METHOD OF TREATING A SUBSTRATE AND VACUUM DEPOSITION APPARATUS

20210040596 · 2021-02-11

    Inventors

    Cpc classification

    International classification

    Abstract

    Vacuum-treating a substrate or manufacturing a vacuum-treated substrate, including the steps: exposing a substrate in a vacuum chamber to a plasma environment, the plasma environment including a first plasma of a material deposition source and a second plasma of a non-deposition source; operating the plasma environment repeatedly between a first and a second state, the first state being defined by: a higher plasma supply power to the first plasma causing a higher material deposition rate and a lower plasma supply power delivered to the second plasma, the second state being defined by: a lower plasma supply power to the first plasma, compared with the higher plasma supply power to the first plasma and causing a lower material deposition rate and a higher plasma supply power to the second plasma, compared with the lower plasma supply power to the second plasma. Also, a vacuum deposition apparatus adapted to perform the method.

    Claims

    1. Method of vacuum-treating a substrate or of manufacturing a vacuum-treated substrate, the method comprising the steps: providing a vacuum chamber (1), providing at least one substrate (30) in said vacuum chamber, generating a plasma environment (2) in said vacuum chamber and exposing said at least one substrate to said plasma environment, wherein said plasma environment comprises a first plasma (11) of a material deposition source and a second plasma (21) of a non-deposition source; operating said plasma environment (2) repeatedly between a first and a second state, said first state (S1) being defined by: a higher plasma supply power to said first plasma causing a higher material deposition rate and a lower plasma supply power delivered to said second plasma, said second state (S2) being defined by: a lower plasma supply power to said first plasma, compared with said higher plasma supply power to said first plasma and causing a lower material deposition rate and a higher plasma supply power to said second plasma, compared with said lower plasma supply power to said second plasma.

    2. The method according to claim 1, comprising operating said plasma environment (2) periodically repeatedly between said first (S1) and said second state (S2), with a varying or constant period.

    3. The method according to claim 1, comprising establishing a treatment time span, during which the at least one substrate is exposed to the plasma environment (2), and wherein during at least 90% of said treatment time span, preferably during at least 99% of said treatment time span, exclusively one of said first state and of said second state prevails.

    4. The method of claim 1 comprising maintaining at least a part of said at least one substrate in said plasma environment (2) while moving said at least one substrate relative to said plasma environment.

    5. The method of claim 1, operating said plasma environment (2) periodically repeatedly between said first and said second state, with a varying or constant period and wherein said period T is selected to be:
    2 secT50 sec.

    6. The method according to claim 1, comprising operating of said plasma environment (2) additionally in a third state being defined by simultaneously delivering said higher plasma supply power to said first plasma and said higher plasma supply power to said second plasma.

    7. The method according to claim 1, comprising exposing a first treatment area (10) for said substrate in said plasma environment (2) at least predominantly to said first plasma and exposing a second treatment area (20) for said substrate in said plasma environment at least predominantly to said second plasma.

    8. The method according to claim 7, wherein there further prevails: said first treatment area (10) coincides with said second treatment area (20) or said first treatment area (10) and said second treatment area (20) overlap or said first treatment area (10) is within said second treatment area (20) or said second treatment area (20) is within said first treatment area (10) or said first and second treatment areas are commonly exposed to said plasma environment.

    9. The method according to claim 7, the method comprising: exposing at least a part of said at least one substrate to said first treatment area (10) during more than one repetition of said operating and then moving at least said part of said substrate into said second treatment area (20).

    10. The method according to claim 9, further comprising exposing at least said part to said second treatment area (20) during more than one of said repetition.

    11. The method according to claim 1, comprising etching said at least one substrate in said second treatment area (20) at least enhanced by said second plasma (21).

    12. The method of claim 1 comprising performing by means of said first plasma (11) one of PVD and of PECVD.

    13. The method of claim 1, wherein said first plasma (11) is the plasma of a sputtering source, preferably of a magnetron sputtering source.

    14. The method of claim 1, wherein at least one of said first plasma (11) and of said second plasma (21) is operated in an atmosphere comprising a reactive gas.

    15. The method of claim 1, wherein said lower plasma supply power to said first plasma (11) is established by supplying said first plasma with at most vanishing supply power.

    16. The method of claim 1, wherein said lower plasma supply power to said second plasma (21) is established by supplying said second plasma with at most vanishing supply power.

    17. The method of claim 1 comprising operating said plasma environment (2) periodically repeatedly between said first (S1) and said second state (S2), with a varying or constant period T thereby establishing said higher plasma supply power to said second plasma simultaneously with said higher plasma supply power to said first plasma during a time span , which is shorter than the period T of said operating.

    18. The method of claim 1, wherein a varying or constant period T of said operating said plasma environment (2) repeatedly is selected to be:
    2 secT50 sec, a third state of said operating being enabled for a third time span t3 for which there is valid:
    0 sec<t32 sec.

    19. The method of claim 1, comprising operating said plasma environment additionally in a third state, defined by simultaneously delivering said higher plasma supply power to said first plasma and said higher plasma supply power to said second plasma and comprising operating said plasma environment still additionally in a fourth state, defined by simultaneously delivering said first lower plasma supply power to said first plasma and said second lower plasma supply power to said second plasma.

    20. The method of claim 19 comprising operating said plasma environment (2) periodically repeatedly between said first and said second state, thereby establishing said higher plasma supply power to said second plasma simultaneously with said higher plasma supply power to said first plasma during a time span , which is shorter than the period T of said operating.

    21. The method according to claim 19 comprising establishing said fourth state during a fourth time span t4 for which there is valid:
    0 sec<t42 sec.

    22. The method according to claim 1 comprising exposing a first treatment area (10) for said substrate in said plasma environment (2) at least predominantly to said first plasma and exposing a second treatment area (20) for said substrate in said plasma environment at least predominantly to said second plasma and positioning simultaneously a first part of said at least one substrate exclusively in said first treatment area (10) and a different, second part of said at least one substrate exclusively in said second treatment area (20).

    23. The method of claim 22 comprising positioning a further different part of said at least one substrate in an overlapping area of said first (10) and second (20) treatment areas.

    24. The method according to claim 1 comprising exposing a first treatment area (10) for said substrate in said plasma environment (2) at least predominantly to said first plasma and exposing a second treatment area (20) for said substrate in said plasma environment at least predominantly to said second plasma and positioning simultaneously a part of a first substrate exclusively in said first treatment area and a part of a second substrate exclusively in said second treatment area.

    25. The method of claim 24 comprising positioning a further part of said first substrate in an overlapping area of said first treatment area (10) and of said second treatment area (20), simultaneously with positioning said one part of said first substrate exclusively in said first treatment area.

    26. The method according to claim 24 comprising positioning a further part of said second substrate in an overlapping area of said first treatment area (10) and of said second treatment area (20), simultaneously with positioning said one part of said second substrate exclusively in said first treatment area.

    27. The method of claim 1, comprising operating said plasma environment (2) periodically repeatedly between said first (S1) and said second (S2) state, with a varying or constant period T, and wherein a first duty cycle DC1 defined as the ratio of time span during which the first state is enabled and said period of said operating, is kept within the range: 25%DC190%.

    28. The method of claim 1, comprising operating said plasma environment (2) periodically repeatedly between said first (S1) and said second state (S2), with a varying or constant period T, and wherein a second duty cycle DC2 defined as the ratio of time span during which the second state is enabled and said period of said operating, is kept within the range: 25%DC295%.

    29. The method of claim 1 comprising exposing a first treatment area (10) for said substrate in said plasma environment (2) at least predominantly to said first plasma and exposing a second treatment area (20) for said substrate in said plasma environment at least predominantly to said second plasma and establishing a gas atmosphere in said first and second treatment areas comprising equal gas or gases or consisting of equal gas or gases.

    30. The method of claim 1, thereby providing at least one of a further plasma adapted to deposit a material on said at least one substrate and of a further plasma adapted to non-deposit a material on said at least one substrate.

    31. A vacuum deposition apparatus comprising: a vacuum chamber (1), in said vacuum chamber a material deposition first treatment area (10) and a material non-deposition second treatment area (20), a drivingly movable substrate holder (31) at least one of movable between said first treatment area and said second treatment area and of movable within at least one of said first treatment area and said second treatment area, a first plasma source (12) generating a first plasma (11) at least predominantly in said first treatment area, a second plasma source (22) generating a second plasma (21) at least predominantly in said second treatment area, a controllable first electric plasma supply arrangement (13), operationally connected to said first plasma source comprising a first control input to control different supply power levels, a controllable second electric plasma supply arrangement (23) operationally connected to said second plasma source and comprising a second control input to control different supply power levels, and a control arrangement operationally connected to said control inputs, wherein said control arrangement is constructed to first control via said first control input said first electric plasma supply to repeatedly supply a first higher, plasma sustaining power level and a first lower power level, wherein said control arrangement is constructed to second control via said second control input said second electric plasma supply to repeatedly supply a second higher, plasma sustaining power level and a second lower power level, and wherein said control arrangement is further constructed to time-synchronize said first and second controls.

    32. The vacuum deposition apparatus of claim 31 said first (10) and second (20) treatment areas are a common area or said first and second treatment areas overlap or said first and second treatment areas are in open communication.

    33. Vacuum deposition apparatus according to claim 31, wherein there is valid at least one of: said first electric plasma supply arrangement is a controlled DC power supply arrangement, said second power supply is a controlled RF power supply.

    34. Vacuum deposition apparatus according to claim 31, wherein said control arrangement controls a temporal overlap of said first higher power level and of said second lower power level and a temporal overlap of said first lower power level and of said second higher power level.

    35. Vacuum deposition apparatus according to claim 31, wherein said control arrangement is adapted and configured to synchronize an operating between a first state and a second state of a method of vacuum-treating a substrate or of manufacturing a vacuum-treated substrate, the method comprising the steps: providing a vacuum chamber (1), providing at least one substrate (30) in said vacuum chamber, generating a plasma environment (2) in said vacuum chamber and exposing said at least one substrate to said plasma environment, wherein said plasma environment comprises a first plasma (11) of a material deposition source and a second plasma (21) of a non-deposition source; operating said plasma environment (2) repeatedly between the first state and the second state, said first state (S1) being defined by: a higher plasma supply power to said first plasma causing a higher material deposition rate and a lower plasma supply power delivered to said second plasma, said second state (S2) being defined by: a lower plasma supply power to said first plasma, compared with said higher plasma supply power to said first plasma and causing a lower material deposition rate and a higher plasma supply power to said second plasma, compared with said lower plasma supply power to said second plasma.

    36. Vacuum deposition apparatus according to claim 31, wherein said first plasma source is a pulsed DC magnetron source, in particular a pulsed DC magnetron source with superimposed RF.

    37. Vacuum deposition apparatus according to claim 31, wherein said first plasma source comprises a rotatable magnet arrangement and/or a rotatable target.

    38. Vacuum deposition apparatus according to claim 31, wherein said substrate holder (31) is adapted to hold a plurality of substrates, in particular wherein said substrate holder is designed as table or as drum rotatable around a respective central axis.

    39. Vacuum deposition apparatus according to claim 38, said substrate holder (31) being adapted to hold two neighboring substrates in said plurality of substrates in a distance from each other, said distance being smaller than or equal to a distance between said first treatment area (10) and said second treatment area (20).

    Description

    [0099] The invention shall now be further exemplified with the help of figures. The figures show:

    [0100] FIG. 1 a schematic cross-section of a vacuum chamber and a substrate in two different states realized in the method according to the invention;

    [0101] FIG. 2 a state diagram illustrating an operating of a plasma environment between a first and a second state;

    [0102] FIG. 3, in six sub-FIGS. 3.a) to 3.f), six individual state diagrams illustrating variants of an operating of a plasma environment between a first and a second state and involving a third and/or fourth state;

    [0103] FIG. 4 a schematic cross-section of a vacuum chamber and a substrate in four different states realized in in an embodiment of the method;

    [0104] FIG. 5 a timing diagram showing a time course of a power delivered to a first plasma and of a power delivered to a second plasma in an embodiment of the method;

    [0105] FIG. 6 a timing diagram showing a time course of a power delivered to a first plasma and of a power delivered to a second plasma in a further embodiment of the method;

    [0106] FIG. 7 a timing diagram showing a time course of a power delivered to a first plasma and of a power delivered to a second plasma in a further embodiment of the method;

    [0107] FIG. 8 a schematic diagram of an embodiment of the apparatus according to the invention;

    [0108] FIG. 9, in subfigures FIG. 9.a) and FIG. 9.b), the time-dependency of voltages measured at the first and second source during two variants of the method according to the invention;

    [0109] FIG. 10 the time-dependency of voltages measured at the first and second source during a method according the state of the art.

    [0110] FIG. 1 shows schematically and simplified, cross-section of a vacuum chamber and a substrate in two different states, one in the upper part and one in the lower part of the figure. In both states, a substrate 30 is positioned in a vacuum chamber 1 and exposed to a plasma environment 2. A first plasma 11 is adapted to deposit a material on the substrate. A second plasma 21 is adapted to non-deposit a material on the at least one substrate, i.e. to treat the substrate in a way that does not lead to deposition of a material. The first plasma 11 can be operated with a higher or a lower plasma power as indicated in the symbolically shown first plasma source 12 by P1: HIGH for the higher power and by P1: LOW for the lower power. Similarly, the second plasma 21 can be operated with a higher or a lower plasma power as indicated in the symbolically shown second plasma source 22 by P2: HIGH for the higher power and by P2: LOW for the lower power. According to the method of the invention, the plasma environment 2 is repeatedly operated between a first state S1, as shown in the upper part of the figure, and a second state S2, as shown in the lower part of the figure. The curved arrows on the left and the right of the figure indicate a repeated alternating between the two states.

    [0111] In the first state S1, a higher plasma supply power is delivered to the first plasma. This causes a higher material deposition rate, as symbolically indicated by the high density of dots. Simultaneously, a lower plasma supply power is delivered to the second plasma, which is indicated by a low density of dots. In the second state S2, a higher plasma supply power, i.e. higher compared to the lower plasma supply power delivered in the first state, is delivered to the second plasma. Simultaneously, a lower plasma supply power is delivered to the first plasma. The power is lower compared with the higher plasma supply power as delivered to the first plasma during the first state.

    [0112] The lower plasma supply power delivered to the first plasma causes a lower material deposition rate. This is symbolically indicated by a lower density of dots in the region of the first plasma 11.

    [0113] The plasma environment 2 is schematically indicated by a region surrounded by a dash-dotted line. Its borders may not be sharply defined. Anyhow, the plasma environment 2 forms a continuous region inside the vacuum chamber.

    [0114] FIG. 2 shows a state diagram illustrating an operating of a plasma environment between a first and a second state. The states involved are defined by the amount of power delivered to the first and the second plasma, respectively. In horizontal rows the plasma supply power to the first plasma is indicated, whereas in vertical rows the plasma supply power to the second plasma is indicated. The diagram shows three horizontal rows and three vertical rows. The top horizontal row corresponds to the lower plasma supply power delivered to the first plasma, as marked by P1: LOW. The bottom horizontal row corresponds to the higher plasma supply power delivered to the first plasma, as marked by P1: HIGH. The middle horizontal row corresponds to a transitional state of the power delivered to the first plasma. The left vertical row corresponds to the lower plasma supply power delivered to the second plasma, as marked by P2: LOW. The right vertical row corresponds to the higher plasma supply power delivered to the second plasma, as marked by P2: HIGH. The middle vertical row corresponds to a transitional state of the power delivered to the second plasma. Five fields in the diagram are marked by cross-hatching. These five fields correspond to transitional power states ST. The first state S1 thus corresponds to the bottom left field of the diagram and the second state S2 of the plasma environment thus corresponds to the top right field of the diagram. According to the invention, the plasma environment comprising the first and the second plasma is operated repeatedly between the first and the second state as indicated by arrows in both directions. The diagram does indicate the succession of states that are reached, but it does not imply any information on the time spent in the first S1 or the second state S2, nor how long it takes to cross the region of transitional state ST.

    [0115] FIG. 3 shows state diagrams similar to the ones explained in relation with FIG. 2. Embodiments of the method are illustrated. The embodiments involve additional states in the succession of states.

    [0116] FIG. 3.a) shows an embodiment, wherein operating of the plasma environment involves a third state S3, which is defined by simultaneously delivering the higher plasma supply power to the first plasma and the higher plasma supply power to the second plasma. Accordingly, the third state is in the bottom right field of the diagram. In the embodiment of the method illustrated in FIG. 3.a), on the transition from the first state to the second state, the plasma supply power to the first plasma is kept on its higher level, while switching the plasma supply power to the second plasma from its lower lever to its higher level, thereby reaching the third state. In the third state, the second plasma may ignite fast and reproducible, or, if already ignited, reach stable operating conditions fast. In the next step, the plasma supply power to the first plasma is reduced to its lower level, thus reaching the second state. In the embodiment shown, the step from the second state S2 back to the first state S1 is taken directly.

    [0117] FIG. 3.b) shows a similar embodiment of the method as FIG. 3.a), but in this case, the third state is established on the transition from the second state S2 to the first state S1.

    [0118] FIG. 3.c) again shows a similar embodiment of the method. In this case, the third state is established on both ways between the first and the second state.

    [0119] FIG. 3.d) shows an embodiment, wherein operating of the plasma environment involves a fourth state S4, which fourth state is defined by simultaneously delivering the first lower plasma supply power to the first plasma and the second lower plasma supply power to the second plasma. In this fourth state, the first and second plasma run both on the respective lower power level, such that the mutual interaction is particularly low. In the present embodiment, the fourth state is established between the second and the first state.

    [0120] FIG. 3.e) shows an embodiment, wherein the fourth state is established on both ways when going forth and back between the first and second state.

    [0121] FIG. 3.f) shows as a further example, an embodiment of the method involving the first, second, third and fourth states as defined above. In the embodiment shown, a sequence of the first state S1, third state S3, second state S2, fourth state S4 and first state S1 (again) is established. Here, during the third state, a higher power level delivered to the first plasma may help to induce a fast ramp up of the second plasma, whereas on the way back to the first state, a clean decoupling may be achieved by switching both plasmas to their lower power level for an intermediate time span, during which the fourth state is established, before reaching the first state again.

    [0122] FIG. 4 shows in a simplified manner, similar to FIG. 1, an embodiment of the method comprising moving at least a part of the at least one substrate in the plasma environment.

    [0123] Specifically, a substrate 30 is positioned on a substrate holder 31, which is rotatable around an axis and thus is able to transport the substrate from a position in the region of the first plasma 11 into a position in the region of the second plasma 21. The plasma environment in the vacuum chamber 1 is switched back and forth between the first state S1 and the second state S2, as indicated by the double arrow between the top and the second partial figure. A moving process may be slow compared to the alternating between the first and second state of the plasma environment. Over a longer time span, as indicated by the curved arrow from the second to the third partial figure, due to the moving, the substrate arrives in the region of the second plasma 21. Irrespective of the position of the substrate, the alternating between the first and the second state of the plasma environment continuous, as shown in the bottom partial figure, where the second state S2 is established again. Over a longer time span, as indicated by the long curved arrow starting at the bottom partial figure, due to the moving, the substrate arrives again in the original position as shown in the top partial figure. Instead of having a complete substrate 30 in the respective position, a part of a large substrate may take the role of the substrate shown here. Several substrates may be placed at various positions of the substrate holder, such that each individual substrate undergoes a similar procedure as the one substrate shown in this figure.

    [0124] FIG. 5 shows a timing diagram for an operating of the plasma environment. A time axis runs from left to right. A time course of a first power P1 delivered to a first plasma and of a second power P2 delivered to a second plasma is displayed each above its own zero power reference marked as thin line. First and second power alternate repeatedly between a higher and a lower level. For easier identification, the respective time spans are denoted by HIGH and LOW for both, first and second power. The diagram is schematic and not to scale. In particular, the absolute power levels of the first power, being related to a material deposition, and the second power, being related to a non-depositing, treatment plasma power, may have different orders of magnitude. Some time is needed to switch from one state to the other, leading in this case to a trapezoidal time course of the power levels. Alternatingly, a first state and a second state of the plasma environment is established, as marked below the time axis. The alternating may be periodic. A period T, i.e. a time of one repetition, may e.g. be in the range of 2 to 50 microseconds.

    [0125] FIG. 6 shows a timing diagram similar to the time diagram shown in FIG. 5. In the present case, the lower power levels delivered to the first and to the second plasma correspond to zero power. Accordingly, the alternating can be seen to switch between an ON and an OFF value instead of a higher and a lower power. In particular, in the case shown here, the power delivered to the first plasma may be applied in form of a pulsed DC current, the pulsing corresponding to the ON/OFF power level P1. The second plasma may be applied in form of RF power, e.g. in form of RF having a frequency of 13.56 MHz. A succession of first state S1 and second state S2 is realized. A synchronization of the two states may e.g. be achieved as follows. A first power supply, which may be a DC power supply, and a second power supply, which may be a RF power supply, produce in response to a trigger signal a rectangular output power pulse. During a time span T.sub.DCon the first power P1 is ON, during a subsequent time T.sub.DCoff the power P1 is OFF. During a time span T.sub.RFoff the second power P2 is OFF, during a time span T.sub.RFon, the second power P2 is ON. The trigger signal sent to the second power supply may be delayed by T.sub.delay with respect to the trigger signal sent to the first power supply. A timing as illustrated in the timing diagram according to FIG. 6 may then be achieved by setting the time spans T.sub.DCon, T.sub.RFoff and T.sub.delay to equal values. Similarly, T.sub.DCoff and T.sub.RFon are set to be equal with respect to each other, but the latter may be different from T.sub.DCon etc. This way, time spans during which power is delivered to the first plasma do not overlap with time spans, during which power is delivered to the second plasma.

    [0126] FIG. 7 show s similar timing diagram as FIG. 6. The same notation regarding the time spans is used. In this embodiment, however, the time span T.sub.DCon is selected to be slightly longer than T.sub.RFoff. Furthermore, T.sub.delay is selected such that at the beginning of a time span with higher plasma supply power to the first plasma and at the end of this time span, an overlap with a time span with higher plasma supply power to the second plasma is created. I.e. the third state S3 of the plasma environment is established between the occurrence of the first state S1 and the second state S2. If the sequence is made periodic, a period T, which in this case is defined as the sum T.sub.DCon+T.sub.DCoff, may lie in the range of 2 to 50 microseconds. T.sub.DCon may be in the range 1 to 48 microseconds. T.sub.RFon may be in the range 2 to 48 microseconds. The sum T.sub.DCon+T.sub.RFon may be larger than the period T due to the time spans, during which the state S3 is established, i.e. during which both ON states overlap.

    [0127] FIG. 8 shows an embodiment of an apparatus according to the invention, schematically and in a partial cross-section. In a vacuum chamber 1, a first plasma source 12 and a second plasma source 22 are provided. The first plasma source 12 creates a first plasma 11. The first plasma 11 is adapted to deposit a material on a substrate 30. In the embodiment shown, the region of the first plasma forms a PVD station. The first plasma source 12 is designed as a magnetron. A first electric plasma supply arrangement 13, in this case a DC power supply, delivers electrical power to the first plasma. A region of dense plasma is indicated by a high density of dots. A second plasma source 22 is driven by an RF power supply, which forms a second electric plasma supply arrangement 23. A second plasma 21 is adapted to treat a substrate in a non-depositing way, e.g. by plasma etching or by applying a compactification treatment by plasma. A dense plasma region in the second plasma 21 is indicated by high density of dots, too. A weak plasma 33, which is in contact with the first and the second plasma, may establish a coupling effect between the first and second plasmas. The weak plasma may comprise ions and/or electrons originating from the first or from the second plasma source. As the inventors have recognized, this coupling or cross-talk may be reduced, may be stopped completely or may be selectively applied to control the process parameters, by applying the method according to the present invention. The method may involve essentially an alternating or anti-synchronous operation scheme regarding operating of the first plasma and the second plasma. In the situation illustrated here, the plasma environment comprises the first plasma 11, the second plasma 21 as well as the weak plasma 33.

    [0128] A substrate holder 31 in form of a table carries substrates 30. In the cross-section shown, two substrates are visible, one substrate 30 in the plasma deposition-area 10 in the region of the first plasma 11, another substrate 30 in the non-deposition plasma treatment area 20 in the region of the second plasma 21. A synchronization unit 32 is operationally connected to power-level control inputs of the first electric plasma supply arrangement and the second electric plasma supply arrangement. The connection may be realized by electrical wire connection or wireless. The synchronization unit and the connections to the electric supplies form a control arrangement. The control arrangement is constructed to control the first supply to alternatingly supply a first higher, plasma sustaining power level and a first lower power level. The control arrangement is constructed to control the second electric plasma supply arrangement to alternatingly supply a second higher, plasma sustaining power level and a second lower power level, and the control arrangement is further constructed to synchronize the supply of the first and of the second electrical supplies.

    [0129] To explain measured results presented further below, reference is made to the following optional measuring devices. One is for measuring a target voltage U.sub.target on the side of the first plasma source. The other is for measuring the rf voltage U.sub.rf and a DC self-bias voltage U.sub.DC self-bias on the side of the second plasma. Temporal stability of these voltages is an important indicator for the stability of the vacuum deposition processes.

    [0130] FIG. 9.a) and FIG. 9.b) show the time-dependency of voltages measured at the first and second source during two variants of the method, both of which are performed in an apparatus as shown in FIG. 8. For both variants, as well as for FIG. 10, the following applies. Source 1 voltage was measured by U.sub.target. Its values in volt are plotted against the axis on the left as a dashed curve. Source 2 voltage was measured by U.sub.rf. Its values in volt are plotted against the axis on the right of the figure a continuous black curve. Source 2 voltage is a rapidly oscillating rf voltage, such that the curve covers an area delimited by its amplitude. The DC self-bias U.sub.DC self-bias at source 2, i.e. a time-average of the rf voltage at source 2, is plotted as continuous grey curve. The horizontal axis displays the time in microseconds.

    [0131] FIG. 9.a) shows the time-dependency of the voltages for a variant of the method, which implements the following sequence of states: S3.fwdarw.S1.fwdarw.S2.fwdarw.S3.fwdarw.S1.fwdarw.S2.fwdarw.S3.fwdarw. . . . . Each of the three states prevail for about 4 microseconds. Both sources are operated in a pulsed way. In the first state S1 and the third state S3, the DC power supply delivers a negative voltage in the range of 400 V to 1 kV to the first plasma, i.e. it delivers the higher plasma supply power to the first plasma. In the second state, the DC power supply is switched off, such that source 1 voltage goes to approximately 0 V. In the second state S2 and the third state S3, the rf power supply delivers rf power, i.e. it delivers the higher plasma supply power to the second plasma. In the first state S1, the rf power supply is switched off. Amplitude and averaged rf voltage at the second source grow during the second state S2 and decrease in the third state S3. This change in the time-dependency is caused by the change from the lower power to the higher power at the side of the first source. As the inventors have recognized, a systematic drifting away of the operating parameters at the second source may be effectively reduced by synchronizing the pulsed operation of the second source with the pulsed operation of the first source.

    [0132] FIG. 9.b) shows a variant of the method with slightly modified timing of the pulses as compared to FIG. 9.a).

    [0133] Here, the rf pulse to the second source overlaps with the DC pulse to the first source at the beginning and at the end of the rf pulse, such that the following sequence of states results: S3.fwdarw.S1.fwdarw.S3.fwdarw.S2.fwdarw.S3.fwdarw.S1.fwdarw. . . . . With this sequence, rf amplitude and averaged rf voltage ramp up much faster at the beginning of the rf pulse and they decay much faster at the end of the rf pulse. In total, rf amplitude and averaged rf voltage, which both are operating parameters that have significant influence on the properties of the layers produced, can be kept close to a desired value over a relatively large fraction of the duration of the rf pulse.

    [0134] FIG. 10 the time-dependency of voltages measured at the first and second source during a method according the state of the art performed in the same apparatus as the variants of the method described in the context of FIG. 9, but with a different timing scheme. Here, rf power is permanently on, whereas the DC power supply is pulsed. Accordingly, the following sequence of states results: S3.fwdarw.S2.fwdarw.S3.fwdarw.S2.fwdarw. . . . . The time course of the DC self-bias voltage is strongly affected by the pulsing on the DC power supply. The DC self-bias voltage even changes sign over time. In this method, the operating parameters of the second plasma source are close to their desired values only for a small fraction of the total operating time.

    Typical Operating Parameters

    [0135] Typical values of operating parameters, which may be combined with various of the embodiments as presented above, may be: pulse repetition rates of 50-100 kHz, duty cycles for pulsed DC power supplies 50% to 80% at power levels of 1 kW to 10 kW (or 1.5 to 15 W per cm2 of target surface area). For RF sources operating in a frequency range of 2-60 MHz (most common at 13.56 MHz) and at a power level of 1-10 kW, at voltage levels of 50-1000 V. Typical deposition rates are in the range of 0.1-10 nm/s. Typical times per deposition step range between a few seconds to about 60 minutes.

    EXAMPLE EMBODIMENT

    [0136] In the following, an example embodiment combining specific features of several of the above-mentioned embodiments is discussed in further detail. Although more specific details are given for this example embodiment, reference signs match with FIG. 8.

    [0137] A substrate 30 is placed on a carrier table, which forms a substrate holder 31 and which rotates underneath two sources, the first plasma source and the second plasma source. This way, the substrate is repetitively exposed to the coating environment of the first plasma source 12 and the plasma treatment environment of the second plasma source 22. A typical distance of 75 mm between the carrier table top surface and the target surface within the first source is used. A silicon target is installed at the first plasma source. The target rotates, and a typical magnetron type magnetic field configuration is generated by a static system of permanent magnets, located behind the back side of the target, for enhanced process rates and target material utilization. The method described here works no matter whether at the first plasma source 12 the target rotates, and the magnet arrangement does not, or the magnet arrangement rotates, and the target does not. A mixture of Argon gas and oxygen molecular gas is let into the vacuum chamber at the first plasma source. The oxygen gas flow is controlled by a feedback loop that regulates the oxygen gas flow to keep the potential of the target at a preset level. Typical gas flows are 35-45 sccm of Ar and 30-45 sccm of oxygen at the first plasma source. The first plasma source is operated with 5-6 kW of DC power, pulsed at a frequency of 80 kHz and a duty cycle of 68% (12.5 us period, 8.5 us on time, 4 us off time). The resulting layer at the substrate surface consists of silicon dioxide. The layer is deposited at a typical deposition rate of 0.25-0.5 nm/s. In addition, 10 sccm of oxygen gas is introduced at the second plasma source 22. The second plasma source is operated using RF power at 13.56 MHz with power levels of 1-2 kW (for comparison in state-of-the-art situation, continuously on) or 2-6 kW (according to the invention, pulsed operation at 80 kHz, 5 us on time). The pulsing cycles of the DC power at the first plasma source 12 and of the second plasma source driven by RF power, are synchronized, so that the relative timing of the on times of the two sources can be adjusted.

    [0138] According to the example embodiment, a capacitively coupled plasma configuration is used as the second plasma source.

    [0139] In this configuration the setup is asymmetric, i.e. the source consists of a powered surface area that is relatively large compared with the grounded surface area of the extraction grid (grey dashed line in FIG. 8), where the plasma flows from the second plasma source towards the substrate surface. Due to the asymmetric configuration, a DC self-bias (typically 100-1000V) develops and the plasma potential inside the plasma source is high, leading to sufficiently high energies of a fraction of the ions reaching the substrate underneath the source to facilitate the desired surface treatment (e.g. smoothening of the growing layer). A static magnetic field generated by DC powered coils around the plasma source or by permanent magnets may be used to further optimize the plasma potential and the spatial distribution of the active species leaving the plasma source at the extraction grid.

    [0140] In this example embodiment the lower plasma supply power delivered to the second plasma corresponds typically to <10% of the higher plasma supply power delivered to the second plasma. It may be as low as zero power. This lower plasma supply power delivered to the second plasma is activated during the on time of the pulsed DC power at the first plasma source. A relatively high power level (typically 1-10 kW, high compared to a known continuous operation of a plasma source with a constant power level) is applied during the off time of the pulsed DC power at the first plasma source.

    [0141] If, contrary to the method of the present invention, the first plasma source is pulsed, and the second plasma source is continuously active, a coupling between the plasmas generated at the two sources during the on time of the first plasma source disturbs the potential distribution. As a result, the RF period averaged potential at the powered electrode of the second plasma source, also known as the DC self-bias voltage, is strongly reduced (it even becomes negative in the present example, see FIG. 10), and the process does not run under the desired conditions, i.e. the energy of the ions reaching the substrate is insufficient, and charged species of the plasma at the first plasma source are drawn towards the second plasma source.

    [0142] According to the example embodiment, however, the first source is kept in pulsed operation and the second plasma source is changed to be in pulsed operation, as well (see FIG. 9.a) and FIG. 9.b).

    [0143] The timing of the DC pulse at the first plasma source and the RF pulse at second plasma source is synchronized, so that a fixed, but adjustable delay between an on time of the first and second plasma source is defined. A synchronization may be accomplished as follows. The DC power supply generates a TTL high signal at the beginning of T.sub.DCon, which triggers a synchronization unit 32. In the synchronization unit, a TTL high signal with length T.sub.RFon is send to the RF power supply after an adjustable delay T.sub.delay. Due to the maximum off time of the specific DC generator at the first plasma source (4 us) used here and the minimum on time of the specific RF generator at the second plasma source (5 us) used here, there must be an overlap. The power timing scheme as shown in FIG. 6 without an overlap and as shown in FIG. 7 with an overlap, may be applied.

    [0144] Typical values for the timing and the power levels at the first plasma source are: 1-15 us for the on time (TDCon) and for the off time (TDCoff), pulse repetition cycles (TDCon+TDCoff) of 2-20 us (or pulse frequencies of 50-100 kHz), duty cycles (TDCon/(TDCon+TDCoff)) of 50-80%, power levels of 1-10 kW (or 1.5-15 W per cm2 of target surface area). The delay Tdelay is set equal to TDCon (no overlap at the beginning of the RF pulse) or slightly (about 1 us) shorter than TDCon (overlap at the beginning of the RF pulse). The RF pulse is on (TRFon) for a time equal or very similar to TDCoff and off (TRFoff) for a time equal or very similar to TDCon. The method considered here works with and without such an overlap, and it also works without an overlap and using a time TRFon smaller than TDCoff, so that none of the two sources is powered for a small fraction of the time period. An overlap of the pulses may actually facilitate a faster plasma ignition at the second plasma source.

    [0145] With the method according to the example embodiment, the following improvements could be observed. For a SiO2 layer (with a thickness of 600 nm) deposited in the present example, we observe a reduction of the surface roughness (Rq value) from about 1.8 nm with the second plasma source turned off to below 0.5 nm with the plasma source turned on at the optimum conditions, i.e. pulsed at 6 kW RF power with a short overlap of the two pulses. The deposition rate was between 0.4 nm/s and 0.5 nm/s on 16 wafers of 200 mm diameter placed on the rotating carrier table. At the same time, the optical losses are reduced from more than 6% to less than <0.1% (measured at a wave-length of 300 nm).

    [0146] The described method may be used in one or multiple process steps of an optical coating consisting of one or multiple layers of different materials (using multiple sputter source in a process sequence). The method may be applied to the vacuum deposition of dielectric coatings for various applications, such as gap filling on high aspect ratio structured wafers in semiconductor manufacturing. If the first plasma source is operated using DC pulsed and superimposed RF power or even using only RF power, the same method described above can be applied. Then, the pulsing scheme of the RF power at the first plasma source must be identical with the scheme for the DC pulsed power as described above. In such cases, it is possible to use one RF power supply for each of the first plasma source and for the second plasma source. Alternatively, it is possible to use a single RF power supply and to switch the output power repetitively between the first and second plasma source (in this case without an overlap between the on times).

    [0147] The method according to the invention may be used with the first plasma source being supplied by a HiPIMS (High-power impulse magnetron sputtering) power supply, where short pulses (typically 50 us-500 us) with high power densities at the target surface (on the order of kW per cm.sup.2, two orders of magnitude above the conventional sputter deposition) are applied at low pulse frequencies (typically 100 Hz-500 Hz). As the duty cycle is small (typically below 10%), the time averaged power stays similar to the conventional sputter deposition, making HiPIMS feasible with the same source hardware. In HiPIMS sputtering, most of the sputtered material is ionized, leading to a metal based plasma and improved control of substrate surface bombardment with kinetic metal particles. This allows for a tuning of film characteristics, such as smoothness, density and structure. Using the operating scheme with HiPIMS utilized at the first plasma source, the duty cycle at the second plasma source may be increased according to the reduced duty cycle of the first plasma source, and/or the times of low power supplied to both sources (state S4) may be increased. The lower pulse frequencies in combination with longer duty cycles at plasma source 2 leads to a stable DC self-bias over a larger fraction of the duty cycle, thereby allowing for an advanced tailoring of the ion energy distribution/ion current distribution at lower peak rf power values. This may result in improved surface treatment conditions and improved source hardware lifetime.

    LIST OF REFERENCE SIGNS

    [0148] 1 Vacuum chamber [0149] 2 plasma environment [0150] 10 first treatment area [0151] 11 first plasma (of a material deposition source) [0152] 12 first plasma source [0153] 13 first electric plasma supply arrangement [0154] 20 second treatment area [0155] 21 second plasma (of a non-deposition source) [0156] 22 second plasma source [0157] 23 second electric plasma supply arrangement [0158] 30 substrate [0159] 31 substrate holder [0160] 32 synchronization unit [0161] 33 weak plasma [0162] P1 power supplied to first plasma [0163] P2 power supplied to second plasma [0164] S1 first state [0165] S2 second state [0166] S3 third state [0167] S4 fourth state [0168] ST transitional power state [0169] T period [0170] T.sub.DCon time during which DC power supply is on [0171] T.sub.DCoff time during which DC power supply is off [0172] T.sub.RFon time during which RF power supply is on [0173] T.sub.RFoff time during which RF power supply is off [0174] t1 first time span during which the first state prevails [0175] t2 second time span during which the second state prevails [0176] t3 third time span during which the third state prevails [0177] t4 fourth time span during which the fourth state prevails [0178] time lag [0179] HIGH/LOW indication of plasma supply power state [0180] ON/OFF indication of plasma supply power state [0181] U.sub.target target voltage [0182] U.sub.rf rf voltage [0183] U.sub.DC self-bias self-bias voltage