4.1. Porosity and microstructure analysis
Figure 7 presents the microstructures of the analyzed samples and Table 4 presents the result obtained in this analysis.
Table 4
Classification of porosity types - SH classes compared to NbC_Ni-967 and WC-Co reference classes
Types
|
SH1
|
SH2
|
NbC-Ni_967
|
WC-Co reference
|
Type A
|
A4
|
A4
|
A6
|
A2
|
Type B
|
---
|
---
|
B2
|
---
|
Type C
|
---
|
---
|
---
|
---
|
There was a reduction in porosity of SH classes compared to NbC_Ni-967. With the addition of Sinter-HIP there was an elimination of Type B porosity (> 10 to ≤ 25 µm) and an improvement in Type A porosity (pores ≤ 10 µm), but the results are lower than the WC-Co reference class. In the analyses carried out previously, all NbC-Ni classes, produced in vacuum sintering and vacuum sintering + argon, had presented Type B2 and B6 porosity and Type A6 porosity [21].
Classes SH1 and SH2 presented “Ni lakes”, and the same had occurred for the previously developed classes, Fig. 6. The distribution of carbides (classes SH1 and SH2) remained heterogeneous, even doubling the milling time, confirmed by the distribution of tungsten in Fig. 8 (a) (b) (c), which show tungsten concentration in regions of area “B”, compared to regions of area “A”.
Comparing the microstructures of SH1 and SH2, it is possible to observe, in Fig. 9 (d) (e), that the grain size of the SH2 class carbides is smaller than that of the SH1 class. This could be due to the cooling time in the sintering process. Class SH2 had a longer cooling time, as the Sinter-Hip oven was kept on during this process, while for Class SH1 the oven was turned off, resulting in a greater drop in temperature during this process.
Two phenomena occurred for the differentiation of grain size between classes SH1 and SH2: the dissolution of secondary carbides, which restricts the growth of NbC grains, and the longer time, which could have increased the size of NbC grains by coalescence. From the result obtained, as shown in Fig. 9 (d) (e), the first phenomenon prevailed over the second, as the grains of SH2 are smaller than those of SH1, even with the cooling time in the sintering process of SH1 being shorter than that of SH2.
Therefore, the microstructure of SH classes, even with the reduction of porosity, remains heterogeneous compared to the WC-Co reference class.
4.2. Binder plasticity analysis
In previous tests [23], the class which had the lowest modulus of elasticity of the binder was WC-Co reference to 192.2 GPa, while the class NbC_Ni-967 showed the lowest elastic modulus of the Ni-NbC classes (235.7 GPa). There was a variation between 235.7 and 367 GPa of the analyzed NbC-Ni classes [23]. In relation to the hardness of the binder (H) the NbC-Ni classes presented a variation between 3.93 and 5.87 GPa, while in the NbC-Ni_967 class the hardness was 4.70 GPa, but statistically equal to the WC-Co reference, which was 4.25 GPa [23].
It was possible to observe that classes SH1 and SH2 had the same behavior as the analyzed parameters. The two classes produced with the addition of the Sinter-HIP process had the same statistical behavior in relation to the NbC-Ni_967 class, but the dispersion of the results of the analyzed parameters was much smaller in the SH classes when compared to the NbC-Ni_967 class, as can be seen in Table 5. This may indicate that the addition of Sinter-HIP reduced the variability of the results of the analyzed parameters.
Table 5
Average parameters obtained via nanoindentation: SH classes compared to NbC-Ni_967 and WC-Co reference classes
Variables
|
SH1
|
SH2
|
NbC-Ni_967
|
WC-Co reference
|
H (GPa)
|
5.03 ± 0.18
|
4.97 ± 0.21
|
4.70 ± 0.71
|
4.25 ± 0.21
|
E (GPa)
|
233.3 ± 19.4
|
232.0 ± 24.6
|
235.7 ± 35.9
|
192.2 ± 16.0
|
With the introduction of the Sinter-HIP process, there was a reduction in the dispersion of the results of classes SH1 and SH2 in relation to class NbC_Ni-967. There was therefore a greater differentiation of results between classes SH in relation to WC-Co reference than the classes NbC_Ni-967 and WC-Co reference.
4.3. Machining analysis
Machining experiments were carried out with classes SH1 and SH2 under the same conditions as the previous tests. Figure 10 shows the comparative performance of each class developed in the Sinter-HIP process (SH1 and SH2) compared to the WC-Co reference and NbC-Ni_967, which was the class that had presented the best result in the machining tests previously, in terms of average lifetime of the cutting edge for a defined flank wear (VB = 0.3 mm).
The grades sintered under vacuum and argon with the addition of the Sinte-HIP process, SH1 and SH2, had a different average performance, in terms of the lifetime of the cutting edge during the machining process. Class SH1 had an average lifetime per cutting edge of 28.8 ± 2.1 minutes, while class SH2 had an average lifetime of 43.8 ± 1.6 minutes. Therefore, grade SH2 had a cutting edge lifetime greater than SH1 by 1.8x.
The main difference between these classes in the manufacturing process was the cooling time in the Sinter-HIP process. Class SH1 had a shorter cooling time with a greater drop in temperature compared to class SH2. Due to this difference in the process, the microstructures they showed were different, and the SH1 had larger NbC carbide grains than the SH2 class, Fig. 9 (d) (e). Hardness and fracture toughness were also different, as described in Table 2.
Comparing the SH classes with the NbC-Ni_967 class, the SH1 class presented a 41% lower performance and the average lifetimes per cutting edge were, respectively, 28.8 and 48.9 minutes. The main difference between them was the hardness and fracture toughness; the SH1 grade had a hardness of 1,402 ± 25 HV30 and KIC = 9.2 ± 0.3 MPa.m1/2, while the NbC-Ni_967 grade had a hardness of 1,460 ± 29 HV30 and KIC= 8.9 ± 0.4 MPa.m1/2. However, the lifetime dispersion of the cutting edge of SH1 was 8%, while the NbC-Ni_967 was 15%, representing a dispersion reduction in the order of 3.5 times. The main difference between the two classes was the reduction in porosity of SH1 as shown in Table 4.
Class SH2 statistically showed equal performance to NbC-Ni_967, with 43.8 and 50.2 minutes respectively. However, there was a reduction in dispersion of 4.6 times, from 15–3.7% for the SH2 class. The hardness of class SH2 was lower than class NbC-Ni_967, 1,460 ± 29 HV30 and 1,407 ± 29 HV30 respectively. The fracture toughness was statistically equal; class SH1 was 9.0 ± 0.3 MPa.m1/2 and class NbC_Ni-967 was 8.9 ± 0.4 MPa.m1/2. However, SH2 had a lower porosity level than NbC-Ni_967 (Type A: A4; Type B: no pores and Type A: A6; Type B: B2, respectively).
Comparing WC-Co grade reference with SH1 class showed a statistically equal performance in the average lifetime of the cutting edge of 25.9 minutes for the WC-Co and 28.8 minutes for the SH1 class. However, the dispersion of WC-Co was 2.6 times lower than SH1 (3.2% and 8.0% respectively).
Class SH2 outperformed the average lifetime of the cutting edge by 68.9%, 43.8 versus 25.9 minutes respectively. However, the dispersion of reference WC-Co was 3.2% while the SH2 was 3.7%. Therefore, with the addition of the Sinter-HIP step there was a reproducibility in the wear behavior. SH2 had a dispersion 1.15 times higher than WC-Co reference and 75.5% lower than NbC-Ni_967while the NbC-Ni_967 class had a dispersion 4 times higher than the WC-Co reference as can be seen in Fig. 11 (a)(b)(c).
Comparisons were made between the relative SH-Ni_967 NbC and WC-Co benchmark for the wear mechanisms that acted during the turning process, shown in Fig. 12.
Class SH1 showed more regular abrasive wear on the cutting edge on the main face compared to NbC-Ni_967. This could be due to the porosity reduction with the addition of the Sinter-HIP process in the SH1 class. Grade SH1 had less spalling on the edge of the edge. These spalls were due to adhesive wear that occurred during the turning process. Adhesive wear caused more damage in the NbC-Ni class than in the WC-Co class because for this system the physicochemical interactions between carbides and binder are different and the microstructure analysis showed that there is greater contiguity in the NbC-Ni classes than the WC-Co reference class.
Comparisons between the SH1 and SH2 grades with the WC-Co reference grade, with the addition of the Sinter-HIP process, show there was a reduction in adhesion wear mechanisms and a better uniformity of abrasion wear. However, compared with the WC-Co reference class, the abrasion is not as uniform as can be seen in Fig. 10. This may be due to the fact that the WC-Co reference class presents a more homogeneous microstructure than the SH classes. The SH classes have a lot of contiguity and a heterogeneous distribution of carbides.
However, regarding the adhesion mechanism, which is the main cause of spalling of the cutting edge, the SH2 class presented damage very similar to the WC-Co reference. While in the SH1 class the damage presented was greater than the reference class, there was less damage than the NbC-Ni_967 class, demonstrating that the addition of the Sinter-HIP process reduced the effect of the adhesion mechanism and left the abrasion wear more uniform.