Up-scaling of algae cultivation

Up-scaling of algae cultivation

Emil Axelsson

Sustainable Process Engineering, masters level 2016

Luleå University of Technology

Department of Civil, Environmental and Natural Resources Engineering

Master thesis on

Up-scaling of algae cultivation

Performed at SP Technical Research institute of Sweden by

Emil Axelsson

Civil engineering in sustainable process engineering, Luleå University of Technology

2016-10-17

Main supervisor - Niklas Strömberg

Additional supervisors - Susanne Ekendahl

, Johan Engelbrektsson

and Mathias Bark

Examiner - Ulrika Rova

Acknowledgments

I would like to express my gratitude to my supervisors at SP. I am very grateful for your guidance through this project.

Thank you Mathias, you were not actually my supervisor in the beginning, but have helped me to understand how algae behave and have helped me examine algae in the microscope for many hours.

Thank you Johan, without your technical knowledge and eye for details it would not have been possible to go through with this project. I am very thankful for that you jumped in and helped me to search for the source of the malfunction of the control system. I think we understood each other and have an interested in common topics.

Thank you Susanne, you have made time for me when I needed it and answered my questions about everything, and made me understand what to do. You helped me in the lab with everything I needed help with.

Niklas, a special thanks to you, who I have worked with the most. You have magically made time available for me and with haste solved the issues that I could not solve myself. I think we are more alike than any of us understood.

I have really enjoyed working with you all!

This work is a part of the MICROBiorefine project which aims to integrate microbial biomass production and refinement for sustainable fuels, chemicals and feed.

Emil Axelsson, Borås, 2016-09-22

Abstract

Microalgae are one of the oldest types of lifeforms on this planet and dead algae are one source for the oil that we extract from the ground. This oil has a major part in the technology advances of humanity, to levels unimaginable not long ago. Unfortunately, this oil is one of the major reasons for the global warming and other environmental issues caused by humans. Therefore, much effort is made on new technologies to decrease the use of fossil oil and other fossil material in favor for so called renewable sources. In this work focus is on production of biomass that can be used for processing to other bulk materials, mainly chemicals. This is also a highly potential market, the amount of materials derived from fossil sources are at least 422 million metric tons per year. The issue though is that the production costs for algae are still fairly high and can’t compete with the market price of fossil raw materials.

Two algae species, Scenedesmus obliquus and Coelastrella sp., were cultivated in 6 pilot size ponds (500 L) and the results were compared to a lab experiment (0.5 L). The lab experiment was earlier performed by the author’s supervisors with the same species. The algae in the ponds were cultivated outdoor with flue gas in semi-closed ponds and the resulting biomass was allowed to sediment spontaneously. Scenedesmus obliquus was successfully cultivated in the pilot, but the system was not suitable for cultivation of Coelastrella sp.

The main aim of this work was to evaluate if it is possible to predict the amount of biomass produced in the pilot cultivation based on the results from the previously performed lab cultivation. The conclusion based on the results in this work is that it not possible to predict the biomass production in the pilot based on lab experiments. The properties and behavior of different algae species can be very different in different systems, and the setups in this study differed too much.

However, the results indicate that the pilot system has a high efficiency and can maintain a monoculture outdoors for at least 18 days as well as that the supply of flue gas highly affects the growth of the algae Scenedesmus obliquus.

Sammanfattning

Mikroalger är en av de äldsta livsformerna på den här planeten. Döda alger är basen för bildandet av den olja som vi nu extraherar från marken, vilket har stor del i den oerhört stora teknologiska utveckling som hade varit otänkbar för inte så länge sedan. Dessvärre så är oljan även en orsak till global uppvärmning och andra miljörelaterade problem människan har orsakat. För att lösa detta problem så sker en stor utveckling av nya tekniker för att minska användningen av fossil olja och andra fossila material och istället använda så kallat förnyelsebara källor. I det här arbetet så är målet att producera alger som kan användas som råvara för andra material, främst bulkkemikalier. Det är en högst potentiell marknad, mängden material framställt från fossila råvaror är minst 422 miljoner ton per år. Problemet är att produktionskostnaden för alger fortfarande är relativt höga och inte kan konkurrera med marknadspriset för fossila råvaror.

Två algarter, Scenedesmus obliquus och Coelastrella sp., har odlats i 6 tankar av pilotstorlek (500 l) och resultatet jämfördes med ett labbexperiment (0,5 l). Labbexperimentet genomfördes av författarens handledare med samma arter. Algerna i tankarna odlades med rökgas täckta, men ej tätade, och kunde spontant sedimentera. Odlingen av Scenedesmus obliquus lyckades, men systemet var inte lämpad för odling av Coelastrella sp.

Huvudmålet med detta arbete var att utvärdera om det är möjligt att förutse mängden biomassa som kan produceras i pilotodlingen baserat på resultat från labbodlingen. Slutsatsen baserat på resultatet i detta arbete är att det inte är möjligt att förutspå resultatet i piloten utifrån

labbexperiment. Egenskaper och beteenden för olika algarter kan vara väldigt olika i olika system, och odlingspremisserna skiljer sig för mycket mellan systemen.

Resultatet indikerar också att pilotsystemet har en hög produktions effektivitet, kan bibehålla en

monokultur under 18 dagar utomhus och att mängden rökgas har en stor påverkan på tillväxten av

Scenedesmus obliquus.

Table of contents

1 Introduction ... 1

1.1 The concept of the algae farming technology at SP ... 1

1.2 The aim of this thesis ... 3

2 Literature review ... 4

2.1 Algae ... 4

2.1.1 Usage of algae today ... 4

2.1.2 Compounds of interest in the algae biomass ... 4

1.1 Cultivation systems ... 5

2.1.3 Up-scaling of cultivation systems ... 5

2.2 Cultivating environment ... 6

2.3 Algae grown in flue gas ... 6

2.3.1 Carbon dioxide ... 6

2.3.2 Nitrogen oxides ... 7

2.3.3 Sulfur oxides ... 7

2.3.4 Heavy metals ... 7

2.4 Separation techniques ... 7

2.4.1 Gravity Sedimentation ... 8

2.4.2 Flocculation ... 8

2.4.3 Flotation ... 9

2.4.4 Centrifugation ... 9

2.4.5 Filtration ... 9

2.4.6 Sun-drying and cross-flow air drying ... 9

2.4.7 Rotary drying, Spray drying and flash drying ... 9

2.4.8 The experiment ... 10

2.5 Economics, a short overview ... 10

3 Method ... 12

3.1 Introduction ... 12

3.1.1 Measured Variables ... 12

3.1.2 Measuring Methods ... 12

3.2 Lab Cultivation ... 13

3.2.1 Setup ... 13

3.3 Inoculum Cultivation ... 14

3.3.1 Setup ... 14

3.4 Pilot cultivation ... 15

3.4.1 General description of the cultivation principle ... 15

3.4.2 Setup ... 16

3.4.3 Sampling ... 17

3.4.4 Experiment ... 18

4 Results and discussion ... 19

4.1 Introduction ... 19

4.1.1 Pilot cultivation ... 19

4.2 Optical density ... 19

4.2.1 OD in Lab cultivation ... 19

4.2.2 OD in Pilot cultivations ... 20

4.3 pH/valve log and gas composition in the pilot ... 21

4.3.1 pH and gas valve log for the pilot ... 21

4.3.2 Flue gas composition ... 22

4.4 Nutrients ... 23

4.4.1 Lab cultivations ... 23

4.4.2 Scenedesmus obliquus pilot cultivations ... 24

4.4.3 Coelastrella sp. pilot cultivations ... 26

4.5 Microscopy counting ... 28

4.5.1 Lab cultivation, final sample ... 28

4.5.2 Scenedesmus obliquus pilot cultivations ... 29

4.5.3 Coelastrella sp. pilot cultivations ... 31

4.5.4 Sediment pond 1-6, pilot cultivations ... 32

4.6 Sun-drying and dry weight ... 33

4.6.1 Sun-drying, pilot cultivations ... 33

4.6.2 Dry weight of algae in lab and pilot cultivations ... 33

4.7 Energy value and photosynthetic efficiency ... 35

4.7.1 Energy value ... 35

4.7.2 Photosynthetic efficiency ... 35

5 Further discussion and conclusions ... 36

6 Recommendations ... 37

7 References ... 38

8 Appendix ... 41

Appendix A – Date and time for sampling ... 41

Appendix B – Solar radiation and outdoor temperature ... 41

1 Introduction

During the industrial age humanity has developed faster than ever before. However, mainly due to the intense use of natural resources, humanity now faces a number of serious global problems including: global warming [1], overfishing [2], deforestation [3], intense use of farmland [4], starvation [5] and unhealthy diets [6].

Because of the effect of global warming, the world leaders have decided that the use of fossil-raw materials needs to come to an end [1]. When fossil raw materials are discussed, the discussion is often about energy products like gasoline, diesel, coal and jet-fuel. These products are of course important, since they make up the majority of the consumption, but there are also a lot of chemicals that are produced from crude-oil. Today almost all bulk chemicals are produced from fossil-raw materials and about 10% of all crude-oil, or 422 million metric tons per year, is processed to chemical products [7,8]. Examples of materials that in general are derived from fossil materials are; plastics, synthetic rubber, synthetic fiber, paint, glue, solvents, explosives and pesticides. To conclude, it is a huge task to replace crude oil but there seems to be few alternatives left if we are to stop global warming.

Algae could be a part of a solution to all of the above mentioned problems. Algae biomass can be produced for food or food supplements [9], energy purposes [10] and industrial raw materials [11].

One of the main advantages of algae is that the production doesn’t need arable land [11].

1.1 The concept of the algae farming technology at SP

At SP Technical Research Institute of Sweden located in Borås, Sweden (SP) the research on

microalgae biomass production is focused on the use of flue gas and waste heat from factories. The research at SP focuses on both micro- and macro- algae, but this work only considers green

microalgae.

Figure 1 Simple illustration of the algae farming technology using the flue gas and excess heat of a suitable facility.

The algae are cultivated using flue gas and excess heat in semi-closed ponds, where the gas and heat

will be provided by a nearby production facility. The flue gas is mainly interesting as a carbon source

and the heat is used to optimize the cultivation temperature, mainly during the colder seasons. The

to sediment during the cultivation. The gas flow is controlled by pH; if it’s high enough the gas flow is turned on and vice versa. The gas flow is the only agitation source in the ponds.

When the cultivation is finished the gas flow stops, so that the suspended algae can sediment. When the sedimentation is sufficient the medium is removed from the cultivation pond and the sediment is sun-dried until a desirable dryness is obtained.

The cultivation concept has been tested and evaluated during three years using the flue gas and excess heat provided by the pulp- and paper-mill Nordic Paper in Bäckhammar. The main purpose of the experiments in Bäckhammar was to examine if the concept worked i.e. the combination of semi- closed cultivars, industry flue gas and heating with excess heat. [12, 13]

To further develop this technology SP has relocated the algae pilot to SP:s facility in Borås (SP Algae Research Facility) to have easier access to the pilot. To more efficiently study the cultivation of algae in flue gas SP has also built an indoor lab cultivation system (without sedimentation). The volume of the indoor lab cultivars is about 0.5 L where the pilot cultivars are 500 – 2300 L per pond. A

comparison of the two systems is found in Table 1.

Table 1 A comparison of the lab- and pilot-cultivations to indicate similarities and differences between the two cultivation systems.

Lab cultivation Pilot cultivation

Volume per pond 0.5 L 500-2 300 L

Flue gas supply Synthetic gas mixture Gas from conventional pellets burner for house heating

Flue gas flow Controlled by pH-set point Controlled by pH-set point

Temperature Regulated room temperature Roughly controlled by heating coils in the ground

Light supply LED-lamps Sun

Agitation Magnetic-stirrer and gas flow Gas flow Sampling during

cultivation The algae concentration in the cultivation system can be assumed to be homogeneous, but sampling reduces the cultivation volume

The algae concentration is by system design not homogeneous, with a cultivation region and a sedimentation region. Sampling barely affects the cultivation volume

The lab-scale is designed for optimal growth conditions with stirring and relative constant

temperature and light supply, whereas the pilot cultivation is designed as a prototype for industrial scale production with continuous sedimentation. The aim with the pilot cultivation is to produce dry algal biomass in a cheap and energy efficient way as well as in enough quantities for making material prototypes and testing. The main applications of the produced algae biomass are for further

processing to chemical bulk products.

1.2 The aim of this thesis

The overall aim was to investigate if it was possible to correlate the biomass production and composition at a pilot scale cultivation with corresponding lab scale cultivation and identify factors that can be used for the correlation. Such a correlation would facilitate predictions of the pilot scale biomass production based on lab scale experiments and thus speed up the development of future cultivation techniques for microalgae at scale.

Apart from the overall aim, the experiment was also carried out to evaluate the recently physically relocated pilot facility, including modified cultivars, and test if Coelastrella sp., in addition to the previously tested Scenedesmus obliquus, can be cultivated with the SP cultivation technique.

To achieve this, it was required to build an inoculum cultivation and new cultivation ponds, which were performed by the student and assisting staff at SP prior to the start of experiment.

2 Literature review

2.1 Algae

Algae are a group of eukaryotic organisms, where the first forms evolved about 1.5 billion years ago [14,15]. There are at least 40 000 known species but it is not known how many that exists, some guess to the hundreds of thousands. Algae can be single cell or multi cell (microscopic and macroscopic). Apart from being based on size, algae are often categorized into four groups; red, brown, diatoms and green. The main energy supply for algae is solar energy via photosynthesis, but some species can also utilize other carbon and energy sources like sugar [16]. This work will focus on green single cell algae, also called green microalgae. The two algae studied only use photosynthesis as the energy source.

2.1.1 Usage of algae today

Macro algae are commonly consumed as food in Japan, Korea and China [9]. Macro algae extractions are also used as addition for consistency purposes in food, paint, toothpaste and glue [11, 14].

Microalgae are today used mainly for exclusive products including health supplements, food,

cosmetics and pharmaceuticals [17,18]. It is also produced for supplements in animal feed. Numbers of the actual biomass production of microalgae are hard to estimate, but in 2004 the annual

production of microalgae was about 5 000 ton of dry biomass [18]. When SP has cultivated algae the energy density of dry algae was in the range of 18-23 kJ/g [13].

Hereafter in this work, microalgae or algae refer to green single cell algae.

2.1.2 Compounds of interest in the algae biomass

Dry algae biomass consists of three major groups; carbohydrates, proteins and lipids. The general compositions of these are 20-40% carbohydrates, 30-50% proteins and 8-15% lipids, but the fractions can be both higher and lower than this [19]. The composition varies between species, but can also differ between cultivations of the same specie, depending on the growth conditions. Table 2 shows an example of the diverse lipid content of algae, ranging from 2-75% (dry weight). [20]

Table 2 Example of the diverse lipid content in algae biomass with a high variation between algae species and between algae of the same specie. [20]

Marine and freshwater microalgae species Lipid content (% dry weight biomass)

Scenedesmus obliquus 11-55

Ankistrodesmus sp. 24-31

Botryococcus braunii 25-75

Chlorella pyrenoidosa 2.0

Dunaliella sp. 18-67

If the three major groups of chemical compounds are separated they can be processed to materials with a number of applications. Carbohydrates can be further processed to biofuels, bioplastics, commodity chemicals and food additives. The proteins can be used as animal feed, nutrient recycling (nitrogen and phosphorus), antibiotics and natural products. [21]

The component that in general gets the most attention in articles is the lipid fraction, mainly because of the possibility to produce biodiesel. Other products that can be produced from algae lipids are chemical products like glycerol, epoxies, fatty alcohols and polymers. [21]

There are also other more valuable compounds that can be extracted from algae. Some applications of these compounds were mentioned in the previous section, 2.1.1, Usage of algae today. The compounds are mainly omega-3 acids, pigments, vitamins and compounds that can be used as pharmaceuticals. [11, 20]

1.1 Cultivation systems

Cultivation systems can either be operated indoor or outdoor and more or less controlled. In general, a controlled system is referred to as a photobioreactor where a less controlled system is called cultivation pond or similar. The indoor cultivations are mainly closed and often systems with high surface/volume ratio and high control of cultivation parameters such as temperature, light supply and nutrients. Outdoor cultivations are in general less controlled systems and a range of different techniques exists, examples are flat-plate photobioreactors, tubular photobioreactors and open ponds (e.g. raceway). The supply of carbon to these systems is mainly through gas sparging and dissolution of carbon dioxide (CO

) and carbonate (CO

) during the time the gas flows upwards through the cultivation medium. [22]

2.1.3 Up-scaling of cultivation systems

Lopes da Silva et al. (2015) claims that the failure to properly scale-up algae facilities is one of the major reasons for the failure of starting commercial algae production sites. In chemical and

fermentation industry the consensus in scale-up is that no step should exceed a volume factor of 10, which to a large extent has been ignored in algae cultivation. [23]

Janssen et al. (2003) concludes that it is difficult to scale-up controlled photobioreactors unless the market price of the products is high. However, this type of reactors is suitable for large scale

inoculum production for mono cultures. It is simply very hard to produce algae at a low cost in these types of reactors (tubular reactors such as vertical air-lift and bubble column and flat panel reactors).

Since scale-up in practice would mean to build more units which would only reduce the investment cost to a lower extent. [24]

Craggs et al. (2012) demonstrated a big-scale raceway system for wastewater cleaning with algae.

Unfortunately, there is no data on the algae production cost since the focus on the experiment was

to remove nutrients from wastewater. However, the experiment showed that it is possible to reduce

the nutrient amount in wastewater with algae and at the same time enhance the degradation of

material by bacteria since algae produces oxygen. [25]

2.2 Cultivating environment

Algae needs a range of nutrients to be able to grow, both macro- and micro- nutrients. The macro nutrients are compounds with carbon, nitrogen, phosphorus and sulfur. According to Grobelaar (2004), the general chemical composition of algae is CO

H

N

P

. [26]

The carbon source is often in the form of CO

, but some species can also utilize CO

. Some algae species can also use organic substances as carbon and energy source, examples of compounds are acetate, glucose, lactate and ethanol. Some algae can only consume one of these compounds where others can consume a number of compounds. The nitrogen may be ammonia (NH

), nitrate (NO

), urea or more complex compounds such as amino acids. Phosphorus can be utilized from phosphate (PO

). [26]

Micro nutrients may be iron, magnesium or zinc. These are often in sufficient amounts in tap-water but lack of micro-nutrients in the medium may sometimes inhibit the growth of the algae. [26]

2.3 Algae grown in flue gas

Lately the interest of growing algae with flue gas has increased [27, 28]. The two main reasons are the possibilities of faster growth and cleaning of flue gas from different substances. This combination could be beneficial both economically and environmentally. The term flue gas refers to the gas mixture that is emitted after combustion of a carbon based material. If the combustion is made with air, the flue gas will mainly consist of nitrogen, oxygen, CO

, nitrogen oxides and sulfur oxides. A general composition range for flue gases from the combustion of woody materials from a paper and pulp mills are shown in Table 3. The concentrations vary between different burners (bark burner, recovery boiler etc.) and fluctuates over time. [29]

Table 3 General gas composition range for flue gases in the paper industry. [29]

Component CO2 (%) O2 (%) NOx (ppm) SO2 (ppm)

Range 5-20 2-20 50-150 <1-40

2.3.1 Carbon dioxide

The concentration of CO

in the atmosphere is a bottleneck in the cultivation of microalgae with air, even though the concentration is higher than preferred due to the combustion of fossil raw

materials, it is still only about 400 ppm [30, 31]. Using air as the carbon source would therefore

require a substantial gas volume and corresponding energy input for the gas pump, since the algae

biomass mainly consist of carbon, 36-65% dry-weight [27]. With the use of flue gas the gas volume

can be substantially lowered since the CO

concentration ranges between 5-20%, or about 100 to 500

times higher than air.

When CO

dissolves in water in the optimum pH interval for most microalgae species (6.4 to 10.3), the dominant form will be bicarbonate (HCO

) [27]. CO

can pass through the cell membrane for direct uptake, while only some microalgae can use HCO

. Therefore, HCO

needs to be converted back to CO

before the majority of algae species can utilize the carbon. If the algae consume CO

, more CO

will be formed from HCO

, since the chemical equilibrium will then be in favor for CO

formation, according to Le Chatelier's principle. However, the formation rate to CO

may be so low that even when there is an abundance of dissolved HCO

the growth will be carbon limited [32].

2.3.2 Nitrogen oxides

In general, flue gases contain some portions of NO

in ppm scale. On average, the NO

composition is 95% NO and 5% NO

, the concentrations of other NO

are in general negligible. NO is barely soluble in water, only 0.032 g/L at 1 tam and 25° C, while NO

have a considerable higher solubility of 213.0 g/L at 1 atm and 25° C. When NO and NO

dissolves in water they will react with the water and form nitric acid (HNO

) and nitrous acid (HNO

). If the pH is higher than 4, which is normal during

microalgae cultivation, the dominant forms of dissolved NO and NO

will be NO

and NO

. Since the concentration of nitrogen oxides is fairly low, it is not a sufficient nitrogen source, if the amount of CO

determines the amount of gas to be introduced to the cultivation. In that case, some other source of nitrogen needs to be used when cultivating algae. [27]

2.3.3 Sulfur oxides

Matsumoto et al. (2015) have shown that microalgae can be cultivated with flue gases having high concentrations of SO

. When cultivating algae using flue gas with a SO

concentration of 400 ppm there was no significant reduction of growth rate compared to 50 ppm. However, this was when the pH was stable at 8.0 and was achieved with addition of NaOH. Without the addition of NaOH the pH dropped to 4.0 and the growth was inhibited due to the acidic environment. Thus, the result

indicates that SO

does not inhibit the growth of algae. [28]

2.3.4 Heavy metals

In general, there are some heavy metals compounds in flue gases with varying concentrations. Algae have a strong metal binding capacity, which is beneficial if the goal of the cultivation is to clean the gases from heavy metals. However, if the aim is to produce biomass for further use it can be of an issue. Then, the heavy metals may either be removed prior to the algae cultivation, or removed in the downstream process. [28]

2.4 Separation techniques

The separation of the algae from the water is an important and problematic step in the cultivation of algae. Since the concentration of algae is very low it may require a lot of energy and costly unit operations to effectively separate the algae from the water. Alternatively, more simple techniques with longer operation times can be used. The tradeoff for which separation technique is best to use is different for different cultivation system. The choice of downstream design is also highly

dependent of the characteristics of the algae and the quality requirements for the final product. For a

visual example of the amount of algae compared to the amount water see Figure 2 c) below. [33, 34]

Usually the separation of algae from the water is performed in a number of stages. The first stage is called bulk separation where the majority of the water is separated from the algae. Examples of bulk separation processes are gravity sedimentation, flocculation and flotation. The second stage is called thickening, examples of common unit operations are centrifugation and filtration. After the

thickening the dry weight is often in the range of 5-15%. The final separation step is dehydration and examples of methods are sun-drying, cross-flow air drying, rotary drying, spray drying and flash drying. [16, 35]

The above mentioned unit operations are briefly explained in the following sections.

2.4.1 Gravity Sedimentation

Gravity sedimentation is one of the most common bulk separation techniques used for algae harvesting. If there are no agitation the algae will eventually fall down and build a sediment on the bottom of the cultivation tank. However, the density and radius of the algae (or any particle) have a big impact on the sedimentation speed and in practice sedimentation is only suitable for larger microalgae species (about > 70 µm). The advantage of gravity sedimentation is that there is no high energy input for the separation, only the removal of the water above the sediment. The downside is that it requires more time than the other techniques. [15, 34]

Figure 2 a) illustrates the gravity sedimentation of microalgae Coelastrella if no agitation is used. Figure b) shows the same flask just after it has been agitated (shaken by hand). Figure c) shows the same flask when it has grown a couple of weeks to illustrate the amount of algae compared to the water amount.

2.4.2 Flocculation

Flocculation is another bulk separation that essentially makes the sedimentation faster. The

chemistry of the cultivation tank is changed so that the algae aggregates, which will make the algae sink faster. Flocculation can be made by either a change of the pH or adding chemical coagulants.

Both ways need some kind of chemical addition and it may be expensive. Also, the addition of coagulants may contaminate the algae biomass. However, since flue gas lowers the pH it can be used to flocculate the algae to some extent. The advantage of flocculation over sedimentation is that it is faster, but might be more expensive. [33, 34]

a) b) c)

2.4.3 Flotation

Flotation is a bulk separation technique where solid particles are separated from a slurry or liquid by being carried to the surface by air bubbles. Small particles (less than 500 µm) will on collision with the bubble get stuck and float to the surface. Therefore, flotation is suitable when the algae is small since sedimentation then is ineffective. It is easier to get the particles to adhere on the bubble if they are hydrophilic, so the method may need some chemical addition. [33, 34]

2.4.4 Centrifugation

Centrifugation is a proposed thickening operation. Centrifugation uses external power to make the sedimentation very fast, from hours, days or even weeks to minutes. The issues with this method are that it requires high energy input, making it more expensive than other methods, and may damage the cells. It may be suitable for applications where degradation is an issue but then the costs need to be motivated by a high product price. However, Adam et al. (2013) tried do make centrifugation more economical and energetically feasible by decreasing the centrifugation time. They found that even though the separation efficiency went down from 90% to 28.5% of the algae biomass, the energy consumption per kg biomass was reduced by 82%. The separation cost is still too high for bulk-products but may be an interesting technique for high to medium value products. [33, 34, 36]

2.4.5 Filtration

Filtration is another proposed thickening operation. The algae slurry is concentrated using filters which will let the water pass but not the algae. Snow et al. (2015) claims that filtration is effective when dealing with low density algae. One issue with filtration is that the filter cake will over time increase the pressure drop and the algae needs to be removed from the filter regularly. [33, 34]

2.4.6 Sun-drying and cross-flow air drying

According to Brennan et al. (2010) the cheapest dehydration operation is sun-drying. The main advantages are that it’s cheap and increases the energy efficiency of the process, since the sun is used as energy source. There is however a number of disadvantages including long drying times, weather dependence and occupation of large areas. Therefore, it is questionable if the operation can be considered practical and efficient enough for large scale production. Show et al. (2015) is more skeptical to the process and only suggests it’s suitable for remote locations with no or unreliable sources of power. Show et al. (2015) also mentions the problem with degradation during longer drying periods. To decrease the drying time fans can be used to increase the air exchange, called cross-flow air drying. [15, 34]

2.4.7 Rotary drying, Spray drying and flash drying

Rotary drying, spray drying and flash drying uses heat to dry the algae slurry. Rotary drying achieves this by passing the slurry through a rotating drum with continuous heating, electrical or with steam.

Spray drying uses a spray nozzle to produce small droplets that are sprayed into a heated chamber.

Spray drying can also be performed using hot gases as the heating medium in direct contact with the biomass slurry, the operation is then called flash drying. The operations take seconds compared to days or weeks for sun-drying and have the advantage of sterilizing the biomass during the process.

The downside is higher equipment and running costs. [34]

2.4.8 The experiment

This work used the unit operation gravity sedimentation followed by sun-drying and skipping the thickening operation. This keeps the energy input and equipment costs as low as possible. See section 3.4 Pilot cultivation for a more detailed description of how it was performed.

2.5 Economics, a short overview

Since there is no commercial algae factory that produces algae for the bulk chemical segment, or any bulk segment, there is not a lot of data from actual facilities. There are however estimations based on both lab-scale and pilot scale experiments. [11, 37]

Acién et al. (2012) have made a cost analysis of a tubular photobioreactor facility of 10 reactors with a total capacity of 30 m

. The reactors were run during two years and the cost analysis is mainly based on the data produced from that experiment. The experiment is a good example of a well- controlled system, it was closed and the medium was highly controlled. With estimated lifetime and thus payback-time of 10 years for the photobioreactors, the results show that the two main costs for the facility are the construction of the facility and the labor (42.6 and 51.6 % respectively), where raw materials and utilities stands for only about 6% of the total cost. The production price was 69.3€/kg.

The authors conclude that if this type of production design is to be economically feasible the investment costs needs to be lowered significantly, a minimal amount of labor costs and the production needs to be performed close to the ideal photosynthetic capacity. They also think that it would be necessary to use flue gas as the carbon source and wastewater as the nutrient source.

Since a production of algae that replaces 10% of the petro-diesel consumption in Spain would consume about 4 times more nitrogen and 10 times more phosphorus than the country currently consumes. [38]

Norsker et al. (2011) compares the production cost of raceway ponds, tubular- and flat panel photobioreactors on commercial scale (100 ha). The article found that the current production costs were about the same for all three system types (4.95, 4.16 and 5.96 €/kg respectively). To be able to produce algae for the production of biodiesel (the focus of the publication) the production cost needs to become significantly lower. The authors discuss in the publication how this can be achieved for the systems and concludes that the tubular and flat panel systems had the possibility to be the cheapest systems. With a theoretical production price of 0.7 and 0.68 €/kg dw respectively, while a raceway system could produce algae at a cost of 1.28 €/kg. The theoretical production costs assume efficiencies in the production, no costs of CO

and medium and optimal location of the production facility. The intention with the theoretical production costs seems to show what is needed to lower the production cost of algae to a sufficient level for bulk production. The authors think that the required improvements are realistically possible in a 10-year period (from 2011). [10]

Some publications points at the possibility to produce algae with the combination of low-price bulk

products and valuable high price products, so to say produce the bulk products with the aid of the

valuable products. However, it is quite possible that the markets for the valuable products will get

saturated if sufficient amounts of bulk-products are to be produced. Therefore, it is of high

importance to ensure that the by-products included in any economical calculations of the algae

farming have a range of applications. [11, 21]

2.6 Photosynthetic efficiency of cultivation systems

The amount of solar energy that is transformed to biomass is called photosynthetic efficiency and can be calculated for any biomass production that are using photosynthesis as the energy source by dividing the energy value of the biomass by the amount of total solar energy radiated on the biomass.

Photosynthetic efficiency is one way to compare different cultivation systems. An article published

by Eustance et al. (2016) examined the photosynthetic efficiency of two Scenedesmus strains in flat

panel and raceway system. The photosynthetic efficiency of the flat panel system was between

1.32 – 2.24 % and 0.30 – 0.68 % for the raceway system. [39]

3 Method

3.1 Introduction

The inoculum cultivation and the cultivation ponds for the pilot cultivation were built by the author with aid and supervision of the SP algae group. Apart from minor contributions of measurements from the author, the lab cultivation was performed by the algae group. The inoculum and pilot cultivations were performed by the author. All methods have been discussed between the author and the algae group at SP.

The name and origin of the cultivate algae are Scenedesmus Obliquus (CAP 276/10) and Coelastrella species (Isolated by Lorenza Ferro, Umeå University, from the municipality water in Umeå, Sweden) A gardening fertilizer, Tg Växupp® Trädgårdsgödsel N P K 14-3-15 produced by Hammenhögs was used as the main source to nutrients in all experiments. The fertilizer is hereafter called NPK. Table 4 shows the composition of the NPK according to the manufacturer.

Table 4 The composition of the nutrient source used according to the manufacturer.

Element/compound Weight % N (total) 14.0%

NH

-N 7.8%

NO

-N 6.2%

P 3.0%

K 15.0%

S 10.0%

3.1.1 Measured Variables

The variables that were measured in the lab and pilot cultivations are found in Table 5. Explanations of the variables can be found in the following section.

Table 5 An overview of the measured variables in the lab and pilot cultivations.

Variable Lab Pilot

Optical density Yes, samples Yes, samples

Nutrients (NH

, NO

, PO

, SO

) Yes, samples Yes, on the iB samples pH Yes, continuously Yes, continuously

Number of cells Yes, final sample Yes, beginning and every week

Temperature Yes No, only outdoor temperature

Light energy Yes, fixed Yes, sun

Gas composition Yes, defined Yes, one-time sample

Sun-drying rate No Yes, continuously

Energy value No, insufficient amount of

sample Yes, for S. obliquus, not for

Coelastrella sp.

3.1.2 Measuring Methods

All measurements were performed at SP by the author under supervision by the algae group or

Optical density (OD) was measured using a Varian 50 Bio spectrophotometer and measure the absorbance at 750 nm, with a corresponding medium as the blank (tap water with NPK, 0.0814 or 0.88 g/L). OD is a common method of estimating the amount of algae in water solutions and is fast and cheap to perform. Duplicate samples were measured and a mean value of these was used as the result.

The amount of nutrients in the medium was measured using ion chromatography (IC) with a

Metrohm 882 Compact IC plus. IC is a reliable method for measuring the amount of a range of anions and cations. The measuring program and calibration was performed by professional staff at SP.

Duplicate samples were measured and the mean value was used as the result. To decrease the measuring costs of the pilot samples the duplicate samples were filtered through the same filter (0.45 µm) and only the iB samples was measured (see section 4.1.1 for explanation of the notation iB).

pH was measured by glass electrodes at the lab and also directly in the pools that was continuously monitored.

The number of cells were counted in a microscope with a hemocytometer from Bürker, 0.100 mm depth, 0.0025 mm

, using 40x magnification. For each measured sample, the cells were counted in 15 random squares or until 300 cells were counted.

The light energy for the pilot was measured by a pycnometer located on house 1 at SP.

The gas composition was measured using gas chromatography (GC) by professional staff at SP.

The sun-drying rate of the sediment was measured with a ruler at the deepest position in the ponds.

The energy value was measured using calorimetry (IKA) performed by professional staff at SP.

3.2 Lab Cultivation

The lab cultivation is an equipment built and used by the algae group at SP. The equipment is used for research on algae cultivated in flue gas and the gas composition can be adjusted. To have

favorable growth conditions there is continuous agitation and the cultivation volume can be assumed to be homogeneous. All equipment was sterilized in an autoclave at 121°C for at least 20 minutes, including the medium.

3.2.1 Setup

Regular 600 mL schott flasks with a modified cork were used as cultivation flasks. Agitation was provided by magnetic stirrers and the gas flow. Figure 7 shows a picture of the lab cultivation. Only the result from 6 flasks is used in this work.

The starting density of the cultures was aimed to be 0.01 OD. The gas was mixed from gas-tubes to the desired gas-mixture, see Table 6.

Table 6 Gas composition of the artificial flue gas used in the lab experiment.

Air CO

NO SO

90 % 10 % 100 ppm 10 ppm

Figure 3 The lab-cultivation setup with light turned on at SP.

The sampling tubes can be seen as well as the gas inlet and the pH-electrodes.

3.3 Inoculum Cultivation

To get a comparable result it is desirable that the starting density of the cultivation is the same for the lab- and pilot-cultivations. A setup for the required amount of algae inoculum did not exist at SP, therefore an inoculum cultivation was built with a capacity of 200 L.

3.3.1 Setup

The setup consisted of 10 new beer brewery barrels in PET plastics with a volume of 23 L each. The barrels were washed and sanitized with the sanitation agent PBWä (manufactured by Five Star Chemicals & Supply Inc.). The carbon-supply consisted of pressurized air of unknown flow and the rest of the nutrients were supplied using NPK. The light was provided by artificial sunlight in a 16/8 (light/dark) cycle per day of unknown effect. Agitation was provided by the air supply and magnetic stirrers. The stirrers were turned on at 18:00 and off at 06:00 Monday-Friday, thus having 12 hours of stirring per 24 hour weekdays and all the time during the weekend. The reason for not having the stirrers turned on all the time was to spare the nearby offices and labs from the disturbing sound from the stirrers.

The main amount of water was sterilized by filtration using Corning® 1000mL Bottle Top Vacuum Filter, 0.22µm Pore 54.5cm² PES Membrane. A smaller portion of the water with the NPK was sterilized in an autoclave for 60 minutes at 121°C, the final medium concentration was 0.88 g/L. The algae were cultivated in room temperature, about 20-25°C.

When it was time to start the pilot cultivation the optical density was measured in all barrels, and the amount of volume to obtain a starting density of 0.01 OD was calculated. To ensure that it was single-cell cultures with the desired algae that was inoculated, samples were investigated in microscope prior to the inoculation. The optical density was high enough to use one barrel for Scenedesmus obliquus and a mixture of three barrels were used for Coelastrella sp. The Coelastrella sp. barrels were mixed into three jerry cans to ensure that the same algae mixture was inoculated.

The S. obliquus barrel was divided into three jerry cans. A picture of the inoculum cultivation is

presented in Figure 4.

Figure 4 The inoculum cultivation with the light turned on. Each barrel have a total volume of 23 l and a liquid volume of about 20 l, the magnetic stirrers can be seen on the floor supporting a

plastic sheet used for stability. The sampling outlets are visible on the side of the barrels.

3.4 Pilot cultivation

3.4.1 General description of the cultivation principle

The pond design used for this project is based on an another design made by the SP algae group called Tethys. The Tethys pond consists of a dome shaped lid covering a “fast-set” plastic pools that can be found in many retail stores, see Figure 5 for illustrations. The liquid capacity is about 2 300 L per pond.

The Tethys reactor principle is that based on the dissolution of the CO

from the dome into the water, rather than on dissolution of gas bubbles in the water column. Gas enters about 10 cm below the water surface that causes a restricted circulation in the upper region of the pond, while further down the algae can spontaneously settle and sediment to the floor. The algae can also spontaneously sediment in the upper level when the gas-flow is off (see section 3.4.2 below). The cultivation system is benefited by a large surface interface between water and gas as well as by a high concentration of CO

in the dome. In this way, the energy required for pumping is reduced dramatically.

When the cultivation is ended and sufficient sedimentation has occurred, the water above the sediment is removed and the construction is used for sun-drying of the biomass. However, one problem with this design of the construction is that the walls partially collapse over the sediment, and may cause incomplete drying of the biomass. To improve the drying process, the fast set pond design was in this work abandoned and instead replaced by a more solid construction based on the same operational principle.

Figure 5 Illustration of the Tethys pond design (not used in this work).

3.4.2 Setup

Six ponds were used, three per specie. The ponds were made from new IBC tanks (food grade plastics) with an original volume of 1 m

. However, after the modification they had a liquid capacity of 500 L, the water height was about 45 cm. Pictures of the ponds can be seen in Figure 6. All ponds were equipped with a pH electrode, a gas inlet and two sampling pipes at different heights (about 5 and 20 cm from the water surface, or 40 and 25 cm from the bottom). The gas inlet was a plastic tube, 2,5 cm in diameter with the outlet about 15 cm from the water surface. The transparent plastic cover used is a greenhouse plastic; Vexithene EVA, Hytiluc difuse. An illustration of the pilot

cultivation can be found in Figure 7.

Figure 6 The pilot cultivation ponds used in this experiment during the algae cultivation.

When CO

dissolves in water the main fraction transforms to carbonic acid. Carbonic acid can release up to two hydrogen ions, which lowers the pH. Therefore, the gas flow needs to be regulated to not lower the pH too much and then kill the algae. Therefore, pH was measured continuously in the ponds to autonomously control the gas flow. If the pH was measured to be higher than 7 the gas valve was opened, and if the pH was lower than 7 the gas valve was closed. Since all gas inlets came from the same gas pump the gas flow was not constant when the gas valve was open.

The only agitation in the ponds was provided by the gas flow, so agitation was only provided if the pH was measured to be lower than 7. The gas pump was set on a timer with 16/8 hours on/off per 24 hours with the gas turned on about 6:00 and off during the night. The carbon supply consisted of flue gas that was provided by a pellets burner designed for house heating (Janfire SWE 20 kW Integral powered with a Janfire NH moody burner). All other nutrients were supplied with NPK, with the initial concentration of 0.0814 g/L. If the measurements indicated that the nutrients were almost consumed, more were added in the form of a concentrated solution equal to a final concentration of 0.0814 g/L in the ponds.

The burner was controlled by the amount of heating required and heated a closed cooling pipe system that was buried in the ground beneath the ponds and besides the ponds. The pipe system can heat the cultivation ponds during colder seasons. However, since the cultivation period was during the summer, the heat was also discarded using the heat-dump section of the pipe system, see Figure 7. If the cooling medium was not cooled enough the burner would not produce flue gas. Therefore, it is possible that the composition of the gas varied. Apart from one gas sample, this was not

measured.

When the cultivation was finished and the sedimentation phase started, the burner was shut down, and thus didn’t produce heat for the ponds during the sedimentation and sun-drying. Also, prior to the cultivation, a thermometer was installed in one pond to register the temperature, but the data was not possible to recover.

Figure 7 A schematic view of the pilot cultivation, the relative positions of the equipment is not the same as in the actual pilot and not to scale.

3.4.3 Sampling

The samples were taken with syringes attached to the two pipes.

The pipes were fixated at different heights in the ponds (25 and 40 cm from the bottom). The sample locations where chosen to measure the growth close to the surface and to see if there was a difference in the algae concentration a bit further down in the pond compared to the top layer.

First the pipes were filled with liquid using the syringe to suck it up. Then the whole amount of liquid that was in the pipe was sucked into the syringe and then discarded (about 60 ml).

Thereafter the sample was taken, about 20-30 ml divided into 2 flasks. The sampling occurred at least 3 times a week. The plots in the results section show how often sampling occurred.

Figure 8 Picture of the flue gas inlet

3.4.4 Experiment

The algae species studied were Scenedesmus obliquus and Coelastrella sp. The first algae, S. obliquus, have been cultivated several times by the algae group at SP and are used as their standard algae.

Coelastrella sp. were successfully cultivated in the lab system by the algae group at SP and was therefore chosen to be the second algae.

When the cultivation time was about the same as the lab experiment the gas flow was turned off and the algae was allowed to fully sediment. Then the water above the sediment was pumped out from the pond. When this was done OD samples was taken from the water above the sediment and samples from the sediment was collected. Thereafter the algae were sun-dried in the ponds with the plastic cover still on top but opened at two sides. When the sun-drying were sufficient the algae were scraped off the bottom and then dried in a vacuum-oven to ensure that all water was drained before it was weighed and the energy analysis was performed.

4 Results and discussion

4.1 Introduction

Prior to the pilot experiment, the expected result was discussed based on previous experience.

Some of the expected results were:

• The mono-cultures in the lab will barely be contaminated

• The mono-cultures in the pilot will be contaminated

• Higher productivity in the lab cultivation compared to the pilot

• The pilot will require a high amount of inoculum volume

4.1.1 Pilot cultivation

The notations iA and iB (i=1-6) refers to the different sampling locations in the ponds, where A refers to the sampling close to the surface (about 5 cm deep) and B the sampling deeper in the ponds (about 20 cm deep). The numbers, 1-6, refers to the different cultivation ponds, Scenedesmus obliquus was cultivated in pond 1-3 and Coelastrella sp. was cultivated in pond 4-6.

4.2 Optical density

4.2.1 OD in Lab cultivation

Figure 9 shows the growth in the lab cultivation. The curves are a mean-value of three flask cultivations per specie and the variations are included in the plot. Since the cultivations are fully mixed the measurements are a good estimation of the amount of algae growth.

0,0

0,5 1,0 1,5 2,0 2,5 3,0

0 2 4 6 8 10 12 14 16 18

Op tic al d en si ty (a bs or ba nc e @ 7 50 n m )

Cultivation time (days)

Algae growth in lab cultivation

S. obliquus Coelastrella sp.

4.2.2 OD in Pilot cultivations

A plot over the algae growth in pond 1-3 can be viewed in Figure 10. After 7 days of cultivation, the pH-regulation stopped to work properly. The pH measurement in pond 1-3 registered other values compared to a control measurement of samples. In pond 1 a lower value compared to the control sample was registered and pond 2 and 3 registered a higher pH compared to the control. This resulted in a higher gas flow to pond 2 and 3, while only a small amount of gas was added to pond 1, see Figure 12. Therefore, the growth is high in pond 2 and 3 and low in pond 1. Since the samples are only from the upper layer in the ponds, and the algae could sediment during the cultivation, the plot only shows an indication of the algae growth. After day 18 the curve shows the sedimentation speed in the ponds between the cultivation and the dewatering. There are only smaller differences in the optical density between the two sample heights. It can be observed that during the sedimentation phase the optical density drops about 50% during the first 4 days, with no additional flocculates or temperature changes with external power.

Figure 10 Algae growth in the upper layer in the Scenedesmus obliquus ponds based on optical density.

Some important actions are also highlighted, like the gas flow stop between day 17 and 18.

Figure 11 shows that the algae concentration in the upper layer of pond 4-6 was negative. It was observed that the Coelastrella sedimented during the first days and formed a layer on the bottom of the pond. The problem with pH-regulation did not occur in pond 4-6. Observe that the y-scale is only between 0-0.1 while the plot in Figure 10 have the scale 0-1. The sedimentation speed wasn’t

0,0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1,0

0 2 4 6 8 10 12 14 16 18 20 22 24 26

Optical density (absorbance @ 750 nm)

Cultivation time (days)

Algae growth and sedimentation speed in upper layer, pond 1-3 (S. Obliquus)

1A 1B

2A 2B

3A 3B

NPK addition Malfunction of pH controll

Malfunction of pelletsburner Gas flow stop Water pumped out

Figure 11 Algae growth in the upper layer in the Coelastrella ponds based on optical density.

Some important actions are highlighted, like the gas flow stop between day 17 and 18.

4.3 pH/valve log and gas composition in the pilot 4.3.1 pH and gas valve log for the pilot

The logs for the pH and gas-valve are presented in Figure 12. The logs are only for the first 12

cultivation days, the raw data for all 18 cultivation days exists, but at the time of this report it has not been processed to graphs, which could not be performed by the author. The pH measurements started to malfunction somewhere after the 7:th cultivation day, 25:th of June. One can note that the gas valves was in general closed for pond 4-6 and pended between closed and open for pond 1-3 until the malfunctioning. After the malfunctioning (until at least day 12) the gas valve was mainly closed to pond 1 and mainly open to pond 2 and 3. Observe that the gas supply was shut down during the night, see section 3.4.2 Setup, but the pH regulation was not. So the gas valve could be open but there is not necessarily a gas flow to the pond.

0,00 0,01 0,02 0,03 0,04 0,05 0,06 0,07 0,08 0,09 0,10

0,0 2,0 4,0 6,0 8,0 10,0 12,0 14,0 16,0 18,0 20,0 22,0 24,0

Optical density (absorbance @ 750 nm)

Cultivation time (days)

Algae growth in pond 4-6 (Coelastrella)

4A 4B

5A 5B

6A 6B

Malfunction of pelletsburner Water pumped out Gas flow stop

Figure 12 pH and gas-valve log for the first 12 cultivation days (16-28 June). The relay refers to the gas-valve and if its equal to 0 it’s closed and if equal to 1 it’s open. Thus, a high frequency of the red lines means that the gas flow is turned

on and off often while a small frequency means that the gas is mainly on (stays at 1) or off (stays at 0).

4.3.2 Flue gas composition

The concentration of CO

in the flue gas (8.6%) in the pilot is a bit lower compared to the concentration used in the lab cultivation (10%), see Table 7 and section 3.2 in methods. The

concentrations are in the general flue gas range for flue gas from pulp and paper mills (5-20 %), see section 2.3 Algae grown in flue gas.

Table 7 Gas composition of the flue gas from the pellets burner at the pilot, one time, duplicate sample.

Component Sample A (vol. %) Sample B (vol. %) Mean (vol. %)

CO

8.6 ± 0.1 8.6 ± 0.1 8.6

O

12.7 ± 0.4 12.6 ± 0.4 12.7

N

78.7 ± 1.0 78.8 ± 1.0 78.8

CH

<0.1 <0.1 <0.1

4.4 Nutrients

4.4.1 Lab cultivations

The concentration of nutrients during the lab cultivation in the S. obliquus flasks can be viewed in Figure 13. Additional nutrients were only added on day 8. Samples were collected before and after the addition, resulting in the drastic increase on day 8 for all nutrients. One can note that the concentration of NO

was quite high, while almost all NH

was consumed. The plot indicates that the NPK was the main source of SO

, since it only increases slightly, expect when NPK was added.

Figure 13 Nutrient concentration in the lab flasks 1-3, the variation between the flasks are displayed by the black lines through all measuring point. The most important compounds are NH4⁺ and NO3- followed by PO43- and lastly SO42-.

0,0 0,2 0,4 0,6 0,8 1,0 1,2 1,4 1,6 1,8 2,0

0 2 4 6 8 10 12 14 16 18

Conentration (mmol)

Cultivation time (days)

Nutrients lab, S. obliquus

NH4(+) NO3(-) PO4(3-) SO4(2-)

The concentration of nutrients during the lab cultivation in the Coelastrella sp. flasks can be viewed in Figure 14. Additional nutrients were added on day 8, 11 and 13, seen as stepwise increase of nutrient concentrations in the plots. The consumption of nutrients where substantially higher compared to the S. obliquus flasks (Figure 13). This was excepted since the growth curve shows a higher growth rate for Coelastrella sp.

Figure 14 Nutrient concentration in the lab flasks 4-6, the variation between the flasks are displayed by the black lines through all measuring point. The most important compounds are NH4+ and NO3- followed by PO43- and lastly SO42-.

4.4.2 Scenedesmus obliquus pilot cultivations

Observe that the x-axis is 0-18 days in the nutrients plots and 0-26 days in the growth plots, see Figure 15-Figure 20 and Figure 9-Figure 10. Apart from the initial amount of nutrients, nutrients were added two times to pond 1-3. Nutrients was only measured in the samples from sample position B.

The sample in pond 1-3 was taken after the nutrients was added day 7 and before on day 13. The concentration of sulphate increases in all ponds. This is expected since algae only consume low amounts of sulphate, and the flue gas and nutrients additions in pond 1-3 increases the concentration, see Figure 15-Figure 17. One can notice that NH

is consumed before NO

. The reason why sample 1 (day 0) has lower concentrations compared to sample 2 (day 2) for pond 1-3 is the inhomogeneity of the system due to low mixing, and that the initial nutrient solution was added about 1 hour before the sampling.

The nutrient consumption in pond 1, see Figure 15, corresponds well with the growth curve (Figure 10). It has a lower consumption compared to pond 2 and 3 and have not consumed all the nutrients when the cultivation was ended. This was expected sine the pond didn’t have sufficient amounts of CO

available due to the malfunction of the pH regulation.

0,00,2 0,40,6 0,81,0 1,21,4 1,61,8 2,0

0 2 4 6 8 10 12 14 16 18

Concentration (mmol)

Cultivation time (days)

Nutrients lab, Coelastrella sp.

NH4(+) NO3(-) PO4(3-) SO4(2-)

Figure 15 The nutrients measurements over time in pond 1. Some important actions are also highlighted, like the nutrient additions. The most important compounds are NH4+ and NO3- followed by PO43- and lastly SO42-.

The nutrient consumption in pond 2 shows that the algae growth was higher compared to pond 1 and 3 (Figure 15 - Figure 17). The concentration of NH

dropped faster and the concentration of NO

was almost consumed and reached a peak of 0.43 mmol compared to pond 3 (0.62 mmol) on day 7.

The consumption of PO

is notable higher compared to pond 1 and a bit higher compared to pond 3.

This was expected based on the result from the growth curve (Figure 10).

0,0 0,1

0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1,0

0 2 4 6 8 10 12 14 16 18

Co nc en tr at io n (m m ol /l )

Cultiavion time (days)

Nutrients in pond 1

NH4(+) NO3(-)

PO4(3-) SO4(2-)

NPK addition Malfunction of pH controll

Malfunction of pelletsburner Gas flow stop

0,0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1,0

0 2 4 6 8 10 12 14 16 18

Co nc en tr at io n (m m ol /l )

Cultiavion time (days)

Nutrients in pond 2

NH4(+) NO3(-)

PO4(3-) SO4(2-)

NPK addition Malfunction of pH controll

Malfunction of pelletsburner Gas flow stop

The nutrient consumption in pond 3 is also as expected and has been discussed in comparison with pond 1 and 2 above. The result is overall as expected based on the growth curve (Figure 10).

Figure 17 The nutrients measurements over time in pond 3. Some important actions are also highlighted, like the nutrient additions. The most important compounds are NH₄⁺ and NO₃^- followed by PO₄^3- and lastly SO₄^2-.

4.4.3 Coelastrella sp. pilot cultivations

No additional nutrients were added to pond 4-6. Nutrients was only measured on samples from sample position B.

The nutrient plots for pond 4-6 are similar, see Figure 18 -Figure 20. There is a low consumption of NH

and barely any consumption of NO

. The concentration of sulphate doesn’t increase, since the gas-flow is fairly low and no additional nutrients were added. Since the algae in pond 4-6 sedimented the OD measurements do not tell if there has been any growth in the ponds. The nutrient results show that there has been a small amount of growth. The consumption of nutrients was highest in pond 5 followed by pond 6 and lastly pond 4. The reason sample 1 (day 0) have lower concentrations compared to day sample 2 (day 2) for pond 4 is the inhomogeneity of the system due to low mixing, and that the initial nutrient solution was added about 1 hour before the sampling. The result indicates that the growth in pond 4-6 was very low.

0,0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1,0

0 2 4 6 8 10 12 14 16 18

Co nc en tr at io n (m m ol /l )

Cultivation time (days)

Nutrients in pond 3

NH4(+) NO3(-)

PO4(3-) SO4(2-)

NPK addition Malfunction of pH controll

Malfunction of pelletsburner Gas flow stop

Figure 18 The nutrients measurements over time in pond 4.

Some important actions are also highlighted, like the gas flow stop.

Figure 19 The nutrients measurements over time in pond 5.

Some important actions are also highlighted, l like the gas flow stop.

0,0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1,0

0 2 4 6 8 10 12 14 16 18

Co nc en tr at io n (m m ol /l )

Cultivation time (days)

Nutrients in pond 4

NH4(+) NO3(-)

PO4(3-) SO4(2-)

Malfunction of pelletsburner Gas flow stop

NPK addition

0,0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1,0

0 2 4 6 8 10 12 14 16 18

Co nc en tr at io n (m m ol /l )

Cultivation time (days)

Nutrients in pond 5

NH4- NO3-

PO4(3-) SO4(2-)

Malfunction of pelletsburner Gas flow stop

NPK addition