation having been effected, the electrical potentials of the disconnected parts must drift in opposite directions: The pipeline will become more negative and the foreign cathode more positive. By use of a low resistance ammeter even the cell current may be measured directly. By means of such a potential measurement also the operator of a pipeline may check whether it is anodically endangered by foreign cathodes or not. Required is a Cu/CuSO4 reference electrode, easily built by laboratory personnel referring to fig. 1, and a high resistance voltage meter, the min. input resistance of which is1 M Ω /V. For a potential measurement to be made, the reference electrode has to be put on moist soil. A potential measurement on an asphalt road surface is impossible! In case there is an indication of corrosion danger, experts should assess its extent and plan protective measures. In connection with this, it must be kept in mind that in the case of an external cathode raising danger, the electrical influence on corrosion strongly prevails and that the properties of the soil are almost of no concern, provided only, a cell current can flow. This means, only in sandy soils with very high resistivities is the danger of corrosion diminished. 5.2.2 Examples of foreign cathodes

Although it is easy to realize that great potential differences mean anodic danger, it is comparatively difficult for the non-expert, when judging from the material, to see which parts of construction might act as foreign cathodes. In the case of a pipeline connected to other installations, e.g. groundings, mainly those parts that consist of nobler materials like, for instance, copper, copper alloys, and stainless steel may be suspected to act as foreign cathodes. In many instances the surface fraction of these materials in the ground is so small that, in accordance with the law of areas, eq. (18), the corrosion danger may be neglected. However, the potential is dependent not only on the kind of metallic material but also on the chemical properties of the environment. In the ground there may be large areas of steel surface covered by concrete or cement mortar. Now, the chemical properties of cement mortar are vastly different from those of soils. Covered by cement mortar, steel is in a special condition,

called passivity. Within cement mortar it behaves like stainless steel surrounded by water. So, foreign cathodes may be represented by all parts of copper, stainless steel and steel covered by cement mortar. Apart from this, old rusted steel surfaces of pipes, storage tanks, and other constructional parts may act as foreign cathodes. Today, steel parts in concrete structures acting as foreign cathodes are the most frequent origins of anodic danger. Possible examples are fixing points and manholes near armatures, points of entry into buildings, connections to building foundations, and to groundings. In this line fig. 15 schematically shows the course of the pipe-tosoil potential near the pipeline contacting a reinforced concrete structure. The funnel-shaped potential distribution above the coating holiday only exists with respect to the IR -free potential which, theoretically, might be determined by using probes. The pipe-to-soil potential as measured on the surface of the ground is much more even due to the ohmic voltage drop of the current leaving the steel surface through the holiday. Were the pipeline disconnected from the rebars in the building, the pipeto-soil potential would assume a value U Cu/CuSO4 < -0.6 V belonging to the steel surface below the holiday.

Reinforced concrete structure

SOIL

   )    V    (    4   o    S   u    C    /   u    C

   U    l   a    i    t   n   e    t   o    P

-0.2 V

-0.4 V

-0.6 V Length

Figure 15 Result of a contact between a pipeline and a reinforced concrete structure

Pipe-to-soil potential mesured with reference electrode on the ground surface Pipe-to-soil potential, free of IR -drops (as if measured with the probe placed on the pipe surface)

5.2.3 Estimation of corrosion danger 

Equation (17), may be used for an assessment of corrosion danger, the indices a and k referring to the pipeline and the foreign cathode respectively. Steel in concrete being passive, eq. (18), may be taken to be valid; and, in the case of large corrosion rates, J a = J A  may be assumed. So it is the area proportion which is decisive, and with it the magnitude of Sa, the total area of the steel surface lying bare in an area of coating damage. The better the coating, i.e. the less mechanical damage occurred along the pipeline, or the higher the coating resistivity is, the larger will be the anodic current density according to eq. (18), or, according to eq. (5), the faster will be the corrosion at places where damage did occur. In such cases corrosion rates may exceed 1 mm/year, and, in saline waters, even reach 1 cm/year. It is true that the corrosion rate will be lower with a pipeline having a badly damaged coating with many holidays or even having no coating at all. So lack of passive protection might help to avoid early failure, but not damage occurring later.

5.2.4 Coating influence

Certainly, doing without a coating is not a protection measure to be recommended, just in order to delay failure. But the user must be aware that, danger by foreign contacts being given, the choice of a top quality coating  cannot be considered a protective measure as long as there is no 100% certainty of the coating being free of  any minute holiday  remaining that during service. This seemingly paradoxical  effect of the extent of coating damage in the case of an external electrical influence is generally valid; it is not limited to foreign cathodes generating danger. The effect is easily understood in view of these relations: •

The electrical resistance in the ground in front of a coating damage (contact resistance, grounding resistance) is inversely proportional to the mean radius of the uncovered steel surface below the damage:

R F = A r -1 •

(22)

The size of this uncovered steel surface is proportional to the square of the same mean radius:

SF = B r 2

(23)

With a voltage U  acting upon the pipeline around the damages, Ohm`s law gives:

I a  =

J a  =

U  R F I a SF

=

=

Ur 

(24)

A U 

AB

r-1

(25)

So the current density J a will increase, if the radius r  decreases!

5.3 Protection measures Protection against danger caused by foreign cathodes may be accomplished in two ways: • Galvanic separation • Local cathodic protection 5.3.1 Galvanic separation Galvanic separation is the most effective means because by it the very source of corrosion cell danger is eliminated.

The reliability of this depends on easy control of all the sites of possible connections, and on the certainty with which accidental contacts may be excluded in practice. Standards and regulations specify galvanic separation to be effected for house connections in public gas delivery systems, because for inside the building, intentional connections for personal protection are made compulsory by DIN/VDE-regulations. With reinforced concrete construction and interconnection of all domestic installations being common today, personal protection observing DIN/VDE results in the formation of spacious foreign cathodes with respect to gas pipelines, that have to be disconnected from these by inserting insulating flanges. 5.3.2 Local cathodic protection, hot spot protection

In ranges of industrial plant, galvanic separation is inapplicable because of the presence of too many interconnections, in many instances also for constructional reasons. Applicable then is local cathodic protection. The techniques of normal CP cannot be followed for these reasons: • The foreign cathodes will consume much more current than the pipeline to be protected. Large ohmic drops in the ground will result. So IR   will become a substantial part of the on-potential. • The pipeline and the foreign cathodes vastly differ with respect to their electrochemical properties. Free corrosion endangers the pipeline anodically. Given cathodic polarization, the two areas are polarized differently. After switching off the protective current for the off-potential to be measured, comparatively large relaxation currents between the pipeline and the foreign cathode will flow. The corresponding ohmic drop will falsify the off-potential too. The measured value is more positive than the IR -free (true) value. The basis of local cathodic protection may be derived from eq. (19). The protection current is rated and fed in so that the foreign cathodes do not deliver cell current but consume protection current. Then eq. (19) gives: I 

> =

J k Sk and J a < = 0

(26)

In other words; Aimed at is the polarization of the foreign cathodes so as to render them harmless instead of breaking their corrosion cell with the pipeline. Within industrial plant many 10 4 m2 of reinforced concrete surface

contact the ground and consume a mean current of about 5 mA m-2, so the order of magnitude of the current consumed for local cathodic protection of a special plant is 102  A. The interference problems raised by such currents can only be solved by all underground installations being incorporated into the protection system. Fig. 16 shows the application of local cathodic protection to a power station with large cooling water ducts (nominal diameters 2000 and 2500 mm) and water conduits for fire extinguishing. For 19 000 m 2 of reinforced concrete foundations and 2000 m 2 of copper groundings a current of 120 A is fed in via eight deep-groundbeds With a resistivity of 150 to 350 Ω m this soil is a comparatively poor conductor; so, for a local potential lowering of specific objects to be protected, a „directed“ current may be fed in via horizontal groundbeds. The values given under  „A“are pipe-to-soil potentials measured with free corrosion going on before local cathodic protection was installed. These potentials range from U Cu/CUSO4  =  -0.5 to -0.1 V. The values given under  „B“,

Figure 16 Local cathodic protection in a power station

Deep groundbrds Horizontal anodes Cooling water lines Fire extinguishing lines

and „C “are on-potentials (off-potentials) measured after 4 months and one year, respectively, of the protection system being in operation. As expected, it was not possible to produce off-potentials more negative than the critical protection potential (-0.85 V), this is in spite of (some) vastly negative on-potentials. However, due to the IR -bound error still present in the off-potential measurements, this impossibility does not imply that the object is not fully cathodically protected. There are ways of measuring that work with polarization probes and/or test coupons, and enable off-potentials at critical points of the structure to be measured sufficiently depleted of IR - errors. This affords a reliable ascertainment of protection. Complicated adjustment steps during the design and construction phases, and later on, controlling measurements and the handling of very large protection currents in view of possible interference with foreign objects, necessitate this type of protection to be managed by expert enterprises having advanced experience. During the last decade progress was manifested in numerous successful applications.

A B C

Pipe-to-soil potential under free corrosion conditions Pipe-to-soil potential after 4 months` polarization Without brackets: On-potential, in brackets: Off-potential Same as B, but after one years` polarization

6

Corrosion caused by stray currents; protection methods

6.1 Origins of stray currents

6.2 Characteristics of stray current action

A stray current is an underground current originating from current-carrying parts of electrical installations. In most cases these are the rails of electric railways. But CP facilities installed for foreign objects to be protected may also be responsible. Related interference problems were discussed already with regard to fig. 14. Depending on the geometric situation of the pipeline in relation to the direction of the current, i.e. the electric field generated by the unit in question, stray current will enter the pipeline over large areas. Within the most positive region the stray current leaves the pipeline for the ground, and here it causes anodic corrosion.

Corrosion caused by stray currents is in many respects similar to that caused by foreign cathodes. However, there are typical differences, too. •  Corrosion danger essentially depends on current density only. Properties of the ground are of no concern, actually, provided a stray current does flow. So only in sandy soils of very high resistivity is the corrosion danger diminished. • The influence of the pipe coating is varied.

If there is no direct connection leading in stray current from foreign objects to the pipeline, a high coating resistivity could be useful, because stray current entry would be cut down. On the other hand in the region of current exit the same conditions are given, as were discussed in section 5.2.4 with respect to foreign cathodes, in view of eqs. (22) through (25), page 24. Hence in this area a high coating resistivity is unfavourable. Useful would be only a coating 100% free of holidays, and that after the laying of the line and during service. Figure 17a Stray currents originating from railways

• As opposed to the situation with foreign cathodes, the origin of danger cannot be eliminated. Stray currents are present in the ground, and they do not need connections to enter the pipeline. A suitable protection measure is not impediment of stray current entry but that of stray current exit. • As in the case of danger caused by foreign cathodes, the corrosion rate is hardly assessable. Values exceeding 1 mm/year may be expected. The endangered regions are characterized by very positive pipe-to-soil potentials. Contrary to the action of foreign cathodes and foreign protection facilities the potential accompanying railway currents is intensely variable with time. Depending on construction and usage of the railway track the regions of entry and exit of stray current are subject to considerable alterations. For these reasons simple measurements of potentials are not sufficient; necessary is recording the potentials for several hours’ duration.

The same way as with foreign cathodes creating danger, the user of the pipeline may, by potential measurements, recognize stray currents to be present. These measurements must by all means be continued for several hours while the neighbouring railway tracks are being used. Assessment of the danger can be achieved by an expert only. The intense danger of early damage necessitates protection measures as soon as possible.

6.3 Protection measures against stray currents Protection may be achieved by leading the stray current through electrical connections back directly to the installation that produces it. Fig. 17 shows schematically stray currents being created by a D.C. railway and flowing through a neighbouring pipeline. Near the rectifier railway feeding station of the railway the stray current leaves the pipeline and causes severe anodic corrosion in this area. This is the very location for protection measures to be taken.

A B

6.3.2 Rectified drainage This is a special case of immediate stray current drainage. It is characterized by heavy duty diodes incorporated in the draining connections to the rails. The diodes are meant to ensure that current can only flow from the pipeline to the rails and not in the opposite direction. This type of draining connection must be applied, if a direct drainage cannot be installed in the immediate neighbourhood of the rectifier railway feeding station. Increasing with the distance from the railway rectifier is the chance of pole changings taking place between pipeline and rails according to varying situations in railway traffic, so that there is the danger of stray current entering through the very drainage. This is to be prevented by diodes. Rectified drainage must be installed in several locations along the pipeline, where stray current may often leave. 6.3.3 Forced drainage With this type of drainage a protective rectifier unit is incorporated into the connection. By means of its voltage the rectifier extracts the stray current from the pipeline and directs it to the rails. Actually this is a normal arrangement for cathodic protection, the rails representing the anode. Since due to varying situations in railway traffic the voltage pipeline/rail may be subject to substantial alterations, the protective rectifier unit should be adjustable to voltage variations with time. There are several possibilities to lower the resulting current variations, e.g. installation of inductivities.

Figure 17 Protection against stray currents originating fro railway stray currents in the soil, pipeline and current drain

Anodic region

of influenced unprotected pipeline

Cathodic region

6.3.1 Immediate drainage Via a heavy duty low ohmic resistor the pipeline is directly connected to the stray current source, i.e. the railway track.Even this measure alone is very effective, provided the connection is made in the immediate neighbourhood of the rectifier railway feeding station of the railway, and the overhead conductor is the positive pole. Thus the negative potential of the rails is transferred to the pipeline, so that even a moderate cathodic protection may be the result.

6.3.4 Use of controlling rectifiers for forced drainage In the case of very large current variations it is advantageous to use rectifiers controlling the potential. By means of the fixed reference electrode RE the rectified current is controlled so as to keep the pipe-to-soil potential in the place of drainage at a fixed value U nom. At larger distances from the reference electrode the stray current influence is different, so there are increases in the temporal variations of the pipe-to-soil potential again, but this is of no concern to the protection effectiveness. If in some region the pipe-to-soil potential is often found to be too far positive, another protection unit must be installed there.

7

High voltage interference, protective measures and effect to cathodic protection

7.1 Types and causes of high voltage   influence In densely populated areas there is not always a way around constructing pipelines running in close proximity to open high voltage lines, crossing them or even parallelling them for some distance. Due to alternating voltages induced in these pipelines high contact voltages may develop, which necessitates protective measures. During construction and operation of pipelines special measures must be taken according to relevant standards and regulations, also with regard to the cathodic protection system. There are two main types of interference which must be assessed in different ways:

7.2.2 Permanent interference caused by operating currents of the high voltage power line

The maximum permissible voltage for long time bodily contact is < 65 V. This means the required groundings may be connected to the pipeline across AC-voltage limiter IVL-10. The Limiter IVL-10 effectively blocks the protective DC-current required for cathodic protection while providing a low ohmic ground connection for induced AC-currents caused by operating or short circuit currents of high voltage overhead line systems. Measuring of true potentials by application of the offpotential technique are guaranteed.

• Short time interference caused by short circuit currents of the high voltage power line

Electromagnetic field

• Permanent interference caused by operating currents of the high voltage power line

In both cases by a suitable grounding of the pipeline the induced contact voltages must be safeguarded against exceeding the admissible extremes. In section 4, however, it was pointed out that disconnection from groundings is a prerequisite for cathodic protection. Groundings consume protection current, and, because of possible relaxation currents flowing after the protection current switch-off, they may cause measuring problems in the determination of true potentials by application of the off-potential technique.

7.2 Protective measures against too high voltages in personal contact According to to relevant standards and regulations the voltages permissible for bodily contact depend on the duration of their action, this enables different protection measures to be taken against the two above mentioned types of interference. 7.2.1 Short time interference caused by short circuit currents of the high voltage power line

In the case of a short circuit of this kind, power is cut off after about 0.2 s. Permissible for bodily contact are voltages exceeding 1000 V. This means the required groundings may be connected to the pipeline across excess voltage drains, working by way of gas discharge with starting voltages (threshold voltages) about 250 V. So during normal operation the groundings are electrically separated from the pipeline, and neither do they interfere with cathodic protection nor with its control by potential measurement. With a short circuit given, secure grounding is provided for the short duration required.

Soil resistivity

Length of  parallelism

Coating quality Coating resistance

Distance OH-line - Pipeline

Figure 18 Parameters regarding the causes of induced AC on pipelines

Important parameters regarding the causes of  induced AC on pipelines in utility corridors Pipeline

Overhead line

Dimensions

Operating current

Coating resistance

Short circuit current

Distance OH-line-Pipeline Phase imbalance Buried depth

Operating frequeny

Soil resistivity

Length of parallelism

7.3

Grounding the pipeline

With the pipeline having a coating resistivity r u   to be lowered to r uG because of high voltage interference and with n groundings of resistances R E each to be installed for this purpose S

S

=

r uG

r u

n

+

(28)

R E

S = π d l R E  is the surface of the pipeline section in question with the diameter d and the length l . So the required number of groundings is n = π d l R E x

r u - r uG

(29)

r u x r uG

In the case of very good coating resistance eq. (29) is simplified to

n = l  x

π d R E

Some experience-based practices are generally followed to “earth“ and “isolate“ the pipeline and/or “divert the induced high voltage/current surges to ground“. These practices are generally not backed-up by analytical field investigations. For optimization of grounding systems computer software is used to determine the necessary locations and the required earth resistances of single groundings along a influenced pipeline section. Results of these computer calculations confirm the predictions of experts but may surprise those who “earth“ and/or “isolate“ based on practice. The groundings may be made of galvanized steel, the zinc layer should at least 70 µm. Strip iron with a dimension 30 x 3.5 mm and rods with about 20 mm diameter are suitable for application as horizontal and deep seated groundings respectively.

(30)

r uG

Required earthing resistance

[ Ω ]

6,00 5,50 5,00 4,50 4,00 3,50

3.0 Ω

3,00 2,50

r     h     o    =     8      r  h   0     o  =   O     1  2    h     m    0  O   x     h  m   r   r     h  o   h     m    o    =  1   x  m   =     0  0   6      0     O  h   O     m   h     x  m   m    x     m   

r  h  o   =  1   5    0  O   h  m   x  m  

2,00 1,50 r ho = 4 0 O h

1,00

m x m

r ho = 20 Ohm x m

0,50 0,00 25

50

75

100

66 m Figure 19 Determination of earthing strip length depending on required earthing resistance and specific soil resistivity

125

150

175

200

250

Lengt of earthing strip Example of application: Required earthing resistance : 3.0 Ω Specific soil resistivity : 94.0 Ω m Diagram: selected ρ  =100 Ω m and 3.0 Ω Result: Lengt of earthing strip = 66 m, taken: 75 m

300

[m]

8

Conditions related to stress corrosion cracking; preventive measures

In the previous sections those types of corrosion were described exclusively which result in a more or less uniform metal consumption, according to eq. (2a. Corrosion damage is given by the diminishing of wall thickness and the development of trough-shaped cavities and pittings. Common to all of these processes are the anodic partial reaction of eq. (2), and the electrochemical protection measure, to shift the potential to < U Cu/CuSO4 = 0.85 V, the protection potential. But there are still other types of corrosion, that may lead to pipeline damage. They are governed by mechanisms different from those of the above discussed types and their dependence on potential is quite varied. Among these are the different kinds of stress corrosion cracking, SCC, characterized by the development of cracks under tensile stresses and without any noticeable material consumption.

8.1 Critical agents causing SCC SCC may occur provided there is a simultaneous action of sufficiently high tensile stresses and critical agents. Strong influences are exerted by the concentrations of these substances, by temperature and potential. In many cases SCC is even restricted to fairly narrow potential regions, so that the question about susceptibility to SCC cannot be answered without reference to the relevant potential. According to the knowledge available so far - mainly chemical in nature - the following substances may trigger SCC in steels: • Intergranular cracks Caustic alkali solutions, alkali carbonate, and bicarbonate; nitrates, fuming nitric acid and condensates containing nitrous oxides (e.g. from flue gases), ammonium salts of weak acids. • Transgranular cracks Aqueous condensates of hydrogen sulphide, prussic acid, carbon dioxide + carbon monoxide (gases common in refineries), liquefied ammonia and pressurized hydrogen.

8.2 Danger caused by corrosion on the inside surface of the pipes; protective measures Examples of critical substances, that may cause stress corrosion cracking at sufficiently high pressures, and pertinent protection measures are combined in this survey: Critical substances Protection measure H2 S, moist

Drying, inhibition

HCN, moist

Drying

CO + CO2, moist

Drying

liquid NH3

Inhibition (by adding H2O); O2 -removal

H2  (gaseous)

Inhibition (by adding O2 ), avoidance of sharp edged notches

SCC of pipelines so far mainly started from the inside of the pipes. As critical substances H2S and CO + CO2 had to be given attention. Corrosion and corrosion protection of the outside surface in the ground are of no concern to these processes. Protection measures are exclusively directed to conditioning the goods to be transported and to the working conditions of the pipelines.

8.3 Danger caused by corrosion on the outside surface of the pipes; protective measures Damage caused by SCC starting from the outside surfaces of pipelines is very rare, and restricted to lines working above ambient temperature, especially those under high pressure. The table below gives a survey concerning this matter. The features of these types of corrosion will be described in the following paragraphs. 8.3.1 SCC caused by nitrates

The critical potentials lie in the region of normal rest potentials. Ordinary cathodic protection totally eliminates the danger of SCC. Nitrates from artificial fertilizers being ubiquitous in farming areas, cathodic protection is always recommendable.

8.3.2 SCC caused by sodium hydroxide

The critical potentials are distinctly more negative than the protection potential. But SCC only occurs with temperatures permanently exceeding 50 °C (which does not happen in normal pipeline operation), and with very high concentrations of sodium hydroxide. Causative for its formation is, following eqs. (2b), (3), (4), cathodic polarization in combination with ion migration, eq. (14b). So, danger will only arise if there are very high protection current densities, if high concentrations of alkali ions are present, and if the sodium hydroxide is impeded from diffusing off by deposits or solid materials on the pipe surface. Lessening the protection current density is doubly effective. In the first place the probability of the critical potentials being reached and in the second place the concentration of sodium hydroxide is decreased. If, due to a lower current density, the protection potential is not fully attained in some areas, this situation has to be put up with. Avoiding SCC that may lead to early failure takes priority of definitely preventing the material consuming type of corrosion, that might - with a decreased probability! - lead to late occurring damage.

Critical substances at the pipe surface that may lead to stress corrosion cracking in soil if critical potentials and sufficiently strong mechanical stress are present Substance; origin

Critical potentials V VS . Cu/CuSO4

Damage became known in

Protection measures

Nitrates; fertilizers

> -0.45

Long-distance heating pipeline

Cathodic protection

(NH4)2CO3; frtilizers

about -0.65

NaOH; cathodes see eq.(4)!

about -1.1

Long-distance heating pipeline

Reduction of temperature (<50 °C)

NaHCO3; NaOH at cathodes CO2 from surroundings

about -0.7

High pressure gas pipeline

Cooling, blasting of steel surface, use of CO2-free coatings, sufficient cathodic protection

Na2CO2; (like with NaHCO3)

about -0.9

not specific (soil and salts)

predominantly with cathodic protection

Only in hard spots on the pipe surface!

Cathodic protection

(like with NaOH) High pressure pipeline Avoidance of hard spot formation (about >400 HV)

8.3.3 SCC caused by sodium bicarbonate

8.3.4 SCC in hard spots

The critical potentials are a bit more positive than the protection potential. Sodium bicarbonate is formed out of sodium hydroxide, which is cathodically generated, and carbon dioxide from the environment:

At the end of the table still another possible type of damage is included that may occur in hard spots of the steel surface, the type of material, and the potential being of no concern whatever. Cathodic protection may slightly further the progress of damage but is not actually causative. This type of SCC is caused by the hydrogen sensitivity of hard spots, the hydrogen being produced by a cathodic partial reaction according to eq. (4).

NaOH + CO2  ---> NaHCO3 .

(31)

The carbon dioxide may come from the soil or from the coating material. Because of the potential range required for this type of corrosion to occur, the bare steel surface below coating holidays is not in danger, provided sufficient cathodic protection is given. Critical potentials may occur below deposits and loose parts of the coatings, especially if there is mill scale still on the steel surface. For this reason, giving adequate cathodic protection, and the removal of mill scale from the steel surface are important protection measures. Another protective arrangement is directed against the formation of sodium bicarbonate after eq. (31). Carbon dioxide may come from the soil and permeate the coating, or, at elevated temperatures, be emitted by the coating material. For this reason only such coating materials should be used, that do not emit carbon dioxide at elevated temperatures, and have a comparatively small permeation coefficient for carbon dioxide. Another protection measure is lowering the temperature. In all these cases by diminishing mechanical stress the danger of SCC may be cut down decisively. In this respect not only stress created by internal pressure, but also additional loads, e.g. by thermal elongation, should be considered.

These hard spots exceeding 400 HV are not common phenomena on pipe surfaces. In case they do occur, danger is created only then if the coating is damaged in the very same place. The chance of such a coincidence is very narrow - and virtually eliminated by application of coatings with high mechanical strength. 8.3.5 SCC caused by cathodlcally produced hydrogen

In theory, hydrogen produced cathodically according to eq. (4), may cause hydrogen-induced SCC of pressurized pipelines, especially in the case of sulfides being present on the steel surface, a situation which cannot be excluded in the soil. This type of corrosion, though, is sharply dependent on the pH of the electrolyte in action. Damage can occur in condensates from sour gas (H2S), see section 8.2, not so, however, in soils with much higher pH values. Moreover by cathodic polarization the pH value at the steel surface is increased according to eqs. (3) and (4), so that this type of damage occuring in soils is definitely excluded even with higher strength pipeline steels. This is in conformance with all practical experience.

9

Behaviour of stainless steel in the ground

9.1 When to use stainless steel Stainless steel pipelines are very rarely used for underground service. One reason for using them is the specification of particular purity of the transported goods, like for instance fuel. Generally a normal austenitic chromium-nickel steel (German Standard material No. 1.4301; AISI 304), containing about 18% Cr and 9% Ni is used, that, due to its passivity, is totally stable in the soil. For particularly aggressive substances to be transported higher alloy steels with specific chemical properties may also be used. For instance in mineral oil and natural gas production the ferritic-austenitic duplex steel AF 22 (German Standard material No. 1.4462) is used for mains. This type of steel is highly resistant against chloride induced pitting and SCC, and apart from this it exhibits a comparatively high strength.

9.2 Stainless steel parts acting as foreign cathodes vs. unalloyed steel Mixed installation of stainless and unalloyed steels causes problems since stainless steel parts will act as foreign cathodes, see section 5.2.2. Corrosion danger may be relieved by electrically separating the parts of different materials and by coating the stainless steel parts. The parts may be wrapped with the simplest insulating tapes; these are only needed to insulate the greatest part of the surface but not to protect the stainless steel from corrosion.

9.3 Overall corrosion properties and influence of alien currents Stainless steel cannot be jeopardized by foreign cathodes in the ground, because, with respect to it they just do not exist. Anodic danger may only be caused by stray currents. This may happen, e.g., in case stainless steel parts are, by means of insulating flanges, kept separate from cathodically protected pipelines. Protection current may enter the stainless steel line and - similar to the situation in Fig. 14, - leave it near the insulating flange. Then, in this region an anodic polarization being present, chloride ions in the soil may cause pitting. For corrosion prevention the stainless steel line should be included into CP by means of a potentialconnection. The protection of stainless steel, though, does not require the potential that is recommended for the protection of unalloyed steel. Pitting is prevented by a protection potential U Cu/CuSO4 = - 0.1 V; this is definitely more positive than the protection potential for unalloyed steel. Stainless steel in the soil may be exposed to any potential more negative than the protection potential. There is no hazard that this might result in a loss of passivity.

9.4 Conditions requiring protective measures Stainless steels in the soil may be subject to corrosion danger on these conditions: • The material is sensitive to intergranular corrosion according to DIN 50 914. • The pipeline is operated at temperatures exceeding 50 °C (approximately) 9.4.1 Sensitizatlon

Stability might be impaired at welds. Improper heat treatment may sensitize stainless steels to intergranular corrosion. Linked to this is an increased sensitivity to pitting corrosion. Sensitization may be avoided by way of proper selection of materials and welding procedures. Danger of corrosion only exists with severely sensitized materials exposed to aggressive soils. Usually a weak cathodic polarization, which is effected simply by galvanic contact with pieces of unalloyed steel, gives full corrosion protection. 9.4.2 Concentrated chloride

With pipelines operated at higher temperatures there is always the danger of chlorides being concentrated below deposits on the steel surface. In the case of stainless steel this causes pitting, and if the steel is austenitic, the additional result is heavily branched transgranular SCC. Application of a cathodic protection potential U Cu/CuSO4 = - 0.1 V may be considered. In any case even a galvanic contact with unalloyed steel is sufficient to eliminate a possible corrosion danger. So, for the protection of stainless steel a piece of unalloyed steel may be used as a galvanic anode.