3.1. X-Ray diffraction analysis
Figures 4a-c present the diffraction patterns obtained for the TiCN, BCN and CrAlN coatings, which are the product of the analysis on the silicon substrates (100). The diffraction lines in the patterns show an order along the 2θ axis corresponding to a face centered cubic structure (FCC), NaCl type and space group Fm3m [16–17]. The substitution mechanism predominated in the formation of the three coatings, both for TiCN [18] and BCN; the carbon atoms (C) substituted the nitrogen atoms (N), which created a Ti-ordered, B-ordered and C-N disordered system. For the CrAlN coating, the aluminum atoms (Al) were replaced by chromium atoms (Cr), creating an ordered Cr-Al system, while the nitrogen atoms (N) were located in interstitial positions. In all three coatings the crystallization process formed a NaCl type FCC structure, the Ti, B, Cr and Al atoms were arranged in the Wyckoff 4a site, on the other hand the C and N atoms were arranged randomly in the Wyckoff 4b site. The international indexing records JCPDF 00-042-1488 and JCPDF 00-035-1293 for titanium and boron carbide nitride were taken as reference, while two indexings were made, taking the structure of chromium nitride (CrN) JCPDF 00-003-1157 and aluminum nitride (AlN) JCPDF 00-025-1495, for the coating of chromium aluminum nitride [18]. It is evident in Fig. 4c, that the ternary material CrAlN was the result of the combination of CrN and AlN, which have the same NaCl type FCC crystalline structure and the space group 225- Fm3m. From the diffraction patterns of the TiCN and BCN coatings, high intensity peaks were obtained with orientation (111) belonging to the angles 2θ = 36.342° and 43.228°; on the other hand, a high intensity peak was obtained for CrAlN with orientation (200) belonging to the angle 2θ = 41.646°. Jointly, small displacements were present in the peaks (111) and (200) towards smaller angles. The results obtained for the lattice parameters (ao) in the TiCN, BCN and CrAlN coatings were the values 4.278 Å, 3.622 Å and 4.334 Å, respectively. We can establish from the analyzed results that the diffraction angles and lattice parameters were affected by the type of coating material.
3.3. Chemical Composition of Coatings by Using XPS Analysis
The Table 1 present the XPS results for all coatings, due to that the elemental signals spectra were obtained for the TiCN, BCN and CrAlN coatings, the signals for the coatings allow locating and defining the respective binding energies, which facilitate determining the chemical composition and stoichiometry of the ternary layers. For TiCN: Ti (2p3) = 458.4, N (1s) = 396.80, C (1s) = 284.8 and Si (2p) = 61.6; for BCN: N (1s) = 400, C (1s) = 285.6 and B (1s) = 192.8; and for CrAlN: Cr (2p) = 475.99, N (1s) = 396.97, Al (2s) = 119 and Al (2p) = 74. Therefore, the binding energy values of the coatings and XRD results (Fig. 4) confirm the formation of CrAlN ternary compound [19, 20]. Finally, the stoichiometry was established for the three coatings (Ti32.45-C35.83-N31.72, B48.63-C31.22-N20.15 and Cr40.27-Al38.01-N21.72) [17].
Table 1
Atomic Percent of the chemical composition and stoichiometric relation for all TiCN, BCN and CrAlN coatings from XPS results.
Coatings
|
Chemical elements
|
C
|
N
|
Ti
|
B
|
Cr
|
Al
|
N/(Ti + C)
|
N/(B + C)
|
N/(Al + Cr)
|
TiCN
|
35.83
|
31.72
|
32.45
|
--
|
--
|
--
|
0.46
|
--
|
--
|
BCN
|
31.22
|
20.15
|
--
|
48.63
|
--
|
--
|
--
|
0.25
|
--
|
CrAlN
|
--
|
21.72
|
--
|
--
|
40.27
|
38.01
|
--
|
--
|
0.27
|
3.4. Coatings Mechanical Properties by Using Nanoindentation Test
The Table 2 presents the mechanical properties for all coatings from nanoindentation results, in this Table 2 the coatings influence in the hardness and elastic modulus can be appreciated; it is also shown that the material with the lowest mechanical properties values was the TiCN (H = 28 GPa, E = 224 GPa), followed by CrAlN (H = 30 GPa, E = 335 GPa) and then BCN (H = 33 GPa, E = 251 GPa). Taking in account the last results, the elastic modulus (E) is related to the type of material, but not to the material microstructure; in this sense the (E) depend on their crystal structure and its own lattice parameter [17]. Also, the hardness difference of the three coatings is due to the different crystallographic directions analyzed from XRD results Fig. 4, therefore the BCN coatings present highest hardness in relation to TiCN and CrAlN coatings.
Table 2
Presents the mechanical properties for all TiCN, BCN and CrAlN coatings from nanoindentation results.
Coatings
|
Mechanical Properties
|
Hardness (GPa)
|
Elastic Modulus (GPa)
|
BCN
|
33
|
251
|
CrAlN
|
30
|
335
|
TiCN
|
28
|
224
|
3.3. Tension and Hardness Tests
The material used as a substrate was AISI 1045 steel and the chemical composition in the state of supply is shown in Table 3. It was necessary to previously carry out the tensile test to determine certain mechanical properties of the substrate, this was fundamental for the planning of the test and the verification of the state of supply of the material. All tests were carried out at room temperature. Table 4 shows the results obtained in the tensile and Rockwell C (HRC and Vickers (HV) hardness tests. Three (3) samples were used, made of calibrated AISI 1045 steel (in supply state), obtaining a yield strength of Sy = 787.783 MPa and a ultimate strength of Sut = 871.624 MPa. The Sy/Sut = 0.904 ratio of the material helped establish the stress relations applied in the fatigue test, which had to be less or equal to the established ratio. This ensured that the behavior of the material remained within the linear-elastic range while being subjected to the load levels used in fatigue.
Table 3
Chemical composition for calibrated AISI 1045 steel.
Standard
|
%C
|
%Mn
|
%Si
|
%P
|
%S
|
SAE 1045
|
0.43–0.50
|
0.60–0.90
|
0.15–0.25*
|
0.030 max.
|
0.050 max.
|
* SAE 1045 standard bars contain silicon as of 1 1/8".
|
Table 4
Tensile and hardness test results for calibrated AISI 1045 steel.
Samples
|
E (GPa)
|
Sy (MPa)
|
Sut (MPa)
|
Srot (MPa)
|
Hardness HRC
|
Hardness HV
|
AISI 1045
|
204.563
|
787.783
|
871.624
|
717.045
|
21.913
|
403.375
|
Standard Deviation
|
± 29.662
|
± 13.625
|
± 22.260
|
± 10.691
|
± 3.517
|
± 21.725
|
3.4. Fatigue Tests
For the fatigue test the stress-life method is used, this model is called the S-N curve or Wöhler diagrams [2, 13, 21–22]. The fatigue tests were performed on a HUNG TA Model No./Serial No.: HT-8129/2031 rotary bending machine, which applies loads that produce fully inverted sinusoidal stresses with a stress ratio of Re = -1. An equivalent load analysis was done for the fatigue machine and this, along with the applied stress levels, were necessary to define the loads that were used. Eq. 1 comes from the equation of normal bending stress, also considering that the moment of inertia of the critical cross section was circular with a diameter ds = 6.35x10− 3 m; this, together with the selected stress levels, was used to determine the weights applied in the test. The following relationships were chosen to determine the applied stress levels 70%, 65%, 60% and 55% of the ultimate strength value [2, 5, 8, 14, 15, 22].
![](data:image/png;base64,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)
Where S = normal bending stress (N/m2), M = maximum bending moment (Nm), c = distance of the element furthest from the neutral axis (m), I = moment of inertia of the area with respect to the reference axis (m4), W = applied weight (N), a = fixed distance (0.2 m) and ds = diameter of the cross section (6.35x10− 3 m).
Standards were established to carry out the fatigue test as follows: the tests were considered complete when the number of technological cycles 1x106 (for infinite life) was exceeded or when the failure of the material occurred before reaching that cycles number, Ng (for finite life). In addition, as a test validity criterion, it was determined that those in which the failure took place in the critical section of the sample would be accepted, otherwise would be discarded. ASTM 468 − 11 and ASTM 739 − 10 standards were used for the representation of the results [23–24]. The points on the graphs represent the data obtained experimentally and the continuous lines represent the trend adjustment lines; additionally, 95% confidence bands are attached. Figure 5 shows the S-N curve of the uncoated material. With respect to the levels tested on the AISI 1045 steel for the transition range, the samples exceeded 1x106 cycles for stress levels of 50% and 45% of the ultimate strength value, which correspond to stresses of 435.81 MPa and 392.23 MPa, respectively. These values obtained in the fatigue tests of the uncoated steel, as well as previous studies of substrates with coatings similar to the ones in this study, served as a reference for the determination of the stress levels applied in the transition range of the materials studied [5, 25].
Figure 6 shows the S-N curve for TiCN coated steel. An increase in fatigue resistance is observed when the graph is placed over the curve of the uncoated steel, which is consistent with previous studies conducted on metallic substrates with titanium carbo-nitride (TiCN) and titanium nitride (TiN) coatings [2, 5, 8, 22]. For the TiCN-coated steel, the increase in the 1x106 cycles was 9.625%. The transition range was located for stress levels of 47.5% and 45% of the ultimate strength value, corresponding to stresses of 414.02 MPa and 392.23 MPa, respectively.
The behavior of the BCN coated steel is shown in Fig. 7. Similar to the previous case, the graph is on top of the base curve (uncoated steel). Previous studies on various metallic substrates coated with tungsten nitride titanium (WTiN), tungsten titanium (WTi), tungsten nitride (WN) and zirconium nitride (ZrN) have shown an improvement in the fatigue behavior [4, 15]. There was an increase in the 1x106 cycles of 4.215%. The transition was established for stress levels of 45% and 40% of the ultimate strength value, corresponding to stresses of 392.23 MPa and 348.65 MPa, respectively.
The data obtained in the fatigue tests for CrAlN-coated steel are detailed and represented in Fig. 8. Similar to the previous cases, the graph is located on top of the curve of the uncoated substrate; however, it presented the lowest improvement compared to the other two coatings (TiCN and BCN). The increase in fatigue resistance is lower to the obtained by BCN, for the 1x106 cycles an improvement of 3.95% was obtained. The transition range was established at stress levels of 45% and 40% of the ultimate strength value, which corresponds to stresses of 392.23 MPa and 348.65 MPa, respectively.
Finally, all the curves for the uncoated steel and with the three coatings (TiCN, BCN and CrAlN) are shown in Fig. 9. For the three cases of the coated samples, an increase in fatigue resistance was found; this improvement in descending order corresponds to TiCN, BCN and CrAlN materials with 9.6%, 4.2% and 3.9%, respectively. The highest gain in fatigue life corresponds to the TiCN coating, followed by BCN; this can be explained by the presence of titanium in the material coating where titanium carbide (TiC) and nitrogen (N) were used, as well as in the final singlelayer of TiCN; providing good adhesion in the coatings and the improvement in fatigue resistance. In this sense is possible to establish that the biaxial deformation in the planes was affected by the type of coating material. The values obtained determine the presence of tensile type stresses, being the lowest of them for the TiCN coating [17]. In the XRD analysis (Fig. 4) it was determined that these materials formed a NaCl type FCC structure; therefore, the Ti and B atoms were located in the Wyckoff 4a site, while the C and N atoms were randomly located in the Wyckoff 4b site; the presence of titanium with a high atomic radius (ra = 2Å) and boron with an intermediate atomic radius (ra = 1.17Å) in both structures, influenced that the mechanical properties (hardness and elastic modulus) (Table 2) affect the improvement of fatigue resistance. The increase of these mechanical properties together with the excellent adhesion of such coatings to the substrate, indirectly causes a delay in the nucleation process of fatigue cracks [4, 17, 26]. The above can be applied to the CrAlN coating, in which titanium is not present and has intermediate mechanical properties compared to the other coatings, which links to its low fatigue resistance.
Figure 10a, establishes the maximum alternating stress for the steels coated with TiCN, BCN, CrAlN and the uncoated steel, with values of 500.93 MPa, 476.21 MPa, 475.01 MPa and 456.91 MPa, respectively. Figure 10b, shows the reported percentages of improvement in fatigue resistance for the three materials TiCN, BCN and CrAlN in comparison to the uncoated steel.
3.5. Fractographic Analysis of Fracture Surfaces
Selected fracture surfaces of the TiCN, BCN, and CrAlN coated and uncoated AISI 1045 steel, which were subjected to fatigue, were analyzed by scanning electron microscopy (SEM). The fractographic analysis of the sample surfaces only included higher maximum alternating stresses, that is, it was done at a stress level of 70% the ultimate strength value. The purpose of the study of these fracture surfaces was to allow the identification of crack nucleation sites [26], the appearance of coating delamination, to explain the fracture process of the substrate-coating systems and particularly to identify the way in which fatigue cracks propagate after nucleation.
Figures 11a-c show the fracture surfaces for the TiCN coated substrate tested at high stress. Figure 11a shows the fracture surface of a coated sample subjected to a maximum alternating stress of 610.137 MPa and which failed at 45422 cycles. It can be stated that the failure of the specimen was the product of the simultaneous propagation of a series of cracks. Nucleation sites were located at the periphery of the surface, which led to the location of the fracture zone in the center of the fault surface. The presence of several fracture traces were examined, indicating the appearance of several nuclei at different sites along the contour of the sample. The results are consistent with information and studies reported in reference with similar deposition processes and types of coatings [2, 5, 8, 14, 15, 22]. Figure 11b shows detail in square (A), where the starting site of one of the fatigue cracks is present. The direction of crack propagation is suggested by the white arrows on the fracture surface and the direction of the arrows determine the subsequent expansion of the cracks. In this case, it can be observed that the core that originates the fatigue crack appeared at the periphery of the sample and not in the substrate, at the interface between coating and substrate. The detailed view of the area represented in square (B) is seen in Fig. 11c, which reveals the fracture of the sample at this crack initiation site.
Figures 12a-c show the fracture surfaces for the BCN coated steel tested at high stress. Figure 12a, defines the failure surface of a sample tested at 610.137 MPa and which failed at 28428 cycles. Image analysis indicates that the sample failure was established by a crack protruding from the surface contour, the fracture occurred as a product of the propagation of the A-nucleated crack, and this led to the fracture zone being located at the opposite end of the sample. Other adjacent cracks are observed on the fracture surface. Similar failure processes as described above with different coatings are consistent with other results obtained [4, 15, 27]. The magnified view of the area identified in square (A) is shown in Fig. 12b, here the direction of propagation of the cracks can be clearly established. An additional magnified view of the crack initiation site is shown in Fig. 12c, which also illustrates the partial delamination of the coating at the site where the crack was probably nucleated in square (B).
The fracture surfaces for the last highly stressed CrAlN coating substrate are shown in Figs. 13a-c. Figure 13a determines the fracture surface of a coated specimen tested with a maximum alternating stress of 610.137 MPa and which failed at 10482 cycles. Image analysis corroborates that the sample fracture was the result of multiple propagations of a series of cracks, which nucleated at the sample contour and guided the fracture zone to the center of the surface. The presence of several traces of cracks indicates that they nucleated at different sites on the periphery of the fault surface. These surface results are consistent with similar coating studies [15, 27]. The area in detail in square (A) determines one of the probable starting points of one of the cracks, shown in Fig. 13b. The white arrows in the section of the fracture suggest the direction of expansion and the direction of the arrows establish the growth of the crack. It can also be stated that the crack nucleated at the edge of the section and not in the substrate, near the substrate-coating interface. Figure 13c is an enlargement of the area represented in square (B), it shows the final fracture of the specimen at the starting site of the analyzed crack.
Finally, for comparison purposes, Figs. 14a-c show the fracture surfaces of an uncoated AISI 1045 steel, tested at high stress. Figure 14a shows the surface of an uncoated specimen tested at 610.137 MPa and which failed at 11312 cycles. It was determined that the sample fracture is a product of the propagation of a single crack instead of multiple cracks initiated at the periphery. The surface did not have a flat appearance typical in fatigue, showing irregularity in the entire section analyzed. The fracture zone was located at the opposite end from the beginning of crack formation at A. Figure 14b, shows an enlargement of the A area, where the direction of crack propagation is defined by the white arrows. An enlargement of the crack nucleation start site is shown in Fig. 14c, where the final fracture of the substrate was verified. The results obtained are consistent with references consulted [7, 13].
The results obtained on the fracture surfaces obey the established theory, fractures have their beginning in material discontinuities where the cyclic stress is maximum. The fracture surfaces are flat and perpendicular to the stress axis, and they follow the determined theoretical sequence: stage I is the beginning of one or more microcracks; in stage II the microcracks form parallel surfaces in the shape of plateaus, known as beach marks; and stage III consists of sudden and fast fracture. Fatigue studies using surface treatment, such as plasma nitriding, establish the same sequence for the fracture; however, they have higher increases in fatigue strength limit (23.76%) [26]. It is necessary to mention that processes of thermal spray coating HP/HVOF of tungsten carbide and hard chrome plating, have adverse effects on the fatigue resistance of steels [28].
3.8. Correlation Between Mechanical and Fatigue Properties
Figure 15a shows the relationship between coating nature, hardness and fatigue resistance for all coatings deposited on the industrial steel substrates. It is clearly shown that the improvement in hardness (H), elastic modulus (E) (Table 2), the increase of the maximum alternating tension (Sa) as a functions of cycles number (Fig. 10a) and the increase of fatigue resistance (Fig. 10b), have incidence when the coatings nature (material’s type) were changes. From this correlation it is possible to determine that one merit index [29] associates the increment coatings hardness and the increase of fatigue resistance (%) at the same coating type. Therefore, the TiCN coating offers the best synergy for mechanical and fatigue properties with good hardness and a high maximum alternating tension (Sa) which is very important for mechanical applications in the devices e.g. automotive applications that require high fatigue demands in service conditions. We can establish from the Fig. 15b results, that the biaxial deformation in the plane is affected by the coating material type, the values obtained determine the presence of tensile-type stresses in the following order: ε = − 3.1794x10− 3 for TiCN, ε = -0.0110 for BCN and ε = -0.0447 for CrAlN. The TiCN coating has the lowest tensile stress value.