Coating apparatus and coating method having divided pulses

Abstract

The present invention relates to a coating method and apparatus for coating an object 40. Magnetron cathodes (22a, 22b, 22c, 22d) having targets (24a, 24b, 24c, 24d) are disposed within a vacuum chamber 12, and power is supplied to the magnetron cathodes (22a, 22b, 22c, 22d) to generate plasma and sputter the targets (24a, 24b, 24c, 24d) to deposit a coating 44 on the object 40. Power is supplied periodically within a period duration (T) according to a HIPIMS method as cathode pulses (60), each cathode pulse (60) including at least two cathode sub-pulses (62) and an intervening cathode sub-pulse break (64). To enable the use of chopped HIPIMS to deposit coatings (44) with desirable properties in a particularly desirable manner, a bias voltage is applied to the substrate 40 to be coated with bias voltage pulses (66), each bias voltage pulse (66) including at least two bias sub-pulses (68) and an intervening bias sub-pulse break (70).

Description

Technical Field

[0001] The present invention relates to a coating method and a coating apparatus. In particular, the present invention relates to a method and an apparatus for coating an object by cathode sputtering, in which power is supplied to at least one cathode according to the HIPIMS method.

Background Art

[0002] It is known to apply a surface coating to an object or a part of an object in order to improve mechanical or chemical properties. In particular, in the case of tools and parts that are easily worn, it is known to apply a coating to the functional surface.

[0003] In particular, a hard material layer is known as a coating. In addition to the CVD method, in particular the PVD coating method, in particular the cathode sputtering method, is known for forming a thin coating.

[0004] In the field of cathode sputtering, the HIPIMS method (High-Power Impulse Magnetron Sputtering) is relevant. In contrast to conventional cathode sputtering in which the cathode operates at a constant power, for example by a DC voltage, in HIPIMS, a short electrical pulse of high voltage is applied to the cathode, achieving a very high peak power. As a result, the ionization of the plasma is significantly increased compared to conventional cathode sputtering, thereby obtaining advantageous properties of the generated coating.

[0005] In the current further development of the HIPIMS method, it has been proposed to subdivide the electrical pulses supplied to the cathode, that is, to apply a series of short pulses (a sequence of short pulses) instead of continuous cathode pulses. These pulses are here called cathode sub-pulses and form a subdivided cathode pulse. This method is called "chopped HIPIMS" or DOMS (Deep Oscillation Magnetron Sputtering).

[0006] For example, Barker et al.'s "Modified High-Power Impulse Magnetron Sputtering Process for Increasing Titanium Deposition Rate," Journal of Vacuum Science & Technology A:Vacuum, Surfaces and Films 31, 060604 (2013), discloses the deposition of titanium layers using a modified HIPIMS method in which the HIPIMS pulse is divided into a series of pulses. A pulse sequence with a duration of 100 μs is applied in a changing sequence with four or two micropulses of varying duration and varying switch-off periods at a frequency of 200 Hz and an average power of 0.75 kW. It has been reported that the divided HIPIMS pulses have a significant effect on achieving higher deposition rates than the conventional HIPIMS method.

[0007] Barker et al., “Investigation of c-HiPIMPS discharge during titanium deposition,” Surface & Coating Technology 258 (2014) 631-638, reports on titanium layers deposited by chopped HIPIMS using pulse sequences with 4 or 8 micropulses. It has been reported that the deposition rate increases significantly with increasing delays between micropulses, which can be explained by plasma ionization.

[0008] EP2587518A1 discloses an apparatus for generating a hydrogen-free (at least substantially hydrogen-free) ta-C layer on a substrate (workpiece) made of a metal or ceramic material. A vacuum chamber is connected to a vacuum pump and an inert gas source. A support device is provided for the substrate (workpiece). At least one graphite cathode having a corresponding magnetic arrangement forming a magnetron serves as a source of carbon material. A bias power supply is used to apply a negative bias voltage to the substrate. A cathode power supply is connected to the graphite cathode and associated anode and is designed to transmit high-power pulse sequences at (preferably programmable) time intervals. Each high-power pulse sequence comprises a series of high-frequency DC pulses, which, if applicable, are tuned to be supplied to at least one graphite cathode after a build-up phase.

[0009] EP3457428A1 discloses a method and apparatus for processing a substrate in a semiconductor processing system. The method begins by activating a pulse-synchronous controller coupled between a pulse RF bias voltage generator and a HIPIM generator. A first time signal is sent from the pulse-synchronous controller to the pulse HF bias voltage generator and the HIPIM generator. A sputtering target and HF electrodes placed in a substrate carrier are energized based on the first time signal. The target and electrodes are de-energized based on the end of the time control signal. A second time signal is sent from the pulse-synchronous controller to the pulse HF bias voltage generator, and the electrodes are energized and de-energized in response to the second time control signal without energizing the target.

[0010] US2008 / 0135400A1 discloses an apparatus for sputtering a target to produce a coating on a substrate. The apparatus comprises a magnetron having a cathode and an anode. A power supply is connected to the magnetron, and at least one capacitor is connected to the power supply. The apparatus also comprises an inductor operably connected to the capacitor. A first switch operably connects the power supply to the magnetron to charge the magnetron according to a first pulse. A second switch is connected to discharge the magnetron according to a second pulse. [Prior art documents] [Patent Documents]

[0011] [Patent Document 1] EP2587518A1 [Patent Document 2] EP3457428A1 [Patent Document 3] US2008 / 0135400A1 [Overview of the project] [Problems that the invention aims to solve]

[0012] An objective of the present invention may be to propose a method and apparatus for depositing a coating with advantageous properties in a particularly advantageous manner using the chopped HIPIMS method. [Means for solving the problem]

[0013] To solve the above problems, the coating method according to claim 1 and the claim 22 The coating apparatus described is proposed.

[0014] Each dependent claim refers to a preferred embodiment of the present invention in its own case.

[0015] According to the present invention, a coating method, or possibly a coating apparatus, for coating an object also called a substrate is proposed. The substrate may be a tool, in particular. Typically, a number of individual objects are coated simultaneously.

[0016] According to the present invention, a vacuum chamber is provided, which includes at least one magnetron cathode having a target, and an object (substrate) is placed inside the vacuum chamber. By supplying power to the magnetron cathode, plasma is generated, the target is sputtered, and a coating is deposited on the object in such a way that the plasma components are deposited on the surface of the substrate. In this process, power is supplied in the form of cathode pulses, periodically with a period duration T, in a pulsed manner according to the HIPIMS method. According to the chopped HIPIMS variation of this method, each cathode pulse is divided into at least two cathode subpulses, between which there is a cathode subpulse break (interruption) or cathode subpulse gap, where there is no power or power significantly reduced compared to the cathode subpulse is applied. Thus, the cathode pulse forms a sequence of cathode subpulses (a series of cathode subpulses) that are spaced apart in time.

[0017] The coating apparatus according to the present invention comprises a vacuum chamber having a receiving means for an object, a target, and at least one magnetron cathode, and also comprises an electrical cathode power supply and a control device. The cathode power supply is designed to supply power to the magnetron cathode according to the HIPIMS method, and cathode pulses are periodically supplied within a period duration T to generate plasma and sputter the target. The control device is designed to control the electrical cathode power supply so that individual, some, or all parameters of the power supplied to the magnetron cathode, in particular power, frequency, pulse shape and / or sequence, etc. The control device is designed to control the electrical cathode power supply according to a sequence such that each cathode pulse supplied to the cathode includes a cathode subpulse break (or gap) interposed with two cathode subpulses. The control device may be a programmable control unit that can, in particular preferably in addition to the cathode power supply, also control other functions of the coating apparatus, in particular the bias power supply and / or the process and / or reactive gas supply, in a time-dependent manner according to the coating program.

[0018] The sequence of cathode subpulses provided according to the chopped HIPIMS method has the advantage of reducing the risk of electric arc discharge formation, or preventing arc discharge formation from being terminated by subsequent subpulse breaks. Furthermore, even higher plasma ionization than in the case of continuous HIPIMS cathode pulses can be achieved with a series of cathode subpulses.

[0019] The cathode pulse is applied as a voltage pulse, and therefore, the following explanation and information regarding duration, pulse shape, sequence, etc., always refers to the voltage considered at the magnetron cathode. The resulting current, and therefore power, depends on the plasma, particularly the ionization conditions.

[0020] The cathode pulse can include various sequences of cathode subpulses. Preferably, these are voltage pulses. For example, a sequence of 2–10 cathode subpulses per cathode pulse, preferably 3–8 cathode subpulses, more preferably 4–6 cathode subpulses, has proven advantageous. The duration of the cathode subpulses may generally be, for example, 4–80 μs, preferably 5–50 μs or 6–35 μs, and particularly preferably 8–25 μs. As described below, the cathode subpulses may always have the same duration, but a sequence in which the cathode subpulses have different durations is preferred.

[0021] Within each sequence of cathode subpulses, a cathode subpulse break (or gap) exists between the cathode subpulses. The duration of the cathode subpulse break is preferably shorter on average than the duration of the cathode subpulse. For example, the duration of the cathode subpulse break is generally 2-30 μs, preferably 5-25 μs, and particularly preferably 8-20 μs. The durations of the cathode subpulses within a sequence may be the same or different.

[0022] According to the present invention, a bias voltage in the form of a bias voltage pulse that is periodic within a periodic duration is applied to an object. Each bias voltage pulse is divided, i.e., comprises at least two bias subpulses and an intervening bias subpulse break (or gap).

[0023] The bias voltage applied to the substrate is negative, and therefore, the positive ions of the plasma are accelerated toward the substrate surface. In contrast to the known use of DC voltage, the bias voltage according to this invention is applied in pulses, and similarly, in a sequence of individual bias subpulses with bias subpulse breaks in between. The frequency or period duration of the pulse sequence of the bias voltage is the same as that of the power or voltage at the cathode.

[0024] Such specifications regarding the temporal progression (progress, elapsed) of the bias voltage synchronized with the temporal progression of the voltage at the cathode provide a certain degree of control over the selection of the plasma components forming the coating. As shown, in the chopped HIPIMS method, i.e., the ion composition of the plasma during cathode operation with time-spaced cathode subpulses is time-dependent, meaning that different ions become dominant in the plasma at different times. By adjusting the temporal progression of the bias voltage to this extent, it is possible to select the type of ions integrated within the layer in a targeted manner.

[0025] This relates in particular to the influence of the respective amounts of metal and gas ions within the coating structure, which is made possible by appropriate timing. It has been observed that the amount of gas ions contained in the plasma, especially the amount of process gas ions, such as argon, has a temporal progression that deviates from the amount of metal ions from the start of the cathode subpulse. Therefore, the respective amounts of gas and metal ions in the coating can be adjusted in a targeted manner by appropriately selecting the duration and start time of the bias subpulse relative to the cathode subpulse. Here, the amount of argon, which is preferably used as the process gas, has a decisive effect on the properties of the coating. Thus, it has been shown that with a relatively high percentage of argon, the coating can achieve very high hardness and high internal stress, i.e., a hard and brittle layer, while layers without argon or with a low percentage of argon are comparatively much more ductile.

[0026] Furthermore, the use of “chopped bias,” i.e., segmented bias voltage, according to the present invention enhances the advantages of the chopped HIPIMS method, namely, the suppression of arc discharge. By dividing the cathode pulse into at least two cathode subpulses and interposing cathode subpulse breaks in between, the power is turned off during the cathode subpulse breaks, preventing the complete formation of an electric arc, thus suppressing the formation of an arc discharge and thus minimizing damage to the system (such as a target), particularly the coating of the object. This effect is assumed to be additionally supported if the bias voltage pulse is segmented so that at least one bias subpulse break is present in each period. During the bias subpulse breaks, the power in the object is also reduced or completely turned off. Therefore, under certain circumstances, the initial electric arc is suppressed and extinguished. This applies not only when the bias subpulse overlaps with or perfectly coincides with the cathode subpulse break, but also in the case of a time delay.

[0027] An advantageous development of the present invention relates particularly to the timing of the bias voltage pulse and the cathode pulse or the bias subpulse and the cathode subpulse.

[0028] Preferably, at least one of the bias subpulses starts with a delay time after the start of one of the cathode subpulses, and more preferably, multiple or all of the bias subpulses can start with a delay time after the start of the associated cathode subpulses. For example, the delay time of multiple or all of the bias subpulses may be the same for each associated cathode subpulse.

[0029] The sequences of the bias subpulses and cathode subpulses can be time-coordinated with each other, but the sequences do not need to match, and for example, each subpulse may have a different number and duration. For example, the time interval between two bias subpulses may correspond at least substantially (i.e., preferably with a deviation of less than ±10%) to the time interval between two cathode subpulses. The time interval is preferably measured between the start of each bias subpulse and cathode subpulse. The bias subpulses and cathode subpulses considered in the comparison may be time-consecutively consecutive with each other, or additional bias subpulses and cathode subpulses may be placed in between. Preferably, one, more, or all bias subpulse breaks may each have substantially the same duration as the associated cathode subpulse break.

[0030] The number of bias subpulses may be the same as or different from the number of cathode subpulses. In some preferred embodiments, the number of bias subpulses may be less than the number of cathode subpulses, i.e., there may be no associated bias subpulses for each cathode subpulse. For example, bias subpulses may be provided in a targeted manner so as to be associated only with higher-power cathode subpulses, while one or more lower-power cathode subpulses are not provided with associated bias subpulses.

[0031] The duration of the bias subpulse and / or bias subpulse break is preferably within the same interval as the duration of the cathode subpulse and / or cathode subpulse break. The delay time between the start of the cathode subpulse and the start of the associated bias subpulse is, for example, 5-200 μs, preferably 10-150 μs, and particularly preferably 10-60 μs.

[0032] The bias voltage pulse can overlap with the cathode pulse at least partially; that is, at least a portion of all bias voltage pulses may be applied simultaneously with at least a portion of one of the cathode pulses. However, this is not always necessary. In some applications, it has proven advantageous to delay the bias voltage pulse relative to the cathode pulse to avoid any actual time overlap.

[0033] The sequences of cathode subpulses and bias subpulses can be coordinated with each other, and in particular, each cathode subpulse may preferably start with a time delay and be associated with a bias subpulse having, for example, at least substantially the same duration. Thus, the number of bias subpulses may be the same as the number of cathode subpulses. However, if the number of bias subpulses is less than the number of cathode subpulses, it is also preferable that, for example, only some of the cathode subpulses, rather than all of them, have associated bias subpulses with, for example, a time delay in each case, for example, at least substantially the same duration. This is particularly preferable in the case of cathode subpulses of different durations. For example, if a somewhat lower peak power is achieved with a shorter cathode subpulse, it is advantageous not to provide an associated bias subpulse, and therefore there is no need to accelerate the ions generated with the lower peak power toward the object being coated.

[0034] In a preferred embodiment, the temporal evolution of the plasma and the formation and transition of various types of ions may be considered in relation to the relevant sequence of cathode subpulses in order to define an optimized sequence of bias subpulses. Thus, for example, gas and metal ions can be considered separately, and the sequence of bias subpulses can be selected so that a desired type of ion is preferably used to form a coating. Metal ions are particularly preferred. For example, the time-dependent transition of the generation of various types of ions, specifically gas and metal ions, can be observed to occur during and after the cathode pulse. When considering one type of metal ion (different types of metal ions differ from each other for the metal in question, and further distinguish between different types based on the degree of ionization), it is often possible to determine its temporal transition using one or more maximum values. In this case, it is preferable to select the sequence of bias subpulses so that at least one of the bias subpulses is applied during at least one maximum value. If the temporal transition includes multiple maximum values, two or more bias subpulses are preferably applied so that they are applied during at least one maximum value in each case. More preferably, a subpulse break may be provided for at least one type of metal ion during the minimum value of the temporal transition between two maximum values. Therefore, it is possible to select an appropriate sequence of bias subpulses so that the desired type of ion is selected according to the time series of metal ion generation and used to build the coating in preference to other ions, especially gas ions.

[0035] The present invention can be complemented and further developed in various embodiments, such as embodiments relating to advantageous sequences of cathode subpulses of different durations, embodiments relating to chopped HIPIMS methods having short period durations or fairly high frequencies, and embodiments relating to advantageous system technologies having capacitors and charging devices. Each of these embodiments offers its own advantages, but combinations of two or more of the above embodiments have proven particularly advantageous.

[0036] According to one aspect of the development of the present invention, at least two cathode subpulses have different durations. The cathode pulse includes at least one first and second cathode subpulse, each lasting (persisting) for at least 8 μs. The first and second cathode subpulses have different durations.

[0037] The names “first” and “second” cathode subpulses should generally be understood to mean that the second cathode subpulse occurs at a later time than the first cathode subpulse in the sequence of cathode subpulses, and that the first and second cathode subpulses do not necessarily have to be successive, nor is it relevant whether the first cathode subpulse is the first cathode subpulse in terms of time in the sequence of cathode subpulses. However, preferably, the first cathode subpulse may be the first, i.e., the earliest cathode subpulse in the sequence, and the second cathode subpulse may be the cathode subpulse that follows immediately after the cathode subpulse break in the sequence.

[0038] By varying the duration of the cathode subpulses in this way, the plasma can be affected in a targeted manner. This is because it has been shown that, surprisingly, the plasma behaves differently in successive cathode subpulses due to residual ionization that persists beyond the cathode subpulse break, even though the applied voltage at the cathode subpulse break that separates the cathode subpulses is zero or near zero. The proposed variation in the duration of the cathode subpulses makes it possible to take advantage of such effects in the plasma, in particular, to define an optimized sequence of cathode subpulses.

[0039] Therefore, for example, pre-ionization can be achieved by one or more shorter cathode subpulses, which can then be used to achieve particularly high peak power in one or more longer cathode subpulses that follow immediately or with a time delay.

[0040] According to a preferred embodiment, the first cathode subpulse described above is the first cathode subpulse in time of a cathode pulse, i.e., a series of cathode subpulses (a sequence of cathode subpulses). The second cathode subpulse described above may follow at a later point in the sequence, immediately following the first cathode subpulse, or an additional cathode subpulse may be provided temporally between the first and second cathode subpulses. The first cathode subpulse is preferably shorter than the second cathode subpulse. For example, the duration of the second cathode subpulse may be 110% to 600%, preferably 150% to 400%, and particularly preferably 200% to 300%, of the duration of the first cathode subpulse. Such a sequence of cathode subpulses makes it possible to achieve particularly high peak power during the second cathode subpulse.

[0041] In a preferred embodiment, the second cathode subpulse can last for, for example, at least 15 μs, preferably at least 20 μs, and particularly preferably at least 25 μs. The first cathode subpulse can last for, for example, up to 25 μs, preferably up to 20 μs, and particularly preferably up to 15 μs.

[0042] In a sequence having two or more cathode subpulses, it is preferable that the pulse duration of temporally consecutive cathode subpulses increases monotonically, that is, that they always last the same amount of time or longer until the next cathode subpulse in the time series. According to a preferred embodiment, the cathode pulse may include, for example, a third cathode subpulse that temporally follows the first and second cathode subpulses and has at least the same length as the first cathode subpulse and at least the same length as the second cathode subpulse.

[0043] According to one aspect of the development of the present invention, the chopped HIPIMS method is performed for an unusually short period of up to 1.5 ms, i.e., an unusually high frequency of pulses at the magnetron cathode of at least 667 Hz. The normal execution of the HIPIMS method proceeds from much lower frequencies or considerably longer period durations. Surprisingly, the inventors have found that using the chopped HIPIMS method with significantly increased peak power enables and is advantageous for short period durations. Preferably, even shorter period durations of 1.25 ms or less (corresponding to a frequency of at least 800 Hz) or 1 ms or less (corresponding to a frequency of at least 1 kHz) are also possible. In some embodiments, even shorter period durations of 0.2–0.6 ms (1.7–5 kHz) can be used.

[0044] At such unusually high frequencies, power is supplied in a very short time, but very high peak power is generated, resulting in high ionization. The high deposition rate is a major advantage of high frequencies; that is, faster layer construction has been demonstrated compared to low frequencies. This has a decisive impact on the economic feasibility of methods and apparatus in industrial applications.

[0045] The total duration of the cathode pulse itself, i.e., the sequence of cathode subpulses, and therefore the time during which power is applied in the form of cathode subpulses, preferably occupies only a portion of the period duration, for example, less than half of the period duration, and particularly preferably at most one-third of the period duration. This is particularly advantageous when the cathode pulse is supplied from a charged capacitor. Thus, the capacitor is discharged during each individual cathode subpulse over the entire duration of the cathode pulse and recharged over the remaining period duration. By limiting the duration of the cathode pulse, sufficient time can be provided for charging the capacitor. The duration of the cathode pulse is preferably in the range of 30 μs–400 μs, particularly preferably 80 μs–300 μs.

[0046] According to one aspect of the development of the present invention, the power supplied to the cathode in the form of a cathode subpulse is supplied from at least one charged capacitor of a cathode power supply. The cathode power supply comprises a capacitor provided in the form of a plurality of individual capacitors (capacitor bank) connected in parallel, and a charging device therefor.

[0047] This type of equipment has proven particularly well suited to supplying power according to the chopped HIPIMS method. In this case, preferably, all power for all cathode subpulses is supplied from the same capacitor, and the electrical connection to the cathode is closed by a switch for the duration of the cathode subpulse and disconnected during the cathode subpulse break and for the remaining period duration. The switch can preferably be controlled by a control device, which can be an IGBT, for example. The charging device can supply power to the capacitor to charge it, at least during the period duration outside of the cathode pulse, preferably during the cathode subpulse break. Particularly preferably, the charging device can remain connected to the capacitor at all times, for example, so as to be connected in parallel during the cathode subpulse.

[0048] According to a preferred embodiment, the charging device is controlled in a way that keeps the power constant, i.e., The power supplied to the capacitor (or, where applicable, the parallel circuit consisting of the capacitor and chamber during the pulse) and averaged over time is designed to be set to a fixed value. This has proven particularly stable, in contrast to alternative concepts such as setting a fixed voltage at which the capacitor is charged.

[0049] While each of the various embodiments and applications of the present invention is advantageous individually, they are particularly advantageous when combined. Hereinafter, several preferred embodiments can be used in conjunction with each individual embodiment, but they can also be used in any combination of two, three, or all four embodiments.

[0050] According to one development, the cathode pulse is a voltage pulse having a peak value of 600-1200V. While this value may vary individually depending on the embodiment, this range has been proven to be preferable for the voltage.

[0051] The cathode subpulse can have various transitions (progressions) and pulse shapes, such as a triangular pulse. Preferably, the cathode pulse is at least substantially rectangular or trapezoidal, i.e., it has a temporal transition with steep rising and falling edges and a nearly linear transition between them. Particularly in the preferred case where the cathode pulse is supplied from a charged capacitor, the transition, upon closer examination, corresponds to a discharge curve, but such a short section of the discharge curve is selected such that the transition does not deviate by more than 20%, particularly preferably 10%, at any point from the linear transition between the preferred starting voltage (after the steep rising edge) and the ending voltage (before the steep falling edge). It has been proven advantageous that the transition between the starting and ending voltages is at least substantially constant, in which case it should be understood that the voltage value does not change by at most 25%, particularly preferably 20%. This type of pulse is considered to be substantially rectangular in this case.

[0052] In a preferred embodiment, at least one peak power of the cathode subpulse is at least 50 kW. More preferably, significantly high peak powers of, for example, over 100 kW, 200 kW or more, and particularly preferably over 300 kW can be achieved. It has been shown that the magnitude of the maximum peak power of the cathode subpulse has a significant effect on the ionization of the metal of the target material. For example, studies have shown that for titanium target materials, the ratio of ionized particles to non-ionized particles (Ti+ / Ti) at a maximum peak power of approximately 500 kW of the cathode subpulse is about twice that at a maximum peak power of 150 kW.

[0053] Preferably, the peak power achieved during the cathode subpulse increases within the temporal sequence (temporal progression) of the cathode pulse. According to a preferred embodiment, the cathode pulse may include, for example, at least a first cathode subpulse and a second cathode subpulse that follows in time. The second cathode subpulse may follow immediately after the first cathode subpulse (after an intervening cathode subpulse break), or an additional cathode subpulse may be provided in between. In this case, the peak power during the second cathode subpulse is preferably at least 30% higher than that during the first cathode subpulse. In this way, particularly high peak power can be achieved overall.

[0054] The present invention is applicable to a wide variety of target materials and combinations of materials. The target can include both metallic and nonmetallic materials. Preferably, at least one component of the target is selected from materials of groups 4 to 6 of the periodic table, and further from the group including boron, carbon, silicon, yttrium, and aluminum. Preferably, all components of the target are selected from the above group. By the method and apparatus of the present invention, coatings can be produced from different material systems formed by the components of the target and, optionally, components supplied in gaseous form. Therefore, in addition to process gases, reactive gases, such as nitrogen, carbon-containing gases, or oxygen, can also be used in particular.

[0055] The bias voltage applied to the object to be coated is preferably pulsed with a bias voltage pulse synchronized with the cathode pulse, i.e., applied at the same frequency and with a fixed phase relationship. The pulse position and duration are preferably selected so that the type of ion is selected in a targeted manner. The position and / or duration can be selected during and after a cathode pulse having at least one maximum value, depending on the progression of the generation of at least one type of metal ion. In this case, it is preferable that the bias pulse is applied during the aforementioned maximum value.

[0056] Embodiments of the present invention will be described in more detail below with reference to the drawings. [Brief explanation of the drawing]

[0057] [Figure 1] This is a schematic diagram in plan view of a coating system as one embodiment of the present invention. [Figure 2] This figure shows an indexable insert (throwaway tip) as an object to be coated. [Figure 3] This is a schematic cross-sectional view of the coating on the substrate material. [Figure 4] This is a circuit diagram of one embodiment of a HIPIMS power supply. [Figure 5a] The typical temporal changes (progress, elapsed time) of cathode voltage and bias voltage are schematically shown. [Figure 5b] The typical temporal changes of cathode voltage and bias voltage are schematically shown. [Figure 6a] In the first exemplary embodiment, the measured temporal changes of cathode voltage and cathode current, as well as the amount and type of ions, are shown. [Figure 6b] In the first exemplary embodiment, the measured temporal changes of cathode voltage and cathode current, as well as the amount and type of ions, are shown. [Figure 7a] In a second exemplary embodiment, the measured temporal changes of cathode voltage and cathode current, as well as the amount and type of ions, are shown. [Figure 7b] In a second exemplary embodiment, the measured temporal changes of cathode voltage and cathode current, as well as the amount and type of ions, are shown. [Figure 8a] This figure shows the measured temporal changes of cathode voltage and cathode current, and bias voltage and bias current, in a third exemplary embodiment. [Figure 8b] This figure shows the measured temporal changes of cathode voltage and cathode current, and bias voltage and bias current, in a third exemplary embodiment. [Figure 9a] This figure shows the measured time-dependent changes of cathode voltage and cathode current, as well as bias voltage and bias current, for the fourth exemplary embodiment. [Figure 9b] This figure shows the measured temporal changes of cathode voltage and cathode current, and bias voltage and bias current, in the fourth exemplary embodiment. [Figure 10a] The graph shows the measured temporal changes of electrical variables in the fifth example. [Figure 10b] The graph shows the measured temporal evolution of electrical variables in the sixth example. [Figure 11a] The graph shows the measured temporal evolution of the electrical variables in the seventh example. [Figure 11b] The graph shows the measured temporal changes of the electrical variables in the eighth example. [Figure 12] This is a schematic diagram illustrating the orientation of the object to be coated relative to the cathode target. [Figure 13a] SEM images of the fracture modes of the coating on the free surface and scoop surface of an object in a comparative example are shown. [Figure 13b] SEM images of the fracture modes of the coating on the free surface and scoop surface of an object in a comparative example are shown. [Figure 14a] SEM images of the fragmentation morphology of the coating on the free and scoop faces of an object according to an exemplary embodiment are shown. [Figure 14b] SEM images of the fragmentation morphology of the coating on the free and scoop faces of an object according to an exemplary embodiment are shown. [Modes for carrying out the invention]

[0058] Figure 1 schematically shows the components of one embodiment of the PVD coating system 10. The interior 20 of the vacuum chamber 12 can be evacuated by a vent 14 to generate a vacuum. A process gas, preferably a noble gas or a mixture of various noble gases, such as argon and / or krypton, can be supplied from inlet 16. A reactive gas, such as nitrogen, can be supplied from inlet 18. In another embodiment, inlets 16 and 18 can be replaced with a common inlet for the process gas and the reactive gas.

[0059] In the exemplary assembly, four magnetron cathodes 22a, 22b, 22c, and 22d, each having a planar sputter target 24a, 24b, 24c, and 24d, are arranged inside the vacuum chamber 12 20. 22a,22b,22c,22d These are connected to controllable electrical HIPIMS power supplies 26a, 26b, 26c, and 26d, respectively, which allow a voltage to be applied to the conductive walls of the vacuum chamber 12, as will be described in detail below.

[0060] The arrangement and circuitry of the four magnetron cathodes 22a, 22b, 22c, and 22d shown in Figure 1 should be understood as an example. In alternative embodiments, other electrode configurations may be provided, for example, one, two, three, or more than four magnetron cathodes 22. In addition to or as an alternative to the illustrated magnetron cathodes 22a, 22b, 22c, and 22d, each connected to the HIPIMS power supplies 26a, 26b, 26c, and 26d, another type of cathode may be provided, for example, a DC magnetron cathode connected to a DC voltage power supply. The magnetron cathodes may also be electrically connected in a different way, facing a separate anode (not shown) rather than the wall of the vacuum chamber 12, as shown in the example.

[0061] Power supplies 26a, 26b, 26c, and 26d are shown only schematically in each case of Figure 1. Figure 4 shows a simplified circuit diagram of the HIPIMS power supply 26a. As capacitors 48, the power supply described above includes a capacitor bank consisting of a number of capacitors connected in parallel, a charger 46 for capacitors 48, and a controllable switch 36 which is preferably designed as an IGBT.

[0062] In the illustrated embodiment, the charging device 46 is a power control voltage source and is connected in parallel with the capacitor 48. The capacitor 48 is connected to the output terminal 28 of the HIPIMS power supply 26a via a switch 36. When the switch 36 is open, the charging device 46 charges the capacitor 48. When the switch is closed, the charge stored in the capacitor 48 is supplied to the output terminal 28, i.e., to the magnetron cathode 22a (Figure 1), where power is supplied by the charging device 46 connected in parallel. The switch 36 is controlled by the controller 38 of the control unit 50 of the system 10.

[0063] The magnetron cathodes 22a, 22b, 22c, and 22d are oriented so that their sputtering targets 24a, 24b, 24c, and 24d face the center of the vacuum chamber 12. Inside the vacuum chamber is a rotatable substrate table 30 on which several rotatable substrate carriers 32 are arranged to hold the substrate 40, i.e., the object to be coated. In the illustrated example, as shown in Figure 2, an indexable insert (throwaway tip) 40 is loaded as the substrate. The substrate 40 consists of a substrate material 42, for example, a WC / Co sintered material in the case of the indexable insert 40. However, this is only one example of a substrate 40 to be coated, and instead, parts or tools of different shapes and materials can be coated.

[0064] The substrate 40 is electrically connected to the substrate table 30 via the substrate carrier 32. The electrical bias voltage V is applied to the wall of the vacuum chamber 12. B A controllable bias power supply 34, which can apply a bias current I to the substrate table 30 and therefore to the substrate 40, is connected to the substrate table 30. This allows a bias current I to be applied at the connection point of the substrate table. B An ion flow is generated in the plasma, which can be measured as such. The connection of the bias power supply to the wall of the vacuum chamber should be understood as an example; instead, the bias power supply may be connected to another anode.

[0065] The HIPIMS power supplies 26a, 26b, 26c, 26d at the inlet and outlets 14, 16, 18, the bias power supply 34, and the pump (not shown) are each connected to the central control unit 50 of the system 10. The central control unit 50 can be programmed so that all parameters of the pretreatment and coating methods performed inside the vacuum chamber 12 20 are controlled by the control unit 50 according to a predetermined stored time-dependent control program. This control program can be changed, and the control unit 50 can store multiple different control programs that can be selectively retrieved. When the operating processes and settings of the system 10 are referred to below, they are predefined (specified) by the control program executed by the control unit 50.

[0066] During the operation of system 10, a vacuum is first created inside the vacuum chamber 12 20 as specified by the respective control programs executed in the control unit 50, followed by the introduction of a process gas, preferably argon. Next, the substrate 40 is pre-treated by ion etching. During etching, the bias power supply 34 is supplied with a high (negative) bias voltage V B The system is controlled to supply (positive) ions, thereby accelerating them onto the substrate. Gases and / or metal ions can be used for etching, enabling a variety of processes.

[0067] In a preferred embodiment, gas ion etching is performed first without operating the magnetron cathode 20 in HIPIMS operating mode. Gas ion etching is performed with a bias voltage I, for example, a DC voltage in the range of -100V to -400V. B As DC etching by applying a bias voltage, or alternatively, MF etching (bias voltage I BIt can be carried out at (-100V to -700V). During DC etching, electrons are generated by a hollow cathode (not shown), and discharged at an anode (not shown). The substrate table 30 is arranged between the anode and the hollow cathode. During MF etching, an MF power supply (not shown) generates electrons, and the electrons ionize the gas. The aforementioned electrons are discharged at the wall of the vacuum chamber 12 (grounded during operation).

[0068] After gas ion etching, metal ion etching may be used to further improve adhesion. For example, one or two magnetron cathodes 22a-d (equipped with targets made of, for example, Cr, Ti, V) are operated in the HiPIMS operation mode at a high peak power such that the donor material is ionized. Further, a bias voltage V B is applied as a DC voltage (DC) or in a pulsed manner, and the pulse of the bias voltage V B can be synchronized with the cathode pulse. The bias voltage V B is preferably between -300 and -1200V.

[0069] Subsequently, a coating 44 (Figure 3) is deposited on the pre-treated substrate surface thus. For this purpose, the HIPIMS power supplies 26a, 26b, 26c, 26d and the bias power supply 34 are controlled to supply an appropriate bias voltage V B and an appropriate cathode voltage V C . This example will be described below. The inlets 16, 18 are controlled to supply a process gas (argon) and, if applicable, a reactive gas (for example nitrogen). Thereby, plasma is generated inside 20 of the vacuum chamber 12, and under this plasma, the target 24a is sputtered. The positive ions of the plasma are accelerated towards the surface of the substrate material 42 of the substrate 40 by a negative bias voltage V B , and a coating 44 is formed there (Figure 3).

[0070] In this process, power is supplied to the magnetron cathodes 22a, 22b, 22c, and 22d from the HIPIMS power supplies 26a, 26b, 26c, and 26d in periodic pulses using the HIPIMS method. This power is supplied according to the chopped HIPIMS method in the form of cathode pulses 60 with pulse duration P for the transition of voltage Vc1 at the first magnetron cathode 22a, as schematically shown in Figures 5a and 5b. These are divided into cathode subpulses 62, three in the example shown in Figure 5a and four in Figure 5b, with pulse durations P1, P2, P3, and P4. Between the cathode subpulses 62, there are cathode subpulse breaks 64 with durations Z1, Z2, and Z3 during which no voltage is applied. The cathode pulses 60 are periodic at frequency f or with period duration T. After the end of the cathode pulses 60, no voltage is applied for the remainder of the period duration T.

[0071] In the example in Figure 5a, the sequence of cathode pulses 60 (a series of cathode pulses 60) includes three cathode subpulses 62, each with the same duration P1, P2, and P3. In the example in Figure 5b, another sequence of cathode subpulses 62a, 62b is shown, first consisting of two short cathode subpulses 62a, followed by two longer cathode subpulses 62b.

[0072] The voltage V at the magnetron cathode shown in Figures 5a and 5b C1 The pulse sequence is predetermined by the control of the HIPIMS power supplies 26a-d by the control unit 50, specifically by the appropriate control of the switches 36 of the HIPIMS power supplies 26a-d. It should be noted that the representation in Figures 5a and 5b is an ideal rectangular pulse shape for simplification, and the actual voltage curve deviates from this, as will be explained and shown individually below. In particular, the actual pulse shape shows a certain degree of decrease in the form of a discharge curve of the capacitor 48 over the duration of the cathode subpulse, but is accompanied by a certain degree of recharging by the charging device 46 during the cathode subpulse break 64.

[0073] Bias voltage VB Depending on the embodiment, the coating may have various temporal transitions (progressions) over the entire duration of the coating. In particular, the bias voltage V B This can be applied as a DC voltage or in a pulsed manner. Examples of preferred embodiments are schematically shown in Figures 5a and 5b.

[0074] In the example in Figure 5a, the bias voltage V B In total duration B ruba A bias pulse 66, applied in the form of an EAS pulse 66 and having the same frequency or period duration T, is periodic, similar to the cathode pulse 60. Similar to the cathode pulse 60, the bias pulse 66 is also divided, i.e., each with a duration Z B1 ,Z B2 Along with the intervening bias subpulse break 70, it is divided into bias subpulses 68 with durations B1, B2, and B3, respectively. The sequence of bias subpulses 68 within each bias pulse 66 corresponds to the sequence of cathode subpulses 64 within each cathode subpulse 60. That is, each cathode subpulse 64 has the same duration but a time offset, i.e., in all cases there is a delay time T D There is a delay associated with the bias subpulse 68.

[0075] In the example in Figure 5b, the bias voltage V B There are two within each period duration T. only Bias subpulse 68 has a bias pulse 66 It is applied in the form of a bias subpulse. 68 The sequence is cathode subpulse 62a, 62b This only partially corresponds to the sequence. This is because the bias subpulse 68 corresponds to each cathode subpulse. 62a, 62b Because it is not associated with the two initial short cathode subpulses. 62a Although no bias subpulse is associated with it, an associated bias subpulse 68 of the same duration is associated with it in both cases, with a delay time T D With the resulting delay, longer third and fourth cathode subpulses 62b followed by.

[0076] The bias voltage V shown in Figures 5a and 5b B The pulse sequence is also shown in an idealized form. Cathode voltage V C1 Similar to the pulse sequences, these are predetermined (defined) by the control unit 50 by appropriately controlling the bias power supply 34. The aforementioned bias power supply also includes a controllable switch (not shown) by which, depending on the embodiment, a capacitor bank is selectively connected to the output, preferably and as in the case of the HIPIMS power supply 26a, or alternatively, a DC voltage source is directly connected to the output.

[0077] Below, based on the first example (Figures 6a, 6b) and the second example (Figures 7a, 7b), we will explain how proper synchronization between the bias pulse 66 and bias subpulse 68 and the cathode pulse 60 and cathode subpulse 62 can affect the composition of the coating 44.

[0078] In the first example, Figure 6b shows examples of the temporal progression (progress, course) after the application of a cathode pulse 60 having the course shown in Figure 6a and having two cathode subpulses 62 with durations P1 and P2 of 20 μs each, and a cathode subpulse break 64 with a duration Z1 of 30 μs between the cathode subpulses 62.

[0079] As shown in the diagram, the plasma contains various types of metal ions and gas ions, but Ti + ,Ti 2+ N2 + , and N + Only the temporal evolution of each of the four types of ions is shown. These ions exhibit different temporal evolutions. For example, Ti +Metal ions show a significant maximum value approximately 120 μs after the start of cathode pulse 60. This is particularly remarkable considering that the total duration P of cathode pulse 60 is only 70 μs.

[0080] As outlined, Ti + It is possible to define a time interval 72 in which metal ions are dominant.

[0081] The positive gas and metal ions in the plasma are affected by a negative bias voltage V B It is accelerated toward the surface of the substrate 42 and therefore becomes part of the coating 44 deposited thereon. DC bias, i.e., bias voltage V B In the case of a continuous DC voltage, all ions are selected without exception for layer formation. As shown in Figures 5a and 5b, the voltage V at magnetron cathodes 22a-d C The bias voltage V synchronized with the temporal progression of B In the case of a pulsed temporal transition, temporal synchronization, i.e., delay time T D By appropriately selecting the duration of each bias subpulse 68, it is possible to select from among the gases and metal ions present in the plasma at various points in time.

[0082] Therefore, in the first example in Figures 6a and 6b, the bias voltage V B For example, Ti + The bias voltage V can only be applied during the time interval of 72 when the ions reach their maximum. B This can be applied in the form of a continuous bias pulse 66 that is not divided into bias subpulses 68.

[0083] In the second example shown in Figures 7a and 7b, the conditions corresponding to those in the first example are shown for a magnetron cathode 22a having a titanium target, supplied with nitrogen as the reactive gas and argon as the process gas. As shown in Figure 7a, the cathode pulse 60 in the second example includes two cathode subpulses 62 with different durations P1 = 10 μs and P2 = 20 μs, with a cathode subpulse break 64 having a duration Z1 of 18 μs between the cathode subpulses 62.

[0084] As shown in Figure 7b, various types of Ti ions + ,Ti 2+ N2 + , and N + Different temporal transitions occur for the sequence of cathode subpulses 62 shown in Figure 7a. These temporal transitions differ from those shown in Figure 6b for the sequence of cathode subpulses 62 shown in Figure 6a. In Figure 7b, in particular, Ti + The ion shows two maximum values ​​at approximately t=30μs and t=90μs, with a minimum value at t=50μs in between. In this case as well, the second high maximum value occurs after the end of the cathode pulse at t=48μs.

[0085] In Figure 7b, unlike the first example shown in Figure 6b, Ti + Two distinct time intervals 72 in which metal ions are dominant can be defined.

[0086] Therefore, in the second example in Figures 7a and 7b, the bias voltage V B This can be applied, for example, in the form of two bias subpulses 68 applied during a time interval 72, where a bias subpulse break 70 intervenes within the range of 40-70 μs. In this way, by selecting an appropriate sequence of bias subpulses 68 in a targeted manner, a high ratio of Ti can be formed in the coating 44. + Ions are guaranteed.

[0087] As a result, the composition of the coating 44 can be predetermined, and the characteristics of the layer are significantly affected by temporal synchronization (duration B1, B2, ... B of the bias subpulse 68). n , and the start time of the bias subpulse, for example, each calculated from the start of the cathode pulse 60, or the cathode subpulse 62 and bias subpulse 68 with their respective delay time T D (As allocated by...). Taking into account the proportion of process gas argon in the coating 44 (however, the temporal evolution of Ar ions is not shown in the examples of Figures 6b and 7b), its properties, particularly with respect to the internal stress and hardness of the coating 44, can be appropriately predetermined. When the Ar content is high, a coating 44 with high internal stress and high hardness is produced. Therefore, a coating 44 with a lower Ar content is considerably more ductile and has significantly lower internal stress.

[0088] Further examples of the pulse sequence of the cathode pulse 60, or rather the sequence of the cathode subpulse 62, are presented and explained below.

[0089] In the third example (Figures 8a, 8b) and the fourth example (Figures 9a, 9b), four magnetron cathodes 22a, 22b, 22c, and 22d are operated by the HIPIMS method at a frequency of 2 kHz, i.e., T = 500 μs. The power averaged over time is 12 kW per cathode in all cases, or 48 kW in total. The magnetron cathodes 22a-d are equipped with titanium, aluminum, and silicon targets 24a-d, for example, two titanium-silicon targets and two titanium-aluminum targets. Argon is accepted as the process gas and nitrogen as the reactive gas.

[0090] In all cases, the layer 44 deposited on the indexable insert serving as the substrate 40 is an Al-Ti-Si-N layer.

[0091] The third and fourth examples differ due to the different sequences of the cathode subpulse 62 (Figures 8a, 9a) and the different sequences of the bias subpulse 68 (Figures 8b, 9b).

[0092] In the third example, the sequence of cathode subpulses 62 includes three cathode subpulses 62, the third of which is longer than the first two. The durations of each cathode subpulse 62 and cathode subpulse break 64 are: P1 = 30 μs, P2 = 25 μs, P3 = 60 μs, Z1 = 25 μs, Z2 = 30 μs, This results in the generation of a total pulse length P of 170 μs within a period duration T of 500 μs.

[0093] The pulse shape of the cathode subpulse 62 is a voltage of approximately 680V in all cases. C1 It is essentially rectangular, and although it decreases slightly during the pulse progression (course), and a voltage drop of approximately 20% occurs in the third cathode subpulse 62, it is considered to be essentially rectangular in shape.

[0094] As shown in the diagram, the cathode current I C1 In each case, the current increases in a ramp-like (sloping) manner, reaching peak values ​​of approximately 75A during the first cathode subpulse 62, approximately 145A during the second cathode subpulse 62, and approximately 220A during the third cathode subpulse 62. Therefore, it is proven that it is advantageous for the third cathode subpulse 62 to be significantly longer than the first and second cathode subpulses 62, in which case the peak power reached is approximately 120kW.

[0095] As shown in Figure 8b, the bias voltage V B The associated sequence includes three bias subpulses 68, the duration of which substantially corresponds to the duration of the associated cathode subpulse 62, however, they have a delay time T of approximately 20 μs. DIt is applied.

[0096] The resulting bias current I B The current reaches a peak value of approximately 15A during the first bias subpulse 68, a peak value of approximately 33A during the second bias subpulse 68, and a peak value exceeding 60A during the third and longest bias subpulse 68, indicating that a large number of ions are being accelerated toward the substrate 42.

[0097] In the third example, coating 44 grows at a deposition rate of 1.9 μm / h.

[0098] As a measure of ionization, the ratio of ionized titanium atoms to the amount of unionized titanium atoms in the plasma can be measured by OES. In the third example, Ti + A very high ionization ratio of approximately 1.2 / Ti is obtained as a result. In a comparative example using undivided HIPIMS pulses under otherwise identical conditions, only a ratio of 0.93 is achieved.

[0099] In the fourth example, as in the third example, the sequence of cathode subpulses 62 includes three cathode subpulses 62, the first being even shorter than in the third example, and the third being even longer, thus increasing the duration of the cathode subpulses: P1 = 20 μs, P2 = 25 μs, P3 = 70 μs, Z1 = 25 μs, Z2 = 30 μs, Generates a sequence of.

[0100] In this case as well, the total pulse length P is 170 μs. As in the third example, the pulse shape is substantially rectangular (with a somewhat significant drop during the third cathode subpulse 62). In all cases, the cathode current I increases in a ramp-like manner. C1The current reaches approximately 25A during the first cathode subpulse 62, approximately 100A during the second cathode subpulse 62, and approximately 215A during the third cathode subpulse 62. Compared to the third example, an even higher peak power of approximately 130kW is achieved.

[0101] In the fourth example (Figure 9b), the bias voltage V B It includes three bias subpulses 68, and the durations of the bias subpulse 68 and cathode subpulse 62, and the durations of the bias subpulse break 70 and cathode subpulse break 64 are substantially corresponding to each other, but the bias subpulse 68 has a delay time T of approximately 25 μs. D This causes a delay. Bias current I B The voltage reaches a peak value of approximately 5A during the first bias subpulse 68, a peak value of approximately 25A during the second bias subpulse 68, and a peak value of approximately 65A during the third bias subpulse 68. Similar to the third example, the deposition rate is 1.9 μm / h.

[0102] In the fourth example, a Ti+ / Ti ratio of 1.57 is obtained, and therefore ionization is increased compared to the third example.

[0103] Thus, the coating 44 generated on the substrate material 42 is shown in Figures 14a and 14b, compared to the coating 44 from a comparative example (Figures 13a and 13b) which uses undivided HIPIMS pulses but otherwise is the same. In both cases, the coating 44 is shown on a free surface 52 that is oriented toward the target 24a during the operation of the coating system 10, and on a scoop surface 54 positioned perpendicular thereto, as shown in Figure 12.

[0104] Figure 14a shows a coating 44 on the free surface 52 of object 40, produced by a special chopped HIPIMS method having a sequence of cathode subpulses 62 as shown in Figures 9a and 9b. The coating 44 exhibits a much finer structure compared to the comparative example shown in Figure 13a.

[0105] As can be seen from Figure 14b, the coating 44 exhibits a much finer structure on the scoop face 54, which is not oriented toward the target 24b (Figure 12), than the comparative example (Figure 13b). Furthermore, in the comparative example shown in Figure 13b, the coating 44 has an oblique growth direction, indicated here by the white arrow as an example, which is not perpendicular to the substrate surface but rather oblique. On the other hand, such oblique growth is not observed in Figure 14b, which is based on the example of the chopped HIPIMS method using a special sequence of cathode subpulses 62 as shown in Figures 9a and 9b; instead, the coating 44 is clearly growing perpendicular to the substrate surface.

[0106] This more uniform growth and finer structure are thought to be the result of more advanced ionization.

[0107] The coating 44 produced according to the fourth exemplary embodiment exhibits particularly favorable properties, especially on the coated tool. In a machining test using an indexable insert (throwaway tip) in which one side was coated according to the comparative example described above (continuous, undivided HIPIMS pulses) and the other side was coated using a sequence of cathode subpulses 62 and bias subpulses 68 according to the fourth example, X6CrNiMoTi 17-12-2 During machining, a 20% longer tool life was achieved in the fourth example, which had a 6 μm coating 44, compared to the comparative example.

[0108] Further examples of the sequences of the cathode subpulse 62 and bias subpulse 68 are shown below.

[0109] First, in comparing the fifth example shown in Figure 10a with the sixth example shown in Figure 10b, the bias voltage V B Various types of synchronization of the temporal progression (progress, elapsed) of the bias voltage V are shown. In both examples, the bias voltage V BThis is divided into a cathode subpulse 62 and a bias subpulse 68 that is synchronized. However, in the fifth example (Figure 10a), the synchronization is without delay time (T D (=0μs), while in the sixth example (Figure 10b), the delay time (T D A (=10μs) interval is provided.

[0110] Figure 10a shows the voltage V at the first magnetron cathode 22a. C1 , current I to the first magnetron cathode 22a C1 , bias voltage V B , and a bias current I for a sequence of two consecutive cathode subpulses 62 having an intermediate cathode subpulse break 64 B This shows the measured temporal progression.

[0111] The frequency is 1000Hz (T=1ms). The pulse durations of cathode subpulse 62 and cathode subpulse break 64 are P1=20s, P2=20μs, and Z1=10μs. The pulse duration of cathode pulse 60 is 50μs (corresponding to 5% of the period duration T). Bias voltage V B These are periodic at the same frequency, with B1=20μs, B2=20μs, and Z B1 It includes two bias subpulses 66 and bias subpulse breaks 70, each with a duration of 10μs, and these have no delay (T D (=0μs) This is time-synchronized with the cathode subpulse 62.

[0112] In the example shown in Figure 10a, the charger 46 of the HIPIMS power supply 26a is regulated to a constant power of 3000W. The capacitor 48 of the HIPIMS power supply 26a is charged to approximately 700V at the start of the first cathode subpulse 62. As shown, the cathode subpulse 62 maintains a voltage V with little overshoot throughout the entire duration of the cathode subpulse 62. C1 It has a substantially rectangular shape with a slight decrease in size.

[0113] Cathode current I C1The current shows a temporal progression of a slight increase to approximately 150A during the first cathode subpulse 62 and a much steeper temporal progression to a peak value (peak power) of approximately 300A during the second cathode subpulse 62. Therefore, the first cathode subpulse 62 reaches a peak power value of approximately 100kW (peak power), and the second cathode subpulse 62 reaches a peak power of approximately 200kW. This increases the ionization degree of the plasma, and therefore generates a large number of plasma charge carriers that collide with the substrate 40, especially during the second bias subpulse 62, which causes the bias current I to increase rapidly. B It can be obtained from.

[0114] Therefore, the example in Figure 10a shows a cathode subpulse 62 of the same length and a bias voltage V that is precisely (without delay) synchronized with the same pulse sequence. B A fifth exemplary embodiment is shown, which has relatively high peak power and current I achieved during the second subpulses 62,68. C1 and I B A rapid increase in this has been shown.

[0115] For comparison with Figure 10a, Figure 10b shows, as a sixth exemplary embodiment, a delay time T of 10 μs between the start of each cathode subpulse 62 and bias subpulse 68. D This shows that the parameters are otherwise identical. Since more ions are already available, the bias current I increases at a faster rate between each bias subpulse 68. B It is possible to observe the different temporal progressions of this phenomenon.

[0116] Below, various sequences of cathode subpulses 62 are presented in comparison with the seventh example shown in Figure 11a and the eighth example shown in Figure 11b. In both examples, sequences of five cathode subpulses 62 are applied to each case. However, in the seventh example (Figure 11a), the first cathode subpulse 62 is shorter in time than the subsequent cathode subpulses 62, while in the eighth example (Figure 11b), the earlier cathode subpulse 62 is longer than the later cathode subpulse 62.

[0117] The two examples are characterized by the following parameters: TIFF0007901583000001.tif151168

[0118] In Figures 8a and 8b, the bias voltage V B In all cases, it is temporally synchronized with the cathode subpulse 62, periodic with the same frequency and pulse sequence, and has the same duration with no delay (T D (=0μs).

[0119] In the example in Figure 11a, only low peak powers of 27 kW and 68 kW are initially generated between the first two cathode subpulses 62. However, during the subsequent longer cathode subpulses 62, the peak power rises sharply to 179 kW and 190 kW, and then only reaches 86 kW during the final cathode subpulse 62.

[0120] In the example in Figure 11b, the peak power only increases to 64 kW and 156 kW during the first three cathode subpulses 62, and then decreases in subsequent cathode subpulses. Therefore, overall, a lower peak power value is achieved than in the example in Figure 11a.

[0121] As a result, it was proven that a sequence having a shorter cathode subpulse 62 initially and a longer cathode subpulse 62 thereafter is advantageous.

Claims

1. A coating method for coating an object (40), The object (40) is placed inside a vacuum chamber (12), and at least one magnetron cathode (22a, 22b, 22c, 22d) having targets (24a, 24b, 24c, 24d) is placed in a vacuum. In order to deposit a coating on the object (40), a plasma is generated and power is supplied to the magnetron cathodes (22a, 22b, 22c, 22d) so that the targets (24a, 24b, 24c, 24d) are sputtered. The aforementioned power is supplied periodically as a cathode pulse (60) within a period duration (T) according to the HIPIMS method. Each of the cathode pulses (60) includes at least two cathode subpulses (62) and an intervening cathode subpulse break (64), The cathode pulse (60) comprises at least first and second cathode subpulses (62), the first and second cathode subpulses (62) having different durations, and each of the first and second cathode subpulses (62) lasting at least 8 μs. A bias voltage pulse (66) is periodically applied to the object within the period duration (T). Each of the bias voltage pulses (66) includes at least two bias subpulses (68) and an intervening bias subpulse break (70). A coating method characterized by the following.

2. At least one of the bias subpulses (68) has a delay time (T) after the start of one of the cathode subpulses (62). Ｄ ) will begin with The coating method according to feature 1.

3. All bias subpulses (68) have a delay time (T) after the start of the associated cathode subpulse (62). Ｄ ) will begin with The coating method according to feature 2.

4. The delay time (T) of the bias subpulse (68) with respect to each of the associated cathode subpulses (62) Ｄ ) are equal, The coating method according to feature 3.

5. The delay time (T) of the bias subpulse (68) with respect to each of the associated cathode subpulses (62) Ｄ ) are different from each other, The coating method according to feature 3.

6. The number of bias subpulses (68) is less than the number of cathode subpulses (62). The coating method according to any one of features 1 to 5.

7. Metal ions are formed in the plasma by the cathode subpulse (62), The amount of one type of metal ion has a temporal progression from the start of the cathode pulse (60) having at least one maximum value, During the period between the aforementioned maximum values, at least one bias subpulse (68) is applied. The coating method according to any one of features 1 to 6.

8. The first cathode subpulse (62) is the first cathode subpulse (62) in the time of the cathode pulse (60), The first cathode subpulse (62) is shorter than the second cathode subpulse (62). The coating method according to any one of features 1 to 7.

9. The second cathode subpulse (62) lasts for at least 15 μs. The coating method according to feature 8.

10. The first cathode subpulse (62) lasts for a maximum of 25 μs. The coating method according to claim 8 or 9, characterized in that it is as described above.

11. The first cathode subpulse (62) lasts for a maximum of 20 μs. The coating method according to claim 8 or 9, characterized in that it is as described above.

12. The cathode pulse (60) includes a third cathode subpulse (62) that follows the first and second cathode subpulses in time, The third cathode subpulse (62) lasts for at least as long as the first and second cathode subpulses (62). The coating method according to any one of features 1 to 11.

13. The cathode pulse (60) is periodic within the period duration (T), The period duration (T) is a maximum of 1.5 ms. The coating method according to any one of features 1 to 12.

14. The cathode pulse (60) lasts for less than half of the period duration (T). The coating method according to feature 13.

15. The cathode power supply (26a, 26b, 26c, 26d) includes a charged capacitor (48) and a charging device (46) therefor, and the power supplied to the magnetron cathode (22a, 22b, 22c, 22d) is provided from the charged capacitor (48). The coating method according to any one of features 1 to 14.

16. The charging device (46) is adjusted to a constant power, The coating method according to feature 15.

17. The cathode pulse (60) is a voltage pulse having a peak value of 600V to 1200V. The coating method according to any one of features 1 to 16.

18. At least one of the cathode subpulses (62) is at least substantially rectangular, or has at least a trapezoidal temporal transition. The coating method according to any one of features 1 to 17.

19. The peak power of at least one of the cathode subpulses (62) reaches at least 50 kW. The coating method according to any one of features 1 to 18.

20. The cathode pulse (60) includes at least first and second cathode subpulses (62), the second cathode subpulse (62) temporally following the first cathode subpulse (62), The peak power supplied to the magnetron cathodes (22a, 22b, 22c, 22d) during the second cathode subpulse (62) is at least 30% higher than that during the first cathode subpulse (62). The coating method according to any one of features 1 to 19.

21. The targets (24a, 24b, 24c, 24d) include materials from groups 4 to 6 of the periodic table, boron, carbon, silicon, yttrium, and / or aluminum. The coating method according to any one of features 1 to 20.

22. A coating apparatus for coating an object (40), A vacuum chamber (12) having at least one magnetron cathode (22a, 22b, 22c, 22d) equipped with receiving means (30, 32) and targets (24a, 24b, 24c, 24d) for the object (40), In order to generate plasma and sputter the targets (24a, 24b, 24c, 24d), cathode power supplies (26a, 26b, 26c, 26d) are provided to periodically supply power to the magnetron cathodes (22a, 22b, 22c, 22d) as cathode pulses (60) within a period duration (T) according to the HIPIMS method, A control device (50) is designed to control the cathode power supplies (26a, 26b, 26c, 26d) to supply power to the magnetron cathodes (22a, 22b, 22c, 22d) such that each of the cathode pulses (60) includes at least two cathode subpulses (62) and an intervening cathode subpulse break (64), and each of the cathode pulses (60) includes at least first and second cathode subpulses (62), the first and second cathode subpulses (62) differ in terms of their duration, and each of the first and second cathode subpulses (62) lasts for at least 8 μs. The system comprises a bias power supply (34) for supplying power to the object (40) in the form of a bias voltage pulse (68), The control device (50) is further designed to control the bias power supply (34) so ​​that the bias voltage pulse (66) is periodically applied to the object (40) within the period duration (T), Each of the bias voltage pulses (66) includes at least two bias subpulses (68) and an intervening bias subpulse break (70). A coating apparatus characterized by the following features.

23. The cathode pulse (60) is periodic within the period duration (T), The period duration (T) is a maximum of 1.5 ms. The coating apparatus according to feature 22.

24. The cathode power supply (26a, 26b, 26c, 26d) includes a charged capacitor (48) and a charging device (46) therefor, and the power supplied to the magnetron cathode (22a, 22b, 22c, 22d) is supplied from the charged capacitor (48). The coating apparatus according to feature 22 or 23.

Images