Influence of AC waveforms on stress corrosion cracking behaviour of pipeline steel in high pH solution

Abstract

The effect of various waveforms AC on stress corrosion cracking (SCC) behaviour and mechanism of X80 pipeline steel was investigated in high pH carbonate/bicarbonate solution by polarisation curves and slow strain rate tensile tests. Under various waveforms AC, the passivity of the steel degrades severely, and pitting corrosion easily forms on the electrode surface. With or without AC application, the SCC behaviour and mechanism of the steels in the solution are distinctly different. Without the presence of AC, the cracks propagation are intergranular, and the mechanism is anodic dissolution. AC significantly increases the SCC susceptibility of pipeline steel. The order of the SCC susceptibility of the steel at various waveforms AC is: sine wave < square wave < triangular wave. Under a long term AC interference, the fracture mode is transgranular, and the mechanism of the steel in high pH solution is controlled by anodic dissolution and hydrogen embrittlement.

Keywords

Pipeline steel AC waveforms Stress corrosion cracking

Introduction

Pipeline stress corrosion cracking (SCC) has been one of the most dangerous failure forms occurring in the buried pipelines used for high pressure natural oil/gas transmission, which usually results in sudden leakage and rupture.¹ It is well known that pipelines generally experience two forms of SCC, i.e. high pH^2–5 and near neutral pH SCC.^6,7 A lot of researchers have conducted on high pH SCC of pipelines in the past.^8–11 High pH SCC usually occurs in concentrated carbonate/biscarbonate electrolyte and causes intergranular cracking. Moreover, the SCC mechanism of ferritic steels is attributed to anodic dissolution.^12–20

The rapidly growing demand for energy requires the construction of an increasing number of high voltage, high power transmission lines and the laying of high pressure pipelines of large diameters. It becomes more difficult to install pipelines with adequate distance from overhead high voltage AC power transmission lines or AC powered rail transit system. Therefore, AC would transfer between the soil and the pipeline at the coating defects, leading to AC corrosion.²¹ Some cases have been reported^22–29 that the corrosion rate of metallic materials (including pipelines) can be accelerated due to the induced AC. A number of studies have been performed to investigating the AC induced corrosion of numerous metallic materials under various corroding systems.^30–36 For example, Lalvani et al. ^31,32 investigated the effects of potential waveforms on the corrosion of carbon steel including positive half cycle rectified, negative half cycle rectified and full wave sinusoidal potential. The results indicated that the reduction in corrosion products under negative half cycle current would increase the corrosion rate of carbon steel. It is reported^37,38 that localised corrosion (pitting corrosion) has been found on pipeline steels as a result of AC interference. It is well accepted that pitting corrosion can easily facilitate the initiation of SCC cracks. Therefore, the SCC failures of pipeline may be induced and facilitated due to AC interference. Our previous work³⁹ found that under a long term effect of AC current, the SCC susceptibility of X80 pipeline steel greatly increases with the increased AC current density. In addition, the mechanism is anodic dissolution with hydrogen embrittlement. To date, however, few studies have been performed on the related research. AC waveform is one of the impact factors that could affect the corrosion behaviour of metal. Herein, in this work, the SCC behaviour of X80 steel under a long term effect of various waveforms AC was investigated in high pH carbonate/bicarbonate solution using slow strain rate tensile (SSRT) tests and electrochemical measurements.

Experimental

Material and solution

All the test specimens are cut from a hot rolled plate of API X80 pipeline steel produced by Bao Steel Co. Ltd of China. The chemical composition of the steel is: C 0.070 wt-%, Si 0.216 wt-%, Mn 1.80 wt-%, P 0.0137 wt-%, S 0.0009 wt-%, Mo 0.182 wt-%, Cr 0.266 wt-%, Cu 0.221 wt-%, Ni 0.168 wt-%, Nb 0.105 wt-%, Al 0.026 wt-%, Ti 0.013 wt-%, V 0.001 wt-%, N 0.003 wt-% and Fe balance. In laboratory test, 0.5M Na₂CO₃+1M NaHCO₃ is usually used as the simulated solution for high pH SCC.¹⁷ In this work, this solution was used to investigate the SCC behaviour of X80 pipeline steel under various waveforms AC. The pH of the solution was ∼9.32.

Electrochemical measurement

Electrochemical measurements were carried out using PARSTAT2273 electrochemical workstation in a three-electrode cell system, where the steel specimen was used as working electrode, a platinum plate as counter electrode and a saturated calomel electrode (SCE) as reference electrode. The schematic diagram of the experimental set-up is identical with Ref. 39. A function generator of AT1645-3 model was used to provide the AC signal to the specimens. The various waveforms AC current with a frequency of 50 Hz was applied between the working electrode and graphite electrode, where a rheostat was used to control the AC current density of 100 A m^{− 2}. An inductor (4 H) was introduced to avoid the interference of the applied AC signal to electrochemical workstation, while a capacitor (25 V, 470 μF) was used to prevent the direct current to flow into the AC power. The AC current in this study was measured with a clamp meter and was root mean square value.

The test specimens were coated with an epoxy, leaving an exposure area of 10 mm × 10 mm as the working surface. The electrode was ground sequentially to 2000 grit emery paper, and then cleaned in distilled water and acetone. Before the test, the electrode was maintained for 30 min in the solution to ensure that a steady state value of corrosion potential was reached. Potentiodynamic polarisation curves were measured at a potential sweeping rate of 2 mV s^{− 1} under various waveform AC, starting from − 1.0 V (versus SCE) and ending at 1.1 V (DC SCE). The test was repeated at least three times.

Slow strain rate tensile tests

The SCC behaviour of the steel at various waveforms AC was investigated by SSRT test. The tests were performed through a WDML-30KN Materials Test System. The tensile specimen was made according to GB T 15970 specification.⁴⁰ Before testing, the gauge length of the specimen was polished to 2000 grit emery paper along the tensile direction, then degreased with acetone, followed by washing with deionised water and finally dried in air. Then, the specimen was soaked in the solution for 24 h before SSRT test. During the whole process of SSRT test, AC was applied between the sample and the platinum plate, and the tensile samples were conducted at a polarised potential of − 580 mV (versus SCE) with a strain rate of 1 × 10^{− 6} s^{− 1}. The electrochemical potential applied to the specimen was in the middle of the potential range causing IGSCC.⁴¹ The sample was maintained at the potential through a PS-12 potentiostat and a three-electrode system, where the sample was used as the working electrode, a platinum plate as counter electrode and a SCE as reference electrode. The test was performed at ambient temperature (∼22°C) and repeated at least three times. The experimental was used to simulate an actual situation that the buried pipeline may be interfered and affected by a long term existence of AC current, and then the SCC failure may more easily occur on the pipeline in carbonate/bicarbonate solution.³⁹ To investigate the SCC susceptibility of the steel under various waveforms AC, reduction in area I _Ψ and elongation loss rate I _δ were calculated.³⁹ After SSRT tests, the fracture surfaces of the steels were observed using scanning electron microscopy.

Results

Polarisation curves⁴²

Figure 1 shows the polarisation curves of X80 steel at various waveforms AC. As seen in Fig. 1, with various waveforms AC, the anodic current density of the sample increases significantly, the passive region significantly narrows, the critical pitting potential shifts negatively, the critical passivation potential shifts to more positive, the critical passivation current density increases and the passive current density sharply increases. The passive current density of the sample under various waveforms AC is different, and the specimen under sine wave has the smallest value. The narrow passive region indicates that AC can prevent the formation of thick passive films, which demonstrates that the SCC behaviour of the steel in the presence of AC may be affected by anodic dissolution. Moreover, a large negative shift of critical pitting potential due to AC application could result in pitting corrosion, which may affect the SCC behaviour.

Polarisation curves of X80 steel at various waveforms AC⁴²

Slow strain rate tensile tests

Figure 2 shows the stress–strain curves of X80 steel at various waveforms AC. When testing in air, there is the highest tensile strength and the largest elongation. The elongation of specimens at various waveforms AC is less than that at the absence of AC, and the order of the elongation of specimens under various waveforms AC is: triangular wave < square wave < sine wave, which indicate that AC application increases the SCC susceptibility of X80 steel in the simulated solution. Especially, the applied AC current of triangular wave has the highest SCC susceptibility, while the specimen tested under sine wave AC has the lowest one.

Stress–strain curves of X80 steel at various waveforms AC

Figure 3 shows AC waveform dependence of the SCC susceptibility of steel. It is seen that both I _δ and I _ψ of specimens under AC superimposition of various waveforms are considerably larger than that of sample without the presence of AC. The order of the SCC sensitive parameters of the specimens under various waveforms AC is: sine wave < square wave < triangular wave. The value of I _δ of samples tested at all three waveforms is >37%, and the value of I _Ψ is larger than 70%, which demonstrates that the fracture characteristic is brittle fracture.⁴³ The result indicates that the steel under AC application of various waveforms has high SCC susceptibility, and the mechanism of the steel may be changed, comparing with that without AC application.

AC waveforms dependence of SCC susceptibility

Figure 4 shows the main fractographs of X80 steel at various waveforms AC. There is a marked difference in fractographs between specimen tested at the absence of AC and those tested at various waveforms AC. It is clearly seen that both the fractographs of the air testing of tensile sample and the specimen tested without AC application show the characteristic of dimple typically, with the remarkable necking phenomenon, indicating that the fracture belong to ductile fracture. In contrast, the samples with various waveforms AC have even macroscopic fracture, slight necking extent and flat microscopic dimple. Moreover, the fractographs of the sample exhibit some features of quasi-cleavage fracture, with some quasi-cleavage plane, which displays a distinct characteristic of brittle fracture. This indicates that the applied AC current increases the SCC susceptibility of pipeline steel. In a further observation of the microfractographs (Fig. 4f, h and j), there are many vertical cracks caused by the infiltration of hydrogen into the steel in the fracture surface, which facilitate the initiation and propagation of SCC cracks. Furthermore, high reduction in area I _Ψ and elongation loss rate I _δ value (Fig. 3), combined with the obvious characteristics of brittle fracture (Fig. 4), demonstrate that there is an effect of hydrogen embrittlement, which plays a key role in the SCC process of pipeline steels with various waveforms AC.

a,b in air, c,d without AC; e,f triangular wave; g,h square wave; i,j sine waveMain fractographs of X80 steel at various waveforms AC

To further analyse the SCC susceptibility of the steel tested at various waveforms AC in high pH solution, the crack morphology of the side near fracture surface and the crack propagation mode of cross-section fractograph are carefully observed. As seen from Fig. 5, there are significant differences in the crack morphology of the side near fracture surface for steels with or without AC application. When the specimen was tested in the absence of AC (Fig. 5a), numerous narrow and shallow cracks exist in the side near fracture surface, and the propagation path of cracks is mainly flexural, which is in accord with the characteristics of the classical intergranular crack. When AC is applied to the specimens, there are a large number of wide and deep cracks, and the crack propagation is linear, which is analogous to the feature of the typical TGSCC crack of pipeline steel. As seen from Fig. 5b, some SCC cracks are very wide, deep and long. This indicates that the resistance of crack propagation is relatively small, and the SCC susceptibility of the steel at triangular wave AC is high.

a without AC; b triangular wave; c square wave; d sine waveMorphologies of side near fracture surface of X80 steel at various waveforms AC

Figure 1 indicates that pitting corrosion is easily formed on the surface of the steel due to AC application, which is in accord with the result of other researchers.⁴⁴ A large number of research works demonstrate that the SCC cracks can be initiated from corrosion pits.⁴⁵ Fig. 6 shows the morphologies of the side near fracture surface of X80 steel at various waveforms AC. In the presence of AC, as seen from Fig. 6b and d, corrosion pits form on the steel, and some cracks are initiated from the bottom of corrosion pits, indicating that corrosion pits induced by AC facilitate the initiation and propagation of the cracks. At the absence of AC, it is seen in Fig. 6a that the cracks show an obvious intergranular propagation path, which is not associated with the corrosion pit. A further observation of Fig. 6b and d shows that there is a phenomenon of the removal of grains. This indicates that anodic dissolution has a certain influence on the SCC behaviour of the steel tested with AC application. Furthermore, the cracks are wide and deep, and the propagation path is linear, which are similar to the characteristics of transgranular SCC (TGSCC) crack. Therefore, the SCC mechanism of X80 steel under AC application in high pH solution is the combination of anodic dissolution with hydrogen embrittlement.³⁹

a without AC; b triangular wave; c square wave; d sine waveMorphologies of side near fracture surface of X80 steel at various waveforms AC

Figure 7 shows the cross-section fractographs of X80 steel at various waveforms AC. It is seen that there is a significant difference in the crack propagation mode of steels with or without AC. Without the application of AC current, as seen from Fig. 7a, the cracks propagation shows an obvious characteristic of intergranular (IGSCC) fracture, which is in very good agreement with the classical IGSCC fracture.⁴⁶ When various waveforms AC current is applied, the crack propagation path is transgranular (TGSCC). It is generally considered that near neutral pH SCC (TGSCC) involves anodic dissolution and the ingress of hydrogen into steel.^47,48 In this paper, the SCC behaviour of the steel under various waveforms AC is analogous to that of TGSCC fracture, indicating that the mechanism is controlled by anodic dissolution and hydrogen embrittlement.

a without AC; b triangular wave; c square wave; d sine waveCross-section fractographs of X80 steel at various waveforms AC

Discussion

According to the mentioned experiment result, various waveforms AC applied to the steel, resulting in a big change in the SCC behaviour and mechanism. The detailed analysis of the result is as follows³⁹:

At the absence of AC application, the potential of − 580 mV (versus SCE) applied to the sample during the SSRT test is in the transition region of the polarisation curves (Fig. 1),⁴² which indicates that the surface passive film of the sample is under the combined effect of formation and dissolution. As seen in Figs. 5a, 6a and 7a, the path of cracks propagation is intergranular, which is in accord with the classical IGSCC. This indicates that the mechanism is attributed to anodic dissolution, which is certified by much research.^49,50

Under AC application, the potential of − 580 mV (versus SCE) applied to the electrode is in the activated region of the polarisation curves (Fig. 1), indicating that the samples are affected by active anodic dissolution. As seen from Fig. 6b and d, the obvious removal of grains demonstrates that the SCC behaviour of the steel is affected by anodic dissolution at a certain extent. While Figs. 2 to 7 show that the SCC susceptibility of the steel greatly increases due to various waveforms AC, moreover, an obvious characteristic of brittle fracture, and the crack propagation is transgranular, which indicates that hydrogen embrittlement plays a significant role in the SCC process of steel with AC superimposition. Therefore, the SCC mechanism is mixed controlled by both anodic dissolution and hydrogen embrittlement.⁵¹

Under the application of various waveforms AC, the relative steady cathode current value is read via the potentiostat. The testing results are − 3.2 mA (triangular wave), − 2.6 mA (square wave) and − 1.5 mA (sine wave) respectively. Thus, the hydrogen evolution reaction under triangular wave is the most intense, and then the most amount of H atoms infiltrate into the steel. Consequently, the hydrogen embrittlement sharply increases, and the SCC susceptibility of the steel tested under triangular wave is the highest, next the square wave, the least sine wave.

According to the above analysis, when various waveforms AC applied to the specimens, it is beneficial to accelerate hydrogen evolution reaction, and the diffusion of hydrogen into the steel would be facilitated.³⁹ Hydrogen enrichment would occur in the defects of the steel, and the high stress concentration region, for instance crack tip, and then, the cracks are generated. Therefore, the ingress of hydrogen could encourage the initiation and propagation of cracks (Figs. 6 and 7). Moreover, the beneficiation of hydrogen would easily occur at the pits induced by various waveforms AC. With the stress concentration caused by the pits, it would easily facilitate the crack initiation and propagation (Fig. 6). In addition, the decrease in pH value usually occurs at the bottom of pits, resulting in the cracks initiation. Consequently, the steel under AC application has higher SCC susceptibility (Figs. 3 and 4).

Conclusions

In the presence of various waveforms AC, the passivity of the steel degrades severely. With or without various waveforms AC, there is an obvious difference in the SCC behaviour and mechanism of the steel. Without the presence of AC, the cracks propagation is intergranular and the mechanism is anodic dissolution. The SCC susceptibility of the steel is significantly enhanced due to the interference of a long term AC current. The order of the SCC susceptibility of the steels at various waveforms AC is: sine wave < square wave < triangular wave. The cracks propagation is transgranular, and the mechanism of the steel at various waveforms AC is controlled by anodic dissolution and hydrogen embrittlement.

Acknowledgements

The authors are grateful for the financial support from the National Natural Science Foundation of China (nos. 51371036 and 51131001) and the Science Foundation of Zhejiang Sci-Tech University.

References

van Boven

Chen

and Rogge

: Acta Mater., 2007 55 29. doi: 10.1016/j.actamat.2006.08.037

Zhang

G. A.

and Cheng

Y. F.

: Corros. Sci., 2010 52 960. doi: 10.1016/j.corsci.2009.11.019

Mustapha

Charles

E. A.

and Hardie

: Corros. Sci., 2012 54 5. doi: 10.1016/j.corsci.2011.08.030

Oskuie

A. A.

Shahrabi

Shahriari

and Saebnoori

: Corros. Sci., 2012 61 111. doi: 10.1016/j.corsci.2012.04.024

Aran

M. A.

and Szpunar

J. A.

: Mater. Sci. Eng. A 2011 A528 4927.

Kang

Y. W.

Chen

W. X.

Kania

Boven

G. V.

and Worthingham

: Corros. Sci., 2011 53 968. doi: 10.1016/j.corsci.2010.11.029

B. T.

Luo

J. L.

and Norton

P. R.

: Corros. Sci., 2010 52 1787. doi: 10.1016/j.corsci.2010.02.020

SadeghiMeresht

ShahrabiFarahani

and Neshati

: Eng. Fail. Anal., 2011 18 963. doi: 10.1016/j.engfailanal.2010.11.014

Jack

T. R.

Erno

Krist

and Fessler

: 2000

NACE, 362

Proc. CORROSION′00, Houston, TX, USA.

10.

Manfredi

and Otegui

J. L.

: Eng. Fail. Anal., 2002 9 495. doi: 10.1016/S1350-6307(01)00032-2

11.

Arafin

M. A.

and Szpunar

J. A.

: Corros. Sci., 2009 51 119. doi: 10.1016/j.corsci.2008.10.006

12.

Wang

J. Q.

and Atrens

: Corros. Sci., 2003 45 2199. doi: 10.1016/S0010-938X(03)00044-1

13.

Parkins

R. N.

: Corrosion 1987 43 130. doi: 10.5006/1.3583125

14.

Parkins

R. N.

Belhimer

and Blanchard

W. K.

: Corrosion 1993 49 951. doi: 10.5006/1.3316023

15.

Pikey

A. K.

Lambert

S. B.

and Plumtree

: Corrosion 1995 51 91. doi: 10.5006/1.3293588

16.

Fang

B. Y.

Atrens

Wang

J. Q.

Han

E. H.

Zhu

Z. Y.

and Ke

: J. Mater. Sci., 2003 38 127. doi: 10.1023/A:1021126202539

17.

Parkins

R. N.

: Corrosion 1996 52 363. doi: 10.5006/1.3292124

18.

B. T.

Song

Gao

and Elboujdaini

: Corros. Sci., 2010 52 4064. doi: 10.1016/j.corsci.2010.08.023

19.

M. C.

and Cheng

Y. F.

: Electrochim. Acta 2008 53 2831. doi: 10.1016/j.electacta.2007.10.077

20.

Song

F. M.

: Corros. Sci., 2009 51 2657. doi: 10.1016/j.corsci.2009.06.051

21.

Hosokawa

Kajiyama

and Fukuoka

: Corrosion 2004 60 408. doi: 10.5006/1.3287751

22.

Tan

T. C.

and Chin

D. T.

: J. Appl. Electrochem., 1988 18 831. doi: 10.1007/BF01016038

23.

Goidanich

Lazzari

Ormellese

Pedeferri

, ; 2005 NACE, 189. Proc. CORROSION′05, Houston, TX, USA.

24.

Wendt

J. L.

and Chin

D. T.

: Corros. Sci., 1985 25 889. doi: 10.1016/0010-938X(85)90019-8

25.

NACE Standard RP0169-96: ‘Control of external corrosion on underground or submerged metallic piping’; 1996

Houston, TX

NACE.

26.

Hanson

H. R.

Smart

; 2004 NACE, 209. Proc. CORROSION′04, Houston, TC, USA.

27.

Nielsen

L. V.

Nielsen

K. V.

Baumgarten

Breuning-Madsen

Cohn

Rosenberg

, ; 2004 NACE, 211. Proc. CORROSION′04, Houston, TX, USA.

28.

Wakelin

R. G.

Sheldon

, ; 2004 NACE, 205. Proc. CORROSION′04, Houston, TX, USA.

29.

Y. T.

Cai

G. W.

and Yang

L. H.

: Corros. Eng. Sci. Technol., 2013 48 322. doi: 10.1179/1743278212Y.0000000076

30.

Qiu

W. W.

Pagano

Zhang

and Lalvani

S. B.

: Corros. Sci., 1995 37 97. doi: 10.1016/0010-938X(94)P4303-X

31.

Lalvani

S. B.

and Zhang

: Corros. Sci., 1995 37 1567. doi: 10.1016/0010-938X(95)00066-S

32.

Lalvani

S. B.

and Zhang

: Corros. Sci., 1995 37 1583. doi: 10.1016/0010-938X(95)00056-P

33.

Song

Kim

Y. G.

Lee

S. M.

Kho

Y. T.

, ; 2002 NACE, 117. Proc. CORROSION′02, Houston, TX, USA.

34.

Lalvani

S. B.

and Lin

: Corros. Sci., 1994 36 1039. doi: 10.1016/0010-938X(94)90202-X

35.

Lalvani

S. B.

and Lin

: Corros. Sci., 1996 38 1709. doi: 10.1016/S0010-938X(96)00065-0

36.

Wakelin

R. G.

Gummow

R. A.

Segall

S. M.

, ; 1998 NACE, 565. Proc. CORROSION′98, Houston, TX, USA.

37.

A. Q.

and Cheng

Y. F.

: Corros. Sci., 2010 52 612. doi: 10.1016/j.corsci.2009.10.022

38.

Linhardt

Ball

, ; 2006 NACE, 160. Proc. CORROSION′06, Houston, TX, USA.

39.

Zhu

C. W.

X. G.

Liu

Z. Y.

and Zhang

D. W.

: Corros. Sci., 2014 87 224. doi: 10.1016/j.corsci.2014.06.028

40.

GB T 15970, ‘ Chinese National Standard for Stress Corrosion Cracking Tests’; 2007.

41.

Wang

Z. F.

and Atrens

: Metall. Mater. Trans. A 1996 27A 2686. doi: 10.1007/BF02652362

42.

Zhu

C. W.

X. G.

Liu

Z. Y.

and W X. G.: Mater. Corros., 2015 presto be published.

43.

Liu

Z. Y.

and Li

X. G.

: J.Chin. Soc. Corros. Prot., 2006 26 360 (in chinese).

44.

Goidanich

Lazzari

and Ormellese

: Corros. Sci., 2010 52 916. doi: 10.1016/j.corsci.2009.11.012

45.

Eslami

Kania

Worthingham

Boven

G. V.

Eadie

and Chen

: Corros. Sci., 2011 53 2318. doi: 10.1016/j.corsci.2011.03.017

46.

Elboujdaini

Fang

Revie

R. W.

and Phaneuf

M. W.

: Corrosion 2006 62 316. doi: 10.5006/1.3280664

47.

M. C.

and Cheng

Y. F.

: Electrochim. Acta 2007 52 8111. doi: 10.1016/j.electacta.2007.07.015

48.

Parkins

R. N.

Delanty

W. K.

Jr. and Blanchard

B. S.

: Corrosion 1994 50 394. doi: 10.5006/1.3294348

49.

Cheng

Y. F.

: J. Mater. Sci., 2007 42 2701. doi: 10.1007/s10853-006-1375-y

50.

Sanchez

Fullea

Andrade

and Alonso

: Corros. Sci., 2007 49 4069. doi: 10.1016/j.corsci.2007.05.025

51.

Luo

and Mao

: Corrosion 1999 55 96. doi: 10.5006/1.3283971

Influence of AC waveforms on stress corrosion cracking behaviour of pipeline steel in high pH solution

Abstract

Keywords

Introduction

Experimental

Material and solution

Electrochemical measurement

Slow strain rate tensile tests

Results

Polarisation curves 42

Slow strain rate tensile tests

Discussion

Conclusions

Acknowledgements

References

Polarisation curves⁴²