Development of Life Cycle Inventory Modules for Semiconductor ProcessingEPA Grant Number: R828208
Title: Development of Life Cycle Inventory Modules for Semiconductor Processing
Investigators: Murphy, Cynthia F. , Allen, David T.
Institution: The University of Texas at Austin
EPA Project Officer: Klieforth, Barbara I
Project Period: April 1, 2000 through March 31, 2003 (Extended to June 1, 2004)
Project Amount: $325,000
RFA: Technology for a Sustainable Environment (1999) RFA Text | Recipients Lists
Research Category: Pollution Prevention/Sustainable Development , Sustainability
The primary objective of the proposed project is to develop generic, use cluster, life cycle inventory (LCI) modules for activities performed during the manufacture of integrated circuits (ICs). This research is intended to facilitate the establishment of standards, encourage the development of predictive rather than historical life cycle analyses, and potentially simplify communication along the materials/product supply chain. The creation of generic rather than product/process specific modules is intended to focus the effort on the gathering and analysis of data that are relatively independent of time and space (i.e., data that will not become obsolete as technology changes are made and which may be applied to multiple manufacturing sites). This will foster standardization and encourage use of the modules by the industry in general rather than by a single company. Generic modules are also less likely to contain sensitive or proprietary manufacturing information and may decrease concerns about sharing of confidential information along the supply chain.
The proposed research will lay the groundwork and develop methodologies for gathering and analyzing data in the area of environmental merit. The intent is to begin the process of capturing and analyzing LCI data such a way that it is immediately useful, while at the same time has extendability across the industry, along supply chain lines, and into the future technologies. A high-level approach will be used to overcome some of the extreme complexity and time barriers that have historically associated with similar efforts. Motorola and SEMATECH, both in Austin, Texas, have agreed to work with the University of Texas to provide technical guidance to the team and permit access to data and process areas as appropriate in the execution of this project.
The general approach to the project will be to create an initial set of use cluster modules and associated LCI modules. The modules will be populated with actual data and validated in a manufacturing setting. Inputs will be described as parametric distributions and sensitivity analyses will be performed using Monte Carlo simulation. The team will then down-select to a small number of use clusters of particular concern or interest. Additional predictive LCI module options, which represent future technologies will be developed for these down-selected modules. These predictive modules will be populated with postulated data and sensitivity analyses will be generated. The results of the predictive LCI modules will be shared with a number of industry representatives. Feedback from these representatives will be incorporated as appropriate.
The set of predictive, generic LCI modules will be disseminated through a significant portion of the semiconductor industry through SEMATECH. Companies that adopt this methodology may use the existing LCI modules and generate additional ones internally to communicate more effectively with their suppliers and with their customers. This is expected to in part satisfy the increasing demand for this type of information along the supply chain and to better meet the needs of the customer. It also provides the basis for a system of standardization so that IC manufacturers are not required to perform separate tests for each customer. While the initial assumptions will be based on current costs, infrastructure, and technology, using metrics that are as independent as possible from specific products and processes will allow these LCI modules to be used widely and for an extended period of time. As the set of assumptions changes, the general methodology can be used to develop new modules.
The training and support of two graduate students will provide them with exposure to the semiconductor industry, a major employer of engineers with advanced degrees in engineering. This level of exposure to the semiconductor manufacturing and its practices is a rare opportunity. The professional contacts the students make will be invaluable to their future careers whether they work in academia or industry.
Expected Improvements to Risk Management: The semiconductor industry can use the life cycle inventory data generated by this project in conjunction with publicly available impact data to identify areas of potential improvement for current processes. The resulting impact assessments can be used to predict, and therefore avoid, negative impacts associated with future processes and/or technologies.