JPH11163070A - Temperature control method for heat treatment process in production of semiconductor device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To control the board temperature accurately even when a board having radiation rate or light absorption (treating temperature) dependent on the film structure, film quality, impurity concentration, or the like, is heat treated by feeding the board temperature back to the lamp output and using it in closed circuit control for controlling the board to a desired temperature. SOLUTION: Temperature control method for measuring the temperature of a board 51 being heated through irradiation with light from a light source 113 by means of a thermocouple 11 without touching the board using a temperature measuring unit 1 contained in a heating furnace having the light source 113 as a heating means comprises a step for measuring the temperature of the board 51 using the temperature measuring unit 1 which is made of a high thermal conductivity material on the periphery of the contact part with the board 51 (temperature measuring part coating member 22) in the coating member 21 of the thermocouple 12 thereof and the coating member 21 (line part coating member 23) is made of a material having high light transmittance or reflectance except the periphery of the contact part, a step for measuring the temperature of the board 51 using the temperature measuring unit 1, and a step performing closed circuit control for feeding a measured value back to the output of the light source 113.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a temperature control method in a heat treatment process for manufacturing a semiconductor device, and more particularly, to a temperature control method by closed circuit control.

[0002]

In recent years, in the MOS device with the miniaturization of semiconductor devices, in order to suppress the short channel effect, in order to improve the cutoff frequency f _T is a bipolar device, the need to form a shallow junction with high accuracy Has occurred. And as one of the methods to form a shallow junction,
Heating method by light irradiation (R
TA: Rapid Thermal Annealing).
RTA is also used for recovery of crystal defects caused by ion implantation, various annealing methods such as sintering, and formation of oxide films and nitride films. Therefore, it is extremely important to accurately control the substrate temperature for substrates having various film structures, substrates having various impurity concentrations, and the like.

However, when the substrate is heated by light irradiation, the emissivity of the substrate changes depending on the film structure, film quality, impurity concentration, etc., so that the light irradiation intensity is constant [Open Loop Co., Ltd.
ntrol)], the amount of light absorption (processing temperature) of the substrate changes. Therefore, it is extremely difficult to accurately control the heating state of the substrate including various variations (variations due to film thickness, film quality, impurity amount, structure, and the like) as the manufacturing process becomes complicated. Further, the processing temperature of the substrate changes due to the light transmittance of the quartz tube constituting the substrate heating device, the light reflectance of the inner wall of the chamber, the output of a lamp serving as a light source, and the like over time. To address this issue,
Closed loop control, which measures the temperature of the substrate and feeds back the measured value to the output of the lamp, has been studied. Thereby, if accurate measurement of the substrate temperature can be realized, excellent substrate temperature control becomes possible.

A device for measuring the temperature of a substrate includes a radiation thermometer. This radiation thermometer has the advantage that temperature can be measured without contact. Another temperature measuring device is a thermocouple. When measuring the temperature with a thermocouple, there are a method of directly contacting the thermocouple with the surface of the substrate, a method of fixing the thermocouple to the surface of the substrate using a heat-resistant adhesive, and the like. These methods have the advantage that the temperature of the substrate (alloy portion) of the thermocouple directly contacts the substrate, so that the substrate temperature can be measured almost accurately.

As another temperature measuring device, a thermocouple is inserted into a covering member made of silicon carbide (SiC),
An apparatus for indirectly measuring the temperature of a substrate by bringing the thermocouple into contact with the substrate via a covering member is disclosed in Japanese Patent Laid-Open No. 4-148546.
No. 6,086,045.

[0006]

However, in the temperature measurement using the radiation thermometer, unlike the contact-type temperature measurement method using a thermocouple, the measurement accuracy is affected by the surface condition of the object to be measured, or the measurement accuracy is affected. Be strongly influenced by the environment. Therefore, the emissivity differs for each substrate having various film structures and impurity concentrations, and it is difficult to perform accurate temperature measurement.

Further, in the method of measuring the temperature by bringing a thermocouple into direct contact with the substrate, problems such as deterioration of the thermocouple due to the reaction between the substrate and the thermocouple and metal contamination of the substrate occur.

Further, in the temperature measurement using a thermocouple provided inside the covering member, the problem of metal contamination of the substrate by the thermocouple is solved, but the temperature measured by the thermocouple is the temperature of the covering member. Also, when the temperature of the substrate rises in the process of heat-treating the substrate by the light irradiation type heat treatment apparatus, not only is the coated member heated by heat conduction, but the coated member itself is directly absorbed by the irradiated light and heated. You. Therefore, in the closed-circuit control that changes the light irradiation intensity, the amount of heating due to the light absorption of the covering member changes depending on the irradiation intensity, and thus the amount of light absorption due to the change in the emissivity of the substrate having various film structures and impurity concentrations. It is difficult to accurately measure a change in (substrate temperature). As described above, since there is no technique for measuring the substrate temperature with high accuracy, there has been a problem in the conventional substrate temperature control by closed circuit control.

Further, heat conduction from the substrate, radiation absorption from the covering member, and absorption of irradiation lamp light differ depending on the material used as the covering member. Quartz and silicon carbide (Si
In the case of C), for example, coating with quartz suppresses light absorption, but poor heat conduction makes it difficult to measure the substrate temperature, resulting in poor thermal response. On the other hand, the coating with silicon carbide is excellent in the conductivity of the substrate temperature, but has a large light absorption, so that the measurement temperature significantly depends on the light irradiation intensity. Due to such thermal characteristics, each material has advantages and disadvantages.

There is also a method in which the covering member is flattened at a contact portion between the covering member and the substrate to form a pseudo-surface contact state, thereby increasing the efficiency of heat conduction from the substrate. This also increases the heat capacity of the member. As a result, heating due to direct absorption of light increases, so that accurate temperature measurement of the substrate cannot be performed.

Therefore, in the closed-circuit control for changing the light irradiation intensity, the amount of heating due to the light absorption of the covering member changes depending on the irradiation intensity, so that the emissivity of the substrate having various film structures and impurity concentrations changes. It is not possible to accurately measure the change in light absorption (substrate temperature) due to light.

[0012]

SUMMARY OF THE INVENTION The present invention is directed to a temperature control method in a heat treatment step for manufacturing a semiconductor device, which is provided to solve the above-mentioned problems, and is housed in a heating furnace having a light source in the heating means. This is a temperature control method using a temperature measuring device that measures the temperature of a substrate heated by light irradiation from a contact point using a thermocouple. In the covering member covering the thermocouple of the temperature measuring device, the periphery of the contact portion with the substrate is formed of a material having a high thermal conductivity, and the covering member excluding the periphery of the contact portion with the substrate is a material having a high light transmittance or A step of measuring the temperature of the substrate using such a temperature measuring device, which is formed of a material having a high light reflectance, and a step of performing closed circuit control for feeding back the measured value to the output of the light source And

In the temperature control method in the heat treatment step of manufacturing a semiconductor device, the substrate temperature is measured by the temperature measuring apparatus having the above-described configuration, and the closed circuit control for feeding back the measured value to the output of the light source is combined. Control accuracy increases. That is, in the above temperature measuring device, the thermocouple is covered with the covering member, and the covering member of the covering member covering the temperature measuring section is made of a material having high thermal conductivity. It becomes easy to conduct to the part. Therefore, although the temperature is measured via the covering member, the temperature of the substrate can be accurately measured.

Since the covering member of the other part except the covering member made of the material having high thermal conductivity is made of a material having a high light transmittance or a material having a high light reflectivity, the covering member of that part is made of a material having a high light reflectance. Hardly receives the light and absorbs the light. Therefore, the temperature rise of the covering member due to the light irradiation hardly occurs, so that a change in the temperature measurement value of the thermocouple due to heat absorption of the covering member and a change in the temperature measurement value due to radiation from the substrate hardly occur. Therefore, the substrate temperature can be measured with high precision, and the output of the light source is fed back based on the temperature, so that the substrate temperature can be controlled with high precision.

[0015]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An example of an embodiment of a temperature control method in a heat treatment step of manufacturing a semiconductor device according to the present invention will be described.
FIG. 1 is a schematic sectional view of a temperature measuring device together with a substrate, and an example of a light irradiation type heat treatment device will be described with reference to a schematic sectional view of FIG.

As shown in FIG. 1, a temperature measuring device 1 used for measuring the temperature of a substrate to be heat-treated in a heat-treating step in the manufacture of a semiconductor device includes a light source 11 as a heating means described with reference to FIG.
The temperature of the substrate 51, which is housed in a heating furnace (quartz tube 112) having a tube 3 and heated by light irradiation from the light source 113, is measured in a contact manner by the thermocouple 11. The periphery of the covering member 21 covering the thermocouple 11 around the contact portion with the substrate 51 (the temperature measuring section covering member 22 of the temperature measuring section 12) is made of a material having high thermal conductivity. Such materials include, for example, silicon carbide, silicon compounds (eg, molybdenum silicide, titanium silicide,
Silicide such as cobalt silicide) or alumina is used. On the other hand, the covering member 21 (the line covering member 23) of the temperature measuring device 1 except for the vicinity of the contact portion with the substrate 51 is made of a material having a high light transmittance or a material having a high light reflectance. As a material having a high light transmittance, for example, quartz is used. Alternatively, as a material having a high light reflectance, for example, alumina is used.

More specifically, the temperature measuring device 1 composed of a contact thermocouple has a structure in which the thermocouple 11 is covered by a covering member 21. The thermocouple 11 is made of, for example, platinum (Pt) -platinum (Pt) · 10% rhodium (R
h) a thermocouple, and a temperature measuring part (alloy part) 12 of the thermocouple 11
Is formed of an alloy of a platinum wire and a platinum / 10% rhodium wire. At least one of the wires is loosely inserted into the insulating tube 31, and the insulating tube 31 is made of, for example, quartz. In this example, the platinum conductive wire 13 is loosely inserted into the insulating tube 31. Naturally, platinum 1
The 0% rhodium wire 14 may be loosely inserted into the insulating tube 31.

The temperature measuring section 12 of the covering member 21
Is covered with a material having high thermal conductivity, and is provided in a state where the temperature measuring unit 12 is sufficiently in contact with the temperature measuring unit 12. The temperature measuring section covering member 22 is made of, for example, a material having a thermal conductivity that is about ten times or more that of quartz (thermal conductivity = 1.66 W · m ⁻¹ · K ⁻¹ ), preferably 100 W · m ⁻¹ · K.
It is made of a material having a thermal conductivity of about K ^-1 or more. As such a material, for example, silicon carbide (thermal conductivity = 2
61 W · m ⁻¹ · K ⁻¹ ).

The temperature measuring section covering member 22 is made of silicon carbide, which is a material having high thermal conductivity, which is in sufficient contact with the temperature measuring section 12. From this, although the temperature is measured via the temperature measuring section covering member 22, the heat of the substrate 51 is sufficiently transmitted to the temperature measuring section 12, and the temperature of the substrate 51 can be measured. Further, the temperature measuring section covering member 22 has a structure having a small surface area in order to minimize direct absorption of light and a small heat capacity in order to enhance thermal responsiveness.
That is, the shape of the temperature measuring section covering member 22 is a cap shape with respect to the temperature measuring section 12, for example, the outer diameter of the cap is 1.4 mm, the inner diameter of the cap is 0.9 mm, and the height of the cap is It is formed to 1.4 mm.

The line covering member 23 other than the temperature measuring unit covering member 22 in the covering member 21 is
It is made of quartz with excellent infrared transmittance, and it is formed in a tubular shape with a circular cross section as a structure that minimizes the direct absorption of light, so it absorbs light even when it is irradiated with light There are few things. Therefore, the temperature rise of the wire covering member 23 due to the light irradiation hardly occurs, so that the temperature measurement value of the thermocouple 11 due to the heat absorption of the wire covering member 23 hardly changes.

In general, the conductors 13 and 14 of the thermocouple 11
Are formed in a state in which light is easily reflected. Therefore, even if the thermocouple 11 is irradiated with light, the thermocouple 11 is hardly affected by the irradiated light. Further, since the temperature measuring section covering member 22 is formed of silicon carbide which has high heat resistance and is thermally stable at a normal heat treatment temperature of a silicon substrate (1200 ° C. or lower), the temperature measuring section covering member 22 is not heat treated. The substrate 51 is not contaminated by the above.

The substrate 51 has a plurality of (for example, two) quartz substrate supporting portions (not shown) protruding from a quartz tray (not shown) and a temperature measuring device to be a tip portion of the temperature measuring device 1. It is horizontally supported by the part covering member 22.

Next, an example of a light irradiation type heat treatment apparatus using the temperature measuring apparatus 1 described in the above embodiment will be described with reference to a schematic sectional view of FIG.

As shown in FIG. 2, a tube 112 made of quartz glass having a high transmittance to infrared rays is installed inside a reaction furnace 111 covered with gold. A halogen lamp serving as a light source 113 is provided. One end of the tube 112 is provided at one end of the reaction furnace 111. The tube 112 is opened and closed at the time of loading and unloading of the substrate 51, and the inside of the tube 112 can be airtight when the tube 112 is sealed. Thus, the door 115 provided with the packing 114 (for example, packing made of resin) is provided.

On the other hand, a gas introduction pipe 116 for introducing a gas is connected to the other end of the tube 112. A quartz tray 117 for supporting the substrate 51 is placed inside the tube 112.
The tray 117 is provided with a substrate support portion 118 made of quartz.
(Temperature measuring section 12 via temperature measuring section covering member 22 described with reference to FIG. 1)
The substrate 51 is supported by. The thermocouple wires 13 and 14 of the temperature measuring device 1 are drawn out through holes 119 provided at the end of the reaction furnace 111. In addition, a window 120 is formed in the reaction furnace 111 so that a temperature can be measured by a pyrometer 131. Thus, the light irradiation type heat treatment apparatus 101 is configured.

The temperature measuring device 1 described with reference to FIG.
In the first step, the temperature of the substrate 51 heated by the light irradiation type heat treatment apparatus 101 described with reference to FIG. 2 is measured.

Next, in a second step, the measured value of the temperature of the substrate 51 is fed back to the output of the light source 113,
The light source 11 is controlled to maintain the temperature of the substrate 51 within a predetermined range.
The closed circuit control for adjusting the output of No. 3 is performed.

In the closed circuit control, for example, when the substrate temperature is high, the difference between the measured temperature value and the set temperature value is obtained,
Based on the difference, the output of the light source 113 is reduced so that the substrate temperature matches the set temperature. On the other hand, when the substrate temperature is low, the difference between the measured temperature value and the set temperature value is obtained. By increasing the output of the light source 113 based on the temperature of the substrate to match the set temperature,
The substrate temperature is always kept at the set temperature. From the difference between the measured substrate temperature and the set temperature, the amount of increase or decrease in the output of the light source may be determined based on, for example, data obtained in advance.

In the temperature control method in the heat treatment step of manufacturing the semiconductor device, the temperature measurement of the substrate 51 by the temperature measuring device 1 having the above-described configuration is combined with the closed circuit control for feeding back the measured value to the output of the light source 113. Thus, the accuracy of controlling the temperature of the substrate 51 is improved. That is,
In the temperature measuring device 1, the thermocouple 11 is covered with the covering member 21, and the temperature measuring unit covering member 22 of the covering member 21 covering the temperature measuring unit 12 is made of a material having high thermal conductivity. In addition, the temperature of the substrate 51 is easily transmitted to the temperature measuring unit 12. Therefore, although the temperature is measured via the temperature measuring section covering member 22, the temperature of the substrate 51 can be accurately measured.

Since the wire covering member 23 other than the temperature measuring portion covering member 22 is made of a material having a high light transmittance or a material having a high light reflectivity, the wire covering member 23 is formed of a material having a high light reflectance. It hardly absorbs the light when it is irradiated. Therefore, the temperature rise of the wire portion covering member 23 due to light irradiation hardly occurs, so that the change in the temperature measurement value of the thermocouple 11 due to the endothermic heat of the wire portion covering member 23 and the change in the temperature measurement value due to the radiation from the substrate 51 hardly occur. Does not happen. Therefore, the temperature of the substrate 51 can be measured with high accuracy, and the output of the light source 113 is fed back based on the temperature, so that the temperature of the substrate 51 can be controlled with high accuracy. Become.

Next, as a comparative example, a conventional temperature measuring device for a contact thermocouple will be described with reference to the schematic sectional view of FIG. FIG. 3 shows a state in which the conventional temperature measuring device 201 is inserted into the quartz tube 112 of the light irradiation type heat treatment device, and the temperature of the substrate 51 is measured.
The same components as those described with reference to FIG. 1 are denoted by the same reference numerals. This conventional temperature measuring device 201
Has a structure in which the thermocouple 11 is covered with a covering member 221 made of silicon carbide (SiC).
The thermocouple 11 is, for example, platinum (Pt) -platinum (P
t) 10% rhodium (Rh) thermocouple, thermocouple 1
The first temperature measuring part (alloy part) 12 is formed of an alloy of a platinum wire and a platinum / 10% rhodium wire. Also, at least one of the wires is loosely inserted into the insulating tube 31,
This insulating tube 31 is made of, for example, quartz. In this example, the platinum conductive wire 13 is loosely inserted into the insulating tube 31.

In the above-mentioned conventional temperature measuring device 201, what the thermocouple 11 measures is the temperature of the covering member 221. In the process of heating the substrate 51 by a light irradiation type heat treatment apparatus (not shown), the substrate 51 is heated and its temperature rises.
Not only is the substrate 1 heated, but also the coating member 221 itself is heated by directly absorbing the light of the light source 113, so that accurate measurement of the substrate temperature becomes difficult. Heat conduction from the substrate 51, radiation of the covering member 221 from the substrate 51, and light source 113
The absorption of light from varies depending on the material used as the covering member 221, and in the case of quartz and silicon carbide as an example,
Since quartz has low radiation absorption and poor heat conduction, it is difficult to measure the substrate temperature and has poor thermal responsiveness. On the other hand, since silicon carbide has a large amount of radiation absorption and good heat conduction, coating with silicon carbide is excellent in substrate temperature conduction, but has a large amount of light absorption, and the measurement temperature is significantly dependent on light irradiation intensity.

Next, the structures of three types of samples in which the film thickness for changing the emissivity of the substrate is changed will be described with reference to FIG.

As shown in FIG. 4A, a first evaluation sample 61 has a silicon oxide (SiO ₂ ) film 63 on one side (front side) of a silicon substrate 62 and a polycrystalline silicon film having a thickness of 150 nm. A capping silicon oxide film 65 having a thickness of 64 nm or 300 nm is laminated, and a silicon oxide (SiO ₂ ) film 66, 1 is formed on the other side (back side) of the silicon substrate 62.
The polycrystalline silicon film 67 having a thickness of 50 nm is laminated. The silicon oxide films 63 and 66 are 70
Although the thickness is changed in the range of 0 nm to 900 nm, the dependency of the light absorption amount on the film thickness is small. The implantation energy of the polycrystalline silicon film 64 is 40 ke.
V and boron difluoride (BF ₂ ) are ion-implanted under the condition that the dose is 5.4 × 10 ¹⁴ / cm ² .

As shown in FIG. 4B, the second evaluation sample 71 is provided on one side (front side) of the silicon substrate 72.
A silicon oxide (SiO ₂ ) film 73, 1 having a thickness of 00 nm;
A polycrystalline silicon film 74 having a thickness of 50 nm and a capping silicon oxide film 75 having a thickness of 300 nm are laminated, and a silicon oxide (SiO ₂ ) film 76 having a thickness of 800 nm is formed on the other side (back side) of the silicon substrate 72. The polycrystalline silicon film 77 is laminated. The thickness of the polycrystalline silicon film 77 on the rear surface side is changed in the range of 150 nm to 350 nm, and the dependency of the light absorption amount on the film thickness is larger than that of the first evaluation sample 61. The implantation energy of 40 k is applied to the polycrystalline silicon film 74 on the front side.
Boron difluoride (BF ₂ ) is ion-implanted under the conditions of eV and a dose of 5.4 × 10 ¹⁴ / cm ² .

As shown in FIG. 4C, a third evaluation sample 81 is composed of a silicon oxide (SiO ₂ ) film 83 on one side (front side) of a silicon substrate 82 and a polycrystalline silicon film having a thickness of 150 nm. A capping silicon oxide film 85 having a thickness of 84 nm and 300 nm is laminated, and a silicon oxide (SiO ₂ ) film 86, 1 is formed on the other side (back side) of the silicon substrate 82.
A polycrystalline silicon film 87 having a thickness of 50 nm is laminated. The third evaluation sample 81 has a thickness of the silicon oxide films 83 and 86 in the range of 100 nm to 600 nm, that is, 100 nm, 200 nm, and 300 n.
Five types of m, 400 nm, and 600 nm are prepared. Therefore, in these third evaluation samples 81, the thickness dependence of the light absorption of the silicon oxide films 83 and 86 is extremely large. Boron difluoride (BF ₂ ) is ion-implanted into the polycrystalline silicon film 84 under the conditions that the implantation energy is 40 keV and the dose is 5.4 × 10 ¹⁴ / cm ² .

The plurality of first evaluation samples 61 having different thicknesses of the silicon oxide film 63, the plurality of second evaluation samples 71 having different thicknesses of the polycrystalline silicon film 77, and the thicknesses of the silicon oxide films 83 and 86 are different. A thermocouple was directly attached to each of the plurality of different third evaluation samples 81 using a heat-resistant adhesive, and accurate temperature measurement was performed using the thermocouple.

In the temperature measurement, the first, second, and third evaluation samples 61, 71, and 81 described with reference to FIG. 4 are used instead of the substrate 51 by using the light irradiation type heat treatment apparatus 101 described with reference to FIG. Heat treatment [RTA (Rapid Therma
l Annealing)] In the RTA sequence, as shown in FIG. 5, an evaluation sample is loaded into a tube 112 (see FIG. 2) set at a temperature of 200 ° C.
Then, after heating to the set temperature T of the RTA at a heating rate of 50 ° C./s, the set temperature T is maintained for t seconds (for example, 10 seconds), and then cooled to 400 ° C. at a cooling rate of 50 ° C./s. The evaluation sample is unloaded from the tube 112. The set temperature T of the RTA is 90
0 ° C, 1000 ° C, 1050 ° C, 1100 ° C, 1150
Set to ° C. The above RTA sequence is an example, and can be changed as appropriate.

FIG. 6 shows the temperature dependence of the sheet resistance of each evaluation sample when processing is performed at each set temperature.
In FIG. 6, the vertical axis represents the sheet resistance ρs, and the horizontal axis represents the RTA.
Shows the set temperature. As shown in FIG.
Up to about 50 ° C., a substantially constant sheet resistance value of about 2140 Ω / □ is shown, and at a temperature higher than about 1000 ° C., the sheet resistance sharply decreases. And 142 at 1050 ° C.
The sheet resistance becomes about 0Ω / □ and becomes 9 at 1100 ° C.
The sheet resistance was about 80 Ω / □, and at 1150 ° C., the sheet resistance was about 800 Ω / □.

FIG. 7 shows the sheet resistance ρs (vertical axis) and the measured temperature (vertical axis) of each evaluation sample measured by the temperature measuring device 1 (see FIG. 1) of the present invention and the conventional temperature measuring device 201 (see FIG. 3). (Horizontal axis).
The values measured by the temperature measuring device 1 are indicated by white circles, triangles, and squares, and the values measured by the temperature measuring device 201 are indicated by black circles, triangles, and squares.

Specifically, in the evaluation sample of the second evaluation sample 71 in which the thickness of the polycrystalline silicon film 77 is 250 nm, a lamp output is used so that the substrate temperature becomes 1050 ° C., and the silicon oxide of the first evaluation sample 61 is used. Membrane 6
3, 66, the thickness of the polycrystalline silicon film 77 of the second evaluation sample 71, and the same for all the evaluation samples in which the thicknesses of the silicon oxide films 83, 86 of the third evaluation sample 81 were changed. Heat treatment is performed with lamp output. That is,
Continuous processing is performed by open circuit control in which heat treatment is performed under a constant light irradiation intensity, and the sheet resistance and the substrate temperature are measured.

As shown in FIG. 7, in open circuit control in which processing is performed at the same light irradiation intensity regardless of the substrate, the sheet resistance (substrate temperature) changes due to the difference in the thickness of the film formed on the substrate surface. In FIGS. 6 and 7, when the temperature-dependent curve of the sheet resistance at the set temperature of the RTA is compared with the temperature-dependent curve of the sheet resistance at the temperature measured by the temperature measuring device 1, the shapes of the curves match. ing. From this,
This shows that the temperature measuring device 1 (see FIG. 1) can accurately measure the substrate temperature with good reproducibility even for various substrates having different substrate structures and different light absorption amounts. On the other hand, the relationship between the measured temperature measured by the conventional temperature measuring device 201 (see FIG. 3) and the sheet resistance at that time is significantly different from the temperature-dependent curve in FIG. The temperature measurement device 201 indicates that the substrate temperature cannot be accurately measured.

The silicon oxide film 8 of the third evaluation sample 81
The samples having different thicknesses of 3,86 were subjected to the closed circuit control using the temperature measuring device 201 (see FIG. 3) and the temperature measuring device 1 (see FIG. 1) in accordance with the sequence described with reference to FIG. The treatment was performed at a measurement temperature of 1050 ° C., and the film thickness dependency of the sheet resistance at this time is shown in FIG. 8 together with the film thickness dependency in the open circuit control. In FIG. 8, the vertical axis indicates the sheet resistance ρs, the horizontal axis indicates the silicon oxide film thickness, black circles indicate open circuit control, and white squares indicate closed circuits by the temperature measurement device 1. Control, white triangles indicate closed circuit control by the conventional temperature measuring device 201. Similarly, FIG. 9 shows the film thickness dependence of the converted temperature from the sheet resistance. In FIG. 9, black circles indicate open circuit control, white squares indicate closed circuit control by the temperature measurement device 1, Open triangles indicate closed-circuit control by the conventional temperature measuring device 201. In FIG. 9, the vertical axis indicates the substrate temperature, and the horizontal axis indicates the silicon oxide film thickness. As apparent from FIGS. 8 and 9, the dependency of the sheet resistance (substrate temperature) on the film thickness, which appears remarkably in the open circuit control, is improved by using the temperature measurement device 1. However, in the temperature measuring device 201, the effect of improving the film resistance of the sheet resistance (substrate temperature) is not seen.

The third circuit is controlled by a closed circuit using the temperature measuring device 1 (see FIG. 1) and the temperature measuring device 201 (see FIG. 3).
When the sample of the evaluation sample 81 in which the thickness of the silicon oxide films 83 and 86 is changed is processed at the measurement temperature of 1050 ° C. by the sequence described with reference to FIG. The lamp output at this time is shown in FIG. In FIG. 10, the vertical axis indicates the lamp output (output ratio with respect to the maximum output), the horizontal axis indicates the silicon oxide film thickness, black circles indicate open circuit control, and white squares indicate temperature. Closed circuit control by the measuring device 1 and open triangles indicate closed circuit control by the conventional temperature measuring device 201.

As shown in FIG. 10, in the closed circuit control using the temperature measuring device 1, it is understood that the substrate temperature is corrected by applying a higher lamp output at the silicon oxide film thickness where the substrate temperature becomes lower. However, as shown in FIG. 7, in the temperature measurement device 201, even if the substrate temperature actually changes, the measurement temperature hardly changes, so that the lamp output becomes almost the same regardless of the silicon oxide film thickness. The substrate temperature cannot be corrected by the thickness. From the above, the temperature measurement device 201 proves that the light from the lamp is directly absorbed by the temperature measurement device 201. Thus, it can be seen that it is necessary to reduce the surface area of silicon carbide to minimize the light absorption as the structure of the temperature measuring device. in short,
In a structure in which the thermocouple 11 is entirely covered like the conventional temperature measuring device 201, light absorption is large and accurate measurement in closed circuit control cannot be performed.

FIGS. 6 to 10 show the results of the closed-circuit control by the temperature measuring device 1 using the results of the third evaluation sample 81 in which the amount of light absorption of the substrate changes extremely depending on the thickness of the silicon oxide film. In an actual controlled semiconductor device manufacturing process, fluctuations in light absorption (substrate temperature) due to process variations in film thickness and film quality are much smaller, and an open circuit is controlled by a closed circuit control using the temperature measuring device 1. The thickness dependence of the sheet resistance (substrate temperature) in control can be sufficiently eliminated.

As described above, in order to perform accurate measurement in the closed circuit control, it is necessary to minimize the structure of silicon carbide covering the temperature measuring section and suppress light absorption.

As an example, the effect of closed-circuit control has been shown by using a substrate having a different film thickness and a different light absorption amount.
In the closed circuit control using the temperature measuring device 1 (see FIG. 1),
In order to measure the actual substrate temperature and feed it back to the output of the lamp serving as the light source 113, the light irradiation type heat treatment apparatus 10
In the case where the processing temperature of the substrate 51 changes due to the light transmittance of the quartz tube 112, the light reflectance of the inner wall of the reaction furnace 111, the output of the lamp serving as the light source 113 over time, or the like. In addition, the present invention realizes highly accurate measurement of the substrate temperature.

In order to evaluate the stability of this control method, a closed circuit control using the temperature measuring device 1 (see FIG. 1)
According to the sequence described with reference to FIG.
A heat treatment is performed at 050 ° C. for 10 seconds. At this time, the temperature measured by the temperature measurement device 1 and the pyrometer, and the light source 1
By reading the output of the lamp No. 13, 135
The stability of the closed circuit control by the temperature measuring device 1 in the continuous processing of zero sheets was evaluated. As a sample, a silicon substrate on which a thermal oxide film (SiO ₂ film) of 600 nm was formed was used, and measurement temperature and lamp output were sampled from the same silicon substrate. Measurement errors were eliminated.

The transition of the measured temperature of the temperature measuring device 1 shown in FIG. 11 is a closed circuit control by the temperature measuring device 1,
Naturally, it becomes constant at 1050 ° C. In FIG. 11, the vertical axis indicates the measured temperature, and the horizontal axis indicates the number of processed substrates.

However, the actual substrate temperature is not constant due to the drift of the temperature measuring device 1, and this drift is substantially constant under the same conditions for the same sample. Transition (Fig. 1
3). This result will be described with reference to FIG. In FIG. 12, the left vertical axis indicates the temperature measured by the pyrometer, the right vertical axis indicates the drift temperature,
The horizontal axis indicates the number of processed substrates. As shown in FIG. 12, a drift from 0 ° C. to about −3 ° C. is observed up to about 200 sheets from the start of using a new temperature measuring device 1, but, for example, −3 ° C. ± as center
It can be seen that very excellent temperature control of 1.0 ° C. or less is possible.

FIG. 13 shows the relationship between the lamp output (vertical axis) and the number of substrates processed (horizontal axis). The lamp output shown here is a ratio to the maximum lamp output. This figure 1
As shown in FIG. 3, it can be seen that the lamp output is stable within a range of about ± 0.5% after continuous processing of about 50 sheets.

As described above, when using a new temperature measuring device 1, it is necessary to perform about 200 heat treatments at a measuring temperature of 1050 ° C. for 10 seconds according to the sequence described with reference to FIG. 5, for example. Very stable substrate temperature control is realized by closed-circuit control by the temperature measuring device 1 that has been subjected to a predetermined number of consecutive heat treatments. Note that the number of heat treatments of 200 is an example, and the number of heat treatments is appropriately selected depending on heat treatment conditions (temperature, time, and the like).

[0054]

As described above, according to the present invention,
Since the covering member of the temperature measuring unit of the temperature measuring device is made of a material having high thermal conductivity, the substrate temperature is sufficiently transmitted to the temperature measuring unit, and the covering members of other portions are made of a material having a high light transmittance or light reflectance. Therefore, the covering member in this portion hardly absorbs light. Therefore, the substrate temperature can be accurately measured by the present temperature measurement device. Since this temperature measuring device is used for closed-circuit control for controlling the substrate to a desired temperature by feeding back the substrate temperature to the lamp output, the emissivity and light absorption (processing temperature) depending on the film structure, film quality, impurity concentration, etc. Even when heat treatment is performed on a changing substrate, the substrate temperature can be controlled with high accuracy. Furthermore, stable control of the substrate temperature can be achieved even if there is a temporal change in the light transmittance of the quartz tube, the light reflectance of the inner wall of the reaction furnace, the output of the lamp as the light source, etc., which constitute the light irradiation type heat treatment apparatus. Can be.

[Brief description of the drawings]

FIG. 1 is an explanatory diagram of an embodiment according to a temperature control method of the present invention.

FIG. 2 is a schematic configuration sectional view of a light irradiation type heat treatment apparatus using a temperature measurement device.

FIG. 3 is a schematic sectional view of a configuration of a temperature measuring device of a comparative example.

FIG. 4 is a schematic sectional view of each evaluation sample.

FIG. 5 is an explanatory diagram of an RTA sequence.

FIG. 6 is a diagram showing the relationship between the sheet resistance of each evaluation sample and the set temperature of RTA.

FIG. 7 is a diagram illustrating a relationship between a measured temperature and a sheet resistance by a temperature measuring device of the present invention and a temperature measuring device of a comparative example.

FIG. 8 is a diagram illustrating a relationship between a sheet resistance and a silicon oxide film thickness of a third evaluation sample.

FIG. 9 is a relationship diagram between a substrate temperature and a silicon oxide film thickness of a third evaluation sample.

FIG. 10 is a diagram showing a relationship between a lamp output ratio and a silicon oxide film thickness in a third evaluation sample.

FIG. 11 is a diagram showing the relationship between the temperature measured by the temperature measuring device of the present invention and the number of substrates processed in the continuous processing of RTA.

FIG. 12 is a diagram showing the relationship between the temperature measured by a pyrometer and the number of substrates processed in a continuous RTA process, and the relationship between the drift amount of the measured temperature of the temperature measuring device of the present invention and the number of substrates processed.

FIG. 13 is a diagram showing the relationship between the lamp output and the number of substrates processed in the continuous processing of RTA.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Temperature measuring device, 11 ... Thermocouple, 21 ... Covering member, 2
2 ... Temperature measuring section covering member, 23 ... Line part covering member, 51 ... Substrate, 113 ... Light source

Claims (4)

[Claims] 1. A semiconductor device manufacturing method using a temperature measuring device which is housed in a heating furnace having a light source in a heating means and measures the temperature of a substrate heated by irradiation of light from the light source with a thermocouple in a contact type. A temperature control method in a heat treatment step, wherein, in a covering member covering a thermocouple of the temperature measuring device, a portion around a contact portion with the substrate is formed of a material having high thermal conductivity, and a portion around a contact portion with the substrate. Excluding the covering member, using the temperature measuring device formed of a material having a high light transmittance or a material having a high light reflectance, a step of measuring the temperature of the substrate, and measuring the measured value of the light source. Performing a closed circuit control to adjust the output of the light source so as to maintain the temperature of the substrate within a predetermined range by feeding back to the output. Temperature control method. 2. The temperature control method in a heat treatment step of manufacturing a semiconductor device according to claim 1, wherein the covering member around a contact portion with the substrate is formed by using the temperature measuring device made of silicon carbide, a silicon compound or alumina. A temperature control method in a heat treatment step of manufacturing a semiconductor device, comprising measuring a temperature of a substrate. 3. The temperature control method according to claim 1, wherein the temperature measurement device performs a predetermined number of heat treatments and then uses the temperature measurement for measuring the temperature of the substrate. A temperature control method in a heat treatment process for manufacturing a device. 4. The temperature control method according to claim 1, wherein the temperature measuring device is used for measuring the temperature of the substrate after performing a predetermined number of heat treatments. A temperature control method in a heat treatment process for manufacturing a device.