You are on page 1of 10

Waste Management xxx (2018) xxx–xxx

Contents lists available at ScienceDirect

Waste Management
journal homepage:

Life cycle assessment of bottled water: A case study of Green2O products

Naomi Horowitz, Jessica Frago, Dongyan Mu ⇑
School of Environmental and Sustainability Science, Kean University, 1000 Morris Ave., Union, NJ 07083, United States

a r t i c l e i n f o a b s t r a c t

Article history: This study conducted a full life cycle analysis of bottled water on four types of bottles: ENSO, PLA (corn
Received 24 August 2017 based), recycled PET, and regular (petroleum based) PET, to discern which bottle material is more bene-
Revised 19 January 2018 ficial to use in terms of environmental impacts. PET bottles are the conventional bottles used that are not
Accepted 19 February 2018
biodegradable and accumulate in landfills. PLA corn based bottles are derived from an organic substance
Available online xxxx
and are degradable under certain environmental conditions. Recycled PET bottles are purified PET bottles
that were disposed of and are used in a closed loop system. An ENSO bottle contains a special additive
which is designed to help the plastic bottle degrade after disposed of in a landfill. The results showed that
of all fourteen impact categories examined, the recycled PET and ENSO bottles were generally better than
the PLA and regular PET bottles; however, the ENSO had the highest impacts in the categories of global
warming and respiratory organics, and the recycled PET had the highest impact in the eutrophication cat-
egory. The life cycle stages that were found to have the highest environmental impacts were the bottle
manufacturing stage and the bottled water distribution to storage stage. Analysis of the mixed bottle
material based on recycled PET resin and regular PET resin was discussed as well, in which key impact
categories were identified. The PLA bottle contained extremely low impacts in the carcinogens, respira-
tory organics and global warming categories, yet it still contained the highest impacts in seven of the
fourteen categories. Overall, the results demonstrate that the usage of more sustainable bottles, such
as biodegradable ENSO bottles and recycled PET bottles, appears to be a viable option for decreasing
impacts of the bottled water industry on the environment.
Published by Elsevier Ltd.

1. Introduction 1997). The intermediates of PET are terephthalic acid (TPA) and
ethylene glycol (EG) which are both acquired using oil feedstock
Bottled water is a fast growing industry, with consumption (Sinha, et al., 2010). Pure PET is a shapeless, glass-like material that
reaching a record high in 2015 with 11.7 billion gallons (Rodwan, crystalizes when certain modifying agents are added or when heat
2016). In 2016 bottled water consumption in Mexico was the high- is applied above 72 °C (Sinha, et al., 2010). Typical PET bottle are a
est worldwide with 67.2 gallons per capita, followed by Thailand major threat to the environment due to the high amount of chem-
with 56.9 gallons per capita, Italy with 47.5 gallons per capita, icals, namely petroleum, required in production, as well as incor-
and the United States with 39.3 gallons per capita (Statista, rect usage and disposal (Revathi, et al., 2017). Approximately 4%
2017). In 2012, worldwide consumption of bottled water totaled of the petroleum used annually in the world in 2016 was for the
288 billion liters, while the projected consumption for 2017 totaled production of plastic (British Plastic Federation 2016). Bottled
391 billion liters (Statista, 2017). As the industry booms, however, water also results in a large amount of waste. According to the
it raises increasing concerns over resource use, human health, and study by the Center for Sustainable Systems, University of
on the negative impacts to ecological systems. Michigan (2015), approximately 7.2–14.1 million tons of plastic
One major concern is the predominant application of plastic waste disposed of in landfills each year accounts for 22% to 43%
bottles made from a petroleum product such as polyethylene of waste disposed in landfills (Gourmelon, 2015). The majority of
terephthalate (PET) (Revathi, et al., 2017). PET is a long-chain poly- plastics are not biodegradeable, and therefore the bulk of the poly-
mer part of the polyester family (Sinha, et al., 2010; Muschiolik, mers manufactured will persist for decades, centuries, and quite
possibly millennia (Hopewell, et al., 2009).
The environmental concerns regarding plastic waste are creat-
Abbreviations: ENSO, Environmental Solution; GHG, greenhouse gas; GWP,
global warming potential; PET, polyethylene terephthalate; PLA, polylactic acid. ing incentives to develop alternatives for petroleum based bottle
⇑ Corresponding author. manufacturing to reduce plastic solid waste disposal (Zia et al.,
E-mail address: (D. Mu). 2007). Currently, scientists have developed many alternatives to
0956-053X/Published by Elsevier Ltd.

Please cite this article in press as: Horowitz, N., et al. Life cycle assessment of bottled water: A case study of Green2O products. Waste Management (2018),
2 N. Horowitz et al. / Waste Management xxx (2018) xxx–xxx

divert plastic bottles from landfills. Intensive research has been put The company proposed three product lines with different bottle
into the field of recycling. Recycling allows for a closed loop system vendors to produce bottled water. Therefore, this study developed
to repurpose the bottle after disposal. Another alternative is the a model that integrated the bottle manufacturing, water fillers, and
usage of biodegradable plastic bottles, such as corn-based polylac- product distribution in a Green2O scenario. The study also aimed
tic acid (PLA) bottles. PLA bottle grade is essentially carbon neutral to identify the key production stages that raise environmental
since it is derived from carbon sequestering plants, and as such is impacts in the bottled water production.
also biodegradable when under appropriate environmental condi- The outcomes of this study are expected to increase the public’s
tions (West, 2016). understanding of environmental impacts of plastic bottles and
Environmental concerns, however, are raised with different enlighten the water bottle industries of the impacts of various
options for diverting plastic bottles from landfills. For example, plastic bottle alternatives. The results can also help the bottle man-
recycling uses energy to sort and process the plastics, which ufacturing industries improve their products to lower environmen-
increases the resource consumption and cost. Many water bottle tal impacts in certain production stages.
industries hesitate to recycle plastics resin because the cost may
be even higher than the new plastic resin (Intagliata, 2012). Simi-
2. Methodology
larly, the PLA resin is derived from renewable resources such as,
biomass of sugar cane or corn starch (Madival, et al., 2009). Grow-
The framework of this study followed the guidelines according
ing raw materials for PLA consumes energy and resources, which
to ISO 14,000 standards for LCA. Data applied in LCA modeling and
increases life cycle impacts of PLA bottles significantly. Another
analysis were either collected from the Green2O product chains or
misconception seems to be that PLA will simply degrade once in
retrieved from existing LCA databases, such as EcoInvent 3.1. A sce-
a landfill, however, the PLA plastic is only compostable under cer-
nario analysis was also conducted in order to determine if combin-
tain environmental conditions – mainly when digested by
ing various percentages of recycled PET resin and regular PET resin
microbes with temperatures reaching 140 ⁰F for ten consecutive
would have a lower impact on the environment.
days (Royte, 2006). In addition, increasing PLA production seems
to be a question of morality when so many people in the world
are starving and malnourished (Royte, 2006). 2.1. Goal and scope
The ENSO bottle is a relatively new alternative created to
increase biodegradability of plastic bottles in landfills. ENSO bot- The goal of this project was to analyze the environmental
tles are regular PET bottles which contain an additive that makes impacts of Green2O bottles by assessing 4 types of plastic materi-
the bottles more enticing to the billions of microorganisms which als including ENSO, PLA, 100% recycled PET, and regular PET. The
normally degrade plastic bottles. The microorganisms break down 100% recycled PET bottle was assumed entirely made from recy-
the bottles into biogases and inert humus leaving no toxic materi- cled plastic resin. The PLA bottle was assumed to be corn based
als behind (ENSO Bottles, 2009). After the ENSO additive is mixed and compostable. The ENSO bottle consists of regular PET resin,
into the plastic bottle components, the final product looks, feels, albeit contains a special additive (1% by weight), which allows it
and performs exactly as a normal bottle would, with the exception to degrade in a landfill extremely quickly. The regular PET was
of being biodegradable. Research has found that ENSO plastics assumed to be petroleum based plastic material. The functional
biodegrade by about 25 percent in only 160 days in ideal environ- unit denoted for this project was 12 bottles, as this amount is typ-
mental conditions (Huff, 2013). In addition, ENSO plastics can be ically found in one pack of Green2O water bottles.
recycled just as any other typical plastic would, however, since a The system boundary for the bottles being studied is illustrated
majority of the bottle is composed of petroleum based plastic resin, in the flow diagram in Fig. 1. The boundary took into account the
carbon dioxide (CO2) and methane (CH4) are released once the bot- complete life-cycle, beginning from the extraction of raw materials
tle decomposes. through the disposal of the product in a landfill. This boundary
In order to evaluate whether PLA and recycled PET bottles are included plastic material manufacturing, bottle manufacturing, fill-
beneficial for the environment, past research was conducted on ing the bottles with water, assembly of the product, distribution of
the Life Cycle Assessment (LCA) of ordinary PET bottles in conjunc- the product to storage and the market, consumption, and disposal/
tion with recycled PET and corn based PLA bottles. A study done by recycling. Generally, the bottle preforms were assumed to be made
Li Shen et al. (2010), analyzed open loop recycling of PET bottles of different raw materials at different locations, and then trans-
with four PET recycling cases: mechanical recycling, semi- ported to Tennessee to be filled with water. The bottle manufactur-
mechanical recycling, back-to-oligomer recycling, and back-to- ing is described in Fig. 1. The products, bottled water were then
monomer recycling. The results were also compared to polylactic distributed across the country. This study assumed identical distri-
acid (PLA) bottles. The results concluded that recycled PET fibers bution and use stages for the four types of water bottles. The final
have lower environmental impacts than virgin fiber production, disposals of four types of bottles were different and described in
specifically, in the categories of abiotic depletion, acidification Fig. 1 as well.
and human toxicity. The recycled fibers were found to have a com-
paratively high environmental impact on freshwater aquatic eco- 2.2. Life cycle Inventory (LCI)
toxicity than the virgin PET, as well as a lower Global Warming
Potential (GWP), than PLA bottles. While Li Shen et al. conducted The goal of Life Cycle Inventory (LCI) was established to identify
a LCA study on recycled PET and PLA bottles, no LCA study has been and quantify the environmental factors crossing the system
conducted on ENSO bottles in conjunction with regular PET, recy- boundaries. LCI data, as shown in Table 1, consisted of raw material
cled PET, and PLA bottles. and energy inputs utilized to create each product. The LCI included
This study analyzed three supposed environmentally friendly all the unit processes and quantity of inputs shown in Fig. 1. The
bottle alternatives: recycled PET bottles, PLA bottles and ENSO bot- impacts of unit processes such as plastic resin production, blow
tles. As a baseline, the regular petroleum based PET bottles were molding, and water injection were attained from Ecoinvent 3.0
analyzed for comparison. The study coordinated with the bottled database on Simapro. The material use and inputs were based on
water manufacturing company Green2O, a New Jersey based water the actual weight of bottles and packaging. All the calculations
bottle company that aims to provide a premium all natural alkaline used to derive the quantitative data are described in the Supple-
spring water, packaged in an environmentally conscious container. mentary Information (SI).

Please cite this article in press as: Horowitz, N., et al. Life cycle assessment of bottled water: A case study of Green2O products. Waste Management (2018),
N. Horowitz et al. / Waste Management xxx (2018) xxx–xxx 3

Bole manufacturing Boled water producon and use Disposal

ENSO bole Water filling Distribuon and ENSO bole Disposal

PET Resin with ENSO
Hauling Truck 72 km
producon 0.27821 kg
Water Producon
Refrigeraon at
5.7493 kg Plasc waste in landfill 0.0556 kg
Market 0.03744
Blow molding 0.00466 kWh
Water Injecon
0.00978 kWh Distribuon to Cardboard waste in landfill 0.0599 g
ENSO Resin transportaon
AZ to OH Storage (from TN
to CA, TX, and NJ)
Product Assembly 1784 km Electricity generaon
ENSO Bole transportaon 0.12061 kWh
OH to TN
Product Packaging
Storage to Market
CH4, CO2
400 km
PLA bole

PLA Resin producon 0.3 kg PLA bole Disposal

Use Hauling Truck 72 km

Blow molding 0.00466 kWh
Other inputs Plasc waste in landfill 0.0556 kg
Consumer Vehicle
PLA Resin transportaon Transport 0.27504
MI to OH km Cardboard waste in landfill 0.0599
Label kg
PLA Bole transportaon
OH to TN Refrigeraon 0.10884
kWh Electricity generaon
Low-density 0.12061 kWh
Recycled PET bole (LPDE)
Recycled PET resin producon
producon 0.27821 kg Recycled PET bole disposal

Blow molding 0.00466 kWh Stretch blow

molding 0.0156 kg Packaging Hauling Truck 72 km

Recycled PET preform

Plasc waste in landfill 0.0556 kg
transportaon Transportaon of
OH to OH, MI Packaging
LPDE Label 1170
k Cardboard waste in landfill 0.0599
Recycled PET bole kg
transportaon Transportaon of
OH, MI to TN cardboard
Cap NJ to NT 1170 km
Recycled PET bole disposal
Typical PET bole
Polypropylene (PP)
PET resin producon 0.27821 producon 0.0216 Packaging Plasc
manufacturing Hauling Truck 72 km
0.0184 kg
Blow molding 0.00466 kWh Injecon molding Plasc waste in landfill 0.334 kg
0.0216 kg Extrusion plasc film
PET preform transportaon 0.059 kg
OH to OH
Transportaon of Cardboard waste in landfill
caps Transportaon of 0.0599 kg
PET bole transportaon 1170 km packing plascs km
OH to TN

Fig. 1. Flow diagram of system boundary.

The bottle manufacturing includes the plastic polymer/resin be transported to Tennessee, blown into bottles, and filled with
manufacturing, preform manufacturing, and blowing preforms spring water before the final product, i.e., Green2O bottled
to bottles. These three steps may occur at different locations. water, was packed and distributed. Similarly to the PLA bottle,
Based on the information provided by the Green2O Company, the ENSO bottle requires the additive which is produced in Ari-
they purchase PLA preform and ENSO preform from the same zona. The study assumed the plastic resin with ENSO additive
vendor located in Dayton, OH. However, the commercialized was made in Arizona. The LCI modeled the transportation for
PLA resin is only provided by NatureWorks, located in Min- the ENSO resin from Arizona to Ohio to make ENSO preforms,
nesota. When modeling the LCI, transportation for PLA resin and then to Tennessee where spring water is filled into ENSO
requires traveling from Minnesota to Ohio where bottle pre- bottles. All the distances inputted into the model are shown in
forms vendor is located. The preforms were then assumed to Table 2.

Please cite this article in press as: Horowitz, N., et al. Life cycle assessment of bottled water: A case study of Green2O products. Waste Management (2018),
4 N. Horowitz et al. / Waste Management xxx (2018) xxx–xxx

Table 1
Direct inputs and outputs of each type of bottle.

Unit Recycled PET PLA ENSO Regular PET

Direct inputs
PET kg 0.278 0 0.275 0.278
Packaging Film kg 0.0184 0.0184 0.0184 0.0184
PLA kg 0 0.3 0 0
ENSO Additive kg 0 0 0.00278 0
LPDE kg 0.0156 0.0156 0.0156 0.0156
PP kg 0.0216 0.0216 0.0216 0.0216
Cardboard kg 0.0599 0.0599 0.0599 0.0599
Water kg 0.479 0.479 0.479 0.4791
Transportation km 3156 3975 5719 3292
Direct discharge/emissions
Carbon Dioxide kg CO2 eq 0 0 0.786 0
Electricity kWh 0 0 1.206 0
Methane kg CH4 eq 0 0 0.292 0
Waste Stream kg 0.0556 0.356 0.334 0.334

Table 2
The distance traveled during each process.

PLA ENSO Recycled PET Regular PET

Polymer manufacturer Minnetonka, MI Mesa, AZ Bowling Green, OH Columbus, OH
Preform manufacturer 1148 km 2892 km Average 61.6 km 237 km
Dayton, OH Dayton, OH Holland, OH and Manchester, MI Holland, OH
Filler 571 km 571 km Average 839 799 km
Dandridge, TN Dandridge, TN km Dandridge, TN Dandridge, TN
Distribution centers 1784 km+ (California, Texas and New Jersey)
Distribution center to stores 400 km
Stores to consumers 0.27 km
Disposal 72 km
The average distance from Tennessee to California, Texas, and New Jersey was found to be 1784 km (as shown in the table below).
Distance for consumers was calculated on the basis that supplementary items besides the water bottles were bought together. The average distance driven to the store
was taken to be 27 km. The quantity of water bottles compared to the entire purchase was assumed to be 1%.

Green2O purchases regular PET and recycled PET preforms from plastics contain 62.5 wt% of carbon, which was assumed to be
a vendor located in Holland, OH. As the vendor did not provide decomposed entirely in a landfill and converted to either CO2 or
detailed information on their supply chain, this study made an CH4. The study then assumed half of total carbon emissions were
assumption that the regular PET and recycled PET plastic resin CH4, in which 90% would be captured to generate electricity and
were produced locally. As a result, the 100% recycled PET bottle 10% would escape into the air. The electricity generation was based
and regular PET bottle do not require a large amount of plastic on the quantity of captured CH4 and the biogas combustion effi-
resin transportation to the bottle preform manufacturers. In the ciency (35% per MJ CH4 combusted) provided in the GREET model.
LCI, the preforms were made in Ohio and then transported to Ten- The total emissions of ENSO bottle disposal in landfills included
nessee for water filling. All the bottle manufacturing and trans- CO2 from plastic decomposition, the CH4 released to air, and emis-
portation scenarios were based on the actual vendors and supply sions from biogas combustion. Those emissions were counted into
chains proposed by the Green2O Company, seen in Table 2. the final impacts because they were derived from petroleum-based
Disposal was also different regarding each bottle in this study. fossil fuels. The electricity generated was treated as a benefit in the
The 100% recycled PET bottle was assumed to be sorted out from LCI, in which the impacts of electricity generated were subtracted
other wastes at the landfills, shredded into flakes and pellets, and from the total impacts of the ENSO bottle.
sent to the manufacturers to mold the PET pellets and make bottle In regards to the packaging, this study assumed that all the
preforms. This study also assumed landfills were located close to packaging cardboard was disposed of in landfills and degraded.
the recycled PET resin manufacturers in Ohio, therefore, the trans- The bottle’s cap, label and plastic film/wrap disposed in landfills
portation of PET pellets to the resin makers were not incorporated were assumed to not be biodegradable. The caps were composed
into the model. of polypropylene (PP) and the labels and plastic film/wrap and
The regular PET bottles were assumed to be disposed of in land- were composed of low density polyethylene (LPDE).
fills, and to not degrade for millennia. Therefore, the regular PET
bottles did not release emissions, albeit they could raise other envi- 2.3. Life cycle impact analysis (LCIA)
ronmental impacts such as ecotoxicity and land occupation. PLA
bottles were assumed to be disposed of in landfills and compostable The Life Cycle Impact Analysis (LCIA) stage, aimed at
but not degradable under standard landfill conditions (Royte, comprehending and analyzing the data for significance in possible
2006). The emissions released from PLA bottles were not accounted environmental impacts in a system. Impact 2002+ (IMPact
for in the final impact assessment because the carbon in PLA was Assessment of Chemical Toxics) analysis method was applied in
extracted from the atmosphere in the corn growing stage. the LCIA using SimaPro 8. The categories analyzed included carcino-
The emission from the ENSO bottles degrading in landfills were gens (kg C2H3Cl eq), non-carcinogens (kg C2H3Cl eq), respiratory
assumed to be recaptured and converted into electricity. The calcu- inorganics (kg PM2.5 eq), ozone layer depletion (kg CFC-11 eq), res-
lation of emissions was based on information provided by the piratory organics (kg C2H4 eq), aquatic ecotoxicity (kg TEG soil), ter-
ENSO manufacturer and is further explained in the SI. The ENSO restrial ecotoxicity (kg TEG soil), terrestrial acidification/

Please cite this article in press as: Horowitz, N., et al. Life cycle assessment of bottled water: A case study of Green2O products. Waste Management (2018),
N. Horowitz et al. / Waste Management xxx (2018) xxx–xxx 5

nutrification (kg SO2 eq), land occupation (m2orgarable), aquatic 4–7 times greater than for the PLA and recycled PET bottles. The
acidification (kg SO2 eq), aquatic eutrophication (kg PO4 P-lim), glo- stage that contains the highest impact in carcinogens for the regu-
bal warming (kg CO2 eq), non-renewable energy (MJ primary) and lar PET bottle was the bottle manufacturing stage, with a value of
mineral extraction (MJ surplus). The impact of nonrenewable 91.1% of all the stages. The highest stage for the ENSO bottle was
energy counted the primary energy use, which included petroleum, also the bottle manufacturing stage, which constitutes 105.9% of
natural gas and coal. Climate change was characterized by utilizing the category. A value over 100% was attained due to the disposal
CO2 equivalency, with 100 years developed by the Intergovernmen- stage having a negative impact of 0.0553 C2H3Cl eq. As the PET
tal Panel on Climate Change (IPCC). Greenhouse gases, such as CO2, resin manufacturing usually releases high carcinogens, the regular
CH4 and N2O, were converted to CO2 equivalency to create a sum PET and ENSO bottles were high in this category because they both
total of global warming potential. Acidification was accounted for contain high petroleum based PET.
using SO2 equivalents, which included NOx and NH3. The eutrophi- The non-carcinogenics of the four bottles were close to each
cation included discharges of nutrients, mainly nitrogen and phos- other, ranging from the highest for the PLA bottle with a value of
phorus, into water bodies. The definition and calculation of other 0.178 C2H3Cl eq. to the lowest for the ENSO bottle of 0.164 C2H3Cl
impacts can be found in Impacts 2000+ as well. eq. Non-carcinogenics was highest in the stage of the distribution
to storage for all four bottles, with a second value in the bottle man-
ufacturing and a third of the customer use stage. The sum of the
3. Results and discussion three categories amounted to 89% of the category. As the electricity
generation has a high non-carcinogen release, the higher impact
3.1. Environmental impacts of bottled water with different bottle occurs in stages using more electricity than others. The product dis-
materials tribution and the customer use stages used electricity for refrigera-
tors and storage, which raised higher impacts on non-carcinogens.
LCA contains categories with units measuring each impact cate- In addition, the fossil use in transportation raised non-
gory. Each category may not be compared to each other (i.e. car- carcinogenics impact, particularly in the distribution to storage
cinogenics and eutrophication), albeit, each category within the stage.
stage/bottle may be compared to the identical category in a differ- Respiratory inorganics was highest for the regular PET bottle,
ent stage/bottle (i.e. carcinogenics in PLA and carcinogenics in with a value of 0.00446 kg PM2.5 eq., followed by the PLA bottle,
ENSO). The data may be analyzed both in terms of impacts within with a value of 0.00426 kg PM2.5 eq. Whereas, the respiratory
each stage for each bottle that contributes the greatest to the envi- organics were highest for the ENSO bottle with a value of 0.0035
ronmental impact, as well as the sum of the impacts for each stage kg C2H4 eq, albeit, it was the lowest in respiratory inorganics with
to compare the total impacts of each bottle. As such, the data can a value of 0.00389 kg C2H4 eq. The stages which contributed to the
depict both the bottle that has the highest impact on the environ- regular PET bottle’s impact in respiratory inorganics was 42.6% due
ment, as well as the stage in each bottle that contributes the most to the distribution to storage stage, and 40.1% due to the bottle
to impacting the environment. The data for each impact category manufacturing stage. Respiratory organics was primarily due to
was analyzed and compared to each other. The total impacts of four the disposal stage in the ENSO bottle, with a value of 0.00167 kg
types of bottled water are shown in Table 3 and the impacts of dif- C2H4 eq. The high impact during the disposal stage was likely
ferent production stages of each bottle type are shown in Fig. 2. due to the release of organic chemicals into the atmosphere during
the decomposition of the bottle in a landfill.
3.1.1. Carcinogens, non-carcinogens and respiratory effects
The carcinogens, non-carcinogens and respiratory effects are all 3.1.2. Ozone layer depletion
environmental impacts juxtaposed alongside human health. The The ozone layer lies in the Earth’s atmosphere and plays a crit-
results showed that the carcinogens were highest for the regular ical role in protecting live forms from strong ultraviolet radiation
PET bottle, with a value of 0.411 kg C2H3Cl eq. A close second in emanating from the sun. Ultraviolet radiation has the capacity to
the category of carcinogens was the ENSO bottle, with a value of destroy organic matter and, in humans, can lead to cancer and cat-
0.355 kg C2H3Cl eq. The carcinogens of regular PET and ENSO were aracts. Certain substances break down the ozone layer, which is

Table 3
LCIA results of the bottled water with different bottle materials.

The bottle that raises the highest and 2nd highest impact in each impact categories are highlighted.

Please cite this article in press as: Horowitz, N., et al. Life cycle assessment of bottled water: A case study of Green2O products. Waste Management (2018),
6 N. Horowitz et al. / Waste Management xxx (2018) xxx–xxx

Fig. 2. Life cycle impacts in life cycle stages of the four bottles. A. Impact category percent composition for each PLA bottle. B. Impact category percent composition for each
100% recycled PET bottle. C. Impact category percent composition for each regular PET bottle. D. Impact category percent composition for each ENSO bottle.

composed of three oxygens bonded together. These substances are bottle manufacturing stage is due to the electricity use for bottle
popularly known as Ozone-Depleting Substances (ODS), and typi- manufacturing. However, the disposal stage accounted a value of
cally contain anthropogenic chemicals composed of chlorine and 1.00  10 7 kg CFC-11 eq, without which, the total impact for
bromine. In this study, the PLA bottle was found to have the high- the bottle would be 1.02  10 7, as opposed to 1.36  10 9. The
est impact in this category with a value of 1.78  10 7 kg CFC-11 enormous negative impact in the disposal stage has on the
eq. The second highest bottle in this category was the regular PET environment may have occurred because the methane was
bottle, with a value of 1.23  10 7 kg CFC-11 eq. The ozone layer captured during the decomposition of the bottle which generates
depletion of the PLA bottle was 130 times the impact of the ENSO electricity and thus reduced the amount of ODS released into the
bottle and 2.2 times the impact of the recycled PET bottle. The pri- environment when generating electricity.
mary stage responsible for the high value is the bottle manufactur-
ing stage, with a value of 87.0% of the impact category for the PLA 3.1.3. Aquatic and terrestrial ecotoxicity
bottle, and 63.9% for the regular PET bottle. This is because the Aquatic ecotoxicity and terrestrial ecotoxicity are poisonous
polymer manufacturing raised high ODS. substances released into the environment that are toxic to organ-
The ENSO bottle was the lowest in this category, containing a isms. Aquatic ecotoxicity was highest for the PLA bottle, followed
value of 1.36  10 9 kg CFC-11 eq. The stage that contributed the closely by the recycled PET bottle and the ENSO bottle, with values
most positive impacts in the category for the ENSO bottle was bot- of 1138, 1118 and 1110 kg TEG water. These results demonstrated
tle manufacturing, which constitutes more than 80% of the positive that the four bottles contributed very closely in quantity to the
impacts for the ENSO bottle. The egregiously high value in the impact category of aquatic ecotoxicity. The life cycle stage which

Please cite this article in press as: Horowitz, N., et al. Life cycle assessment of bottled water: A case study of Green2O products. Waste Management (2018),
N. Horowitz et al. / Waste Management xxx (2018) xxx–xxx 7

contributed the most to the aquatic ecotoxicity was the distribu- 3.1.6. Aquatic eutrophication
tion to storage, which was 77.8% of aquatic ecotoxicity for the recy- Aquatic eutrophication occurs when there is an abundance of
cled PET bottle, 76.5% of the PLA bottle and 80.3% of the regular PET nutrients in a body of water, typically due to runoff, which results
bottle. This was caused by the use of transportation fuels in the in an egregious amount of plant growth, and the death of zoologi-
product distribution. The results showed the aquatic ecotoxicity cal organisms from oxygen deprivation. Aquatic eutrophication
raised by the ENSO bottle was the highest in the distribution to was the highest for the recycled PET bottle, followed by the regular
storage stage, which was 79% of the category. However, there PET bottle, the PLA bottle, and lastly, the ENSO bottle, with values
was also a negative impact of 47.4 kg TEG water for the ENSO of 4.57  10 4, 3.82  10 4, 3.72  10 4 and 2.92  10 4 kg PO4
bottle in the disposal stage, which was raised by the electricity P-lim. The main stage that contributes to the recycled PET bottle
generation from biogas. having a higher magnitude in the category was the bottle manufac-
Terrestrial ecotoxicity was highest for the PLA bottle, with a turing stage, which is 80.2% of the impact category for the bottle.
value of 52.3 kg TEG soil. The high value was mainly due to the bot- The processes of bottle sorting and cleaning contributed to the high
tle manufacturing stage, which was 85.6% of the total impact cat- eutrophication impact of recycled PET bottle. Out of expectation,
egory. Factors, such as pesticide runoff from corn production the PLA bottle was not the highest in the eutrophication, although
contributed to the PLA having high quantities in the impact cate- the fertilizers used to grow corn cause eutrophication.
gories of ecotoxicity. In contrast, the recycled PET bottle was the
lowest in the terrestrial ecotoxicity category with a value of 17.0
kg TEG soil. This is mainly because recycled material was reused 3.1.7. Global warming potential
after consumption, which diverted the wastes that can raise eco- Greenhouse gases (GHGs), such as carbon dioxide and methane,
toxicity in the landfills. The ENSO bottle was the second to lowest are frequently released into the atmosphere by natural causes (i.e.
in terrestrial ecotoxicity, because it contained a negative impact volcanoes, respiration), and anthropogenic origins (i.e. automobiles
value of 9.04 kg TEG water in the disposal stage. The negative and smokestacks), where they then become part of Earth’s atmo-
value was caused by the electricity generation from biogas. sphere and help capture heat to warm the planet, therefore, the
GHGs are used to evaluate the global warming impact. The ENSO
bottle contained the highest quantity, 6.42 kg CO2 eq, in this cate-
3.1.4. Aquatic and terrestrial acidification/nutrification gory. A very likely explanation for the high GWP was the release of
Aquatic and terrestrial acidification occurs when the environ- greenhouse gases as it decomposed in landfills, which contributed
ment is affected by certain chemicals which leads to a lower than 46.8% of the total GWP for the ENSO bottle. The PLA bottle con-
normal pH in that environment. Aquatic acidification regularly tained the second to lowest impact, 3.58 kg CO2 eq, in this cate-
occurs when gases, such as CO2 or SO2, are absorbed by the water gory, which was only 55% of the ENSO bottle. This is because the
and react to forms acidic compounds. When there is a higher than PLA bottle consists of corn, which is a plant that extrapolates CO2
normal concentration of gases in the atmosphere, the ocean from the atmosphere for photosynthesis, thereby reducing the
absorbs more gases and becomes more acidic. In aquatic acidifica- greenhouse gases in the atmosphere causing global warming.
tion, the PLA, recycled PET and typical PET contained values of Finally, the recycled PET bottle had the lowest global warming
0.0223, 0.0222 and 0.0217 kg SO2 eq. The aquatic acidification of impact with a value of 3.57 kg CO2 eq. because it used recycle
the ENSO bottle was not significant lower than other three. For PET polymer and reduced the use of petroleum, a major contribu-
all four bottles, the highest contributing stage was the distribution tor to GHG emissions.
to storage, followed by bottle manufacturing. This was due to the
fact that during distribution to storage the transportation fuels
were consumed, which affected the ocean’s acidity by releasing 3.1.8. Non-renewable energy
gases, such as carbon dioxide into the atmosphere. Those gases Non-renewable energy is energy utilized that cannot, within a
have the potential to be absorbed by the ocean and lower the pH reasonable duration of time, be reused again, i.e. petroleum and
of the water. coal. The typical PET bottle had the highest impact in this category,
Terrestrial ecotoxicity impact quantities for four bottles were with a value of 69.4 MJ primary. The recycled PET and the ENSO
not different significantly. Identical to aquatic acidification, all bottle contained almost identical lowest impacts in this category,
three bottles contained the highest contributing stage in the distri- with values of 55.5 MJ primary. The distribution to storage and
bution to storage stage, with the second to highest being the bottle the bottle manufacturing were two stages contributing to this
manufacturing stage. impact category, because electricity was used for polymer manu-
facturing and fossil fuels were used for transportation.

3.1.5. Land occupation

Land occupation was the highest for the PLA bottle with a value 3.1.9. Mineral extraction
of 0.233 m2orgarable, approximately 4–5 times to other three bot- Mineral extraction is a measurement of the amount of resources
tles. The stage contributed the greatest to this high quantity for the required to create each bottle. The bottle with the highest impact
PLA bottle was the bottle manufacturing stage with a value of in the category of mineral extraction was the regular PET bottle
0.218 m2orgarable, or 93.5% of the total PLA bottle land occupa- with a value of 0.0840 MJ Surplus. The PET required continuous
tion impact. Since the PLA polymer is corn based, which requires extraction of fossil fuels to create the bottle, which was confirmed
extensive arable land to grow and harvest a crop of corn, the bottle by the highest stage impact being bottle manufacturing, and thus is
manufacturing stage of the PLA bottle was the biggest contributing likely why the regular PET bottle contained the highest impact in
factor to this impact. The ENSO bottle contains the smallest quan- the category of mineral extraction. The PLA and ENSO bottles con-
tity in the category with a value of 0.0418 m2orgarable. The low tained the median amount of impact in this category, likely due to
impact was due to the fact that the ENSO bottle is able to fully the fact that they also required continuous extraction of resources.
degrade once it is disposed of. In addition, the electricity genera- The recycled PET was the lowest in this category, with a value of
tion created a benefit in land occupation, which attributed to a 0.0236 MJ Surplus, because the polymer was a reused material,
negative impact 9.08  10 3 m2orgarable in the bottle disposal and thus required less mineral extraction than the other three
stage. bottles.

Please cite this article in press as: Horowitz, N., et al. Life cycle assessment of bottled water: A case study of Green2O products. Waste Management (2018),
8 N. Horowitz et al. / Waste Management xxx (2018) xxx–xxx

3.2. Examination of four bottles in the life cycle stages highest in non-carcinogens, respiratory inorganics, aquatic ecotox-
icity, aquatic acidification and terrestrial acid/nutria. This may
The results showed the recycled PET bottle was the lowest in have occurred due to the pollution emitted from automobiles in
impact categories of carcinogens, terrestrial ecotoxicity, global the distribution process. The bottle manufacturing was the highest
warming potential, non-renewable energy, and mineral extraction. in the remaining categories because the ENSO bottle contains
The PET recycling and reuse contributed significantly to reduce chemicals in order to make the additive, and new PET material to
environmental impacts. Low values in carcinogens and terrestrial make the bottle. In this process the disposal was negative for many
ecotoxicity occurred because the plastic was being reused, and categories because some of the carbon dioxide and methane
thus less petroleum and additional chemicals were required to cre- released is able to be recaptured and used as a co-product to gen-
ate the bottle. The non-renewable energy was lowest because less erate electricity, and thus it positively benefits the environment.
fossil fuel was required to make the polymer. In addition, the study Lastly, the PLA bottles demonstrated the worst environmental
assumed the PET bottles were recycled locally, which reduced fos- performance with the highest impacts in seven categories. The
sil fuel use in polymer transportation and thus lowered down the results showed that generating a product from natural materials
non-renewable energy use. Land occupation of the recycled PET was not necessary better than petroleum based or synthetic mate-
bottle may be second to lowest because the bottle was reused, rials. The PLA bottle resulted in the bottle manufacturing process
and so no land was required as a landfill during the disposal stage. and distribution to storage having the highest impacts in almost
The recycled PET bottles resulted in three categories being the every category. The distribution of the PLA bottle to storage was
highest for the bottle manufacturing stage, four categories being the highest in non-carcinogens, respiratory inorganics, respiratory
the highest in the packaging of the recyclable PET bottles, and eight organics, aquatic ecotoxicity, terrestrial acidity/nutrificiation,
categories being the highest for the distribution to storage stage. aquatic acidification, global warming and non-renewable energy.
The carcinogens may be higher for packaging than recycled PET This high impact was due to the automobile emissions released
manufacturing, because the recycled PET bottles are being made into the environment. Distribution, on the other hand, had abso-
from recycled material, whereas the packaging is new raw material lutely no impact on the impact categories of land occupation or
which must be extracted. Packaging may also be highest in mineral mineral extraction. A possible cause for the other impact categories
extraction since the materials to make the packaging must be new, being highest in bottle manufacturing is because the extraction of
whereas the material to make the bottle is previously extracted the materials, transport of materials to the factory, and assembly of
recycled material. The distribution to storage is highest in eight materials, require an intensive amount of energy. The PLA bottle
categories possibly due to the pollution emitted from automobiles manufacturing composes approximately 93.5% of the land occupa-
in the distribution process. The recycled PET bottle manufacturing tion impact because the area to harvest corn requires an expansive
stage may be the highest in the remaining categories due to the area of land. The eutrophication category was impacted 83.0% from
energy required to assemble and form the bottle. the bottle manufacturing process, which may be a result of the fer-
The typical PET bottle was the highest in the categories of car- tilizers applied to the corn crops. The PLA bottle manufacturing
cinogens, respiratory inorganics, non-renewable energy, and min- composes 82.6% of mineral extraction most likely as a result of
eral extraction. It was also the second highest in other seven the quantity of electricity required to create the bottles.
categories. The impacts of the regular PET bottle were raised A past study conducted by Groot and Borén found that PLA bot-
mainly in the bottle manufacturing stage and product distribution tles resulted in significantly less emissions of greenhouse gases and
stage, in which the petroleum was used for polymer manufacturing usage of non-renewable energy when compared to fossil-based
and for product transportation. The typical PET bottle resulted in polymers, such as PET. The results found by Groot and Borén are
seven of the highest categories being impacted by the bottle man- consistent with the global warming potential and the non-
ufacturing and the remaining seven categories being the highest in renewable energy impacts found in this study. GWP and
the distribution process of the PET bottles. Contrary to the recycled non-renewable energy impacts in both studies were lower than
PET, the regular PET does not contain the highest impacts from the the regular PET alternative impacts. Madival, et al., conducted a life
packaging because the bottles are made from scratch each time in cycle assessment comparing PLA and PET clamshell containers.
the bottle manufacturing stage as opposed to the 100% recycled Their results found that PET contained higher impacts than PLA
PET bottles. bottles in the categories of global warming, ozone layer depletion,
The ENSO bottle had the highest values in the categories of res- aquatic eutrophication, aquatic ecotoxicity and land occupation,
piratory organics and global warming, and lowest in the categories while PLA contained greater magnitudes for aquatic acidification,
non-carcinogens, respiratory inorganics, ozone layer depletion, respiratory organics and respiratory inorganics. The data similar
aquatic and terrestrial acidification, land occupation and aquatic between these studies is that the GWP and aquatic eutrophication
eutrophication. ENSO contained 99% of same material as regular in PLA is lower than for the regular PET bottle, as well as that PLA
PET, but impacts of ENSO were lower than regular PET signifi- contains higher impacts in the impact category aquatic acidifica-
cantly. As the ENSO decomposition can create biogas to generate tion compared to the regular PET. These results may differ between
electricity, the bottle disposal stage had negative impacts in all cat- each other due to the fact that clamshell containers and water
egories. This contributed to the lower impacts of the ENSO bottle bottles contain different additional components, such as packaging
than other bottles. However, when the bottle degrades, it releases and labels, which may have led to contrasting results.
CO2 and CH4 into the atmosphere, which in turn intensifies the
greenhouse effect. The bottle degradation in landfills also caused 3.3. Analysis of bottles with combined materials
the highest respiratory organics. Similar to the regular PET, the
ENSO raised impacts in the stages of bottle manufacturing and pro- The LCA results showed the recycled PET and ENSO bottles are
duct distribution to storage. more environmental friendly bottles for bottled water production,
The LCI for the ENSO bottle demonstrated that the process of whereas, the regular PET and PLA bottle are less environmentally
bottle manufacturing was the highest in seven categories, distribu- favorable. Bottles are also able to be made by combining various
tion to storage in five categories, and disposal in two categories. materials. This study carried out an analysis to determine whether
Disposal was the highest in respiratory organics and global warm- a more sustainable PET bottle would be able to be produced if the
ing because once it was brought into a landfill it degrades, releas- recycled PET resin and regular PET resin were mixed together in
ing chemicals into the atmosphere. The distribution centers were certain quantities to produce a bottle.

Please cite this article in press as: Horowitz, N., et al. Life cycle assessment of bottled water: A case study of Green2O products. Waste Management (2018),
N. Horowitz et al. / Waste Management xxx (2018) xxx–xxx 9

Mineral extracon

Non-renewable energy

Global warming

Aquac eutrophicaon

Aquac acidificaon

Land occupaon

Terrestrial acid/nutri

Terrestrial ecotoxicity

Aquac ecotoxicity

Respiratory organics

Ozone layer depleon

Respiratory inorganics



-25.00% -20.00% -15.00% -10.00% -5.00% 0.00% 5.00% 10.00% 15.00% 20.00% 25.00%

Recyclable PET Bole Manufacturing 10% R 90% N Recyclable PET Bole Manufacturing 50% R 50% N

Fig. 3. Sensitivity analysis on the recycled PET bottle and the regular PET bottle. Tornado diagram portraying the percent change in each impact category when the
composition of materials each bottle is made with is changed. The base line is 30% R recycled PET and 70% of N normal/regular PET.

The analysis utilized the 30% recycled PET resin mixed with 70% 4. Conclusion
regular PET as a baseline to compare the 10% recycled PET - 90%
regular PET with the 50% recycled PET - 50% regular PET. Fig. 3 The study analyzed the Life Cycle Assessment for four types of
depicts the percent change in impact categories when there is a water bottles in an effort to determine the impact each bottle
percent change in recyclable and regular PET resin. The analysis has on the environment. The process found to compose most of
demonstrated that the more recycled PET, the greater the improve- the impacts for each bottle was the bottle manufacturing stage
ment for the impacts in the categories of carcinogens, respiratory and the distribution to storage. The bottle manufacturing stage
inorganics, respiratory organics, terrestrial ecotoxicity, terrestrial was composed of creating the materials for the bottles as well as
acid/nutri, aquatic acidification, global warming, non-renewable shipping the material to be molded, which thus may have led to
energy and mineral extraction. Specifically, there were high extremely high impacts for that stage. Distribution to storage
improvements (>5%) in the categories of carcinogens, respiratory was composed only of transportation, yet contained extremely
organics, global warming, and non-renewable energy, with modifi- high impacts because the transportation was assumed to be fueled
cations of 5.85%, 19.94%, 5.89% and 10.17%. However, some by oil which caused many deleterious environmental impacts.
impacts were increased, including non-carcinogens, ozone layer These results are concordant with a study done by Gleick and Coo-
depletion, aquatic ecotoxicity, and aquatic eutrophication. The ley who found that the total energy requirement for manufacturing
increases of those impacts were significant by 18%, 1.5%, 21%, and transporting the plastic bottles were both the most energy
and 24%. Land occupation was the one category that demonstrated intensive stages. The high impacts in these two stages demonstrate
a 0% change between the two variations of bottles. The results also that it may be possible to decrease human impacts on the environ-
showed that some impact categories were more likely to change ment if material acquisition or transportation were more sustain-
than others when materials were mixed. Those impacts included ably carried out. The transportation for the recycled PET and the
eutrophication, aquatic ecotoxicity, respiratory organics, and regular PET during the bottle manufacturing stage was altogether
non-carcinogens. The small change of combination ratios would 900.6 km and 1036 km, with only a 135.4 km difference, yet the
raise high change of those impacts. bottle manufacturing stage had a percent difference in carcinogens
Due to varying results, with some categories lowering while of 182% and for mineral extraction of 154%. The continuous re-
others increased, it is difficult to determine the best percentage extraction of resources for the regular PET resin most likely led
of polymers to mix together to achieve a bottle with the least to the extremely high impacts in the bottle manufacturing stage.
detrimental impacts on the environment since each category can- The LCIA of the four bottles showed that overall, each bottle had
not be compared to each other in terms of degree of impact. Cer- high impacts in various categories; however, the ENSO bottles had
tain impact categories may be more detrimental to the the lowest impacts in more categories, with 6/14 of the categories
environment and/or humans, while other categories may cause lit- being the lowest compared to the other three bottles, and 2/14
tle to no harm. Other measures such as a cost benefit analysis being the highest. The 100% recycled PET bottle had the overall sec-
should be conducted to determine the best ratio of recycled PET ond to lowest impacts, with 4/14 of the smallest impacts in the cat-
with regular PET. egories and 1/14 of the highest impacts, when compared to the

Please cite this article in press as: Horowitz, N., et al. Life cycle assessment of bottled water: A case study of Green2O products. Waste Management (2018),
10 N. Horowitz et al. / Waste Management xxx (2018) xxx–xxx

other three bottles. The regular PET bottle contained mainly med- Huff, Ethan, 2013. ENS Plastic Bottle Breakthrough: These Bottles Decompose in
Landfills and They’re Recyclable, too!. Retrieved from https://
ian impacts compared to the other three bottles, but was the high-
est in 4/14 of the impact categories and lowest in 1/14 of the Intagliata, C., 2012. Does Recycling Plastic Cost More Than Making It? Retrieved
categories compared to the other three bottles. While the PLA bot- from
tle was shown to have an extremely low impact in the carcinogens, than-making-it.html.
Madival, S., Auras, R., Singh, S.P., Narayan, R., 2009. Assessment of the
respiratory organics and global warming categories, it still con- environmental profile of PLA, PET and PS clamshell containers using LCA
tained 7/14 of the highest impacts compared to the other three methodology. J. Clean. Product. 17 (13), 1183–1194.
bottles. The use of more sustainable bottles, such as biodegradable Muschiolik, G., 1997. The Wiley Encyclopedia of Packaging Technology. Second
Edition. John Wiley & Sons, Inc., New York, Chichester, Weinheim, Brisbane,
ENSO bottles and recycled PET bottles, appears to be a viable Singapore, Toronto. pp 742–745.
option for decreasing humanity’s carbon footprint on the planet. Revathi, R., Kumar, T. R., Raman, M. S., Umanath, B., & Student, U. G. (2017). Reuse of
Bottles for Wall Construction and Crafting. International Journal of Engineering
Science, 6693.
Appendix A. Supplementary material Rodwan, John, 2016. U.S. and International Developments and Statistics. Retrieved
Supplementary data associated with this article can be found, in
the online version, at Royte, Elizabeth. (2006). Corn Plastic to the Rescue. Smithsonian Magazine.
043. Retrieved from
Shen, L., Worrell, E., Patel, M.K., 2010. Open-loop recycling: A LCA case study of PET
References bottle-to-fiber recycling. Resources, Conservat. Recycl. 55 (1), 34–52.
Sinha, V., Patel, M.R., Patel, J.V., 2010. PET waste management by chemical
British Plastic Federation, 2016. Oil consumption Retrieved from http://www.bpf. recycling: a review. J. Poly. Environment 18 (1), 8–25. Statista, 2017. Per capita consumption of bottled water worldwide in 2016,
Center for Sustainable Systems, University of Michigan. 2015. ‘‘Municipal Solid by leading countries (in gallons) Retrieved from https://www.statista.
Waste Factsheet.” Pub. No. CSS04-15. com/statistics/183388/per-capita-consumption-of-bottled-water-worldwide-in-
ENSO Bottles. (2009). Making Plastic Bottles Environmentally Friendly. Retrieved 2009/.
from Pros and Cons of the Corn-Based Plastic PLA PLA is Carbon Neutral and Burns Clean,
Gourmelon, Gaelle, 2015. Global Plastic Production Rises, Recycling Lags. Retrieved But Has a Host of Unsolved Problems. Retrieved from News & Issues, http://
lags-0. Zia, K.M., Bhatti, H.N., Bhatti, I.A., 2007. Methods for polyurethane and polyurethane
Hopewell, J., Dvorak, R., Kosior, E., 2009. Plastics recycling: challenges and composites, recycling and recovery: a review. React. Funct. Poly. 67 (8), 675–
opportunities. Philosoph. Transact. Royal Soc. London B: Biolog. Sci. 364 692.
(1526), 2115–2126.

Please cite this article in press as: Horowitz, N., et al. Life cycle assessment of bottled water: A case study of Green2O products. Waste Management (2018),