JPH07512B2 - Crystal growth equipment - Google Patents

Description

TECHNICAL FIELD The present invention relates to a crystal growth apparatus for a silicon single crystal or the like, which is mainly used as a semiconductor material.

[Conventional technology]

Generally, the Czochralski method (CZ method) is widely used as a crystal growth apparatus of this type. FIG. 3 is a schematic diagram showing a crystal growth apparatus by the CZ method, and 1 shows a crucible arranged in a water-cooled chamber 10 or the like. The crucible 1 is constructed by concentrically arranging a graphite outer layer container 1a and a quartz inner layer container 1b, each of which has a cylindrical shape with a bottom, into which a raw material for crystallization is charged, and this is placed around the crucible 1. Heater 2 placed in
After being heated and melted in, the pulling shaft 6 is placed in the melt L.
The seed crystal 7 hung at is soaked, and the seed crystal 7 is rotated and pulled upward to grow a single crystal 8 at the lower end of the seed crystal 7.

By the way, when a single crystal is usually used as a semiconductor substrate or the like, an impurity element is added to the melt in the crucible 1 in order to adjust the electric resistivity and electric conductivity of the single crystal.
Such impurities segregate in the pulling direction of the single crystal 8, and it is extremely difficult to maintain a uniform concentration distribution over the entire length of the single crystal 8 in the growth direction.

The impurity segregation is caused by the impurity concentration C _S in the single crystal and the impurity concentration C _L in the melt at the growth interface between the melt and the single crystal.
Due to the fact that the ratio C _S / C _L , that is, the effective segregation coefficient ke does not become 1, the impurity concentration in the melt and the impurity concentration in the single crystal change during crystal pulling due to the growth of the single crystal. It depends.

A melt layer method is known as a method for suppressing such segregation. FIG. 4 is a schematic diagram of a crystal growth apparatus using a general melt layer method. The heater 2 controls the solid layer S of the crystal raw material at the bottom of the crucible 1 and the melt layer L of the crystal raw material above it. The single crystal 8 is grown in the same manner as in the process shown in FIG.

In the pulling process of the single crystal 8, the decrease in the thickness of the molten layer due to the pulling of the single crystal 8 is supplemented by the melting of the solid layer S by the control of the heater 2 on the way to keep the volume of the molten layer L constant. An apparatus for continuously adding an impurity element during crystal pulling (referred to as a constant melt layer thickness method) to grow crystals while maintaining a constant impurity concentration in the melt (Japanese Patent Publication No. 34-8242), or an intention. By changing the volume of the melt layer,
There is an apparatus (called a melt layer thickness change method) for growing a crystal while maintaining a constant impurity concentration in a melt without adding an impurity element during crystal pulling.

The conditions for pulling a single crystal using the melt layer method without segregation of impurities can be roughly explained as follows. A one-dimensional model diagram as shown in FIGS. 5 (a) to (e) showing changes in layer thickness between the molten layer L and the solid layer S in the crucible in FIG. 4 (the left side is the upper side of the crucible). Will be explained.

FIG. 5 (a) shows the state at the end of raw material filling, and FIG. 5 (b) shows the state at the end of initial dissolution.
(D) shows the state during crystal pulling, and FIG. 5 (e) shows the state at the end of crystal pulling.

Note that, for example, even if high-purity polycrystalline silicon is used as the raw material, the impurity concentration C _P in the solid layer S is not zero (C _P ≠ 0) here.

FIG. 5 (a) shows a state immediately after the raw material is charged into the crucible, where the weight f _{P of} the raw material first charged into the crucible is 1, and the weight ratio x measured from the top surface side to the bottom surface side of the raw material is The impurity concentration at the position is C _P (x).

Now, as shown in FIG. 5 (a), the crucible is filled with the raw material for crystallization, and then the crucible is melted from its upper surface side. FIG. 5 (B) shows a state at the end of the initial melting, in which a melt layer L having a melt rate f _L (= f ₀ ) and a solid layer S of a lower solid layer f _P coexist in the upper part of the crucible. Then, pulling of the single crystal is started in this state. Note that the impurity concentration in the initial melt is C ₀ .

Next, as shown in FIG. 5C, the weight ratio (separation rate) of the molten layer L when the crystal is pulled up by f _S at the weight ratio (crystal pulling rate) with respect to the initial filling raw material weight of 1 is f _L , F _P is the weight ratio (solid content) of the raw materials remaining as solids, and C is the impurity concentration in the solid phase at the solid layer-melt layer interface at this time.
_P, also when the impurity concentration in the liquid phase and _{C L,} crystal pulling rate _{f S,} the melt rate _{f L,} below between solid ratio _{f P} (1) equation is established.

f _S + f _L + f _P = f ₀ + f _P = 1 (1) where f _S + f _S = f ₀ Next, the crystal pulling rate is changed from f _S to f from the state of FIG.
_When it is increased to _S + Δf _S , the melting rate f _L becomes f _L + Δf
_L , the solid fraction f _P changes to f _P + Δf _P ,
Further, during this period, the impurity concentration in the solid phase at the crystal-melt interface is C _S , or the impurity concentration in the melt layer is C _L + ΔC.
_{If L} is set, the impurities in the A region are preserved, so the following (2)
The formula holds.

C _L · f _L + C _a · Δf _S + C _P · Δf ₀ = C _S · Δf _S + (C _L + ΔC _L ) · (f _L + Δf _L )
(2) where C _a · Δf _{S is the} impurity addition amount, C _{a is the} unit impurity addition amount with respect to the crystallization rate, and the impurity concentration C _S in the solid phase and the impurity concentration C _{L in the} liquid phase at the crystal-melt interface. since the following equation (3) the relationship is established _{between, C S = ke · C L} ... (3) where, ke: omit secondary small term of the effective segregation coefficient (2) where ( By applying the expression (3), the following expression (4) is obtained.

Thus, in the melt layer method, the non-segregation condition is dC in equation (4).
_L / df _S = 0 and C _P = 0 are given by the following equation (5).

In the constant melt layer thickness method in which the solid layer S is melted during the crystal pulling process and the melt layer thickness is kept constant, df _L / df _S = 0, so the impurities represented by the following formula (6) are continuously added. If added, the condition of the above formula (5) can be satisfied.

C _a = keC _L = keC ₀ (6) where C ₀ : impurity concentration in the initial melt layer On the other hand, in the melt layer thickness change method, impurities are not continuously added, so C
_The molten layer thickness may be changed so as to satisfy the following formula (7) with crystal pulling so that _a = 0 and df _L ÷ df _S = −ke are satisfied.

f _L = f _L0 −kef _S (7) where f _L0 : initial melt ratio In many cases, the effective segregation coefficient ke of impurities is smaller than 1, so by selecting f _L0 = ke the segregation condition is maintained until the end. can do.

After the solid layer is completely melted, the non-segregation condition is not established and the segregation proceeds according to the following equation (8).

_{C S = keC 0 · {1-} (f S -f S a) / f L a} ke-1 ...
(8) ^{_{where, f S a, f L a}} : crystal pulling rate at the time of each solid layer is melted entirely, it is the melt index.

FIG. 6 is an explanatory view showing an example of temperature control in the molten layer thickness changing method, and FIG. 6 (a) shows a crucible and a melt in the crucible,
The dimension of the solid layer and the temperature distribution of each part are shown in Fig. 6 (b). The heat generation length of the heater, the crucible depth, the shape of the support shaft, and the material selection conditions are determined by simulation and experiment. Qualitatively, it is explained as follows.

The sum of the amount of heat supplied to the melt from the heater conductive heat through the single crystal, a quantity of heat Q _U and raw material in the crucible, the conduction heat Q _L is dissipated through the crucible axis 1c of such radiant heat from the melt surface Becomes To stabilize the crystal pulling conditions, the heat quantity Q _U
Is set to be almost constant throughout the pull-up period, and
Heater power (heating value), the following (9) is established from the quantity of heat Q _L can be regarded as constant.

However, T _m : boundary temperature between the molten liquid layer and the solid layer T ₀ : temperature at the bottom of the crucible axis λ ₆ : thermal conductivity of the solid layer S λ _c : thermal conductivity of the outer layer container λ ₃ : thermal conductivity of the inner layer container rate S _C: crucible sectional area S _P: cross-sectional area of the crucible shaft lambda _P: thermal conductivity of the crucible axis l _{P: T b} from the crucible axis length (9) of the chamber, clearing the _{T P} follows (10 ) It becomes like a formula.

In the normal crystal pulling process, the position of the melt surface is kept constant (l = constant in Fig. 6) Δl ₂ + Δl ₆ + Δl _P
= 0. Also, Since Δf _S + Δf _L + Δf _P = 0, Δf _S ∝Δl _P is established.

Therefore, the following equation (12) is obtained from equations (10) and (11).

If now are equal ease lambda _P S _P conveyed heat of heat transferred ease lambda ₆ S _C and the crucible axis of the solid layer S can maintain molten layer thickness constant, and in _{λ P S P> λ 6 S} C If so, the melt layer thickness will decrease with the pulling of the crystal, and the melt layer thickness can be controlled by appropriately setting these conduction conditions.

[Problems to be Solved by the Invention]

By the way, in the fusion layer method, as described above, (5), (6),
Theoretically, the impurity concentration distribution in the single crystal could be made uniform as long as the condition shown in the equation (7) is obeyed, and a single crystal with a uniform impurity concentration should be obtained. However, in reality, the single crystal was charged into the crucible. It is difficult to single crystallize the entire raw material. In particular, since the silicon raw material as a semiconductor material has a solid density lower than that of the melt, when the thickness of the solid layer S becomes a certain value or less, the solid layer S floats in the melt and pulls up the single crystal. This is an obstacle, and the pulling of the single crystal has to be stopped with a considerable amount of the raw material left in the crucible, which causes a problem of low raw material yield.

The present invention has been made in view of such circumstances, and an object thereof is to suppress the floating of the solid layer, which is a factor of the decrease in the yield of the crystal raw material, and to significantly improve the yield. A crystal growth apparatus is provided.

[Means for Solving the Problems]

The crystal growth apparatus according to the present invention, in a water-cooled chamber, a raw material for crystal in a crucible in which a heater and a heat insulating material are disposed around the heater and the outer periphery thereof is moved from the upper side to the lower side by the heater. An apparatus for pulling a crystal from the melt and growing it while melting toward the end is characterized in that a substance having a high thermal conductivity is interposed in a part of the heat insulating material.

[Action]

According to the present invention, the temperature of the portion facing the water-cooled chamber, which is in direct contact with the substance having a high thermal conductivity, and the portion facing the same, are relatively higher than those of the other portions. Will be reduced.

〔principle〕

As described above, in the crystal growth apparatus by the melt layer method, the crystal growth work is completed with the raw material, that is, the solid layer remaining in the crucible. At this time, the crystal production yield is the crystal pulling rate, in other words, the solid layer. It is determined by the ratio of the sum of the molten layer remaining in the crucible at the levitation limit of and the solid layer.

Therefore, if the floating limit of the solid layer S is delayed and the raw material in the crucible can be reduced, that is, crystallized, the yield can be improved. Since the floating limit of the solid layer depends on the contact area between the solid layer and the crucible layer, generally, as shown in FIG.
The interface between the solid layer S and the solid layer S has a convex shape. However, if this is made concave or flat, the contact area of the solid layer S with the crucible peripheral wall increases even if the solid layer ratio does not change, and (5)
From the conditional expressions given by the equations (7) to (7), the smaller the initial melt rate f _LO , the smaller the liquid separation rate at the floating limit of the solid layer S, in other words, the larger the crystallization rate.
Therefore, as is clear from equation (10), the temperature T ₀ below the crucible shaft
By decreasing the _{value, the} solid layer thickness l ₀ can be increased, that is, the melt rate can be reduced, and the crystallization rate can be increased.

〔Example〕

Hereinafter, the present invention will be specifically described with reference to the drawings illustrating the embodiments. FIG. 1 is a schematic vertical sectional view of the device of the present invention,
In the figure, 10 is a water-cooled chamber, 1 is a crucible, 2 is a heater, 3 is a heat retaining cylinder, 4 is a pull chamber, and 11 is a solid raw material supply device.

A crucible 1 is arranged at the center of the inside of a chamber 10 which is water-cooled, a heater 2 is provided around the crucible 1, and a heater 2 is provided around the heater 2.
A heat insulating cylinder 3 is arranged around the. A pull chamber 4 facing the top of the crucible 1 and opened / closed by a shutter 12 is provided upright on the upper wall of the chamber 1, and a pulling shaft 6 which can be raised, lowered, and rotated is hung from the pull chamber 4. .

The crucible 1 has a double structure in which an outer layer container 1a made of graphite is arranged on the outer periphery of an inner layer container 1b made of quartz, and an upper end of a shaft 1c penetrating the bottom wall of the chamber 1 is provided at the center of the bottom part thereof. It is connected and can be raised and lowered while being rotated by the shaft 1c.

The heat insulating cylinder 3 is formed into a cylindrical shape by using a material having a low thermal conductivity, for example, a carbon fiber molding material, and the crucible 1,
The heater 2 is arranged concentrically with the outer circumference of the heater 2. Coolants 3a and 3b made of a material having a high thermal conductivity, such as graphite, are annularly interposed in the middle of the heat insulating cylinder 3 in the height direction. Since the coolants 3a and 3b have high thermal conductivity, the temperature drop due to heat transfer in the surrounding water-cooled chamber 10 is generally larger than the temperature drop of the heat insulating cylinder 3 having low heat conductivity,
Therefore, the surface of the coolant on the crucible side has a temperature lower than that of the surrounding material of the heat insulating cylinder 3, absorbs heat of the peripheral wall of the crucible 1 facing this, and the solid layer S at the solid-liquid interface has the first surface at the peripheral wall portion. As shown in the drawing, it is flat, or as shown in FIG.

Rotation / elevation mechanism (not shown) from above the pull chamber 4
The upper end of a pulling shaft 5 connected to the above is introduced, and a seed crystal 7 held by a chuck is hung at the lower end thereof. The seed crystal 7 is made to adapt to the melt in the crucible 2 and then rotated. By raising the temperature while raising it, the silicon single crystal 8 can be grown on the lower end of the seed crystal 7.

The solid raw material supply device 11 is provided with a raw material feeding pipe 11b at a lower portion thereof, and a lower end thereof is exposed above a peripheral portion of the crucible, and the raw material dropped from the hopper 11a passes through the raw material feeding pipe 11b to form the crucible 1 It will be supplied inside.

Although not shown in the drawing, a feeder for impurities is also provided in the same manner.

In such an apparatus of the present invention, first, the crucible 1 is filled with the solid material for crystal, for example, high-purity polycrystalline silicon through the feeder 11.

Next, after replacing the atmosphere gas in the chamber 10 with an inert gas or the like, the electric power of the heater 2 is set to a melting condition, and the filled solid raw material is melted while an impurity supply device (not shown) is used in the melt. Add impurities.

The filled solid raw material is melted until a predetermined initial melt rate is obtained, and the molten layer L of impurities and the lower solid layer S thereof are set to coexist. In this state, the temperature is slightly lowered by the coolants 3a and 3b facing the peripheral surface of the crucible 1, so that the solid layer surface at the interface between the molten layer and the outer layer container has a concave shape in which the periphery in contact with the peripheral wall of the crucible 1 is raised.

When the initial melting is completed, the pulling of crystals is started, and the decrease in the thickness of the molten layer due to this is supplemented by melting the solid layer S by the heater 2.

The pulled single crystal 8 is passed through the shutter 12 to pull chamber 4.
After retracting inside, the shutter 12 is closed. Feeder 11 again
Then, the solid layer raw material is charged into the crucible 1, the solid raw material is melted by the control of the heater 2, and the above process is repeated. The single crystal 8 pulled into the pull chamber 4 is
It is taken out by removing.

FIG. 2 is a partial sectional view showing another embodiment of the present invention,
In this embodiment, the number of the coolant 3a provided in the heat insulating cylinder 3 is one. Other configurations and operations are substantially the same as those of the embodiment shown in FIG. 1, and corresponding parts are designated by the same reference numerals and the description thereof will be omitted.

(Numerical example)

Inner diameter as crucible: 150 mm, depth 200 mm, thickness 4 mm made of quartz inner layer container, graphite outer layer container, also made of graphite crucible shaft, 7 kg of polycrystalline silicon as raw material, and segregation to silicon as impurities Coefficient ke
= 0.35 phosphorus was used to pull up the crystal by using the crystal growth apparatus as shown in Fig. 1 by the melt layer thickness variation method.

As a result, in the conventional apparatus, the solid layer floated in the molten layer at the crystallization rate f _S = 0.62, but in the apparatus of the present invention, the floating of the solid layer was not observed up to the crystallization rate f _S = 0.71. The electrical resistivity ρ in the pulling axis direction of the crystal obtained by the device of the present invention was measured. The segregation prevention effect was also confirmed. As a result, the yield of products has improved significantly.

〔effect〕

As described above, in the device of the present invention, the floating limit of the solid layer can be delayed, the crystallization rate can be increased, and segregation can be prevented at the same time, so that the yield of the product can be improved. Is.

[Brief description of drawings]

FIG. 1 is a schematic vertical sectional view of the device of the present invention, FIG. 2 is an explanatory view showing the shape of a solid layer in a crucible, FIGS. 3 and 4 are explanatory views of the Czochralski method and the molten layer method, and FIG. FIG. 6 is an explanatory view showing the phase change of the filling material in the melt layer method, and FIG. 6 is an explanatory view showing the contents of the temperature control of the crucible. 1 ... crucible, 2 ... heater, 3 ... heat retaining tube, 3a, 3b ...
Coolant, 4 ... Pull chamber, 6 ... Pulling shaft, 7 ...
Seed crystal, 8 …… single crystal, 10 …… chamber, 11 …… solid source feeder

Claims (1)

[Claims] 1. A raw material for crystallization in a crucible in which a heater and a heat insulating material located outside of the heater are disposed in a chamber which is water-cooled, is melted from the upper side to the lower side by the heater. At the same time, in a device for pulling up crystals from the melt and growing them, a substance having a high thermal conductivity is interposed in a part of the heat insulating material.