WO2007115419A2

WO2007115419A2 - Wear-resistant layer for parts, method for coating a part with a wear-resistant layer, and device for carrying out said method

Info

Publication number: WO2007115419A2
Application number: PCT/CH2007/000167
Authority: WO
Inventors: Erich Bergmann
Original assignee: Ecole D'ingenieurs De Geneve (Eig)
Priority date: 2006-04-07
Filing date: 2007-04-02
Publication date: 2007-10-18
Also published as: WO2007115419A3; WO2007115419A9; JP2009532581A; EP2007922A2

Abstract

The invention relates to a coating material which protects parts against wear, a novel method for coating parts with a wear-resistant layer, and a device that allows said coating material to be applied to part surfaces by means of the inventive method. The coating material is embodied as a multiphase material that is composed of a hard material compound and a free nonmetallic element which is provided with a low friction coefficient and very high resistance against wear caused by sliding friction. In the inventive method, multiphase materials are reactively deposited, the two components of the compound being produced with the aid of physical vapor sources. The system required for carrying out said method comprises at least two physical vapor sources that are arranged in a special manner as well as a special turntable.

Description

Component wear protection layer, component wear protection coating method and apparatus for performing a component wear protection coating method

The invention relates to a multiphase component wear protection layer according to the preamble of patent claim 1, a PVD component wear protection coating method according to the preamble of claim 8, as well as an apparatus for carrying out a component wear protection coating method according to the preamble of claim 15.

The invention relates to a novel improved laminate which protects components against wear, a novel process for the Verschleissschutzbe- coating of components, and a device which allows this laminate with the inventive method applied to component surfaces. As components elements of machines, devices and tools are called.

The requirements for a component wear protection coating have been described in numerous publications and there are many suggestions for solutions from the past decades. The requirement profile here is component-specific and combines the weighted resistance to the wear mechanisms: abrasion, erosion, corrosion, surface disruption, "scuffing" with operating requirements such as friction coefficients in certain temperature ranges and media, wettability with certain liquids, temperature resistance, fracture toughness, etc. It has been shown that the wear resistance requirements can be met with hard materials in the sense of the monograph by Kieffer and Benesovsky (R.Kieffer, R. Benesovsky, Hartstoffe, Springerverlag, Vienna 1963), but these are very popular in almost all component applications high coefficient of friction, which leads to energy loss, seizure and wear of the counter body.

In order to meet the requirements for good sliding properties, ie low coefficients of friction, carbon has been produced in the state of the art for 25 years. holding component wear protection layers, z. As pure and modified carbon layers, as proposed by Hübsch and Dimigen in EP0087836. However, these layers have a high layer wear and an insufficient mechanical and thermal stability. The improvement of these properties has since been the subject of numerous proposals. The structure of these carbon layers is generally described with 3 parameters: the content of hydrogen, the ratio of sp ² - (graphitic) bonds to sp ³ - (diamond) bonds, the proportion of foreign substances - are common metals, silicon or nitrogen. The suggestions for improvement can be divided into two groups: demands on binding character and / or hydrogen-nitrogen content, formation of composite materials made of carbon and hard materials. Combinations of both groups have also been proposed. For example, D. Teer et al. in US6726993 suggested that in the carbon-containing component wear protection layer, the carbon should be substantially graphitic bound. In contrast, C. Strondl et al. in EP1123989 that in the carbon-containing component wear protection layer, the carbon should be bound substantially diamond. No. 6,722,6993 proposes Raman spectroscopy as the test method. Not the slightest problem with these proposals is that nobody knows how to determine the proportion of graphitic and diamond bonds on a complex component, but everyone knows that the layer structure depends very much on the geometry of the component, as a function of the time average Incident angle of the vapor determined on a particular functional area. This technical knowledge of the differences in the layer structure in different surface areas of components has been described for numerous layer systems and is described, for example, by GI van der Kolk et al. in DE10005612 to provide different surface areas with different carbon-containing wear protection layers having different technical properties.

The same problem exists with demands for a certain hydrogen content, as described by O. Massler et al. in DE10018143. The one Construction of nitrogen and the advantages of the presence of CN bonds has been described, for example, in US6726993, column 4 lines 13-17. The hydrogen content of a layer can be reliably determined only on level samples. Also for the hydrogen content GI van der Kolk et al. shown how to use it to provide various surface areas with different carbon-containing wear protective layers that have different technical properties.

For this reason, it has repeatedly been proposed not to use the structure but the technical properties of the layer as the test criterion according to the invention. Massler et al. that the layers should have a minimum hardness of 15 Gpa, preferably 20 GPa and an adhesion of at least HF3. Teer et al. again suggest that the wear rate of the layer should not exceed the value of ^10'16 rn ³ / Nm, that the Vickers hardness should be above 1000VHN and the critical load in the bond should exceed 70 N and that the dry friction coefficient should be below 0.1. G. Steffen et al. propose in DE19513614 that the ultramicrohardness should be in the range of 15-40 GPa and the coefficient of friction against a steel body should be <0.2, while H. Holleck et al. in DE19625329 propose to test friction coefficient, hardness and thermal conductivity. The fact that component wear protection layers are said to be hard, to have a low coefficient of friction and to adhere to the components is nothing other than the demand for the generic properties of a component wear protection layer. The fact that a component wear protection layer should be a component wear protection layer can not be sufficient as a sufficient test criterion for a layer structure. Component wear protection layers in the form of composites of carbon and metals had already been proposed by Hübsch and Dimigen. Details of these layers were later reported by Bergmann et al. in the Journal of Vacuum Science and Technology A4, 25867 ff and Le vide, les couches minces 235 Supplement p. 297 ff. Layers of this prior art state of the art consist of a matrix of hydrogen-containing carbon with embedded carbide crystals, with tungsten, titanium, chromium and others being used as the carbide formers. In addition, a multi-load genschicht by modulating the concentration of the carbide crystal at a frequency of about 4 nm. The method used is also the hybrid process of cathode sputtering and cathodic sputtering-activated acetylene pyrolysis, also referred to as dimpling process, the details of which are also described in DE10005612.

These coatings are sufficient in many applications such as slide bearings, roller bearings, various engine parts such as cam followers, camshafts, rocker arms, Kobenringen, parts of gasoline and diesel injection systems such as pistons, needles, injectors - and discs, gears of automotive transmissions, parts of hydraulic systems like valve pushers but also molds like injection molding and plastic molds do not meet the requirements, which is why we were looking for improvements. One of the requirements for a component wear protection layer is the requirement for a minimum layer thickness on all functional surfaces. The surface of components is not perfectly smooth, but has a roughness, resulting from the last processing step. This roughness is determined by standardized methods and characterized by the two parameters R ₃ and R ₂ . To be effective, the layer thickness of a component wear protection layer should be at least 4 times the R ₂ value, which corresponds to a layer thickness of at least 5 μm for most components.

To solve this problem, Teer et al. to use metal and carbon multilayers instead of the carbide concentration modulation layers. In this layer system, a minimum thickness of the individual layers of at least 3 nm is required in addition to the graphitic carbon bond (column 3 line 34). With a lower layer thickness, according to the inventors, the risk exists that the carbon reacts with the metal. The presence of carbides destroys the advantages of this layer system. Also, the presence of metal atoms in the carbon layer does not make a positive contribution to the layer properties. These layers are preferably hydrogen-free and are produced by non-reactive cathode sputtering.

Strondl et al. however, instead of the carbide concentration modulation layers, multiple layers of carbide and carbon are proposed. In these Layers with a diamond carbon bond should be the layer thicknesses of the carbide layers 1-3 nm, preferably 2 nm and the layer thicknesses of the carbon layers 1-20 nm. The layers preferably contain hydrogen. Burger et al. in US6869676 use instead of the layers with carbide concentration modulation multiple layers of hard materials and carbon, wherein the Schichstärken should be in the range of 1-10 nm, preferably 2-5 nm. The production of these layers succeeds them by replacing the PVD process by a CVD process, ie not using the sputtering plasma for the deposition of the carbon layers, but instead using special plasma sources that are spatially separated from the carbide sources. The carbon layers contain hydrogen. But even the so-improved component wear protection layers of the prior art have inadequate properties for many applications.

For slide bearings and gears and engine parts, the sliding wear rate is too high to meet the life cycle requirements demanded by modern automobiles. In roller bearings, the multilayer structure leads to failure due to fatigue. In the case of piston rings, it has been found that the layers of the prior art do not improve the "scuffing." The reasons for the failure are manifold: for example, in the development of the hydrogen-containing layers, their instability at temperatures above 350 ^° C. was also overlooked Due to friction, so-called hot spots occur regularly at certain points where the surface reaches temperatures of up to 800 ° C. The occurrence of hot spots is stochastic and linked to the surface roughness of the component The decomposition product, graphite, is rapidly abraded It could be said that the hydrogen-containing layers of the prior art burn off in spots - pyrolyzing would be the more accurate expression In turn, their inventors have proved that the layers in the frictional contact area to convert turbostratic graphite. Also this conversion leads to a constant wear.

The object of the present invention is to eliminate at least some of the disadvantages of the prior art.

The object is achieved according to the invention by a mer-phase component wear protection layer according to the features of claim 1, as well as by a method according to claims 8 and 15. Dependent claims 2 to 7, 9 to 14 and 16 to 22 relate to preferred embodiments of the invention ,

According to the invention, these problems can be avoided by avoiding hydrogen in contrast to the previous suggestions for improvement in the component wear protection layers, ie the layer contains no hydrogen, and suppresses the multilayer formation as far as possible. An investigation of the binding character of the layers can be omitted. Since the quality of this component wear protection coating according to the invention does not depend on the two parameters carbon bond type and hydrogen content, these layers can also be applied with uniform quality to all functional surfaces of a component.

There are two families of component wear protection coating methods for carbon-containing component wear protection coatings: chemical vapor deposition methods, usually called CVD, and physical vapor deposition methods, usually called PVD. Details of the distinction can be found in the Mater of Material, Volume 4, Analysis and Technology of the Surfaces, Chapter 15 (H.-J.Martieu, E. Bergmann, R. Gras: Presses Polytechniques et Universitaires Romandes, Lausanne 2003). There are numerous distinguishing features for CVD and PVD processes that lead to clear differences in the structure and properties of the layers. These unique differences in construction and properties make it easy for a person skilled in the art to distinguish between PVD layers and CVD layers. As a classic example of a CVD method may serve the document DE19513614, in which also numerous references to the older state of the art can be found: The components are placed in a vacuum oven on an electrically insulated table and under vacuum to a process-specific Tempe- brought. Thereafter, the reaction gases are introduced into the furnace and it is set a certain pressure. Thereafter, a plasma is ignited on the table on which the components are mounted by applying a voltage. There are numerous variants of the type of voltage, as described in detail in the other examples DE10018143 and US6869676. The disadvantages of the CVD methods are also presented in the Trait des materiaux, Volume 4 Analysis et Technology of the Surfaces, Chapter 15 (H.-J. Mathieu, E. Bergmann, R. Gras: Presses Polytechniques et Universitaires Romandes, Lausanne 2003): Difficulty of gas guidance due to the most commonly used pressure range of 10 - 1000 Pa, in that the flow is not molecular and not turbulent but laminar; In addition, plasma-assisted CVD processes, such as those used exclusively for the carbon-containing component wear protection layers, have difficulty in controlling the composition of the gas mixture to deposit layers of reproducible and uniform composition. This latter problem is even further enhanced in hybrid processes but also used positively, as described in DE10005612 (GI van der Kolk et al mistakenly refer to their hydrocarbon pyrolysis as PVD).

The basis of physical coating processes is the incorporation of coating components into the transport phase through the physical processes of evaporation, sputtering and peeling. In the case of physical coating processes, a distinction has hitherto been made between reactive and non-reactive processes. In non-reactive processes, the layers are produced by condensation on the component surface. Depending on the material, similar compounds are formed in the condensation, as they were also present in the starting material: When using a physical vapor source or using several physical vapor sources with identical vaporization material, The coating material speaks the sputtering material, the sputtering of AISn20Cu alloy targets provides AISn20Cu alloy layers, the evaporation of BaTiθ ₃ tablets provides BaTiO ₃ layers. Laser peeling of M0S ₂ targets provides M0S ₂ layers. It is now also possible to form more complex layers by co-evaporation from several crucibles and / or atomization and / or peeling of several vapor deposition material plates, also called "targets." AISn20Cu layers can also be obtained by sputtering aluminum from an aluminum target and from bronze from a bronze target or make (BaTiO ₃ ) _m (SrTiO ₃ ) _n layers from two different crucibles by evaporation of BaTiO ₃ and SrTiO ₃ , with the sputtering sources and / or crucibles arranged so that they are on all functional surfaces This is achieved in the prior art by using only physical vapor sources with very small areas that are as close together as possible and keeping the workpieces at a very large distance.

In reactive processes, in addition to the steam introduced with physical processes, a gas is introduced into the coating system which reacts with the steam introduced on the workpiece surface by physical processes. Suitable gases are oxygen, nitrogen, water vapor, hydrocarbons, boranes and hydrogen sulphide or simple compounds such as carbon oxides or nitrogen oxides, in order to precipitate metal oxides, oxides, nitrides, carbides, borides, sulfides or oxycarbides and oxynitrides. If a pyrolyzed product of the introduced gas, such as free carbon or free sulfur, is also incorporated into the layers, this is referred to as hybrid processes. Processes in which only small amounts of reactive gas are introduced in order to avoid loss of stoichiometry are referred to as non-reactive processes. For decades, the widespread industrial process of oxide coating by oxide evaporation with the addition of small amounts of oxygen may serve as an example. Most of the processes for depositing metal carbon layers are ameliorative variants of the hybrid method already proposed by Hübsch and Dimigen, in which cathode sputtering of a metal or carbide is combined with plasma gaseous hydrocarbon pyrolysis, a plasma activated CVD process. The improvements relate either to the process design of the CVD process as exemplified by US6372303 or the replacement of cathode sputtering by cathodic arc evaporation, as proposed by J. Lauzarita and A. Alberdi in EP0607736 or Wang Da-Yung in TW495553 or Sato Toshiki in JP2001172763. Also, a combination hybrid method has been proposed by I. Yoshinori and U. Yoshinaru in JP2003082458. The other teaching is described by D.Teer et al. in US6726993, who propose a pure PVD process, in their case the sputtering. The systems of the prior art are described, for example, in DE10005612 or US6726993. In both cases it is also possible to use laser peeling devices or arc evaporators as physical vapor sources, as described, for example, in US Pat. No. 6,869,676. The implementation of the inventive method is not possible in these systems. The method according to the invention requires substantial changes in the device design which ensure that no multilayers are deposited for which these prior art systems have been designed. Details will be explained with reference to examples.

We have found that the problems of rapid wear of the prior art component wear protective layers can all be attributed to the inadequate wear resistance of the hydrogen-containing carbon matrix or layers used in these layers. It has been found that by avoiding the presence of carbon layers, the component wear protective layers are much more resistant to wear and this can be achieved without adversely increasing the coefficient of friction, as has been the case in the prior art component wear protection layers. The starting point of the invention was the idea The wear characteristics of cathodic evaporation deposited carbon layers for sliding elements, described as excellent by RP Welty et al. WO2005 / 015065 described as starting point for one of the metal-containing carbon layers, as they are described in the prior art nik technology to use and so to develop a suitable layer for components.

definitions:

Metalloids - also called nonmetals - are chemical elements that form covalent or ionic compounds with metals. These are the halides, the chalcogenides, nitrogen, phosphorus, arsenic, carbon, silicon, germanium and boron. (See Materials Science and Engineering: An Introduction, William D. Callister Jr., John Wiley & Sons Inc. 5th edition, section 2.4)

All substances appear as different phases. Phases are: gas, liquid, crystals of different symmetry. The number of phases under which a substance can occur is determined by the Gibbs phase rule. A material consists of one or more phases of one or more substances. Accordingly, it is referred to as single-phase, two-phase, three-phase, etc. All non-single-phase materials are also referred to as multiphase

The invention will now be explained in more detail below, for example, and with schematic figures for preferred embodiments. Show it:

Figure 1 shows a cylindrical coating chamber 1. It corresponds to the device used in JP2003082458. It is equipped with means for generating vacuum, which are connected to the pump nozzle 2, and devices for monitoring the vacuum, which are not shown in the picture, equipped. In its center is a turning device 7, are attached to the workpiece and / or component rotary support 9. In turn, the workpieces 19 and components 20 are fastened to these workpiece and component carriers. The rotation Direction 7 is constructed so that they both put all the workpiece carrier in the same rotational movement 21, as well as the individual Werkstückträgem a self-rotation 22 gives. The bottom of the coating chamber can be equipped with various physical vapor sources 3, 4, 5, 6. In this and the following examples, magnetron sputtering devices and cathodic arc evaporators have been used in various combinations. The four physical vapor deposition devices furthermore comprise power supplies 10, 11, 12, 13, whose negative output is connected with cables 13, 14, 15, 16 to a plate of material to be introduced into the transport phase with physical processes, referred to below as the target plate. the surface of which faces the workpiece and / or component carriers 9. The target plate is also the surface of the physical vapor sources. The positive outputs of the power supplies 10, 11, 12, 13 were connected to earth. The chamber was also grounded with a ground wire 18. The magnetron sputtering devices and the cathodic arc evaporators were state of the art. The rotating device 7 also includes a gas mixture inlet system 8, by means of which gas mixtures generated by mass flow regulators can be introduced into the plant. It will be clear to those skilled in the art that this coating chamber may be constructed differently, for example as described in JP2003082458 or US6726993 - in the latter case, the four 4 physical vapor sources would form unbalanced magnetron sputtering magnetic field forming devices - and like the inventions other system geometries must be transferred.

FIG. 2 schematically shows a detail of the system 1. By way of example, it contains a magnetron cathode sputtering device 3 and a cathodic arc evaporator 5. Details of these devices are known from the prior art. The magnetron sputtering apparatus and the cathodic arc evaporator each consist of a base body 23, 24, which contains the cooling devices and the devices for generating the magnetic fields, a target plate 24, 27 and target plate fastening devices 25, 28. The magnetron sputtering apparatus 3 is surrounded by an anode screen 26, while the cathodic arc evaporator is surrounded by an arc-limiting screen 29.

FIGS. 3 to 8 are electron and light microscope photographs of layers and layers according to the invention which correspond to the prior art. They will be explained using the examples.

FIG. 3 shows electron microscopic multilayer films according to the prior art. FIG. 3a is a bright light image with the usual magnification. FIG. 3b is a high resolution detail of FIG. 3a. Figure 3c shows the distribution of the element chrome for the same section as Figure 3b. 3d shows the distribution of the element carbon for the same section as 3b and 3c.

FIG. 4 is a greatly enlarged photomicrograph of a section in a calotte cut.

FIG. 5 is an electron micrograph of a component wear protection layer according to the invention.

FIG. 6 is a greatly enlarged photomicrograph of a section in a calotte cut of a component wear protection layer according to the invention.

Figure 7a is a high resolution section of Figure 5. Figure 7b shows the distribution of the element chrome for the same section as Figure 7a. Fig. 7c shows the distribution of the element carbon for the same section as Figs. 7a and 7b.

FIG. 8 is an example of an arc-root-point guide according to the invention for a round cathode. It is explained in Example 12.

The invention will now be explained in more detail by way of examples. The examples represent preferred embodiments. Example 1 :

A system as shown in FIG. 1 was used. In a first attempt, attempts were made to deposit layers according to the prior art. For this purpose, the system was equipped with 4 magnetron sputtering devices 3, 4, 5 and 6, which were equipped with target plates 24 made of chrome. As power supplies 4 current controlled high voltage supplies 10, 11, 12, 13 were selected. Finely ground martensitic stainless steel discs and hardened steel piston ring portions and bushings 20 of aluminum alloy common to these components were used as workpieces 19. These were cleaned by a prior art industrial cleaning method and mounted on the workpiece and component carriers 9. The chamber was pumped to a pressure of 4 mPa. The component carriers were set in a rotational movement 21 at a speed of 3 revolutions / minute and the component carriers of the cylindrically symmetrical components were additionally set in an autorotation 22. After a conditioning step in an argon plasma, an argon pressure of 0.3 Pa was set in the chamber. Thereafter, the high voltage supplies 10, 11, 12, and 13 were turned on and a current of 7 amps each was set. After 10 minutes, the argon gas was replaced with a 1: 4 mixture of argon: acetylene and the pressure was simultaneously raised to 1 Pa. After 1 hour, the high voltage power supplies 10, 11, 12, 13 were switched off and the system was opened. Under these conditions, an adhesive layer of 10 μm chromium had previously been deposited in an acetylene-free test. On the workpieces were layer fragments of different thicknesses. The floor of the chamber was covered with black stratified fragments. More detailed investigations showed that a layer of a carbon matrix with enclosed carbide crystals according to the prior art had been deposited on the parts, but that this layer peeled off in itself, since apparently its elongation strength was insufficient. The adhesion of the chrome undercoat was impeccable. Attempts to improve by changing the substrate stress or the transition from the chromium coating to the coating with chromium-containing carbon. material did not bring any significant improvement. Even the deposition of multilayers brought only an insignificant improvement. These were generated by periodically stopping the acetylene flow for 40 seconds every 40 seconds.

Example 2:

The same device was used as in Example 1 except that the method described by D. Teer et al. implement proposed improvements. The magne- ton sputtering devices at positions 3, 4 and 5 were equipped with target plates made of graphite. The workpieces 19, 20 were cleaned as in the previous example, charged and conditioned in argon plasma. After the conditioning step in the argon plasma, an argon pressure of 0.3 Pa was set in the chamber. Thereafter, the high voltage supplies 10, 11, and 12 were turned on and set a current of 6 amps each. Different coating times have been tried. However, layers with a layer thickness of more than 1.5 μm burst. Thereafter, at position 6, a fourth magnetron sputtering device was added, which was equipped with a target plate made of chromium. The procedure up to and including the conditioning again corresponded to that of Example 1. After the conditioning step in argon plasma, an argon pressure of 0.3 Pa was set in the chamber. Thereafter, the high voltage power supply 13 was turned on and its current set to 8 amps. After 10 minutes, the high voltage supplies 10, 11, and 12 were turned on and a current of 6 amps each was set. These powers were chosen so that the rotational movements 21 generated by the rotary device movement 7 should deposit multiple layers of about 5 nm thickness of alternating chromium and carbon. The component carriers were set into a rotational movement 21 with an increased speed of 10 revolutions / minute compared to Example 1 and the previous experiment, and the component carriers of the cylindrically symmetrical components were additionally set in an autorotation 22. A rotational speed of 10 revolutions / minute corresponds to the maximum which can be used with rotary devices 7 according to the prior art. There were tried different coating times. Only layers of less than 5 μm could be deposited without flaking. An adhesive with rock wave print on a sample of hardened stainless steel showed HF5. However, most of the spalling was observed in the film. Obviously, the layers according to US6726993 have too low tensile strength and ductility, which may be due to their graphitic bonding structure. Graphite is known to have a particularly low tensile strength of all materials and is hardly plastically deformable.

To solve the problems, we wanted to replace the graphitic with diamond bonding and replaced the magnetron sputtering of graphite by cathodic arc evaporation, as described in the prior art by Welty et al. especially recommended for sliding applications.

Example 4:

Again, a device as in Example 1 was used. Position 6 was equipped with a cathodic arc evaporator equipped with a carbon target plate. As power supply 12, a high-power supply KEMPPI 320 was used, as is customary in welding technology. Position 5 was equipped with a cathodic arc evaporator equipped with a chromium target plate. As power supply 13 again a high power KEMPPI 320 was used. Positions 3 and 4 were equipped with magnetron sputtering devices equipped with chromium target plates. For the power sources 10 and 11 again high voltage supplies were used. Of course, all other changes necessary to convert cathode sputtering to cathodic arc evaporation were also made. For example, the high-frequency insulated high-voltage cables 15 and 16 were replaced by copper cables with 120 mm ² cross section. The arrangement corresponded to JP 2003082458. The workpieces were cleaned, charged and conditioned in argon plasma as in the previous example. After the conditioning step in the argon plasma, an argon pressure of 0.3 Pa was set in the chamber. The component carriers were set in a rotational movement 21 at a speed of 10 revolutions / minute and the component carrier of the cylindrically symmetrical components were additionally offset in an internal rotation 22. The current sources 10 and 11 were turned on and the current was regulated to 8 amps in both. After 5 minutes, nitrogen was admitted into the chamber and the total pressure was increased to 1 Pa with this nitrogen. After 20 minutes, the power source 12 was turned on and set a current of 120 A. An arc was ignited on the cathodic arc evaporator. Thereafter, the nitrogen flow and the two power sources 10 and 11 were turned off. After 45 minutes, the power source 12 was turned off. When the parts were opened, the majority of the carbon layer had broken down, that is, due to cohesive failure, parts of the layer had flaked off. The adhesion of the chromium nitride layer to the substrate and the adhesion of the carbon layer to the chromium nitride layer were satisfactory. The method corresponded to the prior art JP200382458 and WO2005 / 015065.

Example 5:

After the methods of the prior art JP200382458 and

WO2005 / 015065 have not yielded a satisfactory result, the prior art methods have been tried with the layering ideas as used in other methods, in particular for CVD methods in US6869676 for PVD methods in US6726993 and for hybrid methods of Strondl et al. in EP1123989 were proposed to combine. Again, a device as in Example 4 was used. However, due to the asymmetry created by the cathodic arc evaporators over the arrangement of Example 2, it was no longer possible to provide a closed magnetic circuit for the unbalanced magnetron sputtering devices. The workpieces were cleaned, charged and conditioned in argon plasma as in the previous example. After the conditioning step in the argon plasma, an argon pressure of 0.3 Pa was set in the chamber. The component carriers were placed in a rotational movement 21 at a speed of 10 revolutions / minute and the component carriers of the cylindrically symmetrical components were additionally offset in a spin 22. The current sources 10 and 11 were turned on and the current was regulated to 8 amps in both. After 3 minutes, the power source 14 was turned on and a current of 120 A was set. An arc was ignited on the cathodic arc evaporator 6. After 45 minutes, the power sources 11, 12 and 14 were turned off. The parts had been coated with a layer of 7 μm. There were no spontaneous flaking. In the adhesion test, however, cohesive flaking was noted. Calotte cuts and cross sections for transmission electron microscopy were made and examined. The transmission electron micrograph is reproduced as image 3. Figure 3a shows that the layer has a multilayer structure, with carbon lamellae 32 having a thickness of less than 2 nm alternating up to 8 nm with chromium layers 33 having a thickness of 12 nm. Figure 3b shows the enlargement of a section confirming the presence of continuous carbon layers and continuous metal layers. The calotte finish showed why these layers fail in the adhesion test. The chrome layers break off from the carbon layers. In further experiments, the set currents of the current sources 11, 12 and 14 were varied, so that the layer thicknesses of the chromium layers between 2 nm and 12 nm and the layer thicknesses of the carbon layers varied between 1 nm and 20 nm. No significant improvement was found. On the contrary, it has been shown that Strondl et al.'S proposal to make the metal layers thinner than the carbon layers results in deterioration. In contrast to the magnetron sputtering methods and the hybrid method, the layer thickness of the carbon layers varies greatly during cathodic arc evaporation. This is due to the slow movement of the cathode footprint on carbon targets due to the process and can not be suppressed. There are numerous publications on the problem of controlling arc movements on carbon targets. The control used in this example was more powerful than the prior art. Example 6.

The procedure was as in Example 5. After arc ignition, an additional 40 sccm (standard cubic centimeter / minute) of acetylene was added to the equipment to complete the process by adding a CVD component to the proposal of Burger et al. Approach US6869676. The so deposited layers showed larger flakes in the calotte cut than the layers of Example 5.

Example 7: Again, a device as in Example 4 was used. The workpieces and components were cleaned, charged and conditioned in argon plasma as in previous examples. After the conditioning step in the argon plasma, an argon pressure of 0.3 Pa was set in the chamber. The component carriers were set in a rotational movement 21 at a speed of 10 revolutions / minute, and the component carriers of the cylindrically symmetrical components were additionally set in an autorotation 22. The current source 13 was turned on, the current was set to 85 amps. Thereafter, an arc was ignited on the cathodic arc evaporator 5 and carried out a metal plasma conditioning, as known from the prior art. At the end of metal plasma conditioning, the current of current source 13 was increased to 120 amps and a chrome plating was initiated by lowering the workpiece bias. Thereafter, the current sources 10 and 11 were turned on and the current was regulated at both to 6 amps. After 3 minutes, the power source 14 was turned on and a current of 120 A was set. On the cathodic arc evaporator 6, an arc was ignited. After 35 minutes, the power sources 11, 12, 13 and 14 were turned off. The parts had been coated with a layer of 8 μm. There were no spontaneous flaking. In the adhesion test, however, cohesive flaking was noted. Calotte slices were made and examined. The layer had a multilayer structure with carbon lamellae of less than 2 nm thickness up to 8 nm alternating with 20 nm thick chromium layers. The calotte finish showed why these layers fail in the adhesion test. The chromium layers burst from the carbon layers in many places. Compared to Example 5 but a noticeable improvement was found.

Example 8: Again, a device as in Example 5 was used. The workpieces were cleaned, charged and conditioned in argon plasma as in previous examples. After the conditioning step in the argon plasma, an argon pressure of 0.3 Pa was set in the chamber. The component carriers were set in a rotational movement 21 at a speed of 10 revolutions / minute and the component carriers of the cylindrically symmetrical components were additionally set in an autorotation 22. The current source 13 was turned on, the current was set to 85 amps. Thereafter, an arc was ignited on the cathodic arc evaporator 5 and carried out a metal plasma conditioning, as known from the prior art. At the end of metal plasma conditioning, the current of current source 13 was increased to 150 amps and a chrome coating was started by lowering the workpiece bias. After 3 minutes, the power source 14 was turned on and a current of 100 A was set. On the cathodic arc evaporator 6, an arc was ignited. After 45 minutes, the current sources 13 and 14 were turned off. The parts had been coated with a layer of 7.5 μm. There were no spontaneous flaking. In the adhesion test, however, cohesive flaking was noted. Calotte slices were made and examined. The layer had a multilayer structure with carbon lamellae less than 2 nm thick up to 6 nm alternating with 10 nm thick chrome lamination layers. The calotte grinding is shown in Figure 4. He shows why these layers fail in the adhesion test. The chromium layers 34 burst from the carbon layers 33 at a plurality of locations 37. Compared to the layers of Example 7 but a noticeable improvement was found. Example 9:

The plant used in the previous examples was rebuilt. The magnetron sputtering devices 3 and 4 were removed. So that the system was equipped only with the two cathodic arc evaporators 5 and 6 on the positions shown in Figure 1. Here, the cathodic arc evaporator 6 was equipped with a carbon target, the cathodic arc evaporator 5 with a chrome target. The rotary device and the component carrier 9 were replaced by a new construction as follows: The rotary device allowed a rotational speed of 100 revolutions / minute. To achieve this, a new powerful engine with a corresponding gear was used. All other assemblies were also redesigned: All plain bearings were replaced by ball bearings. The substrate power supply was disconnected from the motion supply and the electrical contact was made directly to the substrate receiving spindles. The plug-in connections to the substrate receptacles were replaced by screw connections or, preferably, form-fitting connections. All components that were exposed to the centrifugal forces have been reinforced compared to the prior art. The workpieces used were precision ground martensitic stainless steel discs and hardened steel piston ring portions and bushings 20 of aluminum alloy common to these components. These were cleaned with an industrial cleaning method according to the prior art and fixed to the workpiece and component carrier 9. The chamber was pumped to a pressure of 4 mPa. The component carriers were placed in a rotation movement 21 at a speed of 100 revolutions / minute and the component carriers of the cylindrically symmetrical components were additionally set in an autorotation 22. After a conditioning step in an argon plasma, an argon pressure of 0.3 Pa was set in the chamber. The current source 13 was turned on, the current was set to 85 amps. Thereafter, an arc was ignited on the cathodic arc evaporator 5 and a metal plasma conditioning, as known from the prior art, performed. At the end of metal plasma conditioning, the Current of the current source 13 is increased to 150 amps and started by lowering the workpiece bias a chrome plating. After 3 minutes, the power source 14 was turned on and a current of 100 A was set. On the cathodic arc evaporator 6, an arc was ignited. After 50 minutes, the current sources 13 and 14 were turned off. The parts had been coated with a layer of 8 μm. There were no spontaneous flaking. Calotte cuts and cross sections for transmission electron microscopy were made and examined. These studies show that the layer has no multilayer structure. Carbon layers and chrome layers are weakly recognizable. The structural analysis shows that the layers consist essentially of chromium carbide and amorphous carbon. The adhesion test no longer shows any cohesive failure. In the Kalottenschliff one can still recognize isolated local flaking in the layer. Friction coefficient and wear rate measurement was performed on the coated wheels. Both measurements were carried out with a sintered alumina ball. A sliding friction coefficient of 0.3 was measured. The wear rate was 1.5 x 10 ^"16 m ³ / Nm. An analysis of the composition with electron beam induced endergiedispersiver analysis of X-ray radiation revealed 62 at% carbon and 38% chromium.

Example 10:

The plant was rebuilt again. The cathodic arc evaporator equipped with the chrome target was moved from position 5 to position 3. Wiring and cooling water supply have been changed accordingly. Thereafter, a coating was carried out as in Example 9. The parts had been coated with a layer of 8 μm. There were no spontaneous flaking. Calotte cuts and cross sections for transmission electron microscopy were made and examined. The transmission electron micrograph is reproduced as FIG. FIG. 5 shows that the layer has no multilayer structure. There are no carbon layers and chrome layers visible. However, you still can Recognize a composition modulation 39 with a frequency of 2 nm. This corresponds to a layer deposited during one revolution. The calotte grinding is shown in FIG. You can not see a single local flaking. The refined structural analysis, Figure 7 shows that the layers consist essentially of chromium carbide whose grain size is in the nm range, and amorphous carbon. The grain size varies irregularly in the course of the shift. The layers contain no detectable metallic chromium. A multi-layered structure is still partially present, but there are no longer continuous carbon louvers. The layer has portions 38 in which the structure is nearly isotropic, that is, chromium carbide grains and carbon particles. The size of the chromium carbide grains is in the range of 1 to 5 nm. It is also crucial that the carbon lamellae of carbide grains 39 are also interspersed in the region of the layer in which a multilayer structure still exists. The absence of continuous carbon lamellae probably results in excellent tribological properties.

An analysis of the composition with end-dispersive analysis of the electron-beam-induced X-radiation showed 62 at% carbon and 38%

Chrome.

Friction coefficient and wear rate measurement were performed on the coated wheels. Both measurements were carried out with a sintered alumina ball. A coefficient of sliding friction of 0.22 was measured. The wear rate was 3 × 10 ^-17 m ³ / Nm.

Example 11: The construction of the plant was the same as in Example 10. The coating process was also the same as in Example 10. However, 3 minutes after ignition of the arc on the arc evaporator 6, nitrogen was introduced into the plant namely 160 sccm. The coating was then continued for 50 minutes with constant parameter setting. Upon opening, it was found that the parts were coated with a layer of 10 microns. There were no spontaneous flaking. The transmission electron micrograph showed that the layer has no multilayer structure. There were no carbon layers and chrome layers recognizable. However, one can still see a composition modulation with a frequency of 2 nm. This corresponds to a layer deposited during one revolution. In the Kalottenschliff you could not detect a single local flaking. The structural analysis showed that the layers consist essentially of chromium carbide or chromium carbonitride whose grain size is in the nm range, and consist of amorphous carbon. The two chromium compounds could not be distinguished with the analyzes used. The grain size varies irregularly in the course of the shift. The layers contain no detectable metallic chromium. The structure is similar to that of the layers of Example 10.

An analysis of the composition with end-energy dispersive analysis of the electron beam-induced X-radiation showed 58 at% carbon, 32% chromium and 10% nitrogen.

Friction coefficient and wear rate measurement were performed on the coated wheels. Both measurements were carried out with a sintered alumina ball. A coefficient of sliding friction of 0.20 was measured. The wear rate was 1.2 × 10 ^-17 m ³ / Nm.

Example 12:

The cathodic arc evaporation device 6 was rebuilt. The magnetic guidance of the cathode footpoint has been changed. The magnetic guidance of the cathode footpoint of Examples 4-11 is described in Figure 8a. The magnetic guide of the cathode foot of Examples 4-11 had consisted of magnetic lines 39 whose vertices formed 3 rings 41. The unillustrated magnetic field generating device rotated about an axis normal to the target plate, causing the rings to orbit the target plate center. In this magnetic field, the cathode footpoint of a web 40 has followed, which has repeatedly left the imprinted leadership and has slowly changed from one ring to another. This magnetic field has now been replaced by a simpler one, shown in Figure 8b. The vertex Points of the lines 42 of this magnetic field form a circle. This magnetic field has numerous disadvantages compared with the magnetic field 8a, which will not be discussed in detail here. But, the cathode foot point performs a much faster and, above all, even circular meandering movement 43. The other assemblies were left unchanged from Example 11. The coating process was carried out as in Example 11 and 10 μm of layer were again deposited. Calotte cuts and cross sections for transmission electron microscopy were made and examined. The result of transmission electron microscopy is reproduced as a dark field image in FIG. The structure of the layer corresponded to that of Examples 10 and 11. But the grain size varied regularly in the course of the layer. This grain size variation can be observed in the dark field image taken with the chromium carbide line as "Milky Way Lanes" 44. A layer with a larger particle size corresponds to one "Milky Way". The frequency of the grain size layers corresponds to the frequency of Kathodenfusspunktrotation. It is believed that depending on the position of the cathode root point to the parts, the crystallite size of the deposited layer is larger or smaller. Whether the distance of the Katodenfusspunkts from the component surface or the angle between the connecting line from the cathode foot to the component surface and the normal to the target surface is the decisive parameter could not be determined with our device. Nor could it be determined whether the coarser grains corresponded to a shorter or longer distance. Friction coefficient and wear rate measurement were performed on the coated wheels. Both measurements were carried out with a sintered alumina ball. A coefficient of sliding friction of 0.18 was measured. The wear rate was 2 × 10 ^-18 m ³ / Nm.

The observations made in these examples can be explained as follows. Substantially single-phase device wear protective layers or layers consisting of a carbon matrix with carbide inclusions, such as those described in US Pat According to the state of the art, they have such poor mechanical properties that they burst spontaneously with larger layer thicknesses. Multi-layer coatings are an improvement, but they still fail cohesively. Good wear properties require the absence of metal and the absence of continuous carbon louvers. The experiments were carried out only with chromium and carbon. However, it is clear that a similar behavior and similar good properties can be achieved with other carbide-forming metals. Also, alloys that consist essentially of carbide-forming metals seem suitable. The choice of the metal or the alloy will be oriented by the person skilled in the art to the conditions of use of the component. Instead of carbide layers and layers of sulfides, silicides and borides are conceivable. In which then, for example, sulfur, silicon and boron would be embedded. We assume that the periodic variation of the concentration of the metalloid and the metals in some cases facilitates the formation of compound crystals and prevents the deposition of amorphous layers. In the investigated system, a variation with a period of 2 nm has proven to be useful. The addition of nitrogen, which is likely to be in the form of a carbonitride, to the gas atmosphere in which the coating is carried out has also been proven. To ensure a low coefficient of friction, the nitrogen content of the layer should not exceed the carbon content.

The composition of the layer is determined by the ratio of the evaporator streams in the preferred method of Kolfbogenverdampfung for their preparation. In other methods, such as sputtering, one would use the target performance for composition optimization. The optimal composition should be within a certain range. Details depend on the area of application of the component and the carbide formation during the coating. The latter can be controlled, for example, in the system chromium-carbon by the coating temperature and the workpiece preload during the coating process. A lower range of 20, preferably 40 atomic percent carbon should not be undercut to T / CH2007 / 000167

26

not to suppress its free carbon. An upper range of 85, preferably 70 atomic percent carbon, should likewise not be exceeded in order to avoid the formation of a carbon matrix, which, as shown, subsequently affects the mechanical properties.

Wear protection layers according to the prior art were generally thin, 1-3 μm. Our experiments have shown that such thin layers are unsuitable for component wear protection, as they are rubbed too quickly at the roughness peaks. The prior art used such thin layers because the layers have insufficient mechanical properties. The last example has shown that a periodic variation of the grain size has a particularly advantageous effect on the wear rate. The reactive co-evaporation process we invented can be applied to all materials consisting of compounds whose components can be vaporized by physical methods. It is naturally suitable for the deposition of carbides, bonding sulphides and silicides of the metals, but also more exotic compounds such as tellurides may be presentable by this process.

An essential characteristic of the method according to the invention is that the workpieces and components are exposed simultaneously or at very short intervals to the vapor of a physical metal source and a physical metalloid source. This is achieved in that the two physical vapor sources are arranged at a certain distance from each other, which of course also depends on the distance of the functional surfaces to be coated of the workpieces and components of the surfaces of the two sources. A maximum distance of 150 mm should only be exceeded if it is compensated by other plant engineering measures.

If the method according to the invention is realized in such a way that the workpieces and components are alternately exposed frontally to different physical vapor sources, the period during which the workpieces and components should be exposed 167

27

are exposed to a physical vapor source, be so short that no more like 2-5 nm are deposited.

To carry out the method, cathodic arc evaporators have proven particularly useful. In the studied example of the combination

Chromium / carbon, the co-use of magnetron cathode sputtering devices for the production of carbon vapor had a negative effect on the mechanical properties of the layers. On the other hand improved in this

For example, the use of a nitrogen-containing atmosphere has the layer properties, while the use of a hydrocarbon-containing atmosphere has degraded the layer properties.

The device according to the invention is characterized by a special turntable, which allows to move the workpieces and components very quickly from one to the next physical vapor source. Such turntables have not been known in the prior art since they have many disadvantages with respect to the support structure. The brackets for devices according to the invention will generally be more expensive to manufacture and to assemble. However, this is more than offset by the unique properties of the layers that can be produced in them.

A further improvement of the carbonaceous component wear protection coating has been achieved by guiding the functional surfaces of the components at fairly rapid intervals in the vicinity of the cathode footprint on the carbon target. This is achieved by a special guidance of the cathode foot of the carbon evaporating arc evaporator. The illustrated but not exclusive example uses a strong magnetic field forming an elliptical channel. A particular embodiment of the ellipse is the circle. The oscillatory motion normal to the axis of rotation is the projection of the path of cathode footing 43 onto the diagonal of the target plate which is normal to the axis of the turntable movement. The time period of the oscillatory motion should, however, be much larger than the period in the functional surface is moved from one physical vapor source to the next. A period of 1 - 6 seconds has been proven.

Claims

claims

1. Multiphase component wear protection layer consisting of a metal of the subgroups of the Periodic Table of the chemical elements or an alloy thereof, preferably Cr, W, Ti, V, Zr, Nb, Mo, Ta and a single metalloid from the group C, Si, B , and S, characterized in that i) the device wear protection layer consists essentially of compounds of the metals with the metalloid and the metalloid, ii) the device wear protection layer contains substantially no hydrogen iii) the device wear protection layer substantially no detectable layers of metal - except any Adhesive layer on the workpiece or component surface - has, iv) the component wear protection layer has substantially no continuous layers of the metalloids.

2. Multiphase component wear protection layer according to claim 1, characterized in that the ratio of the concentration of the metalloid to the sum of the concentrations of the metals from the layer surface to

Workpiece or component surface varies periodically.

3. Multiphase component wear protection layer according to claim 1 or 2, characterized in that i) is used as metalloid carbon, ii) as metal Cr, W, Ti, Zr, Nb, Mo or Ta is used, iii) the carbon concentration 20 - 85% , preferably 40-70%.

4. Multiphase component wear protection layer according to claim 3, characterized in that the layer additionally contains nitrogen.

5. Multiphase component wear protection layer according to claim 4, characterized in that the nitrogen concentration does not exceed the carbon concentration.

6. Multiphase component wear protection layer according to one of claims 2 to 5, characterized in that the periods at which the ratio of the concentration of the metalloid to the metals from the layer surface to the workpiece or component surface varies 5 nm, preferably not exceed 2 nm.

7. Multiphase component wear protection layer according to one of claims 1 to 6, characterized in that i) it has a multilayer structure, ii) the individual layers differ substantially only in the grain size of the compound of the metals with a metalloid.

8. PVD component wear protection coating method, characterized in that i) at least one physical vapor source is used to produce metal vapor, ii) at least one physical vapor source is used to generate metalloid vapor, iii) the metal vapor and the metalloid vapor strike the workpiece surface in such a way react to a connection.

9. PVD component wear protection coating method according to claim 8, characterized in that the metal vapor is a vapor of the metals of the subgroups of the periodic system of the chemical elements, preferably Cr, W, Ti, V, Zr, Nb, Mo, Ta or alloys, in which these metals form the main alloying fraction and that the metalloid vapor is steam of the elements C, Si, B, and S is preferably C.

10. PVD - component wear protection coating method according to claim 9, characterized in that i) the metal oxide vapor is carbon ii) the time average of the ratio of the incident on the workpieces and component functional surfaces molar flow of carbon to the molar flow of the metal greater than 0.2 and smaller is 4.5, preferably, greater than 0.4 and less than 2.5

11. PVD - component wear protection coating method according to one of claims 8 to 10, characterized in that i) the ratio of the incident on the workpieces and component functional surfaces molar flow of the metal vapor to the molar metalloid vapor periodically changes with a time period At, ii) the time period .DELTA.t less than which is 5 × 10 ^-9 times, preferably 2 × 10 ^-9 times the coating rate expressed in m / sec.

12. PVD - component wear protection coating method according to any one of claims 8 to 1 1, characterized in that as one of the two physical vapor sources, a cathodic arc evaporator is used.

13. PVD - component wear protection coating method according to any one of claims 8 to 12, characterized in that the coating is carried out in a nitrogen-containing atmosphere.

14. PVD - component wear protection coating method according to any one of claims 8 to 13, characterized in that the coating is carried out in an atmosphere substantially free of hydrogen or hydrogen-containing compounds.

15. Apparatus for carrying out a component wear protection coating method comprising a vacuum chamber (1), physical vapor sources (3), (4), (5) and (6), a rotary device (7) and component rotary carriers (9), characterized in that i ) The device contains at least 2 physical vapor sources, ii) at least 2 physical vapor sources are equipped with different materials to be introduced into the vapor phase with physical processes, iii) at least one physical vapor source for the production of carbon, sulfur, silicon, or Iv) each 2 physical vapor sources and the rotating device are arranged so that all components are periodically exposed to both physical vapor sources, v) that the time period Dt, in which the component carriers have reached the same position again smaller than the 5th 10 ^"9 times, preferably 2 10 ^{" 9} times the inverse coating rate expressed in m / sec.

16. A device for carrying out a component wear protection coating method according to claim 15, characterized in that the rotational speed of the rotary device is at least 10 revolutions / minute, preferably at least 50 revolutions / minute.

17. Device for carrying out a component wear protection coating method according to claim 15, characterized in that the rotational speed of the rotary device is at least 5 × 10 ⁹ times more preferred

2 10 ^'9 times the inverse coating rate expressed in m / sec.

18. Device for carrying out a component wear protection coating method according to one of claims 15 to 17, characterized in that i) the average distance transversely to the direction of rotation of the target plates (30) of the respective two physical vapor sources is smaller than the middle one Ii) the average distance transversely to the direction of rotation of the target plates (30) of the respective two physical vapor sources is smaller than the average distance of the component carriers (9) from the surface of the target plates (31 ) of the two physical vapor sources.

19. A device for carrying out a component wear protection coating method according to claim 15, characterized in that the mean distance transversely to the direction of rotation of the evaporation material plates (30) of the respective two physical vapor sources is less than 150 mm.

Device for carrying out a component wear-protection coating method according to claim 15, characterized in that the time required for a component carrier to pass from the middle of one of the two physical evaporation devices to the center of the other physical evaporation devices is less than 6 seconds, preferably less than 1 second.

21. Device for carrying out a component wear protection coating method according to claim 15, characterized in that the respective two physical evaporation devices are cathodic arc evaporators.

22. A device for carrying out a component wear protection coating method according to claim 21, characterized in that i) the cathodic arc evaporator for carbon is a magnetic

Ii) the magnetic cathode foot point guide is designed so that the

Kathodefusspunkt performs an oscillatory movement with a time period δt transverse to the rotational movement (21), iii) the time period δt of the oscillatory motion of the cathode footpoint is at least 10 to 100 times the time period Δt in which the component carriers have again reached the same position.