This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/copyright

Author's personal copy

Materials Science and Engineering A 523 (2009) 173–177

Contents lists available at ScienceDirect

Materials Science and Engineering A

journa l homepage: www.e lsev ier .com/ locate /msea

Small-scale mechanical property characterization of ferrite formed duringdeformation of super-cooled austenite by nanoindentation

Tae-Hong Ahn a, Kyung-Keun Um b, Jong-Kyo Choi b, Do Hyun Kim a, Kyu Hwan Oh a,Miyoung Kim a, Heung Nam Han a,∗

a Department of Materials Science and Engineering, Seoul National University, San 56-1, Shinrim-dong, Kwanak-gu, Seoul, 151-744, Republic of Koreab Plate Research Group, POSCO Technical Research Laboratories, Goedong-dong 1, Nam-gu, Pohang, 790-785, Republic of Korea

a r t i c l e i n f o

Article history:Received 20 February 2009Received in revised form 23 May 2009Accepted 25 May 2009

Keywords:Ultrafine-grained microstructureDynamic transformationPhase transformationNanoindentationMechanical propertiesHardness

a b s t r a c t

The mechanical properties of dynamically and statically transformed ferrites were analyzed using ananoindentater-EBSD (Electron BackScattered Diffraction) correlation technique, which can distinguishindenting positions according to the grains in the specimen. The dilatometry and the band slope and con-trast maps by EBSD were used to evaluate the volume fractions of two kinds of ferrite and pearlite. Fineferrites induced by a dynamic transformation had higher nano-hardness than the statically transformedcoarse ferrites. Transmission electron microscopy (TEM) showed the dynamic ferrites to have a higherdislocation density than the statically transformed ferrites.

Crown Copyright © 2009 Published by Elsevier B.V. All rights reserved.

1. Introduction

Recently, there has been increasing demand for the manufactureof ultrafine-grained ferritic steels to satisfy the increasing demandfor structural steels with improved strength and toughness [1–9].The key technology for producing ultrafine-grained ferritic steels isto impose heavy deformation to super-cooled austenite to inducea dynamic formation of ferrite grains [10]. The Strain-InducedDynamic Transformation (SIDT) of super-cooled austenite has beenhighlighted as a process that can overcome the limitations ofconventional ThermoMechanical Control Processing (TMCP). Bothdynamically and statically transformed ferrites coexist in SIDTedsteels due to the lower volume fraction of SIDTed ferrite in low car-bon steel than the equilibrium ferrite fraction at a given processingtemperature [5,7,10].

It is known that the grain size of the dynamic ferrite is finerthan that of static one [5]. Since the strength of the fine-grainedsteel has been described by the Hall-Petch relationship, which istaken the interaction between grain boundary and dislocation intoaccount, the amount and the grain size of dynamic ferrite are veryimportant in achieving the higher strength and toughness of thesteel. However, this strength evaluation from the Hall-Petch rela-

∗ Corresponding author. Tel.: +82 2 880 9240; fax: +82 2 872 8785.E-mail address: hnhan@snu.ac.kr (H.N. Han).

tionship is not based on the mechanical property itself of the insideof ferrite grain. On the contrary, there might be a possibility thatfine dynamic ferrites could show the softer mechanical behaviorbecause this dynamic ferrite transforms at higher temperature thanstatic one. In order to compare the small-scale mechanical charac-teristics of each ferrite itself by grains, a nanoindentation would bea good candidate.

In this study, quantitative dilatometric analysis was used todetermine the volume fractions of dynamic and static ferrite in lowcarbon steel during deformation and subsequent cooling, respec-tively. The small-scale mechanical properties of the two types offerrite were characterized by nanoindentation. A nanoindentater-EBSD (Electron Backscatter Diffraction) correlation technique wasused to distinguish the indenting positions according to the grainsin the specimen. TEM observations of the dislocation densitiesshowed a difference in nano-hardness between the two types offerrite.

2. Experimental

The composition of steel used in this study wasFe–0.1C–1.5Mn–0.25Si–0.05V–0.01Ti–0.04Nb–0.0036N (wt.%).The Ae3 temperature of this steel was estimated to be 810 ◦Cunder the para-equilibrium condition using Thermo-Calc [11]. Avacuum induction melted ingot was soaked and hot-rolled to a15 mm thick plate at temperatures between 1000 and 1200 ◦C.

0921-5093/$ – see front matter. Crown Copyright © 2009 Published by Elsevier B.V. All rights reserved.doi:10.1016/j.msea.2009.05.050

Author's personal copy

174 T.-H. Ahn et al. / Materials Science and Engineering A 523 (2009) 173–177

Fig. 1. Comparison of relative dilatation curves between deformed (SIDT) and unde-formed specimens during cooling at the rate of 2 ◦C/s.

Cylindrical specimens, 10 mm in diameter and 15 mm in height,were machined from those plates for thermo-mechanical pro-cessing on a Gleeble 3800. The specimens were reheated to1200 ◦C for 5 min resulting in an austenite grain size of 77 �m.The specimens were then cooled to 720 ◦C at a rate of 2 ◦C/s, anddeformed immediately with upto 70% compression at a constantstrain rate of 1.0/s. The deformation temperature was chosen tobe 20 ◦C above the Ar3 temperature of 700 ◦C. After deformation,the specimens were cooled immediately to room temperature at2 ◦C/s. The dilatometric change in the diametrical direction in themiddle of the specimens was measured over the entire process,including heating, deformation and cooling.

After thermo-mechanical processing, sample preparation fornanoindentation was performed by mechanical grinding using adiamond suspension and chemical etching with 3% nital solution toexpose clearly the grain boundary by SPM (Scanning Probe Micro-scope) equipped in the nanoindenter. The nano-hardness of thedynamically and statically transformed ferrites was measured usinga Hysitron Tribolab nanoindenter. Total 37 indentations were car-ried out inside grains one by one. The maximum load for each indentwas 800 �N. The grains could be classified into dynamic and staticferrite by measuring the grain size in the EBSD image.

An EBSD (HKL Nordlys Channel 5) system was used to measurethe crystallographic orientation and size of each ferrite grain. A totalarea of 1400 × 1000 pixels was indexed with a step size of 0.1 �m.The indentation positions were determined by the grains using bothEBSD and SPM image. In addition, the dislocation densities in thedynamically- and statically-transformed ferrites were observed byTEM (Tecnai F20). Thin foils were prepared for the TEM observationby mechanical polishing and electropolishing with a solution con-taining 10% perchloric acid and 90% methanol followed by Ar+ ionmilling to remove the oxidized or contaminated surface layer.

3. Results and discussion

The austenite remaining in the specimen that is subject to imme-diate cooling to room temperature after deformation at above theAr3 temperature and below the Ae3 temperature undergoes a staticaustenite-to-ferrite and austenite-to-pearlite transformation. Thevolume fraction of the statically transformed phase after SIDT couldbe evaluated by a comparison with the relative dilatation curveduring continuous cooling of the specimen without deformation.Fig. 1 shows the relative dilatation curves for both the deformedand undeformed samples during cooling at 2 ◦C/s. The volume frac-tion of the statically transformed phase, fS, in the SIDTed specimenwas determined by applying the lever rule [12–17] to the dilatation

Fig. 2. (a) Grain size map measured by EBSD for the specimen undergone SIDT,(b) band contrast and slope combined map for classifying pearlite and (c) imageof microstructure classified into the dynamically and statically transformed ferritesand the pearlite.

curves. For the lever rule, the dilatations curve of the undeformedspecimen was used as the linear segment in the single austeniteregion. The relative dilatometric curve was used to compensate forlarge changes in the length of the dilatometric specimen due to theheavy compressive deformation. The volume fraction of dynamicferrite, fD, was obtained from the equation, fT = fS + fD , where fT isthe volume fraction of the total transformed phase. In this case,the volume fractions of dynamic ferrite and the static phase weredetermined to be 0.48 and 0.52, respectively. The volume fractionof the statically-transformed pearlite after SIDT was determined tobe approximately 0.1 from optical microscopy and the combinedband contrast and slope map [18] obtained from EBSD, as shown inFig. 2(b). Therefore, the volume fraction of static ferrite is 0.42.

Author's personal copy

T.-H. Ahn et al. / Materials Science and Engineering A 523 (2009) 173–177 175

Fig. 3. (a) Band contrast image measured by EBSD after indentations conducted, (b) SPM images with indentation marks: red, blue and black circles indicate the dynamicallyand statically transformed ferrite and the pearlite, respectively, and (c) load–displacement curves for the clear indents, which exist inside a grain (For interpretation of thereferences to color in this figure legend, the reader is referred to the web version of the article)

Author's personal copy

176 T.-H. Ahn et al. / Materials Science and Engineering A 523 (2009) 173–177

Fig. 4. Bright field images obtained by TEM for (a) dynamically and (b) statically transformed ferrites.

Fig. 2(a) shows the grain size map of the SIDTed specimenmeasured by EBSD. The fraction of well-indexed pixels was approx-imately 96%, and the un-indexed pixels around grain boundarieswere filled by extrapolation using the information of the adjacentpixels. The grain boundaries were defined as high angle ones witha misorientation angle >15◦. As shown in this figure, two typesof ferrite grain can be observed: very fine ferrite grains formednear the prior austenite grain boundaries and coarse ferrite grainsin the prior austenite grains. The dynamically transformed ferritegrains were much finer than the statically transformed grains. Thisbimodal grain size distribution in the ferrite phase might makeit possible to classify the dynamically and statically transformedferrite. The critical grain size was estimated to be 5.7 �m, which cor-responds to a volume fraction of 0.48 for dynamic ferrite based onthe dilatometry results in Fig. 1. The statically transformed pearliteafter SIDT was classified from the band contrast and slope combinedmap [18], as shown in Fig. 2(b). An image of the microstructure clas-sified into the dynamically- and statically-transformed ferrite, andthe statically-transformed pearlite was obtained from the criterionof grain size, as shown in Fig. 2(c).

Nanoindentation was used to compare the small-scale mechani-cal properties of the two types of ferrite. The nanoindentater-EBSDcorrelation technique was suggested due to the difficulty in clas-sifying grains, i.e. dynamic and static ferrites and pearlite, usingSPM only. Fig. 3(a) shows a band contrast image by EBSD. As shownin this figure, two types of ferrite grain can be well classified.In five blue marked regions, the indentations were performed asshown in Fig. 3(b). Among the indentation points, only the clearindents inside a grain were selected and classified into dynamic andstatic ferrites based on the criterion of grain size. Fig. 3(c) showstheir load–displacement curves. The mean nano-hardness of thedynamic and static ferrites was 3.71 ± 0.11 GPa and 3.23 ± 0.02 GPa,respectively. The nano-hardness of pearlite was much higher thanthose of the ferrites.

The dynamic ferrite was approximately 15% harder than thestatic ferrite. This is in contrast to the reports showing that thetransformation temperature of dynamic ferrite is higher than thatof static ferrite. There are some reports showing that supersaturatedcarbon in dynamic ferrite forms very small cementite particles onthe ferrite boundary [10,19]. This supersaturated carbon in dynamicferrite is expected to be a strengthening mechanism. In addition, theplastic deformation occurring after SIDT might increase the disloca-tion density in dynamic ferrite. A large number of TEM observationsof the inside of the grains were made to confirm the difference indislocation density between the two types of ferrite grains. The dis-location density was calculated from about 40 observed images foreach ferrite grains. Fig. 4 shows typical TEM images of the two typesof ferrite grains that contain dislocation structures. The averagedislocation densities of the dynamic and static transformed were2.7 × 1014 m−2 and 8.7 × 1013 m−2, respectively. A higher density ofdislocations of dynamic ferrite is due to work-hardening after SIDT.The higher dislocation density may also be one of the strengthen-ing mechanisms of dynamic ferrite. This difference in dislocationdensity between the two types of ferrites may be used to help dis-tinguish them from a microstructural analysis based on anotherEBSD technique, such as a local misorientation map or grain averagemisorientation.

4. Conclusion

In conclusion, this study examined the nanoindention prop-erties of fine-grained ferrite transformed dynamically by heavydeformation using a nanoindentater-EBSD correlation technique,which can distinguish the indenting positions according to thegrains in the specimen. The critical grain size for classifying thedynamically and statically transformed ferrites was determinedby dilatometry analysis. In addition, the volume fraction of thestatically-transformed pearlite after SIDT could be determined

Author's personal copy

T.-H. Ahn et al. / Materials Science and Engineering A 523 (2009) 173–177 177

from the combined band contrast and slope map from EBSD. Thefine ferrites induced by the dynamic transformation had highernano-hardness than the static ones. Overall, fine dynamically-transformed ferrite grains have a higher dislocation density thanstatically transformed ferrite grains.

Acknowledgement

This work was supported by the Korea Science and EngineeringFoundation (KOSEF) grant funded by the Korea government (MOST)(R0A-2007-000-10014-0). Drs. Um and Choi would like to thank forthe support by one of “National mid-term key technology projects”(10028396-2008-13).

References

[1] H. Yada, C.M. Li, H. Yamagata, ISIJ Int. 40 (2000) 200.[2] P.J. Hurley, P.D. Hodgson, Mater. Sci. Eng. A 302 (2001) 206.

[3] D.H. Shin, I. Kim, J. Kim, K.-T. Park, Acta Mater. 49 (2001) 1285.[4] H. Hidaka, T. Tsuchiyama, S. Takaki, Scripta Mater. 44 (2001) 1503.[5] J.-K. Choi, D.-H. Seo, J.-S. Lee, K.-K. Um, W.-Y. Choo, ISIJ Int. 43 (2003) 746.[6] H. Hou, Q. Chen, Q. Liu, H. Dong, J. Mater. Process. Technol. 137 (2003) 173.[7] H. Beladi, G.L. Kelly, A. Shokouhi, P.D. Hodgson, Mater. Sci. Eng. A 371 (2004)

343.[8] Y. Adachi, K. Tsuzaki, Metall. Mater. Int. 11 (2005) 341.[9] Y. Kimur, T. Inoue, F. Yin, K. Tsuzaki, Science 320 (2008) 1057.

[10] H.-C. Lee, K.-K. Um, ISIJ Int. 48 (2008) 1050.[11] B. Sundman, B. Jansson, J.-O. Andersson, Calphad 9 (1985) 153.[12] ASTM International, Designation: A 1033-04.[13] D.-W. Suh, C.-S. Oh, H.N. Han, S.-J. Kim, Acta Mater. 55 (2007) 2659.[14] D.-W. Suh, C.-S. Oh, H.N. Han, S.-J. Kim, Metall. Mater. Trans. A 38 (2007)

2963.[15] H.N. Han, J.K. Lee, D.-W. Suh, S.-J. Kim, Philos. Mag. 87 (2007) 159.[16] H.N. Han, S.-J. Kim, M. Kim, G. Kim, D.-W. Suh, S.-J. Kim, Philos. Mag. 88 (2008)

1811.[17] X. Sun, H. Dong, Q. Liu, Y. Weng, Mater. Sci. Eng. A 487 (2008) 93.[18] H.-C. Lee, K.H. Oh, Y.-W. Kim, H.N. Han, J.Y. Kang, S.I. Baek, D.H. Kim, T.-H. Ahn,

Quantitative Characterization of Phase Volume Fraction by EBSD Technique,Technical Report of POSCO Research Institute, 2007.

[19] H. Mabuchi, T. Hasegawa, T. Ishikawa, ISIJ Int. 39 (1999) 477.