Top Banner
1 FATIGUE CRACK INITIATION AND PROPAGATION IN FERROUS POWDER METALLURGY ALLOYS S.J. Polasik, J.J. Williams, and N. Chawla Department of Chemical and Materials Engineering Arizona State University Tempe, AZ 85287-6006 K.S. Narasimhan Hoeganaes Corp. 1001 Taylors Lane Cinnaminson, NJ 08077 ABSTRACT Many of the targeted applications for powder metallurgy materials, particularly in the automotive industry, undergo cyclic loading. It is, therefore, essential to examine the fatigue mechanisms in these materials. The mechanisms of fatigue crack initiation and propagation in ferrous powder metallurgy components have been investigated. The fatigue mechanisms are controlled primarily by the inherent porosity present in these materials. Since most, if not all, fatigue cracks initiate and propagate at the specimen surface, surface replication was used to determine the role of surface porosity in relation to fatigue behavior. Surface replication provides detailed information on both initiation sites, and propagation path of fatigue cracks. The effect of microstructural features such as pore size, mean pore spacing, as well as the heterogeneous microstructure on crack deflection was examined and is discussed. Fracture surfaces were examined to elucidate a mechanistic understanding of fatigue processes in these materials. INTRODUCTION P/M components are increasingly replacing wrought materials in high performance applications. Many of the targeted P/M applications, particularly in the automotive industry, involve cyclic loading. Thus, an understanding of fatigue mechanisms in these materials is essential. IN diffusion alloyed powder bonding between iron and the alloying particles is promoted during an annealing step, typically at an intermediate temperature for partial diffusion to take place, prior to pressing and sintering. Binder treatment of the powder mixture prior to pressing and sintering is a new and effective technique to minimize segregation and eliminates the diffusion-alloying step [1-4]. Advances in binder treated versus conventional diffusion alloyed powders have been driven by the cost effectiveness of binder treatment, particularly since the annealing step is eliminated. In the binder treated process, a polymeric binder mechanically bonds the alloying additions to the larger iron particles, resulting in a smaller path for diffusion during sintering and enhanced densification upon sintering. In order to avoid large-scale residual porosity or harmful by-products, the binder is completely burned out prior to sintering. Burnout is accomplished either by a debinding step (at an intermediate temperature below the sintering temperature), or by heating at a relatively slow rate until the sintering temperature is reached. There are many other advantages of binder-treatment over conventional P/M processes. For example, faster and more consistent flow into the die cavity and increased green strength from
15

FATIGUE CRACK INITIATION AND PROPAGATION IN FERROUS ...

Jan 02, 2017

Download

Documents

trancong
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: FATIGUE CRACK INITIATION AND PROPAGATION IN FERROUS ...

1

FATIGUE CRACK INITIATION AND PROPAGATION INFERROUS POWDER METALLURGY ALLOYS

S.J. Polasik, J.J. Williams, and N. Chawla

Department of Chemical and Materials EngineeringArizona State UniversityTempe, AZ 85287-6006

K.S. Narasimhan

Hoeganaes Corp.1001 Taylors Lane

Cinnaminson, NJ 08077

ABSTRACT

Many of the targeted applications for powder metallurgy materials, particularly in the automotiveindustry, undergo cyclic loading. It is, therefore, essential to examine the fatigue mechanisms inthese materials. The mechanisms of fatigue crack initiation and propagation in ferrous powdermetallurgy components have been investigated. The fatigue mechanisms are controlledprimarily by the inherent porosity present in these materials. Since most, if not all, fatiguecracks initiate and propagate at the specimen surface, surface replication was used to determinethe role of surface porosity in relation to fatigue behavior. Surface replication provides detailedinformation on both initiation sites, and propagation path of fatigue cracks. The effect ofmicrostructural features such as pore size, mean pore spacing, as well as the heterogeneousmicrostructure on crack deflection was examined and is discussed. Fracture surfaces wereexamined to elucidate a mechanistic understanding of fatigue processes in these materials.

INTRODUCTION

P/M components are increasingly replacing wrought materials in high performance applications.Many of the targeted P/M applications, particularly in the automotive industry, involve cyclicloading. Thus, an understanding of fatigue mechanisms in these materials is essential. INdiffusion alloyed powder bonding between iron and the alloying particles is promoted during anannealing step, typically at an intermediate temperature for partial diffusion to take place, priorto pressing and sintering. Binder treatment of the powder mixture prior to pressing and sinteringis a new and effective technique to minimize segregation and eliminates the diffusion-alloyingstep [1-4]. Advances in binder treated versus conventional diffusion alloyed powders have beendriven by the cost effectiveness of binder treatment, particularly since the annealing step iseliminated. In the binder treated process, a polymeric binder mechanically bonds the alloyingadditions to the larger iron particles, resulting in a smaller path for diffusion during sintering andenhanced densification upon sintering. In order to avoid large-scale residual porosity or harmfulby-products, the binder is completely burned out prior to sintering. Burnout is accomplishedeither by a debinding step (at an intermediate temperature below the sintering temperature), or byheating at a relatively slow rate until the sintering temperature is reached.

There are many other advantages of binder-treatment over conventional P/M processes. Forexample, faster and more consistent flow into the die cavity and increased green strength from

Page 2: FATIGUE CRACK INITIATION AND PROPAGATION IN FERROUS ...

2

better bonding of particles takes place by binder treatment. Additionally, there is a reduction offine particle dusting, resulting in more efficient use of alloy powders and plant cleanliness.Furthermore, a pre-alloyed powder, with less compressibility than elemental powders can beused since the binder improves lubricity, and therefore compressibility of the powder [4].

The microstructure of the binder treated or diffusion alloyed materials is characterized by aporous and heterogeneous microstructure that develops from incomplete diffusion of alloyingelements during sintering. Due to this incomplete diffusion of alloying elements, a variety ofphases are formed. Several investigators have shown that porosity adversely affects themechanical properties of P/M components [5-13]. Under monotonic tensile loading, porosityreduces the effective load bearing cross-sectional area [7], decreasing both strength and ductility[10]. Two types of porosity are typically observed in P/M materials: interconnected and isolatedporosity. Interconnected porosity has a more pronounced effect on ductility than isolatedporosity. Isolated pores imply a larger contribution from the original powder particles and thus alarger deformed volume, effectively increasing the macroscopic ductility. With interconnectedporosity, on the other hand, there are small sinter bonds between particles that fail by highlylocalized ductile rupture, thus reducing macroscopic ductility of P/M materials [6,10]. In thismanner, much of the metallic phase does not contribute to mechanical strength. A usefulanalogy describing the influence of interconnected porosity on mechanical strength is proposedby Danninger et al. [6]. They compare a P/M steel with interconnected pores with a chaincomposed of intermittent thick (heavy) and thin (light) links. The weight of the chain may bedetermined by the heavy links, but the strength is limited by the thin links.

Porosity significantly affects fatigue performance, although the role of porosity in fatigue issomewhat different than that in tension. In many investigations [5,6,8,10,14-16], the preferredsite for crack initiation occurred at pores or pore clusters located at or near the surface or thespecimen. Holmes and Queeney [7] proposed that the relatively high stress concentration atpores, particularly surface pores, is responsible for localized slip leading to crack initiation. Ingeneral, an angular pore creates a higher stress concentration and stress intensity factor than around pore [10]. Christian and German [9] showed that total porosity, pore size, pore shape, andpore separation are important factors that control the fatigue behavior of P/M materials. Poreshave also been proposed to act as linkage sites for crack propagation through interpore ligaments[8]. Recent limited data indicate that binder-treatment has equivalent tensile and fatigueproperties to those of diffusion alloyed materials [1,4].

Since the preferred initiation site for fatigue cracks is frequently the specimen surface, surfacereplication is an ideal means to study crack initiation and growth during fatigue [9,17-19]. Thus,we have used surface replication as a means to studying the role of surface porosity duringfatigue of P/M materials. Surface replication allows the determination of crack initiation and thenature and duration of crack propagation during the fatigue process.

Surface replication is a particularly valuable tool to investigate short fatigue cracks. Short fatiguecracks are on the order of microstructural features such as grains or inclusions. The study ofshort cracks is very important because they often propagate at much faster rates than long fatiguecracks (typically greater than 2mm) under the same driving force. Additionally, short crackspropagate at stress intensities below the long-crack stress-intensity factor, ∆Kth, determined usinga linear elastic fracture mechanics (LEFM) approach. Since cracks that grow until they reach thecritical size needed for failure must have gone through a period when they can be consideredshort cracks, long crack growth data alone is therefore not sufficient to describe fatigue behavior

Page 3: FATIGUE CRACK INITIATION AND PROPAGATION IN FERROUS ...

3

or predict fatigue life [15, 20]. In this paper we have conducted a detailed examination of fatiguecrack initiation and propagation mechanisms of Fe-0.85Mo prealloy, binder-treated P/Mmaterials. By using surface replication and fractographic techniques, a detailed understanding offatigue processes in these materials was obtained.

MATERIALS AND EXPERIMENTAL PROCEDURE

Alloys of Fe-0.85Mo-1.5Cu-1.75Ni-0.6 graphite were fabricated by binder treating a mixture ofpre-alloyed Fe-0.85Mo with the alloying additions. Binder treatment was conducted using aproprietary process developed by Hoeganaes Corp. [2,3]. All powders were pressed intorectangular blanks, to a green density of 7.0 g/cm3. Sintering was conducted at 1120ºC for 30minutes in a 90% N2–10% H2 atmosphere. The as-sintered microstructure of the P/M alloys wasexamined by etching with a 2% Nital solution. Digital image analysis techniques were used todetermine the pore morphology (pore size and shape distribution as well as mean pore spacing)of the material. Tensile tests were conducted in strain control at a constant strain rate of 10-3/s.Cylindrical axial fatigue specimens were machined from the sintered rectangular blanks. Fatiguetesting was conducted on a servohydraulic load frame in load control, an R ratio (σmin/σmax) of –1, and a frequency of 40 Hz (40 cycles per second). The cylindrical axial fatigue geometry,dimensions, and testing procedures are given elsewhere [21].

Fatigue specimens were hand polished, using diamond paste, to a 1µm finish. Surfacereplication was conducted by interrupting the fatigue test intermittently placing the sample undera small tensile load (~ 0.25σmax) to avoid closure of any fatigue cracks during the replicatingprocedure. The entire gauge section was then bathed with acetone and subsequently coveredwith cellulose acetate tape. After the solvent (acetone) evaporated and the replicating tape dried,the replicas were flattened onto double-stick tape and placed between two microscope slides.Using an optical microscope, digital micrographs were taken of the entire replicated surface forthe various fatigue cycles. The largest crack was identified on the last replica prior to failure,and this crack (or cluster of cracks) was identified on prior replicas until the point of crackinitiation. With the current surface replication technique, a crack of length greater than 15µmcould be identified.

RESULTS AND DISCUSSION

Microstructure

During sintering a variety of phases are formed due to localized variations in composition. Themicrostructure in the as-sintered condition of the Fe-0.85Mo-1.5Cu-1.75Ni-0.6 graphite alloy isshown in Figure 1. Without etching, only porosity is observed (Figure 1a). Upon etching, aheterogeneous microstructure consisting of martensite, nickel-rich ferrite, divorced pearlite(“divorced” since it is not lamellar, as in conventional pearlite), and nickel-rich regionssurrounding pores is observed.

Page 4: FATIGUE CRACK INITIATION AND PROPAGATION IN FERROUS ...

4

(a) (b)Figure 1. Micrographs of P/M material tested in as–sintered condition: (a) unetched and (b) etched. Note theprimary and secondary porosity as well as the heterogeneous nature of the microstructure

The porosity is a combination of primary porosity (always present after sintering these materials)and “secondary porosity.” Secondary porosity consists of residual porosity from liquid phaseformation and diffusion of alloying additions, such as copper, at the sintering temperature. Inorder to characterize the pore structure, both the pore size and shape distribution weredetermined from optical micrographs of cross-sections of the samples. Pore size was estimatedby measuring the pore area. Pore shape was then characterized using a shape form factor, F:

2

4

P

AF

π== 1

where A is the measured pore area and P is the measured pore perimeter. For the shape formfactor, a value of “1” denotes a perfectly round pore, and values that approach zero denoteincreasingly irreglar pores. Results of the pore size and shape analyses are shown in Figure 2.

(a)

Page 5: FATIGUE CRACK INITIATION AND PROPAGATION IN FERROUS ...

5

(b)Figure 2. Distributions in (a) pore size and (b) pore shape in P/M alloy.

The pore size and shape analyses show that the P/M material tested consisted mainly of poresbelow 300µm2 in area and relatively irregular shape (most shape factors are between 0.4 and0.7). The irregularity of the pores suggests a greater amount of local stress concentration andsubsequently, a larger amount of potential fatigue crack initiation sites than in a P/M materialwith perfectly spherical porosity.

Tensile Behavior

The tensile stress-strain behavior of the Fe-0.85Mo-1.5Cu-1.75Ni-0.6 graphite alloy is shown inFigure 3. Table 1 summarizes the tensile properties of this alloy. Porosity clearly decreases themacroscopic ductility due to localization of ductile rupture in the sinter bonds. The hard regionsshown in Figure 1 could contribute to a higher work hardening rate and lower ductility than inwrought materials.

Figure 3. Tensile stress-strain behavior of the binder-treated alloy.

Table 1. Tensile Properties of Fe-0.85Mo binder-treated alloyElastic Modulus

(GPa)0.2% Offset YieldStrength (MPa)

Ultimate TensileStrength (MPa)

Strain-to-failure(%)

121 526 774 1.83

Page 6: FATIGUE CRACK INITIATION AND PROPAGATION IN FERROUS ...

6

As is true with most P/M materials, a combination of localized ductile and brittle fracture wasobserved, Figure 4. Ductile rupture, due to microvoid coalescence occurred locally in isolatedsinter bonds and brittle failure generally occurred in large pearlitic grains. Damage was morepredominant in the form of localized ductile rupture of sintered necks, since these necks see mtohave lower strength than larger, brittle pearlitic grains.

(a) (b)Figure 4. SEM micrographs of tensile fracture: (a) ductile rupture in localized sinter bonds and(b) cleavage in large pearlitic grains.

Fatigue BehaviorThe stress versus cycles behavior of the Fe-0.85Mo-1.5Cu-1.75Ni-0.6 graphite P/M steel isshown in Figure 5 (fatigue run-out was taken as 107 cycles). The results of the presentinvestigation are compared to a similar alloy with 0.5% Mo, studied by Chawla et al. [1,4].Notice the clear increase in fatigue resistance with an increase in Mo content. An increase insolid solution strengthening by increasing barriers for dislocation motion and cyclic slip likelyaccount for this behavior.

Figure 5. Comparison of binder treated alloys with 0.5% Mo and 0.85% Mo. An increase in Mo results in anincrease in fatigue life.

Page 7: FATIGUE CRACK INITIATION AND PROPAGATION IN FERROUS ...

7

The stress-strain behavior was also quantified during fatigue. The width of the stress-strainhysteresis loop was used to determine the cyclic plastic strain amplitude and the slope of theloop, i.e., the secant modulus, was used to estimate fatigue damage. In this analysis a usefulparameter to quantify the evolution of damage is the damage parameter [22], DE, defined as

O

E E

ED −−== 1 2

where E is the secant modulus at any given number of cycles and Eo is the elastic modulus of theas-sintered material. Changes in the plastic strain amplitude and damage parameter can be usedto shed some light into the processes of cyclic hardening or cyclic softening operating duringfatigue. In the low cycle fatigue (LCF) regime, the plastic strain amplitude showed a slightdecrease very early in the fatigue life, indicating cyclic hardening, followed by a gradualincrease, i.e., cyclic softening, until failure took place, Figure 6. The damage parameter,however, does not seem to be as sensitive to cyclic hardening and softening as the plastic strain.

(a)

(b)Figure 6. (a) Damage parameter and (b) plastic strain data from stress-strain hysteresis measurements duringfatigue. In both cases, (a) and (b), the LCF regime exhibits cyclic hardening and gradual cyclic softeningwhile in the HCF regime remains relatively unchanged. Ni, the number of cycles to crack initiation, for theLCF specimen is indicated.

In the high cycle fatigue regime (HCF), both the damage parameter and plastic strain amplituderemained relatively unchanged since the local stresses and strains are more elastic than in theLCF regime. For the LCF specimen, the cycle at which crack initiation occurred in a similarsample at the same stress level was determined from surface replication and is indicated in the

Page 8: FATIGUE CRACK INITIATION AND PROPAGATION IN FERROUS ...

8

figure. It is interesting to note that the onset of crack initiation correlates fairly well with thepoint at which DE shows a noticeable increase.

As mentioned above, one of the most important controlling factors in the fatigue resistance ofP/M materials is porosity. Fatigue cracks tend to initiate near pores or pore clusters because ofthe higher localized stress intensity associated with these defects. Pores with a more irregularshape have higher local stresses and stress intensities than smoother, rounded pores. Typically,cracks initiate at pores located at or near the surface of the specimen, because the stress intensityis higher than for a pore within the interior [13]. The number of cycles required for crackinitiation, Ni, was determined from surface replication, and the fraction of the life of thespecimen spent initiating a crack, Ni/Nf, was plotted as a function of cycles to failure, Nf, Figure7. This approach provides a quantitative estimate of the fraction of fatigue life spent in initiationand propagation. There are two pronounced differences between the LCF and HCF regimes;cracks initiate very early (15%) in LCF, yet initiation life accounts for more than 80% in HCF.

0.0

0.2

0.4

0.6

0.8

1.0

103 104 105 106

Figure 7. Ratio of cycles to initiation to cycles to failure. Note that in LCF regime a significant fraction of lifeis spent in crack propagation, while in the HCF regime, most of the fatigue life is spent in crack initiation.

After a crack has initiated at surface or subsurface pores, it tends to propagate and grow throughthe interpore ligaments, using pores as linkage sites, as seen in Figure 8. While otherinvestigators have proposed microcrack linkage as a mechanism, from fractographicobservations [15,23], surface replication allows in situ monitoring of the crack initiation andgrowth process. The initiation and linkage of short fatigue cracks is monitored during cyclicloading, as opposed to post-mortem fractography. The surface replication technique used in thisstudy confirms the linkage of smaller cracks to form one final crack that causes failure of thematerial.

Nf

Page 9: FATIGUE CRACK INITIATION AND PROPAGATION IN FERROUS ...

9

Figure 8. Crack propagation illustrating the role of pores as crack linkage sites (arrows indicate loading axis).

Page 10: FATIGUE CRACK INITIATION AND PROPAGATION IN FERROUS ...

10

Surface replication provides invaluable information on short fatigue cracks, which behavedifferently than long fatigue cracks. As an example, the stress intensity ranges andcorresponding crack growth rate per cycle were calculated. Crack measurements from surfacereplication were used to determine crack growth rates of the small cracks. The stress intensity,K, was calculated using the expression for a surface crack in a cylinder (seen in Figure 9)developed by Newman and Raju [24],

Q

aFK πσ∆=∆ 3

where a is the crack depth, Q is the shape factor for an ellipse (defined as

(( )) 1/for /464.11 65.1 ≤≤++ caca ), ∆σ is the tensile component of the fatigue stress, and F is aboundary correction size factor that accounts for crack size, crack shape, and the ratio of cracksize to specimen diameter. Note that only the tensile component of the fatigue stress range isused to compute ∆K since only tensile loads result in crack growth. Newman and Rajucalculated the values for F using finite element modeling (FEM). A third order polynomial wasfitted to the F values given in [24] and these projected values were used to calculate the stressintensity at the interior of the crack tip, assuming the crack had an a/c of 0.6.

Figure 9. Schematic of surface crack in cylindrical rod used for calculations of stress intensityof fatigue cracks.

The value for the a/c ratio used for the calculations in this study was determined from an SEMmicrograph of a fracture surface, Fig 10. The initiation site was identified and the angle, θ,between two radii of a circle superimposed on the fracture surface was measured. The arclength, 2c, was calculated using the following equation

D360

c2 π

θ

= 4

where D is the diameter of the specimen. Finally, a was measured and a/c was calculated to be0.58, or about 0.6.

Figure 11 illustrates crack growth in LCF and HCF regimes. Due to crack deflection, the valuefor the crack length (2c) was taken as the horizontal projection of the crack (the length of thecrack was taken to be perpendicular to the loading axis). The da/dN data in Figure 11 shows thatwhile several cracks initiate and link together, some fatigue cracks show a deceleration in crackgrowth rate, followed by arrest. Conversely, some cracks decelerate, but grow subsequently.Crack arrest or deceleration may be attributed to microstructural barriers such as grainboundaries, locally hard regions (such as Ni-rich regions), interaction with other cracks, or evenporosity. In addition, many of these cracks measured through surface replication are small

Page 11: FATIGUE CRACK INITIATION AND PROPAGATION IN FERROUS ...

11

cracks and the plastic zone is of comparable size to the crack, requiring high stresses forpropagation [25]. Crack deflection may also take place due to microstructural barriers in thematerial.

(a) (b)

Figure 10. SEM fractograph used to determine a/c crack ratio used for stress intensity calculations.

(a)

(b)Figure 11. da/dN versus stress intensity ranges (∆K) as calculated using Raju and Newman analysis [24] for (a)LCF and (b) HCF.

1mm 1mm

a1/c1 = 0.58

Page 12: FATIGUE CRACK INITIATION AND PROPAGATION IN FERROUS ...

12

Surface replication also enabled the identification of the initiation area on the fracture surface,through scanning electron microscopy (SEM). A SEM micrograph of a fatigue crack initiationsite is shown in Figure 12. The intermittent nature of fatigue crack growth during each cycle canbe observed through the existence of localized fatigue striations in various broken sinter bonds,Fig. 13. Striations occur when plastic strains at the crack tip cause localized slip on planes ofmaximum shear. The crack front therefore undergoes repetitive blunting and sharpening duringpropagation during both the tensile and compressive portion of the cycle. Unlike striations inwrought materials, the striations in P/M materials are highly localized. This is due to the multiplesites of favorable orientation and size that undergo blunting and sharpening and give the step-likeappearance of the striations.

Figure 12. Fatigue crack initiating pore on specimen surface.

Figure 13. Fracture surface after fatigue showing fatigue striations and localized ductile rupture.

Ductile rupture is also present on the fatigue fracture surface, as seen in Figure 13. The ductilefailure regions are highly localized in sinter bonds and develop from microvoid coalescence.

Page 13: FATIGUE CRACK INITIATION AND PROPAGATION IN FERROUS ...

13

The localized necking and microvoid growth, while present in the crack initiation stage, arepredominately found in the later stages of fatigue. Small spherical inclusions, which likelycontributed to microvoid formation, were found at the bottom of some microvoids, Figure 13.The composition of the inclusions was identified as MnS, as determined by energy dispersivespectroscopy (EDS). Cleavage fracture was also observed on the fracture surface. Since this typeof failure was seen in the tensile tests and therefore fast fracture, cleavage failure was associatedwith the fracture surface region corresponding to the final, fast crack propagation leading tofailure. Generally when this cleavage fracture occurred, it was found to be in large pearliticgrains, similar to what is shown in Figure 4.

CONCLUSIONS

The following conclusions can be made concerning the fatigue behavior of binder-treated Fe-0.85Mo-1.5Cu-1.75Ni-0.6 graphite alloys:

• The microstructure of the alloy consisted of a heterogeneous microstructure with areas of“divorced” pearlite, martensite, and nickel-rich ferrite. In addition conventional porositywas accompanied by secondary porosity formed due to the copper particles forming a liquidphase and diffusing into Fe particles during sintering. The pores had relatively irregularshape.

• An increase in Mo content increases the fatigue resistance of the alloy, presumably due toincreased solid solution strengthening and the hindering of cyclic slip mechanisms at poretips.

• The material showed cyclic hardening followed by a gradual cyclic softening, being mostpronounced in LCF. Additionally, the approximate cycles to crack initiation, determinedby surface replication of a similar specimen at the same stress level, coincided with thebeginning of the rapid increase in plastic strain and damage parameter in the LCF regime.Thus, a significant increase in the damage parameter can be correlated to small crackinitiation.

• The fraction of the life required to initiate a fatigue crack, Ni/Nf, tends to increase with adecrease in fatigue stress amplitude. Thus, crack initiation occurs early in the life of LCFspecimens and crack propagation mechanisms, such as deflection due to microstructuralfeatures, are more significant. In contrast, since crack initiation occupies a much higherfraction of fatigue life in the HCF regime, porosity has a more predominant effect in thisregime.

• Short fatigue cracks initiate at surface pores or pore clusters and propagate at faster ratesand lower stress intensities than long fatigue cracks. Linkage of these small fatigue cracksto form the final critical size crack that causes failure was documented by in situ surfacereplication.

ACKNOWLEDGEMENTS

The authors gratefully acknowledge the financial support for this work from Hoeganaes Corp.

Page 14: FATIGUE CRACK INITIATION AND PROPAGATION IN FERROUS ...

14

REFERENCES

[1] N. Chawla, S. Polasik, K.S. Narasimhan, M. Koopman, K.K. Chawla, “Fatigue Behavior ofBinder-Treated Powder Metallurgy Steels,” International Journal of Powder Metallurgy,(2001) in press.

[2] F.J. Semel, “Properties of Parts Made from ANCORBOND Processed Carbon-Nickel-SteelPowder Mix,” in Advances in Powder Metallurgy and Particulate Materials, Metal PowderIndustries Federation, (1989), Princeton, NJ, 9.

[3] S.H. Luk, J.A. Hamill Jr., “Dust and Segregation-Free Powders for Flexible P/MProcessing,” in Advances in Powder Metallurgy and Particulate Materials, Metal PowderIndustries Federation, (1993), Princeton, NJ, 153.

[4] N. Chawla, T.F. Murphy, K.S. Narasimhan, M. Koopman, K.K. Chawla, “Axial FatigueBehavior of Binder-Treated Versus Diffusion Alloyed Powder Metallurgy Steels,”Materials Science and Engineering A, (2001) in press.

[5] D.A Gerard, D.A. Koss, “Low Cycle Fatigue Crack Initiation: Modeling the Effect ofPorosity,” The International Journal of Powder Metallurgy, 26:4 (1990), 337-343.

[6] H. Danninger, D. Spoljaric, B. Weiss, “Microstructural Features Limiting the Performanceof P/M Steels,” The International Journal of Powder Metallurgy, 33:4 (1997), 43-53.

[7] J. Holmes, R.A. Queeney, “Fatigue Crack Initiation in a Porous Steel,” Powder Metallurgy,28:4 (1985), 231-235.

[8] K.D. Christian, R.M. German, “Relation Between Pore Structure and Fatigue Behavior inSintered Iron-Copper-Carbon,” The International Journal of Powder Metallurgy, 31:1(1995), 51-61.

[9] D.A. Gerard, D.A. Koss, “Porosity and Crack Initiation during Low Cycle Fatigue,”Materials Science and Engineering A, A129 (1990), 77-85.

[10] U. Lindstedt, B., Karlsson, R. Masini, “Influence of Porosity on Deformation and FatigueBehavior of P/M Austenitic Stainless Steels,” The International Journal of PowderMetallurgy, 33:8 (1997), 49-61.

[11] S. Carabajar, C. Verdu, A. Hamel, R. Fougeres, “Fatigue Behavior of a Nickel AlloyedSintered Steel,” Materials Science and Engineering A257 (1998), 225-234.

[12] D. Rodzinak, M. Slesar, “The Fatigue Curve of Sintered Iron and its Microstructural andFractographic Interpretation,” Powder Metallurgy International 12:3 (1980), 127-130.

[13] A. Hadrboletz, B. Weiss, “Fatigue Behavior of Iron Based Sintered Material: A Review,”International Materials Review 42:1 (1997), 1-44.

[14] K.V. Sudhakar, “Fatigue Behavior of a High Density Powder Metallurgy Steel,”International Journal of Fatigue, 22 (2000), 729-734.

[15] H. Drar, A. Bergmark, “Initial Fracture Mechanisms in Nickel Alloyed PM Steel,” Fatigueand Fracture of Engineering Materials and Structures, 20:9 (1997), 1319-1330.

[16] H. Drar, “Metallographic and Fractographic Examination of Fatigue Loaded PM-Steel Withand Without MnS Additive,” Materials Characterization, 45 (2000), 211-220.

[17] M.A. Gomes, A.S. Wronski, C.S. Wright, “Fracture Behavior of a Highly Alloyed HighSpeed Steel,” Fatigue and Fracture of Engineering Materials and Structures, 18:1 (1995),1-18.

[18] S.M. McGuire, M.E. Fine, “Prediction of Fatigue Crack Formation in 304 Stainless Steel,”Metallurgical and Materials Transactions A, 27A:5 (1996), 1267-1271.

[19] M.H. Swain, in Small Crack Test Methods, ASTM STP 1149 (1992), 34-56.

Page 15: FATIGUE CRACK INITIATION AND PROPAGATION IN FERROUS ...

15

[20] M. Caton, J.W. Jones, J.M Boileau, J.E. Allison, “The Effect of Solidification Rate on theGrowth of Small Fatigue Cracks in a Cast 319-Type Aluminum Alloy,” Metallurgical andMaterials Transactions A, 30A:12 (1999), 3055-3068.

[21] N. Chawla, C. Andres, J.W. Jones, J.E. Allison, “Effect of SiC Volume Fraction andParticle Size on the Fatigue Resistance of a 2080 Al/SiCp Composite,” Metallurgical andMaterials Transactions A, 29:11 (1998), 2843-2854.

[22] Z.R. Xu, K.K. Chawla, A. Wolfenden, A. Neuman, G.M. Liggett, N. Chawla, “StiffnessLoss and Density Decrease due to Thermal Cycling in an Alumina Fiber/Magnesium AlloyComposite,” Materials Science and Engineering A, 203A (1995), 75-80.

[23] H. Drar, “Fracture Mechanisms Near Threshold Conditions of Nickel Alloyed P/M Steel,”Engineering Fracture Mechanics, 55:6 (1996), 901-917.

[24] I.S. Raju, J.C. Newman, “Stress Intensity Factors for Circumferential Surface Cracks inPipes and Rods under Tension and Bending Loads,” Fracture Mechanics: ASTM STP 905,edited by J.H. Underwood et al., American Society for Testing and Materials, 789-805.

[25] K.J. Miller, “Materials Science Perspective of Metal Fatigue Resistance,” Materials Scienceand Technology, 9 (1993), 453-462.