JPS63310793A - Device for molecular beam epitaxial growth - Google Patents

Abstract

PURPOSE:To obtain the title device for molecular beam epitaxial growth capable of being easily controlled, by which a crystal is excellently grown, and useful for a semiconductor, etc., by providing a nude ion gauge for measuring the intensity of a molecular beam in the cover having an opening for an incident molecular beam and cooling and heating parts. CONSTITUTION:The cover 7 consisting of a heater 10, the cooling part, the opening 7a, and the nude ion gauge 6 and a manipulator 8 connected to the cover and a crystal substrate 3 are arranged in an ultrahigh-vacuum chamber 9. Molecular beam sources 1 and 2 contg. As, etc., are then heated to about 300 deg.C to generate molecular beams 1a and 1b which are projected on the substrate 3 by opening shutters 4 and 5. Liquefied nitrogen is then introduced into the cover 7, the intensities of the molecular beams 1a and 1b made incident into the cover 7 are measured by the gauge 6 while cooling the gauge with the nitrogen or heating the gauge to accurately control the growing time of a crystal and the composition of the compd. such as AlGaAs and InGaAsP, and the crystal is epitaxially grown on the crystal substrate 3.

Description

[Detailed Description of the Invention] [Object of the Invention] (Industrial Field of Application) The present invention relates to a molecular beam epitaxy full growth device, and particularly to a method for controlling the growth rate, composition, etc. of a crystal that performs epitaxial growth on a crystal substrate. This invention relates to a system in which the intensity of molecular beams is measured using a nude ion gauge for control purposes.

(Prior art) In the molecular beam epitaxial growth method, the constituent elements of a crystal to be grown in an ultra-high vacuum are supplied in atomic or molecular form to the surface of a heated crystal substrate, resulting in 1-pita 4-shuffle growth. Compared to liquid phase growth and vapor phase growth, it is a crystal growth method that can produce crystals with the same quality of culmness by controlling the film thickness on a monoatomic layer basis. Therefore, especially
Applications to high-performance optical and electronic devices using compound semiconductors centered on GaAs (gallium arsenide) have begun, and some have already passed the research and development stage and are being used as production means. There are also examples.

Conventionally, a molecular beam epitaxial growth apparatus using such a one-pitch sial deposition method accommodates an evaporation source element material and heats it to a predetermined temperature to produce molecular beams 1a,
It has a pair of molecular beam sources 1.2 that emit molecular beams 1a and 2a, and each of these molecular beam sources 1 and 2 generates molecular beams 1a and 2a on the surface of the crystal base 3.
A shutter 4.5 is provided in front of each of the molecular beam sources 1.2 to irradiate the molecular beams 1a and 2a with VC control a.

The mate ion gauge 6 also has an opening 7 that is oriented in the direction in which the molecular beams 1a and 2a come in during molecular beam measurement.
The crystal substrate 3 is housed in a cover 7 provided with a portion a, and is supported by a manipulator 8 together with the crystal substrate 3. Then, the nude ion gauge 6 is changed to the position of the crystal substrate 3 by the manipulator 8 at a fixed time of the molecule [1].

The nude ion gauge 6 is a type of vacuum gauge that measures high vacuum, and usually has a filament, a collector,
Although parts such as the grid are housed within the glass tube, this is not necessary when installing it in a vacuum chamber, and the above-mentioned parts are exposed in the vacuum chamber 1.

In addition, the molecular beam sources 1 and 2, the crystal substrate 3, the nude ion gauge 6, etc.
It is located within 9.

Using such a molecular beam epitaxial growth apparatus, Ii
When growing J-products, the crystal growth rate, composition, etc.
In order to measure the molecular weight, a step of measuring the molecule 121 intensity is performed before and after the crystal growth step. The above-mentioned nude ion gauge 6 is widely used as this molecular beam intensity monitor, and the intensity of the molecular beams 1a and 2a supplied from the molecular beam source 1.2 such as a Knudsen cell is measured at the position of the crystal substrate 3 or its vicinity. The mate ion gauge 6 is placed at

Units such as Torr are used).

Conventionally, the nude ion gauge 6 used as this molecular beam intensity monitor includes one in which the gauge itself does not have a cover, and the other in which the gauge itself does not have a cover, and the molecular beam 1a as described above.

A cover 7 having an opening 7a in the direction from which the particles 2a come flying.
There are two types of things surrounded by.

By the way, lf! on a unit area! (If the molecular beam intensity is expressed as the number of gas molecules or gas atoms flying in 17 hours, the molecular beam intensity J (molecules /ct
isec ) is expressed as J −=n ov. Here, no is the molecular density of the molecular beam (molecules
/ai) is the average velocity of the molecular beam (α/5ec).

Then, the nude ion gauge 6 is connected to the molecular beam 1a.

When measuring in 2a, nude ion gauge 6
If the molecular beams 1a and 2a that have passed through do not return to the nude ion gauge 6, the molecular beams 1a and 2a involved in the measurement
The molecular density of a and 2a is nQ, but if there is a molecular beam 1a and 2a that can be handled once through the nude ion gauge 6 and return to the nude ion gauge 6 again, the molecular density is nQ. Molecular beams 1a and 2 reflected by
a is added, and the molecular density becomes nl, which is larger than no,
This change in molecular density causes deviations in measured values.

Now, the molecular beams 1a and 2'ah (Ga
(gallium), M (aluminum), etc., the molecular beams 1a, 2a are attached to the inner wall of the ultra-high vacuum chamber 9 at a temperature near or below room temperature, or to the inner wall of the cover 7 of the nude ion gauge 6. In many cases, the count can be considered to be approximately 1 (almost all of the molecular beams 1a and 2a are attached), and the measurement using the reflected molecules 1i11a and 2a (the direct deviation is expected to be small.On the other hand, A
Molecular beam 1a of nonmetallic elements such as s (arsenic) and P (phosphorus)
, 2a, molecular beam 1a for a cryosurface cooled to near the liquid nitrogen temperature.

The adhesion count of 2a is approximately 1, but the adhesion count is low on the inner wall of the ultra-high vacuum chamber 9 at a temperature near room temperature and the inner wall of the cover 7 of the nude ion gauge 6. The molecular beam 1a comes.

The suspension of 2a cannot be ignored compared to the intensity of the molecular beams 1a and 2a directly incident on the nude ion gauge 6.

Therefore, in the configuration of the conventional molecular beam epitaxial growth apparatus, when the nude ion gauge 6 is not equipped with the cover 7, the influence of the reflected molecular beam when measuring the molecular beam intensity is reduced by the cover 7. (The influence of the reflected molecular beam in this case will be described later), but the nude ion gauge 6 is placed near the crystal substrate 3 in the ultra-high vacuum chamber 9 even during the period when the molecular beam intensity is not measured. Because it is constantly exposed to molecules 1ii11a, 2a that wrap around, reactions with various molecular beams 1a, 2a may cause the lifespan of the nude ion gauge 6's filament, grid, and other components to become unknown, or electrical insulation may be defective. There are problems such as being more likely to cause

On the other hand, when the nude ion gauge 6 is equipped with the cover 7, the molecule I! An example of the 11aIll constant results is shown in FIG.

Here, 50g of As is put into the molecular beam sources 1 and 2, and the molecular beam intensity is measured by heating it to 8 degrees and 300 degrees Celsius.
Further, the cover 7 is heated by thermal radiation from the filament of the nude ion gauge 6, and the temperature reaches 100 to 200°C depending on the location. Then, using the manipulator 8, the nude ion gauge 6 is attached to the crystal substrate 3.
, open the shack 4.5, and place the molecular beam source 1.
When the molecular beam intensity of As from 2 was measured, as shown in Figure 4, the measured value of the nude ion gauge 6 changed with the rt period, rising rapidly immediately after the shutter 4.5 was opened, and then gradually decreasing. showed an increase. Therefore, it is presumed that the molecular density in the space of the nude ion gauge 6 within the cover 7 changes with time.

In addition, Fig. 5 shows another measurement example, and when the shutter 4.5 was opened under the same conditions as the example in Fig. 4, the
As in the figure, a rapid rise was observed, but after that, the indicated value of the nude ion gauge 6 showed a tendency to gradually rise and then fall.

As is clear from these measurement examples, when the cover 7 is attached to the nude ion gauge 6, the molecular beam intensity changes over time, and even if the measurement conditions are the same, the measured values differ, making it difficult to achieve good reproducibility. Difficult to measure.

That is, when the nude ion gauge 6 includes the cover 7, deterioration of the mate ion gauge 6 due to the wraparound molecular beams 1a and 2a is reduced, but the adhesion count of the incident molecular beam to the inner wall surface of the cover 7 is reduced by the molecular beam 1a. ,
It is particularly small when 2a is As, p, etc., and the measured value of the nude ion gauge 6 largely reflects the influence of the molecular beam reflected within the cover 7. In such a case, the adhesion coefficient ``)1'' measured with the nude ion gauge 6 is, in order to calculate the molecular density J expressed by J = nov above,
Special consideration is required.

Furthermore, in an actual growth apparatus, the degree of contamination on the inner wall of the cover 7 of the nude ion gauge 6 changes as the crystal growth process is repeated, and the temperature of the cover 7 changes depending on the temperature of the molecular beam cell, substrate heater, etc. Therefore, the intensity of the molecular beam introduced into the cover 7 through the opening 7a and repulsed by the inner wall of the cover 7 with respect to the incident molecular beam changes with time, making it impossible to obtain reproducibility of measured values.

(Problems to be Solved by the Invention) As mentioned above, in the conventional molecular beam epitaxy full growth apparatus, a cover is required to prevent deterioration of the nude ion gauge due to the molecular beams that wrap around. This caused ambiguity in terms of reproducibility and reliability of the measured values of molecular beam intensity, and it was difficult to accurately control the crystal growth rate, composition, etc.

The present invention was developed in view of the above-mentioned problems, and by always keeping the same level of the reflected molecular beam that returns to the nude ion gauge inside the cover, it ensures reproducibility of measurement and prevents crystal growth. It is an object of the present invention to provide an apparatus for growing the molecule l11vita4-schiffle in which time and composition can be precisely controlled.

[Structure of the invention]

(Means for Solving the Problems) The present invention involves forming a molecular beam of at least one element in a vacuum, and irradiating this molecular beam onto a crystal substrate.
In this molecular beam epitaxial growth apparatus that performs full epitaxial growth of a substance containing the above elements on a crystal substrate, a mate ion gauge is installed inside the cover to measure the intensity of the molecular beam irradiated to the crystal substrate. The cover is provided with an opening through which the molecular beam enters and is configured to be capable of cooling or heating.

(Function) The molecular beam epitaxial growth apparatus of the present invention cools or heats the cover of the nude ion gauge.
The molecular beam intensity is measured by always keeping constant the number of adhesion counts of incident molecular beams entering the cover on the inner wall of the cover, that is, the number of reflected molecular beams returning to the nude ion gauge.

(Embodiment) Hereinafter, the configuration of an embodiment of the present invention will be described with reference to FIGS. 1 and 2.

This device has basically the same structure as the device shown in FIG. 3 above, that is, molecular beams 1a, l
a pair of molecular beam sources 1.2 that generate b, a crystal substrate 3, a shutter 4.5, a nude ion gauge 6, a cover 7 having an opening 7a, a manipulator 8, and an ultra-high vacuum chamber.
9, the same %B portions will be given the same reference numerals and their explanation will be omitted.

In Fig. 1, the 7J bar 7 of the nude ion gauge 6 that measures molecular beam intensity has a double wall, and a liquid or gaseous cold sweat heating medium is introduced into the gap between the walls to cool or heat it. It is formed into a shroud structure that allows for Also, a heater 1 is installed in 9 of the j-shaped brim 7.
0 is provided, and the cover 7 can be heated if necessary.

Then, in this shrimp all-four-shuffle growth apparatus, 50g of As was put into molecular beam sources 1 and 2, and the temperature was raised to 300°C.
The molecular beam intensity was measured by heating.

At this time, the results of measuring the intensities of the molecular beams 1a and 2a without particularly cooling or heating the cover 7 are similar to the results shown in FIGS. had poor reproducibility and changed over time.

Next, liquid nitrogen was introduced into the inside of the cover 7 to cool it down, and under the same conditions molecule 1! When we measured the IT strength, we found that
The results shown in FIG. 2 were obtained.

That is, it can be seen that the measured value of the nude ion gauge 6 shows an instantaneous rise at the moment when the shutter 4.5 is opened, and remains stable thereafter. When the molecular beam intensity was measured 20 times under the same conditions, the same results shown in Figure 2 were obtained. I understand.

This is because by cooling the cover 7 with liquid nitrogen, the number of adhesion targets of the incoming molecular beams becomes substantially 1. l! 21a and 2a are not reflected by the inner wall of the cover 7, and it all depends on how many times they are worn.

Based on the results of accurately measuring the molecular beam intensity with the nude ion gauge 6 in this way, the growth time of the crystal epitaxially grown on the crystal substrate 3, IVGaAs (aluminum gallium arsenide) and 1nGaAsP (indium gallium arsenide phosphide) are determined. It is possible to precisely control the composition of compounds such as

Furthermore, when measuring with the cover 7 cooled, the flying molecules i! Since the j11a and j11a 2a essentially adhere to the inner wall of the cover 7, there is a problem in that the adhered substances accumulate over a long period of use. This is expected to result in a decrease in the cooling efficiency of the cover 7 or a short circuit in the nude ion gauge 6 due to detachment of the deposits, and therefore it is necessary to remove them by evaporation in a vacuum. Therefore, when the inner wall humidity of the cover 7 was heated to about 300° C. using the heater 10, all of the attached As could be removed within several hours. In addition,
Such heating is not necessarily caused by the heater 10.
However, the removal effect can also be achieved by electron collision using the filament of Nude Ion Gauge 6.

In addition, in the above embodiment, the cover 7 is cooled using liquid nitrogen to make the adhesion number of the molecules 11a and 2a substantially 1, so that the molecular beam intensity can be measured accurately. The cover 7 is heated to generate the molecular beam 1a.

The adhesion count of 2a is made substantially 0, and the molecular beams 1a, 2a
Even if the ions are totally reflected, the amount of reflected molecules 1i11a, 2a returning to the nude ion gauge 6 is maintained at a constant degree Celsius, so stable measurement values can be obtained and measurements with good reproducibility can be performed.

〔Effect of the invention〕

According to the present invention, by cooling or heating the cover of the nude ion gauge, the adhesion count of the incident molecular beam entering the cover on the inner wall of the cover, that is, the number of reflected molecular beams returning to the nude ion gauge is constantly monitored. The molecular beam intensity can be measured while keeping it constant, the reproducibility of measurements with a nude ion gauge is good, and the crystal growth rate, composition, etc. can be accurately controlled.

[Brief explanation of the drawing]

Fig. 1 is a block diagram showing one embodiment of the device of the present invention, Fig. 2 is a diagram of molecular beam intensity measurement in the device, Fig. 3 is a block diagram of a conventional device, and Figs. 4 and 5 respectively. It is a molecular beam intensity measurement diagram in the apparatus. 1a, 2a...Molecular beam, 3...Crystal substrate, 6...Nude ion gauge, 7...Cover, 7a...Opening. Shi\

Claims (1)

[Claims] (1) Molecular beam epitaxial growth in which a molecular beam of at least one or more elements is formed in a vacuum, and a crystal substrate is irradiated with this molecular beam to epitaxially grow a substance containing the above element onto the crystal substrate. The apparatus includes a nude ion gauge that measures the intensity of the molecular beam irradiated to the crystal substrate, and the nude ion gauge is placed inside a cover that has an opening through which the molecular beam enters and is configured to be cooled or heated. A molecular beam epitaxial growth apparatus characterized in that it is installed in a.