Preparation of fibre-like silicon nitride from silicon powder

Abstract

Silicon nitride is known to be a very good ceramic material for high-temperature structural applications because of its interesting properties, such as high mechanical strength, light weight and good resistance to thermal shock and oxidation [1]. This material in the form of whiskers or fibres is very useful as a reinforcing constituent in ceramic-matrix composites. Recent reports [2-4] on the use of silicon carbide and silicon nitride whiskers in ceramic composites have shown the role of whiskers in improving the fracture toughness. There are very few reports on the preparation and characterization of fi-silicon nitride whiskers [5, 6]. Recently there has also been a report [7] on the preparation of fibre-like 0l-silicon nitride by the carbothermal reduction and nitridation of diatomaceous earth, which is 82.5% silica with metallic oxides and organic impurities. While working on the synthesis of silicon nitride by the ammonolysis of silicon tetrachloride, the need for a method that does not involve any chloride was felt by us because of the problems involved in the complete removal of the chloride. A recent patent [8] suggested that silicon nitride can be prepared by reacting silicon directly with liquid ammonia at a temperature of about 100 °C and heating the intermediate imide to high temperatures. Since this involves high pressures, a reaction between heated silicon and ammonia gas was attempted. This resulted in the formation of silicon nitride in fibre-like and whisker forms. Silicon powder (2 g; Riedel; major impurity about 0.5% iron) with particle size of about 20/xm was spread on a high-purity graphite (Poco) boat and was kept in a mullite tubular furnace. While passing dry ammonia gas (300 mlmin -1) over the sample, the furnace temperature was raised to 1350 °C. It was maintained at this temperature for 4 h and furnace-cooled. It was observed that about 20% of the silicon was converted into the nitride. The product was white in colour and was formed both on the graphite boat and on the inner wall of the mullite tube just above the graphite boat. The possible reactions under the experimental conditions leading to the formation of silicon nitride could be