What EPA Means When it Says, "Life Cycle Assessment"

Kenneth R. Stone
Life Cycle Assessment Team Leader
National Risk Management Research Laboratory
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
513/569-7474

The EPA is internally placing heavy emphasis on developing and implementing decision-making tools based on Life Cycle Assessment (LCA). EPA has found instances where a technology intended to reduce wastes has created unanticipated impacts in other media and/or stages of the life cycle. LCA is being developed as a means to identify and deal with these impacts before they occur. LCA differs from other pollution prevention techniques in that it views all the resource and energy inputs to a product (Life Cycle Inventory), as well as the associated wastes, health and ecological burdens (Impact Assessment), and evaluates opportunities to reduce environmental impacts (Improvement Analysis) from cradle to grave. LCA is often confused with other assessment tools, such as DoDs life cycle cost (LCC) and it is sometimes referred to as "environmental life cycle costing." However, as the term is applied by the EPA and the international community, LCA is significantly different from these techniques.

The purpose of this paper is to explain what EPA means when we say, "LCA" and to show how LCA methodology can be used in practical applications. This paper also presents an LCA conducted by EPA and DoD under the Strategic Environmental Research and Development Program (SERDP) to demonstrate difficulties and lessons learned while conducting the inventory, impact and improvement assessments.

Perhaps the simplest way to demonstrate the conceptual basis for LCA is to compare and contrast it with another tool currently being used by the DoD for procurement, entitled, "Life Cycle Costing (LCC)." LCC was developed by DoD during the 1960s and incorporated as policy by DoD Directive 5000.1, "Acquisition of Major Defense Systems," in 1971. The impetus for LCC had developed along with increasingly complicated and technically advanced weapons systems which entailed much higher post-acquisition costs in areas such as training, maintenance, technical upgrade and operation. In this situation, purchasing a system on the basis of lowest bid would not be beneficial if that same system proved to be more costly to deploy and maintain over its life cycle. By taking into account the costs of ownership, DoD developed an ability to better identify the probable cost impacts over the entire life cycle, including situations wherein costs were transferred from one part of the system to another.

If we examine current initiatives to bring pollution prevention and environmental metrics into the acquisition process, we find DoD services expanding their LCC focus to include costs and cost-definable impacts up through the concept and design phases (See the shaded blocks in Figure 1). This approach and the techniques applied thereunder is often referred to as Life Cycle Environmental Assessment (LCEA). However, because the life cycle of the weapon itself (product life cycle) is not fully assessed, this approach is still limited to identifying impacts within the DoD acquisition process.

The white blocks in Figure 1 show what would be brought into the analysis in order to make it a full LCA. Note the addition of the upstream stages, "Raw Materials Acquisition," and "Materials Manufacturing" to the product life cycle. These stages refer to the processes involving the extraction of raw materials (e.g., mining ore, pumping crude oil, etc.) and processing materials necessary to production (e.g., smelting steel, refining oil into kerosene, etc.). Disposal is separated from DEMIL, because a demilitarized part might be re-used or recycled in the civilian market. Adding these stages provides the full product life cycle. In order to integrate environmental considerations in the acquisition, the designer needs to examine the product life cycle throughout the process, beginning during the first stage, "Concept Exploration/Definition."

generally require pretreatment prior to discharge to a POTW to adjust the pH, remove oil, grease and solids, and to precipitate phosphates and inactive chelating agents. Another potential impact is that energy use may be higher than that required for chlorinated solvent degreasers. In this example, environmental impacts have shifted across media and life cycle stages as a result of the change-over to aqueous cleaners. While these deficiencies can be overcome with sufficient technical effort, this example shows the necessity to fully assess a change before implementation to ensure that the consequences of that change can be managed.

When a change to a system or product is contemplated, a comprehensive LCA is a means to identify the cross-media transfer of impacts, and shifts of impacts from one stage to another. A decision-maker may then choose to implement the change and accept the transfer of impacts, looking for ways to improve the system and reduce those impacts. Another possibility is that the LCA will identify an unacceptable impact transfer, and the decision-maker chooses not to make the change, even though it appears to have short term pollution prevention benefits.

METHODOLOGY

LCA methodology, as EPA applies the term, consists of three overlapping analyses: Life Cycle Inventory (LCI); Impact Assessment (LCIA) and; Improvement Analysis (LCImA). However, the first step in every LCA is to set down the goals of the study and scope out the parameters. LCA is an expansive systems analysis methodology and the study must be carefully focused in order to acquire meaningful data. Therefore, the concept of LCA has goal definition and scoping as its center, a necessary first step before the analysis begins. The LCI is an inventory of resource, materials and energy consumed, as well as environmental releases produced for each stage in the life of a product, from raw material extraction to ultimate disposal - this is what is meant by the terms, "cradle to grave" and "dust to dust" so often used in the LCA field (Note: EPA has published a manual for conducting LCIs). After this information has been collected, an LCIA of the environmental and health effects related to resource consumption and environmental releases can be conducted. In fact, the LCIA begins to develop before the LCI is completed as impacts of priority concern are rapidly identified. The LCIA is both a quantitative and qualitative process to catagorize, characterize and value environmental impacts to form a basis for comparison between dissimilar impacts (e.g., global warming vs. ozone depleting chemicals). As the LCIA shapes up, the basis for the LCImA is formed, which identifies and provides a preliminary assessment of the changes needed to reduce environmental burdens of the subject product or process.

The importance of the LCA approach in capturing upstream and downstream impacts can be demonstrated by two examples. In one instance, a chemical alternative designed to replace to methyl ethyl ketone, an EPA-17 chemical, was found to have benzene as an upstream precursor. Since benzene is another EPA-17 chemical, making this change would have eliminated the EPA-17 chemical impact from the operations and support stage by moving it upstream to the raw materials acquisition and materials manufacture stages, which is relevant to the decision to move to the alternative. This information would not have been found within the parameters of an LCEA because the materials manufacture stage is outside the DoD gates. The second example, aqueous cleaning, has been presented. In that example, the impacts were both transferred to another media (aquatic toxicity) and moved downstream to the treatment and disposal stage.

Due to it's responsibilities for national defense and the effective implementation of modern warfare, the DoD has requirements for extremely unique products that have no equal in the private sector. One such product is CARC.

Chemical warfare agents are capable of penetrating the matrix of the alkyd paints that were used on military vehicles before CARC became available. Once contaminated, the paint would tend to slowly release the toxins over time, endangering personnel. Decontaminating a vehicle would require stripping the paint coating away and treating the hazardous waste. By contrast, CARC is an extremely hard coating that resists penetration. This allows a vehicle to be decontaminated in the field without stripping the paint. Further, because the CARC paint must also camouflage military vehicles, it has to deliver a low visibility finish and minimal heat and radar signatures. CARC is used on a variety of military vehicles and equipment from armored tanks and artillery pieces to unarmored vehicles and equipment. Therefore it has to perform well on composite armor, steel, aluminum, fiberglass and wood substrates.

CARC is formulated with a variety of pigments, heavy metals, and solvents. It cures rapidly as a rough, hard, and inflexible coating. The most commonly used primer is a Niles two-part epoxy, thinned with alcohols and xylene.

THE CARC LCA: LIFE CYCLE INVENTORY; IMPACT ASSESSMENT; IMPROVEMENT ANALYSIS

Two Army installations participated in the SERDP study, and originally both had used the Bink's Model 7 spraying system, but due to environmental considerations, had changed over to the Mark 1 HVLP guns. While this change led to better transfer efficiency, it also created several problems. CARC is a much heavier, higher solids paint than found in most commercial applications for which the HVLPs had been designed. Installations typically found problems with plugging, transfer efficiency (e.g., HVLP guns not meeting commercial claims), extended production times to deal with cleaning clogged equipment, and increased use of thinner.

Prior to getting the LCA underway, its purpose must be clearly articulated. LCAs are comparative by nature and are best used when either evaluating a potential alternative to an existing activity, or in seeking to identify what choices are available and evaluating them. When a problem, such as an undesirable environmental impact, has been identified, or when there is an interest in generally improving or upgrading a system, the questions being asked will determine the goal and scope of the study. This process, which may be brief, is nonetheless important because LCA is a systems wide approach that needs to be focused in order to avoid unnecessary investment in time and resources to activities that will not contribute to the final decision. In this case, the LCA was conducted for a baseline CARC paint system, which included typical topcoat, thinner, and primer combinations determined from a survey of 13 major U.S. Army installations. The scoping survey was used to identify a typical CARC paint system, based primarily on the products and techniques used at Ft. Eustis, Virginia.

The Inventory (LCI):

In order to accomplish the goals of the LCA, data must be collected. The LCI methodology collects data by focusing on material and energy balances for each operation within the system and for the entire life cycle of the system. The LCI has to be as complete as possible in order to provide the data relevant to the final decision. In instances where estimates have to be made, they must be conservative and clearly noted. Data quality will likely vary upon the source and care must be taken in order to ensure the data used is the best available and that the sources are contemporary. For CARC, the LCI involved collection of environmental and utility data that describe the production of components for the system, their raw materials, paint application and depainting, and disposal of spent CARC and blast media, including the raw materials used, water and energy, air emissions, liquid wastes, and solid wastes. Where primary process information was missing, engineering estimates were made.

Of the total LCI emissions and wastes, the depainting and painting operations contributed greater than 80 percent of each of the following emissions: methyl isoamyl ketone, zylene, aromatic hydrocarbons, and hazardous solid wastes. The hazardous solid wastes included spent CARC and primer and discarded aluminum oxide from the depainting operations.

All of the criteria air pollutant emissions (CO, TSP, PM₁₀, SO_x, Lead, NO_x), including the majority (88%) of the VOCs, were released during the materials manufacturing stage of the CARC life cycle. Essentially all of the water emissions and solid wastes, except hazardous waste, were also released during the materials manufacturing stage.

The Impact Assessment (LCIA):

Since life cycle assessment is primarily about making comparisons and incorporating dissimilar impacts via the analysis, there has to be a methodology for making the comparisons on an equitable basis. An LCIA examines potential and actual environmental and human health effects from the use of resources (energy and materials) and environmental releases. An LCIA is divided into three phases: Classification, Characterization, and Valuation.

Classification is the process of assigning and aggregating results from the inventory into a relatively homogeneous impact categories. This process involves identifying stressors and organizing them by impact on the ecosystem. For example, carbon monoxide, chlorine and methane are all stressors with the potential to impact the environment under the category of global warming. Classification includes the creation of complex stressor/impact chains, because a single pollutant can have multiple impacts, and a primary impact can result in secondary (or greater) impacts as one impact results in another along the cascading impact chain. For CARC, nine impact categories were selected for characterization: smog formation potential, ozone depletion potential (ODP), acid rain potential (AP), global warming potential (GWP), human health inhalation toxicity, terrestrial toxicity, aquatic toxicity, land use, and natural resource depletion.

Characterization assesses the magnitude of impacts for each of the stressor categories in order to translate LCI data to impact descriptors. For example, BOD data for wastewater discharges may be translated to fish mortality. For CARC, equivalency factors were established and multiplied by the quantity of a resource or pollutant associated with painting 1000 ft² of CARC. A subset of the chemicals identified during the LCI had already been assigned impact equivalency units in published documents. New impact equivalency units were created for some chemicals identified in the baseline or alternative LCIs, by adapting the hazard ranking approach described in an EPA report.

Valuation involved assigning relative values or weights to different impacts, economic and performance measures, so that a total score for all decision criteria could be determined. The valuation method used in this study is known as the Analytical Hierarchy Process (AHP). The AHP process involves a structured description of the hierarchical relationships among the problem elements, beginning with an overall goal statement and developing a decision tree, where the branches of the tree include major and minor decision criteria. Assignment of weights was done as a group exercise, where a four member team was asked to reach a consensus on the weight factors prior to their being entered into the model. Because the team included one cost engineer, one paints/coatings specialist, a civil engineer, and an ecologist, the valuation team mix, and the resulting weights, were considered reasonable.

The Improvement Assessment (LCImA):

The impact valuation scores for the baseline CARC system indicated that the impact categories of greatest concern are ozone depletion potential, acid rain potential, and human inhalation toxicity. The ozone depletion potential impact is primarily due to the carbon tetrachloride air emissions associated with manufacturing Part B of the primer. The acid deposition impact is primarily due to the SO_x air emissions associated with manufacturing Parts A & B of the primer. The human inhalation toxicity impact potential is primarily due to the chlorine air emissions associated about equally with manufacturing Parts A & B of the primer and the CARC topcoat. On the basis of the LCIA, it was determined that the alternatives with the best potential for reductions in these impact categories were the substitution of a water-thinnable primer and use of turbine-powered HVLP spray equipment.

The alternative spray equipment is the Can-am turbine HVLP, which uses turbine technology instead of the traditional method of passing compressed air through a conversion zone in order to convert high pressure, low volume air into HVLP. This technology decreases system turbulence which in turn reduces overspray significantly.

The LCA found that a combination of an alternative primer, thinner and topcoat resulted in the lowest impact across the greatest number of impact categories, although it did not have the lowest impact for aquatic toxicity. This alternative also exhibited a potential cost savings of $230,000 per year for each facility working at the Ft. Eustis level of painting operations. In order to test these conclusions, a technical evaluation was performed at two installations on test coupons and full-sized vehicles. The evaluation supported the LCA's findings and demonstrated that cross-media impact of higher solvent usage by the HVLP guns over their predecessors would be eliminated by the new turbine-based HVLP systems. Further, the turbine HVLP dramatically improved transfer efficiencies, resulting in a 40% reduction in product use. Finally, the new system was well-received by the painters, who saw several benefits in terms of ease of cleanup and operations in the new systems.

PROCESS INCONSISTENCIES BETWEEN SITES

During the study, we found that, despite technical orders and standing operating procedures, CARC painting systems are not consistently applied. Some installations found that, in order to be able to force CARC topcoat through an HVLP gun, it had to be thinned by as much as 20% to avoid plugging problems, although one of the sites in the study made limited use of thinner in the topcoat. Therefore, Army installations would either thin or use CARC directly on the basis of judgement and experience, causing significant variations in environmental impacts. This activity demonstrates how an alternative technology, intended to achieve pollution prevention, can in practice fail to do so because it's use transfers impacts to other media. In this instance the effort to reduce the amount of CARC consumed caused increased consumption of solvents as thinner.

We also found two sites using a lacquer thinner not approved for the CARC system. It seemed to perform better than the approved thinners in the painting process, but the installations had begun to notice a shortening of the life span of the CARC topcoat. The installations in question has since changed over to the approved products.

We also found that some installations would bypass the priming system entirely, using the CARC topcoat as a kind of "unicoat" material. This eliminates emissions associated with priming, but it is not known what long term impacts may be created (e.g., changes in the endurance of the topcoat and the frequency of the painting cycles). The vehicles being painted in this fashion operated in a benign environment - low humidity, no exposure to corrosive salts - which may account in part for the success of this application. It does raise the issue of whether there are instances in which the priming system can be avoided without an operational impact, or perhaps that the painting of some vehicles, such as those reserved for training, can be done without the CARC product.

Applying LCA methodology to CARC resulted in a series of discoveries concerning upstream and downstream impacts, problems in the field not previously known to the designers, variances in procedures and potential improvements for the system. These issues came to light precisely because LCA is more than a gate-to-gate analysis and they raise several concerns that can impact the design process.

For example, while the change to HVLP guns did result in a decreased use of CARC paint via improved transfer efficiency, that impact was offset by an increase in organic solvent usage and extended cleaning operations. The LCA indicates that this may not have been a very good trade-off, particularly since VOC emissions are a major health concern. The LCA did identify a new alternative, the turbine HVLP, which appears to eliminate this offset.

Another example, is the use of CARC as a unicoat product. This activity suggests that it may be worthwhile to investigate the use of CARC across DoD and identify instances wherein the primer system may not be needed or the CARC topcoat might be replaced with a non-CARC paint (e.g., vehicles and equipment that will not leave CONUS). Since CARC is an expensive product, this approach may offer significant cost savings. It is due to the fact that LCA is a systems-wide analysis, that it can identify and "flag" these situations for the benefit of the designers, enabling them to make "real world" decisions that are responsive to how these products are used.

Finally, it appears that for CARC, while the LCEA and LCC methods are a part of weapon system acquisition, there does not seem to be an established method or approach for designers to come back and evaluate the system after it has been deployed, as in the case of LCA. Integrating LCA for this purpose has significant potential for savings on cost and environmental impacts by making design a fully integrated process throughout the product life cycle.