Effect of Repeated Welding on the Mechanical Properties of the Heat-Affected Zone of Nuclear Structural Materials

Yongjoon Kang; Jinhyeok Bang; Chang-Young Oh; Seung-Gun Lee

doi:10.5781/JWJ.2025.43.3.6

Abstract

This study aimed to evaluate the effects of repeated welding on the heat-affected zone (HAZ) of nuclear structural materials. Carbon steel, alloy steel, and stainless steel, which are widely used in nuclear structural materials, were subjected to up to four repeated welding cycles. Changes in hardness, strength, and impact toughness were observed with each repetition. The results confirmed that for the carbon steel, alloy steel, and stainless steel used in this study, the mechanical properties of the HAZ were maintained even after up to four repeated welds at the same location.

Key words: Repeated welding, Heat-affected zone, Tensile strength, Impact toughness

1. Introduction

Nuclear structural materials are critical to the safety of nuclear power plants; therefore, verifying their mechanical properties is essential. In particular, the heat-affected zone (HAZ) adjacent to the weld fusion line exhibits mechanical characteristics that differ from those of the unaffected base metal¹^,⁵^,⁶⁾, necessitating detailed evaluation. Such evaluations are typically performed through the welding procedure qualification process.

In cases where existing welds are removed and rewelded for purposes such as equipment alignment or maintenance, the HAZ is subjected to repeated thermal cycles. The mechanical properties of the HAZ exposed to multiple thermal cycles can differ from those of the original HAZ, making verification necessary. However, there is a lack of experimental data regarding the effects of repeated welding thermal cycles on the structural integrity of nuclear components, and the theoretical understanding remains unclear. In practice, two to three instances of repeated welding are commonly accepted, but there are no explicit regulations limiting the number of repetitions²^,³⁾.

Lomozik et al.⁴⁾ simulated the HAZ of low-alloy steel and reported that both hardness and impact toughness of the simulated HAZ decreased with an increasing number of thermal cycles. Meanwhile, Gonçalves de Mello et al.²⁾ conducted up to six repeated welding cycles on carbon steel, and Lee et al.³⁾ performed up to five cycles on 2.25Cr-1Mo steel to investigate changes in the mechanical properties of the HAZ, such as strength and impact toughness; these studies found that the effects of repeated welding were either negligible or showed no clear trend. Thus, not only is research on the influence of repeated thermal cycles on HAZ properties limited, but the existing results are also inconsistent.

The present study aims to evaluate the effects of repeated welding thermal cycles on the HAZ of nuclear structural materials. Repeated welding was performed up to four times on carbon steel, alloy steel, and stainless steel-materials commonly used in nuclear applications-and the resulting changes in hardness, strength, and impact toughness were investigated. The results demonstrated that, for the specific carbon steel, alloy steel, and stainless steel used in this study, the mechanical properties of the HAZ were maintained even after up to four repeated welds at the same location.

2. Experimental procedures

One material was selected from each of the ASME P-No. 1, 5A, and 8 groups, which are widely used in nuclear power plants. The chemical compositions and tensile properties of the selected materials, as provided in the manufacturers’ certificates, are summarized in Table 1.

Table 1

Chemical compositions and tensile properties of the steels

P-No.	ASME specification	Tensile properties			Chemical compositions (wt%)
P-No.	ASME specification	YS (MPa)	TS (MPa)	EL (%)	C	Si	Mn	Cr	Ni	Mo
1	SA516-70	341	520	29	0.20	0.33	1.08	-	-	-
5A	SA387-22-2	469	600	28	0.10	0.18	0.48	2.24	-	1.00
8	SA240-316L	257	600	58	0.018	0.54	1.28	16.63	10.06	2.04

As illustrated in Fig. 1, base metals with a thickness of 25 mm were cut into plates measuring 160 mm in width and 150 mm in length. A 45° groove with a depth of 6 mm and a root radius of 3 mm was machined along the longitudinal centerline of each plate. The groove was initially formed by wire electrical discharge machining, and then the surface was finished by grinding to ensure surface quality. Welding was performed using the gas tungsten arc welding (GTAW) process along the machined groove, as shown in Fig. 2. The welding parameters are listed in Table 2. For the P-No. 1 material, for which post-weld heat treatment (PWHT) is not always required depending on joint thickness and other considerations, test coupons were prepared under both PWHT and non-PWHT conditions.

Fig. 1

Schematic showing the dimensions of the base metal for the test coupons

Fig. 2

Pictures showing the process for the test coupons

Table 2

Welding parameters for the test coupons

Condition ID	Base metal	Filler metal	Process	Current (A)	Voltage (V)	Speed (cpm)	PWHT
P1 (No PWHT)	SA516-70 (P-No. 1)	ER70S-6	GTAW	260-280	12-14	12	Not performed
P1 (PWHT)	SA516-70 (P-No. 1)	ER70S-6	GTAW	260-280	12-14	12	610 °C for 1h
P5A	SA387-22-2 (P-No. 5A)	ER90S-B3	GTAW	260-280	12-14	12	610 °C for 1h
P8	SA240-316L (P-No. 8)	ER316L	GTAW	260-280	12-14	12	Not performed

To prepare repeated welding test coupons, the deposited weld bead was removed after the initial weld, and the same groove geometry was remachined at the same location. Welding was then repeated under identical conditions, and this process was carried out up to three times. The overall procedure for preparing the repeated welding test coupons is schematically illustrated in Fig. 3.

Fig. 3

Schematic showing the steps for the test coupons with repeated welding

For mechanical testing, sub-sized plate-type tensile specimens with a thickness of 3 mm and a gauge section width of 6 mm, as well as sub-sized Charpy V-notch specimens (2.5 mm × 10 mm × 55 mm), were prepared in accordance with ASME SA-370. For the Charpy specimens, notches were machined as close as possible to the fusion line to evaluate the toughness of the HAZ.

Hardness distribution across the cross-section of the repeated welding test coupons was measured using a Vickers hardness tester with a 200 g load. In addition, the microstructure was observed using optical microscopy (OM). Specimens for microstructural analysis were mechanically polished and etched with a 3% nital solution.

3. Results and discussion

The tensile test results for the repeated welding test coupons are summarized in Table 3. For the P-No. 1 and P-No. 5A materials, fracture occurred in the base metal, whereas for the P-No. 8 material, fracture occurred in the weld metal. Regardless of the fracture location, all specimens exhibited tensile strengths exceeding the specified minimum tensile strength of the base metal, thus meeting the acceptance criteria for welding procedure qualification. Variations in tensile strength with respect to the number of repeated welds were negligible.

Table 3

Tensile strength in the transverse direction of the test coupons with different numbers of welds

Welding condition	Number of welds	Tensile strength	Failure location
P1 (No PWHT)	1	538	Base metal
	2	540	Base metal
	3	537	Base metal
	4	543	Base metal
P1 (PWHT)	1	518	Base metal
	2	511	Base metal
	3	508	Base metal
	4	507	Base metal
P5A	1	566	Base metal
	2	559	Base metal
	3	554	Base metal
	4	553	Base metal
P8	1	584	Weld metal
	2	581	Weld metal
	3	584	Weld metal
	4	595	Weld metal

The Charpy impact test results for the HAZ of the repeated welding test coupons are presented in Fig. 4. The austenitic P-No. 8 material, for which impact testing is not required, was excluded from the tests. No clear trend was observed in the variation of HAZ impact energy with increasing number of welds. Although some values were lower than those from the initial weld, they remained higher than those of the base metal in all cases. It is well known that the HAZ consists of sub-regions with significantly different microstructures and mechanical properties depending on the peak temperature of the thermal cycle. Therefore, even slight variations in notch location within the HAZ can considerably affect the test results, likely contributing to the observed variation and lack of a consistent trend in impact energy.

Fig. 4

Impact toughness of the test coupons with different numbers of welds

As illustrated in Fig. 5, hardness profiles were obtained across the fusion boundary of each test coupon at 0.2 mm intervals, and the results are presented in Fig. 6. For ferritic steels, solid-state phase transformations typically occur in both the HAZ and the weld metal. Due to the rapid cooling associated with the welding process, low-temperature transformation products may form. Accordingly, the hardness of the HAZ and weld metal in the P-No. 1 and P-No. 5A materials was higher than that of the base metal. After PWHT, the hardness difference was reduced due to tempering effects. In the case of the P-No. 5A material, softening was observed in the HAZ at a distance from the fusion boundary; however, as indicated by the transverse tensile test results in Table 3, the extent of softening was not considered significant. Additionally, no appreciable change in hardness was observed with increasing numbers of welding repetitions. For the P-No. 8 material, in which no solid-state phase transformations occur in the HAZ or weld metal, the hardness remained similar to that of the base metal and showed no variation with repeated welding.

Fig. 5

Optical micrographs of the test coupons (number of welds: 1), (a) P1 (No PWHT), (b) P1 (PWHT), (c) P5A, and (d) P8

Fig. 6

Hardness profile of the test coupons with different numbers of welds, (a) P1 (No PWHT), (b) P1 (PWHT), (c) P5A, and (d) P8

Microstructures of the base metal and HAZ were examined on the cross-sections of each test coupon. No significant differences were observed with increasing number of weld repetitions; therefore, representative micrographs from the specimens welded four times are shown in Figs. 7-11. For ferritic steels, the HAZ is typically divided into sub-regions according to the peak thermal cycle temperature: coarse-grained HAZ (CGHAZ), fine-grained HAZ (FGHAZ), intercritical HAZ (ICHAZ), and subcritical HAZ (SCHAZ)⁵^,⁶⁾. As shown in Figs. 7-9, the ICHAZ, FGHAZ, and CGHAZ of the P-No. 1 and P-No. 5A materials exhibited significantly different microstructures from the base metal, but no notable changes were observed with repeated welding.

Fig. 7

Optical micrographs of the base metal and HAZ of the P1 (no PWHT) (number of welds, 3), (a) base metal, (b) SCHAZ/ICHAZ, (c) FGHAZ, and (d) CGHAZ

Fig. 8

Optical micrographs of the base metal and HAZ of the P1 (PWHT) (number of welds: 3), (a) base metal, (b) SCHAZ/ICHAZ, (c) FGHAZ, and (d) CGHAZ

Fig. 9

Optical micrographs of the base metal and HAZ of the P5A (number of welds: 3), (a) base metal, (b) SCHAZ/ICHAZ, (c) FGHAZ, and (d) CGHAZ

Fig. 10

Optical micrographs of the base metal and HAZ of the P8 (number of welds: 3), (a) base metal and (b) HAZ/weld metal

Fig. 11

SEM micrograph of the HAZ of the P8 (number of welds: 3)

As shown in Fig. 10, the base metal of the P-No. 8 material consisted of an austenitic matrix with a small amount of retained delta ferrite. Since no solid-state phase transformations occur upon welding, no microstructural differences were observed near the fusion line. With increased welding repetitions, prolonged exposure to high temperatures can potentially lead to sensitization-precipitation of Cr₂₃C₆ carbides and formation of chromium-depleted zones near grain boundaries which may reduce intergranular corrosion resistance⁷⁾. However, as shown in Fig. 11, no Cr₂₃C₆ carbides were observed at the grain boundaries of the HAZ even after four repeated welds. Thus, it can be concluded that repeated welding, under the conditions of this study, did not cause sensitization or degrade the intergranular corrosion resistance of the P-No. 8 material.

4. Conclusions

This study evaluated the changes in mechanical properties of the HAZ in nuclear structural materials with increasing numbers of repeated welds. Carbon steel, alloy steel, and stainless steel-commonly used in nuclear applications-were subjected to up to four repeated welds at the same location. The results demonstrated that the mechanical properties of the HAZ were maintained even after four repetitions. Although slight variations in impact toughness were observed, the values consistently remained higher than those of the base metal. Changes in hardness and tensile strength were negligible, indicating that repeated welding, within the tested range, does not significantly degrade the mechanical integrity of the HAZ. In addition, no noticeable microstructural changes were observed in the HAZ with increasing welding repetitions, and no precipitation of Cr₂₃C₆ carbides-known to cause sensitization-was found in the HAZ of the stainless steel.