- Email: [email protected]
Ion-induced phase transformations in nanostructural TiZrAlN films
SCT-19254; No of Pages 6 Surface & Coatings Technology xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Surface & Coatings Technology journal homepage: www.elsevier.com/locate/surfcoat
Ion-induced phase transformations in nanostructural TiZrAlN films V.V. Uglov a,⁎, S.V. Zlotski a, I.A. Saladukhin a, A.Y. Rovbut a, P.I. Gaiduk a, G. Abadias b, G.N. Tolmachova c, S.N. Dub d a
Belarusian State University, 4 Nezavisimosti av., 220030 Minsk, Belarus Institut P′, Département Physique et Mécanique des Matériaux, Université de Poitiers-CNRS-ENSMA, SP2MI, Téléport 2, F86962 Chasseneuil-Futuroscope, France Kharkov Institute of Physics and Technology, 61108 Kharkov, Ukraine d Institute for Superhard Materials, NAS of Ukraine, 04074 Kiev, Ukraine b c
a r t i c l e
i n f o
Available online xxxx Keywords: Phase formation Irradiation Stress Hardness Magnetron sputtering TiZrAlN
a b s t r a c t (Ti,Zr)1 − xAlxNy thin films (300 nm) with Ti:Zr ratio of ~1:1 and Al content in metal sublattice up to x = 0.312 were deposited at Ts = 270 °C using reactive unbalanced magnetron co-sputtering in Ar + N2 plasma discharges. The nitrogen content, y, was understoichiometric when x ≥ 0.102, despite constancy of N2 partial pressure used during growth. The influence of Xe ion irradiation (180 keV, 1 × 1015 −1 × 1017 cm−2) on the structure, phase formation and mechanical properties of the films was investigated. The increase in Al content resulted in gradual evolution of the microstructure from single-phase, nanocrystalline cubic (c) solid solution (x ≤ 0.072) to dualphase nanocomposite (x = 0.102–0.200) and then to amorphous (x = 0.312) one. Nanocrystalline and amorphous films were stable under irradiation, while crystallization occurred in nanocomposite films for a typical ion fluence of 1 × 1016 cm−2. This phase transformation is associated with the reduction of compressive stress and eventually leads to the development of a net tensile stress inside crystallites. For all films, a decrease of the nanoindentation hardness was observed after an ion fluence of 5 × 1016 cm−2, likely related to the incorporation of Xe impurities, as revealed by RBS data. © 2014 Elsevier B.V. All rights reserved.
1. Introduction It is known that Al-containing hard coatings based on transition metal nitride (Me–Al–N), in which Al substitutes for Me element in the MeN-based lattice, possess improved tribological and thermal properties [1–3]. The crystal structure and mechanical and thermal properties of Me–Al–N coatings are strongly determined by the Al content. While keeping cubic lattice, Al content rise leads to improvement of the mechanical properties and to increase in oxidation resistance of the coatings [4–6]. However, the mechanical characteristics of Me– Al–N coatings become worse when Al content exceeds its maximum solubility in the cubic phase (~ 67 аt.%) that is accompanied by a mixed (cubic-NaCl and wurtzite-ZnS) structure formation [1,6–8]. We revealed the enhancement of the nanohardness and oxidation resistance of quaternary Ti–Zr–Al–N system when Al concentration increases and coating's structure is characterized as a single-phase c-(Ti, Zr,Al)N solid solution [9,10]. Therefore, the properties of Ti–Zr–Al–N coatings, as well as in the case of ternary Me–Al–N systems, are determined by the limit of Al solubility in cubic Me–N lattice, which depends on the deposition ⁎ Corresponding author. Tel.: +375 17 2095512. E-mail address: [email protected] (V.V. Uglov).
conditions (deposition temperature, nitrogen pressure, substrate bias). The value of Zr concentration also plays an important role. As it was shown by Yang et al. [11], the increase in Zr concentration is accompanied by the reduction of Al solubility limit in the Ti–Zr–N lattice. In a previous work, single-phase (Ti,Zr)1 − xAlxN sputter-deposited films with cubic structure could be stabilized up to only x = 0.072 [9]. Possible reasons for this low solubility limit were the limited adatom diffusivity due to low substrate temperature Ts = 270 °C and nitrogen deficiency concomitant with further Al increase above x ~ 0.102. Consequently, it is necessary to increase the Al solubility limit in c-(Ti,Zr)N solid solution to design TiZrAlN coatings with improved mechanical and thermal properties and resistance to oxidation. One possible strategy is to use post-growth ion-beam modification of metastable (Ti,Zr)1 − xAlxNy coatings to enhance the mobility of Al atoms in the films by means of ion irradiation-induced atomic displacements in the collision cascades. Formation of radiation defects and ‘Al–vacancy’ complexes can cause radiation-induced elemental redistribution processes, either within the grains and/or at the grain boundary. In the present study, (Ti,Zr)1 − xAlxNy films in various structural states (nanocrystalline, nanocomposite and amorphous) were synthesized, subsequently irradiated with Xe ions to study the influence of the ion-induced phase transformations on the nanohardness of (Ti,Zr)1 − xAlxNy films.
http://dx.doi.org/10.1016/j.surfcoat.2014.03.003 0257-8972/© 2014 Elsevier B.V. All rights reserved.
Please cite this article as: V.V. Uglov, et al., Surf. Coat. Technol. (2014), http://dx.doi.org/10.1016/j.surfcoat.2014.03.003
V.V. Uglov et al. / Surface & Coatings Technology xxx (2014) xxx–xxx
(Ti,Zr)1 − xAlxNy thin films were deposited on (001) Si wafer covered with native SiO2 (~ 2 nm thick) layer using reactive unbalanced magnetron co-sputtering from elemental metallic targets [9]. Deposition was carried out at the substrate temperature Ts = 270 °C and constant bias voltage of − 56 V. The film thickness was ~ 300 nm. The Al content, x, in the films was varied from 0 to 0.312 by tuning the rf power supply of the Al target from 0 to 200 W, while maintaining the dc power supply of Ti and Zr target constants at 300 and 220 W, respectively [9]. The Ti and Zr target powers were chosen so as to obtain practically equal Ti and Zr concentrations in the synthesized films that correspond to the optimal Ti:Zr ratio of 1:1 of ternary TiZrN films which possess enhanced physical–mechanical properties [9,12–14]. The Ar + N2 working pressure was fixed at 0.20 Pa, corresponding to 10 sccm of Ar and N2 flows from 1.3 to 1.6 sccm. This corresponds to N2 partial pressure in the range of (1.8–2.4) × 10−3 Pa. These deposition conditions were chosen based on a previous study in which the process parameters were optimized to obtain stoichiometric ternary TiZrN films in metallic target mode [15]. Other details of the deposition process are given in Refs. [9,10]. Samples irradiation by Xe2 + ions (with the energy of 180 keV) was carried out on EATON NV3206 ion implantor at the Institute P′ of the University of Poitiers (France). The integral doses were 1 × 1015, 1 × 1016, 5 × 1016 and 1 × 1017 cm− 2. The energetic parameters of irradiation were chosen so as the implanted Xe impurities distribution depth did not exceed the coating thickness, as calculated using SRIM-2013 code [16]. XRD analysis was employed for structural identification using a DRON-4 X-ray diffractometer equipped with CuKα wavelength (0.15418 nm) and a D8 Bruker AXS X-ray diffractometer operating in Bragg–Brentano configuration and equipped with CuKα wavelength (0.15418 nm) and an energy dispersive Si(Li) detector (Sol-X detector) defined with a 0.2 mm opening angle slit. XRD survey was carried out for the detailed analysis of a phase formation and transformation of crystal structure in (Ti,Zr)1 − xAlxNy films, and the subsequent line profile fitting of the diffraction lines was made using Lorentz function. XRD stress analysis was performed on as-deposited and ionirradiated films using the crystallite group method (GCM), which is a pseudo-sin2ψ method adapted to the case of textured layers [17], which relies on the measurement of the elastic strain εhkl of crystallites at different ψ angles with respect to their out-of-plane preferred orientation. In the present case, measurements were carried out on (200)-oriented c-TiZrAlN crystallites, using a four circle Seifert XRD 3000 diffractometer operating at 40 kV and 40 mA with a CuKα radiation. More details on the experimental procedure, as well as (hkl) planes and corresponding ψ values used for the measurements can be found elsewhere [17]. Nitrogen concentration and xenon implantation profile were evaluated by the method of Rutherford backscattering (RBS) with 2.0 MeV He+ ions at the High Voltage Engineering Tandetron system accelerator (Skobeltsyn Institute of Nuclear Physics, Lomonosov Moscow State University). The obtained spectra were fitted using SIMNRA software [18]. The mechanical properties of the films were studied by nanoindentation using a Nano Indenter-G200 system (Agilent Technologies, USA) equipped with a continuous stiffness measurement (CSM) attachment option (for more details, see Ref. [9]). A diamond Berkovich tip with some tip blunting was used. Eight indentations were made on each sample. Load (P) and displacement (h) were continuously recorded up to a maximum displacement of 200 nm at a constant indentation strain rate of 0.05 s−1. The values of E and H were determined from the ACP − h and E − h data. 3. Results and discussion The evolution of the phase composition, obtained from XRD, of the as-deposited (Ti,Zr)1 − xAlxNy films with the x value from 0 to 0.312 is
shown in Fig. 1. From a comparison with previously published data for thinner films (180–200 nm) [9] it follows that the increase of Al concentration results in the transformation of the structure from nanocrystalline state (x ≤ 0.072) to nanocomposite (0.102 ≤ x ≤ 0.200) and then to amorphous (x = 0.312) one (see Table 1). The incorporation of Al in TiZrN lattice up to x = 0.072 is also accompanied by a decrease of intensity and broadening of the (111) reflection of the cubic (Ti,Zr,Al)N solid solution (Fig. 1a). For higher Al content, more drastic changes are observed: a broad reflection appears in the vicinity of the (200) reflection of the c-(Ti,Zr,Al)N solid solution. The intensity and position of this broad peak evolves with further Al content increase. A closer inspection of the XRD patterns of the films with x = 0.102–0.200 (Fig. 1b) indicates that the broad and diffuse peak observed in the range of 2θ = 34–41° is not symmetric and it can be shared by two contributions which correspond to the reflections from c-(Ti,Zr,Al)N and a-Al(Ti)N phases [9]. Al(Ti)N phase is identified as amorphous for a presence of a broad diffuse scattering at the angular interval of diffraction around 2θ = 36–38° that covers the (002) and (101) reflections of w-AlN peaks (Fig. 1b). Thus, x = 0.200 is the maximum aluminum content for the films which are characterized by the microstructure of type II, and for this composition the crystal phase grain size becomes minimal (2 nm, according to the evaluation based on Scherrer's equation). For x = 0.312, the broad diffuse peak is symmetric. Additional experiments using transmission electron microscopy (not reported here) have shown that this film is amorphous. It can be concluded that phase composition of the films changes from single-phase (c-(Ti,Zr,Al)N with (111) preferred orientation) to dualphase (c-(Ti,Zr,Al)N with (200) preferred orientation + amorphous
a c-(Ti,Zr,Al)N (200)
c-(Ti,Zr,Al)N (111)
x=0.072
Intensity (arb. units)
2. Experimental details
32
x=0.058 x=0.041 x=0.024 c-(Ti,Zr)N (200)
c-(Ti,Zr)N (111)
w-AlN c-AlN c-ZrN c-TiN
34
36
38
40
42
44
2θ (degrees)
b c-(Ti,Zr)N (111)
c-(Ti,Zr)N (200)
x=0.312 x=0.200
Intensity (arb. units)
2
32
x=0.160 x=0.128
x=0.102 w-AlN c-AlN c-ZrN c-TiN
34
36
38
40
42
44
2θ (degrees) Fig. 1. X-ray diffraction patterns of (Ti,Zr)1 − xAlxNy films with different Al contents (a): x = 0.024–0.072; (b): x = 0.102–0.312. Patterns corresponding to cubic TiN (JCPDS card no. 38-1420), cubic ZrN (JCPDS card no. 35-753), cubic AlN (JCPDS card no. 5-1495) and hexagonal w-AlN (JCPDS card no. 25-1133) are presented at the lower part.
Please cite this article as: V.V. Uglov, et al., Surf. Coat. Technol. (2014), http://dx.doi.org/10.1016/j.surfcoat.2014.03.003
V.V. Uglov et al. / Surface & Coatings Technology xxx (2014) xxx–xxx
3
Table 1 Structural characteristics of as-deposited magnetron sputtered (Ti,Zr)1 − xAlxNy films (300 nm-thick) with the different Al contents, x, and the corresponding values of grain size of c-(Ti,Zr,Al)N solid solution calculated from XRD line broadening of (200) peak. Denomination of micro-structure
Range of Al contenta
Range of Ti:Zr ratio
Range of N concentration (at.%)
Nature of structure
XRD pattern features
Grain size (nm)
Type I
x ≤ 0.072
1.09–1.12
49–51
0.102 ≤ x ≤ 0.200
1.11–1.16
34–39
Mixed (111) and (200) preferred orientations Strong (200) preferred orientation of c-(Ti,Zr,Al)N nanograins + broad diffuse scattering around 2θ = 36–38°
14–17
Type II
Type III
x = 0.312
1.16
24–31
Nanocrystalline (Ti,Zr,Al)N solid solution with cubic structure Dual-phase nanocomposite consisting of c-(Ti,Zr,Al)N nanograins surrounded by highly disordered matrix (a-Al(Ti)N) XRD-amorphous TiZrAlN phase
Broad diffuse scattering around 2θ = 35–40°
–
In metal sublattice.
b
Counts
Concentration of Xe, at.%
4,0
8
3,5 3,0 2,5
Ti
2,0 1,5
Zr
1,0 0,5 0,0
20
40
N
60 80 Depth, nm
100
120
Al
Si
Concentration of Xe, at.%
a
Counts
a
2–6
7 6
Zr
5
Ti
4 3 2 1 0 0
40
80 120 Depth, nm
160
200
N
240
Xe
Si Al
Xe
Hf
100
200
300
400
500
600
700
800
Hf
900
1000
50
100
150
200
250
Channel
350
400
450
500
12 10 8 6
Zr
Ti
4
Counts
2 0 0
40
80
120 160 Depth, nm
N
200
240
Xe
Si
Concentration of Xe, at.%
d Concentration of Xe, at.%
c
Counts
300
Channel
21 18 15
Zr
12 9
Ti
6 3 0 0
40
80 120 Depth, nm
160
N
200
240
Xe
Si Al
Al
Hf
Hf
50
100
150
200
250
300
350
400
450
50
500
100
150
200
Channel
e
300
350
400
450
500
24 Concentration of Xe, at.%
Counts
250
Channel
20
Zr
16 12 8
Ti
4 0 0
40
80
120 160 Depth, nm
N
200
240
280
Xe Si Al
Hf
50
100
150
200
250
300
350
400
450
500
Channel Fig. 2. RBS spectra of (Ti,Zr)1 − xAlxNy films with different Al concentrations and irradiated by Xe2+ (180 keV, 5 × 1016 cm−2 (a) and 1 × 1017 cm−2 (b–e)): x = 0.058 (a), 0.102 (b), 0.160 (c), 0.200 (d), 0.312 (e). There is Xe distribution in the inserts.
Please cite this article as: V.V. Uglov, et al., Surf. Coat. Technol. (2014), http://dx.doi.org/10.1016/j.surfcoat.2014.03.003
4
V.V. Uglov et al. / Surface & Coatings Technology xxx (2014) xxx–xxx
with the aluminum concentration. In the case of the amorphous film (type III microstructure), the concentration of Xe reaches 21.8 аt.%. The evolution of the structure and phase composition of (Ti,Zr)1 2+ −xAlxNy films (x = 0.058, 0.102, 0.160, 0.200 and 0.312) under Xe ion irradiation is shown in Fig. 3. Differences in the evolution of XRD patterns with irradiation dose are clearly observed, depending on the structure type of as-deposited (Ti,Zr)1 − xAlxNy films. For x = 0.058 (type I) no significant changes in the intensity and broadening of the solid solution diffraction peaks are revealed (Fig. 3a). The same observation holds for the film with x = 0.312 (type III), see Fig. 3e. These observations suggest that ion irradiation does not change the structure type of these films, at least in the investigated dose range. This can be due to the lack of ion energy and irradiation dose for the nanocrystalline structure destabilization and amorphous structure crystallization.
a-Al(Ti)N phase) and then to amorphous a-TiZrAlN state (Fig. 1). Dualphase (Ti,Zr)1 − xAlxNy film (type II microstructure) is a nanocomposite based on the c-(Ti,Zr,Al)N solid solution grains (size of 2–6 nm) surrounded by the amorphous a-Al(Ti)N phase. Amorphous a-TiZrAlN state (type III microstructure) is characterized by a broad diffuse scattering around 2θ = 35–40° (see Table 1, Fig. 1b). To investigate possible elemental redistribution in (Ti,Zr)1 − xAlxNy films after irradiation with Xe ions the RBS experiments have been performed. Redistribution of the constituent elements in the films was not revealed even at the maximum dose of 1 × 1017 cm−2 (Fig. 2). The distributions of the implanted Xe ions along the film depth calculated from RBS spectra are presented in the inserts of Fig. 2. From the obtained Xe distribution profiles, one can see that the penetration depth (~150 nm) does not exceed half of the film thickness. The maximum concentration of the implanted Xe in the film rises with the irradiation dose as well as
a
b
16
c-(Ti,Zr,Al)N (200)
c-(Ti,Zr,Al)N (200)
Intensity (arb. units)
Intensity (arb. units)
c-(Ti,Zr,Al)N (111)
-2
5x10 cm
1x1016 cm-2
1x1017 cm-2
5x1016 cm-2
1x1016 cm-2 a-Al(Ti)N 1x1015 cm-2
as-deposited 34
35
36
37
38
39
40
41
42
43
34
35
36
37
2θ (degrees)
38
39
40
41
42
43
44
2θ (degrees)
d
c 17
c-(Ti,Zr,Al)N (200)
-2
Intensity (arb. units)
5x1016 cm-2
a-Al(Ti)N 1x1016 cm-2 1x1015 cm-2 c-(Ti,Zr,Al)N (200)
as-deposited 34
35
36
37
38
39
40
41
c-(Ti,Zr,Al)N (200)
1x1017 cm-2
1x10 cm
Intensity (arb. units)
c-(Ti,Zr,Al)N (200)
a-Al(Ti)N
as-deposited
42
43
5x1016 cm-2
a-Al(Ti)N 1x1016 cm-2
x4
x4
1x1015 cm-2
x5
x4
as-deposited
44
34
35
36
37
2θ (degrees)
38
39
40
41
42
43
44
2θ (degrees)
e Intensity (arb. units)
a-TiZrAlN
1x1017 cm-2 5x1016 cm-2 1x1016 cm-2 as-deposited
34
35
36
37
38
39
40
41
42
43
44
2θ (degrees) Fig. 3. X-ray diffraction patterns of (Ti,Zr)1 − xAlxNy films with different Al contents: x = 0.058 (а), 0.102 (b), 0.160 (c), 0.200 (d), 0.312 (e), as-deposited and irradiated by Xe2+ ions with the energy of 180 keV at the doses of 1 × 1015–1 × 1017 cm−2.
Please cite this article as: V.V. Uglov, et al., Surf. Coat. Technol. (2014), http://dx.doi.org/10.1016/j.surfcoat.2014.03.003
V.V. Uglov et al. / Surface & Coatings Technology xxx (2014) xxx–xxx
irradiation can be an effective means to obtain a higher Al solubility in the c-TiZrN lattice. To elucidate the origin of the evolution in intensity and angular position of XRD lines after ion irradiation, a detailed stress analysis was carried out for the film with x = 0.102. The evolutions of the lattice parameter with ion fluence are reported in Fig. 4, using sin2ψ plots. In such a representation, the slope of the sin2ψ line is proportional to the inplane biaxial stress in the film. In the as-deposited state, the cubic TiZrAlN crystallites with (200) orientation are under large compressive stress. One can see that the slope of the sin2ψ lines gradually decreases with an ion fluence in the range 1 × 1015 to 1 × 1016 cm−2, and becomes eventually positive at 5 × 1016 cm−2, attesting of the development of a net tensile stress. At the highest ion fluence used (1 × 1017 cm−2), the stress appears almost relaxed. Interestingly, all sin2ψ lines intersect at a single point, whose ordinate corresponds to a stress-free lattice parameter a0 ~ 4.402 ± 0.004 Å. This value is lower than that of cubic TiZrN compound with similar Ti:Zr ratio which equals to 4.45 Å, taking into account the positive deviation from Vegard's law [19], suggesting that Al substitutes at metal sublattice of the c-TiZrAlN solid solution. The development of a net tensile stress after irradiation could be related to ion-induced annihilation of growth-induced point defects together with improvement of crystallinity, despite the accumulation of substantial Xe impurities in the film (see Fig. 2). This suggests that Xe may interact with the radiation vacancies and form gas–vacancy complexes (e.g. voids and/or bubbles) at preferential sites (grain boundary) without inducing significant compressive stress source. In Fig. 5, the evolution of nanoindentation hardness of (Ti,Zr)1 − xAlxNy films vs. Al content x is reported for both as-deposited and Xe irradiated states. With the rise of Al content the hardness of the nanocrystalline films (type I) increases and reaches a maximum value of 26 GPa for x = 0.072. The further increase in aluminum concentration leads to a drop of hardness and for the amorphous film it equals to 14 GPa (Fig. 5). Such decrease in hardness is apparently connected with the gradual decrease in crystallinity degree of the structure at the high aluminum concentrations. Under irradiation with Xe ions (1 × 1016 cm− 2) the hardness of nanocrystalline films does not essentially change but the hardness of nanocomposite and amorphous films starts to decrease. At the dose of 5 × 1016 cm− 2 the hardness of nanocrystalline films decreases by 10–17%, and that of nanocomposite and amorphous films by 32–35%. The reduction in hardness can be explained by the xenon accumulation and concurrent growth of radiation damages of the crystal lattice, despite the higher Al solubility achieved in the cubic phase.
4.50 as-dep 1x1015
4.48
1x1016 4.46
lattice parameter (Å)
However, irradiation of nanocomposite films (0.102 ≤ x ≤ 0.200, type II) leads to their transformation into a nanocrystalline state at intermediate doses (1 × 1015 to 5 × 1016 cm−2), as revealed by the emergence of a XRD line in the vicinity of the (200) reflection of the c-(Ti,Zr,Al)N solid solution. An increase in intensity and reduction of peak width is noticed with ion dose (Fig. 3b–d), suggesting improvement of film crystalline quality, until this latter eventually degrades at the maximum dose of 1 × 1017 cm−2. Radiation-stimulated diffusion processes may lead to the migration of Al and Ti atoms to form a homogenous c-(Ti,Zr,Al)N solid solution from the initial phase-separated state. It may be envisaged that grain growth of this homogeneous cubic solid solution occurs from the ‘dissolution’ of the partially segregated а-Al(Ti)N interfacial phase. At higher dose, the presence of excess impurity Xe atoms reduces the coherency length of the crystals, which explains the loss in diffracted intensity of the c-(Ti,Zr,Al)N (200) line at 1 × 1017 cm−2. Therefore, it can be concluded that ion irradiation can cause a radiation-induced crystallization for highly metastable (Ti,Zr)1 − xAlxNy films, and that this structural evolution occurs mainly because of the generation of non-equilibrium point defects (‘radiation transport’) which can ensure the diffusion of elemental species. Further chemical analysis, using atomic resolution probe techniques, would be desirable to identify the diffusion species. Additionally, a splitting of the diffraction peaks of the c-(Ti, Zr,Al)N solid solution into two components is found for irradiated (Ti,Zr) 1 − x Al x N y films with x = 0.058, 0.102 and to a lesser extent with x = 0.160 (Fig. 3). In the cases of x = 0.160 (at the doses of 1 × 1016 cm−2 and higher) and x = 0.200 the peak sharing is not noticeable because of their low intensity starting from asdeposited state. A satisfactory fit of the XRD line profile is obtained when considering two contributions (using Lorentz functions) centered at a region of larger and smaller angles (x = 0.058, Fig. 3a) or both to the region of larger angles (x = 0.102, 0.160, Fig. 3b,c) as compared to the initial peak position. It is possible to interpret these results from the following ‘double-layer’ structural model. According to RBS data analysis, the coatings irradiated with Xe can be considered as the films consisting of two layers: the first one (near the surface) is implanted by Xe ions and the second one (near the substrate) is not subjected to implantation. In this case, the first layer will have the high level of the compressive stresses caused by radiation defects and accumulation of implanted Xe atoms that likely occurs at the grain boundaries. And the second layer will be characterized by the tensile stresses as a response of the coating to compensate the stresses in the first (upper) layer. The difference in peak splitting for irradiated (Ti,Zr)1 − xAlxNy films with x = 0.058 (Fig. 3a) and x = 0.102 (Fig. 3b) can be understood based on the pristine structural type: nanocrystalline vs. nanocomposite. The fact that for x = 0.102 the two contributions to the (200) XRD line are shifted to higher angles, as compared with the peak position for the as-deposited film, can be explained by the enrichment in aluminum of the (Ti,Zr)1 − xAlxNy solid solution after irradiation (likely in the first layer). This is a consequence of the dissolution of the Al(Ti)N amorphous phase under Xe irradiation, where the Al content in this partially segregated phase is higher than in c-(Ti,Zr,Al)N solid solution counterpart. With the increase of irradiation dose the intensity and broadening of the peak of this solid solution are being changed that is connected with the competing processes of radiation crystallization and defect formation. Similar processes are revealed for x = 0.160 and 0.200 also. The lattice parameter of c-(Ti,Zr,Al)N solid solution, deduced from XRD patterns on irradiated films reported in Fig. 3, decreases from 4.455 to 4.366 Å with the rise of Al content in the (Ti,Zr)1 − xAlxNy films from 0.102 to 0.200, as expected since Al atoms have a lower atomic radius than Ti and Zr ones. Although the stress state may influence the values of the lattice parameter derived from a simple analysis of peak position (which corresponds here to out-of-plane lattice spacings) this variation supports the main finding that Xe
5
5x1016 1x1017
4.44 4.42 4.40 4.38 4.36 4.34 0
0.2
0.4
0.6
0.8
1
sin2 ψ Fig. 4. sin2ψ plot evolution with ion irradiation fluence for the (Ti,Zr)1 − xAlxNy film with x = 0.102.
Please cite this article as: V.V. Uglov, et al., Surf. Coat. Technol. (2014), http://dx.doi.org/10.1016/j.surfcoat.2014.03.003
6
V.V. Uglov et al. / Surface & Coatings Technology xxx (2014) xxx–xxx
28
type I
type II
type III
26
Hardness (GPa)
24
as-deposited irradiated by Xe (1x1016 cm-2) irradiated by Xe (5x1016 cm-2)
22 20 18 16 14
At the Xe irradiation with the dose of 1 × 1016 cm−2 the hardness of nanocrystalline films practically does not change and it starts to decrease for nanocomposite and amorphous films. The increase in irradiation dose up to a value of 5 × 1016 cm−2 leads to the reduction of the hardness of (Ti,Zr)1 − xAlxNy films (by 10–17% for nanocrystalline and 32–35% for nanocomposite and amorphous films). Therefore, in spite of the achievement of a higher Al solubility in the cubic TiZrN lattice under Xe ion irradiation, there is no expected enhancement of film hardness because of high Xe accumulation in the film structure.
12 10
Conflict of interest
8 0,00
0,04
0,08
0,12
0,16
0,20
0,24
0,28
0,32
Al content, x Fig. 5. The evolution of the hardness of as-deposited and irradiated (Ti,Zr)1 − xAlxNy films with varying Al contents.
4. Conclusions Quaternary (Ti,Zr)1 − xAlxNy thin films were synthesized by a reactive unbalanced magnetron sputtering method in the following structural states depending on the aluminum content in metal sublattice, x: • single-phase, nanocrystalline, cubic c-(Ti,Zr,Al)N solid solution with a typical grain size of 17–14 nm, for 0 ≤ x ≤ 0.072, • dual-phase nanocomposite consisting of c-(Ti,Zr,Al)N (grain size of 6–2 nm) and a-Al(Ti)N, for 0.102 ≤ x ≤ 0.200, • amorphous a-TiZrAlN phase, for x = 0.312. The maximum values of hardness (24–26 GPa) are registered for the films with 0.058 ≤ x ≤ 0.128. When further increasing the Al content, their hardness considerably decreases due to the amorphization of the films. Irradiation by xenon ions (180 keV, 1 × 1015–1 × 1017 cm−2) does not affect the structural-phase state of the nanocrystalline (x = 0.058) and amorphous (x = 0.312) films. However, the structure of nanocomposite films (0.102 ≤ x ≤ 0.200) is transformed into the nanocrystalline state, with the formation of a homogeneous cubic solid solution at intermediate doses (1 × 1015 to 5 × 1016 cm−2). Such structural phase transformation is mediated by radiation-stimulated atomic transport in the collision cascades, and dissolution of the segregated a-Al(Ti)N interfacial phase. Redistribution of the constituent elements along the film depth does not occur under irradiation by Xe ions. It was revealed that the maximum concentration of implanted Xe rises with the increase in Al content and it reaches 21.8 at.% for x = 0.312 at the dose of 1 × 1017 cm−2.
None.
Acknowledgments The work was supported by the Belarusian Republic Foundation for Fundamental Research (F13F-003). The authors are very grateful to the engineers of the Institute P′ of the University of Poitiers (France) Dr. Philippe Guerin and Marc Marteau for their help in the realization of synthesis and irradiation of the films. References [1] S. PalDey, S.C. Deevi, Mater. Sci. Eng. A 342 (2003) 58. [2] F. Vaz, L. Rebouta, M. Andritschky, M.F. Silva, J.C. Soares, J. Eur. Ceram. Soc. 17 (1997) 1971. [3] L. Rogström, L.J.S. Johnson, M.P. Johansson, M. Ahlgren, L. Hultman, M. Odén, Thin Solid Films 519 (2010) 694. [4] Z.-J. Liu, P.W. Shum, Y.G. Shen, Thin Solid Films 468 (2004) 161. [5] J.M. Castanho, M.T. Vieira, J. Mater. Proc. Technol. 143–144 (2003) 352. [6] L. Chen, J. Paulitsch, Y. Du, P.H. Mayrhofer, Surf. Coat. Technol. 206 (2012) 2954. [7] M.-H. Tuilier, M.-J. Pac, M. Gîrleanu, G. Covarel, G. Arnold, P. Louis, C. Rousselot, A.-M. Flank, J. Appl. Phys. 103 (2008) 083524. [8] P.H. Mayrhofer, D. Music, J. Schneider, J. Appl. Phys. 100 (2006) 094906. [9] I.A. Saladukhin, G. Abadias, A. Michel, S.V. Zlotski, V.V. Uglov, G.N. Tolmachova, S.N. Dub, Thin Solid Films 538 (2013) 32. [10] G. Abadias, I.A. Saladukhin, V.V. Uglov, S.V. Zlotski, D. Eyidi, Surf. Coat. Technol. 237 (2013) 187. [11] B. Yang, L. Chen, Y.X. Xu, Y.B. Peng, J.C. Fen, Y. Du, M.J. Wu, Int. J. Refract. Met. Hard Mater. 38 (2013) 81. [12] V.V. Uglov, V.M. Anishchik, S.V. Zlotski, G. Abadias, Surf. Coat. Technol. 200 (2006) 6389. [13] V.V. Uglov, V.M. Anishchik, V.V. Khodasevich, Zh.L. Prikhodko, S.V. Zlotski, G. Abadias, S.N. Dub, Surf. Coat. Technol. 180–181 (2004) 519. [14] V.V. Uglov, V.M. Anishchik, S.V. Zlotski, G. Abadias, S.N. Dub, Surf. Coat. Technol. 202 (2008) 2394. [15] G. Abadias, L.E. Koutsokeras, S.N. Dub, G.N. Tolmachova, A. Debelle, T. Sauvage, P. Villechaise, J. Vac. Sci. Technol. A 28 (2010) 541. [16] http://www.srim.org. [17] G. Abadias, Surf. Coat. Technol. 202 (2008) 2223. [18] http://www.rzg.mpg.de/mam/. [19] G. Abadias, V.I. Ivashchenko, L. Belliard, Ph. Djemia, Acta Mater. 60 (2012) 5601
Please cite this article as: V.V. Uglov, et al., Surf. Coat. Technol. (2014), http://dx.doi.org/10.1016/j.surfcoat.2014.03.003
Contents lists available at ScienceDirect
Surface & Coatings Technology journal homepage: www.elsevier.com/locate/surfcoat
Ion-induced phase transformations in nanostructural TiZrAlN films V.V. Uglov a,⁎, S.V. Zlotski a, I.A. Saladukhin a, A.Y. Rovbut a, P.I. Gaiduk a, G. Abadias b, G.N. Tolmachova c, S.N. Dub d a
Belarusian State University, 4 Nezavisimosti av., 220030 Minsk, Belarus Institut P′, Département Physique et Mécanique des Matériaux, Université de Poitiers-CNRS-ENSMA, SP2MI, Téléport 2, F86962 Chasseneuil-Futuroscope, France Kharkov Institute of Physics and Technology, 61108 Kharkov, Ukraine d Institute for Superhard Materials, NAS of Ukraine, 04074 Kiev, Ukraine b c
a r t i c l e
i n f o
Available online xxxx Keywords: Phase formation Irradiation Stress Hardness Magnetron sputtering TiZrAlN
a b s t r a c t (Ti,Zr)1 − xAlxNy thin films (300 nm) with Ti:Zr ratio of ~1:1 and Al content in metal sublattice up to x = 0.312 were deposited at Ts = 270 °C using reactive unbalanced magnetron co-sputtering in Ar + N2 plasma discharges. The nitrogen content, y, was understoichiometric when x ≥ 0.102, despite constancy of N2 partial pressure used during growth. The influence of Xe ion irradiation (180 keV, 1 × 1015 −1 × 1017 cm−2) on the structure, phase formation and mechanical properties of the films was investigated. The increase in Al content resulted in gradual evolution of the microstructure from single-phase, nanocrystalline cubic (c) solid solution (x ≤ 0.072) to dualphase nanocomposite (x = 0.102–0.200) and then to amorphous (x = 0.312) one. Nanocrystalline and amorphous films were stable under irradiation, while crystallization occurred in nanocomposite films for a typical ion fluence of 1 × 1016 cm−2. This phase transformation is associated with the reduction of compressive stress and eventually leads to the development of a net tensile stress inside crystallites. For all films, a decrease of the nanoindentation hardness was observed after an ion fluence of 5 × 1016 cm−2, likely related to the incorporation of Xe impurities, as revealed by RBS data. © 2014 Elsevier B.V. All rights reserved.
1. Introduction It is known that Al-containing hard coatings based on transition metal nitride (Me–Al–N), in which Al substitutes for Me element in the MeN-based lattice, possess improved tribological and thermal properties [1–3]. The crystal structure and mechanical and thermal properties of Me–Al–N coatings are strongly determined by the Al content. While keeping cubic lattice, Al content rise leads to improvement of the mechanical properties and to increase in oxidation resistance of the coatings [4–6]. However, the mechanical characteristics of Me– Al–N coatings become worse when Al content exceeds its maximum solubility in the cubic phase (~ 67 аt.%) that is accompanied by a mixed (cubic-NaCl and wurtzite-ZnS) structure formation [1,6–8]. We revealed the enhancement of the nanohardness and oxidation resistance of quaternary Ti–Zr–Al–N system when Al concentration increases and coating's structure is characterized as a single-phase c-(Ti, Zr,Al)N solid solution [9,10]. Therefore, the properties of Ti–Zr–Al–N coatings, as well as in the case of ternary Me–Al–N systems, are determined by the limit of Al solubility in cubic Me–N lattice, which depends on the deposition ⁎ Corresponding author. Tel.: +375 17 2095512. E-mail address: [email protected] (V.V. Uglov).
conditions (deposition temperature, nitrogen pressure, substrate bias). The value of Zr concentration also plays an important role. As it was shown by Yang et al. [11], the increase in Zr concentration is accompanied by the reduction of Al solubility limit in the Ti–Zr–N lattice. In a previous work, single-phase (Ti,Zr)1 − xAlxN sputter-deposited films with cubic structure could be stabilized up to only x = 0.072 [9]. Possible reasons for this low solubility limit were the limited adatom diffusivity due to low substrate temperature Ts = 270 °C and nitrogen deficiency concomitant with further Al increase above x ~ 0.102. Consequently, it is necessary to increase the Al solubility limit in c-(Ti,Zr)N solid solution to design TiZrAlN coatings with improved mechanical and thermal properties and resistance to oxidation. One possible strategy is to use post-growth ion-beam modification of metastable (Ti,Zr)1 − xAlxNy coatings to enhance the mobility of Al atoms in the films by means of ion irradiation-induced atomic displacements in the collision cascades. Formation of radiation defects and ‘Al–vacancy’ complexes can cause radiation-induced elemental redistribution processes, either within the grains and/or at the grain boundary. In the present study, (Ti,Zr)1 − xAlxNy films in various structural states (nanocrystalline, nanocomposite and amorphous) were synthesized, subsequently irradiated with Xe ions to study the influence of the ion-induced phase transformations on the nanohardness of (Ti,Zr)1 − xAlxNy films.
http://dx.doi.org/10.1016/j.surfcoat.2014.03.003 0257-8972/© 2014 Elsevier B.V. All rights reserved.
Please cite this article as: V.V. Uglov, et al., Surf. Coat. Technol. (2014), http://dx.doi.org/10.1016/j.surfcoat.2014.03.003
V.V. Uglov et al. / Surface & Coatings Technology xxx (2014) xxx–xxx
(Ti,Zr)1 − xAlxNy thin films were deposited on (001) Si wafer covered with native SiO2 (~ 2 nm thick) layer using reactive unbalanced magnetron co-sputtering from elemental metallic targets [9]. Deposition was carried out at the substrate temperature Ts = 270 °C and constant bias voltage of − 56 V. The film thickness was ~ 300 nm. The Al content, x, in the films was varied from 0 to 0.312 by tuning the rf power supply of the Al target from 0 to 200 W, while maintaining the dc power supply of Ti and Zr target constants at 300 and 220 W, respectively [9]. The Ti and Zr target powers were chosen so as to obtain practically equal Ti and Zr concentrations in the synthesized films that correspond to the optimal Ti:Zr ratio of 1:1 of ternary TiZrN films which possess enhanced physical–mechanical properties [9,12–14]. The Ar + N2 working pressure was fixed at 0.20 Pa, corresponding to 10 sccm of Ar and N2 flows from 1.3 to 1.6 sccm. This corresponds to N2 partial pressure in the range of (1.8–2.4) × 10−3 Pa. These deposition conditions were chosen based on a previous study in which the process parameters were optimized to obtain stoichiometric ternary TiZrN films in metallic target mode [15]. Other details of the deposition process are given in Refs. [9,10]. Samples irradiation by Xe2 + ions (with the energy of 180 keV) was carried out on EATON NV3206 ion implantor at the Institute P′ of the University of Poitiers (France). The integral doses were 1 × 1015, 1 × 1016, 5 × 1016 and 1 × 1017 cm− 2. The energetic parameters of irradiation were chosen so as the implanted Xe impurities distribution depth did not exceed the coating thickness, as calculated using SRIM-2013 code [16]. XRD analysis was employed for structural identification using a DRON-4 X-ray diffractometer equipped with CuKα wavelength (0.15418 nm) and a D8 Bruker AXS X-ray diffractometer operating in Bragg–Brentano configuration and equipped with CuKα wavelength (0.15418 nm) and an energy dispersive Si(Li) detector (Sol-X detector) defined with a 0.2 mm opening angle slit. XRD survey was carried out for the detailed analysis of a phase formation and transformation of crystal structure in (Ti,Zr)1 − xAlxNy films, and the subsequent line profile fitting of the diffraction lines was made using Lorentz function. XRD stress analysis was performed on as-deposited and ionirradiated films using the crystallite group method (GCM), which is a pseudo-sin2ψ method adapted to the case of textured layers [17], which relies on the measurement of the elastic strain εhkl of crystallites at different ψ angles with respect to their out-of-plane preferred orientation. In the present case, measurements were carried out on (200)-oriented c-TiZrAlN crystallites, using a four circle Seifert XRD 3000 diffractometer operating at 40 kV and 40 mA with a CuKα radiation. More details on the experimental procedure, as well as (hkl) planes and corresponding ψ values used for the measurements can be found elsewhere [17]. Nitrogen concentration and xenon implantation profile were evaluated by the method of Rutherford backscattering (RBS) with 2.0 MeV He+ ions at the High Voltage Engineering Tandetron system accelerator (Skobeltsyn Institute of Nuclear Physics, Lomonosov Moscow State University). The obtained spectra were fitted using SIMNRA software [18]. The mechanical properties of the films were studied by nanoindentation using a Nano Indenter-G200 system (Agilent Technologies, USA) equipped with a continuous stiffness measurement (CSM) attachment option (for more details, see Ref. [9]). A diamond Berkovich tip with some tip blunting was used. Eight indentations were made on each sample. Load (P) and displacement (h) were continuously recorded up to a maximum displacement of 200 nm at a constant indentation strain rate of 0.05 s−1. The values of E and H were determined from the ACP − h and E − h data. 3. Results and discussion The evolution of the phase composition, obtained from XRD, of the as-deposited (Ti,Zr)1 − xAlxNy films with the x value from 0 to 0.312 is
shown in Fig. 1. From a comparison with previously published data for thinner films (180–200 nm) [9] it follows that the increase of Al concentration results in the transformation of the structure from nanocrystalline state (x ≤ 0.072) to nanocomposite (0.102 ≤ x ≤ 0.200) and then to amorphous (x = 0.312) one (see Table 1). The incorporation of Al in TiZrN lattice up to x = 0.072 is also accompanied by a decrease of intensity and broadening of the (111) reflection of the cubic (Ti,Zr,Al)N solid solution (Fig. 1a). For higher Al content, more drastic changes are observed: a broad reflection appears in the vicinity of the (200) reflection of the c-(Ti,Zr,Al)N solid solution. The intensity and position of this broad peak evolves with further Al content increase. A closer inspection of the XRD patterns of the films with x = 0.102–0.200 (Fig. 1b) indicates that the broad and diffuse peak observed in the range of 2θ = 34–41° is not symmetric and it can be shared by two contributions which correspond to the reflections from c-(Ti,Zr,Al)N and a-Al(Ti)N phases [9]. Al(Ti)N phase is identified as amorphous for a presence of a broad diffuse scattering at the angular interval of diffraction around 2θ = 36–38° that covers the (002) and (101) reflections of w-AlN peaks (Fig. 1b). Thus, x = 0.200 is the maximum aluminum content for the films which are characterized by the microstructure of type II, and for this composition the crystal phase grain size becomes minimal (2 nm, according to the evaluation based on Scherrer's equation). For x = 0.312, the broad diffuse peak is symmetric. Additional experiments using transmission electron microscopy (not reported here) have shown that this film is amorphous. It can be concluded that phase composition of the films changes from single-phase (c-(Ti,Zr,Al)N with (111) preferred orientation) to dualphase (c-(Ti,Zr,Al)N with (200) preferred orientation + amorphous
a c-(Ti,Zr,Al)N (200)
c-(Ti,Zr,Al)N (111)
x=0.072
Intensity (arb. units)
2. Experimental details
32
x=0.058 x=0.041 x=0.024 c-(Ti,Zr)N (200)
c-(Ti,Zr)N (111)
w-AlN c-AlN c-ZrN c-TiN
34
36
38
40
42
44
2θ (degrees)
b c-(Ti,Zr)N (111)
c-(Ti,Zr)N (200)
x=0.312 x=0.200
Intensity (arb. units)
2
32
x=0.160 x=0.128
x=0.102 w-AlN c-AlN c-ZrN c-TiN
34
36
38
40
42
44
2θ (degrees) Fig. 1. X-ray diffraction patterns of (Ti,Zr)1 − xAlxNy films with different Al contents (a): x = 0.024–0.072; (b): x = 0.102–0.312. Patterns corresponding to cubic TiN (JCPDS card no. 38-1420), cubic ZrN (JCPDS card no. 35-753), cubic AlN (JCPDS card no. 5-1495) and hexagonal w-AlN (JCPDS card no. 25-1133) are presented at the lower part.
Please cite this article as: V.V. Uglov, et al., Surf. Coat. Technol. (2014), http://dx.doi.org/10.1016/j.surfcoat.2014.03.003
V.V. Uglov et al. / Surface & Coatings Technology xxx (2014) xxx–xxx
3
Table 1 Structural characteristics of as-deposited magnetron sputtered (Ti,Zr)1 − xAlxNy films (300 nm-thick) with the different Al contents, x, and the corresponding values of grain size of c-(Ti,Zr,Al)N solid solution calculated from XRD line broadening of (200) peak. Denomination of micro-structure
Range of Al contenta
Range of Ti:Zr ratio
Range of N concentration (at.%)
Nature of structure
XRD pattern features
Grain size (nm)
Type I
x ≤ 0.072
1.09–1.12
49–51
0.102 ≤ x ≤ 0.200
1.11–1.16
34–39
Mixed (111) and (200) preferred orientations Strong (200) preferred orientation of c-(Ti,Zr,Al)N nanograins + broad diffuse scattering around 2θ = 36–38°
14–17
Type II
Type III
x = 0.312
1.16
24–31
Nanocrystalline (Ti,Zr,Al)N solid solution with cubic structure Dual-phase nanocomposite consisting of c-(Ti,Zr,Al)N nanograins surrounded by highly disordered matrix (a-Al(Ti)N) XRD-amorphous TiZrAlN phase
Broad diffuse scattering around 2θ = 35–40°
–
In metal sublattice.
b
Counts
Concentration of Xe, at.%
4,0
8
3,5 3,0 2,5
Ti
2,0 1,5
Zr
1,0 0,5 0,0
20
40
N
60 80 Depth, nm
100
120
Al
Si
Concentration of Xe, at.%
a
Counts
a
2–6
7 6
Zr
5
Ti
4 3 2 1 0 0
40
80 120 Depth, nm
160
200
N
240
Xe
Si Al
Xe
Hf
100
200
300
400
500
600
700
800
Hf
900
1000
50
100
150
200
250
Channel
350
400
450
500
12 10 8 6
Zr
Ti
4
Counts
2 0 0
40
80
120 160 Depth, nm
N
200
240
Xe
Si
Concentration of Xe, at.%
d Concentration of Xe, at.%
c
Counts
300
Channel
21 18 15
Zr
12 9
Ti
6 3 0 0
40
80 120 Depth, nm
160
N
200
240
Xe
Si Al
Al
Hf
Hf
50
100
150
200
250
300
350
400
450
50
500
100
150
200
Channel
e
300
350
400
450
500
24 Concentration of Xe, at.%
Counts
250
Channel
20
Zr
16 12 8
Ti
4 0 0
40
80
120 160 Depth, nm
N
200
240
280
Xe Si Al
Hf
50
100
150
200
250
300
350
400
450
500
Channel Fig. 2. RBS spectra of (Ti,Zr)1 − xAlxNy films with different Al concentrations and irradiated by Xe2+ (180 keV, 5 × 1016 cm−2 (a) and 1 × 1017 cm−2 (b–e)): x = 0.058 (a), 0.102 (b), 0.160 (c), 0.200 (d), 0.312 (e). There is Xe distribution in the inserts.
Please cite this article as: V.V. Uglov, et al., Surf. Coat. Technol. (2014), http://dx.doi.org/10.1016/j.surfcoat.2014.03.003
4
V.V. Uglov et al. / Surface & Coatings Technology xxx (2014) xxx–xxx
with the aluminum concentration. In the case of the amorphous film (type III microstructure), the concentration of Xe reaches 21.8 аt.%. The evolution of the structure and phase composition of (Ti,Zr)1 2+ −xAlxNy films (x = 0.058, 0.102, 0.160, 0.200 and 0.312) under Xe ion irradiation is shown in Fig. 3. Differences in the evolution of XRD patterns with irradiation dose are clearly observed, depending on the structure type of as-deposited (Ti,Zr)1 − xAlxNy films. For x = 0.058 (type I) no significant changes in the intensity and broadening of the solid solution diffraction peaks are revealed (Fig. 3a). The same observation holds for the film with x = 0.312 (type III), see Fig. 3e. These observations suggest that ion irradiation does not change the structure type of these films, at least in the investigated dose range. This can be due to the lack of ion energy and irradiation dose for the nanocrystalline structure destabilization and amorphous structure crystallization.
a-Al(Ti)N phase) and then to amorphous a-TiZrAlN state (Fig. 1). Dualphase (Ti,Zr)1 − xAlxNy film (type II microstructure) is a nanocomposite based on the c-(Ti,Zr,Al)N solid solution grains (size of 2–6 nm) surrounded by the amorphous a-Al(Ti)N phase. Amorphous a-TiZrAlN state (type III microstructure) is characterized by a broad diffuse scattering around 2θ = 35–40° (see Table 1, Fig. 1b). To investigate possible elemental redistribution in (Ti,Zr)1 − xAlxNy films after irradiation with Xe ions the RBS experiments have been performed. Redistribution of the constituent elements in the films was not revealed even at the maximum dose of 1 × 1017 cm−2 (Fig. 2). The distributions of the implanted Xe ions along the film depth calculated from RBS spectra are presented in the inserts of Fig. 2. From the obtained Xe distribution profiles, one can see that the penetration depth (~150 nm) does not exceed half of the film thickness. The maximum concentration of the implanted Xe in the film rises with the irradiation dose as well as
a
b
16
c-(Ti,Zr,Al)N (200)
c-(Ti,Zr,Al)N (200)
Intensity (arb. units)
Intensity (arb. units)
c-(Ti,Zr,Al)N (111)
-2
5x10 cm
1x1016 cm-2
1x1017 cm-2
5x1016 cm-2
1x1016 cm-2 a-Al(Ti)N 1x1015 cm-2
as-deposited 34
35
36
37
38
39
40
41
42
43
34
35
36
37
2θ (degrees)
38
39
40
41
42
43
44
2θ (degrees)
d
c 17
c-(Ti,Zr,Al)N (200)
-2
Intensity (arb. units)
5x1016 cm-2
a-Al(Ti)N 1x1016 cm-2 1x1015 cm-2 c-(Ti,Zr,Al)N (200)
as-deposited 34
35
36
37
38
39
40
41
c-(Ti,Zr,Al)N (200)
1x1017 cm-2
1x10 cm
Intensity (arb. units)
c-(Ti,Zr,Al)N (200)
a-Al(Ti)N
as-deposited
42
43
5x1016 cm-2
a-Al(Ti)N 1x1016 cm-2
x4
x4
1x1015 cm-2
x5
x4
as-deposited
44
34
35
36
37
2θ (degrees)
38
39
40
41
42
43
44
2θ (degrees)
e Intensity (arb. units)
a-TiZrAlN
1x1017 cm-2 5x1016 cm-2 1x1016 cm-2 as-deposited
34
35
36
37
38
39
40
41
42
43
44
2θ (degrees) Fig. 3. X-ray diffraction patterns of (Ti,Zr)1 − xAlxNy films with different Al contents: x = 0.058 (а), 0.102 (b), 0.160 (c), 0.200 (d), 0.312 (e), as-deposited and irradiated by Xe2+ ions with the energy of 180 keV at the doses of 1 × 1015–1 × 1017 cm−2.
Please cite this article as: V.V. Uglov, et al., Surf. Coat. Technol. (2014), http://dx.doi.org/10.1016/j.surfcoat.2014.03.003
V.V. Uglov et al. / Surface & Coatings Technology xxx (2014) xxx–xxx
irradiation can be an effective means to obtain a higher Al solubility in the c-TiZrN lattice. To elucidate the origin of the evolution in intensity and angular position of XRD lines after ion irradiation, a detailed stress analysis was carried out for the film with x = 0.102. The evolutions of the lattice parameter with ion fluence are reported in Fig. 4, using sin2ψ plots. In such a representation, the slope of the sin2ψ line is proportional to the inplane biaxial stress in the film. In the as-deposited state, the cubic TiZrAlN crystallites with (200) orientation are under large compressive stress. One can see that the slope of the sin2ψ lines gradually decreases with an ion fluence in the range 1 × 1015 to 1 × 1016 cm−2, and becomes eventually positive at 5 × 1016 cm−2, attesting of the development of a net tensile stress. At the highest ion fluence used (1 × 1017 cm−2), the stress appears almost relaxed. Interestingly, all sin2ψ lines intersect at a single point, whose ordinate corresponds to a stress-free lattice parameter a0 ~ 4.402 ± 0.004 Å. This value is lower than that of cubic TiZrN compound with similar Ti:Zr ratio which equals to 4.45 Å, taking into account the positive deviation from Vegard's law [19], suggesting that Al substitutes at metal sublattice of the c-TiZrAlN solid solution. The development of a net tensile stress after irradiation could be related to ion-induced annihilation of growth-induced point defects together with improvement of crystallinity, despite the accumulation of substantial Xe impurities in the film (see Fig. 2). This suggests that Xe may interact with the radiation vacancies and form gas–vacancy complexes (e.g. voids and/or bubbles) at preferential sites (grain boundary) without inducing significant compressive stress source. In Fig. 5, the evolution of nanoindentation hardness of (Ti,Zr)1 − xAlxNy films vs. Al content x is reported for both as-deposited and Xe irradiated states. With the rise of Al content the hardness of the nanocrystalline films (type I) increases and reaches a maximum value of 26 GPa for x = 0.072. The further increase in aluminum concentration leads to a drop of hardness and for the amorphous film it equals to 14 GPa (Fig. 5). Such decrease in hardness is apparently connected with the gradual decrease in crystallinity degree of the structure at the high aluminum concentrations. Under irradiation with Xe ions (1 × 1016 cm− 2) the hardness of nanocrystalline films does not essentially change but the hardness of nanocomposite and amorphous films starts to decrease. At the dose of 5 × 1016 cm− 2 the hardness of nanocrystalline films decreases by 10–17%, and that of nanocomposite and amorphous films by 32–35%. The reduction in hardness can be explained by the xenon accumulation and concurrent growth of radiation damages of the crystal lattice, despite the higher Al solubility achieved in the cubic phase.
4.50 as-dep 1x1015
4.48
1x1016 4.46
lattice parameter (Å)
However, irradiation of nanocomposite films (0.102 ≤ x ≤ 0.200, type II) leads to their transformation into a nanocrystalline state at intermediate doses (1 × 1015 to 5 × 1016 cm−2), as revealed by the emergence of a XRD line in the vicinity of the (200) reflection of the c-(Ti,Zr,Al)N solid solution. An increase in intensity and reduction of peak width is noticed with ion dose (Fig. 3b–d), suggesting improvement of film crystalline quality, until this latter eventually degrades at the maximum dose of 1 × 1017 cm−2. Radiation-stimulated diffusion processes may lead to the migration of Al and Ti atoms to form a homogenous c-(Ti,Zr,Al)N solid solution from the initial phase-separated state. It may be envisaged that grain growth of this homogeneous cubic solid solution occurs from the ‘dissolution’ of the partially segregated а-Al(Ti)N interfacial phase. At higher dose, the presence of excess impurity Xe atoms reduces the coherency length of the crystals, which explains the loss in diffracted intensity of the c-(Ti,Zr,Al)N (200) line at 1 × 1017 cm−2. Therefore, it can be concluded that ion irradiation can cause a radiation-induced crystallization for highly metastable (Ti,Zr)1 − xAlxNy films, and that this structural evolution occurs mainly because of the generation of non-equilibrium point defects (‘radiation transport’) which can ensure the diffusion of elemental species. Further chemical analysis, using atomic resolution probe techniques, would be desirable to identify the diffusion species. Additionally, a splitting of the diffraction peaks of the c-(Ti, Zr,Al)N solid solution into two components is found for irradiated (Ti,Zr) 1 − x Al x N y films with x = 0.058, 0.102 and to a lesser extent with x = 0.160 (Fig. 3). In the cases of x = 0.160 (at the doses of 1 × 1016 cm−2 and higher) and x = 0.200 the peak sharing is not noticeable because of their low intensity starting from asdeposited state. A satisfactory fit of the XRD line profile is obtained when considering two contributions (using Lorentz functions) centered at a region of larger and smaller angles (x = 0.058, Fig. 3a) or both to the region of larger angles (x = 0.102, 0.160, Fig. 3b,c) as compared to the initial peak position. It is possible to interpret these results from the following ‘double-layer’ structural model. According to RBS data analysis, the coatings irradiated with Xe can be considered as the films consisting of two layers: the first one (near the surface) is implanted by Xe ions and the second one (near the substrate) is not subjected to implantation. In this case, the first layer will have the high level of the compressive stresses caused by radiation defects and accumulation of implanted Xe atoms that likely occurs at the grain boundaries. And the second layer will be characterized by the tensile stresses as a response of the coating to compensate the stresses in the first (upper) layer. The difference in peak splitting for irradiated (Ti,Zr)1 − xAlxNy films with x = 0.058 (Fig. 3a) and x = 0.102 (Fig. 3b) can be understood based on the pristine structural type: nanocrystalline vs. nanocomposite. The fact that for x = 0.102 the two contributions to the (200) XRD line are shifted to higher angles, as compared with the peak position for the as-deposited film, can be explained by the enrichment in aluminum of the (Ti,Zr)1 − xAlxNy solid solution after irradiation (likely in the first layer). This is a consequence of the dissolution of the Al(Ti)N amorphous phase under Xe irradiation, where the Al content in this partially segregated phase is higher than in c-(Ti,Zr,Al)N solid solution counterpart. With the increase of irradiation dose the intensity and broadening of the peak of this solid solution are being changed that is connected with the competing processes of radiation crystallization and defect formation. Similar processes are revealed for x = 0.160 and 0.200 also. The lattice parameter of c-(Ti,Zr,Al)N solid solution, deduced from XRD patterns on irradiated films reported in Fig. 3, decreases from 4.455 to 4.366 Å with the rise of Al content in the (Ti,Zr)1 − xAlxNy films from 0.102 to 0.200, as expected since Al atoms have a lower atomic radius than Ti and Zr ones. Although the stress state may influence the values of the lattice parameter derived from a simple analysis of peak position (which corresponds here to out-of-plane lattice spacings) this variation supports the main finding that Xe
5
5x1016 1x1017
4.44 4.42 4.40 4.38 4.36 4.34 0
0.2
0.4
0.6
0.8
1
sin2 ψ Fig. 4. sin2ψ plot evolution with ion irradiation fluence for the (Ti,Zr)1 − xAlxNy film with x = 0.102.
Please cite this article as: V.V. Uglov, et al., Surf. Coat. Technol. (2014), http://dx.doi.org/10.1016/j.surfcoat.2014.03.003
6
V.V. Uglov et al. / Surface & Coatings Technology xxx (2014) xxx–xxx
28
type I
type II
type III
26
Hardness (GPa)
24
as-deposited irradiated by Xe (1x1016 cm-2) irradiated by Xe (5x1016 cm-2)
22 20 18 16 14
At the Xe irradiation with the dose of 1 × 1016 cm−2 the hardness of nanocrystalline films practically does not change and it starts to decrease for nanocomposite and amorphous films. The increase in irradiation dose up to a value of 5 × 1016 cm−2 leads to the reduction of the hardness of (Ti,Zr)1 − xAlxNy films (by 10–17% for nanocrystalline and 32–35% for nanocomposite and amorphous films). Therefore, in spite of the achievement of a higher Al solubility in the cubic TiZrN lattice under Xe ion irradiation, there is no expected enhancement of film hardness because of high Xe accumulation in the film structure.
12 10
Conflict of interest
8 0,00
0,04
0,08
0,12
0,16
0,20
0,24
0,28
0,32
Al content, x Fig. 5. The evolution of the hardness of as-deposited and irradiated (Ti,Zr)1 − xAlxNy films with varying Al contents.
4. Conclusions Quaternary (Ti,Zr)1 − xAlxNy thin films were synthesized by a reactive unbalanced magnetron sputtering method in the following structural states depending on the aluminum content in metal sublattice, x: • single-phase, nanocrystalline, cubic c-(Ti,Zr,Al)N solid solution with a typical grain size of 17–14 nm, for 0 ≤ x ≤ 0.072, • dual-phase nanocomposite consisting of c-(Ti,Zr,Al)N (grain size of 6–2 nm) and a-Al(Ti)N, for 0.102 ≤ x ≤ 0.200, • amorphous a-TiZrAlN phase, for x = 0.312. The maximum values of hardness (24–26 GPa) are registered for the films with 0.058 ≤ x ≤ 0.128. When further increasing the Al content, their hardness considerably decreases due to the amorphization of the films. Irradiation by xenon ions (180 keV, 1 × 1015–1 × 1017 cm−2) does not affect the structural-phase state of the nanocrystalline (x = 0.058) and amorphous (x = 0.312) films. However, the structure of nanocomposite films (0.102 ≤ x ≤ 0.200) is transformed into the nanocrystalline state, with the formation of a homogeneous cubic solid solution at intermediate doses (1 × 1015 to 5 × 1016 cm−2). Such structural phase transformation is mediated by radiation-stimulated atomic transport in the collision cascades, and dissolution of the segregated a-Al(Ti)N interfacial phase. Redistribution of the constituent elements along the film depth does not occur under irradiation by Xe ions. It was revealed that the maximum concentration of implanted Xe rises with the increase in Al content and it reaches 21.8 at.% for x = 0.312 at the dose of 1 × 1017 cm−2.
None.
Acknowledgments The work was supported by the Belarusian Republic Foundation for Fundamental Research (F13F-003). The authors are very grateful to the engineers of the Institute P′ of the University of Poitiers (France) Dr. Philippe Guerin and Marc Marteau for their help in the realization of synthesis and irradiation of the films. References [1] S. PalDey, S.C. Deevi, Mater. Sci. Eng. A 342 (2003) 58. [2] F. Vaz, L. Rebouta, M. Andritschky, M.F. Silva, J.C. Soares, J. Eur. Ceram. Soc. 17 (1997) 1971. [3] L. Rogström, L.J.S. Johnson, M.P. Johansson, M. Ahlgren, L. Hultman, M. Odén, Thin Solid Films 519 (2010) 694. [4] Z.-J. Liu, P.W. Shum, Y.G. Shen, Thin Solid Films 468 (2004) 161. [5] J.M. Castanho, M.T. Vieira, J. Mater. Proc. Technol. 143–144 (2003) 352. [6] L. Chen, J. Paulitsch, Y. Du, P.H. Mayrhofer, Surf. Coat. Technol. 206 (2012) 2954. [7] M.-H. Tuilier, M.-J. Pac, M. Gîrleanu, G. Covarel, G. Arnold, P. Louis, C. Rousselot, A.-M. Flank, J. Appl. Phys. 103 (2008) 083524. [8] P.H. Mayrhofer, D. Music, J. Schneider, J. Appl. Phys. 100 (2006) 094906. [9] I.A. Saladukhin, G. Abadias, A. Michel, S.V. Zlotski, V.V. Uglov, G.N. Tolmachova, S.N. Dub, Thin Solid Films 538 (2013) 32. [10] G. Abadias, I.A. Saladukhin, V.V. Uglov, S.V. Zlotski, D. Eyidi, Surf. Coat. Technol. 237 (2013) 187. [11] B. Yang, L. Chen, Y.X. Xu, Y.B. Peng, J.C. Fen, Y. Du, M.J. Wu, Int. J. Refract. Met. Hard Mater. 38 (2013) 81. [12] V.V. Uglov, V.M. Anishchik, S.V. Zlotski, G. Abadias, Surf. Coat. Technol. 200 (2006) 6389. [13] V.V. Uglov, V.M. Anishchik, V.V. Khodasevich, Zh.L. Prikhodko, S.V. Zlotski, G. Abadias, S.N. Dub, Surf. Coat. Technol. 180–181 (2004) 519. [14] V.V. Uglov, V.M. Anishchik, S.V. Zlotski, G. Abadias, S.N. Dub, Surf. Coat. Technol. 202 (2008) 2394. [15] G. Abadias, L.E. Koutsokeras, S.N. Dub, G.N. Tolmachova, A. Debelle, T. Sauvage, P. Villechaise, J. Vac. Sci. Technol. A 28 (2010) 541. [16] http://www.srim.org. [17] G. Abadias, Surf. Coat. Technol. 202 (2008) 2223. [18] http://www.rzg.mpg.de/mam/. [19] G. Abadias, V.I. Ivashchenko, L. Belliard, Ph. Djemia, Acta Mater. 60 (2012) 5601
Please cite this article as: V.V. Uglov, et al., Surf. Coat. Technol. (2014), http://dx.doi.org/10.1016/j.surfcoat.2014.03.003