2022 Fiscal Year Research-status Report
Development of GaN superjunction devices with pGaN ArF laser activation for high power application
| Project/Area Number |
22K20438
|
|---|
| Research Institution | Toyota Technological Institute |
|---|
Principal Investigator |
VILLAMIN MARIAEMMA 豊田工業大学, 工学(系)研究科(研究院), ポストドクトラル研究員 (00960874)
|
|---|
| Project Period (FY) |
2022-08-31 – 2024-03-31
|
| Keywords | GaN / Laser activation / excimer laser |
|---|
| Outline of Annual Research Achievements |
Main goal of the study is to realize Gallium nitride superjunction (GaN SJ) power device with p-GaN layer activated using an ArF laser. This will be an original proof of concept for this fabrication method. GaN SJ device is a candidate for next generation power devices. However, fabrication of SJ structure is very difficult, due to the alternating p- and n- layers needed. Although this can be achieved by selective ion implantation, a more elegant approach is to use selective excimer laser annealing, which can define the p-region (p-GaN) by laser irradiation. Research implementation plan includes three phases. Phase 1 includes optimization of the Mg-activation using ArF laser. Phase 2 includes feedback of optimized laser parameters to further tune the next batch of GaN wafer properties, and purchase the designed photomask for SJ device fabrication. And phase 3 includes the used of the optimized wafer and laser parameters to demonstrate actual SJ power device and characterization. Few months after we received Kakenhi Startup, initial results of ArF laser annealing of pGaN is presented during the JSAP spring conference 2023. The effectiveness of ArF laser annealing in the activation of Mg-doped GaN using cloverleaf mesa structure is investigated. Using hall measurement, laser irradiated pGaN showed that the resistivity increases as laser power is increased, implying that the laser power may have damage the surface. ArF laser annealing method using cloverleaf mesa structure is first successfully demonstrated. Further investigation needed to confirm pGaN activation.
|
| Current Status of Research Progress |
Current Status of Research Progress
3: Progress in research has been slightly delayed.
Reason
TThe preparation and alignment of the optical setup and UV camera for laser activation are done. Moreover, the fabrication of pGaN device using cloverleaf mesa structure has been established. This cloverleaf mesa structure will be used in the next fabrication of pGaN device for laser annealing optimization. Also, our laboratory has never used Indium (In) contact before, so after doing several deposition trials, now we can do In metal deposition via resistive evaporation. Also, we have shown initial data on the laser activation of pGaN with cloverleaf structure via ArF laser annealing during the JSAP spring 2023. However, the pGaN devices have high resistance contact, which causes problem during Hall measurement because Hall voltage signal is low. Thus, low pGaN contact resistance is needed. Currently, InAu metal is investigated as low resistance pGaN contact (ie thickness and contact annealing temperature). This needs to be addressed first before we can continue the optimization of laser activation.
|
| Strategy for Future Research Activity |
In the near future, we want to optimize the InAu metal contact for better measurement of the resistivity and Hall voltages of the pGaN wafers. Whereas, in the future, we expect to finish the Phase 1 of the project, which is the optimization of the ArF laser pGaN activation using the current wafers. Then we will buy new batch of GaN wafers. Next is the purchase and design of the photomask for the GaN SJ device (Phase 2) and lastly, the fabrication of the GaN SJ device with pGaN activated using laser annealing (Phase 3).
|
| Causes of Carryover |
We will buy new batch of GaN wafers before the year ends. The cost of the new GaN wafers are expensive, which will use most of my Kakenhi budget. It is important to properly evaluate the wafers that we have now, and then use the data to decide the new batch wafer configuration. Only then, we can buy the new batch of GaN wafers. Thus, the buying of new GaN wafers takes time.
|
