Introduction

Over the years, 13% Cr steel has beenwidely accepted in the oil country tubulargoods (OCTG) segment because of itsgood corrosion resistance. The use of13% Cr steel is recognized as a standardcountermeasure to corrosion attack initi-ated by carbon dioxide in oil and gaswelds. However, conventional type 12–13% Cr steel has not been utilized somuch in line pipe because of its relatively

bad weldability, which requires preheat-ing prior to welding and postweld heattreatment (PWHT) (Ref. 1).

Recently, supermartensitic stainlesssteels (SMSS) have been developed as anattractive technical alternative to high-strength, low-alloy (HSLA) steels mainlyin applications related to the oil and gasindustry (Refs. 2, 3). Welding these mate-rials plays a crucial role in structural com-ponents, influencing their weldability,toughness, and resistance to sulfide stress

cracking. The SMSS were developedbased on classic martensitic stainless steels(11–14% Cr), reducing C content to en-hance weldability, toughness, and corro-sion resistance, plus adding Ni to promotea free ferrite structure and Mo (Refs. 4, 5),which also improves corrosion resistance(Refs. 6, 7).

For welding these materials, gas metalarc welding (GMAW) using SMSS metal-cored wires has been recognized as a suit-able technological option, and its use hasrecently been improved (Ref. 5). This typeof consumable presents several advan-tages such as low slag generation and highdeposition rate (Ref. 8). Shielding gasesemployed for welding this type of materialusually are inert gas mixes (Ar-He) or Ar-rich mixtures. The type of shielding gascan affect the chemical composition of theweld metal, principally O, N, and C con-tents (Ref. 9).

Usually, PWHT is necessary to adjustthe properties of weldments in SMSS. Inthese cases, the heat treatment times typ-ically vary between 5 and 30 min. In thisrespect, it has been reported that theaforementioned short PWHT provides animprovement in sulfide stress cracking(SSC) resistance (Ref. 7). In certain gradesof SMSS, these treatments provide a de-crease of hardness and residual stressalong with increased toughness, whilegrades with higher Mo content are lesssensitive to PWHT (Refs. 7, 10).

Different PWHTs lead to microstruc-tural modifications producing differentcombinations of phases present in SMSSweld deposits (tempered and untemperedmartensite, austenite, carbides, etc.) witheach microstructural pattern affectingtoughness in a specific way.

The objective of this work was to sys-tematically study the effects of differentshielding gas mixtures and PWHT (650°C ×15 min) on the all-weld-metal properties ob-

Supermartensitic Stainless Steel Deposits:Effects of Shielding Gas and

Postweld Heat Treatment

Detailed are the results as CO2 content in the shielding gas increased,plus the postweld heat treatment utilized in this work

BY S. ZAPPA, H. G. SVOBODA, AND E. S. SURIAN

KEYWORDS

Supermartensitic StainlessSteel

Shielding GasPostweld Heat Treatment

(PWHT)Mechanical Properties

S. ZAPPA (zappasebastian@hotmail.com) iswith the Research Secretariat, Faculty of Engi-neering, University of Lomas de Zamora, BuenosAires, Argentina. H. G. SVOBODA is with theMaterials and Structures Laboratory, Faculty ofEngineering, Intecin, University of Buenos Aires,Conicet, Buenos Aires, Argentina. E. S. SURIANis with the Research Secretariat, Faculty of Engi-neering, University of Lomas de Zamora, BuenosAires and Deytema, Regional Faculty of SanNicolás, National Technological University, SanNicolás, Argentina.

ABSTRACT

Welding supermartensitic stainless steel plays a crucial role in structural com-ponents, influencing their toughness and resistance to sulfide stress cracking. Post-weld heat treatment (PWHT) adjusts the final properties of the weldments, bearingon microstructural evolution. The objective of this work was to study the effects ofdifferent shielding gas mixtures and PWHT on supermartensitic stainless steel all-weld-metal properties. Three all-weld-metal test coupons were prepared accordingto standard ANSI/AWS A5.22:95, Specification for Stainless Steel Electrodes for FluxCored Arc Welding and Stainless Steel Flux Cored Rods for Gas Tungsten Arc Weld-ing, using a 1.2-mm-diameter tubular, metal-cored wire under Ar-5% He, Ar-2%CO2, and Ar-18% CO2 gas shielding mixtures in the flat position with a nominalheat input of 1 kJ mm–1. The PWHT used was 650°C for 15 min. All-weld metalchemical composition analysis, metallurgical characterization, hardness and ten-sile property measurements, and Charpy V-notch tests were carried out. It wasfound that as CO2 increased in the shielding gas C, O, and N contents increased aswell as mechanical properties varied; hardness and ultimate tensile strength in-creased, and toughness decreased. The PWHT improved toughness. Technologicalproperty of the consumable was also studied.

297-sWELDING JOURNAL

WE

LD

ING

RE

SE

AR

CH

Zappa et al Supplement October 2013_Layout 1 9/13/13 4:32 PM Page 297

tained with a SMSS metal-cored wire usedwith the semiautomatic welding process.

Experimental Procedure

The oxidation potential (OP) of ashielding gas is related to O2 and CO2 con-tents of the mentioned gas, according tothe following equation (Ref. 11):

OP = O2 + 0.5 CO2

Table 1 shows the chemical composi-tion of shielding gases used and their re-spective OP, the samples identification,and welding parameters utilized. Threeall-weld-metal coupons were welded ac-cording to the standard ANSI/AWSA5.22-95, Specification for Stainless SteelElectrodes for Flux Cored Arc Welding and

Stainless Steel Flux Cored Rods for GasTungsten Arc Welding, using a tubularmetal-core wire of 1.2 mm diameter withthree different shielding gases. In allcases, the welding position was flat, andthe preheating and interpass tempera-tures were 100°C. Gas flow was 18 L/minand stickout 20 mm. After welding, thethree coupons were X-rayed followingANSI B31.3-96, Chemical Plant and Pe-troleum Refinery Piping.

A PWHT at 650°C for 15 min was ap-plied to the different samples in an in-duction heating furnace. This PWHTcondition was chosen according to therecommendations of the consumablemanufacturer and previous studies (Refs.12, 13). From each welded coupon, trans-versal sections for macro and microstruc-ture studies, chemical compositiondetermination, and hardness measure-ments were extracted. To study the oper-ability of the consumable, width andpenetration of the last bead were meas-ured as well as the spatter level as a func-tion of the shielding gas used.

On the cross sections of each coupon,chemical composition by optical emissionspectrometry (OES) was determined, ex-cept the C, O, N, and S contents, whichwere analyzed using a LecoTM. The inclu-sionary levels and critical temperatureswere determined. The microstructuralcharacterization was performed by lightmicroscopy (LM), scanning electron mi-croscopy (SEM), and X-ray diffraction(XRD). The delta ferrite fraction presentin each deposit was measured throughquantitative metallography according to

ASTM E562-99, Standard Test Method forDetermining Volume Fraction by SystematicManual Point Count, and the retainedaustenite contents were determined usinga peak comparison method from the XRDpatterns (Ref. 14).

In addition, Vickers hardness HV1 wasmeasured. From each coupon, 1 longitu-dinal tensile specimen was machined ac-cording to (Ref. 15), and 15 CharpyV-notch (CVN) impact specimens weremachined according to ASTM E23-05,Standard Test Methods for Notched Bar Im-pact Testing of Metallic Materials. Tensileand impact CVN samples were tested inthe as-welded (AW) and PWHT condi-tions. Tensile property was measured atroom temperature, and CVN impact testswere performed at 20°, 0°, –20°, –40°,and –60°C.

Results and Discussion

According to ANSI B31.3-96, allcoupons were approved as they presentedvery low defect levels.

Figure 1 shows the surface appearanceof samples HA, CA, and AA welded undereach shielding gas used. It can be seen thatthere was a clear increase in the amountof spatter as the CO2 content in the shield-ing gas increased. This could be associatedwith an increased partial pressure of CO2in the arc atmosphere.

For the welding parameters used, themetal transfer occurred in a mixed glob-ular/spray mode. The change from atransfer mode to another one is charac-terized by a transition current, which de-pends on the shielding gas and increaseswith its CO2 content. In the gas atmos-phere, the presence of this gas limits theelectromagnetic forces that control spraytransfer mode (Ref. 16). In this sense,when using a gas mixture with higheramounts of CO2, the globular to spraytransfer mode transition current in-creases, then increasing the level of spat-ter associated with the globular transfermode.

Figure 2 shows the images of the lastbead for HA, CA, and AA specimens. Inthese figures, it can be seen that as CO2content increased in the shielding gas, thegeometry of the bead changed, decreasingthe ratio width/penetration as is shown inTable 2.

OCTOBER 2013, VOL. 92298-s

WE

LD

ING

RE

SE

AR

CH

Fig. 1 — Surface appearance of welded samples for the three gases.

Fig. 2 — Last weld bead for the different shielding gases.

Fig. 3 — Inclusions volume fraction vs. oxidation potential of the shielding gas.

Table 1 — Chemical Composition of Shielding Gases Used, Samples Identification and Welding Parameters

As Heat Shielding CO2 Ar He OP Tension Current Welding Rate Heat InputWelded Treated Gas (%) (%) (%) (V) (A) (mm/s) (kJ/mm)

HA HP Ar + 5% He 0 70 5 0 25 226 5 1.1CA CP Ar + 2% CO2 2 98 0 1 25 230 6 1.0AA AP Ar + 18% CO2 18 82 0 9 26 232 6 1.0

Zappa et al Supplement October 2013_Layout 1 9/16/13 9:19 AM Page 298

Table 3 shows the results of the chemicalcomposition of all-weld-metal coupons.Carbon contents, fluctuating between 140and 180 ppm, were higher than expected forthis consumable. The value reported by themanufacturer was less than 100 ppm (Refs.13, 17).

In general, the measured values of therest of the analyzed elements were belowthe nominal values with the exception ofSi and Mo, according to the product datasheet (Ref. 13). In the SMSS, to achievegood toughness and appropriate hardnessvalues, it is necessary to have low C, N, O,and S contents (Refs. 18, 19). The C con-tent controls the martensite hardness andtoughness; it is the more influential ele-ment on MS temperature. The precipi-tated carbides influence corrosionresistance and hydrogen damage (Ref. 20),among others. Also, the presence of N andMo has a great influence on the sequenceof carbides and carbides/nitrides precipi-tation (Ref. 21).

Regarding the O content present in theweld metal, in all cases it exceeded 300 ppm,the limit beyond which, according to the lit-erature (Ref. 17), the absorbed energy fallsbrusquely. The variation observed in thechemical composition of the all-weld metalcorresponds, among other things, to themetallic elements oxidation process occur-ring in the electric arc. The greater O con-tent in the shielding gas, the higheroxidation potential of the gaseous atmos-phere, and the greater the effect of oxida-tion mentioned above (Ref. 22). The highinterstitial elements contents in the depositsare associated with a high content of O andC in the atmosphere of the arc from the de-composition of CO2 in the shielding gas(Ref. 23). These variations were small butcould have affected certain properties. Inthis sense, Ni, Cu, and Mn are elements thatstabilize the austenite, so that higher con-tents of these elements could result in an in-crease of retained austenite in themicrostructure (Refs. 16, 24).

Also, Mo and Cr could affect the deltaferrite fraction as they both stabilize thisphase. Toughness and, to a lesser extent,tensile property and corrosion resistance,could be affected by these small varia-tions in chemical composition, mainly inthe coupons welded under a high OP gasmixture.

The increase of CO2 content in theshielding gas produced a higher O contentin the deposits. One way to demonstratethe increase of this last element is throughthe determination of inclusionary levels.Some authors (Ref. 25) reported that thesizes of the inclusions are the order of mi-crons. Table 4 shows that quantity, averagediameter, and volume fraction of inclu-sions in the AW sample increased with theCO2 content in the shielding gas.

Figure 3 presents the effect of theshielding gas on the inclusion volume frac-tion (IVF) of the deposits, showing thetendency mentioned. These differences inthe inclusionary level may affect, amongother properties, the toughness and cor-rosion resistance of the deposits (Refs. 17,18, 25, 26).

In all cases, in the AW samples, the mi-crostructure consisted of a martensitic ma-trix with small amounts of delta ferrite and

retained austenite. The literature (Refs. 6,7, 27) reports for this type of steel that re-tained austenite contents in the AW con-dition may vary between 2 and 30%, andthat this phase cannot be identifiedthrough the microscopy techniques usedin this work (Refs. 12, 17, 26). Samplessubmitted to PWHT presented a darken-ing of the martensitic structure associatedwith the carbide and/or carbonitride pre-cipitates mainly in the grain boundary.

299-sWELDING JOURNAL

WE

LD

ING

RE

SE

AR

CH

Fig. 4 — LM and SEM in the coupon AP.

Table 2 — Relationships Width/Penetrationfor the HA, CA, and AA Samples

Sample RelationshipWidth/Penetration

HA 2.7CA 2.2AA 1.9

Table 3 — Chemical Composition

Sample C Mn Si S Cr Ni Mo Cu V O N(ppm) (%) (%) (%) (%) (%) (%) (%) (%) (ppm) (ppm)

HA 140 1.72 0.44 0.015 11.8 6.2 2.69 0.48 0.09 380 50CA 150 1.75 0.45 0.014 11.7 6.2 2.66 0.48 0.09 440 60AA 180 1.57 0.42 0.016 11.7 6.1 2.47 0.41 0.09 710 140

Specif. <100 1.8 0.4 NI 12.5 6.7 2.5 0.5 NI NI <100

NI: not informed

Zappa et al Supplement October 2013_Layout 1 9/13/13 4:36 PM Page 299

Figure 4 shows the microstructure ofthe coupon AP obtained by LM and SEM,respectively. In this figure the followingidentification was used: M, martensite; F,ferrite; and Pr, precipitates. Similar mi-crostructures were found in the rest of thecoupons.

In the literature (Ref. 17), there are re-ported several ferrite morphologies foundin SMSS deposits. Two types of ferritecould be identified based on their locationand morphology. Most common was fer-rite with morphology very similar to thatof ferrite found in duplex stainless steelweld metals. The presence of this ferrite isa consequence of incomplete ferrite-to-austenite transformation in weld metalssolidifying as ferrite and was most com-mon for more highly alloyed weld metals(Ref. 17). Another ferrite morphology,similar to that seen in austenitic stainlesssteel weld metals, was found in weld met-als higher in Ni, solidifying as a mixture offerrite and austenite (Ref. 17). This ferritewas located in the last solidifying inter-dendritic regions.

Figure 5 shows the different ferritemorphology for the HA coupon. Thesedifferent morphologies of ferrite in the mi-crostructure show different solidificationmodes generated simultaneously in thedeposits.

The critical temperatures AC1, AC3,and MS were determined by dilatometrywith a heating rate of 1°C/min in the HAand AA coupons. Table 5 shows the resultsobtained. The coupon welded with high

CO2 content in the shielding gas (AA) hada higher austenite transformation startingtemperature regarding the coupon weldedunder Ar + 5% He (HA). Marshall andFarrar (Ref. 28) indicated that the actualAC1 temperature for a low-grade stainlesssteel is around 650°C while high-grade al-loys are around 630°C.

The use of equations for the selectionof PWHT temperature should be carefullyconsidered and the results be verified ex-perimentally to prevent undesired hard-ening of the structure due to the fact thatpartial austenitizing can generate freshmartensite (Ref. 5). The PWHT is usuallyrecommended in SMSS deposits for im-proving both toughness and ductility; ten-sile strength is reduced by only 10 to 20%(Ref. 29). Ni has a strong effect on de-creasing AC1 and Mo the contrary effect.In this sense, the small variations of cer-tain elements observed in the chemicalcomposition of these samples could haveled to changes in critical temperatures.

The different types of ferrite werequantified for the three coupons in AWconditions through 20 measurements on agrid of 660 points. Table 6 shows these re-sults. There were no significant variationsin the measured ferrite contents, indicat-ing no shielding gas important influenceon these values. Table 7 shows the retainedaustenite content.

From a microstructural point of view,the objective of PWHT is to temper themartensite and increase the stable re-tained austenite at room temperature re-

sulting in softening and improving bothtoughness and corrosion resistance of thedeposit (Refs. 2, 3, 19, 30, 31). The litera-ture reports (Refs. 5, 32) that a PWHT at650°C for 5 min is the heat treatment com-monly used in manufacturing SMSS pipewhen using duplex and superduplex stain-less steel consumables; longer heat treat-ment can cause sigma phase to form inthese weld metals. During the PWHT, inthese steels, the tempering of martensiteis generally followed by a softening asso-ciated with incoherent carbide precipita-tion, achieving the maximum softeningwith the M23C6 carbide precipitation atabout 500°C (Ref. 32).

In alloys without Ni, the PWHT is per-formed at about 700°C to obtain a high re-action rate and maximum softening (Ref.32). The presence of Ni reduces the criti-cal transformation temperature AC1. Thistemperature, for a given steel, depends onthe chemical composition, heating rate,and in alloys with high Ni contents can beas low as 500°–550°C (Refs. 32, 33). In thiswork, heat treatment was carried out at650°C for 15 min, which seemed to be ad-equate for matching filler metals but notappropriate for duplex or superduplexones. This temperature/time parameterdid not generate changes that could havebeen observed by LM and SEM. Only dif-ferences in the austenite content in the mi-crostructure could be established by X-raydiffraction.

As mentioned above, in alloys withhigh contents of Ni, the AC1 temperaturemay be around 550°C. At this tempera-ture, the kinetics of carbide formation isvery slow and, it is normal that austeniteprecipitates (Ref. 32). In this way, througha diffusion mechanism, the austeniteformed during PWHT will have a differ-ent chemical composition from the austen-ite retained during the welding process(Refs. 32, 34). In this instance, the marten-sitic matrix and retained austenite pro-duced during welding will have identicalchemical composition while the newaustenite formed during PWHT will be

OCTOBER 2013, VOL. 92300-s

WE

LD

ING

RE

SE

AR

CH

Fig. 5 — Different morphologies of ferrite in the HA coupon. Fig. 6 — Relationship between the austenite con-tent and oxidation potential of the shielding gas.

Table 4 — Inclusionary Analysis

Sample IT Area ID IAD IVF OP O Content(μm2) (incl/ μm2) (μm) (incl/μm3)

HA 5 1695 0.003 1.14 0.002 0 380CA 14 1695 0.008 1.26 0.007 1 440AA 32 1695 0.019 1.40 0.019 9 710

IT: inclusions total; ID: inclusions density; IAD: inclusions average diameter; IVF: inclusions volume fraction; OP (oxygen potential) = O2 + 0.5 CO2

Zappa et al Supplement October 2013_Layout 1 9/13/13 4:32 PM Page 300

richer in N, C, and Ni. This degree of en-richment determines the stability ofaustenite formed during the heat treatment.

If PWHT is performed at temperaturesslightly above AC1, the enriched austeniteis stable at room temperature (Refs. 32,34). If PWHT is carried out at tempera-tures well above AC1, the austenite formedduring PWHT will be transformed to“fresh” martensite during cooling (Ref.32). If PWHT is carried out below AC1temperature, the austenite content will de-crease with respect to that of the AW sam-ple because part of it will transform to“fresh” martensite after cooling to roomtemperature.

Figure 6 shows the relationship be-tween austenite content and OP of theshielding gas for both AW and PWHTconditions. The contents of retainedaustenite increased with increasing CO2 inthe shielding gas in AW samples. This in-crease could be related to higher C and Ncontent generated using higher amountsof CO2 in the shielding gas as both ele-ments are strong promoters of austenite.

Also, it has been reported (Ref. 35)

that the austenite contents are stronglycontrolled by the MS temperature. Re-tained austenite levels, at room tempera-ture, are higher when MS decreases (Ref.35). As seen previously, increasing the OPin the shielding gas slightly decreased thetransformation temperature of austeniteto martensite during cooling. This slightdecrease in the MS temperature generateda higher retained austenite content atroom temperature, in accord with whatwas reported elsewhere (Ref. 35).

When applying PWHT, retainedaustenite content increased for low OP,while for high OP retained austenite val-ues decreased. The higher OP, the highercritical temperature measured. The HAcoupon (AC1 = 590°C) showed low

austenite values in the AW condition andapplication of PWHT, at a temperatureslightly above AC1, then generated higheramounts of retained austenite. On theother hand, the AA coupon (AC1 =660°C) had a higher content of austenitein the AW condition and applying PWHT,which was slightly below the critical tem-perature, generated lower retainedaustenite content regarding the AW sam-ple. In conclusion, applying a PWHT at650°C for 15 min to this type of material,austenite content can decrease or increasedepending on the PWHT temperature lo-cation regarding the AC1 critical tempera-ture of the material.

Table 8 presents the values of hardness,tension property, and toughness deter-

301-sWELDING JOURNAL

WE

LD

ING

RE

SE

AR

CH

Fig. 7 — Influence of OP on hardness, ultimate tensile, and yield strengths.

Fig. 9 — The CVN at 20°C vs. austenite content.

Fig. 8 — Effect of O content on the CVN absorbed energy in AW andPWHT conditions.

Fig. 10 — Hardness vs. CVN in AW and PWHT conditions at 20°C.

Table 5 — Critical Temperatures for Samples H and A

Sample Heating Rate AC1 AC3 MS(°C/min) (°C) (°C) (°C)

HA 1 590 670 125

AA 1 660 710 115

Zappa et al Supplement October 2013_Layout 1 9/13/13 4:32 PM Page 301

mined on the all-weld-metal coupon forthe different conditions studied. Figure 7shows the influence of OP on hardness, ul-timate tensile strength, and yield strengthfor the AW and PWHT samples. It can beseen that as the gas OP increased, hard-ness increased; this fact can be related tothe augmentation of C and N in the de-posits as the OP increased (Ref. 18).

It is known (Refs. 36–38) that C contentcontrols the martensite hardness. Thehigher C content the higher measured hard-ness. The literature also reports (Refs. 39,40) that N is a promoter of carbonitridesthat generates a hardening of the structurein these materials. Then, higher both C andN contents will produce higher hardness.The tensile and yield strengths showed thesame behavior that hardness did. The ap-plication of PWHT resulted in lower valuesof hardness, tensile strength, and yieldstrength in all cases. As mentioned above,the PWHT employed generated martensitetempering and in some cases higher austen-ite content at room temperature; these twofacts produced lower values of mechanicalproperties.

It was not possible to establish a rela-tionship between austenite contents and thetensile properties and hardness. The ab-sorbed energy values obtained were in theorder of these reported in the literature(Refs. 17, 37, 41, 42). In the AW samples,the biggest value was achieved in sampleHA (no CO2 in the shielding gas) with 41 Jat 20°C. The effect of OP on the toughnessof these materials was clearly seen: Thehigher CO2 in the shielding gas, the lowertoughness for both AW and heat-treatedsamples. According to the literature (Ref.17), O content strongly influences thetoughness values of the SMSS.

Figure 8 shows the relationship betweentoughness results at 20°C and the O content

in all welded metal coupon in both AW andPWHT samples. This figure shows clearlythat the toughness decreased when the Ocontent increased for all coupons. The sametrends could be determined for all test tem-perature. In all cases, the absorbed energyvalues achieved after PWHT improved forall temperatures, resulting over 50% higherto those obtained in AW samples. ThePWHT applied, as discussed above, pro-duced martensite tempering and, accordingto the location of PWHT temperature withrespect to AC1 temperature, generatedlower or higher austenite content in the mi-crostructure (Ref. 43) regarding the mi-crostructure in AW condition. Themicrostructure of PWHT samples consistedof tempered martensite, delta ferrite, andaustenite fractions.

It is known (Ref. 43) that higher stableaustenite content at room temperature in-creases the absorbed energy in the CVN im-pact test. It was not possible to establish adirect relationship between austenite con-tent and toughness. However, according toFig. 9, it can be seen that as the austenitecontent increased, toughness tended to in-crease. A low R2 could be indicating thataustenite content is not the only factor in-fluencing toughness.

Figure 10 shows the relationship be-tween the hardness and toughness in AWand PWHT conditions at 20°C.

Conclusions

As CO2 content in the shielding gas in-creased, the following occurred:

1) Higher spatter and a lower width-penetration relationship in the last beadwere obtained.

2) In the all-weld metal, C, O, and Ncontents increased. Mn, Si, Cr, Ni, and Mocontents decreased.

3) Inclusionary levels increased.4) The microstructure, in all cases, was

constituted by a martensitic matrix withdelta ferrite and retained austenite. Theaustenite content increased.

5) Hardness and tensile propertyslightly augmented.

6) Toughness decreased.Regarding the PWHT used in this

work, the following took place:1) The microstructure did not show dif-

ferences with the PWHT.2) Hardness and tensile properties de-

creased slightly in comparison to the AWvalues.

3) Toughness was improved in all sam-ples. In most cases, the absorbed energyincreased in more than 50% compared tothe AW samples.

4) The best toughness was achieved forthose samples welded under inert gas pro-tection and submitted to heat treatment.

In those cases where toughness is not arequirement, it is possible to use CO2 gasshielding, especially lower cost Ar/20 CO2.

Acknowledgments

The authors wish to express their grat-itude to ESAB-Sweden for the consum-able donation and Leco chemical analysis;Conarco-ESAB Argentina for performingchemical analysis; Air Liquide Argentinafor donating welding gases; the LatinAmerican Welding Foundation, Ar-gentina, for welding and mechanical test-ing facilities; the Scanning ElectronMicroscopy Laboratory of Inti-Mechanics,Argentina, for SEM analysis facilities; andAPUEMFI, Argentina, along with AN-PCyT, Argentina, for financial support.

References

1. Kondo, K., et al. 1999. Alloy design ofsuper 13Cr martensitic stainless steel (develop-ment of super 13 Cr martensitic SS for linepipe-1). Supermartensitic Stainless Steel ’99.Brussels, Belgium.

2. Marshall, A. W., and Farrar, J. C. M. 1998.Welding of ferritic and martensitic 13% Crsteels. Preliminary report (draft 2). IIW Doc IX-H-422-98: 1–18.

3. Farrar, J. C., and Marshall, A. W. 1998.Supermartensitic stainless steel — Overviewand weldability. IIW Doc No. IX-H 423-98: 1–3.

4. Bilmes, P. D. 2000. Role of austenite onthe mechanical properties of soft martensiticstainless steels weld metals. PhD thesis. Na-tional University of La Plata, Faculty of Engi-neering. La Plata, Argentina: 147.

5. Lippold, J. C., and Kotecki, D. J. 2005.

OCTOBER 2013, VOL. 92302-s

WE

LD

ING

RE

SE

AR

CH

Table 6 — Ferrite Content

Sample HA CA AA

Ferrite 8.5 10.8 9.3Measured (%)

Table 7 — Retained Austenite

Sample HA CA AA HP CP AP

Retained Austenite 7 9 20 28 30 15Measured (%)

Table 8 — Mechanical Properties

Sample Hardness UTS YS E AR CVN (J)/(°C)(HV1) (MPa) (MPa) (%) (%) 20 0 –20 –40 –60

HA 339 1134 920 17 45 41 33 31 29 28CA 345 1174 955 15 40 37 36 34 33 32AA 357 1189 980 12 33 26 24 21 19 18HP 318 1052 875 19 53 63 51 45 43 43CP 331 1142 925 15 40 59 48 45 43 38AP 337 1159 950 13 37 40 35 34 32 32

UTS: ultimate tensile strength; YS: yield strength; E: elongation to fracture; AR: area reduction

Zappa et al Supplement October 2013_Layout 1 9/13/13 4:32 PM Page 302

Welding Metallurgy and Weldability of Stain-less steels. John Wiley & Sons, Inc.: 1–356.

6. Karlsson, L., et al. 1999. Development ofmatching composition supermartensitic stain-less steel welding consumables. Svetsaren No. 3:3–7.

7. Kvaale, P. E., and Olsen, S. 1999. Experi-ence with supermartensitic stainless steels inflowline applications. Stainless Steel World. TheHague, The Netherlands.

8. Lyttle, K., et al. 1996. Metal cored wires:Where do they fit in your future? Welding Jour-nal 74(10): 35–38.

9. Zappa, S., et al. 2006. Effect of shieldinggas on supermartensitic stainless steel all weldmetal properties. CONAMET/SAM. Santiagode Chile, Chile.

10. Karlsson, L., et al. 1999. Matching com-position supermartensitic stainless steel weld-ing consumables. Supermartensitic Stainless Steel’99. Brussels, Belgium.

11. Stembacka, N. 1990. The influence ofshielding gas on cored wires. Welding Journal69(11): 43–45.

12. Zappa, S., et al. 2007. Effect of post-weldheat treatment on supermartensitic stainlesssteel properties deposited with e tubular metal-cored wire. Soldagem & Inspecao 12(2):115–123.

13. Technical Sheet OK Tubrod, 15-55.2004. ESAB.

14. Cullity, B. D., and Stock, S. R. 2001. El-ements of X ray Diffraction. Prentice-Hall, 3rdEdition, New Jersey.

15. Schnadt, H. M., and Leinhard, E. W.1963. Experimental investigation of the sharp-notch behavior of 60 steels at different temper-ature and strain rates. IIW-IIS Doc. 196-343-63.

16. Lancaster, J. F. 1987. Metallurgy of Weld-ing. Allen & Unwin, 4th ed.

17. Karlsson, L., et al. 1999. Matching com-position supermartensitic stainless steel weld-ing consumables. Supermartensitic StainlessSteels ’99. Brussels, Belgium.

18. Karlsson, L., et al. 2001. Mechanicalproperties and aging response of supermarten-sitic weld metals. Stainless Steel World. TheHague, The Netherlands.

19. Karlsson, L., et al. 2002. Welding of su-permartensitic stainless steels. Recent develop-ments and application experience. Svetsaren,No. 2: 15–22.

20. Zappa, S., et al. 2008. Properties, mi-crostructure and corrosion resistance of super-martensitic stainless steel all weld metal.CONAMET/SAM. Santiago de Chile, Chile.

21. Pickering, F. B. 1978. Physical Metallurgyand the Design of Steels. Applied Science Pub-lishers. London.

22. Vaidya, V. 2002. Shielding gas mixturesfor semiautomatic welds. Welding Journal 81(9):43–48.

23. Stembacka, N., and Persson, K. A. 1989.Shielding gases for gas metal arc welding. Weld-ing Journal 68(11): 41–47.

24. Lippold, J., and Alexandrov, B. 2004.Phase transformation during welding and post-weld heat treatment of a 12Cr-6.5Ni-2.5Mo su-permartensitic stainless steel. Stainless SteelWorld. Houston, Tex.

25. Kimura, M., et al. 2003. Mechanicalproperties and corrosion resistance of matchingGMA welded joint for martensitic stainlesssteel line pipe. Corrosion 2003, NACE, Hous-ton. Paper No. 03100: 1–10.

26. Karlsson, L., et al. 1999. Efficient weld-

ing of supermartensitic stainless steels withmatching composition consumables. StainlessSteel World. The Hague, The Netherlands.

27. Zappa, S., et al. 2007. Effect of postweldheat treatment on supermartensitic stainlesssteel all weld metal toughness.SAM/CONAMET. San Nicolas, Argentina.

28. Marshall, A. W., and Farrar, J. C. M.2001. Welding of ferritic and martensitic11–14% Cr steels. Welding in the World 45 (5/6):32–55.

29. Surian, E., et al. 2010. Welding proce-dure effect on SMSS all weld-metal properties.FABTECH. AWS ed. Atlanta, Ga.

30. Bilmes, P. D., et al. 2006. Microstructureand pitting corrosion of 13CrNiMo weld met-als. Corrosion Science 48: 3261–3270.

31. Bilmes, P. D., Llorente, C. L., and Solari,M. 1998. Effect of postweld heat treatment onthe microstructure and mechanical behavior of13Cr-4NiMoL and 13Cr-6NiMoL weld metals.The 18th ASM Heat Treating Society Conferenceand Exposition. Chicago, Ill.

32. Gooch, T. G., Wolling, P., and Haynes,A. G. 1999. Welding metallurgy of low carbon13% chromium martensitic steels. Super-martensitic Stainless Steel. Brussels, Belgium.

33. Arwidson, S., Baggastrom, G., and Hell-ner, L. 1968. New steels for the power industry.Proc. Conf. Stainless Steels. Birmingham: ISIpublication, 117.

34. Folkhard, H. 1988. Welding metallurgyof stainless steels. Springer-Verlag Wien: 1–280.New York.

35. Wu, W., et al. 2000. The relationship be-tween alloying elements and retained austenitein martensitic stainless steel welds. Scripta Ma-

teriala 42: 1071–1076.36. Ramirez, J. E. 2007. Weldability evalua-

tion of supermartensitic stainless steels. Weld-ing Journal 86(5): 125-s to 134-s.

37. Perteneder, E., and Tosch, J. 1999. Im-portant metallurgical aspects in the manualwelding of soft martensitic 13% chrome steels.Brussels, Belgium.

38. Deleu, E., Dhooge, A., and Dufrane, J.J. 1999. Weldability and hot deformability ofdifferent supermartensitic stainless steel gradesby weld simulation testing. SupermartensiticStainless Steels ’99. Brussels, Belgium.

39. Bilmes, P. D., Solari, M., and Llorente,C. L. 2001. Characteristics and effect of austen-ite resulting from tempering of 13Cr-NiMomartensitic steel weld metals. Materials Charac-terization 46: 285–296.

40. Bilmes, P. D., Llorente, C. L., and Solari,M. 2000. Role of the retained austenite on themechanical properties of 13Cr-4NiMo weldmetals. The 20th ASM Heat Treating Society.America’s Center, St. Louis, Mo.

41. Heuser, H., et al. 1999. GMAW andSAW matching filler metals for supermarten-sitic stainless steels. Supermartensitic StainlessSteel ’99. Brussels, Belgium.

42. Bonnel, J .M., and Butter, D. J. 1999. Su-permartensitic 13% Cr stainless steel weldingconsumables for girth welding line pipe. Super-martensitic Stainless Steels ’99. Brussels, Bel-gium.

43. Zappa, S., et al. 2008. Effect of PWHTon supermartensitic stainless steel all weldmetal toughness. Trends in Welding ResearchProceedings of the 8th International Conference.Pine Mountain, Ga.

303-sWELDING JOURNAL

WE

LD

ING

RE

SE

AR

CH

Win Potential Clientsby Showing Them Videos on

How to Save Money and TimeBy Using Your Products!

For more information visit our website at http://videos.aws.org or please contact:

Rob Saltzsteinsalty@aws.org(800) 443-9353, ext. 243

Sandra Jorgensensjorgensen@aws.org(800) 443-9353, ext. 254

Lea Panecalea@aws.org(800) 443-9353, ext. 220

Zappa et al Supplement October 2013_Layout 1 9/13/13 4:32 PM Page 303