Biotechnology Advances 30 (2012) 1344–1353

Contents lists available at SciVerse ScienceDirect

Biotechnology Advances
journal homepage: www.elsevier.com/locate/biotechadv

Research review paper

Production cost of a real microalgae production plant and strategies to reduce it

F.G. Acién a,⁎, J.M. Fernández a, J.J. Magán b, E. Molina a
a
Department of Chemical Engineering, University of Almería, Almería 04071 Spain
b
Department of Greenhouse Technology, Estación Experimental Fundación Cajamar, Almería 04710, Spain

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

Available online 14 February 2012 The cost analysis of a real facility for the production of high value microalgae biomass is presented. The facil-
ity is based on ten 3 m3 tubular photobioreactors operated in continuous mode for 2 years, data of Scenedesmus
Keywords: almeriensis productivity but also of nutrients and power consumption from this facility being used. The yield of
Costs analysis the facility was close to maximum expected for the location of Almería, the annual production capacity being
Microalgae 3.8 t/year (90 t/ha·year) and the photosynthetic efficiency being 3.6%. The production cost was 69 €/kg.
Tubular photobioreactor
Economic analysis shows that labor and depreciation are the major factors contributing to this cost. Simpli-
Production cost
fication of the technology and scale-up to a production capacity of 200 t/year allows to reduce the produc-
tion cost up to 12.6 €/kg. Moreover, to reduce the microalgae production cost to approaches the energy or
commodities markets it is necessary to reduce the photobioreactor cost (by simplifying its design or mate-
rials used), use waste water and flue gases, and reduce the power consumption and labor required for the
production step. It can be concluded that although it has been reported that production of biofuels from
microalgae is relatively close to being economically feasible, data here reported demonstrated that to
achieve it by using the current production technologies, it is necessary to substantially reduce their costs
and to operate them near their optimum values.
© 2012 Elsevier Inc. All rights reserved.

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1345
2. Materials and methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1345
2.1. Microorganism and culture conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1345
2.2. Production facility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1345
2.3. Cost analysis methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1346
3. Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1346
3.1. Biomass productivity, mass balance and energy balance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1346
3.2. Determination of production cost . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1347
4. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1347
4.1. Overall analysis of the performance of the facility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1347
4.2. Microalgae production cost of case study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1350
4.3. How to reduce the production costs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1351
4.4. Minimum microalgae production cost . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1352
5. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1353
Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1353
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1353

⁎ Corresponding author at: Department of Chemical Engineering, University of Almería, Cañada San Urbano, E-04120-Almería, Spain. Tel.: + 34 950 015443; fax: + 34 950
015484.
E-mail address: facien@ual.es (F.G. Acién).

0734-9750/$ – see front matter © 2012 Elsevier Inc. All rights reserved.
doi:10.1016/j.biotechadv.2012.02.005
 F.G. Acién et al. / Biotechnology Advances 30 (2012) 1344–1353
1. Introduction
Recently microalgae biomass has been proposed as a raw material
for the production of energy and other commodities for several rea-
sons such as their high productivity, the possibility of using low qual-
ity water including seawater, and the fact that no fertile land is
needed (Chisti, 2007, 2008; Gouveia and Oliveira, 2009; Patil, et al.,
2008; Wigmosta et al., 2011). However, until now microalgae have
been only produced with the purpose of obtaining high value prod-
ucts mainly related to applications for humans (health, cosmetics, nu-
traceutical and foods) and aquaculture (Borowitzka, 1999; Richmond,
2000). For these applications, the amount of biomass needed is very
small compared to the requirements of markets such as energy or
commodities. The microalgae biomass market produces only 5 kt/year
at production costs of 25000 $/t (Pulz and Gross, 2004). However, to re-
place only a 5% of the US demand of fuel for transport it is necessary to
produce more than 66000 kt/year of oil rich biomass at production
costs below 400 $/t (Chisti, 2007). Moreover, to replace all transport
fuels in Europe by biodiesel from microalgae, 9.25 million ha (almost
the surface area of Portugal) would be needed, assuming a productivity
of 40000 l/ha·year (Wijffels and Barbosa, 2010).
Although the potential of microalgae to contribute to the world en-
ergy and commodities demand is high, there is a large gap between the
current available technology and the one needed to supply the potential
world demand. The technology must be scaled-up several orders of
magnitude to significantly contribute to the biofuels market, and the
biomass production cost must be also reduced. Thus, it is still necessary
to solve a large number of bottlenecks related with biological, engineer-
ing and economic aspects (Richmond, 2000). Biological and engineering
aspects of the problem have been studied in the last century and still are
being analyzed. Concerning economic analysis of the problem, due to
the lack of existing facilities and including of a defined technology,
only approximations can be performed, all of them assuming large un-
certainty. Recently, several approximations to this problem have been
published (Douskova, et al., 2009; Norsker, et al., 2011; Richardson, et
al., 2010; Singh and Gu, 2010; Wigmosta et al., 2011; Wijffels, et al.,
2010; Williams and Laurens, 2010). Alternatively, more accurate data
can be obtained from the facilities that are now existing.
In the present paper, the cost analysis of a medium-scale plant for
the production of microalgae is presented. The validity of the
manufacturing process proposed has been tested and it is used here
to estimate the costs of production at large scale. Moreover, this anal-
ysis can be used to identify the major factors determining the produc-
tion costs, helping to identify the key technical problems to be solved
to achieve economic viability.
2. Materials and methods
2.1. Microorganism and culture conditions
The microalga used was Scenedesmus almeriensis (CCAP 276/24,
Culture Collection of Algae and Protozoa of the Centre for Hydrology
and Ecology, Ambleside, UK). The optimal conditions for the produc-
tion of this strain have been reported elsewhere (Sánchez et al.,
2008a, 2008b). The culture medium used was Mann&Myers (Mann
and Myers, 1968) prepared using agricultural fertilizers instead of
pure chemicals. The microalga was grown photoautotrophically in tu-
bular photobioreactors aerated to avoid dissolved oxygen accumula-
tion, under pH controlled by on-demand injection of pure CO2, and
temperature below 35 °C. The cultures were operated in continuous
mode at a dilution rate of 0.34 1/day.
2.2. Production facility
The microalgal production facility used is located in “Estación Exper-
imental Las Palmerillas”, property of Fundación CAJAMAR (Almería,
1345
Spain). This facility is similar in design and operation to an industrial
plant although it is used for research. The core of the process is a set
of ten tubular fence-type photobioreactors built as previously described
(Acién Fernández et al., 2001; Alías et al., 2004; Molina, et al., 2001).
Each photobioreactor is made of 400 m long tube of 0.09 m diameter,
with a bubble column 3.5 m high and 0.4 m diameter for degassing
and heat exchange. Diameter of the tube was optimized to maximize
the volume of culture per reactor but minimizing yield losses by exces-
sive light path to photosynthesis. The tubes are optimally arranged to
maximize the interception of solar radiation. Liquid and gas flow rates
entering each photobioreactor are measured using flowmeters; the
pH, temperature and dissolved oxygen at the end of the loop are mea-
sured using Crison probes (Crison Instruments, Spain), connected to a
control-transmitter unit Crison MM44, that sends the information to a
PC control unit, allowing a complete monitoring and control of the facil-
ity. The simplified flowchart of the production process used is shown in
Fig. 1. In addition to photobioreactors, the facility is equipped with all
the necessary ancillaries as an RITEC Fertilizer Unit (Almería, Spain)
used for the automatic preparation of culture medium from fertilizers
and fresh water. The sterilization of the culture medium is performed
by filtration/ozonization online. The culture medium is pumped daily
to the photobioreactors and the harvest is continuously centrifuged by
using a continuous decanter (solid–liquid centrifugation unit, Flotweg,
Germany) to obtain sludge with a 15% dry matter content. The biomass
sludge is freeze-dried in a Cuddom Freeze-dryer (Blenheim, New
Zealand) to obtain dry biomass as final product. Each reactor is bubbled
at constant airflow rate of 200 l/min and the pH is controlled by on-
demand injection of pure CO2 at 3 l/min. The temperature of the culture
is controlled by passing, when needed as determined by the computer
control, cooling water at 1500 l/h through an internal heat exchanger
Fig. 1. Schematic block-diagram of the production process used for economic analysis.
Culture medium is prepared on-line by a fertirrigation unit of 4 m3/h capacity. The cul-
ture medium is sterilized by filtration up to 1 μm pore size and ozone at doses of 10 g/
m3, previously to be introduced into the photobioreactors on daylight period. Harvest
is centrifuged at 9500 rpm at a flow rate of 2 m3/h, a sludge of 15–20%d.wt. being
obtained. The sludge is freeze-dried for 24 h in a equipment with capacity for 80 kg/
day of water, dry biomass with humidity lower than 4% being obtained. Finally, the
dry biomass is milled till a 300 μm particle size.
1346 F.G. Acién et al. / Biotechnology Advances 30 (2012) 1344–1353

located in the degasser column of each photobioreactor. The volume-to-

surface (V/S) ratio of the reactors is 0.07 m 3/m 2. The culture is circulat-
ed through the photobioreactors using a centrifugal pump at 0.9 m/s.
Additional energy is consumed by the air blower that supplies the aer-
ation stream, by the pumps used to prepare medium and operate the
decanter, and by the freeze-dryer. The complete system has been design
and built by the department of Chemical Engineering of University of
Almería (Spain), as well as the data capture system and the control soft-
ware (developed in DaqFactory 5.0, Azeotech Inc.) that controls and
monitors all the activity in the facility.

2.3. Cost analysis methodology

To assess the production cost of the microalgae biomass produced

in the described facility it is necessary to describe the flowchart of the
process in detail, including a list of equipment and its size (Kalk and
Langlykke, 1986). The cost of the major equipment has to be deter-
mined from bibliography or, if possible, from the suppliers. The total
fixed capital can be calculated after the major equipment cost is
known, by multiplying the corresponding Lang factors according to
the nature of the item. The value of these factors is available for a
wide variety of processes. The total production cost can be calculated
as the sum of the depreciation plus the direct production costs. The
depreciation includes not only amortization of the fixed capital,
which is a function of the lifetime considered, but also the property
tax, insurance and the purchase tax. The direct production costs in-
clude raw materials, utilities, labor and others (supervision, mainte-
nance, tax, contingency, etc.). The amount of raw materials required
can be calculated from mass balances accordingly to the specified
flowchart, whereas the consumption of utilities can be calculated
from the power and water use of the process. Labor includes the man-
power necessary for the correct operation of the process and the
general costs of supervision and management, in addition to mainte-
nance, taxes and contingency. To determine the direct production
cost it is necessary to know the cost of the raw materials, power,
water and labor, while the other costs can be calculated by factors
previously defined.
Following this routine, it is possible to ascertain the production
cost of microalgae biomass at any other scale just by modifying the
cost of the major equipment according to the scale chosen. The cost
of a process or equipment can be scaled up or down from a basic
size by using an exponential law for which a value of 0.85 is consid-
ered in this particular scope. This equation is not valid for very large
changes of scale because a certain technology can be feasible at one
scale but not be available at larger scale. Thus, a maximum scale-up
factor of 10 has been considered acceptable in this study without re-
vising the technology. Whenever larger requirements are needed, the
Fig. 2. Variation of (a) Solar radiation and temperature, (b) biomass concentration and
scale-up has been solved by multiplying the number of units.
(c) productivity along the year for the production of Scenedesmus almeriensis in a semi-
  industrial facility 30 m3 at Almería, Spain.
SizeB 0:85
CostB ¼ CostA ð1Þ
SizeA
all year, with mean daily values ranging from 19 to 30 °C. Under these
conditions of high solar radiation and moderate temperature, the cul-
3. Results tures were operated satisfactorily all year, with biomass concentra-
tions from 0.7 to 2.0 g/l (Fig. 2). The biomass productivity varies
3.1. Biomass productivity, mass balance and energy balance along the year according to the solar radiation, ranging from 0.3 to
0.7 g/l·day on a volumetric basis, and from 15 to 45 g/m 2·day on an
The yield of the facility has been determined on an annual basis. areal basis.
For this study, it was operated in a continuous mode at 0.34 1/day To sustain this high biomass productivity it is necessary to supply
dilution rate all year around. Figures from the data acquisition and CO2 and fertilizers. The CO2 gas is injected on-demand to control the
control system, complemented with dry weight determinations of pH, which is an indicator of CO2 depletion. The total amount of CO2
the biomass concentration, were used. The results show that the injected was measured by registering the CO2 injection time and the
solar radiation in Almería (Spain) is high all year, ranging from gas flow when the valves are open (3 l/min). A mean daily value of
10 MJ/m 2·day in winter to 27 MJ/m 2·day in summer (Fig. 2). During 40 kg/day is used (Table 1), which changed seasonally from summer
the spring, cloudy/rainy days were more frequent whereas in the to winter time, and daily accordingly to the solar daylight cycle. The
summer and fall no rain was registered. The temperature is moderate CO2 in the gas outlet the reactor, mixed with the aeration stream
 F.G. Acién et al. / Biotechnology Advances 30 (2012) 1344–1353 1347

Table 1 (Fig. 3). The chemical energy fixed in the biomass has been calculated
Mass balances of the main compounds supplied to the photobioreactors. using a heat combustion of 20 kJ/kg, which was multiplied by the bio-
Inlet, Inlet, Outlet, Outlet, Yield, % Ys/b mass productivity of the plant results in a value of 72 kWh/day total
mM kg/day mM kg/day chemical energy fixed. According to these data, the photosynthetic ef-
CO2 40.00 10.20 74.5% 2.30 ficiency of the cultures was 3.4% (chemical energy fixed to impinging
Carbonate 0.00 0.00 0.00 0.00 solar radiation ratio) whereas the power efficiency of the facility is
Bicarbonate 2.01 1.25 4.80 3.00 − 3.1% 16.0% (energy produced to energy consumed ratio).
Sulphate 3.62 3.56 1.11 1.10 69.2% 0.19
Nitrate 7.68 4.87 0.90 0.57 88.2% 0.33
Potassium 2.28 0.91 1.89 0.76 17.1% 0.01 3.2. Determination of production cost
Calcium 5.28 2.16 4.29 1.76 18.8% 0.03
Magnesium 6.15 1.53 0.84 0.21 86.3% 0.10 The cost of producing microalgae using this semi-industrial scale
Phosphorus 2.02 0.64 0.69 0.22 65.8% 0.03 facility has been determined using as basis a complete flowchart of
the actual process and a detailed description of the equipment, in-
cluding size and type. Flowchart and size of the equipments are deter-
used to remove oxygen, was also measured, resulting in a mean value mined by the scale of the process and its yield, i.e. how much biomass
of 10.2 kg/day being lost. Thus, the efficiency of use of CO2 was 74.5% is required to produce and the productivity of the system. Other oper-
corresponding to a substrate yield coefficient of 2.31 kgCO2/kgbiomass. ation variables, such as working time, nutrients and power consump-
No carbonate or bicarbonate was added to the medium but some bi- tion must also be defined according to the operating conditions of the
carbonate was present in the water used for the medium preparation. system. In this case, the value of the main parameters used for the
The bicarbonate concentration at the inlet was lower than at the out- case study analysed here are shown in Table 3. According to previous
let, indicating that some of the CO2 added for pH control ended up in- data, the mean annual biomass productivity that can be expected
creasing the total inorganic carbon in the culture broth, and this from the system is 0.42 g/l·day at 0.34 1/day dilution rate, for a
inorganic carbon is lost with the culture medium. The data obtained total culture volume of the system of 30 m 3. Using these data, a bio-
showed that a 3.1% of the CO2 injected was lost as bicarbonate with mass production capacity of 3.83 t/year is obtained.
the culture medium outlet the reactor (Table 1). The fertilizers sup- The list of the equipment required is shown in Table 4, along with its
plied to the culture medium were magnesium sulphate, potassium individual cost. The most costly equipment are the photobioreactors,
phosphate and calcium nitrate in addition to micronutrients. The data followed by the freeze-dryer and then by the decanter used to concen-
obtained show that all the fertilizers were supplied in excess, since all trate the harvest. The total cost of the major equipment sums up 348 k€.
of them were found in significant concentrations in the effluent flow The installation cost, including deployment, instrumentation, piping,
(Table 1). Potassium and calcium were supplied in greater excess, and other accessory elements increase the cost as summarized in
resulting in a lower use yield, under 20%, whereas nitrate and magne- Table 5, showing that main entry is the cost of equipment followed by
sium were used more efficiently, with a yield surpassing 85%. The sub- installation and piping cost, service facilities and, finally, the contingen-
strate yield coefficient determined experimentally was higher for cy cost. The total fixed capital required to get the facility setup and ready
nitrate and sulphate, 0.33 kgNO3/kgbiomass and 0.19 kgSO4/kgbiomass, re- to operate is 1.024 k€. Considering a lifetime of 10 years, the annual
spectively, whereas it was very low for potassium, calcium and phos- fixed capital results of 116 k€.
phorous, with values of 0.01 kgK/kgbiomass, 0.03 kgCa/kgbiomass and 0.03 The direct production costs must be also calculated for the analy-
kgP/kgbiomass, respectively. sis. The raw materials, the utilities and labor are the main entries. The
Finally, the facility also consumes power. The power consumption itemized data show that the total raw materials summarizes 3.7 k€
of the major equipment is listed in Table 2. The decanter is the equip- per year, the consumption of CO2 being the main cost. The utilities
ment that consumes more power, but it is only used 6 h/day, resulting sum a value of 13.5 k€ with power consumption as the major contri-
in an average consumption of 33 kWh/day. The highest daily con- bution (Table 6). However, the highest direct production cost is labor
sumptions are related with equipments working 24 h/day as the cen- and others, summing 132.5 k€ per year. Using this information, the
trifugal pumps that move the culture through the photobioreactors, total cost of producing biomass can be calculated from the total
the air blower and the freeze-dryer. The highest daily power con- costs summed to the biomass production ratio, resulting in 69.3 €/kg.
sumption is done by the ten centrifugal pumps used to recirculate
the culture through the tubular photobioreactors, summing up 4. Discussion
240 kWh/day (Table 2). The consumption of the air blower is close
to 96 kWh/day, and the power consumption of the freeze-dryer is 4.1. Overall analysis of the performance of the facility
52 kWh/day, although it does not have to be included in the budget
if wet-biomass is produced instead freeze-dried. The energy fixation To analyze the performance of the facility and to what extent can
yield of the system has also been evaluated. For this, the solar radia- it be improved, an overall analysis of the optimal performance of
tion impinging on the area surface occupied by the reactors has microalgae production systems is carried out. In most situations, the
been calculated, resulting a mean annual value of 2100 kWh/day productivity of microalgae intensive production systems is limited

Table 2
Power consumption of the main equipment used in the facility according to its power rating and time of use.

Equipment Power, W Time operation, h/day Power consumption, kWh/day Units Power consumption, kWh/day

Centrifugal pump photobioreactor 1000 24 24.0 10 240.0

Air blower 4000 24 96.0 1 96.0
Decanter 5500 6 33.0 1 33.0
Culture Medium preparation unit 1100 6 6.6 1 6.6
Ozone unit 1100 6 6.6 1 6.6
Freeze dryer 2200 24 52.8 1 52.8
Control unit 550 24 13.2 1 13.2
Total power consumption, kWh/day 448.2
Total power required, kW 14460.0
1348 F.G. Acién et al. / Biotechnology Advances 30 (2012) 1344–1353

Table 5
Fixed capital estimated for the case study analysed.

Item Detail Factor Cost, €

1 Major purchased equipment 1.00 348,000.00

2 Installation costs 0.20 69,600.00
3 Instrumentation and control 0.15 52,200.00
4 Piping 0.20 69,600.00
5 Electrical 0.10 34,800.00
6 Buildings 0.23 78,300.00
7 Yard improvements 0.12 41,760.00
8 Service facilities 0.20 69,600.00
9 Land 0.06 20,880.00
10 Engineering and supervision 0.30 104,400.00
11 Construction expenses 0.05 39,237.00
12 Contractor's fee 0.03 19,618.50
13 Contingency 0.08 75,839.64
Fig. 3. Energy balance for the production facility located at Estación Experimental Fun-
Total fix capital 1,023,835.14 €
dación Cajamar (EEFC) located in Almería, Spain.
Item Detail Cost, €
1 Lifetime 10
2 Depreciation 98,207.51
Table 3
3 Property tax (at 0.01 depreciation) 0.010 982.08
Main parameters defining the process of the case study analysed.
4 Insurance (at 0.006 depreciation) 0.006 589.25
Case study 5 Purchase tax (at 0.16 of items 1-12/10) 0.160 16,381.36
Total fix capital per annun 116,160.20 €
Ratio V/S 0.07 m3/m2
Dilution rate 0.34 1/day
Biomass productivity 0.42 g/L day
Biomass concentration 1.26 g/L
is necessary to avoid nutrients gradients and favour light exposure
Operation time 300 day/annun of the cells, although it must be reduced to a minimum that does
Operation time 10 h/day not diminish the productivity. Systems with a high volumetric energy
Biomass production capacity 3.83 Tn/year uptake or with a high V/S ratio have a high energy uptake per surface
Air flow rate 0.20 v/v/min
unit, approaching the limit value of 0.9 MJ/m 2·day (Fig. 4). Conse-
Power consumption 15 kWh/m3
Total culture volume 30 m3 quently, the use of photobioreactors with low V/S ratio is recom-
Total culture surface 0.04 Ha mended, being also necessary to reduce the volumetric energy
uptake as much as possible to keep the energy uptake under the ac-
ceptable maximum of self-uptake energy. It is essential, therefore,
by the solar radiation availability inside the reactor, as long as other that the energy consumption in the production process be lower
factors as temperature, pH and nutrients supply are adequately con- than the maximum photosynthetic efficiency. Otherwise the process
trolled. Solar radiation availability is mainly a function of the location would not be a net energy producer. Typical values of self-uptake in
of the reactor. For the facility location (Almería) the daily mean value energy processes range from 10% to 40%, thus accepting a 20% of
measured is 17.58 MJ/m 2·day. Microalgae can conserve a maximum self-uptake the objective of maximum energy consumption for a
of 9–10% of solar energy (photosynthetic efficiency) but microalgal microalgae production process would be 0.18 MJ/m 2·day.
outdoor production systems so-far rarely exceed 6% (Carvalho, et al., On the other hand, the V/S ratio of the reactor also determines the
2006). Thus, considering a mean value of photosynthetic efficiency relationship between the volumetric and superficial biomass produc-
of 5%, a mean annual energy fixation rate, whatever the culture sys- tivity and the total culture volume. According to the 0.9 MJ/m 2·day
tem used, of 0.9 MJ/m 2·day is calculated. This energy from the sun- maximum energy fixation rate for the considered location, and con-
light can be fixed as chemical energy, producing microalgae in sidering a mean combustion heat of the biomass of 20 MJ/kg, it can
different photobioreactors, with designs varying in parameters such be calculated that the mean annual biomass productivity in the
as the V/S ratio or the volumetric power consumption. Power supply

Table 6
Direct production cost for the case study analysed.
Table 4
List and cost of the major equipment for the case study analysed. Item Raw materials Units €/und. Cost, €

1 Fertilizers (kg) 2688 0.40 1,075.07

Item Detail Capacity Cost €/und. No. of Total
2 Carbon dioxide (kg) 8835 0.30 2,650.45
units cost, €
Item Utilities
1 Medium preparation 4 m3/h 6,000.00 1 6,000.00 4 Water (m3) 3070 0.10 307.02
unit (4 m3/h) 5 Power consumption (kWh) 134460 0.10 13,446.00
2 Sterilization process 2 m3/h 15,000.00 1 15,000.00 Item Labor and others Units €/und. Cost, €
(2 m3/h) 6 Labor 3.00 30,000.00 90,000.00
3 Air blower 200 m3/h 2,500.00 2 5,000.00 7 Supervision (at 0.2 labor) 0.20 18,000.00 3,600.00
(200 m3/h) 8 Payroll charges (at 0.25 0.25 23,400.00 5,850.00
4 Photobioreactors 3 m3 15,000.00 11 165,000.00 (labor + supervision))
(3 m3) 9 Maintenance (at 0.04 MEC) 0.04 13,920.00 556.80
5 Harvest storage tank 1 m3 500.00 1 500.00 10 Operating supplies 0.00 2,296.80 9.19
(1 m3) (at 0.004 items 1–5)
6 Decanter (4 m3/h) 4 m3/h 45,000.00 1 45,000.00 11 General plant overheads (at 0.55 0.55 51,786.24 28,482.43
7 Harvest pump 2 m3/h 1,000.00 1 1,000.00 (labor + supervision + maintenance))
(2 m3/h) 12 Tax (at 0.16 items 1–7, 11 and 12) 0.16 90.56 14.49
8 CO2 supply unit 4 Kg/h 500.00 1 500.00 13 Contingency (at 0.05 items 1–7) 0.05 28,710.00 1,435.50
(4 kg/h) 14 Marketing (at 0.05 items 1–13) 0.05 51,191.76 2,559.59
9 Freeze-dryer and 70 Kg/day 110,000.00 1 110,000.00 Total raw materials 3,725.52 €
storage (70 kg/day) Total utilities 13,753.02 €
Total (€) 348,000.00 Total labor and others 132,508.00 €
 F.G. Acién et al. / Biotechnology Advances 30 (2012) 1344–1353
Fig. 4. Influence of volumetric energy uptake and V/S ratio of the reactor in the surface
energy uptake, and comparison with limit values determined by maximum solar effi-
ciency and maximum self uptake energy consumption into the process. According to
this figure the volumetric energy uptake is limited to a different value to be energeti-
cally positive, as a function of V/S ratio of the photobioreactor.
location of Almería approaches to 44 g/m 2·day. Dividing this value by
the V/S ratio, the maximum volumetric biomass productivity can be
calculated. Fig. 4 shows that the volumetric biomass productivity in-
creases as the V/S ratio of the reactor decreases. For reactors with a
large V/S ratio, such as raceways (V/S > =200 l/m 2), the maximum
biomass productivity achievable is lower than 0.22 g/l·day. For
reactors with low V/S ratio, as tubular photobioreactors (V/S b =
100 l/m 2), the maximum biomass productivity can be 0.9 g/l·day or
higher. These values correspond to the maxima that can be expected
according to the solar energy availability and considering an optimal
yield of the cultures. To operate close to the optimal conditions the
reactors must work in continuous mode all year round and at the di-
lution rate that maximises the productivity. The optimal dilution rate
for Scenedesmus almeriensis has been found to be between 0.3 and 0.4
1/day (Sánchez et al., 2008a, 2008b). Considering the dilution rate
used in the production of this strain at large scale, of 0.34 1/day, the
biomass concentration into the culture can be calculated as a function
of V/S ratio of the reactor. The data show that the lower the V/S ratio
of the reactor the higher is the biomass concentration that can be
achieved. This variation is similar to that previously observed for
the biomass productivity (Fig. 5).
The volume of reactor and harvest required to produce a unit mass
of biomass can also be calculated (Fig. 5). The results demonstrate
that the higher the V/S of the reactor is the higher is the volume
of culture required to produce a unit mass of biomass. In raceways
(V/S >=200 l/m2), values over 4.5 m3/kg should be expected whereas
in tubular photobioreactors (V/S b =100 l/m2) the volume needed is
less than 2.2 m3/kg. Additionally, the harvest volume to be managed
to produce the same amount of biomass is much larger in photobior-
eactors with a large V/S ratio (1.5 m 3/kg) than in the ones with a low
V/S ratio (0.77 m 3/kg). The consequence of this is that although race-
way reactors are cheaper than tubular reactors, the low yields of reac-
tors with high V/S ratio imply that larger volumes of reactors are
needed to produce the same amount of biomass, and in consequence
the cost of these reactor can be higher than for low V/S reactors
(Norsker, et al., 2011). Moreover, the higher the V/S of the reactor
the higher the volume of harvest to be managed, thus also needing
larger and more expensive equipment to process the harvest than
when using low V/S ratio reactors. The selection of an optimal V/S
ratio or design of the reactor can be only carried out from an econom-
ical analysis that includes the productivity and the direct operation
costs at each V/S ratio.
The experimental data obtained during the operation of the semi-
industrial facility here reported confirm the maximum values that can
1349
Fig. 5. Influence of V/S ratio of the reactor in (a) the biomass concentration of steady
state, and (b) total and harvested volume of the reactor, for the continuous operation
at dilution rate of 0.34 1/day. Numbers printed for most frequent V/S ratio of photo-
bioreactors reported.
be reached. Considering that the V/S ratio of the reactor is 0.07 m 3/m2,
maximum values of biomass concentration and productivity of 1.82 g/l
and 0.62 g/l·day, respectively, can be obtained. The experimental re-
sults obtained were 1.26 g/l and 0.42 g/l·day for biomass concentration
and productivity respectively, approaching the maximum values by
70%. Regarding volume of the reactor and harvest, the theoretical min-
imum values in these conditions are of 1.62 and 0.55 m 3/kg respective-
ly, whereas the experimentally determined values were 2.36 and
0.80 m 3/kg for reactor volume and harvest volume respectively. These
data are 44% higher than the theoretical estimations, indicating that
the yield of the system can be yet improved. Thus, the photosynthetic
efficiency of the culture was 3.64%, lower than maximum value of 5%,
but it is high considering the size of the facility and the operation all
year around.
Regarding the mass and energy balances, the data here reported
demonstrated that the main nutrient is CO2 followed by nitrate. The
theoretical substrate yield for CO2 is 1.8 kgCO2/kgbiomass, not too differ-
ent from the yield obtained experimentally, 2.3 kgCO2/kgbiomass, indi-
cating that the efficiency of use is high and therefore that the
management of CO2 injection is adequate. Considering that the nitro-
gen content of the biomass is close to 5%, the nitrate theoretical sub-
strate yield is 0.24 kgNO3/kgbiomass. The experimental yield for nitrate
was higher, 0.33 kgNO3/kgbiomass, revealing that nitrate is partially lost
or nitrogen content of the biomass is slightly higher than 5%. Not
enough data of elemental analysis of the biomass all year around
are available corroborate it. According to the experimental data
reported here, for the production of 1 kg of biomass it is necessary
to supply 2.3 kgCO2, 0.33 kgNO3 and 0.03 kgP. These values are highly
1350 F.G. Acién et al. / Biotechnology Advances 30 (2012) 1344–1353

relevant because at a scale as the required for the biofuels market the
supply of these compounds can be a serious limitation. The consump-
tion of petrodiesel in Spain in 2009 was 31100·kt/year. Replacing a
10% of this petrodiesel with biodiesel from microalgae would request
at least 3100·kt/year of biodiesel, equivalent to 17300 kt/year of oil-
rich microalgae biomass (30% oil content). Producing this amount of
biomass would request 40000 ktCO2/year (equivalent to CO2 emis-
sions of 5.5 GWe in carbon power stations), 5800 ktNO3/year (four
times the nitrate produced in Spain), and 600 ktPO4/year (ten times
the phosphate produced in Spain). To reduce these requirements
the composition of the culture medium must be carefully adjusted
to avoid the waste of nutrients with the effluents. The use of the nu-
trients contained in waste water from other uses would obviously be
very advantageous.
Regarding energy, the power consumption in this facility is
3.75 MJ/m 2·day, much higher than the stated limit value of self-
Fig. 6. Major production costs for the production of dry biomass of Scenedesmus almeriensis
uptake energy consumption that allows the operation to be energet-
in the facility considered. Data from Tables 5 and 6.
ically positive, 0.2 MJ/m 2·day. This is a handicap derived from the
fact that this facility was not designed to produce energy but high
value biomass. In this facility, the culture is recirculated through the
400 m long tubular loop of each photobioreactor by using a low stress
centrifugal pump equipped with turbine of 5 mm that can also recir-
culate particles that help clean the inner walls of the photobioreactor,
prevent fouling and allow long-term operation. The liquid velocity is
therefore high, 0.9 m/s, which is also needed to avoid dissolved oxy-
gen accumulation in the loop and to increase the productivity by im-
proving the light–dark exposure of the cells inside the dense culture
in the tube. Therefore, a centrifugal pump of 1100 W is used in each
photobioreactor. However, the theoretical minimum power required
to recirculate the liquid through the solar loop in the described condi-
tions is much lower, 300 W, which indicates a low yield of the centrif-
ugal pumps used, 27%. The power consumption of the impulsion
system can be reduced with the use of more efficient pumps, reducing
head losses in the loop, diminishing the liquid velocity and/or avoid-
ing accessories. Concerning the air blower, the power consumption of
this equipment is close to 53 kWh/day, and although it cannot be
eliminated, it can be used to reduce the power required from centrif-
ugal pump by operating in airlift mode. However, to operate in airlift
mode the photobioreactor must be redesigned because high of riser
and downcomer must be at least 4 m higher than high of the loop.
In any case, using an optimal aeration rate to remove the oxygen
while minimizing the power consumption is also indispensable. The
actual power consumption of the blower is 0.86 MJ/m 2·day while
the maximum compatible with the energetic efficiency goal is
0.20 MJ/m 2·day. The improvement of mass transfer into the bubble
column by using optimal diffusers can reduce the airflow rate re-
quired, and thus the power consumption of blower.

4.2. Microalgae production cost of case study

Regarding the microalgae production cost determined for the fa-

cility designed and operated at Estación Experimental Fundación
Cajamar (EEFC) in Almería (Spain), the obtained data of 69.3 €/kg is
high when considering the energy market but not so much when
comparing with high value biomass for human or nutraceutical uses
(Pulz and Gross, 2004). To reduce this production cost an economic
and technical evaluation needs to be performed, in order to identify
major factors determining the technical and economical yield. In
this sense, an important reduction in the investment requested to
build a microalgae production facility can be attained if standard ma-
terials and equipment can be used, avoiding special or custom made
designs or expensive materials. The same applies to the scale of
equipment. This is important because the selected size and type of
technology limit the scale-up possibilities, especially in the centrifu- Fig. 7. Distribution of major components of (a) equipment cost, (b) raw materials and
gation and freeze-dry steps. The facility shown in this report was utilities costs, and (c) labor costs, for the production of dry biomass of Scenedesmus
built according to these rules, and thus most of the materials and almeriensis in the facility considered. Data from Tables 5 and 6.
 F.G. Acién et al. / Biotechnology Advances 30 (2012) 1344–1353 1351

equipment were standard items provided by food and greenhouses
technology companies. The cost analysis of the facility finally built
shows that labour and depreciation are the major costs, summing
51.6% and 42.6% of the total production cost respectively, whereas
costs related with utilities and raw materials were much lower, total-
ling 3.2% and 2.7% respectively (Fig. 6). The analysis of each item in
these major costs shows that the photobioreactors represent a cost
close to the 48% the total facility, followed in by the freeze-dryer
and the decanter (Fig. 7). Regarding the raw materials and the utili-
ties, the power consumption is the major factor contributing to the
production cost, more than 78% (Fig. 7). The last cost entry corre-
sponds to labour and management. As Fig. 8 shows, direct labour is
the major contributing factor, accounting for as much as a 68%. The
data here reported have been obtained in real conditions. The re-
ferred 30 m 3 plant has been in operation for 2 years producing Scene-
desmus almeriensis biomass at the productivities and costs reported
here. In addition, operation of the reactors allowed identifying the as-
pects that could be improved in design and disposition in order to in-
crease the yield and reduce the production cost.
redesign of the photobioreactor, but it has still been included to ac-
complish the energy balance of the system.
According to these considerations, the production cost at large
scale has been calculated with the methodology described. The re-
sults obtained show how at the larger scale assumed, the total pro-
duction cost decreases to 12.6 €/kg. Now, depreciation is the major
cost, summing up 78% of the total production expenses (Fig. 8). The
other costs are much lower by comparison, representing from 7.0%
to 7.8% of total production cost. According to the detailed cost analy-
sis, the photobioreactor is the main contributing factor representing
94% of total cost of major equipment (Fig. 9). The main contributions
to the direct production cost are power and carbon dioxide (44.4 and
36.5% of raw materials and utilities) followed by labor (38% of labor

4.3. How to reduce the production costs

The main hint from the data presented is that to reduce the pro-
duction costs it is necessary to reduce the labor requested, to one
men/ha or less, by implementing extensive automatization of as
many of the operations taking place in the plant as possible. A reduc-
tion of the cost of the major equipment must also be performed, by a
simplification of design based in the extensive knowledge accumulat-
ed. The equipment cost can also be reduced when the size of the
equipment, and hence the production capacity, is increased. A new
scenario introducing several changes according to the propositions
above is considered next. Sterilization of the culture medium is now
done only by filtration with 1 μm membranes, and the freeze-dryer
is now omitted because the product considered is wet biomass
(15%d.wt.). The centrifugation costs are also reduced by introducing
previously a new flocculation–decantation step. A new size of 200 t/
year for the plant has also been assumed. The proposed scale-up of
global capacity has been refereed to use the biomass production and
not to the overall surface because costs in this analysis are expressed
per unit mass or size of equipment so that the analysis remains inde-
pendent of the photobioreactor technology used. Assuming that the
relationship V/S and productivity of photobioreactors remain the
same for this new size, achieving this production capacity requires
1570 m 3 of installed photobioreactors occupying a surface of 2.24
Ha. Finally, the power consumption must be reduced at least to 3.6
kWh/m3·day, which is the energy fixed chemically in the biomass.
Achieving this power consumption would request a profound

Fig. 9. Distribution of major components of (a) equipment cost, (b) raw materials and util-
Fig. 8. Major production costs for the production of dry biomass of Scenedesmus almeriensis ities costs, (c) and labor costs, for the production of dry biomass of Scenedesmus almeriensis
at large scale but using the same technology that used in the facility considered. at large scale but using the same technology that used in the facility considered.
1352 F.G. Acién et al. / Biotechnology Advances 30 (2012) 1344–1353
and management). The production cost found in this scenario is close
to the value of 4.15 €/kg calculated for closed tubular photobioreac-
tors at a large scale of 100 ha (Norsker, et al., 2011), which is the low-
est compared to open reactors or flat panels due to their lower
productivity and their higher biomass harvesting costs (Posten,
2009). At this scale, the unit production costs reported for open race-
ways and flat panels were 4.95 €/kg and 5.96 €/kg, respectively
(Norsker, et al., 2011). These data revealed how simplifying the pro-
cess and increasing the scale allows reducing the production cost
four-fold. In spite of this reduction, the production cost remains
close to 10 €/kg which is still high and needs to be reduced by, at
least, one order of magnitude to meet the expectations of the energy
and commodities markets.
4.4. Minimum microalgae production cost
The economic analysis described allows determining the production
cost of a facility and assessing the influence of each factor in the final
production cost, as well as the minimum production cost achievable.
The data obtained for large scale showed that the cost of photobioreac-
tors is the major factor in the production cost. If that could be reduced
while maintaining the productivity, the production cost would decrease
which is analysed next. The data calculated with this assumption show
that the reduction of the photobioreactor cost dramatically decreases
the biomass unit production cost (Fig. 10), but it must be highlighted
that even for photobioreactor costs lower than those currently refer-
enced for open raceways (0.1 €/l) (Norsker, et al., 2011), the calculated
production cost still results higher than 3 €/kg. The next factor in rele-
vance is the consumption of raw materials. Since, as has been previously
shown, the carbon dioxide is the most expensive consumable (Fig. 10),
its influence is analysed next. Using flue gases from industrial sources
can reduce the cost of CO2 to values as low as zero if flue gases are read-
ily available. This way, the biomass production cost could be decreased
to 2.5 €/kg (Fig. 10). To further reduce the biomass production cost
Fig. 10. Influence of major factors in the production cost of Scenedesmus almeriensis at large scale.
below this value, the best option is to reduce the personnel necessary
for the operation of the plant. The data obtained from the analysis
showed that one person per 10 hectares is required to achieve a bio-
mass production cost of 2.1 €/kg. Finally, the uptake of water, and espe-
cially of fertilizers, also contributes significantly to the production cost.
The utilization of wastewater containing mineral nutrients (from other
industries, agricultural or urban uses) is highly recommended for other
reasons and could reduce the production cost to 1.8 €/kg, although it is
necessary to define the pre-treatment required to be able to use this
type of waters. Including in these conditions the obtained production
cost is still higher than what would be acceptable for the biofuels mar-
ket. Reducing it further requests enhancing the yield of the system
approaching to the maximum value of 5% of photosynthetic yield.
This analysis is in agreement with previous data (Norsker, et al.,
2011) that indicates that selecting an optimized location to improve
the productivity, improving the photosynthetic efficiency of microalgae
by adequate culture conditions and their control, and using free carbon
dioxide and medium are the key actions to reduce the production costs.
Using tubular or flat panel photobioreactors, the unit production cost
can be reduced to 0.70 and 0.68 €/kg, respectively, whereas for open
raceways the cost cannot be reduced below 1.28 €/kg (Norsker, et
al., 2011). Thus, the bottleneck for the production of microalgae for
the production of energy or commodities is to develop more produc-
tive photobioreactor systems while reducing their cost dramatically.
It has been reported that to guarantee an economic design for pro-
duction of energy products the investment costs cannot exceed 40
€/m 2 (Hankamer, et al., 2007). The data here reported at small
scale (0.04 ha) are much higher than this value, up 2400 €/m 2,
though they can be reduced to 750 €/m 2 at large scale even assuming
the same photobioreactor cost of 5 €/l. According to the last calcula-
tions, if the photobioreactor cost could be reduced to 0.1 €/l the in-
vestment costs drops to 49 €/m 2, substantially closer to the
referenced limit value (Hankamer, et al., 2007). However, the mass
and energy balances must also be taken into account. The data here
 F.G. Acién et al. / Biotechnology Advances 30 (2012) 1344–1353
reported indicates that the nutrients can limit the development of
energy production processes based on microalgae unless waste
from other sources are used. On the other hand, the power consump-
tion of the actual production systems must be sharply reduced to be-
come energetically positive.
5. Conclusions
It can be concluded that although it has been reported that micro-
algal fuel production is relatively close to being economically feasible
(Stephens, et al., 2010) the analysis shown here and the data reported
demonstrate that achieving economic feasibility using the current
production technologies requests a substantial reduction of the cur-
rent costs and to operate them near the optimal photosynthetic
yield. Closed photobioreactors are considered the most productive
systems but to be competitive their cost must be reduced below the
cost of the current open raceways reactors. In addition, large facilities
capable of producing more than 150 t/Ha·year must be operated with
low labor costs, using flue gases as carbon source and wastewater as
growth medium in the largest possible extent.
Acknowledgements
This research was supported by “Algae and aquatic biomass for a sus-
tainable production of 2nd generation biofuels” Project (FP7-ENERGY-
2009-1), Ministerio de Ciencia e Innovación (CTQ2008-06741-CO2-01/
PPQ) and PlanE Spanish programme on phytoplacton production for bio-
fuel and carbon dioxide capture, Junta de Andalucía Plan Andaluz de
Investigación (CVI 173&131), ACCIONA Biocombustibles and Fundación
CAJAMAR.
References
Acién Fernández FG, Fernández Sevilla JM, Sánchez Pérez JA, Molina Grima E, Chisti Y.
Airlift-driven external-loop tubular photobioreactors for outdoor production of
microalgae: assessment of design and performance. Chem Eng Sci 2001;56:
2721–32.
Alías CB, García-Malea López MC, Acién Fernández FG, Fernández Sevilla JM, García
Sánchez JL, Molina Grima E. Influence of power supply in the feasibility of
Phaeodactylum tricornutum cultures. Biotechnol Bioeng 2004;87:723–33.
Borowitzka MA. Commercial production of microalgae: ponds, tanks, tubes and fer-
menters. J Biotechnol 1999;70:313–21.
1353
Carvalho AP, Meireles LA, Malcata FX. Microalgal reactors: a review of enclosed system
designs and performances. Biotechnol Prog 2006;22:1490–506.
Chisti Y. Biodiesel from microalgae. Biotechnol Adv 2007;25:294–306.
Chisti Y. Biodiesel from microalgae beats bioethanol. Trends Biotechnol 2008;26:
126–31.
Douskova I, Doucha J, Livansky K, MacHat J, Novak P, Umysova D, et al. Simultaneous
flue gas bioremediation and reduction of microalgal biomass production costs.
Appl Microbiol Biotechnol 2009;82:179–85.
Gouveia L, Oliveira AC. Microalgae as a raw material for biofuels production. J Ind
Microbiol Biotechnol 2009;36:269–74.
Hankamer B, Lehr F, Rupprecht J, Mussgnug JH, Posten C, Kruse O. Photosynthetic bio-
mass and H2 production by green algae: from bioengineering to bioreactor
scale-up. Physiol Plant 2007;131:10–21.
Kalk J, Langlykke A. Cost estimation for biotechnology projects. In: American Society of
Microbiology, editor. Manual of Industrial Microbiology and Biotechnology; 1986.
p. 363–85. Washington, DC.
Mann JE, Myers J. On pigments, growth, and photosynthesis of Phaeodactylum tricornutum.
J Phycol 1968;4:349–55.
Molina E, Fernández J, Acién FG, Chisti Y. Tubular photobioreactor design for algal cul-
tures. J Biotechnol 2001;92:113–31.
Norsker NH, Barbosa MJ, Vermuë MH, Wijffels RH. Microalgal production—a close look
at the economics. Biotechnol Adv 2011;29:24–7.
Patil V, Tran KQ, Giselrød HR. Towards sustainable production of biofuels from micro-
algae. Int J Mol Sci 2008;9:1188–95.
Posten C. Design principles of photo-bioreactors for cultivation of microalgae. Eng Life
Sci 2009;9:165–77.
Pulz O, Gross W. Valuable products from biotechnology of microalgae. Appl Microbiol
Biotechnol 2004;65:635–48.
Richardson JW, Outlaw JL, Allison M. The economics of microalgae oil. AgBioForum
2010;13:119–30.
Richmond A. Microalgal biotechnology at the turn of the millennium: a personal view.
J Appl Phycol 2000;12:441–51.
Sánchez JF, Fernández JM, Acién FG, Rueda A, Pérez -Parra J, Molina E. Influence of culture
conditions on the productivity and lutein content of the new strain Scenedesmus
almeriensis. Process Biochem 2008a;43:398–405.
Sánchez JF, Fernández-Sevilla JM, Acién FG, Cerón MC, Pérez -Parra J, Molina-Grima E.
Biomass and lutein productivity of Scenedesmus almeriensis: influence of irradi-
ance, dilution rate and temperature. Appl Microbiol Biotechnol 2008b;79:719–29.
Singh J, Gu S. Commercialization potential of microalgae for biofuels production.
Renew Sustain Energy Rev 2010;14:2596–610.
Stephens E, Ross IL, King Z, Mussgnug JH, Kruse O, Posten C, et al. An economic and
technical evaluation of microalgal biofuels. Nat Biotechnol 2010;28:126–8.
Wigmosta MS, Coleman AM, Skaggs RJ, Huesemann MH, Lane LJ. National microalgae
biofuel production potential and resource demand. Water Resour Res 2011;47:
W00H04.
Wijffels RH, Barbosa MJ. An outlook on microalgal biofuels. Science 2010;329:796–9.
Wijffels RH, Barbosa MJ, Eppink MHM. Microalgae for the production of bulk chemicals
and biofuels. Biofuels Bioprod Biorefin 2010;4:287–95.
Williams PJLB, Laurens LML. Microalgae as biodiesel & biomass feedstocks: review &
analysis of the biochemistry, energetics & economics. Energy Environ Sci 2010;3:
554–90.