文章信息/Info

Title:

AI-Driven Design of Next-Generation Interconnect Materials for Integrated Circuits: Progress and Challenges

文章編號(hào):

1674-3962(2025)05-0436-14

作者:

崔國(guó)祥;  李瑞;  袁昌馳;  吳蘊(yùn)雯;  鞠生宏

1. 上海交通大學(xué), 材料科學(xué)與工程學(xué)院, 上海, 200240 2. 上海交通大學(xué), 中英國(guó)際低碳學(xué)院, 上海, 201306

Author(s):

CUI Guoxiang;  LI Rui;  YUAN Changchi;  WU Yunwen;  JU Shenghong

1. School of Materials Science and Engineering, Shanghai Jiao Tong University, Shanghai 200240 2. China-UK Low Carbon College, Shanghai Jiao Tong University, Shanghai 201306

關(guān)鍵詞:

互連材料; 二元合金; 二維材料; 人工智能

Keywords:

interconnect materials;  binary alloys;  topological semimetals;  two-dimensional materials;  artificial intelligence

分類(lèi)號(hào):

TN405;TP18

DOI:

10.7502/j.issn.1674-3962.202409016

文獻(xiàn)標(biāo)志碼:

A

摘要:

隨著芯片在通信、汽車(chē)電子與高性能計(jì)算等領(lǐng)域的深入應(yīng)用，低功耗、高性能的芯片需求持續(xù)上升。在摩爾定律推動(dòng)下，器件微型化帶來(lái)量子隧穿效應(yīng)和布線電阻增加等挑戰(zhàn)，尤其在5 nm及以下工藝節(jié)點(diǎn)，芯片互連成為性能瓶頸。Cu互連面臨尺寸效應(yīng)導(dǎo)致的電阻激增，推動(dòng)對(duì)新型低電阻材料的探索。綜述了集成電路互連在先進(jìn)節(jié)點(diǎn)下的核心挑戰(zhàn)，分析Co、Ru等替代金屬及二元合金、拓?fù)浒虢饘佟⒍�維材料的發(fā)展前景，并探討人工智能在互連材料設(shè)計(jì)中的應(yīng)用，為工業(yè)界開(kāi)發(fā)新一代互連材料提供參考路徑。

Abstract:

With the increasing integration of chips into communications, automotive electronics and high-performance computing, the demand for low-power and high-performance devices continues to grow. Driven by Moore’s Law, the device miniaturization presents critical challenges such as quantum tunneling effects and increased wiring resistance. With technology nodes of 5 nm and below, the interconnects have emerged as a key performance bottleneck. Particularly, the copper interconnects face significant resistance increases due to size effects, driving the search for novel low-resistivity materials. This review summarizes the major challenges of interconnect technology at advanced nodes. In details, the potential applications of alternative metals such as cobalt and ruthenium, binary alloys, topological semimetals and two-dimensional materials are evaluated. At the same time, we also explores the application of artificial intelligence in interconnect material design, providing strategic insights for the development of next-generation interconnect solutions in the semiconductor industry.