JP2022136227A - Plasma cvd apparatus - Google Patents

Abstract

To provide a plasma CVD apparatus capable of solving the problem that since although the plasma CVD apparatus is used for manufacturing a liquid crystal display, an organic EL display and the like, a standing wave is generated between a pair of electrodes, a film thickness is uneven in applications to a large area substrate of, for example, 2 mx2 m, and to a substrate of 1 mx1 m in a VHF plasma CVD.SOLUTION: In a plasma CVD apparatus, supplying first and second electric powers having a retardation linearly phase-shifting to a plurality of first feed points provided in one end part of a non-ground electrode and a plurality of second feed points provided in the other end part to move plasma to be generated in one direction can equalize the distribution of plasma and as the result, uniform the distribution of film thickness.SELECTED DRAWING: Figure 1

Description

The present invention relates to a large-area plasma CVD apparatus for forming various thin films on the surface of a substrate using plasma.

A plasma CVD apparatus is utilized for forming photoelectric conversion films, insulating films, passivation films, barrier films, etc. in heterojunction (SHJ) solar cells, TFTs for liquid crystal displays, organic ELs, and various semiconductor devices. A plasma CVD apparatus that is generally put into practical use comprises an electrically grounded ground electrode capable of holding a substrate, and an electrically ungrounded non-grounded electrode arranged to face the grounded electrode. It has parallel plate electrodes with
Conventional high-frequency plasma CVD equipment using parallel plate electrodes has a non-uniformity due to the influence of standing waves formed by the interference phenomenon of high-frequency power waves supplied with an electric field between a pair of electrodes, resulting in a thin film. thickness distribution becomes non-uniform. As a result, it is difficult to manufacture ultra-large-screen products, such as ultra-large-area liquid crystal displays and organic EL displays, which are intended for substrates having a length similar to the wavelength of the power wave. Moreover, in manufacturing small-area products, it is difficult to reduce the manufacturing cost because the number of substrates that can be manufactured at one time by the plasma CVD apparatus cannot be increased.
Mechanisms and characteristics of standing waves generated in high-frequency plasma generation using parallel plate electrodes are generally known, as described in Non-Patent Documents 1 to 3, for example.
Regarding the problems faced by practical plasma CVD apparatuses using conventional parallel plate electrodes, for example, Patent Document 1 discloses that the high-frequency voltage application method used in the conventional plasma CVD apparatus uniformly distributes the film forming gas into a plasma. It is described that it is difficult in principle to make it uniform, and that this is a problem of standing waves peculiar to high frequencies. Also, for example, US ^Pat . It is stated that sexual problems will increase.

Several methods (ideas) have been proposed so far to solve the problem of non-uniformity in film thickness distribution caused by standing waves in a plasma CVD apparatus. Broadly categorized as
It can be divided into the following three categories. In addition, the numerical formula shown below expresses propagation of a high frequency electric power wave one-dimensionally.
(1) Phase modulation method: For example, the method disclosed in Patent Document 3. In this method, continuous wave high-frequency powers W ₁ (x, t) and W ₂ (x, t) are supplied from, for example, two opposing feeding points provided on the electrode end faces facing each other, and the high-frequency This is a method of temporally modulating the phase difference Δθ between the powers W ₁ (x, t) and W ₂ (x, t). In principle, Δθ in the following formula (1) is changed with time, for example, by Δθ=sin(Ωt). where Ω is the angular frequency.
W ₁ (x, t)=A sin {ωt+2πx/λ+θ ₁ }
W ₂ (x, t)=A sin {ωt−2πx/λ+θ ₂ }
where A is the amplitude, ω is the angular frequency, x is the propagation distance from the electrode end face, λ is the wavelength of the power wave, t is the time, and θ ₁ and θ ₂ are the phases.
The intensity of electric power between a pair of electrodes has a distribution of I(x) represented by the following formula (1). II(x)∝cos ² {2π(x−L ₀ /2)/λ−Δθ/2} (1)
where L ₀ is the length of the pair of electrodes, Δθ=θ ₁ −θ ₂ .
In this method, the plasma fluctuates alternately in the positive and negative directions of x with time, and as a result, is averaged with time, so that the distribution of plasma intensity is made uniform. It should be noted that the plasma intensity distribution expressed by Equation (1) does not change with time unless Δθ is changed.
(2) Standing wave superposition method: For example, the method disclosed in Patent Document 4 and Patent Document 5.
In this method, pulse power W ₁₁ (x, t) and pulse W ₁₂ (x, t) are supplied in a first time zone from at least two feeding points provided on electrode end faces facing each other, and A standing wave of 1 is generated, and pulse power W ₂₁ (x, t) and pulse W ₂₂ (x, t) are supplied in a second time period different from the first time period to generate a second standing wave. It is characterized by generating waves and superimposing the two standing waves. That is, it is a method of alternately supplying and superimposing the two pulse wave powers.
This method uses the powers W ₁₁ (x, t) and W ₁₂ (x, t) to
I ₁ (x, t)∝cos ² {2π(x−L0/2)/λ−Δθ/2}
and by the pulse wave powers W ₂₁ (x, t) and W ₂₂ (x, t),
I ₂ (x, t)∝sin ² {2π(x−L0/2)/λ−Δθ/2}
to generate By superimposing two standing waves with different generation times, a plasma intensity distribution with a constant value is formed.
I(x,t)∝cos ² {2π(x−L0/2)/λ−Δθ/2}
+sin ² {2π(x−L0/2)/λ−Δθ/2}=1 (2)
This method forms a plasma intensity distribution with a uniform spatial distribution by superimposing two standing waves.
(3) Phase adjustment method: For example, the method disclosed in Patent Document 1. In this method, feeding points are provided at a plurality of positions equidistant from the central point of the rectangular electrode, and the phases of the power waves supplied to the plurality of feeding points are adjusted. As a result, high-frequency voltages having the same frequency and phases shifted from each other are input to the respective feeding points.

Patent Document 1 describes the following. That is, it has a vacuum chamber, a substrate holder arranged inside the vacuum chamber, a hollow shower head arranged at a position facing the substrate holder, and a power supply device for applying voltage to the shower head, The shower head includes a container-shaped head body and a single shower plate that covers an opening of the head body and forms the hollow, the shower plate facing the substrate holder and facing the shower plate. is provided with a plurality of outlets, and a film-forming gas is supplied to the inside of the shower head, and a voltage is applied to the shower head while discharging the film-forming gas from the outlets, so that the film-forming gas is discharged into a plasma. a plasma CVD apparatus configured to spray the film-forming gas that has been transformed into plasma onto the substrate holder,
A plurality of the power supply devices are provided, each of the power supply devices is connected to a phase control device, the phase control device is configured to control the phase of the high-frequency voltage output from each of the power supply devices, The power supply devices are respectively connected to power supply points arranged at different positions equidistant from the center on the bottom surface of the head body, and the high-frequency voltages supplied from the power supply units to the power supply points are respectively supplied to the shower plate. A plasma CVD apparatus configured to generate electric discharge using the shower plate as a cathode.
Patent Document 3 describes the following. That is, a substrate to be processed held by a single holding electrode and a single discharge electrode are arranged facing each other in a discharge container with a space therebetween, and a substantially uniform discharge is generated between the discharge electrode and the substrate to be processed. A method of supplying power to a discharge electrode that produces a wide range of states, comprising:
When power is supplied to the discharge electrode via a plurality of feeding points, the phase of the voltage waveform of the high-frequency power supplied to one feeding point and the voltage of the high-frequency power supplied to at least one other feeding point By changing the phase difference of the waveform with time, the voltage distribution generated in the discharge electrode is changed, so that the average value per unit time or the integral value per unit time of the voltage distribution is substantially changed. A method of supplying electric power to the discharge electrodes, wherein the voltage distribution of the discharge electrodes is uniform to suppress the occurrence of standing waves in the voltage distribution of the discharge electrodes.

Patent Document 4 describes the following. That is, high-frequency plasma generation used in a surface processing apparatus for processing the surface of a substrate placed in a vacuum vessel by using plasma generated using power output from a high-frequency power source whose oscillation frequency belongs to the VHF band or UHF band. The apparatus generates a first standing wave and a second standing wave having different antinode positions of standing waves of electromagnetic waves having an electric field in substantially the same direction as the normal direction of the surface of the substrate, and and means for superimposing said first and second standing waves.
Patent Document 5 describes the following. That is, at least a pair of electrodes consisting of a non-grounded electrode and a grounded electrode installed facing each other, and a power supply for supplying high-frequency power to the non-grounded electrode from a plurality of feeding points provided on the non-grounded electrode. means for generating plasma in a plasma generation region sandwiched between the non-grounded electrode and the grounded electrode, wherein the plurality of feeding points are arranged along a line passing through the centers of the upper and lower surfaces of the non-grounded electrode. is provided at the end of the surface of the non-grounded electrode that does not face the grounded electrode or on the side surface of the non-grounded electrode so as to be symmetrical with respect to each other,
The power supply means has at least two states: an in-phase state in which the phases of the plurality of feeding points of the non-grounded electrode are aligned, and an out-of-phase state in which the phases of the feeding points obtained by dividing the plurality of feeding points into two are reversed. A high-frequency plasma generator characterized in that high-frequency power is supplied by switching.

Patent 4404303 Patent 5506379 Patent 3316490 JP 2005-123654 Patent 5489803

M. A. Lieberman, J. P. Booth, P. Chabert, J. M. Rax and M.M. Turner, Standing wave and skin effects in large-area, high-frequency capacitive discharge ,PLASMA SOURCES SCIENCE AND TECHNOLOGY, 11 (2002) ,283-293 A. Perret, P. Chabert, J.-P. Booth, J. Jolly, J. Guillon, and Ph. Auvray, Ion flux nonuniformities in large-area high-frequency capacitive discharges, APPLIED PHYSICS LETTERS, VOLUME 83, NUMBER 2 , 14 July 2003, 243-246 H. Kaneko, A. Kodera, Y. Soejima, S. Tsuji, M. Murata, Production of VHF excited H2 Plasma by New Method of Superposing the Standing Waves, Plasma Processes and Polymers, 2009,6, s269-s272

However, the conventional device has the following problems. The device described in Patent Document 1 provides feeding points at a plurality of positions equidistant from the central point of the rectangular electrode, and adjusts the phase of the electric power wave supplied to the plurality of feeding points. A standing wave is generated by interference of a plurality of electric power waves propagating in the direction of the central point of the rectangular electrode, resulting in an electric field distribution having a concentric intensity distribution. Moreover, since the phase of the voltage of the power supplied to each feeding point is temporally fixed, it is difficult to obtain a uniform electric field distribution.
The method and apparatus described in Patent Document 3 are expected to have the effect of temporally averaging the plasma intensity distribution by varying the phase difference of the voltages of power supplied to a plurality of feeding points over time. However, there is a problem that the effect is not large. It is known that the method described in Patent Document 3 yields a film thickness distribution of about ±17% in a substrate area of 1.4 m×1.1 m.
Moreover, Patent Document 3 describes the following method. That is, in advance, for example, the film thickness distribution at phase differences of 0 °, 45 °, 90 °, 135 ° is measured, and based on the data, the ratio of the film formation time at each phase difference required to achieve a uniform distribution is calculated by a computer, and the information of the calculation result is used to superimpose a plurality of plasma distributions at the selected plurality of phase values. This method is based on the idea that various patterns of standing waves generated between a pair of electrodes are measured in advance and the patterns are superimposed to perform a spatial averaging operation. There is a problem that the effect is unknown.
The device described in Patent Document 4 generates two standing waves alternately in terms of time, and is expected to have the effect of uniforming the plasma distribution of cos ² {2πx/λ}+sin ² {2πx/λ}=1. It is. Since this apparatus uses pulse-modulated power, a plasma with a high electron temperature is generated each time the pulse power rises, which poses the problem of increased ion damage. The pulse modulated plasma method has a problem that its application is limited.
The device described in Patent Document 5 alternately generates two standing waves in time similar to the device described in Patent Document 4, and cos ² {2πx/λ}+sin ² {2πx/λ}=1. This is expected to have the effect of uniforming the plasma distribution. Since this apparatus uses pulse-modulated power, a plasma with a high electron temperature is generated each time the pulse power rises, which poses the problem of increased ion damage. The pulse modulated plasma method has a problem that its application is limited.
In the above-mentioned conventional plasma CVD apparatus, there is a problem that it is difficult to form a uniform film when a substrate having a large area of, for example, approximately 1,300 mm x approximately 1,500 mm (or approximately 2.0 m ² ) or larger is used. .
In view of the problems described above, it is an object of the present invention to provide a thin film forming apparatus using plasma that can solve the problems described above.

A first aspect of the present invention for solving the above-mentioned problems is a reaction vessel equipped with an exhaust system, and a non-grounded electrode and a substrate disposed inside the reaction vessel and facing each other. a source gas introducing means for introducing a source gas between the pair of electrodes; high-frequency power supply means for supplying first power and second power to the feeding point and a plurality of second feeding points provided at the other end, respectively;
In a plasma CVD apparatus for forming a thin film on the substrate by converting the raw material gas into plasma in a plasma generation region sandwiched between the pair of electrodes,
The high-frequency power supply means includes: a first system line that supplies the first power to the plurality of first feeding points; a second system line that supplies the second power to the plurality of second feeding points; A phase shifter that controls the phase difference between the voltages of the first power and the second power,
The phase shifter is arranged such that the phase of the voltage of the output section of the first system line continues to lag behind the phase of the voltage of the output section of the second system line at a constant speed or advances at a constant speed. It is characterized by controlling the former phase and the latter phase so as to continue.
In a second invention based on the first invention, the phase shifter shifts the phase difference between the voltage of the output section of the first system line and the voltage of the output section of the second system line from 0° to 360°. is repeatedly phase-shifted at a constant speed with .
In a third invention based on the first invention, the phase shifter shifts the phase difference between the voltage of the output section of the first system line and the voltage of the output section of the second system line from 0° to 180°, It is characterized by repeatedly phase-shifting in the form of a ramp wave, or repeatedly phase-shifting in the form of a ramp wave from 0° to minus 180°.
In a fourth aspect of the invention, in any one of the second aspect and the third aspect, the phase shifter has a voltage at the output portion of the first system line and a voltage at the output portion of the second system line. is repeatedly phase-shifted in one period of approximately 1 μs to 1 ms.
In a fifth invention based on any one of the first invention to the fourth invention, the distance between the plurality of first feeding points adjacent to each other is one-eighth or less of the wavelength of the voltage of the first power. and the distance between the plurality of second feeding points adjacent to each other is one-eighth or less of the wavelength of the voltage of the second power.
A sixth invention is the invention according to any one of the first invention to the fifth invention, wherein the first feeding point and the second feeding point are ends of a surface of the non-grounded electrode that does not face the grounded electrode. or the side surface of the non-grounded electrode or the end of the surface facing the grounded electrode.
In a seventh invention based on any one of the first invention to the sixth invention, the high-frequency power supply means comprises a two-output phase variable oscillator, a divider, a power amplifier, a matching box, characterized by comprising
An eighth invention is the invention according to any one of the first invention to the seventh invention, wherein the frequency of the power generated by the high-frequency power supply means is a frequency selected from 13.56 MHz or a VHF band of 30 MHz to 300 MHz. It is characterized by
A ninth invention is characterized in that, in any one of the first to eighth inventions, the raw material gas contains at least silane gas or organic silane gas.

ADVANTAGE OF THE INVENTION According to this invention, the problem which the conventional plasma CVD apparatus has can be solved. A plurality of first feeding points and a plurality of second feeding points provided at opposite ends of the non-grounded electrode supply the first power and the second power generated by the high-frequency power supply means, which is a component of the plasma CVD apparatus of the present invention. and is propagated from the plurality of first feeding points and the plurality of second feeding points to between a pair of electrodes consisting of a non-grounded electrode and a grounded electrode on which the substrate is placed, and the first power and The phase difference Δθ of the voltage of the second power is shifted in one direction by repeatedly phase-shifting, for example, 0° to 360° with a period of 1 ms as a ramp wave with a phase shifter. Plasma can be generated. By moving the plasma in one direction, there is an effect that the film formed on the substrate can be made uniform. As a result, it is possible to form a film with a uniform thickness by plasma CVD on a substrate having a very large area of, for example, 2 m×2 m, which is difficult with the conventional apparatus.
The plasma CVD apparatus according to the present invention is capable of forming large-area uniform thin films in fields such as large-screen liquid crystal displays, large-screen organic EL displays, large-area solar cells, and various semiconductor devices. The degree of contribution to improvement is remarkably large.

FIG. 1 is a schematic cross-sectional view showing the configuration of a plasma CVD apparatus according to the first embodiment of the present invention. FIG. 2 is a schematic perspective view showing the configuration of the plasma CVD apparatus according to the first embodiment of the present invention. FIG. 3 shows a state in which power supplied between a pair of electrodes from a plurality of first feeding points and a plurality of second feeding points of the plasma CVD apparatus according to the first embodiment of the present invention propagates as power waves. It is a conceptual diagram. FIG. 4 is a conceptual diagram showing power waves propagating facing each other between a pair of electrodes of the plasma CVD apparatus according to the first embodiment of the present invention. FIG. 5 is a conceptual diagram showing the phase relationship between two output voltage signals of a two-output phase-variable oscillator, which is a component of the plasma CVD apparatus according to the first embodiment of the present invention. FIG. 6 is a conceptual diagram showing one-dimensional distribution of power supplied between a pair of electrodes of the plasma CVD apparatus according to the first embodiment of the present invention. FIG. 7 is a conceptual diagram showing the relationship between the power intensity distribution and the phase difference of supplied power in the plasma generation region of the plasma CVD apparatus according to the first embodiment of the present invention. FIG. 8 is a conceptual diagram showing uniformity of plasma intensity distribution in the plasma generation region of the plasma CVD apparatus according to the first embodiment of the present invention. FIG. 9 is a schematic perspective view showing the configuration of the power supply means of the plasma CVD apparatus according to the second embodiment of the invention. FIG. 10 is a conceptual diagram showing the intensity distribution of plasma generated between a pair of electrodes of the plasma CVD apparatus according to the second embodiment of the present invention.

EMBODIMENT OF THE INVENTION Hereinafter, the form for implementing this invention is demonstrated, referring drawings. In each figure, the same reference numerals are given to the same members, and redundant explanations are omitted.
It should be noted that the present invention is not limited to the following description, and can be modified as appropriate without departing from the gist of the present invention. Also, in the drawings shown below, for convenience of explanation, the scale of each member may differ from the actual scale. Also, the scale of each drawing may differ from the actual scale.

(First embodiment)
A plasma CVD apparatus according to a first embodiment of the present invention will be described with reference to FIGS. 1 to 8. FIG. Although a rectangular non-grounded electrode and a rectangular grounded electrode are used as a pair of electrodes in this embodiment, the present invention is not limited to this.
FIG. 1 is a schematic cross-sectional view showing the configuration of a plasma CVD apparatus according to the first embodiment of the present invention. FIG. 2 is a schematic perspective view showing the configuration of the plasma CVD apparatus according to the first embodiment of the present invention. FIG. 3 shows a state in which power supplied between a pair of electrodes from a plurality of first feeding points and a plurality of second feeding points of the plasma CVD apparatus according to the first embodiment of the present invention propagates as power waves. It is a conceptual diagram. FIG. 4 is a conceptual diagram showing power waves propagating facing each other between a pair of electrodes of the plasma CVD apparatus according to the first embodiment of the present invention. FIG. 5 is a conceptual diagram showing the phase relationship between two output voltage signals of a two-output phase-variable oscillator, which is a component of the plasma CVD apparatus according to the first embodiment of the present invention. FIG. 6 is a conceptual diagram showing one-dimensional distribution of power supplied between a pair of electrodes of the plasma CVD apparatus according to the first embodiment of the present invention. FIG. 7 is a conceptual diagram showing the relationship between the power intensity distribution and the phase difference of supplied power in the plasma generation region of the plasma CVD apparatus according to the first embodiment of the present invention. FIG. 8 is a conceptual diagram showing uniformity of plasma intensity distribution in the plasma generation region of the plasma CVD apparatus according to the first embodiment of the present invention.

First, the configuration of the plasma CVD apparatus according to the first embodiment of the present invention will be described with reference to FIGS. 1 to 8. FIG.
The reaction container 50 is a vacuum container capable of maintaining the inside in a high vacuum state and does not generate impurities, and has exhaust ports 51a and 51b connected to a vacuum pump (not shown). By operating the exhaust ports 51a and 51b in combination with vacuum pumps (not shown), the inside of the reaction vessel 50 can be evacuated to a predetermined degree of vacuum.
In the reaction vessel 50, a non-grounded electrode 1 described later and a grounded electrode 60 described later are arranged as a pair of electrodes for plasmatizing a raw material gas described later.
The ground electrode 60 is arranged at a predetermined location of the reaction container 50, for example, on the bottom surface. The ground electrode 60 has a main surface 60a, and contacts and holds the substrate 61 at the main surface 60a. The temperature of the substrate 61 is controlled to a predetermined temperature by a substrate heater (not shown) provided inside the ground electrode 60 . The shape of the ground electrode 60 is, for example, a rectangle having short sides and long sides, similar to the non-ground electrode 1 described later. The dimensions are one size larger than the later-described non-grounded electrode 1, for example, 1.2 m x 2.2 m.
By opening and closing the substrate loading/unloading valve 62, the substrate 61 is loaded and placed on the main surface 60a of the substrate holding table 60 from the atmospheric side, formed with a desired film, and then unloaded to the atmospheric side.

The non-grounded electrodes 1 are arranged at predetermined intervals so as to face the grounded electrode 60 . The non-grounded electrode 1 is, for example, a rectangle having a short side 1a and a long side 1b, and its dimensions are, for example, 1.1 m×2.1 m. The non-grounded electrode 1 includes a raw material gas introduction pipe 3 for introducing raw material gas, a raw material gas dispersion cavity 6, and a raw material gas ejection hole 5 for ejecting the raw material gas. The non-grounded electrode 1 is fixed to the electrode support shelf 53 of the reaction container 50 via the insulator 55 . The raw material gas introduced from the raw material gas introduction pipe 3 is dispersed in the raw material gas dispersion cavity 6 and is evenly ejected from the raw material gas ejection holes 5 between the non-grounded electrode 1 and the grounded electrode 60 .
The source gas ejection hole 5 is formed in a circular shape with a diameter of approximately 0.4 mm to approximately 1 mm. Here, it is 0.8 mm. If the diameter of the raw material gas ejection hole is set to approximately 1 mm or more, the spatial distribution of the ejection amount of the gas becomes uneven. and costs. The hole diameter of 0.8 mm can be said to be a reasonable value because the spatial distribution of the gas ejection amount is uniform and the manufacturing cost can be suppressed.
The non-grounded electrode 1 is electrically insulated from the reaction container 50 and the grounded electrode 60 by the insulator 55 and is in a non-grounded state.
Here, the surface of the non-grounded electrode 1 facing the grounded electrode 60 is called the surface 1 d of the non-grounded electrode 1 , and the back surface is called the back surface 1 c of the non-grounded electrode 1 .
Power supplied from one output terminal of a two-output phase-variable high-frequency oscillator 70 to be described later to one short side 1a of the back surface 1c of the non-grounded electrode 1 via a first system line 69a to be described later. First feed points 77a and 77b are provided. Power supplied from the other output terminal of the two-output phase-variable high-frequency oscillator 70 described later to the other short side 1aa facing the plurality of first feeding points 77a and 77b via the second system line 69b described later. A plurality of second feed points 77c and 77d are provided to supply the .
The distance between the plurality of first feeding points 77a and 77b and the distance between the plurality of second feeding points 77c and 77d are set to one-eighth or less of the wavelength λ of the high-frequency power supplied from the first system line 69a, which will be described later.
Here, as will be described later, a frequency of 60 MHz is used, so λ/8=40 cm. Further, the interval between the plurality of second feed points 77c and 77d is set to one-eighth or less of the wavelength λ of the high-frequency power supplied from the later-described second system line 69b. Here, as will be described later, a frequency of 60 MHz is used, so λ/8=40 cm. The wavelength in vacuum at a frequency of 60 MHz is 5 m (3.25 m in plasma with a wavelength shortening rate of 0.65).
The reason is as follows. That is, if the interval is set to one-eighth or less of the wavelength of the power, the potential on the line connecting adjacent feeding points becomes substantially the same along the line.

Reference numeral 70 is a two-output phase-variable high-frequency oscillator. A two-output phase-variable high-frequency oscillator 70 outputs a high-frequency sinusoidal voltage in the range of 10 MHz to 100 MHz. The frequency of the two-output phase-variable high-frequency oscillator 70 can be arbitrarily selected from the range of 10 MHz to 100 MHz. Here, for example, it is 60 MHz. The wavelength in vacuum at a frequency of 60 MHz is 5 m (3.25 m in plasma with a wavelength shortening rate of 0.65).
The two-output phase-variable high-frequency oscillator 70 is not a special device, but a commercially available device.
The two-output phase-variable high-frequency oscillator 70 has two output terminals 68a and 68b, and linearly controls the phase difference between the two high-frequency sinusoidal voltages output from the two output terminals 68a and 68b. It is possible. 5(a) and 5(b), the phase difference between the voltages of the two high-frequency sine waves output from the two output terminals 68a and 68b of the two-output phase-variable high-frequency oscillator 70 is It can be controlled in the form of a ramp wave such as
That is, the two-output phase-variable high-frequency oscillator 70 controls, for example, the phase difference between the voltages of the two output signals in the form of a ramp wave whose period T0 is ₀ ° to 360°, or 0° to minus It is possible to control in the form of a ramp wave with 360° as one cycle and one cycle _T0 . Then, for example, the phase difference between the voltages of the two output signals is controlled in the form of a ramp wave with one cycle T0 of ₀ ° to 180°, or in the form of a ramp wave with one cycle of 0° to minus 180°. morphology can be controlled. Here, the one period _T0 is selected from the range of 1 μs to 1 msec, for example, and is set to 1 msec.
A voltage signal output from the output terminal 68a of the two-output phase-variable high-frequency oscillator 70 can be expressed by the following equation.
V ₁ (t)=sin {ωt+θ ₁₁ } (3)
where ω is the angular frequency and θ ₁₁ is the phase. A voltage signal output from the output terminal 68b of the two-output phase-variable high-frequency oscillator 70 can be expressed by the following equation.
V ₂ (t)=sin {ωt+θ ₂₂ } (4)
where ω is the angular frequency and θ ₂₂ is the phase.
The phase difference between the two voltage signals is linearly phase-shifted by a phase shifter (not shown) included in the two-output phase-variable high-frequency oscillator 70 . That is, the two-output phase-variable high-frequency oscillator 70 can shift the phase difference between the two voltage signals at a constant speed as shown by the equation (5).
δθ= _θ11 − _θ22
=αt (5)
where α is a constant of proportionality and t is time.
Then, the two-output phase-variable high-frequency oscillator 70 repeats the phase difference between the two voltage signals with a phase shifter (not shown) as 0° to 360° as one cycle T0 or ₀ ° to 180° as one cycle. can be phase-shifted. That is, as shown in equation (6), the phase can be shifted from 0° to 360° or from 0° to 180° at a constant speed in the form of a ramp wave with a constant period T ₀ .
δθ= _θ11 − _θ22
= t-floor(t) (6)
However, as shown in FIG. 5, t-floor(t) represents a phase shift in the form of a ramp wave that shifts from 0° to 360° at a constant speed and repeats 360° as one period _T0 . ing. Alternatively, it represents a phase shift in the form of a ramp wave that shifts from 0° to 180° at a constant speed and repeats 180° as one period _T0 . It should be noted that the phase may be shifted repeatedly from 0° to minus 360°, or may be shifted from 0° to minus 180°.
Here, an important technical point in the thin film forming apparatus using plasma according to the first embodiment of the present invention is that the phase difference between the two voltage signals is expressed by Equation (5) or Equation (6). or to control the phase difference between the two voltage signals as shown in FIG. As a result, there is an effect that the standing wave generated between the non-grounded electrode 1 and the grounded electrode 60 can be moved in one direction.

Reference numeral 72a is a first power amplifier. The first power amplifier 72a amplifies the power of the high frequency signal transmitted from the output terminal 68a of the two-output phase variable high frequency oscillator 70 via the coaxial cable. The maximum output of the first power amplifier 72a is, for example, 2 kW to 20 kW. Here for example. 5 kW.
A first power amplifier 72a transmits output power to a first matching box 73a, which will be described later, via a coaxial cable. Here, the power output from the first power amplifier 72a is called first power.
Reference numeral 73a is a first matching unit. The first matching box 73a adjusts the first electric power so that the reflected waves at the first feeding points 77a and 77b are small, and transmits the first electric power. The first matching device 73a transmits the first power through a coaxial cable to a first power distributor 79a, which will be described later.
Reference numeral 79a is a first power distributor. The first power distributor 79a divides the first power transmitted from the first matching device 73a into two, and distributes one of the first powers through a coaxial cable to a first vacuum current introduction terminal, which will be described later. While transmitting power to 75a, the other first power is transmitted to a second current introduction terminal 75b for vacuum through a coaxial cable. Reference numeral 75a is a first current introduction terminal for vacuum. The first vacuum current introduction terminal 75a feeds the first power transmitted from the first power distributor 79a to the first feeding point 77a through the vacuum conducting wire 76a. The first feeding point is composed of 77a and 77b, which will be described later.
Reference numeral 75b is a second vacuum current introduction terminal. The second vacuum current introduction terminal 75b feeds the first power transmitted from the first power distributor 79a to the first feeding point 77b via the second vacuum conductor 76b.
Here, from the output terminal 68a of the two-output phase-variable high-frequency oscillator 70, the first power amplifier 72a, the first matching device 73a, the first power divider 79a, the first vacuum current introduction terminal 75a, the second A first electric power transmission line including two vacuum current introduction terminals 75b, a first vacuum conducting wire 76a and a second vacuum conducting wire 76b is referred to as a first system line 69a.

Reference numeral 72b is a second power amplifier. The second power amplifier 72b amplifies the power of the high frequency signal transmitted from the output terminal 68b of the two-output phase variable high frequency oscillator 70 via the coaxial cable. The maximum output of the second power amplifier 72b is, for example, 2 kW to 20 kW. Here for example. 5 kW.
The second power amplifier 72b transmits the output power through a coaxial cable to a second matching box 73b, which will be described later. Here, the power output from the second power amplifier 72b is called second power.
Reference numeral 73b is a second matching device. The second matching device 73b adjusts the second power so that the reflected waves at the second feeding points 7c and 7d are reduced, and transmits the second power. The second matching device 73b transmits the second power through a coaxial cable to a second power distributor 79b, which will be described later.
Reference numeral 79b is a second power distributor. The second power distributor 79b splits the second power transmitted from the second matching device 73b into two, and one of the second powers is sent to a third vacuum current introduction terminal described later via a coaxial cable. While transmitting power to 75c, the other second power is transmitted to a fourth current introduction terminal 75d for vacuum via a coaxial cable. Reference numeral 75c is a third current introduction terminal for vacuum. The third vacuum current introduction terminal 75c feeds the second power transmitted from the second power distributor 79b to the second feeding point 77c through the vacuum conducting wire 76c. The second feeding point is composed of 77c and 77d which will be described later.
Reference numeral 75d is a fourth vacuum current introduction terminal. The fourth vacuum current introduction terminal 75d feeds the second power transmitted from the second power distributor 79b to the second feeding point 77d through the fourth vacuum conducting wire 76d.
Here, from the output terminal 68b of the two-output phase-variable high-frequency oscillator 70, the second power amplifier 72b, the second matching device 73b, the second power divider 79b, the third vacuum current introduction terminal 75c, the second A transmission line for the second electric power including the four vacuum current introduction terminals 75d, the third vacuum conducting wire 76c, and the fourth vacuum conducting wire 76d is called a second system line 69b.
Further, here, the two-output phase variable high frequency oscillator 70, the first system line 69a, and the second system line 69b are provided, and the first electric power is fed to the plurality of first feeding points 77a and 77b, and High-frequency power supply means for supplying the second power to the plurality of second power supply points 77c and 77d is called first high-frequency power supply means.

The distance between the non-grounded electrode 1 and the grounded electrode 60 is determined in consideration of plasma generation conditions. That is, the plasma generation conditions of the raw material gas generated between the non-grounded electrode 1 and the grounded electrode 60 mainly depend on the distance d between the non-grounded electrode 1 and the grounded electrode 60, the pressure p, and the applied high-frequency voltage V. . Here, according to Paschen's law, it is set in the region of the minimum value of the Paschen curve. For example, the pd product selects a value that satisfies approximately 133 Pa·cm to 1333 Pa·cm. Here, for example, the distance d between the non-grounded electrode 1 and the grounded electrode 60 is set to 0.5 cm, and the pressure p is set to 133 Pa to 1333 Pa.

The raw material gas is selected according to the type of film to be used. For example, it is a mixed gas of silane gas, hydrogen gas and rare gas. Here, for example, silane gas and hydrogen are selected, and the flow ratio is set to hydrogen flow rate/silane gas flow rate=100. A mixed gas of silane gas and hydrogen is supplied from a raw material gas source (not shown) and controlled at a predetermined flow rate by a raw material gas mass flow controller (not shown). 5 is ejected.

With the raw material gas introduced between the non-grounded electrode 1 and the grounded electrode 60, from the first high-frequency power supply means to the plurality of first feeding points 77a and 77b and the plurality of second feeding points 77c and 77d, respectively. Plasma is generated between the pair of electrodes 1 and 60 when the first power and the second power are supplied.
The first electric power supplied to the plurality of first feeding points 77a and 77b flows from the side 1a of the non-grounded electrode 1 to between the side surface of the electrode and the wall of the vacuum vessel 50, as indicated by _Vp1 in FIG. It propagates, enters between the non-grounded electrode 1 and the grounded electrode 60, and propagates between the pair of electrodes 1,60.
The second electric power supplied to the plurality of second feeding points 77c and 77d flows from the side 1aa of the non-grounded electrode 1 to between the side surface of the electrode and the wall of the vacuum vessel 50, as indicated by _Vp2 in FIG. It propagates, enters between the non-grounded electrode 1 and the grounded electrode 60, and propagates between the pair of electrodes 1,60.
FIG. 4 schematically shows how _Vp1 and _Vp2 propagate. V _p1 (x,t) and V _p2 (x,t) are propagating waves traveling opposite to each other.
Assuming that the origin of the one-dimensional coordinate axis x is one end of the surface 1d between the pair of electrodes 1 and the length of the long side of the non-grounded electrode 1 is _L0 , the propagating wave shown in FIG. is represented by
V _p1 (x, t)=V ₀₁ sin {ωt+2πx/λ+θ ₁₁ } (7)
where V ₀₁ is the amplitude, ω is the angular frequency, and θ ₁₁ is the phase.
V _p2 (x, t)=V ₀₂ sin {ωt−2πx/λ+θ ₂₂ } (8)
where V ₀₂ is the amplitude, ω is the angular frequency, and θ ₂₂ is the phase.
The two propagating waves V _p1 (x, t) and V _p2 (x, t) generate a power wave I(x, t) traveling in a fixed direction. This is represented by equation (9).
I(x)∝cos ² {2π(x−L ₀ /2)/λ−Δθ/2} (9)
However, _L0 is the length of the pair of electrodes, and Δθ is given by the following equation.
Replacing Δθ with a time-varying Δθ(t), equation (9) becomes a power wave traveling in the x-direction. It is represented by the following (10).
I(x, t)∝cos ² {2π(x−L ₀ /2)/λ−Δθ(t)/2} (10)
Here, Δθ(t) is changed in the form of Δθ(t) shown in Equation (5) or Equation (6) using a phase shifter (not shown) incorporated in the two-output phase-variable high-frequency oscillator 70. .
Δθ(t)=αt (5)
or
Δθ(t)=t−floor(t) (6)
However, as shown in FIG. 5(a), Δθ(t) represents a ramp wave-like phase shift at a constant speed with ₀ ° to 360° as one cycle T0. Alternatively, Δθ, as shown in FIG. 5B, represents that 0° to minus 180° is one period T ₀ and that the phase shifts in a ramp-like manner at a constant speed. Note that _T0 can be arbitrarily selected within the range of 1 μs to 1 ms. T ₀ can be arbitrarily selected in the range of 1 μs to 1 ms.
Δθ(t) may be repeatedly phase-shifted with 0° to minus 360° as one cycle _T0 . Further, the phase may be shifted repeatedly with 0° to 180° as one cycle _T0 . Note that _T0 can be arbitrarily selected within the range of 1 μs to 1 ms.
That is, using a phase shifter (not shown) incorporated in the two-output phase-variable high-frequency oscillator 70, the first
The phase difference between the voltage of the output portion of the system line 69a and the voltage of the output portion of the second system line 69b is shifted in the form of the above equation (5) or (6).

The power wave represented by Equation (10) above is affected by Δθ(t). If Δθ(t) is phase shifted in time, the power wave will move accordingly. A power wave generated between the pair of electrodes 1 and 60 generates plasma having a similar shape between the pair of electrodes 1 and 60 .
The power wave I(x, t) moving in a certain direction has a distribution as shown in FIG. 6 when time is stopped.
When Δθ is zero, a standing wave having an antinode at the center point of the non-grounded electrode 1 is obtained as shown in FIG. When .DELTA..theta. is positive, as shown in FIG. 7, it becomes a standing wave having an antinode at a position shifted in one direction from the center point of the non-grounded electrode. When .DELTA..theta. is negative, as shown in FIG. 7, it becomes a standing wave having an antinode at a position shifted in the other direction from the center point of the non-grounded electrode.
When Δθ is represented by the above formula (5) or the above formula (6), the power wave I(x, t moves continuously at a constant speed in the x direction or the negative direction of x. Fig. 8(a) shows that plasma is generated in a form proportional to the intensity of the power wave, so Fig. 8(a) and Fig. 6 have similar forms.
Looking at the plasma moving in a certain direction at an arbitrary point B on the main surface 60a of the ground electrode 60, strong plasma and weak plasma pass in a certain direction. averaged. This averaging has the same averaging effect as the uniformization of the film thickness in the film forming process in the plasma CVD apparatus by the substrate moving system shown in FIG. 8(b).
That is, the film forming process in the plasma CVD apparatus according to the first embodiment of the present invention is similar to the film forming process in the plasma CVD apparatus using the substrate transfer method shown in FIG. This has the effect of making the thickness uniform.

Next, operation of the plasma CVD apparatus according to the first embodiment of the present invention will be described with reference to FIGS. 1 to 8. FIG. For convenience of explanation, the formation of a microcrystalline silicon film will be described below as an example.
The raw material gas is selected according to the type of film to be used. Silane gas, ammonia, organic silane gas, hydrogen, oxygen, etc. can be selected according to the type of film to be used. Here, for example, a microcrystalline silicon film is formed. For gas supply conditions, for example, silane gas and hydrogen are selected, and the flow rate ratio is set to hydrogen flow rate/silane gas flow rate=100.
First, the substrate loading/unloading valve 62 of the reaction chamber 50 is opened, and the substrate 61 is placed on the main surface 60 a of the substrate holding table 60 . Next, after closing the substrate loading/unloading valve 62, the inside of the reaction chamber 50 is evacuated to a predetermined degree of vacuum through the exhaust port 51a and the exhaust port 51b by a vacuum pump (not shown).
After that, from a silane gas source (not shown) and a hydrogen gas source (not shown), silane gas and hydrogen gas whose flow rates are respectively controlled by mass flow controllers (not shown) for silane gas and hydrogen gas are introduced into the source gas introduction pipe 3 and the source gas dispersion cavity 6. It is ejected from the raw material gas ejection hole 5 via.
Here, it is assumed that the flow ratio of hydrogen gas and silane gas is 100 times, that is, flow rate of hydrogen gas/flow rate of silane gas=100.
Next, the opening/closing degree of the exhaust valve (not shown) is controlled by an exhaust valve control device (not shown) to maintain the internal pressure of the reaction container 50 at approximately 133 Pa to approximately 1333 Pa. It should be noted that the vacuum pump used for vacuuming the reaction vessel 50 and the vacuum pump used during film formation may be different. Here, for example, it is set to 500 Pa and maintained.

Next, the first power is applied to the plurality of first feeding points 77a, 77 using the first high-frequency power supply means.
b, and the second power is fed to a plurality of second feed points 77c, 77d. That is, using a phase shifter (not shown) incorporated in the two-output phase-variable high-frequency oscillator 70, the phase difference between the voltage of the output portion of the first system line 69a and the voltage of the output portion of the second system line 69b is calculated as described above. Phase shift in the form of equation (5) or equation (6).
Then, a power distribution of the form shown in FIG. 6 is generated. As a result, the raw material gas ejected from the raw material gas ejection holes 5 is turned into plasma, radicals such as _{SiH 3} and H, which are precursors of the microcrystalline silicon film, are generated, and microcrystalline silicon is deposited on the substrate 61 .
Since the plasma is generated in the form shown in FIG. 8A, a microcrystalline silicon film having a uniform thickness distribution is formed.

Next, since the thickness of the microcrystalline silicon film is proportional to the plasma generation time, the first high-frequency power supply means is turned on when a predetermined time has elapsed after the first high-frequency power supply means started supplying the output. to zero the output of
The film-forming time is determined based on previously acquired data. Here, it is 1 minute to 30 minutes, for example, 5 minutes. The film forming time can be determined based on previously obtained data relating to the relationship between the electrode spacing, substrate temperature, silane gas flow rate, hydrogen gas flow rate, pressure, electric power, and the like.
After the desired formation of the microcrystalline silicon film is completed, the supply of the silane gas and hydrogen gas is stopped, and the inside of the reaction vessel 50 is once evacuated to a high degree of vacuum. The reaction vessel 50 is then returned to atmospheric conditions. After the reaction vessel 50 is returned to atmospheric conditions, the substrate loading/unloading valve 62 is opened and the substrate 61 is taken out.
A uniform microcrystalline silicon film is formed on the substrate 61 taken out from the reaction vessel 50 .

In the thin film forming apparatus using plasma according to the first embodiment of the present invention, as described above, the phase of the voltage of the first power supplied to the plurality of first feeding points 77a and 77b and the phase of the voltage of the plurality of second feeding points The phase difference Δθ between the phases of the voltages of the second electric power supplied to points 77c and _77d is, as shown in FIG. , the phase shifts like a ramp wave, so the plasma generated between the pair of electrodes 1 and 60 continues to move in a fixed direction. As a result, the thickness of the film deposited on the substrate placed on the ground electrode 60 is uniform.
As a method for continuously moving the plasma generated between the pair of electrodes 1 and 60 in a certain direction, the phase of the voltage of the first power supplied to the plurality of first feeding points 77a and 77b and the phase of the voltage of the plurality of first feeding points 77a and 77b The phase difference Δθ(t) between the voltages of the 21 powers fed to the two feeding points 77c and 77d is set to 0° to minus 180° or 0° to 180° as shown in FIG. 5(b). The phase may be ramp-wave-like at a constant speed with one period _T0 . Also, one period T ₀ can be arbitrarily selected within the range of ₁ μs to 1 ms.
As a result, the plasma CVD apparatus according to the first embodiment of the present invention can uniformly form a film by plasma CVD on a very large substrate, which is difficult with the conventional apparatus.

(Second embodiment)
A plasma CVD apparatus according to a second embodiment of the present invention will be described with reference to FIGS. 9 and 10. FIG. See also FIGS.
FIG. 9 is a schematic perspective view showing the configuration of the power supply means of the plasma CVD apparatus according to the second embodiment of the invention. FIG. 10 is a conceptual diagram showing the intensity distribution of plasma generated between a pair of electrodes of the plasma CVD apparatus according to the second embodiment of the present invention.
First, the configuration of the plasma CVD apparatus according to the second embodiment of the present invention will be described. The difference between the configuration of the plasma CVD apparatus according to the second embodiment of the present invention and the configuration of the above-described plasma CVD apparatus according to the first embodiment of the present invention is as follows.
(a) The feed point is located at the end of the back surface 1c of the non-grounded electrode in the former, while it is located at the end of the surface 2d of the non-grounded electrode in the latter. As a result, in the latter case, the supplied power wave can effectively propagate between the non-grounded electrode and the grounded electrode compared to the former.
(b) In the former case, two feed points are provided on each of the opposing sides 1a and 1aa, whereas in the latter case, four feed points are provided on each of the opposing sides 2a and 2aa. . As a result, in the latter case, it is possible to form a uniform film on a substrate having a large area as compared with the former case.
(C) In the plasma CVD apparatus according to the first embodiment of the present invention, the phase difference between the voltages of the first power and the second power is controlled using a phase shifter provided in the two-output phase-variable high-frequency oscillator 70. However, in the plasma CVD apparatus according to the second embodiment of the present invention, in addition to the phase shifter included in the two-output phase-variable high-frequency oscillator 70, another phase shifter is additionally arranged. be. This makes it possible to match the phase difference of the power supplied to the non-grounded electrodes.

Reference numeral 2 is a rectangular plate-shaped non-grounded electrode. The rectangular plate-shaped non-grounded electrode 2 is used in combination with the grounded electrode 60 . The rectangular plate-shaped non-grounded electrode 2 has a surface 2 d facing the ground electrode 60 and a back surface 2 c not facing the ground electrode 60 . Further, the rectangular plate-shaped non-grounded electrode 2 has two sides 2a and 2aa facing each other among the four sides of the surface 2d. The non-grounded electrode 2 in the shape of a rectangular flat plate has, for example, two sides 2a and 2aa with lengths of, for example, 2.1 m×2.1 m.
A rectangular plate-shaped non-grounded electrode 2 is provided with a raw material gas introduction pipe 3 for introducing raw material gas, a raw material gas dispersion cavity 6, and a raw material gas ejection hole 5 for ejecting the raw material gas. The rectangular plate-shaped non-grounded electrode 2 is fixed to the electrode support shelf 53 of the reaction vessel 50 via an insulator 55 . The raw material gas introduced from the raw material gas introduction pipe 3 is dispersed in the raw material gas dispersion cavity 6 and is evenly ejected from the raw material gas ejection hole 5 between the non-grounded electrode 2 and the grounded electrode 60 in the shape of a rectangular flat plate.
Reference numerals 7a, 7b, 7c and 7d are third feeding points. The third feeding points 7a, 7b, 7c, and 7d have a plurality of feeding points and are arranged at regular intervals along the side 2a of the non-grounded electrode 2 in the shape of a rectangular flat plate.
Reference numerals 7e, 7f, 7g and 7h are fourth feeding points. The fourth feeding points 7e, 7f, 7g, and 7h have a plurality of feeding points and are arranged at regular intervals along the side 2aa of the non-grounded electrode 2 in the shape of a rectangular flat plate.
The distance between each of the plurality of third feeding points 7a and 7b, 7b and 7c, and 7c and 7d is one-eighth or less of the wavelength λ of high-frequency power supplied from a third system line 69c, which will be described later. Here, as will be described later, a frequency of 40 MHz is used, so λ/8=61 cm. The wavelength λ of the electromagnetic wave of 40 MHz is λ=7.5 m in vacuum (4.9 m in plasma with a wavelength shortening rate of 0.65). λ/8=0.61 m.
The respective intervals between the plurality of fourth feeding points 7e and 7f, 7f and 7g, and 7g and 7h are set to one-eighth or less of the wavelength λ of the high-frequency power supplied from the fourth system line 69d described later. Here, as will be described later, a frequency of 40 MHz is used, so λ/8=61 cm.
The reason is as follows. That is, if the interval is set to 1/8 or less, the potentials on the line connecting adjacent feeding points become substantially the same along the line.

The two-output phase-variable high-frequency oscillator 70 outputs a high-frequency sinusoidal voltage in the range of 10 MHz to 100 MHz, as described in the first embodiment of the present invention. Here, for example, it is 40 MHz. The wavelength λ of an electromagnetic wave of 40 MHz is λ=7.5 m in vacuum (4.9 m in plasma with a wavelength shortening rate of 0.65). λ/8=0.61 m.
The two-output phase-variable high-frequency oscillator 70 is not a special device, but a commercially available device.
The two-output phase-variable high-frequency oscillator 70 has two output terminals 68a and 68b, and two high-frequency signals output from the two output terminals 68a and 68b, as described in the first embodiment of the present invention. It is possible to linearly control the phase difference of the sinusoidal voltage. As described in the first embodiment of the present invention, the two-output phase-variable high-frequency oscillator 70 detects the phase difference between the two high-frequency sinusoidal voltages output from the two output terminals 68a and 68b. , can be controlled to form a ramp wave as shown in FIGS. 5(a) and (b). The two-output phase-variable high-frequency oscillator 70 transmits sinusoidal voltage signals from two output terminals through coaxial cables to a third distributor 10a and a fourth distributor 10b, which will be described later.

Reference numeral 10a is a third distributor. The third divider 10a divides the sinusoidal voltage signal transmitted from the output terminal 68a of the two-output phase-variable high-frequency oscillator 70 into four phase shifters 11a and 11b, which will be described later. , to the third phase shifter 11c and the fourth phase shifter 11d.
Reference numerals 11a, 11b, 11c and 11d are respectively a first phase shifter, a second phase shifter, a third phase shifter and a fourth phase shifter. The first phase shifter 11a, the second phase shifter 11b, the third phase shifter 11c, and the fourth phase shifter 11d can adjust the phase of the input sinusoidal voltage signal and output it. It is possible and used in combination with a first phase monitor 15a, which will be described later. As described later, the first phase shifter 11a, the second phase shifter 11b, the third phase shifter 11c and the fourth phase shifter 11d are monitored by a first phase monitor 15a described later. Also, the phases of the outputs of the phase shifters 11a, 11b, 11c, and 11d are adjusted so that the phases of the sine wave voltage signals are aligned. The first, second, third and fourth phase shifters 11a, 11b, 11c and 11d transmit respective output signals via coaxial cables to power amplifiers 12a, 12b, 12c and 12d which will be described later.
Reference numerals 12a, 12b, 12c and 12d are power amplifiers. Power amplifiers 12a, 12b, 12c and 12d amplify the power of the high frequency sinusoidal signals transmitted from the first, second, third and fourth phase shifters 11a, 11b, 11c and 11d, respectively. The maximum output of the power amplifiers 12a, 12b, 12c, 12d is, for example, 2 kW to 20 kW. Here, for example, it is 5 kW. The power amplifiers 12a, 12b, 12c and 12d respectively transmit power to matching units 13a, 13b, 13c and 13d, which will be described later, via coaxial cables.

Reference numerals 13a, 13b, 13c and 13d are matching units. Matching devices 13a, 13b, 13c, and 13d adjust the power transmitted from power amplifiers 12a, 12b, 12c, and 12d, respectively, so that reflected waves at the plurality of third feeding points 7a, 7b, 7c, and 7d are reduced. It has a function to adjust impedance. The matching devices 13a, 13b, 13c, and 13d respectively transmit impedance-adjusted power to first, second, third, and fourth vacuum current introduction terminals 75aa, 75bb, 75cc, and 75dd, which will be described later.
Numerals 75aa, 75ab, 75ac and 75ad are fifth, sixth, seventh and eighth vacuum current introduction terminals. The fifth, sixth, seventh, and eighth vacuum current introduction terminals 75aa, 75ab, 75ac, and 75ad receive the electric power transmitted from the matching units 13a, 13b, 13c, and 13d, respectively, through the vacuum coaxial cable 76aa. to a plurality of third feeding points 7a, 7b, 7c, 7d.
The vacuum coaxial cable 76aa is a coaxial cable using a dielectric that does not generate impurities.
Reference numeral 15a is a first phase monitor. The first phase monitor 15a observes the voltage phase of the electric power input to the fifth, sixth, seventh and eighth vacuum current introduction terminals 75aa, 75ab, 75ac and 75ad. When the phases of the voltages of the electric powers input to the fifth, sixth, seventh and eighth vacuum current introduction terminals 75aa, 75ab, 75ac and 75ad are not aligned, the first, second and third and fourth phase shifters 11a, 11b, 11c, 11d, and matching devices 13a, 13b, 13c, 13d. The first, second, The third and fourth phase shifters 11a, 11b, 11c and 11d and matching devices 13a, 13b, 13c and 13d are adjusted. In this phase adjustment operation, the setting of phase adjustment conditions is prioritized over the setting of matching conditions of the matching units 13a, 13b, 13c, and 13d.
Here, the electric power input to the fifth, sixth, seventh, and eighth vacuum current introduction terminals 75aa, 75ab, 75ac, and 75ad has the same voltage phase within ±10° as the third called electricity. If the phases of the voltages are aligned within ±10°, the wavefronts of the electric power waves propagating between the pair of electrodes 2 and 60 from the plurality of third feeding points are substantially aligned.
Further, here, from the output terminal 68a of the two-output phase-variable high-frequency oscillator 70, the third distributor 10a, the phase shifters 11a, 11b, 11c, 11d, the power amplifiers 12a, 12b, 12c, 12d, the matching device 13a. , 13b, 13c, 13d, fifth, sixth, seventh and eighth vacuum current introduction terminals 75aa, 75ab, 75ac, 75ad and a vacuum coaxial cable 76aa are connected to a third system line 69c. call.

The code|symbol 10b is a 4th distributor. The fourth divider 10b divides the sinusoidal voltage signal transmitted from the output terminal 68b of the two-output phase-variable high-frequency oscillator 70 into four phase shifters 11e and 11f, which will be described later. , to the seventh phase shifter 11g and the eighth phase shifter 11h.
References 11e, 11f, 11g and 11h are a fifth phase shifter, a sixth phase shifter, a seventh phase shifter and an eighth phase shifter, respectively. The fifth phase shifter 11e, the sixth phase shifter 11f, the seventh phase shifter 11g, and the eighth phase shifter 11h can adjust the phase of the input sinusoidal voltage signal and output it. possible and used in combination with a second phase monitor 15b, which will be described later. As will be described later, the fifth phase shifter 11e, sixth phase shifter 11f, seventh phase shifter 11g and eighth phase shifter 11h are monitored by a second phase monitor 15b described later. Also, the phases of the outputs of the phase shifters 11e, 11f, 11g and 11h are adjusted so that the phases of the sine wave voltage signals match. The fifth phase shifter 11e, the sixth phase shifter 11f, the seventh phase shifter 11g, and the eighth phase shifter 11h transmit respective output signals via coaxial cables to power amplifiers 12e, 12e, and 12e, which will be described later. 12f, 12g and 12h.
Reference numerals 12e, 12f, 12g and 12h are power amplifiers. Power amplifiers 12e, 12f, 12g and 12h amplify the power of the high frequency sinusoidal signals transmitted from the fifth, sixth, seventh and eighth phase shifters 11e, 11f, 11g and 11h, respectively. The maximum output of the power amplifiers 12e, 12f, 12g, 12h is, for example, 2 kW to 20 kW. Here, for example, it is 5 kW. The power amplifiers 12e, 12f, 12g and 12h respectively transmit power to matching units 13e, 13f, 13g and 13h, which will be described later, via coaxial cables.

Reference numerals 13e, 13f, 13g and 13h are matching units. Matching boxes 13e, 13f, 13g, and 13h convert the power transmitted from power amplifiers 12e, 12f, 12g, and 12h, respectively, into a plurality of fourth feeding points 7e, 7f, 7g, and 7h, respectively, so that the reflected waves are reduced. It has a function to adjust the impedance as follows. The matching devices 13e, 13f, 13g, and 13h respectively transmit impedance-adjusted electric power to ninth, tenth, eleventh, and twelfth vacuum current introduction terminals 75ba, 75bb, 75bc, and 75bd, which will be described later.
Reference numerals 75ba, 75bb, 75bc, and 75bd denote ninth, tenth, eleventh, and twelfth vacuum current introduction terminals. The ninth, tenth, eleventh, and twelfth vacuum current introduction terminals 75ba, 75bb, 75bc, and 75bd receive the power transmitted from the matching units 13e, 13f, 13g, and 13h, respectively, through the vacuum coaxial cable 76bb. to a plurality of fourth feeding points 7e, 7f, 7g, 7h.
The vacuum coaxial cable 76ba is a coaxial cable using a dielectric that does not generate impurities.
Reference numeral 15b is a second phase monitor. The second phase monitor 15b observes the voltage phase of the power input to the ninth, tenth, eleventh, and twelfth vacuum current introduction terminals 75ba, 75bb, 75bc, and 75bd. When the phases of the voltages of the electric powers input to the ninth, tenth, eleventh, and twelfth vacuum current introduction terminals 75ba, 75bb, 75bc, and 75bd are not aligned, the fifth phase shifter 11e, The sixth phase shifter 11f, the seventh phase shifter 11g and the eighth phase shifter 11h are adjusted, and the matchers 13a, 13b, 13c and 13d are adjusted. The fifth phase shifter is arranged so that the phases of the voltages of the electric powers input to the ninth, tenth, eleventh, and twelfth vacuum current introduction terminals 75ba, 75bb, 75bc, and 75bd are aligned within ±10°. 11e, sixth phase shifter 11f, seventh phase shifter 11g, eighth phase shifter 11h and matching devices 13e, 13f, 13g and 13h are adjusted. In this adjustment operation, the setting of phase adjustment conditions is prioritized over the setting of matching conditions of the matching units 13e, 13f, 13g, and 13h.

Here, the electric power input to the ninth, tenth, eleventh, and twelfth vacuum current introduction terminals 75ba, 75bb, 75bc, and 75bd has the same voltage phases within ±10°, and the fourth called electricity.
Further, here, from the output terminal 68b of the two-output phase-variable high-frequency oscillator 70, the fourth divider 10b, the first, second, third, fourth and fifth phase shifters 11e, 11f, 11g, 11h, power amplifiers 12e, 12f, 12g, 12h, matching devices 13e, 13f, 13g, 13h, 9th, 10th, 11th, 12th vacuum current introduction terminals 75ba, 75bb, 75bc, 75bd, and vacuum A power transmission circuit including the coaxial cable 76ba is called a fourth system line 69d.
Further, here, the two-output phase variable high frequency oscillator 70, the third system line 69c, and the fourth system line 69d are provided, and the third electric power is supplied to the plurality of third feeding points 7a, 7b, 7c, and 7d. A power supply means for supplying power and supplying the fourth power to the plurality of fourth power supply points 7e, 7f, 7g and 7h is called a second high frequency power supply means. The second high-frequency power supply means uses a phase shifter (not shown) incorporated in the two-output phase-variable high-frequency oscillator 70 to shift the voltage of the third power output from the third system line 69c to the fourth system line. The phase difference of the voltage of the fourth power output from 69d is phase-shifted in the form of Equation (5) or Equation (6) above.

Next, the operation of the thin film forming apparatus using plasma according to the second embodiment of the present invention will be described with reference to FIGS. 9 and 10. FIG. See also FIGS. For convenience of explanation, the formation of a microcrystalline silicon film will be described below as an example.
First, the substrate loading/unloading valve 62 of the reaction chamber 50 is opened, and the substrate 61 is placed on the main surface 60 a of the substrate holding table 60 . Next, after closing the substrate loading/unloading valve 62, the inside of the reaction chamber 50 is evacuated to a predetermined degree of vacuum through the exhaust port 51a and the exhaust port 51b by a vacuum pump (not shown).
After that, from a silane gas source (not shown) and a hydrogen gas source (not shown), silane gas and hydrogen gas whose flow rates are respectively controlled by mass flow controllers (not shown) for silane gas and hydrogen gas are introduced into the source gas introduction pipe 3 and the source gas dispersion cavity 6. It is ejected from the raw material gas ejection hole 5 via.
Here, it is assumed that the flow ratio of hydrogen gas and silane gas is 100 times, that is, flow rate of hydrogen gas/flow rate of silane gas=100.
Next, the opening/closing degree of the exhaust valve (not shown) is controlled by an exhaust valve control device (not shown) to maintain the internal pressure of the reaction container 50 at approximately 133 Pa to approximately 1333 Pa. It should be noted that the vacuum pump used for vacuuming the reaction vessel 50 and the vacuum pump used during film formation may be different. Here, for example, it is set to 500 Pa and maintained.

Next, using the second high-frequency power supply means, the third power is fed to the plurality of third feeding points 7a, 7b, 7c, 7d, and the fourth power is fed to the plurality of fourth feeding points 7e, 7f, 7g. , 7h. Then, when the time is stopped and the plasma is observed, a plasma having a wavy intensity distribution as shown in FIG. 10 is generated. FIG. 10 is a conceptual diagram showing plasma intensity on the xy plane, with the side 2b of the surface 2d of the non-grounded electrode 2 facing the grounded electrode 60 as the x-axis and the side 2a as the y-axis.
The phase of the voltage of the third power supplied to the plurality of fourth feeding points 7e, 7f, 7g and 7h and the voltage of the third power supplied to the plurality of fourth feeding points 7e, 7f, 7g and 7h The phase of is shown in FIG. 10 because the phase difference Δθ shifts like a ramp wave with 0° to 360° as one period T ₀ as shown in FIGS. 5(a) and 5(b) Plasma continues to move in the plus or minus direction of the x-axis. Here, the one period _T0 is selected from the range of 1 μs to 1 msec, for example, and is set to 1 msec.
When the plasma shown in FIG. 10 is generated, that is, when the raw material gas ejected from the raw material gas ejection holes 5 is turned into plasma, radicals such as _{SiH 3} and H, which are precursors of the microcrystalline silicon film, are generated. and microcrystalline silicon is deposited on the substrate 61 .
As shown in FIG. 10, the microcrystalline silicon film is moving averaged because the plasma moves in one direction with respect to the substrate. As a result, the thickness of the film formed on the substrate 61 is made uniform. This is similar to the averaging effect caused by the relationship between the plasma between the pair of electrodes 1 and 60 and the substrate in the film deposition process in the plasma CVD apparatus of the substrate moving system as shown in FIG. 8(b). It is due to the effect.

Next, since the thickness of the microcrystalline silicon film is proportional to the plasma generation time, when a predetermined time has elapsed after the start of output supply from the second high-frequency power supply means, the second high-frequency power supply means to zero the output of
The film-forming time is determined based on previously acquired data. Here, it is 1 minute to 30 minutes, for example, 5 minutes. The film forming time can be determined based on previously obtained data relating to the relationship between the electrode spacing, substrate temperature, silane gas flow rate, hydrogen gas flow rate, pressure, electric power, and the like.
After the desired formation of the microcrystalline silicon film is completed, the supply of the silane gas and hydrogen gas is stopped, and the inside of the reaction vessel 50 is once evacuated to a high degree of vacuum. The reaction vessel 50 is then returned to atmospheric conditions. After the reaction vessel 50 is returned to atmospheric conditions, the substrate loading/unloading valve 62 is opened and the substrate 61 is taken out.
A uniform microcrystalline silicon film is formed on the substrate 61 taken out from the reaction vessel 50.

In the thin film forming apparatus using plasma according to the second embodiment of the present invention, as described above, the voltage of the third electric power supplied to the plurality of third feeding points 7a, 7b, 7c, and 7d and the plurality of third The phase difference Δθ(t) of the voltage of the fourth electric power supplied to the four feeding points 7e, 7f, 7g, and 7h is, for example, 0° to 360° for one cycle T, as shown in FIG. ₀ , the phase shifts like a ramp wave at a constant speed, so the plasma generated between the pair of electrodes 1 and 60 continues to move in a constant direction. As a result, the thickness of the film deposited on the substrate placed on the ground electrode 60 is uniform.
As a method for continuously moving the plasma generated between the pair of electrodes 1 and 60 in a constant direction, the phase of the voltage of the third electric power fed to the plurality of third feeding points 7a, 7b, 7c, and 7d and the , the phase difference Δθ between the phases of the voltages of the fourth electric power fed to the plurality of fourth feeding points 7e, 7f, 7g, and 7h, as shown in FIG. 0° to 180° may be defined as one cycle T ₀ , and the phase may be shifted in a ramp-wave manner at a constant speed. One cycle _T0 can be arbitrarily selected in the range of 1 μs to 1 ms.
As a result, the plasma CVD apparatus according to the second embodiment of the present invention can uniformly form a film by plasma CVD on a very large substrate, which is difficult with the conventional apparatus.

1... non-grounded electrode,
3... Raw material gas introduction pipe,
5... Raw material gas ejection hole,
6... Raw material gas dispersion cavity,
55 Insulator,
60... ground electrode,
60a ... the main surface of the ground electrode,
61 Substrate,
70 2 output phase variable high frequency oscillator,
72a ... the first power amplifier,
72b second power amplifier,
73a ... the first matching box,
73b... second matching box,
75a, 75b... first and second current introduction terminals for vacuum,
75c, 75d... fourth and fifth current introduction terminals for vacuum,
76a, 76b, 76c, 76d... vacuum leads,
77a, 77b... first feeding point,
77c, 77d... second feeding point,
79a, 79b... First and second distributors.

Claims (9)

a reaction vessel equipped with an exhaust system; a pair of electrodes arranged inside the reaction vessel and composed of a non-grounded electrode and a grounded electrode on which a substrate is mounted; a raw material gas introducing means for introducing a raw material gas therebetween; a plurality of first feeding points provided at one end of the non-grounded electrode facing each other; and a plurality of second feeding points provided at the other end of the non-grounded electrode. high-frequency power supply means for supplying the first power and the second power to the
In a plasma CVD apparatus for forming a thin film on the substrate by converting the raw material gas into plasma in a plasma generation region sandwiched between the pair of electrodes,
The high-frequency power supply means includes: a first system line that supplies the first power to the plurality of first feeding points; a second system line that supplies the second power to the plurality of second feeding points; A phase shifter that controls the phase difference between the voltages of the first power and the second power,
The phase shifter is arranged such that the phase of the voltage of the output section of the first system line continues to lag behind the phase of the voltage of the output section of the second system line at a constant speed or advances at a constant speed. A plasma CVD apparatus characterized by controlling the former phase and the latter phase so as to continue. The phase shifter repeatedly phase-shifts the phase difference between the voltage of the output section of the first system line and the voltage of the output section of the second system line at a constant speed with one period from 0° to 360°. 2. The plasma CVD apparatus according to claim 1, wherein the phase shift is repeated at a constant speed with one period from 0° to minus 360°. The phase shifter repeatedly shifts the phase difference between the voltage of the output section of the first system line and the voltage of the output section of the second system line from 0° to 180° in the form of a ramp wave, or 0 2. The plasma CVD apparatus according to claim 1, wherein the phase shift is repeated in the form of a ramp wave from degrees to minus 180 degrees. The phase shifter repeatedly shifts the phase difference between the voltage of the output section of the first system line and the voltage of the output section of the second system line in one period of approximately 1 μs to 1 ms. 4. The plasma CVD apparatus according to claim 2 or 3. Adjacent intervals between the plurality of first feeding points are arranged to be one-eighth or less of the wavelength of the voltage of the first power, and adjacent intervals between the plurality of second feeding points are the second 5. The plasma CVD apparatus according to any one of claims 1 to 4, wherein the plasma CVD apparatus is arranged so as to be one-eighth or less of the wavelength of the voltage of the electric power. The first feeding point and the second feeding point are provided at an end of a surface of the non-grounded electrode that does not face the ground electrode, a side surface of the non-grounded electrode, or an end of a surface that faces the grounded electrode. 6. The plasma CVD apparatus according to any one of claims 1 to 5, characterized by: 7. The high-frequency power supply means according to any one of claims 1 to 6, characterized in that it comprises a two-output phase-variable oscillator, a divider, a power amplifier, and a matching box. plasma CVD equipment. The plasma according to any one of claims 1 to 7, wherein the frequency of the power generated by the high-frequency power supply means is a frequency selected from a VHF band of 13.56 MHz or 30 MHz to 300 MHz. CVD equipment. 9. The plasma CVD apparatus according to claim 1, wherein said raw material gas contains at least silane gas or organic silane gas.

Images