THE EFFECT OF SOIL ENVIRONMENT ON THE CORROSION OF PIPELINE

The aim of this work was to investigate the effect of soil environment on steel pipeline corrosion. The corrosion of the gas pipeline DN 500 after 48-year operation was studied by electrochemical methods. The samples of the pipeline were tested in soil electrolyte for 6 hours. Corrosion resistance of the steel samples was evaluated by open circuit potential and potentiodynamic polarization resistance measurements. Corrosion rate icorr of studied steel was determined using a potentiodynamic polarization method. Corrosion behavior of surface without and with corrosion products was compared. The corrosion rate icorr of the sample with corrosion layer reached the lower value compared to the sample without corrosion layer.


Introduction
Underground metal structures also called pipelines represent highly integrated distribution network that transport natural gas and crude oil over long distances in the world.The most commonly used material in gas industry is carbon steel due to low cost, mechanical properties and easy availability [1][2].These pipelines are exposed to different environmental aggressive conditions (soil, underground water, seawater) which can cause the corrosion attack.Corrosion is the electrochemical process that involves the flow of electrical currents on a micro or macro scale.For corroding steel, anodic and cathodic reactions produce the electrochemical cell.The anodic, Equation 1, and cathodic electrochemical reaction, Equation 2, must occur simultaneously.The corrosion process takes place in neutral environment according to Equation 3 and 4; in environment with higher concentration of oxygen according to Equation 5 and with lower concentration of oxygen according to Equation 6 [3][4][5][6].( Pipelines corrosion belongs to the most significant problems in the gas industry, which are caused by several factors such as: types of soil, moisture content in soil, pH, soil resistivity, aeration degree, redox potential, oxidation-redox potential, bacterial activity and concentration of chlorides, sulphates, HCO 3 -and oxygen [7][8][9].The presence of sulphates and chlorides in the soil is considered as corrosive environment for steel pipelines.If chloride and sulphate concentration in soil is below 100 mg.L -1 and 200 mg.L -1 respectively, the soil environment is evaluated as slightly aggressive.On the other hand, sulphate concentration higher than 200 mg.L -1 presents a significant corrosion risk for natural gas pipelines.The microbiological induced corrosion of the pipelines widely occurs in soil environment, too.The same types of bacterial strains are able to damage both outer and inner surface of the pipe and to form corrosion products, which may be observed as deposits.The largest group of microorganisms that are responsible for the corrosion degradation of pipelines are among the sulphate -reducing bacteria [10][11][12][13][14].The corrosion resistance of steel pipelines can be divided into four classes.This classification is given in Table 1 [14]. Table 1 Classification of corrosion resistance [14] Corrosion rate (mm/year) Corrosion resistance < 0,05 excellent 0,05 -0,1 good 0,1 -0,5 middle  0,5 bad Outer surface corrosion of pipelines causes more than 80 per cent of corrosion failure in distribution pipelines.Therefore, there is a need of external pipeline surface protection.The corrosion control can have a major impact on economics, safety and environmental protection of natural gas pipeline operation [10,[15][16].Passive protection is a set of preventive measurements against outer surface corrosion.The aim of passive protection is the prevention the metal structure against to electrochemical corrosion and to support active protection.The 3-layer polyethylene, polyethylene (3 LPE), fusion bonded epoxy (FBE), polyethylene (PE), polyurethane (PUR), bitumen coatings are the most commonly used passive protection system of the pipelines.However, the passive protection is frequently used in combination with active protection.There are two ways to obtain electrochemical controlled current flow: using sacrificial anode system or impressed current cathodic protection system [17][18][19][20].
As mentioned above the corrosion process of the experimental steel pipeline has the electrochemical character.For this reason it was important to determine the electrochemical characteristic of the process using the Tafel and Stern methods.
where: B (mV) the Stern and Geary coefficient R p (Ω.cm 2 ) the polarization resistance ß a (V/decade)the anodic Tafel slope ß c (V/decade)the cathodic Tafel slope j/E (Ωcm 2 )slope of the polarization curve; so called R p Linear polarization resistance is known as derived an equation relating the slope of linear region to the corrosion rate and Tafel plot.The cathodic (ß c ) and anodic (ß a ) branch of Tafel are intersected at the point which determines the corrosion potential E corr and corrosion density i corr of samples in a given electrolyte.This theory explains the corrosion reactions on the basis of cathodic and anodic partial reactions that are occurring at the interface electrode of metal and electrolyte [21].

Experimental material and methods
The investigated material in these experiments was steel EN 10028/2-92 after 48 -year operation.Both the surface samples after grinding (surface without corrosion products) and the surface samples with corrosion layer were tested.The thickness of corrosion products was 53.09 m.The chemical composition of the steel pipeline is given in Table 2.The structure of the steel is documented in Fig. 1.The experimental samples sizes of 20 x 15 x 10 mm were impressed in the soil electrolyte, which presented corrosion environment, for 6 hours.Macroscopic analysis of the samples surface after corrosion tests was carried out using the macroscope LeicaWild M3Z.During the measurement, the specimens were gradually polarized by direct current from an external source in an interval from -900 to + 900 mV at the change of potential rate 5 mV.s -1 .Linear polarization was measured using the Voltalab P6P 201.During the measuring the three -electrode setup composed of tested specimen, reference saturated calomel electrode (SCE) and auxiliary Pt electrode was used.The results of corrosion potential E corr , current density j corr and polarization resistance R p measurement were evaluated using software Voltmaster 4.
The corrosion rate was evaluated from the polarisation curves by Tafel extrapolation.Polarization resistance R p was measured in the area of corrosion potential 20 mV [23].
Corrosion tests were performed by immersing the steel samples in the prepared corrosion electrolyte at room temperature for 6 hours.The corrosion rate of the steel samples immersed in the experimental electrolyte was determined according to the Equation 11 [24]: CR = ((j corr x M) / ( Fe x Z)) x 3270 (11) where: CR (mm.year -1 )the corrosion rate j corr (A/cm 2 ) the corrosion current density M (g/mol) the atomic weight of the metal  Fe (g/cm 3 ) the density of the metal z (-) number of electrons of the dissolving metals

Results and discussion
Microstructure of the experimental material, shown in Fig. 1, is ferrite -pearlite with areas of line ordering grains.The surface state of steel samples (surface without protection and with corrosion layer) before exposure and after 6 hours exposure in the soil is documented in Figs. 2 -Fig. 5. On the surface without protection it can be seen the signs of corrosion attack after exposure in the soil electrolyte.The appearance of surface with corrosion products did not change significantly.
Time dependence E corr is shown in Fig. 6.On the steel surface without corrosion products the corrosion potential decreased from the initial value -499 mV to -685 mV after 60 minute exposure in the soil.This decrease was caused by anodic dissolution of the steel (according to   4.   materials with and without corrosion layers.Based on the corrosion tests of E corr measurements and also basic corrosion characteristics corr , i corr and R p by Tafel and Stern it may be concluded:  after only a short exposure time significant changes were observed on the samples without protection; whereas no visible changes were observed on the samples with corrosion products,  the corrosion potential E corr was stabilized at values which corresponded with formation of corrosion layer on the steel surface; the values of potential reached the value of -698 mV.Corrosion potential E corr of the samples with corrosion layer reached the value of + 26 mV,  results according to Tafel and Stern confirmed that the corrosion layer provided the sufficient barrier against corrosion.Polarization resistance R p of the sample with and without the corrosion layer had the value of 41.08 k.cm and 7.98 k.cm, respectively.

Fig. 2 Fig. 3 Fig. 4 Fig. 5 Fig. 6
Fig. 2 Steel surface state without corrosion layer before exposure in the soil

Fig. 11
Fig. 11 Stern plot of the sample with corrosion layer after 6 hours of exposure in the soil The measurements of polarization resistance R p based on Stern-Geary equation represent slope of the linear current j and potential E dependence at the zero current and lower potential than 20 mV.The current density, which is proportional to the inverse R p , provides instantaneous electrode corrosion rate i corr , using Stern-Geary Equations 7 -10 [21-23]:

Table 2
Chemical composition of EN 10028/2-92 steel used in corrosion tests, (wt %) The corrosion behaviour of the pipeline steel was determined in soil electrolyte by measuring of basic corrosion characteristics.The soil electrolyte was prepared in the ratio of 2:1, what basically means distilled water (volume) and soil (weight).Corrosion potential E SCE was established against that of the saturated calomel electrode (SCE) during exposure of samples in soil electrolyte, employing the digital voltmeter type of 5 ½ Digit Multimeter Agilent 34405 A. The characteristics of soil environment for corrosion tests are given in Table3.Corrosion potential E corr and corrosion rate i corr were determined on the basis of the relationship: corrosion current densitycorrosion potential, by means of potentiodynamic polarization.

Table 4
Measured corrosion characteristics according to Tafel and Stern Sample

E corr (mV) I corr (A.cm -2 ) Rp (kΩ.cm) CR (mm.year -1 )
Fig. 8 Tafel plot of the sample without corrosion layer after 6 hours of exposure in the soil