1
This is the accepted manuscript version of the journal article, which is made
available for scholarly purposes only, in accordance with the journal's author
permissions. The final publication is available at http://link.springer.com.
The full citation is:
Milledge JJ and Heaven S (2013) A review of the harvesting of micro-algae
for biofuel production. Reviews in Environmental Science and Biotechnology
12, 165-78. http://dx.doi.org/10.1007/s11157-012-9301-z
2
A Review of the Harvesting of Micro-algae for Biofuel Production
John J Milledge* and Sonia Heaven
Engineering and the Environment, University of Southampton, *email
j.milledge@soton.ac.uk
Acknowledgement -This work was partially supported by the EU FP7 ALL-GAS
project (268208).
Introduction
Although algal biomass can be ‘energy rich’, the growth of algae in dilute
suspension at around 0.02% - 0.05% dry solids (Zamalloa et al. 2011) poses
considerable challenges in achieving a viable energy balance in algal process
operations. Additional challenges of algae harvesting come from the small size of
micro-algal cells (most algae are below 30µm) (Molina Grima et al. 2003); the
similarity of density of the algal cells to the growth medium (Reynolds 1984); the
negative surface charge on the algae that results in dispersed stable algal
suspensions, especially during the growth phase (Edzwald 1993; Moraine et al.
1979; Packer 2009); and the algal growth rates which require frequent harvesting
compared to terrestrial plants .
The cost effective harvesting of micro-algae is considered to be the most
problematic area of algal biofuel production (Greenwell et al. 2010) and a key
factor limiting the commercial use of micro-algae (Olguín 2003). It has been
suggested that 20 to 30% of the costs of micro-algal biomass is due to the costs of
harvesting (Mata et al. 2010; Molina Grima et al. 2003; Verma et al. 2010), but
estimates as high as 50% of micro-algal biomass cost have been given (Greenwell
et al. 2010). It has been estimated that 90% of the equipment cost for algal
biomass production in open systems may come from harvesting and dewatering
(Amer et al. 2011). The need for continuous harvesting of the dilute suspension
makes the harvesting of micro-algae 'inherently more expensive' than harvesting
land plants (Benemann et al. 1977), and the separation of micro-algae by
settlement and centrifugation can have a harvesting energy requirement of 1 MJ
kg
-1
of dry biomass, greater than the energy cost of harvesting wood at 0.7 – 0.9
MJ kg
-1
(Sawayama et al. 1999). The cost of harvesting micro-algae needs to be
3
reduced. Unfortunately a recent report by UK's Biotechnology and Biological
Sciences research council (BBSRC) on algal research has concluded that: “hardly
any commercial activity exists in downstream processing”(Schlarb-Ridley 2011).
Most work on micro-algal species selection for biofuel production has been
focused on yield and composition rather than on ease of recovery (Brennan and
Owende 2010).
Table 1 Comparison of microalgal harvesting methods (Mohn 1988; Molina
Grima et al. 2003; Shen et al. 2009)
Advantages Disadvantages
Dry Solids
Output
Conc'
Centrifugation
Can handle most algal types with rapid efficient cell
harvesting.
High capital and operational costs. 10-22%
Filtration
Wide variety of filter and membrane types available.
Highly dependent on algal species; best suited to
large algal cells. Clogging or fouling an issue.
2-27%
Ultrafiltration
Can handle delicate cells. High capital and operational costs 1.5-4%
Sedimentation
Low cost, potential for use as a first stage to reduce
energy input and cost of subsequent stages.
Algal species specific, best suited to dense non-
motile cells. Separation can be slow. Low final
concentration.
0.5 -3%
Chemical Flocculation
Wide range of flocculants available, price varies
although can be low cost.
Removal of flocculants, chemical contamination. 3-8%
Flotation
Can be more rapid than sedimentation. Possibility to
combine with gaseous transfer.
Algal species specific. High capital and operational
cost.
7%
Algae can be harvested by a number of methods; Sedimentation, Flocculation,
Flotation, Centrifugation and Filtration or a combination of any of these. Despite
the importance of harvesting to the economic and energy balance viability of
micro-algal biofuel, however, there is no universal harvesting method for micro-
algae (Mata et al. 2010; Shen et al. 2009). A recent extensive review of
dewatering micro-algal cultures concluded that “currently there is no superior
method of harvesting and dewatering” (Uduman et al. 2010). A summary of
advantages and disadvantages of the various methods to harvest micro-algae is
given in Table 1(Milledge and Heaven 2011).
The final moisture content of the harvested algal biomass is an important criterion
in the selection of the harvesting method (Molina Grima et al. 2003). Micro-algal
biomass can spoil in hours if the moisture content is greater than 85 % (Mata et al.
2010) and high moisture content can have a substantial influence on the costs and
methods of further processing (Molina Grima et al. 2003) and energy extraction
from the biomass.
Sedimentation
4
In sedimentation gravitational forces cause liquid or solid particles to separate
from a liquid of different density, but the process can be extremely slow
especially if density difference or particle size is small. Sedimentation can be
described by Stokes’ Law which assumes that sedimentation velocity is
proportional to the square of the (Stokes’) radius of the cells and the difference in
density between the micro-algal cells and the medium as shown below:
(Equation 1)
where r is cell radius, η is fluid dynamic viscosity and ρ
s
and ρ
l
are the solid and
liquid densities.
The density of micro-algae is close to that of water and of salt water at 1024.7 kg
m
-3
(Millero and Lepple 1973) and therefore there is little density difference
driving micro-algal settlement. The cytoplasm of marine micro-algae has a
density between 1030 and 1100 kg m
-3
(Smayda 1970), the density of
cyanobacteria is between 1082 and 1104 kg m
-3
(Kromkamp and Walsby 1990),
density of marine diatom and dinoflagellates between 1030 and 1230 kg m
-3
and
the density of the freshwater green microalgae (Chlorococcum) between 1040 and
1140 kgm
-3
(Van Lerland and Peperzak 1984).
A settlement velocity of 0.1 m day
-1
can be calculated using Stokes’ Law
(equation 1) for a common spherical shaped micro-algae, Chlorella (density 1070
kg m
-3
and average cell diameter 5 µm (Edzwald 1993)), in freshwater (density at
20°C 998 kg m
-3
and viscosity 1 x 10
-3
Pa s
-1
(Weast 1985)). An experimental
study found a considerably higher settling rate for Chlorella at 3.6 m day
-1
(Collet
et al. 2011), but Chlorella does not normally settle readily (Nurdogan and Oswald
1996). The calculated settlement velocity of Cyclotella, a similar sized alga to
Chlorella, is 0.04 m day
-1
, but the observed settlement rate was higher at 0.16 m
day
-1
(Smayda 1970). The observed sinking rates of micro-algae have been found
to deviate from calculated rates, being up to several times higher or lower than the
calculated rate (Reynolds 1984; Smayda 1970). The settling velocity is very
dependent upon the type of micro-algae present, but average settling velocity of
0.2 m day
-1
for diatoms, 0.1 m day
-1
for green micro-algae and 0.0-0.05 m day
-1
for cyanobacteria have been suggested for water quality models (Cole and Wells
1995).
5
Stokes’ law holds for spheroid shapes, but micro-algae are most often not
spherical (Peperzak et al. 2003). Micro-algae can have a diverse range of shapes, a
fact that is often suggested as an evolutionary development to prevent settling
from the euphotic zone (Smayda 1970; Sournia 1978). The sinking rate of 24
autotrophic micro-algae ranging in size from under 10 to 1000 µm was found to
be between -0.4 to over 2.2 m day
-1
with an average of 0.6 m day
-1
, but no
straightforward correlation was found between size and sinking rate and no
relationship was found between cell size and sinking rates for diatoms (Peperzak
et al. 2003). In a study of 20 micro-algae only four always settled readily,
although 14 settled out occasionally (Peperzak et al. 2003). In a study of 30
species of micro-algae found in wastewater most were found reluctant to settle,
with needle like or long cylindrical micro-algae being particularly resistant to
settling (Choi et al. 2006). Filamentous algae (Spirulina) and colonial algae
(Micractinim, Scenesdesmus) with a cluster diameter of ~60 µm have been shown
to be harvestable by settlement, but smaller algae (Chlorella) and motile micro-
algae (Euglena, Chlorognium) do not readily settle out of suspension (Nurdogan
and Oswald 1996). Dinoflagellates have been found to be able to swim at speeds
of up to 0.03 m min
-1
(Smayda 1970) and many species of micro-algae have been
shown to move upwards towards light (Kromkamp and Walsby 1990; Smayda
1970; Sournia 1978).
The settlement of micro-algae varies between species, but can also alter within the
same species. Settlements rates have been shown to vary with light intensity
(Waite et al. 1992), nutrient deficiency has been shown to decrease settlement rate
(Bienfang 1981) and sinking rate increases in older cells especially in senescent
cells (non-dividing cells between maturity and death) (Smayda 1970) and spore-
producing cells (Bienfang 1981). The average density of carbohydrate is 1500 kg
m
-3
, protein 1300 kg m
-3
and lipid 860 kg m
-3
(Reynolds 1984), and micro-algae
with a high lipid content are likely to settle less readily due to the lower density.
Sedimentation has not been widely used for separation of micro-algae (Uduman et
al. 2010) and although settling has been demonstrated in pilot-scale wastewater
treatment systems (Lundquist et al. 2010), it has not yet been achieved on a large