Prepared for presentation at the AIChE 1998 Annual MeetingAIChE Topical Conference on Pollution Prevention and Environmental Risk Reduction Session:Integrated Life Cycle Analysis and Material Accounting for Pollution Prevention

The default impact potentials in P2P are being derived taking into account the necessary trade-offs between data availability and uncertainty. Each potential consists of a fate, exposure and effect component. Unlike a number of comparison tools for life-cycle analysis (LCA) and pollution prevention, P2P takes into account the ability of chemicals to be transported between environmental media after release. The models used to derive the potentials are made readily available to provide designers with the flexibility of extending the database or modifying the default values for case-specific scenarios using minimal resources.

The mass or concentration of an emission is normally restricted by regulations to limit impacts. Legislation is typically process orientated, i.e. addresses emissions from individual facilities. Hence, in theory, adverse impacts associated with single emissions are typically prevented or minimized. Today, impacts are commonly the result of long-term exposure to many sources rather than particular emissions. Therefore all emissions can be considered relevant in pollution prevention, whether above or below a no effect or threshold level. In addition to identifying potentially applicable regulations, P2P Mark III is intended to facilitate the comparison of process and product alternatives in terms of their contribution to potential regional and global scale impacts.

Introduction
The Pollution Prevention Act of 1990 charges the US EPA with the responsibility for developing national pollution prevention (P2) goals, and for devising an approach to measure P2 on a national scale. Similarly, companies are under increasing pressure, in part from the public disclosure of toxic release information required under SARA Title III, to practice P2 and to communicate successes to the public. In response, a computer based system known as "P2P" (Pollution Prevention Progress) was developed (Stephan et al. 1994).

P2P is a user-friendly computer-based tool developed to help quantify the progress achieved by implementing P2 projects. P2 can be achieved through a wide variety of changes in the way a product is designed, manufactured, used or disposed of, or changes in the way a service is rendered. P2P provides a way of structuring and analyzing this information in terms of differences in pollution before and after changes take place to facilitate judgements as to the amounts, types and relative significance of the pollution prevented (or, in some cases, increased). Application domains include process design, "gate to grave" product analysis and product comparison projects based on a complete life cycle analysis. The original program and ongoing revisions are described in Section 2.

Given release (inventory) data associated with a process or product, potential contributions to impacts can be assessed in four stages:

• classificatio
• characterization
• normalization
• valuation

A number of approaches have been presented in the literature to evaluate the significance of a change within and across impact categories. Normalization provides a basis to interpret the significance of a change within a given impact category (Bare 1998). Similarly, valuation is the relative weighting or significance across the different impact categories. Both of these stages are dependent on the study results and the stakeholder�fs preferences. There is currently no scientific approach or consensus and in many case studies these decisions may be intuitive. This multi-objective decision making phase is therefore not automated within P2P. P2P Mark III provides the user with a detailed breakdown of the potential contribution to impacts in each category and the percentage changes.

It should be noted that P2P is for advisory purposes only and is intended to provide a general assessment of the amounts, kinds and relative significance of the pollution that would be prevented (or sometimes increased) as a result of product/process redesign. Reports and values generated by P2P reflect best judgements and do not reflect official US EPA policy.
 

Conclusions
P2P (Pollution Prevention Progress) Mark III is a computer-based tool being developed to facilitate the comparison of process and product designs in the context of their potential contribution to environmental and human health impacts. Unlike risk assessment, actual impacts associated with specific releases are not considered. P2P III provides a relative scale for "beyond compliance" comparisons in terms of "burdens" or contributions to potential impacts.

In addition to regulatory information, eleven impact categories and a number of subcategories are represented in Mark III. The database will include default impact potentials for approximately 2800 chemicals. As this data is to be used to facilitate relative comparisons, a number of analyses are being conducted to ensure consistency and to minimize uncertainty in the approaches and parameters.

The main thrust of P2P Mark III is to develop or identify appropriate relative comparison methodologies that maintain scientific rigor but facilitate the development of a database for a large number of chemicals. Due to the number of chemicals requiring consideration, these efforts are predominantly focused on the development of relative comparison potentials for the human health and ecotoxicological impact categories. However, P2P will rely exclusively on the approaches developed for the TRACI project for the categories of eutrophication, photochemical ozone creation and acidification (Bare 1997). As reasonable consensus exists in terms of ozone depletion and global warming potentials, current approaches will be adopted for these categories.

"Scoring and ranking" approaches, recognized for their relative simplicity and lower data requirements, can introduce significant uncertainty that is typically not quantified. However, the number of chemicals and emissions that can be considered using more scientifically defensible approaches can be limited by the availability of comparable data, particularly in the human health and ecological impact categories. Therefore a number of multimedia fate-exposure-effect approaches are being developed and evaluated in terms of uncertainty to facilitate the derivation of potentials for relative comparison in the absence of complete data sets or by using correlation predictions.