Construction and Building Materials 190 (2018) 170–177

Contents lists available at ScienceDirect

Construction and Building Materials

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

Life cycle assessment of geopolymer concrete

Daniel A. Salas a, Angel D. Ramirez a,⇑, Nestor Ulloa a,b, Haci Baykara a, Andrea J. Boero a
a
Escuela Superior Politécnica del Litoral, ESPOL, Facultad de Ingeniería en Mecánica y Ciencias de la Producción, Campus Gustavo Galindo, Km 30.5 Vía Perimetral, P.O.
Box 09-01-5863, Guayaquil, Ecuador
b
Facultad de Ingeniería Mecánica, Escuela Superior Politécnica de Chimborazo, ESPOCH, Panamericana Sur km 1½, Riobamba, Ecuador

h i g h l i g h t s g r a p h i c a l a b s t r a c t

An LCI of geopolymer concrete was

scaled-up from laboratory to
industrial scale.

The source of NaOH for geopolymer
concrete production influences LCA
results.

Using NaOH made from solar salt is
favorable in terms of environmental
impacts.

The energy mix affects geopolymer
concrete LCA results.

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

Article history: The Life Cycle Assessment (LCA) of a Geopolymer Concrete (GC) was elaborated after scaling up the Life
Received 10 February 2018 Cycle Inventory (LCI) from laboratory scale to industrial scale. The most relevant raw materials and pro-
Received in revised form 27 August 2018 cesses contributing to its environmental performance were identified. Besides, the influence in the envi-
Accepted 19 September 2018
ronmental impacts of both, the electricity generation mix considered (2012 and 2018 energy mix for
Ecuador), and the source of alkali activators (produced in Ecuador and imported from Europe), was
demonstrated. The production of sodium hydroxide is the most relevant process in all life cycle impact
Keywords:
categories. An energy mix with a higher contribution of hydroelectricity (2018 energy mix: 85% hydro-
Building
Sustainability
electricity) entails favorable results. The differences between locally produced and imported sodium
Scale-up hydroxide are the energy mix considered (Ecuadorian vs. average European), and the type of sodium
Sodium hydroxide chloride used as raw material (obtained through seawater evaporation in Ecuador vs. solution and rock
Energy mix mining in Europe). GC entails an environmental performance advantage compared to a conventional con-
Carbon footprint crete (CC) if the following two conditions are applied: sodium hydroxide is produced using local solar
salt, and the electricity mix is the expected energy mix for 2018 in Ecuador. Under this condition,
Global Warming Potential (GWP) characterization for GC is 64% lower than CC.
Ó 2018 Elsevier Ltd. All rights reserved.

1. Introduction mental impacts related to cement production are the main issue
regarding the sustainability of its use, as cement represents the
Concrete is one of the most widely used man-made materials in highest CO2 emissions contribution to concrete [2–5].
the world [1]. It is a composite material resulting from the mixture Approaches aiming at improving the sustainability of the build-
of cement, aggregates, water, and chemical admixtures in different ing and construction sector involve the use of cement with reduced
proportions depending on its intended performance. The environ- environmental burden (e.g., blended cement, cement with clinker
replacement with industrial by-products, or less emission intense
⇑ Corresponding author. clinker), and replacement of virgin materials with recycled or
E-mail address: [email protected] (A.D. Ramirez). reclaimed substitutes [6,7]. However, a technological shift towards

https://doi.org/10.1016/j.conbuildmat.2018.09.123
0950-0618/Ó 2018 Elsevier Ltd. All rights reserved.
 D.A. Salas et al. / Construction and Building Materials 190 (2018) 170–177 171

the use of new low CO2 binders, and the application of new con- GWP), Acidification Potential (AP), Eutrophication Potential (EP),
struction techniques are necessary as well [7]. Photochemical Oxidants Formation (POF) and Ozone Depletion
Life Cycle Assessment (LCA) is an analytical tool for assessing (ODP).
the environmental performance of products during their lifespan.
Several LCA studies have been elaborated for estimating environ- 2.1. Goal and scope of the study
mental impacts of cement [4,8–15], and some for concrete
[2,3,5,16–22]. LCA has also been applied to geopolymer cement/- The LCI of GC is scaled up from laboratory scale to industrial
concrete products [23–33]. scale. Subsequently, the environmental performance of GC blocks
Geopolymers, are made by activating amorphous alumino- of standard size (10  20  40 cm – 50% hollow) is assessed under
silicate materials, such as Fly Ash (FA), Blast Furnace Slag (BFS), different circumstances, and a comparison is made with CC blocks
or natural zeolites; with alkali-based chemicals, such as hydrox- using a cradle to gate approach. Fig. 1 shows the system bound-
ides and silicates [34]. The use of natural zeolites as a raw material aries for GC blocks production for the base case scenario.
in synthesizing geopolymeric materials is not so common, but System boundaries include raw materials extraction and pro-
recently zeolites have been used more often as the raw materials duction, alkali activators production, and GC production. The func-
in geopolymer preparation [35–41]. Sodium hydroxide is the most tional unit is ‘‘1 m3 of hollow blocks made of a GC with a specific
widely used alkali activator [34]. There is an abundance of natural compressive strength”. Specific compressive strengths used in this
zeolite sources in Ecuadorian coastal region, which are suitable for study are presented in Table 1. The most relevant processes to the
geopolymer synthesis [42–45]. overall environmental impacts are identified using LCA. Besides,
Geopolymer technology applied to the construction industry the influence of the source of the alkali activators and the energy
presents an environmental performance improvement potential mix considered for its local production is addressed with scenario
at the building material level. Previous research has shown envi- analysis.
ronmental performance advantages of Geopolymer Concrete (GC)
over Conventional Concrete (CC), when used as alternative cemen- 2.2. Geopolymer concrete mix design
titious material [18–20,22,24,26,28,29,31,32,46]. LCA has already
been used to assess the environmental impacts of GC. Main envi- Ulloa et al. [41] developed and tested several mix designs. Three
ronmental burdens within GC life cycle has been associated with representative GC mixes (low, medium, and a high proportion of
the use of alkali activators [19,21,22,24,26,29,31–33]. Using alkali alkali activators) are analyzed. A CC design mix is included for a
activators such as sodium silicate based on rice husk ash has comparative analysis, in order to determine whether an environ-
shown great potential in reducing the environmental impacts of mental improvement is attained by the alternative GC technology.
GC [21]. Although the exact proportions for each mix of GC were used for
The influence on the environmental impacts of the process of calculations, only ranges of components quantities are presented
obtaining the alkali activators and the type of electricity generation for GC in Table 1. The strength requirements for concrete masonry
technology considered at every production process has not been units in Ecuador are met by all the GC mixes [55].
addressed in previous research. It has been recognized that
methodological LCA choices may influence GC environmental 2.3. Case scenarios
impact results and research conclusions [19,20,22,24,26]. The pre-
sent study analyses a novel GC (GC), which contains the following In Ecuador, a shift in the power production matrix towards a
raw materials: natural zeolites, sodium hydroxide (NaOH), sodium greater share of hydroelectricity has been promoted by the Master
silicate (Na2SiO3), and sand. The aim of the present study is to scale Electrification Plan during the last decade. In 2012, 62% of the
up the Life Cycle Inventory (LCI), developed at laboratory scale, to energy mix was comprised of hydroelectricity, and, by 2018,
industrial scale in order to evaluate the environmental perfor- hydroelectricity participation is expected to reach 85% [56]. The
mance of a GC developed by Ulloa et al. [41], and to assess its present LCA is elaborated upon three case scenarios, which are
potential benefits to improve the sustainability of the building sec- detailed in Table 2.
tor when compared to a conventional concrete (CC) mix. The cir- Scenarios S1 and S2 represent the ideal situation, in which
cumstances under which environmental performance sodium hydroxide is manufactured locally with local raw materi-
improvement is more likely to occur are addressed using scenario als. In scenario S3, sodium hydroxide is imported from Europe.
analysis. The detailed LCI differences of these two situations are addressed
in Section 2.4.3. Regarding the LCI for the CC mix, 2018 energy
mix is considered for its production, thus it is compared to S1 G3
2. Materials and methods mix.

The LCI of GC production was developed at laboratory scale 2.4. Life cycle inventory
[41,47]. However, this study aims to estimate the potential envi-
ronmental impacts of its production at an industrial scale. Thus, 2.4.1. Inventory scale-up process
a scale up from laboratory to industrial scale is modeled. The factors that influence the scale-up process of the system are
This LCA study follows the ISO 14,040 and ISO 14,044 method- analyzed, and some processes are modified as explained below.
ology [48,49]. Calculations were performed using the SimaPro soft- Shibasaki et al. [57] analyze a method for undertaking a scale-up
ware [50]. A part of the inventory data was obtained from of a process from a pilot scale to industrial scale within an LCA
literature and testing. Another part was obtained from databases study, in which processes proved to be relevant should be consid-
such as Ecoinvent [51,52] and USLCI [53]. The inventory data ered for scaling-up purposes. Piccinno et al. [58] developed a
sources are detailed in Section 2.4. framework for scaling-up from laboratory to industrial scale, com-
The impact assessment method known as CML 2001, developed plementing the aforementioned method by including earlier points
by the Institute of Environmental Sciences (CML) of Leiden Univer- of products development. Although Shibasaki et al. [57] focus on a
sity [54], was used for the impact evaluation. The following impact more developed phase than laboratory scale, they propose some
categories were considered for the analysis: Abiotic Depletion for considerations pertinent to the present study, such as: paying
Fossil Fuels (ADPF), Climate Change (Global Warming Potential - attention to processes that undergo a change of technology (e.g.
172 D.A. Salas et al. / Construction and Building Materials 190 (2018) 170–177

Raw materials extraction/ Intermediate raw Geopolymer concrete

production materials production production

Evaporation, Other raw

Electricity
Sea materials

Electricity
water crystallization,
harvesting

Water
Chlor-alkali
Sodium process
chloride
Gravel/ Extraction, Sodium hydroxide
Sand grinding, Block
quarry sifting. Mixing making

Sand Silicate
production
Emissions
Natural Extraction, Curing
zeolite grinding, Sodium silicate Fossil fuel
quarry sifting.

Natural Geopolymer
zeolite concrete
block
Fig. 1. System boundaries for GC blocks production.

Table 1
GC and CC mix composition.

Components (kg/m3) G1 - 5 MPa G2 - 5,5 MPa G3 - 15 MPa CC - 15 MPa

Sodium silicate (Na2SiO3) in 45% solid solution 240–260 330–350 400–420 –
Sodium hydroxide (NaOH) 100% solid (pellets) 45–55 45–55 50–60 –
Water 155–165 120–130 70–80 200
Natural zeolite 700–730 700–730 700–730 –
Sand 1070–1100 1070–1100 1070–1100 1100
Gravel – – – 670
Cement – – – 315
Total dry (kg/m3) 1923–2002 1964–2043 2000–2079 2085

Table 2
Case scenarios description. The sodium silicate LCI is based on the work by Fawer et al. [60],
Scenario Description
and the sodium hydroxide LCI is based on the Ecoinvent v3.3 data-
base [51]. The manufacture of 1000 kg of a 48% solid solution of
S1 (Base (G1, G2, G3, CC) - Blocks - Activators manufactured locally
Scenario) with 2018 energy mix (85% hydroelectricity), and sodium
sodium silicate, requires 209 kg of 100% solid sodium hydroxide
chloridea produced by seawater evaporation [60]. This means that the source of the latter may entail a great
S2 G3 blocks - Activators manufactured locally with 2012 effect in the environmental burden related to the silicate as well.
energy mix (62% hydroelectricity), and sodium chloride Thus, in the analysis, attention is paid to the sodium hydroxide
produced by sea water evaporation
LCI, which is produced (along with chlorine and hydrogen) through
S3 G3 blocks - Sodium hydroxide is imported from Europe:
average European energy mix, and sodium chloride produced the chlor-alkali process. Through this process sodium hydroxide is
by solution mining and rock mining obtained by three common techniques: membrane, diaphragm,
a
and mercury cells. In the present study, it is assumed that the
Sodium chloride is the main raw material for sodium hydroxide production
membrane cell technology is used, as it is the most recent break-
through within the chlor-alkali industry [61]. Chemical facilities
mixing, and curing processes in this case) and considering the construction (raw materials, buildings, equipment installation) is
emissions variations related to it. The following sections detail included in the LCI of both alkali-activators.
the LCI scale-up procedure for each process and raw material. The main raw material for the chlor-alkali process is sodium
chloride. The most common sodium chloride production methods
are solar salt, rock mining salt, and solution mining salt [62,63].
2.4.2. Concrete mix plant facilities From these, solar salt is the less energy intensive [63]. A significant
The inventory for the infrastructure and concrete production influence of using different salt sources on the environmental
machines for the mixing plant facilities is based on the Ecoinvent impacts of the chlor-alkali process has already been demonstrated
database [52] and estimations by Borbor et al. [59]. Average [62]. Thus, the source of the sodium hydroxide (which depends on
monthly production of 144000 blocks is considered. the type of salt and the energy mix of the region), is expected to
influence the environmental performance of GC as well.
2.4.3. Alkali activators In Ecuador, sodium chloride is produced mostly by evaporating
Sodium silicate was obtained as a 45% solid content solution sea water, while in Europe, it is produced mainly by rock mining
with a 2,2 Si2O/Na2O relation. Sodium hydroxide was obtained in and solution mining [51]. In the present study, both cases are
the form of pellets. It was used as a solution with 14 M molarity addressed in the scenario analysis, as described in Table 2. For
solution, which was prepared manually at the laboratory scale. the local manufacture of sodium hydroxide (scenarios S1 and S2),
 D.A. Salas et al. / Construction and Building Materials 190 (2018) 170–177 173

solar salt is considered. For this type of salt, the energy consump-
tion of pumps, conveyors, and harvesting machines estimated by
Goetzfried et al. [64] is considered for the LCI. On the other hand,
for the imported sodium hydroxide (scenario S3), the average LCI
of sodium chloride for Europe (rock mining and solution mining)
was considered instead. The brine preparation for the chlor-alkali
process is included in the LCI of sodium hydroxide [51].

2.4.4. Natural zeolite, sand, and gravel

Researchers gathered sand manually from a riverbank and
obtained natural zeolites from a local supplier. Sand and gravel
inventories are based on Ecoinvent v3.3 database [52]. For scaling
Fig. 2. Curing energy estimation model.
up purposes, sand is assumed to be obtained from a local supplier
and produced from sand and gravel quarries. The same is assumed
for gravel (for the CC mix). In the case of natural zeolite, an inven- Table 3
tory was not available. Thus the LCI of a mineral, for which its Transport inventory.
extraction depends on a similar mining method, was selected:
Raw materials Distance from supplier Freight
Limestone. It is assumed that the environmental burden related
to the mining process of these two minerals is similar. Natural zeolite 20,8 km 14–15 tkm
Silicate 63,9 km 15–28 tkm
Hydroxide 12 km 2,9–3,3 tkm
2.4.5. Concrete mixing and block making processes Sand 4,7 km 4,9–5,1 tkm
At the laboratory scale, a small mixer suitable for the applica- Cement* 4,7 km 1,3 tkm
tion of ASTM cement pastes and mortars mixing normative [65] Sand* 4,7 km 4,9 tkm
Gravel* 4,7 km 4,9 tkm
was used. The GC test specimens were then elaborated manually
in a wooden mold. Given the assumed monthly production of *
For CC mix only.
144000 blocks [59], the energy use related to these processes
was estimated at industrial scale. It is assumed that a concrete
pan mixer and a vibrating block making machine are used for these Mgp : geopolymer blocks mass
processes. Suitable commercialized machines for these processes
were selected, and the electricity consumption was estimated C gp : geopolymer specific heat
based on their technical specifications and operation times.
Regarding the mixing process, an appropriate mixing time of 60 s DT : temperature increase
[65,66] was considered. For the block molding process, the mold-
ing time was 40 to 50 s, which depends on the machine U : ov erall heat transfer coefficient
characteristics.
A : heat transfer surface area
2.4.6. Curing process energy estimation
For the present study, steam curing at 60 °C during 24 h is con- 2.4.7. Cement for conventional concrete
sidered for all design mixes. Since the characterization of the local production of cement is
At the laboratory scale, the test specimens were cured using an not available, an average LCI for Cement Portland production based
electrical laboratory oven. At the industrial scale, fossil fuel-based on Ecoinvent [52], which contains 0,9 kg of clinker per kg of
curing is considered. Few studies have addressed the energy use cement, was adapted in order to use it for calculations. The elec-
[67] or GHG emissions [25,32] related to this process. Thus avail- tricity use amount in the LCI and in the elements included in it
ability of estimates for comparison is narrow. (e.g. clinker and sand), was replaced by the 2018 Ecuadorian
The curing process was modeled considering that the curing energy mix LCI.
chamber is made of low thermal conductivity concrete (0,8 W/
mK) with 10 cm thick walls and a gate made of steel and an insu- 2.4.8. Transport
lator. Its dimensions are 2.1  4  6.5 m (height, width, and Including transport of raw materials from suppliers to concrete
length) and it has a capacity of batches of 7000 hollow blocks mixing facilities represents the actual situation at the industrial
(10  20  40 cm – 50% hollow). The specific heat of the GC scale. The concrete mixing plant is assumed to be located in the
(751 J/kg*K) was obtained from Colangelo et al. [68]. The energy same place as the laboratory facilities, and, from that location, dis-
required to heat up the bricks to the curing temperature, and the tances were estimated using Google Maps. Depending on the quan-
input energy required to compensate the heat loss through the cur- tities of each raw material, different lorry freights were considered.
ing kiln walls were estimated (146 MJ/m3). Fig. 2, shows the model Transport inventory for GC and CC mixes is shown in Table 3.
used for energy estimation and the main Eqs. (1–3), based on heat Specifically, for S3, the transoceanic freight for sodium hydroxide
transfer theory [69]. It is assumed that diesel is used as fuel. The importation, available in the Ecoinvent LCI, is considered as well.
curing energy estimation is enough to assess the life cycle impact
of steam generation. The Ecoinvent inventory [52] for ‘‘steam pro- 3. Results and discussion
duction in the chemical industry” for heating was considered.
Q steam ¼ Q heating þ Q keep temp ð1Þ 3.1. Characterization results

Q heating ¼ M gp C gp DT ð2Þ Table 4 shows the environmental impact category results for GC
(G1, G2, and G3) and CC mixes for all scenarios.
Q keep temp ¼ Q loss ¼ UADT ð3Þ As shown in Table 4, within S1, environmental impacts for GC
mixes differ, depending on their compressive strength, which, for
174 D.A. Salas et al. / Construction and Building Materials 190 (2018) 170–177

Table 4
Life cycle impact category results for GC and CC mixes for all scenarios.

Impact category S1 S2 S3
G1 - 5 MPa G2 - 5,5 MPa G3 - 15 MPa CC - 15 MPa G3 - 15 MPa G3 - 15 MPa
ADPF (MJ) 669 771 900 1213 1480 2796
AP (kg SO2 eq.) 0,546 0,625 0,727 0,674 1,237 1,263
EP (kg PO4 eq.) 0,120 0,136 0,155 0,174 0,201 1,245
GWP (kg CO2 eq.) 84 95 110 302 163 254
ODP (104 kg CFC-11 eq.) 1,13 1,33 1,61 0,13 1,67 1,78
POF (kg C2H4 eq.) 0,022 0,025 0,029 0,026 0,049 0,056

instance, depends on the content of activators. The higher the com- GWP, and 17% of ADPF), and mixing facilities (15% of EP) entail
pressive strength, the higher the environmental impacts, due to lower environmental burdens. Zeolites present even lower envi-
higher activators quantities. ronmental burdens in all impact categories (less than 8%). For G1
For comparison with the CC mix, only the G3 is considered, as and G2 mixes, the contribution of activators decreases, due to its
these present the same compressive strength (15 MPa). When lower content. However, the process contribution for these mixes
comparing the G3 (S1) with the CC mix (S1), GWP improves con- is not further analyzed, as G3 presents the highest compressive
siderably in favor of G3 (G3 is 64% lower than CC mix), followed strength, and thus can be compared to the CC mix.
by ADPF (26% lower), and EP (11% lower). AP and POF categories For ODP category, G3 characterization is one order of magnitude
vary slightly (G3 is 7 and 9% higher than CC respectively), and higher. For this category, the combined contribution of both activa-
ODP, which is the worst comparative characterization result, is tors is 96%. Most of the overall ODP burden (approximately 89%) is
12 times higher for G3 than the CC mix. generated during the chlor-alkali process (sodium hydroxide pro-
It is worth to note that the current comparison between GC and duction), due to the use of carbon tetrachloride for chlorine recov-
CC entails limitations due to the uncertainty arose by the charac- ery from gas streams [70].
terization of cement within the LCI of CC. The LCI of local cement Regarding overall results for G3 (S1), the contribution of the sil-
production is not available. Thus an average cement LCI is consid- icate is 3 to 7 times higher than that of the hydroxide in all cate-
ered instead. Besides, it contains cement with a given clinker to gories, due to the higher content of silicate in the design mix
cement ratio (90%), which is subject to change and may possibly (the solid content of silicate is 3,17 times the solid content of
decrease when environmental improvement measures are imple- hydroxide). Acknowledging that the hydroxide is a main raw mate-
mented within the industry, as has been shown in the literature rial in the manufacture of sodium silicate (representing 21% of the
[4]. Thus, the present comparison between GC and CC is pertinent weight of the selected silicate solution) [60], it is observed that 56%
only under the conditions established in the present study. of GWP (and more than 56% of the environmental impact in rest of
categories) of sodium silicate is related to the production of
sodium hydroxide.
3.2. Contribution analysis
This means that, indirectly, through the sodium silicate envi-
ronmental burden, and directly, through its use as raw material,
In order to assess the most relevant life cycle processes regard-
sodium hydroxide is the main contributor to the life cycle impacts
ing environmental impacts, and to interpret the difference
of GC. Regarding sodium hydroxide (produced with sodium chlo-
between GC and CC mixes in each impact category, the contribu-
ride obtained by seawater evaporation in S1), the use of electricity
tion to the overall life cycle impacts of GC of each process is ana-
is the most relevant process (in most impact categories excepting
lyzed. Fig. 3 shows the process contribution to the environmental
ODP), followed by the chemical factory facilities characterization
impacts of G3 for S1.
(EP category), and the production process itself (ODP category).
The use of alkali activators represents the most important envi-
ronmental burden for G3. In this scenario (S1), sodium silicate bur-
den ranges from 55 to 74% in all impact categories, while sodium 3.3. Scenario analysis
hydroxide contribution ranges from 10 to 22%, as shown in
Fig. 3. Other items, such as curing (17% of GWP), transport (9% of Fig. 4 shows the scenario comparison for CC (S1) and G3 (S1, S2,
S3) mixes. Results differences between S1 and S2 are determined

Sand Zeolite Water

Sodium Hydroxide Sodium Silicate Mixing facilies
S1 (CC) S1 (G3) S2 (G3) S3 (G3)
Curing Transport Mixing process
100% 100%
9%
17%
15% 17%

74%
56% 61%
55% 58%
55%

12% 14% 10%

10% 22% 13%
0%
ADPF

EP

GWP
AP

0%
ODP

POF
EP
ADPF

AP

POF
ODP
GWP

Fig. 4. Scenarios comparison for G3 design mix (S1, S2, and S3) and CC design mix
Fig. 3. Process contribution for the G3 mix (15 MPa) for S1. (only S1).
 D.A. Salas et al. / Construction and Building Materials 190 (2018) 170–177 175

solely by the environmental performance of the energy mix con- which presents the lowest carbon footprint in the reviewed litera-
sidered for each scenario: the energy mix with a higher share of ture (approximately 89 kg CO2 eq./m3), included in their study only
hydroelectricity (S1: 2018 energy mix) results in lower environ- raw materials extraction and geopolymer production, excluding
mental impacts in most categories. On the other hand, S3 considers transport and other processes. On the other hand, McLellan et al.
that sodium hydroxide is imported from Europe. This means that a [29] analyzed different transport distances, showing that GC emis-
different energy mix is considered and that sodium chloride (for sions present a higher sensitivity to transport distances variations
sodium hydroxide production) is obtained through a different than CC. In addition, the values presented in Fig. 5 correspond to
method. concrete mixes with different mechanical properties, aimed at dif-
The difference between these sodium hydroxide sources is ferent applications. Regarding the high temperature curing pro-
caused by the higher carbon intensity of the average European cess, most studies do not include it. Yang et al. [25] estimated
energy mix compared to Ecuadorian electricity, and by the source the GHG emissions related to the steam curing process
of sodium chloride (method of production) used as raw material. (85 °C/24 h) as 38,5 CO2 eq./m3, while Turner and Collins[32] esti-
The solution/rock mining salt production method, considered in mated it (50 °C/24 h + 9 h gradual heating/cooling) as 40 kg CO2 -
S3, entails considerably higher environmental burdens than sea- eq./m3. Curing energy has been estimated by Nisbet et al. [67] for
water evaporation, considered for local production in S1 and S2. CC steam curing (54 °C/24 h) as approximately 62 MJ/m3. No other
Sodium chloride contribution to sodium hydroxide environmental estimation of GHG emissions, or energy, related to curing of
impacts ranges from 14 to 21% in most impact categories (except- geopolymers materials, or concrete, has been found in the
ing ODP: 2%) in the S3 scenario, where solution/rock mining salt is reviewed literature.
used as raw material. When seawater evaporated salt is consid- The GC under study will be further developed, as different
ered, its contribution to the hydroxide burden is lower than 1% mixes are required to widen its applications. Thus, LCA is recom-
in all impact categories (S1 and S2). Sodium hydroxide in S3 mended at later stages of development to update results and con-
includes transoceanic freight as well, but its influence is not clusions regarding the environmental performance for each
relevant. intended application. The assessment of the life cycle impacts of
As shown in Fig. 4, EP for S3 is remarkably higher than S1 and cement production in Ecuador is recommended as well, as it will
S2. This is due, mostly, to the considerably higher EP burden of allow reflecting in a more appropriate manner the potential envi-
the average energy mix for Europe. Such a high EP is derived from ronmental benefits of substituting CC with GC. The environmental
processes like lignite electricity generation (spoil from lignite min- performance of CC is sensitive to the clinker content of cement, and
ing treatment), or wastewaters from the production of construc- different blended cement mixes, which maintain the required
tion elements of biogas electricity generation plants in Europe. properties, can be developed while reducing the clinker content.
An analysis addressing this issue will allow a better understanding
3.4. Comparison with other studies of the potential benefits of using GC instead of CC.

In Fig. 5, GWP results of the present study for GC (G1, G2, and 4. Conclusions
G3 for scenario S1) and CC mixes, are compared to values found
in the literature [18,24–26,28,31–33]. G1, G2, and G3 GWP are The production of sodium hydroxide is the most relevant life
below the results of most of the available studies, excepting Weil cycle process regarding the environmental performance of GC, as
et al. [33]. CC mix result is slightly below the mean value of liter- it is a main raw material in the production of sodium silicate as
ature results for conventional concretes. Most studies lack results well. Other items such as concrete mixing facilities, curing process,
for the other impact categories. Besides, different scopes and calcu- and transport, affect few impact categories each one.
lation methodologies have been applied, and therefore a compar- Due to the high environmental burden related to electricity use
ison for the rest of the impact categories is not possible. within the life cycle of sodium hydroxide, the share of renewable
Fig. 5 shows a framework for results comparison. However, energy in the energy mix considered for the life cycle assessment
GWP results of different geopolymer cement/concrete LCA studies influences overall results. LCA results for GC considering the
are not straightforwardly comparable. For example, Weil et al. [33], 2012 energy mix (62% of hydroelectricity) are from 30 to 70%

Fig. 5. Comparison of GWP results of the present study, with those found in literature, taking into account the compressive strength considered in each case when available.
References: LG1 and LCC2 [33]; LG2, LG4, LG10, LCC1, and LCC4 [25]; LG3, LG6, LCC3, and LCC6 [18]; LG5 and LCC5 [24]; LG8 and LCC8 [28]; LG9 and LCC9 [26].
176 D.A. Salas et al. / Construction and Building Materials 190 (2018) 170–177

higher than 2018 energy mix (85% of hydroelectricity) results in all incineration in cement plants through a life cycle approach, J. Clean. Prod. 19
(2011) 1615–1621, https://doi.org/10.1016/j.jclepro.2011.05.014.
impact categories, excepting ODP, which is not influenced by elec-
[15] C. Valderrama, R. Granados, J.L. Cortina, C.M. Gasol, M. Guillem, A. Josa,
tricity use. Implementation of best available techniques in cement manufacturing: a life-
The source of sodium hydroxide influences greatly the overall cycle assessment study, J. Clean. Prod. 25 (2012) 60–67, https://doi.org/
environmental impacts, due to both, the energy mix (average Euro- 10.1016/j.jclepro.2011.11.055.
[16] C. Knoeri, E. Sanyé-Mengual, H.-J. Althaus, Comparative LCA of recycled and
pean vs. Ecuadorian energy mix) and the type of sodium chloride conventional concrete for structural applications, Int. J. Life Cycle Assess. 18
(solar salt vs. solution/rock mining salt) used for its production. (2013) 909–918, https://doi.org/10.1007/s11367-012-0544-2.
The average European energy mix performs worse environmen- [17] M. Valipour, M. Yekkalar, M. Shekarchi, S. Panahi, Environmental assessment
of green concrete containing natural zeolite on the global warming index in
tally than the Ecuadorian energy mixes, due to higher shares of marine environments, J. Clean. Prod. 65 (2014) 418–423, https://doi.org/
thermoelectricity. Besides, the characterization of local sodium 10.1016/j.jclepro.2013.07.055.
hydroxide, produced with solar salt, entails lower environmental [18] R. Rouwette, LCA of Geopolymer Concrete (E-Crete), Australia, 2012.
[19] S.H. Teh, T. Wiedmann, A. Castel, J. de Burgh, Hybrid life cycle assessment of
impacts than average European sodium hydroxide, produced with greenhouse gas emissions from cement, concrete and geopolymer concrete in
solution mining and rock mining salt. Australia, J. Clean. Prod. 152 (2017) 312–320, https://doi.org/10.1016/J.
The production of GC (G3) entails a potential environmental JCLEPRO.2017.03.122.
[20] S. Marinković, J. Dragaš, I. Ignjatović, N. Tošić, Environmental assessment of
advantage over CC if sodium hydroxide is produced with solar salt green concretes for structural use, J. Clean. Prod. 154 (2017) 633–649, https://
while considering an electricity mix with a high share of hydro- doi.org/10.1016/J.JCLEPRO.2017.04.015.
power. The GWP of GC under these conditions (G3) is 64% lower [21] A. Passuello, E.D. Rodríguez, E. Hirt, M. Longhi, S.A. Bernal, J.L. Provis, A.P.
Kirchheim, Evaluation of the potential improvement in the environmental
than that of CC. However, GC performs worse in the ODP category,
footprint of geopolymers using waste-derived activators, J. Clean. Prod. 166
due to CFC emissions during the chlor-alkali process (due to the (2017) 680–689, https://doi.org/10.1016/J.JCLEPRO.2017.08.007.
use of carbon tetrachloride), which are not present in cement pro- [22] L. Nguyen, A.J. Moseson, Y. Farnam, S. Spatari, Effects of composition and
duction. The rest of the impact categories vary in a lesser degree transportation logistics on environmental, energy and cost metrics for the
production of alternative cementitious binders, J. Clean. Prod. 185 (2018) 628–
(ADPF and EP perform slightly better; AP and POF slightly worse). 645, https://doi.org/10.1016/J.JCLEPRO.2018.02.247.
[23] L. Caldas, M. Martins, D. Lima, R. Sposto, Literature Review of Life Cycle
Assessment Applied to Green Concretes, in: ResearchGate, Cali, 2016.
5. Declaration of interest [24] G. Habert, J.B. d’Espinose de Lacaillerie, N. Roussel, An environmental
evaluation of geopolymer based concrete production: reviewing current
research trends, J. Clean. Prod. 19 (2011) 1229–1238, https://doi.org/
None.
10.1016/j.jclepro.2011.03.012.
[25] K.-H. Yang, J.-K. Song, K.-I. Song, Assessment of CO2 reduction of alkali-
Acknowledgments activated concrete, J. Clean. Prod. 39 (2013) 265–272, https://doi.org/10.1016/
j.jclepro.2012.08.001.
[26] G. Habert, J.B. D’Espinose de Lacaillerie, E. Lanta, N. Roussel, Environmental
This research was supported by Escuela Superior Politécnica del evaluation for cement substitution with geopolymers Ancona, Italy, Universita
Litoral, ESPOL (Project Code G7-DI-2014). Politecnica delle Marche, 2010.
[27] A. Heath, K. Paine, M. McManus, Minimising the global warming potential of
clay based geopolymers, J. Clean. Prod. 78 (2014) 75–83.
References [28] P.S. Matheu, K. Ellis, B. Varela, Comparing the Environmental Impacts of Alkali
Activated Mortar and Traditional Portland Cement Mortar using Life Cycle
Assessment, 2015.
[1] T.R. Naik, Sustainability of concrete construction, Pract. Period. Struct. Des.
[29] B.C. McLellan, R.P. Williams, J. Lay, A. van Riessen, G.D. Corder, Costs and
Constr. 13 (2008).
carbon emissions for geopolymer pastes in comparison to ordinary Portland
[2] E. Crossin, Comparative Life Cycle Assessment of Concrete Blends Centre for
cement, J. Clean. Prod. 19 (2011) 1080–1090, https://doi.org/10.1016/j.
Design, . RMIT University, Melbourne, 2012.
jclepro.2011.02.010.
[3] D.J.M. Flower, J.G. Sanjayan, Green house gas emissions due to concrete
[30] C. Ouellet-Plamondon, G. Habert, LCA of Alternative Concrete: How do we do?
manufacture, Int. J. Life Cycle Assess. 12 (2007) 282, https://doi.org/10.1065/
in: Boston, 2014.
lca2007.05.327.
[31] A. Petrillo, R. Cioffi, C. Ferone, F. Colangelo, C. Borrelli, Eco-sustainable
[4] D.A. Salas, A.D. Ramirez, C.R. Rodríguez, D.M. Petroche, A.J. Boero, J. Duque-
geopolymer concrete blocks production process, Agric. Agric. Sci. Procedia. 8
Rivera, Environmental impacts, life cycle assessment and potential
(2016) 408–418, https://doi.org/10.1016/j.aaspro.2016.02.037.
improvement measures for cement production: a literature review, J. Clean.
[32] L.K. Turner, F.G. Collins, Carbon dioxide equivalent (CO2-e) emissions: a
Prod. 113 (2016) 114–122, https://doi.org/10.1016/j.jclepro.2015.11.078.
comparison between geopolymer and OPC cement concrete, Constr. Build.
[5] J. Sjunnesson, Life Cycle Assessment of Concrete, Lund University, 2005.
Mater. 43 (2013) 125–130, https://doi.org/10.1016/
[6] T. Blankendaal, P. Schuur, H. Voordijk, Reducing the environmental impact of
j.conbuildmat.2013.01.023.
concrete and asphalt: a scenario approach, J. Clean. Prod. 66 (2014) 27–36,
[33] M. Weil, K. Dombrowski, A. Buchwald, Life cycle analysis of geopolymers,
https://doi.org/10.1016/j.jclepro.2013.10.012.
Woodhead Publishing, 2009.
[7] G. Habert, C. Billard, P. Rossi, C. Chen, N. Roussel, Cement production
[34] L. Provis, J.S.J. Van Deventer, Geopolymers. Structures, processing, properties
technology improvement compared to factor 4 objectives, Cem. Concr. Res.
and industrial applications, Geopolymers, Woodhead Publishing, 2009.
40 (2010) 820–826, https://doi.org/10.1016/j.cemconres.2009.09.031.
[35] C. Villa, E.T. Pecina, R. Torres, L. Gómez, Geopolymer synthesis using alkaline
[8] C. Chen, G. Habert, Y. Bouzidi, A. Jullien, Environmental impact of cement
activation of natural zeolite, Constr. Build. Mater. 24 (2010) 2084–2090,
production: detail of the different processes and cement plant variability
https://doi.org/10.1016/j.conbuildmat.2010.04.052.
evaluation, J. Clean. Prod. 18 (2010) 478–485, https://doi.org/10.1016/j.
[36] A. Nikolov, I. Rostovsky, H. Nugteren, Geopolymer materials based on natural
jclepro.2009.12.014.
zeolite, Case Stud. Constr. Mater. 6 (2017) 198–205, https://doi.org/10.1016/j.
[9] W. Chen, J. Hong, C. Xu, Pollutants generated by cement production in China,
cscm.2017.03.001.
their impacts, and the potential for environmental improvement, J. Clean. Prod.
[37] E.A. Ortega, C. Cheeseman, J. Knight, M. Loizidou, Properties of alkali-activated
(2014), https://doi.org/10.1016/j.jclepro.2014.04.048.
clinoptilolite, Cem. Concr. Res. 30 (2000) 1641–1646, https://doi.org/10.1016/
[10] T.D. Eatmon, A life-cycle assessment of Portland cement manufacturing:
S0008-8846(00)00331-8.
comparing the traditional process with alternative technologies, J. Clean. Prod.
[38] D. Bondar, C.J. Lynsdale, N.B. Milestone, N. Hassani, A.A. Ramezanianpour,
17 (2009) 668–675, https://doi.org/10.1016/j.jclepro.2008.04.007.
Effect of type, form, and dosage of activators on strength of alkali-activated
[11] R. Feiz, J. Ammenberg, L. Baas, M. Eklund, A. Helgstrand, R. Marshall, Improving
natural pozzolans, Cem. Concr. Compos. 33 (2011) 251–260, https://doi.org/
the CO2 performance of cement, part I: utilizing life-cycle assessment and key
10.1016/j.cemconcomp.2010.10.021.
performance indicators to assess development within the cement industry, J.
[39] H. Baykara, M.H. Cornejo, R. Murillo, A. Gavilanes, C. Paredes, J. Elsen,
Clean. Prod. 98 (2015) 272–281, https://doi.org/10.1016/j.jclepro.2014.01.083.
Preparation, characterization and reaction kinetics of green cement:
[12] D. García-Gusano, D. Garraín, I. Herrera, H. Cabal, Y. Lechón, Life Cycle
Ecuadorian natural mordenite-based geopolymers, Mater. Struct. 50 (2017)
Assessment of applying CO2 post-combustion capture to the Spanish cement
188, https://doi.org/10.1617/s11527-017-1057-z.
production, J. Clean. Prod. (2013), https://doi.org/10.1016/j.
[40] N. Ulloa, H. Baykara, A. Rigail, M.H. Cornejo, J.L. Villalba, An investigation of the
jclepro.2013.11.056.
effect of migratory type corrosion inhibitor on mechanical properties of
[13] C. Li, S. Cui, Z. Nie, X. Gong, Z. Wang, N. Itsubo, The LCA of portland cement
zeolite-based novel geopolymers, J. Mol. Struct. 1146 (2017) 814–820, https://
production in China, Int. J. Life Cycle Assess. 20 (2014) 117–127, https://doi.
doi.org/10.1016/j.molstruc.2017.06.066.
org/10.1007/s11367-014-0804-4.
[41] N.A. Ulloa, H. Baykara, M.H. Cornejo, A. Rigail, C. Paredes, J.L. Villalba,
[14] C. Strazza, A. Del Borghi, M. Gallo, M. Del Borghi, Resource productivity
Application-oriented mix design optimization and characterization of
enhancement as means for promoting cleaner production: analysis of co-
 D.A. Salas et al. / Construction and Building Materials 190 (2018) 170–177 177

zeolite-based geopolymer mortars, Constr. Build. Mater. 174 (2018) 138–149, Standards. I: LCA in Perspective. IIa: Guide. IIb: Operational Annex. III:
https://doi.org/10.1016/J.CONBUILDMAT.2018.04.101. Scientific Background, Kluwer Academic Publishers, Dordrecht, 2002.
[42] D. Garcés, Caracterización mineralógica del depósito de zeolitas naturales en el [55] NTE INEN 643. Hollow Blocks of Concrete. Requirements, 2014.
río Guaraguau (Isidro Ayora, provincia del Guayas) y su aplicación en la [56] CONELEC, Plan, Maestro de Electrificación 2013–2022, 2013.
remoción de amonio en aguas residuales, ESPOL, 2013. [57] M. Shibasaki, M. Fischer, L. Barthel, Effects on life cycle assessment — scale up
[43] L. Machiels, D. Garcés, R. Snellings, W. Vilema, F. Morante, C. Paredes, J. Elsen, of processes, in: S. Takata, Y. Umeda (Eds.), Adv. Life Cycle Eng. Sustain. Manuf.
Zeolite occurrence and genesis in the Late-Cretaceous Cayo arc of Coastal Businesses, Springer, London, 2007, pp. 377–381.
Ecuador: evidence for zeolite formation in cooling marine pyroclastic flow [58] F. Piccinno, R. Hischier, S. Seeger, C. Som, From laboratory to industrial scale: a
deposits, Appl. Clay Sci. 87 (2014) 108–119, https://doi.org/10.1016/ scale-up framework for chemical processes in life cycle assessment studies, J.
j.clay.2013.10.018. Clean. Prod. 135 (2016) 1085–1097, https://doi.org/10.1016/j.
[44] L. Machiels, F. Morante, R. Snellings, B. Calvo, L. Canoira, C. Paredes, J. Elsen, jclepro.2016.06.164.
Zeolite mineralogy of the Cayo formation in Guayaquil, Ecuador, Appl. Clay Sci. [59] F. Borbor, S. Flores, M. Padilla, G. Bastidas, Creación de una Fábrica de Bloques
42 (2008) 180–188, https://doi.org/10.1016/j.clay.2008.01.012. en la Ciudad de, ESPOL, Guayaquil, 2009.
[45] F. Morante, LAS ZEOLITAS DE LA COSTA DE ECUADOR (GUAYAQUIL): [60] M. Fawer, M. Concannon, W. Wolfram Rieber, Life Cycle Inventories for the
GEOLOGÍA, CARACTERIZACIÓN Y APLICACIONES, Universidad Politénica de Production of Sodium Silicates, EMPA, St Gall, 1997.
Madrid, 2004. [61] I. Garcia-Herrero, M. Margallo, R. Onandía, R. Aldaco, A. Irabien, Environmental
[46] R. Robayo-Salazar, J. Mejía-Arcila, R. Mejía de Gutiérrez, E. Martínez, Life cycle challenges of the chlor-alkali production: seeking answers from a life cycle
assessment (LCA) of an alkali-activated binary concrete based on natural approach, Sci. Total Environ. 580 (2017) 147–157, https://doi.org/10.1016/j.
volcanic pozzolan: a comparative analysis to OPC concrete, Constr. Build. scitotenv.2016.10.202.
Mater. 176 (2018) 103–111, https://doi.org/10.1016/J. [62] I. Garcia-Herrero, M. Margallo, R. Onandía, R. Aldaco, A. Irabien, Connecting
CONBUILDMAT.2018.05.017. wastes to resources for clean technologies in the chlor-alkali industry: a life
[47] ASTM International, ASTM C109/C109M-16a, Standard Test Method for cycle approach, Clean Technol. Environ. Policy (2017) 1–14, https://doi.org/
Compressive Strength of Hydraulic Cement Mortars (Using 2-in. or [50-mm] 10.1007/s10098-017-1397-y.
Cube Specimens) West Conshohocken, PA, 2016. [63] V. Sedivy, Environmental balance of salt production speaks in favour of solar
[48] ISO, Environmental Management- Life Cycle Assessment- Principles and saltworks, Glob. Nest J. 11 (2009) 41–48.
Framework, International Organisation for Standardisation, Geneva, [64] F. Goetzfried, B. Stratmann, Q. Dietlinde, Life Cycle Assessment of Sodium
Switzerland, 2006. Chloride Production and Transport, in: Sevilla, Spain, 2012.
[49] ISO, Environmental Management-Life Cycle Assessment-Requirements and [65] ASTM C305-14, Standard Practice for Mechanical Mixing of Hydraulic Cement
Guidelines, International Organisation for Standardisation, Geneva, Pastes and Mortars of Plastic Consistency, ASTM International, West
Switzerland, 2006. Conshohocken, PA, 2014.
[50] Pre Consultants, SIMAPRO, <https://simapro.com/>, 2014. [66] J.K. Cable, Evaluation of Mix Time on Concrete Consistency and Consolidation,
[51] H.-J. Althaus, M. Chudacoff, R. Hischier, N. Jungbluth, M. Osses, A. Primas, Life 1998.
Cycle Inventories of Chemicals. ecoinvent report No. 8, v2.0, EMPA Dübendorf, [67] M.A. Nisbet, M.L. Marceau, M.G. VanGeem, Environmental Life Cycle Inventory
Swiss Centre for Life Cycle Inventories, Dübendorf, CH, 2007. of Portland Cement Concrete, 2002.
[52] G. Wenet, C. Bauer, B. Steubing, J. Reinhard, E. Moreno-Ruiz, B. Weidema, The [68] F. Colangelo, G. De Luca, C. Ferone, A. Mauro, Experimental and numerical
ecoinvent database version 3 (part I): overview and methodology, Int. J. Life analysis of thermal and hygrometric characteristics of building structures
Cycle Assess. 21 (9) (2016) 1218–1230. employing recycled plastic aggregates and geopolymer concrete, Energies 6
[53] National Renewable Energy Laboratory, U.S. Life Cycle Inventory Database, (2013) 6077–6101, https://doi.org/10.3390/en6116077.
<https://www.nrel.gov/lci/>, 2012. [69] T. Bergman, A. Lavine, F. Incropera, D. Dewitt, Fundamentals of Heat and Mass
[54] M.A.J. Guinée, J.B. Gorrée, M. Heijungs, R. Huppes, G. Kleijn, R. Koning, A.D. Transfer, 7th ed.,., John Wiley & Sons Inc, U.S.A, 2011.
Oers, L.V. Sleeswijk, A. Suh, S. Udo de Haes, H.A. Bruijn, H.d. Duin, R.V. [70] Atmospheric Emissions from Chlor-Alkali Manufacture, Environmental
Huijbregts,, Handbook on Life Cycle Assessment. Operational Guide to the ISO Protection Agency: Air Pollution Control Office, North Carolina, 1971.