JPS626643B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an improved method for supplying a source molecular beam onto a substrate in vacuum and growing a thin film containing the source molecules on the substrate. More specifically, the present invention relates to a molecular beam epitaxy method (hereinafter referred to as MBE) in which a molecular beam is supplied onto a crystalline substrate to grow a thin film crystal on the substrate.

In MBE, the radiation source and the substrate are thermally and geometrically separated and independent from each other, the number and type of radiation sources can be arbitrary, and the temperature of each can be controlled independently. Material molecules for crystal growth are transported unidirectionally from a radiation source onto a substrate as molecules traveling straight through an ultra-high vacuum. Therefore, by independently controlling the intensities of a plurality of molecular beams, a crystal having a composition corresponding to the intensity of each material molecular beam hitting the substrate can be produced with good controllability.

Conventionally, as a method for accurately controlling the molecular beam intensity, a control system as shown in FIG. 1, for example, has been used. That is, a plurality of molecular beam sources (hereinafter referred to as radiation sources) are installed in the ultra-high vacuum layer 1 schematically indicated by the dotted line in the figure.
A temperature-controlled substrate 4 is irradiated with a molecular beam from the radiation source, and a part of the molecular beam is introduced into an ionization section 5 of a mass spectrometer, and passes through a mass spectrometry section 6. The detection unit 7 converts the ions into electrical signals, and the multiplexer 8 sends signals corresponding to each molecular beam intensity to the temperature control devices 9 and 10 of each radiation source.
A method of controlling the temperature of each radiation source 2, 3, etc. is adopted. Since this method directly detects the molecular beam intensity of each radiation source and controls the radiation source temperature, it is theoretically possible to construct an extremely accurate control system. For this purpose, it is necessary to establish the assumption that the gain of the feedback system is always unchanged from the detection of the molecular beam intensity.

However, the conventional method has the disadvantage that the output signal of the mass spectrometer decreases as the device is used. The main reason for this is that the element incident as a molecular beam adheres to the surface of the electron multiplier in the detection section 7, thereby reducing the gain of the multiplier. In order to minimize this, as shown in Figure 1, the incident direction of the molecular beam and the direction of the main axis of the mass spectrometer are arranged almost perpendicularly, and the area around the mass spectrometer is cooled with liquid nitrogen. Care is taken so that only the molecular beam incident through the hole 12 in the shroud is introduced into the ionization section 5. In this way, only the elements ionized in the ionization section 5 are introduced into the mass spectrometry section 6, and only the ions that have passed through the analysis section 6 can reach the detection section 7. Only ions fly to the surface of the electron multiplier of the detection unit 7, and all others adhere to the inner wall of the shroud 11 cooled with liquid nitrogen. Therefore, in FIG. 1, it was thought that contamination of the surface of the electron multiplier by substances incident as molecular beams could be kept to a minimum.

This can be said to be true if the following two assumptions hold. That is, (i) Molecules that enter through the hole 12 as a molecular beam adhere to the inner surface of the liquid nitrogen cooling shroud with an adhesion coefficient of 1, regardless of the type of the molecule. (ii) Once attached, molecules will not evaporate again even if the temperature inside the shroud rises to room temperature.

However, the above assumption does not hold true for molecular beams of group V elements with high vapor pressure.

In order to explain the reason for this, first - group V compounds
We will discuss MBE, followed by the properties of group V elements. - In MBE of group V compounds, the adhesion coefficient of group atoms to the substrate is 1, and
When a free group atom exists on the substrate, the group V molecule combines with that group atom and is incorporated into the crystal, but when there is no free group atom, the group V molecule alone does not attach to the substrate and remains in vacuum. Re-released. Therefore, when growing group V crystals,
It is necessary to supply on the substrate a sufficiently larger amount of group element molecules than group atoms. Therefore, the vacuum pressure of the atmosphere during crystal growth is mostly occupied by the vapor of the group molecules.

If the surface of the liquid nitrogen cooling shroud has an adhesion coefficient of 1 for molecules incident on it, regardless of the type of molecule, the pumping speed will be approximately 10 l/sec cm ² . When the effective displacement of the system is calculated from the amount of As consumed in MBE and the pressure during crystal growth, the adhesion coefficient of _four As molecules to the liquid nitrogen cooling surface is smaller than 1. Furthermore, even after crystal growth is complete, a large amount of excess group molecules released during crystal growth remain in vacuum for a long time because their vapor pressure is high and they are not effectively evacuated by titanium getter pumps or ion pumps. , as will be described later, causes contamination of the vacuum system and has an extremely adverse effect on the electron multiplier.

This is mainly due to the difference in the properties of the two allotropes of the group elements. There are two types of group elements: stable at room temperature and unstable at room temperature (metastable).
There are allotropes of a species. When a substance that is stable at room temperature is heated and evaporated in a vacuum, it is released as metastable molecules. The metastable group molecules thus released into the vacuum have extremely high vapor pressure, activity, and toxicity compared to substances that are stable at room temperature. Further, a metastable substance gradually changes to a room temperature stable substance at room temperature, and the rate of change is P<<As<Sb. In particular, when using phosphorus, metastable yellow phosphorus, which is obtained by heating red phosphorus, which is stable at room temperature, and evaporating, remains in a vacuum in a metastable state for a long time, so after heating is stopped, However, the pressure within the system did not subside. For example, when a -V group compound such as GaP is heated and evaporated, the group element evaporates in the form of _P2 , which also gradually changes to metastable yellow phosphorus.

As mentioned above, crystals using V group, especially P,
When fabricating by MBE, the vacuum chamber is filled with active and highly toxic metastable _P4 molecules with high vapor pressure for a long time, and once the vacuum chamber is exposed to the atmosphere, the phosphorus is immediately oxidized and becomes extremely toxic. The hygroscopic P ₂ O ₃ and P ₂ O ₅ become phosphoric acid due to moisture in the air, and the walls of the vacuum chamber become sticky and contaminated, which is the main cause of changing the gain of the electron multiplier. It was becoming. Moreover, such contamination occurs not only with P but also with arsenic and antimony, which are also group V elements, although there are differences in degree, and countermeasures have been desired.

By the way, the inventors discovered that the adhesion coefficient of group V molecules to a surface kept at a relatively low temperature (temperature range varies depending on the material) changes greatly depending on whether group atoms are present or absent on the image. Even when the adhesion coefficient of group V molecules alone is extremely small, by simultaneously supplying group atoms onto the surface, the adhesion coefficient increases significantly, and as a film with an excess of group V than the stoichiometric ratio, V
Patent application No. 53-106790 for stably fixing group molecules
It was revealed in the issue. Therefore, by providing a surface on which group atoms are deposited in vacuum during and after crystal growth, excess group V molecules existing as back pressure can be solidified as a chemically stable compound.

To further explain this situation with reference to a specific example, in the apparatus shown in FIG. When the heating of the source is stopped, the supply of molecules from the radiation source disappears, and the metastable _As4 molecules that filled the vacuum chamber are transferred to the liquid nitrogen cooling shroud 11.
As the metastable molecules are condensed on top, the pressure is temporarily lowered. However, as the temperature of the liquid nitrogen cooling shroud rose, _As4 condensed on the cooling surface.
Reevaporation of molecules occurs and the pressure increases again. As time passes, the re-evaporated As ₄ molecules gradually adhere to the room-temperature walls as room-temperature-stable metallic arsenic, causing the pressure in the vacuum chamber to drop again. This situation is shown at 21 in FIG. On the other hand, after the completion of crystal growth, only the Ga source 2 was heated for about 1 hour to deposit Ga on the wall of the vacuum chamber, and the subsequent pressure change was as shown at 22 in Figure 2, and the pressure increase was on the Ga deposition surface. It was about 1/10 of what it would be without it. This is because Ga atoms combine with ₄ As molecules to form a stable compound in the form of GaAsx, and we can consider the getter effect of ₄ As molecules due to the Ga-deposited surface.

Based on these results, the present invention aims to minimize the decrease in the gain of the electron multiplier of the mass spectrometer when performing MBE using active substances with high vapor pressure as materials. It is characterized by providing a surface near the meter, preferably near the electron multiplier, on which a stabilizing substance that can solidify the vapor of these substances as a stable substance with a lower vapor pressure is deposited. .

Next, an embodiment of the present invention will be described with reference to FIG.
31 is an ionization section of the mass spectrometer, and an analysis section and a detection section (not shown) are arranged on the right side thereof.
These are surrounded by a liquid nitrogen cooling shroud 32;
The molecular beam enters through a hole 33 made in the shroud. A small chamber 34 for stabilization is provided in front of the ionization section 31, and a Ga molecular beam source 3 for stabilization is installed inside.
Install 5. 36 was evaporated in a stabilization chamber.
This is a chevron that prevents Ga atoms from flying toward the ionization section 31. The stabilization chamber 34 is cooled by conduction from the shroud and is also provided with a Ga source 35.
effectively spread Ga over as wide an area as possible in the stabilization chamber.
is placed in such a position that it can be vapor-deposited. The stabilizing radiation source 35 is heated during or after the crystal growth, or both, to deposit Ga on the inner surface of the stabilizing chamber 34 and remove most of the As ₄ molecules incident on the mass spectrometer.
It was possible to condense it on the inner surface of the stabilization chamber 34 in the form of GaAsx. As a result, the gain of the electron multiplier can be kept stable over a long period of time, and the molecular beam intensity can be easily controlled precisely, making it possible to grow stable films over a long period of time.

Note that the stabilizing radiation source of the present invention is not limited to Ga, and other group elements or other elements can be used.

To enumerate the effects of the present invention again, (1) By vapor-depositing group elements in the stabilization chamber during and after film growth, group V molecules with high vapor pressure and low adhesion coefficient can be concentrated. The proportion of Group V molecules adhering to the surface of the electron multiplier of the mass spectrometer can be minimized. As a result, (2) the gain of the electron multiplier can be kept stable over a long period of time, and (3) stable film growth can be achieved over a long period of time. In addition, (4) the possibility of problems due to poor insulation due to the adhesion of group V molecules is also reduced.

Furthermore, the present invention can be applied not only to -V group thin film growth but also to -group and -group thin film growth by appropriately selecting a stabilizing substance, and can extend the life of the electron multiplier. .

[Brief explanation of the drawing]

Figure 1 is a schematic layout diagram of the conventional MBE equipment, and Figure 2 shows the GaAs crystal growth after GaAs crystal growth.
FIG. 3 is a diagram illustrating a configuration in which a stabilization chamber is attached to a mass spectrometer according to the present invention. 1... Ultra-high vacuum chamber, 2, 3... Radiation source, 4... Substrate, 31... Ionization section, 32... Cooling shroud, 33... Hole, 34... Stabilization chamber, 35... Stabilization Ga source for use.

Claims (1)

[Claims] 1. When a molecular beam of a material containing constituent elements of a thin film to be grown is supplied onto a substrate placed in a vacuum chamber and deposited on the substrate to grow a thin film. , in the vicinity of a mass spectrometer installed in the vacuum chamber to measure the intensity of the molecular beam of the material, vapor generated by evaporating the material or evaporated substances adheres to the surface in the vacuum chamber. A surface is deposited with a stabilizing substance that chemically combines with the back pressure vapor generated by re-evaporation to convert the material into a chemically stable solid with a lower vapor pressure. A method for growing a thin film, comprising: solidifying on the surface. 2. At least one of the materials consists of an allotrope of an element that has two or more allotropes among the elements constituting the thin film, and the vapor pressure of another allotrope obtained by evaporating the above-mentioned one allotrope is higher than the vapor pressure of this one allotrope. 2. The thin film growth method according to claim 1, wherein: 3. A patent claim characterized in that the material substance consists of a compound containing the thin film constituent element to be grown, and the vapor pressure of the substance obtained by evaporating the compound is higher than the vapor pressure of this compound. The thin film growth method according to scope 1. 4. The thin film growth method according to claim 1, wherein the stabilizing substance is composed of a simple substance of a thin film constituent element. 5. The thin film growth method according to claim 1, wherein at least one of the back pressure vapors is a group V element and the stabilizing substance is a group element. 6. The thin film growth method according to claim 1, wherein the stabilizing surface is a cooling surface. 7. A substrate placed in a vacuum, a radiation source that supplies a molecular beam of a material containing the constituent elements of the thin film to be grown on this substrate to grow the thin film, and a source for measuring the intensity of the molecular beam. A mass spectrometer is installed near the mass spectrometer, and back pressure vapor is generated by evaporating the material or by re-evaporating the evaporated material after it adheres to the surface in the vacuum chamber. and a stabilizing surface deposited with a stabilizing substance that combines with the substance to convert the substance into a more stable chemically stable solid with a low vapor pressure.