Microstructure and properties of AlCr and AlCrFe coatings deposited by magnetron sputtering

In this work, AlCr and AlCrFe coatings were deposited via magnetron sputtering on technical substrates of carbon steel C45 and titanium grade 2. The coatings feature an amorphous structure, a thickness of 3–4 μm, and are uniform in terms of thickness and chemical composition. No major defects were visible at the interface; however, there is no diffusion at the interface, which indicates the adhesive type of bonding and results in relatively low adhesion. The coatings are characterized by a hardness of about 8–10 GPa and a Young modulus above 150 GPa. Both values are slightly higher for steel substrate. The coatings assure good corrosion resistance for steel substrate while underperforming those of Ti grade 2.


Introduction
One of the crucial challenges in materials science is to protect materials against the aggressive environment and extend the lifetime of products, thus reducing the cost, materials usage, and energy. In this context, a number of surface modification methods have been developed. In recent years, a growing research interest has been directed to ternary and quaternary metallic systems with complex crystal structures, which exhibit a unique combination of properties such as low friction coefficient, low surface energy, high electrical resistivity, hardness, low thermal conductivity, good oxidation, and corrosion resistance. This makes them a very attractive and versatile coating material. However, research in this area is rather scarce. The first approaches to produce such coatings based on Al were thermal spraying methods such as plasma spraying or high-velocity oxy-fuel spraying (HVOF), often used to improve surface properties such as hardness, wear, and/or corrosion resistance. Zhou et al. investigated the friction and wear resistance of AlCuFeCr coatings deposited on Ti alloy via the low-pressure plasmaspraying method. They revealed that the investigated coatings exhibit excellent wear resistance under dry sliding wear test conditions [1].
This approach has, however, many drawbacks. During plasma spraying, the particles of initial powder are accelerated, melted partially or completely, and deposited on the substrate in the form of coating, which results in relatively high porosity [2,3]. Additionally, rapid cooling generates internal stress that often leads to the formation of cracks. The HVOF method is more effective as the deposited coatings are less porous, but during sputtering, the initial powder undergoes a phase transition, and the particles partially oxidize.
Other routes of obtaining ternary and quaternary coatings are based on physical deposition. In this method, the initial material is transformed into a vapor state and then deposited onto a substrate. Produced coatings of thickness up to several microns are usually characterized by low contamination and uniform thickness and morphology. The first ternary coating was AlCuFe alloy, as this system has a high structural quality and is easier to process than other ternary or quaternary systems [4,5]. Klein et al. [5] obtained AlCuFe coating of high chemical and phase homogeneity by first sputtering Al, Fe, and Cu layers consecutively on SrTiO 3 substrate at room temperature using the RF magnetron sputtering and then annealing. Yoshioka et al. [6] investigated samples of AlCuFe alloy produced by the PVD method. A couple of years later, Eisenhammer et al. [7,8] investigated the possibility of applying AlCuFe coatings as selective absorbers using sputter deposition without post-annealing treatment. The first quaternary system, AlCuFeCr, was deposited from a metal alloy target on MgO and Al 2 O 3 substrate using magnetron sputtering [9]. Authors reported that as-deposited coating was amorphous and, upon subsequent annealing, transformed into a mixture of quasicrystalline and crystalline phases. However, no further investigations concerning the functional properties of these coatings were presented.
In the following years, only a few attempts were made to deposit AlCuFeCr of AlCrFe coatings by different PVD methods [10,11]. Daniels [11]. However, all of the mentioned works were focused on the characterization of deposited coatings in terms of their thickness, microstructure, chemical and phase composition, while their functional surface properties were not investigated.
The aim of the present work was to deposit such a type of coatings on technical substrates such as carbon steel and titanium and thoroughly characterize them in terms of not only microstructural features but also functional properties such as corrosion performance, hardness and adhesion to substrates. Our previous studies on bulk AlCuFeCr alloy revealed that the addition of copper significantly decreases the resistance to uniform corrosion in a strongly acidic and strongly alkaline environment, and the chromium content is crucial for high corrosion resistance [12]. Therefore, we decided to concentrate on AlCrFe and AlCr systems, as they show the potential to exhibit very high protective properties against corrosion.

Substrates
The coating was deposited on two substrates, i.e., Ti grade 2 and carbon steel C45. The nominal compositions of Ti grade 2 acc. to ASTM B 348 is as follows: Fe max. 0.3, O max. 0.25, C max. 0.08, N max. 0.03, H max. 0.015 (all in weight %). The nominal composition for steel C45 is as follows: C 0.45, Mn 0,65%, Si 0.2%, P max. 0.035, S max. 0.04 (all in weight %). The materials were provided in the form of rods whose diameters were 25.2 mm. Before magnetron sputtering, the substrate was grinded and polished up to 1 μm and cleaned with acetone in an ultrasonic bath.

Coatings
Two different coatings, AlCrFe and AlCr, were deposited both on Ti grade 2 and C45 substrates in a DC magnetron sputtering process using original magnetron systems made by Łukasiewicz Research Network-Institute for Sustainable Technology in Radom (Ł-ITeE Radom) with three circular magnetrons placed at an angle of 120° to each other. Three targets made of Al, Fe and Cr were used in the deposition process. The diameter of targets was d = 140 mm and thickness g = 7 mm. The diagram of the magnetron system used is shown in Fig. 1. The AlCr and AlCrFe coatings were deposited in an atmosphere of pure argon (Ar 100%). Prior to being placed in the process chamber, the samples were washed in pure alcohol 99.9%. Immediately before the coating process, the samples were ion-etched in the Ar + plasma. The sample temperature was stabilized at 240 °C using resistance heaters during the entire process. The deposition time of each coating was 1 h. The parameters of the process are listed in Table 1.

Microstructure
To reveal the phase composition and/or crystalline structure of the coatings, X-ray diffraction has been carried out on a Bruker D8 Advance diffractometer (goniometer radius 280 mm), equipped with parallel beam optics and a Cu Kα radiation (λ = 0.154056 nm) source that operates at 40 kV and 40 mA. The scan optics are a Göbel mirror on the incident beam side and twin secondary Soller slits in the diffracted beam. The measurements were performed with a fixed angle of incidence of the primary beam equal to 2˚, over the 2θ range between 15° and 120°, with a step size of 0.025° and 3 s counting time per step.
Scanning electron microscopy (SEM) observations were performed using a Hitachi Su8000 SEM microscope equipped with an Energy Dispersive X-Ray Spectroscope (EDS) to investigate the coating's thickness and chemical composition. The observations were made on the crosssections prepared by mechanical grinding and subsequent polishing. The area for observations was prepared using an IM4000 Hitachi ion milling system.

Nanoindentation
Nanohardness measurements were carried out using Vantage Alpha Nanotester (Micromaterials Ltd, UK) equipment. Maximum load of 5 mN, loading time of 10 s, dwell period of 5 s and 10 s unloading time were applied. For every sample, 25 indentations were performed, and mean values with standard deviation were calculated.

Scratch test
Adhesion of the coatings to the substrate was analyzed using a scratch tester Revetest (CSM Instruments, Switzerland) with a Rockwell diamond indenter with 0.2 mm diameter. During the test, the pressure force, friction force, penetration depth, and the acoustic emission signal associated with layer cracking and decohesion were recorded continuously. The tests were performed with the following parameters: scratch length 8 mm, progressive normal force from 1 to 20 N. The microscopic observation of the scratch was carried out on the light microscope Eclipse LV150N (Nikon, Japan).

Corrosion testing
All the corrosion measurements were performed using Autolab PGSTAT302N potentiostat/galvanostat (Metrohm, Switzerland). The test solutions were 0.5 M H 2 SO 4 and 0.6 M NaCl. A standard three-electrode setup with a platinum plate as a counter electrode, a silver chloride electrode (Ag|AgCl|Cl − ) as a reference electrode, and the sample as the working electrode (exposed surface of 0.2 cm 2 ) was used. The potentiodynamic polarisation was carried out after 1 h of immersion, with a 0.001 V/s scan rate, starting from 0.3 V below the open circuit potential (OCP) and stopped when the current reached a value of 1 mA/cm 2 or when the potential value reached 1 V/Ref. The electrochemical impedance spectroscopy (EIS) test was conducted only in 0.5 M H 2 SO 4 after 1 h of immersion. The frequency scans were conducted using a ± 10 mV sinusoidal wave perturbation versus the OCP. The frequency range was from 100 kHz to 10 mHz, obtaining 10 points per decade.

Microstructure characterization
The XRD patterns of the four coatings investigated here are shown in Fig. 2. Wide diffraction peaks can be seen, indicating that there are no obvious crystalline phases in the coatings, and they have a high content of amorphous phase. The XRD peaks that appeared for AlCrFe coating on Ti grade 2 correspond to the substrate.
SEM images, together with chemical analysis of crosssections of all the coatings studied, are shown in Fig. 3. The coatings feature a thickness of 3-4 µm and a lack of diffusion interface, which indicates the adhesive type of bonding. The coatings are compact and homogenous, and the elements are uniformly distributed throughout the coatings. No large discontinuities or pores are observed for AlCr and AlCrFe coating on steel. However, the AlCrFe coating deposited on Ti grade 2 has a columnar structure, and the discontinuities between the columns can be seen, as illustrated in Fig. 3d.
The averaged chemical composition of the investigated coatings, as measured by EDS is given in Table 2. Both AlCr coatings consist of about 67% of aluminium and 33% of chromium. For both AlCrFe coatings, aluminum content is lower (about 50%), chromium is about 20%, and iron is nearly 30%.

Hardness and young modulus
The results of hardness and reduced Young's modulus measurements using the nanoindentation method are shown in Fig. 4. AlCr coatings, regardless of the substrate on which they were produced, have a hardness of 8.1 GPa and a reduced Young's modulus of about 172 GPa for the layer made on steel and 157 GPa for the layer made on titanium. The AlCrFe layer on the steel has a hardness of 9.8 GPa. In this case, the value obtained significantly differs from that measured for the AlCrFe coating on a titanium substrate, which is approximately 5.1 GPa, while reducing the value of the reduced Young's modulus to 136 GPa compared to 183 GPa obtained for the coating made on steel. This might be caused by the significantly higher intercolumnar porosity of this coating, as described in the previous section. These results are in good agreement with the literature data. He et all [13] revealed that the microhardness of AlCr magnetron sputtered coatings depends on the Cr content. With higher Cr content, the microhardness increased from 205 HV for magnetron-sputtered pure aluminium coating to 1011 HV for Al50Cr50 coating. The values obtained for the coating with 30% Cr content were about 784 HV, corresponding to 7.67 GPa. This value is comparable with the results obtained for our AlCr coatings. Similar nanohardness results were obtained for bilayered AlCr thin film [11] with an atomic ratio of 1:1 where the measured value was 830 HV, corresponding to 8.04 GPa.
The hardness for AlCrFe coating depends not only on the Cr but also on Fe content. For Al94fe3Cr3 cold sprayed coatings [14], the hardness values were significantly lower (1.95 GPa) when compared to our AlCrFe coatings, which can be related to lower Cr and Fe content. Młynarek-Żak et al. [15] examined the mechanical properties of AlCrFe bulk material. They obtained the hardness value of 900 HV (8.83 GPa) for Al65Cr20Fe15 and lower value of 700 HV (6.87 GPa) for higher aluminium content Al71Cr24Fe5. These values correspond with the nanohardness values obtained for AlCrFe coating on the steel substrate. Lower nanohardness values of 6.0 Gpa were obtained for Al85Cr5Fe10 PVD coating [16] with higher Al content.

Scratch testing
The results of coatings adhesion measurements are presented in Fig. 5. Two graphs are showing the acoustic emission signal as a function of the normal force and the friction force and penetration depth as a function of the indenter displacement, correlated with the scratch image and enlarged places of the first damage to the coatings and the end of the scratch for each coating. In the case of the AlCr coating on C45 steel, the first layer cohesion cracks (LC1) occurred at a load not exceeding 2N and were present until the end of the crack (Fig. 5a). No delamination of the coating was observed in the entire range of load, which proves its high adhesion to the substrate despite low cohesion. The AlCr coating on Ti delaminated at a load of 1.7N. Delamination of the coating occurred both in the indenter-substrate contact area and in extensive areas on both sides of the scratch (Fig. 5b). From the moment of the first delamination to the end of the scratch, the coating was removed from the substrate along the entire length.
AlCrFe coatings (Fig. 5c and Fig. 5d) on steel and Ti substrates delaminate at a load of approx. 2.2N. On the steel substrate, after the first delamination, the coating was still locally present in the crack area up to a load of approx. 3.5 N. The AlCrFe coating produced on the Ti substrate, similarly to the AlCr coating, was removed along the entire scratch length after the first delamination. The resulting damage covered large areas on both sides of the scratch.
There is no literature data regarding the adhesion of AlCr or AlCrFe coatings produced by magnetron sputtering on steel and titanium. The adhesion of coatings depends mainly on the substrate material, its mechanical properties, and surface morphology. He et al. [13] examined the adhesion of AlCr magnetron sputtered coatings on the Nd-Fe-B magnets and the obtained adhesive strength was much higher (53 to 65 N LC2). The poor performance of our AlCr and AlCrFe coatings indicates that no strong bonding is formed between the coating and substrate during the magnetron sputtering. Simultaneously, their high stiffness and cohesion lead to delamination on the large area due to tangential forces. The delamination is observed as soon as the intender penetrates the coatings. Low surface roughness increases the susceptibility to delamination due to the limited contact area between substrate and coating and the absence of a geometrical barrier that would limit the coating displacement. Higher surface roughness would lead to the formation of local spots where transverse cracking of the coating occurs. Transverse cracks of the coatings limit their cohesion, stress transfer through the coating and delamination.

Corrosion resistance
The potentiodynamic polarization curves, recorded after 1 h immersion in 0.5 M H 2 SO 4 , are shown in Fig. 6a. Despite the difference in chemical composition between AlCrFe and AlCr coatings, the corrosion potential of AlCrFe coating on steel and both AlCr coatings is very similar. Only AlCrFe coating on Ti grade 2 showed more noble corrosion potential. All the coatings showed pseudo-passive behavior as the current within the plateau is higher than µA per area. The highest current density, related to the most severe corrosion attack, was observed for AlCrFe coating on steel. The AlCrFe coating on Ti grade 2 and AlCr coating on steel had current density within the pseudo-passive plateau one order of magnitude lower, which indicated improved resistance against corrosion in an acidic environment. For AlCr coating on Ti grade 2, the current density was significantly lower than for other coatings and maintained the value of 20 µA/ cm 2 up to 1 V/Ref.
The impedance data are shown as Nyquist plots in Fig. 6b. Both AlCrFe coatings showed a very small semi-circle with an inductive loop at the high frequencies related to the adsorption process of corrosion products on the surface. The semi-circle for AlCr coatings is much larger, indicating higher corrosion resistance in an acidic environment. Yet still, inductive loops at high frequencies indicate ongoing corrosion processes.
The observations of OCP during 1 h immersion in 0.6 M NaCl are shown in Fig. 7a, while potentiodynamic polarization curves recorded after immersion are shown in Fig. 7b. The OCP of all the coatings, except AlCr coating, on Ti stabilized between − 0.5 and − 0.6 V/Ref. For AlCr coating on Ti, the increase from the potential of about − 1.1 up to − 0.8 V in the first 1200 s was observed and then the OCP stabilized at about − 0.75 V/Ref. The polarization curves for all the coatings showed a similar shape: upon reaching the corrosion potential, a fast increase of current density is observed up to the passive plateau. The following abrupt increase of the corrosion current at the breakdown potential is related to the onset of pitting corrosion. The lowest corrosion potential and widest passive domain was observed for AlCr coating on Ti. The highest breakdown potential was observed for AlCrFe coating also on Ti. For both coatings on steel, the breakdown potential was lower. However, the AlCr coating showed the lowest passive current density while AlCrFe coating had one order of magnitude higher passive current, indicating its highest susceptibility to localized attack.
There is almost no literature data regarding the electrochemical behavior of AlCr and AlCrFe magnetron sputtering coatings. Based on our findings, it can be concluded that the higher content of Cr leads to the higher corrosion resistance of AlCr coating compared to AlCrFe coating on 45 steel substrate [12]. However, the obtained resistance is still significantly below the resistance obtained for the bulk material in the AlCrFe system [17]. Simultaneously, the AlCr and AlCrFe coatings on titanium grade 2 exhibit poorer corrosion performance than the substrate. Balbyshev et al. [18,19] examined the corrosion resistance of AlCuFeCr and AlCoFeCr coatings deposited via magnetron sputtering in Harrison. However, those results referred only to the impedance modulus of the examined coatings in a neutral solution, which was higher than one measured for aluminium alloy 2024, which was the substrate. No results regarding potentiodynamic behavior were discussed.  homogenous in terms of chemical composition. They also feature a relatively high hardness of 8-10 GPa. 3. Corrosion resistance study indicates that they can be used to protect steel substrate, but their corrosion resistance is insufficient for titanium. as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.