Çift fazlı çeliklerin deformasyon davranışı

Abstract

ÖZET Bu çalışmada, `Ferrit + Martensit` mikroyapısı kazandırmak ama cıyla, A1-A3 sıcaklıkları arasındaki beş ayrı sıcaklıktan su verilen a- laşımsız düşük karbonlu (%0.074 C) çelik saçın oda sıcaklığındaki meka nik özellikleri ve deformasyon davranışları incelenmiştir. Çift faz mikroyapısına (ferrit + martensit) sahip olan ve su verme sıcaklığı arttıkça, martensit içeriklerinin de arttığı tesbit e- dilen alaşımsız %0.074 C'lu çeliklerin, mekanik özelliklerinin belirlen mesi amacıyla yapılan sertlik ve çekme deneylerinden, mikroyapıda bulu nan martensit miktarının artmasıyla, sertlik ve mukavemetin arttığı, sü- nekliğin ise azaldığı saptanmıştır. Bu ısıl işlem koşullarında optimum mekanik özellikler, mikroyapısında %17'den daha az oranlarda martensit içeren çift bazlı çeliklerden elde edilmiştir. Alaşımsız az karbonlu çelikten üretilen çift fazlı çelikle rin deformasyon davranışlarının belirlenmesi amacıyla, plastik defor masyon bölgesinde yapılan gerilme gevşemesi, deformasyon hızını değiş tirme ve yükleme-boşaltma deneylerinden, bu çeliklerin deformasyon dav ranışlarını kontrol eden iki bölgenin mevcut olduğu tesbit edilmiştir. Gerilme gevşemesi ve deformasyon hızını değiştirme deneyleri yardımıy la, bu bölgelerdeki etkin dislokasyon mekanizmalarının, dislokasyon ke sişmesi (I. Bölge) ve Peierls-Nobarro gerilme engellerinin aşılması (II. Bölge) olduğu saptanmıştır. Yükleme-boşaltma deneylerinden ise, plastik def ormasyonun, serbest dislokasyonların bağlanması sonucu alt yapıda mevcut dislokasyon engellerinin kabaşlamasma neden olduğu ve söz konusu engeller arasındaki mesafenin artan deformasyonla önceleri azalıp, sonra sabit kaldığı tesbit edilmiştir. Çekme deneyi esnasında, elastik def ormasyondan plastik defor- masyona geçerken sürekli akma gösteren, alaşımsız çift fazlı çelikle rin, gerilme gevşemesi deneylerinden sonra tekrar yüklenmesi, defor masyon hızını değiştirme deneylerinde deformasyon hızının aniden artı rılması ve yükleme-boşaltma deneylerinde yükün boşaltılıp derhal yeni den uygulanması sonucu, gerilme-birim şekil değiştirme eğrilerinde sü reksiz akma olayının ortaya çıktığı gözlenmiştir. Bu olayı açıklamak amacıyla yapılan incelemelerde, süreksiz akmanın, gerilme gevşemesi deneylerinde, gerilme gevşemesi esnasında serbest dislokasyon yoğunlu ğunun azalmasından, deformasyon, hızını değiştirme deneylerinde, defor masyon hızı artırıldığında serbest dislokasyon yoğunluğunun da artma sından ve yükleme-boşaltma deneylerinde, yükün boşaltılması sırasında serbest dislokasyon yoğunluğunun azalmasından kaynaklandığı sonuçları na varılmıştır. Ayrıca gerilme gevşemesi ve deformasyon hızını değiş tirme deneyleriyle, çift fazlı çeliklerin oda sıcaklığında, gerilme altında deformasyon yaşlanmasına uğradıkları da saptanmıştır. iv

DEFORMATION BEHAVIOR OF DUAL-PHASE STEELS SUMMARY Dual-phase steels are a new class of low carbon high strength sheet steels characterized by a micros tructure consisting of a disper sion of about 20 percent of hard martens ite particles in a soft ducti le ferrite matrix. The term `dual-phase` refers to the presence of essentially two phases, ferrite and martensite, although small amounts of bainite, pearlite or retained austenite may also be present in the micros tructure. These steels have a number of unique properties, which include; continuous yielding, i.e. no sharp yield point, and a relati vely low yield-to-tensile stress ratio together with a rapid rate of work hardening and high elongations which gives excellent formability compared to traditional high strength steels at the same strength level. As the automotive industry shows a strong tendency to use of high strength sheet steels in order to reduce automotive weight and thus improve its fuel economy, dual-phase steels found weight reduction applications in this area. The simplest way of achieving dual-phase micros tructure from low carbon (the overall carbon content required is 0`1 percent or lower) f erritic-pearlitic steels (plain carbon or High Strength Low Alloy Steels) is to use intercritical annealing, in which the steel is heated to the `ferrite + austenite` field between A^ and A3 and held for several minutes to form small regions of austenite. Depending on the hardenability of austenite grains and/or cooling rate, marten- site forms in the microstructure on cooling. The transformation of the austenite phase into martensite in dual-phase steels occurs at low tempuratures, so that, the ferrite must plastically deform to ac comodate the volume expension (^2 to ^4 percent) arising from austeni te to martensite transformation. As a result, both high mobile dislo cation density and residual stresses are generated in the ferrite grains. The continuous yielding and low yield stress behaviour of dual-phase steels is a consequence of this mobile dislocations and residual stresses surrounding the martensite particles. The excellent combination of strength and ductility displayed by dual-phase steels results from its composite microstructure which is a mixture of ferrite and martensite. To obtain, appropriate micro- structure; cooling rate and temperature of intercritical annealing are critical factors. Depending on cooling rate, microstructure might be a lot complex, due to the formation of retained austenite, bainite, pearlite and carbides besides martensite. The temperature of intercritical annealing is of great importance, as it will deter mine the amount of martensite in the microstructure. The increase in martensite volume fraction increases strength but decreases ductility of dual-phase steels. Optimum mechanical properties can be obtained if the martensite content is less than 20 percent. The features of ferrite affects the properties of these steels, as well. For a givenmartensite volume fraction, optimum strength/ductility combination can be obtained if ferrite grains are small, polygonal and clean. Another microstructural constituent which is believed to improve ductility of dual-phase steels is retained austenite. A mild rate of cooling after intercritical annealing causes to increase the amount of retained austenite which may be as high as 10 percent in some dual phase steels. During tensile deformation, most of the retained austenite transforms to martensite with a few percent of plastic strain and it is claimed that this transformation improves the uniform ductility. The automobile parts produced from steel sheets, such as outher body panels, require high resistance to denting besides excellent for- mability. The denting resistance is a function of sheet thickness and the yield strength of the formed part. In dual-phase steels, high rate of work hardening results rapid increase in yield strength after a few percent plastic deformation,, On the other hand, it is possible to increase the yield strength of prestrained dual-phase steels (espe cially water quenched and tempered ones) by baking at low temperatures carried out after painting operation, due to strain aging. The increase in yield strength after prestraining and aging at low tempera tures (^200°C) is primarily related to the total carbon content of the ferrite, but interaction of carbon with microalloying elements may also be important. The decrease in martensite volume fraction and/or ferrite grain size accelerates the rate of strain aging. Because of their superior mechanical properties than those of similar steels with f errite-pearlite micros true tur e, work hardening behaviour of dual-phase steels became very attractive» In studies made on the work hardening behaviour of these steels the stress-strain curves separated into three regions according to Crussuard-Jauol analy sis and the changes in the substructure at each region are examined with various microscopes. In the first region of deformation, ferrite matrix homogenously deforms by the movement of mobile dislocations surrounding the martensite particles and rapid rate of work hardening is present because of the elemination of residual stresses introduced by martensitic transformation. Second region, covers a region on stress-strain curve where work hardening rate of ferrite is reduced as the plastic flow of the ferrite is constrained by the hard undefor- ming martensite particles and by the effect of `transformation of retained austenite to martensite. In this region, ferrite matrix, builts up a dislocation density higher than that for pure ferrite phase, and, generally cell structure begins to form near martensite particles. Finally, in the third region well developed cell structure forms in the ferrite matrix but the cell size is smaller near marten site than away from martensite. Further deformation does not change the sizes of the cells and causes yielding of the martensite particles. The plastic deformation behaviour of metals can also be inter- prated with mechanical tests which are called `indirect methods`in this study. The two common indirect methods are; stress relaxation and strain rate change tests. In both of these tests, Taylor-Orowan equation is utilized fundementally and deformation parameters are vimeasured. Another indirect method is loading-uploading test which is not as popular as stress relaxation and strain rate change tests and based on different concept, but can be successively used in defiermitiinp the effect of def°rmati-on. on substructures. These indirect methods are briefly described below. I. Stress Relaxation Test: Stress relaxation is a time depen dent decrease in stress in a body, which is constrained to a certain fixed deformation at a constant tempurature. The decrease in stress is the result of dislocation moving, to overcome localized barriers,, The major assumption necessary to analyse stress relaxation, which have been widely used to obtain fundemental parameters of microdeformation modes in metals, such as internal and effective stress, activation volume and dislocation velocity stress exponent, is that the mobile dislocation density (pm) is unchanged during the test» But it is sho wed that, in some metals this is not the case and mobile dislocation density changes with time, during stress relaxation. If relaxed stress- logarithmic relaxation time graph plotted, to determine the value of activation volume, is linear with a single slop, one can claim that, density of mobile dislocations is constant. The simplest method that takes into account the change in mobile dislocation density with rela xation time, is the determination of sequential relaxation curves from the referance stress for a fixed relaxation time. The decreas in the amount of relaxed stress with relaxation cycle indicates that, mobile dislocation density is decreasing during stress relaxation» The decrease in density of mobile dislocations might cause changes in the macroscopic flow stress of the metal in a way of discontinuous yielding (which is known as transient yielding) by reloading after stress rela xation. II. Strain Rate Change Test : The strain rate applied to a specimen has an important influence on the flow stress of metals. During tensile or compression tests, performed at a constant tempera ture, it is possible to determine the deformation parameters such as strain rate sensitivity exponent, internal and effective stress and activation volume, by changing the strain rate suddenly (strain rate change test) and measuring instantaneous change in the stress. At low tempuratures, a change in the density of mobile dislocation during strain rate change test might be very possibly occur. One might expect to observe transient yield point due to an increase in the dislocation density when the strain rate is increased. However, there are some other explanations in the litarature on this type of yielding. III. Loading-Unloading Test : In the mathematical treatment of elastisity, certain assumptions are made, one of which is that the state of the system has time to follow the load variation. However, even in the region below the propartional limit, metals are not per fectly elastic, since this would require coincidence of the curves for loading and unloading. The important point is that the direct relationship between stress and strain is disturbed and a hysterisis loop occurs on the stress-strain diagram. Loading-unloading cycle performed at the plastic deformation region also modifies the stress- strain diagram. That is, if the tensile specimen is stressed beyond viithe macroscopic yield stress value, then unloaded and reloaded without appreciable delay (this kind of interrupted tensile test is named as `loading-unloading test` in this study), strain enegry is dissipated in the f.öım.of heat produced by internal friction. The loss of strain energy causes formation of a hysterisis loop in the plastic deformation region of the stress-strain curve. The area of the hysterisis loop is a measure of the internal friction. The important source of internal friction is the reverse motion and/or reverse curvature of dislocation during unloading. By relo ading after unloading, in the course of plastic deformation, a transient yield point has been occured an the flow stress of some metals. The majority of mechanisms about this kind of yield points are based on the immobilization of mobile dislocations during unlo ading, but some researchers proposed different models. The purpose of the present investigation was to study the plastic deformation behaviour of as-quenched dual-phase steels by utilizing indirect methods. It should be mentioned that, according to the our litarature survey, up to date, indirect methods (stress relaxation, strain rate change and loading-unloading tests) have not been applied to dual-phase steels, yet. The material used in this investigation was temper rolled, about 1.0 mm thick plain carbon sheet steels with a composition of 0.074 %C, 0.41 %Mn, 0.03 %Si, 0.071 %Cu, 0.073 %A1, 0.026 %P and 0.021 %S. Tensile test specimens were machined from as-received sheet, according to ASTM E8 standart. To eliminate the effect of temper rolling and provide normalized starting microstructure, all specimens were annealed 1 hr at 960°C and cooled in air. Normali zed specimens were then intercritically annealed at five different tempuratures between 740-780°C for 9,Q;s,in salt bath and water quenched. By this heat treatment, f erritic-martensitic microstruc ture with five different martensite volume fractions were obtained. The mechanical properties of as-quenced dual-phase steels which contain different amounts of martensite, were determined with tensile test at an initial strain rate of 1.3x10`^ l/s. To interprete the plastic deformation behaviour of these steels, stress relaxation, strain rate change and loading-unloading tests were applied at various plastic strains until necking. For stress relaxation, selected initial strain rate was 1.3x10-4 i/s. strain rate change test were performed by decreasing the strain rate from 1.3x10`^ l/s to 1.3xl0~5 l/s or by increasing the strain rate from 1.3xl0-5 l/s to 1.3x10-4 i/s. in lo ading-unloading tests, the utilized initial strain rate was also 1.3xl0`4 l/s and unloading was done until all the stress on the speci men has been removed. All of these test were carried out with Instron Universal Testing machine at room tempurature, using extansometer. At each plastic strain, transient yield effect was observed by reloading after stress relaxation, increasing the strain rate and reloading after unloading in these as-quenched dual-phase steels. These trainsient yield effects were preciously determined by utilizing the `Ten Step Zero Load Suppresion` unit of the Instron Universal Testing Machine. It should be emphasized that, no attempt was made to improve the mecha nical properties of these steels. viiiThe following results were obtained from the experiments, 1) The mar tens ite content of dual-phase steels, produced from plain carbon (0.074 %C) sheet steels by intercritiçal annealing and water quenching, increases with annealing tempurature parabol icly. Increase in martensite volume fraction causes linear increase of yield and tensile strengths and parabolic decrease of uniform and total elongations of as-quenched dual-phase steels. For this production method, optimum mechanical properties of as-quenched dual-phase steels can be obtained if martensite volume fraction is less than 17 percent. 2) It is possible to seperate the stress-strain curves of as- quenched dual-phase steels into three regions according to Crussuard- Jauol analysis, which is used to determine the deformation behaviour of dual-phase steels, from the view point of changes in the substruc ture. Increase in martensite volume fraction, makes the seperation of I. and II. regions difficult, but in general, causes a decrease in the critical transition strains from region I to II and region II to III. 3) According to stress relaxation, strain rate change and loa ding-unloading tests, the deformation behaviour of as-quenched dual- phase steel is controlled by two regions. The strain ranges which correspond to these two regions depend on the martensite content of the dual-phase steels. In the first region, dislocation interaction and in the second region, overcoming Peierls-Nobarro stress barrier are the effective dislocation mechanisms. The coarsening of disloca tion obstacles occur during plastic deformation by binding of mobile dislocations. The distance between these obstacles decreases and then stays constant with increasing deformation. The effect of de formation on the distance between dislocation obstacles is in accord with the change in cell size with strain. 4) Reloading after stress relaxation or increasing the rate of straining suddenly, during strain rate change test or reloading after removing all of the stress on the specimen during tensile test cause the occurance of transient yield effect on a small region of stress- strain curves of dual-phase steels, which exhibit continuous yielding during transition from elastic to plastic deformation. Ip all these tests, observed transient yield condition, exhibit three characteristic stages: a yield point with increment in stress (Aay ı) f followed by a yield drop (AOyyd.) and/or a plateau and finally the return to the ori jinal trajectory o 5) For each test, the value of increment in stress (Aay £), even if different in magnitude, generally increase with plastic strain and become constant when a stable cell structure is formed. The in crease of martensite content of dual-phase steels, decreases the va lue of strain at which Aay £ becomes constant. On the other hand, yield drop (Aoyyd)' is not observed until a critical plastic strain (Ep c) is reached. The effect of martensite volume fraction on the ixvalue of ep c depends on the type of testing method. That is, for stress relaxation and strain rate change tests, ep>c increases, but for loading-unloading test, epjC decreases with increasing martensite content of dual-phase steels. 6) Transient yield points which arise from reloading after stress relaxation performed at plastic deformation region are, due to the decrease of mobile dislocation density during the test. Pinning of mobile dislocations by solute atoms is the main mechanism. Thus, it is concluded that as-quenched dual-phase steels are exposed to strain aging under stress, at room tempurature. 7) The reason of transient yield point observed in strain rate change test is, pinning of dislocations and effective dislocation sour ces by solute atoms at low strain rates and the sudden increase in dis location density to compensate to high rates of strain when the strain rate is increased. 8) Transient yield point, occured by reloading after unloading to zero stress level in the plastic deformation region of as-quenched dual-phase steels, is attributed to the decrease of mobile dislocation density during unloading. The decrease in the density of mobile dislo cations is due to the interaction of mobile dislocations with other dislocations and/or dislocation obstacles. Strain aging phenomenan is not effective for such transient yielding which arises from reloa ding after unloading. x