Study on the interference corrosion of cathodic protection system

Gan Cui; Zili Li; Chao Yang; Xu Wei

doi:10.1515/corrrev-2014-0061

Article Publicly Available

Study on the interference corrosion of cathodic protection system

Gan Cui
Gan Cui is a doctoral student at the China University of Petroleum. He is involved in corrosion testing and laboratory experiments. During his studies he received a first-class learning scholarship, a national inspirational scholarship, an excellent organizer of science and technology, outstanding cadres, and other honors. During his MS degree studies he won a first-class learning scholarship and a national scholarship. He has participated in more than 10 projects that are relevant to corrosion, and has published more than 10 papers.
, Zili Li
Zili Li is a Professor at the China University of Petroleum. His research focuses on numerical simulation in liquid-liquid separation hydro cyclone flow fields, cyclone structure optimization, gas pipe network dynamic simulation and optimization design, oilfield gathering pipe network optimal operation management software development, oil and gas storage and transportation safety evaluation, light oil storage and transportation systems of oil vapor recovery technology, oil and gas storage and transportation engineering anticorrosion technology. He presides over and participates in the ninth 5-year plan of CNPC and Sinopec and the oil group of young and middle-aged innovation fund. He has published more than 80 papers and is a member of NACE International.
, Chao Yang
Chao Yang is an MS student at the China University of Petroleum. He is involved in corrosion testing and laboratory experiments. During his undergraduate studies, he received a national inspirational scholarship, he was a first-class student, and has be awarded other honors. During his MS studies he won a second-class learning scholarship and was named a first-class student. He has participated in several projects relevant to corrosion.
and Xu Wei
Xu Wei, is a worker in Beijing Oil & Gas Transportation Center. In recent years, he completed three national and province departmental level topics. His innovative research results have been applied to corrosion and protection in oil and gas storage and transportation facilities, marine splash zone corrosion and protection, etc. He was awarded the first prize for ocean engineering science and technology and holds three national invention patents. He has published more than 20 academic papers.

Published/Copyright: August 31, 2015

Published by

Become an author with De Gruyter Brill

Submit Manuscript Author Information Explore this Subject

From the journal Corrosion Reviews Volume 33 Issue 5

Abstract

DC stray current has a great influence on the corrosion of buried pipelines. In this article, first, we deduce the equation of DC stray current interference on pipeline and determine the corresponding boundary conditions. Second, we discretize the mathematical model with boundary element method. Third, the numerical simulation software BEASY is applied to study the interference corrosion of cathodic protection system. Five types of cathodic protection (CP) interference are considered, namely, anodic, cathodic, combined, induced, and joint interference. Moreover, the effects of different parameters on the degree of the CP interference are investigated. The results show that the degree of interference will decrease with the increase of soil conductivity; smaller coating damage rate will make interference more concentrated; anode parameters have a significant impact on anodic interference but have almost no effect on cathodic interference; for combined, induced, and joint interference, the likelihood for corrosion occurrence will be high at the current outflow point on pipeline.

Keywords: BEM; cathodic protection; DC stray current; interference corrosion

1 Introduction

Stray current refers to current that does not flow in an intended current path, and it is one of the important reasons that leads to corrosion and leakage of underground metal pipeline (Li, Wu, Zheng, & Zhang, 2010). Stray current corrosion means corrosion that is caused by stray current and its essence in electrolysis of the electrochemical corrosion (Bertolini, Carsana, & Pedeferri, 2007). Because of the electrical conductivity of buried steel pipeline, potential difference is formed when stray current flows in the pipe (Brichau, Deconinck, & Driesens, 1996), which establishes corrosion cell. The corrosion that is caused by stray current is more serious than soil corrosion under normal condition (the potential difference of corrosion battery is only approximately 0.35 V without stray current, and the pipe-to-soil potential can be as high as 8–9 V when stray current exists) (Brichau et al., 1996), which also has a great influence on the corrosion, service life, and normal safe use of buried pipeline (Ding, Li, Jiang, Guo, & Bai, 2010). Therefore, it is of great significance to study stray current corrosion.

Stray current has three kinds of states: DC, AC, and the telluric currents in the earth, and among them, DC stray current does the greatest harm to the buried pipeline (Gao et al., 2010). DC stray current mainly derives from DC electrified railways, DC electrolytic equipment grounding electrodes, anode bed of cathodic protection system, and so on (Wang, Yan, Dong, Qian, & Li, 2010). Protective current in the cathodic protection system flows into the earth and makes the near metal components subject to corrosion. It will change the soil potential and form interference corrosion at the same time. The situations leading to the corrosion are different and basically can be divided into five types: anodic, cathodic, combined, induced, and joint interference.

Numerical methods have been demonstrated to be a powerful tool in the analysis of corrosion problems in the last two decades. Numerical methods applied to corrosion studies have included the finite difference method (FDM), the finite element method (FEM), and the boundary element method (BEM) (Metwally, Al-Mandhari, Gastli, & Nadir, 2007). BEM has been utilized to model the cathodic protection systems since early 1980s (Lan et al., 2012). Compared with FDM and FEM, BEM only requires the meshing of boundary. As a result, BEM requires fewer equations and a smaller matrix size than FEM and can solve both finite and infinite domain problems (Jia et al., 2004). The last but not the least, BEM is especially developed to calculate the DC stray current interference on pipeline networks and is able to model the complete soil and the entire traction system consisting of rails, traction stations, overhead wires, and trains (Bortels, Dorochenko, Vanden, Weyns, & Deconinck, 2007).

In this article, first, we establish the mathematical model of the cathodic protection (CP) interference and discretize the mathematical model using BEM. Then the five types of CP interference corrosion mentioned previously are overviewed, and various factors affecting the interference corrosion are considered and investigated by using simulation software. Simulation results will play a guiding role in CP interference problems in practice.

2 Mathematical model of the problem

2.1 Governing equation

To facilitate the computation, some simplifications are made: the solutions in the control domain are uniform and electroneutral; there is no concentration gradient in the electrolyte solution. On the basis of these assumptions, Ohm’s law applies, and the current density J follows from

(1) J = σ e E , (1)

where σ_e is the electric conductivity of soil (S/m), J is the current density (mA/m²), and E is the electric field (V/m). Then the static form of the equation of continuity can be written as

(2) ∇ J = ∇ ( σ e E ) = 0. (2)

Under static conditions, the electric potential is defined by the following equivalent equation:

(3) ∇ ϕ = E . (3)

Consequently, the governing equation for the electric potential is the Laplace equation:

(4) Δ ϕ = 0. (4)

2.2 Boundary conditions

Boundary conditions can be divided into the anode boundary condition, the cathode boundary condition, and the insulation boundary condition.

2.2.1 Anode boundary condition

The output current or potential of auxiliary anode is regarded as constant, as shown in the following equations:

(5) I = ∂ ϕ ∂ n = const (5)

(6) ϕ = const . (6)

2.2.2 Insulation boundary condition

The anode current cannot go through the soil into the air medium and only flows along the soil surface because of the infinite resistance of air. Therefore, we can use the ground as insulating surface, and the normal current density on the surface is zero:

(7) i | Γ = - σ ∂ ϕ ∂ n = 0 . (7)

At infinity, we believe that the potential and the normal current densities on the surface Γ_∞ are zero:

(8) { ϕ | Γ ∞ = 0 i | Γ ∞ = - σ ∂ ϕ ∂ n = 0. (8)

2.2.3 Cathode boundary

On the surface of cathode, many complex electrochemical reactions happen; the polarization character is the reflection of these reactions. The polarization data will be regarded as the cathode boundary to solve the model.

The polarization could be obtained through experiment measurements in the laboratory. In this work, the polarization curve of steel was measured in soil environment using a conventional three-electrode cell assembly. The physical and chemical properties of the soil (a sample) are as shown in Table 1. Rectangular platinum (Baoji Zhiming Special Metal Corporation, Baoji, Shanxi province, China) was used as the counter electrode, and saturated calomel electrode was used as the reference electrode. The working electrode (Baoji Zhiming Special Metal Corporation, Baoji, Shanxi province, China) was a cuboid, and its material was Q235. The steel electrode was mounted in epoxy resin (Qingdao Baiwei Corporation, Qingdao, Shandong province, China) such that only one surface of the cuboids was exposed to the soil. The area of the surface was 6.25 cm². Before the experiment, the working electrode was polished gradually using 600–1200 grid water proof abrasive papers (Shenzhen Jiuding Abrasives Corporation, Shenzhen, Guangdong province, China) then washed with distilled water [Chemical Sample Center of China University of Petroleum (East China), Qingdao, Shandong province, China], degreased with acetone [Chemical Sample Center of China University of Petroleum (East China), Qingdao, Shandong province, China] and washed again with ethanol [Chemical Sample Center of China University of Petroleum (East China), Qingdao, Shandong Province, China], and finally dried in a cold air stream. Afterward, the surface was coated by liquid epoxy with a damage rate of 5% (this would be varied when we study on the influence of coating damage rate). The electrochemical experiments were conducted using Parstat2273 (AMETEK Inc., Process & Analytical Instruments, Pittsburgh, PA, USA). The polarization curve measurement performed a scan starting from -400 to -1200 mV at a scan rate of 0.3 mV/s.

Table 1

Physical and chemical properties of the soil.

Moisture content (%)	Soluble salt (%)	(mmol/kg)				pH	Electrical conductivity (S/m)
Moisture content (%)	Soluble salt (%)	CO₃^2-	HCO₃^-	Cl^-	SO₄^2-	pH	Electrical conductivity (S/m)
12.31	0.04	0	2.22	1.2	2.38	7.35	0.01

The polarization curve, which will be applied as the cathode boundary condition, is a nonlinear curve; therefore, we must use polarization data in a piecewise linear interpolation approach (Jia et al., 2004). The polarization curve is shown in Figure 1, and we have made it to be a piecewise linear curve. All the pipe-to-soil potentials in the article are with respect to the saturated copper-copper sulfate electrode.

Figure 1:

The polarization curve. (A) The polarization curve of experimental test. (B) The piecewise linear curve.

3 Boundary element method

3.1 Discretization of models

When the BEM is applied, only boundaries of the domain need to be discretized. The pipeline is discretized using pipe elements method, and the result is shown in Figure 2. The number of nodes and elements reduces a lot when we use this method, and the computational complexity is much smaller.

Figure 2:

The pipe surface discretization.

It should be noted that there are prerequisite preconditions using the pipe elements method (Bortels et al., 2007): (1) the geometry of the protected body is suitable for cylinder unit subdivision and (2) the potential of the same cylinder unit can be considered as equal.

3.2 Method of calculation

It is known that the fundamental solution of the boundary integral equation (Liu, Sun, Wang, & Li, 2013) is 14πr;r refers to the distance between the boundary node and the source node. Therefore, the integration will encounter singularity when the boundary node coincides with the source node, which needs to be handled singly.

3.2.1 Computation of nonsingular coefficient

By using the previously mentioned pipe element method, with standard BEM formula, to make the boundaries discrete, the coefficient matrix of each element is obtained by integral transformation, as follows (Meng et al., 1998):

(9) G i , j m ( t ) = 4 R L 4 π | J | ϕ m ( t ) 4 K ( k ) [ ( R + B ) 2 + ( L t - Z r n ) 2 ] 1 / 2 , (9)

(10) H i , j m ( t ) = | J | ϕ m ( t ) - R L [ ( R + B ) 2 + ( L t - Z r n ) 2 ] 1 / 2 π ( L t - Z r n ) E ( k ) [ ( R - B ) 2 + ( L t - Z r n ) 2 ] 2 . (10)

In these two formulas, i≠j. K(k) is the first kind of ellipse integral, and E(k) is the second kind of ellipse integral. J and B are the results of the coordinate transformation, and t is the local coordinate. Z_rn is the third coordinate of the last pipe element node.

3.2.2 Computation of singular coefficient

For semi-infinite region (Brichau & Deconinck, 1994),

(11) H i i = - ∑ H i j , i ≠ j . (11)

The analytical method to solve G_ii is as follows (Wu, 2008):

(12) G i i = L 2 π [ 1 -ln ( L 16 r ) ] . (12)

Finally, the standard BEM formula is simulated by numerical integral, and the result is as follows:

(13) { G } × { Q } = { H } × { ϕ - η ( Q ) } . (13)

4 Results and discussion

4.1 Anodic interference

The positive potential area is formed near the anode bed of the cathodic protection system, and the potential depends on the ground bed form, soil conductivity, and anode output current. If there are other metal pipes passing through the area, part of the current will flow into these pipes and flow along the pipes, and then flow into the earth from the appropriate location of the pipes. The position of the pipe that current flows out will be the region where corrosion happens. The reason for this corrosion is that the interfered pipe and the anode bed of CP system are close to each other; hence, it is called anodic interference. The diagram is as shown in Figure 3.

Figure 3:

Sketch map of anode interference.

In Figure 3, the parameters in the model are set as follows: protected pipe – length 1600 m, depth 2 m, diameter 0.7 m, and 50 m away from auxiliary anode; unprotected pipe – length 1600 m, depth 2 m, diameter 0.7 m, and 10 m away from auxiliary anode; and auxiliary anode – diameter 0.1 m, depth 2 m, and length 2 m. On the basis of these basic parameters, a simulation software is used to build a model to study the influence of soil electrical conductivity, coating damage rate, anode location, anodic output current, and anode depth on anodic interference corrosion.

4.1.1 Soil conductivity

Figure 4 is the simulated results when the soil conductivity is 0.005, 0.01, 0.1, and 0.5 S/m. From Figure 4A, it can be seen that when the soil conductivity is low (0.005 and 0.01 S/m), the potential distribution of the whole pipeline is funnel shaped. The potential near the anode reaches the most negative value (-1500 mV) and reaches maximum value (-300 mV) at both ends of pipeline. The potential distribution through the whole pipe is very nonuniform. From Figure 4B, it can be seen that the current density value is negative near the anode, which means that the current flows into pipe and the pipe is protected to some extent. At both ends of the pipeline, the current value is positive, which means that the current flowing out of the pipe and the likelihood for corrosion occurrence is high. With the increase of soil conductivity (0.1 and 0.5 S/m), the potential along pipe gets lower (-600 mV) overall and is close to the natural potential. The potential distribution is more uniform and the degree of the interference decreases. Therefore, with the increase of soil conductivity, the potential along the pipe gets lower, and as the distribution gets uniform, the interference degree decreases. The reason is that when the soil electrical conductivity is high, the loop resistance is low, and more current can flow into the pipe from far end. When soil conductivity is low, loop resistance is high; most of the current flows into pipe near the anode bed, and the current flows along the pipeline then flows back into the earth at the ends of the pipe where serious corrosion will occur with a high possibility.

Figure 4:

Profiles of potential and current density along the unprotected pipe as a function of soil conductivity. (A) Potential profile along the unprotected pipe. (B) Current density profile along the unprotected pipe.

4.1.2 Coating damage rate

Figure 5 is the simulated results when the coating damage rate is 1%, 5%, and 10%. Figure 5 shows that when the coating damage rate is small, the whole pipe potential distribution is extremely nonuniform (0 to -2200 mV). At both ends of the unprotected pipe, the potential reaches positive, which indicates that the likelihood for corrosion occurrence is high. As the coating damage rate increases, the whole pipe potential gets lower, the distribution gets uniform, and the interference degree decreases. This shows that the smaller coating defects often make interference more concentrated, which makes it easier to corrode.

Figure 5:

Effect of coating damage rate of the unprotected pipe on potential and current density profiles along the unprotected pipe. (A) Potential profile along the unprotected pipe. (B) Current density profile along the unprotected pipe.

4.1.3 Anode parameters

Figure 6 is the simulated results that exhibit the influence of the distance d between the anode and the unprotected pipe on the anodic interference corrosion. Figure 6A shows that with the increase of the distance between unprotected pipe and anode, the whole pipe potential gets lower, the distribution gets uniform, and the interference degree decreases. Figure 6B shows that when the pipe is close to the anode (d=10 m), a large amount of current flows into the pipe near the anode ground bed and flows out from the remote end, which leads to the high likelihood for corrosion. With the increase of the distance between pipe and anode, the current density distribution is more uniform. The reason is that when the distance between pipe and anode is short, there will be not enough space for the current to get uniform distribution before flowing into the pipe; if the distance is long, there will be a big media space for the current to get more uniform before flowing into the pipe.

Figure 6:

Effect of anode location on potential and current density profiles along the unprotected pipe. (A) Potential profile along the unprotected pipe. (B) Current density profile along the unprotected pipe.

Figure 7 is the simulated results that show the influence of the anode output current I on the anodic interference corrosion. Figure 7A shows that with the increase of anode output current, the potential of the unprotected pipe near the anode bed gets lower and the protected effect is improved. However, the potential at the ends of the unprotected pipe gets higher instead of getting lower, which increases the likelihood for corrosion. This phenomenon can be explained with the result of the current distribution shown in Figure 7B. When anode output current increases, more current flows into the unprotected pipe near the anode bed, and the pipe-to-soil potential decreases, so the protection effect gets better. Meanwhile, because the total current is constant, more current will outflow from the ends of the pipe, which will increase the likelihood for corrosion.

Figure 7:

Effect of anode current on potential and current density profiles along the unprotected pipe. (A) Potential profile along the unprotected pipe. (B) Current density profile along the unprotected pipe.

Figure 8 is the simulated results that demonstrate the influence of the anode depth h on the anode interference corrosion. From the simulated results, it can be seen that with the increase of anode depth, the whole unprotected pipe potential gets lower, the distribution gets uniform, and the interference degree decreases. The reason is that with the anode depth increasing, the current becomes more evenly distributed before flowing into the pipe, so there will be no concentrated inflowing and outflowing region. However, it should be noted that with the anode depth increasing, the distance between the anode bed and the unprotected pipe will get longer than that between the two pipes, so the interference corrosion style will change from anodic to cathodic interference.

Figure 8:

Effect of anode depth on potential and current density profiles along the unprotected pipe. (A) Potential profile along the unprotected pipe. (B) Current density profile along the unprotected pipe.

From the previously mentioned simulated results, we can conclude that for anodic interference, the anode parameters such as anode location, anode output current, and anode depth have great influence on the interference corrosion. Thus, in reality, more attention should be paid when the unprotected pipe passes through the anode bed.

4.2 Cathodic interference

The soil potential near the protected pipe of the cathodic protection system is lower than that of the other areas. If other metal pipes pass by this area, current inflowing from the remote end of the pipe will outflow, so severe interference corrosion will occur with a high possibility. The reason why corrosion occurs is that the interfered pipe and the cathode of CP system are close to each other; hence, this interference is called cathodic interference. The diagram is as shown in Figure 9. The range of cathodic interference is limited and only at the intersection of two pipes.

Figure 9:

Sketch map of cathodic interference.

In Figure 9, the parameters in the model are set as follows: protected pipe – length 1600 m, depth 2 m, diameter 0.7 m, 50 m away from auxiliary anode, and horizontal runs; unprotected pipe – length 1600 m, depth 4 m, diameter 0.7 m, vertical runs, and the crossing location is 1400 m far from the start of the protected pipe; and auxiliary anode – diameter 0.1 m, depth 2 m, and length 2 m. On the basis of these basic parameters, a simulation software is used to build a model to study the influence of soil electrical conductivity, the cross angle of two pipes, the vertical distance of two pipes, and the anodic parameters on cathodic interference corrosion.

4.2.1 Soil conductivity

Figure 10 shows the simulated results when the soil conductivity is 0.005, 0.01, 0.1, and 0.5 S/m, respectively. According to the simulation results, the corrosion potential and the current density distribution of the unprotected pipeline become extremely uneven with the decrease of the soil conductivity. When the soil conductivity is very large, the stray current density of the unprotected pipeline is substantially zero, the corrosion potential is self-corrosion potential, and basically the unprotected pipeline is not affected. When the soil conductivity is very small, the current density nearby the pipeline crossing location becomes positive, the corrosion potential is much more positive than the self-corrosion potential, and the pipeline will suffer a high likelihood for severe corrosion.

Figure 10:

According to this figure, there are two intersections in every curve: the current density between two intersections, which is positive (when current flows out, the corrosion of pipeline aggravates), and the current density outside two intersections, which is negative (when the current flows in, the pipeline is under a certain extent of cathodic protection). Therefore, the change of soil resistivity will not result in the change of the location and length of the pipe sections where corrosion aggravates; it just makes the degree of the corrosion aggravation different.

4.2.2 Inclination angle between two pipes

Figure 11 shows the simulated results when the inclination angle between two pipes is 30°, 60°, and 90°, respectively (as marked in Figure 9). The simulated results show that at the intersection of two pipes (Y=0), with the angle increasing, the potential and the current values of unprotected pipe change slightly. When the angle changes from 30° to 90°, the change of the potential value is <10 mV. Thus, the change of the inclination angle of two pipes basically has no influence on the cathodic interference corrosion.

Figure 11:

Effect of angle between the protected pipe and unprotected pipe on potential and current density profiles along the unprotected pipe. (A) Potential profile along the unprotected pipe. (B) Current density profile along the unprotected pipe.

However, the potential fluctuations at two ends of the pipeline are relatively large for the position of auxiliary anode to remain unchanged when the pipeline inclination angle is changing. Because the relative location of pipeline and auxiliary anode has been changed, the potential at each end of the pipeline has large changes. It is worth pointing out that the locations of the anode and the intersection are always unchanged, and this result is meaningful when comparing the potential of intersections.

4.2.3 Vertical distance between two pipes

Figure 12 is the simulated results when the vertical distance d between two crossing pipes is 2, 10, and 20 m. The simulated results show that with the vertical distance increasing, the potential in the crossing location of the unprotected pipe is gradually decreasing and the corresponding current density is decreasing; thus, the interference degree is decreasing. However, in the other location, the increase of vertical distance has no influence on the potential and current distributions. Therefore, the decrease of the vertical distance only makes cathodic interference degree increase and corrosion severe at the intersection between two pipes, and it has no effect on the other locations of the unprotected pipeline.

Figure 12:

Effect of vertical distance between the protected pipe and unprotected pipe on potential and current density profiles along the unprotected pipe. (A) Potential profile along the unprotected pipe. (B) Current density profile along the unprotected pipe.

4.2.4 Anode parameters

Figure 13 is the simulated results when the anode location X is 0, 100, and 200 m. With the X increasing, i.e. the distance between the anode and the unprotected pipe decreasing, the potential at the intersection of the unprotected pipe decreases, and the interference degree decreases. However, the decrease is not significant. When there is a change of 200 m in X, the potential changes approximately 20 mV correspondingly. Thus, the change in the anode location basically has no effect on the cathodic interference corrosion.

Figure 13:

Figure 14 shows the simulated results when the anode output current I is 1500, 2400, and 3000 mA. The simulated results show that with the anode output current increasing, the potential at the intersection of the unprotected pipe increases, current density increases, and interference degree gets bigger. At the ends of the unprotected pipe, with the anode output current increasing, the potential decreases and the current density decreases, which lead to better protection. Therefore, with the anode output current increasing, the potential distribution of the unprotected pipe gets more nonuniform and interference corrosion gets more severe. However, with the output current changes from 1500 to 3000 mA, the potential changes for approximately 20 mV. Therefore, the change of anode output current basically has no effect on cathodic interference corrosion.

Figure 14:

Effect of anode current on potential and current density profiles along the unprotected pipe. (A) Potential profile along the unprotected pipe. (B) Current density profile along the unprotected pipe.

Figure 15 shows the simulated results when the anode buried depth h is 2, 20, and 50 m. The simulated results show that with the depth increasing, the potential distribution curve and the current distribution curve of the unprotected pipe basically coincide; the cathodic interference degree of unprotected pipe basically has no change.

Figure 15:

Effect of anode depth on potential and current density profiles along the unprotected pipe. (A) Potential profile along the unprotected pipe. (B) Current density profile along the unprotected pipe.

From the previously mentioned simulated results, for the cathodic interference, because the unprotected pipe is far away from the anode bed, the changes of anodic related parameters basically have no effect on the cathodic interference corrosion. Thus, the anode is not the factor that changes the cathodic interference degree.

4.3 Combined interference

If a pipe passes through the anode bed of a cathodic protection system and meanwhile crosses with the cathode of this system, the interference corrosion has two aspects. On the one hand, the current flows into the pipe near the anode and then flows into soil from another location of the pipe, so anodic interference occurs. On the other hand, the current flowing into the pipe at the remote location will flow into soil at the intersection, which causes corrosion. Because corrosion depends on anodic interference and cathodic interference at the same time, it is called the combined interference. The diagram is as shown in Figure 16.

Figure 16:

Sketch map of combined interference.

In Figure 16, the parameters in the model are set as follows: protected pipe – length 1600 m, depth 2 m, diameter 0.7 m, 50 m away from auxiliary anode, and vertical runs; unprotected pipe – length 1600 m, depth 2 m, diameter 0.7 m, 50 m away from auxiliary anode, and the intersection between the two pipes is 1400 m far away from the end of the protected pipe; and auxiliary anode – diameter 0.1 m, depth 2 m, length 2 m, and output current 2400 mA. The soil conductivity is 0.01 S/m. On the basis of these basic parameters, a simulation software is used to build a model to simulate the combined interference.

Figure 17 shows the potential and the current density distribution curves of the unprotected pipe. It shows that there is a peak value and a valley value in the potential distribution curve. In the valley, where the unprotected pipe is near the anode bed, the potential value is low (-760 mV). This means that a lot of cathodic current flows into the unprotected pipe, which makes it under protection to some extent. In the peak where the two pipes cross with each other, the potential value is high (-450 mV). This means that a lot of current flows out of the unprotected pipe at intersection, which aggravates corrosion. Therefore, combined interference can aggravate the corrosion of the unprotected pipeline, and the corrosion only occurs at the intersection of two pipes.

Figure 17:

Potential and current density profiles along the unprotected pipe. (A) Potential profile along the unprotected pipe. (B) Current density profile along the unprotected pipe.

4.4 Induced interference

The current in soil using a metal structure as media to cause interference is called induced interference. As shown in Figure 18, when an underground pipe (which is called pipe 1) passes by the anode of the cathodic protection system, another pipe (which is called pipe 2) crosses with the cathode of the cathodic protection system but does not pass by the anode. Pipes 1 and 2 cross with the third pipe (which is called pipe 3) at the same time. Obviously, the interference of pipe 1 is anodic interference, and the interference of pipe 2 is cathodic interference. Pipe 3 is an induced pipe, which is suffered by induced interference. The location of the three pipes where the current outflows corrodes severely with a high possibility.

Figure 18:

Sketch map of induced interference.

In Figure 18, the parameters in the model are set as follows: protected pipe – length 800 m, depth 2 m, diameter 0.7 m, 50 m away from auxiliary anode, and horizontal runs; unprotected pipe 1 – length 1600 m, depth 2 m, diameter 0.7 m, 50 m away from auxiliary anode, and horizontal runs; unprotected pipe 2: length 800 m, depth 3 m, diameter 0.7 m, and crosses with the protected pipe; unprotected pipe 3 – length 800 m, depth 4 m, diameter 0.7 m, vertical runs, and crosses with pipes 2 and 3; and auxiliary anode – diameter 0.1 m, depth 2 m, length 2 m, and the output current is 2400 mA. Soil conductivity is 0.01 S/m. On the basis of these basic parameters, a simulation software is used to build a model to simulate the induced interference.

Figure 19A and B shows the potential and the current density distribution curves of the unprotected pipe 1, respectively. The interference of unprotected pipe 1 is anodic interference. Near the anode bed, a lot of current flows into the pipe and the potential is low; thus, the pipe gets some protection. However, at the intersection of pipes 1 and 3, the current flows from pipe 1 to pipe 3; thus, pipe 1 corrodes. Figure 19C and D shows the potential and the current density distribution curves of unprotected pipe 2, respectively. The current flows from pipe 3 to unprotected pipe 2 at the intersection of pipes 2 and 3; thus, the potential of pipe 2 at the intersection is low and gets some protection. However, the current outflows from pipe 2 at the intersection of pipe 2 and the protected pipe, and the potential of the intersection is high, which will aggravate corrosion. The interference of pipe 2 is cathodic interference. Figure 19E and F shows the potential and the current density distribution curves of unprotected pipe 3, respectively. The current flows from pipe 1 to pipe 3 at the intersection of pipes 1 and 3; thus, the potential of pipe 3 at this point is low and gets some protection. However, the current outflows from pipe 3 at the intersection of pipes 2 and 3; thus, the potential of pipe 3 at this point is high, which will aggravate corrosion. The interference of pipe 3 is induced interference.

Figure 19:

Potential and current density profiles along unprotected pipes. (A) Potential profile along the unprotected pipe 1. (B) Current density profile along the unprotected pipe 1. (C) Potential profile along the unprotected pipe 2. (D) Current density profile along the unprotected pipe 2. (E) Potential profile along the unprotected pipe 3. (F) Current density profile along the unprotected pipe 3.

From the previously mentioned analysis, it can be seen that all of three pipes in Figure 18 will suffer a high likelihood for corrosion. The difference is the interference style. However, there will be high likelihood for corrosion at the location where the current outflows. It is noted that in this style of interference, for the unprotected pipe, corrosion does not necessarily occur at the intersection of two pipes. For example, at the intersection of pipes 1 and 3, pipe 3 gets some protection.

4.5 Joint interference

The corrosion caused by the unbalance of potential in the joint is called joint interference. For example, cathodic interference in the insulation flange belongs to this corrosion. Potential on two sides of the insulation flange is different, and the difference is more than 0.5 V. The potential of the unprotected pipe is higher than the protected pipe. Thus, the current flows from the high potential area to the low potential area, which causes the unprotected pipe interference and aggravates its corrosion. This kind of corrosion mostly occurs within 5–10 m away from the insulation flange. Therefore, it is regulated that the pipe anticorrosion coating should be strengthened near the insulation flange.

In Figure 20, the parameters in the model are set as follows: protected pipe – length 1600 m, depth 2 m, diameter 0.7 m, 50 m away from auxiliary anode, horizontal runs, and there is an insulation flange at the end; unprotected pipe – length 400 m, depth 2 m, diameter 0.7 m, and connected with the protected pipe by the insulation flange; auxiliary anode – diameter 0.1 m, depth 2 m, and length 2 m. Soil conductivity is 0.01 S/m. On the basis of these basic parameters, a simulation software is used to build a model to study the influence of coating damage rate and anodic output current on joint interference corrosion.

Figure 20:

Sketch map of joint interference.

4.5.1 Coating damage rate

Figure 21 shows the simulated results when the coating damage rate is 1%, 5%, and 10%. The simulated results show that when the coating damage rate is low, the potential of the unprotected pipe near insulation joint is high and the interference corrosion degree is high. With the damage rate increasing, the potential and the interference degrees decrease gradually. However, the reduced potential value is small, which is approximately 20 mV. Therefore, the coating damage rate has no significant effect on the joint interference. However, once the coating is damaged, a lot of current will outflow, which makes corrosion severe. Thus, near the insulation joint, coating grades should be strengthened to avoid the joint interference corrosion.

Figure 21:

4.5.2 Anode current

Figure 22 is the simulated results when the anodic output current I is 1500, 2400, and 3000 mA. The simulated results show that with the anodic output current increasing, the potential of the unprotected pipe near insulation joint obviously increases, current density increases, and the possibility of corrosion gets higher. Therefore, the increase of anodic output current makes the joint interference corrosion severe. The reason is that when the anode output current increases, the polarization of the protected pipe enhances, and the cathodic protected potential of the whole pipe reduces. Thus, as the potential difference between the protected and the unprotected pipe increases, the outflowing current density increases correspondingly. As a result, the joint interference corrosion gets severe.

Figure 22:

Effect of anode current on potential and current density profiles along the unprotected pipe. (A) Potential profile along the unprotected pipe. (B) Current density profile along the unprotected pipe.

5 Conclusions

The simulation software based on BEM was used to establish a three-dimensional model to simulate five styles of cathode protection interference corrosion, namely, anodic, cathodic, combined, induced, and joint interference. The following conclusions have been reached:

For anodic interference
- With the soil conductivity increasing, stray current distribution is more uniform, and the anodic interference corrosion degree of unprotected pipe decreases.
- The small coating damage rate can make the interference concentrated and aggravates anodic interference corrosion.
- The changes of anodic parameters have significant effects on anodic interference corrosion. With the distance between the unprotected pipe and the anode increasing, anodic output current decreasing, and anodic buried depth increasing, the anodic interference degree decreases.
For cathodic interference
- With the soil conductivity increasing, stray current distribution is more uniform, and the cathodic interference corrosion degree of unprotected pipe decreases.
- The inclination angle between two pipes basically has no effect on the cathodic interference degree of unprotected pipe.
- With the vertical distance between two pipes changing, the interference corrosion of the unprotected pipe at the intersection decreases, but it has no obvious changes in other locations.
- Because the protected pipe is far away from the anode bed, the change of anodic parameters has no obvious effect on the cathodic interference corrosion degree of unprotected pipe.
For combined interference and induced interference, the location where the current outflows in the unprotected pipe has a high likelihood for corrosion.
For joint interference
- The coating damage rate of unprotected pipe has no obvious effect on the interference corrosion degree. However, once the coating is damaged, severe corrosion will occur with a high possibility. Thus, near the insulation joint, coating grades should be strengthened.
- With the anodic output current increasing, the potential difference between the protected and the unprotected pipe increases. It will aggravate interference corrosion.

Corresponding author: Gan Cui, College of Pipeline and Civil Engineering, China University of Petroleum (East China), The Yangtze River West Road No. 66 in Huangdao, Qingdao, Shandong 266580, China, e-mail: chennacuigan@163.com

About the authors

Gan Cui

Gan Cui is a doctoral student at the China University of Petroleum. He is involved in corrosion testing and laboratory experiments. During his studies he received a first-class learning scholarship, a national inspirational scholarship, an excellent organizer of science and technology, outstanding cadres, and other honors. During his MS degree studies he won a first-class learning scholarship and a national scholarship. He has participated in more than 10 projects that are relevant to corrosion, and has published more than 10 papers.

Zili Li

Zili Li is a Professor at the China University of Petroleum. His research focuses on numerical simulation in liquid-liquid separation hydro cyclone flow fields, cyclone structure optimization, gas pipe network dynamic simulation and optimization design, oilfield gathering pipe network optimal operation management software development, oil and gas storage and transportation safety evaluation, light oil storage and transportation systems of oil vapor recovery technology, oil and gas storage and transportation engineering anticorrosion technology. He presides over and participates in the ninth 5-year plan of CNPC and Sinopec and the oil group of young and middle-aged innovation fund. He has published more than 80 papers and is a member of NACE International.

Chao Yang

Chao Yang is an MS student at the China University of Petroleum. He is involved in corrosion testing and laboratory experiments. During his undergraduate studies, he received a national inspirational scholarship, he was a first-class student, and has be awarded other honors. During his MS studies he won a second-class learning scholarship and was named a first-class student. He has participated in several projects relevant to corrosion.

Xu Wei

Xu Wei, is a worker in Beijing Oil & Gas Transportation Center. In recent years, he completed three national and province departmental level topics. His innovative research results have been applied to corrosion and protection in oil and gas storage and transportation facilities, marine splash zone corrosion and protection, etc. He was awarded the first prize for ocean engineering science and technology and holds three national invention patents. He has published more than 20 academic papers.

Acknowledgments

This work was financially supported by the National Natural Science Foundation of China (grant no. 50804053) and the National Science and Technology Major Projects of Large Oil & Gas Fields and Coal Bed Methane Development (grant no. 2008ZX05017-04-01).

References

Bertolini L, Carsana M, Pedeferri P. Corrosion behavior of steel in concrete in the presence of stray current. Corros Sci 2007; 49: 1056–1068.10.1016/j.corsci.2006.05.048Search in Google Scholar

Bortels L, Dorochenko A, Vanden BB, Weyns G, Deconinck J. Three-dimensional boundary element method and finite element method simulations applied to stray current interference problems. A unique coupling mechanism that takes the best of both methods. Corrosion 2007; 6: 561–576.10.5006/1.3278407Search in Google Scholar

Brichau F, Deconinck J. A numerical model for cathodic protection of buried pipes. Corrosion 1994; 1: 39–49.10.5006/1.3293492Search in Google Scholar

Brichau F, Deconinck J, Driesens T. Modeling of underground cathodic protection stray currents. Corrosion 1996; 52: 480–488.10.5006/1.3292137Search in Google Scholar

Ding HT, Li LJ, Jiang MH, Guo ZW, Bai HL. Lab research of DC & AC stray current for the failure effect of the gas pipelines buried underground. Guizhou Chem Ind 2010; 5: 9–12 (in Chinese).Search in Google Scholar

Gao B, Shen LS, Meng XQ, Yan LJ, Han HC, Ma LX. DC stray current corrosion and protection of oil gas pipeline. Pipeline Tech Equip 2010; 4: 42–43 (in Chinese).Search in Google Scholar

Jia JX, Song G, Atrens A, John DS, Baynham J, Chandler G. Evaluation of the BEASY program using linear and piecewise linear approaches for the boundary conditions. Mater Corros 2004; 11: 845–852.10.1002/maco.200403795Search in Google Scholar

Lan ZG, Wang XT, Hou BR, Wang ZF, Song JW, Chen SL. Simulation of sacrificial anode protection for steel platform using boundary element method. Eng Anal Bound Elem 2012; 36: 903–906.10.1016/j.enganabound.2011.07.018Search in Google Scholar

Li YT, Wu MT, Zheng F, Zhang SQ. The stray current monitoring for pipelines and the application example in Chuanxi gas field. Total Corros Control 2010; 1: 25–28 (in Chinese).Search in Google Scholar

Liu GC, Sun W, Wang L, Li L. Modeling cathodic shielding of sacrificial anode cathodic protection systems in seawater. Mater Corros 2013; 6: 472–477.10.1002/maco.201206726Search in Google Scholar

Meng XJ, Wu ZY, Liang XW, Zhuang XG, Yuan ZK, Liu YH. Improvement of algorithm for regional cathodic protection model. J Chin Soc Corros Protect 1998; 9: 221–226 (in Chinese).Search in Google Scholar

Metwally IA, Al-Mandhari HM, Gastli A, Nadir Z. Factors affecting cathodic protection interference. Eng Anal Bound Elem 2007; 31: 485–493.10.1016/j.enganabound.2006.11.003Search in Google Scholar

Wang Y, Yan YG, Dong CF, Qian JH, Li XG. Effect of stray current on corrosion of Q235, 16Mn and X70 steels with damaged coating. Corro Sci Protect Technol 2010; 2: 117–119 (in Chinese).Search in Google Scholar

Wu HT. The application of boundary element method in heat transfer. Beijing: National Defense Industry Press, 2008.Search in Google Scholar

Received: 2014-12-11

Accepted: 2015-06-23

Published Online: 2015-08-31

Published in Print: 2015-09-01

Articles in the same Issue

https://doi.org/10.1515/corrrev-2014-0061

Keywords for this article

BEM; cathodic protection; DC stray current; interference corrosion