Influence of Graphite Morphology and Matrix Structure on Fatigue Strength and Wear Resistance
of Ductile and Austempered Ductile Iron

Khaled Ibrahim* and Ahmed ElSawy**
*Central Metallurgical Research & Development Institute CMRDI),
P.O. Box: 87 Helwan, Cairo, Egypt
**College of Engineering, Tennessee Technological University,
Cookeville, TN 38505,USA

ABSTRACT
The effect of graphite morphology and matrix structure on both fatigue strength and wear behavior of the unalloyed ductile iron (DI) and ADI were studied. The graphite morphology has been changed from spheroidal as in the as-cast condition to ellipsoidal shape after applying a forging process. The matrix structure changed from pearlite-ferrite as in the as-cast and forged conditions to an ausferrite structure that was produced by applying an austempering heat treatment at 400 °C for 1 hour. Optimum mechanical properties were achieved for the austempered ductile iron (ADI), where it obtained an ultimate strength of 1060 MPa and hardness value of 390 HB. The lowest value of tensile strength and hardness were reported for the as-cast DI of 620 MPa and 236 HB, respectively. ADI also showed optimum values of fatigue strength and wear resistance, while the forged DI had a lower fatigue strength value due to the harmful effect of graphite ends that facilitate crack initiation and propagation in the matrix. Forged DI showed better wear resistance than the as-cast DI due to the matrix strengthening effect observed by forging. It was found that optimum mechanical properties can be obtained by using ADI rather than the as-cast and forged DIs.

INTRODUCTION
The effect of graphite configuration and matrix structure on static and fatigue properties of cast iron has been the subject of previous investigations [1]. The fatigue performance of ductile iron depends on quantity, size and shape of the nodular graphite as well as its interaction with the matrix structure. Riposan and Chisamera [2] compared the fatigue behavior of various cast irons containing different graphite configurations. Ductile iron exhibited the best fatigue resistance, while gray iron displayed relatively poor fatigue strength. The superior performance of ductile iron over gray iron has been attributed to the dissimilarity in graphite morphology between the two materials. In recent years, there has been significant interest to increase the fatigue strength of ductile cast iron by strengthening its matrix [3]. This can be achieved by applying the austempering heat treatment to produce the austempered ductile iron (ADI), by performing mechanical deformation on the ductile iron or through a surface hardening treatment such as shot peening or roller burnishing [4]. Janowak et al. [5] concluded that fatigue limit is increased with increasing matrix hardness, pearlite content and graphite nodule count.

The purpose of this study was to explore the correlation between the fatigue properties of as-cast ductile iron, forged ductile iron and austempered ductile iron to the microstructural parameters of graphite nodule configuration and matrix structure.

EXPERIMENTAL PROCEDURE
The material used in this investigation was unalloyed DI (GGG60) nodular cast iron. The chemical composition of this material is 3.63% C, 2.32% Si, 0.51% Mn, 0.023% P, 0.014% S, and 0.27% Cu. One-inch Y-blocks were cast in a medium frequency induction furnace. The bottom of the cast Y-blocks were cut into slabs and divided into three groups. One group was tested in the as-cast condition with no change. The second group was subjected to the austempering heat treatment thermal cycle shown in Fig. 1. The third group was hot forged to an 80% reduction in area. The fatigue strength was determined using the S-N curve approach per ASTM standard E-466 [6]. The rotary bending fatigue test technique was used to evaluate the fatigue strength of the studied material. Smooth cylindrical specimens were machined from the test slabs. The dimensions for these specimens are given in Fig. 2. The specimens were loaded under constant stresses, and the number of cycles for failure was recorded. Adhesion wear testing was carried out in a dry condition against a hardened stainless steel ring with a hardness of 63 HRC. The wear test was carried out using block-on-ring testing machine at different rotating speeds of 100, 150, 200, 250 and 300 rpm under a constant applied load of 150 N for one hour. Dry wear of specimens was measured as a function of weight loss with a sensitivity of 0.1 mg. The microstructural analysis of the samples was completed using optical metallographic methods. The fracture surface of the fatigue samples and the worn surface of the wear samples were examined by using scanning electron microscopy.

Fig. 1 Thermal cycle of the austempering heat treatment

Fig. 2 Configuration of the rotary bending fatigue sample

RESULTS AND DISCUSSION
Metallographic investigation
The microstructure of the three different forms of the studied ductile cast iron (DI) is shown in Figs. 3 a-d. The as-cast DI structure consists of nodular graphite within a pearlitic matrix with about 15-20% ferrite (Fig. 3-a). The forged DI has the same matrix structure as the as-cast DI, but the difference lies in the graphite form. The forging process deformed the as-cast DI spheroidal graphite to oval shape with acute edges in both sides (Fig. 3-b). It is interesting to note that the deformed graphite is still surrounded by the ferrite phase, as illustrated at a higher magnification in Fig. 3-c. The austempered ductile iron (ADI) shows another ausferrite matrix structure, in which the matrix consisted of acicular ferrite and high carbon austenite (Fig. 3-d). The austempering heat treatment process does not affect the graphite form. The volume fraction of retained austenite in ADI was measured using XRD and found to be approximately 21% with the remainder of the matrix being acicular ferrite (Fig. 3-d.)

a.  as-cast DI           X100	b.  forged DI           X100
c.  magnification of (b)           X50	d.  A DI           X100
Fig. 3 Microstructure of the investigated ductile iron

Mechanical Properties
The mechanical properties of the different forms of DI were studied. The fatigue behavior and wear characteristics of each type of DI were thoroughly investigated. Figure 4 shows the hardness values of the three different forms of DI. It is obvious that ADI obtains the maximum hardness value of 390 HB while the as-cast DI demonstrates the lowest value of 236 HB. The hot forged DI develops a higher hardness value of (250 HB) than the as-cast DI due to the work hardening effect from the forging process. The tensile properties of the investigated DI were also determined, as seen in Fig. 5. It is clear that the tensile strength behavior of the three forms of DI follow the same trend as the hardness, where ADI showed the maximum tensile strength and the minimum was for the as-cast structure. It is known that mainly the matrix constituents determine the tensile properties of ductile iron. As a result, the pearlitic-ferritic matrix should obtain lower tensile strength than the ausferritic one, because ausferrite is considered to be a homogenous, strong, fine structure. Therefore, ADI has a high tensile strength value of 1060 MPa, while the lowest value of 620 MPa was exhibited for the as-cast DI due to the low strength value of the pearlite and ferrite matrices. On the other hand, the forged DI shows a relatively higher strength value compared to the as-cast DI due to the effect of work hardening applied on the matrix by the forging process. A maximum elongation of 17% was recorded for the as-cast DI and the lowest one was reported for the forged DI (Fig. 6). The reduction in the ductility of the forged DI can be attributed to the acute edges formed at the ends of the ellipsoidal shaped graphite, which can easily initiate cracks that can propagate in matrix. It could be said that the forged DI looks somewhat like the gray iron, which has very low or no ductility. On the other hand, ADI shows better elongation (10%) than the forged DI, but lower than the as-cast DI. The ductility of ADI can be observed from the high carbon austenite existing in matrix as well as the homogeneity and microstructural scale of the ausferrite matrix [7,8].

Fig. 4 Hardness measurements of the different forms of ductile iron	Fig. 5 Tensile properties of the investigated ductile iron alloy

---

Fig. 6 Elongation percentage of the investigated ductile iron

Fig. 7 S-N curves for evaluating the fatigue limit of different forms of ductile iron

---

Fatigue behavior
Ductile iron became a very attractive engineering material because of high fatigue strength compared to its tensile properties [9]. Thus, it is very important to understand the factors affecting the fatigue properties of ductile iron, such as graphite shape, nodule count and size, matrix structure, and thermal-mechanical surface hardening (i.e. heat treatment, shot-peening, surface rolling), etc [10-12 & 18-21]. Unfortunately, there is no published work on using the forging process to change the DI graphite nodules to elliptical shape and its effect on the fatigue strength. Figure 7 shows the S-N curve for three different types of DI. Results show that the highest fatigue limit is obtained for ADI and the lowest one observed is for the forged DI. This preliminary result shows the effect of both matrix structure and the graphite shape on fatigue properties of DI. The main reason for decreasing the fatigue strength of the forged DI may be attributed to the elliptical graphite shape formed inside the pearlitic-ferritic matrix and its inherent fracture mechanics. Figure 3-c shows the elliptical graphite edges, which have harmful effect on fatigue property of the forged DI. Therefore, the main reasons for this harmful effect can be explained as follows:

• High stress concentration may arise at the edges of the elliptical-formed graphite, which can facilitate fatigue crack initiation and propagation under cyclic load.
• The length of the elliptical graphite is increased; therefore it could be considered as an internal defect that decreases the fatigue strength.
• The distance between edges of the elliptical graphite is shorter than that in the as-cast condition with spheroidal graphite. This may aid in faster fatigue crack propagation.

In spite of the matrix structure, the forged specimen is much harder and has more strength than the as-cast DI. However, the fatigue limit of the as-cast DI is higher than the forged one. This result can affirm that the graphite shape plays a very important role in determining the fatigue strength of ductile iron.

On the other hand, the ADI structure has a higher fatigue limit than both the as-cast and forged DI. This is due to the effect of the fine-grained lath microstructure that also contains some austenite (ausferrite). This high carbon austenite can be transformed into martensite as a result of the applied work hardening during the fatigue test. Such transformation occurring in the plastic zone ahead of the crack would relax the stress concentration at the crack tip [13-15]. The accompanying volume change (transformation of high carbon austenite to martensite) also encourages plastically induced crack closure to occur, reducing the fatigue crack rate and consequently increasing the fatigue strength. Therefore, the ausferrite matrix structure plays a very important role in determining the fatigue strength of ADI when compared to the as-cast DI that has the same graphite shape but different matrix structure. As a result, the ausferrite structure develops higher fatigue strength than the as-cast DI. In conclusion, the matrix structure and graphite shape play an important role in determining the fatigue strength of ductile iron. Therefore, for obtaining a high fatigue limit, the as-cast DI should be austempered to produce ausferrite with a nodular graphite shape.

Fracture surface study
Figure 8 shows a typical fracture surface of the investigated samples. In general, the crack initiation and fracture feature are associated with the matrix constituents and imperfections at the surface of the specimens. As shown in Figure 8-a, the as-cast DI has a cleavage fracture surface due to the existence of a pearlite matrix, which is normally fractured in a brittle mode. The fracture surface of the forged DI illustrates the effect of the graphite shape in developing fatigue crack growth in the matrix, as shown by the arrow in Figure 8-b. These deformed graphite nodules can easily initiate a crack through cyclic loading and then the crack can propagate to the next nodule. This observation is in agreement with the obtained results, in which the forged DI demonstrated the lowest fatigue limit as a result of its graphite shape. Moreover, the forging process destroyed or fractured some graphite nodules, which in turn they can extensively pull out from the matrix. On the other hand, ADI obtained quasi-cleavage fracture (Fig. 8-c.) The observed dimples in this fractography indicating the locations of the existing retained austenite, while the small cleavage areas indicate the bainitic matrix locations, which contain Fe3C as a constituent of its structure.

a- as-cast DI

b- forged DI

c- ADI Fig. 8 Fractographic investigation of the studied specimens

Wear characteristics
The results of wet abrasion wear tests carried out at different rotating speed ranging from 100 to 300 rpm are shown in Fig. 9. The wear rate was calculated as the weight loss (gm) per testing time (sec). As expected, the wear rate increased with increasing the rotating velocity due to the increase in shear stress over the specimen surface. This increase in shear stress leads to an increase in the surface temperature and softening of the face layer of the specimen so it can easily be removed by an adhesion wear mechanism, resulting in an increased wear rate. It can be seen that the lowest wear rate was recorded for the ADI, while the highest wear rate was observed for the as-cast DI. The forged DI wear rate is relatively lower than the as-cast DI. The difference in the wear rate may be attributed to the role of both graphite nodule shape and matrix hardness [16]. It is known that the ADI with the ausferritic matrix has higher hardness than the as-cast and forged DI with pearlitic-ferritic matrices. Generally, ADI shows an increase in hardness of about 40% than the as-cast DI and about 36% than the forged DI. This is the reason that the wear rate of ADI is lower than the other two studied conditions. On the other hand, the matrix strengthening effect due to forging caused the wear rate of the forged DI to be lower than the as-cast DI. In the light of this study it is safe to say that the forging strengthening effect is not enough to decrease the wear rate of the as DI and that the austempering heat treatment is necessary to obtain ausferritic matrix and subsequently increase the wear resistance than the as-cast and forged DI. Furthermore, the effect of the graphite shape is not as clear in determining the wear rate of the studied cases, as is the effect of matrix. Generally, it is known that the graphite has very important role in forming a graphite layer over the contact surfaces, which in turn will act as a lubricant layer that decreases the wear rate [17]. This effect is clear in the difference between the wear rate of both the as-cast DI and the forged one. The as-cast DI has a spheroidal graphite shape that has a geometrically larger surface area as compared to the forged DI that has ellipsoidal graphite shape. This factor of surface area is very important in forming a thick graphite layer between the specimen and the rotating ring in order to protect the specimen from adhesion wear. As a result, the forged DI should obtain a higher wear rate than the as-cast DI. However, the deformed matrix (or strengthening of matrix by forging) dominates over the graphite shape difference, resulting in enhancing the wear resistance of the forged DI as compared to as-cast DI.

Morphology of worn surfaces
The worn surfaces of the investigated specimens were examined using SEM to elucidate the mode of the wear process. Figures 10 a-c shows the worn surfaces of the three different forms of ductile iron, in which the testing velocity was 250 rpm. As shown in Fig. 10-a, ADI obtains the lowest surface deterioration, where the worn surface is characterized by a scratch mechanism due to adhesion wear. This observation coincides with the results obtained for ADI. ADI has a high hardness value, which in turn can well resist the adhesion wear by causing only some scratches on the worn surface. For the as-cast and forged DIs that have lower hardness values, the worn surfaces exhibit another phenomenon. The worn surface of the as-cast DI shows the main basic modes of a severe wear condition (Fig. 10-b.) This severe wear mode can also be described as a lamination and delamination wear mechanism. This mechanism of wear produces lamella automatically from the faced surface due to the high-applied shear stress (as a result of severe condition) over the worn surface. These layers or lamellas will fail out as wear debris from the surface (which is known a delamination mechanism). This process can also produce a worn surface with a rougher wear scar and consequently a higher wear rate. The worn surface (Fig. 10-c) of the hot forged DI demonstrates the same feature (lamination and delamination mechanisms) as the as-cast DI with a less rough worn surface.

a- ADI

b- as-cast DI

c- forged DI Fig. 10 - SEM of the worn surfaces observed at revolution speed of 250 rpm

CONCLUSIONS
Based on this study, it may be concluded that:

1. The as-cast and hot forged DI had a microstructure of pearlite-ferrite, but the forged DI showed elongated graphite nodules. The ADI structure had an ausferrite matrix with approximately 21%-retained austenite.
2. ADI had the highest hardness value of 390 HB. The as-cast DI demonstrated the lowest value of 236 HB, while the forged DI had a slightly higher value (250 HB) than the as-cast DI.
3. The tensile properties showed the same trend as the hardness, where ADI developed the highest tensile value of 1060 MPa and the lowest value of 620 MPa was recorded for the as-cast DI.
4. The lowest elongation was reported for the forged DI as a result of the edges formed at the graphite ends, and the highest value of 17% was obtained for the as-cast DI.
5. The highest fatigue strength was observed for the ADI, while the lowest value was reported for the forged DI due to the harmful effect of graphite ends that easily initiate cracks in the matrix.
6. The maximum wear resistance was reported for the ADI and the minimum one was observed for the as-cast DI. Forged DI obtained a relatively better wear resistance than the as-cast DI due to the matrix strengthening effect developed by forging.
7. The ADI worn surface showed a scratched wear mechanism; while both as-cast and forged DIs exhibited lamination and delamination wear mechanisms.
8. The optimum wear resistance and fatigue strength of DI can be achieved by applying an austempering heat treatment rather than the forging process.

ACKNOWLEDGMENTS
This work was partially supported by NSF OISE-0431481 grant. The authors would like to thank Dr. Kathy Hayrynen, Technical Director of Applied Process Inc., for reviewing the manuscript and providing valuable suggestions. Also, thanks are due to Dr. Mervat Ibrahim of the Egyptian Central Metallurgical Research & Development Institute (CMRDI), for conducting the scanning electron microscopy.

Back to top

REFERENCES

1. D. F. Socie and J.Fash, “Fatigue behavior and crack development in cast irons”, AFS Transactions, Vol. 8, pp. 385-392, 1982.
2. I. Riposan and M. Chisamera, “The influence of graphite morphology on the austempered cast iron properties”, Material Research Society Conference Proceedings, Cast Iron IV, pp. 195-202, 1990.
3. L. Bartosiewicz, A. R. Krause and S. Putatunda, “Influence of microstructure on high-cycle fatigue strength behavior of austempered ductile cast iron”, Materials Characterization, Vol. 30, pp. 221-234, 1993.
4. T. Nakamura, K. Kishima and F. Nagai, “Fatigue strength of austempered ductile iron cylinders”, JSME International journal, series A, Vol. 36, No. 4, pp. 348-353, 1993.
5. S. F. Janowak, A. Alagarsamy D. Venugopalan, “Fatigue strength of commercial ductile irons”, AFS Transactions, Vol. 98, pp.511-518, 1990.
6. American Society for Testing and Materials, Philadelphia, PA, ASTM standard E-466, Annual Book. of ASTM Standard, Vol. 03.01, pp.543-559, 1986.
7. M. Grech and J. M. Young, “Influence of austempering temperature on the characteristics of austempered ductile iron alloyed with Cu and Ni”, AFS Transactions, Vol. 160, pp. 345-352, 1990.
8. M. Bahamani, R. Elliott, “Effect of pearlite formation on mechanical properties of austempered ductile iron”, Materials Science and Technology, Vol. 10, pp. 1068 -1071, 1994.
9. H. T. Angus; Cast Iron; Physical and Engineering Properties, 2nd Edition, Butterworths, London.
10. A. A. Nofal, M. Ramadan and R. Abdel-Karim, “Effect of graphite nodularity on structure and properties of austempered cast iron”, 10th International Conference on Austempered Ductile Iron - your means to improved performance and cost, ASM International Conference Proceedings, USA, September 25-27, 2002.
11. M. Sofue, S. Okada and T. Sasaki, “High quality ductile cast iron with improved fatigue strength”, AFS Transactions, Vol. 86, pp.173-182, 1998.
12. J. Kim, S. Kim and M. Kim, “A study of several factors governing the fatigue limits of austempered cast iron with various microstructures”, Metals and Materials, Vol. 6, No. 3, pp. 249-254, 2000.
13. K. M. Ibrahim, H. N. Eldin, A. A. Nofal and M. M. Ibrahim, “Wear resistance of cold rolled ADI”, International Journal of Steel Grips, Vol. 1, No. 4, pp. 278-283, 2003.
14. C. K. Lin, P. K. Lai and T. S. Shih, “Influence of microstructure on fatigue properties of austempered ductile iron-High cycle fatigue”, International Journal of Fatigue, Vol. 18, No. 5, pp. 297-307, 1996.
15. M. M. Ibrahim, “Thermomechanical processing of austempered ductile cast iron”, Ph.D. Thesis, Helwan University, Cairo, Egypt, 2004.
16. E. P. Fordyace and C. Allen, “The dry sliding wear behavior of an austempered spheroidal cast iron”, Wear, Vol. 135, pp. 265-278,1990.
17. L. Ping, S. Bahadur and J. D. Verhoeven, “Friction and wear behavior of high silicon bainitic structures in austempered cast iron and steel”, Wear, Vol. 138, pp. 269-284, 1990.
18. A.G. Fuller, P.J. Emerson, and G.F. Sergeant, “A Report on the Effect Upon Mechanical Properties of Variation in Graphite Form in Irons Having Varying Amounts of Ferrite and Pearlite in the Matrix Structure and the Use of Nondestructive Tests in the Assessments of Mechanical Properties of such Irons”, AFS Trans., No. 80-90, pp. 21-50, 1980.
19. P.J. Emerson, and W. Simmons, “Final Report on the Evaluation of the Graphite Form in Ferritic Ductile Iron by Ultrasonic and Sonic Testing, and the Effect of Graphite Form on Mechanical Properties”, AFS Trans., No. 76-26, pp.109-128, 1976.
20. B.I. Imasogie, A.A. Afonja, and J.A. Ali, “Properties of Ductile Iron Nodularised with Multiple Calcium-Magnesium based Master Alloy”, Mater. Sci. and Technol., Vol 16, pp. 194-201, 2000.
21. B.I. Imasogie and U. Wendt, “Characterization of Graphite Particle Shape in Spheroidal Graphite Iron using a Computer-based Image Analyzer” Journal of Minerals & Materials Characterization & Engineering, Vol. 3, No. 1, pp 1-12, 2004.

Back to top