KOIZUMI, Satoshi / Group Leader

mail：KOIZUMI.Satoshi@nims.go.jp

KOIZUMI's SAMURAI page

Diamond Semiconductor Research

Overview

Diamond is a wide-gap semiconductor material that is expected to be applied to deep-ultraviolet optoelectronics, power devices, and more.
The NIMS Diamond Research Group established diamond vapor growth technology in the 1980s and n-type semiconductor technology in the 1990s, and has succeeded in forming pn junctions in the 21st century. All of these technologies are world-firsts, and are global research achievements that Japan can be proud of.
Currently, these research projects are continuing and developing as an important part of NIMS's research into ultra-wide-bandgap semiconductors.

Characteristics

Diamond's large band gap of 5.5 eV and the unique property of its surface, the negative electron affinity, have the potential to produce a variety of devices with unique characteristics. Applications that take advantage of this property have been difficult due to imperfect doping and growth control.
We aim to develop technology for controlling impurities and crystal perfection to the level of silicon semiconductors, thereby realizing usable diamond materials.

Major reserch

In our research into semiconductor diamond CVD, we are focusing on n-type doping in particular, and developing technology for controlling materials to suit the specific specifications of various devices, with the aim of forming “usable” semiconductor devices.
In addition, based on the knowledge gained through CVD technology, we are also conducting crystal evaluations at the device level. Hetero devices with Ga₂O₃, etc. are also an important research subject. We are considering applications to power electronics, sensors and semiconductor circuits used in extreme environments.

Summary

• Successfully fabricating various diamond devices through advanced n-type doping
• Exploring diamond vapor phase growth technology as the ultimate power device semiconductor material
• Peer-reviewed papers: 186, patents granted: 7

LIAO, Meiyong

mail：Meiyong.LIAO@nims.go.jp

LIAO's SAMURAI page

Diamond MEMS, Electronic, and Optical Devices and Systems

Overview

Diamond is the ultimate semiconductor material, possessing physical properties that far exceed those of currently used semiconductors. It is expected to deliver unparalleled performance in applications for microelectromechanical systems (MEMS) and electronic devices.
Diamond is the most promising semiconductor candidate for highly sensitive and reliable MEMS sensors (for magnetic fields, temperature, vibration, acceleration, etc.), high-frequency mechanical oscillators, and electronic devices capable of withstanding extreme conditions, including high voltage, high temperature and high radiation that existing semiconductor materials cannot endure.
Our research focuses on the development of high-quality diamond wafers through CVD growth, as well as the advancement of MEMS devices, optical and electronic devices and their systems.

Characteristics

• Leadership in smart-cut single-crystal diamond MEMS technologies
• Development of ultrahigh-quality factor and high-reliability MEMS resonators
• Innovation in diamond MEMS sensing chip
• Demonstration of n-channel diamond transistors

Major reserch

We have achieved a number of cutting-edge research breakthroughs, including batch manufacturing of single-crystal diamond MEMS, the highest-temperature MEMS magnetic sensors, actuators, and diamond transistors.

1. We developed the smart-cut method for diamond cantilevers with a record-high quality factor (Q > 1,000,000) at room temperature and theoretically demonstrated the superiority of diamond MEMS by elucidating its mechanical energy loss mechanisms.
2. We developed a universal single-crystal diamond MEMS chip, which exhibited outstanding properties, including high sensitivity, low operating voltage, and high-temperature operation (600℃).
3. We developed a diamond MEMS magnetic sensor and sensor array capable of operating at high temperatures (500℃) while maintaining high magnetic sensitivity (10nT/Hz^0.5).
4. These achievements are expected to enable applications for wear-resistant scanning probe microscopy, high-precision temperature sensing in microscale, internal combustion engines control, petroleum and mineral exploration. Furthermore, we realized the n-channel diamond MOSFETs, paving the way for the development of diamond CMOS.

Summary

We developed the smart-cut method that is highly reproducible and well-controlled for diamond MEMS (i.e. cantilevers, bridges, membranes). The developed diamond MEMS sensors have been proven to possess significantly higher reliability and sensitivity compared to other semiconductor materials. Microscale high-precision and high-reliability sensor arrays (magnetic, temperature, light, etc) that the existing ones cannot reach are under development.

OSHIMA, Yuichi

mail：OSHIMA.Yuichi@nims.go.jp

OSHIMA's SAMURAI page

Development of high-speed, high-quality epitaxial growth technology for ultra-wide-gap semiconductors

Overview

To realize energy conservation for a decarbonized society, development of low-loss power devices is an urgent issue. Currently, most power semiconductor devices are made using Si, but their performance is close to the material limit. Therefore, there is a need to realize and popularize ultra-wide bandgap (UWBG) semiconductor power devices with higher performance than that of Si.
In this study, we aim to establish high-quality and high-speed epitaxial growth technologies necessary to extract the excellent potential of such new materials and to fabricate high-performance devices with sufficient cost effectiveness.

Characteristics

• Application of halide vapor phase epitaxy (HVPE) for UWBG materials
• Much higher growth rate than that of MOCVD, etc. by effectively preventing parasitic reaction.
• High purity crystal growth technology that contributes to good control of electrical properties
• Defect density reduction technology that makes full use of growth mode control and selective growth

Major reserch

We are aiming to establish the HVPE method, which has a proven track record in high-speed deposition of III-V compound semiconductors such as GaN, for the growth of Ga₂O₃, which is attracting attention as a UWBG semiconductor.
So far, we have achieved ultrahigh-speed growth of over 100 µm/h through our unique reactor design and parasitic reaction suppression technology by optimizing the balance between growth and etching reactions, and have succeeded in growing thick films of over 100 µm using this method. We also succeeded in achieving both selective area growth and high-speed growth, and succeeded in significantly reducing dislocation density.
These results will greatly contribute to the realization of high-voltage Ga₂O₃ power devices with thick drift layers exceeding several tens of µm.

Summary

We have demonstrated ultrafast thick film growth of Ga₂O₃, which is attracting attention as a next-generation high-performance power device material, by HVPE method and defect reduction by selective area growth.
The technique for high-speed growth with suppression of parasitic reactions even in the case of precipitation reactions with large equilibrium constants, obtained in this study, is expected to be applicable to new material systems other than Ga₂O₃.

OSHIMA, Takayoshi

mail：OSHIMA.Takayoshi@nims.go.jp

OSHIMA's SAMURAI page

New structural control of oxide power semiconductors

Overview

To achieve carbon neutrality, the semiconductor industry requires the practical implementation of next-generation power semiconductors that enable ultra-low-loss power devices. Among these materials, Ga₂O₃ and rutile-type oxides (such as SnO₂ and GeO₂) have a relatively short research history. However, compared to more established materials like SiC and GaN, these oxide semiconductors exhibit superior fundamental properties for power device applications and have recently attracted increasing attention.
In this study, we employ selective area growth and selective gas etching techniques on these relatively new oxide power semiconductors to promote facet formation, enabling structural control essential for enhancing device performance.

Characteristics

• Plasma-free microfabrication technology
• High-aspect-ratio structures that cannot be fabricated using conventional techniques
• Device fabrication utilizing plasma-damage-free and high-aspect-ratio structures

Major reserch

As an example of this study, we explain the plasma-free microfabrication of Ga2O3 semiconductors.

This semiconductor is n-type unipolar, and thus it is unable to form a pn homojunction. Therefore, to enhance the performance of unipolar devices, it is necessary to control the electric field in microscopic regions using high-aspect-ratio structures such as fins and trenches. In conventional microfabrication, plasma-based dry etching is commonly used. However, this approach has limitations, including the difficulty of forming high-aspect-ratio structures and the introduction of processing damage to the surface due to reactive ions.
In this study, we demonstrated selective growth and selective gas etching as new processing methods capable of overcoming these challenges. These techniques expose chemically stable faceted sidewalls, enabling the fabrication of fin and trench structures, as shown in the figure, without the use of reactive ions.
Currently, we have also attempted fabricating devices that utilize these fin and trench structures.

Summary

By utilizing crystallographic growth and etching microfabrication techniques, it becomes possible to fabricate high-aspect-ratio structures that were previously difficult to fabricate using conventional techniques. We will verify the effectiveness of this method through device fabrication and evaluation. Additionally, we plan to extend this research beyond Ga₂O₃ to recently proposed rutile-type oxide power semiconductors.

IROKAWA, Yoshihiro

mail：IROKAWA.Yoshihiro@nims.go.jp

IROKAWA's SAMURAI page

Characterization of semiconductor devices

Overview

Reliability of semiconductor devices is a critical issue. For example, ambient atmosphere is known to alter the device characteristics, but the fundamental mechanism of the phenomenon is still unclear.
In this research, the effects of ambient atmosphere on nitride-based semiconductor devices are investigated.

Characteristics

Proposal of the physical model and development of hydrogen sensors

Major reserch

We investigated the metal/semiconductor interfaces. Figure 1 shows Nyquist plots of Pt-AlGaN/GaN SBDs. As shown in Fig. 1, a new semi-circle representing electric double layers are not confirmed. Figure 2 shows the calculated average potential at Pd-SiO₂ interfaces. As shown in Fig. 2, the average potential at Pd-SiO₂ interfaces does not changed after hydrogen introduction.
Based on these results, formally proposed hydrogen induced dipole model would not be the responsible mechanism. Through vigorous study, we have confirmed that dielectric layers at metal/semiconductor interfaces play a critical role.

Summary

Obtained results would be applicable not only to contributing to the reliability of semiconductor devices but also to the improvement of hydrogen sensors.