Grantee Research Project Results
2004 Progress Report: Comprehensive Tools to Assess Environmental Impacts of and Improve the Design of Semiconductor Equipment and Processes
EPA Grant Number: R831456Title: Comprehensive Tools to Assess Environmental Impacts of and Improve the Design of Semiconductor Equipment and Processes
Investigators: Dornfeld, David
Institution: University of California - Berkeley
EPA Project Officer: Richards, April
Project Period: December 5, 2003 through December 4, 2006
Project Period Covered by this Report: December 5, 2003 through December 4, 2004
Project Amount: $324,969
RFA: Technology for a Sustainable Environment (2003) RFA Text | Recipients Lists
Research Category: Sustainable and Healthy Communities , Pollution Prevention/Sustainable Development
Objective:
With rapid growth in the electronics and computer sectors, environmental and health issues associated with semiconductor manufacturing are growing in importance. The high economic value associated with the cost of acquisition, use, and disposal of water, energy, chemicals, and materials leads to strong incentives to reduce resource use and minimize waste. Additionally, semiconductor manufacturing requires the mobilization of large flows of materials and energy in other industries throughout the economy leading to a wide range of upstream environmental impacts. Also, microchips are used in numerous ways and their contribution to environmental impacts during the use and disposal of electronic products should be quantified. The objectives of this research project are to: (1) develop a comprehensive design for environment (DFE) tool to assess the environmental and health impacts of semiconductor manufacturing, (2) feed this information back into semiconductor equipment and process development cycles, and (3) promote the broader use of this tool to support industrial ecology.
Progress Summary:
The work in Year 1 of the project has emphasized an experimental approach. This work builds on previous research on bottom-up semiconductor DFE tools at Berkeley, specifically the environmental value system analysis tool, EnV-S. The tool is comprehensive in terms of: (1) scope—by considering upstream life-cycle impacts and facilitating integration into downstream environmental assessments of electronics; and (2) metrics—by supporting a wide range of local and global environmental and health metrics. Ultimately, feedback loops from the tool will be constructed to inform the design of equipment and processes. A key element of the development and application of the tool, however, is the extent to which outputs of this analysis could also be used (with varying scope) to inform environmental decisionmaking by regulators, industry suppliers, utilities, etc.
The work to date has assessed several significant intellectual hurdles, for example: What level of detail is required to insure implementation? How can we adequately link upstream life-cycle impacts and facilitate downstream environmental assessments of electronics? What local and global environmental metrics are needed (especially for health hazard issues)? Can we design the tool to be effective in policy planning by regulators and utilities? Our experience so far with a much reduced scope of effort in EnV-S and our close collaborations with industry have so far given some insight to the first two, linking upstream impacts (life cycle analysis) and metrics. This has been done in collaboration with the semiconductor industry.
Review of Results to Date - Supply Chain (Upstream and Downstream) Impact Assessment
To evaluate total environmental impacts associated with semiconductor manufacturing, it is necessary to evaluate impacts associated with the entire supply chain. Environmental analysis in semiconductors has therefore been progressing towards life cycle assessment (LCA). Given the complexity of semiconductor processes and the rapid rate of change, it is essential to develop LCA tools and studies that can be adapted rapidly with changes in technology and that can account for absent or inconsistent data. Currently, a generic LCA tool to support semiconductor manufacturing does not exist.
Detailed life cycle studies may also require significant time to complete. And it is challenging to ensure that the boundary of the assessment is comprehensive and that most significant contributions to the life cycle inventory have been accounted for.
We base our analysis directly on individual equipment sets. We also seek to develop methodologies for the collection of life cycle data throughout the semiconductor supply chain.
To summarize, several obstacles to effective semiconductor LCA exist, including: (1) significant effort and time required to perform detailed life cycle inventory analyses; (2) boundary problems related to inclusion of all pertinent industries and secondary flows; (3) updating studies as equipment, processes, and chemicals change; and (4) tracking environmental impact data (resource and waste) throughout the supply chain and quantifying uncertainty in results.
This project has adopted a framework to address some of these LCA challenges through the use of a hybrid approach, incorporating Society of Environmental Toxicology and Chemistry and economic input-output strategies. We are applying this hybrid LCA approach to assess the life cycle environmental impacts of a state-of-the-art logic device (~130 nm technology node) fabricated on 300 mm wafers with copper interconnects.
An equipment centric approach offers a convenient way to build bottom-up environmental analyses for the entire manufacturing facility. For the purposes of this work, four key manufacturing modules are identified: shallow trench isolation, gate stack, via, and interconnect. Each module consists of numerous individual process steps performed on different equipment sets.
To perform this analysis, we organize facility mass/flow data according to process and equipment sets to construct a library of environmental process models. For each process step, a hierarchy of up to four data components may be required (Figure 1): (1) recipes for individual process steps, which includes information on identity of and mass of chemicals and other resources used (electricity, ultra-pure water); (2) other information associated with the process equipment, such as consumables used by the tool (brushes, pads), number of process chambers, equipment idle time, etc; (3) point-of-use infrastructure requirements, where each process tool is supported by facility infrastructure located below the cleanroom floor (referred to as the facility gray area or subfab) and it is necessary to track mass/flow data associated with equipment in the subfab, such as pumps, abatement devices, monitoring equipment, gas panels, etc; and (4) backpad infrastructure, where emissions from equipment may require additional treatment at the overall facility scale prior to discharge. Also, resources used by equipment may require additional facilities support. We track mass/flow data associated with systems, such as fluoride waste treatment, house scrubbers, ultra-pure water generation, etc.
Figure 1. Hierarchy of Equipment Models
This approach allows us to analyze different recipe data, emissions data, and facility infrastructure options (pumps, abatement devices, HVAC systems, etc.) for each process step. Such a hierarchical organization of data offers three advantages: (1) It allows a separation of proprietary information (from recipes, for instance) from nonproprietary, generic process tools (such as facilities infrastructure), or shared information among companies. (2) It separates databases that change rapidly (such as recipe databases or platform models) from facility infrastructure data and models that may change less rapidly. (3) The data collected and models developed may be used (to the extent possible) in inventory analysis of different product types.
A hybrid LCA approach is then used to evaluate upstream effects (Figure 1). Hybrid LCA is an emerging approach that involves developing detailed inventories and process models for specific items and using a input-output LCA method for other inventory items. A schematic of the proposed approach is shown in Figure 2 below.
Figure 2. Hybrid LCA Approach for Semiconductor Manufacturing
Future Activities:
Our research will continue to develop along the lines outlined above. Another objective of this research is to promote the broader use of this tool to support industrial ecology. As part of this broader use, a second complimentary study will include the automotive industry with specific interest in alternative energy sources for use in the industry and assessments of manufacturing process alternatives for reducing energy and environmental impacts of manufacturing processes. We are discussing with two automakers, one domestic and one European, to set up means for accessing data for use in this area of the study. The intention is to base our analysis on the EnV-S software and to adapt and evaluate the software for this sector also. That will indicate the extent to which the software can address broader manufacturing issues across different industries.
Journal Articles:
No journal articles submitted with this report: View all 12 publications for this projectSupplemental Keywords:
ecological effects, chemicals, integrated assessment, life-cycle analysis, alternatives, clean technologies, innovative technology, waste reduction, environmentally conscious manufacturing, pollution prevention,, RFA, Scientific Discipline, TREATMENT/CONTROL, Sustainable Industry/Business, Environmental Chemistry, Sustainable Environment, Technology, Technology for Sustainable Environment, Environmental Engineering, industrial design for environment, clean technologies, cleaner production, environmental hazard assessment, environmentally conscious manufacturing, waste minimization, semiconductor industry, alternative materials, industrial ecology, semi-conductor processing, electronics industry, environmentally benign alternative, Design for EnvironmentRelevant Websites:
http://me.berkeley.edu/lmas/LMAS_Web/lmas/news.html Exit
Progress and Final Reports:
Original AbstractThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.