부산대, 값비싼 백금 대신 구리로 AI 반도체 자기 제어 효율 10배 높였다

  • 입력 2026.10.22 13:58
글자 크기
프린트
이 기사를 공유합니다
부산대학교 제공

A research team led by Professor Se-Young Jeong at Pusan National University has developed a key technology for next-generation energy-efficient AI semiconductors by replacing an expensive platinum with more cost-efficient copper.

The technology forms a thin oxide layer on the surface of copper, enabling the efficient conversion of electrical current into signals that control magnetization.

The research team precisely oxidized the copper foil, creating a highly uniform and well-defined interface between copper and copper oxide.

As a result, the core capability for electrically controlling magnetization was enhanced by approximately tenfold compared with naturally oxidized copper.

The efficiency of magnetic control reached a level comparable to that of platinum, a commonly used but more expensive material. At the same time, the researchers achieved higher spin-torque conductivity, reaching a world-leading level in this key performance metric.

By replacing expensive metals with copper, the researchers were able to address the long-standing challenges of high power consumption and heat generation. These achievements are expected to pave the way for more energy-efficient AI semiconductors and next-generation memory devices.

[Editor's note = This video was produced using InfoX, Yonhap Infomax's AI-based video content production system, and Yonhap has not edited the content in any way, nor does it reflect the editorial direction of Yonhap.]

Source: Pusan National University

◆ 한눈에 보기

· 발행 기관 : 부산대학교

· 배포 : 2026년 10월 22일 오후 1시 36분

출처 : 부산대학교 보도자료

· · · · · · · · · ·

한국사회복지저널은 AI 기술의 도움을 받아 보도자료와 복지 정보를 보다 쉽고 정확하게 전달하기 위해 노력하고 있습니다.

머드노블 — 글을 읽고, 고르고, 이야기 속을 걷는다

저작권자 © 한국사회복지저널 무단전재 및 재배포 금지
개의 댓글
0 / 400
댓글 정렬
BEST댓글
BEST 댓글 답글과 추천수를 합산하여 자동으로 노출됩니다.
댓글삭제
삭제한 댓글은 다시 복구할 수 없습니다.
그래도 삭제하시겠습니까?
댓글수정
댓글 수정은 작성 후 1분내에만 가능합니다.
/ 400
내 댓글 모음
모바일버전