Content uploaded by Elena Corella Puertas
Author content
All content in this area was uploaded by Elena Corella Puertas on Feb 26, 2024
Content may be subject to copyright.
Journal of Cleaner Production 418 (2023) 138197
Available online 21 July 2023
0959-6526/© 2023 Published by Elsevier Ltd.
MarILCA characterization factors for microplastic impacts in life cycle
assessment: Physical effects on biota from emissions to
aquatic environments
Elena Corella-Puertas
a
,
*
, Carla Hajjar
a
, J´
erˆ
ome Lavoie
b
,
c
, Anne-Marie Boulay
a
a
CIRAIG, Department of Chemical Engineering, Polytechnique Montr´
eal, 3333 Queen Mary Road, suite 310, Montr´
eal, Qu´
ebec, H3V 1A2, Canada
b
CIRAIG, Department of Strategy and Corporate Social Responsibility, ESG UQAM, 3333 Queen Mary Road suite, 310, Montr´
eal, Qu´
ebec, H3V 1A2, Canada
c
Environmental Sciences Institute, UQAM, Pr´
esident-Kennedy Av. 201, Montr´
eal, Qu´
ebec, H2X 3Y7, Canada
ARTICLE INFO
Handling Editor: Maria Teresa Moreira
Keywords:
Marine litter
Plastic impacts
Life cycle impact assessment
Fate factor
Effect factor
Plastic degradation
ABSTRACT
Although plastics emissions pose hazards to ecosystem quality, life cycle assessment (LCA) methodologies do not
yet include quantication of the potential impacts of plastic leakage. To address this gap in LCA, the MarILCA
working group was founded. This work contributes to MarILCA’s output by providing characterization factors for
assessing the impacts of aquatic (marine and freshwater) microplastic emissions through the impact category of
physical effects on biota and ultimately the ecosystem quality damage category. First, the existing exposure and
effect factor (EEF) for micro- and nanoplastic emissions in aquatic compartments (Lavoie et al., 2021) is updated
using additional toxicity data, delivering a generic EEF of 1067.5 PAF m
3
/kg
in compartment
. Second, fate factors
(FFs) are developed for eleven different polymers (EPS, PS, PA/Nylon, PP, HDPE, LDPE, PET, PVC, PLA, PHA,
TRWP), three shapes (sphere/microbead, cylinder/microber, microplastic lm fragments) and ve sizes (1, 10,
100, 1000, 5000
μ
m). To calculate the FFs, a detailed degradation model and a simplied sedimentation model
are proposed. Polymer density and size play a major role in the fate, whereas the inuence of the shape is less
relevant. Ultimately, the EEF and FFs are combined to deliver midpoint and endpoint characterization factors
(CFs). Uncertainty is calculated with Monte Carlo analysis. Default and archetype-based CFs are recommended to
LCA practitioners in case the details of emission parameters are not fully known. CFs for 1
μ
m microplastic size
are proposed as interim CFs for nanoplastic emissions. Finally, endpoint CFs are tested in case studies within the
UNEP report “Supermarket food packaging and its alternatives: Recommendations from life cycle assessments”,
assessing the relative magnitude of potential microplastic impacts compared to complete LCA results. The case
studies conrm that the proposed methodology contributes to lling a gap in LCA and can assist environmental
decision-making on single-use plastics and their alternatives.
1. Introduction
Life cycle assessment (LCA) is a tool commonly used to assist envi-
ronmental decision-making. LCA aims to be holistic and already covers a
variety of impact categories (for example climate change, ozone deple-
tion or water use) that may cause damage to ecosystem quality and
human health as the two main areas of protection. Nevertheless, to date
there is no ofcial LCA methodology for assessing the potential impacts
of plastic emissions to the environment. This limits the applicability of
LCA as a tool for comparing the environmental impacts of different types
of plastic packaging and products, and their alternatives (Woods et al.,
2021).
In terms of life cycle inventory (LCI), the Plastic Leak Project (PLP)
proposed guidelines for estimating the plastic leakage of different
sources (macroplastics from plastic waste, microbers from textiles,
microparticles from tire abrasion and microplastic pellets from plastic
production) released into different environmental compartments
(ocean, freshwater, air, soil and other terrestrial compartments)
(Quantis and EA, 2020). Based on the PLP methodology, Loubet et al.
(2021) presented guidelines for plastic losses from shing devices lost at
* Corresponding author. CIRAIG, Department of Chemical Engineering, Polytechnique Montr´
eal, 3333 Queen Mary Road, suite 310, Montr´
eal, Qu´
ebec, H3V 1A2,
Canada.
E-mail addresses: maria-elena.corella-puertas@polymtl.ca (E. Corella-Puertas), carla.alchahir-alhajjar@polymtl.ca (C. Hajjar), lavoie.jerome.3@courrier.uqam.ca
(J. Lavoie), anne-marie.boulay@polymtl.ca (A.-M. Boulay).
Contents lists available at ScienceDirect
Journal of Cleaner Production
journal homepage: www.elsevier.com/locate/jclepro
https://doi.org/10.1016/j.jclepro.2023.138197
Received 9 December 2022; Received in revised form 3 July 2023; Accepted 20 July 2023
Journal of Cleaner Production 418 (2023) 138197
2
sea and marine coatings.
In life cycle impact assessment (LCIA), plastic emissions calculated in
LCI are linked with their potential environmental impacts through
characterization factors. To tackle the lack of LCIA methodologies for
plastic emissions, in late 2018 the international working group MarILCA
(MARine Impacts in LCA) was founded, supported by the UN Environ-
ment Life Cycle Initiative and the Forum for Sustainability through Life
Cycle Innovation (FSLCI) (Boulay et al., 2019). In recent years, some
approaches have emerged in parallel to MarILCA, addressing certain
aspects of the LCIA of plastic emissions (Maga et al., 2022; Salieri et al.,
2021; Zhao and You, 2022). These studies enrich this research eld and
will be addressed throughout this work.
The MarILCA framework addresses a variety of impact categories,
some of which are already commonly used in LCA (e.g. human toxicity
or ecotoxicity), whereas others are new (e.g. physical effects on biota)
(Woods et al., 2021). Note that physical effects on biota includes physical
impacts of plastics through external (entanglement, smothering) and
internal (ingestion) pathways; whereas chemical impacts of plastic ad-
ditives are covered separately in the impact category of ecotoxicity
(Tang et al., 2022). Both physical effects on biota and ecotoxicity ulti-
mately cause damage to ecosystem quality.
For the different impact categories, MarILCA aims at developing
characterization factors for emissions of plastics of different sizes
(macro, micro and nano) into different environmental compartments
(marine, freshwater, terrestrial, air) (Woods et al., 2021). In
emission-based LCIA, (midpoint) characterization factors are commonly
modelled with the following structure (Jolliet et al., 2006):
Characterization f actor =Fate factor ×Exposure factor ×Effect f actor (1)
For microplastics, the aquatic exposure and effect factor proposed by
Lavoie et al. (2021) was combined with simplied marine fate factors for
two types of microplastics – expanded polystyrene (EPS) and tire and
road wear particles (TRWP) – to deliver midpoint and endpoint char-
acterization factors (Corella-Puertas et al., 2022). These characteriza-
tion factors were tested in a case study of to-go food containers
(Corella-Puertas et al., 2022).
As a continuation of Corella-Puertas et al. (2022), the objective of
this work is to provide more advanced characterization factors for
physical effects on biota impacts of aquatic (marine and freshwater)
microplastic emissions of different polymers (EPS: expanded poly-
styrene, PS: polystyrene, PA: polyamide (Nylon), PP: polypropylene,
HDPE: high-density polyethylene, LDPE: low-density polyethylene, PET:
polyethylene terephthalate, PVC: polyvinyl chloride, PLA: polylactic
acid, PHA:, polyhydroxyalkanoate, TRWP: tire and road wear particles),
shapes (sphere/microbead, cylinder/microber, microplastic lm frag-
ments) and sizes (1, 10, 100, 1000, 5000
μ
m). (E)PS, PA, PP, HDPE,
LDPE, PET and PVC are among the most abundant (micro)plastics found
in the marine environment (Alimi et al., 2021; Curren et al., 2021; Li
et al., 2016; Orona-N´
avar et al., 2022; Suaria et al., 2016). PLA and PHA
were selected to cover “biodegradable” polymers. TRWP were included
to update the work of Corella-Puertas et al. (2022). The shapes were
chosen based on the most frequently reported microplastic shapes
(Koelmans et al., 2022).
2. Methods
To develop a methodology for quantifying the impacts of micro-
plastic emissions, the following steps are addressed:
1. Exposure and effect factor (EEF) (sections 2.1 and 3.1): Updating the
existing EEF (Lavoie et al., 2021) for nano- and microplastic emis-
sions in aquatic compartments.
2. Fate factors (FFs) (sections 2.2 and 3.2): Proposing FFs addressing
the transport (sedimentation) and degradation of microplastic
emissions in aquatic compartments.
3. Characterization factors (CFs) (sections 2.3 and 3.3): Developing CFs
for assessing the physical effects on biota of microplastic emissions by
combining proposed FFs and EEF, and calculating their uncertainty.
4. Case studies (sections 2.4 and 3.4): Testing the developed CFs in case
studies within the UNEP report “Supermarket food packaging and its
alternatives: Recommendations from life cycle assessments” (UNEP,
2022) and assessing the relative importance of physical effects on
biota impacts compared to other impact categories.
5. Recommendations for LCA practitioners (section 3.5): Providing
guidance for the application of the proposed methodology.
2.1. Exposure and effect factor
The exposure and effect factor of Lavoie et al. (2021) was updated. In
their methodology, the exposure and effect factor are combined into a
single factor (EEF), with the assumption that the exposure factor is:
XF =1kgbioavailable plastic
kgplastic in compartment
(2)
This is a conservative approach that assumes that the total plastic
quantity reaching an aquatic compartment is available for exposure to
organisms. For simplicity, the combined exposure and effect factor was
called effect factor (EF) in Lavoie et al. (2021), but in this work, it will be
called EEF to be more descriptive.
Within the EEFs described in Lavoie et al. (2021), they recommended
their “BEST” EEF. The “BEST” EEF was based on chronic EC50 (Effect
Concentration 50%) and LC50 (Lethal Concentration 50%) values and
thus required fewer extrapolations than the “ALL” EEF, which included
EC50s extrapolated from acute data, NOEC (No Observed Effect Con-
centration) and LOEC (Lowest Observed Effect Concentration) values.
There was only one exception within the “BEST” EEF: Lavoie et al.
(2021) included one extrapolated value from a chronic LOEC of Dania
rerio, since they were missing a sh species representing secondary
consumers. With this additional data point, their data represented three
trophic levels and thus were compliant with USEtox recommendations
(Fantke et al., 2018).
Since this work focuses on updating the “BEST” EEF – from now on
called just EEF, the literature search concentrated rst on gathering
additional chronic EC50 and LC50 values. In the following months, the
USEtox group (Owsianiak et al., 2023) published new recommendations
for modelling ecotoxicity effects – proposing HC20
EC10eq
(Hazardous
Concentration for 20% of the based on EC10 equivalent data, EC10:
Effect Concentration 10%) instead of HC50
EC50eq
(Hazardous Concen-
tration for 50% of the species based on EC50 equivalent data). This work
aims to be in line with USEtox recommendations and agrees that EC10
data are closer to environmentally relevant concentrations than EC50
data. Thus, additional research was carried out focusing on EC10 values.
The keywords of the searches can be found in Multimedia component
1. Data from polymers containing additives were discarded, as physical
effects on biota EEFs aim at specically quantifying the effects of the
physical attributes of polymers and not their chemical toxicity. This
work and Lavoie et al. (2021) assume that virgin polymers can best
represent these physical effects. Despite the effort to completely exclude
chemical toxicity effects in this approach, the authors acknowledge that
virgin polymers could contain residual, non-polymerized monomers
which could contribute to toxicity effects. The authors recognize that
this could mean that the impact category physical effects on biota en-
compasses both the physical attributes of polymers and the potential
effects of residual monomers.
The EEF was calculated from HC20
EC10eq
, following the same
methodology as Owsianiak et al. (Owsianiak et al., 2023). The unit of the
EEF is [PAF m
3
/kg
in compartment
] (PAF: Potentially Affected Fraction of
species). The 95% condence interval (CI) calculations followed the
methodology proposed by Lavoie et al. (2021), based on the work of
Payet (2004). The EEF calculations can be found in Multimedia
E. Corella-Puertas et al.
Journal of Cleaner Production 418 (2023) 138197
3
component 2.
2.2. Fate factors
2.2.1. Marine water
The fate factors (FFs) proposed in this work build on and are an
update to the work published in Corella-Puertas et al. (2022). FFs are
developed for different polymers, sizes and shapes. For the sphere and
microber, the size refers to the diameter, whereas for the lm it refers
to its thickness. The FFs comprise fate mechanisms in a single sub-
compartment of the marine environment, which includes both surface
water and the water column. Ongoing work in the MarILCA working
group will cover other marine subcompartments (onshore beaches and
sediments), as well as make a distinction between continental and global
seawater – this goes beyond the scope of this work. Similar to Cor-
ella-Puertas et al. (2022), the FFs comprise the main microplastic fate
mechanisms of degradation and sedimentation. Other microplastic fate
mechanisms include otation, resuspension and advection (Hajjar, C.,
Bulle, C., Boulay, 2020), and will be addressed in future work. Note that
degradation addresses the deterioration of plastics via biological and
chemical pathways – which can lead to dissociation into dissolved or-
ganics and ultimately achieve the conversion of organic constituents
into CO
2
; whereas fragmentation is dened as the breaking of plastics
into smaller plastic particles via mechanical pathways (Grobert et al.,
2020; Pfohl et al., 2022; Woods et al., 2021; Zhang et al., 2021). How-
ever, it is challenging to distinguish these two processes experimentally,
and an overall “mass loss” comprising mass losses through degradation
and fragmentation is often found in the literature (Chamas et al., 2020;
Maga et al., 2022). In this work, we assume that microplastics do not
undergo signicant mechanical abrasion in marine water, and thus
degradation processes dominate their mass loss. This hypothesis is
supported by a recent study proposing and testing a protocol to measure
the fragmentation and degradation rates of microplastics: Pfohl et al.
(2022) found that degradation rates were 1–2 orders of magnitude
higher than fragmentation rates.
Following the USEtox framework, this work assumes that the marine
water subcompartment is a homogenous box at steady-state. In this case,
FFs can be calculated by inverting the rate constant matrix K (Fantke
et al., 2018; Rosenbaum et al., 2007):
FF = − K−1(3)
In this work, the rate constant matrix only has one element – kwater,total
since there is only one marine water subcompartment – and thus delivers
one FF. Since the emission and receiving compartment are the same, this
FF can be interpreted as the total residence time of the microplastics in
the marine water subcompartment (Fantke et al., 2018). The rate con-
stant kwater,total comprises the removal mechanisms of degradation and
sedimentation:
kwater,total =kdegradation+ksedimentation (4)
Where kdegradation is the degradation rate constant and ksedimentation is the
sedimentation rate constant, both in [kg
mass loss
∕(kg
in compartment
×
year)].
In this work, degradation rates are polymer-, size- and shape-specic,
whereas sedimentation rates are only polymer-specic. The FFs have a
unit of [kg
in compartment
∕(kg
emitted
∕day)]. The calculations for obtaining
the FFs can be found in Multimedia component 3.
The next sections present the methodology for calculating the marine
microplastic degradation and sedimentation rates.
2.2.1.1. Degradation rates. Based on Chamas et al. (2020), this work
assumes that the degradation of microplastics occurs perpendicular to
their surface, with:
−dm(t)
dt=SA(t)kSSDR
ρ
(5)
Where dm/dt is the differential mass loss per unit time, SA is the surface
area in [cm
2
], k
SSDR
is the specic surface degradation rate (SSDR) in
[
μ
m/year] and
ρ
is the polymer density in [g/cm
3
]. k
SSDR
can be found in
the literature for different polymers (Chamas et al., 2020; Maga et al.,
2022). Whereas Corella-Puertas et al. (2022) proposed a simplied
approach in which the surface area of EPS microparticles was constant
over time, this work goes further and looks into how the surface area of
different microparticle types, shapes and sizes changes over time. Below,
the degradation rate is developed for microplastic beads – the derivation
of rates for microbers and lm fragments can be found in Multimedia
component 1.
In the approach proposed by Chamas et al. (2020), (micro)plastic
beads are perfectly spherical and thus SA(t) = 4
π
r2(t). This is a
simplication since microparticles are not necessarily smooth and might
have pores. This is conrmed by a literature review of the specic sur-
face area of different types of microplastics (see Multimedia component
3). This work proposes a surface area correction factor f
C
, which ac-
counts for polymer surface roughness and porosity. With this correction,
the polymer degradation can be described as:
−dm(t)
dt=4
π
r2(t)kSSDR
ρ
fCwith fC=SSAactual
SSAtheoretical
(6)
Where SSAactual is the actual specic surface area of microparticles
measured experimentally and SSAtheoretical is the specic surface area
calculated for the same mass of perfectly spherical microparticles (see
Multimedia component 3). This correction factor is particularly relevant
for the degradation rate of EPS. So far, the literature only proposes
specic surface degradation rates k
SSDR
for non-expanded PS (Chamas
et al., 2020; Maga et al., 2022) and none for EPS. Since EPS is expected
to have signicant porosity, representing EPS beads as smooth spheres
would be an oversimplication since experimental data conrms that
the SSA of EPS is up to an order of magnitude higher than the one of PS
(see Multimedia component 3). Therefore, the higher SSA and thus
faster degradation of EPS over PS can be accounted for with the surface
area correction factor.
For the other polymers studied in this work, specic surface degra-
dation rates k
SSDR
are directly available in the literature (see Multimedia
component 3). The experiments used to determine the specic surface
degradation rates k
SSDR
of these polymers were mostly performed on
macroplastics. This work assumes that the specic surface degradation
occurring on macroplastics will be similar to the one on microplastics. A
further assumption is that the specic surface roughness of these poly-
mers will be roughly the same for macro- and microplastics. For these
reasons, for all non-expanded polymers studied in this work, the surface
area correction factor is set to fC=1. This factor is varied by 10% in the
Monte-Carlo uncertainty calculations, to consider a potential variability
in the surface area of polymers (see Multimedia component 4).
Next, a degradation rate adapted for LCIA is developed. According to
the theorem of Heijungs (1995), a steady-state approach can be used to
model pulse emissions. At steady state, we can assume that there is a
constant inux of microparticles of radius r
max
and that degradation
occurs in a way that the total mass of the system M
total
remains constant.
Over time, the microparticles decrease in size, eventually become
nanoparticles and ultimately the particle mass is completely consumed.
Assuming that the degradation occurs perpendicular to the surface of the
particle at a constant rate, its radius would decrease as follows (Chamas
et al., 2020; Maga et al., 2022):
dr
dt = − kSSDR =const.(7)
This would create a uniform distribution of particles of different
radii, as illustrated in Figure A1 in Multimedia component 1. Based on
E. Corella-Puertas et al.
Journal of Cleaner Production 418 (2023) 138197
4
these assumptions, the mathematical derivation in Multimedia compo-
nent 1 delivers the following degradation rate constants for different
microplastic shapes:
kdegradation,sphere =4kSSDRfc
rmax
(8)
kdegradation,fiber =3kSSDRfc
rmax
(9)
kdegradation,film =2kSSDRfc
rmax
(10)
Where r
max
is the initial radius of spherical microbeads or microbers, or
half the initial thickness of microplastic lm fragments.
2.2.1.2. Sedimentation rates. Sedimentation rates are proposed for three
classes of polymers: low density (positively buoyant, <0.8 g/cm
3
such as
EPS), medium density (close to seawater density, 0.8–1.1 g/cm
3
such as
HDPE, LDPE, PP and PS) and high density (negatively buoyant, >1.1 g/
cm
3
such as PA, PLA, PHA, PET, PVC and TRWP). These rates are expert
estimates based on the approach proposed by Corella-Puertas et al.
(2022). Whereas low-density polymers are only expected to sink if
biofouling and/or aggregation occurs, high-density polymers are ex-
pected to sediment within a few months – even without biofouling or
aggregation (Corella-Puertas et al., 2022). For medium-density poly-
mers, the sedimentation rates are assumed to be between the
low-density and high-density ones. Due to the uncertainty in the sedi-
mentation behaviours, slow, medium, and fast sedimentation scenarios
are proposed for each polymer density category (see Multimedia
component 3).
2.2.2. Freshwater
According to the PLP guidelines, around 30% of microplastics
emitted into freshwater compartments will be trapped in freshwater
sediments, whereas the other 70% are estimated to be transferred into
marine compartments (Quantis and EA, 2020). This estimation was
based on eld data gathered in 40 sites across various river catchments
in England (Hurley et al., 2018). Recently, Domercq et al.’s (2022)
one-year simulations of the fate of microplastics in a generic river
showed that around 90% of high-density microplastics (PVC) accumu-
lated in the river’s sediments, whereas for medium-density microplastics
(PE, PA) less than 20% were trapped in the sediments. Domercq et al.’s
(2022) simulations included various fate mechanisms in the river –
degradation, settling, resuspension, and burial, among several others.
Based on these observations, the hypothesis that around 70%–80% of
microplastics emitted into freshwater are transferred to the ocean might
be reasonable for low and medium-density polymers but less realistic for
high-density polymers.
Relying on the hypothesis that the residence time of microplastics in
freshwater is much shorter than in marine water, the impacts in fresh-
water subcompartments are assumed to be negligible compared to the
impacts in marine water subcompartments. Based on this simplied
approach, this work proposes FFs for emissions to freshwater as 75% of
the FFs for emissions to marine water for low- and medium-density
polymers and 10% for high-density polymers. The resulting freshwater
FFs can be found in Multimedia component 3.
2.3. Characterization factors
2.3.1. Midpoint and endpoint CFs
Following the structure commonly used in LCIA (Bulle et al., 2019;
Hauschild and Huijbregts, 2015; Huijbregts et al., 2016), CFs for physical
effects of biota of microplastic emissions are developed at the midpoint
(problem) and endpoint (damage) levels. Midpoint CFs have units of
[PAF m
3
day/kg
emitted
], the same as the ecotoxicity impact category
(Bulle et al., 2019; Fantke et al., 2018). Endpoint CFs for damage to
ecosystem quality are calculated as:
Endpoint CF =Midpoint CF ∗SF
Water depth (11)
Where SF is the severity factor in [PDF/PAF] (PDF: Potentially Dis-
appeared Fraction of species) and the water depth is measured in [m]
(Bulle et al., 2019; Corella-Puertas et al., 2022; Fantke et al., 2018;
Jolliet et al., 2003). This water depth denes the size of the aquatic
compartment in which microplastics are potentially causing an effect on
ecosystems. The rationale for the choice of the SF, water depth and their
uncertainties can be found in Multimedia component 4. This CF con-
version is done to arrive at an endpoint in [PDF m
2
year/kg
emitted
] that is
compatible with other impact categories on ecosystem quality (Bulle
et al., 2019; Jolliet et al., 2003). Additionally, an optional conversion
step can be done to transform the CFs into units that are harmonized
with the upcoming Global Guidance for Life Cycle Impact Assessment
Indicators and Methods (GLAM) Phase 3 methodology (see Multimedia
component 1).
2.3.2. Uncertainty
A Monte Carlo analysis was performed to calculate the uncertainty of
the CFs. The parameters used in the calculations and rationale for the
choice of distributions can be found in Multimedia component 4. The
calculations were done in Python with 5000 iterations. Histograms and
Shapiro-Wilk tests (both on the logarithm of the CF results) conrmed a
log-normal distribution for most CFs – see the discussion in 3.3.3 for
more details. Therefore, for the recommended CFs, the geometric mean
of Monte Carlo distribution was chosen and 95% CI were calculated
assuming log-normality. An exemplary Python script for calculating the
CFs and their uncertainties is available in Multimedia component 1.
Although this exemplary script contains specic calculations for EPS, it
can be adjusted with the parameters in Multimedia component 4 to
reproduce the results of different polymers.
2.4. Case studies
The methodology developed in this work was tested on two food
packaging case studies in the context of the UNEP report “Supermarket
food packaging and its alternatives: Recommendations from life cycle
assessments” (UNEP, 2022). The rst case study compared the potential
impacts of single-use bags for fresh-cut vegetables, made of PP or PLA
(Vigil et al., 2020), whereas the second case study compared single-use
cardboard boxes versus reusable plastic (HDPE or PP) crates for vege-
table and fruit transportation (Abej´
on et al., 2020). Both case studies
originally covered several impact categories, but did not include the
potential impacts of plastic emissions. This work added potential phys-
ical effects on biota impacts at the endpoint level for both studies,
allowing to compare the magnitude of potential impacts of microplastic
emissions to other impact categories.
2.4.1. Life cycle inventory
Two sources of marine (micro)plastic emissions were identied and
quantied following the PLP guidelines (Quantis and EA, 2020): 1. The
leakage of primary microplastics (pellets) at the polymer production
stage; 2. The leakage of macroplastics at the end-of-life stage, which can
fragment into secondary microplastics. The leakage of TRWP from
transportation was out of the scope of this work. Since the region where
plastics are used and discarded is expected to have an inuence on the
macroplastic leakage, the calculations were not only done for the spe-
cic country of the original case study, but also for the following sce-
narios: high-income country, upper-middle-income country,
lower-middle-income country, low-income country. For the fragmenta-
tion of macroplastics into microplastics, scenarios of 10%, 50%, 100%
fragmentation were tested (similar to Corella-Puertas et al. (2022)). The
calculations of the plastic inventory can be found in Multimedia
E. Corella-Puertas et al.
Journal of Cleaner Production 418 (2023) 138197
5
component 6 and Multimedia component 7.
2.4.2. Life cycle impact assessment
Endpoint CFs for physical effects on biota were applied to the marine
microplastic emissions. Secondary microplastics were approximated as
spherical particles with a diameter of 1000
μ
m for the crates, and
microplastic lm fragments with a thickness of 100
μ
m for the single-use
bags. Note that a preliminary version of the CFs was used in the UNEP
report since the current CFs were not ready at the time of the UNEP
report submission. The preliminary versus nal versions of the CFs and
physical effects on biota impacts are shown in Multimedia component 6
and 7 and discussed in Multimedia component 1.
3. Results and discussion
3.1. Exposure and effect factor
Within the searched publications, 10 suitable, chronic EC50 or LC50
data points were found (7 species and 5 phyla). Yet, as USEtox recom-
mendations regarding the EEF calculations recently changed (see 2.1), a
new search focused on EC10 data but found only 3 data points, which
was not sufcient to generate a recommendable EEF. Joining them with
extrapolated data from Lavoie et al.’s (2021) work (LOEC, NOEC and
EC50) delivered an updated EEF of 1067.5 (358.1–3182.1) PAF
m
3
/kg
in compartment
(see calculations and details about the selection of
data in Multimedia component 2). As a comparison, the original EEF was
72.9 (7.2–736.4) PAF m
3
/kg
in compartment
(Lavoie et al., 2021). Because
of the increase in data points, the uncertainty range of the updated EEF
is smaller (one order of magnitude) than the one of the former EEF (two
orders of magnitude). Despite the new EEF being more than 10 times the
value of the former EEF, the severity factor (see Multimedia component
4) will rebalance things out at the damage level. In fact, multiplying the
EEF and SF values now delivers an endpoint of 26.7 PDF m
3
/kg
in
compartment
instead of the previous 36.5 PDF m
3
/kg
in compartment
(Cor-
ella-Puertas et al., 2022). As the endpoint remains within the same order
of magnitude, this supports its robustness. Furthermore, this example
shows that, when EEF are calculated based on different hazardous
concentrations (HC20
EC10eq
in this work vs. HC50
EC50eq
in Lavoie et al.
(2021)), it is preferable to compare the endpoint level instead of the
midpoint level.
Although Salieri et al. (2021) and Zhao and You (2022) only reported
EEF at the midpoint level for microplastic emissions to freshwater, a SF
can be applied to get to an endpoint level (see Multimedia component
4). Salieri et al.’s (2021) EEF of 22.6 PAF m
3
/kg
in compartment
is based on
HC50
EC50eq
and would become 11.3 PDF m
3
/kg
in compartment
at the
endpoint, which is in the same order of magnitude as this work’s result.
Note that Salieri et al.’s (2021) did not report an uncertainty for their
EEF, which covers less data points than this work. Zhao and You (2022)
proposed freshwater EEFs for different microplastic emissions, which
spanned over several orders of magnitude depending on the type of
polymer and size (119–25380 PAF*m
3
/kg
in compartment
based on
HC20
NOEC
). Since NOEC and EC10 data are expected to be close, the
same SF as this work can be applied in a simplied approach, getting
3–635 PDF m
3
/kg
in compartment
at the endpoint for different microplastic
emissions. Note that, for all polymers and sizes – except PS <100
μ
m –
the data used by Zhao and You (2022) for calculating their EEFs did not
cover three trophic levels. Furthermore, their study did not exclude
toxicity data of plastics containing additives, and thus their data covered
both physical effects and chemical (additive) effects. Despite the dif-
ferences in the methodologies, this work’s endpoint lies within the range
of Zhao and You’s (2022).
All in all, the EEF proposed in the present work is generic for all types
of micro- and nanoplastic emissions since there was not enough specic
data to calculate polymer, size or shape-specic EEFs.
3.2. Fate factors
3.2.1. Marine water
FFs were obtained for various polymers, shapes and sizes, as well as
different degradation and sedimentation scenarios (see Multimedia
component 3). In this case, the FFs can be interpreted as the potential
residence time of microplastics in marine water (see 2.2.1). Overall, for
different types of microplastics, the average residence time spans be-
tween 15 days and 3600 years (scenario of medium degradation and
medium sedimentation).
First, the inuence of polymer density and size will be examined in
the example of spherical microplastics. Since the size of microplastics
ranges between 1 and 5000
μ
m (Woods et al., 2021), 5000
μ
m will be
addressed as “large” microplastics and 1
μ
m as “small” microplastics. To
discuss this inuence of polymer density on the fate, PLA was chosen as
an exemplary high-density polymer, HDPE as a medium-density poly-
mer and EPS as a low-density polymer (see Fig. 1). Note that, within
each polymer density category (see 2.2.1.2), the FFs were in the same
order of magnitude for same-sized particles. There were exceptions for
small-sized particles (10
μ
m and 1
μ
m): 1) the FFs of medium-density
particles varied up to two orders of magnitude for different polymers
(see Figure A2 in Multimedia component 1) and 2) the FFs of PHA were
up to three orders of magnitude smaller than FFs of other high-density
polymers (see Multimedia component 3). This variability is linked to
the different degradation rates of different polymers – the degradation
rate of PHA is particularly high (see Multimedia component 3). As
explained below, degradation mechanisms become particularly impor-
tant for small-sized particles with high specic surface areas.
Overall, higher polymer density is linked with smaller FFs and thus
shorter residence time in the marine water subcompartment. This in-
dicates that sedimentation mechanisms largely inuence marine fate
(see 2.2.1). For large microplastics of high-density polymer (see PLA of
5000
μ
m in Fig. 1), sedimentation occurs so quickly that the residence
time is too small for degradation to have a signicant inuence. Small
microplastics of high density behave similarly, except for the fast
degradation scenario (see PLA of 1
μ
m in Fig. 1). Small microplastics
have a much higher specic surface area and thus degrade much faster
than larger microplastics (see Eqs. (5) and (6)). Therefore, the combi-
nation of a high specic surface area with a high specic surface
degradation rate leads to a signicant inuence of the degradation rate
on the fate of small, high-density polymers.
For medium-density polymers, sedimentation occurs slower than for
high-density polymers. Therefore, in the fast degradation scenario, even
large microparticles have sufcient time to undergo signicant degra-
dation before being removed from the water subcompartment via sedi-
mentation (see HDPE of 5000
μ
m in Fig. 1). For small microplastics with
larger specic surface areas, the inuence of degradation is signicant
across all degradation scenarios (see HDPE of 1
μ
m in Fig. 1).
Low-density polymers have the highest uncertainty, due to large
sedimentation and degradation uncertainties. The overall uncertainty is
four to ve orders of magnitude between the best- and worst-case sce-
narios, for any microparticle size. The degradation uncertainty is related
to the lack of experimental data quantifying the degradation of EPS (see
hypotheses in 2.2.1.1). The sedimentation of low-density polymers is
highly uncertain since these polymers are positively buoyant and can
only sediment through an increase in density, such as through biofouling
or aggregation. Ongoing work is modelling the sedimentation rates of
(low-density) microplastics and is expected to reduce sedimentation
uncertainty.
Regarding the inuence of the shape on the FFs, similar trends can be
observed for microbeads/spheres, microbers/cylinders and lm frag-
ments of the same sizes and polymer densities (see Fig. 1, and
Figures A.3 and A.4 in Multimedia component 1). The only difference is
that degradation occurs fastest for spherical microplastics and slowest
for microplastic lm fragments – linked to Eqs. 8–10. Although this
change in degradation rate affects the FFs, this inuence is minor since
E. Corella-Puertas et al.
Journal of Cleaner Production 418 (2023) 138197
6
the FFs of different shapes remain around the same order of magnitude.
Similar to this work, Maga et al. (2022) proposed FFs for aquatic
(marine and freshwater) microplastics emissions of different polymer
types, shapes and sizes. Their work could not be used directly, with the
following differences found between the two approaches:
1. Sedimentation: Maga et al. (2022)’s FFs include a transfer coefcient
to a different environmental compartment (e.g. transfer from marine
water to marine sediments), however, this is not a rate but a nal
transfer percentage. For high-density polymers, this means that their
FF and residence time in water compartments are zero – in this work,
these FFs are not zero and are calculated from the degradation and
sedimentation rates. In Maga et al. (2022)’s work, low-density
polymers never sediment. In contrast, sedimentation mechanisms
linked to biofouling and aggregation are considered in this work,
thus covering scenarios in which positively buoyant polymers can
sediment.
2. Degradation: Maga et al. (2022)’s model was derived from the
degradation of a single microplastic object (either bead, ber or
lm). Their model, combined with experimental data, was useful for
obtaining specic surface degradation rates for different polymer
types – which were implemented in this work (see Multimedia
component 3). The present work goes one step further and proposes a
steady-state approach for modelling pulsed microplastic emissions,
aligned with USEtox methodologies (Fantke et al., 2018).
3. Units: The FFs proposed by Maga et al. (2022) are normalized by a
ctional reference emission with a residence time of one year and
have the unit of [kg
PPe
∕kg
emitted
] (PPE: plastic pollution equivalent).
Whereas this might be useful to compare the fate of different poly-
mers, these FF cannot be directly combined with existing EEF to form
CFs.
4. Time horizon: Maga et al. (2022) reported FFs for different time
horizons (100, 500, 1000 years). This work proposes steady-state FFs
(integrated to innite time). Other time horizons were outside the
scope of this work but could be developed in the future.
3.2.2. Freshwater
FFs for freshwater emissions of microplastics of different polymer
types, shapes and sizes are available in Multimedia component 3. These
freshwater FFs (and ultimately the freshwater CFs) link freshwater
microplastic emissions with their impacts in marine compartments.
Until a more detailed fate model for freshwater emissions is developed,
using these factors assumes that the residence time in freshwater is much
shorter than residence time in marine water (see 2.2.2) – which may be
true in some basins and less accurate in others.
In order to test this hypothesis, the proposed marine FFs are
compared to Salieri et al.’s (2021) simplied freshwater FFs, which are
based on worst-case hypotheses and account for the degradation of the
microplastics only in freshwater. In this work, most of the medium-case
marine FFs are 2–3 orders of magnitude larger than Salieri et al.’s (2021)
worst-case freshwater FFs, supporting the hypothesis that the residence
time in freshwater is negligible compared to the one in marine water.
There are however two exceptions:
Fig. 1. Fate factors of marine microplastic emissions,
for the impact category of physical effects on biota.
This example shows spherical microplastics of
different polymer types (PLA: high density, HDPE:
medium density, EPS: low density) and sizes (5000
μ
m, 1
μ
m diameters) – results of microbers and
microplastic lm fragments can be found in Multi-
media component 1. Different degradation and sedi-
mentation scenarios are tested. Worst-case scenario:
slow degradation and slow sedimentation. Best-case
scenario: fast degradation and fast sedimentation.
E. Corella-Puertas et al.
Journal of Cleaner Production 418 (2023) 138197
7
1. For the smallest microplastics (1
μ
m), this work’s medium-case
marine FFs are in the same order of magnitude as Salieri et al.’s
(2021) worst-case freshwater FFs. This can be explained by the fact
that Salieri et al.’s (2021) degradation rates are based on half-lives of
macroplastics (Chamas et al., 2020), for which the specic surface
area is several orders of magnitude smaller than the one of micro-
plastics and thus are expected to degrade much slower. Therefore,
the comparison between Salieri et al.’s (2021) FFs and this work’s
FFs might be more appropriate for large microplastics that are close
in size to macroplastics.
2. High-density polymers have a shorter residence time in this work’s
marine FFs than in Salieri et al.’s (2021) freshwater FFs. This is not
surprising, since section 3.2.1 showed that sedimentation mecha-
nisms dominate for high-density polymers, and Salieri et al.’s (2021)
did not include any sedimentation mechanisms.
3.3. Characterization factors
3.3.1. Midpoint and endpoint CFs
Midpoint and endpoint CFs for physical effects on biota and their 95%
CI are developed for different polymers, shapes, sizes and emission lo-
cations (Fig. 2, Fig. 3 and Multimedia component 5). All CFs are also
available in units that are harmonized with the upcoming GLAM Phase 3
methodology (Multimedia component 1 and Multimedia component 8) –
the conversion of units does not alter the trends in physical effects on biota
CFs and thus will not be further discussed here.
A literature review of different polymers and shapes found no TRWP
and EPS in the shape of lms, and thus these CFs are excluded. The
average marine endpoint CFs range between 2.81E-04 and 1.18E-02
PDF m
2
year/kg
emitted
for high-density polymers (2.81E-05 to 1.18E-
03 PDF m
2
year/kg
emitted
for freshwater), between 2.70E-02 and
1.48E+01 PDF m
2
year/kg
emitted
for medium-density polymers (2.02E-
02 to 1.11E+01 PDF m
2
year/kg
emitted
for freshwater), and between
7.79E-02 and 4.13E+02 PDF m
2
year/kg
emitted
for low-density polymers
(5.85E-02 to 3.10E+02 PDF m
2
year/kg
emitted
for freshwater). Fresh-
water CFs show the same trends as marine CFs, since the former are
calculated as a percentage of the latter (see Fig. 2, and Figure A.5 in
Multimedia component 1). The only dissimilarity is that the difference
between high-density and medium-density CFs is more pronounced for
freshwater CFs, due to the choice in calculations described in 2.2.2.
Overall, the following observations for marine CFs of different polymers,
shapes and sizes are also valid for freshwater CFs. Furthermore,
midpoint and endpoint CFs show similar trends since the midpoint-to-
endpoint conversion is identical for all microplastic types (see Figs. 2
and 3).
A visual comparison of all endpoint CFs for marine microplastic
emissions is shown in Fig. 2. Since there is only one EEF for all micro-
plastic types at this point (see 3.1), the differences in CFs can be directly
attributed to the fate of the microplastics. Therefore, the observations
made for the FFs (see 3.2.1) are also valid for the CFs. Summarizing, the
shape of the microplastics only has a small inuence on the CFs, whereas
polymer density and size play relevant roles (see Fig. 2). High-density
polymers (PA, PHA, PLA, PET, PVC, TRWP) have the lowest CFs and
show little variability – the exception is PHA for the sizes of 1 and 10
μ
m,
which have lower CFs linked with remarkably quick degradation (see
3.2.1). For high-density polymers other than PHA, sedimentation dom-
inates the fate, leading to short residence times in the marine water
compartment and thus low CFs. Because degradation plays a secondary
role for these high-density polymers, the particle size does not greatly
affect the CFs – in this work’s model, size inuences degradation but not
sedimentation. The CFs of medium-density polymers (HDPE, LDPE, PP,
PS) are similar for microplastics of 5000
μ
m and 1000
μ
m sizes but show
higher variability for smaller sizes. For smaller microplastics with larger
specic surface areas, degradation becomes more relevant (see 3.2). The
Fig. 2. Endpoint characterization factors (CFs) for physical effects on biota of
marine microplastic emissions, for different polymers, shapes and sizes. The
error bars represent the 95% condence interval. On the y-axis, the polymers
are sorted according to their density (EPS has the lowest density and TRWP the
highest). The microplastic size varies between 5000, 1000, 100, 10 and 1
μ
m,
and is represented by the marker size (e.g. smallest marker =1
μ
m micro-
plastic). For a sphere/microbead and the cylinder/microber, the size refers to
the diameter, whereas for the lm fragment it refers to its thickness. PDF:
Potentially Disappeared Fraction of species.
Fig. 3. Midpoint characterization factors (CFs) for physical effects on biota of
marine microplastic emissions, for different polymers, shapes and sizes. The
error bars represent the 95% condence interval. On the y-axis, the polymers
are sorted according to their density (EPS has the lowest density and TRWP the
highest). The microplastic size varies between 5000, 1000, 100, 10 and 1
μ
m,
and is represented by the marker size (e.g. smallest marker =1
μ
m micro-
plastic). For a sphere/microbead and the cylinder/microber, the size refers to
the diameter, whereas for the lm fragment it refers to its thickness. PDF:
Potentially Disappeared Fraction of species.
E. Corella-Puertas et al.
Journal of Cleaner Production 418 (2023) 138197
8
degradation rates of different polymers differ, having an inuence on
the CFs of smaller, medium-density polymers. The low-density polymer
(EPS) shows similar behaviour to the medium-density polymers, with
the difference that there is a higher uncertainty for EPS and the CFs span
over more orders of magnitude. This is linked to both high degradation
and sedimentation uncertainties (see 3.2).
3.3.2. Comparison to marine ecotoxicity CFs
To assess the magnitude of the potential damage to ecosystem
quality from physical effects on biota, the marine endpoint CFs obtained
in this work are compared to endpoint CFs for marine ecotoxicity from
ImpactWorld+(Bulle et al., 2019) (Fig. 4). For visual clarity, the
microplastics represented in Fig. 4 are spheres/microbeads. Neverthe-
less, the observations would not change if microbers or microplastic
lm fragments were represented – as seen in Fig. 2, the CFs of the three
studied shapes are around the same order of magnitude for the same
polymers and sizes.
Overall, the large variability of physical effects on biota CFs – linked
with the polymer and size – makes them comparable to a wide range of
USEtox substances. For high-density polymers, the physical effects on
biota CFs are in the range of the lowest marine ecotoxicity CFs. Only
around 3% of the USEtox substances have lower potential damage to
ecosystem quality than most high-density polymers (PHA of 10
μ
m or
smaller size are even lower, at around 1% of USEtox substances). This is
close to the observations of Corella-Puertas et al. (2022) for TRWP (5%
of the USEtox substances). Medium- and low-density polymers are
spread over a large range. The highest physical effects on biota CF is the
largest EPS microbead (5000
μ
m) – its average physical effects on biota CF
is higher than 83% of the marine ecotoxicity CFs of USEtox substances.
As a comparison to commonly known substances, the average EPS CFs of
1000 and 5000
μ
m microparticles show smaller potential damage to
ecosystem quality than aldrin but higher than nicotine. The average EPS
CF reported by Corella-Puertas et al. (2022) (2.85E+02 PDF m
2
year/kg
emitted
) is between the average EPS CFs for 1000
μ
m (7.35E+01
PDF m
2
year/kg
emitted
) and 5000
μ
m (3.22E+02 PDF m
2
year/kg
emitted
)
microbeads in this work. The former was based on a simplied degra-
dation methodology which could not be adjusted for size or shape, but
had as input the experimental specic surface area of 830–2000
μ
m EPS
microparticles. The observations for TRWP and EPS indicate that the
preliminary, simplied methodology of Corella-Puertas et al. (2022) was
a reasonable rst approach. Now that the updated, polymer-, size-, and
shape-specic CFs are available, these are recommended over the ones
of Corella-Puertas et al. (2022).
3.3.3. Parameter uncertainty
In most LCIA impact categories, information on the uncertainty of
CFs is lacking – this limits the inclusion of LCIA uncertainty within LCA
studies (UNEP, 2019). Phase 2 of the Global Guidance for Life Cycle
Impact Assessment Indicators and Methods (GLAM) strongly recom-
mends method developers share sufcient underlying uncertainty in-
formation (UNEP, 2019). To provide uncertainty information for the
impact category of physical effects on biota, the CF uncertainty was
calculated with a Monte Carlo analysis, as described in 2.4 and Multi-
media component 4. This uncertainty addresses parameter uncertainty
(HC20
EC10eq
, specic surface degradation rates, surface area correction
factors, sedimentation rates, water depth, severity factor).
The CF uncertainty (95% CI) tends to increase with decreasing
polymer density (see Fig. 3 and Multimedia component 5): For high-
density polymers at the midpoint, the uncertainty spans over one order
of magnitude for high-density polymers (except for PHA, for which
uncertainty can reach up to two orders of magnitude, linked with the
uncertainty of its degradability, see Multimedia component 3). For
medium-density polymers at the midpoint, uncertainty ranges between
one to three orders of magnitude (three orders of magnitude are only
reached by smaller particles of 10 and 1
μ
m, linked with degradation
uncertainty). For low-density polymers at the midpoint, uncertainty
reaches two to three orders of magnitude for low-density polymers.
The uncertainty of endpoint CFs is about one order of magnitude higher
than for midpoint CFs (see Fig. 2).
The proposed CFs are based on the geometric mean of the 5000
values generated by Monte Carlo, based on the assumption of a log-
normal distribution. 71% of the all CFs pass the Shapiro-Wilk log-
normality test – for the endpoint CFs, the success rate is higher at 85%.
Furthermore, two methods of calculation of the 95% CI are compared:
The rst used the geometric standard deviation and assumed log-
normality, whereas the second is a manual selection of the values
125th and 4975th values generated by Monte Carlo (see Multimedia
component 4). The methods show reasonable agreement (<10% dif-
ference) for 73% of all CFs. Since most CFs match a log-normal distri-
bution, it was deemed reasonable to select the average CFs based on the
geometric mean and calculate their 95% CI from the geometric standard
deviation.
3.4. Case studies
The LCIA methodology and CFs developed in this work were applied
to two food packaging case studies (see 2.4). The results are presented
and discussed in the report “Supermarket food packaging and its
Fig. 4. Endpoint characterization factors (CFs) for
physical effects on biota of marine (spherical) micro-
plastic emissions compared to endpoint CFs for ma-
rine ecotoxicity for all organic chemicals represented
in USEtox (total of 2454 substances, examples of
commonly known chemicals indicated by the ar-
rows). For physical effects on biota CFs, the error bars
represent the 95% condence interval. The micro-
plastic size varies between 5000, 1000, 100, 10 and 1
μ
m, and is represented by the marker size (e.g.
smallest marker =1
μ
m diameter). On the left edge of
the x-axis, all markers of high-density polymers (PET,
PLA, PVC, TRWP) overlap. 2,3,7,8 TCDD: 2,3,7,8-
Tetrachlorodibenzo-p-dioxin.
E. Corella-Puertas et al.
Journal of Cleaner Production 418 (2023) 138197
9
alternatives: Recommendations from life cycle assessments” (UNEP,
2022). The summarized results of the case studies can be found in
Multimedia component 1, F and G. The fact that a preliminary version of
the CFs was used in the UNEP report does not signicantly affect the case
studies’ conclusions (see Multimedia component 1). In a nutshell, these
are the key insights from the case studies:
1. The location of plastic use and end-of-life can inuence physical ef-
fects on biota impacts by up to two orders of magnitude. This is linked
with the plastic inventory, and independent of the CFs – since the
latter are not regionalized at this point.
2. For this work’s case studies, physical effects on biota impacts from
microplastic emissions remain smaller than climate change impacts
on ecosystem quality – even in the worst-case scenario of high plastic
leakage due to inadequate waste management practices. In a sepa-
rate case study on to-go food containers used and discarded in
Canada (relatively low leakage), EPS results have high uncertainty,
and physical effects on biota impacts could be potentially higher than
climate change impacts in a worst-case CF scenario (Corella-Puertas
et al., 2022; UNEP, 2022). Although for average-case EPS CFs phys-
ical effects on biota impacts remain smaller than climate change im-
pacts, this might change if the food containers were used in a region
with higher plastic leakage.
3. The magnitude of physical effects on biota impacts is polymer
dependent, with the trend PLA <HDPE, PP <EPS. This is in
agreement with the CF trends for polymers of different densities (see
3.3.1 and 3.3.4).
4. In some cases, the concluding performance of product alternatives
could be reversed by the addition of physical effects on biota impacts.
For the example of to-go food containers, the EPS container performs
better than its compostable alternatives without the inclusion of
microplastic impacts. Once physical effects on biota impacts of
microplastic emissions are added, the EPS container becomes com-
parable or worse (depending on the scenario) than the wood pulp
container (Corella-Puertas et al., 2022; UNEP, 2022).
The application of physical effects on biota CFs to case studies is an
example of how this work’s methodology can be implemented within
LCA. Hopefully, this work will contribute to putting the impacts of
microplastic emissions in perspective of other impact categories and
support decision-making for plastic pollution solutions that avoid
burden-shifting.
3.5. Recommendations for LCA practitioners
To simplify the application of physical effects on biota CFs for
microplastic emissions, default CFs are proposed (see Table 1). These
default CFs are meant to facilitate the work of LCA practitioners, who
might not have access to an exact description of microplastics emitted.
In case LCA practitioners have detailed information on the microplastic
Table 1
Default physical effects on biota characterization factors (CFs) of aquatic microplastic emissions of different polymer densities (low, medium, high) and shapes
(abbreviated in the table: microbeads/spheres/fragments of unspecied shape, microbers/cylinders, microplastic lm fragments). More specic CFs for different
polymers, shapes and sizes, as well as the CF uncertainties, can be found in Multimedia component 5.
Shape Polymer type Default size Midpoint CFs (PAF*m
3
*day/kg
emitted
) Endpoint CFs (PDF*m
2
*year/kg
emitted
)
Marine Freshwater Marine Freshwater
Microplastic beads Low-density <0.8 g/cm
3
1000
μ
m 1.08E+08 8.12E+07 7.35E+01 5.51E+01
Medium-density 0.8–1.1 g/cm
3
2.12E+07 1.59E+07 1.44E+01 1.08E+01
High-density >1.1 g/cm
3
1.74E+04 1.74E+03 1.18E-02 1.18E-03
Plastic microbers Low-density <0.8 g/cm3 10
μ
m 1.53E+06 1.15E+06 1.04E+00 7.79E-01
Medium-density 0.8–1.1 g/cm
3
8.90E+06 6.68E+06 6.05E+00 4.53E+00
High-density >1.1 g/cm
3
1.73E+04 1.73E+03 1.17E-02 1.17E-03
Microplastic lm fragments Medium-density 0.8–1.1 g/cm
3
100
μ
m 1.92E+07 1.44E+07 1.31E+01 9.80E+00
High-density >1.1 g/cm3 1.74E+04 1.74E+03 1.18E-02 1.18E-03
Microplastic beads EPS 1000
μ
m 1.08E+08 8.12E+07 7.35E+01 5.51E+01
HDPE 1.93E+07 1.45E+07 1.31E+01 9.83E+00
LDPE 1.75E+07 1.31E+07 1.19E+01 8.93E+00
PA (Nylon) 1.73E+04 1.73E+03 1.18E-02 1.18E-03
PET 1.74E+04 1.74E+03 1.18E-02 1.18E-03
PHA 1.63E+04 1.63E+03 1.11E-02 1.11E-03
PLA 1.74E+04 1.74E+03 1.18E-02 1.18E-03
PP 1.35E+07 1.01E+07 9.14E+00 6.86E+00
PS 2.12E+07 1.59E+07 1.44E+01 1.08E+01
PVC 1.71E+04 1.71E+03 1.16E-02 1.16E-03
TRWP 1.73E+04 1.73E+03 1.18E-02 1.18E-03
Plastic microbers EPS 10
μ
m 1.53E+06 1.15E+06 1.04E+00 7.79E-01
HDPE 3.85E+06 2.89E+06 2.61E+00 1.96E+00
LDPE 2.41E+06 1.81E+06 1.64E+00 1.23E+00
PA (Nylon) 1.44E+04 1.44E+03 9.78E-03 9.78E-04
PET 1.66E+04 1.66E+03 1.12E-02 1.12E-03
PHA 3.97E+03 3.97E+02 2.70E-03 2.70E-04
PLA 1.73E+04 1.73E+03 1.17E-02 1.17E-03
PP 5.16E+05 3.87E+05 3.50E-01 2.63E-01
PS 8.90E+06 6.68E+06 6.05E+00 4.53E+00
PVC 1.71E+04 1.71E+03 1.16E-02 1.16E-03
TRWP 1.70E+04 1.70E+03 1.15E-02 1.15E-03
Microplastic lm fragments HDPE 100
μ
m 1.45E+07 1.09E+07 9.84E+00 7.38E+00
LDPE 1.16E+07 8.68E+06 7.86E+00 5.89E+00
PA (Nylon) 1.71E+04 1.71E+03 1.16E-02 1.16E-03
PET 1.73E+04 1.73E+03 1.17E-02 1.17E-03
PHA 1.35E+04 1.35E+03 9.17E-03 9.17E-04
PLA 1.74E+04 1.74E+03 1.18E-02 1.18E-03
PP 5.60E+06 4.20E+06 3.80E+00 2.85E+00
PS 1.92E+07 1.44E+07 1.31E+01 9.80E+00
PVC 1.71E+04 1.71E+03 1.16E-02 1.16E-03
E. Corella-Puertas et al.
Journal of Cleaner Production 418 (2023) 138197
10
emissions, the use of specic CFs is recommended (see Multimedia
component 5).
Based on the commonness of (micro)plastic sizes found in the envi-
ronment and industry, default sizes are proposed for spheres/microbe-
ads/fragments of unspecied shape (1000
μ
m diameter) (Castillo et al.,
2016; Firdaus et al., 2020; Mohamed Nor and Obbard, 2014; Zhang
et al., 2020), cylinders/microbers (10
μ
m diameter) (Dreillard et al.,
2022; Geyer et al., 2022; Henry et al., 2018; Kim et al., 2021; Pinlova
and Nowack, 2023) and microplastic lm fragments (100
μ
m thickness)
(Vismaya et al., 2021) (uspackagingandwrapping.com, 1bagatatime.
com, plasticbags.com.au, all accessed in November 2022). In case of
hesitation between two particle size categories (e.g. 100
μ
m vs 1000
μ
m
for microbeads), the larger one was chosen since small microplastics in
the environment might come from larger microplastics that degraded
over time. Furthermore, this is a more conservative approach since
larger microplastics take longer to degrade than smaller ones and thus
are linked with higher potential impacts (higher CFs). As polymers
within the same density categories tend to behave similarly (see 3.3.1),
default CFs based on polymer density are suggested. These more generic
CFs can be applied to microplastic emissions of polymers other than the
ones studied in this work.
This work’s CFs can be applied to both primary and secondary
microplastics (see 2.4). For secondary microplastics, to date, there is no
ofcial fragmentation model quantifying the fragmentation of macro-
plastics into microplastics. Nevertheless, LCA practitioners might still
want to assess the potential impacts of secondary microplastics, since
these are probably more abundant than primary microplastics in the
environment (Bergmann et al., 2015; Hidalgo-Ruz et al., 2012; Koel-
mans et al., 2022). In this case, different fragmentation scenarios can be
tested (see examples in 2.4.1 and Corella-Puertas et al. (2022)). For
fragments from wrappings, bags and other lm-like macroplastics,
microplastic lm fragment CFs are recommended. For fragments from
textiles, ropes, and other ber-containing macroplastics, microber CFs
are advised. For foam fragments and microbeads, spherical microplastic
CFs can be used. In a recent study, Alkema et al. (2022) observed that
plastic fragmentation was associated with particle rounding. Therefore,
for unspecied microplastic fragments, the application of spherical
microplastic CFs is suggested. In any case, CFs are similar for different
shapes, and thus the choice of microplastic shape is expected to have a
minor inuence on the LCIA results.
While CFs for nanoplastic emissions are not available, CFs for 1
μ
m
microplastic size are proposed as interim CFs. 1
μ
m is the dened
boundary between micro- and nanoplastics (Woods et al., 2021). For the
fate of nanoplastics, the choice of 1
μ
m is a conservative approach, since
smaller nanoparticles will degrade faster. The sedimentation of
micro-vs. nanoplastics remains to be studied. Regarding the effects, this
work’s EEF is applicable to both micro- and nanoplastics (see 3.1).
4. Conclusion and outlook
This work proposes a methodology to assess the impacts of micro-
and nanoplastic emissions on aquatic ecosystems, through the new LCA
impact category of physical effects on biota. The fate methodology of
Corella-Puertas et al. (2022) and the exposure-effect methodology of
Lavoie et al. (2021) were updated to deliver CFs for eleven polymers,
three shapes and ve sizes. The CFs were tested in two case studies
within the UNEP report “Supermarket food packaging and its alterna-
tives: Recommendations from life cycle assessments” (UNEP, 2022),
conrming their applicability to assess the relative importance of phys-
ical effects on biota impacts compared to other impact categories.
A limitation of the fate factor is that sedimentation only depends on
the polymer density, and not on the size or shape. This approach might
describe well high-density polymers, which are expected to sediment
independently of their size and shape. However, medium-to low-density
polymers depend on biofouling and aggregation to enable sedimentation
– these processes might be size and shape-dependent and are being
studied in ongoing MarILCA work. Further ongoing MarILCA work aims
at integrating the transport of microplastics to global waters into the fate
of marine microplastic emissions. Moreover, the fate of micro- and
nanoplastic emissions into the freshwater compartment was simplied
and should be developed further in future work.
Due to the scope of this work, the CFs quantify potential impacts in
the water surface and water column, but not in the sediments. The water
subcompartments are richer in species than the sediment subcompart-
ment, and research has focused mostly on the former. To date, there is a
lack of data on the effects of micro- and nanoplastics in sediments.
However, as more data becomes available, it would be relevant to
develop an EEF for the sediment subcompartment. In order to apply such
an EEF, a distinction between the fate of microplastics in sediments and
deep burial would be required – in the latter, microplastics are assumed
to remain there, with no exposure to organisms (Van Colen et al., 2021).
Furthermore, polymer degradation rates for the sediment subcompart-
ment would need to be developed.
Finally, the CFs presented in this work only cover part of the
complexity of the potential impacts of plastic emissions. Further impact
categories related to plastic emissions are currently developed by Mar-
ILCA (physical effects on biota of macroplastics, invasive species, eco-
toxicity of plastic additives, etc.) and will allow for a more
comprehensive assessment of the plastic impacts on the environment.
CRediT authorship contribution statement
Elena Corella-Puertas: Conceptualization, Methodology, Investi-
gation, Data curation, Formal analysis, Visualization, Validation,
Writing – original draft, Writing – review & editing, Funding acquisition.
Carla Hajjar: Methodology, Writing – review & editing. J´
erˆ
ome Lav-
oie: Methodology, Data curation, Formal analysis, Writing – review &
editing. Anne-Marie Boulay: Supervision, Conceptualization, Method-
ology, Writing – review & editing, Funding acquisition, Resources.
Declaration of competing interest
The authors declare that they have no known competing nancial
interests or personal relationships that could have appeared to inuence
the work reported in this paper.
Data availability
Data available in article supporting information.
Acknowledgements
The authors wish to acknowledge Manuele Margni and C´
ecile Bulle
for sharing their expertise on the development of fate factors, Olivier
Jolliet for sharing his expertise on midpoint-to-endpoint conversion of
characterization factors, and Francesca Verones for the exchanges on
GLAM Phase 3 recommendations. The authors also thank Jorge Corella-
Puertas for his support on mathematical derivations, and Maxime Agez
and Nadim Saadi for their support on Python coding. Furthermore, the
authors thank Clo´
e Delcourt, Ameed Shehayeb, Greyson He and Nadim
Saadi for their contribution to the collection of supporting data. More-
over, the authors appreciate the contribution of Mari MCCV to the
design of the graphical abstract. Finally, the authors wish to acknowl-
edge the support of the UN Environment Life Cycle Initiative and the
Forum for Sustainability through Life Cycle Innovation (FSLCI) to
MarILCA, the nancial support of the FRQNT PBEEE scholarship, as well
as the nancial support of CIRAIG’s partners: ArcelorMittal, Hydro-
Qu´
ebec, LVMH, Michelin, Nestl´
e, Optel Group, Solvay, TotalEnergies,
Umicore, Richemont and OCP Group.
E. Corella-Puertas et al.
Journal of Cleaner Production 418 (2023) 138197
11
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.jclepro.2023.138197.
References
Abej´
on, R., Bala, A., V´
azquez-Rowe, I., Aldaco, R., Fullana-i-Palmer, P., 2020. When
plastic packaging should be preferred: life cycle analysis of packages for fruit and
vegetable distribution in the Spanish peninsular market. Resour. Conserv. Recycl.
155, 104666 https://doi.org/10.1016/j.resconrec.2019.104666.
Alimi, O.S., Fadare, O.O., Okoffo, E.D., 2021. Microplastics in African ecosystems:
current knowledge, abundance, associated contaminants, techniques, and research
needs. Sci. Total Environ. 755, 142422 https://doi.org/10.1016/j.
scitotenv.2020.142422.
Alkema, L.M., Van Lissa, C.J., Kooi, M., Koelmans, A.A., 2022. Maximizing realism:
mapping plastic particles at the ocean surface using mixtures of normal distributions.
Environ. Sci. Technol. https://doi.org/10.1021/acs.est.2c03559.
Bergmann, M., Gutow, L., Klages, M., 2015. Marine Anthropogenic Litter. Springer
International Publishing AG Switzerland. https://doi.org/10.1021/acs.est.9b01360.
Boulay, A., Woods, J.S., Verones, F., Vazquez, I., 2019. A new international working
group on marine impacts in life cycle assessment. In: SETAC Europe 29th Annual
Meeting (Helsinki).
Bulle, C., Margni, M., Patouillard, L., Boulay, A.M., Bourgault, G., De Bruille, V., Cao, V.,
Hauschild, M., Henderson, A., Humbert, S., Kashef-Haghighi, S., Kounina, A.,
Laurent, A., Levasseur, A., Liard, G., Rosenbaum, R.K., Roy, P.O., Shaked, S.,
Fantke, P., Jolliet, O., 2019. IMPACT World+: a globally regionalized life cycle
impact assessment method. Int. J. Life Cycle Assess. 24, 1653–1674. https://doi.org/
10.1007/s11367-019-01583-0.
Castillo, A.B., Al-Maslamani, I., Obbard, J.P., 2016. Prevalence of microplastics in the
marine waters of Qatar. Mar. Pollut. Bull. 111, 260–267. https://doi.org/10.1016/j.
marpolbul.2016.06.108.
Chamas, A., Moon, H., Zheng, J., Qiu, Y., Tabassum, T., Jang, J.H., Abu-Omar, M.,
Scott, S.L., Suh, S., 2020. Degradation rates of plastics in the environment. ACS
Sustain. Chem. Eng. 8, 3494–3511. https://doi.org/10.1021/
acssuschemeng.9b06635.
Corella-Puertas, E., Guieu, P., Aufoujal, A., Bulle, C., Boulay, A.-M., 2022. Development
of simplied characterization factors for the assessment of expanded polystyrene and
tire wear microplastic emissions applied in a food container life cycle assessment.
J. Ind. Ecol. 26, 1882–1894. https://doi.org/10.1111/jiec.13269.
Curren, E., Kuwahara, V.S., Yoshida, T., Leong, S.C.Y., 2021. Marine microplastics in the
ASEAN region: a review of the current state of knowledge. Environ. Pollut. 288,
117776 https://doi.org/10.1016/j.envpol.2021.117776.
Domercq, P., Praetorius, A., MacLeod, M., 2022. The Full Multi: an open-source
framework for modelling the transport and fate of nano- and microplastics in aquatic
systems. Environ. Model. Software 148, 105291. https://doi.org/10.1016/j.
envsoft.2021.105291.
Dreillard, M., Barros, C.D.F., Rouchon, V., Emonnot, C., Lefebvre, V., Moreaud, M.,
Guillaume, D., Rimbault, F., Pagerey, F., 2022. Quantication and morphological
characterization of microbers emitted from textile washing. Sci. Total Environ. 832
https://doi.org/10.1016/j.scitotenv.2022.154973.
Fantke, P., Bijster, M., Guignard, C., Hauschild, M., Huijbregts, M., Jolliet, O.,
Kounina, A., Magaud, V., Margni, M., McKone, T., Posthuma, L., Rosenbaum, R.K.,
van de Meent, D., van Zelm, R., 2018. USEtox® 2.0 Documentation. Version 1.1.
Firdaus, M., Trihadiningrum, Y., Lestari, P., 2020. Microplastic pollution in the sediment
of Jagir Estuary, Surabaya city, Indonesia. Mar. Pollut. Bull. 150, 110790 https://
doi.org/10.1016/j.marpolbul.2019.110790.
Geyer, R., Gavigan, J., Jackson, A.M., Saccomanno, V.R., Suh, S., Gleason, M.G., 2022.
Quantity and fate of synthetic microber emissions from apparel washing in
California and strategies for their reduction. Environ. Pollut. 298, 118835 https://
doi.org/10.1016/j.envpol.2022.118835.
Grobert, N., Keskitalo, C., Kondorosi, ´
E., Kruusmaa, M., Melloni, A., Nakicenovic, N.,
Nurse, P., Dykstra, P., Fortunato, E., Heuer, R.-D., 2020. Biodegradability of Plastics
in the Open Environmen, European Comission. https://doi.org/10.2777/284409.
Hajjar, C., Bulle, C., Boulay, A., 2020. The Fate of Microplastics in the Marine
Environment. SETAC SciCon, Dublin.
Hauschild, M.Z., Huijbregts, M.A.J., 2015. Life Cycle Impact Assessment, LCA
Compendium – the Complete World of Life Cycle Assessment. Springer, Dordrecht.
https://doi.org/10.1007/BF02978760.
Heijungs, R., 1995. Harmonization of methods for impact assessment. Environ. Sci.
Pollut. Res. 2, 217–224. https://doi.org/10.1007/BF02986769.
Henry, B.B., Laitala, K.K., Klepp, I.G., 2018. Microplastic Pollution from Textiles: A
Literature Review.
Hidalgo-Ruz, V., Gutow, L., Thompson, R.C., Thiel, M., 2012. Microplastics in the marine
environment: a review of the methods used for identication and quantication.
Environ. Sci. Technol. 46, 3060–3075. https://doi.org/10.1021/es2031505.
Huijbregts, M., Steinmann, Z.J.N., Elshout, P.M.F.M., Stam, G., Verones, F., Vieira, M.D.
M., Zijp, M., van Zelm, R., 2016. ReCiPe 2016: A Harmonized Life Cycle Impact
Assessment Method at Midpoint and Endpoint Level Report I: Characterization.
National Institute for Public Health and the Environment, The Netherlands.
Bilthoven, The Netherlands.
Hurley, R., Woodward, J., Rothwell, J.J., 2018. Microplastic contamination of river beds
signicantly reduced by catchment-wide ooding. Nat. Geosci. 11, 251–257.
https://doi.org/10.1038/s41561-018-0080-1.
Jolliet, O., Margni, M., Charles, R., Humbert, S., Payet, J., Rebitzer, G., Rosenbaum, R.,
2003. Impact 2002+: a new life cycle impact assessment methodology olivier. Int. J.
Life Cycle Assess. 324–330. https://doi.org/10.1007/BF02978505.
Jolliet, O., Rosenbaum, R.K., Chapman, P.M., Mckone, T., Margni, M., Scheringer, M.,
Straalen, N. Van, Wania, F., 2006. Establishing a framework for life cycle toxicity
assessment - ndings of the Lausanne review workshop. Int. J. Life Cycle Assess. 11,
209–212. https://doi.org/10.1065/lca2006.03.002.
Kim, L., Kim, S.A., Kim, T.H., Kim, J., An, Y.J., 2021. Synthetic and natural microbers
induce gut damage in the brine shrimp Artemia franciscana. Aquat. Toxicol. 232,
105748 https://doi.org/10.1016/j.aquatox.2021.105748.
Koelmans, A.A., Redondo-Hasselerharm, P.E., Nor, N.H.M., de Ruijter, V.N., Mintenig, S.
M., Kooi, M., 2022. Risk assessment of microplastic particles. Nat. Rev. Mater.,
0123456789 https://doi.org/10.1038/s41578-021-00411-y.
Lavoie, J., Boulay, A., Bulle, C., 2021. Aquatic micro- and nano-plastics in life cycle
assessment: development of an effect factor for the quantication of their physical
impact on biota. J. Ind. Ecol. 1–13. https://doi.org/10.1111/jiec.13140.
Li, W.C., Tse, H.F., Fok, L., 2016. Plastic waste in the marine environment: a review of
sources, occurrence and effects. Sci. Total Environ. 566–567, 333–349. https://doi.
org/10.1016/j.scitotenv.2016.05.084.
Loubet, P., Couturier, J., Arduin, R.H., Sonnemann, G., 2021. Life cycle inventory of
plastics losses from seafood supply chains: methodology and application to French
sh products. Sci. Total Environ. 804, 150117 https://doi.org/10.1016/j.
scitotenv.2021.150117.
Maga, D., Galafton, C., Bl¨
omer, J., Thonemann, N., ¨
Ozdamar, A., Bertling, J., 2022.
Methodology to address potential impacts of plastic emissions in life cycle
assessment. Int. J. Life Cycle Assess. https://doi.org/10.1007/s11367-022-02040-1.
Mohamed Nor, N.H., Obbard, J.P., 2014. Microplastics in Singapore’s coastal mangrove
ecosystems. Mar. Pollut. Bull. 79, 278–283. https://doi.org/10.1016/j.
marpolbul.2013.11.025.
Orona-N´
avar, C., García-Morales, R., Loge, F.J., Mahlknecht, J., Aguilar-Hern´
andez, I.,
Ornelas-Soto, N., 2022. Microplastics in Latin America and the Caribbean: a review
on current status and perspectives. J. Environ. Manag. 309 https://doi.org/10.1016/
j.jenvman.2022.114698.
Owsianiak, M., Hauschild, M.Z., Posthuma, L., Saouter, E., Vijver, M.G., Backhaus, T.,
Douziech, M., Schlekat, T., Fantke, P., 2023. Ecotoxicity characterization of
chemicals: Global recommendations and implementation in USEtox. Chemosphere
310. https://doi.org/10.1016/j.chemosphere.2022.136807.
Payet, J., 2004. Assessing Toxic Impacts on Aquatic Ecosystems in LCA. ´
Ecole
Polytechnique F´
ed´
erale De Lausanne, Switzerland. https://doi.org/10.5075/ep-
thesis-3112.
Pfohl, P., Wagner, M., Meyer, L., Domercq, P., Praetorius, A., Hüffer, T., Hofmann, T.,
Wohlleben, W., 2022. Environmental degradation of microplastics: how to measure
fragmentation rates to secondary micro- and nanoplastic fragments and dissociation
into dissolved organics. Environ. Sci. Technol. 56, 11323–11334. https://doi.org/
10.1021/acs.est.2c01228.
Pinlova, B., Nowack, B., 2023. Characterization of ber fragments released from
polyester textiles during UV weathering. Environ. Pollut. 322, 121012 https://doi.
org/10.1016/j.envpol.2023.121012.
Quantis and EA, 2020. Plastic Leak Project - Methodological Guidelines V1, vol. 3.
Rosenbaum, R.K., Margni, M., Jolliet, O., 2007. A exible matrix algebra framework for
the multimedia multipathway modeling of emission to impacts. Environ. Int. 33,
624–634. https://doi.org/10.1016/j.envint.2007.01.004.
Salieri, B., Stoudmann, N., Hischier, R., Som, C., Nowack, B., 2021. How relevant are
direct emissions of microplastics into freshwater from an lca perspective?
Sustainability 13. https://doi.org/10.3390/su13179922.
Suaria, G., Avio, C.G., Mineo, A., Lattin, G.L., Magaldi, M.G., Belmonte, G., Moore, C.J.,
Regoli, F., Aliani, S., 2016. The mediterranean plastic soup: synthetic polymers in
mediterranean surface waters. Sci. Rep. 6, 1–10. https://doi.org/10.1038/
srep37551.
Tang, Y., Nangah, R., Marzia, M., 2022. An effect factor approach for quantifying the
impact of plastic additives on aquatic biota in life cycle assessment. Int. J. Life Cycle
Assess. https://doi.org/10.1007/s11367-022-02046-9.
UNEP, 2019. Global Guidance for Life Cycle Impact Assessment Indicators (GLAM).
UNEP, 2022. Single-use Supermarket Food Packaging and its Alternatives:
Recommendations from Life Cycle Assessments.
Van Colen, C., Moereels, L., Vanhove, B., Vrielinck, H., Moens, T., 2021. The biological
plastic pump: Evidence from a local case study using blue mussel and infaunal
benthic communities. Environ. Pollut. 274, 115825 https://doi.org/10.1016/j.
envpol.2020.115825.
Vigil, M., Pedrosa-Laza, M., Cabal, J.V.A., Ortega-Fern´
andez, F., 2020. Sustainability
analysis of active packaging for the fresh cut vegetable industry by means of
attributional & consequential life cycle assessment. Sustain. Times 12. https://doi.
org/10.3390/su12177207.
Vismaya, A., Simon, M., Jayasree, P.K., Peter, L., 2021. Strength characteristics of
subgrade soil stabilized with plastic bag strips. In: Patel, S., Solanki, C.H., Reddy, K.
R., Shukla, S.K. (Eds.), Proceedings of the Indian Geotechnical Conference 2019.
Springer Singapore, Singapore, pp. 485–493.
Woods, J.S., Verones, F., Jolliet, O., V´
azquez-Rowe, I., Boulay, A., 2021. A framework for
the assessment of marine litter impacts in life cycle impact assessment. Ecol. Indicat.
129, 107918 https://doi.org/10.1016/j.ecolind.2021.107918.
Zhang, C., Wang, S., Sun, D., Pan, Z., Zhou, A., Xie, S., Wang, J., Zou, J., 2020.
Microplastic pollution in surface water from east coastal areas of Guangdong, South
E. Corella-Puertas et al.
Journal of Cleaner Production 418 (2023) 138197
12
China and preliminary study on microplastics biomonitoring using two marine sh.
Chemosphere 256, 127202. https://doi.org/10.1016/j.chemosphere.2020.127202.
Zhang, K., Hamidian, A.H., Tubi´
c, A., Zhang, Y., Fang, J.K.H., Wu, C., Lam, P.K.S., 2021.
Understanding plastic degradation and microplastic formation in the environment: a
review. Environ. Pollut. 274 https://doi.org/10.1016/j.envpol.2021.116554.
Zhao, X., You, F., 2022. Life cycle assessment of microplastics reveals their greater
environmental hazards than mismanaged polymer waste losses. Environ. Sci.
Technol. 56, 11780–11797. https://doi.org/10.1021/acs.est.2c01549.
E. Corella-Puertas et al.
