JPS61279121A - Diffusing device - Google Patents

Abstract

PURPOSE:To control the temperature of a wafer at the prescribed value by a method wherein the temperature measuring probe of the wafer is formed using a retaining quartz pipe and a lens, a quartz rod and a total reflection prism which are provided in the quartz pipe, thereby enabling the direct measure of the wafer temperature when a diffusing process is performed. CONSTITUTION:Quartz rods 15A and 15B are provided inside a retaining quartz pipe 18 through the intermediary of a support 19, a lens 17 is provided on the port opposing to a wafer 2, total reflection prisms 16A and 16B are provided, and a temperature measuring probe 8 is constituted. The radiant light coming from the wafer 2 is measured by the temperature measuring probe 8, and the result is transmitted to a light-temperature converter 11 by an optical fiber 10 through the intermediary of a connector 9. In a diffusion device controlling part 12, the difference between the measured temperature and the preset reference temperature is outputted to a controlling part 13, and the heating power of a heater 5 is controlled in accordance with the value of said difference. When an oxidation and diffusion process is going to be performed to the wafer 2, treatment gas is introduced into a quartz pipe 1 from a treatment gas inlet 7.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Application of the Invention] The present invention relates to a diffusion device for semiconductor manufacturing, and in particular to a wafer thermometer i11! suitable for controlling the temperature of a wafer in the device. Regarding the probe for l.

[Background of the invention]

A conventional method for measuring wafer temperature a1 in a diffusion device is described in Japanese Patent Laid-Open No. 56-6429. However, in this known example, the temperature of the wafer holder on which the wafer is placed and held is measured instead of measuring the temperature of the wafer itself.

The temperature difference between the wafer and the wafer holder is greatest during a transient state in which the wafer is moved in and out of the diffusion device processing chamber.

Temperature control in a transient state is becoming increasingly important in processes with higher integration and higher precision, and known measurement methods are
The amount of heat generated by the heater is actually controlled by the indicated value of a thermocouple outside the processing chamber. In this case, the temperature difference between the wafer and the thermocouple outside the processing chamber becomes even larger.

This is the problem with the current temperature control of the diffusion device.

[Purpose of the invention]

The purpose of the present invention is to directly adjust the wafer temperature to a+, including transient conditions.
! The object of the present invention is to provide a probe that enables the determination of

[Summary of the invention]

Thermocouples, which are often used for general temperature measurement, cannot be used in the diffusion device processing chamber because they themselves become impurities for the wafer. Optical fibers are used as a means of guiding radiant light to the outside, but commercially available products are heat resistant to 200°C, so
It is not possible for eight people to be inside the diffuser for processing at temperatures between 1000°C and 1000°C. The heat resistance limit of optical fiber is the coating material (synthetic resin)
It's ready. The function of the coating material is to provide flexibility to the optical fiber body (core + cladding) and not to act directly as a light guide, so there is no problem in terms of heat resistance. It is conceivable to extract the radiant light to the outside.

However, the cladding of commercially available optical fibers contains germanium as an impurity and cannot be introduced into the diffuser. However, even if the optical path is made of quartz alone, it can achieve almost the same effect as an optical fiber. However, since it is not flexible, all you have to do is bend the quartz rod a little in advance, or use an optical system that also uses a prism to bend the light and extract it to the outside.

[Embodiments of the invention]

An embodiment of the present invention will be described below with reference to FIG. In the figure, 1 is a quartz tube, and inside it there are multiple tubes (for example, 50-
100 wafers 2 are placed on boat X'3. Bows 1 to 3 are moved in and out of the quartz tube by a fork 4. Wafer No. 2 in quartz tube I is heated by Peter 5 in quartz tube I.

Between the heater 5 and the quartz tube 1, there is a soaking tube 6 that makes the wall temperature distribution uniform so that it is suitable for uniform heating (there may be no soaking tube). When oxidizing and diffusing the wafer 2 in the quartz tube I, a processing gas from the processing guide 117/+1 is introduced into the quartz tube I (in the case of annealing 1-, the processing gas is not flowed).

In the figure, 8 is a temperature measurement probe according to the present invention.

A signal (radiant light from the wafer) is led out of the quartz tube 1 and externally transmitted via a connector 9 to an optical fiber 10 to a radiant light-to-temperature converter II. The temperature signal from the radiant light-temperature converter 11 is processed by the dispersion device control section 12, and the heat generation amount of the heater is transmitted to the control section 13, which controls the heat generation amount of the heater and brings the temperature of the wafer 2 to a predetermined temperature. Control.

In the above explanation, it is assumed that the boat and wafer are inside the quartz tube, but in reality, the boat and wafer may be outside the quartz tube.

In this case, temperature control is performed using the designated value of the thermocouple 14 provided on or near the wall of the soaking tube (it is known to provide a thermocouple on the wall or near the wall of the soaking tube). Note that an infrared thermometer may be used instead of the thermocouple 14.

FIG. 2 is a detailed view of the temperature measuring probe 8.

Quartz rods 15A, 15B total reflection prism 16A.

16B, the lens 17 is the main component, and is fixed inside the holding quartz tube 18 via a support 19. (
Holding quartz tube 18. No support 19, quartz rod 1.5
A, 1.5B total reflection prism 16A.

16B lens] Even if the combination of 7 is used in a bare state, the function will not change). The total reflection prism 1112 and lens 3 are also made of quartz (in order to prevent impurities from entering the wafer, no material other than quartz can be used in the quartz tube that processes the wafer).

The radiant light from the wafer 2 enters the total reflection value rhythm 16A, which is focused by the lens 17, is refracted, and passes through the quartz rod 1.5A,
It reaches the connector 9 via the total reflection prism 1.6B and the quartz rod 1513. The subsequent route is as already described, so it will be omitted.

Normally, the gap between wafers is about 4 to 10 mm, and in order to prevent the temperature measurement probe from adversely affecting the wafer processing process, it is necessary to make it as small as possible.

Since both the lens and the total reflection prism can be made with a diameter or one side of one stroke, there is no need to worry about the influence on the processing process of a wafer whose temperature is distributed at about 511II+gap.

According to this embodiment, the wafer temperature inside the quartz tube can be directly measured. By feeding this signal back to the diffusion device control section, it becomes much easier to control the wafer temperature to a desired temperature than in the past. In particular, when a wafer is placed into a quartz tube, transient changes in wafer temperature are severe. In this case, it is not easy to quickly bring the wafer temperature close to a steady value just by measuring the temperature of the boat or the temperature near the heater as in the conventional method. The effects of this embodiment are clearly impaired by such a transient state.

〔Effect of the invention〕

According to the present invention, since the wafer temperature inside the quartz tube can be directly measured, the wafer temperature can be easily controlled to a desired temperature.

Especially when placing a wafer into a quartz tube, it is possible to bring the wafer to the desired temperature much more quickly than with conventional indirect measurement (board temperature or temperature near the heater).

[Brief explanation of drawings]

Figure 1 is a three-dimensional illustration of an embodiment of the present invention;
The figure is a cross-sectional view of a temperature measurement probe that is central to the present invention.

Claims (1)

[Claims] 1. In a diffusion device that heats a plurality of wafers in a quartz tube by a heater outside the quartz tube, a lens, a prism, and a light guide are provided in the gap between adjacent wafers. A first radiant light guide is provided, which guides the radiant light from the wafer to the outside of the wafer's outer circumference, and further includes a prism,
Guide one end of a second radiant light guide equipped with a light guide to the outside of the diffuser and connect it to an optical fiber with a connector, and guide the radiant light from the wafer to a radiant light-temperature converter outside the diffuser. A diffusion device characterized by directly measuring wafer temperature. 2. A diffusion device according to claim 1, wherein the lens, prism, and light guide are all made of quartz. 3. A diffusion device according to claim 2, wherein the wafer temperature directly measured in the diffusion device is used to control the amount of heat generated by the heater to control the wafer temperature to a desired value.