UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The effect of pre-deformation on the ageing behavior of 7030.60 and 7108.72 alloys Huang, Jin 1999

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1999-0196.pdf [ 6.15MB ]
Metadata
JSON: 831-1.0078698.json
JSON-LD: 831-1.0078698-ld.json
RDF/XML (Pretty): 831-1.0078698-rdf.xml
RDF/JSON: 831-1.0078698-rdf.json
Turtle: 831-1.0078698-turtle.txt
N-Triples: 831-1.0078698-rdf-ntriples.txt
Original Record: 831-1.0078698-source.json
Full Text
831-1.0078698-fulltext.txt
Citation
831-1.0078698.ris

Full Text

T H E EFFECT OF PRE-DEFORMATION O N T H E AGEING BEHAVIOR OF 7030.60 A N D 7108.72 ALLOYS by Jin Huang B. Eng., Shanghai Jiao Tong University, 1991 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF T H E REQUIREMENTS FOR THE DEGREE OF MASTER OF APPLIED SCIENCE in T H E FACULTY OF GRADUATE STUDIES (Department of Metals and Materials Engineering) We accept this thesis as conforming to the required standard T H E UNIVERSITY OF BRITISH COLUMBIA March 1999 ® Jin Huang, 1999 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of yfr{f>fV/l a<*J j/ft,4tf Tkn I l „ : . ~l r » _ : i : _ L _ r~ _ I L * V The University of British Columbia Vancouver, Canada Date DE-6 (2/88) ABSTRACT This study is part of a research project whose goal is to develop a process model for the production of aluminum alloy automotive bumpers. The problem originates from the industrial bumper manufacturing practice which uses aluminum alloys, AA7030.60 and AA7108.72 in particular. In the industrial production schedule, a forming (pre-deformation) process occurs after the solution heat treatment and water quenching, followed by a T6 ageing process. Due to the specific shape of the bumper product, the forming process introduces a heterogeneous distribution of deformation i n the bumper. A systematic investigation of pre-deformation on the ageing behaviour of these alloys has been carried out. A series of pre-deformations (pre-strain from 0 to 1.2) was imposed on the materials, whose subsequent ageing responses were recorded in terms of the mechanical properties (yield stress, tensile strength, etc.) and electrical resistivity. Some transmission electron microscopy measurements were also conducted by other members of the research group. W i t h a combination of these results the evolution of the material's microstructure and mechanical properties can be traced during the ageing process. Some of the significant results from the present study include the observation that the natural ageing response is reduced and the artificial ageing kinetics are accelerated for the pre-deformed alloys. Lower peak ageing strength is also observed for the pre-deformed alloys. These changes in the material behaviour are related to the changes i n the microstructure when pre-deformation is present in the materials. A t a microscopic level, dislocations are observed to arrange themselves into a loose cell structure. Pre-deformed alloys were observed to have larger average precipitate sizes. In addition, they also have different precipitate size distribution. For pre-deformed samples two peaks can be 11 recognized in the precipitate size distribution curves, corresponding to the precipitates in the bulk and the precipitates at dislocation sites. Based on this study, it has been observed that deformation prior to artificial ageing leads to inferior final properties. A modified bumper manufacturing sequence is proposed in an attempt to eliminate this effect. However, further studies are needed to test the feasibility of the proposal. 111 T A B L E OF C O N T E N T S ABSTRACT ii LIST O F FIGURES vi LIST O F TABLES xi ACKNOWLEDGEMENTS xii CHAPTER 1 INTRODUCTION 1 CHAPTER 2 B A C K G R O U N D 4 2.1 7XXX ALUMINUM ALLOYS USED IN AUTOMOTIVE INDUSTRY 4 2.2 INDUSTRIAL BUMPER MAKING PROCESS 6 2.3 AGE HARDENING 9 2.3.1 The Heat Treatment 9 2.3.2 Precipitation Process 12 2.3.2.1 Formation of G-P Zones 13 2.3.2.2 Formation of r\' 14 2.3.2.3 Formation of r| 16 2.3.2.4 Precipitation Kinetics 16 2.3.3 Dislocation-Precipitate Interaction 17 2.4 T H E EFFECT OF PRE-DEFORMATION O N T H E PRECIPITATION BEHAVIOR 21 2.5 T H E STATEMENT OF OBJECTIVES 23 CHAPTER 3 EXPERIMENTAL PROCEDURE 24 3.1 INTRODUCTION 24 3.2 MATERIAL 24 3.3 METALLOGRAPHIC PREPARATION 25 3.4 EXPERIMENTAL HEAT A N D MECHANICAL TREATMENT 26 3.5 PROPERTY MEASUREMENTS 27 CHAPTER 4 RESULTS 30 4.1 GRAIN STRUCTURE OF T H E AS-EXTRUDED MATERIAL 30 4.2 AGEING BEHAVIOR 33 4.2.1 Evolution of Mechanical Properties During Ageing. 33 4.2.1.1 Typical Tensile Curves 34 IV 4.2.1.2 Ageing Curves in Yield Stress, UTS and Strain-to-necking 37 4.2.2 Evolution of Electrical Resistivity During Ageing. 44 4.3 SUMMARY OF SOME SIGNIFICANT OBSERVATONS 47 CHAPTER 5 DISCUSSION 53 5.1 EFFECT OF Zr ON THE GRAIN STRUCTURES 53 5.2 OVERVIEW' OF THE AGEING BEHAVIOUR 54 5.2.1 Yield Stress Evolution During Ageing 55 5.2.2 Electrical Resistivity Evolution During Ageing. 58 5.3 EFFECT OF PRE-DEFORMATION ON THE YIELD STRESS OF THE AS-DEFORMED MATERIAL 59 5.4 EFFECT OF PRE-DEFORMATION ON THE NATURAL AGEING BEHAVIOUR 62 5.5 EFFECT OF PRE-DEFORMATION ON THE ARTIFICIAL AGEING BEHAVIOUR 64 5.5.1 Artificial Ageing Stage I): AgeingatlOO°C 64 5.5.2 Artificial Ageing Stage II): Ageing at 66 5.5.3 Kinetics of Precipitate Coarsening. 74 5.6 RECOVERY DURING THE ARTIFICIAL AGEING 75 5.7 A SIMPLE MODEL FOR PREDICTING STRESS-STRAIN CURVES 78 5.8 OPTIMUM HEAT TREATMENT 85 CHAPTER 6 CONCLUSIONS AND FUTURE WORK 87 6.1 CONCLUSIONS : 87 6.2 FUTURE WORK 88 BIBLIOGRAPHY 91 APPENDIX 94 v LIST O F F I G U R E S Figure 2.1: Schematic illustration of a typical European bumper ...7 Figure 2.2: Schematic illustration of the thermal-mechanical history of the 7030.60 and 7108.72 aluminum alloys before bumper forming process 7 Figure 2.3: Schematic illustration of the bumper forming and ageing schedule for the 7030.60 and 7108.72 aluminum alloys 9 Figure 2.4: Schematic illustration of the route of phase transformation during quench 11 Figure 2.5: Schematic illustration of dislocation-particle interactions. (1) - (3): moving dislocation shearing a weak particle; (i) - (iii): moving dislocation bypassing a strong particle 19 Figure 2.6: Schematic illustration of the breaking angle <pc for a dislocation holding by an obstacle 21 Figure 2.7: Influence of the precipitate size distribution width A on the evolution of yield stress during ageing 23 Figure 3.1: Schematic illustration of the experiment schedule for the 7030.60 and 7108.72 aluminum alloys 26 Figure 3.2: Schematic illustration of two types tensile sample used in this research, a) for MTS tensile machine and b) for Instron tensile machine 29 Figure 4.1: Tri-planar optical microscopic graphs showing the grain structure of the 7030.60 alloy in the as-received condition 31 Figure 4.2: Tri-planar optical microscopic graphs showing the grain structure of 7108.72 alloy in the as-received condition 32 Figure 4.3a: Engineering stress-strain curves for 7030.60 alloy with no pre-strain, in various stages of ageing till the peak 35 vi Figure 4.3b: Engineering stress-strain curves for 7030.60 alloy wi th no pre-strain, in various stages of ageing after the peak 35 Figure 4.4a: Engineering stress-strain curves for 7030.60 alloy wi th a pre-strain of 0.30, i n various stages of ageing t i l l the peak 36 Figure 4.4b: Engineering stress-strain curves for 7030.60 alloy wi th a pre-strain of 0.30, i n various stages of ageing after the peak 36 Figure 4.5a: Evolut ion of yield stress for the 7030.60 and 7108.72 alloys wi th no pre-strain 39 Figure 4.5b: Evolut ion of U T S and strain-to-necking for the 7030.60 and 7108.72 alloys wi th no pre-strain 39 Figure 4.6a: Evolut ion of yield stress for the 7030.60 and 7108.72 alloys wi th 0.03 pre-strain 40 Figure 4.6b: Evolut ion of U T S and strain-to-necking for the 7030.60 and 7108.72 alloys wi th 0.03 pre-strain 40 Figure 4.7a: Evolut ion of yield stress for the 7030.60 and 7108.72 alloys wi th 0.10 pre-strain 41 Figure 4.7b: Evolut ion of U T S and strain-to-necking for the 7030.60 and 7108.72 alloys wi th 0.10 pre-strain 41 Figure 4.8a: Evolut ion of yield stress for the 7030.60 and 7108.72 alloys wi th 0.30 pre-strain 42 Figure 4.8b: Evolut ion of U T S and strain-to-necking for the 7030.60 and 7108.72 alloys wi th 0.30 pre-strain 42 Figure 4.9a: Evolut ion of yield stress for the 7030.60 and 7108.72 alloys wi th 1.20 pre-strain 43 Figure 4.9b: Evolut ion of U T S and strain-to-necking for the 7030.60 and 7108.72 alloys wi th 1.20 pre-strain 43 Figure 4.10: Evolut ion of electrical resistivity for the 7030.60 and 7108.72 alloys wi th no pre-strain 44 V l l Figure 4.11: Evolut ion of electrical resistivity for the 7030.60 and 7108.72 alloys wi th 0.03 pre-strain 45 Figure 4.12: Evolut ion of electrical resistivity for the 7030.60 and 7108.72 alloys wi th 0.10 pre-strain 45 Figure 4.13: Evolut ion of electrical resistivity for the 7030.60 and 7108.72 alloys wi th 0.30 pre-strain 46 Figure 4.14: Evolut ion of electrical resistivity for the 7030.60 and 7108.72 alloys wi th 1.20 pre-strain 46 Figure 4.15: Y i e l d stress evolution during the ageing process for the 7030.60 and 7108.72 alloys without pre-deformation. (The dashed line corresponds to the industrial T6 schedule.) 48 Figure 4.16: Y i e l d stress evolution during the ageing process for the 7030.60 alloy wi th different levels of pre-deformation 49 Figure 4.17: Y i e l d stress evolution during the ageing process for the 7108.72 alloy wi th different levels of pre-deformation 49 Figure 4.18: Electrical resistivity evolution during the ageing process for the 7030.60 alloy wi th different levels of pre-deformation 51 Figure 4.19: Electrical resistivity evolution during the ageing process for the 7108.72 alloy wi th different levels of pre-deformation 51 Figure 5.1: Schematic illustration of the flow stress as a function of the obstacle radius r 57 Figure 5.2: Y ie ld stress of the 7030.60 and 7108.72 alloys in the as-deformed condition as a function of pre-strain 60 Figure 5.3: Plots of [(a—CT0)/ M a G b ] 2 vs. strain, illustrating the increases of dislocation density after the pre-deformation for the 7030.60 and 7108.72 alloys 62 Figure 5.4: Changes in yield stress during the 24-hour natural ageing process as a function of pre-strain 63 Figure 5.5: Change i n yield stress between the end of natural ageing and after 1 hour at 100°C 65 viii Figure 5.6: Variation in the peak strength as a function of pre-strain 67 Figure 5.7: TEM-micrographs from a) the non-deformed and b) pre-deformed (0.10 strain) conditions for the 7030.60 alloy aged for 70 hours at 150°C 68 Figure 5.8: The time to peak strength as a function of pre-strain 70 Figure 5.9: T E M micrographs of 7030 alloy wi th 0.1 pre-strain, aged for 5 hours at 100°C plus 20 hours at 150°C. a) dislocations i n contrast and b) dislocations out of contrast, showing the precipitates 71 Figure 5.10: Precipitate size distributions for pre-deformed and non-deformed a) 7108.72 alloy and b) 7030.60 alloy, aged at 100°C for 5 hours followed by 70 hours at 150°C (From Saeter, et. al. [33]) 72 Figure 5.11: Schematic illustration of the size distributions for bulk precipitates and precipitates on dislocations for the pre-deformed materials 73 Figure 5.12: Plot of logarithm of the precipitate size vs. the logarithm of ageing time 75 Figure 5.13: F l o w stress components for the 7030.60 alloy wi th pre-strain of 1.20 in various ageing stages 77 Figure 5.14: Plot of the work hardening rate vs. stress for the 7030.60 alloy wi th no pre-deformation in the as water quenched condition 80 Figure 5.15: Plots of the experimental stress-strain curve and the flow stress contribution from the dislocation hardening for the 7030.60 alloy wi th no pre-deformation in the as water quenched condition 81 Figure 5.16: Plots of the experimental stress-strain curve and the calculated ones using different averaging laws for the non-deformed 7030.60 alloy naturally aged for 24 hours 83 Figure 5.17: Plots of the experimental stress-strain curve and the calculated ones using different averaging laws for the non-deformed 7030.60 alloy aged at 100°C for 1 hour 83 Figure 5.18: Plots of the experimental stress-strain curve and the calculated, ones using different averaging laws for the non-deformed 7030.60 alloy aged at 100°C for 5 hours , 84 ix Figure 5.19: Plots of the experimental stress-strain curve and the calculated ones using different averaging laws for the non-deformed 7030.60 alloy aged at 150°C for 10 hours 84 Figure 5.20: Plots of the experimental stress-strain curve and the calculated ones using different averaging laws for the non-deformed 7030.60 alloy aged at 150°C for 60 days 85 Figure 6.1: Schematic illustration of the extrusion and bumper manufacturing processes 89 Figure 6.2: Schematic illustration of the modified bumper manufacturing process 90 Figure A . l : Hardness distribution through the thickness of the as-received a) 7030.60 and b) 7108.72 plate 95 Figure A . 2 : Ageing curves (in hardness) for 7030.60 alloy wi th different re-solution time at 480°C 96 Figure A . 3 : Ageing curves (in resistivity) for 7030.60 alloy wi th different re-solution time at 480°C 97 Figure A .4 : Ageing curves (in hardness) for 7108.72 alloy wi th different re-solution time at 480°C 97 Figure A . 5 : Ageing curves (in resistivity) for 7108.72 alloy wi th different re-solution time at 480°C 98 Figure A . 6 : Ageing curves (in hardness) for 7030.60 alloy wi th different quenching conditions 99 Figure A . 7 : Ageing curves (in resistivity) for 7030.60 alloy wi th different quenching conditions 99 Figure A . 8 : Ageing curves (in hardness) for 7108.72 alloy wi th different quenching conditions 100 Figure A . 9 : Ageing curves (in resistivity) for 7108.72 alloy wi th different quenching conditions 100 L I S T O F T A B L E S Table 2.1: Compositions of some typical 7xxx aluminum alloys (wt.%) 5 Table 2.2: Typical mechanical properties of some 7xxx aluminum alloys 5 Table 3.1: Chemical composition of the 7030.60 and 7108.72 alloys (wt.%) 25 Table 4.1: Y i e l d stress data for 7030.60 and 7108.72 alloys wi th different levels of pre-deformation in the as-quenched, as-deformed and overaged conditions 38 Table 5.1: Summary of precipitate sizes after ageing at 100°C/5hr . + 150°C /70hr (From Saeter, et. al. [33]) 69 Table 5.2: Summary of precipitate sizes for the non-deformed alloys i n the overaged conditions 74 Table 5.3: Y ie ld stress data for 7030.60 alloy wi th pre-strain of 1.20 in the as-quenched, as-deformed and overaged conditions 75 x i A C K N O W L E D G M E N T S I wish to thank my supervisor, D r . Warren J . Poole, for his instructions and guidance throughout this study. H e has taught me so much, not only what to learn, but also, most importantly, how to learn. It is on his chassis that this thesis can run to an end. I would also like to acknowledge D r . Jan A . Saster and D r . Alexis Deschamps for the fruitful discussion, and many thanks to the group members: Shahrzad Esmaeili, Mar ion Charleux, Partha Ganguly, and Gilles Guiglionda for your friendship and inspiring free discussions. I ' l l remember those exciting group hikes and the good time I had wi th you. Special thanks extend to Gary Lockhart, P.Eng., who had made it easy for me to get accustomed to the surroundings at the beginning, and helped me to overcome many problems a newcomer may face. Being very versatile and handy, he also gave me so much hands-on help wi th the experimental set-ups. In the two years of study in the department of Metals and Materials Engineering, U B C , I find it really a pleasure to be in such a friendly atmosphere. Thanks to everybody's efforts to make it such an enjoyable place. So many good things left i n memory. Finally, I would like to present my gratefulness to my wife, D u y u n L i u , though no words are enough for your love and support that make this happen. A l l the happiness and success as I have, I share them with you. xu Chapter 1 I N T R O D U C T I O N A l u m i n u m alloys have been flying high in the sky for decades wi th many applications found in the aerospace industry. However, in response to the growing concern about environmental issues, more and more aluminum alloys have landed down on the ground, finding their applications in the automotive industry. The most significant advantage of applying aluminum i n automotive applications is the weight saving, not only in the component itself, but also on the other surrounding systems. For example, a light car body made out of aluminum makes it unnecessary to have a very strong and heavy suspension system matching it. Less weight leads to lower fuel consumption and green house gas emissions, and culminates i n a better environment. This advantage can offset the higher cost of aluminum compared to steel, its chief competitor in the automotive industry. A s a result of these considerations, there has been a growing utilization of aluminum alloys in automobiles during recent years. Their applications are diverse and one can find their existence in almost all the structures of an automobile, ranging from the cylinder head, block and pistons, to suspension components, wheels, body panels and bumpers. A s an example, A u d i A . G . has made an all-aluminum car, the A u d i A 8 , where its aluminum body-in-white weighs 40% less than the steel counterpart. The use of aluminum as a bumper material is the subject of this research project. 1 In manufacturing of aluminum bumpers, a heat treatable aluminum alloy is usually selected. The manufacturing process can then be designed to take advantage of the alloy's age hardening potential. Bumpers are usually formed in the solid solution condition (low yield stress and high formability) and then go through a heat treatment to age harden the alloy. In attempt to improve the production process for aluminum bumpers, Hydro-Raufoss (a Norwegian company who supplies aluminum bumpers to many European automotive manufacturers) is currently developing a comprehensive model to predict the final properties of their bumpers. The process model w i l l include finite element method calculations to examine the fabrication process and a microstructurally based mechanical property model for the heat treatment cycle. Preliminary calculations and experimental observations show that there is a large spatial variation in the plastic deformation within the bumper after forming. Furthermore, initial experiments by Hydro-Raufoss suggest that the final yield stress of the alloy actually decreases when the alloy is pre-deformed prior to ageing process. This result is not intuitive and requires further investigation. Therefore, the goal of this work is to examine the relationship between deformation after solution heat treatment and the age-hardening response. The driving force comes not only from the academic curiosity, but also from the industry's desire to improve the performance of the product. O n the basis of this work, as well as results from other research work, a process model that predicts the yield strength of the products after the manufacturing process w i l l ultimately be developed, although it does not fall under the focus of the present study. In order to examine the effect of pre-deformation on the ageing behavior of two 7xxx series aluminum alloys (AA7030.60 and AA7108.72), a set of experiments simulating the industrial bumper fabrication process was conducted. Samples were taken from as-received hot extruded plate and given a solution heat treatment at 480°C for 20 minutes. Then, a pre-determined strain ranging from 0 to 1.2 was imposed on the sample by either 2 uniaxial tension or cold rolling. After the pre-deformation, samples were naturally aged for 24 hours, then they went through a two step artificial ageing processes: 100°C for 5 hours, and 150°C for up to 60 days. The ageing response of these alloys was recorded in terms of yield stress, U T S , strain-to-necking and electrical resistivity properties, by conducting tensile tests and electrical conductivity measurements during various ageing stages. The results from the present study confirm the preliminary studies by Hydro-Raufoss. Deformation after solution heat treatment has a negative effect on the final strength of the material. This is attributed to the effects of dislocations on the ageing behaviour of both 7030.60 and 7108.72 alloys. It is found that the presence of a high dislocation density lowers the natural ageing response of alloys, probably due to the dislocation acting as sinks for quenched in vacancies. The higher dislocation density also accelerates the ageing kinetics by providing short circuit diffusion path due to pipe diffusion along dislocation lines. Moreover, the peak strength of the material is lowered since the presence of dislocations changes the precipitate size distribution so that not all the precipitates can make their full contribution to the strength simultaneously. There is also evidence of static recovery that occurs in the pre-deformed samples during the artificial ageing processes. 3 Chapter 2 BACKGROUND This chapter provides a brief review of the alloys that are used for bumpers and their processing history for this product. This is followed by a literature review of precipitation processes and the strengthening mechanisms of relevance to these alloys. 2.1 7XXX ALUMINUM ALLOYS USED IN AUTOMOTIVE INDUSTRY In Europe, 7xxx series aluminum alloys are widely used for making automotive bumpers. The use of these alloys offers substantial weight savings compared to traditional steel bumpers. The choice of 7xxx series alloys is made due to their potential for producing high final strengths through precipitation hardening. The alloying elements for 7xxx alloys are mainly zinc and magnesium, which form a fine distribution of MgZn2 precipitate under the appropriate heat treatment conditions. Usually copper is also added to increase the resistance to stress corrosion cracking [1], and some alloys have trace elements content such as zirconium [2, 3] or silver which are added to refine the grain structure of the material. The highest strength commercial alloys (usually aerospace alloys) contain larger amounts of zinc and magnesium, with the Zn content in the range of 5-7 wt.%. For lower strength alloys the Zn and Mg contents may be reduced [4]. Table 2.1 and 2.2 show the compositions of some typical 7xxx series aluminum alloys 4 and their mechanical properties, respectively. The alloys for the present study, 7030.60 and 7108.72, have also been included for easy comparison. Table 2.1: Compositions of some typical 7xxx aluminum alloys (wt.%) Alloy Zn Mg Cu Si Fe Mn Cr T i Zr A l 7090 7.3-8.7 2.0-3.0 0.6-1.3 0.12 0.15 - - - - bal. 7475 5.2-6.2 1.9-2.6 1.2-1.9 0.10 0.12 0.06 0.18-0.25 0.06 - bal. 7050 5.7-6.7 1.9-2.6 2.0-2.6 0.12 0.15 0.10 0.04 0.06 0.08-0.15 bal. 7075 5.1-6.1 2.1-2.9 1.2-2.0 0.40 0.50 0.30 0.18-0.35 0.20 - bal. 7009 5.5-6.5 2.1-2.9 0.6-1.3 0.20 0.20 0.10 0.10-0.25 0.20 - bal. 7108 5.45 1.20 0.27 - - - - - 0.16 bal. 7030 5.45 1.22 0.3 - - - - - - bal. 7004 3.8-4.6 1.0-2.0 0.05 0.25 0.35 0.20-0.7 0.05 0.05 0.10-0.20 bal. 7039 3.5-4.5 2.3-3.3 0.1 0.3 0.40 0.10-0.40 0.10 0.10 - bal. Table 2.2: Typical mechanical properties of some 7xxx aluminum alloys Alloy Temper Yield Stress (MPa) UTS (MPa) Elongation (%) 7090 T7E71 580 620 9 7475 T651 560 590 12 7050 T736 510 550 11 7075 T6 500 570 11 7009 T6 470 535 12 7108 T6 431 482 15 7030 T6 388 427 19 7004 T6 340 400 12 7039 T61 345 415 13 5 The most significant strengthening mechanisms normally utilized by the 7xxx series aluminum alloys are precipitation hardening and solid solution hardening. However, effectiveness of precipitation strengthening is much more pronounced than that of the solid solution strengthening [5]. When the material is age hardened, the contribution to the strength from the solid solution strengthening is further reduced, since only a small fraction of the total available solid atoms is retained in the matrix. In some cases, further strengthening from dislocation hardening may be important, particularly when the alloy is formed into the shape of the product by plastic deformation. While it is clear that the increase in strength due to the precipitation hardening during ageing is much more than enough to offset the loss of strength due to the reduced contribution from solid solution, the interaction between precipitation hardening and dislocation hardening is more complicated. 2.2 INDUSTRIAL BUMPER MAKING PROCESS T w o 7xxx series aluminum alloys, 7030.60 and 7108.72, are currently being employed i n industry by the Norwegian company Hydro-Raufoss for bumper manufacturing. Figure 2.1 schematically illustrates the shape of a typical bumper. It has an aluminum body, typically covered by a matching plastic shell. These materials are A l - Z n - M g alloys wi th moderately high peak ageing strength. The processing history for these alloys prior to bumper forming is shown schematically in Figure 2.2. First the alloys are smelted and D C cast into billets. They are then homogenized and hot extruded into C shaped extrusions (i.e. a precursor shape for the final bumper). 6 Figure 2.1: Schematic illustration of a typical European bumper. Temperature Casting Time Figure 2.2: Schematic illustration of the thermal-mechanical history of the 7030.60 and 7108.72 aluminum alloys before bumper forming process. In the bumper forming process, the extrusions are given a solution heat treatment and quench before they are fed directly into the stretch bending line, where the bumpers are 7 formed. The thermo-mechanical process is accomplished as follows: i) the extrusions are heated to 480°C and held for 20 minutes when they are fed on a conveyer belt into a tunnel furnace, ii) a water spray quenches the material upon emerging from the furnace, iii) the extrusions are then immediately formed into the shape of bumper by a series of stretch bending operations. After the forming process, the material is given an ageing treatment consisting of three processes, 1) 24-hour natural ageing (this is the typical value, the actual natural ageing time varies depending on the industrial production schedule, etc.), 2) 100°C/5 hour ageing, 3) 150°C/6 hour ageing. This is the industrial T6 heat treatment schedule, and it is conducted to utilize the age hardening potential of the material to improve the final strength of the product. The whole forming and ageing process is shown in Figure 2.3. It is worth noting that the strain imposed on the plates during stretch bending varies, depending on the shape of the bumper. The material in the corner areas experiences a larger strain, while the material in the centre part of the bumper receives a very small amount of strain. A finite element method analysis predicts strains in the range of 0 -0.30 [6]. This difference in the degree of deformation leads to a different dislocation density and amount of work hardening in the various sections of the bumper. 8 Temperature Solution @480°C/20min. IWater quench Ageing @100°C/5hr. Ageing @150°C/6hr. Stretch bending ^ Natural ageing/Id. I Time Figure 2.3: Schematic illustration of the bumper forming and ageing schedule for the 7030.60 and 7108.72 aluminum alloys. 2.3 A G E H A R D E N I N G The very incentive for the development of 7xxx series alloys is their ability to achieve high mechanical properties through age hardening. The age hardening is the result of the interaction of dislocations with precipitates, which are formed by decomposition of a supersaturated solid solution during ageing process. 2.3.1 T H E H E A T T R E A T M E N T The 7xxx series aluminum alloys are heat-treated in the usual way as other age hardenable aluminum alloys. The first step is homogenization, which is followed by a 9 solution heat treatment, quench and an ageing treatment. Dur ing these treatments, the microstructures of the materials undergo a variety of changes wi th variations of mechanical properties associated wi th the different stages of heat treatment. The homogenization process is employed to even out the distribution of alloying elements in the casting materials by high temperature diffusion of the atoms. It is a necessary process for cast aluminum alloys, however, we w i l l not go further into the details of this process as it falls outside the scope of this work. In the following we w i l l focus mainly on the remaining three stages of heat treatment of the alloys. The goal of the solution heat treatment is to obtain a high level of alloying elements in solid solution. The solution temperature is chosen to fall i n the single phase (solid solution) region i n the phase diagram, to obtain a solid solution. This temperature depends on the chemical composition of a particular alloy. A n important factor that is associated wi th the solution treatment is the formation of an equilibrium vacancy concentration at the solution temperature. Dur ing quenching, i f the cooling rate is sufficiently high, a non-equilibrium vacancy concentration may be produced which can significantly affect subsequent ageing steps. Higher solution temperature favors the higher degree of non-equilibrium vacancy concentration that can be obtained by quenching. However, undesired local melting of the alloy due to the segregation around second phase particles may occur if the solution temperature is too high (close to the solidus temperature line). The solution time is usually short, i n the order of minutes, compared wi th homogenization times in hours, since homogenization relies upon long range diffusion to even out the distribution of these alloying elements [7]. O n the other hand, the solution treatment only requires short range diffusion to dissolve precipitates formed during previous processes. After solution heat treatment, the material w i l l undergo a quenching process, which brings it from solution temperature to a predetermined lower temperature. A s shown 10 schematically i n Figure 2.4, the alloy in solid solution region (point A ) goes into a multiphase region (point B) during the quenching process. The solid solution at high temperature is "frozen" in position thus a supersaturated solid solution (SSS) is obtained. This is essential to the following ageing process. After quenching, the material is in a non-equilibrium state, and there is a strong driving force for it to reach equilibrium conditions by the change from a single phase to a multiphase structure. This process occurs during ageing, either naturally (i.e. at room temperature) or artificially (i.e. at elevated temperatures). Often this process is very complicated and involves the formation of Guinier-Preston (G-P) zones, clusters, and formation of a series or sequence of metastable phases such as intermetallic precipitates. Furthermore, the precipitate composition and crystal structure often go through a transition during ageing. These phenomena w i l l be discussed in the following sections. Temperature Composition Figure 2.4: Schematic illustration of the route of phase transformation during quench. 11 2.3.2 PRECIPITATION PROCESS After quenching, the system is frozen into a supersaturated solid solution condition that is in a high free energy state and is therefore unstable. Thus, driven by the nature of the material to stay on the lowest energy level, the material undergoes a series of changes in microstructures and consequently, mechanical properties, during the various stages of heat treatment. The generally accepted sequence of precipitation in 7xxx series aluminum alloys is [8, 9]: SSS -> G-P zones -> n ' -> q( M g Z n ^ where r\' is the metastable transition phase, and n is the equilibrium M g Z n 2 phase. The structure of the G-P zones is not fully understood. However, their role in determining the density of nuclei are thought to be extremely important. The precipitation kinetics usually display the classical behavior of nucleation and growth, followed by cubic coarsening, as Guyot and Cottignies [10] stated i n their study using 7050 aluminum alloy as an example. The decomposition of a supersaturated solution can occur even during the quenching process, as soon as the temperature drops below the solute solvus temperature [11, 12]. Usually, this leads to a rather coarse distribution of precipitates (often on grain boundaries or dispersoids), which adversely affects subsequent hardening. This is one of the reasons why the quenching rate, as well as the rate of heating up to ageing temperature (heating ramp), can affect the size and distribution of the precipitates, as we w i l l discuss later. 12 2.3.2.1 FORMATION OF G-P ZONES The first result when ageing at low temperature (below the G-P zone solvus temperature, Tc) is the formation of the G-P zones. The G-P zones are very small in their initial stage, and apparently, nucleate homogeneously. They have a size in the range of 0.5-1.0 nanometers, which is beyond the capability of resolution of the conventional T E M . However, their existence can be traced by the change in the hardness/yield stress and resistivity of the alloy [13], since these material properties are very sensitive to the microstructure of the material. Direct observation of G-P zones is now possible using Field Ion Microscopy or H R T E M (High Resolution Transmission Electronic Microscope). Fo r example, G-P zones have been directly observed by Mukhopadhyay, et al. [14]. These observations have shown that the shape of G-P zones in A l - Z n - M g is spherical in their initial stages. A s the Z n atom is smaller than A l atom, and M g is larger, the atomic size misfit can be compensated by the formation of clusters containing both Z n and M g atoms resulting i n stress-free spherical shaped coherent clusters, to reduce the free energy of the system. Most researchers agree that there are two types of precipitate nuclei: solute rich clusters that form during low temperature ageing [15], and vacancy rich clusters that form during quenching [16]. The solute rich clusters dissolve partially or completely during artificial ageing, depending on temperature. The vacancy rich clusters are thought to be more stable and may act as nucleation sites for the metastable phase r| ' at appropriate ageing temperatures [17]. It had once been considered that only the solute supersaturation should control the nucleation of the precipitates. However, the distribution of precipitates i n alloys of similar solute supersaturations is frequently very different, and in many alloys the 13 distribution is sensitive to the quench rate, the time of natural ageing, the rate of heating to the ageing temperature and the presence of trace elements, none of which appreciably affect the solute supersaturation. A s a result, other factors must be taken into consideration. A s the research into precipitation hardening systems proceeded, the role of quenched-in vacancies on the nucleation of precipitates was recognized [18]. In fact, the precipitate distribution in A l - Z n - M g alloys is largely controlled by the distribution and concentration of lattice defects, particularly vacancies, as well as the dislocations and grain boundaries. Al though traditional T E M is not applicable for direct observation of the nucleation process since point defects in the crystal lattice are not visible in this technique, information on the role of vacancies can be still obtained by a detailed study of the microstructure of the alloy after various heat treatments. After quenching, a typical aluminum alloy has a vacancy supersaturation of the order of 10 1 0 times higher than equilibrium, i.e. much greater than the solute supersaturation. They tend to precipitate as loops, helices and possibly small clusters or they may be absorbed into vacancy sinks such as grain boundaries or dislocations. The depletion of vacancies around these defects can lead to the formation of precipitate free zones. B y gathering information on the form and number of precipitates in these regions, the role of vacancies in the precipitation process can be revealed [18, 19]. 2.3.2.2 FORMATION OF r)' Further ageing causes the precipitation of particles in the shape of th in plates. This is a transition phase, r\', and can be identified as having a hexagonal structure wi th lattice parameters a=0.496 nm and c= 1.403 nm, as Park and Arde l l [20] reported i n their study of 7075 aluminum alloy i n the peak aged T651 and overaged T7 tempers. They also 14 observed that the peak strength mainly comes from the contribution of these small particles, which are finely dispersed. Thomas and Nut t ing [21] also confirmed that the n ' precipitates are responsible for the maximum hardness when ageing at 130°C. Some researchers have suggested that the n ' transition phases would form on the basis of initial G-P zones [8], while others reported that on increasing the ageing temperature to an intermediate range (100-130°C), the coherent n ' transition phase was found to form without evidence of prior G-P zone formation [22]. Details of heat treatment conditions should therefore be considered to determine the actual precipitation behaviour. The role of G-P zones on the precipitation of the n ' precipitates has received considerable attention. A useful approach to this question is to consider the metastable solvus boundary for G-P zones. A critical temperature is reached where G-P zone formation is no longer favorable. However, at lower temperatures, G-P zones may form wi th smaller zone sizes being stable at lower temperatures. The next critical question is what w i l l happen to these zones when the temperature is raised. G-P zones formed at a lower temperature w i l l dissolve on heating to a higher temperature if they are smaller than a critical size but w i l l transform to the intermediate precipitate at higher temperature if they are larger than a critical size. Whichever being the case, the nucleation stage is of most importance in controlling the precipitate distribution, and hence, the final strength of the material. The heat treatment after the formation of G-P zones can affect the final properties very significantly. Staley [23] found in his research of copper bearing 7050 alloy that the G-P zones formed during natural ageing dissolve on subsequent exposure at ageing temperature above a critical temperature (the G-P zone solvus) if the rate of heating to the elevated temperature is high. The crystalline precipitate initiates heterogeneously under these conditions (on grain boundaries and dislocations), and the strength is low. W i t h lower heating rates many of the G-P zones nucleated during natural ageing do not 15 dissolve at the elevated temperature. They in fact can grow to larger sizes which are stable at higher temperatures, and can therefore serve as nuclei for the crystalline precipitate. This provides a large number of very fine particles and increases the strength attainable by a treatment at that ageing temperature. 2.3.2.3 FORMATION OF r\ The last stage of ageing for the 7xxx aluminum alloys is the formation of the equilibrium crystalline r| phase, M g Z n 2 , and the subsequent coarsening of the n precipitates. The crystal structure of the n phase is hexagonal (C14) wi th a = 0.520nm and c = 0.862 nm [18]. It has a lath like morphology. The n precipitates can be transformed from the metastable plate shaped TI ' phase. In some cases, it can also directly nucleate at dislocations [24]. The formation and coarsening of r\ phase occur in the overageing stage, in which the strength of the material is decreased from the peak strength. Therefore, it is of less interest to the industry, where high strength is usually desirable. 2.3.2.4 PRECIPITATION KINETICS Guyot et al. [10] suggested that the precipitation kinetics for an A l - Z n - M g - C u alloy follows the classical behavior of nucleation, growth and coarsening. The nucleation and growth periods are difficult to quantify, due to the existence of transients occurring 16 before the stabilization of the ageing temperature. However, the volume fraction of the precipitates seems to follow a Johnson-Mehl-Avrami law wi th a time exponent n of ~1.3, which is close to the theoretical value of 1.5 expected for the growth by long range diffusion of a constant number of precipitates. The coarsening, or Ostwald ripening occurs at a nearly constant precipitate volume fraction. 2.3.3 DISLOCATION-PRECIPITATE INTERACTION The plastic deformation of metallic materials occurs by the motion of dislocations. In the absence of dislocations, the strength of a material would be very high (i.e. G/20). Practically, it is not possible to remove all dislocations and therefore strengthening requires that obstacles be added which make the dislocation motion more difficult. It is well established that the age hardening of the material results from the interaction between precipitates and mobile dislocations. The theory was developed i n the 1950's and 1960's, and a wealth of literature on this subject is available. Several excellent reviews may be found including those by Ke l ly and Nicholson [25] and by Embury [26]. Although we have a solid theoretical basis for the origin of the age hardening behaviour, the details of the precipitate-dislocation interaction is still of considerable interest. However, i n most cases, it is difficult to quantify the exact nature of the precipitate-dislocation interaction. Generally, the presence of pre-precipitates (G-P zones or clusters) and precipitates produced by the ageing process described in the previous section would increase the stress required for dislocations to pass through them, thus increasing the flow stress of 17 the plastic deformation. The resistance for dislocations to pass the obstacles may arise from several factors. First, when the particles are small, such as the clusters formed at the beginning of the ageing process, they can be sheared by dislocations. The resistance to shearing derives from a number of possible sources including chemical strengthening, modulus strengthening, coherency strain strengthening, and stacking fault energy strengthening [25]. The resulting flow stress contribution generally has the form: shearable = q f V (2.1) where f is the volume fraction of the shearable particles, r the particle radius, cu m and n are constants, and for most dislocation-particle interactions, both m and n have the value of 0.5 [27]. The dislocation interaction wi th shearable particles is shown schematically in Figure 2.5 (1) - (3). When the particles grow larger as the ageing process advances, they become stronger obstacles and can no longer be sheared by dislocations. Instead, it becomes easier for the dislocations to bypass the precipitate, leaving a dislocation loop around the unbreakable particle (Figure 2.5 (i) - (iii)). This also gives rise to the flow stress of the dislocation slip. The flow stress contribution from this mechanism has the form: CT ~ c z (2.2) non-snearable 2 L where q is a constant, G the shear modulus, b the Burgers vector, and L the particle spacing. L can be expressed in the following way: 18 3) iii) Figure 2.5: Schematic illustration of dislocation-particle interactions. (1) - (3): moving dislocation shearing a weak particle; (i) - (iii): moving dislocation bypassing a strong particle. 19 L = c3 ^ (2-3) where c 3 is a constant, r is the radius of the particle, f the volume fraction of the particle. Substituting Equation 2.3 into Equation 2.2 gives: Gbf 1 / 2 (2.4) non-shearable r where c 4 contains all the constants in Equation 2.2 and 2.3. Foreman and M a k i n [25] studied the dislocation-particle interaction i n a more complicated situation by using computer simulations, where a random array of particles of varying strength is presented. Each particle is characterized by its breaking angle cpc, illustrated schematically in Figure 2.6, which is the angle of the two arms of dislocation line when it overcomes the particle and advances to new obstacles. Thus the strength of a particle can be expressed i n its cpc value. A weak particle has a (pc « 180°, while a strong particle has a cpc « 0°. The results from the simulations showed that the flow stress can be expressed as follows: where G , b and L have the same meaning as in Equation 2.2. For strong particles whose critical breaking angle cpc « 0°, Equation 2.5 reduces to Equation 2.2 (with a factor of 0.8, which is related to the difference between the random (9C < 100°) (2.5) (cpc > 100°) (2.6) 20 distribution and square arrangement of the particles), which gives the flow stress for non-shearable particles. Figure 2.6: Schematic illustration of the breaking angle cpc for a dislocation holding by an obstacle. 2.4 T H E EFFECT O F PRE-DEFORMATION O N T H E PRECIPITATION BEHAVIOR In industrial production of automobile bumpers using 7xxx series aluminum alloys, the material undergo a stretching process, where a strain of up to 0.3 is applied to the material at room temperature in the as-quenched condition. It is therefore necessary to consider the effect of this deformation on the age hardening response of the 7xxx series aluminum alloys. Pre-deformation prior to ageing can affect the precipitation behaviour in several ways. C o l d deformation increases the dislocation density of the material significantly. This increased dislocation density can have two main effects: 21 i) dislocations can act as a preferential nucleation sites for precipitates ii) enhanced growth and coarsening of precipitates can occur due to the high rates of solute diffusion along dislocations. The role of the increased dislocation density introduced by the pre-deformation on the ageing behaviour varies for different aluminum alloys series [5]. L o w levels of pre-strain (less than strains of 0.1) have only a minimal effect on the yield strength in T6 tempers and negative effect on that in T7 tempers for 7xxx series alloys, while the reverse is observed in the behavior of 2xxx alloys. In research on the influence of dislocations on precipitation in an Al-6.1%Zn-2.35%Mg (wt.%) alloy, Deschamps et al. [28] found that the pre-deformation decreases the overall mechanical properties attained by precipitation strengthening, particularly at peak hardness as well as during the coarsening stage. Deschamps et. al. also observed changes in the precipitation sequence. The change in precipitation sequences was attributed to the decrease of the activation energy for the formation of the stable n phase on dislocations, due to the relaxation of elastic misfit in the presence of the dislocation. It also introduces spatial heterogeneity of precipitates and wide precipitate size distribution, which were proposed to be related to two effects: first, the dislocations introduced after quenching act as vacancy sinks, decreasing the precipitation rate of G-P zones at room temperature. The low level of G-P zones near dislocations then makes subsequent formation of n ' and r\ phases more difficult (A similar effect as i n classical precipitate free zone near grain boundaries). The authors attributed the lowering of the peak hardness to a smaller solute content available for homogeneous finely dispersed precipitation and to a broader precipitate size distribution. Deschamps and Brechet [29] have theoretically studied the effect of precipitate size distribution and have shown that as the size distribution widens, the peak strength decreases, as shown i n Figure 2.7. 22 0.01 0.1 1 10 100 1000 Time (h) Figure 2.7: Influence of the precipitate size distribution width A on the evolution of yield stress during ageing. (From Deschamps and Brechet [29].) 2.5 THE STATEMENT OF OBJECTIVES The objective of this study is to obtain better understanding of the effect of pre-deformation on the ageing behaviour and mechanical properties (yield stress, in particular) of 7030.60 and 7108.72 aluminum alloys. We hope that this effort w i l l help the industry to improve the performance of the aluminum bumper product, and ultimately, help to build a better environment. Having reviewed the basic mechanisms involved in precipitation hardening alloys, the following chapters w i l l describe the experimental techniques and results and then discuss the results. 23 Chapter 3 E X P E R I M E N T A L P R O C E D U R E 3.1 INTRODUCTION The general approach that has been taken to study this problem is to simulate the industrial process i n the laboratory. Samples were taken from the as-extruded plate and then processed in a similar manner to the industrial process described in Section 2.2. However, the samples were pre-deformed by tensile loading or rolling in the laboratory, instead of stretch bending which is used i n the industrial process. Different levels of pre-deformation w i l l be imposed on the samples to study the response of the material behavior. 3.2 MATERIAL The aluminum alloys involved in this study were received from the Hydro-Raufoss Automotive Research Centre in the form of hot extruded plate of dimension 150x750x5 mm. They were taken from the production line and cut to length. T w o alloys (7030.60 and 7108.72) were examined and their chemical composition is shown in Table 3.1. Note that the major difference between these two alloys is the intentional addition of zirconium to the 7108.72 alloy. The effect of Z r on grain structure w i l l be discussed later 24 in Chapter 5. Table 3.1. Chemical composition of the 7030.60 and 7108.72 alloys (wt.%) A l l o y Z n M g C u Z r A l 7030.60 5.45 1.22 0.3 - bal. 7108.72 5.45 1.20 0.27 0.16 bal. 3.3 METALLOGRAPHIC PREPARATION The metallographic examinations were carried out to observe the grain structure of 7030.60 and 7108.72 alloys. For each alloy, three samples representing the normal, extrusion and transverse directions of the extruded plate were taken by cutting from the as-received plates using a S iC cutoff wheel wi th l iquid cooling. The samples were mounted in the L E C O S E T 100 polymer resin, and then they went through a mechanical grinding and polishing process to obtain smooth surfaces. The last polishing step was done using a 0.06 p m colloidal A l 2 0 3 powder on a polishing cloth. The etchant used to reveal the grain structure was Graff and Sargent's etchant, which has the following composition: H N 0 3 15.5 m l H F 0.5 m l C r 0 3 3 g H 2 0 84 m l The samples were immersed in the solution of the etching reagents for 80 seconds wi th 25 mild agitation, then rinsed and dried for the optical microscopic observation. The Unimet U N I T R O N 8919 microscope and the Polaroid 55 black and white instant sheet fi lm were employed to take the pictures of the grain structure. 3.4 EXPERIMENTAL HEAT AND MECHANICAL TREATMENT Following the industrial route described in Section 2.2, the experiments were carried out so that the materials would experience a similar heat and mechanical treatment. The experimental schedule is shown schematically in Figure 3.1. Temperature Solutionizing @480°C/20min. IWater quench t- - - Ageing @100°C/5hr. t .' Ageing @150°C/up to 60days Pre-deformation V f ^, Natural ageing/24hr./ J T i m e Figure 3.1: Schematic illustration of the experiment schedule for the 7030.60 and 7108.72 aluminum alloys. The samples from both alloys were solution heat treated at 480°C for 20 minutes, followed by water quench (water temperature was ~20°C). The samples were then pre-26 deformed to a pre-determined strain. A l l the pre-deformations were finished wi th in 15 minutes after water quench. After the pre-deformation, the samples were aged for 24 hours at room temperature (20°C) followed by a two step artificial ageing procedure which consisted of 100°C for 5 hours (in a boiling water bath) and 150°C for up to 60 days in an air furnace. Various levels of pre-deformation (0, 0.03, 0.10, 0.30 and 1.20) were imposed on the samples of both the 7030.60 and 7108.72 alloys to study its effect on the ageing behavior. For the pre-deformation levels of 0.03 and 0.10 strains, the samples were taken along the extrusion direction of the as-received plates. The tensile and resistivity samples were prepared. These samples were elongated using a M T S (100 K N frame) servo-hydraulic testing machine, at a strain rate of 0.02 s"1, to give the designated pre-strain. The deformation was measured using an extensometer on the sample. For the higher pre-deformation levels of 0.30 and 1.20 strain, small pieces of specimen (30x100x5 mm) were sheared from the as-received plate. They were rolled in a small-scale laboratory rolling mi l l , wi th the rolling direction parallel to the extrusion direction of the original plate, at an estimated strain rate of 7.7 s'1. It is worth noting that the samples experienced a mi ld temperature rise (i.e. the sample temperature rose to 60-70°C) when they were rolled through the mi l l . After rolling, tensile samples were punched out from the rolled material using a die. 3.5 PROPERTY MEASUREMENTS The ageing behavior of the alloys studied was measured by using hardness, resistivity, 27 and tensile tests, which were conducted on the samples at various stages in the ageing process. The hardness property of the alloys was measured using the M I C R O M E T 3 micro hardness tester wi th the pyramidal diamond indentor and a test load of 1 kgf. The electrical resistivity of the alloys for various ageing times was also measured in order to study the microstructure change. This was done by conducting measurements of the electrical conductivity of the samples, which had gone through the thermal-mechanical treatments described i n Section 3.3, using a Verimet 4900C eddy current probe, and converting the conductivity data into resistivity values. The tensile tests were conducted either on an Instron screw driven tensile test machine or on the above mentioned M T S testing machine at a nominal strain rate of 1.5x1c4 s'1. T w o different sizes of tensile sample were used as shown in Figure 3.2. The width and thickness of the gauge section area of each tensile sample were measured using a micrometer before the test, so that the value of the initial gauge section area could be obtained. A n extensometer was attached to the specimen during each tensile test, and the displacement-load data were acquired and stored in a data file, from which the engineering stress and strain, true stress and strain data, and, upon on further data analysis, the yield stress data, can be obtained. The engineering strain, engineering stress, true strain, and true stress were calculated using the following equations. 51 (3-1) •eng. 0 C7, _ _ F o (3.2) t^rue — l n 0 + Seng.) (3.3) tftrue = CTeng.*ln(l + S e ng.) (3-4) 28 where 81 is the elongation, 10 the gauge length, F the tensile load, and AQ the initial gauge section area. The yield stress value was determined using the 0.2% offset method. 12.5 mm 55 mm 155 mm 4.5 mm - ' J 5.3 mm 20 mm A « • 70 mm • b Figure 3.2 Schematic illustration of two types tensile sample used i n this research, a) for M T S tensile machine and b) for Instron tensile machine. In summary, the industrial process has been simulated in the laboratory. The evolution of properties was measured wi th tensile tests and electrical resistivity. In the next chapter, the main results from these experiments w i l l be summarized. 29 Chapter 4 R E S U L T S The experimental results, including the grain structure examinations and ageing curves (in terms of yield stress, U T S , strain-to-necking and electrical resistivity) for both 7030.60 and 7108.72 alloys wi th different levels of pre-deformation, are given in this section. Several important observations are also summarized. Some preliminary experimental work was conducted on the as-received, hot-extruded plates of the 7030.60 and 7108.72 alloys to obtain the basic characteristics of these materials, such as the homogeneity of the strength through the plate thickness, the sensitivity to solution time and the quench sensitivity. These results, which helped to design the schedule for experiments thereafter, are given in the Appendix. 4.1 GRAIN STRUCTURE OF THE AS-EXTRUDED MATERIAL The grain structure of these two alloys (7030.60 and 7108.72) i n the as-extruded condition is shown in Figures 4.1 and 4.2. They show very different grain morphologies and grain sizes, although these alloys have been processed wi th the same thermal-mechanical treatment (see Figure 2.2 in Section 2.2). The 7030.60 alloy has an equiaxed grain structure (with an average grain size of 60 pm), indicating that the material recrystallized after hot extrusion. The 7108.72 alloy has a elongated, fibrous grain structure (with an average size of 5 p m in the transverse direction and 20.5 p m i n the longitudinal direction), a remnant of the deformation structure, wi th no evidence of recrystallization. 30 Figure 4.1: Tri-planar optical microscopic graphs showing the grain structure of the 7030.60 alloy in the as-received condition. 31 Figure 4.2: Tri-planar optical microscopic graphs showing the grain structure of 7108.72 alloy in the as-received condition. 32 Obviously the grain size for the 7108.72 alloy is much smaller than that of the 7030.60 alloy. The role of Z r in the 7108.72 alloy is undoubtedly critical in making this difference, as w i l l be discussed later. 4.2 AGEING BEHAVIOUR During the ageing process, the evolution of the microstructure in the alloys is reflected by changes in the microstructure sensitive properties, such as mechanical properties (yield stress and U T S ) and electrical resistivity. The ageing behaviour of the 7030.60 and 7108.72 alloys was studied by experimentally measuring the ageing curves in terms of mechanical properties (yield stress, U T S and strain-to-necking) and electrical resistivity. The results are described in the following sections. 4.2.1 EVOLUTION OF MECHANICAL PROPERTIES DURING AGEING The mechanical properties are among the most important material characteristics as that they affect both the production process and the final service properties for the bumpers. The ageing behaviour of these alloys was examined by studying the evolution of mechanical properties (yield stress, U T S and strain-to-necking) of these alloys wi th different levels of pre-deformation after the solution heat treatment. Tensile tests were conducted on samples in various stages of ageing: 1) as-quenched, 2) after the pre-deformation (as deformed), 33 3) after natural ageing for 24 hours, 4) during ageing at 100°C, 5) during ageing at 150°C. 4.2.1.1 TYPICAL TENSILE CURVES Figure 4.3 shows some typical engineering stress-strain curves for the 7030.60 alloy wi th no pre-strain. A number of important observations can be obtained from this Figure. 1) the yield stress increases during the natural ageing and artificial ageing up to 10 hours at 150°C (Figure 4.3a), after which it decreases (Figure 4.3b). 2) serrations are observed in the engineering stress-strain curves for the samples in as-quenched, naturally aged and artificially aged (100°C/1 hr.) conditions. However, no serrations are observed in the samples artificially aged for longer times. 3) qualitative changes in work hardening behaviour as a function of ageing time can be observed. In particular, the work hardening rate decreases as the strength increases. 4) i n the overaged sample (Figure 4.3b curve c), an unusual stress-strain curve was observed. The initial work hardening rate is low and then there is an increase in the work hardening rate after a strain of around 1%. Figure 4.4 gives some typical engineering stress-strain curves for the 7030.60 alloy wi th a pre-strain of 0.30. While observations are similar to those in the non-deformed cases, there are a number of differences between these two pre-deformation conditions. 34 0.05 0.1 Strain 0.15 0.2 - as water-k quenched — 24 hr. natural c a 9 e i n 9 — 1 hr. at100°C -5hr. at100°C •10 hr. at150°C (peak) Figure 4.3a: Engineering stress-strain curves for 7030.60 alloy wi th no pre-strain, in various stages of ageing t i l l the peak. 0 0.05 0.1 0.15 0.2 Stra in Figure 4.3b: Engineering stress-strain curves for 7030.60 alloy wi th no pre-strain, in various stages of ageing after the peak. 35 450 -400 350 (0 300 CL 250 U) CO <D 200 \ W 150 100 50 ' 0 -• as water-quenched - as rolled • 24 hr. natural ageing •1 hr. at100°C •5hr. at150°C (peak) Figure 4.4a: Engineering stress-strain curves for 7030.60 alloy wi th a pre-strain of 0.30, in various stages of ageing t i l l the peak. -5hr. at150°C (peak) -70 hr. at150°C -60 days at 150°C Figure 4.4b: Engineering stress-strain curves for 7030.60 alloy wi th a pre-strain of 0.30, in various stages of ageing after the peak. 36 1) stepped serrations occur in the engineering stress-strain curves for the samples in 24 hour natural ageing condition and 1 h o u r / 1 0 0 ° C artificial ageing condition (Figure 4.4a). 2) a substantial change in work hardening behaviour from the as-quenched condition to the as-deformed condition. 3) relatively large post-necking elongation in the overaged samples (Figure 4.4b). 4.2.1.2 AGEING CURVES IN YIELD STRESS, UTS AND STRAIN-TO-NECKING In this section, the results of mechanical properties evolution for these alloys w i l l be displayed. The ageing curves in yield stress, U T S and strain-to-necking are plotted for each pre-stain level. Figure 4.5a shows the yield stress evolution for these alloys without pre-deformation during two step artificial ageing. Generally, the yield stress of the alloys first increases and reaches a peak, then it begins to decrease. The same trend is observed for the evolution of U T S , as shown in Figure 4.5b. It is also shown in Figure 4.5b that the strain-to-necking decreases during the second step of artificial ageing. It is interesting to note that while the yield stresses for the non-deformed alloys run parallel, they converge at 70 hours of artificial ageing at 150°C (see Figure 4.5a). This is observed only in the non-deformed alloys. Figure 4.6a shows the yield stress evolution for the alloys wi th a pre-strain of 0.03. Similar to the curves shown i n Figure 4.5a, single-peak ageing curves are observed. However, the yield stresses for these pre-deformed alloys run at a nearly constant offset, and the yield stress for the 7108.72 alloy is consistently higher that the 7030.60 alloy 37 throughout the ageing process. Figure 4.6b gives the evolution of U T S and strain-to-necking for the pre-deformed alloys (0.03 pre-strain). The U T S follows the same trend as the yield stress in Figure 4.6a. The strain-to-necking increases slightly i n the first stage of artificial ageing, then it decreases in the rest of the ageing period. The mechanical properties for the alloys wi th pre-strain of 0.10, 0.30 and 1.20 are given in Figure 4.7, Figure 4.8 and Figure 4.9, respectively. Similar behaviour is observed for these alloys pre-deformed to larger levels, although the magnitudes of each curve are different. A s the pre-strain level increases, the magnitude of the changes in the yield stress and U T S becomes less, and the strain-to-necking also decreases. A detailed comparison between the behaviour of materials wi th different levels of pre-strain w i l l be summarized later. Table 4.1 gives the yield stress data for 7030.60 and 7108.72 alloys wi th different levels of pre-deformation i n the as-quenched, as-deformed and overaged (60 days at 150°C) conditions. Table 4.1: Y ie ld stress data for 7030.60 and 7108.72 alloys wi th different levels of pre-deformation in the as-quenched, as-deformed and overaged conditions. Al loy Pre-strain Yield stress (MPa) Yield stress (MPa) Yield stress (MPa) As-quenched As-deformed Overaged (60 d.) 7030.60 0 60 / 182 0.03 60 127 194 0.10 60 184 154 0.30 60 223 164 1.20 60 300 191 7108.72 0 106 / 193 0.03 106 165 217 0.10 106 235 188 0.30 106 255 188 1.20 106 346 211 38 500 450 400 350 300 250 200 150 100 50 0 -•-7108.72, yield stress -•-7030.60, yield stress ageing at ageing at • 100°C 150°C no pre-strain 10 100 Artificial Ageing Time / Hr. 1000 10000 Figure 4.5a: Evolution of yield stress for the 7030.60 and 7108.72 alloys with no pre-strain. 1 10 100 1000 10000 Artificial Ageing Time / Hr. Figure 4.5b: Evolution of UTS and strain-to-necking for the 7030.60 and 7108.72 alloys with no pre-strain. 39 500 450 400 350 ^ 300 « 250 £ 200 ro CL CO 150 100 50 0 -• -7108.72, yield stress —•—7030.60, yield stress i—— ageing at ageing at 100°C 150°C 0.03 pre-strain 10 100 Artificial Ageing Time / Hr. 1000 10000 Figure 4.6a: Evolution of yield stress for the 7030.60 and 7108.72 alloys with 0.03 pre-strain. ro w (A <D co 10 100 1000 Artificial Ageing Time / Hr. 10000 Figure 4.6b: Evolution of UTS and strain-to-necking for the 7030.60 and 7108.72 alloys with 0.03 pre-strain. 40 500 450 400 350 300 250 200 150 100 50 0 -•-7108.72, yield stress —•—7030.60, yield stress ageing at ageing at 0.10 pre-strain " 100°C 150°C 10 100 Artificial Ageing Time / Hr. 1000 10000 Figure 4.7a: Evolut ion of yield stress for the 7030.60 and 7108.72 alloys wi th 0.10 pre-strain. ro CL C O C O CO 10 100 1000 Artificial AgeingTime / Hr. 10000 Figure 4.7b: Evolut ion of U T S and strain-to-necking for the 7030.60 and 7108.72 alloys wi th 0.10 pre-strain. 41 500 450 400 350 300 250 200 150 100 50 0 -•-7108.72, yield stress —•—7030.60, yield stress ageing at ageing at 0.30 pre-strain 100°C 150°C 10 100 Artif icial A g e i n g T i m e / Hr. 1000 10000 Figure 4.8a: Evolution of yield stress for the 7030.60 and 7108.72 alloys with 0.30 pre-strain. 7108.72, UTS 7030.60, UTS 7108.72, strain-to-necking 7030.60, strain-to-necking 0.30 pre-strain 0.4 0.35 0.3 0.25 c 0.2 '5 CO 0.15 0.1 0.05 100 1000 Artif icial A g e i n g T i m e / Hr. 10000 Figure 4.8b: Evolution of UTS and strain-to-necking for the 7030.60 and 7108.72 alloys with 0.30 pre-strain. 42 500 450 400 350 300 250 200 150 100 50 0 -• -7108.72, yield stress -• -7030.60, yield stress ageing at ageing at " 100°C 150°C 1.20 pre-strain 10 100 Artificial Ageing Time / Hr. 1000 10000 Figure 4.9a: Evolution of yield stress for the 7030.60 and 7108.72 alloys with 1.20 pre-strain. 1 10 100 1000 10000 Artificial Ageing Time / Hr. Figure 4.9b: Evolution of UTS and strain-to-necking for the 7030.60 and 7108.72 alloys with 1.20 pre-strain. 43 4.2.2 EVOLUTION OF ELECTRICAL RESISTIVITY DURING AGEING T h e evolut ion of electrical resistivity was also measured as another method to study the ageing behaviour of the materials, since it is also a microstructure dependent parameter for the materials. T h e results are compi led and plotted i n Figures 4.10 - 4.14. In Figure 4.10, it is observed that the electrical resistivity for the non-deformed alloys decreases slightly in the 100°C ageing stage. In the fo l lowing 150°C ageing stage, it decreases more rapidly. T h e electrical resistivity of 7108.72 al loy runs above the 7030 al loy in the beginning, then they come close at 20-70 hours of artificial ageing at 150°C, and finally, they separate again u p o n further ageing. E d o> O X J > C O 'co CD 48 46 44 42 40 38 36 34 32 1 — - » - 7108.72 • • 1 > — » -•-7030.60 ageing at 100°C ageing at 150°C 1 i i i i i , i no pre-strain 10 100 Artificial Ageing Time / Hr. 1000 10000 Figure 4.10: Evo lu t i on of electrical resistivity for the 7030.60 and 7108.72 alloys w i th no pre-strain. F o r the alloys w i th a pre-strain of 0.03, the electrical resistivity runs almost parallel whi le decreasing i n a similar manner as the non-deformed alloys, as shown i n Figure 4.11. T h e same trend is also observed for the alloys w i th 0.10, 0.30 and 1.20 pre-strain, as shown in 44 Figure 4.12, 4.13 and 4.14, respectively. It is worth noting that the rate at which the electrical resistivity decreases is much higher for the alloys with 1.20 pre-strain. 10 100 Artificial Ageing Time / Hr. 1000 10000 Figure 4.11: Evolution of electrical resistivity for the 7030.60 and 7108.72 alloys with 0.03 pre-strain. E d > cn 'c/> CD CC 10 100 Artificial Ageing Time / Hr. 1000 10000 Figure 4.12: Evolution of electrical resistivity for the 7030.60 and 7108.72 alloys with 0.10 pre-strain. 45 E d > to 'tn or 10 100 Artificial Ageing Time / Hr. 1000 10000 Figure 4.13: Evolution of electrical resistivity for the 7030.60 and 7108.72 alloys with 0.30 pre-strain. E d 0> O > or 48 46 44 42 40 38 36 34 32 - • - 7 1 0 8 . 7 2 II • 1 it * 1 - • - 7 0 3 0 . 6 0 ageing at ageing at ^ 1 ~rZr 1.20 pre-strain _^  ,— i 1 1 — 100°C 150°C 10 100 Artificial Ageing Time / Hr. 1000 10000 Figure 4.14: Evolution of electrical resistivity for the 7030.60 and 7108.72 alloys with 1.20 pre-strain. 46 4.3 SUMMARY OF SOME SIGNIFICANT OBSERVATONS • The grain structure of the as-received (hot-extruded) 7030.60 alloy is a recrystallized structure, while it is a unrecrystallized, fibrous structure for the 7108.72 alloy. • Generally, the yield stress and U T S for both alloys increase during ageing unti l they reach a peak, after which decrease upon further ageing. • For the alloys wi th pre-deformation, both show a similar ageing behaviour in the evolution of mechanical properties. Their ageing curves run parallel to each other. • For the alloys without pre-deformation, a difference in yield stress evolution was observed for the two alloys during the overageing region. The yield stress first converges at 70 hours of artificial ageing at 150°C and then at longer ageing times separates again. Similar behaviour can be observed in the evolution of electrical resistivity for these alloys. Figure 4.15 summarizes the ageing curves (for ageing times of up to 70 hours at 150°C) for the solution heat treated and water quenched 7030.60 and 7108.72 alloys without pre-deformation, showing the evolution of yield stress. The industrial standard T6 heat treatment schedule is also located on the plot. This is the baseline wi th which the other results w i l l be compared. The first data point in the plot stands for the yield stress of the alloy in the as-quenched condition. The 7030.60 alloy in such a condition has a yield stress of 60 M P a , while the 7108.72 alloy starts off at a higher value of 106 M P a . In the next stages, these two alloys behave similarly. Dur ing the 24 hours of natural ageing, the yield stress of both alloys increases; and i n the 5-hour artificial ageing at 100°C, the yield stress increases at a higher rate. In the following artificial ageing at 150°C, the yield stress keeps rising unti l it 47 reaches a peak, after which it decreases. It is worth noting that although these two curves run at an almost constant offset at the beginning, they meet at a time of 70 hours of artificial ageing at 150°C. T i m e after W a t e r Q u e n c h / Hr. Figure 4.15: Y ie ld stress evolution during the ageing process for the 7030.60 and 7108.72 alloys without pre-deformation. (The dashed line corresponds to the industrial T6 schedule.) The ageing curves (for ageing times of up to 70 hours at 150°C) for these alloys with different levels of pre-deformation are summarized in Figures 4.16 - 4.17. Generally, the pre-deformed 7030.60 alloy behaves in a similar manner, i.e., the yield stress rises to a peak and then decreases. However, important differences can be observed between these ageing curves. 48 CO CO CO S> CO co >-12 24 36 48 60 72 T i m e after W a t e r Q u e n c h / Hr. 84 96 Figure 4.16: Y ie ld stress evolution during the ageing process for the 7030.60 alloy wi th different levels of pre-deformation. 12 24 36 48 60 72 T i m e after W a t e r Q u e n c h / Hr. 84 96 Figure 4.17: Y i e l d stress evolution during the ageing process for the 7108.72 alloy wi th different levels of pre-deformation. 49 First, the yield stresses at the 'starting point'. A l l the ageing curves start at the as-quenched condition, which shows a yield stress value of 60 M P a . However, according to the experiment schedule (Figure 3.1), the materials were pre-deformed immediately after water quench, then naturally aged at room temperature. The data points which fall along the y-axis (i.e. at time « 0) in Figure 4.16 correspond to the yield stress of the alloy in the as-deformed condition, and they are designated as the 'starting point' of the ageing process. A s one can see, as the pre-deformation level increases (from the strain of 0 to 1.2), the yield stress of the as-deformed material increases significantly. Second, there is a yield stress drop for the samples pre-deformed for a strain of 0.1 and up at the first hour of artificial ageing at 100°C. Moreover, the drop in yield stress becomes larger as the pre-strain level increases. Thi rd , the peak yield stress for those samples wi th pre-deformation are always lower than the peak yield stress of the non-deformed sample. Furthermore, wi th careful scrutiny of these curves, it can be observed that the time to reach peak yield stress is decreasing as the level of pre-deformation imposed on the material increases. F r o m Figure 4.17, the similar observations can be made on the ageing behaviour of the 7108.72 alloy wi th pre-deformation. These phenomena i n each ageing stage w i l l be discussed in details in Chapter 5. The resistivity evolution for both alloys during ageing is summarized i n Figure 4.18 and 4.19, respectively. Fo r the non-deformed materials, the electrical resistivity increases during the natural ageing, then drops to a lower step during the artificial ageing at 100°C. Dur ing the artificial ageing at 150°C, the electrical resistivity keeps decreasing, firstly at a 50 higher rate, then the rate of decrease is reduced. 0 12 24 36 48 60 72 84 T i m e after W a t e r Q u e n c h / Hr. Figure 4.18: Electrical resistivity evolution during the ageing process for the 7030.60 alloy wi th different levels of pre-deformation. 0 12 24 36 48 60 72 84 96 T i m e after W a t e r Q u e n c h / Hr. Figure 4.19: Electrical resistivity evolution during the ageing process for the 7108.72 alloy wi th different levels of pre-deformation. 51 For the pre-deformed materials, the electrical resistivity is increased i n the as-deformed condition. The more the pre-deformation is imposed on the sample, the larger the value of electrical resistivity increases. Dur ing the artificial ageing period, the electrical resistivity of the pre-deformed samples follows the same trend as the non-deformed material. However, it is worth noting that the rates at which the electrical resistivity decreases are different for the materials pre-deformed to different level. The rate is higher as more pre-deformation is imposed on the material. 52 Chapter 5 D I S C U S S I O N There is a wealth of information to be explored in the experimental results we obtained. In this chapter, we w i l l give a detailed analysis on the results in order to unveil the ageing characteristics of these materials. The ageing behaviour of the materials in each individual ageing stages w i l l be discussed. 5.1 EFFECT OF Zr ON THE GRAIN STRUCTURES The grain structure of an alloy is the result of its chemical composition and the thermal-mechanical history it experienced. The grain structures of the 7030.60 and 7108.72 alloys after the hot extrusion appear to be very different, as illustrated i n Figures 4.1 and 4.2. Considering the fact that these two alloys have the same thermal-mechanical history (see Figure 2.2), and almost the same chemical composition in the major alloying elements of Z n , M g and C u (see Table 3.1), it is suggested that the difference in the trace elements (Zr addition) plays a role on the microstructure of the material that leads to the present result. The 7030.60 alloy has no Z r content. Dur ing or after the hot extrusion, recrystallization may occur wi th the help of the thermal energy, leading to the equiaxed coarse grain structure. 53 The 7108.72 alloy has a Z r addition of 0.16 wt.%. It is well documented that Z r acts as a grain refiner and recrystallization inhibitor in aluminum alloys [3, 4]. It also simultaneously improves the stress corrosion cracking resistance of the 7xxx series aluminum alloys [30]. Zi rconium has low solubility in the 7xxx series aluminum alloys. In the as-cast ingot, Z r usually stays in supersaturated solid solution. Dur ing the preheating in the extrusion process, Z r precipitates as A l 3 Z r intermetallic particles to form dispersoids in the aluminum matrix. Due to the very low diffusion coefficient of zirconium in aluminum, the A l 3 Z r particles are resistant to coarsening. Moreover, since they are formed by a solid-solid reaction, their size is small, ranging from 0.02 to 0.5 p m [5]. So the A l 3 Z r dispersoids act effectively as grain refiner and recrystallization inhibitor because of the Zener drag phenomenon [31], resulting in the fibrous, unrecrystallized grain structure for the Z r bearing 7108.72 alloy after hot extrusion. The difference in the grain structure between these two alloys may affect their mechanical properties, as we w i l l see later. 5.2 OVERVIEW O F T H E A G E I N G BEHAVIOUR Referring to the T E M works done by Stiller, et. al. [32] and Saeter, et. al. [33] on the same alloys, the general sequence of precipitation is given by: SSS -> G-P zones -> n ' - » n where n ' is the metastable precipitate and n the equilibrium precipitate. Many variants of the precipitates, such as r\u n 2 and r| 4, etc., have been reported [33]. G-P zones nucleate 54 and grow during the natural ageing period, and they are also found to survive even at the beginning of the 100°C artificial ageing process. The r\' phase is formed during the artificial ageing process, and may transform into rj phase at some time i n the 150°C ageing process [32]. Fol lowing the series of microstructure evolution, the mechanical properties (yield stress, U T S , etc.) and electrical resistivity of these alloys changes accordingly. These microstructure-sensitive properties provide us a window through which the changes i n the microstructure can be traced. The two-step ageing process is designed to take the full advantage of the precipitation hardening. The natural ageing and the first step of artificial ageing are employed to generate a high density of G-P zones, so that they can act as nucleation sites for the formation of the metastable n ' precipitates. In such a heat treatment schedule, very finely dispersed T I ' precipitates can be obtained, which provide high ageing strength, according to the precipitation hardening mechanism discussed later. 5.2.1 YIELD STRESS EVOLUTION DURING AGEING The yield stress is an important mechanical property for the material that is to be adopted for making bumper product. Therefore, it receives the most attention in this study. The yield stress is determined by many factors, such as chemical composition, thermo-mechanical history, temperature, etc. It is a microstructure sensitive parameter. In this study, the most significant contributors to the yield stress property are precipitation hardening, dislocation hardening and solid solution hardening. These factors either make their own contributions independently, or they can interact wi th each other and make the result no longer accountable by a simple linear addition law. 55 According to the classic theories of age hardening [34], the precipitates contribute to the strength of the materials in two ways. When they are small and weak, the flow stress for dislocation shearing is less than that for dislocation bypassing, so they are sheared by dislocations. The overall strength is dominated by the shearing mechanism. Since a dislocation has some flexibility, the number of precipitates it touches per unit length increases as r increases and the precipitates grow stronger. So the flow stress increases. Their flow stress contributions can be expressed in Equation 2.1, namely, s^hearablf = q f V (2.1) When the precipitates grow larger in the coarsening stage and become strong obstacles to the dislocations, the dislocation-particle interaction changes to bypassing. The flow stress is then dominated by the bypassing mechanism and the flow stress is determined by Equation 2.4, namely, G b f 1 / 2 , x CT = c (2.4) non-shearable 4 j- v ' In the coarsening stage, the increases in the size of the precipitates usually leads to the increasing of the precipitate spacing L , since the volume fraction of the precipitates remains nearly constant. Thus, under the dislocation bypassing mechanism, the flow stress decreases as the precipitates grow larger. A t a certain point where the radius of the precipitates is r 0, the flow stresses for dislocation shearing and bypassing are equal, and it reaches its maximum. This is where the transition from dislocation shearing to bypassing occurs, and also the peak strength occurs during the ageing process. Figure 5.1 illustrates the evolution of yield stress as a function of the precipitate size r, schematically. Bearing in mind that the precipitate size r is a function of ageing time, this qualitatively explains the shape of the ageing curves in Figure 4.15. 56 • Precipitation *% strength due to \ dislocations by-*» passing particles * • _ . * * * * * Precipitation * strength due to dislocations shearing particles \ ^ >— Overall precipitation strength Figure 5.1: Schematic illustration of the flow stress as a function of the obstacle radius r. It is widely accepted that the age hardening of the 7xxx aluminum alloys is i n most cases associated wi th the disperse precipitation of the metastable n ' phase, however, the details of the strengthening mechanism remain unclear. The dislocation-n 1 interaction is a complicated process. A t present, the information on the dislocation-n' interaction is very limited, due to the experimental difficulties. It is even unknown that whether the n ' phase is shearable or non-shearable, so the transition point can not be determined. A n alternate approach to examine the age hardening response is to consider the precipitates as a series of obstacles lying on the glide plane. Dur ing ageing, the obstacle strength (which may be characterized by the critical dislocation breaking angle (pj increases. Simultaneously the averaging, L , is also changing, first due to nucleation of new precipitates and later due to coarsening. F r o m this point of view, the problem can be examined using the results of Foreman and M a k i n (i.e., Equation 2.5 and 2.6): 57 (9C < 100°) (2.5) ( 9 e > 100°) (2.6) where L is the precipitate spacing and 9 C the critical breaking angle which is related to the precipitate strength. A t the initial stage of precipitation, the increasing cos(9 c/2) term, due to the precipitates growing stronger, is more than enough to compensate for the decreasing term (1/L), due to the increase in precipitate size r. So the overall flow stress increases. However, the increase in the value of cos(9 c/2) term is limited. A t a certain point where the decreasing of the term (1/L) can not be compensated by the increasing i n the cos(9 c/2) term, the overall flow stress due to precipitation hardening then begin to decrease. This is an alternative explanation for the occurrence of the peak in the age hardening curve. It is worth noting that these two views are actually complementary. For example, Equation 2.1 can be derived from Equation 2.6 if it is assumed that the obstacle strength of the precipitate is proportional to the size of the precipitate [29]. It is also worth noting that all the arguments above assume that all precipitates have the same size at the same time. We shall see later that the precipitate size distribution may have an interesting influence on the precipitation hardening response. 5.2.2 ELECTRICAL RESISTIVITY EVOLUTION DURING AGEING The electrical resistivity measurement has been adopted by many authors to study the 58 precipitation process [10, 35, 36]. In fact, the electrical resistivity can be used as an indirect size measurement of the precipitates [35]. In this work quantitative analysis of the electrical resistivity data proved to be more complicated than expected, however, useful information, for example, on the precipitate coarsening kinetics, can be obtained by a qualitative analysis. Very small G-P zones can scatter free electrons and give a rise to an increase in electrical resistivity. This happens during the formation of the G-P zones, and explains the resistivity increase during the natural ageing period, shown i n Figures 4.18 and 4.19. The maximum resistivity is reached when the size of these clusters is of the same order as the mean free path of conduction electrons [35]. When the precipitates grow larger, they are less effective in scattering the electrons, and their population also decreases. Furthermore, the matrix is depleted as well , therefore the electrical resistivity decreases. B y this means the electrical resistivity provides a statistical measurement characteristic of the precipitate size, particularly in the artificial ageing stages. F r o m Figure 4.18, it can be observed that in the 150°C ageing stage, it takes less time for the 7030.60 alloy wi th higher level of pre-deformation to reach the same resistivity value. This implies that the precipitates grow faster as the pre-strain level increases. The same is also true for the 7108.72 alloy, as seen from Figure 4.19. 5.3 EFFECT OF PRE-DEFORMATION ON THE YIELD STRESS OF THE AS-DEFORMED MATERIAL F r o m Figures 4.16 - 4.17, it is evident that pre-deformation has several effects on ageing behaviour of both alloys. Starting wi th the pre-deformation process, we w i l l examine the effect of pre-deformation on the as-deformed materials in this section. 59 The pre-deformation process was carried out immediately after solution and water quench. Figure 5.2 illustrates the yield stresses of the 7030.60 and 7108.72 alloys in the as-deformed condition as a function of pre-strain. The curve quite resembles the plastic region in a tensile curve in shape (although to much higher strain that is achievable in a tensile test). The yield stress increases as the pre-strain increases, and it does not seem to have saturated, even at the pre-strain of 1.2. 450 j 400 -Stra in Figure 5.2: Y ie ld stress of the 7030.60 and 7108.72 alloys i n the as-deformed condition as a function of pre-strain. It can be observed that there is a nearly constant difference in the yield stress for the two alloys, as shown i n Figure 5.2, and this offset remains more or less constant during ageing process. This may be attributed to the different grain structures (i.e., grain size and sub-grain size effects) and textures [37] between these two alloy extrusions. However, this phenomenon is not comprehensively understood since the yield stress is a complicated function of many material parameters. 60 After solution and water quench, the material is brought into supersaturated solid solution condition. The solute atoms make their contribution to the flow stress. Meanwhile, the pre-deformation process introduces new dislocations into the material, and makes the material work harden due to the increased dislocation density. A s the level of pre-strain increases, the dislocation density becomes higher. A n approximation for the increases in dislocation density can be made from the following equation [38], a = a 0 + M a G b p 1 / 2 (5.1) where M is the Taylor factor (3.06 for random texture; for 7030.60 and 7108.72 alloy, texture effects are ignored in this approximation), a is a constant wi th a value of 0.2 for F . C . C . metals, G the shear modulus, b the magnitude of the Burgers vector and p the dislocation density. The dislocation density data calculated from Equation 5.1 is shown i n Figure 5.3. The higher dislocation density in the alloy wi th higher level of pre-deformation results in the increased flow stress. The presence of a high dislocation density in the pre-deformed materials can have many effects on the behaviour of the material. It is proposed that the dislocations can act as vacancy sinks [39]. Due to this effect, the excess vacancy concentration obtained by quenching from solution temperature is lower i n the pre-deformed materials where the dislocation density is higher than the non-deformed material. The dislocations can also act as heterogeneous nucleation sites for precipitates. So the distribution of dislocations i n the material has a strong, if not dominant, influence on the distribution of precipitates. The dislocation lines can also provide a high diffusivity short circuit diffusion path for solute atoms. This enhances the overall diffusion process, which dominates the kinetics of precipitate growth and coarsening. The dislocations also deplete solute atoms from 61 the matrix, due to a solute flux to dislocations caused by the interaction between elastic stress field around the dislocations and solute atoms. The result is the lower solute fraction available for bulk precipitation. 30 0 0 0.2 0.4 0.6 0.8 1 1.2 Strain Figure 5.3: Plots of [ (a-a 0 ) / M a G b ] 2 vs. strain, illustrating the increases of dislocation density after the pre-deformation for the 7030.60 and 7108.72 alloys. 5.4 EFFECT OF PRE-DEFORMATION ON THE NATURAL AGEING BEHAVIOUR During the natural ageing period, the yield stresses of the two alloys rise, as seen from Figures 4.16 - 4.17. Figure 5.4 summarizes the yield stress gain for these alloys wi th different levels of pre-strain during the 24-hour natural ageing period. 62 It is clear that the gain in yield stress becomes less as the pre-strain level increases. This indicates that the natural ageing response is lowered for the pre-deformed material. It is known from the literature that the precipitation reaction during natural ageing is the formation of G-P zones. These G-P zones contribute to the strength of the alloys by acting as weak obstacles to dislocations. The nucleation of G-P zones is strongly affected by the excess vacancy concentration in the material. In the non-deformed alloys, the excess vacancy concentration introduced by water quenching is available to enhance the nucleation of G-P zone zones, resulting i n a high density of G-P zones. O n the other hand, for the materials wi th pre-deformation, the excess vacancy concentration is lowered by the increased number of dislocations acting as vacancy sinks. Therefore the nucleation of the G-P zones is suppressed, leading to a lower density of G-P zones. In this case, the contribution to the strength of the alloy is less, resulting in the lowered natural ageing response for the pre-deformed materials. 0.2 0.4 0.6 St ra in 0.8 1.2 Figure 5.4: Changes i n yield stress during the 24-hour natural ageing process as a function of pre-strain. 63 5.5 EFFECT OF PRE-DEFORMATION ON THE ARTIFICIAL AGEING BEHAVIOUR Following the natural ageing process, the materials were then artificially aged i n a two step ageing process. The ageing temperature for the first step is 100°C and the second is 150°C. The ageing behaviour of the material in the artificial ageing process is discussed i n this section. 5.5.1 ARTIFICIAL AGEING STAGE I): AGEING AT 100°C The yield stress of the non-deformed material rises rapidly in this stage, as seen in Figure 4.15. The increasing yield stress is attributed to the increasing number of the r|' precipitates which are formed in this stage [32]. A n interesting material behaviour was observed in this ageing stage. A yield stress drop occurs for some of the pre-deformed alloys in the first hour of artificial ageing at 100°C, as shown in Figures 4.16 and 4.17. The changes in yield stress during this period are summarized i n Figure 5.5. For the 7030.60 alloy wi th 0 and 0.03 pre-strain, the yield stress increases. For all other cases, the yield stress drops wi th the amount of drop becoming larger as the pre-strain level increases In the stage of artificial ageing at 100°C, there are two main strength contributors for the pre-deformed materials (the intrinsic strength of the aluminum matrix is always default), one is the precipitation strengthening and the other is the dislocation strengthening. Since no yield stress drop is observed in the non-deformed materials (which would be an indication of no G-P zone dissolution), the other contributor, the dislocation 64 strengthening, seems to be responsible for the observed yield stress drop. Stra in Figure 5.5: Change in yield stress between the end of natural ageing and after 1 hour at 100°C. We suggest that static recovery occurs in the pre-deformed materials during the artificial ageing process. After cold deformation, dislocation density becomes higher, and energy is stored in the material. This provides the driving force for the recovery to occur. When the material is artificially aged at a higher temperature, the thermal energy makes the recovery process kinetically possible. A s a result of static recovery, the dislocation density decreases, therefore its contribution to the strength becomes less. The higher degree of pre-strain imposed on the material, the higher the driving force for recovery, so the dislocation density recovers more quickly, showing a larger drop in the yield stress. Further evidence for the static recovery w i l l be provided by analyzing the data from the overageing stage. 65 5.5.2 ARTIFICIAL AGEING STAGE II): AGEING AT 150°C It can be clearly observed from Figures 4.16 and 4.17 that the peak strength and T6 strength for the materials vary as the pre-deformation is imposed on the materials. It is worth emphasizing again that the T6 strength and the peak strength for the pre-deformed materials are lower than the non-deformed material. Referring to Figure 2.1, the corner area of the bumper receives a larger deformation (approximately 0.3 of strain) than the center during the bumper forming process. Therefore, the material in this area should have a relatively higher level of work hardening, due to dislocation hardening. Superimposed wi th the precipitation hardening during the heat treatment, it is natural to think that the strength of the material at the corner area should be higher than the material at other area of the bumper. However, the experiments simulating the industrial process give the reverse result (see Figures 4.16 and 4.17 for the T6 strength). In fact, the T6 strength for the material wi th 0.3 pre-strain is significantly lower than the non-deformed material. This leads to the spatial heterogeneity i n the strength throughout the bumper product. A similar effect is observed for the peak strength of the materials. It is unusual that the sum of the dislocation hardening and precipitation hardening would be less than if we only had the precipitation hardening. Figure 5.6 summarizes the peak strength for both alloys as a function of pre-strain. Generally, the peak strengths for the materials wi th pre-deformation (all levels for the 7030.60 alloy, 0.03 and up for the 7108.72 alloy) are lower than that without pre-deformation. Each of these two strengthening mechanisms, the dislocation hardening and the precipitation hardening, contributes to the strength of the material positively, when they are acting individually. However, the total amount of strengthening obtained from a 66 combination of these two mechanisms, is less than the contribution from only one of them working alone for this alloy system. This scenario leads us to the following suggestion: some sort of negative interaction between these two mechanisms exists under these circumstances. In the following, we w i l l discuss the negative effects of dislocations on the precipitation strengthening. 450 St ra in Figure 5.6: Variation in the peak strength as a function of pre-strain. A s it is known, the precipitation hardening is dominated by the evolution of precipitate size and spacing. The precipitate size measurements were conducted in the T E M works on the same alloys used i n this study by Salter, et. al. [33]. Figure 5.7 shows T E M micrographs for the 7030.60 alloy, demonstrating influence of pre-deformation on the precipitation behaviour. In the non-deformed alloy, the precipitates appear to be small and have a more or less rounded shape. When dislocations are introduced into the material, the precipitates are larger, they vary more in size and tend to differ more in shape than i n the non-deformed condition. 67 b) Figure 5.7: TEM-micrographs from a) the non-deformed and b) pre-deformed (0.10 strain) conditions for the 7030.60 alloy aged for 70 hours at 150°C. (From Saeter, [40].) The results of the precipitate size measurement are summarized in Table 5.1. 68 Table 5.1: Summary of precipitate sizes after ageing at 100°C/5hr . + 150°C/70hr . (From Salter, et. al. [33]) Condi t ion A v g . size (nm) S T D (nm) Max size (nm) A v g . s ize /STD 7030.60 0 pre-strain 6.3 3.3 19.3 1.9 7030.60 0.1 pre-strain 11.3 8.5 53.2 1.3 7108.72 0 pre-strain 9.4 4.3 24.7 2.2 7107.72 0.1 pre-strain 13.4 8.6 56.5 1.6 It is clear that the precipitates are larger i n the pre-deformed materials than in the non-deformed ones. (Our electrical resistivity measurements also indirectly confirm this qualitatively, see Figures 4.18 and 4.19). Since the precipitate coarsening process is controlled by diffusion, it is apparent that diffusion is enhanced by additional short circuit diffusion paths, such as along dislocation cores, which are added to the normal bulk diffusion mechanism when more dislocations are introduced into the materials by cold work. If this occurs, the precipitates are able to grow more quickly and rapidly reach the critical size where peak strength is developed. Therefore the shortening of time to reach peak strength can also be expected. In fact, this has been observed in our results, as summarized i n Figure 5.8. The time to peak strength is shortened from 10 hours for the non-deformed materials to approximately 4 hours for the materials wi th a pre-strain of greater than 0.3. The T E M studies by Salter, et. al. [33] on the same alloys used i n this study also give some other important observations. First, they found that the dislocations organize themselves into a loose cell structure, consisting of very broad cell walls and dislocation-free cell interiors i n the 7030.60 alloy wi th 0.1 pre-strain, shown i n Figure 5.9a. This 69 feature appears in the 7108.72 alloy as well . In Figure 5.9b, it is shown that the precipitates in the cell walls are much larger where the dislocation density is higher, indicating that the dislocations play an important role in contributing to the nucleation and growth of precipitates. A n i n situ small angle X-ray scattering (SAXS) study of an A l - Z n - M g - C u alloy by Gomiero, et. al. [41] shows that there is a significant difference in the size of precipitates at the dislocation sites and in the bulk. The precipitates at the dislocations are larger than the bulk precipitates. They also found that deformation prior to ageing accelerates precipitate coarsening during ageing at 160°C (which is the second step in a two step artificial ageing process), especially heterogeneous precipitation on dislocations. 70 Figure 5.9: T E M micrographs of 7030 alloy wi th 0.1 pre-strain, aged for 5 hours at 100°C plus 20 hours at 150°C. a) dislocations in contrast and b) dislocations out of contrast, showing the precipitates. (From Saeter, et. al. [33].) 0.3 jm b) 71 Saeter, et. al. [33] also studied the precipitate size distributions on our alloys. The results are shown in Figure 5.10. The log-normal distribution for the precipitates are observed for these curves. The difference between the pre-deformed and non-deformed materials is that the tail of the size distributions is much longer for the former one. Furthermore, the existence of two peaks, one large and one small, in these distributions indicates that there are two sets of precipitates in the system. We suggest that two groups of precipitates can be distinguished in the pre-deformed materials: one for the bulk precipitates, and the other for precipitates on dislocation sites. It is clear that the precipitates on dislocations are larger in size than the bulk precipitates (see Figure 5.9). This is due to the enhanced diffusion provided along dislocation cores. Figure 5.11 gives a schematic illustration of the precipitate size distribution in the pre-deformed materials, based on the measurement of Salter [33]. Size [nm] Size [nm] (a) (b) Figure 5.10: Precipitate size distributions for pre-deformed and non-deformed a) 7108.72 alloy and b) 7030.60 alloy, aged at 100°C for 5 hours followed by 70 hours at 150°C. (From Saeter, et. al. [33].) 72 A s the levels of pre-deformation increases, the dislocation density i n the material increases as well (a sample estimate was made in Section 5.3). The leads to the following results: the heterogeneous nucleation of precipitates at dislocation sites is enhanced. O n the other hand, as more solute atoms are depleted from the matrix, the bulk precipitation becomes more difficult. These two process culminate in the variation in the distributions of precipitates between the bulk and dislocations. Higher levels of pre-deformation w i l l result in the rising of the peak for precipitation on dislocations and the lowering of the peak for the bulk precipitation in Figure 5.11. Bulk precipitates - Precipitates on - dislocations - / t\ I 1 1 1 , 1 1 1 1 1 1 0 Size Figure 5.11: Schematic illustration of the size distributions for bulk precipitates and precipitates on dislocations for the pre-deformed materials. A s a brief summary, the presence of pre-deformation leads to higher dislocation density, larger precipitate size, and a two peak precipitate size distributions consisting bulk precipitation and precipitation on dislocations. A s the pre-deformation level increases, the precipitation on dislocations is enhanced. Fol lowing the arguments i n Section 5.2.1, we now consider the evolution of strength under the condition that there is a precipitate size distribution as shown i n Figure 5.10. 73 Assuming that the peak strength occurs at a precipitate size r c, regardless of the mechanisms whether it is a shearing-bypassing transition or the precipitate spacing (L) term offsetting the precipitate strength term (cpj. Under the condition that all the precipitates have the same size at the same time, all the precipitates make their full contributions to the flow stress. However, when there is a precipitate size distribution, only a fraction of precipitates has the critical value of r c. The remaining fraction contains precipitates either larger or smaller i n size than r c, and make a smaller contribution to the flow stress than those wi th a size of r c (i.e., there are weak links in the chain). A s a result, the whole population of precipitates do not make same contribution to the flow stress simultaneously. Thus the lower peak strength is expected. This explains why the presence of pre-deformation leads to a lowered peak ageing strength of the alloys. The work by Deschamps [42] shows a similar results (see Section 2.4). 5.5.3 K I N E T I C S O F P R E C I P I T A T E C O A R S E N I N G The precipitate coarsening occurs in the overageing stage. Table 5.2 summarized the data of the precipitate size for the non-deformed alloys in the overaged conditions. Figure 5.12 illustrates the average precipitate size versus ageing time at 150°C. For the 7108.72 alloy, a straight line can be fitted to the log-linear data wi th a slope of 0.33, giving a coarsening law of: r oc t°- 3 3 It is therefore apparent that the coarsening of the precipitates during the overageing stage follows the cubic coarsening law (i.e. exponent = 0.333). 74 Table 5.2: Summary of precipitate sizes for the non-deformed alloys i n the overaged conditions. Condition Avg. size (nm) STD (nm) Max size (nm) Avg. size/STD 7030.60 20 hr./150°C 5.4 - - -7030.60 70 hr./150°C 6.3 3.3 19.3 1.9 7030.60 60 days/150°C 7108.72 20 hr./150°C 23 6.3 10 2.4 72 16 2.3 2.7 7107.72 70 hr./150°C 9.4 . 4.3 24.7 2.2 7107.72 60 days/150°C 24 12 80 2.0 5.6 RECOVERY DURING THE ARTIFICIAL AGEING The overageing process (60 days/1440 hours at 150°C) greatly reduces the yield strength of the alloys. The evidence for static recovery can be revealed by the following analysis 75 on the experimental data. Referring to the data i n Table 4.1 in Section 4.2.1.2, we collect the yield stress data for the 7030.60 alloy wi th pre-strain of 1.20 i n the as-quenched condition, o"a^ , as-deformed condition, CTAJ, and overaged (150°C/60 days) condition, CT^, as shown in Table 5.3. Table 5.3. Y i e l d stress data for 7030.60 alloy wi th pre-strain of 1.20 in the as-quenched, as-deformed and overaged conditions. Alloy Pre-strain Yield stress (MPa) Yield stress (MPa) Yield stress (MPa) As-quenched As-deformed Overaged (60 d.) 7030.60 1.20 60 300 191 A t each ageing stage, the flow stress components are illustrated i n Figure 5.13. In the as-quenched condition, a a . q - o 0 + o„ (5.2) where CT0 is the intrinsic strength (including grain size effect) for the aluminum matrix, which can be estimated to a value of approximately 25 M P a [34], and is the strength contribution from the supersaturated solid solution. In the as-deformed condition, CTa-d = CTa-q + a d i s ( t o t [ i e first order of approximation) (5.3) 76 where the a& is the amount of work hardening (dislocation hardening) obtained during the pre-deformation. Therefore, ^ d , s = ^ - a ^ = 2 4 0 M P a . CD CL CO 350 300 250 200 CO 150 CD 100 50 0 ° 0 As quenched °"dis. As deformed Experiment °>pt residual + 0"ss. residual °"o Overaged 150°C/60d. Assuming no recovery °"di! Overaged 150°C/60d. A g e i n g C o n d i t i o n s Figure 5.13: F l o w stress components for the 7030.60 alloy wi th pre-strain of 1.20 in various ageing stages. In the overaged condition, only a fraction of solid solution strengthening (ajres idual and precipitation strengthening ( o " p p t ) r e s i dua i remain. If there is no recovery of the dislocation hardening at all, then, 77 CTo-a~ °"0 + (°"ss)residual + CTdis + ( a P P t ) res idua l " (5.4) Even taking the values of both (CTj r e s i d u a l and (a p p t ) r e s i d u a l to zero, we can see that has a value of at least (25 + 240) =265 M P a (see the data column to the right of the Figure 5.13), which is substantially larger than the actual value of 191 M P a in Table 5.3. Therefore, the recovery of a d i s has to be taken into account to accommodate this discrepancy. The presence of recovery make the separation of dislocation hardening and precipitation hardening more complicated for the alloys wi th pre-deformation, and this problem needs to be further explored. 5.7 A SIMPLE MODEL FOR PREDICTING STRESS-STRAIN CURVES In this study, two principle strengthening mechanisms exist for the two alloys, i.e., the dislocation hardening and precipitation hardening. They originate from two sets of obstacles, forest dislocations and precipitates. In this section, we w i l l consider how the two mechanisms contribute to the strength of the alloys. Foreman and M a k i n [25] studied this problem of averaging over the various strength of obstacle in their computer experiments. Obstacles of two different strengths distributed randomly in the slip plane wi th different volume fraction are considered. The flow stress if the first type of obstacle was acting alone is au and the flow stress to overcome the second type alone is a 2 . T w o addition laws were proposed. The first is a simple linear addition law: 78 0"™. = o-j + c 2 (5.5) where a t o t is the total flow stress. The second is a mean square law: <*«. = ( ^ 2 + (5-6) B y comparing the computer calculations by Foreman and M a k i n and the results from Equation 5.5 and 5.6, it is found that the linear addition law only applies well to the circumstance where a few strong obstacles are introduced among many weak ones, while the mean square law is in good agreement except for the special circumstance mentioned above [25]. In our case, two types of obstacles, forest dislocations and precipitates, are present. Their contributions to the total flow stress w i l l be discussed next. First, in a simple situation of the stress-strain curve of the 7030.60 alloy in the as-quenched condition (with no pre-strain), the flow stress is the result of only dislocation hardening plus the intrinsic strength of the matrix CT0 (which also includes the solid solution hardening factor for the consideration of simplicity here), i.e., tr t o t = CTQ + CT^. The a t o t can be modeled using Voce equation [43]: CT-a s - s s = e x p [ - - — L ] (5.7) CT. -CT 8 I s r where CT is the flow stress, CT; and e ; are the yield stress and yield strain, respectively, CTs is the saturation stress, which is the interception of the work hardening rate 0 (=dCT/ds) wi th the stress axis shown in Figure 5.13, s r is the relaxation strain, which equals to CT/0( 0) 79 where 0O is the interception of work hardening curve linearly extrapolated to the 0 axis, also illustrated in Figure 5.14. C O ra c ' c d) T 3 i ra X 3500 3000 2500 2000 1500 1000 500 e 0 -7030.60, 0 pre-strain, as water quenched. -+-50 100 150 200 S t r e s s / MPa 250 300 Figure 5.14: Plot of the work hardening rate vs. stress for the 7030.60 alloy wi th no pre-deformation in the as water quenched condition. Using the Equation 5.7 and the parameters described above, the calculated total flow stress curve can be obtained. Subtracting the a 0 from the total flow stress curve, then we obtain the flow stress contribution from the dislocation, which is shown in Figure 5.15. Assuming that the work hardening curve does not change during ageing, the stress-strain curves for different levels of precipitation hardening can be predicted using either the linear addition law, or the mean square law mentioned above. 80 300 j 250 -CO 200 D . 2 to 150 --to CO 100 -50 • 0 -• 0 a 7030.60, 0 pre-strain, as water quenched. dislocation ,ltk^i a hardening b 0.02 0.04 0.06 Strain 0.08 0.1 Figure 5.15: Plots of the experimental stress-strain curve and the flow stress contribution from the dislocation hardening for the 7030.60 alloy wi th no pre-deformation in the as water quenched condition. For the linear addition law, the flow stress is given by: 0\ot. = °"o + 0"Ppt. +0-^ (5.8) where cr0 is the intrinsic strength of the aluminum matrix, a p p t is the flow stress contribution from precipitation hardening, and is the flow stress contribution from dislocation hardening. For the mean square law, the flow stress is given by: CTtot. = CT0 + (CTppt.2 + ^dis.2) 1/2 (5.9) where a 0 , a p p t and a d i s have the same meaning as in Equation 5.8. The results of the predicted stress-strain curves as well as the experimental curves are 81 plotted i n Figures 5.16 - 5.20. Figure 5.16 illustrates the predicted stress-strain curves for the non-deformed 7030.60 alloy in the natural ageing condition. It is clear to observe that the linear addition law applies properly for this case, except for the stress at larger strains (0.07 and up), while the mean square law under-estimate the flow stress. Figure 5.17 illustrates the predicted stress-strain curves for the non-deformed 7030.60 alloy aged at 100°C for 1 hour. The same observation can be made as in Figure 5.16. This implies that there are "a few strong obstacles introduced among many weak ones", i.e. forest dislocations among weak G-P zones in the 7030.60 alloy in such ageing conditions. For the 7030.60 alloy aged at 100°C for 5 hours, the linear addition law begins to over-estimate the flow stress, while the mean square law still under-estimate the flow stress as the former cases, as shown in Figure 5.18. A s the ageing process proceed, both the over-estimate of flow stress for the linear addition law and the under-estimate for the mean square law become more significant, as clearly illustrated in Figures 5.19 and 5.20. The deviation of the models and the experimental results indicates that the addition of the dislocation strengthening and precipitation strengthening is complicated. Since the mean square law actually averages the density of the two obstacles, the under-estimate of flow stress implies that the work hardening behaviour of the 7030.60 alloy may be changed during ageing wi th the presence of precipitates. More dislocations may be generated due to the dislocation-precipitates interactions and extra flow stress may be added accordingly, as might be expected for non-shearable precipitates (i.e. dispersion hardened material) where geometrically necessary dislocations play an important role and thereby affect the dislocation work hardening [25]. Nonetheless, further study is needed to fully understand this problem. 82 I CO Q _ 200 150 «! 100 D D 50 -linear addition law - experiment - mean square law 0.02 0.04 0.06 Strain 0.08 0.1 Figure 5.16: Plots of the experimental stress-strain curve and the calculated ones using different averaging laws for the non-deformed 7030.60 alloy naturally aged for 24 hours. D_ 200 150 <» 100 b » 50 a linear addition law ——experiment — — mean square law a c 0.02 0.04 0.06 Strain 0.08 0.1 Figure 5.17: Plots of the experimental stress-strain curve and the calculated ones using different averaging laws for the non-deformed 7030.60 alloy aged at 100°C for 1 hour. 83 (0 200 150 «! 100 b D 50 o-l -linear addition law -experiment - mean square law 0.02 0.04 0.06 Strain 0.08 0.1 Figure 5.18: Plots of the experimental stress-strain curve and the calculated ones using different averaging laws for the non-deformed 7030.60 alloy aged at 100°C for 5 hours. ro 0_ 200 150 «? 100 b 50 -linear addition law -experiment -mean square law 0.02 0.04 0.06 Strain 0.08 0.1 Figure 5.19: Plots of the experimental stress-strain curve and the calculated ones using different averaging laws for the non-deformed 7030.60 alloy aged at 150°C for 10 hours (peak aged). 84 200 7 a linear addition law b experiment mean square law 150 + C co a c b 0 0.02 0.04 0.06 0.08 0.1 S t r a i n Figure 5.20: Plots of the experimental stress-strain curve and the calculated ones using different averaging laws for the non-deformed 7030.60 alloy aged at 150°C for 60 days. 5.8 OPTIMUM HEAT TREATMENT Conforming to criteria that the bumper should have a highest min imum strength, the optimum heat treatment can be determined wi th the existing facility, based on this study. Since the maximum level of strain imposed on the material during bumper forming is 0.3, and the material wi th a pre-strain level of 0.3 has the lowest peak strength among the materials wi th pre-strain levels of up-to 0.3 (referring to Figures 4.16 and 4.17), the ageing behaviour of the material at the local section of the bumper where the deformation strain is 0.3 controls the final mechanical properties of the whole bumper product. Therefore the heat treatment schedule should be selected such that the material w i th the maximum strain (0.3) i n the bumper can develop a peak strength after the ageing process. 85 Referring to Figure 5.8, the time-to-the-peak for the 7030.60 and 7108.72 alloys wi th pre-strain of 0.3 are 5 hours and 4 hours at 150°C, respectively. So the heat treatment schedule can be chosen as: 4 8 0 ° C / 20 min. Water quench Bumper forming natural ageing V l d a y + 7030.60 , k 150°C/ 5 hrs. 100°C/ 5 hrs. *< 150°C/ 4 hrs. 7108.72 Different from the above schedule, the actual industrial heat treatment has an ageing time of 6 hours at 150°C, i.e. the pre-deformed alloys are slightly overaged. Some other considerations, such as the susceptibility to stress corrosion cracking, may have been taken into account in the selection of this schedule. 86 Chapter 6 C O N C L U S I O N S A N D F U T U R E W O R K Experiments simulating the industrial bumper manufacturing process have been conducted in this study. Mechanical properties (yield stress, U T S and strain-to-necking) and electrical resistivity evolution during aging for the materials wi th pre-determined levels of pre-deformation were obtained. After analyzing these data, we come to the following conclusions. 6.1 CONCLUSIONS 1. The presence of pre-deformation leads to heterogeneous distribution of mechanical property i n the bumper products. 2. The presence of pre-deformation lowers the natural ageing response of both 7030.60 and 7108.72 alloys. This is suggested to occur due to the loss of excess vacancies to dislocations sinks. 3. Static recovery is suggested to occur during artificial ageing, simultaneously wi th the precipitation process. 4. The presence of pre-deformation lowers the peak strength that can be obtained during artificial ageing, possibly due to the dislocations affecting the precipitate size distribution. 87 5. The overageing kinetics are accelerated due to additional short circuit diffusion paths provided by dislocations. 6. According to the criteria that the bumper should have the highest min imum strength, the optimum heat treatment schedule is determined as: Water 480°C/ quench natural ageing + 100°C/ 20 min . Bumper / l d a y 5 hrs. + ' 7030.60 150°C/ 5 hrs. forming 150°C/ 7108.72 4 hrs. 6.2 FUTURE WORK F r o m this study, it is known that the presence of pre-deformation between quenching and natural ageing culminates in the lower peak ageing strength of the alloys, which is not appreciable for the industry to produce high performance bumper products. We would like to examine the possibilities where the negative effect of the pre-deformation on the final mechanical properties of the materials is reduced or eliminated. The deformation (the shaping of the bumper) is obviously necessary i n the bumper manufacturing process. However, the sequence in which the deformation is taking place may be changed to avoid the negative effects. So it is worthwhile to examine the manufacturing process again. Referring to Figures 2.2 and 2.3 shown again here in Figure 6.1, the layout of the whole production line is in such a way that the dislocations introduced by the deformation process can not be eliminated 88 before the following ageing treatment is carried out. Thus the dislocations remain through the ageing treatment and adversely affect the strength of the materials. Temperature Ageing lOCC/Shr . Time Time Figure 6.1: Schematic illustration of the extrusion and bumper manufacturing processes. According to this study, the dislocations introduced by the deformation process should be removed before the ageing process. So we suggest a modified production schedule in which the deformation is moved forward to the end of the hot extrusion process, as shown in Figure 6.2. In the modified process, it is expected that the dislocation introduced by the deformation process w i l l be eliminated during the following solution heat treatment. Therefore the adverse effect of the dislocations w i l l be avoided in this new procedure. O n the other hand, if the heterogeneous distribution of strength in the bumper is acceptable, the solution heat treatment i n the modified production schedule may be removed, which means energy and resource savings. 8 9 Temperature feasting Time Figure 6.2: Schematic illustration of the modified bumper manufacturing process. The details of the proposed process need to be further investigated. Fo r example: • the hot extrusion temperature. The temperature should be selected and controlled more carefully so that a solid solution can be obtained during the stretch bending process. This is to reduce the resistance of the material to the deformation. • water quench after the hot extrusion. Water quench, instead of air cooling in the present process, should be employed, also to in favor of the solid solution formation. 90 B I B L I O G R A P H Y 1. M . O . Speidel, Metall . Trans. A , 6 A 631 (1975). 2. N . R y u m , Acta. M e t a l l , 17, 269 (1969). 3. E . Nes, Acta. Metall . , 20, 499 (1972). 4. A . J . Bryant, J . Inst. Metals, 97, 311 (1969). 5. J . T . Staley, Proc. of the 3 r d Int. Conf. on A l Al loys , Their Physical and Mechanical Properties. E d . L . Arnberg, O . Lohne, E . Nes and N . R y u m , V o l . HI, N T H and S I N T E F , Trondheim, Norway , 107 (1992). 6. Private communications, S. R . Skjervold, Hydro-Raufoss. 7. W . Rosenkranz, A l u m i n u m , 36, 250, 397 (1950). 8. L . F . Mondolfo, N . A . Gjostein and D . W . Levinson, Trans. A I M E . , 206, 1378 (1965). 9. J . Lendvai, Materials Science Forum, vol . 217-222. 43-56 (1996). 10. P. Guyot and L . Cottignies, Acta. M e t a l l , 44, 10, 4161 (1996). 11. P. N . T . U n w i n and R . B . Nicholson, Acta. M e t a l l , J7, 1379 (1969). 12. J . T . Staley, R . H . B r o w n and R . Schmidt, Metall . Trans., 3, 191 (1972). 13. M . Murakami, O . Kawano and Y . Murakami, Acta. M e t a l l , 17, 29 (1969). 14. A . K Mukhopadhyay, Q . R . Yang and S. R. Singh, Acta M e t a l l , 42, 9, 3083 (1994). 15. G . W . Lorimer and R . B . Nicholson, The Mechanisms of Phase Transformations in Crystalline Solids, Inst, of Metals, London, 43 (1969). 16. A . Ke l ly and R . B . Nicholson, Prog. Mater. Sci., 10, 245 (1963). 17. G . W . Lorimer and R . B . Nicholson, Acta. Met., 14,1009 (1966). 91 18. J . D . Embury and R . B . Nicholson, Acta. Metall. , 12, 403 (1965) 19. P. N . T. U n w i n , G . W . Lorimer and R . B . Nicholson, Acta. Metall . , 17,1363 (1969). 20. J . K . Park and A . J . Arde l l , Metall . Trans. A , 14, 1957 (1983). 21. G . Thomas and J . Nutt ing, J . Inst. Metals, 88, 81 (1959-60). 22. D . W . Pashley, M . H . Jacobs and J . T . Vietz, Ph i l . Mag., 16, 51 (1967). 23. J . T . Staley, Metall . Trans., 5, 929 (1974). 24. S. P. Ringer, B . C . Muddle and I. J . Polmear, Metall . Trans. A , 26A, 1659 (1995). 25. L . M . B r o w n and R . K . H a m , Strengthening Methods in Crystals, E d . A . Ke l ly and R. B . Nicholson, Halsted Press Divis ion, John Wiley & Sons, Inc., N e w Y o r k (1971). 26. J . D . Embury, Materials Sci. Forum, 217-222. 57 (1996). 27. E . A . Starke Jr., Mater. Sci. Eng., 29, 99 (1977). 28. A . Deschamps, Y . Brechet, P. Guyot and F . Livet, Z . Metallkd., 88, 601 (1997). 29. A . Deschamps and Y . Brechet, Acta. Mater., 47, no. 1, 293 (1999). 30. I. J . Polmear, Light Al loys , Metallurgy and Materials Science Series, 2 n d edn., Edward Arno ld , London, 97 (1989). 31. J . Wert, Microstructural Cont ro l i n A l u m i n u m Al loys : Deformation, Recovery and Recrystallization, E d . E . H . Ch ia and H . J . McQueen, T M S , 67 (1986). 32. K . Stiller, V . Hasen, M . Knuston-Wedel, G . Waterloo and J . Gjonnes, Proc. of the 6 t h Int. Conf. on A l and Al loys , Their Physical and Mechanical Properties, ed. T . Sato, S. Kumai , T . Kobayashi and Y . Murakami, Toyohashi, Japan, 2, 615 (1998). 33. J . A . Saeter, G . Waterloo and W . J . Poole, in the Proceedings of I C A A - 6 , ed. T . Sato, S. Kumai , T . Kobayashi and Y . Murakami, Toyohashi, Japan, 2, 745 (1998). 34. H . R . Shercliff a n d M . F . Ashby, Acta. Metall . Mater., 38,10, 1789 (1990). 92 35. V . Gerold, in "Metallic Solid Solutions", ed. by J . Friedel and A . Guinier, W . A . Benjamin Inc. N e w Y o r k , X L (1963). 36. Y . Quere, in "Metallic Solid Solutions", ed. by J . Friedel and A . Guinier, W . A . Benjamin Inc. N e w Y o r k , X X X V (1963). 37. A . Fjeldly and H . J . Roven, Material Science and Engineering, A234-236, 606 (1997). 38. F . R . N . Nabarro, Z . S. Basinski, and D . B . Hol t , A d v . Phys., 13, 193 (1964). 39. M . Wintemberger, Acta. M e t a l l , 7. 549 (1959). 40. J . A . Saeter, unpublished work, (1998). 41. P. Gomiero, A . Reeves, A . Pierre, F . Bley, F . Livet and H . Vichery, in the Proceedings of I C C A - 4 , ed. T . H . Sanders, Jr. and E . A . Starke, Jr., Atlanta, Georgia, U S A , 644 (1994). 42. A . Deschamps and Y . Brechet, Acta. Mater., 47, n o . l , 281 (1999). 43. Y . Estrin and H . Mecking, Acta. M e t a l l , 32, no. 1, 57 (1984). 93 A P P E N D I X Some preliminary experimental work was conducted on the as-received, hot-extruded plates of the 7030.60 and 7108.72 alloys to obtain the basic characteristics of these materials. In particular, the homogeneity of the strength through the plate thickness, the sensitivity to solution time and the quench sensitivity, were examined. These results, which helped to design the schedule for experiments thereafter, are given i n this section. The Through Thickness Hardness Distribution in the As-received Plates The hardness profile through the thickness of the 7030.60 and 7108.72 aluminum plates was measured by taking a piece of material sheared from the as-received plate and grinding to different levels. The Vickers hardness at different levels in the plate is shown in Figure A . l . The results indicate that there is only a small difference in hardness between the surface and center of the plates (6.8% for 7303.60 plate and 2.7% for 7108.72 plate). However, to ensure the consistency of the results, all the samples i n the following experiments were taken from the center in transverse direction of the plates by grinding off the surface layer, where applicable. 94 a) 7030.60 Extruded Plate Thickness: 5mm a v e r a g e o f 3 h a r d n e s s m e a s u r e m e n t s Roll ing surface, Hv= 104.1 0 .1mm below, Hv=105.4 1 m m below, Hv=111.2 2 m m below, Hv=111.6 Centre, Hv=111.7 b) 7108.72 Extruded Plate Thickness: 5mm a v e r a g e o f 3 h a r d n e s s m e a s u r e m e n t s surface, Hv=110.4 0.1mm below, Hv=111.0 1 mm below, Hv=111.3 2 mm below, Hv=112.2 Cent re , Hv=113.5 Figure A . l Hardness distribution through the thickness of the as-received a) 7030.60 and b) 7108.72 plate. 95 Sensitivity to Solution Time The sensitivity of the materials to solution hold time was examined to determine the re-solution heat treatment time for the experiment schedule. Samples of 7030.60 and 7108.72 alloy were solution treated at 480°C for 10, 20 and 60 minutes i n a salt bath, followed by water quench. The samples were then naturally aged at room temperature for 3 days, followed by an artificial ageing at 100°C (in boiling water bath) for 5 hours. The hardness and resistivity measurements were conducted for different ageing times to study the ageing response of the materials. The ageing curves are shown in Figures A . 2 -A . 5 . F r o m these results, it can be concluded that there is no substantial difference in ageing behavior of the 7030.60 alloy for solution heat treatment times of 10, 20 and 60 minutes. The same is true for the 7108.72 alloy. For convenience, a solution heat treatment time of 20 minutes was adopted for subsequent experiments. > X (/> co cu c "E CO X 140 120 100 80 60 40 20 0 ageing at room temperature —% 60 min. _ a _ 20 min. i i _o_10min. 10 100 1000 Ageing Time / min. 10000 > x tn CD c T J L-CO X 60 120 180 240 300 360 Ageing Time / min. Figure A . 2 Ageing curves (in hardness) for 7030.60 alloy wi th different re-solution time at 480°C. 96 E d o> cb > tn co cu Cr: 48 47 46 45 44 43 42 41 40 39 38 . ageing at room temperature % 60 min. _ £ 20 min. — 0 — 10 "in-10 100 1000 10000 Ageing Time / min. 60 120 180 240 300 360 Ageing Time/min. Figure A . 3 Ageing curves (in resistivity) for 7030.60 alloy wi th different re-solution time at 480°C. > x CO co Cl) c "S CO X 140 120 100 80 60 40 20 0 ageing at room temperature % 60 rrin. _ £ _ 2 0 rrin. _o—10 min. 10 100 1000 Ageing Time / min. 10000 > x CO tn cu c "E ro x 60 120 180 240 300 360 Ageing Time / min. Figure A . 4 Ageing curves (in hardness) for 7108.72 alloy wi th different re-solution time at 480°C. 97 a CD o X J ;> tn ' to <u 48 47 46 45 44 43 42 41 40 39 1 38 ageing at room temperature 1 .60 rrin. . 20 min. .10 min. 10 100 1000 10000 Ageing Time / min. 60 120 180 240 300 360 Ageing Time / min. Figure A . 5 Ageing curves (in resistivity) for 7108.72 alloy wi th different re-solution time at 480°C. Sens i t iv i ty t o Q u e n c h i n g M e d i u m The quench sensitivity of the 7030.60 and 7108.72 materials was also studied. Samples for both alloys were solution treated at 480°C for 20 minutes, then quenched in four different media: water at room temperature, o i l , ice-water mixture, and air. After quenching, the samples were given a natural ageing at room temperature for 3 days, followed by an artificial ageing at 100°C for 5 hours. Again hardness and resistivity were measured to observe the ageing behaviors of the materials, and the results are shown in Figures A . 6 - A . 9 . For 7030.60 alloy, only a minor difference in ageing curves was observed against the differences in quenching condition, although there are some deviations in the initial stage of natural ageing. 98 For 7108.72 alloy, the ageing curve for air quenching condition is different from other curves in that it showed lower hardness and resistivity values. This way be attributed to some loss of precipitation hardening potential due to the precipitate of course precipitate on the A l 3 Z r dispersoids during the relatively slow cooling for air quenching. > x co co OJ c "S CO X 140 120 100 80 60 40 20 0 ageing at " room temperature _ • — w ater Q — A — oil Q . _ 0 — ice W.Q. —*— air cool 10 100 1000 Ageing Time / min. 10000 > x co co <D c T3 i 03 X 140 120 100 80 60 40 20 0 ageing at100°C -waterQ. . oil < . icel . air cool eW.Q. 60 120 180 240 300 360 Ageing Time / min. Figure A . 6 Ageing curves (in hardness) for 7030.60 alloy wi th different quenching conditions. E d CD b x J '> '-4—* co to d> Cd 48 47 46 45 44 43 42 41 40 39 38 ageing at room temperature - . ~+--4d^ —0—water Q. oil Q. _ o — ice W.Q. I I M i l l I » air cool 10 100 1000 10000 Ageing Time / min. a CD CD co CO a> Cr: 48 47 46 45 44 43 42 41 40 39 38 ageing at100°C . water Q. . oil Q. . ice W.Q. . air cool H 1 h 60 120 180 240 300 360 Ageing Time / min. Figure A . 7 Ageing curves (in resistivity) for 7030.60 alloy wi th different quenching conditions. 99 > X to tn CD c "s ro X 140 120 100 80 60 40 20 0 ageing at room temperature % water Q _ A _ oil Q . _ 0 — ice W.Q. — air cool 10 100 1000 Ageing Time / min. 10000 > x tn tn CD c i CD X 140 120 100 80 60 40 20 0 ageing at100°C .water Q. .oilQ. . ice W.Q. . air cool 60 120 180 240 300 360 Ageing Time / min. Figure A .8 Ageing curves (in hardness) for 7108.72 alloy wi th different quenching conditions. a O "> '-4—* tn tn CD or 48 47 46 45 44 43 42 41 40 39 38 ageing at room temperature .water Q. -oil Q. . ice W.Q. . air cool 10 100 1000 10000 Ageing Time / min. a cn <3 X J > '-4—" CO to CD a: 48 47 46 45 44 43 42 41 40 39 38 ageing at100°C =8= * • • .water Q. .oilQ. . ice W.Q. . air cool —I 1 1 1 1 60 120 180 240 300 360 Ageing Time / min. Figure A . 9 Ageing curves (in resistivity) for 7108.72 alloy wi th different quenching conditions. 100 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0078698/manifest

Comment

Related Items