Tokyo Electron Pushes the Limits of Chip Density with Simulation
Derek Bassett and Michael Carcasi
Tokyo Electron America
Austin, TX
Chip manufacturers have been adhering to Moore’s law, a law stating that the number of transistors that can be economically placed on an integrated circuit doubles every year, since 1965. As the number of transistors increases, however, the process of manufacturing integrated circuit, called photolithography, becomes more difficult. Each device requires approximately 200 cleaning and photolithography steps and any failures in this process can cost millions.
Tokyo Electron America (TEL), a producer of manufacturing tools vital to the processing of integrated circuits, used simulation to understand a type of failure called pattern collapse. Pattern collapse occurs when cleaning fluid between two features evaporates and the changing surface tension occurring during the evaporation causes the features to bend. Ideally, those features return to their normal shape, but sometimes they are permanently deformed.
Using COMSOL Multiphysics, researchers at TEL created a 2D structural mechanics model based on a series of steady-state calculations with the surface tension forces as boundary conditions. They compared those results with experimental data from literature and found that the model accurately predicted the critical aspect ratio for collapse.
Suggested Stories
Optimizing Ultrasound Piezo-disk Transducers
Esaote S.p.A. produces medical diagnostic systems, including the manufacturing and sales of ultrasound imaging ...
Increasing Lifespans of High-Power Electrical Systems
ABB Semiconductors increased the lifespan of an insulated-gate bipolar transistor (IGBT) using multiphysics ...
Adding Quantitative Systems Pharmacology to Pharmaceutical Science Curriculum
The FDA is encouraging pharmaceutical companies to include simulation in their product development cycles, so ...
Nuclear-Safety-Related SQA Procedure Automation with Custom Applications
In an environment as highly regulated and potentially dangerous as the nuclear industry, software quality assurance ...
