*3.7. Discussion about the Results and Corresponding Thermal Cycles*

After collecting all the results and analyzing them individually, it is now necessary to gather this information and relate it to the thermal cycles initially performed.

It remains clear that the welding process generates martensite in the weld and HAZ, increasing the mechanical resistance, but limiting the ductility, as can be seen in the tensile tests and optical microscopic analysis carried out on sample P\_APP\_T00. This is valid for both room temperature and elevated temperature, which creates limitations regarding the use of this kind of joint in their traditional applications, heat exchangers, due to creep phenomenon. The microhardness measured clearly confirms these results, showing an increase in hardness in the MZ and HAZ of these samples. However, regarding the bending tests, no cracks have been reported, but the high hardness presented in the weld moves the bending center from the weld to the base material. Thus, it is clear that martensite needs to be tempered after welding, and some heat treatments can help to decrease the stress accumulated in and close to the weld, improving the overall mechanical behavior of the joints.

Sample P\_APP\_T01 was prepared following the codes' recommendations, including preheating, post-heating, and PWHT. The procedure is time consuming and costly, but the results are excellent. The mechanical resistance remains within the range of values taken as acceptable for this kind of joint, and the ductility is not severely compromised, remaining as well in an acceptable value, although it is more affected at elevated temperatures than at room temperature. The microstructure shows the presence of tempered martensite, and the hardness values are lower than in the case of the P\_APP\_T00 samples. As expected, no cracks have been developed during or after the bending tests. Thus, the main disadvantage of this procedure is the time spent to perform it and the energy consumed, making this process less friendly to the environment.

Sample P\_APP\_T02 was prepared saving the post-heating treatment and keeping the pre-heating and PWHT. The phases' transformation time included in the P\_APP\_T01 sample was suppressed in this and the following thermal cycles. This thermal cycle gave interesting mechanical resistance values at room temperature, but it did harm the ductility a little. However, analyzing the mechanical resistance at elevated temperature, a severe decrease in the yield strength and ultimate strength can be observed, presenting as well the lowest elongation value. Thus, suppressing the post-heat-treatment, the properties at elevated temperature are severely affected. The grain seems to not be significantly affected by the thermal cycles in the HAZ, and regarding the MZ, it seems to be less affected than the other samples. Therefore, the absence of post-heat-treatment affects the properties at elevated temperature, which is drastic for this kind of material regarding the typical applications referred above. It is worth noting that no cold cracking effect was felt in the samples due to the waiting time imposed between the welding process and PWHT.

The P\_APP\_T03 sample followed the same principles of the P\_APP\_T02 sample, but in this case, a post-heat-treatment was included after the welding process and before the six-month waiting time. These samples presented the lowest ultimate strength values at room temperature, although the values relative to tensile at elevated temperature were improved relative to the similar P\_APP\_T02 sample. This is the main concern of these samples. It is also worth noting that these samples present a very good ductility behavior at room temperature and the best ductility at elevated temperature. At this point, it needs to be referred that the only difference between the procedure applied to these samples and the procedure used for sample P\_APP\_T01 is the inclusion of the six-month waiting time after the post-heat-treatment and before the PWHT. However, the inclusion of this waiting time was detrimental to the mechanical properties at room temperature, although other properties were improved, such as the ductility.

Regarding the P\_APP\_T08 sample, the thermal cycles used are very similar to the thermal cycles applied to P\_APP\_T01, having as the only difference suppression of the phases' transformation time. This suppression is important in terms of productivity and energy saving. The results obtained based on these samples allow concluding that there is a good balance between the mechanical resistance obtained and the ductility presented by these samples; both for room temperature and elevated temperature, the grain is a little more acicular than in the other samples, which is not translated in terms of hardness, which presents values extremely close to the average results obtained for samples P\_APP\_T01, P\_APP\_T02, and P\_APP\_T03. Moreover, regarding the bending tests, no cracking or its initiation has been detected. Thus, although these samples show a slight decrease in mechanical resistance and ductility, this is a good alternative to the thermal cycle P\_APP\_T01.

In summary, Table 10 gives a brief overview of the thermal cycles applied to each set of samples and the corresponding qualitative results, following the criteria pointed out at the bottom of Table 11. It is worth noting that samples P\_APP\_T08 and P\_APP\_T01 present slight differences in the procedure, as well as in the final results, not compromising the behavior of welds produced in this material in its main applications.

Regarding the main parameters provided by tensile tests, and in order to make a decision regarding the selection of the best strategy based on quantitative data, Table 11 was elaborated classifying from 1 to 4 (1 = worst value and 4 = best value) the different quantitative data obtained from that test. As can be seen in Table 11, after the recommended procedure corresponding to the P\_APP\_T01 sample, the best alternative is the P\_APP\_T08 strategy, with a sum of 6.1 points. This confirms the analysis previously performed.


**Table 10.** Summary of the thermal cycles applied and corresponding results by test type.

(1) Protected cooling until reaching room temperature; (2) PWHT performed 6 months after welding; Rating Scale: Not Acceptable–Acceptable–Good–Very Good.


**Table 11.** Summary of the thermal cycles applied and corresponding results by test type.
