VISUALIZATION OF CHLORELA LGAL CELS AT 
BUBLE SURFACES 
 
 
 
Except where reference is made to the work of others, the work described in this thesis is 
my own or was done in collaboration with my advisory commite. This thesis does not 
include proprietary or clasified information. 
 
 
 
 
 
 
Stephen Alexander Tuin 
 
 
Certificate of Approval: 
 
 
 
 
Gopal A. Krishnagopalan    Steve R. Duke, Chair 
Profesor      Asociate Profesor 
Chemical Engineering    Chemical Engineering 
 
 
 
 
Ronald A. Putt     Jin Wang 
Asistant Research Profesor    Asistant Profesor 
Chemical Engineering    Chemical Engineering 
 
 
 
 
Joe F. Pitman 
Interim Dean 
Graduate School 
VISUALIZATION OF CHLORELA LGAL CELS AT 
BUBLE SURFACES 
 
 
 
Stephen Alexander Tuin 
 
 
 
 
 
 
A Thesis 
Submited to 
the Graduate Faculty of 
Auburn University 
in Partial Fulfilment of the 
Requirements for the 
Degre of 
Master of Science 
 
 
 
 
Auburn, Alabama 
May 10, 2008
 ii 
VISUALIZATION OF CHLORELA LGAL CELS AT 
BUBLE SURFACES 
 
 
 
Stephen Alexander Tuin 
 
Permision is granted to Auburn University to make copies of this thesis at its discretion, 
upon request of individuals or institutions and at their expense. The author reserves al 
publication rights. 
 
 
 
 
 
 
       Signature of Author 
 
 
 
 
    Date of Graduation 
 iv 
VITA 
 
 
 Stephen Alexander Tuin, son of Van and Patricia Tuin, was born on July 7, 1983 
in Brighton, Colorado. He graduated from Paonia High School in 2002. He graduated 
from the University of Colorado at Boulder with a Bachelor of Chemical Engineering in 
May 2006.  He entered Graduate School at Auburn University in the Department of 
Chemical Engineering in August 2006. 
 
 v 
THESIS ABSTRACT 
VISUALIZATION OF CHLORELA LGAL CELS AT 
BUBLE SURFACES 
 
Stephen Alexander Tuin 
Master of Science, May 10, 2008 
(B.S., Chemical Engineering, 2006) 
 
135 Typed Pages 
 
Directed by Steve R. Duke 
This thesis examines flotation as a method for harvesting the gren algae chlorela 
for its use as a fedstock for biofuel.  Lipids extracted from algae can be converted to 
biodiesel for use as an alternative renewable energy source.  Visualization and image 
procesing techniques were utilized to study algae flocculation and flotation both 
qualitatively and quantitatively. It is hypothesized that imaging techniques may be used 
to gain insight into the proces of algal flotation. 
Flocculation of algae with iron nitrate, alum, chitosan, gelatin, and celulose was 
explored.  Samples of flocculated algae were imaged with light microscopy and 
procesed to determine the average equivalent floc diameter.  Iron nitrate and chitosan 
chemistries produce larger flocs than alum and gelatin chemistries.  Algae flocculated 
with 50 ppm iron nitrate and a 10:1 celulose to algae mas ratio was found to produce
 vi 
the largest flocs of 190 ?m diameter. The smalest average equivalent diameter of 24 ?m 
was observed for alum and gelatin concentrations of 100 ppm and 6.25 ppm respectively. 
Increasing the concentration of the secondary flocculant decreased the average equivalent 
diameter. For iron nitrate and chitosan chemistries, the addition of celulose can replace 
the need for chitosan. 
Algal cel and bubble interactions were imaged in thre bubble facilities:  The 
stationary bubble facility which suspends a bubble on the tip of a needle in a quiescent 
fluid, the suspended bubble facility which suspends a bubble in a down flow of fluid, and 
the electrochemical flotation cel which generates very smal diameter bubbles. 
No algal adsorption to bubbles was observed in the stationary bubble facility. 
This was atributed to the large diameter of the bubbles produced in this facility 
(approximately 1 m). Images were useful to visualize the approach of individual algae 
cels and larger algae flocs to bubble surfaces. Algae adsorption to bubbles was observed 
in the suspended bubble facility during filing of the facility but adsorption was not 
observed during normal operation. Algae adsorption to bubble surfaces was succesfully 
imaged in the electrochemical flotation cel. 
Flotation runs were conducted in an electrochemical flotation cel and a Denver 
D-12 flotation cel. Algae foam collected in these facilities was procesed into algae pads 
via vacuum filtration and the mas of algae floated was determined.  The flotation 
eficiencies for various flocculation chemistries was evaluated. Iron nitrate and chitosan 
concentrations of 50 ppm and 1.25 ppm respectively produced the highest flotation 
eficiency of 83%.  Flotation eficiency increased with increasing average equivalent 
diameter. 
 vii 
ACKNOWLEDGMENTS 
The author would like to thank Dr. Steve Duke for support during this research. 
His help, motivation, and understanding were esential to the succes of this research. 
The author would like to thank Dr. Ron Putt for his asistance and insight into algal 
culture. 
The author would also like to thank al of the undergraduate, graduate, and 
exchange students that developed the methods and equipment used in this research.  A 
special thanks to Lily Raines for her asistance with high-magnification visualization. 
Finaly, the author would like to thank his parents, Van and Trish, his brother, 
Andrew, and his sister, Heather for their support through this research. The author would 
especialy like to thank Jil Blecha, whom without her love and support this research 
could not have succeded. 
 vii 
Style Manual: TAPI JOURNAL, ACS        
Computer Software Used: Microsoft Word, Microsoft Excel, Corel PhotoPaint 11,  
    ImageJ, VideoMach      
 ix 
TABLE OF CONTENTS 
 
LIST OF TABLES.....................................................xi 
LIST OF FIGURES....................................................xii 
CHAPTER 1: INTRODUCTION..........................................1 
CHAPTER 2: BACKGROUND...........................................4 
2.1 Algae as a Fedstock for Biofuel...............................5 
2.2 Chlorela as a fedstock for Biofuel.............................7 
2.3 Induced and Disolved Air Flotation...........................10 
2.4 Flocculation of Algae.......................................14 
CHAPTER 3: EQUIPMENT AND EXPERIMENTAL PROCEDURES...........17 
3.1 Algal Culture.............................................18 
3.2 Chemicals...............................................24 
3.3 Stir Plate Calibration.......................................25 
3.4 Flocculation Beaker Tests...................................28 
3.5 Microscope Slide Photographs................................34 
3.6 Stationary Bubble Facility...................................41 
3.7 Suspended Bubble Facility...................................47 
3.8 Electrochemical Flotation Cel................................51 
3.9 Denver D-12 Flotation Cel..................................57 
3.10 Algae Pads...............................................61
 x 
3.11 Imaging System...........................................64
CHAPTER 4: RESULTS...............................................65 
4.1 Flocculation Beaker Tests and Microscope Slide Results............66 
4.2 Stationary Bubble Facility Results.............................74 
4.3 Suspended Bubble Facility Results.............................77 
4.4 Electrochemical Flotation Cel Results..........................84 
4.5 Denver D-12 Flotation Cel Results............................92 
CHAPTER 5: DISCUSION OF RESULTS................................95 
5.1 Discussion of Algal Flocculation Results........................96 
5.2 Discussion of Algal Cel Adsorption to Bubbles..................101 
5.3 Discussion of Bubble Imaging Results.........................107 
5.4 Recommendations........................................108 
CHAPTER 6: CONCLUSIONS.........................................109 
BIBLIOGRAPHY....................................................112 
APENDIX A: BLOB ANALYSIS PROCEDURE..........................115 
APENDIX B: MOVIE BUILDING PROCEDURE.........................117 
APENDIX C: LIST OF VIDEOS.......................................119 
APENDIX D: FLOC SIZE ANALYSIS..................................121
 xi 
LIST OF TABLES 
Table 3-1: Flocculation beaker test run chemistries.........................31 
Table 3-2: Chemistries tested in the bubble facilities........................46 
Table 4-1: Chemistries for Figure 4-1 and Figure 4-2.......................67 
Table 4-2: Chemistries for Figure 4-3...................................69 
Table 4-3: Chemistries for Figure 4-4...................................70 
Table 4-4: Chemistries for Figure 4-5...................................71 
Table 4-5: Chemistries for Figure 4-6...................................72 
Table 4-6: Chemistries for Figure 4-7...................................73 
Table 4-7: Chemistries for Figure 4-9...................................76 
Table 4-8: Chemistries for Figure 4-12..................................81 
Table 4-9: Chemistries for Figure 4-14..................................83 
Table 4-10: Chemistries for Figures 4-16 through 4-19.......................87 
Table 4-11: Flotation eficiency calculations for the EFC.....................90 
Table 4-12: Chemistries for Figure 4-20 and Figure 4-21.....................93 
Table 5-1: Chemistries for Figure 5-3..................................106 
Table C-1: List of Videos............................................119 
Table D-1: Flocculation beaker test calculations...........................121 
 xii 
LIST OF FIGURES 
Figure 2-1: TEM micrograph of an individual chlorela vulgaris cel.............9 
Figure 2-2: Voith Sulzer Eco flotation cel................................12 
Figure 2-3: Disolved air flotation cel...................................13 
Figure 2-4: Structure of chitin and chitosan...............................16 
Figure 3-1: Outdoor algal growth pond...................................21 
Figure 3-2: Relationship betwen algal concentration and 
transmitance at 550 nm.....................................22 
Figure 3-3: Indoor algal growth tanks....................................23 
Figure 3-4: Fisher stir plate 210T calibration curve..........................27 
Figure 3-5: Typical flocculation beaker test 1 minute after 
addition of the secondary flocculant............................32 
Figure 3-6: Typical flocculation beaker test after 15 minutes of setling..........33 
Figure 3-7: Low magnification (10x) microscope image 
of a typical floc...........................................37 
Figure 3-8: High magnification (40x) microscope image 
of a typical floc...........................................38 
Figure 3-9: Low magnification (10x) microscope image 
 of a typical floc with celulose................................39
 
 xii 
Figure 3-10: High magnification (100x) microscope image 
 of fresh algae and celulose...................................40 
Figure 3-11: Schematic of the stationary bubble facility.......................43 
Figure 3-12: Stationary bubble facility....................................44 
Figure 3-13: Typical frame from the stationary bubble facility..................45
Figure 3-14: Schematic of the suspended bubble facility.......................49 
Figure 3-15: Suspended bubble facility....................................50 
Figure 3-16: Schematic of the electrochemical flotation cel....................54 
Figure 3-17: Electrochemical flotation cel.................................55 
Figure 3-18: Algae foam generated in the electrochemical flotation cel...........56 
Figure 3-19: Schematic of the Denver D-12 flotation cel......................59 
Figure 3-20: Denver D-12 flotation cel...................................60 
Figure 3-21: Algae pads from a typical flotation run..........................63 
Figure 4-1: Flocculation beaker test results while stiring.....................67 
Figure 4-2: Flocculation beaker test results after setling......................68 
Figure 4-3: Microscope slide results for alum/gelatin chemistries...............69 
Figure 4-4: Microscope slide results for 
alum/gelatin/celulose chemistries.............................70 
Figure 4-5: Microscope slide results for 
iron nitrate/chitosan chemistries...............................71 
Figure 4-6: Microscope slide results for 
iron nitrate/chitosan/celulose chemistries........................72 
Figure 4-7: Floc to fiber ratio for celulose chemistries.......................73 
 xiv 
Figure 4-8: Series of frames in the stationary bubble facility...................75 
Figure 4-9: Frames of the stationary bubble facility for 
bubble facility test chemistries................................76 
Figure 4-10: Series of frames during filing of the suspended 
 bubble facility............................................79 
Figure 4-11: Frame from filing of the suspended bubble 
facility highlighting adsorbed algae............................80 
Figure 4-12: Frames of the suspended bubble facility during  
filing for bubble facility test chemistries........................81 
Figure 4-13: Series of frames during normal operation  
of the suspended bubble facility...............................82 
Figure 4-14: Frames of the suspended bubble facility during  
normal operation for bubble facility test chemistries................83 
Figure 4-15: Algal foam in the EFC during operation.........................86 
Figure 4-16: Algal foam in the EFC after shutdown..........................87 
Figure 4-17: Algae pads generated from the EFC............................88 
Figure 4-18: Algae pads generated from the EFC............................89 
Figure 4-19: Flotation eficiency in the EFC................................91 
Figure 4-20: Denver D-12 Flotation Cel during operation.....................93 
Figure 4-21: Algae pads from Denver D-12 Flotation Cel.....................94 
Figure 5-1: Comparison of flocculation beaker images......................100 
Figure 5-2: Comparison of large and smal bubbles in the suspended facility.....105 
Figure 5-3:  Flotation eficiency vs. average equivalent floc diameter...........106
 1 
CHAPTER 1 
INTRODUCTION 
 As the world?s energy needs continue to increase and with oil reserves predicted 
to run out sometime this century, development of an alternative energy source is of 
paramount importance.  Alabama has a unique opportunity to become self sufficient in 
liquid fuel production for transportation through algaculture. Alabama could produce its 
annual 3 bilion galons of fuel using les than 3% of the state?s land [Putt, 2007]. 
This thesis addreses harvesting of the gren algae chlorela using flotation 
proceses for use as a fedstock in the production of biofuels. The lipids extracted from 
algae can be procesed into biofuels. One of the major bottlenecks of biofuel production 
from algae is the harvesting step. The algae must be removed from production water and 
it must be concentrated before the lipids can be extracted.  Algae is at a very dilute 
concentration, in the range of a few hundred parts per milion when it is ready to be 
harvested.  The work in this thesis explores flotation as a method to separate chlorela 
from water.  Chlorela is a very fast growing and robust algae species that thrives in 
warm climates such as those of the southeastern United States. With adequate nutrients, 
it can double in concentration in 8 hours and has a lipid content betwen 20-30% by 
weight giving it large potential as a fedstock for biofuel [Putt, 2007; Sheehan et al., 
1998].
 
 2 
Flotation proceses are used to separate a suspended species from water, in this 
case algae cels or flocs. During flotation, air or another gas is introduced into the water 
and suspended particles adsorb to the bubble surfaces.  The bubbles then rise to the 
surface and create a foam rich in the particles to be separated. The hydrophilic nature of 
algal cels makes them respond poorly to flotation proceses, and prior atempts at large-
scale separation of chlorela using flotation have ben largely unsuccesful [Putt, 2007]. 
 Addition of flocculants to algae before flotation may lead to an increase in 
flotation eficiency.  Flocculation is the proces of individual cels interacting with one 
another to form larger networks of cels caled flocs. Larger particles are generaly more 
conducive to flotation, therefore it is proposed that flocculation of algae prior to flotation 
wil increase flotation eficiency. 
 This work focuses on determining appropriate flocculants and flocculant 
concentrations to produce large and stable flocs, and then evaluating these systems in 
various model flotation facilities. By imaging and observing qualitatively the interactions 
of algal cels and flocs with bubble surfaces, this study aims to develop quantification 
methods and to identify phenomena that wil lead to an efective and eficient algae 
flotation proces. 
Chlorela was grown in the laboratory and in outdoor ponds for use in 
experiments.  Qualitative and quantitative flocculation analyses were performed to 
determine efective flocculants and flocculant concentrations. Algae floc samples were 
imaged using light microscopy techniques to evaluate floc size. Flotation eficiency was 
evaluated using a bench scale flotation facility constructed by Dr. Ron Putt to model the 
disolved air flotation proces and a Denver D-12 laboratory scale flotation cel to model 
 3 
an induced air flotation proces.  Thre facilities were used to observe and image 
interactions betwen algal cels and bubbles.  This thesis presents methods and 
procedures for capturing high-speed digital video of algal cel and bubble interactions. 
This study hypothesizes that imaging techniques may be used to gain valuable insight to 
the mechanism of algal cel adhesion to bubble surfaces 
 4 
CHAPTER 2 
BACKGROUND 
 The primary objective of this research was to study algal cel interactions with 
bubble surfaces by utilizing existing imaging techniques developed by Davies [2000] and 
Emerson et al. [2003] as wel as novel laboratory scale flotation apparatuses. The goal of 
this study was to develop laboratory scale methods to separate algal cels from water 
using flotation proceses.  More broadly, the objective was to understand flotation of 
algae to afect harvesting proceses that wil increase the potential of algae as a fedstock 
for biofuel. 
 5 
2.1 Algae as a Fedstock for Biofuel 
 Due to the ever increasing world demand for oil and the fact that the world?s 
supply of oil is finite, avenues for production of alternative renewable sources of energy 
are of paramount importance. Biofuels such as biodiesel, gren diesel, and ethanol are 
currently being explored and produced to met the worlds expanding energy needs. 
Cultivation of micro-algae in the United States as a fedstock for biofuel has the potential 
to aleviate the nation?s dependence on foreign oil.  The possibility of becoming self-
sufficient in energy production via micro-algae cultivation has widespread economical 
and environmental impacts. 
 Micro-algae store energy in the form of carbohydrates and lipids.  Lipids 
produced by these algae can be converted into biofuels. Biodiesel or gren diesel can be 
produced from these algal lipids. Biodiesel is produced via a transesterification reaction 
betwen algal lipids and methanol [Han et al., 2006].  Gren diesel is very similar to 
traditional petroleum based diesel [Putt, 2007]. 
 The United States Department of Energy?s National Renewable Energy 
Laboratory (NREL) first conducted extensive research into algae cultivation for biofuels 
in the mid 1970?s, continuing their eforts into the mid 1990?s. The project was known as 
the Aquatic Species Program (ASP).  The project focused on identifying algal strains 
capable of producing a large quantity of lipids and maximizing growth conditions for 
lipid production. Many promising strains were isolated and their lipid distributions under 
various conditions were characterized. In some algal strains, it was discovered that lipid 
production could be as high as 50% by weight [Shehan et al., 1998]. 
 6 
 After promising algal strains were identified by the ASP, several laboratory and 
pilot scale facilities were constructed for feasibility asesments.  A facility in Roswel 
New Mexico consisting of two 1000 m
2
 growth ponds was constructed. Several strains 
were tested in this facility including tetraselmis suecica and monoraphidium minutum 
[Sheehan et al., 1998].  Two important conclusions can be drawn from the tests 
performed at this facility.  First, only strains native to the area where algal ponds are 
located should be used for cultivation.  Each strain tested at the Roswel site was 
eventualy overun by native strains.  It is too dificult to maintain an alien species in 
outdoor growth facilities. Therefore only native species should be used because they wil 
eventualy dominate the pond.  Second, location of algal farms should be considered 
carefully.  Algae production at the Roswel site was considerably reduced during the 
winter months due to low temperatures.  Algal farms should be located in climatic 
regions with an abundance of sunlight, water, and year-round temperature ranges 
conducive to algal growth [Sheehan et al., 1998].  Regions in the southeastern United 
States, including Alabama, met these requirements [Putt, 2007]. 
 7 
2.2 Chlorela as a Fedstock for Biofuel 
 The genus of gren algae chlorela is a strong candidate for algae farming in the 
southeastern United States for production of biofuel.  Chlorela cels are spherical in 
shape with celular diameters ranging from 2-10 microns.  Under nominal growth 
conditions, chlorela reproduces very quickly.  Cel concentrations are capable of 
doubling in 8 hours [Putt, 2007]. Chlorela requires only sunlight, carbon dioxide, and 
smal amounts of minerals such as phosphate and nitrogen to reproduce. Chlorela is also 
atractive for mas production because photosynthetic eficiency can theoreticaly 
approach 8% [Morita et al., 2001].  For comparison, the photosynthetic eficiency of 
most tres is about 0.5% and solar panels are typicaly 10-20%.  The lipid content of 
chlorela is typicaly 20% by weight of the dried cel mas [Sheehan et al., 1998]. Figure 
2-1 is a micrograph of a chlorela vulgaris cel [Stead et al., 1995]. 
There are several factors that make chlorela an atractive candidate for algae 
farming in the southeastern U.S. and particularly Alabama.  Chlorela is native to 
Alabama and is found naturaly growing throughout the state. Atempts to cultivate alien 
strains would likely be overtaken by chlorela. The lipid content is sufficiently high that 
cost efective production of biofuel could be realized.  Alabama has a large amount of 
available land and water suitable for chlorela algae production.  Lastly, the climate is 
conducive to year round growth of chlorela.  Production during the winter months of 
January and February and the summer month of July may be reduced but stil possible 
[Putt, 2007]. 
Harvesting of micro-algae presents a dificult chalenge.  The concentration of 
algae during the harvesting stage is very smal, on the order of a few hundred ppm. 
 8 
Separation of the algae from the water in such dilute concentrations can be problematic. 
Filtration systems have been succesfully implemented for harvesting of algae [Hung and 
Liu, 2006], however they require a large throughput and are prone to blinding. Setling 
tanks are a second option in which the algae is alowed to setle to the bottom of a 
quiescent pond in series with the growth pond.  The water above the algae is then 
removed leaving a concentrated algal mas on the bottom of the pond.  This approach 
requires separate setling tanks which can double the amount of required land.  The 
proces also takes several days, leading to a highly ineficient proces.  This study is 
focused on flotation of algae as an alternative algae harvesting technique. 
 After harvesting, the lipids can be removed from the algae in a solid-liquid 
extraction proces similar to those employed to separate soybean oil from the bean. The 
algal lipids can then be procesed into biofuels.  The remaining algal meal after lipid 
removal is very rich in protein and can be used as an animal fed.  Discussion of the 
conversion of harvested algae to biofuels can be found in Han et al. [2006]. 
 9 
 
Figure 2-1: TEM micrograph of an individual chlorela cel [Stead et al., 1995].  
c-chloroplast, p-pyrendoid, n-nucleus, and cw-cel wal. 
 
 
 10 
2.3 Induced and Disolved Air Flotation 
 Flotation is a proces that utilizes gas bubbles (typicaly air) to separate particles 
from a liquid. Flotation was first used in the mineralogical industry to separate mineral 
ore from water. It is also commonly employed in the paper recycling industry to remove 
contaminants from recycled paper. During flotation, air bubbles rising through a liquid 
collect particles that then rise to the surface of the water with the bubbles. The bubbles 
create a foam at the surface of the water which is skimed off. Flocculants or coagulants 
are often added to the fed water to flocculate contaminants. The contaminant flocs are 
more conducive to flotation. There are two main types of flotation proceses, induced air 
flotation and disolved air flotation (DAF). 
 Induced air flotation, also know as froth flotation, utilizes air bubbles injected into 
the vesel containing the liquid to be clarified. Air can be injected through the bottom of 
the vesel with a gas sparger or entrained via cavitation.  Contaminants contacted by 
rising bubbles are adsorbed to the bubble surface and travel to the water surface with the 
bubbles.  The bubbles form a contaminant rich foam on the surface of the water.  The 
foam is then skimed off to separate the contaminants from the clarified water. Bubble 
diameters in induced air flotation cels are typicaly betwen 1 and 2 m. These units are 
typicaly employed to remove larger contaminants.  Froth flotation cels are often 
operated in series to achieve a desired level of contaminant removal.  Figure 2-2 is a 
schematic of a Voith Sulzer EcoCel froth flotation unit [Martin and Britz, 1996]. 
 Disolved air flotation is similar to induced air flotation in that rising bubbles 
contact contaminants that rise to the surface and form a foam.  The primary diference 
betwen disolved and induced air flotation is the method air bubbles are generated. In 
 11 
DAF, the contaminated water to be clarified is presurized.  Air is sparged into the 
presurized stream until the water is saturated with air at the elevated presure.  The 
saturated water is then depresurized in the flotation cel. The presure drop causes the 
saturated air to come out of solution as tiny air bubbles. These bubbles are smaler than 
those produced in induced air flotation.  Typical bubble diameters range from 0.05-0.7 
m. Contaminants adsorb to the bubbles in the same manner as in induced air flotation 
and form a foam on top of the water.  The foam is then skimed to separate the 
contaminants from the water.  DAF is more expensive to operate than induced air 
flotation because of the presurization step. It is used to remove smaler contaminants. 
Figure 2-3 is a schematic diagram of a typical DAF cel [Biermann, 1996].  
The induced air flotation and DAF cels and proceses that may be used for algae 
are likely to be similar to those described here that are established for paper recycling and 
deinking proceses.  Yao-de and Jameson [2003] have studied algae flotation from 
wastewater maturation ponds utilizing Jameson flotation cel technologies. Bench scale 
and pilot scale experiments were used to validate the technology.  They have 
demonstrated that algal removal can approach 99% in a full-scale implementation facility 
in Wagga Wagga, Australia [Yao-de and Jameson, 2003].  Jameson flotation cel 
technologies for algal flotation are also discused in Jameson [1999]. Harvesting algae 
via froth flotation technology is discussed in Levin et al. [1961], Chen et al. [1998], and 
Liu et al. [1999, 2007]. DAF flotation principles are discused in Rodrigues and Rubio 
[2007]. 
 12 
 
Figure 2-2: Voith Sulzer Eco Flotation Cel [Martin and Britz, 1996]. 
 13 
 
Figure 2-3: Disolved air flotation cel [Bierman, 1996]. 
 
 14 
2.4 Flocculation of Algae 
 Flocculation is the proces in which a solute or suspended particles in a solution 
clump together to form a larger agglomerate refered to as a floc.  Algal flocculation 
describes the proces in which individual algae cels or groups of cels clump together to 
form an algae floc. Algae flocs are typicaly unstable and easily disturbed. Agitation of 
flocs wil cause them to break apart into smaler flocs or individual cels. Flocculants are 
chemicals that induce flocculation. 
 Flocculants are typicaly multivalent cations such as alum or iron chlorides or 
sulfates. These positively charged cations interact with the surface of algae cels, which 
typicaly cary a negative charge. This charge mediation alows cels to interact and form 
flocs.  The cel wals of gren algae are very hydrophilic in nature.  This hydrophilic 
nature is not very conducive to flotation [Yao-de and Jameson, 2003]. Charge mediation 
through flocculation may help to reduce the hydrophilic nature of the cel wals and 
increase the efectivenes of flotation.  Bilanovic and Shelef [1988] have shown that 
salinity is an important factor governing flocculation.  Increasing ionic strength lead to 
decreased flocculation efectivenes for chitosan, indicating that marine algal species 
could be more dificult to flocculate [Bilanovic and Shelef, 1988]. 
Several naturaly occurring biological polymers are known to induce flocculation, 
chief among these are chitosan [Kivakaran and Pilai, 2002] and gelatin [Kragh and 
Langston, 1961]. Chitosan is derived from chitin, the primary structural element in the 
exoskeleton of crustaceans [Planas, 2002].  Chitosan is produced by deacetylation of 
chitin, forming a high molecular weight polymer.  Figure 2-4 shows the chemical 
structure of chitin and chitosan. The amino group interacts with the negatively charged 
 15 
surface of algae cels [Planas, 2002]. Gelatin is derived from the connective tisue and 
bones of animals such as catle, pigs, and horses.  Algae may spontaneously flocculate 
without the addition of any flocculants.  This proces is known as auto-flocculation. 
Algae that is several days old wil often auto-floculate [Oh et al., 2001; Alonso et al., 
2000]. 
 16 
 
Figure 2-4: Structure of chitin (a) and chitosan (b) [Planas, 2002]. 
(a) 
(b) 
(a) 
(b) 
 17 
CHAPTER 3 
EXPERIMENTAL PROCEDURES AND EXPERIMENTS 
 The goal of this research was to visualize algal cel and bubble interactions to 
further understand the proces of algal flotation. In order to evaluate viable flocculation 
chemistries to be tested for flotation, several flocculants were examined in various 
concentrations and the resulting flocs were imaged under a microscope.  The average 
equivalent floc diameter was determined for each chemistry tested. After screning for 
viable flocculation chemistries, algal cel and bubble interactions were imaged in thre 
facilities.  The suspended bubble facility suspends a bubble on the tip of a needle in a 
quiescent fluid and algae is injected onto the bubble.  The suspended bubble facility 
suspends a bubble in a down flow of water alowing visualization of algae floc and 
bubble collisions.  The EFC is an electrochemical flotation cel used to evaluate the 
flotation eficiency of each flocculation chemistry tested. 
 18 
3.1 Algal Culture 
 The chlorela algae samples used in this work were obtained from outdoor algae 
growth ponds used in the research program of Dr. Ron Putt.  Dr. Putt utilized two 
concrete outdoor growth ponds at the North Auburn Fisheries Unit. The ponds measured 
2.75 x 7.6 m and were filed with water to a depth of 20 cm.  The water was pumped 
from Farm Pond 11, which is the Auburn Fisheries station reservoir that collects 
rainwater from the local watershed.  Pond algae was established from an inoculum of 
chlorela and supplied nutrients from 5 galons of poultry liter.  A paddlewheel and 
center wals provided mixing in the system.  Figure 3-1 is a photograph of one of the 
outdoor growth ponds.  Dr. Ron Putt developed a relationship betwen algal 
concentration and light transmitance at 550 nm.  Algae concentration was measured in 
the suspension via centrifugation, drying, and weighing the algal mas [Putt, 2007]. 
 In order to maintain a consistent stock of algae for experiments, an indoor 
chlorela culture was established in the laboratory.  Two 25 galon black plastic tanks 
were filed with 15 galons of tap water and alowed to come to room temperature. Each 
tank was agitated with a submergible fountain pump manufactured by Garden Treasures 
model MD170. Pumps were magnetic drive pumps with a range of 130-170 gph, 0.5 inch 
outlet diameter, 120 V, 60 Hz, 11 W, with maximum pumping head of 3.5 ft.  Pumps 
were atached in the center of each tank bed via rubber suction cups and set to the 
maximum of 170 gph.  Per the manufacturer?s instructions, 7.5 mL of amonia and 
chloramine remover (Jungle Laboratories) was added to each tank.  One tablespoon of 
al-purpose plant food manufactured by Expert Gardener was disolved in 1 cup of 
distiled water and added to each tank. An 18 inch fluorescent light bulb was mounted on 
 19 
a Unistrut support over each tank. Each light bulb was controlled with a timer and set to 
a 15 hour light/9 hour dark cycle. Room lights were also turned off during the dark cycle 
each night. Initialy each tank was iluminated with a diferent bulb. Tank A was lit by 
an 18 inch, 15 W GE AquaRays 9325K aquarium light bulb and Tank B was lit by an 18 
inch, 15 W Tropic Sun 5500K light bulb.  Each bulb was designed to simulate the 
ultraviolet spectrum of natural sunlight. (Algae growth was greater for the tank with the 
Tropic Sun bulb, thus the GE bulb was replaced with a Tropic Sun bulb.) Each tank was 
alowed to equilibrate for 24 hours. 
 After 24 hours 2 galons of algae inoculum supplied from Dr. Ron Putt?s outdoor 
pond at a concentration of 300 ppm was transfered to Tank A. The inoculum, contained 
in 2 one-galon containers, was imersed in the tank to alow the temperature to 
equilibrate. After 30 minutes the inoculum was transfered to the tank. After two days, 2 
galons of the algae in Tank A was transfered to 2 one-galon containers and alowed to 
equilibrate in the second tank.  After 30 minutes the algae in the 2 containers was 
released into Tank B. 
 After inoculation, the pH, temperature, and Sechi depth of each tank was 
monitored every day.  The Sechi depth provides a quick estimate of the algae 
concentration. A Sechi disk was constructed by mounting a circular white plastic lid to 
a wooden dowel. The dowel was marked with a ruler in 0.5 cm increments. The disk is 
inserted into the algae and a distance measurement is taken when the disk is no longer 
visible.  Dr. Putt developed a relationship betwen the Sechi depth and the algae 
concentration via measurements with a spectrophotometer.  Algae concentration was 
monitored periodicaly via the transmitance correlation developed by Dr. Ron Putt using 
 20 
a Milton Roy Spectronic 20D spectrophotometer. The transmitance correlation is shown 
in Figure 3-2. De-chlorinated tap water was added to the tanks as needed when the water 
level dropped as the algae was used for experiments.  If the pH fel below 7, sodium 
bicarbonate was added to the tank in order to maintain the pH in the optimal growth 
range of 7-9 [Mayo and Noike, 1994]. Figure 3-3 is an image of the indoor growth tanks; 
the water level is low because the photograph was taken just after samples were removed 
for experiments. 
 21 
Figure 3-1: Outdoor algal growth pond. 
 22 
Figure 3-2: Relationship betwen algal concentration and transmitance at 
550 nm. Data obtained by Dr. Ron Putt. 
 
 23 
 
Figure 3-3: Indoor algal growth tanks. 
 24 
3.2 Chemicals 
 Several flocculating agents were employed.  Feric nitrate nonahydrate was 
purchased from Fisher Scientific. Solid feric nitrate was disolved in distiled water to 
obtain a 10,000 parts per milion by mas (ppm) stock solution.  Aluminum sulfate 
16?H
2
O (alum) was purchased from Fisher Scientific.  Solid alum was disolved in 
distiled water to obtain a 10,000 ppm stock solution. Powdered chitosan, MW 100,000?
300,000 Da was purchased from Argos chemicals.  Stock solution was prepared by 
disolving 100 mg in 10 mL of 1 N hydrochloric acid and alowed to stir for 30 minutes. 
This solution was added to 90 mL of distiled water to obtain a 1000 ppm stock solution. 
Gelatin from an animal source, MW 100,000?300,000 Da was purchased from Fisher 
Scientific. Stock solution was prepared by adding 50 mg of gelatin to 100 mL of distiled 
water heated to 60 ?C on a hot plate and alowed to stir for 2 hours to obtain a 
concentration of 500 ppm. 
 The indoor chlorela growth tanks were supplied one tablespoon of Expert 
Gardener?s al-purpose plant food disolved in 1 cup of water once a wek. This plant 
food has a nitrogen-phosphorous-potasium distribution of 24-8-16.  Tap water used in 
the growth tanks was treated with Jungle Pond amonia and chloramine remover. 
Sodium bicarbonate purchased from Sigma-Aldrich was used to control the pH of the 
growth tanks.  The pH of each experimental condition was controlled using 0.05 M 
hydrochloric acid (HCl) and sodium hydroxide (NaOH) purchased from Fisher Scientific. 
 Celulose stock was prepared by mixing copier paper with water and was re-
pulped in a household blender to obtain a 5% by weight stock solution of celulose fibers. 
 
 25 
3.3 Stir Plate Calibration 
 To standardize experimental procedures for the flocculation beaker tests 
(described in section 3.4) it was necesary to calibrate the magnetic stir plate used in 
these studies.  A standard magnetic stir plate manufactured by Fisher Scientific (model 
210T) was used to stir algae samples with a 1 inch octagonal stir bar manufactured by 
Fisher Scientific. A relationship betwen the speed seting and the velocity in the fluid 
was developed. 
 One side of the stir bar was marked for reference and placed on the stir plate. The 
stir bar was filmed in motion for 4 stir plate setings with a high-speed camera recording 
at 500 frames per second (FPS). By observing the resulting video, the number of frames 
per 1 revolution of the stir bar was measured. Using the frame rate, the revolutions per 
minute (RPM) of the stir bar was calculated. 
 To verify the results of the high-speed video, the RPM of the stir bar was also 
measured with a Monarch Instruments Nova-Strobe DA/DB strobe light.  The stir bar 
was observed under the strobe with the room lights turned off for each of the 
manufacturer?s stir plate setings. The frequency of the strobe light was adjusted until the 
stir bar appeared to stop moving, making sure that the mark on the stir bar stayed in the 
same position. This frequency was recorded for comparison with the high-speed movies. 
The stir bar was also observed under the strobe while spinning in 200 mL of distiled 
water placed in a 250 mL Pyrex beaker used in the flocculation beaker test studies. 
 The angular velocity measured with the strobe light and high-speed videos was 
used to calculate the linear velocity for any stir plate seting in a 250 mL beaker at a 
distance of 2/3 the radius of the beaker. This velocity is denoted the two-thirds velocity 
 26 
(TV).  Equation 3-1 was used to determine the TV.  A typical rising air bubble in 
water has a rise velocity of about 1 ft/sec [Clift et al., 2005]. To simulate these velocities 
in our studies, the stir plate seting was adjusted to give a TV of 1 ft/sec. The Reynolds 
number of mixing (Re
mix
) was defined in Equation 3-2 and calculated for each stir plate 
seting. The relationship betwen the stir plate seting and the TV is shown in Figure 3-
4. 
 
water
tbar
mix
RPSD
?
!"
=e
Equation 3-1 
R = Radius of the beaker (0.2083 ft) 
 
D
bar
 = Diameter of the stir bar (0.0254 m) 
 
RPS = Revolutions per second of the stir bar 
?
water 
 = Density of water at 20 ?C (998.2 
3
kg
) 
?
water
 = Viscosity of water at 20 ?C (0.001002 Pa?s) 
RPSTV!=
3
2
Equation 3-2 
 27 
 
Figure 3-4: Fisher stir plate 210T calibration curve. 
 28 
3.4 Flocculation Beaker Tests 
 The flotation proces often employs the use of flocculants or coagulants to 
increase the size of particles to be floated, resulting in enhanced flotation eficiency 
[Rodrigues and Rubio, 2007]. A qualitative and semi-quantitative method for evaluating 
the ability of feric nitrate, alum, chitosan, gelatin, and celulose to flocculate algae alone 
and in combination was developed. Table 3-1 shows 33 combinations of concentrations 
of additives that were tested. Al runs were performed at 22 ?C and pH 7. 
 Algae stock for these experiments was taken from Tank A and stired throughout 
the experiments at room temperature (22 ?C). The pH of the stock algae was adjusted to 
7 with 0.05 M HCl and NaOH and monitored throughout the experiment. 
 Images of the algae flocs in the stired beaker for each run were taken with a 
Nikon D-40x camera with a Nikon G-Type AF-S DX lens operated in manual mode. The 
camera was mounted on a rail system to ensure the same field of view and scale for each 
image. The camera was mounted 16.5 cm from the edge of a breadboard (seventh row). 
The breadboard was positioned against the Fisher stir plate; the stir plate calibration 
procedure is described in section 3.3.  A white piece of display board was positioned 
verticaly behind the stir plate for background. Lighting was provided by a standard desk 
lamp with a 75 W bulb on the left side of the stir plate and a clamp lamp with a 75 W 
bulb on the right side.  To find the center of the stir plate, a stir bar was placed on the 
plate and alowed to come to rest in the middle. Target guidelines to mark the center of 
the plate were marked with a permanent marker. A metal 150 m ruler was positioned 
verticaly and held in place with a ring stand and clamp on the left side of the stir plate 
 29 
positioned on the center guide mark paralel to the breadboard. The camera lens zoom 
was set to 35 and was focused manualy on the ruler. 
 For each run, 200 mL of the algae stock was transfered to a 250 mL Pyrex beaker 
containing a Fisher 1 inch octagonal stir bar. The beaker was placed in the middle of the 
target guideline on the stir plate. The stir plate was set to 6 corresponding to a TV of 
about 1 ft/sec. The algae was alowed to stir for 1 minute alowing the flow patern to 
develop. After 1 minute the appropriate amount of the primary flocculant (iron nitrate or 
alum) was added with a pipete to a point 2/3 from the center of the beaker and the 
solution was alowed to stir for 15 seconds. After 15 seconds, the appropriate amount of 
the secondary flocculant (chitosan or gelatin) was added at the same location.  The 
solution was alowed to stir for an additional minute. After 1 minute an image was taken 
with the Nikon D40x on the rail and visual observations were recorded.  For runs 
containing celulose, the appropriate amount of celulose was first added to the algae 
sample and alowed to stir on the stir plate on seting 10 until the celulose was evenly 
dispersed. 
 After the image was taken, as quickly as posible a smal volume of the 
flocculated algae was taken from the sample using a disposable plastic pipete. The end 
of the pipete was cut to enlarge the opening to minimize disruption of the algae flocs. 
The sample was transfered to the center of a glas microscope slide and was covered 
with a plastic cover slip for further analysis. Figure 3-5 is an image of a typical beaker 
test run one minute after the addition of the secondary flocculant. After the image and 
microscope slide samples were taken, the stir plate was switched off and the sample was 
alowed to setle for 15 minutes.  After setling, a second image of the flocculated and 
 30 
setled sample was taken.  Figure 3-6 is an image of a typical beaker test run after 15 
minutes of setling. 
 31 
 
Table 3-1: Flocculation beaker test run chemistries. 
 32 
 
 
 
 
Figure 3-5: Typical flocculation beaker test 1 minute after addition of secondary 
flocculant.  Corresponds to iron nitrate and chitosan concentrations of 50 ppm and 
1.25 ppm respectively. 
 33 
 
 
 
 
 
Figure 3-6: Typical flocculation beaker test after 15 minutes of setling. 
Corresponds to iron nitrate and chitosan concentrations of 50 ppm and 1.25 ppm 
respectively. 
 34 
3.5 Microscope Slide Photographs 
 As was described in section 3.4, algae samples were extracted from stired 
solutions in beakers and placed on microscope slides. The samples from the flocculation 
beaker tests summarized in Table 3-1 presented in section 3.4 were analyzed by a 
microscope to determine the average floc size for each chemistry. Each slide was imaged 
under a microscope with the Nikon D40x digital camera.  Slides were mounted on a 
Macromaster CK microscope manufactured by Fisher Scientific. The right eyepiece was 
fited with a microscope camera adapter system. The telescoping portion of the camera 
adapter was set to its longest position to reduce vigneting.  The adjustable eyepieces 
were set as far apart as possible for al images.  The camera was connected to a video 
monitor. 
 Ten images at 10 times (10x) magnification (low mag) as wel as ten at 40 times 
(40x) magnification (high mag) were taken of each microscope slide sample. For each 
image the camera was operated in manual mode.  The shutter speed for al low mag 
images was set to 130 sec
-1
 and 13 sec
-1
 for al high mag images. Scale was established 
by imaging a 2 m ruler with 100 ?m major divisions and 10 ?m minor divisions 
mounted on a microscope slide. The focus was kept in the same position while imaging 
the algae samples to maintain the same scale. Images of a USAF 1951 resolution target 
were also taken.  Figures 3-7 and 3-8 are images of a typical floc under low and high 
magnification with the coresponding scale images of the ruler and resolution target, the 
iron nitrate and chitosan concentrations for these samples were 50 ppm and 1.25 ppm 
respectively. 
 35 
 Average floc area was determined quantitatively using a blob analysis technique 
similar to that used by Emerson et al. [2006].  ImageJ, shareware developed at the 
National Institutes of Health, was used for blob analysis. The average number of pixels 
of 30 distance measurements on images of the ruler was used to establish a pixels/?m 
scale for the algae samples.  For each algae sample image, the area of each floc (A
floc
) 
was traced with the ImageJ polygon tool and the area in ?m
2
 was recorded. Only flocs 
whose entire area was visible in an image were measured.  A floc thicknes of 10 ?m 
(t
floc
) was asumed to calculate an average equivalent diameter of the flocs.  Visual 
observation of many flocs under the microscope showed that most flocs were about two 
algae cels deep.  Based on an average chlorela cel diameter of 5 ?m [Stead et al., 
1995], two cels deep would indicate a floc depth of about 10 ?m. The area reported by 
ImageJ and the asumed thicknes were used to calculate a floc volume.  This volume 
was converted to an equivalent diameter of a sphere with the same volume (D
floc
) with 
Equation 3-3.  Refer to Appendix A for a detailed description of the blob analysis 
procedure. 
 
 
3
4
2
!"
"=
flocfl
floc
tA
D 
 
 For runs containing celulose, the area of celulose fibers atached to algae flocs 
was measured. A floc to fiber ratio was determined as the area of algae flocs atached to 
a celulose fiber, divided by the area of the celulose fiber.  Figure 3-9 is an image of 
algae flocs atached to a celulose fiber corresponding to iron nitrate and chitosan 
Equation 3-3 
 36 
concentrations of 50 ppm and 1.25 ppm respectively and a celulose to algae mas ratio of 
1:1. 
 Samples of fresh algae and celulose with no added flocculants were prepared. 
Images at 100 times (100x) magnification and 40 times (40x) magnification with 
corresponding scale images were taken.  Figure 3-10 is an image of fresh algae and 
celulose along with a corresponding scale image at 100x. 
 37 
 
 
Figure 3-7: Low magnification (10x) microscope images of (a) typical floc, (b) 
ruler scale, and (c) resolution target.  Corresponds to iron nitrate and chitosan 
concentrations of 50 ppm and 1.25 ppm respectively. 
 38 
 
Figure 3-8:  High magnification (40x) microscope images of (a) typical floc, 
(b) ruler scale, and (c) resolution target.  Corresponds to iron nitrate and 
chitosan concentrations of 50 ppm and 1.25 ppm respectively. 
 39 
Figure 3-9:  Low magnification (10x) microscope images of (a) typical floc 
with celulose, (b) ruler scale, and (c) resolution target.  Corresponds to iron 
nitrate and chitosan concentrations of 50 ppm and 1.25 ppm respectively and a 1:1 
celulose to algae mas ratio. 
 40 
Figure 3-10:  High magnification (100x) microscope images of (a) fresh 
algae, (b) clean celulose, and (c) ruler scale. 
 41 
3.6 Stationary Buble Facility 
 The stationary bubble facility was originaly constructed in 1998 by Andrew 
Davies and Duke [2000] to study the interaction of ink particles with air bubbles. The 
stationary bubble facility was employed in this work to study the interaction of algal cels 
with stationary bubbles suspended on the tip of a needle in a quiescent fluid.  Imaging 
algae absorption to stationary bubbles removes many of the dificulties asociated with 
imaging absorption to a moving bubble in a flow field. 
 The stationary bubble facility consists of a luer lock needle connected to a check 
valve. The check valve is atached via ? inch stainles stel tubing and soft-wal tygon 
tubing to a 50 mL Hamilton gas tight syringe. The needle is placed in a clear hexagonal 
vesel containing a quiescent fluid. Air is introduced into the line slowly by hand until an 
air bubble is suspended on the tip of the needle. Algal cels or flocs are injected from a 
plastic disposable pipete whose end has ben widened to minimize disturbing flocs. The 
pipete is directed such that the injection strikes the top surface of the stationary bubble. 
The injection event is recorded with a high-sped CD video camera with a high 
magnification lens. Figure 3-11 is a diagram of the stationary bubble facility and Figure 
3-12 is an image of the facility.  Figure 3-13 is a typical frame captured by the CD 
camera depicting a suspended bubble. Prior to each test run, a few frames of a ruler were 
captured in the field of view to establish scale. 
 Algal cel and bubble interactions in water were studied over a pH range of 2?10 
in 0.5 increments.  Algal cel and bubble interactions in water were studied over a 
temperature range of 10-50 ?C in 10 ?C increments. Interaction betwen air bubbles and 
algae that had been flocculated were observed. Based on the 33 chemistries evaluated in 
 42 
the beaker tests described in section 3.4, seven floculation chemistries were selected for 
further study. These seven chemistries are presented in Table 3-2. Each run in Table 3-2 
was conducted at pH 7 and room temperature of 22 ?C. The pH was adjusted with 0.05 
M HCl and NaOH.  For those runs containing celulose, the appropriate amount of 
celulose was first dispersed in the hexagonal vesel containing the test solution via a 1 
inch magnetic stir bar on a stir plate at the highest seting. 
 43 
 
Figure 3-11: Schematic of the stationary buble facility. Adapted from Ham 
[2004]. 
 44 
 
Figure 3-12: Stationary buble facility. 
 45 
 
Figure 3-13: Typical frame from the stationary buble facility. 
 46 
Table 3-2: Chemistries tested in the buble facilities. 
 
 47 
3.7 Suspended Buble Facility 
 Rossi [1998] constructed the suspended bubble facility in 1998 and it has been 
used extensively for study of ink interactions with bubbles. Bubbles are suspended in a 
vertical tube by a controled flow of fluid. The viewing tube is constructed of Plexiglas 
to alow visualization of the proces. The facility is about 2.75 m tal and can hold 7.5 L 
of fluid. Figure 3-14 is a diagram of the suspended bubble facility and Figure 3-15 is an 
image of the facility. 
 The Plexiglas viewing tube is secured with flanges alowing interchange of 
varying diameter viewing tubes. A 1/8 horsepower Fasco magnetic drive pump rated at 
3000 RPM drives the fluid in the loop.  It can deliver up to 53 LPM under 0.93 m of 
head. Cole-Palmer rotometers are used to monitor the flowrate through the system. They 
are polysolfanone direct-reading-in-line rotometers with 316 stel floats. There are thre 
flow meters in paralel capable of measuring 0.379-3.79 LPM, 0.758-7.58 LPM, and 
7.58-75.8 LPM. Air is injected via a compresed air source at the bottom of the Plexiglas 
viewing chamber through a luer lock needle. The size of the bubbles is afected by the 
flow rate of air introduced to the system and the diameter of the needle orifice. Bubbles 
are prevented from reaching the pump by traveling into a holding chamber located above 
the viewing chamber.  There are several vertical 0.64 cm tubes at the bottom of the 
holding chamber to reduce turbulence of the flow entering the column. Fully developed 
flow is achieved at a distance of about 0.3 m downstream of the entrance, calculated by 
Davies and Duke [2000].  Images are recorded 43 cm above the air injection point to 
ensure fully developed pipe flow. 
 48 
 For each test run in the suspended bubble facility, 7.5 L of algae from Tank A at a 
concentration of 150 ppm was stired in a stainles stel bucket. The pH was adjusted to 
7 with 0.05 M HCl and NaOH.  The seven chemistries in Table 3-2 were tested in the 
facility.  The appropriate chemicals were added to the stock algae in the same manner 
described in the beaker test experiments in section 3.4.  The solution was quickly 
transfered from the bucket to the flow loop. Video was taken during startup (while the 
facility was being filed) of each run at 500 FPS with the high-speed camera. After the 
facility was loaded with the test solution the pump was turned on and bubbles were 
injected through the needle.  The bubbles were suspended within the camera?s field of 
view by adjusting the flow control valves. Video of the proces was recorded with the 
high-speed camera at 500 FPS. 
 49 
 
Figure 3-14: Schematic of the suspended buble facility [Ham, 2004]. 
 50 
 
Figure 3-15: Suspended buble facility. 
 
 51 
3.8 Electrochemical Flotation Cel 
 The diameter of bubbles in the flotation proces is a key parameter; typicaly 
smaler bubbles float contaminants more efectively and eficiently [Casel et al., 1975]. 
The flotation scenario for large bubbles in induced air flotation is often several flocs or 
particles atached to a single bubble; the scenario for smal bubbles in DAF flotation is 
several smal bubbles atached to a single floc.  It is very dificult to produce bubbles 
smaler than about 750 ?m utilizing induced air flotation methods.  Therefore, to 
determine the efect of smaler bubbles on algae flotation, we have used a smal 
electrochemical flotation cel (EFC) that generates smal hydrogen gas bubbles for 
flotation. The size of the bubbles can be controlled by varying the current supplied to the 
fuel cel. The EFC was developed by Dr. Ron Putt at Auburn University [Putt, 2007]. 
 The EFC consists of a flotation chamber and a collection basin, supported by 4 
plastic legs.  The flotation chamber is filed with algae to be separated.  Hydrogen 
bubbles produced in the flotation chamber rise to the surface carying adsorbed algae and 
form an algal foam. Algal foam builds on the surface of the flotation chamber and spils 
into the collection basin thus separating the algae from the water.  Figure 3-16 is a 
schematic of the EFC and Figure 3-17 is an image of the cel.  
A hydrogen atmosphere is created under the flotation chamber by a bench scale 
fuel cel.  Smaler hydrogen bubbles are produced in the flotation chamber by a high 
carbon surface area Nafion membrane with a 1% platinum catalyst.  The hydrogen gas 
produced by the fuel cel is pumped under the flotation chamber where it contacts the 
anode of the Nafion membrane. The hydrogen gas is oxidized at the anode and protons 
pases through the membrane due to a voltage diference generated by connecting a D 
 52 
batery across the membrane.  The protons are reduced at the cathode and form smal 
hydrogen bubbles that rise through the flotation chamber.  The D batery is atached to 
screws in contact with the anode and cathode via aligator clips. 
 The seven chemistries listed in Table 3-2 were studied in the EFC.  Each 
condition was tested twice.  Stock algae was supplied from Tank A and stired 
throughout the experiment. The pH of the algae stock was adjusted to 7 with 0.05 M HCl 
and NaOH. Prior to each run, the fuel cel was loaded with distiled water and alowed to 
operate for 10 minutes to build up a hydrogen atmosphere under the facility.  The fuel 
cel was supplied with 2 V of power, corresponding to 0.5 A of current through the cel. 
For each run, 650 mL of algae was transfered to a 2 L glas beaker and stired with a 1 
inch octagonal Fisher stir bar on the Fisher stir plate on seting 6.  The appropriate 
amount of flocculants for each run were added in the same manner described in section 
3.4 and the algae was transfered to the EFC. The D cel batery was connected and the 
EFC was operated for 5 minutes. 
 After 1 minute had elapsed, video recorded at 500 FPS with the high-speed 
camera was taken with the camera focused on a point 2 cm from the front of the vesel 
and 2 cm from the top of the water level. A few frames of a ruler were taken before each 
run to establish scale. A second video was recorded after 4 minutes had elapsed, with the 
camera focused on a point 2 cm from the front of the vesel and at the water line. Images 
of the foam were taken throughout the experiment. Figure 3-18 is an image of the foam 
generated in a typical run, corresponding to iron nitrate and chitosan concentrations of 50 
ppm and 1.25 ppm respectively. 
 53 
 After 5 minutes had elapsed, the D cel batery and the power to the fuel cel were 
disconnected. The foam was collected with a flat rectangular piece of glas that had been 
cut to just fit in the facility.  The foam was skimed with this piece of glas 8 times, 
twice from each side of the vesel, spiling the foam into the collection basin.  The 
collected foam was transfered to a 50 mL Pyrex beaker and the remaining water in the 
facility was transfered to a 2 L Pyrex beaker.  Algae pads were generated from the 
collected foam and the water remaining in the facility.  The procedure for constructing 
algae pads is described in section 3.10. 
 The EFC can be set up to produce either hydrogen or air bubbles.  Both gases 
were used to determine the efects of gas type on flotation. Four 1 inch aeration stones 
were placed in the bottom corners of the facility. Air was pumped to these stones with a 
standard smal aquarium pump. Air bubbles produced in the EFC are in the range of 1-4 
m while hydrogen bubbles are in the range of 25-500 ?m. Each condition in Table 3-2 
was tested with air as wel as hydrogen. 
 54 
 
Figure 3-16: Schematic of the EFC. 
 
 
Top View 
 
 55 
 
Figure 3-17: Electrochemical flotation cel. (a) Top view, (b) bottom view, and 
(c) front view. 
 56 
 
Figure 3-18: Algae foam generated in the EFC. Corresponds to iron 
nitrate and chitosan concentrations of 50 ppm and 1.25 ppm respectively. 
 57 
3.9 Denver D-12 Flotation Cel 
 The Denver D-12 flotation cel manufactured by Svedala Industries is a laboratory 
scale flotation device used to model industrial induced air flotation proceses.  The 
Denver Cel was used to evaluate the seven chemistries listed in Table 3-2. 
 An aluminum column situated on a cast iron base supports the Cel. The height of 
the column is adjusted via a suspended type mechanism fited with an enclosed anti-
friction spindle bearing stainles-stel shaft. The suspended type mechanism is raised or 
lowered with a hand crank located on the side of the cel. The mechanism wil lock in 
any position by releasing a spring-loaded pin.  This pin must be pulled while adjusting 
the height of the column. Air is pulled into the machine via cavitation producing bubbles 
for flotation. The volume of air is related to the RPM of the motor. There are 2 plastic 
fluidizers and 2 plastic impelor sizes. The smaler fluidizer and impelor were used for 
al experiments. The cel is driven by a ? horsepower, 900-1800 RPM, single phase, 60 
Hz, 115/230 V, TEFC bal bearing motor. The motor speed is controlled by a knob on 
the top of the column and monitored via a mounted tachometer. Test samples are loaded 
in a stainles-stel vesel mounted under the difuser. The machine is capable of testing 
samples of 250, 500, 1000, and 2000 g. The 250 g vesel was used for al experiments. 
Figure 3-19 is a schematic of the Denver Cel and Figure 3-20 is an image of the cel 
[Svedala, 1996]. 
 The procedure employed for the Denver Cel was similar to that of the EFC. For 
each run the 250 g stainles-stel vesel was used and the machine was operated at 900 
RPM for 5 minutes using the smaler impelor and fluidizer.  Algae pads described in 
section 3.10 were constructed from collected foam, and images of the foam produced 
 58 
were taken throughout the experiments. It was not possible to generate algae pads from 
the remaining water in the Denver Cel after each run due to blinding of the filter paper 
during filtration. 
 59 
 
Figure 3-19: Denver D-12 Flotation Cel [Svedala, 1996]. 
 60 
 
Figure 3-20: Denver D-12 Flotation Cel. 
 61 
3.10 Algae Pads 
 Algae pads were constructed from samples generated in the EFC and the Denver 
D-12 Flotation Cel to determine the mas of algae in the foam layer and thus the 
flotation eficiency.  The method involves producing dried algae cakes on filter paper 
substrates from the collected foam as wel as unfloated algae in the clarified water of the 
flotation runs. 
 For each flotation run, 2 pieces of filter paper were dried in an oven at 65 ?C for 4 
hours to remove moisture.  The filter paper was removed from the oven and the dry 
weight was imediately recorded. The filter paper was transfered to a B?chner funnel 
atached to a vacuum pump.  The filter paper was weted with distiled water and the 
vacuum pump was turned on to provide suction to the funnel.  Algae samples from 
flotation runs were filtered through the funnel producing an algae cake on the filter paper. 
Algae from collected foam was collected on 7 cm diameter filter paper. Unfloated algae 
from flotation runs was collected on 11 cm diameter filter paper.  Algae pads were 
removed from the filter and transfered to an oven at 65 ?C and alowed to dry for 24 
hours.  After 24 hours the algae pads were removed from the oven and imediately 
weighed. Images of each algae pad were taken after drying. Figure 3-21 is an image of 
algae pads produced from algae collected from the foam and unfloated algae from a 
typical flotation run, corresponding to iron nitrate and chitosan concentrations of 50 ppm 
and 1.25 ppm respectively. 
 A quantitative measure of the flotation eficiency was calculated from the weight 
of the algae pads. The total mas of algae and added chemicals collected on a given pad 
was calculated as the diference in the final and initial weight of each algae pad. In order 
 62 
to acount for the weight that flocculants and celulose added to the algae pads, it was 
asumed that al added chemicals were bound to the algae.  The amount of added 
chemicals in the foam and the amount remaining bound to the unfloated algae was then 
proportional to the amount of algae floated.  For example, for a run containing 50 mg 
total of algae and 20 mg of additional chemicals total, if 40 mg of algae were collected in 
the foam and 10 mg remained unfloated, then it was asumed that 16 mg of added 
chemicals were present in the foam with the remaining 4 mg bound to the unfloated 
algae. Flotation eficiency was defined as the diference betwen the total mas of algae 
at the start of the run and the mas of unfloated algae, divided by the total mas of algae 
at the start of the run. 
 63 
 
Figure 3-21: Algae pads generated from a typical flotation run. (a) Algae from 
collected foam and (b) unfloated algae.  Corresponds to iron nitrate and chitosan 
concentrations of 50 ppm and 1.25 ppm respectively. 
 64 
3.11 Imaging System 
 The high-speed camera utilized in this study for observation of algae and bubble 
interactions was a Kodak SR-500 Motion Corder Analyzer. The camera system has been 
used for other studies in this laboratory [Emerson, 2006]. The camera can record at 100, 
250, 500, and 1000 frames per second for very high temporal resolution. The camera is 
capable of generating images with a maximum spatial resolution of 512 x 480 pixels. At 
a frame rate of 250 FPS, the camera records 1365 images for a run time of about 5.5 
seconds. Video recorded at 500 FPS has a run time of about 2.75 seconds. The camera 
generates 1365 time-stamped bitmap images.  The camera outputs directly to a PC for 
image procesing without losing resolution. The camera also outputs to a video monitor 
to observe what the camera is imaging. A Micro-Nikkor 60 m AF close-up lens was 
used for al images captured. The large aperture of this lens alows the use of standard 
lighting. The field of view ranges from 9.7 cm
2
 to 25 cm
2
. 
 The bitmap images produced by the camera are in an uncompresed format. 
These files are very large, and video generated from them can be very large, as much as 
700-800 MB.  In order to reduce file size, al bitmap images were converted to 
compresed JPEG files with Corel Photopaint 11. The JPEGS were then asembled into 
an AVI format movie using VideoMach software.  Refer to Appendix B for a more 
complete movie building protocol. 
 65 
CHAPTER 4 
 
RESULTS 
 A systematic study was performed to determine the flotation eficiency of 
chlorela algae flotation. In order to determine efective flocculation chemistries for use 
in flotation, flocculation beaker tests were performed and flocs formed were imaged with 
light microscopy.  The average equivalent diameter of the flocs formed for each 
chemistry was calculated with blob analysis. 
 Seven chemistries were chosen based on the flocculation beaker tests for study in 
the 3 available bubble facilities.  Algal cel and bubble interactions were imaged in the 
stationary bubble facility, which suspends a bubble on the tip of a needle in a quiescent 
fluid. Algal cel and bubble interactions were imaged in the suspended bubble facility, 
which suspends bubbles in a down flow of water. The flotation proces was imaged in 
the EFC during flotation runs for the seven selected chemistries. 
 Algae pads were constructed from the algae foam colected in the EFC and 
Denver D-12 Flotation Cel for each chemistry tested. The mas of algae was calculated 
and used to determine the flotation eficiency for each run. 
 66 
4.1 Flocculation Beaker Tests and Microscope Slide Results 
 The results of the flocculation beaker tests provided valuable qualitative data, 
while the corresponding microscope slides prepared from these tests provided useful 
quantitative data. Al flocculation beaker tests were performed at 22 ?C and pH 7. The 
33 chemistries listed in Table 3-1 were tested.  Based on the results of these 33 tests, 
seven chemistries that yielded the largest average floc size were chosen for additional 
testing in the bubble facilities. Figure 4-1 shows digital images of the flocculation beaker 
tests one minute into the test and Figure 4-2 after fiften minutes of setling for the seven 
chemistries listed in Table 3-2.  Table 4-1 acompanies these figures and provides the 
details of the chemical concentrations in each run. 
Blob analysis was performed on each of the microscope slides prepared during the 
flocculation beaker tests to determine the average equivalent diameter of the flocs 
generated in each run.  Figures 4-3 and 4-4 show the blob analysis results for the 
chemistries containing alum, gelatin, and celulose. Figures 4-5 and 4-6 show the blob 
analysis results for the chemistries containing iron nitrate, chitosan, and celulose. Tables 
4-2 through 4-5 acompany these figures and provide the details of the chemical 
concentrations in each run.  For those runs containing celulose, blob analysis was 
employed to determine the floc to fiber ratio, defined as the average floc area divided by 
the average fiber area.  Figure 4-7 presents the floc to fiber results for chemistries 
containing celulose.  Table 4-6 provides the details of the chemical concentrations in 
each run. Refer to Appendix D for statistical data. 
 67 
Figure 4-1: Flocculation beaker images while stiring. 
Table 4-1: Chemistries for Figure 4-1 and Figure 4-2. 
 68 
 
Figure 4-2: Flocculation beaker images after 15 minutes of setling. 
 69 
 
 
 
Figure 4-3: Microscope slide results. Average equivalent floc diameter for 
chemistries containing alum and gelatin. 
Table 4-2: Chemistries for Figure 4-3. 
 
 70 
Figure 4-4: Microscope slide results. Average equivalent floc diameter for 
chemistries containing alum, gelatin, and celulose. 
Table 4-3: Chemistries for Figure 4-4. 
 
 71 
 
Iron Nitrate and Chitosan Chemistries Average Equivalent Diameter
0
20
40
60
80
100
120
140
1 2 3 4 5 6 7 8 9 10 11 12 33
Run
A
v
e
r
a
g
e
 
E
q
u
i
v
a
l
e
n
t
 
D
i
a
m
e
t
e
r
 
(
u
m
)
Figure 4-5: Microscope slide results. Average equivalent floc diameter for 
chemistries containing iron nitrate and chitosan. 
Table 4-4: Chemistries for Figure 4-5. 
 72 
 
Figure 4-6: Microscope slide results. Average equivalent floc diameter for 
chemistries containing iron nitrate, chitosan, and celulose. 
Table 4-5: Chemistries for Figure 4-6. 
 
 73 
Figure 4-7: Floc to fiber ratio for celulose chemistries. 
Table 4-6: Chemistries for Figure 4-7. 
 
Table 4-6: Chemistries for Figure 4-7. 
 74 
4.2 Stationary Buble Facility Results 
 Interactions betwen algal cels and bubbles were imaged in the stationary bubble 
facility for the seven chemistries listed in Table 3-2.  Algae containing no added 
flocculants was also imaged under varying solution pH and temperature. Adsorption of 
algae to bubbles was not sen for any of the chemistries tested. This may be atributed to 
the generaly large size of bubbles produced in this facility as wel as other factors. 
Typical bubble diameters produced in the stationary bubble facility range from 0.75-1.25 
m. In al cases, algae approached and followed the curve of the bubble. Once the algae 
reached the bottom of the bubble, it did not adsorb and continued past the bubble and 
setled on the bottom of the vesel. 
 Images captured in the stationary bubble facility were converted to avi format 
movies.  Refer to Appendix C for a complete list of al movies generated and the 
corresponding conditions evaluated.  Figure 4-8 is a series of frames captured by the 
high-speed camera during a run in the stationary bubble facility.  The concentration of 
iron nitrate was 100 ppm and the concentration of chitosan was 1.25 ppm. The series of 
images tracks an algae floc as it travels down the bubble but does not adsorb. Figure 4-9 
presents individual frames from the seven chemistries tested showing no algae adsorption 
to bubbles. 
 75 
Figure 4-8: Series of frames from the stationary buble facility. Image 
sequence is from left to right starting from the top left. 
 76 
Figure 4-9: Individual frames from the stationary buble facility showing no 
algae adsorption. 
Table 4-7: Chemistries for Figure 4-9. 
 77 
4.3 Suspended Buble Facility Results 
 Interactions betwen algal cels and bubbles were imaged in the suspended bubble 
facility for the seven chemistries listed in Table 3-2. Frames were captured by the high-
speed camera for each run during normal operation (filed and flowing at steady state) as 
wel as while the facility was being loaded (filed with test algae).  During normal 
operation, algae was not sen to adsorb to bubbles for any chemistry tested.  However, 
algae was clearly sen to adsorb to bubbles during filing of the facility for al chemistries 
tested except the control run (algae and water only). The primary diference betwen the 
bubbles present during normal operation and those present during filing of the facility is 
size.  Typical bubble sizes observed during normal operation were in the range of 1-5 
m, while those observed during loading were in the range of 50-650 ?m.  
Figure 4-10 is a series of frames captured by the high-speed camera during filing 
of the facility for Run B, which corresponds to iron nitrate and chitosan concentrations of 
50 ppm and 1.25 ppm respectively and a celulose to algae mas ratio of 1:1. It is clearly 
sen that a large amount of algae has adsorbed to the bubble in the frame and is rising 
through the column. Several smaler bubbles are also completely covered in flocculated 
algae.  This network of algae and bubbles is actualy sinking in the column due to the 
weight of the algae. Several separate flocs are sen to adsorb to bubbles forming a larger 
network of flocculated algae on the bubble. Figure 4-11 presents individual frames from 
the same run highlighting the bubbles with adsorbed algae.  Algae was observed to 
adsorb to bubbles of similar sizes during loading of the facility for al chemistries tested 
except the control run, although no images were captured for the control run. Figure 4-12 
 78 
shows individual frames for the seven chemistries tested in Table 3-2 during filing of the 
facility showing algae adsorption to bubbles. 
Figure 4-13 is also a series of frames from Run B, except they were taken under 
normal operation of the facility. The bubble diameters in Figure 4-13 are in the range of 
5 m. Bubbles that are around 2 m or greater show a large departure from a spherical 
morphology. The bubbles in Figure 4-13 are sen to distort under the flow field during 
operation. Algae flocs are clearly sen traveling past the bubble without adsorbing. No 
algae was observed to adsorb to bubbles during normal operation of the suspended 
bubble facility for the chemistries tested. Figure 4-14 presents individual frames for the 
seven chemistries tested in Table 3-2 during normal operation of the facility, showing no 
algae adsorption. 
 79 
 
 
Figure 4-10: Suspended buble facility. Frames captured during loading of the 
facility for Run B. Image sequence is from left to right starting with the top left 
image. 
 80 
Figure 4-11: Individual frames from Figure 4-10 highlighting bubles with 
adsorbed algae. Bubbles atached to the same mas of algae are highlighted in 
yelow. 
 81 
Figure 4-12: Individual frames during filing of the suspended buble facility 
showing algae adsorption. 
Table 4-8: Chemistries for Figure 4-12 
 82 
Figure 4-13: Suspended buble facility. Frames captured during normal 
operation for Run B. Image sequence is from left to right starting with the top left 
image. 
 83 
 
Figure 4-14:  Individual frames during normal operation of the suspended 
buble facility showing no algae adsorption. 
Table 4-9: Chemistries for Figure 4-14 
 84 
4.4 Electrochemical Flotation Cel Results 
The seven chemistries listed in Table 3-2 were tested in the EFC utilizing both 
hydrogen and air as the gas for flotation. The foam generated in each run was collected 
and procesed into an algae pad in order to determine the flotation eficiency.  Robust 
foams were developed in each run except the control (algae and water only) when the test 
gas was hydrogen. There was a smal amount of foam for the control run when the test 
gas was hydrogen, however there was only a smal volume of foam present that was 
easily disturbed.  No foam was observed for any of the conditions tested while the test 
gas was air.  Figures 4-15 is an image of the foam generated for Run A right before 
shutdown for both hydrogen and air as the test gas, which corresponds to an iron nitrate 
and chitosan concentration of 50 ppm and 1.25 ppm respectively.  A robust foam is 
clearly sen for the hydrogen run, with smal hydrogen bubbles visible reaching the 
surface and breaking through the foam. There is no foam sen when the test gas is air. 
Very large bubbles can be sen on the surface without any appreciable algae. Figure 4-16 
presents images of the foam generated for the seven chemistries tested in the EFC for 
both air and hydrogen as the test gas. These images were taken right after shutdown of 
the facility. 
Figure 4-19 presents the flotation eficiency for the seven chemistries evaluated in 
the EFC. The flotation eficiency was defined as the diference betwen the total mas of 
algae present in each run and the mas of algae remaining unfloated, divided by the total 
mas of algae present in each run.  The flotation eficiency calculated for the runs 
utilizing air are very smal, in the range of 1%. The surface of the water was skimed in 
the same manner for the runs utilizing hydrogen, however most of the algae that was 
 85 
collected was a result of bubbles bursting on the surface of the water and causing smal 
amounts of water to spil into the collection chamber. Very litle algae was collected as a 
foam using the skimer for the air runs.  Table 4-11 presents the calculations used to 
determine the flotation eficiency. Figures 4-17 and 4-18 are a series of digital images of 
the algae pads generated from the EFC.  Pads for each of the seven runs tested are 
presented together for comparison of hydrogen and air as the test gas. In each case, the 
algae pads produced from the collected foam were a much darker gren when the test gas 
was hydrogen. The algae pads generated from the unfloated algae remaining in the test 
solution are a much lighter gren when the test gas is hydrogen, indicating a larger mas 
of algae was floated for the hydrogen runs. 
 86 
 
Figure 4-15: Algae foam generated in the EFC. Foam generated after 5 
minutes for Run A utilizing hydrogen (a) and air (b) as the flotation gas. 
Corresponds to iron nitrate and chitosan concentrations of 50 ppm and 1.25 
ppm respectively. 
 87 
Figure 4-16:  Images of foam generated in the EFC right after shutdown for 
hydrogen (top photo) and air (bottom photo) as the test gas. 
Table 4-10: Chemistries for Figures 4-16 through 4-19. 
 88 
Figure 4-17: Algae pads generated in the EFC. In each set of 4 pictures: Algae 
from foam using hydrogen (top left), unfloated algae remaining using hydrogen 
(bottom left), algae from foam using air (top right), and unfloated algae remaining 
using air (bottom right). Figure continued on the following page. 
 89 
Figure 4-18: Algae pads generated in the EFC. In each set of 4 pictures: Algae 
from foam using hydrogen (top left), unfloated algae remaining using hydrogen 
(bottom left), algae from foam using air (top right), and unfloated algae remaining 
using air (bottom right). Figure continued from the previous page. 
 90 
Table 4-11: Flotation eficiency calculations for the EFC. 
 91 
 
Figure 4-19: Flotation eficiency in the EFC. 
 92 
4.5 Denver D-12 Flotation Cel Results 
 The seven chemistries listed in Table 3-2 were tested in the Denver D-12 
Flotation Cel.  Algae pads were constructed from the foam collected in each run. 
Atempts to generate algae pads from the unfloated algae remaining in the tank were 
unsuccesful. The volume of liquid remaining in the tank after each run was about 1.4 L. 
The remaining volume was too large to filter using the vacuum pump and the B?chner 
funnel due to extensive blinding of the algae on the filter paper. Therefore the flotation 
eficiency could not be calculated for the Denver Cel runs. 
 Although floatation eficiency could not be calculated, it is clear from the digital 
photographs of the cel in operation that no appreciable foam was produced for any of the 
chemistries tested. Bubbles in the range of 2-6 m can be sen bursting on the surface, 
however there is litle or no algal foam acumulation at the surface of the water. Most of 
the algae in the pads generated from the collected foam was due to large bubbles bursting 
on the surface causing smal amounts of the test solution to spil into the collection 
chamber. Figure 4-20 is a series of images of the Denver Cel in operation at 5 minutes 
showing large bubbles bursting on the surface with litle or no algae foam produced. 
Images of the algae pads generated from the collected foam also indicate 
qualitatively that only a smal amount of algae is present on the pads. The algae pads are 
mostly white with a very light gren tinge indicating very litle algal mas. Figure 4-21 is 
a series of images of the algae pads generated from the collected foam from the Denver 
Cel. In comparison, algae pads produced from collected foam in the EFC when the test 
gas was hydrogen (shown in Figures 4-17 and 4-18) show a very dark gren color 
indicating a large mas of algae present. 
 93 
 
Figure 4-20: Denver D-12 Flotation Cel after operation for 5 minutes. 
Table 4-12: Chemistries for Figure 4-20 and 4-21. 
Table 4-12: Chemistries for Figure 4-20 and Figure 4-21 
 94 
 
Figure 4-21: Algae pads produced from the foam collected in the Denver 
Flotation Cel. 
 95 
CHAPTER 5 
DISCUSION OF RESULTS 
Image procesing techniques were utilized to calculate the average equivalent 
diameter of algae flocs produced with iron nitrate, alum, chitosan, gelatin, and celulose 
chemistries. Promising flocculation chemistries were evaluated in the stationary bubble 
facility, the suspended bubble facility, the EFC, and the Denver D-12 flotation cel. 
Algae pads were constructed from the flotation runs and the mas of algae 
removed from the water was calculated. The flotation eficiency of each of the selected 
chemistries was determined. 
 96 
5.1 Discusion of Algal Flocculation Results 
 The algae formed stable flocs when exposed to various flocculation chemistries. 
The size of the flocs varied with chemistry type, concentration, and the presence or 
absence of celulose. Figure 4-6 shows that the largest flocs were observed for Run 26, 
corresponding to an iron nitrate concentration of 50 ppm and a celulose to algae mas 
ratio of 10:1.  Run 26 yielded an average equivalent diameter of 187 ?m.  Figure 4-3 
shows that the smalest flocs were observed for Run 17, corresponding to alum and 
gelatin concentrations of 100 ppm and 6.25 ppm respectively. Run 17 yielded an average 
equivalent diameter of 24 ?m. In the absence of any chemicals, the algae did not form 
flocs. 
There are a few generalizations that can be drawn from the flocculation beaker 
tests and the corresponding microscope slides for each run.  Figures 4-3 through 4-6 
show that algae flocculated with iron nitrate and or chitosan produce larger flocs than 
algae flocculated with alum and or gelatin. The largest and smalest average diameters 
for iron nitrate and chitosan chemistries are 187 ?m (Run 26) and 43 ?m (Run 12) 
respectively. The largest and smalest average diameters for alum and gelatin chemistries 
are 57 ?m (Run 14) and 24 ?m (Run 17) respectively.  Iron nitrate and chitosan 
chemistries typicaly produce flocs with diameters about twice as large as alum and 
gelatin chemistries.  
 Secondly, Runs 13-17 in Figure 4-3 show that increasing the concentration of 
gelatin leads to a decrease in the average equivalent floc diameter of about 40% for algae 
treated with alum and gelatin.  Runs 18-21 in Figure 4-3 show that addition of alum 
beyond about 75 ppm neither increases nor decreases the average equivalent floc 
 97 
diameter for algae treated with alum and gelatin.  These results of alum and gelatin 
chemistries evaluated with the flocculation beaker tests provided guidelines for selecting 
the alum and gelatin chemistries evaluated with the various bubble facilities. 
Thirdly, Runs 1-5 in Figure 4-5 show that increasing the concentration of chitosan 
leads to a decrease in the average equivalent floc diameter of about 30% for algae treated 
with iron nitrate and chitosan.  Runs 6-10 in Figure 4-5 show that there is no clearly 
evident trend in average equivalent floc diameter for varying concentrations of iron 
nitrate for algae treated with iron nitrate and chitosan.  The results of iron nitrate and 
chitosan chemistries evaluated with the flocculation beaker tests provided guidelines for 
selecting the iron nitrate and chitosan chemistries evaluated with the various bubble 
facilities. 
 Lastly, the addition of celulose has a positive efect on the average equivalent 
floc diameter for iron nitrate and chitosan chemistries and litle efect on alum and gelatin 
chemistries. Addition of celulose increases the average equivalent diameter about 30% 
for iron nitrate and chitosan chemistries.  The largest average equivalent diameter was 
observed for an iron nitrate concentration of 50 ppm and a celulose to algae mas ratio of 
10:1. Figure 4-7 shows that the floc to fiber ratio is about 6 for iron nitrate and chitosan 
chemistries, indicating the majority of the flocs are algae as opposed to celulose. A floc 
to fiber ratio of 6 corresponds to floc/fiber networks with algae occupying 85% of the 
total area with the remainder celulose. This suggests that chitosan can be replaced with 
celulose. From a procesing perspective, replacing chitosan with celulose would greatly 
reduce the cost asociated with flocculating the algae to be floated.  Also, if the algae 
meal remaining after lipid extraction is to be used as an animal fedstock, celulose 
 98 
would have to be added to the meal before it can be used as a fedstock [Putt, 2007]. 
Therefore addition of celulose at the flocculation step would not require a removal step 
further down the procesing line. 
 Flocs produced with iron nitrate and chitosan chemistries were very stable 
compared to those produced with alum and gelatin chemistries.  While stiring in the 
flocculation beaker tests, al flocs reached a maximum size within a few seconds of 
adding the secondary flocculant (chitosan or gelatin). Flocs that collided did not combine 
to form larger flocs but rather bounced of of each other and remained the same size. 
After 15 minutes of setling, flocs that were stired a second time returned to a similar 
size, indicating that flocs do not combine into larger flocs when setled on the bottom of 
the vesel. While pipeting samples from the flocculation beaker tests for the microscope 
slides, the iron nitrate and chitosan flocs stayed together while the alum and gelatin flocs 
were easily disturbed. During the foam collection step in the EFC, the alum and gelatin 
foams were easily disturbed by the glas skimer and were partialy re-dispersed in the 
water while the iron nitrate and chitosan foams remained stable during skiming. 
Qualitative asesments of the flocculation beaker tests do not always agre with 
quantitative data drawn from the corresponding microscope slide photographs. In several 
cases very large flocs of m scale diameter were observed in the beaker while the test 
solution was stiring. However, when samples from the same conditions were analyzed 
with blob analysis, the average floc diameters were smaler than 0.2 m in the largest 
case.  Comparison of flocculation beaker images for Runs 1 and 5 (Figure 5-1) is an 
example.  The images show that Run 5 exhibits larger flocs than Run 1 while the 
solutions were stiring.  However when samples were drawn from the beakers and 
 99 
analyzed under the microscope, Run 1 had an average equivalent diameter of 112 ?m and 
Run 5 had a diameter of 75 ?m. 
The discrepancy is atributed to the method in which the microscope slides were 
prepared. Each slide shows only a very smal representation of the test run and may not 
be representative of the actual average size.  Also, the flocs sampled may have been 
disturbed by the action of the pipete while collecting samples.  The pipete openings 
were widened to minimize this efect.  Lastly, the asumed depth of the flocs on the 
microscope slides of 10 microns has a large efect on the calculated average floc diameter 
and may be underestimated.  
 100 
 
Figure 5-1: Flocculation beaker test images for Run 5 (a) and Run 1 (b). 
Corresponds to iron nitrate and chitosan concentrations of 75 ppm and 0.5 ppm 
for Run 5 and 75 ppm and 2.5 ppm for Run 1 respectively. 
 101 
5.2 Discusion of Algal Cel Adsorption to Bubles 
 The results from the four bubble facilities utilized in this study suggest that algal 
cels and algae flocs do not adhere to large bubbles. For the purposes of this discussion a 
large bubble is considered to have a diameter of 0.75 m or greater.  No algal cel 
adsorption to bubbles was observed for any of the conditions tested in the stationary 
bubble facility. Bubbles produced in this facility are typicaly on the order of 1-2 m; it 
is very dificult to produce a bubble with a diameter smaler than 0.75 m in this facility. 
The pH variation study further supports this asesment.  No algal adsorption was 
observed for any pH condition tested in this facility, however other studies have had 
succes floating algae in a low pH environment without additional chemicals [Levin et 
al., 1961].  The injection technique used for the stationary bubble facility sets up an 
almost guaranted collision of particles with the bubble. This may support the idea that 
algal cels did not adsorb to the bubbles in this facility because of their large diameters. 
 The suspended bubble facility provides anecdotal evidence that algal cels can 
adhere to smaler bubbles, but not larger ones. Figure 5-2 shows algae flocs adsorbed to 
a bubble of 500 ?m diameter while a bubble of 3 m diameter in the same frame does 
not have any algae adsorbed to it.  Images captured by the high-speed camera during 
filing of the suspended bubble facility clearly show algae adsorption to bubbles 
introduced during loading. Most of the bubbles produced during filing are 25-700 ?m in 
diameter. Figure 4-12 clearly shows a large mas of algae atached to several bubbles. In 
some cases the volume of algae atached to a bubble was many times greater than the 
volume of the bubble.  In other cases, several bubbles were atached to the same algal 
mas. However, during normal operation of the facility, no algae was observed to adsorb 
 102 
to any bubbles.  The bubbles produced during normal operation are very large, in the 
range of 1-6 m. 
 The EFC provides concrete evidence that algal cels can adsorb to smal bubbles. 
When the test gas was hydrogen, al chemistries tested in this facility, including to some 
extent the control run, showed algal floc and algal cel adsorption to bubbles. The size of 
the bubbles produced is very smal, on the order of 50 ?m, and can be controlled by 
varying the amount of curent supplied to the fuel cel. Evidence suggests that it is the 
smal bubble size and not the composition of the gas that has a greater efect on algae 
adsorption.  When the test gas in the EFC was air, there was no algal cel adsorption 
observed for any chemistry tested.  However this is atributed to the size of the air 
bubbles produced by the aeration stones, which were in the range of 1-3 m.  The 
anecdotal evidence captured by the suspended bubble facility during filing supports this 
claim.  The seven chemistries tested in the suspended bubble facility were identical to 
those tested in the EFC with both air and hydrogen as the test gas.  No algal cel 
adsorption was observed in the suspended bubble facility during normal operation or in 
the EFC when air was the test gas.  However algal cel adhesion was observed in the 
suspended bubble facility during filing and in the EFC when the test gas was hydrogen, 
suggesting that the size of the bubbles is the main factor in determining the extent of algal 
cel adsorption. 
 Although production of algae pads from the water in the D-12 Denver Flotation 
Cel was unsuccesful, these tests further support the claim that algal cels do not adsorb 
wel to large bubbles.  Typical bubble sizes produced by the Denver Cel under the 
 103 
conditions tested were on the order of 1-3 m, and no algal cel adsorption was observed 
for any of the seven chemistries evaluated in the Denver Cel. 
 The flotation tests conducted in the EFC also provide an asesment of the 
flocculation strength of the iron nitrate/chitosan and alum/gelatin chemistries tested. 
Figure 4-19 clearly shows that in al cases, iron nitrate and chitosan chemistries lead to 
higher flotation eficiencies than alum and gelatin chemistries.  However, these higher 
flotation eficiencies may be related to the stability of the foam produced by each 
chemistry. Qualitative asesments of the water clarity and amount of foam produced for 
both chemistries suggested that either chemistry is efective for algae flotation. However, 
it was more dificult to collect foam generated with alum and gelatin chemistries than to 
collect foam generated with iron nitrate and chitosan chemistries. During collection, the 
foams generated with iron nitrate and chitosan chemistries remained intact even when 
disturbed by the glas skimer.  On the other hand, foams generated with alum and 
gelatin chemistries were easily disturbed by the glas skimer. A significant amount of 
algae left the foam and was resuspended in the test solution by the action of the skimer 
for the alum and gelatin chemistries.  A foam stabilizer may improve the flotation 
eficiency of alum and gelatin chemistries.  The stability of the foam layer and the 
incorporation of algae into it appears to be related to the primary flocculant (iron nitrate 
or alum), while the secondary flocculant (chitosan or gelatin) has litle or no efect. In al 
cases that iron nitrate was the primary flocculant, the foam was very stable regardles if 
chitosan was present or not. In al cases that alum was the primary flocculant, the foam 
was relatively unstable regardles if gelatin was present or not. 
 104 
Flotation eficiency increases with increasing average equivalent floc diameter. 
Figure 5-3 shows that flotation eficiency increased by 50% when the average floc 
diameter increased from 40 ?m to 130 ?m.  Figure 5-3 reveals that iron nitrate and 
chitosan chemistries are superior to alum and gelatin chemistries for both flocculation 
and flotation efectivenes. The figure also shows that algae without flocculants does not 
float to any significant extent and flocculants are necesary in the flotation of algae even 
for smal diameter bubbles. 
 105 
 
Figure 5-2: Frame from the suspended buble facility during filing for 
Run A. The larger circled bubble is 3 m in diameter and shows no algal 
adsorption. The smaler circled bubble is 0.5 m in diameter and algal 
adsorption is highlighted with an arow. 
 106 
 
Figure 5-3:  Flotation eficiency evaluated in the EFC versus average 
equivalent floc diameter calculated from microscope samples of flocculation 
beaker tests. 
Table 5-1: Chemistries for Figure 5-3. 
 
 107 
5.3 Discusion of Buble Imaging Results 
 Although no algal cel adsorption to bubbles was observed in the stationary 
bubble facility, imaging of algal cel and bubble interactions in this facility were useful to 
demonstrate how single cels and algae flocs approach and interact with bubbles. Single 
cels and smal diameter flocs were shown to approach the bubble surface and follow 
streamlines around the bubbles and continue to the bottom of the vesel.  Larger flocs 
have a larger inertial force and interact with bubbles diferently. Larger flocs were sen 
to approach the bubble and strike the bubble surface. These flocs then bounced off of the 
bubble surface and continued to the bottom of the vesel.  Contact with the bubble is 
crucial to adsorption, and therefore larger flocs have a greater likelihood to adsorb to 
bubbles. 
 The studies in the suspended bubble facility were succesful in imaging algal cel 
adsorption to bubble surface during filing of the facility.  No adsorption was observed 
during normal operation of the facility. Incorporation of smaler bubbles, on the order of 
25-800 microns, into the facility under normal operation is recommended.  Evidence 
discussed in section 5.2 suggests algae would succesfully adsorb to bubbles in this size 
range. 
 Imaging of algal cel adsorption to bubble surfaces in the EFC was succesful. 
Images captured at the water air interface show that some bubbles with adsorbed algae 
approaching the surface are forced back down into the test solution. This suggests that 
flow paterns in the cel are important and the structural design of the flotation vesel 
should be considered in detail. 
 
 108 
5.4 Recommendations 
 Imaging of algal cel and bubble interactions provides a useful method for 
studying algal flotation.  Videos offer a quick procedure to determine the viability of 
diferent flocculation chemistries to float algae. Visualization of the entire procedure and 
not just the end results offers insights into how to improve the flotation proces. Future 
work may employ blob analysis on individual frames to determine the volume of algae 
atached to a bubble. 
 The results of this work indicate that iron nitrate and chitosan are efective 
flocculants that improve algae flotation.  Future work to determine concentrations of 
these chemicals that further enhance flotation is recommended.  Other flocculants are 
available that may be efective for algal flotation that should be evaluated, such as 
polyacrylamides. 
 Results from the bubble facilities indicate that smal bubbles in the range of 20-
750 ?m are more efective for algal flotation.  The ability to produce bubbles of these 
sizes in the stationary and suspended bubble facilities are needed to validate this 
asertion.  A smaler needle capable of producing bubbles in this size range should be 
used in the stationary bubble facility. Smaler bubbles may be produced in the suspended 
bubble facility with a smaler needle or through electrochemical methods.  Preliminary 
findings suggest that hydrogen and air are both efective for algal flotation, however the 
ability to produce bubbles of the same size in al of the bubble facilities are needed for 
concrete evidence.  Based on this asertion, DAF is more promising than induced air 
flotation for the flotation of algae because DAF generates bubbles in the 20-750 ?m 
range. 
 109 
CHAPTER 6 
CONCLUSIONS 
Flocculation beaker tests were used to determine the average equivalent diameter 
of algae flocs.  Iron nitrate and chitosan chemistries as wel as alum and gelatin 
chemistries were evaluated. Iron nitrate and chitosan chemistries were found to produce 
larger and more stable algae flocs than alum and gelatin chemistries.  The addition of 
celulose can replace the need for chitosan when iron nitrate is the primary flocculant. 
Samples of the flocculation beaker tests were mounted on microscope slides and imaged. 
Image procesing techniques were used to determine the average equivalent diameter of 
the flocs. The largest average equivalent diameter of 187 ?m was observed for an iron 
nitrate concentration of 50 ppm and a celulose to algae mas ratio of 10:1. The smalest 
average equivalent diameter of 24 ?m was observed for alum and gelatin concentrations 
of 100 ppm and 6.25 ppm respectively.  Increasing the concentration of the secondary 
flocculant (chitosan or gelatin) beyond a few ppm decreased the average equivalent 
diameter. Iron nitrate, alum, chitosan, and gelatin concentrations of 50 ppm, 100 ppm, 
1.25 ppm, and 3.75 ppm respectively were selected for testing in the bubble facilities. 
Observations in the stationary bubble facility were unsuccesful in imaging algal 
cel adsorption to bubbles. They were useful to visualize how individual cels and larger 
flocs interact diferently with bubble surfaces in a quiescent fluid. 
 110 
Adsorption of algal cels to bubble surfaces was succesfully imaged in the 
suspended bubble facility while the apparatus was being filed, however no adsorption 
was observed during normal operation of the facility.  Future studies may include 
alteration of the suspended bubble facility to alow production of bubbles in the 0.1-0.7 
m diameter range. Algal flocs were sen to adsorb more easily to bubbles in this size 
range. 
Algae pads were constructed from collected algae separated in the EFC.  The 
facility was operated using both hydrogen and air as the flotation gas. The mas of algae 
on the pads was used to quantify flotation eficiency. Algal foams generated with iron 
nitrate and chitosan chemistries are more stable than those generated with alum and 
gelatin chemistries.  Iron nitrate and chitosan concentrations of 50 ppm and 1.25 ppm 
respectively produced the highest flotation eficiency of 83%.  The lowest flotation 
eficiency of 29% was observed for alum and gelatin concentrations of 100 ppm and 3.75 
ppm respectively.  Flotation eficiency increases with increasing average equivalent 
diameter. 
Determination of flotation eficiency in the Denver D-12 flotation cel was not 
possible due to blinding of the filter paper while filtering the clarified water. Algae pads 
were constructed from the collected foam.  The Denver Cel provided evidence that 
supports the claim that algae adsorb to smaler bubbles more easily than larger bubbles. 
Future studies may include the incorporation of an electrochemical cel within the 
suspended bubble facility in order to produce smaler bubbles for flotation.  A smaler 
needle fabricated from glas capilary tubing may be used in the stationary bubble facility 
to image algal cel and bubble interactions in a quiescent fluid for bubble sizes in the 
 111 
range of a few hundred ?m.  Imaging techniques may be used to calculate the mas of 
algae atached to an imaged bubble with blob analysis.  Through these techniques a 
clearer picture of the algal flotation proces may be achieved. 
 
 
 112 
BIBLIOGRAPHY 
 
Alonso, D.L., Belarbi, E.H., Fern?ndez-Sevila, J.M., Rodr?guez-Ruiz, J. and Grima, 
  E.M. [2000] Acyl Lipid Composition Variation Related to Culture Age and 
  Nitrogen Concentration in Continuous Culture of the Microalga Phaeodactylum 
 Tricornutum, Phytochemisty, 54, 461-471. 
 
Biermann, C.J. [1996] Handbook of Pulping and Papermaking. Academic Pres. 
 
Bilanovic, D. and Shelef, G. [1988] Flocculation of Microalgae with Cationic Polymers ? 
 Efects of Medium Salinity, Biomass, 17, 65-76. 
 
Casel, E.A., Kaufman, K.M. and Matijevi?. [1975] The Efects of Bubble Size on  
 Microflotation, Water Research, 9, 1017-1024. 
 
Chen, Y.M., Liu, J.C. and Ju, Y.H. [1998] Flotation Removal of Algae from Water, 
 Colloids and Surfaces B, 12, 49-55. 
 
Clift, R., Grace, J.R. and Weber, M.E. [2005] Bubbles, Drops, and Particles, Dover 
Publications, Mieola, NY. 
 
Davies, A. [2000] Visualization of Flexographic and Ofset Ink at Bubble Surfaces, 
  Masters Thesis, Auburn University. 
 
Davies, A. and Duke, S.R. [2000] Visualization of Flexographic and Ofset Ink at Bubble 
Surfaces, TAPI Recycling Symposium, Vol. I, Hyat Crystal City, Washington 
DC, March 5-8, 2000. 
 
Davies, A. and Duke, S.R. [2000] Visualization of Ink Removal Proceses, Paper Age, 
  July 2000. 
 
Davies, A., Rossi, L. and Duke, S.R. [2000] A Method for Visualization and 
  Measurement of Ink Adsorption Rates at Bubble Surfaces, Fundamentals and 
  Numerical Modeling of Unit Operations in the Forest Products Industries, 
 AICHE Forest Products Symposium, Series No. 324, Volume 96, 2000. 28-36. 
 
Emerson, Z.I. [2003] Visualization of Toner Ink and Adhesive Particles at Bubble 
Surfaces, Masters Thesis, Auburn University.
 
 
 113 
Emerson, Z.I., Bonometi, T., Krishnagopalan, G.A. and Duke, S.R. [2006] Visualization  
of Toner Ink Adsorption at Bubble Surfaces, TAPI Journal, 5(4). 
 
Ham, J. [2004] Enzyme Enhanced Deinking of Toner Inks, Masters Thesis, Auburn  
University. 
 
Han, X., Miao, X. and Wu, Q. [2006] High Quality Biodiesel Production from a 
icroalga Chlorela Protothecoides by Heterotrophic Growth in Fermenters, 
Journal of Biotechnology, 126(4), 499-507. 
 
Hung, M.T. and Liu, J.C. [2006] Microfiltration for Separation of Gren Algae from 
 Water, Colloids and Surfaces B, 51, 157-164. 
 
Jameson, G.J. [1999] Hydrophobicity and Floc Density in Induced-Air Flotation for 
  Water Treatment, Colloids and Surfaces A, 151, 269-281. 
 
Kivakaran, R. and Pilai, V.N.S. [2002] Flocculation of Algae Using Chitosan, Journal of 
 Applied Phycology, 14, 419-422. 
 
Kragh, A.M. and Langston, W.B. [1961] The Flocculation of Quartz and Other 
Suspensions with Gelatine, Journal of Colloid Science, 17(2), 101-123. 
 
Levin, G.V., Clendenning, J.R., Gibor, A. and Bogar, F.D. [1961] Harvesting of Algae by 
 Froth Flotation, Resources Research Inc. 
 
Liu, X., Xu, J. and We, Q. [2007] Large Scale Biodiesel Production from microalga 
  Chlorela Protothecoids Through Heterotrophic Cultivation in Bioreactors, 
  Biofuels and Environmental Biotechnology, Acepted Preprint. 
 
Liu, J.C., Chen, Y.M. and Ju, Y.H. [1999] Separation of Algal Cels from Water by 
 Column Flotation, Separation Science and Technology, 34(11), 2259-2272. 
 
Martin, T. and Britz, H. [2002] The New EcoCel Flotation for Primary and Secondary  
Stages. Voith Sulzer Paper Technology. 
 
Mayo, A.W. and Noike, T. [1994] Response of Mixed Cultures of Chlorela Vulgaris and  
Heterotrophic Bacteria to Variation of pH, Water Science and Technology, 30, 
285-294. 
 
Morita, M., Watanabe, Y., Okawa, T. and Saiki, H. [2001] Photosynthetic 
Productivity of Conical Helical Tubular Photobioreactors Incorporating Chlorela 
sp. Under Various Culture Medium Flow Conditions, Biotechnology and 
Bioengineering, 74, 136-144. 
 
 
 114 
Oh, H.M., Le, S.J., Park, M.H., Kim, H.S., Kim, H.C., Yoon, J.H., Kwon, G.S. and 
Yoon, B.D. [2001] Harvesting of Chlorela Vulgaris Using a Bioflocculant from 
Paenibacilus sp. AM49, Biotechnology Leters, 23, 1229-1234. 
 
Planas, M.R. [2002] Development of Techniques Based on Natural Polymers for the 
Recovery of Precious Metals, Disertation, Universitat Polit?cnica de Catalunya. 
 
Putt, R. [2007] Algae as a Biodiesel Fedstock: A Feasibility Asesment, Draft 
Submited to Center for Microfibrous Materials Manufacturing. 
 
Putt, R. [2007] Personal communications, Auburn University. 
 
Rodrigues, R.T. and Rubio, J. [2007] DAF-Disolved Air Flotation: Potential 
Applications in the Mining and Mineral Procesing Industry, International 
Journal of Mineral Procesing, 82, 1-13. 
 
Rossi, L. [1998] Auburn University Internship Report, University of Toulouse, France. 
 
Sheehan, J., Dunahay, T., Benemann, J. and Roesler, P. [1998] A Look Back at the U.S. 
  Department of Energy?s Aquatic Species Program ? Biodiesel from Algae, 
  NREL/TP-580-24190. 
 
Stead, A.D., Cotton, R.A., Ducket, J.G., Goode, J.A., Page, A.M. and Ford, T.W. [1995] 
The Use of Soft X Rays to Study the Ultrastructure of Living Biological Material, 
Journal of X-Ray Science and Technology, 5, 52-64. 
 
Svedala Corporation [1996] Denver D-12 Laboratory Flotation Cel User?s Manual. 
 
Yao-de, Y. and Jameson, G.J. [2003] Application of the Jameson Cel Technology for 
Algae and Phosphorus Removal from Maturation Ponds, International Journal of 
Mineral Procesing, 73, 23-28. 
 115 
APENDIX A 
BLOB ANALYSIS PROCEDURE 
 The following steps outline the procedure to calculate a specified area in an 
image, which was used to determine the average equivalent floc diameter. 
Scale Calibration 
1. Open ImageJ 
2. Click File > Open and locate the appropriate image of the ruler on a microscope 
slide to set the scale 
3. Click on the ?Straight line selections? button on the toolbar 
4. Draw a straight of known distance on the image betwen two ruler tick marks 
5. Click Analyze > Set Scale 
6. The distance in pixels of the line is displayed in the ?Distance in Pixels? box 
7. Type the distance of the line in the ?Known Distance? box 
8. Set the unit of the distance of the known line in the ?Unit of Length? box 
9. Check the ?Global? checkbox and click the ok button 
Floc Area Measurement 
1. Close the image of the ruler on the microscope slide 
2. Click File > Open and locate the first microscope slide image to be analyzed 
3. Click the ?Polygon selections? button on the toolbar 
 116 
4. For each floc to be measured, click on the outside edge of the floc and trace the 
area of the floc 
5. When the entire floc area is highlighted, double click the mouse to close the 
polygon 
6. Click Analyze > Measure (alternatively pres ctrl-m) 
7. The area of the floc in the specified units is displayed in another window 
8. Repeat the procedure for al flocs to be measured 
 
 117 
APENDIX B 
MOVIE BUILDING PROCEDURE 
 The following steps outline the procedure to convert frames captured by the 
Kodak MotionCorder High-Speed Camera into a digital movie in avi format. 
File Batch Change 
1. Open Corel PhotoPaint 11 
2. Click File > Batch change on the toolbar 
3. In the new indow, click Add file 
4. Navigate to the folder the images are stored in and click on the first file 
(f_000001.bmp) in the directory and then scroll down to the 800
th
 file (f_000800) 
and click it while holding the SHIFT key. 
5. Confirm that ?Save as new file? is highlighted 
6. Set the save to folder as the destination where the converted files are to be saved 
7. Set the ?save as? type as JPG 
8. Click PLAY 
9. Corel wil open and close a new indow for each image, after al images have 
been converted, open Windows Explorer 
10. Copy and run the batch file change script (change.bat) to the directory where the 
converted images were saved. The change.bat script renames the JPG files for 
movie building. 
 118 
Movie Building 
1. Open VideoMach 
2. Click on the ?Open video and audio files? folder icon 
3. In the new indow, locate the directory where the JPG images are saved 
4. Select the entire range of images to be compiled into a movie and click open 
5. Click File > Save 
6. In the new indow, set the ?Output Mode? dropbox to video only 
7. Under the ?Write VIDEO and AUDIO to this file? dropbox, specify the 
directory where the movie is to be saved 
8. Click the video tab at the top of the window 
9. Click the resize button 
10. In the new indow uncheck the ?automatic box? and change the ?Make 
resolution divisible by? to 4 and click the ok button 
11. Click the rotate button 
12. Click on the ?90 degres right? button and click the ok button 
13. Under the ?Frame Rate (fps)? section, uncheck the ?automatic? box 
14. Enter the frame rate in frames per second that the images were captured at 
15. In the dropbox, select ?Kep original number of frames? 
16. Click save 
17. When the movie building is complete (1-10 minutes) the movie is created in 
the specified directory 
 
 119 
APENDIX C 
LIST OF VIDEOS 
 
 The following is a list of al movies captured by the high-speed camera.  File 
names that include ?RepX? refer to the replicate number. For replicated videos, 3 Reps 
were performed for each condition listed except for the EFC hydrogen Runs which have 
2 Reps for each condition listed. 
 120 
 
Table C-1: List of Videos 
 121 
APENDIX D 
FLOC SIZE ANALYSIS 
  
Table D-1: Flocculation beaker test calculations