3.1. Chemical Composition
The results of the chemical composition analyzed by GDS of the conventional and manufactured by WAAM 316L stainless steel are shown in Fig. 2. In terms of global chemical composition, from the data presented, it is possible to observe similarity between the values of the chemical elements in each of the samples, with a subtle elevation in the contents of Cr, Ni, Mo, Mn and Si in the WAAM specimen. Observing, in terms of chemical composition, an equivalence of the part deposited with the ER316LSi wire with the AISI 316L alloy produced by conventional process.
Furthermore, in both manufacturing conditions, the chemical compositions found are in accordance with the nominal chemical composition of the alloy, presented in Table 1. An exception, however, was observed in the chromium content of the conventional material, 3% lower than that the minimum quantity desired for 316L stainless steel. This difference, however, can be considered insignificant and may have occurred due to systematic errors inherent to the test performed to determine the chemical composition.
Among the elements present in the samples, Cr and Ni play an important role in the formation of the passivating layer characteristic of this alloy, which makes austenitic stainless steel more resistant to corrosion. They also act by increasing the mechanical strength of the alloy and increasing its hardness [23]. Furthermore, the addition of Mo in the material contributes to its repassivation, strengthening its passivating layer, reducing the propagation of pitting [24], [25]. There are still some papers, as presented by Botton [26] which indicates that the molybdenum act facilitating the formation of the passivation state of steel, since the presence of this element leads to the most positive pit potentials and lower values of critical current density and passivation potential. Thus, as they present similar chemical composition, the passivating layer of the deposited alloy and the conventional alloy must present similar behavior. However, its formation is a complex phenomenon and depends on a number of factors, in addition to chemical composition, such as pretreatment, metal surface composition, electrode potential, polarization time, chemical environment and temperature [27].
The chemical composition of the alloy also influences its solidification mode, thus causing changes in its mechanical properties. In the case of austenitic stainless steel, a higher concentration of ferritic elements such as chromium, molybdenum and silicon can promote the formation of ferrite as a primary phase, instead of austenite, which is present in the alloy as a secondary phase [23], [28]. Thus, a higher concentration of these elements, associated with the conditions of the WAAM process, can promote a greater amount of ferrite in the alloy, in addition to greater hardness and greater mechanical strength.
3.3. Microstructure
The microstructure of the AISI 316L alloy of the annealed pipe is presented in Fig. 4, observing the twins and polygonal grains of austenite (γ), whose size and shape vary, and can be measured from 16 µm to 50 µm. This is the equilibrium microstructure of austenitic stainless steels, being formed in slow cooling conditions [31]. Despite the identification of ferrite in the X-ray diffraction test, the micrographic images performed did not allow observing the presence of this phase in this material, instating that it may be present in low content in conventional alloy, as also observed by Rhouma et al. [32].
A macroscopic image of the 316L stainless steel deposited by WAAM is visualized in Fig. 5a. The fusion lines of the layers formed during deposition are evidenced by means of the yellow horizontal lines in the image (Fig. 5a). A microscopic image of the region between layers 1 and 2 (dashed lines in Fig. 5a) is presented in Fig. 5b. From this, it is verified that the microstructure in the transition region between the layers (highlighted zone between dashed lines as region 1–2) is different from the microstructure of layers 1 and 2. This behavior is due to the different cooling rate that each region is subjected to during the additive manufacturing process, since during deposition a part of the previously deposited layer ends up being partially resonated when a new layer is deposited [30]–[32]. Larger enlargements of the microstructure of layer 1 and the interlayer zone (1–2) are presented in Fig. 6a and Fig. 6b, respectively, allowing to observe clearly the difference between them and how this change occurs abruptly.
From the images presented in Fig. 6a and Fig. 6b, it is perceived, in addition to the austenitic matrix (lighter region), the presence of ferrite (darker region) in different morphologies. Its presence is favored by the rapid rates of heating and/or cooling of the order of 103 K/s that occur in the WAAM process [4], [33], [34]. Stainless steel under these conditions can solidify in four different ways, as shown in Table 3 and the prediction of this mode of solidification is made based on the Creq/Nieq ration. In type I, only the austenite phase is formed. In modes II and III there is formation of the ferrite and austenite phase, however in type II, austenite forms as the primary phase, and by means of a eutectic reaction, due to the segregation effect of ferritic elements, ferrite is formed, commonly in the centers of austenitic dendrites. In type III, ferrite is formed as the primary phase and austenite formation takes place at the ferrite/liquid interface, with ferrite solidification in the interdenticle spaces at the end of solidification, thus differentiating the microstructure obtained by this mode of solidification of mode II. On the other hand, type IV, ferrite is the only resulting phase, with austenite formation after solidification [35], [36].
Table 3
Mode of solidification and influence of the Creq/Nieq ratio following the solidification of stainless steels.
Solidification
|
Solidification mode
|
Mechanism
|
Creq/Nieq Ratio
|
Austenitic
|
I (A)
|
\(\text{L}\text{i}\text{q} \to \text{L}\text{í}\text{q}+ {\gamma } \to {\gamma }\)
|
< 1. 38
|
Austenitic-ferritic
|
II (AF)
|
\(\text{L}\text{i}\text{q} \to \text{L}\text{í}\text{q}+ {\gamma } \to \text{L}\text{í}\text{q}+ {\gamma }+{\delta } \to {\gamma }+{\delta }\)
|
1.38–1.50
|
Ferritic-austenitic
|
III (FA)
|
\(\text{L}\text{i}\text{q} \to \text{L}\text{í}\text{q}+{\delta } \to \text{L}\text{í}\text{q}+{\delta } + {\gamma }\to {\delta } + {\gamma }\)
|
1.50- 2.00
|
Ferritic
|
IV (A)
|
\(\text{L}\text{i}\text{q} \to \text{L}\text{í}\text{q}+ {\delta } \to {\delta }\)
|
> 2. 00
|
For the 316L stainless steel sample produced by WAAM, using Eq. 1 and Eq. 2, the Creq/Nieq value obtained is 1.74, which corresponds to solidification mode III (FA). In processes such as welding, this type of solidification is desirable for this alloy, since it can give the material greater resistance to cracking and traction [36], [37]. However, the presence of ferrite may decrease the ductility of the alloy [29], [38]. The morphology of ferrite in this mode of solidification, although difficult to be accurately predicted, is commonly vermicular or lathy [34], [39]. These forms were identified in some regions of the analyzed material, deposited by WAAM, as can be observed in Fig. 6a and Fig. 6b. The identification of these morphologies was performed by visual analysis, comparing the images made with those available in the literature for 316L stainless steel manufactured by WAAM in studies such as those of Chen et al. [38], Belotti et al. [40] and Wu et al. [11]. However, other ferrite morphologies, such as columnar and globular, were also identified in Fig. 6b. Wang et al. [41], report that despite the prediction of solidification mode III for 316L stainless steel manufactured by WAAM, the refunding zone of the deposited layers presents a higher temperature gradient and higher cooling rate, which can modify the solidification mode, from type FA to AF. Similar behavior was also reported by Belotti et al. [40] who observed the vermicular and lathy morphologies predominantly along the 316L stainless steel part deposited by WAAM, however, at the fusion interface, columnar and globular ferrite structures are perceived.
Thus, as the solidification process of deposited parts begins in the farthest region of the molten pool, and as the region higher than the weld bead cools transfers heat towards the already crystallized region, the grains that are formed first remain longer in contact with high temperatures, favoring the growth of these temperatures and allowing part of the ferrite to decompose into austenite [42], [43]. Thus, the region farther from the melting zone tends to present more uniform microstructure, greater grain size and lower concentration of ferrite. As can be seen in Fig. 7a and Fig. 7b, which show the spacing between ferrite grains, respectively layer 1 and interlayer 1–2, 316L stainless steel deposited by WAAM. It is then noticed that the region of layer 1 (farther from the melting zone) has between 7 µm and 14 µm, while for the microstructure in the fusion line (interlayer 1–2) the spacing between ferrite dendrites can reach 9 µm and have an even smaller size when the ferrite precipitates into globular morphology (4 µm). In addition, it is noted that the distance of the dendrites in 316L stainless steel manufactured via WAAM is less than the size of the polygonal grains of the conventional sample. The different morphologies and variations in grain spacing influence the mechanical behavior of 316L stainless steel produced by WAAM. Since the spacing of the primary cells is one of the parameters that has a strong relationship with the characteristics of strength and hardness in the analyzed alloy, as will be discussed later [44].
Also based on the images of Figure 7a and Figure 7b, an estimate of the ferrite content present in each region was performed, with the aid of the ImageJ software. The data found indicate 9% of this phase in the layer 1 region, and 10% in the interlayer zone 1-2. The values obtained are in accordance with the expected for fused austenitic stainless steels, which according to Pessanha [45] who present between 5% and 20% ferrite. Although there are few studies in the literature in which the authors calculate the ferrite fraction present in the 316L alloy produced by WAAM, Chen et al. (2018) [38][39] and Wen et al. (2020) [46] obtained, respectively, 7% and 17% of this phase in preforms with chemical composition equivalent to the alloy in question and using the WAAM technique. Although distinct, these values are within the range presented by Pessanha [45] and allow us to conclude that despite the difficulty in predicting the microstructure and the contents of the phases present in various metal alloys produced by WAAM, as reported by Örnek (2018) [16], from a correct selection of the metal wire and the parameters of the WAAM process, one can obtain a component with ferritic -austenitic structure with a ferrite content within the expected range for the molten material.
The chemical microcomposition analysis by EDS of the alloy studied in the austenitic matrix region and in ferrite grains was performed at the points highlighted in Figure 8a (Conventional), Figure 8b (WAAM layer 1) and Figure 8c (WAAM interlayer1-2). In the analysis of the material manufactured by WAAM point 1 corresponds to the austenite phase and points 2 and 3 to ferrite, already for the conventional material, the three points refer to the austenite phase. The values found are presented in Figure 9 and allow us to observe that the variation of chemical composition in the conventional sample is small, compared to the WAAM component, since only one phase was identified in the annealed material. In the case of the region of layer 1 and interlayer 1-2, in point 1 (austenite) higher concentrations of austenitizing elements (Ni and Mn) were identified and in points 2 and 3 higher concentrations of ferritizing elements (Cr, Mo and Si). Among the components observed, Cr is the main responsible for the formation of the passivator layer and Mo has great relevance in strengthening this layer. Thus, these microregions with smaller molybdenum compositions may be more susceptible to the rupture of this protective film, favoring the occurrence of some forms of localized corrosion, such pitting.