The Czochralski process is a technique for growing crystals in order to obtain solitary crystals of semiconductors like silicon, metals like silver and gold as well as several salts. The process is named after Jan Czochralski, who first demonstrated it in 1916. He was studying the rates of crystallisation of different metals.
The most significant application of the Czochralski process is the growth of large ingots of single crystal silicon. Initially highly pure semiconductor-grade silicon is subjected to high temperatures in a crucible made of quartz until it melts. Quartz is used as the crucible’s material because of its inertness. At this stage, doping can be initiated with impurity atoms such as boron or phosphorus being added to the molten silicon. Fixed amounts of dopants can yield desired characteristics in the semiconductor formed as a result of this procedure.
Doping with boron or phosphorus can change the intrinsic silicon present in the crucible to either an n-type or a p-type semiconductor. Thus the electrical conductivity of the semiconductor produced can be altered to the desired levels. The so-called ‘seed crystal’ which is mounted on a rod is dipped into the molten-state silicon. After the seed crystal’s rod is retrieved in an upward direction and simultaneously rotated, a large, single-crystal ingot cylindrical in shape is produced. The temperatures involved, the rate of retrieval and the speed of rotation of the rod is controlled and maintained to high degrees of accuracy. The entire process is carried out in an environment engulfed in a noble gas like argon.
The Czochralski process has a noteworthy phenomenon taking place when it is carried out in the quartz crucible. The crucible walls can be dissolved by the molten silicon, bringing oxygen particles along. These oxygen particles cause undesired metal impurities to be trapped and thus prevent them from affecting the final product. Furthermore, these particles help improve the mechanical strength of the wafers, which is necessary during the processing of devices.
Silicon wafers are obtained from large silicon ingots by slicing and polishing to achieve a high degree of flatness. These chips, normally about 0.75 mm in thickness are thus used for the manufacture of integrated circuits. The ingots from which they are cut may range from anywhere between 1 and 2 metres in length with diameters up to 400 mm.