US 12,326,711 B2
Method and computing device for manufacturing semiconductor device
Sooyong Lee, Yongin-si (KR); Mi-Jin Kwon, Anyang-si (KR); Dongho Kim, Hwaseong-si (KR); and Seunghune Yang, Seoul (KR)
Assigned to SAMSUNG ELECTRONICS CO., LTD., Suwon-si (KR)
Filed by SAMSUNG ELECTRONICS CO., LTD., Suwon-si (KR)
Filed on Dec. 30, 2021, as Appl. No. 17/566,151.
Claims priority of application No. 10-2021-0065022 (KR), filed on May 20, 2021.
Prior Publication US 2022/0382249 A1, Dec. 1, 2022
Int. Cl. G05B 19/4099 (2006.01)
CPC G05B 19/4099 (2013.01) [G05B 2219/45031 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method for manufacturing a semiconductor device, the method comprising:
receiving a first layout including a plurality of patterns of the semiconductor device;
generating a second layout by performing machine learning-based process proximity correction (PPC) based on process features of the plurality of patterns of the first layout, the plurality of patterns comprising first-type patterns having a first characteristic and second-type patterns having a second characteristic different from the first characteristic;
generating a third layout by performing optical proximity correction (OPC) on the second layout; and
performing a multiple patterning process based on the third layout,
wherein first process features of the first-type patterns comprise an influence that each of the first-type patterns experiences from neighboring patterns of the first-type patterns,
wherein second process features of the second-type patterns comprise an influence that each of the second-type patterns experiences from neighboring patterns of the second-type patterns,
wherein the performing the multiple patterning process comprises:
performing a first patterning process of etching the first-type patterns; and
performing a second patterning process of etching the second-type patterns,
wherein the performing the machine learning-based process proximity correction comprises:
performing a first machine learning-based process proximity correction based on only the first process features of the first-type patterns, among the first-type patterns and the second-type patterns, and
performing a second machine learning-based process proximity correction based on only the second process features of the second-type patterns, among the first-type patterns and the second-type patterns, and displacement features of the first-type patterns, and
wherein the first machine learning-based process proximity correction and the second machine learning-based process proximity correction are separately performed.