CORROSION TESTING Corrosion fatigue assessment of extruded ...€¦ · sion resistance of magnesium...

© Carl Hanser Verlag, München. Der Nachdruck, auch auszugsweise, ist nicht gestattet und muss beim Verlag schriftlich genehmigt werden.

CORROSION TESTING 15

60 (2018) 1 © Carl Hanser Verlag, München Materials Testing

Materials

The investigated wrought alloys AZ31 and ME20, whose nominal compositions are given in Table 1, were cast in a modified gravity casting process. Billets with diam-eters of 50 mm and lengths of 150 mm were prepared from the cast slaps for the extrusion process. Extrusion tests were carried out at Helmholtz-Zentrum Geesthacht (HZG) (see Figure 1) using an indirect extrusion process, as schemati-cally shown in Figure 2. In this process, there is no relative movement of the billet and the container, which provides the ad-vantage, that a constant pressing force is applied whilst the whole process to achieve homogeneous properties of the extruded material along the profile. The extrusion ratio was 1/25 as the resulting profiles had a diameter of 10 mm. Before the extrusion process, the billets were heated up to a tem-perature of 450 °C for 1800 s. The ram speed for extrusion was constantly 3.4 mm × s-1.

Grain refinement, in turn, may lead to an improvement of the generally low corro-sion resistance of magnesium alloys, which is a major obstacle for their broad applica-tion in light weight applications [1, 3, 4].

Corrosion behavior improvement, as a result of grain refinement, was attributed in some studies to improved passivation be-havior due to formation of a more stable surface oxide layer [1, 3]. However, other studies also reported deterioration of the corrosion resistance, e. g., when the pre-cipitation state was changed by extrusion [4]. In the present study, the influence of corrosion on the microstructure and the dependent mechanical properties under cyclic load for extruded rare earth contain-ing magnesium alloy ME20 as well as standard magnesium wrought alloy AZ31 were characterized and compared by means of load increase tests (LIT) in air and 5 wt.-% NaCl solution. The reduction of the estimated corrosion fatigue strengths was correlated with the different corrosion be-haviors in the same electrolyte.

Because of their low density and high spe-cific strength, magnesium alloys have a high potential for light weight structural applications. Especially extruded profiles made of wrought magnesium alloys show superior mechanical properties compared to cast material [1, 2]. They provide a finer and more homogeneous microstructure than casted materials, but on the other hand, the formability of the extruded pro-files is hampered by the development of a strong texture during the thermomechani-cal processing. The texture has its origin in the hcp lattice structure of magnesium and its alloys, which reduces the number of slip systems that could be activated during plastic deformation [2].

Whilst extrusion of round bars to round profiles, typically, a texture develops in which the basal planes of the hcp crystals are aligned parallel to the extrusion direc-tion. Alloying of rare earth elements, like Ce, to magnesium can result in a less pro-nounced texture and change properties of the microstructure like grain refinement.

The corrosion fatigue behavior of extruded magne-sium alloys AZ31 and ME20 was characterized by means of load increase tests in 5 wt.-% NaCl solution. The results were correlated with the respective corro-sion behaviors investigated in immersion tests and by means of microstructural investigations after precor-rosion. ME20 exhibited a superior corrosion resist-ance compared to AZ31, probably as a result of an en-hanced passivation tendency due to grain refinement because of the presence of rare earth alloying ele-ments. Accordingly, no differences in the cyclic defor-mation behaviors of ME20 in air and 5 wt.-% NaCl so-lution were observed in the load increase tests, while AZ31 exhibited a reduction of the estimated fatigue strengths as well as signs of embrittlement due to the mechanical-medial load.

Martin Klein, Philipp Wittke, Dortmund, Roland Hoppe, Dietmar Letzig, Geesthacht, and Frank Walther, Dortmund, Germany

Corrosion fatigue assessment of extruded magnesium alloys AZ31 and ME20

Article Information

Correspondence AddressMartin KleinTU Dortmund University,Department of Materials Test Engineering (WPT)Baroper Str. 303D-44227 DortmundE-mail: [email protected]. Dr.-Ing. Frank WaltherDepartment of Materials Test Engineering (WPT)TU Dortmund University

KeywordsMagnesium alloys, AZ31, ME20, extrusion, corrosion fatigue behavior


16 CORROSION TESTING

60 (2018) 1

After extrusion the profiles were cut in parts of 200 mm length for manufacturing of specimens for the fatigue tests, accord-ing to the geometry in Figure 3. In addi-tion, specimens from different sections of the profiles were metallographically pre-pared to investigate their homogeneity in terms of microstructure and texture (see Figure 4).

Figure 5 shows exemplary light micro-scope images of the extruded profiles, which were taken parallel to the extrusion direction. Estimated grain sizes were found not to vary along the extruded profile in both cases as well as the occurrence of pre-

cipitations. The mean grain size of AZ31 was estimated to be 15.6 ± 1.6 μm, while in case of ME20, the estimated mean grain size was 8.9 ± 0.8 μm, respectively. Ac-cordingly, the ME20 alloy shows a much finer microstructure compared to the AZ31 reference alloy, as was expected before, be-cause of the well-known grain refinement effect of the contained rare earth elements. For both materials, precipitation of parti-cles can be observed. According to litera-ture, the precipitations are expected to be Al6Mn or Al8Mn5 particles in case of AZ31 [1] and Mg12Ce precipitates in case of ME20 [2].

For analysis of the textures of the extruded profiles, specimens were polished perpen-dicular to the extrusion direction. The pole figures in Figure 6 show, that there is no dif-ference in the shape of textures between both alloys. Furthermore, no mentionable varia-tion of the shape of the textures along the profiles was found. Therefore, an influence of the texture developed during the extrusion process on the results of the fatigue tests can probably be excluded, making these results more comparable.

Alloy Mg Mn Al Zn RE

AZ31 bal. - 3 1 -

ME20 bal. 2 - - < 0.5

Table 1: Nominal chemical compositions (wt.-%) of AZ31 and ME20 Figure 1: Extrusion facility at HZG

Figure 2: Schematic of indirect extrusion process

Figure 3: Specimen geometry for fatigue investigations (all dimensions in mm)

Figure 4: Sampling plan of extruded profiles; specimens 1 to 10 were used for manufacturing of fatigue specimens; specimens a to c were used for metallographic analyses

Experimental methods

The corrosion behavior was quantitatively investigated by means of immersion tests in 5 wt.-% NaCl solution. In the immer-sion tests, corrosion rates were deter-mined according to the formed specific hydrogen volume as a function of the im-mersion time, which is the most reliable method of corrosion rate determination for magnesium and magnesium alloys be-sides direct measurements of the corro-sive weight loss [5]. For this purpose, the experimental setup introduced by Song et al. [5] was applied, in which the formed hydrogen is collected in a burette. Speci-mens in initial corrosion state as well as macrographs in a proceeded corrosion state were taken to investigate the corro-sion mechanisms.Figure 5: Light microscope images of longitudinal sections of a) AZ31 and b) ME20 extruded samples

a) b)



60 (2018) 1

For the investigation of the corrosion fa-tigue behavior, instrumented continuous load increase tests were performed in 5 wt.-% NaCl solution and as a reference also in air. The specimens were subjected to a stress ratio of R = 0.1 at a frequency of f = 10 Hz using sinusoidal load-time func-tions at ambient temperature. Fatigue strengths for a maximum number of cycles of Nmax = 2 × 106 were estimated based on recorded material responses [6]. For this purpose, plastic and total strains were de-termined using an extensometer. In the tests in 5 wt.-% NaCl solution, a self-devel-oped miniaturized in-situ electrochemical cell with a three electrode system was ap-plied to measure corrosion potential re-sponses as well (see Figure 7). The corro-sion potential responses were recorded with a high data acquisition rate using a self-programmed LabVIEW software, which also allows analysis of the oscillation characteristic of the corrosion potential, which is induced by the cyclic straining of the specimens.

Results and discussion

In Figure 8, the formed specific hydrogen volumes for AZ31 and ME20 in 5 wt.-% NaCl solution are plotted as a function of immersion time. ME20 exhibits a much lower corrosion rate than AZ31, given by the much lower hydrogen formation rate. Corrosion rates were determined according to the nearly linear slope of the specific hy-drogen volume as a function of time after approx. 4 × 105 s and are given in Table 2.

To investigate the reason for the differ-ences in the corrosion behaviors on micro-structural basis, micrographs in initial state of corrosive attack as well as in a pro-ceeded corrosion state were taken by light microscope. In Figure 9a, a micrograph of AZ31 after initial corrosive attack is given. The image shows typical features of fili-form corrosion. Filiform corrosion typically occurs in the presence of a partially protec-tive passivating film on the specimen sur-face and mostly in combination with pit-ting corrosion at cathodic active precipi-tates or impurity particles [7]. Once a corrosion pit has formed at precipitates or impurities, the anodic active corrosion front can move along the surface. In this context, the process is driven by the diffu-sion of Cl ions into the anodic active re-gions, which enhances dissolution of the passive film, as well as by repassivation of the tail of the corrosion front, having also a cathodic function, leading to the filiform Figure 9: Microstructure after initial corrosive attack, a) AZ31, b) ME20

a) b)

m corr(g × s-1 × cm-2)

AZ31 1.07 × 10-8

ME20 1.84 × 10-9

Table 2: Corrosion rates m corr determined in immersion tests

Figure 7: Experimental setup for corrosion fatigue tests, a) photograph, b) schematic drawing

a) b)

a) b)

Figure 6: Textures of a) AZ31 and b) ME20 extruded samples

Figure 8: Results of immersion tests in 5 wt.-% NaCl solution



60 (2018) 1

σmax,start = 10 MPa until failure of the speci-men. It can be observed, that εa,p degres-sively increases as a function of the load cycles over the whole test duration. This deformation behavior is unexpected, since continuous load increase normally causes a progressive εa,p increase [8]. Therefore, this behavior could be attributed, e. g., to approaching a saturation state due to the already mentioned poor formability of ex-truded magnesium profiles or to a load-de-pendent change in the deformation mecha-nism, e. g., through activation of additional slip systems or a change in the deformation mechanism from slip to twinning or vice versa. εm,t increases linearly as a function of load cycles until approx. N = 16 × 104, followed by another linear range from ap-prox. N = 22 × 104 with a larger slope until exponential increase right before failure at a maximum failure stress of σmax,f (LIT) ≈ 250 MPa. The fatigue strength of AZ31 in air was then estimated according to the transition between both linear ranges of εm,t, based on determination of the inter-sections of the asymptotes (dashed lines) [8]. Therefore, the fatigue strength was in this case estimated to be approx. σmax,e (LIT) ≈ 200 MPa.

In case of ME20, the fatigue strength was estimated in the same manner to be σmax,e (LIT) ≈ 110 MPa. The specimen failed at σmax,f (LIT) ≈ 150 MPa.

Subsequently, identical load increase tests were performed in 5 wt.-% NaCl solu-tion. Figure 12a shows the results of a load increase test with AZ31. Besides σmax, εa,p and εm,t, also the corrosion potential mean value Ecorr,m and the corrosion potential am-plitude Ecorr,a are plotted as a function of the number of cycles. The progression of εm,t is identical with that from the test in air until earlier failure of the specimen in NaCl solu-tion at σmax,f (LIT) ≈ 210 MPa, as displayed in Figure 12b. In contrast, εa,p values in NaCl solution are much lower than in air. A de-gressive increase can be observed again until N ≈ 15 × 104 and afterwards εa,p starts even to decrease. The much lower values of εa,p could possibly be attributed to an em-brittlement of the material in the corrosive environment. With regard to the electro-chemical measurands, Ecorr,a shows a linear increase as a function of the number of cy-cles. Just before failure, Ecorr,a increases ex-ponentially due to the growth of macro-cracks [9]. Ecorr,m shows a very fluctuating progression with several activation and passivation peaks, which can probably be attributed to breakdown of passivating lay-ers due to the cyclic loading of the speci-

Figure 10c). However, the pits seem to be repassivated afterwards, instead of acting as initiation sites for corrosion propagation as in the case of AZ31. Therefore, ME20 seems to have a superior passivating ten-dency.

For a comparative evaluation of the cor-rosion fatigue behaviors, continuous load increase tests (LIT) were performed. Fig-ure 11 presents the results of a load in-crease test with an AZ31 specimen in air. In the diagram, the maximum stress σmax (controlled variable), the plastic strain am-plitude εa,p and the total mean strain εm,t are plotted as functions of the load cycles N. σmax was linearly increased as a function of the load cycles with a load increase rate of dσmax/dN = 10 MPa × 10-4, starting from

corrosive deterioration [7]. Figure 10a shows a cross section of AZ31 after long-term immersion for 1 week in 5 wt.-% NaCl solution. The specimen exhibits a pit-shaped corrosive deterioration, which is a result of the corrosion processes men-tioned before.

The corresponding micrograph of ME20 in initial corrosion state in Figure 9b shows only very localized, spot-like corrosive at-tack at the Mg12Ce particles (see Figure 5), while the rest of the surface remains unaf-fected. The long-term corrosion morphol-ogy in Figure 10b underlines, that even af-ter 1 week immersion the corrosive deteri-oration is limited to these particular spots. Obviously, the Mg12Ce particles are under-cut, leaving small pits on the surface (see

Figure 11: Strain developments in con-

tinuous load increase test with AZ31 in air

Figure 10: Corrosion morphologies after immer-

sion in 5 wt.-% NaCl solution for 1 week, a) AZ31,

b) ME20, c) detail of formed pits on ME20

a)

b)

c)



60 (2018) 1

men and subsequent formation of corro-sion pits in these activated areas. At around N ≈ 15 × 104, Ecorr,m shows a significant ac-tivation peak (dashed line), which corre-sponds with the angular point of εa,p. Thus, the combined material response of Ecorr,m

and εa,p could be attributed to formation of a corrosion pit with critical size, which acts as an initiation site for microcrack growth, leading to strain localization and embrittlement of the crack tip, which is given by the following decreasing progres-sion of εa,p. Therefore, the corrosion fatigue strength was estimated to be around σmax,e (LIT) ≈ 160 MPa in this case.

Figure 13a displays the corresponding results for the load increase test with ME20 in 5 wt.-% NaCl solution. The εm,t pro-gression is almost completely identical to that from the test in air, as shown in Fig-ure 13b, and failure occurs at approx. the same maximum failure stress of σmax,f (LIT) ≈ 150 MPa. εa,p progressions are also qualita-tively similar in both tests and εa,p values are in the same magnitude. Therefore, ap-parently no embrittlement occurs, as ob-served for AZ31, which could be attributed to the observed passivation tendency of ME20, since embrittlement of a passivated specimen is unlikely [10]. However, also for ME20, εa,p shows unexpected progressions in both tests. Altogether, εa,p exhibits a quasi-progressive progression. However, e. g., in the test in NaCl solution, first there is increasing progression until an angular point around N ≈ 3 × 104 and then decreas-ing progression. From N ≈ 7 × 104, εa,p starts slowly to increase again and turns into a linear increase until failure given by exponential increase. The changing εa,p progression could be attributed again to a change in the deformation mechanism. Ecorr,a increases along a linear base line, as observed for AZ31, but there are several jumps, after which Ecorr,a returns to the base line. As mentioned before, an increase of Ecorr,a is a specific indicator for crack ini-tiation and propagation. Therefore, the Ecorr,a jumps could possibly be attributed to formation of microcracks at weak spots of the passivating layer which are then repas-sivated, so that it comes to crack closure, as a result of the superior passivating behav-ior of ME20 as observed in the corrosion in-vestigations. Ecorr,m shows a much smoother progression than for AZ31 with a first passi-vation plateau around N ≈ 3 × 104, which corresponds to the first angular point of εa,p. With decreasing εa,p values, also Ecorr,m decreases again, which supports the as-sumption, that a change in the deformation

Figure 12: a) Strain and corrosion potential devel-opments in continuous load increase test with AZ31 in 5 wt.-% NaCl so-lution, b) comparison of plastic strain amplitude and total mean strain progressions in air and 5 wt.-% NaCl solution

a)

b)

Figure 13: a) Strain and corrosion potential devel-opments in continuous load increase test with ME20 in 5 wt.-% NaCl so-lution, b) comparison of plastic strain amplitude and total mean strain progressions in air and 5 wt.-% NaCl solution

a)

b)



60 (2018) 1

of the process parameters, Materials Science and Engineering A 528 (2010), pp. 342-354 DOI:10.1016/j.msea.2010.09.025

3 K. Ralston, N. Birbilis: Effect of grain size on corrosion: A review, Corrosion 66 (2010), pp. 075005-075005-13 DOI:10.5006/1.3462912

4 P. Minárik, R. Král, B. Hadzima: Substantially higher corrosion resistance in AE42 magne-sium alloy through corrosion layer stabiliza-tion by ECAP treatment, Acta Physica Polonica A 122 (2012), pp. 614-617 DOI:10.12693/APhysPolA.122.614

5 G. Song, A. Atrens: Understanding magnesium morrosion – A framework for improved alloy performance, Advanced Engineering Materials 5 (2003), pp. 837-858 DOI:10.1002/adem.200310405

6 F. Walther: Microstructure-oriented fatigue as-sessment of construction materials and joints using short-time load increase procedure, Mate-rials Testing 56 (2014), No. 7 – 8, pp. 519-527 DOI:10.3139/120.110592

7 R. Zeng, J. Zhang, W. Huang, W. Dietzel, K. U. Kainer, C. Blawert: Review of studies on corrosion of magnesium alloys, Transactions of Nonferrous Metals Society of China 16 (2006), pp. 763-771 DOI:10.1016/S1003-6326(06)60297-5

8 P. Wittke, M. Klein, H. Dieringa, F. Walther: Corrosion fatigue assessment of creep-resist-ant magnesium alloy Mg-4Al-2Ba-2Ca in aque-ous sodium chloride solution, International Journal of Fatigue 83 (2016), pp. 59-65 DOI:10.1016/j.ijfatigue.2015.04.001

9 T. Magnin: Recent advances for corrosion fatigue mechanisms, ISIJ International 35 (1995), pp. 223-233 DOI:10.2355/isijinternational.35.223

10 M. Bobby Kannan, W. Dietzel: Pitting-induced hydrogen embrittlement of magnesium-alu-minium alloy, Materials Design 42 (2012), pp. 321-326 DOI:10.1016/j.matdes.2012.06.007

Bibliography

DOI 10.3139/120.111113Materials Testing60 (2018) 1, pages 15-21© Carl Hanser Verlag GmbH & Co. KGISSN 0025-5300

The authors of this contribution

Martin Klein, born in 1987, studied Chemical En-gineering with a focus on Technical Chemistry at Hochschule Niederrhein in Krefeld, Germany, where he received a Masters’ degree in August 2012. Since then, he has been working as a re-search assistant at Materials Test Engineering (WPT) of TU Dortmund University, Germany, working in the field of corrosion and corrosion fa-tigue testing. Philipp Wittke, born in 1982, studied Mechan-ical Engineering with specialization in Materials Science and Engineering at Ruhr-Universität Bo-chum (RUB), Germany. After his diploma thesis, he was a scientific assistant for two months at Werkstoffprüfung at RUB and, since September

the precipitated, cathodic particles, leav-ing only small pits after undercutting, which seem to be repassivated after-wards, suggesting a very good passivation behavior.

In the corrosion fatigue tests, AZ31 showed a reduction of the estimated cor-rosion fatigue strength of approx. 20 % compared to the test in air, which can be probably attributed to the pitting ten-dency, which was illustrated in the test by the fluctuating mean corrosion potential progression. A further possible effect of the pitting tendency is the observed re-duction of the plastic strain amplitude compared to the test in air. Through the pit formation bare material is exposed to the corrosive environment, which can lead to hydrogen embrittlement and sub-sequent acceleration of crack initiation and propagation. In contrast, ME20 showed no reduction of the estimated cor-rosion fatigue strength compared to the specimen tested in air. The smooth pro-gression of the mean corrosion potential suggests, that a stable passive layer is formed, which only breaks down at high cyclic loads. As a result, ME20 showed no signs of embrittlement compared to the test in air, since the base material of the specimen is probably protected by the passive layer.

In summary, the load increase test and the measured material responses showed an interesting material behavior of both materi-als, which has to be further investigated in future mechanism-oriented tests to pre-cisely assign the measured responses to the corresponding microstructural processes.

References

1 J. Liao, M. Hotta, N. Yamamoto: Corrosion be-haviour of fine-grained AZ31B magnesium al-loy, Corrosion Science 61 (2012), pp. 208-214 DOI:0.1016/j.corsci.2012.04.039

2 M. Huppmann, S. Gall, S. Müller, W. Reimers: Changes of the texture and mechanical proper-ties of the extruded Mg alloy ME21 as a function

mechanism occurs, since changes of the surface microstructure of the specimen, e. g., through formation of slip bands or twin boundaries, are indicated by the Ecorr,m decrease. The Ecorr,m decrease in turn cor-responds with the first Ecorr,a jump. Thus, it can be assumed, that the activation of the specimen surface enhances anodic dissolu-tion of the activated areas, which act then as stress concentration sites. After this first decrease, Ecorr,m starts increasing again and reaches an angular point at around N ≈ 10 × 104, accompanied by a large Ecorr,a plateau. After this angular point, Ecorr,m monotonously decreases until failure. The angular point can be probably attributed to final breakdown of the formed passivating layer under the high cyclic load at this point of the load increase test, since afterwards no repassivation can be ob-served again. Therefore, the corrosion fa-tigue strength was in this case estimated to be around σmax,e (LIT) ≈ 110 MPa, as in the test in air.

In the bar graph in Figure 14, the quan-titative results of the load increase tests are summarized. In case of ME20, σmax,e (LIT) and σmax,f (LIT) are identical for both tests in air and in 5 wt.-% NaCl solution. For AZ31, a reduction of both parameters of approx. 20 % is found when comparing the test in air to the test in NaCl solution.

Conclusions

The corrosion investigations revealed a strong pitting corrosion tendency of AZ31 and subsequent filiform corrosion pro-cesses. As a result, the corrosion resist-ance of AZ31 is much lower than that of ME20, given by the determined corrosion rate, which was approx. six times higher than that of ME20. The developed pitting corrosion morphology of AZ31 is rather unfavorable in context with corrosion fa-tigue tests, since the formed pits act as crack initiation sites. In contrast, ME20 exhibited only corrosive deterioration at

Figure 14: Comparison of estimated fatigue

strengths and maximum failure stresses

determined in load increase tests



60 (2018) 1

aurach, Germany, he took responsibility for Pub-lic Private Partnership within Corporate Develop-ment from 2008 to 2010. Since 2010, he is Pro-fessor for Materials Test Engineering (WPT) at TU Dortmund University, Germany. His research portfolio includes determination of structure-property relationships of metal- and polymer-based materials and components under fatigue loading from LCF to VHCF range, taking the in-fluence of manufacturing and joining processes as well as service loading and corrosion deterio-ration into account.

and is an expert in the processing of magnesium sheet via twin roll casting technology and their performance optimization. Prof. Dr.-Ing. Frank Walther, born in 1970, studied Mechanical Engineering majoring in Ma-terials Science and Engineering at TU Kaiser-slautern University, Germany. There he finished his PhD on the fatigue assessment of railway wheel steels in 2002 and his habilitation on phys-ical measurement techniques for microstructural-based fatigue assessment and lifetime calculation of metals in 2007. At Schaeffler AG in Herzogen-

2011, he has been working as a scientific assis-tant at Materials Test Engineering (WPT) of TU Dortmund University, Germany. Since March 2014, he is also serving as group leader of Light Metals at WPT, where he is working in the field of destructive materials testing. Dr. Roland Hoppe, born in 1968, studied Phys-ics at Christian-Albrechts University in Kiel, Ger-many. Since his Diploma in 2002, he has been working as a research scientist at the Helmholtz-Zentrum, Geesthacht (HZG), Germany. After working at the Department of Metalphysics, in-cluding finishing his PhD on deformation and fa-tigue induced internal stresses of TiAl alloys, he changed to the Magnesium Innovation Centre at HZG and since then, he has been working on the process and alloy development of magnesium wrought alloys. Dr. Dietmar Letzig, born in 1961, studied Physics at Christian-Albrechts University in Kiel, Germany, and at University Hamburg, Germany. From 1989 to 1995, he worked as a scientist in the Department of Physical Metallurgy at Max-Planck Institute for Iron Research in Düsseldorf, Germany, and obtained his PhD on intermetallic phases at Technical University RWTH Aachen, Germany in 1995. After that, he was employed as a research scientist in the R&D Center at Alcan International Limited in Banbury, UK, till 2000, and developed new 6xxx series alloys and their processing for the European automotive manufac-turing system. Since then, he is head of the De-partment Wrought Magnesium Alloys at the Helmholtz-Zentrum, Geesthacht (HZG), Germany. He is the author of many papers in peer-reviewed scientific journals and proceedings of interna-tional conferences. His area of expertise is the process-microstructure and process-property re-lationship of lightweight materials, especially of wrought magnesium alloys. He is leading funda-mental as well as application-oriented research projects in the field of metal forming processes

Abstract

Bewertung des Schwingungsrisskorrosionsverhaltens der strangge-pressten Magnesiumlegierungen AZ31 und ME20. Das Schwingungs-risskorrosionsverhalten der stranggepressten Magnesiumlegierungen AZ31 und ME20 wurde in Laststeigerungsversuchen in 5 Gew.-% NaCl-Lösung untersucht. Die Ergebnisse wurden in Beziehung zum jeweiligen Korrosionsverhalten gesetzt, das sowohl in Immersionstests als auch durch mikrostrukturelle Untersuchungen nach Vorkorrosion charakteri-siert wurde. ME20 zeigte gegenüber AZ31 einen überlegenen Korrosions-widerstand, der vermutlich einer besseren Passivierungsneigung auf-grund von Kornfeinung durch die Anwesenheit seltener Erden als Legie-rungselemente zugeschrieben werden kann. Entsprechend dieser Ergebnisse zeigte ME20 in vergleichenden Laststeigerungsversuchen an Luft und in 5 Gew.-% NaCl Lösung keine Unterschiede im zyklischen Ver-formungsverhalten, wohingegen für AZ31 sowohl eine Reduzierung der Ermüdungsfestigkeit als auch Anzeichen für Versprödung aufgrund der mechanisch-medialen Belastung beobachtet werden konnten.

CORROSION TESTING Corrosion fatigue assessment of extruded ...€¦ · sion resistance of magnesium...

Documents

Transcript of CORROSION TESTING Corrosion fatigue assessment of extruded ...€¦ · sion resistance of magnesium...