Content uploaded by Ola Eriksson
Author content
All content in this area was uploaded by Ola Eriksson
Content may be subject to copyright.
Plastic waste as a fuel - CO
2
-neutral or not?
Ola Eriksson*
a
and G€
oran Finnveden
b
Received 23rd April 2009, Accepted 24th June 2009
First published as an Advance Article on the web 28th July 2009
DOI: 10.1039/b908135f
Municipal solid waste (MSW) is not only a societal problem addressed with environmental impact, it is
also a resource that can be used for energy supply. In Northern Europe combustion of MSW
(incineration with energy recovery) in combination with district heating systems is quite common. In
Sweden, about 47% of the household waste is treated by incineration with energy recovery. Most
incineration plants are CHP, summing up to 0.3% of the total electricity generation. MSW is to a high
extent a renewable fuel, but plastic, rubber etc. can amount to 50% of the carbon content in the waste.
Recycling of plastic is in general environmentally favourable in comparison to landfill disposal or
incineration. However, some plastic types are not possible to recycle and some plastic is of such low
quality that it is not suitable for recycling. This paper focuses on the non-renewable and non-recyclable
plastic in MSW. A CO
2
assessment has been made for non-recyclable plastic where incineration
with energy recovery has been compared to landfill disposal. In the assessment, consideration has been
taken of alternative fuel in the incinerator, emissions from waste treatment and avoided emissions from
heat and power supply. For landfill disposal of plastic the emissions of CO
2
amounts to 253 g kg
1
plastic. For incineration, depending on different discrete choices, the results vary from 673 g kg
1
to
4605 g kg
1
. Results indicate that for typical Swedish and European conditions, incineration of plastics
has net emissions of greenhouse gases. These emissions are also in general higher for incineration
than for landfill disposal. However in situations where plastics are incinerated with high efficiency and
high electricity to heat ratios, and the heat and the electricity from incineration of plastics are replacing
heat and electricity in non-combined heat and power plants based on fossil fuels, incineration of plastics
can give a net negative contribution of greenhouse gases. The results suggest that efforts should be
made to increase recycling of plastics, direct incineration of plastics in places where it can be combusted
with high efficiency and high electricity-to-heat ratios where it is replacing fossil fuels, and reconsider
the present policies of avoiding landfill disposal of plastics.
Introduction
Municipal solid waste (MSW) contains resources from society
that can be recovered as material or energy. In a global
perspective MSW is mostly disposed of at landfill with no or poor
resource recovery. The waste production per capita changes
widely between different countries, in Fig. 1 municipal solid
waste generation in European countries is depicted.
MSW will be more important as a resource in the future as the
amount of waste increases. For example, the amount of house-
hold waste has increased by 23.8% during 1998–2007 in Sweden.
2
This trend is also valid for waste from other sources and in other
countries.
1
In Northern Europe combustion of MSW (incineration) in
combination with district heating systems is quite common. In
Sweden, a country with a high degree of material recycling
(37%) and biological treatment (12%), about 47% of the
household waste is treated by incineration with energy
recovery.
2
Incineration accounts for approx. 25% of the district
a
Division of Building Quality, Department of Technology and Built
Environment, University of G€
avle, G€
avle, SE 801 76, Sweden
b
Division of Environmental Strategies Research – fms, Department of Urban
Planningand Environment, School of Architecture and the Built Environment,
KTH (Royal Institute of Technology), Stockholm, SE 100 44, Sweden
Broader context
The awareness of climate change has increased dramatically during the last years. Many societal activities contribute to emissions of
gases contributing to climate change, amongst them waste management including emissions of fossil CO
2
from incineration of waste
and CH
4
from degradation of organic material in landfills. Recycling of plastics is often possible and many studies have shown that
material recycling is in general the best waste treatment option from an environmental perspective. But what is the best treatment
option for non-recyclable plastic; incineration or landfill disposal? In this paper we compare and assess these two options. For both
methods energy (both as electricity and heat) can be recovered. Our results indicate that for plastics, landfill disposal is often
preferable over incineration regarding emissions of gases contributing to climate change.
This journal is ªThe Royal Society of Chemistry 2009 Energy Environ. Sci., 2009, 2, 907–914 | 907
ANALYSIS www.rsc.org/ees | Energy & Environmental Science
heating generated in Sweden where residential heating as well as
process heat is included.
2
Most incineration plants recover not
only heat but also electricity (Combined Heat and Power,
CHP), summing up to 18% of the electricity generation from
CHP in Swedish district heating
3
which corresponds to 0.3% of
the total electricity generation.
4
In many other countries, energy
from waste incineration is mostly recovered as electricity with
varying efficiencies. Energy recovery from waste can also be
achieved by anaerobic digestion producing biogas that can
replace fossil fuels for transport or be used for heat and power
production.
Compared to other fuels used in heat and power supply,
municipal solid waste has both advantages and disadvantages
with respect to technology, environmental impact and financial
costs. One advantage is that MSW often has a low initial burden.
Only some collection and transportation is required. Other fuels
often have costs and environmental impacts also from the
production of the fuels. MSW is also a fuel available close to the
end user. Due to urbanisation people tend to live in cities or
greater residential areas. This means high waste generation per
km
2
which in turn makes the cost for waste collection less
expensive than in sparsely populated areas. Another aspect is
that, compared to material recycling and biological treatment,
incineration plants are robust with respect to waste quality. The
plants are designed to be able to treat bulky heterogeneous waste.
Another advantage can be observed for fuel economy with
respect to the owner of an incinerator. Even if some costs are
higher due to expensive cleaning equipment, revenue from sold
energy is supported by a reception fee for the waste fuel. This
makes MSW competitive with e.g. biomass as base load in
district heating systems.
Nevertheless, MSW is a more troublesome fuel compared to
other fuels due to its heterogeneity. Here pre-treatment and
recycling helps, as the combustible fraction then becomes more
homogenous since a well developed sorting (both at source or at
a MRF (Material Recovery Facility) can produce more clean
combustible flows where different materials are not mixed. The
variable fuel quality gives rise to problems during the combus-
tion process which requires an advanced and expensive air
pollution control and management of residues. Even if the
overall degree of efficiency can be high due to flue gas conden-
sation, the electricity-to-heat ratio is in general lower than for e.g.
wood combustion, which in turn in general is lower than for oil
and natural gas combustion.
Within the European Union as well as in many other coun-
tries, waste policy is based on the waste hierarchy:
5
1. Waste prevention
2. Re-use of products
3. Recycling of material
4. Recovery of energy
5. Final disposal
A large number of studies have been looking into the envi-
ronmental aspects of different waste management strategies.
There are also several synthesis reports available.
6–9
These studies
seem to suggest that the waste hierarchy in general is valid from
an environmental perspective. There can however be situations,
or waste fractions, which deviate from the hierarchy. In this
paper we will investigate one such possible exception and that is
the treatment of plastic wastes concerning carbon dioxide and
other gases contributing to climate change.
MSW is partly a renewable fuel which is regarded as not
contributing to climate change. However, the waste often
Fig. 1 Waste generation in European countries during 2007 in kg/capita.
1
908 | Energy Environ. Sci., 2009, 2, 907–914 This journal is ªThe Royal Society of Chemistry 2009
includes plastic and (synthetic) rubber that is produced from
fossil fuels, thus being non-renewable fuels contributing to
climate change. The fossil content in MSW can be measured in
different ways; by weight percent, energy percent or carbon
percent. The balance between biogenic and fossil carbon dioxide
will vary between different countries and incinerators depending
on the type of waste that is incinerated, recycling programs etc.
One recent Swiss study indicated that the biogenic fraction is
about 50% or slightly more.
10
One American study
11
indicated
that the biogenic fraction was approx. 65%. Recycling of plastic
is in general environmentally favourable in comparison to
landfill disposal or incineration.
6–9
However, some plastic types
are not technically possible to recycle, some may be too expensive
to recycle and some plastic is of such poor quality that it cannot
be recycled. That means, that even with 100% source separation
of plastic (no plastic found in the combustible fraction of MSW),
there will be some non-recyclable plastic that has to be treated.
Plastic is usually made out of crude oil (bio-plastics are not
of fossil origin and give no net increase of CO
2
when com-
busted) and is chemically of two types: thermoplastic and
thermoset. Thermoplastic, if exposed to heat, will melt and is
therefore possible to reshape and recycle. Thermoset on the
other hand will keep their shape until they are a charred,
smoking mess. Thermoset are however possible to shred and
the chips can then be used as filling material or reinforcement
material in new products. Thermoset plastics go under
product names such as polyurethane (PUR), acryl, epoxy and
different variants with formaldehyde like Urea Formaldehyde
(UF), Melamine Formaldehyde (MF) and Phenol Formalde-
hyde (PF). Common applications are e.g. small boats, pipes
and cable isolation.
12
Thermoplastic can be recycled several
times but eventually the quality is too poor. Another reason
why a potentially recyclable plastic is sorted out at the recy-
cling plant is due to impurity. Therefore not only thermoset,
but also thermoplastic can be subject to energy recovery.
Besides, plastic products other than packages are collected at
recycling centres. This plastic, which can be of both types, is
in Sweden often incinerated in cement kilns or regular incin-
erators due to it not being included in the system for
producers’ responsibility. Landfill disposal is prohibited in
many countries including Sweden. Most of the plastic pack-
ages in MSW are incinerated, cf. Table 1.
Aim of the paper
This paper will determine under which circumstances incinera-
tion of plastics as a part of MSW may have a zero or negative
impact on global warming;
1. Combustion of the fossil part (plastic) will cause CO
2
emissions, but are the avoided emissions from substituted
fuels higher?
2. Is non-recyclable plastic better treated in a sanitary landfill
with respect to global warming in a short time perspective?
3. Are steps 4–5 in the waste hierarchy correct for non-
recyclable plastic?
Method
Goal and scope definition
ACO
2
accounting is made for waste treatment of 1 kg of non-
recyclable plastic. The methodology used is based on established
methods for environmental Life Cycle Assessment (LCA) of
waste management systems
13,14
and the ISO-standard for LCA
(ISO, 2006).
15
Two types of plastic are investigated: non-recy-
clable and a mixed fraction of presumed poor quality. The plastic
waste itself has a zero burden when entering the waste manage-
ment system. This assumption can be made since the plastic
waste (in terms of density) is identical in the two cases compared
(landfill disposal and incineration) and identical parts of the
systems can be excluded in a comparative study.
13,14
The waste
treatment options are incineration with energy recovery which
has been compared to landfill disposal. In the assessment,
consideration has been taken of GHG emissions (CO
2
,CH
4
and
N
2
O) from waste treatment and avoided emissions from heat and
power supply.
Two scenarios are being compared. In the first scenario the
plastics are subject to sanitary landfill disposal. Parameters of
concern are
Emissions from landfill management
Carbon sequestration in landfill (not carbon sink as the
carbon is fossil)
In the second scenario the plastics are subject to incineration
with combined heat and power (CHP). Parameters of concern
are
Emissions from the incineration plant
Avoided emissions from other electricity generation
Avoided emissions from other heat generation
The assessment is based on data for Swedish and European
conditions. Choice of data has been made to achieve CO
2
-lean
and CO
2
-intense alternatives to explore as wide a span as
possible. The assessment is thought to cover both consequential
and book-keeping conditions (c.f. ref. 16) and represents a short
time perspective with existing plants rather than a long time
perspective with improved technology.
Data inventory
Parameters crucial to GWP impact have been identified and data
acquisition has been made. Concerning the plastic, two proper-
ties are of interest: heat value and chemical composition (in this
case the fossil carbon content). Heat values vary between
different plastics depending on the molecular structure. The
calculations use the higher heating values because incinerators in
Sweden are equipped with flue gas condensation and the
condensation heat recovered is included in figures on efficiency
and total energy recovery. For example PVC has a heat value of
19 MJ kg DM
1
meanwhile one of the most common
Table 1 Treatment of plastic packages in 2007
2
Waste management Amount (ton) Percentage (%)
Source separated, material
recycling
49 119 30.1
Source separated, energy recovery 56 434 34.5
Not source separated, incineration
and landfill disposal
57 842 35.4
This journal is ªThe Royal Society of Chemistry 2009 Energy Environ. Sci., 2009, 2, 907–914 | 909
thermoplastics Polyethene has 46 MJ kg DM
1
. The values
depicted in Table 2 have been used.
The next step would be collection and transport of the plastic
to a treatment plant. This step is omitted, based on the afore-
mentioned assumption and also that we have assumed the
distance from the point of collection to a sanitary landfill and to
an incinerator to be the same.
Regarding the landfill it is primarily a question of to what
extent the plastic is decomposed and also which time frame to
use. If plastic is degraded in the landfill we will attain air emis-
sions of CO
2
and CH
4
, but on the other hand some of the landfill
gas could be extracted and collected with energy recovery and
therefore saving other fuels. The energy recovery suffers from
a low efficiency compared to incineration with energy recovery.
The result for the landfill is the accumulated estimated emissions
(and energy recovery) during a period of approx. 100 years,
a time frame commonly used for landfills in LCA and also to
comply with the weighting factors used.
In this assessment we have assumed that the plastic will be
degraded by 3% during the first century due to a chemical reac-
tion
19
whereas 70% of the degraded carbon will form methane
with a yield of 0.025 kg CH
4
kg
1
plastic. We have assumed that
50% of the methane is collected and combusted. The rest is
emitted, but 15% is released as CO
2
due to soil oxidation in the
top layer.
19
The collected landfill gas is combusted in a gas engine
with 85% degree of efficiency where 37.5% of the useful energy is
recovered as electricity.
20
Besides emissions of CO
2
the gas engine
emits methane (430 mg MJ
1
CH
4
) and also N
2
O (31 mg MJ
1
CH
4
).
20
There will also be some emissions of CO
2
due to landfill
management (mainly for compacting and creating a sealing
cover). The energy consumption is 0.04 MJ kg
1
plastic
21
and
emission factors are taken from ref. 18.
If the plastic is put in an incinerator we assume complete
combustion. Following parameters are then of interest for the
incineration plant:
1. Degree of efficiency
2. Heat and power plant versus heat plant
3. Emission factor for N
2
O
The degree of efficiency may vary widely from plant to plant.
The following figures are taken from ref. 22. In Sweden 20
incineration plants are operative while ten plants are for heat
generation only and combust 31% of all MSW and the others are
CHP. For this purpose we have not chosen data from one single
plant, instead we use Swedish average values for the total heat
and power production. In the Swedish case we have used a mean
value for all Swedish incinerators in 2006 which is 80%. Of the
useful energy 8.2% is released as electricity and 91.8% as heat. If
we had used an average Swedish CHP the figures would be 12%
electricity and 88% heat. In the same report we found mean
values for European incinerators with data from CEWEP. The
average degree of overall efficiency is reported to be 41% and
energy recovered as electricity is 26% and heat 74%. Another
study in this field
23
calculates electricity production at 13%, 24%
or 28% and the corresponding values for heat generation of 70%,
55% and 0%. Both degree of efficiency and the electricity-to-heat
ratio may have a strong influence on the results; therefore we
will test the influence of these parameters in a sensitivity ana-
lysis. Emissions of N
2
O may also vary for different plants and
is not restricted in the incineration directive. We use the figure
0.69 mg MJ
1
.
24
The alternatives to heat and power from combustion of plastic
come next. Let’s begin with district heating. The fossil intense
alternative was chosen to oil combustion. Oil is no longer
a common fuel in Swedish district heating, in 2007 it accounted
for just about 3.5%.
3
Hard coal accounted for 4%. But even if
these remaining parts soon will be substituted there may be
a secondary effect as some residential oil heating still can be
replaced by district heating. On a European level there is still
high oil consumption for heating purpose.
1
Emission factors for
CO
2
,CH
4
and N
2
O were collected from ref. 25.
The fossil lean alternative is biomass combustion in a CHP. In
a decision situation where a new base load heat plant is to be
built in Sweden, an energy company often has a choice between
building a biomass fired boiler, or an incinerator when consid-
ering costs and technical aspects. Bringing in a CHP as an
alternative heat source makes the assessment a bit more
complicated due to the fact that not only heat but also electricity
is generated. Fig. 2 shows a simplification of the alternatives in
terms of energy balance. A CHP for MSW generates roughly
80% of the energy as heat and 20% as electricity. In the first case
(alternative I) oil combustion will be substituted for heat and also
a corresponding amount of avoided electricity (we will come to
the fuels for electricity later). In the second case (alternative II)
the incinerator will substitute biofuel CHP based on heat
production. For every unit of heat the biofuel CHP will also
generate some electricity. This electricity production is ‘‘lost’’ and
may even be larger than the electricity from the incinerator (30%
in Fig. 2) as a biomass CHP often has a higher electricity-to-heat
ratio than an MSW CHP. Compared to the oil case where the
sum of heat and electricity from MSW and alternative 1 is zero,
the biomass CHP alternative leads to a net loss of electricity
(10%). This is compensated for by adding the missing electricity
production, shown as the last bar to the right in the figure
(alternative II). The biomass CHP in this study has energy
recovery with 27% electricity and 72% heat and with related
emissions; taken from ref. 25.
Last but not least, consideration must be made to compensa-
tory electricity. In the literature researchers and others use
different approaches depending on the purpose and choose data
from different sources and years. As the uncertainties are huge
we follow the recommendation to use both a fossil rich and
a ‘‘fossil free’’ alternative.
26
We have not tried to collect and
review as many studies as possible, but more to find relevant
options but with extreme values (reasonable figures while
Table 2 Data on plastics
Plastic property Value Unit Ref.
Dry matter, non-recyclable 95 % 17
Dry matter, mixed plastic
a
93 % 18
Heat value, non-recyclable 32.26 MJ kg
1
17
Heat value, mixed plastic 38.94 MJ kg
1
18
Fossil carbon content, non-
recyclable
0.6560 kg kg
1
wet waste
17
Fossil carbon content, mixed
plastic
0.6935 kg kg
1
wet waste
18
a
The values for mixed plastic are mean values from four different
analyses of household waste fractions.
910 | Energy Environ. Sci., 2009, 2, 907–914 This journal is ªThe Royal Society of Chemistry 2009
covering both high and low GWP). We have included wind
power as fossil lean and coal condense power as fossil intense.
Wind power has an emission factor of 0.0607 g CO
2
kWh
125
and for coal condense the emission factor has been chosen/set to
1000 g CO
2
kWh
1
.
Impact assessment
Table 3 gives the characterisation factors for GWP:
27
Calculations
The calculation of CO
2
-emissions from landfill disposal consists of
a sum of decomposition, soil oxidation of methane, direct methane
emissions and emissions and electricity generation from methane
combustion. These calculations are left out due to space limits.
Given the properties for plastic (Table 2) and parameters for the
incineration plant (Table 4) the CO
2
emissions and the generation of
heat and power can be calculated using the following formula.
E
CO
2
¼Mc(n
CO
2
/n
C
)1000 + (E
N
2
O
/1000)
cf
N
2
O
HHV h(1)
Wheat ¼HHV hfheat
3:6(2)
Wel ¼HHV hfel
3:6(3)
Results from insertion of waste specific data in (1), (2) and (3)
above are presented in Table 5.
Results
Given the data above a fossil carbon accounting has been carried
out. Results for landfill disposal of plastic are depicted in Table 5.
The majority of the carbon in the material will be stored inside
the landfill and prevented from release to the atmosphere.
Table 3 Characterisation factors
Emission CO
2
-equivalents
Carbon dioxide, CO
2
1
Methane, CH
4
25
Nitrous oxide, N
2
O 298
Fig. 2 Energy balance for different fuel alternatives. Units in % of energy release.
Table 5 Energy recovery and emissions of CO
2
-eq. for landfill disposal
of plastic
Landfill disposal of 1 kg of plastic Non-recyclable Mixed
Waste treatment
Degradation 5.90 6.24
Landfill gas emissions 271 271
Landfill management 3.01 3.01
Energy recovery
Heat, kWh 0 0
Electricity, kWh 0.061 0.061
Air emissions gas combustion 35 35
Avoided electricity
Coal condense power (1000 g CO
2
kWh
1
)
61 61
Sum 253.971 254.311
Table 4 Input data for calculation of emissions and energy release from
incineration (in order of appearance)
a
Quantity Unit Symbol Value
Carbon content kg C kg
1
plastic M
c
waste specific
Molar weight for CO
2
g mol
1
n
CO
2
44
Molar weight for C g mol
1
n
C
12
Emission factor N
2
OmgMJ
1
E
N
2
O
0.69
Emission factor CO
2
gCO
2
E
CO
2
waste specific
Characterisation
factor N
2
O
1cf
N
2
O
298
Higher Heating Value MJ kg
1
HHV waste specific
Degree of efficiency 1 h0.80 or 0.41
Heat recovery kWh W
heat
waste specific
Part energy to heat 1 f
heat
91.8% or 74%
Electricity recovery kWh W
el
waste specific
Part energy to electricity 1 f
el
8.2% or 26%
a
Note: 1 kWh ¼3.6 MJ.
This journal is ªThe Royal Society of Chemistry 2009 Energy Environ. Sci., 2009, 2, 907–914 | 911
However, air emissions (mainly as methane) and the generated
electricity cause maximum savings of 61 g (coal condense)
resulting in approx. 254 g CO
2
eq. kg
1
of plastic. The results for
the more complex assessment of incineration of 1 kg of plastic are
presented in Table 6. From this a number of combinations can be
derived which are listed in Table 7.
Discussion and conclusions
As can be seen from Table 7, most cases result in net emissions of
greenhouse gases. Also in most cases, the emissions are higher
than for the landfill case.
Two combinations out of sixteen in Table 7 end up as
a reduction of CO
2
emissions, namely for Swedish conditions
and when fossil fuel alternatives for both heat and power are
present. As noted above, the present situation in Sweden with
current infrastructure, only small amounts of fossil fuels are
used, making these scenarios less relevant for the current Swedish
situation. If wind power is applied, when all other preconditions
are kept constant, a minor positive contribution (35 g) is found
for mixed plastic whereas non-recyclable plastic is higher with
327 g. The result for mixed plastic (327 g) cannot be considered as
robust, as the other combinations with wind power electricity
(odd numbers) have much higher emissions. Looking at the
Table 6 Energy recovery and emissions of CO
2
-eq. for incineration of plastic
Incineration of 1 kg plastic
Sweden Europe
Non-recyclable Mixed Non-recyclable Mixed
Waste treatment
GWP emissions 2411 2549 2408 2546
Specific GWP emission, g CO
2
kWh
1
337 296 650 569
Energy recovery
Heat, kWh 6.6 7.9 2.8 3.3
Electricity, kWh 0.6 0.7 0.9 1.1
Avoided heat
Biomass CHP (19 g CO
2
kWh
1
)123 148 52 62
Electricity from biomass CHP,
kWh
2.46 2.97 1.03 1.25
Oil (318 g CO
2
kWh
1
)2083 2515 876 1058
Avoided electricity
Wind power (0.06 g CO
2
kWh
1
)0.04 0.04 0.06 0.07
Coal condense power (1000 g CO
2
kWh
1
)
586 707 949 1145
Compensatory electricity
Wind power (0.06 g CO
2
kWh
1
) 0.11 0.14 0.01 0.01
Coal condense power (1000 g CO
2
kWh
1
)
1873 2261 86 103
Table 7 Net contribution of carbon dioxide
a
Combination Incineration of plastic Avoided heat Avoided electricity Compensatory electricity Sum
Non-recyclable plastic
1 Sweden, bio CHP (he), wind (el) 2411 23 7 +0.11 ¼2241
2 Sweden, bio CHP (he) coal cond.
(el)
2411 23 7 + 1873 ¼4114
3 Sweden, oil (he) wind (el) 2411 083 .04 +0 ¼327
4 Sweden, oil (he) coal cond. (el) 2411 083 86 +0 ¼59
5 Europe, bio CHP (he), wind (el) 2408 20 +0.01 ¼2337
6 Europe, bio CHP (he), coal cond.
(el)
2408 20 +86 ¼2422
7 Europe, oil (he), wind (el) 2408 76 .06 +0 ¼1532
8 Europe, oil (he), coal cond. (el) 2408 76 49 +0 ¼583
Mixed plastic
9 Sweden, bio CHP (he), wind (el) 2549 48 6 +0.14 ¼2345
10 Sweden, bio CHP (he), coal
cond. (el)
2549 48 6 +2261 ¼4605
11 Sweden, oil (he), wind (el) 2549 515 .04 +0 ¼35
12 Sweden, oil (he), coal cond. (el) 2549 515 07 +0 ¼73
13 Europe, bio CHP (he), wind (el) 2546 24 +0.01 ¼2460
14 Europe, bio CHP (he), coal
cond. (el)
2546 24 +103 ¼2563
15 Europe, oil (he), wind (el) 2546 058 .07 +0 ¼1488
16 Europe, oil (he), coal cond. (el) 2546 058 1 145 +0 ¼343
a
he ¼heat; el ¼electricity.
912 | Energy Environ. Sci., 2009, 2, 907–914 This journal is ªThe Royal Society of Chemistry 2009
overall results the highest net contributions are found for
combinations including biomass CHP. The two most extreme in
this regard (# 2 and 10) suffer a high contribution from
compensatory electricity, i.e. lost electricity generation from
biomass CHP that has to be compensated for by coal condense
power. The combinations with biomass CHP and wind power
(1, 5, 9 and 13) are lower but still higher than all combinations
involving oil heat. The big benefit is substitution of oil heat,
which is logical as most of the released energy will be recovered as
heat (else the degree of efficiency is poor as the heat is lost).
In the discussion we will elaborate on the results from the
assessment. First the robustness of the results is examined. For
robustness it is meant that if vital preconditions are slightly
changed the conclusions will not change. This can be tested in
a sensitivity analysis.
The importance of degree of efficiency
To test this we apply the European figure to the Swedish incin-
eration plants. Now no combination ends up with total negative
emissions. The degree of efficiency must be at least 0.65–0.7 to
shift at least one of the possible two options (4 and 12) into total
negative emissions.
The importance of MSW CHP
When we assume that the Swedish plants are for district heat only
combinations 3–4 result in 141 g CO
2
and combinations 11–12
end up as 190 g CO
2
. Given oil heat substitution (which is
common for all these combinations) MSW CHP is not beneficial
if the electricity is fossil lean (initial emissions for combinations 3
and 11 were higher) where minor savings on ‘‘clean’’ electricity
are less favourable than major savings in oil heat. Likewise MSW
CHP is preferred if the electricity is fossil intense (compare
emissions for combinations 4 and 12). These results suggest that
CHP is important, particularly when fossil intense electricity like
coal condense power is substituted.
The importance of high electricity-to-heat ratio
By changing the degree of efficiency for the European incinerator
from 0.41 to 0.80 we will compare two incinerators with different
electricity-to-heat-ratios (8.2/91.8 ¼0.09 and 26/74 ¼0.35). Now
combinations 8 and 16 become the most favorable options with
1106 (583) and 1696 (343) g CO
2
. When comparing the new
results with the initial results for all combinations we conclude
that given coal condense power it is better to have a high ratio.
For wind power and oil heat it is worse to have a high ratio (less
beneficial to replace fossil lean electricity generation), but for
wind power and biomass CHP it does not matter as the results
are almost the same.
The importance of fuel for CHP plant
In the study it was assumed that the fuel for the CHP plant
was biofuel. If the fuel had been a fossil fuel, two contradictory
things would have happened. On the one hand incineration of
plastic waste would look better, since it is replacing a fossil
fuel. On the other hand, more electricity would be produced
from the CHP plant when a fossil fuel is used instead of
a biofuel since the electricity-to-heat ratio typically is higher for
oil and gas plants than for biofuel plants. If this extra elec-
tricity replaced fossil fuels instead, incineration of plastics
would appear worse.
The next step would be to explore the limitations of the study. Are
accuracy and scope important to the conclusions? The first limita-
tion to address is that the time frame is short. In the long term the
plastic will be degraded further, the incinerators may operate
differently and the alternative fuel options may be others. The
second limitation is that the assessment covers only CO
2
.In
a complete LCA, where toxic substances would be included, the
conclusions may be different, due to the fact that additives in
the plastic may leach and products of degradation areformed in the
landfill. On the other hand emissions of toxic substances from
incineration and landfill disposal of combustion residue can be
avoided. This could be worked around by further steps in the
assessment beyond the carbon accounting. Emissions to air and
water of other pollutants, the lack of resource recovery at many
landfills worldwide, problems with uncontrolled landfill fires and
pests are some of the problems not addressed in this paper. We
believe that more detailed calculations using computer models
(waste models like ORWARE or EASEWASTE and eventually full
LCA using SimaPro or similar) is called for if (1) the carbon
accounting does not show clear results (if different scenarios/alter-
natives are almost equal to each other) and (2) we want to capture
other environmental impacts than GWP. Another assumption that
may inflict on the results is if the collection and transport work
differs depending on treatment. Collection and transport of waste is
usually very expensive in comparison to the treatment cost. The
environmental impact is the opposite (low for collection and
transport in comparison to the treatment). Therefore collection and
transport is often made as efficient as possible. There may however
be situations where this assumption is not correct.
Coming back to the questions raised earlier we will draw
conclusions on
1. What is the environmentally best treatment option for non-
recyclable plastic—landfill disposal or incineration? If the
question is narrowed down to GWP, landfill disposal can
often be the least polluting treatment option.
2. Under which circumstances is landfill disposal or incineration
preferable? Incineration is to be preferred given a high effi-
ciency, a high electricity-to-heat ratio and when energy from
incineration replaces fossil fuels, at least for district heating.
3. Under which circumstances can MSW be considered as
CO
2
- neutral? See answer to question 2.
Finally we expand the discussion to encompass conclusions on
what to do with plastics presently not being recycled in Sweden:
1. Develop recycling technologies for thermoset plastics.
2. Stop incineration of plastic unless the incinerator is an
efficient CHP replacing oil heat and coal condense power.
3. Convert the plastic into a waste fuel which can be exported
to countries where it can be combusted with high efficiency
and substitute fossil fuels.
4. Reconsider the policy not to landfill plastic materials.
Acknowledgements
Financial support from the Swedish Environmental Protection
Agency is appreciated.
This journal is ªThe Royal Society of Chemistry 2009 Energy Environ. Sci., 2009, 2, 907–914 | 913
References
1Eurostat, 2009, http://epp.eurostat.ec.europa.eu/.
2 AvfallSverige, 2008a, Swedish Waste Management. Report available
at http://www.avfallsverige.se.
3 Svensk Fj€
arrv€
arme, 2009, http://www.svenskfjarrvarme.se.
4 AvfallSverige, Energi fr
an avfall ur ett internationellt perspektiv
(Energy from waste in an international perspective). Report 2008:13
(In Swedish), 2008b.
5Directive 2008/98/EC of the European Parliament and of the Council of
19 November 2008 on Waste and Repealing Certain Directives, EU,
2008.
6A.Bj
€
orklund and G. Finnveden, Recycling revisited - life cycle
comparisons of waste management strategies, Resour., Conserv.
Recycl., 2005, 44, 309–317.
7 M. Olofsson, J. Sundberg & J. Sahlin, Evaluating waste incineration
as treatment and energy recovery method from an environmental
point of view, 13th Annual North American Waste-to-Energy
Conference, 2005, Orlando, Florida, May 23–25.
8 S. Tyskeng and G. Finnveden, Comparing energy use and
environmental impacts of recycling and waste incineration,
J. Environ. Eng., 2009, submitted.
9 WRAP, Environmental benefits of recycling - an international review of
life cycle comparisons for key materials in the UK recycling sector,
Water and Resource Action Program, Oxon, UK, 2006.
10 J. Mohn, S. Szidat, J. Fellner, H. Rechberger, R. Quartier,
B. Buchmann and L. Emmenegger, Determination of biogenic and
fossil CO
2
emitted by waste incineration based on
14
CO
2
and mass
balances, Bioresour. Technol., 2008, 99, 6471–6479.
11 B. Bahor, K. Weitz and A. Szargot, Updated analysis of greenhouse
gas emissions and mitigation from municipal solid waste management
options using a carbon balance, Proceedings of Global Waste
Management Symposium, Copper Mountain, Colorado, USA, 2008.
12 National Encyclopaedia, 2009, http://www.ne.se.
13 R. Clift, A. Doig and G. Finnveden, The Application of Life Cycle
Assessment to Integrated Solid Waste Management, Part I -
Methodology, Trans. IChemE, 2000, 78(Part B), 279–287.
14 G. Finnveden, Methodological Aspects of Life Cycle Assessment of
Integrated Solid Waste Management Systems, Resour., Conserv.
Recycl., 1999, 26, 173–187.
15 ISO 14040 International Standard. Environmental management - Life
cycle assessment - Principles and framework, International
Organisation for Standardization, Geneva, Switzerland, 2006.
16 T. Ekvall and B. P. Weidema, System Boundaries and Input Data in
Consequential Life Cycle Inventory Analysis, Int. J. LCA, 2004, 9(3),
161–171.
17 EASEWASTE. Data available in computer model software http://
www.easewaste.dk, 2008.
18 J.-O. Sundqvist, A. Baky, M. Carlsson Reich, O. Eriksson and
J. Granath, Hur ska hush
allsavfallet tas omhand – utv€
ardering av
olika behandlingsmetoder (How shall the municipal solid waste be
managed – assessment of different treatment methods), 2002 IVL
Report 1462 (In Swedish with English summary).
19 G. Finnveden, A. C. Albertsson, J. Berendson, E. Eriksson,
L. O. H€
oglund, S. Karlsson and J.-O. Sundqvist, Solid waste
treatment within the framework of life-cycle assessment, J. Clean.
Prod., 1995, 3, 189–199.
20 A. Baky and O. Eriksson, Systems Analysis of Organic Household
Waste Management in Denmark: Environmental Project #822,
Danish Environmental Protection Agency, Copenhagen, Denmark,
2003.
21 A. Bj€
orklund, Environmental System Analysis of Waste Management
with emphasis on Substance Flows and Environmental Impact,
licentiate thesis, Department of Chemical Engineering and
Technology, Division of Industrial Ecology, Royal Institute of
Technology, Stockholm, Sweden (TRITA-KET-IM 1998:16, AFR
Report 211) 1998.
22 AvfallSverige 2008c, Energy Report (Status 2006), Results of Specific
Data for Energy, Efficiency Rates, Plant Efficiency Factors, NCV and
Determination of the Main Energy Results of 20 Investigated Swedish
W-t-E Plants, Report F2008:06.
23 H. Merrild, A. Damgaard and T. H. Christensen, Life cycle
assessment of waste paper management: The importance of
technology data and system boundaries in assessing recycling and
incineration, Resour., Conserv. Recycl., 2008, 52, 1391–1398.
24 Renova, Milj€
orapport 2007 f€
or avfallskraftv€
armeverket och
f€
orbehandlingsanl€
aggningen, inklusive
atervinningscentralen vid
S€
aven€
as (Environmental report 2007 for waste CHP and pre-
treatment facility, including the recycling central at S€
aven€
as), dnr
0222/08 (In Swedish), 2008.
25 S. Uppenberg, M. Brandel, L.-G. Lindfors, H.-O. Marcus,
A. Wachtmeister and L. Zetterberg, Milj€
ofaktabok f€
or br€
anslen Del
2. Bakgrundsinformation och teknisk bilaga (Environmental data on
fuels Part 2: Background information and technical appendix),
Swedish Environmental Research Institute (IVL), Stockholm,
Sweden (IVL report B 1334 B) (In Swedish), 1999.
26 G. Finnveden, A world with CO
2
caps, Int. J. LCA, 2008, 13, 365–367.
27 IPCC, Climate change 2007 - The Physical Science Basis: Contribution
of Working Group I to the Fourth Assessment Report of the
Intergovernmental Panel on Climate Change, Cambridge University
Press, Cambridge, United Kingdom and New York, NY, USA,
2007, 996 pp.
914 | Energy Environ. Sci., 2009, 2, 907–914 This journal is ªThe Royal Society of Chemistry 2009
