DE102013011073A1 - TlxSi1-xN layers and their preparation - Google Patents

Abstract

The invention relates to a workpiece with coating which coating comprises at least one Ti _x Si _1-x N layer, characterized in that x ≤ 0.85 and the Ti _x Si _1-x N layer nanocrystals and the nanocrystals contained an average grain size of not more than 15 nm and has a (200) texture. The invention also relates to a method for the production of the aforementioned layer, characterized in that a sputtering method is used for the production in which the target surface of the sputtering target has current densities of greater than 0.2 A / cm 2 and the target is a TixSi1-x target, where x ≤ 0.85.

Description

The present invention relates to a coating comprising at least one layer of silicon.

Silicon is a chemical element which is sometimes used in conjunction with hard coatings to increase the layer stresses. If the layer tension increases, this generally leads to a greater hardness of the layer. This is also used, for example, in conjunction with titanium nitride. This results in layers which can be chemically described by the structural formula Ti _x Si _1-x N, where x is the concentration of Ti expressed in at% when only the metallic elements are taken into consideration. In this notation, the atomic concentrations given in percent add up to 100%.

Such layers can be produced in very hard form by means of the so-called cathodic spark evaporation. In this case, a spark is ignited between a target supplying the metallic elements target, which is used as a cathode and an anode over which from the target surface, an electron current of high density is pulled out. Due to the strongly localized very high current density at the target surface, the target surface is locally heated strongly and the material evaporates in ionized form.

The thus vaporized and ionized material is then accelerated to the substrates by means of a negative voltage applied to the substrates. If, in addition, a reactive gas is introduced into the coating chamber, the vaporized ions combine with the reactive gas and form a corresponding layer on the substrate surface.

In this process, however, it often comes to the so-called droplet problem: Due to the sudden local heating on the target surface, it comes to an explosive melting of these are thrown through the whole drop of the target material into the environment. These drops partly land on the surface of the substrate, which generally has negative effects on the coating properties and their quality. Although there are now methods to filter out these droplets. However, such filters make the coating rate very small and it is hardly possible to operate the coating economically.

On the other hand, a silicon content of greater than 15 at% very often leads to damage of the target during the spark evaporation. In extreme cases, then the target must be replaced after each coating, which in turn is detrimental to the efficiency of the process.

These problems are not exposed to the expert in the conventional deposition from the gas phase by means of magnetron sputtering (magnetron sputtering). However, the particles ejected from the target surface by ion bombardment are not or hardly ionized and therefore can not be accelerated to the substrates with a substrate bias applied to the substrates. Accordingly, such conventionally sputtered layers have a relatively low density and hardness.

One known way to push the density and hardness of sputtered layers in the spark-evaporation similar areas is the so-called HiPIMS method. HiPIMS = High Power Impulse Magnetron Sputtering). In this sputtering method, a sputtering cathode is subjected to high power pulse densities, resulting in that the cathode sputtered material is ionized to a high percentage. If a negative voltage is now applied to the workpieces to be coated, these ions are accelerated in the direction of workpieces, which leads to very dense layers.

The sputter cathode must be pulsed with power to give it time to dissipate the heat input associated with the power. The HiPIMS method therefore requires a pulse generator as the power source. This pulse generator must be able to deliver very high power pulses, but they are very short. The pulse generators available today show little flexibility in terms of, for example, pulse height and / or pulse duration. Ideally, a square pulse should be delivered. In most cases, however, the power output within a pulse is highly time-dependent, which has an immediate effect on the layer properties, such as hardness, adhesion, residual stress, etc. In addition, the coating rate is adversely affected by the deviation from the rectangular profile.

In particular, these difficulties raise issues of reproducibility.

Accordingly, as far as the inventors are aware, it has not yet been attempted to produce Ti _x Si _1-x N layers by means of the HiPIMS method.

Therefore, there is a need for a method according to which Ti _{2 ×} Si _1-x N layers can be produced by means of magnetron sputtering at high power.

According to the invention, the layers are produced by means of a sputtering method in which the power output of the power source is constantly high. Several sputtering cathodes are used. Unlike the conventional HiPIMS method, no pulse generator is used, but at first only a first sputtering cathode is charged with the full power of the power source and thus with high power density. Subsequently, a second sputtering cathode is connected to the outputs of the power source. At first, there is little happening because the impedance of the second sputtering cathode at this time is much higher than the impedance of the first sputtering cathode. Only when the first sputter cathode is disconnected from the outputs of the power source does the power output essentially take place via the second sputtering cathode. The corresponding high performance magnetron sputtering method is more specific in WO 2013060415 described. Typically, the power source is driven in the order of 60 kW. Typical powers to which the sputtering cathodes are exposed on average over the average time are 8 kW.

The inventors have now found, completely surprisingly, that if such a method is operated with TiSi targets which contain a silicon content of greater than or equal to 15 at%, it is possible to reproducibly produce nanocrystalline layers having very good mechanical properties. It is particularly interesting that, from an Si concentration in the target of 15 at%, the nanocrystals have on average a particle size of less than 15 nm, as in 1 shown. The concentration ratios in the target used for the coating are reflected almost directly in the coated layer 2 shown. It should be noted at this point that, as soon as a target with a certain Si concentration has been selected, the grain size can be finely adjusted by the nitrogen consumption, such as 7 shows.

That this is a very robust phenomenon is on 3 seen. Here, the grain sizes of layers were measured, which were coated at different positions on the rotating carousel. The measurement series with the black filled circular symbols refers to a Ti ₉₅ Si ₅ target. The measurement series with the white filled circular symbols refers to a Ti ₉₀ Si ₁₀ target. The measurement series with the black filled square symbols refers to a Ti ₈₅ Si ₁₅ target. The measurement series with the white filled triangular symbols refers to a Ti ₈₀ Si ₂₀ target. The measurement series with the black filled triangular symbols refers to a Ti ₇₅ Si ₁₅ target. Obviously, the grain size is maintained over the entire coating height of the chamber.

The layers then exhibit increasing hardness and decreasing modulus of elasticity as the silicon content increases 4 shown. There are not the concentration ratios in the layers but the concentration ratios Ti / Si indicated in the targets used for the preparation of the layers.

According to a further embodiment of the present invention, the Ti _x Si _1-x N layer with at least 15 at% Si content of the metallic components is not applied directly to the substrate to be coated but between the substrate and the inventive layer a TiAlN layer as Intermediate layer provided. One of the advantages of this intermediate layer is that it mediates between the less brittle substrate and the extremely hard Ti _x Si _1-x N layer, which is exposed to very high residual stresses, with regard to the stress and / or pressure conditions. This results in much less chipping and the layer adhesion is improved accordingly. 5 shows a series of such two _- layer according to the invention, wherein for the coating of the Ti _x Si _1-x layers turn the already discussed different targets were used according to the information in the figure. Clearly visible in the series are the different structures of the Ti _x Si _1-x N layers, which are getting finer and finer with increasing Si content. In the present example, a target comprising 40 at% of titanium and 60 at% of aluminum was used to produce the intermediate layer. It has been found to be particularly advantageous if both layers TiAlN and TiSiN have a (200) texture.

Such bilayers with different Si content of tools were tested. For the cutting tests were carried out under the following conditions: workpiece steel DIN 1.2344 hardened to 45 HRC, tool diameter 10 mm solid carbide bur, cutting speed 220 m / min, feed per tooth 0.1 mm, axial feed 10 mm, radial feed 0.5 mm. It was measured by how many Meters a corresponding tool can work without taking damage. Tool which is coated with a commercial coating, survives for just over 200 m. Tool of about the same length survives, coated with the double layer described above, with the outer layer containing only 5% of silicon. In contrast, the tests showed that the tool survived more than 500 m when the outer layer contains at least 15% silicon. Table 1 lists the wear values measured with the tools after a cutting path of 140 m. It is clearly evident that wear in the coating with 30% silicon is the lowest.

According to a further advantageous embodiment, a transition layer is provided between the TiAlN intermediate layer and the Ti _x Si _1-x layer, which was produced by means of cosputters. With the sputtering method described above, cosputters can be reliably performed such that, for example, the pulse lengths for the different targets are selected such that the maximums of the reactive gas consumption curves substantially overlap one another as a function of the pressure prevailing in the coating chamber. This is possible because the pulse duration has a direct influence on the position of the corresponding maxima. This is for an example in 6 where sputtering was done with 3 different pulse durations (0.05 ms, 0.2 ms and 2 ms) in this way it is possible to optimally operate both targets at the same pressure and gas flow ratios prevailing in the chamber.

According to a further embodiment of the present invention, the transition layer is realized as a gradient layer, which has a decreasing proportion of TiAlN and an increasing proportion of Ti _x Si _1-x N with increasing distance from the substrate surface.

According to a further embodiment of the present invention, the final TixSi1-xN-layer is not a pure TixSi1-x-layer but still contains amounts of TiAlN.

According to a further embodiment of the present invention, a first TixSi1-x target and a second TiySi1-y target are used for the coating, where 0 ≤ x ≤ 1 and 0 ≤ y ≤ 1, but y ≠ x, ie first and second target differ in their composition and (x + y) / 2 ≤ 0.85, so that continue to produce layers with a Si concentration of ≥ 15 at%. In the process, both targets can then be operated according to the co-sputter methodology described above. This allows to vary the Si concentration during the coating, ie to realize a Si concentration curve. Table 1 coating Flank wear (μm) Corner wear (μm) benchmark 53.7 80.6 TiAlN + TiSiN Ti / Si = 90/10 45.5 61.7 TiAlN + TiSiN Ti / Si = 70/30 33.1 49.6

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• WO 2013060415 [0013]

Cited non-patent literature

• DIN 1.2344 [0018]

Claims (6)

Workpiece with coating which coating comprises at least one Ti _x Si _1-x N layer, characterized in that x ≤ 0.85 and the Ti _x Si _1-x N layer contains nanocrystals and the nanocrystals contained have an average grain size of not more than 15 nm where x is the concentration of Ti expressed in at% when only the metallic elements are taken into consideration. Workpiece according to claim 1, characterized in that between the Ti _x Si _1-x N layer and the substrate body of the workpiece, an intermediate layer is provided with TiAlN Workpiece according to claim 2, characterized in that between the intermediate layer and the Ti _x Si _1-x N layer, a transition layer is provided which contains both TiAlN and Ti _x Si _1-x N. Workpiece according to claim 3, characterized in that the transition layer is a gradient layer whose silicon content increases with increasing distance from the substrate surface. Process for producing a layer according to one of Claims 1 to 3, characterized in that a sputtering process is used for the production in which current densities of greater than 0.2 A / cm 2 are obtained on the target surface of the sputtering target and the target is a TixSi1-xN target where x ≤ 0.85. Method for producing a layer according to Claim 3, characterized in that the transition layer is produced by means of co-sputtering.

Images