Journal of Chemical Technology and Metallurgy, 51, 3, 2016

328

Journal of Chemical Technology and Metallurgy, 51, 3, 2016, 328-334

SLOW COOLING OF MEDIUM CARBON MOLYBDENUM STEEL BARS TO ELIMINATE HYDROGEN INDUCED CRACKS

Ravichandar Dhakshanamoorthy1, Nagashanmugam Krishnan Bommannan2, Balusamy Thangavel3

1 GCE, Salem & CEO, JSW Steel Ltd, Salem Works, Tamil Nadu, India 2 R&D Center, JSW Steel Ltd, Salem works, Salem 636 453, Tamil Nadu, India3 Department of Mechanical Engineering, Govt. College of Engineering, Salem, Anna University, Chennai, India E-mail: d.ravichandar@jsw.in naga.shanmugam@jsw.in balusamyt@rediffmail.com

ABSTRACT

Hydrogen induced cracks were among the defects observed by ultrasonic testing of medium carbon molybdenum steel at JSW Steel Limited, Salem Works (JSWSL). To reduce hydrogen concentration and hence hydrogen cracks in steel, a slow cooling technique was employed as hydrogen tends to escape during cooling due to its smaller atomic radius and high diffusivity in a solid phase. Cooling experiments were conducted wherein a few hot rolled bars of JDM1046 grade steel were slow-cooled while other bars were air-cooled. Hardness, microscopic and ultrasonic examinations were carried out on these bars after cooling. It was found that the bars slow cooling resulted in defect free steel against 15 % - 35 % defect levels observed in air-cooled bars. The hardness measurements indicated that the slow-cooled bars exhibited reduced hard-ness when compared to that of the air-cooled one. The paper discusses the results of experiments conducted to facilitate the elimination of hydrogen in JDM 1046 grade steel, which finds application in making crank shafts for automobiles.

Keywords: JDM 1046, ultrasonic testing, hydrogen assisted cracks, slow cooling, hardness, micro-structure.

Received 21 December 2015Accepted 18 March 2016

INTRODUCTION

When liquid steels cool from a temperature value above that of austenitization, it transforms into other phase configurations according to the austenite com-position and the cooling rate. Hence, the steel crystal structure and consequently, both the shape and the lattice parameter of the unit cell, change. These changes may introduce dilatational strains into the microstructure, which result in the creation of residual stress concentra-tion zones within the microstructure. They are vulnerable to the formation of micro cracks or flaws growth [1]. Three processes are involved pertaining to hydrogen: (a) hydrogen evolution from the molten steel and segrega-

tion of hydrogen in blooms during solidification, (b) homogenization and redistribution of hydrogen in steel during solidification and (c) hydrogen diffusion from blooms during cooling. Hydrogen diffusion is considered of paramount importance in respect to obtaining high quality blooms of low hydrogen content [2]. Diffusible hydrogen is considered to be mobile at or near room temperature, whereas the remaining residual hydrogen is trapped in the metal. Residual hydrogen can be retained through an interaction with micro structural discontinui-ties or by formation of hydrides with alloying elements [3]. The presence of hydrogen and other factors lead to the delayed formation of internal defects that appear as hairline cracks, shatter cracks or flakes. The latter

Ravichandar Dhakshanamoorthy, Nagashanmugam Krishnan Bommannan, Balusamy Thangavel

329

develop only after an incubation period at temperatures below 200oC in a section ranging from approximately the mid-radius to the center. Flaking is more difficult to avoid in ultra clean steels. There hydrogen flaking is observed even at a hydrogen content less than 2 ppm. It is worth adding that steels containing 1.5 ppm of hydrogen cannot be considered immune in respect to hydrogen flaking [4].

Hydrogen accumulates on the surface of manganese sulphide inclusions and forms molecular hydrogen (H2) and develops sufficient pressure to create internal cracks. However, if the product is slow cooled from the rolling temperature, the atomic hydrogen has sufficient time to diffuse from the product, thus providing no hydrogen damage [5]. At JSWSL, hydrogen induced cracks and center unsoundness contributed to a great extent to ul-trasonic rejections of chrome-moly and high manganese steels. This effect was somewhat minimized by increase of the argon flow rate during vacuum degassing and optimization of the superheats [6]. The present paper discusses the study conducted to eliminate the hydrogen induced cracks by slow cooling of JDM 1046, a medium carbon molybdenum grade steel.

EXPERIMENTAL

The chemical composition of steel was determined by ARL-4460 optical emission spectrometer, while the hydrogen content was evaluated by LECO RH-402 hy-drogen analyser. The steel microstructure was studied by Leica optical microscope, model DMI5000M and a scanning electron microscope, model EV018, Carl-Zeiss, UK. The ultrasonic testing was carried out as per ASTM A388. The hardness was measured by using Brinell Hardness testing machine, model TKB 3000. The bars temperature was recorded by an optical pyrometer, model Raynger 3I-Raytek of ±1 % uncertainty.

The chemical composition of JDM 1046 is presented in Table 1.

The route followed for steel production at JSWSL was as follows: Blast furnace – Energy Optimising Furnace – Ladle Refining Furnace – Vacuum Degas-sing – Continuous Casting. 340 mm x 400 mm blooms of JDM 1046 cast by a continuous caster were further rolled into bars of 125 mm diameter. The initial tem-perature at the exit pass was within the 920oC - 900oC range. A few hot rolled bars were slow-cooled in the box while others were air-cooled as per normal practice. The micro-structure, the hardness and the ultrasonic proper-ties of these bars were studied after the cooling aiming to outline the effect of slow cooling effect.

The bars slow cooling was carried out in mild steel box and its schematic presentation is given in Fig. 1.

The method involved subjection of a few hot rolled bars to slow cooling in a mild steel box. Aiming a com-parison the rest of the same batch was cooled at air. The mild steel box, used for the study, was of a capacity of 20 tons and could house 6 meter length hot rolled bars. Its inner surface was lined by 40 mm thick glass wool to insulate it thermally from atmosphere during the cooling process. After rolling the bars were placed inside the box. It was immediately closed with a top door lined by 40 mm thick glass wool. The bars remained in the box

Grade

C (%)

Si (%)

Mn (%)

P(%)

S (%)

Al (%)

Cr (%)

Mo (%)

H

(ppm)

JDM1046 0.49 0.22 1.03 0.016 0.004 0.026 0.19 0.10 1.89

Table 1. Chemical composition of JDM 1046 (in mass %).

Fig. 1. Schematic diagram of mild steel box.

Journal of Chemical Technology and Metallurgy, 51, 3, 2016

330

until it cooled down to a room temperature. The box was provided with a sliding door to measure the bars temperature. This was done at various intervals start-ing right from the exit pass. It continued until the bars temperature decreased to the ambient value.

RESULTS AND DISCUSSION

Cooling data is plotted as temperature versus time. Fig. 2 presents the temperature profiles of the air-cooled and slow-cooled bars.

The cooling experiments carried out reveal that 640 min are required by slow-cooled bars, against only 168 min in case of the air-cooled one, to cool to 200oC. This indicates approximately a fourfold increase in time for

the slow-cooled bars.After cooling all bars investigated are subjected

to hardness, microscopic and ultrasonic examinations and the results obtained are juxtaposed to determine the effect of slow cooling on these properties. Table 2 illustrates the comparison of the hardness of three air-cooled and three slow-cooled bars.

It is found that the slow-cooled bars exhibit less hardness at all locations measured when compared to that of the air-cooled bars. This can be attributed to the richer and coarser ferrite microstructure of the steel in this case. Furthermore, the reduced hardness minimizes crack formation due to material’s increased ductility and facilitates formability during forging. This in turn is favorable for JDM 1046 as it is supplied to forging industries for making crankshafts.

Few slow-cooled and air-cooled bars were subjected to hydrogen measurements by LECO Hydrogen analyser and the results are presented in Table 3.

The values listed in the Table 3 refer to total hydro-gen, which includes both trapped and diffusible hydro-gen. It is seen that the hydrogen concentration is much higher in the air-cooled bars than that in slow-cooled one at all locations measured. This indicates that most of the hydrogen diffuses out during slow cooling. Microscopic examination is carried out after etching these bars with 2 % nital at various locations such as the sub-surface (5 mm from surface), the mid-radius (30mm from surface) and the core. Micrographs referring to the sub-surface and mid-radius are presented in Figs. 3 and 4.

Fig. 2. Temperature profiles of slow cooled and air cooled bars.

Table 2. Comparison of air cooled and slow cooled bars hardness.

Type of cooling Grade Bars * Hardness (BHN)

At center At mid radius At sub surface

Air cooling

JDM 1046

A1 223 223 229

A2 212 217 207

A3 212 212 197

Slow cooling

S1 197 197 187

S2 207 207 195

S3 202 204 194

*A & S signifies air cooling and slow cooling

Ravichandar Dhakshanamoorthy, Nagashanmugam Krishnan Bommannan, Balusamy Thangavel

331

All the micrographs shown are recorded at 200 X magnification for easy comparison. It is seen that, those of the slow-cooled bars are richer in ferrite and the grains are coarser and uniformly distributed when compared to those of the air-cooled bars.

Only a small amount of the hydrogen will diffuse out of the steel bars during the process of air cooling, which is in fact a fast process. This results in faster

cooling of the surface than that of the core and hence, in a higher hydrogen concentration in the bar interior as hydrogen diffuses towards hotter zone of the product [7]. According to literature data [8] hydrogen is col-lected in micro-pores, voids and shrinkage cavities and causes pinholes and porosity in steel. Furthermore, as its solubility decreases with temperature decrease, local hydrogen pressure builds up in the steel matrix, which

Table 3. Juxtaposition of hydrogen concentration values obtained for air-cooled and slow-cooled bars.

Type of cooling Grade Bars *

Hydrogen (ppm)

At center At mid radius At sub surface

Air cooling

JDM 1046

A1 1.92 1.79 1.36 A2 1.87 1.84 1.76

Slow cooling S1 0.92 0.89 0.75 S2 0.98 0.94 0.84

*A & S signifies air cooling and slow cooling

Fig. 3. Micrographs of JDM1046 sub surface: (a) air-cooled bars; (b) slow-cooled bars.

Fig. 4. Micrographs of JDM1046 mid radius: (a) air-cooled bars; (b) slow-cooled bars.

a) a)

b) b)

Journal of Chemical Technology and Metallurgy, 51, 3, 2016

332

can subsequently lead to cracks formation in the course of rolling. The ultrasonic inspection carried out shows that the air-cooled bars have core defects, which in turn is in agreement with the consideration presented above. Fig. 5 illustrates the presence and absence of a core defect in an air-cooled and a slow-cooled bar, correspondingly.

The defect observed in air-cooled bars can be attrib-uted to the steel chemical composition [9]. It is known that sulphur forms manganese sulphide during cooling,

which traps hydrogen at low temperatures (< 200oC). As the sulphur content of JDM 1046 is very low, only 0.004 %, the sulphide inclusions will be few, i.e. there is an insufficient number of sulphide hydrogen traps to pro-vide an adequate hydrogen distribution. This increases the local hydrogen pressure and hence, facilitates crack-ing appearance in air-cooled bars. Therefore hydrogen flaking may occur at lower hydrogen contents in case low sulphur steels when compared to those observed in high sulphur steels. Hence, less than 1 ppm hydrogen

Fig. 5. Micrographs of (a) a core hydrogen crack in an air-cooled bar; (b) a defect free slow-cooled bar.

Fig. 6b. SEM image of a hydrogen crack and an EDS spectrum recorded at the middle of the crack.

Fig. 6a. SEM image of a hydrogen crack and an EDS spectrum recorded at the end of the crack.

Ravichandar Dhakshanamoorthy, Nagashanmugam Krishnan Bommannan, Balusamy Thangavel

333

low sulphur steels [4]. Thus, JDM 1046 porosity and low sulphur content contributes to hydrogen induced cracking which is observed as a defect in the course of air-cooled bars ultrasonic testing. The experimental results obtained indicate steel desulphurisation has to proceed only to the level pointed above.

But, the scenario in case of slow cooling is different. Most of the diffusible hydrogen escapes out of the steel matrix and does not concentrate in the inner zones to the extent reached by air cooling and hence no cracks

1959. No defects are found in any of the slow-cooled bars in agreement with the consideration presented. Fig. 5b illustrates the slow-cooled core free from any defects.

The defective bars were examined by scanning

observed in air-cooled bars refer only hydrogen induced cracks but not to inclusions. SEM-EDS micro analysis (spot analysis) is also carried out at various locations to determine the samples chemical composition.

SEM-EDS analysis does not reveal the presence of any oxides at these locations and thus rules out the possibility of crack appearance due to inclusions. The spot analysis also rules out the possibility of crack ap-pearance due to sulphide and H2S inclusions. These observations indicate that the crack is determined by hydrogen presence only. The crack shape, which is typical for a hydrogen crack, can be considered as an

Ultrasonic testing of the air-cooled and slow-cooled bars is performed and the data obtained is listed in Table 4. It is evident that 15 % - 35 % defect level is observed at a depth of 60 mm - 70 mm in air-cooled bars, whereas the slow-cooled bars are free from defects.

The hardness of a defective bar is tested. It is found it has hardness of 229 BHN, 241 BHN and 235 BHN at the sub surface, the mid radius and at the center, respec-tively. These values are much higher when compared to those of any of the slow-cooled bars.

CONCLUSIONS

maximum removal of diffusible hydrogen due to in-crease of the time available for diffusion and that of the ferrite phase content in the steel microstructure. The hydrogen measurements reveal that the hydrogen con-centration measured at all locations of the slow-cooled bars is much lower than that of the air-cooled one, which is evidence that most of the hydrogen diffuses out in this case. Slow cooling results in defect free steel against 15 % - 35 % defect level of the air-cooled bars. The results obtained suggest that steel desulphurization has to be limited as the low sulphur content contributes also to the appearance of hydrogen cracks which in turn are responsible for the defects observed in the course of the ultrasonic testing carried out. Slow-cooled bars show hardness less of up to 40 BHN when compared to that of air-cooled one, which is advantageous con-sidering subsequent forging operation. The successful slow cooling experiments imposed this method on bulk manufacture of steel.

Acknowledgements

The authors are thankful to the Management of M/s. JSW Steel Limited, Salem Works, for granting permis -sion and for providing facilities to carry out the research work in JSWSL, R&D center & Blooming mill.

Table 4. Ultrasonic testing data referring to air-cooled and slow-cooled steel bars investigated.

Type of cooling

Bars

Defect level (%)

Defect at depth from

surface (mm)

Air cooling A1 15-35% 60-68

A2 15-30% 60-70

A3 20-35% 60-65

Slow cooling All bars Nil Nil

Journal of Chemical Technology and Metallurgy, 51, 3, 2016

334

REFERENCES

1. Mohammad Reza Allazade, Effect of cooling rate on the microstructure configuration of continuously cast steel slabs, Ph.D. Thesis, University of Pittsburgh, Pittsburgh, 2009.

2. XIAN Aiping, LI Peiji, CHEN Wenxiu, WANG Yikang, Effect of removing hydrogen from heavy rail steel blooms by stack cooling in Panzhihua iron and steel company, Acta Metal.l Sinica, Series A, 6, 6, 1993, 415-419.

3. C.G. Interrante, R. D. Stout, Delayed cracking in steel weldments, Welding J., 43, 4, 1964, 145-160.

4. R.J. Fruchan, A review of hydrogen flaking and its prevention, Proceedings of 13th International Forge

masters Meeting, Pusan, Korea, 1997, 41-42.5. Bruce L. Bramfitt, Carbon and Alloy Steels, Reprinted

from Handbook of Materials Selection, Wiley Publication, New York, 2002, 24.

6. D. Ravichandar, T. Balusamy, K.B. Nagashanmugam, Reducing UT rejections in Cr-Mo and high Mn steels by controlling hydrogen and optimizing superheat, J. Appl. Mech. Mat., 591, 2014, 38-42.

7. P.T. Basedin, The causes of flake formation, Metal lov, 2, 1959, 14-18.

8. Hydrogen in steels, Total materia article, key to met-als.com, 2007.

9. P.V. Sklyuev, Flakes in Steel, Translated from Metallovedenie i Termicheskaya Obrabotka Metallov, 9, 1972, 57-60.