Enhancing Algae Biofuels: Nannochloropsis oculata
Part of the Young Naturalist Awards Curriculum Collection.
Introduction
Growing up by the rich lakes and streams of Colorado, I had often noticed dark masses of algae clouding the clear water. The idea that these simple life forms could somehow help power our modern society without the environmental drawbacks of fossil fuels is inherently appealing—a truly green solution.
I first became familiar with the concept of algae for fuel three years ago as I was looking for other lipid sources for the biofuel I was making at home. Interested in a clean, renewable alternative for our world's energy needs, I first prepared biodiesel with recycled vegetable oil, as a neighbor was doing for use in his vehicles. Soon, however, I learned drawbacks of relying on commonly used corn and soybean oils: relatively low oil yields (under 50 gallons per acre per year) and reliance on resources already needed for valuable food crops.
Then I found out about algae. Depending on the strain, microalgae can yield from 5,000 to 15,000 gallons of oil per acre per year. Algae grow extremely fast and can be cultivated in areas not practical for other crops, such as on arid lands in Colorado, where they are currently being tested. Microalgae can also reduce carbon emissions due to their high CO2 sequestration rates. However, many factors must be optimized for algae biodiesel to become viable for widespread use.
When I visited an algal research facility and talked with the scientists, I thought about what I could do to contribute. One aspect of algae fuel that hasn't been thoroughly researched is that of the precise growing conditions necessary to optimize oil yields.
Therefore, this project analyzes the effects of two source of growth stress—nitrogen limitation and carbon dioxide infusion—on the biomass and lipid production of Nannochloropsis oculata, a microalgal strain common in current research.
Testable Question: How do nitrogen limitation and carbon dioxide infusion affect the cellular lipid content and biomass of the microalgae Nannochloropsis oculata?
Hypothesis: If nitrogen limitation and carbon dioxide infusion are imposed upon cultures of the microalgae Nannochloropsis oculata during growth, and the algae are evaluated gravimetrically by biomass and cellular lipid content, the following are hypothesized:
- Nitrogen-limited algae will have the highest lipid content but the lowest dry weight.
- Carbon-dioxide-infused algae will have the highest dry weight and medium lipid content.
- Control algae will have the lowest lipid content and medium dry weight.
When lipid content and biomass are combined to find the overall lipid production (oil yield), carbon dioxide-infused algae will produce the most overall lipids, followed by nitrogen-limited algae and finally the control algae.
Background Research
Algae are aquatic, eukaryotic, photosynthetic organisms ranging in size from single-celled forms to giant kelps. Algae can be grown on nonarable land so as not to compete with other crops and are highly capable of efficiently using solar energy and CO2 to create biomass. Like terrestrial plants, algae produce neutral storage lipids in the form of triglycerides, long-chain fatty acids used in transesterification to produce biodiesel; they are capable of utilizing the nutrients in wastewater streams and, most importantly, they have high growth rates. For all these reasons, algae are attractive for meeting today's energy demands. Microalgae are most often used for biodiesel since they tend to have higher lipid concentrations than macroalgae such as seaweed.
Algae for biodiesel are most commonly grown in a closed and controlled system rather than an open pond (which is too uncontrolled to yield optimum oil). A photobioreactor (PBR) is often used; this is defined as any bioreactor involving a source of light (a bioreactor is a vessel used to carry out some sort of biochemical process). PBRs are usually closed systems that allow for control over the purity of the culture, the temperature and makeup of the growth medium, the amount and intensity of light, and the levels of nutrients and carbon dioxide allowed to reach the algae. These variables should be as closely controlled as possible for optimum culture growth (Rocha 2003).
Studies have shown that some environmental stress may cause microalgae to produce more lipids; however, such stress may also cause a drop in growth rates. Factors that might affect algal lipid production include nitrogen limitation, high CO2 concentrations, phosphorus limitation, sulfate limitation, desiccation, auto-inhibitor formation and salinity (Chiu 2008, Wang 2009, Lembi 1988, Renaud 1994). In this project, the first two factors listed will be studied due to their comparatively easy implementation.
Several methods have been used to extract oil from microalgae, including hexane-isopropanol and chloroform-methanol solvent extractions, a dry press and supercritical carbon dioxide solvent extraction. Due to the small scale of my experiment and my limited access to high-tech equipment, I decided solvent extraction to be the most practical for me to use (Lee 1998). This method involves separating the algae from the growth media using a centrifuge, drying the harvested algae, lysing the cells, mixing in hexane-isopropanol (ratio 3:2 v/v) to extract the lipids from the lysed cells, separating the phases and then evaporating the hexane (top) phase to leave behind the extracted oil. Chloroform-methanol (ratio 3:1 v/v) can also be used for general lipid quantification.
I chose the algal strain Nannochloropsis oculata for research because both a review of the literature and personal interviews with algal researchers indicated it is a hearty strain with great potential for fuel due to its rapid growth and sufficient oil yields. It is a eustigmatophyte, a photosynthetic unicellular autotroph with coccoid cells and polysaccharide cell walls. It is often grown in f/2 medium, an enriched seawater derivative. I used a recipe for this media from the Center for Culture of Marine Phytoplankton and made it from their stock solutions.
Nile Red (9-diethylamino-5H-benzo[alpha]phenoxazine-5-one) is a lipophilic fluorescing red dye which has been shown to be especially helpful in staining intracellular lipid droplets (Lee 1998, Cooksey 1987). Since its fluorescence is fully quenched in water, it acts as a fluorescent hydrophobic probe. Studies have demonstrated the efficacy of this stain in quantifying lipids in microalgae; a stock solution of 0.1 mg/ml acetone can be added to a known quantity and concentration of culture, excited in a fluorometer at 485 nanometers, and its emission wavelength measured to determine overall fluorescence and therefore the overall lipids in the culture. A fluorometer can be used to measure parameters of fluorescence: its intensity and wavelength distribution of emission spectrum after excitation by a certain spectrum of light. These parameters are used to identify the presence and amount of specific molecules in a medium. From this, the individual lipid content of cells in a culture should be able to be determined accurately.
Materials
Procedural Design
I developed and carried out the following protocols in my research. First, I built a photobioreactor (PBR) to allow for controlled home culturing of N. oculata. It consisted of 12 tubes (bio cells) positioned around a central light source (four aquarium grow lights).
Next, I prepared f/2 medium by adding Instant Ocean sea salt (35g/L) to 1 ml/L of the stock solutions N, P, Si and TM, and 0.5 ml of solution V. To implement nitrogen stress, I limited levels of NaNO3 stock in the stressed medium to 25% of the levels used in the control medium (0.25 ml/L). The media were buffered with sodium bicarbonate and Tris buffer.
Next, I cultured N. oculata for two weeks at 22°C in my PBR, with a starting dilution of 75 ml dense inoculum (ordered from Algagen LLC) in 500 ml of medium. Four bio cells were cultured under control conditions (control medium, air bubbled); four were cultured under nitrogen-limited conditions (nitrogen-limited medium, air bubbled); and four were cultured under CO2 infusion (control medium, bubbled with CO2 instead of air). All bubbling was implemented after three days and began lightly while growth was still young. I monitored the temperature, pH and qualitative growth daily. Culturing in my PBR was performed twice.
After growth, I analyzed my cultures. I separated the algal cells from the set volumes of culture first with vacuum filtration (first PBR run only); I then used centrifugation. I dried and weighed the separated cells to determine the algal dry weight (biomass). I used centrifuged samples of dry algae from the previous step to perform gravimetric lipid analysis, with two types of chemical solvent extractions: chloroform-methanol (3:1 v/v) extraction (first PBR run only), and hexane-isopropanol (3:2 v/v) extraction. I first lysed algal cells with mechanical blending, microwave irradiation and ultrasonification, then added the respective solvent mixture to solubilize and extract cellular lipids from the cell mass. I used separatory funnels to draw off the phase containing dissolved lipids, and dried and weighed the extracted oils with a hot plate. I also used a fluorometric analysis with the fluorescing lipophilic red dye Nile Red to determine cellular lipid levels. I stained fresh cell samples with a solution of Nile Red in acetone (0.1 mg/ml), then used a calibrated fluorometer to measure sample fluorescence, which I would then normalize to cell concentrations.
I evaluated the relative weights of the dry mass of the algae to determine the relative biomass of cultures. I divided the extracted lipid quantities by the original dry mass of the algae sample to obtain the percentage of dry mass that was lipid; I compared these percentages to determine the relative lipid productions of the algae samples. I found the overall lipid production (oil yield) of the samples by multiplying the lipid content by the dry algae biomass.
Data
Analysis
Data Discussion: The following general statements are supported by my data: the nitrogen-limited samples showed a higher lipid content but less biomass; the control samples showed a lower lipid content and higher biomass; all the carbon dioxide-infused samples initially grew well but then quickly died off, most likely due to an acidic medium, an unexpected side effect of extra CO2.
I assessed the overall lipid concentration of the tested samples by evaluating their lipid content and biomass concentration; this data determines which variable ultimately prompts algae to produce the most lipids overall. For both the hexane and chloroform extractions, the highest producers were the N cultures; their higher-percentage production rates overcame their lower biomasses.
There were a few unusual points in my data, one being the biomass of Sample C1A via centrifugation, which is inconsistently higher than both C1A via filtration and C1B. This result was probably due to measuring errors, as it was my first trial with centrifugation, and I suspect I had not yet fallen into a standard protocol for evaluation. The only other real discrepancy is between the ND2, ND3 and ND4 triglyceride and total lipid extractions; this could be explained partially by the fact that a drop or so of diluted ND3A was lost during extraction, and by a possible irregularity of the total lipids-to-triglycerides ratio within the samples themselves. However, given the amount of data and taking into account unavoidable irregularities, all results were very consistent for at least internal comparison.
Procedural Analysis and Observations: The first time I attempted to culture the algae, I met with no success. I used small dilutions of the initial seed culture (only 1:250 on my first attempt, then 3:100, 1:25 and finally 1:10, but by that point my seed cultures were dying), which meant that the culture wasn't able to start properly when very dilute. Another factor possibly contributing to my initial failure was that of too much agitation. Upon discussion with other algal researchers, I found that more success might be gained by delaying agitation until a day or two after inoculation. Too much bubbling could also have had an effect on the next issue with growth: tests with indictor paper revealed that the pH of my media was much too low: only 6.5, when the optimum level for N. oculata growth is 8.2 to 8.4. Most likely the extra agitation introduced too much carbon dioxide into the media, forming carbonic acid and making it far too acidic for the cultures to survive. The final aspect of my setup that may have been an issue was the temperature of my PBR, which was consistently a few degrees lower than optimum (about 20°C rather than 22°C).
To rectify all of this, I did the following: First, I ordered a larger and denser seed culture (2L high density) and cultured at a ratio of 3:20 (75 ml seed in 500 ml media). Also, I did not aerate for the first three days after culturing and then bubbled gently, with only a few bubbles a second. To rectify the low pH, I added 50 ml of a saturated solution of control media and sodium bicarbonate with a pH of 9 to every liter of media to bring it to a pH of 8 to 8.5, as well as monitoring pH throughout growth. Finally, I moved my entire setup to a closet with its own thermostat so that I could regulate the temperature closely at 22°C. These alterations succeeded, and the cultures grew well. I was surprised by how hard it was to cultivate algae in the first place, and then amazed by how quickly it grew once I had gotten it started!
However, there were some issues even in this last batch: the bubbling was very hard to regulate. The tubing, gang valves and check valves did not allow for precise aeration, and as a result some slight inconsistencies may have resulted. Also, after an initial period of seemingly good growth (by visual inspection), by the end of the first week of culturing, all the carbon dioxide-infused trials had died due to low pH. Despite my best efforts, the extra carbon dioxide still made the growth media too acidic. This effect surprised me, since I had expected to find that additional carbon dioxide would increase the CO2 sequestration of the algae and increase its potential for both lipid content and growth. However, since the carbon dioxide-infused trials crashed despite all my work to keep them alive, my results indicate that without extremely careful pH regulation, carbon dioxide infusion is not a viable course for increasing the lipid content or the biomass of microalgae.
To harvest the algae, I poured the bio cells off into clean containers. As all the carbon dioxide trials were by this time clear media without even a tinge of green, I chose not to waste laboratory time on analyzing them. Any remaining live cells would not be quantifiable. I noticed that the algae had turned into sediment at the base of my bio cells; this necessitated several rinses with already harvested culture to ensure maximum collection. Finally, at harvest, I noted the presence of a cream-colored, filmy sediment at the bottom of my culture tubes. The cultures with the least growth (judged visually) seemed to have the most of this substance. Research has lead me to believe that this layer was probably dead cells and other biological detritus, and it stands to reason that this would accumulate to the greatest degree in the samples with the least biomass.
To my dismay, fluorometric analysis of the cultures failed altogether, despite many attempts and exhaustive research into procedures. Serial dilutions of samples of a subculture, stained using techniques found in the literature and excited at several different wavelengths, failed to show a peak at about 580 nm, which is characteristic of Nile Red dissolved in lipids. Instead, the spectrographs were similar to ones of unstained cells; to test this hypothesis, I measured the fluorescence of unstained dilutions as well. The spectrographs were nearly identical, so I concluded that the dye had not penetrated N. oculata's thick cell walls. I tried again using a different method designed specifically for N. oculata; I included an incubation step; and I even attempted lysing the cells with a detergent to solubilize the lipids before staining. However, the fluorometric spectrographs still did not demonstrate Nile Red fluorescence. My speculations about these results include imperfect execution procedures, or possibly the thickening of N. oculata's cell walls due to stress.
Gravimetric analyses proved more successful, but there were still some bumps along the way. My initial use of filters to measure biomass started well; it was simple and quick, though it turned out not to collect as much of the algae as centrifugation. However, once dried, the algae proved impossible to scrape off for lipid extraction, so I used centrifugation instead. My procedures changed slightly between Trials A and B of lipid extraction to improve the amount of lipid captured for triglyceride analysis: I used more initial culture (80 ml rather than 56 ml) as well as using more ultrasonification in lysing. However, though my procedures improved during experimentation to maximize data collection, the two sets of data were clearly reflective of each other (evaluations showed the same correlations between the independent and dependent variables), thereby allowing for quite accurate internal, and even external, comparison of the samples when the sets of data were averaged. Making my analysis more accurate in the second trial improved my study overall.
Another issue I came across in gravimetric analysis was that of inconsistent weight readings; I soon took up weighing everything twice, not wearing gloves when handling weighted vials and beakers to avoid a transfer of static charge, and waiting for a count of ten for the scale to stabilize. Once I had established these weighing protocols, I got consistent readings that made sense.
Also, I used hexane-isopropanol rather than chloroform-methanol extraction in Trial B of lipid analysis since I wanted indications of both total lipids and triglycerides, the lipids used in biodiesel. Chloroform-methanol extraction (Trial A) showed far higher amounts than hexane-isopropanol (Trial B), as expected; but for internal comparison of samples, the two sets of data are very comparable. In fact, this relation implies something more about the lipids in microalgae in general: that a distinct correlation between cellular triglycerides and lipids exists, and that evaluating for one is indicative of the other. Though I had guessed as much from my knowledge of the nature of microalgae, I had decided to test for both to ensure the veracity of my claim, and this study establishes the truth of this assumption.
Summary
Since there is no easy formula for growing and evaluating algae successfully with minimal hassle, much of this year's study was the discovery of the best ways to grow, harvest and lipid-analyze algae. In each of these processes my protocols became more accurate as I familiarized myself with them, and each new method I tried added to my data. Thus, this experiment provided me with information not only on how CO2 infusion and nitrogen limitation affect N. oculata growth and lipid content, but also on how certain procedures tend to affect those results. For instance, I learned that CO2 buildup in media without a buffer causes pH to drop; that filtration is not as effective at biomass evaluation as centrifugation; that extra lysing of cells helps increase the lipid yielded during solvent extraction; that weighing everything twice and without charged rubber gloves increases accuracy. Understanding techniques like these will help me in future forays into the mysteries of algae.
Executing this experiment taught me a lot regarding laboratory procedures, chemical processes and certainly growing algae! It is definitely much more difficult than I thought it would be, and more complicated. In this one study, I came across so many issues and questions that I wanted to pursue: the effects of other nutrients, differences in protocols, effectiveness of analyses, enzymatic factors behind lipid accumulation, and more.
Were I to perform this experiment again, I would first of all use a buffer to keep my carbon dioxide trials alive. In addition, I would perhaps use a different or additional algal strain, one with thinner cell walls more suited to Nile Red staining. Something else that this study has made me interested in is investigating is whether algae initially grown under control conditions but nitrogen-starved later in their growth could produce even more overall lipids, combining the high biomass I experienced with the carbon dioxide-infused trials with the characteristic high-lipid content of nitrogen depletion. Finally, I would continue work on optimizing procedures, perhaps coming up with easier and more effective methods of growing and analysis to utilize. With all I have learned about laboratory procedures, growing algae and extraction processes, I am far better prepared for more extensive research into the world of algae biofuels.
Conclusion
In conclusion, my hypothesis was partially supported. Samples stressed by nitrogen limitation did indeed produce higher percentages of lipid though they yielded less biomass; the control cultures produced more biomass than the nitrogen-limited samples, and lower percentages of lipid. Also, the total lipid yield per unit of culture was slightly higher in the nitrogen-limited samples, as predicted. Surprisingly, however, the carbon dioxide-infused trials died after a week of culturing, despite my best efforts at pH regulation.
Thus, carbon dioxide infusion without thorough pH regulation is clearly not ideal for algae growth or lipid content. And in order to maximize lipid production from Nannochloropsis oculata, growing under nitrogen limitation seems a good alternative to control conditions. Despite the fact that the nitrogen-stressed samples actually produced fewer algae, they had a higher cellular lipid content and higher overall oil yields, and therefore offer greater potential for use in biofuels. With the knowledge that this factor can optimize the lipid content of algae, future work can be done in increasing biomass. Growing algae first under control conditions and then imposing nitrogen starvation could produce an ideal situation for maximizing the efficiency of biofuel production.
Application
Particularly pertinent in today's energy-dependent world, this study adds to ongoing work into optimizing algal oil yields for making algae biodiesel widely practicable. Finding a relatively easy-to-produce, renewable, high-yielding natural oil not competitive with food crops is key for making environmentally friendly biofuels a feasible alternative to fossil fuels. Developing an energy source not reliant on limited petroleum supplies is vital for reasons ranging from environmental preservation to national security and economic stability.
The idea that a fast-growing, durable crop that can grow on non-arable land could produce the oil to fuel the modern world is, quite frankly, amazing, and algae biodiesel holds this promise. My results illustrate the potential. Even on an extremely small scale, without commercial procedures and equipment, I was able to grow enough algae to obtain oil! My study also indicates the importance of further research into optimizing algae growth to make algae biodiesel truly practical for widespread use.
My experiment shows the beneficial impact that nitrogen limitation during growth can have on N. oculata's oil yields. Though this factor decreases biomass, it increases cellular lipid content, more than accounting for the lost algal matter. Added work into optimizing the biomass of nitrogen-limited microalgae, such as implementing nitrogen starvation after the biomass peaks under control conditions, could increase oil yield even more.
This study augments existing research on growing N. oculata by highlighting some essential issues in growth. For instance, the low pH resulting from bubbling air-and especially carbon dioxide-was a problem I had not anticipated. Clearly, pH monitoring is vital to keeping the cultures alive, especially those with high amounts of added carbon dioxide.
Overall, the need for finding a reliable, efficient alternative fuel has never been greater. As this study underlines, with optimized oil yields, algae biodiesel has the potential to meet that aim.
Acknowledgments
Many thanks to the following people for their great help:
- Dr. Matthew Reuer, in the Environmental Science Program at Colorado College in Colorado Springs, for advising me in gravimetric analysis and allowing me the use of his laboratory and equipment. Most of my analytical work was done in Dr. Reuer's laboratory.
- Dr. Nancy Huang, assistant professor in the Biology Department, and Dr. Murphy Brasuel, in the Department of Chemistry and Biochemistry at Colorado College, for advising me in fluorometry and allowing me use of their lab and equipment.
For interviews and advice, particularly about algal growth:
- Dr. Bryan Willson, professor of mechanical engineering, research director of the Engines & Energy Conversion Lab at Colorado State University, and founder of Solix Biofuels.
- Dr. Tracy Yates, senior biologist at Solix Biofuels, in Fort Collins, Colorado.
- Dr. Al Darzins, principal group manager of the National Bioenergy Center at the National Renewable Energy Laboratories in Golden, Colorado.
- Dr. Jerry Brand, professor in the School of Biological Sciences at the University of Texas in Austin.
- Eric Stenn, president of AlgaGen LLC in Vero Beach, Florida.
Bibliography
Barsanti, Laura, and Paolo Gualtieri. Algae: Anatomy, Biochemistry, and Biotechnology. Boca Raton, FL: Taylor & Francis Group, 2006.
"Biodiesel from Algae Oil - Oilgae - Information, News, Links for Algal Fuel, Alga Biodiesel, Biofuels, Algae Biofuel, Energy." Retrieved from http://www.oilgae.com
Chiu, Sheng-Yi, et al. "Lipid Accumulation and CO2 Utilization of Nannochloropsis oculata." Bioresource Technology 100 (January 2009): 833-838. Retrieved on 12 January 2010 from http://www.ncbi.nlm.nih.gov/pubmed/18722767
Cooksey, Keith E., James B. Guckert, Scott A. Williams, and Patrik R. Callis. "Fluorometric Determination of the Neutral Lipid Content of Microalgal Cells Using Nile Red." Journal of Microbiological Methods 6 (1987): 333-345. Retrieved on 13 January 2010 from http://www.montana.edu/wwwmb/uploads/pdfs/faculty/kcooksey/fluorometric_cooksey.pdf
Holmes-Farley, Randy. "Refractometers and Salinity Measurement." Reefkeeping Magazine 5.11 (December 2006). Retrieved on 27 December 2010 from
http://reefkeeping.com/issues/2006-12/rhf/index.php
Lavens, P., and P. Sorgeloos, eds. "Manual on the Production and Use of Live Food for Aquaculture." FAO Fisheries Technical Paper. Rome: Food and Agriculture Organization of the United Nations, 1996. Retrieved from http://www.fao.org/docrep/003/w3732e/w3732e06.htm
Lee, Seong June, Byung-Dae Yoon, and Hee-Mock Oh. "Rapid Method for the Determination of Lipid from the Green Algae Botrycoccus braunii." Biotechnology Techniques 12.7 (July 1998): 553-54. Retrieved from http://www.springerlink.com/content/u588655367v71477/
Lembi, Carole A., and J.R. Waaland. Algae and Human Affairs. Cambridge, England: Cambridge University Press, 1988.
"Lipid Extraction, General Methods." Cyberlipid Center: Resource Site for Lipid Studies. Retrieved on 10 January 2010 from http://www.cyberlipid.org/extract/extr0002.htm
Renaud, S.M., and D.L. Parry. "Microalgae for Use in Tropical Aquaculture II: Effect of Salinity on Growth, Gross Chemical Composition and Fatty Acid Composition of Three Species of Marine Microalgae." J. App. Phycol 6 (1994): 347-356.
Rocha, Jorge M., Juan E. Garcia, and Marta H. Henriques. "Growth Aspects of the Marine Microalga Nannochloropsis gaditana." Biomolecular Engineering 20.4-6 (July 2003): 237-242. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/12919803
Sieg, David. Making Algae Photobioreactors at Home.
Retrieved from http://www.making-biodiesel-books.com/algae-photobioreactor.html
"UTEX The Culture Collection of Algae." University of Texas at Austin. Retrieved from http://web.biosci.utexas.edu/utex/
Wang, Zi Teng, Nico Ullrich, Sunjoo Joo, Sabine Waffenschmidt, and Ursula Goodenough. "Algal Lipid Bodies: Stress Induction, Purification, and Biochemical Characterization in Wild-Type and Starchless Chlamydomonas reinhardtii." Eukaryotic Cell 8.12 (December 2009): 1856-1868. Retrieved from http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2794211/