Silicon not heating with induction heater?

[Patrick Underwood]

I’m using an inductive heater to try to melt some 99.85% polycrystalline silicon, as would occur in the Czochralski process, but the silicon workpiece is not melting let alone even getting hot. The silicon workpiece is about the size of a playing dice. It becomes conductive at 200 F and becomes fully conductive at 230 F. Even after preheating the workpiece well above the point at which it becomes conductive, the inductive heater still doesn’t “take over” to raise the temperature. The inductive heater works fine with other conductive materials such as iron, aluminum, and copper and heats them quite effectively. Heating silicon inductively is a well established process but due to confidentiality I have not come across any details regarding the frequency needed.
- The frequency of the heater I am using is 90KHz and was confirmed with an oscilloscope.

 

[Patrick Underwood]

 

[trurle]

I’m using an inductive heater to try to melt some 99.85% polycrystalline silicon, as would occur in the Czochralski process, but the silicon workpiece is not melting let alone even getting hot. The silicon workpiece is about the size of a playing dice. It becomes conductive at 200 F and becomes fully conductive at 230 F. Even after preheating the workpiece well above the point at which it becomes conductive, the inductive heater still doesn’t “take over” to raise the temperature. The inductive heater works fine with other conductive materials such as iron, aluminum, and copper and heats them quite effectively. Heating silicon inductively is a well established process but due to confidentiality I have not come across any details regarding the frequency needed.
- The frequency of the heater I am using is 90KHz and was confirmed with an oscilloscope.

Your heater frequency is a way too low for silicon. You need at least 100 MHz for effective heating of high-resistance substrates like silicon. Also, powdering silicon actually would make heating even less effective.

 

[Patrick Underwood]

Thank you trurle for the reply.
See that's what I was also thinking... When you say high-resistance substrates I assume you mean electrical resistivity. I do see now mathematically that if the resistivity is higher it would require a higher frequency. Yeah...so we have a mathematical error. The resistivity of intrinsic silicon only gets to about .006 ohm cm when heated not 2 * 10^-6.