**5. The TTT Diagram, CPT Temperature, and the Ferrite Number Determined for Duplex Steels**

In high-alloy duplex steels of older generations (SDSS and HDSS), the precipitation start time is short [34]. Exceeding the activation energy of the secondary phase precipitations, expressed by the intersection of the cooling line with the TTT curve increases the sensitivity to accumulation of heat exposure effects resulting from welding the steel. It is because the energy of the precipitation reaction is always lower than the activation energy of this process [35].

The formation of secondary phases in duplex steels, which are hard and brittle and thus harmful, takes place in two temperature ranges (Figure 7). The upper curve, corresponding to the range of 600–1050 ◦C, shows the precipitation of nitrides, carbides, and intermetallic phases as a result of prolonged thermal exposure of steel due to insufficiently rapid cooling. Thus, the welding with high linear energies facilitates the transformation of δ to δ + γ, which is favorable but also increases the probability of the secondary phase formation within the ferrite at insufficiently fast cooling. The consequence of this is the need to strictly adhere to the recommended linear energies of welding and to control the cooling process of the joint during welding. The lower TTT curve, corresponding to the temperature of 300–550 ◦C, shows the remaining secondary changes in the steel microstructure. The most important is the change of δ-ferrite to acicular secondary α -ferrite, significantly reducing the ductility and toughness of steel. The lower limit of the occurrence of the unfavorable α -ferrite determines the highest temperature of the long-term thermal exposure, which is about 300 ◦C.

The advancement of harmful changes in the microstructure, lower ductility, and corrosion resistance depends on the total heat exposure time to both critical ranges on the TTT curves (Figure 8a,b). The exposure of the lowest alloyed LDSS to temperatures between 800–1000 ◦C may exceed 10 h without the release of harmful phases. However, this time for the standard DSS 22% Cr steel is reduced to about 30–60 min, and for the HDSS steel is even 5–10 min [36]. The presence of 0.5% secondary phases at the boundaries of δ-ferrite causes a dramatic decrease in the breaking work (Figure 8a). The same volume of the harmful and easily released Cr2N phase lowers the critical pitting temperature (CPT) by up to 20 ◦C (Figure 8b).

**Figure 8.** Changes in the microstructure lowering ductility and corrosion resistance of duplex steels, including (**a**) breaking energy KV as a function of intermetallic phase content in duplex steels (according to [37]), (**b**) CPT (Critical Pitting Temperature) index as a function of the content of intercrystalline Cr2N precipitates (according to [38]).

Nitrogen in duplex steel increases the kinetics of austenite formation (Figure 9). With increasing nitrogen content, the high-temperature equilibrium δ + γ area expands towards lower Ni concentrations, and the temperature of austenite formation increases. It may increase even to the temperature of liquids, which means crystallization of a small part of austenite directly from the liquid metal. A significant increase in the transformation rate δ → δ + γ occurs, which, with sufficiently rapid cooling, makes it possible to avoid the effects of shifting the TTT curves on the time axis towards the origin of the coordinate system. For this reason, nitrogen-rich steels can be cooled in the temperature range of 800–1200 ◦C faster, without the fear of exceeding the final ferrite content above the permissible level (70%). The addition of N2 to the forming and shielding gases in Tungsten Inert Gas Arc Welding (TIG, 141) is of fundamental importance for obtaining a properly balanced weld microstructure.

**Figure 9.** TTT diagram for DSS 25% Cr duplex steels with different alloyed nitrogen content (according to [28]).

The balanced microstructure of duplex steel is obtained in a two-stage supersaturation heat treatment. The first stage is homogenizing annealing. It is most effective at the temperature of maximum nitrogen solubility in solid solution, which, depending on the steel grade, is 1050–1150 ◦C (Figure 10a,b). At this temperature, harmful secondary precipitates in δ-ferrite dissolve, and the released atomic nitrogen diffuses into austenite since its solubility in the interstitial solution is many times greater (see Table 1). This applies especially to thick-walled elements made of the SDSS and HDSS steels, where the practical cooling rate in the critical temperature range is too low to avoid the precipitation of secondary phases. The second stage is rapid cooling of the steel in water, limiting the possibility of re-separation. As a result, the steel microstructure is balanced, i.e., the proportion of both components equals ferrite austenite <sup>≈</sup> 50%−10% 50%+10% and the resistance to pitting corrosion and also the KV impact strength is increased.

**Figure 10.** Heat treatment of duplex steel, including (**a**) solubility of Nitrogen in ferrite and austenite (according to [38]), (**b**) heat treatment of the thin-walled and thick-walled duplex steel elements (according to [23]).

The volumetric fraction of δ-ferrite in the structure can be verified experimentally either by microscopic examination (image analysis) or by magnetic methods [4,5]. Values of the ferrite number (FN) are converted into the volumetric fraction of δ-ferrite (%δ) based on the following relationship [39]:

$$\text{\textquotedblleft}\boldsymbol{\delta} = \text{0.7FN}\tag{5}$$

In this equation, %δ (%) is a volumetric fraction of δ-ferrite in the steel microstructure. If it is not possible to perform the appropriate tests, the estimation of the volumetric fraction of %δ-ferrite in duplex steel can be made using the data from the metallurgical certificates in conjunction with the following relationships [36]:

$$\text{\%\\$} = 4.01 \text{Cr}\_{\text{6\%}} - 5.6 \text{Ni}\_{\text{6\%}} + 0.016 \text{T} - 20.93 \text{ (\%\text{wt.})} \tag{6}$$

$$\text{Cr}\_{\text{eq}} = \text{Cr} + 1.73 \text{Si} + 0.88 \text{Mo (\% wt.)} \tag{7}$$

$$\text{Ni}\_{\text{eq}} = \text{Ni} + 24.55\text{C} + 21.75\text{N} + 0.4\text{Cu} \text{ (\%wt.)}\tag{8}$$

In the equations, T is the homogenizing annealing temperature, and the remaining factors are element weight fractions.

The area of the weld with the lowest resistance to pitting corrosion is the heat-affected zone (HAZ). The HAZ is particularly endangered by the time of impact near the fusion line of the temperature exceeding the δ-solvus level when the microstructure of duplex steel is a single-phase and the free growth inhibitory factor δ of the ferrite grains is missing. Lowering the temperature below the level of δ-solvus again, when the joint is cooling down, activates the δ → δ + γ reaction, which, as mentioned earlier, is a diffusion-controlled transformation. The larger the original grain size, the longer it takes to reach thermodynamic equilibrium. It is difficult to re-achieve a microstructure with a balanced proportion of both matrix phases from the developed ferrite grains. The cumulative effect of the thermal cycles of welding successive weld beads leads to an increase in the ferrite content in the HAZ at the expense of austenite. Furthermore, it leads to an increased tendency to form secondary phases from the ferrite and thus to lower both the ductility and resistance to pitting corrosion. The increased content of austenite-forming Ni in the weld and the absorption of strongly austenitic-forming N from the shielding gases intensify the kinetics of the δ → δ + γ transformation, thereby preventing the reduction in austenite content in the weld microstructure. Therefore, the selection of the parent material with the highest possible austenite content, within the limits of the correct balance of the initial duplex steel microstructure, is important for the subsequent resistance of the HAZ to pitting corrosion and its KV impact strength.
