JP7445071B1 - Film forming equipment and film forming method using magnetron sputtering method - Google Patents

Abstract

An object of the present invention is to form a reaction film with good reproducibility by stabilizing the sputtering rate in a film forming apparatus and film forming method using magnetron sputtering. SOLUTION: According to a magnetron sputtering apparatus 10 according to the present invention, an inert gas as a discharge gas is supplied into a vacuum chamber 12 at a constant flow rate Qd. In addition, the angle θ of the conductance valve 22 is kept constant, that is, the effective exhaust speed at the exhaust port 14 of the vacuum chamber 12 is kept constant. Furthermore, the sputtering power Es is kept constant, and the arc discharge power Ed is kept constant. Then, the flow rate Qr of the reactive gas introduced into the vacuum chamber 12 is automatically controlled so that the pressure P within the vacuum chamber 12 is constant. This stabilizes the sputtering rate. In addition, the substrate bias voltage Vb is kept constant. This improves the reproducibility of the reaction film, particularly the reproducibility of film thickness and color tone. [Selection diagram] Figure 1

Description

The present invention relates to a film forming apparatus and a film forming method using a magnetron sputtering method, and more particularly to a film forming apparatus and a film forming method using a so-called reactive magnetron sputtering method for forming a reactive film (compound film) on an object to be processed.

An example of this type of technology is disclosed in Patent Document 1. According to the technique disclosed in Patent Document 1, a magnetron sputtering cathode (magnetron sword) having a target that is a material for a reaction film is provided in a vacuum chamber set to a ground potential. Additionally, a workpiece is provided in the vacuum chamber so as to face the sputtering surface of the target. Further, an earth shield is provided so as to surround the outer periphery of the magnetron sputter cathode with (only) the sputtered surface of the target of the magnetron sputter cathode exposed. This earth shield is also at ground potential. Then, the inside of the vacuum chamber is evacuated by a vacuum pump serving as an exhaust means, and an inert gas as a discharge gas is introduced into the vacuum chamber. In this state, the vacuum chamber is used as an anode, strictly speaking, the earth shield is used as an anode, and the magnetron sputtering cathode is used as a cathode, and sputtering power is supplied to both of them. As a result, particles of the inert gas are discharged and magnetron plasma is generated so as to stick to the surface of the target to be sputtered. The discharge mode of this magnetron plasma is a glow discharge of high voltage and small current. Furthermore, the reason why the magnetron plasma sticks to the sputtered surface of the target is due to the action of the magnetic field formed by the magnet as the magnetic field forming means provided in the magnetron sputtering cathode.

When particles of the inert gas, particularly ions, in the magnetron plasma collide with the sputtering surface of the target, particles constituting the target are ejected from the sputtering surface, that is, sputtered. At this time, the sputtered area, which is the area to be sputtered, is limited to the sputtered surface of the target, and in other words, the above-mentioned earth shield is provided to make this possible. Furthermore, a reactive gas that becomes a material for the reaction membrane is introduced into the vacuum chamber. The particles of this reactive gas are decomposed by the magnetron plasma. Then, the vacuum chamber is used as an anode and the object to be treated is used as a cathode, and bias power is supplied to both of them. As a result, inert gas particles, reactive gas particles, and sputtered particles sputtered from the sputtered surface of the target are accelerated toward the object to be processed. In particular, when the reactive gas particles and sputtered particles adhere to the surface of the workpiece and react with each other, a reaction occurs on the surface of the workpiece, which is composed of the reactive gas particles and the sputtered particles. A film is formed. In addition, the density of the reaction film is improved by the bombardment effect of the inert gas particles on the surface of the object to be treated.

In addition, a cathode filament is provided between the sputtering surface of the target and the object to be processed in the vacuum chamber. Then, by supplying heating power (thermionic emission amount power) to the cathode filament, the cathode filament is heated and thermionic electrons are emitted from the cathode filament. Further, the vacuum chamber is used as an anode, strictly speaking, the earth shield is used as an anode, and the cathode filament is used as a cathode, and arc discharge power (discharge power) is supplied to both of them. Then, thermionic electrons emitted from the cathode filament are accelerated toward the earth shield, and the accelerated thermionic electrons collide with inert gas particles, reactive gas particles, and sputtered particles. Since the aforementioned magnetic field is formed around the cathode filament, the electrons accelerated from the cathode filament toward the earth shield undergo spiral motion (cycloid motion or trochoid motion) due to the action of the magnetic field. This increases the frequency at which the thermoelectrons collide with inert gas particles, reactive gas particles, and sputtered particles, inducing a low-voltage, high-current arc discharge around the cathode filament. That is, in addition to magnetron plasma caused by glow discharge, extremely high-density plasma caused by arc discharge is generated around the cathode filament, that is, generated between the sputtering surface of the target and the object to be processed.

Therefore, sputtered particles sputtered from the surface of the target to be sputtered pass through an extremely high-density plasma space on the way to the object to be processed. As a result, the sputtered particles are activated, have a higher energy than at least the ground state, and are particularly efficiently ionized. Similarly, reactive gas particles are also activated and efficiently ionized. At the same time, the inert gas particles are also activated and efficiently ionized. This improvement in the ionization rate increases the amount of ions incident on the surface of the object to be treated, thereby increasing the hardness of the reaction film formed on the surface of the object to be treated. Further, since the mutual bonding force between the sputtered particles and the particles of the reactive gas increases, the reaction film can be made denser.

JP2017-66483A

By the way, although it is not specified in Patent Document 1, the above-mentioned exhaust means is provided in the middle of the piping (exhaust route) connecting the exhaust port of the vacuum chamber and the intake port of the vacuum pump, in addition to the vacuum pump. It has a conductance valve. This conductance valve includes a plurality of elongated plate-shaped blade members (louvers) arranged in parallel like blinds, and controls the effective pumping speed at the exhaust port of the vacuum chamber by the angle θ of the blade members. In the technology disclosed in Patent Document 1, the angle θ of the blade member of the conductance valve is automatically controlled (feedback control) using the pressure P in the vacuum chamber as a parameter so that the pressure P in the vacuum chamber is constant. ) to be done.

However, the technique disclosed in Patent Document 1 has a problem in that reproducibility of the reaction film cannot be obtained, particularly reproducibility of the color tone and film thickness of the reaction film. This is because the sputtering speed (per unit time and per unit area) of the sputtering surface of the target is determined each time a film formation process for forming a reaction film is performed, so to speak, for each batch, and as the batch is repeated. This is the number of particles sputtered per unit time, and is sometimes called the "sputter evaporation rate," but since this cannot be directly measured, it is conventionally calculated as the mass of particles sputtered per unit time (g/ It is presumed that this is caused by a change in (expressed as min).

For example, the color tone of a reaction film is affected by the composition of the reaction film. Taking a titanium nitride (TiN) film as an example of a reaction film, based on past experience, the color tone of a titanium nitride film with a stoichiometric composition is golden, but when it deviates from the stoichiometric composition, for example, titanium atoms The more the number of titanium atoms than nitrogen, the more the golden tone tends to fade, and when the number of titanium atoms is twice that of nitrogen (Ti ₂ N), the tone becomes metallic (silver). When observing the sputtering surface of the target after the titanium nitride film formation process, it is found that the titanium nitride film is attached to the non-erosion area of the sputtering surface, and the titanium nitride film is attached to the erosion area. Not attached. The area near the boundary between the erosion area and the non-erosion area differs from batch to batch; for example, it may be a pale gold color or it may be a metallic color. As mentioned above, magnetron plasma is generated in the vacuum chamber so that it sticks to the surface of the target to be sputtered, and in addition to this, extremely high-density plasma is generated due to arc discharge. The space where the surface to be sputtered is exposed is extremely chemically active. Therefore, when looking at the sputtered surface of the target microscopically, nitrogen enters the vicinity of the boundary between the erosion region and the non-erosion region, which means that the vicinity of the boundary is nitrided (or a phenomenon similar to nitridation occurs). It can be considered that As the degree of nitridation differs from batch to batch, the sputtering speed changes for each batch, even over a relatively short period (short-term span), which causes the composition ratio of the titanium nitride film to change. It is presumed that this will result in a change in the color tone of the titanium nitride film, and that the reproducibility of the color tone of the titanium nitride film will no longer be obtained.

Further, when the sputtering speed changes, the formation speed (film formation speed) of the reaction film changes, and the film thickness of the reaction film changes. In particular, as batches are repeated (long-term span), the sputtering rate decreases, the reaction film formation rate decreases, and the thickness of the reaction film tends to decrease. For example, if the thickness of the target is 8 mm and the target is used until the depth of the erosion area reaches 7 mm, the surface area of the erosion area of the target at the end of use is the same as the erosion area of the target at the beginning of use. It is approximately 1.1 times larger than the surface area of the region. Therefore, when the sputtering power is constant, the power density of the sputtering power acting on the target at the end of use is reduced by about 10% compared to the power density of the sputtering power acting on the target at the beginning of use. Therefore, as described above, as batches are repeated, the sputtering rate decreases, which in turn decreases the deposition rate of the reaction film, and as a result, the thickness of the reaction film tends to decrease.

Therefore, the present invention aims to provide a new technology that can improve the reproducibility of a reactive film by stabilizing the sputtering speed in a film forming apparatus and a film forming method using reactive magnetron sputtering. , the purpose.

To achieve this object, the present invention includes a first invention relating to a film forming apparatus using reactive magnetron sputtering, and a second invention relating to a film forming method using reactive magnetron sputtering.

Among these, the first invention relating to a film forming apparatus using reactive magnetron sputtering method includes a vacuum chamber, a magnetron sputtering cathode, an earth shield, an exhaust means, an inert gas introduction means, a sputtering power supply means, a reactive gas introduction means, a bias It includes a power supply means, a cathode filament, a heating power supply means, an arc discharge power supply means, a heating power control means, and a reactive gas flow rate control means. Specifically, the vacuum chamber is grounded, that is, has a ground potential, and the object to be processed is housed inside the vacuum chamber. The magnetron sputtering cathode has a target that is a material for the reaction film, and is provided inside the vacuum chamber so that the surface of the target to be sputtered faces the object to be processed. The earth shield is provided so as to surround the outer periphery of the magnetron sputter cathode with (only) the target sputtered region of the magnetron sputter cathode exposed. Further, the earth shield is grounded, that is, has a ground potential. The evacuation means includes a vacuum pump and a conductance valve. The vacuum pump evacuates the inside of the vacuum chamber through the exhaust port of the vacuum chamber. The conductance valve then controls the effective pumping speed at the exhaust port of the vacuum chamber. The inert gas introducing means introduces an inert gas as a discharge gas into the vacuum chamber at a constant flow rate. The sputtering power supply means uses the vacuum chamber as an anode, strictly speaking, the earth shield as an anode, and the magnetron sputtering cathode as a cathode, and supplies sputtering power to both of them. As a result, particles of the inert gas are discharged and magnetron plasma is generated so as to stick to the surface of the target to be sputtered. The discharge mode of this magnetron plasma is a glow discharge of high voltage and small current. Further, the reason why the magnetron plasma sticks to the sputtering surface of the target is due to the action of the magnetic field formed by the magnetic field forming means provided in the magnetron sputtering cathode.

When particles of the inert gas, particularly ions, in the magnetron plasma collide with the sputtering surface of the target, particles constituting the target are ejected from the sputtering surface, that is, sputtered. At this time, the sputtered area, which is the area to be sputtered, is limited to the sputtered surface of the target, and the above-mentioned earth shield is provided to make this possible. Furthermore, the reactive gas introduction means introduces a reactive gas that becomes a material for the reaction film into the vacuum chamber. The particles of this reactive gas are decomposed by the magnetron plasma. The bias power supply means uses the vacuum chamber as an anode and the object to be treated as a cathode, and supplies bias power with a constant predetermined component to both of them. As a result, the inert gas particles, the reactive gas particles, and the sputtered particles sputtered from the sputtered surface of the target are accelerated toward the object to be processed at a constant acceleration. In particular, when the reactive gas particles and sputtered particles adhere to the surface of the workpiece and react with each other, a reaction occurs on the surface of the workpiece with the reactive gas particles and sputtered particles as components. A film is formed. In addition, the density of the reaction film is improved by the bombardment effect of the inert gas particles on the surface of the object to be treated. Note that the predetermined component of bias power is, for example, an average value or an effective value of a bias voltage that is a voltage component of the bias power.

In addition, a cathode filament is provided between the target surface to be sputtered and the object to be processed inside the vacuum chamber. The heating power supply means supplies heating power to the cathode filament, thereby heating the cathode filament and emitting thermoelectrons from the cathode filament. Further, the arc discharge power supply means uses the vacuum chamber as an anode, strictly speaking, the earth shield as an anode, and the cathode filament as a cathode, and supplies arc discharge power to both of them. Then, thermionic electrons emitted from the cathode filament are accelerated toward the earth shield, and the accelerated thermionic electrons collide with inert gas particles, reactive gas particles, and sputtered particles. Since the above-mentioned magnetic field is formed around the cathode filament, the electrons accelerated from the cathode filament toward the earth shield undergo spiral motion due to the action of the magnetic field. This increases the frequency at which the thermoelectrons collide with inert gas particles, reactive gas particles, and sputtered particles, inducing a low-voltage, high-current arc discharge around the cathode filament. That is, in addition to magnetron plasma caused by glow discharge, extremely high-density plasma caused by arc discharge is generated around the cathode filament, that is, generated between the sputtering surface of the target and the object to be processed.

Therefore, sputtered particles sputtered from the surface of the target to be sputtered pass through an extremely high-density plasma space on the way to the object to be processed. Thereby, the sputtered particles are activated and efficiently ionized. Similarly, reactive gas particles are also activated and efficiently ionized. At the same time, the inert gas particles are also activated and efficiently ionized. This improvement in the ionization rate increases the amount of ions incident on the surface of the object to be treated, thereby increasing the hardness of the reaction film formed on the surface of the object to be treated. Further, since the mutual bonding force between the sputtered particles and the particles of the reactive gas increases, the reaction film can be made denser.

Then, the heating power control means controls such that the current component of the arc discharge power, so to speak, the arc discharge, is kept constant while the voltage component of the arc discharge power, so to speak, the arc discharge voltage is kept constant. Controls heating power. This stabilizes the density of plasma caused by arc discharge.

Further, the reactive gas flow rate control means controls the flow rate of the reactive gas into the vacuum chamber, and controls, for example, the reactive gas introduction means so that the pressure inside the vacuum chamber is constant. At this time, sputtering power is kept constant. In addition, the effective pumping speed at the exhaust port of the vacuum chamber by the vacuum pump is kept constant by the conductance valve. This keeps the sputtering speed constant.

Note that in the first invention, a plurality of units may be provided. The unit referred to here is a so-called plasma generation source having a magnetron sputtering cathode, an earth shield, a sputtering power supply means, a cathode filament, a heating power supply means, an arc discharge power supply means, and a heating power control means. That is, a plurality of units serving as plasma generation sources may be provided.

In addition, the bias power depends on whether the object to be processed is a conductive (in other words, insulating) material, and whether the reaction film is a conductive (in other words, insulating) film. Depending on the situation, either DC power, asymmetric bipolar pulse power, or high frequency power is selectively employed.

A second aspect of the present invention relating to a film forming method using reactive magnetron sputtering includes a step of installing an object, an evacuation step, an inert gas introduction step, a sputtering power supply step, a reactive gas introduction step, and a bias power supply step. The method includes a heating power supply step, an arc discharge power supply step, a heating power control step, and a reactive gas flow rate control step. In the processing object installation step, the processing object is installed in a vacuum chamber provided with a magnetron sputtering cathode having a target that is a material for a reaction film, so as to face the sputtering surface of the target. Here, the vacuum chamber is grounded, that is, at ground potential. Then, an earth shield is provided so as to surround the outer periphery of the magnetron sputter cathode with (only) the sputtered surface of the target of the magnetron sputter cathode exposed. This earth shield is also grounded or at ground potential. In the evacuation step, the inside of the vacuum chamber is evacuated by a vacuum pump through the exhaust port of the vacuum chamber. In addition, the effective exhaust speed at the exhaust port of the vacuum chamber is controlled by the conductance valve. In the inert gas introduction step, an inert gas as a discharge gas is introduced into the vacuum chamber at a constant flow rate. In the sputtering power supply step, the vacuum chamber is used as an anode, strictly speaking, the earth shield is used as an anode, and the magnetron sputtering cathode is used as a cathode, and sputtering power is supplied to both of them. As a result, particles of the inert gas are discharged and magnetron plasma is generated so as to stick to the surface of the target to be sputtered. The discharge mode of this magnetron plasma is a glow discharge of high voltage and small current. Further, the reason why the magnetron plasma sticks to the sputtering surface of the target is due to the action of the magnetic field formed by the magnetic field forming means provided in the magnetron sputtering cathode.

When particles of the inert gas, particularly ions, in the magnetron plasma collide with the sputtering surface of the target, particles constituting the target are ejected from the sputtering surface, that is, sputtered. At this time, the sputtered area, which is the area to be sputtered, is limited to the sputtered surface of the target, and in other words, the above-mentioned earth shield is provided to make this possible. Furthermore, in the reactive gas introduction step, a reactive gas that becomes a material for the reaction film is introduced into the vacuum chamber. The particles of this reactive gas are decomposed by the magnetron plasma. In the bias power supply step, the vacuum chamber is used as an anode and the object to be treated is used as a cathode, and bias power with a constant predetermined component is supplied to both of them. As a result, the inert gas particles, the reactive gas particles, and the sputtered particles sputtered from the sputtered surface of the target are accelerated toward the object to be processed at a constant acceleration. In particular, when the reactive gas particles and sputtered particles adhere to the surface of the workpiece and react with each other, a reaction occurs on the surface of the workpiece, which is composed of the reactive gas particles and the sputtered particles. A film is formed. In addition, the density of the reaction film is improved by the bombardment effect of the inert gas particles on the surface of the object to be treated. Note that the predetermined component of bias power is, for example, an average value or an effective value of a bias voltage that is a voltage component of the bias power.

Further, in the heating power supply step, heating power is supplied to the cathode filament. The cathode filament is provided inside the vacuum chamber between the target surface to be sputtered and the object to be processed. The cathode filament is heated by being supplied with heating power and emits thermoelectrons. In the step of supplying electric power for arc discharge, the vacuum chamber is used as an anode, strictly speaking, the earth shield is used as an anode, and the cathode filament is used as a cathode, and electric power for arc discharge is supplied to both of them. Then, thermionic electrons emitted from the cathode filament are accelerated toward the earth shield, and the accelerated thermionic electrons collide with inert gas particles, reactive gas particles, and sputtered particles. Since the above-mentioned magnetic field is formed around the cathode filament, the electrons accelerated from the cathode filament toward the earth shield undergo spiral motion due to the action of the magnetic field. This increases the frequency with which the thermoelectrons collide with inert gas particles, reactive gas particles, and sputtered particles, inducing a low-voltage, large-current arc discharge around the filament. That is, in addition to magnetron plasma caused by glow discharge, extremely high-density plasma caused by arc discharge is generated around the cathode filament, that is, generated between the sputtering surface of the target and the object to be processed.

Therefore, sputtered particles sputtered from the surface of the target to be sputtered pass through an extremely high-density plasma space on the way to the object to be processed. Thereby, the sputtered particles are activated and efficiently ionized. Similarly, reactive gas particles are also activated and efficiently ionized. At the same time, the inert gas particles are also activated and efficiently ionized. This improvement in the ionization rate increases the amount of ions incident on the surface of the object to be treated, thereby increasing the hardness of the reaction film formed on the surface of the object to be treated. Further, since the mutual bonding force between the sputtered particles and the particles of the reactive gas increases, the reaction film can be made denser.

Then, in the heating power control step, while the voltage component of the arc discharge power, that is, the arc discharge voltage is kept constant, the current component of the arc discharge power, that is, the arc discharge, is kept constant. Heating power is controlled. This stabilizes the density of plasma caused by arc discharge.

Furthermore, in the reactive gas flow rate control step, the flow rate of the reactive gas into the vacuum chamber is controlled so that the pressure inside the vacuum chamber is constant. At this time, sputtering power is kept constant. In addition, the effective pumping speed at the exhaust port of the vacuum chamber by the vacuum pump is kept constant by the conductance valve. This keeps the sputtering speed constant.

According to the present invention, sputtering speed is stabilized. This improves the reproducibility of the reaction film, particularly the reproducibility of film thickness and color tone.

FIG. 1 is a diagram showing a schematic configuration of a magnetron sputtering apparatus according to an embodiment of the present invention. FIG. 2 is a top view of the inside of a magnetron sputtering apparatus according to an embodiment of the present invention. FIG. 3 is a diagram showing a schematic configuration of a magnetron sputtering cathode in one embodiment of the present invention. FIG. 4 is a diagram showing the mutual positional relationship between the magnetron sputtering cathode and the cathode filament in one embodiment of the present invention. FIG. 5 is a diagram showing experimental results using a conventional technique as a comparison target for an embodiment of the present invention. FIG. 6 is a diagram showing experimental results according to an embodiment of the present invention. FIG. 7 is a diagram showing film forming conditions in another experiment according to an embodiment of the present invention. FIG. 8 is a diagram showing the properties of a titanium nitride film formed in another experiment according to an embodiment of the present invention. FIG. 9 is a diagram illustrating a comparison of control procedures according to an embodiment of the present invention and a conventional technique for comparison. FIG. 10 is a diagram showing a comparison of the reproducibility of titanium nitride films formed by an example of the present invention and a conventional technique for comparison. FIG. 11 is a diagram showing an expanded example of an embodiment of the present invention.

An embodiment of the present invention will be described with reference to FIGS. 1 to 10.

As shown in FIGS. 1 and 2, the magnetron sputtering apparatus 10 according to this embodiment includes a generally cylindrical vacuum chamber 12 with both ends closed. The vacuum chamber 12 is installed with the ends of its generally cylindrical shape facing upward and downward, that is, with the central axis Xa of the cylindrical shape extending in the vertical direction. The inner diameter of the vacuum chamber 12 is, for example, about 1100 mm, and the height inside the vacuum chamber 12 is, for example, about 800 mm. The shape and dimensions of the vacuum chamber 12 are merely examples, and are appropriately determined depending on various circumstances such as the size, shape, and number of objects 100 to be processed, which will be described later. The vacuum chamber 12 itself is made of a metal with high corrosion resistance and heat resistance, for example, stainless steel such as SUS304, and its wall is grounded, that is, the potential of the wall is at the ground potential, which is a reference potential. It is said that

An exhaust port 14 is provided at an appropriate position on the wall of the vacuum chamber 12, for example at an appropriate position on the wall forming the side surface. The exhaust port 14 is connected to a vacuum pump 18 via an exhaust pipe 16 outside the vacuum chamber 12, and more precisely, to an inlet port (not shown) of the vacuum pump 18. Note that the vacuum pump 18 may be, for example, a diffusion pump, a turbomolecular pump, a cryopump, or the like, but is not limited thereto. Further, in FIG. 2, some elements including the exhaust port 14 are omitted for ease of viewing.

In the middle of the exhaust pipe 16, a main valve, ie, an on-off valve 20, is provided as an opening/closing means for opening and closing the inside of the exhaust pipe 16. Additionally, a conductance valve 22 is provided in the middle of the exhaust pipe 16 to control the effective exhaust speed at the exhaust port 14 of the vacuum chamber 12 by the vacuum pump 18. These vacuum pump 18 and conductance valve 22 constitute exhaust means. Although detailed illustrations are omitted, the conductance valve 22 includes a plurality of elongated plate-shaped blade members arranged in parallel like a blind. Control pumping speed. For example, the smaller the angle θ of the conductance valve 22 (blade member), the greater the conductance, and when the angle θ is 0 degrees, the valve is fully open. On the contrary, the larger the angle θ of the conductance valve 22, the smaller the conductance becomes, and when the angle θ is 90 degrees, the conductance valve 22 is fully closed. This configuration of the conductance valve 22 is just an example, and the configuration is not limited thereto.

Furthermore, magnetron sputtering is applied to an appropriate position inside the wall forming the side surface of the vacuum chamber 12 (the right position in FIGS. 1 and 2) in a state that is electrically insulated from the wall of the vacuum chamber 12. A cathode 24 is provided. Referring also to FIG. 3, the magnetron sputtering cathode 24 includes a roughly rectangular flat target 242 that is a material for the coating, and a magnet unit 244 provided on the back side, which is one main surface of the target 242. . The magnet unit 244 includes a permanent magnet 246 as an example of a magnetic field forming means, and a housing 248 that accommodates the permanent magnet 246. Further, the permanent magnet 246 has a generally rectangular frame-shaped one magnetic pole, for example, an N pole 246a, which is provided along the periphery of the target 242 while being in close contact with the back surface of the target 242, and a N pole 246a, which is attached to the inside of this N pole 246a, and The target 242 has an S pole 246b as the other magnetic pole in the form of a long and narrow protrusion, which is provided so as to be in close contact with the back surface of the target 242 and extend along the longitudinal direction of the target 242. Note that the dimensions of the target 242 are, for example, 457 mm in the longitudinal direction (length dimension), 127 mm in the lateral direction (width dimension), and 8 mm in the thickness direction (thickness dimension). Furthermore, a substantially rectangular groove-shaped gap 246c is provided between the N pole 246a and the S pole 246b of the permanent magnet 246. The housing 248 is provided with a suitable water cooling mechanism (not shown) as a cooling means for cooling the entire magnetron sputtering cathode 24 including the housing 248.

The magnetron sputtering cathode 24 is configured such that the sputtered surface, which is the other main surface (front surface) of the target 242, is directed toward the central axis Xa of the vacuum chamber 12, and the longitudinal direction of the target 242 is aligned along the central axis Xa of the vacuum chamber 12. , that is, it is provided so as to extend along the vertical direction. The magnetron sputtering cathode 24 is covered with an earth shield 26 except for the surface of the target 242 to be sputtered. In other words, the earth shield 26 is provided so as to surround the outer periphery of the magnetron sputter cathode 24 with (only) the sputtered surface of the target 242 of the magnetron sputter cathode 24 exposed. This earth shield 26 is made of a metal with high corrosion resistance and heat resistance, for example, stainless steel such as SUS304. This earth shield 26 is electrically insulated from the magnetron sputtering cathode 24 and electrically connected to the wall of the vacuum chamber 12, that is, at ground potential.

To explain in more detail, the earth shield 26 is provided so as to surround the outer periphery of the magnetron sputter cathode 24 with the sputtered surface of the target 242 of the magnetron sputter cathode 24 exposed as described above. The distance between the outer surface of the sputter cathode 24 and the inner surface of the earth shield 26 is, for example, about 2 mm, which is quite close. The earth shield 26 is substantially flush with the sputtered surface of the target 242 and has a portion that protrudes away from the sputtered surface (that is, along the direction in which the central axis Xa of the vacuum chamber 12 extends). However, the protruding dimension of this protruding portion is, for example, 20 mm to 30 mm. Further, the earth shield 26 is electrically and mechanically connected to the wall of the vacuum chamber 12, but directly (vacuum (separately from tank 12).

As shown in FIG. 2, a portion 12a of the wall of the vacuum chamber 12 where the magnetron sputter cathode 24 and the earth shield 26 are provided has a structure suitable for the magnetron sputter cathode 24 and the earth shield 26. It is said that It is desirable that this portion 12a be able to be opened and closed like a sliding door or a hinged door in consideration of workability during maintenance of the magnetron sputtering cathode 24 including replacement of the target 242.

Referring again to FIG. 1, the magnetron sputter cathode 24 is connected to a DC sputter power supply 28 as an example of sputter power supply means outside the vacuum chamber 12. The magnetron sputter cathode 24 is supplied with sputter power Es, which is DC power at a negative potential with respect to the ground potential, from a sputter power supply 28. In other words, the vacuum chamber 12 is used as an anode, the earth shield 26 is used as an anode, and the magnetron sputtering cathode 24 is used as a cathode, and the sputtering power Es is supplied to both of them. Note that the sputter power supply device 28 operates in a constant power mode in which the sputter power Es is operated in a constant power value, and in a constant power mode in which the sputter voltage (also referred to as "target voltage") Vs, which is a voltage component of the sputter power Es, is constant. There are three operating modes: a constant voltage mode in which the sputtering current (or "target current") Is, which is the current component of the sputtering power Es, is constant. However, here it is set to operate in constant power mode.

In addition, a cathode filament 30 as a thermionic emission means is provided in front of the magnetron sputtering cathode 24, specifically in front of the sputtered surface of the target 242. The cathode filament 30 is a linear body with a diameter of approximately 1 mm, and is made of, for example, tungsten (W), but may also be made of other materials such as molybdenum (Mo), tantalum (Ta), carbon (C), etc. It may be made of melting point metal.

Here, with reference also to FIG. 4, and especially with reference to FIG. When viewed from the central axis Xa side of the tank 12, the center of the sputtered surface of the target 242 is aligned vertically, that is, along the longitudinal direction of the target 242, in other words, the sputtered surface of the target 242 is aligned vertically. It is provided so as to extend parallel to the plane. Further, as shown in FIGS. 4(b) and 4(c), the cathode filament 30 is provided with an appropriate distance D between it and the surface of the target 242 to be sputtered. If this distance D is too small, for example, the cathode filament 30 may come into contact with the sputtered surface of the target 242 or with the ground shield 26, which is a serious disadvantage. On the other hand, if the distance D is too large, the effect of the magnetic field by the magnet unit 244 (permanent magnet 246) described above around the cathode filament 30 will be weakened, which is inconvenient for inducing arc discharge, which will be described later. For these reasons, it is appropriate for the distance D to be approximately 5 mm to 50 mm, for example, 25 mm. The length of the cathode filament 30 is equal to or greater than the length of the target 242, and strictly speaking, the length of the target 242 is equal to or greater than the length of an erosion region 242a, which will be described later. It is 500mm. Although not shown, a suitable tension is applied to both ends or one end of the cathode filament 30 to maintain the straight state of the cathode filament 30. A tension applying mechanism as a tension applying means is provided.

Referring again to FIG. 1, both ends of the cathode filament 30 are connected to a heating power supply device 32 as an example of heating power supply means outside the vacuum chamber 12. The cathode filament 30 is supplied with alternating current cathode power Ec as heating power from a heating power supply device 32 . As a result, the cathode filament 30 is heated to 2000° C. or higher, and thermoelectrons are emitted from the cathode filament 30. Note that the cathode power Ec is not limited to alternating current power, but may be direct current power.

Furthermore, one end of the cathode filament 30 is connected to an arc discharge power supply device 34 as an example of arc discharge power supply means outside the vacuum chamber 12. The cathode filament 30 is supplied with arc discharge power Ed, which is DC power at a negative potential with respect to the ground potential, from the arc discharge power supply device 34 . In other words, the vacuum chamber 12 is used as an anode, the earth shield 26 is used as an anode, and the cathode filament 30 is used as a cathode, and the electric power Ed for arc discharge is supplied to both of them. The arc discharge power supply device 34 is, for example, a constant voltage power supply device, and its maximum output voltage, that is, the maximum value of the arc discharge voltage Vd, which is the voltage component of the arc discharge power Ed, is, for example, 100V.

Further, a current detector 36 as arc discharge current detection means is provided between the arc discharge power supply device 34 and the ground. This current detector 36 detects an arc discharge current Id that is a current component of the arc discharge electric power Ed. The detected value of the arc discharge current Id by the current detector 36 is given to a heating controller 38 as an example of heating power control means.

The heating controller 38 controls the heating power supply device 32 so that the detected value of the arc discharge current Id by the current detector 36 is constant, that is, so that the arc discharge current Id is constant. The cathode power Ec is controlled via the heating power supply device 32. Thereby, the heating temperature of the cathode filament 30, that is, the amount of thermoelectrons emitted by the cathode filament 30, is adjusted as appropriate. That is, the amount of thermoelectrons emitted by the cathode filament 30 is automatically controlled (feedback control) using the arc discharge current Id as a parameter.

When paying attention to the inside of the vacuum chamber 12 from the position where the cathode filament 30 is provided, a plurality of objects to be processed 100, 100, . . . are arranged within the vacuum chamber 12. Specifically, the objects to be processed 100, 100, . . . are arranged at equal intervals along the circumferential direction of a circle centered on the central axis Xa of the vacuum chamber 12. Each object to be processed 100 has, for example, an elongated approximately cylindrical or approximately cylindrical shape, and is held so as to extend in the vertical direction, that is, in the direction along the central axis Xa of the vacuum chamber 12. It is held by a holder 40 as a means. Each holder 40 is coupled to the vicinity of the periphery of a disc-shaped revolution table 44 via a gear mechanism 42 . The center of this revolution table 44 is located on the central axis Xa of the vacuum chamber 12, and one end of a rotating shaft 46 extending along the central axis Xa of the vacuum chamber 12 is fixed to the center of the revolution table 44. Ru. The other end of the rotating shaft 46 is coupled to a shaft 48a of a motor 48 serving as rotational driving means outside the vacuum chamber 12.

That is, when the motor 48 is driven and the shaft 48a of the motor 48 rotates, for example, in the direction shown by the arrow 200 in FIG. 1, the revolution table 44 rotates in the same direction, that is, also shown by the arrow 200 in FIG. Rotate in the direction. Along with this, each of the objects 100 to be processed rotates around the central axis Xa of the vacuum chamber 12, so to speak. At the same time, due to the rotational driving force transmission action of each gear mechanism 42, each holder 40 rotates about a vertical line Xb passing through it, for example, in the direction shown by an arrow 202 in each of FIGS. 1 and 2. As the holder 40 itself rotates, the workpiece 100 also rotates in the same direction, so to speak, rotates on its own axis. Note that the diameter (PCD) of the orbital path of the object to be processed 100 is, for example, about 600 mm. The revolution speed of the object to be processed 100 (the rotation speed of the revolution table 44) is, for example, 0.5 rpm to 1 rpm. On the other hand, the rotation speed of the workpiece 100 (the rotation speed of the holder 40 itself) is, for example, 30 rpm to 60 rpm, that is, 60 times the revolution speed. Although FIGS. 1 and 2 show an example in which the number of objects to be processed 100 (holder 40 and gear mechanism 42) is 12, the number of objects to be processed 100 is not limited to this. Furthermore, the shape of the object to be processed 100 is not limited to the aforementioned elongated, generally cylindrical shape or generally cylindrical shape.

In addition, each workpiece 100 is supplied with power from a bias power supply device 50 as an example of bias power supply means located outside the vacuum chamber 12 via the holder 40, gear mechanism 42, revolution table 44, and rotating shaft 46. Substrate bias power Eb is supplied. The substrate bias voltage Vb, which is a voltage component of the substrate bias power Eb, alternates between a positive high level value with a ground potential as a reference and a negative potential low level value with the ground potential as a reference. It is a so-called asymmetric bipolar pulse power that transitions. The high level value of the substrate bias voltage Vb is constant, for example, +37V with respect to the ground potential. On the other hand, the low level value of the substrate bias voltage Vb can be set arbitrarily, and the average value (DC equivalent value) of the substrate bias voltage Vb is defined by this low level value. Furthermore, the frequency of the substrate bias power Eb can also be arbitrarily set, for example, within the range of 50 kHz to 250 kHz. The duty ratio of the substrate bias power Eb (ratio of the period during which the value of the substrate bias voltage Vb is a high level value in one cycle of the substrate bias voltage Vb) can also be arbitrarily set. Here, the frequency of the substrate bias power Eb is, for example, 100 kHz, and the duty ratio is, for example, 30%. Note that the bias power supply device 50 may have a configuration in which the effective value of the substrate bias voltage Vb is defined instead of the average value of the substrate bias voltage Vb.

Further, a suitable position inside the wall forming the side surface of the vacuum chamber 12 and an appropriate position outside the orbit of each workpiece 100, 100, . . . , for example, the central axis Xa of the vacuum chamber 12. For example, a carbon heater 52 as a temperature control means is provided at a position opposite to the position where the magnetron sputtering cathode 24 is provided (a position on the left side in FIGS. 1 and 2). This carbon heater 52 is connected to a heater heating power supply device (not shown) outside the vacuum chamber 12 . The carbon heater 52 heats the inside of the vacuum chamber 12 by receiving direct current or alternating current power for heating the heater from the heater heating power supply device, thereby heating the inside of the vacuum chamber 12 and, in particular, heating the objects 100, 100, . . . Heat. As shown in FIG. 2, the portion 12b of the wall of the vacuum chamber 12 where the carbon heater 52 is provided is similar to the portion 12a where the magnetron sputter cathode 24 and the earth shield 26 are provided. The structure is suitable for providing the carbon heater 52.

Further, as shown in FIG. 1, a gas inlet for introducing various gases such as an inert gas into the vacuum chamber 12 is provided at an appropriate position within the vacuum chamber 12, preferably at a position near the cathode filament 30. 54 is provided. A gas introduction pipe 56 is coupled to the gas introduction port 54 outside the vacuum chamber 12, and a plurality of, in this case two, branch pipes 58 and 60 are further coupled to the gas introduction pipe 56. .

One branch pipe 58 is a pipe for introducing an inert gas as a discharge gas into the vacuum chamber 12, and is connected to a supply source (not shown) of the inert gas. This branch pipe 58 for introducing inert gas has an opening/closing valve 58a as an opening/closing means for opening and closing the inside of the branch pipe 58, and a flow rate Qd of the inert gas flowing inside the branch pipe 58 is controlled. A mass flow controller 58b is provided as a flow rate control means for this purpose. The branch pipe 58 including the on-off valve 58a and the mass flow controller 58b cooperates with the gas introduction pipe 56 to constitute an example of inert gas introduction means. Further, as the inert gas, for example, argon gas is employed.

The other branch pipe 60 is a pipe for introducing a reactive gas into the vacuum chamber 12, which becomes a material for a coating, in particular, a material for a reaction film as the coating. The sources shown are supplied. This branch pipe 60 for introducing reactive gas also has an opening/closing valve 60a as an opening/closing means for opening and closing the inside of the branch pipe 60, and a flow rate Qr of the reactive gas flowing inside the branch pipe 60 is controlled. A mass flow controller 60b is provided as a flow rate control means for this purpose. The branch pipe 60 including the on-off valve 60a and the mass flow controller 60b cooperates with the gas introduction pipe 56 to constitute an example of reactive gas introduction means. Note that as the reactive gas, an appropriate gas is employed depending on the type of reaction film to be formed. For example, when a nitride film is formed as the reaction film, nitrogen (N ₂ ) gas is used as the reactive gas, and when a carbonide film is formed as the reaction film, acetylene (C _{2 )} gas is used as the reactive gas. A hydrocarbon gas such as H ₂ ) is employed. When an oxide film is formed as the reactive film, oxygen (O ₂ ) gas is used as the reactive gas.

In addition, the pressure gauge 62 is provided such that the gauge part (measuring part) 62a of the pressure gauge 62 is disposed at an appropriate position in the vacuum chamber 12, for example, at an appropriate position in the upper part of the vacuum chamber 12. It will be done. This pressure gauge 62 is a pressure measuring means for measuring the pressure inside the vacuum chamber 12, more specifically, the pressure P at the position where the gauge part 62a is arranged. Then, the measured value of the pressure P by the pressure gauge 62 is given to a gas flow rate controller 64 as an example of a reaction gas flow rate control means located outside the vacuum chamber 12. Note that as the pressure gauge 62, for example, a diaphragm vacuum gauge that can measure absolute pressure with relatively high accuracy and high resolution is employed. Furthermore, the placement position of the gauge part 62a is not limited to the upper part of the vacuum chamber 12, but may be any position where it is less affected by the plasma 300, which will be described later. The main body of the pressure gauge 62 is provided outside the vacuum chamber 12 .

The gas flow controller 64 controls the mass flow controller 60b for introducing the reactive gas so that the measured value of the pressure P by the pressure gauge 62 is constant, that is, so that the pressure P is constant. The flow rate Qr of the reactive gas introduced into the vacuum chamber 12 via the mass flow controller 60b is controlled. That is, the flow rate Qr of the reactive gas is automatically controlled (feedback control) using the pressure P in the vacuum chamber 12 as a parameter. Note that the gas flow rate controller 64 may be incorporated into the mass flow controller 60b, for example.

Further, although detailed description including illustrations will be omitted, the relationship between the cathode filament 30 and the workpiece 100 located closest to the cathode filament 30 (that is, the part closest to the cathode filament 30 on the revolution path of the workpiece 100) A shutter is provided in between. This shutter appropriately transitions between an open state in which the sputtered surface of the target 242 of the magnetron sputtering cathode 24 is exposed toward the space in which the workpiece 100 is placed, and a closed state in which it is shielded from the space.

According to the magnetron sputtering apparatus 10 having such a configuration, various reaction films can be formed on the surface of the object to be processed 100. For example, a case where a titanium nitride film is formed as the reaction film will be described. In this case, titanium with a purity of 3N or higher is used as the target 242, for example. In addition, as the reactive gas, for example, nitrogen gas having a purity of 5N or more is employed.

Then, the object to be processed 100 is first placed in the vacuum chamber 12, that is, attached to the holder 40. Then, the inside of the vacuum chamber 12 is evacuated by the vacuum pump 18 to a pressure P of about 2×10 ⁻³ Pa, and so-called evacuation is performed. Thereafter, the motor 48 is driven, and the workpiece 100 starts rotating around its axis. At the same time, heater heating power is supplied to the carbon heater 52, and the workpiece 100 is heated to about 150°C. As a result, the impurity gas contained in the object to be processed 100 is discharged, and a so-called degassing process is performed.

After this degassing process is performed for a predetermined period of time (for example, about 30 minutes to 1 hour), the supply of heater heating power to the carbon heater 52 is stopped, and then a discharge cleaning process is performed. In this discharge cleaning process, cathode power Ec is supplied to the cathode filament 30. In response to this, the cathode filament 30 is heated and thermoelectrons are emitted from the cathode filament 30. At the same time, arc discharge power Ed is supplied to the cathode filament 30. That is, the earth shield 26 is used as an anode, the cathode filament 30 is used as a cathode, and arc discharge power Ed is supplied to both of them. As a result, thermoelectrons emitted from the cathode filament 30, which is a cathode, are accelerated toward the earth shield 26, which is an anode. In this state, argon gas is introduced into the vacuum chamber 12. Then, the accelerated thermoelectrons collide with the argon gas particles, and the impact ionizes the argon gas particles, generating plasma 300.

Here, since a magnetic field is formed by the above-mentioned permanent magnet 246 in the vicinity of the sputtering surface of the target 242 including the periphery of the cathode filament 30, thermionic electrons accelerated from the cathode filament 30 toward the earth shield 26 undergoes a spiral motion due to the action of the magnetic field. As a result, the frequency with which thermoelectrons collide with argon gas particles increases, and the plasma 300 becomes denser. The discharge mode of the plasma 300 is a low voltage, large current arc discharge.

Note that arc discharge is induced, for example, even when the pressure P in the vacuum chamber 12 is in the range of 0.01 Pa to 1 Pa, that is, even when the pressure P is relatively low. Further, the arc discharge power Ed is kept constant, and more specifically, the arc discharge voltage Vd, which is the voltage component of the arc discharge power Ed, is constant, and the arc discharge power Ed, which is the current component of the arc discharge power Ed, is kept constant. The cathode power Ec is controlled so that the discharge current Id is constant, that is, the amount of thermoelectrons emitted from the cathode filament 30 is controlled. This stabilizes the density of plasma caused by arc discharge.

Substrate bias power Eb is supplied to the workpiece 100 while plasma 300 is generated by this arc discharge. That is, the vacuum chamber 12 is used as an anode, the object to be processed 100 is used as a cathode, and the substrate bias power Eb is supplied to both of them. Then, argon particles, especially argon ions, in the plasma 300 are actively incident on the surface of the object 100 to be processed. The impact removes impurities from the surface of the object to be processed 100, and a so-called discharge cleaning process is performed. Note that since the object to be treated 100 is rotating around its axis as described above, it is subjected to the discharge cleaning treatment when it is exposed to the plasma 300 during the process of this rotation.

After this discharge cleaning process is performed for a predetermined period of time (for example, about 30 minutes), a film formation process for forming a titanium nitride film is performed. For this purpose, sputtering power Es is supplied to the magnetron sputtering cathode 24. That is, the earth shield 26 is used as an anode, the magnetron sputter cathode 24 is used as a cathode, and sputtering power Es is supplied to both of them. Then, the argon ions in the plasma 300 collide with the sputtering surface of the target 242 of the magnetron sputtering cathode 24, and titanium particles are ejected from the sputtering surface of the target 242, that is, sputtered. At this time, the sputtered region, which is the region to be sputtered, is limited to the sputtered surface of the target 242, and in other words, the earth shield 26 is provided to make this possible. Furthermore, nitrogen gas is introduced into the vacuum chamber 12 as a reactive gas. The nitrogen gas particles are decomposed by the plasma 300.

As described above, titanium particles as sputtered particles sputtered from the sputtered surface of the target 242 are accelerated toward the workpiece 100 to which the substrate bias power Eb is supplied. At the same time, the decomposed nitrogen gas particles are also accelerated toward the object 100 to be processed. These titanium particles and nitrogen particles adhere to the surface of the object to be treated 100 and react with each other, so that titanium nitride, which is a reaction film containing titanium particles and nitrogen particles, is formed on the surface of the object to be treated 100. A film is formed. Furthermore, argon ions in plasma 300 are also accelerated toward workpiece 100. Due to the bombardment effect of the argon ions on the surface of the object to be processed 100, the density of the titanium nitride film formed on the surface of the object to be processed 100 is improved.

In this film-forming process, argon gas particles are also discharged by the supply of sputtering power Es, and magnetron plasma is induced by glow discharge of high voltage and small current. That is, the plasma 300 includes (combines) magnetron plasma caused by glow discharge in addition to the above-mentioned arc discharge. The magnetron plasma caused by this glow discharge is generated so as to stick to the sputtering surface of the target 242 due to the action of the magnetic field from the above-mentioned permanent magnet 246.

Then, the titanium particles, nitrogen particles, and argon particles pass through a space of extremely high-density plasma 300 while flying toward the object to be processed 100 . Thereby, the titanium particles, nitrogen particles, and argon particles are more activated and ionized more efficiently. This improvement in the ionization rate increases the amount of ions that are incident on the surface of the object to be processed 100, thereby increasing the hardness of the titanium nitride film formed on the surface of the object to be processed 100. In addition, since the mutual bonding force between titanium particles and nitrogen particles increases, the titanium nitride film can be made denser.

In this film-forming process as well, the film-forming process is performed while the object 100 to be processed is exposed to the plasma 300 during rotation and revolution. Further, the pressure P in the vacuum chamber 12 during the film forming process is controlled within a range of, for example, 0.1 Pa to 1 Pa.

Particularly, referring to FIG. 4 and paying attention to the sputtering surface of the target 242, in the vicinity of the sputtering surface where the plasma 300 is generated, the plasma 300 has a higher density. Therefore, on the sputtered surface of the target 242, a region where the plasma 300 is more dense, that is, a region that follows the gap 246c of the permanent magnet 246, is sputtered more efficiently (intensively). As a result, a roughly rectangular loop (or oval loop) sputter mark appears on the sputtering surface of the target 242, following the gap 246c of the permanent magnet 246, that is, an erosion region 242a is formed.

After this film forming process is performed for a predetermined time (the time required to form a titanium nitride film with a predetermined thickness), the introduction of argon gas and nitrogen gas into the vacuum chamber 12 is stopped. At the same time, the supply of sputtering power Es to the magnetron sputtering cathode 24 is stopped, and the supply of cathode power Ec and arc discharge power Ed to the cathode filament 30 is also stopped. As a result, plasma 300 disappears. Further, the supply of substrate bias power Eb to the workpiece 100 is stopped. Then, while the inside of the vacuum chamber 12 is maintained at a high vacuum by the vacuum pump 18, the inside of the vacuum chamber 12 is waited for to drop to a predetermined temperature (for example, 150° C.), that is, an appropriate cooling period is set. . After that, the drive of the motor 48 is stopped, and the rotation and revolution of the workpiece 100 is stopped. Then, the inside of the vacuum chamber 12 is opened to the outside, and the workpiece 100 is taken out from the vacuum chamber 12 to the outside. This completes a series of processes including the film forming process for forming the titanium nitride film.

In the explanation related to this series of processing, especially in the explanation related to the film forming process, the gas flow rate controller 64 was not specifically mentioned, but as mentioned above, the gas flow rate controller 64 is The mass flow controller 60b for introducing the reactive gas is controlled so that the pressure P of is kept constant, and more specifically, the flow rate Qr of the reactive gas introduced into the vacuum chamber 12 via the mass flow controller 60b is controlled. On the other hand, the flow rate Qd of the inert gas is kept constant. In addition, by keeping the angle θ of the conductance valve 22 (of the blade member) constant, the effective exhaust speed at the exhaust port 14 of the vacuum chamber 12 is kept constant. The functions and effects of performing the film forming process in this way will be explained in detail later.

On the other hand, when a film formation process is performed using the so-called conventional technique disclosed in Patent Document 1, for example, the flow rate Qd of the inert gas is constant, and the flow rate Qr of the reactive gas is also It is assumed to be constant. Then, the angle θ of the conductance valve is automatically controlled so that the pressure P in the vacuum chamber is constant.

Specifically, for example, when a film formation process is performed to form a titanium nitride film using the conventional technology, the flow rate Qd of argon gas as an inert gas (hereinafter referred to as "Q[Ar]") is ) is constant, and the flow rate Qr of nitrogen gas as a reactive gas (hereinafter, expressed by the symbol "Q[N ₂ ]") is constant. Then, the angle θ of the conductance valve is automatically controlled so that the pressure P in the vacuum chamber is constant. Additionally, the arc discharge power Ed is kept constant, the sputtering power Es is kept constant, and the substrate bias voltage Vb (average value) is kept constant.

An experiment was conducted to determine what kind of properties the titanium nitride film would have when a film formation process for forming a titanium nitride film was performed using such a conventional technique. The results are shown in FIG. Note that FIG. 5(a) shows the conditions of the film-forming process (film-forming conditions) in this experiment, and FIG. 5(b) shows the properties of the titanium nitride film formed by the film-forming process. The measurement results of the film thickness and color tone of the titanium nitride film and the film formation rate converted from the film thickness are shown. FIG. 5(c) shows a photograph of the appearance of the titanium nitride film.

As shown in FIG. 5(a), in this experiment, the flow rate Q [Ar] of argon gas was set to 100 mL/min (constant), and the flow rate Q [N ₂ ] of nitrogen gas was set to 40 mL/min (constant). did. Then, the pressure P in the vacuum chamber was set to 0.3 Pa (constant), and strictly speaking, the angle θ of the conductance valve was automatically controlled so as to maintain it. Further, when the arc discharge power Ed is set to 700 W (constant), and more specifically, the arc discharge voltage Vd, which is the voltage component of the arc discharge power Ed, is set to 50 V (constant), the current component of the arc discharge power Ed is The cathode power Ec was automatically controlled so that the arc discharge current Id was 14A (constant). In addition, the sputtering power Es was set to 8 kW (constant), and the substrate bias voltage Vb (average value) was set to -100 V (constant). The film forming process under these conditions was performed twice (two batches) over a time (film forming time) T of 80 minutes.

As a result, as shown in FIG. 5(b), the titanium nitride film formed by the first (first batch) film formation process, so to speak, sample 1, had a film thickness of 0.75 μm. . When this is converted into a film forming rate, the film forming rate is 0.56 μm/h. On the other hand, the titanium nitride film formed by the second (second batch) film forming process, so to speak, sample 2, had a film thickness of 0.96 μm. When this is converted into a film forming rate, the film forming rate is 0.72 μm. That is, the film thickness difference, which is the difference between the film thickness of sample 1 and the film thickness of sample 2, was 0.21. This film thickness difference of 0.21 μm is relatively large compared to the film thicknesses of Sample 1 and Sample 2, which means that good reproducibility cannot be obtained with respect to the film thickness. The film-forming speed difference, which is the difference between the film-forming speed for sample 1 and the film-forming speed for sample 2, is 0.16 μm/h. , is relatively large compared to the film forming speeds of Sample 1 and Sample 2, which means that the variation in film forming speed is relatively large. Note that the film thickness was measured by measuring the difference in level using a contact type surface roughness meter. Furthermore, a silicon (Si) wafer was used as the object to be processed for film thickness measurement.

On the other hand, focusing on the color tone, L ^* , a ^* , and b ^* , which are the color tone elements of Sample 1, were 56.35, 8.74, and 32.61, respectively. In contrast, L ^* , a ^* , and b ^* of sample 2 were 60.56, 2.97, and 16.91, respectively. That is, the differences ΔL ^* , Δa ^* , and Δb ^* between the two were 4.21, 5.77, and 15.70, respectively. The values of these differences ΔL ^* , Δa ^* , and Δb ^* are all relatively large compared to the values of L ^* , a ^* , and b ^* for Sample 1 and Sample 2, which means that the color tone also has good reproducibility. is not obtained. The color tone was measured using a color difference meter (CR-400) manufactured by Konica Minolta, Inc. In addition, a mirror-finished SUS304 plate (30 mm x 30 mm x 1 mm) was used as the object to be processed for color tone measurement.

Regarding the color tone, as shown in FIG. 5(c), it is clear from the appearance that Sample 1 and Sample 2 are completely different. For example, Sample 1 has a golden color characteristic of a titanium nitride film, but Sample 2 has a faded gold color, so to speak, a whitish golden color. In addition, in the external appearance photograph shown in FIG. 5(c), tracing was applied to Sample 1 and Sample 2 in order to reduce the gloss of each sample, taking into consideration the ease of viewing the appearance of Sample 1 and Sample 2. It is covered with paper. In addition, a ruler is attached so that the sizes of Sample 1 and Sample 2 can be intuitively understood.

As is clear from the results of this experiment, the conventional technology does not provide the reproducibility of the film thickness and color tone of the titanium nitride film. Considering the reason for this, as mentioned above, in the conventional technology, each time a film formation process for forming a titanium nitride film is performed, in other words, for each batch, and as the batch is repeated, It is presumed that this is because the sputtering speed of the sputtered surface of the target changes.

Therefore, in this embodiment, when a film formation process for forming a titanium nitride film is performed, the flow rate Q [Ar] of argon gas as an inert gas is kept constant. Then, the sputtering power Es is kept constant, and the angle θ of the conductance valve 22 is kept constant, that is, the effective exhaust speed at the exhaust port 14 of the vacuum chamber 12 is kept constant. Then, the flow rate Q [N ₂ ] of nitrogen gas as a reactive gas is automatically controlled so that the pressure P in the vacuum chamber 12 is constant. In addition, when the arc discharge power Ed is kept constant, more specifically, the arc discharge voltage Vd, which is the voltage component of the arc discharge power Ed, is kept constant, the arc discharge power Ed, which is the current component of the arc discharge power Ed, is kept constant. The cathode power Ec is automatically controlled so that the discharge current Id is constant. By performing the film forming process in this manner, the sputtering rate can be stabilized. Further, the substrate bias voltage Vb (average value) is kept constant. As a result, the sputtering speed of the sputtering surface of the target 242 is stabilized, and the reproducibility of the titanium nitride film is improved, and in particular, the reproducibility of the thickness and color tone of the titanium nitride film is improved. This will be explained in detail below.

First, in this example, an experiment was conducted to confirm how the pressure P in the vacuum chamber 12 and the flow rate Q [N ₂ ] of nitrogen gas change when a titanium nitride film is formed. I did it. The results are shown in FIG. Note that FIG. 6 shows an example of log data in which measurement results of the pressure P in the vacuum chamber 12 and the flow rate Q [N ₂ ] of nitrogen gas are recorded in chronological order. Further, log data of the sputtering power Es and the arc discharge current Id are also shown for reference in FIG. 6 . The horizontal axis in FIG. 6 represents time, and one division on the horizontal axis corresponds to 10 minutes. A time t0 on the horizontal axis is the time when various controls including automatic control of the nitrogen gas flow rate Q[N ₂ ] are started. On the other hand, the vertical axis in FIG. 6 represents the flow rate Q [N ₂ ] of nitrogen gas. Therefore, with respect to the pressure P in the vacuum chamber 12, the sputtering power Es, and the arc discharge current Id other than the nitrogen gas flow rate Q [N ₂ ], the scale on the vertical axis is irrelevant and shows only the respective transition states.

Further, in this experiment, the flow rate Q [Ar] of argon gas was set to 100 mL/min (constant), the sputtering power Es was set to 8 kW (constant), and the angle θ of the conductance valve 22 was set to 40 degrees. In addition, the pressure P in the vacuum chamber 12 was set to 0.220 Pa (constant), and strictly speaking, the flow rate Q [N ₂ ] of nitrogen gas was automatically controlled so as to keep it that way. In addition, when the arc discharge power Ed is 700 W (constant), and more specifically, the arc discharge voltage Vd, which is the voltage component of the arc discharge power Ed, is 50 V (constant), the current of the arc discharge power Ed is The cathode power Ec was automatically controlled so that the component arc discharge current Id was 14A (constant).

In FIG. 6 showing the results of this experiment, for example, focusing on the flow rate Q[N ₂ ] of nitrogen gas, the flow rate Q[N ₂ ] of nitrogen gas is as follows until just before time t0 when automatic control is started: It shows a constant value of 40 mL/min. In other words, the flow rate Q[N ₂ ] of nitrogen gas is controlled by the mass flow controller 60b to be a constant value of 40 mL/min until just before time t0. After time t0, the flow rate Q [N ₂ ] of nitrogen gas is automatically controlled, so that the flow rate Q [N ₂ ] of nitrogen gas becomes a value corresponding to the pressure P in the vacuum chamber 12. . Note that the pressure P in the vacuum chamber 12 after time t0 is basically the constant value of 0.220 Pa mentioned above, and in other words, the flow rate Q [N ₂ ] of nitrogen gas is automatically controlled so as to maintain it. .

Here, when looking at the pressure within the vacuum chamber 12 after time t0, the pressure P within the vacuum chamber 12 fluctuates somewhat (subtly) and occasionally rises instantaneously. Correspondingly (so to speak, synchronously), the flow rate Q [N ₂ ] of nitrogen gas also fluctuates somewhat, but occasionally decreases instantaneously when the pressure inside the vacuum chamber 12 increases instantaneously. However, in detail, it decreases from a value of about 40 mL/min to a value of about 30 mL/min. Note that the period in which the pressure P in the vacuum chamber 12 instantaneously increases, in other words, the period in which the flow rate Q [N ₂ ] of nitrogen gas instantaneously decreases, is not constant, but ranges from several minutes to more than ten minutes. and is indeterminate. Further, the time period during which the pressure P in the vacuum chamber 12 is momentarily increased, in other words, the time period during which the flow rate Q [N ₂ ] of nitrogen gas is momentarily reduced, is several seconds. is 2 to 3 seconds.

From such a transition in the pressure P in the vacuum chamber 12 and the flow rate Q [N ₂ ] of nitrogen gas, it is presumed that the following phenomenon occurs in the process of forming a titanium nitride film.

That is, in the process of forming a titanium nitride film, it is considered that the vicinity of the boundary between the erosion region 242a and the non-erosion region on the sputtered surface of the target 242 is nitrided as described above. This nitriding reduces the amount of titanium particles sputtered from the sputtered surface of the target 242, and accordingly reduces the amount of nitrogen gas particles that react with the titanium particles (sputtered particles). The amount of nitrogen gas particles consumed is reduced by reacting with As a result, the pressure P within the vacuum chamber 12 increases. On the other hand, the flow rate Q[N ₂ ] of nitrogen gas is automatically controlled so that the pressure P inside the vacuum chamber 12 is constant, so that the flow rate Q[N ₂ ] of nitrogen gas is rapidly decreased. However, in detail, as mentioned above, the flow rate decreases from a value of about 40 mL/min to a value of about 30 mL/min. As a result, the ratio (relative ratio) of the amount of argon gas particles to the amount of nitrogen gas particles in the vacuum chamber 12 increases, which reduces the frequency with which argon gas particles, especially argon ions, collide with the sputtering surface of the target 242. (number of times) increases. As a result, the nitride layer formed on the sputtering surface of the target 242 by the above-mentioned nitriding is scraped away by the argon ions, and the amount of titanium particles sputtered from the sputtering surface of the target 242 increases. Then, the amount of nitrogen gas particles consumed by reacting with the increased titanium particles increases, and the pressure P in the vacuum chamber 12 decreases by that amount. In response to this, in order to compensate for the decrease in the pressure P in the vacuum chamber 12, in other words, to replenish the nitrogen gas particles necessary to react with the increased titanium particles, the flow rate Q [N ₂ ] of nitrogen gas is increased. increases and returns to the original value of about 40 mL/min. It is presumed that by repeating such a phenomenon, the pressure within the vacuum chamber 12 will occasionally rise instantaneously and the flow rate [N ₂ ] of nitrogen gas will decrease instantaneously, as described above. Note that since the degree of nitridation of the surface of the target 242 to be sputtered and the concentration of nitrogen gas in the vacuum chamber 12 are not constant, the pressure P in the vacuum chamber 12 increases instantaneously, and the flow rate Q of nitrogen gas increases. It is assumed that the period in which [N ₂ ] instantaneously decreases is not constant, that is, there is no regularity in the period.

In this way, the pressure P in the vacuum chamber 12 momentarily increases and the flow rate Q [N ₂ ] of nitrogen gas momentarily decreases, but the nitrided surface formed on the sputtering surface of the target 242 The layer is appropriately scraped off, so to speak, the sputtered surface of the target 242 is refreshed. As a result, the sputtering speed of the sputtered surface of the target 242 is stabilized, and the reproducibility of the titanium nitride film is improved, particularly the reproducibility of the film thickness and color tone.

In order to verify this, in this example, a titanium nitride film was actually formed and an experiment was conducted to check whether the reproducibility of the titanium nitride film could be obtained, especially whether the reproducibility of the color tone and film thickness could be obtained. went. FIG. 7 shows the film forming conditions in this experiment, and FIG. 8 shows the properties of the titanium nitride film formed in this experiment.

Specifically, as shown in FIG. 7, in this experiment, the flow rate Q [Ar] of argon gas was set to 100 mL/min (constant), the angle θ of the conductance valve 22 was set to 40 degrees (constant), Furthermore, the flow rate Q [N ₂ ] of nitrogen gas was automatically controlled so that the pressure P in the vacuum chamber 12 was 0.220 Pa (constant). Then, the arc discharge power Ed is set to 1000 W (constant), and more specifically, the arc discharge voltage Vd, which is the voltage component of the arc discharge power Ed, is set to 50 V (constant), and the current component of the arc discharge power Ed is The cathode power Ec was automatically controlled so that a certain arc discharge current Id was 20 A (constant). In addition, the sputtering power Es was set to 8 kW (constant), and the substrate bias voltage Vb (average value) was set to -100 V (constant). The film forming process under these conditions was performed 10 times (10 batches) over a time (film forming time) T of 65 minutes.

The film thickness and L ^* , a ^* , and b ^* , which are elements of color tone, were measured for each titanium nitride film formed by these 10 film-forming processes (batches), and the results are shown in FIG. The results were as follows. The film thickness was measured using a contact type surface roughness meter to measure the difference in level, similar to the film thickness experiment in the prior art described above, and a silicon wafer was used as the object to be processed for the film thickness measurement. . The measurement of color tone was also carried out using a color difference meter (CR-400) manufactured by Konica Minolta, Inc., as in the experiment regarding color tone in the prior art described above, and the object 100 for color tone measurement was A mirror-finished SUS304 plate (30 mm x 30 mm x 1 mm) was used. Furthermore, FIG. 8 also shows the flow rate Q [N ₂ ] (variation range) of nitrogen gas, the processing temperature (the surface temperature of the object to be processed 100), and the film formation rate in each film formation process. There is. Additionally, FIG. 8 also shows the measurement results of the internal stress of each titanium nitride film.

As shown in FIG. 8, for example, the difference in film thickness, which is the difference in film thickness, was 0.06 μm. This film thickness difference of 0.06 μm is much smaller than the film thickness difference (0.21 μm) according to the prior art described above. Furthermore, when this film thickness difference of 0.06 μm is expressed as a ratio to the average film thickness of each titanium nitride film formed by 10 film forming processes, it is ±5.36%, which is ±5. The ratio of .36% is at a level sufficient for practical use. That is, according to this example, it was confirmed that good reproducibility of the film thickness could be obtained.

When paying attention to the film forming speed, the difference in film forming speed was 0.05 μm/h. This film-forming speed difference of 0.05 μm/h is much smaller than the film-forming speed difference (0.16 μm/h) according to the prior art described above. Furthermore, when this film-forming speed difference of 0.05 μm/h is expressed as a ratio to the average value of the film-forming speed in each of the 10 film-forming processes, it is ±6.17%; This ratio is a sufficiently small value for practical use. That is, according to this example, it was confirmed that the variation in the film formation rate was sufficiently small, that is, the film formation rate could be stabilized.

Furthermore, focusing on the color tone, the differences ΔL*, Δa ^* , and Δb ^* of L ^* , a ^* , and b ^{* ,} which are the elements of the color tone, are 0.64, 0.85, and 1.16, which means that any was also less than 2. The values of the differences ΔL ^* , Δa ^* , and Δb ^*, which are less than 2, are greater than the values of the differences ΔL ^* , Δa ^* , and Δb ^* (4.21, 5.77, and 15.70) according to the prior art described above. Much smaller. Furthermore, the values of the differences ΔL ^* , Δa ^* , and Δb ^* are expressed as a ratio to the individual average values of L ^* , a ^* , and b ^* of each titanium nitride film formed by 10 film-forming processes. , ±0.51%, ±7.25%, and ±1.67%, and these ratio values are at a level that is sufficiently practical. That is, according to this example, it was confirmed that good reproducibility of color tone was also obtained.

Note that when paying attention to the flow rate Q [N ₂ ] (variation range) of nitrogen gas in each film forming process, the minimum value is approximately 30 mL/min, and in detail, it is 25.8 mL/min to 36.6 mL. /min. The maximum value of the nitrogen gas flow rate Q [N ₂ ] is approximately 40 mL/min, specifically 39.2 mL/min to 43.5 mL/min. Regarding the flow rate Q [N2] of this nitrogen gas, it appears that the variation is relatively large, but as a result, it has been confirmed that good reproducibility can be obtained in terms of film thickness and color tone as mentioned above. It was done.

Further, when paying attention to the processing temperature in each film forming process, the processing temperature is approximately 240°C, and the difference therebetween is 10°C. If this difference in processing temperature of 10°C is expressed as a ratio to the average processing temperature in each film forming process, it is ±2.07%, and this ratio of ±2.07% is an extremely small value. . That is, according to this example, it was confirmed that the variation in processing temperature was sufficiently small.

Looking at the internal stress (compressive stress) of the titanium nitride film formed by each film forming process, the internal stress is approximately -7 GPa, and the difference therebetween is 1.15 GPa. When this difference of 1.15 GPa is expressed as a ratio to the average value of internal stress of each titanium nitride film, it is ±8.34, and this ratio of ±8.34% is an extremely small value. That is, according to this example, it was confirmed that the variation in internal stress was sufficiently small.

The various control points in the titanium nitride film forming process according to this embodiment and the control points according to the prior art described above are summarized in a list as shown in FIG.

As shown in FIG. 9, in the process of forming a titanium nitride film according to the prior art described above, the flow rate Q [Ar] of argon gas as a discharge gas is kept constant, and the flow rate Q [Ar] of argon gas as a discharge gas is constant, and It is assumed that the gas flow rate Q[N ₂ ] is constant. Then, the angle θ of the conductance valve is automatically controlled so that the pressure P in the vacuum chamber is constant. Then, the arc discharge power Ed is kept constant, the sputtering power Es is kept constant, and the substrate bias voltage Vb (average value) is kept constant. In addition, the film forming time T is kept constant.

On the other hand, in the titanium nitride film forming process according to this embodiment, the flow rate Q [Ar] of argon gas as the discharge gas is kept constant, and the angle θ of the conductance valve 22 is kept constant. . Then, the flow rate Q [N ₂ ] of nitrogen gas as a reactive gas is automatically controlled so that the pressure P in the vacuum chamber 12 is constant. Then, the arc discharge power Ed is kept constant, the sputtering power Es is kept constant, and the substrate bias voltage Vb (average value) is kept constant. In addition, the film forming time T is kept constant.

That is, according to the prior art, the angle θ of the conductance valve is automatically controlled so that the pressure P in the vacuum chamber is constant while the flow rate Q [N ₂ ] of nitrogen gas is constant. On the other hand, according to this embodiment, the flow rate Q [N ₂ ] of nitrogen gas is adjusted so that the pressure P in the vacuum chamber 12 is constant while the angle θ of the conductance valve 22 is constant. Automatically controlled. In these respects, the configurations of this embodiment and the prior art are fundamentally different from each other. Due to this difference in structure, according to this example, a titanium nitride film can be formed with better reproducibility than in the conventional technology, and in particular, the titanium nitride film has good reproducibility in terms of film thickness and color tone. This results in extremely beneficial effects.

FIG. 10 is a list showing a comparison of the film thickness difference, film formation speed difference, and color tone difference ΔL ^* , Δa ^* , and Δb ^* according to the present example and the conventional technique. As is clear from FIG. 10, according to this example, better reproducibility in film thickness and color tone can be obtained than in the conventional technology, and the difference in film formation speed can be suppressed to a sufficiently small value for practical use. It will be done.

As described above, according to this embodiment, the sputtering rate is stabilized, thereby improving the reproducibility of a reaction film such as a titanium nitride film, and particularly the reproducibility of film thickness and color tone. This greatly contributes to the development of ornaments, for example. Furthermore, good reproducibility can be obtained over a long period of time (until the end of use of the target 242).

Note that this example is a specific example of the present invention, and does not limit the technical scope of the present invention. The present invention can also be applied to situations other than this example.

For example, as shown in FIG. 11, a plurality of magnetron sputter cathodes 24, 24, . One may also be provided. Specifically, each unit 400 includes a magnetron sputter cathode 24, a ground shield 26, and a cathode filament 30. Although not shown in FIG. 1, each unit 400 includes a sputter power supply 28, a heating power supply 32, an arc discharge power supply 34, a current detector 36, and a heating controller 38. That is, each unit 400 is, so to speak, a generation source of plasma 300, and a plurality of units 400, that is, four units, that are generation sources of plasma 300 may be provided. These units 400, 400, . . . are provided at equal intervals along the circumferential direction of a circle centered on the central axis Xa of the vacuum chamber 12, that is, provided at 90 degree intervals here. In addition, each unit 400, 400, . . . is provided at the same distance from the central axis Xa of the vacuum chamber 12.

According to the configuration shown in FIG. 11, the deposition rate of the reaction film can be improved (speeded up). Also in the configuration shown in FIG. 11, the flow rate Q [N ₂ ] of nitrogen gas is automatically controlled so that the pressure P in the vacuum chamber 12 is constant, thereby improving the reproducibility of the reaction membrane. is planned. Furthermore, it is not necessary to provide a plurality of elements responsible for automatic control, that is, a pressure gauge 62, a gas flow rate controller 64, and a mass flow controller 60b for introducing reactive gas. Just one is enough.

Note that the number of units 400 is not limited to four. In any case, the units 400, 400, ... are provided at equal intervals along the circumferential direction of a circle centered on the central axis Xa of the vacuum chamber 12, and are arranged at equal intervals from the central axis Xa of the vacuum chamber 12. It is essential that they are kept at a distance.

Furthermore, in this example, the case where a titanium nitride film is formed as a reaction film has been described, but various reaction films other than the titanium nitride film, such as a nitride film, a carbonitride film, a carbide film, etc., may also be formed. , the present invention can be applied. The various reaction films mentioned here include titanium carbonitride (TiCN) film, titanium carbide (TiC) film, titanium oxide (TiO ₂ ) film, zirconium nitride (ZrN) film, zirconium carbonitride (ZrCN) film, and zirconium carbide film. (ZrC) film, titanium aluminum nitride (TiAlN) film, titanium aluminum carbonitride (TiAlCN) film, aluminum chromium nitride (AlCrN) film, chromium nitride (CrN) film, chromium carbonitride (CrCN) film, chromium carbide (CrC) film, aluminum nitride (AlN) film, aluminum oxide (Al ₂ O ₃ ) film, silicon nitride (SiN), silicon carbonitride (SiCN) film, silicon carbide (SiC) film, silicon oxide (SiO ₂ ) film, yttrium oxide (Y ₂ O ₃ ) film, etc. In this case, a target 242 made of a material depending on the type of reaction film is used. Further, when forming a carbonitride film, in addition to nitrogen gas, a hydrocarbon gas such as acetylene (C ₂ H ₂ ) gas is used as the reaction gas. When forming a carbonized film, a hydrocarbon gas such as acetylene is used as the reaction gas instead of nitrogen gas. Furthermore, when forming an insulating reaction film such as an aluminum nitride film or a silicon nitride film, the sputter power supply 28 may output bipolar pulse power similar to the bias power supply 50 instead of a DC power supply. A pulsed power supply is used.

In addition, in practice, before forming the reaction film, a film formation process is performed to form an intermediate layer for improving the adhesion of the reaction film. For example, when the reaction film is a titanium nitride film, a titanium (Ti) film is formed as an intermediate layer, but the intermediate layer is appropriately selected depending on the type of reaction film.

Further, in this embodiment, a pulse power supply device is employed as the bias power supply device 50, but the present invention is not limited to this. For example, if the object to be processed 100 is a conductive substance and the reaction film to be formed is also a conductive film, a DC power supply device may be employed. Note that even if the object to be processed 100 is a conductive substance, if the reaction film to be formed is an insulating film, a pulse power supply device or a high frequency power supply device is used as the bias power supply device 50. , is appropriate. Further, when the object to be processed 100 is an insulating material, a high frequency power supply device is employed as the bias power supply device 50.

The present invention is not limited to application to the magnetron sputtering apparatus 10, but can also be applied to a film forming method using magnetron sputtering.

10... Magnetron sputtering device 12... Vacuum chamber 18... Vacuum pump 20... Opening/closing valve 22... Conductance valve 24... Magnetron cathode 26... Earth shield 28... Sputter power supply device 30... Filament 32... Power supply device for filament heating 34... Power supply for arc discharge Device 36... Current detector 38... Heating controller 50... Bias power supply device 56... Gas introduction pipe 58, 60... Branch pipe 58a, 60a... Opening/closing valve 58b, 60b... Mass flow controller 62... Vacuum gauge 64... Gas flow rate controller 100 ... Processing object 242 ... Target 300 ... Plasma 400 ... Unit

Claims (4)

A film forming apparatus that forms a reaction film on a processed object by magnetron sputtering,
a vacuum chamber that is at ground potential and in which the object to be processed is housed;
a magnetron sputtering cathode that has a target that is a material for the reaction film and is provided inside the vacuum chamber so that the sputtered surface of the target faces the object to be processed;
an earth shield provided to surround the outer periphery of the magnetron sputtering cathode with the sputtering surface exposed in order to limit the sputtering region of the magnetron sputtering cathode to the sputtering surface;
Exhaust means comprising: a vacuum pump that evacuates the inside of the vacuum chamber through an exhaust port of the vacuum chamber; and a conductance valve that controls an effective pumping speed at the exhaust port;
an inert gas introducing means for introducing an inert gas into the vacuum chamber at a constant flow rate;
Sputtering power supply means for supplying sputtering power to the earth shield and the magnetron sputter cathode for discharging the particles of the inert gas, using the earth shield as an anode and the magnetron sputter cathode as a cathode;
reactive gas introduction means for introducing a reactive gas that will become a material for the reaction membrane into the vacuum chamber;
Using the vacuum chamber as an anode and the object to be treated as a cathode, the particles of the inert gas, the particles of the reactive gas, and the sputtered particles sputtered from the sputtered surface of the target are directed toward the object. bias power supply means for supplying bias power with a constant predetermined component for acceleration to the vacuum chamber and the object to be processed;
a cathode filament provided between the sputtering surface of the target and the object to be processed;
heating power supply means for supplying heating power to the cathode filament to heat the cathode filament and cause the cathode filament to emit thermoelectrons;
Arc discharge power supply means for supplying arc discharge power to the earth shield and the cathode filament for accelerating the thermoelectrons toward the earth shield, with the earth shield serving as an anode and the cathode filament serving as a cathode. ,
heating power control means for controlling the heating power so that the current component of the arc discharge power is constant while the voltage component of the arc discharge power is constant;
While the sputtering power is kept constant and the effective pumping speed is kept constant by the conductance valve, the reactive gas is supplied to the inside of the vacuum chamber so that the pressure inside the vacuum chamber is constant. A film forming apparatus comprising a reaction gas flow rate control means for controlling the flow rate of the reaction gas. 2. A method according to claim 1, comprising a plurality of units including the magnetron sputter cathode, the earth shield, the sputtering power supply means, the cathode filament, the heating power supply means, the arc discharge power supply means, and the heating power control means. The film forming apparatus described. 3. The film forming apparatus according to claim 1, wherein the bias power is DC power, asymmetric bipolar pulse power, or high frequency power. A film formation method for forming a reaction film on a processed object by magnetron sputtering, the method comprising:
The object to be processed is installed in a vacuum chamber which is connected to a ground potential and is provided with a magnetron sputtering cathode having a target that is the material of the reaction film, so as to face the surface to be sputtered of the target. installation step,
an evacuation step of evacuating the inside of the vacuum chamber with a vacuum pump through the exhaust port of the vacuum chamber, and controlling the effective pumping speed at the exhaust port with a conductance valve;
an inert gas introduction step of introducing an inert gas into the vacuum chamber at a constant flow rate;
an earth shield provided to surround the outer periphery of the magnetron sputtering cathode with the sputtering surface exposed in order to limit the sputtering area of the magnetron sputtering cathode to the sputtering surface; is used as an anode, and the magnetron sputter cathode is used as a cathode, and a sputter power supply step of supplying sputter power for discharging the particles of the inert gas to the earth shield and the magnetron sputter cathode;
a reactive gas introducing step of introducing a reactive gas to be a material for the reaction film into the vacuum chamber;
Using the vacuum chamber as an anode and the object to be treated as a cathode, the particles of the inert gas, the particles of the reactive gas, and the sputtered particles sputtered from the sputtered surface of the target are directed toward the object. a bias power supply step of supplying bias power with a constant predetermined component for acceleration to the vacuum chamber and the workpiece;
a step of supplying heating power to the cathode filament to heat the cathode filament provided between the sputtering surface of the target and the object to be processed and cause the cathode filament to emit thermoelectrons; ,
a step of supplying arc discharge power to the earth shield and the cathode filament, with the earth shield serving as an anode and the cathode filament serving as a cathode, for accelerating the thermoelectrons toward the earth shield; ,
a heating power control step of controlling the heating power so that the current component of the arc discharge power is constant while the voltage component of the arc discharge power is constant;
With the sputtering power kept constant and the effective pumping speed kept constant by the conductance valve, the reactive gas is supplied to the inside of the vacuum chamber so that the pressure inside the vacuum chamber becomes constant. A method for forming a film, comprising a step of controlling a flow rate of a reactant gas.

Images