Ãëàâíàÿ ñòðàíèöà | Êðåìíèé Main page | Silicon

“PLASTMA” LIMITED LIABILITY COMPANY
ABBREVIATED TITLE “PLASTMA LTD”


SILICON

Project description:

Fluoride-hydride technology of obtaining the poly-crystal solar silicon

As for the contents of electrically active admixtures, the solar gradation silicon (SG) occupies the intermediate place between the metallurgic gradation silicon (MG) and the electronic gradation silicon (EG). There is a hypothetic balance between its price and the quality level determining efficiency of photo-voltaic systems (PV-systems), manufactured with its application. For example, ratio of the price of 1 kg of silicon to efficiency of PV-systems equaling to 25$/14,5% is presently considered satisfactory for manufacturers of PV-systems, however, it can hardly be achieved for silicon manufacturers. At least, there are no sufficient data about possible SG silicon production for the price of 25$/kg. Provision of a low consuming technology of the big-scale production of SG silicon is a strategically important task for the PV industry with its market of products being notable of the incrementing deficit.

Pricing for the poly-crystal silicon with the existing possibility of production automation is presently determined by three factors: cost of raw materials and consumable process materials, power expenditures, expenditures for provision of ecological and technological production safety.

Influence of the first two factors on the price of silicon is determined by chemism of the used processes, thermodynamics and kinetics of chemical reactions and thermal physical properties of the used materials imposing certain physical and chemical limitations on possible improvement of technology with the aim of price reduction of the product.

Influence of the third factor is mainly determined by possible exclusion of accumulation of big by-products often forming dumps of liquid wastes which can not be sold, and hence, create unsolvable ecological problems.

In opinions of authors of numerous publications, improvement of the chlorine-silane production technology of poly-crystal silicon over 50 years of its existence has made it possible to essentially reduce power consumption, expenditures of consumable materials, and has provided the increased silicon output from 12% to 60% of the original raw materials. However, this method has exhausted capabilities of its development and can not provide acceptable features of SG-silicon.

The fluoride-hydride production technology of poly-crystal silicon was industrially developed at MEMC plant in Pasadena, USA, owing to researches of Ethyl Co. During 1999-2005 the silicon production volume by using this technology has grown from 1400 to 2700 tons per annum. According to information of MEMC the technology provides the price level acceptable for SG silicon with orientation of the main production to EG silicon of the highest quality. A problem for MEMC includes a big tonnage accumulation of a by-product, sodium-alum-fluoride NaAlF4 (about 4 tons per 1 ton of silicon), which can not fully be used, and hence, it is accumulated in dumps.

Proposed technology

The main advantage of the fluoride-hydride technology is high selectivity and exothermic feature of chemical reactions determining homogeneity of the obtained products, which provides the high silicon output from the original raw materials without essential power expenditure.

He proposed technology has been studied on the laboratory level and may be provided with patent protection declared along with the potential investor by implementation of the pilot project stage.

During 1999-2005 we performed an analysis of problematic issues related with industrialization of the fluoride-hydride technology, and a big volume of experimental works has been conducted, when on the basis of their results the following can be concluded.

As for composition of admixtures, the obtained pilot samples of poly-crystal silicon complies with silicon of SG solar gradation, what is confirmed with the protocol of tests Giredmet No. 9549.01, dated December 14, 2001.

In accordance with the protocol of tests of poly-crystal silicon, dated February 23, 2003, performed by the Belgian center of microelectronics IMEC after modification of silicon poly-crystal to a mono-crystal, its separation to wafers and manufacture of solar components, electro-physical mono-crystal parameters (spread of the specific electrical resistance, the service life of non-major charge carriers) and electrical features of solar components (Isc, Voc, FF, Eff) comply with commercial samples of mono-silicon SG ND22 and pi 04.

The basic technological process is the process of silicone tetrafluoride SiF4 hydrogenesation to mono silane SiH4 along with application of calcium hydride CaH2 as a hydrogen donor. Selection of structural materials and development of the original construction of the hydrogenesation unit resolves basic problems related with corrosion resistance of the equipment , elimination of emergency situations, removal of the production by-product.

The process of silicone tetrafluoride has been optimized. The initial orientation to the process of obtaining SiF4 from the by-product of apatite procession – H2SiF6, by means of transfer of siliconfluorinehydrogen acid to Na2SiF6 and its subsequent thermolysis, has not justified itself due to formation of big tonnage wastes of sodium fluoride NaF. Another method of SiF4 obtaining by means of the reaction of gaseous HF with quartzite SiO2, turned into a suspension in sulfuric acid, is rather expensive and does not provide SiF4 high quality, as far as the admixture level is concerned. The process of obtaining SiF4 by means of a reaction of F2 with metallurgic silicon requires gaseous fluorine by means of HF electrolysis (electric power consumption 20 kWh per 1 kg of F2), and therefore, this process if unprofitable due to its high cost and big power consumption.

The following technology is based on exothermic chemical reactions and it provides the acceptable admixture composition of products with the low level of power consumption and silicon output from the original raw materials at least 85%.

The original raw material is silicon MG with the contents of the main products at least 98% of the mass.

Brief description of the method

Stage 1. Production of silicon tetrafluoride

The process is characterized by a high selectivity towards admixtures remaining in slag, except for carbon. The reaction takes place in a boiling layer of particles of metallurgic silicon having size 1-1.5 mm with pressure not exceeding 2bar. Carbon fluoride gaseous compounds and hydrocarbons are converted to methane with the use of a catalyst. SiF4 is purified in a regenerated absorber with removal of traces of HF and H2O and condensed in a cryogenic condenser-evaporator under pressure. The low boiling hydrogen with the methane admixture is burnt along with obtaining the useful heat.

Stage 2. Production of calcium hydride.

The process is carried out in a boiling layer of granulated metallurgic calcium under pressure not exceeding 2 bar in the mixture of hydrogen with argon. Argon, not participating in the reaction is comprimed to a gasholder and is used again.

Stage 3. Production of mono-silane.

The reaction is carried out in the bubbling reactor in the ion melt of triple salt mixtures containing calcium hydride partially in the form of the suspension, partially in the dissolved form. The maximum pressure in the reactor is equal to 2,5 bar.

The worked out salt melt containing the limit quantity of calcium fluoride is subjected to recycling, when CaF2 is separated by means of filtration, and the mixture of salts is returned to the process.

Mono-silane is purified at sorbents and filtered for removal of mechanical particles and then comprimed to a gasholder by means of a membrane compressor.

Calcium fluoride in the form of the synthetic feldspar is supplied to the manufacturer of waterless HF as a give-and-take raw material, wherein the reaction takes place:

Stage 4. Production of granulated poly-crystal silicon.

The process takes place in a boiling layer of silicon particles in the mono-silane and hydrogen mixture. The reactor is made of quartz, the internal surface of the reactor is covered with a Si3N4 layer in the mixture of mono-silane with ammonia. The maximum pressure in the reactor is equal to 2 bar. Hydrogen extracted according to reaction (4) is comprimed with the membrane compressor and used in the process of obtaining calcium hydride (stage 2) and again at stage 3.

The expected self-cost of silicon production with the output volume of 1000 tons per annum is equal to 23USD/kg (the calculation is shown in the table).

The cost of the pilot project having capacity of 20 tons per annum is evaluated for 1,7 mln. USD with productivity of silicon sets of 3 kg/h.

Investment expenditures for creation of the industrial production of 1000 tons per annum is evaluated for 56 mln. USD.

Table 1

Calculation of self-cost of 1 t of granulated poly-crystal silicon

Annual output
of silicon 1000 tons
commodity CaF2 5600 tons

Description of articles of expenditures

Unit

Price, USD

Q-nty per volume, t

Expenditure per volume, thousand USD

Expenditure per 1 ton, USD

1. Raw material

Metallurgic silicon

t

2 000

1 100

2 200

2 200

2. Auxiliary materials

- metal calcium Ca

t

4 000

2 900

11 600

11 600

- hydrogen fluoride HF

t

1 500

2 900

4 350

4 350

Granulated sodium fluoride NaF

t

1 462

352

515

515

Lithium chloride LiCl

t

2 885

23

67

67

Potassium chloride KCl

t

1 110

29

32

32

3. Power expenditures

- electrical power

thsnd kW/h

50

19 000

950

950

- nitrogen N2

t

117

1 800

420

420

- deionized water

thsnd.m3

409

44

18

18

Main and additional salary

-

-

-

900

900

Deductions for social insurance (35%)

-

-

-

315

315

Depreciation

-

-

-

704

704

- of equipment

-

-

-

650

650

- of buildings and structures

-

-

-

54

54

Expenditures for maintenance and operation of equipment

-

-

-

460

460

Shop expenditures

-

-

-

270

270

Shop self cost

-

-

-

23 347

23 347

General plant expenditures

-

-

-

300

300

Production self-cost

-

-

-

23 647

23 647

Innovation fund

-

-

-

22

22

Out of production expenditures

-

-

-

16

16

Full self-cost

-

-

-

23 685

23 685

Self-cost of calcium fluoride CaF2

-

-

-

1 100

196

Self-cost of silicon production

-

-

-

22 585

22 585


Copyright © 2001-2008, “Plastma” Ltd. Co.