rus
Issue #6/2015
K.Bazaleeva, O.Hovova, E.Tsvetkova, P.Lukyanov
Study Of Structural Stability Of Austenitic Steel Obtained By Selective Laser Melting
Study Of Structural Stability Of Austenitic Steel Obtained By Selective Laser Melting
By the level of the electrical resistivity it was found that the lattice defectiveness of the steel obtained by SLM is 15% higher than that of the same composition of steel after standard water quenching. The structural stability of austenitic steel 316L obtained by SLM was detected by the temperature dependence of the relative electrical resistivity. The difference in the temperature dependences with temperature increasing for steels after the SLM and water quenching indicates the occurrence of recrystallization processes in the steel obtained by SLM method, which as a whole completed by heating to 1100K.
Теги: austenitic steel defectiveness laser melting structural stability аустенитная сталь дефектность селективное лазерное плавление структурная стабильность
Introduction
Selective laser melting (SLM) is one of additive technologies or the so-called 3d-printing method. The idea of synthesis of bulk metal parts of arbitrarily complex shape by additive selective recrystallization of thin layers of metal powder is practically applied nowadays. It acquires a special urgency in piece production, when no need in additional equipment can significantly reduce the cost of a part manufacturing.
The principle of the LM unit operation is based on applying a thin (approximately several tens of micrometers) layer of powder material, its leveling with a roller and exposing to a scanning laser beam. Thus, a part microlayer with a predetermined shape profile is created; then the next layer of powder is applied, and the following microlayer is created; and so the process is repeated until the complete formation of the object. Since the laser action has high local property (molten pool has a diameter of about 50 microns), and the solid substrate provides an intense heat removal, the structure of the alloy formation occurs under the conditions of ultra-rapid cooling from the liquid state (105ё106 K/s). Furthermore, the fusion zones are overlapped in both width and depth, which leads to the repeated recrystallization of the material, and by laser melting of adjacent sites, the recrystallized material is exposed to additional thermal cycling.
As a result of such complex thermal effects, explicitly non-equilibrium state is observed in the structure of the alloy, which has a significant influence on the properties of the synthesized object. The study of the structural stability of the alloy produced by SLM is not only of scientific, but also of practical interest, as it will allow a conscious approach to the solution of different applications.
One of the properties of the material, the most sensitive to defects in the crystal structure, is the resistivity. It is known that the resistivity of the metal is determined by two factors: the scattering of the conduction electrons on the phonon vibrations of the crystal lattice, naturally, taking into account all available alloying components (the heat-sensitive part of the resistivity, which increases with heating), and the scattering of the conduction electrons in the crystal structure defects (the electrical resistance, which depends on the amount of defects which can only be diminished by heating, if the density of the defects decreases) [1]. Thus, the determination of the specific resistivity of the experimental alloy at a fixed temperature low enough and comparing with a reference structural state will allow to reliably estimating the degree of presence of defects. At the same time, studying the relative change in the resistivity of the experimental sample under the conditions of continuous heating and comparing with the typical behavior of the relevant standard, we can get a clear idea of the extent of the structural stability of the object being studied.
These are the studies which we have conducted in this research.
Materials and research methods
Selective laser melting of powder of austenitic steel 03H16N15M3 (Fe-17% Cr-12% Ni-2% Mo-1% Mn-0.7% Si-0.02% C) was conducted using PHENIX-PM100 unit. The powder was produced by gas atomization with its major fraction of the dispersion less than 25 microns. The SLM process parameters were used which allowed obtaining a solid object of minimum porosity [2]: laser power was 50 W, scanning speed of the laser beam on the surface was 100 mm/s, diameter of the laser spot on the powder surface was 70 microns. Also, to improve the quality of the synthesized object a cross-melting strategy was used, i. e. the laser scanning direction in each following layer was changed to a perpendicular one. The process was conducted in a nitrogen atmosphere at a temperature 80 °C; austenitic nickel-chromium alloy was used a substrate. The synthesized samples were hemispheres with a diameter of 15 mm.
The phase-structural state of steel after laser recrystallization was studied by the methods of metallographic and X-ray diffraction analysis, as well as scanning and transmission electron microscopy.
The distribution of alloying elements in the steel structure was monitored by X-ray microspectral analysis. The analysis was conducted on the foils in a transmission microscope Technai G2 20 TWIN, which reduced the local property of the method to the values comparable to the thickness of the foil, i. e., ~ 50 nm. The relative measurement error was 5%. The concentrations of oxygen and nitrogen in steel were determined by reductive melting of trial samples in the helium atmosphere followed by detection of oxygen content in the form of CO by infrared absorption, and of nitrogen – by variation of thermal conductivity of the gas mixture. The relative error in determining the concentration of impurities was 5%.
This paper studied the temperature dependence of the relative resistivity (RT – resistivity at a measurement temperature; R10K – at 10 K) in the range of 10 to 1473K. The value of the relative electrical resistivity was practically independent of the size of the experimental sample, and therefore it allowed assessing changes in its structural condition. The measurement of resistivity in the temperature range of 10 ч 1473K was performed using a potentiometer circuit with continuous heating of the sample. The measurements in the temperature range of 10 ч 473K were performed with the use a special metallic chamber of varying temperatures, which is filled with helium in a gaseous state and is placed in a vessel with liquid helium. The measurements were carried out in increments of 1 K, wherein the temperature variations were ± 0.03 K. In order to measure the voltage drop in the sample, a nanovoltmeter was used; the current monitoring carried out according to the voltage drop on the reference resistance coil of 100 Ohm. The relative error in determining the resistivity did not exceed 0.5%. The studies were conducted on two samples of steel 03H16N15M3 obtained by SLM and a reference sample of steel obtained according to SLM technology, exposed to a special annealing in vacuum at 1160 °C for 10 h and conventionally quenched in water that guarantees the achievement of a certain structural stability of austenitic steel. The sizes of samples for measurement of resistivity were 25Ч2Ч2 mm.
Results and discussion
The results of X-ray phase analysis showed that after SLM, the studied steel was in a single-phase austenitic state (see Figure 1). The diffraction patterns obtained from the longitudinal (laser scanning plane) and transverse cross-sections of the sample showed that the intensity of X-ray peaks was slightly redistributed, indicating some oriented properties of the object: a longitudinal section of the sample largely coincided with the crystallographic plane {111}.
The structure of the austenitic steel synthesized by selective laser melting is shown in Figure 2. Figure 2 (a, b) clearly shows structural hierarchy characteristic for the objects received by laser recrystallization of powder materials [3—5]. Namely, the structure had molten pools within which crystallization cell were observed with a diameter of about 0.5 microns. Furthermore, the molten pools were divided into areas (fragments) sized 20—30 microns, inside which the orientation of crystallization cells was similar. The size and similar orientation of the fragment material suggests that they were formed from individual grains of powder: under the conditions of ultra-fast heating and cooling rates from the liquid state the atoms of matter, apparently, do not have time to be redistributes in space, retaining their original orientation.
Figure 2 (c) shows the microstructure of one fragment: here you can see the crystallization cells and boundaries therebetween. Spot nature of electron diffraction pattern indicates the same crystallographic orientation of the cells belonging to the same fragment; here the axis of area <100> corresponds to the plane of micrograph. Earlier, research [6] has shown that the boundaries of crystallization cells of austenitic steel obtained by SLM are three-dimensional plexuses of dislocations, such as those observed in strong plastic deformation [7, 8]. The authors of researches [9] and [10] have obtained similar results using alloys of different composition, as well as those synthesized by SLM.
Formation of the structure similar to deformation during laser recrystallization was due to the high thermal stresses arising at superfast speeds of the object cooling from a liquid state [6, 11, 12]. Under the effect of these stresses excess concentration of crystal defects can be generated, particularly dislocations and vacancies.
Figure 3 shows the temperature dependence of the relative resistivity of the studied and reference samples of steel 03H16N15M3. The experimental curves for the two experimental samples obtained by SLM coincided within the measurement error. The graph shows that in the temperature range from 10 to 1473K, the relative resistivity of the sample obtained by SLM is less than that of the reference sample, and this difference ranged from 10 to 15% at different temperatures. Thus, as one would expect, resistivity of steel R10K after laser recrystallization associated primarily with the scattering of the conduction electrons in the crystal structure defects was higher than in the reference standard sample.
It should also be noted that with increasing temperature, the difference between the relative values of the resistivity of the studied and reference samples of steel increased. It was found that the temperature dependence of the experimental sample was expressed much weaker, wherein the difference compared to the reference sample has been increasing up to 1100K. This pattern may indicate that the thermal component of the resistivity increase of the experimental sample was gradually impacted by thermally activated processes of redistribution of defects that should reduce the electrical resistance. The data show that the recrystallization processes in the sample obtained by laser recrystallization, were generally terminated by heating to 1100 K since in further heating the kinetics behavior of the experimental and reference samples generally coincided. This allows us to conclude that the SLM-steel at such high temperatures already possesses certain stability. Its causes have been established during further study.
An estimate of the difference of resistivity at 10K of the steel obtained by SLM and standard tempering showed that Δρ was ~ 7 microohm•cm, and this difference was approximately equal to 15% of resistivity of the quenched sample. Assuming vacancy concentration after laser recrystallization equal to 0.1%, the concentration of dislocations equal to 1011 cm-2 and determining the specific surface area of the crystal boundaries as the average cell size (5·104cm–1), according to the contributions of various defects in the increase of the resistivity disclosed in the literature, [1] the approximate increase in resistivity in the presence of these defects was calculated: it was almost one order less than the value observed experimentally. Thus, vacancies, dislocation plexuses and cell boundaries cannot lead to such a significant increase in resistivity.
It has been suggested that a significant contribution to the resistivity of steel synthesized by SLM, was made by individual impurity atoms and their groups.
Presumably, as a result of thermal cycling of steel during SLM, segregations of alloying elements were formed at the cell boundaries. This hypothesis is supported by the fact that the dislocation cell structure was usually not detected by metallographic analysis, but in this case it was observed (see Figure 2 (a)). According to the results of X-ray microspectral analysis of foils, the molybdenum concentration at the cell boundaries was 2.7 wt.%, and in the "body" of the cells – 2%; the chromium concentration at the boundaries was 17.6%, and 16.8% within the cells. That is, the concentration of Mo and Cr at the boundaries was higher, and the difference was several times greater than the measurement error. It is known that the formation of segregations of alloying elements can stabilize the dislocation structure of the alloy.
Furthermore, during laser recrystallization, the atoms of the protective atmosphere and the oxygen atoms of the oxide film from the surface of the powder could dissolve in γ-solid solution. Determination of the concentration of gaseous impurities showed that the studied alloy contained 0.16 wt.% of nitrogen and 0.09% of oxygen. These concentrations were approximately one order of higher than the nitrogen and oxygen content in the composition of austenitic steel.
Thus, the increased defectiveness of the lattice immediately after SLM was established, which upon subsequent heating was the cause of structural instability and relaxation processes. However, the redistribution of alloying elements, primarily Mo and Cr, makes a certain contribution to stabilizing, whereby the perspectives of applying SLM-steels become quite possible including at different operating temperatures.
Conclusions
It is found that specific resistivity of the austenitic steel obtained by SLM is 15% higher than that in the steel of the same composition in the hardened state. A higher value of specific resistivity of steel, synthesized by SLM is explained by the increased dislocation density, extensive grain boundary surface, excess concentration of vacancies, the presence of segregations of alloying elements Mo and Cr at the cell boundaries, and impurity atoms N and O dissolved in γ-solid solution.
Based on the temperature dependence of the relative resistivity, we can argue that during heating the recrystallization processes occur in the steel obtained by SLM which are largely completed at the temperature of 1100K, and further heating shows certain structural of the studied stability comparable with the steel obtained by conventional methods.
Selective laser melting (SLM) is one of additive technologies or the so-called 3d-printing method. The idea of synthesis of bulk metal parts of arbitrarily complex shape by additive selective recrystallization of thin layers of metal powder is practically applied nowadays. It acquires a special urgency in piece production, when no need in additional equipment can significantly reduce the cost of a part manufacturing.
The principle of the LM unit operation is based on applying a thin (approximately several tens of micrometers) layer of powder material, its leveling with a roller and exposing to a scanning laser beam. Thus, a part microlayer with a predetermined shape profile is created; then the next layer of powder is applied, and the following microlayer is created; and so the process is repeated until the complete formation of the object. Since the laser action has high local property (molten pool has a diameter of about 50 microns), and the solid substrate provides an intense heat removal, the structure of the alloy formation occurs under the conditions of ultra-rapid cooling from the liquid state (105ё106 K/s). Furthermore, the fusion zones are overlapped in both width and depth, which leads to the repeated recrystallization of the material, and by laser melting of adjacent sites, the recrystallized material is exposed to additional thermal cycling.
As a result of such complex thermal effects, explicitly non-equilibrium state is observed in the structure of the alloy, which has a significant influence on the properties of the synthesized object. The study of the structural stability of the alloy produced by SLM is not only of scientific, but also of practical interest, as it will allow a conscious approach to the solution of different applications.
One of the properties of the material, the most sensitive to defects in the crystal structure, is the resistivity. It is known that the resistivity of the metal is determined by two factors: the scattering of the conduction electrons on the phonon vibrations of the crystal lattice, naturally, taking into account all available alloying components (the heat-sensitive part of the resistivity, which increases with heating), and the scattering of the conduction electrons in the crystal structure defects (the electrical resistance, which depends on the amount of defects which can only be diminished by heating, if the density of the defects decreases) [1]. Thus, the determination of the specific resistivity of the experimental alloy at a fixed temperature low enough and comparing with a reference structural state will allow to reliably estimating the degree of presence of defects. At the same time, studying the relative change in the resistivity of the experimental sample under the conditions of continuous heating and comparing with the typical behavior of the relevant standard, we can get a clear idea of the extent of the structural stability of the object being studied.
These are the studies which we have conducted in this research.
Materials and research methods
Selective laser melting of powder of austenitic steel 03H16N15M3 (Fe-17% Cr-12% Ni-2% Mo-1% Mn-0.7% Si-0.02% C) was conducted using PHENIX-PM100 unit. The powder was produced by gas atomization with its major fraction of the dispersion less than 25 microns. The SLM process parameters were used which allowed obtaining a solid object of minimum porosity [2]: laser power was 50 W, scanning speed of the laser beam on the surface was 100 mm/s, diameter of the laser spot on the powder surface was 70 microns. Also, to improve the quality of the synthesized object a cross-melting strategy was used, i. e. the laser scanning direction in each following layer was changed to a perpendicular one. The process was conducted in a nitrogen atmosphere at a temperature 80 °C; austenitic nickel-chromium alloy was used a substrate. The synthesized samples were hemispheres with a diameter of 15 mm.
The phase-structural state of steel after laser recrystallization was studied by the methods of metallographic and X-ray diffraction analysis, as well as scanning and transmission electron microscopy.
The distribution of alloying elements in the steel structure was monitored by X-ray microspectral analysis. The analysis was conducted on the foils in a transmission microscope Technai G2 20 TWIN, which reduced the local property of the method to the values comparable to the thickness of the foil, i. e., ~ 50 nm. The relative measurement error was 5%. The concentrations of oxygen and nitrogen in steel were determined by reductive melting of trial samples in the helium atmosphere followed by detection of oxygen content in the form of CO by infrared absorption, and of nitrogen – by variation of thermal conductivity of the gas mixture. The relative error in determining the concentration of impurities was 5%.
This paper studied the temperature dependence of the relative resistivity (RT – resistivity at a measurement temperature; R10K – at 10 K) in the range of 10 to 1473K. The value of the relative electrical resistivity was practically independent of the size of the experimental sample, and therefore it allowed assessing changes in its structural condition. The measurement of resistivity in the temperature range of 10 ч 1473K was performed using a potentiometer circuit with continuous heating of the sample. The measurements in the temperature range of 10 ч 473K were performed with the use a special metallic chamber of varying temperatures, which is filled with helium in a gaseous state and is placed in a vessel with liquid helium. The measurements were carried out in increments of 1 K, wherein the temperature variations were ± 0.03 K. In order to measure the voltage drop in the sample, a nanovoltmeter was used; the current monitoring carried out according to the voltage drop on the reference resistance coil of 100 Ohm. The relative error in determining the resistivity did not exceed 0.5%. The studies were conducted on two samples of steel 03H16N15M3 obtained by SLM and a reference sample of steel obtained according to SLM technology, exposed to a special annealing in vacuum at 1160 °C for 10 h and conventionally quenched in water that guarantees the achievement of a certain structural stability of austenitic steel. The sizes of samples for measurement of resistivity were 25Ч2Ч2 mm.
Results and discussion
The results of X-ray phase analysis showed that after SLM, the studied steel was in a single-phase austenitic state (see Figure 1). The diffraction patterns obtained from the longitudinal (laser scanning plane) and transverse cross-sections of the sample showed that the intensity of X-ray peaks was slightly redistributed, indicating some oriented properties of the object: a longitudinal section of the sample largely coincided with the crystallographic plane {111}.
The structure of the austenitic steel synthesized by selective laser melting is shown in Figure 2. Figure 2 (a, b) clearly shows structural hierarchy characteristic for the objects received by laser recrystallization of powder materials [3—5]. Namely, the structure had molten pools within which crystallization cell were observed with a diameter of about 0.5 microns. Furthermore, the molten pools were divided into areas (fragments) sized 20—30 microns, inside which the orientation of crystallization cells was similar. The size and similar orientation of the fragment material suggests that they were formed from individual grains of powder: under the conditions of ultra-fast heating and cooling rates from the liquid state the atoms of matter, apparently, do not have time to be redistributes in space, retaining their original orientation.
Figure 2 (c) shows the microstructure of one fragment: here you can see the crystallization cells and boundaries therebetween. Spot nature of electron diffraction pattern indicates the same crystallographic orientation of the cells belonging to the same fragment; here the axis of area <100> corresponds to the plane of micrograph. Earlier, research [6] has shown that the boundaries of crystallization cells of austenitic steel obtained by SLM are three-dimensional plexuses of dislocations, such as those observed in strong plastic deformation [7, 8]. The authors of researches [9] and [10] have obtained similar results using alloys of different composition, as well as those synthesized by SLM.
Formation of the structure similar to deformation during laser recrystallization was due to the high thermal stresses arising at superfast speeds of the object cooling from a liquid state [6, 11, 12]. Under the effect of these stresses excess concentration of crystal defects can be generated, particularly dislocations and vacancies.
Figure 3 shows the temperature dependence of the relative resistivity of the studied and reference samples of steel 03H16N15M3. The experimental curves for the two experimental samples obtained by SLM coincided within the measurement error. The graph shows that in the temperature range from 10 to 1473K, the relative resistivity of the sample obtained by SLM is less than that of the reference sample, and this difference ranged from 10 to 15% at different temperatures. Thus, as one would expect, resistivity of steel R10K after laser recrystallization associated primarily with the scattering of the conduction electrons in the crystal structure defects was higher than in the reference standard sample.
It should also be noted that with increasing temperature, the difference between the relative values of the resistivity of the studied and reference samples of steel increased. It was found that the temperature dependence of the experimental sample was expressed much weaker, wherein the difference compared to the reference sample has been increasing up to 1100K. This pattern may indicate that the thermal component of the resistivity increase of the experimental sample was gradually impacted by thermally activated processes of redistribution of defects that should reduce the electrical resistance. The data show that the recrystallization processes in the sample obtained by laser recrystallization, were generally terminated by heating to 1100 K since in further heating the kinetics behavior of the experimental and reference samples generally coincided. This allows us to conclude that the SLM-steel at such high temperatures already possesses certain stability. Its causes have been established during further study.
An estimate of the difference of resistivity at 10K of the steel obtained by SLM and standard tempering showed that Δρ was ~ 7 microohm•cm, and this difference was approximately equal to 15% of resistivity of the quenched sample. Assuming vacancy concentration after laser recrystallization equal to 0.1%, the concentration of dislocations equal to 1011 cm-2 and determining the specific surface area of the crystal boundaries as the average cell size (5·104cm–1), according to the contributions of various defects in the increase of the resistivity disclosed in the literature, [1] the approximate increase in resistivity in the presence of these defects was calculated: it was almost one order less than the value observed experimentally. Thus, vacancies, dislocation plexuses and cell boundaries cannot lead to such a significant increase in resistivity.
It has been suggested that a significant contribution to the resistivity of steel synthesized by SLM, was made by individual impurity atoms and their groups.
Presumably, as a result of thermal cycling of steel during SLM, segregations of alloying elements were formed at the cell boundaries. This hypothesis is supported by the fact that the dislocation cell structure was usually not detected by metallographic analysis, but in this case it was observed (see Figure 2 (a)). According to the results of X-ray microspectral analysis of foils, the molybdenum concentration at the cell boundaries was 2.7 wt.%, and in the "body" of the cells – 2%; the chromium concentration at the boundaries was 17.6%, and 16.8% within the cells. That is, the concentration of Mo and Cr at the boundaries was higher, and the difference was several times greater than the measurement error. It is known that the formation of segregations of alloying elements can stabilize the dislocation structure of the alloy.
Furthermore, during laser recrystallization, the atoms of the protective atmosphere and the oxygen atoms of the oxide film from the surface of the powder could dissolve in γ-solid solution. Determination of the concentration of gaseous impurities showed that the studied alloy contained 0.16 wt.% of nitrogen and 0.09% of oxygen. These concentrations were approximately one order of higher than the nitrogen and oxygen content in the composition of austenitic steel.
Thus, the increased defectiveness of the lattice immediately after SLM was established, which upon subsequent heating was the cause of structural instability and relaxation processes. However, the redistribution of alloying elements, primarily Mo and Cr, makes a certain contribution to stabilizing, whereby the perspectives of applying SLM-steels become quite possible including at different operating temperatures.
Conclusions
It is found that specific resistivity of the austenitic steel obtained by SLM is 15% higher than that in the steel of the same composition in the hardened state. A higher value of specific resistivity of steel, synthesized by SLM is explained by the increased dislocation density, extensive grain boundary surface, excess concentration of vacancies, the presence of segregations of alloying elements Mo and Cr at the cell boundaries, and impurity atoms N and O dissolved in γ-solid solution.
Based on the temperature dependence of the relative resistivity, we can argue that during heating the recrystallization processes occur in the steel obtained by SLM which are largely completed at the temperature of 1100K, and further heating shows certain structural of the studied stability comparable with the steel obtained by conventional methods.
Readers feedback
