United States National Risk ManagementEnvironmental Protection Research LaboratoryAgency Cincinnati, OH 45268Research and Development EPA/600/SR-97/082 October 1997

Project SummaryLife Cycle Design of Milk andJuice Packaging

David V. Spitzley, Gregory A. Keoleian, and Jeff S. McDaniel

A life cycle design demonstrationproject was initiated between the U.S.Environmental Protection AgencyNational Risk Management ResearchLaboratory, Dow Chemical Company,and the University of Michigan to in-vestigate milk and juice packaging de-sign. The primary objective of thisproject was to develop design metricsand guidelines for environmental im-provement of milk and juice packagingsystems. Material production energy ac-counted for a large portion of the totallife cycle energy for these systems.Conversely, post-consumer waste wasresponsible for a majority of their lifecycle solid waste generation. Packag-ing systems were also evaluated withrespect to key performance criteria, lifecycle costs, and regulatory trends atthe local, state and national levels. En-vironmentally preferable containerswere identified, and tradeoffs and cor-relations between design criteria werehighlighted.

This Project Summary was developedby EPAs National Risk ManagementResearch Laboratory, Cincinnati, OH,to announce key findings of the re-search project that is fully documentedin a separate report of the same title(see Project Report ordering informa-tion at back).Introduction

Integration of environmental consider-ations into the design process representsa complex challenge to designers, man-agers and environmental professionals. Alogical framework including definitions,objectives, principles and tools is essen-tial to guide the development of more eco-

logically and economically sustainableproduct systems. In 1991, the U.S. Envi-ronmental Protection Agency collaboratedwith the University of Michigan to developthe life cycle design framework. Using thisframework, environmental, performance,cost and legal criteria are specified andused to investigate design alternatives.

A series of demonstration projects withDow Chemical Company, Ford Motor Com-pany, General Motors Corporation, UnitedSolar and 3M Corporation have been initi-ated with Cleaner Products through theLife Cycle Design Research CooperativeAgreement. Life cycle assessment and lifecycle costing tools are applied in thesedemonstration projects in addition to es-tablishing key design requirements andmetrics. This is a summary report of theDow Chemical packaging project that in-vestigated the life cycle design of milkand juice containers.Project Description

This study considered the life cycle as-pects of both milk and juice packaging forsale to households. Packages used forthe delivery of fresh dairy milk and/or re-constituted orange juice were selected forstudy. Systems for delivering milk and juiceto on-site users, such as school lunchprograms, were not included in this study.Additionally, this study did not addressimpacts associated with beverage produc-tion.

A total of nine different container typeswere included in this study. Glass bottles,HDPE bottles, paperboard gable-top car-tons, flexible pouches, polycarbonatebottles, aseptic cartons, PET bottles, steelcans, and composite cans were studied

2using previously published life cycle in-ventory data. Many of these containerswere included with various sizes, refillrates, and recycling rates in the study.Although complete inventory data werenot available for PET bottles, steel cans,or composite cans; these containers wereonly partially analyzed based on data avail-ability.

In order to compare containers on anequivalent use basis, a functional unit of1000 gallons was selected. All criteria wereevaluated based on quantities necessaryto deliver 1000 gallons to the consumer.

Although both juice and milk contain-ers were included in all aspects of thestudy, for some categories, only resultsfor milk containers are presented in thissummary since more complete informa-tion was available for these containers.

ResultsA multicriteria analysis of the perfor-

mance, cost, environmental, and regula-tory issues influencing each container sys-tem was conducted. The scoring table,

(Table 1) summarizes the results of thisanalysis. In this table the life cycle dataare normalized to arrive at life cycle scoresranging from 0 (best) to 10 (worst). Scoresrepresent the ratio of the data for a spe-cific container relative to the highest re-ported value in the same category. A com-plete explanation of the scoring system iscontained in the full report. The followingsections briefly examine each assessmentcategory. Scores based on energy useand solid waste data were averaged todetermine the total environmental score.

EnvironmentalEnergy Use

Total life cycle energy for selected con-tainers is shown in Table 2. Material pro-duction energy is also shown for thesecontainers when known.

Our analysis indicated a key designtrend in the container systems; materialproduction accounted for a majority of thelife cycle energy. On average 91% of milkcontainer life cycle energy was consumed

in the material production process, and85% of juice container energy was usedin material production.

Solid WasteFor each container system the published

life cycle solid waste was collected andpost-consumer solid waste data were cal-culated based on container mass. Totallife cycle solid waste values reported in-clude waste from industrial processing inaddition to post-consumer waste. Ouranalysis indicated that for both milk andjuice packaging, post-consumer solidwaste accounts for approximately 80% oftotal life cycle solid waste. Both post-con-sumer and life cycle waste can be greatlyaffected by changing refill rates and unitcontainer size.

The one-gallon, 50-trip refillable HDPEbottle generated the least solid waste overits life cycle (4 kg/1000 gal). In contrast,the single-use, one-liter glass bottle gen-erated the greatest mass of life cycle solidwaste (1220 kg/1000 gal).

Table 1. Milk Container Evaluation

Container Type Energy Use Solid Waste Total Environmental* Cost Performance Overall**

Flexible Pouch 2.1 0.14 1.1 1.1 6.2 2.8Gable Top Carton 10.0 1.1 5.6 1.8 5.0 4.1Glass Bottle refillable 4.9 1.1 3.0 1.2 10.0 4.7 single use 8.8 10.0 9.4 10.0 7.5 9.0HDPE Bottle refillable 2.9 0.05 1.5 0.7 3.8 2.0 single use 9.7 0.55 5.1 3.4 1.2 3.2Polycarbonate 3.3 0.04 1.7 1.0 5.0 2.6

Each energy use, solid waste and cost rating is based on data from the full report, using a scale form best to worst of 0 - 10, where the highest energy, waste and costdata for the selected containers receive a 10 and all other data normalized to this point; performance ratings convert the subjective evaluations in the full report tonumerical values.

* Total environmental score is the average of energy use and solid waste.** Overall score is an average of total environmental, cost and performance.

Table 2. Energy Use, MJ/1000 Gal Delivered

Container Trippage Total LIfe Cycle Material Production

Glass Bottle refillable 20 3900 1910 single use - 7000HDPE Bottle refillable 50 2320 470 single use - 7720 6930Gable Top Carton single use - 8000Polycarbonate refillable 40 2630 1020Flexible Pouch single use - 1700 1550

3Airborne/Waterborne EmissionsDue to the fact that emissions data were

available for only a limited number of con-tainers and these data were highly vari-able, emissions data were not used inevaluation. Both airborne and waterborneemissions are reported, where available,in the full report.

CostCosts representative of the life cycle of

each container product were determinedfrom published information and industrysources. The costs of seven processes/stages in the container life cycle weresummed to arrive at the total life cyclecontainer cost. These costs were: emptycontainer, transportation (fuel use), fillingequipment, municipal waste collection, re-cycling, incineration, and landfill disposal.These costs are shown in Table 3. In thistable, waste collection, recycling, incin-eration, and landfill disposal costs are com-bined to give the End-of-Life cost. Com-plete cost data appear in the full report.Costs for milk and juice containers wereevaluated using the same method.

The price of empty containers accountsfor the majority of total life cycle costs ascalculated by the NPPC. For the con-tainer systems examined empty containercost represented 75% of the total on aver-age. Costs for refillable container systemsare less dependent on empty containercosts than single-use systems.

PerformancePerformance requirements for beverage

packaging were determined with a mul-tiple-step process. First, a literature searchwas conducted to determine which physi-

cal characteristics and other properties in-fluence beverage retailers and consum-ers. Next a set of six performance mea-sures were chosen based on their appar-ent importance. These criteria are clarity,burst/shatter resistance, ease of opening,weight, resealability, and necessity of stor-ing empties.

Each container was then subjectivelyevaluated for the six performance mea-sures and ranked as follows: good (+),neutral (0) or poor (-). Using this rankingthe performance measures were weightedequally to determine overall performance.Additional market research would be nec-essary to establish more accurate weight-ing factors.

Two containers for both milk and juicewere determined to have a good overallranking. The single use HDPE bottle wasthe only container to receive a good rank-ing overall for use with both milk andjuice. Of the milk containers, refillableHDPE was the only other container toreceive a positive ranking overall. The onlyother juice container to receive a positiveranking was the PET bottle. The refillableglass bottle received a poor performanceranking overall for use in both milk andjuice packaging.Legal

A complex set of legal requirementsexist for milk and juice packaging in theUS and other countries. These require-ments vary substantially and have impactsthroughout the life cycle. Legal require-ments detailed in the full report aregrouped into five categories: fees andtaxes, municipal/state/federal goals, bansand mandates, recycling/waste minimiza-

tion requirements, and manufacturer re-quirements. Different packages and mate-rials might be favored under some of theregulations, but none of them optimallymeet every requirement. In general, cur-rent regulations target post consumerwaste, however, the trend is towardbroader more flexible packaging laws.

Design Guidelines Simplified guidelines for evaluating the

environmental performance of milk andjuice packaging were developed based onanalysis of previous life cycle inventorystudies. The following guidelines were pro-posed to address life cycle energy and lifecycle solid waste issues in packaging de-sign and management.

1. Life cycle energy can be approxi-mated by computing the materialproduction energy of the package.For this reason, less energy-inten-sive materials should be encour-aged along with less material-in-tensive containers. For refillablecontainers, high refill rates shouldbe achieved to best exploit the ini-tial energy investment in the pro-duction of the container.

2. Life cycle solid waste is largely deter-mined by post-consumer packag-ing waste; consequently less mate-rial-intensive containers in generalshould be emphasized. The full re-port details the analysis done andthe conclusions drawn from thisanalysis.

ConclusionsThis project used the life cycle design

framework and tools to develop environ-mental and cost metrics for guiding milkand juice packaging design. In addition,analysis of milk and juice container sys-tems highlighted both tradeoffs and someconsistent patterns among criteria andmetrics. Refillable HDPE and polycarbon-ate bottles and the flexible pouch wereshown to be the most environmentally pref-erable containers with respect to life cycleenergy and solid waste criteria. These con-tainers were also found to have the leastlife cycle costs. The strong correlation be-tween least life cycle cost and least lifecycle environmental burden indicates thatthe market system is encouraging envi-ronmentally preferable containers in thesecases. In other cases, significant exter-nalities (environmental burdens) not re-flected in the market system may create abarrier for market penetration of an envi-ronmentally preferable container.

Table 3. Cost, $/1000 gal delivered

Container Trips Empty Container Transportation/Filling End-of-Life Total LC

Glass Bottle refillable 20 $64.00 $37.37 $14.14 $115.51 single use - $773.00 $24.40 $171.39 $968.79HDPE Bottle refillable 20 $45.00 $23.73 $1.21 $69.94 single use - $300.00 $19.97 $8.27 $328.22Gable Top Carton single use - $132.00 $21.43 $21.13 $174.56Polycarbonate refillable 40 $70.00 $23.52 $1.00 $94.52Flexible Pouch single use - $80.00 $19.83 $3.41 $103.24

End-of-Life = recycling, incineration, and landfill disposal

4EPA/600/SR-97/082

United StatesEnvironmental Protection AgencyCenter for Environmental Research InformationCincinnati, OH 45268Official BusinessPenalty for Private Use $300

BULK RATEPOSTAGE & FEES PAID

EPAPERMIT No. G-35

Several performance criteria presentpotential barriers to otherwise preferablecontainers. For example, containers thatrequire significant changes in merchan-dising and/or consumer practices may en-counter market resistance. In the case ofrefillable containers, merchants must ac-commodate returns of refillable containerswhile consumers must be responsible forrinsing and returning them to the grocerystore. In the case of the pouch, perfor-mance issues must be addressed in orderto achieve successful market penetration.A pitcher, which must be cleaned periodi-cally, is required to hold the pouch andfacilitate pouring and storage. Thus, al-though this system is currently popular inCanada, both the pouch and refillablebottles exhibit clear performance tradeoffs.Public education about the environmental

merits of these systems is required toinfluence their acceptance.

Based on the findings of this study andother life cycle assessments, regulationsshould be reviewed to encourage moreenvironmentally preferable packaging.Regulations that support post-consumersolid waste minimization should be en-couraged, but they should not prohibit sys-tems such as the flexible pouch which areamong the most environmentally prefer-able container systems.

Dows overall objective in this projectwas to use the life cycle design frame-work as a method of enhancing their stra-tegic planning capabilities for producingand marketing milk and juice packagingresins. Dows participation in this projectdemonstrates how a material supplier cantake a proactive role in life cycle manage-

ment of its products. Their efforts in lifecycle design enable the company to part-ner with their customers (package fabrica-tors) in a more effective way to both en-hance environmental performance andeconomic success. Partnerships are par-ticularly valuable in addressing the com-plex parameters that affect multiple stagesof a product life cycle.

This report was submitted in partial ful-fillment of Cooperative Agreement num-ber CR822998-01-0 by the National Pollu-tion Prevention Center at the University ofMichigan under the sponsorship of theU.S. Environmental Protection Agency.This work covers a period from November1, 1994 to August 30, 1996 and was com-pleted as of September, 1996.

David V. Spitzley, Gregory A. Keoleian, and Jeff S. McDaniel are with the NationalPollution Prevention Center, Ann Arbor, MI 48109-1115.

Kenneth R. Stone is the EPA Project Officer (see below).The complete report, entitled "Life Cycle Design of Milk and Juice Packaging,"

(Order No. PB98-100 423; Cost: $25.00, subject to change) will be available onlyfrom:

National Technical Information Service5285 Port Royal RoadSpringfield, VA 22161Telephone: 703-487-4650

The EPA Project Officer can be contacted at:National Risk Management Research LaboratoryU.S. Environmental Protection AgencyCincinnati, OH 45268