metals

Article Influence of Exposure to Environment on Degradation of a Friction Stir Welded Aluminum Alloy

Shengli Lv 1,2,3,*, Zhi Li 1,2, Xiaosheng Gao 4 and Tirumalai S. Srivatsan 4

1 National Key Laboratory of Science and Technology on UAV, Northwestern Polytechnical University, Xi’an 710072, China; [email protected] 2 School of Aeronautics, Northwestern Polytechnical University, Xi’an 710072, China 3 Innovation Group of Marine Engineering Materials and Corrosion Control, Southern Marine Science and Engineering Guangdong Laboratory, Zhuhai 519080, China 4 Department of Mechanical Engineering, The University of Akron, Akron, OH 44325, USA; [email protected] (X.G.); [email protected] (T.S.S.) * Correspondence: [email protected]; Tel.: +86-1399-122-1186



Received: 28 September 2020; Accepted: 26 October 2020; Published: 29 October 2020


Abstract: We carried out a comparative study on both the stress corrosion response and corrosion damage characteristics of aluminum alloy 2219, both the base material and a friction stir welding (FSW) counterpart upon exposure to exfoliation corrosion (EXCO) solution. The results reveal that the test specimen containing an FSW joint reveals better electrochemical corrosion resistance than that taken from the base metal. When test specimens upon exposure to EXCO solution are concurrently subjected to a tensile stress, since the mechanical properties of the FSW joint are lower than the base metal, a test specimen containing an FSW joint is more easily prone to the early initiation of fine microscopic cracks. This makes the test specimen containing the FSW joint to be less resistant to stress corrosion damage than that taken from the base metal for the various levels of applied stress and exposure time to EXCO solution. The average corrosion depth of the test specimen containing the FSW joint is less than that of the base metal, while the maximum corrosion depth of it is greater than that of the base metal. This reveals that test specimen containing the FSW joint is more susceptible to damage and degradation than test specimen taken from the base metal.

Keywords: aluminum alloy 2219; FSW; stress corrosion damage; corrosion morphology

1. Introduction Aluminum alloy 2219 is an Al-Cu-Mn alloy that is receptive to heat treatment. This alloy offers the characteristics of low density, high specific strength (σ/ρ), good heat resistance coupled with high sensitivity to stress corrosion. Friction stir welding (FSW) is a new material joining technology invented by The Welding Institute (TWI) in 1991. During FSW process, a rotating stirring pin is inserted between the workpieces to be welded and moves along the welding path. The stirring head shoulder and the stirring pin rub against the workpiece to be welded and generate a lot of heat. The workpiece is heated to below the melting point and in a plastic state under the action of heat, and the solid-phase connection of the material is realized under the action of stirring and clamping force [1]. The technique of FSW is essentially a solid-phase bonding technology. Compared one-on-one with the traditional fusion welding methods, the FSW joint of an aluminum alloy has a more homogeneous microstructure coupled with the presence of fewer defects [2,3]. During FSW, the chosen aluminum alloy is subjected to stirring caused by a stirring needle coupled with heat input. This favors the occurrence of plastic deformation at the fine microscopic level coupled with dynamic recrystallization and a resultant change to the microstructure. A change in microstructure often results in a change in mechanical

Metals 2020, 10, 1437; doi:10.3390/met10111437 www.mdpi.com/journal/metals Metals 2020, 10, 1437 2 of 21 properties of both the FSW joint and the adjacent material. The literature shows that in the FSW process, the metal material is stirred by the stirring pin, resulting in a large amount of heat input, plastic flow, and dynamic recrystallization to occur. Consequently, the microstructure of the formed welded joint changes compared to the base material, which leads to the performance of the material in the welded joint zone becoming different from the base material. Fratini studied the correlation between the flow of material and the microstructure of the weld nugget during friction stir welding of 7075 aluminum alloy and predicted the grain size of the weld joint zone [4]. Gan discovered the abnormal growth of grains in the weld zone and heat-affected zone of aluminum alloy 5083 and 6111 friction stir welding and described the relationship between process parameters, microstructure changes, and mechanical properties of the joint zone [5]. An FSW joint does result in three microscopic zones [6,7]. These are the following: The nugget zone (NZ). • Thermo-mechanical affected zone (TMAZ). • Heat-affected zone (HAZ). • In the NZ, which is subjected to the action of stirring caused by the stirring needle coupled with high temperature, the end result is the formation and presence of fine equiaxed grains. The TMAZ is not directly stirred by the stirring needle. Consequently, it experiences a lower degree of plastic deformation at the “local” level coupled with lower thermal influence than the NZ. This often results in the formation and presence of elongated grains with the grain size being noticeably larger than that of the NZ. The HAZ is also subjected to a thermal cycle caused by the stirring action that causes the grains in this zone to grow and eventually coarsen. Due essentially to plastic damage caused by the crushing action of stirring and welding-related defects, such as looseness and holes, which are difficult to avoid in the weld zone, there does exist an overall weak joint whose strength is about 55 pct. to 80 pct. of the base metal. A noticeable change in both microstructure and resultant properties does make the damage due to stress corrosion experienced by the FSW joint to be noticeably different from that of the base metal. The FSW region of an aluminum alloy is a complex zone that essentially has an uneven structure and resultant unpredictable mechanical performance. Different materials coupled with different welding process parameters will tend to make the microstructure at the region of the joint to be different. Furthermore, different environments in synergism with different loads will tend to promote varying degree of environment-induced interactions and/or degradation. Therefore, FSW joint of an aluminum alloy does exhibit complex characteristics when interacting with the environment and the resultant environment-induced degradation, which is referred to as corrosion. During service, the material chosen is often subjected to the conjoint influence of corrosive medium and static stress. Thus, there does exist the risk of inducing conditions that are favorable for the initiation of stress corrosion cracking. Over the years, few researchers have systematically conducted a sizeable amount of research on the corrosion characteristics of both aluminum alloys and their FSW counterparts. Some of these research scholars are of the belief that FSW tends to reduce the overall corrosion resistance of the joint. Wang used water-cooled friction stir welding to improve the microstructure of aluminum alloy 7055 and found the mechanical properties of the joint were improved, but the corrosion behavior of the alloy was affected [8]. Gharavi and Matori and co-workers studied the corrosion behavior of the FSW joint of aluminum alloy 6061-T6 [9,10]. They found the corrosion resistance of the NZ to decrease because of two reasons, and these are the following: Since pitting corrosion occurred both at and along the edges of the intermetallic compounds • present in the microstructure, an increase in both the presence and distribution of intermetallic compounds in the NZ contributes to a gradual increase in the corrosion galvanic couple. A refinement in grain structure of the NZ results in an increase in corrosion sensitivity. • The experiments of Lumsden et al. also support the above findings and conclusion [11,12]. Through a series of corrosion experiments systematically conducted on FSW joints of aluminum Metals 2020, 10, 1437 3 of 21 alloy 7075-T7651, they found that, due to the depletion of copper in the precipitation free zone (PFZ) present both at and along the grain boundary regions, the pitting potential of the NZ decreased favoring the occurrence of selective grain boundary corrosion that only resulted in a gradual decrease in resistance to intergranular corrosion. The interface between the NZ and zone of incomplete recrystallization became the most sensitive zone for the occurrence of damage due to stress corrosion cracking. In this zone, it was easy for the fine microscopic cracks to propagate, eventually resulting in failure by intergranular fracture. This contributed to reducing the overall resistance of the test specimen containing the FSW joint to stress corrosion cracking. Chen studied and documented the microstructure, mechanical properties, and corrosion behavior of friction stir welded joints of aluminum alloy 5086 and aluminum alloy 6061 [13]. He observed noticeable differences in microstructure, mechanical properties and corrosion behavior of the different zones for the two aluminum alloys. He also found the corrosion resistance of the FSW joint of the chosen aluminum alloy to reveal noticeable improvement. Esmaily and co-workers studied the corrosion performance of aluminum alloy 6005-T6 having a double shoulder FSW joint using the atmospheric corrosion experiment [14]. They found and recorded the overall corrosion performance of the joint zone to be superior to that of the base metal. Venugopal studied the pitting corrosion resistance of FSW joint of aluminum alloy 7075 in a 5 percent sodium chloride (NaCl) solution [15]. He found the corrosion resistance of the NZ to be better than both the TMAZ and the base metal. Wadeson and co-workers studied and documented the overall corrosion resistance of aluminum alloy 7108-T79 containing an FSW joint [16]. They found the edge of the TMAZ to be receptive to environment-induced degradation, or corrosion, which did gradually extend to the HAZ due essentially to the uneven precipitation of the strengthening precipitates (both η and η’) in the TMAZ. Paglia and co-workers found from their experiments that a noticeable difference in both microstructure and microchemistry of an FSW high-strength aluminum alloy contributes to deteriorating the overall environment-induced degradation, or corrosion, while coarsening of the strengthening precipitates present both at and along the grain boundaries promoted the occurrence of intergranular corrosion at the region of the FSW joint [17,18]. Corrosion occurred both at the NZ and the HAZ. The HAZ was found to be least resistant to degradation by corrosion. This can essentially be attributed to the width of the copper-depleted PFZ that is present both at and along the grain boundary regions. Lumsden and co-workers pointed out that for the FSW joint of aluminum alloy 7075-T651, the NZ had a lowest pitting potential, and pitting potential of the HAZ was noticeably lower than that of the base metal [11]. Emilie and co-workers found and documented the FSW joint of aluminum alloy 2024-T3 to experience most severe intergranular corrosion at the HAZ [19]. Both pitting corrosion and intergranular corrosion occurred in the base metal and the TMAZ, and pitting corrosion in NZ. Litwhiski and co-workers in their independent study pointed out that the TMAZ was easily susceptible to corrosion for the FSW aluminum alloy 2195 [20]. Padgett studied both the cracking and “local” corrosion behavior of FSW joint of aluminum alloy 2099 under conditions of environmental impact [21]. They observed the edge of the HAZ, on the retreating side of a joint, to be most sensitive to stress corrosion cracking (SCC). However, the experiment of Chen revealed the retreating side to have much better corrosion resistance than the other zones of the aluminum alloy chosen and studied [13]. Paglia and co-workers found the microstructure of each zone of an FSW joint of aluminum alloy 2219-T87 to be different [18]. However, overall SCC behavior was found to be the same. The test specimen when deformed in tension failed at the junction of the NZ and the TMAZ. Astarita and co-workers studied the stress corrosion cracking behavior of the following FSW aluminum alloys: (i) 2024-T3, (ii) 2139-T3, (iii) 2198-T3, and (iv) 6056-T4 [22]. These researchers found the FSW joints of the chosen and studied aluminum alloys to have noticeably different stress corrosion characteristics. For the 2XXXseries aluminum alloy, the FSW joint exhibited anodic characteristics relative to the base metal, while for the 6XXX-series aluminum alloy, the FSW joint exhibits cathodic characteristics relative to the base metal. A careful review of the published literature reveals the different studies conducted on different aluminum alloys resulted in observable differences in the conclusions. The reason for this being that corrosion of an aluminum alloy is a complex process, which is affected by the conjoint and mutually Metals 2020, 10, 1437 4 of 21

interactive influences of several factors. For example, composition of the chosen aluminum alloy is different, or the process parameters chosen for FSW are marginally different. This results in observable differences in the microstructure. Further, while in service different loads on an aluminum alloy will tend to promote different corrosion processes occurring at the fine microscopic level often culminating in different degree and/or severity of damage. Among the variables, the nature of corrosion medium chosen and applied stress level are two key factors that exert an influence on both the occurrence and severity of damage due to stress corrosion at the FSW joint. Therefore, in this study, exfoliation corrosion (EXCO) solution was chosen as the corrosive medium to study the stress corrosion characteristics of aluminum alloy 2219 when under the influence of a tensile stress. The objective was to help both in establishing and understanding the influence of intrinsic microstructural changes occurring at the joint on overall damage caused to the chosen aluminum alloy due to stress corrosion.

2. Materials and Methods The test material chosen was aluminum alloy 2219. The as-provided plate stock had a thickness of 2.5 mm. The chemical composition of the as-provided alloy is as shown in Table1. The physical and mechanical properties of the as-provided alloy are summarized in Table2. Two of the as-provided aluminum alloy plates were welded using the technique of FSW. The test specimens were precision cut from the welded plates using the technique of wire cutting. Two kinds of test specimens chosen for this study were prepared from both the as-provided aluminum alloy plate and the FSW alloy plate. The test specimens taken from the base metal, i.e., as provided aluminum alloy 2219, were cut along the edges of the as-provided plate. The test specimen taken from the FSW aluminum alloy plate was cut along the weld direction at the region of the welded joint. Both the FSW joint specimen and test specimen taken from the base metal, i.e., as-provided aluminum alloy, had the same shape and size. This is shown in Figure1. Tensile strength of the FSW joint specimen of aluminum alloy 2219 was 295 MPa, which is 62% of the tensile strength of test specimen of the base metal, i.e., as-provided aluminum alloy.

Table 1. Chemical composition of aluminum alloy 2219 (in weight percent).

Cu Si Zr Fe Zn V Ti Mn Mg Al 5.8 ~ 6.5 0.2 0.1 ~ 0.25 0.3 0.1 0.05 ~ 0.15 0.01 ~ 0.1 0.2 ~ 0.4 0.02 Balance ≤ ≤ ≤

Table 2. Physical and mechanical properties of aluminum alloy 2219.

Density Tensile Strength Yield Strength Elongation Hardness Heat Treatment State (g/cm3) (MPa) (MPa) (%) (Hv) Metals 20202.84, 10, x FOR PEER C10SREVIEW 475 395 10 105 5 of 22

(a) (b)

FigureFigure 1.1. SchematicSchematic showing showing size size and and shape shape of test of specimentest specimen used forused the for corrosion the corrosion test. (a) Testspecimentest. (a) Test specimenof the base of metal; the base (b) metal; test specimen (b) test specimen containing containing the friction the stir fr weldingiction stir (FSW) welding joint. (FSW) joint.

Both stress level and duration of exposure to an aqueous solution were chosen as the control variables. Consequently an experimental scheme was designed to carry out the accelerated corrosion tests on both FSW test specimens of aluminum alloy 2219 and the unwelded alloy, i.e., the base metal (aluminum alloy 2219). The experiments were carried out at room temperature (24 °C) and laboratory air environment (relative humidity of 55 %). The test specimens were subject to three different stress levels, i.e., (i) 0 MPa, (ii) 79 MPa, and (iii) 118.5 MPa, corresponding to 0 % , 20%, and 30% of the yield strength of the chosen aluminum alloy. At each stress level, the duration of exposure of test specimens of the aluminum alloy to the environment was (i) 8 h, (ii) 16 h, (iii) 24 h, and (iv) 48 h. By observing the morphology and severity of environment-induced degradation and concurrently measuring the depth of corrosion-induced damage experienced by test specimens of the chosen aluminum alloy, the overall severity of damage experienced by the test specimens taken from the base metal and the FSW joint are compared. In an attempt to avoid deviation of the test results arising from a dispersion of material properties, the experiment was designed to be conducted in triplicate, i.e., testing three specimens for each condition. The environment chosen for purpose of this study was an EXCO solution prepared very much in conformance with specifications and procedures detailed in Standard ASTM G34-01 [23]. Composition of the solution is provided in Table 3 and had a starting pH value of 0.4.

Table 3. Composition and ratio of exfoliation corrosion (EXCO) solution.

Composition NaCl KNO3 HNO3(70%) Content 234 g/L 50 g/L 63 ml/L

The key equipment required for the experiment was the following: • A loading device for the stress corrosion test was used to apply a constant tensile stress on the test specimen while being exposed to the environment during loading; • A continuous zoom video microscope (Model: UNION DZ3; Union.Co., Chuo-ku, Kobe, Japan) was used to observe the morphology, nature, extent and severity of environment-induced damage, or corrosion experienced by the test specimens taken from both the base metal and the FSW joint. • A low magnification scanner (Model: HP 8200, Hewlett-Packard, Palo Alto, CA, USA) used to establish the macroscopic corrosion morphology both at and surrounding the zone of corrosion. • A laser displacement sensor (Model: Keyence LK-G30, Keyence (China) Co. Ltd., Shanghai, China) to measure the depth of pits on the surface of test specimens exposed to the aggressive aqueous solution (i.e., EXCO solution) while concurrently obtaining distribution data specific to depth of the pit. The measurement accuracy of the instrument can reach 0.02 μm.

For test specimens taken from the base metal, i.e., as provided aluminum alloy plate, and test specimens containing the FSW joint, the following steps were repeated. The chosen test specimens

Metals 2020, 10, 1437 5 of 21

Both stress level and duration of exposure to an aqueous solution were chosen as the control variables. Consequently an experimental scheme was designed to carry out the accelerated corrosion tests on both FSW test specimens of aluminum alloy 2219 and the unwelded alloy, i.e., the base metal (aluminum alloy 2219). The experiments were carried out at room temperature (24 ◦C) and laboratory air environment (relative humidity of 55%). The test specimens were subject to three different stress levels, i.e., (i) 0 MPa, (ii) 79 MPa, and (iii) 118.5 MPa, corresponding to 0%, 20%, and 30% of the yield strength of the chosen aluminum alloy. At each stress level, the duration of exposure of test specimens of the aluminum alloy to the environment was (i) 8 h, (ii) 16 h, (iii) 24 h, and (iv) 48 h. By observing the morphology and severity of environment-induced degradation and concurrently measuring the depth of corrosion-induced damage experienced by test specimens of the chosen aluminum alloy, the overall severity of damage experienced by the test specimens taken from the base metal and the FSW joint are compared. In an attempt to avoid deviation of the test results arising from a dispersion of material properties, the experiment was designed to be conducted in triplicate, i.e., testing three specimens for each condition. The environment chosen for purpose of this study was an EXCO solution prepared very much in conformance with specifications and procedures detailed in Standard ASTM G34-01 [23]. Composition of the solution is provided in Table3 and had a starting pH value of 0.4.

Table 3. Composition and ratio of exfoliation corrosion (EXCO) solution.

Composition NaCl KNO3 HNO3(70%) Content 234 g/L 50 g/L 63 ml/L

The key equipment required for the experiment was the following:

A loading device for the stress corrosion test was used to apply a constant tensile stress on the test • specimen while being exposed to the environment during loading; A continuous zoom video microscope (Model: UNION DZ3; Union.Co., Chuo-ku, Kobe, Japan) • was used to observe the morphology, nature, extent and severity of environment-induced damage, or corrosion experienced by the test specimens taken from both the base metal and the FSW joint. A low magnification scanner (Model: HP 8200, Hewlett-Packard, Palo Alto, CA, USA) used to • establish the macroscopic corrosion morphology both at and surrounding the zone of corrosion. A laser displacement sensor (Model: Keyence LK-G30, Keyence (China) Co. Ltd., Shanghai, • China) to measure the depth of pits on the surface of test specimens exposed to the aggressive aqueous solution (i.e., EXCO solution) while concurrently obtaining distribution data specific to depth of the pit. The measurement accuracy of the instrument can reach 0.02 µm.

For test specimens taken from the base metal, i.e., as provided aluminum alloy plate, and test specimens containing the FSW joint, the following steps were repeated. The chosen test specimens were polished to a near mirror-like surface finish with progressively finer grades of silicon carbide (SiC) impregnated emery paper (i.e., 400-grit, 600-grit, 800-grit, 1000-grit, 1200-grit, 1500-grit, 2000-grit, 2500-grit, 3000-grit, and 5000-grit). The as-polished test specimens were cleaned using anhydrous ethanol and subsequently dried in ambient air. The middle part of the test specimen, measuring 20 mm by 6 mm, was reserved for the corrosion test. Other portions of the chosen test specimen were coated and sealed with paraffin in a hot melting state. This is shown in Figure2. The corrosion zone of the chosen test specimen was then exposed to the chosen aqueous environment, i.e., EXCO solution. According to ASTM standard, the quantity of solution taken should be 70 ml. A specified tensile stress was applied at both ends of the chosen test specimen. When the specified duration of exposure to the aggressive aqueous environment was reached, the test specimen was unloaded and removed. Concentrated nitric acid (65%) was used to wipe the “exposed” area of the test specimen with the prime intent of removing the presence of corrosion products. An ultrasonic vibration cleaner was then Metals 2020, 10, x FOR PEER REVIEW 6 of 22

were polished to a near mirror-like surface finish with progressively finer grades of silicon carbide (SiC) impregnated emery paper (i.e., 400-grit, 600-grit, 800-grit, 1000-grit, 1200-grit, 1500-grit, 2000- grit, 2500-grit, 3000-grit, and 5000-grit). The as-polished test specimens were cleaned using anhydrous ethanol and subsequently dried in ambient air. The middle part of the test specimen, measuring 20 mm by 6 mm, was reserved for the corrosion test. Other portions of the chosen test specimen were coated and sealed with paraffin in a hot melting state. This is shown in Figure 2. The corrosion zone of the chosen test specimen was then exposed to the chosen aqueous environment, i.e., EXCO solution. According to ASTM standard, the quantity of solution taken should be 70 ml. A specified tensile stress was applied at both ends of the chosen test specimen. When the specified duration of exposure to the aggressive aqueous environment was reached, the test specimen was Metalsunloaded2020, 10 and, 1437 removed. Concentrated nitric acid (65%) was used to wipe the “exposed” area 6of of the 21 test specimen with the prime intent of removing the presence of corrosion products. An ultrasonic vibration cleaner was then used to clean the test specimen surface using anhydrous ethanol with the usedprime to purpose clean the of test removing specimen any surface residual using corrosion anhydrous products. ethanol The with test the specimen prime purpose was then of dried removing using anyhigh-velocity/speed residual corrosion air products. from a hair The blower. test specimen The laser was depth then driedmeasuring using high-velocityinstrument was/speed used air to from both ameasure hair blower. and Therecord laser any depth change measuring in depth instrument of the exposed was used surface to both along measure the longitudinal and record any symmetry change incenter depth line of theof the exposed corroded surface test alongspecimen the longitudinalof aluminum symmetry alloy 2219, center and lineabout of 600 the corrodedsurface depth test specimenchanges can of aluminum be obtained alloy per 2219, millimeter. and about Approximatel 600 surface depthy 12,000 changes data points can be ca obtainedn be obtained per millimeter. for each Approximatelymeasurement line, 12,000 which data can points quantitatively can be obtained reflect for the each damage measurement of the specimen. line, which For can a joint quantitatively specimen, reflectthe data the quantitatively damage of the represent specimen. the For damage a joint specimen,in the NZ. theA damage data quantitatively model, due representto corrosion the of damage the test inspecimen the NZ. Asurface, damage can model, be establishe due tod corrosion using the of data the testcollected specimen on depth surface, of the can test be establishedspecimen surface. using theThis data provides collected a measure on depth of the of thecorrosion test specimen damage surface.experienced This by provides the test specimen a measure for of different the corrosion levels damageof applied experienced stress and by theduration test specimen of exposure for di fftoerent the levelsaqueous of appliedenvironment stress and(i.e., duration EXCO ofsolution). exposure A toscanner the aqueous was used environment to record the (i.e., macroscopic EXCO solution). morpho Alogy scanner of corrosion was used on to the record surface the macroscopicof the chosen morphologytest specimen. of The corrosion microscopic on the morphology surface of the of chosen the corroded test specimen. region of The the microscopic test specimen morphology was observed of thein a corroded high magnification region of the microscope. test specimen was observed in a high magnification microscope.

FigureFigure 2. 2.Schematic Schematic showingshowing corrosioncorrosion zone zone on on the the test test specimen specimen and and nature nature of of loading. loading.

SinceSince therethere areare two types of of specimens specimens ((i) ((i) sp specimenecimen taken taken from from FSW FSW alloy alloy plate plate and and (ii) (ii)specimen specimen taken taken from from the as-provide the as-providedd aluminum aluminum alloy plate), alloy plate),three levels three of levels applied of appliedstress, and stress, four anddurations four durations of exposure of exposure to the environment, to the environment, a total of a 24 total tests of was 24 tests carried was out. carried Each out. test Each was testrepeated was repeatedthree times three for times the purpose for the purposeof ensuring of ensuring consistency consistency in the test in results. the test results. 3. Results and Discussion 3. Results and Discussion 3.1. Analysis of Morphology of Electrochemical Corrosion 3.1. Analysis of Morphology of Electrochemical Corrosion 3.1.1. Morphology at the Macroscopic Level 3.1.1. Morphology at the Macroscopic Level When the value of applied stress is 0 MPa, the test specimen experiences pure electrochemical corrosion.When When the value the test of specimenapplied stress of aluminum is 0 MPa, alloy the test 2219 specimen was exposed experiences to the aqueous pure electrochemical environment, i.e.,corrosion. EXCO solution,When the pitting-induced test specimen of damage aluminum was favoredalloy 2219 to was occur. exposed With an to increase the aqueous in the environment, duration of exposure to the aqueous environment, corrosion of the surface of the test specimen gradually increased and tended to become uniform, appearing to the naked eye as spalling corrosion. Results of the tests reveal that corrosion of the test specimen containing the FSW joint and test specimen of the base metal gradually increases with an increase in the duration of exposure to the aqueous environment. For the same duration of exposure to the aqueous environment, i.e., EXCO solution, the corrosion experienced by test specimen of aluminum alloy 2219 containing the FSW joint was noticeably less than that of the test specimen taken from the base metal, i.e., alloy 2219. The macroscopic corrosion morphology of the test specimen containing the FSW joint and test specimen of the base metal, under identical conditions, is shown in Figure3. It is observed from (a) that when duration of exposure to the aqueous environment is 8 h, small pits were observed on the surface of the test specimen containing the FSW joint providing an indication of the occurrence of pitting. The pits were independent of each other Metals 2020, 10, x FOR PEER REVIEW 7 of 22 i.e., EXCO solution, pitting-induced damage was favored to occur. With an increase in the duration of exposure to the aqueous environment, corrosion of the surface of the test specimen gradually increased and tended to become uniform, appearing to the naked eye as spalling corrosion. Results of the tests reveal that corrosion of the test specimen containing the FSW joint and test specimen of the base metal gradually increases with an increase in the duration of exposure to the aqueous environment. For the same duration of exposure to the aqueous environment, i.e., EXCO solution, the corrosion experienced by test specimen of aluminum alloy 2219 containing the FSW joint was noticeably less than that of the test specimen taken from the base metal, i.e., alloy 2219. The macroscopic corrosion morphology of the test specimen containing the FSW joint and test specimen Metalsof the2020 base, 10 metal,, 1437 under identical conditions, is shown in Figure 3. It is observed from (a) that when7 of 21 duration of exposure to the aqueous environment is 8 h, small pits were observed on the surface of the test specimen containing the FSW joint providing an indication of the occurrence of pitting. The but also easily distinguishable. The surface of the test specimen of the base metal revealed a dense pits were independent of each other but also easily distinguishable. The surface of the test specimen distribution of pits with the edges of a pit being difficult to distinguish. When duration of exposure to of the base metal revealed a dense distribution of pits with the edges of a pit being difficult to the aqueous environment (EXCO solution) was extended to 24 h, it can be observed from Figure3b that distinguish. When duration of exposure to the aqueous environment (EXCO solution) was extended pitting corrosion on the surface of the test specimen containing the FSW joint was visibly increased, to 24 h, it can be observed from Figure 3b that pitting corrosion on the surface of the test specimen while the extent and severity of corrosion of the test specimen taken from the base metal was noticeably containing the FSW joint was visibly increased, while the extent and severity of corrosion of the test observable, i.e., several areas of the test specimen revealed an evidence of flakes that had peeled off. specimen taken from the base metal was noticeably observable, i.e., several areas of the test specimen Therefore, a test specimen containing the FSW joint revealed much better corrosion resistance than test revealed an evidence of flakes that had peeled off. Therefore, a test specimen containing the FSW specimen taken from the base metal. joint revealed much better corrosion resistance than test specimen taken from the base metal.

FSW joint base metal (a)

FSW joint base metal (b)

Figure 3. A comparison of macroscopic morphology due to pure electrochemical corrosion:

(a) duration of exposure to environment is 8 h at applied stress (σ) of 0 MPa; (b) duration of exposure to environment is 24 h at applied stress (σ) of 0 MPa.

3.1.2. Microscopic Morphology of Pure Electrochemical Corrosion The friction stir welding process makes it possible for the weld zone to develop a precipitation free zone at and along the grain boundaries. Therefore, when the test specimen of aluminum alloy 2219 containing an FSW joint is exposed to the aqueous EXCO solution, the formation and presence of pits both at and along the grain boundary region is favored to occur. The morphology of corrosion, at the fine microscopic level, for the test specimen containing the FSW joint following 8 h of exposure to Metals 2020, 10, x FOR PEER REVIEW 8 of 22

Figure 3. A comparison of macroscopic morphology due to pure electrochemical corrosion: (a) duration of exposure to environment is 8 h at applied stress (σ) of 0 MPa; (b) duration of exposure to environmentMetals 2020, 10, x is FOR 24 PEERh at appliedREVIEW stress (σ) of 0 MPa. 8 of 22

3.1.2. MicroscopicFigure 3. A Morphology comparison of of macroscopic Pure Electrochemical morphology due Corrosion to pure electrochemical corrosion: (a) duration of exposure to environment is 8 h at applied stress (σ) of 0 MPa; (b) duration of exposure to The environmentfriction stir is welding 24 h at applied process stress makes (σ) of 0 it MPa. possibl e for the weld zone to develop a precipitation free zone at and along the grain boundaries. Therefore, when the test specimen of aluminum alloy 2219 3.1.2.containing Microscopic an FSW Morphology joint is exposed of Pure Electrochemical to the aqueous Corrosion EXCO solution, the formation and presence of pits bothThe at friction and along stir welding the grain process boundary makes region it possibl is favorede for the toweld occur. zone The to develop morphology a precipitation of corrosion, at thefree fine zone microscopic at and along level, the for grain the boundaries. test specimen Theref containingore, when the the FSWtest spec jointimen following of aluminum 8 h of alloyexposure Metalsto the22192020 aqueous, containing10, 1437 environment, an FSW joint i.e., is exposedEXCO solution, to the aqueous is shown EXCO in solution, Figure 4.the It formation can be seen and thatpresence there8 ofare 21 severalof pitspits both in atthe and exposed along the zone, grain boundaryall of which region are is favoredinitiated to occur.at the The grain morphology boundary. of corrosion, A change in morphologyat the fine of microscopic corrosion level,for the for test the testspecimens specimen containing containing the FSWFSW joint joint following and exposed 8 h of exposure to the EXCO the aqueousto the aqueous environment, environment, i.e., EXCO i.e., EXCO solution, solution, is shown is shown in Figure in Figure4. It 4. can It becan seen be seen that that there there are are several solution for different duration of times is shown in Figure 5. With an increase in duration of exposure pits inseveral the exposed pits in the zone, exposed all of zone, which all are of initiatedwhich are at initiated the grain at boundary.the grain boundary. A change Ain change morphology in to the aqueous EXCO solution results in the following: of corrosionmorphology for theof corrosion test specimens for the test containing specimens the containing FSW joint the and FSW exposed joint and toexposed the EXCO to the solution EXCO for di1. fferentThesolution magnitude duration for different of and times duration severity is shown of of ti mescorrosion in is Figure shown expe5 in. WithFigurerienced an 5. Withby increase the an test increase in specimens duration in duration ofincreases. exposure of exposure to the aqueous2. Theto the EXCOdensity aqueous solution of EXCO pits increases resultssolution in results with the following: the in the smaller following: pits gradually growing and developing to become larger1. The pits. magnitude and severity of corrosion experienced by the test specimens increases. 1. The magnitude and severity of corrosion experienced by the test specimens increases. 3. The2. Thegrain density boundaries of pits increases tend to with gradually the smaller dissolv pits graduallye indicating growing the andoccurrence developing of tointergranular become 2. The density of pits increases with the smaller pits gradually growing and developing to become corrosion.larger pits. 3.larger The pits. grain boundaries tend to gradually dissolve indicating the occurrence of intergranular 3. WhenThecorrosion. grainduration boundaries of exposure to tend the EXCO to gradually solution is 48 dissolve h, the overall indicating integrity the of the occurrence test specimen of surfaceintergranular is severely corrosion.damaged, which has a deleterious influence on overall performance of the test When duration of exposure to the EXCO solution is 48 h, the overall integrity of the test specimen specimensurface containing is severely the damaged, FSW joint. which has a deleterious influence on overall performance of the test specimen containing the FSW joint.

FigureFigure 4. MicroscopicMicroscopic 4. Microscopic morphology morphology morphology of of ofcorrosion corrosion corrosion experienced experienced by by test test by specimen testspecimen specimen of aluminum of aluminum of aluminum alloy alloy 2219 alloy2219 2219containingcontaining an FSW an an FSW FSWjoint. joint. joint.

Metals 2020, 10, x FOR PEER REVIEW 9 of 22 (a) (b)

(a) (b)

(c) (d)

FigureFigure 5. Influence 5. Influence of duration of duration of exposure of exposure to the environmentto the environment on morphology on morphology of corrosion of corrosion experienced by testexperienced specimens by of test aluminum specimens alloy of aluminum 2219 containing alloy 2219 the containing FSW joint: the ( aFSW) duration joint: (a of) duration exposure of= 8 h; (b) durationexposure of = exposure8 h; (b) duration= 16 h; (ofc) exposure duration = of 16 exposure h; (c) duration= 24 h; of ( dexposure) duration = 24 of h; exposure (d) duration= 48 of h. exposure = 48 h.

3.2. Analysis of Morphology of Stress Corrosion

3.2.1. Morphology of Stress Corrosion at the Macroscopic Level When the chosen test specimen is exposed to the EXCO solution and concurrently subjected to a tensile stress, stress corrosion is favored to occur for both the specimen containing the FSW joint and the specimen taken from the base metal. Stress corrosion is more dangerous than pure electrochemical corrosion, since the conjoint influence of stress and corrosion medium favor the initiation of pits, whose gradual growth and eventual coalescence results in the initiation of fine microscopic cracks. Furthermore, due to the lower tensile strength of the test specimen containing the FSW joint, it is easier to initiate fine microscopic cracks at the grain boundary region and to grow these into the grain eventually culminating in failure by fracture. This suggests that the formation, presence, and growth of fine microscopic cracks is far more dangerous culminating in failure of the component. However, it was difficult to observe the development and presence of pits and their ensuing growth through the depth, or thickness, of the test specimen and the concomitant initiation of fine microscopic cracks only by observing the macro corrosion morphology of the specimen. Therefore, from a macroscopic viewpoint corrosion experienced by test specimen containing the FSW joint was noticeably less than that of the base metal. A comparison of the macroscopic morphology of corrosion of the test specimen containing the FSW joint and test specimen of the base metal at an applied stress value of 79 MPa, and duration of exposure to the EXCO solution for 8 h and 24 h is shown in Figure 6. It is easily observed that the degree of environment-induced damage experienced by the test specimen of the base metal (i.e., aluminum alloy 2219) was more severe than the test specimen containing the FSW joint for a given duration of exposure to the aqueous environment, i.e., EXCO solution. Therefore, from macroscopic observation of the environment-induced damage, it is clear that specimens containing the FSW joint revealed better resistance to damage resulting from exposure to the aqueous environment.

3.2.2. Morphology of Stress Corrosion at the Microscopic Level For a test specimen, which experiences environment-induced damage, such as corrosion, the presence of pits and fine microscopic cracks can also be observed at the same time. Some of the fine microscopic cracks originate from the pits and often tend to extend along the grain boundaries. A few of these fine microscopic cracks will tend to grow through the grain. The conjoint influence of applied stress and exposure to an aggressive environment, such as EXCO solution, will promote the development of pits through the thickness, while concurrently promoting the initiation, coalescence, and propagation of cracks. Due to a gradual loss of the second-phase particles present both at and

Metals 2020, 10, 1437 9 of 21

When duration of exposure to the EXCO solution is 48 h, the overall integrity of the test specimen surface is severely damaged, which has a deleterious influence on overall performance of the test specimen containing the FSW joint.

3.2. Analysis of Morphology of Stress Corrosion

3.2.1. Morphology of Stress Corrosion at the Macroscopic Level When the chosen test specimen is exposed to the EXCO solution and concurrently subjected to a tensile stress, stress corrosion is favored to occur for both the specimen containing the FSW joint and the specimen taken from the base metal. Stress corrosion is more dangerous than pure electrochemical corrosion, since the conjoint influence of stress and corrosion medium favor the initiation of pits, whose gradual growth and eventual coalescence results in the initiation of fine microscopic cracks. Furthermore, due to the lower tensile strength of the test specimen containing the FSW joint, it is easier to initiate fine microscopic cracks at the grain boundary region and to grow these into the grain eventually culminating in failure by fracture. This suggests that the formation, presence, and growth of fine microscopic cracks is far more dangerous culminating in failure of the component. However, it was difficult to observe the development and presence of pits and their ensuing growth through the depth, or thickness, of the test specimen and the concomitant initiation of fine microscopic cracks only by observing the macro corrosion morphology of the specimen. Therefore, from a macroscopic viewpoint corrosion experienced by test specimen containing the FSW joint was noticeably less than that of the base metal. A comparison of the macroscopic morphology of corrosion of the test specimen containing the FSW joint and test specimen of the base metal at an applied stress value of 79 MPa, and duration of exposure to the EXCO solution for 8 h and 24 h is shown in Figure6. It is easily observed that the degree of environment-induced damage experienced by the test specimen of the base metal (i.e., aluminum alloy 2219) was more severe than the test specimen containing the FSW joint for a given duration of exposure to the aqueous environment, i.e., EXCO solution. Therefore, from macroscopic observation of the environment-induced damage, it is clear that specimens containing the FSW joint revealed better resistance to damage resulting from exposure to the aqueous environment.

3.2.2. Morphology of Stress Corrosion at the Microscopic Level For a test specimen, which experiences environment-induced damage, such as corrosion, the presence of pits and fine microscopic cracks can also be observed at the same time. Some of the fine microscopic cracks originate from the pits and often tend to extend along the grain boundaries. A few of these fine microscopic cracks will tend to grow through the grain. The conjoint influence of applied stress and exposure to an aggressive environment, such as EXCO solution, will promote the development of pits through the thickness, while concurrently promoting the initiation, coalescence, and propagation of cracks. Due to a gradual loss of the second-phase particles present both at and along the grain boundaries of the FSW joint, the intergranular region is relatively weak causing mechanical properties of the joint to progressively degrade. Therefore, for the same magnitude of applied stress and exposure to an aggressive aqueous environment, the test specimen with a friction stir welded joint is noticeably more susceptible to stress corrosion cracking than test specimen of the base metal. Metals 2020, 10, x FOR PEER REVIEW 10 of 22 along the grain boundaries of the FSW joint, the intergranular region is relatively weak causing mechanical properties of the joint to progressively degrade. Therefore, for the same magnitude of applied stress and exposure to an aggressive aqueous environment, the test specimen with a friction stirMetals welded2020, 10 ,joint 1437 is noticeably more susceptible to stress corrosion cracking than test specimen 10of ofthe 21 base metal.

FSW joint base metal (a)

FSW joint base metal (b)

Figure 6. AA comparison of of the macroscopic morphology of of damage resulting as a consequence of exposure to the environmentenvironment (EXCO solution): ( a)) duration duration of of exposure exposure = 88 h h at at an an applied applied stress stress ( (σσ)) of 79 MPa; ( b) duration of exposure = 2424 h h at at an an applied stress ( σ) of 79 MPa.

A comparison of the microscopic morphology of th thee test specimen containing an FSW joint with the test specimen ofof thethe as-providedas-provided aluminumaluminum alloy alloy upon upon exposure exposure to to the the EXCO EXCO solution solution for for full full 16 16 h hunder under the the action action of of a tensilea tensile stress stress of of 79 79 MPa MPa is shownis shown in in Figure Figure7. It7. is It observedis observed that that degradation degradation of ofthe the test test specimen specimen surface surface was was evident evident for both for specimens.both specimens. However, However, the extent the extent and severity and severity of surface of surfacedamage damage experienced experienced by test specimenby test specimen containing cont theaining FSW the joint FSW is noticeablyjoint is noticeably smaller smaller and reveals and revealsan overall an betteroverall resistance better resistance to damage to damage induced induced by the aqueous by the environmentaqueous environment (i.e., EXCO (i.e., solution) EXCO whensolution) compared when compared one-on-one one-on-one with test specimenwith test takenspecimen from taken the base from metal, the base i.e., metal, aluminum i.e., aluminum alloy 2219. alloyFurther, 2219. fine Further, microscopic fine microscopic cracks arising cracks from aris environment-induceding from environment-induced damage were damage found were on the found test onspecimen the test containingspecimen containing the FSW, as the shown FSW, inas Figureshown8 ,in but Figure no fine 8, but microscopic no fine microscopic cracks were cracks observed were observedon the surface on the of surface test specimen of test specimen taken from taken the from base the metal. base metal. The most The most appealing appealing rationale rationale for this for thisobservation observation is that is that upon upon exposure exposure to the to the same same aqueous aqueous medium medium (EXCO (EXCO solution) solution) and and same same level level of applied stress, the fine microscopic cracks initiated earlier in the test specimen containing the FSW joint due to its lower tensile strength than the test specimen taken from the base metal. Metals 2020, 10, x FOR PEER REVIEW 11 of 22 Metals 2020, 10, x FOR PEER REVIEW 11 of 22 of applied stress, the fine microscopic cracks initiated earlier in the test specimen containing the FSW Metals 2020, 10, 1437 11 of 21 jointof applied due to stress, its lower the finetensile microscopic strength than cracks the initia test tedspecimen earlier takenin the fromtest specimen the base metal.containing the FSW joint due to its lower tensile strength than the test specimen taken from the base metal.

(a) (b) (a) (b) Figure 7. A comparison of the morphology stress-corrosion-induced damage at the fine microscopic levelFigure for 7. the A comparisontest specimen of containing the morphology an FSW stress-corrosion-induced joint and test specimen damage of the base at the metal finefine (duration microscopic of exposurelevel for the= 16 testtest h， specimenspecimenσ = 79 MPa): containing (a) FSW an joint; FSW (b joint) base and metal. test specimen of the base metal (duration of exposure = 1616 h h,，σσ= = 7979 MPa):MPa): ((aa)) FSWFSW joint;joint; ((bb)) basebase metal. metal.

Figure 8. A fine microscopic micro-crack on the surface of test specimen containing an FSW joint. Figure 8. A fine microscopic micro-crack on the surface of test specimen containing an FSW joint. (Duration of exposure = 16 h, applied stress (σ) of 79 MPa). (DurationFigure 8. Aof fineexposure microscopic = 16 h, appliedmicro-crack stress on (σ the) of surface79 MPa). of test specimen containing an FSW joint. (Duration of exposure = 16 h, applied stress (σ) of 79 MPa). When the stress value increases to 118.5 MPa, fine microscopic cracks were observed on the When the stress value increases to 118.5 MPa, fine microscopic cracks were observed on the surfaceWhen of test the specimen stress value of the increases base metal to 118.5 following MPa, 16 fi hne of microscopic exposure to cracks the EXCO were solution, observed as shownon the surfacein Figure of9 a.test For specimen the test of specimen the base containing metal followin the FSWg 16 jointh of exposure under identical to the EXCO conditions, solution, more as serious shown insurface Figure of 9a. test For specimen the test ofspecimen the base containing metal followin the FSWg 16 hjoint of exposure under identical to the EXCO conditions, solution, more as serious shown stress-corrosion-inducedin Figure 9a. For the test specimen microscopic containing cracking the was FSW observed, joint under as identical shown inconditions, Figure9b; more the cracksserious stress-corrosion-induceddeveloped were significantly microscopic wider than cracking the cracks wa formeds observed, and present as shown on the in surface Figureof 9b; test the specimen cracks developedstress-corrosion-induced were significantly microscopic wider thancracking the cracwasks observed, formed andas shown present in onFigure the surface9b; the ofcracks test ofdeveloped the base were metal. significantly Under identical wider conditions,than the crac multipleks formed cracks and were present found on to the be surface distributed of test on specimenthe surface of ofthe the base test metal. specimen Under containing identical conditions, the FSW joint. multiple Therefore, cracks were under found identical to be distributed conditions, onspecimen the surface of the of base the metal.test specimen Under identical containing conditions, the FSW multiplejoint. Therefore, cracks were under found identical to be conditions,distributed theon the environment-induced surface of the test specimen damage experiencedcontaining th bye theFSW specimen joint. Therefore, containing under the identical FSW joint conditions, was more thesevere environment-induced than the specimen damage of the base experienced metal, i.e., by aluminum the specimen alloy containing 2219. The the most FSW appealing joint was reason more severethe environment-induced than the specimen of damage the base experienced metal, i.e., aluminumby the specimen alloy 2219. containing The most the appealingFSW joint reasonwas more for forsevere this than is that the tensile specimen strength of the of base the testmetal, specimen i.e., aluminum containing alloy the 2219. FSW The joint most is lower appealing than thereason tensile for thisstrength is that of tensile test specimen strength taken of the from test thespecimen base metal, containing and the the intergranular FSW joint is corrosionlower than experienced the tensile strengththis is that of testtensile specimen strength taken of the from test the specimen base me tal,containing and the theintergranular FSW joint corrosionis lower thanexperienced the tensile by bystrength the test of specimenstest specimen upon taken exposure from the to thebase EXCO metal, solution and the causesintergranular the grain corrosion boundary experienced to gradually by weakenthe test andspecimens concurrently upon reducingexposure theto bindingthe EXCO force solution between causes the grains, the grain so when boundary the test to specimen gradually is weakenthe test andspecimens concurrently upon exposurereducing theto thebinding EXCO force solution between causes the grains,the grain so whenboundary the test to graduallyspecimen subjectedweaken and to aconcurrently tensile stress, reducing it becomes the noticeablybinding force easier between to initiate the grains, fine microscopic so when the cracks test specimen along the grain boundary region . The cracks in the test specimens containing the FSW joint tend to progress

Metals 2020, 10, x FOR PEER REVIEW 12 of 22 Metals 2020, 10, 1437 12 of 21 is subjected to a tensile stress, it becomes noticeably easier to initiate fine microscopic cracks along the grain boundary region . The cracks in the test specimens containing the FSW joint tend to progress along the weakened grain boundaries resulting in conditions conducive for premature failure by along the weakened grain boundaries resulting in conditions conducive for premature failure by intergranular fracture. This is shown in Figure 10. Consequently, overall resistance of test specimen of intergranular fracture. This is shown in Figure 10. Consequently, overall resistance of test specimen aluminum alloy 2219 containing the FSW joint upon exposure to the EXCO solution is lower than that of aluminum alloy 2219 containing the FSW joint upon exposure to the EXCO solution is lower than of the base metal. that of the base metal.

(a) (b)

(c)

FigureFigure 9. 9.A A comparisoncomparison of the the stress-corrosion-related stress-corrosion-related dama damagege of test of testspecimen specimen of the of base the metal base and metal andtest test specimen specimen containing containing an FSW an FSW join joint.t. (Duration (Duration of exposure of exposure = 16 =h, 16σ = h, 118.5σ = 118.5MPa) MPa)showing: showing: (a) (a)microscopic microscopic crack crack on onsurface surface of base of base metal, metal, i.e., aluminum i.e., aluminum alloy 2219; alloy (b 2219;) microscopic (b) microscopic crack on surface crack on of test specimen containing the friction stir welded joint; (c) microscopic cracks on surface of specimen Metalssurface 2020, 10, of x FOR test specimenPEER REVIEW containing the friction stir welded joint; (c) microscopic cracks on surface of13 of 22 specimencontaining containing FSW joint. FSW joint.

Figure 10. Weakened grain boundaries initiate fine microscopic cracks under the influence of a tensile Figure 10. Weakened grain boundaries initiate fine microscopic cracks under the influence of a tensile stress (Duration of exposure = 16 h and applied stress (σ) of 118.5 MPa). stress (Duration of exposure = 16 h and applied stress (σ) of 118.5 MPa).

3.3. Analysis of Damage due to Corrosion

3.3.1. Model for Corrosion-Induced Damage The polished specimen had a flat surface prior to the initiation of environment-induced damage, i.e., corrosion. Subsequent to the highly localized damage induced by the environment by way of corrosion, both pits and cracks were observed on the surface of the test specimen. The gradual loss of material from the surface was used to express damage experienced by the specimen intuitively. By using a laser displacement sensor, the data specific to change in depth of the test specimen surface were obtained. If the original height of the test specimen surface is taken to be zero, then surface height of the test specimen following exposure to the aqueous environment (i.e., EXCO solution) should be either less than or equal to 0 due to the damage. The height value at each point on the test specimen surface does provide a measure of both the extent and severity of environment-induced damage at that point. Therefore, the damage resulting as a direct consequence of exposure of the test specimen to the aqueous environment can be established by the change in surface depth caused by exposure to the environment and the resultant damage due to corrosion. The laser measurement data come from the longitudinal symmetry centerline of the specimen. For the joint specimen, this is the location of the NZ. The average distance between the measurement points is 5 μm, and the line connecting these points indicates the change in the corrosion depth on the measured surface. The corrosion depth curves for the test specimen containing the FSW joint for different duration of exposure to the chosen environment, i.e., EXCO solution, safely reflects the amount of environment- induced damage for the test specimen containing the FSW joint. The change curve of stress corrosion depth taken on a symmetrical section of the test specimen containing the FSW joint, along the direction of applied stress, following exposure to the EXCO solution for 8 h, 16 h, 24 h, and 48 h under the influence of an applied stress of 118.5 MPa, is shown in Figure 11. Since measurement of the depth was made along the longitudinal symmetry plane, the curves also provide information related to the damaged section of the nugget zone. It can be seen from this figure that with increased duration of exposure to the chosen aqueous environment, i.e., EXCO solution, the amount of damage increases.

Metals 2020, 10, 1437 13 of 21

3.3. Analysis of Damage Due to Corrosion

3.3.1. Model for Corrosion-Induced Damage The polished specimen had a flat surface prior to the initiation of environment-induced damage, i.e., corrosion. Subsequent to the highly localized damage induced by the environment by way of corrosion, both pits and cracks were observed on the surface of the test specimen. The gradual loss of material from the surface was used to express damage experienced by the specimen intuitively. By using a laser displacement sensor, the data specific to change in depth of the test specimen surface were obtained. If the original height of the test specimen surface is taken to be zero, then surface height of the test specimen following exposure to the aqueous environment (i.e., EXCO solution) should be either less than or equal to 0 due to the damage. The height value at each point on the test specimen surface does provide a measure of both the extent and severity of environment-induced damage at that point. Therefore, the damage resulting as a direct consequence of exposure of the test specimen to the aqueous environment can be established by the change in surface depth caused by exposure to the environment and the resultant damage due to corrosion. The laser measurement data come from the longitudinal symmetry centerline of the specimen. For the joint specimen, this is the location of the NZ. The average distance between the measurement points is 5 µm, and the line connecting these points indicates the change in the corrosion depth on the measured surface. The corrosion depth curves for the test specimen containing the FSW joint for different duration of exposure to the chosen environment, i.e., EXCO solution, safely reflects the amount of environment-induced damage for the test specimen containing the FSW joint. The change curve of stress corrosion depth taken on a symmetrical section of the test specimen containing the FSW joint, along the direction of applied stress, following exposure to the EXCO solution for 8 h, 16 h, 24 h, and 48 h under the influence of an applied stress of 118.5 MPa, is shown in Figure 11. Since measurement of the depth was made along the longitudinal symmetry plane, the curves also provide information related to the damaged section of the nugget zone. It can be seen from this figure that with increased duration of exposure to the chosen aqueous environment, i.e., EXCO solution, the amount of damage increases.

3.3.2. An Analysis of Damage Due to Electrochemical Corrosion The fitting curve depicting the average depth of corrosion for the two chosen specimens with duration of exposure to the aqueous environment (EXCO solution) is shown in Figure 12. The fitting curve depicting maximum depth of corrosion experienced by the two types of chosen test specimens with duration of exposure to the aqueous environment (EXCO solution) under conditions that promote pure electrochemical corrosion is shown in Figure 13. To avoid the influence of random factors, the data provided here are the average of three replicates. In an attempt to compare the risk of damage, the ratio (α) of maximum depth to the average depth is chosen for the purpose of comparison. A change in the ratio of α (maximum depth/average depth) for the two types of specimens over duration of expourse is shown in Figure 14. Here each "maximum corrosion depth" represents the average value of the maximum corrosion depth of three parallel specimens, and each “average corrosion depth” is the average value of the average corrosion depth of the three parallel specimens. α is the ratio of the mean of the maximum depth to the mean of the average depth. It is observed that the value of α for the two chosen specimens, i.e., (i) test specimen containing the FSW joint and (ii) test specimen of the base metal, i.e., aluminum alloy 2219, was 1.25. This helps us to conclude that the extent of damage experienced by the specimens is roughly the same. For the four chosen duration of exposure to the aqueous environment, i.e., EXCO solution, the average depth of corrosion and maximum depth of corrosion for the test containing the FSW joint is noticeably less than the corresponding values for the base metal, i.e., AA2219. This leads to the conclusion that when pure electrochemical corrosion occurs upon exposure to the EXCO solution, the environment-induced damage, i.e., corrosion, experienced by test specimen of aluminum alloy 2219 containing an FSW joint, is less than the damage experienced by the base metal, i.e., aluminum alloy 2219. This suggests that overall corrosion resistance of test Metals 2020, 10, 1437 14 of 21 specimen containing the FSW joint is noticeably improved when compared one-on-one with the base metal. " # hmax Metals 2020, 10, x FOR PEER REVIEW α = ,14 of (1) 22 have

(a)

(b)

(c)

(d)

Figure 11. CorrosionCorrosion damage damage of of the the longitudinal longitudinal section section of oftest test specimen specimen containing containing an anFSW FSW joint: joint: (a) (stressa) stress corrosion corrosion damage damage profile profile model model of FSW of joint FSW specimen joint specimen (duration (duration of exposure of exposure = 8 h, σ == 118.58 h, σMPa);= 118.5 (b) stress MPa); corrosion (b) stress damage corrosion profile damage model profile of FSW model joint specimen of FSW joint(duratio specimenn of exposure (duration = 16 ofh, exposureσ = 118.5 MPa);= 16 h,(c)σ stress= 118.5 corrosion MPa); damage (c) stress profile corrosion model damage of FSW profile joint specimen model of (duration FSW joint of specimen exposure (duration= 24 h, σ = of 118.5 exposure MPa);= (d24) stress h, σ = corrosion118.5 MPa); damage (d) stress profile corrosion model damageof FSW joint profile specimen model of(duration FSW joint of specimenexposure = (duration 48 h, σ = of118.5 exposure MPa). = 48 h, σ = 118.5 MPa).

3.3.2. An Analysis of Damage due to Electrochemical Corrosion The fitting curve depicting the average depth of corrosion for the two chosen specimens with duration of exposure to the aqueous environment (EXCO solution) is shown in Figure 12. The fitting curve depicting maximum depth of corrosion experienced by the two types of chosen test specimens

Metals 2020, 10, x FOR PEER REVIEW 15 of 22 with duration of exposure to the aqueous environment (EXCO solution) under conditions that promote pure electrochemical corrosion is shown in Figure 13. To avoid the influence of random factors, the data provided here are the average of three replicates. In an attempt to compare the risk of damage, the ratio () of maximum depth to the average depth is chosen for the purpose of comparison. A change in the ratio of (maximum depth/average depth) for the two types of specimens over duration of expourse is shown in Figure 14. Here each "maximum corrosion depth" represents the average value of the maximum corrosion depth of three parallel specimens, and each “average corrosion depth” is the average value of the average corrosion depth of the three parallel specimens. α is the ratio of the mean of the maximum depth to the mean of the average depth. It is observed that the value of for the two chosen specimens, i.e., (i) test specimen containing the FSW joint and (ii) test specimen of the base metal, i.e., aluminum alloy 2219, was 1.25. This helps us to conclude that the extent of damage experienced by the specimens is roughly the same. For the four chosen duration of exposure to the aqueous environment, i.e., EXCO solution, the average depth of corrosion and maximum depth of corrosion for the test containing the FSW joint is noticeably less than the corresponding values for the base metal, i.e., AA2219. This leads to the conclusion that when pure electrochemical corrosion occurs upon exposure to the EXCO solution, the environment- induced damage, i.e., corrosion, experienced by test specimen of aluminum alloy 2219 containing an FSW joint, is less than the damage experienced by the base metal, i.e., aluminum alloy 2219. This suggests that overall corrosion resistance of test specimen containing the FSW joint is noticeably improved when compared one-on-one with the base metal.

Metals 2020, 10, 1437 ℎ 15 of 21 = , (1) ℎ

Base metal y = Intercept + FSW Equation B1*x^1 + B2*x^ 50 2 Polynomial fit of Base metal Weight No Weighting Residual Sum 4.41989 0.24931 Polynomial fit of FSW of Squares 45 Adj. R-Square 0.98338 0.99877 Value Standard Error Average depth Intercept -3.2663 4.34888 40 Average depth B1 1.47434 0.38302 Average depth B2 -0.00924 0.00646 Average depth Intercept -1.48425 1.03286 35 Average depth B1 1.13602 0.09097 Average depth B2 -0.00541 0.00153

30

25

20

Average depth (μm) depth Average 15

10

5

0 8 16 24 32 40 48 Duration of exposure (h) Metals 2020, 10, x FOR PEER REVIEW 16 of 22 Figure 12. Fitting curve showing the variation of average depth of corrosion with duration of exposure MetalsFigure 2020, 10 12., x FOR Fitting PEER curve REVIEW showing the variation of average depth of corrosion with duration of16 of 22 to the aqueous environment (EXCO solution) at an applied stress (σ) of 0 MPa. exposure to the aqueous environmenty = Intercept + (EXCO solution) at an applied Base stress metal (σ) of 0 MPa. Equation B1*x^1 + B2*x^ 2 FSW 60 Weight No Weighting Polynomial Base metal fit of Base metal Residual Sum y = Intercept6.52475 + 2.30721 of EquationSquares B1*x^1 + B2*x^ Polynomial fit of FSW 2 FSW 55 Adj. R-Square 0.98131 0.99297 60 Weight No Weighting Value Standard Error Polynomial fit of Base metal MaximumResidual depth Sum Intercept6.52475 0.50055 2.30721 5.28389 of Squares Polynomial fit of FSW 50 Maximum depth B1 1.49589 0.46537 55 Adj. R-Square 0.98131 0.99297 Maximum depth B2 -0.00722 0.00785 Value Standard Error Maximum depth Intercept -2.0721 3.14206 45 Maximum depth Intercept 0.50055 5.28389 Maximum depth B1 1.37686 0.27673 50 Maximum depth B1 1.49589 0.46537 Maximum depth B2 -0.00568 0.00467 40 Maximum depth B2 -0.00722 0.00785 45 Maximum depth Intercept -2.0721 3.14206 Maximum depth B1 1.37686 0.27673 35 Maximum depth B2 -0.00568 0.00467 40 30 35 25 30 20 25 Maximum depth (μm) 15 20

Maximum depth (μm) 10 15 5 10

50 8 16 24 32 40 48 Duration of exposure (h) 0 8 16 24 32 40 48 Duration of exposure (h) Figure 13. Variation of maximum depth due to corrosion-induced damage with duration of exposure Figure 13. Variation of maximum depth due to corrosion-induced damage with duration of exposure toFigure the environment 13. Variation at of an maximum applied stress depth (σ due) of to0 MPa.corrosion-induced damage with duration of exposure to the environment at an applied stress (σ) of 0 MPa. to the environment at an applied stress (σ) of 0 MPa. Equation y = a + b*x Base metal Weight No Weighting Residual Sum 0.07062 0.00512 FSW of Squares Equation y = a + b*x 2.0 Pearson's r -0.43474 0.65503 Linear fit of base metal α No Weighting Base metal Adj.Weight R-Square -0.2165 0.1436 Residual Sum 0.07062Value 0.00512 Standard Error Linear FSW fit of FSW α of Squares α Intercept 1.38536 0.17756 2.0 Pearson's r -0.43474 0.65503 Linear fit of base metal α α Slope -0.00429 0.00628 Adj. R-Square -0.2165 0.1436 α Intercept 1.16345 0.04779 Value Standard Error Linear fit of FSW α α Slope 0.00207 0.00169 α Intercept 1.38536 0.17756 1.5 α Slope -0.00429 0.00628 α Intercept 1.16345 0.04779 α Slope 0.00207 0.00169 1.5

1.0 α

1.0 α

0.5

0.5

0.0 0 8 16 24 32 40 48 0.0 Duration of exposure (h) 0 8 16 24 32 40 48 Figure 14. Variation of α withDuration duration of exposure of exposure (h) at applied stress (σ) of 0 MPa. Figure 14. Variation of α with duration of exposure at applied stress (σ) of 0 MPa. Figure 14. Variation of α with duration of exposure at applied stress (σ) of 0 MPa. 3.3.3. A Comparative Analysis of Damage due to Stress Corrosion 3.3.3.When A Comparative the test specimen Analysis is subjected of Damage to thedue conjoint to Stress influence Corrosion of stress and exposure to a corrosive medium,When i.e., the EXCO test specimen solution, isthe subjected occurrence to the of SCCconjoint is favored. influence The of conjoint stress and and exposure mutually to interactivea corrosive influencesmedium, i.e., of both EXCO mechanical solution, theeffect occurrence (applied stress)of SCC and is favored. chemical The effect conjoint (i.e., exposureand mutually to the interactive aqueous environmentinfluences of (EXCOboth mechanical solution) effectcauses (applied the material stress) to and be severely chemical damaged effect (i.e., with exposure resultant to the damage aqueous to theenvironment components. (EXCO Variation solution) of the causes average the materialcorrosion to depth be severely for the damaged test specimen with resultantof aluminum damage alloy to 2219the components. containing the Variation FSW joint of theand average test specimen corrosion of thedepth as-provided for the test aluminum specimen alloyof aluminum 2219, i.e., alloy the base2219 metal,containing with theduration FSW joint of exposure and test specimento the en vironment,of the as-provided is shown aluminum in Figure alloy 15. 2219,Variation i.e., theof maximumbase metal, corrosion with duration depth forof exposurethe test specimen to the en ofvironment, alloy 2219 is containingshown in theFigure FSW 15. joint Variation and test of specimenmaximum of corrosion the base depthmetal specimenfor the test with specimen duration of ofalloy exposure 2219 containingat an applied the stressFSW jointof 79 andMPa test is shownspecimen in Figure of the 16.base The metal average specimen depth with of corrosion duration experienced of exposure by at the an testapplied specimen stress of of aluminum 79 MPa is alloyshown 2219 in containingFigure 16. Thethe FSWaverage joint de ispth noticeably of corrosion smaller experienced than for test by specimenthe test specimen of the base of metal,aluminum i.e., aluminumalloy 2219 containingalloy 2219. theThe FSW variation joint is of noticeably ratio of maximum smaller than corrosion for test depth specimen to the of theaverage base metal,corrosion i.e., depthaluminum (α) for alloy test 2219.specimens The variation of aluminum of ratio alloy of 2219maximum containing corrosion the FSW depth joint to andthe averagetest specimens corrosion of thedepth base (α metal) for test (i.e., specimens aluminum of alloy aluminum 2219), underalloy 2219 an applied containing stress the of FSW 79 MPa, joint is and shown test inspecimens Figure 17. of Itthe is baseseen metalthat the (i.e., ratio aluminum (α) for the alloy test 2219), specimens under containing an applied the stress FSW of joint79 MPa, is noticeably is shown greaterin Figure than 17. thatIt is forseen test that specimen the ratio taken (α) for from the the test base specimens metal, fo containingr all of the the four FSW different joint isdurations noticeably of exposuregreater than to that for test specimen taken from the base metal, for all of the four different durations of exposure to

Metals 2020, 10, 1437 16 of 21

3.3.3. A Comparative Analysis of Damage Due to Stress Corrosion When the test specimen is subjected to the conjoint influence of stress and exposure to a corrosive medium, i.e., EXCO solution, the occurrence of SCC is favored. The conjoint and mutually interactive influences of both mechanical effect (applied stress) and chemical effect (i.e., exposure to the aqueous environment (EXCO solution) causes the material to be severely damaged with resultant damage to the components. Variation of the average corrosion depth for the test specimen of aluminum alloy 2219 containing the FSW joint and test specimen of the as-provided aluminum alloy 2219, i.e., the base metal, with duration of exposure to the environment, is shown in Figure 15. Variation of maximum corrosion depth for the test specimen of alloy 2219 containing the FSW joint and test specimen of the base metal specimen with duration of exposure at an applied stress of 79 MPa is shown in Figure 16. The average depth of corrosion experienced by the test specimen of aluminum alloy 2219 containing the FSW joint is noticeably smaller than for test specimen of the base metal, i.e., aluminum alloy 2219. The variation of ratio of maximum corrosion depth to the average corrosion depth (α) for test specimens of aluminum alloy 2219 containing the FSW joint and test specimens of the base metal Metals 2020(i.e.,, 10, aluminumx FOR PEER alloyREVIEW 2219), under an applied stress of 79 MPa, is shown in Figure 17. It is seen17 that of 22 the ratio (α) for the test specimens containing the FSW joint is noticeably greater than that for test the aqueousspecimen environment taken from the(EXCO base metal,solution). for all The of the reas fouron di ffforerent this durations is that of when exposure a totest the specimen aqueous containingenvironment the FSW (EXCOjoint specimen solution). is The under reason stress, for this the is thatpits when initiated a test both specimen at and containing along the the grain FSW boundaryjoint regions specimen are is more under likely stress, theto develop pits initiated and both grow at andthrough along thethe grain thickness. boundary Consequently, regions are more the damage likelyexperienced to develop by test and specimens grow through containing the thickness. the FSW Consequently, joint is observably the damage more experienced serious than by test the damage specimensexperienced containing by test thespecimens FSW joint of is observablythe as-provided more serious aluminum than thealloy damage 2219. experienced Furthermore, by test the specimens of the as-provided aluminum alloy 2219. Furthermore, the ratio [α] for the two chosen test ratio for the two chosen test specimens increases with an increase in duration of exposure to the specimens increases with an increase in duration of exposure to the chosen aqueous environment, i.e., chosen aqueous environment, i.e., EXCO solution. The increase is noticeably rapid for the specimen EXCO solution. The increase is noticeably rapid for the specimen containing the FSW joint, while containingit is the relatively FSW joint, slow forwhile the it test is specimenrelatively prepared slow for from the thetest base specimen metal, i.e.,prepared aluminum from alloy the 2219.base metal, i.e.,This aluminum clearly indicates alloy that 2219. the extentThis clearly and severity indicates of damage that experiencedthe extent byand the severity specimen of containing damage experiencedan FSW by jointthe specimen is noticeably containing more. an FSW joint is noticeably more.

y = Intercept + B Base metal Equation 1*x^1 + B2*x^2 FSW joint 55 Weight No Weighting Polynomial fit of Base metal Residual Sum 3.74099 0.48133 of Squares Polynomial fit of FSW Adj. R-Square 0.9843 0.99754 50 Value Standard Error Average corrosi Intercept 0.87901 4.00097 Average corrosi B1 1.90615 0.35238 45 Average corrosi B2 -0.01778 0.00594 Average corrosi Intercept 1.73812 1.43513 Average corrosi B1 1.73522 0.1264 Average corrosi B2 -0.01623 0.00213 40

35

30

25

20 Average corrosion depth (μm) 15

10 0 8 16 24 32 40 48 Duration of exposure (h) Figure 15. Variation of average corrosion depth (µm) with duration of exposure (h) to the environment Figure 15. Variation of average corrosion depth (μm) with duration of exposure (h) to the (EXCO solution) at an applied stress (σ) of 79 MPa. environment (EXCO solution) at an applied stress (σ) of 79 MPa.

Base metal y = Intercept + B Equation 1*x^1 + B2*x^2 FSW joint

Weight No Weighting Polynomial fit of base metal Residual Sum 16.80663 16.03989 160 of Squares Polynomial fit of FSW Adj. R-Square 0.98871 0.99447 Value Standard Error Maximum corro Intercept -2.52928 8.48031 140 Maximum corro B1 4.20528 0.74689 Maximum corro B2 -0.03447 0.0126 Maximum corro Intercept -3.5663 8.28461 Maximum corro B1 4.50559 0.72966 120 Maximum corro B2 -0.02408 0.01231

100

80

60

Maximum corrosion depth (μm) Maximum 40

20 0 8 16 24 32 40 48 Duration of exposure (h)

Figure 16. Variation of maximum corrosion depth (μm) with time of exposure (h) to the environment at an applied stress (σ) of 79 MPa.

Metals 2020, 10, x FOR PEER REVIEW 17 of 22

the aqueous environment (EXCO solution). The reason for this is that when a test specimen containing the FSW joint specimen is under stress, the pits initiated both at and along the grain boundary regions are more likely to develop and grow through the thickness. Consequently, the damage experienced by test specimens containing the FSW joint is observably more serious than the damage experienced by test specimens of the as-provided aluminum alloy 2219. Furthermore, the ratio for the two chosen test specimens increases with an increase in duration of exposure to the chosen aqueous environment, i.e., EXCO solution. The increase is noticeably rapid for the specimen containing the FSW joint, while it is relatively slow for the test specimen prepared from the base metal, i.e., aluminum alloy 2219. This clearly indicates that the extent and severity of damage experienced by the specimen containing an FSW joint is noticeably more.

y = Intercept + B Base metal Equation 1*x^1 + B2*x^2 FSW joint 55 Weight No Weighting Polynomial fit of Base metal Residual Sum 3.74099 0.48133 of Squares Polynomial fit of FSW Adj. R-Square 0.9843 0.99754 50 Value Standard Error Average corrosi Intercept 0.87901 4.00097 Average corrosi B1 1.90615 0.35238 45 Average corrosi B2 -0.01778 0.00594 Average corrosi Intercept 1.73812 1.43513 Average corrosi B1 1.73522 0.1264 Average corrosi B2 -0.01623 0.00213 40

35

30

25

20 Average corrosion depth (μm) 15

10 0 8 16 24 32 40 48 Duration of exposure (h)

Figure 15. Variation of average corrosion depth (μm) with duration of exposure (h) to the Metalsenvironment2020, 10, 1437 (EXCO solution) at an applied stress (σ) of 79 MPa. 17 of 21

Base metal y = Intercept + B Equation 1*x^1 + B2*x^2 FSW joint

Weight No Weighting Polynomial fit of base metal Residual Sum 16.80663 16.03989 160 of Squares Polynomial fit of FSW Adj. R-Square 0.98871 0.99447 Value Standard Error Maximum corro Intercept -2.52928 8.48031 140 Maximum corro B1 4.20528 0.74689 Maximum corro B2 -0.03447 0.0126 Maximum corro Intercept -3.5663 8.28461 Maximum corro B1 4.50559 0.72966 120 Maximum corro B2 -0.02408 0.01231

100

80

60

Maximum corrosion depth (μm) Maximum 40

20 0 8 16 24 32 40 48 Duration of exposure (h) Metals 2020,Figure 10, x FOR 16. PEERVariation REVIEW of maximum corrosion depth (µm) with time of exposure (h) to the environment 18 of 22 Figureat an16. applied Variation stress of (maximumσ) of 79 MPa. corrosion depth (μm) with time of exposure (h) to the environment

at an applied stress (σy = )Intercept of +79 B MPa. Equation 1*x^1 + B2*x^2 Base metal Weight No Weighting FSW joint Residual Sum 4.10457E-4 0.00237 3.5 of Squares Polynomial fit of Base metal Adj. R-Square 0.98498 0.99008 Value Standard Error Polynomial fit of FSW abase Intercept 1.79417 0.04191 abase B1 0.01998 0.00369 abase B2 -1.82612E-4 6.22535E-5 3.0 aFSW Intercept 1.87667 0.10068 aFSW B1 0.03661 0.00887 aFSW B2 -1.44652E-4 1.49554E-4

2.5 α 2.0

1.5

1.0 0 8 16 24 32 40 48 Duration of exposure (h) Figure 17. Variation of ratio of maximum depth to average depth with time of exposure to the chosen Figureenvironment 17. Variation at an of appliedratio of stress maximum (σ) of 79 depth MPa. to average depth with time of exposure to the chosen environment at an applied stress (σ) of 79 MPa. 3.3.4. Relationship between Extent of Damage and Applied Stress 3.3.4. RelationshipStress is one between of the basic Extent factors of Damage that exerts and an Applied influence Stress on both the initiation and progression, Stressi.e., growth, is one of of damage the basic due factors to stress that corrosion. exerts Thean influence results of thison both study the clearly initiation indicate and the progression, existence of a relationship between the level of applied stress and the damage induced by exposure to the i.e., growth, of damage due to stress corrosion. The results of this study clearly indicate the existence aqueous environment, i.e., EXCO solution, by way of corrosion. For an exposure time to the aqueous of a relationship between the level of applied stress and the damage induced by exposure to the environment of 48 h, the variation of average corrosion depth with applied stress for the two chosen aqueoustest specimensenvironment, (i.e., (i)i.e., test EXCO specimen solution, containing by way the of FSW corrosion. joint and For (ii) an test exposure specimen time of the to base the metal, aqueous environmenti.e., aluminum of 48 alloyh, the 2219) variation is shown of average in Figure corro 18. Itsion is seen depth that with the average applied corrosion stress for depth the fortwo both chosen test specimensspecimens (i.e., increases (i) test with specimen an increase containing in the level the of FSW applied joint stress and ( σ(ii)), thereby test specimen indicating of thethe depthbase metal, of i.e., aluminum alloy 2219) is shown in Figure 18. It is seen that the average corrosion depth for both specimens increases with an increase in the level of applied stress (σ), thereby indicating the depth of corrosion-induced damage to have a positive correlation with the magnitude of applied stress. At the three chosen applied stress (σ) levels of (i) 0 MPa, (ii) 79 MPa, and (iii) 118.5 MPa, the average depth of environment-induced damage, i.e., corrosion, experienced by the base metal is greater than that for the test specimen containing the FSW joint. This clearly indicates that overall corrosion resistance of the test specimen containing the joint reveals an improvement. Comparing the maximum depth of corrosion-induced damage for the two kinds of test specimens, i.e., test specimen containing the FSW joint and test specimen of the base metal, it is seen that at a value of applied stress of 0 MPa, i.e. pure electrochemical corrosion, the maximum corrosion depth experienced by specimen of the base metal is greater than that for test specimen containing the FSW joint. However, at applied stress levels of 79 MPa and 118.5 MPa, the maximum depth of corrosion for test specimens containing the FSW joint is greater than that of the base metal, as shown in Figure 19. The most appealing reason for this is that, although the FSW process improves the overall resistance to both corrosion and corrosion-induced damage of the specimen subject to pure electrochemical corrosion, the overall resistance to stress corrosion of the test specimen containing the FSW joint decreases due to an observable decrease in mechanical properties. This is shown in Figure 20. For pure electrochemical corrosion, the α values for the two chosen test specimens are the same, while for the case of stress corrosion, the α values for the two types of specimens chosen reveal an increase, with α values for the specimen containing the FSW joint increasing noticeably faster, providing evidence for the overall detrimental influence of the welded joint to stress corrosion.

Metals 2020, 10, 1437 18 of 21 corrosion-induced damage to have a positive correlation with the magnitude of applied stress. At the three chosen applied stress (σ) levels of (i) 0 MPa, (ii) 79 MPa, and (iii) 118.5 MPa, the average depth of environment-induced damage, i.e., corrosion, experienced by the base metal is greater than that for the test specimen containing the FSW joint. This clearly indicates that overall corrosion resistance of the test specimen containing the joint reveals an improvement. Comparing the maximum depth of corrosion-induced damage for the two kinds of test specimens, i.e., test specimen containing the FSW joint and test specimen of the base metal, it is seen that at a value of applied stress of 0 MPa, i.e., pure electrochemical corrosion, the maximum corrosion depth experienced by specimen of the base metal is greater than that for test specimen containing the FSW joint. However, at applied stress levels of 79 MPa and 118.5 MPa, the maximum depth of corrosion for test specimens containing the FSW joint is greater than that of the base metal, as shown in Figure 19. The most appealing reason for this is that, although the FSW process improves the overall resistance to both corrosion and corrosion-induced damage of the specimen subject to pure electrochemical corrosion, the overall resistance to stress corrosion of the test specimen containing the FSW joint decreases due to an observable decrease in mechanical properties. This is shown in Figure 20. For pure electrochemical corrosion, the α values for the two chosen test specimens are the same, while for the case of stress corrosion, the α values for the two types of specimens chosen reveal an increase, with α values for the specimen containing the FSW joint increasing noticeably faster, providing evidence for the overall detrimental influence of the Metalswelded 2020 joint, 10, x to FOR stress PEER corrosion. REVIEW 19 of 22

Base metal FSW joint

80

70

60

50

40

Average corrosion depth (μm) depth corrosion Average 30

20 0 20406080100120 Stress level (MPa) Figure 18. Variation of the average corrosion depth (µm) with level of applied stress (MPa) for the two Figure 18. Variation of the average corrosion depth (μm) with level of applied stress (MPa) for the chosen specimens of aluminum alloy 2219. Duration of exposure to environment is 48 h. two chosen specimens of aluminum alloy 2219. Duration of exposure to environment is 48 h.

y = Intercept + B Base metal Equation 1*x^1 + B2*x^2 FSW joint 300 Weight No Weighting Residual Sum 00 of Squares Adj. R-Square -- -- Value Standard Error Maximum depth Intercept 55.8 -- 250 Maximum depth B1 0.12447 -- Maximum depth B2 0.00866 -- Maximum depth Intercept 51 -- Maximum depth B1 0.48101 -- Maximum depth B2 0.0109 -- 200

150

100 Maximumdepth (μm)

50

0 20406080100120 Stress level (MPa)

Figure 19. Variation of maximum corrosion depth (μm) with applied stress level (MPa) for the two chosen specimens of aluminum alloy 2219. Duration of exposure to environment is 48 h.

Metals 2020, 10, x FOR PEER REVIEW 19 of 22

Base metal FSW joint

80

70

60

50

40

Average corrosion depth (μm) depth corrosion Average 30

20 0 20406080100120 Stress level (MPa)

Figure 18. Variation of the average corrosion depth (μm) with level of applied stress (MPa) for the Metals 2020, 10, 1437 19 of 21 two chosen specimens of aluminum alloy 2219. Duration of exposure to environment is 48 h.

y = Intercept + B Base metal Equation 1*x^1 + B2*x^2 FSW joint 300 Weight No Weighting Residual Sum 00 of Squares Adj. R-Square -- -- Value Standard Error Maximum depth Intercept 55.8 -- 250 Maximum depth B1 0.12447 -- Maximum depth B2 0.00866 -- Maximum depth Intercept 51 -- Maximum depth B1 0.48101 -- Maximum depth B2 0.0109 -- 200

150

100 Maximumdepth (μm)

50

0 20406080100120 Stress level (MPa) Figure 19. Variation of maximum corrosion depth (µm) with applied stress level (MPa) for the two MetalsFigure 2020, 10 19., x FORVariation PEER REVIEWof maximum corrosion depth (μm) with applied stress level (MPa) for the two20 of 22 chosen specimens of aluminum alloy 2219. Duration of exposure to environment is 48 h. chosen specimens of aluminum alloy 2219. Duration of exposure to environment is 48 h. Base metal y = Intercept + B Equation 1*x^1 + B2*x^2 FSW joint 5 Weight No Weighting Residual Sum 00 of Squares Adj. R-Square -- -- Value Standard Error α Intercept 1.21 -- 4 α B1 0.00861 -- α B2 7.05015E-5 -- α Intercept 1.25616 -- α B1 0.02254 -- α B2 4.19101E-5 -- 3 α 2

1

0 050100 Stress level (MPa) Figure 20. Variation of α with the level of applied stress (MPa) for the two types of test specimens of Figure 20. Variation of α with the level of applied stress (MPa) for the two types of test specimens of aluminum alloy 2219 for a duration of exposure be 48 h. aluminum alloy 2219 for a duration of exposure be 48 h. 4. Conclusions 4. Conclusions The following are the key conclusions resulting from this exhaustive study. The following are the key conclusions resulting from this exhaustive study. 1. When pure electrochemical corrosion occurs, the average corrosion depth and maximum corrosion 1. When pure electrochemical corrosion occurs, the average corrosion depth and maximum depth of test specimen containing the joint are less than that of the test specimen of the base corrosion depth of test specimen containing the joint are less than that of the test specimen of the metal for four different durations of exposure to the aqueous environment. This suggests that the base metal for four different durations of exposure to the aqueous environment. This suggests overall corrosion resistance of test specimens containing an FSW joint is better than that of the that the overall corrosion resistance of test specimens containing an FSW joint is better than that base metal. of the base metal. 2. When stress corrosion occurs, due to reduced mechanical properties of the test specimen containing 2. When stress corrosion occurs, due to reduced mechanical properties of the test specimen the FSW joint, the stress corrosion resistance is inferior to that of the base metal, i.e., aluminum alloy containing the FSW joint, the stress corrosion resistance is inferior to that of the base metal, i.e., 2219, and fine microscopic cracks are favored to occur. aluminum alloy 2219, and fine microscopic cracks are favored to occur. 3. When stress corrosion occurs, the average corrosion depth of test specimen of the alloy containing an FSW joint is less than that of the base metal. The maximum depth of corrosion is greater than that of the base metal. This indicates that the test specimen containing the FSW joint is more susceptible to damage and degradation than the base metal upon exposure to the same aqueous environment. 4. When stress corrosion occurs, the ratio of maximum depth to average depth (denoted as α) increases with an increase in applied stress and duration of exposure to the aqueous environment (EXCO solution). 5. The ratio (α) for test specimens containing the FSW joint is significantly higher than that for specimens of the base metal. Further, the observed increase in α with applied stress is faster for the test specimens containing an FSW joint than for test specimens of the base metal.

Author Contributions: Methodology, S.L.; validation, S.L. and X.G.; formal analysis, S.L.; investigation, Z.L.; data curation, Z.L.; writing—original draft preparation, S.L. and Z.L.; writing—review and editing, T.S.S.; funding acquisition, S.L. All authors have read and agreed to the published version of the manuscript.

Funding: This research was funded by Aeronautical Science Foundation of China, grant number 41402020401”.

Acknowledgements: Shengli Lv are grateful for the support of this research under the Aeronautical Science Foundation of China and program 41402020401.

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

Metals 2020, 10, 1437 20 of 21

3. When stress corrosion occurs, the average corrosion depth of test specimen of the alloy containing an FSW joint is less than that of the base metal. The maximum depth of corrosion is greater than that of the base metal. This indicates that the test specimen containing the FSW joint is more susceptible to damage and degradation than the base metal upon exposure to the same aqueous environment. 4. When stress corrosion occurs, the ratio of maximum depth to average depth (denoted as α) increases with an increase in applied stress and duration of exposure to the aqueous environment (EXCO solution). 5. The ratio (α) for test specimens containing the FSW joint is significantly higher than that for specimens of the base metal. Further, the observed increase in α with applied stress is faster for the test specimens containing an FSW joint than for test specimens of the base metal.

Author Contributions: Methodology, S.L.; validation, S.L. and X.G.; formal analysis, S.L.; investigation, Z.L.; data curation, Z.L.; writing—original draft preparation, S.L. and Z.L.; writing—review and editing, T.S.S.; funding acquisition, S.L. All authors have read and agreed to the published version of the manuscript. Funding: This research was funded by Aeronautical Science Foundation of China, grant number 41402020401. Acknowledgments: Shengli Lv are grateful for the support of this research under the Aeronautical Science Foundation of China and program 41402020401. Conflicts of Interest: The authors declare no conflict of interest.

References

1. Threadgill, P.L.; Leonard, A.J.; Shercliff, H.R.; Withers, P.J. Friction stir welding of aluminium alloys. Int. Mater. Rev. 2009, 54, 49–93. [CrossRef] 2. Surekha, K.; Murty, B.S.; Prasad, R.K. Comparison of corrosion behaviour of friction stir processed and laser melted AA 2219 aluminium alloy. Mater. Des. 2011, 32, 4502–4508. [CrossRef] 3. Williston, D.H. Comparison of joining processes for Haynes 230 nickel based super alloy. Ph.D. Thesis, Mississippi State University, Starkville, MI, USA, 2013. 4. Fratini, L.; Buffa, G.; Palmeri, D. Material flow in FSW of AA7075-T6 butt joints: Continuous dynamic recrystallization phenomena. J. Eng. Mater. Technol. 2006, 128, 2857–2863. [CrossRef] 5. Gan, W.; Okamoto, K.; Hirano, S.; Chung, K.; Kim, C.; Wagoner, R.H. Properties of friction-stir welded aluminum alloys 6111 and 5083. J. Eng. Mater. Technol. 2008, 130, 389–392. [CrossRef] 6. Nandan, R.; Debroy, T.; Bhadeshia, H.K.D.H. Recent Advances in Friction-Stir Welding-Process, Weldment Structure and Properties. Prog. Mater. Sci. 2008, 53, 980–1023. [CrossRef] 7. Sullivan, A.; Robson, J. Microstructural Properties of Friction Stir Welded and Post-Weld Heat-Treated 7449 Aluminum Alloy Thick Plate. Mater. Sci. Eng. A 2008, 478, 351–360. [CrossRef] 8. Wang, Q. Corrosion behavior of spray formed 7055 aluminum alloy joint welded by underwater friction stir welding. Mater. Des. 2015, 68, 97–103. [CrossRef] 9. Gharavi, F.; Matori, K.A.; Yunus, R.; Othman, N.K.; Fadaeifard, F. Corrosion evaluation of friction stir welded lap joints of AA6061-T6 aluminum alloy. Trans. Nonferrous Met. Soc. China 2016, 26, 684–696. [CrossRef] 10. Gharavi, F.; Matori, K.; Yunus, R.; Othman, N.; Fadaeifard, F. Corrosion behavior of Al6061 alloy weldment produced by friction stir welding process. J. Mater. Res. Technol. 2015, 4, 314–322. [CrossRef] 11. Lumsden, J.B.; Mahoney, M.W.; Poliock, G.; Rhodes, C.G. Intergranular Corrosion Following Friction Stir Welding of Aluminum Alloy 7075-T651. Corrosion (Houston) 1999, 55, 1127–1135. [CrossRef] 12. Lumsden, J.B.; Pollock, G.; Mahoney, M.W. Corrosion behavior of friction-stir-welded AA7075-T7651. Corrosion 2003, 59, 212–219. [CrossRef] 13. Chen, Z. Microstructure characterization, mechanical properties, and corrosion behaviors of friction stir welded AA5086 and AA6061. Master’s Thesis, . University of HAWAI’I at MANO,¯ Honolulu, HI, USA, 2014. 14. Esmaily, M.; Mortazavi, N.; Osikouicz, W.; Hindsefelt, H.; Svensson, J.E.; Halvarsson, M.; Martin, J.; Johansson, L.G. Bobbin and conventional friction stir welding of thick extruded AA6005-T6 profiles. Mater. Des. 2016, 108, 114–125. [CrossRef] Metals 2020, 10, 1437 21 of 21

15. Venugopal, T.; Rao, K.S.; Rao, K.P. Studies on Friction Stir Welded AA7075 Aluminum Alloy. Trans. Indian Inst. Met. 2004, 57, 659–663. 16. Wadeson, D.A.; Zhou, X.; Thompson, G.E.; Skeldon, P.; Djapic, O.L.; Scamans, G. Corrosion Behaviour of Friction Stir Welded AA7108 T79 Aluminium Alloy. Corros. Sci. 2006, 48, 887–897. [CrossRef] 17. Paglia, C.S.; Buchheit, R.G. A Look in the Corrosion of Aluminum Alloy Friction Stir Welds. Scr. Mater. 2008, 58, 383–387. [CrossRef] 18. Paglia, C.S.; Carroll, M.C.; Pitts, B.C.; Reynolds, T.; Buchheit, R.G. Strength, Corrosion, and Environmentally Assisted Cracking of A 7075-T6 Friction Stir Weld. In Materials Science Forum; Trans Tech Publications Ltd.: Zurich, Switzerland, 2002; Volume 396, pp. 1677–1684. 19. Bousquet, E.; Poulon-Quintin, A.; Puiggali, M.; Devos, O.; Touzet, M. Relationship between microstructure, microhardness and corrosion sensitivity of an AA 2024-T3 friction stir welded joint. Corros. Sci. 2011, 53, 3026–3034. [CrossRef] 20. Litwinski, E. Effect of FSW Process and Parameters on Corrosion Susceptibility. In Proceedings of the Physical Sciences Symposia—Falilure Analysis and Prevention. Microscopy and Microanalysis, Honolulu, HI, USA, 31 July–4 August 2005. 21. Barbara, N.P. Investigation into the Stress Corrosion Cracking Properties of AA2099, an Al-Li-Cu Alloy. Ph.D. Thesis, The Ohio State University, Columbus, OH, USA, 2008. 22. Astarita, A.; Bitondo, C.; Squillace, A.; Armentani, E.; Bellucci, F. Stress corrosion cracking behavior of conventional and innovative Aluminium alloys for aeronautic applications. Surf. Interface Anal. 2013, 45, 1610–1618. [CrossRef] 23. ASTM G34-01. Standard Test Method for Exfoliation Corrosion Susceptibility in 2 and 7 Series Aluminum ××× ××× Alloys; IHS Markit: Englewood, CO, USA, 2014.

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

© 2020 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 (http://creativecommons.org/licenses/by/4.0/).