Influence of Metallic Oxide Nanoparticles on the Mechanical Properties of an A-TIG Welded 304L Austenitic Stainless Steel

materials

Article Inﬂuence of Metallic Oxide Nanoparticles on the Mechanical Properties of an A-TIG Welded 304L Austenitic Stainless Steel

Sebastian Balos 1,*, Miroslav Dramicanin 1 , Petar Janjatovic 1 , Nenad Kulundzic 1, Ivan Zabunov 2, Branka Pilic 3 and Damjan Klobˇcar 4,* 1 Faculty of Technical Sciences, University of Novi Sad, Trg Dositeja Obradovica 6, 21000 Novi Sad, Serbia; [email protected] (M.D.); [email protected] (P.J.); [email protected] (N.K.) 2 Faculty of Special Technology, Alexander DubˇcekUniversity of Trenˇcín, Študentská 2, 911 50 Trenˇcín, Slovakia; [email protected] 3 Faculty of Technology, University of Novi Sad, Bulevar Cara Lazara 1, 21000 Novi Sad, Serbia; [email protected] 4 Faculty of Mechanical Engineering, University of Ljubljana, Aškerˇcevac. 6, 1000 Ljubljana, Slovenia * Correspondence: [email protected] (S.B.); [email protected] (D.K.); Tel.: +381-214-852-339 (S.B.)

Received: 25 August 2020; Accepted: 11 October 2020; Published: 12 October 2020

Abstract: Austenitic stainless steels represent a significant aerospace material, being used for various castings, structural components, landing gear components, afterburners, exhaust components, engine parts, and fuel tanks. The most common joining process is tungsten inert gas (TIG) welding, which possesses many advantages such as suitability to weld a wide range of ferrous and non-ferrous metals and alloys, providing high quality welds with good mechanical properties. Its major disadvantage is low productivity due to low penetration and welding speed. This can be overcome by introducing an activating flux before welding. The activating flux reverses the material flow of the weld pool, significantly increasing penetration. Therefore, shielding gas consumption is reduced and welding without a consumable is enabled. However, the consumable in conventional TIG also enables the conditioning of the mechanical properties of welds. In this study, Si and Ti metallic oxide nanoparticles were used to increase the weld penetration depth, while bend testing, tensile, and impact toughness were determined to evaluate the mechanical properties of welds. Furthermore, optical emission spectroscopy, light, and scanning electron microscope were used to determine the chemical compositions and microstructures of the welds. Chemical compositions and weld mechanical properties were similar in all specimens. The highest tensile and impact properties were obtained with the specimen welded with the flux containing 20% TiO2 and 80% SiO2 nanoparticles. Although lower than those of the base metal, they were well within the nominal base metal mechanical properties.

Keywords: oxide ﬂux; activated tungsten inert gas welding; mechanical properties; 304L stainless steel

1. Introduction Austenitic stainless steels are used in various applications due to their excellent corrosion resistance in atmospheric and pure water environments, marine environments as well as in acid and alkaline solutions. Their advantages also present their unique properties such as hardness, strength, density, malleability, ductility at lower temperatures, elasticity, conductivity, thermal expansion, good heat, and ﬁre resistance. Austenitic stainless steels are also easily fabricated using forming, machining, cutting, and welding, which makes them good materials for producing components in the aerospace industry. Common applications of austenitic stainless steels in aerospace applications include various castings, structural components, landing gear components, afterburners, exhaust components, engine

Materials 2020, 13, 4513; doi:10.3390/ma13204513 www.mdpi.com/journal/materials Materials 2020, 13, 4513 2 of 11 parts, and fuel tanks. Depending on the use and desired properties of the components, the following austenitic stainless steels are usually used in aerospace applications: 304, 316, and 321 [1–3]. The dominant welding process common for welding austenitic stainless steels is tungsten inert arc (TIG) welding. To increase productivity, the activated flux can be used prior to TIG welding, resulting in the modified process called activated tungsten inert gas welding (A-TIG), which has significantly increased penetration that is sufficient for full penetration of the whole plate thickness. Therefore, in A-TIG, common V-preparation can be replaced by a simpler and cheaper closed square preparation of the base metal, since the whole contact surface between base metals can be welded in one pass. This renders multi-pass welding with repeated welding of incremental welds with consumable material unnecessary. Consumables are not needed, a considerably lower amount of shielding gas is expended, and less power and time is needed. A-TIG has been demonstrated to be effective in welding various base metals such as different types of steels, titanium alloys, aluminum alloys, magnesium alloys, and dissimilar materials [4,5]. All these are accompanied by a considerable energy consumption and a lower man-power demand [6,7]. A-TIG was invented in the former USSR (Union of Soviet Socialist Republics), more accurately, at the Paton Welding Institute in Kiev, present-day Ukraine [8]. It usually consists of metallic oxides (SiO2, TiO2, Al2O3, NiO, MnO3, Cr2O3, CeO2, and V2O5) in alcohol or acetone solvents [5,9,10]. Aside from the flux particle type, the particle size is also influential, with a preference for nanoparticles. Several studies have proven that nano-sized particles possess a high efficiency. This is because the high specific surface area of the nano-sized particles results in a faster thermal dissociation and decomposition during arc heating compared to that of the micron-sized particles [11–13]. The flux is applied on the surface by a brush or in the form of a spray prior to welding. There are two major mechanisms of increasing the penetration: the reversal of molten metal flow and arc constriction. In conventional TIG welding, the molten metal flows from the center of the weld (low surface tension) toward the edges (high surface tension area), creating a relatively wide but shallow weld. In A-TIG welding, this is reversed due to the influence of oxygen from the flux, resulting in a completely different weld shape: narrow and deep [14,15]. The arc constriction is the result of the electronegativity of the metallic constituents in the flux [16,17]. A-TIG has been successfully used for welding a number of materials, ferrous and non-ferrous as well as dissimilar materials [18,19]. Although A-TIG possesses many advantages, the lack of a consumable may affect the mechanical properties of the weld itself. Namely, the consumable material in conventional TIG in overmatching welds has a higher ultimate tensile strength (UTS) compared to the base metal, thus influencing the rise in the UTS of the weld itself. However, there have been few studies addressing this important issue with A-TIG. Sharma and Dwivedi [20] experimented with the dissimilar welding of P92 and 304H steels, where multipass TIG (with Inconel 82 wire consumable) and A-TIG (without consumable) were compared. It was found that the weld metal obtained by A-TIG contained untempered martensite, compared to austenite in the weld metal obtained by TIG. This resulted in a higher strength, but a lower impact toughness of the weld metal in A-TIG compared to TIG. In A-TIG welded specimens, tensile testing resulted in the fracture that occurred in the 304H base metal, while in TIG, the fracture occurred in the weld metal. Ramkumar et al. [21] studied the influence of activated flux in forming dissimilar welds between 316L austenitic stainless steel and S32750 super-duplex stainless steel. It was shown that the tensile fracture occurred in 316L steel and not in the weld metal or stronger S32750 steel. Hdhibi et al. [22] experimented with the remelting of 316L steel with TIG and various compositions of activated flux. The remelted specimens were tensile tested and it was determined that δ-ferrite plays a major role in increasing the tensile properties of predominantly austenitic material. The influence of nitrogen content in shielding gas (Ar based) composition was studied in [23] where it was found that the nitrogen content had a positive effect on the tensile properties and hardness of 304 austenitic stainless-steel A-TIG welds due to interstitial solid solution strengthening and grain refinement. Materials 2020, 13, 4513 3 of 11

Instead of aiming at optimizing the flux to achieve the maximum penetration as in our previous study [12], in this study, the influence of nanoparticle-based fluxes on the mechanical properties of the weld was determined. The obtained mechanical properties were compared to the base metal and procedures for determining the weld quality, which may be the impeding factor of the wider acceptance of A-TIG technology in the industrial sector. The main novelty identified was finding the effect of nanoparticles in the flux and their effect on the weld shape and mechanical properties of welds.

2. Materials and Methods In this study, the base metal used was AISI 304L austenitic stainless steel. Its chemical composition is presented in Table1, while its mechanical properties are shown in Table2. The base material size was cut from the 10 mm plate to 200 mm 50 mm dimensions by waterjet, and then butt welds were × done without a gap.

Table 1. Chemical composition of the base metal wt.%.

C Si Mn Cr Ni P S Fe 0.02 0.042 1.55 18.63 8.23 0.032 0.0004 balance

Table 2. Mechanical properties of the 304L base metal and nominal base metal properties (ASTM A240/240M; EN ISO 10088-2).

Tensile Properties Impact Energy

Material Yield Ultimate Reduction Crack Crack Speciﬁcations Tensile Elongation Impact Strength Strength A (%) of Area Z Formation Propagation Energy (J) Rp (MPa) Rm (MPa) (%) (J) (J) Base metal 474 697 57 60 107 166 273

304L nominal ––– values >170 >485 >40 >100

Prior to welding, several nanoparticle-based fluxes in acetone solvents were prepared. The active element of the fluxes was 20 nm TiO2 and SiO2 nanoparticles, weighed by a Tehtnica Type 2615 (Zelezniki, Slovenia) analytic balance, mixed with the solvent by means of a Tehtnica mm530 (Zelezniki, Slovenia) magnetic stirrer for 10 min. After mixing, the fluxes were smeared over the surface of the specimens to be welded by a brush, having the width of 20 mm at both base metal sides and were 10 µm thick, measured immediately after application. Subsequently, the welding was performed by using an EWM Tetrix300 TIG welding machine (EWM, Mündersbach, Germany). Welding current was 250 A and the WTh2 electrode tip was sharpened to 90◦. The distance from the tip to the base metal was 2 mm, while the welding speed was set and kept constant at 40 mm/min. All specimens were welded without a consumable and with Ar shielding gas at a 14 L/min flow rate as well as a ceramic backing plate. The welded specimens were designated in relation to the flux applied during welding: 5Ti (100% TiO2), 4Ti1Si (80% TiO2; 20% SiO2), 3Ti2Si (60% TiO2; 40% SiO2), 2Ti3Si (40% TiO2; 60% SiO2), 1Ti4Si (20% TiO2; 80% SiO2), and 5Si (100% SiO2). Weld characterization was done in macro imagery, microstructures, microhardness and tensile, bend, and impact testing. Tensile testing was conducted in accordance to EN ISO 4136:2012 standard, impact energy in accordance to EN ISO 9016:2012, bend testing with EN ISO5173:2009, and microhardness with EN ISO 6507:2018. Specimens were cut by a waterjet, in accordance with Figure1. In Figure1, the weld direction is indicated. Metallographic preparation was done on Struers equipment (Bellerup, Denmark) and consisted of cutting, mounting, abrasive paper grinding (150 to 2000 grit), polishing with diamond suspensions (6 to 0.25 µm particle size), and etching by using aqua regia (three parts 37% HCl; 1 part 70% HNO3). Examinations were done with a Leitz Orthoplan light microscope (Oberkochen, Germany) and JEOL JSM6460LV (Tokyo, Japan) scanning electron microscope (SEM). Microhardness was measured by a Wilson Tukon 1102 (Uzwil, Switzerland) device Materials 2020, 13, x FOR PEER REVIEW 4 of 11 Materials 2020, 13, x FOR PEER REVIEW 4 of 11 Materials 2020, 13, 4513 4 of 11 device with 100 gf loading in the base metal and weld metal, as shown in Figure 2. The distance device with 100 gf loading in the base metal and weld metal, as shown in Figure 2. The distance between the indentations was 0.5 mm. Weld metal chemical analysis was done by an ARL ISpark between the indentations was 0.5 mm. Weld metal chemical analysis was done by an ARL ISpark with8860 100(Waltham, gf loading MA, in USA) the base optical metal emission and weld spectrometer metal, as shown (OES). in Tensile Figure 2and. The bend distance testing between were done the 8860 (Waltham, MA, USA) optical emission spectrometer (OES). Tensile and bend testing were done indentationsby the Schenck was Hydropuls 0.5 mm. WeldPSB 250 metal (Darmstadt, chemical Germany) analysis was universal done by hydraulic an ARL ISparktesting 8860machine, (Waltham, while by the Schenck Hydropuls PSB 250 (Darmstadt, Germany) universal hydraulic testing machine, while MA,impact USA) testing optical was emission performed spectrometer on a Testing (OES). Equipment Tensile and IE bendJWT‐ testing450 (Jinan, were China) done by instrumented the Schenck impact testing was performed on a Testing Equipment IE JWT‐450 (Jinan, China) instrumented HydropulsCharpy pendulum PSB 250 tester. (Darmstadt, Image Germany)analyses to universal obtain the hydraulic δ‐ferrite content testing machine, was done while by ImageJ impact software testing Charpy pendulum tester. Image analyses to obtain the δ‐ferrite content was done by ImageJ software was(version performed 1.51). on a Testing Equipment IE JWT-450 (Jinan, China) instrumented Charpy pendulum tester.(version Image 1.51). analyses to obtain the δ-ferrite content was done by ImageJ software (version 1.51).

Figure 1. Specimens in a welded plate: M‐macro; T‐tensile testing; B‐bend testing; I‐impact testing, Figure 1.1. Specimens in a welded plate: M M-macro;‐macro; T T-tensile‐tensile testing; B-bendB‐bend testing; I I-impact‐impact testing, with indicated welding direction (WD) and specimen sizes, the thickness being kept as the plate itself. with indicated welding direction (WD) and specimen sizes, the thickness being kept as the plate itself. itself. The dashed line indicates the position of the weld. The dashed lineline indicatesindicates thethe positionposition ofof thethe weld.weld.

Figure 2. Microhardness measurement scheme. Figure 2. Microhardness measurement scheme. 3. Results and Discussion 3. Results and Discussion 3. Results and Discussion 3.1. Metallographic Testing and Chemical Compositions 3.1. Metallographic Testing and Chemical Compositions 3.1. MetallographicMacro-sections Testing of the and welded Chemical specimens Compositions are shown in Figure3. Although similar in shape and Macro‐sections of the welded specimens are shown in Figure 3. Although similar in shape and columnarMacro morphology,‐sections of the typical welded for A-TIGspecimens weldments, are shown there in wasFigure a marked 3. Although difference similar in in width. shape In and all columnar morphology, typical for A‐TIG weldments, there was a marked difference in width. In all specimens,columnar morphology, the root was typical significantly for A‐TIG wider weldments, than the there face, was as was a marked obtained difference in [12], whichin width. was In the all specimens, the root was significantly wider than the face, as was obtained in [12], which was the resultspecimens, of the the heat root accumulation was significantly and the wider ceramic than backing the face, plate, as was which obtained limited in excessive [12], which penetration. was the result of the heat accumulation and the ceramic backing plate, which limited excessive penetration. Thisresult indicates of the heat that accumulation the maximum and penetration the ceramic with backing applied parametersplate, which can limited be higher, excessive that is, penetration. the present This indicates that the maximum penetration with applied parameters can be higher, that is, the parametersThis indicates can that be applied the maximum to weld a penetration thicker base with metal applied of the same parameters type. Alternatively, can be higher, this that means is, thatthe present parameters can be applied to weld a thicker base metal of the same type. Alternatively, this thepresent same parameters thickness as can applied be applied in this to experiment weld a thicker could base be weldedmetal of with the same a lower type. welding Alternatively, current, orthis a means that the same thickness as applied in this experiment could be welded with a lower welding highermeans weldingthat the same speed. thickness The widest as applied weld was in obtainedthis experiment in the 1Ti4Si could specimen.be welded However, with a lower it is welding difficult current, or a higher welding speed. The widest weld was obtained in the 1Ti4Si specimen. However, it tocurrent, assess or the a higher maximum welding penetration speed. The of tested widest fluxes weld withwas obtained the parameters in the 1Ti4Si used inspecimen. this experiment However, just it is difficult to assess the maximum penetration of tested fluxes with the parameters used in this byis difficult evaluating to theassess weld the metal maximum shape frompenetration the macro-sections. of tested fluxes Moreover, with the it is parameters difficult to compareused in this the experiment just by evaluating the weld metal shape from the macro‐sections. Moreover, it is difficult to presentexperiment macro-sections just by evaluating to the the macro-sections weld metal shape obtained from by the the macro same‐sections. fluxes, becauseMoreover, in it [12 is ],difficult different to compare the present macro‐sections to the macro‐sections obtained by the same fluxes, because in [12], weldingcompare parameters the present weremacro used.‐sections At higher to the currents,macro‐sections larger obtained particles mayby the be same optimal fluxes, for because achieving in high[12], different welding parameters were used. At higher currents, larger particles may be optimal for penetration,different welding but this parameters remains to were be proven used. inAt future higher studies. currents, larger particles may be optimal for achieving high penetration, but this remains to be proven in future studies. achieving high penetration, but this remains to be proven in future studies.

Materials 2020, 13, x FOR PEER REVIEW 5 of 11

In Table 3, the widths of the weld metals are shown. The addition of SiO2 nanoparticles makes the weld face narrower due to the arc constriction effect of the Si, with the only exception being 1Ti4Si. This is in contrast to the results shown in [12], where the 3Ti2Si specimen proved to have the highest penetration by using the same fluxes. However, it must be noted that the current used in this study was 300 A, which is considerably higher than that in [12], where it was 200 A. Obviously, 1Ti4Si represents the optimal mixture from the point of view of the welding parameters, most notably the welding current. The microstructures obtained with the light microscope of specimen 5Ti is shown in Figure 4, where typical dendritic weld metal and transition weld interface microstructures were obtained. The SEM microstructure revealed the typical δ‐ferrite morphology that occurs between austenitic dendrites. The SEM micrograph is shown in Figure 5, presenting δ‐ferrite in the austenite matrix of the weld metal. There were no differences between the microstructures obtained with different experimental conditions.Figure 3. TheMacro-sections δ‐ferrite content of specimens in the welded base metal with variouswas 6%, types while of flux. in the weld metal of Figure 3. Macro‐sections of specimens welded with various types of flux. all specimens, it was 7.63% to 8.01%. In Table3, the widths of the weld metals are shown. The addition of SiO 2 nanoparticles makes Table 3. Weld metal widths. the weld face narrower due to the arc constriction effect of the Si, with the only exception being 1Ti4Si. This is in contrastSpecimen to the results shown5Ti in [12],4Ti1Si where the 3Ti2Si3Ti2Si specimen 2Ti3Si proved to1Ti4Si have the highest5Si penetration by using the sameFace fluxes. 10.5 However, 9.9 it must be noted7.7 that the8.7 current used14.0 in this6.7 study Weld width (mm) was 300 A, which is considerablyRoot higher9.4 than8.5 that in [12],8.5 where it was8.6 200 A. Obviously,10.8 8.8 1Ti4Si represents the optimal mixture from the point of view of the welding parameters, most notably the welding current.

Table 3. Weld metal widths.

Specimen 5Ti 4Ti1Si 3Ti2Si 2Ti3Si 1Ti4Si 5Si

Weld width Face 10.5 9.9 7.7 8.7 14.0 6.7 (mm) Root 9.4 8.5 8.5 8.6 10.8 8.8

The microstructures obtained with the light microscope of specimen 5Ti is shown in Figure4, where typical dendriticFigure 3. weld Macro metal‐sections and of transitionspecimens welded weld interface with various microstructures types of flux. were obtained. The SEM microstructure revealed the typical δ-ferrite morphology that occurs between austenitic Table 3. Weld metal widths. dendrites.Figure The4. Microstructures SEM micrograph of the is 5Ti shown specimen in Figure obtained5, presenting by light microscope:δ-ferrite weld in the metal austenite (a); weld matrix of theinterface weld metal. Specimen(b). There were no diﬀ5Tierences 4Ti1Si between the3Ti2Si microstructures 2Ti3Si obtained1Ti4Si with di5Siﬀerent experimental conditions. TheFaceδ-ferrite 10.5 content in9.9 the base metal7.7 was 6%, while8.7 in the14.0 weld metal6.7 of all Weld width (mm) specimens, it was 7.63% to 8.01%.Root 9.4 8.5 8.5 8.6 10.8 8.8

Figure 4. MicrostructuresMicrostructures of ofthe the 5Ti 5Ti specimen specimen obtained obtained by light by light microscope: microscope: weld weldmetal metal(a); weld (a); weldinterface interface (b). (b).

Materials 2020, 13, 4513 6 of 11 Materials 2020, 13, x FOR PEER REVIEW 6 of 11

Figure 5. Scanning electron microscope (SEM) image of the δ-ferrite in the weld metal of the Figure 5. Scanning electron microscope (SEM) image of the δ‐ferrite in the weld metal of the 5Ti 5Ti specimen. specimen. Chemical compositions of weld metals obtained by OES as well as the 304L steel nominal chemical compositionChemical are compositions shown in Table of4 .weld It can metals be seen obtained that all obtained by OES chemical as well compositions as the 304L ofsteel weld nominal metals correspondedchemical composition well to the are nominal shown in chemical Table 4. composition It can be seen of 304Lthat all austenitic obtained stainless chemical steel. compositions Additionally, of theweld weld metals metal corresponded chemical compositions well to the nominal corresponded chemical well composition to the chemical of 304L compositionaustenitic stainless of the steel. base metalAdditionally, shown in the Table weld1, indicatingmetal chemical that the compositions flux does not corresponded influence changes well to in the chemical compositioncomposition of the base weld metal metal. shown in Table 1, indicating that the flux does not influence changes in the chemical composition of the weld metal. Table 4. Chemical compositions of the weld metals and 304L austenitic stainless steel in accordance withTable ASTM 4. Chemical A240/A240M, compositions wt.%. of the weld metals and 304L austenitic stainless steel in accordance with ASTM A240/A240M, wt.%. Specimen C Si Mn Cr Ni P S Fe Specimen C Si Mn Cr Ni P S Fe 5Ti 0.02 0.45 1.5 18.57 8.29 0.032 0.0006 balance 5Ti 0.02 0.45 1.5 18.57 8.29 0.032 0.0006 balance 4Ti1Si 0.02 0.45 1.51 18.55 8.24 0.032 0.0006 balance 3Ti2Si4Ti1Si 0.02 0.02 0.450.45 1.531.51 18.5618.55 8.24 8.2 0.032 0.03 0.0006 0.0007 balance balance 2Ti3Si3Ti2Si 0.02 0.02 0.450.45 1.511.53 18.5718.56 8.2 8.24 0.03 0.032 0.0007 0.0007 balance balance 1Ti4Si2Ti3Si 0.025 0.02 0.450.45 1.531.51 18.618.57 8.24 8.26 0.032 0.032 0.0007 0.0006 balance balance 5Si1Ti4Si 0.025 0.025 0.44 0.45 1.551.53 18.6318.6 8.26 8.23 0.032 0.031 0.0006 0.0004 balance balance AISI 304L5Si 0.02 0.0250.045 0.44 1.551.55 18.6318.63 8.23 8.23 0.031 0.032 0.0004 0.0004 balance balance ≤ ≤ AISI 304L 0.02 ≤0.045 ≤1.55 18.63 8.23 0.032 0.0004 balance 3.2. Mechanical Properties of Welds 3.2. Mechanical Properties of Welds The tensile properties of the welds are shown in Figure6, while the typical fracture mode is shown in FigureThe 7tensile. Tensile properties properties of the in welds all specimens are shown were in Figure lower compared6, while the to typical the base fracture metal, mode which is shown could bein Figure the result 7. Tensile of the properties replacement in all of specimens polygonal were austenitic lower compared by a cast dendriticto the base microstructure, metal, which could that be is, athe polygonal result of withthe replacement dendritic austenite. of polygonal Tensile austenitic strengths by and a reductionscast dendritic of areamicrostructure, were well within that is, the a limitspolygonal set upwith by dendritic ASTM A240 austenite./A240M Tensile relating strengths to 304L and austenitic reductions stainless of area were steel. well The within highest the tensile limits strengthset up by and ASTM reduction A240/A240M in area relating of all welded to 304L specimens austenitic were stainless obtained steel. in The the highest 1Ti4Si specimen, tensile strength as the resultand reduction of the widest in area weld of all metal welded and thespecimens widest spacewere obtained for deformation. in the 1Ti4Si This isspecimen, also important as the fromresult the of pointthe widest of view weld of penetrationmetal and the potential, widest space since for this deformation. mixture influences This is thealso largest important weld from metal. the Fracture point of modeview of is penetration fully ductile, potential, with a notable since this necking, mixture as influences shown in the Figure largest7. This weld was metal. proven Fracture by the mode SEM is imagefully ductile, analysis with (Figure a notable8). A necking, dimpled as specimen shown fracturein Figure surface 7. This was was obtained, proven by typical the SEM of ductile image fracture.analysis (Figure This is the 8). resultA dimpled of microvoid specimen formation fracture around surface non-metallic was obtained, inclusions, typical of microvoid ductile fracture. growth, andThis coalescenceis the result [ 24of]. microvoid At the bottom formation of the around dimples, non circular‐metallic phases inclusions, can be observed,microvoid which growth, can and be morphologicallycoalescence [24]. attributedAt the bottom to non-metallic of the dimples, inclusions circular (Figure phases8c,d). can be observed, which can be morphologically attributed to non‐metallic inclusions (Figure 8c,d).

Materials 2020, 13, x FOR PEER REVIEW 7 of 11 Materials 2020, 13, 4513x FOR PEER REVIEW 7 of 11 Materials 2020, 13, x FOR PEER REVIEW 7 of 11

Figure 6. TensileTensile properties of the welds.welds. Figure 6. Tensile properties of the welds.

Figure 7. NeckingNecking in in two two tensile tensile 1Ti4Si specimens: closer to the weld beginning ( a); closer to the weld Figure 7. Necking in two tensile 1Ti4Si specimens: closer to the weld beginning (a); closer to the weld end ( b)).. end (b).

Figure 8. SEM images of fracture surface obtained by tensile testing of the 4Ti1Si4Ti1Si specimen:specimen: macro Figure 8. SEM images of fracture surface obtained by tensile testing of the 4Ti1Si specimen: macro image ( a); micro images showing ductile fracture ( b); dimples ( (c); dimples with indicated indicated non non-metallic‐metallic image (a); micro images showing ductile fracture (b); dimples (c); dimples with indicated non‐metallic inclusions (d). inclusions (d).

MaterialsMaterials 20202020,, 1313,, 4513x FOR PEER REVIEW 88 ofof 1111

The highest tensile strength is the result of the highest absolute amount of δ‐ferrite in the austeniticThe highest matrix, tensile which strength is proportional is the result to ofthe the weld highest metal absolute width, amount most particularly of δ-ferrite inthe the weld austenitic metal matrix,width, as which shown is proportionalin Table 3. The to thebody weld‐centered metal cubic width, crystal most lattice particularly (BCC) had the weldhigher metal mechanical width, asproperties shown in compared Table3. The to body-centeredthe austenitic face cubic‐centered crystal lattice cubic (BCC)lattice. had This higher is similar mechanical to the findings properties of comparedHdhibi et toal. the [22], austenitic who also face-centered stressed the cubic significance lattice. This of δ‐ isferrite similar in to obtaining the findings the of convenient Hdhibi et al.tensile [22], whoproperties also stressed of 316L the austenitic significance stainless of δ-ferrite steel inA‐ obtainingTIG remelting. the convenient Furthermore, tensile the properties tensile strengths of 316L austeniticobtained in stainless this study steel were A-TIG very remelting. similar Furthermore,to those obtained the tensile in Hdhibi strengths et al. obtained [22], who in used this study 316L wereaustenitic very stainless similar to steel those welded/remelted obtained in Hdhibi by A‐TIG et al. with [22], the who flux used based 316L on SiO austenitic2 and TiO stainless2. However, steel weldedin this study,/remelted a positive by A-TIG correlation with the between flux based SiO2 on nanoparticle SiO2 and TiO content2. However, and weld in thismetal study, root width a positive and correlationtensile strength between could SiO be2 nanoparticle observed. Additionally, content and weld when metal single root component width and fluxes tensile are strength considered, could be it observed.was found Additionally, that the specimen when singlewelded component with the flux fluxes containing are considered, SiO2 nanoparticles it was found exhibited that the specimen a higher weldedtensile strength with the and flux containinga lower reduction SiO2 nanoparticles in area compared exhibited to athe higher specimen tensile welded strength with and the a lower flux reductioncontaining in TiO area2 nanoparticles. compared to the specimen welded with the flux containing TiO2 nanoparticles. ImpactImpact energiesenergies obtainedobtained byby thethe instrumentedinstrumented CharpyCharpy pendulum tester, withwith V-notchV‐notch situatedsituated inin thethe centercenter ofof thethe weldweld metalmetal areare shownshown inin FigureFigure9 9.. As As in in base base metal metal (Table (Table2 ),2), crack crack propagation propagation valuesvalues werewere higherhigher comparedcompared to to thethe crackcrack initiation energyenergy values.values. However, allall impact energy values werewere lowerlower thanthan thosethose ofof thethe basebase metal.metal. ValuesValues forfor overalloverall impactimpact energiesenergies werewere betweenbetween 80%80% andand 95%95% ofof thethe basebase metal.metal. The highesthighest overalloverall impact energies were obtainedobtained inin specimensspecimens 2Ti3Si2Ti3Si andand 1Ti4Si.1Ti4Si. TheThe presencepresence ofof thethe lessless ductileductile δ‐δ-ferriteferrite mightmight explainexplain whywhy weldweld metalsmetals havehave aa slightlyslightly lowerlower impactimpact energiesenergies comparedcompared toto thethe basebase metal.metal. FigureFigure9 9a,ba,b depicts depicts the the typical typical force–time force–time curves curves from from thethe instrumentedinstrumented CharpyCharpy pendulumpendulum testertester ofof thethe basebase metalmetal andand 3Ti2Si3Ti2Si specimen.specimen. ItIt cancan bebe seenseen thatthat thethe curvecurve shapeshape waswas very similar, however,however, the the overalloverall impactimpact energyenergy ofof thethe basebase metalmetal specimenspecimen waswas higherhigher (blue(blue accumulativeaccumulative curve).curve).

FigureFigure 9.9. Impact energy of welded specimens:specimens: crack initiation, crack propagation, andand overalloverall impactimpact energyenergy ((aa);); force–timeforce–time curvecurve ofof basebase metalmetal ( (bb);); force–time force–time curve curve of of specimen specimen 3Ti2Si 3Ti2Si ( c(c).).

AlthoughAlthough impact impact energies energies (crack (crack initiation, initiation, propagation, propagation, and overalland overall impact impact energies) energies) were slightly were lowerslightly in lower the welded in the specimens welded specimens compared compared to the base to metal, the base bend metal, testing bend did testing not result did in not any result cracks, in

Materials 2020, 13, x FOR PEER REVIEW 9 of 11 Materials 2020, 13, x FOR PEER REVIEW 9 of 11 Materials 2020, 13, 4513 9 of 11 any cracks, on either side of any specimen tested (Figure 10). This is in accordance with the relatively any cracks, on either side of any specimen tested (Figure 10). This is in accordance with the relatively high values in the reduction of area (Z) obtained in tensile testing (Figure 6). The deformed area of high values in the reduction of area (Z) obtained in tensile testing (Figure 6). The deformed area of theon either weld sidemetal, of anydespite specimen the fact tested that (Figure the surface 10). This was is inground accordance prior with to testing, the relatively showed high signs values of the weld metal, despite the fact that the surface was ground prior to testing, showed signs of columnarin the reduction morphology, of area (Z)which obtained is in accordance in tensile testing with the (Figure macro6). images The deformed (Figure 2). area The of values the weld obtained metal, columnar morphology, which is in accordance with the macro images (Figure 2). The values obtained indespite all specimens, the fact that with the crack surface propagation was ground energies prior tohigher testing, than showed crack initiation signs of columnar energies, morphology, proved that in all specimens, with crack propagation energies higher than crack initiation energies, proved that whichthe suitability is in accordance of welds with to pressure the macro vessel images application (Figure2). is The maintained. values obtained Namely, in a all higher specimens, crack the suitability of welds to pressure vessel application is maintained. Namely, a higher crack withpropagation crack propagation energy ensures energies a minimized higher than sudden crack initiationfailure of energies,the weld proved in the thatcase theof overpressure suitability of propagation energy ensures a minimized sudden failure of the weld in the case of overpressure weldsoccurrence. to pressure vessel application is maintained. Namely, a higher crack propagation energy ensures aoccurrence. minimized sudden failure of the weld in the case of overpressure occurrence.

Figure 10. BendBend testing testing of of the 1Ti4Si specimen; weld root (a) and weld face (b). Figure 10. Bend testing of the 1Ti4Si specimen; weld root (a) and weld face (b). The microhardness results results are are shown shown in in Figure Figure 11. 11 .In In the the majority majority of of specimens, specimens, the the hardness hardness of The microhardness results are shown in Figure 11. In the majority of specimens, the hardness of theof the weld weld metal metal was was higher higher compared compared to the to hardness the hardness of fusion of fusion zones, zones, probably probably due to a due higher to a content higher the weld metal was higher compared to the hardness of fusion zones, probably due to a higher content ofcontent δ‐ferrite of δ and-ferrite a lower and a distance lower distance between between δ‐ferriteδ-ferrite phases: phases: grain size grain in size base in metal base metal 20–100 20–100 μm andµm of δ‐ferrite and a lower distance between δ‐ferrite phases: grain size in base metal 20–100 μm and interdendriticand interdendritic distance distance in weld in metal weld 10–40 metal μ 10–40m. Thisµ m.is depicted This is depictedin Figure in4a,b. Figure Although4a,b. Althoughthe differences the interdendritic distance in weld metal 10–40 μm. This is depicted in Figure 4a,b. Although the differences werediﬀerences small, were the microhardness small, the microhardness of specimens of specimens2Ti3Si and 2Ti3Si 1Ti4Si and were 1Ti4Si the lowest were the (Figure lowest 11c), (Figure which 11 c),is were small, the microhardness of specimens 2Ti3Si and 1Ti4Si were the lowest (Figure 11c), which is whichinteresting, is interesting, since the since tensile the tensileproperties properties and impact and impact energies energies of these of these specimens specimens were were the the highest. highest. interesting, since the tensile properties and impact energies of these specimens were the highest. Apparently, there was an optimal δ‐δ-ferriteferrite content, content, which which provided provided the the highest highest mechanical mechanical properties. properties. Apparently, there was an optimal δ‐ferrite content, which provided the highest mechanical properties.

Figure 11. AverageAverage microhardness microhardness results results for weld face horizontal measurements ( a); weld root Figure 11. Average microhardness results for weld face horizontal measurements (a); weld root horizontal measurements (b); weld metal vertical measurements ((cc).). horizontal measurements (b); weld metal vertical measurements (c).

Materials 2020, 13, 4513 10 of 11

4. Conclusions According to the presented results and within the limitations of this study,the following conclusions can be drawn:

Chemical compositions of the weld metals are not dependent on the composition of the flux. • Different SiO to TiO ratios result in obtaining similarly shaped weld metals, with the same • 2 2 columnar structure, but with different widths. The widest weld resulted in the highest absolute amount of δ-ferrite, the widest space for • deformation, resulting in the highest strength and ductility. Tensile properties and impact toughness of weldments were marginally lower compared to the • base metal, due to the replacement of strain hardened polygonal austenite by the mixture of cast dendritic austenite. Tensile strength and reduction of area as well as impact toughness of the optimal specimen and • the majority of other specimens, were within the limits set by the base metal standard limits. Bend testing did not result in the occurrence of any cracks. • Microhardness values in the weld metal in the majority of specimens were higher compared to • the fusion zones. Additionally, specimens that had the lowest average microhardness had the highest tensile and impact toughness.

Author Contributions: Conceptualization, S.B.; Formal analysis, M.D.; Investigation, M.D. and P.J.; Resources, I.Z. and B.P.; Data curation, M.D. and N.K.; Writing—original draft preparation, S.B.; Writing—review and editing, D.K. and B.P. All authors have read and agreed to the published version of the manuscript. Funding: The authors acknowledge the financial support from the state budget by the Slovenian Research Agency Programme No. P2-0270 and projects No. L2-8181 and L2-8183. Acknowledgments: This article is based on work from the project entitled “Modern technologies in production engineering: lectures, science and collaboration with industry”. Conflicts of Interest: The authors declare no conflict of interest.

References

1. Baddoo, N.R. Stainless steel in construction: A review of research, applications, challenges and opportunities. J. Constr. Steel. Res. 2008, 64, 1199–1206. [CrossRef] 2. Garrison, W.M. Ultrahigh-strength steels for aerospace applications. JOM 1990, 42, 20–24. [CrossRef] 3. Wells, M.G.H. Advances in steels for aerospace applications. Key Eng. Mater. 1992, 77–78, 71–80. [CrossRef] 4. Tseng, K.-H.; Hsu, C.-Y. Performance of activated TIG process in austenitic stainless steel welds. J. Mater. Process. Technol. 2011, 211, 503–512. [CrossRef] 5. Klobˇcar, D.; Tušek, J.; Bizjak, M.; Simonˇciˇc,S.; Lešer, V. Active flux tungsten inert gas welding Of austenitic stainless steel AISI 304. Metalurgija 2016, 55, 617–620. 6. Vidyarthy, R.S.; Dwivedi, D.K. Activating flux tungsten inert gas welding for enhanced weld penetration. J. Manuf. Process. 2016, 22, 211–228. [CrossRef] 7. Dramicanin, M.; Balos, S.; Janjatovic, P.; Zabunov, I.; Grabulov, V. Activated flux TIG welding of stainless steel pipes. Chem. Ind. Chem. Eng. Q. 2019, 25, 353–360. [CrossRef] 8. Gurevich, S.M.; Zamkov, V.N.; Kushnirenko, N.A. Improving the penetration of titanium alloys when they are welded by argon tungsten arc process. Avtom. Svarka. 1965, 18, 1–5. 9. Yangchuan, C.; Zhen, L.; Zunyue, H.; Yida, Z. Effect of cerium oxide flux on active flux TIG welding of 800 MPa super steel. J. Mater. Process. Technol. 2016, 230, 80–87. 10. Vora, J.J.; Badheka, J.B. Experimental investigation on mechanism and weld morphology of activated TIG welded bead-on-plate weldments of reduced activation ferritic/martensitic steel using oxide fluxes. J. Manuf. Process. 2015, 20, 224–233. [CrossRef] 11. Tseng, K.-H.; Lin, P.-Y. UNS S31603 stainless steel tungsten inert gas welds made with microparticle and nanoparticle oxides. Materials 2014, 7, 4755–4772. [CrossRef][PubMed] Materials 2020, 13, 4513 11 of 11

12. Balos, S.; Dramicanin, M.; Janjatovic, P.; Zabunov, I.; Klobcar, D.; Busic, M.; Luisa Grilli, M. Metal oxide nanoparticle-based coating as a catalyzer for A-TIG welding: Critical raw material perspective. Metals 2019, 9, 567. [CrossRef] 13. Balos, S.; Dramicanin, M.; Janjatovic, P.; Zabunov, I.; Pilic, B.; Goel, S.; Szutkowska, M. Suppressing the use of critical raw materials in joining of AISI 304 stainless steel using activated tungsten inert gas welding. Metals 2019, 9, 1187. [CrossRef] 14. Takeuchi, Y.; Takagi, R.; Shinoda, T. Effect of bismuth on weld joint penetration in austenitic stainless steel. Weld. Res. Suppl. 1992, 71, 283–290. 15. Mills, K.C.; Keene, B.J.; Brooks, R.F.; Shirali, A. Marangoni effects in welding. Philos. Trans. R. Soc. A Math. Phys. Eng. Sci. 1998, 356, 911–925. [CrossRef] 16. Skvortsov, E.A. Role of electronegative elements in contraction of the arc discharge. Weld Int. 1998, 12, 471–475. [CrossRef] 17. Tanaka, M.; Shimizu, T.; Terasaki, T.; Ushio, M.; Koshiishi, F.; Terasaki, H. Effects of activating flux on arc phenomena in gas tungsten arc welding. Sci. Technol. Weld Join. 2000, 5, 397–402. [CrossRef] 18. Thakur, P.P.; Chapgaon, A.N. A review on effects of GTAW process parameters on weld. IJRASET 2016, 4, 136–140. 19. Cibi, A.J.; Thilagham, K.T. High frequency gas tungsten arc welding process for dressing of weldment. IJAERS 2017, 4, 229–235. 20. Sharma, P.; Dwivedi, D.K. Comparative study of activated flux-GTAW and multipass-GTAW dissimilar P92 steel-304H ASS joints. Mater. Manuf. Process. 2019, 34, 1–10. [CrossRef] 21. Ramkumar, K.D.; Bajpai, A.; Raghuvanshi, S.; Singh, A.; Chandrasekhar, A.; Arivarasu, M.; Arivazhagan, N. Investigations on structure–property relationships of activated flux TIG weldments of super-duplex/austenitic stainless steels. Mater. Sci. Eng. A 2015, 638, 60–68. [CrossRef] 22. Hdhibi, A.; Touileb, K.; Djoudjou, R.; Ouis, A.; Bouzazi, M.L.; Chakhari, J. Effect of single oxide fluxes on morphology and mechanical properties of ATIG on 316 L austenitic stainless steel welds. Eng. Technol. Appl. Sci. Res. 2018, 8, 3064–3072. 23. Huang, H.-Y. Effects of shielding gas composition and activating flux on GTAW weldments. Mater. Des. 2009, 30, 2404–2409. [CrossRef] 24. Sabiha, A.; Nemes, J.A. Internal ductile failure mechanisms in steel cold heading process. J. Mater. Process. Technol. 2009, 209, 4292–4311. [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/).