Hard Material Layer

Abstract

A hard material layer is deposited on a workpiece as a functional layer by an arc-PVD method. The layer is essentially an electrically insulating oxide of at least one of the metals (Me) of the transition metals of the sub-groups IV, V, VI of the periodic table and Al, Si, Fe, Co, Ni, Co, or Y and the functional layer (32) contains no noble gas or halogen.

Description

FIELD AND BACKGROUND OF THE INVENTION

[0001] The invention relates to a hard material layer deposited as oxidic arc PVD functional layer (32) on a workpiece (30) according to the preamble of claim 1 as well as to a method for coating a workpiece with a hard material layer according to the preamble of claim 21.

[0002] The operation of arc evaporator sources, also known as spark cathodes, by feeding with electrical pulses has been known in prior art for a relatively long time. With arc evaporator sources high evaporation rates, and therewith high deposition rates, can be achieved economically in coating. In addition, the structure of such a source can technically be realized relatively simply. These sources operate at currents typically in the range of approximately 100 A and more and at voltages of a few volts to a few tens of volts, which can be realized with relatively cost-effective DC power supplies. A significant disadvantage with these sources comprises that in the proximity of the cathode spot very rapidly proceeding melting occurs on the target surface, whereby drops are formed, so-called droplets, which are hurled away as splatters and subsequently condense on the workpiece and consequently have an undesirable effect on the layer properties. For example, thereby the layer structure becomes inhomogeneous and the surface roughness becomes inferior. With high requirements made of the layer quality, layers generated thusly, can often not be commercially applied. Attempts have therefore already been made to reduce these problems by operating the arc evaporator source in pure pulse operation of the power supply. However, until now only marginal improvements in the splatter formation could be achieved therewith.

[0003] The use of reactive gases for the deposition of compounds from a metallic target in a reactive plasma was until now limited to the production only of electrically conductive layers. In the production of electrically nonconducting, thus dielectric layers, such as for example of oxides using oxygen as the reactive gas, the problem of splatter formation is intensified. The re-coating of the target surfaces of the arc evaporator and of the counterelectrodes, such as the anodes and also other parts of the vacuum process installation, with a non-conducting layer leads to entirely unstable conditions and even to the quenching of the arc. In this case the latter would have to be repeatedly newly ignited or it would thereby become entirely impossible to conduct the process.

[0004] EP 0 666 335 B1 proposes for the deposition of purely metallic materials with an arc evaporator to superimpose onto the DC current a pulsing current in order to be able to lower hereby the DC base current for the reduction of the splatter formation. Pulse currents up to 5000 A are herein necessary, which are to be generated with capacitor discharges at relatively low pulse frequencies in the range of 100 Hz to 50 kHz. This approach is proposed to prevent the droplet formation in the non-reactive evaporation of purely metallic targets with an arc evaporator source. A solution for the deposition of non-conducting dielectric layers is not stated in this document.

[0005] In the reactive coating by means of arc evaporator source there is a lack of reactivity and process stability, especially in the production of insulating layers. In contrast to other PVD processes (for example sputtering), insulating layers can currently only be produced by means of arc evaporation with electrically conducting targets. Working with high frequency, such as is the case during sputtering, has so far failed due to the lacking technique of being able to operate high-power supplies with high frequencies. Working with pulsed power supplies appears to be an option. However, in this case the spark, as stated, must be ignited repeatedly or the pulse frequency must be selected so large that the spark is not extinguished. This appears to function to some degree in applications for special materials, such as graphite, as described in DE 3901401. It should, however, be noted that graphite is not an insulator, but rather is electrically conductive, even if it is a poorer conductor than normal metals.

[0006] In oxidized target surfaces a renewed igniting with mechanical contact and by means of DC supplies is not possible. The actual problem in reactive arc evaporation are the coatings with insulating layers on the target and the anode, or on the coating chamber connected as the anode. In the course of their formation, these insulating coatings increase the burn voltage of the spark discharge, lead to increased splatters and sparkovers, an unstable process, which ends in an interruption of the spark discharge. Entailed therein is a coating of the target with island growth, which decreases the conducting surface. A highly diluted reactive gas (for example argon/oxygen mixture) can delay the accretion on the target, however it cannot solve the fundamental problem of process instability. While the proposal according to U.S. Pat. No. 5,103,766 of alternately operating the cathode and the anode with renewed ignition each time results in process stability, it does however lead to increased splatters.

[0007] The resolution via a pulsed power supply as is possible for example in reactive sputtering, cannot be applied in classic spark evaporation. The reason lies therein that a glow discharge has a "longer life" than a spark when the power supply is interrupted. In order to circumvent the problem of the coating of the target with an insulating layer, in reactive processes for the production of insulating layers either the reactive gas inlet is locally separated from the target (in that case the reactivity of the process is only ensured if the temperature on the substrate also permits an oxidation reaction) or a separation between splatters and ionized fraction is carried out (so-called filtered arc) and after the filtering the reactive gas is added to the ionized vapor.

[0008] There is further the wish for additional reduction or scaling capability of the thermal loading of the substrates and the ability to conduct low-temperature processes in cathodic spark coating.

[0009] In WO 03018862 the pulse operation of plasma sources is described as a feasible path for reducing the thermal loading on the substrate. However, the reasons offered there apply to the field of sputter processes. No reference is established to spark evaporation.

[0010] In the application field of hard material coatings there has in particular been for a long time the need to be able to produce oxidic hard materials with appropriate hardness, adhesive strength and under control according to the desired tribological properties. Herein aluminum oxides, in particular aluminum chromoxides, could play an important role. Prior art in the field of PVD (Physical Vapor Deposition) deals herein most often only with the production of gamma and alpha aluminum oxide. The method most frequently mentioned is dual magnetron sputtering, which in this application entails great disadvantages with respect to process reliability and costs. Japanese patents concentrate more on layer systems in connection with the tools and cite, for example, the arc ion plating process as the production method. There is the general wish to be able to deposit alpha aluminum oxide. However, in current PVD methods, substrate temperatures of approximately 700.degree. C. or more are required in order to obtain this structure. Some users elegantly attempt to avoid these high temperatures through nucleation layers (oxidation of TiAlN, Al--Cr--O system). However, this does not necessarily make the process less expensive and faster. Until now it also did not appear possible to be able to produce satisfactorily alpha aluminum oxide layers by means of arc evaporation.

[0011] With respect to prior art the following disadvantages are summarized, in particular regarding the production of oxidic layers with reactive process: [0012] 1. Stable conduction of the process is not possible for the deposition of insulating layers, if there is no spatial separation between arc evaporator cathode or anode of the arc discharge and the substrate region with reactive gas inlet.

[0013] 2. There is no fundamental solution of the problematic of splatters: conglomerates (splatters) are not fully through-reacted, wherein metallic components occur in the layer, increased roughness of the layer surface is generated and the constancy of the layer composition and the stoichiometry is disturbed. [0014] 3. Insufficient possibilities for realizing low-temperature processes, since insufficiently the thermal loading of the substrate is too high for the production of oxides with high-temperature phases. [0015] 4. The production of flat graduated intermediate layers for insulating layers has so far not been possible by means of arc evaporation.

[0016] In contrast to sputtering, coating by means of cathodic spark is substantially a evaporation process. It is supposed that in the transition between hot cathode spot and its margin parts are entrained which are not of atomic size. These conglomerates impinge as such onto the substrate and result in rough layers, and it has not been possible fully to react-through the splatters. Avoidance or fragmentation of these splatters was so far not successful, especially not for reactive coating processes. In these, on the cathode of the arc evaporator source, for example in oxygen atmosphere, additionally a thin oxide layer forms, which tends to increased splatter formation.

SUMMARY OF THE INVENTION

[0017] The present invention addresses the problem of eliminating the listed disadvantages of prior art. The problem addressed is in particular to deposit economically layers with better properties with at least one arc evaporator source, such that the reactivity in the process is increased through better ionization of the vaporized material, and of the reactive gas involved in the process is increased. In this reactive process the size and frequency of the splatters is to be significantly reduced, in particular in reactive processes for the production of insulating layers. Further, better process control is to be made possible, such as the control of the evaporation rates, increase of the layer quality, settability of the layer properties, improvement of homogeneity of the reaction, as well as the reduction of surface roughness of the deposited layer. These improvements are in particular also of importance in the production of graduated layers and/or alloys. The process stability in reactive processes for the production of insulating layers is to be generally increased.

[0018] In particular, an arc evaporation process is to be made possible which permits the economic deposition of oxidic hard material layers, aluminum oxide and/or aluminum chromoxide layers which preferably have substantially alpha and/or gamma structure. Moreover, a low-temperature process should be realized, preferably below 700.degree. C., also at high economy of process. Furthermore the expenditure for the device and in particular for the power supply for pulsed operation should be kept low. Said tasks may occur singly as well as also combined with one another, depending on the particular required application area.

[0019] The problem is solved according to the invention through a hard material layer applied with an arc evaporation PVD method according to claim 1 and by proceeding according to a method as claimed in claim 21 for the production of such a layer on a workpiece. The dependent claims define further advantageous embodiments.

[0020] The problem is solved according to the invention thereby that a hard material layer is deposited as arc PVD functional layer onto a workpiece, this layer substantially being formed as an electrically insulating oxide, comprised of at least one of the metals (Me) Al, Cr, Fe, Ni, Co, Zr, Mo, Y and the functional layer comprises a content of inert gases and/or halogens of less than 2%. The content of inert gases is preferably less than 0.1%, in particular less than 0.05% or even better is zero and/or the content of halogens is less than 0.5%, in particular less than 0.1%, or even better is zero. These gases should be incorporated into the layer to as small an extent as possible and the arc evaporation process should therefore exclusively take place with pure reactive gas or a pure reactive gas mixture without inert gas component, such as He, Ne, Ar, or halogen gases, such as F.sub.2, Cl.sub.2, Br.sub.2, J.sub.2, or halogen-containing compounds such as CF.sub.6 or the like.

[0021] The known CVD processes use halogen with which at undesirably high temperatures of approximately 1100.degree. C. a layer is deposited. Even under reactive process conditions, the known sputter processes are operated with a high proportion of inert gas, such as with argon. The content of such gases in the layer should be below said values or preferably be zero. The pulse arc evaporation process according to the invention also permits sufficing without such process gases.

[0022] The preceding patent application with the application number CH00518/05 shows essentially already an approach to a solution. A first solution is specified which is especially well suited for completely reacted target surfaces and has a marked reduction of splatter formation compared to DC-operated arc evaporator targets. This application proposes superimposing a high-current pulse onto the DC feed of an arc evaporator source with a pulsed power supply, as is shown schematically in FIG. 2. A further reduction of the splatters and their size at higher economy is attained through the approach according to the succeeding patent application CH 01289/05 which claims priority of CH 00518/05 and represents a further development. In this application a vacuum process installation for the surface working of workpieces with at least one arc evaporator source is provided comprising a first electrode connected to a DC power supply, a second electrode disposed separated from the arc evaporator source being provided and that the two electrodes are connected to a single pulsed power supply. Between the two electrodes, consequently an additional discharge gap is operated with only a single pulsed power supply which makes possible an especially high ionization of the involved materials at very good controllability of the process.

[0023] The second electrode can herein be a further arc evaporator source, a workpiece holder or the workpiece itself, whereby in this case the second electrode can also be implemented as an evaporation crucible forming the anode of a low-voltage arc evaporator.

[0024] An especially preferred embodiment comprises that both electrodes are the cathodes of one arc evaporator source each and that each of these arc evaporator sources by itself is connected directly to a DC power supply for the purpose of maintaining a holding current and wherein the two cathodes are connected to a single pulsed power supply such that the arcs, or the arc discharges, of the two sources are not extinguished in operation. In this configuration, consequently, only one pulsed power supply is required since this supply is interconnected directly between the two cathodes of the arc evaporators. Apart from the high degree of ionization and the good controllability of the process, high efficiency of the configuration also results. Between these two electrodes and the pulse discharge gap additionally generated thereby, compared to this discharge gap, a bipolar pulse forms electrically from negative and positive components, whereby the entire period duration of this fed AC voltage can be utilized for the process. In fact, no unused pulse pauses are generated and the negative as well as also the positive pulses without interruption contribute overall to the process. The deposition rate can thereby be additionally increased without having to employ additional expensive pulsed power supplies. This configuration with two arc evaporator sources is especially suited for the deposition of layers from a metallic target utilizing reactive gas. With this configuration it becomes even possible to omit entirely supporting inert gases, such as argon, and it is possible to work with pure reactive gas, even unexpectedly with pure oxygen. Through the high degree of ionization attainable therewith of the vaporized material as well as also of the reactive gas, such as for example oxygen, nonconducting layers with high quality are generated which nearly reach the quality of the bulk material. The process runs very stably and herein the splatter formation is, unexpectedly, also reduced or entirely avoided. However, said advantages can also be attained by using other sources as the second electrode, such as, for example, a bias electrode or a low-voltage arc evaporator crucible, although said advantageous effects are not attained to the same degree as in the implementation of the configuration with two arc evaporators.

[0025] The present application claims priority of the two cited preceding applications CH 00518/05 and 01289/05 which substantially disclose a first approach to a solution for the present problem formation of the deposition of electrically nonconducting oxidic layers. The invention introduced in the present patent application represents a further development regarding the conduction of the process and the application. These two applications are consequently an integrating component of the present application.

[0026] The various features of novelty which characterize the invention are pointed out with particularity in the claims annexed to and forming a part of this disclosure and are entirely based on the Switzerland priority application no. 518/05, filed Mar. 24, 2005, and Switzerland priority application no. 1289/05, filed Aug. 3, 2005.

BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In the following the invention will be described in further detail by example and schematically with Figures. Therein depict:

[0028] FIG. 1 schematically an illustration of an arc evaporator coating installation, such as corresponds to prior art,

[0029] FIG. 2 a first configuration according to the invention with a DC-fed arc evaporator source in operation with superimposed high-current pulse,

[0030] FIG. 3 a second configuration with two DC-fed arc evaporator sources and high-power pulsed supply connected between them according to the invention, a dual pulse arc evaporator configuration,

[0031] FIG. 4 a cross section through a deposited layer as a multilayer according to the invention,

[0032] FIG. 5 an enlarged cross section of the layer according to FIG. 4.

[0033] FIG. 1 shows a vacuum process installation which depicts a configuration known from prior art for operating an arc evaporator source 5 with a DC power supply 13. The installation 1 is equipped with a pump system 2 for setting up the required vacuum in the chamber of the vacuum process installation 1. The pump system 2 permits the operation of the coating installation at pressures<10.sup.-1 mbar and also ensures the operation with the typical reactive gases, such as O.sub.2, N.sub.2, SiH.sub.4, hydrocarbons, etc. The reactive gases are introduced via a gas inlet 11 into the chamber 1 and here distributed accordingly. It is additionally possible to introduce additional reactive gases through further gas inlets or also inert gases, such as argon, as is necessary, for example, for etching processes or for the deposition of nonreactive layers in order to use the gases singly and/or in mixtures. The workpiece holder 3 located in the installation serves for receiving and for electrical contacting of the workpiece, not shown here, which are conventionally fabricated of metallic materials, and for the deposition of hard material layers using such processes. A bias power supply 4 is electrically connected with the workpiece holder 3 for applying a substrate voltage or a bias voltage to the workpieces. The bias power supply 4 can be a DC, an AC or a bipolar or a unipolar pulse substrate power supply. Via a process gas inlet (11) an inert or a reactive gas can be introduced in order to set and to control process pressure and gas composition in the treatment chamber.

[0034] Component parts of the arc evaporator source 5 are a target 5' with cooling plate placed behind it, and an ignition finger 7, which is disposed in the peripheral region of the target surface, as well as an anode encompassing the target. A switch 14 permits selecting between a floating operation of the anode 6 of the positive pole of the power supply 13 and operation with defined zero or ground potential. When igniting the arc of the arc evaporator source 5 a brief contact is established of the ignition finger 7 with the cathode and the former is subsequently withdrawn whereby a spark is ignited. The ignition finger 7 is for this purpose connected via a current limiter resistor to anode potential.

[0035] The vacuum process installation 1 can additionally optionally, should the conduction of the process require such, be equipped with an additional plasma source 9. In this case the plasma source 9 is implemented as a source for generating a low-voltage arc with a hot cathode. The hot cathode is, for example, formed as a filament disposed in a small ionization chamber, in which with a gas inlet 8 a working gas, such as for example argon, is introduced for the generation of a low-voltage arc discharge which extends into the main chamber of the vacuum process installation 1. An anode 15 for developing the low-voltage arc discharge is located at an appropriate position in the chamber of the vacuum process installation 1 and is operated, in known manner, with a DC power supply between cathode and plasma source 9 and anode 15. If required, additional coils 10, 10' can be provided, such as for example Helmholtz-like configurations which are placed about the vacuum process installation 1 for the magnetic focusing or guiding of the low-voltage arc plasma.

[0036] Proceeding according to the invention, as depicted in FIG. 2, the arc evaporator source 5 is operated being fed additionally with a pulsed high-power supply 16'. This pulsed power supply 16' is advantageously directly superimposed onto the DC power supply. It is understood that for their protection the two supplies must be operated electrically decoupled with respect to each other. This can be carried out in conventional manner with filters, such as with inductors, such as is familiar to a person of skill in the art. With this configuration it is already possible according to the invention to deposit layers exclusively with pure reactive gas or reactive gas mixtures, such as oxides, nitrides, etc., without undesirable support gas components, such as for example argon in PVD sputter processes or halogens of the precursors in CVD processes. It is, in particular, possible to generate therewith the pure, electrically nonconducting oxides, which are very difficult to obtain economically, in the desired crystalline form and to deposit them as layers. This reactive pulsed arc evaporation method is herewith denoted as RPAE method.

[0037] In a further improved and preferred embodiment of a vacuum process configuration, apart from a first arc evaporator source 5, a second arc evaporator source 20 is provided with the second target electrode 20', as is shown in FIG. 3. Both arc evaporator sources 5, 20 are operated with one DC power supply 13 and 13' each, such that the DC power supplies ensure with a base current the maintenance of the arc discharge. The DC power supplies 13, 13' correspond to prior art and can be realized cost-effectively. The two electrodes 5', 20', which form the cathode of the two arc evaporator sources 5, 20, are connected according to the present invention to a single pulsed power supply 16, which is capable of outputting to the two electrodes 5', 20' high pulse currents with defined form and edge slope of the pulses. In the depicted configuration according to FIG. 3 the anodes 6 of the two arc evaporator sources 5, 20 are referred to the electrical potential of the ground of the process installation 1. This is herewith also denoted as Dual Pulsed Arc Evaporation (DPAE).

[0038] It is possible to operate the spark discharges with reference to ground or also floatingly. In the preferred case of floating operation, the first DC power supply 13 is connected with its negative pole to the cathode 5' of the first arc evaporator source 5 and its positive pole with the opposing anode of the second arc evaporator source 20. The second arc evaporator source 20 is operated analogously and the second power supply 13' is connected to the positive pole of the anode of the first arc evaporator source 5. This opposing operation of the anodes of the arc evaporator sources leads to better ionization of the materials in the process. However, the ground-free operation, or the floating operation, of the arc evaporator source 5, 20 can also take place without using the opposing anode feed. In addition, it is possible to provide a switch 14, as shown in FIG. 1, in order to be able to change over optionally between floating and ground-tied operation.

[0039] The supply for this "Dual Pulsed Mode" must be able to cover different impedance ranges and yet not be "hard" in the voltage. This means that the supply must supply high current, yet, in spite of it, be largely operable voltage-stably. An application of an example of such a supply was filed under the No. CH 518/05 parallel with the same date as said patent application No. CH 1289/05.

[0040] The first and preferred application field of this invention is that of cathodic spark evaporation with two pulsed arc evaporator sources (5, 20) as is depicted in FIG. 3. For these applications the impedances are at intervals of approximately 0.01 .OMEGA. to 1 .OMEGA.. It should be noted here that usually the impedances of the sources, between which "dual pulsing" is carried out are different. The reason may be that these are comprised of different materials or alloys, that the magnetic field of the sources is different or that the material erosion of the sources is at a different state. The "Dual Pulsed Mode" now permits a balance via the setting of the pulse width such that both sources draw the same current. This leads consequently to different voltages at the sources. The supply can, of course, also be loaded asymmetrically with respect to the current if such appears desirable for the process conduction, which is the case, for example, for graduated layers of different materials. The voltage stability of a supply is increasingly more difficult to realize the lower the impedance of the particular plasma. The capability of change-over switching or the controlled active tracking of a supply to different output impedances is of therefore of special advantage if the full range of its power is to be utilized, thus for example in the range of 500 V/100 A to 50 V/1000 A or as it is realized in the parallel application No. CH 518/05.

[0041] The advantages of such dual pulsed cathode configuration and in particular one comprised of two arc evaporator sources are summarized as follows: [0042] 1. Increased electron emission at steep pulses results in higher current (also substrate current) and increased ionization of the vaporized material and of the reactive gas. [0043] 2. The increased electron density contributes also to a fast discharge of the substrate surface in the production of insulating layers, i.e. relatively short charge-reversal times on the substrate (or also only pulse pauses of the bias voltage) are sufficient in order to discharge the insulating layer which is forming. [0044] 3. The bipolar operation between the two cathodic arc evaporator sources permits a quasi-100% pulse pause ratio (duty cycle), while the pulsing of a source along always necessarily requires a pause and therefore the efficiency is not so high. [0045] 4. The dual pulsed operation of two cathode spark sources, which are opposite to one another, immerses the substrate region into a dense plasma and increases the reactivity in this region even of the reactive gas. This is also reflected in the increase of the substrate current. [0046] 5. In reactive processes under oxygen atmosphere in pulsed operation still higher electron emission values can be attained, and it appears that a melting of the spark region, as is the case in classic evaporation of metallic targets, can be largely avoided. Working in purely oxidic reactive mode without further foreign or support gases is now readily possible. To be able to attain said advantageous process properties in said different possible embodiments of the invention, the pulsed power supply 16, 16' must satisfy different conditions. In bipolar pulse presentation it should be possible to carry out the process at a frequency which is in the range of 10 Hz to 500 kHz. Due to the ionization conditions, herein the maintainable edge slopes of the pulses is important. The magnitudes of the leading edges U2/(t2-t1), U1/(t6-t5), as well as also of the trailing edges U2/(t4-t3) and U1/(t8-t7) should have a slope in the range of 0.02 V/ns to 2 V/ns and this at least in open-circuit operation, thus without load, however preferably also under load. It is understood that the edge slope has an effect in operation, depending on the corresponding magnitude of the load or the connected impedance of the corresponding settings. The pulse widths in bipolar presentation for t4 to t1 and t8 to t5 are advantageously >1 .mu.S, the pauses t5 to t4 and t9 to t8 can advantageously be essentially 0, however, under certain conditions, they can also be .gtoreq.0 .mu.s. If the pulse pauses are >0, this operation is referred to as time-gapped and through, for example, variable time shift of the pulse gap widths the specific and purposeful introduction of energy into a plasma and its stabilization can be set. It is especially advantageous if the pulsed power supply is laid out such that a pulse option up to 500 A at 1000 V voltage is possible, wherein herein the pulse/pause ratio (duty cycle) must be appropriately taken into consideration or must be adapted for the laid out possible power of the supply. Apart from the edge slope of the pulse voltage it is necessary to observe that the pulsed power supply (16) is capable of handling a current rise to 500 A in at least 1 .mu.s.

[0047] With the operation introduced here of arc evaporator sources with DC feed and superimposed high-current pulsed feed (RPAE, DPAE) it is possible to deposit with high quality starting from one or several metal targets with reactive gas atmosphere corresponding metal compounds onto a workpiece 30. This is in particular suited for the generation of purely oxidic layers, since the method does not require additional support gases, such as inert gases, customarily argon. The plasma discharge of the arc evaporator 5, 20 can thus, for example and preferably, take place in pure oxygen atmosphere at desired working pressure without the discharge being unstable, is prevented or yields unusable results, as too high a splatter formation or poor layer properties. It is also not necessary to use, as is the case in CVD methods, halogen compounds. This permits, first, to produce economically wear-resistant oxidic hard material layers of high quality at low process temperatures, preferably below 500.degree. C., which, as a result, are nevertheless high temperature-resistant, preferably >800.degree. C. and which are chemically highly stable, such as, for example, have high resistance to oxidation. Furthermore, to attain a stable layer system the diffusion of oxygen with the oxidation entailed therein in the deeper layer system and/or on the workpiece should as much as possible be avoided.

[0048] It is now readily possible to produce oxidic layers in pure oxygen as reactive gas from the transition metals of the subgroups IV, V, VI of the periodic system of elements and Al, Si, Fe, Co, Ni, Y, with Al, Cr, Mo, Zr as well as Fe, Co, Ni, Y being preferred. The functional layer 32 is to contain as the oxide one or several of these metals, no inert gas and/or halogen, such as Cl, however at least less than 0.1% or better less than 0.05% inert gas and less than 0.5% or better less than 0.1% halogen in order to attain the desired layer quality.

[0049] Such functional layers 32 or multiple layer system 33 (multilayer) should, in particular, as hard material layer have a thickness in the range of 0.5 to 12 .mu.m, preferably from 1.0 to 5.0 .mu.m. The functional layer can be deposited directly onto the workpiece 30 which is a tool, a machine part, preferably a cutting tool, such as an indexable insert. Between this layer and the workpiece 30 at least one further layer or a layer system can also be deposited, in particular for the formation of an intermediate layer 31, which forms in particular an adhesion layer and comprises preferably one of the metals of the subgroups lVa, Va and Vla of the periodic system of elements and/or Al or Si or a mixture of these. Good adhesive properties are achieved with compounds of these metals with N, C, O, B or mixtures thereof, the compound comprising N being preferred.

[0050] The layer thickness of the intermediate layer 31 should be in the range of 0.05 to 5 .mu.m, preferably 0.1 to 0.5 .mu.m. At least one of the functional layers 32 and/or of the intermediate layer 31 can advantageously be implemented as a progression layer 34, whereby a better transition of the properties of the particular layers is brought about. The progression can be from metallic over nitridic to nitrooxidic and up to the pure oxide. Thus a progression region 34 is formed where the materials of the abutting layers, or, if no intermediate layer is present, the workpiece material, are mixed into one another.

[0051] On the functional layer 32 a further layer or a layer system 35 can be deposited as cover layer, should this be required. A cover layer 35 can be deposited as additional friction-reducing layer for further improvement of the tribological behavior of the coated workpiece 30.

[0052] Depending on the requirements, one or more layers of said layers or layer systems can be developed as progression layers 34 in the region where they border on one another or within individual layers concentration gradients of any type can be generated. In the present invention this is simply possible through the controlled introduction of the reactive gases into the vacuum process installation 1 for setting the particular types of gas necessary for this purpose and of the gas quantities for the reactive arc plasma process.

[0053] As functional layer 32 with the desired hard material properties, now aluminum oxide layers (Al.sub.2O.sub.3), layers can now readily be produced which even have substantially stoichiometric composition. Especially advantageous hard material layers as functional layer 32 are substantially comprised of an (Al.sub.xMe.sub.1-x).sub.yO.sub.z, where Me is preferably one of the metals Cr, Fe, Ni, Co, Zr, Mo, Y singly or also in mixtures, settable depending on the desired proportions x, y and z of the involved substances. Further is especially preferred chromium as the metal Me in the metal mixed oxide of the (Al.sub.xMe.sub.1-x).sub.yO.sub.z which consequently forms (Al.sub.xCr.sub.1-x).sub.yO.sub.z or (AlCr).sub.yO.sub.z. Herein the proportion 1-x of the metal chromium in the layer should be 5 to 80 atom %, preferably 10 to 60 atom %. Well suited as hard material functional layer 32 is also a metal nitride, in particular the aluminum chromium nitride (AlCr).sub.yN.sub.z or at most also (AlTi).sub.yN.sub.z.

[0054] Through the intentional capability of process conduction it is now also possible in the case of aluminum and aluminum chromoxides to be able to attain the especially desired alpha and/or gamma structure.

[0055] Due to said simple settability of the layer conditions with their composition via the control of the supply of the reactive gases and due to the stable process condition, it is for the first time possible to produce multilayer systems (multilayer) 33 with any number of layers and any composition and even with progressions. Several layers can herein be generated of different materials or, and this appears often to be of advantage, with the alternating identical materials as a type of sandwich. For functional hard material layers 32, a layer system with repeated layer sequence pairs 33, in which the material composition changes periodically, is advantageous. Especially a structure from Me.sub.1 to an Me.sub.2-oxide and/or from an Me.sub.1-nitride to an Me.sub.1-oxide and/or from an Me.sub.1-nitride to an Me.sub.2-oxide yields excellent results with respect to endurance and less fissuring of the functional layer or of this layer system. An example of a functional layer 32 as a multilayer 33 is shown in FIG. 4 and in enlarged cross section in FIG. 5. Shown is a preferred material pairing of alternating aluminum chromium nitride (AlCr).sub.xN.sub.y with aluminum chromoxide (AlCr).sub.xO.sub.y produced with the method according to the invention, preferably in stoichiometric material composition. The layer packet in this example comprises 42 layer pairs with alternating materials, as stated above. The entire layer thickness of this functional layer 32 as multilayer system 33 is approximately 4.1 .mu.m, the thickness of a layer pair, thus two deposits, being 98 nm. Further preferred material pairings are alternating aluminum zirconium nitride (AlZr).sub.xN.sub.y with aluminum zirconium oxide (AlZr).sub.xO.sub.y produced with the method according to the invention, preferably in stoichiometric material composition. For hard material layers as functional layer 32 it is of advantage if the multilayer system 33 includes at least 20 deposits, preferably up to 500 deposits. The thickness per deposit should be in the range from 0.01 to 0.5 .mu.m, preferably in the range from 0.2 to 0.1 .mu.m. In the region of the individual bordering deposits of the layers progressions 34 are also evident, which ensure for good behavior of the transitions. In the example according to FIG. 4 as an example a cover layer 35 is also deposit as a friction-reducing layer over the functional layer 32, 33. The cover layer is comprised of titanium nitride and is approximately 0.83 .mu.m thick. Under the functional layer as an example additionally an intermediate layer 31 is disposed as adhesion layer which is approximately 1.31 .mu.m thick and has been deposited as an Al--Cr--N intermediate layer with RPAE onto the workpiece 30.

[0056] The coatings introduced here, whether single layer or multilayer system should preferably have an R, value of not less than 2 .mu.m and/or an R.sub.a value of not less than 0.2 .mu.m. These values are in each instance measured directly on the surface before a potential after-treatment of the surface, such as brushing, blasting, polishing, etc. Thus, the values represent a purely process-dependent surface roughness. By R.sub.a value is understood the mean rough value according to DIN 4768. This is the arithmetic mean of all deviations of the roughness profile R from the center line within the total measuring path I.sub.m. By R.sub.z is understood the mean roughness depth according to DIN 4768. This is the mean value of the individual roughness depths of five successive individual measuring paths le in the roughness profile. R.sub.z depends only on the distance of the highest peaks to the deepest valleys. By forming the mean value the effect of an individual peak (valley) is reduced and the mean width of the band, in which the R profile is included, is calculated. The introduced coating according to the invention is especially suited for workpieces such as cutting, forming, injection molding or punching and stamping tools, however, very specifically for indexable inserts.

[0057] In the following a typical sequence of a substrate treatment in a reactive pulse arc evaporation coating process is described using the present invention. Apart from the coating process proper, in which the invention is realized, the other process steps will also be described, which involve the pretreatment and posttreatment of the workpieces. All of these steps allow wide variations, some can also be omitted under certain conditions, shortened or extended or be combined differently. In a first step the workpieces are customarily subjected to wet-chemical cleaning, which, depending on the material and prior history, is carried out in different manner.

Example 1

[0058] Description of a typical process sequence for the production of an Al--Cr--O layer 32 (as well as of an Al--Cr--N/Al--Cr--O multilayer 33) and Al--Cr--N intermediate layer 31 by means of RPAE (reactive pulse arc evaporation) for coating workpieces 30, such as cutting tools, preferably indexable inserts.

[0059] 1. Pretreatment (cleaning, etc.) of the workpieces (30) (substrates) as known to the person of skill in the art.

[0060] 2. Placing the substrates into the holders intended for this purpose and transfer into the coating system.

[0061] 3. Pumping the coating chamber 1 to a pressure of approximately 10.sup.-4 mbar by means of a pump system as known to the person of skill in the art (forepumps/diffusion pump, forepumps/turbomolecular pump, final pressure approximately 10.sup.-7 mbar attainable).

[0062] 4. Starting the substrate pretreatment in vacuo with a heating step in an argon-hydrogen plasma or another known plasma treatment. Without restrictions, this pretreatment can be carried out with the following parameters: Plasma of a low-voltage arc discharge with approximately 100 A discharge current, up to 200 A, to 400 A, the substrates are preferably connected as anode for this low-voltage arc discharge: [0063] Argon flow 50 sccm [0064] Hydrogen flow 300 sccm [0065] Substrate temperature 500.degree. C. (partially through plasma heating, partially through radiative heating) [0066] Process time 45 min [0067] It is preferred that during this step a supply is connected between substrate 30 and ground or another reference potential, which can act on the substrates with DC (preferably positive) or DC pulsed (unipolar, bipolar) or as IF (intermediate frequency) or RF (high frequency).

[0068] 5. As the next process step etching is started. For this purpose the low-voltage arc is operated between the filament and the auxiliary anode. A DC, pulsed DC, IF or RF supply is connected between substrates and ground and the substrates are preferably acted upon with negative voltage. In the pulsed and IF, RF supplies positive voltage is also impressed on the substrates. The supplies 4 can be operated unipolarly or bipolarly. The typical, however not exclusive, process parameters during this step are: [0069] Argon flow 60 sccm [0070] Discharge current low-voltage arc 150 A [0071] Substrate temperature 500.degree. C. (partially through plasma heating, partially through radiative heating) [0072] Process time 30 min [0073] To ensure the stability of the low-voltage arc discharge during the production of insulating layers, the work is either carried out with a hot, conductive auxiliary anode 15, or a pulsed high-power supply is connected between auxiliary anode and ground.

[0074] 6. Start of coating with the intermediate layer 31 (approximately 15 min) CrN intermediate layer 300 nm by means of spark evaporation (source current 140 A, Ar 80 sccm, N2 1200 sccm, with bias of -80 V or of -100 V down to -60 V or 40 V, respectively. [0075] The coating can take place with and without low-voltage arc.

[0076] 7. Transition to the functional layer 32 (approximately 5 min) In the transition to the functional layer proper, onto the spark sources are additionally superposed unipolar DC pulses of a second power supply connected in parallel, which can be operated with 50 kHz (FIG. 2). An Al target is additionally operated in the same manner in order to produce AlCr as a layer. In the example work took place with 10 ps pulse/10 ps pause and in the pulsed currents up to 150 A generated. Oxygen at 200 sccm was subsequently let in.

[0077] 8. Driving back of the AlCrN coating After the oxygen gas flow has been stabilized, the AlCrN coating is brought down. For this purpose the N2 gas flow is reduced. This ramp takes place over approximately 10 min. The Ar flow is subsequently reduced to zero (unless work is carried out with low-voltage arc).

[0078] 9. Coating with functional layer 32 The coating of the substrates with the functional layer proper takes place in pure reactive gas (in this case oxygen). The most important process parameters are: [0079] Oxygen flow 400 sccm [0080] Substrate temperature 500.degree. C. [0081] DC source current 60 A [0082] Onto the DC source current a pulsed DC current (unipolar) of 150 A is superimposed with a pulse frequency of 50 kHz and a pulse characteristic of 10 ps pulse/10 .mu.s pause. Process pressure in the coating chamber 9.times.10.sup.-2 mbar. The bias at the substrates is reduced to -40 V. Since aluminum oxide layers are insulating layers, a bias supply is utilized, which is operated either DC pulsed or as IF (50 kHz-350 kHz). [0083] The coating can also be carried out simultaneously with the low-voltage arc. In this case a higher reactivity is attained. The simultaneous use of the low-voltage arc during the coating has furthermore the advantage that the DC component in the sources can be reduced. At higher arc current, it can be further reduced. [0084] The coating process conducted in this way is stable even over several hours. The target 5, 5' is covered with a thin smooth oxide layer. However, no insulating islands are formed, although the target surface changes through the oxygen, which is also reflected in the increase of the burn voltage. The target surface remains significantly smoother. The spark runs quieter and divides into several smaller sparks. The number of splatters is significantly reduced.

[0085] The described process is a fundamental preferred version since it keeps the requirements made of the pulsed power supply low. The DC supply supplies the minimum or holding current for the spark and the pulsed high-power supply 16, 16' serves for avoiding the splatters and ensures the process. One feasibility of generating multilayer systems 33, thus multiple layers 33, for the above layer example comprises that the oxygen flow during the layer deposition is decreased or even switched off entirely, while the nitrogen flow is added. This can take place periodically as well as aperiodically, with layers of exclusive or mixed oxygen-nitrogen concentration. In this way multilayers 33 are produced such as are shown in FIG. 4, and enlarged in FIG. 5, by example in cross section. In many cases this functional layer 32 forms the termination of the coating to the outside, without a further layer following thereon.

[0086] Depending on the application and requirement, wear properties can be "topped" with one or several cover layers 35. The example of the AlCrN/AlCrO multilayer already described above with a TiN top layer is also shown in FIG. 4. The at least one cover layer 35 can in this case be, for example, a friction-reducing layer, wherein in this case the hard material layer 32, or the functional layer or the multiple layer serves as support layer for the friction-reducing layer 35.

[0087] If there is the wish to produce multilayer functional layers 33 or multilayer intermediate layers with especially thin oxide-containing layer thickness, in a preferred process variant this can also take place thereby that the operation of the oxide-forming target under oxygen flow takes place just until the target exhibits first poisoning signs (voltage rise, most often after a few minutes) and then switching again to, for example, nitrogen flow. The process variant is especially simple and can be realized with the existing prior art (FIG. 1) thus without target pulse operation. However, this does not permit a free adaptation of the layer thickness to the particular requirements.

[0088] The implementation of said example in dual pulsed operation with two or more arc evaporator sources yields, in addition, advantages with respect to the conduction of the process and economy.

Example 2

[0089] Coating of workpieces 30, such as cutting tools, preferably indexable inserts, with an Al--Cr--O hard material layer system 32 and Cr--N intermediate layer 31 by means of DPAE (Dual Pulsed Arc Evaporator)

[0090] Steps 1 to and including 5 analogous to Example 1.

[0091] 6. Starting the coating with the intermediate layer (approximately 15 min) AlCrN intermediate layer 300 nm by means of spark evaporation (target material AlCr (50%, 50%), source current 180 A, N2 800 sccm, with bipolar bias of -180 V (36 .mu.s negative, 4 .mu.s positive). [0092] The coating can take place with and without low-voltage arc. [0093] Up to this point the method follows prior art such as is shown for example in FIG. 1.

[0094] 7. Transition to functional layer 32 (approximately 5 min) In the transition to the functional layer 32 proper, the nitrogen is ramped down from 800 sccm to approximately 600 sccm and subsequently an oxygen flow of 400 sccm is switched on. The nitrogen flow is now switched off.

[0095] 8. Coating with the functional layer 32

[0096] The bipolar pulsed high-power supply 16, as shown in FIG. 3, between both arc evaporator cathodes 5, 20 is now taken into operation. In the described process work took place with a positive or negative time mean value of the current of approximately 50 A. The pulse durations were each 10 .mu.s for the positive as well as negative voltage range with 10 ps pauses each in between at a voltage of 160 V. The peak value of the current through the bipolar pulsed power supply 16 depends on the particular pulse form. The difference of DC current through the particular arc evaporator cathode 5, 20 and peak value of the bipolarly pulsed current must not fall below the so-called holding current of the arc evaporator cathode 5, 20, since otherwise the arc (spark) is extinguished. [0097] During the first 10 minutes of the coating the bias is ramped from -180 V to '60 V. The typical coating rates for double rotating workpieces 30 are between 3 .mu.m/hr and 6 .mu.m/hr. [0098] The coating of the workpieces 30 with the functional layer 32 proper thus takes place in pure reactive gas (in this example oxygen). The most important process parameters are once again summarized: [0099] Oxygen flow 400 sccm [0100] Workpiece temperature 500.degree. C. [0101] DC source current 180 A, for the Al as well as also for the Cr source. [0102] The bipolarly pulsed DC current between the two cathodes has a frequency of 25 kHz. [0103] Process pressure approximately 9.times.10.sup.-3 mbar.

[0104] As already stated, the coating can also take place simultaneously with the operation of the low-voltage arc. In this case a further increase of the reactivity especially in the proximity of the workpiece is attained. In addition, the simultaneous utilization of the low-voltage arc during the coating has also the advantage that the DC component at the sources can be reduced. With higher arc current, this can be further reduced. The coating process conducted in this way is stable even over several hours. Targets 5', 20' of the arc evaporators 5, 20 are covered with thin, smooth oxide layer. This is desirable and is also the precondition for a largely splatter-free and stable process. The covering is manifested in an increase of the voltage at the target.

[0105] Workpieces were coated with different coatings and under the same conditions subjected to a practical comparison test.

[0106] Test conditions for the rotation tests: As the measure for these tests known TiAlN layers and known alpha aluminum oxide layers deposited by means of CVD are used. In all test layers a layer thickness of 4 .mu.m was tested. As test material were used stainless steel (1.1192). As rotation cycles were selected 1, 2 and 4 min each. The cutting rate was 350 m/min, advance 0.3 mm/rev. Engagement depth 2 mm. The conditions were selected such that short test times are attainable at high temperatures on the cutting edge of the workpiece. The wear on the end flank and the chipping edge as well as the surface roughness of the worked steel were tested, and the length of time was determined before a certain increased roughness occurred. As the quantitative measure for wear, this service time was determined.

[0107] Results: [0108] a) CVD layer alpha aluminum oxide (prior art) [0109] layer thickness d=4 .mu.m [0110] The tool survived the 4-minute test. However, after the test in the SEM there was no longer any layer material on the chipping edge. [0111] b) TiAlN layer (prior art), d=4 .mu.m This layer showed already after less than 2 min initial signs of destructions and supplied a rough surface on the workpiece.

[0112] Invention: [0113] c) AlCrN intermediate layer, d=0.4 .mu.m [0114] AlCrN/AlCrO multilayer, d=3.6 .mu.m [0115] TiN top layer, d=0.8 .mu.m [0116] Endurance 4 min [0117] d) AlCrN intermediate layer, d=0.4 .mu.m [0118] AlCrN/AlCrO multilayer, d=3.6 .mu.m 3 min 40 s [0119] e) AlCrN intermediate layer, d=0.3 .mu.m [0120] AlCrO single layer, d=2.9 .mu.m [0121] TiN top layer, d=0.9 .mu.m 4 min [0122] f) AlCrN intermediate layer, d=0.35 .mu.m [0123] AlCrO single layer, d=3.5 .mu.m 3 min 20 s [0124] g) ZrN intermediate layer, d=0.3 .mu.m [0125] ZrN/AlCrO multilayer, d=3.8 .mu.m [0126] ZrN top layer, d=0.5 .mu.m 3 min 10 s [0127] h) ZrN intermediate layer, d=0.2 .mu.m [0128] ZrO/AlCrO multilayer, d=6.4 .mu.m [0129] ZrN top layer, d=0.8 .mu.m 4 min [0130] i) AlCrN intermediate layer, d=0.5 .mu.m [0131] AlCrO/alpha alumina multilayer, d=8.2 .mu.m 4 min [0132] k) (Ti, AlCrN) intermediate layer, d=0.4 .mu.m [0133] AlCrO/TiAlCrN multilayer, d=4.5 .mu.m 3 min 50 s

[0134] Layers of or multilayers comprising oxidic layers of the stated materials show markedly less wear at high cutting rates. Conducting layers (TiAlN) according to prior art at high cutting rates are markedly inferior to the oxide systems according to the invention. Systems according to the present invention of (AlCr).sub.yO.sub.z, and (AlZr).sub.yO.sub.z show similarly low wear as known CVD layers of .alpha.-aluminum oxide, however without its disadvantage of high temperature loading or loading through aggressive chemicals of the workpiece during the coating process. The conduction of the process, furthermore, can be carried out substantially simpler, for example through changing-over of gases or controlled change of the gas components (for example O.sub.2 to N.sub.2) and/or changing-over from one target, or changing of the components of the target feed under control, to the other, while in CVD processes intermediate flushing as well as adaptation of the temperature level for individual layers of a multilayer layer system are necessary.

Claims

1. Hard material layers deposited on a workpiece (30) as Arc-PVD functional layer (32) wherein this layer is formed substantially as an electrically insulating oxide of at least one of the metals (Me) from the transition metals of subgroups IV, V, VI of the periodic system of elements and Al, Si, Fe, Co, Ni, Y, characterized in that the functional layer (32) has a content of inert gas and/or of a halogen of less than 2%.

2. Hard material layer as claimed in claim 1, characterized in that the content of inert gas in the functional layer (32) is maximally 0.1%, preferably maximally 0.05%, and/or that of halogen maximally 0.5%, preferably maximally 0.1%, preferably comprises substantially no inert gas and/or halogen.

3. Hard material layer as claimed in claim 1, characterized in that the functional layer (32) has a thickness in the range from 0.5 .mu.m to 12 .mu.m, preferably 1.0 to 5 .mu.m.

4. Hard material layer as claimed in claim 1, characterized in that the functional layer (32) is substantially an aluminum metal mixed oxide of the form (Al.sub.xM.sub.1-x).sub.yO.sub.z wherein Me is preferably one of the metals Al, Cr, Mo, Zr, Fe, Co, Ni, Y, singly or also in mixtures thereof.

5. Hard material layer as claimed in claim 4, characterized in that Me is the metal chromium and forms the form (Al.sub.xCr.sub.1-x).sub.yO.sub.z.

6. Hard material layer as claimed in claim 5, characterized in that the fraction of the metal chromium in the layer is 5 to 80 atom %, preferably 10 to 60 atom %.

7. Hard material layer as claimed in claim 1, characterized in that the functional layer (32) is substantially a stoichiometric aluminum oxide layer of the form Al.sub.2O.sub.3.

8. Hard material layer as claimed in claim 1, characterized in that the functional layer (32) forms the outermost layer or an additional support layer, with at least one superjacent cover layer (35), such as in particular a friction-reducing layer (35).

9. Hard material layer as claimed in claim 1, characterized in that the functional layer (32) has a temperature resistance of greater than 800.degree. C. and that it is chemically stable.

10. Workpiece with a hard material layer as claimed in claim 1, characterized in that the workpiece (30) is a tool, a machine part, preferably an indexable insert.

11. Workpiece as claimed in claim 10, characterized in that between the functional layer (32) and the workpiece (30) a further layer forming an intermediate layer (31) is disposed, and this layer forms in particular an adhesion layer (31) and such adhesion layer preferably comprises one of the metals of the subgroups IV, V and VI of the periodic system of elements and/or Al, Si, Fe, Co, Ni, Y or a mixture thereof.

12. Workpiece as claimed in claim 11, characterized in that the metals of the intermediate layer (31) are compounds with N, C, O, B or mixtures thereof, the compound with N being preferred.

13. Workpiece as claimed in claim 11, characterized in that the layer thickness of the intermediate layer (31) is 0.05 to 5 .mu.m, preferably is in the range of 0.1 to 0.5 .mu.m.

14. Workpiece as claimed in claim 10, characterized in that at least one of the layers, in particular the functional layer (32) and or the intermediate layer (31) is implemented as a progression layer (34), such as from metallic over nitridic and/or from nitridic to nitrooxidic and up to the oxide.

15. Workpiece as claimed in claim 10, characterized in that at least one of the layers, in particular the functional layer (32), is implemented as a multilayer system (33) with different material composition, in which preferably several deposits (33) alternately repeat with respect to their essential composition and that the multilayer system (33) preferably comprises at least 3 deposits.

16. Workpiece as claimed in claim 15, characterized in that the repeating layer sequence pairs of the layer system alternately change the material composition, such as preferably from an Me.sub.1 to an Me.sub.2-oxide and/or from an Me.sub.1-nitride to an Me.sub.1-oxide and/or from an Me.sub.1-nitride to an Me.sub.2-oxide.

17. Workpiece as claimed in claim 15, characterized in that the repeating layer sequence pair of the layer system alternately comprises the material composition of (Al.sub.xCr.sub.1-x).sub.yN.sub.z and (Al.sub.xCr.sub.1-x).sub.yO.sub.z and these preferably in stoichiometric composition such as (Al.sub.xCr.sub.1-x)N and (Al.sub.xCr.sub.1-x).sub.zO.sub.3.

18. Workpiece as claimed in claim 15, characterized in that the repeating layer sequence pairs of the layer system comprise alternately the material composition (AlZr).sub.xN.sub.y and (AlZr).sub.xO.sub.y and these preferably in stoichiometric composition such as (Al.sub.xZr.sub.1-x)N and (Al.sub.xZr.sub.1-x).sub.2O.sub.3.

19. Workpiece as claimed in claim 15, characterized in that the multilayer system (33) comprises at least 20 deposits, preferably up to 500 deposits.

20. Workpiece as claimed in claim 15, characterized in that the layer thickness of one deposit of the multilayer system (33) is in the range of 0.01 to 0.5 .mu.m, preferably in the range of 0.02 to 0.1 .mu.m.

21. Method for coating a workpiece (3) in a vacuum process installation (1) with a hard material layer (32) deposited as functional layer, which is implemented as an electrically insulating oxide of at least one of the metals (Me) of the transition metals of the subgroups IV, V, VI of the periodic system of elements and Al, Si, Fe, Co, Ni, Co, Y, and that the layer is deposited with an arc evaporator source (5) operated with a DC power supply (13), characterized in that a pulsed power supply (16, 16') is superimposed, wherein the target (5', 20) of the arc evaporator source (5, 20) comprises one of the metals and the metal vapor-deposited in this way is reacted to the oxide in an oxygen-containing reactive gas atmosphere.

22. Method as claimed in claim 21, characterized in that in the reactive gas atmosphere of the process chamber of the vacuum installation (1) so small a quantity of inert gas and/or halogen gas is supplied that in the deposited layer maximally 0.5% of such gases, preferably substantially none of these gases, are incorporated.

23. Method as claimed in claim 21, characterized in that two DC-fed arc evaporator sources (5, 20) are operated, wherein additionally a single pulsed power supply (16) is connected to the two sources (5, 20) and in this manner forms a dual pulse arc evaporator configuration (5, 20).

24. Method as claimed in claim 21, characterized in that the workpiece is substantially comprised of steel, an iron-, chromium-, cobalt- or nickel-containing alloy of one or several metals, a hard metal, a ceramic, a cermet, or cubic boron mononitride, wherein at least one further layer is deposited by means of a PVD method and one of the layers is an adhesion layer (31) which borders directly on the workpiece (30), wherein the or at least one of the following layers, the functional layer (32), is substantially comprised of Al.sub.2O.sub.3 or (AlMe).sub.2O.sub.3, wherein Me comprises at least one transition metal of the group IV, V or VI of the periodic system of elements or silicon and at least the aluminum or aluminum metal oxide layer is deposited with an arc evaporator (5, 20), in which from at least one target (5', 20'), poisoned on the surface, aluminum oxide, metal oxide or aluminum metal oxide is vaporized in an oxygen-containing atmosphere.

25. Method as claimed in claim 21, characterized in that the coating attains a roughness value R.sub.a of not less than 0.2 .mu.m.

26. Method as claimed in claim 21, characterized in that at least one further layer is deposited which comprises substantially an aluminum-free one or several metal oxides comprising oxide layer, wherein the metal oxide comprises at least one transition metal of group IV, V or VI of the periodic system of elements or silicon, however preferably chromium or zirconium.

27. Method as claimed in claim 24, characterized in that the adhesion layer (31) comprises at least one of the transition metals of group IV, V or VI of the periodic system of elements and/or aluminum or silicon.

28. Method as claimed in claim 24, characterized in that the adhesion layer (31) comprises a hard layer which comprises a nitride, carbide or boride, at least one of the transition metals of group IV, V or VI of the periodic system of elements and/or aluminum or silicon or a mixture of these compounds.

29. Method as claimed in claim 21, characterized in that the functional layer (32) is deposited as hard material layer system which comprises several deposits (33) of a nitride, carbide, boride or oxide of at least one of the transition metals of group IV, V or VI of the periodic system of elements and/or aluminum or silicon or a mixture of these compounds, wherein at least directly succeeding deposits differ by the stoichiometry of their metal or nonmetal content.

30. Method as claimed in claim 29, characterized in that the deposition of the hard material layer system (32) takes place with one or several deposits (33) of aluminum chromoxide-containing layers.

31. Method as claimed in claim 29, characterized in that transitions between the individual deposits (33) of the hard material layer system (32) with respect to the stoichiometry of their metal or nonmetal content are increased or decreased smoothly and continuously or stepwise.

32. Method as claimed in claim 29, characterized in that the layer of the individual deposits of the hard material layer system (32) is deposited with a thickness between 0.01 and 0.5 .mu.m, preferably betwen 0.02 and 0.1 .mu.m.

33. Method as claimed in claim 30, characterized in that nitride-, carbide- or boride-containing layers are deposited alternately with aluminum chromoxide-containing layers.

34. Method as claimed in claim 24, characterized in that at least one transition from the adhesion layer (31) to the aluminum oxide-containing layer or to the hard material layer system (32) or from the hard material layer system (32) or the aluminum oxide-containing layer to the cover layer (35) with respect to the stoichiometry of their metal or nonmetal content are increased or decreased smoothly and continuously or stepwise.

35. Method as claimed in claim 21, characterized in that the aluminum oxide-containing layer is substantially deposited as (Al.sub.1-xCr.sub.x).sub.2O.sub.3, wherein 0.05<.times.<0.80, however preferably 0.01<.times.<0.60.

36. Method as claimed in claim 21, characterized in that as workpiece (30) a tool, in particular a cutting, forming or injection molding tool is coated.

37. Method as claimed in claim 21, characterized in that as workpiece (30) a part, in particular a part for an internal combustion engine or a turbine, is coated.