Fatigue Corrosion Behavior of Friction Welded Dissimilar Joints in Diﬀerent Testing Conditions

metals

Article Fatigue Corrosion Behavior of Friction Welded Dissimilar Joints in Diﬀerent Testing Conditions

Stefano Rossi , Francesca Russo, Alberto Maria Lemmi, Matteo Benedetti * and Viglio Fontanari Department of Industrial Engineering, University of Trento, Via Sommarive 9, 38123 Trento (TN), Italy; [email protected] (S.R.); [email protected] (F.R.); [email protected] (A.M.L.); [email protected] (V.F.) * Correspondence: [email protected]; Tel.: +39-0461-282430

Received: 8 July 2020; Accepted: 25 July 2020; Published: 29 July 2020

Abstract: Our study will be focused on stainless steel AISI 304—carbon steel ASTM A105 joints obtained by rotary friction welding and on their fatigue corrosion behavior in different testing environments. As a first thing, the joints will be characterized by microscopy and electrochemical techniques. The manuscript will then describe the optimization of the experimental setup and the validation of the testing procedure. After that, the fatigue behavior of the joints will be tested in different aggressive environments. This study pointed out that it is possible to build a simple and low-cost setup for the study of fatigue corrosion behavior of dissimilar joints while exploiting in situ electrochemical measurements to follow the fatigue corrosion process.

Keywords: fatigue corrosion; rotary friction welding; dissimilar joints; AISI 304; ASTM A105

1. Introduction The possibility of joining distinct materials is a very attractive opportunity in the assembly and design of high-performance components for high-duty technological applications, such as in the automotive or in the aeronautic field [1]. Many of the structural elements in pressure vessels, spacecraft, and transport applications, as well as in offshore and nuclear structures are made of welded joints [2–4]. The coupling of different materials is very intriguing as it is possible to take advantage of the best properties of each material and build components with good structural and lightweight characteristics, but, on the other hand, it represents a challenge because new phenomena, arising from their interaction, must be taken under consideration [5]. Dissimilar joints can be realized using different techniques, such as threading or the use of adhesives [6], but the production rate and life expectancy of these joints is not so high. High-performance joints can be obtained exploiting solid-state welding methods [7], such as rotary friction welding (RFW) and linear friction welding (LFW) [8,9]. Friction welding (FW) techniques allow coupling dissimilar materials, such as metals, ceramics, polymers, and composites [10–12] that cannot be joined by conventional fusion welding techniques, operating at temperatures lower than the melting point of the materials being welded together. Despite all these positive aspects, it is necessary to point out that FW techniques could affect the material composition and structure and could also cause the precipitation of brittle phases [13] as well as the formation of cracks [13,14]. Friction welding is exploited using the relative motion of the two pieces pressed against each other, thus producing the needed heat by the conversion of mechanical energy into thermal energy, without any external heat input [6]. Rotation speed, applied load, and welding time are the main process parameters to be considered [5,6,15]. These can be adjusted to reach the plastic temperature range close to the interface where atomic diffusion happens and metallurgical bonds form. Austenitic stainless steel-steel joints are very common and possess a

Metals 2020, 10, 1018; doi:10.3390/met10081018 www.mdpi.com/journal/metals Metals 2020, 10, 1018 2 of 14 good combination of mechanical properties, formability, weldability, and resistance to corrosion [16–18]. Many studies are present in the literature about the microstructural characteristics, phase formation at the interface, and tensile properties of these type of joints, but only few data are present about extensive fatigue studies and fatigue corrosion investigations. Our study will be focused on stainless steel AISI 304—carbon steel ASTM A105 joints. The AISI 304 grade stainless steel has good mechanical properties and high corrosion resistance [19], ASTM A105 carbon steel has attractive properties such as high strength, good wear resistance, and affordable cost. In many applications, dissimilar welded joints play a key structural role and they are often subjected to cyclic stresses. Such loading could lead to worrisome fatigue failure and to catastrophic failure as well [20]. Therefore, it is necessary to evaluate the integrity of the joints and deeply study their mechanical properties, especially fatigue resistance [20] and failure modes. The design of fatigue experiment could be very challenging because fatigue behavior does not only depend on the chosen material, but it is also dependent on the sample’s geometrical features (sharp edges, notches) [21–23] and post-welding treatments [24,25], that could influence surface roughness and create preferential cracking points. When considering the integrity of structural components, such as dissimilar metal welded joints (DMWJ), it is important to take under consideration the effect of the environment and the role of corrosion in the cracking mechanisms and failure of these components. Cracking caused by a combination of environmental, mechanical, and metallurgical variables is commonly referred to as “environmental cracking” [26] and could manifest in fatigue corrosion (FC), stress corrosion cracking (SCC), and hydrogen embrittlement [27]. FC is the result of the combined action of an alternating or cyclic stress and a corrosive environment; the fatigue process is thought to cause the rupture of the protective passive film, upon which corrosion is accelerated [28]. Contrary to a pure mechanical fatigue, there is no fatigue limit load in corrosion-assisted fatigue. SCC is the cracking induced from the combined influence of tensile stress and a corrosive environment; the required tensile stresses may be in the form of external applied stresses or in the form of residual stresses set up as a result of machining, deformation, welding, and grinding [29,30]. Hydrogen embrittlement is a type of deterioration that could be linked to corrosion processes: It usually involves the diffusion of hydrogen into the crystal lattice of metallic component subjected to tensile stress and it could even cause cracking and catastrophic brittle failure [30]. Conditions favorable to corrosion fatigue could also lead to stress-corrosion cracking phenomena and to hydrogen embrittlement, therefore the transition between these environmental cracking processes is not so sharp and many features are shared between them. Corrosion can happen in a huge variety of conditions and it can develop in a large variety of forms. It is then necessary to investigate the behavior of structural components in the realistic conditions in which corrosion can develop, such as their behavior in fatigue and fatigue corrosion tests. Fatigue and corrosion interact in a complex manner, failure can occur at different mode and times, and the component damage can be never described as the simple sum of the two effects. Only sporadic experiments were conducted about the study of fatigue behavior in an aggressive environment of hybrid joints between dissimilar materials. Considering the large range of potential applications, the testing of stainless steel-steel joints could be both interesting from a scientific point of view and improve economic sustainability. The main aim of this work is to develop a method to characterize fatigue and fatigue corrosion behavior of dissimilar friction welded joints. The experimental setup development and validation of the testing procedure will be the crucial point of this work. This paper will include a description of how the specimen geometry and the testing setup were developed to obtain robust measurements. Then, the effectiveness of the testing method will be probed in different experimental conditions.

2. Materials and Methods Dissimilar rotary friction welded joints were realized using AISI304 stainless steel and ASTM A105 medium-low carbon steel. The two alloys chemical composition, according to the international standards [31,32], is reported in Table1. Metals 2020, 10, x FOR PEER REVIEW 3 of 16 Metals 2020, 10, 1018 3 of 14 Table 1. Chemical composition of AISI 304 and ASTM A105 alloys.

2 Table 1. ChemicalAlloying compositionElement AISI of AISI 304 304 ASTM and ASTM A105 A105 alloys. C 0.08 1 0.35 1 Alloying ElementMn AISI2.00 304 1 0.60–1.05ASTM A105 2 C P 0.080.0451 1 0.0350.35 1 1 Mn S 2.000.031 1 0.0400.60–1.05 1 1 1 P Si 0.0451.00 1 0.10–0.350.035 S 0.03 1 0.040 1 1 Si Cr 18.00–20.001.00 1 0.300.10–0.35 CrNi 18.00–20.008.00–10.50 0.400.30 1 1 NiMo 8.00–10.50- 0.120.40 1 1 1 MoCu - - 0.400.12 1 Cu - 0.40 1 1 V-V - 0.080.08 1 FeFe Bal.Bal. Bal. Bal. 1 Maximum1 Maximum value; 2 value;percentage 2 percentage sums (Cu,sums Ni, (Cu, Cr, Ni, Mo, Cr, V) Mo,= 1.00V) = wt%;1.00 wt%; maximum maximum percentage percentage (Cr +(CrMo) + Mo)= 0.32 wt%. = 0.32 wt%. The samplesThe samples were were produced produced by by Tomet Tomet MECHANICS (Castelgomberto, (Castelgomberto, Vicenza, Vicenza, Italy) joining Italy) a joining a 21 mm21 diametermm diameter rod rod of of ASTM ASTM A105 steel steel with with an 18 an mm 18 diameter mm diameter rod of AISI304 rod of stainless AISI304 steel. stainless The steel. The samplessamples were were produced produced throughthrough the the rotary rotary friction friction welding welding technique. technique. The as-provided The as-provided samples samples neededneeded further further turning turning operations operations toto achieve a shape a shape suitable suitable to be tested to be in tested fatigue in experiments. fatigue experiments. The samples were lathe machined and then grinded until reaching a surface roughness value of about 2 The samples were lathe machined and then grinded until reaching a surface roughness value of about μm (Ra). The final design of the samples was developed after preliminary tests in a universal testing µ 2 m (Rmachinea). The (UTM), final starting design from of the the samples one suggested was developedby the UNI EN after ISO preliminary 11782-1 standard tests [33]. in aThe universal testingspecimen machine cross (UTM), section starting in the welded from the area one was suggested reduced to byabout the one UNI third EN of ISO the 11782-1original so standard as to [33]. The specimenavoid influences cross section of surface in thedefects welded and to area reduce was the reduced load bearing to about capability one within third ofthe the allowable original so as to avoidranges influences of the UTM of surface used during defects fatigue and testing. to reduce The weld the load is located bearing approximately capability in within the middle the allowableof ranges ofthe the investigated UTM used cylindrical during volume, fatigue having testing. 6 mm The diameter weld is cross located section approximately and 10 mm length. in the A middle large of the fillet radius and a smooth bevel transition up to the 12 mm diameter of the cross section in the investigated cylindrical volume, having 6 mm diameter cross section and 10 mm length. A large fillet gripping regions was realized, in order to avoid the risk of failure localization at the fillet’s base. The radius andrecess a in smooth the shaft bevel was made transition in order up to toease the the 12 sealing mm diameterof the electrochemical of the cross cell. section A sketch in of the the gripping regionsfinal was sample realized, geometry in order is reported to avoid in Figure the risk 1. of failure localization at the fillet’s base. The recess in the shaft was made in order to ease the sealing of the electrochemical cell. A sketch of the final sample geometry is reported in Figure1.

Metals 2020, 10, x FOR PEER REVIEW 4 of 16 (a)

(b)

FigureFigure 1. (a) Selected1. (a) Selected specimen specimen design design for for fatigue fatigue tests tests (unit: (unit: mm); mm); (b) image (b) image of the of as-made the as-made sample. sample.

Metals 2020, 10, 1018 4 of 14

The morphology and internal structure of the welded joints were studied by optical and electron microscopy, by means of Nikon SMZ25 stereomicroscope (Nikon Instruments, Amstelveen, The Netherlands) and JEOL IT300 scanning electron microscope (SEM, Jeol Ltd., Akishima, Tokyo, Japan), respectively. Moreover, optical and SEM observations were employed in the failure mechanism and failure analysis assessment. Energy dispersive X-ray spectrometry (EDS) measurements, carried out using a Bruker “Quantax Micro-XRF” analyser (Bruker, Billerica, MA, USA), were useful to gain further information on the composition of the samples and the distribution of the different alloying elements. Micro-hardness tests were carried out using a Vickers digital micro-hardness tester (Model FM-310, Future tech, Kawasaki, Japan) along the weld joint. A load of 200 g was applied for a duration of 10 s. The measurements are the average of three repetitions. Electrochemical analyses were performed to determine the corrosion resistance of the two different alloys and study their electrochemical behavior. Cyclic potentiodynamic measurements were carried out using a potentiostat PARSTAT 2273 (Princeton Applied Research, Oak Ridge, TN, USA) in a three-electrode configuration where the reference electrode (RE) was an Ag/AgCl sat. (+207 mV SHE) electrode and the counter electrode (CE) was a platinum wire. The testing solution was a 3.5 wt% NaCl solution. As a first thing, the free corrosion potential (Ecorr) was determined. Then potentiodynamic measurements were exploited, starting from a potential value of (E = Ecorr 0.1 V) and going into − anodic polarization. The test was stopped when the potential reached the breakdown potential for samples showing a passivity state. The scanning rate was set to be 0.166 mV/s. The fatigue and fatigue corrosion behavior of the dissimilar friction welded joints was studied performing fatigue tests in different conditions: The samples were tested in air (F), in an aggressive solution with no imposed potential (fatigue corrosion (FC)), in an aggressive solution with a negative imposed potential (cathodically protected fatigue corrosion (FC-P)), and in an aggressive solution with a positive imposed potential (accelerated fatigue corrosion (FC-A)). The fatigue tests were performed on an Instron 8800 universal testing machine (Instron, Norwood, MA, USA) with a TRIO RT3 control system. The specimens were axially loaded in tension, with stress ratio (minimum over maximum stress in a cycle) R = 0.1. A 4 Hz sine wave was chosen to perform the tests, a trade-off aimed at making detectable the environmental effects on the fatigue behavior in a reasonable testing time. The tests were stopped at the specimen’s failure or after two million fatigue cycles, as the samples were considered run outs. As a first thing, fatigue in air tests were performed. In order to exploit corrosion fatigue tests and apply a given potential to the specimens, it was necessary to develop an easy-to-build electrochemical cell, paying attention not to interfere with the fatigue tests and to avoid electrical contact between the samples and the testing apparatus. The design of the cell accounted for various factors, such as the ease to be built, to be operated, and to be replaced as well. The possibility of easily building and replacing the cell allows performing customizable tests. Figure2 shows a sketch of the electrochemical cell and a photograph of the experimental setup. The first challenge to be overcome arose since the hydraulic actuator of all UTMs available in the lab displaces the bottom grip of the specimen holder, thus making it impossible to position the cell on the bottom grip of the UTM. This problem was solved by providing the electrochemical cell an external support, realized with a T structure anchored on the bottom plate of the UTM and provided with an adjustable prong clamp. The second challenge consisted of guaranteeing a static waterproof seal between the cell and the specimen; very compliant polymeric elastic tubular elements were fixed to the cell and the specimen using mechanical retainers (steel cable ties). As a solution container, a polyethylene (PE) 200 mL flask three times wider in diameter with respect to the specimen was chosen. Holes were cut in the cap of the flask and in the bottom of the flask to allow the specimen to pass through. Two small holes were provided in the flask to support the electrodes without the need to use external supports. The electrochemical cell, mounted around the specimen, was also useful to perform reference electrochemical measurements and thus interpret the time-current data that will be collected during corrosion fatigue experiments. These measurements were carried out with the same parameters used Metals 2020, 10, 1018 5 of 14 in the electrochemical characterization of the samples. After that it was possible to exploit fatigue tests in the presenceMetals of an 2020 aggressive, 10, x FOR PEER REVIEW environment, as described in Table2. 5 of 16

(a) (b)

Figure 2. (a)Figure Electrochemical 2. (a) Electrochemical cell cell design; design; ( bb) )photograph photograph of the experimental of the experimental setup. setup.

The first challengeTable to be overcome 2. Testing arose parameters since the hydraulic of fatigue actuator tests. of all UTMs available in the lab displaces the bottom grip of the specimen holder, thus making it impossible to position the cell on the bottom grip of the UTM. This problem was solved by providing the electrochemical cell Test Name Description Solution Potential an external support, realized with a T structure anchored on the bottom plate of the UTM and provided Fwith an adjustable Fatigue prong clamp. in air The second challengeN/A consisted of guaranteeingN/A a static waterproofFC seal between the Fatigue cell and corrosion the specimen; very3.5 compliant wt% NaCl polymeric elasticN tubular/A elements FC-P Protected fatigue corrosion 3.5 wt% NaCl 1.3 V (vs. Ag/AgCl) were fixed to the cell and the specimen using mechanical retainers (steel− cable ties). As a solution container,FC-A a polyethylene Accelerated (PE) 200 fatigue mL corrosionflask three times3.5 wt% wider NaCl in diameter+0.4 Vwith (vs. respect Ag/AgCl) to the specimen was chosen. Holes were cut in the cap of the flask and in the bottom of the flask to allow the specimen to pass through. Two small holes were provided in the flask to support the electrodes Regardingwithout the FC-P the need test, to use the external imposed supports. potential was chosen in order to have a complete cathodic protected materialThe [34 electrochemical,35]. This cell, choice mounted could around also the sp proveecimen, verywas also useful useful to to perform verify reference the e ﬀectiveness of electrochemical measurements and thus interpret the time-current data that will be collected during cathodic protectioncorrosion in fatigue the presence experiments. of These an increased measurements hydrogen were carried productionout with the same due parameters to the used cathodic reduction of water. In FC-Ain the testing electrochemical the imposed characterization potential of the has sample a values. After comparable that it was possible to the to exploit stainless-steel fatigue breakdown potential observedtests in experimentally. the presence of an aggressive The applied environment, potential as described during in Table fatigue2. tests was used to hinder or enhance corrosion phenomena. Table 2. Testing parameters of fatigue tests. After all the testsTest were Name completed, Description a detailed microscopic Solution analysis Potential of the specimens was carried out in order to gain furtherF insights on Fatigue the in failure air mechanisms N/A of the joints N/A in diﬀerent conditions. FC Fatigue corrosion 3.5 wt% NaCl N/A FC-P Protected fatigue corrosion 3.5 wt% NaCl −1.3 V (vs. Ag/AgCl) 3. Results and DiscussionFC-A Accelerated fatigue corrosion 3.5 wt% NaCl +0.4 V (vs. Ag/AgCl)

3.1. Morphological andRegarding Metallographic the FC-P test, the Analysis imposed potential was chosen in order to have a complete cathodic protected material [34,35]. This choice could also prove very useful to verify the effectiveness of Figure3 showscathodic a protection SEM micrograph in the presence of of the an increased welded hydrogen specimen production interface. due to Onthe cathodic the left the AISI 304 reduction of water. In FC-A testing the imposed potential has a value comparable to the stainless- stainless steel, onsteel the breakdown right sidepotential the observed ASTM experimentally. A105 carbon The steel.applied Thepotential interface during fatigue between tests was the two materials is not straight andused to continuous, hinder or enhance but corrosion there phenomena. is a gradual intermixing of the two steels, due to the rotary friction welding process. In addition to that, it is possible to highlight the presence of some black precipitates in the carbon steel side. The composition of the precipitates is not constant, but they are mainly composed of carbon, sulphur, molybdenum, and manganese. Figure4 shows optical micrographs of the specimen interface. Welding defects, such as porosities and cracks are not present, thus underlining the good quality of the welding process. The dissimilar joint microstructure can be classified in three different regions. The first region, astride the interface and with an extension of 470 µm on either side of the interface, is the fully plastically deformed region. This region contains very fine small recrystallized grains of ferrite and perlite [36]. It is well known that pressure used to join the materials results in recrystallization leading to a grain refinement in the central region of the weld [37]. The second region is constituted by an area of the sample which underwent a partial plastic deformation; here the grains are larger than in the first region near the interface, but their shape is modified and stretched along the load axial direction. Its structure is constituted by ferrite Metals 2020, 10, x FOR PEER REVIEW 6 of 16

After all the tests were completed, a detailed microscopic analysis of the specimens was carried out in order to gain further insights on the failure mechanisms of the joints in different conditions.

3. Results and Discussion Metals 2020, 10, 1018 6 of 14 3.1. Morphological and Metallographic Analysis Figure 3 shows a SEM micrograph of the welded specimen interface. On the left the AISI 304 stainless steel, on the right side the ASTM A105 carbon steel. The interface between the two materials and perlite, as alreadyis not straight reported and continuous, for medium-carbon but there is a gradual steelsintermixing which of the underwenttwo steels, due to rotary the rotary friction welding processes [38]. Thefriction extension welding process. of this In addition area is to 1500 that, itµ ism. possible The thirdto highlight area the is presence represented of some byblack the undeformed precipitates in the carbon steel side. The composition of the precipitates is not constant, but they are base material microstructure,mainly composed of consistingcarbon, sulphur, of molybdenum, ferrite and and perlite. manganese.

Figure 3. Scanning electron microscopy image of the welded specimen interface. FigureMetals 2020 3., 10Scanning, x FOR PEER REVIEW electron microscopy image of the welded specimen interface.7 of 16

Figure 4 shows optical micrographs of the specimen interface. Welding defects, such as porosities and cracks are not present, thus underlining the good quality of the welding process. The dissimilar joint microstructure can be classified in three different regions. The first region, astride the interface and with an extension of 470 μm on either side of the interface, is the fully plastically deformed region. This region contains very fine small recrystallized grains of ferrite and perlite [36]. It is well known that pressure used to join the materials results in recrystallization leading to a grain refinement in the central region of the weld [37]. The second region is constituted by an area of the sample which underwent a partial plastic deformation; here the grains are larger than in the first region near the interface, but their shape is modified and stretched along the load axial direction. Its structure is constituted by ferrite and perlite, as already reported for medium-carbon steels which underwent rotary friction welding processes [38]. The extension of this area is 1500 μm. The third area is represented by the undeformed base material microstructure, consisting of ferrite and perlite.

(a) (b)

(c)

(d) (e) Figure 4. OpticalFigure images 4. Optical of images the welded of the welded sample sample microstructures. microstructures. (a) Interface (a) Interface region (AISI region 304 on the (AISI 304 on the left side and ASTM A105 on the right side of the interface); (b) intermediate region; (c) schematic left side and ASTMrepresentation A105 of on the sample; the right (d) intermediate side of theregion interface); at high magnification; (b) intermediate (e) base metal. region; (c) schematic representation of the sample; (d) intermediate region at high magniﬁcation; (e) base metal. Figure 5 shows the longitudinal hardness measurements distribution of the test specimen. Hardness was measured across the weld interface, from 2.0 mm left and right on the AISI 304 and ASTM A105 metals.

Metals 2020, 10, x FOR PEER REVIEW 8 of 16

Metals 2020, 10, 1018 7 of 14

Figure5 shows the longitudinal hardness measurements distribution of the test specimen. Hardness was measured across the weld interface, from 2.0 mm left and right on the AISI 304 and Metals 2020, 10, x FOR PEER REVIEW 8 of 16 ASTM A105 metals.

Figure 5. Vickers hardness distribution through the weld interface.

A maximum hardness of 247 HV has been obtained at the weld interface. The increase in Figure 5. Vickers hardness distribution through the weld interface. hardness at the weldingFigure interface 5. Vickers is probablyhardness distribution due to throughfriction the and weld oxidation interface. processes which take place duringA maximum welding hardness processing. of 247 Similar HV has hardness been obtained measur atements the weld were interface. obtained The by increase [36] for in the hardness AISI 304 stainless steelA maximum base metal hardness and offor 247 the HV weld has been interface obtained of at AISI the weld304 stainlessinterface. The steel—mild increase in steel friction at the weldinghardness interface at the welding is probably interface is due probably to friction due to friction and oxidation and oxidation processes processes which take take place place during weldingwelded processing.during joints. welding Other Similar processing. studies hardness [13,39] Similar measurementshardnessreport also measur highements wereer values were obtained obtained for dissimilar by by [36 [36]] for for jo thetheints AISI AISI with 304 304 304 stainless stainless steelsteel. base stainless metal andsteel forbase the metal weld and interface for the weld of AISIinterface 304 of stainless AISI 304 stainless steel—mild steel—mild steel frictionsteel friction welded joints. welded joints. Other studies [13,39] report also higher values for dissimilar joints with 304 stainless Other studies [13,39] report also higher values for dissimilar joints with 304 stainless steel. 3.2. Electrochemicalsteel. Analysis 3.2. ElectrochemicalElectrochemical3.2. Electrochemical Analysis analysesAnalysis of the plain AISI 304 specimen, plain ASTM A105 specimen, and weldedElectrochemical specimenElectrochemical are analyses useful analyses of to the ofset plainthe up plain AISIthe AISIelectr 304 304 specimen,ochemical specimen, plain plainparameters ASTMASTM A105 A105during specimen, specimen, the corrosion and and welded fatigue specimentests andwelded are to useful discuss specimen to the set are up obtaineduseful the to electrochemical set results. up the electrFigureochemical parameters 6 shows parameters the during potentiodynamic during the corrosionthe corrosion fatiguepolarization fatigue tests andscans to of all the samplestests and to taken discuss under the obtained consideration. results. Figure 6 shows the potentiodynamic polarization scans of discuss theall obtainedthe samples results. taken under Figure consideration.6 shows the potentiodynamic polarization scans of all the samples taken under consideration.

Figure 6. Potentiodynamic polarization curves in a 3.5 wt% NaCl solution of: (a) AISI 304; (b) ASTM A105; (c) welded AISI304—ASTM A105.

Figure 6. Potentiodynamic polarization curves in a 3.5 wt% NaCl solution of: (a) AISI 304; (b) ASTM A105;Figure (c) welded6. Potentiodynamic AISI304—ASTM polarization A105. curves in a 3.5 wt% NaCl solution of: (a) AISI 304; (b) ASTM A105; (c) welded AISI304—ASTM A105. The curve (a), relative to the AISI 304 specimen, starts in the cathodic region, it reaches the corrosion potential ( 0.206 V vs. Ag/AgCl) and then, it goes through the passivation region, where an − Metals 2020, 10, x FOR PEER REVIEW 9 of 16

The curve (a), relative to the AISI 304 specimen, starts in the cathodic region, it reaches the Metalscorrosion2020, 10 potential, 1018 (−0.206 V vs. Ag/AgCl) and then, it goes through the passivation region, 8where of 14 an increase of the potential does not imply a variation of the current density. When the breakdown increasepotential of is the reached potential (+0.355 does V notvs. Ag/AgCl) imply a variation the curve of rapidly the current deviates density. to higher When values the of breakdown the current potentialdensity isand reached corrosion (+0.355 takes V vs.place. Ag /BeyondAgCl) the this curve potential rapidly the deviates pitting tophenomenon higher values is ofstable, the current and the densitymaterial and is not corrosion able to takesreform place. a permanent Beyond passivit this potentialy layer the anymore. pitting The phenomenon return scan is clearly stable, highlights and the materialthe hysteresis is not able phenomenon, to reform a that permanent implies passivityno immediate layer repassivation anymore. The of return the metal scan clearlysurface, highlights in fact it is thenecessary hysteresis to reach phenomenon, a potential that lower implies than nothe immediate breakdown repassivation potential for the of the passive metal layer surface, to onset in fact again. it isThe necessary curve (b) to reachrepresents a potential the linear lower polarization than the breakdown curve of the potential ASTM forA105 the specimen. passive layer At the to onsetlowest again.potential The values, curve (b)it is represents possible theto identify linear polarization the cathodic curve region of theunder ASTM diffusion A105 control, specimen. due At to the the lowestoxygen potential reduction. values, After it passing is possible the to co identifyrrosion thepotential cathodic of the region material under (− di0.454ffusion V vs. control, Ag/AgCl) due tothe theanodic oxygen current reduction. starts Afterto arise. passing The thecorrosion corrosion rate potential of the carbon of the steel material specimen ( 0.454 increases V vs. Ag with/AgCl) the − thepotential, anodic currentas typical starts of an to alloy arise. in Thean active corrosion state rateand ofno thepassivation carbon steel tendency. specimen The increasescurve (c) represents with the potential,the potentiodynamic as typical of an polarization alloy in an curve active of state the and welded no passivation AISI 304—ASTM tendency. A105 The sample. curve (c) The represents shape of thethe potentiodynamic curve is very similar polarization to the one curve of the of ASTM the welded A105 AISIspecimen, 304—ASTM but the A105corrosion sample. potential The shape is shifted of theto curvelower isvalues very similar(−0.554 toV thevs. Ag/AgCl). one of the ASTMThe electrochemical A105 specimen, situation but the is corrosion very complex, potential as the is shifted sample tois lower constituted values (by0.554 three V different vs. Ag/AgCl). zones: The The electrochemical stainless steel situationregion, the is verycarbon complex, steel region, as the sampleand the − iswelding constituted area, by which three has diff aerent different zones: composition The stainless with steel respect region, to the the plain carbon steels. steel A galvanic region, andcoupling the weldingbetween area, AISI which 304 and has aASTM different A105 composition with the increase with respect of anodic to the current plain steels. density A galvanicdue to increase coupling of betweencorrosion AISI rate 304 of carbon and ASTM steel A105was expected. with the The increase presence of anodic of a mixed current phase density withdue modification to increase of ofthe corrosionmicrostructure rate of carbonseems steelto further was expected. worsen Thethe presence specimen of abehavior. mixed phase This with fact modification can be explained of the microstructureconsidering that seems the towelding further process worsen produced the specimen phas behavior.es with a Thisdifferent fact cancomposition be explained with considering detrimental thateffects the weldingon the corrosion process producedresponse. phases with a different composition with detrimental effects on the corrosion response. 3.3. Fatigue Testing 3.3. FatigueThe results Testing obtained by the fatigue testing of welded AISI304—ASTM A105 samples under differentThe results conditions: obtained Fatigue by thein air fatigue (F), fatigue testing corrosion of welded (FC), AISI304—ASTM protected fatigue A105 corrosion samples (FC-P), under and diacceleratedfferent conditions: fatigue corrosion Fatigue in(FC-A) air (F), are fatigue plotted corrosion in Figure (FC), 7. Fatigue protected tests fatiguestarted corrosionin such a way (FC-P), that andtheaccelerated first tests were fatigue carried corrosion out at (FC-A) stress levels are plotted near but in Figure lower7 than. Fatigue the tensile tests started strength in such(580 MPa) a way of thatthe the stainless first tests steel were [40] carried and carbon out at stresssteel (min. levels 485 near MPa) but lower[41]. The than stress the tensile levels strength were also (580 selected MPa) of to thehave stainless single steeltest durations [40] and carbonless than steel 50 h (min. each. 485 MPa) [41]. The stress levels were also selected to have single test durations less than 50 h each.

Figure 7. S-N curves for AISI304—ASTM A105 welded specimens under diﬀerent testing conditions. Figure 7. S-N curves for AISI304—ASTM A105 welded specimens under different testing conditions. The fatigue tests in air (F) were carried out at three diﬀerent loading conditions. The most severe loading condition resulted in a lower number of cycles to failure whereas the progressive reduction Metals 2020, 10, x FOR PEER REVIEW 10 of 16

The fatigue tests in air (F) were carried out at three different loading conditions. The most severe loading condition resulted in a lower number of cycles to failure whereas the progressive reduction Metalsof2020 the, loading10, 1018 led to longer sample lives. The specimen tested at the lower stress level has survived9 of 14 two million cycles, thus suggesting a transition from finite life regime to a fatigue limit. A further of theinvestigation loading led in to this longer stress sample range lives. was beyond The specimen the scope tested of this at the research lower stressalso considering level has survived the limited twonumber million of cycles, samples thus available. suggesting The a FC, transition FC-P, and from FC-A ﬁnite tests life were regime carried to a fatigueout at the limit. two A more further severe investigationloading conditions. in this stress The range presence was beyondof an aggressive the scope environment of this research led alsoto a consideringdetectable decrease the limited of the numberfatigue of life. samples The imposition available. The of the FC, anodic FC-P, andpotential FC-A drastically tests were reduced carried out the atspecimen’s the two more life due severe to the loadingstrong conditions. corrosion attack The presence on the specimen. of an aggressive On the environmentother side, the led cathodic to a detectable protection decrease seems to of exert the a fatiguepositive life. influence The imposition on the offatigue the anodic behavior potential by enhancin drasticallyg the reduced fatigue life the at specimen’s the lower lifestress due level to the up to strongvalues corrosion higher attackthan those on the obtained specimen. for tests On the in a other non-aggressive side, the cathodic environment. protection seems to exert a positive inﬂuence on the fatigue behavior by enhancing the fatigue life at the lower stress level up to valuesElectrochemical higher than those Characterization obtained for during tests in Fatigue a non-aggressive Experiments environment. In order to have further insights on the electrochemical behavior of the samples during fatigue Electrochemical Characterization during Fatigue Experiments testing conditions, galvanic coupling tests were performed on the unloaded (reference) sample as wellIn order as on tothe have loaded further sample insights during on the fatigue electrochemical corrosion behavior tests. Figure of the 8 represents samples during the evolution fatigue of testingpotential conditions, over time galvanic of coupled coupling AISI tests 304—ASTM were performed A105 onin thea 3.5 unloaded wt% NaCl (reference) solution sample for the as two wellconditions. as on theloaded sample during the fatigue corrosion tests. Figure8 represents the evolution of potential over time of coupled AISI 304—ASTM A105 in a 3.5 wt% NaCl solution for the two conditions.

Figure 8. Potential evolution over time of coupled AISI304—ASTM A105 in a 3.5 wt% NaCl solution unloadedFigure sample 8. Potential and onevolution a loaded over sample time tested of coupled to fatigue AISI304—ASTM corrosion at 500A105 MPa. in a 3.5 wt% NaCl solution unloaded sample and on a loaded sample tested to fatigue corrosion at 500 MPa. Looking at the galvanic coupling potential of the unloaded sample, it is possible to observe that it reachedLooking a stable state at the after galvanic a transient. coupling This potential behavior of can the be un explainedloaded sample, considering it is possible that the to two observe alloys that haveit areached different a behaviorstable state in the after environment, a transient. thus, This the behavior transient can regime be explained represent considering the on-set of that diff erentthe two corrosionalloys have mechanisms. a different In behavior the case in of the an appliedenvironmen load,t, thusthus, during the transient fatigue regime corrosion represent experiments, the on-set theof potential different of thecorrosion galvanic mechanisms. couple approaches In the morecase negativeof an applied potentials load, with thus respect during to fatigue what happens corrosion forexperiments, the unloaded the samples. potential This of isthe clearly galvanic caused couple by theapproaches mechanical more stress: negative Cyclic potentials loading with seems respect to activateto what the systemhappens and for enhance the unloaded corrosion samples. mechanisms. This is clearly caused by the mechanical stress: Cyclic loading seems to activate the system and enhance corrosion mechanisms. 3.4. Failure Analysis 3.4.Optical Failure and Analysis SE microscope analyses were carried out in order to get further insights on the failure modes ofOptical these specimensand SE microscope and to identify analyses the were fatigue carrie fractured out growthin order lines to get and further the probable insights area on the of crackfailure initiation. modes of Thethese observation specimens withand to SEM identify allowed the fatigue reaching fracture higher growth magnifications lines and and the betterprobable investigating the crack origin area.

Metals 2020Metals, 10, 1018 2020, 10, x FOR PEER REVIEW 11 of 16 10 of 14

area of crack initiation. The observation with SEM allowed reaching higher magnifications and better investigating the crack origin area. 3.4.1. Fatigue in Air 3.4.1. Fatigue in Air Figure9 shows the fracture surface of two di fferent AISI304—ASTM A105 specimens tested by fatigue in air.Figure Figure 9 shows9a shows the fracture thepresence surface of two oftwo different di ff AISI304—ASTMerent areas: The A105 fatigue specimens fracture tested by area at the fatigue in air. Figure 9a shows the presence of two different areas: The fatigue fracture area at the bottom of the sample, containing the origin where the fatigue cracks started and the unstable fracture bottom of the sample, containing the origin where the fatigue cracks started and the unstable fracture area on thearea top on (also the top called (also “sudden called “sudden fracture fracture zone”) zone”) [42]. [42]. It is It possible is possible to to observe observe that that the the crackcrack probably originatedprobably in a di originatedfferent plane in a different with respectplane with to re thespect plane to the where plane where the crack the crack propagated, propagated, andand this fact is highlightedthis fact by is thehighlighted presence by the of apresence shadow of a in shadow this area. in this The area. sudden The sudden fracture fracture area, area, on onthe the contrary, is characterizedcontrary, byis characterized a higher roughness, by a higher becauseroughness, of because the sudden of the sudden propagation propagation that that led led to to the the specimen specimen failure. The same observations are valid for the sample tested at 450 MPa, shown in Figure failure. The9c,d. sameThe samples observations both failed are in validthe region for theplasticized sample during tested welding at 450 at MPa,about 1 shown mm from in the Figure 9c,d. The samplesinterface. both failed in the region plasticized during welding at about 1 mm from the interface.

(a) (b)

Figure 9.FigureOptical 9. Optical images images of the of the fracture fracture surface surface of of a welded a welded AISI304—ASTM AISI304—ASTM A105 specimen A105 specimentested to tested fatigue in air (F) (a) Top view of the sample fracture surface tested at 500 MPa; (b) close-up of the to fatigue in air (F) (a) Top view of the sample fracture surface tested at 500 MPa; (b) close-up of the fatigue fracture surface and crack nucleation area of the sample (a); (c) top view of the sample fracture fatigue fracturesurface tested surface at 450 and MPa; crack (d) close-up nucleation of the area fatigue of thefracture sample surface (a); and (c) crack top viewnucleation of the area sample of the fracture surface testedsample at (c 450). MPa; (d) close-up of the fatigue fracture surface and crack nucleation area of the sample (c). 3.4.2. Fatigue Corrosion 3.4.2. FatigueFigure Corrosion 10 shows the fracture surface of two different AISI304—ASTM A105 specimen tested to FigureFC in10 a shows3.5 wt% theNaCl fracture solution. surfaceThe fatigue of fracture two di ﬀareaerent is very AISI304—ASTM limited in dimension A105 with specimen respect to tested to one of the samples tested to FA and the transition to the unstable fracture area is very marked. The FC in a 3.5unstable wt% fracture NaCl solution. area is characterized The fatigue by the fracture presence area of circular is very lines, limited due into the dimension rotary friction with respect to one of the samples tested to FA and the transition to the unstable fracture area is very marked.

The unstable fracture area is characterized by the presence of circular lines, due to the rotary friction welding process: This fact gives some information about the failure mode of these specimens, as the unstable fracture area is very near to the weld interface. In addition to that, it is also possible to observe that only one main fracture plane was identiﬁed in the analyzed samples and the fracture happened at 0.75 mm from the weld interface. The behavior of the specimens and their lower fatigue cycle number resistance is probably related to the aggressive environment that weakened the specimen causing a localized corrosion attack, as it is possible to see in Figure 10d. In addition to that, it is also possible to observe that only one main fracture plane was identiﬁed in the analyzed samples. Metals 2020, 10, x FOR PEER REVIEW 12 of 16

welding process: This fact gives some information about the failure mode of these specimens, as the unstable fracture area is very near to the weld interface. In addition to that, it is also possible to observe that only one main fracture plane was identified in the analyzed samples and the fracture happened at 0.75 mm from the weld interface. The behavior of the specimens and their lower fatigue cycle number resistance is probably related to the aggressive environment that weakened the Metals 2020specimen, 10, 1018 causing a localized corrosion attack, as it is possible to see in Figure 10d. In addition to that, 11 of 14 it is also possible to observe that only one main fracture plane was identified in the analyzed samples.

(a) (b)

Figure 10.FigureOptical 10. Optical images images of the of fracturethe fracture surface surface of of a a welded welded AISI304—ASTM A105 A105 specimen specimen tested tested to fatigue corrosionto fatigue corrosion (FC): (a) (FC): Top ( viewa) Top of view the of sample the sample fracture fracture surface surface tested tested at at 450 450 MPa; MPa; ((b) close-up close-up of the of the fatigue fracture surface and crack nucleation area of the sample (a); (c) top view of the sample fatigue fracture surface and crack nucleation area of the sample (a); (c) top view of the sample fracture fracture surface tested at 450 MPa; (d) external surface of the sample in (c) near the fracture surface. surface tested at 450 MPa; (d) external surface of the sample in (c) near the fracture surface. 3.4.3. Fatigue Corrosion under Cathodic Protection 3.4.3. Fatigue Corrosion under Cathodic Protection Figure 11 shows the fracture surface of an AISI304—ASTM A105 specimen tested to corrosion Figurefatigue 11 inshows a 3.5 thewt% fracture NaCl solution surface with of ancathodic AISI304—ASTM protection (FC-P). A105 The specimen cathodically tested protected to corrosion fatigue inspecimen a 3.5 wt% shows NaCl a behavior solution similar with cathodic to one of protectionthe specimens (FC-P). tested The in cathodicallyair: The transition protected from the specimen shows afatigue behavior fracture similar area to (at one the ofbottom the specimens of the sample) tested to the in air:unstable The transitionfracture area from is smoother the fatigue with fracture area (atrespect the bottom to those of of the the sample) FC tested to specimens. the unstable Figure fracture 11b shows area a isdetail smoother of the area with between respect these to those two of the regions: It is possible to identify the typical progression marks present in the samples tested to fatigue FC tested[41]. specimens. The failureFigure of the sample11b shows happened a detail at about of the 50 area μm betweenfrom the theseinterface, two as regions: evident from It is possiblethe to identifystructure the typical of the progression failure surface. marks present in the samples tested to fatigue [41]. The failure of the sample happened at about 50 µm from the interface, as evident from the structure of the failure surface. Metals 2020, 10, x FOR PEER REVIEW 13 of 16

(a) (b)

FigureFigure 11. (a 11.) Optical (a) Optical image image of of the the fracture fracture surface surface ofof a welded welded AISI304—ASTM AISI304—ASTM A105 A105 specimen specimen tested tested to protectedto protected fatigue fatigue corrosion corrosion (FC-P); (FC-P); (b ()b SEM) SEM micrograph micrograph of of the the area area in in between between the the fatigue fatigue fracture fracture surfacesurface and theand brittle the brittl fracturee fracture surface. surface.

3.4.4. Accelerated Fatigue Corrosion Figure 12 shows a welded AISI304—ASTM A105 specimen tested to accelerated corrosion fatigue in a 3.5 wt% NaCl solution. The behavior of the accelerated test samples was different, and most of the failure originated from the ASTM A105 side. The sudden fracture area is very irregular, and the fatigue fracture area is very limited in dimension, as is possible to see from Figure 12b.

(a) (b)

Figure 12. (a) Optical image of the fracture surface of a welded AISI304—ASTM A105 specimen tested to accelerated fatigue corrosion (FC-A); (b) SEM micrograph of the area in between the fatigue fracture surface and the unstable fracture surface.

The fractographic analysis was very important to highlight the distinct behavior of the samples tested under different conditions. In particular, the coupling between microscopic analysis with fatigue testing pointed out remarkable differences between the samples. The experimental method and setup seemed to be effective for studying the fatigue corrosion of dissimilar welded joints, although the obtained results should be implemented with further fatigue testing at different stress levels.

4. Conclusions This study pointed out the possibility of studying the fatigue corrosion behavior of dissimilar metal welded joints using a simple testing apparatus and performing the tests in different aggressive environments.

Metals 2020, 10, x FOR PEER REVIEW 13 of 16

(a) (b)

Figure 11. (a) Optical image of the fracture surface of a welded AISI304—ASTM A105 specimen tested Metalsto2020 protected, 10, 1018 fatigue corrosion (FC-P); (b) SEM micrograph of the area in between the fatigue fracture12 of 14 surface and the brittle fracture surface.

3.4.4.3.4.4. Accelerated FatigueFatigue CorrosionCorrosion FigureFigure 12 12 shows shows a weldeda welded AISI304—ASTM AISI304—ASTM A105 A105 specimen specimen tested test to accelerateded to accelerated corrosion corrosion fatigue infatigue a 3.5 wt%in a 3.5 NaCl wt% solution. NaCl solution. The behavior The behavior of the accelerated of the accelerated test samples test samples was diﬀ erent,was different, and most and of themost failure of the originated failure originated from the from ASTM the A105 ASTM side. A105 The si suddende. The fracturesudden fracture area is very area irregular, is very irregular, and the fatigueand the fracture fatigue areafracture is very area limited is very in limited dimension, in dimension, as is possible as is topossible see from to see Figure from 12 Figureb. 12b.

(a) (b)

FigureFigure 12.12. (a) Optical image of the fracture surface of a welded AISI304—ASTM A105A105 specimenspecimen testedtested toto acceleratedaccelerated fatigue fatigue corrosion corrosion (FC-A); (FC-A); (b) SEM(b) SEM micrograph micrograph of the of area the in ar betweenea in between the fatigue the fracturefatigue surfacefracture and surface the unstable and the un fracturestable surface.fracture surface.

TheThe fractographicfractographic analysisanalysis waswas veryvery importantimportant toto highlighthighlight thethe distinctdistinct behaviorbehavior ofof thethe samplessamples testedtested underunder di differentfferent conditions. conditions. In particular,In particular, the couplingthe coupling between between microscopic microscopic analysis analysis with fatigue with testingfatigue pointed testing outpointed remarkable out remarkable differences difference betweens thebetween samples. the Thesamples. experimental The experimental method and method setup seemedand setup to beseemed effective to be for effective studying for the studying fatigue corrosion the fatigue of dissimilarcorrosion weldedof dissimilar joints, welded although joints, the obtainedalthough resultsthe obtained should results be implemented should be withimplemented further fatigue with further testing fatigue at different testing stress at different levels. stress levels. 4. Conclusions 4. ConclusionsThis study pointed out the possibility of studying the fatigue corrosion behavior of dissimilar metal weldedThis joints study using pointed a simple out testing the possibility apparatus and of studying performing the the fatigue tests in corrosion different aggressive behavior environments.of dissimilar metalThe welded home-built joints using setup a provedsimple testing to be e apparatuffective tos and be adaptedperforming to thethe studytests in of different fatigue aggressive corrosion • environments.behavior of the samples under investigation, with the possibility of monitoring the progression of the tests by electrochemical measurements. The tests highlighted an evident coupling between environmental and mechanical testing • conditions by the local electrochemical environment. The cathodic protection of the samples is effective in improving their fatigue behavior, in particular at low stress values. On the contrary, fatigue corrosion or even anodic activation worsen the behavior of the samples. The failure analysis of the samples tested in different conditions showed that in all cases their • failure occurs near the welded interface, in the plastically deformed regions and not in the parent metal. A significant difference was observed among the fracture surface produced by different testing conditions thus confirming that the different severity of the test conditions may change the local damage mechanisms.

Author Contributions: Conceptualization and methodology, S.R., V.F. and M.B.; investigation, A.M.L. and F.R.; formal analysis, F.R.; writing—original draft preparation, F.R.; writing—review and editing, S.R., V.F. and M.B.; supervision, S.R., V.F. and M.B. All authors have read and agreed to the published version of the manuscript. Funding: This research received no external funding. Metals 2020, 10, 1018 13 of 14

Acknowledgments: We gratefully acknowledge Tomet Mechanics (Castelgomberto, Vicenza, VI, Italy) for the samples supplied for the experiments. Conﬂicts of Interest: The authors declare no conﬂict of interest.

References

1. Sakundarini, N.; Taha, Z.; Abdul-Rashid, S.H.; Ghazila, R.A.R. Optimal Multi-material Selection for Lightweight Design of Automotive Body Assembly Incorporating Recyclability. Mater. Des. 2013, 50, 846–857. [CrossRef] 2. Varelis, G.E.; Papatheocharis, T.; Karamanos, S.A.; Perdikaris, P.C. Structural behavior and design of high-strength steel welded tubular connections under extreme loading. Mar. Struct. 2020, 71, 102701. [CrossRef] 3. Wirsching, P.H. Fatigue reliability in welded joints of offshore structures. Int. J. Fatigue 1980, 2, 77–83. [CrossRef] 4. Fan, Y.; Jiang, D.; Mo, D.; Zhu, D.; Luo, Z. A review on dissimilar metals’ welding methods and mechanisms with interlayer. Int. J. Adv. Manuf. Technol. 2019, 102, 2845–2863. 5. Uday, M.B.; Ahmad Fauzi, M.N.; Zuhailawati, H.; Ismail, A.B. Advances in friction welding process: A review. Sci. Technol. Weld. Join. 2010, 15, 534–558. [CrossRef] 6. Martinsen, K.; Hu, S.J.; Carlson, B.E. Joining of dissimilar materials. CIRP Ann. Manuf. Technol. 2015, 64, 679–699. [CrossRef] 7. Cai, W.; Daehn, G.; Li, G.; Mishra, R.; Vivek, A.; Khan, H.; Komarasamy, M. A state-of-the-art review on solid state metal joining. J. Manuf. Sci. Eng. 2019, 141, 031012. [CrossRef] 8. Wenya, L.; Vairis, A.; Preuss, M.; Tiejun, M. Linear and rotary friction welding review. Int. Mater. Rev. 2015, 61, 71–100. 9. Vairis, A.; Papazafeiropoulos, G.; Tsainis, A. A comparison between friction stir welding, linear friction welding and rotary friction welding. Adv. Manuf. 2016, 4, 296–304. [CrossRef] 10. Klomp, J.T.; De With, G. Strong metal-ceramic joints. Mater. Manuf. Process. 1993, 8, 129–157. [CrossRef] 11. Meshram, S.D.; Mohandas, T.; Reddy, G.M. Friction welding of dissimilar pure metals. J. Mater. 2007, 184, 330–337. [CrossRef] 12. Rupinder, S.; Ranvijay, K.; Feo, L.; Fraternali, F. Friction welding of dissimilar plastic/polymer materials with metal powder reinforcement for engineering applications. Compos. Part B Eng. 2016, 101, 77–86. 13. Arivazhagan, N.; Surendra, S.; Prakash, S.; Reddy, G.M. Investigation on AISI 304 austenitic stainless steel to AISI 4140 low alloy steel dissimilar joints by gas tungsten arc, electron beam and friction welding. Mater. Des. 2011, 32, 3036–3050. [CrossRef] 14. Carlone, P.; Astarita, A. Dissimilar metal welding. Metals 2019, 9, 1206. [CrossRef] 15. Sathiya, P.; Aravindan, S.; Noorul Haq, A.; Paneerselvam, K. Optimization of friction welding parameters using evolutionary computational techniques. J. Mater. 2009, 209, 2576–2584. [CrossRef] 16. Nascimento, M.P.; Souza, R.C.; Pigatin, W.L.; Voorwald, J.H.C. Effects of surface treatments on the fatigue strength of AISI 4340 aeronautical steel. Int. J. Fatigue 2001, 23, 607–618. [CrossRef] 17. Shushan, S.M.; Charles, E.A.; Congleton, J. The environment assisted cracking of diffusion bonded stinless steel to carbon steel joints in an aqueous chloride solution. Corros. Sci. 1996, 38, 673–686. [CrossRef] 18. Sarkari Khorrami, M.; Mostafaei, M.A.; Pouraliakbar, H.; Kokabi, A.H. Study on microstructure and mechanical characteristics of low-carbon steel and ferritic stainless steel joints. Mater. Sci. Eng. A 2014, 608, 35–45. [CrossRef] 19. Prasad, K.S.; Rao, C.S.; Rao, D.N. A review on welding of AISI 304L austenitic stainless steel. J. Manuf. Sci. Prod. 2012, 14, 1–11. [CrossRef] 20. Maddox, S.J. Fatigue behaviour of welded joints. In Advances in Fatigue Science and Technology; Branco, C.M., Rosa, L.G., Eds.; Springer: Dordrecht, The Netherlands, 1989; Volume 159. 21. Akiniwa, Y.; Miyamoto, N.; Tsuru, H.; Tanaka, K. Notch effect on fatigue strength reduction of bearing steel in the very high cycle regime. Int. J. Fatigue 2006, 28, 1555–1565. [CrossRef] 22. Casavola, C.; Giordano, V.; Pappalettere, P.; Pruncu, C.I. Influence of geometrical shape of specimen in fatigue life characterization on welded joint in titanium alloy. In Proceedings of the Twelfth Meeting New Trends in Fatigue and Fracture, Brasov, Romania, 27–30 May 2012. 23. Alam, M.M.; Barsoum, Z.; Jonsén, P.; Kaplan, A.F.H.; Häggblad, H.Å. The influence of surface geometry and topography on the fatigue cracking behaviour of laser hybrid welded eccentric fillet joints. Appl. Surf. Sci. 2010, 256, 1936–1945. [CrossRef] Metals 2020, 10, 1018 14 of 14

24. Bertini, L.; Fontanari, V.; Straffelini, G. Influence of post weld treatments on the fatigue behaviour of Al-alloy welded joints. Int. J. Fatigue 1998, 20, 749–755. [CrossRef] 25. Sabzi, M.; Dezfuli, S.M. Post weld heat treatment of hypereutectoid hadfield steel: Characterization and control of microstructure, phase equilibrium, mechanical properties and fracture mode of welding joint. J. Manuf. Process. 2018, 34, 313–328. [CrossRef] 26. Revie, R.W. Environmental cracking of metals. In Modern Aspects of Electrochemistry; Conway, B.E., Bockris, J.O., Eds.; Springer: Boston, MA, USA, 1994; Volume 26. 27. Andersen, P.L. ASM Handbook—Vol. 19. Corrosion Fatigue Testing; ASM International: Geauga County, OH, USA, 1998. 28. Schijve, J. Fatigue of Structures and Materials; Springer: Dordrecht, The Netherlands, 2009; pp. 457–479. 29. Andersen, P.L.; Jones, R.H. Stress Corrosion Cracking—Material Performance and Evaluation; ASM International: Geauga County, OH, USA, 1992. 30. Sieradzki, K.; Newman, R.C. Stress-corrosion cracking. J. Phys. Chem. Solids 1987, 48, 1101–1113. [CrossRef] 31. ASTM International. ASTM A276 Standard, Standard Specification for Stainless Steel Bars and Shapes; ASTM International: West Conshohocken, PA, USA, 2017. 32. ASTM International. ASTM A105 Standard, Standard Specification for Carbon Steel Forgings for Piping Applications; ASTM International: West Conshohocken, PA, USA, 2018. 33. UNI Ente Nazionale Italiano di Unificazione. UNI EN ISO 11782-1 Standard, Corrosion of Metals and Alloys—Corrosion Fatigue Testing Cycles to Failure Testing; UNI Ente Nazionale Italiano di Unificazione: Milano, Italy, 2008. 34. Fontana, M.G. Corrosion Engineering; Tata McGraw-Hill Education: New York, NY, USA, 2005. 35. ASTM International. ASTM G8-96 Standard, Standard Test Methods for Cathodic Disbonding of Pipeline Coatings; ASTM International: West Conshohocken, PA, USA, 2019. 36. Ananthapadmanaban, D.; Seshagiri Rao, V.; Abraham, N.; Prasad Ra, K. A study of mechanical properties of friction welded mild steel to stainless steel joints. Mater. Des. 2009, 30, 2642–2646. [CrossRef] 37. Özdemir, N. Investigation of the mechanical properties of friction-welded joints between AISI 304L and AISI 4340 steel as a function rotational speed. Mater. Lett. 2005, 59, 2504–2509. [CrossRef] 38. Sahin, M. Joining with friction welding of high-speed steel and medium-carbon steel. J. Mater. Process. Technol. 2005, 168, 202–210. [CrossRef] 39. Yoon, H.-K.; Kong, Y.-S.; Kim, S.-J.; Kohyama, A. Mechanical properties of friction welds of RAFs (JLF-1) to SUS304 steels as measured by the acoustic emission technique. Fusion Eng. Des. 2006, 81, 945–950. [CrossRef] 40. Rowlands, D.P. The Mechanical Properties of Stainless Steel. In Sassda Stainless Steel Information Series No. 3; Southern Africa Stainless Steel Development Association: Sandton, South Africa, 2006. 41. MT Acciai Speciali. Acciai al Carbonio. Available online: http://www.mtacciai.com/it/acciai/al-carbonio/ astm-a105 (accessed on 21 July 2020). 42. Sachs, N.W. Understanding the Surface Features of Fatigue Fractures: How They Describe the Failure Cause and the Failure History. J. Fail. Anal. Prev. 2005, 2, 11–15. [CrossRef]

© 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/).