Introduction

Austempering ductile iron (ADI) is a widely used material for manufacturing engineering structure components and wear-resistant parts due to its advantages of low cost, ideal strength-to-weight radio, high fatigue resistance, good ductility and wear resistance performance.1,2,3,4 Researchers and designers have developed many quality ADI castings to substitute cast steel, forgings and other spheroidal graphite cast iron components.5,6,7 Excellent comprehensive mechanical performance is also the advantages of ADI in applications.8 Further improvement of the mechanical properties of ADI may increase the safety and weight reduction of components, which is also the long-term aim for materials researchers.9,10,11

A novel two-step austempering process was developed to simultaneously increase strength and toughness.12,13,14,15 The nucleation of ferrite was enhanced during the first-step austempering, which contributed to ausferrite refinement and mechanical property improvement. However, this process needs to be optimized and improved. For example, the effect of first-step austempering process time at different temperatures has not been studied. Considering practical applications, a short soaking time for the first-step austempering may cause operation difficulties during industrial manufacturing. A uniform temperature of the castings is difficult to attain in the whole cross-section. Further research on the strengthening mechanism of this process is necessary to enhance the potential applications of this conditional material.

Therefore, some basic studies have been carried out in our previous work, the composition, the second step temperature and time of Cu-alloyed ADI were optimized in our earlier work.16,17 Cu is a common strengthening element in ADI, and it is typically added into ADI with other strengthening elements, such as Ni and Mo, to improve the hardenability and expand the process window.18,19 However, adding Ni and Mo may increase the cost. This study aims to use Cu as a strengthening element combined with the two-step austempering process for substitution of Mo and Ni to reduce cost and achieve high performance ADI simultaneously.

In this study, the effect of the first-step austempering temperature and time on two-step ADI is examined, and then the mechanism of microstructure evolution and the mechanical property improvement are discussed. Seven groups of samples treated by a two-step austempering process were prepared. The first-step austempering parameters of these samples were different. Optical microscope (OM), scanning eletron microscope (SEM) and transmission electron microscope (TEM) were used to observe the microstructure. Tensile tests, impact tests and plane strain fracture toughness tests are carried out to evaluate the mechanical properties of ADI.

Experimental Procedure

Materials and Heat Treatment

The chemical composition of the Cu-alloyed ductile iron used in this study is shown in Table 1. Scrap steel, pig iron and alloy material were melted in a 50 kg medium-frequency induction furnace. The iron melt was heated up to over 1500 °C and then poured into a handle ladle for nodularization through the sandwich method. The addition of the nodularizer and inoculant is listed in Table 2 and added to the pocket of the sandwich treatment lade. Iron chips were evenly covered over the nodulizer and inoculations. The treated liquid iron was then post inoculated through adding the floating silicon to liquid iron surface in amount of 0.2%. As liquid iron temperature approximately dropped to 1300 °C, it was cast into Y-shaped blocks in phenolic bonded resin sand models.

Table 1 Chemical Composition of Cu-alloyed ADI Used in this Study (wt%)
Full size table
Table 2 The Addition of Nodularizer and Inoculant for the Sandwich Nodulization
Full size table

The Y-shaped block shown in Figure 1a with sizes of 25 mm in thickness and 165 mm in length was used for tensile and impact tests. The block shown in Figure 1b with sizes of 75 mm in thickness and 90 mm in length was used for the plane strain fracture toughness test. It was cut into three pieces with a thickness of 25.5 mm before austempering. The sampling locations are also shown in Figure 1.

Figure 1
figure 1

Y-shaped block and sampling locations:17 (a) block used for tensile and impact tests; (b) block used for fracture toughness tests.

Full size image

The heat-treatment route is shown in Figure 2. All samples were austenitized at 900 °C for 90 min, and then rapidly quenched in a salt bath at a lower temperature for 5–38 min. After, samples were immediately transferred to another salt bath at 360 °C for 45 min. Finally, these samples were air cooled at room temperature. The detailed parameters of the first step austempering are listed in Table 3. The B1, A3 and B2 samples were controlled to achieve a similar transformation amount after first-step austempering, which was estimated from isothermal transformation kinetics curves of the same composition Cu-alloyed DI as shown in Figure 3. These curves were obtained by dilatometry and microscopic test in a Gleeble 1500D system.20

Figure 2
figure 2

Routes of two-step austempering heat-treatment process17.

Full size image
Table 3 Parameters of First-step Austempring Process
Full size table
Figure 3
figure 3

TTT curves and relationship between transformed austenite and isothermal holding time at different temperatures for Cu-alloyed ADI.20

Full size image

Microstructure Characterization

Metallographic specimens were prepared by mechanical grinding followed by polishing, and then etching with 2% nital. The microstructure was observed by an optical microscope (OM, GX71, OLYMPUS, Tokyo, Japan) and a scanning electron microscope (SEM, Quanta200, FEI, Eindhoven, The Netherlands) to examine the morphology and distribution of high carbon austenite and acicular ferrite. The transmission electron microscope (TEM) specimens were prepared by ion-milling (691, Gatan, Pleasanton, CA, USA) and examined with a transmission electron microscope (JEM-2100, JEOL, Tokyo, Japan) operating at 200 kV.

X-ray diffraction (XRD) was used to calculate the austenitic phase volume fraction and its carbon content. The test was performed according to Standard ASTM E 975.21 An X-ray diffractometer (D/max-2600/PC, Rigaka, Japan) was used with Cu Kα radiation at 40 kV and 150 mA, and a scanning rate of 3 min−1 in the range of 30–100°. The mechanical stability of austenite was evaluated by comparing the austenite phase volume fraction variation from the XRD results. The XRD samples were cut from the place of the impact specimen fracture and far away from the fracture. The size of the specimen was 10 mm×10 mm×1.5 mm. The volume fraction of the retained austenite was estimated using the following relationship,

$$\frac{{I}_{\gamma \{hkl\}i}}{{I}_{\alpha \{hkl\}j}}\text{=}\frac{{R}_{\gamma \{hkl\}}{X}_{\gamma }}{{R}_{\alpha \{hkl\}}{X}_{\alpha }}$$

where \({I}_{\gamma \{hkl\}i}\) and \({I}_{\alpha \{hkl\}j}\) are the integrated intensities of a given {hkl} plane from the austenite and ferrite, respectively, Xγ and Xα are the volume fractions of austenite and ferrite, respectively; and \(\frac{{R_{{\gamma \left\{ {hkl} \right\}i}} }}{{R_{{\alpha \left\{ {hkl} \right\}j}} }} \frac{{R_{{\gamma \left\{ {hkl} \right\}i}} }}{{R_{{\alpha \left\{ {hkl} \right\}j}} }}\) is the ratio of the intensity factor corresponding to the crystal plane of {hkl}i from austenite and {hkl}j from ferrite. The {200}, {220} planes of austenite and {200} and {211} planes of ferrite were used to analyze the volume fraction of austenite. The volume fraction of austenite was calculated by each \(\frac{{I_{{\gamma \left\{ {hkl} \right\}i}} }}{{I_{{\alpha \left\{ {hkl} \right\}j}} }} \frac{{I_{{\gamma \left\{ {hkl} \right\}i}} }}{{I_{{\alpha \left\{ {hkl} \right\}j}} }}\), and the result was caculated by summing the four ferrite and austenite peaks and the ratio of the measured integrated intensity to the R-values.

The carbon content of austenite was determined by the equation:

$${a}_{\gamma }=0.358+0.0044{C}_{\gamma }$$

where aγ was the lattice parameter of austenite (nm) and Cγ the carbon content of austenite (wt.%).

Mechanical Properties Test

The tensile test was performed according to Standard ASTM E8 in an electronic universal testing machine (44300, CCS, Changchun, China).22 Specimens were machined according to Standard ASTM E8, and the gauge length and diameter were 30 mm and 6 mm respectively. Tensile strength, yield strength and percentage elongation values were obtained in this test, and the values were the average of three tests.

An unnotched Charpy impact test was performed according to Standard ASTM A897 and ASTM E23 in a pendulum impact testing machine (NI300, NCS, Beijing, China) at room temperature.23,24 The size of the specimen was 10 mm×10 mm×55 mm. The impact energy result was the average value of the highest three test values of four tested samples.

Plane fracture toughness was completed according to Standard ASTM E 1820 in a servo-hydraulic test system (MTS-810, MTS Systems Corporation, USA).25 The compact tension (CT) specimen was 25 mm in thickness and 50 mm in effective width. The CT specimens were ground and polished with 800-grit waterproof sandpaper and then precracked in fatigue at a ΔK level of 10 MPa m1/2 to produce a 2 mm long crack. The fracture toughness result was the average value from three tests.

Hardness test was performed according to Standard ASTM E 10 in a Brinell hardness tester(HB3000D, Laihua, Hangzhou, China) at room temperature.26 Brinell hardness was measured on fractured CT samples using 3000 kg load and a 10 mm diameter tungsten carbide ball. The result was the average value of three tests.

Results and Discussion

Microstructure

Micrographs of Cu-alloyed DI in as-cast condition are shown in Figure 4. The microstructure characteristics are listed in Table 4, which was evaluated by five fields of view by through the Image Pro Plus software. The spheroidization of different thickness Cu-alloyed ductile iron blocks was satisfactory. The graphite nodule count with a maximum Féret diameter of 5μm in 75 mm thick block were slightly less than 25 mm thick blocks, while pearlite content was more. The microstructure and mechanical properties of the two-step ADI were likely not affected from the minimal difference in graphite nodule counts and pearlite content.

Figure 4
figure 4

As-cast micrographs of Cu-alloyed ductile iron for different Y-shaped blocks with magnification, 100x: (a), (b) 25 mm in thickness; (c) and (d) 75mm in thickness.

Full size image
Table 4 The Microstructure Characterization of as-cast Cu-alloyed Ductile Iron
Full size table

First Step Temperature

Figure 5 shows optical micrographs of the two-step Cu-alloyed ADI treated by different first-step temperatures. Plate martensite was clearly distinguished, as marked by the blue dash line in A1 in Figure 5a. This result indicated that retained austenite (RA) was not stable enough. Therefore, a part of RA was transformed to plate martensite during cooling to room temperature.27 As the first-step temperature increased from 240 to 280 °C, ausferrite was significantly refined without martensite transformation. With a further increase in the first-step temperature, the ausferrite is coarsened.

Figure 5
figure 5

Optical microstructure of the two-step Cu-alloyed ADI treated by different first-step temperatures: (a) 240 °C; (b) 260 °C; (c) 280 °C; (d) 300 °C; and (e) 320 °C.

Full size image

Two kinds of ferrite morphologies were distinguished from the SEM images, as marked in Figure 6. The sharper and finer ferrite needles are marked by yellow dashed lines, while thicker and blunt ferrite bunches are marked by red dashed line. No significant change in ferrite bunch morphology was observed with increasing the first-step temperature. However, the thickness of the ferrite needle was slightly increased with the first-step temperature in the range of 240–280 °C. As the first step temperature increased to 320 °C, thicker ferrite needle was observed. It consisted of several parallel arranged ferrite laths, which could not be clearly distinguished in the other samples in which the first step temperature was below 300 °C.

Figure 6
figure 6

SEM micrographs of the two-step Cu-alloyed ADI treated by different first step temperatures: (a) 240 °C; (b) 260 °C; (c) 280 °C; (d) 300 °C; and (e) 320 °C.

Full size image

Finer ferrite laths belonged to ferrite needles compared to that in ferrite bunches in A1 and A3 were observed in TEM images marked by yellow arrows. The ferrite lath was thickened from 40 to 150 nm with a first-step temperature increasing from 240 to 320 °C. The austenite film was distributed between ferrite laths. Moreover, ferrite bunches were observed in the TEM images. Ferrite laths were much thicker than ferrite needles. This result indicated that the ferrite needle and ferrite bunch both consisted of the alternating structures of ferrite laths and austenite films, which were formed at relevant temperatures. The difference between the ferrite needle and ferrite bunch was the thickness of the ferrite lath and austenite film.

As shown in Figure 3b, transformed austenite in A1, A3 and A5 sample respectively was roughly 2, 25 and 55%. The A1 sample transformation percentage was low due to low transformation rate and long incubation period of bainitic transformation at 240 °C. Therefore, few ferritic needles were formed at this stage. Finally, the bainitic ferrite in A1 was mostly formed during second step austempering at 360 °C, which resulted the ausferrite coarsened. As first step increasing to 280 °C, transformation austenite was increased to 25%. These finer ferrite needles separated untransformed austenite, which restricted ferrite bunches growth. Therefore, the ausferrite gradually refined. With further increase of first step temperature, more ferrite needles were formed in A5 sample during first step austempering, which were little different morphologic comparing to ferrite bunches. This was the reason for ausferrite coarsened for A5 sample (Figure 7).

Figure 7
figure 7

TEM micrographs of the two-step Cu-alloyed ADI treated by different first step temperatures: (a) 240 °C; (b) 280 °C;17 and (c) 320 °C.

Full size image

Two kinds of RA were also distinguished from the figures. Blocky RA was distributed among ferrite needles and ferrite bunches, and was easily observed in OM and SEM images due to its bright morphology. The austenite film was distributed between ferrite laths that were only observed in TEM images. The blocky RA size significantly decreased with an increase in first step temperature from 240 to 280°C, and then increased with a further increase in first step temperature. The austenite film was thickened in the ferrite needle with an increase in the first step temperature, while the thickness minimally changed in ferrite bunch. The RA volume fraction Xγ and average carbon content of RA of each sample are shown in Figure 8. Xγ initially increased from 35.17 to 42.21%, and then decreased to 29.48% as the first-step temperature increased to 280 °C. Xγ increased to 36.12% with a further increase in the first-step temperature. The austenitic carbon content variation trend was the same as the austenitic volume fraction, which was rising and falling as the first-step temperature increased from 240 to 280 °C. With the first step temperature further increasing, the austenitic carbon content was gradually increased.

Figure 8
figure 8

X-ray diffraction results from the two-step Cu-alloyed ADI treated by different first step temperatures: (a) austenitic volume fraction; and (b) austenitic carbon content.

Full size image

First Step Time

As shown in Figure 9, with the first step time extending, ausferrite was gradually refined, especially for blocky austenite. Blocky austenite tended to decrease with the extension of the first step time. Some obscure martensite was distributed among ferrite in the B1 sample, which indicated that austenite was not stable enough to entirely remain at room temperature.

Figure 9
figure 9

Optical microstructure of the two-step Cu-alloyed ADI treated by different first step times: (a) 0 min20; (b) 5 min; (c) 15 min; and (d) 25min.

Full size image

Ferrite needles and ferrite bunches were simultaneously distinguished from the SEM images, as shown in Figure 10. However, the sizes of ferrite bunches and blocky austenite were significantly reduced with the extension of the first step time. The width of ferrite needles was minimally influenced by the first step time.

Figure 10
figure 10

SEM micrographs of the two-step Cu-alloyed ADI treated by different first step times: (a) 5 min; (b) 15 min; and (c) 25 min.

Full size image

The thickness of ferrite laths was measured from the TEM images, as shown in Figure 11. The results showed that the thickness of ferrite laths in needles was in the range of 40–80 nm, while the thickness in bunches was thicker at approximately 150 nm. Overall, the first step holding time had minimal influence on the thickness of the ferrite laths. This result indicated that the refinement was essentially a result of the increasing ferrite needle content. Moreover, blocky austenite was separated into smaller pieces. Hence, the microstructure showed a refinement trend with extension of the first step time.

Figure 11
figure 11

TEM images of the two-step Cu-alloyed ADI treated by different first step times: (a) 5 min; (b) 15 min; and (c) 25 min.

Full size image

As shown in Figure 12, Xγ gradually decreased from 32.91 to 29.48% with the first-step time extending from 5 min to 15 min. Then, the austenitic volume fraction was reduced to 23.81% with further extending the first step time to 25 min. The carbon content slightly decreased at first and then increased with the extension of the first step time.

Figure 12
figure 12

Retained austenitic content and carbon content of the two-step ADI treated with different first step times: (a) retained austenitic volume fraction, and (b) retained austenitic carbon content.

Full size image

Combined with optical micrographs, the Xγ of B1 being lower than that of A3 resulted from martensite transformation during cooling to room temperature, which warrants discussion. The ausferrite in the A1 and B1 samples was both finer than the ADI treated by a single-step process in earlier research at 360 °C, as shown in Figure 9a), this result was caused by finer acicular ferrite formation as noted by other researchers.12,28 Moreover, the second step process duration was long enough for Cu-alloyed ductile iron to reach nearly complete, as shown in the transformation curves in Fig.3. However, martensite was actually observed in the blocky austenite in A1 and B1, which indicated that the austenite was not stable. This was the phenomenon of austenitic thermal stabilization,28,29 which were also observed in martensite or bainite transformation. Moreover, the inhomogeneous carbon distribution in large blocky austenite was another reason, which resulted the localized relative carbon-poor and martensite transformation. Finally, although the second step temperature was high and the holding time was long enough for the ausferrite reaction, part of the retained austenite is still transformed into martensite during cooling to room temperature.

As shown in Figure 3b, more ferrite was transformed in this stage with the first step time elongated or first step temperature rising. More nuclei also existed, and as the temperature increases to 360 °C, nucleation and growth rapidly divided the grains into smaller pieces with the first-step ferrite needles together. The carbon diffusion distance was shorted and easily homogenously distributed, which increased the stability of blocky austenite. Therefore, plate martensite could not be distinguished in other samples.

It was concluded from the above phenomenon that acicular ferrite was both transformed during the two-step austempering, which caused the different ferrite morphologies in the two-step ADI. The ferrite needle was nucleated and grown during the first step of austempering. At the end of the first-step process, a number of ferritic needles and nucleation occurred. Nucleation grew rapidly at the second step of austempering, which promoted untransformed austenite separation into smaller blocks. Therefore, the size of blocky austenite tented to decrease with increasing ferritic needles. Moreover, new ferrite nucleation was simutaneously formed in untransformed austenite and grew into bunches. The ferrite bunch was unable to grow through the existing ferrite needle during the second step. Hence, the growth behavior of ferrite bunches was limited in these separated austenite grains. To maintain the ausferrite reaction, more ferrite was nucleated. Moreover, the preformed interface between ferrite and austenite has relative high energy, which promoted ferrite nucleation.30 Therefore, the ferrite bunch was refined, resulting in a decrease in diffusion distance of carbon atoms in the austenite film and block, which made it easier for carbon to be homogenously distributed and increased the austenitic stability.

The microstructure was compared between different times at 240 and 320 °C for the first step austempering. As shown in Figure 13a and b, with extension of the first step time at 240 °C, the transformed martensite was supressed, and the ausferrite was significantly refined, especially for the blocky RA. The ferrite needles and bunches were easily distinguished in SEM images. The ferrite lath thickness in the needles measured from TEM images was approximately 40–70 nm, and the ferritic bunch thickness was approximately 100–150 nm, regardless of the extension time during first step. With the first step time extending at 320 °C, ausferrite minimally changed under OM. However, ausferrite was still refined, as shown in Figure 14. Especially for the blocky RA, the size significantly decreased. Moreover, ferrite lath distance was clearly reduced.

Figure 13
figure 13

Optical micrographs of two-step Cu-alloyed ADI treated by different first step times at 240 °C and 320 °C: (a) 240 °C×15 min; (b) 240 °C×38 min; (c) 320 °C×6.5 min; and (d) 320 °C×15 min.

Full size image
Figure 14
figure 14

SEM images of the two-step Cu-alloyed ADI treated by different first step times at 240 °C and 320 °C: (a) 240 °C×15 min; (b) 240 °C×38min; (c) 320 °C×6.5min; and (d) 320 °C×15 min.

Full size image

The microstructure variation indicated that the ausferrite was refined with the extension of the first step time. This was caused by the increase in the finer ferrite needle amount. The first step temperature is the main influencing factor for ferrite lath thickness in needles and austenitic transformation percentage during first step austempering of two-step ADI.

In summary, the blocky RA size decreasing and ferrite needle amount increasing was the reason for the microstructure refinement with the extension of the first step time (Figure 15).

Figure 15
figure 15

TEM images of the two-step Cu-alloyed ADI treated by different first step times at 240 °C and 320 °C: (a) 240 °C×15 min; (b) 240°C×38 min;16 (c) 320 °C×6.5 min; and (d) 320 °C×15 min.

Full size image

Mechanical Properties

First Step Temperature

As shown in Figure 16, the ultimate tensile strength, yield tensile strength and elongation both initially increased with the first step temperature rising, and the maximum ultimate tensile strength, yield strength and elongation reached to 1280, 1190 MPa, 10.2% at 280 °C, respectively. When the first step temperature was over 280 °C, the ultimate strength, yield strength and elongation were reduced with a further increase in the first-step temperature. The impact energy initially decreased from 77.5 to 73.2 J, and then increased with the first step temperature. When the first step temperature was over 280 °C, minimally changes were observed for the impact energy at approximately 90 J. The maximum was 93.5 J at 320 °C. The fracture toughness increased with the first step temperature. When the first step temperature reached 300 °C, the fracture toughness was slightly reduced. With a further increase in the first step temperature, the maximum fracture toughness reached to 66.72 MPa m1/2. The hardness was initially slightly decreased with first step temperature increasing. With the first step temperature further increasing, the trend of hardness variation was similar to tensile strength. The maximum was 349 HB at 280 °C.

Figure 16
figure 16

Mechanical properties of the two-step ADI treated by different first step temperatures: (a) tensile test; (b) impact energy; (c) fracture toughness and (d) Brinell hardness.

Full size image

First Step Time

As shown in Figure 17, with the extension of the first time, the ultimate tensile strength, yield strength and hardness continuously increased. As the first step time was extended to 25 min, the maximum ultimate tensile strength, yield strength and Brinell hardness reached 1375 MPa, 1280 MPa and 360 HB respectively. The elongation, impact energy and fracture toughness gradually decreased. The maximum impact energy, elongation and fracture toughness reached to 97.3 J, 10.5% and 66.27 MPa m1/2, respectively, when the first step time was 5 min at 280 °C.

Figure 17
figure 17

Mechanical properties of two-step ADI treated with different step times: (a) tensile test; (b) impact energy; and (c) fracture toughness; (d) Brinell hardness.

Full size image

As mentioned above, the soft and ductile austenitic volume fraction decreased, while hard and brittle bainitic ferrite volume fraction increased with first step time extension. It was the main reason for the strength increasing and the plasticity decreasing.

As shown in Table 5, the strength increased with the first step time for two-step ADI treated at 240 and 320 °C for first step austempering. The elongation and impact energy were simultaneously decreased.

Table 5 Mechanical Properties of Two-Step Cu-ADI Treated by Different Time at 240 °C and 320 °C for First Step Austempering
Full size table

Mechanical Properties Strengthening Mechanism

In this work, the strength and hardness of the two-step ADI were mainly influenced by ferrite needle size and amount with first step temperature or time variation. All samples treated by different parameters clearly confirmed to this inference. The special case was the A1 sample, which contained plate martensite in matrix and elevated the hardness.

The FCC crystal structure and the low-dislocation-density characteristic in austenite were easier to deformation and more ductile, which was contributed for ADI ductility compared to high-dislocation-density acicular ferrite. Therefore, the ductility of ADI was generally increased with austenite volume fraction.31 However, in this study, a different trend for the A1-A5 samples was observed, which suggested other factor was influenced to ductility of two-step ADI. Firstly, the ausferrite refinement increased the amount of coordination deformation grain during tensile test, which decreased the stress concentration at grain or phase boundaries and delayed crack initiation. Therefore, the ductility was enhanced by ausferrite refinement. Secondly, austenite stability was another factor, which was influenced by austenitic volume fraction, morphology and austenitic carbon content.32 For A1–A5 samples, although RA volume fraction decreasing with ausferrite refinement, the reduced RA was almost blocky, which was less stable than film austenite. For materials containing carbon-richment retained austenite, the loss of austenite before yield behavior was negative for ductility, which was attributed to transformation of high-carbon martensite namely stress-induced martensite.33 In this study, stress-induced martensite transformed from metastable blocky austenite was another reason for elongation decreasing. The amount of loss austenite after fracture was decreased with ausferrite refinement and elongation increasing, which also confirmed to austenite stability improvement was benefit to ductility of two-step ADI.34 Moreover, more amount of stable austenite remained which was transformed after yield behavior, was also benefit for the ductility improvement, which was attributed to TRIP effect. When size of ausferrite was close, the ductility was mainly depended on the austenite volume fraction.

The impact energy and fracture toughness were also influenced by acicular ferrite size, RA volume fraction and stability.14,35,36 It was known that refinement was benefit for toughness due to higher crack propagation resistance by more grain and phase boundaries. Therefore, the ausferrite refinement was contributed to impact energy and fracture improvement. The austenitic plastic deformation and strain induced transformation were both benefit for impact energy due to the increase of strain energy and phase transformation energy consuming. However, the stress-induced martensite in front of crack tips might cause the stress concentration, which was promoted to crack propagation. While the strain-induced martensite ahead of crack tip during fracture was simultaneously increase the plastic energy and transformation energy. It was benefit for toughness. Therefore, the austenitic stability also influenced the toughness of Cu-alloyed ADI. Differing from ductility under tensile test, the plastic deformation was inadequately deformed under the high strain rate condition during impact test, while the plastic deformation was also minimal under the plane strain condition during fracture toughness test. Therefore, the impact energy and fracture toughness were relative less influenced by the RA volume fraction, while the austenite stability was more influential.

Conclusion

  1. 1.

    The matrix of the two-step Cu-alloyed ADI consisted of different morphologies ferrite and retained austenite. The ferrite needle size was mainly influenced by the first-step austempering temperature, while the ferrite needle content depended on the first step austempering time. The ferrite bunches were minimally influenced with first step temperatures and times. A low ferrite needle content caused the ausferrite coarsening and martensite transformation. The overmuch ferrite needles led to minimal morphology variation compared to a single-step ADI.

  2. 2.

    The austenitic volume fraction initially decreased and then increased with an increase in first-step temperature. The martensite transformation was the reason for austenite decreased at 240 °C for first-step temperature. The austenitic stability had a similar trend to austenitic volume fraction as first-step temperature increased. The austenitic volume fraction gradually decreased with the extension of first-step time, while austenitic stability increased.

  3. 3.

    As the first-step temperature increased to 280 °C, the maximum ultimate tensile strength, yield strength and elongation reached 1280, 1190 MPa, 10.2%, respectively. The impact energy and fracture toughness continued to increase with the first step temperature, and the maximum were 93.2 J and 66.72 MPa m1/2, respectively, as the first step temperature increased to 320 °C.

  4. 4.

    As the first step time extended, the ultimate tensile strength and yield strength gradually increased and reached a maximum at 1375 and 1280 MPa, respectively at 25 min. The elongation, impact energy and fracture toughness decreased with the first step time with maximum of 97.3 J, 10.5% and 66.27 MPa m1/2 respectively, when the first step time is 5 min The samples treated for different times at 240 and 320 °C for first step austempering exhibited the same trend.

  5. 5.

    In the first-step austempering process, the decrease of ferrite needle size and the increase of ferrite needle volume fraction were the main reason for ausferrite refinement, which also influenced the austenitic stability. These factors together influenced the mechanical properties of two-step Cu-alloyed ADI. An optimal combination of ausferrite refinement, austenitic volume fraction and austenitic stability could be achieved through the regulation the appropriate first step transformation amount, which ensure the mechanical properties improvement.