INFLUENCE OF DEFORMATION ON MICROSTRUCTURE OF TI-15V-3CR-3SN-3AL ALLOY

Transcription

1 INFLUENCE OF DEFORMATION ON MICROSTRUCTURE OF TI-15V-3CR-3SN-3AL ALLOY Abstract Zdenek SPOTZ a, Tonu LEEMET b, Pawel ROKICKI a, Lenka FUSOVA a, Karel SAKSL a Veli-Tapani KUOKKALA b, Carsten SIEMERS c a Slovak Academy of Sciences, Institute of Materials Research, Watsonova 47, Kosice, Slovakia, me2dd@seznam.cz b Tampere University of Technology, Department of Materials Science, Korkeakoulunkatu 10, Tampere, Finland c Technische Universität Braunschweig, Institut für Werkstoffe, Langer Kamp 8, Braunschweig, Germany Titanium and its alloys are widely used in industrial applications due to their attractive properties.ti-15v-3cr- 3Al-3Sn, a solute-rich, precipitation hardenable, beta titanium alloy shows a unique combination of low density, high strength, ductility, and corrosion resistance. During the production of semi-finished parts, several deformation steps (e.g., rolling or forging) are necessary, leading to different microstructures and properties especially at the rod ends. The aim of the current study is therefore to investigate the microstructure-property relationships of Ti-15V- 3Cr-3Al-3Sn exposed to deformation at different strain rates and temperatures. Samples of Ti-15V-3Cr-3Al- 3Sn alloy have been deformed in quasi-static and Hopkinson Split Pressure Bar experiments between room temperature and 900 C at strain rates between 10-3 s -1 and 10 3 s -1. The resulting microstructures were investigated by means of optical microscopy, scanning electron microscopy and hard x-ray diffraction. In addition, the hardness of the samples was investigated. The size and shape of grains were determined and compared with the undeformed state. The microstructure and properties strongly depend on the initial experiment temperature and strain rate. The highest increase of lattice parameter has been observed at low strain rates and low temperature, which has been identified by the x-ray diffraction analyses. 1. INTRODUCTION Titanium is one of the most common metals on the Earth. After aluminium, iron and magnesium, it is the fourth most common metal in the planet with a concentration of about 0.6% [1,2]. Titanium exists in two crystallographic forms. At room temperature, pure titanium has a hexagonal close-packed (hcp) crystal structure, also referred to as the α-phase. At 883 C, the α-phase transforms to a body-centred cubic (bcc) structure, known as the β-phase. The manipulation of these crystallographic variations through alloying and thermal processing is the basis for the development of a wide range of alloys and properties. Over the last 40 years, the commercial production of titanium and its alloys has increased steadily. As these materials have very attractive properties, they have found applications in many industrial fields. The Ti-15V-3Cr-3Al- 3Sn alloy is a solute-rich β titanium alloy, developed primarily to lower the cost of titanium sheet metal parts by reducing processing cost through the capability of being strip producible and its excellent roomtemperature formability characteristics. The stabilization of β-phase is done by addition of vanadium and chromium, while aluminium is a β -phase stabilizer. In addition, aluminium, vanadium, chromium and tin as alloying elements increase the mechanical properties of the titanium alloy. Titanium alloys can also be aged

2 to attain a wide range of strength levels to meet the requirements of a variety of applications, in many cases replacing hot-formed Ti-6Al-4V and reducing the prize of the final product [1,3,4]. The mechanical behaviour of materials at high strain rates differs considerably from that observed at quasistatic or intermediate strain rates. High strain rates occur in many processes and events of practical importance, such as in dynamic structural loadings, landing gears, etc [5]. The aim of this work is therefore to investigate the microstructure-property relationships of Ti-15V-3Cr-3Al-3Sn exposed to deformation at different strain rates and temperatures. 2. MATERIAL For the experiments, a solute rich β titanium Ti-15V-3Al-3Cr-3Sn was used in conditions typical to many industrial applications. Ti-15V-3Al-3Cr-3Sn alloy was produced by VAR melting followed by one remelting step. The final ingot was deformed by rotary swaging at 850 C followed by water quenching from 700 C [6]. Samples of the Ti-15V-3Cr-3Al-3Sn alloy were electro discharge machined to cylinders with a diameter of 8mm and a length of 6mm. The specimens were deformed at room temperature, 600 C and 900 C at three different strain rates, 10-3 s -1, 1 s -1 and 10 3 s -1. Split Hopkinson Pressure Bar (SHPB) is the most widely used technique for conducting high strain rate tests in the range of 10 2 to 10 4 s -1. The compression test apparatus consists of two slender bars, between which the small cylindrical specimen is sandwiched. The actual test is performed by impacting a striker bar to the free end of the first (incident) bar and by the consequent travel of the elastic stress pulse through the specimen, deforming it at a high rate. For the current research, a SHPB device with Maraging steel (YS ~ 1850MPa) pressure bars was used. Room temperature and elevated temperature tests were performed with an identical setup. For the high temperature SHPB tests, a special technique developed at the Department of Materials Science of Tampere University of Technology was used. In this method, fast pneumatically driven mechanical specimen and bar manipulation systems with precise timing are applied. Low strain rate tests were conducted on standard 100kN servohydraulic testing machines Instron 8800 and MTS 810 [7]. After the deformation test, samples were embedded in Polyfast embedding powder (Struers) and metallographic cuts were made by the following procedure: wet grinding of the samples using silicon carbide papers with water down to grit size P2500, followed by fine polishing with diamond, particle size 1 μm, and rinsing with oil. To remove the surface oxide layer, the samples were finally polished by a mixture of OPS (grain size 0,06μm, Sommers company) and hydrogen peroxide 2:5. Finally, the samples for metallographic observation were etched in the Kroll reagent (100ml of H 2 O, 6ml of HNO 3, 3ml of HF). The general microstructures were studied by means of optical microscopy (Olympus GX 71), while the microstructural details were investigated by a scanning electron microscope (JEOL JSM LV-5600) equipped with an EDX analyzer. The microstructures of deformed specimens were compared with an undeformed asprepared sample (in figures marked as ref ). From the micrographs the grain sizes and shapes were determined by analytical software ImageJ [8]. Hardness measurements were carried out by a Vickers hardness tester Heckert 309/54. The conditions for all hardness measurements were the same, i.e., an indentation load of 10kg was applied for 10 seconds. High-energy X-ray diffraction measurements were performed at HASYLAB at DESY (Hamburg, Germany) on the experimental station BW5 using monochromatic synchrotron radiation of 100.6keV (λ = Å). The samples measured at room temperature were illuminated for 10 seconds by a well collimated incident beam of 1mm 2 in cross-section. XRD patterns were recorded using a 2D detector (mar235 Image plate) in

3 symmetric mode. The obtained XRD patterns were integrated to the scattering angle-intensity space by using the program Fit2D [9]. Phase analysis was performed by the CMPR (with database Logic) [10] and lattice parameter refinement by the PowderCell [11] computer programs. 3. RESULTS 3.1 Optical and Electron Microscopy The microstructures of specimens deformed at the lowest and at the highest strain rates, 0.001s -1 and 1000s - 1, are shown in Fig.1 and Fig.2, respectively. The specimen deformed at room temperature at strain rate 1000s -1 shows significantly higher amount of twins, as seen in Fig 2.a). A detailed view of the twins at a grain boundary is shown in Fig.3b). The average grain sizes of the deformed specimens were compared with the reference sample, as shown in Fig.4. The average grain size of the sample deformed at 900 C is considerably smaller, which is a consequence of thermally induced dynamic recrystallisation. The elemental analysis (EDX) did not show any significant changes in the chemical composition at grain boundaries, twins, or grain interiors for any of the tested specimens. a) b) c) Fig. 1. Microstructure of samples deformed at strain rate s -1 at 600 C b) and at 900 C c) at room temperature a), a) b) c) Fig. 2. Microstructure of samples deformed at strain rate 1000 s -1 at room temperature a), at 600 C b) and at 900 C c) a) b) Fig. 3. Microstructure of samples deformed at room temperature at strain rate s -1 a) and 1000 s -1 b)

4 a) grains size D crossection b) grains size D crossection grains size D (μm) RT 600 C 900 C ref grains size D (μm) s-1 1 s s-1 ref strain rate (s-1) temperature ( C) Fig. 4. Average grain size of deformed specimens as a function of strain rate a) and temperature b) 3.2 Hardness Fig.5 shows the hardness dependence of the specimens on temperature and applied strain rate. Samples deformed at room temperature exhibit significantly higher (~12%) hardness compared to the reference sample, while specimens deformed at elevated temperatures (600 C and 900 C) show the same or even lower values compared to the reference (Fig.5b). This observation can be explained by the relaxation of the dislocation structure due to the high temperature. For the tests performed at 900 C, hardness of the specimens decreases with increasing strain rate, which is a consequence of progressive sample recrystallization, confirmed also by the following analysis. a) Hardness b) Hardness RT 600 C s-1 1 s-1 HV C HV s strain rate (s-1) temperature ( C) 3.3 X-Rays Diffraction Fig.6 shows the XRD patterns of all tested specimens. Phase analysis proves the presence of only one phase in all samples, i.e., bcc Ti (S.G: Im3m). The Bragg peaks of this phase are labelled by squares in Fig.6. The calculated lattice parameters from the phase are shown in Fig.7 as a function of strain rate and initial test temperature. The samples deformed at room temperature show in all cases lattice parameter values that are above those of the reference sample, and there also is a clear trend for the lattice parameter to decrease with increasing strain rate. At high strain rates and with increasing strain, the deformation energy stored temporarily in the microstructure is responsible for the increase in the overall sample temperature,

5 leading further to the annihilation of stacking faults, dislocations, vacancies etc., because of the progressive recrystallisation of the sample. a) Fig. 6. XRD patterns of all tested specimens. a) b) RT 600 C 900 C s-1 1 s s-1 strain rate (s -1 ) temperature ( C) Fig. 7. Lattice parameters of deformed specimens as a function of strain rate a) and temperature b) 4. CONCLUSIONS In this work, we have studied the microstructure of Ti-15V-3Cr-3Al-3Sn alloy after compressive deformation at strain rates ranging from 0.001s -1 to 1000s -1 in the temperature range of 20 C to 900 C. The sample deformed at room temperature at the strain rate of 1000s -1 showed significantly higher amount of twins compared to the other deformation states. The twins result from the high initial energy required for the dislocation movement at this test condition. Reduction of the average grain size is clearly visible for all experiments performed at 900 C due to the dynamic recrystallisation. The chemical analysis revealed no significant chemical changes at grain boundaries, twins, etc. for any of the tested specimens. Hardness of the specimens decreases with increasing initial test temperature for all strain rates due to the relaxation and annihilation of the dislocation arrangements. For tests performed at 900 C, the hardness additionally decreases with increasing strain rate, which probably is caused by the excess adiabatic heating of the specimen during deformation. The lattice parameter decreases with increasing strain rate at all

6 studied temperatures. For the two lowest strain rates (0.001s -1 and 1s -1 ), the lattice parameter decreases with increasing initial temperature of the experiment. This could be explained by the improving mobility of crystallographic defects and their subsequent annihilation. ACKNOWLEDGEMENTS The research leading to these results has received funding from the European Union Seventh Framework Programme (FP7/7-2013) under grant agreement No. PITN-GA , project MaMiNa. Financial support of the European Commission is therefore gratefully acknowledged. The hard-x-ray investigations were carried out at HASYLAB (DESY), Germany, beam line BW5. REFERENCES [1] R. BOYER ET AL., Materials Properties Handbook-Titanium Alloys, ASM International, 1994 [2] T. BELL AND H. DONG, Proceedings of the 12th IFHT and SE Congress Vol. 2, 0 [3] A. ZHECHEVAA et al., Enhancing the microstructure and properties of titanium alloys through nitriding and other surface engineering methods, Surface and Coatings Technology, Vol., Issue 7, 5 [4] W. SMITH, Structure and Properties of Engineering Alloys (2nd ed.), McGraw-Hill, New York, 1993 [5] M. HOKKA ET AL., Mechanical Testing of Materials with the Hopkinson Split Bar Technique, [6] C. SIEMERS et al., Development of advanced and Free-Machining Titanium Alloys, COM2010, Canada, 2010 [7] J. RÄMO, V.-T. KUOKKALA, T. VUORISTO, Influence of strain rate and adiabatic heating on the deformation behaviour of cold heading steels, Journal of Material Processing Technology, 9 [8] ImageJ, software, actualised 2010 [9] Fit2D, software, actualised 4 [10] CMPR and LOGIC, software and database, actualised 6 [11] PowderCell, software,