Gravity Assisted Ultrasound Cell Concentrator Patrick Gallant 1
Gravity Assisted Ultrasound
Cell Concentrator
Patrick Gallant
1
TABLE OF CONTENTS
Page
Chapter I Introduction 3
Chapter II Background
2.1 Physical Properties of Microalgae 5
2.2 Methods for Microalgae Harvesting 5
2.3 Acoustic Cell Separation Methods 7
2.3.1 Ultrasonic Laminar Flow Filtration 7
2.3.2 Ultrasonic Cross-Flow Filtration 9
2.3.3 Ultrasonic Mesh Filtration 10
2.3.4 Ultrasonic Sedimentation 11
2.3.5 Zhaowei Wang’s Ultrasonic-assisted Filter 13
2.4 Ultrasound Effects on Algae 13
Chapter III Theory and Methods
3.1 Theory 15
3.2 Materials 17
3.3 Setup and Operation 18
Chapter IV Results 19
Chapter V Discussion 20
Schematics 21
Works Cited 28
2
Chapter I
INTRODUCTION
With the huge growth in the biomedical and biotechnology fields today there is an ever
increasing need for large batches of cell cultures for research and industrial purposes. These
large batches of cells are used for everything from tissue generation to biodiesel. The one thing
that these varied applications have in common is that they need to be concentrated and separated
from the media in which they are grown. There are multiple effective separation techniques on a
small scale but on a large scale many of them either lose effectiveness or are not cost effective
due to their huge energy consumption.
Sedimentation and centrifugation are some of the original separation techniques used yet
sedimentation is often very time consuming while centrifugation is ineffective at separating cells
with densities similar to their growth medium and can damage the cells. These methods gave
way to others such as laminar flow filtration, cross flow filtration and membrane filtration.
These methods have shown success with separating cells in different ways. Membrane filtration
can be very effective at trapping cells in a membrane with a cell suspension flowing through it
though the membrane has to be emptied periodically. Cross flow filtration is able to easily
separate smaller cells or particles from a heterogeneous cell or particle solution but it can
become clogged so it has to be back-flushed or replaced periodically. Laminar flow filtration
3
can effectively increase cell concentration but can have problems when separating different types
of cells from each other.
All of these techniques of sedimentation and filtration can be improved through a variety
of methods such as mechanical vibration and electric fields. In this work, cell sedimentation will
be enhanced by means of ultrasound standing waves. This research builds on the work of my
predecessor Zhaowei Wang [1], in which he had designed and created a system using an
ultrasound assisted inclined gravity settler. His research on this topic focused on finding the
optimal flow rate for maximum retention of polystyrene particles and HB-159 hybridoma cells.
In this thesis, this device has been redesigned and used to harvest the algae strain Scenedesmus
dimorphous by using ultrasound to concentrate algae cells suspended in an aqueous in a growth
media.
An algae solution will pass though the gravity settler once and will exit the settler in two
streams. The basic equipment used will be the gravity settler, a transducer, a frequency
generator, a small bioreactor and an amplifier. The concentration of the algae in each of the two
outlet streams and the inlet stream will be measured using a spectrophotometer.
4
CHAPTER II
BACKGROUND
2.1 Physical Properties of Microalgae
Scenedesmus dimorphus are classified as microalgae in that they grow as a suspension of
individual or small groups of cells on a microscopic scale in water, as opposed to large
multicellular photosynthetic protist colonies such as seaweed. This makes Scenedesmus
dimorphus more difficult to isolate from water then the photosynthetic protists or macroalgae,
yet scenedesmus dimorphus is used for biofuel applications due to its high lipid content which
ranges from 16-40% by weight [8]. This species of algae tends to be found in singles, doubles,
triplets and quads and has a size of roughly 10µm and has a sticky consistency which causes it to
clump and form thick sediments if it is not under agitation. This property can cause
complications in some separation methods yet it can also be a boon for separation methods such
as sedimentation [8].
2.2 Methods for microalgae harvesting
There are many well established methods of microalgae harvesting and separation
available such as: filtration, centrifugation, flocculation and flotation. Filtration is by far the
simplest of the four methods and has a high rate of recovery. However, filtration methods are
known to easily clog and have high maintenance costs. The method only tends to be cost
effective when used for large colonial microalgae [9].
5
Centrifugation is a process that is known for having a high recovery rate and to be
extremely fast. This process works due to Stroke’s law which states that the denser and larger
the particle, the faster it will move through the media. By applying centrifugal force to a cell
suspension the different materials in the suspension will forms layers based on density and size
so the layer of interest (algae layer) can be removed and used. Centrifugation is a highly
efficient system and is great for small scale operations but it has a very high energy use (1 kW
m3) so it is not used as the primary method of separation for algae [9].
Flocculation can be thought of as enhanced sedimentation. Microalgae generally have a
negative charge on them due to the different ions they take in which causes them to repel one
another. Flocculation is the process of adding a chemical that negates this charge and allows the
microalgae to clump or form flocs which causes the particles to settle. It is also possible to use
autofloculation where the charge of on the microalgae is negated by changing parameters such as
pH and oxygen levels. Flocculation has a high recovery rate reaching upwards of 80 percent but
it is a very sensitive process. The efficiency of the flocculation depends on maintaining the
levels of all materials in the cells medium. In one variable is out of place the efficiency can
drop. The process also runs the risk of contaminating the end product with the chemicals that are
added for flocculation [9].
Flotation can be thought of as reversed sedimentation. The process starts out the same as
flocculation in that chemicals are added in order for the microalgae to floc together. Air or gas is
then added to the fluid which attaches itself to the cell flocs and the overall decrease in density
causes them to rise to the surface. The resulting slurry is then scrapped off the top of the media.
This process is mostly used for the removal of microalgae from drinking water. Flotation has the
6
advantage of working well is large scale applications but the end result does tend to
contaminated by the flocculation chemicals [9].
2.3 Acoustic-assisted Cell Separation Methods
There are a variety of different acoustic-assisted cell separation techniques that can be
used to harvest or recover cells from the bioreactor fluid. This chapter will present a review of
the four methods that use ultrasound in the concentration of cell solutions: laminar flow
filtration, cross flow filtration, mesh filtration, and sedimentation.
2.3.1 Ultrasonic-assisted Laminar Flow Filtration
By definition laminar flow is simply orderly fluid flow that runs parallel to the walls of
the object it is flowing through. This means there are no cross currents, eddies or swirls within
the fluid that would disrupt the motion of the fluid through the object. When this is applied to
filtration it simply means that a particle or cell suspension flows through a filter without any
disruption in the movement of the fluid. When ultrasound is not applied to a laminar flow filter
the separation depends on the filter to guide the cells into their proper positions for separation
using designs such as asymmetric pinched-flow fractionation or deterministic lateral
displacement [2].
When using ultrasound to enhance laminar flow filtration, the cells' positions can be
altered by means of a standing wave. A standing wave is a wave that is passed perpendicularly
through the flow of fluid and is then met by an identical wave or the original wave is reflected
back on itself. This process creates a series of nodes and antinodes or areas of low acoustic force
and high acoustic force. The nodes can be found at every half and full wavelengths of the
7
ultrasonic wave and the antinodes at every quarter and three quarter wavelengths. When a fluid
containing cells or particles enter the area where the ultrasound wave is applied the cells are
pushed away from the antinodes due to the high acoustic force and pushed towards the nodes.
This effectively lines up the cells or particles into discreet lines at the position of the nodes. This
phenomena is well illustrated by the 1998 paper by Hawkes [3] in which the purpose of the
experiment was not to implement better filtration techniques but to actually observe the behavior
of the cells when the ultrasound wave was introduced. This was done by passing yeast cells
through an acoustic chamber where an ultrasound wave was produced by a transducer and then
reflected back on to itself by a brass reflector. This was done in a narrow area and then once the
cells had left the ultrasonic field the area widened so that the results of the ultrasonic field could
be more easily seen. This worked due to the properties of laminar flow that state there are no
cross currents or eddies so there would be no movement that would disrupt the position of the
cells in straight lines. The experiment showed that it was possible to form distinct bands of cells
using an ultrasound wave and to expand them using laminar flow.
Furthering on with this work Hawkes had another paper published three years later in
2001 [4]. This paper actually took the results of their previous work and put them to practical
use for cell filtration. The basic idea was to have one inlet bringing in cell filled media which
would then be separated into two outlets, one with all the cells flowing out of it and one with
cell-free media flowing out of it. The design of the filter itself was brilliant in its simplicity. The
fluid entered the bottom of the filter where it found itself in a chamber 10 mm wide where a
single node formed at 5 mm, half wavelength at the frequency used, of 3 MHz. The yeast cells
then passed through this ultrasonic field and were shaped into a straight line at the 5 mm node.
The cells then left the ultrasonic field but did not disperse due to the laminar flow. A splitter was
8
then used to separate the cell filled media from the clarified media and each flowed out of its
respective outlets. They concluded that this method was comparable to others such as membrane
filtration except without the downsides of long residence time and membrane fouling.
2.3.2 Ultrasonic Cross-Flow Filtration
Cross-flow filtration is a fairly simple method of separating cells. The suspension of
cells is pushed into a filter where it has two different paths to choose from. The first path simply
takes the solution out of the filter with no resistance or filtering. The second path takes the fluid
through a membrane with small holes that does not allow large particles or cells through. This
allows small cells or particles to become trapped within the membrane while allowing the fluid
to pass through. For this reason cross-flow filtration is an excellent way of isolating small
particles or cells from large particles or cells in a heterogeneous mix. Theoretically a cross flow
filtration system can run until the membrane is filled with small particles or cells and becomes
fouled. Since the fluid flow will cause some or the larger particles to impact the membrane they
can sometimes remain stuck to the membrane and cause clogging. Once the membrane becomes
fouled it loses permeability so it is not as effective in letting the small particles or cells through.
This is where ultrasound comes in.
There are two traditional options to overcome the problem of membrane fouling. The
first is to simply remove the membrane and to replace it with a new one. This can get very
expensive. The second is to stop the filtration process and run cleaning fluid through the
membrane followed by a rinse to remove the remaining cleaning fluid. The problem with using
chemicals to clean the filter is twofold: The cost of the chemicals can decrease the profit margin
and depending on the chemical to be used, special disposal methods may be in order. Ultrasound
9
opens up a third option. Ultrasound applied to the cross flow filtration membrane can cause
cavitation [4], which is pressure variations which cause rapid fluid movement. The result of this
is micro bubbles rapidly forming and popping on any available surface. The agitation by the
bubbles helps to break up the membrane fouling layers and helps to increase the permeability of
the membrane.
A good example of this came from a 1998 paper by Xijun Chai [5]. The authors used 1 wt
% dextran solutions of different molecular weights. It was shown that by applying ultrasound to
the membrane, a significant increase in the permeate flux through the polyacrylonitrile
ultrafiltration membrane was obtained, with an increased change of flux with an increase in
molecular weight.
2.3.3 Ultrasonic Mesh Filtration
Mesh filtration is fairly simple. A cell suspension flows through a mesh and the larger
cells are trapped by the mesh while smaller cells simply pass through. This is due to the holes in
the mesh being similar in size to the larger cells. Ultrasonic mesh filtration makes things more
complicated. The pores in the mesh do not trap the cells. Rather, the mesh causes a
rearrangement of the acoustic field, leading to the entrapment of cells at discrete nodal points.
Consequently, the pore size of the mesh can be significantly larger than the particle size and thus
larger than the pore size of the standard mesh filter. This allows the cells or particles to be easily
flushed from the mesh when the mesh is full since the particles due not stay trapped once the
ultrasound is turned off. While we are unsure of what exactly are the forces behind this
phenomenon, experiments and modeling have shown patterns that would help to explain it.
Modeling by Grossner et al. gives three possible scenarios or a combination of the three
scenarios: the particles are forced along paths that create dendritic structures, the particles are
10
held in place purely by the acoustic force, and/or the particles are clumped together so they are
too large to pass through the pores [6].
Ultrasound can also to be used to assist regular mesh filters in a manner similar to that of
ultrasonic-assisted cross-flow filtration. Meshes with pore sizes smaller than the particle size
will foul or clog over time and the ultrasound wave can be used to break up the fouling. A nice
paper put out by the Israel Institute of Technology in 2005 shows how ultrasound helps improve
the live span of everyday HEPA filter [07]. They found that at a decibel range of 110-130 dB,
the lifetime of the HEPA filter can be increased up to 10 times at low frequencies.
2.3.4 Ultrasonic Sedimentation
Sedimentation is the simplest cell separation technique available. The basic technique is
to allow a cell suspension to sit in a container until all the cells clump together and sink to the
bottom since they are not being agitated. There are two major problems with this method
though: time and clumping. If cell separation is achieved this way it tends to take an excessive
amount of time. This makes it unsuitable for large scale cell separation since a large amount of
space would have to be allocated for a relatively small amount of cells. Clumping becomes a
problem simply due to the fact that some cells will not clump together without something to
encourage clumping as in the flocculation process. Adding chemicals to induce clumping will
result in problems in cell use and waste disposal.
Ultrasonic sedimentation helps to solve both of those problems. Ultrasonic sedimentation
uses a standing ultrasound wave in order to force cells closer together to induce clumping. All
the cells in an area exposed to the ultrasonic standing wave are drawn to the nodes of the wave,
clump together, and fall to the bottom of the container. This greatly decreases the amount of
11
time needed for the sedimentation and it makes sure the cells will come into contact with each
other.
One of the earliest examples of this method was in a paper by Trampler et al. in 1994
[10]. The system was very efficient reaching regular separation efficiency of 97%. The system
drew cells into a 32 ml acoustic resonator from a stirred cell reservoir which was placed directly
under the resonator. The cells form lines within one second of entering the resonator and clump
together, until the gravitational force on the cells was greater than the fluid flow force so they
settled back into the reservoir. The fluid that continued through the resonator was nearly cell
free and eventually recycled back into the reservoir.
The technology has continued to be used for cell separation and the major producer of
ultrasonic sedimentation filters is Applikon Biotechnology. Their ultrasonic sedimentation
filters are referred to a BioSep (biological separaters) and range from 1 L/day models to 1000
L/day models. A 2003 paper by Bosma et al. demonstrates the effectiveness of one of the
BioSep models quite well. A cell suspension was pumped into the BioSep through the side of
the unit. The fluid was then pulled upwards by a second pump where the cells where exposed to
a standing ultrasonic wave. The standing wave forced the cells together until the force of gravity
on the clumps of cells was greater than drag force of the fluid and the cells settled into an
outflow on the bottom of the BioSep. The outflows flow rate was equal to the different between
the fluid flows through the pumps. The results for microalgae harvesting using species Monodus
subterraneus where encouraging as they were able to reach 90% efficiency at flow rates of 4-6
liters per day. Efficiency was calculate using the formula:
Efficiency=100%-100%*(Creturn*Фreturn)/(Cin*Фin)
where C is the cell number and Ф is the flow rate [11].
12
2.3.5 Zhaowei Wang’s Ultrasonic-assisted Filter
As my experiment is based in the work done by Zhaowei Wang there are many
similarities in set up, filter design, and implementation. The basic set up is nearly identical with
the only noticeable difference being I declined to connect a computer to the oscilloscope during
the experiments. The dimensions of my design mimic Zhaowei’s except for the slope angle
which is 10 degrees in my design and 6.8 degrees in Zhaowei’s. The outer dimensions are also
different due to my filter being designed for easy disassembly using bolts to hold the filter
together where Zhaowei’s design is slightly smaller and is held together using epoxy.
Implementation of the designs is where you begin to find differences. Zhaowei used polystyrene
particles and hamster hybridoma cells in his design where I used algae (Scenedesmus
dimorphus) cells. Zhaowei’s experimental procedure was to have constant voltage across the
transducers that allowed for cell filtration and then alter the flow rate to determine the fastest
flow rate possible while maintaining high cell retention. My experiments were a reversal of that.
I kept a constant flow rate which displayed high cell retention at high voltages and then ran tests
at different levels of voltages to determine how much voltage was actually needed for cell
retention to occur. Zhaowei’s experiments were designed with making his process as fast as
possible where my experiments where focused on making my process as energy efficient as
possible.
2.4 Ultrasound Effects on Algae
When energy is introduced to a system of living organism the effect of those living
organisms must be taken into account. In this experiments described herein, Scenedesmus
dimorphus algae is exposed to an ultrasound standing wave. Studies have been done on the
effects of ultrasound or sonication on similar strains of algae. Meghana Gavand published a
13
paper in 2007 on the effects of sonication and oxidants on algae which show a sizable mortality
rate for algae when ultrasound is introduced [12]. The mortality rate reaches 38% when exposed
to the ultrasound wave over a period of 20 min. While this is an alarming statistic it should not
affect the functionality of the experiment. Gavand’s experiments also showed a 0% mortality
rate below five minutes of constant exposure. A paper by T. J. Lee helps to explain the effect of
the ultrasound on the algae [13]. His study showed that ultrasound decreases the photosynthetic
rate by damaging the system of photosynthesis. In the work presented here the algae will only be
exposed for brief periods of time and thus are not expected to be negatively affected.
14
Chapter III
Theory and Methods
3.1 Theory
The success of this experiment relies on the principle of reflected standing waves. A
reflected standing wave is in ultrasound wave that is produced from a transducer and which is
then reflected back on the wave. This creates the presence of nodes and antinodes or areas of
high and low acoustic force. In ultrasound-assisted laminar flow applications the purpose of
creating an ultrasound standing wave is to separate the particles or cells into distinct lines and to
direct those lines of cells into specific exit positions. The operating principle of the ultrasound
assisted gravity filter are slightly different. The ultrasound standing wave is not intended to keep
the particles or cells in straight lines; rather, it is used to hold them in their horizontal positions
which causes them to collect on the slope of the gravity filter so they can be siphoned off through
the outlet port.
Basic filter design internal
15
Basically the cells in the ultrasound assisted gravity filter will be under the effect of three
forces: gravity, acoustic force and hydraulic drag force. The effects from gravity should be
minimal. The fluid and cells are flowing in the direction of gravity, and gravity may just
accelerate slightly the cell at accumulation on the lower surface. The primary forces the
experiment will have to deal with are the acoustic force and the hydraulic drag force. The
acoustic force is calculated using the following equation [1]:
Fac=4πR3κEacsin(2κx)F 3.1
where R is the radius of the particle, Eac is the energy density of the ultrasound wave which can
be expressed as the power per square meter, x is how far the cells is from a node, κ is the wave
number and F is the acoustic contrast factor. Now the wave number and the acoustic contrast
factor are not constants but have formulas of their own which are
κ=2 πλ
3.2
F= 13 ( 5 ρP−2 ρf
ρf +2 ρP
−Ɣ P
Ɣ f) 3.3
where λ is the wavelength, ρP is the particle density, ρf is the fluid density, Ɣf is the fluid
compressibility and ƔP is the particle compressibility. The acoustic force formula (Eq. 3.1) is
used to calculate the force that is applied on the cells and the acoustic contrast factor (Eq. 3.3)
determines their direction. If the acoustic contrast factor is positive then the cells go towards the
nodes, if the acoustic contrast factor is negative then the cells go towards the antinodes. In this
experiment it should not matter whether the cells go towards the nodes or the antinodes as long
as they are held in position. This will occur as long as the acoustic force is greater than the
hydraulic drag force in the horizontal direction. This will follow the formula
FD=6 πμ ( v f −v p ) R sin θ 3.4
16
where µ is the fluid viscosity, vf is the fluid velocity, vp is the particle velocity, R is the particle
radius and θ is the angle of the slope from the y-axis.
3.2 Materials
For the actual design layout and its evolution see appendix A at the back of the paper.
The main filter design was made up of: a central filtration chamber, two cooling chambers, two
polycarbonate films, two end pieces with transducers attached, and six gaskets. There were a
variety of materials used to build the filter and some were replaced over time as the experiment
went on. The central chamber was the only consistent part throughout the experiment and it was
constructed out of a block of acrylic. The cooling chambers originally started as one cooling
chamber and one reflecting block both made out of polycarbonate. The block was changed later
on to a second cooling chamber made out of acrylonitrile butadiene styrene (ABS). There was
originally only one end piece made of borosilicate glass which was switched for polycarbonate
when it was determined the acoustic impedance of borosilicate was a bad match for the
experiment. A steel plate was added to the design as a second end piece at that time to act as a
reflector with a transducer being attached to the polycarbonate end piece. The steel plate was
replaced with a second piece of polycarbonate and a second transducer was attached to it.
Further design changes saw the removal of the polycarbonate which was replaced by two solid
pieces of silicon rubber. The transducers were held to the silicon by two frames made of ABS.
The gaskets in the filter were also made out of silicon rubber with two of the gaskets being
replaced by the solid pieces of silicon that the transducers rested on. The filter was held together
by four ¼ in, 100 mm bolts and four ¼ in wing nuts.
17
3.3 Setup and Operation
The basic setup of the experiment was the same throughout the experiment. This can be
seen in appendix B at the back of the paper. The operation of the filter was kept fairly simple.
The pump connected to the central chamber outlet closest to the slope and the pump connected to
the second outlet and the cooling water flow were kept at flow rate ratio of 1:2 and later switched
to a flow rate ratio of 1:4. Various flow rates were attempted ranging from total flow rates of 20
ml/min to 10 ml/min. The transducer(s) attached to the sides of the filter had a variety of
voltages applied ranging from 0.1 Vp to 0.3 Vp in increments of 0.02 Vp after these voltages
were run through a 50 db amplifier. The frequency of the voltage applied was varied in order to
obtain the best possible power factor but was consistently around 2.3 MHz. Samples were taken
from the input line and the two output lines from the filter in intervals ranging from 15 min to 30
min, increasing the time as the flow rate decreased. Three sets of samples were taken at each
voltage setting.
18
Chapter IV
Results
No significant qualitative data was obtained from the performed experiments. All
samples taken showed no increase in cell concentration in the concentrate line. When only one
side of the filter had a transducer attached there was no visible change to the filter. When
transducers were placed on polycarbonate and put on opposite sides of the filter the epoxy
holding the transducers to the polycarbonate began to blacken from excessive heat. The
polycarbonate plates were replaced with the ABS frames and silicone rubber sheets to avoid
using epoxy to hold them together. This was unsuccessful as the excess heat melted the solder
that connected the wires to the transducers. Epoxy was placed over the solder in an attempt to
keep the wires connected but the epoxy again blackened and proceeded to vibrate off the
transducers.
19
Chapter V
Discussion
While the experiment has failed to show positive results there are steps that can be taken
to possibly make future endeavors in this line of experimentation successful. The largest issue to
solve is the excess heat buildup causing part failure. The actual cause of this is unknown at this
time but it is thought to be due to either insufficient cooling or due to the transducers being held
two tightly changing vibrational energy to heat energy. Possible solutions to this are to decrease
the temperature of the cooling water to combat the excess heat and redesign the frames holding
the transducers in place in order to give the transducers room to vibrate. Leaking was an issue at
times but it may be easily solved by increasing the thickness of the silicon rubber gaskets to
allow for more compression and a better seal. These are positive steps in fixing the experiment
but it all hinges on whether or not Scenedesmus dimorphus can be concentrated using
ultrasound. The physical properties (crescent shape, clumping) of Scenedesmus dimorphus may
stop it from acting in the manor proposed by the experimental theory. This is something that
must be taken into consideration before continuing this line of experimentation.
20
Schematics
21
22
23
24
25
26
27
Works Cited
[1] Zhaowei Wang. “Two Approaches for Cell Retention in Perfusion Culture Systems”.Thesis for Doctorate of Engineering in Applied Biomedical Engineering. Cleveland State University, December 2009.
[2] Hideaki Tsutsuiand Chih-Ming Ho. “Cell separation by non-fluidic forces in microfluidicsystems”. Mechanics Research Communications 36 (2009) 92–103
[3] Jeremy J. Hawkes and associates. “A laminar flow expansion chamber facilitating downstream manipulation of particles concentrated using an ultrasonic standing wave”.
Ultrasonics 36 (1998) 901-903
[4] E. S. Tarleton and R. J. Wakeman. “Electro-Acoustics Crossflow Microfiltration”. Filtration Society. Manchester May 19, 1992.
[5] Xijun Chai, Takaomi Kobayashi and Nobuyuki Fujii. “Ultrasound effect on cross-flow filtration of polyacrylonitrile ultra®ltration membranes”. Journal of Membrane Science 148 (1998) 129-135.
[6] Joanne M. Belovich and Associates. “Single fiber model of particle retention in an acoustically driven porous mesh”. Ultrasonics 41 (2003) 65–74
[7] L. Moldavsky, M. Fichman and C. Gutfinger. “Enhancing the performance of fibrous filters by means of acoustic waves”. Aerosol Science 37 (2006) 528–539
[8] Carlos Encarnaciόn and Associates. Maximization of Scenedesmus Dimorphus Lipid Yield for the Production of Biodiesel. Diss. Polytechnic University of Puerto Rico, 2010.
[9] Luisa Goveia. Microalgae as a Feedstock for Biofuels. Heidelberg: Springer. 2011
[10] Felix Trampler and Associates. “Acoustic Cell Filter for High Density Perfusion Culture ofHybridoma cells”. Bio/Technology 12 (1994) 281-284
[11] Rouke Bosma and Associates. “Ultrasound, a new separation technique to harvest microalgae”. Journal of Applied Phycology 15 (2003) 143-153
[12] Meghana Gavand and Associates. “Effects of sonication and advanced chemical oxidants on the unicellular green alga Dunaliella tertiolecta and cysts, larvae and adults of the brine shrimp Artemia salina: A prospective treatment to eradicate invasive organisms from ballast water”. Marine Pollution Bulletin 54 (2007) 1777–1788.
[13] T.J. Lee and Associates. “Ultrasound Irradiation for blue-green algae bloom control”. Environmental Technology 22 (2001) 383-390.
28