1 Introduction

Every year, there is an increasing demand for high-strength steels, which are used in the manufacturing process of welded constructions in cars, trucks, cranes, bridges, mining machinery, and other welded constructions [1, 2]. The effort to reduce the weight of these constructions is driven by the desire to lower the carbon footprint generated by these mechanisms during their lifetime. Among the steels employed for this purpose are the high-strength low-alloy (HSLA) structural steels S690QL, S960QL, S1100QL, S700MC, S960MC, and S1100MC, which are used in the previously mentioned sectors owing to their high strength and ductility. HSLA steels typically have a lower carbon content (0.05 to 0.25%) compared to mild or low-carbon steels. Therefore, HSLA steels are often considered better for welding than mild steels, due to its resistance to hydrogen-induced cracking (HIC). However, practical experience makes it evident that the requirement of verified weldability needs to be reconsidered for thin material thicknesses (t ≤ 3 mm) [3]. The reason is unsatisfactory heat transfer in thin materials, which results in the heat-affected zone’s (HAZ) microstructural changes (softening effect) being enforced. As an example, S960MC steel’s thin sheets welded using the gas metal arc welding (GMAW) method fail to achieve the minimum requirements for yield strength (YS) and ultimate tensile strength (UTS), whereas thicker sheets (t > 3 mm) fulfill the requirements despite requiring a higher heat input [3,4,5,6,7,8]. The EN ISO 15614–11 standard defines these requirements; it specifies that a minimum YS and UTS must be at least the minimum value of YS/UTS of BM specified in the product standard (EN ISO 10149–2, EN ISO 10025–6) or by the manufacturer (for steels with YS above 960 MPa). Therefore, for welding thin HSLA steel sheets, a laser beam welding (LBW) is suggested. Several studies have stated that the application of LBW is appropriate for quenched and tempered steels (Q&T) [9,10,11,12] and for steels that have undergone thermomechanical controlled processing (TMCP) [13,14,15,16,17].

Minimizing the heat input makes it possible to significantly reduce the zone affected by irreversible microstructure changes in the HAZ, and at the same time, it also affects the mechanical properties of the HAZ. One of the approaches to explain the irreversible microstructural changes in the HAZ responsible for the formation of the soft zone (SZ) can be through the theory of polycrystalline strengthening [18] expressed in Eq. (1):

$$\upsigma ={\upsigma }_{0}+{\upsigma }_{{\mathrm{s}}}+ {\upsigma }_{{\mathrm{p}}}+{\upsigma }_{{\mathrm{d}}}+{\upsigma }_{{\mathrm{g}}}$$
(1)

where σ0 is the internal lattice stress, σs is the solid solution strengthening, σp is the precipitation strengthening, σd is the dislocation strengthening, and σg is the grain boundary strengthening.

According to the polycrystalline strengthening theory, considering the HAZ as a single entity is not possible owing to various microstructural changes. When it comes to HSLA steels, the majority of the changes occurred during welding are accompanied by a decrease in the effect of strengthening mechanisms; thus, it reduces the material’s yield strength in the HAZ (softening effect). In the zones exposed to temperatures exceeding Ac3 (the temperature at which the ferrite is fully transformed into austenite), recrystallization occurs and so the subsequent change in the size of the prior austenitic grain (PAG). In the HAZ, they are coarse grain heat-affected zone (CGHAZ) and fine grain heat-affected zone (FGHAZ). The change in the PAG size can be determined by Eq. (2):

$${\mathrm{D}}^{\mathrm{m}}-{\mathrm{D}}_{0}^{\mathrm{m}}\mathrm{ = }{\mathrm{K}} \, \times \, {\mathrm{t}}$$
(2)

where D (mm) is the mean grain diameter, D0 (mm) is the initial grain diameter, t (s) is the soaking time at a given temperature, K (mm2·s−1) is the proportional constant depending on the temperature T (K) and the activation energy Q (kJ·mol−1) required for grain growth, and m expresses grain growth coefficient dependent on temperature, chemical composition, processing conditions, and impurities in alloy [19]. As the grain size increases in the CGHAZ, the strengthening effect of the σg decreases, resulting in a decrease in the YS within the range defined by the Hall–Petch relation Eq. (3) [20]. The opposite phenomenon occurs when the grain is refined (may occur in the FGHAZ).

$${\upsigma }_{{\mathrm{g}}} = {k}_{{\mathrm{y}}} \times {d}_{\mathrm{\alpha }}^{-1/2}$$
(3)

where ky (MPa·mm1/2) is a coefficient expressing the effect of grain size on the increase in YS and dα (mm) is the average ferritic grain size. The coefficient ky has a different value for each material, which can differ significantly and is therefore determined experimentally for each material. In subzones where recrystallization does not occur, the increase/decrease in σg is not observed.

A substantial proportion of the overall YS of HSLA steels is dislocation strengthening [21, 22]. The dislocation strengthening component is expressed in Eq. (4):

$${\upsigma }_{{\mathrm{d}}} = 2 \times \alpha \times G \times b \times {\rho }^{1/2}$$
(4)

where α is a constant depending on the crystal structure (0.5 for the ferritic lattice), G is the shear modulus (8 × 104 MPa for Fe), b is the Burgers vector (0.25 nm), and ρd (m−2) is the dislocation density. According to Eq. (4), the dislocation density is the only factor influencing any change in dislocation strengthening in HSLA steels. In the welding process, when steel is heated to elevated temperatures, dislocations have more thermal energy and can move more freely. At these higher temperatures, a process called “recovery” occurs in which the dislocations rearrange to form lower energy configurations. This process can lead to a decrease in dislocation density. This phenomenon can be observed in sub-critical heat-affected zone (SCHAZ). Reduction in dislocation density may not occur in weld joints welded under rapid cooling conditions (e.g., LBW) as the material becomes hardened and dislocations increase again. Such a phenomenon can be observed in the weld metal (WM), CGHAZ, FGHAZ, and partially in the inter-critical heat-affected zone (ICHAZ). The ICHAZ is specific, as the original martensitic microstructure is partially dissolved into ferrite and carbides. Such a transformation can be accompanied by both a decrease in the ρd (ferritic transformation) and an increase in the ρd (martensitic transformation).

The YS of HSLA steels is also significantly influenced by precipitation strengthening. By suppressing dislocation motion with incoherent and non-deformable precipitates (Orowan bypassing), precipitation strengthening σp improves the overall YS [22]. This relationship can be mathematically described by the Ashby-Orowan model (5):

$${\upsigma }_{\mathrm{p}}\mathrm{ = 9.995 }\times \, {10}^{3}\times \, \frac{{\mathrm{f}}^{1/2}}{\mathrm{d}} \, \times \, {\mathrm{ln}}\mathrm{(2.417 }\times \, {\mathrm{d}}\mathrm{)}$$
(5)

where f (%) is the volume fraction of precipitates in the matrix and d (nm) is the average size of the precipitates. Due to precipitate coarsening at high temperatures, the strengthening effect in the HAZ might decrease. The HAZ decreases most significantly in the areas that are exposed to the highest temperatures (CGHAZ, FGHAZ). As previously stated, in ICHAZ, carbides may form during the decay of the initial martensitic structure. However, due to their size, such carbides do not contribute to the increase in strength, and therefore, their formation cannot be considered desirable.

The knowledge provided above makes it evident that the base metal’s (BM) mechanical properties and delivery state affect the ratio of each strengthening mechanism. As a result, the material’s reaction to the temperature cycle and the subsequent changes in the HAZ will differ. It may be assumed that fusion-welded HSLA steel joints will eventually undergo changes that cause the HAZ to soften. The resulting SZ can significantly negatively affect the mechanical properties of weld joints. Above all, it is a decrease in the YS, UTS, and elongation (A).

However, the theory implies that the existence of SZ does not necessarily reduce the mechanical properties. For instance, Maurer et al. [23], Amraei et al. [24], Rodrigues et al. [25], or de Meester [26] reported that with a sufficiently narrow SZ, YS and UTS is at least at the level of BM. In the literature, the term soft zone is often used, but information about its determination is largely absent. Thus, the requirement to understand the SZ’s width (wsz) forms the cornerstone for the application of theoretical findings. A possible approach for determining the wsz is based on the line hardness measurements that were published by Mičian et al. [3] or Frátrik et al. [14]. In the instance, a section of the material that is softer than 90% of the BM’s hardness is thought to be the SZ.

When compared to arc welding technologies, the HAZ of laser-welded joints have different properties because of their higher heating and cooling rates and lower heat input. While a higher heating rate increases Ac1 (the lowest temperature at which austenite can form on heating at a specified heating rate) and Ac3 temperature values, decreasing the width of the CGHAZ, FGHAZ, and ICHAZ subzones [27, 28], a higher cooling rate promotes the martensitic transformation. The lower heat input has also a partial effect on the microstructure in the SCHAZ, which, due to the narrowed temperature fields and lower soaking times, is less tempered because of the welding process [29]. Such an increase in the values of Ac1 and Ac3 is desirable, as it leads to a narrowing of the SZ. In all subzones, an increase in the cooling rate associated with LBW is advantageous because it prevents the decrease in dislocation density or, in certain circumstances, increases it.

It is also important to consider the possibility of very high cooling rate in the case of LBW, which would lead to brittle decay structures. Such microstructure may significantly raise a given steel’s susceptibility to the HIC [30, 31].

Apart from the wsz, the overall mechanical properties of weld joints are also influenced by the ratio between the UTS of the HAZ (\({\upsigma }_{\mathrm{HAZ}}^{\mathrm{T}}\)) and BM (\({\upsigma }_{\mathrm{BM}}^{\mathrm{T}}\)) and between the UTS of the WM (\({\upsigma }_{\mathrm{WM}}^{\mathrm{T}}\)) and BM (\({\upsigma }_{\mathrm{BM}}^{\mathrm{T}}\)). For example, Mochizuki et al. [32] claim that the strength of the welded joint increases with an increase in the ratio of \({\upsigma }_{\mathrm{HAZ}}^{\mathrm{T}}\)/\({\upsigma }_{\mathrm{BM}}^{\mathrm{T}}\) and \({\upsigma }_{\mathrm{WM}}^{\mathrm{T}}\)/\({\upsigma }_{\mathrm{BM}}^{\mathrm{T}}\). According to Mochizuki et al. [32], Maurer et al. [23], or Amraei et al. [33], an increase in these ratios results in an increase in the overall YS and UTS of the weld joint. It can be difficult to determine the \({\upsigma }_{\mathrm{HAZ}}^{\mathrm{T}}\) value at that point. One of the ways to express the \({\upsigma }_{\mathrm{HAZ}}^{\mathrm{T}}\) is by physical simulation. For example, Mičian et al. [34] determined the YS and UTS of the individual HAZ subzones of S960MC steel. It was found that the UTS in the ICHAZ reaches only 75% of the S960MC BM’s value. Such a determination, although it is considered very accurate, may not fully reflect the actual processes in the HAZ of weld joints. Therefore, a determination of the \({\upsigma }_{\mathrm{HAZ}}^{\mathrm{T}}\) can be obtained on the basis of hardness. For example, Pavlina et al. [35] applied such a calculation for YS (Eq. (6)) and UTS (Eq. (7)):

$$\mathrm{YS }= -90.7 + 2.876 \times {{\mathrm{H}}}_{{\mathrm{v}}}$$
(6)
$$\mathrm{UTS }= -99.8 + 3.734 \times {{\mathrm{H}}}_{{\mathrm{v}}}$$
(7)

where YS and UTS have units of (MPa) and Hv (kgf·mm−2) is Vickers hardness.

The same method as for the HAZ can also be applied to the WM. Apart from the temperature cycle, the primary factor affecting the \({\upsigma }_{\mathrm{WM}}^{\mathrm{T}}\) is the chemical composition of the filler metal (FM) or BM, respectively. The fact that the FM or BM contains an increased amount of elements that contribute to hardenability (Mn, Cr, Mo, and Ni) in this instance is advantageous for the strength increase. In general, the carbon equivalent (CET) value of BM may be applied to indicate the hardenability of WM of weld joints that are welded without filler metal, such as LBW. It is anticipated that the \({\upsigma }_{\mathrm{WM}}^{\mathrm{T}}\) will increase as the CET value increases (under comparable conditions and temperature cycles).

In most HSLA steels, a significant softening effect also occurs in the CGHAZ [19, 32]. The reason is an increase in the PAG size. As mentioned, the decrease in YS in the CGHAZ is caused by the decrease in the strengthening mechanisms σg, σd, and σp. As it was already demonstrated, the absolute PAG size causing a large decrease in σg is dependent on the maximum temperature reached and soaking time [19]. However, the welding technology can greatly affect the width of the CGHAZ through the shape of the temperature field. Therefore, to narrow the CGHAZ, it is recommended to apply LBW technology prior to GMAW technology. One of the factors that, in addition to temperature, may contribute to the overall increase in the PAG size is the chemical composition of the BM. In this case, precipitates (carbides, nitrides, carbonitrides) contribute to the suppression of PAG growth.

In particular, the precipitates of Nb, V, and Ti are significant for the properties of the CGHAZ and FGHAZ because they remain stable even at high temperatures, contributing to the so-called Zener pinning, which controls grain growth in the CGHAZ and FGHAZ [36, 37]. It can be stated that the presence of dispersed particles (carbides, nitrides, carbonitrides), which contribute to Zener pinning, can indirectly affect grain growth by influencing the grain boundary mobility. The interaction between the dispersed particles and the grain boundaries can result in a reduced mobility of the boundaries, hindering their migration and impeding grain growth. When particles are present at the grain boundaries, they can act as barriers or obstacles, restricting the movement of the grain boundaries. The particles pin the grain boundaries and prevent their migration, which limits the grain growth process.

The SCHAZ is the farthest-off subzone of the HAZ, whose mechanical properties are affected by the temperature cycle. No phase transformations or changes in the PAG’s size take place in the given zone because the temperature Ac1 is not exceeded. However, BM’s tempering occurs in the case of HSLA steels, which are composed of non-equilibrium structural components. In the context of strengthening mechanisms, it is possible to look at the BM’s softening as a process that decreases the dislocation density and, therefore, as a decrease in the YS due to a decrease in σd. The tempering process in steel is influenced by several key factors that affect its final properties, including hardness, toughness, and strength.

Despite the fact that there has been numerous research on the weldability of HSLA steels by LBW, this work provides a distinctive overview based on a comparison of the characteristics of the weld joints of different steels. The characteristics of the soft zone and how they affect the overall mechanical properties of the welded joints are also the primary objective of the investigation of the welded joints.

2 Materials and methods

2.1 Materials

Experimental materials utilized for this study are quenched and tempered (Q&T) structural steels S690QL, S960QL, and S1100QL and thermo-mechanically controlled processed (TMCP) structural steels S700MC, S960MC, and S1100MC. Microstructure of Q&T steels as well as S960MC and S1100MC is composed by the mixture of martensite, tempered martensite, bainite, and residual austenite. The amount of individual compounds and its ratio differs based on individual materials. Steel S700MC is composed by the mixture of fine-grained bainite and acicular ferrite. All materials were delivered in the form of rolled sheets with a thickness of 5 mm (S690QL, S960QL, S1100QL, S700MC) and 6 mm (S960MC, S1100MC), respectively. The chemical composition of the assessed steels obtained by own spectral analysis using Q2 ION device is shown in Table 1. Mechanical properties of assessed steels obtained by own measurements are listed in Table 2. The values of YS, UTS, and A50 were determined by static tensile test as average values from three independent measurements. The value of HV0.1 was determined as average value of ten independent measurements (Vickers hardness test). CET (wt.%) expresses the carbon equivalent as defined by EN ISO 1011–2. The calculation was carried out according to Eq. (8).

Table 1 Chemical composition of assessed steels (wt.%)
Full size table
Table 2 Mechanical properties and CET values of assessed steels
Full size table
$$\mathrm{CET = }{\mathrm{C}}\mathrm{ + }\frac{\mathrm{Mn + Mo}}{10}\mathrm{ + }\frac{\mathrm{Cr + Cu}}{20}\mathrm{ + }\frac{\mathrm{Ni}}{{40}}$$
(8)

2.2 Experimental welds

All welded joints were made as butt-welded I-shaped joints without filler metal with zero root gap on the TRUMPF TruDisk 5001 device. Welding edges were aligned by milling and cleaned and degreased before welding. The sheets were clamped in the jig to prevent deformations during welding. The dimensions of the welded sheets were 150 mm × 150 mm × 5 mm and 150 mm × 150 mm × 6 mm, respectively. Welded joints were welded in two variants for each material. One variant was welded with lower heat input (without designation), while the second variant was welded with higher heat input designated with additional letter “H.” The welding parameters are listed in Table 3. All joints were shielded by Ar 4.6 shielding gas. The welding setup is shown in Fig. 1.

Table 3 Welding parameters for all assessed welded joints
Full size table
Fig. 1
figure 1

Welding setup for all experimental welded joints

Full size image

During welding, thermal cycles were measured from the root side. Thermal cycles were measured with K-type NiCr-NiAl thermocouples (TC), which were placed on the surface of the welded sheet (from the root side). The exact values of the TC distance from the weld edge are given in Table 3. The aim of the measurement was to determine the heating rate, cooling rate, and time t8/5 (cooling time from 800 to 500 °C) in the HAZ of the welded joints. The temperature was recorded at a frequency of 80 Hz. For all temperature cycles, a 1st derivative was constructed to evaluate the maximal heating rate and the maximal cooling rate. Figure 2 shows cross-sections of experimental welded joints.

Fig. 2
figure 2

Cross-sections of experimental welded joints: a S690QL, b S690QL-H, c S960QL, d S960QL-H, e S1100QL, f S1100QL-H, g S700MC, h S700MC-H, i S960MC, j S960MC-H, k S1100MC, l S1100MC-H

Full size image

2.3 Experimental methods

Two samples were taken from each welded joint for the purpose of a static tensile test. The preparation, shape, dimensions of the samples, and the test conditions corresponded to the requirements of the EN ISO 4136 standard. The test was performed on the INSTRON 5985 device.

All considered weld joints were subjected to a microhardness test. Microhardness was measured by the Vickers method (HV0.1) in the form of line measurements. The line was guided in cross-section through the weld joint in the middle of the material thickness and covered the entire HAZ, WM, and part of the BM. The line consisted of 121 measurements with a mutual distance of 0.1 mm between indentations. The purpose of the test was to measure the hardness changes and the width of the SZ (wsz) in all of the weld joints. The SZ for every weld joint was defined as the zone with a decrease in hardness below 90% of the hardness value of the BM. The width of this zone was determined from the X-axis of the graphical plot. Hardness of the BM was taken from Table 2. An example can be seen in Fig. 3.

Fig. 3
figure 3

Example of determining the width of the soft zone

Full size image

For the purposes of mutual comparison, it will be necessary to consider the thickness of the BM in addition to the wsz. The parameter considering the thickness of the BM is called the relative thickness of the soft zone (XSZ). The determination of the XSZ was done according to Eq. 9.

$${\mathrm{X}}_{\mathrm{SZ}}\mathrm{ = }\frac{{\mathrm{w}}_{\mathrm{SZ}}}{\mathrm{t}}$$
(9)

where XSZ (-) is relative thickness of the soft zone, wsz (mm) is the width of the SZ, and t (mm) is the thickness of the BM.

A bend test was performed to obtain an overview of the ductility and deformability of the weld joints. The dimensions, preparation of the samples, and the test procedure were made according to the EN ISO 5173. Four samples (2 × TRBB, transverse root bend test specimen for a butt weld; 2 × TFBB, transverse face bend test specimen for a butt weld) were tested on each weld joint, while the diameter of the load indenter and the distance of the supports were calculated separately for each material according to the minimum elongation given by the product standard (Eq. 10):

$${\mathrm{d}}_{\mathrm{i}}\mathrm{ = }\frac{\mathrm{100 }\times {\mathrm{t}}_{\mathrm{s}}}{\mathrm{A}}-{\mathrm{t}}_{\mathrm{s}}$$
(10)

where di (mm) is the diameter of the test indenter, ts (mm) is the thickness of the test sample, and A (%) is the minimum ductility based on the product standard, or material data sheet from the manufacturer.

Microstructural analysis was performed by light optical microscopy (LOM) on the Leica DMi8 device. Samples were etched with 3% Nital. The objective of the analysis was to determine the mean PAG size in the CGHAZ and the width of the CGHAZ. The CGHAZ was considered to be the part of the HAZ with a coarsened grain from the fusion line to the zone with a grain size equal to the original grain of the BM (Fig. 4). Mean PAG size was determined by a linear intercept method. Both the CGHAZ width and the mean PAG size were evaluated at the center of the material thickness. To provide additional support for the measurements, images of the CGHAZ were obtained by scanning electron microscope (SEM) using the Phenom XL device.

Fig. 4
figure 4

Example of determining the width of the CGHAZ

Full size image

3 Results and discussion

Figure 5 shows the temperature cycles for the S1100QL, S700MC-H, S960MC, and S960MC-H weld joints so that all thickness variations and all heat input variations are covered.

Fig. 5
figure 5

Thermal cycles of the selected welded joints: a S1100QL, b S700MC-H, c S960MC, d S960MC-H

Full size image

Table 4 shows selected parameters of the weld joints thermal cycles. From the measured results, it is possible to observe the effect of increasing heat input on the decreasing cooling rate. However, with respect to the measured values of the cooling rate and the characteristics of the continuous cooling transformation (CCT) diagrams of the assessed materials, the observed differences can be considered irrelevant for the resulting decay structures. According to the published CCT diagrams [38,39,40], the cooling rate for all materials leads to the formation of martensite in the HAZ. Figure 6 shows a CCT diagram for S960MC steel, where it can be observed that cooling rates exceeding 50 °C·s−1 (typical for LBW) lead to martensitic transformation.

Table 4 Selected thermal cycle parameters
Full size table
Fig. 6
figure 6

CCT diagram of S960MC steel [39]

Full size image

By the static tensile test, the YS, UTS, and A50 were evaluated. The results of the test are shown in Table 5.

Table 5 Mechanical properties of all assessed weld joints
Full size table

From the given values, it can be observed that the amount of heat input does not directly affect the resulting values of the YS, UTS, and A50. For steels S690QL, S960QL, and S960MC, this fact is confirmed by the location of the fracture during the static tensile test in the BM. In the case of welded joints of S700MC, S1100QL, and S1100MC, fracture initiation is localized to the SCHAZ. Furthermore, every welded joint fulfilled the standard requirements for the minimum YS and UTS.

Based on the tests performed, it can be concluded that all welded joints of steels S690QL, S960QL, S700MC, and S960MC met the requirements for the bend test. In the case of steels S1100QL and S1100MC, the test specimens did not meet the requirements specified in the standard. According to EN ISO 5173, the test is considered satisfactory if the sample is bent to an angle of 180° without the appearance of cracks larger than 3 mm. If cracks with a length of more than 3 mm appear during the test, the test is considered unsatisfactory.

The line microhardness measurements showed differences in the individual subzones of the welded joints. From the graphical plots (Fig. 7), a significant increase in hardness compared to the BM can be observed in the WM, CGHAZ, and FGHAZ for the welded joints of S690QL, S960QL, and S960MC steels. Similar hardening of the microstructure also occurred in the weld joints of S1100QL and S1100MC steels; however, due to the high hardness of the BM, there was observed no increase in hardness compared to the BM.

Fig. 7
figure 7

Microhardness distribution across the welded joints: a S690QL, b S960QL, c S1100QL, d S700MC, e S960MC, f S1100MC

Full size image

The wsz was measured using Fig. 7. The measured values of the wsz and the decrease in hardness are shown in Table 6.

Table 6 Values of the width of the soft zone (wsz) and max. hardness drop of assessed weld joints
Full size table

Based on Fig. 2, it is possible to observe how an increase in the heat input leads to an increase in the width of the HAZ. With greater heat input, the fusion zone’s characteristic shape does not alter appreciably. The measured values of the visible HAZ and WM widths are shown in Table 7. With the given values, it must be considered the fact that the SCHAZ is not visible through LOM, although in many welded joints, the SCHAZ is the area with the greatest decrease in hardness. In Fig. 8, it is possible to observe the microstructure of the zones with the greatest hardness drop. For all weld joints, this is the zone between the ICHAZ and SCHAZ.

Table 7 Values of the width of the HAZ and WM
Full size table
Fig. 8
figure 8

Microstructure of the ICHAZ: a S690QL, b S960QL, c S1100QL, d S700MC, e S960MC, f S1100MC

Full size image

Figure 9 shows the microstructure of the CGHAZ by LOM (left) and SEM (right). The values of the mean PAG size and the width of the CGHAZ are given in Table 8.

Fig. 9
figure 9figure 9

Microstructure of the CGHAZ: a S690QL, b S960QL, c S1100QL, d S700MC, e S960MC, f S1100MC

Full size image
Table 8 The mean PAG size and the width of the CGHAZ of the assessed steels
Full size table

3.1 Effect of the chemical composition on the properties of WM and CGHAZ

In the laser welding process, conditions of an increased cooling rate (above 120 °C·s−1) occur in all welded joints. At such cooling rates, the formation of martensite is assumed, the properties of which, in addition to the cooling rate, depend on the chemical composition of the steel. The fraction of elements that enhance hardenability (C, Mn, Mo, Cr, etc.) primarily determines the volume fraction of martensite and its overall hardness. If simplified, the influence of these elements on hardenability can be described by the CET parameter. In Fig. 10, the effect of CET on the WM hardness is shown. For this purpose, the WM hardness was determined as the average of 10 WM hardness values closest to the weld axis (1-mm section of the WM).

Fig. 10
figure 10

Effect of CET on the WM hardness

Full size image

Furthermore, it is possible to observe a comparable, if not identical, impact of the chemical composition on the hardness in the CGHAZ and ICHAZ. However, this phenomenon occurs only under conditions of a rapid cooling rate (comparable to laser welding). The chemical composition also affects the size of the PAG in the CGHAZ. According to this theory, the precipitates of Nb, V, and Ti create an obstacle to PAG growth that develops during heating. As a result, their higher content can inhibit considerable PAG growth and narrow the CGHAZ. The mean PAG size and CGHAZ width measurements verified these hypotheses satisfactorily. In Fig. 11, the effect of Nb, V, and Ti content on the increase in the width of the CGHAZ and the mean PAG size in this zone can be observed.

Fig. 11
figure 11

Effect of Nb and Ti content on the PAG size and width of the CGHAZ

Full size image

3.2 Effect of \({\upsigma }_{\mathrm{WM}}^{\mathrm{T}}\) and \({\upsigma }_{\mathrm{HAZ}}^{\mathrm{T}}\) on the mechanical properties of welded joints

The level of contribution of the chemical composition can be demonstrated by the influence of \({\upsigma }_{\mathrm{WM}}^{\mathrm{T}}\) on the overall mechanical properties of the welded joints. There was a correlation established between the relative UTS of the welded joint and the \({\upsigma }_{\mathrm{WM}}^{\mathrm{T}}\)/\({\upsigma }_{\mathrm{BM}}^{\mathrm{T}}\) and \({\upsigma }_{\mathrm{HAZ}}^{\mathrm{T}}\)/\({\upsigma }_{\mathrm{BM}}^{\mathrm{T}}\) (Figs. 12 and 13). Research by Mochizuki et al. [32] and Maurer et al. [23], for instance, confirmed this correlation. Mochizuki et al. [32] verified that the overall strength of the welded joint increases in proportion to the value of \({\upsigma }_{\mathrm{WM}}^{\mathrm{T}}\). In this case, the overall strength is expressed through the value of the relative strength of the weld joint, and thus through the ratio between the UTS of weld joint (see Table 5) and UTS of BM (see Table 2). To determine the value of \({\upsigma }_{\mathrm{WM}}^{\mathrm{T}}\), \({\upsigma }_{\mathrm{HAZ}}^{\mathrm{T}}\), and \({\upsigma }_{\mathrm{BM}}^{\mathrm{T}}\), Eq. 8 was used. \({\upsigma }_{\mathrm{BM}}^{\mathrm{T}}\) was calculated using hardness value of the BM (see Table 2).

Fig. 12
figure 12

Effect of \({\upsigma }_{\mathrm{WM}}^{\mathrm{T}}\)/\({\upsigma }_{\mathrm{BM}}^{\mathrm{T}}\) on the relative UTS of weld joints

Full size image
Fig. 13
figure 13

Effect of \({\upsigma }_{\mathrm{HAZ}}^{\mathrm{T}}\)/\({\upsigma }_{\mathrm{BM}}^{\mathrm{T}}\) on the relative UTS of weld joints

Full size image

It can be observed that as \({\upsigma }_{\mathrm{WM}}^{\mathrm{T}}\) increases, correspondingly, the relative UTS of welded joints increases. The welded joints of S1100QL, S1100MC, and S700MC show the lowest relative UTS values, reflected by the smallest value of the \({\upsigma }_{\mathrm{WM}}^{\mathrm{T}}\)/\({\upsigma }_{\mathrm{BM}}^{\mathrm{T}}\).

According to Fig. 13, it is possible to observe the dependence between the relative UTS and \({\upsigma }_{\mathrm{HAZ}}^{\mathrm{T}}\)/\({\upsigma }_{\mathrm{BM}}^{\mathrm{T}}\) values. The decrease of the relative UTS was most noticeable for S1100MC and S1100QL steels (welded joints with highest \({\upsigma }_{\mathrm{HAZ}}^{\mathrm{T}}\) decrease).

The results of the static tensile tests on the welded joints (Table 5) also showed a significant decrease in the elongation of the welded joints. The decrease in elongation compared with the original BM values ranged from 29 to 62%. The most common reason for the decrease in elongation is often associated with microstructural changes that promote brittleness, which are accompanied by the formation of phases that reduce the ductility of the material (e.g., martensite and M-A compound). However, as in the case of the overall YS and UTS, the issue of reduced elongation is more complex, and thus, the overall value A50 of the welded joint is also influenced by other factors. According to Figs. 14 and 15, it is possible to observe the influence of ratios \({\upsigma }_{\mathrm{WM}}^{\mathrm{T}}\)/\({\upsigma }_{\mathrm{BM}}^{\mathrm{T}}\) and \({\upsigma }_{\mathrm{HAZ}}^{\mathrm{T}}\)/\({\upsigma }_{\mathrm{BM}}^{\mathrm{T}}\), respectively on the overall elongation of welded joints. Provided values demonstrate that as the soft zone’s strength decreases, the elongation reduces as well.

Fig. 14
figure 14

Effect of \({\upsigma }_{\mathrm{WM}}^{\mathrm{T}}\)/\({\upsigma }_{\mathrm{BM}}^{\mathrm{T}}\) on the relative A50 of welded joints

Full size image
Fig. 15
figure 15

Effect of \({\upsigma }_{\mathrm{HAZ}}^{\mathrm{T}}\)/\({\upsigma }_{\mathrm{BM}}^{\mathrm{T}}\) on the relative A50 of welded joints

Full size image

3.3 Effect of the XSZ on the mechanical properties of welded joints

One of the most crucial assessed parameters for the HAZ of HSLA steels that are susceptible to softening is the relative thickness of the soft zone (XSZ). The given parameter also influences the welded joint’s overall behavior of deformations during loading, which has an impact on the YS, UTS, and A50 values [24]. Theoretical knowledge makes it obvious that as the XSZ decreases, the YS and UTS values increases. Moreover, in circumstances where the value of XSZ is small enough, there is no observed decrease in the YS and UTS [25, 26]. These conclusions were also experimentally confirmed based on the given results (Fig. 16).

Fig. 16
figure 16

Effect of the relative thickness of the soft zone (XSZ) on the relative YS/UTS

Full size image

It was recently confirmed that when there is not any (or negligible) wsz, there is no decrease in the relative YS and UTS of the weld joints. The only cases where there was a noticeable decrease in the relative YS and UTS were in the welded joints made of S1100QL and S1100MC steels, where the relative YS and UTS, respectively, dropped to values between 0.90 and 0.96. It is obvious that the XSZ also has a significant effect on the decrease in elongation. It is possible to observe how the value of A50 decreases as the XSZ increases from the provided graph (Fig. 17).

Fig. 17
figure 17

Effect of the relative thickness of the soft zone (XSZ) on relative elongation A50

Full size image

It is evident from Fig. 17 that the decrease in A50 occurred even in cases when the soft zone’s width was negligible (wsz ≈ 0 mm).

4 Summary

This investigation demonstrated several specifics corresponding to the welding of HSLA steels by LBW technology. From a metallurgical perspective, the most notable changes are caused by the high cooling rate, which effects all HAZ subzones and the WM’s overall properties. This was apparent in the weld joints of S960MC, S960QL, and S690QL steels, where the higher cooling rate was demonstrated in the WM, CGHAZ, and FGHAZ increase in hardness. For these steels, the hardness increased from 19 to 45% compared to the BM. Such an increase in hardness can be considered a desirable phenomenon if the weld joint is subjected to static loading.

Examining the effect of the chemical composition, it was determined that the elements of C, Mn, Mo, Cr, Cu, and Ni, whose impact on hardenability in this study was expressed by the CET parameter, are responsible for an increase in hardness/strength in the WM, CGHAZ, and FGHAZ. The influence of the chemical composition was also demonstrated on the size of the PAG and on the width of the CGHAZ. The elements Nb, V, and Ti play an important role in this instance because they prevent PAG growth, which also reduces the width of the CGHAZ. Suppression of such PAG growth may be desirable in the case of weld joints welded with lower cooling rates (e.g., MAG welding), as by preventing the PAG growth, the decrease of YS due to decrease in σg can be reduced.

The primary indicator of the softening effect—a decrease in hardness—was most visible to steels S1100QL, S1100MC, S700MC, and partially S960MC. The strength ratio \({\upsigma }_{\mathrm{HAZ}}^{\mathrm{T}}\)/\({\upsigma }_{\mathrm{BM}}^{\mathrm{T}}\) in the provided work was used to express the influence of this decrease on the overall mechanical properties of the weld joints. From the review of the literature, it is also evident that as the XSZ increases, there will likely be a significant decrease in the YS, UTS, and A50. These statements were completely verified in the case of experimental welded joints, where the influence of XSZ on the value of relative YS, UTS, and A50 was confirmed.

Based on the findings obtained and their analysis, it can be stated that the values of strength and elongation of welded joints can be influenced by the obtained value of \({\upsigma }_{\mathrm{WM}}^{\mathrm{T}}\), \({\upsigma }_{\mathrm{HAZ}}^{\mathrm{T}}\), and XSZ. The results showed that the increase in \({\upsigma }_{\mathrm{WM}}^{\mathrm{T}}\) is significantly influenced by the CET parameter, the cooling rate, and the initial hardness of the BM. Therefore, the highest possible cooling rate should be utilized while welding HSLA steels. However, in this particular instance, it is also necessary to consider the manufacturer’s recommendations in order to prevent hydrogen-induced cracking.

It needs to be stated that one of the technologies that can produce welded joints with the highest possible YS, UTS, and A50 is a LBW method. However, a paradox occurs in the case of S1100MC welded joints: experimental laser-welded joints indicate considerable softening, while in the study of GMAW welds published by Amraei et al. [33, 41], the HAZ exhibits hardening rather than softening. Further research is therefore needed to determine the causes of this phenomenon.

5 Conclusions

Based on the knowledge discussed in the introduction and the experiments carried out, several findings were made with respect to the weldability of HSLA steels with a LBW. Knowledge regarding the influence of chemical composition on the selected properties of welded joints can be summarized as follows:

  • With increasing value of the CET parameter, the WM’s hardness increases;

  • With increasing content of Nb, V, and Ti, the PAG size in the CGHAZ decreases;

  • With increasing content of Nb, V, and Ti, the CGHAZ width decreases.

Several findings were also made concerning the influence of the mechanical properties of weld joint individual zones on the overall mechanical properties of the welded joint. These can be summarized as follows:

  • The relative UTS of welded joints increases with the increasing ratio of \({\upsigma }_{\mathrm{WM}}^{\mathrm{T}}\)/\({\upsigma }_{\mathrm{BM}}^{\mathrm{T}}\);

  • The relative UTS of welded joints increases with the increasing ratio of \({\upsigma }_{\mathrm{HAZ}}^{\mathrm{T}}\)/\({\upsigma }_{\mathrm{BM}}^{\mathrm{T}}\);

  • The relative A50 of welded joints increases with the increasing ratio of \({\upsigma }_{\mathrm{WM}}^{\mathrm{T}}\)/\({\upsigma }_{\mathrm{BM}}^{\mathrm{T}}\);

  • The relative A50 of welded joints increases with the increasing ratio of \({\upsigma }_{\mathrm{HAZ}}^{\mathrm{T}}\)/\({\upsigma }_{\mathrm{BM}}^{\mathrm{T}}\);

  • The relative YS and UTS, respectively, of welded joints increases with the decreasing value of XSZ;

  • The relative A50 of welded joints increases with the decreasing value of XSZ.

According to mechanical tests performed and evaluated according to valid standards, it is possible to draw following conclusions:

  • The YS and UTS values of the welded joints reached the minimum values required by standards for all welded joints;

  • The results of the bend test were in accordance with the requirements for all welded joints, except for the welded joints of S1100QL and S1100MC steels, for which the test results were not satisfactory;

  • For all welded joints, the values of YS and UTS decreased to the level of 90–96% compared to the BM;

  • For all welded joints, the values of A50 decreased to the level of 23–79% compared to the BM.