Jiangsu Laser Alliance Guide:

This study investigates the effect of process parameters on the thin wall structure of graded 316L and Inconel 718 stainless steel produced by direct laser metal deposition.

Key words: laser deposition, cladding, functional gradient

Summary

Direct laser metal deposition (LDMD) has evolved from a prototyping method to the production of products from one or more metals. This makes it possible to create gradient compositions with different composition of elements, phases and microstructures in different places. In this paper, we study the process of direct laser deposition of thin-walled structures made of 316L stainless steel and Inconel 718 with a continuous gradient. This paper examines the effect of process parameters such as laser power level and powder mass flow on the deposition of SS316L and Inconel 718 in steel-nickel gradient structures.

Description of the microstructure and identification of phases by optical microscopy and X-ray diffraction analysis. The structure was subjected to mechanical testing through hardness, wear resistance and tensile tests. The XRD results showed that the NbC and Fe2Nb phases formed during the deposition. The effect of experimental parameters on the microstructure and physical properties is determined and discussed. It is shown that the mechanical properties can be controlled by input parameters, and carbide generation makes it possible to selectively control the hardness and wear resistance of functionally graded materials.

1. Introduction

In 1984, materials scientists from Sendai, Japan proposed the concept of functionally graded materials (FGM) as a means of making thermal barrier materials. Functionally graded materials are a class of advanced materials whose composition and microstructure gradually change from one side to the other, resulting in corresponding changes in properties. These materials can be designed for specific functions and applications. In addition, the gradient change in the material makes it possible to reduce the stress concentration that occurs near a sharp interface between two different phases. Today, the FGM concept has spread to various sectors around the world. Functionally classified materials have found their way into areas such as biomedicine, automotive and aerospace, electronics, optics and nuclear, reactor components and energy conversion.

Microstructure and element distribution of graded TiC-Ni materials with Ni content from 0.10 to 30 wt.% 7

At present, several methods are used to produce functionally classified materials, such as compression molding, plasma spraying, slip casting, and powder metallurgy. In the preparation of functionally graded coatings, compression molding and plasma spraying are usually used, and the coating of a functionally graded material obtained by plasma spraying is not dense. Although powder metallurgy can be used to produce bulk functional graded materials, their shape and size are usually limited due to the use of pressurized compaction molds.

The Laser Direct Metal Deposition (LDMD) process allows complex prototypes to be fabricated almost clean, saving time and tooling costs. Various metals and alloys are deposited with this process, such as steels such as H13, WC-Co and stellite. Jasim et al. pioneered the laser deposition process to fabricate cermet functionally graded materials. Since then, many researchers have applied this concept to create a series of functionally graded materials for various applications. Pei and De Hosson used Nd:YAG lasers to produce functionally graded AlSi40 materials, and Tivillon et al. analyzed the fabrication of cobalt-based stellite 6 and Inconel 625 nickel-based superalloy by laser deposition. developed WC-(NiSiB alloy) cermet/tool ​​steel graded functional material (FGM) laser cladding for high temperature tribological applications. Lin et al. studied the solidification behavior and morphological evolution of stainless steel during composition change to Rene88DT.

As shown, first, using the same aluminum based system results in similar thermal performance. Second, compositional gradients can be generated for desired FGMs even at high local dilutions. Finally, primary Si particles can serve as rigid reinforcement for FGM, the size of which can be controlled during solidification. This is very important for in situ FGM formation in laser cladding. The powder prepared by spray atomization technology has a spherical shape with a particle size of 50-125 microns.

Nickel and steel alloys are widely used in the energy and nuclear industries. Austenitic stainless steels have high corrosion resistance due to the formation of an oxide film rich in chromium on the surface. It can vary in thickness and, under certain conditions, can also form a double layer with an additional outer layer, while maintaining its good corrosion resistance. Joining steel is usually not a problem because austenitic welds are resistant to hot cracking, stress and high impact in service conditions. Nickel-Chromium Alloy Inconel 718 alloy is suitable for applications where high temperatures are encountered and the atmosphere is highly carburized and oxidized. The properties of nickel and stainless steel alloys make them suitable for many applications, such as those found in nuclear power and refineries. However, the two alloys are still commonly joined by fusion welding, which can result in less solidification cracking resistance. To overcome the problem of cracking, functional classification of alloys is a possible solution, but there are many process parameters that need to be controlled.

The figure above shows the cross-sectional morphology of oxide scale formed on the surface of a sample oxidized at 950°C for 128 hours, as well as the morphology and redistribution of alloy elements oxidized at 1000°C for 140 hours.

This study investigates the effect of process parameters on the thin wall structure of graded 316L and Inconel 718 stainless steel produced by direct laser metal deposition. The aim of the study was to determine the effect of technological parameters on the microstructure, hardness and wear resistance. While the two materials were well studied on their own, this was a study that had never before been done using this combination. In an earlier paper, Wu et al. investigated the hardness, wear, and microstructure of 316L-Inconel718 graded walls, but considered only one type of pipe wall fabricated with one set of parameters.

The most relevant parameters in LDMD are usually specific energy (defined as power/(beam diameter x travel speed)) which gives the energy density at the surface, line mass (defined as powder flow/travel speed) and given the mass of material available for deposition per unit path length. In this article, laser power and powder mass flow are used as the main process variables and their influence is tested. The wire feed speed and other "secondary" process parameters such as gas flow rate, roughness and substrate temperature are kept constant. Existing studies have shown that different fixed values ​​of these parameters can affect the absolute values ​​of the measurements, but not the determined underlying LDMD processes and trends.

2. Experiment procedure

In the process of laser deposition, a LaserlineLDL160-1500 diode laser with a power of 1.5 kW is used. The 316L stainless steel substrate is positioned such that the beam is perpendicular to a 2.5 mm (fast axis) × 3.5 mm (slow axis) surface on the substrate. Before the experiment, the SS316L substrate was sandblasted on a Guyson sandblaster and then degreased with ethanol. The movement of the x-axis and y-axis (horizontal plane) is controlled by the CNC table. The entire setup was stored in an argon-filled room (glove compartment). FST PF-2/2 powder disc feeder containing two 1.5 liter powder containers to supply 316L stainless steel powder (particle diameter 50-120 µm) and Inconel 718 powder (particle diameter 53-150 µm). 316L stainless steel and Inconel 718 chemistry: 0.03% C, 2.0% Mn, 1.0% Si, 16.0-18.0% Cr, 10.0-14.0% Ni, 2.0 –3.0% Mo, 60% Fe, and 0.042% C. 18% Fe, 19% Cr, 0.5% Al, 1% Ti, 3% Mo, 5% Nb, and 55% Ni, respectively.

First, weigh the 316L and Inconel 718 stainless steel powders separately and then mix them with a mechanical stirrer for 30 minutes as desired.weight percentage. The two materials were calibrated by applying 3 coats of 100wt% SS316L to a block of SS316L substrate, then increasing the weight fraction of Inconel718 in the application mix by 25% every 3 coats while decreasing the weight fraction of SS316L powder. The powder is delivered to the melt bath by argon gas from a coaxial nozzle. A dual-channel powder feeder is used, allowing the powder to be mixed in the nozzle before application. The nozzle is aligned with the center of the laser beam on the substrate and moves parallel to the slow axis at a travel speed of 4 mm/s. The process parameters used in the experiment are shown in Table 1. The exact values ​​of the process parameters were chosen empirically based on previous experimental setups which showed a specific energy operating window of about 42-80 J/mm2 for good mass orbitals.

Table 1 Experimental parameters of the process.

Prepared wall samples were cut, filled with Struers epoxy and polished to 4000 grains. All samples were subjected to electrolytic etching in 10% oxalic acid at a constant voltage of 6 V. The microstructure of the gradient was studied using optical microscopy. The elemental composition was analyzed by X-ray energy spectroscopy (EDS). Microhardness tests were carried out on the cross sections of the erected walls. Wear tests were carried out using a Teer Coated Disc Pin Wear Tester (POD-2) in which specimens were subjected to rotational wear in contact with WC-Co balls.

3. Results

3.1. Macrostructure and dimensions

All parameters together form a thin-walled continuous structure. All samples analyzed for the effect of power and powder consumption adhered well to the substrate with no signs of track breakage and were well structured. On fig. 1 shows an example of a graduated wall (sample 6) made at 550 W power and a powder mass flow rate of 0.834 g/s.

Figure 1. Direct laser cladding of SS316L/Inconel 718 metal with a functional gradient of 6 (A) 100% SS 316L, (B) 75% SS 316L, (C) 50% SS 316L, (D) 25% SS 316L and (E) 0% 316L stainless steel.

An analysis of the track size shows that there is a clear trend towards increased deposit volume with increasing power. Changes in the layer height and layer width for the main input variables of the feed power and mass flow rate of the powder are shown in fig. 2(a) and fig. 2b.

Fig. 2 (a) Effect of laser power and powder consumption on the average layer height. (b) Effect of laser power and powder consumption on the average layer width.

The average layer height of samples with low powder consumption ranges from 0.54 mm to 0.76 mm, while the average layer height of samples with high powder consumption ranges from 0.63 mm to 0.84 mm, and the track height increases with an increase in power. A similar trend was observed in the track width: at low powder consumption, the average layer width was the smallest and amounted to 1.31 mm at a power of 450 W and 1.65 mm at a power of 750 W. At high powder flow rates, the average layer width is 1.37 mm at 450 W and 1.68 mm at 750 W.

For all samples, the layer size was limited by the available laser power, but it began to plateau when the laser power exceeded the critical values ​​for layer height and layer width. This indicates that when the laser power is increased above a critical value, there is not enough powder to use the full laser power, so the deposition moves from the laser power limited region to the powder limited region. This explains the flattening of the plots in Fig. 2 (a and b).

3.2. Microstructure

Good adhesion between multilayer walls. The results showed that the growth of columnar dendrites predominated in all excised tissues. Case 3 in fig. 3 illustrates how this happens. At the edges of the cross section, a growth transition from columnar to dendritic cells can be observed, with these structures also predominating at the layer boundaries of all sections, as shown in Figure 4.

Figure 3. Cross section of Sample 3 showing the addition of Inconel 718 along the wall.

Figure 4. An equiaxed granular structure was observed on the side of the wall (this image was taken from sample 4).

The spacing between branches of the secondary dendrite (SDAS) of each layer was measured for all samples, and the average of the three layers was taken to represent deposition at each powder composition. The block averaging procedure for 0.632 g/s and 0.834 g/s powder spraying is shown in fig. 5 (a and b). For a range of parameters, a flow measurement method from 4.49 µm to 8.44 µm for low powder settling was analyzed, giving the highest performance with 718 100% Inconel powder. For high powder flow deposition parameters, the SDAS ranges are 4.13 µm and 7.76 µm.

Fig. 5 (a) Mean SDAS change from lowerpart of the wall to the top of the wall at a low powder flow rate (0.632 g/s). (b) Average SDAS change from bottom of wall to top of wall at high powder flow rate (0.834 g/s).

This trend indicates that SDAS decreases with increasing powder consumption. This effect can be explained by the observed faster quenching as a result of increased powder mass flow, resulting in a finer dendritic structure. In addition, SDAS increased with building height, suggesting that the local cooling rate also decreased with distance from the baseplate, which acts as a heat sink. This is because the average temperature gradient in the assembly direction decreases with increasing distance from the substrate.

There is no significant difference in the microscopic characteristics of samples prepared at different powder rates, with the exception of the distance between the branches of the secondary dendrite (SDAS).

Despite the different compositions of the molten bath formed during deposition on the entire wall, no evidence of fluidization cracking or solid state cracking due to thermal stresses occurring in the wall was observed. The tendency of austenitic alloys (such as 316L) and nickel alloys (such as Inconel 718) to liquefaction cracking increases significantly with grain size, so the fine microstructure described above may provide some protection against this.

3.3. Phase analysis by X-ray diffraction

In fig. Figure 6 (a–e) shows X-ray diffraction patterns of representative sections of the wall of sample 8 perpendicular to the direction of the gradient. A similar picture was obtained for other samples. On fig. 6(a and b) show that structures made from 100% and 75% SS316L are fully austenitic with the lowest ferrite content. It also shows that there is no clear orientation in the microstructure. As the mass fraction of Inconel 718 increases, a new phase is formed consisting of niobium carbide (NbC) and Fe2Nb, as shown in fig. 6(c-e).

Fig. 6(a-e) X-ray diffraction patterns of sample 8 at various positions along the SS 316L-Inco718 gradient. (a) 100% 316L stainless steel, (b) 75% 316L stainless steel, (c) 50% 316L stainless steel, (d) 25% 316L stainless steel, and (e) 0% 316L stainless steel.

3.4. Tensile test

Tension tests were carried out at room temperature parallel to the direction of deposition or laser scanning using an INSTRON 4507 universal tensile testing machine (traverse speed 1 mm/s). All samples were tested after deposition and the effective tensile strength was calculated for each sample. After the tensile test, failure was observed using a Hitachi S-3400N scanning electron microscope.

All specimens failed due to ductile fracture under moderate strain. In all samples, the destruction ofinitially started on the 100% 316L stainless steel side and then quickly spread to the 100% Inconel 718 side. 7 shows the fracture surface of specimen 1 as an example of the fracture surfaces of all specimens.

Figure 7. The surface of sample 1 is broken and enlarged.

Summary of tensile strength data presented in fig. 8, as a function of laser power, shows a trend towards lower tensile strength with increasing laser power. In addition, tensile strength appears to increase with increasing mass power consumption. The tensile strength of the low flow rate powder deposition process was found to be 526-573.5 MPa, and the tensile strength of the high flow powder deposition process was found to be 531 MPa to 596 MPa. The decrease in tensile strength can be explained by the fact that as the heat input increases, the grain size becomes larger and, therefore, the resistance to the applied load becomes smaller.

Figure 8. Effective tensile strength versus powder consumption.

3.5. Hardness distribution

Measurements of Vickers microhardness were carried out on walls made at different mass flow rates of powder and power, along the height of the walls from bottom to top. The results are presented in Figure 9 depending on the number of substrate layers. On fig. 9(a) shows an approximately parabolic hardness distribution with a maximum measured hardness of 186.1 HV0.1 for the last layer and 168.1 HV0.1 for the first layer applied at high powder rates.

Fig. 9(a–d) Hardness curves for different power and powder mass flow rates.

With light powder spraying, the hardness values ​​of the lower and upper layers were 155.6 HV0.1 and 179.3 HV0.1, respectively. For three other power levels, a similar distribution was measured with variations in the measured values. In addition, in most cases deposits with a high powder flow rate have a higher hardness than deposits with a low flow rate. On fig. 9(a-d) it is clearly seen that in the multilayer structures studied, the hardness first decreases until the weight fraction of Inconel 718 increases to about 50%. Starting from this percentage, an increase in the hardness of the coating is observed.

It is noteworthy that the hardness of the top layer increases significantly for all samples. This can be explained by the fact that the last layer was not reheated, unlike the other layers below it.

3.6. Wear test

All sample walls were stripped from the 316L base plate, cut lengthwise down the middle of the wall, and mounted on a Teer Coated Disc Pin Wear Tester (POD-2). Calculate the specific wear rate using the software available for the machine.

In fig. 10 shows the average specific wear rate of threelayers of constant composition at different power levels. In all samples, MSWR was inversely proportional to hardness when changing composition from 316L stainless steel to Inconel 718. In the first three coats at 100% MSWR SS316L is low, then it increases with the addition of Inconel 718 in the next three coats. In the next 9 coats add MSWR decreases with increasing amount of Inconel 718. This trend is noted for all samples. The higher the power level, the higher the overall MSWR.

In fig. 10 shows the specific wear rates of (a) 450 W, (b) 550 W, (c) 650 W, and (d) 750 W along the gradient at different powers at a load of 20 N and 30 mm/s. sliding speed.

4. Discussion

By varying the laser power and powder flow, the microstructure can be controlled using high-power semiconductor laser deposition. The study showed that the minimum power and maximum flow rate of the powder provide the finest grain structure, but these changes in microstructure occur at the expense of changes in the height and width of the layer.

As the power increases, the degree of grain refinement decreases, and the distance between the branches of the secondary dendrites increases. The reduced cooling rate means that there is enough time for the small dendritic branches to melt and disappear; thus, the distance between branches of secondary dendrites in the alloy increases. In addition, the slower the cooling rate during solidification, the longer it takes for the grains to coarsen.

This answer is echoed by Wu et al. for titanium alloys and Griffith et al. for austenitic stainless steels, who also found that a high cooling rate at low power results in a fine grain structure.

It is also noted that SDA increases with build height, indicating that local solidification time increases with distance from the substrate acting as a heat sink. The increase in solidification time is associated with a decrease in the average cooling rate with distance from the substrate. This is because the average temperature gradient in the assembly direction decreases with increasing distance from the substrate.

An analysis of the tensile strength of gradient structures created under various operating conditions shows that the tensile strength is inversely proportional to the laser power. This is due to the same factors discussed above, except that here the cooling rate and temperature gradient are reduced by increasing power consumption, rather than by changing the characteristics of the heatsink.

It is worth noting that the hardness distribution in the sample follows a roughly parabolic curve, in contrast to conventional single material deposition where the sample hardness generally decreases when measured perpendicular to the substrate. This can be explained by the X-ray diffraction pattern, which shows the presence of NbC and Fe2Nb as the percentage of Inco increases.nel 718. According to Fujita et al., NbC and Fe2Nb increase the strength of the material, so when more Inconel 718 is added and more Nb is present to form NbC, the hardness increases.

The wear rate of laser-sprayed SS316L and Inconel 718 gradient walls is inversely proportional to hardness. Currently, niobium is used to modify stainless steel to achieve a good combination of thermal fatigue resistance, heat resistance and corrosion resistance, such as in automotive exhaust systems. It is well known that wear-resistant materials can be obtained by hardening soft phases with harder phases, and the presence of hard niobium carbides (NbC) and Fe2Nb in laser-deposited samples not only increases their hardness, but also increases wear resistance. This functional sorting method provides the means to design hardness and wear resistance according to the specific needs of the end user.

5. Conclusion

To be able to fabricate new 316L stainless steel and Inconel 718 stainless steel gradient structures using a diode laser deposition process, a series of parametric studies were conducted to study the effects of laser power and powder mass flow.

High-resolution optical and scanning electron microscopy, X-ray diffraction analysis, tensile tests, microhardness tests, and wear rate tests were used to analyze functionally graded walls made from these materials.

The following conclusions can be drawn from the obtained results:

• Functional classification of SS316L and Inconel 718 by laser deposition is a viable manufacturing method.

• The distance between the branches of the secondary dendrite (SDAS) strongly depends on the power and mass flow of the powder.

• The tensile strength of functional parts is inversely proportional to laser power and increases with powder mass flow.

• Formation of carbide-like NbC at higher ratios of inconel creates opportunities for selective control of hardness and wear resistance of functional materials.

• Gradient structures can be used in applications where the details of the loading conditions are known, where the microstructure and properties can be designed to best match the loading of each part of the component.

Source: Parametric Study of the Development of Functionally Graded Inconel Steel Materials by Direct Laser Metal Deposition, Materials & Design (1980-2015), doi.org/10.1016/j.matdes.2013.08.079

References: M. Koizumi, FGM Activities in Japan, Compos Part B: Eng, 28 (1997), pp. 1–4, W. Liu, J.N. DuPont, Fabrication of Functionally Graded TiC/Ti Composites Using Laser Engineering. , Scripta Mater, 48 (2003), p. 1337

Original work HJen Changjun of Jiangsu Laser Alliance!