Influence of Selective Laser Melting Process and Heat Treatment Parameters on the Corrosion Resistance of 17-4 Precipitation Hardening Stainless Steel

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

 

1.      Improve clarity and english language within the abstract

2.      Reference is missing. Taguchi experiment table with the orthogonal array setting L9 (3ˆ4) was used for the 76 experimental design to limit the number of experiments.

3.      What is ANOVA? This needs to be clarified and explained

4.      What is Minitab software? This needs to be clarified and explained

5.      I highly doubt this! You need to specify. Why does heat treatment reduce the residual stress. The temperature seems to be too high: The fabricated samples were then heat treated, before being cut from the plate, at 103 650° C for one hour to reduce residual stresses [19] …. (The entire first paragraph of the results and discussion belongs to the introduction. These are no new results but standard.

6.      The section needs to be completely rearranged (The presence of manufacturing defects, like porosity, in SLM-made samples can af-130 fect not only the mechanical properties but also their susceptibility to localized corrosion 131 attack [23]. Furthermore, these defects reduce their ability to repassivate when exposed to 132 corrosive environments [24]. Micrographs of the samples obtained before and after poten-133 tiodynamic polarization are presented in Figure 1. Optical microscopy observations 134 showed a significant number of pits on the surface of the sample after being subjected to 135 polarization tests, which confirmed that the passivity breakdown was associated with pit-136 ting corrosion [15]. Moreover, it can be observed that some samples exhibit pre-existing 137 small pores before corrosion. For instance, sample 3 (Figure 1c) presents a higher level of 138 porosity before corrosion due to the insufficient energy density (47.22 Jmm−3). However, 139 as the energy density is increased to more than 68 Jmm−3, the pores are basically absent 140 (Figure 1d, Figure 1g, and Figure 1h) in comparison to samples printed with lower energy 141 density between 53 and 55 Jmm−3, as depicted in Figure 1b and Figure 1f, where the pores 142 are marked with yellow circles. The presence of a high porosity level leads to a decrease 143 in corrosion resistance. Suryawanshi, et al. [25] confirmed that a higher porosity level de-144 grades the corrosion resistance in chloride-rich environments due to the penetration of 145 chloride ions into pores and other defects, which destroys the passive film.)

7.      Figure 1 is not clear. What do the authors want to express? How did they decide upon the significance.

8.      This finding is not clear: The plots show otherwise. It can be noticed that raising the aging time to 150 min leads to a reduction in icorr, 268 suggesting an improvement in corrosion resistance. Conversely, elevating the aging time 269 to 240 min results in an increase of icorr, signifying a reduction in corrosion resistance.

9.      The SLM process and its specific microstructure is not addressed. This is necessary because it may not be compared to bulk material. Properties differ throughout components due to melt pool lines, epitactical growth, thermal gradients within components, microstructural gradients, phase precipitation and many more. All of the above affect corrosion initiation, corrosion procedure, passivation and (local) degradation. So far this has not been addressed at all.

10.  A comparison to bulk material is mission. The title promises otherwise.

11.  The entire results and discussion section is more a description of the analytical measurements. It does not explain the corrosion resistance, corrosion mechanism, influence of parameters or connects to previous findings regarding initial corrosion, difference of mechanical impact and corrosion, passivation and corrosion protection. All of these need to be addressed to round of the findings and make the paper worthwhile reading.

 

Author Response

1. Improve clarity and english language within the abstract.

Page 1

Selective laser melting (SLM) is an advanced additive manufacturing technique that enables the fabrication of complex metal components with high precision. However, inadequate parameter optimization can lead to defects that compromise the corrosion resistance of fabricated parts. Therefore, optimizing both SLM and heat treatment parameters is essential for enhancing electrochemical properties. The present work aims to determine the effect of the SLM process and heat treatment parameters on the corrosion resistance of SLM-made 17-4 PH stainless steel. A set of SLM and heat treatment parameters (laser power, scanning speed, aging time, and aging temperature) was determined by employed Taguchi method and a set of cyclic potentiodynamic polarization and electrochemical impedance experiments was performed in 3.5 wt% NaCl solution to generate corrosion data. The Taguchi method and statistical analysis of variance reveal the effect of laser power, scanning speed, aging time, and aging temperature on corrosion current density and passive film resistance of the SLM-made 17-4 PH samples. Laser power and aging temperature had the most significant effects, with lower laser power and higher aging temperature leading to decreased corrosion resistance, as indicated by higher corrosion current density and lower passive film resistance. Additionally, this study proposes empirical predictive models to estimate the electrochemical properties of SLM-made 17-4 PH stainless steel.

2. Reference is missing. Taguchi experiment table with the orthogonal array setting L9 (3ˆ4) was used for the 76 experimental design to limit the number of experiments.

Page 2

Agree, the reference has been added: 

• Taguchi experiment table with the orthogonal array setting L₉ (3ˆ4) was used for the experimental design to limit the number of experiments [18]. 
• [18] Mason, R.L., R.F. Gunst, and J.L. Hess, Statistical design and analysis of experiments: with applications to engineering and science. 2003: John Wiley & Sons.

3. What is ANOVA? This needs to be clarified and explained

Page 2 & 3

Agree, the following paragraph has been added:

• An ANOVA, a statistical method, was used to evaluate the significance of each process parameter (P, V, A, T) in terms of percent contribution to corrosion resistance. This method determines the effect of each factor by analyzing variance in the output variable calculating associated p-values, which indicate the probability of observing the obtained results under the null hypothesis, assuming the factor has no effect. A p-value less than 0.05 signifies that the respective factor has a statistically significant influence on the output variable [38].

4. What is Minitab software? This needs to be clarified and explained

Page 3

• The results were analyzed using Minitab, a statistical software designed for Design of Experiments (DOE). Minitab provides tools for generating ANOVA tables and regression equations. The stepwise method was employed, excluding non-significant terms after each iteration, without relying on a hierarchical model, to minimize errors and computation time. 

5. I highly doubt this! You need to specify. Why does heat treatment reduce the residual stress. The temperature seems to be too high: The fabricated samples were then heat treated, before being cut from the plate, at 103 650° C for one hour to reduce residual stresses [19] …. (The entire first paragraph of the results and discussion belongs to the introduction. These are no new results but standard.

Page 3

Agree, the following paragraph has been added, and the first paragraph of the results and discussion has been removed:

• Distortion and residual stresses are significant challenges in AM, negatively affecting dimensional precision and part performance. Distortion refers to the deviation of the AM-built part from its intended design or dimensions, which can occur during the additive process or when the part is separated from the substrate [40]. As a result , post-processing operations, such as stress relief, are currently necessary to address these effects [36]. Heat treatments are among the post-processing operations to control microstructures, phases transformations, and mitigate residual stresses.

6. The section needs to be completely rearranged (The presence of manufacturing defects, like porosity, in SLM-made samples can af-130 fect not only the mechanical properties but also their susceptibility to localized corrosion 131 attack [23]. Furthermore, these defects reduce their ability to repassivate when exposed to 132 corrosive environments [24]. Micrographs of the samples obtained before and after poten-133 tiodynamic polarization are presented in Figure 1. Optical microscopy observations 134 showed a significant number of pits on the surface of the sample after being subjected to 135 polarization tests, which confirmed that the passivity breakdown was associated with pit-136 ting corrosion [15]. Moreover, it can be observed that some samples exhibit pre-existing 137 small pores before corrosion. For instance, sample 3 (Figure 1c) presents a higher level of 138 porosity before corrosion due to the insufficient energy density (47.22 Jmm−3). However, 139 as the energy density is increased to more than 68 Jmm−3, the pores are basically absent 140 (Figure 1d, Figure 1g, and Figure 1h) in comparison to samples printed with lower energy 141 density between 53 and 55 Jmm−3, as depicted in Figure 1b and Figure 1f, where the pores 142 are marked with yellow circles. The presence of a high porosity level leads to a decrease 143 in corrosion resistance. Suryawanshi, et al. [25] confirmed that a higher porosity level de-144 grades the corrosion resistance in chloride-rich environments due to the penetration of 145 chloride ions into pores and other defects, which destroys the passive film.)

Page 4

Agree, the following paragraph has been rearranged:

• The presence of porosity in SLM-fabricated samples significantly influences their susceptibility to localized corrosion and impairs their ability to repassivate in corrosive environments [47, 48]. To investigate the correlation between porosity and corrosion behavior, micrographs of the samples before and after potentiodynamic polarization testing were analyzed and presented in Figure 1. Optical microscopy observations showed a significant number of pits on the surface of the sample after being subjected to polarization tests, which confirmed that the passivity breakdown was associated with pitting corrosion [15]. Moreover, it can be observed that some samples exhibit pre-existing small pores before corrosion.
• For instance, sample 3 (Figure 1c) presents a higher level of porosity before corrosion due to the insufficient energy density (47.22 J×mm^-3). However, as the energy density is increased to more than 68 J×mm^-3, the pores are basically absent (Figure 1d, Figure 1g, and Figure 1h) in comparison to samples printed with lower energy density between 53 and 55 J×mm^-3, as depicted in Figure 1b and Figure 1f, where the pores are marked with yellow circles. The presence of a high porosity level leads to a decrease in corrosion resistance. Suryawanshi, et al. [49] confirmed that a higher porosity level degrades the corrosion resistance in chloride-rich environments due to the penetration of chloride ions into pores and other defects, which destroys the passive film.

7. Figure 1 is not clear. What do the authors want to express? How did they decide upon the significance.

Page 5

• We aim to demonstrate the effect of energy density on porosity and show that increased porosity significantly accelerates the corrosion of the samples.

8. This finding is not clear: The plots show otherwise. It can be noticed that raising the aging time to 150 min leads to a reduction in icorr, 268 suggesting an improvement in corrosion resistance. Conversely, elevating the aging time 269 to 240 min results in an increase of icorr, signifying a reduction in corrosion resistance.

• Disagree, the figure 4a shows that raising the aging time (A) to 150 min leads to a reduction in icorr and elevating the aging time (A) to 240 min results in an increase of icorr.

9. The SLM process and its specific microstructure is not addressed. This is necessary because it may not be compared to bulk material. Properties differ throughout components due to melt pool lines, epitactical growth, thermal gradients within components, microstructural gradients, phase precipitation and many more. All of the above affect corrosion initiation, corrosion procedure, passivation and (local) degradation. So far this has not been addressed at all.

• The primary objective of this study was to quantify corrosion resistance of the SLM-made 17-4PH material and develop predictive models of the corrosion behavior. While we agree that a detailed microstructural analysis would be valuable, it would be more suitable for a future study.

10. A comparison to bulk material is mission. The title promises otherwise.

• Disagree. The comparison to bulk material was only part of the discussion section. In the abstract, the content was reorganized, and the sentence comparing it to bulk material was removed.

11. The entire results and discussion section is more a description of the analytical measurements. It does not explain the corrosion resistance, corrosion mechanism, influence of parameters or connects to previous findings regarding initial corrosion, difference of mechanical impact and corrosion, passivation and corrosion protection. All of these need to be addressed to round of the findings and make the paper worthwhile reading.

• As mentioned in notice 10's response, the objective of this study was to quantify corrosion resistance and develop predictive models. The methodology involved investigating a large number of parameters, with tests conducted for 9 sets of parameters, each replicated 3 times, for a total of 27 tests.

Reviewer 2 Report

Comments and Suggestions for Authors

The paper investigates the effects of selective laser melting (SLM) process and heat treatment parameters on the corrosion resistance of 17-4 precipitation hardening stainless steel, utilizing the Taguchi method and statistical analysis of variance. Key findings indicate that increasing laser power enhances corrosion resistance by reducing corrosion current density (icorr) and increasing passive film resistance (Rf), while higher aging temperatures negatively impact these properties. The study identifies optimal conditions for corrosion resistance, including a laser power of 220 W, scanning speeds between 800 and 900 mm/s, aging times of 60 to 150 minutes, and an aging temperature of 480 °C. The research contributes to a better understanding of how SLM parameters influence the microstructure and corrosion behavior of stainless steel, providing valuable insights for improving material performance in corrosive environments. Some revisions are recommended.

 

1. Can you elaborate on the specific experimental setup used for the cyclic potentiodynamic polarization and electrochemical impedance tests? Understanding the precise conditions under which these experiments were conducted, including the configuration of the electrochemical cell, the type of reference electrode used, and the temperature control during testing, is crucial for evaluating the reliability and reproducibility of the corrosion resistance data presented in your study.

 

2. What criteria were used to select the specific levels of laser power, scanning speed, aging time, and aging temperature in your Taguchi experimental design?

 

3. How did you ensure the homogeneity of the samples produced under different SLM parameters, and what measures were taken to minimize variability in the microstructure? Addressing the consistency of the microstructure across different samples is essential for validating the findings.

 

4. Could you provide more details on the statistical analysis performed, particularly regarding the ANOVA results? It would be helpful to understand how the significance of each factor was determined, including the thresholds for p-values and the interpretation of the percentage contributions of each parameter to the overall variation in corrosion resistance. This information is vital for assessing the robustness of your conclusions.

 

5. What were the specific mechanisms identified that explain the observed improvements in corrosion resistance with increased laser power? A deeper exploration of the microstructural changes that occur with varying laser power, such as changes in phase composition or grain size, would strengthen your claims.

 

6. How did the aging process affect the distribution and precipitation of Cu-rich particles within the matrix, and what implications does this have for corrosion resistance?

 

7. Some references may help to understand the underlying mechanism of heat treatment on the corrosion behavior of the 3D-printed alloys.

Author Response

1. Can you elaborate on the specific experimental setup used for the cyclic potentiodynamic polarization and electrochemical impedance tests? Understanding the precise conditions under which these experiments were conducted, including the configuration of the electrochemical cell, the type of reference electrode used, and the temperature control during testing, is crucial for evaluating the reliability and reproducibility of the corrosion resistance data presented in your study.

Page 3

Disagree, the experimental setup was mentioned:

• The corrosion behavior of the SLM-made 17-4 PH samples was evaluated by conducting three electrochemical tests: open circuit potential (OCP), electrochemical impedance spectroscopy (EIS), and cyclic potentiodynamic polarization (CPP), using a potentiostat (CH Instruments model 760E, Austin, TX, USA). All tests were performed in 3.5 wt% NaCl solution prepared with 18.2 MΩ deionized water, maintained at room temperature 20 ± 1 °C, in a three-electrode cell setup with the 17-4 PH sample, saturated calomel electrode (SCE), and graphite rod as working, reference, and counter electrodes, respectively. Before each test, the sample surface was polished with SiC papers to 800 grit, then polished with 9 and 3 µm diamond suspension, and further polished with 0.05 µm colloidal silica suspensions, rinsed and cleaned with water and ethanol, and then dried immediately. The OCP tests were conducted for 1 h, subsequently followed by the AC mode EIS tests at a frequency range of 10⁵ to 10^-2 Hz, and an AC amplitude of 10 mV from the E_OCP. The CPP test was then conducted, after 1 h at OCP, with a scan rate of 1 mV×s^-1 from -0.6 to 0 V_SCE. All tests were repeated three times to ensure reproducibility and reliability. Test data were collected and analyzed by using the CH Instruments and EC-Lab software (BioLogic, Seyssinet-Pariset, France). After being subjected to the electrochemical test, the corroded samples were rinsed, dried with alcohol, and observed by using optical microscopy LECO 300 to reveal the corrosion morphology.

2. What criteria were used to select the specific levels of laser power, scanning speed, aging time, and aging temperature in your Taguchi experimental design?

Page 2

 Agree, the following references has been added:

• The levels were chosen based on prior studies [16, 19-37].

3. How did you ensure the homogeneity of the samples produced under different SLM parameters, and what measures were taken to minimize variability in the microstructure? Addressing the consistency of the microstructure across different samples is essential for validating the findings.

• The primary objective of this study was to quantify corrosion resistance of the SLM-made 17-4PH material and develop predictive models of the corrosion behavior. While we agree that a detailed microstructural analysis would be valuable, it would be more suitable for a future study.

4. Could you provide more details on the statistical analysis performed, particularly regarding the ANOVA results? It would be helpful to understand how the significance of each factor was determined, including the thresholds for p-values and the interpretation of the percentage contributions of each parameter to the overall variation in corrosion resistance. This information is vital for assessing the robustness of your conclusions.

Page 8

Agree, the following paragraph has been rearranged:

• It can be observed that the corrosion current density is influenced by P and V with a total contribution of 67.64%, the influence of other factors is relatively small. For example, A and T, though statistically significant, contribute only 5.26% and 13.50%, respectively, to the overall variation. The quadratic effect of aging temperature (T²), with a p-value of 0.081, is insignificant, meaning it has little effect on the corrosion current density. Similarly, the error term contributes only 0.02%, further indicating that the model fits well.

  On the other hand, the passive film resistance is influenced mainly by the aging time (A), with a percentage of 65.92% and a significant p-value of 0.006 < 0.05, and the speed (V), with a percentage of 19.79%, as well as by the other parameters with a percentage of 14.29%. The remaining parameters, which include factors like P, T, and their interactions, contribute a smaller percentage of 14.29% to the overall variation in passive film resistance. This suggests that aging time and laser speed play a dominant role in determining R_f, while the other factors have a lesser statistically significant impact.

5. What were the specific mechanisms identified that explain the observed improvements in corrosion resistance with increased laser power? A deeper exploration of the microstructural changes that occur with varying laser power, such as changes in phase composition or grain size, would strengthen your claims.

Page 11

Agree, the following paragraph has been rearranged:

• It can be observed that increasing laser power results in a decrease in i_corr and an increase in R_f, which is beneficial for corrosion resistance. Irrinki, et al. [23] have also confirmed that increasing the energy density by increasing laser power from 150 W to 195 W improves the material density and corrosion resistance of SLM-made 17-4 PH when exposed to a 0.5 M NaCl environment. They emphasized that porosity causes the stagnation of the sodium chloride solution within the part, leading to the breakdown of the passive layer and the initiation of pitting corrosion. Wang, et al. [30] reported that increasing the energy density, by increasing laser power to 230 W and decreasing scanning speed to 886 mm.s^-1, results in better corrosion resistance of SLM-made 15-5PH SS in a 0.1 M NaCl solution. This improvement was attributed to the absence of manufacturing defects (fractional porosity). They further highlighted that manufacturing defects have a complex influence on pitting corrosion resistance and the stability of the passivation film. Zhang, et al. [51] studied the effect of laser power on the corrosion behavior of LPBF-made 316L stainless steel, demonstrating that higher laser power refines the grain structure, reduces defects like lack of fusion pores, and improves pitting corrosion resistance. This was further supported by optical microscopy images (Figure 1), which show that higher laser power effectively eliminates pores.

6. How did the aging process affect the distribution and precipitation of Cu-rich particles within the matrix, and what implications does this have for corrosion resistance?

Page 11 & 12

Agree, the following paragraph has been rearranged:

• The high corrosion resistance of samples heat treated at 480 °C can be attributed to the precipitation of Cu-rich precipitates. Copper is commonly used as an alloying element to enhance the general corrosion resistance of stainless steels. It is enriched at the surface during the anodic dissolution of the alloy, leading to a reduction in its corrosion rate [36]. The 17-4PH contains approximately 3-5 wt.% of copper. The Cu-rich precipitates are generated during specific post-process heat treatments, which can be tailored to change the precipitates size, shape, and hardness and strength of the post-processed components [52]. E. Oguzie, et al. [53] reported that the addition of Cu generally improved the corrosion resistance through suppression of the active dissolution. In a similar study by Pan, et al. [24], it was reported that the formation of reverted austenite, elimination of δ-ferrite and precipitation of Cu-rich particles at the grain boundaries contributed to an improvement in the pitting corrosion resistance of the heat-treated 17-4 PH samples.

7. Some references may help to understand the underlying mechanism of heat treatment on the corrosion behavior of the 3D-printed alloys.

Page 12

Agree, the following paragraph has been added:

• Heat treatment after AM process parameters can have a significant influence on the microstructure and corrosion behavior of precipitation-hardening SS. The heat treatment process results in the formation of a more protective passive film, which leads to improved corrosion and pitting resistance [48]. Stoudt confirmed that after homogenization, the AM samples could be more resistant to pitting than wrought SS17-4 in NaCl environment, and reported that homogenizing at 1150°C for 120 min followed by solutionizing at 1050°C for 30 min can eliminate the segregation and produce a uniform microstructure with a volume fraction of austenite of less than 0.1 [64]. Jiang et al. [65] showed that improper peak-aging treatment and phase segregation lead to increased hardness and susceptibility to high hydrogen embrittlement (HE) in 17-4PH SS, resulting in the failure of the valve stem 29 manufactured from this material and used in petrochemical pipeline. The increase in aging temperature and soaking time does not have a significant effect on various mechanical properties, but it does have a negative impact on corrosion properties. To ensure homogeneity in microstructure, it is advisable to perform solution annealing before aging. Corrosion was observed to be the least in samples subjected to the H900 aging heat treatment, while the highest corrosion levels were observed in the as-built SLM specimens. The reason for the higher corrosion rate at higher aging temperatures and longer soaking times is attributed to the formation of more Cr₂₃C₆ precipitates [56]. The presence of copper-rich precipitates (CRPs) is crucial for enhancing the material's strength. However, when the aging temperature surpasses 560°C, the structural changes and enlargement of CRPs can cause a swift decline in strength and hardness. Furthermore, the corrosion performance is also affected adversely [66].

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Accepted in the present state

Reviewer 2 Report

Comments and Suggestions for Authors

The authors have replies all my comments.