Materials and Design 30 (2009) 129–134

Contents lists available at ScienceDirect

Materials and Design

journal homepage: www.elsevier .com/locate /matdes

Weather resistance of low carbon high performance bridge steel

Jia Guo *, Chengjia Shang, Shanwu Yang, Hui Guo, Xuemin Wang, Xinlai HeSchool of Materials Science and Engineering, University of Science and Technology Beijing, Handian District, No. 30 Xueyuan Road, Beijing 100083, PR China

a r t i c l e i n f o

Article history:Received 5 November 2007Accepted 16 April 2008Available online 25 April 2008

Keywords:A. metal matrixE. corrosionG. metallography

0261-3069/$ - see front matter � 2008 Elsevier Ltd. Adoi:10.1016/j.matdes.2008.04.038

* Corresponding author. Tel.: +86 1062334979; faxE-mail address: guojia1017@sohu.com (J. Guo).

a b s t r a c t

Weathering resistance of low carbon (0.03–0.05 wt.% C) micro-alloyed bainitic steel made in laboratoryscale was studied and compared with that of conventional weathering steel 09CuPCrNi by wet–dry cycletest in a 3.5 wt.% NaCl aqueous solution. The results show that corrosion resistance of the bainitic steel isbetter than that of 09CuPCrNi, attributed to homogenous bainite microstructure. Subsequently, this steelwas trial-manufactured in industrial plant, and the yield strength of which achieves 500 MPa grade, addi-tionally, the gauge of plate steel reaches 80 mm, and excellent mechanical properties across the thicknesssection are obtained. Microstructures of the plate are homogenous bainite at surface layer, which contrib-utes to excellent corrosion resistance of the industry trail steel.

� 2008 Elsevier Ltd. All rights reserved.

1. Introduction

Recently, bridge steel without painting or other protective coat-ings has been used widely for the sake of environment protectionand low cost [1]. Therefore, it is of great importance to improvethe weathering resistance of steel, especially for the high perfor-mance steel, as the thickness of which is thinner than that of con-ventional weathering steel.

Microstructures of conventional weathering bridge steels aremainly ferrite and pearlite, such as Cor–Ten (conventional carbonand weathering steel) series steel developed by American in1933, BS968 steel made by Britain and SMA steel made by Japa-nese. However, those microstructures often lead to low strength le-vel. Therefore, low carbon bainitic steels serving as bridge steelbecome a trend because of its high strength, excellent weld prop-erties and economical manufacture process. On the other hand,reducing carbon contents is advantage not only to improve weldproperties but also to obtain homogenous bainite in heavy platefor the low carbon microalloyed steel. Meanwhile, strength lostdue to decrease of carbon content can be compensated thoughalloying technology and thermo-mechanical control process(TMCP) [2].

Chloride ion in marine atmosphere is one of the important fac-tors leading to corrosion. Many authors have focused on the influ-ence of alloying elements [3–5] and rust structure [6–8]. Recently,several studies about effect of microstructure on corrosion behav-ior have been carried out. Zhang et al. [9] reported that pearlite inweathering steel 09CuPCrNi deteriorated the corrosion resistanceas revealed using an artificial atmosphere-salt spray testing unit.Chen [10] studied corrosion resistances of low carbon micro-al-

ll rights reserved.

: +86 1062332428.

loyed steels with different microstructures (pure ferrite, ferriteand pearlite, ferrite and bainite), the results showed that corrosionrates were different even with the same chemical composition.Zhao et al. [11] developed 750 MPa grade low carbon microalloyedsteels, the corrosion resistance of which was better than that ofCor–Ten A in wet–dry cyclic corrosion test, because negative effectof pearlite on corrosion was eliminated through optimizing themicroalloying elements technology and TMCP process. However,little research has been done with respect to the corrosion behav-iors of 500 MPa grade level thick plates in chloride environment,which are applied to bridge structure parts.

The aims of the present work were to study the initial corrosionbehavior of bainitic steel with carbon content ranged in 0.03–0.05 wt.% and compare it with conventional weathering steel(WS) 09CuPCrNi using wet/dry cyclic test. 09CuPCrNi steel wasindustrial produced according to China National Standards GB/T4171-2000, the chemical composition range and manufacturingprocess of which are same with that of Cor–Ten A. In addition, thissteel was trial-manufactured in industrial plant, and 500 MPagrade steels with thickness of 80 mm was expected to obtain. Cor-rosion resistance of the industry trial steel was also measured bywet/dry accelerated corrosion test, because it is difficult to obtainthe same results completely as that of the other tests even undersame test conditions.

2. Materials and experimental procedures

Steels A1 and A2, respectively, with 0.03% and 0.05% carbon contents were de-signed, the corrosion resistance of which was compared with that of WS 09CuPCrNi.Chemical compositions of these steels are listed in Table 1. Steels A1 and A2 weremelted in 25 kg vacuum induction furnace and cast in a metal mould. The forgedingots were soaked at 1250 �C for 2 h, and then hot-rolled to 6 mm-thick plateswith a finish rolling temperature 880 �C before water quenched. Subsequently, thissteel was trial-manufactured in industrial plant, chemical composition of which is

Table 1Chemical composition of various carbon contents steels and WS (wt.%)

Heat C Si Mn S P Nb Ti Cu Ni Cr

WS 0.082 0.28 0.38 0.006 0.018 0.035 0.008 0.29 0.24 0.33A1 0.03 0.30 1.77 0.005 0.01 0.04 0.021 Cu, Ni, Cr, Mo(total) 6 1.2%A2 0.05 0.29 1.77 0.005 0.01 0.05 0.019 Cu, Ni, Cr, Mo(total) 6 1.2%B 0.045 0.30 1.40 0.005 0.008 0.04 0.021 Cu, Ni, Cr, Mo(total) 6 1.2%

Fig. 1. Typical optical micrographs of steels A1 and A2: (a) 0.03C% with quasi-polygonal ferrite, granular bainite and bainite ferrite; (b) 0.05C% with granular2bainite, bainitic ferrite and a little bit of acicular ferrite and (c) WS with ferrite andpearlite.

130 J. Guo et al. / Materials and Design 30 (2009) 129–134

also shown in Table 1. The bridge steel was processed by TMCP with acceleratedcooling rate more than 10 �C/s, and finish cooling temperature 550 �C, for obtainingoptimum mechanical properties. The thickness of plates reached 80 mm.

Pieces of 60 � 40 � 5 mm were cut from steels A1, A2, WS and B for corrosiontest, it should be noted that the pieces cut from steel B near the surface layer. Allpieces were mechanically ground with 400–1000 grit silicon carbide abrasive paper,and cleaned with acetone, then rinsed with distilled water. A1, A2 and B were com-pared with WS in the cyclic wet–dry corrosion test. The test included two parts: (1)immersing samples into 3.5 wt.% NaCl solution at 40 �C for 18 min and (2) dryingthe specimen at 42 �C, 50% RH for 62 min. The accelerated corrosion test lastedfor 360 cycles, and the specimens were taken out in three batches after 36, 72,108, 144 and 180 cycles, respectively, for weight loss measurement. The rust layersformed on specimens were removed by immersing the specimens into Clark’s solu-tion (100 ml HCl, 1.19 g/ml + 20 g Sb2O3 + 50 g SnCl2) and stirring for around 10 minat 25 �C.

Samples for metallographic observations were prepared by conventional grind-ing and polishing techniques, and then etched with 3% nital solution. The cross-sec-tion of the rust layers was analyzed using scanning electron microscopy (SEM).

The impedance measurements were carried out in 0.5 wt.% NaCl aqueous solu-tion to evaluate the protection performance of rust formed on the steels. Electro-chemical impedance spectra (EIS) measurements were performed using aSolartron 1255B frequency response analyzer in combination with a PAR 1287Apotentiostat at room temperature. The sinusoidal potential for measurementswas close to open circuit potential, amplitude 10 mv, and frequency ranged from106 to 10�2 Hz.

3. Results and discussion

3.1. Effects of low carbon content on the weathering resistance

The microstructures of steels A1, A2 and WS are exhibited inFig. 1. The predominant microstructures of A1 and A2 are granular

Table 2Mechanical properties of steels A1 and A2

Sample no. Rm (MPa) Rel (MPa) A (%) Ak (half size)/J

20 �C �20 �C �40 �C

A1 843 712 14.5 99.3 102.7 94.7A2 955 770 11.3 76.3 74 78.3

-20 0 20 40 60 80 100 120 140 160 180 200

0.00

0.04

0.08

0.12

0.16

0.20

corr

osio

n de

pth

(mm

)

corrosion cycles

0.03C-DQ 0.05C-DQ WS

Fig. 2. Corrosion depth of LCS and WS vs. corrosion cycles.

J. Guo et al. / Materials and Design 30 (2009) 129–134 131

bainite (GB), bainitic ferrite (BF) and a small amount of acicular fer-rite (AF), as shown in Fig. 1a and b with arrows. The microstruc-tures of WS are polygonal ferrite (PF) and pearlite (P) structures,as shown in Fig. 1c with arrows.

Table 2 lists the mechanical properties of steels A1 and A2. Fromthe table, it can be seen that the yield strength increases from 712 to770 MPa with increasing carbon content from 0.03% to 0.05%. Ten-sile strength of the two steels, as well as yield strength, increases

Fig. 3. SEM images of rust layers formed on steels: (a) steel A1; (b) steel A2 and (c)WS.

from 843 to 955 MPa. Elongation of two steels, on the contrary,decreases from 14.5% to 11.3% with increasing carbon content.

The good balance of strength, ductility and toughness is as-cribed to the homogeneous microstructures of bainite. Further-more, the microstructures are not influenced significantly bycooling rate in this carbon content range. Therefore, to maintainconsistent cross-section strength, low carbon content should beconsidered for heave plates. For 0.03–0.05 wt.% C bridge weather-ing steels, the yield strength of 500–700 MPa grade can be ob-tained by accelerated cooling process.

The relations between corrosion depth and test time of steelsA1, A2, and WS are illustrated in Fig. 2. Corrosion rates of the threesteels are rapid in the first stage of test, and then reduce with pro-

0 50 100 150 200 250 300 350 400

0

-50

-100

-150

-200

Zre(ohm-cm2)

Zim

(ohm

-cm

2 )

Experimental ipedance spectra steel A1 steel A2 WS

Fig. 4. Nyquist plots of steels A1, A2 and WS in 0.1 M NaCl solution after 180 cyclesaccelerated corrosion.

Fig. 5. Equivalent circuit for rusted steel A1 and A2 in 0.1 M NaCl solution.

0 50 100 150 200 250 300 350 400 450 500 550

0

-50

-100

-150

-200

-250

-300

Zre(ohm-cm2)

Zim

(ohm

-cm

2 )

Simulated impedance spectra steel A1 steel A2 WS

Fig. 6. Simulated spectra of EI after 180 cycles of cyclic wet–dry acceleratedcorrosion.

Table 3Mechanical properties of steel B

Gauge(mm)

Position Rel(MPa)

Rm(MPa)

A(%)

Rel/Rm AKV(�40 �C)/J

80 1/4t 535 610 27.0 0.88 260 294 2841/2t 500 615 26.5 0.81 115 280 174

Fig. 7. Cross section microstructures of 60 mm thickness steel plate: (a) 1/2t;(b) 1/4t and (c) near surface.

132 J. Guo et al. / Materials and Design 30 (2009) 129–134

longing corrosion test. In general, final corrosion depth of the twosteels is less than that of WS. Corrosion rate of these steels in accel-erated tests decreases slightly, which indicates that, under thesesevere cyclic wet/dry test conditions, enough dry time are neces-sary to form dense rust layers. Moreover, weight loss of the lowcarbon steels is lower than that of WS because of their uniform bai-nite microstructure, which attributed to their low carbon contentsand rapid cooling rate [2].

Fig. 3a–c shows the SEM micrograph of the cross-sections ofsamples after 180 cycles of accelerated corrosion tests. The figuresdisplay that rust layers of the low carbon microalloyed steels aredenser and thinner than that of WS. In the case of WS, there aremany large voids and microcracks in the rust layers facilitatingpenetration of chloride solution to the substrate and promotingcorrosion process. These voids are probably due to large amountsof pearlite appeared in hot rolled WS steel [11]. Thus, the thicknessof rust layers on low carbon steels and WS after 180 cycles are100–200 and 200–400 lm, respectively.

The Nyquist plots of the test steels after 180 cycles of cyclicdry/wet condition are shown in Fig. 4. It presents one depressedsemicircle with a long tail at low frequency region, indicating thatthe electrochemical reaction is controlled by diffusion process.The semicircle diameter of low carbon steels are larger than thatof WS, which means that protection ability of rust layers on thetwo steels are better than that on WS. As illustrated in Fig. 5,an equivalent circuit model was used to describe the rustlayers/substrate interface. In this figure, Rs represents the electro-lyte resistance, crust the rust capacitance, Rr the rust resistance,Cd the double-layer capacitance, Rt the charge-transfer resistanceand W the Warburg resistance. The Faraday impedance is dividedinto two parts: charge transfer resistance Rt and Warburgimpedance W. The latter represents influence of concentrationpolarization and element diffusion on the electrode reaction,which exists only at low frequencies. In equivalent circuit, theimpedance at high and intermedia frequencies can be expressedas follows:

Z ¼ Rs þ1

jxCrust þ 1Rrþ 1

jxCd1þ1Rt

ð1Þ

Eq. (1) can be simplified in the following way:

Z ¼ Rs þa

a2 þ ðxCrust � bÞ2

" #� xCrust � b

a2 þ ðxCrust � bÞ2

" #j ð2Þ

-20 200 40 60 80 100 120 140 160 180 200-0.02

0.00

0.02

0.04

0.06

0.08

0.10

0.12

0.14

0.16

0.18

corr

osio

n de

pth

(mm

)

corrosion cycles

WS steel B

Fig. 8. Corrosion depth of steels B and WS vs. corrosion cycles.

Fig. 9. SEM images of rust layers formed on steel B.

J. Guo et al. / Materials and Design 30 (2009) 129–134 133

where a and b are represented as follows:

a ¼ Rr þRt

1þ ðxCd1RtÞ2ð3Þ

b ¼ xCd1R2t

1þ ðxCd1RtÞ2ð4Þ

Make the real part of the equation be X, and the imaginary partbe Y. When x is decreased to zero, the electrochemical impedancespectra will have tendency to intersect with the Zreal axis(Zreal = Rs + Rr + Rt). When x is increased to infinity, the electro-chemical impedance spectra will also have tendency to intersectthe Zreal axis (Zreal = Rs). Simulations of the equivalent circuit modelare in good agreement with experimental data, as shown in Fig. 6.

3.2. Industrial trial of 500 MPa grade high performance bridge steel

According to the microstructure evolution characteristics ofsteel B during continuous cooling and isothermal treatment, thecontrolled cooling processes were applied. Accelerated coolingrate, start and finish cooling temperature were chosen to obtaina good combination of strength, elongation and toughness.Mechanical properties of the steel are illustrated in Table 3, itcan be seen that yield strength is higher than 500 MPa, and elonga-tion and toughness are adequate. Because of low carbon content,the weldability should be excellent [12].

Microstructure of the as-rolled steel B at 1/2 and 1/4 thicknesspositions and surface position are shown in Fig. 7. Typical micro-structures are quasi-polygonal ferrite, acicular ferrite and granularbainite at 1/2, 1/4 and surface positions along thickness direction.A small amount of large M/A island and/or degenerated pearlitedistribute intra and/or inter the ferrite at 1/2 and 1/4 thicknesspositions. However, the surface layer of plate is mainly homoge-neous bainite, in which the amount of M/A is less and the size isquite small. The results are ascribed to rapid cooling rate in surfacelayer of the steel.

The relation between corrosion depth and test cycles of steel Band WS are exhibited in Fig. 8. Corrosion rates of the two steels ininitial stage are rapid, and then reduce with prolongs the test,especially after about 140 cycles. Final corrosion depth of steel Bis less than that of WS.

Fig. 9 diagrammatizes the cross-section of steel B after 180 cy-cles observed by SEM. Thickness of the rust layers are 100–200 lm,which is thinner and denser significantly than that on WS(�350 lm). These results reveal that the corrosion resistance ofsteel B is better than that of WS in cyclic wet/dry test.

4. Conclusions

(1) Corrosion resistance of steel with carbon content range in0.03–0.05% is better than that of WS as observed by a labo-ratory-accelerated test that involving cyclic wet/dry condi-tions in a chloride environment. SEM photographs revealthe rust layers of these steels (A1, A2) are thinner and denserthan that of WS, and corresponding EIS results show thatrust layers resistance of these steels (A1, A2) is larger thanthat of WS.

(2) The steel was trial-manufactured in industrial plant, and500 MPa grade weathering bridge steel was obtained. Micro-structures of the large gauge plates are mainly quasi-polyg-onal ferrite, acicular ferrite, granular bainite and a smallamount of pearlite (degenerate pearlite) at 1/2 and 1/4 posi-tion. However, microstructure of the surface layer is homo-geneous bainite with less and smaller second phases,which is similar to that of steels A1 and A2. Therefore, corro-sion resistance of the industrial trail steel is as good as thatof steels A1 and A2.

Acknowledgements

This research was supported by National Nature Science Foun-dation of China (Nos. 50571016 and 50571089), National Key BasicResearch and Development Programme of China (No.2004CB619102), National High Technology Research and Develop-ment Programme of China (No. 2006AA03Z507).

References

[1] Mathay WL. Highway structures design handbook. American Institute of SteelConstruction Inc.; 1993.

134 J. Guo et al. / Materials and Design 30 (2009) 129–134

[2] Shang CJ, Zhao YT, Wang XM, Yang SW, He XL. Formation and control of theacicular ferrite in low carbon microalloying steel. In: Proceedings of the 3rdinternational conference on advanced structural steels, Gyeongju, Korea; 2006.p. 475–9.

[3] Chen XH, Dong JH, Han EH, Ke W. Effect of Ni on the ion-selectivity of rust layeron low alloy steel. Mater Lett 2007;61:4050–3.

[4] Tahara A, Shinohara T. Influence of the alloy element on corrosion morphologyof the low alloy steels exposed to the atmospheric environments. Corros Sci2005;47:2589–98.

[5] Ishikawa T, Kumagai M, Yasukawa A, Kandori K, Nakayama T, Yuse F.Influences of metal ions on the formation of c-FeOOH and magnetite rusts.Corros Sci 2002;44:1073–86.

[6] Hara S, Kamimura T, Miyuki H, Yamashita M. Taxonomy for protective abilityof rust layer using its composition formed on weathering steel bridge. CorrosSci 2007;49(3):1131–42.

[7] Yamashita M, Miyuki H, Matsuda Y, Nagano H, Misawa T. The long termgrowth of the protective rust layer formed on weathering steel by atmosphericcorrosion during a quarter of a century. Corros Sci 1994;36:283–99.

[8] Kamimura T, Hara S, Miyuki H, Yamashita M, Uchida H. Composition andprotective ability of rust layer formed on weathering steel exposed to variousenvironments. Corros Sci 2006;48:2799–812.

[9] Zhang CL, Cai DY, Liao B, Zhao TC, Fan YC. A study on the dual-phase treatmentof weathering steel 09CuPCrNi. Mater Lett 2004;58:1524–9.

[10] Chen YY. Corrosion resistance and mechanical properties of low-alloy steelsunder atmospheric conditions. Corros Sci 2005;47:1001–23.

[11] Zhao YT, Yang SW, Shang CJ, Wang XM, Liu W, He XL. The mechanicalproperties and corrosion behaviors of ultra-low carbon microalloying steel.Mater Sci Eng A 2007;454:695–700.

[12] Kimihiro N, Kazuyuki M, Naoyoshi T. High performance steel plates for bridgeconstruction-high strength steel plates with excellent weldability realizingadvanced design for rationalized fabrication of bridge. JFE Technical Report,vol. 5; 2005. p. 30–6.