Algal Research 31 (2018) 347–362

Contents lists available at ScienceDirect

Algal Research
journal homepage: www.elsevier.com/locate/algal

Techno-economic evaluation of microalgae harvesting and dewatering T

systems
⁎
F. Fasaei, J.H. Bitter, P.M. Slegers, A.J.B. van Boxtel
Biobased Chemistry and Technology, Wageningen University Research, P.O. Box 17, 6700AA Wageningen, The Netherlands

A R T I C L E I N F O A B S T R A C T

Keywords: Microalgal biomass is processed into products by two main process steps: 1) harvesting and dewatering; and 2)
Microalgae extraction, fractionation and conversion. The performance of unit operations for harvesting and dewatering is
Harvesting often expressed in qualitative terms, like “high energy consumption” and “low in operational cost”. Moreover,
Dewatering equipment is analysed as stand-alone unit operations, which do not interact in a chain of operations. This work
Cost
concerns a quantitative techno-economic analysis of different large-scale harvesting and dewatering systems
Energy
System analysis
with focus on processing cost, energy consumption and resource recovery. Harvesting and dewatering are
considered both as a single operation and as combinations of sequential operations. The economic evaluation
shows that operational costs and energy consumption are in the range 0.5–2 €·kg− 1 algae and 0.2–5 kWh·kg− 1
of algae, respectively, for dilute solutions from open cultivation systems. Harvesting and dewatering of the dilute
systems with flocculation results in the lowest energy requirement. However, due to required chemicals and loss
of flocculants, these systems end at the same cost level as mechanical harvesting systems. For closed cultivation
systems the operational costs decrease to 0.1–0.6 €·kg− 1 algae and the energy consumption to
0.1–0.7 kWh·kg− 1 algae. For all harvesting and dewatering systems, labour has a significant contribution to the
total costs. The total costs can be reduced by a high level of automation, despite the higher associated investment
costs. The analysis shows that a single step operation can be satisfactory if the operation reaches high biomass
concentrations. Two-step operations, like pressure filtration followed by spiral plate technology or centrifuga-
tion, are attractive from an economic point of view, just as the operation chain of flocculation followed by
membrane filtration and a finishing step with spiral plate technology or centrifugation.

1. Introduction like lipids, proteins and carbohydrates. Furthermore, specific compo-

The increasing demand for food, energy and materials raised the
role of microalgae feedstock in the biobased economy. However,
commercial production of algal products is still in its infancy. To
commercialize algal biomass as a commodity, the production costs for
algal products should be decreased at least by a factor 10 [1].
The production of algal based products has three main steps: 1)
biomass cultivation, 2) harvesting and dewatering, and 3) biomass
extraction, fractionation and conversion. Algal biomass cultivation oc-
curs in open or closed photobioreactors. These reactors deliver a very
dilute algal solution ranging from 0.05–0.075% dry matter for open
pond systems to 0.3–0.4% for closed systems. The function of har-
vesting and dewatering is to increase the total solid matter up to
10–25% of total dry matter [2] or even to a dry product. Harvesting and
dewatering can be done in one or more successive steps, depending on
the type of applied equipment. In the last stage of processing, the
harvested biomass is split into fractions towards the aimed components,
nents of interest are processed into user products, such as biodiesel from
lipids.
Cultivation is the main cost contributor for algal based products
[3,4]. However, harvesting and dewatering of microalgae biomass are
also considered as an important contributor to the total costs. Several
studies report the harvesting costs at 20–30% of the total production
costs [2,5–8]. The high capital expenditure and energy consumption
result from the dilute algae solutions, the large volumes to be pro-
cessed, and the small size of microalgal cells [1,5,9].
Various unit operations show potential to be implemented for har-
vesting and dewatering. These technologies range from proven tech-
nologies to innovative process unit operations. Application of the
technologies is not straightforward due to the physical and chemical
properties of dilute algal solutions. Table 1 gives an overview and
qualifications, from existing literature, of possible unit operations for
harvesting and dewatering. Harvesting and dewatering of algal biomass
can be carried out by using a single technology with high impact

⁎
Corresponding author at: P.O. Box 17, 6700 AA Wageningen, The Netherlands.
E-mail address: ton.vanboxtel@wur.nl (A.J.B. van Boxtel).

https://doi.org/10.1016/j.algal.2017.11.038
Received 29 June 2017; Received in revised form 29 November 2017; Accepted 30 November 2017
2211-9264/ © 2017 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/BY-NC-ND/4.0/).
F. Fasaei et al. Algal Research 31 (2018) 347–362

Table 1
Overview of available technologies for harvesting and dewatering of microalgae with main qualifications.

Technology Strength Weakness Reference

Centrifugation • Continuous • High capital cost [2,9]

• Efficient for large scale processing
• High recovery
Spiral plate technology (SPT) • Efficient for small scale processing
• High recovery
Pressure filtration • Low energy demand
• High recovery
Vacuum filtration • Continuous
Membrane filtration • Efficient for small scale processing
• High recovery
Sedimentation • Easy application
• Low energy demand
Chemical flocculation • Low energy demand
• Low equipment cost
Drum drying • Mature technology
Spray drying • Suitable for high value products
Solar drying • Low cost
performance or by combining multiple unit operations in a sequence.
The effectivity of combination of unit operations in sequence depends
on the individual performance of each unit. The choice of a unit op-
eration for the first concentration step or harvesting also affects the
choice and performance of the following units in the dewatering step
[10].
Fig. 1 shows a structure of possible combinations of unit operations.
Concentrating microalgae from the cultivation medium can follow
three strategies: 1) a single-step harvesting and dewatering to the aimed
concentration; 2) one step of harvesting followed by a separate dewa-
tering step; and 3) one step of harvesting followed by two steps of
dewatering. These three strategies can be followed by drying to extend
the shelf-life and to make the product accessible for further downstream
processing [5]. The choice for the strategy is also set by the constraints
of an operation, such as the maximal feasible concentration, the visc-
osity of the concentrate, etc. For example, flocculation is effective up to
2–2.5% dry matter [22,26–28] and membrane filtration to 5–7% dry
matter [29]. These operations need a third operation to reach a final
concentration of 10–25% dry matter, as a result of the mentioned
constraints.
Sedimentation, driven by the gravitational force, has long settling
times (10 h or longer) and can reach only total solid contents up to
2–3% [9]. Therefore, this method is not attractive for large scale ap-
plications [20] and is outcompeted by flocculation. Solar drying is also
slow, requires large areas and has a high risk for contamination and loss
of biomass [8,25]. Therefore, these technologies are not given in Fig. 1
as options for processing algal biomass in large scale applications.
Harvesting and dewatering are often assessed qualitatively (e.g.
qualifications used in Table 1) or papers report experimental results of
just a single unit operation [2,9,30,31]. Generally, quantitative as-
sessments of technologies for harvesting and dewatering focus on en-
ergy demand and yield [31]. A common drawback of existing evalua-
tions is due to the stepwise approach. In this type of approach, each
technology is considered for a specific task, while the interaction of all
operations in a chain, and subsequent overall performance, are not
evaluated or discussed. The main goal of this study is, therefore, to
quantitatively analyse combinations of harvesting-dewatering systems.
This quantitative analysis is based on a techno-economic assessment of
harvesting and dewatering systems available at industrial scale. In this
analysis, feasible configurations of proposed unit operations, as given in
Fig. 1, are considered. The main addressed criteria are biomass
• High capital cost [9,11]
• Limited throughput capacity
• Discontinuous [5,12,13]
• Clogging or fouling
• Relative high harvesting cost [5,12–14]
• Clogging or fouling
• Fouling [15–19]
• High capital cost
• Slow rates [9,16,20]
• Large operational area
• Low recovery
• Limited application: suitable for large size algae
• Difficult recovery of flocculants [13,16,21,22]
• High energy demand [23]
• High energy demand [5,24]
• Large drying surface [8,25]
• Slow drying rate
• High risk for contamination and loss of mass
• Not for food grade products
recovery, energy requirement, capital, labour and other operational
expenditures per kg of harvested biomass. Moreover, aspects such as
chemical consumption, resource recovery and opportunities to recycle
the medium to cultivation site are discussed. In an effect analysis, the
role of different feed concentrations obtained in different cultivation
systems, the role of seasonal changes, production characteristics related
to the latitude, and the role of automation are discussed. The results of
the analysis gives a clear view on the efficiency of harvesting-dewa-
tering processing chains in terms of cost, energy consumption, and re-
source recovery.
2. Approach and methods
Fig. 1 illustrates the succeeding steps and unit operations that are
applied for harvesting and dewatering in this work. The available
technologies for the harvesting step are membrane filtration, chemical
flocculation, vacuum and pressure filtration, centrifugation, and spiral
plate technology. For dewatering step, membrane filtration, vacuum
and pressure filtration, centrifugation, and spiral plate technology, can
be applied. A short description of the technologies is given in Appendix
A. Technologies, such as centrifugation, vacuum and pressure filtration
and spiral plate technology, have the potential to achieve high biomass
concentrations and possibly do not need an additional dewatering step.
Membrane filtration and flocculation are limited in the maximal con-
centration and require a successive step (centrifugation, vacuum or
pressure filtration or spiral plate technology) to achieve a high con-
centration solution. Moreover, an initial harvesting step reduces the
volume size significantly and it is, therefore, meaningful to quantify the
role of volume reduction on the performance of a chain of operations.
2.1. Model based analysis
A model-based approach is applied for the techno-economic eva-
luation. For each unit operation a simulation model is defined. The
models concern the input-output mass and energy balances for each
unit operation. Additional relations are included to connect the energy
demand and product yield to economic estimation elements (see
Appendices B and D). The models for the unit operations are made in
Excel. A flexible structure is used to connect all unit operations with
each other in any combination, as given in Fig. 1.
The function of harvesting and dewatering is to split feed streams,

348
F. Fasaei et al. Algal Research 31 (2018) 347–362

Centrifuge

Membrane Drum dryer

Filter
Spiral plate
technology
Spray dryer

Spiral plate
Pressure technology
Filter

Centrifuge
Vacuum
Filter
Pressure
None
Filter

Flocculation
Membrane Vacuum
(Cationic/
Filter Filter
Chitosan)

None
Centrifuge

Spiral plate
Centrifuge Spiral plate
technology
technology

None

Centrifuge
Spiral plate
technology
None

Fig. 1. Combination of operations for harvesting and dewatering of algal biomass at large scale.

with or without the aid of extra chemicals, into a concentrated product
stream and a co-stream. The mass balance for algal biomass in each step
is:
Fin Cin = Fout Cout + Fco,out Cco,out
consumption of each unit operation, which is linked to the amount of
materials being processed. The data was derived from information by
industrial equipment suppliers, literature and engineering databases
(Appendix C).
(1)

where F represents the in/out volumetric flow rates for streams con-
taining algal biomass (m3 ⋅ h− 1) and co is the co-stream that contains
water and a small fraction of lost algal biomass. C represents the con-
centration of algal biomass (kg ⋅ m− 3) in each stream. This equation is
extended with additional information on the degree of concentration,
the efficiency of separation etc. (see Appendix B).
The energy input is based on data for the specific energy
2.2. Operational conditions and equipment constraints
The analysis is performed for a standard system. This system con-
cerns the harvesting and dewatering of cultivated algae in a 100 hectare
open pond under chemostat conditions with a continuous flow of bio-
mass. Related data and conditions are compiled in Table 2. The average
feed flow rate and concentration throughout the year is applied.

349
F. Fasaei et al.
Table 2
Settings and conditions for the standard system.
Reference condition for system evaluation
Feed concentration 0.05% Dry matter
Feed stream 400 m3·hr− 1
Product concentration in concentrate/to 15.00% Dry matter
dryer
Water content in product after drying 5% kg water/kg
product
Product concentration between harvesting Variable kg·m− 3
and dewatering
Operational hours 7200 hr·yr− 1
The capacity of unit operations is not unlimited. For each operation,
maximal sizes of the equipment are applied. If the required capacity
exceeds the maximal size of equipment, additional units are installed.
For example, if the required capacity exceeds that of three units oper-
ating at maximal capacity, then a fourth unit is installed. It is assumed
that all installed units have the same capacity. Information on the
maximal sizes/capacity of the unit operations are given in Table D.1,
Appendix D. During harvesting and dewatering the concentration in-
creases and with the increase in concentration the ability to process the
concentrate can reach the maximal feasible values (for example, due to
fouling or clogging, solution is too viscous, flocked algae still contain a
large amount of water, etc.). Therefore, the concentration values are
constrained. The maximum feasible values for the concentrate are also
given in Table D.1.
2.3. Economic analysis
The yearly costs (€·yr− 1) for harvesting-dewatering are given by
[32]:
CI
Pc = (0.5I + M ) × CI + + Ce + Cc + Cl + (Closs)
A (2)
then the production costs per kilogram produced algal biomass are:
Pc
Pc,algae =
Ytotal
where CI is the total investment cost (€), I the rate of interest (%) and M
the % maintenance costs, CI/A the yearly depreciation (€·yr− 1) over the
depreciation time A (yr), Ce the energy costs (€·yr− 1), Cc the costs for
consumables (€·yr− 1), Cl the costs for labour (€·yr− 1) and Ytotal is the
total produced biomass in a year (kg·yr− 1). Data related to the costs is
given in Appendix D, Table D.4.
Closs is an optional term which represents the loss of algae (€·yr− 1).
This term is relevant if the co-stream from a unit operation is not reused
in the cultivation system. For the analysis of the standard system it is
assumed that all co-streams can be recycled to the cultivation system.
However, the impact of non-recycled streams is discussed in the results
and discussion section.
For each scenario, the investment costs are calculated from the total
purchase costs of equipment multiplied with a Lang factor. The Lang
factor is a multiplier to the purchase costs in order to cover the costs for
piping, instrumentation, electrical facilities, buildings, engineering,
construction and contractors fee. In industries such as dairy and sugar
processing, it is common to apply a Lang factor of 3.5 for liquid/solid
separation, which is also used in this work [33,34]. Investment related
costs such as amortization, interest and maintenance are based on the
Lang corrected investments cost. The purchase costs of equipment are
based on prices from data sources from 2014 to 2015.
Table D.1 in Appendix D gives the investment costs for all unit
operations at a given capacity. Extrapolation of the purchase costs to
other capacities and scales is based on the scaling factor rule:
Algal Research 31 (2018) 347–362
Typical dry matter for an open pond system
Corresponds to 100 ha open pond system with productivity of 15 ton·ha− 1·year− 1,
7200 production hours·year− 1 in chemostat operation
For flocculation, a maximum value of 2.5% dry matter is applied, for membrane filtration 5.0% kg
dry matter, for the other operations 15.0% dry matter.
Continuous operation for 360 days per year for 20 h per day.
costB size B ⎞n
=⎛
costA ⎝ size A⎠ (4)
where costA and costB represent the purchase costs of a unit operation
with size or capacity size A and size B. n is the corresponding scaling
factor for the equipment. Scaling factors that cover the range 10–100%
of maximal capacity of the considered units are derived from en-
gineering data bases, supplier information or literature and are given in
Table D.1.
The operational costs include energy, consumables, labour and costs
related to the loss of biomass due to incomplete biomass recovery. The
energy costs follow directly from the energy uptake of the installations.
These were specified according to the equipment supplier data and
literature (see Table C.2 in Appendix C). Electricity price and the costs
for direct heating by natural gas are based on EuroStat information for
2014 for Northwest Europe [35]. The price for steam used for the
drying operations is based on a 80% efficiency from steam generation
by gas heating. The applied values are given in Table D.3 in Appendix
D. The costs related to consumables are applicable to the consumption
of flocculants and membrane replacement (Table D.5 in Appendix D).
The costs for labour are based on the work load for continuous
operation in a 4-shift system. Salaries are based on the 2015 minimum
wage in Northwest Europe (the Netherlands) and are adapted by edu-
cational level and responsibility with relevant factors of 3, 4.3 and 6.7
for operator, supervisor and plant manager respectively. Each operator
supervises 5 medium level automated continuous unit operations while
(3)
a supervisor coordinates 4 operators and the plant manager is re-
sponsible for 20 operators. Moreover, overhead costs of 20% are taken
into account for administration, laboratory etc. [3]. Detailed informa-
tion is given in Table D.2 in Appendix D.
2.4. Effect analysis
The aim of the effect analysis is to project the impact of different
algae cultivation systems on total annual cost, energy demand, and the
individual cost contributors. Cultivation systems vary in productivity,
biomass concentration and feed flow to the harvesting and dewatering
system. Three 100 ha cultivation systems are considered, each with a
different yearly productivity and algae concentration (Table 3). Next,
the impact of varying the size of the cultivation system (1, 10 and
Table 3
Productivity, biomass concentration and flow rate for the effect analysis.
Productivity Biomass Feed flow rate
(ton·ha− 1·yr− 1) concentration (m3·h− 1) for
(kg·m− 3) 100 ha
Open pond 15 0.5 400
Tubular 30 1.5 280
systems
Flat panel 45 2.5 250
systems
350
F. Fasaei et al.
Fig. 2. Combined graphs of energy demand and total costs for the harvesting and de-
watering systems. The numbering of the systems is given in Table 4, the results are
documented in Appendix E, Table E.1. (A): all results, (B): details for the boxed area from
figure A, (C): details from boxed area from figure B.
100 ha) is considered. The effect of a higher cultivation productivity at
other latitudes is investigated by doubling the required average capa-
city. The impact of seasonal changes is estimated by varying feed over a
year for a system with a capacity ± 50% around the average value
given in Table 2. Finally, the role of energy prices, labour costs, and
purchase costs of equipment are examined by a ± 10% variation to the
applied values.
3. Results and discussion
3.1. Harvesting and dewatering
Both processing costs and energy consumption are considered as
bottlenecks in harvesting and dewatering [6]. Therefore, first the total
costs and energy demand for the standard system (Table 2) are eval-
uated. Fig. 2 shows the energy demand and total costs per kg of har-
vested biomass for each harvesting-dewatering combination as depicted
in Fig. 1. The combinations are given in Table 4 and numerical data is
given in Table E.1 in Appendix E. The left bottom corner of Fig. 2-A
concerns the group of scenarios with low total costs and energy de-
mand. This area is enlarged in Fig. 2-C.
351
Algal Research 31 (2018) 347–362
Fig. 2-A shows that scenarios 25 and 26 have the highest total costs.
These scenarios start with spiral plate technology as the harvesting step.
Spiral plate technology can concentrate algal biomass to a concentra-
tion of 15–25% dry matter, while the capacity of the largest commercial
available unit is limited to a feed rate of 4 m3·hr− 1. To process all feed
from the cultivation system around 100 spiral plate technology (SPT)
units are required. As a result, the investment and labour costs increase
significantly. Pre-concentration before applying spiral plate technology
reduces the costs significantly, for example by flocculation (numbers 14
and 22 in Fig. 2-C), filtration methods (numbers 2,4 and 6), or cen-
trifugation (number 24). For small scale processing a few spiral plate
technology (SPT) units are required. For example, for 1 ha, one spiral
plate technology unit is sufficient and, therefore, the operating and
investment-related costs will be more balanced. Nevertheless, the spiral
plate technology is an expensive solution as a first harvesting step
compared to technologies as centrifugation or pressure filtration. The
strongest benefit of spiral plate technology is the high final dry matter
contents. Therefore, this operation is considered as a strong finishing
step.
Centrifugation reaches biomass concentrations up to 15%–20% dry
matter. This technology is proven in several industries. The processing
capacity, up to 120 m3·hr− 1 feed, is high. The energy consumption of
modern centrifuges has been optimized. This operation can be applied
as a single unit (scenario 23).
Harvesting with membrane filtration followed by one-step dewa-
tering with centrifugation (scenario 1) has slightly lower cost than the
combination of membrane filtration with SPT (scenario 2). The energy
requirement of these systems with membrane filtration is high
(4.2 kWh·kg− 1) due to the large volumetric flow of permeate.
Vacuum filtration, as a single step (scenario 28), is more energy
efficient compared to vacuum filtration combined with a dewatering
step (scenarios 5 and 6), while the production costs are in the same
range. The main reason for this result is related to the lower required
filter area for equipment in series than for a single unit. In other words,
the costs for an additional unit (centrifuge or SPT) are compensated by
reduction in required filter area. Comparison of the results for three
different filtration systems in the harvesting step shows that pressure
filtration is more energy and cost efficient than membrane and vacuum
filtration.
Single-step filtration methods (scenario 27 and 28) are the lowest in
costs and attractive due to the simplicity of the operations. However,
there is a risk of fouling leading to a lower performance in higher
product concentrations such as 15% dry matter. As an alternative, fil-
tration can be done to a lower level of dry matter to reduce the risk of
fouling (intermediate concentration of 10% dry matter) in combination
with an effective finishing step like centrifugation or spiral plate tech-
nology (scenarios 3 and 4, and 5 and 6). A single harvesting step with
centrifugation (scenario 23) is also economically attractive. Such a
single operation results in a relative simple processing system.
However, two step operations offer better possibilities for extension of
production capacity.
The left side of Fig. 2-A and -B are enlarged in Fig. 2-C and majorly
concerns chemical flocculation (cationic and chitosan) for harvesting.
Flocculation is followed either by one-step or two-step dewatering. The
two-step dewatering starts with membrane filtration followed by cen-
trifugation, SPT, pressure filtration, or vacuum filtration. Flocculation
first needs a short rapid mixing phase, then a longer slow mixing phase,
followed by a period of settling for phase separation. Flocculation is
energy efficient due to the low energy requirement for mixing. More-
over, flocculation provides, with 2.5% dry matter in the concentrated
stream, a significant volume reduction (50 times for the cultivation
system from Table 2). As a consequence, the required dimensions and
energy consumption of the downstream equipment are significantly
smaller.
Applying cationic flocculants results in higher costs compared to
chitosan flocculants. This is due to the lower required dosage of
F. Fasaei et al. Algal Research 31 (2018) 347–362

Table 4
Combinations of unit operations for harvesting and dewatering steps.

Combinations Harvesting Dewatering

1 Membrane filter Centrifuge –

2 Membrane filter Spiral plate technology –
3 Pressure filter Centrifuge –
4 Pressure filter Spiral plate technology –
5 Vacuum filter Centrifuge –
6 Vacuum Filter Spiral plate technology –
7 Cationic flocculation Membrane filter Pressure filter
8 Cationic flocculation Membrane filter Vacuum filter
9 Cationic flocculation Membrane filter Centrifuge
10 Cationic flocculation Membrane filter Spiral plate technology
11 Cationic flocculation Pressure filter –
12 Cationic flocculation Vacuum filter –
13 Cationic flocculation Centrifuge –
14 Cationic flocculation Spiral plate technology –
15 Chitosan flocculation Membrane filter Pressure filter
16 Chitosan flocculation Membrane filter Vacuum filter
17 Chitosan flocculation Membrane filter Centrifuge
18 Chitosan flocculation Membrane filter Spiral plate technology
19 Chitosan flocculation Pressure filter –
20 Chitosan flocculation Vacuum filter –
21 Chitosan flocculation Centrifuge –
22 Chitosan flocculation Spiral plate technology –
23 Centrifuge – –
24 Centrifuge Spiral plate technology –
25 Spiral plate technology – –
26 Spiral plate technology Centrifuge –
27 Pressure filter – –
28 Vacuum filter – –

chitosan for fresh water cultivated algae, compared to the required
dosage of cationic flocculants [36,37]. Under marine conditions the
flocculating effect of chitosan is lower and requires pH adjustment and
modification in chitosan properties [26].
Combinations of flocculants with other operations are, from an
economic point of view, competitive to other combinations of opera-
tions. However, there are some important remarks to the application of
flocculants:
microalgae and limit the possible applications [40].
4. Alkaline conditions are proposed (pH 9.9) to improve the degree of
algae recovery with chitosan flocculation [41]. Despite the increase
in recovery, these conditions contaminate the algal biomass with
mineral precipitation [42]. Flocculation with chitosan and neutral
pH is preferred (applied in this study) to prevent any contamination.
However, the lower recovery of biomass at neutral condition in-
creases the costs of harvesting and dewatering by 25–30%.

1. Flocculation systems have a lower biomass recovery yield compared 3.2. Drying operations
to the other harvesting operations. Part of the flocculants remains in
the co-stream, which contains significant amounts of algae. The co- After harvesting and dewatering the water content in the standard
stream has to be recycled to the cultivation unit to recover the re- system, from Table 2, is 85%. The water content of harvested algal
mainder of algae and to limit the water usage in cultivation. The biomass should be reduced to about 5% to extend the shelf life of the
available reports on the impact of flocculants in the recycle streams biomass, to reduce the weight for transport, and also to allow dry
are contradictory. Granados et al. [38] did not observe adverse ef- processing downstream.
fects on the growth rate of microalgae with a reused supernatant The total costs per kg of dry algae for drum and spray drying are
stream from flocculation (polyelectrolytes). However, Beim et al. given in Fig. 3. The maximum capacity of spray dryer is higher than
reported a negative impact of cationic polymeric flocculants on that of drum dryers, respectively up to 10,000 and 1000 kg water
water ecosystems, especially on the cell growth rate of protococcal evaporation per hour. For the standard system one spray dryer or two
algae [39]. Further investigation on the impact of reuse of residual drum dryers are required. Thus, the drum dryers lead to higher capital
streams containing flocculants on cultivation is, therefore, advised.
In addition, extra operations to remove retained flocculants from the
co-stream will give an upward shift to costs and energy use, which
can affect the positive results for costs and energy consumption of
flocculation.
2. If the co-stream cannot be reused for cultivation, a significant value
loss of biomass occurs, thus, the costs of this operation will increase.
For example, for flocculants with a recovery of 90% the lost value is
0.56 €·kg− 1 harvested biomass, and at 80% recovery 1.25 €·kg− 1
harvested biomass. These costs are further increased by the waste-
water treatment costs.
3. The flocculants are also attached to the microalgae in the con-
centrated stream. The presence of flocculants and their interaction
with microalgae may affect the performance of the following ex-
traction and fractionation steps. Moreover, the presence of poly-
meric flocculants can alter the carbon profile composition of Fig. 3. Costs for drum and spray drying per kg of dry product.

352
F. Fasaei et al. Algal Research 31 (2018) 347–362

and maintenance cost. The energy consumption of a drum dryer is

around 0.9 and of the spray dryer 1.09 kWh·kg− 1 evaporated water
(specifications from [1], [23,24]).
Drum drying is, therefore, lower in energy consumption compared
to spray dryer, respectively 5.1 and 6.1 kWh·kg− 1 algae. This leads to
0.25 and 0.30 €·kg− 1 dried algae as total energy costs. The costs of
labour are comparable for both dryers as one operator team can manage
more units. The costs for consumables are negligible. The overall costs
are just below 0.5 €·kg− 1 dried algae for both drying systems.
The results in Fig. 3 are based on the algal biomass from a culti-
vation unit as specified in Table 2. One spray dryer can manage large
volumetric flow rates from harvesting and dewatering, as they have
larger capacities. For increased drying capacities, the capital costs of
spray dryers become beneficial compared to the drum dryer. However,
for the drum dryer the lower energy costs compensate the increased
capital and related costs. As a result, both systems operate at similar
cost. For smaller scale processing, similar results are obtained.
The drying costs are related to the product concentration after
harvesting and dewatering. The total drying costs of 0.69, 0.48, and
0.32 €·kg− 1 dried biomass is derived for concentrates with biomass
concentrations of 100, 150 and 200 kg·m− 3, respectively.

3.3. Effect analysis

Harvesting and dewatering characteristics of 100 ha open pond,

tubular and flat plate systems are compared (see specifications in
Table 3). In these systems, the biomass concentration and yearly pro-
ductivity increase from pond, to tubular, to plates, while the average
flow rate towards the harvesting and dewatering system decreases. The
effect of these systems on cost and energy consumption per kg of dry
algae are shown in Fig. 4.
The results indicate that increasing the cultivation concentration, in
combination with a lower feed rate, leads to a shift of the energy
consumption towards the lower-left corner. This is the energy efficient
area. Fig. 4-C shows that cultivation setup results in processing costs
below 1 €·kg− 1 algae for nearly all scenarios, and an energy con-
sumption below 1 kWh·kg− 1 algae. This trend is due to the 3–5 fold
increase in the amount of biomass in the feed from the closed photo-
bioreactors. Simultaneously, there is a decrease in volumetric flow rate
of feed, which results in smaller equipment dimensions and a lower
energy consumption. Thus, the costs and energy consumption for har-
vesting and dewatering are highly influenced by the biomass con-
centration in the feed.
For a most beneficial combination of cultivation and harvesting- Fig. 4. Cost and energy comparison for harvesting-dewatering for three different 100 ha
dewatering system, the production costs of both systems have to be cultivation systems (specified in Table 3). A: standard system (open pond), B: tubular
summarized. The results from this work show a difference of system, C: flat plate system.
0.35–0.40 €·kg− 1 (see Table E.1) between harvesting-dewatering of
biomass from an open pond and flat panel system by pressure filtration
and centrifugation. So, the harvesting-dewatering system can compen-
sate for a maximum of 0.35–0.40 €·kg− 1 higher production costs in a
cultivation system. The results for more productive regions are in the
same order of magnitude (see effect analysis Northwest Europe versus
South Europe). The potential compensation depends on the combina-
tion of methods used for harvesting and dewatering. The data in Table
E.1 shows that the difference in costs between harvesting-dewatering
systems that start with flocculation is around 0.30 €·kg− 1, with mem-
brane filtration 1.10 €·kg− 1, and vacuum filtration 0.80 €·kg− 1.
The effect of cultivation size on the production costs are illustrated
in Fig. 5 for the combination of pressure filtration/centrifugation
(scenario 3) and three cultivation sizes. The production costs are 9.45,
1.24 and 0.50 €·kg− 1 for 1, 10 and 100 ha respectively. Results show a
significant reduction of production costs with the increase in cultivation Fig. 5. Effect of the cultivation size (1, 10 and 100 ha) on the costs per kg harvested
size. Labour is the major contributor to the total costs and reduction of algae. The results concern harvesting with pressure filtration to 10% dry matter and
subsequent centrifugal dewatering to 15% dry matter.
labour costs is essential to improve the total costs. The same trend was
found for other combinations of unit operations. Therefore, the main
challenge for cost reduction is reducing the labour costs. This reduction

353
F. Fasaei et al.
Fig. 6. Effect of algae concentration after harvesting on the total costs for the combina-
tion of pressure filtration to 10% dry matter and centrifugal dewatering to 15% dry
matter.
can be achieved by applying harvesting systems with large capacities
(flocculation, centrifuges) or by reducing the labour by implementing a
high degree of automation. A higher degree of automation, however,
implies a higher Lang factor to be applied. An increase in Lang factor
from 3.5 to 4.0, in combination with a reduction of labour costs of 50%,
results for the system defined in Table 2 in a reduction of total costs of
15.4%. Automation costs resulting in Lang factors up to 6.6 are com-
pensated by a 50% reduction in labour costs. A consequence of auto-
mation is a shift in the type of labour based on operators to labour
based on information technology and electronics.
The impact of the algae concentration after harvesting on total costs
is given in Fig. 6. The results show a decrease in total costs with in-
creasing concentrations after harvesting. The reduction in costs is
possible due to the reduction of investment costs and a lower energy
consumption by the smaller equipment in the dewatering step. The
decrease in costs is marginal above harvesting concentrations of
50 kg·m− 3. Fig. 6 also illustrates the major contribution of labour to the
total costs. The same trends of the effect of harvesting concentration on
the total costs are found for all other harvesting and dewatering sce-
narios.
In the presented results, the amortization time (15 years) and costs
for maintenance (5% of investment) were based on fresh water algae
cultivation. Marine algae cultivation with salt water will cause corro-
sion in stainless steel equipment. As an alternative, marine algae can be
processed in equipment with coated surfaces [43]. By applying these
surface coatings, part of the equipment can be made from carbon steel
instead of stainless steel. This lowers the investment costs, but si-
multaneously the life time of equipment is shorter and maintenance
requirements are higher. We estimated an overall increase of 20% in
the total costs.
The system as defined in Table 2 and other used data are related to
Northwest Europe. Production capacity of cultivation systems, labour
costs and energy prices in South Europe differ from the applied values.
Simulations were also performed for conditions that correspond to
South Europe (see Table D.6 in Appendix D). It is assumed that the costs
of equipment are the same in both situations. For the harvesting-de-
watering system with pressure filtration and centrifugation the results
are graphically presented in Fig. 7. The results for all combinations are
Fig. 7. Comparison of costs for harvesting by pressure filtration and centrifugal dewa-
tering in Northwest and South Europe.
354
Algal Research 31 (2018) 347–362
Fig. 8. Effect of varying capacity of chemostat cultivation systems on the costs for har-
vesting and dewatering by pressure filtration and centrifugation. A (top): relative varia-
tions in feed flow rate to harvesting system during a year, B (bottom): associated costs.
Dashed lines: averaged values.
given in Table E.2 in Appendix E. The total costs in Northwest Europe
are slightly higher than those in South Europe. This result is a combi-
nation of the higher labour costs, which are partly compensated by the
lower energy costs. The higher production rate in South Europe requires
more and larger equipment and, thus, more (in this case double) in-
vestments and, as a result, the investment costs per kg dry mass remains
about the same. On this scale of harvesting and dewatering, the max-
imum size of equipment is used and, then, no benefits from the scaling
rule (Eq. (4)) can be obtained.
The system defined in Table 3 concerns a harvesting and dewatering
system with an averaged feed rate. The productivity of cultivation
systems, however, varies during the seasons and results in variation of
feed rate for chemostat operated systems. The harvesting and dewa-
tering system must be designed for the maximum expected feed rate
from the cultivation units. As a consequence, in periods of low feed
rates the equipment is partially used, and then the costs are shared over
a lower amount of products.
Fig. 8-A shows an assumed pattern of variations of the feed rate over
a period of 12 months. The feed rate varies between 200 and
600 m3·hr− 1, with average value of 400 m3·hr− 1. The cost results for
the combination of pressure filtration and centrifugation are given in
Fig. 8-B. It is obvious that in periods of high cultivation productivity,
the costs are the lowest. In periods of low cultivation productivity the
costs increase by a factor 2.6. The average costs (taking seasonal
changes into consideration) given by the dashed line are about 60%
higher than those during the high productivity periods.
In periods of low capacity, installations can temporarily be swit-
ched-off, but the capital and maintenance burden of the equipment
continues. The energy consumption, which is related to the capacity, is
constant over all periods. For Fig. 8-B it is assumed that the number of
employees remains constant over the year, and this assumption has a
strong impact on the costs during the low capacity season. By using
seasonal labour, linked to the capacity, the contribution of labour will
F. Fasaei et al.
Fig. 9. Sensitivity analysis of selected parameters on the total harvesting and dewatering
costs (drying excluded).
be constant over the year (0.24 €·kg− 1). The costs in the lowest pro-
duction periods (period 1 and 12) will even reduce from 1.24 to
1 €·kg− 1. The average costs over all periods are then 36% higher than
those in the high productivity season.
3.4. Sensitivity analysis
Fig. 9 shows the sensitivity of the harvesting and dewatering costs
(combination of pressure filtration and centrifugation, excluding
drying) to ± 10% variations in investment costs, energy costs, Lang
factor, amortization, interest rate, maintenance, operation duration and
costs of labour. The effect of the individual variations on the total costs
is less than ± 4%. Labour, investment, Lang factor, and operation time
are the most important sources for variation. The impact of their var-
iation on the total costs is similar. It must be noted that these factors are
correlated, i.e., with a longer operational time the equipment is more
efficiently used and, thus, results in a reduction of investment costs.
Also, a 10% increment in the Lang factor has a similar effect as a 10%
increment in investment costs. The contribution of capital costs to the
total costs is in the range of 30–40%, and therefore variations of 10% in
these parameters does not show more than 4% change in the total costs.
In this work values for the specific energy uptake by equipment
were taken from the literature and by consulting equipment suppliers.
These values may have some uncertainty. The models were used to
evaluate the effect of variation of specific assumptions on the costs. This
was evaluated for the specific energy consumption, maximal capacity
per unit operation, biomass recovery, and the concentration after de-
watering. The model evaluations show that ± 10% variation in the
specific energy consumption has ± 2% effect on the costs and ± 10%
on the energy uptake per kg of algae.
Doubling the maximal capacity per unit operation reduces the costs
by 20% and has only a minor effect on the energy consumption. For
most unit operations, the recovery is in the range 0.95–0.99, except for
the flocculation units. Each percent improvement in recovery gives 1%
more biomass, and hence both costs and energy consumption decrease
by 1%. Ending harvesting and dewatering at a concentration of 10% dry
matter instead of 15% dry matter increases the costs by only 2%. The
reason for this last result is related to dimensions. Larger volumes of the
harvested biomass requires larger dewatering units. The dimensions of
this unit are based on the feed rate (product from harvesting step) and
not on the end concentration.
The quantitative analysis from this study revealed bottlenecks and
strengths of technologies. Pahl et al. addressed that the costs of har-
vesting and dewatering arise from the application of capital expensive
unit operations with a high energy demand [9]. However, the results in
this study showed that energy efficient technologies with low capital
costs (such as flocculation) suffer from a low biomass recovery, which
increases the costs by 25–30%. Moreover, not only capital and energy
355
Algal Research 31 (2018) 347–362
are the main contributors to the costs, labour has also a significant
contribution.
The statement by Grima et al. that the costs of harvesting are in the
range of 20–30% of the total costs of biomass production find, amongst
others, its origin in the research of Gudin et al. from the late 80s [5,44].
Estimated microalgae production costs are in the range of 4–6 € per kg
biomass in commercial scale cultivation systems [4,45]. With these
numbers and the costs from present study, the contribution of har-
vesting and dewatering to the total production costs are in the range
3–15%. Further, the contribution of the harvesting and dewatering to
the production of algal biomass decreased by upscaling and using state-
of-the-art process equipment. Continuous performance improvements
by equipment suppliers will further reduce the energy consumption and
the costs.
4. Conclusion
The performance of unit operations for harvesting and dewatering
microalgae is often expressed in qualitative terms. Moreover, the in-
teraction between unit operations in a chain is not addressed. This work
assessed the techno-economic performance of 28 scenarios for large
scale microalgae harvesting and dewatering. We found for harvesting
and dewatering of algal streams of 0.05% to 15% dry matter (open
pond system), that the cost range is between 0.3 and 2.0 €·kg− 1 algae
and the energy consumption goes up to 4.5 kWh·kg− 1 algae. For algal
broth from closed systems with a higher dry matter content the pro-
duction costs and energy consumption decrease to below 0.5 €·kg− 1
algae and below 0.5 kWh·kg− 1 algae. With these results, harvesting and
dewatering contribute 3–15% of the production costs of algae biomass.
The application of spiral plate technology for harvesting is currently
outside the given cost ranges. The maximum capacity for this method is
limited and requires a large number of units for large scale cultivation,
which raises the contribution of investments and labour to the total
costs. The lowest cost and energy consumption was achieved by ap-
plying pressure filtration for harvesting and centrifugation for dewa-
tering.
Single-step harvesting and dewatering requires unit operations that
can process algae to high biomass concentrations. Two-step operations
like pressure filtration followed by spiral plate technology or cen-
trifugation are attractive from a cost point of view, just as the chain of
flocculation followed by membrane filtration and a finishing step with
spiral plate technology or centrifugation. Flocculation for harvesting
followed in combination with a second unit operation require less than
0.1 kWh·kg− 1 algae. The low costs for energy are partially cancelled by
the additional costs for flocculants and the relatively low biomass re-
covery. The costs for flocculation systems are, therefore, comparable to
those of mechanical concentration methods. The impact of flocculants
in the water recycle stream to cultivation units, and on fractionation
and extraction steps, however, may limit the use of flocculants.
In all scenarios labour was a major cost contributor. The results are,
thus, sensitive to the choices related to labour. Additional investments
for a higher degree of automation can be compensated by the lower
labour cost.
Although there is no doubt over the important role that qualitative
data based analysis provides to pre-screening of existing potential
technologies for harvesting and dewatering steps, a quantitative model
based approach can provide deeper economic insight.
Acknowledgment
This work is performed within the TKI AlgaePARC Biorefinery
program with financial support from the Netherlands' Ministry of
Economic Affairs in the framework of the TKI BioBased Economy under
contract nr. TKIBE01009.Declaration of authors contributions
The conception and design of the study, all the calculations, analysis
F. Fasaei et al. Algal Research 31 (2018) 347–362

and interpretation of data has been done by F. Fasaei supported by
A.J.B. van Boxtel. The manuscript is written by F. Fasaei and A.J.B. van
Boxtel. The work was supervised and supported for improvement
with critical questions by J.H. Bitter. Edit was done by P.M.
Slegers.Declaration of conflicts
There are no known conflicts of interest associated with this
publication.Declaration of consent and/or animal use
The work concerned modelling and simulation. There are “no con-
flicts, informed consent, human or animal rights applicable”.

Appendix A. Short description of unit operations

A.1. Centrifugation

The driving force for separation during centrifugation is the difference in density between the microalgae cells and solvent. Different types of
industrial centrifuges can be used for continuous flows. The disc-stack centrifuge is suitable for the harvesting of microalgae with a size of around
5–10 μm. Moreover, disk-stack centrifuges require minimal manual intervention and they are more suitable for harvesting the microalgae compared
to multi-chamber and solid bowl centrifuges [29,46].

A.2. Spiral Plate Technology

Spiral plate technology (SPT) is a three phase separator (liquid/liquid/solid). Biomass is collected between rotating plates where, due to the
rotation, increased centrifugal g-forces exists. At given times the operation is shortly interrupted to discharge the collected biomass between the
plates. The main difference between spiral plate technology and disc-stack centrifugation is the short settling distance (3–6 mm). As a result the dry
solid content can reach higher values (up to 30% dry matter) than in disc-stack centrifugation units [47]. The energy efficiency is similar to that of a
centrifugation system.

A.3. Pressure and vacuum filtration

Filtration separates algae cells due to their size by a pressure difference over a filter. The fluid passes the filter, but oversized particles are
retained. With pressure filtration the pressure at the feed side is above atmospheric pressure, while for vacuum filtration a vacuum is created at the
filtrate side. Pressure and vacuum filtration are simple and efficient methods, which can recover large quantities of biomass and can work in
continuous operation [46,48].

A.4. Membrane filtration

Algae solutions can be concentrated by membrane filtration. Several types of membrane technologies can be applied. In reverse osmosis (RO),
water and small salt molecules permeate the filter. Microfiltration and ultrafiltration can be applied for harvesting of biomass and also for isolation of
components. Ultrafiltration (UF) is used to retain larger organic molecules like proteins and carbohydrates. Microfiltration (MF) can separate algae
cells from the solution. The operational pressure is 1–2 bar, 5 bar and 40 bar for MF, UF and RO, respectively. Microfiltration and also ultrafiltration
are applied for harvesting and dewatering of rather large algae particles from the cultivated solution [49–51]. With increasing dry matter content,
the performance of the membrane decreases due to concentration polarization and fouling. Membrane systems can concentrate the algal biomass up
to 5% dry matter. At that level of concentration polarization and fouling are severe and the flux declines too far to be effective. Several membrane
modules are available in the market. Table A.1 provides a summary of the specifications for each module. The energy consumption in the membrane
system is related to flow rate of permeate and pressure requirements. The pumping energy required to achieve sufficient cross flow velocity is also
part of the costs, which is dependent on flow rate.

Table A.1
Summary of characteristics of different membrane modules.

Membrane modules Tendency to fouling Plant investment Ref.

Spiral wound Average Low [49]

Tubular membranes Low High and low [49]
Flat sheet Average High [49]

A.5. Flocculation

Flocculants interact with the surface of algae cells resulting in coagulation of algae. The aggregated particles coalesce into larger flocs. These flocs
are separated from the medium by sedimentation. Flocculation occurs in three steps, 1) intense mixing during 3 min, 2) moderate mixing for 20 min,
and 3) settling over 60 min [52]. Because of the different requirements in these phases, the operation occurs best in three different tanks. The energy
demand for flocculation is related to the mixing phases. Several flocculants can be applied, all adhere to this principle, like chitosan, poly-glutamate
and also polymeric flocculants. Polymeric flocculants, such as cationic polymers, are Zetag 7557 (BASF, Germany), Synthofloc 5080H (Sachtleben,
Germany) and SNF H536 (SNF-Floerger, France).

A.6. Spray drying

In spray drying hot air (over 100 °C) is used to evaporate the water from atomized algae droplets [53]. In the initial phase of spray drying algae
droplets are wet, and as a result the temperature of the algae remains at the wet bulb temperature. Towards the exhaust of the dryer the particles

356
F. Fasaei et al. Algal Research 31 (2018) 347–362

heat-up, and the air temperature falls. In standard spray drying operations, there is still a significant difference between the product and air
temperature at the exhaust of a spray dryer. The inlet air temperature should be chosen such that the exhaust air temperature remains in the range
60–90 °C, where the dried particles' temperature remains below 45 °C [46].

A.7. Drum drying

An alternative for spray drying is drum drying. For drum drying the algae paste is distributed over a rotating drum, which is internally heated by
steam [46]. Due to heat transfer water evaporates from the paste [29]. After about half a rotation of the drum the paste is dry and collected. In some
cases the drum is placed in a vacuum system, which results in an increased evaporation rate. This last system is, however, a batch wise operation
with the same disadvantages as freeze drying. The product temperatures in drum dryers can exceed protein denaturation temperature. These systems
are therefore not preferred for drying of high quality proteins. If only lipids are required as an end product, these systems can be considered.

Appendix B. Mass and energy balances

The general mass balance for algae biomass is:

Fin Cin = Fout Cout + Fco,out Cco,out (B.1)
Eq. (B.1) is complemented with expressions for the degree of recovery:
Fin Cin R = Fout Cout (B.2)

Fin Cin (1 − R) = Fco,out Cco,out (B.3)

−1
where F represents the volumetric in/out flow rates for streams containing algal biomass (m ⋅ h ), and co is the co-stream that mainly contains
3

water or flocculants and a small fraction of algae. C represents the concentration of algal biomass (kg ⋅ m− 3) in each stream, and R is the recovery of
the biomass.
The energy consumption for, centrifugation, spiral plate technology, pressure and vacuum filtration and microfiltration is based on the specific
energy consumption of these operations. For centrifugation and spiral plate technology the specific energy consumption is defined per m3 feed:
H = EFin (B.4)
For pressure and vacuum filtration, the energy consumption is related to volumetric flow of feed while for membrane filtration is related to
volumetric flow rate of the permeate:
H = EFco,out (B.5)
Values/expressions for the specific energy consumption are given in Table C.1.
The energy consumption for flocculation concerns the mixing steps. The energy consumption is based on a propeller stirrer according to Doran
[54] with height-impeller diameter ratio of 0.33 to tank diameter:
The mixing with turbulent flow:
E = kρN 3D5 (B.6)
The diameter of the tanks is calculated as:
0.33
4
D = ⎛Vol⎜
⎞ ⎟

⎝ πHD ratio ⎠ (B.7)

H = 2.778e−7 Et (B.8)
−3
where E is the energy required for mixing (W), k = 0.4 is constant, N is stirring speed (rps), ρ is mass density of the fluid (kg·m ), D is the diameter
of mixing tank, HDratio is the height to diameter ratio of the tank, Vol is the volume of tank for the required residence time, and t is the mixing time
(s).

Appendix C. General data: properties, process conditions and specific energy requirements

The physical properties of water-algae solution is given in Table C.1. Specific energy requirement and reachable total solid matter for unit
operations derived from literature are given in Table C.2. Information for the flocculation systems is given in Table C.3.

Table C.1
Physical properties of algal biomass at ambient temperature.

Symbol Value Unit Reference

ρw Density water 1000 kg·m− 3

ρA Density algae biomass 1030 kg·m− 3 [55]
Cp, WA heat capacity of algae water mixture 4181 J kg− 1 K− 1 Equal to water
Cp, W heat capacity of water 4181 J kg− 1 K− 1
Cp, air heat capacity of air 1 J kg− 1 K− 1

357
F. Fasaei et al. Algal Research 31 (2018) 347–362

Table C.2
Feasible outputs and specific energy requirement of unit operations.

Technology Maximal solid output concentration Biomass recovery Energy requirement (kWh·m− 3 Reference
(%) (%) feed)

Centrifuge 10–20 95–99 0.70–1.30 [2,9,24]

Spiral plate technology (SPT) 20–22 95–99 0.95–2.00 [1,9,11]
Pressure filter 22–27 98 0.50–0.90 [5,13,56]
Vacuum filter 18–22 98 1.22–5.90 [5,12–14]
Membrane filtration (spiral 1.5–10 99 0.80–2.51 (kWh·m− 3 permeate) [15–19,57]
wound)
Chemical flocculation 3–8 80–98 0.15 [13,16,21,22]
Drum dryer 90–95 99 0.90 (kWh·kg− 1 evaporated [23,24]
water)
Spray dryer 90–95 99 1.0–1.2 (kWh·kg− 1 evaporated [24]
water)

Table C.3
Operational conditions for harvesting with flocculation.

Technology pH Maximum biomass recovery (%) Dosage Other conditions Reference

Cationic flocculation 7.5 80–90 162–167 mgflocculant·g− 1 biomass Fresh water [36]
Chitosan flocculation 7 85–98 4–38 mgflocculant·g− 1 biomass Fresh water [22,37]

Appendix D. Tables with specific data

Table D.1
Scale-up information and capital costs for each unit operation for harvesting and dewatering.

Equipment/materials Capacity/size Costs at given Scale Max. feasible Max capacity/size Ref.
capacity/size factor n concentration

Centrifuge 80 m3·h− 1 250,000 € 0.6 200 kg·m− 3 120 m3·hr− 1 [24]

Microfiltration – 300 €·m− 2 1.00 50 kg·m− 3 5000 m2 [49]
Stainless steel tanks for flocculation 100 m3 125,000 € 0.35 400 m3 [2,58]
10,000 € for each stirrer
Pressure filter (plate) 50 m2 75,900 € 0.55 200 kg·m− 3 100 m2 [59]
Vacuum rotary filter 50 m2 240,000 € 0.66 200 kg·m− 3 100 m2 [59,58]
Spiral plate technology (SPT) 50 4 m3·h− 1 229,000 € 0.6 200 kg·m− 3 4 m3·h− 1 [60]
Spray dryer 190 kg·h− 1 water 841,323 € 0.60 5% water in 10,000 kg·h− 1 water [24,60]
evaporated product evaporated
Drum dryer 1000 kg·h− 1 water 270,000 € 0.33 5% water in [2,23,61]
evaporated product
Chitosan – – 25 kg·m− 3 [37,62]
Cationic polymer – – 25 kg·m− 3 [63]

Table D.2
Specific elements for labour cost estimation [3,35].

Labour

Base salary 9 (€·hr− 1)

Operator 3 × base salary 5 units per operator
Supervisor 4.3 × base salary 4 operators per supervisor
Manager 6.7 × base salary 20 operators per manager
Additional overhead 20% of total labour

358
F. Fasaei et al. Algal Research 31 (2018) 347–362

Table D.3
Industrial prices for energy [35].

Energy Price (€·kWh− 1)

Electricity 0.10
Natural gas heating 0.04
Steam 0.05

Table D.4
Applied economic parameters in capital cost estimation [3].

Applied values for cost estimation

I 6% Interest rates (% of investment)

M 5% Maintenance (% of investment)
A 15 years Amortization period
W 7200 h Number of hours in a year
Lfactor 3.5 Lang factor

Table D.5
List of applied consumables.

Equipment/materials Consumables Ref.

Microfiltration Replacement once in 3 years, 100 €·m− 2 [49]

Chitosan 10–25 (€·kg− 1) [37,62]
Cationic polymer 3.5–4.5 (€·kg− 1) [63]

Table D.6
Specific data for two different production locations.

Northwest Europe South Europe

Productivity (ton·ha− 1·year− 1) 15 30

Minimum wage (€·hr− 1) 9.0 4.5
Electricity costs (€·kWh− 1) 0.10 0.12

Appendix E. Results

Table E.1
Overview of the scenarios, energy consumption and costs for harvesting and dewatering for open ponds (400 m3·hr− 1, 0.5 kg·m− 3), tubular
(280 m3·hr− 1, 1.5 kg·m− 3) and flat panel systems (250 m3·hr− 1, 2.5 kg·m− 3).

Routes Harvesting Open pond Tubular systems Flat panel systems

Energy Cost Energy Cost Energy Cost

(kWh·kg− 1) (€·kg− 1) (kWh·kg− 1) (€·kg− 1) (kWh·kg− 1) (€·kg− 1)

1 Membrane filter Centrifuge – 4.24 1.45 1.41 0.51 0.84 0.30

2 Membrane filter Spiral plate – 4.25 1.52 1.42 0.64 0.85 0.42
technology
3 Pressure filter Centrifuge – 1.03 0.50 0.36 0.18 0.22 0.12
4 Pressure filter Spiral plate – 1.04 0.54 0.36 0.25 0.22 0.18
technology
5 Vacuum filter Centrifuge – 2.64 1.07 0.89 0.38 0.54 0.24
6 Vacuum filter Spiral plate – 2.65 1.12 0.90 0.44 0.55 0.30
technology
7 Cationic Membrane filter Pressure filter 0.06 1.20 0.05 1.02 0.05 0.97
flocculation

359
F. Fasaei et al. Algal Research 31 (2018) 347–362

8 Cationic Membrane filter Vacuum filter 0.07 1.21 0.07 1.03 0.07 0.98
flocculation
9 Cationic Membrane filter Centrifuge 0.08 1.25 0.08 1.06 0.08 1.01
flocculation
10 Cationic Membrane filter Spiral plate 0.09 1.32 0.09 1.16 0.09 1.12
flocculation technology
11 Cationic Pressure filter – 0.03 1.13 0.02 0.98 0.02 0.94
flocculation
12 Cationic Vacuum filter – 0.06 1.15 0.05 0.99 0.05 0.95
flocculation
13 Cationic Centrifuge – 0.07 1.18 0.06 1.02 0.06 0.98
Flocculation
14 Cationic Spiral plate – 0.09 1.40 0.09 1.25 0.09 1.22
Flocculation technology
15 Chitosan Membrane filter Pressure filter 0.06 0.43 0.05 0.23 0.05 0.18
flocculation
16 Chitosan Membrane filter Vacuum filter 0.08 0.44 0.07 0.24 0.07 0.18
flocculation
17 Chitosan Membrane filter Centrifuge 0.09 0.46 0.08 0.25 0.08 0.19
Flocculation
18 Chitosan Membrane filter Spiral plate 0.09 0.53 0.09 0.35 0.09 0.31
flocculation technology
19 Chitosan Pressure filter – 0.03 0.36 0.02 0.19 0.02 0.15
flocculation
20 Chitosan Vacuum filter – 0.06 0.38 0.05 0.21 0.05 0.16
flocculation
21 Chitosan Centrifuge – 0.07 0.39 0.06 0.21 0.06 0.16
Flocculation
22 Chitosan Spiral plate – 0.09 0.61 0.09 0.46 0.09 0.40
flocculation technology
23 Centrifuge – – 1.94 0.48 0.76 0.19 0.48 0.12
24 Centrifuge Spiral plate – 2.10 0.62 0.82 0.30 0.53 0.21
technology
25 Spiral plate – – 2.11 13.50 0.70 4.50 0.42 2.71
technology
26 Spiral plate Centrifuge – 2.23 13.73 0.76 4.76 0.46 2.87
technology
27 Pressure filter – – 0.96 0.44 0.32 0.15 0.19 0.10
28 Vacuum filter – – 2.49 0.96 0.83 0.33 0.50 0.21

Table E.2
Harvesting and dewatering costs and energy consumption for an open pond system in South Europe with double capacity, compared to Northwest
Europe with adapted productivity, labour, and energy base costs are given in Table D.6 in Appendix D.

Routes Harvesting Dewatering Open pond

Energy (kWh·kg− 1) Cost (€·kg− 1)

1 Membrane filter Centrifuge – 4.244 1.465

2 Membrane filter Spiral plate technology – 4.250 1.551
3 Pressure filter Centrifuge – 1.032 0.400
4 Pressure filter Spiral plate technology – 1.036 0.436
5 Vacuum filter Centrifuge – 2.640 1.006
6 Vacuum filter Spiral plate technology – 2.647 1.041
7 Cationic flocculation Membrane filter Pressure filter 0.059 1.106
8 Cationic flocculation Membrane filter Vacuum filter 0.074 1.114
9 Cationic flocculation Membrane filter Centrifuge 0.084 1.152
10 Cationic flocculation Membrane filter Spiral plate technology 0.093 1.242
11 Cationic flocculation Pressure filter – 0.027 1.071
12 Cationic flocculation Vacuum filter – 0.058 1.087
13 Cationic flocculation Centrifuge – 0.070 1.121
14 Cationic flocculation Spiral plate technology – 0.092 1.320
15 Chitosan flocculation Membrane filter Pressure filter 0.060 0.325
16 Chitosan flocculation Membrane filter Vacuum filter 0.075 0.333

360
F. Fasaei et al. Algal Research 31 (2018) 347–362

17 Chitosan flocculation Membrane filter Centrifuge 0.085 0.347

18 Chitosan flocculation Membrane filter Spiral plate technology 0.094 0.442
19 Chitosan flocculation Pressure filter – 0.028 0.297
20 Chitosan flocculation Vacuum filter – 0.059 0.313
21 Chitosan flocculation Centrifuge – 0.071 0.323
22 Chitosan flocculation Spiral plate technology – 0.093 0.533
23 Centrifuge – – 1.940 0.434
24 Centrifuge Spiral plate technology – 2.065 0.514
25 Spiral plate technology – – 2.110 10.946
26 Spiral plate technology Centrifuge – 2.230 11.544
27 Pressure filter – – 0.960 0.360
28 Vacuum filter – – 2.490 0.936

References

[1] R.B. Draaisma, et al., Food commodities from microalgae, Curr. Opin. Biotechnol. 24 (2) (2013) 169–177.
[2] A.I. Barros, et al., Harvesting techniques applied to microalgae: a review, Renew. Sust. Energ. Rev. 41 (2015) 1489–1500.
[3] J. Ruiz, et al., Towards industrial products from microalgae, Energy Environ. Sci. 9 (2016) 3036–3043.
[4] N.-H. Norsker, et al., Microalgal production — a close look at the economics, Biotechnol. Adv. 29 (1) (2011) 24–27.
[5] E. Molina Grima, et al., Recovery of microalgal biomass and metabolites: process options and economics, Biotechnol. Adv. 20 (7–8) (2003) 491–515.
[6] N. Uduman, et al., Dewatering of microalgal cultures: a major bottleneck to algae-based fuels, J. Renewable Sustainable Energy (2010) 2(1).
[7] I. Rawat, et al., Dual role of microalgae: phycoremediation of domestic wastewater and biomass production for sustainable biofuels production, Appl. Energy 88 (10) (2011)
3411–3424.
[8] L. Brennan, P. Owende, Biofuels from microalgae—a review of technologies for production, processing, and extractions of biofuels and co-products, Renew. Sust. Energ. Rev. 14 (2)
(2010) 557–577.
[9] S.L. Pahl, et al., Harvesting, thickening and dewatering microalgae biomass, in: M.A. Borowitzka, N.R. Moheimani (Eds.), Algae for Biofuels and Energy, Springer, Netherlands:
Dordrecht, 2013, pp. 165–185.
[10] P.M. Slegers, et al., A model-based combinatorial optimisation approach for energy-efficient processing of microalgae, Algal Res. 5 (0) (2014) 140–157.
[11] H.A. Boele, Separating Device and Method, Google Patents, 2013.
[12] A. Wileman, A. Ozkan, H. Berberoglu, Rheological properties of algae slurries for minimizing harvesting energy requirements in biofuel production, Bioresour. Technol. 104 (2012)
432–439.
[13] Y. Shen, W. Yuan, Z.J. Pei, Q. Wu, E. Mao, Microalgae Mass Production Methods, Transactions of the ASABE. 52 (4) (2009) 1275–1287.
[14] G. Shelef, A.S., M, Green, Microalgae Harvesting and Processing, A Literature Review. Technion Research and Development Foundation Ltd, Haifa, Israel, 1984.
[15] M.L. Gerardo, M.A. Zanain, R.W. Lovitt, Pilot-scale cross-flow microfiltration of Chlorella minutissima: a theoretical assessment of the operational parameters on energy con-
sumption, Chem. Eng. J. 280 (2015) 505–513.
[16] A. Darzins, P.P., L. Edye, Current status and potential for algal biofuels production, A Report to IEA Bioenergy Task 39, R. T39-T2, 2010.
[17] M.R. Bilad, et al., Harvesting microalgal biomass using submerged microfiltration membranes, Bioresour. Technol. 111 (2012) 343–352.
[18] I. Rawat, et al., Biodiesel from microalgae: a critical evaluation from laboratory to large scale production, Appl. Energy 103 (2013) 444–467.
[19] J. Milledge, S. Heaven, A review of the harvesting of micro-algae for biofuel production, Rev. Environ. Sci. Biotechnol. 12 (2) (2013) 165–178.
[20] J.K. Pittman, A.P. Dean, O. Osundeko, The potential of sustainable algal biofuel production using wastewater resources, Bioresour. Technol. 102 (1) (2011) 17–25.
[21] E.S. Beach, et al., Preferential technological and life cycle environmental performance of chitosan flocculation for harvesting of the green algae Neochloris oleoabundans, Bioresour.
Technol. 121 (2012) 445–449.
[22] G.P. ˋt Lam, et al., Cationic polymers for successful flocculation of marine microalgae, Bioresour. Technol. 169 (2014) 804–807.
[23] ANDRITZ, Available from: http://www.andritz.com.
[24] GEA, Available from: http://www.gea.com/nl/index.jsp.
[25] C.-L. Chen, J.-S. Chang, D.-J. Lee, Dewatering and drying methods for microalgae, Dry. Technol. 33 (4) (2015) 443–454.
[26] M.S. Farid, et al., Using nano-chitosan for harvesting microalga Nannochloropsis sp, Bioresour. Technol. 131 (2013) 555–559.
[27] C.O. Letelier-Gordo, et al., Effective harvesting of the microalgae Chlorella protothecoides via bioflocculation with cationic starch, Bioresour. Technol. 167 (2014) 214–218.
[28] P.A. Hansel, R. Guy Riefler, B.J. Stuart, Efficient flocculation of microalgae for biomass production using cationic starch, Algal Res. 5 (2014) 133–139.
[29] M.A. Borowitzka, et al., Harvesting, thickening and dewatering microalgae biomass, Algae for Biofuels and Energy, Springer, Netherlands, 2013.
[30] M.K. Danquah, et al., Dewatering of microalgal culture for biodiesel production: exploring polymer flocculation and tangential flow filtration, J. Chem. Technol. Biotechnol. 84 (7)
(2009) 1078–1083.
[31] A.E.M. Abdelaziz, G.B. Leite, P.C. Hallenbeck, Addressing the challenges for sustainable production of algal biofuels: II. Harvesting and conversion to biofuels, Environ. Technol. 34
(13–14) (2013) 1807–1836.
[32] A.K. Lee, D.M. Lewis, P.J. Ashman, Energy requirements and economic analysis of a full-scale microbial flocculation system for microalgal harvesting, Chem. Eng. Res. Des. 88 (8)
(2010) 988–996.
[33] H.J. Lang, Engineering approach to preliminary cost estimates, Chem. Eng. 54 (1947) 130–133.
[34] H.J. Lang, Cost relationships in preliminary cost estimation, Chem. Eng. 54 (1947) 117.
[35] eurostat Statistics Explained, Available from: http://ec.europa.eu/eurostat/statistics-explained/index.php/Energy_price_statistics.
[36] D. Vandamme, et al., Flocculation of microalgae using cationic starch, J. Appl. Phycol. 22 (4) (2010) 525–530.
[37] F. Roselet, et al., Screening of commercial natural and synthetic cationic polymers for flocculation of freshwater and marine microalgae and effects of molecular weight and charge
density, Algal Res. 10 (2015) 183–188.
[38] M.R. Granados, et al., Evaluation of flocculants for the recovery of freshwater microalgae, Bioresour. Technol. 118 (2012) 102–110.
[39] A. Beim, A. Beim, Comparative ecological–toxicological data on determination of maximum permissible concentrations (MPC) for several flocculants, Environ. Technol. 15 (2)
(1994) 195–198.
[40] L. Borges, et al., Effects of flocculants on lipid extraction and fatty acid composition of the microalgae Nannochloropsis oculata and Thalassiosira weissflogii, Biomass Bioenergy 35
(10) (2011) 4449–4454.
[41] S. Şirin, et al., Harvesting the microalgae Phaeodactylum tricornutum with polyaluminum chloride, aluminium sulphate, chitosan and alkalinity-induced flocculation, J. Appl. Phycol.
24 (5) (2012) 1067–1080.
[42] D. Vandamme, I. Foubert, K. Muylaert, Flocculation as a low-cost method for harvesting microalgae for bulk biomass production, Trends Biotechnol. 31 (4) (2013) 233–239.
[43] Z. Ahmad, Principles of Corrosion Engineering and Corrosion Control, Butterworth-Heinemann, Oxford, 2006.
[44] C. Gudin, Bioconversion of solar energy into organic chemicals by microalgae, J. Adv. Biotechnol. Processes 6 (1986) 73–110.
[45] R.H. Wijffels, M.J. Barbosa, M.H.M. Eppink, Microalgae for the production of bulk chemicals and biofuels, Biofuels Bioprod. Biorefin. 4 (3) (2010) 287–295.
[46] E.M. Grima, F.A. Fernández, A.R. Medina, 10 Downstream processing of cell-mass and products, Handbook of Microalgal Culture: Biotechnology and Applied Phycology, 2004, p.
215.
[47] T. Minowa, et al., Oil production from algal cells of Dunaliella tertiolecta by direct thermochemical liquefaction, Fuel 74 (12) (1995) 1735–1738.
[48] P.M. Doran, 10 - Unit operations, Bioprocess Engineering Principles, Academic Press, London, 1995, pp. 218–253.
[49] J. Wagner, Membrane Filtration Handbook: Practical Tips and Hints, (2001).
[50] A.S. Grandison, Chapter 5 - microfiltration, Separation Processes in the Food and Biotechnology Industries, Woodhead Publishing, 1996, pp. 141–153.
[51] M.J. Lewis, Chapter 4 - ultrafiltration, Separation Processes in the Food and Biotechnology Industries, Woodhead Publishing, 1996, pp. 97–139.

361
F. Fasaei et al. Algal Research 31 (2018) 347–362

[52] M. Eddy, B.J. CLark, J.M. Morriss (Eds.), MetCalf & Eddy: Wastewater Engineering, 1991.
[53] K. Masters, Spray Drying, Leonard Hill Books, London, 1972.
[54] P.M. Doran, 7 - Fluid flow and mixing, Bioprocess Engineering Principles, Academic Press, London, 1995, pp. 129–163.
[55] R. Bosma, et al., Ultrasound, a new separation technique to harvest microalgae, J. Appl. Phycol. 15 (2) (2003) 143–153.
[56] M.K. Danquah, L. Ang, N. Uduman, N. Moheimani, G.M. Forde, Dewatering of microalgal culture for biodiesel production: exploring polymer flocculation and tangential flow
filtration, J. Chem. Technol. Biotechnol. 84 (2009) 1078–1083.
[57] D.M. Krstić, et al., Energy-saving potential of cross-flow ultrafiltration with inserted static mixer: application to an oil-in-water emulsion, Sep. Purif. Technol. 57 (1) (2007)
134–139.
[58] NETL, Available from: https://www.netl.doe.gov.
[59] DACE, Dutch Association of Cost Engineering ICEC member, Ed. 30. The Huge, The Netherlands: Michael Sprong.
[60] evodos, Available from: http://www.evodos.eu.
[61] K.K. Sharma, S. Garg, Y. Li, A. Malekizadeh, P.M. Schenk, Economics of producing fuel pellets from biomass, Appl. Eng. Agric. 22 (3) (2006) 421–426.
[62] K.K. Sharma, S. Garg, Y. Li, A. Malekizadeh, P.M. Schenk, Critical analysis of current microalgae dewatering techniques, Biofuels 4 (2013) 397–407.
[63] G.P. ˋt Lam, et al., Dosage effect of cationic polymers on the flocculation efficiency of the marine microalga Neochloris oleoabundans, Bioresour. Technol. 198 (2015) 797–802.

362