SCT-19736; No of Pages 15 Surface & Coatings Technology xxx (2014) xxx–xxx Contents lists available at ScienceDirect Surface & Coatings Technology journal homepage: www.elsevier.com/locate/surfcoat Development of (Cr,Al)ON coatings using middle frequency magnetron sputtering and investigations on tribological behavior against polymers N. Bagcivan, K. Bobzin, T. Brögelmann ⁎, C. Kalscheuer Surface Engineering Institute, RWTH Aachen University, Kackertstr. 15, 52072 Aachen, Germany a r t i c l e i n f o Available online xxxx Keywords: PVD (Cr,Al)ON Plastic processing Polymers Adhesion Tribological tests a b s t r a c t Plastic processing is an enormous and expanding commercial sector. Injection molding and extrusion are common techniques for worldwide efficient mass production of plastic components with high shape accuracy and high surface quality for a broad variety of applications. Adhesive and abrasive wear as well as corrosion taking place during the production of plastic products and high deforming forces lead to the necessity of developing new material concepts. Ternary nitride and oxide hard coatings deposited by PVD (physical vapor deposition) find widespread application as hard protective coatings against wear and corrosion due to their outstanding tribological, mechanical and chemical properties. Besides ternary nitrides and oxides, quaternary chromium based oxy-nitride coating systems as (Cr1 − xAlx)ON are promising candidates for tribological applications revealing high potential for decreasing adhesion and deforming forces between PVD coated tool and plastics melt during the production, e.g. of optical components or microstructered parts. The present work deals with the development of chromium based oxy-nitride hard coatings (Cr1 − xAlx)ON on tool steel AISI 420 (X42Cr13, 1.2083) using middle frequency (mf) pulsed magnetron sputtering (MS) PVD technology. The aluminum content of the (Cr1 − xAlx)ON coatings was varied in the range of 10.3 at.-% and 68.0 at.-%. By means of optical emission spectroscopy (OES) the deposition process was monitored regarding Cr/Cr+ and Al/Al+ ratios and the working points were varied via oxygen gas flow. Morphology, mechanical properties, phase and chemical composition were analyzed. Adhesion behavior between (Cr1 − xAlx)ON coatings towards plastics by high temperature contact angle measurements revealed a significant impact of the coatings' chemical composition determined by depth resolved electron probe micro analysis (EPMA). Tribological model tests in a pin-on-disk-tribometer verified a positive influence of (Cr1 − xAlx)ON coatings on the adhesion behavior towards polymers which is directly linked to lowering of deforming forces. This makes oxy-nitride (Cr1 − xAlx)ON hard coatings a promising candidate for the production of optical components or microstructured parts by injection molding or extrusion. © 2014 Elsevier B.V. All rights reserved. 1. Introduction Plastics are well known as versatile materials for a broad range of applications and gaining more importance with regard to increase of efficiency and greener mobility [1]. Therefore, plastic processing is an enormous and expanding commercial sector since the worldwide manufacturing of plastic products reaches 288 mio. tons of plastic products in 2012 [1,2]. Injection molding and extrusion embossing are common techniques for efficient mass production of plastic components with high shape accuracy and high surface quality for a broad variety of applications since they easily allow geometric shaping of components and a simultaneous functionalization within one process step [3–10]. Generally, machine components and tools in plastic processing are often subject to a high complex thermo-mechanical loading and corrosive environments strongly dependent on the plastic and its filler materials and additives [11]. Especially the production of structural plastic ⁎ Corresponding author. Tel.: +49 241 809 6282; fax: +49 241 809 9306. E-mail address: broegelmann@iot.rwth-aachen.de (T. Brögelmann). parts puts high demands on surface quality of molding tools. Different kinds of wear like adhesion and abrasion especially during the filling and release phase as well as corrosion appear to have a significant influence on machine performance, reducing lifetime and increasing material costs [11,12]. With regard to microstructured tools as the embossing roll for the production of functionalized plastic parts and optical products, a chemically inert tool surface ensuring low adhesion towards the plastic melt is necessary to reduce release forces and minimize contaminations of the produced optical products [12–15]. In order to modify the surface properties of molding tools or embossing rolls, physical vapor deposition (PVD) is one of the most promising technologies to meet the challenges like wear, corrosion and low release forces [16, 17]. The low coating thickness of typically 2 and 5 μm and the high uniformity of the coating allow to coat microstructured surfaces and to preserve their geometry [13]. Due to the high mechanical loads during plastic processing a high wear resistance of the PVD coating is one of the main requirements in order to improve the durability of the tool [13–18]. Transition metal nitrides like titanium-nitride (TiN) and chromium-nitride (CrN) deposited using PVD offer a high potential as http://dx.doi.org/10.1016/j.surfcoat.2014.09.016 0257-8972/© 2014 Elsevier B.V. All rights reserved. Please cite this article as: N. Bagcivan, et al., Surf. Coat. Technol. (2014), http://dx.doi.org/10.1016/j.surfcoat.2014.09.016 2 N. Bagcivan et al. / Surface & Coatings Technology xxx (2014) xxx–xxx Fig. 1. Schematic of the configuration of industrial coating unit CC800/9 SinOx (a) and various targets used for oxy-nitride coating deposition (b). protective coatings due to good wear and corrosion resistance [19–21]. First CrN coatings were deposited in the 1980s [22]. Although hardness of TiN could not be reached [23], increased corrosion resistance in various media was observed. Furthermore, addition of aluminum (Al) into the binary system Cr–N results in an increase of hardness and influences the coating microstructure [22,24,25]. A high potential as wear protective coating is also observed [25–27]. In subsequent years it was shown that PVD technology offers a wide range of possibilities to produce ternary metastable material compounds [12,28]. Especially, ternary nitrides as (Cr,Al)N have gained much attention within the last years and are therefore intensively investigated [29–33]. Due to the high solubility of CrN to AlN, (Cr,Al)N coatings can be deposited within a wide range of chemical compositions. (Cr,Al)N coatings have been reported exhibiting a good oxidation and corrosion resistance as well as good tribological properties and high hardness [33–40]. Besides transition metal nitrides, alumina (Al2O3) and chromia (Cr2O3) coatings have gained much interest for various industrial applications due to their high hardness, chemical inertness, and wear and corrosion resistance [41–45]. Among them are wear protective coatings on highly stressed components in tribological systems and tools for machining [46,47] as well as thermal barrier coatings for gas turbine blades [48,49]. Regarding challenging tool applications as cutting, corundum-type Al2O3– Cr2O3 ((Al,Cr)2O3) coatings were subject to extensive research since they are considered having properties comparable to α-Al2O3 in terms of hardness, wear and oxidation resistance and thermal stability [45, 50–56]. The studies on (Al,Cr)2O3 coatings included investigations on growth, microstructure and phase evolution of these coatings deposited by reactive radio frequency (rf) magnetron sputtering [43–45], investigations on microstructure and thermal stability of corundum-(Al,Cr)2O3 Table 1 Process parameters for deposition of (Cr1 − xAlx)ON coatings with variable oxygen content using CC800/9 SinOx. Parameters Unit Value Target mfMS cathode 1 Power mfMS cathode 1 Target mfMS cathode 3 Power mfMS cathode 3 Pressure Process gas Gas flow (Ar) Gas flow (N2) Gas flow (O2) Bias voltage Pulse mode Frequency Reverse time – W – W mPa – sccm sccm sccm V – kHz ns CrAl20 2000 CrAl20 2000 550 Ar 150 Pressure controlled 5, 10, 15, 20, 25, 30, 35 −40 Bipolar 1850 500 coatings [52] and analysis of the temperature-dependent structural and mechanical properties of corundum and cubic type (Al,Cr)2O3 coatings [42,50,56] deposited by reactive cathodic arc evaporation (CAE). Further studies were carried out to analyze the influence of silicon on the target poisoning [41] and the microstructure and chemical composition of macroparticles incorporated during the deposition of (Al,Cr)2O3 coatings by CAE [57]. Recently, quarternary oxy-nitride coatings as (Cr,Al) ON and (Ti,Al)ON are attracting increased attention as potential hard coatings for tools in cutting and forming applications due to their mechanical properties and oxidation resistance [58–60]. Besides increasing the oxidation resistance, the incorporation of oxygen in ternary nitride coatings as (Cr,Al)N strongly influences the properties of oxy-nitride (Cr,Al)ON coatings attributed to different charges and differences in the nature of metal–anion bonds [59,61,62]. Here, the incorporation of oxygen in the nitride structure and the incorporation of the nitrogen atoms in the oxide structure in combination with the transition region between the oxide and nitride structures can be considered responsible for the favorable mechanical properties and thermal stability of the oxynitride coatings [62–65]. This is in good agreement with the results of Sjölén et al. [66] revealing an improved oxidation resistance, improved chemical stability, alloy hardening and grain size hardening after incorporation of oxygen in multi-element nitride coatings accompanied by a toughness enhancement. Recent scientific studies focused on investigations on phase transformation, structural and mechanical properties of (Cr,Al)ON coatings deposited by CAE [62,67] and magnetron sputtering techniques [62,68–70] revealing the existence of three different regions of microstructure and properties dependent on the oxygen content in the oxy-nitride coating. Further studies showed that the incorporation of active elements as yttrium (Y) and silicon (Si) influences the growth and phase development of (Cr,Al)ON coatings as well as the mechanical properties [58,63,71]. These favorable properties offer the possibility to use oxy-nitride coatings in a variety of industrial applications as diffusion barrier in gas turbines [68], in microelectronics [44] and in tribological applications for wear and corrosion protection [44,62,68,72]. With regard to the use of nitride and oxy-nitride coatings as wear and corrosion protective coatings in plastic processing, binary nitrides as TiB and CrN as well as ternary nitrides as (Ti,Al)N and (Cr,Al)N and oxy-nitrides as (Ti,Al)ON and (Cr,Al)ON were tested on components of injection Table 2 Target configuration for deposition of (Cr1 − xAlx)ON coatings at constant oxygen gas flow. Coating mfMS cathode 1 mfMS cathode 3 Oxygen gas flow [sccm] No. 1 No. 2 No. 3 CrAl20 AlCr20 AlCr20 CrAl20 AlCr20 Al 20 20 20 Please cite this article as: N. Bagcivan, et al., Surf. Coat. Technol. (2014), http://dx.doi.org/10.1016/j.surfcoat.2014.09.016 3 N. Bagcivan et al. / Surface & Coatings Technology xxx (2014) xxx–xxx Table 3 Process parameters for deposition of (Cr1 − xAlx)ON coatings with variable aluminum content using CC800/9 SinOx. Parameters Unit Value Target mfMS cathode 1 Power mfMS cathode 1 Target mfMS cathode 3 Power mfMS cathode 3 Pressure Process gas Gas flow (Ar) Gas flow (N2) Gas flow (O2) Bias voltage Pulse mode Frequency Reverse time – W – W mPa – sccm sccm sccm V − kHz ns Variable (see Table 2) 2000 Variable (see Table 2) 2000 550 Ar 150 Pressure controlled 20 −40 Bipolar 350 500 molding machines and were proven to be efficient to protect the tools from wear and corrosion [12,14,15,25,73–75]. Although these nitride coatings revealed great potential for wear and corrosion protection in plastic processing, application-oriented investigations on the adhesion behavior of chromium based quaternary oxy-nitride hard coatings (Cr1 − xAlx)ON towards the plastic melt has not been in focus yet. The objective of this research was to determine the adhesion behavior of (Cr1 − xAlx)ON towards two commercially available optical plastics (polymethyl methacrylate, PMMA) by means of application-oriented tribological model tests to reveal the huge potential of (Cr1 − xAlx)ON coated tools regarding the reduction of deficient parts, simplifying cleaning and shortening cycle times in plastic processing. Therefore, low-temperature (Cr1 − xAlx)ON coatings were deposited using middle frequency (mf) pulsed magnetron sputtering (MS) PVD technology in an industrial coating unit below 200 °C to avoid annealing of temperature-sensitive cold work steel AISI 420 (X42Cr13, 1.2083). After definition of process window by means of optical emission spectroscopy (OES), oxygen content of the (Cr1 − xAlx)ON coatings was varied in the range of 10.8 at.-% and 61.7 at.-% to investigate the oxygen influence on the adhesion towards the plastics. Afterwards, aluminum content of the oxy-nitride coatings was set between 10.3 at.-% and 68.0 at.-% in order to avoid deterioration of mechanical properties due to exceeding the maximum solubility of face-centered cubic (fcc)-AlN in (fcc)-CrN and the formation of hexagonal AlN phase [34,76]. In addition, OES was applied for monitoring of coating processes. Mechanical properties, morphology and phase composition depending on oxygen content and Cr:Al ratio measured by electron probe microanalysis (EPMA) were investigated by means of nanoindentation, scanning electron microscopy (SEM) and X-ray diffraction (XRD). Adhesion behavior between (Cr1 − xAlx)ON coatings towards commercially Table 4 Specific data of optical plastics Plexiglas 7N™ and Plexiglas Resist zk6HF™. Trade name Producer Parameters Optical parameters Transmittance Refractive index Other parameters Density Processing conditions Predrying temperature Mass temperature Tool temperature Plexiglas 7N Plexiglas Resist zk6HF Standard Unit Evonik Roehm Value Evonik Roehm Value ISO 13468-2 ISO 489 % – 92 1.49 91 1.49 ISO 1183 g/cm3 1.19 1.16 °C °C °C max. 93 220–260 60–90 max. 80 220–260 50–70 Table 5 Parameters for tribological tests in PoD-tribometer. Parameter Substrate Coating Counterpart Temperature Rel. humidity Load Rel. velocity Distance Unit Value °C % N cm·s−1 m Disk: AISI 420 (∅ = 25 mm, =8 mm) (Cr1 − xAlx)ON: O2-variation, Al-variation Pin (∅6 mm): Plexiglas 7N, Plexiglas Resist zk6HF 110 ± 5 30 ± 5 2 10 100 available polymethacrylate was determined by high temperature contact angle measurements. Application-oriented tribological model tests in a pin-on-disk-tribometer were carried out to analyze friction behavior of oxy-nitride (Cr1 − xAlx)ON coatings against plastics. 2. Experimental details Oxy-nitride (Cr1 − xAlx)ON hard coatings were deposited on cemented carbide (WC/Co, THM12, SNUN120412, Kennametal Widia Produktions GmbH & Co. KG, Essen, Germany) and on case hardened cold work steel AISI 420 (X42Cr13, 1.2083, (52 ± 3) HRC, ∅ = 25 mm, h = 8 mm) via an industrial CC800/9 SinOx coating unit from CemeCon AG, Wuerselen, Germany, equipped with two bipolar pulsed mf cathodes (Fig. 1a) using a rectangular pulse with a frequency of 18.51 kHz. Fig. 1b illustrates the targets for coating deposition. Deposition temperature of reactive mfMS process was kept below 200 °C in order to avoid annealing of temperature-sensitive cold working steel AISI 420. With regard to mechanical loads in tribological testing an (Cr,Al)N interlayer was deposited to ensure sufficient adhesion between oxynitride (Cr1 − xAlx)N coatings and AISI 420. Specimens (THM12, AISI 420) were polished to a mean roughness of Ra = 0.01 μm. For hightemperature contact angle measurements, AISI 420 was brought to Ra = 0.05 μm by sand-blasting to match the expected surface roughness of the mfMS coated specimens. Prior to coating deposition, substrate material was cleaned in a multi-stage ultrasonic bath containing alkaline solvents and dried by cleaned compressed air. Additionally, an insitu plasma etching process was provided for cleaning and activation of the substrates. During the coating deposition of the oxy-nitride coatings, the samples were moved in planetary motion (double rotation) to ensure uniform distribution of coating thickness. For investigation of the plasma emission and monitoring of the PVD deposition processes an optical spectrometer (OES) Plasus EMICON HR/2MC, Kissing, Germany was applied outside the vacuum chamber. The emitted light was conducted via collimator optic attached together with an optical fiber to the entrance of three spectrometers. The three spectrometers divided the light in three spectral areas, which are collected from CCD cameras. Thus, a spectrum with a high resolution (0.15 nm) was generated in area from 20 nm up to 860 nm without influencing process parameter Table 6 Process parameters for recording of nitrogen hysteresis using CC800/9 SinOx. Parameters Unit Value Target mfMS cathode 1 Power mfMS cathode 1 Target mfMS cathode 3 Power mfMS cathode 3 Heating power Gas flow (Ar) Gas flow (O2) – W – W W sccm sccm CrAl20 2000 CrAl20 2000 1500 150 5, 10, 15, 20, 25, 30, 35 Please cite this article as: N. Bagcivan, et al., Surf. Coat. Technol. (2014), http://dx.doi.org/10.1016/j.surfcoat.2014.09.016 4 N. Bagcivan et al. / Surface & Coatings Technology xxx (2014) xxx–xxx Fig. 2. Cathode voltage and OES Cr I and Al I intensity during nitrogen hysteresis as a function of pressure (a) and nitrogen flow (b) for oxygen variation between 5 sccm and 20 sccm. selection as heat output and cathode power. The collimator optic was clamped in front of cathode 1 (C1: CrAl20 target) with a distance of 5 mm. This configuration allows plasma monitoring close to magnetic field lines of cathode 1. 2.1. Coating deposition — variation of oxygen content Prior to deposition of oxy-nitride (Cr1 − xAlx)ON hard coatings, OES measurements were carried out to set the process window in reactive mfMS PVD process with regard to nitrogen and oxygen gas flow to avoid abrupt formation of non-conductive films on the sputtering target (target poisoning effect) resulting in non-uniform coatings [41,55,57]. Variation of the oxygen content was achieved by setting oxygen gas flow within the limits of the process window during the deposition of oxy-nitride (Cr1 − xAlx)ON coatings. Cr:Al ratio was kept constant by using two CrAl20 targets consisting of a Cr target with 20 pieces of diameter 15 mm Al plugs (purity: Cr 99.9% and Al 99.5%) (Fig. 1b). Table 1 shows the process parameters for coating deposition with variable oxygen content. 2.2. Coating deposition — variation of aluminum content In order to obtain different aluminum contents within the oxynitride (Cr1 − xAlx)ON coatings different target configuration were used (Table 2). Oxygen gas flow was kept constant at 20 sccm O2 during coating deposition. Besides a CrAl20 target consisting of a Cr target with 20 pieces of diameter 15 mm Al plugs (purity: Cr 99.9% and Al 99.5%) an AlCr20 target (Al target with 20 pieces of diameter 15 mm Cr plugs, purity: Al 99.5% and Cr 99.9%) and an Al-Target (purity: 99.5%) respectively, was used to achieve variable chemical composition (Cr:Al ratio). The aim was to set the aluminum content below the theoretical maximum solubility of cubic (fcc) AlN in cubic (fcc) CrN at 77 at.-% Al to avoid the formation of hex AlN phase accompanied by deterioration of mechanical properties and tribological behavior [34,35,76]. Process parameters can be obtained from Table 3. 2.3. Coating characterization In order to evaluate coating morphology and thickness, scanning electron microscope (SEM) ZEISS DSM 982 Gemini, Jena, Germany, micrographs of fractured cross section were taken using a secondary electron (SE) detector. For this purpose, cemented carbide specimens were coated. Chemical composition was quantitatively measured by means of electron probe microanalysis (EPMA) CAMEBAX SX 50, CAMECA SES, Gennevilliers, France. For this purpose, ball-crater measurements were carried out followed by a line scan in EPMA as shown in [77]. The low detection limit in combination with a relative accuracy of quantification the chemical composition between 1% and 5% makes EPMA a powerful tool to determine oxygen and nitrogen content within the deposited oxy-nitride (Cr1 − xAlx)ON coatings. Mechanical properties, universal hardness, HU, and modulus of indentation, EIT, of the oxy-nitride coatings were determined using the method of nanoindentation. A Nanoindenter XP by MTS Nano Instruments, Oak Ridge, TN, USA was applied for this purpose. The indentation depth did not exceed 1/10 of Fig. 3. Cathode voltage during nitrogen hysteresis as a function of pressure (a) and OES Cr I and Al I intensities as a function of nitrogen flow (b) for oxygen variation between 25 sccm and 40 sccm. Please cite this article as: N. Bagcivan, et al., Surf. Coat. Technol. (2014), http://dx.doi.org/10.1016/j.surfcoat.2014.09.016 5 N. Bagcivan et al. / Surface & Coatings Technology xxx (2014) xxx–xxx Fig. 4. Voltage at mfMS cathode 1, pressure and OES intensity of Cr I and Al I as a function of oxygen flow. coating thickness. The evaluation of the measured results was based on the equations according to Oliver and Pharr [78]. A Poisson's ratio of ν = 0.25 was assumed. The surface topography was analyzed by means of a confocal laser scanning microscope Keyence VK-X210, Tokio, Japan, according to ISO 4287 (line profile) and ISO 25178 (area measurement). Crystallographic phase analysis was carried out via high angle (HA) X-ray diffraction (XRD) using grazing incidence (GI) X-ray diffractometer XRD 3003, General Electric, Munich, Germany. Analysis was done using Cu-Kα radiation (λ = 0.15406 nm) operated at 40 kV and 40 mA using the following parameters: GI: ω = 3°; diffraction angle 2θ: 20° to 90°; step width: s = 0.05°; step time: t = 60 s. 2.4. Characterization of adhesion behavior and tribological performance Besides wear and corrosion resistance, lowering of deforming forces between tool surface and plastics can be considered to be the main objective of improving efficiency and cost effectiveness of plastic processing as injection molding and extrusion embossing. Since deforming forces are closely linked to the adhesion between tool surface and the plastics, temperature and time dependent adhesion behavior of oxynitride (Cr1 − xAlx)ON coatings and two commercially available optical plastics was analyzed. Thus, in particular with respect to plastic processing, tool temperature and dwell time of plastics in contact to the molding tool are important factors since occurring chemical interaction can be assumed to influence the adhesion between tool surface and plastics. Therefore, a contact angle measurement device Kruess DSA 10, Hamburg, Germany, supplemented by a heating chamber was used for adhesion analysis of two optical plastics based on polymethacrylate (PMMA), Plexiglas 7N™ and Plexiglas Resist zk6HF™ of Evonik Roehm GmbH, Darmstadt, Germany, and the oxy-nitride (Cr1 − xAlx)ON coatings. Specific data of the optical plastics can be obtained from Table 4. Before high contact angle measurements were done, annealing temperature was set at half of the processing temperature of the optical plastics (T = 110 °C) and kept constant for 10 min to ensure homogeneous heating of the oxy-nitride coating and the plastic granulate. Afterwards, the chamber was heated up to the processing temperature (T = 220 °C). Contact angles were measured equidistantly every 30 s at both three-phase points using a CCD camera recognizing the shape of the melted plastic granulates. Temperature was kept constant during the entire measurement time of 50 min. Derived from real boundary conditions in plastic processing, application-oriented tribological tests were carried out to determine tribological performance of the oxy-nitride (Cr1 − xAlx)ON coatings regarding continuous sliding wear and friction behavior. The focus of the tribological evaluation was set on the friction behavior of the oxy-nitride coatings against Plexiglas 7N and Plexiglas Resist zk6HF since correlation between adhesion behavior observed in high temperature contact angle measurements and friction behavior in tribological tests was expected [79]. Tribological tests were performed in a pin-on-disk (PoD) tribometer, CSM instruments, Peseux, Switzerland. Pins (∅ = 6 mm) extruded from plastic granulate were chosen as counterpart. Pins were pressed in off-center position onto the coated specimen with a constant load of 2 N. Coated and uncoated specimens were clamped into a rotation holding device and heated up to half of processing temperature of the plastics (Table 4). The parameters of the tribological tests can be taken from Table 5. 3. Results 3.1. Plasma monitoring by optical emission spectroscopy (OES) As a first step towards the development of oxy-nitride (Cr1 − xAlx) ON coatings, nitrogen hysteresis at constant argon gas flow using variable oxygen gas flow in the range between 5 sccm and 40 sccm O2 were recorded (Table 6). Target configuration remained unchanged (C1: CrAl20, C3: CrAl20). Voltage at mfMS cathode 1 (CrAl20) and intensity of Cr I (λ = 357.869 nm [80]) and Al I (λ = 763.511 nm [80]) recorded via OES during nitrogen hysteresis at oxygen gas flow in the range between 5 sccm and 20 sccm O2 as a function of pressure (a) and nitrogen flow (b) are depicted in Fig. 2. As increasing nitrogen content leads to an increasing pressure due to constant Ar flow, voltage at mfMS cathode 1 drops with increasing oxygen flow at lower pressure since formation of non- Table 7 Chemical composition and thickness of the (Cr1 − xAlx)ON coatings deposited at different oxygen flow with constant target configuration. Oxygen flow Cr Al O N Thickness Deposition rate [sccm] [at.-%] [at.-%] [at.-%] [at.-%] [μm] [μm/h] 5 10 15 20 25 30 35 45.6 34.5 36.5 38.9 32.6 28.9 28.9 3.22 2.85 2.82 2.09 1.37 1.23 1.17 1.23 1.09 1.08 0.80 0.52 0.47 0.45 ± ± ± ± ± ± ± 1.4 6.0 0.2 1.2 1.4 1.2 0.8 6.1 4.0 4.2 4.5 4.9 6.2 6.4 ± ± ± ± ± ± ± 0.3 0.6 0.2 0.2 0.1 0.5 0.2 10.8 40.0 31.8 21.9 51.0 61.0 61.7 ± ± ± ± ± ± ± 2.2 25.5 1.2 1.2 7.4 5.3 2.8 37.5 21.5 27.4 34.7 11.5 3.9 2.9 ± ± ± ± ± ± ± 2.5 19.3 1.5 0.9 6.1 2.0 1.4 Please cite this article as: N. Bagcivan, et al., Surf. Coat. Technol. (2014), http://dx.doi.org/10.1016/j.surfcoat.2014.09.016 6 N. Bagcivan et al. / Surface & Coatings Technology xxx (2014) xxx–xxx Fig. 5. SEM pictures of oxy-nitride (Cr1 − xAlx)ON coating topography as a function of oxygen gas flow. conductive Al-rich oxide islands is promoted [41,50,55,57]. OES Cr I intensity decreases with higher nitrogen content reaching a constant level at about 30 sccm N2. Beyond this point, OES Cr I intensity remains stable from which a constant sputter rate can be concluded. In contrast, OES Al I intensity decreases at nitrogen gas flow between 20 sccm N2 and 60 sccm N2 accompanied by a continuous drop in cathode voltage. Further hysteresis tests were carried out at oxygen flows between 25 sccm and 40 sccm O2 (Fig. 3). Similar to Fig. 2, cathode voltage drops as a function of pressure at oxygen gas flow in the range between 25 sccm and 40 sccm O2 (Fig. 3a). OES Cr I and Al I intensity reveal similar behavior dependent on nitrogen flow (Fig. 3b). Starting from 35 sccm O2, hysteresis changes significantly. Although increasing nitrogen flow leads to significant drop in cathode voltage, reverse process of increasing cathode voltage at lower nitrogen flow does not occur (Fig. 4a). Similar behavior can be obtained from OES measurements in Fig. 3. OES Cr I intensity and therefore sputter rate decrease with increasing nitrogen flow. Reverse behavior at decreasing nitrogen flow cannot be observed. Significant reduction of chromium intensity already occurs for 35 sccm O2 at 100 sccm N2 and for 40 sccm O2 at 50 sccm N2, respectively. A comparable reduction of intensity and sputter rate cannot be seen for aluminum. It is assumed Table 8 Results of surface roughness analysis for oxy-nitride coating deposited at different oxygen gas flow. Coating 5 sccm O2 10 sccm O2 15 sccm O2 20 sccm O2 25 sccm O2 30 sccm O2 35 sccm O2 2D roughness (ISO 4287) 3D roughness (ISO 25178) Ra Rz Rdc Sa Sz Sq Sk [μm] [μm] [μm] [μm] [μm] [μm] [μm] 0.0959 0.0756 0.0470 0.0456 0.0302 0.0279 0.0251 2.4578 1.9747 2.0718 1.1983 1.1178 0.8258 0.8100 0.1517 0.1190 0.0703 0.0665 0.0477 0.0438 0.0421 0.0997 0.0779 0.0481 0.0468 0.0310 0.0287 0.0265 2.5509 2.0777 2.1802 1.2706 1.2003 0.8716 0.8605 0.1331 0.1042 0.0699 0.0640 0.0417 0.0381 0.0370 0.2974 0.2268 0.1269 0.1249 0.0880 0.0799 0.0750 that voltage drop can be attributed to continuous formation of nonconductive layers on aluminum plugs within the CrAl20 targets with increasing oxygen [50,55,57,81,82]. Finally, oxygen hysteresis in the range between 0 and 50 sccm O2 at constant gas flow of Ar (200 sccm) and N2 (45 sccm) was carried out to determine the start of target poisoning. Fig. 4 illustrates the typical course of cathode voltage in oxygen hysteresis. Cathode voltage remains stable over a broad oxygen gas flow range until it abruptly falls off to a significant lower level. With decreasing oxygen flow cathode voltage keeps constant at low level until it jumps over to the higher level. OES Al I intensity starts to decrease at about 35 sccm O2. Complete fall of aluminum intensity occurs simultaneously to chromium intensity at about 47 sccm O2. Formation of non-conductive AlN films on the target surface leads to electrical charging of non-conductive surface decreasing the potential between target and plasma. Potential difference is responsible for reduction of ion bombardment and sputter rate due to lower sputter yield. Based on this results, oxygen flow for the deposition of oxy-nitride (Cr1 − xAlx)ON hard coatings was set between 5 sccm and 35 sccm O2. 3.2. Coating deposition — variation of oxygen content Taking into account OES measurements, oxy-nitride (Cr1 − xAlx)ON hard coatings with almost constant Cr:Al ratio and different oxygen contents ranging from 10.8 at.% O2 to 61.7 at-% O2 were successfully deposited on tool steel AISI 420 below 200 °C via mfMS PVD process using an industrial coating unit CC800/9 SinOx. Process temperature below 200 °C was determined using a drag pointer clamped on the substrate holder. Table 7 reveals the results of chemical composition determined by EPMA line scans. In order to evaluate statistical reliability of the data obtained, error bars are given in Table 7. Due to target configuration (C1: CrAl20, C3: CrAl20), the oxy-nitride coatings exhibit a very high chromium content between 28.9 at.-% and 45.6 at.-% compared to the aluminum content ranging between 4.0 at.-% and 6.4 at.-%. Topography and morphology analysis of the oxy-nitride (Cr1 − xAlx)ON hard coatings was performed by means of SEM. The SEM micrographs of the surfaces in Fig. 5 exhibit a compact and closed coating surface. Topography changes with increasing oxygen Please cite this article as: N. Bagcivan, et al., Surf. Coat. Technol. (2014), http://dx.doi.org/10.1016/j.surfcoat.2014.09.016 N. Bagcivan et al. / Surface & Coatings Technology xxx (2014) xxx–xxx 7 Fig. 6. SEM pictures of oxy-nitride (Cr1 − xAlx)ON coating morphology as a function of oxygen flow. flow. At 5 sccm O2 (a), columnar morphology takes a triangular form exhibiting sharp edges and lower density. At 15 sccm O2 (c) transformation to a denser topography accompanied by a grain agglomeration at 25 sccm O2 (e) and 30 sccm O2 (f) can be observed. The smoothing of the coating surface can be verified by means of surface roughness and topography analysis via confocal laser scanning microscopy (Table 8). 2D roughness values (Ra, Rz, Rdc) as well as 3D roughness values (Sa, Sz, Sq, Sk) decrease with higher oxygen gas flow during deposition process. SEM cross section fractures of the oxy-nitride (Cr1 − xAlx)ON Fig. 7. HAXRD for phase analysis of oxy-nitride coatings as a function of oxygen flow. Please cite this article as: N. Bagcivan, et al., Surf. Coat. Technol. (2014), http://dx.doi.org/10.1016/j.surfcoat.2014.09.016 8 N. Bagcivan et al. / Surface & Coatings Technology xxx (2014) xxx–xxx Table 9 Chemical composition and thickness of the (Cr1 − xAlx)ON coatings deposited at constant oxygen flow (20 sccm) with different target configuration. Target Cr Al O N Thickness Deposition rate [C1/C3] [at.-%] [at.-%] [at.-%] [at.-%] [μm] [μm/h] CrAl20/CrAl20 AlCr20/AlCr20 AlCr20/Al 38.9 ± 1.2 14.7 ± 0.5 8.9 ± 0.3 4.5 ± 0.2 21.9 ± 0.3 18.9 ± 0.2 21.9 ± 1.3 60.2 ± 0.9 67.8 ± 0.5 34.8 ± 1.1 3.1 ± 0.7 4.4 ± 0.3 2.09 0.71 0.63 0.80 0.27 0.24 hard coatings deposited at different oxygen gas flows in the range between 5 sccm O2 and 35 sccm O2 are shown in Fig. 6. As it is also apparent from the coatings' surface topography at 5 sccm O2 (a) and 10 sccm O2 (b) in Fig. 5, the coatings reveal a clearly visible columnar morphology in SEM cross section (Fig. 6). Starting at 15 sccm O2 (c), higher oxygen flow leads to a finer columnar morphology accompanied by a higher density in comparison to lower oxygen flow. Decreasing coating thickness from 3.22 μm to 1.17 μm for (Cr1 − xAlx)ON coatings deposited at 5 sccm O2 (a) and 35 sccm O2 (g), respectively, can also be observed in Fig. 6. The reduction of the deposition rate can be directly attributed to the higher oxygen flow which leads to preferred reaction of aluminum and oxygen and subsequently to premature formation of nonconductive layers on the CrAl20 targets. High angle (HA) XRD patterns of (Cr1 − xAlx)ON deposited with different oxygen flows in the range between 5 sccm O2 and 35 sccm O2 can be taken from Fig. 7. Aluminum content was kept constant since target configuration (CrAl20, CrAl20) has been retained unchanged. The spectra are dominated by the peaks characteristic of corundumtype alumina (α-Al2O3, JCPDS 46-1212) and chromia (Cr2O3, JCPDS 381479) which is in good agreement with the results in [43,68]. With regard to the quaternary system Cr–Al–O–N, the formation of Al2O3 and Cr2O3 is energetically favorable compared to the formation of AlN and CrN compounds since creation of Al–O and Cr–O bonds contributes to a decrease of the Gibbs free energy much more than that of Al–N and Cr–N bonds [67]. Therefore, formation of pure phases can be excluded under the deposition conditions. Instead, formation of solid solution of at least one or more of the three structures occurs: NaCl-type (Cr,Al) ON, (fcc)-(Cr,Al)2O3 − x and corundum-type (Cr,Al)2O3, whereby the latter may contain dissolved nitrogen [42,62,67,68,83]. For completeness, diffraction angles of face-centered cubic (fcc)-CrN (JCPDS 110065) and (fcc)-AlN (25-1495) are plotted in Fig. 7 suggesting a possible formation of (Cr,Al)N solid solution [67]. The assumption of the formation of corundum-type (Cr,Al)2O3 can be confirmed by the fact that the peaks are located between the peak positions characteristic of α-Al2O3 and Cr2O3 [42,68]. As described by Khatibi et al. [42] the bottom curves seem to show a broad peak at about 2θ = 44° characteristic of NaCl- type-(Cr,Al)ON or (fcc)-(Cr,Al)2O3 − x. This can be found in good agreement with the observations in [43,67,83]. With higher oxygen gas flow during the deposition process, formation of corundum-type (Cr,Al)2O3 is promoted [67,68]. The peak shift from the positions for pure Cr2O3 in the XRD spectra in Fig. 7 can be attributed to the dissolution of aluminum in the (Cr,Al)2O3 and (Cr,Al)ON structure, respectively. Consequently, the phase composition of the oxy-nitride coatings strongly depends on the N2/O2 ratio in the reactive gas mixture and the Al and Cr contents in the coating [67]. A higher content of O2 and Cr in the coating will higher the probability of the formation of corundum-type (Cr,Al)2O3 coatings. A low content of O2 combined with a high Al content will result in formation of (fcc)-(Cr,Al)2O3 coatings [67]. It must be stated that the formation of corundum structured (Cr,Al)2O3 depending on the oxygen gas flow during the deposition is not fully understood yet. By means of theoretical calculations in Alling et al. [84] it was proved that formation of (Al,Cr)2O3 solid solutions stabilized with one third-metal vacancies is possible resulting in metastable coatings with respect to the corundum (Al,Cr)2O3 solid solutions [42]. Increasing intensity of WC peaks (JCPDS 00-025-1047) with higher oxygen flow can be attributed to the decreasing deposition rate leading to reduced coating thickness (see Table 7). 3.3. Coating deposition — variation of aluminum content Since ternary nitride coatings as (Cr1 − xAlx)N can be deposited within a wide range of chemical compositions due to the high solubility of cubic (fcc) AlN to cubic (fcc) CrN [34,76], aluminum content of the oxy-nitride (Cr1 − xAlx)ON coatings was varied via target configuration (Table 2) at a constant oxygen gas flow of 20 sccm O2. Improvements of oxidation and corrosion resistance as well as favorable tribological properties due to higher hardness of aluminum content within the coatings were expected [36–40,59,60]. Chemical composition of the deposited coatings as measured by EPMA is listed in Table 9. As expected, variation of target configuration leads to higher aluminum content in the coatings. A significant decrease of coating thickness and therefore, deposition rate with higher aluminum content can be observed. Since target configuration provides more aluminum supply, Fig. 8. SEM pictures of oxy-nitride (Cr1 − xAlx)ON coating morphology as a function of aluminum content. Please cite this article as: N. Bagcivan, et al., Surf. Coat. Technol. (2014), http://dx.doi.org/10.1016/j.surfcoat.2014.09.016 N. Bagcivan et al. / Surface & Coatings Technology xxx (2014) xxx–xxx Table 10 Results of surface roughness analysis deposited with variable target configuration. 2D roughness (ISO 4287) Target configuration Ra [μm] AlCr20/AlCr20 AlCr20/Al 3D roughness (ISO 25178) Rz Rdc Sa Sz Sq Sk [μm] [μm] [μm] [μm] [μm] [μm] 0.5136 4.7525 0.8422 0.5223 4.8465 2.2310 0.5636 0.5559 5.8527 0.8727 0.5650 5.8621 3.0120 0.5450 formation of aluminum oxide on the target surface is accelerated leading to decreasing sputter yield [41,55,57]. Cross section SEM micrographs for coating morphology analysis as a function of aluminum content can be taken from Fig. 8. Independent from target formation (CrAl20, CrAl20 (a), AlCr20, AlCr20 (b) and AlCr20, Al (c)), the oxynitride coatings exhibit a fine columnar morphology. The results of the surface roughness analysis by means of confocal laser scanning microcopy can be found in Table 10 presenting a higher roughness for the oxy-nitride coatings compared to the coatings deposited at different oxygen flow (Table 8). Fig. 9 presents the HAXRD spectra of oxy-nitride (Cr1 − xAlx)ON coatings deposited at constant oxygen flow of 20 sccm O2 and various target configurations at cathode 1 and cathode 3, respectively CrAl20/ CrAl20, AlCr20/AlCr20 and AlCr20/Al. Similar to the previous HAXRD spectra in Fig. 7, peaks are characteristic of corundum-type alumina (α-Al2O3, JCPDS 46-1212) and chromia (Cr2O3, JCPDS 38-1479) as described in [43,68]. As described in Section 3.2, formation of solid solution of at least one or more of the three structures occurs: NaCl-type (Cr,Al) ON, face centered cubic (fcc)-(Cr,Al)2O3 − x (with or without dissolved nitrogen) and corundum-type (Cr,Al)2O3 (with or without dissolved nitrogen). The broad peak at about 2θ = 44° can be considered as characteristic for NaCl-type (Cr,Al)ON or (fcc)-(Cr,Al)2O3 − x [42]. The higher Al content in the coatings compared to the XRD spectra shown in the spectra in Fig. 7 leads to the assumption that the formation of fcc (Cr,Al)2O3 coatings is promoted [67]. Moreover, rising intensity of WC peaks is caused by reduced coating thickness due to decreasing deposition rate with increasing aluminum supply (see Table 9). 9 3.4. Chemical composition of the oxy-nitride hard coatings The depth-resolved chemical composition analysis of the oxynitride (Cr1 − xAlx)ON measured by means of EPMA reveals differences in distribution of the elements chromium, aluminum, oxygen and nitrogen in the coatings. Fig. 10 depicts the chemical composition of four different coatings (CrAl20, CrAl20, 5 sccm O2 (a), CrAl20, CrAl20, 20 sccm O2 (b), AlCr20, AlCr20, 20 sccm O2 (c) and AlCr20, AlCr20, 20 sccm O2 (d)). The scales of the measurements were fitted to the SEM cross section fractures. (Cr1 − xAlx)ON-toplayer and (Cr1 − xAlx)N interlayer are marked by dashed lines and can be derived from changing of oxygen and nitrogen contents. The course of the investigated elements along toplayer thickness is mainly smooth. According to variation of oxygen gas flow (5 sccm O2 a) vs. (20 sccm O2 b) and target configuration (AlCr20, AlCr20 c) vs. (AlCr20, Al d), distribution of the elements varies (Tables 7, 9). Fig. 11 shows the results of the mechanical properties for oxy-nitride (Cr1 − xAlx)ON hard coatings deposited at different oxygen flows between 5 sccm O2 and 35 sccm O2 with two CrAl20 targets. Furthermore, universal hardness HU and modulus of indentation EIT for the oxynitride coatings deposited with different target configurations at constant oxygen flow (20 sccm O2) can be obtained from Fig. 11. With rising oxygen flow, hardness and modulus of indentation increase from HU = (9.1 ± 1.5) GPa and EIT = (190.2 ± 27.3) GPa at 5 sccm O2 to their maximum HU = (14.1 ± 2.0) GPa and EIT = (303.0 ± 44.0) GPa at 25 sccm O2. Further increase of oxygen flow results in a decrease of hardness and modulus of indentation, respectively. With regard to the oxy-nitride coatings deposited with different target configuration, comparable higher aluminum content leads to improvement of mechanical properties since increase of hardness and modulus of indentation can be observed in Fig. 11. This hardness enhancement of the coatings can be attributed to the solid solution hardening effect as described in Section 3.2. Concerning high universal hardness HU and low modulus of indentation EIT, the H3/E2IT ratio can be considered as being an important parameter to evaluate the resistance of materials to plastic deformation, e.g. with regard to tribological applications [85]. Coatings with higher H3/E2IT ratio exhibit higher toughness. Derived from hardness and modulus of indentation in Fig. 11, the oxy-nitride coatings with higher aluminum content reveal the highest Fig. 9. HAXRD for phase analysis of oxy-nitride coatings as function of aluminum content. Please cite this article as: N. Bagcivan, et al., Surf. Coat. Technol. (2014), http://dx.doi.org/10.1016/j.surfcoat.2014.09.016 10 N. Bagcivan et al. / Surface & Coatings Technology xxx (2014) xxx–xxx Fig. 10. Depth-resolved chemical composition measured by EPMA. HU3/E2IT ratios, 0.046 for (Cr1 − xAlx)ON deposited with two AlCr20 targets and 0.050 for (Cr1 − xAlx)ON deposited with one AlCr20 and one Al target. 3.5. Characterization of adhesion behavior and tribological performance Fig. 12 shows the dynamic behavior of high-temperature contact angles of optical plastic Plexiglas 7N on oxy-nitride (Cr1 − xAlx)ON coatings measured at the processing temperature of 220 °C for 50 min. Case hardened tool steel AISI 420 served as reference. All oxy-nitride (Cr1 − xAlx)ON coatings exhibit similar dynamic contact angle behavior presented as a slight decrease as function of time. This compares with a progressive decrease of the plastic contact angle on uncoated steel surface (AISI 420). Furthermore, a significant gap between the contact Fig. 11. Universal hardness HU and Modulus of Indentation EIT as function of oxygen flow and target configuration. angles of oxy-nitride coatings with different oxygen and aluminum contents, respectively, and the uncoated steel surface is clearly visible. This can be traced in Fig. 13 presenting photographs of the dynamic behavior of the contact angles of Plexiglas 7N on (Cr1 − xAlx)ON (CrAl20, CrAl20, 10 sccm O2) (a), (Cr1 − xAlx)ON (AlCr20, AlCr20, 20 sccm O2) (b) and AISI 420 (c). This leads to the assumption that the adhesion between oxy-nitride coatings and Plexiglas 7N is much lower compared to the uncoated tool surface. Closer investigations on contact angles of Plexiglas 7N distinguishing between the oxy-nitride (Cr1 − xAlx)ON coatings can be seen in Fig. 14. Application of oxy-nitride (Cr1 − xAlx)ON coatings deposited at 10 sccm O2, 15 sccm O2 and 25 sccm O2 leads to the highest contact angles of Plexiglas 7N. In contrast, any lower or higher oxygen content with regard to the aforementioned coatings results in lower contact angles of the optical plastic and thus, leading to stronger adhesion. Increasing of aluminum content induces even stronger adhesion. In order to correlate the adhesion behavior between the optic plastics and the oxy-nitride (Cr1 − xAlx)ON coatings determined by hightemperature contact angle measurements and the deforming forces in plastic processing, tribological model tests in a pin-on-disk tribometer were carried out. Parameters can be taken from Table 5. Fig. 15 depicts the Coefficient of Friction (CoF) observed when testing Plexiglas 7N pins (Ø = 6 mm) against oxy-nitride coatings with different oxygen and aluminum contents. Again, the uncoated steel serves as reference. From the friction curves it can be concluded that the tribological application of all oxy-nitride coatings presents a significantly lower Coefficient of Friction (maximum CoF ~ 0.70) than the tool steel (CoF ~ 0.82). With regard to the oxy-nitride coatings and the tool steel a correlation can be drawn between the adhesion behavior in hightemperature contact angle measurements (Fig. 14) and the friction behavior in Fig. 15. The oxy-nitride coatings exhibiting the highest contact angles in contact with Plexiglas 7N (10 sccm O2–25 sccm O2 in Fig. 14) Please cite this article as: N. Bagcivan, et al., Surf. Coat. Technol. (2014), http://dx.doi.org/10.1016/j.surfcoat.2014.09.016 N. Bagcivan et al. / Surface & Coatings Technology xxx (2014) xxx–xxx 11 Fig. 12. High-temperature contact angles of optical plastic Plexiglas 7N against oxy-nitride coatings. reveal the lowest CoF (min. CoF ~ 0.6) in tribological testing (Fig. 15). Since the oxy-nitride coatings exhibit smoother surface with higher oxygen gas flow (Tables 8 and 10), any correlation between improved surface quality and favorable adhesion behavior can be excluded. Thus, from these results it can be concluded that lower adhesion (higher contact angles) based on physical and chemical interactions in hightemperature contact angle leads to lower CoF and therefore, might contribute to decreasing of deforming forces in plastic processing. The dynamic behavior of high-temperature contact angles of Plexiglas Resist zk6HF on oxy-nitride coatings can be derived from Fig. 16. Again, AISI 420 was used as reference. Focusing on the oxy-nitride (Cr1 − xAlx)ON coatings, the dynamic behavior of the contact angles almost reveals similar behavior while the time dependency is considerably less pronounced and therefore, is in marked contrast to Plexiglas 7N. Plexiglas Resist zk6HF and oxynitride coatings are basically in thermodynamic equilibrium since contact angle shows almost no temperature and time dependency. This leads to the assumption that no chemical interactions occur at the testing temperature. In turn, the contact angles between Plexiglas Resist zk6HF and the uncoated tool steel surface (AISI 420) rapidly decrease with test duration. The depicted photographs in Fig. 17 verify the different adhesion behavior of Plexiglas zk6HF on the coated surfaces (Cr1 − xAlx)ON (CrAl20, CrAl20, 10 sccm O2) (a), (Cr1 − xAlx)ON (AlCr20, AlCr20, 20 sccm O2) (b) and on the tool steel (AISI 420) (c). The influence of the chemical composition of the oxy-nitride coatings on the adhesion towards the plastic can be understood in Fig. 18. As observed for Plexiglas 7N, contact angles between Plexiglas Resist zk6HF and oxy-nitride coatings vary with their chemical composition. Oxy-nitride coatings deposited at oxygen flow between 5 sccm O2 and 25 O2 lead to lower adhesion against the plastic compared to higher oxygen flow. Increasing aluminum content results in further deterioration of the adhesion towards Plexiglas Resist zk6HF. For correlation analysis between the adhesion behavior in hightemperature contact angle measurements and the deforming forces in plastic processing, friction behavior was determined in tribological model tests. Fig. 19 shows the Coefficient of Friction (CoF) observed when testing Plexiglas Resist zk6H pins (Ø = 6 mm) against oxynitride coatings with different oxygen and aluminum content. For comparison, uncoated tool steel was tested against Plexiglas Resist zk6HF. It can be observed that the tribological application of all oxy-nitride coatings presents a significantly lower Coefficient of Friction (maximum CoF ~ 0.71) than the tool steel (CoF ~ 0.85). With regard to the oxynitride coatings and the tool steel a correlation can be drawn between the adhesion behavior in high-temperature contact angle measurements (Fig. 18) and the friction behavior in Fig. 19. The oxy-nitride coatings exhibiting the highest contact angles in contact with Plexiglas 7N (5 sccm O2–25 sccm O2 in Fig. 18) show the lowest CoF (min. CoF ~ 0.62) in tribological testing (Fig. 15). Since the oxy-nitride Fig. 13. Time dependent development of high-temperature contact angles of Plexiglas 7N on different surfaces. Please cite this article as: N. Bagcivan, et al., Surf. Coat. Technol. (2014), http://dx.doi.org/10.1016/j.surfcoat.2014.09.016 12 N. Bagcivan et al. / Surface & Coatings Technology xxx (2014) xxx–xxx Fig. 14. High-temperature contact angles of optical plastic Plexiglas 7N against oxy-nitride coatings (contact angles on tool surface are hidden). Fig. 15. Tribological tests of oxy-nitride coatings against Plexiglas 7N in PoD-tribometer. coatings exhibit smoother surface with higher oxygen gas flow, higher surface quality does not correlate with favorable adhesion behavior. Thus, from these results it can be concluded that lower adhesion (higher contact angles) based on physical and chemical interactions in hightemperature contact angle leads to lower CoF and therefore, might contribute to decreasing of deforming forces in plastic processing. As adhesion and tribological behavior strongly depend on chemistry at the interface between optical plastic and oxy-nitride hard coatings, metal ratio (Al/(Cr + Al) as well as gas to metal ratios (O/(Cr + Al), N(Cr + Al)) and gas ratio (O/(O + N) were calculated from EPMA data to identify correlations (Fig. 20). Due to highly complex interactions in the interface between the coating and the plastic, no specific parameter, e.g. element ratio can be found to be responsible for adhesion of the plastics on the oxy-nitride coatings. Nevertheless, correlations between the O/(Cr + Al), O/(O + N) and N/(Cr + Al) ratios and the adhesion behavior can be identified since the coatings deposited Fig. 16. High-temperature contact angles of optical plastic Plexiglas Resist zk6HF against oxy-nitride coatings. Please cite this article as: N. Bagcivan, et al., Surf. Coat. Technol. (2014), http://dx.doi.org/10.1016/j.surfcoat.2014.09.016 N. Bagcivan et al. / Surface & Coatings Technology xxx (2014) xxx–xxx 13 Fig. 17. Time dependent development of high-temperature contact angles of Plexiglas Resist zk6HF on different surfaces. Fig. 18. High-temperature contact angles of optical plastic Plexiglas Resist zk6HF against oxy-nitride coatings (contact angles on tool surface are hidden). at high oxygen gas flow above 25 sccm O2 reveal a comparably high O/ (Cr + Al) above 1.3 (corresponding to a low N/(Cr + Al) ratio) leading to a higher CoF and lower contact angle, respectively. Since these coatings have an nearly constant O/(O + N) ratio (higher than for the other coatings) it seems obvious that the reactive gas ratio strongly influences the interactions between the coatings and the plastics. Fig. 19. Tribological tests of oxy-nitride coatings against Plexiglas Resist zk6HF in PoD-tribometer. Please cite this article as: N. Bagcivan, et al., Surf. Coat. Technol. (2014), http://dx.doi.org/10.1016/j.surfcoat.2014.09.016 14 N. Bagcivan et al. / Surface & Coatings Technology xxx (2014) xxx–xxx References Fig. 20. Al/(Cr + Al), O/(Cr + Al), N/(Cr + Al), O/(O + N) ratios in the investigated oxynitride coatings as a function of oxygen flow and target configuration. Furthermore, a very low O/(O + N) combined with a high N/(Cr + Al) ratio for the coating deposited at 5 sccm O2 is also assumed to influence the adhesion towards the investigated optical plastics negatively. With regard to the low N/(Cr + Al) and high O/(O + N) ratios for the coatings deposited with modified target configuration (AlCr20 + AlCr20, AlCr20 + Al), Al can be considered influencing the adhesion towards the investigated optical plastics negatively. As in contrast, the coatings deposited at oxygen gas flow between 10 sccm O2 and 20 sccm O2 contribute to low adhesion (high contact angles, low CoF), it is likely that element ratios must be within a certain scatter range to ensure a favorable adhesion behavior for plastic processing. Therefore, the oxy-nitride (Cr1 − xAlx)ON coatings need to be adjusted regarding their chemical composition and the plastics to be processed. 4. Summary Within the scope of this work, low-temperature oxy-nitride (Cr1 − xAlx)ON hard coatings were deposited using middle frequency (mf) pulsed magnetron sputtering (MS) PVD technology in an industrial coating unit on cold work steel AISI 420 (X42Cr13, 1.2083). Oxygen content varied in the range of 10.8 at.-% and 61.7 at.-% via oxygen gas flow during the deposition process. Aluminum content varied between 10.3 at.-% and 68.0 at.-% by target configuration. EPMA revealed different element distribution within the coatings. High-temperature contact angle measurements using two commercially available plastics (polymethyl methacrylate) Plexiglas 7N and Plexiglas Resist zk6HF revealed a significant influence of chemical composition of the oxy-nitride coatings on the adhesion behavior. Tribological model tests carried out in a pin-on-disk tribometer (oxy-nitride coating against plastic pin) are proving the correlation between low adhesion (high contact angle) determined in high-temperature contact angle measurements and low Coefficient of Friction (CoF) in tribological model tests. Taking into account the EPMA data, parameter areas of metal ratio Al/(Cr + Al), gas to metal ratios O/(Cr + Al) and N(Cr + Al) and gas ratio (O/(O + N) were identified to contribute to favorable adhesion and friction behavior against the investigated plastics. Acknowledgment The authors gratefully acknowledge the financial support of the German Research Foundation, Deutsche Forschungsgemeinschaft (DFG), within the Cluster of Excellence “Integrative Production Technology for High-Wage Countries” and the research area “Multi-Technology Products”. The authors also kindly thank for the financial support of the DFG within the transregional collaborative research center TRR87/ 1 (SFB-TR 87) subproject A1. [1] Plastics Europe, The Facts: An Analysis of European Plastics Production, Demand and Waste Data for 2011, Association of Plastics Manufacturers, 2012. (http://www. plastics-europe.org). [2] Plastics Europe, The Facts: An Analysis of European Latest Plastics Production, Demand and Waste Data for 2013, Association of Plastics Manufacturers, 2014. (http://www.plastics-europe.org). [3] W. Michaeli, S. Hessner, F. Klaiber, CIRP Ann. Manuf. Technol. 56 (1) (2007) 545–548. [4] J. Bekesi, J.J.J. Kaakkunen, W. Micheali, F. Klaiber, M. Schöngart, J. Ihlemann, P. Simon, Appl. Phys. A 99 (2010) 691–695. [5] E. Puukilainen, T. Rasilainen, M. Suvanto, T.A. Pakkanen, Langmuir 23 (13) (2007) 7263–7268. [6] M. Callies, Y. Chen, F. Marty, A. Pépin, D. Quéré, Microelectron. Eng. 78–79 (2008) 100–105. [7] C. Hopmann, A. Neuss, R. Schoeldgen, M. Schoengart, Proceedings of the Polymer Processing Society 28th Annual Meeting, December 11–15, 2012, Pattaya (Thailand), 2012. [8] C. Hopmann, C. Kremer, F. Petzinka, M. Köpf, S. Eilbracht, Proceedings of the Polymer Processing Society 28th Annual Meeting, December 11–15, 2012, Pattaya (Thailand), 2012. [9] W. Michaeli, C. Hopmann, S. Eilbracht, S. Schöngart, M. Scharf, International Symposium of Assembly and Manufacturing, ISAM 2011, Tampere Finland, 2011. [10] C. Hartmann, S. Eilbracht, S. Theiß, M. Schöngart, M. Scharf, W. Michaeli, F. Klaiber, N. Bagcivan, S. Beckemper, K. Bobzin, A. Bührig-Polaczek, A. Gillner, J. Holtkamp, T. Ivanov, C. Brecher, Integrative Produktionstechnik für Hochlohnländer, Springer, Berlin Heidelberg, 2011. [11] O. Durst, J. Ellermeier, C. Berger, Surf. Coat. Technol. 203 (2008) 848–854. [12] E.J. Bienk, N.J. Mikkelsen, Wear 207 (1–2) (1997) 6–9. [13] K. Bobzin, N. Bagcivan, A. Gillner, C. Hartmann, J. Holtkamp, W. Michaeli, F. Klaiber, M. Schöngart, S. Theiß, Prod. Eng. Res. Dev. 5 (4) (2011) 415. [14] M. Heinze, Surf. Coat. Technol. 105 (1998) 38–44. [15] L. Cunha, M. Andritschky, K. Pischob, Z. Wang, A. Zarychta, A.S. Miranda, A.M. Cunha, Surf. Coat. Technol. 153 (2002) 160–165. [16] K. Bobzin, N. Bagcivan, S. Theiß, Thin Solid Films 528 (2013) 180–186. [17] C. Hopmann, M. Schöngart, K. Bobzin, N. Bagcivan, T. Brögelmann, S. Theiß, T. Münstermann, M. Steger, Proceedings of the 4 M Conference, San Sebastian, 2013. [18] K. Bobzin, N. Bagcivan, T. Theiß, W. Michaeli, C. Hopmann, S. Eilbracht, M. Schöngart, M. Scharf, C. Hartmann, J. Holtkamp, A. Gillner, IEEE, (International Symposium on Assembly and Manufacturing (ISAM), 2011. [19] Y.Q. Fu, H.J. Du, S. Zhang, Surf. Coat. Technol. 167 (2003) 129. [20] Y.J. Zhang, P.X. Yan, Z.G. Wu, W.W. Zhang, G.A. Zhang, W.M. Liu, Q.J. Xue, Phys. Status Solidi 202 (2005) 95. [21] Z.G. Wu, G.A. Zhang, M.X. Wang, X.Y. Fan, P.X. Yan, T. Xu, Appl. Surf. Sci. 256 (2006) 2733–2738. [22] K. Bobzin, E. Lugscheider, R. Nickel, P. Immich, Mater. Werkst. 37 (2006) 833–841. [23] C. Friedrich, G. Berg, E. Broszeit, C. Berger, Datensammlung zu Hartstoffeigenschaften, Mater. Werkst. 28 (1997) 59. [24] S.H. Lee, J.J. Lee, Am. Vac. Soc. A 13 (1995) 2030–2034. [25] K. Bobzin, R. Nickel, N. Bagcivan, F.D. Manz, Plasma Process. Polym. 4 (2007) 144–149. [26] Y.J. Lin, A. Agrawal, Y. Fang, Wear 264 (2008) 226–234. [27] J.L. Mo, M.H. Zhu, B. Lei, Y.X. Leng, N. Huang, Wear 263 (2007) 1423–1429. [28] H. Holleck, V. Schier, Surf. Coat. Technol. 76–77 (1995) 328. [29] L. Wang, G. Zhang, R.J.K. Wood, S.C. Wang, Q. Xue, Surf. Coat. Technol. 204 (2010) 3517–3524. [30] J. Vetter, E. Lugscheider, S.S. Guerreiro, Surf. Coat. Technol. 98 (1998) 1233. [31] H. Scheerer, T.H. Hoche, E. Broszeit, B. Schramm, E. Abele, C. Berger, Surf. Coat. Technol. 200 (2005) 203. [32] A.E. Reiter, T.V.H. Derflinger, B. Hanselmann, T. Bachmann, B. Sartory, Surf. Coat. Technol. 200 (2005) 2114. [33] M. Uchida, N. Nihira, A. Mitsuo, K. Toyoda, K. Kubota, T. Aizawa, Surf. Coat. Technol. 177–178 (2004) 627. [34] A. Sugishima, H. Kajioka, Y. Makino, Surf. Coat. Technol. 97 (1997) 590. [35] Y. Makino, K. Nogi, Surf. Coat. Technol. 98 (1998) 1008. [36] Y. Lv, L. Ji, X. Liu, H. Li, H. Zhou, J. Chen, Appl. Surf. Sci. 258 (2012) 3864–3870. [37] J.A. Freeman, P.J. Kelly, G.T. West, J.W. Bradley, I. Iordanova, Surf. Coat. Technol. 204 (2009) 907–910. [38] J. Lin, B. Mishra, J.J. Moore, W.D. Sproul, Surf. Coat. Technol. 202 (2008) 3272–3283. [39] K. Bobzin, N. Bagcivan, P. Immich, S. Bolz, R. Cremer, T. Leyendecker, Thin Solid Films 517 (2008) 1251–1256. [40] B. Podgornik, B. Zajec, N. Bay, J. Vizintin, Wear 270 (2011) 850–856. [41] J. Paulitsch, R. Rachbauer, J. Ramm, P. Polcik, P.H. Mayrhofer, Vacuum 100 (2014) 29–32. [42] A. Khatibi, A. Genvad, E. Göthelid, J. Jensen, P. Eklund, L. Hultman, Acta Mater. 61 (2013) 4811–4822. [43] A. Khatibi, J. Lu, J. Jensen, P. Eklund, L. Hultman, Surf. Coat. Technol. 206 (2012) 3216–3222. [44] D. Diechle, M. Stueber, H. Leiste, S. Ulrich, Surf. Coat. Technol. 206 (2011) 1545–1551. [45] D. Diechle, M. Stueber, H. Leiste, S. Ulrich, V. Schier, Surf. Coat. Technol. 204 (2010) 3258–3264. [46] L. Khizhnyakova, M. Ewering, G. Hirt, K. Bobzin, N. Bagcivan, Trans. Nonferrous Metals Soc. China 20 (3) (2010) s954–s960. [47] K. Bobzin, N. Bagcivan, A. Reinholdt, M. Ewering, Surf. Coat. Technol. 205 (5) (2010) 1444–1448. Please cite this article as: N. Bagcivan, et al., Surf. Coat. Technol. (2014), http://dx.doi.org/10.1016/j.surfcoat.2014.09.016 N. Bagcivan et al. / Surface & Coatings Technology xxx (2014) xxx–xxx [48] O. Knotek, E. Lugscheider, F. Löffler, W. Beele, Surf. Coat. Technol. 68 (69) (1994) 22–26. [49] J. Müller, D. Neuschütz, Vacuum 71 (2003) 247–251. [50] M. Pohler, R. Franz, J. Ramm, P. Polcik, C. Mitterer, Thin Solid Films 550 (2014) 95–104. [51] J. Ramm, M. Ante, T. Bachmann, B. Widrig, H. Brändle, M. Döbeli, Surf. Coat. Technol. 202 (2007) 876–883. [52] V. Edlmayr, M. Pohler, I. Letofsky-Papst, C. Mitterer, Thin Solid Films 534 (2013) 373–379. [53] L. De Abreu Vieira, M. Döbeli, A. Dommann, E. Kalchbrenner, A. Neels, J. Ramm, H. Rudigier, J. Thomas, B. Widrig, Surf. Coat. Technol. 204 (2010) 1722–1728. [54] K. Pedersen, J. Bøttiger, M. Sridharan, M. Sillassen, P. Eklund, Thin Solid Films 518 (2010) 4294–4298. [55] J. Ramm, A. Neels, B. Widrig, M. Döbeli, L. de Abreu Vieira, A. Dommann, H. Rudigier, Surf. Coat. Technol. 205 (2010) 1356–1361. [56] H. Najafi, A. Karimi, P. Dessarzin, M. Morstein, Surf. Coat. Technol. 214 (2013) 46–52. [57] M. Pohler, R. Franz, J. Ramm, P. Polcik, C. Mitterer, Surf. Coat. Technol. 206 (2011) 1454–1460. [58] H. Najafi, A. Shetty, A. Karimi, M. Morstein, Thin Solid Films 519 (2010) 319–324. [59] M. Stueber, U. Albers, H. Leiste, K. Seemann, C. Ziebert, S. Ulrich, Surf. Coat. Technol. 203 (5–7) (2008) 661–665. [60] K. Tönshoff, B. Karpuschewski, A. Mohlfeld, T. Leyendecker, G. Erkens, H.G. Fuß, R. Wenke, Surf. Coat. Technol. 108–109 (1998) 535–542. [61] S.G. Ebbinghaus, H.P. Abicht, R. Dronskowski, T. Müller, A. Reller, A. Weidenkaff, Prog. Solid State Chem. 37 (2009) 173. [62] H. Najafi, A. Karimi, P. Dessarzin, M. Morstein, Thin Solid Films 520 (2011) 1597–1602. [63] H. Najafi, A. Karimi, M. Morstein, Surf. Coat. Technol. 205 (2011) 5199–5204. [64] P.E. Gannon, A. Kayani, C.V. Ramana, M.C. Deibert, R.J. Smith, V.I. Gorokhovsky, Electrochem. Solid-State Lett. 11 (2008) B54. [65] Q.M. Wang, Y.N. Wu, M.H. Guo, P.L. Ke, J. Gong, C. Sun, et al., Surf. Coat. Technol. 197 (2005) 68. [66] J. Sjölén, L. Karlsson, S. Braun, R. Murdey, A. Hörling, L. Hultman, Surf. Coat. Technol. 201 (2009) 6392. 15 [67] A. Khatibi, J. Sjölén, G. Greczynski, J. Jensen, P. Eklund, L. Hultman, Acta Mater. 60 (2012) 6494–6507. [68] M. Stueber, D. Diechle, H. Leiste, S. Ulrich, Thin Solid Films 519 (2011) 4025–4031. [69] H. Najafi, A. Karimi, M. Morstein, Thin Solid Films 519 (2010) 319–324. [70] A. Karimi, M. Morstein, T. Cselle, Surf. Coat. Technol. 204 (2010) 2716–2722. [71] H. Najafi, A. Karimi, D. Alexander, P. Dessarzin, M. Morstein, Thin Solid Films 549 (2013) 224–231. [72] W. Liu, X.P. Su, S. Zhang, H.B. Wang, J.H. Liu, L.Q. Yan, Vacuum 82 (2008) 1280–1284. [73] L. Cunha, M. Andritschky, L. Rebouta, K. Pischow, Surf. Coat. Technol. 116–119 (1999) 1152–1160. [74] L. Cunha, M. Andritschky, K. Pischow, Z. Wang, A. Zarychta, A.S. Miranda, A.M. Cunha, Surf. Coat. Technol. 133–134 (2000) 61–67. [75] R. Kaindl, R. Franz, J. Soldan, A. Reiter, P. Polcik, C. Mitterer, B. Sartory, R. Tessadri, M. O'Sullivan, Thin Solid Films 515 (2006) 2197–2202. [76] X.-Z. Ding, X.T. Zeng, Y.C. Liu, F.Z. Fang, G.C. Lim, Thin Solid Films 519 (2008) 1710–1715. [77] K. Bobzin, N. Bagcivan, M. Ewering, R.H. Brugnara, S. Theiß, Surf. Coat. Technol. 205 (2011) 2887–2892. [78] W.C. Oliver, G.M. Pharr, J. Mater. Res. 7 (1992) 1564–1583. [79] M. Rebelo de Figueiredo, C. Bergmann, C. Ganser, C. Teichert, C. Kukla, C. Mitterer, Int. Polym. Process. 4 (2013) 415–420 XXVIII. [80] A. Kramida, Y. Ralchenko, J. Reader, NIST Atomic Spectra Database, National Institute of Standards and Technology, Gaitherburg, MD, USA, 2014. available http://physics. nist.gov/asd, 19.02. [81] N. Wiberg, Inorganic Chemistry, de Gruyter, Berlin, 2001. [82] C. Kunze, R.H. Brugnara, N. Bagcivan, K. Bobzin, G. Grundmeier, Surf. Interface Anal. 45 (2013) 1884–1892. [83] A. Khatibi, J. Palisaitis, C. Höglund, A. Eriksson, P.O.Å. Persson, J. Jensen, J. Birch, P. Eklund, L. Hultman, Thin Solid Films 519 (2011) 2426–2429. [84] B. Alling, A. Khatibi, S.I. Simak, P. Eklund, L. Hultman, J. Vac. Sci. Technol. A 31 (2013) 03602. [85] J. Musil, Surf. Coat. Technol. 125 (2000) 322. Please cite this article as: N. Bagcivan, et al., Surf. Coat. Technol. (2014), http://dx.doi.org/10.1016/j.surfcoat.2014.09.016