Evaluation of Heat-Affected Zone Hydrogen-Induced Cracking in Navy Steels

The implant test was conducted on HY-100, HSLA-100, and HSLA-65, plus the hydrogen-induced cracking susceptibility was quantitatively evaluated
BY X. YUE AND J. C. LIPPOLD ABSTRACT
The implant test was conducted on HY-100, HSLA-100, and HSLA-65 to evaluate their susceptibility to heat-affected zone (HAZ) hydrogen-induced cracking (HIC). The stress vs. time to failure curve was plotted, and the normalized critical stress ratio (NCSR) and embrittlement index for each steel were determined, which can be used to quantitatively evaluate HIC susceptibility. The coarse-grained HAZ (CGHAZ) microstructure of the three steels was characterized by means of optical and transmission electron microscopy. In addition, SEM fractography was conducted to study the HIC fracture behavior. Intergranular (IG), quasi-cleavage (QC), and microvoid coalescence (MVC) fracture modes were found to occur sequentially during the crack initiation and propagation process. The fracture behavior observed in the present investigation is in good agreement with Beachems model. It can be concluded based on the implant test results that, among the three steels, HY-100 is the most susceptible to HAZ HIC while HSLA-100 and HSLA-65 exhibit good resistance. The difference in the HIC susceptibility of the three steels is further explained by combining the microstructure characterization of the CGHAZ and fracture behavior. These results can serve as a benchmark for the future development of high-performance Navy steels. of steels. Among various weldability issues of high-strength steels, hydrogen-induced cracking (HIC) (also referred as cold cracking) in the heat-affected zone (HAZ) following welding is of concern (Refs. 46), and thereby it is important to evaluate the naval steels susceptibility to HAZ HIC. Within the HAZ, the coarsegrained HAZ (CGHAZ) is the most susceptible to formation of untempered martensite with coarse grain size (Refs. 7, 8), and therefore potentially the most susceptible to HIC (Ref. 9). Based on the strong microstructure influence on HIC, CCT diagrams for the CGHAZ of HY100, HSLA-100, and HSLA-65 have been constructed, as described in an earlier publication (Ref. 10). In parallel with that study, the susceptibility to HIC has been evaluated for the same three steels. In the present investigation, the implant test is used to evaluate susceptibility to HIC. The microstructure of the weld CGHAZ from this test is characterized, and fractography is conducted to illustrate the HIC fracture behavior. The HIC test experimental results will be used to develop a weldability database of current Navy steels, which can serve as a benchmark for the future development of highperformance steels.

WELDING RESEARCH

Materials and Experimental Procedure

HY-100, HSLA-100, and HSLA-65 were provided in the form of rolled plate by the Naval Surface Warfare Center, Carderock Division, West Bethesda, Md. Table 1 summarizes the chemical compositions of the three steels used in this investigation. The steel plates were machined into the implant specimens, whose dimensions are listed in Table 2. The implant test used in the present investigation was first developed by Henri Granjon at the Institut de Soudure (French Welding Institute) (Ref. 11). In the implant test, a cylindrical sample with a 0.5-in.- (12.7-mm-) long 10-32 UNF thread on one end is inserted into a clearance hole in the center of the specimen plate. The other end with a 0.5-in.- (12.7mm-) long 1/4-20 UNC thread is inserted into a threaded connection rod of the loading system so that it is possible to apply a tensile load on the specimen after welding. A weld bead was then deposited on the top surface of the specimen plate directly over the threaded sample and hole, creating a HAZ in the 10-32 UNF thread region, as shown in Fig. 1. The thread serves to create a stress concentration in the HAZ region, thereby causing HIC to occur in the HAZ instead of the fusion zone. Two minutes after completion of welding, the sample is loaded in tension when the temperature of the weld assembly is in the range of 100150C. The tensile load is provided by The Ohio State University Modified Implant Testing System (OSU-MITS), as shown in Fig. 2, which was specially designed and built to

Introduction
Ship structures are subject to a complex dynamic loading during service that is superimposed on residual stress present as a result of fitup and fabrication. Therefore, high-performance steels for ship structure applications have been a constant goal pursued by the United States Navy. In order to meet the requirement for good combination of high strength and low-temperature fracture toughness, high-yield-strength steels (HY series) and high-strength, low-alloy steels (HSLA series) have been under development by the Navy for the last 50 years. Among them, HY-100, HSLA-100, and HSLA-65 are used extensively in surface ship and submarine construction today, and they will continue to be the principal structural materials in the foreseeable future (Refs. 13). Navy shipbuilding has been heavily reliant on welding as a fabrication technique, and it has been of great practical importance to conduct weldability testing
X. YUE (yuexinosu@gmail.com) and J. C. LIPPOLD are with the Welding Engineering Program at The Ohio State University, Columbus, Ohio. Based on a paper presented at FABTECH 2012 in Las Vegas, Nev., November 1214, 2012.

KEYWORDS
Implant Test Hydrogen-Induced Cracking CGHAZ Microstructure Fracture Behavior HSLA-100 HY-100 HSLA-65

20-s JANUARY 2013, VOL. 92

Fig. 1 Schematic drawing of the implant test.

accurately control loading during the test. The design of the OSU-MITS ensures the implant specimen is free of bending, torsion, or shock loading. It has the capability of providing tensile loads of up to 10,000 lb (4500 kg), and the entire system is easily moved. The times to failure for a series of tests performed at various stress levels were recorded by a computer equipped with a data-acquisition system connected to the OSU-MITS. The load (stress) applied is then plotted against the time to failure. The highest stress at which no failure occurs after 24 h loading is defined as the lower critical stress (LCS), which is taken as an index to determine susceptibility to HIC. Flux cored arc welding (FCAW) was used to deposit the weld bead on the top surface of the specimen plates. The welding consumable used was the 0.047 in. (1.2 mm) Pipeliner111M (AWS E111T1GM) provided by The Lincoln Electric Co. Welding parameters were as follows: 25 V, current 225235 A, travel speed 12 in./min (5.1 mm/s), and wire feed speed 300 in./min (127 mm/s). This corresponds to a heat input in the range of 28.1 to 29.4 kJ/in. (1.11 to 1.16 kJ/mm). Ar + 25%CO2 is the recommended shielding gas for this consumable; however, in order to minimize the hydrogen loss, pure argon at a flow rate of 45 ft3/h (21.2 L/min) was used instead. Welding with this consumable can produce diffusible hydrogen content in the range of 45 mL/100 g for typical performance as stated in the product specification. Previous tests show that cracking did not occur without intentional introduction of diffusible hydrogen. Then, in order to introduce sufficient diffusible hydrogen to cause HIC in HAZ, a thin film of lubricating oil was applied evenly on the specimen plate surface before welding, and the amount of oil was carefully controlled each time. This produced an average diffusible hydrogen content of 8.1 mL/100 g (four samples were tested with a standard devia-

Fig. 2 The OSU Modified Implant Testing System (OSU-MITS) and implant specimen. A Full view of the testing system; B close-up view showing an implant specimen under loading and an unloaded one on the top right corner; C the implant specimen.

tion of 0.2 mL/100 g), which was measured using the gas chromatograph method in accordance with AWS A4.3. Metallographic samples were sectioned perpendicular to the welding direction through the axis of the implant speci-

mens. Then they were mounted, polished, and etched with 5% nital and examined using an optical microscope. The TEM samples were evaluated in a Philips CM200 TEM operated at 200 kV. Vickers hardness measurements were conducted

Table 1 Chemical Composition of the Test Steels Element (wt-%) C Mn Si P S Cu Ni Cr Mo V Nb Ti HY-100 0.18 0.28 0.21 0.008 0.002 0.15 2.32 1.37 0.26 <0.01 <0.01 <0.01 HSLA-100 0.051 0.90 0.25 0.008 0.002 1.17 1.58 0.60 0.37 <0.01 0.017 <0.01 HSLA-65 0.074 1.35 0.24 0.011 0.006 0.25 0.34 0.14 0.06 0.058 0.018 0.012

WELDING JOURNAL 21-s

WELDING RESEARCH

Fig. 3 Typical implant test specimen sectioned near the implant axis. HY-100. 5% nital etch.

Table 2 Specimen Plate/Implant Specimen Dimensions Specimen Plate Material Plate thickness in. (mm) Plate width in. (mm) Plate length in. (mm) Length of the test bead in. (mm) Hole diameter in. (mm) Implant Specimen Material Total length of implant specimen in. (mm) Type of thread Pitch in. (mm) Major diameter in. (mm) Minor diameter in. (mm) Thread length in. (mm) Thread angle Thread root radius in. (mm) HY-100, HSLA-100, and HSLA-65 1 (25.4) 10-32 UNF 1/32 (0.79) 0.1900 (4.83) 0.1517 (3.85) 0.5 (12.7) 60 deg 0.004 (0.1) A36 Steel 0.5 (12.7) 2 (50.8) 4 (101.6) 3.5 (88.9) 0.201 (5.1)

WELDING RESEARCH

on the as-polished samples using a 1-kg load, in accordance with ASTM E 384-10.

Results and Discussion

Weld Macrostructure and Hardness

Figure 3 shows a transverse section of a typical test weld taken along the axis of the implant specimen. The fusion boundary separating the weld metal and implant specimen is clearly discernable. Due to the excessive grain growth and possible formation of susceptible (high hardness) microstructure, HIC will most likely occur in the CGHAZ region, which is just adjacent to the fusion boundary. Note that the HAZ in the implant specimen is much wider than that in the adjacent plate due to the difference in heat flow and temperature gradient. Vickers hardness measurements were taken along the axis of the implant specimens of the three steels, as shown in Fig. 4AC. The hardness variation from weld metal to HAZ and the base metal is apparent with the hardness of the weld metal

is in the range of 360 to 380 HV. The hardness of CGHAZ of HY-100 is higher than that of the weld metal, which is in the range of 420 to 440 HV. While for HSLA-100 and HSLA-65, their CGHAZ hardness is lower than the weld Fig. 4 Vickers hardness measurements taken along metal, in the range of 325 to 340 HV and the axis of the implant specimen. A HY-100; B 300 to 317 HV, respectively. The location HSLA-100; C HSLA-65. of the CGHAZ, as shown in Fig. 4AC, is determined by metallographic obsertained austenite (Refs. 13, 14). The forvation. It should be noted that the red dotmation of lath martensite is consistent ted line is only the approximate boundary with a CGHAZ in HY-100 with hardness between the CGHAZ and adjacent finein the range of 420 to 440 HV. grained HAZ (FGHAZ). The CGHAZ microstructure of HSLA-100 is shown in Fig. 6AC. MartenMicrostructure Characterization of CGHAZ site and bainite with a needle-like morof the Three Steels phology form in the CGHAZ as shown in Fig. 6A. Similar to HY-100, because the Figure 5 shows the optical and TEM carbon content of HSLA-100 is relatively bright-field microstructure of the low (0.051 wt-%), the martensite formed CGHAZ from the HY-100 implant speciis of the lath type, whose morphology is men. Martensite forms in the CGHAZ of clearly seen in Fig. 6B. It is shown in Fig. this steel, and the packet of the martensite 6C that parallel laths with small intra-lath laths (Ref. 12) can be seen in the higher platelet-like precipitates form in the mimagnification TEM microstructure, as crostructure. The precipitates are most shown in Fig. 5B. The dark thin film belikely cementite, and they are directionally tween martensite laths is probably reoriented. This morphology is characteris-

22-s JANUARY 2013, VOL. 92

Fig. 5 CGHAZ microstructure of HY-100. A Optical; B bright-field TEM.

Fig. 6 CGHAZ microstructure of HSLA-100. A Optical; B and C bright-field TEM.

tic of lower bainite (Ref. 15). Therefore, the HSLA-100 CGHAZ microstructure is a mixture of lath martensite and bainite, which has hardness in the range of 325 to 340 HV. The CGHAZ microstructure of HSLA-65 is shown in Fig. 7AC. It can be seen in Fig. 7A that a small amount of ferrite forms along the prior austenite grain boundaries. The morphology of packets of parallel laths free of precipitates can be seen in Fig. 7B, which is the feature of lath martensite. Similar to HSLA-100, directionally aligned intra-lath cementite

platelets can be observed in Fig. 7C, which confirms the presence of lower bainite. The difference is that these cementite platelets are coarser than the ones forming in HSLA-100. The formation of a mixture of ferrite, bainite, and martensite results in a HSLA-65 CGHAZ with the lowest hardness (300 to 317 HV) among the three steels.
The Implant Test Results

A series of composite weldments of specimen plates and steel implants

welded with the same parameters was subject to different levels of tensile loading after welding, and the time to failure at each stress level was recorded. These data were used to generate the implant test curves for the three steels, as shown in Fig. 8AC. The tensile stress is equal to the load divided by the cross-sectional area of the root diameter of the 10-32 thread. Because the lower critical stress (LCS) is an important index to quantify HIC susceptibility, tests run at the highest stress at which the implant specimen does not fail after 24 h tensile loading

Table 3 Implant Test Results Steel CGHAZ Max Hardness (HV) 440 340 317 CGHAZ Tensile Strengtha ksi (MPa) 212 (1462) 154 (1062) 145 (1000) Lower Critical Stress ksi (MPa) 72 (496) 83(572) 76 (524) Nominal Yield Strength ksi (MPa) 100 (689) 100 (689) 65 (448) NCSRb Embrittlement Indexc

HY-100 HSLA-100 HSLA-65

0.72 0.83 1.17

0.34 0.54 0.52

(a) The CGHAZ tensile strength is converted from the CGHAZ max hardness using the ASTM hardess conversion chart. (b) NCSR stands for normalized critical stress ratio, which is the ratio of lower critical stress to nominal yield strength. (c) Embrittlement index is the ratio of lower critical stress to the CGHAZ tensile strength.

WELDING JOURNAL 23-s

WELDING RESEARCH

Fig. 7 CGHAZ microstructure of HSLA-65. A Optical; B and C bright-field TEM.

WELDING RESEARCH

Fig. 8 The implant test result curves of the following: A HY-100; B HSLA-100; C HSLA-65.

Fig. 9 Fracture morphology of HY-100 implant specimen at a stress level of 91.3 ksi (629 MPa) that failed after 3 min of loading. A General fracture appearance (white arrow indicates the direction of crack growth); B region I (IG); C region II (QC); D region II (MVC).

were repeated twice to verify the LCS. The delayed nature of HIC can be seen from Fig. 8AC, and, as expected, there is a longer incubation time before implant specimen fracture with lower ap-

plied stress levels. It is also found that at comparable stress levels, the incubation time of HY-100 is shorter than that of HSLA-100. The implant test results for the three

steels are provided in Table 3. The LCS is determined to be 72 ksi (496 MPa), 83 ksi (572 MPa), and 76 ksi (524 MPa) for HY100, HSLA-100, and HSLA-65, respectively. Because the three steels have different strength levels, a normalization procedure for critical stress is required. Also, in order to avoid the effect of variation in notch tensile stress on quantitatively rating HIC susceptibility, instead of using the embrittlement index used by Sawhill (Ref. 16), a normalized critical stress ratio (NCSR) is used in the present investigation. This is simply the ratio of LCS to nominal yield strength of the test steels (Ref. 17). Therefore, an exact measure of HIC on percent degradation from nominal yield strength could be determined. As the nominal yield strength is 100 ksi (689 MPa), 100 ksi (689 MPa), and 65 ksi (448 MPa) for HY-100, HSLA100, and HSLA-65, respectively. The NCSR is determined accordingly to be 0.72, 0.83, and 1.17 for HY-100, HSLA100, and HSLA-65, respectively. Beside the NCSR, a new embrittlement index, which is the ratio of the lower critical stress to CGHAZ tensile strength, is proposed in the present investigation to compare the HIC susceptibility of the three steels. Due to martensite formation in the CGHAZ, the maximum CGHAZ hardnesses is 440, 340, and 317 HV for HY-100, HSLA-100, and HSLA-65, respectively. And they are much higher than the three steels re-

24-s JANUARY 2013, VOL. 92

spective base metal hardnesses, which are 283, 284, and 201 HV for HY-100, HSLA-100, and HSLA-65, respectively. That means the weld CGHAZ has a higher tensile strength than the base metal for the three steels. If the tensile strength of the CGHAZ can be determined, then the influence of diffusible hydrogen on the CGHAZ degradation can be expressed by the ratio of lower critical stress to CGHAZ tensile strength. However, it is difficult to measure the CGHAZ tensile strength directly from the implant test, that is because if no hydrogen is introduced into the weld, the failure will occur in the lower strength base metal rather than in the higherstrength CGHAZ. As a result, the maximum CGHAZ hardness is converted into the CGHAZ tensile strength according to the ASTM hardness conversion chart, which are determined to be 212 ksi (1462 MPa), 154 ksi (1062 MPa), and 145 ksi (1000 MPa) for HY-100, HSLA-100, and HSLA-65, respectively. The embrittlement index is determined accordingly to be 0.34, 0.54, and 0.52 for HY-100, HSLA-100, and HSLA-65, respectively. The higher the embrittlement index, the lower the HIC susceptibility, which means the degradation of CGHAZ tensile strength due to diffusible hydrogen is not serious. Note that the CGHAZ tensile strength is not experimentally determined but only an approximation; however, it can still be used as an index to evaluate the steels HIC susceptibility. Based on the above implant test results, both the NCSR and embrittlement index show that HY-100 undergoes the most serious degradation due to the effect of diffusible hydrogen among the three steels, while HSLA-100 and HSLA65 are less susceptible to HIC compared with HY-100. For HSLA-100 and HSLA65, their embrittlement index is almost the same, that is because of their relatively lower carbon and alloy addition (lower hardenability as shown in Table 3, Ref. 10) as well as their finer grain size compared with HY-100. However, it should be noted that NCSR of HSLA-65 (1.17) is higher than that of HSLA-100 (0.83), which means the CGHAZ degradation from base metal yield strength due to the effect of diffusible hydrogen for HSLA-100 is more severe than HSLA-65. Thereby, it indicates HSLA-65 has better resistance to HIC than HSLA-100.
Fracture Behavior

Fig. 10 Fracture morphology of HY-100 implant specimen at a stress level of 80.3 ksi (554 MPa) that failed after 12 min of loading. A General fracture appearance (white arrow indicates the direction of crack growth); B region I (IG); C region II (QC); D region II (MVC).

Fig. 11 Fracture morphology of HSLA-100 implant specimen at a stress level of 102.4 ksi (706 MPa) that failed after 1.5 min of loading. A General fracture appearance (white arrow indicates the direction of crack growth); B region I (IG); C region II (QC); D region II (MVC).

Figure 9AD shows the fracture morphology of the HY-100 implant specimen at a stress level of 91.3 ksi (629 MPa) that failed after 3 min of loading. It shows that the fracture surface can be divided into three regions, which are region I, region

II, and final failure region. Region I is in close vicinity to the thread root, where the highest stress concentration exists. A coarse intergranular (IG) fracture mode is dominant in region I, where the grain size is in the range of 7090 m as shown in Fig. 9B. It can be concluded that cracking

initiates in the location where CGHAZ and thread root coincides, or somewhere closely behind the thread root (Ref. 18), as a result of stress concentration as well as the presence of coarse-grained lath martensite in the CGHAZ. The white arrow in Fig. 9A indicates the crack prop-

WELDING JOURNAL 25-s

WELDING RESEARCH

Fig. 12 Fracture morphology of HSLA-100 implant specimen at a stress level of 85.8 ksi (592 MPa) that failed after 60 min of loading. A General fracture appearance (white arrow indicates the direction of crack growth); B region I (IG); C region II (QC); D region II (MVC).

Fig. 13 Fracture morphology of HSLA-65 implant specimen at a stress level of 77.5 ksi (534 MPa) that failed after 23 min of loading. A General fracture appearance (white arrow indicates the direction of crack growth); B region I; C region II (QC); D region II (MVC).

agation direction. As the crack propagates, region II with different features is shown on the fracture surface. Both quasicleavage (QC) and microvoid coalescence (MVC) can be observed in region II, which are shown in Fig. 9C and D. With

further propagation, overload failure will take place. Note that the boundaries separating the three regions are not distinct, and the division of the fracture surface is based on the fracture morphology. Figure 10AD shows the fracture mor-

phology of an HY-100 implant specimen at a stress level of 80.3 ksi (554 MPa) that failed after 12 min of loading. Similar to the sample shown in Fig. 9, three distinct regions can also be seen on the fracture surface. The fracture mode at crack initiation in region I is essentially intergranular. Again, it is shown that the CGHAZ is the most susceptible to HIC among the HAZ regions. Relative to the previous sample, a small difference in fracture morphology exists in that the area of IG failure increases with decreasing the tensile loading. Both QC and MVC can be observed in region II, as shown in Fig. 10C and D. The fracture morphology of an HSLA100 implant specimen at a stress level of 102.4 ksi (706 MPa) that failed after 1.5 min is shown in Fig. 11AD. Similar to HY-100, the fracture surface can also be divided into three regions as shown in Fig. 11A. Region I with predominant IG fracture can only be observed in a small area close to the thread root, as shown in Fig. 11B. In addition to the clear faceted IG shown on the fracture surface, the prior austenite grain boundary can also be observed on the thread surface, and it is continuous across the boundary separating the fracture surface and thread surface. It shows that cracking initiates in the CGHAZ when a critical amount of hydrogen diffuses to the stress concentration area. The prior austenite grain boundary becomes the weak link under the influence of both hydrogen and stress so that the relative grain boundary sliding occurs in the CGHAZ. That is why the prior austenite grain boundary can be observed on the thread surface. In region II, both QC and MVC fracture modes can be observed as shown in Fig. 11C and D. By decreasing the tensile stress in HSLA-100 to 85.8 ksi (592 MPa), the implant specimen failed after 60 min of loading. The fracture morphology of this sample is shown in Fig. 12AD. It is shown in Fig. 12B that IG fracture can be observed in a small area of region I. Both QC and MVC can be observed in region II, as shown in Fig. 12C and D. The fracture morphology of the HSLA-65 implant specimen is shown in Fig. 13AD. As shown in Fig. 13B, there is some faceted IG fracture with a smaller grain size in region I near the thread root even though it is not so clear as compared to the fracture surface of the HY-100 specimen. This is probably because of the mixture of ferrite, bainite, and martensite in the CGHAZ. In region II adjacent to region I, both QC and MVC can be observed, as shown in Fig. 13C and D. The occurrence of IG, QC, and MVC fracture modes on the fracture surface can be explained using Beachems model (Refs. 19, 20), as shown in Fig. 14. As-

WELDING RESEARCH

26-s JANUARY 2013, VOL. 92

Fig. 14 Combined effect of stress intensity factor and hydrogen concentration at crack tip on the fracture mode.

sume when the implant specimen is subject to loading after welding, the combination of stress intensity factor and hydrogen concentration at the crack tip corresponds to point a in Fig. 14. The hydrogen concentration is not sufficient to initiate a crack, so cracking will not occur immediately. During the incubation period, atomic hydrogen continuously diffuses to the triaxially stressed region, and after some time, it will reach the critical level indicated by point b in Fig. 14. A crack will then be initiated in the CGHAZ and grow intergranularly. As the crack propagates, the stress intensity factor increases while the hydrogen level decreases to point c, promoting a QC fracture mode. As the crack continues to grow, and if the combination of stress intensity factor and hydrogen concentration reaches point d, the fracture mode will change to MVC. If the stress intensity factor continues to increase to the critical value K C, ultimate failure will take place. Microstructure and the fracture behavior taken together can explain the difference in HIC susceptibility of the three steels. As shown from fractography, cracking will always initiate in the CGHAZ, and the intergranular fracture occurs first. For the same welding conditions, the HY-100 CGHAZ microstructure is high-hardness (420 to 440 HV) martensite, while a mixture of bainite and martensite with lower hardness (325 to 340 HV) forms in the HSLA-100 CGHAZ. For HSLA-65, the CGHAZ has the lowest hardness (300 to 317 HV) among the three steels as a result of the

presence of ferrite, bainite, and martensite. It has been shown previously (Ref. 10) that the prior austenite grain size is the largest in HY-100, and the smallest in HSLA-65, with HSLA-100 intermediate. Based on the fracture surface observations from the implant tests, HY-100 has the coarsest IG fracture and the largest area of IG fracture region among the three steels, while both of these features are the smallest for HSLA-65. As a result of different grain size, microstructure and associated hardness of the CGHAZ at the same welding condition, the HIC susceptibility of the three steels is different, which is indicated by the value of NCSR and embrittlement index of the three steels. Therefore, it can be concluded that HY-100 is the most susceptible to HIC among the three steels, while HSLA-65 is the least.

gen content is 8.1 mL/100 g, the lower critical stress (LCS) is 72 ksi (496 MPa), 83 ksi (572 MPa), and 76 ksi (524 MPa) for HY100, HSLA-100, and HSLA-65, respectively. The normalized critical stress ratio (NCSR) is determined accordingly to be 0.72, 0.83, and 1.17 for HY-100, HSLA100, and HSLA-65, respectively. 4. A new embrittlement index is proposed, that is the ratio of the LCS and tensile strength of the CGHAZ, which is approximated by the hardness. Using this approach, the embrittlement index is determined to be 0.34, 0.54, and 0.52 for HY-100, HSLA-100, and HSLA-65, respectively. 5. Based on morphology and fracture mode, the fracture surface of the three steels can be divided into three regions. In region I, the crack will initiate in the CGHAZ and grow intergranularly. Both quasi-cleavage (QC) and microvoid coalescence (MVC) can be observed in region II. Final failure occurs under overload conditions. 6. As the crack initiates and propagates, IG, QC, and MVC fracture mode will occur in sequence. The observation of the three fracture modes on the fracture surface can be explained using Beachems model. 7. Among the three steels, HY-100 has the coarsest IG fracture and the largest area of IG fracture, while both of these are the smallest for HSLA-65. 8. Based on the implant test results, HY-100 is the most susceptible to HIC because of the formation of a high-hardness martensitic microstructure and large prior austenite grain size in the CGHAZ. HSLA-100 is less susceptible as a result of formation of bainite and martensite with lower hardness and smaller grain size. HSLA-65 shows resistance to HIC, resulting from the mixture of ferrite, bainite, and martensite with the lowest hardness and smallest grain size in the CGHAZ. Acknowledgments The authors gratefully acknowledge the financial support of the Office of Naval Research, Award No. N000140811000. Grant Officers: Dr. Julie Christodoulou and Dr. William Mullins. The authors would also like to thank Johnnie DeLoach, Matthew Sinfield, and Jeffrey Farren with the Naval Surface Warfare Center Carderock Division, West Bethesda, Md., for providing the steels used in this study and valuable discussions regarding the weldability of these steels. Dejian Liu and Geoffrey Taber are acknowledged for their constructive ideas and help on building the implant testing system. In addition, Badri Narayanan, John Procario, and Garr Eberle with The Lin-

Conclusions
The results of the present investigation can be summarized as follows: 1. In the present welding condition, the hardness of the CGHAZ is in the range of 420440, 325340, and 300317 HV for HY-100, HSLA-100, and HSLA-65, respectively. 2. Lath martensite with a thin film of retained austenite is observed in the CGHAZ of HY-100. For HSLA-100, lath martensite and bainite form in the CGHAZ. While for HSLA-65, a mixture of ferrite, bainite, and martensite forms in the CGHAZ. 3. When the average diffusible hydro-

WELDING JOURNAL 27-s

WELDING RESEARCH

coln Electric Co. are thanked for providing the Pipeliner111M welding consumable used in the implant test.
References 1. Montemarano, T. W., Sack, B. P., Gudas, J. P., Vassilaros, M. G., and Vanderveldt, H. H. 1986. High strength low alloy steels in naval construction. Journal of Ship Production 2(3): 145162. 2. Czyryca, E. J., Link, R. E., Wong, R. J., Aylor, D. A., Montemarano, T. W., and Gudas, J. P. 1990. Development and certification of HSLA-100 steel for naval ship construction. Naval Engineers Journal 102(3): 6382. 3. DeLoach, J. J., Null, C., Flore, S., and Konkol, P. 1999. The right welding wire could help the U.S. Navy save millions. Welding Journal 78(6): 5558. 4. Devletian, J. H., and Fichtelberg, N. D. 2001. Controlling hydrogen cracking in shipbuilding. Welding Journal 80(11): 46-s to 52-s. 5. Mayudeeswaran, G., Balasubramanian, V., and Reddy, G. M. 2008. Hydrogen induced cold cracking studies on armour grade high strength, quenched and tempered steel weldments. International Journal of Hydrogen Energy 33(7): 18971908. 6. Savage, W. F., Nippes, E. F., and Homma,

H. 1976. Hydrogen induced cracking in HY-80 steel weldments. Welding Journal 55(11): 368-s to 376-s. 7. Adonyi, Y. 2006. Heat-affected zone characterization by physical simulations. Welding Journal 85(10): 4247. 8. Shome, M., and Mohanty, O. N. 2006. Continuous cooling transformation diagrams applicable to the heat-affected zone of HSLA80 and HSLA-100 steels. Metallurgical and Materials Transactions A 37A: 21592169. 9. Nawrocki, J. G., DuPont, J. N., Robino, C. V., and Marder, A. R. 2001. The postweld heat treatment response of simulated coarsegrained heat-affected zones in a new ferritic steel. Metallurgical and Materials Transactions A 32A: 25852594. 10. Yue, X., Lippold, J. C., Alexandrov, B. T., and Babu, S. S. 2012. Continuous cooling transformation behavior in the CGHAZ of naval steels. Welding Journal 91(3): 67-s to 75-s. 11. Granjon, H. 1969. The implants method for studying the weldability of high strength steels. Metal Construction and British Welding Journal 1(11): 509515. 12. Morito, S., Huang, X., Furuhara, T., Maki, T., and Hansen, N. 2006. The morphology and crystallography of lath martensite in alloy steels. Acta Materialia 54: 53235331. 13. Spanos, G., Fonda, R. W., Vandermeer, R. A., and Matuszeski, A. 1995. Microstructural

changes in HSLA-100 steel thermally cycled to simulate the heat-affected zone during welding. Metallurgical and Materials Transactions A 26A: 32773293. 14. Moon, D. W., Fonda, R. W., and Spanos, G. 2000. Microhardness variations in HSLA-100 welds fabricated with new ultralow-carbon weld consumables. Welding Journal 79(10): 278-s to 285-s. 15. Bhadeshia, H. K. D. H. 2001. Bainite in steels: Transformations, Microstructure and Properties. pp: 6388, London, UK, IOM Communications Ltd. 16. Sawhill, J. M. Jr., Dix, A. W., and Savage, W. F. 1974. Modified implant test for studying delayed cracking. Welding Journal 53(12): 554-s to 560-s. 17. Dickinson, D. W., and Ries, G. D. 1979. Implant testing of medium to high strength steel A model for predicting delayed cracking susceptibility. Welding Journal 59(7): 205-s to 211-s. 18. McMahon, C. J. Jr. 2001. Hydrogeninduced intergranular fracture of steels. Engineering Fracture Mechanics 68: 773788. 19. Beachem, C. D. 1972. A new model for hydrogen-assisted cracking (Hydrogen embrittlement). Metallurgical Transactions 3: 437451. 20. Gedeon, S. A., and Eagar, T. W. 1990. Assessing hydrogen-assisted cracking fracture modes in high-strength steel weldments. Welding Journal 69(6): 213-s to 220-s.

WELDING RESEARCH

Authors: Submit Research Papers Online

Peer review of research papers is now managed through an online system using Editorial Manager software. Papers can be submitted into the system directly from the Welding Journal page on the AWS Web site (www.aws.org) by clicking on submit papers. You can also access the new site directly at www.editorialmanager.com/wj/. Follow the instructions to register or log in. This online system streamlines the review process, and makes it easier to submit papers and track their progress. By publishing in the Welding Journal, more than 69,000 members will receive the results of your research. Additionally, your full paper is posted on the American Welding Society Web site for FREE access around the globe. There are no page charges, and articles are published in full color. By far, the most people, at the least cost, will recognize your research when you publish in the world-respected Welding Journal.

Welding Journal Becomes Digital and Mobile

The Welding Journal can now be enjoyed by AWS members for free using their computers or Internet-ready mobile phones or tablets, including iOS (iPad and iPhone), Android, Windows 7, and HP Web OS devices. Presented below are three key areas with details for reading the magazine in various ways. For desktops/laptops, the Welding Journal can be read online, or downloaded for offline reading and archiving. You can also do custom searches within a single issue or throughout all the issues in the archive (dating back to December 2011), and print individual pages or whole issues. Active links add convenience. For mobile devices, the Welding Journal can be used on Internet-capable tablets or phones. A Web app runs in your mobile browser and automatically recognizes the device, optimizing presentation and functionality. It is the best choice for those who prefer reading content online vs. downloading to their mobile device. For iPad and iPhone users, go to iTunes, type in Welding Journal in the search window, and download the app to retrieve the magazine online or download for immediate access. This is the best choice for those who prefer to store issues in their iOS devices for offline reading. In addition, build your Welding Journal catalog with all the available issues in the archive. The digital edition will be sent to you automatically every month, but make sure your e-mail address is up to date by logging in at www.aws.org.

28-s JANUARY 2013, VOL. 92