JP2010093096A - Temperature control method of heat treatment apparatus and heat treatment apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To suppress deterioration in in-furnace temperature stability due to film deposition, change in the number of produced wafers, etc. <P>SOLUTION: A heat treatment apparatus includes a heating means 1 of heating the inside of a processing chamber for processing a substrate, a heating control part 71 for controlling the heating means, and a first and a second temperature detecting means for detecting the temperature in the processing chamber, wherein the first temperature detecting means is arranged closer to the substrate than the second temperature detecting means 2, which is arranged closer to the heating means than the first temperature detecting means 3. The heating control part finds the time difference between the maximum temperature point of time of detected temperature by the second temperature detecting means 2 and the maximum temperature point of time of detected temperature by the first temperature detecting means 3, and controls the heating means such that when the temperature difference is larger than a predetermined time difference, the time difference becomes equal to or less than the predetermined time difference. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a temperature control method for a heat treatment apparatus and a heat treatment apparatus.

Conventionally, in a semiconductor manufacturing apparatus, since it is necessary to maintain the temperature in the furnace at an appropriate temperature or to follow the temperature change designated in the furnace, the control device uses a heater based on a preset temperature change pattern. (For example, refer to Patent Document 1).
WO 2006/070552

  Even if the temperature control method and the temperature control adjustment method, which are conventional techniques, are used, it may not be possible to cope with changes in temperature characteristics depending on the film deposited on the tube, the number of wafers, and the wafer type, and the temperature stabilization time may be delayed.

  The present invention has been made in order to solve the above-described problems. A heat treatment apparatus and a heat treatment apparatus capable of updating a part of a preset operation amount pattern in response to a change in furnace temperature characteristics. An object is to provide a temperature control method.

  In order to solve the above-described problem, a temperature control method of a heat treatment apparatus according to the present invention includes a heating unit that heats a processing chamber for processing a substrate, a heating control unit that controls the heating unit, and a temperature in the processing chamber. First and second temperature detecting means for detecting, the first temperature detecting means is disposed closer to the substrate than the second temperature detecting means, and the second temperature detecting means is A temperature adjustment method in a heat treatment apparatus disposed closer to the heating means than the first temperature detection means, wherein the first temperature detection means starts from the maximum temperature point of the temperature detected by the second temperature detection means. A step of obtaining a time difference between the detected temperature of the detected temperature and the maximum temperature, and a process for controlling the heating means so that the time difference is not more than a predetermined time difference if the time difference is not less than a predetermined time difference. It is characterized in that it has a.

  Further, the heat treatment apparatus according to the present invention includes a heating means for heating the processing chamber for processing the substrate, a heating control unit for controlling the heating means, and first and second temperature detections for detecting the temperature in the processing chamber. And the first temperature detection means is disposed closer to the substrate than the second temperature detection means, and the second temperature detection means is more heated than the first temperature detection means. A heat treatment apparatus disposed at a position close to the means, wherein the heating control unit includes a maximum temperature time point of the detected temperature by the first temperature detecting means from a maximum temperature time point of the detected temperature by the second temperature detecting means. When the time difference is equal to or greater than a predetermined time difference, the heating means is controlled so that the time difference is equal to or less than the predetermined time difference.

  Embodiments of the present invention will be described below with reference to the drawings. The present invention relates to a control method for controlling a control amount output from a heating unit by using feedback control based on a proportional / integral / differential (PID) operation for an operation amount input to the heating unit in temperature control in a heat treatment apparatus. This relates to the automatic temperature adjustment procedure.

(First embodiment)
FIG. 1 is a functional block diagram for explaining a heat treatment apparatus according to the first embodiment of the present invention, and FIG. 2 is a view for explaining details of a configuration around a reaction tube in the apparatus.

  The heat treatment apparatus according to the present embodiment is configured to include a processing furnace 91 and a main control unit 7. In FIG. 1, only the temperature control unit 71 is extracted from the main control unit 7 and displayed.

  The processing furnace 91 includes a heater (heating means) 1, a quartz cap 1a, a heater thermocouple (second temperature detection means) 2, a cascade thermocouple (first temperature detection means) 3, a boat 4, a reaction tube 5, a leveling unit. Heat pipe 6, exhaust pipe 8, boat elevator 9, base 10, gas supply pipe 11, mass flow controller (MFC) 12, pressure regulator (APC) 13, pressure sensor 14, O-ring 15, seal cap 16, rotating shaft 17, An introduction port 18 and an exhaust port 19 are provided.

    The main control unit 7 includes a temperature control unit (heating control unit) 71, a gas flow rate control unit 72, a pressure control unit 73, and a drive control unit 74.

The soaking tube 6 is made of a heat-resistant material such as SiC, and is formed in a cylindrical shape with the upper end closed and the lower end opened. For example, a reaction vessel (hereinafter referred to as reaction tube 5) made of a heat resistant material such as quartz (SiO ₂ ) is formed in a cylindrical shape having an opening at the lower end, and is disposed concentrically with the heat equalizing tube 6 in the heat equalizing tube 6. Yes. A gas supply pipe 11 made of, for example, quartz and an exhaust pipe 8 are connected to the lower part of the reaction tube 5, and an introduction port 18 connected to the gas supply pipe 11 extends from the lower part of the reaction tube 5 to the side of the reaction tube 5. For example, it has a structure that rises into a thin tube and reaches the inside of the reaction tube 5 at the ceiling. The exhaust pipe 8 is connected to the exhaust port 19 of the reaction pipe 5. The gas flows from the gas supply pipe 11 through the ceiling of the reaction tube 5 into the reaction tube 5 and is exhausted from the exhaust pipe 8 connected to the lower part of the reaction tube 5.

A gas for processing is supplied into the reaction tube 5 through the gas supply tube 11 to the inlet 18 of the reaction tube 5. The gas supply pipe 11 is connected to a mass flow controller (MFC) 12 as a gas flow rate control means or a moisture generator (not shown). The mass flow controller 12 is connected to the gas flow rate control unit 72 and has a configuration capable of controlling the flow rate of the supplied gas or water vapor (H ₂ O) to a predetermined amount.

  The gas flowing through the reaction tube 5 is discharged from the exhaust port 19 of the reaction tube 5. A gas exhaust pipe 8 connected to a pressure regulator (for example, APC) 13 is connected to the exhaust port 19 of the reaction pipe 5, and the pressure in the reaction pipe 5 is detected by pressure detection means (hereinafter, pressure sensor 14). And the APC 13 is controlled by the pressure control unit 73 so that the pressure in the reaction tube 5 becomes a predetermined pressure.

At the lower end opening of the reaction tube 5, a disk-shaped holding body (hereinafter referred to as a base 10) made of, for example, quartz is detachable so as to be airtightly sealed through an O-ring 15, and the base 10 is a disk-shaped lid. (Hereinafter referred to as seal cap 16). The seal cap 16 is connected to a rotating means (hereinafter referred to as a rotating shaft 17).
The quartz cap 1a), the substrate holding means (hereinafter referred to as boat 4), and the substrate (hereinafter referred to as wafer 4a) held on the boat 4 are rotated. The seal cap 16 is connected to lifting means (hereinafter referred to as a boat elevator 9) so that the boat 4 is lifted and lowered. The rotary shaft 17 and the boat elevator 9 are controlled by the drive control unit 74 so as to be driven at a predetermined speed.

On the outer periphery of the reaction tube 5, heating means (hereinafter referred to as heater 1) are arranged concentrically. The heater 1 detects the temperature by temperature detection means (hereinafter referred to as heater thermocouple 2 and cascade thermocouple 3) so that the temperature in the reaction tube 5 becomes the processing temperature set by the host controller Uc. Control. Here, the heater thermocouple 2 has a role of detecting the temperature of the heater 1, and the cascade thermocouple 3 has a role of detecting the temperature between the soaking tank 6 and the reaction tank 5. Specifically, the heater 1 is divided into a plurality of zones (for example, U zone, CU zone, CL zone, L zone, etc.) in order to control the temperature in the furnace with higher accuracy. Based on the deviation between the temperature detection values and the temperature set values from the plurality of heating zones, a power control signal to the heater is output for each heating zone, and temperature control is performed. Note that both the heater thermocouple 2 and the cascade thermocouple 3 have a plurality of detection points so as to be detected at positions corresponding to the respective zones so as to correspond to the plurality of zones of the heater 1.

  Here, the cascade thermocouple 3 is installed between the reaction vessel 5 and the boat 4 so that the temperature in the reaction vessel 5 can be detected. However, the cascade thermocouple 3 and the heater thermocouple 2 are connected to each other. The arrangement is arranged between the heater 1 and the wafer 4a, the cascade thermocouple 3 is arranged closer to the wafer 4a than the heater thermocouple 2, and the heater thermocouple 2 is arranged closer to the heater 1 side than the cascade thermocouple 3. It only has to be done.

    Next, an example of an oxidation / diffusion treatment method in the treatment furnace 91 will be described. First, the boat 4 is lowered by the boat elevator 9. The boat 4 holds a plurality of wafers 4a. Next, the temperature in the reaction tube 5 is set to a predetermined temperature while being heated by the heater 1. The inside of the reaction tube 5 is filled with an inert gas in advance by the MFC 12 connected to the gas supply pipe 11, and the boat 4 is lifted and moved into the reaction tube 5 by the boat elevator 9, and the internal temperature of the reaction tube 5 is changed. Maintain a predetermined processing temperature. After maintaining the inside of the reaction tube 5 at a predetermined pressure, the boat 4 and the wafer 4 a held on the boat 4 are rotated by the rotating shaft 17. At the same time, a processing gas is supplied from the gas supply pipe 11 or water vapor is supplied from a moisture generator. The supplied gas descends the reaction tube 5 and is uniformly supplied to the wafer 4a.

  In the reaction tube 5 during the oxidation / diffusion treatment, the pressure is controlled by the APC 13 so as to reach a predetermined pressure through the exhaust pipe 8, and the oxidation / diffusion treatment is performed for a predetermined time.

    When the oxidation / diffusion process is completed in this way, the gas in the reaction tube 5 is replaced with an inert gas and the pressure is changed to normal pressure, and then the boat elevator is moved to the next oxidation / diffusion process of the wafer 4a. The boat 4 is lowered by 9 and the boat 4 and the processed wafer 4 a are taken out from the reaction tube 5. The processed wafer 4a on the boat 4 taken out from the reaction tube 5 is replaced with an unprocessed wafer 4a, and is again raised into the reaction tube 5 in the same manner as described above, and oxidation / diffusion processing is performed.

  Next, the temperature control unit 71 shown in FIG. 1 will be described. The temperature control unit 71 performs control according to each of the heater 1, the cascade thermocouple 3, and the heater thermocouple 2 for each of the heating zones described above, but in the following description, one of them is provided unless otherwise specified. The explanation is for the heating zone.

  The temperature controller 71 includes switchers 20 and 22, subtractors 21 and 25, PID calculators 23 and 26, PIDC calculator 24, and Power pattern output unit 27.

  The switcher 20 selectively switches the control method according to the set control mode. Specifically, selection switching such as PID control (described later) and Power control (described later) is performed.

    The subtractor 21 calculates the difference F by subtracting the control amount (detected temperature) A from the target value Sc set by the host controller Uc, and sends it to the PID calculator 23 or PIDC calculator 24 via the switch 22. Output.

  The switch 22 selectively switches the control method according to the set control mode. Specifically, selection switching between PIDC control (described later) and PID control is performed.

  Next, processing (PID control) in the PID computing unit 23 will be described with reference to FIG.

As shown in FIG. 3, the PID calculator 23 includes an adder 30, an integral calculator 31, a proportional calculator 32, and a differential calculator 33. The integral calculator 31 receives the deviation F, and outputs a value obtained by multiplying the deviation F by a time integral calculation (I calculation) by a preset parameter Ki as an integral value N. Assuming that the deviation F at a specific time t is F (t) and the integration value N at that time is N (t), the integration value N is obtained according to the equation (1). In equation (1), the integration range of ∫F (u) du is between 0 and t.
N (t) = Ki · ∫F (u) du (1)

The proportional calculator 32 receives the deviation F and outputs a value (P calculation) multiplied by a preset parameter Kp as a proportional value O. When the deviation F at a certain specific time t is expressed as F (t) and the proportional value O at that time is expressed as O (t), the proportional value O can be obtained according to the equation (2).
О (t) = Kp · F (t) (2)

The differential calculator 33 inputs the deviation F, and outputs as a differential value R a value obtained by multiplying the deviation F by the time differential calculation (D calculation) by a preset parameter Kd. Assuming that the deviation F at a specific time t is F (t) and the differential value R at that time is R (t), the differential value R is obtained according to the equation (3).
R (t) = Kd · dF (t) / dt (3)

The adder 30 receives the integral value N, the proportional value O, and the differential value R, calculates the sum of these, and outputs the manipulated variable X. Assuming that the deviation F at a specific time t is F (t) and the manipulated variable X at that time is represented by X (t), the manipulated variable X can be calculated from the above-mentioned formulas (1), (2), and (3). Is obtained according to Equation (4), and such a calculation process in the PID calculator 23 is referred to as a PID calculation. In equation (4), the integration range of ∫F (u) du is between 0 and t.
X (t) = Kp.F (t) + Ki.∫F (u) du + Kd.dF (t) / dt (4)

    That is, as shown in FIG. 1, the target value (target temperature) Sc from the host controller Uc and the control amount (detected temperature) A from the cascade thermocouple 3 are input to the temperature control unit 71 in the main control unit 7. The subtractor 21 in the temperature control unit 71 outputs a deviation F obtained by subtracting the control amount A from the target value (target temperature) Sc. In the PID calculator 23, the PID calculation is performed using the deviation F, and the operation amount X is determined. This manipulated variable X is converted to a target value X ′, and this target value X ′ and the control amount (detected temperature) B from the heater thermocouple 2 are input to the subtractor 25, and the subtractor 25 controls from the target value X ′. A deviation E obtained by subtracting the amount B is output. The PID calculator 26 performs a PID calculation using the deviation E, and an operation amount Z is output as an output of the temperature control unit 71 and input to the heater 1. The control amounts A and B output from the heater 1 are fed back to the temperature control unit 71 again. In this manner, the operation amount Z output from the temperature control unit 71 is changed every moment so that the deviation F between the target value Sc and the control amount A becomes zero. Such a control method is called PID control.

  Next, the PIDC calculator will be described with reference to FIG.

  As shown in FIG. 4, the PIDC computing unit 24 includes an adder 40, an integral computing unit 41, a proportional computing unit 42, a differential computing unit 43, a switching unit 44, an integration pattern output unit 45, and an integration pattern output unit 46. The

  The switch 44 performs selection switching based on a preset control switching time. Specifically, the integration pattern output unit 45 or the integration pattern output unit 46 + the integration calculation unit 41 is switched at a timing t set in advance from the start of control.

The integration calculator 41 receives the deviation F, and outputs a value obtained by multiplying the deviation F by time integration calculation (I calculation) by a preset parameter Ki as an integration value N. Assuming that the deviation F at a specific time t is F (t) and the integral value N at that time is N (t), the integral value N is obtained according to the equation (5). In equation (5), the integration range of ∫F (u) du is between 0 and t.
N (t) = Ki · ∫F (u) du (5)

Here, the integral output pattern is to set the output value for the integral calculation in advance according to the process instead of the integral calculation, and the integral pattern output units 45 and 46 are set to the preset output patterns. Based on the above, the integration pattern value J is output. FIG. 5 is a diagram illustrating an integral output pattern. The host controller Uc can set the output value C, the rate (Rate), and the time time for each of a plurality of steps (Step 1 to Step 4) when performing temperature control. Assuming that a certain step is the I step, the output value C (I-1) of the previous step is changed at the rate Rate (I) toward the output value C (I), and after reaching the output value C (I), the output value Continues output with C (I). After the start of the I step, when the time time (I) has elapsed, the process proceeds to the next step I + 1. Assume that an output value J at a specific time t is J (t).

  The adder 49 receives the integration pattern value J from the integration pattern output unit 46 and the integration calculation value N from the integration calculator 41, calculates the sum of these values, and outputs an integration operation amount W.

That is, when the deviation F at a specific time t is represented by F (t) and the integral operation amount W at that time is represented by W (t), the switching unit 44 causes the integral pattern output unit 45, the integral pattern output unit 46 + the integral computing unit 41 to be used. Is switched according to the equation (6). The integral operation amount W is the sum of the integral pattern value (J (t)) and the integral calculation value. In equation (6), the integration range of ∫F (u) du is between 0 and t.
W (t) = J (t) + Ki∫∫ (u) du (6)

The proportional calculator 42 receives the deviation F and outputs a value (P calculation) multiplied by a preset parameter Kp as a proportional value O. When the deviation F at a certain specific time t is expressed as F (t) and the proportional value O at that time is expressed as O (t), the proportional value O is obtained according to the equation (7).
О (t) = Kp · F (t) (7)
The differential calculator 43 receives the deviation F and outputs a value obtained by multiplying the deviation F by a time differential calculation (D calculation) by a preset parameter Kd as a differential value R.
Assuming that the deviation F at a specific time t is F (t) and the differential value R at that time is R (t), the differential value R is obtained according to the equation (8).
R (t) = Kd · dF (t) / dt (8)

The adder 40 receives the integral operation amount W or the integral pattern value J, the proportional value O, and the differential value R, calculates the sum of these, and outputs the operation amount X. Assuming that the deviation F at a specific time t is F (t) and the manipulated variable X at that time is represented by X (t), the manipulated variable X can be calculated from the above-mentioned formulas (6), (7), and (8). Is obtained according to equation (9a) or equation (9b), and this is called PIDC calculation.
X (t) = W (t) + Kp.F (t) + Kd.dF (t) / dt (9a)
X (t) = J (t) + Kp.F (t) + Kd.dF (t) / dt (9b)

That is, as shown in FIG. 1, when the target value Sc from the host controller Uc and the control amount A from the cascade thermocouple 3 are input to the temperature controller 71, the subtractor 21 in the temperature controller 71. Then, the deviation F obtained by subtracting the control amount A from the target value Sc is output. When the deviation F is input to the PIDC calculator 24 by the switcher 22, the PIDC calculator 24 determines the manipulated variable X using the deviation F, a preset integral pattern integral value, a proportional differential calculator, and the like. The The manipulated variable X is converted into a target value X ′, and the target value X ′ and the control amount B from the heater thermocouple 2 are input to the subtractor 25, and the subtractor 25 subtracts the control amount B from the target value X ′. Deviation E is output. The PID calculator 26 performs a PID calculation using the deviation E, and an operation amount Z is output as an output of the temperature control unit 71 and input to the heater 1. The control amount A and the control amount B output from the heater 1 are fed back to the temperature control unit 71 again. In this manner, the operation amount Z output from the temperature control unit 71 is changed every moment so that the deviation F between the target value Sc and the control amount A becomes zero. Such a control method is called PIDC control.

  Next, the power pattern output unit 27 will be described.

Here, the power pattern includes integral calculation, differential calculation, and proportional calculation. Instead of heater output values, heater output values including integration, differentiation, and proportional calculation are set according to the process. The power pattern output unit 27 outputs an operation amount Z based on a preset pattern. Specifically, for example, as shown in FIG. 6, a certain operation amount Z1 is output for a certain time t1, and after a lapse of t1, a certain operation amount Z2 is output for a certain time t2. . At this time, it is also possible to perform ramping by a preset inclination (operation amount / time) when shifting from the operation amount Z1 to the operation amount Z2.
A control method in which the operation amount is determined and output based on a pattern (patterned) created in advance by the time, the operation amount, and the inclination in this way is called power control.

  FIG. 7 is a flowchart for explaining the temperature adjustment method according to the present embodiment. Here, a case where the initial temperature S is ramped up to the target temperature S ′ to reach a stable process will be described. Here, the ramp-up means a ramp-up, that is, a temperature raising step.

  First, by using the above-described standard PID parameters (Ki, Kp, Kd) in PID control, the temperature control unit 71 controls the temperature of the heater so that the temperature detected by the cascade thermocouple becomes a predetermined target temperature (first step). Step 1), temperature characteristic basic data (temperature data detected by the cascade thermocouple) is acquired (S11a).

Next, as shown in FIG. 8, a temperature adjustment evaluation time A1 [min] from the ramp-up start (temperature rise start) (based on the temperature measured by the cascade thermocouple) is determined based on the temperature characteristic basic data. In addition, the maximum overshoot time of the temperature indicated by the cascade thermocouple 3 (time from the start of ramp-up to the maximum overshoot) a [min], target temperature S ′ arrival time b [min], target temperature S ′ A stable time c [min] and an operation amount d [%] when the target temperature is stable are obtained. Further, the manipulated variable e (t) [%] at each time point within the temperature adjustment evaluation time A1, the temperature f (t) [° C.] (not shown) indicated by the heater thermocouple 2, and the temperature indicated by the cascade thermocouple 3 g (t) [
° C].

Next, the total amount of heat B1 [% * min] during the temperature adjustment evaluation time A1 of the temperature indicated by the cascade thermocouple 3 (hereinafter also referred to as the cascade temperature) is obtained. In equation (10), the integration range of ∫e (t) dt is between 0 and A1.
B1 = ∫e (t) dt (10)

Further, it is assumed that the temperature is stable at the target temperature S ′ throughout the temperature adjustment evaluation time A1, and only the operation amount d when the target temperature S ′ is stable is output. Assuming that the total heat amount is a stable heat amount C1 [% * min], C1 is obtained by the following equation (11).
C1 = d * A1 (11)
Using B1 and C1 obtained from the equations (10) and (11), the amount of heat obtained by the following equation (12) is defined as a heating amount D1 [% * min].
D1 = B1-C1 (12)
Here, the heat-up amount will be described. The amount of heat obtained by the PID calculation from the initial temperature to the desired target temperature S ′ until it stabilizes until the target temperature S ′ is maintained at the target temperature S ′, and the target temperature is set to the set value. Two types of calories, the amount of calorie required to raise the temperature while following, are included.

  Therefore, if subtracting the stable heat amount when it is assumed that the manipulated variable when the target temperature is stabilized is continuously output from the total heat amount obtained by the equation (10), the heat amount necessary for raising the temperature, A quantity is obtained.

  As mentioned above, the temperature rises following the set value from the start of temperature rise to the target temperature, but if the amount of heat rise is too large at this time, overshoot will continue to rise even if the target temperature is exceeded. Resulting in. This overshoot is an unnecessary phenomenon in temperature control and should be eliminated as much as possible. Here, the overheat factor is hidden in the heat-up amount, not the stable heat amount in the total heat amount, and the output pattern reflecting the heat amount ratio of the overshoot factor hidden in the heat-up amount is used. Thus, the inside of the furnace can be quickly stabilized at an ideal temperature.

Next, as indicated by the hatched portion in FIG. 9, the sum of the cascade temperatures overshooted from the target temperature S ′ arrival time b to the target temperature S ′ stabilization time c from the start of evaluation is obtained. Assuming that this is the overshoot temperature sum F ′, F ′ is obtained by the following equation (13). In Expression (13), the integration range of ∫g (t) dt is between b and c.
F ′ = ∫g (t) dt−S ′ * (c−b) (13)

Next, as shown by the hatched portion in FIG. 10, the temperature sum during the target temperature S ′ stabilization time c from the start of evaluation is obtained. When this is the temperature sum G, G is obtained by the following equation (14). In Expression (14), the integration range of ∫g (t) dt is between 0 and c.
G = ∫g (t) dt−S * c (14)

Next, an overshoot temperature ratio H [percentage] is obtained. This is the ratio of the overshoot temperature sum F ′ to the temperature sum G as shown in FIG. Therefore, it is calculated | required by the following formula | equation (15).
H = F '/ G (15)

This overshoot temperature ratio H is the ratio of the amount of heat of the overshoot factor concealed in the temperature rise heat amount D1. Accordingly, by reducing the temperature rise amount D1 by the overshoot temperature ratio H and reflecting it in the temperature rise amount D1, overshoot can be suppressed and the interior of the furnace can be quickly stabilized at an ideal temperature. Assuming that the heat-up heat amount after reducing the overshoot temperature ratio H is D1 ′, D1 ′ is obtained by the following equation (16).
D1 '= D1 * (1-H) (16)

Then, a power output pattern (corresponding to the first output control pattern) is obtained by using a, d, and D1 ′ obtained as described above (second step), and power control is performed (third process). Process). That is, based on the detected temperature detected by the cascade thermocouple in the heater control in the first step, the operation amount (first operation amount) for controlling the heater is patterned (see the operation amount shown in FIG. 11). ), A first output control pattern is obtained, and the heater is controlled based on the first output control pattern. That is, based on the temperature data detected by the cascade thermocouple, the first operation amount, which is the operation amount that defines the POWER output pattern, is patterned to obtain the POWER output pattern.

As the Power output pattern, the pattern shown in FIG. 11 is obtained (S11b). If the temperature rising ramp rate (° C./min) is h, each parameter in FIG. 11 is obtained by the following equations (17) to (20). The ramp rate here means the speed (gradient) of temperature change. For example, the ramp rate when raising the temperature from 100 ° C. to 200 ° C. in 5 minutes is (200 ° C.-100 ° C.) / 5. Min = 20 ° C./min.
E0 = (D1 ′ / T1) + d (17)
E1 = d (18)
T1 = (S′−S) / h (19)
T2 = a−T1 (20)
Here, E0 is the amount of heat rise D1 ′ as a constant output, and D1 ′ is divided by the time T1 to reach the target temperature S ′ from the initial temperature S, and the manipulated variable d (%) when the target temperature is stable. (Ie, the manipulated variable at T1). S indicates the initial temperature.

  After T1, only the manipulated variable d when the target temperature is stable is output until the time (maximum overshoot time) a (min) for reaching the maximum cascade temperature is reached.

  By using this Power output pattern, it is possible to obtain a temperature waveform indicated by the heater thermocouple 2 that can follow the ramp rate h and can be stabilized quickly. This is acquired as power control temperature characteristic data (S12).

  Subsequently, the overshoot amount of the basic temperature characteristic data during power control is determined (S13). When the overshoot amount is larger than a preset overshoot allowable value (S13, Yes), the above-described equations (13) to (13) Using (20), the power output pattern is corrected from the ratio between the entire temperature waveform and the overshooted portion, and adjustment is performed to reduce the overshoot (S14). Then, power control is performed again using the power output pattern thus corrected (third step), and temperature characteristic data at the time of control is acquired (S12).

  Next, when the overshoot amount is not larger than the preset overshoot allowable value (S13, No), as shown in FIG. 12, the temperature transition with respect to the cascade thermocouple during the PID control acquired in the first step (Temperature waveform) The temperature adjustment evaluation time J1 [min] from the start of Power control is determined from the power control temperature characteristic basic data acquired in the third step, and the maximum temperature indicated by the heater thermocouple 2 is determined. Time j [min] and time n [min] at which the temperature indicated by the heater thermocouple 2 is sufficiently stable are obtained. Here, the temperature indicated by the heater thermocouple 2 during power control at each time point within the temperature adjustment evaluation time J1 (hereinafter also referred to as a heater waveform) is k (t) [° C.], and the temperature indicated by the cascade thermocouple 3 is Let m (t) [° C.].

  From these data, as shown in FIG. 13, based on the heater waveform L (t) obtained by subtracting the amount of P operation from the waveform of the temperature k (t) indicated by the heater thermocouple 2 during power control as shown in FIG. An integral output pattern (corresponding to the second output control pattern) M (t) is obtained (fourth step) (S15). That is, based on the heater waveform L (t), at least a part of a second operation amount (refer to the operation amount shown in FIG. 5) that is an operation amount that defines an integral output pattern is patterned to obtain an integral output pattern. .

  Here, the reason for subtracting the P (proportional) operation from the waveform of the temperature k (t) indicated by the heater thermocouple 2 is that a power output pattern is created for k (t) based on the output calculated by PID calculation during PID control. Therefore, this waveform is based on the sum of P (proportional) output, I (integral) output, and D (differential) output.

  Therefore, a desired integral output pattern M (t) can be obtained by subtracting the P output and the D output. Here, in general, the D output is negligible when the temperature rises, so the D output can be ignored.

Next, by subtracting k (0) · j from the heater temperature waveform L (t) obtained by subtracting the P operation, the amount of heat for the time from the start of evaluation to the maximum heater temperature time j indicated by the heater waveform k (t) is obtained. . Let Q be the case. In equation (21), the integration range of ∫L (t) dt is between 0 and j.
Q = ∫L (t) dt−k (0) · j (21)
By dividing this Q by the time until the maximum heater temperature time j, a value M having the same area (heat amount) as Q and a constant heater temperature can be obtained from the start of evaluation to the time j.
(Area (heat quantity) = Q area (heat quantity) of base j, height M)
M = Q / j (22)
When this M is doubled, a triangle with a base j height M * 2 equal to the area (heat amount) of Q can be obtained. By using the height and inclination of this triangle and k (0) · j, the same amount of heat as the amount of heat due to the temperature indicated by the heater thermocouple (hereinafter also referred to as the heater thermocouple temperature) excluding the P operation indicated by the heater thermocouple. Is an appropriate integral output pattern.

  As shown in FIG. 13, step 1 is performed between the start of evaluation and the maximum temperature point j indicated by the heater thermocouple during power control. From the maximum heater temperature point j during power control (shown by the heater thermocouple) to the maximum cascade temperature during PID control. Let step 2 be during time a.

  Step 1 temperature is an appropriate integrated output pattern from the start of evaluation to the maximum temperature point j indicated by the heater thermocouple during power control if the height and inclination of the triangle are applied. In other words, the heater thermocouple maximum time during power control is the time when there is no influence on the output of the heater thermocouple, that is, the temperature detected by the heater thermocouple due to the amount of temperature rise. It is appropriate to create an integral output pattern.

  After reaching the maximum temperature indicated by the cascade thermocouple during PID control, which is the end time of step 2, the temperature detected by the cascade thermocouple is not affected by the fluctuation of the heater heating. That is, if the cascade temperature reaches the target temperature S ′ at this point, the target temperature S ′ is stabilized without overshooting thereafter. That is, when the maximum temperature indicated by the cascade thermocouple is reached, the output for raising the temperature by heating the heater, that is, the temperature at which the temperature detected by the cascade thermocouple is no longer affected by the amount of temperature rise is increased. It is appropriate to create an integral output pattern.

  Therefore, if the integrated output pattern is also patterned so that the output at the time of stabilization when the temperature detected by the cascade thermocouple at the time of PID control becomes maximum, the integrated output pattern is the appropriate integrated output pattern of Step 2, and the Step 1 target The pattern may be linearly patterned from the temperature V to the heater temperature k (n) of the time n when the heater thermocouple is sufficiently stable. Note that the temperature detected by the cascade thermocouple during power control may be maximized, not during PID control.

Parameters L (t) [° C.] (heater temperature obtained by subtracting P), V [° C.] (STEP 1 target temperature in FIG. 13), T [° C./min] (STEP 1 slope in FIG. 13) ) And U [° C./min] (STEP2 slope in FIG. 13) are obtained by the following equations (23) to (26).
L (t) [° C]
= (Heater waveform-P operation output)
= K (t)-[P constant Kp * (target temperature S '-cascade temperature during POWER control m (t))]
... (23)
From formula (21) and formula (22)
V [° C.] = (∫L (t) dt / j) × 2 (24)
T [° C./min]=(V−k(0))/j (25)
U [° C./min]=(k(n)−V)/(a−j) (26)
An integral output pattern is created from the above equations and PIDC control is performed (S15). In Expression (24), the integration range of ∫L (t) dt is between 0 and j.

  Specifically, as shown in FIG. 13, an initial value, a target value, a rate, and a time are set for each step and output along the step. Further, the integral output pattern can be created using more steps than the above-described two steps in the transition period.

As shown in FIG. 14, for example, when r integration pattern output steps are set every certain time q [min] in a temperature transition period, the target temperature Vr and ramp rate (tilt) of each step STEP (r) are set. Yr is obtained by the following equations (27) and (28).
Vr [° C.] = (∫L (r * q) dt / q) * 2 (27)
Yr [° C./min]=(V(r−1)−V(r))/q (28)

  By creating the integral output pattern as described above, it is possible to prevent the integral output from increasing during the temperature transition period, and to stabilize the interior of the furnace at the target temperature S ′ as short as possible by using the power control waveform as a base. be able to.

  Furthermore, the heat treatment apparatus according to the present embodiment is used when overshoot and undershoot are large even when PIDC control is performed with the above-described integral output pattern, or when deviation between heating zones needs to be greatly improved. It has a function to adjust. For example, by using the integral output pattern, an excessive integral calculation value is excluded, but on the other hand, due to the influence, the temperature raising timing is delayed, and the proportional calculation value or the like may be excessively increased. Therefore, there are cases where these need to be adjusted.

  The overshoot / undershoot improvement adjustment will be described below. As described above, in the present invention, the integrated output in the temperature transition period is patterned, and in the stable temperature period, the integrated output pattern is created so that the integrated output amount becomes that in the stable state, and the temperature is adjusted in addition to the integrated calculation output. ing. Here, during the temperature stabilization period, the integral calculation output is used because the adverse effect of a small temperature change such as unpredictable disturbance tends to be more noticeable than when the temperature rises. This is because the incompatibility cannot be dealt with by using the integral calculation output.

  That is, for example, in the temperature control shown in FIG. 13, if the deviation from the target temperature S ′ to the temperature indicated by the cascade thermocouple is set to 0 at the time of switching between Step 2 and Step 3 from Step 3, The temperature then stabilizes and this is considered to be an appropriate adjustment.

  Therefore, when an overshoot or undershoot occurs (S16, Yes), an integrated output is obtained from the deviation between the target temperature S ′ at the switching point of Step 2 and Step 3 (the PID control switching point) and the temperature indicated by the cascade thermocouple. The pattern is corrected and adjusted to reduce overshoot or undershoot (S17a).

As shown in FIG. 15, when the difference between the temperature indicated by the cascade thermocouple at the switching point between step 2 and step 3 and the target temperature S ′ is p, overshoot, It is considered that the undershoot occurs because of the influence of the P output in the transition period. The P output W1 calculated from the deviation p at the switching point between step 2 and step 3 is obtained by equation (29) as in equation (7).
W1 = Kp · p (29)
That is, it is considered that the P output W1 is an extra output and appears as a deviation p at the switching point between Step 1 and Step 2.

  Accordingly, it is considered that overshoot and undershoot improvement can be realized by adjusting this extra P output W1 with an integral output pattern.

Here, the temperature transition period is step 1 to step 2, and the adjustment amount W1 is subtracted from the step 1 target temperature V to recalculate the step 1 inclination T and the step 2 inclination U. -The operation amount of step 2 can be adjusted to an appropriate value.
That is, if undershooting is performed, the integral operation amount in the temperature transition period is increased, and if overshooting is performed, the integral operation amount in the temperature transition period is decreased, so that the target temperature S ′ at the switching point between step 2 and step 3 is reduced. And the temperature indicated by the cascade thermocouple can be reduced, and overshoot and undershoot can be improved.

Assuming that the adjustment amount W1, the integrated output initial temperature k (0), the step 1 target temperature V, the time j of step 1, the time (a-j) of step 2, and the heater stable temperature (step 3 target temperature) k (n) The step 1 target temperature V ′, the step 1 ramp rate T ′, and the step 2 ramp rate U ′ after adjustment are obtained by the following equations (30) to (32).
V ′ = V−W1 (30)
T ′ [° C./min]=(V′−k(0))/j (31)
U ′ [° C./min]=(k(n)−V′)/(a−j) (32)

  As shown in Fig. 16, even if the temperature adjustment is appropriate, the operation amount is created and adjusted for each heating zone, so there are early and late zones until the target temperature stabilizes, and this is an inter-zone deviation. May affect the film thickness.

  Therefore, when the inter-zone deviation is larger than the preset allowable value (S18, Yes), the zone with fast target temperature stabilization is slowed down in accordance with the slow zone by adjusting the integral output pattern, and the inter-zone deviation is improved. (S19).

  The time deviation dev_t [min] at the maximum time of the maximum inter-zone deviation between a zone 901 with a certain target temperature stability is early and a zone 902 with a slow target temperature stability is obtained.

  This is to obtain the temperature B ′ of the zone 902 at the time when the inter-zone deviation is maximum, and the difference between the time when the zone 901 becomes the temperature B ′ and the maximum time between the zones is obtained as the time deviation dev_t.

  Next, as shown in FIG. 17, the target temperature stabilization is delayed according to the zone 902 by delaying each step of the integrated output pattern of the zone 901 by dev_t.

Each parameter shown in FIG. 17 is obtained by the following equations (33) and (34).
Here, T is the gradient T [° C./min] in Step 1 described above.
a ′ = a + dev_t (33)
k (0) ′ = k (0) + T · dev_t (34)
As described above, by delaying the integral output pattern by the time deviation dev_t, it is possible to delay the zone where the target temperature stability is fast.

  On the other hand, if the inter-zone deviation is not larger than the preset allowable value (S18, No), the temperature adjustment process is terminated.

  Thus, in the semiconductor device manufacturing method according to the present embodiment, the substrate is heat-treated by performing temperature control using the second output control pattern described above as at least part of the operation amount of the heater (fifth). Process).

  Specifically, based on the integral output pattern obtained above, the host controller Uc sets the integral output pattern for each heating zone, and sets the output value, rate, and time for each of a plurality of steps. Set the time. Note that the switching time of the switch 44 is set to a time for shifting to the stable state after the temperature rise is completed (for example, a time for shifting from step 2 to step 3 in FIG. 17).

After the setting, temperature control is performed using the PIDC calculator 24, the PID calculator 26, etc., and the substrate is processed. Specifically, the temperature control unit 71 performs the cascade thermocouple 3 with respect to the target value Sc (for example, Step 2 target temperature k (n) in FIG. 17) at the time of temperature increase (for example, Step 1 and Step 2 in FIG. 17). On the basis of the result F obtained by subtracting the control amount A from the subtractor 21, the operation amount (second operation amount) X is calculated by the proportional operation unit 42, the differentiation operation unit 43, and the integration pattern output unit 45. Then, the manipulated variable X is converted into the target value X ′, and the control amount B from the heater thermocouple 2 is subtracted by the subtractor 25 to output a deviation (first deviation) E. Thereafter, the deviation E is calculated by the PID calculator 26.
Is used to calculate the PID, and the operation amount Z is output as the output of the temperature control unit 71 to control the heater 1.

The control amounts A and B output from the heater 1 are returned to the temperature control unit 71 again. In this way, the operation amount Z is controlled so that the deviation F between the target value Sc and the control amount A becomes zero. When the temperature rise is stable (for example, step 3 in FIG. 17), the temperature is switched by the switch 44 after the temperature rise is finished, and the control amount A from the cascade thermocouple 3 is subtracted from the target value Sc. The operation amount (third operation amount) X is calculated by the proportional operation unit 42, the differential operation unit 43, the integration pattern output unit 46, and the integration operation unit 41, and then the operation amount is subtracted by the subtractor 21. X is converted to the target value X ′, and the control amount B from the heater thermocouple 2 is subtracted by the subtractor 25 to obtain the deviation (
The second deviation) E is output. Thereafter, the PID calculator 26 performs a PID calculation using the deviation E, and outputs an operation amount Z as an output of the temperature control unit 71 to control the heater. The control amounts A and B output from the heater 1 are returned to the temperature control unit 71 again. Also at this time, the operation amount Z is controlled so that the deviation F between the target value Sc and the control amount A becomes zero. While performing such temperature control, the substrate is heat-treated.

  FIG. 18, FIG. 19, and FIG. 20 show temperature changes in the furnace when the temperature control according to the present embodiment is performed.

18 and 19 are specific examples at the time of ramp-up. At the time of ramp-up by PID control which is the conventional method shown in FIG. 18, it takes 35 minutes until the overshoot is stabilized at about 40 ° C. and the target temperature ± 3 ° C. At the time of ramp-up by PIDC control in which the temperature is adjusted based on the procedure of the present invention (FIG. 12) shown in FIG. 19, it takes about 17 minutes until the overshoot stabilizes within about 2 ° C. and within the target temperature ± 3 ° C. Thus, it is clear that the present invention is effective during ramp-up.

FIG. 20 shows an improvement result of the inter-zone deviation in the ramp-up.
There is an inter-zone deviation of about 50 ° C. before adjusting the deviation between zones. As a result of adjusting between zones based on the procedure of the present invention, the deviation between zones could be reduced within about 20 ° C.

  From FIG. 20, it is clear that the inter-zone deviation adjustment procedure of the present invention is effective in reducing the inter-zone deviation.

(Second Embodiment)
Next, a second embodiment of the present invention will be described.

  The heat treatment apparatus according to the present embodiment has the same configuration as the heat treatment apparatus according to the first embodiment described above. The present embodiment and the first embodiment are different in the process when an overshoot or undershoot occurs in the procedure of the temperature adjustment method (S16, Yes). Hereinafter, in the present embodiment, the same components as those in the first embodiment and the processing contents are denoted by the same reference numerals, and description thereof is omitted. FIG. 21 is a flowchart for explaining a temperature adjustment procedure in the present embodiment.

In the first embodiment described above, the integral output in the temperature transition period is patterned, and in the temperature stable period, the integral output amount is created to be that in the stable state, and the temperature is adjusted.
That is, Step 1 to Step 2 are the temperature transition period, and Step 3 is the stable period.

  Here, the second embodiment is adjusted using the overshoot temperature ratio instead of the adjustment using the temperature adjustment of S17a.

  Also in the second embodiment, the overshoot temperature ratio H is obtained on the basis of the PIDC control result of S15 using the above-described equations (13) to (15). The calculated overshoot temperature ratio H is the heat ratio of the overshoot factor concealed in the integral output pattern of step 1 to step 2. Accordingly, by reducing the step 1 target temperature V by the overshoot temperature ratio H, it is reflected in the integral output pattern in the temperature transition period, and the inside of the furnace can be quickly stabilized at an ideal temperature (S17b).

If the initial temperature k (0), step 1 target temperature V, step 1 time j, step 2 time (aj), heater stable temperature (step 3 target temperature) k (n), the overshoot temperature ratio H is reduced. The subsequent step 1 target temperature V ′, step 1 ramp rate T ′, and ramp rate U ′ of step 2 are obtained by the following equations (35) to (37).
V ′ = V * (1−H) (35)
T ′ [° C./min]=(V′−k(0))/j (36)
U ′ [° C./min]=(k(n)−V′)/(a−j) (37)
By executing the above procedure, appropriate temperature adjustment can be performed quickly and surely, even if it is not a skilled worker (S17b).

(Third embodiment)
Subsequently, a third embodiment of the present invention will be described.

  The heat treatment apparatus according to the present embodiment has the same configuration as the heat treatment apparatus according to the first embodiment described above. The present embodiment and the first embodiment are different in the heat treatment process. In the first embodiment, the process is mainly from the ramp-up to the time when the target temperature is stabilized, and the third embodiment. Then, it is mainly a boat up recovery process. The boat up recovery process means that when the boat filled with wafers to be heat-treated next time is lifted by the boat elevator and carried into the reaction tube, it is caused by the thermal effect of the boat and wafer at room temperature (about 25 ° C.). Although the temperature in the reaction tube (for example, 200 ° C.) is reduced, the reduced temperature in the reaction tube is repaired, and the wafer and the boat are transferred into the reaction tube to the original temperature (in this case, 200 ° C.). This refers to the process of returning the temperature. The processes from the calculation process of the temperature rise amount D to the improvement adjustment of the deviation between zones (step S19 in the first embodiment) are different. Hereinafter, in the present embodiment, the same components as those in the first embodiment and the processing contents are denoted by the same reference numerals, and description thereof is omitted.

  First, as shown in FIG. 22, the temperature adjustment evaluation time A1 [min] from the ramp-up start is determined from the temperature characteristic basic data obtained by the PID control, and the maximum overshoot of the temperature indicated by the cascade thermocouple 3 is determined. Time a [min], initial temperature S return time b [min], initial temperature S stabilization time c [min], and manipulated variable d [%] when initial temperature is stable are obtained. Further, the manipulated variable e (t) [%] at each time point within the temperature adjustment evaluation time A1, the temperature f (t) [° C.] indicated by the heater thermocouple 2, and the temperature g (t) [shown by the cascade thermocouple 3] ° C]. Further, s [min] is a time until g (t) starts to decrease.

Assuming that the amount of stable heat that is not output during overshoot is the missing heat amount Y [% * min], Y is obtained by equation (38). In equation (38), the integration range of ∫ (d−e (t)) dt is between b and c.
Y = ∫ (d−e (t)) dt (38)
The formula (10) and formula (11) shown in the first embodiment are applied to the temperature characteristic basic data shown in FIG. 22, and the total heat quantity B1 and stable heat quantity obtained from the formula (38) are obtained. Using C1 and the missing heat amount Y, the heat amount obtained by the following equation (39) is defined as the temperature rise heat amount D1 [% * min].
D1 = B1-C1 + Y (39)
Here, the temperature rising heat amount and the missing heat amount will be described. It is considered that the total amount of heat obtained by the PID calculation until the temperature is stabilized at a certain initial temperature S after the boat is up includes the amount of heat necessary for stably maintaining the initial temperature S.

  However, because of the follow-up control, the temperature drops to some extent from the initial temperature S due to boat-up, and after the deviation between the initial temperature and the actual measurement value increases, the output of the PID calculation increases in order to stably maintain the initial temperature S. .

  Here, assuming that the minimum cascade temperature S1 when lowered by boat-up and the time from the start of evaluation to S1 is r1, boat-up recovery is the initial temperature from the lowest cascade temperature S1 when lowered by boat-up after elapse of r1 from the start of evaluation. It can be thought of as a ramp up to S.

  When considering the ramp-up from S1 to S, the amount of heat necessary for stably maintaining the initial temperature S described above is the amount of heat required for raising the temperature from the lowest cascade temperature S1 reduced by the boat-up to the initial temperature S and the initial amount. It is divided into the amount of heat for stably holding at the temperature S.

  Therefore, as the amount of heat for stably maintaining the initial temperature S from the total amount of heat obtained by applying Equation (10) as described above, the manipulated variable when the cascade temperature is stabilized after the boat is up is output. By subtracting the assumed stable heat amount, the amount of heat necessary to raise the temperature and the raised heat amount can be obtained.

  However, in the case of boat up recovery at a low temperature (for example, a temperature range of 50 ° C. to 200 ° C.), the time (c−b) until the target temperature is stabilized after reaching the target temperature after the cascade temperature overshoot becomes too large. End up. Assuming that the temperature adjustment evaluation time A1 is sufficient until the cascade temperature is stabilized, the temperature adjustment evaluation time A1 is also increased in the same manner as in (c−b), and the stable heat quantity obtained by applying the equation (11) is expressed by the equation ( 10), it becomes larger than the total amount of heat obtained by applying 10), and the heat-up amount cannot be obtained correctly.

  Therefore, during the overshooting (c−b) as shown in the equation (38), the heat amount of the difference between the stable operation amount and the actual measurement value of the operation amount was obtained by applying the equation (10). By adding to the total amount of heat, the missing heat amount, which is the amount of stable heat lost between (c−b), can be corrected, and the correct heat-up heat amount can be obtained.

  On the other hand, if the amount of heat generated by heating is too large, an attempt is made to stably maintain the initial temperature S, but overshooting is considered. In other words, it is considered that the inside of the furnace can be quickly stabilized at an ideal temperature by using an output pattern that reflects the heat amount ratio of the overshoot factor hidden in the temperature rising heat amount.

Next, as shown by the shaded portion in FIG. 23, the sum of the cascade temperatures overshooted from the initial temperature S arrival time b to the initial temperature S stabilization time c is obtained. Assuming that this is the overshoot temperature sum F ″, F ″ is obtained by the following equation (40). In equation (40), the integration range of ∫g (t) dt is between b and c.
F ″ = ∫g (t) dt−S (c−b) (40)
Next, as indicated by the hatched portion in FIG. 24, the sum of temperatures from the lowest temperature S1 time point r1 to the initial temperature S stabilization time c is obtained. If this is the temperature sum G, G is obtained by the following equation (41). In equation (41), the integration range of ∫g (t) dt is between r1 and c.
G = ∫g (t) dt−S1 (c−r1) (41)
Next, the temperature rise D1 ′ after the reduction of the overshoot temperature ratio H and the overshoot temperature ratio H is obtained. Note that the calculation of the overshoot temperature ratio H and the heat-up amount D1 ′ is the same as that in the first embodiment described above, and a description thereof will be omitted.

From the above, the power output pattern shown in FIG. 25 is obtained and the power control is performed. Each parameter in FIG. 25 is calculated | required by the following formula | equation (42)-Formula (46).
E0 = (D1 ′ / T1) + d (42)
E1 = d (43)
T0 = s (44)
T1 = (r1-s) (45)
T2 = a−T1−T0 (46)
Subsequently, T0 will be described. At the time of boating up, the influence is greatly different for each heating zone, especially in the L zone and the U zone. In the case of the L zone, when the boat is up, the normal temperature wafers, the boat, etc. are always on the entrance side where the boat rises one by one, so the temperature drop due to the boat up is severe. On the other hand, in the U zone, since the wafers, boats, and the like are warmed to some extent while boating up between the L and CU zones, the temperature drop is relatively slow compared to the L zone. Therefore, if an appropriate temperature rise is added to the stable operation amount in each of the L zone, CL zone, CU zone, and U zone simultaneously with the start of boat up (start of evaluation), the U zone is particularly affected by boat up, that is, U The temperature rises before the cascade temperature generated when wafers, boats, etc. arrive near the zone, and wasteful heat (for example, overshoot) occurs.

  Therefore, it is set in advance so that the heat-up amount of heat is not output until the cascade temperature starts to decrease due to the boat-up effect in each zone, for example, until the temperature decreases by 1 ° C. in one minute.

  This makes it possible to make a difference in the time for starting the output corresponding to the temperature rise in each zone, and it is possible to adjust the temperature appropriately for each zone.

  E0 is obtained by dividing the heating temperature D1 'by the time T1 so as to output at a constant output amount, and adding the operation amount d when the initial temperature is stable. Thereafter, only the manipulated variable d when the target temperature is stable is output until the maximum overshoot time a is reached.

  By using this Power output pattern, a temperature waveform indicated by the heater thermocouple 2 that can quickly stabilize the interior of the furnace at the initial temperature S can be obtained. This is acquired as power control temperature characteristic data.

  Next, as shown in FIG. 26, the temperature adjustment evaluation time J1 [min] from the start of the power control is determined from the power control temperature characteristic basic data, and the maximum temperature time j indicated by the heater thermocouple 2 for each zone. [Min], a time n [min] during which the temperature indicated by the heater thermocouple 2 is sufficiently stabilized is obtained. Further, the temperature k (t) [° C.] indicated by the heater thermocouple 2 and the temperature m (t) [° C.] indicated by the cascade thermocouple 3 at each time point within the temperature adjustment evaluation time J1 are set.

  From these data, an integrated output pattern M (t) as shown in FIG. 27 is obtained based on the heater waveform L (t) obtained by subtracting the P operation from the waveform of k (t) as shown in FIG. Here, the reason for subtracting the P operation from the waveform of the temperature k (t) indicated by the heater thermocouple 2 is that k (t) is obtained by creating a Power output pattern based on the output calculated by the PID calculation during PID control. Therefore, this waveform is due to the sum of P (proportional) output, I (integral) output, and D (differential) output. Therefore, a desired integral output pattern M (t) can be obtained by subtracting the P output and the D output. Here, when the temperature is generally raised, the D output is negligible, so the D output can be ignored.

Next, the area (heat amount) from the output start point s of the heating temperature amount L (t) subtracted by the P operation to the maximum heater temperature point j (actual value not subtracted by P) is obtained. Let Q be the case. In Expression (47), the integration range of ∫L (t) dt is between s and j.
Q = ∫L (t) dt−k (0) ・ (j−s) (47)
By dividing this Q by j and the time until the maximum heater temperature time point, a value M having the same area (heat quantity) as Q and the heater temperature from the time point s to the time point j can be obtained (base (j−s ), Rectangular area (heat quantity) of height M = Q area (heat quantity)).
M = Q / (j−s) (48)
When M is doubled, a triangle having a base (j−s) equal to the area (heat amount) of Q and a height M * 2 can be obtained. By using the height and inclination of this triangle and k · (j−s), the amount of heat is the same as the amount of heat due to the temperature indicated by the heater thermocouple excluding the amount of P operation indicated by the heater thermocouple.

  As shown in FIG. 27, step 0 from the start of evaluation to the output start point s of the amount of heat generated during heating, step 1 from the point s to the maximum heater temperature point j during power control, and PID control from the maximum heater temperature point j during power control. Step 2 is performed during the maximum cascade temperature time a.

  Since the step 0 temperature is a period in which there is still no influence from the boat up, the temperature is stabilized at the temperature at the start of evaluation.

  Step 1 temperature is considered to be an appropriate integrated output pattern from the start of evaluation to the maximum heater temperature point j indicated by the heater thermocouple during power control, if the above-described triangle height and inclination are applied.

  It is considered that the maximum cascade time at the time of PID control, which is the end time of step 2, has no influence on the cascade temperature after this time, such as disturbance in the temperature transition period. That is, if the cascade temperature reaches the initial temperature S at this time, it is considered that the cascade temperature is stabilized at the initial temperature S without overshooting thereafter.

Parameters L (t) [° C.] (heater temperature subtracted by P), V [° C.] (STEP 1 target temperature), T [° C./min] (STEP 1 slope) and U [° C./min] in FIGS. (STEP2 slope) is obtained by the following equations (49) to (52).
L (t) [° C.] = (Heater waveform−P operation output) = k (t) − [P constant Kp * (cascade temperature m (t) at initial temperature S−POWER control)] (49 )
V [° C.] = [∫L (t) dt / (j−s)] × 2 (50)
T [° C./min]=(V−k(0))/(j−s) (51)
U [° C./min]=(k(n)−V)/(a−j) (52)
An integral output pattern is created from the above equations and PIDC control is performed. In Expression (50), the integration range of ∫L (t) dt is between s and j. Specifically, as shown in FIG. 27, an initial value, a target value, a rate, and a time are set for each step and output along the step. In addition, the integrated output pattern may be created using more steps than the above-described three steps in the transition period.

As shown in FIG. 28, for example, when w integration pattern output steps are set every certain time q [min] in a temperature transition period, the target temperature Vw and ramp rate (slope) Yw of each step STEP (w) are It calculates | requires by the following formula | equation (53) and Formula (54).
Vw [° C.] = (∫L (w * q) dt / q) × 2 (53)
Yw [° C./min]=(V(w−1)−V(w))/q (54)
By creating the integral output pattern as described above, it is possible to prevent the integral output from increasing during the temperature transition period, and to stabilize the interior of the furnace to the initial temperature S in the shortest time by using the power control waveform as a base. Can do.

  Further, the present invention has a function of adjusting these when the overshoot and undershoot are large even when the PIDC control is performed with the integral output pattern described above, or when the deviation between zones needs to be greatly improved.

  The overshoot / undershoot improvement adjustment will be described below.

  As described above, in the present invention, the integral output in the temperature transition period is patterned, and in the temperature stable period, the integral output pattern is created so that the integral output amount becomes that in the stable state, and the temperature is adjusted.

  That is, in the temperature control method shown in FIG. 27, for example, the stable period starts from Step 3, and if the deviation between the initial temperature S and the temperature indicated by the cascade thermocouple is set to 0 at the time of switching between Step 2 and Step 3, The temperature will stabilize and this is considered to be an appropriate adjustment.

  Therefore, when overshoot or undershoot occurs, the adjustment amount may be obtained from the deviation between the initial temperature S at the switching point between step 2 and step 3 and the temperature indicated by the cascade thermocouple.

As shown in FIG. 29, when the difference between the temperature indicated by the cascade thermocouple at the switching point between step 2 and step 3 and the initial temperature S is p, overshoot and undershoot are possible although the integral output pattern is appropriate. The occurrence of the shoot is considered to be greatly influenced by the P output in the transition period. The P output W1 calculated from the deviation p at the switching point between step 2 and step 3 is obtained by equation (55) as in equation (7).
W1 = Kp · p (55)
That is, it is considered that the P output W1 is an extra output and appears as a deviation p at the switching point between Step 1 and Step 2. Therefore, it is considered that improvement of overshoot and undershoot can be realized by adjusting this extra P output W1 with an integral output pattern.

  Here, the temperature transition period is step 1 to step 2, and the adjustment amount W is subtracted from the step 1 target temperature V to recalculate the step 1 inclination T and the step 2 inclination U. -The operation amount of step 2 can be adjusted to an appropriate value. In other words, if the undershooting is performed, the integral operation amount in the temperature transition period is increased, and if overshooting is performed, the integral operation amount in the temperature transition period is decreased, so that the initial temperature S at the switching point between step 2 and step 3 is Deviation from the temperature indicated by the cascade thermocouple can be reduced, and overshoot and undershoot can be improved.

Adjustment amount W1, integrated output initial temperature k (0), step 0 time s, step 1 target temperature V, step 1 time (j-s), step 2 time (aj), heater stable temperature (step 3 target temperature) ) K (n), the adjusted step 1 target temperature V ′, step 1 ramp rate T ′, and step 2 ramp rate U ′ are obtained by the following equations (56) to (58).
V ′ = V−W | (56)
T ′ [° C./min]=(V′−k(0))/(j−s) (57)
U ′ [° C./min]=(k(n)−V′)/(a−j) (58)
Next, improvement adjustment of the inter-zone deviation will be described. As shown in Fig. 30, even if the temperature adjustment is appropriate, the operation amount is created and adjusted for each heating zone, so there are early and late zones until the target temperature stabilizes, and this is an inter-zone deviation. May affect the film thickness. Therefore, by adjusting the integral output pattern, the zone where the target temperature stability is fast is slowed down according to the slow zone, and the inter-zone deviation is improved.

  The time deviation dev_t [min] at the maximum time of the maximum inter-zone deviation between the zone 903 with a certain target temperature stability early and the zone 904 with a slow target temperature is obtained. This obtains the temperature B ′ of the zone 904 at the maximum time of deviation between zones, and the difference between the time when the zone 903 reaches the temperature B ′ and the maximum time of deviation between zones is obtained as a time deviation dev_t.

  Next, as shown in FIG. 31, the target temperature stabilization is delayed according to the zone 904 by delaying each step of the integrated output pattern of the zone 903 by dev_t. Each parameter shown in FIG. 31 is obtained by Expression (33) to Expression (34) shown in the first embodiment. Here, T is the gradient T [° C./min] in Step 1 described above.

  As described above, by delaying the integral output pattern by the time deviation dev_t, it is possible to delay the zone where the target temperature stability is fast. The above is the description of the temperature adjustment method in the present embodiment.

  By executing the above procedure, appropriate temperature adjustment can be performed quickly and surely even without a skilled worker. Specific behavior in the furnace is shown in FIGS.

  FIG. 32 and FIG. 33 are specific examples of temperature recovery after boat up is completed. In the boat up recovery by the PID control which is the conventional method shown in FIG. 32, the overshoot is about 10 ° C. and is not stabilized within the target temperature ± 5 ° C. even about 100 minutes after the boat up.

  In the boat up recovery by PIDC control in which the temperature is adjusted based on the procedure of the present invention shown in FIG. 33, the overshoot is about 3 ° C., and is stabilized within the target temperature ± 5 ° C. about 15 minutes after the boat up is completed. It is clear from FIGS. 32 and 33 that the present invention is effective in boat up recovery.

  According to the present invention, even when a relatively large disturbance occurs in the furnace to be controlled, the integrated output pattern that can cause the control amount to follow the target value most quickly from the sum of the operation amounts including the disturbance. Is set in advance, the integrated value is output as a pattern instead of the integral operation from a specific time, and the integral operation is performed again from the point in time when most of the deviation due to the disturbance or the like seems to disappear. Therefore, the control amount output from the controlled object can be quickly and accurately recovered to the initial value.

  In addition, the procedure of the present invention is a natural phenomenon, the actual temperature, operation amount, time, etc., so far most of the relying on the experience and intuition of skilled workers at the time of temperature adjustment can be obtained by a certain calculation formula, The temperature can be adjusted quickly and reliably, and the time and cost can be reduced.

  In addition, by programming the procedure of the present invention and incorporating it as software in a temperature controller or the like, it is possible to perform appropriate temperature adjustment without requiring operator intervention.

(Fourth embodiment)
Subsequently, a fourth embodiment of the present invention will be described.

  The heat treatment apparatus according to the present embodiment has the same configuration as the heat treatment apparatus according to the third embodiment described above. The present embodiment and the third embodiment are different in the process when an overshoot or undershoot occurs in the procedure of the temperature adjustment method. Hereinafter, in the present embodiment, the same components as those in the third embodiment and the processing contents are denoted by the same reference numerals, and description thereof is omitted.

In the third embodiment described above, the integral output in the temperature transition period is patterned, and in the temperature stable period, the integral output amount is made to be that in the stable state, and the temperature is adjusted.
That is, Step 1 to Step 2 are the temperature transition period, and Step 3 is the stable period.

  That is, overshoot and undershoot adjustment during PIDC control in the third embodiment is adjusted using an overshoot temperature ratio instead of adjustment using temperature adjustment.

  Also in the third embodiment, the overshoot temperature ratio H is obtained based on the result of the PIDC control using the above-described equations (59) to (61). The expressions (59) to (61) and their explanation will be described again below.

As indicated by the hatched portion in FIG. 23, the sum of the cascade temperatures overshooted from the initial temperature S arrival time b to the initial temperature S stabilization time c is obtained. Assuming that this is the overshoot temperature sum F ″, F ″ is obtained by the following equation (59). In equation (59), the integration range of ∫g (t) dt is between b and c.
F ″ = ∫g (t) dt−S (c−b) (59)
Next, as indicated by the hatched portion in FIG. 24, the sum of temperatures from the lowest temperature S1 time point r1 to the initial temperature S stabilization time c is obtained. If this is the temperature sum G, G is obtained by the following equation (60). In equation (60), the integration range of ∫g (t) dt is between r1 and c.
G = ∫g (t) dt−S ′ (c−r1) (60)
Next, an overshoot temperature ratio H [percentage] is obtained. This is the ratio of the overshoot temperature sum F ″ shown in FIG. 23 to the temperature sum G shown in FIG. 24. Therefore, it is obtained by the following equation (61).
H = F ″ / G (61)
The calculated overshoot temperature ratio H is the heat ratio of the overshoot factor concealed in the integral output pattern of step 1 to step 2. Therefore, by reducing the overshoot temperature ratio H from the step 1 target temperature V, it is reflected in the integral output pattern in the temperature transition period, and the inside of the furnace can be quickly stabilized at an ideal temperature.

Initial temperature k (0), step 0 time s, step 1 target temperature V, step 1 time (j-s), step 2 time (aj), heater stable temperature (step 3 target temperature) k (n) Then, the step 1 target temperature V ′, the step 1 ramp rate T ′, and the step 2 ramp rate U ′ after the reduction of the overshoot temperature ratio H are obtained by the following formulas (62) to (64). .
V ′ = V * (1−H) (62)
T ′ [° C./min]=(V′−k(0))/(j−s) (63)
U ′ [° C./min]=(k(n)−V′)/(a−j) (64)

(Fifth embodiment)
Subsequently, a fifth embodiment of the present invention will be described.

  The heat treatment apparatus according to the present embodiment has the same configuration as the heat treatment apparatus according to the first embodiment described above. This embodiment is different from the first embodiment in the calculation method for obtaining the operation amount to be ramped up when obtaining the power control output pattern in the procedure of the temperature adjustment method. Hereinafter, in the present embodiment, the same components as those in the first embodiment and the processing contents are denoted by the same reference numerals, and description thereof is omitted.

  First, a power output pattern is obtained using a, d, and D1 'obtained in the same manner as in the first embodiment, and power control is performed.

Each parameter in FIG. 35 is calculated | required by the following formula | equation (65)-Formula (68).
I0 = (D1 '/ T1) + d (65)
I1 = d (66)
T1 = a / 2 (67)
T2 = a (68)
Therefore, when the period until the maximum overshoot time at the time of PID control is divided into two, the first half is a period in which the heater temperature is raised for ramp-up, and the second half is a period in which the heater temperature starts to fall but the cascade temperature rises with a delay. The first half of the period for outputting the power required for temperature rise, that is, the half at the time of the maximum overshoot is considered appropriate, and the parameters are set.

I0 is obtained by adding the manipulated variable d when the target temperature is stable to the manipulated variable when the temperature rise amount D ′ is output at a constant output amount and half the overshoot maximum time a.
Power control temperature characteristic data is acquired using this Power output pattern.

  Next, as shown in FIG. 34, the temperature adjustment evaluation time A1 [min] from the start of the power control is determined from the power control temperature characteristic basic data, and the target temperature S ′ reaching time b [min], the target temperature is determined. The S ′ stabilization time c [min], the maximum temperature j2 [° C.] indicated by the heater thermocouple, and the maximum temperature time a2 [min] are obtained. Further, the temperature f (t) [° C.] indicated by the heater thermocouple 2 and the temperature g (t) [° C.] indicated by the cascade thermocouple 3 at each time point within the temperature adjustment evaluation time A1 are obtained.

Next, as shown in FIG. 9, the sum of the temperatures overshooted from the target temperature S ′ arrival time b to the target temperature S ′ stabilization time c is obtained. If this is the total overshoot temperature K, K is obtained by the following equation (69). In Expression (69), the integration range of ∫m (t) dt is between b and c.
K = ∫g (t) dt−S ′ * (c−b) (69)
Next, as shown in FIG. 10, the temperature sum during the target temperature S ′ stabilization time c is obtained. Assuming that this is the temperature sum L1, L1 is obtained by the following equation (70). In equation (70), the integration range of ∫g (t) dt is between 0 and c.
L1 = ∫g (t) dt (70)
Next, an overshoot temperature ratio M [percentage] is obtained. This is the ratio of the overshoot temperature sum K to the temperature sum L1. Therefore, it is obtained by the following equation (71).
M = K / L1 (71)
Next, as shown in FIG. 35, by reducing the overshoot temperature ratio M from the operation amount obtained from the temperature rise of the power output pattern, the operation amount I0 can be made an appropriate operation amount. Assuming that the manipulated variable I0 after reducing the overshoot temperature ratio H is I0 ', I0' is obtained by the following equation (72).
I0 ′ = (I0−I1) * (1−M) + I1 (72)
As shown in FIG. 35 obtained by the above formulas (69) to (72), the overshoot temperature ratio is reduced, the temperature characteristic data is acquired again by the power control, and the same procedure is repeated several times for proper operation. The amount can be determined and the temperature can be adjusted.

(Sixth embodiment)
Next, a sixth embodiment will be described. The present embodiment is a modification of the configuration in the temperature control unit described in the first embodiment. Hereinafter, in the present embodiment, the same components as those in the first embodiment and the processing contents are denoted by the same reference numerals, and description thereof is omitted.

  As shown in FIG. 36, in the present embodiment, the manipulated variable X is calculated by switching between the PID calculation and the PlDC calculation at a preset time, and the manipulated variable Z is switched at the preset time between the PID calculation and the PlDC calculation. It is configured to calculate or from a power output pattern.

(Seventh embodiment)
Next, a seventh embodiment will be described. The present embodiment is a modification of the above-described first embodiment. Hereinafter, in the present embodiment, the same components as those in the first embodiment and the processing contents are denoted by the same reference numerals, and description thereof is omitted.

  As shown in FIG. 37, in the present embodiment, the manipulated variable X is calculated by PIDC calculation, and the manipulated variable Z is calculated by switching between the PlD calculation and the PIDC calculation at a preset time or calculated from the Power output pattern. It is the composition to do.

(Eighth embodiment)
Next, an eighth embodiment will be described. The present embodiment is a modification of the above-described first embodiment. Hereinafter, in the present embodiment, the same components as those in the first embodiment and the processing contents are denoted by the same reference numerals, and description thereof is omitted.

  As shown in FIG. 38, in the present embodiment, the manipulated variable Z is calculated by PlDC calculation or calculated from a Power output pattern.

(Ninth embodiment)
Next, a ninth embodiment will be described. The present embodiment is a modification of the above-described first embodiment. Hereinafter, in the present embodiment, the same components as those in the first embodiment and the processing contents are denoted by the same reference numerals, and description thereof is omitted.

  As shown in FIG. 39, in the present embodiment, the manipulated variable X is calculated by PID calculation, and the manipulated variable Z is calculated by PlDC calculation or calculated from the Power output pattern.

(Tenth embodiment)
Next, a tenth embodiment will be described. The present embodiment is a modification of the above-described first embodiment. Hereinafter, in the present embodiment, the same components as those in the first embodiment and the processing contents are denoted by the same reference numerals, and description thereof is omitted.

  As shown in FIG. 40, in this embodiment, the manipulated variable X is calculated by PIDC calculation, and the manipulated variable Z is calculated by PID calculation or calculated from a Power output pattern.

(Eleventh embodiment)
Next, an eleventh embodiment will be described. The present embodiment is a modification of the above-described first embodiment. Hereinafter, in the present embodiment, the same components as those in the first embodiment and the processing contents are denoted by the same reference numerals, and description thereof is omitted.

  As shown in FIG. 41, in the present embodiment, the manipulated variable X is calculated by PIDC calculation, and the manipulated variable Z is calculated by PlDC calculation or calculated from the Power output pattern.

(Twelfth embodiment)
Next, a twelfth embodiment will be described. The present embodiment is a modification of the above-described first embodiment. Hereinafter, in the present embodiment, the same components as those in the first embodiment and the processing contents are denoted by the same reference numerals, and description thereof is omitted.

  As shown in FIG. 42, in the present embodiment, the control amount A of the cascade thermocouple is ignored and the deviation E between the control amount B of the heater thermocouple and the target value Sc is used.

(Thirteenth embodiment)
Next, a thirteenth embodiment will be described. The present embodiment is a modification of the above-described first embodiment. Hereinafter, in the present embodiment, the same components as those in the first embodiment and the processing contents are denoted by the same reference numerals, and description thereof is omitted.

  As shown in FIG. 43, in this embodiment, the control amount A of the cascade thermocouple is ignored and the deviation E between the control amount B of the heater thermocouple and the target value Sc is used. The operation amount Z is calculated by PlDC calculation or calculated from a power output pattern.

(Fourteenth embodiment)
Next, a fourteenth embodiment will be described. The present embodiment is a modification of the above-described first embodiment. Hereinafter, in the present embodiment, the same components as those in the first embodiment and the processing contents are denoted by the same reference numerals, and description thereof is omitted.

  As shown in FIG. 44, in the present embodiment, the manipulated variable Z is calculated by switching three of the PlD calculation, the PIDC calculation, and the Power output pattern at a preset time, or calculated from the Power output pattern. It has become.

(Fifteenth embodiment)
Next, a fifteenth embodiment is described. The present embodiment is a modification of the above-described first embodiment. Hereinafter, in the present embodiment, the same components as those in the first embodiment and the processing contents are denoted by the same reference numerals, and description thereof is omitted.

  As shown in FIG. 45, in the present embodiment, the operation amount Z is calculated by switching the PIDC calculation and the Power output pattern at a preset time, or is calculated from the Power output pattern.

(Sixteenth embodiment)
Next, a sixteenth embodiment will be described. The present embodiment is a modification of the above-described first embodiment. Hereinafter, in the present embodiment, the same components as those in the first embodiment and the processing contents are denoted by the same reference numerals, and description thereof is omitted.

  As shown in FIG. 46, in the present embodiment, the operation amount Z is calculated by switching between the PlD calculation and the Power output pattern at a preset time, or is calculated from the Power output pattern.

(Seventeenth embodiment)
Subsequently, a seventeenth embodiment of the present invention will be described.

The heat treatment apparatus according to the present embodiment is not a vertical apparatus structure like the heat treatment apparatuses according to the above-described embodiments, but a single wafer apparatus. Hereinafter, in the present embodiment, constituent parts having the same functions as those described in the above-described embodiments are denoted by the same reference numerals and description thereof is omitted. As shown in FIG. 47, the heat treatment apparatus according to the present embodiment includes a temperature control unit having the same configuration and function as the temperature control unit 71 in the first embodiment shown in FIG.

  Specifically, in the heat treatment apparatus according to the present embodiment, as shown in FIGS. 47 and 48, the processing furnace 86 is configured as a single wafer CVD furnace (single wafer cold wall CVD furnace). And a chamber 223 in which a processing chamber 201 for processing a wafer (semiconductor wafer) 85 as a substrate to be processed is formed. The chamber 223 is formed in a cylindrical shape in which an upper cap 224, a cylindrical cup 225, and a lower cap 226 are combined so that upper and lower end surfaces are closed.

  A wafer loading / unloading port 250 that is opened and closed by a gate valve 244 is opened horizontally in the middle of the cylindrical wall of the cylindrical cup 225 of the chamber 223. The wafer loading / unloading port 250 is a wafer 85 that is a substrate to be processed. Is formed in the processing chamber 201 so that it can be carried in and out by a wafer transfer device (not shown in FIG. 48). In other words, the wafer 85 is transported through the wafer loading / unloading port 250 and loaded into and out of the processing chamber 201 while being mechanically supported from below by the wafer transfer device.

  An exhaust port 235 connected to an exhaust device (not shown) such as a vacuum pump communicates with the processing chamber 201 at the upper part of the wall surface of the cylindrical cup 225 facing the wafer loading / unloading port 250. The inside of the processing chamber 201 is exhausted by an exhaust device.

  Further, an exhaust buffer space 249 communicating with the exhaust port 235 is formed in an annular shape at the upper part of the cylindrical cup 225, and acts so that the exhaust is performed uniformly on the front surface of the wafer 85 together with the cover plate 248.

  The cover plate 248 extends on a part of the susceptor (substrate holding means) 84 so as to cover the edge portion of the wafer 85, and controls a CVD film formed on the edge portion of the wafer 85. Used for.

A shower head 236 for supplying a processing gas is integrally incorporated in the upper cap 224 of the chamber 223. That is, a gas supply pipe 232 is inserted in the ceiling wall of the upper cap 224, and for example, a processing gas 241a such as a raw material gas or a purge gas and a processing gas 241b are introduced into each gas supply pipe 232, and an opening / closing valve 243, a flow rate. A gas supply device comprising a control device (mass flow controller = MFC) 241 is connected. A disc-shaped shower plate (hereinafter referred to as a plate) 240 is horizontally fixed on the lower surface of the upper cap 224 at a distance from the gas supply pipe 232. The outlets (hereinafter referred to as “blow-out outlets”) 247 are provided so as to be uniformly arranged over the entire surface and to circulate through the upper and lower spaces.

  A buffer chamber 237 is formed by an inner space defined by the inner surface of the upper cap 224 and the upper surface of the plate 240, and the buffer chamber 237 diffuses the processing gas 230 introduced into the gas supply pipe 232 evenly as a whole. It is made to blow out from each blower outlet 247 equally in the shape of a shower.

  An insertion hole 278 is formed in a circular shape at the center of the lower cap 226 of the chamber 223, and a support shaft 276 formed in a cylindrical shape is inserted into the processing chamber 201 from below on the center line of the insertion hole 278. . The support shaft 276 is raised and lowered by an elevating mechanism (elevating means) 268 using an air cylinder device or the like.

A heating unit 251 is concentrically arranged at the upper end of the support shaft 276 and is fixed horizontally. The heating unit 251 is moved up and down by the support shaft 276. That is, the heating unit 251 includes a support plate 258 formed in a disc shape, and the support plate 258 is fixed concentrically to the upper end opening of the support shaft 276. A plurality of electrodes 253 that also serve as support columns are vertically erected on the upper surface of the support plate 258, and a heater (heating means) 81 that is formed in a disc shape between the upper ends of these electrodes 253 and is divided and controlled in a plurality of regions. Is cross-linked and fixed. The electrical wiring 257 for these electrodes 253 is inserted through the hollow portion of the support shaft 276.

  In addition, a reflector 252 is fixed to the support plate 258 below the heater 81, and heat generated from the heater 81 is reflected to the susceptor 84 side to act on efficient heating.

  Further, radiation thermometers 87 and 88 as temperature detecting means are introduced from the lower end of the support shaft 276, and the tips of the radiation thermometers 87 and 88 are installed with a predetermined gap with respect to the back surface of the susceptor 84. The radiation thermometers 87 and 88 are composed of a combination of a rod made of quartz and an optical fiber, detect radiation emitted from the back surface of the susceptor 84 (for example, the back surface corresponding to the divided region of the heater 81), and It is used to calculate the back surface temperature (the temperature of the wafer 85 can be calculated based on the relationship between the temperature of the wafer 85 and the susceptor 84 acquired in advance), and the heating condition of the heater 81 is controlled based on this calculation result. .

On the outside of the support shaft 276 of the insertion hole 278 of the lower cap 226, a rotating shaft 277 formed in a cylindrical shape having a larger diameter than the support shaft 276 is disposed concentrically and is inserted into the processing chamber 201 from below. The rotary shaft 277 is lifted and lowered together with the support shaft 276 by a lifting mechanism 268 using an air cylinder device or the like. A rotating drum 227 is concentrically arranged at the upper end of the rotating shaft 277 and fixed horizontally. The rotating drum 227 is rotated by the rotating shaft 277. That is, the rotating drum 227 includes a rotating plate 229 formed in a donut-shaped flat plate and a rotating cylinder 228 formed in a cylindrical shape, and an inner peripheral edge of the rotating plate 229 is an upper end of a cylindrical rotating shaft 277. A rotating cylinder 228 is concentrically fixed to the outer peripheral edge of the upper surface of the rotating plate 229 and is fixed to the opening. A susceptor 84 formed in a disk shape using silicon carbide, aluminum nitride, or the like is placed on the upper end of the rotating cylinder 228 of the rotating drum 227 so as to close the upper end opening of the rotating cylinder 228.

  As shown in FIG. 48, a wafer elevating device 275 is installed on the rotary drum 227. The wafer elevating device 275 is configured by protruding pins (substrate protruding means) 266 and 274 provided on each of two elevating rings formed in a circular ring shape. The rotation side ring is disposed on the rotating plate 229 of the rotating drum 227 concentrically with the support shaft 276. On the lower surface of the rotation side ring, a plurality of (three in the present embodiment) protruding pins (hereinafter referred to as rotation side pins) 274 are arranged at equal intervals in the circumferential direction so as to face downward in the vertical direction. Each rotation-side pin 274 is slidably fitted in each guide hole 255 that is arranged on the rotation plate 229 on a line concentric with the rotation cylinder 228 and opened in the vertical direction. The lengths of the rotation-side pins 274 are set to be equal to each other so that the rotation-side ring can be pushed up horizontally, and are set to correspond to the push-up amount of the wafer from above the susceptor. The lower end of each rotation-side pin 274 faces the bottom surface of the processing chamber 201, that is, the upper surface of the lower cap 226 so as to be separably seated.

On the support plate 258 of the heating unit 251, another lifting ring (hereinafter referred to as a heater side ring) formed in a circular ring shape is disposed concentrically with the support shaft 276. On the lower surface of the heater side ring, a plurality of (three in the present embodiment) protruding pins (hereinafter referred to as heater side pins) 266 are arranged at equal intervals in the circumferential direction and vertically downward. Each heater-side pin 266 is slidably fitted in each guide hole 254 that is arranged on the support plate 258 on a line concentric with the support shaft 276 and opened in the vertical direction. The lengths of these heater-side pins 266 are set to be equal to each other so that the heater-side ring can be pushed up horizontally, and their lower ends are opposed to the upper surface of the rotation-side ring with an appropriate air gap. That is, these heater side pins 266 do not interfere with the rotation side ring when the rotation drum 227 rotates.

In addition, a plurality of (three in the present embodiment) protruding pins (hereinafter referred to as protruding portions) 266 are arranged on the upper surface of the heater side ring at equal intervals in the circumferential direction. The upper end of the projecting upper portion 266 is opposed to the heater 81 and the insertion hole 256 of the susceptor 84. The lengths of these protrusions 266 are set to be equal to each other so that the wafer 85 placed on the susceptor 84 is horizontally floated from the susceptor 84 through the insertion holes 256 of the heater 81 and the susceptor 84 from below. Further, the lengths of these protrusions 266 are such that the upper ends of the protrusions 266 are in the state where the heater-side ring is seated on the support plate 258.
It is set so as not to protrude from the top surface. That is, these protrusions 266 do not interfere with the susceptor 84 when the rotary drum 227 rotates, and do not prevent the heater 81 from being heated.

  As shown in FIG. 48, the chamber 223 is horizontally supported by a plurality of columns 280. Each elevating block 281 is fitted to these columns 280 so as to be movable up and down, and an elevating platform that is raised and lowered by an elevating drive device (not shown) using an air cylinder device or the like is interposed between the elevating blocks 281. 282 is installed. A susceptor rotating device is installed on the lifting platform 282, and a bellows 279 is interposed between the susceptor rotating device and the chamber 223 so as to hermetically seal the outside of the rotating shaft 277.

  A brushless DC motor is used for the susceptor rotating mechanism (rotating means) 267 installed on the lifting platform 282, and an output shaft (motor shaft) is formed as a hollow shaft and configured as a rotating shaft 277. The susceptor rotating mechanism 267 includes a housing 283, and the housing 283 is installed on the elevator base 282 in the vertical upward direction. A stator (stator) 284 composed of an electromagnet (coil) is fixed to the inner peripheral surface of the housing 283. That is, the stator 284 is configured by winding a coil wire (enamel-coated copper wire) 286 around an iron core (core) 285. A lead wire (not shown) is electrically connected to the coil wire 286 through an insertion hole (not shown) formed in the side wall of the housing 283, and the stator 284 receives electric power from a brushless DC motor driver (not shown). Is supplied to the coil wire 286 through a lead wire to form a rotating magnetic field.

Inside the stator 284, a rotor (rotor) 289 is arranged concentrically with an air gap (gap) set, and the rotor 289 is rotatably supported by the housing 283 via upper and lower ball bearings 293. ing. That is, the rotor 289 includes a cylindrical main body 290, an iron core (core) 291, and a plurality of permanent magnets 292, and a rotating shaft 277 is fixed to the main body 290 so as to rotate integrally with a bracket 288. . The iron core 291 is fitted and fixed to the main body 290, and a plurality of permanent magnets 292 are fixed to the outer periphery of the iron core 291 at equal intervals in the circumferential direction. A plurality of magnetic poles arranged in an annular shape are formed by an iron core 291 and a plurality of permanent magnets 292, and the rotating magnetic field formed by the stator 284 cuts off the magnetic fields of the plurality of magnetic poles, that is, the permanent magnets 292. The child 289 rotates.

  The upper and lower ball bearings 293 are respectively installed at the upper and lower ends of the main body 290 of the rotor 289, and a gap for absorbing the thermal expansion of the main body 290 is appropriately set in the upper and lower ball bearings 293. The gap between the ball bearings 293 is set to 5 to 50 μm in order to absorb the thermal expansion of the main body 290 and suppress the minimum shakiness. The gap of the ball bearing means a gap generated on the opposite side when the ball is brought to either the outer race or the inner race.

The opposing surface of the stator 284 and the rotor 289 is opposed to a cover 287 which is an outer and inner enclosure member constituting a double cylindrical wall, and the inner peripheral surface of the housing 283 and the outer peripheral surface of the main body 290 are opposed to each other. A predetermined air gap (gap) is set between each cover 287 and each cover 287. The cover 287 is made of stainless steel, which is a non-magnetic material, and is formed in a cylindrical shape with a very thin cylindrical wall. The entire periphery of the cover 287 is formed by electron beam welding on the housing 283 and the main body 290 at the upper and lower opening ends of the cylinder. It is firmly and uniformly fixed over. The cover 287 is made of stainless steel, which is a non-magnetic material, and is extremely thin. Therefore, the cover 287 not only prevents the magnetic flux from diffusing, but also prevents the motor efficiency from decreasing, and the stator 284 coil wire 286 and the rotor 289 are permanent. Corrosion of the magnet 292 can be prevented, and contamination inside the processing chamber 201 due to the coil wire 286 and the like can be reliably prevented. The cover 287 completely isolates the stator 284 from the inside of the processing chamber 201 in a vacuum atmosphere by surrounding the stator 284 in an airtight seal state.

  Further, a magnetic rotary encoder 294 is installed in the susceptor rotating device. That is, the magnetic rotary encoder 294 includes a detection ring 296 as a detection target made of a magnetic material, and the detection ring 296 is formed in a circular ring shape using a magnetic material such as iron. On the outer periphery of the detection ring 296, a large number of teeth as detection parts are arranged in an annular shape.

A magnetic sensor 295 for detecting each tooth which is a detected portion of the detected ring 296 is installed at a position of the housing 283 facing the detected ring 296. A gap (sensor gap) between the front end surface of the magnetic sensor 295 and the outer peripheral surface of the detected ring 296 is set to 0.06 to 0.17 mm. The magnetic sensor 295 is configured to detect the change in magnetic flux at these opposed positions accompanying the rotation of the detection ring 296 using a magnetoresistive element. The detection result of the magnetic sensor 295 is transmitted to the drive control unit of the brushless DC motor, that is, the susceptor rotation mechanism 267, and is used for position recognition of the susceptor 84 and used to control the rotation amount of the susceptor 84. The processing furnace 86 includes a main control unit 7 including a gas flow rate control unit 72, a drive control unit 74, a heating control unit 71a, a temperature detection unit 71b, and the like. The temperature control unit 71 includes a heating control unit 71a and a temperature detection unit 71b. The temperature control unit 71 includes heater thermocouples 82 and 83 for detecting the temperature of the heater 81.

  The gas flow rate control unit 72 is connected to the MFC 241 and the opening / closing valve 243 and controls the gas flow rate and supply. The drive control unit 74 is connected to the susceptor rotating mechanism 267 and the lifting block 281 and controls the driving thereof. The heating control unit 71 a is connected to the heater 81 via the wiring 257 and controls the heating state of the heater 81. The temperature detector 71b is connected to the radiation thermometers 87 and 88, detects the temperature of the susceptor 84, and is used for heating control of the heater 81 in cooperation with the heating controller 71a.

  Next, a film forming process in the method for manufacturing a semiconductor device in the heat treatment apparatus in this embodiment will be described.

When the wafer 85 is carried in and out, the rotary drum 227 and the heating unit 251 are lowered to the lower limit position by the rotary shaft 277 and the support shaft 276. Then, the lower end of the rotation-side pin 274 of the wafer lifting device 275 abuts the bottom surface of the processing chamber 201, that is, the upper surface of the lower cap 226, so that the rotation-side ring rises relative to the rotation drum 227 and the heating unit 251. . The raised rotation side ring lifts the heater side ring by pushing up the heater side pin 266. When the heater side ring is lifted, the three projecting upper portions 266 standing on the heater side ring pass through the heater 81 and the insertion hole 256 of the susceptor 84, and the wafer 85 placed on the upper surface of the susceptor 84 is seen from below. Support and lift from the susceptor 84.

When the wafer lifting device 275 is in a state where the wafer 85 is lifted from the upper surface of the susceptor 84, an insertion space is formed between the lower space of the wafer 85, that is, the lower surface of the wafer 85 and the upper surface of the susceptor 84. A tweezer which is a substrate holding plate provided in a wafer transfer machine (not shown in FIG. 48) is inserted into the insertion space of the wafer 85 from the wafer loading / unloading port 250. The tweezer inserted below the wafer 85 moves up and receives the wafer 85. Upon receiving the wafer 85, the tweezer retreats from the wafer loading / unloading port 250 and unloads the wafer 85 from the processing chamber 201. Then, the wafer transfer machine that unloads the wafer 85 by the tweezers transfers the wafer 85 to a predetermined storage location such as an empty wafer cassette outside the processing chamber 201.

Next, the wafer transfer machine receives the wafer 85 to be subjected to the next film formation process from a predetermined storage location such as an actual wafer cassette by the tweezer, and carries it into the processing chamber 201 from the wafer loading / unloading port 250. The tweezers transport the wafer 85 to a position above the susceptor 84 so that the center of the wafer 85 coincides with the center of the susceptor 84. When the wafer 85 is transferred to a predetermined position, the tweezers are slightly lowered to transfer the wafer 85 to the susceptor 84. The tweezer that has transferred the wafer 85 to the wafer lifting / lowering device 275 moves out of the processing chamber 201 from the wafer loading / unloading port 250. When the tweezer leaves the processing chamber 201, the wafer loading / unloading port 250 is closed by a gate valve (gate valve) 244.

  When the gate valve 244 is closed, the rotary drum 227 and the heating unit 251 are lifted by the lift table 282 through the rotary shaft 277 and the support shaft 276 with respect to the processing chamber 201. As the rotary drum 227 and the heating unit 251 are raised, the thrust pins 266 and 274 are lowered relative to the rotary drum 227 and the heating unit 251, and the wafer 85 is placed on the susceptor 84 as shown in FIG. It will be in the state completely transferred to. The rotation shaft 277 and the support shaft 276 are stopped at a position where the upper end of the protrusion 266 is at a height close to the lower surface of the heater 81.

  On the other hand, the processing chamber 201 is exhausted by an exhaust device (not shown) connected to the exhaust port 235. At this time, the vacuum atmosphere in the processing chamber 201 and the external atmospheric pressure atmosphere are isolated by the bellows 279.

  Subsequently, the rotating drum 227 is rotated by the susceptor rotating mechanism 267 via the rotating shaft 277. That is, when the susceptor rotating mechanism 267 is operated, the rotating magnetic field of the stator 284 cuts the magnetic field of the plurality of magnetic poles of the rotor 289, so that the rotor 289 rotates, so that the rotation fixed to the rotor 289 The rotating drum 227 is rotated by the shaft 277. At this time, the rotational position of the rotor 289 is detected every moment by the magnetic rotary encoder 294 installed in the susceptor rotation mechanism 267 and transmitted to the drive control unit 74, and the rotation speed and the like are controlled based on this signal. .

  During the rotation of the rotating drum 227, the rotating side pin 274 is separated from the bottom surface of the processing chamber 201, and the heater side pin 266 is separated from the rotating side ring. In addition, the heating unit 251 can maintain a stopped state. That is, in the wafer elevating device 275, the rotation side ring and the rotation side pin 274 are rotated together with the rotation drum 227, and the heater side ring and the heater side pin 266 are stopped together with the heating unit 251.

When the temperature of the wafer 85 rises to the processing temperature and the exhaust amount of the exhaust port 235 and the rotational operation of the rotary drum 227 are stabilized, the processing gas 230 is supplied to the supply pipe 232 as shown by the solid arrow in FIG. To be introduced. The processing gas 230 introduced into the gas supply pipe 232 flows into the buffer chamber 237 functioning as a gas dispersion space and diffuses radially outward in the radial direction. The flow is substantially uniform and blows out toward the wafer 85 in a shower shape. The processing gas 230 blown out in the form of a shower from the group of the outlets 247 passes through the space above the cover plate 248 and is sucked into the exhaust port 235 via the exhaust buffer space 249 and exhausted.

  At this time, since the wafer 85 on the susceptor 84 supported by the rotating drum 227 is rotating, the processing gas 230 blown out in a shower form from the group of the outlets 247 is in contact with the entire surface of the wafer 85 evenly. . Since the processing gas 230 contacts the entire surface of the wafer 85 evenly, the film thickness distribution and film quality distribution of the CVD film formed on the wafer 85 by the processing gas 230 are uniform over the entire surface of the wafer 85.

  Since the heating unit 251 is not rotated by being supported by the support shaft 276, the temperature distribution of the wafer 85 heated by the heating unit 251 while being rotated by the rotating drum 227 is uniformly controlled over the entire surface. The As described above, the temperature distribution of the wafer 85 is uniformly controlled over the entire surface, whereby the film thickness distribution and film quality distribution of the CVD film formed on the wafer 85 by the thermochemical reaction are uniformly controlled over the entire surface of the wafer 85.

  When a predetermined processing time selected in advance elapses, the operation of the susceptor rotation mechanism 267 is stopped. At this time, the rotational position of the susceptor 84, that is, the rotor 289 is monitored every moment by the magnetic rotary encoder 294 installed in the susceptor rotating mechanism 267, so that the susceptor 84 is accurately stopped at the preset rotational position. The In other words, the protrusion 266 and the insertion hole 256 of the heater 81 and the susceptor 84 are matched accurately and with good reproducibility.

  When the operation of the susceptor rotating mechanism 267 is stopped, as described above, the rotating drum 227 and the heating unit 251 are lowered to the loading / unloading position by the lifting platform 282 via the rotating shaft 277 and the support shaft 276. . As described above, the wafer 85 is lifted from the susceptor 84 by the action of the wafer lifting device 275 during the lowering. At this time, since the protruding portion 266 and the insertion hole 256 of the heater 81 and the susceptor 84 are matched accurately and with good reproducibility, there is no possibility that the protruding portion 266 pushes up the susceptor 84 and the heater 81.

  Thereafter, the above-described operations are repeated, whereby a CVD film is formed on the next wafer 85.

  The configuration shown in the first to sixteenth embodiments described above is for a vertical heat treatment apparatus, but is not limited to this, and from the first embodiment described above. The configuration of the sixteenth embodiment can also be applied to a heat treatment apparatus as a single wafer apparatus as shown in the present embodiment.

As described above, according to each of the above-described embodiments, by changing the operation amount input to the heating unit provided in the heat treatment apparatus, the control amount by the heating unit is changed, and the control amount is changed. Control by performing feedback processing based on proportional, integral, and differential operations based on the amount, and maintaining the control amount to the target value or following the control, the output of an appropriate manipulated variable is set in advance and the proportional / A means for selectively switching control for outputting an operation amount according to integral / derivative calculation is provided, and the appropriate operation amount and an operation amount according to the proportional / integral / derivative operation are appropriately selected and input to the heating means. The temperature control method characterized by this can be provided.

  Further, the temperature control method having the above-described configuration may be automatically adjusted to an optimum operation amount by repeatedly executing the temperature control method several times. In addition, it goes without saying that the temperature control method having the above-described configuration may be programmed and mounted on a computer.

  In addition, the calculation method of the target temperature in the integral pattern output shown in the above-described embodiments is an exemplification, and may be calculated by a calculation method as shown in Expression (27), for example.

  In addition, an output control pattern is created based on the output amount obtained as a result of the PID operation, and the I operation amount is obtained from the temperature waveform detected by the heater thermocouple when the heater is output-controlled by the output control pattern. It suffices if the integral output pattern closer to the temperature waveform detected by the subtracted heater thermocouple can be obtained more quickly and efficiently by subtracting the P operation (not only the P operation but also the D operation). .

  As described above, according to the present invention, when the temperature in the processing chamber fluctuates, such as during ramp-up or boat-up recovery, first, the operation is performed by PID control until the temperature stabilizes from the initial temperature to the target temperature. The basic data is acquired, and the amount of heat spent from the initial temperature to stabilization at the target temperature in the basic data is divided into a stable heat amount and a heat-up amount of heat, and an output control pattern is obtained. The temperature control based on the output control pattern thus obtained is performed, and the integral output pattern is obtained by subtracting the P operation (or P operation and operation) from the basic data detected by the thermocouple at this time. It has become.

  In each of the above-described embodiments, an example in which the present invention is applied to a vertical apparatus and a single-wafer apparatus has been shown. However, the present invention is not limited to this and is also applied to a heat treatment apparatus as a horizontal apparatus. It goes without saying that it is possible.

  Even if the temperature inside the furnace is adjusted by automatic temperature adjustment as described above, for example, when a film is deposited in the furnace like a POLY film, or the film accumulation of dummy wafers and supplementary wafers, and the number of production wafers change. Depending on the conditions, the temperature characteristics may change, and the temperature stability may deteriorate. Such a change in temperature characteristic is due to a change in heat absorption rate accompanying film deposition and a change in heat capacity due to change in the number of wafers.

  Along with the change in temperature characteristics, the above-mentioned maximum temperature of the heater thermocouple affects the cascade thermocouple maximum temperature, and the operation amount pattern Time1, Time2 and the actual heater thermocouple maximum temperature of FIG. There is a difference between the time of maximum temperature and this is one of the major factors of deterioration of temperature stability.

  Therefore, in this embodiment, when the temperature characteristic changes, the temperature characteristic data is acquired or acquired in advance, and when the difference between Time1 and Time2 becomes larger than a preset time, Time1 or Time2 is calculated again, Prevents deterioration of temperature stability.

  A flow of calculation at Time 1 and Time 2 will be described. As described above, the difference between the actual heater thermocouple maximum temperature point and the cascade thermocouple maximum temperature point is set in advance due to the film deposition in the furnace, the film deposition on the dummy wafer, the replenishment wafer, and the number of production changes. The target step Time is updated when the threshold value is larger than the threshold value.

  Referring to FIG. 12, when the difference between Time 1 which is the maximum temperature of the heater thermocouple and Time 2 which is the maximum temperature of the cascade thermocouple is equal to or greater than a predetermined threshold value, it is determined that the conditions in the furnace have changed. By changing the manipulated variable input to the heating means provided in the heat treatment apparatus, feedback control is performed by proportional / integral / differential calculation based on the controlled variable that changes the manipulated variable, and input to the heated means .

In short, when the deviation between Time 1 and Time 2 and the actual heater thermocouple and cascade thermocouple becomes G or more for a certain period of time due to the temperature characteristic change, the target step Time is updated.
Conditional expression (1)
Abs (Time 1-Heater thermocouple maximum temperature point)> G
When the above conditional expression is satisfied,
Time1 = Heater thermocouple maximum temperature point Time2 = Cascade thermocouple maximum temperature point-Time1
And
Conditional expression (2)
Abs (Time2-Cascade thermocouple maximum temperature)> G
When the above conditional expression is satisfied,
Time1 = Heater thermocouple maximum point Time2 = Cascade thermocouple maximum point-Time1
And Here, the constant time G in the conditional expressions (1) and (2) is common, but different values may be used in the conditional expressions (1) and (2).
Further, when the conditional expression (1) is satisfied, only the update of Time 1 is satisfied, and when the conditional expression (2) is satisfied, even when only the Time 2 is updated, an effect can be expected to deteriorate the temperature stability.

  With the configuration as described above, even if a relatively large disturbance occurs in the controlled furnace, the integrated output that allows the controlled variable to follow the target value most quickly from the total amount of manipulated variable including the disturbance. Set the pattern in advance, output the integrated value from the specified time instead of the integral operation, and then perform the integral operation again from the point where most of the deviations due to disturbances and others are considered to have disappeared. Therefore, the control amount output from the control target can be quickly and accurately changed to the target value.

  In addition, the procedure of the present invention is a natural phenomenon, the actual temperature, operation amount, time, etc., which have so far depended on the experience and intuition of skilled workers at the time of temperature adjustment, can be determined by a predetermined calculation formula, The temperature can be adjusted quickly and reliably, and the time and cost can be reduced.

  In addition, by programming the procedure of the present invention and incorporating it as software in a temperature controller or the like, it is possible to perform appropriate temperature adjustment without requiring operator intervention.

  Although the present invention has been described in detail according to particular embodiments, it will be apparent to those skilled in the art that various changes and modifications can be made without departing from the spirit and scope of the invention.

  As described in detail above, according to the present invention, it is possible to provide a temperature adjustment method, a heat treatment apparatus, and a semiconductor device manufacturing method that can contribute to improvement in work efficiency and cost reduction.

It is a functional block diagram for demonstrating the 1st Embodiment of this invention. It is a figure for demonstrating the detail of the structure of the reaction tube periphery in the same apparatus. It is a figure for demonstrating the process (PID control) in the PID calculating unit. It is a figure for demonstrating a PIDC calculator. It is a figure for demonstrating an integral output pattern. It is a figure for demonstrating Power control. It is a flowchart for demonstrating the procedure of the temperature adjustment method. It is a figure for demonstrating the parameter calculated | required based on temperature characteristic basic data. It is a figure for demonstrating the sum total of the cascade temperature which overshooted. It is a figure for demonstrating the temperature sum total from the evaluation start to target temperature S 'stabilization time c. It is a figure for demonstrating the relationship between the operation amount and time. It is a figure for demonstrating the heater waveform L (t) which deducted P operation part from the waveform of k (t). It is a figure for demonstrating the integral output pattern M (t). It is a figure for demonstrating the target temperature and ramp rate when r steps of integral pattern output are set for every fixed time. It is a figure for demonstrating generation | occurrence | production of the overshoot and undershoot when an integral output pattern is appropriate. It is a figure for demonstrating the deviation between zones. It is a figure for demonstrating the temperature adjustment procedure in case there exists a deviation between zones. It is a figure which shows the temperature change in a furnace when the temperature control by this Embodiment is performed. It is a figure which shows the temperature change in a furnace when the temperature control by this Embodiment is performed. It is a figure which shows the temperature change in a furnace when the temperature control by this Embodiment is performed. It is a flowchart for demonstrating the temperature adjustment procedure in 2nd Embodiment. It is a figure for demonstrating the parameter calculated | required based on temperature characteristic basic data. It is a figure for demonstrating the sum total of the cascade temperature which overshooted. It is a figure for demonstrating the temperature sum total from the minimum temperature S1 time point r to the initial temperature S stabilization time c. It is a figure for demonstrating a Power output pattern. It is a figure for demonstrating determination of the temperature adjustment evaluation time etc. after Power control start based on the temperature characteristic basic data at the time of Power control. It is a figure for demonstrating the integral output pattern M (t). It is a figure for demonstrating the target temperature and ramp rate at the time of setting w steps of integral pattern output for every fixed time. It is a figure for demonstrating generation | occurrence | production of overshoot and undershoot in the case where an integral output pattern is appropriate. It is a figure for demonstrating the deviation between zones. It is a figure for demonstrating the temperature adjustment in case there exists a deviation between zones. It is a figure which shows the specific example of the temperature recovery after boat loading. It is a figure which shows the specific example of the temperature recovery after boat loading. It is a figure for demonstrating determination of the various parameters from the temperature characteristic basic data at the time of Power control. It is a figure for demonstrating the reduction from the operation amount calculated | required from the temperature rising heat amount of the Power output pattern of the overshoot temperature ratio M. It is a functional block diagram for demonstrating the 6th Embodiment of this invention. It is a functional block diagram for demonstrating the 7th Embodiment of this invention. It is a functional block diagram for demonstrating the 8th Embodiment of this invention. It is a functional block diagram for demonstrating the 9th Embodiment of this invention. It is a functional block diagram for demonstrating the 10th Embodiment of this invention. It is a functional block diagram for demonstrating the 11th Embodiment of this invention. It is a functional block diagram for demonstrating the 12th Embodiment of this invention. It is a functional block diagram for demonstrating the 13th Embodiment of this invention. It is a functional block diagram for demonstrating the 14th Embodiment of this invention. It is a functional block diagram for demonstrating the 15th Embodiment of this invention. It is a functional block diagram for demonstrating the 16th Embodiment of this invention. It is a functional block diagram for demonstrating the 17th Embodiment of this invention. It is a figure for demonstrating the detail of a structure of the periphery of a process furnace in a single wafer apparatus about 17th Embodiment of this invention.

Claims (2)

A heating means for heating the processing chamber for processing the substrate, a heating control section for controlling the heating means,
First and second temperature detecting means for detecting the temperature in the processing chamber, the first temperature detecting means being disposed closer to the substrate than the second temperature detecting means, and the second temperature detecting means The temperature detection means is a temperature adjustment method in a heat treatment apparatus disposed closer to the heating means than the first temperature detection means,
Obtaining a time difference between the maximum temperature point of the detected temperature by the second temperature detecting unit and the maximum temperature point of the detected temperature by the first temperature detecting unit;
The temperature control method of the heat processing apparatus which has the process of controlling the said heating means so that this time difference may be below a predetermined time difference when this time difference is more than a predetermined time difference. A heating means for heating the processing chamber for processing the substrate, a heating control section for controlling the heating means,
First and second temperature detecting means for detecting the temperature in the processing chamber, the first temperature detecting means being disposed closer to the substrate than the second temperature detecting means, and the second temperature detecting means The temperature detecting means is a heat treatment apparatus disposed at a position closer to the heating means than the first temperature detecting means,
The heating control unit obtains a time difference between the maximum temperature point of the detected temperature by the second temperature detecting unit and the maximum temperature point of the detected temperature by the first temperature detecting unit, and the time difference is equal to or greater than a predetermined time difference. If there is, a heat treatment apparatus characterized by controlling the heating means so that the time difference is equal to or less than a predetermined time difference.