Linköpings universitet SE–581 83 Linköping
Linköping University | Department of management and engineering
Master’s thesis, 30 ECTS | Engineering materials
202020 | LIU-IEI/LITH-EX-A--2020/Influence of pearlitic and ferritic matrix, molybdenum and
nickel as alloying elements in compacted graphite iron properties by performing fracture
toughness and fatigue crack propagation rate tests.--SE
Influence of matrix and alloying
on the fatigue crack propagation
and fracture toughness of
com-pacted graphite iron for cylinder
heads
–
Inverkan av matris och legeringsämnen på spricktillväxt och
brottseghet i kompaktgrafitjärn ämnade för cylinderhuvuden
Leny Gonzalez
Supervisor : Viktor Norman Examiner : Johan Moverare
Upphovsrätt
Detta dokument hålls tillgängligt på Internet - eller dess framtida ersättare - under 25 år från publicer-ingsdatum under förutsättning att inga extraordinära omständigheter uppstår.
Tillgång till dokumentet innebär tillstånd för var och en att läsa, ladda ner, skriva ut enstaka ko-pior för enskilt bruk och att använda det oförändrat för ickekommersiell forskning och för undervis-ning. Överföring av upphovsrätten vid en senare tidpunkt kan inte upphäva detta tillstånd. All annan användning av dokumentet kräver upphovsmannens medgivande. För att garantera äktheten, säker-heten och tillgängligsäker-heten finns lösningar av teknisk och administrativ art.
Upphovsmannens ideella rätt innefattar rätt att bli nämnd som upphovsman i den omfattning som god sed kräver vid användning av dokumentet på ovan beskrivna sätt samt skydd mot att dokumentet ändras eller presenteras i sådan form eller i sådant sammanhang som är kränkande för upphovsman-nens litterära eller konstnärliga anseende eller egenart.
För ytterligare information om Linköping University Electronic Press se förlagets hemsida http://www.ep.liu.se/.
Copyright
The publishers will keep this document online on the Internet - or its possible replacement - for a period of 25 years starting from the date of publication barring exceptional circumstances.
The online availability of the document implies permanent permission for anyone to read, to down-load, or to print out single copies for his/hers own use and to use it unchanged for non-commercial research and educational purpose. Subsequent transfers of copyright cannot revoke this permission. All other uses of the document are conditional upon the consent of the copyright owner. The publisher has taken technical and administrative measures to assure authenticity, security and accessibility.
According to intellectual property law the author has the right to be mentioned when his/her work is accessed as described above and to be protected against infringement.
For additional information about the Linköping University Electronic Press and its procedures for publication and for assurance of document integrity, please refer to its www home page: http://www.ep.liu.se/.
Abstract
The constants modernization in the fuels used request improvements in the combustion effectiveness, as a consequence the material for components as the cylinder heads must enhance their properties. Seven different compositions of compacted graphite iron (CGI) are analysed with the aim of characterized and select the most suitable material for the cylinder head service condition. The present master thesis report focuses on the impact of the matrix- either pearlitic or ferritic- and alloying elements such as molybdenum and nickel in the fatigue crack propagation ratedNdaand fracture toughness (kIc).
Tests to determined the fatigue crack growth rate, according to the ASTM standard E647 and fracture toughness (ASTM E399) were conducted. The equipment utilized was a servo hydraulic machine, for the fatigue crack propagation rate test and a electromechani-cal machine for the fracture toughness. Moreover, for measure the crack length a portable microscope camera and a camera connected to a DIC(digital image correlation) software was used. The interpretation of the data obtained from the tests were done by TEMA- a DIC software- and MATLAB.
The test results are analysed describing the influence of the composition and the mi-crostructure have over the mechanical properties achieved. Furthermore, an analyse for relating the graphite average length and the hardness with the fatigue crack growth rate and the fracture toughness of the materials is performed.
Acknowledgments
I would like to express special gratitude to my thesis supervisor Viktor Norman. for trusting me, gave me the opportunity of doing this relevant master thesis project research and for all the support, help, attention and time he gave me.
Secondly I would like to thank all the help and good disposition of Patrik Härnman dur-ing the work at the laboratory, without his help and dur-ingenious ideas the tests would have not result as well.
I would like also to thank Matthieu Verdet, specially for all the help in software manage-ment and all the encouragemanage-ment and support he provide me.
Finally I must thank to my family who made all this opportunity to even be a reality. Special thanks to my lovely mom without her love and trust nothing of this could have been done.
Contents
Abstract iii
Acknowledgments iv
Contents v
List of Figures vi
List of Tables vii
1 Introduction 1
1.1 Background . . . 1
1.2 Rationale of The Study . . . 2
1.3 Objectives . . . 3
1.4 Research questions and hypothesis . . . 4
1.5 Assumptions . . . 5
2 Literature Review 6 2.1 Fatigue . . . 6
2.2 Stress Intensity Factor k . . . 8
2.3 Fatigue Crack Growth RatedNda . . . 10
2.4 Importance of the cast iron microstructure . . . 12
3 Method 14 3.1 Relevant terminology . . . 14
3.2 Fatigue crack propagation rate test . . . 15
3.3 Fracture toughness test . . . 19
4 Results and Discussion 24 4.1 Fatigue Crack Propagation Results . . . 24
4.2 Fracture Toughness Results . . . 33
5 Conclusion 35
List of Figures
1.1 Cross section of head cylinder and Engine block . . . 1
2.1 Description of a cyclic parameters from Comsol Multiphysics. . . 6
2.2 Residual stresses distribution from the surface due to peening surface process, Deformation and fracture mechanics of engineering chapter 9. . . 8
2.3 Fracture modes. . . 8
2.4 Stresses involve in crack surface opening from Chapter 10 Deformation and Frac-ture Mechanics for Engineering. . . 9
2.5 Crack growth stages from Chapter 10 Deformation and Fracture Mechanics for Engineering. . . 10
2.6 Stress effect on the fatigue crack propagation rate from Deformation and fracture mechanics of engineering chapter 10. . . 11
2.7 Classification of Metal Alloys from Concepts in Physical Metallurgy . . . 12
3.1 Specimen configuration (C(T)), for the fatigue crack propagation rate test. . . 15
3.2 Fatigue crack propagation tests equipment. . . 18
3.3 Pearlitic 0.5% Nickel 10MM Crack, tests N°2 5X. . . 19
3.4 Alwetron machine utilized for performing the pre-crack and fatigue toughness test. 21 3.5 Chart of load [N] vs displacement [mm] for the specimen 10MM Pearlitic 0.5% Nickel. . . 22
3.6 Types of load vs displacement curves. . . 22
4.1 Fatigue crack propagation chart of ferritic and pearlitic specimens, tests N°1 and N°2. . . 24
4.2 Fatigue crack propagation chart of pearlitic specimens with molybdenum in its composition and pearlitic reference composition specimen, tests N°2 and N°1. . . . 25
4.3 Fatigue crack propagation chart of ferritic specimens with molybdenum in its composition and ferritic reference composition specimen, tests N°2 N°1. . . 25
4.4 Fatigue crack propagation chart of specimens with Nickel in its composition and pearlitic reference composition specimen, tests N°2 and N°1. . . 26
4.5 Fatigue crack propagation chart of pearlitic specimens N°2 and N°1. . . 27
4.6 Chart of fatigue crack growth rate vs average length for all the specimens tested, N°1 and N°2. . . 28
4.7 Chart of fatigue crack growth rate vs hardness for all the specimens tested, N°1 and N°2. . . 29
4.8 Photos showing the extensometer incident. . . 32
4.9 Chart of fracture toughness vs graphite average length for all the specimens tested. 34 4.10 Chart of fatigue toughness vs hardness for all the specimens tested. . . 34
List of Tables
3.1 Table with specimens denotation and chemical composition. . . 17
3.2 Established parameters for the fatigue crack propagation tests. . . 18
4.1 Paris regime coefficients C and n. . . 30
4.2 Validation for crack length measures. . . 31
1
Introduction
1.1
Background
The cylinder head comprises an essential component for the functioning of a vehicle. The cylinder head is normally positioned at the top of the engine block. The main function of the cylinder heads, as can be seen in the figure 1.1, is to close the combustion chamber. A further role is to maintain lubricated the cylinder otherwise the engine is not possible to do even a simple movement [14].
The design of the cylinder heads contains three separately different passages, where cool-ing fluid and lubricant oil go inside the cylinder in the meantime the combustion gas go out of it. The cylinder design has paths that permit the cooling fluid to pass by the engine block which is indispensable to cool the elements in the engine. To allow simultaneously the gases and fluids to go in or out the cylinder, there are valves that are trigger by the camshafts. If the vehicle works with direct combustion the fuel is injected in the cylinder, for which the cylin-der head will present in his design also injectors for this action. The design also prevents, with a head gasket among the cylinder head and the engine block, for the oil and water to leak inside the combustion chamber [14].
1.2. Rationale of The Study
[47]. The definition of fatigue [18] corresponds to repetitive cyclic loading’s which produce an increase in the stress in the material or component, which can terminate in mechanical failure. The fatigue failure process consists in three stages:
1. The high number of cycles will lead the material to present microstructural damage which can growth until identify macroscopy damage observed as a crack.
2. The crack will continue to grow per cycle until it reaches a critical crack length. 3. The material/component fails since at this stage the crack propagates uncontrollably. The develop of the damage is conditioned by the nature of the material, this encompasses the chemical composition, manufacturing, loading, geometry factor and environment. Therefore, the material and design for a cylinder head should prioritized the fatigue resistance proper-ties, considering that they conditioned their durability. The cylinder heads experience high cycle fatigue (HCF), due to the pressure in the cooling passage and low cycle fatigue (LCF) in the gas passage [49]. A high cycle fatigue condition means that the component can en-dure more than 1000 to 10000 cycles before reach failure, LCF correspond to the opposite [25]. Nowadays the innovations in combustion processes have led to an increase in the com-bustion pressure, this have repercussions in the requirements of the cylinder heads materials properties. The materials normally used for manufacturing of cylinders heads are cast irons, which present a high wear resistance therefore more durable and less expensive, others are made of aluminium alloys which are lighter [32].
The cylinder heads normally can be found made of cast iron, such as lamellar graphite iron (LGI) and compacted graphite iron (CGI) incorporated in a metallic matrix, pearlitic or ferritic, see [21]. The nature of the chemical composition and the synthesis processes gives the component properties. Understood this the modifications in the chemical composition by adding alloying elements can have an important effect on the mechanical properties the material will manifest at operating conditions. The type of matrix present in the material will also defined the mechanical properties limits.
As it was explained before, the morphology of the graphite exposed high influence in the material properties [27]. Therefore, the variables to study are graphite morphology, the alloying elements and the matrix.
The innovation for the cylinder heads function is required due to the increase in the use of bio fuels which demand a higher pressure (A deeper explanation in 1.2) therefore, higher temperature in the combustion chamber. For the improve in the combustion is needed a higher operating loads, to accomplish this it is require a smaller engine packages design [20]. In Europe there was an increase from 180[bar] in 1999 to 220-240 [bar] in 2007 in the pressure in the combustion chamber for vehicles [20]. Compact Graphite Iron fulfill this requirements, doubling the fatigue strength of gray cast iron within a 45% higher stiffens and around 75% higher tensile strength [20].
1.2
Rationale of The Study
The global climate changing, is not a recent new, therefore measures have been taken to re-duce the accelerated impact the planet it is suffering. Sweden contemplate as a pioneer in environmental awareness from many years ago, being the first country in create the Envi-ronmental Protection Agency in 1967. Making consecutive conscious decisions over the past fifty years have carried Sweden to be one of the most innovative countries in environmental technology, considering the use of smart grids and bio-fuels [45].
The bio-fuels are fuels that derived from the processing of biomass by carbon fixation similar as happen in plants. The use of plants or animals’ substances to generate energy is defined as biomass. Therefore, the bio-fuels can be created from industrial, domestic or
1.3. Objectives
agricultural organic wastes. Bio-fuels substitute the consumption of large formation time fuels as fossil fuels.
One goal for the innovation in the transportation system is change to a complete use of bio-fuels. The progress in bio-fuels have developed fourth generation of bio-fuels. The sec-ond generation comprehend the use of biomass to produce fuels (gas or liquid), a technology that has been implemented in Sweden for the pass years. Some consequences of the use of bio-fuels over conventional fuels is that they demand a high efficiency in the combustion. Bio diesel presents some methyl esters with longer carbon chains then the traditional diesel, this exhibit a higher viscosity that it implies filter plugging issues. For the use of Syngas, which is made from biomass, the quality is quite lower than natural gas or gasoline, due to this the temperature of the spark ignition (for a spark ignition engine) must be at 30 to 40 Celsius degrees more than usual [17].
An outcome of the use of bio-fuels innovation in the engineering of the transportation is needed. For the cylinder heads function this one is going to be exposed to higher tempera-tures and pressure; thus a more resistance and resilience material is required to be utilized.
1.3
Objectives
1. Identify the suitable material composition for a combustion head cylinder.
Due to the tough service conditions of head cylinder material (high temperatures and cyclic loads), the laboratory work will include fracture toughness and fatigue crack growth rate tests. The outcome from the fracture toughness will provide relevant in-formation as the kIc, at this value the crack will propagate unstably which can lead to instant or unexpected fracture. From the test is expected to notice a different trend between the matrix type (either ferritic or pearlitic). The aim of this investigation is to accomplish significant results. This signifying to find differences between the spec-imens compositions, this manner real conclusions can take in place when to decide which material suits better the cylinder heads purpose.
2. Obtained relevant and veracious information from the test perform.
To have a clear and real knowledge of the fatigue crack propagation rate and fracture toughness for the materials compositions to study, the procedure is going to follow the respective standards. The standard E 399 for fracture toughness tests [8], meanwhile the fatigue crack propagation rate test is standardized by the E 647 ASTM designation [3]. Within the analysis of the data obtained from the tests, micrographs from the fatigue crack propagation tests are going to be performed to complement the research.
3. Identify the correlation between the synthesis and microstructure in the material
properties.
The microstructure will determine the material properties, this concern: a) The percent of alloys elements (molybdenum and nickel).
b) Morphology of the graphite content, which normally is vermicular but in can possi-bly be found lamellar and nodular as well.
c) Effect of the type of matrix, either pearlitic or ferritic.
These features are going to be study and tested to achieve the aim of characterized the material and even improve it. The synthesis aspects contain the chemical composition, solidification and cooling rate for the cylinder heads.
1.4. Research questions and hypothesis
1.4
Research questions and hypothesis
1. How does the composition affect the microstructure and by extension the fatigue
crack propagation?
The fatigue life is the resistance of a material to undergo cyclic loads without risk of failure. This project approach is the impact of the material composition in the properties related to the fatigue resistance.
The fatigue life of the specimen is influence by the material properties (chemical com-position), design geometry, surface quality, residual stresses, internal defects, grain size, some external factors also affect such as temperature, oxidation and corrosion. The fa-tigue resistance depends on the factors mentioned above and the cyclic load to initiate a crack following by the propagation of this one. Even tough commonly the crack ini-tiation in fatigue occurs at the possible surface defects in the material that concentrate the stresses, this research will focus on the influence of the microstructure in the fatigue crack propagation rate and fracture toughness of the material [18].
The material microstructure factors that affect the fatigue crack propagation compre-hends the graphite morphology, the percentage of the phases found in the matrix, such as ferrite and pearlite and the amount of the alloying elements.
2. What are the advantages and disadvantages of the material microstructure?
In the section 2 is it going be explained in depth the major impact in the material proper-ties, also finding a ferritic or pearlitic matrix in the structure have a specific effect in the material as well. Furthermore, the effect on fatigue crack propagation rate and fracture toughness behaviour of alloying elements -molybdenum and nickel - will be analysed for this investigation.
The influence of the type of matrix and the alloying elements in the fracture toughness and fatigue crack propagation rate will be studied, with the aim of generate a char-acterization of the degree of impact of the material microstructure in their properties. Ones the description for each specimen is done, a comparison between them will assist on determine the advantages and disadvantages of the different materials to find an improved selection for the cylinders heads material.
Finding a ferritic or pearlitic matrix in the microstructure have an specific effect in the material as well. A ferritic matrix allows a percent of ductility but no high mechanical resistance, on the other side a pearlitic matrix will give the material a higher mechanical resistance and hardness [23]. The percentage of alloys elements in the material compo-sition can increase the fracture toughness or decrease it depending on the effect they have over the graphite and the matrix phases. It is indispensable to be aware of the fact that the advantages and disadvantages the material selected are going to rely on the service conditions the component. For components exposed to static loads the me-chanical requirements will be different than for a component under cyclic loads, for the first type of loads the maximum stress of service is higher than for a fatigue condition, however the impact of cyclic loads might be more severe than the static ones.
3. How can the material microstructure be achieved? Is possible to improve the material
properties?
The research will allow to select the material that had the best test performance, from this point is relevant to be aware of the mechanism related to the chemical composition, given that this permit the material properties. As it was explained above, the effect of the graphite morphology and the matrix will have an impact on the way the material resists the cyclic loads and crack propagation. If the time permits it, a further research regarding the processing variables will be studied, to comprehend how to achieve the chemical composition desired. Some variables that affect the synthesis of the material
1.5. Assumptions
are raw material, spheroidization, inoculation, pouring temperature and cooling rate and post heat treatment for the component [27].
The investigation considerate compact cast iron with different matrix and alloying el-ements. Therefore, is relevant to be aware of all the knowledge related to this kind of irons. Both the properties and the synthesis of a compact graphite iron falls in the middle between the lamellar and spheroidal graphite irons. For getting the vermicular microstructure of a CGI, the treatment for transforming the graphite should be less time that the treatment for getting a nodular morphology, this way the microstructure does not have the time to transform all the graphite in nodules[23].
1.5
Assumptions
1.5.1
Temperature effect
Is known that the cylinder heads are exposed at high temperatures due to the gas that pro-ceeds from the combustion chamber. The temperature the cylinder head material must en-dure it has increase due to the use of bio-fuels, as it was explained in the section 1.2 .
It is expected that high temperatures might have serious impact in the properties of the component, regardless this factor the project is focused on the characterization of the material though the performance of fracture toughness and fatigue crack growth rate tests.
According to the study paper Fracture toughness and crack growth rate for ferritic and pearlitic compacted cast irons at 25 °C and 150°C[27]. It was noticed that the temperature does not affect severely the fracture toughness with values of 1 to 4 MPa?m lower at 150 °C than at 25°C. For the crack growth resistance, increasing the temperature benefits this property but for irons with a high percentage of vermicularity the temperature has a considerable effect. The irons to study contain a medium vermicularity, therefore the influence of the temperature can be dismissed since is not going to obstruct the material characterization for this master thesis project.
2
Literature Review
2.1
Fatigue
As it is explained in 1.3 a material that is under fatigue conditions will fail due to the cyclic loads that is subjected. The design for a component that experience fatigue must consider the maximum and minimum load in a cycle that although their values are below the static mechanical material properties- such as yield strength- can lead to fracture.
In the objectives section 1.1 the three stages of the fatigue phenomenon are exposed,first the crack has microscopic dimensions, second the crack growth macroscopically until a crit-ical magnitude and finally it breaks. Depending on the material application it is possible to not observe the second stage, this case can lead to an unexpected fracture [18]. Therefore, it is relevant to study the fatigue behavior of materials exposed to cyclic loads conditions.
2.1.1
Cyclic Loads
The cyclic loads are defined as the variable loads from one pick to other, see figure 2.1. Cyclic loads can not be found in the nature but can be easily exemplify and understood by the passing of a train or truck over a bridge[15]. In a general case the cycles do not have the same amplitude[18], such as the train, truck or cars passing over a bridge since they have different mass they induce different loads. For a component application under fatigue, it is expected that cyclic loads vary in a systematic manner in the operational life of the component [31].
2.1. Fatigue
In the figure 2.1 the σarepresents the amplitude of the stress, this one is defined as equa-tion 2.1, usually is measured in [Pa].
σa = σmax´ σmin
2 (2.1)
Where:
σa: the stress amplitude [Pa].
σmax: the maximum stress in a cycle [Pa].
σmin: the minimum stress in a cycle [Pa].
The minimum stress in the cycle is σminand the σmax the maximum stress. The σm corre-spond to the mean stress which is defined as the following equation 2.2.
σm=
σmax+σmin
2 (2.2)
The∆σ is expressed as ∆σ =σmax´ σmin. In this figure is possible to observe clearly the definition of a cycle as the variation between two stress picks.
The effect of the cyclic loads in the fatigue resistance of the material will depend on the frequency and the stress amplitude of the load applied [18].For example the effect of cyclic loads at high frequency might have severe impact on a corrosive environments[1]. Therefore, at the moment of performing a fatigue analysis the cyclic loads are the ones that matter over the statics loads[12].
2.1.2
Residual stresses
When a component is manufactured, as a cast iron, is predictable that the component presents remaining stresses, this stresses can be related to the molding and post heat treatments that the cylinders heads must be exposed before it can be functional, also can exist due to the forces applied in the assemblage of the pieces, this stresses are called residual stresses [50]. Never-theless the total stresses that a cylinder head experience are the residual ones, the stresses due to the cyclic loads (mechanical function) plus the stresses of the load implied by the pres-sure of the gas at high temperature [50]. The residual stresses that a component contain must be taking into account, this one can have a great effect on the mean stress and therefore the fatigue life can be severely reduced[50].
On the other side, the residual stresses are induced to increase the surface hardness of the material [38] since the surface is the area were the crack it is initiated, which contain superficial defects that concentrated stresses. Inducing stress on the surface will decrease the movement of the dislocations,thereby the hardness of the material and the fatigue lifetime of the components are improved [33].
It is been discovered that compressive residual stresses have a larger depth influence from the surface than tension residual stresses, see figure 2.2. Some treatments as peening or sur-face rolling work as a way to induce compressive residual stresses [38].
2.2. Stress Intensity Factor k
Figure 2.2: Residual stresses distribution from the surface due to peening surface process, Deformation and fracture mechanics of engineering chapter 9.
2.2
Stress Intensity Factor k
Due to the high level of stress that the head cylinder needs to hold under fatigue condition, the concept Stress Intensity Factor (k) becomes relevant. The stress intensity factor is used to have a knowledge of the stress level around a crap tip or a notch, This value is use in fatigue conditions to obtain information regarding the critical stress which lead to an uncon-trolled crack growth that culminates in fracture. k depends on the specimen geometry and is measured in Pa?m [36].
The theory behind the stress intensity factor can be explain understanding the behaviour of a specimen that is subdued to a specific load, the higher the load applied the higher the stress experimented by the specimen and vice versa, thereby if the load applied is too high then the k will be as well, leading the specimen to a very high stress level that it might not hold, the crack will propagates thought the specimen and break. In contrast there is a very low load that permits the specimen to endure this load infinite time always without experi-menting plastic deformation [39]. The American academy Paul Croce Paris postulate that the stress intensity factor works as a function of the stress and the crack length in the specimen [40]. The stress intensity factor is found to control the dynamic fatigue crack propagation as well as the static fracture and environmental assisted [40].
There are three modes of stress intensity presented in the figure 2.3. The first mode, KI, is the one that is going to be utilized in the tests procedure. It indicates a opening tensile where the crack surfaces separate.
2.2. Stress Intensity Factor k
2.2.1
Stress Intensity Factor Range
∆k
In cyclic loading the stress applied is going to be described as sinusoidal, thus the stress inten-sity factor is going to have a maximum and minimum value see equation 2.3. Therefore, the Stress Intensity Factor Range is the difference between the maximum stress intensity factor in a cycle and the minimum stress intensity factor in a cycle [3].
∆k=kmax´kmin (2.3)
Where:
kmax: the maximum stress intensity value in a cycle [Pa ?
m]. kmin: the minimum stress intensity value in a cycle [Pa
? m].
2.2.2
Critical Stress Intensity Factor k
IcThe value necessary to propagate an initial crack that present fracture toughness is defined as the Critical Stress Intensity Factor [37]. It can also be understood as the resistance to fracture due to crack growth under loading [35].
2.2.3
Effective Stress Intensity Factor
∆k
e f fOccasionally when an external load is been applied to the specimen during a cycle, the crack might not open, in case it is necessary to reach a certain optimum stress (Kop). From this optimum value to the maximum stress value (Kmax) the increase in the load will open the crack, see figure 2.4 [41]. This phenomenon is called crack closure and can occur due to plastic deformation, phase transformation during the crack propagation, fluids or corrosion in the crack surfaces [46]. Therefore, the Effective Stress Intensity Factor∆ke f f can define the range where the crack will open, its expression corresponds to the difference between the maximum stress intensity factor and the optimum stress intensity factor see equation 2.4.
∆ke f f =kmax´kop (2.4) Where:
∆ke f f : effective stress intensity factor [Pa ?
m].
kmax: the maximum stress intensity value in a cycle [Pa ?
m]. kop: the optimum stress intensity value in a cycle [Pa
? m].
Figure 2.4: Stresses involve in crack surface opening from Chapter 10 Deformation and Frac-ture Mechanics for Engineering.
2.3. Fatigue Crack Growth RatedNda
2.3
Fatigue Crack Growth Rate
dNdaThe rate at which a crack growth subdued to fatigue correspond to the differential curve of the crack length (da) and cycles (dN) values measure at a a certain crack point [25].
The fatigue crack propagation (FCP) rate is not constant during all the specimen fatigue life, see figure 2.5. It is predictable that in the first stage the rate is going to be slow, then it will have a constant variation, this second stage is known as the Paris law region. The last stage is characterized for having a rapid and unstable growth until fracture [34].
Figure 2.5: Crack growth stages from Chapter 10 Deformation and Fracture Mechanics for Engineering.
The Paris law is defined as the section of the curve of the crack propagation rate and the stress intensity range in logarithmic scale. In this region the FCP rate is going to the controlled by the stress intensity factor [34]. This last affirmation is accomplish during the second stage of the curve 2.5, see figure above. From this region the Paris relation can be obtained see equation 2.5. da dN =C(∆k) m (2.5) Where: da
dN : crack growth rate [m/cycles].
C and m : materials constants for crack propagation. ∆k : stress intensity factor range [Pa?m].
The stresses (residual, cyclic loads and thermal stresses) experimented by the component, or specimen in the case of the laboratory work, in overall with the material properties such as fatigue resistance properties will define the rate of growth of a crack.
2.3.1
Fatigue Crack Propagation rate relation with Stress and Crack length
for obtaining the crack propagation rate, the crack length data during the fatigue the test must be acquire, this can be done by the use of compliance or electric potential difference method. The fatigue researchers have found results from the test performed where it is been observed that for almost all the cases, the FCP rate is raise with increases in the crack length.2.3. Fatigue Crack Growth RatedNda
Another important relation is that the FCP rate have an important magnitude correlation with the stress applied as can be noted in the figure below 2.6 where a higher stress leads the component to endure less number of cycles (N) and a minor crack length (a) until the material breaks [42].
Figure 2.6: Stress effect on the fatigue crack propagation rate from Deformation and fracture mechanics of engineering chapter 10.
As is described in the chapter 10 of the book Deformation and fracture mechanics [43], during the 1960 decade several experiments were performed with the aim of understanding the behaviour of the fatigue crack propagation rate. The results show that the stress intensity factor controlled the FCP rate.
2.3.2
Fatigue Crack Growth Threshold
∆k
thFrom the concept of the fatigue crack growth rate it is important to define a threshold regime. According to the ASTM Standard E647 [3] the ∆kth correspond to the value at which the Fatigue crack growth rate approaches zero see figure 2.5. For almost all materials the value of∆kthis given at a fatigue crack growth rate of 10´10m/cycle.
2.3.3
Plane Strain condition
The fracture toughness test performed in this research follows the standard for plane strain fracture toughness of metallic materials (E399) [8]. In the introduction section 1 are presented the three stages for the fatigue fracture failure. As it is explained in the chapter 10 of deforma-tion and fracture mechanics [44], in the first stage the crack will have a 45° with the xy plane in reference to the loading direction, thereafter the direction will change to the loading direction allowing the crack to propagate. During the second stage, the plane of the crack propagation will depend on the stress level the specimen endures. In the section stress intensity factor 2.2, is exposed how the stress apply will influence the stress intensity the specimen experiment. Therefore, a small stress intensity factor range produce a small plastic deformed zone and vice-versa for a big∆K value. A plane strain condition is accomplish when the thickness of the specimen is large in comparison with the plastic zone size, this condition leads to a flat fracture. If the plastic zone size is larger compared to the specimen thickness a plane stress condition rules the fracture and an incline fracture is developed [44].
2.4. Importance of the cast iron microstructure
2.4
Importance of the cast iron microstructure
2.4.1
Classification of cast irons
In the wide field of the metal alloys this ones can be classified in two types, ferrous and non-ferrous. The ferrous alloys have many sub-groups classifications as can be shown in the figure 2.7. The steel and the cast iron differ in the amount of carbon within their structure, a cast iron is compound from a range of 2-4 % of carbon.
Figure 2.7: Classification of Metal Alloys from Concepts in Physical Metallurgy Cast irons can be categorized as white, gray, spheroidal and compact cast iron. This first presents the carbon as iron carbide plates (Fe3C with 6,67 % C) and a white fracture surface, meanwhile the gray iron (also named lamellar graphite iron) exhibit the carbon as graphite (which contains 95-99 % of carbon) in the shape of flakes with a gray fracture surface [28]. A nodular graphite iron - also named spheroidal graphite iron - presents the graphite in spheroidal morphology and lastly the compact graphite iron (CGI)- also known as vermicular cast iron - the graphite has a worm shape with round edges [29]. For a better comprehension of the graphite morphology in the specimens for this project, the vermicularity concept will be introduced. The vermicularity concern the percent of vermicular morphology observed in the specimen graphite [27], therefore a nodular cast iron is going to have a very low or no vermicularity, in contrast a gray cast iron will have a high vermicularity. A compact cast iron is expected to have a medium vermicularity morphology.
The fact that these ferrous alloys are manufactured by casting, permit to achieve different types of shapes for the final product which amplify the uses they might be given, even though they display a tendency to be highly brittle and get rusting, they also have a great wear and deformation resistance [28].
2.4.2
Influence of the cast iron microstructure in the mechanical properties
The properties of the cast irons are given by the chemical composition of them. Therefore, the alloying elements are a relevant parameter. The alloying element that is essential for cast irons is the silicon (Si) since it is the one that have the effect of maintain the carbon out as a second phase forming graphite, the amount of silicon can vary between 1-3% [28]. The mechanical properties are linked to the chemical composition and microstructure of the material. The aspects related to the microstructure of the material include the graphite morphology, which can be presented as nodular (low vermicularity), vermicular (medium vermicularity) or flake (high vermicularity). The impact of flake morphology displays an important aspect, they act as a stress concentration areas, turning the flakes in an inconvenient graphite morphology. The lamellar graphite iron exhibits a low resistance at tension due to the graphite flakes (ten-sile strength 110.9 – 134.7 [MPa]) [27], at compression the resistance can reach higher values, near the double as in tension. A nodular morphology will allow the material to distribute2.4. Importance of the cast iron microstructure
the stress raising up the resistance (tensile strength 389.6 – 532.5 [MPa]) [27]. The compact graphite iron presents an intermediate vermicularity, is not as low as the nodular cast iron or high as the gray cast iron (tensile strength 257.6 – 367.8 [MPa]) [27].
If a comparison is made between lamellar graphite iron and compacted graphite iron, this last one has higher fracture toughness, (kIc47.8 – 56.3 [Pa
?
m] for CGI and for LGI KIC 14-22 [Pa?m])[27]. Furthermore, the properties of the CGI are less dependent than the LGI on the specimen geometry. On the other side some benefits of CGI over the nodular graphite iron are that it has lower thermal expansion coefficient, lower level of stress thermally induced which gives the CGI superior thermoshock resistance along with improve damping capacity, pouring properties as shrinkage, mold-filling and fluidity [29].
The graphite is incorporated in a matrix which can be ferritic or pearlitic. The ferrite is a phase characterized for being soft, ductile and for have a low carbon solubility, around 0.008 % at room temperature and 0.022 % at 727° C as it was explained in the section 1.4. The pearlitic phase is compound of layers of ferrite and cementite, therefore its properties are in the middle of the brittle, hard with high carbon (6.67 % C) cementite and the ductile/soft ferrite. The effect of the microstructure is a relevant analysis, the amount of graphite in it can signify more possibilities of the crack to propagate through the brittle graphite[19]. Further-more a ferritic matrix can help to dissipate the stress the specimen experiment being under fatigue conditions since is more ductile or cannot be resistance enough for the conditions and lead the specimen to failure. The cementite layers present in the pearlitic matrix can have an important contribute to the iron resistance and hardness, increasing the fatigue life of the component [19].
3
Method
3.1
Relevant terminology
1. Crack size (a)Is the lineal length in [mm] of the crack size of the specimen, both surfaces should be measured. The crack size (a) is not propagated under normal conditions unlike the critical crack size which one follow an unstable crack growth.
2. Force range (∆P)
Represent the difference between the maximum force and minimum applied in a cycle, considering the tensile forces as positive and the compressive forces as negatives. Is measure in [N] [3].
∆P=Pmax´Pmin (3.1)
Where:
Pmax: highest force applied in a cycle [3]. Pmin: lowest force applied in a cycle [3]. 3. Stress Ratio (R)
Is the ratio between the minimum force and the maximum force [3]. R= Pmin
Pmax (3.2)
4. Frequency (f )
Is the amount of cycles per second is measured in hertz [Hz] [3]. 5. Strength ratio (Rxx)
It has the purpose of relating the maximum load applied to keep the yield strength and the initial dimensions of the specimen. The subscripts xx indicates the specimen config-uration [8]. For the specimen configconfig-uration for the fatigue toughness test is calculated as the equation below 3.3.
Rsb=
6 ¨ Pmax ¨W B(W ´ a)2 ¨ σ
YS
3.2. Fatigue crack propagation rate test
Where:
Pmax: maximum load that the specimen could endure [8]. W : width of the specimen [8].
B : thickness of the specimen [8]. a : crack length [8].
σYS: yield strength in tension at a 0.2 % offset [8]
3.2
Fatigue crack propagation rate test
For performing the fatigue crack propagation rate test the standard for the test method is followed.
3.2.1
Standard Test Method for Measurement of fatigue crack growth rates, E647
[3].
This standard contains the detailed information from the preparation to procedure, calcula-tions and analysis of the data. A fatigue crack propagation rate test is performed under the linear elasticity behaviour of the material for the applied force. This method it applies to pre-cracked notched specimens. The results are expressed indNda vs∆k this manner of presenting the outcomes is independent of the planar geometry.
The tests are going to be develop in an inert environment, in other words, no high tem-peratures or corrosion and oxidation risk. Due to this condition the fatigue crack propagation is expected to be characterized as a function of the stress intensity factor range (∆k) and the stress ratio (R) [4].
The procedure followed was the one for a K-increasing test, this type of test is suitable for rates over 10´8m/cycle, for this rates the variability is less. For rates under 10´8m/cycle the FCP rate is sensitive to small variations in the stress intensity range. It is also recommended to performed the test at constant force amplitude for rates above 10´8m/cycle to avoid transient rates product of changes in the Pminor the stress ratio [5].
In the section 9 of the standard E647 calculation and interpretation of results [6] it is advised to calculate a crack curvature correction factor(CCC) to consider the difference between the physical and the software crack measures, in the subsection 4.1.4 is possible to see the out-comes related to the CCC.
The specimen dimensions can be observed in the figure 3.1, where W correspond to the width of the specimen. The dimensions fulfill the standard annex A1 for a C(T) configuration [7].
3.2. Fatigue crack propagation rate test
3.2.2
Fatigue crack propagation rate test features
This test is going to be develop with the aim of obtaining data that represent the real fa-tigue crack propagation behaviour of the materials. There are seven different compositions of specimens with three samples per composition, the three specimens by composition were numbered N°1, N°2 and N°3. The fatigue crack propagation test was performed in duplicate to consider the variability of the method. Therefore, the specimens N°1 and N°2 were tested. The specimens compositions are display in the table 3.1, the ones named just as ferritic and pearlitic signify the reference composition.
3.2. Fatigue crack propagation rate test
Specimen denotation Composition
1MM N°1 Pearlitic 4MM N°1 Pearlitic 0.25 % Mo 10MM N°1 Pearlitic 0.5 % Ni 16MM N°2 Pearlitic 1 % Ni 20MM N°1 Ferritic 26MM N°1 Ferritic 0.15 % Mo 36MM N°1 Ferritic 0.25 % Mo
Table 3.1: Table with specimens denotation and chemical composition.
The MM denotation in the first column of the table 3.1 signifies medium solidification rate and medium cooling rate.
3.2.3
Fatigue crack propagation rate data acquisition equipment
A Mini Bionix 858 machine is utilized for the test which works with a servo hydraulic sys-tems, which means that a servo valve will receive the hydraulic fluid that comes from a hy-draulic pressurized cylinder [30], see figure 3.2a. The machine has an axial and torsional limit of 25 [kN][13]. The data acquisition was obtained through the use of a clip-on extensometer, located in the sharp edges of the C(T) specimens (see figure 4.8a). The software named Doli was utilized, which it has a special interface for performing fatigue tests. The crack length were measured (apart from the software crack length data) previously to begin the test with a vernier caliper and compared with the measure obtained from the photos took with the DIC (Digital image correlation) camera equipment, which works with a camera (see figure 3.2b) connected to a computer with a software named TEMA. The software allows to take measures by using the tools offered which are be explained and were use for the fracture toughness test, see subsection 3.3.3. The DIC equipment were also utilized to take measures in between the test.
3.2.4
Fatigue crack propagation rate data acquisition method
The extensometer measures the displacement the specimen experiments. The software Doli works the data from the extensometer to calculate the crack length utilizing the compliance method, as is indicated in the annex corresponding to the C(T) specimen [7]. The equation 3.4 is the expression to calculate the crack length by the compliance method.
a W =c0 + c1¨u + c2¨u 2 + c 3¨u3 + c4¨u4 + c5¨u5 (3.4) u= 1 aB ¨ E ¨ υ P+1 (3.5) Where:
C: Constants that depend on the extensometer position. a: crack length.
W: specimen width. B: specimen thickness. E: elastic modulus.
υ: crack opening displacement.
3.2. Fatigue crack propagation rate test
(a) Mini Bionix 858 machine utilized for the fa-tigue crack propagation rate tests.
(b) Camera used to take initial measures with the DIC equipment.
Figure 3.2: Fatigue crack propagation tests equipment.
It is important to remark that the dimension- width, thickness and notch length (initial crack length)- of each specimen were measure with a vernier caliper before prior to be tested For calculating the stress intensity (K) the software uses the expression 3.6.
K= P B ¨ W0.5¨ f(d) (3.6) f a W = f(d) = 2+d (1 ´ d)1.5 ¨(0.886 + 4.64 ¨ d ´ 13.32 ¨ d 2 + 14.72 ¨ d3 ´5.6 ¨ d4) (3.7) Where:
K: stress intensity factor. P: load.
The stress intensity factor range (∆K) is calculated by multiplying the K for one minus the stress ratio (R).
∆K=K ¨(1 ´ R) (3.8)
• Fatigue crack propagation rate test parameters
Parameter Value Units Maximum Force 4000 [N]
Minimum Force 200 [N] Stress ratio 0.05 [-]
Frequency 10 [Hz] Elastic Modulus 150 [GPa]
3.3. Fracture toughness test
In the table 3.2 are the parameters established for the fatigue crack propagation rate test, this parameters were invariables for all the tests performed. The tests on the specimens N°2 were conducted under the same procedure with the parameters in the table 3.2. Pauses at the crack length of 12.37 [mm] and 15.5[mm] were done to take measures of the crack length by utilizing the DIC software TEMA, although this measures were not taking in consideration since the crack length was not clear to visualize. The test termination for the specimens N°2 were at 16[mm]. The differences among the tests N°1 and N°2 were the test termination which varied for each composition between 15[mm] and 17[mm]. Moreover the tests N°1 were stopped at different crack length to measures.
3.2.5
Leica Optical Microscope Crack Photos
• 1MMFigure 3.3: Pearlitic 0.5% Nickel 10MM Crack, tests N°2 5X.
Afterwards performing all the test for fatigue crack propagation rate, the specimens N°2 were polish with the silicon carbon grinding papers # 220, # 500, # 1200, # 4000. The figure 3.3 is an image took with the Leica optical microscope with the aim of mea-suring the final crack length of the pearlitic specimen which 0.5% Nickel. This images were taking for all the specimens N°2, which were previously polished, to compare the final crack length the software Doli indicates and the measured crack length with the microscope software measurement tools to make a posterior analysis with the values as is display in the table 4.2 in the subsection 4.1.4.
3.2.6
Data interpretation
For analysing the data the software MATLAB was utilized.The Doli software does not allow to continue the test ones is stopped, thus a "new test" is needed to be done. With MATLAB the the several parts (between 4-5) of the same specimen tested could be merged. As a result the fatigue crack propagation rate charts presented in the subsection 4.1.1 are plotted in a logarithmic scale.
3.3
Fracture toughness test
3.3.1
Standard Test Method for Plane-strain Fracture Toughness of Metallic
Materials, E399 [8]
The standard E 399 was followed to performed the fracture toughness test. In the subsec-tion 2.3.3the definisubsec-tion of plane strain condisubsec-tion is indicated, for this test this condisubsec-tion is require, this signify that a crack tip resistance should be under plain strain condition. The plane stress condition specify that the thickness of the specimen must be large in comparison with the plastic deformed zone. In addition the standard is valid for specimens with a thick-ness of 1.6 [mm] as minimum. The specimens must be pre-cracked following the annex A2 [9]. The k value characterizes the material resistance to fracture in an neutral environment
3.3. Fracture toughness test
kIc value can be applied for the designing components, the user must be aware of the dif-ference between the service conditions and the test ones. A component expose to aggressive environments will be fractured at a relevant lower stress[10]. The method presented can be utilized for the research, develop and improvement of new or existing materials for their ser-vice function, it is also possible to evaluate the manufacturing impact such as welding and metallurgic variables like composition or heat treatments, all within the aim of selecting the suitable material for the component service conditions [10].
3.3.2
Fracture toughness test features
For this test there were 3 specimens per composition, the specimens were numbered N°1, N°2 and N°3, but due to errors in the pre-crack process or visible surface imperfections is possible to find in the table with the results 4.3 for the first round of tests specimen N°2. The compositions are the same as the ones in the table 3.1. Two round of tests were performed.
3.3.3
Fracture toughness data acquisition equipment
For performing the pre-crack and the fracture toughness test the Alwetron CTC 50 machine was utilized, which works with a electromechanical system, see figure 3.4. The software named CycliEdc connected to the Alwetron machine allows to create methods for kIc deter-mination and for cyclic loads, this last one was used to perform the pre-crack.
Due to the fact that the notch was to narrow and it did not content the sharp edges to clip-on an extensometer. The crack length for the pre-crack process was measured by using a portable microscope camera. For the fracture toughness test is mandatory to use any type of extensometer, due to this the DIC (Digital image correlation) camera was used to record the test and to analysed the data through the TEMA software. DIC offers a virtual extensometer tool, here two points are selected and tracked during the test record. The DIC allows to get a full strain field of the component. The data obtained by the extensometer added can be plotted versus the time, where from this chart the timetable data is used and processed in MATLAB. Must be remark that using a simple unit conversion from a known dimension of the specimen- in this case the width (W)- and the equivalence in pixels were done, to finally obtain the displacement in [mm].
For get the kIcvalue is needed to plot the force vs the displacement. The force data was acquire from the CycliEdc software while the displacement from the TEMA virtual exten-someter. To merged both data the software MATLAB was utilized.
3.3.4
Pre-crack method
The pre-crack procedure was done following the specifications in the annex A2[9], some of these conditions are the next ones.
1. The pre-crack must be performed under cyclic loading for a number of cycles within the range of 104and 106. The amount of cycles for the pre-crack depends on the specimens dimensions (notch, width and thickness) and stress intensity level at which this process is performed.
2. The crack length, considering the started notch plus the pre-crack must be in the range of 0.45 times the width (W) and 0.55 times the width.
3. The ratio of maximum stress intensity of the fatigue cycle over the young’s modulus of the materialKmax
E
shall not exceed 0.00032 [m0.5].
4. It is recommended that the load chosen for the initial part does not exceed 80 % of the kIc estimated. The stress intensity in the terminal stage, when 97.5% of the final crack length is reached, should not exceed 60 % of the kIc.
3.3. Fracture toughness test
Figure 3.4: Alwetron machine utilized for performing the pre-crack and fatigue toughness test.
5. The frequency advisable to be under 100[Hz].
All these numbering were taking carefully in account since a wrong pre-crack can have pro-duce a severe impact on the kIcresults, due to the high stress the material could experiment.
The kIc estimated value that is indicated in the item number 4 was the maximum stress intensity observed from the fatigue crack propagation test, these values were within the range of 26 to 30 [MPa?m]. For calculate the load to apply for reaching the 80 % of the estimated kIcthe expression in the annex for the specimen configuration specification was used [11], see equation 3.9. P= K ¨ B ¨ W 1.5 f Wa ¨S (3.9) f a W = f(d) =3 ¨ d 0.5¨(1.99 ´ d ¨(1 ´ d)¨(2.15 ´ 3.93 ¨ d+2.7 ¨ d2)) 2 ¨(1+2 ¨ d)¨(1 ´ d)1.5 (3.10) Where: P: load[KN].
K: estimated stress intensity [MPa?m]. B: specimen thickness in [cm].
W: specimen width in [cm]. a: specimen crack length in [cm]. s: span between the supports in [cm].
The units for this calculation have a great impact. Therefore, if the international system is used the units should be the ones presented above.
3.3. Fracture toughness test
and displacement data is plotted is important to verify that the slope for the linear part of the curve is between 0.7 to 1.5, this condition was fulfill for all the specimen tested.
The next step is to calculate the conditional fracture toughness KQ, for getting to this result first a secant line (yellow line in figure 3.5)must be drawn with a 0.95 % of the slope of the tangent line (red line in figure 3.5) to the linear part of the force vs displacement curve. The conditional load (PQ)- use afterward to calculate the conditional K- can be defined in two manners, depending on the curve behaviour, see figure 3.6. If the points that precede the intersection point of the yellow line with the curve are lower than the load at the intersection point than PQis the intersection point (case for the curve 1 and 3 in the figure 3.6), this case is the one for the tested specimens, as can be seen in the figure 3.5. The other situation is when the there it is a higher point in the points previous to the intersection and in this case, as is shown in the second curve in the figure 3.6, here the PQ is the maximum load of the points previous to the interception.
Figure 3.5: Chart of load [N] vs displacement [mm] for the specimen 10MM Pearlitic 0.5% Nickel.
Figure 3.6: Types of load vs displacement curves. Afterwards, the ratio Pmax
PQ
, Pmax corresponds to the maximum load the specimen en-dured, if the value is under 1.10 the following step is to calculate KQas indicated in the annex A3 (see equation 3.12), otherwise the could be invalid meaning that PQmight not have any relation with kIcand the strength ratio must be calculated as indicated in the expression 3.3. The values used for the yield strength were the ones from the table 2 in the "The Importance of the Microstructure for the Mechanical Properties of Compact Graphite Iron" [26]. The expression 3.11 must be calculated, if the value is less than the thickness and the initial crack length, then
3.3. Fracture toughness test
KQcorresponds to the kIcvalue.
2.5 ¨ kIc σys (3.11) KQ= PQ¨S¨ B ¨ W1.5 ¨f a W (3.12) The terms in the equation 3.12 are the same ones as in the equation used for the pre-crack 3.9. For the realization of the interpretation of the data including all the steps described above were done by a code made in MATLAB.
4
Results and Discussion
4.1
Fatigue Crack Propagation Results
4.1.1
Fatigue Crack Propagation Rate and Stress Intensity Factor Range Charts
dadN
vs
∆K
For matter of convenience the pearlitic and ferritic specimens with reference composition will be named as pearlitic RC and ferritic RC respectively.
• Ferritic and pearlitic specimens
Figure 4.1: Fatigue crack propagation chart of ferritic and pearlitic specimens, tests N°1 and N°2.
In the figure 4.1 is possible to observe that the pearlitic specimens display a lower fa-tigue crack propagation rate than the ferritic specimens. The behaviour show in the curves indicates that for a pearlitic material at a given stress intensity factor the present
4.1. Fatigue Crack Propagation Results
crack in the specimen will not propagates as fast as the ferritic specimens displaying a higher fatigue crack propagation resistance, this comportment is beneficial for the head cylinder application although the difference between the curves is not large a trend can be noted.
• Pearlitic specimens with molybdenum
Figure 4.2: Fatigue crack propagation chart of pearlitic specimens with molybdenum in its composition and pearlitic reference composition specimen, tests N°2 and N°1.
From the chart in the figure 4.2 the specimen with molybdenum in a pearlitic matrix (green line), shows a lower fatigue crack propagation rate than the pearlitic RC (red line) specimen by positioning below the pearlitic reference composition curve during all the test. From this chart can be observe that the effect of the molybdenum increase the fatigue crack propagation resistance, especially in the stress intensity range from 15.4 to 17.8 [MPa?m]. The final stage of the test the effect of the molybdenum is not very significant since in this region the crack propagates uncontrolled.
• Ferritic specimens with molybdenum
Figure 4.3: Fatigue crack propagation chart of ferritic specimens with molybdenum in its composition and ferritic reference composition specimen, tests N°2 N°1.
From the figure 4.3 can be noted that the relevance of the amount of molybdenum in specimens with a ferritic matrix do not seems to have a major impact. Both curves
(fer-4.1. Fatigue Crack Propagation Results
while the 0.25% Mo specimen show a deviation from the other curves displaying a slower FCP rate, although this last behaviour is not enough to considerate a positive affect of the molybdenum in a ferritic matrix.
• Nickel specimens
Figure 4.4: Fatigue crack propagation chart of specimens with Nickel in its composition and pearlitic reference composition specimen, tests N°2 and N°1.
The specimens in the figure 4.4 have a pearlitic matrix. . It is possible to observe for the test N°2, from a stress intensity factor of 15.2 [MPa?m] until the end of the test the curve for the specimen with a larger amount of nickel, 1% Ni, present a tendency to be below the other two curves. This can be interpreted as bigger fatigue crack propaga-tion resistance (lower fatigue crack propagapropaga-tion rate). For the curve with 0.5% Ni the behaviour differentiate from the reference composition curve for the short range of 15.8 - 17-8 [MPa?m]. After this range the 0.5% Ni curve pass to locate above the reference composition curve meaning that at this stage the crack grows faster than the other two curves. Finally the 0.5% Ni have nearly identical performance to the reference composi-tion specimen, therefore 0.5 % Ni does not seems to improve considerably the specimen performance, 1% Ni shows that it has a positive impact by reducing the fatigue crack propagation rate. For the test N°1 the curves intersect each other so no clear trends can be analysed for this test. This could be related to the difference in the test performance stops.
4.1. Fatigue Crack Propagation Results
• Pearlitic specimens
Figure 4.5: Fatigue crack propagation chart of pearlitic specimens N°2 and N°1.
From the figure 4.5, in the range for stress intensity factor of 15.2 to 17.8 [MPa?m], is observed that the even though the curves do not present big different between them, a soft trend can be remarked. For the specimens that contain 0.25 % Mo and 1 % Ni the curves display the tendency to located under the rest of the curves, showing a lower fatigue crack propagation rate. This comportment is a consequence of the alloying elements which increase the resistance of the material. Especially the specimens with 0.25 % Mo presents visually a highly fatigue crack propagation resistance.
In the final stage of the fatigue crack propagation, which is represented in the chart from 18 [MPa?m] until the end of the test curves, is shown how the crack growths in a unstable manner, simultaneously the slopes of the curves are increased, representing a rapid crack propagation until the specimens experience failure.
4.1. Fatigue Crack Propagation Results
4.1.2
Fatigue crack propagation rate (
dNda) relation with the graphite average
length (<a>) and hardness
For performing this analysis the data obtained from all the tests (N°1 and N°2) where consid-ered. The values of fatigue crack propagation rate were obtained from a linear approximation to the curvesdNda vs∆k for a fixed stress intensity factor of 18 [MPa?m], since this stress inten-sity math the higher FCP rate for the second region of the fatigue crack propagation curves. It must be remarked that the behaviour of curves it is similar, hereby this method could be done.
The average length data was acquired from the table 2.8 in the report "The casting proce-dure’s impact on the microstructure" [2] and the hardness from the table 3 of the report "The Im-portance of the Microstructure for the Mechanical Properties of Compact Graphite Iron" [26]. These reports and this master thesis worked in collaboration to complete the material characteriza-tion of Scania vermicular cast irons materials for cylinder heads.
• Fatigue crack growth rate vs average length
Figure 4.6: Chart of fatigue crack growth rate vs average length for all the specimens tested, N°1 and N°2.
The molybdenum in quantities of 0.3-1 % in pearlitic cast iron [24]is use as a alloying element with the aim of increasing the hardness and resistance of the material. This improve-ments can be done because of the decrease in the temperature for the pearlite transformation [48]. The molybdenum have a strong effect of refinement over the graphite flakes [22]; it also reduce the inter layers space of the ferrite and cementite in the pearlitic matrix which it expresses as a strength increase [24].
From this figure is possible to see that the molybdenum have a significant impact on the pearlitic matrix, as a result the specimen with a pearlitic matrix and 0.25 % Mo have the larger graphite average length, this mean that the graphite has a fine flake morphology. The graphite contribute with the crack propagation, since is a brittle zone which acts as a stress concentration area where an existing crack can propagates across it. Therefore, if there is graphite flakes present in the specimen microstructure- which is possible considering that correspond to a compact graphite cast iron- is beneficial that this flakes are fine, int his man-ner they do not have a great impact as stress concentration areas [23]. For the ferritic matrix the molybdenum does not have an important impact as can be see in the figure 4.6 the ferritic specimens have the lower graphite average length this imply that the graphite refinement effect of the molybdenum is not as significant as it is for the pearlitic specimens. This differ-ence could be due to the small carbon solubility in the ferrite ( 0.008 % C [19]) to produce a refinement of the graphite, thus consequently have a larger graphite average length. The diffusion phenomenon that occurs is that the carbon going out of the graphite diffuse to the
4.1. Fatigue Crack Propagation Results
matrix, although for a ferritic matrix this one does not accept this carbon flow, therefore the graphite cannot be relevantly refined.
The nickel is added to the composition of cast irons for the graphitizing effect, in other words it promotes the graphite formation, also it have a refinement effect on the graphite even though is not as strong as the molybdenum [22]. The pearlite refinement improve the strength and hardness of the material. Given that the nickel improve the above mentioned mechanical properties by its graphitizing and pearlite refinement characteristics, from the figure 4.6 can be said that the larger quantity of nickel (1% Ni) the graphite average value obtained is non representative of the effect of nickel in the microstruture explained above, therefore this data is going to be considered for this parameter analysis. The specimen with 0.5% Ni the graphite refinement is evidently less than with the molybdenum for the pearlitic specimens.
If the values of the fatigue crack growth rate are observed is possible to confirm the analy-sis made from the fatigue crack propagation rate vs∆K curves, where the pearlitic specimens expose a slower fatigue crack propagation rate. The pearlitic 1% Ni display a higher and more stable resistance in the FCP curves than the 0.25% Mo, this remarks the stronger nickel alloy-ing effect of improvalloy-ing the fatigue crack propagation resistance meanwhile the molybdenum present a relevant effect on graphite refinement for the pearlitic specimens.
• Fatigue crack growth rate vs hardness
Figure 4.7: Chart of fatigue crack growth rate vs hardness for all the specimens tested, N°1 and N°2.
As can be seeing in the figure 4.7 a trend can be visualize, where for a higher fatigue crack propagation rate the hardness values are smaller, this behaviour concurrently is displayed for the ferritic specimens. The trend seen can be understood due to the phases present in the matrix. The ferritic present a low carbon solubility therefore, is a soft phase since the carbon -which contributes to the hardness by disturbing the crystal structure by staying in the middle of the edge or in the center of the faces of the unit cell [16]- is very low 0.008 % C [19]. Therefore, the characteristics explained above will lead the ferritic matrix to ex-hibit a lower resistance to the cyclic loads, as a result an existing crack will propagates faster. For a pearlitic matrix the pearlite constituents are ferrite and cementite layers; were the ce-mentite layers expose a high hardness due to the amount of carbon ( 6.67 % C) and brittle characteristics [19]. Normally the pearlite contains 86.5 % of ferrite and 13.5 % of cementite consequently the properties will not be far from the ferrite ones, thought a higher resistance as a consequence of the increase in the hardness can be observed and expressed as a lower
