Description

The present invention relates to a method, in particular to a method comprising a PVD coating process for producing a MxA -xN-based thin film (also called coating or coating layer or coating layer system, e.g. a multilayer or a monolayer coating) with a metastable cubic crystal structure exhibiting high thermal stability for the use in high temperature applications, to a MxAh-xN-based thin film as well as to a use of such a thin film in high temperature applications.

Coatings comprising metastable cubic MxA -xN-based structures are well known for their wear resistant applications. These coatings show an optimal combination of hardness, fracture resistance, and oxidation resistance. As a result, even applying only a few micron-thin coating causes significant enhancement in life time of tools, and components in cutting, forming, and other related applications in automotive, and aerospace.

With regard to specific M-AI-N compounds (compounds comprising M, Al and N or consisting of M, Al and N, where M stands for one or more metals and/or semi-metals and/or metalloids, Al for aluminium, and N for nitrogen) with a metastable cubic crystal structure, in particular titanium aluminum nitride (TiAIN) compounds have proven to be suitable for use in high- temperature applications. Furthermore, Ti-AI-N coatings are well-established coatings for wear protection of tools and components. Usually, Ti-AI-N coatings are deposited on substrates by means of physical vapor deposition (PVD) techniques. Because of its very good combination of wear resistance properties and thermal stability Ti-AI-N coating systems have been very good studied. Special attention has been given to the influence of the Al-content in the thermal stability of this kind of coatings.

Besides to Ti-AI-N coatings, other similar coatings based on vanadium aluminium nitride (V-AI- N) or zirconium aluminium nitride (Zr-AI-N) have also been well-studied. All these M-AI-N coatings mentioned above (with M = Ti or V or Zr, respectively) should have any advantage for the mentioned high-temperature applications.

CONFIRMATION COPY However, despite the operational capability at high temperatures, even these known M-AI-N coatings comprise metastable phases showing limited thermal stability, in particular in relation to the ability of the M-AI-N coating material to resist the action of heat and to maintain its hardness at a given temperature. In this regard, the thermal stability is evaluated by measuring hardness of the M-AI-N coating material as a function of annealing temperature, showing the likely maximum application temperature for these coatings. Several investigations on this topic have revealed that the metastable alloys of cubic MxA -xN-based structures might display a hardness enhancement at intermediate annealing temperatures between 800°C and 900°C. However, the key challenge is at annealing temperatures above 1000°C, where the metastable alloy decomposes to their respective ground states resulting in loss of hardness.

The above-mentioned decomposition can be illustrated in the following way:

Metastable cubic M-AI-N - ► cubic MN + wurtzite AIN, wherein M is a transition metal, i.e. Ti, V or Zr.

The resultant wurtzite AIN has a lower elastic modulus of about 300 GPa, and a lower hardness of approximately 25 GPa.

The above-mentioned lower hardness caused by phase transformation limits the application temperature of these coatings to a maximum temperature of 800°C for longer exposure times up to 100 hours, and 1000°C for short exposure times up to 1 hour.

Thus, regarding the limited temperature range, there is still need for improvements. Particularly for increasing protection and improving cutting performance of cutting tools in applications involving exposition of the cutting tools to high temperatures, it remains a challenge to attain a sufficiently good combination of toughness and thermal stability which allows meeting the current demands.

With regard to the increase of thermal resistance, it is known from publication WO 2014/037104 A1 that there is a correlation between the thermal stability of coating structures and the defect density of the coating structures within the crystal structure of the coatings. WO 2014/037104 A1 describes a heterogeneous multilayer coating structure consisting of A- and B-layers, in which the A-layers of TiAIN have a higher defect density than the B-layers of TiAITaN. In WO 2014/037104 A1 it is proposed to estimate the defect density via transmission electron microscopy or X-ray diffraction and to vary it via different deposition techniques in combination with the change of the coil current.

However, estimating the thermal stability of a coating structure as suggested in WO 2014/037104 A1 via an estimate of a general defect density has proven to be inaccurate. Likewise, it has proven difficult to vary the defect density in a targeted manner by changing the coil current.

It is therefore the object of the present invention to at least partially overcome the above-mentioned disadvantages of known coating structures and methods for their manufacture. In particular, it is the object of the invention to provide a process for the manufacture of a coating structure with improved thermal stability, which is adaptable in a constructively simple, controllable and targeted manner to enable coating structures with improved thermal stability.

The above problem is solved by a process according to claim 1 , a thin film according to claim 20 and a use of the thin film according to claim 25. Further features and details of the invention result from the respective subclaims, the description and the drawings. Features and details described in connection with the process according to the invention are of course also valid in connection with the thin film or the use of the thin film according to the invention and vice versa, so that with regard to the disclosure of the individual aspects of the invention, reference is or can always be made mutually.

In this respect, according to a first aspect of the invention a method for producing an MxAl.i-xN- based thin film with a metastable cubic crystal structure exhibiting a high thermal stability allowing its use in high temperature applications is disclosed, wherein M = Ti or V or Zr, and wherein the method comprises the following steps: A preparation step, in which a substrate or at least a substrate surface to be coated is prepared at least by cleaning the substrate surface to be coated, a coating step, in which at least the substrate to be coated is coated by depositing a coating comprising at least one MxAh-xN-based thin film, wherein the coating process is a PVD coating process carried out under targeted control of at least one coating process parameter for adjusting the Frenkel defect density within the crystal structure in the MxAh-xN- based thin film. High-temperature applications may be understood here in particular as:

• permanent or intermittent use (i.e. up to 100 hours) of a tool or component, in such a manner that at least a surface of the tool or component is exposed at temperatures of more than 800 °C, or

• short-term use (i.e. up to 1 hour) of a tool or component, in such a manner that at least a surface of the tool or component is exposed at temperatures of more than 1000 °C.

A PVD coating according to the invention may preferably be able to resist such high- temperature applications.

Preferred areas of application for the inventive coating in question could preferably be protection and performance improvement of cutting tools, forming tools, precision components, other kind of components such as turbines or turbine parts.

Short-term use in the context of the present description may be understood as for a duration of less than one hour. As a concrete example of short-term use, the lifetime of a coated insert in a turning machining operation can be mentioned, because the time an insert is used in such kind of machining operations is usually less than 1 hour.

In the context of the present invention, targeted control may be understood to mean in particular that the at least one process parameter (such as total pressure, nitrogen partial pressure, bias voltage, process temperature or substrate temperature, deposition time, etc) is intentionally changed in such a way that the change in the process parameter causes a targeted change in the Frenkel defect density.

In the context of the present invention, a Frenkel defect (also called Frenkel pair) may be understood to mean in particular a defect in which an atom within a crystal lattice is displaced from its ideal lattice position to an interstitial position, in combination with a vacancy at the original position of the atom. The combination an atom displaced to an interstitial position with a vacancy is special for Frenkel defects, in comparison with, for example, vacancies on the nitrogen sublattice which may form to accommodate substoichiometry in MAIN coatings. Furthermore, in the context of the invention, thermal stability of a coating material exhibiting a metastable cubic phase may be understood to mean the ability of the coating material to resist transformation of the metastable cubic phase into an (undesired) wurtzite phase. Formation of wurtzite phase is undesired, since mechanical properties of the coating deteriorate once the wurtzite phase is formed. For example, the hardness of the coating may decrease. A further undesired effect of wurtzite phase formation is an increased tendency towards cracking or cohesive failure of the coating. As such, the suggested maximal temperature of use, would be in particular the initial temperature for wurtzite phase formation, at which a part of the metastable cubic phase transforms into wurtzite phase (which is undesired for the purposes in the present context), e.g., following spinodal decomposition of fcc-Ti-AI-N.

In the context of the present invention, the inventors have identified the possibility of adjusting the Frenkel defect density within the crystal structure of a coating layer by producing or avoiding the production of Frenkel defects within the crystal structure during formation of a coating layer in a controlled manner and/or in combination with a conduction of a pre-treatment and/or a post-treatment, thereby producing a pre-determined adjustment of the initial temperature for wurtzite formation, at which a part of the metastable cubic phase is transformed into a wurtzite phase. In this manner the present invention provides a method for producing an increment of the thermal stability of the coating layer in a simple and reliable manner.

The inventors determined that using the Frenkel defect density is very advantageous compared to a determination of general defect densities, because surprisingly the determination of a Frenkel defect density enables a much more accurate and reliable prediction of the temperature stability of a coating structure, which in turn enables a much more accurate and targeted control of the temperature stability of coating layers.

In the context of the invention, the metastable cubic crystal structure may be understood in particular as a cubic crystal structure which is stable up to a limit temperature and only begins to transform at least partially from a cubic phase into a wurtzite phase at a determined temperature. In this regard, it is known that limit temperatures for MxA -xN-based thin films are usually in a temperature range between 800 °C and 1000 °C The present invention allows attaining higher limit temperatures in this regard, for example up to 1300°C, which results in higher thermal stability in comparison to the state of the art, in a controlled, reliable, and simple manner.

A preferred coating process parameter that can be used as the at least one coating process parameter for attaining a simple and precise adjustment of thermal stability of a MxAli-xN-based thin film according to the present invention, is the total pressure or the nitrogen gas partial pressure used during the coating process. This process parameter is particularly advantageous for adjusting the Frenkel defect density within the crystal structure in the MxA -xN-based thin film. When the nitrogen deposition pressure (also called nitrogen gas partial pressure) is used ♦ as the at least one adjusting coating process parameter, this parameter is preferably adjusted having a value in a range of less than 2 Pa and preferably not lower than 0.05 Pa, in particular in a range from 0.05 to 1.0 Pa, further preferred in a range from 0.05 to 0.5 Pa. In one embodiment of the innovation, a nitrogen deposition pressure of less than 2 Pa is preferably combined with a plasma having energetic ions, specifically a plasma with average ion energies >25 eV. In one further embodiment, a nitrogen deposition pressure of less than 2 Pa and preferably not lower than 0.05 Pa can be used, preferably in combination with a plasma having multiple ion charge states, specifically a plasma with average ion charge states >1.

For attaining a simple and precise control of thermal stability of a MxA -xN-based thin film according to the present invention, preferably at least one further coating process parameter may be targeted controlled in combination with the deposition pressure, wherein the further parameter is at least one of the following: bias-voltage, coating process time for adjusting coating thickness, coating process temperature (in this context, the term coating process temperature is used for referring to the temperature of the surface on which the coating is being formed, normally the substrate surface being coated - the inventors recommends to use low substrate temperatures for attaining lower mobility in the crystal lattice). However, also other coating process parameters can be used with this purpose.

A further manner to influence the generation of Frenkel defects is for example by forming the coating comprising a combination of different coating layers, or also by applying a pretreatment (before coating process) and/or a post-treatment (after coating process) with this purpose. With regard to a higher thermal stability of the coating structure according to the present invention, it can advantageously be provided in particular that the coating process parameters may be adjusted to achieve an increase in the Frenkel defect density within the crystal structure in the MxA -xN-based thin film which has the effect of reducing the mobility of Al in its crystal structure. As has been found out by the inventors, the mobility of Al is required for decomposition of the metastable cubic MxA -xN, while N mobility is not required, wherefore particularly the increase in activation energy for Al mobility retards Al segregations and therefore result in an increased thermal stability. Therefore, the inventors found Frenkel defects particularly useful for increasing the thermal stability. Although an increase in the Frenkel defect density shows an advantageous increase in thermal stability of the coating, it must be considered that the Frenkel defect density cannot be increased without limit. In case of attaining a Frenkel defect density that results of a change in the original (in case of no Frenkel defects) crystal structure, then this Frenkel defect density would not result in the desired technical effect. For example, in a theoretical example of 50% Frenkel defects, then the crystal structure is no longer cubic.

For determining a Frenkel defect density, the quantification of the Frenkel defect density may preferably be estimated via the stress-free lattice parameter due to lower sample preparation needs compared to e.g. transmission electron microscopy methods, wherein the estimation or determination of the stress-free lattice parameter may in particular be performed via X-ray diffraction methods. Further diffraction methods, in particular, electron diffraction or neutron diffraction may also be useful for determination of the stress-free lattice parameter. The determination of the Frenkel defect density hereby may be performed during (for example by means of in-situ diffraction methods) and/or after producing the MxA -xN-based thin film.

The determination of the Frenkel defect density, through X-ray diffraction methods is conducted involving the determination of the stress-free lattice parameter taking into account the coating stoichiometry and the M/AI ratio. The stress-free lattice parameter is influenced by the coating stoichiometry and the M/AI ratio. In other words, two coatings with same density of Frenkel defects, but different stoichiometry, will have different stress-free lattice parameters. In the same manner, two coatings with same density of Frenkel defects, but different M/AI ratio, will have different stress-free lattice parameters. A further preferred determination of the Frenkel defect density, comprises the usage of theoretical calculations, wherein preferably density functional theory (DFT)-calculations may be used, wherein in particular the DFT-calculations may be used to determine the baseline for the MxA -xN-based thin film without Frenkel-defects. A particular advantage of using this method is that coatings of a wide range of M/AI ratios and stoichiometries can be calculated, which may be difficult and/or time consuming to synthesize experimentally without Frenkel defects.

A method for determining the Frenkel defect density, by means of the stress-free lattice parameter, therefore preferably may comprise at least one, in particular all of the following steps: i) Determination of the stress-free lattice parameter of the coating abating, preferably through X-ray diffraction methods. ii) Determination of the stoichiometry and M/AI ratio of the coating. iii) Determination of the stress-free lattice parameter without Frenkel-defects abaseiine (Baseline-value) for the stoichiometry and M/AI ratio determined in step ii). The determination preferably comprises the usage of theoretical calculations, preferably DFT-calculations. iv) Comparing the difference between abating, the stress-free lattice parameter of the coating determined in step i) to abaseiine the stress-free lattice parameter without Frenkel-defects determined in step iii). The magnitude of the difference abating - abaseiine may be a measure of the Frenkel defect density.

In a preferred embodiment with respect to determining the Frenkel defect density, the ratio a _CO ating / abaseiine may be larger than a factor of 1.002 but smaller than a factor 1.03, more preferrable larger than a factor of 1 .003 but smaller than a factor 1.02, in particular larger than a factor of 1.003 but smaller than a factor 1.01 . Coating layers within the aforementioned range of a _coa ting / abaseiine ratio was found to accurately and reliably predict a Frenkel defect density leading to enhanced thermal stability of the coating layers.

Moreover, in a further preferred embodiment with respect to determining the Frenkel defect density, the difference abating - abaseiine may be in the range between 0.01 A and 0.5 A, more preferably in the range between 0.015 A and 0.3A, in particular in the range between 0.015 A and 0.1 A. (A represents the length unit Angstrom, where 1A = 10‘ ¹⁰ m.) Coating layers within the aforementioned range of acting - abaseiine difference was found to accurately and reliably predict a Frenkel defect density leading to enhanced thermal stability of the coating layers.

For providing a high flexibility for controlling the variable process parameters, the coating process comprises the use of physical vapor deposition (PVD) techniques or is preferably a physical vapor deposition (PVD) process.

The coating process (e.g. a PVD process or a process combining PVD techniques and other kind of coating techniques) preferably involves the use of magnetron sputtering techniques and/or, cathodic arc evaporation techniques.

If involving magnetron sputtering techniques, then preferably high-power impulse magnetron sputtering (HiPIMS) techniques, because using HiPIMS techniques allow attaining high plasma ionisation and may particularly be designed to favour the generation of multiply charged species.

More preferably the PVD coating process may be a cathodic arc evaporation process because of the capability to produce plasma with a high degree of ionization, high ion energies, and multivalent ions (for example doubly charged ions M ²⁺ (Metal)).

A plasma comprising ions with high energies and/or a plasma with multivalent ions have been found to be particularly effective in the context of the invention for controlled generation of Frenkel defects. The effectiveness of such plasma was found by the inventors to be caused by the capability for ion irradiation, where highly energetic and/or highly charged ions impinge into the growing film and contribute to the creation of Frenkel defects. The inventors further explored plasma with different characteristics and investigated the effectiveness for the purpose of controlled generation of Frenkel defects. In particular, the parameters average ion energy, most probable energy, average charge state, and average kinetic energy were found to be parameters that, alone or in combination of two or more parameters, predict well the capability for controlled generation of Frenkel defects. In the context of the present invention, it has proved to be particularly advantageous, for the controlled generation of Frenkel defects, when the PVD coating process may be designed selecting the coating process parameters in such a way that the average ion energy of the ions in the coating plasma is greater than 25 eV and smaller than 200 eV, in particular between 25 eV and 100 eV.

It has proved to be particularly advantageous, for the controlled generation of Frenkel defects, if the PVD coating process may be designed in such a way that the plasma used to deposit MxA -xN-based thin film comprise metal ions (for example Ti ⁺ and/or Ti ²⁺ and/or Al ⁺ in the case of TixA -xN) with a most probable energy, corresponding to the maxima of the ion energy distribution (IED), larger than 15 eV and smaller than 150 eV. The most probable energy (maxima of the ion energy distribution (IED) as explained above), can preferably be determined through measurement of the ion energy distribution (IED) for each ion type, where the relative number of ions are measured with respect to energy per charge.

Moreover, it has proved to be particularly advantageous in the context of the invention, if the plasma enhanced PVD coating process may be designed in such a way that the average charge state of the ionic species contained in the plasma is between 1.2 and 1.6, in particular between 1 .3 and 1 .4.

Furthermore, according to a particularly advantageous embodiment of the present invention, the PVD coating process may be conducted by using a coating plasma having an average kinetic energy of the ions arriving at the substrate or arriving at the surface of the growing film that is greater than 85 eV, preferably greater than 90 eV, however preferably smaller than 1000 eV. The average kinetic energy of the ions arriving at the substrate or the growing film (Ek) may be calculated as E _k = Ej° + qe * (V _pi + V _s ), where Ej° denote the average ion energy in eV, q the average ion charge state, e the elementary charge (negative), V _pi the plasma potential, and V _s the magnitude of the substrate bias potential. In other words, the kinetic energy of the ions arriving at the substrate or at the surface of the growing film, may be determined taking into account the kinetic energy gained from acceleration across the sheath (also called substrate sheath or plasma sheath) formed in the vicinity of the substrate due to a difference between the substrate bias potential and the plasma potential. For a plasma comprising ions of known average ion charge state and average ion energy, the kinetic energy of the ions arriving at the substrate or the growing film may preferably be controlled by adjusting the substrate bias potential.

In the context of a constructively simple and fast production of a thin film according to the invention, different target materials and/or different targets may be used, wherein the different target materials preferably may be vaporized simultaneously.

In the context of producing a clean and high-quality coating, the substrate temperature used may preferably be in a range from 300 °C up to 600 °C, more preferably between 400 °C and 500 °C.

In the context of a constructively simple and fast production of a thin film according to the invention, a bias voltage of less than - 150 V, preferably less than - 100 V, in particular less than - 50 V may be used. It is understood that a negative bias voltage is used, so the terms "smaller" or "larger" here refer to the magnitude of the bias voltage (i.e. a bias voltage of -50 V is smaller than a bias voltage of -100 V).

According to a second aspect of the invention a MxA -xN-based thin film with a metastable cubic crystal structure for the use in high temperature applications is disclosed, wherein MxA -xN-based thin film is preferably produced via a method described above, wherein M = Ti, V or Zr, wherein the MxAh-xN-based thin film showing an increased Frenkel defect density determinable via the determination of a difference of the stress-free lattice parameter of the MxA -xN-based thin film (abating) in comparison to the stress-free lattice parameter of a determined baseline-value of a MxA -xN-based thin film of same stoichiometry and same M/AI ratio (abaseiine), wherein the stress-free lattice parameter of the MxA -xN-based thin film (abating) differs from the stress-free lattice parameter of the determined baseline-value of a MxA -xN- based thin film of same stoichiometry and same M/AI ratio (abaseiine) by 0.2 % to 3 %. Thus, the MxAh-xN-based thin film according to the invention shows the same advantages as have already been described in detail with respect to the process according to a first aspect of the invention. The MxAh-xN-based thin film according to the second aspect of the invention can in particular have a crystal structure that is stable up to at least 1050 °C. It is understood with respect to the determination of an increased Frenkel defect density of the MxAh-xN-based thin film that the stress-free lattice parameters of the MxAli. _x N-based thin film with Frenkel defects and without Frenkel-defects (baseline-value) may be determined by a method as described in detail on page 8 of the description.

Preferably, the stress-free lattice parameter of the MxA -xN-based thin film (abating) may differ from the stress-free lattice parameter of the determined baseline-value of a MxAh-xN-based thin film of same stoichiometry and same M/AI ratio (abaseiine) by 0.3 % to 2 %, particularly by 0.3 % to 1 %.

With regard to the exact composition of the MxA -xN-based thin film according to the invention, it has proven to be advantageous if x may be in a range 0.60 s x ^0.25, preferably x may be > 0.25, preferably x > 0.45, in particular x = 0.5.

In the context of the production of thin coatings, it may further be intended that the layer thickness of the MxAh-xN-based thin film may preferably be in the range from 0.5 pm - 20 pm, preferably 1 pm - 10 pm, more preferably 2 pm - 5 pm for attaining the desired thermal stability effect.

According to a third aspect of the invention a use of a M _x Ali- _X N-based thin film for coating cutting tools, forming tools, precision components or turbine applications with the intention of improving their performance is disclosed. Thus, the use of the inventive thin films according to the present invention results in the advantages as have already been described in detail above, when the process according to the present invention was described or when the inventive MxA -xN-based thin film has been described.

Further advantages, features and details of the invention will be apparent from the following description, in which embodiments of the invention are described in detail with reference to the drawings. The features mentioned in the independent claims and in the corresponding parts of the description may be essential to the invention. Further features not included in the independent claims but mentioned in the dependent claims and in the corresponding parts of the description constitutes preferred embodiments of the invention individually or in any combination. The examples of the invention provided in the present description should not be understood as a limitation of the invention but as preferred embodiments or showcases of the present invention. The Figures 1 to 4 are used for facilitating understanding of the description of the invention.

The Figures show:

Fig. 1 a schematic representation of the individual steps of a PVD coating process for producing a MxA -xN-based thin film according to the invention,

Fig. 2 a graphic representation of DSC-measurements of Ti-AI-N probes according to the invention showing the heat flow data of the coatings stem from the difference between two heating cycles,

Fig. 3 a graphic representation of the stress-free lattice parameters as a function of deposition pressure, marked in the figure is also the stress-free lattice parameter for (Tio.56Alo.44)o.48No.52 (corresponding to (Tio.56Alo.44)iNi.o8) calculated by DFT methods, without Frenkel defects,

Fig.4 a graphic representation of lattice parameters of (Tio.5Alo.5)Nx and (Tio.sAlo.sJNx with one Al Frenkel pair as a function of nitrogen content at 0 K (zero degrees Kelvin).

Fig. 1 shows a schematic representation of the individual steps of a PVD coating process for producing a MxAli-xN-based thin film according to the invention.

Within the scope of the method according to the present invention, the method comprises the following steps:

• A preparation step 100, in which a substrate or at least a substrate surface to be coated (the substrate for example be made of Al or Fe, or steel, or cemented carbide or any other material that can be coated) is prepared, for example by cleaning the substrate or at least the substrate surface to be coated, before coating.

• An optional pre-treatment step 200, consisting in one or more pre-treatments, in which the substrate or the substrate surface to be coated is treated before coating, for attaining specific conditions at the substrate surface to be coated.

• A coating step 300, in which at least the substrate to be coated is coated by depositing a coating comprising at least one MxAh-xN-based thin film.

• An optional post-treatment step 400, consisting in one or more post-treatments, in which at least the coating provided in the coating step 300 is treated for attaining specific coating properties or specific conditions at the coated substrate surface.

Fig. 2 shows a graphic representation of DSC-measurements of a MxAh-xN-based thin film in the form of (Tio.sAlo.sJN probes according to the invention showing the heat flow data of the films stem from the difference between two heating cycles in order to obtain only the probes calorimetric response.

In some examples of the present invention Ti-AI-N thin films were synthesized with increased amount of Frenkel defects for adjusting a pre-determined thermal stability. For attaining the desired increment of Frenkel defects in some examples the deposition pressure was controlled during the deposition by maintaining it constant at a value within a range of low deposition pressure from 0.5 Pa to 1 Pa, or by varying it within a range of low deposition pressure values from 0.5 Pa to 1 Pa. In this manner Ti-AI-N coatings with increased amount of Frenkel defects were obtained. In the DSC-measurements, it was found that in this manner the onset temperature for wurtzite phase formation was increased by 56° C in comparison to the state of the art Tio.50Alo.50N coatings.

As shown in Fig. 2 powder from Tio.5Alo.5N coatings deposited Fe substrate foils exhibit three distinct enthalpic reactions during calorimetric analysis: recovery processes and defect annihilation (marked with (1 ) in Figure 2) were observed at 670 °C, the onset of spinodal decomposition forming Al-rich domains is observed at 860 °C (marked with (2) in Figure 2). The wurtzite solid solution formation is marked with (3) in Figure 2) - it is noted that the onset of this peak is shifted by 56 °C to higher temperature for the coating deposited at 0.5 Pa N2 compared to the coating deposited at 3.5 Pa N ₂ coating. Fig. 3 shows: the stress-free lattice parameters as a function of deposition pressure.

As can be seen in Fig. 3 an increase by 0.9% has been noted when the deposition pressure was adjusted to 1.0 Pa N2 or lower. As can also be seen in Fig. 3, an increase of the stress- free lattice parameter of 0.35% was noted compared to the value calculated by DFT methods for (Tio.56Alo.44)o.48N.52 without Frenkel defects. The stress-free lattice parameter can preferably be determined by X-Ray Diffraction methods.

Thus, it has been found here that the presence of Frenkel defects can be identified by studying the stress-free lattice parameters. It was found (as shown in Fig. 3(b)) that coatings deposited at low deposition pressure (for example lower than 1 Pa), with increased amount of Frenkel defects, are characterized by a higher value of the stress-free lattice parameter (higher than 4.20 A, valid for the stoichiometry and M/AI-ratio according to the formula (Tio.56Alo.44)iNi.o8 ).

When evaluating the lattice parameters of M-AI-N coatings, care must be taken to correct the parameters with respect to the coating stoichiometry and M/AI-ratio.

Fig. 4 shows a graphic representation of lattice parameters of (Tio.sAlo.sJNx and (Tio.sAlo.sJNx with one Frenkel pair as a function of nitrogen content at 0 K (zero degrees Kelvin). (Tio.sAlo.sjNx is here used only as an example of one MxA -xN thin film. However similar graphic representations of lattice parameters can be obtained for MxA -xN thin films with M different from Ti, for example M= V or Zr, and also for chemical element concentrations corresponding to values of x different from 0.5.

The stoichiometry, i.e. the ratio of metal (Ti+AI) to non-metal (N) atoms will influence the lattice parameter. This is exemplified in the curve with square symbols in Figure 4 considering the case of (Tio.sAlo.sJNx. The lattice parameters are the highest, with a value of approximately 4.172 A, at nitrogen content of 50 at%, i.e. a metal: non-metal ratio of 1 :1 , or in other words exactly stoichiometric. The curve with square symbols in Figure 4 also shows that if the N-content is lower than 50 at%, for example, in the range between 46 to 50 at%, i.e. a metal: non-metal ratio in the range between 1 :0.85 and 1 :1 , the lattice parameter is lower than the value for 50 at%. As an example, in the case of a N-content of 47 at%, i.e. metal:non-metal ratio of 1 :0.887, the stress free lattice parameter can be read out from the curve with square symbols in Figure 4 to about 4.169 A. The curve with square symbols in Figure 4 furthermore shows that if the N-content is higher than 50 at%, for example, in the range between 50 to 53.5 at%, i.e. a metal:non-metal ratio in the range between 1 :1 and 1 :1.15, the lattice parameter is lower than the value for 50 at%. As an example, in the case of a N- content of 53 at%, i.e. metal: non-metal ratio of 1 :1.128 the stress free lattice parameter can be read out from the curve with square symbols in Figure 4 to about 4.163 A. It is therefore essential to accurately measure the N-content and correct the baseline for the lattice parameter accordingly, before assessing the change due to Frenkel defects. Stress-free lattice parameters for coatings of different stoichiometry can be difficult to obtain experimentally, therefore theoretical methods, preferably density functional theory (DFT) calculations can be used to obtain the baseline curve for MxAh-xN thin films without Frenkel defects.

The influence of Frenkel pairs on the lattice parameter is as well shown in Figure 4. The curve with downward pointing triangular symbols shows the lattice parameter of (Tio.sAlo.sJNx with 1 Aluminium (Al) Frenkel pair per 128 atoms, over the range of nitrogen content from 46 to 53.5 at%, i.e. a metal:non-metal ratio in the range between 1 :0.85 and 1 :1.15. As an example, from the curve with downward pointing triangular symbols, the lattice parameter at 50 at%, i.e. metal:non-metal ratio of 1 : 1 , with 1 Al Frenkel defect per 128 atoms, is 4.196 A. The magnitude of the difference between the lattice parameter determined with Frenkel defects (4.196 A) and the lattice parameter without Frenkel defects (4.172 A) - in this case 0.024 A - is a measure of the Frenkel defect density. In this example of (Tio.sAlo.sJNi, i.e. Ti/AI ratio of 1 and metaknon- metal ratio of 1 :1 , a lattice parameter difference larger than 0.024 A would indicate a Frenkel defect density higher than 1 per 128 atoms, and a lattice parameter difference smaller than 0.024 A would indicate a Frenkel defect density smaller than 1 per 128 atoms.

Finally, two specific TixA -xN coatings were applied to substrates at different N2-coating pressures and measured with respect to the stress-free lattice parameters and thermal resistance. The results are shown in Table 1 below.

Tab. 1 Comparison of two TixA -xN coatings applied to substrates at different N2-coating pressures and measured with respect to the stress-free lattice parameters and thermal resistance.

TiAIN coatings were deposited by means of cathodic arc deposition with TiAl 50:50 (at.%) targets. The substrates were mounted inside the deposition chamber which was evacuated to a base pressure on the order of 3x1 O' ⁴ Pa, and subsequently heated to 450 °C. In particular, a-AhOs substrates and Fe-foils were coated, however, the method can be extended to other types of substrates like cemented carbides, tool steels or the like. The distance between the arc targets and the substrates was kept at 160 mm. An in-situ etching was performed prior to the deposition step. The substrates were held at a DC substrate bias potential of -40V.

In a first comparative process, the iSfe-pressure during deposition was kept at 3.5 Pa. Coated substrates were analyzed by X-ray diffraction. Stress-free lattice parameters were obtained according to the method of Genzel for textured cubic crystals (see C. Genzel, “X-Ray Stress Gradient Analysis in Thin Layers — Problems and Attempts at Their Solution”, physica status solidi (a) 159(2) (1997) 283-296.) using a General Area Diffraction Detection System. (111), (200) and (220) lattice planes were measured in Bragg-Brentano geometry. The elastic compliances Sy were taken from the inverse of the elastic tensor from the publication “Ab initio elastic tensor of cubic Tio.5Alo.5N alloys: Dependence of elastic constants on size and shape of the supercell model and their convergence”, by F. Tasnadi, M. Oden, LA. Abrikosov, Phys Rev B 85(14) (2012) 144112. The stress-free lattice parameter was found to be 4.170±0.02A. The onset temperature for wurtzite formation, was characterized with DSC measurements of powder created from the coating and was found to be about 1020 °C. In a second inventive process, the N2 pressure during deposition was kept at 0.5 Pa. Coated substrates were analyzed by X-ray diffraction according to the method of GenzeL The stress-free lattice parameter was found to be 4.206A. The onset temperature for wurtzite formation, was characterized with DSC measurements of powder created from the coating and was found to be about 56 °C higher than the comparative process.

In order to characterize the plasma conditions during deposition, a plasma process monitor (PPM) mass-energy analyzer was mounted on the deposition system in place of the samples. The time-averaged ion energy distribution functions (IEDF)s were obtained through energy scans at fixed mass-to-charge ratios of the dominant positive ions. The entrance orifice of the analyzer was grounded and all ion energy distributions (IED)s were measured with respect to ground. For each ion species, the average ion energy was calculated as a mean value of the ion energy distribution (IED). The average ion energy of the comparative process was determined to 16.6 eV and of the inventive process to 26.2 eV. The average charge state was determined to 1.31 in the comparative process and to 1.37 in the inventive process.

The aforementioned DSC measurements were performed of powders created from the deposited coating in that the Fe foil substrates were etched in a nitric acid with a HNOs/deionized H2O volume ratio of 1/5 and subsequently dried. The coating flakes were milled into a powder using a mortar. The thermal response of c-(Ti,AI)N powders was investigated in a differential scanning calorimeter (DSC) in continuous heating mode and Ar atmosphere with a purity of 99.9999%. The powders were initially outgassed for 30 min at 150 °C, after which the measurement started by heating up the sample to 1500 °C in 30 seem flow of Ar. The heating and cooling rates were kept constant at 20 and 40 °C/min, respectively.

List of Reference Signs

100 Preparation of a substrate to be coated

200 Optional pre-treatment of the substrate surface 300 Coating of the substrate surface, preferably by means of a PVD coating process according to the invention

400 Optional post-treatment of the coated substrate