Abstract: Additive metal manufacturing has undergone a revolution in recent years, being able to be incorporated in several industries such as aeronautics, automotive and even in medicine, allowing the manufacture of complex parts with fewer steps in the process, which translates in material savings and cost reduction. In this work, the corrosion of low carbon steel obtained by depositing consecutive layers is carried out, using electrochemical impedance spectroscopy and electrochemical noise immersed in a 0.1 M NaCl solution, establishing a comparison between the metal of contribution and deposited material. The layers of the material are characterized microstructurally and mechanically using scanning electron microscopy and Vickers microhardness. Overall, the results show a good response of the material to the action of the electrolyte after the immersion time, on the other hand, the microstructural results allow identifying the formation of 3 zones due to the cooling of the material. The microhardness of the steel does not show great changes between the zones, however, there is a slight increase in the intermediate zone due to the reduction in grain size. These studies allow researchers to know the behavior of these materials in applications that require contact with corrosive solutions of this nature.
Keywords:additive manufacturingadditive manufacturing,electrochemical impedanceelectrochemical impedance,electrochemical noiseelectrochemical noise,low carbon steellow carbon steel,scanning electron microscopyscanning electron microscopy.
Resumen: La manufactura aditiva de metales ha sufrido una revolución en los últimos años, pudiéndose incorporar en diversas industrias como la aeronáutica, automotriz e incluso en la medicina, permitiendo la fabricación de partes complejas con un menor número de pasos en el proceso, lo que se traduce en ahorro de material y reducción de costos. En este artículo se realiza el monitoreo del estado de corrosión de un acero de bajo carbono obtenido mediante la deposición de capas consecutivas, evaluándose a través de la técnica de impedancia electroquímica y ruido electroquímico, inmerso en una solución 0.1 M de NaCl, estableciendo una comparativa entre el metal de aporte y el material depositado. Las capas del material son caracterizadas microestructural y mecánicamente mediante microscopía electrónica de barrido y microdureza Vickers. De forma global, los resultados muestran una buena respuesta del material ante la acción del electrolito transcurrido el tiempo de inmersión, Por otro lado, los resultados microestructurales permiten identificar la formación de 3 zonas debido al enfriamiento en gradiente del material. La microdureza del acero no muestra grandes cambios entre las zonas, sin embargo, hay un ligero aumento en la zona intermedia debido a la reducción en el tamaño del grano. Estos estudios permiten a investigadores conocer el comportamiento de estos materiales en aplicaciones que demanden el contacto con soluciones corrosivas de esta naturaleza.
Palabras clave: acero de bajo carbono, impedancia electroquímica, manufactura aditiva, microscopía electrónica de barrido, ruido electroquímico.
Artículos
Evaluation of the corrosion resistance of an additive manufacturing steel using electrochemical techniques
Evaluación de la resistencia a la corrosión de un acero fabricado por manufactura aditiva mediante técnicas electroquímicas

Recepción: 20 Mayo 2020
Corregido: 10 Julio 2020
Aprobación: 21 Agosto 2020
Additive manufacturing (AM) is a process in which a part or machine element is constructed from a three- dimensional model with the help of CAD-CAM software, by adding successive layers. [1, 2, 3, 4, 5, 6]. Although these technologies are constantly evolving [7], there is still a limited state of the art on the effects that these materials can have under working conditions, especially corrosion. Authors such as Cervantes et al. [2], evaluated the mechanical performance of steels produced by additive manufacturing, Agredo et al. [8] analyzed the performance of these steels coated by electroless nickel plating in saline solutions, using electrochemical techniques. Zhang et al. [9] studied the mechanical and microstructural properties of stainless steel, finding significant variations of properties depending on the layers. Researchers like Liu et al. [10] evaluated the behavior of additively manufactured high-speed steel underwear and corrosion conditions.
It is important to know the performance of these steels subjected to severe environmental conditions, especially when they are in the presence of corrosive media such as industrial environments. For this reason, one of the main tools used is electrochemical techniques, which allow us to study the performance and predict what effects the materials may have when faced with the action of an electrolyte. Electrochemical impedance spectroscopy (EIS) is a technique that allows the study of the corrosion behavior of metallic materials. It is based on the application of a potential signal with magnitudes of the order of mV to the working electrode, a frequency scan is performed to obtain characteristic current values of the electrochemical behavior of the material under the established conditions. The impedance results obtained can be represented in terms of a real and imaginary component (Nyquist diagram), an impedance modulus and a phase angle (Bode diagram). These diagrams can be associated with electrical circuits that represent the electrochemical processes occurring in the system [11, 12, 13, 14, 15, 16].
Another newer and little-used technique is Electrochemical Noise (EN), which provides a detailed study of potential and current signal variations when the metal is in a corrosive medium. Malo and Uruchurtu [17], define noise as stochastic oscillations of potential and current. This allows to obtain information of the corrosion kinetics and to identify the corrosion mechanisms, through the interpretation by several methods, one of the most used is the time series, whose behavior allows to determine the types of corrosion that are appearing in the material. Another of the most implemented methods is the statistical one, where one of the most significant parameters is the standard deviation, for example, the decrease of the standard deviation of the potential is associated with the formation of a protective layer of oxides so it increases the resistance to corrosion [18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28].
An important measure in electrochemical noise is the resistance to noise Rn, which is related as the ratio of the standard deviation of the potential between the standard deviation of the current [29], this parameter allows tracking the corrosion as a function of time and relate it directly to the corrosion resistance of materials, equation 1 shows this correlation.
 (1)
 (1)A second measure to identify the type of corrosion is the location index (LI), Xia et al. [30] defined LI as the ratio of the standard deviation of the current to the RMS current, equation 2 shows the above.
 (2)
(2)This research studies the behavior of low carbon steel manufactured by additive manufacturing when faced with the action of a 0.1 M NaCl solution. The behavior of the filler metal is evaluated (ER70S-6 solid micro- wire, AWS A5.18/A5.18M standard). EIS and EN techniques are used to track the state of corrosion and monitor the system from 0 hours to 1320 hours of immersion
The steel manufactured by AM was obtained from a MIG-Router CNC machine at the Facultad de Ingeniería- Universidad Nacional Autónoma de México. For this purpose, a solid filament is used, designated AWS ER70S-6, which is kept at a distance between the torch and the surface of 8 mm, in addition to the following parameters: a current of 80 A, a voltage of 35 V with an argon flow of 0.4247 m3/h, an electrode advance speed of 15 m/min; This process is done by superimposing 5 layers of material leaving a solidification time of 60 s between every 2 layers, these variables are considered optimal, to generate material with an adequate combination of mechanical properties, according to studies carried out by Cervantes et al. [2]. Figure 1 shows an outline of the process.

The elemental composition of the materials was obtained by atomic emission spectroscopy. The microstructural characterization of the filler material (strand) and the additive manufacturing steel was carried out by scanning electron microscopy (SEM) in a Phillips XL-20 electron microscope at HV=20 kV, WD=8.3 mm, and complemented by optical microscopy, the areas formed in the material after the printing process are characterized. The metallographic preparation was done by conventional polishing techniques implementing the ASTM E3 standard [31], the development of the microstructure was executed by Nital immersion at 2%.
The hardness of the layers was evaluated using a digital microdurometer model HVS-1000, with a load of 100 g- 10 s, with a resolution of 0.1 HV. This characterization is carried out by taking measurements every 500 µm.
A Gill AC potentiostat controlled by the Sequencer software was used for electrochemical characterization. A 0.1 M de NaCl electrolyte was used and evaluated as a function of immersion time, using electrochemical impedance spectroscopy (EIS) and electrochemical noise (ENM) techniques, these were performed in a three- electrode electrochemical cell, composed by an SCE electrode as a reference, a graphite sheet as counter electrode and the material as working electrode, making a frequency sweep from 104 to 10-1 Hz, acquiring 10 points/decade, a sine wave with an amplitude of 10 mV was used as a perturbation. In electrochemical noise, time series of 1024 points every 0.5 s were obtained. Corrosion control was performed at 0, 240, 504, 1080 and 1320 h of immersion.
The elemental composition of the strand is shown in table 1. Table 2 shows the elemental composition of the material produced by AM, which is the product of three random measurements on the sample.

This composition does not show significant variation as supplied by the strand manufacturer [32] and is within the range of AWS A5.18/A5.18M, these materials are classified as low carbon steels with high Mn content.
Figure 2a shows the cross-section of the strand, here the high degree of plastic deformation in its microstructure is evident, which is the product of the wire drawing process applied to obtain this material. In figure 2b, a zoom is observed. This allows to identify in more detail the orientation of the microstructure, which is oriented in the longitudinal direction (in the direction of the deformation process).
The microstructural study of the printed material shows the identification of 3 important areas. Figures 3 to 5 show: a first zone composed of the first two layers, an intermediate zone composed of layers 3 and 4, and an upper zone composed of the last layer, each with significant microstructural changes, each layer has an average thickness of 3000 µm.
Figure 3a shows the microstructure by optical microscopy of zone 1, the light areas correspond to ferrite and the dark areas to pearlite. In these two first layers, when the material is deposited on the support base it undergoes a process of rapid solidification, which favours a dendritic type growth, which is observed in figure 3b (SEM image). It is important to note that inclusions may occur in the deposited material (first two layers, figure 3b, circles in red), however, this is typical of the manufacturing process and is studied in detail by Cervantes et al. in [2], these defects do not exceed 5% of the material by volume.


Zone 2, or intermediate zone, composed of the following two layers of material, shows a microstructure of ferrite and pearlite (Figure 4a)[2], [33]. In this zone a more homogeneous structure is evident with respect to zone 1, this homogeneity is associated to the fact that there are no thermal shocks, which generates a greater stability of the temperature obtaining a better distribution of the phases present, in figure 4b it can be observed at global level what was mentioned before. A zoom is made, which is shown in figure4c, here it is possible to identify in more detail the phases present in the matter, for this case, the darkest area shows the ferrite phase, while the lightest area shows the pearlite phase.

The zone 3, or upper zone of the steel, composed of the last layer is shown in figure 5. Figure 5a shows the right lateral zone of the layer, where the orientation of the grains is observed, this tendency is generated by the high solidification speed in the material, in figure 5b and 5c the left lateral and central zone of the layer is shown, again the formation of a dendritic type structure predominates [34], condition that is linked to the solidification process [33], this high solidification speed is linked to the contact that the material has with the environment, which favors a heat transfer by convection. These results are consistent with those reported in [2].

The microhardness results are shown in Figure 6, it is important to remember that each of the layers has an average thickness of the order of 3000 µm. A behavior that tends to be constant is exhibited, however, there is a slight change in the intermediate zone, this is due to the high homogeneity of the phases, which is linked to the different solidification speeds that make up the material, these results are consistent with those reported by Agredo et al. [8], with a deviation of less than 5% of the reported results. A comparison is established between the layers and the strand, evidencing a greater hardness in the latter, this is related to the severe plastic deformation that the material undergoes in its obtaining process. The steel printing process can be compared to a normalized heat treatment, when it is melted, deposited and slowly cooled to room temperature. In the investigation carried out by Franco and Paz [35], they reveal a hardness after normalized heat treatment of around 130 HV for a material with similar carbon content, in the case of the steel produced in this investigation, a hardness of the order of 150 HV.

Figure 7a shows the Nyquist diagram by EIS test for the strand, in Figure 7b, the behavior for the additive manufacturing steel and the strand is observed. In general it is evident the formation of a capacitive-resistive circuit for both materials, there is a clear difference of the printed steel, in this last one is observed an arc of greater diameter, this increase is linked with a greater resistance to the corrosion for the material deposited by layers [36- 37]. In both cases the materials show a reduction in the diameter of the arc as the immersion time passes, reflecting a progressive decrease in corrosion resistance

Figures 8 and 9 show the Bode diagram for the impedance module and the phase angle respectively. The impedance module displays a constant region at high frequencies, this is maintained for the strand and the AM steel, and at low frequencies the sum of the solution resistance and polarization resistance is obtained. However, the impedance experiences a considerable reduction after 1320 h of evaluation, this is due to the degradation of the material with the time of exposure, that is to say, the resistance to polarization decreases. From the phase angle diagram, Figure 9, the strand evidence the formation of only one time constant until 504 hours of evaluation, however, for 1080 and 1320 hours of immersion it is observed the appearance of 2- time constants one located at medium frequencies and another at low frequencies, aspect linked to the appearance of oxides on the surface, depending on the electrochemical behavior seen in the curves, the literature reports that the formation of oxides on the surface occurs as the appearance of a porous and poorly adherent layer, which does not limit the corrosion process to continue. [38, 39, 40, 41, 44]. The AM steel, shows the appearance of a single time constant located at medium frequencies for the first hours of evaluation, for 1320 h, a shift towards low frequencies is detailed [13,38,42]. Overall, these results indicate the degradation of the material (AM), due to the electrochemical phenomena that occur in the reaction of the electrolyte and the immersed material, a phenomenon given by the lack of corrosion products that promote a barrier effect on the surface of the material [37,38,43], however, AM steel shows greater resistance to corrosion than the strand.
In AM steel there is a small variation between 504 h and 240 h, achieving a very reduced increase in corrosion resistance, this is due to the formation of a small layer of protective oxides, however, the general behavior of steel is in decay, and the layer of oxides formed does not limit the corrosion process [8], [44].
The average calculation of the corrosion rate for AM steel is made, obtaining a value of 2.34±0.05 mm/year

Moreover, the results acquired by electrochemical noise were analyzed by means of statistical tools. Figure 10 shows the results of the location index calculation for filler metal (strand) and additive manufacturing steel. For the 0 hours of evaluation the filler metal and the additive manufacturing steel present a location index between 0.01 and 0.1, which indicates that mixed corrosion is being presented, after 120 hours of evaluation the strand presents a location index between 0.1 and 1, which identifies that pitting corrosion is occurring, this prevails until 1320 hours of evaluation. In the other hand, the AM steel at 120 hours of evaluation maintains a mixed corrosion state, already from 504 to 1320 hours this evolves to pitting corrosion [30].


The results of the calculation of the noise resistance (Rn) are shown in Figure 11. This parameter is directly related to the corrosion resistance of the materials, this tool has allowed to see that AM steel has a greater resistance to corrosion that the strand, however, again details the degradation of the material, product of the electrochemical phenomena that appear in the system [20,25-2628]. For 0 hours of immersion both materials show the highest noise resistance value, for the strand, it begins to decline until the end of the test, in the case of AM steel, the noise resistance value drops until 504 h and later evidence a slight increase, this is in accordance with the results presented in EIS.
The stable state of the standard deviation ratio between potential and current density shows that no significant corrosion reactions occur in the samples [26].

The microstructure of the additive manufacturing steel allows the identification of 3 zones, a more homogeneous structure towards the intermediate zone of the steel is evident, with a better distribution of the phases, this translates into a slight increase in the hardness of the material. A dendritic-type microstructure generated by the solidification process is shown for the upper zone of the material.
The corrosion study has determined that the additive manufacturing steel has a greater resistance to corrosion than the strand; however, the nonprotective corrosion products continue to cause the material to degrade as the immersion time passes.
The value of the corrosion rate is calculated from the results in electrochemical noise, obtaining an average value for the additive manufacturing steel of 2.34 mm/year.
The location index shows mixed corrosion for the first hours of evaluation of the 2 materials; however, this is transformed to pitting corrosion from 120 hours of immersion for the strand and from 504 hours for the steel by additive manufacturing, demonstrating that the steel by additive manufacturing presents greater resistance to corrosion.
From the corrosion study, it is considered that it is necessary, for this type of steel produced by the additive manufacturing technology, to think about coating it to increase its useful life.
Subsequent studies will study the corrosion products generated by the system.
How to cite: D. G. Agredo-Diaz, A. Barba-Pingarrón, N. Ortiz-Godoy, J. R. González-Parra, J. J. Olaya-Florez, J. J. Cervantes-Cabello, C. A. Ortiz-Otalora, “Evaluation of the corrosion resistance of an additive manufacturing steel using electrochemical techniques,” Rev. UIS Ing., vol. 19, no. 4, pp. 213-222, 2020, doi: https://doi.org/10.18273/revuin.v19n4-2020018
The authors would like to thank the PAPIIT IT101318 and PAPIME PE100218 Projects of the DGAPA UNAM for the support for this work.
To the Universidad Nacional de Colombia for the promotion of scientific research and the provision of its facilities, to PhD. Irma Angarita Moncaleano for her collaboration, encouragement and inspiration to work for science.
To the Universidad Pedagógica y Tecnológica de Colombia-UPTC and the research group GSEC, for their collaboration in the development of this project.











