Improved wafer alignment model algorithm for better on-product overlay

Jeong, Ikhyun; Kim, Hyun-Sok; Kong, Yeong-Oh; Song, Jang Ho; Ju, Jae-Wuk; Kim, Young‐Sik; Lambregts, Cees; Yu, Miao; Rahman, Rizvi; Karssemeijer, Leendertjan; McNamara, Elliott; Böcker, Paul; Choi, Jung Min; Oh, Nang-Lyeom; Lee, Kang-San; Lee, Jin-Seo

doi:10.1117/12.2516259

Cited by 4 publications

(7 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Higher order wafer alignment (HOWA) is usually used to accurately describe the shape and position of the wafer in the scanner. The HOWA model mainly contents the following three parts: (a) physical shape and position of the wafer; (b) model extrapolation error in regions without alignment marks; (c) alignment mark measurement errors 14 . However, in order to accurately correct the grid deviation during alignment, the contents of the model that are not related to (a), namely (b) and (c), should be removed, as they do not reflect the grid deviation (wafer deformation, WD) but are related to the mark deformation (MD) which may be caused by processes, or they will introduce the overlay error.…”

Section: Wafer Alignment Model Mapping Algorithmmentioning

confidence: 99%

“…The optimal wavelength ratio is found by the optimal color weighting (OCW) algorithm, reducing the alignment error caused by the asymmetry of the alignment marks 13 . Fourth, by separating the physical shape and position of the wafer from alignment mark errors and model extrapolation errors, the wafer alignment model mapping (WAMM) algorithm can obviously improve the grid corrections and reduce the overlay variations in the manufacturing process 13,14 . Fifth, The choice of alignment marks can be determined by the sampling scheme optimization (SSO) algorithm which picks out a set of suitable alignment marks and allows for a good balance between the throughput and the alignment accuracy in high volume manufacturing (HVM) 15 .…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Review of overlay error and controlling methods in alignment system for advanced lithography

Hao¹,

Qi²

2022

Thirteenth International Conference on Information Optics and Photonics (CIOP 2022)

View full text Add to dashboard Cite

Acting as one of the three critical indicators to evaluate the performance of lithography scanners, overlay has been a vital factor which seriously affects the electric property, life span and reliability of integrated circuits (IC). With continuous developments of technology nodes and applications of resolution enhancement technologies (RET), current resolution of the scanners has shrunk to 7 nm and beyond, proposing more stringent requirements for measuring, monitoring and correcting the overlay error. To ensure that the overlay budget fully met by the scanner, advanced optical technologies and holistic alignment approaches have been gradually applied to the alignment system of the scanner, revealing that an integrated methodology in which multiple technologies are adopted proved to be valid and effective in reducing the overlay error in present technology nodes.In this article, the exact concept of overlay error is provided, and on this basis different methods to describe the overlay error are comprehensively analyzed from the perspectives of classifications, sources, quality control and correcting models. Then the systematic relationship between alignment and overlay is illustrated. Recent improvements in the alignment system are also the key points here to systematically reduce the overlay error, including the optimization of the alignment optical path, mark types and design procedures, as well as the application of optimal color weighting (OCW), wafer alignment model mapping (WAMM) and sampling scheme optimization (SSO) algorithms. Based on the combined application and iteration of these technologies, it is believed that the on-product overlay (OPO) can be further compressed to support the continuous tightening of the overlay budget in the industry of IC.

show abstract

Section: Wafer Alignment Model Mapping Algorithmmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Review of overlay error and controlling methods in alignment system for advanced lithography

Hao¹,

Qi²

2022

Thirteenth International Conference on Information Optics and Photonics (CIOP 2022)

View full text Add to dashboard Cite

show abstract

“…Schmitt-Weaver et al [22] proposed an efficient highdensity wafer alignment metrology method by combining deep learning and wafer leveling metrology of photolithography equipment. Jeong et al [23] proposed an improved wafer alignment model algorithm for better product overlay. Lee [24] proposed a marker layout to optimize overlay alignment in photolithography.…”

Section: Semiconductor Mask Alignermentioning

confidence: 99%

Improved MSRN-Based Attention Block for Mask Alignment Mark Detection in Photolithography

Park

Jeong

2022

Applied Sciences

View full text Add to dashboard Cite

Wafer chips are manufactured in the semiconductor industry through various process technologies. Photolithography is one of these processes, aligning the wafer and scanning the circuit pattern on the wafer on which the photoresist film is formed by irradiating light onto the circuit pattern drawn on the mask. As semiconductor technology is highly integrated, alignment is becoming increasingly difficult due to problems such as reduction of alignment margin, transmittance due to level stacking structure, and an increase in wafer diameter in the photolithography process. Various methods and research to reduce the misalignment problem that is directly related to the yield of production are constantly being conducted. In this paper, we use machine vision for exposure equipment to improve the image resolution quality of marks for accurate alignment. To improve image resolution quality, we propose an improved Multi-Scale Residual Network (MSRN) that combines Attention Mechanism using a Multi-Scale Residual Attention Block to improve image resolution quality. Our proposed method can extract enhanced features using two different bypass networks and attention blocks with different scale convolution filters. Experiments were used to verify this method, and the performance was improved compared with previous research.

show abstract

“…Dozens of alignment marks are distributed on the entire wafer, and each alignment mark's position is measured by an alignment sensor to obtain wafer distortion. According to the wafer distortion, the exposure process parameters are adjusted to accurately transfer the circuit patterns from mask to wafer [12,13]. The overlay accuracy is generally 1/3 that of the process nodes, and the alignment accuracy is about 1/5~1/3 of the overlay accuracy.…”

Section: Introductionmentioning

confidence: 99%

Adaptive Weighted Error-Correction Method Based on the Error Distribution Characteristics of Multi-Channel Alignment

Song,

Wang,

et al. 2024

Sensors

View full text Add to dashboard Cite

As process nodes of advanced integrated circuits continue to decrease below 10 nm, the requirement for overlay accuracy is becoming stricter. The alignment sensor measures the position of the alignment mark relative to the wafer; thus, sub-nanometer alignment position accuracy is vital. The Phase Grating Alignment (PGA) method is widely used due to its high precision and stability. However, the alignment error caused by the mark asymmetry is the key obstacle preventing PGA technology from achieving sub-nanometer alignment accuracy. This error can be corrected using many methods, such as process verification and multi-channel weighted methods based on multi-diffraction, multi-wavelength and multi-polarization state alignment sensors. However, the mark asymmetry is unpredictable, complex and difficult to obtain in advance. In this case, the fixed-weight method cannot effectively reduce the alignment error. Therefore, an adaptive weighted method based on the error distribution characteristic of a multi-channel is proposed. Firstly, the simulation result proves that the error distribution characteristic of the multi-alignment result has a strong correlation with the mark asymmetry. Secondly, a concrete method of constructing weight values based on error distribution is described. We assume that the relationship between the weight value of each channel and the deviations of all channels’ results is second-order linear. Finally, without other prior process correction in the simulation experiment, the residual error’s Root Mean Square (RMS) of fixed weighted method is 14.0 nm, while the RMS of the adaptive weighted method is 0.01 nm, when dealing with five typical types of mark asymmetry. The adaptive weighted method exhibits a more stable error correction effect under unpredictable and complicated mark asymmetry.

show abstract

Improved wafer alignment model algorithm for better on-product overlay

Cited by 4 publications

References 1 publication

Review of overlay error and controlling methods in alignment system for advanced lithography

Review of overlay error and controlling methods in alignment system for advanced lithography

Improved MSRN-Based Attention Block for Mask Alignment Mark Detection in Photolithography

Adaptive Weighted Error-Correction Method Based on the Error Distribution Characteristics of Multi-Channel Alignment

Contact Info

Product

Resources

About