화학공학소재연구정보센터
학회 한국재료학회
학술대회 2018년 가을 (11/07 ~ 11/09, 여수 디오션리조트)
권호 24권 2호
발표분야 A. 전자/반도체 재료 분과
제목 Aluminium Oxide Based Resistive Random Access Memory  for Neuromorphic Devices  
초록 Recently artificial intelligence (AI) technologies have been used widely used in various ways, such as image-voice recognition and industry related with data analysis. They have played an important role in artificial neural networks imitating human brain.[1] In spite of these exceptional performances, there is a limit to the artificial intelligence which works in the traditional Von Neumann structure. It is inefficient to separately process and save a lot of data information. To overcome these problems, many researches have been conducted to implement the functions of neurons and synapses. We investigated the characteristics of aluminium oxide (AlOx) based resistive random access memory (ReRAM). AlOx based ReRAM has many advantages to be used as a neuromorphic memory. It has good CMOS compatibility as well as its process is simple. It was fabricated with the simple structure of metal-insulator-metal (MIM), and AlOx  was used as an active layer, which was sandwiched between an inert electrode Pt and W. We measured not only I-V curve but the integration and firing characteristic. It demonstrated the set voltage of –2V, the reset voltage of 2.5V, high resistance state(HRS) current of 4.3x10-9 A and low resistance state(LRS) current of 7.32x10-4 A. Finally, we present how the device can operate as a neuromorphic device. In addition, the uniformity and the reliability of the device such as endurance cycling and retention time were investigated.


[1] Burr, G. W. et al. Neuromorphic computing using non-volatile memory. Adv. Phys. X 2, 89–124 (2017).

*This research was supported by Nano·Material Technology Development Program through the National Research Foundation of Korea(NRF) funded by Ministry of Science and ICT (NRF-2016M3A7B4910249)
저자 김지연, 권기현, 김동원, 진수민, 김혜지, 양훈모, 박재근
소속 한양대
키워드 <P>aluminium oxide; neuromorphic device</P>
E-Mail