278-s | OCTOBER 2000

RE

SE

AR

CH

/DE

VE

LO

PM

EN

T/R

ES

EA

RC

H/D

EV

EL

OP

ME

NT

/RE

SE

AR

CH

/DE

VE

LO

PM

EN

T/R

ES

EA

RC

H/D

EV

EL

OP

ME

NT

ABSTRACT. Microstructural evolution re-sulting from gas metal arc welding ofHSLA-100 steel plate with an experi-mental ultra-low-carbon weld consum-able, designated CTC-03, was investi-gated by transmission electronmicroscopy, optical microscopy and mi-crohardness analyses. A color micro-hardness map was generated from morethan 1600 diamond indentation mea-surements taken across the transversecross section of the weldment, to make adirect correlation between the micro-hardness variations and the correspond-ing microstructures. Eight characteristicareas of the CTC-03/HSLA-100 weldwere selected to study the correspondingmicrostructure by TEM and optical mi-croscopy, and the microstructures werethen correlated with the microhardnessmap. The microstructure throughout thefusion zone (FZ) consists of a majority oflath ferrite with varying amounts (de-pending on the location) of untemperedfine-lath martensite, some interlath re-tained austenite, and spherical oxide in-clusions. The softest regions of the fusionzone are in curved white bands locatedat the lower portions of the heat-affectedzones of the weld beads. The mi-crostructure within these white bands ispredominantly lath ferrite. The hardestregions in the fusion zone are located be-tween the bead boundaries and thesewhite bands, and contain significantamounts of fine untempered lath marten-site. The base plate heat-affected zone(HAZ) consists of a mixture of untem-

pered lath martensite and coarse au-totempered plate martensite. The as-received HSLA-100 base metal exhibits aquench-and-tempered microstructurewith a majority of fine-lath martensiteand significant amounts of coarsemartensite. Despite its low heat input, theCTC-03 weldment exhibits hardness val-ues comparable with those of the HSLA-100 base metal and similar to some weld-ments made with other filler metals. Themicrohardness map of this weld wascompared to four other microhardnessmaps developed for welds made withother candidate ULC filler metals and/ordifferent heat inputs. In all five weld-ments studied, the midsection of the HAZof the base metal (i.e., the region locatedabout midway between the weld inter-face and the outer boundary of the HAZ)is the hardest region of the weld, regard-less of base metal, filler metal type orheat input.

Introduction

The family of high yield (HY) strengthsteels, including HY-80, HY-100 and HY-

130, has been extensively used by theU.S. Navy since its development in 1946(Ref. 1). The strength and toughness ofthese steels are derived from the forma-tion of quenched and tempered marten-site. While carbon plays a major role inthe strengthening mechanism of thesesteels through the formation of fine dis-persions of carbides, it can also promotethe formation of deleterious untemperedmartensite microstructures during fastcooling. These deleterious microstruc-tures have an enhanced susceptibility tohydrogen-assisted cracking (Ref. 2).Thus, welding of these steels requiresstrict control of the heat input and pre-heat/interpass temperature to ensure theproduction of acceptable microstruc-tures and mechanical properties. Sincethese constraints on the welding opera-tional envelope are very costly, the U.S.Navy has pursued, since about 1980,high-strength low-alloy (HSLA) steels as areplacement for HY steels.

HSLA steels are lower carbon, cop-per-precipitation-strengthened steels thatmeet the strength and toughness require-ments of HY steels, yet are more easilywelded without preheat. Use of theseHSLA steels in surface ship and subma-rine construction can result in significantreductions in production costs throughthe minimization or elimination of pre-heat, utilization of a wider range of en-ergy inputs and employment of more re-laxed fabrication control (Refs. 3–7).However, welding consumables that takefull advantage of the benefits of HSLAbase metal have yet to be developed, cer-tified and put into production (Refs. 1, 5).Welding consumables originally de-signed for HY-steels are currently beingused to weld HSLA steels for limited ap-

Microhardness Variations in HSLA-100 WeldsFabricated with New Ultra-Low-Carbon Weld

Consumables

BY D. W. MOON, R. W. FONDA AND G. SPANOS

HSLA-100 steel weldments made with ultra-low-carbon consumables wereinvestigated by TEM and microhardness mapping analyses

KEY WORDS

HSLA-100MicrohardnessUltra-Low CarbonHigh-Strength SteelNaval ShipsFiller MetalHydrogen CrackingD. W. MOON, R. W. FONDA and G. SPANOS

are with the Naval Research Laboratory,Washington, D.C.

plications in naval ships, but when madewithout preheat these weldments havebeen found to be susceptible to hydro-gen-assisted cracking. Therefore, theNavy has been spearheading the devel-opment of new welding filler metals forwelding both HSLA and HY steels (Ref. 6).The aim of this program is to develop newwelding consumables that can bewelded with 1) no preheat or no interpasstemperature controls when welding thickHSLA steel plates, 2) reduced costs for in-spection and hydrogen damage controlsand 3) reduced preheat/interpass tem-peratures and heat input controls, in-cluding elimination of postweld soak,during welding of HY-100 base plate(Ref. 6).

The major design concept in develop-ing these new consumables is to signifi-cantly reduce the carbon content. Thesealloys are therefore referred to as ultra-low-carbon (ULC) (Ref. 5) or ultra-low-carbon-bainitic (ULCB) consumables(Refs. 1, 5). The loss in strength due to thecarbon reduction is at least partially off-set by judiciously increasing the Mn, Moand Ni concentrations (Ref. 8). Usingsuch a welding consumable could widenthe welding process window and reducethe Navy’s shipbuilding cost by millionsof dollars (Ref. 9). The fine scale of themicrostructures present in these ULCsteels cannot be adequately resolved byoptical microscopy to characterize anddifferentiate the different microcon-stituents. Thus, higher-resolution tech-niques such as transmission electron mi-croscopy (TEM) must be used. Only alimited number of studies have employedTEM to characterize the microstructuresin welds made with these new ULC fillermetals, e. g., Refs. 10–13. The majority ofinvestigations have used optical mi-croscopy to correlate the weld mi-crostructures with their mechanicalproperties.

In order for the Navy (and industry) todevelop, certify and employ advancedULC filler metals in ship and submarineconstruction, it is essential that the mi-crostructural and property evolutionthroughout these new ULC weldments bewell understood. This type of knowledgeis critical in developing weld metals andwelding procedures that minimize bothcost and the potential factors for failure,yet provide for equivalent or enhancedfinal properties. Therefore, the aim of thepresent study is to determine the mi-crostructural evolution and microstruc-ture-microhardness correlations through-out the fusion zone and the HAZ of anHSLA-100 weldment made using an ex-perimental ULC filler metal, designatedCTC-03. In addition, direct comparisonsare made between the color microhard-

ness map developed for thecurrent weld and those forwelds made with differentfiller metals, base metalsand heat inputs.

ExperimentalProcedure

Using an exploratoryultra-low-carbon Mn-Ni-Mo weld consumable,CTC-03, 1-in.-thick HSLA-100 steel plates werewelded by the NationalCenter for Excellence inMetalworking Technology(NCEMT). The CTC-03weld consumable was de-veloped by the NCEMTunder its Navy Manufac-turing Technology pro-gram (Ref. 14). The multi-pass GMA weldingprocess employed a sin-gle-V joint configurationwith a 45-deg includedangle. A backing plate ofthe same HSLA-100 baseplate was used beneath theroot. This weldment, CTC-03D, was fabricated in thespray transfer mode with aheat input of 30 kJ/in.(1200 kJ/m), a 16 in./min(41 cm/min) weldingspeed, a 125°F (52°C) pre-heat temperature and a125°F interpass tempera-ture. For further details ofthe welding procedure,see Ref. 14. The chemicalcompositions of the basemetal, the welding con-sumable and the as-deposited weld metal areshown below in Table 1.The mechanical propertiesof the HSLA-100 baseplate are 113 ksi (779 MPa)yield strength (YS), 127 ksi (876 MPa) ul-timate tensile strength (UTS), 22% elon-gation (El), 71% reduction in area (RA)and 91 ft-lb (123 J) CVN impact tough-ness at –120°F (–84°C) (Ref. 8). The aver-age mechanical properties taken from thecenter of the as-deposited weld metal ofthe CTC-03D weld (the D designates the30 kJ/in. heat input) are 114 ksi (786 MPa)YS, 119.2 ksi (822 MPa) UTS, 19.3% El,63.3% RA and a CVN impact toughnessof 124 ft-lb (168 J) at 0°F (–18°C) and 86ft-lb (117 J) at –60°F (–51°C) (Ref. 14).

All microstructural observations weretaken from transverse cross sections fromthe weldment. These cross sections in-cluded the fusion zone, the heat-affected

zone and the unaffected base metal —Fig. 1A. After coarse polishing with suc-cessively finer grit papers, weld cross sec-tions were polished first with 3-µm and 1-µm diamond paste, and then with a 0.06-µm silica suspension. They were subse-quently etched with a solution of picricacid saturated in 100 mL of water with 10g of sodium tridecylbenzene sulfonate(Ref. 15) to reveal the weld solidificationstructure. To study the solid-state mi-crostructure for optical microscopy a 1%nital etchant was used. Microhardnessmaps across the base metal, heat-affectedzone and fusion zone were developedusing a diamond indentor with a 1-kgload. After polishing away the etched sur-

WELDING RESEARCH SUPPLEMENT | 279-s

RE

SE

AR

CH

/DE

VE

LO

PM

EN

T/R

ES

EA

RC

H/D

EV

EL

OP

ME

NT

/RE

SE

AR

CH

/DE

VE

LO

PM

EN

T/R

ES

EA

RC

H/D

EV

EL

OP

ME

NT

Fig. 1 — Transverse cross section of the CTC-03D weldment. A— Macrograph; B — microhardness map; C — macrographwith the selected areas for TEM and optical microscopy studiescircled.

face, indentations were made at 1-mmspacing across a 59 x 28-mm area of theweldment. The data were then interpo-lated and converted into a color micro-hardness map using NCSA Image andDatascope software on a Macintosh com-puter — Fig. 1B. In some regions moreclosely spaced microhardness measure-ments were made (again with a 1-kg load)to more clearly determine microhardnesstrends across certain features. These ad-ditional measurements were performedwith a spacing of 0.25 mm along each offour parallel lines, with a 0.25-mm spac-ing between these lines. Indents fromevery other line were offset (parallel to theline direction) by 0.125-mm, relative tothe adjacent lines. These microhardnessdata are thus presented in Fig. 2 as aver-aged data points every 0.125 mm parallelto the direction of the four lines, but avoidthe potential interference that might ac-company measurements in which the in-dents were actually 0.125 mm apartalong a single line. Additionally, mea-surements taken at progressively furtherspacings than 0.25 mm showed no ap-

preciable changein the microhard-ness values, indi-cating a negligibleinteraction be-tween the indents.

To prepareTEM specimens,3.0-mm-diametercylindrical speci-mens were elec-tro-discharge ma-chined fromselected regions ofthe weldment.Disks of 0.25-mmthickness were cutwith a diamond saw from these cylinders,mechanically thinned to 0.05 mm, thenslightly dimpled. These slices were thenelectropolished to perforation in a Fis-chione twin-jet electropolishing unitwith a solution of 100 g of anhydroussodium chromate in 500 mL of glacialacetic acid. The foils were electropol-ished at room temperature with a currentof about 24 mA (55 V). TEM observations

were performed on a Philips CM 30 op-erating at 300 kV.

For comparison, microhardness mapswere also constructed for welds madewith two other exploratory ULC fillermetals: ARC100 and CTC-08. ARC100was developed by ESAB (a welding con-sumable manufacturer) to meet theNavy’s MIL-100S specifications (82–120ksi yield strength); these are the same

280-s | OCTOBER 2000

RE

SE

AR

CH

/DE

VE

LO

PM

EN

T/R

ES

EA

RC

H/D

EV

EL

OP

ME

NT

/RE

SE

AR

CH

/DE

VE

LO

PM

EN

T/R

ES

EA

RC

H/D

EV

EL

OP

ME

NT

Fig. 2 — Hardness variation across the fusion boundary and heataffected zone within the weld metal of a weld bead in region 5in Fig. 1C. In this plot, the “upper bead” was deposited after andon top of the “lower bead.”

Fig. 3 — Tempered martensite microstructure of the unaffected HSLA-100 base metal. A — Light micrograph; B — TEM micrograph; arrowsindicate some of the coarse martensite.

A

B

Table 1 — Compositions of Base Metals, Welding Consumable and As-Deposited Weld Metal (wt-%)

C Ni Mn Mo Cu Cr Si Al Ti P S V Nb Zr

HY80 base plate (a) 0.16 2.49 0.30 0.43 0.016 1.39 0.25 0.010 0.004 0.004(welded w/ARC100)HSLA100 base plate (b) 0.055 3.41 0.81 0.57 1.53 0.73 0.22 0.009 0.008 0.07 0.001 0.0011 0.032(welded w/CTC03 andCTC08)HSLA100 base plate (a) 0.035 3.14 0.79 0.57 1.37 0.57 0.22 0.027 0.004 0.001 0.028(welded w/ARC100)CTC03 consumable (d) 0.028 3.78 1.54 0.52 0.001 0.01 0.34 0.028 0.001 0.002 0.001CTC03D weld metal (c) 0.035 3.72 1.42 0.46 0.09 0.09 0.33 0.006 0.014 0.008 0.002 0.005CTC08 consumable (d) 0.03 3.72 1.80 0.99 0.001 0.01 0.33 0.025 0.001 0.002 0.003CTC08A weld metal (e) 0.040 3.70 1.56 0.89 0.08 0.06 0.30 0.013 0.005 0.014CTC08C weld metal (e) 0.053 3.70 1.5 0.83 0.34 0.16 0.27 0.009 0.005 0.015ARC100 consumable (f) 0.018 2.67 1.63 0.46 0.001 0.01 0.29 0.002 0.006 0.003 0.001 0.005 0.002ARC100 weld metal (f) 0.025 2.72 1.40 0.47 0.067 0.05 0.26 0.004 0.004 0.003 0.001 0.009 0.001

The chemical compositions of (a) and (b) were determined ny NLR, (c) by Ref. 14, (d) by VIM ingots (Ref. 23), (e) by Ref. 23 and (f) by Electric Boat Corp.

specifications for which CTC-03 was de-veloped. ESAB has worked in conjunc-tion with Electric Boat Corporation and aU.S. Navy working group (Ref. 6) to com-mercialize this alloy. CTC-08 was devel-oped by NCEMT to satisfy the Navy’shigher strength weld specification, MIL-120S (102–123 ksi yield strength) (Ref. 16).The compositions of these consumablesare also provided in Table 1. While theCTC03D-HSLA100 and ARC100-HY80welds were prepared with a heat input of30 kJ/in. (1200 kJ/m), the CTC08A-HSLA100 and CTC08C-HSLA100 weldswere prepared with heat inputs of 55kJ/in. (2200 kJ/m) and 110 kJ/in. (4300kJ/m), respectively.

Results and Discussion

Microstructure-Microhardness Variationsand Correlations Across Weldment CTC-03D

The low-magnification solidificationstructure and the microhardness map ofthe transverse cross section of the CTC-03D weldment are shown in Figs. 1A and1B, respectively. A transparency of thecolor microhardness map (Fig. 1B) wasoverlaid onto the macrograph in Fig. 1Ato directly correlate the microhardnessvariations to specific macroscopic fea-tures, including the individual weldbeads, the bead heat-affected zoneswithin the fusion zone (B-HAZs, where Bstands for bead), the base metal heat-affected zone and the unaffected regionof the base metal. This direct correlationbetween the macrostructure and the mi-crohardness was then used to select re-gions of the weldment (Fig. 1C) for de-tailed microstructural analysis. Theseregions are discussed below.

The HSLA-100 base plate (region 8 inFig. 1C) has a quench-and-tempered mi-crostructure (Fig. 3), which consists pri-marily of fine-lath martensite and signifi-cant amounts of coarse martensite (Ref. 17).TEM shows both of these phases containnumerous small copper precipitates.

The base metal HAZ, region 3 in Fig.1C, contains the hardest regions(340–360 HV) of the entire weldment —Fig. 1B. This high hardness is due to thehigher carbon content of the base metal(Table 1) in combination with the rela-tively high cooling rate due to the heatsink provided by the base metal. Accord-ingly, the microstructure of this region ismartensitic and consists of both a fine un-tempered lath martensite and a coarseautotempered martensite — Fig. 4. Thiscoarse martensite is typically an order ofmagnitude wider than the fine-lathmartensite and often exhibits a planar in-terface at one end. However, optical mi-croscopy tends to underestimate the

amount of coarse martensite inthe sample due to the small sizeof some of these martensiteplates. The hardest regionwithin the HAZ is located aboutmidway across the HAZ — Fig.1B. This is contrary to predic-tions based on grain size argu-ments alone (Ref. 18), whichsuggest the hardness maximumshould occur adjacent to the fu-sion zone boundary. One con-tributing factor is a significantamount of softer coarse au-totempered martensite forms inthe coarse-grained HAZ, whichreduces the microhardness di-rectly adjacent to the fusionboundary.

The overlaying procedureshows the white bands in themacrograph (Fig. 1A) are softand appear even softer wher-ever multiple bands intersect.These white bands exhibitgreater surface roughness thanthe surrounding areas and thusappear whiter when viewedwith oblique lighting — Fig. 1.Each white band is curved, andis clearly associated with theheat-affected zone of the corre-sponding weld bead lyingabove it. However, it is not ob-vious from Fig. 1B preciselyhow the local microhardnessvaries in the vicinity of a whiteband. Thus, finer-scale micro-hardness measurements weremade across the white band thatpasses through region 5 — Fig. 2. This fig-ure shows the hardness variation travers-ing from the fusion zone of a weld bead(B-FZ), across the bead boundary andthrough the bead heat-affected zone inthe weld metal. The hardness initially in-creases upon crossing the bead boundaryfrom the B-FZ into the B-HAZ, then dropssharply to a minimum at the white band.After passing the white band, the hard-ness gradually increases up to the as-de-posited B-FZ hardness of about 300 HV.It is thus clear that the white bands are thesoftest regions of the weldment.

The largest soft region of the weld-ment, Region 1 in Fig. 1, is located in anarea near where three white bands mergetogether. The microstructure of this regionappears by optical microscopy (Fig. 5A) toconsist of lath ferrite with some alignedferrite laths at the grain boundaries. TEMobservations (Fig. 5B) confirm that themicrostructure is predominantly lath fer-rite, with small amounts of retainedaustenite between the ferrite laths. Thisinterlath-retained austenite often appearsdark in TEM images (see arrow in Fig. 5B)

and was identified by electron diffractionin this and other ULC weld metal inves-tigations (Refs. 10, 22). Only minoramounts of martensite were observed inthis region.

The predominance of ferrite at thewhite bands is caused by the low peaktemperature and the relatively slow cool-ing rate at this distance from the beadboundary (Refs. 19, 20). As a result ofheat flow from the bead deposited di-rectly above this white band, this areawas reheated and most likely experi-enced reaustenitizing within or directlyabove the intercritical range. The amountof ferrite may also have been enhancedby recalescence (increase in temperaturecaused by the liberation of the latent heatof transformation associated with theaustenite-to-ferrite transformation) dur-ing cooling. Significant recalescenceduring cooling was observed in similarULC alloys (Ref. 21) and ULC as-deposited weld metal (Ref. 22).

Region 2 is one of the hardest regionswithin the fusion zone. This region is inthe last deposited weld bead, located at

WELDING RESEARCH SUPPLEMENT | 281-s

RE

SE

AR

CH

/DE

VE

LO

PM

EN

T/R

ES

EA

RC

H/D

EV

EL

OP

ME

NT

/RE

SE

AR

CH

/DE

VE

LO

PM

EN

T/R

ES

EA

RC

H/D

EV

EL

OP

ME

NT

Fig. 4 — The HAZ microstructure of the base metal (re-gion 3 in Fig. 1C). A — Light micrograph; B — TEM mi-crograph showing fine untempered lath martensite (LM)and coarse autotempered martensite (CM).

A

B

the top right of the weld in Fig. 1C.Since this is the last-depositedbead, the microstructure was notaltered by subsequent bead depo-sition. Optical microscopy of Re-gion 2 (Fig. 6A) reveals long lathsof ferrite (up to more than 50 µmin length) within large prioraustenite grains. TEM reveals thepresence of significant amounts offine-lath martensite betweenthese large ferrite laths, as shownin Fig. 6B. This martensite is re-sponsible for the high microhard-ness of this region. Optical mi-croscopy often tends tooverestimate the amount of ferritein a microstructure when both fer-rite and martensite are present inULC steels (Ref. 10).

While no solid-state precipi-tates, such as carbides or car-bonitrides, were ever observed inthe ULC fusion zone, widely scat-tered spherical oxide inclusions(usually less than 1 µm in diame-ter) are present throughout the fu-sion zone — Figs. 6B, 7B. Energy-dispersive spectrometry revealsthese inclusions to contain nearlyequal amounts of manganese andtitanium oxides in addition tosmall amounts of silicon oxideand aluminum oxide.

Region 4 (Fig. 1C) is one of thehardest regions of the fusion zoneand lies within the root pass. Thehigh microhardness of this regionappears to be due to dilution fromthe backing plate in combinationwith the fast cooling rates from thebacking plate and the base metal

on either side of the V. Optical mi-croscopy observations of Region 4 reveala microstructure of acicular ferrite in abasket-weave morphology below thewhite band (Fig. 7A), while lath ferritepredominates above the white band. TEMobservations (Fig. 7B) reveal the presenceof isolated small packets of lath marten-site within the acicular ferrite, contribut-ing to the high hardness of this region.

Region 5 is a relatively hard region(330–340 HV) in the middle of the fusionzone — Fig. 1. This region, located be-tween a bead boundary and its whiteband, appears somewhat similar to Re-gion 1 in optical microstructure, exceptwith finer features. The large ferrite lathsobserved in Regions 2 and 7 are not ob-served in this region. TEM observationsinstead reveal a majority of finer-lath fer-rite with significant amounts of lathmartensite.

Region 6, a soft region in the center ofthe fusion zone, is located below two ad-jacent bead boundaries — Fig. 1C. Thereappears to be no significant difference inmicrostructure between this area and Re-gion 1 — Fig. 5. It consists mostly ofaligned lath ferrite and possibly somelath martensite.

Region 7 lies in the B-HAZ of the lastbead, about midway between the beadboundary and the white band — Fig. 1C.Both optical microscopy and TEM reveala microstructure of long laths of ferritevery similar to that observed in Region 2(Fig. 6B), with lath martensite and somefine-lath ferrite between the larger ferritelaths. The relatively fast cooling rate ofthis region produced a significant amountof lath martensite, resulting in the ob-served high hardness. The microhardness

282-s | OCTOBER 2000

RE

SE

AR

CH

/DE

VE

LO

PM

EN

T/R

ES

EA

RC

H/D

EV

EL

OP

ME

NT

/RE

SE

AR

CH

/DE

VE

LO

PM

EN

T/R

ES

EA

RC

H/D

EV

EL

OP

ME

NT

Fig. 5 — The microstructure of a large soft region in thefusion zone, in region 1 of Fig. 1C. A — Light micro-graph; B — TEM micrograph of predominantly lath fer-rite with some retained austenite (arrowed).

Fig. 6 — Microstructure of the fusion zone in region 2of Fig. 1C. A — Light micrograph; B — TEM micrographof relatively coarse laths of ferrite (F) within a lathmartensite (LM) microstructure.

A

A

B

B

values in this region of the B-HAZ areslightly higher than those in the fusionzone of the bead, as shown in Fig. 2.

Comparison of Microhardness Maps fromWelds Made with Different ULCConsumables

To compare the local property varia-tions of welds made with different ULCfiller metal candidates, the microhard-ness map of the weldment of HSLA-100steel plate welded with the CTC-03 con-sumable (CTC-03D/HSLA-100) is com-pared in Fig. 8 with microhardness mapsof welds made with other ULC consum-ables with various heat inputs. Theseweldments are identified as CTC-08A/HSLA-100, CTC-08C/HSLA-100,ARC-100/HY-80 and ARC-100/HSLA-100. The microhardness maps of the twoCTC-08 welds (CTC-08A/HSLA-100 andCTC-08C/HSLA-100) were previously re-ported by Fonda, et al. (Ref. 10), while themicrohardness maps of the two ARC-100welds (ARC-100/HY-80 and ARC-100/HSLA-100) were constructed for thepresent study. In all five weldments, thehardest region is located in the HAZabout midway between the fusion zoneboundary and the outer edge of the HAZ— Fig. 8. The weldments also show sim-ilar local hardness variations in the fusionzone — Fig. 8. The softest regions are lo-cated at the white bands, while the hard-est regions are between those whitebands and the weld bead boundaries.

CTC-03D vs. CTC-08A and CTC-08C

The CTC-08 welding consumable wasdesigned to weld HSLA-100 steels, pro-ducing weldments that meet the MIL-120Sspecifications (Ref. 16). The CTC-08A andCTC-08C weldments shown in Fig. 8 wereprepared by welding 1-in.-thick baseplates of HSLA-100 with the CTC-08 con-sumable using heat inputs of 55 kJ/in.(2200 kJ/m) and 110 kJ/in. (4300 kJ/m), re-spectively. The fusion zone of the CTC-08C weld is clearly softer than that of theCTC-08A weld (Fig. 8), due to the higherheat input and resultant slower coolingrate of the CTC-08C weld, producingsofter, ferritic microstructures. On theother hand, the HAZ hardnesses of thesewelds are comparable despite their differ-ences in heat input.

Although the CTC-03 weldments sat-isfy the lower strength MIL-100S specifi-cation (Ref. 14) and the CTC-08 weld-ments meet the higher strength MIL-120Sspecification (Ref. 16), the fusion zonehardness of the CTC-03D weld is com-parable to that of CTC-08A and actuallyexceeds that of CTC-08C. This is due tothe high cooling rate of CTC-03D, de-

spite its lean composition (Table1). Nevertheless, CTC-03D ex-hibits weld metal cracking forhigh cooling rates (Refs. 23, 24).Therefore, the U.S. Navy shiftedits emphasis onto a more promis-ing ULC filler metal, designatedARC-100, which exhibits lesssusceptibility to weld metalcracking and better impacttoughness at high cooling rates(Refs. 23, 24).

CTC-03D vs. ARC-100 on HY-80or HSLA-100

ARC-100 was developed tomeet the MIL-100S specificationfor welding of both HY and HSLAsteels. Figure 8 shows a compari-son between 2-in. -thick (50-mm)weldments of these two baseplates welded with ARC-100 atthe same heat input (30 kJ/in.). Al-though the fusion zone of theARC-100/HY-80 weldment isvery similar to the ARC-100/HSLA-100 weldment over-all, the HAZ hardness of the HY-80 weldment is much higher(about 410 HV) than that of theHSLA-100 weldment (about 320HV), as anticipated due to thehigher carbon content of HY-80— Table 1. This comparison pro-vides a quantitative example ofthe benefits of the HSLA baseplate over HY-80 steel plate. Even at alow heat input (30 kJ/in.) and withoutpreheat, the HAZ of the higher strengthHSLA-100 has lower hardness than thatof lower strength HY-80. Therefore, theHSLA-100 HAZ should be much lesssusceptible to hydrogen-assisted crack-ing than the HY-80 HAZ. Nevertheless,the potential for hydrogen-assisted crack-ing in the HSLA-100 HAZ should still beconsidered due to its high hardness.

The ARC-100/HSLA-100 fusion zoneis softer than that of CTC-03D, reflectingtheir compositional differences. For thisdouble-V weld configuration, the rootpass region of the ARC-100/HSLA-100weld is harder than the surrounding areaof the fusion zone, as shown in Fig. 8E.The higher hardness of this region is at-tributed both to dilution from the baseplate and to the use of a compositionallyrich flux coated consumable (MIL-10718-M) for the first two passes. Fastcooling rates due to the large heat sink ofthat area also contribute to the high hard-ness in this region. It is obvious from themicrohardness map the fusion zone islower in hardness than the HSLA-100base plate (undermatched), even for thelow 30 kJ/in. heat input used. This sug-

gests even lower heat inputs might beconsidered, increasing the fusion zonehardness and matching with the unaf-fected base plate hardness. However, afurther reduction in heat input might fur-ther elevate the already high hardness ofthe HAZ and increase the HAZ suscepti-bility to hydrogen-assisted cracking.

Summary

Microstructure and microhardnessvariations have been investigated andcorrelated in an HSLA-100 steel weld-ment fabricated with an exploratoryultra-low-carbon (ULC) welding con-sumable designated CTC-03. These re-sults are compared to the microhardnessmaps of four other weldments, madewith different ULC candidate filler metals(designated CTC-08 and ARC-100), baseplates and heat inputs. The principal find-ings from this investigation are summa-rized as follows:

1) A color microhardness mappingmethod in conjunction with detailedtransmission electron microscopy andoptical microscopy from local regionsacross weldments enables the direct cor-relation of microhardness variations tothe corresponding microstructures.

WELDING RESEARCH SUPPLEMENT | 283-s

RE

SE

AR

CH

/DE

VE

LO

PM

EN

T/R

ES

EA

RC

H/D

EV

EL

OP

ME

NT

/RE

SE

AR

CH

/DE

VE

LO

PM

EN

T/R

ES

EA

RC

H/D

EV

EL

OP

ME

NT

Fig. 7 — Microstructure from the hardest area within theroot pass, region 4 in Fig. 1C. A — Light micrograph; B— TEM micrograph showing acicular ferrite with a smallpacket of lath martensite (LM).

A

B

2) The fusion zone consists predomi-nantly of lath ferrite with varyingamounts (depending on location) of un-tempered fine lath martensite, as well assmall amounts of interlath retainedaustenite and oxide inclusions. Nopolygonal ferrite or solid-state precipi-tates such as carbides or carbonitrideswere observed in the fusion zone. Thelocal variations in microhardness corre-late well with the local variations in themicrostructure.

3) The heat-affected zone of the basemetal was the hardest region in eachweldment examined, regardless of fillermetal type, base metal or heat input. Thehardness reaches a maximum about mid-way through the heat-affected zone ofeach weldment studied, rather than adja-cent to the fusion boundary. In the CTC-03D weldment, the HAZ microstructureconsists predominantly of untempered

lath martensite, which is potentially sus-ceptible to hydrogen cracking. The hard-ness decreases toward the fusion zone,where significant amounts of autotem-pered coarse martensite are also ob-served. Although the potential for failureis much worse for the HAZ of the HY-80weldment, the hardness of the HSLA-100HAZ remains significantly high. The mi-crostructure and properties of the basemetal HAZ, therefore, still deserve atten-tion, aside from the improved fusionzone properties resulting from new ULCfiller metals.

4) The softest regions in the fusionzone are in (and just outside) the curvedwhite bands observed by low-magnifica-tion optical microscopy. The microstruc-ture within these white bands consistspredominantly of lath ferrite and alignedgrain boundary ferrite, with smallamounts of retained austenite.

5) In the vicinity of a weld bead, thehardness increases upon crossing theweld bead boundary into the bead heat-affected zone within the weld metal, thenfalls sharply to its lowest value at thewhite band. The hardness then graduallyincreases again up to that exhibited bythe untempered microstructure of thenext bead. The harder regions corre-spond to an increasing volume fraction offine untempered lath martensite.

6) Despite the low heat input (30 kJ/in.),the fusion zone hardnesses of two of thenew ultra-low-carbon filler metals, CTC-03D and ARC-100, are comparable to thebase metal hardness. In fact, the low heatinput of the CTC-03D weldment pro-duced a fusion zone with a hardnesscomparable to that of the high heat input(but richer composition) CTC-08A weld-ment. However, when compared to theCTC-03 filler metal, the ARC-100 fillermetal exhibits better impact toughnessand better resistance to weld metal crack-ing at high cooling rates.

Acknowledgments

The authors would like to acknowl-edge funding through the Office of NavalResearch (George Yoder, program man-ager) under contract 96PR04294-00. Weacknowledge many helpful discussionswith Paul Konkol and coworkers atNCEMT and Neil Fichtelberg of EB. Wealso appreciate the leadership of andhelpful discussions with the managers ofthe Navy’s Ultra Low Carbon Weld Con-sumable Development Program, JohnDeloach of NSWC-CD (technical man-ager) and Charles Null of NAVSEA (tech-nical sponsor). We would particularlylike to acknowledge the frequent assis-tance of Ed Pierpoint with many aspectsof this research.

References

1. Holsberg, P. W., Gudas, J. P., and Ca-plan, I. L. 1989. Metallurgical design andprocesses in the U.S. Navy high strength steelwelding. Recent Trends in Welding Scienceand Technology TWR’89, edited by S. A.David and J. M. Vitek, p. 593.

2. Flax, R. W., Keith, R. E., and Randall, M.D. 1971. Welding the HY Steels. ASTM Spe-cial Technical Publication 494, p.13.

3. Losz, I. M. B., and Challenger, K. D.1989. Microstructure and properties of a cop-per-precipitation strengthened HSLA steelweldment. Proceedings of the 2nd Interna-tional Conference on Trends in Welding Re-search, Gatlinburg, Tennessee, p. 229.

4. Czyryca, E. J., Link, R. E., Wong, R. J.,Aylor, D. A., Montemarano,T. W., and Gudas,J. P. 1990. Development and certification ofHSLA-100 steel for naval ship construction.

284-s | OCTOBER 2000

RE

SE

AR

CH

/DE

VE

LO

PM

EN

T/R

ES

EA

RC

H/D

EV

EL

OP

ME

NT

/RE

SE

AR

CH

/DE

VE

LO

PM

EN

T/R

ES

EA

RC

H/D

EV

EL

OP

ME

NT

Fig. 8 — Microhardness maps of welds made with three different filler metals and different weld-ing parameters. The corresponding microhardness scale is included at the bottom of this figure.

Naval Engineers Journal, 102(3): 63.5. Cullison, A. 1994. Two paths, one goal:

a consumable to weld HSLA100. WeldingJournal 73(1): 51.

6. DeLoach, Jr., J. J. 1995. An overview of theU.S. Navy filler metal development program.Proceedings of a TMS Symposium on Weldingand Weld Automation in Shipbuilding, editedby R. DeNale, Cleveland, Ohio, p. 85.

7. Holsberg, P. W., Guda, J. P., and Caplan,I. L. 1995. Navy’s welding research picks upsteam. Advanced Materials and Processes138(1): 45.

8. Sampath, S. K., Civis, S. K., Dong, H.,Konkol, P. J., and Green, R. S. 1995. Evaluationof new performance electrodes for GMA weld-ing of HSLA-100 steel. Proceedings of the ASMInternational Symposium on High StrengthSteels for Structural Applications, p.180.

9. Irving, B. 1995. U.S. Navy maintainshigh interest in funding for welding research.Welding Journal 74 (3): 41.

10. Fonda, R. W., Spanos, G., and Vander-meer, R. A. 1995. Processing-microstructure-property relationships. Proceedings of the 4thInternational Conference Trends in WeldingResearch, Gatlinburg, Tenn., p. 277.

11. Spanos, G., Fonda, R. W., Vandermeer,R. A., and Matuszeski, A. 1995. Microstruc-tural changes in HSLA steel thermally cycled

to simulate the heat-affected zone duringwelding. Metallurgical and Materials Transac-tions A, vol. 26A, p. 3279.

12. Blackburn, J. M., Vassilaros, M., Bran-demarte, A., Fox, A., and Franke, G. 1995. TheEffects of thermal cycling on ULCB weldmetal. A Proceedings on Welding and Au-tomation in Shipbuilding, by Structural Mate-rials Division of TMS, edited by R. DeNale,p.167.

13. Devletian, J. H., Signh, D., and Wood,W. E. 1995. Ultra low carbon bainitic weldmetal for joining 65 to 100 ksi yield steels. Aproceedings on Welding and Automation inShipbuilding, by Structural Materials Divisionof TMS, edited by R. DeNale, p.151.

14. Konkol, P. J. 1997. Commercializationof Advanced Welding Consumables for HSLA-80 and HSLA-100 Steels, reported by NationalCenter for Excellence in Metal Working Tech-nology, in Johnstown, Pa.

15. Brewer, A. W., Erven, K. A., and Krauss,G. 1991. Materials Characterization, p. 53, 27.

16. MIL-E-23765/2E 1994. Electrodes andRods — Welding Bare, Solid, or Alloy Cored;and Fluxes, Low Alloy Steel.

17. Fonda, R. W., Spanos, G., and Vander-meer, R. W. 1994. Observations of latemartensite in a low carbon steel,. Scripta Met-allurgica 3: 683.

18. Ion, J. C., Eastering, K. E., and Ashby,M. F. 1984. A second report on diagrams ofmicrostructure and hardness for heat-affectedzones in welds. Acta Metallurgica 32(11):1949.

19. Davenport, E. S., and Aborn, R. H.1936. Metallurgical Aspects of the Welding ofSteel, presented at Annual Meeting of AWS,Cleveland, Ohio. A contribution from the Re-search Laboratory of United States Steel Corp.

20. Nippes, E.F. 1959. The weld heat-affected zone. Welding Journal 38(1): 1-s to 18-s.

21. Fonda, R. W., Vandermeer, R. A., andSpanos, G. 1998. Continuous Cooling Trans-formation (CCT) Diagram for Advanced NavyWelding Consumables. Naval Research Labo-ratory Report NRL/MR/6324-98-8185.

22. Fonda, R. W., and Spanos, G. 1999.Microstructural evolution in ultra-low carbonsteel weldments — part I: controlled thermalcycling and CCT diagram of the weld metal.Metallurgical Transactions A, in press.

23. Konkol, P. 1998. Concurrent Tech-nologies Corp., private communication.

24. Deloach, J. D., Null, C., Fiore, S., andKonkol, P. 1999. The right welding wire couldhelp the U.S. Navy save millions. WeldingJournal 78(6): 55.

WELDING RESEARCH SUPPLEMENT | 285-s

RE

SE

AR

CH

/DE

VE

LO

PM

EN

T/R

ES

EA

RC

H/D

EV

EL

OP

ME

NT

/RE

SE

AR

CH

/DE

VE

LO

PM

EN

T/R

ES

EA

RC

H/D

EV

EL

OP

ME

NT

All authors should address themselves to the followingquestions when writing papers for submission to theWelding Research Supplement:

◆ Why was the work done?◆ What was done?◆ What was found?◆ What is the significance of your results?◆ What are your most important conclusions?With those questions in mind, most authors can

logically organize their material along the following lines,using suitable headings and subheadings to divide thepaper.

1) Abstract. A concise summary of the majorelements of the presentation, not exceeding 200 words, tohelp the reader decide if the information is for him or her.

2) Introduction. A short statement giving relevantbackground, purpose and scope to help orient the reader.Do not duplicate the abstract.

3) Experimental Procedure, Materials, Equipment.4) Results, Discussion. The facts or data obtained

and their evaluation.5) Conclusion. An evaluation and interpretation of

your results. Most often, this is what the readers remember.6) Acknowledgment, References and Appendix.Keep in mind that proper use of terms, abbreviations

and symbols are important considerations in processing amanuscript for publication. For welding terminology, theWelding Journal adheres to ANSI/AWS A3.0-94, StandardWelding Terms and Definitions.

Papers submitted for consideration in the WeldingResearch Supplement are required to undergo PeerReview before acceptance for publication. Submit anoriginal and one copy (double-spaced, with 1-in. margins on8

1⁄2 x 11-in. or A4 paper) of the manuscript. Submit the

abstract only on a computer disk. The preferred format isfrom any Macintosh® word processor on a 3.5-in. double-or high-density disk. Other acceptable formats includeASCII text, Windows™ or DOS. A manuscript submissionform should accompany the manuscript.

Tables and figures should be separate from themanuscript copy and only high-quality figures will bepublished. Figures should be original line art or glossyphotos. Special instructions are required if figures aresubmitted by electronic means. To receive completeinstructions and the manuscript submission form, pleasecontact the Peer Review Coordinator, Doreen Kubish, at(305) 443-9353, ext. 275; FAX 305-443-7404; or write tothe American Welding Society, 550 NW LeJeune Rd.,Miami, FL 33126.

Preparation of Manuscripts for Submissionto the Welding Journal Research Supplement