NCER Assistance Agreement Final Project Summary

Date of Final Report: February 28, 2007

EPA Agreement Number: X-83254101-0

Center: Center for Environmental and Energy Research (CEER)

Project Title: Recycling of Silicon-wafers Production Wastes to SiAlON Based Ceramics with Improved Mechanical Properties

Investigator(s): James R. Varner and David A. Earl

Institution(s) of PI(s): Alfred University

Research Category: Congressionally Mandated Center

Project Period: October 1, 2005 – December 31, 2006

Description and Objective of Project: The output of highly pure semiconductor silicon for integrated circuits and memories is increasing year by year. During wafer production process, about 60% (2400 tons) of silicon ingot after trimming is scrapped with the waste water disposal from cutting and polishing. The recycling to highly pure silicon is very costly. If the silicon sludge can be converted to nitride-based structural ceramics (SiAlON), it is helpful for semiconductor industry and ecological problems.

The objective of this project was to recycle the silicon sludge to SiAlON ceramics by using the combustion synthesis process, and to demonstrate that the fracture toughness of the SiALON based ceramics can be improved by adding reinforcing secondary particles such as ZrO₂ into the SiAlON matrix. The anticipated benefits are high fracture toughness, low cost, reduced environmental pollution, significant energy saving due to microwave sintering, and reduced emission due to the self-propagating exothermic synthesis reaction. The sintered products can be used for abrasives, corrosion-resistant filters, and wear-resistant materials below 1000° C.

Objectives of the project:

1) Reduce the environment pollution due to silicon sludge produced in the semiconductor industry by recycling silicon sludge through converting it into SiAlON ceramics by the combustion synthesis process.

2) Reduce the energy requirement of the SiAlON ceramic processing by microwave heating.

3) Demonstrate that fracture toughness of SiAlON ceramics can be improved by transformation toughening of ZrO₂ secondary particles added as reinforcment into the SiAlON matrix.

Technical Background: The output of semiconductor silicon for large-scale integrated circuits and memories in USA is about 2,000 tons/year in recent years [1]. It is produced as a single crystalline ingot and processed to wafers through cutting, polishing, and washing. Large edges of a silicon ingot cut by trimming (~10% of an ingot) are used as a source material for polycrystalline silicon solar batteries. The silicon sludge contains silicon and a large amount of ceramic abrasives such as Al₂O₃, SiC, Si₃N₄, and ZrO₂, coagulants, polymers, grinding oils, and water. If the silicon sludge is left outside and dried, there is potential danger of pollution by diffusing out of fine powders into the air, or through fire. Since recycling to high-purity silicon is very difficult and costly, the general practice is adding the source material to cement or disposing of it in land reclamation. An alternative solution is to use the silicon sludge to produce SiAlON ceramics.

SiAlON (Silicon Aluminum Oxy-nitride) is a high-technology structural ceramic material used for many commercial applications requiring wear resistance, high hardness, chemical stability, and heat resistance due to its excellent high-temperature properties [2]. SiAlON ceramics that are iso-structural with silicon nitride offer the advantage of incorporating some of the sintering additives into the silicon nitride lattice, thus reducing the overall amount of secondary phase and potentially improving high-temperature properties. They offer advantages of easier fabrication compared with silicon nitride ceramics because of the lower viscosity of the M-Si-Al-O-N liquid phase, where M is one of the cations Li, Mg, Ca, Y, Sc and most of the rare-earths, which facilitate easier densification at sintering temperatures [3]. Cost remains a major barrier to the more widespread use of SiAlON-based structural ceramics, and will remain so until large-scale supplies of less-expensive raw materials become available [4]. The raw-material cost could be significantly reduced if it is possible to recycle any industrial waste material which contains a significant amount of silicon as the major phase. Silicon wafers production waste from semiconductor industry is one of the ideal precursors for the synthesis of SiAlON powder due to its small particle size and sufficient amount of other phases, such as Al₂O₃, which are necessary for liquid-forming agents for successful pressureless sintering. Thus, these waste silicon sludges could be used without any special preparation to produce very-low-cost SiAlON-based structural ceramics.

One of the economical methods of producing SiALON from silicon is by nitriding combustion known as Self-propagating High-temperature Synthesis (SHS). Nitriding combustion was discovered by A.G. Merzhanov and his coworkers in 1967 as a solid-gas combustion mode of self-propagating SHS [5]. Many other compounds, such as carbides, borides, silicides, aluminides, and others, are produced from the mixture of metal and non-metal elements by SHS [6]. Nitriding combustion is similar to oxidation combustion, since it involves a highly exothermic reaction, but is different in that it leaves solid products of metal nitrides without discharging carbon dioxide. Nitriding combustion is regarded as an energy-saving process to produce various nitride ceramics, because the synthesis reaction propagates spontaneously after the initiation of combustion. The nitriding combustion is based on the following reaction [7].

3Si + 2N₂ = Si₃N₄ -748 kJ/mol

This exothermic reaction propagates spontaneously and rapidly when the reactant is charged with a powder form in a pressurized nitrogen atmosphere.

Summary of Findings: A high-temperature self-propagating high-temperature synthesis (SHS) reactor capable of operating at a maximum pressure of 300 psi was designed, developed and successfully tested. Silicon wafers production wastes were collected and characterized for particle size, phases using X-ray diffraction (XRD), and morphology of the particles using scanning electron microscopy (SEM). The silicon sludge was milled into fine particles and subsequently converted into β-SiAlON ceramic by high-temperature SHS reaction. The XRD results clearly show that only 50% of the silicon sludge was converted into β–SiAlON due to low nitrogen pressure. The β-SiAlON powder was then mixed with 25wt% of Y₂O₃-stabilized ZrO_2, consolidated into pellets, and sintered to 92% of the theoretical density at 1600ºC. These materials exhibited a maximum Vickers hardness of 4.6GPa, which is much lower than the expected value of 12-15GPa. Through this project it was demonstrated that it is possible to convert silicon wafers production wastes into β-SiAlON ceramics by high-temperature SHS reaction. However, it was observed that the complete conversion of silicon sludge into β-SiAlON requires a higher nitrogen operating pressure than the present SHS reactor system. The SHS autoclave was also tested for other potential ceramic material development such as TiN from Ti or AlN from Al waste.

Conclusions:

• · A high-temperature SHS reactor capable of operating at a maximum pressure of 300 psi was designed, developed and successfully tested.
• · A recycling process of the silicon wafer production waste in semiconductor industries to SiAlON-based ceramics has been demonstrated by using an SHS nitriding combustion process.
• · Only 50% of the reactant mixture converted to β-SiAlON at the operating nitrogen pressure of 240 psi of the SHS reactor chamber. This nitrogen pressure was not sufficient to complete the combustion reaction. However, our experimental results clearly indicate that combustion efficiency could be improved by designing an SHS reactor capable of operating at higher pressure.
• · The Vickers hardness of the sintered sample was very low due to the presence of un-reacted silicon.
• · The SHS reactor developed in Alfred University through CEER funding was successfully tested for the conversion of Ti to TiN. It has the potential to develop advanced ceramic materials such as AlN from aluminum waste. It is also worth noting that aluminum and titanium are much more reactive (combustible) than silicon, hence complete reaction could be possible by applying a lower gas pressure than required for conversion of silicon to SiAlON.

References:

1. 1. U. S. Environmental Protection Agency: Advanced Search for Silicon producers for

semiconductor Industries in USA.

2. 2. Thommy Ekstrom and Mats Nygren, “SiAlON Ceramics,” J. Am. Cer. Soc.,75 [2], p. 259-76 (1992).

3. Mukerji,J. and Bandyopadhyay, S., “SiAlONs from natural aluminosilicates,” Adv.Ceram.Mater.,1988,3[4]369.

4. Perere, D. S, “Silicon nitride and SiAlON made from New Zealand raw materials,” J. Aust. Ceram. Soc., 23[1] 11(1987).

5. A. G. Merzhanov, “History and Recent Developments in SHS,” Ceramics International 21 pp.371-379 (1995).

6. A. G. Merzhanov, “ Twenty years of search and finding” in Combustion and Plasma Synthesis of High-Temperature Materials, ed. Z. A. Munir and J. B. Holt. VCH Publ., Inc., New York, 1990,

7. Zheng J, Miyamoto Y, Yamada O. Combustion synthesis of sialon powders,” J Am Ceram Soc, 1991, 73, 3700-2

Publications/Presentations:

Supplemental Key Words: Self Propagating High Temperature Synthesis (SHS Reaction), SiALON, Recycling, Wastes

Relevant Web Sites: http://ceer.alfred.edu