Advertisement

Welding in the World

, Volume 62, Issue 2, pp 339–350 | Cite as

Influence of microalloy design on heat-affected zone toughness of S690QL steels

  • Lei Zhang
  • Thomas Kannengiesser
Research Paper
  • 146 Downloads

Abstract

Three high-strength Nb-, Ti- and Ti + V-bearing S690QL steels were welded to investigate and compare the effects of microalloy addition on heat-affected zone (HAZ) toughness. Charpy V notch impact tests from three microalloyed welds under different cooling rates have been performed. Fractographic examination shows that several factors, including large-sized grain, upper bainite or hard second phase, interact to determine brittle fracture and impaired toughness in Nb-bearing weld with high heat input. In contrast to this reduced toughness, Ti-bearing welds exhibits satisfied toughness regardless of at fast or slow cooling. This is attributed to its limited austenite grain and refined favourable intragranular acicular ferrite structure. Moreover, in the case of such refined structure as matrix, TiN particles are found to be irrelevant to the fracture process. The crystallographic results also confirm that high-angle boundaries between fine ferrites plates provide effective barriers for crack propagation and contribute to improved toughness.

Keywords

High-strength steels Microalloyed steels Toughness Cooling rate Microstructure 

1 Introduction

The microalloy concept with Nb, Ti and V elements has been widely applied during high-strength steels development due to their remarked improvement of mechanical properties even by a small addition [1]. Hence, the understanding of microalloy elements on influencing the steel performance is always of great research interest. One key issue about their influence on steel weldability is of importance when considering the fabrication process and safe quality of welded structure. In particular, the effect of microalloy elements during phase transformation and on the subsequent HAZ toughness properties has received much interest.

It is reported that several factors affecting HAZ toughness in microalloyed steel have to be considered: the solubility of microalloy elements, the precipitates characteristics during welding, austenite grain growth, effect on phase transformation and the precipitation hardening [2]. As regards the solubility of microalloy precipitates, TiN has the lowest solubility and could keep stable even at high temperature above 1400 °C [1]. During weld thermal cycle, the precipitates in general will be dissolved to a certain degree depending on the peak temperature and dwell time, with a consequent matrix enrichment of solute elements. The most possible dissolved precipitates in the coarse grained HAZ are those with lower dissolution temperature comparatively with peak temperature, like AlN, VN, Nb(C,N) and V(C,N) [1]. This indicates that sufficient fine microalloy precipitates that can provide effective pinning effect during austenite coarsening is closely linked with the particle characteristics and peak temperature. Another beneficial effect with the presence of Ti or V rich particles is to facilitate formation of intragranular acicular ferrite [3]. However, on the other hand, one major concern is to avoid the tendency of coarse TiN acting as potent cleavage initiator. Although this negative effect is often reported to strongly depend on a complex interplay of the matrix microstructure, alloy composition, grain size and TiN particle size, one deeper understanding on this is still essential for high-strength steels [4, 5, 6].

Some controversial ideas exist towards Nb contribution to microstructural evolution and HAZ property. Some of the opinions are that coarse grain size, bainite or martensite-austenite (M–A) constituent are the major cause of low toughness in Nb treated steels and HAZ [7]. Hattingh and Pienaar have claimed that the adverse influence of Nb on HAZ toughness at higher C level (0.19%) and high heat inputs but Nb addition could also play a beneficial effect on toughness, depending on the C level and cooling rates [8]. Similarly, in V-bearing steel at high heat input, the suppression of M–A constituent will diminish, causing reduced toughness. Hence, the role of Nb and V on determining HAZ toughness is conditionally related with heat input and generally high heat input is not suggested.

The purpose of this study is to investigate and compare, by means of Charpy V notch impact test, the effects of different microalloy addition (Nb, Ti, Ti + V) in the same strength grade S690QL steels on HAZ toughness. Different cooling rates will be applied to three steels in order to investigate the influence of microalloy on phase transformation and HAZ toughness. High-resolution scanning electron microscopy (SEM) combined with electron backscatter diffraction (EBSD) are used to observe and analyse the microstructure, crystallographic information and HAZ toughness properties.

2 Experimental

The materials investigated in this study are Nb-, Ti- and Ti + V-microalloyed S690QL steels, which have yield strength of minimal 690 MPa. The as-received steel plates were in thickness of 6, 6.5 and 8 mm, respectively. Their chemical compositions are given in Table 1. The same welding wire with yield strength of around 720 MPa was selected to obtain a matched weld metal (WM), and its chemical composition is present in Table 2.
Table 1

Chemical composition of three base materials

Materials

Chemical composition (wt%)

C

Mn

Si

Cu

Cr

Ni

Mo

Al

Ti

Nb

N

V

Fe

Steel A

0.14

1.20

0.29

0.04

0.31

0.04

0.21

0.100

0.03

0.006

Bal.

Steel B

0.15

1.29

0.40

0.03

0.32

0.02

0.035

0.02

0.010

Bal.

Steel C

0.14

1.15

0.31

0.01

0.30

0.06

0.17

0.038

0.01

0.0064

0.01

Bal.

Table 2

Chemical composition of filler material

 

C

Mn

Si

Cu

Cr

Ni

Mo

Al

Ti

Zr

V

P

S

Wt%

0.09

1.71

0.59

0.02

0.22

1.49

0.49

0.008

0.045

0.006

0.004

0.005

0.011

All steel plates were welded using fully automated gas metal arc welding (GMAW) system. Modified spray arc welding process is applied to conduct the V-groove single-pass butt welding (30° in angle) under a shielding gas of 82% Ar and 18% CO2. The thermocouples are inserted into weld pool to record thermal cycle and measured ∆t8/5 time represents the cooling rate. Applied welding parameters and cooling rate are listed in Table 3.
Table 3

Welding parameters selected in present study

Weld

Welding speed (cm/min)

Current (A)

Voltage (V)

Heat input (kJ/cm)

Cooling rate (∆t8/5/s)

A1

40

283.9

28.9

12.31

20.4

A2

50

291.7

29.3

10.26

15.3

A3

60

293

29.3

8.58

11.4

B1

80

379.8

34.5

9.83

17.3

B2

70

350.8

32.1

9.65

15.0

B3

80

328.1

31.2

7.68

11.2

C1

46

340

28.6

12.68

18.0

C2

55

340

28.6

10.61

15.0

C3

67

342

28.7

8.79

11.2

Hardness measurements were completed on etched samples using a Vickers indenter with a 10 kgf load for 15 s dwell time. The Charpy toughness was tested at − 40 °C using a 300J Wolpert impact test machine. Three samples were cut from each weld and then prepared into the size of 5 mm × 10 mm × 55 mm. The V-notch is located at the fusion line according to DIN EN ISO 9016:2013-02 and tests were conducted according to standard DIN EN ISO 148-1:2011 [9, 10]. For metallurgical analysis, the samples were etched with 2% nital and then observed by light optical microscopy (LOM) and field emission gun scanning electron microscopy (FEG-SEM) LEO Gemini 1530 VP operating at 15 kV. EBSD is also carried out to provide the crystallographic information of ferrites with different morphologies and the retained austenite. The measurement were made at 15 or 20 kV accelerating voltage, a spot size of approximately 10 nm and working distance of 18 mm. EBSD maps were obtained with a step size of 0.13 to 0.22 μm. All EBSD raw data were postprocessed with Channel 5 software provided by HKL Technology (Oxford Instruments). Fracture surfaces of Charpy samples were analysed using another SEM (Tescan VEGA3).

3 Results and discussions

3.1 Hardness profiles

The hardness for all steels is almost at the same level, around 270 HV10 ± 5 HV10. But the prior austenite grain size measured for steel A, B and C are 8.6 μm ± 0.1 μm, 21.3 μm ± 4.7 μm and 9.7 μm ± 1.2 μm, respectively. Such different austenite grain sizes combined with different chemical compositions will influence phase transformation in the HAZs and final mechanical properties.

The hardness profiles across the welds from different cooling rates are shown in Fig. 1. An overview of the trends in the hardness variation as a function of cooling rate reveals that different types of microalloyed welds exhibit great discrepancy. In the present paper, particular focus will be made on coarse grained heat affected zone (CGHAZ) and fine grained heat affected zone (FGHAZ) region. For weld A with Nb addition, the hardness increases to a larger extent with increasing cooling rate. The highest hardness reaches up to around 370 HV10. Contrary to this great increase, weld B shows almost constant hardness level even under fast cooling rate, where the hardness keeps approximately 300 HV10. In the case of weld C, the hardness tends to increase with increasing cooling rate but at a relatively small extent. The highest value has reached to around 340 HV10. Therefore, the weld A is the most sensitive to the changes of cooling rate, which indicates that fast cooling rate will result in remarkable hardness increase in CGHAZ and FGHAZ, whereas weld B experiences only limited increase even at fast cooling rate.
Fig. 1

Hardness profiles of all welds (a), (b) and (c) as a function of cooling rate

3.2 Microstructure observation

Figure 2 shows low magnification optical micrographs and high resolution SEM images of the CGHAZs from Nb-bearing welds A. The CGHAZs of all Nb-bearing welds exhibit a mixed microstructure containing upper, lower bainite and martensite. The resultant upper bainite is represented by rather coarse packets through optical metallography. In weld A1 with high heat input, the time above austenitization and during cooling extends, whereby grain growth and diffusion process are favoured. The resultant bainitic ferrite products have a comparatively large size. Some coarse carbides precipitate inside or retained austenite films exist between laths. As fast cooling is applied in weld A3, limited grain growth occurs and subsequently much small bainitic ferrite with fine precipitates are formed. The effect of fast cooling on bainite formation as a result of small ∆t8/5 time can clearly be seen on the SEM images presenting reduced bainitic ferrite lath size. The resultant finer microstructure consisting of predominantly lower bainite and martensite indicates good agreement with the high hardness presented in Fig. 1. Similarly, the FGHAZ microstructure also contains lower bainite and martensite that are not shown here.
Fig. 2

Optical and SEM images in the CGHAZ of A1 (a, b) and A3 (c, d). B, Bc, M and RA are bainite, bainitic carbide, martensite and retained austenite, respectively

Figure 3 shows the Ti-bearing CGHAZ microstructure. In spite of at slow or fast cooling, the microstructure consists of newly formed polygonal ferrite, intragranular acicular ferrite (IGF), bainite and martensite. Some retained austenite was also occasionally observed between ferrite plates. Intragranular interlocked ferrite exhibits a chaotic structure which is well known to be similar with acicular ferrite. Another notable characteristic is the much limited austenite grain growth compared with those in Nb-bearing CGHAZ. Concerning the austenite grain growth in the HAZ with the presence of different microalloy precipitate, another paper has compared this behaviour and confirmed the most effective pinning effect from Ti-rich precipitates [3]. The increased cooling has contributed to a high fraction of martensite (Fig. 3c, d) and an increased hardness value (Fig. 1). The FGHAZs contain a mixed microstructure similar to that in the CGHAZ. The small differences are less martensite formation which results in slight hardness decrease.
Fig. 3

Optical and SEM images in the CGHAZ of B1 (a, b) and B3 (c, d). B, IGF, M and PF are bainite, intragranular acicular ferrite, martensite and polygonal ferrite, respectively

The CGHAZ microstructure obtained in Ti + V-bearing welds is given in Fig. 4. In weld C1 with high heat input, the microstructure is made of IGF, bainite and martensite (Fig. 4a). When the cooling rate increases (Fig. 4d), the IGF fraction is decreased and bainitic ferrite size becomes smaller with much finer carbides inside. Such a microstructure variation involving bainite and martensite as predominant phases is consistent with a corresponding hardness value increase (Fig. 1). The hardness in the FGHAZs shows a slight tendency of increase due to the mixed bainite–martensite microstructure and limited fine grain size.
Fig. 4

Optical and SEM images in the CGHAZ of C1 (a, b) and C3(c, d). B, IGF and M are bainite, intragranular acicular ferrite and martensite, respectively

3.3 EBSD analysis

The EBSD technique aims to assess the crystallographic features of two typical microstructures in the CGHAZ that are bainite with second phase between ferrite laths and intragranular acicular ferrite. The presence of the second phase (retained austenite) is strongly linked with bainite transformation. The chemistry analysis of retained austenite has been performed by EDX mapping method to reveal its formation mechanism. The crystallographic misorientation between ferrite plates were analysed to evaluate the resistance of ferrite with different morphologies to brittle fracture.

An example of EBSD data obtained with specimen A1 is presented in Fig. 5. The image quality map represents the quality of Kikuchi line for each point measurement. It strongly depends on the specimen preparation condition. If image quality is low (see Fig. 5a), the grey-scale level becomes unknown dark point due to the difficulty of obtaining clear Kikuchi-line diffractions in this region. This occurs very often at grain boundaries (indicated by arrow in Fig. 5a) and retained austenite with lattice defects or high dislocations. Accordingly, the quality map image (Fig. 5a) looks like a conventional SEM image and clearly reveals a bainitic and martensitic structure. Figure 5b illustrates the distribution of different phases, in which red area represents ferrite phase and green represents austenite. Quite limited austenite is identified, as indicated by black arrows. The presence of retained austenite tends to be mostly located at grain boundaries or between bainitic ferrite laths, where meanwhile unknown black points are often seen. Therefore, this leads to further difficulties in indexing the EBSD patterns. It was confirmed that the retained austenite combined martensite in the M–A constituent contains many defects such as dislocations. For this reason, they show especially low image quality, which is often observed in bainite transformation [11]. The colours in Fig. 5c correspond to the crystallographic orientation on the transverse direction of observed plane, as indicated by the stereographic triangle. The bainitic ferrites that have retained austenite between laths show the same crystallographic orientation, corresponding to the prevailing red colour. This indicates that they belong to one crystallographic packet group. Additionally, the retained austenite has the same crystallographic orientation because they are retained from the same prior austenite grain.
Fig. 5

Typical orientation imaging maps obtained from the EBSD measurement. a Pattern quality. b Phase map. c Inverse pole figure (IPF) colour map

Another EBSD measurement has been performed at localised retained austenite, as seen in Fig. 6. In Fig. 6a, the selected region is indicated by red line rectangle in SEM image. In the middle of the selected area, the bright white structure is distinguishable from its surrounding structure. Although this bright structure is of low image quality (Fig. 6b), one part of it can still be confirmed to be retained austenite (Fig. 6c). The surrounding matrix is bainitic ferrite with a low misorientation angle (around 3.5°) between ferrite plates.
Fig. 6

Typical orientation imaging maps obtained from the EBSD measurements. a SEM image. b Pattern quality of the region indicated by red line. c Phase map. d Inverse pole figure (IPF) colour map

In order to reveal alloy distribution in retained austenite, EDX elements mapping is present in Fig. 6. The measured area is indicated by blue rectangle (see Fig. 6a). Both observed needle-like region and block-type region have less Fe content compared with the surrounding ferrite matrix (Fig.  7 b). Instead, they are enriched with high Ni (Fig. 7 c). Moreover, only an increase of C in needle-like shape region is seen in Fig. 7 d. These element distribution implies that alloying elements Ni and C are enriched in retained austenite and make it stable to room temperature after transformation. In addition, there are no significant changes in Cr, Mn, Mo and Si contents (Fig. 8).
Fig. 7

EDX mapping of the region indicated by blue line in Fig. 6. a SEM image. b Fe. c Ni. d C

Fig. 8

EDX mapping of the region indicated by blue line in Fig. 6. a Cr. b Mn. c Mo. d Si

As is the case for the intragranular ferrite microstructure, another EBSD analysis was made on specimen B2. Figure 9 gives the IPF colour map combined with boundary misorientation distribution and IPF colour map for austenite. The grain boundaries are based on the misorientation between adjacent points of the EBSD data. Different criteria with varied colours are used to draw boundaries (Fig. 9a). The low-angle misorientation boundaries are highlighted by red, dark yellow or magenta colours in bainitic ferrite packet. However, high angle boundaries coloured with blue or violet are seen to interrupt the parallel bainite structure. These are grain boundaries between IGF and bainite. This point could be deduced based on the morphology characteristics of fine interlocked structure from intragranular ferrite. It is well known that the fine acicular ferrite can grow from the same inclusions but with different crystallographic orientation. Also, some retained austenite is visible in Fig. 9b, which is accompanied by the formation of IGF and bainite.
Fig. 9

Typical orientation imaging maps obtained from the EBSD measurements. a IPF bcc colour map combined with boundary misorientation distribution. b IPF fcc colour map for austenite

A further study on the misorientation was performed in this mixed structure. Figure 10 shows the point to point and point to origin misorientation profiles of line A and line B (indicated in Fig. 9a). It demonstrates that the orientation changes between and within ferrite laths. From Fig. 10a, b, it is evident that nearly all acicular ferrite plates have high angle boundaries with misorientation angle greater than 45° or sometimes with a misorientation of approximately 60°. However, within bainitic ferrite, low-angle boundaries less than 10° are often found. This characteristic orientation feature is derived from the formation mechanism and the nature of different microstructure (bainite or IGF). The influence of the misorientation between neighbouring crystallographic ferrite grains is strongly correlated with their ability to resist crack propagation.
Fig. 10

Misorientation along the line A (a) and line (b), including point to point misorientation and point to origin misorientation

3.4 Charpy V notch toughness results

Table 4 summarises the Charpy toughness results from all HAZs. Note that the observed large toughness value scattering is on the one side predetermined by the additional effect of weld shape on impact toughness values. Such difference in weld shape has already been seen in other paper [12]. On the other hand, the controlling factor to determine toughness is related with hardness and microstructure features that involve austenite grain size, matrix structure and second phases as well. Contrary to the relationship between hardness and toughness in the weld metal, high HAZ hardness does not always indicate low toughness values particularly for Nb- and Ti + V-bearing HAZs.
Table 4

Toughness results of different HAZs

Weld

Cooling rate (∆t8/5/s)

Hardness/HV10 (highest value)

CVN impact energy (J)

Experimental values (average)

A1

20.4

312

14, 16, 11 (14)

A2

15.3

338

12, 14, 16 (14)

A3

11.4

368

18, 36, 38 (31)

B1

17.3

273

52, 64, 47 (54)

B2

15.0

272

20, 49, 28 (32)

B3

11.2

299

66, 76, 42 (61)

C1

18.0

294

24, 37, 34 (32)

C2

15.0

322

24, 18, 18 (20)

C3

11.2

341

72, 67, 74 (71)

In the case of Nb-bearing HAZs, the most noticeable low value of impact toughness occurred at a slow to medium cooling rate, indicating that the cleavage resistance in the HAZ is weak. However, at much higher cooling rate, the toughness exhibits a tendency to increase and approaches a satisfactory level of standard requirements. Contrary to this obvious toughness variation with heat input changes, Ti-bearing weld B mostly have much higher energy values irrespectively of a slow or fast cooling rate.

A further fractographic examination was performed to reveal the fracture mode. The fractographic observation on radial zone is given in Fig. 11. Figure 11a, b illustrates that the fracture at high heat input is predominantly in brittle manner. The river lines within the large facet corresponds well with the large bainitic ferrite packets where the crack can propagate easily without crossing a high-angle boundaries as indicated by the EBSD results. Such an observation confirmed that the microstructural coarseness results in a subsequent coarseness of the cleavage facet size. Indeed, it would be desirable to know exactly the initiation factor to get an impression of brittle fracture mechanism and this is often reported to be related with M–A constituent. But the exact role of M–A constituent on fracture process in present study is difficult to determine due to lack of sufficient evidence.
Fig. 11

Fractographs at radial zone of Charpy specimens in the HAZ of welds A (a-c), B (d-f) and C (g-i)

Figure 11d–f refers to the fracture surfaces from Ti-bearing HAZs. All three images suggest that the radial zone has a quasi-cleavage fracture mode, corresponding well with the higher impact energy. A close examination at fracture surfaces has been done in order to search for TiN particles and further determine their roles during fracture process. The observation shows that no identifiable TiN act as origin of fracture or cause the secondary initiation while no surrounding facet emanated from TiN particles. Therefore, in steel B, even when potential fracture triggering TiN particles are present in the HAZ, they did not cause any brittle fracture initiation and lower the impact toughness. Additionally, reviewing the matrix microstructure, a small austenite grain size combined with fine interlocking acicular ferrite and small bainite packet has inter-related contribution to small facet size and excellent toughness. Hence, it is confirmed that the possible deleterious role of TiN particles as crack initiator was deactivated in Ti-bearing HAZ by the presence of pronouncedly refined microstructure.

The fracture behaviour of weld C (Fig. 11g–i) was in a manner comparably similar with that in Nb-bearing HAZ, in which the river pattern is clearly visible. This reveals that, owing to the potential austenite growth and the subsequent large-sized bainite products, which occurred to both two welds, the fracture mode is varied in a similar tendency and strongly influenced by the grain size and microstructure composition. Nevertheless, the comparatively better toughness in weld C than weld A can be explained by the presence of certain intragranular ferrite and better pinning efficiency of precipitates. Moreover, the Charpy toughness is significantly improved at fast cooling rate due to the refined microstructure and the absence of upper bainite. This overall tendency among three welds reveal that it should be therefore possible to increase the HAZ toughness by ensuring proper low heat input.

3.5 Effect of microstructure on HAZ toughness

In the case of Nb-bearing steel A, HAZ toughness can be impaired by excessive grain coarsening, large bainitic ferrite packets, upper bainite or the presence of M–A phase. The relatively coarse austenite grain in the HAZ was attributed to the limited pinning effect from Nb-rich carbonitrides and AlN that have comparatively low dissolution temperatures, as discussed in another paper [3]. As a consequence of coarse austenite grains, large bainite packet or bainitic ferrite laths were subsequently formed during austenite decomposition, especially when welding at high or medium heat input. The upper bainite structure is then characterised using EBSD to understand the relevant ‘grain size’ and bainitic packet size which are critical for influencing crack propagation. The high volume fraction of upper bainite, accompanied with coarse matrix grain, holds responsible for the loss in toughness, that is one reason why heat input needs to be controlled at low level for Nb-bearing welds A when considering high HAZ toughness. Under low heat input, limited time is available for austenite grain growth. Meanwhile, upper bainite is mostly substituted by lower bainite, where a mixture of refined lower bainite and martensite with much less retained austenite is formed. Such refined microstructure can further contribute to excellent toughness and the present fractographic observation supports this point. Other researchers have reported similar tendency of HAZ toughness changes as a function of heat input in Nb containing welds, and the combination of Nb/Ti is also suggested to avoid the rapid coarsened austenite [7, 8, 13]. Also, the detrimental effect of high heat input on the toughness of weld containing niobium has been well documented and explained [8, 13, 14]. It is reported that Nb in solution can strongly delay austenite to ferrite transformation and suppresses the transformation start temperature of bainite formation. The reason is due to the grain-boundary segregation of Nb, which raises the energy barrier at boundaries and retards the transformation of austenite to grain boundary ferrite or pearlite by solute drag effect [1, 15]. The possible detrimental effect of Nb in the present study could be correlated with the formation of brittle upper bainite and untempered martensite or granular bainite and M–A constituent.

By utilising Ti-rich particles as well as by finely balanced alloy design, it can prevent excessive austenite grain coarsening and meanwhile promote IGF formation. The EBSD results have confirmed that high angle boundaries between IGF plates present effective barriers for crack to propagate and thus contribute to high HAZ toughness. In spite of beneficial effect from IGF on improved toughness, another feature about TiN particle acting as positive or negative factor of toughness must be additionally considered. Some researchers claimed that TiN particles could act as fracture initiation sites in the case of bainitic structure with large praket size, but they are irrelevant to fracture process in the case of fined-grained acicular ferrite as matrix structure [6]. Support of this observation can be found in other literature [16], which points out that the susceptibility to cleavage fracture did not depend on the size of fracture initiating particle but additionally on local ferrite grain size. Some suggested that when a fine balance of Ti and N element, where Ti/N weight ratio is close to 2.2, dispersed fine TiN particle will bring beneficial effect on toughness [17]. This has been well approved by our observation of Ti-bearing steel with Ti/N ratio of 2. Therefore, based on the above results and comparison between the findings of others, it is concluded that Ti addition with an acceptable balance with N has assisted in improved HAZ toughness by the formation of desirable IGF, fine grain size and dispersed fine particles.

To explain HAZ toughness behaviour of Ti + V-bearing welds, it should highlight that weld C experienced the similar toughness variation tendency as heat input changes compared with that of weld A. The same reasons can be interpreted here by the coarser grain structure than that in Ti-bearing HAZs, combined with upper bainite present in the HAZs when welding at high or medium heat input. It should note that the presence of IGF structure in weld C has also positive contribution to toughness. This is however absent at medium cooling rate or when more amount of upper bainite is formed, toughness will decrease. When welding at low heat input, likewise in other two welds, a satisfactory toughness can be approached thanks to the refined matrix microstructure of lower bainite and martensite.

4 Conclusions

a) Considering the fracture behaviour of Nb-bearing HAZ at high heat input, several factors, such as the high volume fraction of upper bainite, accompanied with coarse matrix grain, hold responsible for loss in toughness and determine the brittle fracture mode. Fortunately, HAZ toughness is improved in the case of the refined lower bainite and martensite as predominant microstructure when low heat input is applied.

b) Ti addition with an acceptable balance with N has assisted in improved HAZ toughness by the formation of desirable IGF, fine grain size and fine dispersed particles. On the fracture surface, Ti-rich particles are not observed to be the origin of fracture indicating no any deterioration effect on fracture behaviour in the case of fine IGF structure as matrix.

c) EBSD results show that fine intragranular ferrite has higher angle boundaries and these provide effective barriers for crack propagation, thus improving HAZ toughness. Regarding the role of M–A constituent surrounding the upper bainite structure, although some crystallographic and chemical information are revealed, the exact role of M–A constituent on fracture process is difficult to determine due to lack of sufficient evidence.

Notes

Acknowledgements

The authors thank R. S. Neumann, M. Buchheim, S. Brunow, D. Schroepfer and E. Steppan from Federal Institute for Materials Research and Testing (Germany) for their kind support. One of the authors (Lei Zhang) also appreciates the funding support from China Scholarship Council.

References

  1. 1.
    Gladman T (1997) The physical metallurgy of microalloyed steels. Institute of Materials, LondonGoogle Scholar
  2. 2.
    Mitchell P, Hart P, Morrison W (1995) The effect of microalloying on HAZ toughness. In: Microalloying ‘95 Conf Proc pp 149–162Google Scholar
  3. 3.
    Zhang L, Kannengiesser T (2014) Austenite grain growth and microstructure control in simulated heat affected zones of microalloyed HSLA steel. Mater Sci Eng A 613:326–335.  https://doi.org/10.1016/j.msea.2014.06.106 CrossRefGoogle Scholar
  4. 4.
    Fairchild DP, Howden DG, Clark WAT (2000) The mechanism of brittle fracture in a microalloyed steel: part I. Inclusion-induced cleavage. Metall Mater Trans A 31(3):641–652.  https://doi.org/10.1007/s11661-000-0007-4 CrossRefGoogle Scholar
  5. 5.
    Zhang LP, Davis CL, Strangwood M (2001) Dependency of fracture toughness on the inhomogeneity of coarse TiN particle distribution in a low alloy steel. Metall Mater Trans A 32(5):1147–1155.  https://doi.org/10.1007/s11661-001-0125-7 CrossRefGoogle Scholar
  6. 6.
    Echeverria A, Rodriguez-Ibabe JM (2003) The role of grain size in brittle particle induced fracture of steels. Mat Sci Eng a-Struct 346(1–2):149–158.  https://doi.org/10.1016/S0921-5093(02)00538-5 CrossRefGoogle Scholar
  7. 7.
    Lazor RB, Glover AG, McDonald RD (1984) Properties of welds in thick section Nb-containing steels. In: Welding in energy-related projects. Pergamon pp 1–9. doi: https://doi.org/10.1016/B978-0-08-025412-8.50005-X
  8. 8.
    Hattingh RJ, Pienaar G (1998) Weld HAZ embrittlement of Nb containing C–Mn steels. Int J Press Vessel Pip 75(9):661–677.  https://doi.org/10.1016/S0308-0161(98)00066-0 CrossRefGoogle Scholar
  9. 9.
    DIN EN ISO 9016:2013-02. Destructive tests on welds in metallic materials—impact tests—test specimen location, notch orientation and examination (ISO 9016:2012); German version EN ISO 9016:2012 (2013). vol DIN EN ISO 9016:2013–02Google Scholar
  10. 10.
    DIN EN ISO 148-1:2011. Metallic materials—Charpy pendulum impact test—part 1: test method (ISO 148-1:2009); German version EN ISO 148–1:2010 (2011)Google Scholar
  11. 11.
    Terasaki H, Komizo YI (2013) Morphology and crystallography of bainite transformation in a single prior-austenite grain of low-carbon steel. Metall Mater Trans A 44(6):2683–2689.  https://doi.org/10.1007/s11661-013-1610-5 CrossRefGoogle Scholar
  12. 12.
    Zhang L, Kannengiesser T (2016) HAZ softening in Nb-, Ti- and Ti + V-bearing quenched and tempered steel welds. Weld World 60(2):177–184.  https://doi.org/10.1007/s40194-016-0299-7 CrossRefGoogle Scholar
  13. 13.
    Zhang Y-Q, Zhang H-Q, Li J-F, Liu W-M (2009) Effect of heat input on microstructure and toughness of coarse grain heat affected zone in Nb microalloyed HSLA steels. J Iron Steel Res Int 16(5):73–80.  https://doi.org/10.1016/S1006-706X(10)60014-3 CrossRefGoogle Scholar
  14. 14.
    Wang G, North T, Leewis K (1990) Microalloying additions and HAZ fracture toughness in HSLA steels. Weld J 69(1):14–22Google Scholar
  15. 15.
    Yang JR, Huang CY, Chiou CS (1999) Microstructures of heat-affected zone in niobium containing steels. Mater Trans JIM 40(3):199–208.  https://doi.org/10.2320/matertrans1989.40.199 CrossRefGoogle Scholar
  16. 16.
    Petch NJ (1986) The influence of grain boundary carbide and grain size on the cleavage strength and impact transition temperature of steel. Acta Metall 34(7):1387–1393.  https://doi.org/10.1016/0001-6160(86)90026-X CrossRefGoogle Scholar
  17. 17.
    Rak I, Gliha V, Koçak M (1997) Weldability and toughness assessment of Ti-microalloyed offshore steel. Metall Mater Trans A 28(1):199–206.  https://doi.org/10.1007/s11661-997-0096-4 CrossRefGoogle Scholar

Copyright information

© International Institute of Welding 2018

Authors and Affiliations

  1. 1.Federal Institute for Materials Research and TestingBerlinGermany

Personalised recommendations