Environmental Implications of Consumer Convenience

Environmental Implications of Consumer Convenience Coffee as a Case Study Andrea L. Hicks Keywords: coffee consumers disposal industrial ecology life cycle assessment (LCA) products of convenience Supporting information is linked to this article on the JIE website Summary Products of convenience are playing an increasingly large role in today s society. These products provide a competitive advantage over their conventional counterparts by requiring less time and effort to produce a similar service or experience. At the same time, these products are often also more materials intensive to produce and create a greater amount of waste. A comparative midpoint life cycle assessment of different coffee brewing systems is presented in order to explore the comparative impact of three different systems: drip filter, french press, and pod style (a product of convenience). Utilizing a comparative functional unit, the drip filter system method was found to have the greatest environmental impact in all impact categories, whereas the pod style had the least in six of the impact categories (with the french press having the least in two of the impact categories, and a tie between pod style and french press in a single impact category). This suggests that contrary to popular belief, the pod style coffee may be the more environmentally friendly option. The two most significant contributors to environmental impact in all of the categories considered was the amount of dry coffee utilized and the energy needed to brew the coffee, although in some categories considered transportation was also significant. There is the potential for the environmental impact of coffee brewing to shift if coffee wastage occurs (likely in the case of the drip filter and french press system) or if substantial changes in materials or energy consumption were to occur (in the case of the pod-style brewing system). From the perspective of industrial ecology, this analysis suggests that, in regard to products of consumer convenience, the convenient alternative may not have a significantly greater environmental impact than its conventional counterpart, and that it may be time to question that often-held assumption. Introduction Coffee is a ubiquitous beverage in the world today. It is currently grown in 50 countries throughout the world, with its location of production significantly influencing the quality and flavor. Consumption of coffee in America is significant, with 59% of adults drinking it daily and 71% at least weekly (National Coffee Association 2015). The consumption and cultivation of coffee is not new and has been documented since the fifteenth century. However, since 1950, consumption has grown by 200%, with the majority being consumed by Europe and the United States (Salomone 2003). The methods of preparation have also changed significantly, particularly in recent history, moving from slowly brewed batches over open flame to single-serve pods, which brew a single serving in mere seconds. The changes in preparation not only have the potential to affect the quality and taste of the resulting coffee, but also the environmental impact of a serving of coffee. This evolution in preparation method represents a shift to a product of convenience, which will be explored in this work through the lens of industrial ecology (IE). Address correspondence to: Andrea L. Hicks, University of Wisconsin Madison, Department of Civil and Environmental Engineering, Room: 2208, 1415 Engineering Hall, Madison, WI 53706, USA. Email: hicks5@wisc.edu; Web: https://directory.engr.wisc.edu/cee/faculty/hicks_andrea/ 2017 by Yale University DOI: 10.1111/jiec.12487 Editor managing review: Miguel Brandão Volume 22, Number 1 www.wileyonlinelibrary.com/journal/jie Journal of Industrial Ecology 79

The birth of products of consumer convenience in the United States is largely associated with the 1950s and the freeing of women from the kitchen (Neuhas 1999; Szmigin 2006). Products of convenience in the food sector have been characterized by multiple attributes, with the chief attribute being reduced time usage. An accepted working definition that will be used in this work is as follows: Convenience is associated with reducing the input required from consumers in either food shopping, preparation, cooking or cleaning after the meal (IGD Business Publications 2002, 1). A relevant example of a widely adopted convenience product is the introduction of the television dinner by Swanson in 1954 (Wessels Living Farm History 2015). By 1959, consumers were spending $50 million annually (equivalent to $400 million today) on television dinners (Bureau of Labor Statistics 2009), suggesting that in a period of 5 years these products had become widely accepted by consumers. A significant body of research in the food industry has been conducted as to who uses these products of convenience. Brunner and colleagues (2010) surveyed consumers in Switzerland and found a negative correlation with the use of products of convenience and nine household indicators: age (older respondents consumed fewer); sex (women consumed fewer); number of children; working full time; knowledge of nutrition; cooking skills; cherishing naturalness; seeing cooking as women s work; and thinking convenience products are expensive. Buckley and colleagues (2007) categorized consumers into four typologies and ranked their likelihood to use food convenience products: home meal preparer (least: 25%); food connoisseurs (less: 26%); convenience-seeking grazers (more: 33%); and kitchen evaders (most: 16%). Their work suggests that around half the population is more or most likely to use food products of convenience. Quality of the food has historically been an issue with these convenience products; however, in the case of home-brewed coffee, differences in quality attributed to the brewing methods are expected to be minimal. The analysis of consumer products fits squarely within the purview of IE. The 1969 container study by Coca-Cola, which compared different packing systems (such as bottles vs. cans) is often heralded as the first life cycle assessment (LCA) study completed by industry, although the final results were never made public (Hunt and Franklin 1996). In the 1990s, a series of studies compared cloth diapers with disposable diapers (a product of consumer convenience), with significantly divisive results (Franklin Associates Ltd 1990; Lehrburger et al. 1991). Those results, however, illustrated how the tool of LCA could be manipulated to have conclusions reach a particular end, through choices regarding boundaries and inventory assumptions. From that point on, there has been a near constant debate as to the relative environmental impact of disposable and reusable diapers. Multiple studies have concluded that trade-offs exist between the two diapering options (LeVan 1995; Cordella et al. 2015). Products of consumer convenience represent a class of products that has the potential shift how consumers achieve a particular goal, whether it be diapering a baby or brewing a cup of coffee. With the shift in consumption comes the potential for a significant change in the environmental impact of achieving that goal. IE and its tools, such as LCA, allow for the potential shifts in environmental impact to be quantified. This quantification serves as a guide to understanding which products and product components should be targeted for change in order to mitigate their environmental impact. The environmental impact of coffee has been assessed previously in the literature. Coltro and colleagues (2006) developed an environmental profile and inventory of Brazilian coffee, based on data collected from farmers. Hassard and colleagues (2014) found the cultivation and farming of the coffee to be the greatest contributor to the carbon footprint, across multiple methods of coffee reparation in Japan. Busser and Jungbluth (2009) completed an LCA on ground and instant coffee, sold in pouches and stick packs, finding the brewing and coffee production (farming and processing) steps as the most environmentally relevant portions of the life cycle, with the contributions of transportation and retail packaging to be fairly minor. This was also found to be true by Salomone (2003), on a case study of a coffee company in Sicily. De Monte and colleagues (2005) analyzed the environmental impact of different conventional coffee packaging, ranging from metal cans to polylaminate bags (containing various amounts of coffee), excluding the actual production and use of the coffee. Their study found that with respect to the metal cans, landfill disposal was more environmentally costly than recycling, and both disposal options had a greater environmental impact than the raw materials and manufacturing stages of the product life. Humbert and colleagues (2009) compared the environmental impacts of freeze-dried (often called instant coffee ) coffee with drip filter coffee preparation and capsule espresso. They found the freeze-dried coffee to have the least environmental impact when compared to the drip filter and capsule expresso preparations. This suggests that in the case of coffee, a product of convenience, such as instant coffee, may have a smaller environmental impact than conventional coffee brewing technology, although the question exists as to whether this is the case for pod-style coffee. Three methods of coffee preparation drip filter, french press, and pod style (a product of convenience) are investigated in a midpoint LCA to further the understanding of the environmental implications of products of convenience, when compared to conventional products and preparation methods. Methods and Materials The midpoint LCA is conducted utilizing the framework established in the International Organization for Standardization (ISO) 14040 standards (ISO 2006). As such, the methods section of this work is organized into the LCA stages of goal and scope, inventory analysis, and impact assessment. SimaPro software (version 8.0.1) (PRé Consultants 2015) is used to perform the LCA, and the results are reported using the Tool for the Reduction and Assessment of Chemical and othe Environmental Impacts (TRACI) environmental impact categories (EarthShift 2015). Multiple databases are utilized as sources of data, and the exact assumptions and allocations are presented in the Supporting Information found on the journal s website. 80 Journal of Industrial Ecology

Figure 1 System boundaries as included in the study: red (products and processes relevant to all three coffee brewing systems); green (relevant to drip filter and pod-style processes); and violet (relative only to the pod-style process). Goals and Scoping The goal of this work is to determine the life cycle environmental impact of three different methods of brewing coffee: drip filter, french press, and pod style. The first two methods represent conventional options for the brewing of coffee, whereas the third represents a product of consumer convenience. The major motivation for carrying out this study is to address the environmental impact of pod-style coffee relative to other methods of coffee brewing. The functional unit employed is 0.275 liters (L), which is representative of the yield of a single pod of coffee drink, although the environmental impact of each coffee brewing technology will first be presented in its native yield units. The boundaries of the system are presented in figure 1. The system analyzed includes the raw materials acquisition, manufacturing, transportation, use, and end-of-life stages. The figure uses color to illustrate which portions of the system are relevant to each coffee brewing process. The raw materials portion of the life cycles includes the growing of the coffee and the environmental contributions attributed to the plastic, aluminum, and wood pulp (used in the creation of the filters in the drip filter and pod-style coffee processes). In the manufacturing phase, the raw materials are transformed into roasted and ground coffee, a plastic cup, filter, and foil covering (although the energy required to assemble the coffee pod into a single unit is not included because of a current unavailability of data to produce a valid estimate). Transportation of the coffee from the farm to the distribution center is included, as well as the transportation from the distribution center to the point of sale. During the use phase, water and energy are used to brew coffee that is then assumed to be consumed. When the coffee is done being brewed, the coffee grounds, filter, foil, cup, and some residual water are then disposed of. It should be noted that figure 1 presents the boundaries for the most material-intensive and complicated of the brewing options of the single-serve coffee pod. It is still applicable to the other coffee systems analyzed in this work by omitting the components that are not utilized in each method (as indicated by the coloring of the figure). Table 1 presents the sources of data used to create the inventory, along with a brief description of their quality and the time scale of which they represent. An effort was made to use the most recent data available, given that the coffee brewing technology evaluated in this work is relevant to 2015. The data used in this work span the time scale from 2002 through 2015; the oldest data present a detailed inventory of the coffee farming in Brazil based on surveys of the farmers and is a critical component of this inventory. A more detailed assessment of the data used in this work is presented in the supporting information on the journal s website. Life Cycle Inventory Analysis The assumptions made in each stage of the life cycle in order to create the inventory are discussed with respect to each stage. A detailed inventory is presented along with each database allocations in the supporting information associated with this article and hosted on the journal s website. Hicks, Environmental Implications of Consumer Convenience 81

Table 1 Sources of data used to create the inventory, with a brief description and relevant time scale Data Source Brief description of data quality Raw materials Coffee growing Coltro et al. (2006) Life cycle inventory of Brazilian Green Coffee per 1,000 kg, appropriate to coffee grown in Brazil in 2002; data were obtained from farmer surveys. Drip filter Disassembly (table 2) Data obtained experimentally using technology relevant to 2015 French press Disassembly (table 2) Coffee: pod Disassembly (table 2) Manufacturing/transportation Roasting Schwartzberg (2013) Analysis of the potential to save energy through different coffee roasting techniques and conditions relevant to 2013 Grinding Hassard et al. (2014) Carbon footprint assessment of alternative coffee products in Japan; energy consumption use in grinding was obtained from the technical specification of the grinder, relevant to 2010. Transportation Humbert et al. (2009) LCA on different coffee products (including drip filter and instant coffee), transportation distance, and mode assumptions are utilized, relevant to 2008. Use Water for brewing Experimental (table 2) Experimentally determined, using coffee brewing technology relevant to 2015 Drip filter: energy Experimental (table 2) French press: energy Calculated (table 2) Coffee: pod Tear down (table 2) Drip filter: yield Experimental (table 2) French press: yield Experimental (table 2) Coffee pod: yield Experimental (table 2) End of life Drip filter French press Coffee: pod Note: kg = kilograms; LCA = life cycle assessment. Assumption based on observation Filter and grounds to landfill, modeled as biodegradable waste Grounds to landfill, modeled as biodegradable waste Entire pod to landfill; coffee and filter modeled as biodegradable waste; plastic and aluminum foil modeled as inert waste Life Cycle Stages The assumptions regarding each life cycle stage are presented in the following subsections. The raw materials data were acquired through disassembly and measurement of coffee pods and the components of the other coffee brewing systems. Analytical measurements were performed in triplicate. In table 2, the components are presented as a function of the amount of coffee produced natively or the native yield. The yield varies with the method of coffee production utilized. The drip filter coffee is produced is a small conventional coffee pot, which makes around four cups. The french press makes slightly more; however, with this method of coffee production, it is not possible to make a smaller quantity than the press is designed for and have the product be optimal. The coffee machine used with the singleserve coffee pod has three yield settings; the largest was selected for this study. Two functional units are utilized in this work; the first is the native unit, which is the unit that the coffee is brewed in and presented in table 2. The second functional unit is a volume of 0.275 L, or the yield produced by one single-serve coffee pod. Raw Materials Acquisition and Manufacturing The environmental impact of the growing of coffee is modeled using the inventory work of Coltro and colleagues (2006) on Brazilian green coffee. In this case, the coffee beans are wet processed, to remove the fruit covering them before they are dried. The amount of energy used to roast the coffee is obtained from Schwartzberg (2013), using the roasting data for a medium-sized roaster to produce a medium roast, at a cost of 1,621 kilojoules per kilogram (kj/kg) of coffee. The quantity of energy used to grind the coffee is calculated using the work of Hassard and colleagues (2014), at a rate of 550 watts and a throughput rate of 4.7 grams of coffee per second. 82 Journal of Industrial Ecology

Table 2 Disassembly and experimental data Drip filter coffee French press coffee Single serve coffee pod Component Amount Material Component Amount Material Component Amount Material Coffee 52.17 (g) Coffee Coffee 42.75 (g) Coffee Coffee 11.73 (g) Coffee Filter 0.89 (g) Paper Water 0.95 (L) Water Filter 0.27 (g) Paper Water 0.71 (L) Water Heat 323.07 (kj) Electricity Foil 0.24 (g) Aluminum Electricity 0.13 (kwh) Electricity Cup 2.52 (g) Mixed Plastic Water 0.28 (L) Water Electricity 0.02 (kwh) Electricity Yield 0.70 (L) Yield 0.83 (L) Yield 0.28 (L) Note: g = grams; L = liters; kwh = kilowatt-hours; kj = kilojoules. The raw materials acquisition includes the environmental impacts of producing the filter, foil, and cup used to brew the coffee. It does not include the energy used in assembling the components of the cup nor the impact of brewing equipment itself. It is assumed that a single brew is a very small percentage of the lifetime of the equipment (such as the drip coffee makers, french press, and coffee pod machine), that its impact on the production of a yield of brewed coffee would be very small. And thus, the impact of manufacturing the coffee brewing equipment is not included in this work. The plastic cup used in the production of pod coffee is a proprietary mix of different plastics; however, in this work, it is modeled as polypropylene. A major single-serve coffee pod manufacturer has committed to producing a cup utilizing a single type of plastic by 2020, with the goal of it being made of polypropylene (Green Mountain Coffee 2015). Transportation The transportation associated with the coffee is estimated using the work of Humbert and colleagues (2009). In their work (as in this work), the transportation is allocated in two groups. The first is from the grower to the coffee processing plant, and the second is from the plant to the point of purchase for the consumer. Moving the coffee from the grower to the processing plant is modeled to require a 410-kilometer (km) journey by a large truck (capacity greater then 32 tonnes [t]), then a 15,000- km journey by ocean freight vessel, and, finally, a 133-km trip by a large truck. Essentially, this accounts for transporting the coffee from the farm to the port, where it will be loaded onto the ship, across the waterway, and then to the plant by truck. The transportation is allocated to the modes using the mass-distance unit of tonnes-kilometers. The trip from the processing plant to the conumser is modeled to require 420 km by large truck and then 300 km by smaller truck (capacity between 16 and 32 t). This is presented in more detail in the supporting information. Use The amount of water used to brew the coffee was determined experimentally for each method, along with the amount of brewed coffee yielded (table 2). The amount of energy consumed during brewing for the drip coffee and pod coffee was found using a device-specific energy meter. The amount of energy needed for the french press coffee was determined by using the average temperature for tap water in Chicago of 12.17 C (Pacific Northwest Laboratory 2001) and heating it to 93.33 C (How to brew coffee 2015), a typical value for the water used in french press coffee. The impact of water consumption is modeled using water from a treated residential drinking source and the impact of energy from an electricity mix relevant to the United States. This calculation is presented further in the supporting information. End of Life At the end of its life, once the coffee has been brewed, the coffee grounds, filter, cup, foil, and residual water are disposed of in a municipal landfill setting. The coffee grounds are modeled as biologically degradable material, and each component is modeled based on its material composition. The plastic cup and aluminum foil are modeled as inert material, given that they will not degrade in a short time scale. The assumption was made to model the waste as being disposed of in a landfill setting. Currently, alternative disposal methods of coffee pods are not common, although the potential does exist to dismantle the pod into its primary components and dispose of them in a municipal recycling setting. With respect to the coffee grounds, there is also the option to dispose of them in a home compost pile or to apply them to a garden; however, those options are not considered in this study. Impact Assessment The environmental impacts of the three coffee systems were found using the nine impact categories available in TRACI: ozone depletion, global warming potential, smog, acidification, eutrophication, carcinogenics, noncarcinogenics, respiratory effects, and eco-toxicity (EarthShift 2015). The TRACI impact categories were selected because of their applicability to the United States, which is the geographical focus area of this study. Interpretation The results of the midpoint LCA are discussed both individually and comparatively. The results of each brewing method are first evaluated to determine which portion of the process life Hicks, Environmental Implications of Consumer Convenience 83

Figure 2 Environmental impact of coffee production per 1 kg of green coffee (OZ = ozone depletion; GW = global warming; SM = smog; AC = acidification; EU = eutrophication; CN = carcinogenics; NC = noncarcinogenics; RE = respiratory effects;ec = eco-toxicity). kg = kilogram. cycle has the greatest environmental impact, along with the potential to reduce that impact. Then, the three brewing methods are compared in each of the nine impact categories utilized. This is done to address the overall goal of the LCA, which is to understand the environmental impact of the pod-style coffee brewing system (a product of consumer convenience) when compared to its conventional coffee brewing counterparts. Results The results of the growing and processing of coffee are presented in figure 2, based on the work of Coltro and colleagues (2006). Above, the results bar on each figure in each impact category is the total environmental impact score for that impact category. The impact units are as follows: ozone depletion (kilograms chlorofluorocarbon equivalent, kg CFC-11-eq), global warming (kilograms carbon dioxide equivalent, kg CO 2 - eq), smog (kilograms ozone equivalent, kg O 3 -eq), acidification (molecular hydrogen cation equivalent, mol H + -eq), eutrophication (kilograms nitrogen equivalent, kg N-eq), carcinogenics (Comparative Toxic Unit for human; CTUh), noncarcinogenics (CTUh), respiratory effects (kilograms particulate matter of 10 microns or less equivalent, kg PM 10 -eq), and ecotoxicity (Comparative Toxic Unit for ecosystem, CTUe). The impact of growing and processing the coffee varies as a function of the impact category considered. In most categories, fertilizer usage dominates, suggesting that the ability to produce similar yields while using less fertilizer would significantly reduce the environmental impact of the production of coffee. In the case of the ozone depletion category, the use of diesel fuel contributes most significantly to the environmental impact, whereas for the eco-toxicity category, it is the use of pesticides. Most coffee in Brazil is washed and processed utilizing natural water sources, and this is characterized as such in the analysis. A significant quantity of water is used in the production of green coffee, at a mass ratio of 11:1 for both washing the coffee and wet method processing (Coltro et al. 2006). Coffee Brewing The results of the overall life cycle impact for the three coffee brewing options modeled are presented in figure 3. The results are presented in their native yield units, as described in table 2. In the case of drip filter coffee (figure 3a), the greatest environmental impacts are attributed to the production of green coffee and the electricity consumption in the brewing method. In the case of smog and ozone depletion categories, transportation also contributes significantly to the environmental impact. This is, again, true for the french press coffee (figure 3a) and the single-pod system, although in the smog category, the greatest impact occurs because of transportation. Although the pod in the pod-style coffee system (figure 3c) is more materials intensive to produce, it had the least environmental impact in six of the impact categories considered (OZ, EU, NC, RE, EC, CN). The french press has the least environmental impact in two categories (GW, SM), and the impacts of the two systems were approximately the same in one category (AC). One method that has been suggested to reduce the environmental impact of a coffee pod system would be changing the material used in the cup to potentially be a biodegradable material, which has been suggested as a future potential innovation (Hamblin 2015). However, the plastic cup does not represent a significant source of environmental impact during the life cycle of the coffee pod. Additionally, transitioning from a material that is relatively stable in a landfill setting, such as plastic, to a biodegradable material has environmental impact implications from a global warming potential standpoint. In general bioplastics are considered to be more environmentally friendly than their nonrenewable counterparts; however, environmental impact trade-offs occur as a function of increased carbon production for bioplastics (Tabone et al. 2010; Gironi and Piemonte 2011; Weiss et al 2007; Renouf et al. 2013; Madival et al. 2009; Kendall 2012; Copeland et al. 2013). Another basis of the perceived environmental friendliness is the ability for many bioplastics to be disposed of in an industrial composter; however, access to such a composter is not guaranteed (Hottle 84 Journal of Industrial Ecology

a) b) c) Figure 3 Midpoint LCA results (a) french press (0.70 L), (b) drip coffee (0.83 L), and (c) single-pod coffee (0.275 L) in native units (OZ = ozone depletion; GW = global warming; SM = smog; AC = acidification; EU = eutrophication; CN = carcinogenics; NC = noncarcinogenics; RE = respiratory effects; EC = eco-toxicity). L = liter; LCA = life cycle assessment. et al. 2014). As such, the bioplastics are often disposed of in a landfill setting, which will shift the life cycle environmental impact of the product (Madival et al. 2009; Hakkinen and Vares 2010; Potting and van der Harst 2015). Although presenting the impact of utilizing the three methods to brew coffee in their native units is useful, the real question is whether the use of a product of convenience, such as the single-pod coffee system, has a greater environmental impact than the other methods using a consistent functional unit. In figure 4a and 4b, the environmental impact of brewing a single serving of coffee (0.275 L) is shown for two representative impact categories across the three technologies: global warming Hicks, Environmental Implications of Consumer Convenience 85

a) b) Figure 4 Midpoint LCA results in common units by impact category: (a) global warming and (b) eutrophication; functional unit of 0.275 L. L = liter; LCA = life cycle assessment. potential and eutrophication potential. These two categories illustrate the two main results found in the analysis. The results of the analysis with respect to the remaining seven impact categories is available in the Supporting Information on the Web. The results of the LCA are unexpected. It is fairly simple to assume that the product of convenience would have a greater environmental impact because of its single-serve nature and the requirement of additional raw materials. However, in this instance, it has the least environmental impact across six impact categories considered, whereas the drip filter coffee has the greatest. The french press has the least environmental impact across two of the categories, suggesting that having the least environmental impact is not solely dependent on whether the coffee brewing system is a product of convenience. The two most significant contributors to the environmental impact of the coffee brewing options are the amount of ground coffee and electricity utilized per serving of coffee drink produced. An important consideration in this analysis is that different amounts of ground coffee are used in each preparation method to yield the same volume of drinkable coffee, and different amounts of electricity are utilized to heat the water. This analysis did not include so-called vampire power, which is the amount of electricity electronics use when they are plugged in, but not actively in use. If this were included, it would increase the amount of energy consumed by the drip filter coffee and pod coffee processes. Additionally, coffee wastage is not taken into account in this analysis, where although a whole pot or press of coffee may be brewed, not all of it is consumed. If it were to be included, the environmental impact per serving of conventionally brewed coffee consumed would be greater. Interestingly, the additional raw materials utilized by the pod-style brewing system did not significantly increase that system s environmental impact beyond that of the drip filter system. The greatest environmental impact contribution of the additional materials was due to the plastic cup in the pod style system. Nor did their disposal in the landfill at the end of their life. The filter and spent coffee grounds were modeled as biodegradable waste, whereas the foil and cup portions of the coffee pod were modeled as an inert material in the landfill. 86 Journal of Industrial Ecology

Figure 5 Sensitivity analysis of the pod-style coffee as a function of the amount each input would need to increase, to increase the overall category environmental impact by 25% (Disp. = disposal; OZ = ozone depletion; GW = global warming; SM = smog; AC = acidification; EU = eutrophication; CN = carcinogenics; NC = noncarcinogenics; RE = respiratory effects; EC = eco-toxicity). A higher percentage indicates that, for the given environmental impact, the results are less sensitive to estimates used in the LCA for inputs. Potential shifts in the materials utilized in coffee pods would have the ability to shift the environmental impact of the coffee production system. Shift to biodegradable cups most likely would increase the global warming potential impact of the podstyle coffee system, given that they would emit methane in the landfill disposal setting, whereas the current plastic cups do not on a reasonable time scale. Sensitivity Analyses A sensitivity analysis was performed on the suite of coffee brewing systems modeled. The analysis takes the form of a oneway sensitivity analysis, as described by Bjorklund (2002), where the amount that an individual input parameter needs to change in order for the overall impact in each category to change by a certain percentage is quantified. In this analysis, an overall impact category increase of 25% was selected as the target, and the results are presented relative to that. Figure 5 presents the percentage increase necessitated of the amount of each input needed to increase the overall impact in that category by 25%. The sensitivity analysis is presented for the pod-style coffee brewing system, whereas the other two systems are presented in the supporting information. For the single-serve coffee pod, electricity consumption (ozone depletion [OZ], global warming [GW], and smog [SM]) and coffee usage (acidification [AC], eutrophication [EU], carcinogenics [CN], noncarcinogenics [NC], respiratory effects [RE], and eco-toxicity [EC]) were found to be the two inputs which had the greatest impact on the environmental impact. This is interesting as the concern often voiced in public discussion of environmental impacts of single-serve coffee pods has related to their material intensity. The environmental impact was least sensitive to the quantity of foil (OZ, EU), the mass of the plastic in the cup (CN, NC, and EC), and the quantity of water utilized for brewing (GW, SM, AC, and RE). Similar results were found for the two conventional brewing methods, where water consumption was found to be the input generating the least change in impacts and coffee and electricity consumption were the inputs causing the greatest change in impacts. When the raw materials were increased in the sensitivity analysis, it was also assumed that the disposal scaled accordingly, such that an increase in the mass of the plastic cup for the singleserve coffee pod would also include a corresponding increase in the mass of plastic disposed of at the end of life. As noted previously, the three coffee brewing systems considered utilize different amounts of ground coffee per unit of liquid coffee drink produced. Presented as grams of ground coffee per liter of coffee drink produced, they vary as follows: drip filter (74 g/l), french press (45 g/l), and pod style (43 g/l). This is relevant to consider as the quantity of ground coffee utilized, and its upstream environmental impacts, affect the overall life cycle impacts and comparison of the three brewing processes. Results presented above were calculated using native serving sizes and using equal volumes. Figure 6, in contrast, presents the environmental impact per serving (.275 L) when the strength of the brewed coffee is the same for all 3 systems. The strength is measured as coffee density, set at 43 g/l (the same as the pod style coffee) and shown for the global warming and Hicks, Environmental Implications of Consumer Convenience 87

a) b) Figure 6 Global warming and eutrophication impacts per serving (.275 L) when the strength of the brewed coffee is the same for all 3 systems (43 g/l). Figure 7 Forecasted U.S. environmental impact of coffee makers annually. eutrophication potential impact categories. The analysis for the remaining impact categories is included in the supporting information. In completing the analysis, the quantity of water and electricity were held steady, while the quantity of ground coffee and its associated disposal were varied. A product of convenience, such as the coffee pod, has very little potential for environmental impact if it is not widely adopted. In 2010, drip filter coffeemakers accounted for 75% of retail coffeemaker sales, whereas pod-style machines accounted for only 19% (french presses were not considered in the analysis by ENERGY STAR) (ENERGY STAR 2011). Figure 3 presents projections of the potential environmental impact attributed to the use of both drip filter and coffee pod brewing systems for the United States, assuming that 90% of households own and use a coffeemaker to produce one serving of coffee per day (0.275 L) (Lighthouse Marketing Research 2013; United States Census Bureau 2015). Twenty-four percent of the households with coffeemakers have the coffee pod variety, whereas the rest are assumed to have drip filter (Lighthouse Marketing Research 2013). The typical U.S. household replaces their coffeemaker every 4 to 5 years, with 5 years assumed in this forecast (International Housewares Association 2004). Four different scenarios are projected in figure 7. The first, current, utilizes population growth and the current ownership rate of 24% of those owning coffeemakers to have pod style and the remainder drip filter. The second, replacement, is when 20% of the population replaces their coffeemakers annually with pod style. The pod and drip forecasts refer to the environmental impact that would occur if each technology were used exclusively to meet the population s coffee brewing. This suggests that, in the case of greenhouse gas emissions, there is the potential for a decrease in environmental impact as more consumers adopt the pod-style brewing system. Given that the pod-style coffee system produces around half the amount of global warming potential of the drip filter system. It should be noted that this chart is most likely an underestimate of the annual environmental impact of coffee consumption in the United States, because it assumes that each household only consumes two servings (0.275L each) of coffee per day, when in reality consumption patterns are far more diverse than that (Lighthouse Marketing Research 2013). 88 Journal of Industrial Ecology

Discussion Coffee and its consumption are ubiquitous in the United States. The introduction of a product of convenience to home coffee brewing and consumption has the potential to change the environmental impact of the activity. Of the nine impact categories modeled in this work, the drip filter coffee had the greatest environmental impact of all of them, with the pod-style coffee having the least impact in six. When the steps of making coffee are considered, the greatest environmental impact is attributed to the production of the ground coffee itself and the energy needed to brew the coffee. This brings to light the fact that differing amounts of ground coffee are utilized to produce the same amount of coffee beverage in the different brewing processes. In two of the processes (drip filter and french press), a set amount of beverage must be made, and there is the potential for coffee wastage, potentially shifting the environmental impact of the brewing method. Opportunities do exist to potentially reduce the environmental impact of the pod-style coffee system through disassembling the product into its components and recycling them, although currently the pods are not easy to dissemble. Challenges also exist as to how to recycle the plastic cup, with it being a proprietary mixture of different plastics. Some newer coffee pods have been produced with a single type of plastic (such as explored in this work) to allow for recycling of the cup after use; however, they still require disassembly (Green Mountain Coffee 2015). There is the potential for the development of biodegradable coffee pods; however, there is the potential to shift the environmental impact through the use of alternate raw materials (as discussed previously). Additionally, there is the challenge of creating a product that will stand up to heat and water (albeit for a very brief time) without degrading significantly, but will then degrade in a municipal compost setting. Energy consumption may be significant in the production of the coffee pod. In this analysis, the energy used to assemble the final pod plus the minimal amount of adhesive used is not included. If the amount of energy, which is not currently available in literature or on the websites of the coffee pod manufacturers, is significant, it may potentially shift the relative environmental impact of the coffee pod system and would then cause the french press brewing system to have the least environmental impact. This suggests that, based on the midpoint LCA presented, coffee pods have a lower environmental impact than the conventional drip filter method and sometimes a lower impact than the french press (as a function of the impact category considered). That may not hold true if an excessive amount of energy is required during the manufacturing phase or if an excessive amount of waste material is produced. Conclusions Products of convenience are attractive to consumers through offering the opportunity to save time and effort. However, these products often come at environmental cost attributed to issues such as single-serve packaging. In the case of coffee pods, this midpoint analysis utilizing the TRACI impact categories found single-serve coffee pods to have a significantly lower environmental impact when compared with the conventional drip filter method. This has the potential to shift, however, if more energy and materials are utilized in the production of coffee pods than expected. The coffee pod also requires more materials that are not biodegradable when compared with the two other options. One issue associated with these pods is that they are a monstrous hybrid, in that they are made of multiple materials, which are not easily separable, making them nearly impossible to recycle in their current form. The only current viable option for their disposal is in a landfill or incinerator setting, whereas the two conventional coffee preparation methods present the option of composting their coffee grounds and paper filter. If the coffee pods did utilize a single plastic, such sa polypropylene, then the option of recycling that plastic into a new feedstock would exist, although this would not fix the challenge of dissembling the pods, which is both messy and time-consuming further increasing the likelihood of consumers choosing not to disassemble them into their potentially recyclable components. The aluminum foil topping the pod is currently recyclable, although the issue of ease of disassembly persists. Biodegradable coffee pods are also a potential method for keeping coffee pods out of the landfill, although this would likely increase the global warming potential of the coffee pods, and in the case of industrial composting, many consumers simply do not currently have access (Hottle et al. 2014). At the least, this study suggests that the environmental impact of a product of convenience in this case is not greater than its conventional product counterpart, and may actually be less. It also points to the need for further work as to the amount of energy used in the manufacturing of singleserve coffee pods. A major question generated by this work is whether this finding will hold true for other products of convenience that are in various stages of development and consumer adoption. Acknowledgments The author acknowledges the use of startup funds from the University of Wisconsin Madison Department of Civil and Environmental Engineering to fund this work. Although some brands are identified by named in this work in order to provide historical context, they are in no way endorsed by the author. The author also thanks the anonymous reviewers whose comments significantly improved the quality of the article. References Bjorklund, A. E. 2002. Survey approaches to improve reliability in LCA. The International Journal of Life Cycle Assessment 7(2): 64 72. Brunner, T. A., K. van der Horst, and M. Siegrist. 2010. Convenience food products: Drivers for consumption. Appetite 55(3): 498 506. Hicks, Environmental Implications of Consumer Convenience 89

Buckley, M., C. Cowan, and M. McCarthy. 2007. The convenience food market in Great Britain: Convenience food lifestyle (CFL) segments. Appetite 49: 600 617. Bureau of Labor Statistics. 2009. Consumer price index. http://data. bls.gov/cgi-bin/cpicalc.pl?cost1=1&year1=1959&year2=2015. Accessed 26 August 2015. Busser, S. and N. Jungbluth. 2009. The role of flexible packaging in the life cycle of coffee and butter. The International Journal of Life Cycle Assessment 14(Suppl 1): S80 S91. Coltro, L., A. Lucia, P. A. Oliveira, J. P. Baddini, and R. M. Kletcke. 2006. Environmental profile of Brazilian green coffee. The International Journal of Life Cycle Assessment 11(1): 16 21. Copeland, A. M., A. A. Ormsby, and A. M. Willingham. 2013. Assessment and comparative analysis of a reusable versus disposable to-go system. Sustainability 6(6): 353 358. Cordella, M., I. Bauer, A. Lehmann, M. Schulz, and O. Wolf. 2015. Evolution of disposable baby diapers in Europe: Life cycle assessment of environmental impacts and identification of key areas of improvement. Journal of Cleaner Production 95: 322 331. De Monte, M., E. Padoano, and D. Pozzetto. 2005. Alternative coffee packaging: An analysis from a life cycle point of view. Journal of Food Engineering 66(4): 405 411. EarthShift. 2015. TRACI 2 impact assessment method. www.earth shift.com/software/simapro/traci2. Accessed 26 August 2015. ENERGY STAR. 2011. Market & industry scoping report: Coffee makers. Washington, DC: U.S. Environmental Protection Agency. Gironi, F. and V. Piemonte. 2011. Bioplastics and petroleum-based plastics: Strengths and weaknesses. Energy Sources, Part A: Recovery, Utilization, and Environmental Effects 33(21): 1949 1959. Green Mountain Coffee. 2015. Recyclable Keurig pods. www.keurig. com/content/recyclable. Accessed 25 August 2015. Franklin Associates Ltd. 1990. Energy and environmental profile analysis of children s disposable and cloth diapers. Report to the American Paper Institute Diaper Manufacturers Group. Prairie Village, KS, USA: Franklin Associates Ltd. Hakkinen, T. and S. Vares. 2010. Environmental impacts of disposable cups with special focus on the effect of material choices and end of life. Journal of Cleaner Production 18: 1458 1463. Hamblin, J. 2015. A brewing problem: What s the healthiest way to keep everyone caffeinated? The Atlantic,2March. Hassard, H. A., M. M. Couch, T. Techa-Erawan, and B. C. McLellan. 2014. Product carbon footprint and energy analysis of alternative coffee products in Japan. Journal of Cleaner Production 73: 310 321. Hottle, T. A., M. M. Bilec, D. L. Meeks, and A. E. Landis. 2014. Biopolymer wastes: Assessing the shift to compost-based waste management. American Center for Life Cycle Assessment, Proceedings of LCA XIV, 6 9 October, San Francisco, CA, USA. How to brew coffee. 2015. Hot to brew coffee.com. www.howtob rewcoffee.com/french.html. Accessed 21 July 2015. Humbert, S., Y. Loerincik, V. Rossi, M. Margni, and O. Jolliet. 2009. Life cycle assessment of spray dried soluble coffee and comparison with alternatives (drip filter and capsule espresso). Journal of Cleaner Production 17:1351 1358. Hunt, R. G. and W. E. Franklin. 1996. LCA How it came about Personal reflections on the origin and development of LCA in the USA. The International Journal of Life Cycle Assessment 1(1): 4 7. IGD Business Publications. 2002 Consumer attitudes to meal solutions. Watford, UK: IGD. International Housewares Association. 2004. The Home Authority looks at the U.S. consumer. Market Watch, the Home Authority 3(4): 1. ISO (International Organization for Standardization). 2006. Environmental management Life cycle assessment Principles and framework. Geneva, Switzerland: ISO. Kendall, A. 2012. A life cycle assessment of biopolymer production from material recovery facility residuals. Resources, Conservation and Recycling 61: 69 74. Lehrburger, C., J. Mullen, and C. Jones. 1991. Diapers: Environmental impacts and lifecycle analysis. Philadelphia, PA, USA: The National Association of Diaper Service. LeVan, S. L. 1995. Life cycle assessment: Measuring environmental impact. In 49th Annual Meeting of the Forest Products Society, 25 28 June, Portland, OR, USA. Lighthouse Marketing Research. 2013. The coffee pod market. www. gorbycoffee.com/wp-content/uploads/2015/02/sierra-report- Updated-3-27-13.pdf. Accessed 25 August 2015. Madival, S., R. Auras, S. P. Singh, and R. Narayan. 2009. Assessment of the environmental profile of PLA, PET and PS clamshell containers using LCA methodology. Journal of Cleaner Production 17: 1183 1194. National Coffee Association. 2015. All about coffee. www.ncausa. org/i4a/pages/index.cfm?pageid=68. Accessed 22 July 2015. Neuhas, J. 1999. The way to a man s heart: Gender roles, domestic ideology, and cookbooks in the 1950 s. Journal of Social History 32(3): 529 555. Pacific Northwest Laboratory. 2001. December 1. Appendix D, Cold water inlet temperatures for selected U.S. locations. www.pnl. gov/fta/3_appd.htm. Accessed 20 July 2015. Potting, J. and E. van der Harst. 2015. Facility arrangements and the environmental performance of disposable and reusable cups. The International Journal of Life Cycle Assessment 20(8): 1143 1154. PRé Consultants. 2015. SimaPro. www.pre-sustainability.com/simapro. Accessed 26 August 2015. Renouf, M., R. J. Pagan, and M. K. Wegener. 2013. Bio-production from Australian sugarcane: An environmental investigation of product diversification in an agro-industry. Journal of Cleaner Production 39: 87 96. Salomone, R. 2003. Life cycle assessment applied to coffee production: Investigating environmental impacts to aid decision making for improvements at company level. Food, Agriculture & Environment 1(2): 295 300. Schwartzberg, H. 2013. Batch coffee roasting; Roasting energy use; reducing that use. In Advances in Food Process Engineering Research and Applications, edited by S. Yanniotis et al. 173 195. New York: Springer Science+Business Media. Szmigin, M. 2006. Mothers of invention : Maternal empowerment and convenience consumption. European Journal of Marketing 40(9 10): 1122 1142. Tabone, M. D., J. J. Cregg, E. J. Beckman, and A. E. Landis. 2010. Sustainability metrics: Life cycle assessment and green design in polymers. Environmental Science & Technology 49(2): 742 749. United States Census Bureau. 2015. State and county quick facts. http://quickfacts.census.gov/qfd/states/00000.html. Accessed 26 August 2015. Weiss, M., M. Patel, H. Heilmeier, and S. Bringezu. 2007. Apply distance-to-target weighing methodology to evaluate the environmental performance of bio-based energy, fuels, and 90 Journal of Industrial Ecology