Materials & Design

Materials and Design 25 (2004) 219230 www.elsevier.com/locate/matdes

Improvement in strength and toughness of austempered ductile cast iron by a novel two-step austempering process
Jianghuai Yang, Susil K. Putatunda
Received 7 May 2003; accepted 12 September 2003

Department of Chemical Engineering and Materials Science, Wayne State University, 5050 Anthony Wayne Drive, Detroit, MI 48202, USA

Abstract Austempered ductile cast iron (ADI) has emerged as an important engineering material in recent years because of its excellent mechanical properties. These include high strength with good ductility, good wear resistance, fatigue strength and fracture toughness. It is therefore considered as an economical substitute for wrought or forged steel in several structural applications especially in the automotive industry. In this investigation, a nodular or ductile cast iron with predominantly pearlitic as-cast structure was processed by a novel two-step austempering process. Two batches of samples were prepared. All the specimens were initially austenitized at 927 C (1700 F) for 2 h. The rst batch of samples were processed by conventional single-step austempering process at several temperatures such as 260 C (500 F), 273 C (525 F), 288 C (550 F), 302 C (575 F), 316 C (600 F), 330 C (625 F), 343 C (650 F), 357 C (675 F), 371 C (700 F), 385 C (725 F) and 400 C (750 F) for 2 h, whereas the second batch of samples were processed by the two-step austempering process. These samples were initially quenched for 5 min in a salt bath maintained at 260 C (500 F) and then austempered for 2 h at several austempering temperatures. These temperatures were 288 C (550 F), 302 C (575 F), 316 C (600 F), 330 C (625 F), 343 C (650 F), 357 C (675 F), 371 C (700 F), 385 C (725 F) and 400 C (750 F). Inuence of this two-step austempering process on microstructure and mechanical properties of ADI was examined. Test results show that this two-step austempering process has resulted in signicant improvement in yield and tensile strengths and fracture toughness of the material over the conventional single-step austempering process. 2003 Elsevier Ltd. All rights reserved.
Keywords: Austempered ductile cast iron (ADI); Two-step austempering process; Microstructure; Strength; Toughness; Fracture toughness

1. Introduction Austempered ductile cast iron (ADI) is considered to be an important engineering material because of its attractive properties such as good ductility at high strengths [14], good wear resistance [5,6], fatigue strength [79] and fracture toughness [1014]. Because of these combinations of properties, ADI is now used extensively in many structural applications in automotive industry, defense and earth moving machineries, etc. The attractive properties of ADI are related to its unique microstructure that consists of ferrite (a and high carbon austenite (cHC ). This is dierent from the austempered steels where the microstructure consists of
Corresponding author. Tel.: +1-313-577-3808; fax: +1-313-5773810. E-mail address: sputa@chem1.eng.wayne.edu (S.K. Putatunda). 0261-3069/$ - see front matter 2003 Elsevier Ltd. All rights reserved. doi:10.1016/j.matdes.2003.09.021
*

ferrite and carbide. Because of this dierence, the product of austempering reaction in ductile iron is often referred to as ausferrite rather than bainite. This terminology will be used in the present paper. Large amount of silicon present in ductile iron suppresses the precipitation of carbide during austempering reaction and retains substantial amount of stable high carbon austenite (cHC ). Small amounts of alloying elements such as nickel, molybdenum and copper are generally added to ADI so that it has sucient hardenability to be quenched to the austempering temperature without forming pearlite. Austempered ductile cast iron is an alloyed and heattreated nodular cast iron. The starting material for the development of ADI is the high quality ductile or nodular cast iron. It is then subjected to an isothermal heat treatment process known as austempering. Conventional austempering process (referred to as single-step

220

J. Yang, S.K. Putatunda / Materials and Design 25 (2004) 219230

austempering here) consists of austenitizing the casting in the temperature range of 871982 C (16001800 F) for sucient time to get a full austenite (c) matrix, and then quenching it to an intermediate temperature range of 260400 C (500750 F). The casting is maintained at this temperature for 24 h. A schematic of the conventional single-step austempering process is shown in Fig. 1. During austempering, a two-stage phase transformation reaction takes place in ADI. In the rst reaction, austenite (c decomposes into ferrite (a and high carbon or transformed austenite (cHC ): c ! a cHC : 1

If the iron is held at the austempering temperature for too long, a second reaction occurs, which causes further decomposition of the high carbon austenite into ferrite and e-carbide: cHC ! a e 2

The microstructure in this case will contain e carbide which makes the material brittle [15]. Since the e carbide is a detrimental phase constituent this reaction must be avoided during austempering process. It has been well established [16,17] that the best combination of mechanical properties (tensile strength and ductility) is obtained in ADI after the completion of the rst reaction but before the onset of the second reaction. The time period between the completion of the rst reaction and the onset of the second reaction is termed as process window. One important function of the alloying elements added to ADI is to enlarge this process window. It is now well established [7,16,17] that the important microstructural features inuencing the mechanical properties of ADI are:

1. transformed austenite content (volume fraction of austenite, Xc ); 2. carbon content of the austenite (Cc ) as well as 3. morphology of ferrite (acicular a and austenite. When ADI is austempered at lower temperatures, e.g., 260 C (500 F), it has ner ferrite and austenite and this results in higher yield and tensile strength but with lower ductility. On the other hand, when ADI is austempered at higher temperatures, e.g., 371 C (700 F), it has coarser or feathery ferrite and austenite and this reduces the yield and tensile strengths but imparts higher ductility. The fracture toughness and fatigue strength of the ADI have been found to be dependent on the parameter Xc Cc [18]. This parameter (Xc Cc ) increases with increase in austempering temperature. Thus, in summary, mechanical properties of ADI have been found to be dependent on the neness of ferrite and austenite and the austenitic carbon [7,1618] [(Xc Cc ), where Xc is the volume fraction of austenite and Cc is the carbon content]. Therefore, it will be possible to optimize the mechanical properties of ADI if one can produce very ne ferrite and austenite as well as very high austenitic carbon. It has also been observed that the transformation of austenite to ferrite during the rst reaction of ADI occurs by the nucleation and growth process [710]. Nucleation depends on the supercooling and it is possible to obtain higher carbon in austenite (Cc ) if the austempering temperature is higher. Thus, it was theorized here that by combining large supercooling and higher austempering temperature, it may be possible to increase the parameter (Xc Cc ) and at the same time we can have very ne ferrite and austenite in the matrix. This is expected to enhance the mechanical properties of ADI. Hence, the novel concept of two-step austempering was conceived by these investigators. This two-step

1200

1. A-B: heat up to the

B C

austenitizing temperature; 2. B-C: hold at the

Austempering temperature (C)

austenitizing temperature
800

(usually 2 hrs); 3. C-D: quench to austempering temperature; 4. D-E: hold at the austempering temperature

400

(usually between 2- 4 hrs); 5. E-F: air cool to room temperature.

20

A 1 2 3 4 Austempering time (Hour)

F 5

Fig. 1. Schematic of conventional (single step) austempering process.

J. Yang, S.K. Putatunda / Materials and Design 25 (2004) 219230

1.

221

A-B: heat up to the austenitizing temperature; B-C: hold at the austenitizng temperature (2 hours); C-D: quench to the first austempering temperature; D-E: hold at the first austempering temperature (for a few minutes until nucleation is completed);

1200 2. B Austempering temperature (C) C 3. 800 4.

F 400 D E

5.

E-F: raise temperature immediately to second austempering;

6.
HH

F-G: hold at the second austempering temperature (usually 2 hous);

A 20 1 2 3 4 Austempering time (Hour)

H 7. 5

G-H: air cool to room temperature.

Fig. 2. Schematic of the conceived two-step austempering process.

process involves rst quenching the alloy to a lower temperature after austenitizing and thus increasing the supercooling, and then immediately (once the nucleation is complete), raising the temperature of the quenching media to a higher temperature to facilitate faster diusion of carbon and thereby increasing the carbon content of the austenite and hence increasing the austenitic carbon (Xc Cc ) in the matrix. A schematic of the conceived two-step austempering process is shown in Fig. 2. A preliminary study by one of these investigators (S.K.P.) on a low manganese ADI [18] has shown that the two-step austempering is a feasible process for development of ADI. The present investigation is a continuation of that study where the two-step process has been applied to an ADI with conventional composition. The primary objective of this investigation was thus to examine the validity of the hypothesis, i.e., whether large supercooling and carbon partitioning will indeed produce the desired microstructure. The secondary objective was to examine whether ner ferrite and austenite together with higher austenitic carbon will improve the mechanical properties as well as fracture toughness of ADI of conventional composition. The nal objective was to compare these properties with ADI processed by single-step process, so that to establish a novel route by which ADI can be further strengthened and toughened. 2. Experimental procedure 2.1. Material The material used in the present investigation is an alloyed nodular cast iron of conventional composition.

The chemical composition of the material is reported in Table 1. The microstructure of the as-cast material is shown in Fig. 3. The as-cast material was predominantly
Table 1 Chemical composition of the material (wt%) C-3.45 Si-2.48 S-0.012 P-0.013 Mn-0.40 Ni-1.50 Mo-0.30 Mg-0.15

Fig. 3. Microstructure of the ductile iron in as-cast state. Etched by 2% nital, 500.

222

J. Yang, S.K. Putatunda / Materials and Design 25 (2004) 219230

(over 80%) pearlitic in nature. The material was cast in the form of round cylindrical bars of 25.4 mm (1 in.) diameter and 370 mm (14 in.) long and keel blocks. Cylindrical samples for tensile testing were prepared as per ASTM standard E-8 [19] from these round bars. For fracture toughness tests, compact tension samples were prepared as per ASTM standard E-399 [20] from these cast keel blocks. The compact tension specimens had a thickness B 20:2 mm and the width of W 40:4 mm. The aspect ratio a=W was kept at a=W 0:45. The details of these specimens have been reported elsewhere [21]. Two batches of samples were prepared. One batch of samples was processed by single step, i.e., conventional austempering process and the second batch of samples was processed by two-step process. This was done for comparison purpose. The rst batch of samples were initially austenitized at 927 C (1700 F) for 2 h and then austempered at several temperatures like 260 C (500 F), 288 C (550 F), 302 C (575 F), 316 C (600 F), 330 C (625 F), 343 C (650 F), 357 C (675 F), 371 C (700 F), 385 C (725 F) and 400 C (750 F), respectively, for 2 h, and then air cooled. The austempering treatment parameters were so chosen so that the processing will be within the process window of ADI. The second batch of samples was processed by the twostep austempering process. These samples were also initially austenitized at 927 C (1700 F) for 2 h and then quickly quenched into molten salt bath maintained at 260 C (500 F) for 5 min to obtain sucient supercooling and nucleation of ferrite. Immediately after that, the samples were transferred to another salt bath which was maintained at predetermined austempering temperatures, such as 288 C (550 F), 316 C (600 F), 330 C (625 F), 343 C (650 F), 357 C (675 F), 371 C (700 F), 385 C (725 F) and 400 C (750 F). The samples were austempered at these temperatures for 2 h and nally air cooled. 2.2. Tensile testing Tensile testing of samples was performed per ASTM standard E-8 [19]. Three identical samples were tested from each heat-treated conditions. The tests were carried out at a constant engineering strain rate of 4 104 s1 on an MTS (material test system) servo-hydraulic machine at room temperature and in ambient atmosphere. Load and displacement plots were obtained on an XY recorder and from these loaddisplacement diagrams yield strength, ultimate tensile strength and % elongation values were calculated. The average values from three test samples are reported in this paper. 2.3. Fracture toughness testing After heat treatment, the compact tension samples were ground and then polished with 600-grit emery pa-

per. The specimens were pthen precracked in fatigue at a DK level of 10 MPa m with a stress ratio R 0:1 to produce a 2-mm long sharp crack front in accordance with ASTM standard E-399 [20]. After fatigue precracking, the specimens were loaded in tension in a servo-hydraulic MTS test machine and the loaddisplacement diagrams were obtained with a clip gage in the knife edge attachment on the specimens. From these loaddisplacement diagrams, PQ values were calculated using the 5% secant deviation technique per ASTM standard E-399 [20]. From these PQ values, KQ values were determined using the standard stress intensity factor calibration function for compact tension specimens [20]. Since these KQ values satised all the requirements of a valid KIC test per ASTM standard E-399 [20], they were all valid KIC values. Four identical test specimens were tested from each heat-treated condition. Values reported in this paper are averages from these four tests. 2.4. Metallography and X-ray diraction Microstructures of the heat-treated samples were examined by optical microscopy after etching with 2% nital solution. X-Ray diraction (XRD) analysis was performed to estimate the austenite content and the percentage of carbon in austenite, following the procedure of Rundman and Klug [22]. XRD was done using a monochromatic copper Ka radiation at 40 KV and 100 mA. A Rigaku rotating head anode diractometer was used to scan the angular 2h range from 42 to 46 at a scanning speed of 0.25/min and 7292 at a scanning speed of 1/min. The proles were analyzed with Jade 5 software to obtain the peak positions and the integrated intensity of {1 1 1}, {2 2 0} and {3 1 1} planes of FCC (face centered cubic) austenite and {1 1 0} and {2 1 1} planes of BCC (body centered cubic) ferrite. The volume fraction of ferrite (Xa ) and austenite (Xc ) were determined by the direct comparison method using the integrated intensities of {1 1 0} and f2 1 1g planes of ferrite and {1 1 1}, {2 2 0} and f3 1 1g planes of austenite by the following equation [23]: Ic h k l Rc h k l Xc ; Ia h k l Ra h k l Xa 3

where Ic h k l is the integrated intensity from a given plane h k l of austenite phasec; Ia h k l is the integrated intensity from a given plane h k l of ferrite (a phase and Rc h k l and Ra h k l are constants, where   R 1=v2 j F j2 pL e2m ; 4 where v is the atomic volume of unit cell; F is the structure factor; p is the multiplicity factor; L is the Lorentz-polarization factor and e2m is the temperature factor.

J. Yang, S.K. Putatunda / Materials and Design 25 (2004) 219230

223

The carbon content of austenite was determined by the equation [24]: ac 0:3548 0:0044Cc ; 5 where ac is the lattice parameter of austenite in nanometer and Cc the carbon content of austenite in wt%. The {1 1 1}, {2 2 0} and {3 1 1} planes of austenite were used to estimate the lattice parameter. Four samples were examined and the average from these samples is reported here.

3. Results and discussions 3.1. Microstructure Some of the microstructures of the samples processed by single-step austempering process are shown in Figs. 4(a)(d). Similarly, some of the microstructures of the samples austempered by two-step process are reported in Figs. 5(a)(d). The austempered microstructure shows

a matrix of dark needle-shaped ferrite plus bright etching austenite with nodular graphite dispersed in the matrix. Signicant coarsening of both ferrite and austenite grains were observed as the austempering temperature increased. At lower austempering temperatures, very ne ferrite and austenite was detected. On the other hand, at higher austempering temperature, coarse and feathery ferrite characteristic of upper bainitic structure was observed in both single- as well as twostep process. An estimate of the ferritic cell size (d) was obtained from the X-ray diraction prole from the breadth of {2 1 1} diractometer peaks of ferrite using the Scherrer equation [23]: d 0:9k=b cos h; 6 where k is the wave length, b is the breadth of (2 1 1) peak of ferrite at half height in radians and h is the Bragg angle. In Fig. 6, the ferritic cell size (d) has been plotted against the austempering temperature. It shows evidently that the ferritic cell size (d) increases as austempering temperature increases in both single- and

Fig. 4. Microstructures of ADI heat-treated by the single-step austempering process. All samples were austenitized at 927 C (1700 F) for 2 h and austemperd in salt bath maintained, respectively, at: (a) 399 C (750 F); (b) 371 C (700 F); (c) 343 C (650 F) and (d) 288 C (550 F). Samples were etched by 2% nital, 1000.

224

J. Yang, S.K. Putatunda / Materials and Design 25 (2004) 219230

Fig. 5. Microstructures of ADI heat-treated by the two-step austempering process. All samples were austenitized at 927 C (1700 F) for 2 h and quenched to the 260 C (500 F) salt bath (rst austempering temperature) for 5 min, and then quickly transformed to the higher temperature salt bath (second austempering temperatures) maintained, respectively, at: (a) 399 C (750 F); (b) 371 C (700 F); (c) 343 C (650 F) and (d) 288 C (550 F). Samples were etched by 2% nital, 1000.

two-step austempering processes. However, the two-step austempering has generally resulted in ner ferrite size above 288 C (550 F) than single step. At 288 C (550 F), no signicant dierence was observed in ferritic cell size. These observations are consistent with the nucleation-growth mechanism in metallurgy. For two-step austempering, samples were rst quenched to a lower temperature (260 C (500 F)) for the purpose of creating greater supercooling to facilitate more ferrite nucleation. When the austempering temperature was raised to the second austempering temperature, the nucleated ferrite grows at the expense of austenite domains. Since there is greater nucleation of ferrite, smaller ferrite particles are obtained by two-step austempering process. At 288 C (550 F), since the temperature dierence between the rst and second austempering temperature is very small (only 50 F), there is no predominant eect on ferrite cell size, i.e., the resultant ferrite cell size in both single and two steps are very similar. On the other hand, when the dierence between rst and second

austempering temperature is large, the resultant ferritic cell size is much smaller in two-step process than the single-step process. Thus, the greater supercooling in two-step process has resulted in ner ferrite and austenite and it appears the transformation reaction is predominantly controlled by the nucleation process rather than growth. Two-step austempering in salt bath is preferable over the oil quenching process since quenching rate will not be too severe to cause some distortion or even micro cracking. Moreover, it will prevent any possibilities of martensite formation. The volume fraction of austenite and its carbon content were also determined by X-ray diraction. Fig. 7 is a plot of volume fraction of austenite against the austempering temperature. The austenite content (Xc ) was observed to increase with increase in austempering temperature. During the austempering process, the ferrite forms out of austenite and since ferrite dissolves very little carbon, for the ferrite needles to grow, the carbon from ferrite must diuse out into the sur-

J. Yang, S.K. Putatunda / Materials and Design 25 (2004) 219230

225

55

Single-step austempering Two-step austempering

45

35

25

15

5 240

290

340

390

440

Fig. 6. Inuence of austempering temperature on the mean particle size d (nm) of ferrite.

0.5
Single-step austempering Two-step austempering

0.4

temperature, growth rate of ferrite is low. Hence, the ferrite becomes ner in nature. Thus, at lower austempering temperature, ferrite volume fraction (Xa ) is larger and at the same time ferrite and austenite are ner in nature. This means austenite volume fraction (Xc ) is lower. On the other hand, at higher austempering temperature because of lower supercooling, the nucleation of ferrite is lower and at the same time higher diusion rate of carbon causes the ferrite to become coarse or feathery in nature. Thus, at higher austempering temperature, e.g., 371 C (700 F) and above, we obtain higher volume fraction of austenite, but ferrite and austenite are both coarser in nature. As it is evident from Fig. 7, the two-step process has resulted in a larger volume fraction of austenite in the matrix. This is because two-step austempering process has resulted in ner ferrite cells than single-step process, as shown in Fig. 6. For a given surface area of microstructure, ner ferrite particles will occupy smaller spaces of the matrix (with smaller (Xa ), therefore, the austenite content (Xc ) becomes larger. As mentioned earlier, the nucleation process is more signicant than growth, hence due to larger supercooling, more ferrite particles are nucleated in two-step process. Since the growth process is not a dominant factor, the growth rate of ferrite remains relatively low. Hence, it results in ner ferrite and austenite in the matrix. Since ner ferrite occupies a smaller area, the austenite volume fraction is higher in two-step process. The inuence of austempering temperature on the carbon content of austenite is reported in Fig. 8. As the
2.2

Volume fraction of austenite (X)

Mean partice sized (nm) of ferrite

0.3
2

Single-step Austempering Two-step Austempering

0.2

Carbon content in austenite (C) (%)

1.8

0.1

1.6

0 240 280 320 360 400

1.4

Fig. 7. Inuence of austempering temperature on volume fraction of transformed austenite (Xc ).

1.2

rounding austenite. Since nucleation depends on the supercooling, at lower temperature because of larger supercooling more ferrite is nucleated and at the same time because of the lower diusion rate of carbon at this

240

290

340

390

440

Fig. 8. Inuence of austempering temperature on carbon content (Cc , %) of austenite.

226

J. Yang, S.K. Putatunda / Materials and Design 25 (2004) 219230

55
0.8

Single-step austempering two-step austempering

50 45 40 Hardness (HRC) 35 30 25

Single-step austempering Two-step austempering

0.7

Austenitic carbon(X*C) (%)

0.6

0.5

0.4

0.3

0.2

20
0.1

15
0 240 280 320 360 400

10 240

290

340
o

390

440

Fig. 9. Inuence of austempering temperature on austenitic carbon (Xc Cc , %).

Austempering temperature ( C)

Fig. 10. Inuence of austempering temperature on hardness.

austempering temperature increases, the carbon content of the austenite also increases due to higher diusion rate of carbon at higher temperature. It can be seen from Fig. 8 that two-step process has resulted in a much higher carbon content in austenite than the single-step process. During the two-step austempering process, the samples were transferred to the second austempering salt bath maintained at higher temperatures. Hence, diusion rate of carbon has increased and this has resulted in greater partitioning of carbon and as a result the carbon content of the austenite was higher. In Fig. 9, the austenitic carbon content (Xc Cc ), i.e., total carbon in austenite has been plotted against the austempering temperature. The two-step process has generated a higher austenitic carbon content (Xc Cc ) than the singlestep process. Since both austenite content and the carbon content of austenite are higher in two-step process, the austenitic carbon content (Xc Cc ) is accordingly higher in two-step process. This has indeed validated the hypothesis of this investigation. 3.2. Tensile properties

Single-step austempering Two-step austempering

1400

0.2% Yield strength (Mpa)

1200

1000

800

600

400 240

290

340

390

Fig. 11. Inuence of austempering temperature on 0.2% yield strength.

Figs. 10 and 11 show inuence of austempering temperature on hardness and yield strength of the ADI processed by single- and two-step processes, respectively. Fig. 12 shows the eect of austempering temperature on the tensile strengths of the materials (single step as well as two-step process). From these gures, it is evident that both hardness and yield strength decrease as the austempering temperature increases. Since both

the austenite and ferrite becomes coarser with the increase in austempering temperature, the hardness, yield and tensile strengths decrease. It must be pointed out that the yield and tensile strengths of the materials obtained by single-step process are comparable to the values reported in literature [713]. For example, ASTM

J. Yang, S.K. Putatunda / Materials and Design 25 (2004) 219230

25

227

Single-step austempering Two-step austempering 1600

20

Single-step austempering Two-step austempering

Ultimate tensile strength (MPa)

Percentage elongation (%)

240 290 340 390

1400

15

1200

1000

10

800
5

600
0 240 280 320 360 400 440

Fig. 12. Inuence of austempering temperature on ultimate tensile strength. Fig. 13. Inuence of austempering temperature on ductility (percentage elongation).

Grade 4 ADI species a minimum yield strength of 1100 MPa for Grade 4 ADI. After austempering at 260 C (500 F) by single-step process, an average yield strength of 1210 MPa was obtained. Similarly, yield strength acquired in ADI by single-step process after austempering at 330 C (625 F) was 950 MPa, which was more than 10% higher than the specied value of 850 MPa for the ASTM Grade 3 iron. This implies that the austempering processes were successful and produced the anticipated results. Figs. 1012 show that the two-step austempering process has resulted in higher hardness, higher yield and higher tensile strengths than the singlestep process. This is due to the ner ferrite and austenite produced by two-step austempering process. It is well known (HallPetch equation) that the ner grains result in higher yield and tensile strengths. The ductility of the materials (two-step as well as single step) is displayed in Fig. 13. Increased ductility was obtained at higher austempering temperature. However the two-step process has resulted in lower ductility than the single-step process. This reduction in ductility is probably related to two reasons. First of all, ductility generally decreases as the strength increases. Moreover, it is possible that there was the formation of e or H agg carbides within ferrite matrix during the two-step process. As mentioned earlier, the two-step austempering process forms ner ferrite as well as austenite cells. During austempering, carbon atoms diuse out of ferrite needles into the surrounding austenite. Since the diusion paths are much shorter and the diusion rate higher, in the two-step

process, time needed to saturate austenite with carbon will be less. In this sense, the current 2-h austempering time would be probably beyond the process window of the material, causing the second reaction to take place. Our X-ray diraction proles showed that small peaks appeared in between c2 2 0, a2 1 1 and c3 1 1 peaks in the 7292 2h range for samples processed by two-step austempering process. This observation indicated the possible formation of carbides in the two-step process. Dubensky and Rundman [25] observed unstable transition carbides of e (Fe2 C with a hexagonal structure) and H agg carbide (Fe5 C2 with a monoclinic structure) at prolonged austempering process. They argued that the presence of a particular carbide or carbides will greatly depend on alloy composition and heat treatment, and it is also likely that the times when they rst appear will also vary signicantly from alloy to alloy. Dorazil and Svejcar [26] indicated the existence of carbides at an early austempering time. It is therefore reasonable to deduce the presence of carbides may have partly contributed to the reduction of ductility. However, a TEM analysis is being carried out to examine any presence of carbides. Interestingly, it appears that when we apply two-step austempering process, full 2 h may not be necessary for successful austempering of ADI. In this way, there may be an economical advantage for two-step process since comparable mechanical properties can be obtained at a much shorter duration of austempering.

228

J. Yang, S.K. Putatunda / Materials and Design 25 (2004) 219230

200

3.3. Tensile toughness Tensile toughness of materials was determined by the equation [27]: K e1n ; U 1n

Single-step austempering Two-step austempering

160

7
Tensile toughness U (MPa)
120

where U is the tensile toughness of material, K is the strength coecient, e is the true plastic strain and n is the strain hardening exponent. Tensile toughness is a measure of a materials ability to undergo plastic deformation without fracture. It is measured by the area under the true-stress and true-strain diagram of a material. Fig. 14 is the plot of tensile toughness against austempering temperature. Tensile toughness increases with increase in austempering temperature in both single- and two-step austempering processes. However, at the same austempering temperature, toughness obtained in single step is much higher than that obtained from two-step austempering. This is because toughness is a parameter which comprises both strength and ductility and it is usually determined by the total area under the true stressstrain curve of a material [28]. Though the strength of ADI heat-treated by two-step austempering process is higher than single-step process, its ductility is lower. Fig. 15 reports tensile toughness against the pa1=2 rameter Xc Cc , square root of total carbon content in austenite. Tensile toughness increases linearly with the 1=2 increase of Xc Cc . As the carbon content of the austenite increases austenite becomes tougher, higher

80

40

0.2

0.4

0.6

0.8
1/2

1.2

(XC)

(%)

Fig. 15. Plot of tensile toughness U (MPa) against parameter Xc Cc 1=2 (%).

Single-step austempering Two-step austempering

200

Tensile toughness U (MPa)

150

austenitic carbon has a major role in increasing the tensile toughness of the material. In both austempering processes, higher austempering temperature produced higher toughness, and vice versa. Again, austenite content and its carbon concentration played important roles in determining tensile toughness. Higher volume fraction of austenite and higher carbon content in it gave rise to higher tensile toughness. Therefore, it can be concluded that if an appropriate two-step austempering process could be designed which avoids the formation of carbides in the matrix, i.e., loss of ductility, then it would result in much higher tensile toughness also than the single-step austempering process, since larger amount of austenite and carbon content can be produced by this two-step process. 3.4. Fracture toughness

100

50

0 240 290 340 390

Fig. 14. Inuence of austempering temperature on tensile toughness U (MPa).

The inuence of austempering temperature on the fracture toughness of the material is shown in Fig. 16. Very high fracture toughness values were obtained in the upper bainitic temperature range above 300350 C (575675 F). The two-step process has produced signicantly higher fracture toughness than the single-step process over all the temperatures. In case of single-step process, the fracture toughness was found to initially increase with temperature, reaching a peak value around 350 C (675 F) and then started to decrease with further rise in temperature. However, in the case of two-step process, fracture toughness was found to increase

J. Yang, S.K. Putatunda / Materials and Design 25 (2004) 219230

160

229

Single-step austempering Two-step austempering 70 Fracture toughness K IC (MPa m )

120
1/2

Single-step austempering 140 Two-step austempering

60

Tensile toughness U (MPa)

240 290 340
o

100

80

50

60

40

40

20

30 390
0 35 45 55 65 75 85

Austempering temperature ( C)

Fig. 16. Inuence of austempering temperature on fracture toughness KIC .

Fig. 17. Plot of tensile toughness (U, MPa) against parameter KIC p (MPa m).

continuously till 371 C (700 F), and further elevation in austempering temperature caused fall in fracture toughness. The maximum fracture toughness obtained in ADI by conventional single-step process p (as reported in literature) is between 60 and 66 MPa m [15,16]. But in case of the current two-step process, a p maximum fracture toughness of 73.6 MPa m was obtained, which is 10% higher than the traditional single-step process. Thus, two-step process appears to be a viable process for improving the fracture toughness of ADI. Another interesting aspect of this process is that, generally fracture toughness of a structural material decreases with the increase in its yield strength, however, the present test results have shown that by the two-step process, it is possible to achieve simultaneous high yield strength and high fracture toughness in ADI. This undoubtedly illustrates that the two-step austempering process is an innovative heat treatment approach for ADI. Fig. 17 is a plot of fracture toughness against tensile toughness of the material for both two- and single-step processes. This gure shows that fracture toughness of ADI increases with the increase in tensile toughness for two austempering processes, respectively. At any given tensile toughness level, two-step austempering lead to higher fracture toughness. As discussed above, higher tensile toughness is determined by higher strength and ductility. This again demonstrates the possibility that the proper two-step austempering process will make the ADI simultaneously much stronger and tougher.

In previous publications [29,30], one of the authors with another co-worker developed a mathematical model relating fracture toughness to the parameter ry Xc Cc 1=2 , where ry is the yield strength and Xc Cc the
8000 Single-step austempering Two-step austempering

7000

6000

5000

K IC2 [(MPa) 2 m]

4000

3000

2000

1000

0 0 200 400 600 800 1000 1200 1400

y (X C)1/2 (MPa)
2 Fig. 18. Plot of square of fracture toughness KIC [(MPa)2 m] against parameter ry Xc Cc 1=2 (MPa).

230

J. Yang, S.K. Putatunda / Materials and Design 25 (2004) 219230 [9] Bartosiewicz L, Krause AR, Sengupta A, Putatunda SK. Relationship between Fatigue threshold and Fatigue Strength in Austempered Ductile Cast Iron. International Symposium for Testing and Failure Analysis, vol. 16. ISTFA, ASM; 1990. p. 323 36. [10] Janowak JF, Norton PA. A guide to mechanical properties possible by austempering, 1.5% Ni, 0.3% Mo iron. AFS Trans 1985;88:12335. [11] Wilkinson G, Grupke C. Design Consideration and Product Applications of Casting, 2nd International Conference on Ductile Iron, Ann Arbor, MI; March 1986. p. 34958. [12] Panasiewicz J, Grupke C, Huth J. Chryslers Experience with Austempered Ductile Iron, World Conference on Austempered Ductile Iron, Bloomingdale, IL; March 1991. p. 17694. [13] Okazaki K, Asai H, Tokuyoshi M, Kusonoki H, Sakahara H. Proceedings of the World Conference on Austempered Ductile Iron, Bloomingdale, IL; March 1991. p. 28899. [14] Kovacs BV. Austempered ductile iron, facts and ction. Modern Cast 1990;36:3841. [15] Putatunda SK, Singh I. Fracture toughness of unalloyed austempered ductile cast iron. J Test Eval 1995;23(5):32532. [16] Doong JL, Chen C. Fracture toughness of bainitic-nodular cast iron. Fatigue Fract Engrg Mater Struct 1989;12:15565. [17] Moore DJ, Rouns TN, Rundamn KB. The eect of heat treatment, mechanical deformation and alloying elements on rate of bainitic formation in austempered ductile cast iron. J Heat Treat 1985;4(1):724. [18] Putatunda SK. Development of austempered ductile cast iron (ADI) with simultaneous high yield strength and fracture toughness by a novel two-step austempering process. Mater Sci Engrg A 2001;315(September):7080. [19] ASTM E-8. Standard test method for tensile testing of metallic materials. Annual book of ASTM standards, vol. 3.01. ASTM; 1993. p. 54266. [20] ASTM E-399. Standard test method for plane strain fracture toughness testing of metallic materials. Annual book of ASTM standard, vol. 3.01. Pennsylvania; 1992. p. 745. [21] Rao PP, Putatunda SK. Inuence of microstructure on fracture toughness of austempered ductlie cast iron. Metal Mater Trans 1997;28A(7):145770. [22] Rundman KB, Klug RC. AFS Trans 1982;90:499508. [23] Cullity BD. Elements of X-ray diraction. Reading, MA: Addison-Wesley; 1974. p. 411412. [24] Roberts CS. Trans AIME 1953;197:2034. [25] Dubensky WJ, Rundman KB. An electron microscope study of carbide formation in austempered ductile iron. AFS Trans 1985;64:38994. [26] Dorazil E, Svejcar B. Study of upper bainite in silicon steels. Arch Eisenhuttenaves 1979;50:2938. [27] Alder PH, Golson, Owen WS. Metal Trans A 1986;17A:1725. [28] Dieter GE. Mechanical metallurgy. 3rd ed. New York: McGrawHill; 1986. [29] Rao PP, Putatunda SK. Comparative study of fracture toughness of austempered ductile iron with upper and lower ausferritic microstructure. Mater Sci Technol 1998;14(December): 125763. [30] Rao PP, Putatunda SK. Dependence of fracture toughness of austempered ductile cast iron on austempering temperature. Metal Mater Trans 1998;29A(December):300516.

total carbon contained in austenite. The model implies 2 that the square of fracture toughness (KIC ) has a linear relationship with the parameter ry Xc Cc 1=2 . In Fig. 18, 2 square of fracture toughness KIC was plotted against 1=2 the parameter ry Xc Cc . A directly proportional relationship between the two variables is obvious for both the austempering processes. The higher the value of ry Xc Cc 1=2 , the higher was the fracture toughness, thus validating the earlier model. Since the two-step austempering resulted in higher yield strengths (ry ) and austenitic carbon content Xc Cc 1=2 , a larger product of ry Xc Cc 1=2 and this resulted in higher fracture toughness.

4. Conclusions A novel concept of two-step austempering process for ADI was conceived. This two-step process has resulted in higher yield strength, tensile strength and fracture toughness than the conventional single-step austempering process. The two-step process has also resulted in ner ferrite and austenite as well as higher austenitic carbon (Xc Cc ) in the matrix. This has contributed to improved strength and fracture toughness in ADI. The tensile toughness was lower in the two-step process which can be attributed to the reduction in ductility as a result of two-step process.

References
[1] Dodd J. High strength, high ductility ductile irons. Mod Cast 1978;68(5):606. [2] Gundlach RB, Janowak JF. Development of a ductile iron for commercial austempering. AFS Trans 1983;94:37788. [3] Harding RA, Gilbert GNJ. Why the properties of ductile irons should interest engineers. British Foundryman 1986;79:48996. [4] Schmidt I, Schuchert A. Unlubricated wear of austempered ductile cast iron. Z Metal 1987;78:8715. [5] Johansson M. Austenitic bainitic ductile iron. AFS Trans 1977; 85:11722. [6] Putatunda SK et al. Inuence of microstructure on high cycle fatigue behavior of austempered ductile cast iron. Mater Character 1993;30:22134. [7] Shanmugam P, Rao PP, Udupa KR, Venkataraman N. Eect of microstructure on the fatigue strength of an austempered ductile iron. J Mater Sci 1994;29:493340. [8] Bartosiewicz L, Duraiswamy S, Sengupta A, Putatunda SK. Near Threshold Fatigue Crack Growth Behavior of Austempered Ductile Cast Iron. Morris Fine Symposium, TMS, Detroit; 1991. p. 13538.