*3.1. Non-Destructive Test Results*

Some of the initially performed tests were merely qualitative, thus providing information about whether the samples were ready to pursue the analysis or not. These tests were already described before in this work.

## *3.2. Hardness and Microhardness Analyses*

The hardness tests performed allowed assessing the hardness profile across the joint, from the base material 1 (BM1) to the base material 2 (BM2), passing by the thermal-affected zones 1 and 2 (HAZ1 and HAZ2) and the melted zone (MZ). The microhardness values can be observed in Table 6, and the profiles corresponding to the different samples can be seen in Figure 5, resulting from the

average values of five samples measured in the same places (15 indentations per sample), taking into account the central area of each sample. The calculated standard deviation can be observed in Table 6 as well. The letter B included in the label of samples P\_APP\_T02 and P\_APP\_T03 in Figure 5 represents the values obtained before the PWHT was carried out six months after the welding and corresponding heat treatments performed in the first stage of the samples' preparation. As expected, the hardness values before PWHT are higher in the HAZ and MZ, because the martensite is not tempered yet, being extremely hard in that state. After PWHT, the values achieved are within the expected range. Since the sample P\_APP\_T00 was not subjected to heat treatments, its hardness is extremely high in the MZ, as well as in the HAZ. Thus, sample P\_APP\_T00 does not fulfill the properties required by the usual applications of this kind of material, and the procedure used in this sample can be considered inadequate. It is worth noticing that the remaining samples show hardness values within the usual range for this material; thus, at this stage, excluding the P\_APP\_T00 sample, the other samples can be considered for the next tests. Performing this test only lets us validate in a first moment, in the shop floor, the ability of the work pieces to advance to the next stages, because the accurateness of the hardness measuring equipment is ±40 HV. Thus, in order to evaluate the surface hardness of the samples with adequate accuracy, microhardness tests were performed using the microhardness measuring equipment previously described. The microhardness profile can be seen in Figure 6, showing clearly that sample P-APP\_T00 presents a different hardness behavior resulting from the lack of heat treatments. In this case, the microhardness measurements were performed only after the PWHT treatment; thus, there are no curves corresponding to this situation, as shown in Figure 5. The values reported in the curves can be considered common for this kind of material and joint. The increased hardness reported in the MZ and HAZ is typical (250–280 HV10) as well in these conditions, and the values registered are within the range of acceptable values (200–275 HV10, but they can reach 300 HV10 without special concerns). Excluding the sample P\_APP\_T00, which presents a singular behavior due to the absence of heat treatments before the PWHT applied to this sample, the samples were mainly constituted of non-tempered martensite, which was hard and brittle. The pattern followed by the remaining samples is very similar among them, with microhardness values of around 230 HV10 in the base metal and slighting increasing across the HAZ, reaching steady values between 250 and 280 HV10 in the MZ. Obviously, sample P\_APP\_T00 is not acceptable, reaching microhardness values out of the range of acceptable values. This corroborates the unconditional need for applying heat treatments to this base material when welding is included in the manufacturing processes. According to ISO 18265 (Metallic materials—Conversion of hardness values), it is possible to estimate the ultimate strength through hardness values [34]. This is because both hardness and ultimate strength are indicators of the mechanical resistance of a metal to plastic deformation, and they are approximately proportional. However, this proportionality is not valid for all metals. For this case, it may be pointed out that the breaking stress in MPa is 3.13 times higher than the hardness in HV.


**Table 6.** Surface microhardness of the different samples (HV10). HAZ: heat-affected zone, MZ: melted zone.

**Figure 5.** Hardness profile of the different samples across the welded and neighboring areas. The label B in some graphs represents the values obtained before the PWHT was carried out 6 months after the welding.

**Figure 6.** Microhardness profile of the different samples across the welded and neighboring areas.

The microhardness tests were also performed having as the main goal identifying soft spots in the cross-section, i.e., very small areas where the hardness is lower than in the base metal or in the fine grain region of the HAZ, which is also known as the intercritical region. The summary of the results by the crossed region can be seen in Table 7. The intercritical region is located in the connection zone of the HAZ, which is very close to the base material. This is where the so-called Type IV crack usually appears, which often occurs both at the manufacturing stage and during the life of the component.


**Table 7.** Cross-section microhardness of the different samples (HV1).

Regarding the soft spot analysis, no special attention was paid to the P\_APP\_T00 samples, due to its non-heat-treated condition. Taking into attention the P\_APP\_T01 samples, lower microhardness values were identified in the range of 7.5 to 10 mm from the center line of the weld. Indeed, this is the most critical area in the weld cross-section because this is a HAZ that is close to the base material, which is an area that is well known as the region where the grain size is finer and the hardness can drop about 20 HV of the substrate's usual level. Regarding the samples P\_APP\_T01, some microhardness measurements provided values under 200 HV (187 HV, while the base material presents 220–230 HV) in the intercritical area, indicating that there are soft spots that can induce Type IV cracks. The samples P\_APP\_T02 and P\_APP\_T03 did not present values under 200 HV. Indeed, the lowest value obtained in both samples was 204 HV, showing that these samples do not present soft spots and are not prone to develop Type IV cracks in the intercritical area. The sample P\_APP\_T08 presented just one (one in 14) value below 200 HV (170 HV), which was located into the HAZ and close to the surface. Since all the other results were close to 220 HV in this sample, the result was not valued also because in the root and middle indentations rows, the same effect was not felt. Thus, it was considered that sample P\_APP\_T08 presents good conditions to avoid the development of Type IV cracks, presenting less concerns than sample P\_APP\_T01.

## *3.3. Welds Chemical Analysis*

In order to check the final composition of the weld, optical spectroscopy was used to obtain the weld composition. However, as the chemical composition of the base material and filler metal are very similar, it would be expected that the weld follows a chemical composition very similar to the base material and filler metal. The results can be seen in Table 8, where the compositions related to Pipe 1, Pipe 2, and the filler metal have been taken from the information provided by the suppliers and the composition related to the weld face and root face were taken from analyses performed by optical spectroscopy. As can be observed, the chemical composition of the weld, both in the face and in the root, are very similar to the compositions provided by the suppliers to the base materials and filler metal. Thus, the dilution in the welding process will not significantly affect the weld composition.


**Table 8.** Chemical analysis of the materials used in the welding and the weld composition (wt %).
