公开(公告)号：US12203163B2

公开(公告)日：2025-01-21

申请号：US17334630

申请日：2021-05-28

Applicant: Applied Materials, Inc.

Inventor： Goichi Yoshidome ,  Suhas Bangalore Umesh ,  Sushil Arun Samant ,  Martin Lee Riker ,  Wei Lei ,  Kishor Kumar Kalathiparambil ,  Shirish A. Pethe ,  Fuhong Zhang ,  Prashanth Kothnur ,  Andrew Tomko

IPC: C23C14/35 ,  H01J37/34

Abstract: Methods of processing a substrate in a PVD chamber are provided herein. In some embodiments, a method of processing a substrate in a PVD chamber, includes: sputtering material from a target disposed in the PVD chamber and onto a substrate, wherein at least some of the material sputtered from the target is guided to the substrate through a magnetic field provided by one or more upper magnets disposed about a processing volume of the PVD chamber above a support pedestal for the substrate in the PVD chamber, one or more first magnets disposed about the support pedestal and providing an increased magnetic field strength at an edge region of the substrate, and one or more second magnets disposed below the support pedestal that increase a magnetic field strength at a central region of the substrate.

公开(公告)号：US20220245307A1

公开(公告)日：2022-08-04

申请号：US17166965

申请日：2021-02-03

Applicant: Applied Materials, Inc.

Inventor： Prashanth Kothnur ,  Karthik Ramanathan ,  Ajit Balakrishna ,  Kartik Shah ,  Umesh Kelkar ,  Vishwas Pandey ,  Prasoon Shukla ,  Sushil Arun Samant

IPC: G06F30/27 ,  G05B13/02 ,  G05B17/02 ,  H01L21/67 ,  H03H17/02

Abstract: Embodiments described herein include processes for generating a hybrid model for modeling processes in semiconductor processing equipment. In a particular embodiment, method of creating a hybrid machine learning model comprises identifying a first set of cases spanning a first range of process and/or hardware parameters, and running experiments in a lab for the first set of cases. The method may further comprise compiling experimental outputs from the experiments, and running physics based simulations for the first set of cases. In an embodiment, the method may further comprise compiling model outputs from the simulations, and correlating the model outputs with the experimental outputs with a machine learning algorithm to provide the hybrid machine learning model.