Journal of Materials Science and Nanomaterials
Open Access

Selective laser melting (SLM); Functionally graded material (FGM); Magnetic grading; Martensitic transformation; Deformation martensite; Thermal strain

Introduction

The magnetic response shifts from paramagnetic to ferromagnetic with the solid-state transition from austenite to martensite. By selectively allowing or suppressing martensitic transformation, this makes it possible to construct an in-situ magnetically graded material from a single alloy composition.

According to reports, 17-4PH stainless steel has a martensite start temperature of 105–132 °C. At room temperature, conventionally processed 17-4PH is mostly martensitic, with less than 10% retained austenite. However, selective laser melting (SLM)-constructed 17- 4PH has been shown to retain up to 80 weight percent of austenite. This effect has been observed in other techniques with similar rapid solidification rates (105–106 K s1), such as melt spinning and splat quenching [1]. This effect has been attributed to rapid solidification driving fine austenite grain size, which in turn suppresses Ms.

Due to the extreme thermal gradients caused by rapid melting and solidification, additively manufactured (AM) parts frequently experience thermal strain. These can distort the finished part, but they could also cause phase change in an alloy that is vulnerable. Metastable retained austenite transformed into martensite under applied deformation in SLM-built 17-4PH, demonstrating transformation induced plasticity (TRIP) [2].

The cementing states of SLM-constructed 17-4PH are supposed to fundamentally settle austenite, possibly to the degree of holding a completely austenitic design. Although thermal strain is typically regarded as an undesirable aspect of additively manufactured components, it may be possible to use it to control the extent of deformation-driven martensitic transformation and, as a result, produce a single-composition material that is magnetically and microstructurally graded.

Studies have already demonstrated grading for composition, precipitate density, grain size, and twinning-an important area of interest within additive manufacturing (AM). Electrical machine architectures, such as switched reluctance motors and internal permanent magnet machines, but especially synchronous reluctance motors, benefit from the enabling technology of magnetic grading [3].

To create a rotating magnetic field, synchronous reluctance motors run current through stator windings. The rotor is pulled in the same direction (synchronously) by this field after locking into it. The rotor is made up of ferromagnetic flux guides arranged in radial shells that are separated by insulating (non-magnetic) gaps that restrict the flux within the guides. Support struts are necessary for the mechanical integrity of the shells, but they can also serve as flux leakage paths, which reduces efficiency.

Currently, the rotors are made by punching slices from stacks of thin sheets of ferromagnetic electrical steel and laminating them together. Because the support struts that hold the shells together are made of the same ferromagnetic steel, their width needs to be kept as low as possible to prevent flux leakage. Rotors can be constructed using axial lamination, but this method is difficult to manufacture [4]. Powder sintering, in which two different powders are used, is another option. When AM is used to functionally grade during construction, magnetic variation from a single alloy composition can be achieved with high spatial resolution.

Experimental method

The first set of builds covered a range of geometries at fixed energy density, while the second set covered a range of energy density, E. The 17-4PH powder’s chemical composition can be found. Supplementary Figure contains SEM images of the powder and the distribution of particle sizes.

Under argon, a Renishaw SLM125 was used to construct all of the samples. A meander scan method with a 67° rotation between layers was used to construct the samples. Electro-discharge machining (EDM) was used to remove the samples from the baseplate to reduce accidental deformation.

For each of the S1–S5 conditions, a set of cylinders with a height of 8 mm were constructed for individual characterisation. A reviewed cuboid was worked from 4 mm wide blocks of each condition requested in diminishing energy thickness, with evaluating in the form plane so the various circumstances were constructed at the same time. The stereolithography (STL) files for each condition were overlapped by 0.1 mm to improve bonding between the interfaces.

Additionally, a set of samples with distinct build plane crosssectional areas was constructed under the S1 and S5 conditions. All of the samples measured 15 mm in height and 10 mm in width, but their lengths ranged from 2 mm to 20 mm. EDM wire cutting was used to cut these samples into 1 mm-thick slices working from the top surface toward the baseplate.

Before the graded sample was taken out of the baseplate, it was tested with a handheld magnet. In the event of unintentional deformation during removal, this was intended for qualitative evaluation of the genuine as-built condition.

EDM-cut 1.5 mm-thick discs from the top of each condition’s cylindrical samples were used for magnetic characterisation. At the University of Manchester, the vibrating sample magnetometer (VSM) was a MicroSense Model 10. The sample mass was used as a basis for normalizing the results. The field was set so that it was parallel to the build direction for each and every measurement.

At PTS (TQM) Ltd., measurements were taken with a Fischer Feritscope MP30 on two-millimeter-thick discs cut from the cylinders below the VSM discs. Each sample received an average of four measurements. Estimations were likewise taken from each of the cuts of the examples with various cross-sectional region [5].

Using a PANalytical X’Pert3 Powder and Cu K radiation with a step size of 0.0394° and a time interval of 5000 s per step, XRD was performed on the discs that had been used for Feritscope measurements previously. The “Peak Analyzer” method and either a Gaussian or a Gauss-Lorentz curve were used to deconvolute data from sections that corresponded to the four major austenite peaks in Origin. Average lattice parameters were calculated using the peak positions.

At the University of Sheffield, a 20 kV FEI Nova FEGSEM was used for scanning electron microscopy (SEM). Images were taken from a vertical section through the graded cuboid that was perpendicular to the build plane. The example was ground and cleaned, then, at that point, carved with Kallings #2 reagent [6].

Results

Graded sample

We discovered that the cuboidal sample produced with graduated 4 mm thick slices of the five conditions successfully demonstrated an in-situ magnetically graded response, ranging from magnetic (ferromagnetic) at the high energy density (S1) end to non-magnetic (paramagnetic) at the low energy density (S5) end. This was accomplished by employing a handheld magnet. Before and after removal from the baseplate, this was confirmed [7].

Attractive characterisation - VSM

The high energy thickness test (S1) showed a delicate ferromagnetic reaction with some paramagnetic material commitment clear at high applied field. Energy density decreased with decreasing ferromagnetic behavior in the intermediate conditions. The behavior was mostly paramagnetic in the low energy density (S5) condition, with only a small amount of ferromagnetic activity.

Because it is a structure-independent characteristic that is unaffected by porosity or dislocations and follows a rule-of-mixtures from the values for the various phases that are present, we decided to use the saturation magnetisation, ms, to evaluate magnetic behaviour [8].

Fully martensitic low sulphur 17-4PH has a saturation magnetization of 162.4 emu g-1. We avoided porosity corrections and used emu g1 throughout because it was consistent with weight percent.

According to our analysis, the high energy density (S1) sample contained 71 wt percent austenite and 29 wt percent martensite. The proportion of martensite decreased and that of austenite increased as the energy density decreased. Austenite weighed more than 99.5 percent and martensite weighed less than 0.5% in the low energy density sample. Repeatability within one weight percent was found in subsequent scans of the S1 sample [9].

Feritscope measurements were used for magnetic characterisation to provide spatially resolved phase information for the various crosssectional samples. For the high energy thickness (S1) condition, all calculations showed martensite content expanding with distance from the baseplate. The relative effects of sample area and aspect ratio on deposition time and sample rigidity (second moment of inertia) may have influenced the various martensite levels found in various samples. Throughout the construction, each of the low energy density (S5) samples displayed an entirely austenitic structure [10].

All of the samples had a dual phase martensite/austenite structure, as demonstrated by phase identification XRD. A double austenite peak was observed in the traces, indicating the presence of two distinct austenite phases (phases 1 and 2). The material was not suitable for Rietveld analysis because it had a crystallographic texture [11]. Instead, we independently extracted the peak positions and areas for the two austenite phases by deconvoluting portions of the data in origin.

A forest of solidification cells with diameters ranging from 0.5 m to 2 m was found to have identical microstructures under all build conditions, as demonstrated by SEM microstructure examination. Energy density had no effect on the size of solidification cells. In addition, the SEM revealed that all samples contained grains ranging from 10 to 20 micrometers. Grain size and solidification cell size did not change with depth through the sample in the build direction, as shown in additional SEM images [12].

Discussion

The primary objective of this study was to suppress thermally driven martensitic transformation in order to produce an in-situ magnetically graded microstructure. Thermal strain was then used to control the rate of deformation-driven martensitic transformation.

Melt pool modeling predicted that cellular solidification would be 15 m across the entire SLM build conditions, with solidification rates ranging from 104 to 106 K s1. Even at very high energy densities, the experimental microstructures seen by SEM proved this. Additionally, it is in line with the reported microstructures of SLM-built 17-4PH and other steels. Grain sizes of 10–20 m were consistent across all build conditions, and there was no observable trend in grain size with increasing energy density [13].

Additionally, the model predicted that any in-situ heat treatment brought about by the addition of subsequent layers and hatches would be too brief to cause this structure to become coarser. SEM images through the depth demonstrated that the solidification cells did not coarsen even under the highest energy density condition, confirming this.

In the absence of a secondary driving force (such as strain), it was anticipated that the SLM-built 17-4PH grain and solidification cell size would be fine enough to suppress thermally activated martensite and produce a fully metastable austenite structure at room temperature [14]. Based on reports that dense dislocation walls that were comparable to the cell boundaries stopped martensite growth, solidification cells were thought to play a role in reducing thermal Ms.

It was anticipated that the low energy density condition would have the lowest thermal strain and, consequently, the highest levels of retained austenite. The paramagnetic response to an applied field and a weighted martensite content of less than 0.5% were confirmed by VSM. Separately, the XRD revealed a predominant austenitic structure with extremely low martensite intensity. This demonstrated that the thermally driven martensitic transformation in this 17-4PH composition could be completely suppressed with a grain size of 10–20 m and a solidification cell size below 2 m [15].

It was also necessary to produce a material with a proportion of ferromagnetic martensite that could be controlled in order to produce a material with a magnetic grade. Thermal martensite would be suppressed in all build conditions because it was anticipated that all build conditions would have comparable solidification cell sizes. Thermal strain was predicted to rise with an increase in energy density, but it was not possible to tell without calibration whether this would be enough to drive controllable martensitic transformation through TRIP [16].

According to the results of the experiments, the high energy density condition contained 29% ferromagnetic martensite, indicating that enough thermal strain had been applied to cause transformation. In addition, the intermediate conditions revealed an increase in martensite with increasing thermal strain, demonstrating that the extent of transformation could be controlled by varying build parameters.

Conclusions

The characteristic microstructure of small grains (10–20 m) containing elongated solidification cells of less than 2 m diameter is driven by the rapid solidification caused by SLM in 17-4PH. Changes in energy density and sample depth were found to hold this consistent.

In the absence of thermal strain and low energy density, this microstructure was fine enough to prevent the thermally driven martensitic transformation in this 17-4PH composition, resulting in a fully austenitic, paramagnetic material.

Even though the microstructure continued to prevent thermal martensite from forming, the increased thermal strain was sufficient to cause martensite to deform at higher energy densities. It was demonstrated that the martensite phase fraction grows in proportion to the thermal strain.

The martensite content was also found to be influenced by the effect of component geometry and the rise in thermal strain with build height.

Utilizing this effect, an in-situ magnetically graded cuboidal sample and a magnetically graded motor that was operated by a demonstrator motor were constructed.

1. Hsiao CN, Chiou CS, Yang JR (2002) Aging reactions in a 17-4 PH stainless steel. Mater Chem Phys 74: 134-142.

Indexed at, Google Scholar, Crossref

2. Christien F, Telling MTF, Knight KS (2013) A comparison of dilatometry and in-situ neutron diffraction in tracking bulk phase transformations in a martensitic stainless steel. Mater Charact 82: 50-57.

Google Scholar

3. Koistinen DP, Marburger RE (1959) A general equation prescribing the extent of the austenite-martensite transformation in pure iron-carbon alloys and plain carbon steels. Acta Metall 7: 59-60.

Indexed at, Google Scholar, Crossref

4. Facchini L (2010) Metastable austenite in 17-4 precipitation-hardening stainless steel produced by selective laser melting. Adv Eng Mater 12: 184-188.

Indexed at, Google Scholar, Crossref

5. Luecke WE, Slotwinski JA (2014) Mechanical properties of austenitic stainless steel made by additive manufacturing. J Res Natl Inst Stand Technol 119: 398-418.

Indexed at, Google Scholar, Crossref

6. Murr LE (2012) Microstructures and properties of 17-4 PH stainless steel fabricated by selective laser melting. J Mater Res Technol 1: 167-177.

Indexed at, Google Scholar, Crossref

7. Rafi HK, Pal D, Patil N, Starr TL, Stucker BE (2014) Microstructure and mechanical behavior of 17-4 precipitation hardenable steel processed by selective laser melting. J Mater Eng Perform 23: 4421-4428.

Indexed at, Google Scholar, Crossref

8. Gu H (2013) Influences of energy density on porosity and microstructure of selective laser melted 17-4PH stainless steel. Solid Freeform Fabrication Symposium 474-489.

Indexed at, Google Scholar

9. Yang HS, Bhadeshia HKDH (2009) Austenite grain size and the martensite–start temperature. Scr Mater 60: 493-495.

Indexed at, Google Scholar

10. García-Junceda A, Capdevila C, Caballero FG, de Andrés CG (2008) Dependence of martensite start temperature on fine austenite grain size. Scr Mater 58: 134-137.

Indexed at, Google Scholar, Crossref

11. Takaki S, Fukunaga K, Syarif J, Tsuchiyama T (2004) Effect of grain refinement on thermal stability of metastable austenitic steel. Mater Trans 45: 2245-2251.

Indexed at, Google Scholar, Crossref

12. Kurz W, Trivedi R (1994) Rapid solidification processing and microstructure formation. Mater Sci Eng A 179: 46-51.

Indexed at, Google Scholar, Crossref

13. Buchbinder D, Schleifenbaum H, Heidrich S, Bültmann J (2011) High power selective laser melting (HP SLM) of aluminum parts. Phys Procedia 12: 271-278.

Indexed at, Google Scholar, Crossref

14. Inokuti Y, Cantor B (1976) Splat-quenched Fe-Ni alloys. Scr Metall 10: 655-659.

Google Scholar

15. Mercelis P, Kruth JP (2006) Residual stresses in selective laser sintering and selective laser melting. Rapid Prototyp J 12: 254-265.

Indexed at, Google Scholar, Crossref

16. Mumtaz KA, Hopkinson N (2007) Laser melting functionally graded composition of Waspaloy and Zirconia powders. J Mater Sci 42: 7647-7656.

Indexed at, Google Scholar, Crossref