- Email: [email protected]
Effect of hydrogen concentration on fatigue crack growth behaviour in pipeline steel
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 6 ) 1 e8
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevier.com/locate/he
Effect of hydrogen concentration on fatigue crack growth behaviour in pipeline steel I.M. Dmytrakh*, R.L. Leshchak, A.M. Syrotyuk, R.A. Barna Karpenko Physico-Mechanical Institute of National Academy of Sciences of Ukraine, 5 Naukova Street, 79060, Lviv, Ukraine
article info
abstract
Article history:
The ambiguous relationship between fatigue crack growth rate and hydrogen concentra-
Received 23 September 2016
tion CH in the bulk of metal under cyclic loading of the ferrite-pearlite low-alloyed steel in
Received in revised form
hydrogen-contained environments has been found: there is a certain CH value at which the
15 November 2016
crack growth resistance of steel increases. At these test conditions fracture surface dem-
Accepted 29 November 2016
onstrates some increasing of the plastic component on relief. The received results point on
Available online xxx
key importance of the determining of such threshold hydrogen concentration values, which correspond the transition “enhanced plasticity e embrittlement”.
Keywords:
© 2016 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.
Pipeline steel Hydrogen charging of metal Hydrogen concentration Cyclic loading Enhanced plasticity Hydrogen embrittlement
Introduction The phenomenon of hydrogen embrittlement of metallic materials, which lead to the loss of their plasticity, decreasing of fracture toughness and degradation of fatigue properties is well-known and the numerous records of such effects can be found in the literature [1e3]. Despite of that, there is the number of works where the positive hydrogen effect, i.e. the hydrogen-induced plasticity, was shown [4e14]. For example, in work [12] the dramatic phenomenon was found in which charging a supersaturated level of hydrogen into specimens of austenitic stainless steels of types 304 and 316L drastically improved the fatigue crack growth resistance, rather than accelerating fatigue crack growth rates. Results presented in
Ref. [14] confirmed the existence of a hydrogen-induced plasticity effect within a particular range of cathodic polarization of X70 pipeline steel. The possible explanation for the contradictory results of hydrogen effects on macroscopic deformation and the model to interpret the criterion for the application of local softening concept can be found in work [13]. Therefore it is important and actual problem to study and to clarify all aspects of specific hydrogen effects of metallic structural materials with the aim to establish the conditions for the transition “enhanced plasticity e embrittlement”. This may be the crucial key factor for deeper understanding of the hydrogen influence mechanism and also it will promote to review the existed conceptions where the hydrogen is
* Corresponding author. Fax: þ380 322 649427. E-mail address: [email protected] (I.M. Dmytrakh). http://dx.doi.org/10.1016/j.ijhydene.2016.11.193 0360-3199/© 2016 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved. Please cite this article in press as: Dmytrakh IM, et al., Effect of hydrogen concentration on fatigue crack growth behaviour in pipeline steel, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.11.193
2
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 6 ) 1 e8
considered as a negative agent. Some disputable aspects of this problem are currently discussed and considered [15e18]. Within the frame of above-mentioned problem the presented paper is dedicated to evaluation of the fatigue behaviour of hydrogen charged pipeline steel under known hydrogen concentration in a bulk of metal. Here the parameters of fatigue crack growth resistance of steel were assessed as the function of hydrogen concentration. This work can be considered as further development of our previous work [19] where the specificity of strain behaviour of pipeline steel depending on the hydrogen concentration in bulk of metal was studied. There the existence of some characteristic value of the hydrogen concentration at which the mechanism of hydrogen influence changes, namely: below this value the enhanced plasticity (decreasing of the yield stress value) takes places and above e the hydrogen embrittlement occurs, was shown.
Experimental procedure The object of study was low alloyed pipeline steel (sY ¼ 260 MPa and sU ¼ 440 MPa) with nominal chemical composition (in weight %): C ¼ 0.17e0.24; Si ¼ 0.17e0.37; Mn ¼ 0.35e0.65; S < 0.04; remainder Fe. This material consists of grains of ferrite-pearlite, typical for all carbon steels (Fig. 1). The rectangular cross-section beam specimens (Fig. 2) were manufactured with real pipes, which were supplied from two different manufacturers (test series A and test series B). The longitudinal cracks were studied and cut off of specimens from pipe was corresponded to this case. For realisation of experimental studies on the hydrogen charging of specimens, determination of the hydrogen concentration in a bulk of steel and the fatigue crack growth under joint action of the cyclic loading and hydrogenation environmental conditions the special testing stand was developed, which based on the fatigue testing machine for pure bending of specimens at the environmental conditions [20] and the dynamic electrochemical laboratory VoltaLab40 [21]. The general view of developed facility is presented in Figs.
Fig. 1 e Structural specificity of studied pipeline steel (x 1000).
Fig. 2 e Geometry of the beam specimen. 3 and 4. Here the standard three-electrode electrochemical cell was used where the auxiliary (counter) electrode consists of four round bars. The mutual location of working (specimen) and auxiliary (counter) electrodes is given in Fig. 5. The hydrogenation of specimens was made by electrochemical method under cathodic polarisation at some constant potential Ecath ¼ const. With the aim to simulate the hydrogen entry at real operating conditions of the buried pipeline, the following procedure has been applied [22e26]. The special deoxygenated, near-neutral pH NS4 solution, which is the model of underground water, was chosen as the electrolyte for hydrogen charging of steel. The chemical composition of the NS4 solution is given in Table 1. Taking into account the situation of freely corroding system that exists for the real pipeline, the potential of polarisation Ecath was slightly more negative than the free corrosion potential Ecorr for given steel, i.e.: Ecorr ¼ 600 mV (SCE) and Ecath ¼ 800 mV (SCE). Hydrogen concentration in a bulk of steel has been determined on the base of hydrogen discharging process under anodic polarisation with using of the hydrogen electrochemical oxidation method proposed in work [27]. The detailed description of the application of this method for the pipelines hydrogenation problems can be found in works [27,28]. As preliminary stage of study the experimental dependence “hydrogen concentration CH in specimen e time of exposure t” was received. Following to work [26] these experimental data were described by power relation:
Fig. 3 e Special fatigue testing stand.
Please cite this article in press as: Dmytrakh IM, et al., Effect of hydrogen concentration on fatigue crack growth behaviour in pipeline steel, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.11.193
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 6 ) 1 e8
Fig. 4 e Electrochemical cell.
Fig. 5 e Mutual location of beam specimen and auxiliary (counter) electrodes in electrochemical cell for fatigue test.
CH ¼ A$106$tm, [mol/cm3]
(1)
where A and m are some constants that depend on system “material e environment” and testing conditions. For our case the values of these constants are: A ¼ 0.28,106 and m ¼ 0.65. This analytic relation served as the reference curve under hydrogen charging of the specimens for fatigue test. The specimens, which hydrogen charged to assigned level of CH (CH ¼ 0.001; 0.209 and 0.456 ppm), were subjected to fatigue loading under pure bending conditions in NS4
Table 1 e Chemical composition of NS4 solution (g/l) [26]. NaHCO3
KCl
CaCl2
MaCl2$H2O
0.483
0.120
0.137
0.131
3
solution. Here the cathodic polarisation at Ecath ¼ 800 mV (SCE) was applied during all time of test. The rectangular cross-section beam specimens (Fig. 2) with initial edge crack of length a0 z 2.5 mm were subjected by cyclic pure bending with frequency f ¼ 1 Hz under stress ratio R ¼ 0. The stress intensity factor K in the crack tip for such specimen in case of its loading by pure bending was calculated from the known formula [20]. The initial level of loading, i.e. the initial range of stress intensity factor, was the same for all pffiffiffiffiffi tested specimens: DK0 z11 MPa m. The details of the test procedure are the following. As environment is transparent the crack lengths were measured on a plane surface of specimens by the optical microscope with accuracy ±0.005 mm. The specimens were subjected by pure bending under constant displacement mode. The subjects of tests were pre-cracked specimens with initial edge cracks of length a0 z 2.5 mm. The initial cracks were nucleated in air conditions (f ¼ 12 Hz and R ¼ 0) at final stress inpffiffiffiffiffi tensity factor range DKi z9:5 MPa m that is lower than the initial range of stress intensity factor for the main tests in pffiffiffiffiffi environment (DK0 z11 MPa m). Such choice of test conditions can be explained by following. As it is known from literature records [29] the most significant environmental effects can be observed at stress ratio R ¼ 0. The frequency of cyclic loading f is also very important parameter and it is accepted that fatigue crack growth rate increases with decreasing of frequency. However as it was shown [30] there is some maximum in the curve da/ dN ¼ F(f) and for considered steel this maximum corresponds to value fy1:0 Hz. From these reasons the conditions of fatigue test were chosen as R ¼ 0 and f ¼ 1.0 Hz for receiving the maximal environmental effect on fatigue crack growth rate. Resulting data under different initial hydrogen concentrations in the bulk of specimens were received as the sequence of the following parameters: number cycles of loading Ni, crack length ai, fatigue crack growth rate (da/aN)i, and stress intensity factor rang DKi. After fatigue tests the fracture surfaces of specimens were examined by the scanning electron microscopy EVO-40XVP [31] and some number of special selected images were received. For quantitative evaluation of the fracture surface relief the images were digitally processed with the aim to detect the possible difference in fracture mechanisms under different hydrogen concentration in the steel. Taking into account the complicated inhomogeneous relief of fracture surfaces, because the fatigue crack growth mechanism is gradually mixed (tearing and shearing), the special technique for computer analysis of characteristic elements of fractographic images was applied [32,33]. This method is based on the statement that for each fracture mechanism the own typical features of image's structure are inherent [32]. In particular, the ductile fracture characterizes with image's structure with the closed contours, in the form of ovals. This is the ground for procedure of image segmentation, which uses the multilevel determination of the brightness threshold [32]. Here, with the aim to get the less distortion of the segmenting image not only the large details were used but also the small details with low peaks in the histogram of brightness were taken into account. It enables to separate the image fragment of the fracture surface into
Please cite this article in press as: Dmytrakh IM, et al., Effect of hydrogen concentration on fatigue crack growth behaviour in pipeline steel, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.11.193
4
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 6 ) 1 e8
connected regions, and then to determine and analyze their quantitative parameters. The processing of received images was focused on the detection of sites where the shear fatigue crack growth mechanism was occurred. Each image had the square of 786,432 pixels or S ¼ 42,219 mm2. Using the above-mentioned procedure the corresponding sites were detected (see below Fig. 10 where these sites are marked with blue rectangles). It should be noted that the real area of ductile fracture is bounded of the fracture relief and it is smaller than the blue rectangle square. Received data were used for calculation of the portion of fracture surface area with the shear fatigue crack growth mechanism. The detailed description of applied method is given in Refs. [32,33].
Results and discussion Two series of fatigue tests (A and B) were conducted under different initial hydrogen concentration in the bulk of specimens. Primary test results were presented as the separate scatter plots of fatigue crack growth rate da/dN versus stress intensity factor DK (Fig. 6). The comparison of these data showed on some small difference between series A (Fig. 6a) and B (Fig. 6b) that can be considered as natural scattering, which is always inherent for such tests. However, when we impose these two graphs on each other a full coincidence of data can be seen for both tests at CH ¼ 0.209 ppm (red dots). It indicates on the ambiguous relationship between fatigue crack growth rate da/dN and hydrogen concentration CH for two series of tests. For comparative assessment and analysis of results these scatter plots were described analytically with using the wellknown Paris equation [34]:
da/dN ¼ C$(DK)n;
(2)
where C and n are the constants of material and test conditions. The values of constants C and n in equation (2) and mean square deviation r2 are given in Table 2. Fig. 7 presents an analytic description fatigue crack growth rate diagrams by power function (2). The received results showed on the ambiguous effect of an initial hydrogen concentration in specimens on fatigue crack growth in steel for both test series. For given pipeline steel there is some characteristic value of the hydrogen concentration CH y0:209 ppm, which causes the decreasing of fatigue crack growth rate and, consequently, leads to increasing of fatigue resistance of material. Especially, it can be clear seen when we pffiffiffiffiffi make the section of diagrams at DK ¼ 20 MPa$ m (Fig. 8) and at da/dN ¼ 104 mm/cycle (Fig. 9). Such tendency may be explained [12] by interaction between hydrogen and dislocations. The hydrogen affects on dislocations by two ways: promotes of their pinning (or dragging) or enhances of their mobility. Competition between these two roles defines whether the resulting phenomenon is positive or negative with point of view of fatigue crack growth resistance of material. It is important to note that above-presented results are restricted by Paris region and their analysis is valid only to this part of the fatigue crack growth rate diagram [34]. For consideration of the threshold zone as well as the final failure zone the special study is required and it can be the subject of further research. The fracture surface examination was made for the restricted area, which correspond of the range of stress inpffiffiffiffiffi tensity factor DK ¼ 20 MPa$ m, i.e. for conditions were the CH effect is the most significant on fatigue crack growth rate (see Figs. 7 and 8). The visual observation of received images showed on the modification of fracture surface depending on the hydrogen concentrations CH in bulk of material (Fig. 10). For very low hydrogen charged specimen (CH ¼ 0.001 ppm) the surface has inhomogeneous relief indicating a mixed mechanism of fatigue crack growth: tearing and shearing (Fig. 10a). Here also the relief of typical quasicleavage brittle
Fig. 6 e Fatigue crack growth rate diagrams under different initial hydrogen concentrations in the bulk of specimens: a e test series A; b e test series B. Please cite this article in press as: Dmytrakh IM, et al., Effect of hydrogen concentration on fatigue crack growth behaviour in pipeline steel, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.11.193
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 6 ) 1 e8
5
Fig. 7 e Analytic description fatigue crack growth rate diagrams by power function (Paris equation): a e test series A; b e test series B.
Fig. 8 e Dependence of fatigue crack growth rate da/dN at pffiffiffiffiffi DK ¼ 20 MPa$ m on the initial hydrogen concentrations CH in material.
fracture can be also found. With increasing of the hydrogen concentrations CH in material (CH ¼ 0.209 ppm) the increasing of the homogeneity of surface relief can be seen due to increasing of area where the shear fatigue crack growth mechanism was occurred (Fig. 10b). It points on increasing of the plastic deformation under fatigue process of steel. Further increasing of the initial hydrogen concentration in specimen (CH ¼ 0.514 ppm and CH ¼ 1.231 ppm) leads to increasing of the heterogeneity of surface relief and to appearance of some traces of tearing and quasicleavage, i.e. the fatigue crack growth mechanism becomes gradually mixed again (Fig. 10c and d). The digital processing of these images with using the special technique [32,33] gave the possibility to recognize the sites where the shear fatigue crack growth mechanism were realized. In Fig. 10 these sites are marked with blue rectangles. From these results it can be seen that the area marked with blue rectangles in Fig. 10b is visibly bigger than in Fig. 10a, c
Fig. 9 e Dependence of stress intensity factor value DK at da/dN ¼ 10¡4 mm/cycle on the initial hydrogen concentrations CH in material. and d. It indicates on the predominance of the shear fatigue crack growth mechanism at CH ¼ 0.209 ppm in comparison with others considered cases. For final presentation of received data we used the parameter k, which is the portion of image area (in percents to total area) where the shear fatigue crack growth mechanism was realized. The dependence of parameter k on the of the initial hydrogen concentrations CH in steel is shown in Fig. 11. It can be clear seen that this plot has the maximum at the hydrogen concentration CH y0:209 ppm that can serve as additional justification of the received findings. These data and their analysis give the basis for the following statement: there is a certain value of the hydrogen concentration in a bulk of material CH ¼ C*H , at which the crack growth resistance of steel increases and the diagrams of fatigue crack growth rate are shifted to higher values of stress intensity factor. For given pipeline steel C*H ¼ 0:209 ppm. It should be noted that value is close to early established characteristic value of the hydrogen concentration at which the
Please cite this article in press as: Dmytrakh IM, et al., Effect of hydrogen concentration on fatigue crack growth behaviour in pipeline steel, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.11.193
6
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 6 ) 1 e8
pffiffiffiffiffi Fig. 10 e Specificity of fracture surface during fatigue crack growth in pipeline steel at DK ¼ 20 MPa$ m on dependence of the initial hydrogen concentrations CH in material: a e CH ¼ 0.001 ppm; b e CH ¼ 0.209 ppm; c e CH ¼ 0.514 ppm; d e CH ¼ 1.231 ppm (x 500). The sites of the shear fatigue crack growth mechanism detection are marked with blue rectangles. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
mechanism of hydrogen effect on steel deformation under static loading changes: enhanced plasticity or embrittlement [19]. There it was found that C*H ¼ 0:227 ppm. As a concluding remark it may be noted that the determining of threshold values C*H , which correspond the transition “enhanced plasticity e embrittlement” can be considered as one of the key factors for the assessment of serviceability of structural materials in the presence of hydrogen-contained environments. This is also very important to underline that of the hydrogen effect on mechanical behaviour of structural
metallic materials is multiple and several mechanisms can be realized, which may affect independently or coexist simultaneously. Here, first of all, the recent works [35,36], should be marked where the simultaneous action of the hydrogenenhanced decohesion (HEDE) and hydrogen enhanced
70,00 ΔK = 20 MPa(m)
1/2
= const.
Table 2 e Values of the parameters n and C in Paris equation under different initial hydrogen concentrations CH in the bulk of specimens. " # CH, ppm Test series A 0.001 0.209 0.456 Test series B 0.001 0.209 0.514 1.231
n
C,
mm=cycle pffiffiffiffi ðMPa$ m Þn
4.71 4.86 4.29
1$1010 5$1011 5$1010
5.55 5.16 5.92 5.66
11
1$10 2$1011 4$1012 9$1012
r2
0.96 0.97 0.96
k, %
65,00
60,00
55,00 CH=0,209 ppm
50,00 0
0,5
1
1,5
CH, ppm 0.96 0.96 0.98 0.99
Fig. 11 e Fracture surface characterisation parameter k as the function of the initial hydrogen concentrations CH in pffiffiffiffiffi steel (DK ¼ 20 MPa$ m).
Please cite this article in press as: Dmytrakh IM, et al., Effect of hydrogen concentration on fatigue crack growth behaviour in pipeline steel, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.11.193
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 6 ) 1 e8
localized plasticity (HELP) was shown for low carbon structural steel depending on the local hydrogen concentration in metal. Such results are the base for the development of corresponding engineering models with the aim to predict and to prevent the hydrogen degradation of the structural components under aggressive operating conditions [37]. At the end it should be stated that definition “critical concentration” is often used in studies of problems of hydrogen in metals and alloys [1,38e40]. Although in different works this term has different physical sense. For example, according to work [38], hydrogen treatment below the “critical” content was found to cause the substantial rearrangement of dislocations and de-cohesion of grain boundaries and in overcritical condition, the formation of micro crevices at the grain and phase boundaries has occurred. Others authors use this definition under developing of local fracture criteria [1,39]. In Ref. [41] the term “critical concentration” defines the hydrogen concentration level at which the critical loss of local strength of the material at the notch is occurred. In Ref. [37] the special diagram for structural integrity analysis of boiler pipes exposed to hydrogen damage is proposed and where the hydrogen concentration is the basic parameter. Here the definition “critical concentration” is associated with the increased activity and activation of the HEDE mechanism (HEDE > HELP). Above mentioned results and data confirm the principal importance to take into account the hydrogen concentration in all considerations where the problems of hydrogen degradation of materials are investigated.
Conclusions The specificity of fatigue behaviour of pipeline steel depending on the initial hydrogen concentration in bulk of metal CH was studied. The ambiguous relationship between fatigue crack growth rate and hydrogen concentration CH in the bulk of metal under cyclic loading of the ferrite-pearlitic low-alloyed steel in hydrogen-contained environments has been found. It was shown that there is a certain CH value at which the crack growth resistance of steel increases and the diagrams of fatigue crack growth rate are shifted to higher values of stress intensity factor. This value correlates with early established characteristic value of the hydrogen concentration at which the mechanism of hydrogen effect on steel deformation under static loading changes: enhanced plasticity or embrittlement. The received results point on the fundamental importance of such characteristic (threshold) values of hydrogen concentration in structural steels for the evaluation their strength and serviceability under assigned operating conditions.
Acknowledgements Presented study was conducted within the project number 10ІІІ-108-15 of Fundamental Research Programme of National Academy of Sciences of Ukraine (2015e2017). The authors also
7
would like to thank to Dr. R. Kosarevych for his valuable help in digital processing of the fracture surface images.
references
[1] Hanneken JW. Hydrogen in metals and other materials: a comprehensive reference to books, bibliographies, workshops and conferences. Int J Hydrogen Energy 1999;24:1005e26. [2] Somerday Brian, Sofronis Petros, Jones Russell, editors. Effects of Hydrogen on Materials. Ohio: Published by ASM International e Materials Park; 2009. p. 766. Printed in the USA. [3] Somerday BP, Sofronis P, editors. Hydrogen-Materials Interaction. New York, USA: Published by ASME Press; 2014. p. 821. [4] Matsui H, Kimura H, Moriya S. The effect of hydrogen on the mechanical properties of high purity iron I. Softening and hardening of high purity iron by hydrogen charging during tensile deformation. Mater Sci Eng 1979;40:207e16. [5] Moriya S, Matsui H, Kimura H. The effect of hydrogen on the mechanical properties of high purity iron II. Effect of quenched-in hydrogen below room temperature. Mater Sci Eng 1979;40:217e25. [6] Matsui H, Kimura H, Kimura A. The effect of hydrogen on the mechanical properties of high purity iron III. The dependence of softening in specimen size and charging current density. Mater Sci Eng 1979;40:227e34. [7] Kimura H, Matsui H. Mechanism of hydrogen-induced softening and hardening in iron. Scr Metall 1987;21:319e24. [8] Gavriljuk VG, Shivanjuk VN, Foct J. Diagnostic experimental results on the hydrogen embrittlement of austenitic steels. Acta Mater 2003;51:1293e305. [9] Liang Y, Sofronis P, Aravas N. On the effect of hydrogen on plastic instabilities in metals. Acta Mater 2003;51:2717e30. [10] Ahn DC, Sofronis P, Dodds RH. On hydrogen-induced plastic flow localization during void growth and coalescence. Int J Hydrogen Energy 2007;32:3734e42. [11] Ren XC, Chu WY, Su YJ, Li JX, Qiao LJ, Jiang B, et al. The effects of atomic hydrogen and flake in mechanical properties of a tyre steel. Mater Sci Eng 2008;A491:164e71. [12] Murakami Y, Kanezaki T, Mine Y. Hydrogen effect against hydrogen embrittlement. Metall Mater Trans 2010;41A:2548e62. [13] Miresmaeili R, Liu L, Kanayama H. A possible explanation for the contradictory results of hydrogen effects on macroscopic deformation. Int J Press Vessels Pip 2012;99e100:34e43. [14] Liu ZY, Wang XZ, Du C, Li JK, Li XG. Effect of hydrogeninduced plasticity on the stress corrosion cracking of X70 pipeline steel in simulated soil environments. Mater Sci Eng 2016;A 658:348e54. [15] Dadfarnia M, Sofronis P, Neeraj T. Hydrogen interaction with multiple traps: can it be used to mitigate embrittlement? Int J Hydrogen Energy 2011;36:10141e8. [16] Turnbull A. Perspectives on hydrogen uptake, diffusion and trapping. Int J Hydrogen Energy 2015;40:16961e70. [17] Stashchuk M, Dorosh M. Evaluation of hydrogen stresses in metal and redistribution of hydrogen around crack-like defects. Int J Hydrogen Energy 2012;37:14687e96. [18] Jothi S, Croft TN, Wright L, Turnbull A, Brown SGR. Multiphase modelling of intergranular hydrogen segregation/ trapping for hydrogen embrittlement. Int J Hydrogen Energy 2015;40:15105e23. [19] Dmytrakh IM, Leshchak RL, Syrotyuk AM. Effect of hydrogen concentration on strain behaviour of pipeline steel. Int J Hydrogen Energy 2015;40:4011e8.
Please cite this article in press as: Dmytrakh IM, et al., Effect of hydrogen concentration on fatigue crack growth behaviour in pipeline steel, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.11.193
8
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 6 ) 1 e8
[20] Dmytrakh IM. Corrosion fracture of structural metallic materials: effect of electrochemical conditions in crack. Strain 2011;47(Suppl. 2):427e35. http://dx.doi.org/10.1111/ j.1475-1305.2010.00784.x. [21] VoltaLab 40 (PGZ301 & VoltaMaster 4). Dynamic Electrochemical Laboratory; 2009. p. 125. Instruction. Radiometer Analytical. [22] Cheng YF. Fundamentals of hydrogen evolution reaction and its implications on near-neutral pH stress corrosion cracking of pipelines. Electrochim Acta 2007;52:2661e7. [23] Cheng YF, Niu L. Mechanism for hydrogen evolution reaction on pipeline steel in near-neutral pH solution. Electrochem Commun 2007;9:558e62. [24] Niu L, Cheng YF. Corrosion behavior of X-70 pipe steel in near-neutral pH solution. Appl Surf Sci 2007;253:8626e31. [25] Dey S, Mandhyan AK, Sondhi SK, Chattoraj I. Hydrogen entry into pipeline steel under freely corroding conditions in two corroding media. Corros Sci 2006;48:2676e88. [26] Capelle J, Gilgert J, Dmytrakh I, Pluvinage G. Sensitivity of pipelines with steel API X52 to hydrogen embrittlement. Int J Hydrogen Energy 2008;33:7630e41. [27] Yan M, Weng Y. Study on hydrogen absorption of pipeline steel under cathodic charging. Corros Sci 2006;48:432e44. [28] Capelle J, Dmytrakh I, Azari Z, Pluvinage G. Evaluation of electrochemical hydrogen absorption in welded pipe with steel API X52. Int J Hydrogen Energy 2013;38:14356e63. [29] Franc¸ois D, Pineau A, Zaoui A. Environment assisted cracking. In: Mechanical behaviour of materials. Volume II: fracture mechanics and damage. Dordrecht: Springer; 2013. p. 363e406. [30] Syrotyuk AM, Dmytrakh IM. Methods for the evaluation of fracture and strength of pipeline steels and structures under the action of working media. Part І. Influence of the corrosion factor. Mater Sci 2014;50:324e39.
[31] Scanning Electron Microscope Zeiss EVO-40XVP. User Manual, Carl Zeiss SMT Ltd, Cambridge (England). 2005. Version: 1.0. 52 p. [32] RYa Kosarevych, Student OZ, Svirs'ka LM, Rusyn BP, Nykyforchyn HM. Computer analysis of characteristic elements of fractographic images. Mater Sci 2013;48:474e81. [33] Kosarevych RJ, Rusyn BP, Korniy VV, Kerod TI. Image segmentation based on the evaluation of the tendency of image elements to form clusters with the help of point field characteristics. Cybern Syst Anal 2015;51:704e13. [34] Suresh S. Fatigue of materials. 2nd ed. Cambridge, UK: Cambridge University Press; 1998. [35] Djukic MB, Sijacki Zeravcic V, Bakic G, Sedmak A, Rajicic B. Hydrogen embrittlement of low carbon structural steel. Proc Mater Sci 2014;3:1167e72. [36] Djukic MB, Sijacki Zeravcic V, Bakic GM, Sedmak A, Rajicic B. Hydrogen damage of steels: a case study and hydrogen embrittlement model. Eng Fail Anal 2015;58:485e98. [37] Djukic MB, Bakic GM, Sijacki Zeravcic V, Sedmak A, Rajicic B. Hydrogen embrittlement of industrial components: prediction, prevention, and models. Corrosion 2016;72:943e61. [38] Lunarska E, Ososkov Y, Jagodzinsky Y. Correlation between critical hydrogen concentration and hydrogen damage of pipeline steel. Int J Hydrogen Energy 1997;22:279e84. [39] Lufrano J, Sofronis P. Enhanced hydrogen concentrations ahead of rounded notches and cracks e competition between plastic strain and hydrostatic stress. Acta Metal 1998;46:1519e26. [40] Venegas V, Caleyo F, Gonzalez JL, Baudin T, Hallen JM, Penelle R. EBSD study of hydrogen-induced cracking in API5L-X46 pipeline steel. Scr Mater 2005;52:147e52. [41] Capelle J, Gilgert J, Dmytrakh I, Pluvinage G. The effect of hydrogen concentration on fracture of pipeline steels in presence of a notch. Eng Fract Mech 2011;78:364e73.
Please cite this article in press as: Dmytrakh IM, et al., Effect of hydrogen concentration on fatigue crack growth behaviour in pipeline steel, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.11.193
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevier.com/locate/he
Effect of hydrogen concentration on fatigue crack growth behaviour in pipeline steel I.M. Dmytrakh*, R.L. Leshchak, A.M. Syrotyuk, R.A. Barna Karpenko Physico-Mechanical Institute of National Academy of Sciences of Ukraine, 5 Naukova Street, 79060, Lviv, Ukraine
article info
abstract
Article history:
The ambiguous relationship between fatigue crack growth rate and hydrogen concentra-
Received 23 September 2016
tion CH in the bulk of metal under cyclic loading of the ferrite-pearlite low-alloyed steel in
Received in revised form
hydrogen-contained environments has been found: there is a certain CH value at which the
15 November 2016
crack growth resistance of steel increases. At these test conditions fracture surface dem-
Accepted 29 November 2016
onstrates some increasing of the plastic component on relief. The received results point on
Available online xxx
key importance of the determining of such threshold hydrogen concentration values, which correspond the transition “enhanced plasticity e embrittlement”.
Keywords:
© 2016 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.
Pipeline steel Hydrogen charging of metal Hydrogen concentration Cyclic loading Enhanced plasticity Hydrogen embrittlement
Introduction The phenomenon of hydrogen embrittlement of metallic materials, which lead to the loss of their plasticity, decreasing of fracture toughness and degradation of fatigue properties is well-known and the numerous records of such effects can be found in the literature [1e3]. Despite of that, there is the number of works where the positive hydrogen effect, i.e. the hydrogen-induced plasticity, was shown [4e14]. For example, in work [12] the dramatic phenomenon was found in which charging a supersaturated level of hydrogen into specimens of austenitic stainless steels of types 304 and 316L drastically improved the fatigue crack growth resistance, rather than accelerating fatigue crack growth rates. Results presented in
Ref. [14] confirmed the existence of a hydrogen-induced plasticity effect within a particular range of cathodic polarization of X70 pipeline steel. The possible explanation for the contradictory results of hydrogen effects on macroscopic deformation and the model to interpret the criterion for the application of local softening concept can be found in work [13]. Therefore it is important and actual problem to study and to clarify all aspects of specific hydrogen effects of metallic structural materials with the aim to establish the conditions for the transition “enhanced plasticity e embrittlement”. This may be the crucial key factor for deeper understanding of the hydrogen influence mechanism and also it will promote to review the existed conceptions where the hydrogen is
* Corresponding author. Fax: þ380 322 649427. E-mail address: [email protected] (I.M. Dmytrakh). http://dx.doi.org/10.1016/j.ijhydene.2016.11.193 0360-3199/© 2016 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved. Please cite this article in press as: Dmytrakh IM, et al., Effect of hydrogen concentration on fatigue crack growth behaviour in pipeline steel, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.11.193
2
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 6 ) 1 e8
considered as a negative agent. Some disputable aspects of this problem are currently discussed and considered [15e18]. Within the frame of above-mentioned problem the presented paper is dedicated to evaluation of the fatigue behaviour of hydrogen charged pipeline steel under known hydrogen concentration in a bulk of metal. Here the parameters of fatigue crack growth resistance of steel were assessed as the function of hydrogen concentration. This work can be considered as further development of our previous work [19] where the specificity of strain behaviour of pipeline steel depending on the hydrogen concentration in bulk of metal was studied. There the existence of some characteristic value of the hydrogen concentration at which the mechanism of hydrogen influence changes, namely: below this value the enhanced plasticity (decreasing of the yield stress value) takes places and above e the hydrogen embrittlement occurs, was shown.
Experimental procedure The object of study was low alloyed pipeline steel (sY ¼ 260 MPa and sU ¼ 440 MPa) with nominal chemical composition (in weight %): C ¼ 0.17e0.24; Si ¼ 0.17e0.37; Mn ¼ 0.35e0.65; S < 0.04; remainder Fe. This material consists of grains of ferrite-pearlite, typical for all carbon steels (Fig. 1). The rectangular cross-section beam specimens (Fig. 2) were manufactured with real pipes, which were supplied from two different manufacturers (test series A and test series B). The longitudinal cracks were studied and cut off of specimens from pipe was corresponded to this case. For realisation of experimental studies on the hydrogen charging of specimens, determination of the hydrogen concentration in a bulk of steel and the fatigue crack growth under joint action of the cyclic loading and hydrogenation environmental conditions the special testing stand was developed, which based on the fatigue testing machine for pure bending of specimens at the environmental conditions [20] and the dynamic electrochemical laboratory VoltaLab40 [21]. The general view of developed facility is presented in Figs.
Fig. 1 e Structural specificity of studied pipeline steel (x 1000).
Fig. 2 e Geometry of the beam specimen. 3 and 4. Here the standard three-electrode electrochemical cell was used where the auxiliary (counter) electrode consists of four round bars. The mutual location of working (specimen) and auxiliary (counter) electrodes is given in Fig. 5. The hydrogenation of specimens was made by electrochemical method under cathodic polarisation at some constant potential Ecath ¼ const. With the aim to simulate the hydrogen entry at real operating conditions of the buried pipeline, the following procedure has been applied [22e26]. The special deoxygenated, near-neutral pH NS4 solution, which is the model of underground water, was chosen as the electrolyte for hydrogen charging of steel. The chemical composition of the NS4 solution is given in Table 1. Taking into account the situation of freely corroding system that exists for the real pipeline, the potential of polarisation Ecath was slightly more negative than the free corrosion potential Ecorr for given steel, i.e.: Ecorr ¼ 600 mV (SCE) and Ecath ¼ 800 mV (SCE). Hydrogen concentration in a bulk of steel has been determined on the base of hydrogen discharging process under anodic polarisation with using of the hydrogen electrochemical oxidation method proposed in work [27]. The detailed description of the application of this method for the pipelines hydrogenation problems can be found in works [27,28]. As preliminary stage of study the experimental dependence “hydrogen concentration CH in specimen e time of exposure t” was received. Following to work [26] these experimental data were described by power relation:
Fig. 3 e Special fatigue testing stand.
Please cite this article in press as: Dmytrakh IM, et al., Effect of hydrogen concentration on fatigue crack growth behaviour in pipeline steel, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.11.193
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 6 ) 1 e8
Fig. 4 e Electrochemical cell.
Fig. 5 e Mutual location of beam specimen and auxiliary (counter) electrodes in electrochemical cell for fatigue test.
CH ¼ A$106$tm, [mol/cm3]
(1)
where A and m are some constants that depend on system “material e environment” and testing conditions. For our case the values of these constants are: A ¼ 0.28,106 and m ¼ 0.65. This analytic relation served as the reference curve under hydrogen charging of the specimens for fatigue test. The specimens, which hydrogen charged to assigned level of CH (CH ¼ 0.001; 0.209 and 0.456 ppm), were subjected to fatigue loading under pure bending conditions in NS4
Table 1 e Chemical composition of NS4 solution (g/l) [26]. NaHCO3
KCl
CaCl2
MaCl2$H2O
0.483
0.120
0.137
0.131
3
solution. Here the cathodic polarisation at Ecath ¼ 800 mV (SCE) was applied during all time of test. The rectangular cross-section beam specimens (Fig. 2) with initial edge crack of length a0 z 2.5 mm were subjected by cyclic pure bending with frequency f ¼ 1 Hz under stress ratio R ¼ 0. The stress intensity factor K in the crack tip for such specimen in case of its loading by pure bending was calculated from the known formula [20]. The initial level of loading, i.e. the initial range of stress intensity factor, was the same for all pffiffiffiffiffi tested specimens: DK0 z11 MPa m. The details of the test procedure are the following. As environment is transparent the crack lengths were measured on a plane surface of specimens by the optical microscope with accuracy ±0.005 mm. The specimens were subjected by pure bending under constant displacement mode. The subjects of tests were pre-cracked specimens with initial edge cracks of length a0 z 2.5 mm. The initial cracks were nucleated in air conditions (f ¼ 12 Hz and R ¼ 0) at final stress inpffiffiffiffiffi tensity factor range DKi z9:5 MPa m that is lower than the initial range of stress intensity factor for the main tests in pffiffiffiffiffi environment (DK0 z11 MPa m). Such choice of test conditions can be explained by following. As it is known from literature records [29] the most significant environmental effects can be observed at stress ratio R ¼ 0. The frequency of cyclic loading f is also very important parameter and it is accepted that fatigue crack growth rate increases with decreasing of frequency. However as it was shown [30] there is some maximum in the curve da/ dN ¼ F(f) and for considered steel this maximum corresponds to value fy1:0 Hz. From these reasons the conditions of fatigue test were chosen as R ¼ 0 and f ¼ 1.0 Hz for receiving the maximal environmental effect on fatigue crack growth rate. Resulting data under different initial hydrogen concentrations in the bulk of specimens were received as the sequence of the following parameters: number cycles of loading Ni, crack length ai, fatigue crack growth rate (da/aN)i, and stress intensity factor rang DKi. After fatigue tests the fracture surfaces of specimens were examined by the scanning electron microscopy EVO-40XVP [31] and some number of special selected images were received. For quantitative evaluation of the fracture surface relief the images were digitally processed with the aim to detect the possible difference in fracture mechanisms under different hydrogen concentration in the steel. Taking into account the complicated inhomogeneous relief of fracture surfaces, because the fatigue crack growth mechanism is gradually mixed (tearing and shearing), the special technique for computer analysis of characteristic elements of fractographic images was applied [32,33]. This method is based on the statement that for each fracture mechanism the own typical features of image's structure are inherent [32]. In particular, the ductile fracture characterizes with image's structure with the closed contours, in the form of ovals. This is the ground for procedure of image segmentation, which uses the multilevel determination of the brightness threshold [32]. Here, with the aim to get the less distortion of the segmenting image not only the large details were used but also the small details with low peaks in the histogram of brightness were taken into account. It enables to separate the image fragment of the fracture surface into
Please cite this article in press as: Dmytrakh IM, et al., Effect of hydrogen concentration on fatigue crack growth behaviour in pipeline steel, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.11.193
4
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 6 ) 1 e8
connected regions, and then to determine and analyze their quantitative parameters. The processing of received images was focused on the detection of sites where the shear fatigue crack growth mechanism was occurred. Each image had the square of 786,432 pixels or S ¼ 42,219 mm2. Using the above-mentioned procedure the corresponding sites were detected (see below Fig. 10 where these sites are marked with blue rectangles). It should be noted that the real area of ductile fracture is bounded of the fracture relief and it is smaller than the blue rectangle square. Received data were used for calculation of the portion of fracture surface area with the shear fatigue crack growth mechanism. The detailed description of applied method is given in Refs. [32,33].
Results and discussion Two series of fatigue tests (A and B) were conducted under different initial hydrogen concentration in the bulk of specimens. Primary test results were presented as the separate scatter plots of fatigue crack growth rate da/dN versus stress intensity factor DK (Fig. 6). The comparison of these data showed on some small difference between series A (Fig. 6a) and B (Fig. 6b) that can be considered as natural scattering, which is always inherent for such tests. However, when we impose these two graphs on each other a full coincidence of data can be seen for both tests at CH ¼ 0.209 ppm (red dots). It indicates on the ambiguous relationship between fatigue crack growth rate da/dN and hydrogen concentration CH for two series of tests. For comparative assessment and analysis of results these scatter plots were described analytically with using the wellknown Paris equation [34]:
da/dN ¼ C$(DK)n;
(2)
where C and n are the constants of material and test conditions. The values of constants C and n in equation (2) and mean square deviation r2 are given in Table 2. Fig. 7 presents an analytic description fatigue crack growth rate diagrams by power function (2). The received results showed on the ambiguous effect of an initial hydrogen concentration in specimens on fatigue crack growth in steel for both test series. For given pipeline steel there is some characteristic value of the hydrogen concentration CH y0:209 ppm, which causes the decreasing of fatigue crack growth rate and, consequently, leads to increasing of fatigue resistance of material. Especially, it can be clear seen when we pffiffiffiffiffi make the section of diagrams at DK ¼ 20 MPa$ m (Fig. 8) and at da/dN ¼ 104 mm/cycle (Fig. 9). Such tendency may be explained [12] by interaction between hydrogen and dislocations. The hydrogen affects on dislocations by two ways: promotes of their pinning (or dragging) or enhances of their mobility. Competition between these two roles defines whether the resulting phenomenon is positive or negative with point of view of fatigue crack growth resistance of material. It is important to note that above-presented results are restricted by Paris region and their analysis is valid only to this part of the fatigue crack growth rate diagram [34]. For consideration of the threshold zone as well as the final failure zone the special study is required and it can be the subject of further research. The fracture surface examination was made for the restricted area, which correspond of the range of stress inpffiffiffiffiffi tensity factor DK ¼ 20 MPa$ m, i.e. for conditions were the CH effect is the most significant on fatigue crack growth rate (see Figs. 7 and 8). The visual observation of received images showed on the modification of fracture surface depending on the hydrogen concentrations CH in bulk of material (Fig. 10). For very low hydrogen charged specimen (CH ¼ 0.001 ppm) the surface has inhomogeneous relief indicating a mixed mechanism of fatigue crack growth: tearing and shearing (Fig. 10a). Here also the relief of typical quasicleavage brittle
Fig. 6 e Fatigue crack growth rate diagrams under different initial hydrogen concentrations in the bulk of specimens: a e test series A; b e test series B. Please cite this article in press as: Dmytrakh IM, et al., Effect of hydrogen concentration on fatigue crack growth behaviour in pipeline steel, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.11.193
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 6 ) 1 e8
5
Fig. 7 e Analytic description fatigue crack growth rate diagrams by power function (Paris equation): a e test series A; b e test series B.
Fig. 8 e Dependence of fatigue crack growth rate da/dN at pffiffiffiffiffi DK ¼ 20 MPa$ m on the initial hydrogen concentrations CH in material.
fracture can be also found. With increasing of the hydrogen concentrations CH in material (CH ¼ 0.209 ppm) the increasing of the homogeneity of surface relief can be seen due to increasing of area where the shear fatigue crack growth mechanism was occurred (Fig. 10b). It points on increasing of the plastic deformation under fatigue process of steel. Further increasing of the initial hydrogen concentration in specimen (CH ¼ 0.514 ppm and CH ¼ 1.231 ppm) leads to increasing of the heterogeneity of surface relief and to appearance of some traces of tearing and quasicleavage, i.e. the fatigue crack growth mechanism becomes gradually mixed again (Fig. 10c and d). The digital processing of these images with using the special technique [32,33] gave the possibility to recognize the sites where the shear fatigue crack growth mechanism were realized. In Fig. 10 these sites are marked with blue rectangles. From these results it can be seen that the area marked with blue rectangles in Fig. 10b is visibly bigger than in Fig. 10a, c
Fig. 9 e Dependence of stress intensity factor value DK at da/dN ¼ 10¡4 mm/cycle on the initial hydrogen concentrations CH in material. and d. It indicates on the predominance of the shear fatigue crack growth mechanism at CH ¼ 0.209 ppm in comparison with others considered cases. For final presentation of received data we used the parameter k, which is the portion of image area (in percents to total area) where the shear fatigue crack growth mechanism was realized. The dependence of parameter k on the of the initial hydrogen concentrations CH in steel is shown in Fig. 11. It can be clear seen that this plot has the maximum at the hydrogen concentration CH y0:209 ppm that can serve as additional justification of the received findings. These data and their analysis give the basis for the following statement: there is a certain value of the hydrogen concentration in a bulk of material CH ¼ C*H , at which the crack growth resistance of steel increases and the diagrams of fatigue crack growth rate are shifted to higher values of stress intensity factor. For given pipeline steel C*H ¼ 0:209 ppm. It should be noted that value is close to early established characteristic value of the hydrogen concentration at which the
Please cite this article in press as: Dmytrakh IM, et al., Effect of hydrogen concentration on fatigue crack growth behaviour in pipeline steel, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.11.193
6
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 6 ) 1 e8
pffiffiffiffiffi Fig. 10 e Specificity of fracture surface during fatigue crack growth in pipeline steel at DK ¼ 20 MPa$ m on dependence of the initial hydrogen concentrations CH in material: a e CH ¼ 0.001 ppm; b e CH ¼ 0.209 ppm; c e CH ¼ 0.514 ppm; d e CH ¼ 1.231 ppm (x 500). The sites of the shear fatigue crack growth mechanism detection are marked with blue rectangles. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
mechanism of hydrogen effect on steel deformation under static loading changes: enhanced plasticity or embrittlement [19]. There it was found that C*H ¼ 0:227 ppm. As a concluding remark it may be noted that the determining of threshold values C*H , which correspond the transition “enhanced plasticity e embrittlement” can be considered as one of the key factors for the assessment of serviceability of structural materials in the presence of hydrogen-contained environments. This is also very important to underline that of the hydrogen effect on mechanical behaviour of structural
metallic materials is multiple and several mechanisms can be realized, which may affect independently or coexist simultaneously. Here, first of all, the recent works [35,36], should be marked where the simultaneous action of the hydrogenenhanced decohesion (HEDE) and hydrogen enhanced
70,00 ΔK = 20 MPa(m)
1/2
= const.
Table 2 e Values of the parameters n and C in Paris equation under different initial hydrogen concentrations CH in the bulk of specimens. " # CH, ppm Test series A 0.001 0.209 0.456 Test series B 0.001 0.209 0.514 1.231
n
C,
mm=cycle pffiffiffiffi ðMPa$ m Þn
4.71 4.86 4.29
1$1010 5$1011 5$1010
5.55 5.16 5.92 5.66
11
1$10 2$1011 4$1012 9$1012
r2
0.96 0.97 0.96
k, %
65,00
60,00
55,00 CH=0,209 ppm
50,00 0
0,5
1
1,5
CH, ppm 0.96 0.96 0.98 0.99
Fig. 11 e Fracture surface characterisation parameter k as the function of the initial hydrogen concentrations CH in pffiffiffiffiffi steel (DK ¼ 20 MPa$ m).
Please cite this article in press as: Dmytrakh IM, et al., Effect of hydrogen concentration on fatigue crack growth behaviour in pipeline steel, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.11.193
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 6 ) 1 e8
localized plasticity (HELP) was shown for low carbon structural steel depending on the local hydrogen concentration in metal. Such results are the base for the development of corresponding engineering models with the aim to predict and to prevent the hydrogen degradation of the structural components under aggressive operating conditions [37]. At the end it should be stated that definition “critical concentration” is often used in studies of problems of hydrogen in metals and alloys [1,38e40]. Although in different works this term has different physical sense. For example, according to work [38], hydrogen treatment below the “critical” content was found to cause the substantial rearrangement of dislocations and de-cohesion of grain boundaries and in overcritical condition, the formation of micro crevices at the grain and phase boundaries has occurred. Others authors use this definition under developing of local fracture criteria [1,39]. In Ref. [41] the term “critical concentration” defines the hydrogen concentration level at which the critical loss of local strength of the material at the notch is occurred. In Ref. [37] the special diagram for structural integrity analysis of boiler pipes exposed to hydrogen damage is proposed and where the hydrogen concentration is the basic parameter. Here the definition “critical concentration” is associated with the increased activity and activation of the HEDE mechanism (HEDE > HELP). Above mentioned results and data confirm the principal importance to take into account the hydrogen concentration in all considerations where the problems of hydrogen degradation of materials are investigated.
Conclusions The specificity of fatigue behaviour of pipeline steel depending on the initial hydrogen concentration in bulk of metal CH was studied. The ambiguous relationship between fatigue crack growth rate and hydrogen concentration CH in the bulk of metal under cyclic loading of the ferrite-pearlitic low-alloyed steel in hydrogen-contained environments has been found. It was shown that there is a certain CH value at which the crack growth resistance of steel increases and the diagrams of fatigue crack growth rate are shifted to higher values of stress intensity factor. This value correlates with early established characteristic value of the hydrogen concentration at which the mechanism of hydrogen effect on steel deformation under static loading changes: enhanced plasticity or embrittlement. The received results point on the fundamental importance of such characteristic (threshold) values of hydrogen concentration in structural steels for the evaluation their strength and serviceability under assigned operating conditions.
Acknowledgements Presented study was conducted within the project number 10ІІІ-108-15 of Fundamental Research Programme of National Academy of Sciences of Ukraine (2015e2017). The authors also
7
would like to thank to Dr. R. Kosarevych for his valuable help in digital processing of the fracture surface images.
references
[1] Hanneken JW. Hydrogen in metals and other materials: a comprehensive reference to books, bibliographies, workshops and conferences. Int J Hydrogen Energy 1999;24:1005e26. [2] Somerday Brian, Sofronis Petros, Jones Russell, editors. Effects of Hydrogen on Materials. Ohio: Published by ASM International e Materials Park; 2009. p. 766. Printed in the USA. [3] Somerday BP, Sofronis P, editors. Hydrogen-Materials Interaction. New York, USA: Published by ASME Press; 2014. p. 821. [4] Matsui H, Kimura H, Moriya S. The effect of hydrogen on the mechanical properties of high purity iron I. Softening and hardening of high purity iron by hydrogen charging during tensile deformation. Mater Sci Eng 1979;40:207e16. [5] Moriya S, Matsui H, Kimura H. The effect of hydrogen on the mechanical properties of high purity iron II. Effect of quenched-in hydrogen below room temperature. Mater Sci Eng 1979;40:217e25. [6] Matsui H, Kimura H, Kimura A. The effect of hydrogen on the mechanical properties of high purity iron III. The dependence of softening in specimen size and charging current density. Mater Sci Eng 1979;40:227e34. [7] Kimura H, Matsui H. Mechanism of hydrogen-induced softening and hardening in iron. Scr Metall 1987;21:319e24. [8] Gavriljuk VG, Shivanjuk VN, Foct J. Diagnostic experimental results on the hydrogen embrittlement of austenitic steels. Acta Mater 2003;51:1293e305. [9] Liang Y, Sofronis P, Aravas N. On the effect of hydrogen on plastic instabilities in metals. Acta Mater 2003;51:2717e30. [10] Ahn DC, Sofronis P, Dodds RH. On hydrogen-induced plastic flow localization during void growth and coalescence. Int J Hydrogen Energy 2007;32:3734e42. [11] Ren XC, Chu WY, Su YJ, Li JX, Qiao LJ, Jiang B, et al. The effects of atomic hydrogen and flake in mechanical properties of a tyre steel. Mater Sci Eng 2008;A491:164e71. [12] Murakami Y, Kanezaki T, Mine Y. Hydrogen effect against hydrogen embrittlement. Metall Mater Trans 2010;41A:2548e62. [13] Miresmaeili R, Liu L, Kanayama H. A possible explanation for the contradictory results of hydrogen effects on macroscopic deformation. Int J Press Vessels Pip 2012;99e100:34e43. [14] Liu ZY, Wang XZ, Du C, Li JK, Li XG. Effect of hydrogeninduced plasticity on the stress corrosion cracking of X70 pipeline steel in simulated soil environments. Mater Sci Eng 2016;A 658:348e54. [15] Dadfarnia M, Sofronis P, Neeraj T. Hydrogen interaction with multiple traps: can it be used to mitigate embrittlement? Int J Hydrogen Energy 2011;36:10141e8. [16] Turnbull A. Perspectives on hydrogen uptake, diffusion and trapping. Int J Hydrogen Energy 2015;40:16961e70. [17] Stashchuk M, Dorosh M. Evaluation of hydrogen stresses in metal and redistribution of hydrogen around crack-like defects. Int J Hydrogen Energy 2012;37:14687e96. [18] Jothi S, Croft TN, Wright L, Turnbull A, Brown SGR. Multiphase modelling of intergranular hydrogen segregation/ trapping for hydrogen embrittlement. Int J Hydrogen Energy 2015;40:15105e23. [19] Dmytrakh IM, Leshchak RL, Syrotyuk AM. Effect of hydrogen concentration on strain behaviour of pipeline steel. Int J Hydrogen Energy 2015;40:4011e8.
Please cite this article in press as: Dmytrakh IM, et al., Effect of hydrogen concentration on fatigue crack growth behaviour in pipeline steel, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.11.193
8
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 6 ) 1 e8
[20] Dmytrakh IM. Corrosion fracture of structural metallic materials: effect of electrochemical conditions in crack. Strain 2011;47(Suppl. 2):427e35. http://dx.doi.org/10.1111/ j.1475-1305.2010.00784.x. [21] VoltaLab 40 (PGZ301 & VoltaMaster 4). Dynamic Electrochemical Laboratory; 2009. p. 125. Instruction. Radiometer Analytical. [22] Cheng YF. Fundamentals of hydrogen evolution reaction and its implications on near-neutral pH stress corrosion cracking of pipelines. Electrochim Acta 2007;52:2661e7. [23] Cheng YF, Niu L. Mechanism for hydrogen evolution reaction on pipeline steel in near-neutral pH solution. Electrochem Commun 2007;9:558e62. [24] Niu L, Cheng YF. Corrosion behavior of X-70 pipe steel in near-neutral pH solution. Appl Surf Sci 2007;253:8626e31. [25] Dey S, Mandhyan AK, Sondhi SK, Chattoraj I. Hydrogen entry into pipeline steel under freely corroding conditions in two corroding media. Corros Sci 2006;48:2676e88. [26] Capelle J, Gilgert J, Dmytrakh I, Pluvinage G. Sensitivity of pipelines with steel API X52 to hydrogen embrittlement. Int J Hydrogen Energy 2008;33:7630e41. [27] Yan M, Weng Y. Study on hydrogen absorption of pipeline steel under cathodic charging. Corros Sci 2006;48:432e44. [28] Capelle J, Dmytrakh I, Azari Z, Pluvinage G. Evaluation of electrochemical hydrogen absorption in welded pipe with steel API X52. Int J Hydrogen Energy 2013;38:14356e63. [29] Franc¸ois D, Pineau A, Zaoui A. Environment assisted cracking. In: Mechanical behaviour of materials. Volume II: fracture mechanics and damage. Dordrecht: Springer; 2013. p. 363e406. [30] Syrotyuk AM, Dmytrakh IM. Methods for the evaluation of fracture and strength of pipeline steels and structures under the action of working media. Part І. Influence of the corrosion factor. Mater Sci 2014;50:324e39.
[31] Scanning Electron Microscope Zeiss EVO-40XVP. User Manual, Carl Zeiss SMT Ltd, Cambridge (England). 2005. Version: 1.0. 52 p. [32] RYa Kosarevych, Student OZ, Svirs'ka LM, Rusyn BP, Nykyforchyn HM. Computer analysis of characteristic elements of fractographic images. Mater Sci 2013;48:474e81. [33] Kosarevych RJ, Rusyn BP, Korniy VV, Kerod TI. Image segmentation based on the evaluation of the tendency of image elements to form clusters with the help of point field characteristics. Cybern Syst Anal 2015;51:704e13. [34] Suresh S. Fatigue of materials. 2nd ed. Cambridge, UK: Cambridge University Press; 1998. [35] Djukic MB, Sijacki Zeravcic V, Bakic G, Sedmak A, Rajicic B. Hydrogen embrittlement of low carbon structural steel. Proc Mater Sci 2014;3:1167e72. [36] Djukic MB, Sijacki Zeravcic V, Bakic GM, Sedmak A, Rajicic B. Hydrogen damage of steels: a case study and hydrogen embrittlement model. Eng Fail Anal 2015;58:485e98. [37] Djukic MB, Bakic GM, Sijacki Zeravcic V, Sedmak A, Rajicic B. Hydrogen embrittlement of industrial components: prediction, prevention, and models. Corrosion 2016;72:943e61. [38] Lunarska E, Ososkov Y, Jagodzinsky Y. Correlation between critical hydrogen concentration and hydrogen damage of pipeline steel. Int J Hydrogen Energy 1997;22:279e84. [39] Lufrano J, Sofronis P. Enhanced hydrogen concentrations ahead of rounded notches and cracks e competition between plastic strain and hydrostatic stress. Acta Metal 1998;46:1519e26. [40] Venegas V, Caleyo F, Gonzalez JL, Baudin T, Hallen JM, Penelle R. EBSD study of hydrogen-induced cracking in API5L-X46 pipeline steel. Scr Mater 2005;52:147e52. [41] Capelle J, Gilgert J, Dmytrakh I, Pluvinage G. The effect of hydrogen concentration on fracture of pipeline steels in presence of a notch. Eng Fract Mech 2011;78:364e73.
Please cite this article in press as: Dmytrakh IM, et al., Effect of hydrogen concentration on fatigue crack growth behaviour in pipeline steel, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.11.193