Development History of the Ion Implanter IMPHEAT

In 2009, Sumitomo Electric approached NIC about developing an ion implanter for SiC power devices. Sumitomo Electric, grasping the demands of the times for EVs, focused on the potential of SiC and had high hopes that NIC’s ion implantation technology would serve as the catalyst for the creation of new power devices. Although the NIC technical team had knowledge of the excellent basic properties of SiC, many of them shared the view that ion implantation was technically difficult.

An ion implanter consists of an ion source, which generates ions of the element to be implanted; an analyzer magnet, which separates multiple types of ions based on differences in mass and charge; an acceleration tube, which accelerates the ions to a specified energy; and an end station, which implants the ions into the wafer. However, there are significant differences between ion implantation on Si and SiC. First of all, the wafers themselves are different. While Si wafers have a high degree of flatness, SiC wafers have a low degree of flatness and are easily deformed by the effects of ion implantation and heat. Additionally, the wafer temperature during implantation is below 100°C with Si, but SiC requires heating to around 500°C. Furthermore, while the ion species implanted in Si were boron, phosphorus, and arsenic, in SiC it was necessary to implant aluminum in addition to phosphorus.

The SiC wafer transport system remains one of the development challenges. In addition to being able to withstand high temperatures, achieving stable transport was also a major challenge. Tomoaki Kobayashi, who currently works as an expert, was also involved in the transport system and high-temperature protection measures.

“In early SiC implanters, wafers were placed on the holder by hand. For transportation, an electrostatic chuck was used, which uses electrical force to attract and hold the holder. It was necessary to consider the insulation between the power to the electrostatic chuck and the power to the high-temperature heater, as well as the appropriate voltages. There were also many challenges to overcome, such as how to raise the temperature quickly, how to achieve an even temperature distribution, and how to measure the temperature. We continued to make steady efforts to overcome these challenges one by one,” says Kobayashi.

Ion implanters for SiC power devices need to ionize aluminum, which is not normally used. Ionization converts gas or vapor containing the intended material into plasma, but there is no such gas that contains elemental aluminum. The person who took on this challenge was Tetsuya Igo, who is currently in charge of  technical sales in the Semiconductor Equipment Division.

“We repeatedly considered what type of aluminum compound would be good to use. And we turned our attention to aluminum nitride. By applying a plasma of a gas containing fluorine to aluminum nitride, aluminum fluoride was generated, and an aluminum ion beam was successfully generated. The key was how to stabilize it. The high voltage applied caused discharges and made the beam generation unstable. Another challenge was to develop an aluminum ion source that could produce large amounts of aluminum ions for a long period of time. This was passed down as a task for the future,” says Igo.

Thus, the first research unit was delivered. However, problems arose to the point that the team members had to commute to the client, Sumitomo Electric, for a year. The issues and new challenges discovered there led to the development of the mass-produced machines IMPHEAT and IMPHEAT-II as well as the next-generation ion implanter.