JP6017482B2 - Method for producing hydrogenated chlorosilane and catalyst for producing hydrogenated chlorosilane - Google Patents

Description

  The present invention relates to a technology for producing hydrogenated chlorosilanes as raw materials for producing polycrystalline silicon and the like, and more particularly, to a method for producing hydrogenated chlorosilanes using metal silicon particles containing a metal reaction accelerator as a tetrachlorosilane reduction catalyst. .

  High-purity polycrystalline silicon is used as a raw material for manufacturing silicon substrates for semiconductor devices or solar cells. This high-purity polycrystalline silicon is generally produced by a Siemens method using hydrogenated chlorosilane, particularly trichlorosilane purified to a high purity, as a raw material gas. In this method, a raw material gas is supplied into a reactor in a hydrogen atmosphere, and polycrystalline silicon is deposited on a silicon core wire heated to a high temperature.

  Such a precipitation reaction produces a large amount of tetrachlorosilane as a by-product. For this reason, in order to increase the utilization efficiency of silicon raw materials, it is necessary to provide a process for separating and recovering tetrachlorosilane from exhaust gas and reducing it to trichlorosilane or the like.

  When dichlorosilane or trichlorosilane is synthesized from metal silicon, generally, a method of reacting metallurgical grade metal silicon with hydrogen chloride is used. Also in this reaction, tetrachlorosilane is by-produced. For this reason, when incorporating the reaction of metallic silicon and hydrogen chloride as a trichlorosilane supply process for producing high-purity silicon, the by-product tetrachlorosilane is separated and recovered from trichlorosilane in order to increase the utilization efficiency of silicon raw materials. A process of reduction is required.

  Various reactions have been proposed as a method for reducing tetrachlorosilane to hydrogenated chlorosilane. (1) Gas phase reaction of tetrachlorosilane and hydrogen at about 900 to 1100 ° C. (2) Reaction promotion It is roughly classified into a method in which hydrogen and tetrachlorosilane are reacted at about 350 to 600 ° C. on the surface of metal silicon particles containing an agent. All of these methods are disclosed in JP-T-2004-532786 (Patent Document 1).

  Among these methods, there are various methods for reacting hydrogen and tetrachlorosilane on the surface of metal silicon particles containing a reaction accelerator. In many cases, a copper compound is used as the reaction accelerator.

  For example, in the method disclosed in Japanese Patent Laid-Open No. 56-73617 (Patent Document 2), copper particles are used as an accelerator. In the method disclosed in JP-A-60-36318 (Patent Document 3), copper (I) chloride is used as an accelerator. Furthermore, Japanese Patent Laid-Open No. 9-235114 (Patent Document 4) points out the problem that aggregation occurs when the catalyst is activated in a hydrogen atmosphere after putting metal silicon and copper or a copper salt in the fluidized tank. And proposes the use of a catalyst in which copper silicide is formed on the surface of metal silicon grains.

  In addition to these, a method of adding a metal other than copper as a reaction accelerator has been proposed. For example, JP-A-10-29813 (Patent Document 5) reports that when a reaction accelerator containing iron and aluminum is used in addition to copper, the reduction reaction of tetrachlorosilane can be performed stably over a long period of time. Has been made.

JP-T-2004-532786 JP-A-56-73617 JP-A-60-36318 JP-A-9-235114 Japanese Patent Laid-Open No. 10-29813

  Thus, in the conventional method, copper is used as a reaction accelerator, but the effect of such a reaction accelerator tends to decrease with the reaction time. Such a decrease in the effect of the reaction accelerator is presumably because copper chloride is regenerated from the copper silicon compound that is the catalytically active center and discharged out of the system. Therefore, when the reduction reaction is continued for a long time, it is necessary to add a reaction accelerator in the middle of the process. That is, when copper is used as a reaction accelerator, it is essential to frequently perform activity management and the like.

  The present invention has been made in view of such problems of the conventional method. The object of the present invention is to stably carry out a reaction for producing hydrogenated chlorosilane by reducing tetrachlorosilane over a long period of time. It is an object of the present invention to provide a novel reaction accelerator that can make hydrogenated chlorosilanes more stable.

  In order to solve the above problems, a method for producing a hydrogenated chlorosilane according to the present invention is a method for producing hydrogenated chlorosilane by reducing tetrachlorosilane in a hydrogen gas atmosphere, wherein tetrachlorosilane is converted to hydrogenated chlorosilane. As the conversion reaction accelerator, a reaction accelerator containing an iron component, a calcium component, and a nickel component is used.

  For example, the reaction accelerator includes metallic iron or an iron compound as an iron component, metallic calcium or a calcium compound as a calcium component, and metallic nickel or a nickel compound as a nickel component.

  For example, the reaction accelerator is prepared by adding the iron component, the calcium component, and the nickel component to metal silicon particles, and is used as a catalyst for reducing tetrachlorosilane to hydrogenated chlorosilane. .

  Preferably, the amount of the iron component, the calcium component, and the nickel component contained in the metal silicon particles is 0.1% of the iron component with respect to the whole metal silicon particles including the reaction accelerator. -5 mass%, the calcium component is 0.1-5 mass%, and the nickel component is 400-3000 mass ppm.

  In one embodiment, the catalyst is prepared by heating the metal silicon particles together with hydrogen gas in a reaction vessel to remove the surface oxide film of the metal silicon particles, and further in a hydrogen atmosphere, the iron component, the calcium component, And it is obtained by adding the nickel component and heating to activate the catalyst.

  In another aspect, the catalyst contains the metal silicon particles, the iron component, the calcium component, and the nickel component in a reaction vessel, and is heated in a hydrogen atmosphere to heat the metal silicon particles. The surface oxide film was removed and the catalyst was activated at the same time.

  In these embodiments, for example, the hydrogen atmosphere includes tetrachlorosilane.

  Preferably, the reaction temperature when reducing the tetrachlorosilane is 350 to 600 ° C.

  The method for activating metal silicon particles for producing hydrogenated chlorosilane according to the present invention is used as a catalyst for reducing tetrachlorosilane to hydrogenated chlorosilane using a reaction accelerator containing an iron component, a calcium component and a nickel component. Activate the metal silicon grains.

  The catalyst for producing hydrogenated chlorosilane according to the present invention contains a reaction accelerator containing an iron component, a calcium component and a nickel component in the metal silicon particles as a reaction accelerator for conversion from tetrachlorosilane to hydrogenated chlorosilane. ing.

  The catalyst for reducing tetrachlorosilane according to the present invention to hydrogenated chlorosilane is prepared by adding an iron component, a calcium component, and a nickel component as reaction accelerators to metal silicon particles. Since the activity of this catalyst is maintained for a long time, the production process of hydrogenated chlorosilane becomes more stable. Even if a reaction accelerator is added to the system during the reduction reaction, a small amount of addition is sufficient.

  In the present invention, when a hydrogenated chlorosilane is produced by reducing tetrachlorosilane in a hydrogen gas atmosphere, a reaction accelerator containing an iron component, a calcium component and a nickel component is used as a reaction accelerator for the conversion from tetrachlorosilane to hydrogenated chlorosilane. Use the agent. When this catalyst system is used, usually no copper component is added.

  Such a reaction accelerator includes, for example, metallic iron or an iron compound as an iron component, metallic calcium or a calcium compound as a calcium component, and metallic nickel or a nickel compound as a nickel component.

  Practically, for example, the above-described reaction accelerator is prepared by adding an iron component, a calcium component, and a nickel component to metal silicon particles, and is used as a catalyst for reducing tetrachlorosilane to hydrogenated chlorosilane. .

  In producing hydrogenated chlorosilane using the reaction accelerator according to the present invention, when a large amount of tetrachlorosilane is treated in a practical plant, a fluidized tank is used for the reduction reaction as disclosed in Patent Document 4. It is preferable.

  The metal silicon particles containing the metal component are not particularly limited as long as the metal content is a metallurgy grade having a silicon content of 98% or more. Preferably, those whose particle size and particle size distribution are adjusted by mechanical grinding are used. In order to improve the flow in the fluid tank, the average particle size of the metal silicon particles is preferably adjusted to 100 to 300 μm 1.

  Each of the iron component, calcium component, and nickel component added as a reaction accelerator can be prepared into metal silicon particles with each element remaining as a single metal. In this case, it is preferable that each single metal is pulverized and prepared (added) as a granular material. The particle size at the time of addition is preferably 300 μm or less, and may be a fine powder of several microns or less so that good mixing can be obtained.

  An iron component, a calcium component, and a nickel component may be added as an iron compound, a calcium compound, or a nickel compound, and respective oxides or metal salts can be used. Of these, metal halide salts are particularly preferred because they are easily activated. Specifically, iron (II) chloride or iron (III) chloride, calcium chloride, nickel (II) chloride or nickel (III) chloride can be exemplified. Since these metal salts are easily available as powders or those having a small particle size, mechanical mixing with metal silicon particles is easy.

  The amount of the iron component, calcium component, and nickel component contained in the metal silicon particles is 0.1% of the iron component (iron element) with respect to the whole metal silicon particles including the reaction accelerator (100 mass%). It is preferable that -5 mass%, a calcium component (calcium element) is 0.1-5 mass%, and a nickel component (nickel element) is 400-3000 mass ppm.

  More preferably, the amount of the iron component, calcium component, and nickel component contained in the metal silicon particles is 1 to 5 mass% for the iron component (iron element) and 1 to 3 mass for the calcium component (calcium element), respectively. %, Nickel component (nickel element) is 1000 to 2000 mass ppm.

  A catalyst prepared by adding an iron component, a calcium component, and a nickel component to metal silicon particles can be activated as a catalyst under the conditions for reducing tetrachlorosilane in a fluidized tank. There are also advantages.

  The catalytic activity when reducing tetrachlorosilane to hydrogenated chlorosilane in a hydrogen atmosphere is thought to be due to the silicidation reaction accelerator. This activation is performed under the conditions under which tetrachlorosilane is reduced. It can be done. That is, by using the catalyst according to the present invention, the reduction reaction of tetrachlorosilane and the activation of the catalyst can proceed simultaneously. In the case of the catalyst according to the present invention, unlike the conventional catalyst using a copper component as a reaction accelerator, poor fluidity hardly occurs in this activation process.

  When each of the iron component, the calcium component, and the nickel component is added in the form of a metal chloride, each metal silicide is formed by heating in a reducing atmosphere with hydrogen gas. At this time, even if the reducing gas is allowed to flow at a high temperature, a part thereof does not disappear as in the case of the copper salt. That is, in the catalyst preparation process according to the present invention, as described in Patent Document 4, it is not necessary to perform the preparation once under different conditions.

  The heating for forming each metal silicide may be 250 ° C. or more and 350 ° C. or less as in the method disclosed in Patent Document 4 using a copper salt, but 350 to 350 corresponding to the reduction reaction temperature of tetrachlorosilane as it is. It can also be performed in a temperature range of 600 ° C.

  The activation of such a catalyst is carried out by heating the metal silicon particles together with hydrogen gas in the reaction tank to remove the surface oxide film of the metal silicon particles, and further in the hydrogen atmosphere, the iron component, calcium component, Further, it can be achieved by adding a nickel component and heating to form a silicide compound.

  Alternatively, after filling the reaction vessel with metal silicon particles, iron component, calcium component, and nickel component, heating in a hydrogen atmosphere to remove the surface oxide film of metal silicon particles and activate the catalyst (Silicon compound generation) may be performed simultaneously.

  That is, in this invention, the metal silicon particle used as a catalyst at the time of reduce | restoring tetrachlorosilane to hydrogenated chlorosilane is activated using the reaction accelerator containing an iron component, a calcium component, and a nickel component.

  In any of the above cases, each metal component charged into the reaction vessel can be sufficiently mixed by flowing with an air flow in the fluidized bed.

  As the gas used at the time of activation, nitrogen or argon may be added in addition to hydrogen, but in actual plants, only hydrogen is usually used.

  Here, since the activation of the metal silicon containing iron, calcium, and nickel proceeds rapidly, the activation may be performed by heating in an atmosphere containing only hydrogen. Later, hydrogen gas containing tetrachlorosilane may be supplied with heating to start the reduction reaction immediately.

  When the inside of the fluidized tank filled with the catalyst thus prepared is maintained at 350 to 600 ° C. and tetrachlorosilane is introduced together with hydrogen, hydrogenation of tetrachlorosilane occurs, and hydrogenated chlorosilane, particularly trichlorosilane, is highly efficient. Is obtained. The ratio of hydrogen and tetrachlorosilane introduced at this time is preferably 2: 1 to 5: 1.

  When a conventional reaction accelerator is used, it is necessary to frequently add a reaction accelerator to the reaction system in order to maintain the reaction activity. However, since the activity of the reaction accelerator of the present invention is maintained for a long time, the production process of hydrogenated chlorosilane becomes more stable. In addition, the metal silicon to be added also contains a trace amount of each metal component, and it is thought that each component is concentrated to the concentration required for activity by long-term operation, but the reaction is promoted during the reduction reaction. Even when the agent is added to the system, the addition is small.

  The activation of metallic silicon by a combination of iron, calcium and nickel can be similarly applied to the production of hydrogenated chlorosilanes and chlorosilanes by the reaction of metallic silicon and hydrogen chloride.

  A test fluid tank made of SUS316 (50A <ID φ49.5 mm> × 100 cm) was used as a reactor, and the following examples and comparative examples were examined.

[Comparative Example 1]
After putting 500 g of metal silicon particles into the reactor and replacing with nitrogen, the pressure was increased to 3.0 MPa while supplying hydrogen gas, and the mixture was heated to 500 ° C. and left for 1 hr.

Thereafter, hydrogen gas was flown in an amount of 0.5 Nm ³ / hr, tetrachlorosilane was supplied at 1.85 kg / hr so that the molar ratio in the reactor was 2, and the product discharged after 1 hour of operation. The fraction of trichlorosilane in the gas was analyzed by gas chromatography.

  As a result, the reaction rate was 26.3 mol%.

  Although it is known that there is almost no reaction activity when high-purity metallic silicon is used as a catalyst, the reaction rate is considered to be due to transition metal impurities contained in the used metallic silicon. Therefore, the value of the reaction rate due to impurities contained in the metal silicon used in Comparative Example 1 was defined as the background (blank).

[Example 1]
After putting 400 g of metal silicon particles into the reactor and replacing with nitrogen, the pressure was increased to 3.0 MPa while supplying hydrogen gas, and the mixture was heated to 500 ° C. and left for 1 hr.

Thereafter, hydrogen gas was flowed in an amount of 0.5 Nm ³ / hr, and 1.85 kg / hr of tetrachlorosilane was supplied so that the molar ratio in the reactor became 2, and after standing for 1 hour, the reactor was put into the reactor. 100 g of metallic silicon, 15 g of metallic iron, 5 g of metallic calcium, and 0.75 g of metallic nickel were mixed and supplied.

  After 1 hour of operation, the fraction of trichlorosilane in the discharged product gas was analyzed by gas chromatography.

  As a result, the reaction rate was 30.3 mol%.

  That is, a higher value was shown compared to the background (reaction rate 26.3 mol%).

[Comparative Example 2]
After putting 400 g of metal silicon particles into the reactor and replacing with nitrogen, the pressure was increased to 3.0 MPa while supplying hydrogen gas, and the mixture was heated to 500 ° C. and left for 1 hr.

Thereafter, hydrogen gas was flowed in an amount of 0.5 Nm ³ / hr, and 1.85 kg / hr of tetrachlorosilane was supplied so that the molar ratio in the reactor became 2, and after standing for 1 hour, the reactor was put into the reactor. 100 g of metallic silicon, 15 g of metallic iron and 5 g of metallic calcium were mixed and supplied.

  After 1 hour of operation, the fraction of trichlorosilane in the discharged product gas was analyzed by gas chromatography.

  As a result, the reaction rate was 26.3 mol%.

  That is, when the nickel component was not included, the value was equal to the background (reaction rate 26.3 mol%).

[Comparative Example 3]
After putting 400 g of metal silicon particles into the reactor and replacing with nitrogen, the pressure was increased to 3.0 MPa while supplying hydrogen gas, and the mixture was heated to 500 ° C. and left for 1 hr.

Thereafter, hydrogen gas was flowed in an amount of 0.5 Nm ³ / hr, and 1.85 kg / hr of tetrachlorosilane was supplied so that the molar ratio in the reactor became 2, and after standing for 1 hour, the reactor was put into the reactor. 100 g of metallic silicon, 5 g of metallic calcium, and 0.75 g of nickel were mixed and supplied.

  After 1 hour of operation, the fraction of trichlorosilane in the discharged product gas was analyzed by gas chromatography.

  As a result, the reaction rate was 26.3 mol%.

  That is, when no iron component was included, the value was equal to the background (reaction rate 26.3 mol%).

[Example 2]
After putting 400 g of metal silicon particles into the reactor and replacing with nitrogen, the pressure was increased to 3.0 MPa while supplying hydrogen gas, and the mixture was heated to 500 ° C. and left for 1 hr.

Thereafter, hydrogen gas was flowed in an amount of 0.5 Nm ³ / hr, and 1.85 kg / hr of tetrachlorosilane was supplied so that the molar ratio in the reactor became 2, and after standing for 1 hour, the reactor was put into the reactor. 100 g of metallic silicon, 15 g of metallic iron, 5 g of metallic calcium, and 0.75 g of metallic nickel were mixed and supplied.

  The operation was carried out for 10 days while supplying only metallic silicon as appropriate. During this period, the fraction of trichlorosilane in the discharged product gas was analyzed by gas chromatography.

  The reaction rate immediately after the start of the reaction was 30.1 mol%, and the reaction rate after 10 days of operation was 30.0 mol%.

  That is, it showed a high value compared with the above background (reaction rate 26.3 mol%), and no significant decrease was observed in the value even after long-time operation.

  Moreover, when the catalyst inside the reaction furnace was taken out after the operation was stopped, the mass of the entire metal silicon particles was 520 g. About 30% by mass of nitric acid (about 10 ml) was added to a part (0.5 g) of the metal silicon particles, and about 5 ml of hydrofluoric acid was added dropwise to complete dissolution. When this solution was analyzed by ICP-MASS (device name: 7500CS manufactured by Agilent), the content of iron, calcium, and nickel with respect to the entire metal silicon particles was 0.28% by mass for iron and 0.95 for calcium. The mass% and nickel were 1400 mass ppm.

[Comparative Example 4]
After putting 400 g of metal silicon particles into the reactor and replacing with nitrogen, the pressure was increased to 3.0 MPa while supplying hydrogen gas, and the mixture was heated to 500 ° C. and left for 1 hr.

Thereafter, hydrogen gas was flowed in an amount of 0.5 Nm ³ / hr, and 1.85 kg / hr of tetrachlorosilane was supplied so that the molar ratio in the reactor became 2, and after standing for 1 hour, the reactor was put into the reactor. 100 g of metallic silicon, 15 g of metallic iron, 5 g of metallic calcium, and 5 g of metallic copper were mixed and supplied.

  The operation was carried out for 10 days while supplying only metallic silicon as appropriate. During this period, the fraction of trichlorosilane in the discharged product gas was analyzed by gas chromatography.

  The reaction rate immediately after the start of the reaction was 30.1 mol%, but the reaction rate after 10 days of operation was reduced to 26.5 mol%.

  That is, although the initial value showed a higher value than the background (reaction rate 26.3 mol%), the reaction rate was significantly reduced during long-time operation.

  Moreover, when the catalyst inside the reaction furnace was taken out after the operation was stopped, the mass of the entire metal silicon particles was 520 g. When a part (0.5 g) of the metal silicon particles was analyzed in the same manner as in Example 2 described above, the content of iron, calcium, and copper with respect to the entire metal silicon particles was 0.28% by mass of iron. The calcium content was 0.95% by mass and the copper content was 0.03% by mass.

  This result indicates that the copper disappears significantly during the reaction process.

As described above, as a reaction accelerator for the conversion of tetrachlorosilane to hydrogenated chlorosilane, a catalyst for producing hydrogenated chlorosilane, in which a reaction accelerator containing an iron component, a calcium component, and a nickel component is contained in metal silicon particles, When used, the reaction when tetrachlorosilane is reduced to produce hydrogenated chlorosilane is stable over a long period of time, thereby making the production of hydrogenated chlorosilane more stable.
[Example 3]

  After putting 500 g of metal silicon particles into the reactor and replacing with nitrogen, the pressure was increased to 3.0 MPa while supplying hydrogen gas, and the mixture was heated to 500 ° C. and left for 1 hr.

Thereafter, hydrogen gas was flowed in an amount of 0.5 Nm ³ / hr, and 1.85 kg / hr of tetrachlorosilane was supplied so that the molar ratio in the reactor would be 2, and then only metal silicon was supplied as appropriate. Then, the operation was performed for 15000 hours, and during this time, the fraction of trichlorosilane in the discharged product gas was analyzed by gas chromatography.

  Although the reaction rate immediately after the start of the reaction was 26.4 mol%, the reaction rate after the operation for 15000 hr was 30.2 mol%.

  Further, when the catalyst inside the reaction furnace was taken out after the operation was stopped, the mass of the entire metal silicon particles was 490 g. About 30% by mass of nitric acid (about 10 ml) was added to a part (0.5 g) of the metal silicon particles, and about 5 ml of hydrofluoric acid was added dropwise to complete dissolution. When this solution was analyzed by ICP-MASS (device name: 7500CS manufactured by Agilent), the content of iron, calcium and nickel with respect to the entire metal silicon particles was 0.35% by mass of iron and 1.25% of calcium. The mass% and nickel were 1900 mass ppm.

  The present invention stabilizes the process of producing hydrogenated chlorosilane by reducing tetrachlorosilane in a hydrogen gas atmosphere.

Claims (11)

A method for producing hydrogenated chlorosilane by reducing tetrachlorosilane in a hydrogen gas atmosphere,
A method for producing hydrogenated chlorosilane, wherein a reaction accelerator containing an iron component, a calcium component, and a nickel component is used as a reaction accelerator for conversion from tetrachlorosilane to hydrogenated chlorosilane.   2. The production of hydrogenated chlorosilane according to claim 1, wherein the reaction accelerator contains metallic iron or an iron compound as an iron component, contains metallic calcium or a calcium compound as a calcium component, and contains metallic nickel or a nickel compound as a nickel component. Method.   The reaction accelerator is prepared by adding the iron component, the calcium component, and the nickel component to metal silicon particles, and is used as a catalyst for reducing tetrachlorosilane to hydrogenated chlorosilane. Item 3. A method for producing a hydrogenated chlorosilane according to Item 2.   The amount of the iron component, the calcium component, and the nickel component contained in the metal silicon particles is 0.1 to 5 mass of the iron component with respect to the whole metal silicon particles including the reaction accelerator. %, The calcium component is 0.1 to 5% by mass, and the nickel component is 400 to 3000 ppm by mass.   The catalyst heats the metal silicon particles together with hydrogen gas in a reaction tank to remove the surface oxide film of the metal silicon particles, and then in a hydrogen atmosphere, the iron component, the calcium component, and the nickel The method for producing a hydrogenated chlorosilane according to claim 3 or 4, which is obtained by adding components and heating to activate the catalyst.   The method for producing hydrogenated chlorosilane according to claim 5, wherein the hydrogen atmosphere includes tetrachlorosilane.   The catalyst contains the metal silicon particles, the iron component, the calcium component, and the nickel component in a reaction tank, and is heated in a hydrogen atmosphere to remove the surface oxide film of the metal silicon particles. The method for producing a hydrogenated chlorosilane according to claim 3 or 4, which is obtained by simultaneously performing catalyst activation.   The method for producing hydrogenated chlorosilane according to claim 7, wherein the hydrogen atmosphere includes tetrachlorosilane.   The method for producing hydrogenated chlorosilane according to any one of claims 1 to 4, wherein a reaction temperature at the time of reducing the tetrachlorosilane is 350 to 600 ° C.   The activity of metal silicon particles for the production of hydrogenated chlorosilanes, which activates metal silicon particles used as a catalyst in the reduction of tetrachlorosilane to hydrogenated chlorosilanes, using a reaction accelerator containing iron, calcium and nickel components. Method.   A catalyst for producing hydrogenated chlorosilane, wherein a metal accelerator contains a reaction accelerator containing an iron component, a calcium component, and a nickel component as a reaction accelerator for conversion from tetrachlorosilane to hydrogenated chlorosilane.