KR20240025894A - Large area plasma generator and matching method - Google Patents

Abstract

A large-area plasma generator and matching method are disclosed.
The characteristic configuration of the disclosed invention includes a reactor provided with a vacuum chamber therein; a substrate holder portion disposed below the vacuum chamber and provided with a seating surface on which a substrate is mounted; An antenna unit provided to be connected to the substrate holder unit, which introduces ultra-high frequencies provided from the ultra-high frequency oscillation member and transmits them to the substrate holder unit, thereby generating a large-area plasma corresponding to the area of the seating surface on the substrate holder unit. Board holder and antenna; A lifting cavity is installed on the upper side of the vacuum chamber to be able to be lifted by a first lifting member, and has an open lower part to shield the upper space of the substrate holder part when lowered, thereby forming a resonance chamber in which plasma is generated; a lifting plunger that is installed to be capable of being raised and lowered within the inner area of the lifting cavity by a second lifting member and performs matching; and controlling the lifting plunger to be positioned at a point where reflected waves are minimized.

Description

Large area plasma generator and matching method}

The present invention relates to a plasma generator and matching method.

In particular, by installing a cavity that can be raised and lowered inside the reactor, a resonance space is formed between the substrate holder and by using the substrate holder as an antenna, it is possible to generate a high-density plasma with a large area larger than the object to be processed; By installing a plunger that can be raised and lowered within the inner area of the cavity, the point where the size of the reflected wave for ultra-high frequencies is lowest can be captured and the impedance matching (tuning) can be precisely controlled; The substrate holder and antenna are configured to be raised and lowered so that the optimal contact distance between the object to be processed and the plasma can be controlled; The point is that the resonance frequency can be finely matched by changing the frequency; It relates to a large-area plasma generator and matching method having features such as the like.

The content described in this section simply provides background information for this embodiment and does not constitute prior art.

In the manufacturing process of semiconductor devices and liquid crystal displays, plasma generators such as plasma etching devices and plasma CVD film forming devices are used to perform plasma processing such as etching or film forming on objects such as semiconductor wafers or glass substrates. It is used.

A method of generating plasma in a plasma generator includes supplying a processing gas into a chamber in which parallel plate electrodes are arranged, supplying a predetermined power to the parallel plate electrodes, and generating plasma by capacitive coupling between the electrodes. Electrons are accelerated by an electric field generated by microwaves and a magnetic field generated by a magnetic field generator placed outside the chamber, and these electrons collide with neutral molecules in the processing gas to ionize the neutral molecules, creating plasma. Methods for generating it are known.

In the case of the latter method using the magnetron effect of the electric field generated by the microwave and the magnetic field generated by the magnetic field generator, microwaves of a certain power are supplied to the antenna placed in the chamber by passing through the waveguide/coaxial tube, and the microwave is transmitted from the antenna. is being radiated into the processing space within the chamber.

A general microwave introduction device is equipped with a microwave oscillator having a magnetron that outputs microwaves adjusted to a predetermined power and a microwave generation power supply that supplies a direct anode current to the magnetron, and sends the microwaves output from the microwave oscillator to the chamber via an antenna. It was configured to radiate into the processing space within.

The area of semiconductor wafers and liquid crystal display glass substrates that are to be processed is increasing every year. Therefore, plasma processing requires a plasma generator capable of generating high-density, uniform plasma over a large area.

In response to this request, a plasma generator has been disclosed that introduces microwaves guided through a square waveguide into an annular space through a coaxial line and into a chamber through a plurality of arc-shaped slits (Patent Document 1).

However, objects to be treated with plasma are becoming increasingly larger in size, and there is a demand for a plasma generator that can stably generate uniform plasma over a large area.

Additionally, during actual plasma processing, there are often cases where the plasma processing conditions must be changed in various ways depending on the type and number of objects to be processed. Therefore, it is desirable to stably generate uniform plasma over a wide range of conditions.

Japanese Patent Publication No. 2003-124193 Registered Patent Publication No. 10-0689037 Registered Patent Publication No. 10-1240842

The present invention was made based on the recognition of the above-described problems. The purpose of the present invention is to provide a large-area plasma generator that can stably form high-density plasma with uniform distribution over a large area.

The technical problem to be achieved by the present invention is not limited to the technical problem mentioned above, and other technical problems not mentioned can be clearly understood by those skilled in the art from the description below. There will be.

The present invention for achieving the above object includes a reactor provided with a vacuum chamber therein; a substrate holder portion disposed below the vacuum chamber and provided with a seating surface on which a substrate is mounted; An antenna unit provided to be connected to the substrate holder unit, which introduces ultra-high frequencies provided from the ultra-high frequency oscillation member and transmits them to the substrate holder unit, thereby generating a large-area plasma corresponding to the area of the seating surface on the substrate holder unit. Board holder and antenna; A lifting cavity is installed on the upper side of the vacuum chamber to be able to be lifted by a first lifting member, and has an open lower part to shield the upper space of the substrate holder part when lowered, thereby forming a resonance chamber in which plasma is generated; a lifting plunger that is installed to be capable of being raised and lowered within the inner area of the lifting cavity by a second lifting member and performs matching; And a matching controller that controls the lifting plunger to be positioned at a point where reflected waves are minimized. A large-area plasma generator including a is provided.

Preferably, a cooling passage may be formed inside the wall of the lifting cavity, and a refrigerant may be circulated through the cooling passage to cool the lifting cavity.

Preferably, the lifting plunger and the substrate holder/antenna can be cooled by contacting a cooling module with their unused portions, respectively.

Preferably, the first lifting member includes: a first fixing flange coupled to the upper surface of the reactor; a first lifting flange disposed above the first fixing flange to be capable of being raised and lowered; A first lifting rod in the form of a pipe whose upper part is supported on the first lifting flange and whose lower end penetrates the first fixed flange and the upper surface of the reactor and is connected to the upper part of the lifting cavity; a first screw rod whose lower part is connected to the first fixing flange and whose upper end is engaged with an arm tab block provided on the first lifting flange; a first lifting control knob installed on the upper end of the first screw rod and allowing the first lifting flange to be raised and lowered by rotating the first screw rod through a rotational operation; a first guide rod whose lower end is supported by the first fixing flange and whose upper end penetrates the first lifting flange; and a first guide block installed on the first lifting flange and guiding the raising and lowering of the first guide rod.

Preferably, the inside of the first lifting rod communicates with the cooling passage of the lifting cavity, and an inflow port through which refrigerant flows into the first lifting rod may be provided on one side.

Preferably, it may further include a bellows for sealing around the first lifting rod in the section between the first fixing flange and the first lifting flange.

Preferably, the second lifting member includes: a second fixed flange installed on the first lifting flange and raised and lowered in conjunction with the lifting and lowering of the first lifting flange; a second lifting flange disposed on an upper side of the second fixing flange to be capable of being raised and lowered in a direction close to or distant from the second fixing flange; a second lifting bar whose upper part is supported on the second lifting flange and whose lower end is connected to the upper part of the lifting plunger through a hollow part of the first lifting bar; a second screw rod whose lower part is connected to the second fixing flange and whose upper end is engaged with an arm tab block provided on the second lifting flange; a second lifting control knob installed on the upper end of the second screw rod and allowing the second lifting flange to be raised and lowered as the second screw rod is rotated through a rotation operation; a second guide rod whose lower end is supported by the second fixing flange and whose upper end penetrates the second lifting flange; and a second guide block installed on the second lifting flange and guiding the raising and lowering of the second guide rod.

Preferably, it may further include a bellows for sealing around the second lifting rod in the section between the second fixing flange and the second lifting flange.

Preferably, the substrate holder and antenna are lifted and lowered by a third lifting member, wherein the third lifting member includes: a third fixing flange coupled to the bottom of the reactor; a third lifting flange disposed on the lower side of the third fixing flange to be capable of being raised and lowered; a third lifting rod in the form of a pipe whose lower part is supported on the third lifting flange and whose upper end penetrates the third fixed flange and the bottom of the reactor and is connected to the substrate holder and antenna; a third screw rod whose upper part is connected to the third fixing flange and whose lower end is engaged with an arm tab block provided on the third lifting flange; a third lifting control knob installed on the upper end of the third screw rod and allowing the third lifting flange to be raised and lowered by rotating the third screw rod through a rotational operation; a third guide rod whose upper end is supported by the third fixing flange and whose lower end penetrates the third lifting flange; and a third guide block installed on the third lifting flange and guiding the raising and lowering of the third guide rod.

Preferably, the first lifting member includes: a first fixing flange coupled to the upper surface of the reactor; a first lifting flange disposed above the first fixing flange to be capable of being raised and lowered; a first lifting rod whose upper part is supported on the first lifting flange and whose lower end penetrates the first fixed flange and the upper surface of the reactor and is connected to the upper part of the lifting cavity; a plurality of first screw rods, the lower part of which is connected to a support block of the first fixed flange, and the upper part of which is engaged with an arm tab block provided on the first lifting flange; a first drive chain gear rotatably installed at a lower portion of one of the support blocks; a plurality of first driven chain gears rotatably installed at lower portions of the remaining support blocks among the support blocks on which the first driving chain gear is not installed; a first chain connecting the first drive chain gear and the first driven chain gear to interlock with each other; a first whim wheel installed on a support block on which the first drive chain gear is installed and integrally fixed with the first drive chain gear and the support block; A first whim gear engaged with the first worm wheel; and a first rotary handle for rotating the first worm gear.

Preferably, the second lifting member includes: a second fixed flange installed on the first lifting flange and raised and lowered in conjunction with the lifting and lowering of the first lifting flange; a second lifting flange disposed on an upper side of the second fixing flange to be capable of being raised and lowered in a direction close to or distant from the second fixing flange; a second lifting bar whose upper part is supported on the second lifting flange and whose lower end is connected to the upper part of the lifting plunger through a hollow part of the first lifting bar; a plurality of second screw rods, the lower part of which is connected to a support block of the second fixing flange, and the upper part of which is engaged with an arm tab block provided on the second lifting flange; a second drive chain gear rotatably installed below one of the support blocks; a plurality of second driven chain gears rotatably installed at lower portions of the remaining support blocks among the support blocks on which the first driving chain gear is not installed; a second chain connecting the second drive chain gear and the second driven chain gear to interlock with each other; a second whim wheel installed on the support block on which the second drive chain gear is installed and integrally fixed with the second drive chain gear and the support block; a second whim gear engaged with the second worm wheel; and a second rotary handle for rotating the second worm gear.

Preferably, the matching controller includes a directional coupler that measures a reflected wave or impedance acting on the lifting plunger; a storage unit that stores reflected waves or impedance measurements fed back from the directional coupler; and a lifting control unit that compares the measured value stored in the storage unit with the set value and, when the measured value exceeds the set value, adjusts the lifting position of the lifting plunger until the measured value falls below the set value. there is.

Another aspect of the present invention is to measure the reflected wave in the measurement unit of the matching controller while the frequency of the ultra-high frequency oscillation member is fixed at 2,450 MHz, and store the real-time measured value measured in the measurement unit in a storage unit, and store the measured value in real time in the storage unit. A matching method is provided that compares the real-time measured value stored in and the preset setting value, and if the measured value exceeds the set value, the lifting control unit readjusts the position of the lifting plunger to the optimal point where reflected waves are minimized.

According to another aspect of the present invention, with the frequency of the ultra-high frequency oscillating member varying within the range of 2,400 to 2,500 MHz, the entire frequency range (bandwidth) is divided into frequency range 1 and scanned to measure the reflected wave, but the entire frequency range is 100 MHz. is scanned at 10 MHz, which is the frequency domain 1, and the reflected wave is measured for each frequency domain of group 1 to find the point where the reflected wave is minimized. When the region width 1 where the reflected wave of the group 1 is minimized is detected, the frequency domain width 2 is detected. Group 2, in which reflected waves are minimized, is formed by reducing the frequency domain width 1 to 1/n and conducting a secondary search. After the secondary search, frequency domain width 2 is reduced to 1/m and frequency domain 3 is searched to minimize reflected waves. A matching method is provided through frequency grouping search to detect the optimal frequency by forming a minimized group 3.

According to an embodiment of the present invention, a uniform and high-density plasma can be generated by limiting the plasma generation space on the upper side of the substrate holder by embedding a cavity in the reactor, and embedding a plunger in the cavity to generate reflected waves through positional displacement. By capturing the point where the minimum is the lowest, the impedance matching can be precisely controlled, and by configuring the substrate holder and antenna to be raised and lowered, there is an advantage of controlling the optimal contact distance between the object to be processed and the plasma.

In addition, as the resonance frequency can be finely matched by changing the frequency, optimal high-frequency energy can be expressed.

In addition, tuning within a fine frequency range from the beginning takes a lot of time, so the optimal frequency can be tracked in a short time by adjusting it through grouping search that divides the frequency into a large and small frequency section.

The effects of the present invention are not limited to the effects described above, and should be understood to include all effects that can be inferred from the configuration of the invention described in the detailed description or claims of the present invention.

Figure 1 is an example diagram (plan view and front view together) for explaining a plasma generator, which is a conventional micro application device.
Figure 2 is an actual illustration of a plasma generator, a conventional micro application device;
Figure 3 is an example showing plasma generation in the power range of 230 to 2,000 Watt in a plasma generator, a conventional micro application device;
Figure 4 is an example showing the phenomenon in which plasma changes according to pressures of 0.5, 1, and 2 torr in a conventional plasma generator using microwaves;
Figure 5 is a configuration diagram of a transmission line (coaxial cable) for microwave transmission;
Figure 6 is an example showing a configuration in which a magnetron, a circulator, and a dummy load are embedded in a power supply among the components of a plasma generator, which is a conventional microwave application device.
Figure 7 shows a configuration in which, among the components of a plasma generator, which is a conventional microwave application device, a magnetron, circulator, dummy load, and stub matcher are built into the power supply, and ultra-high frequencies are transmitted through a coaxial rod (coaxial cable) instead of a waveguide. Example too,
Figure 8 is an example of a conversion adapter for converting a waveguide to a coaxial cable;
9 is a conceptual diagram of a large-area plasma generator according to the present invention;
Figure 10 is an example diagram for forming a magnetic field in the lifting cavity,
Figure 11 is an example of a waveguide sealing window;
12 is a first embodiment of the first and second lifting members according to the present invention;
13 is a first embodiment of the third lifting member according to the present invention;
14 shows a second embodiment of the first and second lifting members according to the present invention;
Figure 15 is a schematic diagram of the main part of Figure 14;
16 is a second embodiment of the third lifting member according to the present invention;
17 is a block diagram for explaining the configuration of according to the present invention,
18 is an exemplary diagram of a plasma generating device using semiconductor oscillation microwaves,
Figure 19 is a graph illustrating [I ² ] according to frequency change in a series resonance circuit.

The present invention may be implemented in many different forms and is therefore not limited to the embodiments described herein. In order to clearly explain the present invention in the drawings, parts that are not related to the description are omitted, and similar parts are given similar reference numerals throughout the specification.

Throughout the specification, when a part is said to be "connected (connected, contacted, combined)" with another part, this means not only "directly connected" but also "indirectly connected" with another member in between. "Includes cases where it is. Additionally, when a part is said to “include” a certain component, this does not mean that other components are excluded, but that other components can be added, unless specifically stated to the contrary.

The terms used herein are only used to describe specific embodiments and are not intended to limit the invention. Singular expressions include plural expressions unless the context clearly dictates otherwise. In this specification, terms such as “comprise” or “have” are intended to indicate the presence of features, numbers, steps, operations, components, parts, or combinations thereof described in the specification, but are not intended to indicate the presence of one or more other features. It should be understood that this does not exclude in advance the possibility of the existence or addition of elements, numbers, steps, operations, components, parts, or combinations thereof.

Hereinafter, preferred embodiments of the present invention will be described with reference to the attached FIGS. 1 to 18.

The attached Figure 1 is a plan view and a front view of a microwave application device.

As shown in the drawing above, the overall configuration of the microwave application device consists of a magnetron (Magnetron: 111), a waveguide (112), a circulator (Circulator: 113), a 3-stub tuner (3stub-tuner: 115), and an applicator (Applicator). :117), a sliding short circuit (119), and a power supply or power (Power:121) that controls and supplies power to the magnetron. In other words, when power is supplied to the filament and anode of the magnetron (Magnetron: 111), the microwaves generated by the magnetron travel through the waveguide, circulator (113), and 3-stub tuner (3 stub-tuner: 115). It is delivered to the applicator (Applicator: 117).

Here, a 3-stub tuner is a device with three metal rods in a waveguide or cavity resonator, and is a matching device that ensures that microwaves are transmitted as much as possible without being reflected. A plunger or sliding short circuit (SSC) is also an additional device installed for matching to minimize reflected waves. Sometimes 4-stub is used instead of 3-stub. In addition, each stub can be manually adjusted and matched (manual tuning), or automatic tuning using a matching algorithm can be performed.

The automatic matcher consists of a stub, a diode, an integrated automatic control board, and a motor drive unit, which matches the load impedance.

A more detailed explanation of matching is as follows.

In general, when two circuits are combined, the impedance seen from the output terminal of the first circuit to the input terminal of the same circuit is Z1 = R1 + jX1, and the input impedance of the second circuit is Z2 = R2 + jX2. The air space between the two impedances is If there is a dual relationship, that is, R1 = R2, X1 = -X2, the condition of minimum loss and maximum output is obtained. Here, the first circuit is a power supply, the second circuit is a load (plasma applicator device), and the matcher serves to match the impedance of the circuits. In other words, matching is a method for transmitting maximum output from a signal transmission path. The principle of matching is that the signal source has internal resistance, and for maximum output transmission, the load must be equal to the impedance of the signal source. Where active devices are included, the input impedance and output impedance are significantly different. However, the input impedance needs to be matched to the signal source, and the output load needs to be matched to the output impedance. Power amplifiers and many active devices deviate from this fundamental principle because of limitations imposed by internal power losses and linearity requirements. The output load impedance must be optimally selected so that the impedance is matched to achieve maximum output transmission under these constraints.

There are several types of matching methods, including matching using a conductor rod, matching using a conductor plate window in a waveguide, matching using an anti-reflective termination circuit, matching using a tapered waveguide, and matching using a transformer (λ/4 impedance converter). There is.

As illustrated in Figure 2, in the case of a 3-stub tuner, a conductor rod is inserted into the waveguide from the wide side of the waveguide for matching, and if a reflected wave exists in the waveguide, the reflected wave is canceled by the electromagnetic field caused by the conductor rod.

On the other hand, the plunger (or sliding short circuit) matching method is a matching method using a conductor plate window in the waveguide, and inserts a plate-shaped object such as a conductor plate at intervals at a right angle to the observation within the waveguide. This is a method of matching by appropriately selecting the distance to the load.

1 and 2 are cases where the 3-stub tuner 115 matching method and the plunger 121 matching method are used together. As in the above case, the stub tuner and plunger can be used interchangeably, or it is also possible to use only one stub tuner or plunger individually.

The present invention discloses a structure for installing such a plunger inside a resonance cavity (see FIG. 10).

If the microwaves are not absorbed by the applicator and are reflected back, these reflected microwaves can cause fatal damage to the magnetron. In this case, the direction of the reflected waves is changed to protect the magnetron, and the changed reflector is used as a dummy load. The energy is protected at (Dummy Load; 123). Refrigerant is supplied to the dummy load and the absorbed heat is released to the outside. The power propagated in the direction of movement of the microwave oscillated from the magnetron is called forward power (forward power: incident wave power, or traveling wave power), and the power that is not absorbed by the applicator and returns is called reflected wave power (reflected power: anti-wave power). Or it is called countercurrent wave power. If the reflected wave, which returns the microwave due to lack of alignment, is transmitted as is, it will damage the magnetron 111, which is an expensive product. Therefore, the direction of the reflected wave is changed separately by the magnet built in the circulator 113, and then the dummy load 123 ) to be absorbed into the absorber and disappear. Refrigerant is supplied to cool the heat-absorbing dummy load 123, the heat-generating magnetron 111, and the power supply module or power supply 121 so that their temperature does not rise, and the reflected wave energy absorbed by this refrigerant is transmitted to the outside. It is released.

To measure this forward power and reflected wave power, a directional coupler is installed between the circulator and the 3-stub tuner and measured using a power meter. By creating two ports that can measure microwaves in the waveguide, one port is directed toward the application device to measure forward power, and the other port is located in the opposite direction, i.e. toward the magnetron, and is used to measure reflected wave power.

Figure 3 shows a photograph of plasma generated when the power is varied in the range of 230 to 2000 Watt using these microwaves.

Figure 4 shows a photograph of plasma being generated by changing the pressure at various pressures, that is, 0.5, 1, and 2 torr, at a fixed power supply. It can be seen that as the pressure increases, the intensity and area of the plasma occurring within the dome decrease.

A waveguide, a conduit through which microwaves are transmitted, is an arbitrary structure that traps waves and guides them to propagate. Mainly, it refers to a hollow conductive metal tube through which electromagnetic waves travel.

Waveguides are applied to high-power, low-loss, millimeter wave systems, etc. Examples of use include radar, satellite repeaters, transmission of radio waves for broadcasting, or transmission of electromagnetic waves for heating and plasma generation, as mentioned above. It is used for. The structural form of the waveguide is a metal conductor (copper, brass, aluminum, etc.) that surrounds the electromagnetic wave waveguide space. The conductor thickness must be sufficiently secured to about 1 to 3 mm for mechanical strength, and the skin penetration depth must be several times greater in the application frequency range. Make it possible.

Typically, the size of the waveguide is determined according to the frequency of the microwave. Microwave frequencies permitted by the International Radio Association for industrial, scientific, and medical fields (Idustrial, Scitific, Medical: ISM) other than the communication frequencies mainly used for communication purposes are 2,450 MHz (2.45 GHz), 915 MHz, and 433 MHz. there is. For 2.45GHz microwaves, the waveguide standards used are mainly WR-284, WR-340, and WR-430, and for 915MHz, the WR-975 waveguide is used. Recently, the 433MHz band, which is a lower frequency than the existing frequency, is being used, and in this case, power is transmitted in the form of a cable rather than a waveguide. In addition, with the recent development of silicon-based Laterally Diffused Metal Oxide Semiconductor LDMOS or GaN or SiC semiconductors, ultra-high frequency oscillation by semiconductors is possible, and this semiconductor oscillation ultra-high frequency generation method is lightweight, simple, and easy to use. Because there are many advantages, such as frequency modulation, there is a groundbreaking transition from the existing magnetron oscillation method to the semiconductor method.

For the transmission of microwave energy, which is the irradiation part of the microwave oscillator, wires cannot be used because they radiate microwaves to the outside, and a hollow metal pipe called a waveguide is usually used. Depending on the cross-sectional shape, waveguides include square or circular waveguides and complex-shaped ridge-type waveguides. Square waveguides are generally used, and circular and ridge-type waveguides are used for special purposes.

Waveguides typically transmit microwaves with spreading characteristics by confining them in a square tube, and are usually made of brass or copper, which have good conductivity. Therefore, these square waveguides become heavy and large when installing a process or device using microwaves, and assembly becomes difficult. In particular, even in the case of low power, the device is heavy and large, making installation difficult and application limited. An example of a plasma generator using such a microwave delivery device (see FIGS. 1 and 2) is shown in FIG. 6.

In addition to waveguides, microwave transmission also includes coaxial cables (coaxial tubes).

of transmission line The shape is composed of two or more conductors, and the two conductors are separated by a dielectric . The energy transfer method transfers energy in the form of round-trip current/voltage waves flowing through two conductors. When viewed in the form of electromagnetic waves, the form of radio waves is transmitted as TEM waves and is transmitted by current waves and voltage waves. Since there is no cut-off frequency, it can operate from direct current to very high frequencies, and is applied to low output and somewhat low frequencies, but is inefficient in ultra-high frequency bands due to skin effect and dielectric loss. Coaxial cable is a type of transmission line. Typically, two-wire parallel cables have a defect in which the effective resistance of the conductor increases at high frequencies due to the skin effect. Coaxial cable was developed to compensate for this problem.

There is a problem that the wave easily escapes to the outside as the frequency increases. In the case of frequencies above 60Hz, a closed waveguide is used to solve this problem. Closed waveguide methods include Co-Axial Cable, Metallic WaveGuide, and Strip Line.

Looking at the coaxial cable structure as illustrated in Figure 5, Two cylindrical conductors and a dielectric share a central axis, and the inner conductor is used for actual signal transmission. The external conductor is a net-shaped shield (braided or foil) made of aluminum/copper, and is made by filling the gap between them with a dielectric/insulator (polyethylene, etc.) to separate them. On the outside, the outer shell is covered with a cable jacket (vinyl, polyethylene, etc.). In order to transmit a higher frequency signal than a regular power transmission line or wire (a cable made of two twisted wires), the inner conductor is surrounded with insulation and shielded with an outer conductor so that the waves do not escape the cable. To reduce outgoings

When transmitting ultra-high frequencies above 3 GHz through a coaxial cable, the resistance of the copper wire increases and the dielectric loss of surrounding insulators also increases, resulting in increased transmission loss and reduced transmission efficiency in cases above 3 GHz. It is usually used to transmit the output signal of a power monitor. It is flammable and easy to handle, but the transmission power tolerance of microwave transmission energy is small, about 300W (VSWR=about 3), so it is not used in microwave devices with a power of more than 300W.

The difference between a transmission line and a waveguide is that a transmission line (including coaxial lines) that propagates by TEM mode can also be viewed as a waveguide, but in general, propagation of radio waves by TEM mode is not called a waveguide. In other words, only the propagation by TE mode (Transverse Electric Field/Wave, transverse electric field) and TM mode (Transverse Magnetic Field/Wave, transverse magnetic field) is referred to as a waveguide.

When using low-power microwaves, there is a method of installing them by using coaxial cables instead of waveguides to reduce volume and weight.

The plasma generator using microwaves illustrated in FIG. 6 will be described as an example as follows. If the magnetron (M), circulator (C), dummy load (D), etc. are outside, they take up separate space, so these M-C-D are installed inside the power supply device to integrate and unify them. As a result, significant space reduction is possible, and the power supply to the magnetron is directly connected and supplied internally, and cooling lines are shared, which has the advantage of shortening the moving wire and making installation easier. 3-Stub tuner (matcher or matcher) is located externally because it must operate the stub for matching in the case of manual operation.

In the case of Figure 6, a coaxial antenna, a dome-shaped microwave vacuum window, a resonance cavity, and a plasma formed on a substrate are illustrated.

In the case of an automatic tuner, as illustrated in FIG. 7, a plurality of stubs are automatically adjusted to match the impedance, so in this case, it is possible for the tuner to be located inside the power supply.

In the present invention, since the tuning and matching function is performed by a plunger plate within the cavity, a stub tuner is not necessary, and if matching is not achieved with the plunger, the tuning and matching function is performed by merging.

In the case of Figures 6 and 7, there is an inconvenience in that a waveguide must be used to connect from the isolator (the sum of the circulator and the dummy load is called an isolator) to the applicator. To solve this problem, a coaxial cable-waveguide conversion adapter (WCCA, see FIG. 8) and a coaxial cable can be used. In this case, the additional advantage is that the installation space is reduced and installation is facilitated. Waveguides and coaxial cables can be converted to each other. In other words, it is possible to switch to a waveguide-coaxial or coaxial-waveguide form. Therefore, it is more accurate to label it as WCCA, or CWCA (Waveguide-CoAxial cable to Conversion Adaptor).

If ultra-high frequencies are oscillated using a semiconductor method and the plasma applicator is in the form of a waveguide, plasma is generated using a coaxial cable-waveguide conversion adapter. The shape of the waveguide-coaxial cable conversion adapter is illustrated in Figure 8.

Figure 7 is an example diagram showing a case where an automatic tuner (matcher) is installed inside a power supply and plasma is generated using a waveguide-coaxial cable conversion adapter. As in this case, the present invention presents a method of generating plasma by simplifying the transmission part of a microwave device using a waveguide-coaxial cable conversion adapter (WCCA), a coaxial cable, and a CWCA. When a waveguide is used to transmit ultra-high frequencies, the device is heavy and large, making assembly difficult. Using the above-described method can solve this difficulty.

On the other hand, if the device is configured so that a 3-stub tuner is not required in the microwave transmission path, the microwave application device can be greatly simplified. In addition, if the frequency is modulated instead of a tuner without using a tuner, or if a plunger is used and fused with an applicator to match impedance, the microwave transmission device can be used in a much smaller size. The present invention includes such a frequency modulation method and an applicator method equipped with a plunger.

Figure 9 is a configuration diagram of a large-area plasma generator according to the present invention.

Existing plasma equipment, including semiconductor equipment such as plasma etcher, asher, and CVD equipment, has mainly used 13.56MHz due to the convenience of use and the efficiency of plasma (compared to direct current plasma). Compared to the 13.56 MHz RF, the use of ultra-high frequency 2450 MHz, which has a frequency 180.67 times higher, has been used in some processes because it has an advantage in generating high-density plasma, but there are problems such as the difficulty in enlarging the area and the heavy and heavy-duty installation of ultra-high frequency transmission devices such as waveguides. Due to inconveniences in terms of use, high-density plasma has not been activated despite its many advantages. In order to improve and improve these points, the present application discloses a new plasma applicator, a cooling-movable resonance cavity, an antenna, and a method and device using semiconductor ultra-high frequency oscillation. These methods and devices allow for larger plasma areas, easier device installation, and increased utilization.

In the existing case, it has been difficult to enlarge the area of ultra-high frequency plasma because the coaxial antenna for ultra-high frequency transmission is small (FIGS. 4 and 6) or the plasma applicator must be smaller than the width of the waveguide (FIG. 3). In this respect, 915MHz has a relatively larger waveguide size than 2450MHz (for 915MHz, the internal size of the waveguide WR975 is 247.65*123.83 mm: for 2450MHz, the internal size of the waveguide WR340 is 86.36*43.18mm), so it can generate a larger plasma.

However, the plasma size is still limited by the waveguide size. In the present invention, it is possible to form a large-area plasma larger than the substrate by forming a cavity inside the reactor and using the substrate holder itself as an antenna, and tuning and matching is possible by moving the plunger plate inside the cavity.

According to Figure 9, the large-area plasma generator 300 according to an embodiment of the present invention may include a reactor 310 provided with a vacuum chamber 311 therein.

The reactor 310 can be used in various processes. Typically, in the semiconductor production equipment field, it is used in plasma asher, plasma etcher, and plasma deposition equipment. In the case of plasma asher, radicals are generated in the plasma zone and transferred to the wafer surface in a downstream manner, and a higher density plasma is required to remove photosensitive polymers that have become difficult to remove due to curing. In the case of deposition or etching equipment, a high-density, large-area plasma source is needed to improve faster deposition and etching rates.

The interior of the reactor 310 must usually maintain a vacuum in order to generate plasma, and gas is supplied through a gas inlet pipe for reaction.

Ultra-high frequency waves must be transmitted to the inside of the reactor 310 to generate plasma. Usually, ultra-high frequencies are generated in a magnetron (or semiconductor) and transmitted through a waveguide (or coaxial cable). The place where energy is transmitted to the end is called a cavity, and a resonance phenomenon occurs within this cavity, which is called a resonance cavity (in the present invention, a lifting cavity). It is decided to name it). Sometimes it is transmitted using only a waveguide, sometimes it is transmitted by configuring a coaxial antenna (by connecting the power source to a coaxial cable), and sometimes it is transmitted by configuring a waveguide-antenna. The case of transmitting ultra-high frequencies using a waveguide-coaxial antenna is illustrated in FIGS. 4, 6, and 7, and the case of transmitting ultra-high frequencies without a coaxial antenna is illustrated in FIG. 3. Even in the case of an atmospheric pressure ultra-high frequency torch, it is possible to generate plasma without a coaxial antenna.

A sight glass 312 may be installed on one external side of the reactor 310 to measure temperature and monitor the status of plasma generation from the outside. Temperature can be measured by observing light coming from the plasma through the sight glass 312.

The sight glass 312 usually uses a thermocouple or the like to measure temperature, but when plasma is generated, temperature measurement is impossible due to interference from the plasma. Therefore, the plasma temperature is measured using a far-infrared thermometer, a non-contact temperature measurement method.

The sight glass 312 may be made of a material such as quartz or sapphire so that it can withstand high temperatures.

Meanwhile, the sight glass 312 is installed with a tube-shaped support (not shown) at the center, considering that the lifting plunger is raised and lowered by the second lifting member, which will be described later, and the sight glass is installed to match the upper center of the support. . Through this sight glass, the temperature is measured by observing the light coming from the plasma. This sight glass uses quartz or sapphire to withstand high temperatures, and uses an integrated sight glass for high temperatures that is brazed with metal. Alternatively, an O-ring groove may be formed in the lifting plunger, and a cooling pipe (or cooling tube) or cooling channel may be formed on the flange fastened to the O-ring groove to allow cooling.

The cooling pipe or cooling tube can be made of a metal tube and related metal swage lock or Cajun fitting without the need for oil sealing. Hereinafter, it is assumed that the cooling module is the same.

The large-area plasma generator 300 may include a substrate holder and antenna 320 disposed below the vacuum chamber 311. Here, the substrate holder and antenna 320 is provided to be connected to a substrate holder portion 321 provided with a seating surface 321a on which the substrate (object to be processed) 1 is mounted, and the ultra-high frequency oscillation member. It can be composed of an antenna unit 322 that introduces ultra-high frequencies provided from and transmits them to the substrate holder unit 321, thereby generating a large-area plasma (P) corresponding to the area of the seating surface on the substrate holder unit. .

The above-described substrate holder and antenna 320 is configured so that the substrate holder itself acts as an antenna instead of a coaxial rod in order to increase the plasma area.

The substrate holder portion 321 is a portion where the substrate 1 for processing is located. In addition, if a special ingredient is needed to promote the reaction occurring on the substrate, a reaction promotion plate (not shown) may be installed around the substrate. The reaction accelerating plate may be a part of the substrate holder portion 321, but may be distinguished from the material of the substrate holder portion. In addition to promoting the reaction, the reaction accelerating plate can also be used as an intermediate part to protect the substrate 1 or protect the substrate from special substances.

Here, if the substrate 1 or the reaction accelerating plate protrudes further than the seating surface 321a of the substrate holder part 321, arcing may occur and process defects may occur, so it is preferable to match it exactly with the seating surface 321a. do.

For reference, radio waves can be transmitted through a metal tube called a waveguide. There is no major problem using a coaxial line below 3 GHz, but above 3 GHz, transmission loss increases and transmission power decreases, making it unsuitable as a microwave transmission line.

A waveguide is the same as removing the central conductor of a coaxial line. There is only air (or vacuum) inside and there is no central conductor or dielectric that causes loss, so there is less attenuation of microwaves than a coaxial line.

A rectangular waveguide is called a spherical waveguide, and because there is no central conductor in a waveguide, the propagation method is different from that of a coaxial line. In a coaxial line, it propagates along the central conductor, and in a waveguide, it propagates while reflecting off the tube wall. Therefore, the propagation distance of the radio wave becomes longer, so even if the distance is the same, the propagation time takes longer inside the waveguide than on the outer wall. Due to tube wall reflection, either an electric field or a magnetic field exists in the direction of travel.

Mode refers to the form in which energy of a specific frequency is concentrated in a certain structure. In the case of a waveguide or transmission line, it refers to the form in which electromagnetic waves (in a specific frequency band) travel, and this is called the propagation mode. In the case of a mode in a resonator, it refers to the resonance frequency and the resonance type and is called a resonant mode. In particular, in resonance mode, energy is concentrated in a spatially limited area. This resonance mode plays a very important role in generating and maintaining high-density plasma.

This is related to the phenomenon where energy is concentrated at a specific frequency depending on the structural characteristics. What is important here is that the mode is ultimately determined by the shape of the structure, so in order for the user to use a specific mode, the structure must be designed so that the desired frequency energy converges to that mode.

There are three major propagation modes of high frequency: TEM, TE, and TM. This is determined by the direction in which the electromagnetic wave travels and whether the E (electric field) and H (magnetic field) fields are vertical.

TEM (Transverse Electromagnetic) is when the direction of movement and the E field and H field are perpendicular. Transmission Line modes include Microstrip, Stripline, Coaxial line, Coplanar Waveguide, and Parrarel Plate. Because the two metals progress in parallel in a certain direction, the E field and H field can exist vertically at the same time in the direction of progress.

If a magnetic field exists in the direction of travel, it is called a TE wave, and if an electric field exists, it is called a TM wave. In other words, TE (Transverse Electric) is when only the E field is perpendicular to the traveling direction, and TM (Transverse Magnetic) is when only the H field is perpendicular to the traveling direction. This is a mode formed in the case of a general metal waveguide. Unlike a transmission line, because the bounce effect of a specific field component of a plane wave occurs within a single metal pipe, either the E field or the H field cannot be perpendicular to the direction of travel. The TE and TM modes of these waveguides are automatically determined according to the structural characteristics, and electromagnetic waves can only be perpendicular to the direction of travel of either the E field or the H field.

To use a specific mode, the size of the waveguide is determined accordingly. In addition, there are not only two types of electromagnetic waves penetrating the waveguide: TE and TM waves, but there are several types of modes, so they are indicated by subscripts as TEmn and TMmn modes. Subscripts such as mn are coefficients meaning order and are related to wave number. Electromagnetic waves have the strongest resonance characteristics in half-wavelength units where voltage alternates between (+) and (-). This refers to the number of half-wavelengths per unit path.

In the present invention, a case in which ultra-high frequency energy in TE mode transmitted from a waveguide is converted to TM mode using the substrate holder and antenna 320 to transmit ultra-high frequency energy is exemplified. The present invention includes generating high-density plasma by switching the above-described ultra-high frequency mode. The present invention does not only include the case of switching to TM mode. In cases other than the cylindrical cavity, other modes can be used, and since a resonant cavity can be formed in these modes, the converted mode is not limited to a specific mode. In other words, it is not limited to specific modes such as the TM mode mentioned above. Here, the frequencies used are 2450 MHz, 915 MHz, and 433 MHz permitted by ISM, but are not limited thereto.

Meanwhile, the substrate holder and antenna 320 can be installed to be lifted up and down by the third lifting member 330. In other words, the reason for raising and lowering the substrate holder and antenna 320 is to optimize the contact distance between the seated substrate 1 and the plasma (P). Here, optimization refers to deriving the best process results such as uniformity, reaction speed, and minimization of damage, and the distance between the substrate and the plasma is one of the very important variables. Important process variables within the reactor include pressure, electric power, gas flow rate, and gas supply type. For example, in the case of a plasma ashing process, the substrate spacing and the bias voltage applied to the substrate are very important to reduce plasma damage, and damage can be minimized by adjusting this.

In this embodiment, the third lifting member 330 can be lifted up and down on the third fixed flange 331 coupled to the bottom of the reactor 310 and on the lower side of the third fixed flange, as shown in FIG. 13. A third lifting flange 332 is disposed, the lower part is supported by the third lifting flange 332, and the upper end penetrates the third fixed flange 331 and the bottom of the reactor 310 to serve as the substrate holder and A third lifting rod 333 in the form of a pipe connected to the antenna 320, the upper part is connected to the third fixed flange 331, and the lower end is an arm tab block (332a) provided on the third lifting flange 332. A third screw rod 334 engaged with the third screw rod 334 is installed on the upper end of the third screw rod 334, and the third screw rod 334 is rotated through a rotation operation to raise and lower the third lifting flange 332. A third lifting control knob 335 that allows the device to be raised and lowered, the upper end of which is supported by the third fixing flange 331, and the lower part of which penetrates the third lifting flange 332, a third guide rod 336 and , It may be composed of a third guide block 337 that is installed on the third lifting flange 332 and guides the raising and lowering of the third guide rod 336.

As another embodiment, the third lifting member 340 includes a third fixing flange 341 coupled to the bottom of the reactor 310, and directly below the third fixing flange, as shown in FIG. 16. A third lifting flange 342 is arranged to be able to lift, the lower part is supported by the third lifting flange 342, and the upper end penetrates the third fixed flange and the bottom of the reactor 310 to serve as the substrate holder and antenna. A third lifting rod 343 in the form of a pipe connected to (320), the upper part is connected to the support block 342a of the third fixed flange 342, and the lower end is an arm tab block provided on the third lifting flange ( A plurality of third screw rods 344 engaged with 342a), a third drive chain gear 345 rotatably installed on the upper part of one of the support blocks 342a, and the support block 342a ), a plurality of third driven chain gears 346 rotatably installed on the upper part of the remaining support blocks where the third driving chain gear 345 is not installed, the third driving chain gear 345 and the third The third chain 347 connecting the driven chain gear 346 to interlock, and the third drive chain gear 345 are installed on the support block, and the third drive chain gear 345 and the support block ( It will be composed of a third whim wheel 348 integrally fixed with 342a), a third whim gear 349 engaged with the third worm wheel, and a third rotary handle 349a for rotating the third worm gear. It may be possible.

Considering that the temperature of the substrate holder-antenna 320 may increase because it is in contact with the plasma P, it is desirable to install cooling units in necessary areas to cool the substrate holder and antenna 320. As another example, a configuration in which a cooling passage is formed inside the substrate holder and antenna 320 and cooling is performed by circulating coolant through this cooling passage may be applied.

Meanwhile, the waveguide 112 for transmitting ultra-high frequencies to the antenna unit 322 is used by oscillating from a magnetron or semiconductor at atmospheric pressure, so it is not used in a vacuum state. Since the waveguide 112 generates and transmits ultra-high frequencies at atmospheric pressure, there is no need for vacuum sealing or pressure sealing. As illustrated in FIGS. 3 and 6, even if the reactor 310 operates in a vacuum, the dielectric and metal parts of the reactor, such as a dielectric tube (see FIG. 3), a dielectric dome (hemisphere: see FIG. 6), and a dielectric plate, are installed in the middle. Just seal it.

However, in the present invention, the waveguide 112 is directly connected to the reactor 310, the substrate holder portion 321 is used as a coaxial antenna, and the gas flowing into the reactor 310 may be a harmful gas, and may be used under high pressure. Considering that it can be operated in , sealing or pressure sealing must be maintained in order to maintain the airtightness of the reactor 310. Therefore, a waveguide sealing window 112a is required (shown in FIG. 9).

Since the waveguide sealing window 112a may generate heat due to high frequency transmission, any one of quartz, Pyrex, sapphire, and alumina, which are high-temperature dielectrics, may be used, and the O-ring may also be made of a high-temperature polymer O-ring or a metal O-ring for sealing. can be used The O-ring made of the high-temperature polymer material may be, for example, a Viton O-ring or a Kalrez O-ring.

As illustrated in FIG. 11, the above problem can be solved by installing an adapter flange 112b for a waveguide sealing window in the middle of the waveguide 112.

The existing waveguide did not have an O-ring or a groove for installing an O-ring, but in the present invention, an O-ring groove (112c) can be made on the front of the adapter flange (112b) and an O-ring (112d) can be installed. On the front side of the adapter flange, a step is created as much as the thickness of the dielectric window so that the dielectric window 112e can be embedded, an O-ring groove is formed here, an O-ring is recessed, and the dielectric window 112e is installed.

Then, in order to fix the dielectric window 112e and the O-ring 112d, a cover 112f with a hole equal to the inner size of the waveguide sealing window 112a is fixed. The fixing screw for the cover forms a female tab between the regular waveguide holes on the adapter flange, and embeds the male screw in the cover (112f) so that it does not protrude, and fastens it with the push tab.

On the other hand, when using a polymer O-ring, waveguide cooling may be necessary. In this case, cooling can be achieved by forming a cooling channel groove on the edge of the waveguide 112, or by welding a circular or square tube to the edge.

As another example, the high-temperature dielectric plate is used after integrating the waveguide by brazing (brazing) it with a metal (any one of aluminum, brass, copper, and stainless steel). This makes it possible to use a waveguide sealing window that is flattened by creating a step equal to the thickness of the dielectric plate, reprocessing it for joint welding, and then brazing it.

The waveguide sealing window 112a may be coupled to a connection portion of the waveguide 112 for ultra-high frequency transmission. Waveguide 112 may enter from the side as illustrated in FIG. 9 . As another method, it is also possible to install it integrally with the support of the antenna unit 322 and install it on the lower flange.

This is similar to the ultra-high frequency transmission device part at the top of Figure 6. In both cases, the antenna support must be sealed to vacuum or gas seal the reactor and must be cooled with a cooling pipe or cooling channel to withstand high temperatures.

As illustrated in FIG. 9, it can be seen that ultra-high frequencies are transmitted through a waveguide, and the ultra-high frequencies are connected to a substrate holder antenna connected to the waveguide and then radiated and transmitted into the cavity. In addition to the waveguide method described above, the ultra-high frequency transmission method can also be transmitted through a coaxial connector. This application also includes a method and device for transmitting ultra-high frequencies using a coaxial connector and a coaxial cable instead of a waveguide. In this case, it is connected to the coaxial tube rod connected to the synchronous plate holder antenna through a coaxial connector. In the semiconductor method, it is usually output in the form of a coaxial connector, so it is connected using a power supply coaxial connector, coaxial cable, etc. In the magnetron method, it is usually connected in the form of a waveguide, so it is connected in the form of a waveguide. However, in the case of low power, for convenience, a CWCA (waveguide-coaxial cable conversion adapter) - copper cable - coaxial antenna is used to connect between the power supply and the antenna. In the case of high power, both magnetron and semiconductor methods use waveguides to transmit microwaves to the antenna.

The large-area plasma generator 300 may include an elevation cavity 350 installed on the upper side of the vacuum chamber 311. The lifting cavity 350 has a shape with an open bottom and serves to shield the upper space of the substrate holder part 321 when it is lowered to form a resonance chamber in which plasma is generated. That is, it is possible to generate high-density plasma by limiting the space size by the lifting cavity 350.

The lifting cavity 350 may have various external shapes, such as cylindrical, square, oval, hemispherical, or elliptical hemisphere, and the lifting plunger below can be manufactured in the same shape according to the corresponding shape. For example, when a deposition operation is performed on a hemispherical substrate, the lifting cavity 350 can be formed in a cylindrical shape, and the lifting plunger built into it can be formed in a hemispherical shape identical to the shape of the substrate. Accordingly, the deposition yield can be increased by conforming the morphology.

The wall of the lifting cavity 350 may be made of metal or a composite material containing metal components.

In addition, the lifting cavity 350 must be capable of being lifted up and down by the first lifting member 360 for interaction between the plasma P and the substrate 1. In the case of high density, considering that various charged particles or radicals may affect the film and base materials on the surface of the substrate 1, the lifting operation of the lifting cavity 350 is very important.

In addition, the temperature of the lifting cavity 350 may increase due to the high density plasma (P) inside, so the cooling effect is also important. For cooling, a cooling passage may be formed inside the wall of the lifting cavity 350, and coolant may be circulated through the cooling passage.

Meanwhile, the refrigerant for cooling the lifting cavity 350 and the refrigerant for the lifting plunger below may be circulated separately.

In contrast, the cold inlet can be branched to a splitter so that the refrigerant passing through one flow path flows into the lifting cavity 350, and the refrigerant passing through the other flow path can be separated and supplied to the lifting plunger. The refrigerant that is heat-exchanged and discharged from each part can be configured to mix in the mixing section and flow out to the outside.

The refrigerant may basically be water, but alternatively, a separate circulating liquid such as silicone oil or a low-temperature refrigerant may be used.

Since the gas injected into the lifting cavity 350 may be highly toxic or corrosive, sealing must be performed to prevent external leakage. Except for atmospheric pressure plasma, most low-temperature plasmas must maintain a vacuum and therefore must be vacuum sealed. Here, the flow rate of gas injected into the lifting cavity 350 can be controlled using an external mass flow controller (MFC) (not shown).

A gas distribution unit (not shown) may be installed at the end of the gas inlet part of the lifting cavity 350. The gas distribution unit can improve uniformity when gas is introduced by evenly distributing the gas and allowing it to flow into the resonance chamber (C). Typically, the form of fluid flowing in through a tubular gas inlet is a lamina flow, so the moving speed of the center is fast, so the gas distribution part is essential for a uniform inflow speed.

Here, the gas distribution unit is connected to the gas inlet, and the material may be metal or dielectric (quartz, alumina, sapphire), etc., and various types of composite distribution layers (Multi- Story Gas Distributor) may be applied.

As an example of a gas distribution unit, it is also possible to install a gas distribution unit on the plate 380 of the plunger plate at the end of the gas inlet. Install a gas pipe inside the metal support pipe 393 and allow gas to flow into it. Since the sight port for measuring plasma temperature must be installed at the upper center of the support pipe 393, the gas pipe can be fastened and connected to the side of the metal support pipe. In this case, the gas distribution hole must be sufficiently small (e.g., smaller than Lamda/4) to prevent ultra-high frequencies from escaping.

Meanwhile, the first lifting rod below may be applied to the gas inlet of the lifting cavity 350.

In this embodiment, the first lifting member 360 for lifting the lifting cavity 350 includes a first fixing flange 361 coupled to the upper surface of the reactor 310, as shown in FIG. , a first lifting flange 362 disposed to be able to be lifted on the upper side of the first fixed flange, the upper part is supported on the first lifting flange 362, and the lower part is connected to the first fixed flange 361 and the reactor ( A first lifting bar 363 in the form of a pipe that penetrates the upper surface of 310 and is connected to the upper part of the lifting cavity 350, the lower part is connected to the first fixed flange 361, and the upper end is connected to the first lifting cavity 350. The first screw rod 364 is engaged with the arm tab block 364a provided on the flange 362, and the first screw rod 364 is rotated by rotating the first screw rod 364 while being installed on the upper end of the first screw rod. A first lifting control knob 365 that allows the first lifting flange 362 to be raised and lowered, the lower end of which is supported on the first fixing flange 361, and the upper part of which penetrates the first lifting flange 362. 1 It may be composed of a guide rod 366 and a first guide block 367 that is installed on the first lifting flange 362 and guides the raising and lowering of the first guide rod 366. The first guide rod is shown as being installed in plural pieces to support the weight and enable stable elevation, but it may also be installed as a single guide rod.

As another embodiment, the first lifting member 370 includes a first fixing flange 331 coupled to the upper surface of the reactor 310, and an upper side of the first fixing flange, as shown in FIGS. 14 and 15. A first lifting flange 332 arranged to be capable of being raised and lowered, the upper part is supported by the first lifting flange 372, and the lower end penetrates the first fixed flange 331 and the upper surface of the reactor 310. The first lifting rod 373 is connected to the upper part of the lifting cavity 350, the lower part is connected to the support block 371a of the first fixed flange 371, and the upper end is connected to the first lifting flange 372. A plurality of first screw rods 374 engaged with the provided arm tab block 372a, a first drive chain gear 375 rotatably installed at the lower part of one of the support blocks 371a, and Among the support blocks 371a, a plurality of first driven chain gears 376 rotatably installed in the lower portion of the remaining support blocks on which the first drive chain gear 375 is not installed, and the first drive chain gear 375 ) and the first driven chain gear 376 are connected to each other to interlock, and the first driving chain gear 375 is installed on a support block on which the first driving chain gear 375 is installed. and a first whim wheel 378 integrally fixed with the support block 371a, a first whim gear 379 engaged with the first worm wheel, and a first rotary handle 379a for rotating the first worm gear. ) may also be composed of.

Meanwhile, in both of the above embodiments, a bellows (corrugated pipe) (b) for sealing may be installed around the first lifting rod (363,373) in the section between the first fixing flange (361,371) and the first lifting flange (362,372). . The bellows (b) can maintain the seal while expanding and contracting according to the lifting operation of the first lifting flanges (362, 372).

Bellows (corrugated pipe) include formed bellows and welded bellows. Welding bellows are manufactured by welding ultra-small thin plates of 0.05 to 0.2 mm into a two-layer or multi-layer diaphragm depending on the application using inert gas on a non-consumable electrode and arc heat generated between the base materials. It has elasticity, pressure resistance, and bending. Although it has excellent properties, it has the disadvantage of being expensive because it is manufactured by hand. Molded bellows is made by making a tube from a thin plate material with excellent elongation, then inserting a bellows molding mold on the outside and applying high fluid pressure to the inside of the tube to form it. It has the advantage of being able to be produced quickly and in mass quantities, but like a welded bellows, the shape of the wrinkles is different. It has the disadvantage of poor elasticity, pressure resistance, and bendability as it cannot be configured in various ways.

When applying a wider range of cavity movement, a welding bellows is used. Bellows require the use of materials that resist corrosion in environments where corrosive fluids are used. The invention of this application includes all cases where a welded or molded bellows made of a material with strong corrosion resistance and good vacuum tightness is used in the transfer device.

On the other hand, as shown in FIG. 10, the lifting cavity 350 can form a magnetic field to increase the density and temperature of the plasma. To this end, by installing a permanent magnet (M) or an electromagnet on the outside of the lifting cavity 350 to create a magnetic field, the density of plasma can be increased through the rotational movement of electrons caused by the magnetic field. The magnets are preferably installed at the bottom of the outer wall of the lifting cavity, that is, in an area similar to the position of the substrate 1, and are located below the lowest moving height of the lifting plunger 380 to avoid interference.

The large-area plasma generator 300 may include a lifting plunger 380 that is installed to be lifted up and down within the inner region of the lifting cavity 350 by a second lifting member 390 and performs matching.

The lifting plunger 380 functions to match the plasma (P) in the resonance chamber (C), and its temperature may rise when it comes into contact with the plasma, so cooling is necessary. For cooling, the cooling module is brought into contact with the necessary part of the lifting plunger 380 to cool it.

The cooling module can use a circular or square fluid pipe and can be fixed to the lifting plunger 380 by welding.

The fluid pipe of the cooling module may be made of a high-temperature polymer or metal material such as Teflon or metal (e.g., copper plate, stainless steel 304, or stainless steel 316 material) to withstand high temperatures. Additionally, the fluid pipe is preferably made of a flexible material so that it can be easily installed even in a narrow space.

Here, a plurality of fluid pipes can be installed and used in combination in the required cooling unit. Several cooling tubes are distributed by a splitter (divider) and supplied to each cooling tube, or the refrigerant that comes out through the outlet after cooling is combined and discharged through a mixer, etc., and the discharged refrigerant is sent to the chiller. It can be re-cooled by a cooler (not shown) and used in circulation. If the chiller is not used, the discharged refrigerant is discarded and completely new refrigerant can be introduced. However, since proper cooling is difficult with this method, it is desirable to use a chiller that recovers waste heat and recirculates it. The external chiller cooler uses a refrigerant circulation pump and has a temperature control function, so it is advantageous to set the desired temperature value and send it to a desired location, such as the cooling pipe of the lifting cavity 350 and the lifting plunger 380 in the reactor 310. do.

Unlike the cooling structure using the cooling tube described above, a method in which the wall of the lifting plunger 380 is composed of a double wall and coolant is circulated inside the double wall can also be applied.

In this embodiment, the second lifting member 390 for lifting the lifting plunger 380 is installed on the first lifting flange 372, as shown in FIG. 12, to lift and lower the first lifting flange. A second fixed flange 391 that is lifted and lowered in conjunction with the second fixed flange 392, which is disposed on the upper side of the second fixed flange to be capable of being raised and lowered in a direction close to or distant from the second fixed flange, and the upper part is A second lifting bar 393 is supported on the second lifting flange, and the lower end is connected to the upper part of the lifting plunger 380 through the hollow part of the first lifting bar 363, and the lower part is connected to the second fixing flange. It is connected to (391), the upper end of which is engaged with the arm tab block (392a) provided on the second lifting flange (392), and the second screw rod (394) is installed on the upper end of the second screw rod, and is rotated by rotation. A second lifting control knob 395 that allows the second lifting flange 392 to be raised and lowered by rotating the second screw rod 394, the lower end of which is supported on the second fixed flange 391, and the upper part of A second guide rod 396 penetrating the second lifting flange 392, and a second guide block installed on the second lifting flange 392 and guiding the raising and lowering of the second guide rod 396 ( 397).

In another embodiment, the second lifting member 410 is installed on the first lifting flange 372, as shown in FIG. 14, and is a second fixed flange that is raised and lowered in conjunction with the lifting and lowering of the first lifting flange. (411) and a second lifting flange 412 disposed on the upper side of the second fixing flange to be capable of being raised and lowered in a direction close to or distant from the second fixing flange 411, and the upper part is attached to the second lifting flange. It is supported, and the lower end is connected to the upper part of the lifting plunger 380 through the hollow part of the first lifting bar 373, and the lower part is a support block of the second fixing flange 411. It is connected to (411a), and the upper end is engaged with the arm tab block (412a) provided on the second lifting flange (412), a plurality of second screw rods (414), and one of the support blocks (411a) A second drive chain gear 415 rotatably installed at the bottom, and a plurality of support blocks rotatably installed at the bottom of the remaining support blocks on which the second drive chain gear 415 is not installed among the support blocks 412a. A second driven chain gear 416, a second chain 417 connecting the second driving chain gear 415 and the second driven chain gear 416 to interlock, and the second driving chain gear 415 A second whim wheel 418 is installed on a support block and is integrally fixed with the second drive chain gear 415 and the support block, and a second whim gear 419 meshed with the second worm wheel. , It may be composed of a second rotary handle 419a that rotates the second worm gear.

In common with the two above-mentioned embodiments, a bellows (b) for sealing may be installed around the second lifting rod (393,413) in the section between the second fixing flange (391,411) and the second lifting flange (392,412). The bellows (b) can maintain the seal while expanding and contracting when the second lifting flanges (392, 412) are lifted and lowered.

The lifting members and methods of FIGS. 12 and 14 can be applied interchangeably. For example, the lifting device of FIG. 14 can be applied to move the cavity, and the lifting device of FIG. 12 can be applied to the lifting device of the plunger.

Additionally, the knob in FIG. 12 and the handle 379a in FIG. 14 can be operated manually or automatically, and in the case of automatic operation, they can be operated by a motor.

The lifting device of the lifting plunger, lifting cavity, or substrate holder is very complicated, as shown in Figures 12, 14, and 13. In addition to the complicated method described above, a lifting device using a metal pipe with O-ring sealing is also possible in a simple method. The reason for using a lifting device using a metal pipe with O-ring sealing is, firstly, that it is used as a simple and convenient transfer device, and secondly, that this metal pipe can contain a refrigerant or a metal pipe that can flow refrigerant. The following part is moved to another part and described. On the other hand, for the cooling pipe (not shown), for example, a tab is placed on the outside, the inner surface of the lowest part is inclined, a rubber ring is placed on it, a metal ring is installed on it, and a tab is installed on the inside of the cap. It is put out and connected to the tabbed part on the outside of the mating part, and is rotated to move, and then moved downward to press the metal ring. This pressure presses the O-ring and presses the rod, sealing it in a biting manner. The reason for using these parts is that the reactor must withstand pressure such as a vacuum, so sealing is required and at the same time, it must be possible to move the position.

The sealing O-ring and metal ring are installed as a single unit, but a plurality of metal rings and O-rings can be installed to tighten the sealing.

A bias applicator (not shown) may be installed inside the second lifting rods 393 and 413 so that DC or RF power can be applied by installing an insulated wire. Since the bias application unit described above is installed within the second lifting rod, it is not affected by ultra-high frequencies. Externally, power is supplied after installation using a feedthrough for power supply to the high vacuum flange. The feedthrough uses an insulating material in the middle, so power is supplied to the metal penetrating the inside, so even if power is supplied by connecting to the flange, no current flows in the reactor connected to the flange. When supplying DC or RF power, install a filter and cover it with EMI shielding material to prevent interference from ultra-high frequencies. If it is necessary to separate the substrate holder through which ultra-high frequencies are transmitted, a thin (partially EMI shielded) dielectric is placed within the substrate and the substrate holder/antenna, and a bias is applied only to the substrate.

Alternatively, ultra-high frequency power can be pulsed to supply ultra-high frequency to the substrate holder and antenna, and by applying a bias to the substrate during the ultra-high frequency pulse-off time, the bias is applied to the substrate without being interfered with by ultra-high frequency, thereby reducing the charge in the plasma. By controlling the supply of particles (ions or electrons), the required process can be implemented.

1 to 8 mostly show an ultra-high frequency oscillation method using a magnetron. In this case, a 3-stub or 4-stub tuner is used by tuning manually or automatically, as shown in Figures 1 and 2.

At this time, when the stub tuner is installed, the plunger can be used without additional installation. In other words, the stub tuner and plunger are installed selectively and used by tuning or matching.

In the case of the present invention, since the lifting plunger 380 is movably installed in the lifting cavity 350, it is possible to tune (match) and use it without the stub tuner illustrated in FIG. 1.

On the other hand, in the case of unusual shapes such as complex-shaped substrates or cavity structures, the alignment may not be good by the lifting plunger 380. In this case, the lifting plunger 380 and the stub tuner can be used in combination. In the case of the present invention, all of the various matching methods described above may be included. In the case of a semiconductor oscillation method rather than a magnetron method, matching by frequency modulation is also possible.

The large-area plasma generator 300 may include a matching controller 420 that controls the lifting plunger 380 to be positioned at a point where reflected waves are minimized.

In this embodiment, as shown in FIG. 17, the matching controller 420 includes a measuring unit 421 consisting of a directional coupler and a power meter to measure reflected power, and storing the measurement value measured by the measuring unit. The storage unit 422 compares the measured value stored in the storage unit with the pre-entered set value, and when the measured value exceeds the set value, the genital plunger 380 is operated until the measured value falls below the set value. It may include a lifting control unit 423 that adjusts the lifting position.

That is, when tuning is performed by moving the lifting plunger 380 up and down, the position of the lifting plunger can be adjusted manually or automatically, and the lifting position of the lifting plunger 380 is automatically adjusted by a power device such as a motor. In this case, the optimal position where reflected waves are minimized can be automatically tracked using the position division tracking matching method.

As described above, the Multi-Position-Division-Search Tuning Method measures the reflected wave or impedance through the directional coupler 421 and feeds it back to the lifting control unit 423 so that the measured value is preset. If the value is exceeded, the position of the lifting plunger 380 is readjusted to track the optimal point where impedance is minimized.

If you divide a fine section from the beginning, the matching time may take a long time, so the entire section is subdivided into multiple sections, and the lowest section among these sections is found to perform the first matching.

Next, the first matching section is further subdivided into multiple sections, and the lowest section among these sections is found to perform the second matching.

Next, the second matching section is further subdivided into multiple sections, and the lowest section among these sections is found to achieve third matching.

In this way, the method of achieving n-order matching by gradually narrowing the internal space of the lifting cavity 350 is called the position division tracking matching method, and the present invention determines the point at which the impedance (or reflected wave) is lowest by this method. . The above method can be applied to the magnetron oscillation method, and the allowable high frequency in the magnetron oscillation method is 2,450 MHz. The MPDS TM method can also be applied to the semiconductor method. In the semiconductor method, the frequency is additionally variable, so a higher density plasma can be generated by detecting and tuning the resonance frequency while changing the frequency.

In the semiconductor oscillation method, the high frequency is variable from 2,400 to 2,500 MHz. In the case of variable frequency, the above-described frequency modulation tuning process is performed. Here, the variable range of the above-mentioned frequency may be wider, and this is determined by the frequency variable range that can be varied by the driver (or oscillator) chip that provides the frequency signal to the semiconductor chip that generates ultra-high frequencies. In the present invention, the case of 2,400 ~ 2,500 MHz is exemplified.

First, the lifting plunger 380 is located at an appropriate height where reflected waves are minimal.

Next, the reflected wave is measured while the frequency is fine-tuned within the range of 0.1 to 10 MHz, and based on this, comparative analysis is performed and fine-tuning is performed.

In this process, tuning within a fine frequency range from the beginning takes a lot of time, so it is adjusted through grouping search that divides the frequency into a large and small frequency section.

The frequency grouping search method initially measures the reflected wave by scanning the entire frequency range (bandwidth) divided by the frequency range width 1.

For example, scan 2,400 ~ 2,500 MHz, bandwidth 100 MHz to 10 MHz, which is frequency range 1, measure reflected waves in each frequency range of group 1, and search for area 1 where reflected waves can be minimized.

Next, when area 1 where the reflected wave of group 1 is minimized is detected, the frequency width 2 is searched by reducing the frequency width 2 to 1/n of the previous frequency width 1 (for example, 1/10, 1 MHz). The area detected in this way becomes group 2.

Next, if the range of frequency domain 2 is detected after the second search, this frequency domain 2 is reduced by 1/m and searched with a frequency width of 3 to find the optimal frequency (for example, 0.1 MHz, which is 1/10 of frequency domain 2). . The area detected in this way becomes group 3.

As an example, frequency areas 1, 2, and 3 are set in units of 10 MHz, 1 MHz, and 0.1 MHz, respectively, to find the optimal path that minimizes the reflected wave in a short time. This is called Frequency-Division-Gruoping Search Tuning Method (FDGS TM).

As before, since a certain level of matching position has already been determined by the movement method of the lifting plunger, it is possible to search for a more fine matching position using the frequency division grouping tracking matching method. Therefore, in the case of semiconductor ultra-high frequency oscillation, fine tuning and matching is possible without using a stub tuner or coaxial tuner.

Meanwhile, as another example, a method of forming optimal high-density plasma by tracking and setting the resonant frequency that maximizes the Q-factor will be described as follows.

A cavity structure capable of resonance was formed in the reactor by the substrate holder antenna and the cavity. If the inner wall of the reactor through which ultra-high frequencies are transmitted is made of metal, the inner wall of the reactor can be considered a type of cavity structure.

However, it is complicated because various parts and modules are installed inside the reactor, and related components are installed for sealing or vacuum extraction.

In order to avoid such a complicated shape and simplify it, a simple structure cavity was formed to confine ultra-high frequencies by a substrate holder/antenna and a cavity wall, as illustrated in FIG. 10.

Of course, Figure 10 is a structure when using a flat substrate such as a wafer, and may become complicated depending on the shape of the substrate. For example, it may be oval or oval-cylindrical.

When an ultra-high frequency signal is applied through a coaxial line or slot inside a metal container, resonance occurs at a specific frequency depending on the size of the cavity.

Typically, ultra-high frequencies move in TE mode in a waveguide structure, and then the mode is switched through a coaxial antenna. The mode inside the cavity is determined by the shape of the cavity.

When ultra-high frequencies transmitted in this way are trapped in the cavity structure surrounding the substrate holder and antenna, there are cases where resonance occurs. In the case of resonance like this, the formation of this cavity is called a resonating cavity, and the frequency at this time is called the resonance frequency.

Resonance is a physical phenomenon that means that when the periodicity of a structure matches the periodicity of a signal, the energy of the frequency corresponding to the period is preserved (or transmitted) without being lost. So, when you want to select only a specific frequency by using this resonance phenomenon, you can extract only that bandwidth without loss.

Accordingly, the resonating frequency varies depending on the shape or volume of the structure in Figure 10, but in the existing magnetron method, the 2450 MHz frequency is fixed, so it is impossible to vary the frequency. Of course, how sharp or broad the power transmitted at the 2450 MHz frequency is depends on the cavity structure or material of each manufacturer's magnetron.

On the other hand, in the case of ultra-high frequency oscillation by semiconductors, the frequency can be varied. For example, frequency changes are possible in the range of 2400 to 2500 MHz, that is, the bandwidth (and width) of 100 MHz. By using this to find and match the resonance frequency, it is possible to make the cavity resonate in the currently set cavity structure, and the algorithm is explained as follows.

To track the frequency of the resonant cavity, the concept of Q-factor (Q-factor: ‘quality factor’) is used. The present invention discloses a method of finding and finding the resonance frequency by tracking the highest point of the Q factor.

Q factor is a quantity that represents the quality of the resonant circuit and is also called 'quality factor'. The larger this value, the sharper the response curve to the frequency of the resonance circuit and the better the frequency selectivity. The Q factor can be measured with or without a load. In the case of a load, it must include the loss of the external circuit (for example, the signal source of the generator).

On the other hand, this is not the case when there is no load. The Q factor is as follows:

When there is no load: Q = ω (total stored energy) / (average power loss) =

= ω (W _E + W _M ) / P _L = Per cycle (stored energy / dissipated energy) where ω is the angular frequency (=2πf), W _E is the energy (stored in) the electric field, W _M is the energy stored in the magnetic field, and P _L is the average power delivered to the load.

If there is a load, the Q factor is called the Loaded Q Factor and is as follows.

With load: Q = ω (W _E + W _M ) / P _T

The power consumed is _P , which means it includes not only the load but also the power consumed in external circuits. In resonance, W _E = W _M , in other words The electric field energy and the magnetic field energy become the same, and the impedance has only a real part without an imaginary part.

No-load Q factor in resonance Q ₀ = 2 ω ₀ (W _E ) / P _L =2 ω ₀ (W _M ) / P _L

Here, ω ₀ is the angular frequency 2πf ₀ , where f ₀ is the resonance frequency.

In addition to the above methods, there are various other types of concepts and definitions, of which the most widely used definitions are as follows.

Q = f ₀ / ( f ₂ - f ₁ ) = 1 / B _F

Here, f ₁ and f ₂ are the frequencies at which the power when rising and falling, respectively, is half the maximum value at the resonance frequency f ₀ . B _F is the non-bandwidth (non-bandwidth), which is the ratio of the pass bandwidth and the resonance frequency of the tuning circuit in the receiver's tuning circuit.

The power consumed in a specific circuit is proportional to the square of the magnitude of the current I, and the power is proportional to [I ² ]. Therefore, [I ² ] is measured as power P and the frequency is changed (see FIG. 19).

Since the ultra-high frequency allowed for ISM purposes is 2450 MHz, the magnetron oscillation method is used by fixing the above frequency. If semiconductor oscillation ultra-high frequency is used, the frequency can be varied (for example, in the range of 2400 to 2500 MHz).

Using the Multi-Position-Division-Search Tuning Method (MPDS TM) mentioned above, the position of the lifting plunger is located at an appropriate height to reduce reflected waves to some extent.

Afterwards, the frequency is finely adjusted within the range of Δf=10 to 0.01 MHz, the current [I] is measured, and the power [I ² ] is calculated. I is the current in the plasma generated in the case of the present invention, and can be measured using a Langmuir Probe, etc.

Alternatively, forward power (and reflected power) can be measured using a directional coupler and power meter.

In the present invention, since tuning within a fine frequency range from the beginning takes a lot of time, a grouping search algorithm is disclosed that divides the frequency into a large section and a small section (by dividing the large section). For example, if frequency modulation is possible in the range of 2400 ~ 2500 MHz, the frequency variable range is bandwidth = 100MHz. Forward power is measured while changing the frequency change width from Δf = 1 to 0.01 MHz, and the measured forward power is fed back to the resonance frequency search PC.

If Δf = 0.01 MHz from the beginning, 10,000 frequency changes and forward power measurements are required. Since this method consumes a lot of time and therefore energy consumption, the problem is solved in the following way.

The frequency grouping search method initially scans the entire frequency range (bandwidth) by dividing it by frequency width 1 (=Δf ₁ ) and measures forward power (for example, 2400 ~ 2500 MHz, bandwidth 100 MHz is scanned at 10 MHz with frequency width 1). ) Measure the forward power in each frequency region of Group 1 and search for region 1 of the maximum forward power FP10 where this is maximized. When frequency region 1-0 (i.e. resonance frequency f _0i , i=1 stage) where the forward power of this group 1 is maximized is detected, frequency region 1-1 and frequency region 1-2 where this forward power is 1/2 Find . Frequency 1-1 area and frequency 1-2 area are likely to have a symmetrical structure centered on frequency area 1-0. . After finding the frequencies f11 and f21 that are closer to FP10/2 among the frequency areas 1-1 and 1-2, the Q factor Q1 is calculated (i.e., f _1i , f _2i , Qi = where i is found step by step is the order of the order).

The frequency range found in the first tracking process by reducing the frequency width 2 to 1/n (= frequency width 2 (Δf ₂ )) of the previous frequency width 1 (Δf ₁ ) (e.g., 1/10, 1 MHz). Search within the range 1-0. At this time, the detected area becomes frequency area 2-0 (i.e., resonance frequency f _0i , i=2 steps).

When the frequency range 2-0 range is found and the maximum forward power FP ₂₀ is calculated, a value that is 1/2 of this power is calculated. Search frequency area 1-1 with frequency width 2 (Δf ₂ ), which is 1/n of the primary frequency width, to find frequency area 2-1, which is 1/2 of the maximum forward power FP ₂₀ . Find frequency domain 2-2 using a method similar to the above. After finding the frequencies f ₁₂ and f ₂₂ that give a value closer to FP ₂₀ /2 among the frequency areas 2-1 and 2-2, the Q factor Q2 is calculated.

The frequency range 2-0 is reduced by 1/m to frequency width 3 (Δf ₃ ) (for example, frequency width 3 (Δf ₃ ) 0.1MHz = 1/10 of frequency width 2 (Δf ₂ ) = frequency width 1 ( Search for 1/100 of Δf ₁ ) to find the optimal frequency region 3-0 (i.e., resonance frequency f _0i , i=3 steps) that gives the maximum forward power FP ₃₀ (this detected region becomes group 3) ). Afterwards, the frequencies f13 and f23, which are 1/2 of FP ₃₀ , are found and the Q factor Q3 is calculated.

As an example, frequency width 1, frequency width 2, and frequency width 3 are set in units of 10 MHz, 1 MHz, and 0.1 MHz, respectively, to find the optimal path that maximizes forward power in a short time. In a similar way, after searching for f04, which gives forward power FP40, with a four-step frequency width 4 Δf ₄ (for example, 0.01 MHz), FP41 and FP42 can be determined, and thus f14 and f24 can be searched with an accuracy of 0.01 MHz. there is. This calculates the Q factor Q4.

Using the above method, in order to track the 2400 ~ 2500 MHz, 100 MHz bandwidth area with an accuracy of 0.01 MHz frequency width (Δf), the resonant frequency f0i (and f1i and f2i) must be found through 10,000 paths, but using the above method It is possible to detect the resonance frequency f0i using a method within 100 times. In the above example, the initial search frequency width 1 (Δf ₁ ) was searched in units of 10 MHz, which can be searched in units of 1 MHz. This is expected to take a little longer in the initial search and detection time.

The accuracy of the above-described resonance frequency search is determined by the frequency variable width that modulates the frequency of the driver chip that provides a frequency signal to the semiconductor chip that generates ultra-high frequencies.

As described above, the method of finding the resonant frequency is called the Frequency-Division-Gruoping Search Tuning Method (FDGS TM).

In addition to the method of tracking the resonant frequency by calculating [I ² ] or Q factor while changing the frequency described above, the first derivative of [I ² ] and Q factor with respect to frequency (=d[I ² ] It is possible to search for the resonance frequency by calculating /df and dQ/df) or, if necessary, by calculating the second derivative (=d ² [I ² ]/df ² and d ² Q/df ² ). .

In other words, at the highest point of the Q factor, the first derivative (=d[I ² ]/df) is a 0 value, which is the point where it changes from positive to negative. High accuracy is achieved by the ‘frequency division grouping tracking matching method’ described above. Detection of the resonant frequency f _0i is possible.

The present invention discloses a method of forming the highest density plasma by maximizing energy transfer by tracking, detecting, and determining the resonant frequency while varying the frequency as described above. This illustrates a method of determining the resonance frequency that gives the highest point of [I ² ] and Q factor (= (f ₂ -f ₁ )/f ₀ ) according to the above-mentioned frequency change, and the invention of this application is not limited to this method. .

If the resonance frequency is found using the above method and power is supplied according to the frequency, all power is input into the generation of plasma, thereby maximizing the Q factor. This makes it possible to generate the highest density plasma for each delivered power.

For reference, the semiconductor oscillation method described above is a method of oscillating microwaves using a semiconductor instead of a magnetron, as shown in FIG. 18. Since the semiconductor microwave generation module requires the supply of direct current power, an AC-DC conversion inverter 225 is used to convert the AC input power 231 supplied to the power supply 121 into direct current. The inverter supplies power to the power control board (Control Board: 221) and the semiconductor ultra-high frequency generation module (223).

Additionally, in order to set and control the microwave power to be supplied, the power control board exchanges signals with the outside world through a communication gateway (GateWay: 233). When the frequency is variable, an oscillator (not shown) serving as a signal generator generates a signal and provides it to the ultra-high frequency generator semiconductor. That is, the semiconductor ultra-high frequency generation module 223 includes an oscillator for signal generation and an amplifying semiconductor 222.

As shown in Figure 18, in the case of relatively small power, the output comes from the power supply through a coaxial cable, is converted into a waveguide form through a coaxial cable-waveguide conversion adapter, and is then connected to an applicator device, and from the waveguide again through a coaxial antenna. After all conversions are made, ultra-high frequency energy is delivered to the dome shape to generate plasma. Recently, high-power semiconductor power has also been released. As mentioned above, coaxial cables have a high loss rate when power increases, so when composing total power by merging multiple power modules inside a semiconductor power supply, a combiner, etc. is used. It is output in the form of a waveguide.

Microwave generators using these semiconductor devices have the following advantages.

1) 1 ~ Maximum allowable output (for example, maximum allowable power = about 500W) Accurate power adjustment is possible in 1W increments.

2) For example, in the case of 2,450 MHz, frequency modulation is possible up and down 50 MHz, that is, 2450 MHz ± 50 MHz (range 2400 ~ 2500 MHz). Of course, semiconductor devices with a wider frequency range are also possible.

3) Longer lifespan than the magnetron method.

4) A small isolator can be embedded inside the power supply, and when reflected waves are generated, the generator circuit is protected through immediate automatic reduction and interruption of the supplied power.

5) It is small in size and lightweight.

6) It is possible to accurately measure incident wave power and reflected wave power.

7) Ripple is low and power transmission efficiency is high.

8) For industrial applications that require multiple applicators (such as the need to cover a large area), this can be achieved. In other words, operation is possible by using a plurality of semiconductor power supplies or by branching one power supply into a divider and easily installing it on a plurality of applicators. Since it does not use a waveguide, it can be easily installed by connecting with a coaxial cable.

The high-power generation of semiconductor-type ultra-high frequency devices (combiner and coaxial-waveguide conversion adapter) and the generation of large-area ultra-high-density plasma by its transmission are explained. Currently, low-power ultra-high frequency power supplies (power) are produced using a single power semiconductor chip that can typically achieve an output of about 500 Watt. Since it is a method used by connecting a coaxial cable to the output terminal connector, it is currently mainly applied in low power areas of 500 Watt or less. Typically, in the case of ultra-high frequency generation using a semiconductor method, a coaxial cable (see Figure 5) is used for low power of 300 to 500 W or less, but for high power of 300 to 500 W or more, it is difficult to apply a coaxial cable and the microwave must be output in the form of a waveguide. . In other words, for power within 500W, coaxial rods, coaxial connectors, and coaxial cables are used.

The problem with coaxial cables that cannot be used in the case of high power is that the ultra-high frequencies coming from each semiconductor are combined and modularized using a coaxial rod and a primary combiner (mixer or combiner), and at the final stage, the ultra-high frequencies coming from the modules are combined in a secondary combiner. After combining, a coaxial-waveguide conversion adapter (CoAxial-Waveguide Converter) is used to oscillate high-output ultra-high frequency waves in the form of a waveguide at the end.

To explain in detail, as a method of making a semiconductor-type high-power power for ultra-high frequency, the above-mentioned single low-power ultra-high frequency power is modularized using a plurality of modules, and a plurality of modules configured in this way are combined to form a 1-2Kw high-power power with a cooling function. These modules are combined (for example, about 2 sets) and placed in a cooled box to create greater power, and then placed in a standard-sized rack to supply the semiconductor-type high-frequency high-power device desired by the user. .

Currently, semiconductor-type ultra-high frequency power (power supply) can be produced up to 10Kw in the 2450MHz range, and in the case of 915MHz, 60~90KW can be produced. In the future, it is possible to further combine and improve the above-mentioned methods to produce a power supply with a larger capacity.

In the case of the semiconductor method, it has a relatively long lifespan and is more stable than the magnetron method, so it can be used for a very long time and has the great advantage of being able to vary the frequency. Typically, within 10kW, 2450MHz is used, and above 10kW, 915MHz is used.

In the present invention, by changing the frequency, detecting and setting tuning the resonant frequency by changing the frequency, not only in the case of low power of 1KW or less, but also in the case of high power of 100KW or more, by using the above-described method, Ultra-high-density plasma can be generated, and this includes all devices and methods.

In the case of a waveguide-type plasma applicator as illustrated in FIG. 18, it is applied using a coaxial cable-waveguide conversion adapter 153.

When the semiconductor oscillation ultra-high frequency generator is operated, if the semiconductor microwave module is not properly matched, reflected waves are generated, and a cooling line 131 is installed to cool them. In the case of air cooling, installation is relatively simple because only a fan is installed without the need to supply cooling water, and the air supplied from the fan contains heat and is discharged to the outside of the power supply. However, when the reflected wave power is large, it generates a lot of heat and can be fatal to microwave semiconductor devices, so it is safer to install a device-type isolator and coolant line. In the case of large power, if a semiconductor device-type isolator cannot be used due to insufficient capacity, a waveguide-type isolator cooled by water cooling can be installed externally to prevent transmission of reflected waves.

The various embodiments described above are merely illustrative, and those skilled in the art will recognize that various modifications and other equivalent embodiments are possible.

Therefore, it will be understood that the present invention is not limited to the forms mentioned in the detailed description above. Therefore, the true scope of technical protection of the present invention should be determined by the technical spirit of the attached patent claims. In addition, the present invention should be understood to include all modifications, equivalents and substitutes within the spirit and scope of the present invention as defined by the appended claims.

1: substrate
111: Magnetron
112: Waveguide
113: Circulator
115: 3-stub tuner (3stub-tuner)
117: Applicator
119: Plunger or Sliding Short Circuit
121: Microwave Power Supply
131: Cooling Line
151: Co-Axial Cable
153: Connector
155: Co-Axial Cable-Waveguide Conversion Adaptor
171: Plasma
173: Antenna
161: [Wafer, etc.] Substrate [Substrate]
165: Holder with cooling line (Wafer Holder with Cooling Line)
219: Oscillator signal generator
221: Automatic control board (Control Board)
223: Semiconductor microwave generation module
225: AC-DC conversion inverter
231: Power supply unit
233: Gateway for External Communication
300: Large-area plasma generator
310: reactor
311: Vacuum chamber
312: sight reading
320: Substrate holder and antenna
321: Substrate holder part
322: antenna unit
330, 340: Third lifting member
350: lifting cavity
360, 370: First lifting member
380: lifting plunger
390, 410: Second lifting member
420: Matching controller
P: plasma

Claims (14)

A reactor equipped with a vacuum chamber inside;
a substrate holder portion disposed below the vacuum chamber and provided with a seating surface on which a substrate is mounted; An antenna unit provided to be connected to the substrate holder unit, which introduces ultra-high frequencies provided from the ultra-high frequency oscillation member and transmits them to the substrate holder unit, thereby generating a large-area plasma corresponding to the area of the seating surface on the substrate holder unit. Board holder and antenna;
A lifting cavity is installed on the upper side of the vacuum chamber to be able to be lifted by a first lifting member, and has an open lower part to shield the upper space of the substrate holder part when lowered, thereby forming a resonance chamber in which plasma is generated;
a lifting plunger that is installed to be capable of being raised and lowered within the inner area of the lifting cavity by a second lifting member and performs matching; and
a matching controller that controls the lifting plunger to be positioned at a point where reflected waves are minimized;
A large-area plasma generator comprising a.
In claim 1,
A large-area plasma generator, characterized in that a cooling passage is formed inside the wall of the lifting cavity, and a refrigerant is circulated through the cooling passage to cool the lifting cavity.
In claim 1,
A large-area plasma generator, characterized in that the lifting plunger and the substrate holder and antenna are cooled by contacting a cooling module with each unused portion.
In claim 1,
The first lifting member,
A first fixed flange coupled to the upper surface of the reactor;
a first lifting flange disposed above the first fixing flange to be capable of being raised and lowered;
A first lifting rod in the form of a pipe whose upper part is supported on the first lifting flange and whose lower end penetrates the first fixed flange and the upper surface of the reactor and is connected to the upper part of the lifting cavity;
a first screw rod whose lower part is connected to the first fixing flange and whose upper end is engaged with an arm tab block provided on the first lifting flange;
a first lifting control knob installed on the upper end of the first screw rod and allowing the first lifting flange to be raised and lowered by rotating the first screw rod through a rotational operation;
a first guide rod whose lower end is supported by the first fixing flange and whose upper end penetrates the first lifting flange; and
a first guide block installed on the first lifting flange and guiding the raising and lowering of the first guide rod;
A large-area plasma generator comprising a.
In claim 4,
The interior of the first lifting rod is in communication with the cooling passage of the lifting cavity, and an inflow port is provided on one side through which refrigerant flows into the interior of the first lifting rod.
In claim 4,
A large-area plasma generator, characterized in that it further comprises a bellows for sealing around the first lifting rod in the section between the first fixing flange and the first lifting flange.
In claim 4,
The second lifting member,
a second fixed flange installed on the first lifting flange and raised and lowered in conjunction with the raising and lowering of the first lifting flange;
a second lifting flange disposed on an upper side of the second fixing flange to be capable of being raised and lowered in a direction close to or distant from the second fixing flange;
a second lifting bar whose upper part is supported on the second lifting flange and whose lower end is connected to the upper part of the lifting plunger through a hollow part of the first lifting bar;
a second screw rod whose lower part is connected to the second fixing flange and whose upper end is engaged with an arm tab block provided on the second lifting flange;
a second lifting control knob installed on the upper end of the second screw rod and allowing the second lifting flange to be raised and lowered as the second screw rod is rotated through a rotation operation;
a second guide rod whose lower end is supported by the second fixing flange and whose upper end penetrates the second lifting flange; and
a second guide block installed on the second lifting flange and guiding the raising and lowering of the second guide rod;
A large-area plasma generator comprising a.
In claim 7,
A large-area plasma generator, further comprising a bellows for sealing the area around the second lifting rod in the section between the second fixing flange and the second lifting flange.
In claim 1,
The substrate holder and antenna are raised and lowered by a third lifting member,
The third lifting member,
A third fixed flange coupled to the bottom of the reactor;
a third lifting flange disposed on the lower side of the third fixing flange to be capable of being raised and lowered;
a third lifting rod in the form of a pipe whose lower part is supported on the third lifting flange and whose upper end penetrates the third fixed flange and the bottom of the reactor and is connected to the substrate holder and antenna;
a third screw rod whose upper part is connected to the third fixing flange and whose lower end is engaged with an arm tab block provided on the third lifting flange;
a third lifting control knob installed on the upper end of the third screw rod and allowing the third lifting flange to be raised and lowered by rotating the third screw rod through a rotational operation;
a third guide rod whose upper end is supported by the third fixing flange and whose lower end penetrates the third lifting flange; and
a third guide block installed on the third lifting flange and guiding the raising and lowering of the third guide rod;
A large-area plasma generator comprising a.
In claim 1,
The first lifting member,
A first fixed flange coupled to the upper surface of the reactor;
a first lifting flange disposed above the first fixing flange to be capable of being raised and lowered;
a first lifting rod whose upper part is supported on the first lifting flange and whose lower end penetrates the first fixed flange and the upper surface of the reactor and is connected to the upper part of the lifting cavity;
a plurality of first screw rods, the lower part of which is connected to a support block of the first fixed flange, and the upper part of which is engaged with an arm tab block provided on the first lifting flange;
a first drive chain gear rotatably installed at a lower portion of one of the support blocks;
a plurality of first driven chain gears rotatably installed at lower portions of the remaining support blocks among the support blocks on which the first driving chain gear is not installed;
a first chain connecting the first drive chain gear and the first driven chain gear to interlock with each other;
a first whim wheel installed on a support block on which the first drive chain gear is installed and integrally fixed with the first drive chain gear and the support block;
A first whim gear engaged with the first worm wheel; and
a first rotary handle for rotating the first worm gear;
A large-area plasma generator comprising a.
In claim 10,
The second lifting member,
a second fixed flange installed on the first lifting flange and raised and lowered in conjunction with the raising and lowering of the first lifting flange;
a second lifting flange disposed on an upper side of the second fixing flange to be capable of being raised and lowered in a direction close to or distant from the second fixing flange;
a second lifting bar whose upper part is supported on the second lifting flange and whose lower end is connected to the upper part of the lifting plunger through a hollow part of the first lifting bar;
a plurality of second screw rods, the lower part of which is connected to a support block of the second fixing flange, and the upper part of which is engaged with an arm tab block provided on the second lifting flange;
a second drive chain gear rotatably installed below one of the support blocks;
a plurality of second driven chain gears rotatably installed at lower portions of the remaining support blocks among the support blocks on which the first driving chain gear is not installed;
a second chain connecting the second drive chain gear and the second driven chain gear to interlock with each other;
a second whim wheel installed on the support block on which the second drive chain gear is installed and integrally fixed with the second drive chain gear and the support block;
a second whim gear engaged with the second worm wheel; and
a second rotary handle for rotating the second worm gear;
A large-area plasma generator comprising a.
In claim 1,
The matching controller,
A measuring unit that measures reflected waves or impedance;
a storage unit that stores real-time measurement values fed back from the measurement unit; and
A lifting control unit that compares the real-time measured value stored in the storage unit with the set value and, when the measured value exceeds the set value, adjusts the lifting position of the lifting plunger until the measured value falls below the set value;
A large-area plasma generator comprising a.
With the frequency of the ultra-high frequency oscillating member fixed at 2,450 MHz,
Measure the reflected wave in the measurement unit of the matching controller,
Store the real-time measurement values measured by the measurement unit in the storage unit,
A matching method that compares the real-time measured value stored in the storage unit with a preset setting value, and if the measured value exceeds the set value, the position of the lifting plunger is readjusted by the lifting control unit to change it to the optimal point where reflected waves are minimized.
In a state where the frequency of the ultra-high frequency oscillation member is varied within a predetermined range,
The frequency bandwidth area is divided into frequency area 1 and scanned to measure the reflected wave. The frequency bandwidth area is scanned to 10MHz, which is the frequency area 1, and the reflected wave is measured for each frequency area to form group 1 in which the reflected wave is minimized. ,
When the region width 1 in which the reflected wave of group 1 is minimized is detected, the frequency domain width 2 is reduced to 1/n of the frequency region 1 and performs a secondary search to form group 2 in which the reflected wave is minimized,
After the secondary search, the frequency domain width 2 is reduced to 1/m and the frequency domain 3 is searched, thereby forming group 3 that minimizes the reflected wave and detecting the optimal frequency. A matching method by frequency grouping search.

Images