UDC

Mechanical with

By

Properties

Duplex

Noboru

of Ductile

Cast

WADE**

and

Yoshisada

UEDA***

Introduction The heat treatment of producing fine duplex structures has been given attention as an effective method of improving the toughness of steels, and intensive studies have been carried out systematically for 9%Ni,l~ Ni-Cr-Mot-4) and stainless steels.5~ As for ductile cast iron, however, there are only few studies,s''~ in which it is reported that ductile cast iron with a ferrite-pearlite duplex matrix produced by an isothermal heat transformation from the pearlitic structure is improved in tensile strength by 15 kg/mm2 and in elongation by 2 to 3 % compared with those of the conventional bull's-eye ductile cast iron. However, more detailed studies will be required in practice about kinds and proportions of the second phases. In the previous works8 13) concerned with the heat transformation of ductile cast iron, it was elucidated that the heat transformation characteristics of ferritic ductile cast iron such as, the austenitizing mechanism, rate of transformation, heat transformation temperature (Ac1) and volume change during transformation are considerably different from those of steels and pearlitic ductile cast irons, and that (1)

*

**

Iron

Matrix*

Synopsis The intention of the paper was to improve the mechanicalproperties of ductilecast iron by a duplexmatrix which is used in steels. Frritic ductile cast iron was heat-treated to producethefollowing duplex matricesof various proportions;ferrite-bainite, ferrite pearlite and ferritetemperedtroostite. The tensile and impact tests wereperformed on the irons with a duplex matrix. The 0.2% proof stress, tensile strength and hardness increase with increasingvolumefraction of the secondphase, but there is no linear relationship known as the law of mixture. The harder the secondphase is, the higher the strength becomes. In the elongationand impact energy of the alloy with higher silicon content, two peaks appear at volumefractions of upper bainite of about 50 and 95%, and the transition temperatures drop to minimums; the elongation values are 18 and 12%, the absorbed and upper shelf energies are 14.5 to 15 kg . m/cm2 and the transition temperaturesare - 45° to - 47°C in the un-notchedspecimen. Thus, the strength and toughnessof ductileiron can be improvedby the proper secondphase of a proper volumefraction in ferritic structure. The improvementcomesfrom thefine duplex matrix structure and low carbon content of the secondphase, which is a characteristic in the austenitizing of ferritic ductile cast iron, and it also comesfrom the secondphase with high strength and high ductility, such as upper bainite, formed mainly around the graphite nodules,probably because of preventinga crack initiation at the graphite-secondphase interface.

I.

669.131.7:539.42:669.112.24:539.537

ferrite-austenite duplex mixture of various proportions can readily be produced practically in a relatively wide range of austenitizing temperature and time, and (2) austenite forms mostly around graphite nodules at higher temperatures but forms at grain boundaries at lower temperatures. The improvement of mechanical properties will be expected14~ by using these duplex matrices obtained from the transformation of ferritic ductile cast irons. The present study was, therefore, performed to examine the mechanical properties of ductile cast iron with duplex matrices of various proportions, and to obtain the optimum condition of improving the strength and toughness. Moreover, an attempt was made to interpret the improvement of the mechanical properties in terms of the observed microstructural change. II. Experimental Procedures Pig iron for ductile cast iron and commercial pure iron were melted in an induction furnace, and commercial metallic silicon and copper were added into the melt. The melt was then treated with a Fe45%Si-10%Mg alloy and cast into the C02 molds having Y-type block of 25 mm thick. The chemical composition of the specimens is given in Table 1. To obtain the ferritic structure, the specimens were annealed at 900°C for 2 hr, furnace-cooled to 720°C, held at 720°C for 20 hr and then air-cooled. The specimens were then heat treated, as shown in Fig. 1, in molten salts (BaC12plus KCl for austenitizing, and NaN03 for austempering and tempering) to produce the following duplex matrices of various proportions; ferrite-bainite, ferrite-pearlite and ferritetempered troostite. All the specimens were preheated at 700°C for 20 min to obtain the homogeneous heating temperature. The austempering Table 1. Chemical composition of specimens. (wt%)

Originally published in Imono (J. Japan Foundrymen's Society), 50 (1978), 305 and 51 (1979), 480, in Japanese. English received February 15, 1980. Department of Iron and Steel Engineering, Faculty of Engineering, Nagoya University, Furo-cho, Chikusa-ku, Nagoya 465. Department of Metallurgy, Faculty of Engineering, Nagoya University.

Research

Article

version

(117)

Transactions

(118)

Fig. Heat trix

ISIJ,

Vol.

21,

1981

1. treatments

for

obtaining

duplex

ma-

structures.

Fig. 2. Effect of the second phase on tensile properties and hardness. (All the specimens were austenitized at 900°C.)

Fig.

3.

Effect properties

was performed at 400° and 300°C to obtain upper and lower bainite, respectively. The holding time of austenitizing in obtaining ferrite-austenite mixtures was pre-determined by metallographic observations. The volume fraction of the second phase (bainite, pearlite and troostite) was measured by an automatic scanning microscope and photographs. After the heat treatment, the specimens of 8 mm diameter and 50 mm gage length for tensile test were finished by polishing with a fine sand paper and sticked on a foil-type strain gage. Tensile test was carried out with an Instron-type universal testing machine of 10 ton capacity at a loading speed of 0.3 cm/min. The instrumented Charpy impact tests (10 kg. m capacity) on the specimens with V-notch and unnotch, 5 X 10X 55 mm in size, were carried out in the temperature range of -196°C to +90°C, and a load-deflection curve was recorded. The test temperatures were adjusted with an appropriate mixture of liquid nitrogen, methyl alcohol and isopentane, an ice water and a warm water. Fractographic and microscopic observations, hardness measurement and calorimetric analysis8'15~were performed.

of

the and

second

phase

on

tensile

hardness.

111. Results 1.

Tensile Properties and Hardness Figures 2 and 3 show the relation of the tensile properties and hardness to the volume fraction of various second phases. As indicated in Fig. 2(a), the values of 0.2% proof stress, tensile strength and hardness increase with increasing volume fraction of the second phases, and the increase is more remarkable in cases of harder second phases of larger proportions. The linear relationship between these values and the volume fraction of the second phase known as the law of mixture does not exist in this case. The similar phenomenon is also recognized in steels,l6~ and it is suggested that the deviation from the linear relationship will come from the difference between the strains in the first and the second phases. on the other hand, the elongation, reduction of area and tensile energy (the energy absorbed by the specimen up to failure, calculated from the stressstrain curve) show considerably complicated variations as indicated in Fig. 2 (b). For instance, the elongation fairly decreases at volume fraction of the second phase of 10 to 20%, then recovers gradually and passes through a maximum value with an increase in

Transactions

the volume fraction of the second phase. The maximum value in elongation appears when the second phase is upper or lower bainite, or pearlite except troostite. Similarly, the tensile energy exhibits a maximum value suggesting the improvement of the toughness at room temperature. The ductile cast iron with a ferrite-upper bainite duplex structure is the most excellent one in ductility and toughness. Figure 3 shows the tensile behavior of ductile cast iron of higher silicon content. Tensile test was performed about the iron with a ferrite-bainite duplex matrix, and some specimens were austenitized at 850°C as well as 900°C in order to find the effect of austenitizing temperature on the tensile properties. The values of 0.2% proof stress, tensile strength and hardness indicate an increasing tendency to the volume fraction of bainite approximately similar to the cast iron of lower silicon content shown in Fig. 2, but these are generally higher. For example, the tensile strength of the iron with higher silicon content is about 10 kg/mm2 higher than that of lower silicon content in both fully upper and lower bainitic structures. The effect of austenitizing temperature on the tensile strength is more apparent when the amount of the bainite is large. In the values of elongation and tensile energy of the iron of higher silicon content, two peaks appear at volume fractions of upper bainite mixture of 50 and 95% as shown in Fig. 3(b); the elongation is 18% and the tensile energy is 27 kg. m (the tensile strength is 68 kg/mm2) at volume fraction of 50%, and the elongation is 12% and the tensile energy is 24 kg. m (the tensile strength is 93 kg/mm2) at 95% volume fraction. The iron of higher silicon content, therefore, has higher values in both strength and ductility at optimum ferrite-upper bainite mixtures. A further research was performed in order to examine the effect of copper on the tensile properties.

Fig.

ISII,

Vol.

21,

1981

(119)

The values of 0.2% proof stress, tensile strength and hardness indicate an increasing tendency to the volume fraction of the second phase approximately similar to the unalloyed iron of higher silicon content (No. 2). There appear also two peaks in the elongation and tensile energy at volume fractions of upper bainite of 50 and 95%, where the elongations are 18% (tensile strength is 65 kgfmm2) and 12% (tensile strength is 82 kg/mm2), respectively. 2. Evaluation of Tensile Strength and Elongation The relationship between the tensile strength and elongation of ductile cast iron with a duplex matrix is summarized in Fig. 4 in comparison with the Japanese Industrial Standard (JIS) requirements. These plots are used conventionally for the evaluation of the tensile properties of metals.7"7~ It is apparent that a considerably wide range of tensile strength and elongation can be obtained by the duplex matrices even if the chemical composition is the same, and that the iron with a ferrite-upper bainite duplex matrix has an optimum combination of strength and elongation, where the elongations at volume fractions of upper bainite of 50 and 95% are about two times higher than those of the iron with a ferrite-pearlite duplex matrix at the same level of tensile strength. 3. Impact Properties at Room Temperature (in the Case of the Un-notchedSpecimen) Figures 5 and 6 show the results of impact test performed at room temperature as a function of the volume fraction of bainite. Where, the maximum load (Pm), yield load (Pa) and deflection (o) were obtained from the loaddeflection curve and are defined as illustrated in Fig. 5. In the impact energy (or called the absorbed energy), E1, was evaluated from the swing angle of hammer in the conventional test, and Eo from the

4.

The

relationship

strength

and

cast

with

iron

between elongation a duplex

tensile of

ductile

matrix.

Fig. 5. Impact properties at room temperature. (In the un-notched specimen)

Fig. 6. Impact properties at room temperature. (In the un-notched specimen)

Research

Article

(120)

Transactions

ISIJ, Vol. 21,

1981

integrated area of the load-deflectioncurve. It is apparent from Fig. 5 that there are two peaks in the impact energy and deflection at volume fractions of the upper bainite of 50 and 95%, produced by austenitizing at 900°C and austempering at 400°C. But no apparent peak exists in the upper bainite mixture produced by austenitizing at 850°C and austempering at 400°C. The maximum load and yield load increase with increasing volume of bainite, and particularly become higher at volume fractions over 90%, and there exists a maximum in Pm at volume fraction of 95%. From Fig. 6 about the lower bainite, it is also evident that the impact energy and deflection decrease considerably with increasing volume fraction of the bainite at first and at volume fraction of about 80%, and are lower than those in the upper bainite mixture, while the maximum load and yield load increase to a great extent with increasing amount of lower bainite and are higher than those in the upper bainite mixture. Some of those values show complicated variations to the bainite volume fraction. This phenomenon is similar to the results from the tensile test. The 0.2% proof stress, tensile strength, elongation and tensile energy correspond to the yield load, maximum load, deflection and impact energy, respectively. It is consequently demonstrated that a remarkable improvement of toughness in the ductile cast iron can be achieved with ferrite-upper bainite duplex mixture, as well as ductility, at optimum volume fractions.

the volume fraction of bainite. In the upper shelf energy of the un-notched specimen, there appear two peaks at volume fractions of the upper bainite of 55 and 95% produced by austenitizing at 900°C and austempering at 400°C, where the values are 15 and 14.5 kg • mf cm2, respectively. This is the similar tendency to the impact energy at room temperature in Fig. 5. In the upper bainite mixture produced by austenitizing at 850°C, however, the upper shelf energy decreases simply with the upper bainite volume , and is somewhat lower than that produced by austenitizing at 900°C. Such a dependence on the austenitizing temperature is a characteristic of the impact behavior of the iron with a ferrite-bainite duplex matrix. In the V-notched specimen, however, the upper shelf energy changes scarcely with the upper bainite , and is considerably lower than that in the un-notched specimen. The transition temperature shows a complicated variation with the upper bainite volume as indicated in Fig. 9. There are two minimum values at volume fractions of the upper bainite of 55 and 95% produced by austenitizing at 900°C, where the values are almost the same as or only a few degrees higher than that of the ferritic, and are lower than those of the upper bainite mixture produced by austenitizing at 850°C and also lower than those of pearlitic iron. Consequently, the optimum conditions for the improvement of the toughness of the iron with a ferrite-upper bainite duplex matrix were revealed.

4.

5.

Transition Temperature Figure 7 shows the impact energy-temperature curves for various matrix structures. Where, the impact energy value was obtained from the swing angle of hammer. It is evident that the impact energy of the iron with a proper ferrite-upper bainite duplex matrix is superior to those of the fully ferritic and bainitic, and also to pearlitic irons, in most range of test temperatures. The upper shelf energy and transition temperature ( TrE) evaluated from the impact energy-temperature curve are shown in Figs. 8 and 9 as a function of

Fig. Fig.

7.

Impact

energy-temperature

curve.

8.

Effect on the

Microscopicand Fractographic Observations The characteristic mechanical behavior of the iron with a duplex matrix, especially in the ductility and toughness as mentioned above, appears to be attractive in practice. It is generally recognized18~that the ductility and toughness of the steel with a fine duplex structure is favorable, because that fine dispersed second phases have a little restriction to the deformation of the first phase (or matrix). Microscopic observations were, therefore, conducted to examine the morphology and distribution of duplex matrix. Typical duplex matrix structures are given

of upper upper

bainite shelf

volume

energy.

Fig.

9.

Effect on

the

of upper transition

bainite

volume

temperature.

Transactions

in Photos. 1 and 2 with various matrices and proportions. It is evident that the ferrite-upper bainite duplex structures are relatively finer, and the fine phases exist around graphite nodules at both volume fractions of 50 and 95%, where the ductility and toughness are improved (see Photos. 1(b) and (c)). However, when the upper bainite of a small portion formed along ferrite grain boundaries, a minimum elongation is obtained (see Photo. 1(a)). Photograph 2 shows, furthermore, that the ferritepearlite and ferrite-troostite duplex structures are relatively coarse. It is apparent from Photo. 3 that the formation site of bainite depends on austenitizing temperatures, and that the upper bainite forms mostly around the graphite nodules when it is austenitized at 900°C, while it forms along the ferrite grain boundaries when it is austenitized at 850°C. This phenomenon comes from the change of the preferential precipitation site, as previously observed,9,1o,12~ of austenite at different austenitizing temperatures. It is therefore suggested that an excellent toughness of the iron is caused by the fine bainite formed around graphite nodules.

Photo.

1.

Effect

Iv.

site

austenitizing

of upper

bainite.

The

specimens

after

austenitizing.

were

temperature on the formation (Specimen No. 4) austempered 20 min at 400°C

21,

1981

(121)

Discussion

1. Ratio of 0.2% Proof Stress to Tensile Strength Figure 10 shows the variation of the ratio of the 0.2% proof stress to the tensile strength (o.2/aT) with the volume fraction of various duplex matrices. I t is observed that there exists a minimum in all duplex matrices at an intermediate proportion and that the ratios are in a relatively wide range of 0.53 to 0.86. The lower value of the ratio is a characteristic of the

Ferrite-upper bainite duplex matrix structures. (Specimen No. 2. Austenitized at 900°C and

3. of

Vol.

Photograph 4 is the fracture profile of the ferritic iron tested at various temperatures, which shows typical dimple patterns including a graphite nodule in each dimple, and cleavage facets with river patterns. Photograph 5 shows the fracture profile of the iron with volume fraction of the upper bainite of 95%, where the toughness and ductility are the best. Dimple patterns and cleavage facets with cleavage steps and river patterns are also observed. But they are smaller than those in ferritic iron. Moreover, lots of cleavage steps are seen on the fracture at -98°C .

austempered

Photo. 2. Various duplex No. 1)

Photo.

ISIJ,

at 400°C)

matrices

of

cast

irons.

(Specimen

(122)

Transactions

Photo.

Photo.

ISIJ,

4.

5.

Vol.

Fracture

21,

profile

1981

after

impact

test

at various

temperatures.

(fully ferritic iron with un-notched)

Fracture profile after impact test at various temperatures. (in the un-notched specimen) The specimen was austenitized at 900°C and austempered at 400°C, and the volume fraction of upper bainite of 95% was obtained.

2.

Fig.

10.

Variation of the ratio of 0.2% proof stress to tensile strength with volume fraction of the second phase.

cast iron with a duplex matrix, and will contribute to an excellent workability. The similar phenomenon is also recognized in a steel,19~ which has been given attention as a dual phase steel sheet. A decrease in the ratio up to an intermediate proportion of the second phase may suggest that the plastic deformation of the iron will be accelerated by increasing interface between ferrite and second phases. An increase in the ratio beyond the intermediate proportion, on the other hand, will come from an increase in the harder second phase which has less deformability and restricts the plastic deformation of ferrite, resulting in the enhanced proof stress. Therefore, the ratio will mainly be affected by the deformability of ferrite in the former and that of second phase in the latter.

Exponent of Strain Hardening, n An evaluation for the mechanical properties of metals is frequently performed by means of the exponent of strain hardening.4,20) The n values, which are obtained from the slope of the curve showing the relation between the logarithm of stress and the logarithm of strain, are shown in Figs. 2(b) and 3(b) with a ferrite-bainite duplex matrix. In the case of the iron of lower silicon content as shown in Fig. 2(b), the tendency of the variation of the n value to the volume fraction of the second phase is similar to that of elongation in both cases of upperand lower bainite. In the case of the iron of higher silicon content, however, while the n value shows only an increase with increasing amount of lower bainite, it varies similarly to elongation with increasing amount of upper bainite. It is generally accepted that the increase in the n value scarcely affects on ductility loss in ductile materials, but affects considerably in brittle ones.21> From this respect, ductile cast irons with a ferriteupper bainite duplex matrix produced by austenitizing at 900°C and then austempered at 400°C are excellent in ductility. 3. Relationship betweenthe Impact and Tensile Properties at Room Temperature As described above, the impact values showed a similar tendency to the tensile values as a function of

Transactions

Fig.

Fig.

Relation

of deflection

to

elongation.

Vol.

21,

1981

(123)

Fig. 13. Relation of impact energy to tensile energy. (The tensile energy is obtained from a stressstrain curve in the tensile test.)

11.

Relationship tensile

12.

ISIJ,

between

the

impact

and

properties.

Fig. 14. Relationship between the upper shelf energy and tensile strength. (Percent numbers indicate the volume fraction of upper bainite.)

the volume fraction of the second phase. The relations of these values at the same volume fraction of bainite are illustrated in Figs. 11 to 13. Figure 11 shows that there is an approximately linear relation between the maximum load (and yield load) and tensile strength (and 0.2% proof stress) at room temperature. Figure 12 shows a parabolic relation between the deflection and elongation accompanied with somewhat scattered values. The impact energy exhibits also a parabolic relation with the tensile energy as shown in Fig. 13, but the tendency is different between the upper and lower bainite. Thus, fairly simple relationships are present between the impact and tensile properties. In the impact test, the specimen is deformed dynamically under complicated loading conditions with three-dimentional stress and strain, while in the tensile test, it is deformed statically under a simple uni-axial loading condition.22~ From this respect, the relationship seems to be complicated, but it is relatively simple in fact. Therefore, the instrumented Charpy impact test provides more effective information on the impact properties than the conventional un-instrumented impact test, so that a qualitative or to some extent quantitative evaluation of tensile properties will be possible from the impact test.

Fig. 15. Relationship between the transition temperature and tensile strength. (Percent numbers indicate the volume fraction of upper bainite.)

4. Relation betweenthe Strength and Toughness In this section, the tensile strength and toughness will be related directly. Figures 14 and 15 show the relations of the upper shelf energy and transition temperature to the tensile strength. In the upper shelf energy of the unnotched specimen austenitized at 900°C and austempered at 400°C, there are two peaks at tensile strengths of 68 and 93 kg/mm2, where the energies are 15 and 14.5 kg2mf cm2 and the volume fractions of upper bainite are 55 and 95%, respectively. The upper shelf energy is higher than those of the fully upper bainitic and of the pearlitic, and is also higher than that produced by austenitizing at 850°C. The desirable characteristic can also be seen in the transition temperature shown in Fig. 15 at the same tensile strengths of 68 and 93 kg/mm2. Thus, both the toughness and strength of the iron are improved considerably at these volume fractions of the upper bainite. 5.

Effectof the Site of Bainite Formation on the Toughness As mentioned above, the improvement of toughness can be achieved only by the duplex matrix of proper volume fraction of the upper bainite produced by austenitizing at 900°C and austempering at 400°C. This suggests that the formation site of upper bainite

(124)

Transactions

ISIT, Vol. 21, 1981

will have some important effects on the toughness. From Figs. 5, 8, and 9 and Photo. 3, it is apparent that the austenitizing temperature affects on the formation site of austenite, which results in bainite by austempering. From these experimental results, therefore, it is suggested that the quality of the matrix structure, especially around graphite nodules, may play an important role in the impact and tensile properties. The graphite is generally a source of a crack initiation site since the brittle and low strength graphite acts as an inner crack in cast iron. A bending test of the specimen, which was polished and picraletched for microstructure examination, was carried out to examine the cracking behavior. Photograph 6 shows that the crack propagation occurs by the linking of graphite nodules ahead of the main crack (Photo. 6(a)), and is prevented by the bainite around the graphite nodule, so that the crack propagates preferably in ferrite (Photo. 6(b)). Thus, the toughness can be improved by the upper bainite with high strength and toughness around graphite nodules since the most preferential site of crack initiation is the graphite-ferrite interface. 6.

The Relation between Mechanical Properties and Carbon Content Dissolved in Austenite To obtain more satisfactory explanation for the reason of the complicated mechanical behaviors, a calorimetric analysis was conducted by using the iron specimen of higher silicon content in order to examine the carbon content dissolved in austenite, because it is supposed that the carbon content will give some effect on the mechanical behavior. A small specimen of 4 mm diameter and 12 mm long was isothermally austenitized for a short period of time at 900°C or 850°C, quenched to room temperature and tempered for 2 hr at 475°C, where no graphitization occurred, and then analyzed calorimetrically. The results are shown in Fig. 16 giving the total amount of carbon dissolved (C) and the average carbon content in austenite (CAF) to the volume fraction of austenite (F). The total carbon content increases with increasing volume fraction of austenite, showing a fairly similar increasing tendency to the tensile strength and 0.2% proof stress. It is therefore suggested that the cabon dissolved in austenite will exert a considerable influence on the tensile strength and proof stress. On the other hand, the average carbon content in austenite decreases at first with increasing volume fraction of austenite up to about 40 to 50%, then stays constantly or increases slightly, and after that increases again considerably. Such a tendency of the average carbon content can also be derived by a theoretical analysis.23'24) Carbon dissolved supersaturately in austenite at an unsteady state decreases rapidly to a required value at a steady state of austenitizing, and it increases again because of more rapid carbon diffusion than increasing volume of austenite. After 100% austenitizing, diffusion of Research

Article

Photo.

Fig.

6.

16.

Crack propagation profile of ductile cast irons. (The arrow indicates the propagation.)

Variation during

of carbon

content

in

bending

direction

dissolved

test

of

in

specimen

main

crack

austenite

austenitizing.

carbon in austenite continues to attain the equilibrium carbon content at a given temperature. From the results of the average carbon content in austenite, the complicated variations in ductility and toughness for the iron of higher silicon content with a ferrite-upper bainite duplex matrix can be explained as follows Since a small amount of harder bainite with higher carbon content acts like a source of stress concentration, the ductility and toughness are lowered up

Transactions

to about 10% in volume fraction of the bainite, and then they become higher in spite of an increase in the volume fraction of bainite up to 50% because of a decrease in the average carbon content. In the latter stage from 50% in volume fraction of bainite, however, they become lower again with increasing fraction of higher strength bainite and decreasing fraction of ductile ferrite. However, there appears a maximum in the ductility and toughness at about 95% in volume fraction of the upper bainite, which comes from the fine dispersed bainite with considerably low carbon content mostly around graphite nodules. The bainite around graphite nodules will also contribute to the matrix strengthening. V. Conclusions Mechanical properties of ductile cast iron with duplex matrices of various proportions were examined systematically by means of the tensile and instrumented Charpy impact tests. The 0.2% proof stress, tensile strength and hardness increase with an increase in the volume fraction of the second phase, and the harder the second phase is, the higher the values become. But there exists no linear relationship known as the law of mixture. The tensile strength of the iron of higher silicon content (2.5%Si) with lower bainite matrix arrives to the maximum value of 141 kgfmm2 (with elongation of 3%). A considerable increase in the elongation and tensile energy occurs at intermediate volume fractions of the second phase, while they decrease at small fractions. The tendency is more apparent in a ferritebainite duplex matrix. In the elongation and tensile energy of the iron of higher silicon content, there appear two peaks at volume fractions of upper bainite of 50 and 95%, where the elongations are 18 and 12%, and the tensile energies are 27 and 24 kg.m (tensile strengths are 68 and 93 kgfmm2), respectively. In the iron containing a large amount of silicon and copper, there appear also two peaks in the elongation and tensile energy at volume fractions of upper bainite of 50 and 95%, where the elongations are 18% (tensile strength of 65 kg/mm2) and 12% (tensile strength of 82 kgf mm2), respectively. The results of impact test performed at room temperature show similar tendencies to those of the tensile test as to the bainite volume fraction. When a ferritic ductile cast iron is austenitized properly for a short period of time at 900°C and then austempered at 400°C, and the volume fractions of upper bainite of 50 and 95% are obtained, the impact energy values are improved to a great extent as compared with other irons having the same strength. From the impact tests at various temperatures, it was proved that there appear two peaks in the upper shelf energy and two minimums in the transition temperature at volume fractions of upper bainite of 50 and 95%. Thus, the toughness at these fractions is apparently improved. The instrumented Charpy impact test provides

Is",

Vol.

21,

1981

(125)

more effective information than the conventional uninstrumented test, and a qualitative assumption for the tensile properties is also possible. From the correlated impact and tensile properties, a considerable improvement in both toughness and strength can be achieved when the iron is austenitized at 900°C and austempered at 400°C, and the volume fractions of upper bainite of about 50 and 95% are obtained. Particularly at the volume fraction of 95%, the tensile strength is 93 kg/mm2 (it is about two times higher than that in ferritic), the transition temperature is -45°C (in the un-notched specimen) and the upper shelf energy is 14.5 kg. mf cm2 (it is higher than that in ferritic). At this volume fraction, in addition, the toughness is considerably higher than that in fully bainitic at the same tensile strength, and also in pearlitic structures. The improvement of mechanical properties comes from the fine duplex matrix structures and low carbon content of the second phase, and also comes from the second phase with high strength and ductility, such as upper bainite, formed mainly around graphite nodules, probably because of preventing a crack initiation at the graphite-second phase interface. Acknowledgements The authors wish to express their appreciation to Asahi Malleable Iron Co., Ltd. for the providing of ductile cast irons, and Y. Kondo, Nagoya Government Industrial Research Institute, and S. Aoki, a student of Nagoya University, now Nichiden Seimitsu Co., Ltd., and I. Ogasawara, a student of Nagoya University, now Howa Kogyo Co., Ltd., for the experimental assistance. REFERENCES 1) 2) 3) 4) 5) 6) 7) 8) 9) 10) 11) 12)

13) 14) 15) 16)

C. W. Marschall, R. F. Hehemann and A. R. Troiano : Trans. ASM, 55 (1962), 135. D. P. Edwards : JISI, 207 (1968), 1494. T. Saito and I. Uchiyama : Tetsu-to-Hagane, 63 (1977), 478. Y. Tomita, S. Oki and K. Okabayashi : Tetsu-to-Hagane, 63 (1977), 1321. R. C. Gibson, H. W. Hayden and J. H. Brophy: Trans. ASM, 61 (1968), 85. L. A. Solntsev: Steel in USSR, 4 (1974), 687. Y. Tanaka and K. Ikawa: Imono (J. Japan Found. Soc.), 48 (1976), 622. Y. Ueda and N. Wade : Imono (J. Japan Found. Soc.), 49 (1977), 25. Y. Ueda and N. Wade: Tetsu-to-Hagane, 63 (1977), 1572. Y. Ueda and N. Wade: Tetsu-to-Hagane, 63 (1977), 2355. Y. Ueda, N. Wade and R. Hara: Imono (J. Japan Found. Soc.), 50 (1978), 26. N. Wade : " Fundamental Studies on the Heat Treatment of Spheroidal Graphite Cast Irons," Doctoral Thesis to Nagoya University, (1978), 1. N. Wade and Y. Ueda : The 7th symposium of J. Japan Found. Soc., Nagoya, (1979), 18. Y. Ueda and N. Wade : Imono (J. Japan Found. Soc.), 50 (1978), 305. A. Jamieson: Foundry, 83 (1955), 132. I. Tamura, Y. Tomota, Y. Yamaoka, S. Kanatani, M.

Research

Article

Transactions

(126)

17) 18) 19) 20)

ISIJ,

Vol.

21,

1981

Ozawa and A. Akao: Tetsu-to-Hagane, 59 (1973), 454. D. L. Grews: Trans. Amer. Found. Soc., 82 (1974), 223. H. Suto: Bull. Japan Inst. Metals, 14 (1975), 7. M. Takahashi, K. Kunishige and A. Okamoto : Bull. Japan Inst. Metals, 19 (1980), 10. S. Koda: Symposium of Strength and Toughness Comm. on Controlled Structures for Improving Ductility and Toughness of Steels, ISIJ, Tokyo, (1975), 43.

Research

Article

21) 22)

23) 24)

T. Gladman, B. Holmes and F. B. Pickering: JISI, 208 (1970), 172. A. S. Tetelman and A. J. McEvily, Jr.: Fracture of Structural Materials, I, trans. by H. Miyamoto, Baihukan, Tokyo, (1976), 3. S. Niedzwiedz, Y. Parton, A. Taub and B. Weiss : Trans. ASM, 58 (1965), 253. N. Wade and Y. Ueda: Submitted to Trans. ISIJ.