iii iv TABLE OF CONTENTS ................................ ................................ ................................ ................................ ......... ................................ ................................ ................................ ................................ ..... ................................ ................................ ................................ ................... ................................ ................................ ................................ ..................... 1.1 Motivation ................................ ................................ ................................ ............................ 1 1.2 Background ................................ ................................ ................................ .......................... 3 1.2.1 Introduction of 3D printing technologies ................................ ................................ .......... 3 1.2.2 Self - healing materials ................................ ................................ ................................ ............ 6 1.2.3 Self - healing Effects in Fe - based alloys ................................ ................................ ............... 9 1.2.4 Materials preparation ................................ ................................ ................................ ......... 10 1.2.5 Properties of copper - iron alloy ................................ ................................ .......................... 13 1.2.6 Boron - additives effects on 420 stainless steel ................................ ................................ . 15 ................................ ................................ ................................ ................................ ................................ ............................. 3.1 Samples preparation ................................ ................................ ................................ ......... 18 3.1.1 Powder additive amount ................................ ................................ ................................ ..... 18 3.1.2 Printing ................................ ................................ ................................ ................................ ... 21 3.1.3 Curing ................................ ................................ ................................ ................................ ..... 21 3.1.4 Sintering ................................ ................................ ................................ ................................ . 21 3.1.5 Shapes and density test ................................ ................................ ................................ ........ 22 3.2 Specimens preparation and processes of fatigue test ................................ ..................... 27 3.2.1 Specimens printing ................................ ................................ ................................ ............... 27 3.2.2 Grinding ................................ ................................ ................................ ................................ . 28 3.2.3 Surface finish processing ................................ ................................ ................................ .... 28 3.2.4 WEDM ................................ ................................ ................................ ................................ .... 29 3.2.5 Fatigue test ................................ ................................ ................................ ............................. 30 3.2.6 Heat treatment ................................ ................................ ................................ ...................... 31 ................................ ................................ ....... 4.1 S - N diagram ................................ ................................ ................................ ....................... 33 4.1.1 UTS testing ................................ ................................ ................................ ............................. 33 v 4.1.2 Stress and life cycle for healable samples ................................ ................................ ....... 34 4.1.3 Stress and life cycle for standard samples ................................ ................................ ...... 38 4.2 Result of improvement of fatigue life after healing ................................ ........................ 42 4.3 SEM images ................................ ................................ ................................ ....................... 43 ................................ ................................ ................................ ...................... 5.1 Difficulties and problems ................................ ................................ ................................ .. 46 5.2 Data analysis and conclusions ................................ ................................ .......................... 46 ................................ ................................ ................................ ................................ ........ vi ................................ ................................ ................................ ................ ................................ ................................ ... ................................ ................................ ................................ ................................ .... ................................ .... ................................ ............................. ................................ .............. ..... ................................ ........................ ................................ ........ .............. ................................ ................................ ....................... vii ............................... ................................ .......... ................................ ........... ................................ ................................ .............................. ................................ ................................ ................................ .......................... ................................ ......... ............................. ................................ ................................ ............. ................................ ........ ............................... ................................ ........ ................................ .......... ................................ ............... ........................ ................................ ................. viii 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Chapter 2 RESEARCH OBJECTIVES There are several objectives in different stages of the experiments. 1. Determine the amount of copper powder mixed into SS420 powder based on copper - iron phase diagram in order to achieve the best mechanical properties and the subsequent fatigue test. 2. Investigate the best sintering additives, the mass fraction of additives and the sintering temperature for copper - iron mixtures t o achieve at least over 95% relative density. A cube of 1 x 1 x 1 is modeled and fabricated by 3D printing for density measurement in this step. 3. The s amples of 4.5 x 0.69 x 0.16 are p rinted f or fatigue test . Each sample should have the identical dimension, density and surface finish. They will be cut into a dog - bone shape subsequently by W ire - Electro discharged machine (w - E DM ) . The S - N diagram will be constructed based on the fatigue test results. 4. The ultimate purpose is to investigate whether or not the addition of copper provides the self - healing of SS420 parts produced by 3D printing in our fatigue testing . As a reference, t he samples without copper powder are also fabricated for the fatigue test. 17 5. The planned fatigue tests consisted of two distinct tests, the one to induce fatigue damage without failure and the other to fracture the sample. To induce the fatigue damage, the fatigue cycle is r un for a certain number of cycles definitely below the fatigue life on at leas t two samples and after the initial fatigue cycle, these samples are heat - treat ed to heal the microscopic damages . T he optimal heat treatment temperature should be determined. The second test is the fatigue test under the same stress till samples fails. T he experimental data will prove that the fatigue life of samples with copper is improved much more than those without copper. 6. T he fatigue results must be compared before and after the heat treatment and calculate the improvement in the life cycles on both types of samples. 18 The spherical 420 SS powder (Ex - one, USA), copper powder and boron powder were used for this research. The 420 SS powder is the main powder whose size is with the average The c opper powder with the average size of used to provide the healing of the consolidated stainless steel. The b oron powder with the average size of used as a sintering additive to consolidate the powder mixture to a near full density. 19 20 For each test, the powders are mixed uniformly by the high - speed mixer (DAC 150.1 FVZ, FlackTek inc., Landrum, SC, USA ) which is capable to vary t he speed between 300 and 3500 rpm. Only two minutes are usually enough for mixing them uniformly . The MRF (Materials Research Furnaces) environment - controlled furnace is applied to sinter the samples. The vacuum or argon gas environment can be provided to prevent the oxidation of samples . V acuum environment is found to be better to improve the final density . Different sintering temperature and additives are chosen in this experiment to compare the final densities in order to optimize the sintering condition . Before the sintering process, th e samples are heated to 240°C at a rate of 10°C/min and kept for two hours in air furnace to burn out the binder phase which consists of Ethylene Glycol Monobutyl Ether, Ethylene Glycol and Isopropanol. Each of these binder constituents is burn out the temperatures of 170°C, 197.3°C or 82.6°C, respectively. In the subsequent sintering process , the samples will be heated from 240°C at a rate of 5°C/min to a preset sintering temperature. Once the preset sintering temperature is reached, the furnace wa s kep t at this temperature for six hours to complete the sintering process. The final step is to cool down samples to room temperature at a rate of 10°C/min. T able 1 shows that the samples with different combination s of additives, including without any sin tering a dditive , are sintered between 1150°C and 1350°C. The relative density measurements will be measured using Archimedes method to determine optimal processing conditions . 21 The X1 - Lab 3D printer is used in all of our 3D printing process. Basically, two printing beds, a supply bed and a printing bed , exist in the X1 - Lab . The supply bed has the movable base which pushes a right amount of powder to fill a layer of powder into the printing bed . The powder from the supply bed is pushed to the printing bed by the roller which maintains the smooth layer of the powder on the printing bed . When the printing process begins, the roller moves a layer of power which is set at 0.1 mm from the supply bed to the printing bed . Then, the binder phase is applied selectively on this layer before preparing for next layer. During this process, obviously the supply bed is raised up and printing bed is lowered when print ing in a layer by layer fashion . Binder phase must keep every layer ed sectio n to stick together . T he printed samples are very fragile , which must be c ur ed to stabilize the shapes. This process maintains the shape of samples before burn ing out the binder phase. 22 to preset temperature, for example, 1150°C. Once the preset temperature is reached, the furnace maintains this temperature for 6 hours. The final step is to cool down to room temperature again. In this research, 1150°C is the lowest sintering temperature used. It is better to achieve a high er density at a low er sintering temperature to minimize the any shape distortion . 23 24 25 . On 26 °C to 1350°C , the relative density did not improve . With the additional copper , the final density did not improv e even with the increase in the sintering temperature. The results from the first three experiments show that the sintering temperature of 1300°C is adequate to sinter our samples. B ased on our previous work [ 35 ] , 1% of b oron n itride is added in the experiment run # 4 at 1300°C , which substantially improve the relative density. However, the 3% variation in density from 94% to 97% among the experiment number 4 is still too much . Even though the density is high enough , a full - density 3D printed material is not necessary to prove the self - healing of copper . In the experiment number 5 , two different sizes of 420 SS and 0.5% boron are utilized to achieve the particles in printing . The large stainless - steel powder from Ex - one, USA has the average particle size of 30 and the smaller one from has the average particle size of The mixture of the large and small powders leads to the final parts with almost full density consistently . Considering the energy consumption, sintering temperature must be reduced to 1150°C in the experiment number 6 with the same constituents as the experiment number 5 achieving the equivalent final density . As a result, the ideal powder mixture is determined to be °C for 6 hours. The resulting cubic samples are shown in F igure 6 . 27 28 29 30 3. 2.5 Fatigue test S ample without copper are needed for comparisons. After preparing the samples for the fatigue test, four samples will be used for ultimate tensile strength ( UTS ) test and eight samples are used to construct the S - N diagram. The determination of UTS enable us to limit the highest stress applicable for the fatigue test. To construct the S - N diagram, multiple stress levels below UTS must be applied to our samples . The fatigue life cycles are recorded until the failure occurs on the samples loaded at different stress levels. The endurance li mit is reached if no failure occur s no matter how many numbers of cycles applied below a certain stress threshold . S - N diagrams are plot ted against two variables, magnitude of alternating stress and the number of cycles to failure. The results are usually displayed on logarithmic scales. The S - N diagrams should be built for the healable samples as well as the reference samples. Based on the S - N diagram, two stress levels in the middle range of the fatigue test undergoes about 70% percent of the allowable stress cycle . As a result, the samples attain some microscopic damages such as cracks and voids without failure . Then , these samples after heat - treating for self - healing are put to the fatigue test again until they fail under the cyclic load . T he life cycl es are recorded until it breaks , which are compare d with the life cycle without the heat treatment under the same stress . By comparing the results, we can determine if Also, the same heat t reatment process and fatigue tests are applied to the reference samples without copper to determine the effect of copper. The broken sample after the fatigue test is shown in F igure 8 . 31 3. 2.6 Heat treatment Based on the Fe - Cu phase diagram , the heat treatment cycle is determined to be 950°C, where the phase transformation of iron allows the copper atoms to diffuse into the iron matrix and heal the microscopic damages . This is the anneal ing process used in industries . Annealing alters the physical and sometimes chemical properties of a material to increase its ductility and reduce its hardness for subsequent deformation processing . In our case, the temperature of 950 °C is needed to transform the iron phase from ferrite to austenite. During the annealing, iron atoms migrate in the crystal lattice and the number of dislocations decreases, leading to the change i n ductility and hardness, which means that the samples will heal itself even without the presence of copper. That makes the reference samples necessary in our study. Theoretically, the fatigue life of the samples without copper should be increased after an nealing . The extend of the impro ve ments 32 in the fatigue life for both samples should be compared after heat ing at 950 °C to prove the healing effects by the presence of copper . 33 34 stress ( MPa) number of cycle s number of cycles ( in the logarithmic scale ) 583 20980 4.32 540 19191 4.28 483 70520 4.84 458 66307 4.82 416 88090 4.94 390 83058 4.91 366 219469 5.34 333 1000000 6 333 280000 5.44 333 1300000 6.11 333 1500000 6.17 333 1800000 6.25 333 2000000 6.30 333 10000000 7 35 endurance limit: 333 MPa y = - 199.35x + 1415.7 0 100 200 300 400 500 600 700 4 4.5 5 5.5 6 6.5 7 7.5 stress(MPa) fatigue life cycle(log) linear trend line 36 stress ( MPa) life cycle(log) life cycle 450 4.84 69862.44 380 5.19 156814.06 37 stress ( MPa) 70% life cycle (log) 80% life cycle (log) 85% life cycle (log) 70% life cycle 80% life cycle 85% life cycle 450 3.39 3.87 4.11 48903.71 55889.95 59383.07 380 3.63 4.15 4.41 109769.84 125451.24 133291.95 38 stress ( M P a) life cycle life cycle(log) 550.9 36056 4.55 472.2 78511 4.89 407 142105 5.15 354.15 250000 5.39 354.15 1000000 6 354.15 2000000 6.30 354.15 2500000 6.39 354.15 3000000 6.47 354.15 3500000 6.54 39 stress ( M P a) linear life cycle(log) line life cycle 450 4.97 95292.49 380 5.26 185923.56 Stress (MPa) 70% life cycle (log) 80% life cycle (log) 85% life cycle (log) 70% life cycle 80% life cycle 85% life cycle 450 3.48 3.98 4.23 66704.74 76233.99 80998.61 380 3.68 4.21 4.47 130146.49 148738.85 158035.03 40 endurance limit: 354.15 MPa y = - 241.15x + 1650.7 200 250 300 350 400 450 500 550 600 4 4.5 5 5.5 6 6.5 7 stress(MPa) fatigue life cycle (N) linear trend line 41 0 100 200 300 400 500 600 700 4 4.5 5 5.5 6 6.5 7 7.5 stress(MPa) fatigue life cycle(log) healable sample standard sample 333 42 43 4. 3 SEM images C avities and Cracks 44 The SEM images show the defe c ts of grain boundaries are reduced in the healable samples with copper. As a result, c opper can achieve healing effect for SS420 from structure analysis. 45 46 47 1. 2. 3. 48 49 50 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 51 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 52 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 53 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53.