Periodic Reporting for period 3 - GreenDiamond (Green Electronics with Diamond Power Devices)
Reporting period: 2017-11-01 to 2020-10-31
14 European research centers and firms coming from 6 European countries will during 5 years - May 2015 to October 2020 - develop diamond based power electronics devices. 3 to 30 times better than the actual used materials, diamond will enable to diminish loses, thermal requirements and the weight, through the higher blocking voltages, of the current devices and to enhance at the same time the overall reliability and performances.
Silicon is a well-established starting material that has addressed the requirements of energy conversion for more than 50 years. However, it is widely recognised (as shown in research roadmaps on power semiconductor devices) that a real step-improvement in Power Electronics will be obtained by employing devices based on wide bandgap semiconductor materials. Wide bandgap semiconductor materials have superior electrical characteristics for power devices when compared to silicon. Power electronic devices based on wide bandgap semiconductor materials will result in substantial improvements in the performance of power electronics systems by offering higher blocking voltages, improved efficiency and reliability, as well as reduced thermal requirements thus leading to realisation of more efficient green electronic systems. Among wide bandgap semiconductors, diamond is considered to be the ultimate semiconductor for applications in high power electronics due to its exceptional properties. Its dielectric breakdown strength is 3 times higher than in SiC and more than 30 times better than in Si. In addition, the carrier mobility is very high for both carrier types and the thermal conductivity is unsurpassed.
The GreenDiamond project is a very successful project in term science and technology. The main impact is the demonstration that the technology to build a more efficient power converter with diamond is almost ready. All the work packages, focusing on a particular technology required for the fabrication of the full converter have been successful, except two wafer-level technological steps, ie 4 epilayer stack and etching. This bottleneck have been clearly identified and solutions proposed for further R&D.
GreenDiamond consortium is convinced that a diamond converter can be build with full power diamond transistors within 4-years of R&D.
The work programme to achieve this R&D goal and transfer this new technology to industry is:
• Engage a 4-year R&D programme at EU level to solve last bottleneck
• Inform industrials about diamond potential in energy applications
• Convince diamond producer about this new market, to attract specific production at low cost
Silicon is a well-established starting material that has addressed the requirements of energy conversion for more than 50 years. However, it is widely recognised (as shown in research roadmaps on power semiconductor devices) that a real step-improvement in Power Electronics will be obtained by employing devices based on wide bandgap semiconductor materials. Wide bandgap semiconductor materials have superior electrical characteristics for power devices when compared to silicon. Power electronic devices based on wide bandgap semiconductor materials will result in substantial improvements in the performance of power electronics systems by offering higher blocking voltages, improved efficiency and reliability, as well as reduced thermal requirements thus leading to realisation of more efficient green electronic systems. Among wide bandgap semiconductors, diamond is considered to be the ultimate semiconductor for applications in high power electronics due to its exceptional properties. Its dielectric breakdown strength is 3 times higher than in SiC and more than 30 times better than in Si. In addition, the carrier mobility is very high for both carrier types and the thermal conductivity is unsurpassed.
The GreenDiamond project is a very successful project in term science and technology. The main impact is the demonstration that the technology to build a more efficient power converter with diamond is almost ready. All the work packages, focusing on a particular technology required for the fabrication of the full converter have been successful, except two wafer-level technological steps, ie 4 epilayer stack and etching. This bottleneck have been clearly identified and solutions proposed for further R&D.
GreenDiamond consortium is convinced that a diamond converter can be build with full power diamond transistors within 4-years of R&D.
The work programme to achieve this R&D goal and transfer this new technology to industry is:
• Engage a 4-year R&D programme at EU level to solve last bottleneck
• Inform industrials about diamond potential in energy applications
• Convince diamond producer about this new market, to attract specific production at low cost
The results obtain at the end of the project matches the specific objectives as described in section 1.1 of the DoA:
1. To develop simulation models and parameters suitable for analysis of diamond devices’ analysis.
This first objective has been achieved by WP2 mostly during RP2, and simulation has refined during RP3.
2. To fabricate Prototype MOS transistors in several stages 2028
The process is for the fabrication of the first Diamond MOS transistor in inversion mode, as in the proposal, has not been successful. In fact the process involves 7 main tasks carried out by at least 4 partners, ie p+/p- epitaxy (CNRS/NEEL-CEA), n epitaxy (IMEC-IAF), p+ epitaxy (CNRS/NEEL), etching (UCL-IAF). Each step was studied separately with good results. But the process flow for the full transistor fabrication involves 4 successive epilayers, and the accumulation of growth defects at the end of the epilayer growth process makes the etching impossible. No sample was obtained with sufficiently low defect at the surface, even after optimization of the process, to allow the V-shape etching.
To continue the research activity on the other work packages, we decided to replace the original diamond MOSFET by two other kind of transistors:
- Reverse Blocking MESFET. This transistor is requiring only one epilayer, with less challenging fabrication process. Expected power is much lower than the original MOSFET, but we kept the volage breakdown rating at 1kV. We did the design from scratch, then fabrication and test (WP2, 3, 4 and 5).
- SiC MOSFET. These transistors are already available commercially. Specific transistors have been chosen and integrated into the packaging and the converter to validate the design carried out in WP7, WP8 and WP9.
With this backup solution, GreenDiamond could continue its research activity towards diamond power electronics.
3. To design and develop high quality, cost-effective package solutions suitable for the robust operation of 6.52028kV and 10 kV rated devices under extreme conditions
This task is completed. The package type is SOT-227, and base plate, die-attach and topside interconnections have been developed and tested. Packaging is limited to 250°C, higher that all existing packaging, because of the lack of suitable encapsulant.
4. To develop reliability tests for normal and extreme operating conditions of the developed power devices 2028
The measurement probe stations and equipment are ready to measure any device or diamond transistor prototype.
5. To demonstrate a high voltage three-phase DC/AC high-power converter based on diamond devices 2028
A 30kV – 2W hybrid gate driver has been designed and a first prototype has been realized. Three possible topologies have been considered to deploy a HVDC or MV application. However, due to the lack of diamond power transistor, only a SiC based converter has been fabricated.
6. To disseminate the outcomes within EU and provide the foundation for industrialisation
Results have already been published and presented in scientific conferences.
The most important important event was organised just before confinement on March 22020 in Brussels. For the first time a meeting gathered companies, international bodies and academics to discuss about diamond power electronics for energy applications. In particular, for the first time a roadmap to the first industrial diamond power converter was presented, and 3 companies presented their views about the opportunities for diamond electronics for their company.
Very interesting ideas came up during the discussion time, in order to continue the research activity and transfer the technology to the companies.
Regarding the exploitation of the results, the most important is that R&D will continue toward diamond electronics, more efficient and reliable than any other technology. Activity will be reduced, or re-oriented towards other target (harsh environment signal processing electronics, aeronautics, …) because of the lack of suitable financial support. But activity will continue.
Another important element is that USA just started a 12M€ project for diamond power electronics. So technology will be developed and will emerge in the industry, led by USA or EU depending on the financial support GreenDiamond partners could obtain to continue this project at a larger scale.
1. To develop simulation models and parameters suitable for analysis of diamond devices’ analysis.
This first objective has been achieved by WP2 mostly during RP2, and simulation has refined during RP3.
2. To fabricate Prototype MOS transistors in several stages 2028
The process is for the fabrication of the first Diamond MOS transistor in inversion mode, as in the proposal, has not been successful. In fact the process involves 7 main tasks carried out by at least 4 partners, ie p+/p- epitaxy (CNRS/NEEL-CEA), n epitaxy (IMEC-IAF), p+ epitaxy (CNRS/NEEL), etching (UCL-IAF). Each step was studied separately with good results. But the process flow for the full transistor fabrication involves 4 successive epilayers, and the accumulation of growth defects at the end of the epilayer growth process makes the etching impossible. No sample was obtained with sufficiently low defect at the surface, even after optimization of the process, to allow the V-shape etching.
To continue the research activity on the other work packages, we decided to replace the original diamond MOSFET by two other kind of transistors:
- Reverse Blocking MESFET. This transistor is requiring only one epilayer, with less challenging fabrication process. Expected power is much lower than the original MOSFET, but we kept the volage breakdown rating at 1kV. We did the design from scratch, then fabrication and test (WP2, 3, 4 and 5).
- SiC MOSFET. These transistors are already available commercially. Specific transistors have been chosen and integrated into the packaging and the converter to validate the design carried out in WP7, WP8 and WP9.
With this backup solution, GreenDiamond could continue its research activity towards diamond power electronics.
3. To design and develop high quality, cost-effective package solutions suitable for the robust operation of 6.52028kV and 10 kV rated devices under extreme conditions
This task is completed. The package type is SOT-227, and base plate, die-attach and topside interconnections have been developed and tested. Packaging is limited to 250°C, higher that all existing packaging, because of the lack of suitable encapsulant.
4. To develop reliability tests for normal and extreme operating conditions of the developed power devices 2028
The measurement probe stations and equipment are ready to measure any device or diamond transistor prototype.
5. To demonstrate a high voltage three-phase DC/AC high-power converter based on diamond devices 2028
A 30kV – 2W hybrid gate driver has been designed and a first prototype has been realized. Three possible topologies have been considered to deploy a HVDC or MV application. However, due to the lack of diamond power transistor, only a SiC based converter has been fabricated.
6. To disseminate the outcomes within EU and provide the foundation for industrialisation
Results have already been published and presented in scientific conferences.
The most important important event was organised just before confinement on March 22020 in Brussels. For the first time a meeting gathered companies, international bodies and academics to discuss about diamond power electronics for energy applications. In particular, for the first time a roadmap to the first industrial diamond power converter was presented, and 3 companies presented their views about the opportunities for diamond electronics for their company.
Very interesting ideas came up during the discussion time, in order to continue the research activity and transfer the technology to the companies.
Regarding the exploitation of the results, the most important is that R&D will continue toward diamond electronics, more efficient and reliable than any other technology. Activity will be reduced, or re-oriented towards other target (harsh environment signal processing electronics, aeronautics, …) because of the lack of suitable financial support. But activity will continue.
Another important element is that USA just started a 12M€ project for diamond power electronics. So technology will be developed and will emerge in the industry, led by USA or EU depending on the financial support GreenDiamond partners could obtain to continue this project at a larger scale.
GreenDiamond, as the largest international project on diamond electronics, significantly contributed to the development from TRL2 to TRL4 of promising technologies for Power Electronics with no negative environmental, resource or safety issues. The vision of the consortium is to enable
highly integrated, cost competitive and reliable diamond power electronics devices and components to improve functionality, reliability, and efficiency of next-generation, low-carbon components and systems.
highly integrated, cost competitive and reliable diamond power electronics devices and components to improve functionality, reliability, and efficiency of next-generation, low-carbon components and systems.