The effect of ultrafine-grain size and titanium hydride addition on mechanical properties of tungsten

The relationship between the grain size and mechanical properties of tungsten alloys was studied for potential improvement in ductility through minimization of grain size and for possible modification to the ductile-to-brittle transition temperature (DBTT) in powder metallurgy-produced alloys. In this study, nanoscale W powders were used to produce near full-density tungsten and several tungsten alloys. Pure W, and alloys of W with 0.5 wt.% and 1 wt.% of titanium hydride (TiH2), were prepared at several reduction temperatures (650 °C, 750 °C, and 850 °C) and were also prepared at several sintering temperatures (1200 °C, 1300 °C, and 1400 °C). Three-point bend (3-PB) tests were used to measure strength and ductility at elevated temperatures and to determine the DBTT of these tungsten alloys. Commercially purchased hot-rolled W was also tested to give some gauge of industrial materials currently in use. Different surface conditions of commercially purchased hot-rolled W samples were tested to determine the relationship between the DBTT and sample's surface finish of hot-rolled tungsten. Correlations were examined between ductility, DBTT, and the interrelated aspects of composition and grain size. Samples containing titanium (Ti) exhibited grain sizes of less than half those of pure W samples at all sintering temperatures tested, presumably due to grain boundary pinning by Ti-rich phases in the microstructure that deterred grain growth.

THE EFFECT OF ULTRAFINE-GRAIN SIZE AND TITANIUM HYDRIDE ADDITION ON MECHANICAL PROPERTIES OF TUNGSTEN by Huan Zhang A thesis submitted to the faculty of The University of Utah in partial fulfillment of the requirements for the degree of Master of Science Department of Metallurgical Engineering The University of Utah May 2018 Copyright © Huan Zhang 2018 All Rights Reserved The University of Utah Graduate School STATEMENT OF THESIS APPROVAL Huan Zhang The thesis of has been approved by the following supervisory committee members: Zhigang Zak Fang , Chair 3/9/2017 Date Approved Sivaraman Guruswamy , Member 3/9/2017 Date Approved Mark Koopman , Member 3/9/2017 Date Approved and by Manoranjan Misra the Department/College/School of , Chair/Dean of Metallurgical Engineering and by David B. Kieda, Dean of The Graduate School. ABSTRACT The relationship between the grain size and mechanical properties of tungsten alloys was studied for potential improvement in ductility through minimization of grain size and for possible modification to the ductile-to-brittle transition temperature (DBTT) in powder metallurgy-produced alloys. In this study, nanoscale W powders were used to produce near full-density tungsten and several tungsten alloys. Pure W, and alloys of W with 0.5 wt.% and 1 wt.% of titanium hydride (TiH2), were prepared at several reduction temperatures (650 °C, 750 °C, and 850 °C) and were also prepared at several sintering temperatures (1200 °C, 1300 °C, and 1400 °C). Three-point bend (3-PB) tests were used to measure strength and ductility at elevated temperatures and to determine the DBTT of these tungsten alloys. Commercially purchased hot-rolled W was also tested to give some gauge of industrial materials currently in use. Different surface conditions of commercially purchased hot-rolled W samples were tested to determine the relationship between the DBTT and sample's surface finish of hot-rolled tungsten. Correlations were examined between ductility, DBTT, and the interrelated aspects of composition and grain size. Samples containing titanium (Ti) exhibited grain sizes of less than half those of pure W samples at all sintering temperatures tested, presumably due to grain boundary pinning by Ti-rich phases in the microstructure that deterred grain growth. TABLE OF CONTENTS ABSTRACT .................................................................................................................... iii LIST OF FIGURES ........................................................................................................ vii LIST OF TABLES ........................................................................................................... x ACKNOWLEDGMENTS ............................................................................................... xi Chapters 1 INTRODUCTION ......................................................................................................... 1 1.1 A Brief Introduction to Tungsten (W) ..................................................................... 1 1.2 Nanometal Materials ............................................................................................... 2 1.3 Challenges .............................................................................................................. 2 1.4 Brief Introduction of Nano Tungsten Process .......................................................... 4 1.5 Goals ...................................................................................................................... 6 2 LITERATURE REVIEW .............................................................................................. 8 2.1 The History of Tungsten.......................................................................................... 8 2.2 Crystallographic Properties ..................................................................................... 9 2.3 Physical Properties of Tungsten .............................................................................. 9 2.4 Chemical Properties of Tungsten ........................................................................... 10 2.5 Mechanical Properties of Tungsten........................................................................ 10 2.6 Advantages and Applications of Tungsten ............................................................. 11 2.7 Disadvantages of Tungsten .................................................................................... 14 2.8 Methods to Improve Ductility of Tungsten ............................................................ 16 2.8.1 Alloying ......................................................................................................... 16 2.8.2 Minimizing Grain Microstructure ................................................................... 17 2.9 Methods to Minimize Grain Size ........................................................................... 19 2.9.1 Top-Down Approach ...................................................................................... 19 2.9.2 Bottom-Up Approach ..................................................................................... 20 3 SCOPE AND OBJECTIVES ....................................................................................... 25 3.1 Scope .................................................................................................................... 25 3.2 Research Objectives .............................................................................................. 26 4 THE EXPERIMENTAL PROCESS ............................................................................ 27 4.1 Brief Experimental Introduction ............................................................................ 27 4.2 Ball Milling Process .............................................................................................. 28 4.3 Drying Process ...................................................................................................... 29 4.4 Forming and Compacting Process ......................................................................... 29 4.5 Sintering Process ................................................................................................... 31 4.6 Characterization Methods ...................................................................................... 33 4.6.1 Scanning Electron Microscopy ....................................................................... 33 4.6.2 Brunauer-Emmett-Teller Specific Surface Area Analysis ................................ 33 4.6.3 Particle Size Analysis ..................................................................................... 33 4.6.4 Vickers Hardness Tester ................................................................................. 33 4.6.5 X-ray Diffraction ............................................................................................ 34 4.7 Sample Characterization........................................................................................ 34 4.7.1 Powder Analysis ............................................................................................. 34 4.7.2 Density Measurement ..................................................................................... 35 4.7.3 Grain-Size Measurement ................................................................................ 36 4.8 Mechanical Test Introduction ................................................................................ 36 4.8.1 A Brief Introduction of the Three-Point Bend Test.......................................... 36 4.8.2 MTS System ................................................................................................... 38 4.8.3 Bend Fixture ................................................................................................... 39 4.8.4 Three-Point Bend Samples Introduction.......................................................... 39 4.8.5 Three-Point Bend Test Setup .......................................................................... 41 5 EXPERIMENTAL RESULTS ..................................................................................... 50 5.1 The Effect of Final Compact Pressure on Density.................................................. 50 5.2 Effect of Reduction Temperature on Density ......................................................... 51 5.3 Effect of Reduction Temperature on Grain Size .................................................... 53 5.4 The Effect of Sintering Temperature on Density.................................................... 54 5.5 The Effect of Sintering Temperature on Grain Size ............................................... 55 5.6 Vickers Hardness .................................................................................................. 56 5.7 The Relationship Between Grain Size and Relative Density Across Three Different Compositions .............................................................................................................. 56 5.8 The Relationship Between Grain Size and Vickers Hardness Across Three Different Compositions ............................................................................................... 57 5.9 SEM and EBSD Images ........................................................................................ 57 5.10 Effect of Surface Condition on Commercially Purchased Hot-Rolled W.............. 58 5.11 Effect of Strain Rate on Commercially Purchased Hot-Rolled W ........................ 60 5.12 Effect of TiH2 Addition on the Mechanical Properties of Sintered W .................. 61 5.13 Fracture Surface Analysis .................................................................................... 62 6 DISCUSSION ............................................................................................................. 79 v 6.1 Analysis and Discussion ........................................................................................ 79 7 SUMMARY ................................................................................................................ 91 REFERENCES............................................................................................................... 93 vi LIST OF FIGURES Figures 1 The relationship between the DBTT and grain size of refractory metals, Nb, Mo, and W. (The curve is redrawn from reference [8].) ...................................................................... 7 2 The relationship between the average grain diameter and DBTT of W [2]. ................. 24 3 The sintering program used in this investigation. ........................................................ 42 4 The custom-designed, ultrahigh-energy planetary ball milling machine. ..................... 44 5 The particle size distributions of pure tungsten powder before and after 6 hours of ultrahigh-energy planetary ball milling. (The upper figure represents the particle size of pure tungsten powder before ball milling, and the lower figure represents the particle size distribution of pure tungsten powder after 6 hours of ultrahigh energy ball-milling.) ..... 45 6 Schematic diagram of the three-point bend test. .......................................................... 47 7 The test setup of the high-temperature three-point bend test........................................ 48 8 The as-received EDM commercially purchased hot-rolled W sample and as-sintered pure W sample using a reduction temperature of 750 °C and a sintering temperature of 1300 °C. ........................................................................................................................ 49 9 The relationship between the different compaction forces and the as-sintered relative theoretical density of pure W, with a reduction temperature of 750 °C, a sintering temperature of 1300 °C, with an H2 atmosphere and switched it to an Ar atmosphere after reduction process. .......................................................................................................... 63 10 The relationship between the reduction temperature and density of three different compositions: W, W-0.5% TiH2, and W-1% TiH2. All samples used the same sintering temperature, 1300 °C, and Archimedes' principle was used to measure the relative theoretical density of each sample. ................................................................................ 64 11 The relationship between the reduction temperature and average grain size of three different compositions: W, W-0.5% TiH2, and W-1% TiH2. All samples used the same sintering temperature, 1300 °C, and the linear intercept method was used to measure the grain size of each sample. .............................................................................................. 64 12 The relationship between the sintering temperature and density of three different compositions: W, W-0.5% TiH2, and W-1% TiH2. All samples used the same reduction temperature of 750 °C, and Archimedes' principle was used to measure the relative density of each sample. .................................................................................................. 65 13 The relationship between sintering temperature and average grain size of three different compositions: W, W-0.5% TiH2, and W-1% TiH2. All samples used the same reduction temperature of 750 °C, and the linear intercept method was used to measure the grain size of each sample. .............................................................................................. 65 14 The Vickers hardness of as-sintered W, W-0.5% TiH2, and W-1% TiH2 samples with a reduction temperature of 750 °C and sintering temperature of 1300 °C. ...................... 66 15 The relationship between the average grain size and relative density of W, W-0.5% TiH2, and W-1% TiH2 with the same sintering temperature, 1300 °C, but three different reduction temperatures ranging from 650 °C to 850 °C. ................................................. 66 16 The relationship between the grain size and Vickers hardness of W, W-0.5% TiH2, and W-1% TiH2 with a reduction temperature of 750 °C and a sintering temperature of 1300 °C. ........................................................................................................................ 67 17 An SEM image of polished W-0.5% TiH2 with 0.1 surface finish prepared at a reduction temperature of 850 °C and sintering temperature of 1300 °C. ......................... 68 18 An EBSD image of polished W-0.5% TiH2 with 0.1 surface finish at a reduction temperature of 850 °C and sintering temperature of 1300 °C. ........................................ 68 19 SEM images of as-sintered W, W-0.5% TiH2, and W-1% TiH2 with 0.1 μm surface finish using a sintering program with a reduction temperature of 750 °C, sintering temperature of 1300 °C, and a flowing H2 atmosphere that is then switched to an Ar atmosphere after reduction............................................................................................. 69 20 Three-point bend stress-strain curve of a commercially purchased hot-rolled W sample without polishing (as-received hot-rolled surface condition) at test temperatures ranging from room temperature to 300 °C with a strain rate of 0.3 mm/min. .............................. 70 21 Three-point bend stress-strain curve of a commercially purchased hot-rolled W sample with 1 μm surface finish at test temperatures ranging from room temperature to 300 °C with a strain rate of 0.3 mm/min. ................................................................................... 71 22 The flexural strain-test temperature curves of both polished (1 μm surface finish) and unpolished (as-received hot-rolled surface condition) commercially purchased hot-rolled W samples with a test strain of 0.3 mm/min................................................................... 72 23 Three-point bend stress-strain curve of an unpolished commercially purchased hotrolled W sample (as-received hot-rolled W surface condition) at test temperatures ranging from room temperature to 300 °C with a test strain rate of 0.1 mm/min. ........................ 73 viii 24 Flexural strain vs. test temperature curves of unpolished (as-received hot-rolled surface condition) commercially purchased hot-rolled W samples at test temperatures ranging from room temperature to 300 °C with a strain rate of 0.1 mm/min and 0.3 mm/min. ........................................................................................................................ 74 25 Three-point bend stress-strain curve of a sintered pure W sample with polish at 1 μm surface finish condition and at test temperatures ranging from room temperature to 600 °C with a test strain rate of 0.3 mm/min. ................................................................. 75 26 Three-point bend stress-strain curve of a sintered W-1%TiH2 sample with polish at 1 μm surface finish condition and at test temperatures ranging from room temperature to 450 °C with a test strain rate of 0.3 mm/min. ................................................................. 76 27 SEM images of room-temperature fracture surfaces of as-sintered pure W and W-1% TiH2, sintered samples that are using the same sintering program with a reduction temperature of 750 °C, a sintering temperature of 1300 °C, and an H 2 atmosphere that is the switched to an Ar atmosphere after reduction. .......................................................... 77 28 SEM images of as-sintered pure W and W-1% TiH2 fracture surfaces at 450 °C. These postsintered samples use the same sintering program with a reduction temperature of 750 °C, a sintering temperature of 1300 °C, and an H2 atmosphere that then switches to an Ar atmosphere after reduction. .................................................................................. 78 29 The relationship between oxygen content and the reduction temperature of as-sintered W, W-0.5 %TiH2, and W-1% TiH2 when all have the same sintering temperature of 1300 °C. ........................................................................................................................ 86 30 The relationship between average grain size and Vickers hardness of as-sintered W, W-0.5 %TiH2, and W-1% TiH2 when all have the same sintering program, with a reduction temperature of 750 °C and a sintering temperature of 1300 °C. ...................... 87 31 TEM images of grain morphology of a commercially purchased hot-rolled W bend sample. (These TEM images are provided by Lei Xu.) .................................................. 89 32 Grain-size distribution of as-sintered W and W-1% TiH2 samples with reduction temperatures of 750 °C and sintering temperatures of 1300 °C. ..................................... 90 ix LIST OF TABLES Tables 1 Crystal structures and relevant lattice parameters of tungsten [12] .............................. 23 2 Main physical properties of tungsten [12] ................................................................... 23 3 Impurity composition in raw tungsten powder. ........................................................... 43 4 Specific surface area and average particle size of powders before and after 6 hours of ball milling. ................................................................................................................... 43 5 The chemical composition of commercially purchased hot-rolled W, which is given by the Plansee company. .................................................................................................... 46 6 Basic characteristics of four types of tungsten materials investigated in this work. They are all sintered bend samples with the same sintering program: a reduction temperature of 750 °C, a sintering temperature of 1300 °C, and flowing H2 that is switched to an Ar atmosphere after reduction............................................................................................. 88 ACKNOWLEDGMENTS I would like to express my sincere gratitude to Professor Zhigang Zak Fang for providing me with this opportunity to pursue my master's study under his guidance. Professor Fang is an optimistic and responsible person and always provided me with great suggestions during my research. He also shared his experiences and knowledge to help me overcome difficulties in my work. I would also like to express my thanks to Dr. Mark Koopman for direction and support. His guidance and encouragement helped me progress in my research project. I would like to express my thanks to Dr. Sivaraman Guruswamy, one of my committee members, for his valuable suggestions and patience in guiding my thesis. I also want to thank Dr. Chai Ren for his expertise and in-lab support, which made my research process go smoothly. He also provided many valuable suggestions and much-appreciated advice for my research work. He mentored me to overcome problems myself, which greatly improved my ability to solve practical problems. I also would like to thank everyone in Fang's group, who always gave me help and suggestions on my research work. The harmonious working conditions made my work easy and efficient. I also want to say thank you to Lei Xu for his help on the TEM images. I want to say thank you to my parents, who always give me encouragement and spiritual support during my time abroad. Finally, I would like to express my thanks to the U.S. Department of Energy (DOE) (DESC0008673), Office of Science, Fusion Materials Program and the U.S. Department of Defense (DOD), SBIR Program for financial support. xii CHAPTER 1 INTRODUCTION 1.1 A Brief Introduction to Tungsten (W) Tungsten's elemental symbol is W, and its atomic number is 74. It has not been found in its elemental form in nature. Usually it is extracted from the tungsten minerals like wolframite and scheelite. Wolframite is a monoclinic mineral containing ferberite (FeWO4) and hubnerite (MnWO4); it is also the main ore for extracting tungsten. Tungsten has a steel-gray and silvery white metallic luster. Pure tungsten has the highest melting point among all metals, which is about 3685 K. Tungsten has a number of exceptional properties, including high density, good hardness, and good high temperature strength [1]. The applications of tungsten have continually increased since 1957 due to its high melting point and the excellent properties of tungsten materials. Prior to this time, tungsten was primarily used in incandescent light bulb filaments, electronic tubes, and welding electrodes. Currently, tungsten is more used in varied applications such as rocket nozzles, cutting tools, and ionization surfaces for electrical propulsion in space [2]. Tungsten is also an ideal plasma-facing material in the fusion reactors applications [3]. In the past five years, the price of tungsten has continually decreased. The continually decreasing price of tungsten raw materials also motivates the production and 2 development of tungsten materials, giving tungsten a promising future. 1.2 Nanometal Materials In the mid-1980s, Gleiter [4] made a visionary argument that, if metals and alloys can be made into nanocrystalline size, they would have a number of appealing mechanical characteristic for all fundamental and structural applications. Compared to the conventional coarser grain metal materials, the metal with nanostructure and ultrafine grain has high yield strengths and fracture strengths, superior wear resistance, and possibly superplastic formability at low temperatures. The deformation mechanisms might also change due to the existence of the nanoscale grain size. For the nanoscale materials, the deformation mechanism may be changed to the grain-boundary deformation process [4]. −1 According to the famous Hall-Petch equation, 𝜎𝑦 = 𝜎0 + 𝐾𝑑 2 (1), where d is the average grain diameter, σy is the yield strength, σo is the material-dependent constant for the starting stress for dislocation movement, and K is the material-dependent constant of strengthening coefficient, the yield strength of the metal is increased with decreased grain size [4,5], and a reasonably good ductility is also one of the amazing attributions of getting metal into nanoscale [4]. 1.3 Challenges Tungsten has many good properties. However, tungsten also has many weaknesses. A major drawback of tungsten is that it is very brittle at room temperature [1]. Tungsten is inherently brittle and generally shows intergranular fracture at low 3 temperature and radiant environment. Another problem associated with the tungsten materials is its relatively high ductile-to-brittle transition temperature (DBTT) and relatively low recrystallization temperature [6], which is just around 1200 °C. For the tungsten materials, the boundaries between grains are usually the weakest part, and due to the inherent brittleness of tungsten, cracks always formed in the grain boundaries and propagate along the grain boundaries. Besides tungsten's relatively high DBTT, when the temperature of the working environment is lower than DBTT, tungsten material will show little or no macroscopic plastic deformation prior to fracture [4,6-7]. However, when the working temperature is higher than the DBTT, tungsten will show plastic deformation in most situations. The mechanical behavior of tungsten is characterized by its DBTT. Due to the importance of the DBTT for tungsten material, some researches have already studied the relationship between the DBTT and grain size of tungsten in order to improve the comprehensive properties of tungsten material. L. L. Seigle and C. D. Dickinson have shown the relationship between DBTT and the average grain size of tungsten in their paper, which shows that the DBTT and grain size of tungsten is in a nearly linear relationship [8]. Refining grain size of tungsten can be considered as a great method to decrease the DBTT and improve the mechanical properties of tungsten by changing the deformation mechanism and decreasing the impurities concentration among the grain boundaries [9-11]. However, there is no published literature on the DBTT measurement of ultrafine-grain as-sintered W samples without mechanical working. In this investigation, the author wants to study the relationship between DBTT and average grain size of the as-sintered W, and find a method to decrease the DBTT and improve the 4 mechanical properties of the as-sintered W. In this investigation, during the preparation process, the most difficult part is to make the as-sintered W samples in high density and also combined with ultrafine grain size. Except for the manufacturing problem, there are also many other technical problems associated with it, so the road of improving the mechanical properties and decreasing the DBTT of the as-sintered W by decreasing its grain size is still facing several challenges. 1.4 Brief Introduction of Nano Tungsten Process One objective of this investigation is to find a process that could improve the mechanical properties, especially the ductility of the as-sintered W without mechanical working. Another objective was to deeply study the relationship between the grain size and mechanical properties of the as-sintered W. The three-point bend (3-PB) test was chosen to measure the DBTT of the as-sintered W bend sample. In this investigation, powder metallurgy (PM) was chosen to produce W bulk samples due to the high melting point of W. Mechanical alloying (MA), titanium hydride (TiH2) addition, and a relatively low sintering temperature program were all used to help the as-sintered W to achieve near-full density with ultrafine grain size or nano-microstructure. MA is a solid-state powder-processing technique for producing homogeneous material. During the process of MA, powders are mixed and crushed by collision between metal balls. Both particle size and grain size can be reduced by the collision between balls and powders. In this investigation, nano-tungsten powders and heptane are mixed together for the milling process. There are many advantages of using wet milling for MA: (1) It can reduce the oxidation of W powder 5 (2) It can prevent the agglomeration of W powder. (3) It can decrease powder flying and greatly improve the working environment. Although MA is the most popular method for preparing metal powders, it still has some shortcomings due to the ultrahigh energy during the milling process and collisions between the balls. So, the introduction of contaminants is inevitable during the high energy milling process. This is also the reason why tungsten carbide balls were chosen as the milling balls in this investigation. Tungsten carbide balls can reduce the risk of introducing other impure elements during the milling process. After milling process, a conventional axial mold is used to compact bulk tungsten sample. In the sintering process, due to the high melting point and low diffusion coefficient of W, the grain of W has a strong tendency to grow during the high-temperature sintering environment, which makes keeping the grains at nanosize very difficult. In order to perfectly minimize the grain size, grain-growth inhibitor was used to control grain growth during the sintering process in this investigation. Nano grain size W powders was used and sintered under conventional press and at a relative low sintering temperature to make high-density and fine-grainsized as-sintered W sample. In the practical industrial production, this PM method could also greatly decrease the production cost, saving cost. During the whole manufacture process, the sintering process plays an important role in the full cost, so the low sintering temperatures with conventional press can raise the efficiency of the economic development and also reduce waste of powder and gas. Sintering W into nearly full density at low temperature requires the use of ultrafine-grain W powder that has a very high surface activity. A custom-designed, ultrahigh-energy, planetary ball-milling (HEPM) machine was used in this investigation to help W powder mill into nanosize. 6 The W powder took on many lattice distortions and defects during the HEPM process. Lattice distortion in the powder can decrease the diffusion distance between atoms and also improve the driving force to promote densification during the sintering process [12]. Solid-phase sintering was used for nano-tungsten powder sintering to avoid overly high temperatures, which cause the rapid growth of grain size. The sintering temperature of solid-phase sintering is below the liquid-phase temperature. The general W powder with common ball milling process generally is unable to obtain full density for W when just using solid-phase sintering, but with the help of HEPM machine, the W powder can achieve nanoscale, which makes full densification of the W sample possible at relative low sintering temperatures. 1.5 Goals In this investigation, the additive TiH2 is worked as the grain size inhibitor to suppress the grain growth of W grains. The effect of different sintering conditions and different amounts of TiH2 additions on the density, hardness, and grain size of as-sintered W samples was studied. The main goal of this investigation is to find out a process to prepare nearly full-density W with ultrafine grain size in order to improve the mechanical properties of the as-sintered W. A 3-PB test at an elevated temperature is used to measure the mechanical properties and DBTT of each as-sintered W sample. The effect of TiH2 additions on the mechanical properties of low-temperature as-sintered W is also studied. The DBTT of each as-sintered W sample was measured in an attempt to discern the relationship between grain size and mechanical properties, especially the relationship between the grain size and DBTT of as-sintered W without mechanical working. 7 Figure 1. The relationship between the DBTT and grain size of refractory metals, Nb, Mo, and W. (The curve is redrawn from reference [12].) CHAPTER 2 LITERATURE REVIEW 2.1 The History of Tungsten In 1755, A. F. Cronstedt used the Swedish words "tung" and "sten", meaning "heavy stone", to describe high-density tungsten [13]. In 1781, the Swedish chemist Carl Wilhelm Scheele found that the nature of the composition of tungsten ore is actually the compound of lime with a peculiar acid. Scheele also discovered scheelite. In 1783, two Spaniard brothers, J. J. and F. de Elhuyar, found that tungsten acid can be extracted and isolated from wolframite. In the same year, tungsten powder was first made by reduction of tungsten trioxide by carbon [13]. In 1786, H. I. Duhamel du Monceau discovered that adding tungsten would harden steel. The year 1847 marks the starting point of tungsten industrialization. In 1855, J. Jacob and Koeller made the first attempts to produce tungsten steel. Tungsten steel has high hardness, strength, good wear resistance, and great heat resistance. Tungsten steel is typically used in drilling bits and cutting tools. In 1904, Just and Hannaman found a process of mixing a metal powder with an organic binder to build tungsten filaments. Until 1927, the extreme hardness of tungsten carbide (WC) had been known. Krupp offered the first cemented carbide (WC-6%Co) at the Leipzig tool fair [14]. In summary, the continual developments in industrial manufacture gave tungsten 9 technology an avalanche-like growth. Tungsten technology underwent a huge change from the 19th to the 20th century [14]. 2.2 Crystallographic Properties Tungsten is a transition metal. The atomic number of tungsten is 74 in the periodic table of elements, and the number of electrons is 74. The electron distribution within shells and subshells of the tungsten atom is shown below: (1s)2(2s)2(2p)6(3s)2(3p)6(3d)10(4s)2(4p)6(4d)10(4f)14(5s)2(5p)6(5d)4(6s)2 Tungsten has a body-centered cubic (BCC) lattice. Besides amorphous tungsten, three phases are known; they are σ-W, β-W, and ɣ-W. The σ-W phase has a stable BCC structure under standard temperature and pressure, and σ-W is the only stable phase. β-W is a metastable phase and converts to σ-W when heated above 600 °C to 700 °C, and this conversion is an irreversible process. The ɣ-W phase is found only in thin sputtered layers at the beginning step of sputtering and will transform to the σ-W phase when heated above 700 °C [14]. The crystal structures and relevant lattice parameters of tungsten are shown in Table 1 [12]. 2.3 Physical Properties of Tungsten Tungsten has a series of exceptional physical properties, like high melting point, which is around 3410 °C, the highest melting point among all nonalloyed metals. The density of pure tungsten is similar to that of gold at about 19.3 g/cm3. Tungsten also has low sputtering rate, high tensile strength, good hardness, electrical conductivity, high thermal conductivity and electron emission performance, low vapor pressure, and a low 10 coefficient of thermal expansion [1]. The main physical properties of tungsten are listed in Table 2 [12]. 2.4 Chemical Properties of Tungsten W has been used in many fields due to its exceptional physical properties. The chemical properties of W are also very important because they can determine and limit the metal's application across diverse environments. W can be considered as a rather inert metal that is resistant to many elements and compounds [14]; for example, W will not react with oxygen or water at room temperature, and W is insoluble in hydrochloric, sulfuric, and nitric acids. W is stable in mineral acids at low temperatures, and it will only slightly react at high temperatures [14]. In general, the chemical properties of W are very stable. The reactions of W with some normal gases are shown in detail below. W and carbon dioxide (CO2) will start to react at 1200 °C, and W will oxidize to WO3 through CO2. Under carbon monoxide (CO), W is stable under 1400 °C; when the temperature is higher than 1600 °C, it will produce carbide (WC, W2C). When the reaction temperature is above 400 °C, W will react with sulfur vapor and H2S to form WS2. The reaction will become violent when the temperature is higher than 700 °C. W can react with nitrogen when the temperature is higher than 1500 °C, and it will form WN2 when the temperature reaches 2300 °C [12]. 2.5 Mechanical Properties of Tungsten As mentioned before, W is very brittle at room temperature, which is also the biggest challenge limiting the practical applications of W. The inherent brittleness of W 11 decreases its use in structural applications. According to published literature, the lowtemperature mechanical properties of W have been found to be highly dependent on the purity of the material, the manufacture processing, and the microstructure of the material. For example, chemical composition [15-17], average grain size [8, 18-21], and thermomechanical processing and treatment [19, 21-25] have been reported to greatly influence the mechanical properties of W materials. The manufacture route and thermomechanical process are considered the most important factors in the performance and properties of pure W [26]. The main production methods for pure W materials are casting and PM. Most W products are produced through the PM process due to the technical difficulties associated with the high melting point of W [26]. The mechanical properties of W are highly dependent on the fabrication route; when the different pressing types and different deformation degrees are used, the mechanical properties of W would make it different. Many methods, especially the mechanical working process, have been tried to improve the ductility of W. However, due to the relatively low recrystallization temperature of W, the significant deformation at high temperatures will cause W to recrystallize, making it become brittle again. 2.6 Advantages and Applications of Tungsten Benefiting from a high melting point, W can be applied in some cruel conditions, like extreme high-temperature environments. The use of W in high-temperature incandescent lightbulb filaments is a famous example. This was a revolutionary advancement in the world, providing the first widespread use of artificial lighting. W also has many excellent properties in fusion reactors [27,28], spallation neutron source solid 12 targets [29,30], and high-power density structural materials [28]. Outstanding properties of W include a high melting point, the lowest vapor pressure (Pv=1.3 ×10-7 Pa at Tm), low sputter yield, high thermal conductivity, good thermal shock resistance, good hightemperature strength, and nearly zero tritium retention. W is also considered to be the foremost material candidate for diverters and plasma-facing components in fusion reactors [1,31]. With the combined function of the high melting point, and high density of W, W is also considered to be an ideal material in some military areas and army applications. Tungsten-based alloys, due to their excellent properties, have received significant attention as a potential material to replace depleted uranium (DU) in kinetic energy penetrator (KEP) applications. This ammunition does not contain explosives; it uses kinetic energy (mv2) to destroy the target. However, due to its inherent brittleness and deformation properties at ballistic strain rates, using tungsten-based alloy to replace DU still faces many difficulties [32]. DU has been used as a raw material for armor-piercing bullets and KEPs, but serious environmental pollution problems limited the application of DU. Due to their high density, high melting point, and high-temperature strength, W and tungsten-based alloys are considered ideal candidates for making armor-piercing bullets and KEPs in army applications. For KEPs, kinetic energy powers the penetrator at the launching process. At the same conditions, the penetrator can provide stronger destructive power with higher density compared to other metal materials with lower density. Therefore, W and tungsten-based alloys have received significant attention as potential materials to replace DU in KEP applications. Polycrystalline W heavy metals were developed for radiation containment, like in 13 fusion reactors. They were subsequently used extensively in penetrator applications and as target material as a result of their combination of strength, density, thermal expansion and conductivity, machinability, and especially their excellent strength and modulus retention at high temperatures, refractory characteristics, low tritium retention, and low sputtering yielding [26, 33-36]. Although W is usually not used for structural purposes due to its brittleness at low temperatures, W and tungsten-based alloys are ideal materials for making military weapons; therefore, the development and improvement of the mechanical properties of W materials become extremely urgent. A lot of researchers are working on the improvement of mechanical and structural properties of W. Various W alloys have been made to satisfy different working needs. Due to the high-hardness properties and high-temperature strength of W materials, tungsten alloys are widely applied in tool materials and drilling bits. For example, the tungsten alloy WC-Co, a kind of cemented carbide that is usually used as cutting material, has a series of great properties, like very good hardness, good strength and wear resistance, and heat and corrosion resistance. Another tungsten alloy, W-Cu, has good thermal and electrical conductivity and is usually used in integrated circuit and electronic materials. Tungsten steel, with its hardness and wear resistance, is mainly used in making all kinds of tools. Tungsten steel has many different types, like wire-drawing die steels and high-speed steels. The typical chemical elements of wire-drawing die steels are tungsten, chromium, carbon, and manganese. These steels are characterized by a comparatively high percentage of carbon, approximately two percent. High-speed steel tools can maintain a sharp cutting edge even at high working temperatures [13]. 14 In general, W and tungsten alloys are not only heavily integrated into everyday life, but they also have strong potential to contribute to revolutionary progress for society. 2.7 Disadvantages of Tungsten W has many advantages, as mentioned above; however, W also has some disadvantages. The inherent shortcomings of W greatly limit the development and application of W materials. The lack of room-temperature ductility has been a major issue that limits the use of W in many applications, especially in army applications. W is very brittle and always shows intergranular fracture at room temperature. W also has a relatively high DBTT, which is well above room temperature. W also has a relatively low recrystallization temperature, which is around 1200 °C. When the surrounding temperature is higher than 1200 °C, it will cause the recrystallized brittleness of tungsten. Irradiation will also induce embrittlement of tungsten [37-39]. It is already reported that the DBTT of the as-sintered W is above 400 °C (disc-shaped compact tension) [40-42]. The DBTT is actually a range of temperatures. When the material works at a temperature below its DBTT, the material shows brittle behavior and has a great tendency to shatter. However, when the working temperature is higher than the material's DBTT, the material will show ductile behavior and might show a plastic deformation mechanism. DBTT is a very important characteristic for metal materials. Once a material is cooled below the DBTT, it has a much greater tendency to shatter instead of bending or deforming. The most famous incident was the significant hull cracking of BCC steel rivets used in the production of Liberty ships during World War II. When the Liberty ships literally broke in the middle, the government first thought those ships were being 15 sunk by German submarines; however, after the engineer checked the fracture surface, the real reason for the brittle fractures was revealed: the low-temperature brittleness of the steel rivets holding the hull plating together. The first crack was shown at the welded part between steel plates. The ship broke very quickly due to the brittle fracture of the steel. At that time, the temperature of the cold seawater was lower than the DBTT of steel, and caused the steel plate, which was ductile at room temperature, to become brittle and then break. The result was a tragic loss of life and property. The brittle transition of materials sometimes causes serious consequences. The lack of ductility greatly limited the application and development of W. However, the continually increasing demand for high performance of structural material gives W a promising future. The need for material working in cruel environments is continually increasing. Sometimes, the material is needed to work in some extremely severe environments, like relatively low-temperature environments or very hightemperature conditions, or even some conditions of high radiation concentration. The surrounding working environment and working temperature might have a very important impact on the mechanical properties of materials-as seen with the Liberty ship in cold seawater. Usually, the mechanical properties of materials will change with the change of the surrounding temperature. Therefore, study of DBTT is very important for metal materials, especially the metals that are brittle, like W. W has a relatively high value of DBTT, and this limits the application of W materials on low working temperatures. However, it is also found that recrystallized equiaxed grains will cause W to become brittle [33, 43-47]. Thus, in order to obtain better performance, recrystallization is 16 avoided during the productive process of W. In general, the disadvantages of W materials are the inherent low-temperature brittleness, high DBTT, and a relatively low recrystallization temperature. 2.8 Methods to Improve Ductility of W In this investigation, the author wants to find a method to improve the ductility of the as-sintered W without mechanical working. The reason why the author focuses on the as-sintered W is because the high-melting-point metals like W are usually produced by the PM method in industry. However, the existence of porosity and impure substances in the as-sintered W cannot be eliminated, and the porosity and impurities will aggravate the brittleness of W materials. The lack of ductility of W is the main problem that limits the use of W in many applications. Many researchers have studied to find a process for improving the mechanical properties of W, especially improving the ductility and decreasing the DBTT of W. In general, there are two approaches that could improve the mechanical properties and may increase the ductility of W materials, and they are called alloying and minimizing grain microstructure, respectively. 2.8.1 Alloying A large number of tungsten alloys were studied in the past, but only some of them achieved technical improvement. The aim of tungsten alloying is to improve its chemical, physical, and mechanical properties at both ambient conditions and elevated temperatures. Low-temperature brittleness is the biggest crucial challenge in the manufacture of pure W metal [14]. 17 Tungsten-rhenium (W-Re) alloy has been studied, and it has been proved that the addtive Re can improve the ductility of W; W-Re alloy does have higher ductility than pure W. The ductility of W at room temperature can reach greater than 20% when the content of Re in W is greater than 20%. However, rhenium is a rare element and very expensive, which limits its wide use in practice, and W-Re also has a comparatively higher radiation-induced embrittlement rate [26,48]. Titanium (Ti) has also been considered to add into W for an alloy to improve the mechanical properties of W. The WTi alloy is fully dense, and the average grain size and grain distribution width are significantly decreased compared with pure W [37]. The effect of the addition of tantalum (Ta) on the mechanical behavior of W has also been studied; it has been determined that the addition of Ta enhances the mechanical performance of pure W. WTa alloys have good corrosion resistance and better hot strength [14]. W-5%Ta alloy is three times harder than the pure W sample. However, the addition of Ta has a negative effect on the DBTT [49]. The effect of potassium-doped tungsten (K-doped W) has also been studied; K-doped W showed that the addition of K can improve the mechanical properties of W; it lowers the DBTT and increases the yield strength [50]. 2.8.2 Minimizing Grain Microstructure Besides alloying method, the ultrafine grain microstructure of W makes decreasing its DBTT possible. Much literature shows that the grain size can have a great influence on the DBTT of W; some investigations reported that the DBTT of W can be decreased by decreasing the grain size. Grain size might play an important role in decreasing the DBTT of W. The DBTT has been reported by much literature to decrease 18 with decrease in the grain size. L. L. Seigle and C. D. Dickinson [8] show that by making the tungsten wire at a grain size of around 0.005-0.015 mm by using undoped tungsten wire swaged to 0.01 inch and recrystallized at 1400 °C, the DBTT of the tungsten is between 353k to 448k, which shows that the DBTT of tungsten can be decreased by decreasing the grain size. However, other investigators think that the DBTT of tungsten is decreased with increased grain size at the large grain scale. Klopp, William D, and Witzke [51] have noted that the measured DBTT of electron-beam-melted (EBM) W is around 546-666 K when the grain size of the EBM W is around 0.1 to 1 mm. This result shows that the DBTT of EBM W is increased when the grain size of W is decreased from 1 mm to 0.1 mm. Figure 2 shows the relationship between the average grain diameter and DBTT of W by summarizing the literature that reported the relationship between the grain size and DBTT of W [2]. It shows that for the grain size less than 0.1 mm, the DBTT of W is decreased with decreased grain size, and for grain size larger than 0.1 mm, the DBTT of W is decreased with increased grain size. In Figure 2, when continually decreasing the grain size of W, the DBTT of W is first increased and reaches its maximum value and is then decreased, it shows clearly that when the grain size of W is smaller than 0.1 mm, the DBTT of W is decreased when decreasing the average grain size. Utilizing nanoscale advantage to improve mechanical properties of W has become a mainstream practice in making the as-sintered W by PM method. Furthermore, due to a series of benefits of nanoscale grain size, ultrafine-grain W has become the critical method for improving the ductility of as-sintered W materials. In general, there are two 19 ways usually used to achieve the nano or ultrafine microstructure of W materials. One of the approaches is called the top-down approach, and the other one is called the bottom-up approach. In this investigation, the author wants to study the relationship between the grain size and DBTT of the as-sintered W, and pay attention to the grain size range below 5 μm to study the relationship between the grain size and DBTT of as-sintered W without mechanical working. 2.9 Methods to Minimize Grain Size 2.9.1 Top-Down Approach The top-down method is a process that decreases grain size by the severe plastic deformation (SPD) method. In many cases, SPD was used as the most effective way to produce nano-grain size of W. Furthermore, according to the experimental record, the original coarse grain size of W can be effectively refined into 100 nanometers through the SPD process. Among SPD methods, equal-channel angular pressing (ECAP) and high-pressure torsion (HPT) attracted the most attention in the field due to their good effects on refining grain size. ECAP is the most effective way to decrease grain size; the process decreases grain size through the pure shear deformation. The advantages of ECAP are that the deformation process doesn't change the cross section of the sample and makes the redeformation possible. HPT is a new method for transforming metal into ultrafine grain materials. In this method, the material is set between the plunger and substrate and receives pressure and rotated torsion, decreasing the grain size. 20 The grain size of W specimens can decrease to about 0.3-1.5 μm and 0.5-2.0 μm from the original grain size of 3-20 μm during the two passes of ECAP at 800 °C and three passes of ECAP at 950 °C, respectively [42]. This confirmed that ECAP is effective at refining the grain size of W samples. ECAP-strained W shows simultaneous improvement in both strength and ductility as grain size is refined. The DBTT of the ultrafine-grain tungsten after ECAP processing at 800 °C and 950 °C is decreased to 386 °C and 322 °C, respectively [42]. ECAP is a promising technique for producing lowDBTT bulk ductile W [47]. ECAP can greatly reduce the grain size of W. However, the process of ECAP is expensive and it is impractical for manufacturing. A lot of strain energy is introduced during the ECAP process, and the inside strain energy is very easy to become the driving force for recrystallization, causing the recrystallization brittleness of W. 2.9.2 Bottom-Up Approach The second approach for achieving nano-microstructure is called the bottom-up approach, in which nanosized W powders are used. This is also the approach that was used in this investigation to achieve the nano-grain size of W and to improve the ductility and reduce the DBTT of W. The main part of the bottom-up method is to have nanoscale powder first, and benefit from those nanoscale powders, which also make sintering W into ultrafine grain possible. Nanoscale powder instead of coarse powder is used to achieve the as-sintered W into nano-microstructure. The most crucial part of this bottom-up method is to obtain the nanoscale metal powder. Ball milling is the main approach to obtain the ultrafine and 21 nanoscale metal powder. In order to continue minimizing the grain size of powder, there are some auxiliary methods that can also help to minimize grain size. These methods include using special sintering techniques or adding grain-growth inhibitors. One special sintering technique, hot isostatic pressing, is a sintering technique that combines the action of high temperature and high pressure and can finish the sintering process in a very short time. With shortened sintering time, grain growth during the sintering process can be controlled and decreased. The second way to decrease grain growth is to add some grain-growth inhibitors to suppress the progress of grain growth during the sintering process. Carbide and oxide are the two most common grain growth inhibitors that are used. Many researchers have studied the effects of yttrium oxide (Y2O3), lanthanum oxide (La2O3), and thorium dioxide (ThO2) on the mechanical properties of W. In recent studies, the focus was on manufacturing nanostructured W materials by dispersing oxide nanoparticles into the W matrix [38,52,53]. ThO2 is a great oxide that can improve the mechanical properties of W, and W-ThO2 has relatively high tensile strength and creep strength at 2000 °C. However, ThO2 is radioactive, so recently, La2O3 and Y2O3 have been used to replace ThO2 to improve the properties of W. It has been shown that both La2O3 and Y2O3 can effectively inhibit grain growth and reduce grain size in W [38]. Itoh and Ishiwata studied the effect of sintering temperature on the particle size of Y2O3 and the grain size of W by investigating the microstructure of developed Y2O3 dispersed tungsten alloy [54]. Y2O3 is the most popular oxide addition for reducing grain size and improving the properties of W. The grain size of W can be decreased with an increase in the additive volume of Y2O3 particles. Adding Y2 O3 into W can not only increase density 22 but also reduce grain size. However, the effect of Y2O3 on bending strength depends on sintering temperature. The bending strength of Y 2O3 dispersed tungsten alloy decreases as the sintering temperature increases. The addition of Y2O3, however, greatly reduces the oxidation rate at high temperatures, and the strength and toughness retention of WY2O3 is much better than pure W [37,54]. After much research, it has been found that existing pores and impurities, like the elements carbon (C) and nitrogen (N), will cause the brittleness of W. C and N will concentrate at the grain boundary area and cause poor connection between the two grain boundaries. The interstitial impurities like C, N, O, and H have very little solubility in the W, so they segregate very easily on the grain boundary, which seriously influences the mechanical properties of the as-sintered W. Improving the purity of W powder, especially decreased interstitial impurities, can possibly improve the mechanical properties of W. Besides that, the mechanical properties of W are also strongly dependent on the sample's surface condition. Joseph R. Stephens [55] has shown that, when the surface of the W specimen is removed by electropolishing, it can increase the room-temperature bend ductility. If the condition of the surface is improved, the brittleness situation of W can be greatly refined. The reason for this improvement can be attributed to some of the surface defects being removed during the polishing process. Compared with the top-down approach, like ECAP process, the cost of the bottom-up approach is much less expensive, and the bottom-up approach is also suitable for the industrial manufacture and complexly shaped products because it can greatly decrease the volume of metal removed by cutting. 23 Table 1. Crystal structures and relevant lattice parameters of tungsten [12] Metal Lattice type name Tungsten (W) BCC Coordination number K8 Lattice constant (nm) 0.31652 Table 2. Main physical properties of tungsten [12] Atomic number 74 Relative atomic mass 183.86 Outer electronic structure 5d46s2 Density (g/cm3) 19.3 Melting point (K) 3683±20 Boiling point (K) 5973±200 Atomic radius (nm) 0.13706 24 Figure 2. The relationship between the average grain diameter and DBTT of tungsten. (This curve is redrawn from reference [2].) CHAPTER 3 SCOPE AND OBJECTIVES 3.1 Scope Because W has many great properties, and it cannot be easily replaced by other metals in military applications. Researchers have tried many methods to improve the mechanical properties of W, especially the ductility. ECAP and hot-rolling processes are the most popular methods for refining grain size and improving ductility of W. Most of the literature shows that when the W has undergone ECAP and hot-rolling process, the grain size of W can be greatly refined and also shows a much more improved ductility. In addition, W that has undergone ECAP and hot-rolling shows a decrease in DBTT. However, both ECAP and hot-rolling introduce plastic deformation into W. There is not any published literature on the DBTT measurement of ultrafine-grain as-sintered W without mechanical working. The effect of mechanical working on tungsten's grain size is not just a single impact factor. It is unknown if the lowered DBTT of mechanically worked W is due to the ultrafine grain size or the deformation process. Furthermore, no one has tried to measure the DBTT of ultrafine-grain W without mechanical working to figure it out. Does the reduction of grain size into an ultrafine grain range affect the DBTT of the as-sintered W without mechanical working? Founded on curiosity and the lack of answers in previous studies, the present research seeks to discover the answer to 26 this question. 3.2 Research Objectives In order to improve the mechanical properties of W, especially the as-sintered W, the author wanted to find a process that can be used to produce near-full density with nano-grain-size as-sintered W. As mentioned above, no one has tried measuring the DBTT of ultrafine-grain, as-sintered W without mechanical working. In this investigation, the author explored this issue to find out if the reduction of grain size can improve the mechanical properties of the as-sintered W. In addition, the author wants to focus on the relationship between the grain size and DBTT of as-sintered W without mechanical working. In order to improve the density and reduce the grain size of the as-sintered W, different sintering parameters and the addition of TiH2 were used to control the density and grain size of the as-sintered W sample. The 3-PB test at elevated temperature was used to measure the DBTT of each as-sintered W sample to observe the relationship between the grain size and DBTT. In general, the two main objectives of this research were, first, to study the relationship between grain size and DBTT of as-sintered W without mechanical working and, second, to study the effects of adding TiH2 and of using different sintering parameters on the density and grain size of the as-sintered W. It was important to build a mechanical system that could be used to measure the DBTT. This custom-built MTS system can be used to measure the DBTT of other metals in the future. CHAPTER 4 THE EXPERIMENTAL PROCESS 4.1 Brief Experimental Introduction In this investigation, a bottom-up approach was used to sinter the ultrafine-grain W bulk samples. A custom-designed, unique, HEPM machine was used to obtain nanosize W powder, which makes sintering W at low temperatures more easy. A relatively lowtemperature sintering program was used to maintain the ultrafine grain size of the assintered W and also slow down the grain-growth rate of W sample during the sintering process. The W powders were purchased from the Buffalo Tungsten Company and used as raw W powders in this investigation. The purity level of this raw nano-tungsten powder is over 99.95%. The impurity composition of this raw nano-tungsten powder was disclosed by the Buffalo Tungsten Company and is shown in Table 3. TiH2 powder was added into W to work as a grain-growth inhibitor. TiH2 is unstable and will dissolve into Ti and H2 at around 400 °C. Different amounts of TiH2 were added into W to study the effect of TiH2 addition on the properties of the as-sintered W. In this investigation, three different compositions were studied: pure W, tungsten with a point five weight percent of TiH2 addition (W-0.5%TiH2), and tungsten with a one weight percent of TiH2 addition (W-1%TiH2). All samples underwent the same preparation processes: milling, drying, 28 compacting, and sintering. And after the sintering process, Archimedes' principle and the linear intercept method were used to measure the density and average grain size of each as-sintered sample, respectively. After measurement, a 3-PB test was applied to all assintered samples to measure the mechanical properties and DBTT. 4.2 Ball Milling Process Many methods can be used to prepare insoluble nano-tungsten powders, like MA, spray conversion, chemical vapor deposition, and so on. The most common and simple way to prepare nano-tungsten powder is MA, not only because W powder is very brittle and easy to mill, but also because of the low production cost of this method. In this investigation, a custom-designed, unique, HEPM machine was used to mill raw W powder. W powder was carried out in a stainless steel canister filled with mixed heptane and high-purity alcohol. Tungsten carbide balls at one mm in diameter were used as the mill balls during the HEPM process, with a ball-powder ratio of approximately 5:1. The HEPM machine can mill two bottles of W powder at the same time, and 500 g tungsten carbide balls and 100 g W powder were placed in the bottom of the stainless steel canister. The mixed liquid of heptane and high-purity alcohol was used as a milling medium between balls and powders, and it was used to protect the oxidation of W powder during the lengthy ultrahigh-energy ball milling process. The added alcohol was also used to slightly reduce the aggregation of after-milled nano-tungsten powder. In this investigation, all powders were used in a custom-designed, HEPM machine to mill for 6 hours. Figure 4 is a photograph of the custom-designed, HEPM machine used in this investigation. The centrifugal force of this HEPM machine is about 100 gravitational 29 acceleration (g). This custom-designed, HEPM machine made the milling process more efficient, which also made sintering nearly full-density W sample at relative low sintering temperature become possible. 4.3 Drying Process After 6 hours HEPM process, the tungsten carbide balls and powders were taken out of the milling can together, and an 180 μm sifter was used to separate the milling balls and powders. After sieving, the milling balls were collected for cyclic utilization and the after-milled W powder mud was put into a clean breaker or evaporating dish for air-drying. For safety considerations, the after-milled W powder mud was always put into a hood in order to prevent spontaneous combustion of W powder. Usually, it needed 2 or 3 days for drying the after-milled W powder. 4.4 Forming and Compacting Process Green density is a critical factor for evaluating the maximum after-sintered density. The green density of compacted nano-tungsten bulk is around 34%, which means approximately 60% of shrinkage will occur during the sintering process. Lower green density in a compacted sample can greatly increase the risk of cracking during sintering due to the large amount of shrinkage during the sintering process. Before the final compacting process, some preprocesses were used to improve the green density of compacted W bulk. A lot of internal stress was introduced into the compacted W piece due to the effect of friction between each particle during compaction. To overcome this problem, a 30 process named "pelletization" was used in this investigation before the final compact step to improve green density and release the internal stress of the compacted W bulk. Pelletization is a repeat process of compression and splintering. One effective way to keep the integrity of the as-sintered W sample is to improve the green density of the compacted W sample before sintering. The value of the green density of the compacted sample is actually the weight of the compacted sample directly divided by the volume of the sample. During a series of experiments on the nano-tungsten powder, it was found that the process of pelletization can really improve the green density of a compacted W sample and can also reduce the risk of the sample fragmenting after sintering. A normal axial die molding was used for the compression process. The die used for the pelletization process in this investigation had a circular shape with a 35 mm diameter. Compression processes were applied three times in each pelletization process, and compression pressures were gradually increased. Each time after the compression process, the coffee grinder was used to break the compacted tungsten bulk into powder. After it was broken into powders, the step was repeated over and over again until the final step of compression was finished. After the pelletization process, the tungsten powders were compacted into green parts using uniaxial compression with 54 MPa pressure. Two differently dies shaped were used in this investigation due to the different research purposes. One was a circular die with a 16.25 mm diameter, and this die was used for making small, round pellets of W samples to study the effect of different sintering parameters on the grain size and density of the as-sintered W samples as well as the effect of TiH2 addition on the properties of the as-sintered W. The other die was an 8 31 mm x 72 mm rectangle with round ends and a radius on both ends of 4 mm. This die was used for making the 3-PB samples, which were used for studying bend tests and measuring mechanical properties and DBTT of the as-sintered W samples. 4.5 Sintering Process W usually is very stable at room temperature and will not easily react with oxygen or water, but after the HEPM process, the W powder was milled into nanoscale. Part of W powder will react with oxygen and become tungsten oxide during the milling and airdrying processes. In this situation, the reduction process becomes a very necessary step of W sintering. In this investigation, the whole sintering process can be divided into roughly three sections: reduction, presintering, and final sintering. For the nano-tungsten powder sintering, the reduction process is very important. During reduction, hydrogen was flowing all the time to reduce the tungsten oxide going back to W. The final oxygen content might have a big effect on the mechanical properties of after-sintered tungsten samples, so reduction can not only help the powder compact to decrease oxygen content but can also create fresh and active surfaces that improve the surface diffusion. Tungsten oxide will react with hydrogen, producing pure W and water vapor. The vapor is then brined out of the furnace by the continuous flow of hydrogen gas. The chemical equation for reduction can be simply written as: 𝑊𝑂2 + 2𝐻2 = 𝑊 + 2𝐻2 𝑂. During the presintering part, 1050 °C was used as the temperature during the whole sintering process. Argon (Ar) gas was flowing all the time during the presintering process. The purpose of presintering is to let the W sample keep a small grain size and 32 become more densified before the final sintering process. For the final sintering, due to the benefits of HEPM milled nano-scale W powder, the W sample can obtain great densification at a relatively low sintering temperature. In this investigation, a range temperature between 1200 °C to 1400 °C was chosen as the final sintering temperature in order to study the effects of sintering temperature on the properties of as-sintered tungsten. During the final sintering process, Ar gas was chosen to flow at all times until it is cooled down to room temperature. An Ar atmosphere was used in both the presintering and final sintering process because it was found that W samples sintered in an Ar atmosphere after the reduction part can achieve higher density than the samples sintered just in H2 atmosphere. The main cause of the difference in density of as-sintered W is the different diffusion mechanism. It was found that the dominant mechanism for densification in the sintering of nano-W powder with Ti in an Ar atmosphere is a mixture of volume diffusion and grain-boundary diffusion. For an H2 atmosphere at low temperatures, it is a vapor transport; as the temperature increases, it shifts to volume diffusion. Chai [56] shows those diffusion mechanisms in detail in his paper; sintering nano-tungsten in an Ar atmosphere will show higher after-sintered density. In this investigation, all as-sintered samples were sintered in the H2 atmosphere before reduction and then the atmosphere was switched from H2 to an Ar after the reduction part. The whole sintering program used in this investigation is shown in Figure 3, and the heating rate was 5 °C/min during the whole sintering process. 33 4.6 Characterization Methods 4.6.1 Scanning Electron Microscopy FEI NovaNano 630 scanning electron microscopy (SEM) was used in this investigation. The morphology and grain size of each as-sintered W sample was used (SEM) for imaging. SEM is mainly used for measuring the average grain size of the assintered W samples and also for observing the fracture surface of 3-PB samples. 4.6.2 Brunauer-Emmett-Teller Specific Surface Area Analysis Brunauer-Emmett-Teller (BET) is a useful technique to measure the specific surface area of powders. In this work, BET was mainly used to measure the specific surface area of three different compositions powder before and after HEPM in order to study and understand this custom-designed, HEPM process. 4.6.3 Particle Size Analysis Particle size analysis is a technique used for measuring particle size distribution and average particle size of solid or liquid particulate material. In this investigation, particle size analysis was used to measure the mean particle size and particle size distribution of W powder before and after 6 hours HEPM. 4.6.4 Vickers Hardness Tester LECO's Vickers Hardness Tester LV700AT was used in this investigation to measure the hardness of the as-sintered W samples, and 30 kg load was used to measure the Vickers hardness of each as-sintered tungsten alloy. 34 4.6.5 X-ray Diffraction X-ray diffraction (XRD) is a rapid analytical technique used for phase identification and grain-size measurement of powders. In this investigation, XRD was used to measure the crystallite size of the W powder before and after HEPM milling. 4.7 Sample Characterization XRD, BET surface area analysis, particle size analysis, and SEM were used to measure crystallite size, specific surface areas, and average particle size before and after 6 hours HEPM. The crystallite size of tungsten powder was measured by the XRD linebroadening technique. After sintering, tungsten specimens were first tested using Archimedes' principle to measure the density and then Vickers Hardness Tester was used to measure the Vickers hardness. As for the grain-size measurement, 500-grit SiC paper was first used on the sample for rough polish. Then the polishing machine was used with the help of added diamond-polishing suspension from 9 μm to 0.1 μm. After surface polishing at 0.1 μm, the polished surface of as-sintered sample was cleaned by ethanol to prepare for the subsequent SEM imaging. 4.7.1 Powder Analysis The purity level, grain size, and particle size of raw W powder can have a significant influence on the final mechanical properties of materials. Therefore, the raw W powder plays a crucial role on the properties of the final as-sintered W samples. In this investigation, all nano-tungsten powders were bought from the Buffalo Tungsten Company and were used as the raw metal powder. Different amounts of TiH2 powder are 35 added into W and it is worked as a grain-growth inhibitor during the sintering process. Different amounts of TiH2 powder were mixed with pure W powder together for 6 hours HEPM to make different types of W-Ti alloy powder: W, W-0.5% TiH2, and W-1% TiH2. The BET specific surface area and the mean particle size of W powder before and after 6 hours of HEPM were measured by BET and particle size analysis, respectively. Table 4 shows the BET specific surface area and average particle size of as-received pure W powder, after HEPM pure tungsten powder, after HEPM W-0.5% TiH2 powder, and after HEPM W-1% TiH2 powder. Figure 5 shows the particle size distribution of pure W powder before and after 6 hours of HEPM measured by particle size analysis. It can be known from the results of BET and particle size analysis that the custom-designed, HEPM machine can greatly reduce particle size and improve the specific surface area of W powder. The particle size of W powder reduced from 32.5 μm to 0.496 μm after 6 hours of HEPM, and it is very obvious that the particle size distribution of W powder becomes more uniform after the HEPM process. 4.7.2 Density Measurement Density is a very important parameter for the as-sintered materials. Archimedes' principle was used to measure density of all as-sintered W samples, and then the relative theoretical density can be calculated. The relative theory density was calculated by dividing the measured density by the theoretical density. 36 4.7.3 Grain-Size Measurement XRD was used to measure the crystalline size of the pure W powders before and after 6 hours of HEPM. The grain size of powder can be calculated using the value of the XRD peak width. The average crystalline size of nano-W powder after 6 hours of HEPM is 16 nm based on XRD peak profiles using the Williamson-Hall plot method [57]. The linear intercept method was used to measure the average grain size of each as-sintered W sample. Each as-sintered W sample was polished in 0.1-μm surface conditions in order to get more clear SEM images for the grain size measurement. Software named Nano Measurer was used to measure the grain size through SEM images of each as-sintered W sample. The software is easy to use; open the SEM image, the scale bar length is input by dragging its actual distance, and then lines are drawn through W grains, usually five lines on each image. Once the lines are drawn, the program records them and converts them to actual distances. For more accurate numbers in this investigation, five different SEM images were taken from each as-sintered W sample for the grain size measurement. 4.8 Mechanical Test Introduction 4.8.1 A Brief Introduction of the Three-Point Bend Test The main purpose of this investigation was to measure the mechanical properties, especially the DBTT, of each as-sintered W sample. There are many test methods that can be used for measuring the DBTT of W, such as tension test, compression test, 3-PB test, four-point bend test, and Charpy impact test. In this investigation, the 3-PB test was finally chosen to measure the mechanical properties and DBTT of each as-sintered 37 sample. The reason to choose the 3-PB test to measure the DBTT is due to fact that the 3PB test is the least expensive and simplest test method for measuring DBTT compared with other test techniques. The 3-PB test can also be used for small-specimen geometries [58], which is suitable for experimental study. A sketch of the 3-PB test is shown in Figure 6. It has two points on the bottom for support and one point on the top for loading force. In this investigation, the 3-PB test was performed on the MTS hydraulic system with a buildup high-temperature furnace, which can provide high-temperature test conditions. The temperature range of this furnace is from room temperature to 1200 °C. The 3-PB test was used on the as-sintered specimens of W, W-0.5% TiH2, W-1% TiH2, and commercially purchased hot-rolled W (bought from the Plansee Company) to record the relationship between load and displacement. The relationship between grain size and mechanical properties and the effect of TiH2 additive on the mechanical properties of the as-sintered W were also studied. Commercially purchased hot-rolled W sample was not used for comparison with the as-sintered W sample due to the fact that they have totally different manufacture history. For brittle materials like tungsten, it has a linear stressstrain relationship; the flexural stress and flexural strain can be calculated by these equations [26]: 3𝐹𝐿 Stress: σ= 2𝑏𝑑2 (Equation 1) Strain: є= 6𝐷𝑑 𝐿2 (Equation 2) where σ (MPa) is the stress, є (%) is the strain, b (mm) is the sample width, d (mm) is the sample thickness, D is the sample deflection, F (KN) is the maximum load, and L (mm) is the support span [26]. 38 4.8.2 MTS System The MTS machine system is composed of a load unit, load cell, hydraulic system, high-temperature furnace, MTS temperature controller, cooling system, and gas-flowing system. Figure 7 shows a photograph of the 3-PB test on the MTS system. The load unit is the primary structure for most materials testing. The high-temperature furnace can be tested at temperatures up to 1200 °C. The MTS series 653 high-temperature furnace provides a high-temperature test environment for the 3-PB test; the furnace is installed in the middle of the MTS load unit and the MTS temperature controller is used to control the temperature of the furnace. The furnace has three hot zones produced by two horizontally oriented silicon carbide heating elements; each vertical side of the furnace has three heating elements. The high-temperature furnace comes with three K-type thermocouples that correspond to the top zone, middle zone, and bottom zone of the furnace. An Ar tank was prepared and connected to three stainless steel tubes to provide an Ar atmosphere if needed to prevent tested specimens from oxidizing at hightemperature testing environment. A high-temperature axial extensometer was used to obtain an accurate number for the tested sample's deflection. The measurement limitation of the extensometer is ± 1 mm. It can give a more accurate displacement number than the linear variable displacement transducer. Axial extensometers are used for measuring strain and elastic modulus in high-temperature compression tests at temperatures up to 650 °C. The extensometer, where the end of the extension rod is contacting the specimen, is held in position by a spring-loaded hold-down assembly. The hold-down assembly secures the extensometer while allowing full movement of the rods. The extension rods use ceramic 39 extension rods that are 3.5 mm in diameter. The pressure head is made by SiC, and this material can stand extremely high-temperature environments without damage. 4.8.3 Bend Fixture A custom-designed bend fixture was used in this investigation for the 3-PB test. This fixture has three pieces: the V-shaped compressive head, the upper fixture, and the bottom fixture. For this custom-designed, fixed 3-PB fixture, the span is 28 mm. 4.8.4 Three-Point Bend Samples Introduction W materials prepared by different production methods were investigated. The first type was the 3 mm-thick commercially purchased hot-rolled W bought from Plansee; the purity of this hot-rolled W was above 99.97%. Table 5 shows the chemical composition of hot-rolled W based on the information given from the Plansee Company. The second type included the as-sintered W, W-0.5% TiH2, and W-1% TiH2 specimens prepared by the relatively low-temperature sintering programs discussed earlier. For the as-sintered bend samples, a rectangular die was used for forming and compacting. According to ASTM C1211-02 Standard Test Method for Flexural Strength of Advanced Ceramics at Elevated Temperatures, the 3-PB test sample dimensions are about 3 mm × 4 mm × 46 mm, where 3 mm is the thickness, 4 mm is the width, and 46 mm is the length. Actually, the length of the sample does not affect the final results of flexural stress and flexural strain measurement according to Equations 1 and 2. Based on previous studies on the effects of severe sintering parameters on the density and grain size for three different compositions, W, W-0.5% TiH2, and W-1% 40 TiH2, and also after considering the balance between good density and fine grain size, we finally used reduction temperature at 750 °C and sintering temperature at 1300 °C as the sintering program for all as-sintered 3-PB samples. All bend samples underwent the same sintering program. After sintering, Archimedes' principle was used to measure each sample's density, and then a grinder machine was used to roughly remove excess material and make the sample into the size for testing. For the commercially purchased hot-rolled W sheet, electrical discharge machining (EDM) was used to cut the sample into the right dimensions for the 3-PB test. After the W sample was ground and underwent EDM, all bend samples were polished on the polishing machine for getting more refined the surface condition. Three sides of each bend sample were polished into 1 μm with the help of a polishing machine and diamond suspension. During the polishing process, due to the fact that the W bend sample was too long (46 mm), hot mounting wax was used to mount the bend samples on the polishing holder. Hot mounting wax provides a quick and strong bond between samples and holder. Mounting wax melts at 120 °C and can be dissolved in acetone. The polishing process is considered to increase the room-temperature bend ductility of commercially pure sintered and swaged tungsten rods. The beneficial effects of polishing include the removal of surface notches and cracks resulting from processing or specimen fabrication. As reported by Stephens, just removing around 12.5 μm of surface layer was found to increase the maximum bending angle from 17 ° to 30 ° [26,55]. In order to study the effect of surface roughness on the flexural properties of commercially purchased hotrolled W, two different surface conditions were used in this investigation. One was the as-received EDM surface finish, and the other one was the polished 1 μm surface 41 condition. An investigation into the effect of surface roughness was not performed on the as-sintered tungsten due to the difficult preparation process for making the as-sintered W samples. Therefore, all of the as-sintered bend samples in this investigation were all tested with the 1 μm surface finish condition. Figure 8 shows a photograph of a commercially purchased hot-rolled W that has undergone EDM and an as-sintered W bend sample. 4.8.5 Three-Point Bend Test Setup Commercially purchased hot-rolled W samples were tested with two different surface finish conditions to study the effect of surface condition on the mechanical properties of hot-rolled W. Before the tests, for both as-sintered W and hot-rolled W, lubricant was put on the fixture to decrease the friction between samples and the fixture during the test process. For the polished sample, the unpolished side of the sample was placed facing to the compression head, and the test sample was put in the middle of the fixture. The fixture was placed in the middle of the high-temperature furnace to ensure uniform heating of the test sample. For the high-temperature test, after the furnace reached the setting temperature, we waited for 25 to 35 minutes before running the 3-PB test to ensure that the test sample reached the setting temperature. When the test temperature was higher than 400 °C, an Ar atmosphere was provided in the furnace to make sure the specimen would not be oxidized. After all steps were set up correctly, the 3-PB test was started. After the test, the strain-stress curve, yield strength, and fracture strength were calculated through the 3-PB test. 42 Figure 3. The sintering program used in this investigation 43 Table 3. Impurity composition in raw tungsten powder Element % Element % Al <0.001 Na <0.001 Ca <0.001 Ni <0.001 Co <0.001 P <0.001 Cr <0.001 Pb <0.001 Cu <0.001 Sb <0.001 Fe <0.001 Si <0.001 K <0.001 Sn <0.001 Mg <0.001 Ta <0.001 Mn <0.001 Ti <0.001 Mo 0.001 V <0.001 Zn <0.001 Table 4. Specific surface area and average particle size of powders before and after 6 hours of ball milling Specific Surface Area W (before milling) Average Particle Size 6.0240 (m2/g) 0.3566 32.545 (μm) 45 W (after milling) 9.4518 0.496 W-0.5%TiH2 9.3824 - W-1%TiH2 9.5294 - TiH2 44 Figure 4. The custom-designed, ultrahigh-energy planetary ball milling (HEPM) machine. 45 Figure 5. The particle size distributions of pure tungsten powder before and after 6 hours of ultrahigh-energy planetary ball milling. (The upper figure represents the particle size of pure tungsten powder before ball milling, and the lower figure represents the particle size distribution of pure tungsten powder after 6 hours of ultrahigh energy ball-milling.) 46 Table 5. The chemical composition of commercially purchased hot-rolled W, which is given by the Plansee Company. Al ≤ 15 μg/g Cr ≤ 20 μg/g Cu ≤ 10 μg/g Fe ≤ 30 μg/g K ≤ 10 μg/g Ni ≤ 20 μg/g Si ≤ 20 μg/g Mo ≤ 100 μg/g C ≤ 30 μg/g H ≤ 5 μg/g N ≤ 5 μg/g O ≤ 20 μg/g Cd ≤ 5 μg/g Hg ≤ 1 μg/g Pb ≤ 5 μg/g 47 Figure 6. Schematic diagram of the three-point bend test. 48 Figure 7. The test setup of the high-temperature three-point bend test. 49 Figure 8. The as-received EDM commercially purchased hot-rolled W sample and assintered pure W sample using a reduction temperature of 750 °C and a sintering temperature of 1300 °C. CHAPTER 5 EXPERIMENTAL RESULTS In this chapter, the author reports the effects of compacted pressure and different sintering conditions on the properties of the as-sintered W, W-0.5% TiH2, and W-1% TiH2. Factors that affect especially the density, grain size, and hardness of as-sintered samples were studied. The author also reports effects that might influence the mechanical properties of W samples of two different types: the as-sintered W and commercially purchased hot-rolled W. Archimedes' principle was used to measure the density of all assintered samples. The linear intercept method and Vickers Hardness Tester were used to measure the average grain size and Vickers hardness, respectively. Reduction temperature and sintering temperature played very important roles in sintering samples. To set sintering parameters on the properties of as-sintered W, W-0.5% TiH2 and W-1% TiH2, three different reduction temperatures and sintering temperatures were studied. In this investigation were used reduction temperatures of 650 °C, 750°C, and 850°C and sintering temperatures of 1200 °C, 1300°C, and 1400 °C. 5.1 The Effect of Final Compact Pressure on Density All as-sintered W samples were prepared by the same sintering program: a reduction temperature of 750 °C, a sintering temperature of 1300 °C, in H2 atmosphere 51 and switched to Ar atmosphere after the reduction process. The as-sintered W samples used same sintering program to study the effect of final compact pressure on the density of the as-sintered W sample. The relationship between the different final compact pressure and the relative theoretical density of as-sintered pure W is shown in Figure 9. As observed in Figure 9, it is very obvious that when final compact pressure at 146.3 MPa was used, the as-sintered W bulk sample showed the highest relative theoretical density. When using the final compact pressure at 97.5 MPa, the as-sintered W sample showed the lowest relative theoretical density. It is also shown in Figure 10 that when the final compact pressure continually increased from 146.3 MPa to 243.8 MPa, the relative theoretical density of the as-sintered W dropped a little bit, lower than the relative theoretical density with a compact pressure of 146.3 MPa. However, the relative theoretical density of the as-sintered W sample with the final compact pressure at 243.8 MPa was still higher than the as-sintered W sample with final compact pressure at 97.5 MPa. Based on this experimental result and in order to get the best relative theoretical density of the as-sintered W sample, all sintered samples used 146.3 MPa as their final compact pressure in this investigation. 5.2 Effect of Reduction Temperature on Density In order to study the effect of different reduction temperatures on the density of as-sintered W, W-0.5% TiH2, and W-1% TiH2 samples, All samples were prepared with the same sintering temperature at 1300 °C and three different reduction temperatures, 650 °C, 750 °C, and 850 °C, were used. The relationship between different reduction 52 temperatures and relative theoretical density of three different as-sintered W compositions is shown in Figure 10. All compositions show similar tendencies in Figure 10, which is a parabola going downward. With continually increasing the reduction temperature from 650 °C to 850 °C, the relative theoretical density of all as-sintered samples was first increased and then reached their highest relative theoretical density at reduction temperature of 750 °C. Then the density was decreased at a reduction temperature of 850 °C. In general, the assintered pure W samples with three different reduction temperatures showed the highest relative theoretical density across all compositions, and the as-sintered W-1%TiH2 samples had the lowest relative theoretical density across all compositions. The assintered pure W sample at reduction temperature of 750 °C showed the highest relative density, and the as-sintered W-1% TiH2 sample at reduction temperature of 650 °C showed the lowest relative density among all as-sintered samples. The as-sintered samples of W, W-0.5% TiH2, and W-1% TiH2 showed similar relative theoretical density (around 94.5%) at a reduction temperature of 650 °C, and as-sintered pure W had the highest density (around 98%) at reduction temperature of 750 °C. At reduction temperature of 750 °C, the relative theoretical density differences among the as-sintered W, W-0.5% TiH2, and W-1% TiH2 samples were approximately 1%. The as-sintered pure W sample always showed the highest relative theoretical density among the three compositions, and the as-sintered W-1% TiH2 sample showed the lowest relative theoretical density. Under the same conditions, the density of the assintered W samples decreased when the TiH2 addition contents are increased. These experiment results show that the reduction temperature at 750 °C is the best 53 reduction temperature for the as-sintered W samples to reach their highest relative density. This result also showed that TiH2 addition has a negative effect on the W densification. 5.3 Effect of Reduction Temperature on Grain Size All as-sintered samples in this section were prepared by the same sintering temperature at 1300 °C, and three different reduction temperatures, 650 °C, 750 °C, and 850 °C, were used to study the effect of different reduction temperatures on the average grain size of as-sintered W, W-0.5% TiH2, and W-1% TiH2 samples. The relationship between reduction temperature and average grain size of three different compositions is shown in Figure 11. The as-sintered W, W-0.5% TiH2, and W-1% TiH2 samples showed similar tendencies in Figure 11, which is a parabola that points up. The grain sizes of all compositions are decreased first, and reached their smallest average grain size at the reduction temperature of 750 °C. When higher reduction temperature 850 °C was used, the grain size tended to increase. Among all compositions, pure W samples at reduction temperature of 850 °C showed the biggest grain size, and W-1% TiH2 samples at reduction temperature of 750 °C showed the smallest grain size. The as-sintered W, W0.5% TiH2, and W-1% TiH2 samples all showed their smallest grain size at the reduction temperature of 750 °C. The grain size difference between pure W samples and the W-0.5% TiH2 sample is much larger than the grain size difference between the W-0.5% TiH2 and W-1% TiH2 sample. Even doubling the addition amount of TiH2 from 0.5% to 1%, the grain size difference between W-0.5% TiH2 and W-1% TiH2 was not very big; the grain size of the 54 W-0.5% TiH2 sample was just a litter higher than the W-1% TiH2 sample when using reduction temperatures at 650 °C and 850 °C, and they also showed almost the same average grain size at reduction temperature of 750 °C. However, there is a big gap in grain size between the pure W samples and W-0.5% TiH2 samples; the grain size of the pure W sample was more than twice larger than the W-0.5% TiH2 sample when the same reduction condition was used. This result shows that the TiH2 addition is very effective in refining the grain size of the as-sintered W; just a very small amount of TiH2 addition can greatly decrease the average grain size of the as-sintered W. These experiment results show that the TiH2 addition has a great effect on decreasing the grain size of the as-sintered W and that W grain size is very sensitive to TiH2 contents. 5.4 The Effect of Sintering Temperature on Density All samples in this section were prepared using the same reduction program at 750 °C and also the same sintering atmosphere. Three different sintering temperatures, 1200 °C, 1300 °C, and 1400 °C, were used to study the effect of sintering temperature on the density of as-sintered W, W-0.5% TiH2, and W-1% TiH2. The relationship between sintering temperature and relative theoretical density in three different compositions, W, W-0.5% TiH2, and W-1% TiH2, is shown in Figure 12. It is very obvious that the three compositions all have similar tendencies; for all of them, relative theoretical density is increased when the sintering temperature increased. Pure W samples showed the highest relative theoretical density (around 99.5%) at sintering temperature of 1400 °C, and the W-1% TiH2 sample showed the lowest relative theoretical density (around 94%) at 55 sintering temperature of 1200 °C. Pure W samples showed the highest theoretical density across all compositions. When the same sintering temperature conditions were used, the relative theoretical density of W samples is decreased with increased amount of TiH2, verifying again that the TiH2 addition has a negative effect on the densification of assintered W. This experimental result shows that the density of the as-sintered W is increased with increased sintering temperature. 5.5 The Effect of Sintering Temperature on Grain Size All samples in this section were prepared using the same reduction temperature at 750 °C, with three different sintering temperatures ranging from 1200 °C to 1400 °C. The relationship between sintering temperature and average grain size of the three different compositions is shown in Figure 13. The grain size of those three compositions showed the same tendency: the grain size is increased with increased sintering temperature. Among all samples, pure W sample showed the largest grain size at the sintering temperature of 1400 °C, and the W-1% TiH2 sample showed the smallest grain size at sintering temperature of 1200 °C. When the same sintering temperature was used, the grain size of the pure W sample was almost twice larger than the grain size of the W0.5% TiH2 sample. The grain size difference between W-0.5% TiH2 samples and W-1% TiH2 samples was very small, especially at 1200 °C and 1300 °C. At 1400 °C, the grain size difference became a bit larger. It seems that the additive TiH2 works better with higher content under high-temperature conditions. And 1% TiH2 addition has a better grain size- 56 suppression effect than 0.5% TiH2 addition at high sintering temperature. The TiH2 additions have a great effect on refining the grain size of as-sintered W. 5.6 Vickers Hardness All sintered samples were prepared by the same sintering program, with a reduction temperature of 750 °C and sintering temperature of 1300 °C. A 30 kg load was used to measure the Vickers hardness of as-sintered W, W-0.5% TiH2, and W-1% TiH2 samples. Figure 14 shows the Vickers hardness of the as-sintered W, W-0.5% TiH2, and W-1% TiH2 samples that used the same sintering program, with a reduction temperature of 750 °C and sintering temperature of 1300 °C. It is very clear that the as-sintered W-1% TiH2 sample shows the highest Vickers hardness (around 750 HV), and the as-sintered pure W sample shows the lowest Vickers hardness (about 450 HV). The Vickers hardness of the as-sintered W-0.5% TiH2 sample is around 625 HV. It is very obvious that the Vickers hardness of as-sintered W is increased with increased amount of TiH2 addition. This experiment result shows that the TiH2 addition has a very good effect on improving the Vickers hardness of as-sintered W. 5.7 The Relationship Between Grain Size and Relative Density Across Three Different Compositions All samples were prepared using the same sintering temperature at 1300 °C but with three different reduction temperatures from 650 °C to 850 °C. Figure 15 shows the relationship between the average grain size and relative theoretical density in W, W-0.5% TiH2, and W-1% TiH2 samples at three different reduction temperatures. All 57 compositions at different reduction temperatures showed similar tendencies; both the average grain size and relative theoretical density were decreased with increased the amounts of TiH2. At a reduction temperature of 650 °C, the density decrease was not very obvious when compared to the density decrease at the reduction temperature between 750 °C and 850 °C. However, the effect of TiH2 addition on decreasing the average grain size of as-sintered is valid; TiH2 addition does a great job in refining grain size of W during the sintering process, but it is also has a negative effect on densification. 5.8 The Relationship Between Grain Size and Vickers Hardness Across Three Different Compositions All as-sintered samples in this section were prepared by the same sintering program, with reduction temperature of 750 °C and sintering temperature of 1300 °C. The relationship between the average grain size and Vickers hardness in three different W compositions is shown in Figure 16. As Figure 16 indicates, pure W sample has the largest grain size and also the lowest Vickers hardness. W-1% TiH2 sample has the lowest average grain size and the highest Vickers hardness, and the Vickers hardness of as-sintered W is increased when the amount of TiH2 addition increases. The Vickers hardness is also increased when the average grain size decreased. 5.9 SEM and EBSD Images Figure 17 and Figure 18 feature SEM and electron backscatter diffraction (EBSD) images of the as-sintered W-0.5% TiH2 sample with a reduction temperature of 850 °C and sintering temperature of 1300 °C. 58 Both the grain size and grain shape of the W-0.5% TiH2 sample is shown clearly in the SEM and EBSD images. EBSD is a technique that can be used to measure grain size; here it was used to measure the grain size of as-sintered W-0.5% TiH2 and to check if the grain size result calculated from the SEM image could be matched with the EBSD result. The EBSD image shows that the average grain size of a W-0.5% TiH2 sample with a reduction temperature of 850 °C and a sintering temperature of 1300 °C is around 1 μm. This is a similar result to the calculated result from the SEM intercept method, which proved that the grain size result calculated from the SEM image is credible. The reason why SEM images are used to calculate grain size is their low experimental cost compare with the cost of EBSD. SEM images of as-sintered W, W-0.5% TiH2, and W-1% TiH2 samples are shown in Figure 19. All as-sintered samples were prepared using the same sintering program, with a reduction temperature of 750 °C and sintering temperature of 1300 °C. The SEM image of each as-sintered W, W-0.5% TiH2, and W-1% TiH2 sample shows that the grain size decreases dramatically with increased TiH2 content. 5.10 Effect of Surface Condition on Commercially Purchased Hot-Rolled W For the commercially purchased hot-rolled W sample, two different surface conditions were used in this investigation to study the effect of surface finish on the mechanical properties of commercially purchased hot-rolled W. The two different surface finishes were a 1 μm surface finish and the as-received EDM (unpolished) hot-rolled surface condition. Through a high-temperature 3-PB test on the MTS system, software 59 programs named Station Manager and Multipurpose Elite were used to record the 3-PB load-displacement curve and stress-strain curves. For both polished and unpolished commercially purchased hot-rolled W, test temperatures ranging from room temperature to 300 °C were used. Figure 20 shows the 3-PB bend stress-strain curve of commercially purchased hot-rolled W with as-received EDM surface (unpolished) at test temperatures ranging from room temperature to 300 °C with a test strain rate of 0.3 mm/min. Figure 21 shows the 3-PB stress-strain curve of a commercially purchased hot-rolled W sample with 1 μm surface finishing at test temperatures ranging from room temperature to 300 °C and with a test strain rate of 0.3 mm/min. In order to measure DBTT based on the 3-PB stress-strain curve at different test temperatures, we defined the DBTT in this experiment as the temperature at which a sample undergoes a minimum of 5.0% flexural strain without failure. Therefore, the DBTT of commercially purchased hot-rolled W could be evaluated by the strain-test temperature curve. The strain-test temperature curves of both polished and unpolished commercially purchased hot-rolled W samples with a test strain rate of 0.3 mm/min were put together and shown in Figure 22. Through this strain-test temperature curve, it could be evaluated that the DBTT of unpolished commercially purchased hot-rolled W at a test strain rate of 0.3 mm/min is around 150 °C to 175 °C, and the DBTT of 1 μm surfacepolishing commercially purchased hot-rolled W at a test strain rate at 0.3 mm/min is around 100 °C to 150 °C. It is clear that for commercially purchased hot-rolled W with a 1 μm surface condition, the DBTT was lower than the unpolished sample when the same test strain at 0.3 mm/min was used. It is proven that surface conditions have a great effect on the 60 DBTT of hot-rolled W. For commercially purchased hot-rolled W, the DBTT could be reduced with better surface finish conditions. 5.11 Effect of Strain Rate on Commercially Purchased Hot-Rolled W For commercially purchased hot-rolled W, two different test strain rates were used in this investigation in order to study the effect of test strain rate on mechanical properties, especially the DBTT. One test strain rate used was 0.3 mm/min; another was 0.1 mm/min. All commercially purchased hot-rolled W samples used in this section had surface finishes that had undergone EDM to exclude the effect of surface condition on mechanical properties. Figure 23 shows the 3-PB stress-strain curve of unpolished commercially purchased hot-rolled W bend samples (as-received EDM surface condition) at test temperatures ranging from room temperature to 300 °C and with a test strain rate of 0.1 mm/min. Figure 24 compares 3-PB strain-temperature curves at two test strain rates: 0.1 mm/min and 0.3 mm/min. These curves were based on unpolished commercially purchased hot-rolled W samples (as-received EDM surface condition). By definition, DBTT is the temperature at which a W sample undergoes minimum 5% strain without failure. The DBTT of an unpolished commercially purchased hot-rolled W sample (asreceived EDM surface condition) with a test strain rate of 0.1 mm/min was between 100 °C and 150 °C. The DBTT of an unpolished commercially purchased hot-rolled W sample (as-received EDM surface condition) with a test strain rate of 0.3 mm/min was between 150 °C and 175 °C. The results show that test strain rate does have an influence 61 on the DBTT of commercially purchased hot-rolled W. Tungsten is a strain-rate-sensitive material. In this test, when the test strain rate was lowered from 0.3 mm/min to 0.1 mm/min, the DBTT of commercially purchased hot-rolled W with a surface finish that has undergone EDM can decrease to around 50 °C. 5.12 Effect of TiH2 Addition on the Mechanical Properties of Sintered W Two different compositions, pure W and W-1% TiH2, were used in this investigation to study the effect of TiH2 addition on the mechanical properties of sintered W. The same sintering program was used for making bend samples of both sintered pure W and W-1% TiH2. All sintered samples had the same reduction temperature (750 °C) and sintering temperature (1300 °C), and they all had an H2 atmosphere that was switched to an Ar atmosphere after reduction. To decrease the effect of surface defects on mechanical properties as far as possible, all sintered bend samples were prepared in 1 μm surface conditions. Figure 25 features the 3-PB stress-strain curve of a sintered pure W sample with 1 μm surface condition, a test temperature ranging from room temperature to 600 °C, and with a test strain rate of 0.3 mm/min. It is clear that all sintered pure W bend samples broke at around 0.2% flexural strain, and even raising the test temperature to 600 °C, the sintered W sample did not show any ductile behaviors. Therefore, the DBTT of a sintered pure W sample with 1 μm surface condition and a test strain rate of 0.3 mm/min should be higher than 600 °C. Figure 26 shows the 3-PB stress-strain curve of a sintered W-1%TiH2 sample with 62 1 μm surface condition, a test temperature ranging from room temperature to 450 °C, and with a test strain rate of 0.3 mm/min. It shows clearly in Figure 26 that all sintered W-1% TiH2 bend samples broke at around 0.2% flexural strain, the same results as the sintered pure W bend sample. Sintered W-1% TiH2 bend samples did not show any ductile behavior, even with raising the test temperature to 450 °C. According to the stress-strain curve, the DBTT of sintered W-1% TiH2 samples with 1 μm surface conditions and with a test strain rate of 0.3 mm/min is higher than 450 °C. Among the sintered samples, both pure W and W-1% TiH2 bend samples did not show any ductile behavior during the 3-PB test. TiH2 addition to W didn't show good results in improving the mechanical properties of sintered W samples. 5.13 Fracture Surface Analysis After the 3-PB test, SEM was used to observe the fracture surface of each broken bend sample in detail. Figure 27 shows SEM images of the room temperature 3-PB fracture surfaces of a sintered W sample, sintered W-1% TiH2 sample, and commercially purchased hot-rolled W sample. Figure 28 shows SEM images of the 3-PB fracture surfaces of sintered W and W-1% TiH2 samples at a test temperature of 450 °C. It can be clearly observed in the SEM images that both sintered W and sintered W-1% TiH2 bend samples showed intergranular cleavage fractures at room temperature. However, in contrast with sintered W, commercially purchased hot-rolled W showed mainly transgranular fractures at room temperature. Sintered W and sintered W-1% TiH2 specimens, even at the high test temperature of 450 °C, still mainly showed intergranular 63 fractures. The fracture mode didn't change for sintered bend samples even when the test temperature was increased from room temperature to 450 °C. Compaction pressure vs. density 98.00% 97.80% 97.60% 97.40% Relative 97.20% theoretical 97.00% density 96.80% 96.60% 96.40% 96.20% 96.00% 97.532 146.298 243.829 Compaction pressure (MPa) Figure 9. The relationship between the different compaction forces and the as-sintered relative theoretical density of pure W, with a reduction temperature of 750 °C, a sintering temperature of 1300 °C, and with an H2 atmosphere that was switched to an Ar atmosphere after reduction. 64 Reducation temperature vs. Density 100.00% W 99.00% W-0.5%TiH2 98.00% W-1%TiH2 Relative Density 97.00% 96.00% 95.00% 94.00% 600 650 700 750 800 850 900 Reducation Temperature (°C ) Figure 10. The relationship between the reduction temperature and density of three different compositions: W, W-0.5% TiH2, and W-1% TiH2. All samples used the same sintering temperature, 1300 °C, and Archimedes' principle was used to measure the relative density of each sample. Reduction temperature vs. Grain size 6 W 5 W-0.5%TiH2 4 W-1%TiH2 Grain Size (μm) 3 2 1 0 600 700 800 900 Reducation Temperature (°C ) Figure 11. The relationship between the reduction temperature and average grain size of three different compositions: W, W-0.5% TiH2, and W-1% TiH2. All samples used the same sintering temperature, 1300 °C, and the linear intercept method was used to measure the grain size of each sample. 65 Sintering temperature vs. Density 100.00% W 99.00% W-0.5%TiH2 98.00% W-1%TiH2 Relative Density 97.00% 96.00% 95.00% 94.00% 1150 1200 1250 1300 1350 1400 1450 Sintering Temperature (°C ) Figure 12. The relationship between the sintering temperature and density of three different compositions: W, W-0.5% TiH2, and W-1% TiH2. All samples used the same reduction temperature of 750 °C, and Archimedes' principle was used to measure the relative density of each sample. Sintering temperature vs. Grain size 6 W 5 W-0.5%TiH2 4 W-1%TiH2 Grain Size (μm )3 2 1 0 1150 1200 1250 1300 1350 1400 1450 Sintering Temperature (°C ) Figure 13. The relationship between sintering temperature and average grain size of three different compositions: W, W-0.5% TiH2, and W-1% TiH2. All samples used the same reduction temperature of 750 °C, and the linear intercept method was used to measure the grain size of each sample. 66 Figure 14. The Vickers hardness of as-sintered W, W-0.5% TiH2, and W-1% TiH2 samples with a reduction temperature of 750 °C and sintering temperature of 1300 °C. Figure 15. The relationship between the average grain size and relative density of W, W0.5% TiH2, and W-1% TiH2 samples with the same sintering temperature, 1300 °C, but three different reduction temperatures ranging from 650 °C to 850 °C. 67 Figure 16. The relationship between the grain size and Vickers hardness of W, W-0.5% TiH2, and W-1% TiH2 with a reduction temperature of 750 °C and a sintering temperature of 1300 °C. 68 5 μm Figure 17. An SEM image of polished W-0.5% TiH2 with 0.1 surface finish prepared at a reduction temperature of 850 °C and sintering temperature of 1300 °C. Figure 18. An EBSD image of polished W-0.5% TiH2 sample with 0.1 surface finish at a reduction temperature of 850 °C and sintering temperature of 1300 °C. 69 (a). W (b). W-0.5% TiH2 (c). W-1% TiH2 Figure 19. SEM images of postsintered W, W-0.5% TiH2, and W-1% TiH2 with 0.1 μm surface finish using a sintering program with a reduction temperature of 750 °C, sintering temperature of 1300 °C, and a flowing H2 atmosphere that is then switched to an Ar atmosphere after reduction. 70 Figure 20. Three-point bend stress-strain curve of a commercially purchased hot-rolled W sample without polishing (as-received hot-rolled surface condition) at test temperatures ranging from room temperature to 300 °C with a strain rate of 0.3 mm/min. 71 Figure 21. Three-point bend stress-strain curve of a commercially purchased hot-rolled W sample with 1 μm surface finish at test temperatures ranging from room temperature to 300 °C with a strain rate of 0.3 mm/min. 72 Figure 22. The flexural strain-test temperature curves of both polished (1 μm surface finish) and unpolished (as-received hot-rolled surface condition) commercially purchased hot-rolled W samples with a test strain of 0.3 mm/min. 73 Figure 23. Three-point bend stress-strain curve of an unpolished commercially purchased hot-rolled W sample (as-received hot-rolled W surface condition) at test temperatures ranging from room temperature to 300 °C with a test strain rate of 0.1 mm/min. 74 Figure 24. Flexural strain vs. test temperature curves of unpolished (as-received hotrolled surface condition) commercially purchased hot-rolled W samples at test temperatures ranging from room temperature to 300 °C with a strain rate of 0.1 mm/min and 0.3 mm/min. 75 Figure 25. Three-point bend stress-strain curve of a sintered pure W sample with polish at 1 μm surface finish condition and at test temperatures ranging from room temperature to 600 °C with a test strain rate of 0.3 mm/min. 76 Figure 26. Three-point bend stress-strain curve of a sintered W-1%TiH2 sample with polish at 1 μm surface finish condition and at test temperatures ranging from room temperature to 450 °C with a test strain rate of 0.3 mm/min. 77 (b) W-1% TiH2, RT (a) W, RT (c) Hot rolled W Figure 27. SEM images of room-temperature fracture surfaces of postsintered pure W and W-1% TiH2, sintered samples that are using the same sintering program with a reduction temperature of 750 °C, a sintering temperature of 1300 °C, and an H2 atmosphere that is the switched to an Ar atmosphere after reduction. 78 (a) W, 450 °C (b) W-1% TiH2, 450 °C Figure 28. SEM images of postsintered pure W and W-1% TiH2 fracture surfaces at 450 °C. These postsintered samples use the same sintering program with a reduction temperature of 750 °C, a sintering temperature of 1300 °C, and an H2 atmosphere that then switches to an Ar atmosphere after reduction. CHAPTER 6 DISCUSSION 6.1 Analysis and Discussion There are five main methods that can improve the ductility and decrease the DBTT of W: 1. Add some elements 2. Improve the purity 3. Improve the surface finishing condition 4. Increase the deformation degree 5. Refine the grain size Re has been considered an effective alloying element to improve the ductility of W materials. Severe plastic deformation techniques, like ECAP, have been used to reduce the grain size of tungsten. Tungsten processed using ECAP shows very good data on ductility and a low value of DBTT. The purpose of this investigation was to study how TiH2 addition and refining grain size without mechanical working affect the mechanical properties of W. The purity level of W powders used in this research before being milled was over 99.95%, ensuring its quality and decreasing the impurity content. Tungsten carbon balls were also used to decrease the risk of introducing impurities during the lengthy ball milling process. With 80 the help of a HEPM machine, nanosize W powders were obtained, and a relatively low sintering program and additive TiH2 were used in this investigation to help achieve nearfull density with ultrafine grain as-sintered W. During the experimental process, effects of sintering parameters, additive TiH2 contents, and final compact pressure on the properties of as-sintered samples were studied. The 3-PB test at elevated temperature was used to measure the mechanical properties and DBTT of as-sintered W samples and hotrolled W samples. Commercially purchased hot-rolled W samples were not used for a contrasting experiment to as-sintered W, because they had different thermomechanical histories. Different surface finishes and strain rates were used in a 3-PB test to study their effects on the mechanical properties and DBTT of commercially purchased hot-rolled W samples. Different TiH2 addition contents were used to study the effect of TiH2 addition on the mechanical properties and DBTT of as-sintered W. After 6 hours of the ultrahigh-energy planetary ball milling process, nanotungsten powders-both crystallized size and particle size-were significantly reduced, and the particle size distribution also became more concentrated after milling. The specific surface area and the sinterability of after-milled tungsten powder are increased in comparison to nonmilled tungsten powder. A benefit of nano-scale tungsten powder is that it becomes possible to sinter a tungsten sample into near-full density. This happens when particle size and grain size are smaller; the intrinsic driving force and surface energy of the powder become more active, which makes the sintering process easier and makes it possible to sinter tungsten bulk into full density at relatively low temperatures like 1300 °C. The effect of final compact pressure on the density of as-sintered W samples was 81 studied. Based on the experimental results, when final compacted pressure of 146.3 MPa was used, it showed the highest as-sintered relative theoretical density for conventional- pressure, low-temperature sintered pure tungsten. Usually, with higher final compacted pressure, a compacted sample should show higher as-sintered density; however, for nanosized powder, when compacted pressure has already reached a certain extent, further increased compact pressure will not improve density any more. Instead it has a negative effect on improving as-sintered density when the compacted pressure is too high. When the final compacted pressure of 243.8 MPa was used, the density of as-sintered W was a little bit lower than the density from when a final compacted pressure of 146.3 MPa was used. This is possible because of nanosized tungsten powder; if the particle size is too small and if the compact pressure is increased to a certain extent, it will form a wispy crack inside and cause the density to drop a bit. When all compositions are sintered at the same sintering temperature of 1300 °C, with three different reduction temperatures from 650 °C to 850 °C, the relative theoretical density of sintered tungsten alloys shows a parabolic curve; the highest density is reached at a reduction temperature of 750 °C. Part of the nano-tungsten powder will be oxidized due to the high energy and strong force of friction during the milling process. The reduction process is a very important step. When a reduction temperature of 650 °C is used, W-1% TiH2 shows the lowest relative density. It may be that a reduction temperature of 650 °C is too low, and tungsten oxide cannot be completely reduced during the reduction process. To further understand this phenomenon, Leco oxygen analysis was used to measure the oxygen content of each as-sintered tungsten sample. Figure 29 shows the relationship between oxygen content and the different reduction 82 temperatures of as-sintered W, W-0.5% TiH2, and W-1% TiH2 samples. It clearly shows that for all as-sintered W, W-0.5% TiH2, and W-1% TiH2 samples, the oxygen content decreases with increased reduction temperatures. For all compositions, the oxygen content used with a reduction temperature of 650 °C was higher than other oxygen content with higher reduction temperatures. The higher oxygen content is also a signal that tungsten oxide didn't completely reduce during reduction. The residual oxide in tungsten impacts the densification of tungsten. Sintering of tungsten is very sensitive to oxygen content, as a small amount of oxygen contamination can increase the activation energy for surface diffusion from 300 KJ/mol to nearly 400 KJ/mol [57,59]. The existing oxide, then, is the reason why the density of sintered W at a reduction temperature of 650 °C is lower than the density at a reduction temperature of 750 °C. The experimental result also shows that the density of sintered W used at a reduction temperature of 850 °C is lower than the density used at a reduction temperature of 750 °C. Figure 29 shows that the oxygen content decreases with an increased reduction temperature. However, as-sintered W samples get higher density at a reduction temperature of 750 °C rather than 850 °C; this is due to the effect of surface diffusion. Powders form sinter bonds between particles and then experience coarsening without densification when a reduction temperature of 850 °C is used. The low-temperature coarsening effect during reduction will consume surface energy, which is used for densification. This is also the reason that the grain size with a reduction temperature of 850 °C is larger than the grain size used at a reduction temperature of 750 °C [57]. Figure 29 shows that the as-sintered W samples have the lowest oxygen content among all compositions; W-1% TiH2 samples show the highest oxygen content. One 83 benefit of nano-sized, after-milled tungsten powder is that part of the tungsten powder will react with oxygen and become tungsten oxide during the milling and drying processes. Assume that both W powder and W-1% TiH2 powder contain the same amount of oxide after milling. During the sintering process, tungsten oxide will be reduced into pure W for W-1% TiH2 samples. TiH2 will dissolve into Ti and H2. Ti is very active and will react with residual oxygen and form titanium oxide, which gives W-1% TiH2 samples a higher oxygen content compared with pure W samples under the same conditions. With increased sintering temperatures, both density and grain size of the sintered W, W-0.5% TiH2, and W-1% TiH2 samples were increased, which followed the classic sintering theory. As sintering temperature increases, the diffusion rate of atoms increases, which promotes densification kinetics [57,60], and high sintering temperatures can motivate grain growth. When the same sintering program is used, both the relative theoretical density and average grain size of sintered W decreases with an increase in the amount of TiH2. The hardness of the tungsten alloys increases when the amount of additive TiH2 increases. Figure 30 shows the relationship between the average grain size and Vickers hardness for W, W-0.5 %TiH2, and W-1% TiH2 with a reduction temperature of 750 °C and a sintering temperature of 1300 °C. It is very clear that TiH2 addition can greatly decrease grain size and also improve the Vickers hardness of sintered W. The great improvement on the Vickers hardness of W-1% TiH2 is the effort, using the additive TiH2, to refine grain size and strengthen grain boundary. The hardness of tungsten increases as the grain size is reduced, which follows the Hall-Petch relationship. Note that TiH2 addition has a 84 negative effect on the densification of sintered W. In order to balance between good density and fine grain size, 750 °C and 1300 °C were chosen as the reduction temperature and sintering temperature for all sintered bend tungsten alloy samples. Table 6 shows the basic characteristics and microstructure of commercially purchased hot-rolled W and sintered W, W-0.5% TiH2, and W-1% TiH2 with a reduction temperature of 750 °C, a sintering temperature of 1300 °C, and an H2 atmosphere that switches to an Ar atmosphere after reduction. The TEM images of commercially purchased hot-rolled W are shown in Figure 31. Sintered W, W-0.5% TiH2, W-1% TiH2, and commercially purchased hot-rolled W plate samples undergo high-temperature 3-PB tests on the MTS system to measure the DBTT and mechanical properties; SEM is used to observe the fracture surface of each broken bend sample. For commercially purchased hot-rolled W samples after the test, surface finish condition and test strain rate both have influence on the DBTT. The DBTT of a commercially purchased hot-rolled W sample is decreased when either the surface finish is refined or the test strain rate is decreased. Tungsten is a strain-rate-sensitive material. Samples of both as-sintered W and W-1% TiH2 didn't show any ductile behavior at test temperatures between room temperature and 600 °C. The DBTT of as-sintered W and W-1% TiH2 is higher than 600 °C. As-sintered W and W-TiH2 samples in low sintering programs without mechanical working still have a long way to go in improving mechanical properties, especially improving ductility and decreasing the DBTT. The biggest problem for the poor mechanical properties of sintered W samples is mainly due to the density not being high enough. Even 98% is a good relative density for 85 sintered W with a sintering temperature of 1300 °C. The relative density of commercially purchased hot-rolled W is 99.9%. Compared with a commercially purchased hot-rolled W sample, the density difference is obvious; the sintered W sample still needs an improved density. Through the SEM images of the fracture surface of a sintered W sample, some of the tiny holes and pores inside the sintered W sample are obvious. A sintered W sample still needs to overcome a density improvement need. However, if it is produced by a powder metallurgy method, holes and pores are inevitable. Besides density, the existence of large grain size is another possibility limiting the mechanical properties of sintered W. Figure 32 shows the grain-size distribution of as-sintered W and W-1% TiH2 with a reduction temperature of 750 °C and sintering temperature of 1300 °C. As can be seen from Figure 33, the values of the average grain size of sintered W and the sintered W-1% TiH2 are decreased; however, large grain still exists in the samples. The large grain is twice the size of the average grain size. The DBTT of as-sintered W in this work didn't improve when decreasing the grain size. It shows that the effect of grain size has no big dependence on the DBTT of as-sintered tungsten. Mechanical working can both decrease the grain size and DBTT of tungsten; during mechanical working, the microstructure and grain size is changed and modified by different working process, which is good for decreasing the DBTT of tungsten. 86 Oxygen Content 1 0.8 0.6 W-1%TiH2 Oxygen Content 0.4 W-0.5%TiH2 W 0.2 0 550 650 750 850 Reduction Temperature (°C) Figure 29. The relationship between oxygen content and the reduction temperature of assintered W, W-0.5 %TiH2, and W-1% TiH2 samples when they have the same sintering temperature at 1300 °C. 87 Figure 30. The relationship between average grain size and Vickers hardness of assintered W, W-0.5 %TiH2, and W-1% TiH2 when all have the same sintering program, with a reduction temperature of 750 °C and a sintering temperature of 1300 °C. 88 Table 6. Basic characteristics of four types of tungsten materials investigated in this work. They are all sintered bend samples with the same sintering program: a reduction temperature of 750 °C, a sintering temperature of 1300 °C, flowing H2 and switched it to Ar atmosphere after reduction process. Hot-Rolled W Sintered Sintered Sintered (PLANSEE) W W-0.5% TiH2 W-1% TiH2 99.9% 97.99% 97.1% 96.4% Material Relative Theoretical Density Approximatel Approximately Approximately spherical, with spherical, with average grain size average grain size of 1.17 μm of 1.01 μm 639.78 HV 745.46 HV y spherical, Microstructure "Pancake" shape with average grain size of 3.09 μm Hardness 331.88 HV 456.3HV 89 Figure 31. TEM images of grain morphology of a commercially purchased hot-rolled W bend sample. 90 Figure 32. Grain-size distribution of as-sintered W and W-1% TiH2 samples with reduction temperatures of 750 °C and sintering temperatures of 1300 °C CHAPTER 7 SUMMARY In this work, effects of different sintering parameters and TiH2 addition on the properties of sintered W were studied, and the 3-PB test at elevated temperatures was used to measure the mechanical properties and DBTT of each bend sample. Based on the experimental results, the following conclusions can be drawn: 1. After 6 hours in a custom-designed, ultrahigh-energy planetary ball milling machine, the particle size and crystalline size of tungsten powders can greatly decrease, and tungsten powder may become more concentrated after milling. 2. TiH2 addition can greatly decrease the grain size of sintered W. However, TiH2 addition also has a negative effect on densification of sintered W. 3. TiH2 addition can greatly improve the Vickers hardness of sintered W. 4. Pure W samples show the highest as-sintered density when a final compact pressure of 146.3 MPa is used. 5. When W, W-0.5% TiH2, and W-1% TiH2 all have the same sintering temperature at 1300 °C, they all get their highest density and smallest grain size at a reduction temperature of 750 °C. 92 6. When W, W-0.5% TiH2, and W-1% TiH2 samples are all at the same reduction temperature, both grain size and density increase as the sintering temperature increases. 7. The oxygen content of as-sintered W decreases when the reduction temperature increases. Under the same sintering conditions, the oxygen content of W will increase with increased TiH2 content. 8. Surface condition and test strain rate affect the mechanical properties of W. Either improving the surface condition or decreasing the test strain rate can decrease the DBTT of commercially purchased hot-rolled tungsten. 9. The DBTT of sintered W is higher than 600 °C, and the DBTT of sintered W-1% TiH2 is higher than 450 °C. The TiH2 addition has no significant effect on improving the mechanical properties of sintered tungsten. REFERENCES [1] Gaganidze, E.; Rupp, D.; Aktaa, J. Fracture Behaviour of Polycrystalline Tungsten. J. Nucl. Mater. 2014, 446 (1-3), 240-245. [2] Stephens, J. R. A Review of the Deformation Behavior of Tungsten at Temperatures Less than 0.2 of the Melting Point. Nasa Tm 1972, No. X-2482, 1- 34. [3] Norajitra, P.; Boccaccini, L. V.; Gervash, A.; Giniyatulin, R.; Holstein, N.; Ihli, T.; Janeschitz, G.; Krauss, W.; Kruessmann, R.; Kuznetsov, V.; et al. Development of a Helium-Cooled Divertor: Material Choice and Technological Studies. J. Nucl. Mater. 2007, 367-370 B (SPEC. ISS.), 1416-1421. [4] Clark, J. Flow and Fracture of Tungsten and Its Alloys: Wrought, Recrystallized and Welded Conditions; 1963. [5] Petch, N. J. The Cleavage Strength of Polycrystals. J. Iron Steel InstInst 1953, 174 (19), 25-28. [6] Bechtold, J. H.; Shewmon, P. G. Flow and Fracture Characteristics of Annealed Tungsten. Trans. Amer. Soc. Met. 1954, 46, 397-407. [7] Lassila, D. H.; Gray, G. T. Ductile-Brittle Transition Behavior of Tungsten under Shock Loading. Le J. Phys. IV 1994, 4 (C8), C8-349-C8-354. [8] Seigle, L. and C. D. Effect of Mechanical and Structural Variables on the DuctileBrittle Transition in Refractory Metals. 1963, 17. [9] Nygren, R. E.; Raffray, R.; Whyte, D.; Urickson, M. A.; Baldwin, M.; Snead, L. L. Making Tungsten Work. J. Nucl. Mater. 2011, 417 (1-3), 451-456. [10] Kurishita, H.; Amano, Y.; Kobayashi, S.; Nakai, K.; Arakawa, H.; Hiraoka, Y.; Takida, T.; Takebe, K.; Matsui, H. Development of Ultra-Fine Grained W-TiC and Their Mechanical Properties for Fusion Applications. J. Nucl. Mater. 2007, 367- 370 B (SPEC. ISS.), 1453-1457. [11] Kecskes, L. J.; Cho, K. C.; Dowding, R. J.; Schuster, B. E.; Valiev, R. Z.; Wei, Q. 94 Grain Size Engineering of Bcc Refractory Metals: Top-down and Bottom-upApplication to Tungsten. Mater. Sci. Eng. A 2007, 467 (1-2), 33-43. [12] Wai Fazhan, Feng penfa, W. H. The Tungsten Materials and Its Process; Metallurgical industry publishing company. [13] Li, K. C.; Wang, C. Y. Tungsten, Its History, Geology, Ore-Dressing, Metallurgy, Chemistry, Analysis, Application, and Economics. GFF. 1945, pp 116-118. [14] Lassner, E. (University of T. V.; Schubert, W.-D. (University of T. V. Tungsten: Properties, Chemistry, Technology of the Element, Alloys, and Chemical Compounds; 2000; Vol. 37. [15] Luo, A.; Shin, K. S.; Jacobson, D. L. Ultrahigh Temperature Tensile Properties of Arc-Melted Tungsten and Tungsten-Iridium Alloys. Scr. Metall. Mater. 1991, 25 (10), 2411-2414. [16] Stephens, J. R. Effects of Interstitial Impurities on the Low-Temperature Tensile Properties of Tungsten. 1964. [17] Raffo, P. L. Yielding and Fracture in Tungsten and Tungsten-Rhenium Alloys. J. Less-Common Met. 1969, 17 (2), 133-149. [18] Campbell, R. and C. D. Effect of Melting Variables on Purity and Properties of Tungsten. 1961, 655-668. [19] Farrell, K.; Schaffhauser, A. C.; Stiegler, J. O. Recrystallization, Grain Growth and the Ductile-Brittle Transition in Tungsten Sheet. J. Less-Common Met. 1967, 13 (2), 141-155. [20] Thornley, J. and A. W. The Effects of Grain Size and Annealing Temperature on the Mechanical Properties of Cast Tungsten; 1970. [21] Wronski, A. and A. F. The Ductile-Brittle Transition in Polycrystalline Tungsten. 1965, 149-158. [22] Jaffee, R. and C. S. The Effect of Rhenium on the Fabricability and Ductility of Molybdenum and Tungsten. 1958. [23] Davis, J. W.; Smith, P. D. ITER Material Properties Handbook. J. Nucl. Mater. 1996, 233-237 (Pt B), 1593-1596. [24] Zukas, E. and W. G. High-Density, High-Strength Alloys Having Some Ductility at Ambient Temperature. 1971. [25] Smid, I.; Akiba, M.; Vieider, G.; Plöchl, L.; Plochl, L. Development of Tungsten 95 Armor and Bonding to Copper for Plasma-Interactive Components. J. Nucl. Mater. 1998, 258, 160-172. [26] Krsjak, V.; Wei, S. H.; Antusch, S.; Dai, Y. Mechanical Properties of Tungsten in the Transition Temperature Range. J. Nucl. Mater. 2014, 450 (1-3), 81-87. [27] Bolt, H.; Barabash, V.; Krauss, W.; Linke, J.; Neu, R.; Suzuki, S.; Yoshida, N. Materials for the Plasma-Facing Components of Fusion Reactors. In Journal of Nuclear Materials; 2004; Vol. 329-333, pp 66-73. [28] Ishijima, Y.; Kannari, S.; Kurishita, H.; Hasegawa, M.; Hiraoka, Y.; Takida, T.; Takebe, K. Processing of Fine-Grained W Materials without Detrimental Phases and Their Mechanical Properties at 200-432 K. Mater. Sci. Eng. A 2008, 473 (1- 2), 7-15. [29] Ullmaier, H.; Carsughi, F. Radiation Damage Problems in High Power Spallation Neutron Sources. Nucl. Instruments Methods Phys. Res. Sect. B Beam Interact. with Mater. Atoms 1995, 101 (4), 406-421. [30] Kawai, M. Fabrication of a Tantalum-Clad Tungsten Target for KENS. J. Nucl. Mater. 2001, 312-320. [31] Baluc, N.; Abe, K.; Boutard, J. L.; Chernov, V. M.; Diegele, E.; Jitsukawa, S.; Kimura, A.; Klueh, R. L.; Kohyama, A.; Kurtz, R. J.; et al. Status of R&D Activities on Materials for Fusion Power Reactors. Nucl. Fusion 2007, 47 (10). [32] Butler, B. Demonstration of Shear Localization in Ultrafine Grained Tungtsne Alloys via Powder Metallurgy Processing Route; 2012. [33] Lee, W.-S.; Lin, C.-F.; Xiea, G.-L. Dynamic Shear Deformation and Failure Behaviour of Pure Polycrystalline Tungsten. Mater. Sci. Eng. A 1998, 247 (1-2), 102-112. [34] Norajitra, P.; Boccaccini, L. V.; Diegele, E.; Filatov, V.; Gervash, A.; Giniyatulin, R.; Gordeev, S.; Heinzel, V.; Janeschitz, G.; Konys, J.; et al. Development of a Helium-Cooled Divertor Concept: Design-Related Requirements on Materials and Fabrication Technology. In Journal of Nuclear Materials; 2004; Vol. 329-333, pp 1594-1598. [35] Cottrell, G. A. A Survey of Plasma Facing Materials for Fusion Power Plants. Mater. Sci. Technol. 2006, 22 (8), 869-880. [36] Monge, M. A.; Auger, M. A.; Leguey, T.; Ortega, Y.; Bolzoni, L.; Gordo, E.; Pareja, R. Characterization of Novel W Alloys Produced by HIP. J. Nucl. Mater. 2009, 386-388 (C), 613-617. 96 [37] Aguirre, M. V.; Martín, A.; Pastor, J. Y.; Llorca, J.; Monge, M. A.; Pareja, R. Mechanical Behavior of W-Y2O3 and W-Ti Alloys from 25 °C to 1000 °C. Metall. Mater. Trans. A 2009, 40 (10), 2283-2290. [38] Miao, S.; Xie, Z. M.; Yang, X. D.; Liu, R.; Gao, R.; Zhang, T.; Wang, X. P.; Fang, Q. F.; Liu, C. S.; Luo, G. N.; et al. Effect of Hot Rolling and Annealing on the Mechanical Properties and Thermal Conductivity of W-0.5 Wt.% TaC Alloys. Int. J. Refract. Met. Hard Mater. 2016, 56, 8-17. [39] Gumbsch, P. Controlling Factors for the Brittle-to-Ductile Transition in Tungsten Single Crystals. Science (80-. ). 1998, 282 (5392), 1293-1295. [40] Faleschini, M.; Kreuzer, H.; Kiener, D.; Pippan, R. Fracture Toughness Investigations of Tungsten Alloys and SPD Tungsten Alloys. J. Nucl. Mater. 2007, 367-370 A (SPEC. ISS.), 800-805. [41] Rieth, M.; Dafferner, B. Limitations of W and W-1%La2O3 for Use as Structural Materials. J. Nucl. Mater. 2005, 342 (1-3), 20-25. [42] Hao, T.; Fan, Z. Q.; Zhang, T.; Luo, G. N.; Wang, X. P.; Liu, C. S.; Fang, Q. F. Strength and Ductility Improvement of Ultrafine-Grained Tungsten Produced by Equal-Channel Angular Pressing. J. Nucl. Mater. 2014, 455 (1), 595-599. [43] Vogler, T. J.; Clayton, J. D. Heterogeneous Deformation and Spall of an Extruded Tungsten Alloy: Plate Impact Experiments and Crystal Plasticity Modeling. J. Mech. Phys. Solids 2008, 56 (2), 297-335. [44] Clayton, J. D. Modeling Dynamic Plasticity and Spall Fracture in High Density Polycrystalline Alloys. Int. J. Solids Struct. 2005, 42 (16-17), 4613-4640. [45] Lennon, A. M.; Ramesh, K. T. The Thermoviscoplastic Response of Polycrystalline Tungsten in Compression. Mater. Sci. Eng. A 2000, 276 (1-2), 9- 21. [46] Dümmer, T.; Lasalvia, J. C.; Ravichandran, G.; Meyers, M. A. Effect of Strain Rate on Plastic Flow and Failure in Polycrystalline Tungsten. Acta Mater. 1998, 46 (17), 6267-6290. [47] Zhang, Y.; Ganeev, A. V.; Wang, J. T.; Liu, J. Q.; Alexandrov, I. V. Observations on the Ductile-to-Brittle Transition in Ultrafine-Grained Tungsten of Commercial Purity. Mater. Sci. Eng. A 2009, 503 (1-2), 37-40. [48] Barabash, V.; Federici, G.; Rödig, M.; Snead, L. L.; Wu, C. H. Neutron Irradiation Effects on Plasma Facing Materials. J. Nucl. Mater. 2000, 283-287 (PART I), 138-146. 97 [49] Tejado, E.; Carvalho, P. A.; Munoz, A.; Dias, M.; Correia, J. B.; Mardolcar, U. V.; Pastor, J. Y. The Effects of Tantalum Addition on the Microtexture and Mechanical Behaviour of Tungsten for ITER Applications. In Journal of Nuclear Materials; 2015; Vol. 467, pp 949-955. [50] Sasaki, K.; Yabuuchi, K.; Nogami, S.; Hasegawa, A. Effects of Temperature and Strain Rate on the Tensile Properties of Potassium-Doped Tungsten. J. Nucl. Mater. 2015, 461, 357-364. [51] Witzke, W. D. K. and W. R. Mechanical Properties and Recrystallization Behavior of Electron-Beam-Melted Tungsten Compared with Arcmelted Tungsten. 1996. [52] Liu, R.; Zhou, Y.; Hao, T.; Zhang, T.; Wang, X. P.; Liu, C. S.; Fang, Q. F. Microwave Synthesis and Properties of Fine-Grained Oxides Dispersion Strengthened Tungsten. J. Nucl. Mater. 2012, 424 (1-3), 171-175. [53] Battabyal, M.; Schäublin, R.; Spätig, P.; Walter, M.; Rieth, M.; Baluc, N. Microstructure and Mechanical Properties of a W-2wt.%Y2O 3 Composite Produced by Sintering and Hot Forging. J. Nucl. Mater. 2013, 442, S225-S228. [54] Itoh, Y.; Ishiwata, Y.; Miyazaki, M. Strength Properties of Yttrium-OxideDispersed Tungsten Alloy. Trans. Japan Soc. Mech. Eng. Ser. A 1995, 61, 595- 599. [55] Stephens, J. R. An Exploratory Investigation of Some Factors Influencing the Room-Temperature Ductility of Tungsten. 1960. [56] Ren, C.; Koopman, M.; Fang, Z. Z.; Zhang, H. The Effects of Atmosphere on the Sintering of Ultrafine-Grained Tungsten with Ti. JOM 2016, 68 (11), 2864-2868. [57] Ren, C.; Fang, Z. Z.; Zhang, H.; Koopman, M. The Study on Low Temperature Sintering of Nano-Tungsten Powders. Int. J. Refract. Met. Hard Mater. 2016, 61, 273-278. [58] Lassila, D. H., F. M. and D. F. Ductile-Brittle Transition Temperature Testing of Tungsten Using the Three-Point Bend Test. 1991. [59] Pichaud, M.; Drechsler, M. A Field Emission Measurement of the Influence of Adsorption on Surface Self-Diffusion. Surf. Sci. 1972, 32 (2), 341-348. [60] German, R. M. Sintering Theory and Practice. Theory Pract. 1996, 1, 550.