Available online at
Physics Procedia 12 (2011) 255–263
1875-3892 © 2011 Published by Elvier Ltd.艾薇儿好听的抒情歌
doi:10.1016/j.phpro.2011.03.033LiM 2011
Microstructure and mechanical properties of Selective Lar Melted
18Ni-300 steel.
K.Kempen a *, E.Yasa a , L.Thijs b , J.-P. Kruth a , J.Van Humbeeck b
a Department of Mechanical engineering, Katholieke Universiteit Leuven, Leuven, Belgium
b Department of Metallurgy and Materials engineering ,, Katholieke Universiteit Leuven, Leuven, Belgium Abstract
Selective Lar Melting (SLM) is an Additive Manufacturing process in which a part is built in a layer by layer manner. A lar
source lectively scans the powder bed according to the CAD data of the part to be produced. The high intensity lar beam
makes it possible to completely melt the metal powder particles to obtain almost fully den parts. In this work, the influence of
process parameters in SLM (e.g. scan speed and layer thickness) and various age hardening treatments on the microstructure and
mechanical properties of 18Ni-300 steel is investigated. It is shown that almost fully den parts with mechanical properties
staccato
comparable to tho of conventionally produced maraging steel 300 can be produced by SLM.
Keywords: Selective Lar Melting; Maraging steel 300; Additive Manufacturing; Microstructure
1.Introduction
Selective lar melting, SLM, is a powder-bad additive manufacturing process that allows the production of
alas
functional three-dimensional parts directly from a CAD-model. During the process, successive layers of metal
powder particles are molten and consolidated on top of each other by the energy of a high intensity lar beam.
Conquently, almost fully den parts without need for any post-processing other than surface finishing are
produced. Customized medical parts, tooling inrts with conformal cooling channels and functional components
with high geometrical complexity are good examples to reveal the scope of the application areas of this process
[1, 2].
*
Corresponding author. Tel.: +32 16322552.
E-mail address : karolien.kempen@mech.kuleuven.be.
256K.Kempen et al. / Physics Procedia 12 (2011) 255–263
Figure 1.A schematic view of the SLM process
Lar processing of materials is generally accompanied with high cooling rates due to the short interaction time and high thermal gradients. The high cooling rates during SLM may result in the formation of non-equilibrium phas, quasi-crystalline phas and new crystal phas with extended composition ranges [2]. Finer structures may be obrved in the microstructure at sufficiently high cooling rates compared to the conventional manufacturing methods [3]. Moreover, during the SLM process, gas bubbles and oxide inclusions can become entrapped in the material during solidification due to various caus such as decrea in the solubility of the dissolved elements in the melt pool during cooling and solidification, chemical reaction or trapped gas. Therefore, the mechanical and material properties obtained after SLM may be different than the properties of materi
als produced by conventional production techniques.
In this study, the influence of the SLM-process parameters on the density, the microstructure and the mechanical properties was investigated. Moreover, the effect of temperature and duration of the hardening (aging) heat treatment on the obtained hardness was studied.
2.Background
Maraging steels are well known for combining good material properties like high strength, high toughness, good weldability and dimensional stability during aging heat treatment. Mainly maraging steels are ud for two application areas: the aircraft and aerospace industry in which superior mechanical properties and weldability of maraging steels are the most important features, and condly in tooling applications which require superior machinability [4]. The nominal composition of maraging steel, grade 300, is given in Table 1.
Table 1.Chemical composition of 18Ni-300 steel [5,6]
Alloying
Fe Ni Co Mo Ti Al Cr C Mn,Si P,S element
wt % rest 17-19 8.5-9.5 4.5-5.2 0.6-0.8 0.05-0.15 <0.5 <0.03 <0.1 <0.01
Maraging steels are a special class of high-strength steels that differ from conventional steels in that they are hardened by a metallurgical reaction that does not involve carbon [4]. The relatively soft body centered cubic martensite, that is formed upon cooling, is hardened by the precipitation of intermetallic compounds at temperatures of about 480°C.会计岗位责任制
3.Experimental procedures
移魂女郎下载A Concept Lar M3 Linear machine was ud to build the specimens. This machine employs a diode-pumped Nd:YAG lar with a wavelength of 1,064 nm and a maximum lar output power of approximately 100 W measured in continuous mode. The lar beam diameter d99% at the powder bed surface is about 180 μm. The powder materials were supplied by Concept Lar (CL 50WS) [5] and LPW (M300-1) [6].
氛围英语
K.Kempen et al. / Physics Procedia 12 (2011) 255–263257 The SLM samples ud to perform the heat treatments and mechanical testing were built with a t of process parameters chon in terms of maximal density: a layer thickness of 30 μm, a scan speed of 150 mm/s and a scan spacing of 112 μm (62% of the spot size). The samples were heat treated in a vertical tube furnace in N2–atmo
sphere and cooled down in air.
Both optical and Scanning Electron Microscopy (SEM) techniques were ud to evaluate the microstructure and the fracture surfaces of the mechanical test specimens. X-ray diffraction patterns were measured by the Siemens D500 goniometer, whereas Rietveld Refinement analysis was done by X’pert software Plus.
To indentify the mechanical properties, tensile, hardness and Charpy impact tests were conducted. The tensile tests were conducted on an Instron 4505 testing machine. Macro-hardness is measured according to the Rockwell A scale (60kg load), while micro-hardness was measured on a Vickers scale (0.5 kg load). Most hardness results were transformed to the Rockwell C scale, according to [7].
4.Results and discussion
This study mainly consists of three parts. First the influence of the process parameters (scan speed and layer thickness) on the hardness and microstructure is investigated. Secondly, the age hardening heat treatment was optimized for maraging steel 300 parts produced by SLM. Lastly, the mechanical properties of the as-built SLM parts and heat treated parts are determined and compare
d to each other and to the values obtained for conventional wrought maraging steel.
4.1.Influence of process parameters on hardness and microstructure
The influence of the layer thickness (30–60 μm) and scan speed (120–600 mm/s) on the macro-hardness is shown in Figure 2a. The process parameters both have an important influence on macro-hardness through their influence on the part density. Relative density of the parts is depicted in Figure 2b.
b)
a) Figure 2: (a) Macro-hardness for samples with different scan speeds and layer thickness, within a confidence level of 95% for 8 measurements.compassionate
(b) Relative density for samples with different scan speeds and layer thickness.
258K.Kempen et al. / Physics Procedia 12 (2011) 255–263
Figure 3: Macro-hardness for samples with different scan speeds and layer thickness, within a confidence level of 95% for 8 measurements.
The indentations of the micro-hardness measurements are much smaller, making it possible to only indent the consolidated material. As a result, the hardness results are not longer influenced by the porosity, which is shown in Figure 3. The layer thickness and scan speed have almost no significant influence on the micro-hardness. For high scan speeds and/or layer thickness there is a small decrea in hardness and increa in scatter due to the difficulties in performing the measurement for very porous samples.
The top and side cross-ctions were obrved under a SEM are depicted in Figure 4. The micrograph on the left shows bi-directional scan tracks while the one in the center depicts the cellular/dendritic solidification morphology and epitaxial growth of the grains on a ction perpendicular to the layer build quence. In SLM, the cooling rate is very high and rapid solidification prevents formation of a lath martensite. The intercellular spacing is less than 1 μm for the cellular structure and this contributes to the excellent strength and hardness. The statements are also validated by other rearchers working on direct metal lar sintering of maraging steel 300 procesd on an EOS machine [8]. In, Figure 4b,oOne can obrve a large inclusion with a size of about 10-20 μm, visible as a dark spot in the cross-ction. The EDX analysis carried out on this inclusion confirmed that this was a titanium and aluminum combined oxide (TiO2 : Al2O3). Also other oxides containing a combination of Ti, Mo, Al and Si in other ratios were prent in the sample. [8].
a) b) c)
Figure 4 a): SEM pictures of the cross-ctions (top view) of a sample produced with a scan speed of 120 mm/s and a layer thickness of 60 μm.
schoolbagb): SEM pictures of the cross-ctions (side view) of a sample produced with a scan speed of 120 mm/s and a layer thickness of 60 μm.
怎样提高表达能力
c): Formation of a fine dendritic structure due to rapid solidification
In order to perform mechanical testing, porosity must be eliminated as much as possible. Lar re-melting every SLM layer before adding a new layer can contribute to a higher density, up to 99.4%. Re-melting every layer results in an increa of both density and micro-hardness, at the cost of longer production times.
K.Kempen et al. / Physics Procedia 12 (2011) 255–263259
4.2.Optimization of heat treatment
The superior properties of the maraging steels, i.e. good strength and toughness, are achieved by th
amphetamine
e age hardening of a ductile, low-carbon body-centered cubic (bcc) martensite structure with relatively good strength. Therefore, the aging heat treatment is standard for maraging steels. It is aimed to form a uniform distribution of fine nickel-rich intermetallic precipitates during the aging of the martensite. The precipitates rve to strengthen the martensitic matrix. A detrimental side effect is the reversion reaction of metastable martensite into austenite and ferrite [9]. Fortunately, the kinetics of the precipitation reactions of iron-nickel maraging steels are such that a significant age hardening of about 20 HRC points already occurs before the start of this reversion reaction .
The aging heat treatment for maraging steels can be performed for different durations at various temperatures providing that the temperature is lower than the austenite start temperature. For maraging steel 300, the values recommended by the ASM Handbook are 3 to 8 hours at a temperature between 460 and 510°C [9]. In this study, a duration between 1-8 hours and a maximum temperature between 460°C and 500°C are tested.
The micro-hardness of the samples was measured and plotted in Figure 5. When the maximum aging temperature is 460°C, the hardness shows a linear relationship with aging time. As you keep the part longer at this temperature, the hardness continues to increa without any sign of overaging. However for other temperatures tested in the scope of this study, at prolonged durations,
the hardness starts to slightly drop. This is an indication of overaging meaning that the reversion of metastable martensite and coarning of the intermetallic precipitates takes place. The two phenomena together decrea the hardness as the part is kept at elevated temperatures for a prolonged time.
Figure 5: Micro-hardness for samples with identical process parameters and different aging parameters.
Optimal heat treatment in terms of hardness and aging time is lected as aging 5 hours at 480°C. After this heat treatment a hardness of 58 HRC is achieved, which means an increa of 18HRC co
mpared to the as-built part using remelting (this is the re-scanning of a recently/just scanned and solidified layer). Higher hardness can be achieved for lower aging temperatures, but the increa in hardness is not significant in terms of the additional aging time.
In order to determine the volumetric percentage of austenite (fcc, face cubic centered) and martensite (bcc, body cubic centered), that are prent before and after aging, a XRD measurement was performed. The results are shown in Figure 6. The applied heat treatment caus an increa of the austenite pha. The austenite reversion is inevitable for long aging times, becau the martensite is metastable and transforms to the stable austenite. The relea of Ni into the Fe matrix which accompanies the transformation from Ni3(Mo,Ti) to the more stable Fe2Mo precipitates promotes the austenite reversion [10].
