lei??? I! {Dig i:$5n$u.fl$$zaalfi.. . .fi ‘ _.- 7.3;)! Its-.13, «Sift: rietcéxl: t...“ l u..7,7 1., .v. :1. .(.\. (.52. 5... Cr. 17‘.) A; \ F}.- .. 2Y4...2.,l:.$t .2) 5.....‘iz: E \ b 13‘.) «vi . 5 r) 1...... :(xrgzllrl4 5.51215. it’llclitilrill‘vlliuixt; 1 L... 1:? . . .23.; 5.5.1..) civil} :lll’tlllchgliiig v :1 I. t . r .1 o (p r: . [all5911.4r41.1.33)¢t§fv€lifl1irllticlfr119rl§5z I... i}: u .5 (a ‘ H. 1...: .nl .\,. » .. 5.»... .\ 3 E K ; {xi .5 .. L. r A ‘ ‘ ‘ . (1.17}. ( . r) 4 (if) .4 él‘rl’vzillilr’v'lglnflg I I»: If '31!" (ii-gig . r: v a I p I I. 5.: «(r c all... rrl... «r , It: zlltlllgltllli‘igllllii 2...)... r;Ir(vvizrl.lpil,_(11(l4/I;I t: .15); 5.41 v.9 5. 5.153113 il’fulllr! ‘V ‘ i I, . ‘ .3 . . . s. .l/lillltlv|\l . ,12... ii. . , , V . .. ‘ 3.1.? Einviiuc, 2 p lei .‘vxi ‘ltt , ,.C.. r .. , 1.; .‘ r .v‘.) .31 . ,. . , but not all crystallographically equivalent sets were observed. The recrystallization texture of the supersaturated solid solution was found to be vary according to the annealing temperature but essen- tially a typical transition type recrystallization texture. A texture randomization dur- ing discontinuous precipitation as observed in other system could not be confirmed. To my family TABLE OF CONTENTS LIST OF TABLES ..................................................................................................... vi LIST OF FIGURES ................................................................................................... vii Chapter 1: INTRODUCTION .................................................................................. 1 Chapter 2: LITERATURE SURVEY ...................................................................... 5 Chapter 3: EXPERIMENTAL PROCEDURE ......................................................... 18 3.1 Material preparation ................................................................................. 18 3.2 Specimen Preparation .............................................................................. 18 3.2.1 Rolling Specimens ................................................................................ 18 3.2.2 Annealing Specimens ............................................................................ 22 3.2.3 Optical Microscopy Specimens ............................................................ 24 3.2.4 Transmission Electron Microscopy Specimens ................................... 24 3.3 Tensile Tests ............................................................................................ 25 3.4 Differential Thermal Analysis ................................................................. 25 3.5 Texture Determination ............................................................................. 27 Chapter 4: EXPERIMENTAL RESULTS ................................................................ 29 4.1 Rolling Texture ........................................................................................ 29 4.2 Recrystallization Texture ......................................................................... 29 4.3 Deformation Microstructure .................................................................... 34 4.4 Recrystallization Microstructure .............................................................. 46 iv 4.5 Hardness ................................................................................................... 59 4.6 Stored Energy Measurements .................................................................. 61 4.7 Tensile test ............................................................................................... 61 Chapter 5: DISCUSSIONS ........................................................................................ 68 5.1 Precipitation and Recrystallization .......................................................... 68 5.2 texture ....................................................................................................... 70 Chapter 6: CONCLUSIONS ...................................................................................... 74 LIST OF REFERENCES ........................................................................................... 76 LIST OF TABLES Table No. 2-1 Summary of main characteristics in rolled f.c.c. metals ................................ 3-1 Analyzed chemical composition of experimental alloys ................................ 3-2 Jet polishing conditions for Cu5wt%Ag ......................................................... 4-1 List of the interplanar spacings of matrix and precipitates ........................... 4-2 Yield limit response data ................................................................................ Page 15 19 23 52 66 Figure 1.1 Figure 2.1. Figure 2.2. Figure 2.3. Figure 2.4. Figure 3.1. Figure 3.2. Figure 3.3. Figure 4.1. Figure 4.2. Figure 4.3. LIST OF FIGURES Schematic TIT diagram for the start of recrystallization and precipitation [2,3]. .............................................. 2 Schematic sketch of combined discontinuous reaction [8]. ........................................................................................... 7 {111} pole figures of rolling textures of Cu and Cu-Zn alloys before and after recrystallization [9]. ............................. 9 Sections through the plane at =90° showing probable slip rotations around the cube orientation [16]. ..................................................................................... 10 Changes in the volume fraction of various texture components during recrystallization of cold rolled copper [22]. ..................................................................... 12 Scheme of specimen preparation .......................................................... 20 Vickers hardness as function of temperature for 30 min. isochronal annealing for 90% rolled Cu5%Ag. ................................................................................... 21 Dimensions of sheet tensile specimen .................................................. 26 Rolling texture of a supersaturated Cu5%Ag. ...................................... 30 Rolling texture of a aged Cu5%Ag. ..................................................... 31 Recrystallization texture of a supersaturated Cu5%Ag after annealing at 250° C. ...................................................... 32 vii Figure 4.4. Recrystallization texture of a supersaturated Cu5%Ag after annealing at 450‘ C. ....................................................................... 33 Figure 4.5. Recrystallization texture of a aged Cu5%Ag after annealing at 350° C. ....................................................................... 35 Figure 4.6. Recrystallization texture of a aged Cu5%Ag after annealing at 550° C. ....................................................................... 36 Figure 4.7. TEM micrograph of a single phase Cu5%Ag before rolling. ........................................................................................ 37 Figure 4.8. MIcrostructure of a aged two phase Cu5%Ag. .................................... 38 Figure 4.9. MIcrostructure of 90% cold rolled single phase Cu5%Ag after annealing at various temperature. ........................................................................................... 40 Figure 4.10. TEM micrographs and diffraction pattern of a single phase Cu5%Ag after 90% rolling. ................................... 41 Figure 4.11. TEM micrograph and diffraction patterns of a single phase Cu5%Ag showing deformation bands. .............................................................................. 43 Figure 4.12. MIcrostructure of 90% cold rolled aged two phase Cu5%Ag after annealing at various temperature. ........................................................................................ 44 Figure 4.13. TEM micrograph aged two phase Cu5%Ag before rolling. ...................................................................................... 45 Figure 4.14. TEM micrographs and diffraction pattern of a single phase Cu5%Ag annealed at viii 250° C. ................................................................................................... 47 Figure 4.15. TEM micrographs and diffraction pattern of a single phase Cu5%Ag annealed at Figure 4.16. TEM micrograph of a single phase alloy after annealing at 250° C. ..................................................................... 50 Figure 4.17. TEM micrographs and diffraction pattern of a single phase Cu5%Ag annealed at 450° C. ................................................................................................... 51 Figure 4.18. TEM micrograph aged two phase Cu5%Ag annealed at 100° C. .............................................................................. 53 Figure 4.19. TEM micrographs and diffraction pattern of aged two phase Cu5%Ag annealed at Figure 4.20. TEM micrograph aged two phase Cu5%Ag annealed at 200° C. .............................................................................. 56 Figure 4.21. TEM micrograph aged two phase Cu5%Ag annealed at 300°C. .............................................................................. 57 Figure 4.22. TEM micrographs and diffraction pattern of aged two phase Cu5%Ag annealed at Figure 4.23. Vickers hardness as function of temperature for 30 min. isochronal annealing for 90% rolled Cu5%Ag. .................................................................................... 6O Figure 4.24. Figure 4.25. Figure 4.26. Figure 4.27. Stored energy release of a solid solution alloy rolled to 95% reduction. The engineering stress-strain curve after rolling: (a) solid solution alloy (b) aged two phase alloy. The engineering stress-strain curve of a solid solution alloy: (a) after rolling (b) after annealing at 250° C. (c) after annealing at 450° C. The engineering stress-strain curve of a two phase alloy: (a) after rolling (b) after annealing at 350° C. (c) after annealing at 550° C. 62 63 65 CHAPTER 1 INTRODUCTION The recrystallization behavior of alloys is of primary interest for the processing of commercial materials and thus, has been subject of numerous investigations, which have been reviewed recently [1,2,3,4]. Fundamental contributions to the understand- ing of the complex and often seemingly contradictory observations are due to Hom- bogen and collaborators [2,3], who particularly focussed on the time dependence, and therefore sequence and interaction of the various processes that can occur during the annealing of deformed two phase alloys and supersaturated solid solutions. Their basic approach considers both recrystallization and precipitation as thermally activated processes however with different time constants, such that for the time to start any of these processes ti = tiOCXp(Qi/kT) 1 =R or P where R stands for recrystallization and P for precipitation. Qp depends of course on temperature, because of the existence of an equilibrium temperature for the phase transformation. Hence, in a plot l/T vs. In t,- (Fig. 1) a straight line is obtained for recrystallization, while the precipitation curve degenerates to a ’nose’ close to the equilibrium temperature. The position of both curves relative to each other can be influenced by the degree of deformation. With increasing strain both curves are shifted to smaller times, usually the recrystallization curve more strongly than the precipitation curve. For very high strains the recrystallization curve may not intersect Figure 1.1 Schematic 'I'I'I‘ diagram for the start of recrystallization and precipitation [2,3]. the precipitation curve, and recrystallization will begin (or may even go to comple— tion) before precipitation commences. However, in the more usual case recrystalliza- tion and precipitation curves will intersect, giving rise to basically four different tem- perature regimes: T >T1 : only recrystallization in a solid solution T1>T >T2 : precipitation subsequent to recrystallization T2>T >T 3 : interaction of recrystallization and precipitation T3>T : only precipitation (no recrystallization assumed below T3) The most interesting case is the annealing at T FP , combined discontinuous reaction occurs as shown in Fig. 2-1. If F D + FC < F P, recrystallization proceeds without any motion of large angle grain boundaries ( continuous recrystallization ). Hombogen [8] described the microscopic process by subgrain coalescence or Y-node motion. .No ' N1 PM a) N‘,’6 ° ° Nuc, FctFN C) Figure 2.1. Schematic sketch of combined discontinuous reaction [8]. 2.2 Textures There are two basic types of deformation and corresponding recrystallization textures as shown in Fig. 2-2. During room temperature rolling the copper type tex- ture occurs in metals with a stacking fault energy higher than ~40mJ/m 2, while the brass type texture develops in alloys with a stacking fault energy lower than ~2OmJ/m 2. Textures in rolled sheets can be usually described by a few ideal orienta- tions in terms of the plane {hkl} that lies parallel to the plane of the sheet and the direction [uvw] that is parallel to the rolling direction. The copper type rolling texture can be expressed by {112}<111>, {123}<634> and {011}<211> components while the brass type rolling texture is characterized by {110}<112> plus {110}<001> com- ponents. On recrystallization the copper type rolling texture changes into the cube texture and the brass type rolling texture leads to the brass type recrystallization con- taining {326}<835> as the main orientation. The transition between the two types of rolling texture has been studied in various copper alloys [9-13]. The transition from the copper type to the brass type texture can be inferred by decreasing the tempera- ture of rolling or by increasing of the solute content at constant rolling temperature. This has been shown in copper alloys [14]. An increasing tendency for mechanical twinning during deformation is the mechanism which causes the f.c.c. rolling texture transition with decreasing stacking fault energy [15]. In alloys, two types of recrystallization textures may result from the copper type rolling texture, namely the cube texture and the retained rolling texture. The influence of second phase particles and alloying elements is important for the corresponding recrystallization texture. Dillamore and Katoh [16] developed a model of texture for- mation of f.c.c./b.c.c. alloys which predicts the existence of transition bands in which a central region of the curved lattice contains the cube orientation. Fig. 2-3 demonstrates that certain crystals will tend to approach the cube orientation by . . n Deformatuon ReCrvstalhzatno Figure 2.2. {111} pole figures of rolling textures of Cu and Cu-Zn alloys before and after recrystallization [9] .1 10 Y I I I I toonaoé— ‘ “1 3 I I I 7x I I I I | I L ’4‘ “\ (-— LN ‘ ‘9 [orotuo ’ Q~Iu° 8‘69“ Figure 2.3. Sections through the plane at ¢=90° showing probable slip rotations around the cube orientation [16]. 11 rotation around the normal direction and then subsequently diverge by rotation around the rolling direction, eventually arriving at one of the stable orientations. The cube orientation was actually found in the deformed microstructure of c0pper by electron microscopic investigations [17,18]. These nuclei may occur either as dynami- cally recovered subgrains [17] or highly elongated microbands [18] which recover statically at a very early stage of annealing. This was explained by Ridha and Hutchinson as a unique dislocation symmetry of the cube orientation which favours recovery in these areas compared to the other orientations. The nucleation mechan- ism involved is subgrain growth, which occurs by bulging of the microband boun- daries. Ridha and Hutchinson showed this nucleation mechanism using lightly annealed pure copper in which a cube oriented deformation structure is present and is bowing out along one side as a series of mini bulges into the adjacent substructure. Many observations show that this is the typical pattern of behavior. At a slightly later stage the small bulges link together producing a continuous front which advances into the neighbouring dislocated substructure. They found that the nuclei of cube orienta- tion are located in transition bands. The cube orientation is known to occur at the early stage of recrystallization process [19]. The origin of cube texture has been stu- died recently by many authors [18,20,21]. Virnich et a1. [33] followed the progress of recrystallization in cold rolled copper using quantitative texture measurements. Their results ( Fig. 2-4) show that the cube component first increases at the expense of the copper component {2ll}. The S component {123}<634> is consumed later and both these components continue to diminish subsequently at a stage where the rate of development of the cube texture is greatly reduced. The formation of recrystallization textures can be explained by either oriented growth or oriented nucleation. The theory of oriented growth is based on the observa- tion of the orientation dependence of grain boundary migration in single crystals. Measurements of grain boundary movement in deformed aluminum and copper single I _, . _ 123 (534) 5° {8} I . o .- 100 (001) z 1.0 \I I 10012) J 3 . ’I ’1 I Cpbe ‘fio . . Iwufs 201mm") ‘ ‘53; ‘ .. ' 025 1°“ .\ .. I100 ."4—.. ..| _ 0‘ 6 ‘9' I ' ‘u ‘ \ ‘ 80' 320 1280 5120 Figure 2.4. Changes in the volume fraction of various ' ' texture components during recrystallization of cold rolled copper [22]. 13 crystals [23-25] have revealed that boundaries between crystals with a common <111> axis and an orientation relationship described by a 30" or 40” rotation around this axis move the fastest. According to this theory all kinds of orientations are nucleated and the nuclei which have a favorable orientation relationship with the deformed matrix determine the recrystallization texture. In polycrystalline materials the orientation relationship between the nuclei and its neighbors are more compli- cated. It has to be considered that the grains have to grow into several orientations corresponding to the different component of the deformation texture. The recrystalli- zation texture formed by such a process is called a compromise texture. The brass recrystallization texture is the typical example of a compromise texture. According to the theory of oriented nucleation formulated by Burgers et a1. [26] the formation of recrystallization textures is governed by the nucleation process. They proposed crys- tallites which are present within the deformed structure and which are cube oriented undergo recovery faster than those of other orientations. Cube oriented nuclei pro- duced by rapid polygonization can subsequently grow by large angle grain boundary migration. 2.3 Microstructure of the Deformed State Since a recrystallization texture is determined by the structure formed during deformation, nucleation from the deformed state is of major interest in recrystalliza- tion textures. In heavily deformed metals nucleation initiates in band shaped inhomo- geneities such as deformation bands, shear bands and kink bands [27]. At higher degrees of deformation, the microstructure consists of elongated cells. If the cell have small orientation difference across the boundary [28]. Clusters of these bands are sometimes etched at higher strains as characteristic markings in the optical micros- tructure. Deformation bands are regions with a uniform orientation having a band like 14 structure usually lying parallel to the deformation direction. The structure between two deformation bands is called a transition band. Since the orientation changes gra- dually but continuously from one side of a transition band to the other, large orienta- tion differences up to 300 can be built up. Shear bands [29] are reported to form at angles of about 35° to the rolling direction (Fig. 2-5) and cut through differently oriented regions without significant deviation. Even though the bands have quite different orientations fi'om their neighbors, the copper type deformation texture is not affected by them [30]. The heterogeneities in the microstructure described above have been observed to be nucleation sites for primary recrystallization. The main charac- teristic features of the heterogeneous microstructure in f.c.c. metals and alloys are summarized in Table 2-1 [11]. The important mechanisms of nucleation are strain induced grain boundary migration or bulging, and the processes of subgrain formation and subgrain enlarge- ment by the migration or elimination, respectively, of small angle boundaries. The features of subgrain enlargement can be similar to either normal grain growth or secondary recrystallization. The kinetics of these processes can be the basis for the subgrain enlargement observed in materials containing second phase particles. If par- ticles of a second phase exist, nucleation of primary recrystallization can be accelerated or retarded. The effect of second phase particles is determined by both their size and volume fraction. If coarse particles are present before deformation, the region around them can be a favourable nucleation site since a large orientation gra- dient and deformation deve10p in that area [31]. If very fine particles are present, the dislocation structure is uniform [32]. Precipitation shortly before recovery has a simi- lar effect. Here the major process is subgrain enlargement which is dependent on par— ticle growth which is recrystallization in situ. 15 SFE Strain High Medium Low Low Grain elongation Grain elongation Grain elongation Deformation bands Deformation bands Deformation bands Equixed cells Equixed cells Stacking faults Microbands Deformation twins Medium As above As above As above Some clustering Some alignment and alignment of twins of microbands Shear bands at higher strains High As above As for 1 As for 1 Extensive microband Twin alignment alignment and Continuing replacement clustering of twinned structure Shear bands by shear bands Table 2-1 Summary of main characteristics in rolled f.c.c. metals '16 2.4 Precipitation Precipitation from a supersaturated solid solution can proceed by the spinodal decomposition or by the nucleation and growth mechanisms. In the spinodal process no interface effects are involved, but for the nucleation and growth process the for- mation and motion of an interface are the essential features of the transformation. The classical theory of the nucleation process[33,34] requires that the nucleus forms with the minimum possible value of surface energy per unit volume of precipitate. For a precipitate with isotIOpic interfacial energy, this requires a spherical nucleus shape by the Gibbs Wulff theorem [33,35]. However, of usual case in the solid state precipitation where the two phases have different crystal structures, a low energy interface occurs only for particular well matched planes or directions [35 ,36]. This constraint favors the particular relative orientation relationships between the two phases namely those that give good matching. Nucleation can also occur heterogene- ously on dislocations [37,38]. A special case is the situation where the nucleus and the matrix have the same crystal structure and similar lattice parameter and are found in the so called ’cube-cube’ identical orientation relationship. In this case a low inter- facial energy is expected, showing almost no variation with interface orientation, and giving a nearly spherical equilibrium shape as observed [39]. The migration of a large angle grain boundary can pass through a set of precipi- tates during recrystallization and grain growth. If the precipitates in the consumed matrix grain are coherent with the matrix, then a range of interesting interactions may occur. Many investigations have been made into this process for the case where both phases have the same crystal structure. The following phenomena have been com- monly reported: l7 1) The grain boundary does not pass through the precipitates, which therefore retain their initial orientation and become incoherent after recrystallization [40]. 2) The coherent precipitates dissolve after contact with the moving grain boundary and precipitate coherently with the new grain, either discontinuously at the moving grain boundary or continuously at some distance behind the grain boundary. Both cases have been observed experimentally [41,42,43]. 3) The grain boundary is held by the coherent precipitates which then coarsen, lead- ing to a complete halt of the boundary movement. With additional nucleation of recrystallization the so-called ’necklace structure’ of new grains forms [44,45,46]. This is a network of the recrystallized grains along the prior grain boundaries. 4) The grain boundary could pass through the coherent precipitate so that both phases suffer the same change of orientation and thereby retain the coherent low energy interface between the two phases [47]. It has frequently been observed that small increases ( about 10% ) in the yield strength of cold worked 70-30 brass are obtained when it is heated at a temperature below the recrystallization temperature. Small but significant hardness increases have been reported by Mohan et a1. [48] when cold worked Cu-Zn, Cu-Al, and Cu-Si alloys were heated for about 30 min. at 200° C. It has been generally proposed that the two most probable causes of low temperature hardening of these alloys are short range order [49] and Suzuki locking [50]. CHAPTER 3 EXPERIMENTAL PROCEDURE 3.1 Material Preparation The present experiments were carried out on Cu-5 wt%Ag alloys. To form the alloys Cu and Ag of 99.995% purity were melted in an induction furnace with an argon atmosphere containing 10% hydrogen. For homogenization the alloys were annealed in a vacuum furnace at 850°C for 72h. From this block specimens with 15x10 mm2 rectangular cross section and 20mm length were machined and rolled to 50% reduction. Then these alloys were solutionized for 7h at 850° C ( in air, wrapped in Cu foil ) and water quenched subsequent to solutionization. After solution treatment the alloys were chemically analyzed using a nondestruc- tive X-ray microanalysis technique. Analysis was performed in a IEOL JSM 35C SEM which has an analysis unit which adopts the ZAF method. Cu and Ag of 99.995% purity were used as standards and the alloys were used as specimens. Their chemical analyses are given in Table 3—1. 3.2 Specimen Preparation 3.2.1 Rolling Specimens Two types of rolling samples were prepared as schematically indicated in Fig. 3-1. One set of samples was rolled as quenched, i.e as a supersaturated solid solu- tion (solubility limit at room temperature <1%, at 780° C 7.9%). Another set of sam- ples was annealed for 3 h at 600° C and then water quenched. Since the solubility 18 l9 Area # Composition, Wt. Pct. COP?“ Siver 1 94.49 5.51 2 94.76 5_24 3 94.73 527 4 94.32 1 5,68 5 94.71 5.29 6 94.50 5.50 Table 3.1. Analyzed Chemical Composition of Experimental Alloys 20 ICASTI EIOMOGENIZEDJ IT’REROLLED (SOO/QI r— Ffl—gSOLUTIONIYIZ—EDWOO - [OUENCHEID TO RU IEERW IROLLED (90% - [numb] m Figure 3.1. Scheme of specimen preparation 21 .w<§=u 8:8 e8 é 9: 48:5 Econ—comm SE on .5.— BBEoQEB be .8305: ms amazes: £88; .N.m Emmi Gov mmeimmmzme ozamzfi com com com 00..» com com 00 F o h h F r _ L mafia—on anew I o m 333 25. III. 10m 3 H 100 S m 1 G on: N new. mm S noew \/ 32 am / 2: m noom m8 ( I .I. I . I 1...... -.I..I.. -lzLFrONN 22 limit at 600°C is 3%, this second type of samples contained precipitates of a silver rich solid solution as was confirmed by TEM (ch. 4). All samples were rolled to a total thickness reduction of 95%. The sample thickness after rolling was .Smm. Spe- cial care was taken to ensure homogeneous deformation throughout the thickness by keeping the ratio Id ld >1 (Id - contact length, d - specimen thickness) [56]. The specimens were turned end for end between passes. To remove any contamination introduced during rolling, the specimens were subsequently thinned by mechanical polishing to 0.4mm thickness. 3.2.2 Annealing Specimens To establish the recrystallization temperature for the rolled material, specimens of both types were subjected to 30 min isochronal annealing between 100° C and 700°C. The Vickers hardness number was obtained using the microhardness tester (Buehler Micromet) after each annealing treatment. The applied load in the test was 0.3kg: The loading speed was 50ttm/sec and the loading time was 10 seconds. A significant decrease in hardness, which can be attributed to recrystallization, was observed between 250° C and 300° C for the single phase material and between 350° C and 400°C for the two phase material (Fig. 3-2). No pronounced recovery was observed prior to recrystallization. Corresponding to these results, the rolled two phase samples were then subjected to recrystallization annealing for 30 min at 350°C in order to study the influence of the annealing temperature on texture and micros- tructure. There were two sets of the rolled single phase material. One set was recrys- tallized for 30 min at 250°C and one set was recrystallized for 30 min at 450° C. 23 w<§43mso e8 2529.8 mew—£8 co». Nd 033—. mm no.5 _Eom 35905 EN “538% m._owo> :82 6.350 25mm 28 32.58% 8 e E 25% be? 8:55 2:5 2250 E om§o> Gov meow £8 23385 24 3.2.3 Optical Microscopy Specimens Specimens were prepared from longitudinal sections i.e. with the plane of obser- vation containing the rolling plane normal and rolling directions of the rolled sample. The longitudinal sections ND-RD were cut with a Buehler diamond wheel saw. The microstructure was investigated on the cut face of the sample. Specimens were mounted in lucite after slicing and mechanically ground with abrasive grit papers of 240 grit to 600 grit and polished on a cloth, using alumina powder of the size from Sum to 0.3ttm particles to get a scratch fiee surface. The following etchants were used after polishing. hydrochloric acid 167ml water 333ml iron chloride 25g Microscopy was performed in a Neophot 21 Microscope. 3.2.4 Transmission Electron Microscopy Specimens Slices about 3mm thick were cut from longitudinal sections (i.e. the plane con- taining the rolling plane normal and rolling directions) using a Buehler diamond wheel saw. These slices were handground on 600 grit Emery paper to 100m thick. The final thinning with a Tenupol jet polisher produced thin enough (100nm) speci- mens. The polishing conditions are given in Table 3-2 25 In some cases where the electropolishing was not successful, an ion beam thin- ning technique was used. Argon was used as the bombarding species. The transmis- sion electron microscopy studies were carried out by means of a 200KV Hitachi H800 microscope. 3.3 Tensile Tests Mechanical property data (U.T.S., 0.2 pct. offset Y.S. and elongation ) were determined using an SFM Smart-1 Testing System. Specimens of both single and two phase alloys were tested after 90% cold reduction and after 30 min heat treatments in the range of 250°C to 550°C. All heat treatments were carried out in a vacuum fur- nace. Suitable tensile specimens were machined from the rolled and annealed sheet and pulled to fracture at room temperature in an United Tensile Tester at a strain rate of 0.0048 in. per in. per min. The specimen dimensions are shown in Fig. 3-3. The tensile specimen was oriented such that its tensile axis was parallel to the direction of rolling. The yield stress was taken as the stress measured at 0.2 pct strain. 3.4 Differential Thermal Analysis Differential thermal analysis (D.T.A.) was performed using a Dupont 900 Differential Thermal Analyzer equipped with a differential scanning calorimeter (D.S.C.) cell. The sample was a solid solution alloy which was rolled to 95% reduc- tion and prepared as a 7.4mg cube. An empty copper pan was used as a reference. Both sample and standard were cleaned ultrasonically with acetone prior to thermal analysis to provide a clean flat surface for contact with the D.S.C cell thermocouples and to avoid surface contamination. The specimen was heated with a rate of 5° C/min. to 650° C under a 90% nitrogen plus 10% hydrogen atmosphere. 26 *‘——— 0.43 (10.922) ———‘ 0.125 (3.302) I 0.94 (24.13) I —""" r 0.17 (4.318) 6) 0.24 (6.09 A .~-—.. Figure 3.3. Dimensions of sheet tensile specimen 0.006 (0.152) 27 3.5 Texture Determination The crystallographic orientation of an individual crystallite in a polycrystalline sample is defined by the orientation of its crystal coordinate system Kc with respect to the sample coordinate system K S. The orientation g of Kc with respect to K s can b described by the angles or and B of a direction [uvw] and the rotation angle ‘y[uvw]. If the direction [uvw] is normal to the reflecting lattice plane (hkl) then the reflected intensity is independent of a rotation of the crystal through angle 7. Thus polycrystal diffraction yield the orientation distribution of the crystal direction [uvw] as a func- tion of or and B. This is called a (hkl) pole figure. A Norelco wide range goniometer was mounted on a diffractometer base. An IBM XT computer was adapted to the counter electronics through an OMEGA WB- AlO—B Analog-to digital converter for the automatic determination of pole figures. In order to check the accuracy of this system, an aluminum single crystal and lithium fluoride single crystal were used as the test material. The orientation of the single crystal were verified by Laue patterns. Usually it is not possible to exactly orient the specimen and thus, to indicate the rolling direction in the pole figure. Thus the rolling direction on the measured pole figure is not in a correct position. It is more accurate to determine the rolling direc- tion by symmetry considerations from the pole figure measurements themselves. This is achieved by rotating the pole figure through an appropriate angle, until the correctly adjusted symmetric pole figure is obtained. The calculation of the correction factors (i.e. geometrical factor, defocusing fac— tor, and absorption factor) can be avoided if the reflected intensity is compared with the corresponding intensity reflected from a random sample under the same condi- dons 28 The software system can be divided into three parts. The first is used for data acquisition. The second is the pole figure plotting routine. The third part, which is the plot of the ideal orientation, is applied to the second part to figure out components of the pole figure. The measured intensity data from the data acquisition program are not yet pole density values. The calculation of the correction factor can be avoided if the reflected intensity is compared with the corresponding intensity reflected from a random sam- ple under the same condition. In order to obtain the pole density, corrections have to be applied which are carried out by the normalized pole figure program. This normal- ization was done by the large number of randomly oriented grains, i.e. a powder sample and expressed as times random. The average intensity of the powder sample is calculated in each step of or motion and the measured intensity is divided by this value. The final step of the measuring process is the data output or data storage. After the measurement the representation of the data is made by drawing a pole figure on the printer. The pole figure is represented by contour lines of equal inten- sity level. CHART ER 4 EXPERINIENTAL RESULTS 4.1 Rolling Texture After rolling the texture of the samples was determined by pole figure measure- ments, using the Schultz reflection method on a Norelco pole figure goniometer with spiral scanning. The data were normalized with regard to a powder specimen of pure Cu. Rolling textures of aged and supersaturated samples are represented by {111} and {200} pole figures as shown in Fig. 4-1,2. The solid solution alloy rolled to 95% reduction exhibits a transition type rolling texture consisting of S orientation {123}<634>, brass orientation {011}<211>, and Goss orientation {011}<100>. These three texture components are clearly present in both {111} and {200} pole figures. The two phase alloy rolled to 95% reduction shows essentially the same texture con- sisting of brass orientation, Goss orientation, and also with intensity close to the S orientation. There was no evidence to distinguish the rolling texture of the solid solution alloy from that of two phase alloy except the split of the highest intensity in between {011}<211> and {011}<100> in the {111] pole figure of the solid solution alloy. 4.2 Recrystallization Texture The recrystallization textures of the solid solution alloy after rolling and anneal- ing at 250°C and 450°C respectively are shown in Fig. 4-3,4. The main components 29 RD RD I ‘ i . .‘l \ _ ‘ I ~- '7'— \ \\ / ‘ —‘ \ ' — ‘ J ' ‘ ‘1' (_ Figure 4.1. Rolling texture of a supersaturated Cu5%Ag. Figure 4.2. Rolling texture of a aged Cu5%Ag. 32 Figure 4.3. Recrystallization texture of a supersaturated Cu5%Ag after annealing at 250°C. 33 Figure 4.4. Recrystallization texture of a supersaturated Cu5%Ag after annealing at 450°C. 34 of the specimen annealed at 250°C are {001}<100> and {132}<113>. The recrystall- ization textures of annealing at 450°C are different from those of annealing at 250°C. While cube orientation {001}<100> exists again, a {368}<423> orientation appears instead of the {132}<113> component. The recrystallization of the two phase material results in a texture distinctly different from the deformation texture for both annealing at 350° C and 550° C as shown in Fig. 4-5,6. The recrystallization texture for 350° C annealed specimen con- sists of two components which are {001}<100> and {123}<634>. The recrystalliza- tion texture for the 550°C annealed specimen shows essentially the same texture as annealing at 350° C. By comparison of these two recrystallization textures it can be noted that texture is getting sharper as annealing temperature increases. 4.3 Deformation Microstructure The microstructural development of the deformed single and two phase alloys was investigated through the optical microscope and especially the electron micro- scope. The microstructure of the supersaturated solid solution alloy prior to defbrma— tion was investigated to assure the attainment of a homogeneous solid solution. The electron micrographs shown in Fig. 4-7. reveal that the microstructure is not actually homogeneous. Instead these are needles of submicron size as straight lines or spots depending on their orientation with respect to the angle of observation. The existence of these precipitates may be due to an imperfect quenching procedure or precipitation at room temperature after quenching. The former case is more likely than the latter case, since in general the rate of precipitation from a copper based supersaturated solid solution is very slow. The microstructure of a two phase alloy which was precipitation treated at 600°C for 3 hours to purposely obtain a second phase is given in Fig. 4-8. As 35 Figure 4.5. Recrystallization texture of a aged Cu5%Ag after annealing at 350° C. 36 / ’/ R D Figure 4.6. Recrystallization texture of a aged Cu5%Ag after annealing at 550° C. 37 .weae 883 m crystallographic direction. The pattern comprises both the matrix and precipitate diffraction, each diffraction spot cluster consisting of matrix and precipitate components and the associated double diffraction spots (ac-u laAg = 361/409). A dark field image with the indicated diffraction spot reveals that only rod shaped precipitates which are aligned along a <110> direction belong to the corresponding orientation. Thus weak extra spots near the vicinity of the transmitted spot whose zone axis is close to <111> may arise from smaller globu- lar precipitates or rods extending in a mainly perpendicular direction of the image plane. Optical micrographs of the deformation microstructure of the solution treated alloy are shown in Fig. 4-9(a). Longitudinal sections of the specimen rolled to 90% reduction showed microstructures similar to that found in medium stacking fault energy material. The microstructure consisted of deformation bands and Cu type shear bands. Cu type shear bands are known to make an angle of 35° with respect to the rolling plane [8]. Deformation bands are aligned such that their boundary is paral- lel to the rolling plane. Electron micrographs of the same specimen rendered more detailed features than the optical micrographs. Fig. 4-10(a) shows the Cu-type shear band lying at about 35° to the rolling direction. Higher magnification (Fig. 4-10(b)) shows clusters of microbands aligned parallel to the rolling direction. The micrograph of the highest magnification clearly reveals a long band like feature of ~0.1 mm thick- ness. Selected area diffraction pattern from this regions near [111] shows elongated diffraction spots. This is due to the fact that a distorted grain has split into regions of 40 chafing—boa 32.5, “a weaves“ How“ 33590 0323 Swan 958 28 $8 we 8805.585: .aé 853m 0.63 n ._. A5 use: u ._. 3 0:8». n ._. S Dank-N " h. ADV 0.63 n ._. A3 098” n ._. A8 0.62 n ._. 3 9.8— " ._. A: 41 has m crystallographic direction. In addition to fibrous precipitates much smaller glo- bular precipitates exist also, which which are likely to cause extra weak spots around the transmitted spot whose zone axis is [111]. 43 .353 eoumgfioe wagon: w) can be observed from right to left. The pancake shaped grain structure is retained even after recrystallization. Annealing twins are shown in each pancake shaped grain. The corresponding diffi'action pattern (Fig. 4-14(b)) confirms the occurrence of annealing twins. This diffraction pattern also shows that the precipi- tates are coherent with the matrix since each diffraction spot cluster consists of a matrix and precipitate component and the associated double diffraction spots. Whereas most precipitates have rod shapes, some have sphere shapes of different size. Actually rod shape precipitates are seen as small sphere shape precipitates since they are almost normal to the specimen surface. On the other hand, the formation of large sphere shape precipitates seems to be due to precipitate growth at the expense of smaller precipitates. Fig. 4-15 shows the microstructure of the same sample within a different grain. The overall impression is the same but the concurrent occurrence of recrystallization and discontinuous precipitation is shown more clearly compared with the one shown in Fig. 414. In the upper left grain the rod shape precipitates do not extend to the adjacent grain boundary, and rather large globular precipitates dominate in such regions. This may due to the fact that the area denuded of fibrous precipi- tates corresponds to the nucleus position with its boundary having moved away dur- ing recrystallization or precipitation prior to recrystallization. The specimen annealed at 450°C shows features similar to the specimen annealed at 250°C. A recrystallized grain several micrometers in diameter is shown in the center. The precipitate morphology is more heterogeneous, comprising longer and also shorter rods and even globular shapes. Note that with increased annealing 47 .058 a @0385 w. The comparison of both diffi'action patterns reveals an orientation relationship between them which is a 45° rotation around the axis close to the <100> direction. However the orientation rela- tionship between upper grain and deformed region is a 45° rotation around the axis close to the <110> direction. The microstructure of an aged alloy sample annealed at 200°C (Fig. 4-20(b)) shows features similar to annealed at 100°C. It is clearly shown that fibrous precipi- tates transform to globular precipitates 0.1 to 0.3 pm in size upon recrystallization of the matrix phase. The increase of the annealing temperature enhances the progress of recrystallization and increases the recrystallized area. After annealing at 300°C, a large area was recrystallized which contained large globular precipitates and also fine needle shape precipitates. Fig. 4—21. shows a recrystallized grain several pm in size which is slightly larger than those annealed at lower temperatures. Drastic changes in the microstructure of the aged alloy occur after annealing at a higher temperature of 550° C. Here the morphology of precipitates is totally different from those annealed at a lower temperature. While the large globular shape precipitate and the fine needle shape precipitate do not appear, extremely fine globu- lar precipitates are distributed randomly. As evident from the optical micrograph, the grain boundary can not be found through TEM due to a 50 um diameter grain size after grain growth. A selected area diffraction pattern from the microstructure in Fig. 4-22Cb) shows typical ring pattern. This can only be due the existence of the 55 .0909 a 83093 . . 05w” w<§mso 03:3 23 comm mo Egan “Stowaway Ea Emacs? Emit E v .m E: _ 56 .0 seem 8 BREE“ w<§m50 82m 95 cows nauuwobwd Em? .om.v Rama E: m.._ ,.7./ :;.; 57 . . Rama .0 com 3 Banana w<§nao 82m 95 cows :92?wa 2B. “N v E1 m 412:; m 58 .0 5mm 3 330:5 m as shown in Fig. 4-14 and Fig. 4-17. From the SAD pattern it is apparent that these precipitates are coherent with the matrix phase. These fine precipitates are distributed throughout the volume, seemingly neither to be dissolved nor to be passed through by the motion of the moving grain boundaries. The alternative case is observed in discontinuous precipitation and continuous precipitation of relatively large particles compared with the much smaller continuous precipitate mentioned above. In the aged two phase alloy the elongated fibrous pre- cipitate is dissolved completely by the migrating grain boundary, and large globular particles precipitate at the grain boundary and also inside the newly formed grain. In this case coherent precipitates reprecipitate coherently with the growing grain, as was frequently observed in a variety of alloy systems [41,42,43]. Similar features can be observed in the supersaturated solid solution alloys. Relatively large globular shape 68 69 particles form in the precipitate free zone ( only for large scale precipitates ) behind the grain boundary. This area seems to be the portion where the grain boundary moved away during grain growth immediately following the combined discontinuous reactions. As already mentioned fine continuous precipitates have been observed frequently after annealing of the deformed alloys. It is hard to tell whether the fine scale precip- itates are the result of continuous precipitation or spinodal decomposition on the basis of electron microscopy [51]. However a miscibility gap has been constructed on the equilibrium phase diagram by Boswell et al. [52] using an equation of Cook et al. [53]. The occurrence of spinodally decomposed structures is easier in the deformed alloy compared with the undeformed alloy [54]. It was also shown that the elastic strain energy caused from the coherency strain lowers the temperature of spinodal decomposition [55]. Then, in the present case, the occurring temperature of spinodal decomposition seems to be lowered additionally. Drastic changes in the morphology of the precipitates occurs after annealing of the two phase alloy at 550° C. Small globular precipitates about 20nm in diameter are dispersed within the matrix. The two phases seem to form a coherent interface to attain the lowest free energy since the Ag rich precipitate has the same crystal struc- ture and a similar lattice parameter. Actually the atomic size difference is large (11.6%) compared with a system such as Al-Ag (0.97%), which forms fully a coherent precipitate. The matrix and precipitate must be strained to maintain a coherent interface. These precipitates seem to have been formed after the completion of recrystallization. It seems that long fiber like precipitates have been dissolved completely by the motion of the recrystallization front. Small globular precipitates have also been formed continuously behind the moving grain boundary. Thus precipi- tate are formed in the strain free region different from other cases such as discontinu- ous or localized precipitation. This could be the reason for the formation of different 70 morphology in both size and shape. In the supersaturated solid solution alloy recrystallization and discontinuous pre- cipitation occurred simultaneously after annealing at 250°C and 450°C. The difference in annealing temperature seems not to affect the sequence of recrystalliza- tion and precipitation. But, as seen in the optical micrographs, grain growth occurred after the completion of combined discontinuous reactions in the specimen annealed at 450°C. After recrystallization, grains grow further during continued annealing. Dur- ing this grain growth it seems that moving grain boundaries bypass the coherent pre- cipitates, and then these precipitates become incoherent with the matrix phase. The occurrence of an apparent ring pattern in single phase alloys annealed at 450°C would support this interpretation. 5.2 Texture The rolling textures are very much as expected. While pure copper develops a Cu type rolling texture, its concentrated alloys tend to form a brass type texture [9- 13]. With increasing solute content the Cu type rolling texture changes to a transi- tion type texture and finally to the brass type rolling texture. According to the pole figures both the single and two phase alloy exhibit a transition type texture. It seems obvious that 5% Ag in Cu content can decrease the stacking fault energy low enough to cause texture transition. The transition of the rolling texture observed for the present alloy is similar to that observed in the Cu-Zn [11], Cu-P [12], and Cu-Ge [13]. Only the amount of alloy content to cause the texture transition is different in such a way that 5% Ag corresponds to 6% Zn, 1% P, and 2% Ge respectively. Thus, only one parameter, the stacking fault energy, apprently causes the rolling texture transition. 71 According to present results there is no evidence to distinguish the rolling tex- ture of the solid solution alloy from that of the two phase alloy. Here the aging treat- ment prior to rolling to produce the second phase has little effect on the subsequent deformation behavior. Even though a second phase is distributed throughout the matrix phase as the long fibrous shape already shown in TEM micrographs, the deformation texture is almost the same as that of the single phase alloy. From this result it can be concluded that the motion of dislocation or mechanical twinning was not notably affected by the presence of a second phase. This may be attributed to the fact that the second phase (i.e. silVer rich precipitate) has deformation characteris- tics similar to copper alloys. After annealing one finds different textures for the supersaturated solid solution alloy and the aged two phase alloy. In the first case the resulting texture varies with the annealing temperature. While the cube orientation can be found at both annealing temperatures, other details differ. From the pole figure it can be noted that the inten- sity of the cube component gets stronger with the increase in the annealing tempera- ture. This is probably due to grain growth, since the annealing texture continues to develop after completion of recrystallization. Although there is a minor difference in the texture components the annealing texture of the solid solution alloy develops by combined discontinuous reactions as evident from the TEM micrographs. In the second case ( i.e. aged two phase alloy ), the recrystallization textures annealing at for both 350° C and 550° C is a mixed texture comprising retained rolling and cube texture. This implies that continuous recrystallization ( recrystallization in situ ) is the major mechanism which leads to the partial retention of the rolling texture, while the cube component belongs to the portion which has been recrystallized discontinuously. The occurrence of a cube component seems to be due to rapid growth of cube oriented nuclei since they undergo recovery and recrystallization much faster than the nuclei of other orientations during the early stage of recrystallization [17,18]. 72 The observations in the supersaturated solid solution are far more complicated owing to the concurrent occurrence of precipitation and recrystallization. Hombogen and Kreye [2] (referred to as HK in the following) investigated a variety of commer- cial alloys Al-Cu, Ni-Al, Ni-Cr-Al, Cu-Co, Ni-Be and found that a discontinuous recrystallization always leads to a random texture even from a pronounced deforma- tion texture, and they attribute their result to a loss of growth selection owing to impurity drag because of segregation of the solute to the grain boundaries. The present results cannot confirm the observation of HK. The cube orientation is known as an orientation to develop very early during recrystallization [19]. According to current understanding of cube texture formation [18,20,21], it first nucleates in the Cu component of the rolling texture and then favorably grows into the S component. This is consistent with the present results, since both Cu and S orientation are identified in the rolling texture. In conclusion, the recrystallization textures can be basically accounted for by the known recrystalliza- tion mechanisms for texture formation in solid solutions. This implies also that the observed crystallographic growth of the discontinuous precipitates, which actually provides the major driving force for grain boundary migration during recrystallization does not noticeably affect the growth of the recrystallized grains. It is not quite clear what causes the difference between the current investigation and the results of I-IK on other systems. HK surmise that the texture randomization is due to a loss of growth anisotropy owing to segregation. This may indeed be correct for the systems that they studied. However, it certainly cannot be generalized, other- wise recrystallization texture formation in single phase alloys could be ruled out in general, quite in contrast to common observations, e.g. in the well studied system .Cu-Zn. Without any doubt the extent of segregation and consequently its effect on grain boundary mobility depends strongly on the alloying element. For example it is well known that Cu or Mg additions to A1 effectively reduce the recrystallization 73 kinetics, while alloying of A1 with Ag hardly has any effect on the recrystallization of Al. Therefore, each alloy system has to be considered differently. From our results we have to conclude that also alloying of Cu with Ag does not substantially change the recrystallization mechanisms operating in pure Cu. In fact, the rolling and recrystallization textures as established in this investigation are qualitatively very similar to results on Cu6%Zn [63], essentially consisting of the same components, although different in strength. In the rolled single phase alloy the presence of a cube oriented region was identified in the transition band area by means of SAD patterns in the electron micro- scope. The existence of a cube orientation in the deformed structure is important to the argument about whether the cube texture is developed by oriented growth or oriented nucleation. The cube oriented region located in the thin band region elongated along the rolling direction and the large angle grain boundary separates the cube oriented band from its neighbor deformation band. Dillamore et al. [16] pro- posed a model for the formation of the cube orientation in f.c.c. materials, saying that the cube orientation can be formed in the transition band region during deformation. Their model predicts that certain orientations rotate around the normal direction to approach the cube orientation and then rotate further around the rolling direction to form the cube orientation. In agreement with other authors [16,18] the cube orienta- tion can be identified in the ND-RD plane. From the corresponding SAD pattern it is evident that the cube oriented grain is separated from its neighbor by a large angle grain boundary. It is worthwhile to note that the cube oriented band is rotated around the rolling direction with respect to its neighbor as predicted by Dillamore et a1. [16]. During annealing this large angle grain boundary can migrate rapidly due to its high mobility. CHAPTER 6 CONCLUSIONS The main results are summarized as follows: 1) Discontinuous precipitates present prior to rolling, which are coherent with the matrix phase, tend to deform with the matrix becoming fibrous shaped 2) Annealing of the deformed supersaturated solid solution alloys at 250° C and 450° C leads to the formation of a fine continuous matrix precipitation followed by a combined discontinuous reaction of recrystallization and precipitation. The migrating gain boundary bypasses fine continuous precipitates but leaves fibrous precipitates at the gain boundary and also inside the newly formed gain. The difference in anneal- ing temperature seems not to affect the sequence of recrystallization and precipita- tion. But gain gowth occurs after the completion of combined discontinuous reac- tions in the specimen annealed at 450°C. During this gain gowth it seems that moving gain boundaries bypass the coherent precipitates and then these precipitates become incoherent with the matrix phase. 3) In the two phase alloy the elongated fibrous precipitates are dissolved completely by the migating gain boundary and large globular particles precipitate coherently behind the moving gain boundary. 4) Drastic changes occurs in the morphology of the precipitates during annealing of the two phase alloy at 550° C. While the large globular shape precipitate disapper, small globular precipitates about 20nm in diameter are dispersed within the matrix. These precipitates seem to form after the completion of recrystallization and gain gowth. 74 75 5) The deformed single and two phase alloys exhibit a transition type rolling texture comprising an S orientation {123}<634>, brass orientation {011}<211> and Goss orientation {011}<100>. The aging treatment prior to rolling has little effect on the subsequent deformation texture. 6) The recrystallization textures of the supersaturated solid solution alloy annealed at 250°C and 450°C are essentially of transition type texture with the cube orientation {001 }<100> as a major component. 7) The recrystallization textures of the two phase alloy for both annealing at 350° C and 550°C are mixed textures consisting of partially retained rolling and cube tex- ture. The cube component is getting stronger due to gain gowth as the annealing temperature increases. 8) Texture randomization is not observed after discontinuous recryStallization. The crystallogaphic gowth of the recrystallized gains was not influenced noticeably by the concurrent discontinuous precipitation which actually provides the major driving force for gain boundary migation during recrystallization. 9) In accordance with the findings of other authors in copper [16,18], the presence of a cube oriented region is identified in the transition band area in the rolled single phase alloy. [1] [2] [3] [4] [5] [6] [7] [8] [9] LIST OF REFERENCES P. Cotterill and RR. Mould: Recrystallization and Grain Growth in Metals, John Wiley, New York, (1976) E. Hombogen and H. Kreye: in Texturen in Forschung and Praxis, ed. J. Grewen and G. Wassermann, Springer Verlag, Berlin, (1969) 274 E. Hombogen and U. Kbster: in Recrystallization of Metallic Materials, ed. F. Haessner, Dr. Riederer Verlag, Stuttgart (1978) 159 N. Hansen: Mem. Sci. Rev. Met. 72 (1975) 189 E. Hombogen: Z. Metallk. 69 (1978) 735 S. Saji and E. Hombogen: Z. Metallk. 69 (1978) 741 C. Zener and CS. Smith: Trans. AIME 175 (1949) 15 E. Hombogen: Prakt. Metallogaphie 9 (1970) 349 U. Schmidt and K. Lticke: Texture 3 (1979) 85 [10] RA. Beck: Advances in Physics 3 (1954) 245 [11] J. Hirsch and K. Lllcke: Acta Met. 36 (1988) 2863 [12] U. Schmidt, K. Lllcke and J. Pospieche: Int. Conf. on Textures, Cambridge, (1975) 154 [13] H. Eichellcraut, J. Hirsch and K. Lllcke: Z. Metallk. 75 (1984) 113 [14] R. Alam, H.D. Mengelberg and K. Lllcke: Z. Metallk. 58 (1967) 867 [15] G. Wassermann: Z. Metallk. 54 (1963) 61 [16] LL. Dillamore and H. Katoh: Met. Sci. 8 (1974) 73 [17] T. Noda and J. Huber: Z. Metallk. 69 (1978) 570 76 77 [18] A.A. Ridha and W.B. Hutchinson: Acta Met. 30 (1982) 1929 [19] J.W.H.G. Slakhorst: Acta Met. 23 (1975) 301 [20] J. Hirsch and K. Lllcke: in Annealing Processes - Recovery, Recrystallization and Grain Growth, ed. N. Hansen et al., Riso National Laboratory, Roskilde, Denmark (1986) 361 [21] J. Hirsch, E. Nes and K. Lllcke: Acta Metal]. 35 (1987) 427 [22] C. Zener and CS. Smith: Trans. AIME 175 (1949) 15 [23] B. Liebmann, K. Lllcke and G. Masing: Z. Metallk. 47 (1956) 57 [24] G. Ibe, W. Dietz, A.C. Fraker and K. Lucke: Z. Metallk. 61 (1970) 498 [25] P. Stuijte and G. Gottstein: in Proc. 5th Int. Conf. on Textures of Materials, ed. G. Gottstein and K. Lllcke, (1978) 511 [26] W.G. Burgers, Y.H. Lin and T.J. Tiedema in The influence of crystal orientation on polygonisation of aluminum single crystals, Proc. Nederlandse Akadamie van Wetenschappen. 54 (1951) 459 [27] S.P. Bellier and D. Doherty: Acta Met. 25 (1977) 521 [28] H. Hu: in Recovery and recrystallization of Metals, ed. L. Himmel, Interscience Pub1., New York (1964) 311 [29] M. Hatherly: in Proc. 5th Int. Conf. on Textures of Materials, ed. G. Gottstein and K. Lllcke, (1978) 81 [30] J. Grewen, T. Noda and D. Sauer: Z. Metallk. 67 (1977) 260 [31] A. Kelly and RB. Nicholson: in Progress in Materials Science, ed. B.Chalmrs, Pergamon Press, New York 10 (1963) 149 [32] R]. Humphry and J.W. Martin: Phil Mag. 17 (1968) 146 [33] K.H. Virnich, G.Koh1hoff, K. Lucke and J. Pospiech: in Proc. 5th Int. Conf. on Textures of Materials, ed. G. Gottstein and K. Lllcke, (1978) 475 78 [34] J.W. Christian: in The [theory of transformations in metals and alloys, (1965) Oxford, Pergamon. [35] KC. Russel: in Phase transformations, New York, ASM, Chapman and Hall, (1970) 219 [36] J .W. Martin and RD. Doherty: in The stability od microstructure in metallic systems, Cambridge, Cambridge University Press, (1976) 10, 163, 182 [37] RD. Doherty: in Phase boundaries, ed. D. Smith and GA. Chadwick, London, Academic Press [38] J.W. Cahn: Acta Met, 5 (1957) 168 [39] HI. Aronson, K.C. Russell, and G.W. Lorimer: Met. Trans. 8A (1977) 1644 [40] AR. Jones and B. Ralph: Acta Met.: 23 (1975) 355 [41] H.B. Aaron, D. Fainstein and GR. Kotler: J. Appl. Phys. 41 (1970) 4404 [42] M. Dahlen and L. Winberg: Acta Met. 28 (1980) 41 [43] J .V. Bee, A.R. Jones and RR. Howell: J. Mater. Sci. 15 (1980) 337 [44] DE. Stephens and GR. Purdy: Acta Met, 23 (1975) 1343 [45] J.V. Bee, A.R. Jones and RR. Howell: J. Mater. Sci., 15 (1980) 337 [46] J .V. Bee, A.R. Jones and RR. Howell: in Recrystallization and Grain Growth of Multiphase and Particle Containing Materials, ed. N. Hansen et al., Riso National Laboratory, Roskilde, Denmark ( 1980) 153 [47] RR. Howell and J.V. Bee: in Recrystallization and Grain Growth of Multiphase and Particle Containing Materials, ed. N. Hansen et al., Riso National Labora- tory, Roskilde, Denmark (1980) 171 [48] AP. Rama Mohan, P. Rama, and TR. Anantharaman: Trans. Indian Inst. Met, 18 (1965) 173 79 [49] S. Matsuo and L.M. Clarebrough: Acta Met. 11 (1963) 1195 [50] H. Suzuki: J. Phys. Soc. Japan, 17 (1962) 322 [51] M. Bouchard and C. Thomas: Acta Met. 23 (1975) 1485 [52] P.G. Boswell and GA. Chadwick: J. Mater. Sci. 12 (1977) 1879 [53] H.B. Cook and J.E. Hilliard: Trans. AIME 233 (1967) 142 [54] J. Dutkiewicz: Bull. Acad. Pol. Sci. Ser. Sci. Tech., 22 (1974) 323 [55] J.W. Cahn: Trans. AIME 242 (1968) 166 [56] HO. Asbeck: Doctoral Dissertation, RWTH Aachen (1973) [57] F. Haessner, G. Hoschek and G. Tblg: Acta Met. 27 (1979) 1539 [58] F. Haessner and W. Hemmirger: Z. Metallk. 69 (1978) 553 [59] P.G. Schewmon: in Diffusion in Solids, J. Williams and OK. Jenks (1983) 111 [60] K.H. Virnich: Doctoral Dissertation, RWTH Aachen (1979) "Ill11111111111111