Journal of Cleaner Production 240 (2019) 118202

Contents lists available at ScienceDirect

Journal of Cleaner Production

journal homepage: www.elsevier.com/locate/jclepro

Graphene nanoplatelet reinforced concrete for self-sensing structures

e A lifecycle assessment perspective
Ioanna Papanikolaou a, *, Noemi Arena b, Abir Al-Tabbaa a
a
Department of Engineering, University of Cambridge, Trumpington St, Cambridge, CB2 1PZ, United Kingdom
b
Costain Group Plc., Maidenhead, SL6 4UB, United Kingdom

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

Article history: Concrete is the most widely used construction material, however, concrete structures often suffer from
Received 11 February 2019 poor durability and require frequent inspections and repairs. Concrete production is also associated with
Received in revised form high carbon emissions and the large-scale depletion of natural resources. Recently, the addition of
4 July 2019
graphene nanoplatelets (GNPs) in cementitious materials has shown a potential to improve the per-
Accepted 26 August 2019
Available online xxx
formance and instigate additional functionalities. However, there is limited understanding around the
environmental effects from the production of GNP and its incorporation in concrete. This study has
Handling Editor: M.T. Moreira investigated, by means of a Life Cycle Assessment (LCA), the environmental impact of concrete reinforced
with graphene nanoplatelets by focusing on the “cradle-to-gate” of GNP production and their incorpo-
Keywords: ration in concrete. The production of 1 kg of G2NanPaste (GNPs product) is found to result in 0.17 kgCO2
Life cycle assessment equivalent units which is lower than Portland cement (0.86 kgCO2 eq). Ordinary Portland cement (CEM I) is
Construction materials 248 times more damaging than G2NanPaste in terms of global warming. The superplasticiser addition is
Graphene production found to have a greater environmental impact compared to G2NanPaste. The sensitivity analysis showed
Self-sensing concrete
that if the addition of GNPs results in a 5% reduction of the Portland cement, the effect of the concrete
Eco-design
mix on global warming can be reduced by 21%. This indicates that GNPs could be environmentally
Innovative materials
friendly if used as a supplement for some of the cement. The main aim of this paper is to perform the first
LCA of the addition of graphene nanoplatelets in concrete. It is hoped that this work will pave the way for
further research in this area.
© 2019 Elsevier Ltd. All rights reserved.

1. Introduction
Concrete is one of the most commonly used structural materials
and comprises of a cementitious component (usually CEM I), ag-
gregates, water and air. Over the years it has evolved to include
various additives and admixtures to ensure good strength and to
meet site specific requirements (Neville, 2011). Despite the great
advancements in materials, concrete structures are often damaged
and suffer from poor durability (Gardner et al., 2018). In the UK,
approximately a fifth of the total civil engineering construction
output was due to repair and maintenance (R&M) works from 2011
to 2015 (HM Treasury, 2010) which not only leads to disruptions
and delays for the public (Al-Tabbaa and Paine, 2018) but it also
necessitates a greater consumption of materials and carbon
* Corresponding author.
E-mail addresses: ip324@cam.ac.uk (I. Papanikolaou), noemi.arena@costain.com
(N. Arena), aa22@cam.ac.uk (A. Al-Tabbaa).
emissions.
Concrete production is associated with several environmental
effects, including carbon emissions (mainly from cement manu-
facture), waste, water use and natural resources depletion (The
Concrete Centre, 2016). It is estimated that each tonne of concrete
has approximately 100e300 kg of embodied CO2 (National Ready
Mixed Concrete Association, 2012) depending on its composition.
A standardised mix (based on the 2008 composition and CO2
emissions) results in 73.8 kgCO2 per tonne of concrete which is a
28% reduction compared to the 1990 baseline (The Concrete Centre,
2016). Demand for concrete has also been found to grow along with
the national income to approximately $10e15,000 per person until
demand for new infrastructure has been saturated (Allwood and
Cullen, 2012). In the UK alone, the infrastructure pipeline from
2016 to 2021 accounts for more than £400 billion (HM
Infrastructure and Projects Authority, 2016), hence demand for
concrete is expected to remain very high.
Allwood and Cullen (2012) have undertaken a study on the
environmental impact of concrete consumption and suggest two

https://doi.org/10.1016/j.jclepro.2019.118202
0959-6526/© 2019 Elsevier Ltd. All rights reserved.
2 I. Papanikolaou et al. / Journal of Cleaner Production 240 (2019) 118202
List of acronyms
CEM I Ordinary Portland Cement
GGBS Ground Granulated Blast-furnace Slag
GO Graphene Oxide
GNP Graphene nanoplatelets or multi-layer graphene
(>10 layers)
LCA Life Cycle Assessment
LCIA Life Cycle Impact Assessment
LPE Liquid Phase Exfoliation
RGO Reduced Graphene Oxide
routes for reducing the CO2 emissions: firstly, to optimise the
concrete production methods in order to reduce the energy con-
sumption and secondly, to use less resources (cement, aggregates,
water) whilst providing the same service. Focusing on the latter,
one approach would be to adopt biomimicry principles (Ellen
MacArthur Foundation, 2018), i.e. develop materials that mimic
natural systems. This approach will help in extending the service
life of the structures and reducing in turn the demand for repairs
(Al-Tabbaa and Paine, 2018). Indeed, over the last years there has
been increasing research interest on the development of bio-
mimetic construction materials that can self-heal when damaged
and self-sense their environment and condition (Gardner et al.,
2018). Self-healing concrete has already been trialled successfully
on construction sites in the UK (Davies et al., 2018), (Al-Tabbaa and
Paine, 2018) whilst self-sensing concrete for functional structures
has also been researched (Han et al., 2015). Self-sensing or moni-
toring concrete refers to the material that can monitor its condition
and identify any damage, whilst maintaining or improving the
mechanical and durability performance (Han et al., 2014). Such a
material would lead to a great reduction in the number of in-
spections undertaken on the infrastructure network and will
reduce their impact on the society and the environment.
The self-sensing mechanism depends on the electrical re-
sistivity of the material; a property that defines how strongly the
material will oppose the flow of electric current. In the case of
concrete, the electrical resistivity varies, and it depends on the mix
composition (water/cement ratio, type of aggregates, use of sup-
plementary materials such as silica fume and GGBS), on the age of
the structure and on the exposure conditions. For example, dried
concrete acts as an insulator and does not allow the passage of
current, whilst moist concrete behaves as an electrolyte (Neville,
2011). The presence of ions in moist concrete or in the pore water
allow for the passage of electric current which reduces the re-
sistivity. Therefore, an increase in the water/cement ratio reduces
the resistivity significantly, whilst an increase in the hydration age
will increase the resistivity. This process of transporting electric
current through the concrete is described as electrolytic and it
depends on the ions in the pore water network.
However, the use of functional fillers, allows for a different
method of transporting the electric current through the concrete
and this process is referred to as electronic conduction. Functional
materials are electrical conductors and when added at certain
quantities, they form a continuous path through which the electric
current can travel. The quantity required to form this continuous
conductive path is called the percolation threshold and it varies
depending on the material type, shape and size. When functional
fillers are used at or near the percolation threshold, the electrical
resistivity of concrete is primarily dependent on the formation of
the conductive path rather than the electrolytic conduction
described earlier. Consequently, damage and the effect of cycling
loading can be sensed by passing current through the continuous
conductive path and measuring the resultant effect on the voltage.
In the case of cracking for example, the continuous conductive path
will break, and it will result in an abrupt increase in the electrical
resistivity. The composition of a traditional standard concrete mix
and that of a self-sensing concrete which incorporates functional
fillers and admixtures is illustrated in Fig. 1.
Various functional fillers have been investigated such as steel
and carbon fibers, carbon black, graphite powder (Horszczaruk
et al., 2016), (Chung, 2002). Most recently, some authors have
also investigated the nanomaterial of graphene for self-sensing
application (Shamsaei et al., 2018), (Du et al., 2013), (Pang et al.,
2014). The successful isolation of graphene in 2004, has led to an
increasing research interest both for the material production
(Novoselov et al., 2012) and on its effect on cementitious compos-
ites with over 100 papers published on this topic (Xu et al., 2018).
Graphene is defined by the International Organization for Stan-
dardization (ISO) (2017) as a single layer of carbon atoms which are
bound in a honeycomb structure. Its properties include an intrinsic
strength of 130 GPa, a Young's modulus of 1 TPa, complete imper-
meability to any gases and a million times higher ability than
copper to sustain high densities of electric current (Novoselov et al.,
2012).
There are different forms of graphene nanomaterials including
graphene oxide (GO), reduced graphene oxide (RGO) and graphene
nanoplatelets (GNPs) or multi-layer graphene as otherwise known.
GNPs comprise of layers of graphene sheets (usually between 10
and 100 layers), a thickness of less than 100 nm and a diameter of
several micrometres (Shamsaei et al., 2018). Each graphene mate-
rial possesses different properties with GNPs possessing superior
electrical conductivity performance to GO (Kauling et al., 2018)
which is a pre-requisite to generating a self-sensing mechanism.
This paper focuses on the use of GNPs as functional fillers in
cement composites to generate a self-sensing mechanism. The aim
is to understand, by means of an attributional Life Cycle Assessment
(LCA), the environmental effect of GNPs production and of their
incorporation in a traditional concrete mix design that has been
used in a construction project in the UK.
2. Materials
2.1. GNP production e Liquid Phase Exfoliation
There are two general approaches for graphene production:

a bottom-up approach where graphene can “grow” on silicon
carbide and

a top-down approach where graphene is exfoliated from
graphite (Sharon, 2015).
Each production method will result in a variety in terms of
fundamental properties (eg. size and thickness), quality (e.g.
number of defects) and cost of graphene (Novoselov et al., 2012).
For applications in the construction industry it is essential that the
graphene quality is balanced with the scalability potential (large
volumes of materials required) and cost.
It has been found that Liquid Phase Exfoliation (LPE) is one of the
most scalable techniques and of controlled costs (Raccichini et al.,
2015) to produce graphene of a reasonable quality (Phiri et al.,
2017). Graphite is a layered material and in LPE, it is added in a
liquid solvent and mechanical forces, such as shear and sonication,
are applied in order to exfoliate it (Sharon, 2015). However, a study
of the graphene production with LPE of over 60 producers found
that the graphene quality is rather poor, and many companies
produce “graphene” with over 100 layer thickness (Kauling et al.,
 I. Papanikolaou et al. / Journal of Cleaner Production 240 (2019) 118202
Fig. 1. Composition of a standard concrete mix (top) and self-sensing concrete (bottom).
2018).
The material chosen for this analysis is a multi-layer GNP
product supplied by Nanesa called G2NanPaste. According to the
Technical Data Sheet, the GNP comprises of an average flake
thickness of 40 layers (~14 nm), a carbon content >97% and an
average particle (lateral size) D50 of 25 mm (Nanesa, 2018a). The
increased thickness is found to be a suitable compromise between
resultant properties and cost for construction applications.
2.2. GNP e cement composites
When GNPs are added in cement composites, they have been
found to improve the mechanical and durability properties and to
also reduce the electrical resistivity, which is a prerequisite for a
self-diagnosing mechanism (Shamsaei et al., 2018). However, there
are certain challenges in the literature that do not allow for a
straightforward assessment of the property improvement. The two
main challenges lie in the fact that authors use different GNP
products (with varied thicknesses, sizes and surface areas) and
employ different dispersion techniques to introduce the GNPs in
the cement matrix.
An ideal dispersion is achieved when the nanoparticles are
completely separated from each other and do not form any ag-
glomerates (Korayem et al., 2017), which will help in homoge-
neously realising the benefits of adding GNPs. Dispersion of GNPs in
an aqueous solution is challenging due to their low dispersibility
(Rodríguez-Pe rez et al., 2013), high surface energy and strong van
der Waals attraction (Shamsaei et al., 2018). Limited studies to date
have focused on the dispersion of graphene nanoplatelets in con-
crete. The lack of a dispersion protocol is one of the reasons that an
optimum content has not yet been established. The GNP contents
usually vary from 0.01% to 5% by weight of cement, which is a very
wide range. Different optimum contents have been quoted for each
measured property, for example compressive and flexural strengths
and durability performance. Fig. 2 summarises the literature to date
with a focus on the effect of GNP addition on compressive (Wang
et al. (2016), Tong et al. (2016), Liu et al. (2016), Rehman et al.
(2017), Bai et al. (2018) and Alkhamis and Imqam (2018)) and
flexural strength of CEM I composites (Wang et al. (2016), Alkhamis
and Imqam (2018) and Zohhadi et al. (2015)). As it is shown, the
most frequently tested GNP dosages range from 0.01% to 0.1% by
3
weight of binder. At that dosage range, most authors observe
compressive strength increases from 2% to 32% whilst flexural
strength has a greater improvement, ranging from approximately
7%e40%. The underlying reason for the differences in the strength
improvements is the fact that not all GNPs had the same properties
and they were also dispersed with different methods. Other factors
that could affect the reported mechanical properties include the
type of cementitious material used and the mix composition (such
as w/c ratio). In this paper, only mixes that incorporate CEM I are
presented. Even though the mix compositions vary between au-
thors, this is not considered to be a reason for the differences in the
mechanical properties improvement since they are always
compared to their respective control mixes.
In terms of durability performance, a 5% by weight of cement
GNP addition was found to reduce the water penetration depth of
the mortar by 75% (Du and Pang, 2015) and a reduction in water
sorption between 27 and 50% was also observed for GNP contents
ranging from 0.5 to 2.5% by weight of cement (Du et al., 2016).
Finally, GNP addition have been found to reduce the electrical re-
sistivity, a pre-requisite for generating additional functionalities
such as a self-diagnosing mechanism. A decrease in resistivity of
more than one order of magnitude was observed for GNP dosages
between 2.4% and 3.6% by volume of cement (Le et al., 2014). Liu
et al. (2016) found that mortar containing 6.4% GNPs by weight of
cement, had a stable electrical resistance as well as an immediate
and accurate response to compressive loading (known as piezor-
esistive effect). Therefore, GNPs are a promising candidate for
reinforcing the cement matrix and improving the mechanical
properties and durability performance.
However, the environmental impact of using GNPs as functional
fillers in cement composites has not been investigated to date.
There is limited understanding around the environmental effects of
graphene production and of its effect when incorporated in cement
composites.
The aim of this paper is to establish, by means of an attributional
Life Cycle Assessment (LCA), the environmental effect of using
graphene-nanoplatelets as functional fillers in cement composites
to improve performance and to generate a self-sensing mechanism.
The focus is on the effect of the GNP production and incorporation
in the cement composites, i.e. the ‘cradle-to-gate’ aspect.
4 I. Papanikolaou et al. / Journal of Cleaner Production 240 (2019) 118202
Fig. 2. Summary of the literature on the effect of GNPs on the compressive (top) and flexural (bottom) strength of an ordinary Portland Cement composite at 28 days.
3. Goal and scope definition
Due to the increasing interest in advanced materials for concrete
structures, an insight in the environmental performance is neces-
sary before such products are commercialised. This study is
intended for an audience in the infrastructure sector who are
interested in using advanced materials in their projects to solve
durability, inspection and monitoring challenges.
The goal of the study was to evaluate, by means of a Life Cycle
Assessment, the environmental impacts associated with the pro-
duction and use of graphene nanoplatelets (GNPs) in concrete. The
GNPs considered in this study are produced by Liquid Phase Exfo-
liation (LPE) in Italy, Europe. This GNP product was selected
following a review of GNP manufacturers. GNPs that were very
exfoliated (<10 nm thick) were found to be unsuitable for civil
engineering applications, primarily due to their prohibiting cost but
also due to additional difficulties in dispersing them homoge-
neously in concrete. On the contrary, the GNP product considered
in this study comprises of more layers (approximately 14 nm thick)
and its cost is significantly lower compared to more exfoliated
GNPs. The increased thickness allows for this product to be a cost-
effective solution whilst at the same time dispersion in concrete is
less challenging. The produced GNPs are then introduced in a
standard concrete mix design in the UK, which is assumed to be the
country of the final concrete use. However, the use of concrete
reinforced with GNPs is not considered in this study.
The LCA was carried out according to ISO standards (ISO-14040,
2006); (ISO-14044, 2006). The functional unit is 1 m3 of concrete.
This functional unit was selected because it is commonly used in
the construction industry to calculate the composition of the con-
crete mix, estimate costs and to quantify the environmental impact,
making this study easily comparable to others. Other functional
units (such as the total volume of concrete required for a specific
application) have not deemed to be suitable for this study due to
the many unknowns that still exist on the effect of GNPs in the
resultant properties of concrete.
In the base case, the GNPs are fabricated in Italy (EU) whilst the
concrete mix is produced and used in the UK. In this scenario, the
GNP production is assessed along with the impact of using the
specified 1 m3 of concrete. The boundaries of this study are from
 I. Papanikolaou et al. / Journal of Cleaner Production 240 (2019) 118202
“cradle to gate”, meaning that they include the processing of raw
materials for GNP production to the delivery and processing of the
product (1 m3) concrete to be used in the UK. The use and final
disposal of the product have not been considered due to many
unknowns that would exist without a focused application (eg. a
specific bridge or building where this concrete would be used), but
these two stages have been taken into account in the discussion
section. The system boundaries of the system are illustrated in
Fig. 3.
The data were primarily obtained from commercial companies
(Nanesa for GNP production and Costain Group for concrete pro-
duction) as well as from scientific literature. The remaining data
were obtained from the life cycle inventory databank Ecoinvent 3.0.
The LCA was carried out using the software SimaPro (2015). The life
cycle environmental impacts were assessed using the Impact
2002 þ methodology (Jolliet et al., 2003). This methodology pro-
vides damage and impact categories and gives an overall and
complete analysis on the whole life cycle of the project. These two
types of results allow different audiences to understand the results
as one is more general (damage category) and the other one is more
scientific for a technical audience. In this paper, the results are
mainly reported in terms of mid-point categories and those
selected for this study are: carcinogens and non-carcinogens, res-
piratory inorganics, aquatic and terrestrial ecotoxicity, global
warming, non-renewable energy and mineral extraction. A nor-
malisation process was then carried out to identify the highest
impact categories for the system (ISO-14040, 2006).
4. Life cycle inventory
The lifecycle inventory of the materials, emissions and energy
consumption for graphene production is based on the data supplied
by the industrial producer (Nanesa, 2018b). The concrete mix
design that has been used as the base case, was supplied by the
company Costain and was used in one of their projects. Table 1
summarises the primary and secondary data that are considered
in this study.
In this study, the following assumptions have been made:

GNPs are produced by LPE in Italy, Europe.

G2NanPaste product includes only water and GNPs, no chemical
agent has been considered.

Concrete mix prepared in the UK.

Input energy is representative of the energy mix of the two
countries, Italy and UK. Initiatives to reduce energy consump-
tion or use clean energy sources in the production of either
GNPs or concrete have not been considered.
Fig. 3. System boundaries for the self-sensing graphene-reinforced concrete system under analysis.
5

The dosage of G2NanPaste in concrete that was assumed is
representative of what is commonly used in the literature (as
per Fig. 2).

The avoided impacts related to the use of ground granulated
blast furnace slag (GGBS) have been assumed to replace not
virgin material but the average mix of virgin and recycled ma-
terial actually used in the market (Arena et al., 2017), (Gala et al.,
2015).
Starting with the production of graphene nanoplatelets, the
simplified process is shown in Fig. 4, which includes two main
phases; the graphite expansion phase followed by the exfoliation
phase. The production of 1 kg of G2NanPaste (which is the GNP
product that is then added in concrete to instigate the self-sensing
mechanism) requires graphite and water as raw materials and
electrical and thermal energy.
Following GNP production, to produce 1 m3 of GNP-reinforced
concrete (the functional unit of this study) at least 2% of G2Nan-
Paste by weight of cement is required. The G2NanPaste product,
comprises of approximately 95% water and 5% GNPs.
Table 2 includes the material quantities for the base case, the
GNP-reinforced concrete and also for the sensitivity analysis. The
concrete mix process involves two steps; firstly, an aqueous solu-
tion containing water, superplasticiser and GNPs is prepared, which
is followed by the addition and mixing of the remaining raw ma-
terials - Portland cement CEM I, GGBS, limestone (coarse aggregate)
and sand (fine aggregate). In preparing the aqueous solution of
GNPs, two mixing methods have been considered. The first option
involves sonication of the solution for 2 h, whilst the second option
involves high speed shear mixing of the solution at 4500 and
7000 rpm for a total of 10 min. The electric energy consumption of
the sonicator is taken as 4 kW (Nanesa, 2018b) and that of the high-
speed shear mixer is taken as 1.1 kW (Nanesa, 2018b). Upon prep-
aration of the aqueous solution, a power consumption of 1.9 kW
was assumed for the electricity consumption at the ready-mix
concrete batching plant.
5. Life cycle impact assessment and interpretation
The contributions from the production of GNPs and the manu-
facture of 1 m3 of GNP-reinforced concrete have been quantified in
terms of damage categories and impact categories. The results are
normalized and presented in terms of person equivalent units,
representing the average impact caused by a person for one year in
each respective category (Sleeswijk et al., 2008).
6 I. Papanikolaou et al. / Journal of Cleaner Production 240 (2019) 118202

Table 1
Inventory data source classification.

Primary data Source

Concrete mix design Information supplied by Costain

Production of G2NanPaste GNP manufacturer (Nanesa)
Graphene solution mixing - duration and energy requirements Preliminary experimental work from the authors for duration
and Nanesa for electric energy consumption
Superplasticiser composition Information from the chemical companies

Secondary data Source

Italian energy mix for GNP production Ecoinvent 3.0

UK energy mix for concrete mix requirements Ecoinvent 3.0
GNP dosage Based on the literature review findings
Raw materials such as graphite and water Ecoinvent 3.0

Fig. 4. Simplified methodology for production the G2NanPaste product (GNPs) (top) and Process for producing graphene-reinforced concrete (bottom).

Table 2
Material quantities used in this study to produce 1 m3 of the different concrete mixes.

Material Base case GNP-concrete Sensitivity analysis

Portland cement: CEM I kg/m3 150 150 142.5

GGBS: kg/m3 310 310 310
Limestone (Aggregates): kg/m3 900 900 900
Sand: kg/m3 805 805 805
Water: kg/m3 185 182.151 182.15
Superplasticiser: kg/m3 e 1.35 1.35
G2NAN Paste (95% water): kg/m3 e 3 32
Total density: kg/m3 2350 2351.5 2344

Note1: Water in GNP-reinforced concrete is reduced to take into account the water available in the G2NanPaste. The total water content in the mix remains the same
and is equal to 185 kg/m.3.
Note2: The content of GNP in the sensitivity analysis remains the same as in the remaining of the study (3 kg/m3) and the only variable that changes in the content of
CEM I which reduces by 5%.

5.1. Graphene nanoplatelets (G2NanPaste) production comprises of two main phases - expansion and exfoliation. From
Fig. 5 it is apparent that the greatest effect comes from the elec-
The production of the GNP product, G2NanPaste has been ana- tricity consumption during the exfoliation and expansion phases,
lysed and is presented in Fig. 5. The G2NanPaste production followed by thermal energy and the use of graphite as a raw
 I. Papanikolaou et al. / Journal of Cleaner Production 240 (2019) 118202
Fig. 5. Normalized results in terms of impact (mid-point) categories of life cycle impact assessment of 1 kg of G2NanPaste production. Here and in the following figures the criteria
for normalisation are per person equivalent units, where one person equivalent represents the average global per capita annual average impact in the specific category.
material. The consumption of water appears to be negligible in the
whole GNP manufacturing process. Non-renewable energy con-
sumption is shown to be the most significant localised impact
followed by global warming potential and respiratory inorganics.
The impacts of GNP production could be reduced by considering
these impacts and optimising the manufacturing process accord-
ingly. For example, since non-renewable energy consumption is
found to be the greatest localised impact, GNP manufacturers
should seek to use renewable energy (such as solar and wind) for
generating the required electricity for GNP production. Reducing
the impact from the use of graphite as a raw material can be ach-
ieved by making the mining process more efficient and by reducing
the transport requirements from graphite mining to GNP
manufacture.
In terms of the impact of cement production on climate change,
the production of 1 kg of CEM I corresponds to 0.86 kgCO2 eq. GNP
production in this study has been found to be more sustainable
compared to cement and the production of 1 kg of G2NanPaste
corresponds to only 0.17 kgCO2 eq. This indicates that the effect of
GNPs could be environmentally friendly if used as a supplement for
some of the CEM I or if it led to a reduced need for cementitious
materials due to providing the same level of service.
5.2. Base case
A standard concrete mix design has been used to establish the
baseline for this study and is then compared with the GNP-
reinforced concrete mix. The results for 1 m3 of concrete are pre-
sented in Fig. 6 in terms of impact categories. It is evident that
Portland cement (CEM I) is the largest contributor, followed by
aggregates (sand and limestone as fine and coarse aggregate
respectively). GGBS is a by-product product of the blast furnaces
used in the iron industries (Cementitious Slag Makers Association,
2018) and it has been considered as avoided impact since it is used
as partial replacement of virgin raw materials (CEM I). The benefits
of using GGBS are twofold; firstly, the environmental impact of
7
disregarding GGBS as waste from the manufacturing process of iron
is avoided, and secondly, the need for consuming raw materials in
concrete is reduced by partly replacing CEM I with GGBS. The use of
water is found to have a negligible effect and the same applied to
the electricity. This is because modern concrete batching plants
have improved their energy efficiency and the energy intensity has
reduced from 132.1 kWh/tonne in 2008 to 122.7 kWh/tonne in
2015 (The Concrete Centre, 2016). In addition, in the UK approxi-
mately 22% of the total electricity demand was covered by
renewable energy sources (European Environment Agency, 2017)
whilst around 20% of the electricity comes from nuclear energy
(World Nuclear Association, 2018). Global warming is found to be
the largest mid-point impact category and to produce this 1 m3 of
concrete is found to correspond 19.42 kgCO2 eq. This is significantly
less than what has been quoted in the literature for a standard
concrete mix (National Ready Mixed Concrete Association, 2012),
(The Concrete Centre, 2016) and the difference is attributed to the
high contents of GGBs that substitute Portland cement. The impact
of global warming is followed by the consumption of non-
renewable energy and respiratory inorganics, which shows that
the effect on public health should also be considered.
5.3. Comparison of GNP dispersion methods in concrete
As discussed earlier, homogeneous dispersion of GNPs in the
cement matrix is a key challenge in the literature and very few
studies have investigated the optimum method. Following previous
work by the authors (Papanikolaou et al., 2018) it was established
that superplasticisers are necessary in dispersing GNPs in the
cement matrix and they should also be combined with a mechan-
ical mixing method. Two mechanical mixing methods are analysed
and compared in this study e one is the sonication of the solution
for 2 h and the other is high-speed shear mixing for 10 min. Both
cases involve the addition of a superplasticiser. The results for both
options are tabulated in Fig. 7 and Fig. 8.
In both cases, as observed for the base case scenario, Portland
8 I. Papanikolaou et al. / Journal of Cleaner Production 240 (2019) 118202

Fig. 6. Normalized results in terms of impact (mid-point) categories of life cycle impact assessment of producing 1 m3 of concrete.

Fig. 7. Normalized results in terms of impact (mid-point) categories of life cycle impact assessment of producing 1 m3 of concrete with sonicated G2NanPaste.

Fig. 8. Normalized results in terms of impact (mid-point) categories of life cycle impact assessment of producing 1 m3 of concrete with high-speed shear mixed G2NanPaste.
 I. Papanikolaou et al. / Journal of Cleaner Production 240 (2019) 118202 9

Fig. 9. Comparison between all scenarios investigated in this study in terms of the greatest impact categories.

cement has the greatest contribution followed by limestone and
sand, whilst GGBS has a negative contribution (i.e. positive envi-
ronmental impact). Overall, the plasticiser is found to have a greater
effect compared to G2NanPaste e its effect on the global warming it
is 3.22 times that of the G2NanPaste whilst this increases to 6.13
times for respiratory inorganics. These results are in agreement
with Long et al. (2018) who found that the greenhouse gas emis-
sions of GO are less than that of the polycarboxylate super-
plasticiser when added in a concrete mix (GO has 0.38 kgCO2eq.
compared to 0.91 kg CO2eq. for the superplasticiser).
The addition of G2NanPaste is found to be more damaging when
sonication is used compared to high-speed shear mixing which is
reflected in the higher electricity impact. This was expected due to
the longer duration of sonication required and also because soni-
cators have a greater power requirement compared to the high-
speed shear mixer. GNPs can be considered as an admixture for
concrete, whose effect remains insignificant compared to that of
Portland cement.
A comparison between all scenarios investigated in this study is
presented in Fig. 9 and the global warming coefficient is shown at
the inset. All alternative scenarios have a greater impact compared
to the normal concrete mix, but this is because graphite and
G2NanPaste are considered as additions rather than as re-
placements. The impact is more pronounced when sonication is
used for dispersing G2NanPaste. It is interesting to note, that the
primary impact of concrete production is on human health (res-
piratory inorganics) rather than on global warming or the depletion
of resources.
5.4. Sensitivity analysis e reducing cement content whilst
maintaining performance
The above analysis shows that G2NanPaste has a very small
impact compared to other concrete components such as Portland
cement and aggregates. Therefore, a potential increase in the
G2NanPaste content would not significantly affect the overall
impact of the concrete mix. However, the use of GNPs allows for an
enhancement in the concrete performance which could in turn lead

Fig. 10. Normalized results of the overall impact of using G2Nanpaste to replace 5% of Portland cement on human health, ecosystem quality, climate change and resources (damage
categories).
10 I. Papanikolaou et al. / Journal of Cleaner Production 240 (2019) 118202
Fig. 11. Normalized results of the overall impact of using G2Nanpaste to replace 5% of Portland cement expressed in mid-point categories.
to a reduction of the cement content required. Below are reported
the results of an alternative analysis with a 5% reduction in cement
content is assumed.
From Fig. 10 and Fig. 11, it is shown that the effect of G2NanPaste
is minimal. Portland cement remains the biggest contributor, with
the effect being more pronounced for climate change followed by
human health. As shown in Table 3, to produce 1 m3 of this new
concrete mix that incorporates 5% less cement, the effect on global
warming is 15.29 kgCO2 eq. which is significantly less than the base
case that was associated with 19.42 kgCO2 eq. Therefore, reducing
cement by only 5% can result in a 21% reduction in the global
warming potential. This highlights the significance of reducing the
cement content whilst maintaining the same service and the pos-
itive role that GNP addition can play in the lifecycle impact of
concrete.
6. Conclusions
In this study, the environmental impact of producing GNPs and
using them as a reinforcement for concrete structures has been
investigated by means of a Life Cycle Assessment. It has been found
that the production of 1 kg of G2NanPaste with Liquid Phase
Exfoliation results in 0.17 kgCO2 eq. This is less than the production
of Portland cement, which is 0.86 kgCO2 eq, meaning that GNPs
could be environmentally friendly if used as a supplement for some
of the cement.
The impact of using GNPs as a nano-reinforcement has also been
analysed and it has been found that Portland Cement (CEM I) is 248
times more damaging than G2NanPaste in terms of global warming
and 124 times more damaging in terms of respiratory inorganics
(human health) for the specific concrete mix that was investigated
in this study. Portland cement is shown to be the component with
the greatest impact in all cases. Moreover, the impact of the con-
crete mix is greater in all cases in terms of respiratory inorganics
Table 3
Comparison between the base case and the alternative scenario with GNPs and 5%
less cement.
Impact category Unit Base case GNPs þ5% cement reduction
Global warming kg CO2 eq. 19.42 15.29
compared to global warming and non-renewable energy
consumption.
The sensitivity analysis showed that if the addition of GNPs
results in a 5% reduction of the Portland cement, the effect of the
concrete mix on global warming can reduce by 21%. This highlights
that GNPs are a promising additive for nano-reinforcement in
concrete as their environmental impact is not high and they could
lead to a reduction in concrete volume (and Portland cement)
whilst maintaining the same level of service.
7. Further work
This study has concentrated on the cradle-to-gate stage of
producing GNPs and incorporating them in a concrete mix design.
It has considered only one GNP production from a particular
manufacturer. Further work should explore the comparative impact
of different GNP producers in order to identify the optimum GNP
production method. In terms of next steps, a full LCA, cradle to
grave, with a focused construction application (eg. a tunnel lining)
and waste management of this advanced composite material will
be undertaken in conjunction with further experimental activity
looking at the detailed performance improvement with GNP
dosages.
Funding
This work was supported by the Engineering and Physical Sci-
ences Research Council (Grant No. EP/L016095/1-EPSRC Centre for
Doctoral Training in Future Infrastructure and Built Environment
and Grant No. EP/P02081X/1 - Resilient Materials for Life (RM4L))
and Costain Group.
Acknowledgements
The authors would like to thank Nanesa for providing materials
and technical advice.
References
Al-Tabbaa, A., Paine, K., 2018. Biomimetic cementitious construction materials for
next-generation infrastructure. In: Proceedings of the Institution of Civil Engi-
neers - Smart Infrastructure and Construction, vol. 171, pp. 67e76. https://doi.
 I. Papanikolaou et al. / Journal of Cleaner Production 240 (2019) 118202
org/10.1680/jsmic.18.00005.
Alkhamis, M., Imqam, A., 2018. New cement formulations utilizing graphene nano
platelets to improve cement properties and long-term reliability in oil wells. In:
SPE-192342-MS. https://doi.org/10.2118/192342-MS.
Allwood, J., Cullen, J., 2012. Sustainable Materials with Both Eyes Open. UIT Cam-
bridge Ltd., Cambridge.
Arena, N., Sinclair, P., Lee, J., Clift, R., 2017. Life cycle engineering of production, use
and recovery of self-chilling beverage cans. J. Clean. Prod. 142, 1562e1570.
https://doi.org/10.1016/j.jclepro.2016.11.148.
Bai, S., et al., 2018. Enhancement of mechanical and electrical properties of gra-
phene/cement composite due to improved dispersion of graphene by addition
of silica fume. In: Construction and Building Materials, vol. 164. Elsevier Ltd,
pp. 433e441. https://doi.org/10.1016/j.conbuildmat.2017.12.176.
Cementitious Slag Makers Association, 2018. What is GGBS? URL. https://ukcsma.
co.uk/what-is-ggbs/. accessed 11.17.18.
Chung, D.D.L., 2002. Piezoresistive cement-based materials for strain sensing.
J. Intell. Mater. Syst. Struct. 13, 599e609. https://doi.org/10.1106/
104538902031861.
Davies, R., Teall, O., Pilegis, M., Kanellopoulos, A., Sharma, T., Davies, R., 2018. Large
scale Application of self-healing Concrete : design. Construction , and Testing 5,
1e12. https://doi.org/10.3389/fmats.2018.00051.
Du, H., Gao, H.J., Pang, S.D., 2016. Improvement in concrete resistance against water
and chloride ingress by adding graphene nanoplatelet. Cement Concr. Res. 83,
114e123. https://doi.org/10.1016/j.cemconres.2016.02.005.
Du, H., Pang, S.D., 2015. Enhancement of barrier properties of cement mortar with
graphene nanoplatelet. Cement Concr. Res. 76, 10e19. https://doi.org/10.1016/j.
cemconres.2015.05.007.
Du, H., Quek, S.T., Pang, S.D., 2013. Smart multifunctional cement mortar containing
graphite nanoplatelet. Sensors Smart Struct. Technol. Civil, Mech. Aerosp. Syst.
8692, 1e10. https://doi.org/10.1117/12.2009005.
Ellen MacArthur Foundation, 2018. Circular economy schools of thought [WWW
Document]. URL. https://www.ellenmacarthurfoundation.org/circular-
economy/concept/schools-of-thought. accessed 11.5.18.
European Environment Agency, 2017. Renewable Energy in Europe - 2017 Update.
https://doi.org/10.2800/313983.
Gala, A.B., Raugei, M., Fullana-i-Palmer, P., 2015. Introducing a new method for
calculating the environmental credits of end-of-life material recovery in attri-
butional LCA. Int. J. Life Cycle Assess. 20, 645e654. https://doi.org/10.1007/
s11367-015-0861-3.
Gardner, D., Lark, R., Jefferson, T., Davies, R., 2018. A survey on problems encoun-
tered in current concrete construction and the potential benefits of self-healing
cementitious materials. Case Stud. Constr. Mater. 8, 238e247. https://doi.org/10.
1016/j.cscm.2018.02.002.
Han, B., Ding, S., Yu, X., 2015. Intrinsic self-sensing concrete and structures: a re-
view. Meas. J. Int. Meas. Confed. 59, 110e128. https://doi.org/10.1016/j.
measurement.2014.09.048.
Han, B., Yu, X., Ou, J., 2014. Self-sensing Concrete in Smart Structures. Elsevier Inc.,
Oxford.
HM Infrastructure and Projects Authority, 2016. National Infrastructure Delivery
Plan 2016-2021.
HM Treasury, 2010. Infrastructure Cost Review: Main Report. Group.
Horszczaruk, E., Sikora, P., Lukowski, P., 2016. Application of nanomaterials in
production of self-sensing concretes: contemporary developments and pros-
pects. Arch. Civ. Eng. 62, 61e74. https://doi.org/10.1515/ace-2015-0083.
International Organization for Standardization (ISO), 2017. Nanotechnologies e
Vocabulary e Part 13: Graphene and Related Two-Dimensional (2D) Materials.
BSI Standards Publication, London, UK.
ISO-14040, 2006. Environmental Management- Life Cycle Assessment- Principles
and Framework (Geneva, Switzerland).
ISO-14044, 2006. Environmental Management - Life Cycle Assessment - Re-
quirements and Guidelines (Geneva, Switzerland).
Jolliet, O., Margni, M., Charles, R., Humbert, S., Payet, J., Rebitzer, G., Rosenbaum, R.,
2003. Impact 2002þ: a new life cycle impact assessment methodology. Int. J.
Life Cycle Assess. 8. https://doi.org/10.1007/BF02978505.
Kauling, A.P., Seefeldt, A.T., Pisoni, D.P., Pradeep, R.C., Bentini, R., Oliveira, R.V.B.,
Novoselov, K.S., Castro Neto, A.H., 2018. The worldwide graphene flake pro-
duction. Adv. Mater. 1803784, 1e6. https://doi.org/10.1002/adma.201803784.
Korayem, A.H., Tourani, N., Zakertabrizi, M., Sabziparvar, A.M., Duan, W.H., 2017.
A review of dispersion of nanoparticles in cementitious matrices: nanoparticle
geometry perspective. Constr. Build. Mater. 153, 346e357. https://doi.org/10.
11
1016/j.conbuildmat.2017.06.164.
Le, J.L., Du, H., Pang, S.D., 2014. Use of 2-D Graphene Nanoplatelets (GNP) in cement
composites for structural health evaluation. Compos. B Eng. 67, 555e563.
https://doi.org/10.1016/j.compositesb.2014.08.005.
Liu, Q., Xu, Q., Yu, Q., Gao, R., Tong, T., 2016. Experimental investigation on me-
chanical and piezoresistive properties of cementitious materials containing
graphene and graphene oxide nanoplatelets. Constr. Build. Mater. 127, 565e576.
https://doi.org/10.1016/j.conbuildmat.2016.10.024.
Long, W.J., Zheng, D., Duan, H. bo, Han, N., Xing, F., 2018. Performance enhancement
and environmental impact of cement composites containing graphene oxide
with recycled fine aggregates. J. Clean. Prod. 194, 193e202. https://doi.org/10.
1016/j.jclepro.2018.05.108.
Nanesa, 2018a. G2Nan - Technical Data Sheet (TDS).
Nanesa, 2018b. Resource Consumption - Private Communication.
National Ready Mixed Concrete Association, 2012. Concrete CO2 Fact Sheet. NRMCA
Publ. Number 2PCO2 13.
Neville, A., 2011. Properties of Concrete, fifth ed. Pearson Education Limited.
Novoselov, K.S., Fal, V.I., Colombo, L., Gellert, P.R., Schwab, M.G., Kim, K., Ko, V.I.F.,
Colombo, L., Gellert, P.R., Schwab, M.G., Kim, K., 2012. A roadmap for graphene.
Nature 490, 192e200. https://doi.org/10.1038/nature11458.
Pang, S.D., Gao, H.J., Xu, C., Quek, S.T., Du, H., 2014. Strain and Damage Self-Sensing
Cement Composites with Conductive Graphene Nanoplatelet 9061, p. 906126.
https://doi.org/10.1117/12.2045329.
Papanikolaou, I., Litina, C., Al-tabbaa, A., 2018. A preliminary study into the effect of
superplasticisers on the dispersion of graphene materials in cement. In: 12th
Fib International PhD Symposium in Civil Engineering. Czech Technical Uni-
versity in Prague, Prague, Czech Republic, pp. 123e130.
Phiri, J., Gane, P., Maloney, T.C., 2017. General overview of graphene : production ,
properties and application in polymer composites. Mater. Sci. Eng. B 215, 9e28.
https://doi.org/10.1016/j.mseb.2016.10.004.
Raccichini, R., Varzi, A., Passerini, S., Scrosati, B., 2015. The role of graphene for
electrochemical energy storage. Nat. Mater. 14, 271e279. https://doi.org/10.
1038/nmat4170.
Rehman, S.K.U., et al., 2017. A sustainable graphene based cement composite. Sus-
tainability 9 (7), 1e20. https://doi.org/10.3390/su9071229.
Rodríguez-Pe  Martín, N., 2013. The chemistry of pristine
rez, L., Herranz, M.A.,
graphene. Chem. Commun. 49, 3721e3735. https://doi.org/10.1039/
c3cc38950b.
Shamsaei, E., de Souza, F.B., Yao, X., Benhelal, E., Akbari, A., Duan, W., 2018. Gra-
phene-based nanosheets for stronger and more durable concrete: a review.
Constr. Build. Mater. 183, 642e660. https://doi.org/10.1016/j.conbuildmat.2018.
06.201.
Sharon, M., 2015. Towards Mass Production of Graphene: Lab to Industry (Scaling
up) 167e196.
SimaPro, 2015. LCA, 8.3. Release date: 2015-05-29. Available at: http://www.
simapro.co.uk.
Sleeswijk, A.W., van Oers, L.F.C.M., Guine e, J.B., Struijs, J., Huijbregts, M.A.J., 2008.
Normalisation in product life cycle assessment: an LCA of the global and Eu-
ropean economic systems in the year 2000. Sci. Total Environ. 390, 227e240.
https://doi.org/10.1016/j.scitotenv.2007.09.040.
The Concrete Centre, 2016. 9th Concrete Industry Sustainability Performance
Report, vol. 2015. Performance data 16.
Tong, T., et al., 2016. Investigation of the effects of graphene and graphene oxide
nanoplatelets on the micro- and macro-properties of cementitious materials.
In: Construction and Building Materials, vol. 106. Elsevier Ltd, pp. 102e114.
https://doi.org/10.1016/j.conbuildmat.2015.12.092.
Wang, B., Jiang, R., Wu, Z., 2016. Investigation of the Mechanical Properties and
Microstructure of Graphene Nanoplatelet-Cement Composite. https://doi.org/
10.3390/nano6110200.
World Nuclear Association, 2018. Nuclear Power in the United Kingdom [WWW
Document]. URL. http://www.world-nuclear.org/information-library/country-
profiles/countries-t-z/united-kingdom.aspx. accessed 11.15.18.
Xu, Y., Zeng, J., Chen, W., Jin, R., Li, B., Pan, Z., 2018. A holistic review of cement
composites reinforced with graphene oxide. Constr. Build. Mater. 171, 291e302.
https://doi.org/10.1016/j.conbuildmat.2018.03.147.
Zohhadi, N., Aich, N., Matta, F., Saleh, N.B., Ziehl, P., 2015. Graphene nano-
reinforcement for cement composites. In: Nanotechnology in Construction.
Springer International Publishing, Cham, pp. 265e270. https://doi.org/10.1007/
978-3-319-17088-6_34.