Search engine for discovering works of Art, research articles, and books related to Art and Culture
ShareThis
Javascript must be enabled to continue!

Mask Distortion Analysis for the Fabrication of 1 GBit Dynamic Random Access Memories by X-Ray Lithography

View through CrossRef
In order to fabricate 1 Gbit dynamic random access memories, or DRAMs, with 0.15 µm minimum features using X-ray lithography, the total overlay error must be no more than 0.05 µm. We assign 0.03 µm to the budget for mask distortion overlay error, which can be subdivided into fabrication-process-induced distortion (0.024 µm), fixturing-induced distortion (0.01 µm), and X-ray exposure-induced distortion (0.01 µm). We study, through theoretical models, these sources of distortion. In our mask-making process, a 75-mm-dia., 0.6-mm-thick Si wafer coated with 1-µm-thick SiC is direct-bonded to a 100-mm-dia., 4-mm-thick Si frame. Tungsten absorber is patterned on the SiC film, and then the wafer is back-etched to form the membrane window. The mask frame opening is used as the back-etching mask. We found that in order to meet the budget for the fabrication-process-induced distortion for a 46-mm-dia. SiC membrane (corresponding to the area of two 1 Gbit DRAM chips) and a target membrane stress of 1 × 109 dyn/cm2, the nonuniformity in the membrane stress or thickness must be under 2 % and the absorber stress no more than 5 × 107 dyn/cm2. High X-ray exposure power leads to a temperature rise of the mask. We found that as long as the X-ray exposures are performed in helium, the distortion in the printed image is within budget, even at high exposure intensities (340 mW/cm2 absorbed power). We also determine that a properly designed three-point mask-holding fixture is sufficient to meet the budget for fixturing induced distortion.
Title: Mask Distortion Analysis for the Fabrication of 1 GBit Dynamic Random Access Memories by X-Ray Lithography
Description:
In order to fabricate 1 Gbit dynamic random access memories, or DRAMs, with 0.
15 µm minimum features using X-ray lithography, the total overlay error must be no more than 0.
05 µm.
We assign 0.
03 µm to the budget for mask distortion overlay error, which can be subdivided into fabrication-process-induced distortion (0.
024 µm), fixturing-induced distortion (0.
01 µm), and X-ray exposure-induced distortion (0.
01 µm).
We study, through theoretical models, these sources of distortion.
In our mask-making process, a 75-mm-dia.
, 0.
6-mm-thick Si wafer coated with 1-µm-thick SiC is direct-bonded to a 100-mm-dia.
, 4-mm-thick Si frame.
Tungsten absorber is patterned on the SiC film, and then the wafer is back-etched to form the membrane window.
The mask frame opening is used as the back-etching mask.
We found that in order to meet the budget for the fabrication-process-induced distortion for a 46-mm-dia.
SiC membrane (corresponding to the area of two 1 Gbit DRAM chips) and a target membrane stress of 1 × 109 dyn/cm2, the nonuniformity in the membrane stress or thickness must be under 2 % and the absorber stress no more than 5 × 107 dyn/cm2.
High X-ray exposure power leads to a temperature rise of the mask.
We found that as long as the X-ray exposures are performed in helium, the distortion in the printed image is within budget, even at high exposure intensities (340 mW/cm2 absorbed power).
We also determine that a properly designed three-point mask-holding fixture is sufficient to meet the budget for fixturing induced distortion.

Related Results

Software caching techniques and hardware optimizations for on-chip local memories
Software caching techniques and hardware optimizations for on-chip local memories
Despite the fact that the most viable L1 memories in processors are caches, on-chip local memories have been a great topic of consideration lately. Local memories are an interest...
Effects of Contextual Cues on False Memory: A Comparative Experimental Approach
Effects of Contextual Cues on False Memory: A Comparative Experimental Approach
Research on false memory formation using the Deese-Roediger-McDermott (DRM) paradigm has been extensively conducted in Western contexts. Yet, a significant gap remains in experimen...
Development of An Accurate Optical Proximity Correction System for 1 Gbit Dynamic Random Access Memory Fabrication
Development of An Accurate Optical Proximity Correction System for 1 Gbit Dynamic Random Access Memory Fabrication
In this paper, we demonstrate the feasibility of a refined one-dimensional (1D) optical proximity correction (OPC) method incorporated into a hierarchical mask data processing syst...
Inlet Distortion Studies on a Centrifugal Compressor
Inlet Distortion Studies on a Centrifugal Compressor
Abstract Centrifugal compressors can be subjected to non-uniformity in inlet flows due to complicated inlet pipe geometries arising from space constraints and the wa...
0.15 µm Electron Beam Direct Writing for Gbit Dynamic Random Access Memory Fabrication
0.15 µm Electron Beam Direct Writing for Gbit Dynamic Random Access Memory Fabrication
This paper describes 0.15 µ m electron beam (EB) direct writing techniques for Gbit dynamic random access memory (DRAM) fabrication. In order to use EB direct writing for reliab...
Effect of Rear Engine Concept Distortion on the Aerodynamic Performance of a Fan Rotor
Effect of Rear Engine Concept Distortion on the Aerodynamic Performance of a Fan Rotor
Abstract The civil aviation industry is facing serious environmental and energy issues, and boundary layer ingestion technology (BLI) is a viable solution. However, ...
Critical levels of mask efficiency and of mask adoption that theoretically extinguish respiratory virus epidemics
Critical levels of mask efficiency and of mask adoption that theoretically extinguish respiratory virus epidemics
AbstractUsing a respiratory virus epidemiological model we derive equations for the critical levels of mask efficiency (fraction blocked) and of mask adoption (fraction of populati...

Back to Top