*Article* **The Influence of the Type of Electrolyte in the Modifying Solution on the Protective Properties of Vinyltrimethoysilane/Ethanol-Based Coatings Formed on Stainless Steel X20Cr13**

**Aleksandra Kucharczyk 1, Lidia Adamczyk 1,\* and Krzysztof Miecznikowski <sup>2</sup>**


**Abstract:** The paper reports the results of the examination of the protective properties of silane coatings based on vinyltrimethoxysilane (VTMS) and ethanol (EtOH), doped with the following electrolytes: acetic acid (AcOH), lithium perchlorate LiClO4, sulphuric acid (VI) H2SO4 and ammonia NH3. The coatings were deposited on stainless steel X20Cr13 by the sol–gel dip-coating method. The obtained VTMS/EtOH/Electrolyte coatings were characterized in terms of corrosion resistance, surface morphology and adhesion to the steel substrate. Corrosion tests were conducted in sulphate media acidified up to pH = 2 with and without chloride ions Cl−, respectively. The effectiveness of corrosion protection was determined using potentiometric curves. It has been demonstrated that the coatings under study slow down the processes of corrosion of the steel substrate, thus effectively protecting it against corrosion.

**Keywords:** vinyltrimethoxysilane; silane; corrosion; sol–gel

#### **1. Introduction**

The majority of metals in contact with atmospheric air (electrochemical corrosion) form a protective layer—an oxide (passive) film of their surface. The phenomenon of passivation provides a basis for the natural corrosion resistance of some metals and construction alloys, such as aluminium or stainless steels. However, it is not fully sufficient in more aggressive media, for example, in locations, where the metal is exposed to the action of chloride, bromide, or fluorine ions. The effects of corrosion processes are usually associated with additional, often considerable costs; therefore, various methods for protection against corrosion are used [1,2]. As mentioned above, the process of electrochemical corrosion proceeds in the environment of aggressive solutions of electrolytes. Therefore, this phenomenon affects the correct functioning of different types of current sources and materials (such as concrete, steel, etc.), which many people tend to forget.

Stainless steels, or iron alloys containing 12–18% chromium, have for many years enjoyed great popularity because of their corrosion resistance, availability and price. They have unique characteristics, thanks to which they find increasingly wide application in today's technology. Their corrosion resistance, mechanical or engineering properties enable them to be used in particularly demanding working environments [3]. The martensitic stainless steel X20Cr13 used in the present study, intended for toughening, exhibits high mechanical properties, such as strength, ductility and machinability, while retaining sufficient corrosion resistance. X20Cr13 does not show resistance to chlorine, salts and intercrystalline corrosion. Stainless steel is suitable for operating in the environments of water vapour,

**Citation:** Kucharczyk, A.;

Adamczyk, L.; Miecznikowski, K. The Influence of the Type of Electrolyte in the Modifying Solution on the Protective Properties of Vinyltrimethoysilane/Ethanol-Based Coatings Formed on Stainless Steel X20Cr13. *Materials* **2021**, *14*, 6209. https://doi.org/10.3390/ma14206209

Academic Editor: Vít Kˇrivý

Received: 20 September 2021 Accepted: 15 October 2021 Published: 19 October 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

low concentrated inorganic acids, solvents and pure water [4]. The corrosion resistance of stainless steels can be enhanced by introducing to them alloy additions, such as nickel or molybdenum, but also by forming protective coats on their surface [5,6].

The modification of metal surface belongs to corrosion prevention methods. Apart from classical metallic, inorganic, or paint coatings, also coatings of organosilicon compounds (siloxane compounds) obtained from modifying solution, attract a lot of attention. These are coatings composed of siloxane bonds formed as a result of the reactions of hydrolysis and condensation. One of the simpler and the most common methods of depositing silane-based coatings is the sol–gel method [7,8]. Silane coatings are used especially for protecting poorly passivating metals, such as Al, Fe, Zn, Mg, Ti and stainless steel [3,9].

Silanes are compounds characterized by low toxicity and, most importantly, offer protection against corrosion and good adhesion to a substrate (e.g., steel). During the deposition of silane-based protective coatings, strong covalent bonds form [5,6]. It has also been shown that organosilanes provide effective corrosion protection of materials, such as aluminium and steel [10,11]. Silanes are used in corrosion protection chiefly as interlayers as they provide the adhesion of coatings to metals [12–16].

The protective properties of silane coatings (structure, stability in time, tightness, corrosion resistance) are related to the parameters of the silane solution (silane type, composition, concentration, pH), as well as with the method of application and the process of drying of the deposited coating at a specific temperature. In many publications [17–20] the authors, prior to depositing a sol–gel coating, used pre-treatment of the steel surface with acids with the aim of enhancing the adhesion and anticorrosive properties. Coatings based on organosilicon compounds are deposited on the surface of substrate elements, usually from modifying solutions being sol–gel solutions [21–32]. In the majority of cases, acetic acid and ammonia were included in the composition of modifying solutions [26–32].

The precursor of the reaction of synthesis in the sol–gel method are various alcoholates of metals, salts or nitrates. After immersing the metal in a diluted silane solution, particles are adsorbed on the metal surface through hydrogen bonds. The key reactions are hydrolysis and condensation, whereby a compact protective coating forms at the silane/metal interface. The hydrolysis and condensation (polycondensation) reactions occur simultaneously within the whole volume of the solution. The properties of the end product and the rate of the process are strongly influenced by, e.g., the R≡H2O: silane mole ratio, medium pH, solvent type, the nature and concentration of catalysts, and temperature; for example, individual stages of the sol–gel process run faster when an appropriate (acidic or basic) catalyst is used [33–35].

The present study is devoted to the structural examination and corrosion testing of coatings composed of vinyltrimethoxysilane (VTMS), ethanol and electrolytes, deposited on stainless steel X20Cr13. Within the study, the effect of the addition of an acidic and a basic electrolyte on the structural properties and corrosion protection of the investigated stainless steel was examined. The aim of the investigation was to obtain vinyltrimethoxysilanebased coatings by the dip-coating method of the best possible physical, chemical and anticorrosive protection, which could be used for the corrosion protection of applied metals and their alloys (Fe, Al, Zn, Cu and Cu).

Over a dozen or so years, many papers on the protection of metal surface with silanes have been published; it should be emphasized, however, that those publications did not address the effect of modifiers, i.e., electrolytes of varying pH values, on the process of protecting metals covered with silane coatings against corrosion.

#### **2. Materials and Methods**

Analytically pure reagents and deionized water were used in experiments. Sol– gel solutions were prepared by mixing vinyltrimethoxysilane (VTMS) of the molecular formula CH2=CHSi(OC2H5)3 (supplied by Sigma Aldrich), anhydrous ethyl alcohol EtOH (supplied by Sigma Aldrich) and electrolyte (acidic and basic, respectively). The volumetric VTMS:EtOH:Electrolyte ratio of the obtained coating was 4.84:2.16:3.0.

The following electrolytes were selected for testing:


#### *2.1. The Influence of the Reaction Environment on the Sol–Gel Process*

Four (4) electrolytes (acetic acid CH3COOH (AcOH), lithium perchlorate LiClO4, sulphuric acid (VI) H2SO4 and ammonia NH3) were chosen for the sol–gel process on account of the hydrolysis reaction: acidic and basic hydrolysis.

Table 1 shows the effect of the medium (electrolyte) on individual stages of the sol–gel process. Table 2 gives examples of substances that speed up this process.



**Table 2.** Substances accelerating the sol–gel process.


The sol–gel process can be run by two methods. The first method is one-stage basic or acidic catalysis. In the case of basic catalysis, hydrolysis proceeds with the participation of the hydroxyl anion (OH−). This ion reacts directly with the silicon atom, leading to the formation of silanol and the RO− group. It enables semi-transparent gels of high porosity to be obtained. In the case of acidic catalysis, on the other hand, the hydrolysis process is initiated by acid. Protons H<sup>+</sup> react with the oxygen atoms bonded with the silicon atom in the –OR or OH group. This causes the electron cloud to shift towards the oxygen atom in the Si-O bond. As a consequence, this results in an increase in positive charge on the silicon atom. The water molecule combines with the silicon atom, which is followed by disintegration and the formation of silanols and alcohol. In acidic catalysis, we obtain transparent gels of low porosity [50].

For technological reasons, the sol–gel process is most favourably carried out by using the second of the above-mentioned methods, i.e., two-stage acidic-basic catalysis. This shortens the time necessary for obtaining the gel. The method comprises the first stage—hydrolysis at pH < 7, the second stage—raising the medium reaction up to a pH of approximately 7, whereby we slow down the hydrolysis process and accelerate the gelation process [51].

#### *2.2. Preparation of Test Material*

Stainless steel X20Cr13 of the following composition (in wt%): C-0.17; Cr-12.6; Si-0.34; Ni-0.25; Mn-0.30; V-0.04; P-0.024; and S < 0.005 was used for testing. Test samples were in

the shape of 5 mm-diameter cylinders. Their walls were isolated in polymethyl methacrylate frames using epoxy resin. The geometric working surface area of the samples was 0.196 cm2. Prior to experiments, the samples were each time mechanically polished on abrasive papers with a decreasing grit size up to grade 2000, and then rinsed with distilled water and ethyl alcohol. Before applying a coating, each sample was flushed with acetone to degrease its surface. On the prepared electrodes, coatings were deposited by the dipcoating method following the procedure developed in the paper [36]. The composition and the stirring parameters for the test coatings are shown in Table 3.


**Table 3.** Composition of coatings and mixing parameters.

The anticorrosive properties of the obtained coatings were assessed in two corrosion media: 0.5 mol dm−<sup>3</sup> Na2SO4 (pH = 2) and 0.5 mol dm−<sup>3</sup> Na2SO4 + 0.5 mol dm−<sup>3</sup> NaCl (pH = 2) using a potentiodynamic technique with the use of a scanning potential from— 0.8 V to 1.6 V at a polarization rate of 10 mVs<sup>−</sup>1. The values of all potentials were measured and expressed relative to the saturated calomel electrode.

The composition and surface appearance of the coatings deposited on the investigated steel were assessed using a JEOL JSM-6610 LV scanning electron microscope with an EDS-type X-ray microanalyzer (JEOL, Tokyo, Japan). Microstructural examination was carried out with a KEYENCE VHX 7000 digital microscope and an Olympus GX41 optical microscope (Keyence, Mechelen, Belgium). The surface roughness of the coatings was measured using an Hommel Tester T1000 profilometer (JENOPTIK Industrial Metrology, Jena, Germany). The microscopic surface maps were made using an AFM (Atomic Force Microscope) NanoScope V MultiMode 8 (BRUKER, Bremen, Germany). The characteristics of the coatings were determined using Attenuated Total Reflection Fourier Transform Infrared Spectroscopy (ATR- FTIR) Bruker Optics-Vertex 70 V (BRUKER, Bremen, Germany). The coatings thicknesses were measured using a DT-20 AN 120 157 meter (ANTICORR, Gda ´nsk, Poland). Electrochemical measurements were taken in the classical three-electrode system using a CHI 706 measuring station (CH Instruments, Austin, TX, USA). The working electrodes were steel X20Cr13 coated and uncoated as well as glassy carbon, an auxiliary electrode (platinum), and a reference electrode (the saturated calomel electrode, SCE).

The adhesion of the test coatings was assessed by a qualitative test using ScotchTM Tape (ScotchTM Brand, St. Paul, MI, USA).

#### *2.3. The Preparation of Samples for Structural Examination (Sample Cross-Section)*

Samples for structural examination (sample cross-section) were prepared in four steps: application of a coating on steel, cutting the sample through in the plane perpendicular to the surface, then the samples were embedded in epoxy resin with graphite. So embedded specimens were subjected to grounding and polishing to obtain a mirror surface.

#### *2.4. Corrosion Resistance Test in a Potassium Hexacyanoferrate (III) Solution (Ferroxyl Test)*

For performing a quick test to indicate the barrier nature of the proposed protective coatings, it was applied the well-known and widely used ferroxyl test (ferroxyl indicator) in a modified form. For this purpose, the tested plate was immersed in a solution containing potassium ferricyanide. In the presence of iron ions (formed in the process of corrosion in an acid medium), insoluble blue iron (II) hexacyanoferrate (III) (Prussian Blue) appears, indicating that the material dissolving in anodic locations. A 2-mmol dm−<sup>3</sup> potassium hexacyanoferrate (III) K3[Fe(CN)6] solution was utilized for testing. An X20Cr13 steel sample uncovered and covered with a VTMS/EtOH/AcOH coating in a VTMS concentration of 3.16 mol dm−<sup>3</sup> were immersed in potassium hexacyanoferrate (III) solution. Afterward, electrochemical measurements were taken by the cyclic voltammetry in the potential range from −0.6 V to 1.2 V vs. SCE [49,50].

#### **3. Results**

#### *3.1. Microstructural Observations and Chemical Analysis*

The topography of obtained coatings was assessed using a light microscope. Figures 1 and 2 show the morphology of the investigated coatings deposited on the X20Cr13 stainless steel surface. All of the four coating types uniformly cover the entire surface of the electrodes without any free structural spaces. A structure of the metallic substrate with polishing traces is visible in the photographs (Figure 1). As shown in Figure 1, the coatings morphology is compact, smooth, shiny and transparent over the entire sample surface.

**Figure 1.** Coatings topography: VTMS/EtOH/AcOH (**a**), VTMS/EtOH/LiClO4 (**b**), VTMS/EtOH/ H2SO4 (**c**), VTMS/EtOH/NH3 (**d**), (×200).

**Figure 2.** (**A**) Topography of VTMS coatings deposited on steel X20Cr13 (SEM Jeol JSM-6610 LV), (**B**) Cross-section of coatings deposited on steel X20Cr13 (digital microscope KEYANCE VHX 7000, ×200). Coatings: VTMS/EtOH/AcOH (**a**), VTMS/EtOH/LiClO4 (**b**), VTMS/EtOH/H2SO4 (**c**), VTMS/EtOH/NH3 (**d**).

Figure 2 represents the topography of VTMS/EtOH/AcOH, VTMS/EtOH/LiClO4, VTMS/EtOH/H2SO4, VTMS/EtOH/NH3 coatings, revealed using a scanning electron microscope. The obtained results confirm the previous observations: regardless of the added electrolyte, the morphology of VTMS coatings covers uniformly the sample surface, forming a compact, tight and homogeneous structure.

Figure 2B shows the cross-section of the obtained VTMS/EtOH/Electrolyte coatings deposited on the X20Cr13 stainless steel. The recorded profile indicates that the coatings are uniformly deposited on the steel surface. The structural observations confirm that the coatings are free from any cracking and their surface roughness is negligible, which is a huge asset from the point of view of the corrosion resistance of steel. The average coatings thickness (as measured in four locations on the sample) were the following for respective coatings: VTMS/EtOH/AcOH 11.4 μm (a); VTMS/EtOH/LiClO4 8.05 μm (b); VTMS/EtOH/H2SO4 8.65 μm (c); VTMS/EtOH/NH3 12.8 μm (d). Based on the crosssection, it was demonstrated that the modification of the silane solution with various electrolytes had a significant effect on the coating thickness.

#### 3.1.1. Chemical Composition of the VTMS/EtOH/Electrolyte Coatings

The chemical composition of the produced coatings was determined using an electron scanning microscope equipped with an EDS X-ray chemical analyzer. Based on the performed chemical analysis, the contents of silicon for respective coatings are as follows: VTMS/EtOH/AcOH 32.47%; VTMS/EtOH/LiClO4 29.05%; VTMS/EtOH/H2SO4 30.26%; VTMS/EtOH/NH3 25.87%. The rest was made up of the elements C and O.

#### 3.1.2. Testing for Adhesion to the Substrate

Immediately after depositing VTMS/EtOH/Electrolyte coatings, their adhesion to the X20Cr13 stainless steel substrate was tested using ScotchTM Tape. The produced coatings are characterized by good adhesion to the steel substrate.

#### 3.1.3. Surface Roughness of Obtained Coatings

The measurement of the surface roughness of the VTMS/EtOH/Electrolyte coatings deposited on the X20Cr13 stainless steel taken with a profilometer is represented in Figure 3. Both the topography and profile of the coatings confirm that the coatings are free from cracking, and their surface roughness is little (low Ra values). The testing results differ, depending on the electrolyte used; the closest Ra values can be observed for VTMS/EtOH/AcOH and VTMS/EtOH/NH3 coatings (Ra = 0.40–0.43 μm). Table 4 shows the results of the measurement of parameter Ra.

**Figure 3.** Roughness measurement Ra for coatings deposited on X20Cr13 steel: VTMS/EtOH/AcOH (**a**), VTMS/EtOH/ LiClO4 (**b**), VTMS/EtOH/H2SO4 (**c**), VTMS/EtOH/NH3 (**d**). Profilometer Hommel Tester T1000.


**Table 4.** Roughness parameter Ra for individual coatings deposited on steel X20Cr13.

The examination made using an AFM microscope confirms the previous findings that the addition of electrolyte has an effect on the coating surface roughness. The surface morphologies of VTMS/EtOH/Electrolyte coatings produced on the metal surface, with a varying electrolyte addition, are illustrated in Figure 4. The recorded values of parameter Ra for respective coatings are as follows: VTMS/EtOH/AcOH 0.381 μm; VTMS/EtOH/LiClO4 0.908 μm; VTMS/EtOH/H2SO4 1.45 μm; VTMS/EtOH/NH3 0.389 μm.

**Figure 4.** AFM images of the surface of coatings deposited od steel X20Cr13: VTMS/EtOH/AcOH (**a**), VTMS/EtOH/ LiClO4 (**b**), VTMS/EtOH/H2SO4 (**c**), VTMS/EtOH/NH3 (**d**). Pictures were taken using an AFM NanoScope V MultiMode 8 Bruker.

Protective coatings are generally porous layers; after some time, the surface of steel or metal will come into contact with an aggressive electrolyte solution, water, or oxygen molecules. A discontinuity in the coating may initiate pitting or crevice corrosion. Obtained protective coatings, as compared to the protected elements, are usually extremely thin.

#### 3.1.4. Thickness of Obtained Coatings

One of the key parameters influencing the corrosion resistance of elements is the thickness of their protective coatings. In the present study, this parameter has been analyzed using three examination methods. Based on profile examination (Figure 2B), the thickness of obtained coatings was analyzed. The thickness of each coating is the average of 4 measurements: VTMS/EtOH/AcOH 11.4 μm (a); VTMS/EtOH/LiClO4 8.05 μm (b); VTMS/EtOH/H2SO4 8.65 μm (c); VTMS/EtOH/NH3 12.8 μm (d).

The recorded thicknesses measured with a profilometer are given in Table 5.


**Table 5.** Thickness measurement results for individual coatings on steel X20Cr13.

To compare the thicknesses of the coatings, in addition to the methods described above, thickness measurements were taken using a DT-20 Testan meter with an integrated probe designed for measuring on ferro- and non-ferromagnetic substrates. A series of 10 measurements (at different locations on the sample) was done; Table 6 provides recorded thickness values for VTMS/EtOH/Electrolyte coatings. The obtained coatings thickness values are consistent with those produced with a digital microscope and a profilometer.



Based on the performed measurements using three instruments (a digital microscope, profilometer, and a thickness meter), the mean coatings thickness was determined (Table 7).

**Table 7.** The average thickness of the coatings calculated by measurements from three instruments (a digital microscope, profilometer, and a thickness meter).


The differences in coatings thickness between individual methods were negligible, which confirms the usefulness of the applied methods for coatings thickness measurement.

#### *3.2. Analysis of Coatings Composition*

The characteristics of vinyltrimethoxysilane- based coatings deposited on the X20Cr13 steel substrate were determined using Attenuated Total Reflection Fourier Transform Infrared Spectroscopy (ATR-FTIR). The ATR-FTIR spectra of VTMS/EtOH/Electrolyte coatings are shown respectively in Figure 5. Characteristic absorption peaks were observed for VTMS in the range of 4000–400 cm−1. The absorption bands observed at the values of 2953 cm−1, 2857 cm−<sup>1</sup> and 1290 cm−<sup>1</sup> correspond to the asymmetric tensile and bending vibrations of the C-H bond belonging to the –Si-(OCH3) group. Subsequent peaks were noted at the values of 1602 cm−<sup>1</sup> and 1410 cm<sup>−</sup>1, which correspond to the tensile vibrations of the C=C bond of the CH2=CH- group. The bands of values of 1000 cm−1, 883 cm−<sup>1</sup> and 750 cm−<sup>1</sup> correspond to the vibrations of Si-O-C. The wide band occurring at about 1100 cm−<sup>1</sup> corresponds to the asymmetric tensile vibrations of the Si-O-Si bond. The peak at the value of 704 cm−<sup>1</sup> corresponds to the Si-C bond. The wide absorption band at about 3400 cm−<sup>1</sup> was caused by the –OH groups. The peak at 964 cm−<sup>1</sup> was ascribed to

the asymmetric bending vibrations of the Si-OH bond, whereas the band observed at the wavelength of 3053 cm−<sup>1</sup> matches the vibrations in the alkene group.

**Figure 5.** ATR- FTIR spectra obtained for VTMS coatings with a concentration of 3.16 mol dm−<sup>3</sup> deposited on steel X20Cr13: VTMS/EtOH/AcOH (**A**), VTMS/EtOH/LiClO4 (**B**), VTMS/EtOH/ H2SO4 (**C**), VTMS/EtOH/NH3 (**D**).

#### *3.3. Electrochemical Analysis*

To make the assessment of the kinetic tendency to either general or pitting corrosion, measurements of open circuit potential (OCP) were taken for steel uncovered and covered with a VTMS coating, respectively, with the addition of electrolyte in the form of either: acetic acid (b), lithium perchlorate (c) sulphuric acid (VI) (d), or ammonia (e).

As shown by the open circuit potential measurements illustrated in Figure 6A,B, the steel not covered with a coating directly after being immersed in the test solutions exhibits a potential of approximately −0.4 V, which decreases for longer exposure times and takes on the value of corrosion potential for steel (−0.5 V). For steel X20Cr13 covered with the coatings: VTMS/EtOH/LiClO4, VTMS/EtOH/H2SO4 and VTMS/EtOH/NH3, Figure 6A, the OCP potential increases for the initial 24 h until reaching a value of 0.4 V. As can be seen from Figure 6A (line b), the steel covered with a VTMS/EtOH/AcOH coating directly after being immersed in the corrosion solution shows a potential of 0.2 V, which increases up to 0.4 V after 24 h exposure. The values of potential for all steels covered with coatings after prolonged immersion in the corrosion solution show potential from the passive range, so more positive than E*kor* (0.5 V).

The dependence of the open circuit potential of uncoated and coated steel on the time of holding in the chloride ion-containing corrosion solution is represented in Figure 6B. The uncoated X20Cr13 steel undergoes active dissolution after approximately 50 h of immersion in the corrosion solution. By contrast, the steel covered with VTMS-based coatings, upon immersion in the corrosion solution, exhibits a potential from the passive range. The potential of the steel covered with VTMS/EtOH/AcOH coatings increases, for the initial 24 h, up to a value of approximately 0.45 V and stays on this level for another 13.5 days; for VTMS/EtOH/H2SO4, the potential is −0.25 V and remains for 350 h; for VTMS/EtOH/NH3, after 150 h, it amounts to −0.35 V and holds on this level for subsequent 200 h; and for VTMS/EtOH/LiClO4, the potential stays at the level of 0.35 V for 240 h and then dramatically decreases to a value of 0.0 V.

**Figure 6.** Potential measurement in open circuit potential OCP from exposure time in solution: 0.5 mol dm−<sup>3</sup> Na2SO4 mol dm−<sup>3</sup> pH = 2 (**A**) and 0.5 mol dm−<sup>3</sup> Na2SO4 + 0.5 mol dm−<sup>3</sup> NaCl pH = 2 (**B**) for steel X20Cr13 uncovered (a) and covered with coatings VTMS/EtOH: CH3COOH (b), LiClO4 (c), H2SO4 (d), NH3 (e).

It is worth noting that the stationary potential value of the coated steel, despite the log time of exposure in the chloride ion-containing corrosion solution, is more positive than the stationary potential value of steel. Microscopic observations after the measurement did not reveal any local corrosion effects under the VTMS/EtOH/AcOH coating, which indicates significant substrate protection.

To establish the most effective influence of electrolytes on the anticorrosion properties of the produced VTMS silane coatings deposited on the X20Cr13 steel, the assessment of their capacity for inhibiting general and pitting corrosion was made using potentiodynamic curves. The experiment was conducted in two solutions:


**Figure 7.** Potentiodynamic polarization curves recorded in the solution: 0.5 mol dm−<sup>3</sup> Na2SO4 pH = 2 (**A**) and 0.5 mol dm−<sup>3</sup> Na2SO4 + 0.5 mol dm−<sup>3</sup> NaCl pH = 2 (**B**) for uncoated steel X20Cr13 (a) and covered with coatings VTMS concentrations in a 3.16 mol dm−<sup>3</sup> solution and the addition of an electrolyte: CH3COOH (b), LiClO4 (c), H2SO4 (d), NH3 (e). Polarization rate 10 mVs<sup>−</sup>1, solutions in contact with air.

The potential range of −0.8–1.6 V for the X20Cr13 steel uncoated and coated, respectively.

As follows from Figure 7A, the produced VTMS/EtOH/Electrolyte coatings inhibit the cathodic and anodic processes and shift the corrosion potential of the steel by approximately 0.5 V (the VTMS/EtOH/AcOH coating). The anodic current densities for the steel covered with VTMS/EtOH/Electrolyte coatings in the passive range are smaller by 1–4 times than those for the uncoated steel.

To assess the capacity of the produced coatings to inhibit pitting corrosion, similar potentiodynamic curves were plotted for a sulphate solution acidified to pH = 2, containing an addition of 0.5 mol dm−<sup>3</sup> of chloride ions (Figure 7B).

The corrosion potential of the X20Cr13 steel for all coatings is shifted by approximately 0.1–0.5 V towards positive values relative to the corrosion potential values recorded for the uncoated steel (E*kor* = −0.527 V). Lower values of cathodic and anodic current densities were also observed for the steel covered with these coatings, compared to the uncoated steel.

The shape of the polarization curves shows that the pitting nucleation potential (Epit) amounts to, respectively: for the uncoated steel 0.12 V; for the steel covered with the coatings: VTMS/EtOH/LiClO4 0.18 V; VTMS/EtOH/H2SO4 0.19 V; VTMS/EtOH/NH3 0.64 V. The thermodynamic susceptibility to pitting is similar for the coatings VTMS/EtOH/LiClO4 and VTMS/EtOH/H2SO4. As shown by Figure 7B (line b), in the case of employing the VTMS/EtOH/AcOH coating for steel protection, no puncture potential of the passive film (pitting nucleation potential) was observed. The silane coating modified with acetic acid effectively hinders the access of aggressive anions to the steel substrate, thus protecting the substrate against pitting corrosion. Microscopic observations after the measurement did not reveal any local corrosion effects under the VTMS/EtOH/AcOH coating. Figure 7B implies that the application of coatings on steel protects the substrate against local corrosion.

To verify the resistance of coatings deposited on steel to pitting corrosion, the chronoamperometric method was employed. In this method, variations in current density are recorded as a function of time after applying a constant potential to the working electrode. From chronoamperometric curves, one can infer the nucleation of pits.

To determine the stability of the applied coats, the time of holding the test samples in the corrosion solution containing chloride ions and the value of current density were compared at a preset potential. Chronoamperometric curves were recorded in a 0.5 mol dm−<sup>3</sup> solution of Na2SO4 + 0.5 mol dm−<sup>3</sup> NaCl with pH = 2 at a potential of 0.1 V for uncoated and coated steel, respectively.

Figure 8 shows the chronoamperometric curves for steel plotted at a potential of 0.1 V red out from the polarization curves, Figure 7B. As can be observed, the initiation of pit formation on the steel occurs within several seconds, after which the value of current density dramatically increases. In the case of applying the following coating types, VTMS/EtOH/AcOH, VTMS/EtOH/LiClO4, and VTMS/EtOH/H2SO4, the highest corrosion resistance was achieved. The increase in current density for the above-mentioned coatings occurred within a time span ranging from 250 to 312 h. The best capacity to block the transport of chloride ions responsible for pitting corrosion is shown by the VTMS/EtOH/AcOH coating (312 h).

#### *3.4. Corrosion Resistance Test in a Potassium Hexacyanoferrate (III) Solution (Ferroxyl Test)*

To demonstrate the corrosion resistance of coatings deposited on the X20Cr13 steel, electrochemical tests were carried out in a 2 mmol dm−<sup>3</sup> solution of K3[Fe(CN)6].

Figure 9 shows a typical voltammetric response of the glassy carbon electrode (A) and the VTMS/EtOH/AcOH–coated X20Cr13 steel electrode (B) in the presence of Fe(CN)6 <sup>3</sup><sup>−</sup> sampler ions. In the case of the pure glassy carbon electrode (Figure 9A), we observe a well-developed and quasi-reversible pair of ferrocyanide ions. By contrast, Figure 9B illustrates the voltammetric response of the VTMS/EtOH/AcOH coating on the X20Cr13 steel substrate, on which it does not observe any electrochemical response in the investigated potential range. This is associated primarily with the fact that ferrocyanide ions do not cross through the produced VTMS/EtOH/AcOH layer (pores in the layer are smaller than the size of the ferrocyanide ion). Additionally, no formation of the blue colouring (Prussian Blue formation) was observed on the steel surface, confirming definitely that the obtained coating provides a compact and tight protective barrier. Moreover, the VTMS/EtOH/AcOH layer formed on the X20Cr13 steel blocked the transport of electrons to ferrocyanide ions, has manifested itself by the attenuation of the redox currents (Figure 9B) [49,50].

**Figure 8.** Chronoamperometric curves recorded in a chloride solution (0.5 mol dm−<sup>3</sup> Na2SO4 + 0.5 mol dm−<sup>3</sup> NaCl pH = 2) for X20Cr13 steel not covered with the coating (a) and coated with VTMS in 3.16 mol dm−<sup>3</sup> solution and addition of electrolyte: CH3COOH (b), LiClO4 (c), H2SO4 (d), NH3 (e).

**Figure 9.** Voltammetric response for: glassy carbon (**A**) and coated X20Cr13 steel with VTMS/EtOH/AcOH (**B**). Electrolyte: 2 mmol dm−<sup>3</sup> K3[Fe(CN)6]. Polarization rate 10 mVs<sup>−</sup>1.

#### **4. Conclusions**

The investigation of VTMS/EtOH/Electrolyte coatings has shown that the sol–gel process can be used for producing protective layers on stainless steel X20Cr13.

The selection of the appropriate electrolyte has a significant impact on the corrosion and structural properties of VTMS coatings (a uniform surface with no visible defects in the structure). The produced coatings exhibit good adhesion to the substrate and, in addition, extend the duration of steel resistance to the action of chloride and sulphate ions in an acid

medium. The best ability to block the transport of chloride ions responsible for the pitting corrosion of steel is shown by the VTMS/EtOH/AcOH coating. The surface roughness and thickness of the coating may be influenced by the size of the doped electrolyte ion. Acetic acid-doped silane coatings deposited on the X20Cr13 steel, with low surface roughness and a small thickness of the coating, exhibit the anticorrosion properties.

Data obtained from potentiodynamic measurements show that the produced VTMS/ EtOH/Electrolyte coatings provide stainless steel's anodic and barrier protection. An experiment using a potassium hexacyanoferrate (III) solution has confirmed that the VTMS/EtOH/AcOH coating forms a uniform, tight structure and blocks the transfer of electrons to ferrocyanide ions.

**Author Contributions:** Conceptualization and idea of this study, A.K. and L.A.; Methodology, L.A. and K.M.; Formal analysis, L.A. and K.M.; Writing—original draft preparation, A.K., L.A. and K.M.; Writing—review and editing, L.A. and A.K.; Visualization, L.A. and K.M.; Supervision, L.A.; Project administration, L.A.; Funding acquisition, L.A. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not applicable.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


## *Communication* **Hydrogen-Induced Cracking Caused by Galvanic Corrosion of Steel Weld in a Sour Environment**

**Jin Sung Park 1, Jin Woo Lee <sup>2</sup> and Sung Jin Kim 1,\***


**Abstract:** This study examined the hydrogen-induced cracking (HIC) caused by galvanic corrosion of an ASTM A516-65 steel weld in a wet sour environment using a combination of standard immersion corrosion test, electrochemical analyses, and morphological observation of corrosion damage. This study showed that the weld metal has lower open circuit potential, and higher anodic and cathodic reaction rates than the base metal. The preferential dissolution and much higher density of localized corrosion damage were observed in the weld metal of the welded steel. On the other hand, the presence of weldment can make steel more susceptible to HIC, specifically, in areas of the base metal but not in the weld metal or heat affected zone, which is in contrast to typical expectations based on metallurgical knowledge. This can be explained by galvanic corrosion interactions between the weldment and the base metal, acting as a small anode and a large cathode, respectively. This type of galvanic couple can provide large surface areas for infusing cathodically-reduced hydrogen on the base metal in wet sour environments, increasing the susceptibility of welded steel to HIC.

**Keywords:** ASTM A516-65 steel; weld; hydrogen-induced cracking; galvanic corrosion; sour environment

#### **1. Introduction**

Hydrogen degradation of ferrous alloys has attracted considerable attention in the scientific and engineering community for more than 100 years [1–3]. Among the degradation phenomena, hydrogen-induced cracking (HIC) and sulfide stress corrosion cracking (SSCC) are major technical problems that remain to be addressed in the petrochemical industries. Premature failure caused by hydrogen embrittlement (HE) occurs mainly at the welded joints of the steel structures [4–6]. This is because welds are formed under a welding thermal cycle with rapid heating and cooling processes, and they are normally comprised of dendritic and heterogeneous structures with several metallurgical defects [7,8]. Hence, the welded joint is considered the most critical and problematic part of steel structures.

Sour corrosion is the deterioration that occurs on a metal surface in a highly acidic environment containing H2S. This type of corrosion is expected to occur preferentially at the welds when the welded steel structures are exposed to wet sour environments. The sour corrosion process can be briefly summarized as anodic metal dissolution (M → Mn+ + e−) followed by the infusion of cathodically-reduced hydrogen (H+ + e<sup>−</sup> → H) in steel [9–11]. The poisoning effect by H2S facilitates the absorption of atomic hydrogen, and the hydrogen could be trapped at certain metallurgical defects in steel [9,12,13]. According to internal pressure theory [14], which has been widely accepted as a mechanism of HE in steels, the continuous trapping of atomic hydrogens tends to recombine into molecular hydrogen (H + H → H2) and leads to significant volume expansion, resulting in the nucleation of cracks.

The heat affected zone (HAZ) with a coarse grain size can be the most inferior part of the welded steel, and numerous papers have reported the (hydrogen-induced) mechanical degradation of the HAZ depending on the welding heat input [15,16]. In the case of weld metal (WM), welding consumables can be one of the factors controlling the resistance to

**Citation:** Park, J.S.; Lee, J.W.; Kim, S.J. Hydrogen-Induced Cracking Caused by Galvanic Corrosion of Steel Weld in a Sour Environment. *Materials* **2021**, *14*, 5282. https:// doi.org/10.3390/ma14185282

Academic Editor: Vít Kˇrivý

Received: 25 August 2021 Accepted: 10 September 2021 Published: 14 September 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

hydrogen-assisted cracking (HAC) failures. On the other hand the welding consumable is adopted considering mostly the mechanical properties of the base metal (BM). This welding consumable does not guarantee the WM will have high resistance to corrosion or corrosion-induced HAC. In this regard, Pagotto et al. [17] reported a much higher anodic dissolution current at welds relative to the BM of carbon steel in a neutral aqueous environment, employing a scanning vibrating electrode technique (SVET).

In contrast to the common metallurgical aspects described above, the authors found that the HIC ratio of the BM was higher than that of the WM when the welded steel was exposed to a wet sour environment. Moreover, the cracking problem of the BM became worse in the presence of the WM compared to the unwelded steel sample. This is closely associated with the formation of a galvanic couple with the WM and BM in an acidic aqueous environment. In this study, the National Association of Corrosion Engineers (NACE) standard HIC test, diffusible hydrogen measurement, and several electrochemical evaluations were conducted to clarify the mechanistic reason for the more serious damage by HIC in the BM, which can be protected galvanically.

#### **2. Experimental Procedure**

The test materials under investigation were equivalent to an ASTM A516-65 grade pressure vessel steel plate with a 15 mm thickness produced by an industrial rolling process. The chemical compositions of the two types of steel samples used in this study, termed Steel A and B, are listed in Table 1. The steels were normalized by heating to 910 ◦C for 8 min and cooled to room temperature in air.

**Table 1.** Chemical compositions of the two tested steel samples.


To produce the welded samples for Steel A and B, a double X-groove was produced, and the tandem submerged arc welding (SAW) was performed using two solid wires (OE-SD3, 0.07% C-0.9% Mn-0.3% Si) with diameters of 4 mm each and an OP121TT (0.07% C-1.6% Mn-0.3% Si) flux. The total heat input from the sum of two electrodes was approximately 30 kJ/cm, which was calculated using the welding parameters; these are listed in Table 2.

**Table 2.** Welding parameters and calculated heat input.


A brief schematic diagram of the double-pass welded sample is presented in Figure 1a.

**Figure 1.** (**a**) Brief schematic diagram of the double-pass welded sample; (**b**) Cross-section view of the two welded samples; (**c**) Microstructures in the WM, HAZ, and BM of the two welded samples; (**d**) Vickers hardness profile of the two welded samples.

Metallographic observations of the WM, HAZ, and BM by optical microscopy (OM) (Zeiss, Jena, Germany) and field-emission scanning electron microscopy (FE-SEM) (Hitachi, Tokyo, Japan) were conducted after the steel samples had been polished up to a 1 μm surface finish and etched in a 3% Nital solution. After the macro- and micrographic observations, a Vickers hardness test of the two welded samples was performed with a constant force of 500 gf for 10 s, and the hardness distributions in different zones of the welded joints were obtained.

The HIC sensitivity was evaluated according to the NACE TM0284 standard method [18], and the sensitivity indices of the crack length ratio (CLR (%)) and crack area ratio (CAR (%)) were determined using an ultrasonic detection method. To ensure reproducibility, three samples obtained from the two tested materials (Steel A and B) were evaluated. After the HIC tests, the diffusible hydrogen contents introduced in the samples were measured using a glycerin volumetric method in reference to JIS Z3113 standard [19]. For this, immediately after the HIC tests, the samples were inserted into the glycerin column that was maintained at 45 ◦C, using liquid nitrogen as a medium to prevent hydrogen diffusion out of the samples. After three days, the volume of hydrogen collected at the top of the glycerin column was measured.

For the mechanistic study, three types of electrochemical measurements (open circuit potential (OCP), potentiodynamic (PD) polarization, and galvanic current) were conducted in a simulated wet sour environment (5% NaCl + 0.5% CH3COOH + 0.05 M Na2S solution). A typical three-electrode cell composed of a steel sample, a Pt grid, and a saturated calomel electrode (SCE), which acted as the working, counter, and reference electrode, respectively, was used for the OCP and PD polarization measurements. Before the tests, the samples were ground to 2400 grit sand-paper and cleaned ultrasonically in ethanol. For the PD polarization, the potential was scanned from −500 mV to 500 mV vs OCP at a scan rate of 0.2 mV/s. A potentiostat (Gamry reference 600, Pennsylvania, America) in zero-resistance ammeter (ZRA) mode was used to measure the variations in the galvanic current flow between the WM and BM with dimensions of 1 cm2. The distance between the two electrodes was 20 mm. With these electrochemical analyses, the surface and cross-section morphologies of the welded steel sample were observed after immersion in a simulated wet sour solution for seven days.

#### **3. Results and Discussion**

Figure 1 shows the macrostructure, microstructure, and hardness profile of the two welded steel samples. The major differences between the two BMs of the two samples

lie in the banding index of pearlite and the level of the 2nd phase particles, which was reported previously [20], but they were not the focus of the present study. Under the same welding conditions and consumables, however, there was no significant difference in the macro- and microstructures of weldments of the two samples. The distributions of the hardness profile of the two welded samples showed a similar pattern: the highest and lowest hardness values were measured around the fusion lines and the BM, respectively.

According to common knowledge in welding metallurgy, a HAZ with high hardness and a coarse grain can be considered the most critical and problematic area in a highstrength steel weld [21–23]. For this reason, the SSC of a welded steel sample, which is equivalent to the sample in this investigation, occurred in the HAZ, which has been discussed elsewhere [21,24,25] and is not covered in the present study. The point the authors try to make in this study is to clarify the underlying mechanisms of the changes in the HIC levels with the presence of a narrow weldment in steel samples.

Figure 2 presents the ultrasonically detected HICs of the unwelded and welded steel samples, which had been immersed in a NACE solution saturated with H2S. The differences in the HIC sensitivity (Table 3) between the two unwelded samples are discussed elsewhere [20]. The focus here was on the changes in the HIC levels and distributions after welding the steel samples.

**Figure 2.** (**a**) Schematic diagram showing the dimensions of welded sample for HIC test in reference to NACE TM0284 standard method; ultrasonically detected HICs of the unwelded and welded samples after HIC test: (**b**) Steel A and (**c**) Steel B.


**Table 3.** Mean values (μ) and their standard deviations (σ) of ultrasonically detected CLR (%) and CAR (%) values of the two samples after HIC test in reference to NACE TM0284 standard method.

In contrast to common expectation, most HICs occurred in the BM in the welded samples, not in the HAZ or WM. Moreover, it is interesting to note that the HIC levels in the BM of both welded steel samples were even higher than those of unwelded samples. Under the same materials in the absence of an externally applied stress, the higher susceptibility to HIC could mainly be due to the higher infusion of hydrogen in the materials [26,27]. This suggests that the presence of a weldment in the tested sample could lead to additional hydrogen uptake in the BM some distance from the weldment. This is supported by the increase in the amount of diffusible hydrogen contents ([H]diff.) introduced in the tested samples after welding, as shown in Figure 3. Although the [H]diff. is the sum of the diffusible hydrogen contents obtained from the BM, HAZ, and WM, and each contribution cannot be extracted separately, judging from the HIC occurrence location (Figure 2), most of it may be obtained from the BM.

**Figure 3.** Diffusible hydrogen contents in the unwelded and welded samples: (**a**) Steel A and (**b**) Steel B. (Error bars represent the standard deviations of the mean values).

An electrochemical approach can be adopted to understand the underlying mechanism behind the higher hydrogen infusion and resulting HIC in the BMs of the welded samples. Because the WM and BM are connected electrically, an electrochemical potential difference can be generated through the differences in chemical composition between them, leading to a galvanic current flow from the anode to the cathode. Figure 4a shows that the WM has a slightly lower open circuit potential than the BM, indicating that the WM may be a more active electrode and act as an anode. Although no significant differences in the PD polarization curves (Figure 4b) between the BM and WM of each steel sample were observed, the anodic reaction (Fe → Fe2+ + 2e−) and cathodic reaction (H<sup>+</sup> + e<sup>−</sup> → H) rates of the WM were slightly higher than those of the BMs. From a practical aspect, the difference in corrosion current density (*icorr*) between the WM and BM appears to be insignificant. Nevertheless, there could be sufficient driving force for galvanic corrosion between the WM and BM when they are coupled, which can be supported by the galvanic current flow and current level, as shown in Figure 4c. The measured positive current density also indicates that the galvanic current flows from the WM to the BM, and more electrons can be supplied to the BM. In particular, the geometry of the welded steel sample, or even large-sized welded steel structures such as welded pipes, can provide a large cathode (BM) and small anode (WM) ratio. This ratio can be another significant factor expediting galvanic corrosion. From an electrochemical perspective, the larger the cathode area compared to the anode, the greater the galvanic current, which is the more favorable condition for galvanic interactions. Hence, it can be accepted that the anodic steel dissolution and cathodic hydrogen reduction are dominant on the WM and BM, respectively, in the welded sample. The preferential dissolution and much higher density of localized corrosion damage of the WM in the welded sample, shown in Figure 4d, can be the metallographic observation of galvanic corrosion.

The formation of this type of galvanic couple between the WM and BM in an acidic sour solution leads to more hydrogen reduction (H<sup>+</sup> + e<sup>−</sup> → H) in the BM, resulting in the higher infusion of hydrogen in the BM and more vulnerability to HIC. This process is illustrated schematically in Figure 5.

**Figure 4.** Measurements of (**a**) OCP, (**b**) PD polarization, and (**c**) Galvanic current; (**d**) Metallographic observation of weld metal in Steel A after an immersion test.

**Figure 5.** Brief schematic illustration showing the mechanism of galvanic corrosion between the WM and BM, and hydrogen infusion and cracking in the BM.

#### **4. Summary**

This work elucidates the preferential occurrence of HIC in the BM in the welded steel under wet sour corrosion with a series of experimental results. The major findings are summarized as follows.

The preferential occurrence of HIC in the BM is caused primarily by the fact that the chemical composition of WM, which was optimized for mechanical properties, is slightly anodic to the BM. Even if the WM is close in chemical composition to the BM, the dendritic and inhomogeneous microstructure of the WM can also produce a potential difference with the BM, leading to galvanic corrosion. This leads to the galvanic current flows from the WM to the BM, suggesting that more electrons can be supplied to the BM. Hence, the cathodic hydrogen reduction is more dominant on the BM resulting in the higher infusion of hydrogen in the BM (i.e., welded samples had 5.73% (Steel A) and 28.45% (Steel B) higher [H]diff. than unwelded samples) and more susceptibility to HIC (i.e., welded samples had 66.51% (Steel A) and 167.11% (Steel B) higher CLR (%) than unwelded samples). Therefore, optimizing the WM so that its chemical composition is very close to that of the BM or slightly nobler than that of the BM is an effective strategy to suppress preferential anodic dissolution and the formation of HIC in the WM and BM, respectively, in the welded steels under an acidic sour environment.

**Author Contributions:** Conceptualization, J.W.L. and S.J.K.; methodology, J.S.P., J.W.L., and S.J.K.; investigation, J.S.P. and S.J.K.; data curation, J.S.P. and S.J.K.; writing-original draft preparation, J.S.P. and S.J.K.; review and editing, S.J.K. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was supported in part by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT), No. 2019R1C1C1005007. In addition, this work was partly funded and conducted under the Competency Development Program for Industry Specialists of the Korean Ministry of Trade, Industry and Energy (MOTIE), operated by the Korea Institute for Advancement of Technology (KIAT), No. P0002019, HRD Program for High Value-Added Metallic Materials Expert.

**Data Availability Statement:** The data that support the plots and other findings of the current study are available from the corresponding author on reasonable request.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**

