What are the main factors that affect crystal growth?
Thread Content
The temperature on the surface of the silicon rods in a Siemens reduction furnace should be around 1080–1100 degrees; when this temperature drops, amorphous silicon may form. In PECVD-based polycrystalline silicon film deposition and silane pyrolysis vapor deposition (such as the fluidized bed process), the temperature is much lower than that in the Siemens method; yet amorphous silicon does not form. So what are the main factors that influence crystal growth? Traditionally, it is believed that the temperature of the substrate – which in the Siemens process refers to the surface of the silicon core or rod, in thin-film growth processes it is the substrate on which the film is grown, and in fluidized bed processes it is the seed crystal grains – determines the growth process. However, I believe that what truly determines the crystal growth process is not the absolute temperature of the substrate, but rather the temperature gradient between the core reaction zone and the substrate. In the Siemens process, the temperature of the gas phase is lower than that of the substrate, which allows the reduction reaction to take place on the surface of the silicon rod. The core reaction zone can be considered to be located on the surface of the silicon rod, with no temperature gradient between this zone and the substrate. The reason for maintaining the temperature between 1080 and 1100 degrees is that if the temperature is too low, the reduction reaction will not occur. Therefore, I believe that in the Siemens process, this temperature is used merely to enable the reduction reaction to take place; it is not necessary to maintain this temperature in order to grow polycrystalline silicon. In thin-film growth and fluidized-bed processes, the substrate is in full contact with the gas phase, and the temperature gradient between the core reaction zone in the gas phase and the substrate can also be ignored; however, the absolute temperature of the substrate is lower. Taking all this information into account, I believe the main factor affecting crystal growth is the temperature gradient. If this hypothesis is correct, then if it is possible to enable silicon trichloride and silicon tetrachloride to undergo reduction reactions with hydrogen at lower temperatures (by finding ways to reduce the energy barrier for this reaction), it will be possible to lower the reaction temperature in the reduction furnace (or fluidized bed), thereby achieving energy savings.
The original poster’s analysis makes some sense~
Hasn’t anyone discussed it...
Do you mean that polycrystalline silicon crystals can also grow at low temperatures, but the reduction reaction cannot occur? I’m not sure what you mean by reducing the reaction energy barrier – using a catalyst?
Well, the general idea is that lowering the energy barrier can be achieved using catalysts, but we have other methods as well; however, this information has to remain a secret between me and my supervisor, hehe
Regarding your inference, I hold the following view: (1) The chemical deposition reaction in Siemens reduction furnaces is primarily governed by thermodynamic and kinetic factors. Based on Gibbs free energy and chemical reaction constants, to obtain polycrystalline silicon products, an appropriate reaction temperature range is between 1000–1200 degrees. The reason for this is that the deposition temperature on the silicon surface should be below T_max; as the temperature increases, the crystals of silicon become larger and brighter； The lower the temperature, the finer the crystals become, and their surface takes on a dark gray color. However, the temperature cannot be too low; below 1000°C, loose, dark brown, scale-like, amorphous silicon is formed. The surface temperature of silicon rods should not be too high, as at high temperatures (above 1200°C), silicon undergoes reverse corrosion. The entire deposition process is primarily governed by the reduction reactions on the silicon surface, with little dependence on the temperature gradient (which requires a well-designed reduction furnace) ； Additionally, during the production process, the composition of the reaction exhaust gases changes due to temperature gradients, and adjustments are required. (2) In the chemical vapor deposition used for the production of PECVD polycrystalline silicon films (the main method), in order to enable silicon atoms to crystallize properly on the substrate, practice has shown that in order to obtain high-quality films with a polycrystalline silicon structure through the decomposition of silane, the substrate temperature needs to be between 550 and 600 degrees. If, at this temperature, the amorphous silicon form predominates ； Above this temperature, it is constrained by the substrate material and enhanced chemical deposition techniques (and of course, cost). To enable the reaction to proceed within this temperature range, energy technologies such as heating, plasma, or light radiation are required (there are mainly two approaches for preparing polycrystalline silicon films by chemical deposition: A: First, an amorphous silicon film is prepared, which is then crystallized into a polycrystalline silicon film through various heat treatment processes (two-step method)) ； B: Producing polycrystalline silicon films in one step through the decomposition of silane gas by utilizing relevant energy technologies). It can be seen that the substrate material is the factor determining the bottleneck in deposition temperature; therefore, the approach is to reduce the deposition temperature and employ a technique for preparing amorphous silicon films followed by crystallization. Although the temperature gradient can be neglected during the deposition process, the crystal form of silicon atoms formed by the deposition reaction is primarily determined by the surface temperature of the substrate. Using reduced energy barriers is merely a measure to lower the threshold for reaction formation, but whether reactions will occur under lower conditions depends on the specific results The above are merely my personal views; they can be discussed!
However, if only surface temperature and substrate material are used to explain it, it seems impossible to reasonably account for the temperature conditions in the fluidized bed process...............Whether it is the new silane method (represented by REC) or the chlorosilane fluidized bed method (represented by Wacker)
My guess: the growth of amorphous silicon occurs under conditions of a low mixing ratio, a large feed rate, a slow flow velocity, and becomes severe in the later stages of growth. Fewer nuclei are provided than needed; the fluidized bed method provides more nuclei than required. I don’t quite understand the fluidized bed method and would like to learn about it*
The reason, in my opinion, is that the effective growth surface area inside the reduction furnace is too small, which is the primary cause of excessive waste of raw materials.
The size of the effective growth surface area has little impact on what crystal form is formed....
My main question now is whether there is a theoretical \"absolute lower limit\" for the growth temperature of polysilicon; whether there really is such a temperature threshold at which, if the temperature of the \"substrate\" (whether it’s a silicon core or a seed crystal) falls below this threshold, polysilicon can no longer be grown. In Siemens reduction furnaces, the temperature must remain above 950 degrees (or 1000 degrees). In the PEVCD film fabrication process (where deposition occurs before zone melting), the temperature can be reduced, and in the fluidized bed process the temperature is even lower. In all cases, polycrystalline silicon is produced. Therefore, I wonder whether it might be possible to lower the reaction barrier, provide sufficient energy to the reactants in the gas phase at lower temperatures to generate a sufficient concentration of free radicals, and also provide a sufficient deposition surface in the gas phase, thereby allowing the growth temperature to be reduced further (below that of the current fluidized bed process)..........................