Content uploaded by Paula Pérez-López
Author content
All content in this area was uploaded by Paula Pérez-López on Aug 29, 2017
Content may be subject to copyright.
1
Comparative life cycle assessment of real pilot reactors for microalgae
cultivation in different seasons
Paula Pérez-López*1,2, Jeroen H. de Vree3,4, Gumersindo Feijoo1, Rouke Bosma3, Maria J.
Barbosa3, María Teresa Moreira1, René H. Wijffels3,5, Anton J.B. van Boxtel6, Dorinde M.M.
Kleinegris4,7
1 Department of Chemical Engineering, Institute of Technology, University of Santiago de Compostela. 15782 -
Santiago de Compostela, Spain
2 MINES ParisTech, PSL Research University, Centre Observation, Impacts, Energie (O.I.E.), 1 rue Claude
Daunesse, CS 10207, 06904 Sophia Antipolis Cedex, France
3 Wageningen University, Bioprocess Engineering Group, AlgaePARC, P.O. Box 16, 6700 AA, Wageningen, The
Netherlands
4 Uni Research Environment, P.O. Box 7810, 5020 Bergen, Norway
5 University of Nordland, Faculty of Biosciences and Aquaculture, N-8049, Bodø, Norway
6 Wageningen University – Biobased Chemistry and Technology, P.O. Box 17, 6700 AA - Wageningen, The
Netherlands
7 Wageningen UR Food & Biobased Research, AlgaePARC, P.O. Box 16, 6700 AA, Wageningen, The Netherlands
* Corresponding author: Tel.: +33 (0) 4 97 15 70 55; E-mail address: paula.perez_lopez@mines-paristech.fr
Abstract
Microalgae are promising natural resources for biofuels, chemical, food and feed products.
Besides their economic potential, the environmental sustainability must be examined. Cultivation
has a significant environmental impact that depends on reactor selection and operating
conditions. To identify the main environmental bottlenecks for scale-up to industrial facilities this
study provides a comparative life cycle assessment (LCA) of open raceway ponds and tubular
photobioreactors at pilot scale. The results are based on experimental data from real pilot plants
operated in summer, fall and winter at AlgaePARC (Wageningen, The Netherlands). The energy
consumption for temperature regulation presented the highest environmental burden. The
production of nutrients affected some categories. Despite limited differences compared to the
vertical system, the horizontal PBR was found the most efficient in terms of productivity and
environmental impact. The ORP was, given the Dutch climatic conditions, only feasible under
summer operation. The results highlight the relevance of LCA as a tool for decision-making in
2
process design. Weather conditions and availability of sources for temperature regulation were
identified as essential factors for the selection of geographic locations and for microalgal
cultivation systems based on environmental criteria. Simulation of large-scale reactors with
optimized temperature regulation systems lead to environmental improvements and energy
demand reductions ranging from 17% up to 90% for systems operated in favorable summer
conditions.
Keywords Microalgae cultivation, Life cycle assessment (LCA), tubular photobioreactor, open
raceway pond, pilot plant, weather variations
1. Introduction
The scarcity of natural resources and
particularly the exhaustion of fossil fuels are a
global challenge to be addressed in forthcoming
decades [1-3]. The current production framework
based on non-renewable energies poses several
problems to society, including economic
instability and political conflicts (due to raw
material scarcity and related increasing prices), as
well as environmental concerns [4]. Alternative
sources including biomass feedstocks such as
vegetable oils, waste oils or algal lipids are
currently under development to reduce harmful
effects on environment and ecological threats such
as global warming [2, 5-6].
Microalgae have shown a great potential for
the production of numerous compounds with a
wide variety of applications that include biofuels
and other forms of energy as well as chemicals,
food and feed, among others [1, 7-8]. Some
advantages of microalgae compared to other
bioenergy feedstocks are their higher productivity
per unit, the possibility to cultivate them on
marginal land in fresh- or saltwater avoiding
competition with food crops and the option of
coupling their growth with the treatment of waste
streams [2, 5, 9-12]. Despite the advantages of
microalgae and their lower requirements in
categories such as land competition or
eutrophication [5, 13], some aspects of
environmental sustainability, such as the energy
balance or greenhouse gas emissions, are still
liable to improve, especially for the use of
microalgae for energy applications [3, 5]. Life
Cycle Assessment (LCA) has the potential to be
used as a guiding tool for decision-making in
process design [8, 14]. LCA may contribute to
identify the main bottlenecks to be addressed
during the scale up towards sustainable industrial
facilities.
Algae cultivation has been identified as a
major contributor to the operational and embodied
energy of microalgal processes [3, 7, 15]. The
total energy demand for cultivation stage usually
ranges from 0.1 up to 5 MJ energy input per MJ
energy produced [9, 16]. This is mainly due to
3
addition of nutrients and CO2 [1, 5] and the
specific requirements of the selected reactor
configuration, such as mixing and temperature
control [9, 15, 17].
The embodied energy for nutrients is related
to manufacturing of synthetic fertilizers and the
reactor materials as well as CO2 production,
whereas the operational energy consumption is
linked to pumping for the delivery of culture
medium and CO2 [1, 5, 9]. Since algae are
temperature sensitive, heating and cooling is
required to operate relatively close to the optimal
temperature of the algal species. Temperature
regulation allows high productivities and prevents
growth inhibition, but may increase the energy
demand of the process [18-19]. The integration of
options such as waste heat from power generation
or cold water resources allows reducing the energy
requirements for heating or cooling the water from
room temperature to the set point temperature, and
thus, contributes to the optimization of the
cultivation stage [19-20]. Furthermore, climatic
data including irradiation and temperature depend
on geographic location. Therefore, the heating and
cooling needs of the system vary between
locations [21]. The selection of an appropriate
location according to available resources (energy,
nutrients, waste heat, cooling water) and algal
strains may serve to maintain the optimal
temperature with low heating and cooling
requirements so that the energy consumption is
minimized.
The environmental performance of
cultivation is also influenced by reactor selection.
Open raceway ponds (ORPs) and closed tubular
photobioreactors (PBRs) are currently considered
as the two most feasible existing configurations
for large-scale cultivation of microalgae [10, 15].
Although simple reactors such as ORPs have
fewer elements that consume energy than closed
PBRs, the maximum biomass productivity is also
lower and they are more sensitive to
contamination risks. However, closed PBRs have
higher costs of infrastructure and operation [15,
22]. These aspects should be considered when
comparing the environmental performance of
different configurations.
Numerous studies dealing with the
environmental performance of different reactor
designs for microalgae cultivation have been
published [1, 5, 9, 12, 15, 17, 20, 23-24].
However, most of this work considers
hypothetical simulated scenarios or extrapolations
from lab-scale data rather than existing pilot or
commercial systems [1, 5, 8, 12, 15, 17, 24]. Few
of them make a comparison between different
reactor configurations, often restricted to a very
limited set of indicators that only takes into
account energy requirements and greenhouse gas
emissions [1, 9, 15, 17]. Moreover, they are based
on average growth parameters and omit the
influence of weather fluctuations, affecting reactor
stability, on the environmental results.
This work provides a comparative life cycle
assessment (LCA) of the two most common
reactor configurations (ORPs and tubular PBRs)
to evaluate the main environmental burdens of
each option, to compare their performances and to
4
identify bottlenecks for up-scaling. The evaluation
considers the algal biomass production from the
eustigmatophyte Nannochloropsis sp. due to its
good biomass productivity and capability for high
lipid content when stressed [13, 25]. The
evaluation is based on data from three real
reactors operated in parallel at AlgaePARC pilot
facility (Wageningen, The Netherlands) [26]. The
use of real pilot data is expected to overcome
current concerns of microalgal LCAs related to
the lack of large-scale information [14]. The data
are obtained for comparable weather conditions
for each reactor, during three seasons (summer,
fall and winter). These systems have been
designed and operated as a first step to facilitate
the transition from laboratory research to outdoor
production for industrial applications [18].
2. Materials and methods
2.1. AlgaePARC cultivation systems
AlgaePARC is a research facility of
Wageningen University and Research (The
Netherlands) that was built with the aim of
comparing different PBRs and optimizing process
control strategies for microalgae cultivation and
processing. The main objective of this facility is to
develop systems with low production costs and
energy requirements that can serve as a basis for
the improvement of large-scale microalgae plants
[18].
AlgaePARC outdoor facilities comprise
several pilot-scale photobioreactors, including
horizontal and vertical tubular PBRs and an ORP.
Among the available systems, the operation of the
ORP, a vertical and an horizontal tubular PBR was
monitored throughout the year 2013. Part of these
data on photosynthetic efficiency, areal and
volumetric productivities have been published in
De Vree et al. [26]. The layout of each system is
depicted in Figure 1. As described in Bosma et al.
[18], the ORP consists of a 4.73 m3 oval pond with
a separation plate in the center and two additional
plates that divide each of the rounded corners into
three narrower channels to improve mixing. A
paddle wheel drives the culture at a controlled
speed of the motor. The CO2 is dosed in a gas
circulation loop, by injection at the bottom of the
pond under a gas collection hood. The hood traps
the CO2 enriched air, minimizing CO2 losses.
Liquid culture medium is pumped from nutrient
dosing stations close to the ORP and temperature
is maintained above a set point with a submerged
tubular heat exchanger. Active cooling was not
needed.
The horizontal tubular PBR system (0.56 m3)
consists of three loops with transparent pipes that
are placed in parallel at the same height, whereas
the vertical system (1.06 m3) is composed of
seven loops with pipes “stacked” on top of each
other. In both systems, the culture medium is
divided over the loops by a distribution header.
The accumulation of excess oxygen is avoided by
using an oxygen stripper. This system receives air,
blown by a compressor via a sparger at the
bottom. Oxygen is transferred from the liquid to
the gas phase and leaves the stripper at the top.
The stripper contains three heat exchange spirals
5
to keep the culture temperature between a
minimum and a maximum value.
The operation of the three systems was
performed under different Dutch weather
conditions that can be classified as “summer”,
“fall” and “winter”. However, the “winter”
operation of the ORP was unfeasible as heavy
rainfall in this period resulted in a too high
dilution of the ORP in combination with the low
solar radiation level and low temperatures.
Geometric and average operating parameters in
each reactor and season are specified in Table 1.
a)
b)
c)
d)
e)
5
6
1
4
①
PMMA
transparent tubes
②
Polypropylene
opaque
straight
tubes
③
Polypropylene
opaque
corners
④
Aluminum
support structure
⑤
Oxygen
stripper
⑥
Internal
heat exchanger
①
Main
polypropylene body
④
Heat
exchanger
②
Middle
separation
⑤
CO
2supply system
③
Paddlewheel
and steel shaft
⑥
Channels
to improve
mixing
2
3
4
2
1 3
5
6
Figure 1. Pilot-scale systems at AlgaePARC facilities: a) Horizontal tubular PBR, b) vertically stacked
tubular PBR c) ORP, d) simplified scheme of the main components of the tubular reactors and e) simplified
scheme of the main components of the ORP (excluding, pumps, compressors and nutrient dosage tanks).
6
Table 1. Dimensions and operational average parameters for each reactor and period
Reactor configuration
Horizontal PBR
Vertical PBR
ORP
Season
Summer
Fall
Winter
Summer
Fall
Winter
Summer
Fall
Reactor characteristics
Volume (m3)
0.56
1.06
4.73
Ground area (m2)
27.0
31.0
25.4
Optical path (cm)
4.6
4.6
20
Operational parameters
Start-up date
04/07/13
29/08/13
04/11/13
04/07/13
27/08/13
07/11/13
08/07/13
22/08/13
Shutdown date
22/08/13
31/10/13
16/12/13
27/08/13
04/11/13
17/12/13
22/08/13
18/09/13
Dilution rate (%)
25.3
23.5
11.6
27.1
25.4
14.0
16.0
11.0
Biomass concentration in
the reactor (g/L)
2.5
1.3
0.6
1.9
1.0
0.4
0.5
0.5
Ground areal
productivity
(g·m-2·d-1)
12.1
4.6
1.2
19.4
8.3
2.7
10.5
2.1
Volumetric productivity
(g·m-3·d-1)
654.8
250.2
66.7
568.5
241.7
79.2
56.5
11.3
Recirculation flow
(m3·h-1)
8.1
8.1
8.0
18.7
18.8
18.7
0
0
Air flow (m3·h-1)
5.6
6.0
6.0
11.6
10.0
9.7
0
0
CO2 flow (L·h-1)
40.1
21.3
8.8
53.4
26.1
14.0
30.3
30.3
Weather conditions
Temperature outside (ºC)
21.4
14.4
5.8
21.4
14.2
5.8
21.7
17.5
Light intensity
(mol photons·m-2·d-1)
36.9
18.6
5.0
36.9
18.6
4.9
38.2
23.9
Photosynthetic
efficiency, PE (%)
1.5
1.1
1.0
2.5
2.1
2.6
1.1
0.3
Mean sunlight duration
calculated from global
radiation (h)1
7.7
4.3
2.1
7.7
4.2
2.1
7.5
5.1
Mean daily precipitation
(mm) 1
1.2
4.3
2.9
1.1
4.5
2.0
1.3
3.9
Mean precipitation
duration (h)
0.6
2.3
2.2
0.6
2.4
1.6
0.7
2.2
Mean daily cloud cover
(in octans) 1
4.0
5.5
6.3
4.0
5.6
6.3
4.1
5.3
Mean daily relative
humidity (%)1
76
85
89
76
84
89
76
81
1Calculated from data of the Royal Dutch Meteorological Institute (http://www.knmi.nl/index_en.html), for De Bilt
weather station
2.2. Life cycle assessment methodology
Life Cycle Assessment (LCA) was used in
the comparative evaluation of the environmental
aspects and potential impacts of algae cultivation.
LCA allowed a systematic evaluation of the
environmental performance of the studied systems
throughout the whole process chain, from raw
materials extraction to microalgal cultivation and
waste disposal. Following the ISO 14040
standards [27], four sequential stages were
undertaken, namely i) goal and scope definition,
ii) inventory analysis, iii) impact assessment and
iv) interpretation of the results. The procedures
7
and assumptions considered for each stage are
detailed below.
2.2.1. Goal and scope of the study
The goal of the present study was to assess
three different reactor configurations for the pilot-
scale production of Nannochloropsis sp in three
periods of the year (summer, fall and winter). With
this work, bottlenecks in environmental
performance (referred to as hot spots) of the
systems were identified.
Since the environmental performance of the
different reactors is linked to the biomass
production, the life cycle inventory and impact
assessment are referred to 1 kg of produced
biomass dry weight, contained in a 22% DW
slurry, as the functional unit (FU).
The system boundaries were divided into
foreground and background processes, referring to
the steps that are directly affected by the study,
and the processes that supply energy and materials
via a homogeneous market, respectively [28]. The
processes in the foreground systems were
classified in four subsystems, shown in Figure 2:
i) cleaning of the reactor, ii) preparation of the
culture medium, iii) cultivation and iv) biomass
concentration.
i) Cleaning of the reactor (S1)
In the first stage of the process, tap water was
pumped to a silo (6 m3) and sterilized with
hypochlorite (2 mg/L) and passed through
activated carbon filters to remove the hypochlorite
before being supplied to the systems.
Before inoculation, each reactor was flushed
with sterilized tap water to ensure the absence of
competing algae and protozoa. In the case of the
tubular PBRs, the system was rinsed with tap
water three times. In the second rinse, 3% of a
disinfection agent (containing hydrogen peroxide)
was added, whereas in the third one, 0.5 g/L
plastic beads for biofilm removal were used. For
the ORP, tap water in a quantity equal to three
times the usable volume of the reactor and a
vacuum cleaning system was used for 1-2 h after
the last washing step to remove all water.
ii) Preparation of the culture medium (S2)
The main source of nutrients for the
cultivation of Nannochloropsis was natural
seawater. To avoid contamination, this seawater
was sterilized in an analogous manner as tap water
for cleaning by adding hypochlorite (5 mg/L) and
removing the chlorine with active carbon. Then, it
was passed through a cascade filter (10 m, 5 m
and 1 m) and supplied to the systems.
Additionally, culture medium with NaNO3 as
the main nitrogen source was supplemented to the
reactor. The nitrate solution consisted of 212 g/L
NaNO3, 11.5 g/L KH2PO4, 3 g/mL Na2EDTA, 50
mL of trace mineral stock solution (see
Supplementary material) and 17.5 mL NaOH 4 M
to adjust pH. For the tubular PBRs, a dosage of 10
mL of nitrate solution per L of seawater was added
in the final culture medium, whereas 2 mL/L were
used in the medium of the ORP.
iii) Cultivation (S3)
This stage consisted of a semi-continuous
process in which the biomass was operated with a
fixed daily dilution rate and harvested according
to the scheme explained by de Vree et al. [26]. As
8
no source of artificial light was provided, light
intensity only depended on weather conditions. To
maintain temperature close to the optimal
temperature of the species, a central chiller and
heater were used. To minimize the energy
consumption, the culture temperature varied
between 20 and 30ºC. Heating was applied to
prevent temperatures below 20ºC and cooling
when culture temperature rose above 30ºC. The
set point was selected according to
Nannochloropsis optimal range of 20-30°C [29].
No active cooling was needed for the ORP since
cooling in this system occurs naturally by water
evaporation [18]. The purity of the culture was
checked microscopically (Leica Laborlux S) three
times per week to minimize contamination
problems.
iv) Biomass concentration (S4)
Despite similar environmental conditions, the
final biomass concentration varied for a given
period depending on the cultivation system and
seasonally due to the different weather conditions.
In order to make relevant comparisons between
the performances of the different reactors,
microfiltration and centrifugation were applied to
obtain a defined biomass concentration (22%
DW), regardless of the reactor system and season.
The life cycle inventory and determined
environmental impact assessment are referred to 1
kg of produced biomass (dry weight), so 4.55 kg
of slurry (22% DW).
9
FOREGROUND SYSTEM
SUBSYSTEM S3: CULTIVATION
AIR, SOIL & WATER
EMISSIONS
BACKGROUND SYSTEM
CHEMICALS (Nutrients&
Cleaningagents) MATERIALS OF
THE REACTOR ELECTRICITYAIR SUPPLY
HARVES TED
BIOMASS
WASTE TO
TREATMENT
WATER SUPPLY
SUBSYSTEM S2: PREPARATION OF THE
CULTURE MEDIUM
NUTRIENT
ADDITION MEDIUM
PUMPING
FILTRATION
CHLORINE
ADDITION
SUBSYSTEM S4 : BIOMASS CO NCENTRATION
SUBSYSTEM S1: CLEANING OF THE REACTOR
CHLORINE
ADDITION
MICROFILTRATION CENTRIFUGATION
RAW MATERIALS FOSSIL FUELSWATER
Nutrients
Electricity
FILTRATION THROUGH
ACTIVATED CARBON
Electricity
Electricity
Tap
water Chemicals
Electricity
Residual culture medium (to
wastewater treatment)
CLEANING WITH VACUUM
SYSTEM (ONLY FOR ORP)
Electricity
Chlorine
solution
ADDITION OF PLASTIC BEADS
(ONLY FOR TUBULAR PBRs)
Tap water
RINSE WITH FREEBAC
CLEAROXYL SOLUTION Plastic beads
Natural
seawater
Electricity
Chlorine solution Elec tricity
TEMPERATURE CONTROL
SYSTEM (COOLING/HEATING)
Electricity
CULTIVATION IN
DEMO-SCALE REACTOR
ADDITION OF
INOCULUM MIXING
CO2ADDITION
Electricity
Electricity
Electricity
TRANSP ORT
Electricity
Residual culture medium (to
wastewater treatment)
Figure 2. Process value chain and system boundaries of the cultivation of Nannochloropsis sp. in pilot-scale reactors
at AlgaePARC facilities.
10
2.2.2. Inventory analysis: Data acquisition
and assumptions
The collection of inventory data related to
the significant inputs and outputs to the system
under study is an essential step of LCA
methodology. In this study, the information for the
foreground system mainly consisted of primary
data collected in the facility.
The inputs and outputs for the cleaning stage
(S1) were quantified with respect to the volume of
water, by assuming a water consumption equal to
three times the total volume of the corresponding
reactor for each cleaning event. During the full
monitoring period, the tubular PBRs were cleaned
once after each of the six performed cultivations.
For the ORP the length of each cultivation period
was shorter (due to several washouts resulting
from unfavorable weather conditions and high
dilution rates at some periods of the operations)
and 10 cleaning cycles were applied in total. The
total quantity of inputs for each evaluated period
was estimated according to the ratio between the
duration of the period and the total feasible
operation time per year (approximately 10
months).
Similarly, the water and chemicals for the
preparation of the culture medium (S2) were
calculated by considering the average dilution rate
of each period. The energy requirements were
estimated with respect to the total seawater and
medium, assuming that the equipment was
operating at the maximum allowed capacity.
Regarding the cultivation stage (S3), the
energy consumption for the different operations
(base energy of monitoring system, mixing,
aeration and temperature control) was directly
obtained from the on-line monitoring system. The
energy consumption of the microscope for the
purity check was considered negligible compared
to the requirements of the reactors. The quantities
of building materials for each reactor were
calculated from measurements of the dimensions,
which allowed determining the volume of each
component and obtaining the weight by
multiplying by the corresponding density. Life
spans of 10 or 20 years were considered for the
building materials depending on their properties
and function (see Table S1 in Supplementary
Material).
The inputs for the biomass concentration (S4)
were the energy consumptions for the consecutive
units of microfiltration and centrifuge, which were
calculated according to the total volume of
medium to separate from the biomass in order to
achieve the final 22% DW concentration.
In all subsystems, the solid wastes were
assumed to be disposed in either sanitary or inert
landfills, whereas the resulting wastewater was
collected in the general sewage system and treated
in a conventional wastewater treatment plant. An
average transport distance of 200 km was
considered for chemicals and building materials
and 50 km was estimated for wastes. No seawater
transport was considered, since it was assumed
that a commercial scale facility would be placed
close to the coast. Materials for auxiliary
equipment used in the process (filters, pumps,
centrifuge, etc.) were neglected, since this
11
equipment was shared between several systems
and the corresponding quantities for each system
after applying the appropriate allocation
procedures would be very limited. Moreover, the
equipment was common to the three analyzed
reactors and thus, no additional information for
the comparative purposes of the present work
would be obtained.
The inventory data related to the background
system were obtained from Ecoinvent database
[30]. These inputs include the production of the
chemicals required for the cleaning and the
nutrients for the culture medium, as well as the
production of electricity used throughout the
stages of the processes, the manufacture of the
building materials for each reactor and the waste
disposal. With regard to NaNO3 production, this
process is not defined in the Ecoinvent database.
Therefore, the considered inventory data
correspond to the synthetic process as described
by Pérez-López et al. [7].
No allocation procedure was required,
according to the goal of the study (the comparison
of the different reactor configurations). Thus, all
the environmental burdens were allocated to the
total quantity of biomass harvested from each
reactor.
The inventory data of the assessed scenarios
are shown in Table 2.
Table 2. Inventory table for the cultivation of microalgae Nannochloropsis in pilot-scale systems (FU: 1 kg DW
harvested biomass in 22% DW slurry).
Reactor configuration
Horizontal PBR
Vertical PBR
ORP
Season
Summer
Fall
Winter
Summer
Fall
Winter
Summer
Fall
INPUTS from TECHNOSPHERE
Materials
S1. Cleaning
Tap water (L)
128
398
994
163
439
977
2959
15984
Chlorine solution (g)
4.49
13.9
34.8
5.70
15.4
34.2
104
559
Disinfectant (kg)
1.56
4.84
12.1
1.98
5.34
11.9
0
0
Plastic beads (g)
21.4
66.3
166
27.1
73.2
163
0
0
S2. Nutrient supply
Chlorine solution (g)
17.8
43.3
88.6
21.9
47.9
90.1
145
649
Deionized water (kg)
3.85
9.33
19.1
4.71
10.3
19.4
6.27
28.0
FeSO4∙7H2O (g)
6.45
15.7
32.1
7.90
17.3
32.6
10.5
47.0
MnCl2∙2H2O (g)
0.368
0.893
1.83
0.451
0.988
1.86
0.600
2.68
ZnSO4∙7H2O (g)
0.142
0.344
0.705
0.174
0.381
0.717
0.231
1.03
Co(NO3)2∙6H2O (g)
0.0151
0.0365
0.0748
0.0184
0.0405
0.0760
0.0245
0.110
CuSO4∙5H2O (g)
0.0051
0.0125
0.0256
0.0063
0.0139
0.0261
0.0084
0.0376
Na2MoO4∙2H2O (g)
0.0521
0.126
0.259
0.0638
0.140
0.263
0.0849
0.379
Na2EDTA∙2H2O (g)
22.6
54.8
112
27.7
60.7
114
36.8
164
NaNO3 (kg)
0.912
2.21
4.53
1.12
2.45
4.61
1.49
6.64
KH2PO4 (g)
49.5
120
246
60.6
133
250
80.7
360
NaOH (g)
13.8
33.4
68.4
16.9
37.0
69.5
22.4
100
S3. Cultivation
Biomass (g)
84.1
197
210
100
189
206
399
3045
12
PMMA (kg)
0.122
0.380
0.951
0.301
0.813
1.81
0.012
0.066
PP (kg)
0.077
0.239
0.598
0.068
0.182
0.406
1.49
8.02
Steel (kg)
0
0
0
0.015
0.040
0.089
0.136
0.737
Aluminum (kg)
0.192
0.594
1.49
0.136
0.368
0.819
0
0
Synthetic rubber (g)
0
0
0
8.78
23.7
52.7
4.32
23.3
Compressed air (m3)
479
1363
4653
503
1125
2940
0
0
Carbon dioxide (m3)
3.44
4.83
6.82
2.31
2.94
4.25
3.43
22.7
Energy
S1. Cleaning
Filtration (kWh)
0.026
0.082
0.204
0.033
0.090
0.201
0.608
3.29
Vacuum system (kWh)
0
0
0
0
0
0
1.72
9.29
S2. Nutrient supply
Water pumping (kWh)
0.118
0.287
0.588
0.145
0.318
0.597
0.964
4.31
Filtration (kWh)
0.140
0.340
0.697
0.172
0.377
0.708
1.14
5.10
Mixing (kWh)
0.029
0.070
0.143
0.035
0.077
0.145
0.047
0.209
S3. Cultivation
Base energy (kWh)
14.6
38.6
132
8.22
21.5
57.8
5.67
37.5
Aeration and CO2 (kWh)
51.5
136
466
20.3
53.1
143
26.1
173
Mixing (kWh)
21.7
60.4
139
33.3
89.4
120
89.5
619
Heating (kWh)
35.3
286
2496
61.0
549
3267
198
4873
Cooling (kWh)
157
46.7
0
152
34.5
0
0
0
S4. Biomass concentration
Microfiltration (kWh)
0.176
0.428
0.875
0.216
0.473
0.890
1.44
6.42
Centrifugation (kWh)
0.199
0.484
0.991
0.244
0.536
1.01
1.63
7.26
Transport
S1. Cleaning
Chemicals (kg·km)
313
970
2425
397
1072
2384
20.7
112
Materials (kg·km)
4.28
13.3
33.1
5.43
14.6
32.6
0
0
Wastes (kg·km)
1.07
3.31
8.28
1.36
3.66
8.14
0
0
S2. Nutrient supply
Chemicals (kg·km)
205
496
1016
251
450
1033
357
1593
S3. Cultivation
Materials (kg·km)
78.3
243
607
106
286
635
328
1770
Wastes (kg·km)
19.6
60.7
152
26.5
71.4
159
81.9
442
INPUTS from ENVIRONMENT
S2. Nutrient supply
Seawater (L)
426
1033
2116
522
1144
2151
3500
15628
S3. Cultivation
Occupation, land (m2year)
0.385
1.19
2.98
0.265
0.714
1.59
0.592
2.86
OUTPUTS to TECHNOSPHERE
Product
Microalgal biomass (kg)
1
1
1
Wastes
S1. Cleaning
Plastic beads to landfill (g)
21.4
66.3
166
27.1
73.2
163
0
0
Wastewater to treatment
plant (L)
130
402
1004
165
444
987
2959
15984
S3. Cultivation
PMMA (kg)
0.122
0.380
0.951
0.302
0.813
1.81
0.012
0.066
PP (kg)
0.077
0.239
0.598
0.068
0.182
0.406
1.49
8.02
Steel (kg)
0
0
0
0.015
0.040
0.089
0.136
0.737
Aluminum (kg)
0.192
0.594
1.49
0.136
0.368
0.819
0
0
Synthetic rubber (g)
0
0
0
8.78
23.7
52.7
4.32
23.3
S4. Biomass concentration
13
Wastewater to treatment
plant (L)
425
1038
2131
522
1150
2166
3502
15652
2.2.3. Life cycle impact assessment
The environmental profile of the described
systems was assessed by performing the
classification and characterization stages of the
LCA methodology [27]. Normalization and
weighting were not conducted as these optional
elements were not considered to provide relevant
information for the objectives of the study. Two
methodologies were used: CML 2001, reported by
the Centre of Environmental Science of Leiden
University [31] and the Cumulative Energy
Demand (CED) based on the method published by
Ecoinvent version 2.0 and expanded by PRé
Consultants [32].
The impact categories were selected
according to the most relevant environmental
issues related to microalgal products and used in
previous LCA studies [7, 23, 33-34]. Eleven
potential impact categories (specified in Table 3)
were evaluated according to the CML 2001
methodology. CED methodology was applied to
obtain the total energy consumption throughout
the whole process (Total CED), which included
three categories of non-renewable sources of
energy and three categories of renewable sources.
The inventory data were implemented in the
software SimaPro 8 [35].
3. Results and discussion
3.1. Identification of hot spots for each reactor
configuration
The LCA characterization results for the two
applied methodologies (CML and CED) in each of
the eight evaluated scenarios are listed in Table 3.
Figure 3a shows the average distribution of
impacts for each reactor configuration, for the
subsystems cleaning (S1), nutrient supply (S2),
cultivation (S3) and concentration (S4). Figure 3
shows that, as expected, the cultivation stage is the
main hot spot for all the reactors in nearly all the
analyzed categories, with 80% or more impact.
This result confirms, for all operational systems,
the findings of previous studies based on
hypothetical scenarios with extrapolated data
regarding the importance of cultivation stage [3, 9,
15]. These studies indicated that energy
requirements and GWP of algal cultivation could
represent more than 90% of the total
environmental burdens for tubular PBRs and 55%
for the ORP [9, 15]. For AlgaePARC facility, the
main contributions to the total CED are those of
NR-F (85% of total CED) and NR-N (10% total
CED). These are two important categories that can
be reduced by minimizing the energy
requirements for algal products.
Concerning S2, only the CML categories of
AP, EP, TEP and POFP, as well as the CED
category of NR-B show significant contributions.
Most of these contributions are associated to the
production of the nitrogen source. S1 and S4 have
a very limited contribution to all the evaluated
categories. S4 has for all categories more
14
influence for the ORP than for the tubular PBRs.
This is because the achieved biomass
concentrations in the ORP are significantly lower
than in the tubular PBRs. Both S1 and S4 have
contributions above 5% for the categories of EP,
TEP and R-HYD. For the tubular PBRs, the
highest contribution from S1 is associated with the
category of HTP, with only 4-6%, while the
contribution from S4 only exceeds 1% for the
summer and fall periods in the categories of EP,
TEP and R-HYD.
Table 3. Characterization results for the cultivation of microalgae Nannochloropsis in real pilot-scale systems (FU: 1
kg DW harvested biomass in 22% DW slurry).
Impact category
Horizontal PBR
Vertical PBR
ORP
Summer
Fall
Winter
Summer
Fall
Winter
Summer
Fall
CML 2001 METHODOLOGY
Abiotic depletion, ADP
(kg Sb eq)
1.66
3.40
18.6
1.64
4.43
20.6
1.98
33.0
Acidification, AP (kg SO2
eq)
0.460
0.996
4.36
0.485
1.23
4.77
0.605
7.51
Eutrophication, EP (kg
PO43- eq)
0.361
0.753
3.87
0.363
0.96
4.26
0.514
7.15
Global Warming
Potential, GWP (kg CO2
eq)
216
443
2409
214
574
2665
256
4256
Ozone layer depletion,
ODP (kg CFC-11 eq)
7.68·10-6
1.58·10-5
8.34·10-5
7.56·10-6
2.01·10-5
9.17·10-5
9.19·10-6
1.47·10-4
Human Toxicity, HTP (kg
1,4-DB eq)
52.5
115
525
49.7
130
546
55.2
836
Freshwater Ecotoxicity,
FEP(kg 1,4-DB eq)
50.8
105
565
49.9
134
621
66.8
1027
Marine Ecotoxicity,
MEP(kg 1,4-DB eq)
33.0
68.2
366
32.4
87.1
403
42.4
661
Terrestrial Ecotoxicity,
TEP(kg 1,4-DB eq)
0.007
0.014
0.073
0.007
0.018
0.081
0.009
0.134
Photochemical oxidants’
formation, POFP (kg
C2H4 eq)
0.016
0.035
0.169
0.016
0.043
0.185
0.020
0.295
Land competition, LC
(m2year)
4.32
9.23
46.90
4.13
11.1
50.1
5.29
81.2
CED METHODOLOGY
Non-renewable fossil
energy demand, NR-F
(MJ)
3010
6179
33692
2983
8033
37315
3610
59727
Non-renewable nuclear
energy demand, NR-N
(MJ)
377
771
4187
370
994
4622
455
7449
Non-renewable energy
demand from biomass of
primary forests, NR-B
(MJ)
2.76·10-3
6.38·10-3
1.99·10-2
3.10·10-3
7.30·10-3
2.09·10-2
3.89·10-3
3.13·10-2
15
Renewable energy
demand from food and
agricultural sources, R-B
(MJ)
105
215
1213
104
282
1345
125
2156
Renewable energy
demand from solar, wind
and geothermal, R-WSG
(MJ)
27.4
55.6
314
26.9
72.9
348
32.1
557
Renewable energy
demand from
hydropower, R-HYD
(MJ)
18.7
42.7
180
17.3
45.2
180
20.1
275
TOTAL CUMULATIVE
ENERGY DEMAND,
CED (MJ)
3539
7263
39587
3501
9427
43811
4242
70164
16
Figure 3. Relative contributions to the environmental burdens of a) the subsystems and b) the production processes
involved in the operation of the compared pilot-reactor configurations. For each category, 100% corresponds to the
environmental impacts for each reactor (considering the average of the three operation periods) indicated in Table 2.
Figure 3b illustrates the breakdown of the
contributions per involved production process.
The impact of the electricity production for
cultivation constitutes the main environmental
burden of S3. The figure shows that electricity for
cultivation has contributions between 80-95% in
17
most of the categories. The role of energy
requirements with respect to the total
environmental impacts is consistent with other
studies in which the electricity for cultivation has
been identified as the main hot spot, but with
slightly lower relative contributions than that of
AlgaePARC pilot systems [14-15, 20, 33]. For
example, Stephenson et al. [15] reported that
electrical power during cultivation in ORPs has a
contribution of 74% to fossil energy requirement
and 65% to GWP. Similarly, energy was identified
by Lardon et al. [34] as one of the main causes of
impact for a raceway pond (with contributions
between 42-75% to CED and 18-36% to GWP),
together with fuel combustion and use of
fertilizers.
As further discussed in section 3.2., the main
reason for the high energy consumption is the use
of an electrical heater and chiller in this pilot
plant. These units, used for temperature control,
work fine for a pilot facility, but at large scale
should be replace by much more efficient
technologies, for example, using waste heat or a
storage temperature buffer in the ground. It should
be remarked that the absolute values in Table 3
are based on pilot-scale systems, and may change
significantly after scaling up to commercial scale.
Electrical efficiencies of the equipment are
expected to improve in commercial systems [36].
This improvement jointly with optimized
equipment dimensioning (to avoid oversizing)
could reduce the energy consumption by
approximately 66% when moving from pilot to
full scale [37]. The main trends for up-scaling,
including the expected differences between the
evaluated pilot-scale systems and hypothetical
industrial-scale facilities, are presented in section
3.4.
The production of building materials for the
reactors (plastics, steel and aluminum grouped as
“infrastructure” in Figure 3b) or the compressed
carbon dioxide, also included in S3, have low
contributions. The infrastructure is responsible for
more than 5% of the impacts in six of the eight
assessed scenarios: HTP, POFP, LC and R-HYD.
The contribution of this production process only
exceeds 16% for the category of R-HYD.
However, this category represented less than 1%
of total CED in all scenarios. Most of the impacts
from infrastructure are associated with the
production of metals, specifically aluminum for
the tubular PBRs (used for the supporting
structure) and steel for the ORP (among others
used for the shaft of the paddle wheel and the
tubes of the heat exchanger).
Among the other processes, the production of
nutrients has the highest impact, although it is
restricted to the categories AP (between 14-30%
depending on the season), NR-B (from 40% to
65%) and to a lesser extent to EP, TEP and POFP
(from 4 to 10%). Sodium nitrate, which comprises
more than 90% of the nutrients, has the highest
contribution. Substituting the main nitrogen
source by other alternatives would have a limited
effect on impact reduction under the evaluated
conditions. This finding differs with previous
research that mentions a noticeable influence of
the production of fertilizers in the environmental
18
profile [1, 5, 34]. The low relative contribution of
nutrients is caused by the higher contribution of
electricity of the pilot-scale reactors, which
attenuates the relative contribution of the other
processes. In addition, the use of assumptions and
extrapolated laboratory data for productivity used
in life cycle and techno-economic studies of algal
biofuels may lead to underestimation of the
required raw materials due to the overestimation
of the productivity potential [21].
The relative contribution of nutrient
production in the environmental profile depends
on the season of cultivation. Summer period
presented higher productivities (linked to higher
dilution rates) and lower electricity requirements
than fall and winter operation due to high
temperatures and light intensities. Higher dilution
rates require large quantities of nutrients to replace
the harvested culture volume. Electricity
consumption is lower in summer due to low
heating requirements compared to fall and winter,
which largely compensate energy needs for
cooling. The combined effect of higher dilution
rates and lower electricity requirements lead to a
higher relative contribution of nutrients to the
environmental burdens in summer than in fall and
winter.
Waste treatment has a moderate contribution
(around 10%) for ORP scenarios to EP and
toxicity categories (FEP, MEP, TEP) and to R-
HYD, while its influence is below 3% for the
tubular PBRs in all categories and conditions. This
impact was linked to the treatment of wastewater
from S1 and S4. The difference is mainly due to
the larger volume of water required by the ORP
both for biomass production and cleaning, per
functional unit. Although the tubular PBRs
required the addition of chemicals (disinfectant)
and materials (plastic beads) with higher impacts,
they needed a lower number of cleanings and
produced more biomass per year than ORP,
according to the operation during the reference
year. A minor difference in waste treatment
between the ORP and tubular reactor results from
S4. The harvested biomass from the ORP has a
lower concentration than the biomass from the
tubular PBRs, and therefore a higher volume of
wastewater was generated to achieve the same
biomass concentration.
3.2. Distribution of electricity requirements during
the cultivation stage (S3)
The identification of the hot spots reveals
that production of electricity for the cultivation of
Nannochloropsis sp. is the major contributor to
the environmental impact for all the reactor
designs and operating conditions. These energy
requirements arise from four activities: 1)
temperature regulation (including heating and
cooling of the culture medium), 2) aeration, 3)
mixing and 4) base energy of monitoring system.
To determine the relevance of each activity, the
distribution of electricity consumption is depicted
in Figure 4.
.
19
Figure 4. Distribution of electricity requirements during the cultivation stage for the evaluated pilot reactor
configurations and operating conditions (100% corresponds to the electricity consumption in S3 per FU listed in
Table 2).
Figure 4 shows that temperature regulation
is the main consumer of electricity during
cultivation in all evaluated scenarios, with total
requirements ranging between 60% and 90%. The
relative contributions of heating and cooling for
the tubular systems depend to a large extent on the
season. While cooling requires about 55% of the
electricity consumption for cultivation in the
summer for both tubular PBRs, it takes less than
10% of the requirements during fall and has no
contribution for the winter period. Regardless of
the weather conditions, the ORP needs no cooling
because it cools by evaporation of water. All
systems need additional heating to maintain the
temperature above the set point, even in the
summer period, though during this period the
heating request is almost only during the night.
Electricity requirements for heating are moderate
for the tubular reactors in summer (13% and 22%
respectively for horizontal and vertical systems)
but go above 62% for the ORP. Heating exceeds
50% of the total requirements for all the systems
operated in fall and winter.
20
One of the most significant findings of this
work is the high influence of temperature
regulation system on the global environmental
performance of microalgal cultivation. To the best
of our knowledge, no previous LCA study
discussed this key issue. Most studies exclude this
activity from the system boundaries. In some
cases, this seems a realistic assumption, since the
operating conditions to estimate the inventory data
are based on locations with warm temperatures
and sunlight intensities [1, 33], but for these
locations the contribution of cooling will increase
severely. Indeed, non-cooled closed reactors can
reach temperatures above 60°C [36], which would
make Nannochloropsis cultivation unfeasible,
according to its temperature growth range [29].
For other studies that consider less favorable
locations [5, 15], the effect of temperature
regulation is expected to be relevant for the total
impacts. Among the published LCA studies, only
Taelman et al. [20] was found to specify the use of
waste heat to maintain temperature in winter.
However, no contribution to the environmental
burdens is reported for this input, being treated as
a re-used waste flow, and thus having zero impact.
The applicability of Taelman’s approach is
restricted to scenarios in which a sufficient source
of waste heat is available. Estimating the total
energy needs for temperature regulation is
essential to verify for each possible location
whether this balance is met or not, and hence,
guarantee the sustainability of a large-scale
facility. Thus, this estimation may affect the
decision-making process depending on the
availability of natural cooling or waste heating
sources. The quantification provided in the present
paper highlights, for the first time, the importance
of taking temperature regulation into account
when conducting an LCA of a microalgal
cultivation system.
When comparing both tubular systems, the
electricity consumption for the heating in the
vertical system is between 1.3 and 1.9 times
higher than for the horizontal PBR. This is caused
by the approximately two times larger tube area of
the vertical system compared to the horizontal
system, which involves a higher heating
requirement to keep the temperature at the
minimum set temperature of 20°C. In addition,
during daytime, less light is absorbed per loop in
the vertical system due to its design, light dilution
effect and shading of the tubes. In fall/winter this
effect is even more pronounced, because the lower
tubes in the loops almost receive less light due to
low inclination of the sun and shading.
Aeration and CO2 addition has a relevant
contribution to the impact of cultivation
(associated with the blower) in the horizontal
system (around 20% in the three scenarios), but is
below 10% for all vertical PBR and ORP
scenarios. The high impact of the aeration/CO2 is
due to the back pressure of the stripper on pilot-
scale. Small electrical blowers are not designed to
overcome this pressure and for that an oversized
blower was needed; on large scale, an air
compressor would be used and electricity
consumption would decrease largely. The impact
of mixing (pumping and paddle wheel) is higher
21
for the vertical PBR and ORP contributions than
for the horizontal PBR due to higher volumes that
needed to be mixed. For the ORP, the electricity
consumption of the paddle wheel has already been
pointed out as a relevant contributor by other
authors [1, 15]. Although the impact of the paddle
wheel in fall is seven times higher than in summer
(due to lower productivities), the relative
contribution with respect to the total energy
requirements is significantly higher for summer
due to the lower consumption of other
components (e.g. the heating system). The base
energy used for monitoring had a minor
contribution between 1% and 7%. In a
hypothetical large-scale plant, this contribution
may be even lower since such installation would
have less equipment and fewer sensors (which are
numerous in the case of a pilot plant to allow
measuring a higher number of parameters).
3.3. Comparative environmental assessment of
cultivation scenarios
The comparative evaluation of scenarios is
represented in Figure 5, which reflects the
environmental profiles, with the ORP operation
under fall conditions as a reference.
For the operation in fall and winter, the
difference between configurations is decisive.
While the environmental burdens of the
horizontal PBR operated in fall approximately
double compared to the summer period, the
effects nearly triple for the vertical tubular
system.
For the operation in fall, the horizontal
PBR presents between 15% and 30% lower
impacts than the vertical configuration;
mainly due to a 30% lower electricity
consumption for cultivation. The difference
between the summer and fall scenarios is
much more pronounced in the case of ORP,
for which the impacts in fall are between 12
and 17 times higher than for the summer
operation. Hence, the fall performance of the
ORP is 90% worse than any of the tubular
systems under the same conditions and even
exceeds the environmental profile of both
PBRs operated in winter (with significantly
colder conditions and less irradiation) with
40%.
Although tubular PBRs were operated
under winter conditions and present a better
behavior than ORP in fall, the environmental
burdens are significantly higher, compared to
the relatively efficient performance during
summer and fall periods. The contributions of
the horizontal PBR during winter are about 5
times higher than those of fall, and up to 10
times above those of summer. Similarly, the
vertical PBR in winter conditions has an
average of 4 times the impacts of fall and
more than 12 times the impacts of summer.
The environmental impact of the horizontal
PBR in fall is between 5-10% lower than that
of the vertical PBR.
22
Figure 5. Relative environmental profile of the compared pilot reactor configurations with respect to ORP in fall
conditions for FU=1 kg DW microalgal biomass, according to the impact categories of a) CML methodology and b)
CED methodology.
23
The results show clearly that the efficient
environmental performance of an ORP system is
extremely restricted to the environmental
conditions, whereas tubular PBRs are less
dependent on weather conditions and can maintain
a relatively efficient performance for longer
periods. The ORP may only be feasible during a
limited period of the year, especially for locations
with low sunlight intensities, high rainfall and
moderate to low temperatures. This is due to the
combination of higher electricity consumptions
during cultivation stage for heating and a low
volumetric productivity for the ORP. This finding
is in agreement with the experimental difficulties
that prevented the operation of ORP in winter and
supports the unfeasibility of ponds for locations
with unfavorable thermal and solar conditions.
This outcome differs from the conclusions of
previous LCAs, where it was suggested that ORPs
have lower environmental impacts than tubular
systems due to a more simple operational strategy
[1, 15, 17]. The temperature regulation system and
the variations in productivity during the seasons
are the key factors for this difference, so both
factors should be included in environmental
assessments. Optimized temperature control
strategies (e.g. integration of waste heat, using
ground water for cooling or wider temperature
ranges) are essential to maintain moderate energy
consumption. Moreover, in the aforementioned
studies, different algae productivities are used in
the inventory analysis stage, due to more
favorable locations for the considered facilities. In
addition, the use of assumed values may involve
data inaccuracies and unrealistic assumptions,
which are overcome in this work thanks to the use
of experimental measurements.
3.4. Scalability of the results
The systems at AlgaePARC pilot facility are
pilot-scale reactors built to explore how laboratory
results can be translated to industrial scale
systems. However, at any pilot scale size, there are
limitations with regard to efficiencies of used
equipment (circulation pumps, air blowers and
especially temperature control), which are ruled
out in upscaling. In addition, during the evaluation
period the plant was operated to test the effects of
different conditions rather than maintaining a
stable optimized operation for long periods, which
would be the case in an industrial scale facility.
This situation influences significantly the absolute
values measured at pilot scale. For example,
Taelman et al. [20], indicated that in large scale
installations pumps with an efficiency of 80% are
installed, instead of the 11% efficiency pump in
this pilot-plant study.
As mentioned before, culture temperature
was controlled by a central electrical chiller and
electrical heater. The choice of an electrical
cooling/heating was motivated by the easy
installation and the flexibility in use for a pilot
plant meant for research. However, due to the high
primary energy demand the electrical heaters
should not be used at industrial scale plants. For
large scale applications the use of ground water
for cooling and waste heat from a biorefinery or
power generation plant are much more
24
convenient. If these heat sources are not available,
steam generation by direct burning of fuels instead
of electrical heating will reduce the impact of
heating with a factor 2-2.5 (taking into account
that the generation of electricity from primary
energy has an efficiency of 35-45%, while steam
generation has an efficiency between 80-90%).
Therefore, the values for heating and cooling
reported here are much higher than what is
expected at large scale, yet still give a good
indication on comparison between systems.
Further improvements on the environmental
impact can be obtained by either moving the
production facility to a warmer and sunnier
climate [20, 38], by using a waste heat stream
from e.g. power generation [19] or by choosing a
microalgal species that can grow at a wider range
of temperatures, and therefore decreasing the need
for heating or cooling. Other possibilities would
be to adjust the day-night cycles (i.e. using a
different set point or even avoid temperature
control at night) or to use a buffer tank below the
ground to store water during the night after
heating it during the day. In addition, heating
could be turned off during the night (keeping only
frost protection) to further reduce the energy
impacts. However, just before sunrise the culture
temperature should be above 20°C to prevent low
productivities.
As the inefficiencies of electrical equipment
(e.g. the circulation pumps for the tubular
systems) and the temperature regulation at
AlgaePARC pilot facilities are analogous for all
systems, comparison between systems is still
valid. Consequently, these data should not be used
to calculate absolute impacts for microalgae
cultivation at industrial scale, but they serve well
for analysis and comparison of the environmental
performance of various process designs and to
help debottlenecking these configurations.
Since the energy consumption linked to the
temperature control system has been identified as
the main contributor to the environmental impacts
of the current pilot reactors, we propose the
comparison of a set of hypothetical pilot and
large-scale scenarios. The hypothetical pilot
scenarios are based on the expected reductions on
energy consumption when temperature regulation
system can be omitted. In this case, the same
production conditions are assumed but energy
requirements during cultivation are considered to
be only necessary for mixing, aeration and
monitoring. Operating conditions close to these
scenarios might be achieved by selecting a
suitable location with favorable weather
conditions maintained over long periods of the
year, as well as including approaches to improve
the efficiency of the systems such as using waste
heat or ground buffer deposits. According to the
energy consumptions listed in Table 2, total
electricity consumption in the hypothetical pilot
scenarios would range between 88 and 830 MJ.
Three large-scale scenarios are also proposed,
based on the ratio between pilot and full-scale
energy consumptions in the cultivation stage
reported by Liu et al. [37], which would involve
electricity requirements between 26 and 249
MJ/FU.
25
Figure 6 shows a comparison of the best
performing real and hypothetical scenarios
corresponding to summer operation.
Figure 6. Relative environmental profile of hypothetical pilot and large-scale reactors with respect to pilot ORP in
summer conditions for FU=1 kg DW microalgal biomass, according to the impact categories of a) CML
methodology and b) CED methodology.
26
According to Figure 6, an optimized pilot-
scale system without temperature regulation under
favorable (summer) conditions could present
impact reductions ranging from 17% (for ORP in
NR-B) up to 80% (for vertical PBR in the
categories R-B and R-SWG) with respect to the
pilot summer scenario with the highest impact
(that is, the ORP operated in summer). When
upscaling, total impact reductions are expected to
range from 23% up to 90%. When comparing
hypothetical to real equivalent system, an
optimized pilot horizontal PBR could have
environmental improvements between 23% and
68% with respect to AlgaePARC system, while the
large-scale reactor would allow environmental
impact reductions between 50% and 90%. The
reductions for an optimized pilot-scale vertical
system would range between 22% and 77% and
for an ORP, between 17% and 60%. In the case of
large-scale vertical PBR, we could expect total
environmental improvements from 42% and 93%,
whereas the ORP would improve between 23%
and 84%. If we consider the comparison of these
optimized systems to the worst pilot scenario
(ORP operated in fall), impact reductions between
90% and 99% could be achieved. Despite the
difference of hypothetical large-scale systems with
respect to real pilot reactors a common trend has
been found: The hypothetical ORP reactor has
higher environmental impacts than the
corresponding horizontal and vertical PBR under
all the evaluated operation conditions. This means
that the complexity of tubular systems in terms of
energy-consuming elements is compensated by the
larger volumetric productivity. The results
presented in this section suggest that the upscaling
and optimization of microalgal reactors can
involve significant environmental improvements,
even though the hypothetical large-scale scenarios
were modeled according to a set of assumptions.
The applicability of the results relies on the
possibility to maintain the same biomass
productivities as those obtained with the
temperature control system.
The findings of this paper related to the
dependency of the environmental performance on
the weather conditions may contribute to the role
of LCA as a tool for process design and
optimization towards large-scale systems. The
high influence of the weather conditions should be
taken into account for the selection of appropriate
reactor configurations depending on the
geographic location. High productivities with
reduced heat requirements can be achieved in a
relatively easy manner by placing the facility in a
suitable location with warm temperatures and high
solar irradiations. Thus, an open pond could be
suitable for a location with warm temperatures
and low rainfall, whereas a closed system would
be more efficient from an environmental point of
view in the case of locations with more moderate
temperatures or high fluctuations. However,
locations with too high temperatures should also
be avoided for closed systems to reduce or
eliminate cooling needs. The availability of
natural sources or waste streams that can be used
for temperature regulation is also an essential
27
factor to take into account when selecting the
location.
4. Conclusions
LCA is a powerful tool to quantify the
environmental performance of microalgae
cultivation. A comparative LCA based on real
plant data for outdoor pilot raceway pond,
horizontal and vertical tubular photobioreactors at
AlgaePARC (The Netherlands) identified
temperature control as the main cause of impact
for all systems, regardless of the cultivation
season. ORP showed higher environmental
impacts than both tubular PBRs, as the latter
compensate energy-consuming elements with
higher productivity.
The results of this paper highlight the relevance of
LCA as a tool for process design and optimization.
In the case of microalgal cultivation, weather
conditions and availability of sources for
temperature regulation have been identified as
essential factors to take into account when
selecting a geographic location. For a given
location, this work can contribute to identify an
appropriate reactor configuration according to
environmental and energetic criteria. Moreover,
the simulations of hypothetical optimized
scenarios at pilot and large scale provide
information on the potential environmental
improvements. Despite some differences
influencing the scalability of the results, the use of
experimental data from outdoor pilot systems
instead of limiting to process simulations from
lab-scale data is essential to analyze and
debottleneck the environmental impact towards
large-scale cultivation.
Acknowledgements
The work presented in this paper has been
developed within the framework of the
BAMMBO Project (Project reference: FP7
KBBE-2010-4). The authors P. Pérez-López, G.
Feijoo and M.T. Moreira belong to the Galician
Competitive Research Group GRC 2013-032. P.
Pérez-López would like to express her gratitude to
the Spanish Ministry of Education for awarding a
research scholarship (AP2012-1605).
Furthermore, the authors would like to thank the
Ministry of Economic Affairs, Agriculture and
Innovation and Province of Gelderland, and
BioSolar Cells, BASF, BioOils, Drie Wilgen
Development, DSM, Exxon Mobil, GEA
Westfalia Separator, Heliae, Neste Oil, Nijhuis,
Paques, Cellulac, Proviron, Roquette, SABIC,
Simris Alg, Staatsolie Suriname, Synthetic
Genomics, TOTAL and Unilever for the financial
support of the AlgaePARC research program.
References
[1] Draaisma RB, Wijffels RH, Slegers PM,
Brentner LB, Roy A, Barbosa MJ. Food
commodities from microalgae. Curr Opin
Biotechnol 2013; 24(2):169-177.
[2] Hallenbeck PC, Grogger M, Mraz M, Veverka
D. Solar biofuels production with microalgae.
Appl Energy 2016; 179, 136-145.
28
[3] Lam MK, Lee KT. Microalgae biofuels: A
critical review of issues, problems and the way
forward. Biotechnol Adv 2012; 30(3):673-690.
[4] Nienhueser IA, Qiu Y. Economic and
environmental impacts of providing renewable
energy for electric vehicle charging – A choice
experiment study. Appl Energy 2016;
180:256-268.
[5] Clarens AF, Resurreccion, EP, White MA,
Colosi LM. Environmental life cycle
comparison of algae to other bioenergy
feedstocks. Environ Sci Technol 2010;
44(5):1813-1819.
[6] Zhang H, Xu W, Zhou D, Cao J. Waste
cooking oil-to-energy under incomplete
information: Identifying policy options
through an evolutionary game. Appl Energy
2017; 185(1):547-555.
[7] Pérez-López P, González-García S, Jeffryes C,
Agathos SN, McHugh E, Walsh D, Murray P,
Moane S, Feijoo G, Moreira MT. Life cycle
assessment of the production of the red
antioxidant carotenoid astaxanthin by
microalgae: from lab to pilot scale. J Clean
Prod 2014; 64:332-344.
[8] Van Boxtel AJB, Pérez-López P, Breitmayer E,
Slegers PM. The potential of optimized
process design to advance LCA performance
of algae production systems. Appl Energy
2015; 1122-1127.
[9] Brentner LB, Eckelman MJ, Zimmerman JB.
Combinatorial life cycle assessment to inform
process design of industrial production of
algal biodiesel. Environ Sci Technol 2011;
45(16):7060-7067.
[10] Chisti Y, Yan J. Energy from algae: Current
status and future trends: Algal biofuels – A
status report. Appl Energy 2011; 88(10):3277-
3279.
[11] Fortier MOP, Roberts GW, Stagg-Williams
SM, Sturm B.S.M.. Life cycle assessment of
bio-jet fuel from hydrothermal liquefaction of
microalgae. Appl Energy 2014; 122: 73-82.
[12] Woertz IC, Benemann JR, Du N, Unnasch S,
Mendola D, Mitchell BG, Lundquist TJ. Life
cycle GHG emissions from microalgal
biodiesel - A CA-GREET model. Environ Sci
Technol 2014; 48(11):6060-6068.
[13] Razon LF, Tan R. Net energy analysis of the
production of biodiesel and biogas from the
microalgae: Haematococcus pluvialis and
Nannochloropsis. Appl Energy 2011; 88:3507-
3514.
[14] Chiaramonti D, Maniatis K, Tredici MR,
Verdelho V, Yan J. Life Cycle Assessment of
algal biofuels: Needs and challenges. Appl
Energy 2015; 154:1049-151.
[15] Stephenson AL, Kazamia E, Dennis JS,
Howe CJ, Scott SA, Smith AG. Life-cycle
assessment of potential algal biodiesel
production in the United Kingdom: A
comparison of raceways and air-lift tubular
bioreactors. Energy Fuels 2010; 24:4062-
4077.
[16] Sills DL, Paramita V, Franke MJ, Johnson
MC, Akabas TM, Greene CH, Tester JW.
Quantitative uncertainty analysis of Life Cycle
29
Assessment for algal biofuel production.
Environ Sci Technol 2013; 47:687-694.
[17] Jorquera O, Kiperstok A, Sales EA, Embiruçu
M, Ghirardi ML. Comparative energy life-
cycle analyses of microalgal biomass
production in open ponds and
photobioreactors. Bioresour Technol 2010;
101(4):1406-1413.
[18] Bosma R, de Vree JH, Slegers PM, Janssen
M, Wijffels RH, Barbosa MJ. Design and
construction of the microalgal pilot facility
AlgaePARC. Algal Res 2014; 6:160-169.
[19] Slade R, Bauen A. Micro-algae cultivation
for biofuels: Cost, energy balance,
environmental impacts and future prospects.
Biomass Bioenerg 2013; 53:29-38.
[20] Taelman SE, De Meester S, Roef L, Michiels
M, Dewulf J. The environmental sustainability
of microalgae as feed for aquaculture: A life
cycle perspective. Bioresour Technol 2013;
150:513-522.
[21] Moody JW, McGinty CM, Quinn JC. Global
evaluation of biofuels potential from
microalgae. PNAS 2014; 111(23):8691-8696.
[22] Slegers PM, van Beveren PJM, Wijffels RH,
van Straten G, van Boxtel AJB. Scenario
analysis of large scale algae production in
tubular photobioreactors. Appl Energy 2013;
105:395-406.
[23] Collet P, Lardon L, Hélias A, Bricout S,
Lombaert-Valot I, Perrier B, Lépine O, Steyer
JP, Bernard O. Biodiesel from microalgae -
Life cycle assessment and recommendations
for potential improvements. Renew Energy
2014; 71:525-533.
[24] Montazeri M, Soh L, Pérez-López P,
Zimmerman JB, Eckelman MJ. Time-
dependent life cycle assessment of microalgal
biorefinery co-products. Biofuels Bioprod
Bioref 2016; 10(4):409-421.
[25] Rodolfi L, Chini Zitelli G, Bassi N, Padovani
G, Biondi N, Bonini G, Tredici MR.
Microalgae for oil: Strain selection, induction
of lipid synthesis and outdoor mass cultivation
in a low-cost photobioreactor. Biotechnol
Bioeng 2009; 102(1):100-112.
[26] De Vree JH, Bosma R, Janssen M, Barbosa
MJ, Wijffels RH. Comparison of four outdoor
pilot-scale photobioreactors. Biotechnol
Biofuels 2015; 8:215.
[27] ISO 14040. Environmental Management -
Life Cycle Assessment - Principles and
Framework, Geneva, Switzerland, 2006.
[28] Azapagic A, Clift R. Allocation of
environmental burdens in multiple-function
systems. J Clean Prod 1999; 7(2):101-119.
[29] Sandnes JM, Källqvist T, Wenner D, Gislerød
HR, 2005. Combined influence of light and
temperature on growth rates of
Nannochloropsis oceanica, linking cellular
responses to large-scale biomass production. J
Appl Phycol 2005; 17:515-525.
[30] Frischknecht R, Jungbluth N, Althaus HJ,
Doka G, Heck T, Hellweg S, Hischier R,
Nemecek T, Rebitzer G, Spielmann M, Wernet
G. Overview and methodology. Ecoinvent
30
report No. 1. v2.0. Swiss Centre for Life Cycle
Inventories, Dübendorf; 2007a.
[31] Guine JB, Gorrée M, Heijungs R, Huppes G,
Kleijn R, de Koning A, van Oers L, Wegener
Sleeswijk A, Suh S, Udo de Haes HA, de
Bruijn H, van Duin R, Huijbregts MAJ.
Handbook on life cycle assessment:
Operational guide to the ISO standards.
Dordrecht (The Netherlands): Kluwer
Academic Publishers; 2002.
[32] Frischknecht R, Jungbluth N, Althaus HJ,
Bauer C, Doka G, Dones R, Hischier R,
Hellweg S, Humbert S, Köllner T, Loerinick
Y, Margni M, Nemecek T. Implementation of
Life Cycle Impact Assessment methods.
Ecoinvent report No. 3. v2.0. Swiss Centre for
Life Cycle Inventories, Dübendorf; 2007b.
[33] Collet P, Hélias A, Lardon L, Steyer JP,
Bernard O. Recommendations for Life Cycle
Assessment of algal fuels. Appl Energy 2015;
154:1089-1102.
[34] Lardon L, Hélias A, Sialve B, Steyer JP,
Bernard, O. Life-cycle assessment of biodiesel
production from microalgae. Environ Sci
Technol 2009; 43(17):6475-6481.
[35] Goedkoop M, Oele M, Leijting J, Ponsioen T,
Meijer E. Introduction to LCA with SimaPro.
Pré-Consultants 2013.
[36] Ruiz J, Olivieri G, de Vree J, Bosma R,
Willems P, Reith JH, Eppink MHM,
Kleinegris DMM, Wijffels RH, Barbosa MJ.
Towards industrial products from microalgae.
Energy Environ Sci 2016; 9, 3036-3043.
[37] Liu X, Saydah B, Eranki P, Colosi LM, Greg
Mitchell B, Rhodes J, Clarens AF. Pilot-scale
data provide enhanced estimates of the life
cycle energy and emissions profile of algae
biofuels produced via hydrothermal
liquefaction. Bioresour Technol 2013;
148:163-171.
[38] Norsker NH, Barbosa MJ, Vermuë MH,
Wijffels RH. Microalgal production – a close
look at the economics. Biotechnol Adv
29(1):24-27.
