l1 10 Plastic PDF
l1 10 Plastic PDF
LCA of plastic waste recovery into recycled materials, energy and fuels in T
Singapore
Hsien H. Khoo
Institute of Chemical and Engineering Sciences, 1 Pesek Road, Jurong Island, 627833 Singapore
A R T I C LE I N FO A B S T R A C T
Keywords: In Singapore, 822,200 tonnes of plastic waste was generated in 2016, with only 7% recycled. Due to the complex
LCA nature of plastic waste mixtures, mechanical recycling is often inefficient, leading to a majority of waste being
Plastic wastes incinerated. In this article, alternative solutions are introduced to address the waste problem, such as recovering
Plant capacities valuable fuels from plastics via thermochemical methods. Life cycle assessment (LCA) was adopted to investigate
Mechanical recycling
8 scenarios of plastic waste management options. In a nation with land scarcity, the scales and sizes of each
Pyrolysis
plastic waste recycling/recovery method is also taken into consideration. The results demonstrated how different
Gasification
combinations of four plastics valorization technologies, and associated capacities, affected the potential en-
vironmental benefits and drawbacks of plastic waste treatment systems. In order to enable selecting the best
option among the 8 scenarios, normalization and weighting was carried out.
https://doi.org/10.1016/j.resconrec.2019.02.010
Received 29 November 2017; Received in revised form 8 February 2019; Accepted 8 February 2019
Available online 26 February 2019
0921-3449/ © 2019 Elsevier B.V. All rights reserved.
H.H. Khoo Resources, Conservation & Recycling 145 (2019) 67–77
the percentages of separate collections, and disposal options; all play a recycling, WTE, and potential pyrolysis or gasification plants. In a na-
part in affecting the credibility and acceptability of LCA results. In tion with land scarcity, the scales or sizes of valorization or waste
another case, Slagstad and Brattebø (2013) looked into uncertainty treatment technologies are an important consideration (Whiting et al.,
factors relating to waste compositions and discussed the effect of such 2013).
uncertainties in both accounting LCA and comparative LCA. Herein, the term “recycling” refers to mechanical recycling of
Varying models or approaches to further expand waste-LCA have plastic waste, and the term “recovery” refers to the thermal conversion
been reported. In order to apply optimization objectives in LCA in- of plastic wastes into electrical power, fuels and/or chemicals.
vestigations of waste systems, Münster et al. (2015) presented systems
engineering optimization model, OptiWaste. The proposed model can
assist in prioritizing waste options while incorporating LCA methods. 2. Plastic waste management in Singapore
Laner et al. (2015) included exergy analysis in the LCA of four residual
household waste management scenarios. The first approach considered Singapore is a small nation with a total land area of 719.7 km2 and
exergy flow analysis of the waste treatment systems, and in the second therefore lacks spaces for landfills. The nation also faces the challenge
approach, an “exergetic LCA” using the Cumulative Exergy Extraction of diversifying energy resources to satisfy the needs of a robust, growing
from the Natural Environment (CEENE) was applied. Both approaches industry. Presently, most plastic wastes are sent to WTE plants for the
offered different quantitative results and conclusions regarding mate- twin purpose of reducing waste volumes and deriving energy in the
rial recovery. The environmental profiles of household wastes con- form of electrical power. At the WTE, an estimated 634 kW h of elec-
sidering the LCA of different waste management systems were per- trical power can be generated from 1 tonne of mixed plastics wastes. At
formed by Bassi et al. (2017) for a European-wide case study involving the same time, 90% reduction of waste volume occurs, and the re-
Germany, Denmark, France, UK, Italy, Poland and Greece. They con- maining 10% residual waste consisting of bottom and fly ash are then
cluded that environmental performances scored best with the mini- sent to Singapore’s offshore landfill, Pulau Semakau, by barge.
mization of disposal via landfills.
A list of more LCA applications in waste management is summarized
in Table 1 A, Appendix A. Further methodological reviews, strengths 2.1. Plastic waste recycling
and limitations, waste scenario modelling and guidelines with regard to
the application of LCA in waste management can be found in the works Albeit challenges faced with complex plastic waste mixtures, me-
of Laurent et al. (2014a; 2014b), Manfredi et al. (2011), and Clear chanical recycling is the key technology employed to recycle plastic
(2010). wastes (National Environmental Agency, 2017a). Generally, the first
step of mechanical recycling involves the size reduction of plastics to
finer forms such as pellets or flakes. This can be done via grinding or
1.2. Paper objectives shredding (Zia et al., 2007). Further steps involve the removal of con-
taminants and impurities, and bonding of clean plastics via agglutina-
So far, there are limited reports on the conversion of plastic waste tion process with the addition of pigments and additives. The dis-
into valuable, high-calorific fuels in Singapore. There is also no publicly advantages of mechanical recycling include difficulties in sorting out
available reference pertaining to work done on the environmental each type of plastic fraction from a large amount of mixed plastics
outcomes of plastic waste processed via thermochemical methods based waste (MPW), and high levels of contaminations. In that case, extra
on their associated capacities. This article attempts to provide a new steps are needed in sorting and cleaning (Al-Salem et al., 2017). The
perspective on the LCA of plastic waste treatment by including plant resultant useful products from mechanical recycling are polyethylene
capacities in the LCA system comparison. The environmental impacts of terephthalate (PET) and polyethylene (PE) resins. Mechanical recycling
8 plastic wastes scenarios will be carried out for an LCA model scoped can be economically viable only when these specific types of plastic
around Singapore as a “whole system”. For each case, varying fractions waste streams are available in sufficient quantities.
of 822,200 tonnes of plastic waste will be distributed to mechanical
68
H.H. Khoo Resources, Conservation & Recycling 145 (2019) 67–77
2.1.1. Recycling rates in other countries processed into methanol, ethanol, or other chemicals (Gershman and
The generation of waste plastics per capita in various countries, and Bratton, 2013; Kungkajit et al., 2015). Al-Salem et al. (2017) and Cossu
their respective recycling rates, is illustrated in Fig. 2. It can be ob- et al. (2017) both provided further comprehensive sources of in-
served that Japan is leading the way in plastic waste recycling with an formation pertaining to various plastic waste management strategies
impressive 78% rate. This is achieved via a combination of mechanical and their associated treatment methods.
and chemical recycling methods (Yolin, 2015). The generation of
plastic waste in both Hong Kong and South Korea are nearly similar
2.2. Worldwide industrial pyrolysis and gasification systems
(95 kg/capita), but with different reported recycling rates of 36% in
Hong Kong and 54% in South Korea (EPD, 2015; Vink, 2014). Despite
Pyrolysis and gasification methods employed for the thermo-
varying plastic waste generated per capita, Germany, Denmark, and
chemical conversion of various plastic wastes into valuable fuels or
Finland have reported recycling rates of around 40%–50% (Bundesamt,
other products have already been implemented on industrial scales in
2017; Toft et al., 2016; Hennlock et al., 2014). Among the countries
various parts of the world. The Sapporo Waste Plastics Liquefaction
known to control the use of plastics, the UK is leading the way with
Plant has been operating in Japan since 2000. With a capacity of
34 kg per capita of plastic waste generated in 2015 (WRAP, 2016). In
14,800 tonnes per year the plant is the biggest of its kind in Japan
comparison, Singapore is lagging behind in plastic wastes recycling.
(Fukushima et al., 2009). The Plastics2Oil (2018) or P2O process con-
In wake of the mounting problem of waste plastics generated in the
verts plastic waste into fuel via thermal-catalytic pyrolysis. The US
small city-state, along with the present inefficient recycling rates, al-
proprietary P2O process accepts almost all plastic waste as feedstock
ternative solutions are called for. Pyrolysis and gasification methods are
except for PVC and nylons (Tsiamis, 2013). Two other US-based com-
introduced in the next section for the recovery of high-calorific, valu-
panies featuring pyrolysis technologies are RES Polyflow (2018) Vadxx
able fuels from plastic waste streams.
technology (Vadxx, 2018), which has the capacity to process
25,000 tonnes of plastic waste per year (Karidis, 2017) It was reported
in November 2018 that RES Polyflow aims to start a plant in Ashley,
2.1.2. Thermochemical recovery methods
Indiana, with the capacity to convert 100,000 tonnes of plastic scrap
Various thermochemical methods have been developed and ex-
into “roughly 18 million gallons of diesel fuel and naphtha blend stocks,
tensively explored. Pyrolysis process thermo-chemically converts var-
plus 5 million gallons of wax per year” (Sandoval, 2018). The Ther-
ious types of plastic wastes into a mixture of high calorific gases (pri-
omofuel plastic-to-diesel technology is commercialized by UK-based
marily complex hydrocarbons, hydrogen and carbon monoxide), liquid
company, Cynar. The ThermoFuel conversion plants are designed to
fuels and solid residues (Butler et al., 2011). Plastics-to-fuel (PTF)
process approximately 6,000 tonnes a year of waste plastics (Recycling
technologies have emerged as a potential option to convert plastic
International, 2010). One can refer to reports by Gershman and Bratton
wastes into a valuable resource. The principal output of PTF technol-
(2013) and Ocean Recovery Alliance (2015) for more information on
ogies is a liquid petroleum product – either a synthetic crude oil or
existing pyrolysis and gasification technologies employed worldwide.
refined fuels (Ocean Recovery Alliance, 2015; Kunwar et al., 2016).
To the best of our knowledge, there is no industrial or large-scale
Many pyrolysis technologies offer the advantage of omitting an in-
commercial plant operating in Singapore for the thermochemical con-
tensive sorting process and can accept various types of mixed plastics
version of mixed plastic streams into fuels or chemicals.
(Tsiamis, 2013). One can refer to Sharuddin et al. (2016) for a com-
prehensive review of pyrolysis technologies for the recovery of valuable
fuels from waste plastic streams. 3. Method: life cycle assessment
Unlike pyrolysis, gasification takes place in an oxygen‐controlled
atmosphere to allow the partial combustion of waste materials at high Here, we apply life cycle assessment (LCA) evaluate the potential
temperatures of 600 °C–800 °C (Brems et al., 2013). Gasification tech- environmental impacts or drawbacks of various plastics waste recycling
nology potentially offers feedstock flexibility, including an assortment and recovery options. LCA is carried out according to ISO standards and
of plastic wastes, for generating a wide range of products (Brems et al., frameworks (ISO, 2006a, 2006b). The four steps in LCA are 1) Goal and
2013). The primary output is synthesis gas that can then further be scope definition, 2) Inventory analysis, 3) Impact assessment, and 4)
69
H.H. Khoo Resources, Conservation & Recycling 145 (2019) 67–77
Table 1
Compilation of LCA scenarios.
Scenarios Brief Description Reference flow of waste plastics in tonnes/year and percentage recycled/recovered (in
brackets)
MR WTE P G
1 Recycling rate of 7.24% sent to MR; the rest of is sent to WTE 59,500 762,700 (93%) 0 0
Reference Case (7%)
2 Recycling rate of 10.64% sent to MR; the rest of is sent to 87,500 734,700 0 0
WTE (11%) (89%)
3 Recycling rate of 7.24% sent to MR plus potential P; the rest 59,500 732,700 30,000 0
of is sent to WTE (7%) (89%) (4%)
4 Recycling rate of 7.24% sent to MR plus potential G; the rest 59,500 687,700 0 75,000
of is sent to WTE (7%) (84%) (9%)
5 Recycling rate of 7.24% sent to MR plus potential P + G; the 59,500 657,700 30,000 75,000
rest of is sent to WTE (7%) (80%) (4%) (9%)
6 Recycling rate of 10.64% sent to MR plus potential P + G; the 87,500 629,700 30,000 75,000
rest of is sent to WTE (10%) (77%) (4%) (9%)
7 Recycling rate of 10.64% sent to MR plus potential 2 x P; the 87,500 674,700 60,000 0
rest of is sent to WTE (10%) (83%) (7%)
8 Recycling rate of 10.64% sent to MR plus potential 2 x G; the 87,500 584,700 0 150,000
rest of is sent to WTE (11%) (71%) (18%)
Interpretation. • The mixed plastic waste composition is estimated to consists of: 40%
PE, 17% PVC, 12% PP, 4% PS, 4.8% PET, and 22.2% other mixed
3.1. Goal and scope definition fractions
• Electrical energy is supplied by the national grid, which is mostly
The goal of the LCA is to compare the environmental profiles of the powered by natural gas
following plastic waste management options: mechanical recycling, • Other energy needs are supplied by fossil-based fuels (natural gas,
WTE, pyrolysis and gasification. A few potential scenarios will be diesel)
compared against the reference case of waste recycling rate of 7.24% • Plastic waste combustion at the WTE enables a volume reduction of
via mechanical recycling, as reported in 2016. The functional unit 90%; the remaining 10% is sent to the landfill by barge
serves to define the basis of comparison in the LCA system. Nation-wide • The sea travel distance from the WTE on the mainland to the off-
waste-LCA studies commonly take into account a regional amount of shore landfill is about 25 km
waste output generated in a specific geographical boundary (please • Solid waste from the pyrolysis and gasification plants are also sent to
refer to Table 1 again). In this case, a functional unit is the waste the WTE
treatment of 822,200 tonnes of plastic waste. • Un-recycled plastics from the mechanical recycling plants and un-
The LCA scope considers the geographical boundary covering the processed waste plastics from other waste treatment facilities are
entire nation as a “whole system” – starting from annual waste gen- sent directly the WTE
eration and central collection, before proceeding to the distribution of • Pollution from land transportation of waste is considered minimal
plastic wastes to various recycling options and the WTE. From the WTE, and therefore can be omitted from the scope of the LCA
solid wastes consisting of fly ash and bottom ash are sent by barge to • The output from WTE is in the form of electrical energy; and output
Semakau landfill. from mechanical recycling is recycled PET and PE
• The output from both pyrolysis and gasification are fuels which are
estimated to be 42.9 MJ/kg and 26.8 MJ/kg respectively
3.2. Scenarios and LCA modelling
The LCA model of all 8 scenarios was constructed using Gabi LCA
A total of 8 scenarios will be investigated, out of which, 7 cases will software (Thinkstep, 2018).
be compared against the reference system (scenario 1) of sending
59,500 tonnes of plastics waste for mechanical recycling, and the rest
(762,700 tonnes) to the WTE. In all scenarios, the capacities – in terms 3.3. Life cycle inventory (LCI)
of tonnes per year (tpy) – of the recycling and recovery technologies are
taken into consideration. Existing capacities from several reports (i.e., The input-output data for the WTE and mechanical recycling,
Gershman and Bratton, 2013; Neidel and Jakobsen, 2013; Ocean compiled in Tables 2 and 3, is sourced from Perugini et al. (2005) and
Recovery Alliance, 2015) are taken as representatives for the scenarios. National Environmental Agency (2016) respectively. Missing informa-
Based on the information provided, the following are introduced: ad- tion on energy input for the WTE and emissions are estimated from
ditional 28,000 tonnes-per-year or tpy mechanical recycling plant, po- McDougall et al. (2003) and Gradus et al. (2016). Information on
tential 30,000 tpy PTF facility based on pyrolysis technology, and pyrolysis and gasification plants, documented in Table 4, is extracted
lastly, a potential 75,000 tpy gasification system. It is assumed in all from various available datasets and reports describing the respective
cases that the bulk of leftover, un-recycled plastic wastes are sent di- thermochemical recovery technologies (e.g., RTI International, 2012;
rectly to the WTE. TNO, 2001; Tsiamis, 2013). In all cases, the input-output data for me-
A total of 8 scenarios are tested – combining the types and capacities chanical recycling and WTE are reported in Tables 2 and 3 respectively.
of plastic waste recycling and recovery options. Table1 gives a summary The data does not include the resources and emissions associated with
of the main plastic waste reference flows in tonnes/year, as well as, the the cradle-to-gate production of fossil-based resources needed (e.g.,
percentages of the associated waste fractions (in brackets) for scenarios diesel, natural gas). Finally, Tables 5 and 6 contains the data of air
1 to 8 emissions per MWh electrical power generation in Singapore, and air
Several conditions are modelled in the LCA system: emissions per tonne-km of barge transportation, respectively.
70
H.H. Khoo Resources, Conservation & Recycling 145 (2019) 67–77
Table 2
Input-output data for sorting, shredding and mechanical recycling of plastics (Perugini et al., 2005).
Sorting of 1 tonne Mixed Plastic Waste (MPW) Comments
Table 3 Table 4
Input-output data for WTE combustion of plastic waste (National Input-output data for pyrolysis/gasification of plastic waste (Tsiamis, 2013; RTI
Environmental Agency, 2016). International, 2012; TNO, 2001).
1 tonne of mixed plastics waste sent to WTE Pyrolysis of 1 tonne of mixed plastics waste
b
Energy input (all in MJ) Energy Comments Energy input (all in MJ)a Output Comments
(output)
Natural gas 9.86 2300 MJ Electrical power for use Natural gas 18.3 650 kg diesel Useful product
Electricity 252 Electricity 106
WTE emissions (all in g/m3 except otherwise indicated) Process emissions (all in kg)
CO2 950 kg/tonne c Greenhouse gas emission CO2 150 Greenhouse gas emission
CO 0.010 Emission contributing to CO 0.07 Emission contributing to particulate
particulate formation formation
SO2 0.142 Emissions contributing to NOx 0.6 Emissions contributing to
HCl 0.153 acidification potential SO2 3.36 × 103 acidification potential
PM (dust) 2.5 × 103 Emission contributing to NMVOC 4.48 × 103 Emission contributing to particulate
particulate formation PM (dust) 9.52 × 103 formation
Mercury 3.6 × 106 Substances contributing to air Wastes (kg)
11
Dioxin 3.6 × 10 pollution Solid waste 65 Sent to WTE
11
Furan 3.6 × 10
Solid waste 100 kg Sent to Semakau landfill Gasification of 1 tonne of mixed plastics waste
b
Supplemented by McDougall et al. (2003). Energy input (all in MJ)a Output Comments
c
Adjusted according to Johnke et al. (2001).
Natural gas 376 278 kg Useful product
Electricity 1450 ethanol
3.4. Life cycle impact assessment (LCIA)
Process emissions (all in kg)
CO2 181 Greenhouse gas emission
The ReCiPe impact assessment method was selected due to its CO 0.45 Emission contributing to particulate
flexibility to provide characterization factors that are representative of formation
global scales, with possible adjustments for impact categories to be CH4 0.907 Greenhouse gas emission
NOx 0.49 Emissions contributing to
applied at a country and continental scale (Huijbregts et al., 2017). The SO2 0.091 acidification potential
following impacts were generated with the help of Gabi software: cli- VOC 0.408 Emissions contributing to
mate change (in kg CO2-eq), terrestrial acidification (in kg SO2-eq), and PM (dust) 0.18 particulate formation
particulate matter formation (in kg PM10-eq). These impacts are se- Wastes (kg)
Solid waste 130 Sent to WTE
lected due to the emphasis of Singapore’s national targets to reduce
greenhouse emissions and enhance air quality standards, especially
SO2, NO2 and particulate matter (National environmental Agency, 4 and 5 respectively. In all cases, we observe that the selected re-
2017b). cycling/recovery technologies, along with the associated available (or
In addition to these impact categories, two other sustainability in- combination of) capacities, play a significant role in the potential en-
dicators have been considered. One is the total residual wastes gener- vironmental benefits and drawbacks of plastic waste treatment.
ated from each recycling/recovery methods. Next, the total net energy Greenhouse gas emissions, namely CO2, are mostly from the WTE.
output (in total GJ) will be calculated based on the total potential worth This is because large volumes of plastic wastes are being distributed for
of energy output minus the amount consumed for each of the 8 plastic incineration in all 8 cases, starting from 93% in the reference case, to
waste management option. 71% in the last scenario (Fig. 3). The highest impact from WTE is
748,500 tonnes CO2-eq (scenario 1) and lowest, 625,500 tonnes CO2-eq
(scenario 8). The mechanical recycling of plastic wastes offers the
4. Environmental Impact Results
benefit of redirecting waste material away from the WTE and, ulti-
mately, landfills; nevertheless, the stages involved in mechanical re-
The environmental impact results for climate change, terrestrial
cycling processes (e.g., sorting, shredding, cleaning, etc.) all take up fair
acidification, and particulate matter formation are displayed in Figs. 3,
71
H.H. Khoo Resources, Conservation & Recycling 145 (2019) 67–77
Table 5 (2 × 75,000 tpy). In this last scenario, the total climate change results
Air Emissions per 1 MW h electrical energy generated; adjusted to 90% power are is quite comparable to 6. In all cases, greenhouse gas emissions from
from natural gas based on NGCC power plant operations (Tan et al., 2010). barge transportation are rather minimal.
Input 90% Natural gas, 10% mixed petroleum products As displayed in Fig. 4, terrestrial acidification impacts from WTE are
˜109,000 kg SO2-eq (scenario 1) to ˜93,000 kg SO2-eq (scenario 8). As
Pollutants kg Comments for Fig. 5, particulate matter impacts resulting from sending plastic
waste to the WTE are ˜22,000 to ˜19,000 kg PM10-eq for scenarios 1 and
CO2 430.56 Greenhouse gas emission
CO 0.126 Emission contributing to particulate formation 8 respectively. Unlike climate change, both acidification and particulate
N2O 6.37 × 1004 Greenhouse gas emission formation impacts (Figs. 5 and 6) are observed to decline slightly from
VOC 3.07 × 1002 Emission contributing to particulate formation scenarios 1 to 2; then proceed to climb to scenario 5. From there the
d
NOx 9.90 × 1002 Emissions contributing to acidification potential
d
impacts are reduced again, before ascending to the highest peaks at
SOx 7.02 × 1006
d
PM 8.89 × 1005 Emission contributing to particulate formation
scenario 8. For both impacts, the defining patterns and peaks on the
graphs are mostly from the process emissions (PM, NOx, SOx) emitting
d
Adjusted according to Skone and James (2010). from the gasification of plastic wastes. The highest impacts for acid-
ification and particulate formation are scenario 8. In both cases, the
Table 6 introduction of 2 × 75,000 tpy gasification plants resulted in the
Air Emissions per 1 km-tonne barge travel (Ecoinvent, 2018). highest emissions for these two impact categories.
Input Diesel
Both acidification and particulate matter formation impacts from
Pollutants kg Comments the pyrolysis of plastics are also evident in scenarios 3, 5, 6 and 7, albeit
not as significant as those generated from the gasification of plastic
CO2 0.0296 Greenhouse gas emission wastes. The contributions to both acidification and particulate forma-
CO 2.54 × 105 Emission contributing to particulate formation
N2O 3.11 × 106 Greenhouse gas emission
tion impacts from mechanical recycling pale in comparison to the other
CH4 2.25 × 107 three plastic waste treatment options. Environmental impacts from
NMVOC 9.39 × 106 Emission contributing to particulate formation barge transportation are negligible in all cases.
NOx 4.70 × 104 Emissions contributing to acidification potential Fig. 6 presents the levels of waste volumes ending up for disposal at
SO2 5.64 × 106
the WTE. The two highest peaks, scenario 8 and 4, are mainly due to
PM (unspecified) 9.76 × 106 Emission contributing to particulate formation
waste being sent for gasification, where 130 kg solid is generated for
every tonne of plastic waste gasified. In scenario 4, an accumulated
amounts of energy (Perugini et al., 2005; TNO, 2001) which also result 82,000 tonnes of waste resulted from the waste treatment of plastic
in CO2 emissions. waste via 7% mechanical recycling, 84% WTE and 9% gasification; and
The third scenario, which is the second highest peak among the in scenario 8, about 83,300 tonnes of waste was generated by the dis-
climate change impacts, is altogether contributed mainly by the WTE tribution of plastic waste to 11% mechanical recycling, 71% WTE and
(720,450 tonnes of CO2-eq), mechanical recycling (24,600 tonnes of 18% gasification. The least amount of annual waste is generated (sce-
CO2-eq) and pyrolysis (20,200 tonnes of CO2-eq). With the introduction nario 7) by sending 7% of plastic waste volumes to the pyrolysis fa-
of thermochemical plastic recovery systems in scenario 5–8 greenhouse cility, 10% for mechanical recycling, and the rest to the WTE.
gases from both pyrolysis and gasification are also evident in their
contribution to climate change impacts. 5. Discussion
In scenario 6, the combination of 87,500 tonnes, 30,000 tonnes and
75,000 tonnes of plastic waste stream sent for mechanical recycling, 5.1. Valuable outputs
pyrolysis and gasification respectively – altogether prevents
192,500 tonnes/year of plastic waste sent to the WTE. As a result, an One of the main incentives for the recycling plastic wastes is re-
annual reduction of about 61,000 tons of CO2-eq/year (˜8% annually) is source recovery (Al-Salem et al., 2017; Kunwar et al., 2016). As
achieved in comparison to the reference case (scenario 1). Another naphtha from fossil fuel refineries is the main source of plastic pro-
significant reduction of climate change impacts can be realized in sce- duction, the recovery of plastic to liquid oil through pyrolysis process
nario 8 with the potential installation of 2 gasification systems had a great potential since the oil produced exhibits high calorific value
(Wong et al., 2015; Sharuddin et al., 2016). This benefit is especially
72
H.H. Khoo Resources, Conservation & Recycling 145 (2019) 67–77
73
H.H. Khoo Resources, Conservation & Recycling 145 (2019) 67–77
5.2. Comparison of GHG results with other case studies can be recycled, and at the same time prevent waste being sent for
incineration or WTE (Wilson et al., 2013; Tunesi, 2011). The sizes or
In many other case studies, greenhouse gas (GHG) emissions are scales of such waste treatment technologies have to be considered in a
viewed as the most important environmental aspect for plastics waste nation with a total land area of 719.7 km2.
management or recycling. Apart from climate change impacts, the Gasification is a well-developed technology since 1970s and there-
evaluation of acidification, particulate matter formation, solid waste fore several industrial or large-scale gasification processes for plastic
and energy generation (Figs. 4–7) adds value to this work. Nevertheless, wastes have already been designed (Gershman and Bratton, 2013;
to further enhance the significance of GHG results, this impact is Fabry et al., 2013). Compared to gasification, large-scale pyrolysis
compared with other cases. Due to different functional units employed technologies are yet to be available. For a space-constrained city-state,
by other cases (e.g., Cossu et al., 2017; Gear et al., 2018), the GHG it will be necessary to identify ideal locations to set up large thermo-
results are standardized to kg CO2-eq per tonne mixed plastics waste chemical recycling plants (Tunesi, 2011). Additionally, higher capital
(MPW). The comparison is presented in Fig. 8. cost and larger floor area for high-volume operating equipment will
It can be observed that the GHG results from WTE/incineration from likely be expected (Wilson et al., 2013).
our case study fall within a lower range as compared to those reported Land area requirements for setting up pyrolysis and gasification
by Gear et al. (2018) and Morandin et al. (2016). The lower amount is facilities are evaluated from reports (e.g., Beston Machinery, 2016;
basically due the fact that the energy supply required at the WTE mostly World Energy Council, 2016; Whiting et al., 2013). It is estimated that a
comes from natural gas. The results for mechanical recycling matches 30,000 tpy pyrolysis facility would require approx. 4,500 m2 floor area,
well with WRAP (2008), a UK-based investigation on plastics waste and a 75,000 tpy gasification system requires approx. 13,500 m2 floor
management. Nagy (2017) reported significantly lower GHG emissions area. These floor spaces are projected to include raw material or feed-
for mechanical recycling; this is most likely because the mechanical stock and product storage units.
recycling process reported is powered by 50% hydroelectric energy.
Both gasification and pyrolysis methods employed for the treatment
of plastic waste are expected to differ in operating conditions and 6. Overall comparison
temperatures (e.g., Sandoval, 2018; Tsiamis, 2013; Gershman and
Bratton, 2013) and therefore results in varying GHG emissions as shown 6.1. Normalization and weighted scores
on the graphs: ˜ 600 to 870 kg CO2-eq/t MPW for pyrolysis (PE
International, 2015; WRAP, 2008) and ˜370–880 kg CO2-eq/t MPW for Prado et al. (2017) emphasized that normalization and weighting is
gasification (Cossu et al., 2017; Morandi et al., 2016; Chen et al., 2011). crucial in comparative LCA studies to aid decision making, especially
Gear et al (2018) presented a unique case of waste plastics treatment for cases where alternative scenarios result in better environmental
via thermal cracking as an alternative solution to replace incineration performances in some areas but unfavorable in other impact categories.
and landfilling. A few pilot scale designs were tested and its resultant In the comparison of food waste conversion options, Khoo et al (2010)
GHG emissions fall between the range of pyrolysis and gasification divided the resultant impacts by geographical-based reference values
technologies. per capita to aid in the interpretation of separate impact indicators, and
generated a single score for the comparison of two or more technolo-
gical options.
5.3. Size of recycling plant Normalization is also done for benchmarking purposes. As an ex-
ample, Zhong et al. (2011) carried out normalization as a bench-
Waste plastic recycling or recovery technologies with large capa- marking step in their LCA comparison of a polycrystalline photovoltaic
cities offer the benefit of increasing the volumes of plastic wastes that (PV) module and a wind turbine. In their investigation, all the impact
74
H.H. Khoo Resources, Conservation & Recycling 145 (2019) 67–77
results were divided by a reference score. In order to compare the best environment, in this case Singapore’s National Environmental Agency.
option among this 8 plastic waste investigation, a similar normalization It is also recommended that waste planning policies should include
approach is carried out to calculate the total overall impacts in relation incentives to encourage environmentally responsible investments in
to a selected reference value. Instead of adopting country-wide nor- innovative technologies that offer the dual benefit of minimizing waste
malized impacts (per capita), the results from the reference case is used and at the same time satisfy the need for renewable fuels (Ang et al.,
as the normalized benchmark against scenarios 2 to 8. Therefore the 2015).
impact indicator results from scenario 1 will be employed to scale the The results of scenario 7 is recommended as the best option due to
following: climate change, terrestrial acidification, particulate matter the following:
formation, wastes and energy generation.
After the normalization step, weighting factors are given to each • Reduction of 31,000 tons CO -eq 2
impact category to express their relative importance, and applied to • Potential energy output of 3 × 10 GJ 6
aggregate the normalized results into a single score. Among all en- • Prevention of about 3,000 tonnes of solid waste.
vironmental concerns, we consider Singapore’s importance of green-
house gas mitigation as upmost priority and hence place the highest These annual benefits can be realized by sending 7% of plastic
weighted factor for climate change impacts (Ho et al., 2015; Low, wastes for pyrolysis, 11% for mechanical recycling, and the rest to the
2018). Energy security (Ang et al., 2015) and the reduction of waste WTE. Since a 30,000 tpy pyrolysis facility would require approximately
levels (disposed at the WTE) are both taken as the next two important 4,500 m2 floor area, two would require a total land area of about
sustainability indicators for the nation. Since energy derived from 9,000 m2.
plastic waste is considered as a valuable output, rather than an en-
vironmental burden, inverse (negative) scores will be allocated for
energy generation. The normalized and weighted values are compiled 7. Concluding remarks
in Table 7. The total normalized and weighted scores are displayed in
Fig. 9. Plastics are readily discarded after use and to prevent large volumes
From the results displayed in Fig. 9, we recommend scenario 7 as of disposal ways to step up efforts to convert waste to valuable re-
the most preferred option based the lowest normalized and weighted sources are necessary. Due to the complex nature of plastic waste
scores (total = 3.2). As compared to the reference case (total = 5.0), an mixtures, often accompanied by the presence of impurities, mechanical
annual reduction of 31,000 tonnes CO2-eq can be achieved via the recycling has proven to be difficult and not widely employed in
distribution of 87,500 tonnes of plastic wastes for mechanical recycling, Singapore. An alternative solution was suggested to recover valuable
60,000 tonnes for pyrolysis, and the rest to the WTE. Additionally, fuels from plastic waste via thermochemical conversion technologies.
combined total of 3 × 106 GJ potential worth of energy can be derived, With high contents of carbon and hydrogen in plastics, thermal de-
and about 3,000 tonnes of solid waste/year for disposal at the WTE can gradation process at elevated temperatures via pyrolysis can lead to the
be prevented. production of significant amount of hydrocarbons fuels with high ca-
The second best option is observed to be from scenario 3 (total = lorific contents.
4.2); here, a net worth of 2.33 × 106 GJ can potentially be generated This article presented a new perspective on the LCA of plastic waste
(mostly from the introduction of a 30,000 tpy pyrolysis plant), how- treatment by including waste treatment plant capacities in the in-
ever, a reduction of merely 7,850 CO2-eq greenhouse gas emissions/year vestigation. In the LCA case study, the environmental profile of plastic
are realized. waste management options was generated for a functional unit of
It should be highlighted however that the normalization and 822,200 tonnes of plastic waste in Singapore. A total of 8 scenarios are
weighting factors displayed in Table 8 are given according to the re- compared in the LCA investigation. Three potential plastic treatment/
lative importance of environmental impacts and sustainability in- recycling plants were introduced for comparison against the reference
dicators for Singapore (Ho et al., 2015; Low, 2018; Ang et al., 2015). scenario of distributing plastic waste streams to: 28,000 tpy mechanical
Different normalization methods can be applied (Sala et al., 2012), recycling plant, 30,000 tpy PTF facility, and 75,000 tpy gasifier.
which will result in different outcomes (Heijungs et al., 2007). The article demonstrated that selected plastics recycling/conversion
Weighting factors can also be influenced by altered political views or technologies – along with their associated available (or combination of)
agendas, geographical settings, environmental regulations, or even cost capacities – play a significant role in the potential environmental
(Ahlroth et al., 2011; 2014). Therefore the final scores (Fig. 9) can be benefits and drawbacks of plastic waste treatment. The final normalized
influenced by alterations made to the set of normalization and/or and weighted results from the LCA model concluded that Scenario 7
weighting factors applied. was selected as the optimum option with lowest normalized and
weighted scores (total = 3.2) as compared to the reference case
6.2. Suggestions to policy makers (total = 5.0). However, it should be noted that the final scores are
subjective to the set of normalization and weighting factors that were
It is worth noting that Singapore has strong ambitions towards given according to the environmental impacts and sustainability in-
adopting sustainable practices in waste and resource management. dicators that were viewed to be important. A different set of normal-
Policy makers play the important role of formulating environmental ization and/or weighting factors may alter the conclusions.
regulations and setting national legislations to ensure a clean
Table 7
Normalized and weighted factors.
Normalization and weighting factors Impacts
Climate change (tons/year) Acidification Particulate matter formation (kg/year) Total Waste Energy generation
(kg/year) (tons/year) (GJ/year)
Normalized scores‡ 7.73 × 105 1.11 × 104 2.26 × 104 7.99 × 104 × 106
Weighted factors 3.0 1.0 1.0 2.0 2.0
‡
Scenario 1 as reference.
75
H.H. Khoo Resources, Conservation & Recycling 145 (2019) 67–77
Acknowledgements Chen, X., Xi, F., Geng, Y., Fujita, T., 2011. The potential environmental gains from re-
cycling waste plastics: simulation of transferring recycling and recovery technologies
to Shenyang, China. Waste Manage. 31, 168–179.
With special to appreciation ExxonMobil Chemical Company for Clavreul, J., Guyonnet, D., Christensen, T.H., 2012. Quantifying uncertainty in LCA-
funding this research work. Their feedback and guidance received modelling of waste management systems. Waste Manage. 32, 2482–2495.
during the course of this LCA project were highly valuable. And also, Clear, J., 2010. The incorporation of waste prevention activities into life cycle assess-
ments of municipal solid waste management systems: methodological issues. Int. J.
additional appreciation to colleagues Naraharisetti Pavan Kumar, for Life Cycle Assess. 15, 579–589.
help in evaluating the sizes of the pyrolysis/gasification facilities, and Cossu, R., Garbo, F., Girotto, F., Simion, F., Pivato, A., 2017. PLASMIX management: LCA
Zhou Ying, for data collection and compilation. of six possible scenarios. Waste Manage. 69, 567–576.
Das, P., Tiwari, P., 2018. Valorization of packaging plastic waste by slow pyrolysis. Res
Coserv. Rec. 128, 69–77.
Appendix A. Supplementary data EcoInvent, 2018. Transport – Freight: Barge. Ecoinvent Center, Switzerland.
EPD, 2015. Waste Statistics of Hong Kong in 2013. Environmental Protection Department
(EPD), Hong Kong.
Supplementary material related to this article can be found, in the
Fabry, F., Rehmet, C., Rohani, V.J., Fulcheri, L., 2013. Waste gasification by thermal
online version, at doi:https://doi.org/10.1016/j.resconrec.2019.02. plasma: a review. Waste Biol. Valoriz. 4, 421–439.
010. Fukushima, M., Shioya, M., Wakai, K., Ibe, H., 2009. Toward maximizing the recycling
rate in a Sapporo waste plastics liquefaction plant. J. Mater Cycles Waste Manage. 11,
11–18.
References Gear, M., Sadhukhan, J., Thorpe, R., Clift, R., Seville, J., Keast, M., 2018. A life cycle
assessment data analysis toolkit for the design of novel processes - a case study for a
Ahlroth, S., 2014. The use of valuation and weighting sets in environmental impact as- thermal cracking process for mixed plastic waste. J. Clean. Prod. 180, 735–747.
sessment. Res. Conserv. Recycl. 85, 34–41. Gershman, Brickner, Bratton, Inc., 2013. Gasification of Non-recycled Plastics From
Ahlroth, S., Nilsson, M., Finnveden, G., Hjelmet, O., Hochschorner, E., 2011. Weighting Municipal Solid Waste in the United States. Report No. GBB/12038-01, Prepared for:
and valuation in selected environmental systems analysis tools - suggestions for the American Chemistry Council.
further developments. J. Clean. Prod. 19, 145–156. Gradus, R., van Koppen, R., Dijkgraaf, E., Nillesen, P., 2016. . A Cost-Effectiveness
Al-Salem, S.M., Antelava, A., Constantinou, A., Manos, G., Dutta, A., 2017. A review on Analysis for Incineration or Recycling of Dutch Household Plastics. Tinbergen
thermal and catalytic pyrolysis of plastic solid waste (PSW). J. Environ. Manage. 197, Institute Discussion Paper (Report No. TI 2016-039/VI). VU University Amsterdam,
177–198. Faculty of Economics and Business Administration, Amsterdam, the Netherlands.
Ang, B.W., Choong, W.L., Ng, T.S., 2015. A framework for evaluating Singapore’s energy Heijungs, R., Guinée, J., Kleijn, R., Rovers, V., 2007. Bias in normalization: causes,
security. Acs Appl. Energy Mater. 148, 314–325. consequences, detection and remedies. Int. J. Life Cycle Assess. 12 (4), 211–216.
Astrup, T., Møller, J., Fruergaard, T., 2009. Incineration and co-combustion of waste: Hennlock, M., Zu Castell-Rüdenhausen, M., Wahlström, M., Kjær, B., Milios, L., Vea, E.,
accounting of greenhouse gases and global warming contributions. Waste Manage. et al., 2014. Economic Policy Instruments for Plastic Waste – a Review With Nordic
Res. 27, 789–799. Perspectives. Nordic Council of Ministers, Denmark.
Bassi, S.A., Christensen, T.H., Damgaard, A., 2017. Environmental performance of Ho, J.C., Gautam, J., Low, M., 2015. Energy Studies Institute, SingaporeSingapore’s
household waste management in Europe - an example of 7 countries. Waste Manage. Intended Nationally Determined Contribution (INDC) for COP-21 Climate Conference
69, 545–557. in Paris,2015. Singapore’s Intended Nationally Determined Contribution (INDC) for
Beston Machinery, 2016. Waste Plastic Pyrolysis Plant Project Report. (Accessed: 30 COP-21 Climate Conference in Paris.
November 2018). https://plasticpyrolysisplants.com/. Huijbregts, M.A.J., Steinmann, Z.J.N., Elshout, P.M.F., Stam, G., Verones, F., Vieira, M.,
Blengini, G.A., Fantoni, M., Busto, M., Genon, G., Zanetti, M.C., 2012. Participatory ap- et al., 2017. ReCiPe2016: a harmonized life cycle impact assessment method at
proach, acceptability and transparency of waste management LCAs: case studies of midpoint and endpoint level. Int. J. Life Cycle Assess. 22, 138–147.
Torino and Cuneo. Waste Manage. 32, 1712–1721. ISO, 2006a. Environmental Management - Life Cycle Assessment - Principles and
Brems, A., Dewil, R., Baeyens, J., Zhang, R., 2013. Gasification of plastic waste as waste- Framework, ISO 14040. International Organization for Standardization, Switzerland.
to-energy or waste-to-syngas recovery route. Nat. Sci. 5, 695–704. ISO, 2006b. Environmental Management - Life Cycle Assessment – Requirements and
Bundesamt, Umwelt, 2017. Plastics Waste in Germany. (Accessed 4 December 2018). Guidelines, ISO 14044. International Organization for Standardization, Switzerland.
https://www.umweltbundesamt.de/en/print/topics/waste-resources/product- Johnke, B., Hoppaus, R., Lee, E., Irving, B., Martinsen, T., Mareckova, K., 2001. Emissions
stewardship-waste-management/plastics. From Waste Incineration: Good Practice Guidance and Uncertainty Management in
Butler, E., Devlin, G., McDonnell, K., 2011. Waste polyolefins to liquid fuels via pyrolysis: National Greenhouse Gas Inventories. Intergovernmental Panel on Climate Change
review of commercial state-of-the-Art and recent laboratory research. Waste Biol. (IPCC).
Valoriz. 2 (3), 227–255. Karidis, A., 2017. Waste-to-Energy: Alternative Thermal Conversion Technologies Creep
76
H.H. Khoo Resources, Conservation & Recycling 145 (2019) 67–77
77