Fusion Engineering and Design 133 (2018) 32–38

Contents lists available at ScienceDirect

Fusion Engineering and Design

journal homepage: www.elsevier.com/locate/fusengdes

Effect of preheating on welding cold crack sensitivity of China low T

activation martensitic steel
⁎ ⁎
Yutao Zhai, Bo Huang , Zhenyu Zhang, Xiaodong Mao, Zhumin Zhao
Key Laboratory of Neutronics and Radiation Safety, Institute of Nuclear Energy Safety Technology, Chinese Academy of Sciences, Hefei, Anhui, 230031, China

A R T I C LE I N FO A B S T R A C T

Keywords: The China low activation martensitic (CLAM) steel has been chosen as the primary structure material in Chinese
CLAM steel series PbLi blankets for fusion reactors. The assembly preparation of the blanket would be seriously influenced
Cracking sensitivity by the weld performance of the CLAM steel. In this study, the welding cold crack sensitivity index of the CLAM
Maximum hardness test steel was analyzed, then the maximum hardness in weld heat-affected zone (HAZ) and y-groove tests were
Y-groove test
performed. Based on the analysis to the maximum hardness in weld HAZ and the cracking rate of y-groove test at
different preheating conditions, the results indicated that the HAZ maximum hardness of CLAM increased at
firstly and then decreased with the increase of the preheating temperature (PT). The cracking rate of y-groove
test reduced with the increase of the PT, and the crack rate maintained at a relative stable stage when the PT
exceeded 250 °C. It revealed high welding crack tendency of the CLAM steel. Thus, the preheating before welding
should be applied, and a PT at about 250 °C was proposed in order to prevent welding cracking.

1. Introduction
China low activation martensitic (CLAM) steel is one of three pri-
mary kinds of reduced Activation Ferritic/Martensitic (RAFM) steel,
which is developed in the Institute of Nuclear Energy Safety Technology
(INEST), Chinese Academy of Sciences for about fifteen years [1–3]. It
has been chosen as the primary structure material of China series
blanket for fusion reactors [4,5], CN helium cooled ceramic breeder
(HCCB) test blanket module (TBM) for ITER and the blanket of China
fusion engineering test reactor (CFETR) [6–8].
Due to the structural complexity of the blanket, it is difficult to
fabricate the blanket by casting and one-off molding, and welding is
considered as a flexible way to assemble the blanket [9–13]. However,
the fabrication would be seriously influenced by the weld performance
of the CLAM steel because the martensitic steel is sensitive to cracking
in welding for its obvious hardenability [14]. Thus, it is crucial to in-
vestigate the weldability of CLAM steels as it will be applied in the
blanket fabrication by welding [15–18].
The cold cracking tests for weldments are designed to assess the cold
cracking sensitivity of the parent materials. And the cold cracking
sensitivity is one of most important contents to evaluate the procedure
weldability of the structure materials. The study of the procedure
weldability has important value to the development of the welding
process. As mentioned above, the CLAM steel has been chosen as the
primary structure material of China series fusion blanket. To ensure the
welding precedure of the CLAM steel available and reliable, it should be
qualified by the nuclear safety regulator before the engineering appli-
cations. Therefore, the cold cracking sensitivity of the CLAM steel, as
one of the most important contents of the weldability, it must be clar-
ified before the qualification and engineering applications.
In this paper, the study was focused on the welding cold crack
sensitivity of CLAM steel. On the one hand, the welding cold crack
sensitivity index of the CLAM steel was analyzed based on empirical
formula [19]. On the other hand, the cracking tendency was in-
vestigated by maximum hardness in weld heat-affected zone (HAZ) test.
Meanwhile, the critical preheating temperature (PT) to improve the
cracking tendency of CLAM was obtained by calculating the crack rate
in the y-groove test at different PTs [19–21]. Based on the study on the
weld crack sensitivity of CLAM, it will provide essential guidance and
reference on the welding procedure establishing of the CLAM steel.
2. Experiment
2.1. Materials
The material used in this research is 16 mm HEAT 1105 CLAM steel
plate, which was normalized at 980 °C/30 min, air cooling and tem-
pered at 760 °C/90 min, air cooling [14,22]. Its chemical composition
was listed in Table 1. The welding was performed by tungsten inert gas
(TIG), and the filler wires were cut from the 16 mm thick CLAM plate.

⁎
Corresponding author.
E-mail addresses: bo.huang@fds.org.cn (B. Huang), zhumin.zhao@fds.org.cn (Z. Zhao).

https://doi.org/10.1016/j.fusengdes.2018.05.071
Received 31 October 2017; Received in revised form 16 February 2018; Accepted 21 May 2018
0920-3796/ © 2018 Elsevier B.V. All rights reserved.
Y. Zhai et al. Fusion Engineering and Design 133 (2018) 32–38

Table 1 Table 3
Chemical compositions of CLAM (HEAT 1105) in wt.%. PTs of maximum hardness in weld HAZ tests.
Cr C W V Ta Mn Si Ti No. 1 2 3 4 5
8.73 0.08 1.34 0.21 0.15 0.43 0.14 0.01
PT (°C) RT 100 200 250 300
O N S Al Cu Ni Nb Fe
0.004 0.004 0.004 0.02 0.01 0.01 0.01 Bal.

in the test line with the interval of 0.5 mm and load of 10 kg. The tests
2.2. Weld crack sensitivity index were performed at RT according to ref. [24].

Hardenability and crack tendency in weld HAZ is closely related to
the chemical composition of steels. Therefore, the crack sensitivity
could be evaluated by chemical composition. Based on the results of
extensive experiments, Doctor Ito proposed the cold crack sensitivity
index [19], which was related to chemical composition, thickness and
diffusion hydrogen, as shown in formulas (1)-(3). The PT could be
calculated according to the crack sensitivity index, and the formulas
could be used in the high-strength low alloy steels with the tensile
strength of 400–1000 MPa.
Si Mn + Cu + Cr Ni Mo V the principal rolling direction of the plate can be determined, arrange
Pcm = C + + + + + + 5B
30 20 60 15 10
Pc = Pcm + [H]/60 + t/600
T0 = 1440Pc − 392°C
In the formulas, Pcm – carbon equivalent of low carbon steel; t –
plate thickness, [H] – diffusion hydrogen in weld metal according to
Japanese standard JIS Z3118 (ml/100 g) [23], T0 - lowest preheating
temperature before welding, °C.
According to the chemical compositions of CLAM in Table 1, the Pcm
of CLAM is about 0.59, and Pc is 0.61, while diffusion hydrogen in the
weld wires is usually considered as [H] ≤ 0.02 ml/g [23]. Then, the
critical PT (T0) is calculated as 486 °C. It was shown that the hard-
enability and crack tendency of CLAM steel is obvious, and it has poor
resistance to crack in welding.
2.3. Maximum hardness in weld HAZ test
Maximum hardness in weld HAZ test was performed according to
the standard ISO 6507 [24]. The standard specification test plates were
prepared by quenched and tempered CLAM plate retained the rolling
surface as shown in Table 2. The 1# plate was welded at room tem-
perature (RT), and the 2#–5# plates were welded in different PTs, as
seen in Table 3.
The filler material was CLAM wires with diameter of 2*2 mm, and it
was cut from the CLAM steel plate (the same to the test plate) with the
method of wire electric discharge (WED). Before the welding, it was
dried at 75 °C for 2 h. The welding was performed on the rolling surface
by manual at the environment of 25 °C × 39% relative humidity (RH).
The post weld plates were cooled naturally in the air without any other
heat treatment. The hardness tests were performed on the samples,
which were machined from the plates after 12 h of placement, as shown
in Fig. 1.
The inspection surface of the samples was etched with 20% FeCl3 for
30 s after grinding. The hardness tests were conducted as Fig. 2 shown.
The test line was parallel to the rolling surface and tangent to the
bottom of the fuse line. The hardness test points were located on the
tangent point and more than 7 points on each side of the tangent point
2.4. Y-groove cracking test
The y-groove cracking test relates to more stringent test conditions
and is recommended for parent material testing. The dimensions of the
test pieces shall be in accordance with Fig. 3 based on the standard of
ISO 17642-2 [21].
Key: a-test welding, b-anchor weld, t-thickness of the test plate, g-
root gap 2.0 mm ± 0.2 mm.
The test pieces were machined by sawing, milling and grinding.
Ensure that surfaces to be welded are milled or ground finish. Where
(1) the rolling directions of the plates to be the same and parallel to the
welding direction. Ensure that the surfaces to be welded are ground
(2)
smooth and free from scale, rust, oil, grease and other contaminants.
(3) The anchor welds (shown in Fig. 3 b-area) with a welding con-
sumable with a yield strength equal to or greater than the yield strength
of the material under test. Deposit the welds with a procedure to avoid
hydrogen cracking, using preheat, interpass and post-heating control.
Dry all consumables used for anchor welds in accordance with manu-
facturers’ recommendations to give the lowest possible hydrogen levels.
The filler material was also CLAM wires with diameter of 2*2 mm,
and dried at 75 °C for 2 h, and the welding was performed by manual at
the environment of 25 °C × 39% RH. The tests were executed on 12 test
plates for 6 PTs, as shown in Table 4. In order to reduce the accidental
and statistical error, 2 prototypes were prepared and the results of the
crack rate at one condition were averaged.
The welding temperature in the test section was strictly controlled
in ± 3 °C of preheating temperature. The welding was started when the
temperature of the whole test plate has reached the specified preheating
temperature. And the welding arc started and ended outside the test
section. The test bead was welded with the method of tungsten inert gas
(TIG) welding with argon (Ar) protection and the welding process was
listed in Table 5 [25–28].
After 48 h of placement, samples were prepared for crack observa-
tion in surface, cross section and root of the weld joints, and the posi-
tions of the 5 sectioning cuts at the positions nearest to the starting
point where the width of bead becomes constant and the centre of the
bead crater, and the positions quartering the distance there of as shown
in Fig. 4.
Key: 1-Cutting in the width direction, 2-Examining section position
The length and height of the cracks were measured according to
Fig. 5. The crack length was measured following the strait line even the
crack was curve, as Fig. 5(a) shown. When the cracks overlapped, the
length was measured as one crack.
1) The surface crack rate was calculated by the formula (4):
∑ lf
Cf = × 100%
L (4)

Table 2 2) In the formula, Cf is the surface crack rate, %; ∑lf is the total length
Sizes of testing plate for maximum hardness in weld HAZ tests. of the surface crack, mm; L is the weld length, mm.
Test plate number L (Length), mm B (Width), mm I (Seam length), mm 3) Test pieces were snapped or bent to fracture after dyeing with ap-
propriate methods, then the root crack could be measured according
1# 200 75 125 ± 10
to Fig. 5(b), and calculated by formula (5):
2# ∼ 5# 200 150 125 ± 10

33
Y. Zhai et al. Fusion Engineering and Design 133 (2018) 32–38

Fig. 1. Specimen of maximum hardness in weld HAZ (a) test pieces, (b) weld joint.

Table 4
PTs for the y-groove cracking test.
No. 1# 2# 3# 4# 5# 6#

Temperature (°C) RT 100 200 250 300 350

Table 5
Welding process for the y-groove cracking test.
Welding Welding Voltage, V Welding speed, Ar flow, L/
method current, A mm/min min

TIG 260 14–16 60–80 16

Fig. 2. The position of the hardness test.

∑ lr
Cr = × 100%
L (5)

In the formula: Cr is the surface crack rate, %; L is the weld length,

mm; ∑lr is the total length of the root crack, mm.

1) The crack height could be measured according to Fig. 5(c), and the
cracking rate at the five cross sections was calculated respectively
according to formula (6). Then, the average value was given.

Hc
Cs = × 100%
H (6)

In the formula: Cs is the section crack rate, %; H is the minimum

Fig. 4. Welding method of test bead and crack examining position.
thickness of the weld, mm; Hc is the height of the section crack, mm.
The five section surfaces were determined by the four equal distances of
the weld length. 3. Results and discussion

3.1. Maximum hardness of HAZ

The maximum hardness test results of HAZ were shown in Fig. 6, the

Fig. 3. Dimensions (mm) of the y-groove test pieces.

34
Y. Zhai et al.
Fig. 5. Crack measure (a) surface crack, (b) root crack and (c) section crack.
Fig. 6. The maximum hardness at different PTs.
value of maximum hardness was 402 HV10 at the RT status, which was
higher than the critical hardness value 350 HV10 of cracking tendency
of steel welding by international institute of welding (IIW) re-
commended [24]. The maximum hardness was increased with the PTs
increasing, and the PT reached at 250 °C the maximum hardness was
reached maximum value of 412 HV10, then decreased to 394 HV10
when the PT continuously increasing to 300 °C.
The reasons of the HAZ maximum hardness tendency might be ex-
plained as follows, the function of preheating before welding was equal
to slow down the cooling rate, and the hardness was contributed by the
microstructures and grain sizes. According to the ref. [22,29], the
martensite starting temperature (Ms) was 452 °C. When the PT below
the 250 °C, the cooling rate was fast and the martensitic was more re-
mained. Then the hardness was mainly contributed by the grain coar-
sening and the martensitic. While the PT was higher than 250 °C, which
was close to the martensite starting temperature (Ms), the martensitic
transformation rate was fast and the number of martensite decreased
and it resulted in the decrease in the hardness.
3.2. Y-groove cracking test
3.2.1. Crack ratio
The length of cracks at the surface, cross section and the root of the
weld joints were measured and the crack rate was calculated, the results
were shown in Fig. 7. There was no surface crack in all conditions and
the crack ratio was mainly contributed by the large root cracks. The
large cracks in the root of the weld joints were the main contributions
to the crack ratio, and the micro-cracks were mostly distributed in the
middle of the cross section of the weld joints. So, the crack ratio was the
average of the root crack ratio and section crack ratio.
As Fig. 7 showed, the crack ratio decreased gradually with the in-
creasing of the PT, in particular, the PTs were over 250 °C, there was no
large cracks observed. The average crack rate fluctuated in a small
range of 0.38% when the PTs were over 250 °C, it might be the reasons
the y-groove test relates to the more stringent test conditions than the
normal welding conditions and the welding method was TIG by
manual, then the protection of the Ar might be insufficient during the
movement of the torch. The micro-cracks were easy to crack under the
35
Fusion Engineering and Design 133 (2018) 32–38
Fig. 7. The Crack rate at different PTs.
stringent test conditions and the welding speed by manual hardly to be
controlled at a constant speed. The 0.38% fluctuation of the crack rate
could be considered as a stable condition. Therefore, the CLAM steel
should be preheated to ∼250 °C at least in order to avoid cracking in
the TIG welding.
3.2.2. Cracks analysis
The cracks generated at different preheating conditions, it showed
that the large cracks generated at the bottom and extended to the center
of the weld joints as shown in Fig. 8, it might be resulted by the dif-
ferent cooling rates between the center and the edge of the weld pool.
The cooling rate was faster at the edge than in the center of the pool.
Therefore, the weld stress in the center of the pool was larger than at
the edge and it resulted in the wider cracks at the center than at the
bottom of the welding. Additionally, the end of the cracks were stopped
by some defects, such as voids, impurities and so on, as shown in Fig. 9.
Beside the cracks, the voids also existed in the center of the weld
joints (Fig. 10(a)) and the interface of the melted and the HAZ of the
base metal (BM) (Fig. 10(b)). The reasons might be that the impurities
or dust existed in the pool and the interface of the weld metal (WM) and
BM. The voids generated in the center of the weld joints was with ir-
regular edge, the reason might be that the molten pool was cooled
slowly and with complex heat transfer process. The cooling rate near
the base metal was fast and a large temperature gradient existed near
the interface of the molten pool and base metal, therefore, higher stress
was generated at the interface than in the center of the weld pool,
which induced and promoted cracking.
The shape of the large cracks and micro-cracks was obviously dif-
ferent as shown in Fig. 11, while the large cracks were with jagged
edges (Fig. 11(a)) and the micro ones were with straight edges
(Fig. 11(b)). The reasons might be that the large cracks usually gener-
ated in the center of the weld pool, and where the cooling rate was
lower than other places, the periphery of the cracks were frozen firstly,
the high stress, impurity accumulation and material loss in the center of
the weld pool which led to the jagged cracks generation. Moreover, the
micro-cracks usually generated along the large grain boundary, the
Y. Zhai et al. Fusion Engineering and Design 133 (2018) 32–38

Fig. 8. Position of the large crack at the welded joints (a) root, (b) cross section.

Fig. 9. Positions of the small crack at the welded joints (a) micro-crack, (b) crack ends.

Fig. 10. Void defects of the weld joints (a) weld metal zone, (b) nearby the HAZ.

Fig. 11. Shapes of the large crack (a) and the micro- crack (b).

36
Y. Zhai et al. Fusion Engineering and Design 133 (2018) 32–38

Fig. 12. SEM morphology of the crack section.

Fig. 13. Micro-pores structure at the surface of the cluster.

Fig. 14. Chemical composition analysis in the crack.

small impurity or contaminants accumulated in grain boundary might
be the crack source.
In order to check the section of the cracks, the prototype was brittle
broken along the weld cracks and the morphology of the crack section
was shown in Fig. 12. The fracture of the crack section was pebble-
shaped bulge and with smooth surface, it was probably induced by the
surface tension and viscous force of the peripheral materials.
It was noticed that there were some cluster found in the cracks, the
microstructures were shown in Fig. 13, which was pictured by SEM. It
could be seen that there was some micro-pores structure at the surface
of the cluster. It indicated that during the solidification, parts of ma-
terials might migrate away and the welding pool was poor protected by
the welding atmosphere.
The chemical compositions of the cluster in the crack were analyzed
by energy dispersive spectrometer (EDS) line scanning of SEM, as
Fig. 14 shown. The content of the Fe element was decreased and the Cr
was increased, while the change of other elements were not obvious. It
might be affected by the atom migration energy and the atom size of the
elements. The EDS result showed that the cluster was Cr-rich particles.
4. Conclusions
The welding cold crack sensitivity of CLAM steel was investigated
with the methods of maximum hardness in weld HAZ test and Y-groove
cracking test, the results has been concluded as follows:
1) The carbon equivalent of CLAM steel showed high welding crack
sensitivity of CLAM, it implicated that cracking would easily occur
during the CLAM steel welding.
2) The results of the maximum hardness in weld HAZ test and the y-

37
Y. Zhai et al.
groove cracking test showed that the critical PT to prevent cracking
in welding would be 250 °C. It suggested that in order to decrease
the tendency of welding cracking, the CLAM steel should be pre-
heated over 250 °C at least before the welding.
3) The crack rates was high before the 200 °C PT (over 2%) and the
main factor of the contribution to it was the large cracks at the root
of the weld joints. When the PT was reached to 250 °C and even
more higher, the large crack was disappeared.
4) The main defects in the weld joints were cracks. And the small in-
clusions, impurity or contaminants might be the crack source, but
they were also the reasons for the termination of the cracks.
The welding cold crack sensitivity was not only closely related to
the chemical composition of the materials, but also the welding process.
Based on the study of this manuscript, it was helpful for understanding
the weldability of CLAM steel, optimizing the welding procedure and
improving the mechanical properties of the weld joints.
Acknowledgments
This work was supported by the CAS Pioneer Hundred Talents
Program and funded with the National Magnetic Confinement Fusion
Science Program of China with Grant No. 2014GB112003, and the
International Science & Technology Cooperation Program of China with
Grant No. 2015DFG62120, the National Natural Science Foundation of
China with Grant No. 11632001, the Provincial Natural Science
Foundation of Anhui with Grant Nos. 1704A0902021 and
1608085ME98, and the Open Research Funds for the Jiangsu Key
Laboratory of Nuclear Energy Equipment Materials Engineering with
Grant No. 1006-56XCA17012. The authors are great appreciation to
other members in FDS Team for their support and contribution to this
work.
References
[1] Q. Huang, F.D.S. Team, Development status of CLAM steel for fusion application, J.
Nucl. Mater. 455 (2014) 649–654.
[2] Q. Huang, Status and improvement of CLAM for nuclear application, Nucl. Fusion
57 (2017) 086042(9pp).
[3] Q. Huang, J. Li, Y. Chen, Study of irradiation effects in China low activation mar-
tensitic steel CLAM, J. Nucl. Mater. 329–333 (2004) 268–272.
[4] Y. Wu, F.D.S. Team, Design status and development strategy of China liquid li-
thium-lead blankets and related material technology, J. Nucl. Mater. 367-370
(2007) 1410–1415.
[5] L. Tan, Y. Katoh, A.-A.F. Tavassoli, et al., Recent status and improvement of re-
duced-activation ferritic-martensitic steel for high-temperature service, J. Nucl.
Mater. 479 (2016) 515–523.
[6] Y. Liu, Overview of fusion nuclear technology activities at SWIP, Fusion Eng. Des.
38
Fusion Engineering and Design 133 (2018) 32–38
85 (2002) 2125–2131.
[7] Y. Wu, F.D.S. Team, Conceptual design and testing strategy of a dual functional
lithium lead test blanket module in ITER and EAST, Nucl. Fusion 47 (2007)
1533–1539.
[8] S. Liu, X. Ma, K. Jiang, et al., Conceptual design of the water cooled ceramic breeder
blanket for CFETR based on pressurized water cooled reactor technology, Fusion
Eng. Des. 124 (2017) 865–870.
[9] E. Rigal, G.D. Dinechin, G. Rampal, et al., Manufacturing of a HCLL cooling plate
mockup, Fusion Eng. Des. 83 (2008) 1268–1272.
[10] X. Chen, Y. Huang, B. Madigan, et al., An overview of the welding technologies of
CLAM steels for fusion application, Fusion Eng. Des. 87 (9) (2012) 1639–1646.
[11] X. Chen, Y. Huang, Z. Shen, et al., Effect of thermal cycle on microstructure and
mechanical properties of CLAM steel weld CGHAZ, Sci. Technol. Weld. Joining 18
(4) (2013) 272–278.
[12] X. Chen, Y. Fang, S. Zhang, et al., Effects of LSP on microstructures and residual
stresses in a 4 mm CLAM steel weld joints, Fusion Eng. Des. 94 (2015) 54–60.
[13] A. Cardella, E. Rigal, L. Bedel, et al., The manufacturing technologies of the eur-
opean breeding blankets, J. Nucl. Mater. 133 (2004) 329–333.
[14] Q. Wu, S. Zheng, S. Liu, et al., Effect of post-weld heat treatment on the mechanical
properties of electron beam as-weld joint for CLAM steel, J. Nucl. Mater. 442 (2013)
512–517.
[15] P. Aubert, F. Tavassoli, M. Rieth, et al., Review of candidate welding processes of
RAFM steels for ITER test blanket modules and DEMO, J. Nucl. Mater. 417 (2011)
43–50.
[16] Q. Huang, Q. Wu, C. Li, et al., Progress in development of fabrication of small TBMs
for EAST and ITER, Fusion Eng. Des. 85 (2010) 2192–2195.
[17] C. Li, B. Huang, J. Zhang, et al., Fabrication techniques development of dual
functional lithium lead test blanket module, Fusion Sci. Technol. 66 (2014)
180–186.
[18] Q. Huang, C. Li, Q. Wu, et al., Progress in development of CLAM steel and fabri-
cation of small TBM in China, J. Nucl. Mater. 417 (2011) 85–88.
[19] Y. Li, J. Wang, The Test and Analysis Methods of Metal Weldability, Chemical
Industry Press, 2014.
[20] British Standard, EN 1043 Destructive Test on Welds in Metallic Materials-Hardness
Testing, (1996).
[21] International Standard, ISO 17642-2 Destructive Tests on Welds in Metallic
Materials-Cold Cracking Tests for Weldments-Arc Welding Processes, International
Standardization Organization, 2005.
[22] S. Zheng, Q. Wu, Q. Huang, et al., Influence of different cooling rates on the mi-
crostructure of the HAZ and welding CCT diagram of CLAM steel, Fusion Eng. Des.
86 (2011) 2616–2619.
[23] Japanese Standard, JIS Z3118 Method for Measurement of Amount of Hydrogen
Evolved from Steel Welds, Technical Committee on Welding of Japan, 2007.
[24] International Standard, ISO 6507 Metallic Materials-Vickers Hardness Test,
International Standardization Organization, 2005.
[25] Jay J. Vora, Vishvesh J. Badheka, Experimental investigation on microstructure and
mechanical properties of activated TIG welded reduced activation ferritic/mar-
tensitic steel joints, J. Manuf. Processes 25 (2017) 85–93.
[26] S. Sam, C.R. Das, V. Ramasubbu, et al., Delta ferrite in the weld metal of reduced
activation ferritic martensitic steel, J. Nucl. Mater. 455 (2014) 343–348.
[27] J. Vanaja, K. Laha, S. Sam, et al., Influence of strain rate and temperature on tensile
properties and flow behavior of a reduced activation ferritic-martensitic steel, J.
Nucl. Mater. 424 (2012) 116–122.
[28] Jay J. Vora, Vishvesh J. Badheka, Improved penetration with the use of oxide fluxes
in activated TIG welding of low activation ferritic/martensitic steel, Trans. Indian
Inst. Met. 69 (2016) 1755–1764.
[29] Y. Huang, X. Chen, Z. Shen, et al., Measurement and analysis of SHCCT diagram for
CLAM steel, J. Nucl. Mater. 432 (s1-3) (2013) 460–465.