JP5114251B2 - Vacuum processing equipment - Google Patents

Description

  The present invention relates to the technical field of pressure measurement, and more particularly to the technical field of pressure measurement using a diaphragm vacuum gauge.

In a semiconductor process, the process is often performed in a pressure range of 0.01 to 100 Pa.
The vacuum gauge used in this pressure range is a diaphragm vacuum gauge.

Reference numeral 110 in FIG. 3 denotes a vacuum processing apparatus such as a CVD apparatus or an etching apparatus, and includes a vacuum chamber 111. A thermostat 112 is arranged outside the vacuum chamber 111, and a diaphragm vacuum gauge 113 is arranged inside the thermostat 112. The diaphragm vacuum gauge 113 is connected to the vacuum chamber 111 by a branch pipe 114.
A vacuum exhaust system 119 is connected to the vacuum chamber 111, and the inside of the vacuum chamber 111 is configured to be placed in a vacuum atmosphere by the vacuum exhaust of the vacuum exhaust system 119.

The internal pressure of the vacuum chamber 111 is transmitted by the diaphragm vacuum gauge 113 via the branch pipe 114, and the internal pressure of the vacuum tank 111 is determined by the diaphragm vacuum gauge controller 116 by the capacitance value of the capacitor inside the diaphragm vacuum gauge 113. Is detected and converted to a pressure value.
The temperature of the thermostat 112 is maintained at a constant temperature slightly higher than room temperature, so that no temperature drift occurs in the capacitance value of the diaphragm vacuum gauge 113.

  In the pressure range of 0.01 to 100 Pa, the gas flow in the branch pipe 114 changes to the molecular flow / intermediate flow / viscous flow region. However, if the temperature of the diaphragm vacuum gauge and the vacuum chamber are different, the instruction of the diaphragm vacuum gauge A complex shift (thermal transition effect) depending on the pressure / temperature difference occurs between the value and the actual pressure inside the vacuum chamber.

Control of the process gas pressure with high accuracy, but reproducibility is required, for example, when the CVD chamber 100 ° C. of gas flow measured by the vacuum gauge held in 45 ° C., up to 8.3% by the pressure area Pressure error ranging from 0 to 0%. Further, when a normal temperature gas of 25 ° C. is flowing in the sputtering chamber, a pressure error in the range of 3.3% to 0% at maximum occurs when measured with a vacuum gauge maintained at 45 ° C. This is an error that cannot be ignored as a process variation factor.
However, there is no vacuum processing apparatus that correctly incorporates the thermal transition effect in pressure measurement.

The compensation method for the thermal transition effect is described, for example, in the following non-patent document.
“Thermal Transpiration Effect of Hydrogen, Rare Gases and Methane”, T. Takaishi, Y. Sensui, Trans. Farad. Soc. 59 (1963) 2503

  An object of the present invention is to provide a pressure measurement technique that reflects the thermal transition effect.

In order to solve the above-described problems of the prior art, the present invention provides a vacuum chamber that can be evacuated, a diaphragm vacuum gauge that measures pressure as a capacitance value, and a zero that connects the internal atmosphere of the vacuum chamber and the diaphragm vacuum gauge. Branch pipes having the above-mentioned tube lengths, first and second thermometers for measuring temperatures of the vacuum chamber and the diaphragm vacuum gauge, respectively, measurement results of the diaphragm vacuum gauges, and the first and second thermometers A pressure measurement device to which the measurement result is input, and the pressure measurement device includes the measurement result of the diaphragm vacuum gauge, the temperature of the vacuum chamber, the temperature of the diaphragm vacuum gauge, and the vacuum chamber. The relationship between the internal pressure and the internal pressure is stored, and the calculation device is configured to calculate the internal pressure of the vacuum chamber from the measurement result input to the calculation device, and the pressure measurement device includes the branch pipe When the ratio L / D of the tube length L to the inner diameter D is greater than the reference value Wherein a relationship between a vacuum processing apparatus in which the relationship is stored separately in smaller.
Further, in the present invention, when the reference value is 10 and the ratio value L / D of the tube length L and the inner diameter D is larger than 10, the relationship between the following expressions (1) and (2) is satisfied. This is a vacuum processing apparatus to be used.

The present invention is constructed as described above, can be used a Monte Carlo method for the evaluation of thermal transpiration effects, Monte Carlo method using the simulated lean gas stream is generally D irect S imulation M onte C arlo Method (DSMC Method) (or Monte Carlo direct method), and the procedure will be described below.

[Procedure 1] Setting the analysis area Set the probe, branch pipe, and part of the vacuum vessel as the analysis area. It is arbitrary about which part of the vacuum vessel is included in the analysis region. In this embodiment, the height is selected so as to match the branch pipe diameter and the radius matches the radius of the measuring element.

[Procedure 2] Setting of boundary conditions Boundary conditions related to temperature distribution and particle reflection law are given to the boundary surrounding the analysis region. In this embodiment, the perfect diffuse reflection law is applied to all the boundaries. A uniform temperature of 300K was given to the probe boundary, a uniform temperature of 600K to the vacuum vessel boundary, and a temperature distribution that changed stepwise from 300K to 600K at the center of the branch pipe was given to the branch pipe boundary.

[Procedure 3] Initial Conditions of Particles N _S sample particles are arranged in the analysis region, and a velocity corresponding to the initial gas temperature T ₀ is given. In this example, about 10 ⁵ sample particles were uniformly arranged in the analysis region. In addition, 300K, which is equal to the wall temperature of the probe, was selected as the initial gas temperature.

[Procedure 4] Determination of Mass, Diameter, and Weight of Sample Particles The mass m of the sample particles is calculated from the molar mass M of the gas molecules from the following formula (A),

Decide by. N ₀ is the Avogadro number.
The diameter d of the sample particles is obtained from the following formula (B) using the gas viscosity coefficient η (assumed to be known).

here,
k: Boltzmann constant (J / K)
T: Gas temperature (K)
d: Molecular diameter when the molecular model is a hard sphere model (m)
And In this embodiment, the initial temperature T ₀ is selected as the gas temperature T, and the viscosity coefficient at that temperature is used. In addition, the average temperature of the low temperature part and the high temperature part may be used as the gas temperature, and the viscosity coefficient at that temperature may be used.
One sample particle represents a large number of real particles. The degree of overlap is called weight W,

It is defined as here,
P ₀ : Initial pressure (Pa)
V: Volume of analysis area (m ³ )
And

[Procedure 5] Tracking particle motion
DSMC method is used to track the translation and collision motion of sample particles.
The procedure for tracking particle motion in the DSMC method will be described. Since the shape of the analysis region has axial symmetry, the DSMC method of axial symmetry system was used.

<1> to <6>.
<1>
Divide the analysis area into cells.
<2>
Using the position and velocity of the particle at the time, the position of the particle at the time t + Δt is obtained from the following equation.
r (t + Δt) = r (t) + v (t) Δt
Here, Δt is a time step, and r (t) and v (t) are the position and velocity of the particle at time t, respectively.
<3>
Using r (t + Δt), it is checked which of the cells prepared in <1> the particle belongs to. Hereinafter, this cell is referred to as a belonging cell.
<4>
1: The particle in which the belonging cell is found is updated by updating the particle position to r (t + Δt) and setting the particle velocity v (t + Δt) at time t + Δt equal to v (t).

2: Particles for which no affiliation cell was found crossed the boundary wall surrounding the analysis region and then moved out of the analysis region. In this problem, since all the boundary walls are reflective walls, particles are not allowed to enter the analysis area. The particles are reflected when the boundary wall is crossed by the following procedure. First, using the flight time Δt ′ to the boundary wall, a position that intersects the boundary wall r (t + Δt ′) = r (t) + v (t) Δt ′
And the remaining time
Δt ″ = Δt−Δt ′
Ask for. The velocity v (t + Δt ′) after reflection on the wall surface of the particle is obtained from the conditions of the boundary wall temperature and the diffuse reflectance at the intersection position. Position after reflection r (t + Δt) = r (t + Δt ′) + v (t + Δt ′) Δt ″
To update the position. The belonging cell is examined using this position.
The particle velocity v (t + Δt) at time t + Δt is updated to be equal to v (t + Δt ′).

<5>
According to the above procedure, the affiliation cell of all particles was determined. The following operations are performed in each cell. The number N of particle pairs to be selected is determined according to the probability law. For each of the selected N particle pairs, the presence or absence of collision is determined based on the probability law. For colliding particles, the particle velocity is updated to the velocity after the collision.
<6>
Using the updated position and velocity of the particle, the operations from <2> to <5> are repeated, and the translational / collision motion of the particle for each Δt is tracked.

[Procedure 6] Pressure sampling From time t to t + Δt, the lower end of the analysis region (z = 0: diaphragm surface of the probe) and the upper end (z = H: end of the analysis region in the vacuum chamber) It is assumed that the number of sample particles entering and reflecting on the surface is N _G and N _V , and the change in the component perpendicular to the incident surface of the momentum of each sample particle i entering and reflecting is | Δ (mv _z ) _i |. The areas of the respective incident surfaces are S _G and S _V. The pressure on each surface at time t

It is defined as
[Procedure 7] The thermal transition coefficient is obtained.
The pressure in the steady state of P _G and P _V respectively

And Knudsen number

And thermal transition coefficient

Ask for. here,

It is. D is the diameter of the branch pipe.
[Procedure 8] Creation of Data Set The initial pressure is applied discretely and a data set (Kn _k , γ _k ) is created by the above procedure. k is a data set number, and k is given so that Kn _k <Kn _{k + 1} . In this example, assuming a vacuum gauge of medium vacuum, 13 points were selected for P ₀ between 1 × 10 ^{−3 and} 1.5 × 10 ² Pa.

[Procedure 9] Interpolation of thermal transition coefficient The value calculated by the equation (G) and the indication value of the diaphragm vacuum gauge (steady state P _G )

Is assigned to

Ask for. Interpolate the data set obtained in step 8

Ask for. For example, as an interpolation method

That Do and

To using a linear interpolation for

I ask. Other

High-order interpolation is performed using many neighboring points,

You may ask for.
[Procedure 10] Calibration of vacuum gauge reading

It was determined by T _G, T _V and Step 9

Substituting into equation (F), the estimated value of the vacuum vessel pressure

Ask for.

  Even when there is a pressure difference between the vacuum chamber of the vacuum process device and the diaphragm vacuum gauge due to thermal transition, the pressure of the vacuum chamber can be correctly obtained by using the present invention, so that the controllability / reproduction of the vacuum process is possible. It is possible to improve the property.

  Reference numeral 10 in FIG. 1 denotes a vacuum processing apparatus according to the present invention, which has a vacuum chamber 11. An evacuation system 19 is connected to the vacuum chamber 11, and vacuum processing such as film formation by CVD or sputtering, etching by RIE, etc. can be performed while evacuating the inside of the vacuum chamber 11 by the evacuation system 19. It has become so.

A pressure measuring device 20 and a thermostatic chamber 12 are arranged outside the vacuum chamber 11.
A diaphragm vacuum gauge 13 is disposed inside the thermostatic chamber 12, and the diaphragm vacuum gauge 13 is maintained at a constant temperature. The diaphragm vacuum gauge 13 is connected to the vacuum tank 11 by a branch pipe 14, and the diaphragm vacuum gauge 13 is configured to contact the internal atmosphere of the vacuum tank 11 through the branch pipe 14.

An example of the internal structure of the diaphragm vacuum gauge 13 is shown in FIG.
The diaphragm vacuum gauge 13 is partitioned into a measurement chamber 33 and a high vacuum chamber 34 by a diaphragm 30 made of metal foil (or a ceramic diaphragm and an electrode).
The measurement chamber 33 is connected to the vacuum chamber 11 by a branch pipe 14, and the internal pressure of the vacuum chamber 11 is transmitted to the measurement chamber 33 by the branch tube 14, and the diaphragm 30 is made to be a high vacuum chamber according to the internal pressure of the measurement chamber 33. Swells to the 34th side.

An electrode is disposed in the high vacuum chamber 34. Here, the first and second electrodes 31 and 32 are respectively arranged so as to face the diaphragm 30, and the first electrode 31 and the diaphragm 30 are interposed between the first and second electrodes 31 and 32. The capacitor configured and the capacitor configured between the diaphragm 30 and the second electrode 32 are connected in series. The capacitance of the capacitor between the first and second electrodes 31 and 32 varies depending on the degree of swelling of the diaphragm 30.
The pressure measurement device 20 includes a vacuum gauge control device 16, a calculation device 25, and a display device 28.

  The first and second electrodes 31 and 32 are connected to the vacuum gauge control device 16, and the capacitance value between the first and second electrodes 31 and 32 is detected by the vacuum gauge control device 16 and a signal indicating the pressure. To be output to the calculation device 25.

A first thermometer 21 is attached to the inner wall surface of the vacuum chamber 11, and a second thermometer 22 is attached to the outer surface of the diaphragm vacuum gauge 13. The temperature of the vacuum tank 11 and the diaphragm vacuum gauge 13 is Are measured by the first and second thermometers 21 and 22, respectively, and the measurement results are output to the calculation device 25.
The measured value of the diaphragm vacuum gauge 13 with respect to the internal pressure of the vacuum chamber 11 is affected by the temperature of the vacuum chamber 11, the temperature of the diaphragm vacuum gauge 13, and the shape of the branch pipe 14.

  In the present invention, a straight pipe having a circular cross section is used as the branch pipe 14. When the inner diameter (inner peripheral diameter) D and the pipe length L of the branch pipe 14 are L / D> 10, the following (1) It is known that the formula holds.

  Where X is

It is.
Experimental values obtained as values of A, B, and C are shown in Table 1 below (values described in the above-mentioned non-patent literature).

If the pipe length L of the branch pipe 14 is shorter than the inner diameter D of the branch pipe 14 and L / D> 10 does not hold, an error will increase if the above equations (1) and (2) are used. (e.g., D irect S imulation M onte C arlo method; DSMC method) by and experimental, and the temperature T _G of the diaphragm vacuum gauge 13, and the temperature T _V of the vacuum chamber 11, and the internal pressure P _V of the vacuum chamber 11, the diaphragm relation indicated pressure P _G in the vacuum gauge 13 is determined in advance.

FIG. 4 shows the case of L and D ≦ 10 (calculation was performed under the condition that a cylindrical space of φ50 held at 300K and 600K is connected by an orifice of D = 10 mm (L = 0)). It is a graph showing the relationship, the symbol S ₁ is a curve representing the relationship between X and γ determined by the DSMC method, and the symbol S ₂ is a curve describing the equations (1) and (2) in the same figure.

In the case of L / D ≦ 10, the curve obtained by the DSMC method and the curve obtained from the equations (1) and (2) do not match. Since it is known that the curve S ₁ obtained by the DSMC method is more consistent with the experimental value, the DSMC method is used in the range of L / D ≦ 10 rather than using the equations (1) and (2). It is desirable to use the determined relationship.
FIG. 5 shows a case where L / D> 10. Reference numeral S ₃ is a DSMC method, and reference numeral S ₄ is a curve obtained from equations (1) and (2).

For L / D> 10 and the curve obtained by DSMC method (1), because it matches substantially the curve obtained from the equation (2), so that both can have use of.
The calculation device 25 stores both the above expressions (1) and (2) and the relational expression obtained by the DSMC method.

Temperature T _V and the membrane vacuum gauge 13 First temperature T _G of the vacuum chamber 11, the second thermometer 21, the indicated pressure P _G by a diaphragm vacuum gauge 13, are respectively measured at constant time intervals The measured value is immediately input to the calculation device 25.

The inner diameter D is known, the constants A, B, C and the inner diameter D are input in advance. When the gas type stored in the calculation device 25 is selected, if L / D> 10, the calculation device 25 first, the value input from the second thermometer 21, 22 and the membrane vacuum gauge 13 (1), is substituted into equation (2), X, seek gamma, and calculates the internal pressure P _V, a display device 28.

In the case of L / D ≦ 10, the internal pressure P _V is calculated from the relationship stored with the values input from the first and second thermometers 21 and 22 and the diaphragm vacuum gauge 13, and similarly, the display device 28.
Thereby, the measured value close to the true value is displayed.

  In the above embodiment, both the expressions (1) and (2) and the relational expression obtained by the DSMC method are stored in the calculation device 25, but the L / D value of the branch pipe 14 to be used is determined. If L / D> 10 holds, only the expressions (1) and (2) may be stored, and if L / D ≦ 10, only the relational expression obtained by the DSMC method may be stored.

Further, the DSMC method, and the temperature T _V of the temperature T _G and the vacuum chamber 11 of the diaphragm vacuum gauge 13, and the internal pressure P _V of the vacuum chamber 11, of obtaining a relational expression indicated pressure P _G of the diaphragm vacuum gauge 13 no stores numbers were determined as a database, and compares the temperature T _V of the temperature T _G and the vacuum chamber 11 of the diaphragm vacuum gauge 13, the measured value of the command pressure P _G of the diaphragm vacuum gauge 13 in the database, the vacuum chamber 11 The internal pressure P _V may be obtained.
As the calculation device 25 that stores the relationships according to the equations (1) and (2) and the DSMC method, a control device that controls the process in the vacuum chamber 11 can be used.

  In the said Example, although the diaphragm vacuum gauge 13 measured the capacitance value between the 1st, 2nd electrodes 31 and 32, it is a diaphragm vacuum gauge which measures the capacitance value between a single electrode and a diaphragm. There may be. In short, the diaphragm vacuum gauge that can be used in the present invention can detect the swelling of the diaphragm affected by the internal pressure of the vacuum tank connected by the branch pipe (including the case of zero length) by the capacitance value. If it is.

<How to find A, B, C>
In Table 1 above, known experimental values are shown as the values of A, B, and C. For gases whose coefficients have not been determined by experiments, the values of A, B, and C are determined by the following procedure. Can do.
First, it is assumed that the viscosity coefficient η of a normal temperature gas is known. By substituting these temperatures and viscosity coefficient η into the following equation (3), the molecular diameter d is obtained.

  Next, the coefficients A, B, and C are obtained by substituting the molecular diameter d into the following equation (4).

<Correction factor for short pipe length>
As shown in Fig. 4, in the case of short pipe length (when L / D≤10), (1) and (2) are in error with the experimental results. If so, the relational expression need not be obtained by the DSMC method.
A method for correcting equations (1) and (2) will be described.
Experiments or numerical experiments, the indicated pressure P _G in the internal pressure P _V and diaphragm vacuum gauge 13 for the temperature T _V and the vacuum chamber 11 of the temperature T _G and the vacuum chamber 11 of the diaphragm vacuum gauge 13, a plurality of different pressures and temperatures And each measured value is expressed by the following formulas (11) and (12):

To obtain a plurality of (X, γ).
If there are N pairs of (X, γ), and the k-th group is represented by (X _k , γ _k ), the following equation (13):

The correction coefficient α that gives the minimum value of S2 calculated in step (1) is calculated, and the following correction (1) equation:

And the above equation (2), L / D of the temperature T _G of the diaphragm vacuum gauge 13, the temperature T _V of the vacuum tank 11, the internal pressure P _V of the vacuum tank 11, and the indicated pressure P _G of the diaphragm vacuum gauge 13 stored in the calculator 25 as a relation in the case of ≦ 10, it can be made to determine the internal pressure P _V of the vacuum chamber 11 from the measured value. The value of the correction coefficient α is 0.5 or more and 1.2 or less.

The thermal transition coefficient γ to pair the Ar gas was determined from the numerical experiment by DSMC method. The calculation conditions were L = 10 mm, D = 10 mm, a uniform temperature of 600 K for the vacuum chamber 11 and 300 K for the diaphragm vacuum gauge, and a temperature distribution that changed stepwise at the center of the branch pipe.

The numerical experiment results are plotted in the graph of FIG.
The dotted line is a curve according to the modified equations (1) and (2) with α = 0.73, and the solid line is a curve according to equations (1) and (2). The dotted curve shows a high agreement with the plot.

  The usable range of the length of the branch pipe is determined in advance, and a correction coefficient α is obtained for a plurality of pipe lengths between the upper limit value and the lower limit value, and the correction coefficient of the actually used branch pipe length is calculated. When α is not obtained, the obtained correction coefficient α is subjected to linear interpolation, high-order interpolation, or function approximation to obtain the correction coefficient α of the pipe length to be used.

First, correction coefficients α ₀ , α ₁₀ , α ₅₀ , α ₁₀₀ , and α ₂₀₀ were determined at D = 10 mm, L = ₀ , ₁₀ , ₅₀ , ₁₀₀ , and ₂₀₀ mm. The results are plotted on the graph of FIG.
The solid line is the following equation (21),

Κ ₁ = α ₀ , the coefficient κ ₂ is a value obtained by subtracting α ₀ from the asymptotic value of L / D, and the coefficient κ ₃ is L / D = 0, 1, 5 , 10, and 20 are selected so that the sum of the squares of the differences between the values of the curve α (L / D) and the values of α ₀ , α ₁₀ , α ₅₀ , α ₁₀₀ , and α ₂₀₀ is minimized.
In the case of L = 30 mm, α = 0.683 according to the above equation (21).

The figure for demonstrating the vacuum processing apparatus of this invention The figure for demonstrating an example of the diaphragm vacuum gauge used for the vacuum processing apparatus The figure for demonstrating the vacuum processing apparatus of a prior art The graph explaining the relationship used when L / D ≦ 10 A graph for comparing the relationship obtained by the DSMC method when L / D> 10 with the relationship obtained from equations (1) and (2). Graph for explaining the results of numerical experiments when L = 10 mm and D = 10 mm A graph for explaining a correction coefficient when L = 10 mm and D = 10 mm

10 ...... vacuum processing apparatus 11 ...... vacuum chamber 13 ...... diaphragm vacuum gauge 20 ...... pressure measuring device 21 ...... first thermometer 22 ...... second thermometer gamma ...... heat transition of T _V ...... vacuo Tank temperature (K)
P _V ...... Internal pressure of vacuum chamber (Pa)
T _G …… Temperature of diaphragm vacuum gauge (K)
P _G …… Pressure (Pa) obtained from the capacitance value of the diaphragm vacuum gauge
A, B, C ... Constant depending on gas type D ... Branch pipe inner diameter (mm)
L: Branch pipe length (mm)
α …… Correction factor

Claims (2)

A vacuum chamber that can be evacuated;
A diaphragm gauge for measuring pressure as a capacitance value;
A branch pipe having zero or more pipe lengths connecting the internal atmosphere of the vacuum chamber and the diaphragm vacuum gauge;
A first thermometer and a second thermometer for measuring the temperature of the vacuum chamber and the diaphragm vacuum gauge, respectively;
A pressure measuring device to which the measurement result of the diaphragm vacuum gauge and the measurement result of the first and second thermometers are input;
In the pressure measuring device, the relationship between the measurement result of the diaphragm vacuum gauge, the temperature of the vacuum chamber, the temperature of the diaphragm vacuum gauge, and the internal pressure of the vacuum chamber is stored. The internal pressure of the vacuum chamber is calculated from the measurement result input to the calculation device ,
The pressure measurement device is a vacuum processing device in which the relationship when the ratio L / D of the length L and the inner diameter D of the branch pipe is larger than a reference value and the relationship when the ratio is small are stored separately . The reference value is set to 10, if the value L / D ratio of the inner diameter D and the tube length L is greater than 10, the following (1) and (2) the relationship of claim 1, wherein the use of Vacuum processing equipment.

Images