The structure and mechanical properties of Cr-based Cr-Ti alloy films

Previous studies have dealt with Cr and its alloy films that exhibit promising characteristics as surface modification layers for antiwear, anticorrosive, and decorative applications. However, the effect of Ti alloying on the structure and mechanical properties of Cr films has not been studied. This work aimed to the structure and mechanical properties of Cr-Ti alloy films in the Cr-rich side. To this end, pure Cr, Cr-6 at.% Ti, Cr-11 at.% Ti, Cr-16 at.% Ti, and Cr-21 at.% Ti alloy films were prepared by magnetron sputtering, and the structure and mechanical properties of the films were evaluated. The results indicated that all the films exhibited a Cr-based growth with body-centered cubic structure, and increasing the Ti content decreased the (110) orientation growth of Cr basis. Ti alloying increased the hardness of the films, while leaded to a monotonic decrease in the modulus of the films. The first-principles method was employed to demonstrate that the reduced modulus was determined by the Ti alloying degree, rather than the orientation evolution of the films. The analysis of H/E value suggested that the wear resistance of the films was improved by Ti alloying. The mechanical properties of present Cr-Ti alloy films, and other Cr-based alloy films or metallic glasses in publications were compared and discussed. We proposed that Ti alloying is a considerable way to explore advanced mechanical properties of Cr-based alloy films.


Introduction
Cr films attract a lot of interest due to their high mechanical strength, excellent resistance to corrosion and oxidation, and metallic lustre. These advantages generate large usage of Cr films as antiwear and anticorrosive surfaces with decorative functionalities [1,2]. Physical vapour deposition (PVD) processes, such as magnetron sputtering [3,4], thermal evaporation [5], electrodepositing [1], and cold spraying [6], are mostly used to produce Cr films in different coating systems. The relationships among process parameter, film structure, and properties have been extensively studied because controlling the process parameters is a basic way to optimize the structure and properties of the films [7][8][9]. Moreover, the alloy composition is another important factor on the structure and properties of film materials. Therefore, the present work is aimed to investigate the structure and properties of Cr alloy films.
Generally, the addition of alloy elements can improve mechanical properties of metallic materials if the effects of alloy elements are utilised adequately. On one hand, solid solution hardening is an efficient method to improve the strength of metallic materials. On the other hand, alloying could change the growth morphology, crystalline size, crystallinity, crystalline orientation, and stress state of the films [10][11][12][13][14][15], all of which can produce substantial effects on their properties. These facts have adequately promoted extensive studies on binary alloy films, such as Zr-Ti [16], Ti-Nb [17,18], Ni-W [10], Mg-Ti [19], Cu-Mo [20], and Ag-Cr [21] systems. Regarding Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
to Cr alloy systems, Chen et al [22] demonstrated that Cr-Al alloy coatings could have much better oxidation resistance than pure Cr coatings. Li et al [23] demonstrated that Cr-Ni alloy films could exhibit excellent hightemperature wear resistance. Wang et al [24] demonstrated that Cr-Fe interlayers could improve the wetting behaviours between Al and Zr 2 O 3 /Al 2 O 3 composites. The abovementioned results indicate that alloying of Cr films is a noteworthy attempt to explore the enhanced mechanical and functional properties of the films.
Ti has excellent anticorrosive and decoration properties, which is similar to Cr element. Previous studies have focused their attention on corrosion and oxidation resistances of binary Cr-Ti alloy films as an important issue of these materials [25][26][27][28][29]. More recently, Zhang et al [30] investigated the structure and mechanical properties of Cr-Ti alloy films, however, in the Ti-rich region, and they found that mechanical properties of the films were highly related to the phase constitution of Ti basis. Compared to Ti-based growth, Cr-based growth is a preferred configuration to provide high hardness and strength [31], and good wear resistance [32,33]. Furthermore, Cr-based growth is expected to decrease the fabricating cost because of the low cost of Cr element. As a result, Lee et al [34] proposed the practical application of Cr-based Cr 100-x Ti x (x=0-19 at.%) film as an underlayer for magnetic coating systems. However, to the best of our knowledge, the structure and mechanical properties of Cr-Ti alloy films in the Cr-rich region have not been studied.
This work focused on the structure and mechanical properties of Cr-Ti alloy films in the Cr-rich region. The pure Cr and Cr-Ti alloy films were prepared by magnetron sputtering processes. Firstly, the structure evolutions of the films with increasing Ti content were studied. Secondly, the mechanical property modifications introduced by structure evolution were investigated. Finally, the mechanical properties of the present Cr-Ti alloy films and other Cr alloy systems in publications were compared and discussed.

Film preparation
Cr-Ti alloy films were prepared using high power impulse magnetron sputtering (HiPIMS) processes by codepositing pure Cr and Ti targets (Cr 99.8%, Ti 99.95%, 500 mm×94 mm×8 mm) which were connected to a TruPlusma 4000 G2 HiPIMS source (Freiburg, GER), respective. The pulse width was 200 μs, and the pulse frequency was 1500 Hz, yielding a duty cycle of 21% for all depositions. The applied voltage to the Cr target was≈470 V with an average power of 4.6 kW for pure Cr deposition, and 4.3, 4.2, 4.1, and 4.0 kW for codeposition. The applied voltage to the Ti target was≈350 V with an average power of 0.7, 1.3, 1.9, and 2.5 kW. Five groups of samples, including pure Cr films and four Cr-Ti alloy films with increasing Ti content, were prepared. The HiPIMS conditions were also employed in combination with a negative polarisation to the substrate (bias of −100 V) using a TruPlusma DC-pulsed source working at a frequency of 100 kHz and 90% duty cycle. P-type (100) silicon wafers were used as substrates. Silicon wafers were adopted for the observation on the fractographies of the films. Conductive ability of the wafers was utilized to enhance substrate current density under bias conditions, which may improve the particle bombardment effect on the growing surfaces of the films during depositions. The wafers were chemically cleaned in an ultrasonic bath with acetone and isopropylic alcohol (for 20 min each), followed by Ar etching for 30 min at 800 V inside the chamber before deposition. In addition, the wafers were fixed at the profile of a 250mmdiameter cylindrical steel frame which was rotating at 25 Hz during deposition. The separation between the profile of frame and the target was 150 mm. A base pressure of 7.0×10 −3 Pa was achieved, and the gas flow of Ar was fixed at 80 sccm for a working pressure≈0.5 Pa. The deposition temperature was fixed at 200°C and the deposition time was fixed at 180 min After each deposition, the temperature was maintained for 20 min, and then the films were cooled to 30°C in the vacuum.

Film characterization
We used a PANAlytical X'Pert PRo MPD instrument (Almelo, Netherlands) for the preliminary analysis of the phases in the samples. The XRD patterns were obtained using CuK α (40 kV, 30 mA) radiation with grazing incidence of 2°, and step size of 0.02°. The diffraction angle ranged from 30°to 90°. A Tescan MIRA3 XMU SEM apparatus (Brno, Czech Republic) was used to obtain the growth morphologies of the films. Bulk compositions of the films were detected using a SEM-based INCA energy dispersive spectrometer (EDS) with an X-MAX detector (Oxford Instruments, UK). The calibration was performed using a Ti standard, and the quantisation was processed with a built-in XPP correction procedure. The film thickness was measured using SEM observations. A summary of the main deposition parameters, film thickness, and composition obtained using EDS are displayed in table 1.
TEM and STEM analyses were performed using a FEI Themis Z spherical aberration correction TEM (Hillsboro, OR). Prior to the analyses, the cross section of the sample was prepared using a focused ion beam (FIB) system equipped with a Tescan LYRA3 GMH scanning electron microscope (SEM; Brno, Czech Republic).
Bright field (BF) technique was used to show the growth morphology of the film. The bulk composition was measured by STEM-EDS technique which was performed on an initial High-angle annular dark field (HAADF) image. Selected area electron diffraction (SAED), and high resolution transmission electron microscopy (HRTEM) with fast Fourier transform (FFT) images were used to reveal the crystalline orientation evolution which was affected by the addition of Ti alloy elements.
Nanoindentation tests were performed using an Agilent U9820A Nano Indenter G200 (Santa Clara, CA) with a diamond Berkovich indenter. The Agilent Continuous Stiffness Measurement (CSM) option was used to measure the mechanical properties as a continuous function of penetration depth with a surface approach velocity of 10 nm s −1 , and a strain rate target of 0.05 s −1 . The value of the Poisson ratio was 0.25. To minimise the substrate influence, less than one-tenth of the film thickness was taken into consideration. We averaged a total of nine indents to determine the mean hardness (H) and modulus (E) for the sample at each condition.

Calculation method
A virtual-crystal approximation (VCA) method within the density functional theory (DFT) framework [35] was used to investigate the effect of Ti alloying and crystalline orientation on the modulus of the films from a theoretical perspective. The VCA method is superior in terms of simplicity and efficiency, and has been used in the theoretical research of various alloy systems, including binary [36][37][38] and ternary [39] alloys, high-entropy alloys [40,41], alloy carbides [42,43], and nitrides [44]. In a VCA model of Cr-Ti solid solution, if two elements Cr and Ti are randomly distributed in the atomic ratio of x:1-x, a virtual element Cr x Ti 1−x will be produced to replace each alloying element Cr and Ti at the lattice site. In this work, the value of X was set as a continuous variable from 1 to 0.78 with a step size of 0.01 according to the EDS results displayed in Tab. 1. The following calculations were performed using the CASTEP module within the Materials Studio software package 7.0. Firstly, the pure Cr unit cell (X=1) was optimised in appropriate convergence conditions to obtain a lattice constant of 2.8731 Å. This result is very close to the value 2.8935 Å obtained by the position of (110) main peak of present pure Cr reflections, and the standard value 2.8839 Å of pure Cr (PDF card no. 06-0694). Secondly, the VCA models (0.99X0.78) were optimised in similar conditions, and the elastic modulus on each crystalline orientation including (100), (110), (111), (211), and (310), obtained by continuously changing the X value, were given by [45][46][47]: where S 11 , S 12 , and S 44 are elastic compliances. The terms l 1 , l 2 , and l 3 are the direction cosines: the cosine of the angle between the direction of interest [hkl] and the X-, Y-, and Z-axes (the 〈100〉 directions). , which leads to an increase in the average of the nearest neighbour distance (i.e., interplanar spacing). The peak broaden indicates that the crystalline size of the Cr-Ti alloy films was refined by the alloying processes. This is because the incorporation of Ti element will increase the defect density, and disturb the crystalline growth of the Cr basis during the growth of the films.

Results and discussion
According to the Cr-Ti binary equilibrium phase diagram [48], the solid solubility of Ti in BCC Cr basis is below 2 at.% at 600°C. A heating treatment at about 1250°C is needed to produce BCC-type Cr-20 at.% Ti solid solution under equilibrium conditions. However, we produced BCC-type Cr-based Cr-Ti alloy films with Ti content up to 21 at.%. That is to say, the present Cr-Ti alloy films are supersaturated solid solutions, and they are not stable in thermodynamics. The supersaturated solution behaviours in the films can be attributed to the particle bombardment process of magnetron sputtering. It means an atomic scale heating on the growing surface of the films which companies with a highly non-equilibrium cooling process [49,50]. During the process, the solid solution with enhanced solubility at high temperature can be formed by the energy from particle bombardment, and then fast freezed into a metastable state, i.e supersaturated state. Similar results have been widely demonstrated in sputtered-deposited binary alloy systems, such as Mo-Cu [51], Al-Mo [52], and Ni-Zr [53] alloy films. Figure 2 displays the SEM fractographies of present pure Cr and Cr-Ti alloy films with increasing Ti content. From figures 2(a)-(c), pure Cr, Cr-6 at.% Ti, and Cr-11 at.% Ti films all exhibit a columnar morphology. From figures 2(d)-(e), Cr-16 at.% Ti and Cr-21 at.% Ti alloy films also exhibit a columnar growth, while the size of the column becomes more finer, and the boundary of the column becomes blurred, which suggests the decomposition of the columnar growth with the increasing Ti content. Previous studies [9] demonstrated that the decomposition of columnar growth is highly related to the change of crystalline orientation of the films. Therefore, TEM analyses of the Cr-6 at.% and Cr-21 at.% Ti alloy films were both performed to reveal the crystalline orientation evolution of the films. Firstly, the elemental composition of the Cr-6 at.% and Cr-21 at.% Ti alloy films were verified by STEM-EDS technique, as shown in figure 3. The results are fully consistent with those obtained by SEM-EDS (Tab. 1). Subsequently, we focused on the boundary zone between two columnar crystals in the BF mode, as indexed by the white frames in figure 4(a)  the film has a uniform growth during the deposition process. The SAED pattern of the Cr-6 at.% Ti alloy film shows an intensive reflection only in the (110) orientation, while the reflection in other orientations seems weak or dispersive, as shown in figures 4(e), (f). When the Ti content increases to 21 at.%, the SAED pattern has change into a circular-distributed form, which suggests a reduced orientation growth at high Ti content condition, as shown in figures 4(g), (h). To note that, no additional phase could be identified except the Cr-based phase in above TEM analyses. This result confirms that the present Cr-Ti alloy films are BCC-type Cr-based solid solutions as indicated by the XRD results. Another important conclusion that can be drawn by the TEM analyses is that Ti alloying process decreases the (110) orientation growth of Cr basis. The structure zone model (SZM) from Barna et al [14,54] gives detail on the orientation development in metallic films. The orientation development in the films stems from the minimization of overall surface and interface energies. During the process, crystallizes with densest planes are typically selected, so the film will develop an oriented structure: that is, (110) for BCC, (111) for FCC, and (0001) for HCP. This has been extensively demonstrated in publications [8,9,20]. In addition, the SZM also illustrates that the incorporation of  additives or impurities could decline the orientation growth of the basic structure of the films. For example, the incorporation of Zr will decrease the (110) growth of BCC Ti-Nb basis [12], and the incorporation of O will decrease the (111) growth of FCC Al basis [54]. It can be explained that the presence of alloying or impurity elements and their segregation to surfaces and grain boundaries have an inhibition effect on the orientation development of the basic structures. In present work, we report that the BCC Cr film could have a (110) orientation growth, and Ti alloying process could decease the orientation growth of Cr basis. These results agree well with the SZM, and the experimental results in publications. Figure 5(a) shows the hardness of present pure Cr and Cr-Ti alloy films as a function of penetration depth. The hardness of the films reaches a constant value when the penetration depth reaches 170 nm. The summary of hardness values of the films as a function of Ti content is shown in figure 5(d). The data indicate that: (i) the Cr-6 at.% Ti alloy films have a significantly improved hardness (≈ 12.5 GPa) compared to that of pure Cr films (≈ 9 GPa [55,56]). Such case is attributed to the solid solution strengthening effect and the fine-grain strengthening effect which are both caused by the Ti alloying process. (ii) Cr-Ti alloy films exhibit a slight decrease in the hardness when the Ti content exceeds 6 at.%. This result can be related to the high degree of misorientation between the neighbouring grains in the Cr-Ti alloy films containing high Ti contents, which has been demonstrated by the HRTEM analyses in figure 4(d). As we know, the volume fraction of grain boundaries increases obviously in nanometers. Therefore, the deformation resistance of grain boundaries becomes a crucial factor influencing the film hardness. In the highly textured film, the sliding or movement of grain boundaries is difficult because of the increased degree of coherency associated grain boundaries, i.e., the misorientations between the neighbouring grains are small [57]. When the misorientation between the neighbouring grains increases, the sliding or movement of grain boundary become easier, which could decrease the deformation resistance of grain boundaries, and thus decrease the film hardness. The so-called texture strengthening effect has been demonstrated in nanometals, such as Zr-Ti [15] alloy films or Ni nanophases [58]. Therefore, we can infer that the weakened orientation growth denotes the relatively low hardness of Cr-Ti alloy films containing Ti content exceeds 6 at.%.   figure 5(d). The data show that the modulus decreases monotonically from 270.7 to 188.5 GPa with the increase in Ti content. The XRD and TEM results demonstrate that all the Cr-Ti alloy films grow in BCC-type Cr-based solid solution structure. As a result, two critical factors, including chemical composition and crystalline orientation of the films, can be considered for the change in the modulus. The theoretical results on the modulus of the films with different Ti contents and crystalline orientations obtained using VCA are presented in figure 5(c). The data indicates that: (i) the increasing of Ti content unanimously decreases the modulus of Cr lattice for all conditions, although the curves are not monotonous in some intervals; (ii) the (110) orientation is one of crystalline orientations which have relatively low modulus, and thus the change from (110) to other orientations or polycrystalline growth would increase the modulus. To note that the (111) orientation has the lowest modulus, however, the (111) growth is the weakest in Cr lattice according to PDF card no. 06-0694. In conclusion, theoretical results demonstrate that increasing Ti content and declining (110) orientation growth will have negative and positive effects on the modulus of the films, respectively. In terms of the nanoindentation results, the pure Cr film with (110) growth has a modulus of 270.7 GPa, which is close to the theoretical result of 290.2 GPa on (110) orientation. This discrepancy can be attributed to the high-density defects in the real samples because of the highly non-equilibrium deposition processes. When the Ti content rises, the films show a decrease of modulus, which agrees well with the theoretical results. Therefore, we can derive that the increasing Ti content, rather than the declining (110) texture growth, determines the change of the modulus of the films.
The ultimate tensile strength σ UTS of ductile metallic and alloy films can be estimated from the formula H≈3σ UTS , here H is the hardness of the film [59]. In addition, the evaluation of wear resistance of the film can be proxied by the value of H/E [60][61][62]. Generally, materials possessing a higher H/E ratio often exhibit higher wear resistance. So we can investigate the effects of Ti alloying on the ultimate tensile strength and wear resistance of the films based on the nanoindentation results. Figure 6 depicts the values of σ UTS and H/E obtained for present pure Cr and Cr-Ti alloy films (red region). It is shown that: (i) the values of σ UTS obtained for Cr-Ti alloy films (≈ 4 GPa) are higher than that obtained for pure Cr film (≈ 3 GPa); (ii) the value of H/E increases from 0.037 to 0.063 as the Ti content rises. These results indicate that Ti alloying could improve ultimate tensile strength and wear resistance of the films. We note that the increase of H/E value is denoted by the enhanced hardness when the Ti content below 6 at.%, while by the reduced modulus when the Ti content exceeds 6 at.%, as illustrated in figure 5(d). The enhanced hardness means larger resistance to plastic deformation for compressed film surface during the sliding processes, while the reduced modulus means the stresses under the contact loading can be distributed over a wider area. Accordingly, the changes of H and E values suggest that the above two effects are both conducive to the enhanced wear resistance of the films by Ti alloying. This result is different from some publications on binary alloy films [15,30], in which the alloying processes could improve the values of H/E primarily by a substantial hardening effect, despite often triggering a slight change in the modulus. For example, Wang et al [15] reported that the hardness of Zr-Ti alloy films was much more sensitive to Zr/Ti ratio than modulus was. Zhang et al [30] found that Cr alloying of Ti films could largely increase the film hardness, but slightly increase the film modulus despite of the phase evolution induced by Cr additives. These studies show a feasible way to improve the wear resistance of the films based on the control of H/E, with respect to different effects of alloying process on film hardness and modulus. Figure 6 also depicts the values of σ UTS and H/E obtained for Ti-based Ti-Cr alloy films [30,63], Cr-based Cr-Cu alloy films [56], Cr-based Cr-Ni alloy films [64], and Cr-based bulk metallic glasses (BMGs) [31,65,66] for a comparison. In the group of Cr alloy films (grey region), the present Cr-Ti alloy films exhibit a higher value region of σ UTS than the Cr-Cu and Cr-Ni alloy films, suggesting that Ti alloying is a preferred strengthening way among the alloying processes considered. This result can be simply explained by the solution strengthening effect which is highly related to the disparity of atomic radius between the basic and additive elements. The atomic radius of Ti (1.45 Å) is much larger than that of Cr (1.27 Å), while the atomic radiuses of Cu and Ni (1.28 and 1.24 Å) are very close to that of Cr. Therefore, Ti alloying process can cause great lattice distortions in Cr basis, and thus produce much stronger solution strengthening effect than Cu and Ni alloying processes. For the concerned Cr-Ti binary systems, the Cr-based Cr-Ti alloy films (red region) are characterized by relatively high values of σ UTS compared to the Ti-based Ti-Cr alloy films considered (green region), while inversely the Tibased Ti-Cr alloy films may have a higher value of H/E than the Cr-based ones. The advantage of H/E of the Tibased alloy films can be explained by the intrinsic low modulus of Ti element. The Cr-based BMGs (yellow region) undoubtly exhibit the highest value of σ UTS among the materials considered. It is because BMGs have the extremely strong strengthening effect from the metastable multielement structures. In spite of this, the Cr-based Cr-Ti alloy films could possess the values of H/E comparable to those of BMGs, which are attributed to the significantly reduced modulus under Ti alloying conditions. It follows from the above that Ti alloying is considerable way to explore the enhanced mechanical properties of Cr-based alloy films.

Conclusions
This work focused on the structure and mechanical properties of Cr-based Cr-Ti alloy films. Pure Cr, Cr-6 at.% Ti, Cr-11 at.% Ti, Cr-16 at.% Ti, and Cr-21 at.% Ti alloy films were prepared using magnetron sputtering process by co-depositing pure Cr and Ti targets, respectively. We investigated the structural properties of the films by XRD, SEM, and TEM techniques. The results show that all the films have a body-centered cubic Crbased solid solution structure, and Ti alloying decreases the (110) orientation of Cr basis. We used nanoindentation technique to assess hardness and modulus of the films. The results show that Ti alloying not only increases the hardness, but also decreases the modulus of the films. We investigated the modulus of the films using first-principles calculations by considering the effects of chemical composition and crystalline orientations. The calculation results demonstrate that Ti content plays a key role in determining the decrease in the modulus of the films. We evaluated the effect of Ti alloying on wear resistance of the films by the H/E values. The result shows that the H/E value increases from 0.037 to 0.063, suggesting an improvement of wear resistance of the films by Ti alloying process. Finally, we compared the mechanical properties of present Cr-Ti alloy films with those of other Cr alloy films or Cr-based BMGs in publications. The results suggest that Ti alloying process is a developable way to explore the enhanced mechanical properties of Cr-based alloy films.