1. Introduction

The weight reduction for fuel efficiency as well as thesafety is continuing to be the major concerns in automobilemanufacturing. To meet such requirements, investigation onthe development of high strength steels with excellentformability for the fabrication of various automobile partsis being carried out.

It is well known that transformation induced plasticity(TRIP)-aided steels,1) exhibit a superior combination ofstrength and ductility compared to precipitation hardenedand solution hardened steels. The strain induced transfor-mation of any retained austenite present in TRIP-aidedsteels occurs during deformation leading to the formationof strain induced martensite, so that a good combination offormability and strength can be obtained.2) Employing theTRIP phenomenon to the production of ultra high strengthsteel without loss of formability has been an interest ofmany steel makers.3–5)

It has been known that carbon, manganese and siliconare the most effective alloying elements for TRIP-aidedsteels. Such elements contribute to stabilizing the retainedaustenite due to the inhibition of carbide precipitation dur-ing bainite reaction.6) However, steels with high contents ofsilicon pose difficulties in the welding process. In order toreduce the content of such elements without loss of strengthand ductility, some investigations have been carried out7,8)

and also some other alternatives have been proposed. The

TRIP-aided steel containing aluminum in place of siliconwas also proposed and the effects of aluminum and siliconon the TRIP properties was compared.9) The correspondingphase change was examined for various heat cycles to findthe optimum heat treatment for the TRIP steels.10) The ef-fect of deformation during intercritical annealing on themorphology of phases has been investigated for hot rolledTRIP steels11,12) and the cold rolled steels with microalloy-ing elements like Nb13–15) and P16) have also been studied.

However the effect of nitrogen on TRIP phenomena hasbeen rarely reported though nitrogen addition can form pre-cipitates to affect on behavior of transformation in TRIP-aided steels. In this study, the effect of nitrogen on the coldrolled TRIP-aided steel on microstructure and mechanicalproperties were investigated.

2. Experiments

2.1. Material

In this study, two kinds of steels, S and S-N, were investi-gated in order to study the effects of nitrogen on the me-chanical properties. The chemical compositions of thosesteels are shown in Table 1. The nitrogen content of0.003% in steel S was an impurity that unavoidably re-mained in the melting process. On the contrary, for thesteels with nitrogen content of higher than 0.010%, nitro-gen was intentionally added using nitrogen gas during themelting process. Aluminium was added in order to simulate

ISIJ International, Vol. 46 (2006), No. 4, pp. 599–605

599 © 2006 ISIJ

Effects of Nitrogen on the Mechanical Properties of Cold RolledTRIP-aided Steel Sheets

Seung Chul BAIK,1) Sung-Ho PARK,1) Ohjoon KWON,1) Dong-Ik KIM2) and Kyu Hwan OH3)

1) Technical Research Laboratories, Pohang Iron & Steel Co. Ltd., Pohang, 790-785, Korea. E-mail: scbaik@posco.co.kr2) Advanced Materials Research Center, Korea Institute of Science and Technology, 39-1 Hawolgok-dong, Seongbuk-gu, 136-791 Seoul, Korea. E-mail: dongikkim@kist.re.kr 3) School of Materials Science and Engineering, Seoul National Univ.,Seoul 151-742, Korea.

(Received on October 7, 2005; accepted on December 20, 2005 )

Tensile tests were conducted in order to study the effects of nitrogen on the mechanical properties of the0.2C–1.5Mn–1.5Si–0.04Al–(0.003–0.015)N steels often annealing at 800–830°C, followed by the austemper-ing at 400–450°C. The results show that both tensile strength and elongation increase and the balance oftensile strength�elongation are improved upon nitrogen addition to the TRIP steel. It was found that thedensity of AlN precipitates increases with addition of nitrogen. The volume fraction of retained austenite in-creases due to AlN precipitation, because the precipitates retard the transformation of austenite during cool-ing and austempering. As a larger volume fraction of austenite remained after annealing has been trans-formed to martensite during deformation, elongation as well as strength has increased in the nitrogenadded steel. It is also observed that the average grain size of ferrite and bainite decreases because of AlNprecipitation that hinders grain growth. The refinement of ferrite and bainite by AlN precipitates also con-tributes to the increase in the strength of nitrogen added steels.

KEY WORDS: nitrogen; AlN; precipitate; strength; elongation; strain hardening; TRIP.

the industrial steels. It is to be mentioned here that alumini-um addition was done to scavenge oxygen in the refiningprocess, and it remains as an impurity (0.02–0.05%) forcold rolled steels.

2.2. Processing and Characterization

The steels used in present investigation were vacuummelted and subsequently hot rolled to 30 mm in the labora-tory. The hot rolled plates were reheated at 1 250°C and hotrolled to 3.2 mm with finishing temperature as 910°C, fol-lowed by soaking at 600°C for 1 h and furnace cooling toroom temperature. This was done in order to simulate thecoiling process of hot bands. The hot rolled steels werepickled, cold rolled to a thickness of 1.4 mm followed byannealing using a continuous annealing simulator (ULVACCCT-Y8).

Tensile specimens with a gauge length of 50 mm and awidth of 12.5 mm were made of the cold rolled steel sheetsalong the rolling direction. The specimens were heat treatedaccording to the schedule given in Fig. 1, which reproducesthe industrial continuous annealing process. The effect oftemperatures on the mechanical properties was measuredusing the specimens which were heated at the intercriticaltemperature in the range of 800–830°C and with theaustempering temperature in the range of 400–450°C.

In order to determine the change in the amount of re-tained austenite during deformation, the volume fraction ofretained austenite was measured for the tensile specimenswhich were uniaxially tensioned by 0–25%. The retainedaustenite fractions were calculated by Equation (1) usingthe data measured by X-ray diffraction analysis with Cu-Karadiation.

...........(1)

where Ia(200) is the integrated intensity of (200) peak in

the ferrite, and Ig(200) and Ig(220) are the integrated inten-sities of (200) and (220) peaks in the austenite, respectively,and Vc is the percentage of cementite.

The distribution of retained austenite phase was exam-ined using an electron back scattering diffraction (EBSD).Using the Oxford INCA Crystal EBSD system installed onJEOL 6500F schottky type field emission gun scanningelectron microscope, the EBSD patterns were measured andeach pattern was identified as ferrite and residual austeniteindependently. Transmission electron microscopy was alsocarried out in order to examine the shape of retainedaustenite and precipitation. For this purpose, thin foils wereprepared using conventional mechanical and twin jet elec-tropolishing, and these were examined with JEOL 200CXtransmission electron microscope.

3. Results

3.1. Mechanical Properties

Figure 2 shows the variation of tensile strength (TS),elongation (El) and yield strength (YS) as a function of thenitrogen content in the experimental 0.2C–1.5Mn–1.5Si–0.04Sol.Al steels. As mentioned earlier, a nitrogen contentof 0.003% unavoidably remained as an impurity in themelting process. On the contrary, nitrogen was intentionallyadded using nitrogen gas in the melting process for thesteels with nitrogen higher than 0.010%. It has been shownthat both the strength and the elongation for the steels withnitrogen content 0.010% and above are higher than those ofthe steel with 0.003% nitrogen. Therefore the balance ofstrength�elongation (Ts�El) was found higher for the ni-trogen added steels than the steels without intentional addi-tion of nitrogen. In order to clarify the effect of nitrogen onthe mechanical properties of TRIP-aided steel, two kinds ofsteels, S and S-N, were chosen and analysed in detail. SteelS represents the steel with 0.003% nitrogen (as an impuri-ty), and steel S-N denotes the steel with 0.010% nitrogenthat was intentionally added.

VV

I

I I

γα

γ γ

(%)(%)

.( )

( ) ( )

��

��

100

1 4 27200

200 220

c

ISIJ International, Vol. 46 (2006), No. 4

© 2006 ISIJ 600

Table 1. Chemical composition of examined steels (mass con-tent in %).

Fig. 1. Schematic diagram of heat cycle.

Fig. 2. Mechanical properties as a function of N Content for thesteels annealed at 830°C, followed by austempering at400°C.

The mechanical properties of steel S and S-N were com-pared when specimens were heat treated under various con-ditions. It has been shown in Fig. 3 that both the strengthand the elongation of steel S-N are higher than those ofsteel S when specimens were heat treated in the intercriticalannealing temperature of 800–830°C. Figure 4 also indi-cates that both the strength and the elongation of steel S-Nare higher than those for the steel S regardless of austem-pering temperature in the range of 400–450°C.

The flow curves obtained by uniaxial tension tests arepresented in Fig. 5 along with the curves of strain–strain hardening rate. It is apparent from these plots that thehardening rate of steel S-N becomes higher than that of

steel S at the true strain of 0.05 and the higher value ismaintained up to the strain of necking. In general, the trans-formed martensite increases the strain hardening rates dueto the dislocations generated at the interface between ferriteand martensite.17) In order to examine the change in volumefraction of retained austenite during deformation, X-ray dif-fraction technique was employed. Volume fraction wasmeasured for the specimens uniaxially tensioned by therange of 0–25%. Figure 6 shows that initial volume frac-tion of retained austenite in steel S-N is higher than that ofsteel S. However, the volume fraction of retained austenitein steel S-N became smaller when the specimens were ten-sioned larger than the strain of 0.05.

3.2. Microstructure

The microstructures of steel S and S-N after annealingwere observed using SEM. As shown in Fig. 7, fine secondphase particles form in steel S-N. Since it was difficult todistinguish retained austenite from the microstructure con-taining various second phases using SEM, the distributionof retained austenite was examined using EBSD. Figure 8contains the retained austenite distribution measured usingEBSD with respective SEM images for steel S and S-N.The EBSD maps were obtained in the position correspon-dent to the SEM images. It can be seen that the volumefraction of retained austenite in steel S-N is higher than thatof steel S. The grain size was also measured forferrite�bainite and retained austenite, respectively usingEBSD (Fig. 9). The result indicates that the average grain

ISIJ International, Vol. 46 (2006), No. 4

601 © 2006 ISIJ

Fig. 3. Mechanical properties as a function of intercritical an-nealing temperature for the steels austempered at 400°C.

Fig. 4. Mechanical properties as a function of austempering tem-perature for the steels annealed at 830°C.

Fig. 5. The changes of flow stress (s) and strain hardening rate(ds /de) as a function of true strain.

Fig. 6. The changes of percent of retained austenite as a functionof true strain.

size of ferrite and bainite decreases with the increase in ni-trogen content while the addition of nitrogen has little effecton the average grain size of retained austenite.

It has been understood that nitrogen forms AlN in com-bination with Al (Fig. 10). As compared to steel S, moreprecipitates of AlN were observed in steel S-N due to high-er content of nitrogen. The AlN precipitates in steel S andS-N differ in morphology and consequent grain shapes ofretained austenite (Fig. 11). While the grains of retained

austenite in steel S were smoothly rounded, many irregularboundaries of retained austenite were observed in steel S-Nas shown in Fig. 11(b). It was also found that the AlN pre-cipitates were present around the irregular boundaries of re-tained austenite in steel S-N.

While the volume fraction of retained austenite in steelS-N was higher than that in steel S before deformation, andit became lower as compared to steel S after uniaxial ten-sion. Figure 12 displays the microstructures recorded usingSEM and EBSD. These micrographs clearly reveal that thevolume fraction of retained austenite in steel S is higherthan that of steel S-N after uniaxial tension. These resultsverify the results of austenite volume fraction measure-ments using X-ray diffraction analysis (Fig. 6).

ISIJ International, Vol. 46 (2006), No. 4

© 2006 ISIJ 602

Fig. 7. Scanning electron micrographs of the cold rolled sheet ofsteels (a) S and (b) S-N annealed at 830°C, followed byaustempering at 400°C.

Fig. 8. Scanning electron micrographs and morphology of re-tained austenite observed using EBSD in the cold rolledsheet of steels (a) S and (b) S-N annealed at 830°C, fol-lowed by austempering at 400°C.

Fig. 9. Averaged grain sizes of ferrite+bainte and retainedaustenite obtained by EBSD in the cold rolled sheet ofsteels S and S-N annealed at 830°C, followed by austem-pering at 400°C.

Fig. 12. Scanning electron micrographs and morphology of re-tained austenite observed using EBSD in the cold rolledsheet of steels (a) S and (b) S-N after the deformation inuniaxial tension by true strain of 0.05.

4. Discussion

4.1. Effect of AlN Precipitation on the RetainedAustenite Characteristics

It is well known that microstructure and mechanicalproperties of cold rolled TRIP steels are strongly dependenton the annealing conditions. Referring to the typical heatcycles of TRIP steels applied at the continuous annealingline as shown in Fig. 1, it is clear that during intercriticalannealing, austenite forms in two main steps.19,20) Initially,almost instantaneous nucleation of austenite takes place onpearlite colonies or at grain boundary carbides, and subse-quently these nuclei grow slowly into the ferrite. During an-nealing for a short time, the annealing temperature must besufficiently high to ensure complete dissolution of carbidesand to form a dual phase microstructure consisting of fer-rite and austenite.21) It was confirmed that more than 50%austenite was formed steel S and S-N at the temperaturehigher than 800°C using a dilatometer. The austenite istransformed to bainite and carbon is diffused to austeniteduring austempering. The retained austenite becomes stableand the martensite transformation temperature decreases toa temperature lower than room temperature because of highcontent of carbon in the retained austenite.22) Therefore themechanical properties and the morphology in TRIP-aidedsteels are found strongly dependent on the austemperingcondition.7,10,23)

The precipitates can also affect the transformation andgrain growth in the TRIP-aided steels during the annealingprocess. In order to clarify the effect of precipitate on themorphology and mechanical properties, it is necessary to

form precipitates without solid solution of added elements.Though the effect of Nb has been studied by many re-searchers14,15,24) and Nb(CN) precipitates have been ob-served in the Nb added TRIP steels,22) it is difficult to ex-tract the effect of precipitates in the Nb added steel becauseof Nb in solid solution. However most amount of nitrogenadded in Steel S-N has been precipitated so it is possible toanalyze the effect of precipitate in this study. Figure 13shows the calculated results of AlN mass fraction in steel Sand S-N as a function of temperature in the equilibriumstate that was calculated using the program Thermo-Calc.These results confirm that the entire amount of nitrogengets precipitated to form AlN in the equilibrium state, and

ISIJ International, Vol. 46 (2006), No. 4

603 © 2006 ISIJ

Fig. 10. Transmission electron micrographs showing the distrib-ution of AlN in the cold rolled sheet of steels (a) S and(b) S-N annealed at 830°C, followed by austempering at400°C.

Fig. 11. Transmission electron micrographs showing the shapeof retained austenite in the cold rolled sheet of steels (a)S and (b) S-N annealed at 830°C, followed by austem-pering at 400°C.

Fig. 13. The change of weight fraction of AlN in the equilibri-um state as a function of temperature calculated usingthe program Thermo-Calc.

the fraction of AlN in steel S-N is three times as much asthat of steel S. The observation using TEM also showedhigh density of AlN precipitates in the steel S-N, as shownin Figs. 10 and 11. The effect of small amount of solute ni-trogen can be neglected so that this study focuses on the ef-fect of nitrogen on morphology of precipitates and mechan-ical properties of TRIP.

These precipitates could retard the transformation andgrain growth during cooling from the intercritical annealingtemperature and austempering. The austenite formed in theintercritical annealing was transformed to ferrite or bainiteduring cooling and austempering annealing, and the un-transformed austenite remained as the retained austeniteafter annealing.18) During transformation of austenite tobainite, the precipitates may retard the transformation rateby dragging the grain boundaries, and the irregular grainboundaries of retained austenite (Fig. 11(b)) can be ex-plained by this reaction between the precipitates and grainboundaries during transforming of austenite. This also ex-plains the retention of high volume fraction of austenite insteel S-N as apparent from Fig. 8 and small average grainsize of ferrite and bainite, as shown in Fig. 9.

In addition, it was confirmed in this study that the phasesin the TRIP-aided steels could be well characterized by theanalysis using EBSD as shown in Fig. 8. The accurate de-termination of the volume fraction of retained austenite isof great importance for the optimization of TRIP-aidedsteels and many methods25–27) have been developed to mea-sure retained austenite. Among the developed methods, theanalysis using EBSD is most useful when the size distribu-tion of retained austenite is to be determined.

4.2. Mechanical Properties of Nitrogen Added Steel

As shown in Fig. 2, the tensile strength increases by30–50 MPa and the elongation increases by 5–7%, when nitrogen was intentionally added. The balance of tensilestrength�elongation gets enhanced by about 30% of theoriginal value by the intentional addition of 0.01 wt% nitro-gen. It has been shown in Figs. 3 and 4 that the nitrogenadded steels was superior to the TRIP steel without additionof nitrogen regardless of annealing temperatures. These re-sults reveal that the addition of nitrogen has strong effect toenhance the strength and elongation with low cost.

The changes of volume fraction of retained austenite andstrain hardening rate as a function of strain are presented inFigs. 5 and 6, respectively. These results indicate that thevolume fraction of retained austenite in steel S-N was high-er than that of steel S before the deformation, however, itbecomes smaller after the uniaxial tension by the true strainof 0.05. The hardening rate of steel S-N becomes higherthan that of steel S at the true strain of 0.05 and maintainedthe higher value up to the strain of necking. It is reportedthat the transformation of retained austenite to martensiteduring deformation leads to dislocation generation and thusresults in a flow stress increase.28) Besides, the transformedmartensite causes increase in the strain hardening rates be-cause of dislocations generated at the interfaces betweenferrite and martensite which subsequently get multiplied inthe ferrite matrix as the material underwent further defor-mation.17) Also, the transformation induced deformationgives rise to an increased resistance to necking, thus lead-

ing to an improved uniform elongation.28) A previous studyon simulation of TRIP behavior29) has also shown that theelongation of TRIP steel gets enhanced because the trans-formation of retained austenite suppressed the start of neck-ing. Therefore, the higher strength and elongation of steelS-N can be explained in term of the transformation of largevolume fraction of retained austenite during deformation.The refinement of ferrite and bainite by the precipitate ofAlN (Fig. 9) also contributed to the increase in strength ofnitrogen added steels.

It has been reported that the stability of retained austeniteagainst a strain induced transformation to martensite is in-creased with decreasing size.16) Therefore, large sized re-tained austenite was unstable and transforms to martensiteat low strain, while smaller sized retained austenite wasmore stable and could be retained to higher strains.13) Thework by Jeong et al.30) has shown that most of retainedaustenite with particle size larger than 1 mm gets trans-formed to martensite at strain about 5%. However, there isno unique agreement about the effect of retained austeniteparticle size on ductility. Some researchers13) have suggest-ed that retained austenite with particle sizes smaller than 1mm contributes to the TRIP effect. While the others16) re-ported that the small austenite particles have little effect onductility because the retained austenite with submicron sizeshows mechanical stability even after 10% deformation. Itis shown in Fig. 8(b), that large particles of retained austen-ite larger than 1 mm exist with small particles. It can be sug-gested that large particles gets transformed up to deforma-tion 5%, which results in an increased strain hardening rateand its high value was maintained by the transformation ofsmall particles of retained austenite after this level of defor-mation in steel S-N.

5. Conclusions

In order to analyze the effect of addition of N in the coldrolled TRIP-aided steel, the mechanical properties andmorphology were studied for the 0.2C–1.5Mn–1.5Si–0.04Al–(0.003–0.015)N steels annealed at 800–830°C, andaustempered at 400–450°C. The following conclusions canbe drawn:

(1) Both the tensile strength and the elongation in-creased by addition of nitrogen, so that the balance betweentensile strength and elongation was improved.

(2) The density of AlN precipitates increased when Nwas intentionally added.

(3) The precipitation of AlN had a retarding effect onthe transformation of austenite during cooling and austem-pering, and resulted in an increased volume fraction of re-tained austenite with an increase of the nitrogen content.

(4) The results obtained, viz. the change of the volumefraction of the retained austenite and strain hardening in asteel with nitrogen addition, indicated that the transforma-tion of a large volume of retained austenite gave rise to ahigh strength and good ductility.

(5) The average grain size of ferrite and bainite de-creased due to AlN precipitates that hindered grain growth.The refinement of ferrite and bainite by the AlN precipi-tates also contributed to an increase in strength of N addedsteels.

ISIJ International, Vol. 46 (2006), No. 4

© 2006 ISIJ 604

REFERENCES

1) V. F. Zackay, E. R. Parker, D. Fahr and R. Busch: Trans. Am. Soc.Met., 60 (1967), 252.

2) G. B. Olson and M. Cohen: Metall. Trans. A, 7A (1976), 1897.3) K. Sugimoto, N. Usui, M. Kobayashi and S. Hashimoto: ISIJ Int., 32

(1992), 1311.4) O. Matumura, Y. Sakuma and H. Takechi: Trans. Iron Steel Inst.

Jpn., 27 (1987), 571.5) Y. Sakuma, O. Matumura and H. Takechi: Metall. Trans. A, 22A

(1991), 489.6) H. K. D. H. Bhadeshia and D. V. Edomonds: Metall. Trans. A, 10A

(1979), 895. 7) P. Jacques, E. Girault, T. Catlin, N. Geerlofs, T. Kop, S. van der

Zwaag and F. Delannay: Mater. Sci. Eng. A, A273–275 (1999), 475.8) M. H. Saleh and R. Priestner: J. Mater. Process. Technol., 113

(2001), 587.9) E. Girault, A. Mertens, P. Jacques, Y. Houbaert, B. Verlinden and J.

Van Humbeeck: Scr. Mater., 44 (2001), 885.10) E. Girault, P. Jacques, P. Ratchev, J. Van Humbeeck, B. Verlinden and

E. Aernoudt: Mater. Sci. Eng. A, A273–275 (1999), 471.11) A. Basuki and E. Aernoudt: Scr. Mater., 40 (1999), 1003.12) A. Basuki and E. Aernoudt: J. Mater. Process. Technol., 89–90

(1999), 37.13) D. Q. Bai, A. D. Chiro and S. Yue: Mater. Sci. Forum, 284–286

(1998), 253.14) W. Bleck, K. Hulka and K. Papamentellos: Mater. Sci. Forum,

284–286 (1998), 327.15) A. Z. Hanzaki, P. D. Hodgson and S. Yue: Metall. Trans. A, 28A

(1997), 2405.

16) H. C. Chen, H. Era and M. Shimizu: Metall. Trans. A, 20A (1987),437.

17) M. F. Ashby: Strengthening Methods in Crystals, John Wiley andSons, New York, (1971), 137.

18) T. Minote, S. Torizuka, A. Ogawa and M. Niikura: ISIJ Int., 36(1996), 201.

19) G. R. Speich: Fundamentals of Dual Phases Steels, TMS-AIME,Warrendale, PA, (1981), 4.

20) G. R. Speich, V. A. Demarest and R. L. Miller: Metall. Trans. A, 12A(1981) 1419.

21) D. T. Llewellyn and D. J. Hillis: Ironmaking Steelmaking, 23 (1996),471.

22) S. C. Baik, S. Kim, Y. S. Jin and O. Kwon: ISIJ Int., 41 (2001), 290.23) A. Z. Hanzaki, P. D. Hodgson and S. Yue: ISIJ Int., 35 (1995), 79.24) E. V. Pereloma, I. B. Timokhina and P. D. Hodgson: Mater. Sci. Eng.

A, A273–275 (1999) 448. 25) L. Zhao, N. H. van Dijk, E. Brück, J. Sietsma and S. van der Zwaag:

Mater. Sci. Eng. A, A313 (2001), 145.26) E. Girault, P. Jacques, Ph. Harlet, K. Mols, J. Van Humbeeck, E.

Aernoudt and F. Delannay: Mater. Character., 40 (1998), 111.27) B. Hutchinson, L. Ryde, E. Lindh and K. Tagashira: Mater. Sci. Eng.

A, A257 (1998), 9.28) J. M. Rigsbee and P. J. Vander Arend: Formable HSLA and Dual

Phase Steels, ed. by A. T. Davenport, TMS-AIME, Warrendale, PA,(1979), 56.

29) N. Tsuchida and Y. Tomota: Mater. Sci. Eng. A, A285 (2000), 345.30) W. C. Jeong, D. K. Matlock and G. Krauss: Mater. Sci. Technol., 165

(1993), 9.

ISIJ International, Vol. 46 (2006), No. 4

605 © 2006 ISIJ