How are silicon carbide chips made?

Posted by

Wide-bandgap semiconductor materials, represented by silicon carbide and gallium nitride, break through the performance limitations of original semiconductor materials in high-power, high-frequency, high-speed, and high-temperature environments and are widely used in 5G communications, the Internet of Things, new energy, and cutting-edge national defense weapons. Equipment and other cutting-edge fields play an important role. In the context of Moore’s Law encountering bottlenecks and China’s Intelligent Manufacturing 2025, wide bandgap semiconductor materials are undoubtedly a good opportunity for China’s semiconductor industry!

Silicon carbide chips are made like this

New materials, “core” the future! Silicon carbide chips, replacing traditional silicon-based chips, can effectively improve work efficiency, reduce energy loss, reduce carbon emissions, improve system reliability, reduce volume, and save space.

Taking electric vehicles as an example, the use of silicon carbide chips will reduce the size of the electric drive device to one-fifth, reduce the driving loss of electric vehicles by more than 60%, and significantly increase the mileage with the same battery capacity.

How to manufacture silicon carbide chips for the future? This has to mention a concept: cells. Generally speaking, chips are semi-finished products after wafer cutting. Each wafer integrates hundreds of chips (the number depends on the chip size), and each chip is composed of thousands of cells. So how exactly are cells made?

first step

Injection mask. First, the wafer is cleaned, a layer of silicon oxide film is deposited, then the photoresist pattern is formed through process steps such as dispersion, exposure, and development, and finally the pattern is transferred to the etching mask through the etching process.

Step 2

Ion Implantation. Put the masked wafer into the ion implanter and inject high-energy ions. The mask is then removed and annealed to activate the implanted ions.

Step 2

Ion Implantation. Put the masked wafer into the ion implanter and inject high-energy ions. The mask is then removed and annealed to activate the implanted ions.

third step

Make the gate. A gate oxide layer and a gate electrode layer are sequentially deposited on the wafer to form a gate-level control structure.

the fourth step

Create a passivation layer. Deposit a dielectric layer with good insulating properties to prevent breakdown between electrodes.

the fifth step

Make drain-source electrodes. A hole is opened in the passivation layer, and metal is sputtered to form a drain-source electrode.

When a positive voltage is applied between the drain-source electrode and the gate-source electrode, the channel opens and electrons flow from the source to the drain, generating a current flowing from the drain to the source. At this point, a basic power device, the cell, is completed. Thousands of cells form a chip, and then integrated into the wafer substrate, you will have a wafer as brilliant as a rainbow!

The silicon carbide substrate of the wafer is prepared by the physical vapor transport method (PVT), and is formed through a series of processes such as decomposition and sublimation of silicon carbide powder, gas transmission and deposition, and cutting, grinding and polishing.